Skip to main content
NCBI home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2006 May 9;103(20):7877–7882. doi: 10.1073/pnas.0602661103

Adenosine A2a blockade prevents synergy between μ-opiate and cannabinoid CB1 receptors and eliminates heroin-seeking behavior in addicted rats

Lina Yao *,†,‡,§, Krista McFarland ¶,‖, Peidong Fan *,†,‖, Zhan Jiang *,†, Takashi Ueda *,†, Ivan Diamond *,†,‡,**,††,§
PMCID: PMC1458620  PMID: 16684876

Abstract

Relapse is the most serious limitation of effective medical treatment of opiate addiction. Opiate-related behaviors appear to be modulated by cannabinoid CB1 receptors (CB1) through poorly understood cross-talk mechanisms. Opiate and CB1 receptors are coexpressed in the nucleus accumbens (NAc) and dorsal striatum. These regions also have the highest density of adenosine A2a receptors (A2a) in the brain. We have been investigating the postsynaptic signaling mechanisms of μ-opiate receptors (MORs) and CB1 receptors in primary NAc/striatal neurons. In this article, we present evidence that MOR and CB1 act synergistically on cAMP/PKA signaling in NAc/striatal neurons. In addition, we find that synergy requires adenosine and A2a. Importantly, an A2a antagonist administered either directly into the NAc or indirectly by i.p. injection eliminates heroin-induced reinstatement in rats trained to self-administer heroin, a model of human craving and relapse. These findings suggest that A2a antagonists might be effective therapeutic agents in the management of abstinent heroin addicts.

Keywords: PKA, addiction, nucleus accumbens, gene activation

Opiate addiction is a world-wide public health problem with serious socioeconomic ramifications. A major limitation of effective medical treatment is the craving and relapse that develops during attempted abstinence. Opiates bind to three opioid receptors: the δ-opioid receptor (DOR), μ-opioid receptor (MOR), and κ-opioid receptor (KOR) receptors. MOR and DOR are implicated in reward for heroin and morphine, whereas KOR is implicated in aversion (1). MOR antagonists reduce opiate self-administration, and constitutive deletion of the MOR attenuates opiate-induced conditioned place preference (CPP) (2, 3). Moreover, selective MOR blockade is sufficient to induce conditioned aversion in morphine-dependent animals, presumably because of unopposed activation of KOR (1). The nucleus accumbens (NAc) mediates reward and reinforcement of addictive agents. Thus, inactivation of the NAc core inhibits heroin self-administration (4). We have been investigating postsynaptic signaling mechanisms activated by opiate receptors in NAc/striatal neurons. We reported that brief exposure of primary NAc/striatal neurons to MOR agonists for 10 min activates cAMP/PKA signal transduction followed by stimulation of cAMP response element (CRE)-mediated gene expression hours later (5). We also found that paradoxical stimulation of cAMP/PKA signaling by Gi-coupled MOR depends on preferential binding of the MOR to Gαi3βγ. Thus, activation of the MOR appears to release βγ subunits from Gαi3; released unbound βγ subunits stimulate adenylyl cyclase (AC) II and IV to increase cAMP production. In turn, this transient increase in cAMP activates PKA and CRE-dependent gene transcription (5).

Signaling of G protein-coupled receptors can be modulated by G protein regulators. Recent evidence suggests that an activator of G protein signaling 3 (AGS3), regulates Gαi3-coupled receptor signaling by competing with βγ subunits for binding to Gαi3-GDP (6, 7). By preventing the reassociation of unbound βγ subunits with Gαi3-GDP, AGS3 appears to selectively prolong the action of free βγ dimers while simultaneously inactivating Gαi3. We found that AGS3 is required for MOR-induced PKA signaling via βγ subunits (8). Importantly, antisense oligonucleotide against AGS3 expressed in the core of the NAc (but not the shell) abolishes reinstatement of heroin self-administration (8). Therefore, AGS3 in the NAc appears to regulate MOR-induced cAMP/PKA signaling and heroin-seeking behavior.

Like opiates, marijuana is also widely abused, but the intracellular mechanisms underlying its euphoriant and addictive properties are poorly understood. Tetrahydrocannabinoid (THC), the addictive constituent of marijuana and hashish, is an agonist for the Gi-coupled cannabinoid CB1 receptor (CB1). THC induces CPP in mice (9) and rats (10) and self-administration in squirrel monkeys (11). In addition, robust and consistent self-administration of the potent synthetic cannabinoid agonist WIN 55212-2 in drug-naïve mice has also been reported (12, 13). CB1 is expressed in the NAc (14) and appears to contribute to opiate-related behaviors through poorly understood cross-talk mechanisms. Thus, morphine self-administration (15) and CPP (16) are decreased in mice lacking CB1, and the opiate receptor antagonist naltrexone reduces THC self-administration in squirrel monkeys.‡‡ Cross-talk between MOR and CB1 is also suggested by findings in rats that heroin reinstates cannabinoid-seeking behavior (17) and that the CB1 agonist HU-210 reinstates heroin-seeking behavior (18). Consistent with these observations, CB1 and MOR are coexpressed in the NAc (19), and several investigators propose that cannabinoids and opioids may interact at cellular (20) and intracellular signaling levels (21, 22). However, the postsynaptic molecular mechanisms that enable a signaling interaction between MOR and CB1 in NAc neurons are poorly understood.

The NAc/striatum expresses the highest density of adenosine A2a receptors (A2a) in the brain (23); these receptors are coupled primarily to the stimulatory G protein Gαolf (24). Recent evidence suggests that A2a can modify behavioral actions of ethanol, opiates, and cannabinoids (25–27). In addition, animals treated with A2a agonists or morphine show a similar physical dependence and exhibit bidirectional cross-withdrawal syndromes after blockade by receptor antagonists (28). Thus, A2a antagonists attenuate the development of morphine sensitization in mice (29) and inhibit morphine self-administration in rats (30). Similar findings have been reported in mice lacking A2a (31), suggesting that A2a might be responsible for attenuating heroin-related behaviors.

In this study, we present evidence for synergy between MOR and CB1 for cAMP/PKA signaling in primary NAc/striatal neurons and that synergy requires A2a. Importantly, an A2a antagonist administered either directly into the NAc or indirectly by i.p. injection eliminates reinstatement in rats trained to self-administer heroin. These findings suggest that A2a antagonists might also be useful in attenuating craving and relapse in human heroin addicts.

Results

CB1 Promotes PKA Activation in Primary NAc/Striatal Neurons.

We know that CB1 in NAc/striatal neurons is coupled to a pertussis toxin-sensitive Gαi/o protein that can either inhibit AC via Gαi at 30 min or stimulate cAMP production via βγ at 10 min (5). We then asked whether CB1 activation of PKA signaling induces CRE-mediated gene transcription using the reporter CRE-luciferase (32). A 10-min exposure of primary NAc/striatal neurons to the CB1 agonist methanandamide (Met; 200 nM) induces cAMP/PKA-dependent luciferase activity 5 h later (Table 1). A 30-min exposure to Met also increased luciferase activity 5 h later (data not shown). Taken together, these findings suggest that the stimulated cascade of PKA signaling persists after the increase in cAMP has dissipated. As expected, CB1-induced luciferase activation is blocked by AM 281 (10 μM), a CB1 antagonist (Table 1), as well as by the PKA inhibitor Rp-cAMPS (5). By contrast, activation of Gi/o-coupled adrenergic α2b or muscarinic M4 receptors by UK 14304 (10 μM) or carbachol (10 μM), respectively, does not activate PKA (data not shown) (5).

Table 1.

CB1 induces CRE-mediated luciferase activity in NAc/striatal neurons

Treatment % Increase over control
AGS3 AS 1 ± 4
AGS3 scramble 3 ± 3
Gαi3 AS 2 ± 3
Met 54 ± 5*
Met + AM281 5 ± 5
Met + AGS3 AS 2 ± 4
Met + AGS3 scramble 52 ± 2*
Met + Gαi3 AS 3 ± 2
Met + Gαi2 AS 53 ± 4*
Met + Gαi1 AS 56 ± 3*
Met + Gαo AS 52 ± 3*
Met + Gαolf AS 50 ± 6*
Open in a new tab

AGS3 scramble sequence was 5′-TACCGGCTACGACCGGACCGTCAAG-3′. Data are the mean ± SEM of at least three experiments.

*, P < 0.01 compared with control (one-way ANOVA and Dunnett’s test).

CB1-Induced PKA Signaling Is Mediated by Gαi3/AGS3/βγ.

Like MOR, CB1 may also couple to the specific Gαi subunit, Gαi3 (8, 33), whereas α2b and M4 couple to Gαi2 (34, 35). To test this possibility in primary NAc/striatal neurons, we investigated the specificity of CB1 interaction with Gαi3 subunits. We find that CB1 coimmunoprecipitates predominately with Gαi3 (Fig. 1). CB1 appears to interact preferentially with Gαi3 rather than other G proteins. Specific interaction of CB1 with Gαi3 is also suggested by results in a functional assay using antisense oligonucleotide directed against Gαi3 subunits in NAc/striatal neurons. Knockdown of Gαi3 specifically blocks Met-induced CRE-mediated luciferase activity (Table 1). In contrast, knockdown of other Gαi subunits such as Gαi1, Gαi2, Gαo, or Gαolf is without effect (Table 1). In additional control studies, antisense oligonucleotide directed against Gαolf, but not Gαi3, blocks Gαs/αolf-coupled prostaglandin E1 (PGE1)-induced CRE-luciferase activity (data not shown). Current concepts suggest that AGS3 competes with βγ for binding to a Gαi3-GDP complex (6, 8), thereby prolonging unbound βγ stimulation of AC. Free unbound βγ would be expected to reassociate with Gαi3 in the absence of AGS3, thereby reducing the availability of βγ for stimulation of AC II and IV (36, 37). We find that AGS3 knockdown prevents Met-induced CRE-mediated gene expression in primary NAc/striatal neurons (Table 1). Western blotting analysis confirmed that antisense oligonucleotide for Gα subunits or AGS3 reduces expression of each specific gene by >80% (8). Antisense oligonucleotide for a scrambled peptide had no effect. These findings suggest that the Gαi3/AGS3/βγ signaling pathway mediates CB1 activation of cAMP-dependent gene transcription in primary NAc/striatal neurons.

Fig. 1.

Fig. 1.

Open in a new tab

CB1 preferentially interacts with Gαi3 in NAc/striatal neurons. Coimmunoprecipitation of CB1 with Gαi3 subunits. CB1 was immunoprecipitated with specific polyclonal anti-CB1 antibodies and probed with anti-Gαo, Gαi1, Gαi2, or Gαi3 antibodies. The blot shown is representative of three separate experiments.

MOR and CB1 Exhibit Synergy for PKA Activation in NAc/Striatal Neurons.

We have reported that the MOR also activates the same Gαi3/AGS3/βγ signaling pathway as CB1 in primary NAc/striatal neurons (8). Thus, CB1 and MOR appear to share a common molecular signaling mechanism. Behavioral studies show that MOR and CB1 interact with each other to modulate opiate- or cannabinoid-seeking behavior (15–17, ‡‡). These studies suggested the possibility of synergy or additivity between MOR and CB1 for intracellular signaling. Therefore, we next asked whether subthreshold concentrations of MOR and CB1 agonists act together to promote PKA signaling. Our results show that subthreshold concentrations of [d-Ala2,N-MePhe4,Gly5-ol]enkephalin (DAMGO; 0.01 nM) or Met (0.02 nM) alone are without effect but together, activate CRE-mediated gene expression synergistically (Fig. 2A). Synergy of CRE-mediated gene expression occurs as a function of increasing Met or DAMGO concentrations when coincubated with a constant subthreshold concentration of DAMGO (0.01 nM) or Met (0.02 nM), respectively (Fig. 2B and C). Synergy is characterized by a 231-fold decrease in the EC50 for Met stimulation of gene expression and a 291-fold decrease in the EC50 for DAMGO stimulation. Synergy is blocked by the MOR antagonist d-Phe-Cys-Tyr-d-Trp-Orn-Thr-Pen-ThrNH2 (CTOP), the CB1 antagonist AM 281, Rp-cAMPS, pertussis toxin (PTX), the βγ inhibitor β-adrenergic receptor kinase 1 (βARK1), and by antisense oligonucleotide for Gαi3 and AGS3, respectively (Table 2). These results suggest that synergy between MOR and CB1 for cAMP/PKA signaling requires Gαi3 and AGS3 and is mediated by βγ stimulation of AC. Our previous studies demonstrated that synergy induced by Gαi-coupled receptors involved in addiction requires activation of A2a (5, 32). A2a is mainly coupled to Gαolf in the NAc/striatum (24). Therefore, we searched for evidence that synergy between subthreshold concentrations of MOR and CB1 agonists requires Gαolf. Antisense oligonucleotide for Gαolf blocks synergy between MOR and CB1 for CRE-mediated gene transcription (Table 2). Importantly, Gαolf knockdown has no effect on CB1 or MOR activation by saturating concentration of Met (Table 1) or DAMGO alone (8). Therefore, we predicted that synergy between MOR and CB1 requires A2a activation and Gαolf to prime AC for stimulation by βγ (36).

Fig. 2.

Fig. 2.

Open in a new tab

DAMGO and Met act synergistically to induce CRE-mediated gene expression in primary NAc/striatal neurons. (A) Cells were transfected with HSVCRE-Luc and treated with 0.01 nM DAMGO (DAM) or 0.02 nM Met alone or in combination for 10 min. Cells were then washed and cultured for 5 h before luciferase assay. (B) Cells were treated as above, with various concentrations of Met plus 0.01 nM DAMGO for 10 min. Luciferase was assayed as above. (C) Cells were treated with various concentrations of DAMGO plus 0.02 nM Met. The EC50 value was calculated by using GraphPad prism software. ∗, P < 0.01 compared with control (one-way ANOVA and Dunnett’s test).

Table 2.

Synergistic increase of CRE-mediated luciferase activity in NAc/striatal neurons

Treatment % Increase over control
DAM (0.01 nM) 2 ± 3
Met (0.02 nM) 3 ± 1
DAM + Met 38 ± 3*
DAM + Met + CTOP 5 ± 2
DAM + Met + AM281 3 ± 5
DAM + Met + Rp-cAMPS 1 ± 4
DAM + Met + PTX 2 ± 2
DAM + Met + βARK1 3 ± 2
DAM + Met + Gαi3 AS 5 ± 3
DAM + Met + AGS3 AS 4 ± 5
DAM + Met + Gαolf AS 9 ± 6
PGE1 (10 μM) 57 ± 8*
Open in a new tab

Data are the mean ± SEM of at least three experiments.

*, P < 0.01 compared with control as in Table 1.

A2a Regulates Synergy in NAc/Striatal Neurons.

Neural cells in culture continually release adenosine into the extracellular medium, activating adenosine receptors to promote synergy of CRE-gene expression by subthreshold concentrations of DAMGO and Met. If A2a is involved, then elimination of adenosine or blockade of A2a should prevent synergy. As predicted, adenosine deaminase, which degrades adenosine or the A2a antagonists 3,7-dimethyl-1-(2-propynyl)xanthine (DMPX) or 3,7-dihydro-8-[(1E)-2-(3-methoxyphenyl)ethenyl]-7-methyl-3-(3-phosphonooxy)propyl-1-(2-propynyl)-1H-purine-2,6-dione disodium salt (MSX-3), each prevents a synergy-induced increase of CRE-mediated gene expression (Fig. 3). Adenosine removal or A2a blockade also prevents a synergy-induced increase in cAMP (data not shown). The A1 antagonist 8-cyclopentyl-1,3-dipropylxanthine (DPCPX) is without effect. These findings suggest that adenosine and A2a are required for synergy. Because of the requirement for adenosine and A2a in our in vitro studies, we asked whether A2a regulates heroin-seeking behavior.

Fig. 3.

Fig. 3.

Open in a new tab

Synergy requires adenosine A2a activation. Cells were cultured, treated, and assayed for luciferase activity as in Fig. 2A. Where indicated, cells were preincubated with 1 unit/ml adenosine deaminase (ADA), DMPX (10 μM), MSX-3 (100 nM), or DPCPX (100 nM) for 30 min. Data are the mean ± SE of three experiments. ∗, P < 0.01 compared with control (one-way ANOVA and Dunnett’s test).

A2a Antagonists Eliminate Reinstatement of Heroin Self-Administration in Addicted Rats.

The NAc is implicated in drug-seeking behavior (38). To determine the role of A2a in reinstatement of heroin self-administration, the A2a antagonist DMPX (5 nmol) was directly administered into the NAc of rats. We found that DMPX eliminates heroin-induced reinstatement of lever-pressing for heroin; the A1 antagonist DPCPX (5 nmol) administered into the NAc is without effect (Fig. 4A). These results suggested that A2a in the NAc is required for heroin-seeking behavior. We next asked whether systemic administration of a highly specific A2a antagonist, MSX-3, could achieve the same result. MSX-3 given by i.p. injection completely eliminates heroin-induced reinstatement in seven of eight heroin addicted rats (Fig. 4B).

Fig. 4.

Fig. 4.

Open in a new tab

A2a antagonists eliminate heroin-seeking behavior in rats. (A) DMPX in NAc inhibition of heroin reinstatement. DMPX (5 nmol) or DPCPX (5 nmol) was infused into the NAc. (B) MSX-3 (3 mg/kg) was injected i.p. DMPX (pretreatment time, 25 min) in NAc and MSX-3 i.p. each prevented reinstatement of lever-pressing elicited by a priming injection of heroin (0.25 mg/kg, s.c.). Inactive lever-press responses were not different across drug treatment conditions. Data represent mean active lever presses ± SEM (n = 8 in all groups). ∗, P < 0.01 compared with extinction responding (two-way ANOVA and Tukey’s posttest).

Discussion

The major findings in this article are that (i) MOR and CB1 in NAc/striatal neurons act synergistically through a common molecular mechanism involving Gαi3, AGS3 and βγ to activate cAMP/PKA signaling; (ii) A2a appears to regulate synergy between MOR and CB1 because A2a is required for their synergistic activation of cAMP/PKA signaling; and (iii) in rats withdrawn from self-administering heroin, an A2a antagonist administered directly into the NAc or indirectly via i.p. injection completely prevents heroin-induced reinstatement of heroin self-administration.

We have previously shown in NAc/striatal neurons that MOR preferentially associates with Gαi3; this association appears to provide specificity for MOR-induced PKA activation (8). Our data suggest that CB1 also coimmunoprecipitates predominately with Gαi3. Specific interaction with Gαi3 was also suggested by a functional assay using antisense oligonucleotide directed against Gαi subunits in NAc/striatal neurons. Knockdown of Gαi3 specifically blocks Met-induced CRE-mediated luciferase activity. In contrast, knockdown of other Gαi subunits such as Gαi1, Gαi2, Gαo, or Gαolf is without effect. It seems likely, therefore, that MOR and CB1 preferentially couple to Gαi3 for PKA activation. The specificity of G protein signaling for CB1 is also regulated by AGS3. Knockdown of AGS3 prevents CB1-induced PKA Cα translocation (data not shown) and CRE-mediated gene expression. Our data are consistent with current concepts that AGS3 competes with βγ for binding to Gαi3, prolonging the action of free βγ. In the absence of AGS3, the action of βγ is attenuated because unbound βγ appears to reassociate quickly with Gαi.

Our studies suggest that Gαi3 and therefore, AGS3, appear to associate preferentially with MOR and CB1. This association suggested that in NAc/striatal neurons, MOR and CB1 could act together to promote cAMP/PKA signaling through a common molecular signaling pathway, Gαi3/AGS3/βγ. In this article, we demonstrated that a subthreshold concentration of DAMGO or Met, which is without effect when added separately, together act in synergy to promote CRE gene transcription. Results from other laboratories are consistent with the possibility that synergy mediated by Gαi3/AGS3/βγ may be a molecular mechanism by which MOR and CB1 interact with each other at cellular (20) and intracellular signaling levels (21, 22) to modulate opiate- and cannabinoid-seeking behavior (15–18, ‡‡).

Other investigators have described an opiate-induced increase of cAMP/PKA signaling, which is exaggerated upon opiate withdrawal (39, 40). In vivo, increased cAMP/PKA activity may be associated with opiate-induced tolerance and dependence and is thought to contribute to the reinforcing properties of opiates (41). In addition, cannabinoid-induced increases in cAMP and PKA activities in the cortex and/or cerebellum during treatment (42) and withdrawal (43) have also been reported. Our findings suggest that a transient increase of cAMP at 10 min triggers a cascade of events in cAMP/PKA signaling, including PKA translocation and phosphorylation of cAMP response element binding protein (CREB), leading to CRE-mediated gene transcription hours later. These findings suggest the possibility that activation of CRE gene expression due to a transient increase in cAMP might persist during continued drug exposure despite subsequent inhibition of cAMP production. Thus, a transient increase in cAMP followed by increased CRE-mediated gene transcription may contribute to the acquisition and/or maintenance of drug-seeking behavior.

In this study, we also demonstrate in primary NAc/striatal neurons that synergy between MOR and CB1 for cAMP/PKA signaling is mediated by βγ dimers and requires adenosine activation of A2a. Previously, we used a transformed glioma-neuroblastoma cell line (NG108-15/D2) and demonstrated similar findings, but in that study, synergy involved dopamine D2 receptor and ethanol activation of A2a (32). We have also found synergy involving dopamine D2 receptor and CB1 or DOR but not other Gi/o-coupled receptors (5). Thus far, all instances of synergy between receptors involved in addictive behaviors appear to require adenosine activation of A2a. In the presence of endogenous adenosinergic tone, it is possible that exogenous opiates might induce synergy with endogenous cannabinoids in vivo. The possibility of A2a-dependent synergy suggested that A2a antagonists might attenuate heroin-seeking behavior. We show here that A2a antagonists, whether administered directly into the NAc or indirectly via systemic i.p. injection, completely eliminate reinstatement of heroin self-administration in addicted rats. Moreover, these results are consistent with our studies of ethanol drinking in a different experimental setting: systemic administration of an A2a antagonist strikingly reduces operational ethanol self-administration (25). Our in vitro synergy studies led us to test the possibility that A2a antagonists might attenuate addictive behaviors. However, the effectiveness of an A2a antagonist in eliminating reinstatement of heroin-seeking behavior does not prove that synergy between MOR and CB1 is responsible for heroin reinstatement. Nevertheless, our findings are consistent with well documented evidence for cross-talk between MOR- and CB1-dependent behaviors (17, 18, ‡‡).

The highest density of A2a in the brain is expressed postsynaptically on medium spiny neurons in the NAc/striatum (44). MOR and CB1 are also expressed postsynaptically on the same medium spiny neurons (19). The unique coexpression of A2a with MOR and CB1 on the same NAc/striatum neurons suggests that A2a might interact with and affect MOR- and CB1-addictive behavior. Thus, A2a blockade significantly reduces morphine and cannabinoid reward and the symptoms of withdrawal (27, 31). But, we cannot rule out the possibility that the reinforcing action of opiates might also involve a dopamine-dependent mechanism in ventral tegmental area (VTA) affecting downstream NAc function. Thus, injection of opioids into VTA stimulates locomotor activity and produces CPP in rats (45, 46). Also, MOR and DOR agonists are self-administered into the VTA. Despite these findings, more recent observations show that the NAc core is indeed required for opiate-addictive behaviors. Inactivation of the NAc core inhibits heroin self-administration (4) as well as heroin-induced reinstatement of drug-seeking behavior (our unpublished observation).

In summary, we show that CB1 mimics MOR-induced cAMP/PKA signaling through Gαi3/AGS3/βγ and that synergy between subthreshold concentrations of CB1 and MOR agonists significantly hypersensitizes CB1 or MOR signaling. At a cellular level, synergy between CB1 and MOR for cAMP/PKA signaling requires adenosine A2a, suggesting that A2a blockade might attenuate heroin-addictive behavior. Indeed, an A2a antagonist administered directly into the NAc or indirectly via i.p. injection completely eliminates reinstatement in heroin-addicted rats, a valid model of human craving and relapse. Therefore, it should be possible to design and develop highly specific A2a antagonists targeted to NAc/striatal neurons as potential therapeutic agents to help prevent craving and relapse in human heroin addicts.

Materials and Methods

Reagents and Primary NAc/Striatal Culture.

Reagents were from Sigma except where indicated. Rp-cAMPS was from BioLog (Hatward, CA), p-CRE-luciferase was from Stratagene, and pCMV-β-gal was from Promega. Primary neurons were prepared from the striatum of newborn rat pups, cultured exactly as described in ref. 8, and used on day 10.

Immunoprecipitation, Western Blot Analysis, and CRE-Luciferase Assay.

Solubilized membranes, prepared from cells stimulated with 200 nM Met, were immunoprecipitated with anti-CB1 antibodies (Santa Cruz Biotechnology) and Western blotted with antibodies for rat Gα0, Gαi1, Gαi2, and Gαi3 (Santa Cruz Biotechnology), respectively (8). To assay CRE-luciferase, rat NAc/striatal neurons were plated at 2.5 × 104 cells per 24-well plate and grown for 10 days. Cell transfection, drug treatment, and luciferase assay were carried out as described in ref. 5.

Antisense and Viral Vectors.

Herpes simplex virus (HSV) vectors expressing antisense RNA for AGS3 and Gαi were created by cloning each antisense oligonucleotide, designed as described in ref. 8, into an HSVLacZ vector under control of a mouse U6 polymerase III promoter. HSV at 1 multiplicity of infection was used to transfect primary neurons.

Animal Surgery and Behavioral Assay.

Sprague–Dawley male rats were anesthetized, implanted with jugular catheters and bilateral intracranial guide cannulas, and trained as described (8, 47, §§). After animals met maintenance criterion (active lever presses do not vary by >10% across 3 consecutive days), extinction and reinstatement were conducted as described (8, 47, §§). Reinstatement testing was conducted after responding fell to 15% of maintenance levels. Each rat received a priming injection of heroin (0.25 mg/kg, s.c.) before placement in the self-administration chamber for a 3-h extinction session. After behavioral testing, rats were anesthetized, and the brain was fixed and sectioned as described to verify infusion sites (47).

Acknowledgments

We thank Dr. R. J. Lefkowitz (Howard Hughes Medical Institute, Duke University Medical Center, Durham, NC) for the βARK1 minigene, Dr. R. L. Neve (Harvard Medical School, Boston) for the HSVPrpUC vector, and John Shryock and Adrienne Gordon for critical reading of the manuscript. This research was supported by a National Institutes of Health grant (to I.D. and L.Y.), funds provided by the State of California for medical research on ethanol and substance abuse through the University of California (San Francisco), and a grant from the Department of the Army (to I.D. and L.Y.). The U.S. Army Medical Research Acquisition Activity (Fort Detrick, MD) is the awarding and administering acquisition office.

Abbreviations

DOR

δ-opioid receptor

MOR

μ-opioid receptor

CPP

conditioned place preference

NAc

nucleus accumbens

CRE

cAMP response element

AC

adenylyl cyclase

AGS3

activator of G protein signaling 3

CB1

cannabinoid CB1 receptor

A2a

adenosine A2a receptor(s)

Met

methanandamide

DAMGO

[d-Ala2,N-MePhe4,Gly5-ol]enkephalin

DMPX

3,7-dimethyl-1-(2-propynyl)xanthine

DPCPX

8-cyclopentyl-1,3-dipropylxanthine

MSX-3

3,7-dihydro-8-[(1E)-2-(3-methoxyphenyl)ethenyl]-7-methyl-3-(3-phosphonooxy)propyl-1-(2-propynyl)-1H-purine-2,6-dione disodium salt.

Footnotes

Conflict of interest statement: A patent application concerning this work is under review and will be owned by the University of California at San Francisco. I.D. is Vice President and L.Y. is a Senior Scientist at CV Therapeutics, Inc.

‡‡

Goldberg, S. R., Munzar, P., Justinava, Z. & Tanda, G., 2001 Symposium on Cannabinoids, June 28–30, 2001, Burlington, VT, p. 102 (abstr.).

§§

McFarland, K., 34th Annual Meeting of the Society for Neuroscience, Oct. 23–27, 2004, San Diego, CA, abstr. 119.2.

References

  • 1.Pan Z. Z. Trends Pharmacol. Sci. 1998;19:94–98. doi: 10.1016/s0165-6147(98)01169-9. [DOI] [PubMed] [Google Scholar]
  • 2.Negus S. S., Burke T. F., Medzihradsky F., Woods J. H. J. Pharmacol. Exp. Ther. 1993;267:896–903. [PubMed] [Google Scholar]
  • 3.Matthes H. W., Maldonado R., Simonin F., Valverde O., Slowe S., Kitchen I., Befort K., Dierich A., Le Meur M., Dolle P., et al. Nature. 1996;383:819–823. doi: 10.1038/383819a0. [DOI] [PubMed] [Google Scholar]
  • 4.Alderson H. L., Parkinson J. A., Robbins T. W., Everitt B. J. Psychopharmacology. 2001;153:455–463. doi: 10.1007/s002130000634. [DOI] [PubMed] [Google Scholar]
  • 5.Yao L., Fan P., Jiang Z., Mailliard W. S., Gordon A. S., Diamond I. Proc. Natl. Acad. Sci. USA. 2003;100:14379–14384. doi: 10.1073/pnas.2336093100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Lanier S. M. Biol. Cell. 2004;96:369–372. doi: 10.1016/j.biolcel.2004.03.012. [DOI] [PubMed] [Google Scholar]
  • 7.Sanada K., Tsai L. H. Cell. 2005;122:119–131. doi: 10.1016/j.cell.2005.05.009. [DOI] [PubMed] [Google Scholar]
  • 8.Yao L., McFarland K., Fan P., Jiang Z., Inoue Y., Diamond I. Proc. Natl. Acad. Sci. USA. 2005;102:8746–8751. doi: 10.1073/pnas.0503419102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Ghozland S., Matthes H. W., Simonin F., Filliol D., Kieffer B. L., Maldonado R. J. Neurosci. 2002;22:1146–1154. doi: 10.1523/JNEUROSCI.22-03-01146.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Lepore M., Vorel S. R., Lowinson J., Gardner E. L. Life Sci. 1995;56:2073–2080. doi: 10.1016/0024-3205(95)00191-8. [DOI] [PubMed] [Google Scholar]
  • 11.Tanda G., Munzar P., Goldberg S. R. Nat. Neurosci. 2000;3:1073–1074. doi: 10.1038/80577. [DOI] [PubMed] [Google Scholar]
  • 12.Maldonado R., Rodriguez de Fonseca F. J. Neurosci. 2002;22:3326–3331. doi: 10.1523/JNEUROSCI.22-09-03326.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Fattore L., Martellotta M. C., Cossu G., Mascia M. S., Fratta W. Behav. Brain Res. 1999;104:141–146. doi: 10.1016/s0166-4328(99)00059-5. [DOI] [PubMed] [Google Scholar]
  • 14.Howlett A. C. Annu. Rev. Pharmacol. Toxicol. 1995;35:607–634. doi: 10.1146/annurev.pa.35.040195.003135. [DOI] [PubMed] [Google Scholar]
  • 15.Ledent C., Valverde O., Cossu G., Petitet F., Aubert J. F., Beslot F., Bohme G. A., Imperato A., Pedrazzini T., Roques B. P., et al. Science. 1999;283:401–404. doi: 10.1126/science.283.5400.401. [DOI] [PubMed] [Google Scholar]
  • 16.Martin M., Ledent C., Parmentier M., Maldonado R., Valverde O. Eur. J. Neurosci. 2000;12:4038–4046. doi: 10.1046/j.1460-9568.2000.00287.x. [DOI] [PubMed] [Google Scholar]
  • 17.Spano M. S., Fattore L., Cossu G., Deiana S., Fadda P., Fratta W. Br. J. Pharmacol. 2004;143:343–350. doi: 10.1038/sj.bjp.0705932. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.De Vries T. J., Homberg J. R., Binnekade R., Raaso H., Schoffelmeer A. N. Psychopharmacology. 2003;168:164–169. doi: 10.1007/s00213-003-1422-1. [DOI] [PubMed] [Google Scholar]
  • 19.Salio C., Fischer J., Franzoni M. F., Mackie K., Kaneko T., Conrath M. NeuroReport. 2001;12:3689–3692. doi: 10.1097/00001756-200112040-00017. [DOI] [PubMed] [Google Scholar]
  • 20.Shapira M., Gafni M., Sarne Y. Brain Res. 2003;960:190–200. doi: 10.1016/s0006-8993(02)03842-8. [DOI] [PubMed] [Google Scholar]
  • 21.Thorat S. N., Bhargava H. N. Eur. J. Pharmacol. 1994;260:5–13. doi: 10.1016/0014-2999(94)90003-5. [DOI] [PubMed] [Google Scholar]
  • 22.Manzanares J., Corchero J., Romero J., Fernandez-Ruiz J. J., Ramos J. A., Fuentes J. A. Trends Pharmacol. Sci. 1999;20:287–294. doi: 10.1016/s0165-6147(99)01339-5. [DOI] [PubMed] [Google Scholar]
  • 23.Svenningsson P., Le Moine C., Fisone G., Fredholm B. B. Prog. Neurobiol. 1999;59:355–396. doi: 10.1016/s0301-0082(99)00011-8. [DOI] [PubMed] [Google Scholar]
  • 24.Kull B., Svenningsson P., Fredholm B. B. Mol. Pharmacol. 2000;58:771–777. doi: 10.1124/mol.58.4.771. [DOI] [PubMed] [Google Scholar]
  • 25.Arolfo M. P., Yao L., Gordon A. S., Diamond I., Janak P. H. Alcohol. Clin. Exp. Res. 2004;28:1308–1316. doi: 10.1097/01.alc.0000139821.38167.20. [DOI] [PubMed] [Google Scholar]
  • 26.Halimi G., Devaux C., Clot-Faybesse O., Sampol J., Legof L., Rochat H., Guieu R. Eur. J. Pharmacol. 2000;398:217–224. doi: 10.1016/s0014-2999(00)00275-2. [DOI] [PubMed] [Google Scholar]
  • 27.Soria G., Castane A., Berrendero F., Ledent C., Parmentier M., Maldonado R., Valverde O. Eur. J. Neurosci. 2004;20:2203–2213. doi: 10.1111/j.1460-9568.2004.03682.x. [DOI] [PubMed] [Google Scholar]
  • 28.Coupar I. M., Tran B. L. T. Life Sci. 2001;69:779–790. doi: 10.1016/s0024-3205(01)01155-9. [DOI] [PubMed] [Google Scholar]
  • 29.Weisberg S. P., Kaplan G. B. Neurosci. Lett. 1999;264:89–92. doi: 10.1016/s0304-3940(99)00188-3. [DOI] [PubMed] [Google Scholar]
  • 30.Sahraei H., Motamedi F., Khoshbaten A., Zarrindast M. R. Eur. J. Pharmacol. 1999;383:107–113. doi: 10.1016/s0014-2999(99)00613-5. [DOI] [PubMed] [Google Scholar]
  • 31.Berrendero F., Castane A., Ledent C., Parmentier M., Maldonado R., Valverde O. Eur. J. Neurosci. 2003;17:315–324. doi: 10.1046/j.1460-9568.2003.02439.x. [DOI] [PubMed] [Google Scholar]
  • 32.Yao L., Arolfo M. P., Dohrman D. P., Jiang Z., Fan P., Fuchs S., Janak P. H., Gordon A. S., Diamond I. Cell. 2002;109:733–743. doi: 10.1016/s0092-8674(02)00763-8. [DOI] [PubMed] [Google Scholar]
  • 33.Mukhopadhyay S., Howlett A. C. Eur. J. Biochem. 2001;268:499–505. doi: 10.1046/j.1432-1327.2001.01810.x. [DOI] [PubMed] [Google Scholar]
  • 34.McClue S. J., Milligan G. FEBS Lett. 1990;269:430–434. doi: 10.1016/0014-5793(90)81209-7. [DOI] [PubMed] [Google Scholar]
  • 35.Migeon J. C., Nathanson N. M. J. Biol. Chem. 1994;269:9767–9773. [PubMed] [Google Scholar]
  • 36.Tang W. J., Gilman A. G. Science. 1991;254:1500–1503. doi: 10.1126/science.1962211. [DOI] [PubMed] [Google Scholar]
  • 37.Koch W. J., Hawes B. E., Inglese J., Luttrell L. M., Lefkowitz R. J. J. Biol. Chem. 1994;269:6193–6197. [PubMed] [Google Scholar]
  • 38.Koob G. F. Ann. N.Y. Acad Sci. 1992;654:171–191. doi: 10.1111/j.1749-6632.1992.tb25966.x. [DOI] [PubMed] [Google Scholar]
  • 39.Nestler E. J., Aghajanian G. K. Science. 1997;278:58–63. doi: 10.1126/science.278.5335.58. [DOI] [PubMed] [Google Scholar]
  • 40.Avidor-Reiss T., Nevo I., Levy R., Pfeuffer T., Vogel Z. J. Biol. Chem. 1996;271:21309–21315. doi: 10.1074/jbc.271.35.21309. [DOI] [PubMed] [Google Scholar]
  • 41.Nestler E. J., Malenka R. C. Sci. Am. 2004;290:78–85. doi: 10.1038/scientificamerican0304-78. [DOI] [PubMed] [Google Scholar]
  • 42.Rubino T., Vigano D., Massi P., Spinello M., Zagato E., Giagnoni G., Parolaro D. Neuropharmacology. 2000;39:1331–1336. doi: 10.1016/s0028-3908(99)00196-3. [DOI] [PubMed] [Google Scholar]
  • 43.Tzavara E. T., Valjent E., Firmo C., Mas M., Beslot F., Defer N., Roques B. P., Hanoune J., Maldonado R. Eur. J. Neurosci. 2000;12:1038–1046. doi: 10.1046/j.1460-9568.2000.00971.x. [DOI] [PubMed] [Google Scholar]
  • 44.Svenningsson P., Le Moine C., Kull B., Sunahara R., Bloch B., Fredholm B. B. Neuroscience. 1997;80:1171–1185. doi: 10.1016/s0306-4522(97)00180-2. [DOI] [PubMed] [Google Scholar]
  • 45.Wise R. A. Neurosci. Biobehav. Rev. 1989;13:129–133. doi: 10.1016/s0149-7634(89)80021-1. [DOI] [PubMed] [Google Scholar]
  • 46.van der Kooy D., Mucha R. F., O’Shaughnessy M., Bucenieks P. Brain Res. 1982;243:107–117. doi: 10.1016/0006-8993(82)91124-6. [DOI] [PubMed] [Google Scholar]
  • 47.McFarland K., Kalivas P. W. J. Neurosci. 2001;21:8655–8663. doi: 10.1523/JNEUROSCI.21-21-08655.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES