Protein Kinase C-mediated Phosphorylation and Functional Regulation of Dopamine Transporters in Striatal Synaptosomes*

Dopamine transporters (DATs) are members of a family of Na+- and Cl−-dependent neurotransmitter transporters responsible for the rapid clearance of dopamine from synaptic clefts. The predicted primary sequence of DAT contains numerous consensus phosphorylation sites. In this report we demonstrate that DATs undergo endogenous phosphorylation in striatal synaptosomes that is regulated by activators of protein kinase C. Rat striatal synaptosomes were metabolically labeled with [32P]orthophosphate, and solubilized homogenates were subjected to immunoprecipitation with an antiserum specific for DAT. Basal phosphorylation occurred in the absence of exogenous treatments, and the phosphorylation level was rapidly increased when synaptosomes were treated with the phosphatase inhibitors okadaic acid or calyculin. Treatment of synaptosomes with the protein kinase C activator phorbol 12-myristate 13-acetate (PMA) also increased the level of phosphate incorporation. This occurred within 10 min and was dosedependent between 0.1 and 1 μm PMA. DAT phosphorylation was also significantly increased by two other protein kinase C activators, (−)-indolactam V and 1-oleoyl-2-acetyl-sn-glycerol. The inactive phorbol ester 4α-phorbol 12,13-didecanoate at 10 μm was without effect, and PMA-induced phosphorylation was blocked by treatment of synaptosomes with the protein kinase C inhibitors staurosporine and bisindoylmaleimide. These results indicate that DATs undergo rapidin vivo phosphorylation in response to protein kinase C activation and that a robust mechanism exists in synaptosomes for DAT dephosphorylation. Dopamine transport activity in synaptosomes was reduced by all treatments that promoted DAT phosphorylation, with comparable dose, time, and inhibitor characteristics. The change in transport activity was produced by a reduction inV max with no significant effect on theK m for dopamine. These results suggest that synaptosomal dopamine transport activity is regulated by phosphorylation of DAT and present a potential mechanism for local neuronal control of synaptic neurotransmitter levels and consequent downstream neural activity.

Dopamine transporters (DATs) 1 are integral membrane neuronal proteins that function to terminate dopaminergic neurotransmission by the rapid reuptake of synaptic dopamine into presynaptic neurons. As the primary mechanism for the clearance of synaptic dopamine, DAT is the main determinant that regulates the intensity and duration of dopaminergic neurotransmission (1). DAT is implicated in the etiology of psychostimulant drug abuse, as binding of cocaine and amphetamine to the protein inhibit dopamine transport (2), and the resulting elevation of synaptic dopamine levels is believed to underlie the reinforcing properties of these drugs (3,4). DAT is also a dopaminergic-specific mode of entry for the neurotoxins 6-hydroxydopamine and 1-methyl-4-phenylpyridinium (5,6), implicating it in mechanisms of neurotoxicity that serve as the best current models for Parkinsonian neurodegeneration.
DAT is a member of a large class of neurotransmitter and amino acid transporters, including carriers for norepinephrine, serotonin, ␥-amino butyric acid, glycine, proline, taurine, and betaine, which drive reuptake of transmitter by cotransport of Na ϩ and Cl Ϫ down electrochemical gradients (7)(8)(9). Molecular cloning of DAT demonstrates the presence of 12 potential transmembrane domains, with a presumed topology orienting the N and C termini intracellularly. DAT and the other proteins in this group are extensively glycosylated, and their sequences contain numerous consensus phosphorylation sites for PKA, PKC, and Ca 2ϩ -calmodulin kinase (10 -14).
The presence of potential phosphorylation sites on these proteins suggests that they may be subject to phosphorylationinduced functional regulation, and several studies have shown that transport of neurotransmitters is affected by protein kinase activators. Treatment of striatal synaptosomes and heterologous expression systems with phorbol esters or other protein kinase activators reduces dopamine transport activity of mouse, rat, and human DATs (15)(16)(17)(18). Activation of arachidonic acid pathways can also decrease hDAT activity (19), whereas elevated striatal dopamine uptake occurs after treatment with Ca 2ϩ pathway activators (20). PMA also regulates activity of several other neurotransmitter transporters, including glutamate transporters expressed in in cultured glia and transfected HeLa cells (21), ␥-amino butyric acid transporters expressed in Xenopus oocytes (22), and glycine transporters expressed in cultured embryonic kidney cells (23). Serotonin transport can also be acutely modulated by calmodulin in placental choriocarcinoma cells (24), PMA in HEK 293 cells (25), and cGMP and nitric oxide in rat basophilic leukemia cells (26).
Evidence for the potential involvement of transporter phosphorylation in producing these effects has been obtained only for dopamine and glutamate transporters, which undergo in vivo phosphorylation regulated by PMA in rDAT-LLC-PK 1 cells and C6 glioma cells, respectively (16,21). Purified glutamate transporters and serotonin transporter N-and C-terminal tail fusion proteins have also been shown to be in vitro substrates for PKA and/or PKC (21,27). Thus, although several studies have indicated that protein kinase activators regulate neurotransmitter transport activity, only a small number of studies utilizing in vitro or heterologous expression systems have demonstrated direct transporter phosphorylation, and to date little work has been done examining these properties in the brain. In this report we describe the endogenous phosphorylation and dephosphorylation of DAT in rat striatal synaptosomes and show that the phosphorylation state of the protein is regulated by protein kinase C. Dopamine transport activity is reduced by all treatments that increase DAT phosphorylation, with similar dose, kinetic, and inhibitor characteristics, suggesting that in brain DAT is subject to phosphorylation-induced functional regulation.

EXPERIMENTAL PROCEDURES
Tissue Preparation-Male Sprague Dawley rats, 150 -300 g, were decaptitated, and the striata were rapidly removed and weighed. The tissue was homogenized in a Teflon-glass homogenizer in 10 ml of cold 0.32 M sucrose and centrifuged at 800 ϫ g for 10 min. The supernatant was recentrifuged at 12,500 ϫ g for 10 min, and the resulting synaptosomal pellet was resuspended in 0.32 M sucrose at a concentration of 120 mg/ml original wet weight for phosphorylation experiments or 30 mg/ml original wet weight for uptake assays (0.5-2 mg/ml protein). Proteins were assayed using the Pierce protein assay kit with bovine serum albumin as the standard.
Phosphorylation-Krebs bicarbonate buffer (25 mM Na 2 HCO 3 , 124 mM NaCl, 5 mM KCl, 1.5 mM CaCl 2 , 5 mM MgSO 4 , 10 mM glucose, pH 7.3) was saturated with 95% O 2 /5% CO 2 , and [ 32 P]orthophosphate was added to a final concentration of 1.5 mCi/ml. The synaptosomal suspension was diluted 4-fold in this buffer to a tissue concentration of 30 mg/ml original wet weight in a final volume of 300 l. Treatment compounds were added in 3-l volumes, and incubations were carried out at 30°C for 45 min with shaking at 100 rpm. At the end of the incubation, reaction tubes were transferred to an ice water bath to reduce enzyme activity, and samples were centrifuged at 4,000 ϫ g for 2 min at 4°C. Supernatants were removed, and the tissue was solubilized with 100 l of 0.5% SDS and four passages through a 26-gauge needle. The solubilized samples were centrifuged at 20,000 ϫ g for 10 min at 4°C, and supernatants were removed for analysis.
Immunoprecipitation-Solubilized tissue samples were immunoprecipitated with antibody 16, directed against amino acids 42-59 of the deduced DAT primary sequence (28). This antibody has been shown by immunoprecipitation, immunoblot, and immunohistochemistry to be highly specific for DAT (28,29). Solubilized, 32 PO 4 -labeled synaptosomal tissue was diluted with 50 mM Tris-HCl, pH 7.5, containing 0.1% Triton X-100 and serum 16 diluted 1:400. Samples were incubated at 4°C for 1 h followed by the addition of 20 l of protein Sepharose CL4B (Pharmacia Biotech Inc.) for an additional hour. Immune complexes were washed twice with the Tris-Triton buffer, and samples were eluted with SDS-polyacrylamide gel electrophoresis loading buffer (60 mM Tris-HCl, pH 6.8, 2% SDS, 10% glycerol, and 5 mM dithiothreitol). Samples were electrophoresed on 7% polyacrylamide gels followed by autoradiography using Kodak BioMax film for 1-5 days. Amersham Rainbow molecular mass markers were standards on all gels. Positive controls for immunoprecipitations were provided by parallel precipitation of [ 125 I]DEEP photoaffinity labeled dopamine transporters (30). For peptide blocking experiments, diluted antiserum was preincubated with either 50 g/ml peptide 16 or peptide 18 (amino acids 580 -608) prior to addition of the sample. Levels of DAT phosphorylation were quantitated by Cerenkov counting. All results were obtained in three or more independent experiments.
Dopamine Uptake-For preincubations with test compounds or vehicle, synaptosomal samples were distributed into 500-l volumes and treated at 30°C with shaking at 100 rpm for 20 min or other indicated times. After preincubation, 50-l aliquots were distributed into assay tubes prepared as follows, and [ 3 H]DA was added to inititate transport.
Dopamine uptake was performed in a final volume of 1000 l, in a buffer consisting of 16 mM Na 2 PO 4 , 126 mM NaCl, 4.8 mM KCl, 1.3 mM CaCl 2 , 1.4 mM MgSO 4 , 1 mM ascorbic acid, and 1 M pargyline. Final tissue concentration was 1.5 mg/ml original wet weight, and [ 3 H]DA was 1 nM except for saturation analyses where [ 3 H]DA was increased to 10 nM and unlabeled dopamine varied from 10 nM to 1 M. Nonspecific uptake was defined with 100 M (Ϫ)-cocaine. Uptake assays were carried out for 3 min at 30°C, followed by filtration under vacuum through Whatman GF/B filters soaked in 0.05% polyethyleneimine. Filters were washed rapidly with 5 ml of ice-cold 0.3 M sucrose using a Brandel M48R filtering manifold. Radioactivity bound to filters was counted using a Beckman LS 6000 liquid scintillation counter at 45-50% efficiency. Specific dopamine uptake in control preparations averaged 1.18 Ϯ 0.38 pmol/min/mg protein, and data were analyzed by EBDA and LIGAND computer software. For purposes of comparison across experiments, uptake results were converted to percentages of control values and analyzed statistically with paired t tests using Statview version 4D by Abacus Concepts, Inc. (Berklely, CA). Because of the modest effects produced by the test compounds, the experimental design incorporated substantial assay redundancy. In each experiment, preincubations with test compounds or vehicle were performed on duplicate tissue samples. Each tissue sample was analyzed in six assay tubes, and the resulting values were averaged. Each data point thus consists of the mean of 12 independent determinations.
Alanine Uptake-[ 3 H]Alanine uptake was examined using the same conditions as for dopamine transport, except that [ 3 H]alanine was 10 nM.
Materials-PMA, 4␣PDD, okadaic acid, calyculin, microcystin, cyclosporin A, staurosporine, bisindoylmaleimide I HCl, (Ϫ)-indolactam V, OAG, forskolin, 8-Br-cAMP, and isobutylmethylxanthine were obtained from Calbiochem (San Diego, CA). Compounds were dissolved at high concentrations in Me 2 SO or water, and stock solutions were diluted for use in Krebs bicarbonate buffer, producing final Me 2 SO concentrations of up to 0.5%. Me 2 SO alone at this level did not affect DAT phosphorylation or dopamine transport, but in all experiments control samples were treated with matching concentrations of vehicle.  1-4), and the phosphorylated protein was not extracted from the cerebellum, a brain region devoid of DATs (lane 9). These results identify this phosphoprotein as DAT, with the possible caveat that the observed phosphoprotein is a contaminant that co-precipitates only in the presence of DAT.
Although DATs were phosphorylated at a basal level in untreated synaptosomes, inclusion of the phosphatase inhibitor okadaic acid (OA) during 32 PO 4 labeling resulted in a dramatic increase in the level of DAT phosphorylation (Figs. 2 and 6). This effect was dose-dependent, increasing up to 10 M, the highest concentration tested (Fig. 2, top). Time course studies showed increased levels of phosphorylated DAT by 5 min of treatment, with a maximum reached by 15-20 min (Fig. 2,  bottom). Because of this effect, 10 M okadaic acid was used in all subsequent experiments, unless otherwise indicated.
Treatment of synaptosomes with PMA in the presence of OA during 32 PO 4 labeling increased DAT phosphorylation levels a further 3-5-fold (Fig. 3). This effect was dose-dependent, increasing steeply between 0.1 and 1 M (top). The time course of PMA stimulation showed increases in phosphorylation by 5-10 min, with maximum levels reached by 15-20 min (bottom). The same relative response to PMA was produced in the absence of phosphatase inhibitors (not shown), although the lower baseline phosphorylation made this difficult to observe. The inactive phorbol ester 4␣PDD at 10 M had no effect on DAT phosphorylation (see Fig. 5). To further verify the involvement of PKC in this response, the effects of two other protein kinase C activators, (Ϫ)-indolactam V and OAG, were also examined (Fig. 4). A slight increase above basal was produced by 0.1 M (Ϫ)-indolactam V, whereas 1 and 10 M treatments increased DAT phosphorylation to comparable or slightly higher levels than PMA. OAG, a diacylglycerol analog, caused enhancement of DAT phosphorylation at 0.3 and 1 mM.
These results suggest that activation of protein kinase C results in increased phosphorylation of DAT. Additional evi-dence in support of this conclusion was obtained using the PKC inhibitors staurosporine and bisindoylmaleimide (Fig. 5). 1 M staurosporine blocked the increased DAT phosphorylation produced by 1 M PMA, whereas 10 M staurosporine inhibited even basal phosphorylation. The phosphorylation of a 20-kDa contaminant present in precipitated samples was not affected by staurosporine, indicating that the reduction of DAT phosphorylation below basal levels was not due to toxicity or nonspecific effects. PMA-induced phosphorylation of DAT was also blocked by bisindoylmaleimide, a kinase inhibitor much more specific for PKC than staurosporine. Partial inhibition of stimulated phosphorylation was observed at 100 nM, and almost complete inhibition of the stimulated increase was obtained at 1 M, the highest concentration tested.
Additional characterization of DAT dephosphorylation was done using three other phosphatase inhibitors, calyculin, microcystin, and cyclosporin A (Fig. 6). Calyculin was as effective as okadaic acid at preventing DAT dephosphorylation in the dose range of 1-10 M. Conversely, neither 10 M microcystin (Fig. 6) nor 10 M cyclosporin A (not shown) showed significant effects. Fig. 6 also shows that protein kinase A activators had no effect on DAT phosphorylation. In four separate experiments, neither 50 M forskolin in the presence of the cAMP phosphodiesterase inhibitor isobutylmethylxanthine (Fig. 6 phorylation above the okadaic acid-induced level. These compounds were also without effect in the absence of OA. Dopamine Transport-To examine possible functional consequences of DAT phosphorylation, synaptosomes were assayed for dopamine transport activity after being given phosphorylation-inducing treatments (Table I). When synaptosomes were treated with either 10 M PMA or 10 M okadaic acid, DA uptake activity was reduced to 86.9 Ϯ 3.0% and 86.5 Ϯ 3.0% of control values, respectively. When both compounds were used together the effects were additive, resulting in transport being reduced to 75.9 Ϯ 2.5% compared with the control. The treatment group values are statistically different from control values, and the combined OA and PMA treatment group is also statistically different from the individual treatment groups. The nonphorbol PKC activators (Ϫ)-indolactam V and OAG also reduced DA uptake to 92.8 Ϯ 1.5% and 79.0 Ϯ 2.6% of control values, respectively, whereas 4␣PDD had no effect (average uptake ϭ 97.8 Ϯ 3.2% of control). Staurosporine and bisindoylmaleimide, the protein kinase inhibitors that blocked PMA-induced phosphorylation, also blocked the ability of PMA to decrease transport activity (Table II).
Kinetics and dose analysis of transport reduction showed several similarities to the properties of phosphorylation responses. The PMA effect on transport occurred in the dose range of 1-10 M (Fig. 7), and the time course of uptake reduction in response to treatment with PMA plus OA occurred within 2-3 min and plateaued by 10 min (Fig. 8). Saturation analysis of dopamine uptake following treatment with PMA plus OA showed that the effect on transport was produced by a reduction in V max with no significant change in K m (Fig. 9) Alanine Transport-To assess whether the reduced transport activity in response to PMA and OA was due to an alteration of Na ϩ electrochemical gradients across membranes, we examined the effects of these compounds on Na ϩ -dependent transport of alanine (26). Synaptosomes were treated with 10 M PMA plus 10 M OA or with vehicle, and aliquots of each were assayed in parallel for transport of [ 3 H]DA or [ 3 H]alanine. In four independent experiments, alanine transport in treated tissue was not significantly different from control transport (normalized uptake ϭ 93.0 Ϯ 4.4%, p Ͼ 0.05), whereas dopamine transport activity in the treated tissue was 62.8 Ϯ 3.6% compared with control (p Ͻ 0.01). This indicates that the reduction in DA transport observed after treatment with PMA and OA is unlikely to be due to loss of transmembrane ion gradients or perturbations of membrane integrity.

DISCUSSION
DAT Phosphorylation-This report demonstrates that dopamine transporters expressed in brain undergo endogenous phosphorylation and characterizes the DAT phosphorylation and dephosphorylation properties. The appearance of phosphorylated DAT on gels is somewhat different from its appearance in photoaffinity labeling or Western blots, in which DAT is visualized as a broad, homogenous band. Although basally phosphorylated DAT had this appearance, PMA-stimulated DAT often exhibited reduced electrophoretic mobility. In some experiments, radiolabel was concentrated at the protein's trailing edge, with reduced label intensity at the leading edge ( Figs.  1 and 3; Fig. 5, left), whereas in other experiments the appearance of the stimulated band remained more homogenous but displayed a discrete upward shift relative to the basal band ( Fig. 4; Fig. 5, right). Many phosphoproteins migrate as dou-   blets or triplets on gels and undergo phosphorylation-induced interconversion between forms, although the appearance of DAT in these experiments was not as distinct as a doublet. Despite this previously undescribed appearance, the immunoprecipitation specificity controls argue against the possibility that this reduced electrophoretic mobility phosphoprotein is a contaminant rather than DAT. The significance of this observation is at present unknown. The altered migration pattern may indicate the presence of multiple states of the protein, possibly differentially or preferentially phosphorylated forms.
DAT was phosphorylated at a basal level in untreated synaptosomes, demonstrating the occurrence of spontaneous phosphate turnover on the protein. Okadaic acid and calyculin each increased the phosphorylation state of the protein severalfold, demonstrating that DAT undergoes robust in vivo dephosphorylation. The 5-10-min time course of the okadaic acid effect indicates that DATs undergo rapid and constitutive dephosphorylation. The 1-10 M dose responses to okadaic acid and calyculin may indicate that protein phosphatase 2B (calcineurin) is a major DAT phosphatase, because these compounds inhibit protein phosphatases 1 and 2A at lower concentrations (31). However, cyclosporin A, a protein phosphatase 2B-specific inhibitor (32), had no effect on DAT phosphorylation. These inconsistencies may reflect membrane permeability characteristics of the compounds and indicate the need for further work to clarify the identity of the phosphatase activities involved with DAT.
Several lines of evidence indicate that the phosphorylation level of DAT is controlled by protein kinase C. DAT phosphorylation was strongly increased by PMA, OAG, and (Ϫ)-indolactam V, three different protein kinase C activators. Phosphorylation was not affected by the inactive phorbol ester 4␣PDD, and the protein kinase C inhibitors staurosporine and bisindoylmaleimide blocked PMA-induced increases in DAT phosphorylation. These results are strong evidence that activation of PKC leads to increased phosphorylation of DAT, although it is not known if PKC phosphorylates DAT directly or induces phosphorylation via a downstream event. PKC-induced increases occurred in the presence of okadaic acid, demonstrating that the effects of OA and PKC activators on phosphorylation are additive and indicating the potential for in vivo regulation of DAT phosphorylation to occur through alterations in activity levels of kinases, phosphatases, or both. In contrast to the results with PKC activators, treatment of synaptosomes with the PKA activators forskolin and 8-Br-cAMP did not induce DAT phosphorylation, indicating the lack of involvement of PKA with DAT.
Dopamine Transport-The possibility that dopamine transport activity is regulated by phosphorylation of DAT is indicated by the finding that dopamine uptake in synaptosomes was reduced by all treatments that promoted DAT phosphorylation. PMA and okadaic acid each reduced dopamine transport to about 87% of control values, and use of both compounds together resulted in further reduction to about 76% of control values, mimicking the additivity of these compounds on DAT phosphorylation. Uptake was also reduced by (Ϫ)-indolactam V and OAG, 4␣PDD had no effect on uptake, and staurosporine and bisindoylmaleimide were both able to block PMA-induced reductions in transport activity. The reduction in uptake occurred with dose and time courses similar to those for phosphorylation and was produced by a reduction in transport V max with no apparent change in the K m for dopamine. The close correlation of phosphorylation and transport characteristics strongly suggests that dopamine transport activity is regulated by DAT phosphorylation.
Results similar to these were found when mouse striatal synaptosomes were treated with OA, PMA, and diacylglycerol analogs (18). However, PKA activators did not affect DA transport, consistent with our finding that PKA activators do not induce DAT phosphorylation. The possibility that the reduced dopamine transport obtained in these studies was a nonspecific effect produced by an alteration of transmembrane Na ϩ gradients has been addressed by several different approaches. Our results show that in synaptosomes and rDAT-LLC-PK 1 cells (16), Na ϩ -dependent alanine transport was not altered by PMA and OA treatments that substantially reduced DA uptake. In addition, PMA does not affect synaptosomal viability or alter the activity of synaptosomal Na ϩ -K ϩ ATPase, which controls transmembrane Na ϩ gradients (18). These results suggest that it is unlikely that the phosphorylation activators examined in these studies are inducing alterations in Na ϩ electrochemical gradients or affecting membrane integrity. The effects of these compounds on DA uptake therefore appear to be specific for DA transport and physiologically meaningful.
Although these results strongly indicate that DAT activity is regulated by phosphorylation, the mechanisms underlying this relationship remain to be elucidated. Possible modes of action include direct effects on substrate translocation or changes in DAT surface expression as has been found for heterologously expressed ␥-amino butyric acid and serotonin transporters (22,25). Additional aspects of DAT function such as ion flux or binding of cocaine or other uptake blockers may also be affected by phosphorylation.
Implications-While the pathways controlling DAT phosphorylation in vivo remain to be elucidated, the finding that the diacylglycerol analog OAG stimulates DAT phosphorylation indicates the feasibility for involvement of phospholipase Ccoupled receptors in phosphorylation of DAT, and the rapid kinetics of DAT phosphorylation and dephosphorylation are compatible with receptor-mediated actions. One possible control mechanism is by feedback through presynaptic dopamine autoreceptors. Although dopamine receptors have classically been characterized as coupling to cAMP production (33), recent studies indicate that dopamine receptors can couple either positively (34,35) or negatively (36) with PKC pathways. The latter may be compatible with the observation that D 2 receptor agonists increase striatal dopamine transport (37). If activation of D 2 receptors reduces PKC activity, DAT phosphorylation would be reduced and DA transport activity would be increased. Coupling of dopamine receptors to DAT phosphorylation would also provide a mechanism for cocaine involvement in regulation of transport activity via cocaine's effects on synaptic dopamine levels. It is also possible that other phospholipase C-coupled receptors may regulate DAT phosphorylation and that this may be a convergence point for multiple signaling pathways.
Although the transport of many neurotransmitters is affected by protein kinase activators, until recently there has been little direct evidence that phosphorylation of transporter proteins mediates these effects. The most well documented example of neurotransmitter transporter phosphorylation is DAT, which has now been characterized as a phosphoprotein both in neuronal tissue and heterologous expression systems (16,38). The phosphorylation of DATs in LLC-PK 1 cells displays activator, inhibitor, and kinetic characteristics that parallel those shown here for neuronal DATs, indicating that heterologous expression systems provide valid models for examining DAT phosphorylation and that results from studies such as deletion or site-directed mutagenesis using transfected DATs will be relevant to the neuronal form of the protein.
Regulation of DA transport activity by phosphorylation would provide neurons with a previously unappreciated mechanism for fine temporal and spatial control of synaptic dopamine concentrations. Such functional regulation could have profound effects on the intensity and duration of dopaminergic synaptic transmission, actions of psychostimulant drugs, and mechanisms of neurotoxicity and neurodegeneration.