Address correspondence and reprint requests to Allyn C. Howlett, Department of Physiology and Pharmacology, Wake Forest University Health Sciences, One Medical Center Blvd., Winston-Salem, NC 27157 USA. E-mail: firstname.lastname@example.org
Although biochemical and physiological evidence suggests a strong interaction between striatal CB1 cannabinoid (CB1R) and D2 dopamine (D2R) receptors, the mechanisms are poorly understood. We targeted medium spiny neurons of the indirect pathway using shRNA to knockdown either CB1R or D2R. Chronic reduction in either receptor resulted in deficits in gene and protein expression for the alternative receptor and concomitantly increased expression of the cannabinoid receptor interacting protein 1a (CRIP1a), suggesting a novel role for CRIP1a in dopaminergic systems. Both CB1R and D2R knockdown reduced striatal dopaminergic-stimulated [35S]GTPγS binding, and D2R knockdown reduced pallidal WIN55212-2-stimulated [35S]GTPγS binding. Decreased D2R and CB1R activity was associated with decreased striatal phosphoERK. A decrease in mRNA for opioid peptide precursors pDYN and pENK accompanied knockdown of CB1Rs or D2Rs, and over-expression of CRIP1a. Down-regulation in opioid peptide mRNAs was followed in time by increased DOR1 but not MOR1 expression, leading to increased [D-Pen2, D-Pen5]-enkephalin-stimulated [35S]GTPγS binding in the striatum. We conclude that mechanisms intrinsic to striatal medium spiny neurons or extrinsic via the indirect pathway adjust for changes in CB1R or D2R levels by modifying the expression and signaling capabilities of the alternative receptor as well as CRIP1a and the DELTA opioid system.
Compelling behavioral, anatomical, and physiological evidence suggests a strong relationship between cannabinoid and dopamine systems, especially between CB1 cannabinoid receptors (CB1R) and D2 dopamine receptors (D2R) (reviewed in (Fernandez-Ruiz et al. 2010; Smith and Villalba 2008). Both CB1Rs and D2Rs are G protein-coupled receptors highly expressed in the striatum, and are key proteins in the processing of basal ganglia neurotransmission (Sanudo-Pena et al. 1999; van der Stelt and DiMarzo 2003; Fernandez-Ruiz 2009; Lovinger 2010). CB1Rs and D2Rs have been found to be co-localized in the enkephalin-containing medium spiny neurons (MSN) of the striatum, as well as being co-expressed on the axon terminals at the globus pallidus (indirect striatopallidal pathway) (Gerfen et al. 1990; Mailleux and Vanderhaeghen 1992; Szabo et al. 1998; Hermann et al. 2002; Matyas et al. 2006; Crespo et al. 2008; Martin et al. 2008; Van Waes et al. 2012). In addition, CB1Rs and D2Rs are observed in close proximity on soma and dendritic spines of neurons within the ventral striatum (Pickel et al. 2006). The close alignment of these two receptors within protein complexes has been substantiated with fluorescence resonance energy transfer or multicolor bimolecular fluorescence complementation studies in heterologous expression systems (Marcellino et al. 2008; Przybyla and Watts 2010).
In support of their close anatomical co-localization, functional interactions from biochemical data identified that CB1Rs and D2Rs converge to share Gi/o proteins or adenylyl cyclase effectors in striatal membranes (Meschler and Howlett 2001). Other studies identified a switch in the G protein coupling resulting in an increase in cAMP production upon simultaneous treatment by both CB1R and D2R agonists in cultured neonatal striatal cells (Glass and Felder 1997) or by co-expression of both CB1Rs and D2Rs in host cells (Jarrahian et al. 2004; Kearn et al. 2005). However, to date, a direct functional relationship between CB1Rs and D2Rs in MSNs of the basal ganglia has not clearly been established, and as such remains poorly understood.
On the basis of these data, we hypothesize that CB1Rs and D2Rs can interact in striatal neurons to cooperatively regulate cellular function in the basal ganglia in vivo. To determine how one receptor influences the cellular signaling by the other receptor, we used RNA interference to suppress the synthesis of either CB1Rs or D2Rs in rat dorsal striatum. Here, we present data demonstrating a physiologically relevant coupling of CB1R and D2R regulation at the transcript, protein and signaling level in rat basal ganglia circuitry. These results reveal a pattern of functional mimicry for these two G protein-coupled receptors associated with homeostatic adaptations in basal ganglia signaling. These studies also uniquely identify a contribution of the CRIP1a protein to cellular regulation of signaling by both CB1Rs and D2Rs in vivo. CRIP1a is an accessory protein that has been shown to regulate CB1R-mediated tonic inhibition of voltage-dependent Ca2+ channels (Niehaus et al. 2007), presumably at the pre-synaptic terminals. The present data suggest that CRIP1a is important in regulating signal transduction in the striatopallidal pathway.
Material and methods
WIN55212 was purchased from Tocris, and N-propylnorapomorphine (NPA), [D-Ala2, NMe-Phe4, Gly-ol5]-enkephalin (DAMGO) and [D-Pen2, D-Pen5]-enkephalin (DPDPE) were purchased from Sigma, St Louis, MO, USA. [35S]GTPγS was purchased from PerkinElmer, Waltham, MA, USA. Dulbecco's Modified Eagle's Medium was purchased from Gibco Life Technologies (Gibco, Rockville, MD, USA). All other chemicals were reagent grade and purchased from Sigma-Aldrich or specialized suppliers as indicated.
Cloning of adeno-associated viral plasmids and generation of viruses
For viruses designed for RNA interference-directed knockdown of the expression of CB1Rs and D2Rs, oligos encoding short hairpin RNA (shRNA) sequences were cloned into an adeno-associated viral (AAV) plasmid, EGFP-U6-pACP, described previously (Sadri-Vakili et al. 2010). In this plasmid, the cytomegalovirus promoter drives expression of the enhanced green fluorescent protein (EGFP) gene, which is cloned with an intron/polyA sequence derived from SV40. shRNA expression is driven by a murine U6 pol III promoter which is located downstream of the EGFP cassette. The entire EGFP and U6 transgenes are flanked by AAV2 inverted terminal repeats. Each of the synthetic oligos, encoding the shRNA and its respective complement (Sigma-Aldrich), were annealed and ligated into unique BbsI and NheI sites after the U6 promoter. The target sequences were selected by the siRNA target finder program on the GenScript website (https://www.genscript.com/ssl-bin/app/rnai) using the mRNA sequences NM_012784 (rat Cnr1 mRNA) and NM_012547 (rat Drd2 mRNA). Three different AAV-shCB1R and AAV-shD2R viruses were created, and each was individually tested for knockdown efficiency. A control vector consisted of an identical EGFP transgene but encoded a scrambled shRNA that does not correspond to any known rat mRNA sequence (determined from a BLAST search).
For the virus designed to over-express mouse CRIP1a, total mRNA from mouse cortex was isolated and converted to cDNA using a High-Capacity cDNA Archive Kit (Applied Biosystems, Foster City, CA, USA), and CRIP1a cDNA was amplified by reverse transcriptase polymerase chain reaction (PCR). The forward primer 5′-aatttctagaGCCACCATGGGGGACCTACCC-3′ and the reverse primer 5′-ggccaagcttTCAGAGGAAGGACTCCTTATTCACCCA-3′ provided an XbaI restriction site upstream of the translation initiation codon, a Kozak sequence and a HindIII site downstream of the translation stop codon of the CRIP1a fragment. The 0.5 kb PCR product was verified by sequence analysis, and subcloned into pACP at the XbaI and HindIII sites. This plasmid (CRIP1a-pACP) consists of two AAV2 inverted terminal repeats flanking the cytomegalovirus promoter, CRIP1a cDNA, and an intron and polyadenylation signal sequence derived from SV40.
Packaging of all recombinant AAVs was carried out according to a standard triple transfection protocol to create pseudotyped AAV2/10 virus (Xiao et al. 1998). The AAV2/10 rep/cap plasmid provides the AAV2 replicase and AAV10 capsid genes (Gao et al. 2002; De et al. 2006), and adenoviral helper functions were provided by the pHelper plasmid (Stratagene). AAV-293 cells were transfected with 10 μg of pHelper and 1.15 pmol each of AAV2/10 and AAV vector plasmids to be used in this study. The cells were harvested 48 h later, and clarified viral lysates were isolated from the cell pellets. The virus was pooled, aliquoted, and stored at −80°C. AAV-vector stocks were titered by real-time quantitative PCR (qPCR) (Eppendorf Realplex) using primers and probe sets designed to amplify a sequence in the SV40 intron.
Stereotaxic intracranial injections of AAV-viruses
Adult male Sprague–Dawley rats weighing 280–310 g (Harlan Inc., Indianapolis, IN, USA) were used, and experimental procedures were approved by the Wake Forest University Institutional Animal Care and Use Committee and followed the ARRIVE guidelines (Kilkenny et al. 2010). Prior to surgery, animals were anesthetized with a combination of ketamine (100 mg/kg) and xylazine (8 mg/kg, i.p.). Rats were placed into a stereotaxic frame (Kopf), 1 mm holes were drilled into the skull, and a 10 μL Hamilton syringe fitted within a 30 gauge needle was used for viral delivery. All injections were made into the dorsal striatum using the following coordinates relative to bregma, with the tooth bar set to + 0.5 mm: anteroposterior (AP): + 0.5 mm, mediolateral (ML): ± 2.8 mm, dorsoventral (DV): + 4.4 mm relative to stereotaxic zero (Paxinos and Watson 1997) (see Figure S2). AAV-scramble, a mixture of three AAV-shCB1R, a mixture of three AAV-shD2R, or AAV-CRIP1a (4 μL) were unilaterally injected at a rate of 0.5 μL/min. To reduce back-flow of viral solution, the needle remained in place ten min post-injection before being slowly removed. To avoid issues of lateralization, injections of AAV were made in either the left or right hemispheres with equal frequency within each experimental group of rats. Animals were sutured, and revived on a heating pad before being returned to their home cages (two animals per cage).
Analysis of gene expression in AAV-injected brains
Rats were killed at 3, 5, 10, 17, 21, 30, or 56 days post-injection of AAV, and brains were dissected and placed in an ice cold brain matrix. Coronal brain slices (2 mm) were taken at the site of injection, placed on a chilled dissecting block, and 2 mm-diameter circular brain punches were collected ipsilaterally (treated) and contralaterally (control). Total RNA was isolated and purified using an RNeasy Mini Kit (Qiagen, Valencia, CA, USA). Total RNA (1 μg) was reverse transcribed into cDNA using a High-Capacity cDNA Archive Kit (Applied Biosystems, Foster City, CA, USA). Real-time qPCR was performed using TaqMan Universal PCR Master Mix and specific TaqMan primer-probe MGB assay sets (Applied Biosystems) for the following genes: 18s ribosomal RNA, neuron-specific enolase 2 (eno2), CB1 cannabinoid receptor (cnr1), CRIP1a (cnrip1), delta opioid receptor (Oprd1), D2 dopamine receptor (drd2), mu opioid receptor (Oprm1), pro-enkephalin (PENK), pro-dynorphin (PDYN). Data were analyzed using the ΔΔCT method comparing the ipsilateral to the contralateral side (Gerald et al. 2006), and eno2 on the side contralateral to the injection served as the reference standard.
Generation of CRIP1a antibody
We generated a rabbit polyclonal antibody against rat CRIP1a corresponding to amino acids D20-F32 (AbD20) (Figure S2a). The peptide was synthesized and conjugated by disulfide formation with keyhole limpet hemocyanine, rabbit antibodies generated and affinity-purified, and titered using an ELISA (GenScript, Piscataway, NJ, USA). To verify avidity and specificity for CRIP1a, we compared this novel antibody with a previously characterized CRIP1a antibody, CRIP1a AbK148, which recognizes the last 17 amino acids of CRIP1a (K148-L164) (Elphick et al. 2004; Niehaus et al. 2007). Western blot analyses of striatal crude homogenates identified a band at the same mobility at various antibody dilution factors (1 : 50, 1 : 100, 1 : 300, 1 : 1000) using either CRIP1a AbD20 or AbK148 (data not shown).
Immunohistochemistry of protein levels in brain slices
Effects of shCB1R and shD2R knockdown, and CRIP1a over-expression on total protein levels of CB1R, CRIP1a, D1R, DOR1, MOR1, cAMP-response element binding protein (CREB), phospho-CREB, ERK, and phospho-ERK were quantitated using a modified immunohistochemistry technique (Kearn 2004). Anesthetized animals were decapitated, brains were dissected, snap-frozen in isopentane cooled by dry ice, and stored at −80°C. Frozen brains were sectioned at 40 μm using a cryostat microtome, slices were placed into a 24-well plate containing frozen phosphate-buffered formalin (1.5 mM KH2PO4, 2.7 mM KCl, 8 mM Na2HPO4, 150 mM NaCl; 30% sucrose (w/v); 3% paraformaldehyde (v/v), pH 7.4), and stored at 4°C. Slices were washed in Tris-buffered saline (TBS) (20 mM Tris-HCl, pH 8.0, 150 mM NaCl), blocked and permeabilized overnight at 4°C in Odyssey Blocking Buffer (LI-COR Biosciences, Lincoln, NE, USA) containing 0.1% Tween-20 and 1 mM sodium orthovanadate. Slices were incubated at 4°C for 18 h with primary antibodies: CB1 cannabinoid receptor (CB1R, Santa Cruz Biotechnology, Santa Cruz, CA, USA; 1 : 750), CRIP1a (AbK148; 1 : 300; AbD20; 1 : 300), D2 dopamine receptor (D2DR, Santa Cruz; 1 : 500), D1 dopamine receptor (D1DR, Santa Cruz; 1 : 300), delta opioid receptor 1 (DOR1, Santa Cruz; 1 : 500), mu opioid receptor 1(MOR1, Santa Cruz; 1 : 500), total ERK1/2 (ERK2, Santa Cruz; 1 : 1000), phosphoERK1/2 (p-ERK, Santa Cruz; 1 : 500), total CREB (CREB, Cell Signaling; 1 : 1000), phosphoCREB (p-CREB, Cell Signaling; 1:500). Slices were washed in TBS containing 0.1% Tween 20 (TBS-T), and incubated for 2 h with a secondary goat anti-rabbit (1:1500) or goat anti-mouse antibody (1 : 1500) conjugated to infrared dyes, washed in TBS-T and allowed to dry overnight. The fluorescent immunocomplexes were detected using the LI-COR Odyssey imaging system and software program (LI-COR Biosciences, Lincoln, NE, USA), and integrated densities from gray-scale images were determined for demarcated regions of interest (dorsal striatum, globus pallidus, entopeduncular nucleus) using Image J software (National Institutes of Health, Bethesda, MD, USA, http://rsb.info.nih.gov/ij/) (see Figure S1D,E,F). Changes in protein levels were determined by comparing the integrated densities between the AAV-treated and the untreated contralateral hemisphere. For phospho-CREB and phospho-ERK levels, normalization was established to total CREB or ERK, respectively.
[35S]GTPγS binding in brain slices
Agonist-induced G protein activation by CB1R, D2R, MOR1, and DOR1 was quantitated by [35S]GTPγS, binding assays. Receptor/G protein coupling was assayed in rat brain sections using [35S]GTPγS autoradiography (Sim et al. 1995). Rat brain sections (20 μm) were pre-incubated for 10 min in TME buffer (50 mM Tris-HCl, 3 mM MgCl2, 0.2 mM EGTA, 100 mM NaCl, pH 7.4), then 15 min with 1 mM GDP and 1 μM DPCPX at 25°. Sections were incubated in assay buffer with 1 mM GDP and 1 μM DPCPX, 0.04 nM [35S]GTPγS, with or without various agonists for 2 h at 25°. Agonists included 1 μM WIN 55212-2 (CB1R), 3 μM NPA (D2R), 3 μM DAMGO (mu opioid), and 3 μM DPDPE (delta opioid). The sections were then washed, exposed to X-ray film, and analyzed as described previously (Sim et al. 1995). Agonist-stimulated activity was calculated by subtracting the optical density in basal sections (GDP only) from that of agonist-stimulated sections and results are expressed as percent stimulation over basal activity.
Data from treated (ipsilateral) sides of the brain were compared to untreated control tissue isolated from the same region on the contralateral side of the brain, such that each animal served as its own control. Statistical differences were determined by paired comparisons between treated and untreated tissue using Student's paired two-sided t-test analyses on Prism 4 or InStat software (GraphPad Software Inc., San Diego, CA, USA).
AAV-shRNA-mediated CB1R and D2R knockdown in vivo
CB1Rs are expressed in the GABAergic MSNs of the striatum that project both directly and indirectly to their output nuclei (Herkenham et al. 1991; Mailleux and Vanderhaeghen 1992; Martin et al. 2008). The MSNs of the direct pathway express D1 receptors (striatonigral pathway), whereas the MSNs of the indirect pathway preferentially express D2 receptors (striatopallidal) (Hermann et al. 2002; Pickel et al. 2006; Tepper et al. 2007). Our goal was to directly and selectively manipulate the expression of either CB1Rs or D2Rs in the dorsal striatum to determine the extent to which these receptors regulate signal transduction and gene expression in intact basal ganglia tissue.
We characterized the kinetics of CB1R and D2R knockdown by shRNA designed specifically against either CB1R or D2R. The AAV2/10 virus transduces the cell bodies at the injection site, and is not further delivered by trans-synaptic mechanisms (Xiao et al. 1998). A single unilateral injection of AAV-shCB1R in the rat dorsal striatum produced a reduction in CB1R mRNA levels to 58 ± 4% compared to control at day 21, and remained significantly reduced to day 56. Analysis of AAV-shCB1R-treated striata revealed a time-dependent reduction in D2R mRNA that accompanied the reductions observed for CB1R mRNA (Fig. 1a). Injection of AAV-shD2R into dorsal striatum resulted in noticeable reductions in D2R mRNA levels starting at day 3 (15 ± 3%) and continuing until day 56 (23 ± 6%). The maximum decline in D2R transcript levels that AAV-shD2R was capable of producing occurred between days 10 and 28 (52 ± 6%) (Fig. 1b). Interestingly, CB1R transcript levels were significantly reduced as a result of knockdown of D2Rs in rat striatum (Fig. 1b). It appears that the transcriptional regulation or mRNA stability of these receptors is tightly coupled, as AAV-mediated knockdown of one receptor coincides with mRNA reductions in the other receptor.
Using an immunohistochemistry procedure that could quantify receptor protein in identified regions of interest in brain slices, we established the extent of receptor protein loss following AAV-mediated knockdown of CB1Rs or D2Rs. Because maximum knockdown of CB1R and D2R mRNA occurred between days 15 and 30, we examined immunoreactive receptor densities in slices from brains at 17 days post-injection of AAV-shCB1R or AAV-shD2R. Comparison of regions of interest (see Figure S1) between ipsilateral (injected) and contralateral (non-injected) hemispheres from AAV-shCB1R- injected animals showed a significant decrease in CB1R immunoreactive protein in the dorsal striatum (21 ± 5%), as well as in the outflow projections to the globus pallidus (13 ± 2%) and entopeduncular nucleus (26 ± 3%) (Fig. 1c). This is consistent with a scenario of CB1R synthesis in cell bodies of striatal MSNs, and subsequent translocation to the pre-synaptic termini in the ‘indirect’ pathway of the basal ganglia [see review (Mackie 2005)]. As seen with transcript levels, knockdown of the CB1R was associated with a statistically significant reduction in D2R protein in the dorsal striatum (15 ± 6%) and entopeduncular nucleus (17 ± 3%), with a decrease within the globus pallidus (8 ± 1%) (Fig. 1c).
Brain slices from rat dorsal striata injected with AAV-shD2R exhibited a significant knockdown of immunoreactive D2R in the dorsal striatum (35 ± 6%), globus pallidus (16 ± 1%), and entopeduncular nucleus (16 ± 2%) (Fig. 1d). Significant reductions of immunoreactive CB1R were observed in the dorsal striatum (14 ± 6%) and entopeduncular nucleus (18 ± 4%), as well as a reduction in the globus pallidus (8 ± 1%) (Fig. 1d). Although a parallel relationship was observed between CB1R and D2R knockdown, specific knockdown of either receptor failed to alter D1R transcript levels (data not shown) or total protein levels (Fig. 1c and d). This evidence supports the notion that whereas CB1Rs exert a regulatory influence on D2Rs in the “indirect” pathway, there is very limited or no regulation of the D1 receptor system in neurons comprising the ‘direct’ pathway in the basal ganglia.
AAV-CRIP1a-mediated CRIP1a over-expression in vivo
Analysis of dorsal striatal tissue from AAV-shCB1R or AAV-shD2R-treated animals revealed that knockdown of CB1Rs and D2Rs induced a 3- and 4-fold increase in the mRNA levels of CRIP1a, respectively (Fig. 1a and b). These increases in CRIP1a mRNA were accompanied by an approximately 25% increase in levels of CRIP1a immunoreactive protein in the dorsal striatum as well as significant, but less robust, increases in the outflow nuclei (Fig. 1c and d). Interestingly, the time-course for the increase in striatal CRIP1a mRNA and protein mirrored the decreases in striatal CB1R and D2R mRNA and protein. Our data showing increased levels of protein in the outflow nuclei of the dorsal striatum suggest that CRIP1a gains access to the axonal compartment and can establish a synthesis/degradation equilibrium at that locus as well as at the cell soma.
As striata from both AAV-shCB1R- and AAV-shD2R-injected animals displayed a marked increase in CRIP1a mRNA and protein levels, we took the complimentary approach of over-expressing CRIP1a in vivo to investigate its role in the regulation of CB1R- and D2R-mediated functions in rat striatum. AAV-CRIP1a was unilaterally injected into the rat dorsal striatum and differences in gene expression between ipsilateral and contralateral hemispheres were assessed 3, 5, 10, and 17 days post-injection. CRIP1a mRNA levels were significantly increased at all measured time-points. Striata from AAV-CRIP1a-injected rats displayed 5- to 6-fold increases in levels of CRIP1a transcript as early as 3 days and at least as long as 17 days post-injection (the last time-point examined) (Fig. 2a). During this time-course, no measurable deviations in the expression of D2R mRNA were detected when compared to the contralateral hemisphere (Fig. 2a). However, CB1R mRNA increased transiently at the day 3 time-point following the AAV-CRIP1a injection, before returning to pre-injection expression levels at day 5 (Fig. 2a).
Viral delivery of shRNA transgenes has been shown to induce immunomodulatory responses at the injection site with some, but not all, shRNA sequences (McBride et al. 2008). Evidence suggests that CB1R expressed by both glia and microglia participate in immune responses (Klein et al. 2003; Cabral and Marciano-Cabral 2005; Sheng et al. 2005). Thus, we hypothesized that the transient CB1R up-regulation seen within the earliest days following the AAV-CRIP1a injection might be because of the infiltration of immune cells and/or gliosis in response to the viral injection. To investigate this, RNA from the injection sites of AAV-shCB1R, AAV-shD2R, and AAV-CRIP1a animals were subjected to qPCR, and analyzed for glial fibrillary acidic protein (GFAP), an astrocytic marker (Smith and Eng 1987) and IBA1, a protein specifically expressed and up-regulated during macrophage/microglia activation (Ito et al. 2001). Analysis of striatal gene expression from AAV-injected animals showed that the mRNA levels of GFAP and IBA1 were significantly increased at only the day 3 time-point and returned to baseline by day 5 (data not shown). Of particular note, control rats injected intra-striatally with vehicle (Dulbecco's Modified Eagle's Medium) or AAV-shScramble also displayed increased GFAP and IBA1 expression at the day 3 time-point only, compared with the control brain hemispheres, in which the microglial marker IBA1 was expressed at levels that were 1% of eno2 (data not shown). These findings suggest that the short-lived immune response results from the injection process itself, indicative of inflammatory cell infiltration immediately following the injection. During the time-course of the study, no statistical differences in eno2 were detected between the ipsilateral and contralateral hemispheres. This finding indicates that no decreases in neuronal mass, characteristic of neurodegeneration, were investigated in response to the AAV injection.
For detection of CRIP1a protein, we generated CRIP1a AbD20 (see Figure S2). Rats were killed at 17 days post-injection with AAV-CRIP1a, by which time AAV-mediated transgene expression had peaked and reached equilibrium. AAV-CRIP1a successfully over-expressed CRIP1a protein levels over a 17-day time period, not only in the striatum (42 ± 8%) but also in the globus pallidus (43 ± 5%) and entopeduncular nucleus (42 ± 6%) (Fig. 2b). However, no detectable, sustained differences in CB1R, D2R or D1R mRNA or protein levels were observed as a result of over-expression of CRIP1a in the dorsal striatum (Fig. 2a and b). The lack of changes in CB1R or D2R levels indicates a non-reciprocal relationship between the expression patterns of these proteins in vivo.
CB1R and D2R knockdown influence G protein activation and signal transduction
To investigate the effects of AAV-shCB1R and AAV-shD2R knockdown and CRIP1a over-expression on receptor-stimulated G protein activation, [35S]GTPγS binding assays were performed on coronal brain slices from animals killed 17 days post-injection. Knockdown of either the D2R or CB1R in the striatum was associated with a 19 ± 7% or 19 ± 4% reduction, respectively, of striatal NPA [3 μM]-stimulated G protein activation (Fig. 3a). No appreciable changes in D2R-stimulated G protein activation were observed following striatal over-expression of CRIP1a. Fig. 3b shows the effects of CB1R and D2R knockdown and CRIP1a over-expression on [35S]GTPγS binding stimulated by the CB1R agonist WIN55212 in the globus pallidus. Striatal knockdown of CB1Rs did not result in measurable differences in CB1R-stimulated [35S]GTPγS binding in the globus pallidus, the site of pre-synaptic CB1R (Fig 3b). The lack of significant reduction in WIN55212-stimulated [35S]GTPγS binding may be attributed to the expression of CB1R in post-synaptic globus pallidus neurons, which would not have been altered by the knockdown of the pre-synaptic CB1Rs. D2R knockdown significantly reduced CB1R-stimulated G protein activation (17 ± 3%) in the globus pallidus, an effect that could represent the response of either pre-synaptic because of direct effects of the CB1R or post-synaptic neurons because of the effects of released neurotransmitters (Fig. 3b). CRIP1a over-expression produced a 16 ± 8% decrease in CB1R-stimulated [35S]GTPγS binding in the globus pallidus.
Effects of AAV-shCB1R and AAV-shD2R knockdown and CRIP1a over-expression on levels of phosphoCREB and phosphoERK were quantitated in rats killed at 17 days post-treatment. Comparison of immunofluorescence between hemispheres on coronal brain slices showed that the total protein levels of ERK1/2 and CREB were not affected following any of the viral injections performed (data not shown); therefore, phosphoERK and phosphoCREB immunostaining could be normalized to the total ERK and CREB protein. Striatal CB1R knockdown was associated with a significant decrement in phosphoERK levels in the striatum (18 ± 3%) as well as the outflow projections to the globus pallidus (17 ± 2%) and entopeduncular nucleus (21 ± 2%), reflecting changes in either pre-synaptic or post-synaptic signal transduction or both (Fig. 3c). CRIP1a over-expression in the dorsal striatum was also associated with a marked reduction in phosphoERK levels in the striatum (16 ± 3%), globus pallidus (17 ± 4%) and entopeduncular nucleus (15 ± 1%) (Fig. 3c). The levels of phosphoCREB were not altered by striatal CB1R knockdown or CRIP1a over-expression (Fig. 3d), suggesting that, in contrast to ERK, CREB regulation may not be a mechanism for CB1R regulation of gene expression within neurons in either the striatum or the post-synaptic neurons in the outflow nuclei.
Striatal D2R knockdown had a region-specific effect on the phosphoERK levels, such that levels of phosphoERK were reduced in the striatum (18 ± 2%), but significantly increased in the globus pallidus (11 ± 2%) and entopeduncular nucleus (11 ± 3%) (Fig. 3c). The levels of phosphoCREB were not altered by D2R-knockdown, except in the globus pallidus, where a significant increase in phosphoCREB (9 ± 1%) was observed (Fig. 3d). As phosphoCREB is likely to be found in nuclei of the post-synaptic neurons, one could postulate that increases in both phosphoERK and phosphoCREB are occurring in post-synaptic neurons associated with D2R reduction in the striatal MSNs.
Striatal CB1R and D2R evoke regulation of opioid peptide systems
As CB1R and D2R are co-expressed and co-localized in the enkephalin-containing GABAergic MSNs of the striatum (Pickel et al. 2006), we sought to identify downstream influences exerted by CB1R and D2R knockdown on opioid physiology. Within days following striatal knockdown of either CB1R or D2R, the mRNA expression of opioid peptide precursors PENK and PDYN in the dorsal striatum was significantly decreased (Fig. 4a and b). Gene expression of the delta (DOR1) but not mu (MOR1) opioid receptor was increased in the same striatal tissue samples. Time-course analysis indicated that mRNA levels of DOR1 were up-regulated within 7 days after the maximum reduction of PENK associated with D2R knockdown (Fig. 4b). CRIP1a over-expression also evoked significant reductions in PENK and PDYN transcript levels, and within days, mRNA levels of DOR1 were significantly up-regulated (Fig. 4c). There were no significant alterations in the gene expression of MOR1 at any time-point. This time-course suggested that an alteration in the synthesis or degradation of DOR1 followed a change in availability of the endogenous agonist enkephalin. These observations might suggest a direct effect on gene expression or mRNA stability within neurons, such that modifications in CB1R, D2R, or CRIP1a levels contribute to the suppression of opioid peptides concurrently with increased expression of DOR1. Alternatively, a decrease in enkephalin leading to reduced local DOR1 stimulation might trigger DOR1 up-regulation. Finally, a multi-cellular regulatory mechanism may come into play in the dorsal striatum.
CB1R, D2R, and CRIP1a influence DOR1 protein levels and signal transduction
We investigated the responses to opioid transcriptional alterations with studies on MOR1 and DOR1 receptor expression and function. Fig. 5a shows that there were no changes in the immunoreactive protein levels of MOR1 at 17 day post-injection of any of the AAV transcripts. Striatal CB1R knockdown animals displayed significant increases in immunoreactive DOR1 protein in the dorsal striatum (25 ± 7%), globus pallidus (25 ± 6%), and entopeduncular nucleus (21 ± 7%). Striatal D2R knockdown resulted in significant increases in DOR1 protein in the striatum (27 ± 6%) and globus pallidus (15 ± 4%), and a notable increase in the entopeduncular nucleus (10 ± 4%) that trended toward significant at p < 0.10. CRIP1a over-expression resulted in significantly increased DOR1 protein in the dorsal striatum (18 ± 5%), globus pallidus (11 ± 2%), and entopeduncular nucleus (23 ± 3%) (Fig. 5b).
To determine if these receptors were functionally coupled to Gi/o proteins, we evaluated agonist-stimulated G protein activation. D2R knockdown and CRIP1a over-expression produced significant increases in DPDPE-stimulated [35S]GTPγS binding in dorsal striatum (43 ± 14%; 29 ± 5%, respectively) (Fig. 5c). Although knockdown of CB1R resulted in significantly increased striatal DOR1 protein (27 ± 8%), no changes were observed in DPDPE-stimulated [35S]GTPγS binding. Neither CB1R or D2R knockdown, nor CRIP1a over-expression exerted any significant effect on DAMGO-stimulated [35S]GTPγS binding in the dorsal striatum (Fig. 5c).
Physiological and biochemical evidence for the interaction between cannabinoid and dopamine receptor systems has continued to accumulate over the last two decades [reviewed in (Fernandez-Ruiz et al. 2010)]. Despite their close pharmacological interactions, few studies have directly investigated the biological underpinnings involved in the synthesis and expression of CB1R and D2R during manipulation of the other receptor. In this study, we utilized RNA interference to inhibit the synthesis of CB1R or D2R in rat dorsal striatum to investigate the interactions between cannabinoid, dopamine, and opioid receptor systems.
These studies demonstrate that CB1R and D2R are tightly coupled at the transcript, protein and signaling level in rat dorsal striatum. The knockdown of either CB1R or D2R resulted in the concomitant suppression of the expression of the other receptor in the cell bodies of the dorsal striatum and in the projections to the globus pallidus and entopeduncular nucleus. It is reasonable to postulate that the decrease in receptors in the outflow nuclei was because of the use-dependent, agonist-driven processes of internalization and degradation of pre-synaptic receptors, coupled with the reduced ability to replenish these receptors as a result of the RNA interference at the soma. Our data suggest that a new equilibrium is established within 15–30 days, after which, the RNA interference becomes less effective and the equilibrium tends to become re-established toward control at the end of the 56-day post-injection period.
Consistent with our findings, studies of subchronic pharmacological blockade of the D2R in rats with the antagonist haloperidol resulted in an increase in striatal D2R binding and G protein activation as expected with decreased use, but also an increase in striatal CB1R binding and G protein activation in the outflow nuclei (Andersson et al. 2005). Levels of D2R in the dorsal striatum in response to subchronic pharmacological blockade of the CB1R with rimonabant were increased in rats (Crunelle et al. 2011). Combined electrophysical and biochemical studies demonstrate that activation of the striatal dopaminergic system up-regulates anandamide and CB1 receptor levels in the striatum, and CB1R has been postulated to be a downstream effector of D2 receptors in the inhibition of GABAergic neurotransmission (Centonze et al. 2004). Our results suggest a coupling of the regulation and expression between CB1R and D2R, and as such may provide insight into reward and learning based mechanisms where induction of corticostriatal long-term depression (Gerdeman et al. 2002) or striatal synaptic plasticity is of underlying importance (i.e., drug addiction and motor movement disorders).
In contrast to our current findings, genetic deletion using knockout mice indicated that the expression of CB1R and D2R was inversely related. Life-long, total loss of CB1R resulted in up-regulated expression of D2R in the striatum (Houchi et al. 2005), probably as the result of developmental compensatory alterations in the GABAA and NMDA receptors in the basal ganglia circuitry of CB1R knockout mice (Warnault et al. 2007). Life-long, total loss of D2R resulted in up-regulated expression of CB1R in the striatum (Thanos et al. 2011). The general conclusion from these studies is that loss of cannabinergic or dopaminergic signaling is compensated through the up-regulation of the other receptor system to maintain functional homeostasis as the basal ganglia circuitry develops. Our examination of homeostatic mechanisms in response to partial CB1R or D2R knockdown in normal adult animals after the basal ganglia circuitry has already been established, may be more reminiscent of adult neurodegenerative diseases such as Huntington's disease, in which the MSNs suffer a loss of CB1R and D2R concurrently early in their degeneration process (Glass et al. 2000; Blazquez et al. 2011). Also in contrast to knockout mice, the clearly delineated AAV-shRNA injection locus allows receptor knockdown to be limited to cell soma within the dorsal striatum, and thus, CB1R on glutamatergic terminals and D2R on nigrostriatal terminals are not targeted.
The partial deficit observed in striatal D2R, evoked by either D2R or CB1R knockdown, was reflected in comparable decrements in D2R-G protein activation and reductions in phosphoERK levels in the dorsal striatum. In contrast, the shRNA-mediated deficit in D2R in the globus pallidus was associated with increased phosphoERK as well as phosphoCREB, indicative of signal transduction in the post-synaptic neurons, perhaps because of an amelioration of D2R-mediated suppression of neurotransmission. As knockdown of CB1R or D2R failed to alter D1R transcript, protein or receptor activity, the dynamic changes in receptor expression, activity, and downstream signaling to the ERK pathway appear to be selective for the D2R-positive striatopallidal pathway.
Our findings are unique in reporting an interaction between CB1R and D2R at the transcript level in the striatum. Gene expression of neuronal CB1R can be regulated by ERK (Lim et al. 2003), a potential signaling mechanism by which D2R could influence the expression at the transcription level. Similarly, ERK was able to stimulate the D2R promoter (Takeuchi et al. 2002), providing a common mechanism by which both receptors might be regulated concurrently.
We also identified an augmentation in expression of CRIP1a, a CB1R-associated protein that interacts with the distal carboxy terminus of CB1R, but not CB2R (Niehaus et al. 2007), whose role in cellular physiology has remained elusive. Although a direct functional link between CRIP1a and CB1R has yet to be established, the mRNA levels of both CRIP1a and CB1R were reduced in epileptic hippocampal tissue compared with control, likely associated with decreased CB1Rs in glutamatergic terminals (Ludanyi et al. 2008). Immunofluorescence staining indicated that CRIP1a was in close proximity to CB1R in the pre-synaptic terminals of retina cone photoreceptors (Hu et al. 2010).
Our data on CRIP1a expression revealed the presence of CRIP1a transcript and protein in the striatum, globus pallidus, and entopeduncular nucleus, comprising the striatopallidal pathway. Knockdown of either CB1Rs or D2Rs resulted in up-regulation of CRIP1a transcript and protein, suggesting a novel role for the CRIP1a protein in dopamine receptor systems. We hypothesized that the observed changes in CRIP1a could be related mechanistically to the cellular signaling patterns of CB1Rs and D2Rs, and developed a CRIP1a over-expressing virus to evaluate this hypothesis. CRIP1a over-expression reduced phosphoERK in the striatum and its outflow projections, recapitulating the effects of CB1R knockdown. Niehaus and colleagues (Niehaus et al. 2007) suggested that the physiological role of CRIP1a was to reduce CB1R-mediated tonic inhibition of Ca2+ channels, but data supporting effects on other CB1R-mediated responses is just beginning to be appreciated. One possible mechanism by which CRIP1a could reduce ERK phosphorylation is indirectly through CRIP1a regulation of CB1R signaling, inasmuch as our data indicate that there were no effects of CRIP1a over-expression on CB1R mRNA or protein levels. This finding is consistent with data from co-transfected cell lines, which also showed no effect on CB1R expression when CRIP1a was over-expressed (Niehaus et al. 2007).
A prominent finding was the down-regulation of mRNA for the opioid peptides PDYN and PENK that accompanied knockdown of CB1Rs or D2Rs. The down-regulation in opioid peptide mRNAs was followed in time by increased DOR1 mRNA expression and augmented protein expression along the striatopallidal pathway. Interestingly, these same sequelae were mimicked by over-expression of CRIP1a. Data from our lab in N18TG2 neuronal cells stably over-expressing CRIP1a shows that PENK mRNA is significantly reduced and concomitantly DOR1 mRNA is up-regulated when compared with WT cells (Blume et al. 2010). A possible mechanism for the observed changes in opioid physiology is that CB1Rs, D2Rs (and suppressed CRIP1a levels) contribute to the expression of opioid peptides in MSNs. Intra-striatal administration of selective inhibitors of ERK phosphorylation blocked opioid peptide gene expression in rats (Shi and McGinty 2006). Thus, one possible mechanism for the observed reductions in PENK and PDYN gene expression may be related to the reduction in striatal ERK phosphorylation resulting from CB1R or D2R knockdown, and CRIP1a over-expression. Reduced enkephalin-stimulated DOR1 may lead to DOR1 up-regulation, as increased DOR1 binding has been observed in quantitative autoradiography studies in enkephalin knockout mice (Brady et al. 1999). Alternatively, the observed up-regulation in DOR1 expression may be an adaptive response that mediates recovery of the disinhibition of the striatopallidal pathway produced by the loss of CB1Rs and/or D2Rs receptors in the dorsal striatum.
The ability to achieve specific time-dependent knockdown of CB1Rs and D2Rs in a discrete population of basal ganglia neurons is a significant advantage over knockout mice in the investigation of pathologies where reductions, but not total loss of CB1R or D2R is observed. This study may have therapeutic relevance in diseases of the basal ganglia in which CB1Rs and D2Rs are key players in the underlying pathophysiology, such as Parkinson's and Huntington's diseases as well as drug addiction. The present data add to a growing body of evidence for a cellular and molecular interaction between CB1Rs and D2Rs at the behavioral, systems, and cellular-molecular levels.
This study was supported by US Public Health Services grants: R01-DA03690; R01 DA014030 R21-DA025321; K01-DA024763, P50-DA006634, F32-DA026295, T32-DA00724, and F31-DA032215. The authors have no conflicts of interest to declare.