Abstract
A brain dopamine receptor that modulates phosphatidylinositol (PI) metabolism via the activation of phospholipase Cβ (PLCβ) has been described previously. The present study aims to define the downstream signaling cascade initiated by the PI-linked dopamine receptor. Incubation of rat brain frontal cortical slices with 6-chloro-7,8-dihydroxy-3-methyl-1-(3-methylphenyl)-2,3,4,5-tetrahydro-1H-3-benzazepine (SKF83959), a recently identified selective agonist of the PI-linked D1-like dopamine receptor, elicited transient time- and dose-dependent stimulations of cyclin-dependent kinase 5 (cdk5) and calcium/calmodulin-dependent protein kinase II (CaMK II) activities. The stimulation of these kinases is blocked by 20 μM R-(+)-7-chloro-8-hydroxy-3-methyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine (SCH23390) or the PLCβ antagonist 1-[6-[[17β-methoxyestra-1,3,5(10)-trien-17-yl]amino]hexyl]-1H-pyrrole-2,5-dione (U-73122) and is attenuated by the protein kinase inhibitor calphostin C or by the intracellular calcium chelator BAPTA, indicating that SKF83959 stimulates cdk5 and CaMK II activities via a PI-linked D1-like dopamine receptor, and PLCβ and is dependent on protein kinase C and calcium. Although cdk5 and CaMK II are physically associated in native brain tissue, no change in this association was observed in response to SKF83959 stimulation or to the inhibition of either cdk5 by roscovitine or of CaMK by 2-[N-(2-hydroxyethyl)]-N-(4-methoxybenzenesulfonyl)]amino-N-(4-chlorocinnamyl)-N-methylbenzylamine) (KN93), suggesting that SKF83959-mediated stimulation of cdk5 or CaMK II is independent of the other kinase and that the association of the two kinases is not modulated by change of kinase activity. Moreover, we found that cdk5 phosphorylates dopamine and cAMP-regulated phosphoprotein at Thr75, whereas CaMK II is responsible for the activation of cAMP response element-binding protein in response to SKF83959 stimulation. The present data provide the first insight into the signaling mechanism for the PI-linked dopamine receptor. This information, in turn, may help in exploring the functional consequences of stimulation of this brain receptor.
Dopamine, one of the major neurotransmitters in brain, exerts its action via dopamine receptors. So far, at least five dopamine receptors have been cloned. It is known that the receptors in the D1 receptor family (D1, D5) couple to Gs protein and activate the production of cAMP; the D2 receptor family (D2-4) couples to Gi/G0 protein and inhibits cAMP production (Missale et al., 1998). Thus, these receptors regulate the levels of cAMP, which in turn regulates the activity of protein kinase A (PKA). In addition to the cAMP/PKA pathway, recent evidence indicates that activation of dopamine receptors also elevates intracellular calcium via the activation of phospholipase Cβ (PLCβ), which hydrolyzes phosphatidylinositol (PI) into diacylglycerol and inositol trisphosphate (IP3) (Felder et al., 1989; Undie and Friedman, 1990; Pacheco and Jope, 1997). IP3 is an important second messenger that elicits the release of calcium from intracellular stores. The dopamine receptor that activates the PLC/IP3 pathway has been called the PI-linked dopamine receptor. This receptor seems to pharmacologically resemble the D1 receptor family because it is stimulated selectively by agonists and is inhibited by antagonists of the D1 receptor (Felder et al., 1989; Friedman et al., 1997). SKF83959 is a recently identified selective agonist for the PI-linked dopamine receptor (Panchalingam and Undie, 2001; Jin et al., 2003). However, the functional implications of this novel PI-linked dopamine receptor pathway in brain are completely unknown.
The regulation of intracellular calcium content by IP3 is critical in mediating neuronal responses. Ca2+ and calcium-sensing proteins such as calmodulin form functional complexes that interact with various proteins, thus modulating their functions (Soderling, 2000). Ca2+/calmodulin regulates CaMK, a serine/threonine protein kinase. CaMK phosphorylates cAMP response element-binding protein (CREB), a critical transcriptional factor that is involved in synaptic plasticity and in mediating learning and memory processes (Bito et al., 1996). Thus, it is of interest to investigate whether the PI-linked dopamine receptor regulates the CaMK/CREB pathway.
Recent information indicates that cyclin-dependent kinase 5 (cdk5) is an important kinase in the modulation of dopamine receptor signals (Bibb et al., 1999; Nishi et al., 2000). cdk5 was shown to phosphorylate the dopamine and the cAMP-regulated phosphoprotein (DARPP-32) at Thr75. Phospho-DARPP-32 (Thr75) is an inhibitor of protein kinase A. Activation of the D1 dopamine receptor has been shown to decrease the level of phospho-DARPP-32 (Thr75), whereas activation of the D2 dopamine receptor increases its level (Nishi et al., 2000). cdk5 exhibits exclusively high activity in neurons; it has been shown that cdk5 is required for neuronal development and in regulating neuronal migration, process elongation, cortical lamination, and synaptic plasticity (Dhavan and Tsai, 2001; Smith et al., 2001). Abnormal cdk5 activity has been linked to neurodegenerative diseases such as Alzheimer's disease (Dhavan and Tsai, 2001). This kinase was also shown to be associated with calcium regulation (Liu et al., 2001; Dhavan et al., 2002). The interaction between cdk5 and CaMK II was reported recently (Dhavan et al., 2002). However, the potential role of the PI-linked dopamine receptor in the regulation of cdk5 and CaMK II remains unknown.
The present work demonstrates that stimulation of the PI-linked dopamine receptor by SKF83959 induces the transient activation of cdk5 and CaMK II. Activation of cdk5 and CaMK II is mediated via PLCβ and requires the activation of PKC and the intracellular release of calcium. Receptor activation increases the phosphorylation of DARPP-32 (Thr75), a known substrate of cdk5, whereas activation of CaMK II results in the activation of CREB. This study therefore provides new insights into the signaling pathways that transduce PI-linked dopamine receptor signals. These findings may facilitate the unraveling of the functional role of the PI-linked dopamine receptor in brain.
Materials and Methods
Materials. SKF83959 was kindly provided by the National Institute of Mental Health synthesis program (Menlo Park, CA). R(+)SCH23390 hydrochloride, SKF81297, and SKF83959 were purchased from Sigma/RBI (Natick, MA). PD98059, KN92, and KN93 were from BIOMOL Research Laboratories (Plymouth Meeting, PA). Antibodies to cdk5, CaMKs, and horseradish peroxidase-linked secondary antibodies were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). U-73122, U-73343, calphostin C, and roscovitine were obtained from Calbiochem (San Diego, CA). CaMK assay kit was obtained from Upstate Biotechnology (Charlottesville, VA). Proteinase inhibitors were from Sigma-Aldrich (St. Louis, MO). Anti-phospho-(Thr75 and Thr34) DARPP-32, were from New England Biolabs (Beverly, MA). Electrophoresis reagents were obtained from Bio-Rad (Hercules, CA). [γ-32P]ATP (3000 Ci/mmol) was purchased from PerkinElmer Life and Analytical Sciences (Boston, MA). Other reagents were purchased from standard laboratory suppliers.
Slice Preparation and Treatment. Brain slices of rat frontal cortex (FCX) were prepared as described previously (Zhen et al., 2001a; Dhavan et al., 2002) using a McIllwain tissue chopper (Brinkmann Instruments, Westbury, NY). The prisms were washed in oxygenated Krebs-Ringer buffer. After equilibration at 37°C for 30 min, the slices were subsequently incubated with SKF83959 or other reagents. Reactions were terminated by cooling on ice followed by centrifugation. The tissues were sonicated and lysed in lysis buffer containing 20 mM Tris, pH 7.8, 150 mM NaCl, 50 M NaF, 2 mM EGTA, 1 mM EDTA, 0.5 mM β-glycerophosphotate, 1 mM vanadate, 1% Triton X-100, 1 mM phenylmethysulfoxide, and proteinase inhibitor cocktail. The protein content of the supernatant was determined by the Bradford method (Bio-Rad). Aliquots of supernatants (300 μg) were immunoprecipitated with anti-cdk5, anti-CaMK II or IV, and anti-ERK2 antibody, respectively, using a method published previously, with minor modification (Zhen et al., 2001b). The immunoprecipitates were collected and used for kinase activity assays or in some cases for analyzing the physical association between cdk5 and CaMK II, as described below.
In Vitro Immune Complex Kinase Assays. The immune complex was washed three times with lysis buffer and twice with the corresponding kinase assay buffer (see below). The cdk5 kinase assay was performed in 50 mM HEPES, pH 7.4, containing 10 mM MgCl2, 1 mM dithiothreitol, 100 μM [γ32P]ATP (5 μCi), and substrate peptide (Dhavan et al., 2002) for 20 min at 30°C. ERK kinase activity was assessed for 20 min at 30°C in the presence of 50 μM [γ32P]ATP (5 μCi) and 0.2 mg/ml myelin basic protein, as described previously (Zhen et al., 2001a). The reaction was stopped on ice, and 20 μl of supernatant was spotted onto P81 paper and washed extensively in 0.425% phosphoric acid. Radioactivity incorporated into peptide was determined by scintillation counting. For CaMK assay, immunocomplexes of CaMK II or IV were washed with kinase assay buffer, and the reaction was conducted in the presence of 10 μM [γ-32P]ATP using peptide-γ as substrate (Kasahara et al., 2001), according to the manufacturer's instruction.
Phosphorylated Proteins and Immunoblot Analysis. The lysate was boiled in preparation buffer, and 30 μg of protein was loaded onto 12% SDS-PAGE. The proteins were separated electrophoretically and transferred to nitrocellulose membranes. The membranes were blocked overnight at 4°C with 10% (w/v) fat-free dry milk in 0.1% Tween-20 phosphate-buffered saline (TBS) followed by incubation with anti-phospho-CREB (1:2000) for 2 h. The membranes were washed and then incubated for 1 h with species-specific horseradish peroxidase-conjugated secondary IgG antibody (1:5000-1:10,000 dilution) in 0.1% TBS. The membranes were washed once with 0.3% TBS for 20 min followed by four 10-min washes with 0.1% TBS, and the signals were visualized by Supersignal (Pierce, Rockford, IL). For the analysis of DARPP-32, the brain tissues were sonicated and prepared by boiling in 1% SDS-contained sample preparation buffer as described previously (Snyder et al., 1998). Equal amounts of homogenate protein were loaded onto 12% SDS-PAGE, and the membranes were then probed with anti-phospho-DARPP-32 at Thr34 (1:750) or anti-phospho-DARPP-32 at Thr75 antibody, respectively. For assessing the association of Cdk5 and CaMK II, 300 μg of lysate protein was incubated with 3 μg of anti-cdk5 antibody and was incubated with shaking for 2 h before the addition of Protein A/G PLUS (Santa Cruz Biotechnology, Santa Cruz, CA). Immunoprecipitates were then collected and loaded onto 12% SDS-PAGE. The proteins were separated electrophoretically and transferred to nitrocellulose membranes. Membranes were incubated with anti-CaMK II antibody, or vice versa, CaMK II precipitates were separated by SDS-PAGE, and anti-cdk5 antibody was used to detect the association between the kinases.
Measuring Brain PI Hydrolysis in Brain Slices in Response to SKF83959 Challenge. Brain slices were incubated with Kreb's bicarbonate buffer for 45 min with buffer replacement every 15 min. Tissues were then labeled with [myo-3H]inositol (10 μCi) for 1 h in the presence of 20 mM lithium chloride. Tissues were then incubated with SKF83959 or antagonist SCH23390 for a designated time. The reaction was stopped with 300 μl of 1 M trichloroacetic acid. The 500-μl sample of trichloroacetic acid extraction was transferred to 2-ml Eppendorff tubes. With the addition of 10 mM EDTA and freon/tri-n-ethyllamine (3:1, v/v), samples were thoroughly mixed and centrifuged. A 500-μl sample of the upper layer was taken in a glass tube containing 0.5 M NaHCO3 and 1 ml of distilled water. After pass to the AG-1 × 8 column (0.75 ml), samples were washed extensively. Total immunoprecipitates were then eluted and collected as described previously (Challiss et al., 1988). Results are expressed as counts per minute per milligram of protein.
Cell Culture and Treatment. PC-12 cells that were transfected with D1 dopamine receptor were prepared as described previously (Zhen et al., 2001b). Cells were seeded in 10-cm dishes. After they reached 80 to 90% confluence, cells were cultured in low serum medium for 4 h before incubating with SKF83959 or other drugs for the indicated times. Cells were then harvested, and lysates were prepared for kinase assay.
Data Analysis. Data are expressed as mean ± S.E.M. and analyzed by analysis of variance followed by Newman-Keuls test unless otherwise indicated. Statistical significance was considered at p < 0.05.
Results
SKF83959 Stimulates cdk5 Activity via D1-Like Dopamine Receptor in Rat Brain Slices. FCX slices were incubated with 20 μM SKF83959 for 2, 5, 10, 20, and 30 min, and cdk5 activity was then measured in tissue lysates. SKF83959 induced a time-dependent stimulation of kinase activity (Fig. 1A). SKF83959-induced activation of cdk5 reached a peak within 2 min (2.37 ± 0.21-fold, n = 3) and returned to basal levels at 10 min, indicating that SKF83959 activates cdk5 transiently. A dose-response analysis indicates that SKF83959 activated cdk5 activity at 10 μM, and peak activation occurred at 20 μM (Fig. 1B). Because SKF83959 has been found to be a selective agonist of the PI-linked D1 dopamine receptor, we therefore checked whether activation of cdk5 is also mediated by the receptor. FCX slices were preincubated with 20 μM of the selective D1 receptor antagonist SCH23390 for 20 min before the stimulation of SKF83959. As shown in Fig. 1C, SCH23390 almost completely blocked SKF83959-stimulated cdk5, suggesting that the stimulation of cdk5 by SKF83959 is mediated via a D1-like dopamine receptor. Moreover, we found that another D1 receptor antagonist, SKF83566 (100 nM), also blocked SKF83959- and SKF81297-stimulated cdk5 activity (Fig. 1 D), whereas prazosin (1 μM), spiperone (10 μM), and mesulergine (10 μM) did not (data not shown).
Activation of cdk5 by SKF83959 Is Mediated by PLCβ and Is Dependent on PKC and Intracellular Calcium. Because SKF83959 is a selective agonist for the PI-linked D1-like dopamine receptor and this receptor is known to couple to PLCβ/PKC/IP3 pathway, we tested whether this signaling pathway is involved in cdk5 activation. Brain slices were preincubated with U-73122 (10 μM), U-73343 (10 μM), calphostin C (1 μM), or BAPTA (10 μM) for 20 min before the addition of SKF83959, and cdk5 activity was determined. As shown in Fig. 2A, the putative phospholipase C inhibitor U-73122 blocked SKF83959-induced cdk5 activation. In contrast, its inactive analog, U-73343, did not interfere with activation of this kinase. Moreover, inhibition of PKC by calphostin C or intracellular calcium release by BAPTA also attenuated cdk5 activation, whereas PKA inhibitor had little effect (Fig. 2B). These results suggest that SKF83959-mediated activation of cdk5 is mediated via PLCβ and is dependent on PKC and Ca2+ but not on PKA. Phosphorylation of PLCβ by PKC has been shown to inhibit PLCβ activity (Ryu et al., 1990; Litosch, 1996). We therefore tested whether PKC activation influences SKF83959-induced cdk5 activation by preincubated FCX slices with 1 μM PMA before stimulating with SKF83959, carbachol, or NaF. Carbachol and NaF are known to stimulate the PLCβ pathway, therefore serving as positive controls. As expected, both carbachol and NaF stimulated cdk5 activity in FCX slices, whereas activation of PKC by PMA significantly reduced carbachol- or NaF-stimulated cdk5 activation (Fig. 3). Likewise, the SKF83959-elicited response was also inhibited by prior PKC stimulation (Fig. 3). Taken together, the present data demonstrate that stimulation of the PI-linked D1 dopamine receptor by SKF83959 activates cdk5 through PLCβ-mediated PKC activation and elevation in intracellular calcium.
SKF83959 Activates CaMK II Activity in a Time- and Dose-Dependent Manner. Calcium has been shown to play an important role in regulating CaMK activity. To study the effect of SKF83959 on CaMK activity, FCX slices were treated with SKF83959, and CaMK II and IV activities were measured in anti-CaMK II or anti-CaMK IV immunoprecipitates. As shown in Fig. 4, A and B, SKF83959 induced a transient activation of CaMK II in FCX slices. Similar to the result with cdk5 shown in Fig. 1, the maximal stimulation was achieved at 2 min of agonist stimulation; a return to basal level occurred within 10 min. The dose-response curve indicates that stimulation of CaMK II is observed at 10 μM of SKF83959, and it reaches maximum stimulation at 20 μM. However, significant changes in CaMK IV activity were not detected by this treatment (data not shown), indicating that SKF83959 selectively stimulates CaMK II activity. Similar, we did not detect significant stimulation of ERK activity in the present experimental condition (data no shown). As expected, SCH23390 blunted the drug-induced activation of CaMK II. Another D1 receptor antagonist SKF83566 also blocked the SKF83959-induced CaMK II activation, whereas D2, α1, or 5-hydroxytryptamine2A/2C receptor antagonist did not (Fig. 4C), It is interesting to note that stimulation of CaMK II was also mediated through PLCβ and was dependent on PKC and intracellular calcium because U-73122, calphostin C, or BAPTA blocked the activation of CaMK II (Fig. 4D). Thus, it seems that SKF83959-mediated CaMK II activation shares upstream mechanisms with cdk5 activation in FCX slices.
SKF83959 Did Not Alter cdk5 and CaMK II Activity in D1-Expressing PC-12 Cells. One may argue that classic D1 dopamine receptor contributes to the activation of cdk5 and CaMK II because SKF83959 was originally shown to have a high affinity for the D1 dopamine receptor, and the FCX is enriched with this receptor. We, therefore, tested whether cdk5 and CaMK II activation in brain slices is mediated by the D1 dopamine receptor. PC-12 cells that stably express the D1 dopamine receptor were incubated with either 1 μM SKF83959 or with the selective D1 receptor agonist SKF81297, and kinase activities were assessed. As shown in Fig. 5A, activation of D1 receptors by SKF81297 did not affect cdk5 or CaMK II activity although this treatment resulted in a dramatic stimulation on cAMP (data not shown; Jin et al., 2003). As expected, SKF83959 also did not activate these kinases in the cells (Fig. 5B). However, carbachol induced time-dependent stimulation on the CaMK II activity in the cells (Fig. 5C). Our previous study demonstrated that both SKF83959 and SKF81297 do not stimulate PI hydrolysis in these cells (Jin et al., 2003) but do so in brain slices (Fig. 5D) (Jin et al., 2003), indicating that classic D1 dopamine receptors are not involved in SKF83959-mediated activation of cdk5 and CaMK II in brain slices.
Activation of cdk5 and CaMK II by SKF83959 Is Independent of Each Other. As shown above, the activations of cdk5 and CaMK II by SKF83959 share a similar time course. A previous study also indicated that the cdk5 activators p35 and p39 interact with the α-subunit of Ca2+/CaMK II. We therefore tested whether the activation of the two kinases by SKF83959 is interrelated. Brain slices were pre-treated with 50 μM KN93, a selective inhibitor of CaMK II, or its negative control, KN92, before the addition of SKF83959. Inhibition of CaMK II did not alter SKF83959-induced cdk5 stimulation (Fig. 6A). On the other hand, inhibition of cdk5 by roscovitine also did not affect the activation of CaMK II by SKF83959 (Fig. 6B). Moreover, we have observed a physical association between cdk5 and CaMK II; however, this association was not altered significantly by change of either cdk5 or CaMK II activity (Fig. 6C), suggesting that cdk5 or CaMK II activity may not be required for their association. It is therefore clear that although SKF83959-induced activation of cdk5 and CaMK II share similar upstream pathways, the activations of the two kinases are regulated independently.
Activation of CaMK II or cdk5 Differentially Regulates Phosphorylation of Distinct Substrates in Response to SKF83959. CaMKs are known to phosphorylate CREB at Ser133, and this activates CREB. We first determined whether CREB phosphorylation is altered by SKF83959 in FCX slices. As shown in Fig. 7A, SKF83959 activated CREB transiently; phosphorylation of CREB (Ser133) increased at 2 min, reached maximal at 5 min, and returned to control level within 10 min. To detect the upstream mechanism for CREB activation, brain slices were preincubated with SCH23390, U-73122, and U-73343 for 20 min before the addition of SKF83959. Activation of CREB was blunted by SCH23390 or U-73122 but not by U-73343 (Fig. 7B), indicating that CREB activation by SKF83959 is mediated by the PI-linked dopamine receptor via activation of PLCβ. We further found that KN93, a selective inhibitor of CaMKs, also blocked SKF83959-stimulated CREB phosphorylation, whereas KN92 and roscovitine did not alter the drug-induced activation of CREB (Fig. 7C), indicating that CaMK II is required for CREB activation by SKF83959, whereas cdk5 is not.
We next examined the potential target of cdk5 activation by SKF83959. DARPP-32 phosphorylation in response to SKF83959 stimulation in FCX slices was analyzed using specific anti-phospho-DARPP-32 antibodies (Thr34 and Thr75). As shown in Fig. 7D, SKF83959 induced a significant increase in the level of phospho-DARPP-32 (Thr75), whereas no significant change in phospho-DARPP-32 (Thr34) was observed. Moreover, roscovitine, a selective inhibitor for cdk5, largely attenuated SKF83959-induced phosphorylation of DARPP-32 at Thr75, indicating that cdk5 activation contributes to SKF83959-mediated Thr75 phosphorylation of DARPP-32. The results demonstrate that activation of the PI-linked dopamine receptor which stimulates the PLC/PKC/IP3 pathway results in activation of cdk5 and CaMK II, and these regulate the phosphorylation of DARPP-32 and CREB, respectively.
Discussion
The present studies demonstrate that SKF83959 activates cdk5 and CaMK II in a time- and dose-dependent manner. The stimulations are transient and return to basal levels within 10 min. Activation of cdk5 and CaMK II seems to be mediated by the PI-linked dopamine receptor in brain. We further demonstrated that both cdk5 and CaMK II activation by SKF83959 is mediated by PLCβ and is dependent on PKC and intracellular calcium release, because inhibition of PKC or chelation of intracellular released calcium abolished drug-induced kinase activations. Although the activation of cdk5 and CaMK II share the same upstream pathway and are physically associated, inhibition of cdk5 did not interrupt SKF83959-induced activation of CaMK II, and cdk5 activation was not affected by CaMK inhibition, indicating that the activations of the kinases are independent of each other. We also found that SKF83959 induces the phosphorylation of DARPP-32 (Thr75) and of CREB that are associated with cdk5 and CaMK II activation. To our knowledge, this is the first report that describes the detailed signaling cascades that are activated by the PI-linked D1 dopamine receptor.
Although a number of studies have found that Gαq couples the dopamine receptor to PLC (Felder et al., 1989; Undie and Friedman, 1990; Pacheco and Jope, 1997), the recent identification of a selective agonist, SKF83959 (Panchalingam and Undie, 2001; Jin et al., 2003), for this pathway has allowed the detailed investigation of signal transduction and function of this dopamine receptor. The present data demonstrate that activation of the PI-linked D1 dopamine receptor differentially regulates the cdk5 and CaMK II pathways. The conclusion is reached from the following observations: 1) the selective D1 dopamine receptor antagonist SCH23390 blocked SKF83959-stimulated cdk5 and CaMK II activities; 2) inhibition of PLCβ attenuated SKF83959-stimulated cdk5 and CaMK II activation; 3) SKF83959 did not stimulate PI hydrolysis (Jin et al., 2003), nor did it activate cdk5 or CaMK II (Fig. 5) in D1A-expressing PC-12 cells, indicating that the stimulation of cdk5 or CaMK II by SKF83959 in brain slices is not mediated by the classic Gαs-coupled D1A dopamine receptor; and 4) the D1 dopamine receptor agonist SKF81297 did not stimulate cdk5 and CaMK II activities in D1A/PC-12 cells, although it induced a dramatic stimulation of cAMP/PKA, further confirming that activation of the classic Gαs-coupled D1 receptor is not responsible for SKF83959-stimulated cdk5 and CaMK II activation in brain slices.
cdk5, the catalytic component of the kinase, and its neuron-specific activator p35 have received great deal of attention in physiology and pathology. cdk5 is a proline-directed serine/threonine kinase and is enriched in neurons. It forms a complex with p35 in neurons, and the association is essential for its activation. cdk5 has been demonstrated to play an important role in neurite outgrowth and migration (Dhavan and Tsai, 2001) and in neuroplasticity (Fischer et al., 2003). Although neuronal cytoskeletal proteins such as Tau were originally identified as substrates for cdk5, suggesting that the kinase may be involved in the pathology of neurodegenerative diseases (Smith et al., 2001), there are other intracellular substrates that were identified recently. Among them, DARPP-32 is of particular interest because it is considered to be a central signaling molecule in the dopamine signaling system of the brain (Greengard et al., 1999). Stimulation of both D1- and D2-like dopamine receptors was shown to regulate DARPP-32 phosphorylation. Phosphorylation of DARPP-32 at Thr75 converts DARPP-32 into a potent PKA inhibitor and reduces phosphorylation of DARPP-32 at Thr34 (Bibb et al., 1999). The latter is an inhibitor of protein phosphatase 1. Dopamine D1 receptor stimulation decreases the striatal phospho-DARPP-32 Thr75, whereas D2 receptor stimulation increases phosphorylation at this site probably via alteration in protein phosphatase-2A (Nishi et al., 2000). The present data demonstrate that activation of the PI-linked dopamine receptor stimulates cdk5 activity in frontal cortex, and this activation is responsible for SKF83959-induced change in levels of phospho-DARPP-32 Thr75 in this brain region. Thus, our results reveal additional mechanisms for dopamine receptor-regulated DARPP-32/protein phosphatase 1 pathway in brain. It is interesting to note that stimulation of metabotropic glutamate receptors also results in a transient activation of cdk5 in brain slices. This activation of cdk5 is associated with metabotropic glutamate receptor agonist-mediated stimulation of phospho-DARPP-32 at Thr75 (Liu et al., 2001), thus implying that activation of Gq-linked G-protein coupled receptors may share a common cdk5 pathway that regulates DARPP-32.
Brain dopamine systems have been considered to play a critical role in synaptic plasticity associated with learning and memory (Jay, 2003). Recent evidence indicates that many signaling molecules may be involved in transducing dopamine signals that are involved in neuroplasticity and in cognitive processes. These include mitogen-activated protein kinases, PKA, and CaMKs (Frey et al., 1993; Kornhauser and Greenberg, 1997; Zhen et al., 1998, 2001a; Jay, 2003). These protein kinases are able to phosphorylate CREB at Ser133, and CREB is considered to be a central transcription factor in the formation of memory. The present data demonstrate that the PI-linked dopamine receptor is involved in regulating the activation of CREB via CaMK II, indicating the potential role of this receptor in neuroplasticity.
Acknowledgments
We thank Regine Brioche and Rochelle Ledgister for excellent technical assistance.
Footnotes
-
This work was supported by U.S. Public Health Service grant DA11029 and the City University of New York Incentive Collaborative grant.
-
ABBREVIATIONS: PKA, protein kinase A; PLCβ, phospholipase Cβ; cdk5, cyclin-dependent kinase 5; CaMK II, calcium/calmodulin-dependent protein kinase II; PI, phosphatidylinositol; PKC, protein kinase C; DARPP-32, dopamine and cAMP-regulated phosphoprotein; CREB, cAMP-responsive element-binding protein; IP3, inositol trisphosphate; FCX, frontal cortex; PAGE, polyacrylamide gel electrophoresis; ERK, extracellular signal-regulated kinase; TBS, Tween-20 phosphate-buffered saline; PMA, phorbol 12-myristate 13-acetate; SKF83959, 6-chloro-7,8-dihydroxy-3-methyl-1-(3-methylphenyl)-2,3,4,5-tetrahydro-1H-3-benzazepine; SCH23390, R-(+)-7-chloro-8-hydroxy-3-methyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine; U-73122, 1-[6-[[17β-methoxyestra-1,3,5(10)-trien-17-yl]amino]hexyl]-1H-pyrrole-2,5-dione; BAPTA, 1,2-bis(2-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid; KN93, 2-[N-(2-hydroxyethyl)]-N-(4-methoxybenzenesulfonyl)]amino-N-(4-chlorocinnamyl)-N-methylbenzylamine); SKF81297, R-(+)-6-chloro-7,8-dihydroxy-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzapine; PD98059, 2′-amino-3′-methoxyflavone; KN92, 2-[N-(4-methoxybenzenesulfonyl)]amino-N-(4-chlorocinnamyl)-N-methylbenzylamine; U-73343, 1-[6-((17β-3-methoxyestra-1,3,5(10)-trien-17-yl)amino)hexyl]-2,5-pyrrolidinedione; SKF83566, (±)-7-bromo-8-8-hydroxy-3 methyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine; H89, N-[2-(4-bromocinnamylamino)ethyl]-5-isoquinoline; SKF38393, 1-phenyl-7,8-dihydroxy-2,3,4,5-tetrahydro-1H-3-benzazepine hydrochloride.
- Received May 4, 2004.
- Accepted July 29, 2004.
- The American Society for Pharmacology and Experimental Therapeutics