Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Post-tetanic long-term potentiation (LTP) is considered a physiological form of synaptic plasticity, and its occurrence in either cortical (Bliss and Collingridge, 1993) or subcortical areas (Calabresi et al., 1996) has been regarded as a cellular substrate for memory and learning. More recently, it has been reported that pathological events such as anoxia and energy deprivation may induce long-term changes of excitatory synaptic transmission in hippocampal CA1 pyramidal neurons (Crépel et al., 1993; Hsu and Huang, 1997). Long-term pathological changes of synaptic transmission induced by ischemia and energy deprivation may underlie the differential neuronal vulnerability expressed in different brain areas or even in different neuronal subtypes in the same structure. Nevertheless, the cellular and molecular mechanisms underlying brain ischemic LTP (i-LTP) have not yet been elucidated. Overstimulation of AMPA/kainate and NMDA receptors during ischemia leads to rapid sodium and calcium accumulation; the latter event is worsened by the mobilization of intracellular calcium triggered by the concomitant mGluR activation (Sheardown et al., 1990; Lee et al., 1999; Pellegrini-Giampietro et al., 1999). However, the possible interaction between ionotropic and metabotropic glutamate receptors in the formation of i-LTP has never been investigated. We addressed this issue by using electrophysiological experiments from striatal spiny neurons, which express a marked vulnerability to ischemia and energy deprivation (Ferrante et al., 1985; Pulsinelli, 1985; Chesselet et al., 1990).

We report that brief in vitro ischemia causes, in coincidence with the stimulation of NMDA and mGluR1 receptors, the activation of protein kinase C (PKC) and mitogen-activated protein (MAP) kinase ERK. This complex cascade of biochemical events leads to the induction of i-LTP in striatal spiny neurons.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Preparation, Maintenance of the Slices, and Electrophysiological Recordings. Corticostriatal slices 270 µm thick were prepared from adult Wistar rats and mice. The preparation and maintenance of coronal slices have been described previously (Calabresi et al., 1997, 1999). A single slice was transferred to a recording chamber and submerged in a continuously flowing Krebs' solution (35°C, 2-3 ml/min) gassed with 95% O₂/5% CO₂. To study ischemia in striatal neurons, slices were deprived of glucose by totally removing glucose from the perfusate and by adding sucrose to balance the osmolarity. This solution was gassed with a mixture of 95% N₂/5% CO₂ instead of the normal gas mixture. The composition of the control solution was 126 mM NaCl, 2.5 mM KCl, 1.2 mM MgCl₂, 1.2 mM NaH₂PO₄, 2.4 mM CaCl₂, 11 mM glucose, and 25 mM NaHCO₃. In the majority of experiments, external magnesium was omitted.

Data Analysis and Drug Applications. Quantitative data on postischemic modifications of EPSP and excitatory postsynaptic currents (EPSC) are expressed as a percentage of the controls, which represent the mean of responses recorded during a stable period (15-30 min) before the ischemic episode. Values given in the text and in the figures are mean ± S.E.M. of changes in the respective cell populations. Wilcoxon's test or Student's t test (for paired and unpaired observations) were used to compare the means, and analysis of variance was used when multiple comparisons were made against a single control group. Drugs were applied by dissolving them to the desired final concentration in the saline or in the KCl intraelectrode solution. 2-Amino-5-phosphonovalerate (APV) and staurosporine were from Sigma (St. Louis, MO). BAPTA, PD98059, and Ro 32-0432 were from Sigma/RBI (Natick, MA). Calphostin C, CPCCOEt, LY367385, 2-methyl-6-(phenylethynyl)-pyridine (MPEP), (R,S)-1-aminoindan-1,5-dicarboxylic acid (AIDA), and picrotoxin were from Tocris Cookson (St. Louis, MO). UO126 was from Alexis Corporation (Läufelfingen, Switzerland).

Extract Preparation and Immunoblotting. Corticostriatal slices (270 µm) were prepared as described above. Slices were preincubated for 20 min in magnesium-free control solution alone or containing one of the following compounds: 10 µM PD98059, 30 µM LY367385, 30 µM MPEP, or 1 µM calphostin C. Slices were then subjected to 5-min ischemia as described above in the presence of the appropriate inhibitor, followed by either increasing wash time in magnesium-free control solution or a set 15-min wash in magnesium-free control solution plus the appropriate inhibitor. Striata were microdissected and rapidly homogenized using a Microfuge tube pestle in 40 µl of extraction buffer (10 mM Tris-HCl, 1% Triton X-100, 50 mM NaCl, 50 mM NaF, 5 µM ZnCl₂, 30 mM sodium pyrophosphate, 1 mM dithiothreitol, titrated to pH 7.05, with freshly added protease and phosphatase inhibitors) (Hipskind et al., 1994). Insoluble proteins were removed by centrifugation at 10,000g for 15 min at 4°C, and supernatant proteins were immediately denatured in 50 mM Tris-HCl, pH 6.8, 2% SDS, 2% glycerol, and 1% 2-mercaptoethanol. Western blots containing 5 µg of extract protein per lane were immunodetected as described (Bowler et al., 1999) using antisera specific for ERKs 1 and 2 activated by phosphorylation on Thr 202 and Tyr 204 (New England Biolabs, Beverly, MA), as well as control antisera directed against ERKs 1 and 2 (New England Biolabs). After incubation with horseradish peroxidase-coupled goat anti-rabbit antibodies (Sigma), immune complexes were visualized with the use of Renaissance ECL (PerkinElmer Life Sciences, Boston, MA) and exposure to Kodak XAR-5 film (Eastman Kodak, Rochester, NY).

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Electrophysiological Characterization of the Recorded Neurons. Intracellular and whole-cell patch-clamp recordings were performed from striatal spiny neurons and large aspiny (LA) cholinergic interneurons in corticostriatal slices. These neurons were easily distinguished by their typical electrophysiological properties, as shown previously in intracellular or whole-cell patch-clamp experiments (Kita et al., 1984; Kawaguchi, 1993; Calabresi et al., 1997, 1998a,b). Striatal spiny neurons represented the majority (n = 122; resting membrane potential, 85 ± 4 mV; input resistance, 40 ± 8 M) of the recorded cells, whereas 24 cells had electrophysiological characteristics of LA interneurons (resting membrane potential, 60 ± 4 mV; input resistance, 162 ± 12 M). Also, the physiological and pharmacological properties of cortically evoked EPSPs significantly differ in the two groups of neurons. In fact, in the presence of a physiological concentration of external magnesium (1.2 mM), spiny neurons responded to the stimulation of corticostriatal fibers by producing an EPSP mediated completely by the activation of glutamate AMPA receptors, whereas LA interneurons responded by generating synaptic potentials mediated in part by NMDA receptors. The AMPA glutamate receptor antagonist CNQX (10 µM), in fact, fully suppressed the EPSPs recorded intracellularly from spiny neurons, whereas the subsequent application of the NMDA receptor antagonist APV (50 µM) was required to abolish EPSPs evoked in LA interneurons (n = 6). In contrast, in spiny neurons, omission of external magnesium, a procedure that removes the voltage-dependent block of NMDA receptors, was necessary to reveal the NMDA component of the EPSP that could be blocked by APV (50 µM) (n = 20).

Effects of Ischemia on Spiny Neurons and Cholinergic Interneurons. To test the effect of energy deprivation on glutamatergic corticostriatal transmission in both the physiological condition (1.2 mM Mg²⁺) and magnesium-free medium, both spiny and LA neurons were deprived of oxygen and glucose for 3 min and their responses were recorded intracellularly. According to previous reports (Calabresi et al., 1999; Pisani et al., 1999), in vitro ischemia in striatal spiny neurons caused a membrane depolarization associated with a decreased input resistance (n = 101), whereas the same insult resulted in a membrane hyperpolarization of LA interneurons (n = 15). In the two neuronal subtypes, the ischemia-induced membrane potential and input resistance changes were unaffected by the omission of magnesium ions, and in all neurons used for data analysis, these electrophysiological parameters recovered to the control values within 3 to 6 min of wash (Figs. 1, A and B, and 2A). In both experimental conditions (1.2 mM and free Mg²⁺ medium), a marked blockade of synaptic transmission was observed within 2 min from the onset of the ischemic episode in the two neuronal subtypes. Upon return to control solution, the EPSP amplitude recorded from spiny neurons in the magnesium-free medium were potentiated. This i-LTP lasted throughout the period of observation (usually longer than 30 min) and reached a plateau after 15 to 20 min of wash (n = 45) (Fig. 1). On the contrary, in striatal neurons recorded in 1.2 mM Mg²⁺ medium (data not shown), as well as in LA interneurons recorded either in 1.2 mM or free magnesium medium, cortically evoked EPSPs returned to control value after the brief ischemic episode without showing any significant long-term change (Fig. 2A). Similar results were obtained from LA interneurons in whole-cell patch-clamp experiments. In the presence of 1.2 mM magnesium, in fact, in vitro ischemia (4 min) produced outward currents in all of the cells that were voltage-clamped at 50 mV (+35 ± 18 pA; n = 9), and although cortically evoked EPSCs were partially sensitive to 50 µM APV, they did not exhibit any long-term enhancement at the wash of the ischemic episode (Fig. 2B). Similar results were obtained in magnesium-free medium (n = 3) (data not shown).

View larger version (39K): [in this window] [in a new window]   Fig. 1.   Induction of i-LTP requires NMDA glutamate receptor activation in striatal spiny neurons. A, the electrophysiological trace shows a chart record of the membrane potential changes induced in a striatal spiny neuron by a brief period of ischemia (black bar) in the absence of magnesium ions. The upward deflections reflect EPSPs evoked by cortical stimulation. Note that after a 5-min wash, the recording of the membrane was interrupted for approximately 15 min. The inset shows superimposed EPSPs (averages of four single sweeps) at higher sweep speed before ischemia (a) and 20 min after (b). The RMP of the neuron was 86 mV. B, the electrophysiological trace shows a chart record of the membrane potential changes induced in a striatal spiny neuron by a brief period of ischemia (black bar) in magnesium-free medium plus 50 µM APV. The inset shows superimposed EPSPs (averages of four single sweeps) at higher sweep speed before (a) and 20 min after ischemia (b). The RMP of the neuron was 84 mV. Scale bars in A also apply to B. C, cumulative data obtained by whole-cell patch-clamp recordings in 1.2 mM magnesium. Note that i-LTP was evident either at 80 mV () or at 50 mV () when the ischemic insult was applied, holding the cell at 50 mV. Superimposed traces in the inset represent single experiments showing synaptic currents before ischemia (a) and 20-min washout (b) recorded at holding potential of 80 mV (left) and at 50 mV (right). D, i-LTP was evident neither at 80 mV () nor at 50 mV () when the ischemic insult was applied, holding the cell at 80 mV.

View larger version (28K): [in this window] [in a new window]   Fig. 2.   Large cholinergic aspiny interneurons do not express i-LTP. A, top left, striatal cholinergic interneurons express ischemic LTP neither in control medium () nor in the absence of external magnesium (). The single traces in the top right are EPSPs evoked by cortical stimulation before (a) and 20 min after (b) ischemia. The trace (bottom) shows a chart record of the membrane potential and EPSP amplitude changes (upward deflections) induced in a striatal aspiny cholinergic interneuron by a brief period of ischemia (black bar) in magnesium-free medium (RMP, 60 mV). B, during whole-cell patch clamp recordings (left), striatal cholinergic interneurons do not express i-LTP even when voltage is clamped at 50 mV. The single traces are EPSCs evoked by cortical stimulation before (a) and 20 min after ischemia (b), obtained from a single experiment.

Critical Role of NMDA Receptors in i-LTP Induction. Bath application of the NMDA receptor antagonist APV (50 µM) fully prevented the induction of i-LTP in striatal spiny neurons recorded in magnesium-free medium, supporting the conclusion that this experimental procedure allowed the induction of this form of synaptic plasticity by facilitating NMDA-mediated neurotransmission (n = 6) (Fig. 1B). In physiological conditions, the Mg²⁺-dependent blockade of NMDA receptors is typically relieved by membrane depolarization. Use of the whole-cell patch-clamp technique, therefore, even in the presence of physiological concentrations of Mg²⁺ ions (1.2 mM), an NMDA-mediated component of corticostriatal EPSCs could be unmasked by clamping the membrane potential of the recorded striatal cells at membrane values significantly positive to their resting membrane potential (RMP). Accordingly, when recorded at holding potentials of 50 mV but not at 80 mV, striatal spiny neurons responded to a single stimulation of corticostriatal fibers by producing an EPSC that was partially blocked by 50 µM APV (19 ± 4%, n = 6, p < 0.01).

Role of mGluR1 in i-LTP. Antagonists of group I mGluRs (mGluR1 and mGluR5) have been found to be consistently neuroprotective in a variety of experimental conditions, including brain ischemia (Nicoletti et al., 1999). To investigate the possibility that group I mGluR antagonists exert their protective effects by interfering with the excitotoxic damage that follows the formation of i-LTP, we studied the effects of selective antagonists of these receptors (Schoepp et al., 1999) on this pathological form of synaptic plasticity. We found that the selective mGluR5 antagonist MPEP (30 µM, n = 4, 10-min bath application) failed to affect i-LTP (p > 0.05), whereas LY367385 (30 µM, n = 4), AIDA (300 µM, n = 4), and CPCCOEt (100 µM, n = 4), blockers of mGluR1, fully prevented its induction (p < 0.01 for both experimental conditions) (Fig. 3A). The involvement of mGluR1 in the generation of i-LTP was also confirmed by using transgenic mice that selectively lacked this receptor subtype (Conquet et al., 1994). Striatal spiny neurons recorded intracellularly from these mice (n = 10) were indistinguishable electrophysiologically from their wild-type counterparts (n = 9) and from rat striatal neurons, but they did not exhibit i-LTP after the ischemic challenge (n = 10, p > 0.05). Wild-type mice showed robust i-LTP under the same conditions (n = 9, p < 0.01) (Fig. 3B). This difference could not be attributed to different membrane potential changes induced by in vitro ischemia, because mGluR1 knockout mice and wild-type mice responded to 3 min of energy deprivation with a membrane depolarization of comparable amplitude (26 ± 3 mV and 28 ± 4 mV, respectively; p > 0.05).

View larger version (42K): [in this window] [in a new window]   Fig. 3.   Activation of mGluR1s is required for i-LTP in striatal spiny neurons. A, i-LTP is blocked by LY367385, AIDA, and CPCCOEt but not by MPEP. The superimposed voltage traces represent single experiments using these mGluR antagonists. RMPs were 85 mV (in MPEP), 84 mV (in LY367385), 85 mV (in AIDA), and 86 mV (in CPCCOEt). B, i-LTP is absent in mice lacking mGluR1 but not in normal mice. The superimposed voltage traces represent single experiments obtained from normal (mGluR1 +/+) and mGluR1-lacking mice (mGluR1 /). RMPs were 85 mV (mGluR1 +/+) and 85 mV (mGluR1 /). In each experimental condition, the averages of single traces were obtained before (a) and 20 min after ischemia (b).

Role of Intracellular Calcium and PKC in i-LTP. Stimulation of mGluR1 results in the activation of PKC and in the elevation of intracellular levels of calcium ions (Nicoletti et al., 1999; Schoepp et al., 1999). This latter event also follows NMDA receptor activation and ischemic membrane depolarization and is recognized, along with PKC activation, as an important determinant for the generation of many forms of synaptic plasticity (Bliss and Collingridge, 1993; Calabresi et al., 1996). Therefore, we tested the dependence of i-LTP on intracellular calcium accumulation and PKC stimulation. Intraneuronal injection of high concentrations of the calcium-chelating agent BAPTA (100 mM), which did not affect per se the membrane depolarization produced by ischemia (30 ± 2 mV, p > 0.05, n = 6), fully prevented i-LTP (p > 0.05) (Fig. 4A). Similar results were obtained using the PKC blockers calphostin C (1 µM, 7 min, n = 7) or staurosporine (100 nM, 7 min, n = 6, p > 0.05) and after intracellular administration of Ro 32-0432 (100 µM, n = 6, p > 0.05), another highly selective antagonist of PKC (Fig. 4B). This also demonstrates that the induction of PKC by energy deprivation that is crucial for i-LTP occurs at the postsynaptic site of corticostriatal synapses.

View larger version (28K): [in this window] [in a new window]   Fig. 4.   Intracellular calcium elevation, as well as PKC and MAP kinase activation, is required for i-LTP in striatal spiny neurons. A, buffering of intracellular calcium by BAPTA prevents i-LTP. B, i-LTP is blocked by either bath application of calphostin C and staurosporin or intracellular application of Ro 32-0432. C, the i-LTP is also prevented by PD98059 and UO126.

Inhibitors of the ERK Cascade Suppress i-LTP. Mitogen-activated protein kinases are involved in the generation of NMDA-mediated neurotoxicity (Ghosh and Greenberg, 1995) and ischemic neuronal damage (Alessandrini et al., 1999; Sugino et al., 2000). Thus, we tested the induction and expression of i-LTP in the presence of PD98059 and UO126, two specific inhibitors of MAP kinase kinase (MEK) and thereby of p42/44 MAP kinase activation (Alessi et al., 1995). As shown in Fig. 4C, long-term incubation (2 h) of the slices in 10 µM PD98059 (n = 8) or 30 µM UO126 (n = 5) prevented the induction of striatal i-LTP (p > 0.05). Neither drug affected the intrinsic membrane properties of the striatal spiny neurons, nor did the inhibitors alter the physiological and pharmacological characteristics of corticostriatal EPSPs and the amplitude of ischemia-induced membrane depolarization (30 ± 2 mV in PD98059 and 27 ± 3 mV in UO126; p > 0.05 for each experimental condition).

View larger version (56K): [in this window] [in a new window]   Fig. 5.   ERK is activated by ischemia and is modulated by drugs interfering with i-LTP expression. A, ERK is activated by ischemia in striatal slices. Striatal slices were incubated in Mg-free solution (control), subjected to 5-min ischemia, and followed by increasing wash time in Mg-free solution. Proteins in whole-tissue lysates were separated by SDS-polyacrylamide gel electrophoresis, blotted onto polyvinylidene difluoride, and immunodetected using activated ERK-specific antiserum or control ERK antiserum as indicated. The two species of ERK recognized by these sera, p44ERK1 and p42ERK2, are indicated to the left of each panel. B, striatal slices were subjected to in vitro ischemia and 20-min Mg-free wash in the presence of the following inhibitors: 10 µM PD98059, 30 µM LY367385, 30 µM MPEP, or 1 µM calphostin C. ERK activation was determined by immunoblotting, as described above. In both A and B, the data have been quantified and the phospho ERK1 and ERK2 signals corrected for the amount of the kinase revealed by the control signal. The corresponding numbers are presented in the appropriate panels as relative ERK induction.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

Activation of NMDA receptors is critical for the induction of i-LTP. This evidence suggests that i-LTP occurs in vivo during the "up state" of striatal neurons. In fact, although striatal spiny neurons are highly polarized in vitro and require manipulations to reveal an NMDA-mediated component of excitatory transmission, they show a characteristic membrane oscillatory behavior when recorded in vivo (Calabresi et al., 1996; Wilson and Kawaguchi, 1996). During the depolarized "up state," neurons reach membrane potentials of approximately 50 mV, thereby enabling the inactivation of NMDA receptors and possibly the induction of i-LTP (Wilson and Kawaguchi, 1996). In our in vitro experiments, we were able to mimic this in vivo condition either by removing external magnesium or by holding the membrane potential of the cells at depolarized levels. Although cholinergic interneurons possess a more positive resting membrane potential and show tonic firing activity (Graybiel et al., 1994; Aosaki et al., 1995; Wilson and Kawaguchi, 1996), they do not express i-LTP. This finding is in agreement with those of previous studies showing that this neuronal subtype is resistant to ischemia and energy deprivation (Pulsinelli, 1985; Chesselet et al., 1990).

The observation that i-LTP is blocked by mGluR1 antagonists and is absent in mice selectively lacking this receptor subtype provides a new synaptic mechanism explaining the neuroprotective effects of group I mGluR antagonists in ischemic and excitotoxic neuronal damage (Nicoletti et al., 1999; Pellegrini-Giampietro et al., 1999). The evidence that intracellular application of Ro 32-0432, as well as bath application of staurosporin and calphostin C, prevented i-LTP demonstrates that i-LTP requires the activation of PKC expressed postsynaptically in spiny neurons. PKC activity is critically regulated by intracellular calcium levels (Calabresi et al., 1998a). Accordingly, intracellular application of the calcium chelator BAPTA also blocks i-LTP. Because intracellular calcium elevation and PKC stimulation represent critical biochemical events resulting from the activation of mGluR1 (Nicoletti et al., 1999), we favor the hypothesis that mGluR1s required for i-LTP are postsynaptically located on striatal spiny neurons.

The convergent action of energy deprivation and mGluR1 and PKC stimulation allows the induction of i-LTP through ERK stimulation, because we found that the activation of this enzyme by brief ischemia is significantly reduced by mGluR1 and PKC inhibitors. We also observed that the blockade of the ERK cascade with specific inhibitors (Alessi et al., 1995) prevented i-LTP. In addition, the finding that the inhibition of MAP kinase pathway was able to prevent i-LTP even in the presence of the AMPA receptor antagonist CNQX favors the idea of a close interaction among ischemia, NMDA receptors, and MAP kinase pathway stimulation in the induction phase of i-LTP.

A possible model to explain the interaction between NMDA receptors, mGluR1, PKC, and MAP kinase pathway in the formation of striatal postischemic LTP is shown in Fig. 6. We suggest that during a brief ischemic episode, an abnormal amount of glutamate is released from corticostriatal terminals, causing a membrane depolarization. The ischemia-induced membrane depolarization leads to the activation of NMDA glutamate receptors which, in turn, causes an elevation of intracellular calcium levels. A further increase of intracellular calcium levels is also induced by the activation of mGluR1. The augmented intracellular calcium concentration is critical for the activation of PKC. The activated PKC triggers a cascade of events resulting in the activation of MEK and MAP kinase. Activated MAP kinase probably has multiple targets, including cAMP-response element-binding protein, which mediates its ability to induce long-term adaptive changes in neurons. The resulting synthesis of new proteins mediates the long-term remodeling of the synapse believed to underlie postischemic LTP.

View larger version (47K): [in this window] [in a new window]   Fig. 6.   Putative signal transduction pathways operating in striatal postischemic LTP. See description under Discussion. AMPA, AMPA subtype of glutamate receptor; NMDA, NMDA subtype of glutamate receptor; mGluR1, metabotropic glutamate receptor subtype 1; PLC, phospholipase C; DAG, diacylglycerole; MAPK, mitogen-activated protein kinase; CREB, cAMP response element binding protein.

ERK cascade inhibition has been shown to alter hippocampal LTP (English and Sweatt, 1997) and to block long-term facilitation in the mollusk Aplysia californica (Martin et al., 1997). At present, however, very little is known about the role of the ERK pathway in the synaptic processes after ischemia. It has been hypothesized that cerebral ischemia leads to the activation of this signal transduction cascade via glutamate release and activation of NMDA receptors, which, in turn, causes calcium entry (Wieloch et al., 1996; Xia et al., 1996; Alessandrini et al., 1999).

One would expect that the ERK cascade exerts its effect selectively on late phases of synaptic plasticity by altering the pattern of gene expression. However, we found that the pharmacological inhibitors of the ERK cascade also blocked early stages of i-LTP, thereby suggesting a role early in this phenomenon. In agreement with this, PD98059 completely prevents LTP in the dentate gyrus and long-term depression in the prefrontal cortex, indicating that the activation of the MAP kinase ERK cascade can play a role in the induction phases of different forms of synaptic plasticity in several brain areas (Coogan et al., 1999; Otani et al., 1999; Sweatt, 2001).

Studies dealing with the role of ERK activity in ischemia have proven contradictory. Sustained ERK activation after ischemia was suggested to mediate selective resistance to ischemia in adult (Hu and Wieloch, 1994) and neonatal brains (Hee Han and Holtzman, 2000). Conversely, it has been supposed that ERK activity might favor neuronal death through inappropriate protein phosphorylation in CA3 pyramidal cells and disruption of the cytoskeleton in CA1 neurons (Runden et al., 1998). We provide the first evidence that activation of the mGluR1/PKC/MAP kinase pathway is required for the generation of the ischemia-induced long-term enhancement of excitatory transmission in the brain.

Protein kinase A is involved in several forms of synaptic plasticity in various brain areas (Greengard et al., 1999; Otani et al., 1999; Calabresi et al., 2000) and interacts with both PKC and MAP kinase activity (Bowler et al., 1999; Otani et al., 1999; Calabresi et al., 2000; Sweatt, 2001). Thus, it would be interesting to address the role of this kinase in striatal i-LTP. Experiments are in progress in our laboratory to investigate this important issue.