Abstract
The dopamine D4 receptor in prefrontal cortex (PFC) plays a key role in normal mental functions and neuropsychiatric disorders. However, the cellular mechanisms and physiological actions of D4 receptors remain elusive. In this study, we found that activation of D4 receptors in PFC exerts a complex regulation of Ca2+/calmodulin-dependent protein kinase II (CaMKII), a multifunctional enzyme critically involved in synaptic plasticity that is fundamental for cognitive and emotional processes. In PFC slices with high neuronal activity, application of the D4 receptor agonist [4-phenylpiperazinyl)-methyl]benzamide (PD168077) produced a potent reduction of the CaMKII activity, whereas in PFC slices with low neuronal activity, PD168077 caused a marked increase of the CaMKII activity. The D4 up-regulation of CaMKII activity was through the stimulation of phospholipase C pathway and elevation of intracellular Ca2+ via ionsitol-1,4,5-triphosphate receptors. These results reveal a bidirectional regulation of CaMKII activity by PFC D4 receptors in response to changes in neuronal activity, and a nonclassic signaling pathway underlying the D4 up-regulation of CaMKII activity. This modulation provides a unique and flexible mechanism for D4 receptors to regulate CaMKII activity, which could lead to dynamic regulation of many targets of CaMKII by D4 receptors.
Prefrontal cortex (PFC) is a brain region critically involved in the control of cognition, reasoning, perception, and emotion (Goldman-Rakic, 1995). Dysfunction of PFC has been implicated in a variety of neuropsychiatric disorders, including schizophrenia (Andreasen et al., 1997; Lewis and Lieberman, 2000). PFC functions are highly influenced by the dopaminergic input from the ventral tegmental area (Brozoski et al., 1979; Berger et al., 1988). Aberration of the dopaminergic system in PFC is considered a major factor in the pathophysiology of schizophrenia (Grace, 1991; Carlsson et al., 2001).
Dopamine D4 receptors are highly enriched in PFC neurons (Mrzljak et al., 1996; Wedzony et al., 2000). The elevated D4 receptors found in the PFC of patients with schizophrenia (Seeman et al., 1993) and the high affinities of D4 receptors for antipsychotic drugs (Van Tol et al., 1991; Kapur and Remington, 2001) suggest that D4 receptors may be critically involved in PFC functioning and neuropsychiatric disorders (Oak et al., 2000). In agreement with this, D4 receptor antagonists ameliorate cognitive deficits caused by the psychotomimetic drug phencyclidine (Jentsch et al., 1997; Jentsch and Roth, 1999). Moreover, mice lacking D4 receptors exhibit supersensitivity to psychomotor stimulants (Rubinstein et al., 1997) and reduced exploration of novel stimuli (Dulawa et al., 1999). To understand how D4 receptors regulate PFC functions under normal and pathological conditions, we need to determine the potential targets of D4 receptors that are critically involved in the regulation of cognitive and emotional processes subserved by PFC.
One potential target of such for D4 receptors is the Ca2+/calmodulin-dependent protein kinase II (CaMKII). CaMKII is highly expressed in the forebrain and concentrated at postsynaptic densities at glutamatergic synapses (Kennedy et al., 1983). This ideal position allows the multifunctional enzyme to play a central role in regulating several key postsynaptic targets required for synaptic plasticity that is integral for learning and memory (Malenka and Nicoll, 1999; Soderling et al., 2001). Mice with deficient CaMKII exhibit impairments in spatial learning (Silva et al., 1992) and permanent memory retention (Frankland et al., 2001). In addition to the cognitive deficit, these CaMKII mutant mice exhibit a spectrum of behavioral abnormalities associated with emotional disorders, including a decreased fear response and an increase in defensive aggression (Chen et al., 1994).
The function of CaMKII is shaped by its autoregulation and subcellular localization (Hudmon and Schulman, 2002). CaMKII is autophosphorylated at Thr286 when the enzyme is activated in the presence of Ca2+/calmodulin, leading to the appearance of a sustained, Ca2+-independent activity (Miller and Kennedy, 1986). This autoregulatory property enables CaMKII to act as a molecular memory device to detect synaptic activity and to coordinate and execute Ca2+ signal transduction. CaMKII also dynamically alters its subcellular distribution after N-methyl-d-aspartate receptor stimulation through a mechanism involving Ca2+/calmodulin binding and autophosphorylation (Shen and Meyer, 1999). Given the convergent involvement in PFC functions for D4 receptors and CaMKII, we sought to understand their interactions by examining the D4 regulation of CaMKII activity in this study.
Materials and Methods
Western Blot Analysis. PFC slices were prepared as described previously (Gu et al., 2003). After treatment with different agents as indicated in the text, equal amounts of protein from slice homogenates were separated on 7.5% acrylamide gels and transferred to nitrocellulose membranes. The blots were blocked with 5% nonfat dry milk for 1 h at room temperature. Then the blots were incubated with the anti–Thr286-phosphorylated α-CaMKII antibody (Santa Cruz, 1:2000) for 1 h at room temperature. After being rinsed, the blots were incubated with horseradish peroxidase-conjugated antirabbit antibodies (Amersham, 1:2000) for 1 h at room temperature. After 3 washes, the blots were exposed to the enhanced chemiluminescence substrate. Then the blots were stripped for 1 h at 50°C followed by saturation in 5% nonfat dry milk and incubated with an anti–α-CaMKII antibody (Upstate Biotechnology, 1:5000) for the detection of the total α-CaMKII. Quantification was obtained from densitometric measurements of immunoreactive bands on films.
Dopamine receptor ligands PD168077 maleate, l-745870 trihydrochloride (Tocris Cookson Inc., Ellisville, MO), quinpirole, sulpiride, and SCH23390 (Sigma-Aldrich, St. Louis, MO), as well as second-messenger reagents U73122, genistein, BAPTA/AM, 2-aminoethoxydiphenylborane (2APB), 1,1′-diheptyl-4–4′-bipyridinium (DHBP), thapsigargin (Calbiochem, San Diego, CA), α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid/kainate receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), and N-methyl-d-aspartate receptor antagonist d(-)-2-amino-5-phosphonopetanoic acid (APV) (Sigma-Aldrich) were made up as concentrated stocks and stored at -20°C. The final dimethyl sulfoxide concentration in all applied solutions was less than 0.1%. Stocks were thawed and diluted immediately before use.
Results
Bidirectional Regulation of CaMKII Activity by D4 Receptors in PFC Neurons. CaMKII is activated by the binding of Ca2+/calmodulin, and then it undergoes autophosphorylation at Thr286, which renders the enzyme to obtain Ca2+-independent autonomous activity (Miller and Kennedy, 1986). Thus, the activated CaMKII (Thr286-phosphorylated) should be sensitive to stimuli that can change cellular Ca2+, such as neuronal activity. So we first examined whether the activation levels of CaMKII might be dynamically regulated by D4 receptors in response to different patterns of neuronal activity. PFC slices were incubated for 1 h with either bicuculline (BIC, 10 μM) to increase activity through block of inhibitory transmission, or with CNQX (10 μM) and APV (20 μM) to decrease activity through block of excitatory transmission, followed by a 10-min treatment with the specific D4 receptor agonist PD168077 (Glase et al., 1997; Wang et al., 2002). As shown in Fig. 1A, the basal level of activated CaMKII in PFC slices was higher after BIC treatment compared with after CNQX/APV treatment. PD168077 (20 μM) caused a significant decrease of the activated CaMKII in PFC with high neuronal activity (BIC-treated), but it caused a potent increase of the activated CaMKII in PFC with low neuronal activity (CNQX/APV-treated). In contrast to the bidirectional effect of PD168077, the D2 receptor agonist quinpirole (20 μM) only produced a reduction of the activated CaMKII irrespective of the neuronal activity. Total CaMKII levels exhibited no change with any of the treatment. Quantitative data from a series of experiments are summarized in Fig. 1B. PD168077 decreased Thr286-phosphorylated CaMKII by 65 ± 11% (n = 8) in BIC-treated PFC slices, whereas it increased Thr286-phosphorylated CaMKII by 270 ± 52% (n = 8) in CNQX/APV-treated PFC slices. Quinpirole reduced Thr286-phosphorylated CaMKII by 72 ± 12% (n = 6) or 36 ± 6% (n = 6) in PFC slices treated with BIC or CNQX/APV, respectively.
Similar experiments were performed in PFC slices pretreated with TTX (0.5 μM, 1 h) to suppress spike activity. As shown in Fig. 1, C and D, PD168077 caused a significant decrease (59 ± 10%, n = 8) of the Thr286-phosphorylated CaMKII in PFC with high neuronal activity (no TTX pretreatment), but it caused a marked increase (240 ± 48%, n = 8) of the Thr286-phosphorylated CaMKII in PFC with low neuronal activity (TTX-pretreated). Quinpirole reduced the level of Thr286-phosphorylated CaMKII by 64 ± 12% (n = 6) or 31 ± 5% (n = 6) in PFC slices pretreated without or with TTX, respectively. These results indicate that D4 receptors exert a dynamic bidirectional regulation of CaMKII activity depending on the neuronal activity.
To confirm that neuronal activity is manipulated by drugs that affect synaptic transmission, glutamatergic excitation, or GABAergic inhibition, we compared the level of activated (Thr286-phosphorylated) CaMKII in PFC slices treated with saline or various drugs. As shown in Fig. 1, E and F, compared with saline-treated slices (ctl), slices treated with TTX, CNQX/APV, or the nonselective glutamate receptor antagonist kynurenic acid (1 mM) showed a substantial decrease of the activated CaMKII (TTX: 67 ± 12%, n = 10; CNQX/APV: 72 ± 13%, n = 10; and kynurenic acid: 70 ± 13%, n = 8). Moreover, the reduction of CaMKII activity by TTX or CNQX/APV was not blocked by the PKA activator cpt-cAMP (100 μM, cpt-cAMP + TTX: 58 ± 13%, n = 8; cpt-cAMP + CNQX/APV: 63 ± 10%, n = 8), suggesting that it is not mediated by PKA inhibition. On the other hand, compared with saline-treated slices (ctl), slices treated with bicuculline caused little change on the activated CaMKII (20 ± 4%, n = 8), and slices treated with glutamate/glycine (100/10 μM) or high KCl (30 mM) further increased the level of activated CaMKII (glutamate: 89 ± 17%, n = 8; KCl: 92 ± 17%, n = 8). These data suggest that PFC neurons are switched to the “low activity” state by the treatment with TTX, CNQX/APV, or kynurenic acid, whereas they are at the “high activity” state in saline-(ctl), bicuculline-, glutamate/glycine-, or high KCl-treated slices.
We further compared the effect of PD168077 on CaMKII activity in PFC slices at the high-activity state. As shown in Fig. 1, G and H, PD168077 caused a potent reduction of the activated CaMKII in saline-, glutamate/glycine-, or high KCl-treated slices (saline: 58 ± 12%, n = 8; glutamate: 69 ± 15%, n = 8; KCl: 68 ± 15%, n = 8), similar to the effect of PD168077 in bicuculline-treated slices (Fig. 1, A and B). These data further indicate that D4 receptors decrease the level of CaMKII activation in PFC with high neuronal activity.
Mediation by D4 Receptors of the Up-Regulation of CaMKII Activity in PFC Slices. BecauseD4 receptors couple to the “classic” inhibition of PKA pathway in transfected cell lines (Chio et al., 1994), it is surprising that D4 receptors increased the activation level of CaMKII in PFC neurons with low neuronal activity. Thus, in subsequent experiments, we further examined the mechanisms underlying D4 upregulation of CaMKII activity in TTX-pretreated PFC slices.
The dose-dependence of PD168077-induced CaMKII activation is shown in Fig. 2, A and B. A small effect could be detected after a 10-min exposure to 5 μM PD168077, and a saturating effect was seen at 20 μM PD168077. Quantification data exhibited a 3.4 ± 0.8-fold increase of CaMKII activity (n = 8, p < 0.001, ANOVA) by PD168077 (20 μM, 10 min). The kinetics of PD168077-induced activation of CaMKII was also tested. As demonstrated in Fig. 2, C and D, the CaMKII activation induced by PD168077 (40 μM) showed rapid and transient kinetics, reaching a peak at 10 min and declining to basal levels within 30 to 60 min.
To verify that D4 receptors were mediating the PD168077 activation of CaMKII, we examined the ability of selective D4 receptor antagonists to prevent the action of PD168077. As shown in Fig. 3, A and B, PD168077 (40 μM) produced a potent increase (3.7 ± 0.9-fold, n = 12) of activated CaMKII in PFC slices, and this effect was significantly (p < 0.001, ANOVA) blocked by l-745870 (20 μM, 0.98 ± 0.2-fold, n = 8), a highly selective D4 antagonist (Patel et al., 1997). In contrast to the strong effect of PD168077 on CaMKII activation in PFC slices, PD168077 failed to regulate CaMKII activity in striatal slices (1.1 ± 0.2-fold, n = 6) (Fig. 3, A and B), consistent with the highly enriched expression of D4 receptors in PFC but not in striatum.
To further confirm the involvement of D4 receptors in the up-regulation of CaMKII activity, we tested the effect of dopamine (50 μM) on CaMKII in the presence of D1/D5 antagonist SCH23390 (10 μM) and D2/D3 antagonist sulpiride (10 μM). As shown in Fig. 3, C and D, when D1/D5 and D2/D3 receptors were blocked, dopamine produced an enhancement of CaMKII activity (3.4 ± 0.8-fold, n = 8), mimicking the PD168077 effect. Moreover, this effect of dopamine on CaMKII was blocked by the D4 antagonist l-745870 (1.15 ± 0.2-fold, n = 6). These results suggest that dopamine released on PFC neurons could indeed elevate CaMKII activity via D4 receptors.
Signaling Mechanisms Underlying the D4 Enhancement of CaMKII Activity in PFC Slices. Our previous study has shown that D4 receptors decrease CaMKII activity in PFC slices (no TTX pretreatment) through a cascade involving the inhibition of PKA and ensuing disinhibition of protein phosphatase 1 (Wang et al., 2003). We next examined the signal transduction pathways mediating the increase of CaMKII activity by D4 receptors in PFC slices when the neuronal activity was suppressed by TTX pretreatment. As shown in Fig. 4A, application of the phospholipase C (PLC) inhibitor U73122 (1 μM) but not the broad-spectrum tyrosine kinase inhibitor genistein (100 μM) blocked the PD168077-induced increase of CaMKII activity. Moreover, application of the Gi/o protein alkylating agent N-ethylmaleimide (NEM, 30 μM) failed to prevent PD168077 from elevating CaMKII activity. As summarized in Fig. 4B, the PD168077-induced activation of CaMKII (3.6 ± 0.8-fold, n = 14) was abolished in the presence of U73122 (0.9 ± 0.2-fold, n = 8) but was intact in the presence of genistein (3.5 ± 1.0-fold, n = 8) or NEM (3.9 ± 1.3-fold, n = 8). These results suggest that the D4 enhancement of CaMKII activity is through a mechanism dependent on the stimulation of PLC pathway but not the activation of tyrosine kinases or the coupling to Gi/o proteins.
Given the dependence of CaMKII activation on Ca2+, we then examined whether the D4 enhancement of CaMKII activity required the Ca2+ entry from extracellular regions or the Ca2+ elevation from intracellular stores. To test this, PFC slices were incubated in Ca2+-free solutions or with the membrane-permeable Ca2+ chelator BAPTA/AM. As shown in Fig. 5, A and B, PD168077 induced a potent increase (4.1 ± 1.2-fold, n = 8) in CaMKII activity under Ca2+-free conditions, similar to the PD168077 effect in normal Ca2+-containing solutions (3.8 ± 0.7-fold, n = 15). However, when intracellular Ca2+ increase was blocked by BAPTA/AM (50 μM), PD168077 failed to enhance CaMKII activity (1.12 ± 0.25-fold, n = 8). These results suggest that an increase of intracellular Ca2+ from internal stores is required for the D4 enhancement of CaMKII activity.
In neurons, a major source of internal calcium is the stores present in the endoplasmic reticulum network. Both ionsitol-1,4,5-triphosphate receptors (IP3Rs) and ryanodine receptors (RyRs) on endoplasmic reticulum are responsible for releasing calcium from this internal source (Kostyuk and Verkhratsky, 1994; Simpson et al., 1995). To determine which one was involved in the D4 activation of CaMKII, PFC slices were pretreated with pharmacological agents to block these receptors. As shown in Fig. 5, A and B, application of 2APB (30 μM), a membrane-permeable IP3R antagonist (Hamada et al., 1999), abolished the D4 effect on CaMKII activation (0.9 ± 0.2-fold, n = 10). In contrast, DHBP (30 μg/ml), a potent RyR antagonist (Kang et al., 1994), failed to alter the D4 enhancement of CaMKII activity (3.6 ± 0.9-fold, n = 6). Pretreatment of PFC slices with the intracellular calcium pump inhibitor thapsigargin (5 μM, 30 min) to deplete internal stores of Ca2+ also eliminated the D4 effect on CaMKII activation (1.0 ± 0.2-fold, n = 8). These results suggest that D4 receptors elevate intracellular calcium via IP3Rs to increase CaMKII activity.
As a control, we also examined the involvement of PLC/IP3R signaling in D4 reduction of CaMKII activity in high neuronal activity conditions. As shown in Fig. 5, C and D, application of the PLC inhibitor U73122 (1 μM) or IP3R antagonist 2APB (30 μM) failed to block the PD168077-induced decrease of activated CaMKII (62 ± 13%, n = 8; U73122: 74 ± 12%, n = 8; 2APB: 71 ± 11%, n = 8), suggesting that the PLC/IP3R signaling is not involved in D4 reduction of CaMKII activity.
Discussion
CaMKII has been regarded as a cognitive kinase because of its involvement in regulating learning and memory, and its autoregulatory properties that can be viewed as a type of molecular memory (Hudmon and Schulman, 2002). A variety of extracellular signals triggers the activation of CaMKII by elevating the intracellular Ca2+ level through Ca2+ influx or Ca2+ release from internal stores. We show here that stimulation of D4 receptors can lead to either up- or down-regulation of CaMKII activity depending on basal neuronal activity. In PFC slices with suppressed neuronal activity, the level of activated CaMKII was increased by D4 receptors, whereas in PFC slices with elevated neuronal activity, the level of activated CaMKII was decreased by D4 receptors. This dual regulation of CaMKII activity was unique for D4 receptors, because it was not observed with D2 receptor activation. By regulating CaMKII activity in such a dynamic activity-dependent fashion, D4 receptors could fine-tune the functions of CaMKII flexibly and precisely.
How can D4 receptors either decrease or increase CaMKII activity? Emerging evidence has suggested that G protein-mediated signal transduction is a complex signaling network with divergent and convergent pathways intimately intertwined (Gudermann et al., 1997). The classic signaling cascade for D4 receptors is to couple to Gi/o-type G proteins to inhibit adenylate cyclase and cAMP formation (Chio et al., 1994). The inhibition of PKA could cause the activation of protein phosphatase 1 via decreased phosphorylation of the inhibitory protein I-1 (Ingebritsen and Cohen, 1983), leading to the dephosphorylation of CaMKII and a decrease of CaMKII activity. Our previous study confirmed this mechanism for the D4 down-regulation of CaMKII activity (Wang et al., 2003). In this study, we show that the D4 up-regulation of CaMKII activity is through the stimulation of PLC pathway and elevation of intracellular Ca2+ via IP3Rs. How D4 receptors activate the PLC pathway is not clear. One potential mechanism is that activation of D4 receptors in PFC neurons leads to the release of G protein βγ subunits and thus potentiates the stimulation of PLC by βγ subunits (Camps et al., 1992).
This study mechanistically links together D4 receptors and CaMKII, both of which have been implicated in cognitive and emotional processes associated with PFC. The D4 regulation of CaMKII activity enables D4 receptors to affect many aspects of cellular function via changing numerous CaMKII substrates, such as K+ channels, glutamate receptors, synapsin, cAMP response element-binding protein, tau, and so on (Hudmon and Schulman, 2002). A novel feature of this D4 modulation of CaMKII is that it is bidirectional depending on neuronal activity, and a “dual signaling” (i.e., inhibition of adenylate cyclase and stimulation of PLC) underlies the D4-induced suppression or potentiation of CaMKII activity. This supports the notion that many neuromodulators, such as dopamine and serotonin, can have dual roles not only because they can act on a variety of different targets, but also because they can act differently on the same target under different physiological conditions (Cai et al., 2002). This mechanism ensures that the modulation provides a feedback system to effectively maintain normal neuronal activity.
Footnotes
-
This work was supported by National Institutes of Health grants NS48911, MH63128 and AG21923 (to Z.Y.), National Science Foundation grant IBN-0117026 (to Z.Y.), and Howard Hughes Medical Institute Biomedical Research Support Program grant 53000261 (State University of New York at Buffalo).
-
Article, publication date, and citation information can be found at http://molpharm.aspetjournals.org.
-
doi:10.1124/mol.104.001404.
-
ABBREVIATIONS: PFC, prefrontal cortex; CaMKII, Ca2+/calmodulin-dependent protein kinase II; PLC, phospholipase C; CNQX, 6-cyano-7-nitroquinoxaline-2,3-dione; APV, d(-)-2-amino-5-phophonopetanoic acid; 2APB, 2-aminoethoxydiphenylborane; DHBP, 1,1′-diheptyl-4-4′-bipyridinium; NEM, N-ethylmaleimide; ctl, control; IP3R, ionsitol-1,4,5-triphosphate receptor; RyR, ryanodine receptor, ANOVA, analysis of variance; PD168077, [(4-phenylpiperazinyl)-methyl]benzamide; PKA, protein kinase A; BIC, bicuculline; TTX, tetrodotoxin; U73122, 1-[6-[[17β-methoxyestra-1,3,5(10)-trien-17-yl]amino]hexyl]-1H-pyrrole-2,5-dione; SCH23390, R-(+)-7-chloro-8-hydroxy-3-methyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine; l-745870, (3-([4-(4-chlorophenyl)piperazin-1-yl]methyl)-1H-pyrollo[2,3-b]pyridine; BAPTA/AM, 1,2-bis(2-aminophenoxy)ethane-N,N,N′-tetraacetic acid/acetoxymethyl ester; cpt-cAMP, 8-(4-chlorophenylthio)adenosine 3′,5′-cyclic monophosphate.C
- Received April 13, 2004.
- Accepted June 30, 2004.
- The American Society for Pharmacology and Experimental Therapeutics