Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Repeated administration of cocaine leads to numerous behavioral alterations in both humans and laboratory animals. In humans, cocaine dependence is accompanied by withdrawal symptoms, including anhedonia, anergia, anxiety, depression, and cocaine craving (Gawin, 1991). Many of these symptoms have been successfully modeled in animals (Carroll and Lac, 1987; Kokkinidis and McCarter, 1990; Markou and Koob, 1991; Harris and Aston-Jones, 1993; Mutschler and Miczek, 1998), providing researchers the opportunity to elucidate underlying neuroadaptations. The past decade has witnessed considerable advances in our understanding of many cellular and molecular alterations that accompany dependence on, and withdrawal from, cocaine and other psychostimulants (Hyman, 1996; White and Kalivas, 1998). In many cases, these alterations (and others) have also been observed in humans (Volkow et al., 1990, 1999; Staley et al., 1994, 1997; Ross et al., 1996; Staley and Mash, 1996; Segal et al., 1997; Chen et al., 1999; McLeman et al., 2000).

We have focused our attention on alterations in dopamine (DA) receptor modulation of the mesocorticolimbic DA system after repeated cocaine administration (for reviews, see White et al., 1995a, 1997) because this system is intricately involved in the induction and expression of cocaine sensitization as well as cocaine withdrawal effects (for reviews, see Pierce and Kalivas, 1997; Kalivas et al., 1998; White and Kalivas, 1998). In recent years, we have extended this analysis to the level of voltage-sensitive ion channels that are targets of DA receptor modulation. We have reported that withdrawal from repeated cocaine administration produces a whole-cell depression characterized by a marked reduction in basal Na⁺ current in medium spiny neurons (MSNs) of the rat nucleus accumbens (NAc), rendering these cells less excitable to depolarizing influences (White et al., 1995b; Zhang et al., 1998). We proposed that the down-regulation of Na⁺ currents occurred as a result of increases in signaling through the DA D1 receptor/cAMP/protein kinase A (PKA) system (for review, see Nestler, 1997), resulting in an enhanced basal phosphorylation state of voltage-sensitive Na⁺ channels, an effect known to reduce Na⁺ currents (for review, see Catterall, 2000).

Herein, we have focused on voltage-sensitive calcium current (I_Ca) as a potential target of cocaine-induced neuroadaptations in NAc MSNs. Previous work within the dorsal striatum has demonstrated that DA D1 receptors suppress N- and P/Q-type Ca²⁺ channel conductances, through a cAMP/PKA/protein phosphatase-1 signaling system (Surmeier et al., 1995). Accordingly, we hypothesized that repeated DA D1 receptor stimulation produced by cocaine-induced inhibition of DA reuptake and enhancement of extracellular DA levels would also lead to alterations in the basal state of Ca²⁺ channel phosphorylation and I_Ca. We now report that during withdrawal from repeated cocaine administration, NAc MSNs exhibit a marked reduction in I_Ca through N- and R-type Ca²⁺ channels.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals and Treatments. Adult male Sprague-Dawley rats initially weighing 150 to 175 g were housed in groups of two to four in a temperature- and humidity-controlled vivarium under a 12-h light/dark cycle. Food and water were freely available. After 1 week of acclimation, rats were designated for use in acute experiments regarding I_Ca modulation or were randomly assigned to groups that received once daily i.p. injections of saline (1.0 ml/kg) or ()-cocaine HCl (15.0 mg/kg as the salt) for five consecutive days. Experiments using the treated rats were conducted on the 3rd day after cessation of injections (i.e., day 8).

Acute Dissociation Procedure. On the day of the experiment, rats were anesthetized with halothane and decapitated. Brains were quickly removed and dissected into blocks containing the NAc. Slices of 400 µm were cut with a motorized vibrating microtome while bathed in a low-Ca²⁺ (100 µM), HEPES-buffered salt solution containing 140 mM sodium isethionate, 2 mM KCl, 4 mM MgCl₂, 0.1 mM CaCl₂, 23 mM glucose, and 15 mM HEPES, pH 7.4 (300-350 mOsM/l). Slices were then incubated for 1 to 6 h at room temperature in a NaHCO₃-buffered saline (Eagle's balanced salts solution) bubbled with 95% O₂, 5% CO₂. Slices were then removed into the low-Ca²⁺ buffer, and with the aid of a dissecting microscope, the NAc, including both core and shell, was micropunched and placed in an oxygenated Cell-Stir chamber (Wheaton, Millville, NJ) containing pronase (1-1.5 mg/ml) in HEPES-buffered Hanks' balanced salt solution at 35°C. After 20 to 30 min of enzyme digestion, tissue was rinsed three times in the low-Ca²⁺, HEPES-buffered saline and mechanically dissociated with a graded series of fire-polished Pasteur pipettes. The cell suspension was then plated into a Petri dish mounted on the stage of an inverted microscope containing 2 ml of HEPES-buffered Hanks' balanced salt solution. After allowing the cells to settle, the solution bathing the cells was changed to our normal recording external solution.

Whole-Cell Recordings. Electrodes were pulled from Corning 7052 glass (Corning Glassworks, Corning, NY) and fire polished before use. The I_Ca was isolated by using the following solutions. The internal solution consisted of 180 mM N-methyl-glutamine, 40 mM HEPES, 4 mM MgCl₂, 0.1 mM BAPTA, 12 mM phosphocreatine, 2 mM Na₂ATP, 0.2 mM Na₃GTP, and 0.1 mM leupeptin, pH 7.3 (270-275 mOsM/l); the external solution consisted of 135 mM NaCl, 20 mM CsCl, 1 mM MgCl₂, 10 mM glucose, 10 mM HEPES, 0.001 mM tetrodotoxin, and 5 mM BaCl₂, pH 7.4 (300-305 mOsM/l). Electrodes filled with this solution had a resistance of 2 to 5 M when tested in the bath solution. A 6-mV junction potential was measured between the electrode and bath solution and was not compensated. Recordings were obtained with an Axon Instruments (Union City, CA) 200A patch-clamp amplifier and controlled and monitored with a PC 486 running pCLAMP6 (version 6.01) with a 2-kHz filter and a DigiData 1200 interface (333 kHz; Axon Instruments). After seal rupture, series resistance (<10 M) was compensated (70-80%) and periodically monitored. All currents were leak subtracted using the P/6 procedure in pCLAMP6. Recordings were made only from medium-sized neurons (<14 µm in diameter) that had only a few short proximal dendrites. Drugs were applied with a DAD-12 superfusion drug application system controlled by a 386 computer (ALA Scientific Instruments, Westbury, NY). All experiments were performed at room temperature (20-22°C). SKF 38393, SCH 23390, nifedipine, and okadaic acid were purchased from Sigma/RBI (Natick, MA); -conotoxin GVIA (CgTx) and PKI [4-25] were purchased from Peninsula Laboratories (Belmont, CA); -agatoxin TK (AgTx) was purchased from Peptides International (Louisville, KY); and 8-bromo-cAMP (8-Br-cAMP), Hanks' balanced salt solution, Earle's balanced saline solution, and tetrodotoxin were purchased from Sigma-Aldrich (St. Louis, MO).

Statistics. Differences between drug effects during within cell treatments were evaluated with Student's paired t tests. Differences between cocaine- and saline-pretreated NAc neurons were evaluated with Student's nonpaired t tests.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Whole-Cell I_Ca in NAc MSNs Is Comprised Primarily of High Voltage-Activated (HVA) Calcium Channels. Neurons used for recording were chosen according to their diameter and/or whole-cell capacitance. Only medium-sized neurons (7-14 µm in diameter) with short remaining processes were used. These neurons exhibited whole-cell capacitance of 3 to 9 pF as previously reported for MSNs in both dorsal striatum and NAc (Surmeier et al., 1995; Churchill and MacVicar, 1998; Zhang et al., 1998).

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Whole-cell ICa in an acutely dissociated NAc neuron. A, ICa was obtained by stepping the voltage from the holding potential (80 mV) to various levels of depolarization as shown by the voltage protocol. B, current-voltage relationship plotted from the traces shown in A.

View larger version (34K): [in this window] [in a new window]   Fig. 2.   Contribution of various Ca2+ channel subtypes to whole-cell ICa in NAc neurons. ICa was evoked by stepping the membrane from a holding potential of 80 to 0 mV, which provided maximum ICa (Fig. 1). We sequentially applied the N-type Ca2+ channel blocker CgTx, the P/Q-type Ca2+ channel blocker AgTx, and the L-type Ca2+ channel antagonist nifedipine (NFD) to isolate the contribution of each channel type to the ICa in NAc neurons as demonstrated in the recording traces. The nonselective Ca2+ channel blocker cadmium (CdCl2) abolished all current in each neuron tested. Inset, averages (mean ± S.E.M.) obtained from 14 NAc neurons.

DA D1 Receptor Stimulation Reduces N-Type and P/Q-Type I_Ca in NAc MSNs through a cAMP/PKA Enhancement of Protein Phosphatase Activity. During a thorough characterization of the effects of DA D1 receptor stimulation on I_Ca in MSNs of the dorsal striatum, Surmeier et al. (1995) proposed that D1 receptor stimulation suppresses N- and P/Q-type currents by enhancing a protein phosphatase-1-mediated dephosphorylation of the channels through the cAMP/PKA signaling system. To determine whether similar mechanisms exist for MSNs of the NAc, we next conducted a series of pharmacological studies. We observed that bath application of the DA D1 receptor-selective agonist SKF 38393 at a subsaturating concentration (1 µM) reversibly reduced I_Ca (Fig. 3), with a slow onset and recovery time course; the extent of suppression averaged 21.2 ± 1.6% (n = 21). At lower concentrations (0.1 µM), the modulation was weak and variable (8.9 ± 3.2%, n = 6; Fig. 3). Higher concentrations (100 µM) produced greater inhibition, but the effect was difficult to washout (29.3 ± 2.2%, n = 5; Fig. 3). Accordingly, all subsequent studies used the 1 µM concentration of SKF 38393. The DA D1 receptor-selective antagonist SCH 23390 significantly attenuated the effect of 1 µM SKF 38393 (n = 6; Fig. 3). The modulation produced by D1 receptor activation could be mimicked by 8-Br-cAMP (50 µM), a membrane-permeable cAMP analog, indicating involvement of the adenylyl cyclase-cAMP system coupled to D1 receptors (Fig. 3). The mean reduction was 23.7 ± 1.6% (n = 7). Furthermore, the suppression of I_Ca by 8-Br-cAMP was prevented by intracellular dialysis with the protein kinase inhibitor PKI [4-25] (10 µM, n = 5; Fig. 4A); similar results were obtained using SKF 38393 rather than 8-Br-cAMP (n = 4, 2.1 ± 0.3%; Fig. 4A). Finally, bath application of okadaic acid (1 µM, n = 5), a nonselective inhibitor of protein phosphatases (PPs) type 1 and type 2A, prevented the inhibitory effect of SKF 38393 (Fig. 4B). To identify the Ca²⁺ channel subtypes subject to D1 receptor modulation, we again used the selective N-type and P/Q-type channel toxins -CgTx (2 µM) and -AgTx (100-200 nM) to block these channels and reexamined the effect of SKF 38393 on I_Ca. During application of the two toxins, 1 µM SKF 38393 no longer suppressed I_Ca; in fact, a slight, but not statistically significant, enhancement was usually observed (mean enhancement, 5.2 ± 2.1, n = 5; Fig. 5), indicating that D1 receptor activation specifically acted on N- and P/Q-type channels to suppress current. Taken together, these findings suggest that, as in dorsal striatal MSNs, DA D1 receptor stimulation reduces N- and P-type Ca²⁺ channels in NAc neurons by producing a cAMP/PKA enhancement of PP activity, leading to dephosphorylation of Ca²⁺ channels.

View larger version (24K): [in this window] [in a new window]   Fig. 3.   DA D1 receptor stimulation or direct activation cAMP production reduces whole-cell ICa in NAc neurons. The DA D1 receptor agonist SKF 38393 (1 µM) suppressed ICa when applied in the bathing solution. After washing the neurons, the inhibitory effect of SKF 38393 was prevented by simultaneous application of the DA D1 receptor antagonist SCH 23390 (1 µM). Subsequently, the membrane-permeable cAMP analog 8-Br-cAMP (50 µM) produced inhibition that was comparable with that observed during D1 receptor stimulation. Inset traces, traces shown were taken at the times indicated by the corresponding numbers. Inset graphs, left inset shows the concentration-response curve for SKF 38393-induced suppression of whole-cell ICa (means ± S.E.M.); right inset shows the ability of the DA D1 receptor antagonist SCH 23390 (SCH) (1 µM) to attenuate the inhibitory effects of SKF 38393 (SKF) (t5 = 16.5, p < 0.00001, paired test). Bars represent means ± S.E.M.

View larger version (19K): [in this window] [in a new window]   Fig. 4.   Dopamine D1 receptor modulation of whole-cell ICa in NAc neurons involves PKA and a PP. A, when the PKA inhibitor PKI [4-25] (PKI) (10 µM) was included in the recording solution to allow it to dialyze into the neuron, suppression of current by 8-Br-cAMP (Br) was significantly prevented (t10 = 9.9, p < 0.0001). The left two columns in the graph inset show the results of 8-Br-cAMP alone and in the presence of PKI (means ± S.E.M.). Similar results were obtained with SKF 39393 (SKF), as shown in the right two columns of inset graph (statistics were not performed due to the small sample size in the PKI group; n = 4). B, bath application of the PP1/PP2A inhibitor okadaic acid (OA) (1 µM) also significantly (t4 = 10.7, p = 0.001, paired test) prevented the suppression of current by SKF 38393 (1 µM), as shown in the inset graph.

View larger version (26K): [in this window] [in a new window]   Fig. 5.   DA D1 receptor stimulation suppresses N- and P/Q-type ICa components. The ability of SKF 38393 (SKF) (1 µM) to suppress ICa is abolished after administration of the N-type antagonist CgTx and the P/Q-type antagonist AgTx. Toxin concentrations are as in Fig. 2. Note the slight increase in ICa that occurred during D1 receptor stimulation when N- and P/Q-type channels were blocked. Inset, traces shown were taken at the times indicated by the corresponding numbers.

Repeated Cocaine Treatment Decreases I_Ca Primarily through N- and R-Type Channels. We measured the I_Ca density to compare the basal state of I_Ca in animals treated repeatedly with cocaine or saline. Current density was estimated by using total cell capacitance as an index of membrane area (pA/pF). Repeated cocaine treatment produced a significant 30% reduction in basal I_Ca density (Fig. 6A), which was solely attributable to peak currents because there was no significant difference between the two groups of cells with respect to cell capacitance. There were no differences in the voltage dependence of activation or inactivation when cocaine-pretreated neurons were compared with control neurons (data not shown). The reduction in basal I_Ca was region specific because we failed to observe a similar effect when we recorded from motor cortex neurons after repeated cocaine treatment [I_Ca for saline, 81.1 ± 8.7 pA/pF (n = 7); I_Ca for cocaine, 82.5 ± 6.3 pA/pF (n = 8)]. Pharmacological blockade of selective Ca²⁺ channel subtypes revealed that both N- and R-type currents were significantly reduced by repeated cocaine treatment (Fig. 6B).

View larger version (45K): [in this window] [in a new window]   Fig. 6.   Effects of withdrawal from repeated cocaine treatment on ICa in NAc neurons. A, after 3 days of withdrawal from repeated cocaine administration, there is a significant reduction in basal ICa density (saline n = 44, cocaine n = 34, t76 = 4.12, p < 0.001). B, reduction in ICa density is primarily due to a reduction in N-type (t33 = 2.65, p < 0.02) and R-type currents (t33 = 2.21, p < 0.05). The slight alterations in P/Q-type and L-type currents were not statistically significant (saline n = 16, cocaine n = 19). Toxin and nifedipine concentrations are as described in Fig. 2C. In cocaine withdrawn NAc neurons, the PP1/PP2A inhibitor okadaic acid (OA, 1 µM) significantly (t16 = 2.59, p < 0.02) increases peak ICa (n = 8), an effect not observed in saline withdrawn control neurons (n = 10). D, suppression of peak ICa by the DA D1 receptor agonist SKF 38393 (1 µM) is significantly attenuated (t35 = 4.01, p < 0.001) in cocaine withdrawn NAc neurons (saline n = 21, cocaine n = 16). A similar significant attenuation (t11 = 2.31, p < 0.05) in the modulation was also observed for 8-Br-cAMP (saline n = 7, cocaine n = 6). All bars represent means ± S.E.M.

D1 Receptor Modulation of I_Ca Is Decreased in Cocaine-Pretreated NAc Neurons. Previous in vivo electrophysiological evidence suggested that repeated cocaine administration increases the functional sensitivity of DA D1 receptors in the NAc (Henry and White, 1991, 1995). However, D1 receptor modulation of whole-cell Na⁺ current was not altered after repeated cocaine administration (Zhang et al., 1998). To examine whether DA D1 receptor modulation of I_Ca was altered by repeated cocaine administration, we compared the ability of SKF 38393 to suppress peak I_Ca in NAc neurons dissociated from cocaine- and saline-pretreated rats. We found that the response to 1 µM SKF 38393 in the cocaine-pretreated group was significantly reduced compared with the saline-pretreated group (Fig. 6D). To determine whether this was due to a change in D1 receptors or to signaling events downstream of the receptors, we also evaluated the suppression of current produced by 8-Br-cAMP. As with the D1 agonist, the response to 8-Br-cAMP was significantly reduced in the cocaine-pretreated neurons compared with saline-pretreated neurons (Fig. 6D).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Our findings indicate that during withdrawal from repeated cocaine administration, there is a marked reduction in I_Ca carried through N- and R-type channels in MSNs of the NAc. As with the reductions in whole-cell Na⁺ currents that we reported previously (Zhang et al., 1998), the reductions in I_Ca were region-specific in that we did not observe similar changes in neurons dissociated from motor cortex, an area that lacks appreciable dopaminergic innervation. Our findings suggest that down-regulation of I_Ca results from an alteration in the basal state of phosphorylation of N-type channels as well as additional alterations in R-type channels.

HVA Ca²⁺ Channels Are Primarily Responsible for I_Ca in Most Acutely Isolated NAc MSNs. We observed that over 90% of adult NAc MSNs exhibit only HVA I_Ca with only sporadic neurons exhibiting activation at relatively hyperpolarized potentials. In addition, micromolar concentrations of Cd²⁺, which only weakly (ED₅₀ > 50 µM) block LVA (or T-type) Ca²⁺ channels (for review, see Randall, 1998), completely eliminated I_Ca in every neuron tested. Our findings are in close agreement with those of Surmeier and colleagues who demonstrated that neostriatal MSNs express LVA I_Ca when cultured from embryonic rat brain or dissociated during early postnatal development (P2-P7), but not when dissociated from later postnatal animals, leading these authors to suggest a developmental loss of LVA Ca²⁺ currents (Bargas et al., 1991, 1994). In contrast to our findings, another recent report (Churchill and MacVicar, 1998) indicated that 55% of NAc neurons acutely isolated from rats between P24 and P32 exhibited LVA currents. These authors also reported that 77% of cultured and only 42% of acutely isolated P5 to P11 neurons exhibited prominent LVA currents. These findings suggest that more extensive dendritic arbors in the cultured neurons may contribute to the increased probability of detecting LVA currents. Our findings from neurons isolated from rats between P47 and P52 contained very short (<40 µm) dendritic processes, which may have reduced the likelihood of observing LVA currents. Recent autoradiographic analyses indicate abundant expression of the LVA T-type _1G, _1H, and _1I subunits within the juvenile and adult striatum (McRory et al., 2001) and intracellular recordings of NAc MSNs in brain slices indicate the presence of a presumed LVA current (O'Donnell and Grace, 1993). However, it is possible that the current observed in brain slices is due to L-type channels with a more negative activation threshold (Avery and Johnston, 1996). As for HVA currents, we used selective blockers of various Ca²⁺ channels to demonstrate that the whole-cell current was comprised of N-type (~25%), P/Q type (~25%), and L-type (~30%), with a toxin/antagonist resistant (or R-type) component of about 20%. These values are similar to those reported by Churchill and MacVicar (1998) who further established that Q-type channels might primarily carry the P/Q current.

DA D1 Receptors Suppress N- and P/Q-Type Currents in NAc MSNs by a cAMP/PKA/PP Signaling Cascade. Surmeier et al. (1995) proposed a model for DA D1 receptor suppression of N- and P/Q-type I_Ca in striatal MSNs in which N- and P/Q-type Ca²⁺ channels are fully phosphorylated at rest, presumably by cGMP-dependent protein kinase. Because cGMP-dependent protein kinase and PKA phosphorylate similar substrates, an enhancement of PKA activity exerts no direct effects on N- and P/Q-type channels. Instead, PKA activation phosphorylates targeting proteins for PP1, releasing the catalytic subunit of PP1 into the cytosol to dephosphorylate N- and P/Q-type Ca²⁺ channels. Our results are consistent with this model. We found that DA D1 receptor signaling suppressed I_Ca in NAc MSNs. Stimulation of DA D1 receptors with the selective agonist SKF 38393 decreased I_Ca by approximately 25%, an effect that was prevented by the DA D1 receptor antagonist SCH 23390 and mimicked by the membrane-permeable cAMP analog 8-Br-cAMP. When N- and P/Q-type currents were blocked with -CgTx and -AgTx, respectively, SKF 38393 no longer suppressed I_Ca. In fact, some neurons exhibited a slight enhancement of I_Ca, a finding that is consistent with reports that DA D1 receptors can facilitate L-type I_Ca in striatal MSNs because L-type channels are not subject to the PKA-activated PP1 dephosphorylation (Surmeier et al., 1995; Hernández-López et al., 1997).

Repeated Cocaine Administration Reduces Basal I_Ca Primarily through N- and R-Type Channels. On the 3rd day of withdrawal from five daily cocaine injections, basal I_Ca in NAc MSNs was reduced by 30%. Use of selective antagonists for the various HVA channels indicated that both N- and R-type I_Ca components were significantly reduced. Experiments with okadaic acid support the likelihood that the reduction in basal I_Ca is due to enhanced signaling through the cAMP/PKA/PP1 cascade proposed above because this PP inhibitor increased I_Ca only in NAc neurons dissociated from cocaine-pretreated rats. This finding suggests greater constitutive PP1 activity in the cocaine-pretreated neurons. But because the okadaic acid effect was only about one-half (17%) of what would be expected if greater PP1 activity was fully responsible for the reduction in basal whole-cell I_Ca (30%), it is possible that repeated cocaine reduced basal I_Ca density by additional mechanisms, particularly with respect to decreased R-type current. One possibility would be an alteration in the expression of Ca²⁺ channel subunits as has been reported after chronic DA receptor stimulation in other systems (Fass et al., 1999). However, we can only speculate given that nothing is known regarding neuromodulation of R-type Ca²⁺ channels in MSNs (Foehring et al., 2000).

Functional Implications. We previously reported a whole-cell depression of Na⁺ currents in NAc MSNs during withdrawal from repeated cocaine treatment. We proposed that this neuroadaptation would render the NAc much less responsive to excitatory inputs from glutamatergic neurons in the prefrontal cortex, ventral subiculum, and basolateral amygdala. Given that convergent excitatory drive from these structures is required for NAc neurons to fluctuate between their hyperpolarized "down" state to a more depolarized "up" state necessary for spike activity (O'Donnell and Grace, 1995; Groenewegen et al., 1999; O'Donnell et al., 1999), the reduced excitability caused by down-regulated Na⁺ current would reduce the ability of NAc neurons to relay essential cognitive and motivational commands from the limbic system to the motor system, perhaps accounting for cocaine withdrawal symptoms of anhedonia, anergia, and depression.