Induction of striatal long-term synaptic depression by moderate frequency activation of cortical afferents in rat
1 Laboratory for Integrative Neuroscience, National Institute on Alcohol Abuse and Alcoholism, NIH, Rockville, MD 20852, USA
2 Department of Pharmacology, Vanderbilt University School of Medicine, Nashville, TN 37232, USA
Abstract
The striatum regulates motor output, and it is thought that changes in the synaptic efficacy of inputs to the striatum contribute to motor learning and habit formation. Previously, several laboratories have observed that brief high frequency stimulation (HFS) of cortical afferents innervating the dorsolateral striatum induces a long-term decrease in synaptic efficacy called long-term depression (LTD). We recently showed that HFS-induced striatal LTD requires retrograde signalling involving postsynaptic release of endocannabinoids and activation of presynaptic CB1 cannabinoid receptors. In the present study we have employed whole-cell recording in brain slices to examine a new form of LTD at corticostriatal synapses that can be induced by a 10 Hz, 5 min train. The decrease in synaptic efficacy is associated with a decrease in presynaptic release probability, as demonstrated by a decrease in frequency but not amplitude of spontaneous excitatory postsynaptic currents (sEPSCs) and an increase in the paired pulse ratio (PPR). This form of LTD is blocked by antagonists for CB1 and D2 dopamine receptors and impaired by blockers of L-type calcium channels. However, 10 Hz-induced LTD does not depend on postsynaptic depolarization, unlike HFS-induced LTD. Furthermore, this new form of LTD is not prevented by treatments known to block HFS-induced LTD, including antagonism of metabotropic glutamate receptors (mGluRs), chelation of postsynaptic calcium, or intracellular application of an anandamide membrane transport inhibitor (VDM11). From these findings it is not clear that the endocannabinoid responsible for this form of LTD acts in a retrograde fashion, and the cellular source of endocannabinoid necessary for 10 Hz-induced LTD is as yet unknown. Our results demonstrate that a prolonged moderate frequency train induces cannabinoid-dependent LTD, further supporting the idea that endocannabinoids play a prominent role in the regulation of long-lasting changes in striatal output.
, 百拇医药
Introduction
Information storage at central synapses is widely attributed to long-lasting activity-dependent changes in synaptic efficacy. We have been investigating such changes in the striatum, a brain region with important roles in motor learning and habit formation. High frequency activation of inputs to the striatum can induce either long-lasting increases or decreases in transmission at excitatory synapses onto striatal medium spiny neurones (MSNs) (Calabresi et al. 1992a,b; Lovinger et al. 1993; Partridge et al. 2000; Kerr & Wickens, 2001). The long-lasting decrease in efficacy, long-term depression or LTD, is especially prominent in the dorsolateral striatum (Partridge et al. 2000; Smith et al. 2001). Expression of high frequency stimulation (HFS)-induced striatal LTD (HFS-LTD) is presynaptic and due to a decrease in the probability of glutamate release, as shown by a decrease in the frequency but not amplitude of spontaneous EPSCs (Choi & Lovinger, 1997a) and an increase in the paired pulse ratio (PPR) (Choi & Lovinger, 1997b). However, HFS-LTD induction requires postsynaptic activation. Calcium chelators (i.e. 10 mM EGTA) introduced into the postsynaptic neurone block LTD induction (Choi & Lovinger, 1997b). In addition, HFS-LTD induction requires convergent activation of group I metabotropic glutamate receptors (mGluRs) (Sung et al. 2001; Gubellini et al. 2001), D2 dopamine receptors (D2Rs) (Calabresi et al. 1997; Tang et al. 2001), and L-type voltage-gated calcium channels (Calabresi et al. 1994; Choi & Lovinger, 1997b). The decrease in presynaptic release probability and the dependence on postsynaptic activation demonstrates the necessity for a retrograde messenger to explain HFS-LTD (Choi & Lovinger, 1997b). Indeed, HFS-LTD requires the activation of a presynaptic CB1 cannabinoid receptor, presumably via postsynaptic calcium-dependent synthesis and release of endocannabinoids (Gerdeman et al. 2002).
, http://www.100md.com
The MSNs of the striatum receive convergent glutamatergic inputs from the cortex and thalamus and dopaminergic inputs from the substantia nigra pars compacta. Simultaneous activation of many cortical afferents is necessary to depolarize MSNs to action potential firing threshold. The HFS protocol used to induce LTD mimics firing patterns found in deep-layer cortical pyramidal neurones that innervate the striatum. These neurones can fire action potentials at 100 Hz or more for up to 3 s (Kasper et al. 1994). However, other firing patterns of deep-layer cortical neurones innervating the dorsal striatum have been reported. In vivo, moderate frequency membrane potential oscillations (5 Hz) were recorded in MSNs that were due to the synchronous firing of cortical inputs (Charpier et al. 1999). In addition, in vitro recordings from layer 5 rat prefrontal cortical neurones demonstrated firing at irregular frequencies with an average of 5–10 Hz (Fellous et al. 2003). Given the variable frequencies of in vivo spike trains at the corticostriatal synapse, it is reasonable to speculate that an alternate frequency train may produce plasticity, perhaps including LTD. Recently, a moderate frequency stimulation (13 Hz, 10 min or 10 Hz, 5 min)-induced form of LTD at glutamatergic synapses in the nucleus accumbens was shown to be CB1-dependent (Robbe et al. 2002, Hoffman et al. 2003). Although induction of this form of LTD is dependent on the activation of postsynaptic group I mGluRs and calcium release from internal stores, it is expressed presynaptically. Taken together, these findings led us to speculate that stimuli delivered at the moderate frequency of 10 Hz might induce LTD in the dorsal striatum. In the present study we examined changes in the efficacy of glutamatergic synapses in response to afferent stimulation at this moderate frequency in corticostriatal brain slices from rat.
, http://www.100md.com
Methods
Brain slice preparation
Brain slices were prepared as described previously (Ronesi et al. 2004) from Sprague-Dawley rats aged 16–21 postnatal days (Charles River Laboratories, Wilmington, MA, USA). Animals were anaesthetized with halothane (Sigma, St Louis, MO, USA) and killed by decapitation, in accordance with the National Institutes of Health animal welfare guidelines. The brains were removed and transferred to an ice-cold modified artificial cerebrospinal fluid (aCSF) containing (mM): 194 sucrose, 30 NaCl, 4.5 KCl, 1 MgCl2, 26 NaHCO3, 1.2 NaH2PO4, 10 D-glucose, and brought to pH 7.4 by bubbling with 95% O2–5% CO2. Coronal slices of 350 μm thickness were made in ice-cold modified aCSF using a manual vibroslice (World Precision Instruments, Sarasota, FL, USA). Slices were transferred immediately to a nylon net that was submerged in normal aCSF containing (mM): 124 NaCl, 4.5 KCl, 2 CaCl2, 1 MgCl2, 26 NaHCO3, 1.2 NaH2PO4, and 10 D-glucose. ACSF was maintained at a pH of 7.4 and osmolarity was 325 mosmol l–1. Slices were incubated for at least 30 min at room temperature (19–21°C) prior to the start of recording. Hemislices containing both the cortex and the striatum were then transferred to the recording chamber, submerged in normal ACSF containing, in the majority of experiments, 25 μM picrotoxin (to prevent GABAA receptor-mediated synaptic responses). During electrophysiological recordings, slices were maintained at a temperature between 29 and 32°C, stable within 1°C during any given experiment.
, 百拇医药
Whole-cell voltage-clamp recording
Whole-cell recordings were performed as described previously (Ronesi et al. 2004). Pipettes (2.5–5.0 M) made from borosilicate glass capillaries were pulled on a Flaming-Brown micropipette puller (Novato, CA, USA) and filled with an internal solution containing the following (mM): 120 caesium methane sulphonate, 5 NaCl, 10 tetraethylammonium chloride, 10 Hepes, 4 lidocaine N-ethyl bromide, 1.1 EGTA, 4 Mg-ATP, and 0.3 Na-GTP, pH 7.2, and osmolarity adjusted to 298 mosmol l–1. In some experiments, BAPTA (20 mM, Sigma) was substituted for EGTA in the internal solution. Access to medium spiny neurones 3–4 layers below the slice surface was aided with a differential interference contrast (DIC)-enhanced visual guidance system. Cells were voltage clamped at –70 mV and maintained at this potential prior to, during, and subsequent to the moderate frequency stimulus train. Afferent stimuli were delivered via a master-8 stimulator (A.M.P.I., Jerusalem, Israel) and a bipolar-twisted tungsten wire that was placed in the white matter dorsal to the lateral region of the striatum, and test stimuli before and after the 10 Hz train were given at a frequency of 0.05 Hz. The stimulus intensity was set to yield an excitatory postsynaptic current (EPSC) amplitude between 200 and 400 pA. Series resistance was < 10 M and was uncompensated. Both series and input resistances were monitored throughout the experiment by delivering 5 mV voltage steps. If series resistance changed more than 20% during the course of an experiment, the cell was discarded.
, 百拇医药
Synaptic currents were recorded with either an Axopatch 1D or Axopatch 200A amplifier (Axon Instruments, Foster City, CA, USA), filtered at 5 kHz, digitized at 10 kHz, and stored on a microcomputer. EPSC amplitudes were examined using peak detection software in pCLAMP8 (Axon Instruments, Union City, CA, USA). The moderate frequency stimulation protocol consisted of a 10 Hz, 5 min train of stimuli at the same intensity used for pre- and post-train test pulses. For some experimental conditions, a high frequency train was given. This HFS protocol consisted of four 100 Hz trains (1 s duration) with an intertrain interval of 10 s, where each train was paired with a depolarizing step to 0 mV in the postsynaptic neurone. HFS experiments were performed to provide evidence that drugs known to block HFS-induced LTD that had no effect on 10 Hz-induced LTD (10 Hz-LTD) were actually penetrating slices sufficiently to affect some aspect of synaptic transmission. Effects on HFS-induced LTD were only examined in 2–3 cells per treatment group in the present study because results were similar to those observed in previous studies with larger numbers of cells (Choi & Lovinger 1997b; Tang et al. 2001; Sung et al. 2001; Ronesi et al. 2004). The EPSC amplitude of the response over the 5 min prior to the 10 Hz train was averaged and compared to the average EPSC amplitude of the response 20–30 min post train. LTD was defined as a change in mean EPSC amplitude that was greater than two standard deviations from the baseline EPSC mean. In figures depicting the time course of an LTD experiment, data points were plotted as means ± S.E.M. normalized to baseline. The paired pulse ratio (PPR, interval of 50 ms) was calculated as the ratio of the mean amplitude of the second response during a 10 min time period over the mean amplitude of the first response during the same period (Kim & Alger, 2001). However, for the points plotted in Fig. 1B, the PPR was calculated for each time point in order to show the time course of the PPR change following the train. Mean PPR values 10 min prior to the train were compared to PPR values 20–30 min post train using an unpaired t test, where P < 0.05 was considered statistically significant. To determine if the magnitude of LTD was impaired under certain pharmacological conditions, an unpaired t test was used (significance at P < 0.05) to compare the EPSC amplitude at 20–30 min post train between untreated and treated cells. Spontaneous release events were recorded for at least 2 min beginning 5 min before the 10 Hz train and 30 min after the train. Data were visualized both on an oscilloscope and in the Clampex program, and data were stored directly in Clampex using a free-running data collection protocol. During the collection of sEPSC data, current records were amplified at a gain of 2 mV pA–1, filtered at 5 kHz, and digitized at 10 kHz. Quantal events were detected and analysed offline using the Mini Analysis program by Synaptosoft, version 5.6 (Decatur, GA, USA). The current records were visually inspected following automated analysis to prevent false positive identification and false negative rejection of sEPSCs. Peak amplitude, interevent interval, rise time, and decay time were measured. The threshold for peak amplitude detection (typically 7 pA) was twice the baseline root mean square noise level. Data for cumulative EPSC histograms were compared statistically using the Kolmogorov-Smirnov test, where the statistical criterion for significance was P < 0.005.
, http://www.100md.com
A, MSNs in intact coronal brain slices containing the cortex and striatum were voltage clamped at –70 mV. Time course of LTD during and after a moderate frequency train (train indicated by arrow at time = 0). After at least 10 min baseline, 10 Hz, 5 min stimulation of cortical afferents produced a long-lasting decrease in EPSC amplitude recorded from striatal MSNs. Dotted line represents baseline. Standard error bars are smaller than symbols. Sample traces are averages of 15 evoked EPSCs immediately before (left) and 25 min after (right) the train. Scale bar: 100 pA, 25 ms. B, representative sample traces of evoked EPSCs during start (first 900 ms) and end (last 900 ms) of a 10 Hz, 5 min train. Note that there is no change in baseline current, but a marked decrease in EPSC amplitude. Scale bar: 100 pA, 100 ms. C, representative experiment showing that 10 Hz-induced LTD does not occlude HFS-LTD. D, time course of PPR during and after the 10 Hz stimulation train. Note the lasting increase in PPR.
, 百拇医药
Pharmacology
AM251, sulpiride, CPCCOEt, LY341495 and VDM11 were obtained from Tocris-Cookson (St Louis, MO, USA) and nifedipine was obtained from Calbiochem (San Diego, CA, USA). AM251, sulpiride, and LY341495 were dissolved in DMSO at 10 mM, nifedipine was dissolved in DMSO at 20 mM, CPCCOEt was dissolved in DMSO at 80 mM, and VDM11 was dissolved in DMSO at 50 mM. The final concentration of DMSO in the extracellular solution never exceeded 0.1%, and the concentration of DMSO in the internal solution was 0.02% in internal VDM11 experiments. Picrotoxin and D-aminophosphonovalerate (APV) were purchased from Sigma and dissolved directly into the aCSF. For AM251 wash-on experiments, 0.5 mg ml–1 bovine serum albumin was included in the external solution to prevent this drug from adhering to the perfusion apparatus. aCSF and drugs were bath applied using gravity flow at a rate of 1–2 ml min–1.
, 百拇医药
Results
Induction of striatal LTD with a 10 Hz, 5 min train
During whole-cell recordings from MSNs in coronal slices containing the cortex and striatum, 200–400 pA EPSCs were evoked by stimulation at 0.05 Hz with a stimulating electrode placed in the white matter overlying the striatum. All of the recordings were performed in the dorsolateral striatum, where it was previously shown that LTD is the predominant form of HFS-induced synaptic plasticity (Partridge et al. 2000; Smith et al. 2001). Activating cortical afferents to the dorsolateral striatum at 10 Hz for 5 min induced a long-lasting decrease in EPSC amplitude (Fig. 1A). LTD was defined as a persistent decrease in EPSC amplitude (at 20–30 min post train) that was greater than two standard deviations less than the baseline EPSC amplitude mean. In 21 of 24 cells subjected to the 10 Hz stimulation protocol this criterion was met, and the average post-train EPSC amplitude in these cells was 61 ± 3% of the mean baseline response (Fig. 1A). During the course of the 5 min train, the amplitude of evoked EPSCs was diminished and in some cases the EPSC was completely absent in response to the last few stimuli of the train. Individual EPSCs decayed completely back to baseline current levels between stimuli, and no detectable change in net baseline current developed over the course of the 5 min stimulation (Fig. 1B).
, 百拇医药
To determine if LTD induced by 10 Hz stimulation and HFS could be observed in the same neurone and to determine if we could observe evidence of saturation of LTD expression mechanisms we delivered a 10 Hz stimulus train, which was followed 30 min later by four HFS trains combined with postsynaptic depolarization (Fig. 1C). The decrease in EPSC amplitude was robust in response to both stimulus trains. The mean EPSC amplitude after the 10 Hz train was 58 ± 7% of baseline and was 44 ± 2% of baseline following the HFS train (n = 4). When an HFS train was given first followed by a 10 Hz train, a similar result was produced. The mean EPSC amplitude was 57% of baseline following the HFS train and was 36% of baseline following the 10 Hz train (n = 2). Interestingly, giving an LFS train followed 30 min later by another LFS train did not appear to produce saturation of LTD. The mean EPSC amplitude 20–30 min after the first 10 Hz train was 67 ± 5% of baseline and 45 ± 6% of baseline following the second train (n = 4). Moreover, delivery of two consecutive HFS trains induced LTD, as the mean EPSC amplitude following the first HFS train was 73% of baseline and was 48% of baseline following the second train (n = 2). Thus, we did not observe evidence of LTD saturation with any stimulus protocol. Unfortunately, we were not able to maintain stable recordings for a period long enough to determine if LTD was saturated with more than two stimulus trains, and it was not possible to determine if the two forms of LTD could truly occlude one another. However, our findings do indicate that both 10 Hz and HFS-induced LTD can occur at synapses onto the same neurones.
, 百拇医药
Picrotoxin (25 μM), a GABAA receptor antagonist, was included in the extracellular solution in order to isolate pure EPSCs, since GABAergic interneurones and local axon collaterals are activated by afferent stimulation. In the absence of picrotoxin, 10 Hz-induced striatal LTD was unimpaired, demonstrating that GABAergic transmission does not play an appreciable role in this form of LTD. LTD was observed in 7 of 8 neurones, and had a typical mean LTD magnitude of 59 ± 5% of baseline (P > 0.05, unpaired t test). Therefore, the data obtained from experiments performed in the absence of picrotoxin were included with those from experiments performed in the presence of picrotoxin. All other experiments described in this study were done with picrotoxin in the extracellular solution.
, 百拇医药
Presynaptic expression of 10 Hz-induced striatal LTD
To determine if striatal LTD induced by 10 Hz stimulation is associated with changes in the ratio of responses to paired stimuli, an index of relative neurotransmitter release probability, we examined the amplitude of EPSCs evoked by two stimuli of equal magnitude delivered 50 ms apart. The ratio of the amplitudes of the second evoked EPSC relative to the first (PPR ratio) averaged 1.09 ± 0.05 in the pre-train baseline condition. In the 21 of 24 neurones that demonstrated LTD as defined by the above criterion, PPR increased to 122 ± 3% of the baseline value 20–30 min following the 10 Hz stimulus train (Fig. 1D).
, 百拇医药
We also measured the frequency, amplitude and time course of spontaneous EPSCs (sEPSCs) to determine the possible pre- or postsynaptic locus of the change in synaptic efficacy (Redman, 1990). It should be noted that the frequency and amplitude of sEPSCs recorded in this corticostriatal slice preparation are maintained even in the presence of sodium channel blockade (Tang & Lovinger, 2000), and thus more than likely represent quantal, miniature EPSCs. Due to difficulties in maintaining stable recordings and to a low frequency of spontaneous events in many cells, only a small number of neurones were examined for 10 Hz-induced changes in sEPSC frequency and/or baseline. Representative traces of sEPSCs recorded before and after the train are shown in Fig. 2A. In the pre-train baseline period, sEPSC frequency averaged 1.5 ± 0.6 Hz (n = 3). Following the 10 Hz stimulus train, the frequency of sEPSCs dropped to 42 ± 6% of baseline (Fig. 2B, n = 3), and all three cells exhibited a significant increase in interevent interval, as can be seen in the cumulative probability distributions shown in Fig. 2D (P < 0.005, Kolomogorov-Smirnov test). In contrast, the amplitude of sEPSCs was unchanged. The median event amplitude was 17 ± 1 pA prior to 10 Hz stimulation, and remained essentially the same (101 ± 10% of baseline) following the stimulus train (Fig. 2B, n = 3). All three neurones exhibited no change in amplitude distribution (Fig. 2C; P > 0.005, Kolmogorov-Smirnov test). The 10–90% rise times (Fig. 2B, 103 ± 7% of baseline) and half-decay times (Fig. 2B, 117 ± 5% of baseline) of sEPSCs did not change following the 10 Hz train (P > 0.05). When stimulation was maintained at 0.05 Hz throughout the experiment there were no apparent changes in the frequency or amplitude of sEPSCs, demonstrating that the decrease in sEPSC frequency was not simply due to drift over time during low frequency (0.05 Hz) stimulation of cortical inputs. Under this condition, the frequency of sEPSCs was 101 ± 13% baseline and the amplitude was 85 ± 3% baseline (n = 3).
, http://www.100md.com
A, representative sample traces of spontaneous EPSCs recorded 5 min before (pre-LTD) and 30 min after (post-LTD) a 10 Hz train. Scale bars: 25 pA, 1 s. B, bar graph showing mean percentage change relative to baseline in frequency, amplitude, 10–90% rise time, and decay time of sEPSCs (n = 3). Twenty to thirty minutes after the train, there is a decrease in spontaneous EPSC frequency but not amplitude, rise time, or decay time. C, averaged (n = 3) cumulative probability amplitude distribution of sEPSCs before () and after () a 10 Hz train. D, cumulative interevent interval distribution of sEPSCs before and after a 10 Hz train.
, http://www.100md.com
We wanted to determine if shorter duration 10 Hz trains could induce LTD. Activating corticostriatal synapses for 1 min at 10 Hz did not induce LTD. The mean EPSC amplitude 20–30 min after the train was 92 ± 4% of baseline and the mean PPR was 110 ± 4% of baseline (n = 4). Likewise, a 2 min, 10 Hz train did not induce LTD. The EPSC amplitude was 87 ± 5% of baseline and the PPR was 101 ± 5% of baseline (n = 3). Interestingly, the EPSC amplitude dropped to 7 ± 3% (n = 5) of the value of the start of the train within the first minute of the onset of 10 Hz stimulation. Thus, loss of the EPSC during the stimulus train does not appear to signal the onset of LTD, as several more minutes of 10 Hz stimulation are needed for LTD induction.
, http://www.100md.com
LTD requires activation of CB1 cannabinoid receptors and D2 dopamine receptors
Slices that were treated throughout the duration of the experiment with the CB1 antagonist AM251 (2–4 μM) did not exhibit a long-lasting decrease in EPSC amplitude in the majority of neurones examined (Fig. 3A, P > 0.05, unpaired t test). In 7 of 9 cells receiving 10 Hz stimulation, the mean EPSC amplitude 20–30 min after the train was 103 ± 12%. The PPR was also unchanged and was 106 ± 7%. In the remaining two neurones, the EPSC amplitude was decreased to 77 and 61% of baseline, and PPR was increased by 115 and 130%, respectively. Application of AM251 10 min after the termination of the train did not affect the maintenance of LTD. The mean EPSC amplitude was 64% of baseline and the mean PPR was 122% of baseline (n = 2, Fig. 3A).
, 百拇医药
A, slices treated with the CB1 antagonist AM251 (2–4 μM) prior to the train failed to exhibit LTD, whereas bath application of AM251 (2 μM) 10 min after the train did not affect LTD amplitude (train indicated by arrow at time = 0). Dotted line represents baseline and bar indicates time course of drug perfusion. B, slices treated with the antagonist sulpiride (2–10 μM) failed to exhibit LTD, whereas bath application of sulpiride (4 μM) 10 min after the train did not affect LTD amplitude. Bar denotes time course of drug perfusion. Above, 15 trace averages of EPSCs recorded before (left) and 30 min after (right) 10 Hz train (only pre-train drug condition shown). Scale bars: 100 pA, 25 ms.
, 百拇医药
Bath application of the D2 antagonist sulpiride (2–10 μM) during the train, and 10 min before and after the train, blocked 10 Hz-induced LTD in 5 of 5 cells (Fig. 3B, P > 0.05, unpaired t test). The percentage of baseline EPSC amplitude 20–30 min after the train was 92 ± 7. The PPR was also unchanged and averaged 105 ± 4%. As reported previously, bath application of the D2 antagonist sulpiride for the duration of the experiment also blocked HFS-induced striatal LTD. At 20–30 min after the train, the EPSC amplitude was 95% of baseline (n = 2). Sulpiride alone had no effect on baseline transmission (data not shown, Tang et al. 2001). When sulpiride was applied 10 min after the termination of the train, there was no change in the magnitude of LTD (P > 0.05, unpaired t test). The mean EPSC amplitude 20–30 min after the train was 54 ± 3% of baseline and the mean PPR was 127 ± 3% of baseline (n = 3, Fig. 3B).
, 百拇医药
Differential induction mechanisms of 10 Hz- and HFS-induced LTD
To determine if the activation of group I mGluRs is necessary for the induction of 10 Hz-induced LTD, slices were treated with the non-competitive antagonist CPCCOEt (80 μM) during, and 10 min before and after, the train (Fig. 4A). We observed a typical decrease in EPSC amplitude in 5 of 6 cells to 61 ± 2% of baseline value 20–30 min following the 10 Hz train (P > 0.05, unpaired t test). As in control cells, this decrease in EPSC amplitude was associated with an increase in the PPR. The PPR was 121 ± 3% of baseline. Bath application of CPCCOEt had no effect on baseline EPSC amplitude (data not shown, Sung et al. 2001). Also, extracellular application of the mGluR antagonist LY341495, at a concentration (10 μM) that not only blocks group II mGluRs, but is also non-selective with respect to both group I mGluRs, had no effect on the magnitude of 10 Hz-LTD (Fig. 4B, n = 4; EPSC, 64 ± 4% of baseline; PPR, 124 ± 9% of baseline, P > 0.05, unpaired t test).
, 百拇医药
A and B, slices pretreated with the mGluR1 antagonist CPCCOEt (80 μM) or the broad-spectrum mGluR antagonist LY341495 (10 μM) exhibit LTD with a magnitude similar to that observed in control slices (train indicated by arrow at time = 0). C, slices treated with the L-type calcium channel blocker nifedipine (20 μM) exhibit LTD, but the magnitude of the decrease in EPSC amplitude is smaller than in control slices (train given at time = 0). Dotted line represents baseline and bar denotes time course of drug perfusion.
, http://www.100md.com
The activation of NMDA receptors also does not appear to be crucial for 10 Hz-LTD. When the NMDA receptor antagonist APV (100 μM) was applied extracellularly, the expression of LTD was unimpaired (P > 0.05, unpaired t test). At 20–30 min post train, the average EPSC amplitude was 54 ± 1% of baseline and the PPR was 121 ± 4% of baseline (n = 4). However, bath application of the L-type calcium channel blocker nifedipine (20 μM) significantly impaired the magnitude of LTD (Fig. 4C, n = 6, P < 0.01, unpaired t test) and was not associated with a robust increase in the PPR. The average EPSC amplitude was 74 ± 3% of baseline and the average PPR was 108 ± 3% of baseline.
, 百拇医药
Including the calcium chelator BAPTA (20 mM) in the postsynaptic neurone via the patch pipette failed to block 10 Hz-induced LTD in the majority of neurones examined. Following the train, 5 of 6 neurones exhibited a decrease in EPSC amplitude 20–30 min after the train that was greater than two standard deviations below the baseline mean (Fig. 5A, P > 0.05, unpaired t test). The percentage of baseline EPSC amplitude was 53 ± 8 and the percentage of baseline PPR was 131 ± 4. As previously reported, chelating calcium postsynaptically prevented HFS-induced LTD in three neurones. The mean post-train EPSC amplitude was 85 ± 7% of baseline in these neurones. Intracellular BAPTA has no apparent effect on baseline synaptic efficacy, as the stimulus magnitudes required to elicit 200–400 pA currents (0.4–1.0 mA, 40–50 μs duration) were similar to those used to evoke EPSCs under control conditions (0.3–1.0 mA, 40–60 μs duration).
, 百拇医药
A, cells loaded intracellularly with the calcium chelator BAPTA (20 mM) still exhibit 10 Hz-induced striatal LTD (train indicated by arrow at time = 0). B, the AMT inhibitor VDM11 (10 μM) was loaded into the postsynaptic neurone. In 6 of 9 cells LTD was unimpaired, whereas LTD was reduced in 3 of 9 cells. Mean ± S.E.M. normalized EPSC values from all 9 neurones are plotted as a function of time in the graph. Dotted line represents baseline.
HFS-LTD is blocked when inhibitors of the anandamide membrane transporter (AMT) are included in the postsynaptic neurone, presumably by preventing endocannabinoid release (Ronesi et al. 2004). Therefore, the AMT inhibitor VDM11 (10 μM) was included in the patch pipette in an attempt to block 10 Hz-induced striatal LTD (Fig. 5B). Recordings from nine neurones showed that LTD was not significantly decreased in the presence of postsynaptic intracellular VDM11 (P = 0.06, unpaired t test). The mean EPSC amplitude 20–30 min following the train was 70 ± 6% of baseline and the mean PPR was 115 ± 8% of baseline. The decrease in magnitude largely reflects the fact that three neurones exhibited decreased EPSCs averaging 85 ± 7% of baseline. The other neurones in this experiment exhibited robust LTD (62 ± 5% of baseline). In experiments done within the same batches of slices used for the 10 Hz-LTD experiments, VDM11 included internally blocked HFS-LTD, and the EPSC amplitude 20–30 min after HFS was 104% of baseline (n = 2). Like BAPTA, intracellular VDM11 did not appear to effect baseline synaptic efficacy (stimulus magnitude was 0.2–0.9 mA, 40–60 μs duration).
, 百拇医药
Discussion
Although it is well-established that LTD can be induced by HFS of afferents to the dorsolateral striatum, it was not clear if LTD could be induced by afferent activation at other frequencies within the naturally occurring range of cortical neuronal firing. Here we demonstrate that LTD at corticostriatal synapses can be induced by afferent stimulation at a moderate frequency (10 Hz). This form of LTD requires the activation of CB1 cannabinoid and D2 dopamine receptors, and analysis of PPR and sEPSCs indicates that it is expressed as a change in presynaptic function, possibly as a decrease in the probability of glutamate release. This expression mechanism strongly resembles that of HFS-LTD at the same synapse.
, 百拇医药
Endocannabinoids and dopamine mediate synaptic plasticity in the dorsal striatum
In central nervous system neurones, cannabinoids and endocannabinoids work mainly through activation of the CB1 receptor. This receptor is expressed primarily presynaptically throughout the brain, and densely in the forebrain, where its activation is coupled to a decrease in neurotransmitter release (see Freund et al. 2003 for review). At the corticostriatal synapse, activation of CB1 with various synthetic agonists reduces EPSC amplitude by decreasing glutamate release probability (Gerdeman & Lovinger, 2001). Endocannabinoids (e.g. 2-arachidonylglycerol and anandamide) are endogenous ligands for cannabinoid receptors and are synthesized de novo from lipid precursors embedded in the membrane (Cadas et al. 1996; Stella et al. 1997). Endocannabinoid synthesis is inducible, and it is thought that following prolonged postsynaptic depolarization and/or activation of presynaptic inputs, endocannabinoids are made and immediately released. Endocannabinoids have been shown to be necessary for several forms of plasticity, including LTD in the nucleus accumbens and HFS-LTD in dorsal striatum (Gerdeman & Lovinger, 2003).
, 百拇医药
Like HFS-LTD, 10 Hz-induced LTD requires activation of the D2 dopamine receptor. Previous studies have demonstrated dopamine release in response to stimulation in striatal slices, even at very low stimulus frequencies (Zhou et al. 2001; Schmitz et al. 2002). We have observed transient dopamine release in response to single stimuli delivered using the same electrode placement and stimulus parameters used in the present study (Partridge et al. 2002). In vivo microdialysis experiments revealed that activation of D2 increases anandamide levels 8-fold within the striatum (Giuffrida et al. 1999). Therefore, D2 activation may increase endocannabinoid levels in the striatum and promote subsequent CB1 activation. The role of dopamine and D2 receptors in LTD within the dorsal striatum is different from other forms of endocannabinoid-dependent LTD that do not appear to require D2 (Robbe et al. 2002; Chevaleyre & Castillo, 2003). The reasons for this difference are not yet known, but the dorsal striatum has the highest level of dopaminergic innervation in the brain, and it is only in this region that D2 activation has been linked to endocannabinoid synthesis/release. However, continued activation of either CB1 or D2 receptors is not necessary for maintenance of 10 Hz-LTD, as the application of antagonists for these receptors 10 min after the induction of LTD does not affect expression.
, http://www.100md.com
It is possible that local depolarization at distal dendritic sites might contribute to 10 Hz-LTD induction. However, several observations suggest that this is not the case. First, no net inward current was observed during the 10 Hz stimulus train, indicating that no sizeable current is generated anywhere in the cell. We believe a standing inward current of more than a few picoamps conducted from the dendrites to the soma should be readily observed, because we can easily detect sEPSCs. Second, previous studies demonstrated that HFS-LTD is blocked by holding the neuronal membrane potential at –70 mV (Choi & Lovinger, 1997b), which prevents depolarization to a level needed for LTD induction even in the face of much higher frequency input.
, 百拇医药
10 Hz-LTD versus HFS-LTD
Interestingly, we noted several differences between the induction mechanism of 10 Hz-LTD and HFS-LTD. LTD induced by brief HFS trains requires postsynaptic depolarization, activation of L-type calcium channels, increased postsynaptic calcium concentration, and activation of group I mGluRs (Calabresi et al. 1994; Choi & Lovinger, 1997b; Gubellini et al. 2001; Sung et al. 2001). The observations made at present indicate that LTD is induced when striatal afferents are activated at 10 Hz even in the absence of postsynaptic depolarization. Furthermore, 10 Hz-induced LTD was not prevented by blockade of mGluR1 with an antagonist known to prevent the HFS-induced form of plasticity. Our findings also suggest that continuous strong activation of the postsynaptic AMPA-type glutamate receptors is not necessary for LTD induction, since the EPSC was almost completely eliminated within 1–2 min of initiation of the 10 Hz train, a time point at which LTD is not yet initiated. The observation that a high concentration of nifedipine reduces, but does not fully block LTD suggests that induction of this form of LTD is not fully dependent on L-type calcium channel activation. The reduction in LTD by nifedipine but not intracellular BAPTA suggests the possibility that the 10 Hz stimulus produces calcium channel-dependent cannabinoid production in neighbouring neurones that contributes to LTD induction at synapses impinging on the neurones from which the recording was made. However, the lack of full blockade by nifedipine contrasts with HFS-induced LTD. It is possible that induction of 10 Hz-LTD does not depend on an intracellular calcium increase in any cell in the slice. Alternatively, the relative contribution of different sources for increasing intracellular calcium to the induction of 10 Hz- versus HFS-LTD may differ. The lack of effect of mGluR antagonists on 10 Hz-LTD provides the first evidence that striatal LTD does not require these receptors. This is a new and important finding given that past studies of HFS-LTD suggested a necessary role for these receptors. Overall, our results suggest that molecular mechanisms involved in the induction of 10 Hz-LTD are less complex than those involved in HFS-LTD. It will be interesting to determine if the molecular mechanisms underlying this presynaptic change are the same for 10 Hz- and HFS-LTD.
, 百拇医药
Source of endocannabinoids necessary for 10 Hz-induced striatal LTD
Our earlier studies indicated that the MSNs themselves are the primary source of endocannabinoids necessary for LTD induced by HFS. However, we have no solid evidence to date that a postsynaptic neurone is the site of endocannabinoid production/release for 10 Hz-LTD. This form of LTD appears to be independent of postsynaptic depolarization and insensitive to calcium chelation. These findings do not preclude the involvement of a postsynaptic endocannabinoid in 10 Hz-LTD induction, as endocannabinoid synthesis could be calcium independent and not require depolarization or metabotropic receptor activation. For example, 2-arachidonylglycerol (2-AG) can be synthesized by a phospholipase C-mediated increase in diacylglycerol (DAG), the immediate precursor of 2-AG. This appears to be the case for heterosynaptic inhibitory LTD in the hippocampus, which is not blocked by chelating postsynaptic calcium, but is blocked with inhibitors of PLC and DAG lipase (Chevaleyre & Castillo, 2003). However, 10 Hz-induced LTD was not blocked in a large majority of cells when the AMT inhibitor VDM11 was included postsynaptically. We previously demonstrated that this compound blocks HFS-LTD and release of endocannabinoids from MSNs (Ronesi et al. 2004), consistent with the idea that the AMT is involved in endocannabinoid release in that form of LTD. If 2-AG or another postsynaptically released endocannabinoid were responsible for LTD induction, we would have expected blockade of LTD by intracellular VDM11 loading. It is conceivable that both postsynaptic synthesis and release of the endocannabinoid involved in 10 Hz-LTD induction differ from those required for HFS-LTD induction. However, the existence in the same neurone of two such mechanisms for the generation and release of the same molecular signal would be surprising. Our findings certainly raise the strong suspicion that the source of endocannabinoids necessary for this form of LTD may not be the postsynaptic neurone.
, 百拇医药
Despite the evidence discussed above, the involvement of D2 receptors leaves open the possibility of a postsynaptic locus of LTD induction, because these receptors are highly expressed by a subset of MSNs (Gerfen et al. 1990; Sesack et al. 1994). However, D2 receptors have also been localized to the terminals of the corticostriatal afferents (Wang & Pickel, 2002), and are well known to act as autoreceptors on the striatal dopaminergic terminals themselves (Starke et al. 1989), so we cannot rule out the possibility of a presynaptic induction mechanism or a mechanism involving cross-talk between corticostriatal and dopaminergic afferents. In addition to being found at corticostriatal synapses, D2 receptors are found on aspiny acetylcholinergic interneurones (Alcantara et al. 2003), and it is possible that endocannabinoids are being synthesized and released from these cells. If the endocannabinoid necessary for 10 Hz-LTD is produced by neurones other than the postsynaptic MSNs, then synthesis of the endo-cannabinoid could be calcium dependent, and the endocannabinoid might spill over from neighbouring neurones to activate receptors on presynaptic neurones impinging on the neurone under study. Regardless of the site of endocannabinoid release in 10 Hz-LTD, the differential reliance on postsynaptic activation for induction of HFS- versus 10 Hz-LTD suggests that these two forms of endocannabinoid-dependent plasticity could occur under different in vivo conditions. HFS-LTD would be favoured by brief bouts of strong cortical input associated with postsynaptic activity. The LTD described in the present study might occur during more prolonged periods of moderate cortical input that is uncorrelated with postsynaptic activation.
, 百拇医药
Induction of striatal LTD and LTP by different frequency trains
Throughout the brain, different forms of synaptic plasticity can occur at the same synapse following differential frequencies of synaptic activation. For example, a well-characterized phenomenon at the CA3–CA1 synapse in the hippocampus is the induction of LTP with a brief high frequency train and LTD with a prolonged low frequency train. High frequency activation of corticostriatal inputs can produce either LTD or LTP (Calabresi et al. 1992b; Partridge et al. 2000; Kerr & Wickens, 2001), NMDAR-dependent striatal LTP can be depotentiated with a low-freqency train (2 Hz, 10 min) (Picconi et al. 2003), and thus, in striatum, as in other brain regions, it appears that multiple forms of plasticity can be induced by diverse frequencies of afferent activation.
, 百拇医药
Stimulation of the orofacial motor cortex at 5 Hz (100–200 pulses) has been shown to induce LTP of EPSPs recorded intracellularly from medium spiny neurones in the barbiturate-anaesthetized rat (Charpier et al. 1999). This finding might appear to contradict the present results showing LTD induction by a 10 Hz train. This discrepancy can be accounted for by several possible explanations. All of the recordings in this study were performed in relatively young animals (16–21 postnatal days), whereas the animals in the in vivo study were adults. It was shown previously that young corticostriatal synapses exhibit relatively high release probability, and that activity at corticostriatal synapses undergoes a developmentally regulated form of LTD in vivo (Choi & Lovinger, 1997b). Perhaps synapses from young animals are particularly susceptible to moderate or high frequency-stimulation-induced LTD because of their high probability of release. According to the sliding threshold model of synaptic plasticity, whether a synapse expresses LTD or LTP is dependent on previous activity at that synapse (see Abraham & Tate, 1997 for review). This may explain LTP induction in adult animals, whose corticostriatal synapses may have experienced relatively lower levels of input at a given synapse due to a lower probability of release. We suspect that prolonged moderate frequency firing could be sufficient to induce LTD in vivo, and there is evidence that striatal neurones can be activated at moderate frequencies for several minutes at a time (Charpier et al. 1999; Fellous et al. 2003). It is possible that the duration of moderate frequency synaptic activation necessary to induce LTD might be shorter in vivo, where it is likely that modulatory factors are present that are lost in the slice preparation. Ultimately, the type of synaptic plasticity induced by a given pattern of excitatory input will most likely depend on the activity of the postsynaptic neurone and other afferent inputs during the period of excitatory drive (see Lovinger et al. 2003 for discussion). More information on neuronal and afferent activities contributing to in vivo LTP and in vitro LTD will be needed to determine the conditions under which these two forms of plasticity occur.
, http://www.100md.com
Long-term changes in synaptic plasticity in the striatum are thought to be important for both motor and habit learning (Jog et al. 1999). The ways in which MSNs integrate these various inputs to regulate striatal output is only partially understood. The finding that prolonged activation of cortical afferents to the striatum induces a long-term decrease in synaptic efficacy provides additional evidence for activity-induced changes in synaptic efficacy in the striatum due to simultaneous activation of many inputs. It will be of interest to correlate changes in striatal MSN firing and synaptic efficacy with patterns of cortical input under different environmental and behavioural conditions. This may lead to a better understanding of the roles of synaptic potentiation and depression in learning memory and the shaping of habitual behaviour.
, http://www.100md.com
Conclusions
We have demonstrated that corticostriatal LTD can be induced by a prolonged moderate frequency stimulus train. This form of LTD is expressed presynaptically as a long-lasting decrease in EPSC amplitude that is due to a decrease in neurotransmitter release probability. LTD induced by 10 Hz stimulation requires the activation of both D2 dopamine receptors and CB1 cannabinoid receptors, but is insensitive to treatments that block HFS-induced LTD, including several manipulations that prevent postsynaptic activation and/or postsynaptic endocannabinoid release. It appears that extended activation of inputs to the striatum for minutes may provide the necessary signal to promote endocannabinoid synthesis and subsequent activation of presynaptic CB1 receptors.
, http://www.100md.com
References
Abraham WC & Tate WP (1997). Metaplasticity: a new vista across the field of synaptic plasticity. Prog Neurobiol 52, 303–323.
Alcantara AA, Chen V, Herring BE, Mendenhall JM & Berlanga ML (2003). Localization of dopamine D2 receptors on cholinergic interneurons of the dorsal striatum and nucleus accumbens of the rat. Brain Res 986, 22–29.
Cadas H, Gaillet S, Beltramo M, Venance L & Piomelli D (1996). Biosynthesis of an endogenous cannabinoid precursor in neurons and its control by calcium and cAMP. J Neurosci 16, 3934–3942.
, 百拇医药
Calabresi P, Maj R, Pisani A, Mercuri NB & Bernardi G (1992a). Long-term synaptic depression in the striatum: physiological and pharmacological characterization. J Neurosci 12, 4224–4233.
Calabresi P, Pisani A, Mercuri NB & Bernardi G (1992b). Long-term potentiation in the striatum is unmasked by removing the voltage-dependent magnesium block of NMDA receptor channels. Eur J Neurosci 4, 929–935.
Calabresi P, Pisani A, Mercuri NB & Bernardi G (1994). Post-receptor mechanisms underlying striatal long-term depression. J Neurosci 14, 4871–4881.
, 百拇医药
Calabresi P, Saiardi A, Pisani A, Baik J-H, Centonze D, Mercuri N, Bernardi G & Borrelli E (1997). Abnormal synaptic plasticity in the striatum of mice lacking dopamine D2 receptors. J Neurosci 17, 4536–4544.
Charpier S, Mahon S & Deniau JM (1999). In vivo induction of striatal long-term potentiation by low-frequency stimulation of the cerebral cortex. Neuroscience 91, 1209–1222.
Chevaleyre V & Castillo PE (2003). Heterosynaptic LTD of hippocampal GABAergic synapses. A novel role for endocannabinoids in regulating excitability. Neuron 38, 461–472.
, 百拇医药
Choi S & Lovinger DM (1997a). Decreased frequency but not amplitude of quantal synaptic responses associated with expression of corticostriatal long-term depression. J Neurosci 17, 8613–8620.
Choi S & Lovinger DM (1997b). Decreased probability of neurotransmitter release underlies striatal long-term depression and postnatal development of corticostriatal synapses. Proc Natl Acad Sci U S A 94, 2665–2670.
Fellous JM, Rudolph M, Destexhe A & Sejnowski TJ (2003). Synaptic background noise controls the input-output characteristics of single cells in an in vitro model of in vivo activity. Neuroscience 122, 811–829.
, 百拇医药
Freund TF, Katon I & Piomelli D (2003). Role of endogenous cannabinoids in synaptic signaling. Physiol Rev 83, 1017–1066.
Gerdeman GL & Lovinger DM (2001). CB1 cannabinoid receptor inhibits synaptic release of glutamate in rat dorsolateral striatum. J Neurophysiol 85, 468–471.
Gerdeman GL & Lovinger DM (2003). Emerging roles for enodcannabinoids in long-term synaptic depression. Br J Pharmacol 140, 781–789.
, 百拇医药
Gerdeman GL, Ronesi J & Lovinger DM (2002). Postsynaptic endocannabinoid release is critical to long-term depression in the striatum. Nat Neurosci 5, 446–451.
Gerfen CR, Engber TM, Mahan LC, Susel Z, Chase TN, Monsma FJ Jr & Sibley DR (1990). D1 and D2 dopamine receptor-regulated gene expression of striatonigral and striatopallidal neurons. Science 250, 1429–1432.
Giuffrida A, Parsons LH, Kerr TM, Rodriguez de Fonseca F, Navarro M & Piomelli D (1999). Dopamine activation of endogenous cannabinoid signaling in dorsal striatum. Nat Neurosci 2, 358–363.
, http://www.100md.com
Gubellini P, Saulle E, Centonze D, Bonsi P, Pisani A, Bernardi G, Conquet F & Calabresi P (2001). Selective involvement of mGlu1 receptors in corticostriatal LTD. Neuropharmacology 40, 836–846.
Hoffman AF, Oz M, Caulder T & Lupica CR (2003). Functional tolerance and blockade of long-term depression at synapses in the nucleus accumbens after chronic cannabinoid exposure. J Neurosci 23, 4815–4820.
Jog MS, Kubota Y, Connolly CI, Hillegaart V & Graybiel AM (1999). Building neural representations of habits. Science 286, 1745–1749.
, http://www.100md.com
Kasper EM, Larkman AU, Lubke J & Blakemore C (1994). Pyramidal neurons in layer 5 of the rat visual cortex II. Development of electrophysiological properties. J Comp Neurol 339, 475–494.
Kerr JN & Wickens JR (2001). Dopamine D-1/D-5 receptor activation is required for long-term potentiation in the rat neostriatum in vitro. J Neurophysiol 85, 117–124.
Kim J & Alger BE (2001). Random response fluctuations lead to spurious paired-pulse facilitation. J Neurosci 21, 9608–9618.
, 百拇医药
Lovinger DM, Partridge JG & Tang KC (2003). Plastic control of striatal glutamatergic transmission by ensemble actions of several neurotransmitters and targets for drugs of abuse. Ann N Y Acad Sci 1003, 226–240.
Lovinger DM, Tyler EC & Merrit A (1993). Short and long-term synaptic plasticity in the neostriatum. J Neurophysiol 70, 1937–1949.
Partridge JG, Apparsundaram S, Gerhardt GA, Ronesi J & Lovinger DM (2002). Nicotinic acetylcholine receptors interact with dopamine in induction of striatal long-term depression. J Neurosci 22, 2541–2549.
, http://www.100md.com
Partridge JG, Tang KC & Lovinger DM (2000). Regional and postnatal heterogeneity of activity-dependent long-term changes in synaptic efficacy in the dorsal striatum. J Neurophysiol 84, 1422–1429.
Picconi B, Centonze D, Hakansson K, Bernardi G, Greengard P, Fisone G, Cenci MA & Calabresi P (2003). Loss of bidirectional striatal synaptic plasticity in L-DOPA-induced dyskinesia. Nat Neurosci 6, 501–506.
Redman S (1990). Quantal analysis of synaptic potentials in neurons of the central nervous system. Physiol Rev 70, 165–198.
, http://www.100md.com
Robbe D, Kopf M, Remaury A, Bockaert J & Manzoni OJ (2002). Endogenous cannabinoids mediate long-term synaptic depression in the nucleus accumbens. Proc Natl Acad Sci U S A 99, 8384–8388.
Ronesi J, Gerdeman GL & Lovinger DM (2004). Disruption of endocannabinoid release and striatal long-term depression by postsynaptic blockade of endocannabinoid membrane transport. J Neurosci 24, 1673–1679.
Schmitz Y, Schmauss C & Sulzer D (2002). Altered dopamine release and uptake kinetics in mice lacking D2 receptors. J Neurosci 15, 8002–8009.
, 百拇医药
Sesack SR, Aoki C & Pickel VM (1994). Ultrastructural localization of D2 receptor-like immunoreactivity in midbrain dopamine neurons and their striatal targets. J Neurosci 14, 88–106.
Smith R, Musleh W, Akopian G, Buckwalter G & Walsh JP (2001). Regional differences in the expression of corticostriatal synaptic plasticity. Neuroscience 106, 95–101.
Starke K, Gothert M & Kilbinger H (1989). Modulation of neurotransmitter release by presynaptic autoreceptors. Physiol Rev 69, 864–989.
, http://www.100md.com
Stella N, Schweitzer P & Piomelli D (1997). A second endogenous cannabinoid that modulates long-term potentiation. Nature 388, 773–778.
Sung KW, Choi S & Lovinger DM (2001). Activation of group I mGluRs is necessary for induction of long-term depression at striatal synapses. J Neurophysiol 86, 2405–2412.
Tang KC & Lovinger DM (2000). Role of pertussis toxin-sensitive G-proteins in synaptic transmission and plasticity at corticostriatal synapses. J Neurophysiol 83, 60–69.
, 百拇医药
Tang K, Low MJ, Grandy DK & Lovinger DM (2001). Dopamine-dependent synaptic plasticity in striatum during in vivo development. Proc Natl Acad Sci U S A 98, 1255–1260.
Wang H & Pickel VM (2002). Dopamine D2 receptors are present in prefrontal cortical afferents and their targets in patches of the rat caudate-putamen nucleus. J Comp Neurol 442, 392–404.
Zhou FM, Liang Y & Dani JA (2001). Endogenous nicotinic cholinergic activity regulates dopamine release in the striatum. Nat Neurosci 4, 1224–1229., 百拇医药(Jennifer Ronesi, and Davi)
2 Department of Pharmacology, Vanderbilt University School of Medicine, Nashville, TN 37232, USA
Abstract
The striatum regulates motor output, and it is thought that changes in the synaptic efficacy of inputs to the striatum contribute to motor learning and habit formation. Previously, several laboratories have observed that brief high frequency stimulation (HFS) of cortical afferents innervating the dorsolateral striatum induces a long-term decrease in synaptic efficacy called long-term depression (LTD). We recently showed that HFS-induced striatal LTD requires retrograde signalling involving postsynaptic release of endocannabinoids and activation of presynaptic CB1 cannabinoid receptors. In the present study we have employed whole-cell recording in brain slices to examine a new form of LTD at corticostriatal synapses that can be induced by a 10 Hz, 5 min train. The decrease in synaptic efficacy is associated with a decrease in presynaptic release probability, as demonstrated by a decrease in frequency but not amplitude of spontaneous excitatory postsynaptic currents (sEPSCs) and an increase in the paired pulse ratio (PPR). This form of LTD is blocked by antagonists for CB1 and D2 dopamine receptors and impaired by blockers of L-type calcium channels. However, 10 Hz-induced LTD does not depend on postsynaptic depolarization, unlike HFS-induced LTD. Furthermore, this new form of LTD is not prevented by treatments known to block HFS-induced LTD, including antagonism of metabotropic glutamate receptors (mGluRs), chelation of postsynaptic calcium, or intracellular application of an anandamide membrane transport inhibitor (VDM11). From these findings it is not clear that the endocannabinoid responsible for this form of LTD acts in a retrograde fashion, and the cellular source of endocannabinoid necessary for 10 Hz-induced LTD is as yet unknown. Our results demonstrate that a prolonged moderate frequency train induces cannabinoid-dependent LTD, further supporting the idea that endocannabinoids play a prominent role in the regulation of long-lasting changes in striatal output.
, 百拇医药
Introduction
Information storage at central synapses is widely attributed to long-lasting activity-dependent changes in synaptic efficacy. We have been investigating such changes in the striatum, a brain region with important roles in motor learning and habit formation. High frequency activation of inputs to the striatum can induce either long-lasting increases or decreases in transmission at excitatory synapses onto striatal medium spiny neurones (MSNs) (Calabresi et al. 1992a,b; Lovinger et al. 1993; Partridge et al. 2000; Kerr & Wickens, 2001). The long-lasting decrease in efficacy, long-term depression or LTD, is especially prominent in the dorsolateral striatum (Partridge et al. 2000; Smith et al. 2001). Expression of high frequency stimulation (HFS)-induced striatal LTD (HFS-LTD) is presynaptic and due to a decrease in the probability of glutamate release, as shown by a decrease in the frequency but not amplitude of spontaneous EPSCs (Choi & Lovinger, 1997a) and an increase in the paired pulse ratio (PPR) (Choi & Lovinger, 1997b). However, HFS-LTD induction requires postsynaptic activation. Calcium chelators (i.e. 10 mM EGTA) introduced into the postsynaptic neurone block LTD induction (Choi & Lovinger, 1997b). In addition, HFS-LTD induction requires convergent activation of group I metabotropic glutamate receptors (mGluRs) (Sung et al. 2001; Gubellini et al. 2001), D2 dopamine receptors (D2Rs) (Calabresi et al. 1997; Tang et al. 2001), and L-type voltage-gated calcium channels (Calabresi et al. 1994; Choi & Lovinger, 1997b). The decrease in presynaptic release probability and the dependence on postsynaptic activation demonstrates the necessity for a retrograde messenger to explain HFS-LTD (Choi & Lovinger, 1997b). Indeed, HFS-LTD requires the activation of a presynaptic CB1 cannabinoid receptor, presumably via postsynaptic calcium-dependent synthesis and release of endocannabinoids (Gerdeman et al. 2002).
, http://www.100md.com
The MSNs of the striatum receive convergent glutamatergic inputs from the cortex and thalamus and dopaminergic inputs from the substantia nigra pars compacta. Simultaneous activation of many cortical afferents is necessary to depolarize MSNs to action potential firing threshold. The HFS protocol used to induce LTD mimics firing patterns found in deep-layer cortical pyramidal neurones that innervate the striatum. These neurones can fire action potentials at 100 Hz or more for up to 3 s (Kasper et al. 1994). However, other firing patterns of deep-layer cortical neurones innervating the dorsal striatum have been reported. In vivo, moderate frequency membrane potential oscillations (5 Hz) were recorded in MSNs that were due to the synchronous firing of cortical inputs (Charpier et al. 1999). In addition, in vitro recordings from layer 5 rat prefrontal cortical neurones demonstrated firing at irregular frequencies with an average of 5–10 Hz (Fellous et al. 2003). Given the variable frequencies of in vivo spike trains at the corticostriatal synapse, it is reasonable to speculate that an alternate frequency train may produce plasticity, perhaps including LTD. Recently, a moderate frequency stimulation (13 Hz, 10 min or 10 Hz, 5 min)-induced form of LTD at glutamatergic synapses in the nucleus accumbens was shown to be CB1-dependent (Robbe et al. 2002, Hoffman et al. 2003). Although induction of this form of LTD is dependent on the activation of postsynaptic group I mGluRs and calcium release from internal stores, it is expressed presynaptically. Taken together, these findings led us to speculate that stimuli delivered at the moderate frequency of 10 Hz might induce LTD in the dorsal striatum. In the present study we examined changes in the efficacy of glutamatergic synapses in response to afferent stimulation at this moderate frequency in corticostriatal brain slices from rat.
, http://www.100md.com
Methods
Brain slice preparation
Brain slices were prepared as described previously (Ronesi et al. 2004) from Sprague-Dawley rats aged 16–21 postnatal days (Charles River Laboratories, Wilmington, MA, USA). Animals were anaesthetized with halothane (Sigma, St Louis, MO, USA) and killed by decapitation, in accordance with the National Institutes of Health animal welfare guidelines. The brains were removed and transferred to an ice-cold modified artificial cerebrospinal fluid (aCSF) containing (mM): 194 sucrose, 30 NaCl, 4.5 KCl, 1 MgCl2, 26 NaHCO3, 1.2 NaH2PO4, 10 D-glucose, and brought to pH 7.4 by bubbling with 95% O2–5% CO2. Coronal slices of 350 μm thickness were made in ice-cold modified aCSF using a manual vibroslice (World Precision Instruments, Sarasota, FL, USA). Slices were transferred immediately to a nylon net that was submerged in normal aCSF containing (mM): 124 NaCl, 4.5 KCl, 2 CaCl2, 1 MgCl2, 26 NaHCO3, 1.2 NaH2PO4, and 10 D-glucose. ACSF was maintained at a pH of 7.4 and osmolarity was 325 mosmol l–1. Slices were incubated for at least 30 min at room temperature (19–21°C) prior to the start of recording. Hemislices containing both the cortex and the striatum were then transferred to the recording chamber, submerged in normal ACSF containing, in the majority of experiments, 25 μM picrotoxin (to prevent GABAA receptor-mediated synaptic responses). During electrophysiological recordings, slices were maintained at a temperature between 29 and 32°C, stable within 1°C during any given experiment.
, 百拇医药
Whole-cell voltage-clamp recording
Whole-cell recordings were performed as described previously (Ronesi et al. 2004). Pipettes (2.5–5.0 M) made from borosilicate glass capillaries were pulled on a Flaming-Brown micropipette puller (Novato, CA, USA) and filled with an internal solution containing the following (mM): 120 caesium methane sulphonate, 5 NaCl, 10 tetraethylammonium chloride, 10 Hepes, 4 lidocaine N-ethyl bromide, 1.1 EGTA, 4 Mg-ATP, and 0.3 Na-GTP, pH 7.2, and osmolarity adjusted to 298 mosmol l–1. In some experiments, BAPTA (20 mM, Sigma) was substituted for EGTA in the internal solution. Access to medium spiny neurones 3–4 layers below the slice surface was aided with a differential interference contrast (DIC)-enhanced visual guidance system. Cells were voltage clamped at –70 mV and maintained at this potential prior to, during, and subsequent to the moderate frequency stimulus train. Afferent stimuli were delivered via a master-8 stimulator (A.M.P.I., Jerusalem, Israel) and a bipolar-twisted tungsten wire that was placed in the white matter dorsal to the lateral region of the striatum, and test stimuli before and after the 10 Hz train were given at a frequency of 0.05 Hz. The stimulus intensity was set to yield an excitatory postsynaptic current (EPSC) amplitude between 200 and 400 pA. Series resistance was < 10 M and was uncompensated. Both series and input resistances were monitored throughout the experiment by delivering 5 mV voltage steps. If series resistance changed more than 20% during the course of an experiment, the cell was discarded.
, 百拇医药
Synaptic currents were recorded with either an Axopatch 1D or Axopatch 200A amplifier (Axon Instruments, Foster City, CA, USA), filtered at 5 kHz, digitized at 10 kHz, and stored on a microcomputer. EPSC amplitudes were examined using peak detection software in pCLAMP8 (Axon Instruments, Union City, CA, USA). The moderate frequency stimulation protocol consisted of a 10 Hz, 5 min train of stimuli at the same intensity used for pre- and post-train test pulses. For some experimental conditions, a high frequency train was given. This HFS protocol consisted of four 100 Hz trains (1 s duration) with an intertrain interval of 10 s, where each train was paired with a depolarizing step to 0 mV in the postsynaptic neurone. HFS experiments were performed to provide evidence that drugs known to block HFS-induced LTD that had no effect on 10 Hz-induced LTD (10 Hz-LTD) were actually penetrating slices sufficiently to affect some aspect of synaptic transmission. Effects on HFS-induced LTD were only examined in 2–3 cells per treatment group in the present study because results were similar to those observed in previous studies with larger numbers of cells (Choi & Lovinger 1997b; Tang et al. 2001; Sung et al. 2001; Ronesi et al. 2004). The EPSC amplitude of the response over the 5 min prior to the 10 Hz train was averaged and compared to the average EPSC amplitude of the response 20–30 min post train. LTD was defined as a change in mean EPSC amplitude that was greater than two standard deviations from the baseline EPSC mean. In figures depicting the time course of an LTD experiment, data points were plotted as means ± S.E.M. normalized to baseline. The paired pulse ratio (PPR, interval of 50 ms) was calculated as the ratio of the mean amplitude of the second response during a 10 min time period over the mean amplitude of the first response during the same period (Kim & Alger, 2001). However, for the points plotted in Fig. 1B, the PPR was calculated for each time point in order to show the time course of the PPR change following the train. Mean PPR values 10 min prior to the train were compared to PPR values 20–30 min post train using an unpaired t test, where P < 0.05 was considered statistically significant. To determine if the magnitude of LTD was impaired under certain pharmacological conditions, an unpaired t test was used (significance at P < 0.05) to compare the EPSC amplitude at 20–30 min post train between untreated and treated cells. Spontaneous release events were recorded for at least 2 min beginning 5 min before the 10 Hz train and 30 min after the train. Data were visualized both on an oscilloscope and in the Clampex program, and data were stored directly in Clampex using a free-running data collection protocol. During the collection of sEPSC data, current records were amplified at a gain of 2 mV pA–1, filtered at 5 kHz, and digitized at 10 kHz. Quantal events were detected and analysed offline using the Mini Analysis program by Synaptosoft, version 5.6 (Decatur, GA, USA). The current records were visually inspected following automated analysis to prevent false positive identification and false negative rejection of sEPSCs. Peak amplitude, interevent interval, rise time, and decay time were measured. The threshold for peak amplitude detection (typically 7 pA) was twice the baseline root mean square noise level. Data for cumulative EPSC histograms were compared statistically using the Kolmogorov-Smirnov test, where the statistical criterion for significance was P < 0.005.
, http://www.100md.com
A, MSNs in intact coronal brain slices containing the cortex and striatum were voltage clamped at –70 mV. Time course of LTD during and after a moderate frequency train (train indicated by arrow at time = 0). After at least 10 min baseline, 10 Hz, 5 min stimulation of cortical afferents produced a long-lasting decrease in EPSC amplitude recorded from striatal MSNs. Dotted line represents baseline. Standard error bars are smaller than symbols. Sample traces are averages of 15 evoked EPSCs immediately before (left) and 25 min after (right) the train. Scale bar: 100 pA, 25 ms. B, representative sample traces of evoked EPSCs during start (first 900 ms) and end (last 900 ms) of a 10 Hz, 5 min train. Note that there is no change in baseline current, but a marked decrease in EPSC amplitude. Scale bar: 100 pA, 100 ms. C, representative experiment showing that 10 Hz-induced LTD does not occlude HFS-LTD. D, time course of PPR during and after the 10 Hz stimulation train. Note the lasting increase in PPR.
, 百拇医药
Pharmacology
AM251, sulpiride, CPCCOEt, LY341495 and VDM11 were obtained from Tocris-Cookson (St Louis, MO, USA) and nifedipine was obtained from Calbiochem (San Diego, CA, USA). AM251, sulpiride, and LY341495 were dissolved in DMSO at 10 mM, nifedipine was dissolved in DMSO at 20 mM, CPCCOEt was dissolved in DMSO at 80 mM, and VDM11 was dissolved in DMSO at 50 mM. The final concentration of DMSO in the extracellular solution never exceeded 0.1%, and the concentration of DMSO in the internal solution was 0.02% in internal VDM11 experiments. Picrotoxin and D-aminophosphonovalerate (APV) were purchased from Sigma and dissolved directly into the aCSF. For AM251 wash-on experiments, 0.5 mg ml–1 bovine serum albumin was included in the external solution to prevent this drug from adhering to the perfusion apparatus. aCSF and drugs were bath applied using gravity flow at a rate of 1–2 ml min–1.
, 百拇医药
Results
Induction of striatal LTD with a 10 Hz, 5 min train
During whole-cell recordings from MSNs in coronal slices containing the cortex and striatum, 200–400 pA EPSCs were evoked by stimulation at 0.05 Hz with a stimulating electrode placed in the white matter overlying the striatum. All of the recordings were performed in the dorsolateral striatum, where it was previously shown that LTD is the predominant form of HFS-induced synaptic plasticity (Partridge et al. 2000; Smith et al. 2001). Activating cortical afferents to the dorsolateral striatum at 10 Hz for 5 min induced a long-lasting decrease in EPSC amplitude (Fig. 1A). LTD was defined as a persistent decrease in EPSC amplitude (at 20–30 min post train) that was greater than two standard deviations less than the baseline EPSC amplitude mean. In 21 of 24 cells subjected to the 10 Hz stimulation protocol this criterion was met, and the average post-train EPSC amplitude in these cells was 61 ± 3% of the mean baseline response (Fig. 1A). During the course of the 5 min train, the amplitude of evoked EPSCs was diminished and in some cases the EPSC was completely absent in response to the last few stimuli of the train. Individual EPSCs decayed completely back to baseline current levels between stimuli, and no detectable change in net baseline current developed over the course of the 5 min stimulation (Fig. 1B).
, 百拇医药
To determine if LTD induced by 10 Hz stimulation and HFS could be observed in the same neurone and to determine if we could observe evidence of saturation of LTD expression mechanisms we delivered a 10 Hz stimulus train, which was followed 30 min later by four HFS trains combined with postsynaptic depolarization (Fig. 1C). The decrease in EPSC amplitude was robust in response to both stimulus trains. The mean EPSC amplitude after the 10 Hz train was 58 ± 7% of baseline and was 44 ± 2% of baseline following the HFS train (n = 4). When an HFS train was given first followed by a 10 Hz train, a similar result was produced. The mean EPSC amplitude was 57% of baseline following the HFS train and was 36% of baseline following the 10 Hz train (n = 2). Interestingly, giving an LFS train followed 30 min later by another LFS train did not appear to produce saturation of LTD. The mean EPSC amplitude 20–30 min after the first 10 Hz train was 67 ± 5% of baseline and 45 ± 6% of baseline following the second train (n = 4). Moreover, delivery of two consecutive HFS trains induced LTD, as the mean EPSC amplitude following the first HFS train was 73% of baseline and was 48% of baseline following the second train (n = 2). Thus, we did not observe evidence of LTD saturation with any stimulus protocol. Unfortunately, we were not able to maintain stable recordings for a period long enough to determine if LTD was saturated with more than two stimulus trains, and it was not possible to determine if the two forms of LTD could truly occlude one another. However, our findings do indicate that both 10 Hz and HFS-induced LTD can occur at synapses onto the same neurones.
, 百拇医药
Picrotoxin (25 μM), a GABAA receptor antagonist, was included in the extracellular solution in order to isolate pure EPSCs, since GABAergic interneurones and local axon collaterals are activated by afferent stimulation. In the absence of picrotoxin, 10 Hz-induced striatal LTD was unimpaired, demonstrating that GABAergic transmission does not play an appreciable role in this form of LTD. LTD was observed in 7 of 8 neurones, and had a typical mean LTD magnitude of 59 ± 5% of baseline (P > 0.05, unpaired t test). Therefore, the data obtained from experiments performed in the absence of picrotoxin were included with those from experiments performed in the presence of picrotoxin. All other experiments described in this study were done with picrotoxin in the extracellular solution.
, 百拇医药
Presynaptic expression of 10 Hz-induced striatal LTD
To determine if striatal LTD induced by 10 Hz stimulation is associated with changes in the ratio of responses to paired stimuli, an index of relative neurotransmitter release probability, we examined the amplitude of EPSCs evoked by two stimuli of equal magnitude delivered 50 ms apart. The ratio of the amplitudes of the second evoked EPSC relative to the first (PPR ratio) averaged 1.09 ± 0.05 in the pre-train baseline condition. In the 21 of 24 neurones that demonstrated LTD as defined by the above criterion, PPR increased to 122 ± 3% of the baseline value 20–30 min following the 10 Hz stimulus train (Fig. 1D).
, 百拇医药
We also measured the frequency, amplitude and time course of spontaneous EPSCs (sEPSCs) to determine the possible pre- or postsynaptic locus of the change in synaptic efficacy (Redman, 1990). It should be noted that the frequency and amplitude of sEPSCs recorded in this corticostriatal slice preparation are maintained even in the presence of sodium channel blockade (Tang & Lovinger, 2000), and thus more than likely represent quantal, miniature EPSCs. Due to difficulties in maintaining stable recordings and to a low frequency of spontaneous events in many cells, only a small number of neurones were examined for 10 Hz-induced changes in sEPSC frequency and/or baseline. Representative traces of sEPSCs recorded before and after the train are shown in Fig. 2A. In the pre-train baseline period, sEPSC frequency averaged 1.5 ± 0.6 Hz (n = 3). Following the 10 Hz stimulus train, the frequency of sEPSCs dropped to 42 ± 6% of baseline (Fig. 2B, n = 3), and all three cells exhibited a significant increase in interevent interval, as can be seen in the cumulative probability distributions shown in Fig. 2D (P < 0.005, Kolomogorov-Smirnov test). In contrast, the amplitude of sEPSCs was unchanged. The median event amplitude was 17 ± 1 pA prior to 10 Hz stimulation, and remained essentially the same (101 ± 10% of baseline) following the stimulus train (Fig. 2B, n = 3). All three neurones exhibited no change in amplitude distribution (Fig. 2C; P > 0.005, Kolmogorov-Smirnov test). The 10–90% rise times (Fig. 2B, 103 ± 7% of baseline) and half-decay times (Fig. 2B, 117 ± 5% of baseline) of sEPSCs did not change following the 10 Hz train (P > 0.05). When stimulation was maintained at 0.05 Hz throughout the experiment there were no apparent changes in the frequency or amplitude of sEPSCs, demonstrating that the decrease in sEPSC frequency was not simply due to drift over time during low frequency (0.05 Hz) stimulation of cortical inputs. Under this condition, the frequency of sEPSCs was 101 ± 13% baseline and the amplitude was 85 ± 3% baseline (n = 3).
, http://www.100md.com
A, representative sample traces of spontaneous EPSCs recorded 5 min before (pre-LTD) and 30 min after (post-LTD) a 10 Hz train. Scale bars: 25 pA, 1 s. B, bar graph showing mean percentage change relative to baseline in frequency, amplitude, 10–90% rise time, and decay time of sEPSCs (n = 3). Twenty to thirty minutes after the train, there is a decrease in spontaneous EPSC frequency but not amplitude, rise time, or decay time. C, averaged (n = 3) cumulative probability amplitude distribution of sEPSCs before () and after () a 10 Hz train. D, cumulative interevent interval distribution of sEPSCs before and after a 10 Hz train.
, http://www.100md.com
We wanted to determine if shorter duration 10 Hz trains could induce LTD. Activating corticostriatal synapses for 1 min at 10 Hz did not induce LTD. The mean EPSC amplitude 20–30 min after the train was 92 ± 4% of baseline and the mean PPR was 110 ± 4% of baseline (n = 4). Likewise, a 2 min, 10 Hz train did not induce LTD. The EPSC amplitude was 87 ± 5% of baseline and the PPR was 101 ± 5% of baseline (n = 3). Interestingly, the EPSC amplitude dropped to 7 ± 3% (n = 5) of the value of the start of the train within the first minute of the onset of 10 Hz stimulation. Thus, loss of the EPSC during the stimulus train does not appear to signal the onset of LTD, as several more minutes of 10 Hz stimulation are needed for LTD induction.
, http://www.100md.com
LTD requires activation of CB1 cannabinoid receptors and D2 dopamine receptors
Slices that were treated throughout the duration of the experiment with the CB1 antagonist AM251 (2–4 μM) did not exhibit a long-lasting decrease in EPSC amplitude in the majority of neurones examined (Fig. 3A, P > 0.05, unpaired t test). In 7 of 9 cells receiving 10 Hz stimulation, the mean EPSC amplitude 20–30 min after the train was 103 ± 12%. The PPR was also unchanged and was 106 ± 7%. In the remaining two neurones, the EPSC amplitude was decreased to 77 and 61% of baseline, and PPR was increased by 115 and 130%, respectively. Application of AM251 10 min after the termination of the train did not affect the maintenance of LTD. The mean EPSC amplitude was 64% of baseline and the mean PPR was 122% of baseline (n = 2, Fig. 3A).
, 百拇医药
A, slices treated with the CB1 antagonist AM251 (2–4 μM) prior to the train failed to exhibit LTD, whereas bath application of AM251 (2 μM) 10 min after the train did not affect LTD amplitude (train indicated by arrow at time = 0). Dotted line represents baseline and bar indicates time course of drug perfusion. B, slices treated with the antagonist sulpiride (2–10 μM) failed to exhibit LTD, whereas bath application of sulpiride (4 μM) 10 min after the train did not affect LTD amplitude. Bar denotes time course of drug perfusion. Above, 15 trace averages of EPSCs recorded before (left) and 30 min after (right) 10 Hz train (only pre-train drug condition shown). Scale bars: 100 pA, 25 ms.
, 百拇医药
Bath application of the D2 antagonist sulpiride (2–10 μM) during the train, and 10 min before and after the train, blocked 10 Hz-induced LTD in 5 of 5 cells (Fig. 3B, P > 0.05, unpaired t test). The percentage of baseline EPSC amplitude 20–30 min after the train was 92 ± 7. The PPR was also unchanged and averaged 105 ± 4%. As reported previously, bath application of the D2 antagonist sulpiride for the duration of the experiment also blocked HFS-induced striatal LTD. At 20–30 min after the train, the EPSC amplitude was 95% of baseline (n = 2). Sulpiride alone had no effect on baseline transmission (data not shown, Tang et al. 2001). When sulpiride was applied 10 min after the termination of the train, there was no change in the magnitude of LTD (P > 0.05, unpaired t test). The mean EPSC amplitude 20–30 min after the train was 54 ± 3% of baseline and the mean PPR was 127 ± 3% of baseline (n = 3, Fig. 3B).
, 百拇医药
Differential induction mechanisms of 10 Hz- and HFS-induced LTD
To determine if the activation of group I mGluRs is necessary for the induction of 10 Hz-induced LTD, slices were treated with the non-competitive antagonist CPCCOEt (80 μM) during, and 10 min before and after, the train (Fig. 4A). We observed a typical decrease in EPSC amplitude in 5 of 6 cells to 61 ± 2% of baseline value 20–30 min following the 10 Hz train (P > 0.05, unpaired t test). As in control cells, this decrease in EPSC amplitude was associated with an increase in the PPR. The PPR was 121 ± 3% of baseline. Bath application of CPCCOEt had no effect on baseline EPSC amplitude (data not shown, Sung et al. 2001). Also, extracellular application of the mGluR antagonist LY341495, at a concentration (10 μM) that not only blocks group II mGluRs, but is also non-selective with respect to both group I mGluRs, had no effect on the magnitude of 10 Hz-LTD (Fig. 4B, n = 4; EPSC, 64 ± 4% of baseline; PPR, 124 ± 9% of baseline, P > 0.05, unpaired t test).
, 百拇医药
A and B, slices pretreated with the mGluR1 antagonist CPCCOEt (80 μM) or the broad-spectrum mGluR antagonist LY341495 (10 μM) exhibit LTD with a magnitude similar to that observed in control slices (train indicated by arrow at time = 0). C, slices treated with the L-type calcium channel blocker nifedipine (20 μM) exhibit LTD, but the magnitude of the decrease in EPSC amplitude is smaller than in control slices (train given at time = 0). Dotted line represents baseline and bar denotes time course of drug perfusion.
, http://www.100md.com
The activation of NMDA receptors also does not appear to be crucial for 10 Hz-LTD. When the NMDA receptor antagonist APV (100 μM) was applied extracellularly, the expression of LTD was unimpaired (P > 0.05, unpaired t test). At 20–30 min post train, the average EPSC amplitude was 54 ± 1% of baseline and the PPR was 121 ± 4% of baseline (n = 4). However, bath application of the L-type calcium channel blocker nifedipine (20 μM) significantly impaired the magnitude of LTD (Fig. 4C, n = 6, P < 0.01, unpaired t test) and was not associated with a robust increase in the PPR. The average EPSC amplitude was 74 ± 3% of baseline and the average PPR was 108 ± 3% of baseline.
, 百拇医药
Including the calcium chelator BAPTA (20 mM) in the postsynaptic neurone via the patch pipette failed to block 10 Hz-induced LTD in the majority of neurones examined. Following the train, 5 of 6 neurones exhibited a decrease in EPSC amplitude 20–30 min after the train that was greater than two standard deviations below the baseline mean (Fig. 5A, P > 0.05, unpaired t test). The percentage of baseline EPSC amplitude was 53 ± 8 and the percentage of baseline PPR was 131 ± 4. As previously reported, chelating calcium postsynaptically prevented HFS-induced LTD in three neurones. The mean post-train EPSC amplitude was 85 ± 7% of baseline in these neurones. Intracellular BAPTA has no apparent effect on baseline synaptic efficacy, as the stimulus magnitudes required to elicit 200–400 pA currents (0.4–1.0 mA, 40–50 μs duration) were similar to those used to evoke EPSCs under control conditions (0.3–1.0 mA, 40–60 μs duration).
, 百拇医药
A, cells loaded intracellularly with the calcium chelator BAPTA (20 mM) still exhibit 10 Hz-induced striatal LTD (train indicated by arrow at time = 0). B, the AMT inhibitor VDM11 (10 μM) was loaded into the postsynaptic neurone. In 6 of 9 cells LTD was unimpaired, whereas LTD was reduced in 3 of 9 cells. Mean ± S.E.M. normalized EPSC values from all 9 neurones are plotted as a function of time in the graph. Dotted line represents baseline.
HFS-LTD is blocked when inhibitors of the anandamide membrane transporter (AMT) are included in the postsynaptic neurone, presumably by preventing endocannabinoid release (Ronesi et al. 2004). Therefore, the AMT inhibitor VDM11 (10 μM) was included in the patch pipette in an attempt to block 10 Hz-induced striatal LTD (Fig. 5B). Recordings from nine neurones showed that LTD was not significantly decreased in the presence of postsynaptic intracellular VDM11 (P = 0.06, unpaired t test). The mean EPSC amplitude 20–30 min following the train was 70 ± 6% of baseline and the mean PPR was 115 ± 8% of baseline. The decrease in magnitude largely reflects the fact that three neurones exhibited decreased EPSCs averaging 85 ± 7% of baseline. The other neurones in this experiment exhibited robust LTD (62 ± 5% of baseline). In experiments done within the same batches of slices used for the 10 Hz-LTD experiments, VDM11 included internally blocked HFS-LTD, and the EPSC amplitude 20–30 min after HFS was 104% of baseline (n = 2). Like BAPTA, intracellular VDM11 did not appear to effect baseline synaptic efficacy (stimulus magnitude was 0.2–0.9 mA, 40–60 μs duration).
, 百拇医药
Discussion
Although it is well-established that LTD can be induced by HFS of afferents to the dorsolateral striatum, it was not clear if LTD could be induced by afferent activation at other frequencies within the naturally occurring range of cortical neuronal firing. Here we demonstrate that LTD at corticostriatal synapses can be induced by afferent stimulation at a moderate frequency (10 Hz). This form of LTD requires the activation of CB1 cannabinoid and D2 dopamine receptors, and analysis of PPR and sEPSCs indicates that it is expressed as a change in presynaptic function, possibly as a decrease in the probability of glutamate release. This expression mechanism strongly resembles that of HFS-LTD at the same synapse.
, 百拇医药
Endocannabinoids and dopamine mediate synaptic plasticity in the dorsal striatum
In central nervous system neurones, cannabinoids and endocannabinoids work mainly through activation of the CB1 receptor. This receptor is expressed primarily presynaptically throughout the brain, and densely in the forebrain, where its activation is coupled to a decrease in neurotransmitter release (see Freund et al. 2003 for review). At the corticostriatal synapse, activation of CB1 with various synthetic agonists reduces EPSC amplitude by decreasing glutamate release probability (Gerdeman & Lovinger, 2001). Endocannabinoids (e.g. 2-arachidonylglycerol and anandamide) are endogenous ligands for cannabinoid receptors and are synthesized de novo from lipid precursors embedded in the membrane (Cadas et al. 1996; Stella et al. 1997). Endocannabinoid synthesis is inducible, and it is thought that following prolonged postsynaptic depolarization and/or activation of presynaptic inputs, endocannabinoids are made and immediately released. Endocannabinoids have been shown to be necessary for several forms of plasticity, including LTD in the nucleus accumbens and HFS-LTD in dorsal striatum (Gerdeman & Lovinger, 2003).
, 百拇医药
Like HFS-LTD, 10 Hz-induced LTD requires activation of the D2 dopamine receptor. Previous studies have demonstrated dopamine release in response to stimulation in striatal slices, even at very low stimulus frequencies (Zhou et al. 2001; Schmitz et al. 2002). We have observed transient dopamine release in response to single stimuli delivered using the same electrode placement and stimulus parameters used in the present study (Partridge et al. 2002). In vivo microdialysis experiments revealed that activation of D2 increases anandamide levels 8-fold within the striatum (Giuffrida et al. 1999). Therefore, D2 activation may increase endocannabinoid levels in the striatum and promote subsequent CB1 activation. The role of dopamine and D2 receptors in LTD within the dorsal striatum is different from other forms of endocannabinoid-dependent LTD that do not appear to require D2 (Robbe et al. 2002; Chevaleyre & Castillo, 2003). The reasons for this difference are not yet known, but the dorsal striatum has the highest level of dopaminergic innervation in the brain, and it is only in this region that D2 activation has been linked to endocannabinoid synthesis/release. However, continued activation of either CB1 or D2 receptors is not necessary for maintenance of 10 Hz-LTD, as the application of antagonists for these receptors 10 min after the induction of LTD does not affect expression.
, http://www.100md.com
It is possible that local depolarization at distal dendritic sites might contribute to 10 Hz-LTD induction. However, several observations suggest that this is not the case. First, no net inward current was observed during the 10 Hz stimulus train, indicating that no sizeable current is generated anywhere in the cell. We believe a standing inward current of more than a few picoamps conducted from the dendrites to the soma should be readily observed, because we can easily detect sEPSCs. Second, previous studies demonstrated that HFS-LTD is blocked by holding the neuronal membrane potential at –70 mV (Choi & Lovinger, 1997b), which prevents depolarization to a level needed for LTD induction even in the face of much higher frequency input.
, 百拇医药
10 Hz-LTD versus HFS-LTD
Interestingly, we noted several differences between the induction mechanism of 10 Hz-LTD and HFS-LTD. LTD induced by brief HFS trains requires postsynaptic depolarization, activation of L-type calcium channels, increased postsynaptic calcium concentration, and activation of group I mGluRs (Calabresi et al. 1994; Choi & Lovinger, 1997b; Gubellini et al. 2001; Sung et al. 2001). The observations made at present indicate that LTD is induced when striatal afferents are activated at 10 Hz even in the absence of postsynaptic depolarization. Furthermore, 10 Hz-induced LTD was not prevented by blockade of mGluR1 with an antagonist known to prevent the HFS-induced form of plasticity. Our findings also suggest that continuous strong activation of the postsynaptic AMPA-type glutamate receptors is not necessary for LTD induction, since the EPSC was almost completely eliminated within 1–2 min of initiation of the 10 Hz train, a time point at which LTD is not yet initiated. The observation that a high concentration of nifedipine reduces, but does not fully block LTD suggests that induction of this form of LTD is not fully dependent on L-type calcium channel activation. The reduction in LTD by nifedipine but not intracellular BAPTA suggests the possibility that the 10 Hz stimulus produces calcium channel-dependent cannabinoid production in neighbouring neurones that contributes to LTD induction at synapses impinging on the neurones from which the recording was made. However, the lack of full blockade by nifedipine contrasts with HFS-induced LTD. It is possible that induction of 10 Hz-LTD does not depend on an intracellular calcium increase in any cell in the slice. Alternatively, the relative contribution of different sources for increasing intracellular calcium to the induction of 10 Hz- versus HFS-LTD may differ. The lack of effect of mGluR antagonists on 10 Hz-LTD provides the first evidence that striatal LTD does not require these receptors. This is a new and important finding given that past studies of HFS-LTD suggested a necessary role for these receptors. Overall, our results suggest that molecular mechanisms involved in the induction of 10 Hz-LTD are less complex than those involved in HFS-LTD. It will be interesting to determine if the molecular mechanisms underlying this presynaptic change are the same for 10 Hz- and HFS-LTD.
, 百拇医药
Source of endocannabinoids necessary for 10 Hz-induced striatal LTD
Our earlier studies indicated that the MSNs themselves are the primary source of endocannabinoids necessary for LTD induced by HFS. However, we have no solid evidence to date that a postsynaptic neurone is the site of endocannabinoid production/release for 10 Hz-LTD. This form of LTD appears to be independent of postsynaptic depolarization and insensitive to calcium chelation. These findings do not preclude the involvement of a postsynaptic endocannabinoid in 10 Hz-LTD induction, as endocannabinoid synthesis could be calcium independent and not require depolarization or metabotropic receptor activation. For example, 2-arachidonylglycerol (2-AG) can be synthesized by a phospholipase C-mediated increase in diacylglycerol (DAG), the immediate precursor of 2-AG. This appears to be the case for heterosynaptic inhibitory LTD in the hippocampus, which is not blocked by chelating postsynaptic calcium, but is blocked with inhibitors of PLC and DAG lipase (Chevaleyre & Castillo, 2003). However, 10 Hz-induced LTD was not blocked in a large majority of cells when the AMT inhibitor VDM11 was included postsynaptically. We previously demonstrated that this compound blocks HFS-LTD and release of endocannabinoids from MSNs (Ronesi et al. 2004), consistent with the idea that the AMT is involved in endocannabinoid release in that form of LTD. If 2-AG or another postsynaptically released endocannabinoid were responsible for LTD induction, we would have expected blockade of LTD by intracellular VDM11 loading. It is conceivable that both postsynaptic synthesis and release of the endocannabinoid involved in 10 Hz-LTD induction differ from those required for HFS-LTD induction. However, the existence in the same neurone of two such mechanisms for the generation and release of the same molecular signal would be surprising. Our findings certainly raise the strong suspicion that the source of endocannabinoids necessary for this form of LTD may not be the postsynaptic neurone.
, 百拇医药
Despite the evidence discussed above, the involvement of D2 receptors leaves open the possibility of a postsynaptic locus of LTD induction, because these receptors are highly expressed by a subset of MSNs (Gerfen et al. 1990; Sesack et al. 1994). However, D2 receptors have also been localized to the terminals of the corticostriatal afferents (Wang & Pickel, 2002), and are well known to act as autoreceptors on the striatal dopaminergic terminals themselves (Starke et al. 1989), so we cannot rule out the possibility of a presynaptic induction mechanism or a mechanism involving cross-talk between corticostriatal and dopaminergic afferents. In addition to being found at corticostriatal synapses, D2 receptors are found on aspiny acetylcholinergic interneurones (Alcantara et al. 2003), and it is possible that endocannabinoids are being synthesized and released from these cells. If the endocannabinoid necessary for 10 Hz-LTD is produced by neurones other than the postsynaptic MSNs, then synthesis of the endo-cannabinoid could be calcium dependent, and the endocannabinoid might spill over from neighbouring neurones to activate receptors on presynaptic neurones impinging on the neurone under study. Regardless of the site of endocannabinoid release in 10 Hz-LTD, the differential reliance on postsynaptic activation for induction of HFS- versus 10 Hz-LTD suggests that these two forms of endocannabinoid-dependent plasticity could occur under different in vivo conditions. HFS-LTD would be favoured by brief bouts of strong cortical input associated with postsynaptic activity. The LTD described in the present study might occur during more prolonged periods of moderate cortical input that is uncorrelated with postsynaptic activation.
, 百拇医药
Induction of striatal LTD and LTP by different frequency trains
Throughout the brain, different forms of synaptic plasticity can occur at the same synapse following differential frequencies of synaptic activation. For example, a well-characterized phenomenon at the CA3–CA1 synapse in the hippocampus is the induction of LTP with a brief high frequency train and LTD with a prolonged low frequency train. High frequency activation of corticostriatal inputs can produce either LTD or LTP (Calabresi et al. 1992b; Partridge et al. 2000; Kerr & Wickens, 2001), NMDAR-dependent striatal LTP can be depotentiated with a low-freqency train (2 Hz, 10 min) (Picconi et al. 2003), and thus, in striatum, as in other brain regions, it appears that multiple forms of plasticity can be induced by diverse frequencies of afferent activation.
, 百拇医药
Stimulation of the orofacial motor cortex at 5 Hz (100–200 pulses) has been shown to induce LTP of EPSPs recorded intracellularly from medium spiny neurones in the barbiturate-anaesthetized rat (Charpier et al. 1999). This finding might appear to contradict the present results showing LTD induction by a 10 Hz train. This discrepancy can be accounted for by several possible explanations. All of the recordings in this study were performed in relatively young animals (16–21 postnatal days), whereas the animals in the in vivo study were adults. It was shown previously that young corticostriatal synapses exhibit relatively high release probability, and that activity at corticostriatal synapses undergoes a developmentally regulated form of LTD in vivo (Choi & Lovinger, 1997b). Perhaps synapses from young animals are particularly susceptible to moderate or high frequency-stimulation-induced LTD because of their high probability of release. According to the sliding threshold model of synaptic plasticity, whether a synapse expresses LTD or LTP is dependent on previous activity at that synapse (see Abraham & Tate, 1997 for review). This may explain LTP induction in adult animals, whose corticostriatal synapses may have experienced relatively lower levels of input at a given synapse due to a lower probability of release. We suspect that prolonged moderate frequency firing could be sufficient to induce LTD in vivo, and there is evidence that striatal neurones can be activated at moderate frequencies for several minutes at a time (Charpier et al. 1999; Fellous et al. 2003). It is possible that the duration of moderate frequency synaptic activation necessary to induce LTD might be shorter in vivo, where it is likely that modulatory factors are present that are lost in the slice preparation. Ultimately, the type of synaptic plasticity induced by a given pattern of excitatory input will most likely depend on the activity of the postsynaptic neurone and other afferent inputs during the period of excitatory drive (see Lovinger et al. 2003 for discussion). More information on neuronal and afferent activities contributing to in vivo LTP and in vitro LTD will be needed to determine the conditions under which these two forms of plasticity occur.
, http://www.100md.com
Long-term changes in synaptic plasticity in the striatum are thought to be important for both motor and habit learning (Jog et al. 1999). The ways in which MSNs integrate these various inputs to regulate striatal output is only partially understood. The finding that prolonged activation of cortical afferents to the striatum induces a long-term decrease in synaptic efficacy provides additional evidence for activity-induced changes in synaptic efficacy in the striatum due to simultaneous activation of many inputs. It will be of interest to correlate changes in striatal MSN firing and synaptic efficacy with patterns of cortical input under different environmental and behavioural conditions. This may lead to a better understanding of the roles of synaptic potentiation and depression in learning memory and the shaping of habitual behaviour.
, http://www.100md.com
Conclusions
We have demonstrated that corticostriatal LTD can be induced by a prolonged moderate frequency stimulus train. This form of LTD is expressed presynaptically as a long-lasting decrease in EPSC amplitude that is due to a decrease in neurotransmitter release probability. LTD induced by 10 Hz stimulation requires the activation of both D2 dopamine receptors and CB1 cannabinoid receptors, but is insensitive to treatments that block HFS-induced LTD, including several manipulations that prevent postsynaptic activation and/or postsynaptic endocannabinoid release. It appears that extended activation of inputs to the striatum for minutes may provide the necessary signal to promote endocannabinoid synthesis and subsequent activation of presynaptic CB1 receptors.
, http://www.100md.com
References
Abraham WC & Tate WP (1997). Metaplasticity: a new vista across the field of synaptic plasticity. Prog Neurobiol 52, 303–323.
Alcantara AA, Chen V, Herring BE, Mendenhall JM & Berlanga ML (2003). Localization of dopamine D2 receptors on cholinergic interneurons of the dorsal striatum and nucleus accumbens of the rat. Brain Res 986, 22–29.
Cadas H, Gaillet S, Beltramo M, Venance L & Piomelli D (1996). Biosynthesis of an endogenous cannabinoid precursor in neurons and its control by calcium and cAMP. J Neurosci 16, 3934–3942.
, 百拇医药
Calabresi P, Maj R, Pisani A, Mercuri NB & Bernardi G (1992a). Long-term synaptic depression in the striatum: physiological and pharmacological characterization. J Neurosci 12, 4224–4233.
Calabresi P, Pisani A, Mercuri NB & Bernardi G (1992b). Long-term potentiation in the striatum is unmasked by removing the voltage-dependent magnesium block of NMDA receptor channels. Eur J Neurosci 4, 929–935.
Calabresi P, Pisani A, Mercuri NB & Bernardi G (1994). Post-receptor mechanisms underlying striatal long-term depression. J Neurosci 14, 4871–4881.
, 百拇医药
Calabresi P, Saiardi A, Pisani A, Baik J-H, Centonze D, Mercuri N, Bernardi G & Borrelli E (1997). Abnormal synaptic plasticity in the striatum of mice lacking dopamine D2 receptors. J Neurosci 17, 4536–4544.
Charpier S, Mahon S & Deniau JM (1999). In vivo induction of striatal long-term potentiation by low-frequency stimulation of the cerebral cortex. Neuroscience 91, 1209–1222.
Chevaleyre V & Castillo PE (2003). Heterosynaptic LTD of hippocampal GABAergic synapses. A novel role for endocannabinoids in regulating excitability. Neuron 38, 461–472.
, 百拇医药
Choi S & Lovinger DM (1997a). Decreased frequency but not amplitude of quantal synaptic responses associated with expression of corticostriatal long-term depression. J Neurosci 17, 8613–8620.
Choi S & Lovinger DM (1997b). Decreased probability of neurotransmitter release underlies striatal long-term depression and postnatal development of corticostriatal synapses. Proc Natl Acad Sci U S A 94, 2665–2670.
Fellous JM, Rudolph M, Destexhe A & Sejnowski TJ (2003). Synaptic background noise controls the input-output characteristics of single cells in an in vitro model of in vivo activity. Neuroscience 122, 811–829.
, 百拇医药
Freund TF, Katon I & Piomelli D (2003). Role of endogenous cannabinoids in synaptic signaling. Physiol Rev 83, 1017–1066.
Gerdeman GL & Lovinger DM (2001). CB1 cannabinoid receptor inhibits synaptic release of glutamate in rat dorsolateral striatum. J Neurophysiol 85, 468–471.
Gerdeman GL & Lovinger DM (2003). Emerging roles for enodcannabinoids in long-term synaptic depression. Br J Pharmacol 140, 781–789.
, 百拇医药
Gerdeman GL, Ronesi J & Lovinger DM (2002). Postsynaptic endocannabinoid release is critical to long-term depression in the striatum. Nat Neurosci 5, 446–451.
Gerfen CR, Engber TM, Mahan LC, Susel Z, Chase TN, Monsma FJ Jr & Sibley DR (1990). D1 and D2 dopamine receptor-regulated gene expression of striatonigral and striatopallidal neurons. Science 250, 1429–1432.
Giuffrida A, Parsons LH, Kerr TM, Rodriguez de Fonseca F, Navarro M & Piomelli D (1999). Dopamine activation of endogenous cannabinoid signaling in dorsal striatum. Nat Neurosci 2, 358–363.
, http://www.100md.com
Gubellini P, Saulle E, Centonze D, Bonsi P, Pisani A, Bernardi G, Conquet F & Calabresi P (2001). Selective involvement of mGlu1 receptors in corticostriatal LTD. Neuropharmacology 40, 836–846.
Hoffman AF, Oz M, Caulder T & Lupica CR (2003). Functional tolerance and blockade of long-term depression at synapses in the nucleus accumbens after chronic cannabinoid exposure. J Neurosci 23, 4815–4820.
Jog MS, Kubota Y, Connolly CI, Hillegaart V & Graybiel AM (1999). Building neural representations of habits. Science 286, 1745–1749.
, http://www.100md.com
Kasper EM, Larkman AU, Lubke J & Blakemore C (1994). Pyramidal neurons in layer 5 of the rat visual cortex II. Development of electrophysiological properties. J Comp Neurol 339, 475–494.
Kerr JN & Wickens JR (2001). Dopamine D-1/D-5 receptor activation is required for long-term potentiation in the rat neostriatum in vitro. J Neurophysiol 85, 117–124.
Kim J & Alger BE (2001). Random response fluctuations lead to spurious paired-pulse facilitation. J Neurosci 21, 9608–9618.
, 百拇医药
Lovinger DM, Partridge JG & Tang KC (2003). Plastic control of striatal glutamatergic transmission by ensemble actions of several neurotransmitters and targets for drugs of abuse. Ann N Y Acad Sci 1003, 226–240.
Lovinger DM, Tyler EC & Merrit A (1993). Short and long-term synaptic plasticity in the neostriatum. J Neurophysiol 70, 1937–1949.
Partridge JG, Apparsundaram S, Gerhardt GA, Ronesi J & Lovinger DM (2002). Nicotinic acetylcholine receptors interact with dopamine in induction of striatal long-term depression. J Neurosci 22, 2541–2549.
, http://www.100md.com
Partridge JG, Tang KC & Lovinger DM (2000). Regional and postnatal heterogeneity of activity-dependent long-term changes in synaptic efficacy in the dorsal striatum. J Neurophysiol 84, 1422–1429.
Picconi B, Centonze D, Hakansson K, Bernardi G, Greengard P, Fisone G, Cenci MA & Calabresi P (2003). Loss of bidirectional striatal synaptic plasticity in L-DOPA-induced dyskinesia. Nat Neurosci 6, 501–506.
Redman S (1990). Quantal analysis of synaptic potentials in neurons of the central nervous system. Physiol Rev 70, 165–198.
, http://www.100md.com
Robbe D, Kopf M, Remaury A, Bockaert J & Manzoni OJ (2002). Endogenous cannabinoids mediate long-term synaptic depression in the nucleus accumbens. Proc Natl Acad Sci U S A 99, 8384–8388.
Ronesi J, Gerdeman GL & Lovinger DM (2004). Disruption of endocannabinoid release and striatal long-term depression by postsynaptic blockade of endocannabinoid membrane transport. J Neurosci 24, 1673–1679.
Schmitz Y, Schmauss C & Sulzer D (2002). Altered dopamine release and uptake kinetics in mice lacking D2 receptors. J Neurosci 15, 8002–8009.
, 百拇医药
Sesack SR, Aoki C & Pickel VM (1994). Ultrastructural localization of D2 receptor-like immunoreactivity in midbrain dopamine neurons and their striatal targets. J Neurosci 14, 88–106.
Smith R, Musleh W, Akopian G, Buckwalter G & Walsh JP (2001). Regional differences in the expression of corticostriatal synaptic plasticity. Neuroscience 106, 95–101.
Starke K, Gothert M & Kilbinger H (1989). Modulation of neurotransmitter release by presynaptic autoreceptors. Physiol Rev 69, 864–989.
, http://www.100md.com
Stella N, Schweitzer P & Piomelli D (1997). A second endogenous cannabinoid that modulates long-term potentiation. Nature 388, 773–778.
Sung KW, Choi S & Lovinger DM (2001). Activation of group I mGluRs is necessary for induction of long-term depression at striatal synapses. J Neurophysiol 86, 2405–2412.
Tang KC & Lovinger DM (2000). Role of pertussis toxin-sensitive G-proteins in synaptic transmission and plasticity at corticostriatal synapses. J Neurophysiol 83, 60–69.
, 百拇医药
Tang K, Low MJ, Grandy DK & Lovinger DM (2001). Dopamine-dependent synaptic plasticity in striatum during in vivo development. Proc Natl Acad Sci U S A 98, 1255–1260.
Wang H & Pickel VM (2002). Dopamine D2 receptors are present in prefrontal cortical afferents and their targets in patches of the rat caudate-putamen nucleus. J Comp Neurol 442, 392–404.
Zhou FM, Liang Y & Dani JA (2001). Endogenous nicotinic cholinergic activity regulates dopamine release in the striatum. Nat Neurosci 4, 1224–1229., 百拇医药(Jennifer Ronesi, and Davi)