G-protein Receptor Kinase 5 Regulates the Cannabinoid Receptor 2-induced Up-regulation of Serotonin 2A Receptors*

Background: Cannabinoids regulate serotonin signaling in prefrontal cortex. Results: Cannabinoid-induced up-regulation and enhanced activity of serotonin 2A (5-HT2A) receptors are regulated by G-protein receptor kinase 5 (GRK5) in neuronal cells. Conclusion: Cannabinoids differentially regulate expression of GRKs, which could contribute to the modulation of CB2 receptor signaling. Significance: Cannabinoid-induced up-regulation of 5-HT2A receptors could represent an adverse effect of repeated exposure to cannabinoids. We have recently reported that cannabinoid agonists can up-regulate and enhance the activity of serotonin 2A (5-HT2A) receptors in the prefrontal cortex (PFCx). Increased expression and activity of cortical 5-HT2A receptors has been associated with neuropsychiatric disorders, such as anxiety and schizophrenia. Here we report that repeated CP55940 exposure selectively up-regulates GRK5 proteins in rat PFCx and in a neuronal cell culture model. We sought to examine the mechanism underlying the regulation of GRK5 and to identify the role of GRK5 in the cannabinoid agonist-induced up-regulation and enhanced activity of 5-HT2A receptors. Interestingly, we found that cannabinoid agonist-induced up-regulation of GRK5 involves CB2 receptors, β-arrestin 2, and ERK1/2 signaling because treatment with CB2 shRNA lentiviral particles, β-arrestin 2 shRNA lentiviral particles, or ERK1/2 inhibitor prevented the cannabinoid agonist-induced up-regulation of GRK5. Most importantly, we found that GRK5 shRNA lentiviral particle treatment prevented the cannabinoid agonist-induced up-regulation and enhanced 5-HT2A receptor-mediated calcium release. Repeated cannabinoid exposure was also associated with enhanced phosphorylation of CB2 receptors and increased interaction between β-arrestin 2 and ERK1/2. These latter phenomena were also significantly inhibited by GRK5 shRNA lentiviral treatment. Our results suggest that sustained activation of CB2 receptors, which up-regulates 5-HT2A receptor signaling, enhances GRK5 expression; the phosphorylation of CB2 receptors; and the β-arrestin 2/ERK interactions. These data could provide a rationale for some of the adverse effects associated with repeated cannabinoid agonist exposure.

The clinical manifestations of this CB 2 receptor-induced upregulation of 5-HT 2A receptors are currently under discussion. It is noteworthy that recent and independent clinical studies provide evidence indicating that sustained use of nonselective cannabinoid agonists may precipitate the onset of mental disorders associated with dysfunction of 5-HT 2A receptor neurotransmission in PFCx, such as anxiety, schizophrenia, and psychosis (5)(6)(7)(8)(9). Accordingly, recent preclinical studies have indicated that chronic, but not acute, exposure to non-selective (10,11) or selective CB 2 receptor agonists induced anxiety-like behaviors in rodents (12). CB 2 receptors have been identified in postsynaptic neurons in several brain areas of the limbic brain, including brain areas such as the PFCx, hippocampus, and amygdala (13)(14)(15)(16). The CB 2 receptor is a prototypical G-protein-coupled receptor (GPCR) that couples to the G i/o class of G-proteins and can activate ERK1/2 signaling in either a G-protein-or ␤-arrestindependent pathway (17,18). The different signaling and trafficking profiles of this receptor would depend on the nature of post-translational modifications, such as phosphorylation by G-protein receptor kinases (GRKs) that modify the interaction between this receptor and associated signaling proteins (such as ␤-arrestins and G-proteins) (17) and desensitization of this receptor (19).
Here we study the role of GRKs in the cannabinoid-induced up-regulation of 5-HT 2A receptors. GRKs, such as GRK2, exert important roles in the desensitization and inhibition of ␤-arrestin 2 (␤Arr2) signaling of GPCRs (20,21). Of note, recent results demonstrate that some GRKs, such as GRK5 and/or GRK6, also regulate ␤Arr2 signaling-mediated ERK1/2 activation (20). Here we report that agonists of cannabinoid receptors differentially regulate the expression of GRK proteins, which would contribute to regulation of 5-HT 2A receptors in neuronal cells. We hypothesize that the data presented here could provide, at least in part, a molecular mechanism by which repeated exposure to cannabinoids might be relevant to the pathophysiology of some cognitive and mood disorders by up-regulating and enhancing the activity of 5-HT 2A receptors.
Animal Experimental Protocol-Male Sprague-Dawley rats (225-275 g; Harlan Laboratories, Indianapolis, IN) were housed two per cage in a temperature-, humidity-, and lightcontrolled room (12-h light/dark cycle, lights on 7:00 a.m. to 7:00 p.m.). Food and water were available ad libitum. All procedures were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals as approved by the University of Kansas Institutional Animal Care and Use Committee.
After arrival, the rats were allowed to acclimate to their environment for at least 4 days prior to the start of the treatment period. Eight rats were randomly assigned to each group; cage mates were assigned to the same treatment group. The body weight of each rat was recorded every other day. All solutions were made fresh before administration, and rats were injected with either vehicle (Tween 80/ethanol/saline (1:1:18); 1 ml/kg, intraperitoneally) or CP55940 (0.05 mg/kg, intraperitoneally) once per day for 7 days. Rats were sacrificed by decapitation 48 h after the last CP55940 injection. The brains were immediately removed, and the PFCx was dissected and frozen in dry ice.
Phosphoprotein Purification-Phosphorylated proteins were separated by an affinity chromatography procedure using a phosphoprotein purification kit from Qiagen (Valencia, CA) as described in detail previously (22). Immunodetection by phospho-specific antibodies has shown that the kit yields a complete separation of non-phosphorylated (flow-through) and phosphorylated proteins (elution fraction) (23). Briefly, tissue or cells were homogenized in 200 l of phosphoprotein lysis buffer containing 0.25% (w/v) CHAPS solution, protease inhibitor mixture, and Benzonase. These homogenates were incubated for 30 min at 4°C and then centrifuged for 30 min at 10,000 ϫ g and 4°C. Thermo Scientific Pierce BCA assay reagents were utilized to determine the protein concentrations of the supernatants, and then 3.5 mg of total protein, adjusted to 0.1 mg/ml with phosphoprotein lysis buffer containing 0.25% CHAPS, was run through the phosphoprotein purification columns. The non-phosphorylated proteins were washed out of the columns with 35 ml of phosphoprotein lysis buffer, and the bound phosphorylated proteins were eluted with phosphopro-tein elution buffer. Eluted fractions containing phosphorylated proteins were collected. The isolated phosphoprotein fractions were concentrated using Nanostep ultrafiltration columns with a molecular mass cut-off of 10 kDa. Thermo Scientific Pierce BCA assay reagents were utilized to determine the protein concentrations, and samples were analyzed by Western blot.
Effect of ␤-Arrestin 2, CB 2 , or CB 1 shRNA Lentivirus Transfection on Cannabinoid-induced Up-regulation of GRK5 mRNA-After confirming that treatment with the ␤-arrestin 2, CB 2 , or CB 1 shRNA lentivirus significantly reduced the respective protein levels, ␤-arrestin 2, CB 2 , or CB 1 shRNA-treated cells were treated with either vehicle (ethanol, 0.01% final concentration) or GP1a (1 nM) for 72 h. Cells were washed (three times) with PBS every 24 h, and fresh vehicle or 1 nM GP1a was added.
Calcium Assay to measure 5-HT 2A Receptor Activity-Optimal conditions were elucidated using different buffers, plating densities, agonists, time courses, etc., and through reference to previously published protocols (34,35). Cells were plated at 30,000 cells/well in complete medium and grown to 90% confluence on black-sided 96-well plates. 24 h prior to measuring calcium release, medium was changed to serum-free medium. After a 24-h incubation in serum-free medium, cells were washed (twice) with Kreb's medium (135 mM NaCl, 5.9 mM KCl, 1.5 mM CaCl 2 , 1.2 mM MgCl 2 , 11.6 mM Hepes, 11.5 mM D-glucose, pH 7.3) and incubated with 4 M Fluo 3-AM in 200 l of Kreb's medium for 60 min at 37°C in the dark. After loading, cells were washed (twice) with Kreb's medium and incubated in 200 l of Kreb's medium for 30 min to allow for de-esterfication of intracellular AM esters. Finally, cells were stimulated with a single injection of 5-HT, and the response was recorded for 1 min in 6-s intervals. Fluo 3-AM fluorescence using 485-nm excitation and 528-nm emission was measured with a BioTek fluorescence plate reader (34,35).
Effect of GRK5 shRNA Lentivirus Transfection on Cannabinoid-induced CB 2 Receptor Phosphorylation or ␤-Arrestin 2/ERK Interaction-Cells stably transfected with control or GRK5 shRNA lentivirus particles were treated with vehicle (ethanol, 0.01% final concentration) or GP1a (1 nM) for 72 h. Cells were washed (three times) with Kreb's buffer every 24 h, and fresh vehicle or GP1a was added. Phosphorylated proteins were isolated from cells, and Western blot was used to determine the expression of phosphorylated CB 2 receptors as described above. Co-immunoprecipitation of ␤-arrestin 2/ERK was examined following the protocol listed below. Expression of phosphorylated CB 2 receptors was determined by Western blot as described above.
Co-immunoprecipitation-These experiments were conducted with the Thermo Scientific Pierce co-immunoprecipitation kit following the manufacturer's protocol and as described in detail previously (1). The ␤-arrestin 2 and ERK1/2 antibody was purchased from Santa Cruz Biotechnology, Inc. Samples were analyzed by Western blot using ERK1/2 antibody. The specificity of the ␤-arrestin 2 or ERK1/2 antibody has been verified (1).
Statistics-All data are expressed as the mean Ϯ S.E., where n indicates the number of rats or cell culture plates per group. Data were analyzed by an unpaired Student's t test or ANOVA.

RESULTS
Chronic exposure to cannabinoid receptor agonists could mediate the cannabinoid-induced up-regulation of 5-HT 2A receptors, at least in part, through changes in the phosphorylation status of CB 2 receptors by GRK proteins. We initially examined the effect of repeated exposure to CP55940 (CB 1 /CB 2 receptor agonist) on the phosphorylation status of the CB 2 receptors in rat PFCx (Fig. 1A). Phosphorylated proteins were separated from the PFCx of vehicle-and CP55940-treated rats, and Western blot was conducted as explained previously. We found a significant (p Ͻ 0.01) increase in the phosphorylation of CB 2 receptors in CP55940-treated rats compared with vehicletreated animals (121 Ϯ 12% increase; Fig. 1A). Of note, CP55940 treatment did not significantly (p Ͼ 0.05) modify the total CB 2 receptor protein expression in PFCx homogenate compared with vehicle-treated controls (Fig. 1B).
We then used a neuronal cell line, CLU213 cells, to determine if CP55940 treatment shifts GRK expression similar to shifts found in rat PFCx and to better examine the mechanisms FIGURE 1. CP55940-induced enhanced phosphorylation of CB 2 receptors, increased GRK5 expression levels, and reduced GRK2 expression levels in rat PFCx. Rats were injected with CP55940 (0.05 mg/kg, intraperitoneally) once a day for 7 days. After decapitation, the brains were collected, and PFCx was dissected. A, phosphorylated proteins were separated and detected as described under "Experimental Procedures." 30 g of isolated phosphorylated protein was used in Western blot detection. B-E, CB 2 receptor and GRK protein levels were evaluated by Western blot. Proteins (8 g) were resolved by SDS-PAGE, and antibodies for CB 2 receptor (A and B), GRK5 (C), GRK6 (D), and GRK2 (E) were used to detect the proteins of interest. Representative Western blots are shown, and integrated optical density was calculated as described under "Experimental Procedures." ␤-Actin was used as a loading control. F, GRK5, GRK6, and GRK2 mRNA levels were evaluated by qRT-PCR as described under "Experimental Procedures." **, p Ͻ 0.01; *, p Ͻ 0.05, significant effect of CP55940 treatment compared with vehicle-treated controls. The data represent mean Ϯ S.E. (error bars) (n ϭ 6 -8).
involved in the cannabinoid-induced up-regulation of GRK5 proteins. Here we treated cells with CP55940 for 72 h in order to assess the effect of repeated cannabinoid agonist exposure on the expression of GRKs because our previous findings show that repeated, but not single, exposure to cannabinoid agonists up-regulates 5-HT 2A receptor protein expression (1)(2)(3)(4). CP55940 treatment in these cells significantly (p Ͻ 0.01) increased GRK5 protein levels (67 Ϯ 3% increase compared with vehicle-treated controls) without significant (p Ͼ 0.05) changes in the protein levels of GRK6 (Fig. 2, A and B, respectively). On the other hand, GRK2 protein levels were significantly (p Ͻ 0.01) reduced compared with controls (24 Ϯ 2% decrease; Fig. 2C). CP55940 treatment also significantly (p Ͻ 0.05) increased GRK5 mRNA levels and significantly (p Ͻ 0.05) reduced GRK2 mRNA levels compared with vehicle-treated cells (69 Ϯ 0.2% increase and 24 Ϯ 0.05% decrease, respectively). There was no significant (p Ͼ 0.05) change in GRK6 mRNA levels in CP55940-treated cells compared with vehicletreated controls (Fig. 2D).
We then aimed to identify the cannabinoid receptor involved in the up-regulation of GRK5 because this could mediate the enhanced phosphorylation of CB 2 receptors detected in rat PFCx. Cells were treated with either vehicle, GP1a (1 nM) (selec-tive CB 2 agonist), or ACEA (15 nM) (selective CB 1 agonist) over 72 h. We found that GP1a treatment significantly (p Ͻ 0.01) increased (69 Ϯ 0.05% increase), whereas ACEA did not significantly (p Ͼ 0.5) modify GRK5 mRNA levels compared with controls ( Fig. 3A). This evidence suggested that CB 2 receptors could mediate the up-regulation of GRK5. We then tested the effect of GP1a on GRK5 mRNA levels in either control, CB 1 shRNA, or CB 2 shRNA stably transfected cells over 72 h. We have previously shown that treatment with the CB 1 or CB 2 shRNA lentiviral particles significantly reduces CB 1 or CB 2 receptor expression, respectively (3). We found that treatment with GP1a significantly (p Ͻ 0.01) increased GRK5 mRNA levels in control and CB 1 shRNA-treated cells (73 Ϯ 0.03 and 73 Ϯ 0.04% increase compared with controls, respectively). It is noteworthy that CB 2 shRNA treatment prevented (p Ͻ 0.01) the GP1a-induced increases in GRK5 mRNA levels (Fig. 3B). Neither control, CB 1 , nor CB 2 shRNA lentivirus treatment significantly (p Ͼ 0.01) modified basal GRK5 mRNA levels. The twoway ANOVA for GRK5 mRNA showed significant main effects of transfection (F 2,17 ϭ 20.4, p Ͻ 0.0001) and cannabinoid agonist treatment (F 1,17 ϭ 187, p Ͻ 0.0001). There was a significant interaction between transfection and cannabinoid agonist treatment (F 2,17 ϭ 22.8, p Ͻ 0.0001) on GRK5 mRNA levels. Next, we investigated whether the ERK1/2 signaling pathway may be involved in the cannabinoid-induced up-regulation of GRK5. CB 2 receptors are positively coupled to the ERK1/2 signaling pathway, and cannabinoid agonists, such as ⌬ 9 -THC, can regulate the expression of some GRKs through the ERK1/2 signaling pathway (36). We used PD198306, a selective ERK1/2 inhibitor (37), to study the effect of GP1a-induced ERK1/2 activation on GRK5 up-regulation. GP1a treatment significantly (p Ͻ 0.01) increased GRK5 mRNA levels compared with vehicle-treated cells (77 Ϯ 2% increase; Fig. 3D). This up-regulation of GRK5 was prevented (p Ͻ 0.01) by pretreatment with PD198306. This ERK1/2 inhibitor pretreatment did not significantly (p Ͼ 0.05) modify basal levels of GRK5 mRNA. The two-way ANOVA for GRK5 mRNA showed significant main effects of PD198306 pretreatment (F 1,11 ϭ 22.4, p Ͻ 0.0015) and GP1a treatment (F 1,11 ϭ 43.3, p Ͻ 0.0002). There was a significant interaction between PD198306 and GP1a treatment (F 1,11 ϭ 35.2, p Ͻ 0.0003).
In the next studies, we examined the role of GRK5 in the cannabinoid-induced increases in 5-HT 2A receptor activity. We have previously reported that repeated CP55940 treatment in this neuronal cell culture model significantly enhances 5-HT 2A receptor-mediated phosphoinositol hydrolysis (2). Here, we studied the effects of GRK5 shRNA treatment on the 5-HT 2A receptor-mediated calcium (Ca 2ϩ ) release. We began conducting dose response experiments with 5-HT as described under "Experimental Procedures." We used 5-HT in these experiments because previous studies have shown that the maximal response to (Ϫ)-1-2,5-dmiethoxy-4-iodoamphetamine HCl (a 5HT 2A/2C receptor agonist) is lower than the maximal response to 5-HT in two different cell lines (35). A dose-response experiment in CLU213 cells showed that 5-HT stimulated Ca 2ϩ release in a dose-dependent way (Fig. 5A) with an EC 50 of 0.11 Ϯ 0.02 nM. To confirm that this response was the result of stimulation of 5-HT 2A but not 5-HT 2C receptors, we measured the effect of MDL11,939 or SB242084 (5-HT 2A and 5-HT 2C receptor antagonists, respectively) (34,35) in the 5-HT-induced Ca 2ϩ release in a neuronal cell model. MDL11,939 and SB242084 inhibited the 5-HT-mediated Ca 2ϩ release with different affinities. Whereas the MDL11,939 IC 50 was ϳ1 nM, the SB242084 IC 50 was 0.1 M, suggesting that the 5-HT-mediated Ca 2ϩ release in CLU213 cells is mainly mediated by 5-HT 2A receptors at the concentration of 5-HT (0.1 nM) used in this assay (Fig. 5B). 10 nM SB242084 was added to the preincubation medium in the cannabinoid assays to prevent the activation of 5-HT 2C receptors. Based on the K d provided under "Experimental Procedures," the fractional occupancy of 5-HT 2C and 5-HT 2A receptors at this dose of SB242084 is 95 and 7%, respectively.
Our next aim was to determine whether GRK5 plays a significant role in the 5-HT-mediated Ca 2ϩ release in a neuronal cell model. Control cells or cells stably transfected with GRK5 shRNA were incubated with either vehicle, CP55940 (1 nM) or GP1a (1 nM), for 72 h. In control cells, 5-HT (0.1 nM)-mediated Ca 2ϩ release was significantly increased by both CP55940 and GP1a treatment (205 Ϯ 3% and 201 Ϯ 4% increase over control for CP55940 or GP1a, respectively). It is noteworthy that in cells stably transfected with GRK5 shRNA, neither CP55940 nor GP1a induced significant increases in the 5-HT-mediated Ca 2ϩ release. The two-way ANOVA showed a significant main effect of transfection (F 1,17 ϭ 215.2, p Ͻ 0.0001) and cannabinoid treatment (F 2,17 ϭ 55.97, p Ͻ 0.0001) and a main interaction between them (F 2,17 ϭ 46.58, p Ͻ 0.0001). In summary, these results suggest that GRK5 plays a pivotal role in the CB 2 receptor-induced up-regulation of 5-HT 2A receptors in our neuronal cell model.
We then examined the role of GRK5 in the cannabinoid agonist-induced phosphorylation of CB 2 receptors. Cells stably transfected with either GRK5 or control shRNA lentiviral particles were treated with vehicle or GP1a (1 nM) for 72 h, and phosphorylated proteins were isolated as described under "Experimental Procedures." We found that GP1a treatment significantly (p Ͻ 0.01) enhanced phosphorylation of CB 2 receptors in control shRNA-treated cells by 36 Ϯ 7% (Fig. 7A). Of note, this GP1a-induced enhanced phosphorylation of CB 2 receptors was prevented (p Ͻ 0.01) in cells stably transfected with GRK5 shRNA lentiviral particles. No significant differences in the CB 2 receptor phosphorylation levels were detected between vehicle and GP1a in cells stably transfected with GRK5 shRNA lentiviral particles. The two-way ANOVA for phosphorylated CB 2 showed main effects of transfection (F 1,43 ϭ 21, p Ͻ 0.0001) and GP1a (F 1,43 ϭ 4.8, p Ͻ 0.0333). There was a significant interaction between transfection and GP1a treatment (F 1,43 ϭ 8.8, p Ͻ 0.005) on phosphorylated CB 2 receptors. We then examined the effects of GRK5 shRNA lentivirus treatment and GP1a treatment on CB 2 receptor protein levels in whole cell lysates. Repeated GP1a treatment did not significantly (p Ͼ 0.05) modify CB 2 receptor protein levels in whole cell lysates (Fig. 7B). Furthermore, GRK5 shRNA lentivirus particle transfection did not significantly (p Ͼ 0.05) alter the basal levels of CB 2 receptors (Fig. 7B).

DISCUSSION
Cannabinoid agonists produce their physiological effects through the activation of two G-protein-coupled cannabinoid receptors in the brain, the CB 1 and CB 2 receptors (18,38). CB 1 and CB 2 receptors bind endocannabinoids, synthetic cannabinoids, and cannabinoids found in nature (such as in Cannabis sativa) with high affinity (18,38). Although only CB 1 receptors were initially identified in the brain (39), later studies have also identified CB 2 receptors in several brain areas, including PFCx, hippocampus, amygdala, substantia nigra, and cerebellum (13,14), triggering a reevaluation of the possible roles that CB 2 receptors might play in the brain.
We have previously reported that repeated exposure to either nonselective cannabinoid agonists or selective CB 2 receptor agonists up-regulates and enhances the activity of 5-HT 2A receptors in rat PFCx and neuronal cell models (1)(2)(3)(4). CB 2 receptors can couple to the G i/o class of G-proteins to regulate transient ERK1/2 signaling, whereas ␤-arrestin 2 may be involved in the long term regulation of ERK1/2 signaling (1,17,40,41). Recent evidence has highlighted that neuronal CB 1 receptors can modulate ERK1/2 signaling through G i/o and multiple tyrosine kinase receptors (41). Although G-protein mediated activation of ERK1/2 is transient and peaks within 2-5 min (42,43), ␤-arrestins can form a scaffolding complex with ERK1/2 to regulate long term ERK1/2 activity (42)(43)(44). Although the mechanisms of the cannabinoid-induced up-regulation of 5-HT 2A receptors have not been completely identified, our results suggest that activation of the ␤-arrestin 2 and ERK1/2 signaling pathway mediates this phenomenon that is dependent on CB 2 , but not CB 1 , receptors (2,4). The key role of ␤-arrestin 2 in this up-regulation seems to involve an enhanced cannabinoid-induced interaction between ␤-arrestin 2 and ERK1/2 in rat PFCx (1).
Recent reports suggest that certain GRK proteins could trigger the activation of the ␤-arrestin 2 and ERK1/2 signaling pathway (45). The classical role described for GRK proteins is to trigger the desensitization of GPCRs (42,46). Indeed, GRK2 and GRK3 proteins would phosphorylate the serine and threonine residues within the intracellular loops and carboxyl-terminal tail domains of GPCRs to uncoupled them from their G-proteins and, hence, trigger the desensitization of their corresponding signaling pathway (42,46). GRK2 and GRK3 proteins would also inhibit ␤-Arrestin signaling in a G␤␥-dependent pathway (45,47). On the other hand, GRK5 and GRK6 proteins would have new roles in the signaling of GPCR that would relate to their ability to trigger the activation of the ␤-arrestin 2/ERK1/2 signaling pathway in a G-protein-independent way (21,45,47). Indeed, overexpression of GRK5 and/or GRK6 has been found to enhance ␤-arrestin 2-mediated ERK1/2 activation, whereas overexpression of GRK2 and/or GRK3 abolishes ␤-arrestin 2-mediated ERK1/2 activation (45).
Here we report that repeated CP55940 treatment increases CB 2 receptor phosphorylation and selectively increases GRK5 mRNA and protein expression in rat PFCx and a neuronal cell model without changes in the mRNA or protein levels of GRK6. This was also associated with reduced levels of GRK2 mRNA and protein levels in this area of the limbic brain and in cultured cells.
A limited number of reports have studied the effect of cannabinoids on the regulation of GRK protein expression. For instance, multiple tetrahydrocannabinoil (CB 1 /CB 2 receptor agonist) treatments, but not a single tetrahydrocannabinoil treatment, up-regulates GRK2 and GRK4 in the striatum, GRK4 in the cerebellum, and GRK2 in the PFCx and hippocampus (36). To the best of our knowledge, there are currently no other reports detailing the effects of repeated cannabinoid agonist exposure on the expression of GRK proteins. This limited evidence would suggest that chronic exposure to different classes of cannabinoids may have differential effects on expression of GRKs and subsequent regulation of signaling cascades throughout the brain. Furthermore, our previous evidence suggests that repeated CP55940 or GP1a treatment enhances ␤-arrestin 2-mediated ERK1/2 signaling because we have previously reported that repeated cannabinoid treatment enhances pERK levels over a single cannabinoid exposure (1,2). We have previously found that the ␤-arrestin 2 shRNA lentivirus transfection significantly reduces cannabinoid-induced increases in pERK levels (2). The modulation of GRK protein expression by cannabinoids could be contributing to an intensification of this signaling cascade. Interestingly, different agonists and drugs of abuse have been shown to modulate changes in expression of GRKs, and changes in GRK expression have been described in different pathophysiological conditions (48).
Our evidence indicates that the cannabinoid-induced changes in GRK5 expression could be mediated by changes in transcription because we report here that repeated CP55940 treatment increases GRK5 mRNA and protein in rat PFCx and in our neuronal cell culture model. In neuronal cells, we found that a selective CB 2 receptor agonist (GP1a), but not a selective CB 1 receptor agonist (ACEA), significantly increased GRK5 mRNA levels compared with vehicle-treated controls, suggesting that CB 2 receptors mediate the cannabinoid-induced upregulation of GRK5. Confirmatory evidence of the role of CB 2 receptors in the GRK5 up-regulation was provided by studies with either CB 1 or CB 2 shRNA lentiviral particles (Fig. 3), where the cannabinoid-induced up-regulation of GRK5 was prevented only in CB 2 shRNA lentivirus-treated cells. Although the detailed mechanism of cannabinoid-induced up-regulation of GRK5 was not identified in this paper, we speculate that a transcription factor such as nuclear factor immunoglobulin chain enhancer-B cell (NF-B) could mediate this GRK5 upregulation. The CB 2 receptor is positively coupled to the ERK1/2 signaling pathway, which regulates NF-B (18,49). Rat, human, and mouse GRK5 promoter contains a consensus sequence for NF-B, and activation of NF-B increases GRK5 expression (50).
Here we also investigated the role of GRK5 in the cannabinoid-induced up-regulation of 5-HT 2A receptors. Through the use of GRK5 shRNA lentiviral particles, we identified that GRK5 is involved in the cannabinoid-induced up-regulation and enhanced activity of 5-HT 2A receptors. Indeed, treatment with GRK5 lentiviral particles significantly reduced the CP55940-and GP1a-induced up-regulation of 5-HT 2A receptors without significantly altering basal levels of 5-HT 2A recep-tor mRNA. However, treatment with CP55940 or GP1a significantly increased 5-HT 2A mRNA levels in GRK5 shRNA lentivirus-treated cells compared with vehicle-treated controls. This evidence suggests that the CB 2 receptor can mediate 5-HT 2A receptor up-regulation, at least in part, through GRK5. Here the CP55940-and GP1a-induced increases in 5-HT 2A mRNA levels could be attributed to new rates in synthesis and degradation of GRK5 protein due to the GRK5 shRNA lentivirus particle transduction.
Additionally, we found that repeated CP55940 and GP1a treatment significantly increased serotonin-stimulated 5-HT 2A receptor-mediated Ca 2ϩ release (Fig. 5, D and E). We have previously reported that repeated CP55940 treatment significantly increases 5-HT 2A receptor-mediated phospholipase C␤ activity in rat PFCx-and 5-HT 2A receptor-mediated phosphoinositol hydrolysis in a neuronal cell culture model (2). Interestingly, here we provide more evidence that the enhanced 5-HT 2A receptor activity would involve the cannabinoid-induced upregulation of 5-HT 2A receptors. As shown in Fig. 5E, repeated GP1a treatment significantly increased the E max (2-fold increase) without significantly affecting the EC 50 , which could be explained, at least in part, by the enhanced cannabinoidinduced overexpression (2-fold) of 5-HT 2A receptors in neuronal cells and in rat PFCx. The role of CB 2 receptors in mediating this phenomenon was identified by either CB 2 or GRK5 shRNA treatment. Indeed, CB 2 or GRK5 shRNA lentiviral particle treatment prevented CP55940-and GP1a-induced increases in serotonin-stimulated 5-HT 2A receptor-mediated Ca 2ϩ release (Fig. 6, A and B).
We also examined the role of GRK5 in the cannabinoid-induced phosphorylation of the CB 2 receptor and enhanced ␤-arrestin 2/ERK interaction (Fig. 7). We found that GRK5 shRNA lentiviral particle treatment reduced the cannabinoid-induced enhanced phosphorylation of the CB 2 receptor and the enhanced ␤-arrestin 2/ERK interaction in a neuronal cell culture model. Here the GP1a-induced increases in CB 2 receptor phosphorylation and ␤-arrestin/ERK interaction in GRK5 shRNA lentivirus-treated cells could be attributed to new rates in synthesis and degradation of GRK5 after GRK5 shRNA lentivirus treatment and/or shifts in GRK6 activity. This evidence indicates that GRK5 is necessary for the cannabinoid-induced up-regulation of 5-HT 2A receptors. Although further evidence is required, it is possible to speculate that GRK5-induced phosphorylation of the CB 2 receptor and subsequent formation of the ␤-arrestin 2/ERK scaffolding complex could be an initiating mechanism contributing to the up-regulation and enhanced activity of 5-HT 2A receptors. Further experiments are needed to demonstrate this hypothesis.
In conclusion, this study provides new insight into the cannabinoid agonist regulation of GRK proteins in rat PFCx and neuronal cell culture. Furthermore, this study is the first to show that GRK5 is involved in the cannabinoid-induced upregulation and enhanced activity of 5-HT 2A receptors in neuronal cells. We also identified mechanisms contributing to the up-regulation of GRK5 in a neuronal cell model. Recent and independent clinical studies have provided evidence indicating that sustained use of nonselective cannabinoid agonists may precipitate the onset of mental disorders associated with dys-function of 5-HT 2A receptor neurotransmission in PFCx, such as anxiety, schizophrenia, and psychosis (5)(6)(7)(51)(52)(53)(54). However, a definitive mechanism by which repeated cannabinoid agonist exposure may be precipitating neuropsychiatric disorders has not been identified. The results presented here and our previous studies (1)(2)(3)(4) suggest that GRK5-mediated enhanced phosphorylation of the CB 2 receptor and enhanced ␤-arrestin 2/ERK interaction would drive the up-regulation of 5-HT 2A receptors and GRK5. Interestingly, a recent report has linked enhanced function and expression of 5-HT 2A receptors in PFCx to enhanced anxiety-like behaviors in rodents (55). Furthermore, the therapeutic benefits of atypical antipsychotics are proposed to be mediated by desensitization of 5-HT 2A receptor signaling in PFCx, particularly pyramidal neurons, which are enriched in 5-HT 2A receptors (28,56). Therefore, this study may facilitate a better understanding of mechanisms underlying the etiology of some neuropsychiatric disorders and adverse effects of chronic exposure to cannabinoids. Understanding the mechanisms underlying the adverse effects of repeated cannabinoid exposure is especially critical because accumulating evidence is showing that selective CB 2 receptor agonists have wide therapeutic application in the treatment of a variety of different conditions (57)(58)(59)(60). This evidence could provide insight into mechanisms that can be targeted to prevent the potential adverse effect while deriving the therapeutic benefits of cannabinoids.