Opioid Receptor Heteromer in the Nucleus Accumbens

Home , Opioidergic

ORIGINAL ARTICLE Antidepressant-like and anxiolytic-like effects following activation of the m-d opioid receptor heteromer in the nucleus accumbens

N Kabli1,2, T Nguyen1, G Balboni3, BF O’Dowd1,2 and SR George1,2,4

Treatment-resistant major depressive disorder remains inadequately treated with currently available antidepressants. Opioid receptors (ORs) are involved in the pathophysiology of depression yet remain an untapped therapeutic intervention. The m-d OR heteromer represents a unique signaling complex with distinct properties compared with m- and d-OR homomers; however, its role in depression has not been characterized. As there are no ligands exclusively targeting the m-d heteromer, we devised a strategy to selectively antagonize the function of the m-dOR complex using a specific interfering peptide derived from the dOR distal carboxyl tail, a sequence implicated in m-dOR heteromerization. In vitro studies using a minigene expressing this peptide demonstrated a loss of the unique pharmacological and trafficking properties of d-agonists at the m-d heteromer, with no effect on m-ord-OR homomers, and a dissociation of the m-dOR complex. Intra-accumbens administration of the TAT-conjugated interfering peptide abolished the antidepressant-like and anxiolytic-like actions of the d-agonist UFP-512 (H-Dmt-Tic-NH-CH(CH2-COOH)-Bid) measured in the forced swim test, novelty-induced hypophagia and elevated plus maze paradigms in rats. UFP-512’s antidepressant-like and anxiolytic-like actions were abolished by pretreatment with either mOR or dOR antagonists. Overall, these findings demonstrate that the m-d heteromer may be a potential suitable therapeutic target for treatment-resistant depression and anxiety disorders.

Molecular Psychiatry (2014) 19, 986–994; doi:10.1038/mp.2013.115; published online 24 September 2013 Keywords: anhedonia; emotion; G-protein coupled receptor; heterooligomer; mood; ventral striatum

INTRODUCTION potent antidepressant-like and anxiolytic-like effects in animal 17–25 Treatment-resistant major depression remains a significant health, models. Further, the analgesic effects of d-agonists are social and economic burden that is inadequately treated with attenuated or abolished in mOR-gene-deleted animals and 26,27 currently available antidepressant medications that modulate following pretreatment with mOR antagonists, supporting the serotonergic and noradrenergic neurotransmission.1 Thus, pharma- notion that these ligands may in fact recruit the m-d OR heteromer cological agents with alternative mechanisms of action are critically for certain actions. Second, animals in which the genes for the mOR, needed and necessary to address this unmet therapeutic need. dOR or Gaz (the inhibitory G-protein to which the m-d heteromer Opioid receptors (ORs) and their endogenous ligands are couples selectively) are deleted display depressive-like and involved in the pathophysiology of major depression, yet remain anxiogenic behavior.28 Finally, recent studies localize the m-d an untapped potential therapeutic intervention.2–4 However, heteromer to brain nuclei implicated in mood and emotional although m-opioid agonists such as morphine have been used in processing, including the nucleus accumbens (NAc),29 a subcortical the clinic to successfully manage refractory depression, their brain region whose activity is reduced in patients with major clinical utility is limited by the development of tolerance and depressive disorders.30,31 Furthermore, direct activation of the NAc adverse side effects ensuing from mOR activation.5–11 m-and with deep brain stimulation has shown promise in ameliorating d-ORs have been shown to form a heteromeric receptor complex depression and anxiety in patients.32,33 Thus, overall, several lines possessing unique pharmacological and signalling properties.12–15 of evidence support exploring the activation of the m-d heteromer The m-d heteromer displays less of the functional desensitization in the NAc in models of depression and anxiety. and downregulation that limit the therapeutic utility of drugs In this study, we investigated the function of the m-d heteromer targeting the mOR,12,14 rendering the m-d receptor complex a and its role in regulating depressive-like and anxiogenic behavior promising therapeutic target. by determining its involvement in the mood-elevating properties Although the pharmacological and molecular profile of the m-d of d-agonists. In the absence of selective ligands exclusively heteromer in vitro is being elucidated, its physiological role in vivo targeting the m-d heteromer, we devised a novel strategy to is largely unknown. Exploring the m-d heteromer as a therapeutic selectively antagonize the function of the m-d receptor complex target for treatment-resistant major depression is supported by through physical disruption. To this end, we designed a specific behavioral, genetic and biochemical findings. First, d-opioid interfering peptide aimed at the distal carboxyl tail domain of the agonists, which we and others have shown to bind to and activate dOR, which has been implicated as a site of interaction between the m-d heteromer and regulate its intracellular trafficking,12,16 have m- and d-ORs. After validating the use of the peptide to dissociate

1Campbell Family Mental Health Research Institute, Center for Addiction and Mental Health, Toronto, ON, Canada; 2Departments of Pharmacology and Toxicology, Faculty of Medicine, University of Toronto, Toronto, ON, Canada; 3Department of Life and Environmental Sciences, University of Cagliari, Cagliari, Italy and 4Department of Medicine, Faculty of Medicine, University of Toronto, Toronto, ON, Canada. Correspondence: Dr SR George, Department of Pharmacology and Toxicology, Faculty of Medicine, University of Toronto, Medical Sciences Building Rm 4358, 1 King’s College Circle, Toronto, Ontario, M5S 1A8, Canada. E-mail: [email protected] Received 11 March 2013; revised 16 July 2013; accepted 18 July 2013; published online 24 September 2013 m-d heteromer mediates antidepressant-like action N Kabli et al 987 the m-d heteromer in vitro, we administered it in vivo into the NAc Intact cell internalization assay as a TAT-conjugate in order to interrogate the role of the m-d mOR binding was performed on intact cells stably expressing ORs using the 3 heteromer in the antidepressant-like and anxiolytic-like effects of hydrophilic selective m-agonist [ H]-DAMGO at its KD (2 nM) as described 12 UFP-512 (H-Dmt-Tic-NH-CH(CH2-COOH)-Bid), a d-agonist with high previously. Cells were treated with 1 mM agonist or saline vehicle for 1 h at subnanomolar affinity at the m-d heteromer.12 Overall, our findings 37 1C. In separate experiments, cells were pretreated with either 100 nM of the d-antagonist naltrindole or the m-antagonist CTOP. The disappearance indicate that activation of the m-d heteromer in the NAc by UFP- 3 512 produced antidepressant-like and anxiolytic-like actions, of [ H]-DAMGO binding sites on the cell surface served as an indicator of mOR internalization. which were completely abolished by dissociating the m-d heteromer or antagonizing either one of its component receptors. Animals Adult male Sprague Dawley rats weighing 250–350 g (Charles River, MATERIALS AND METHODS St Constant, Que´bec, Canada) were housed in pairs (and singly-housed following surgery) in a temperature-controlled room with corn chip Drugs bedding and free access to rodent chow and water. Rats were maintained 2 4 Naltrindole hydrochloride, deltorphin II, DAMGO ([D-Ala , N-MePhe , under a standard 12-h/12-h light/dark cycle. Testing was performed during 2 5 Gly-ol]-enkephalin), DPDPE ([D-Pen , D-Pen ]enkephalin), imipramine the light cycle. Animal protocols were approved by the University of and fluoxetine were purchased from Sigma (Saint Louis, MO, USA). CTOP Toronto Animal Care Committee and were in accordance with the (D-Phe-Cys-Tyr-D-Trp-Orn-Thr-Pen-Thr-NH2) was purchased from Tocris guidelines set by the Canadian Council on Animal Care. (Ellisville, MO, USA). UFP-512 was synthesized by Balboni et al.34 Stereotaxic surgery Expression in mammalian cells Rats were mounted on a stereotaxic frame under isoflurane anesthesia. Human embryonic kidney 293T cells stably expressing m- and/or d-ORs Stainless steel guide cannulae were implanted bilaterally into the NAc (AP (density of 150–175 fmol mg À 1 per receptor and 150–350 fmol mg À 1 þ 1.6, ML þ 3.0, DV À 6.9) at a 101 angle from the midline or into the when both receptors were co-expressed) were generated as described lateral cerebral ventricle (AP À 0.8, ML þ 1.4, DV À 3.6) at 901 angle to the previously.12 skull using co-ordinates from.35 Guide cannulae were positioned 1 mm anterior to the NAc to prevent tissue damage at the target site. Cannulae were secured onto the skull using stainless steel screws and dental cement Minigene constructs with the insertion of dummy stylets to maintain patency. Rats were Complementary DNA fragments expressing either the distal 16 or 22 allowed to recover, and behavioral testing was conducted 7 days following amino acids of the dOR carboxyl tail were subcloned into pcDNA3.1. The surgery. Placement of guide cannulae was verified by brain sectioning and Transformer site-directed mutagenesis kit (Clontech, Mountain View, CA, histological examination at the end of the experiment. USA) was used to insert the hemaglutinin epitope (YPYDVPDYA) at the NH2-terminal of the dOR carboxyl tail fragment using the same approach TAT fusion peptides previously described.13,14 Cells were transfected using LipofectAMINE (Invitrogen, Burlington, ON, Canada) according to the manufacturer’s A peptide derived from the distal portion of the dOR carboxyl tail 0 0 protocols. (5 -VTACTPSDGPGGGAAA-3 ) identified as a m-d heteromerization interaction interface was conjugated to the HIV protein transduction domain TAT (50-YGRKKRRQRRR-30) to create a TAT fusion peptide for transduction into Cell membrane preparation neurons.36 A scrambled sequence of the same 16 amino acids was used as Membrane fractions (crude P2 fraction) and cell lysates were prepared control. Dansyl chloride was conjugated to the TAT peptide to allow for from cells as described previously.12,14 easy detection of the fusion peptide in vivo. Interfering or scrambled fusion peptides (15 pmol per side) were injected bilaterally into the NAc at concentrations previously reported in the literature.37–39 Competition radioligand binding Experiments were performed in duplicate on cell membrane preparations Forced swim test À 12 À 4 with increasing concentrations of competing ligand (10 –10 M) This test was adapted from Porsolt et al.40 Rats were placed in a cylindrical as described in detail previously.12 [3H]-DAMGO (36.8 Ci mmol À 1) and 3 À 1 container filled to a depth of 30 cm with water (at 25 þ 1 1C). During the [ H]-diprenorphine (50.0 Ci mmol ) were purchased from Perkin Elmer initial session, rats were allowed to swim without behavioral scoring for Life Sciences (Waltham, MA, USA). 15 min in order to induce immobility on test day. On the next day, animals were tested and their predominant behavior was recorded at 5-s intervals 41 Dot blotting analysis for 5 min using the time sampling method validated by Lucki et al. Rats which floated in the water making only the necessary movements to Cell lysate supernatant was spotted onto nitrocellulose membrane and maintain their head above water were deemed immobile. Following swim then incubated with anti-hemaglutinin antibody diluted 1:2000 (AB 3254, sessions, the rats were dried and placed under a heat lamp and monitored Chemicon, Temecula, CA, USA). Immunoreactivity was revealed by until recovery. incubating with HRP-conjugated anti-chicken secondary antibody diluted Dose-response curves of UFP-512’s behavioral effects following intraNAc 1:1000 (Bio-Rad, Hercules, CA, USA) followed by enhanced chemilumines- and intracerebroventricular (i.c.v.) administration were generated. In all cent reagents (KPL, Gaithersburg, MD, USA), and then exposing to film. subsequent experiments, the 1.5 mg and 80 nmol doses of UFP-512 were microinjected into the NAc and i.c.v., respectively. Naltrindole and CTOP were administered at 1 mg intraNAc and 10 nmol i.c.v. Agonist and Immunoprecipitation studies antagonist doses used were derived from literature examining the effects Membrane fractions were solubilized, incubated with anti-cMyc antibody of microinjection of opioid ligands into the NAc or i.c.v., and behavioral (Millipore, Billerica, MA, USA) and then protein-G-agarose beads, followed testing was conducted 30 min following the last injection.17,23,42–49 by washing and solubilization in SDS sample for gel electrophoresis as described previously.14 Elevated plus maze test The elevated plus maze apparatus (Colbourn Instruments, Whitehall, PA, SDS-polyacrylamide gel electrophoresis and immunoblotting USA) consisted of a black Plexiglas maze (55 cm above ground) with four Membrane proteins were resolved on a 10% Tris-Glycine precast gel arms (45 Â 45 cm) extending out in a cross pattern from a central zone. (Novex, San Diego, CA, USA) and subjected to denaturing SDS- Two arms were open and two arms were covered by 30 cm-high vertical polyacrylamide gel electrophoresis as described previously.14 Membranes black Plexiglas walls. Rats were placed in the central zone and allowed to were incubated in FLAG antibody diluted 1: 5000 (Sigma, Saint Louis, MO, move freely around the maze for 5 min. The session was recorded by video USA) or cMyc antibody diluted 1:1000 (Millipore). and the amount of time spent in the open arms was assessed.

& 2014 Macmillan Publishers Limited Molecular Psychiatry (2014), 986 – 994 m-d heteromer mediates antidepressant-like action N Kabli et al 988 Novelty-induced hypophagia test heteromer internalization (Figure 2a). UFP-512 did not induce the Novelty-induced hypophagia was assessed in singly-housed rats. Animals internalization of mOR expressed alone (Figure 2a). Similar effects were trained to drink a sweetened milk solution in their home cage for 3 were observed with the d-agonist SNC80 (Supplementary Figure days. Only animals that consumed milk solution were included. On the S1D). Thus, the interfering peptide disrupted the m-d heteromer as fourth day, the latency to drink milk solution and the amount consumed it abolished d-agonist-induced internalization of mOR—an effect were assessed in the home cage. On the fifth and final day, rats were seen only when the two receptors are complexed in a heteromer. placed in a novel cage without corn cob bedding under bright light conditions and the latency to drink milk solution and amount consumed were assessed in this novel and anxiogenic environment. Effect of the interfering dOR carboxyl tail peptide on the coimmunoprecipitation of m- and d-Ors Data analysis Dot blotting analysis confirmed the expression of the peptides encoded by the minigenes in transfected cells (Figure 2b). Statistical analyses and graph generation were performed using GraphPad Prism software 3.01 (San Diego, CA, USA). The results were presented as Immunoprecipitation of mOR using a cMyc antibody and means±s.e.m. Data from competition radioligand binding experiments subsequent blotting using a FLAG antibody targeting the dOR were analyzed by nonlinear least-squares regression. An F-test was used to revealed an immunoreactive band atB65–70 kDa consistent with compare the coefficients of goodness-of-fit and to determine whether a immunoprecipitation of the dOR (Figure 2c, lane 1). Expression of two-site (biphasic competition curve with high- and low-affinity sites) or a the interfering peptide resulted in a loss of the immunoreactive one-site (monophasic competition curve with one high- or low-affinity site) band atB65–70 kDa and prevented coimmunoprecipitation analysis was a statistically significant better fit for the radioligand of the dOR with the mOR, suggesting the dissociation of m- and competition binding curves. d-ORs in the m-d heteromer (Figure 2c, lane 2). The control peptide One-way analysis of variance followed by Dunnett’s post hoc analysis had no effect on m- and d-OR coimmunoprecipitation (Figure 2c, was used to assess statistical significance of cell surface radioligand binding, forced swim test, elevated plus maze, novelty-induced hypogha- lane 3). The expression of both receptors and specificity of both gia and locomotor activity data. Po0.05 was deemed significant. antibodies were confirmed (Figure 2c, lanes 4–10).

Effect of intra-accumbens microinjection of the TAT-conjugated RESULTS dOR carboxyl tail interfering peptide on d-agonist-induced Effect of the interfering dOR carboxyl tail peptide on the d-agonist- antidepressant-like action in the forced swim test detected ligand binding pocket in the m-d OR heteromer and the In order to determine the contribution of the m-dOR heteromer in mOR homomer NAc to the antidepressant-like action of the d-agonist UFP-512, To examine whether a minigene expressing the dOR carboxyl tail rats were microinjected with interfering peptide or a scrambled peptide interfered with the m-d physical interaction and disso- control peptide before UFP-512 and examined in the forced swim ciated the m-dOR heteromer, d-agonist modification of high- test model of behavioral despair.40 In our study, we elected to use affinity m-agonist binding to the m-dOR heteromer was assessed. In UFP-512 as it is devoid of convulsive activity,23 has antidepressant- competition radioligand binding studies, the d-agonist UFP-512 like and anxiolytic-like properties in animals,23 does not affect displaced high-affinity [3H]-DAMGO (m-agonist) binding to the m-d locomtor activity,23 recruits the mOR for its actions12 and activates heteromer (Figure 1a; Supplementary Table S1), a phenomenon the m-d heteromer with the highest affinity in vitro.12 Cannula which occurs only when the mOR is complexed with dOR.12 Co- placement in the NAc is illustrated (Supplementary Figure S2). expression of the dOR carboxyl tail peptide minigene abolished A dose-response curve of UFP-512’s effects following intraNAc UFP-512-induced displacement of high-affinity m-agonist binding administration is shown (Supplementary Figure S3). UFP-512 was to the m-dOR heteromer (Figure 1a), and resulted in a right-ward administered at 1.5 mg intraNAc in all subsequent experiments. shift to a monophasic low-affinity competition curve identical to Microinjection of UFP-512 bilaterally into the NAc resulted in a UFP-512’s binding to the mOR expressed alone (that is, mOR significant reduction in the duration of immobility in the forced homomer) (Figure 1b; Supplementary Table S1). The minigene expre- swim test, suggestive of antidepressant-like action (Figure 3a). ssing the control peptide had no effect (Figure 1c; Supplementary Pretreatment with TAT-conjugated interfering peptide reversed Table S1). Similar set of effects was observed with SNC80, another the UFP-512-induced reduction in immobility, whereas the class of d-agonist (Supplementary Figures S1A–C). Furthermore, the scrambled peptide control had no effect (Figure 3a). The tricyclic interfering peptide had no effect on UFP-512 (Figure 1d) or DAMGO antidepressant imipramine administered intraperitoneally served (Figure 1e) binding to the m-receptor ligand binding pocket within as a positive control showing a significant reduction in immobility the mOR homomer, or UFP-512 binding to the d receptor pocket (Figure 3b). Dissociating the m-dOR heteromer in NAc had no within the dOR homomer (Figure 1f; Supplementary Table S2). Thus, effect on imipramine’s antidepressant effects in the forced swim the dOR carboxyl tail peptide selectively disrupted d-agonist test (Figure 3b). Fluoxetine has been shown to have negligible binding to the mOR binding pocket within the m-d heteromer but effect when administered alone in the forced swim test50–52 and not the m-receptor binding pocket within the mOR homomer or the was thus used as a negative control. Thus, the dOR carboxyl tail d-receptor binding pocket within the dOR homomer. interfering peptide selectively antagonized the antidepressant-like effects of agonists acting at the m-dOR heteromer and this Effect of the dOR carboxyl tail peptide on d-agonist-induced receptor complex mediated the antidepressant-like effects of internalization of the m-dOR heteromer d-agonist UFP-512 in the NAc. To determine the effect of dissociating the m-dOR heteromer on d- agonist-induced internalization, receptor endocytosis in intact Role of dOR and mOR in d-agonist-induced antidepressant-like cells stably expressing m- and d-ORs in the presence or absence of effects in the forced swim test the interfering peptide minigene was assessed. The d-agonist UFP- To investigate the contribution of dOR and mOR in UFP-512- 512-induced internalization of mOR in cells co-expressing m- and d- mediated antidepressant-like action in the NAc, rats were ORs, indicative of m-d heteromer internalization, and this was microinjected bilaterally into the NAc with the d-antagonist abolished by co-expression of the interfering peptide minigene naltrindole or the m-antagonist CTOP before UFP-512 administra- (Figure 2a). Pretreatment of the m-d co-expressing cells with tion and behavioral testing. Both naltrindole and CTOP adminis- either the d-antagonist naltrindole or the m-antagonist CTOP tered into the NAc reversed the decrease in immobility induced by abolished d-agonist-induced internalization of mOR, confirming UFP-512 (Figure 3c). Thus, blocking either the mOR or the dOR the requirement of both m- and d-ORs for d-agonist-induced m-d abolished the antidepressant-like action of the d-agonist and

Molecular Psychiatry (2014), 986 – 994 & 2014 Macmillan Publishers Limited m-d heteromer mediates antidepressant-like action N Kabli et al 989

100 100 80 80 60 60 40 40 H-DAMGO bound H-DAMGO bound 3 3 20 20 -OR OR % % 0 -OR + PEP 0 -OR + PEP

-12 -11 -10 -9 -8 -7 -6 -5 -4 -12 -11 -10 -9 -8 -7 -6 -5 -4 log [UFP-512] M log [UFP-512] M

100 100 80 80 60 60 40 40 H-DAMGO bound H-DAMGO bound

3 20 3 20 -OR OR % 0 -OR + CTRL PEP % 0 OR + PEP

-12 -11 -10 -9 -8 -7 -6 -5 -4 -12 -11 -10 -9 -8 -7 -6 -5 -4 log [UFP-512] M log [UFP-512] M

100 100 80 80 60 60 40 40 H-DPN bound 3

H-DAMGO bound 20 3 20 OR % OR % 0 OR + PEP 0 OR + PEP

-12 -11 -10 -9 -8 -7 -6 -5 -4 -12 -11 -10 -9 -8 -7 -6 -5 -4 log [DAMGO] M log [UFP-512] M Figure 1. Effect of the d-opioid receptor (dOR) carboxyl tail peptide minigene on d-agonist UFP-512 modulation of m-agonist binding to the m-dOR heteromer. d-agonist detection of the m-ligand binding pocket using the m-agonist [3H]-DAMGO ([D-Ala2, N-MePhe4, Gly-ol]-enkephalin) in cells co-expressing m- and d-OR in the absence or presence of the interfering (a) or control peptide minigene (CTRL) (c), or in cells expressing m-opioid receptor (mOR) alone (b–d). m-agonist detection of the m-ligand binding pocket in cells expressing mOR alone (e). d-agonist detection of the d-ligand binding pocket using the OR antagonist [3H]-diprenorphine (DPN) in cells expressing dOR in the absence and presence of the interfering or control peptide minigene (f). Gray and black arrowheads point to the high and low ligand binding affinities, respectively. An F-test was used to compare the coefficients of the goodness-of-fit and to determine whether a two-site (biphasic competition curve with high- and low-affinity sites) or a one-site (monophasic competition curve with one high- or low-affinity site) analysis was a statistically significant better fit for the radioligand competition binding curves. Results shown are mean±s.e.m. and curves are representative of n ¼ 3–6 experiments performed in duplicate. demonstrated the involvement of both m- and d-ORs, which aligns significant reduction in latency to drink milk and served as a with the above data demonstrating that dissociating the m-d positive control (Figure 4b). receptor complex reversed the behavioral effects of UFP-512. To investigate the contribution of dOR and mOR to UFP-512’s antidepressant-like and anxiolytic-like action in the novelty- induced hypophagia paradigm, rats received i.c.v. microinjections Effect of intra-accumbens microinjection of the TAT-conjugated of the d-antagonist naltrindole or the m-antagonist CTOP before dOR carboxyl tail interfering peptide on d-agonist-induced UFP-512 administration. A dose-response curve of UFP-512’s anxiolytic-like and antidepressant-like action in the novelty- effects following i.c.v. administration is shown (Supplementary induced hypophagia test Figure S4). UFP-512 was administered at 80 nmol i.c.v. in all In order to determine the contribution of m-dOR heteromers in the subsequent experiments. i.c.v. pretreatment with either naltrin- NAc to the anxiolytic-like and antidepressant-like effects of UFP- dole or CTOP followed by UFP-512 reversed the UFP-512-induced 512 in a model of anxiety and depression featuring anhedonia,53 a decrease in latency to drink milk (Figure 4c). Thus, both m- and significant clinical issue in patients with mood disorders, rats were d-ORs are required for the anxiolytic-like and antidepressant-like microinjected bilaterally into the NAc with interfering peptide action of UFP-512 in this behavioral paradigm. before UFP-512 and examined in the novelty-induced hypophagia test. Intra-accumbens injection of UFP-512 resulted in a significant reduction in the latency to drink milk in a novel environment Effect of intra-accumbens microinjection of the TAT-conjugated (Figure 4a). Pretreatment with TAT-conjugated interfering peptide dOR carboxyl tail interfering peptide on d-agonist-induced followed by UFP-512 reversed the UFP-512-induced reduction in anxiolytic-like action in the elevated plus maze latency to drink milk, whereas the scrambled peptide control had In order to determine the contribution of the m-dOR heteromer no effect (Figure 4a). The benzodiazepine diazepam resulted in a in NAc to the anxiolytic-like effects of UFP-512, rats were

& 2014 Macmillan Publishers Limited Molecular Psychiatry (2014), 986 – 994 m-d heteromer mediates antidepressant-like action N Kabli et al 990 70 μ-δOR μOR 60

OR 50 μ μ–δOR +++ 40 PEP - + - 30 CTRL PEP - - + 20 IB: HA

Internalized 10 *** *** ***

(% of surface receptors) 0

UFP-512 UFP-512 PEP - + - - NALT - - + - CTOP - - - +

μOR + + + + + + - + + + + δOR + + + + + + + + + + - PEP --+ --+ - --+ - CTRL PEP - -+ - -+- - -+-

IP IB: cMyc IB: FLAG cMyc cMyc cMyc

IB: FLAG 70 kDa 70 kDa 70 kDa

Figure 2. Effect of the d-opioid receptor (dOR) carboxyl tail peptide minigene on d-agonist-induced internalization of the m-opioid receptor (mOR) within the m-dOR heteromer and on coimmunoprecipitation of m- and d-ORs. (a) d-agonist-induced internalization of cell surface mOR in cells co-expressing m- and d-ORs in the absence or presence of the interfering peptide minigene or pretreatment with either the d-antagonist naltrindole (NALT) or the m-antagonist CTOP (D-Phe-Cys-Tyr-D-Trp-Orn-Thr-Pen-Thr-NH2) or in cells expressing mOR alone. Agonist-induced reduction in [3H]-DAMGO ([D-Ala2, N-MePhe4, Gly-ol]-enkephalin) labeling in intact human embryonic kidney (HEK) 293T served as an index of mOR internalization. Data shown represent drug-induced loss of cell surface receptors as a percentage of cell surface receptors in vehicle- treated control cells, and are expressed as mean±s.e.m. for n ¼ 3–4 experiments performed in triplicate. (b) Dot blot depicting the expression of the hemaglutinnin (HA)-tagged dOR carboxyl tail peptide minigenes in HEK 293T cells co-expressing m- and d-ORs. (c) Immunoprecipitation of mOR using a cMyc antibody and subsequent blotting using a FLAG antibody targeting the dOR in lysates from cells expressing m- and d-ORs alone (lane 1) or in the presence of interfering (lane 2) or control peptide (lane 3). In cells co-expressing m- and d-ORs, an immunoreactive band was revealed at B65–70 kDa which is consistent immunoprecipitation of the dOR (lane 1); this band was abolished in cells expressing the interfering peptide (lane 2) but not the control peptide (lane 3). Immunoblots depict the expression of mOR and dOR, respectively, in lysates from cells expressing m- and d-OR alone (c) lanes 4 and 8, respectively) or in the presence of expressed interfering (c) lanes 5 and 9, respectively) or control peptide (c) lanes 6 and 10, respectively). The speciﬁcity of the cMyc and FLAG antibodies, respectively, is demonstrated by the absence of an immunoreactive signal in membranes from cells expressing FLAG-tagged dOR and cMyc-tagged mOR, respectively (c) lanes 7 and 11, respectively). Blots have been cropped to enhance clarity and conciseness.

microinjected with interfering peptide before UFP-512 and DISCUSSION examined in the elevated plus maze, a model of anxiety with Our findings indicate that activation of the m-d heteromer in the 54 great face and construct validity. Intra-accumbens injection of NAc produced potent antidepressant-like and anxiolytic-like UFP-512 resulted in a significant increase in the time spent in the effects similar in magnitude to the effects of clinically-relevant open arms of the elevated plus maze, which is suggestive of antidepressants and anxiolytics in rodent models. Dissociating the anxiolytic-like action (Figure 5a). Pretreatment with TAT-conju- m-d heteromer with an interfering peptide which had no effect on gated interfering peptide reversed the UFP-512-induced increase m or d homomeric ORs or antagonizing either one of the com- in time spent in open arms, whereas the scrambled peptide ponent receptors of the heteromer in NAc abolished d-agonist- control had no effect (Figure 5a). The benzodiazepine diazepam induced antidepressant-like and anxiolytic-like effects in animal served as a positive control (Figure 5b). Both naltrindole and CTOP models of behavioral despair, anhedonia and anxiety. reversed the increased time spent in open arms of the maze The antidepressant-like and anxiolytic-like effects of m-d induced by UFP-512, but had no effect on their own (Figure 5c). heteromer activation in the NAc may be accounted for by Thus, both m- and d-ORs are also required for the anxiolytic-like modulation of dopamine release, as supported by anatomical, action of UFP-512. In separate experiments, UFP-512 did not alter nociceptive biochemical, pharmacological and behavioral lines of evidence. In thresholds in the tail immersion assay at any dose or time point the NAc, m- and d-ORs are co-expressed on nerve terminals of GABAergic interneurons which synapse onto local and projection tested (Supplementary Figure S5). Thus, at the doses used, UFP- 55–59 512 selectively produced antidepressant-like and anxiolytic-like dopaminergic neurons. Activation of the m-d heteromer effects—but not analgesia—through the NAc. Further, neither would inhibit GABA (gamma-aminobutyric acid) release and thus UFP-512 nor interfering or scrambled peptides altered locomotor disinhibit dopaminergic neurons resulting in dopamine release, as 56 activity counts (Supplementary Figure S6A) or patterns supported by anatomical data. Indeed, dopamine release in the (Supplementary Figure S6B). Thus the behavioral effects of the NAc occurs within minutes following infusion of d-agonists into 60,61 TAT peptides and the UFP-512-induced reduction in immobility, the same region, coincides with the peak behavioral effects increased time spent in open arms and reduced latency to drink we observed in our study and likely accounts for the rapid reversal milk in the forced swim test, elevated plus maze and novelty- of depressive-like and anxiogenic behavior in our animals.62 In induced hypophagia paradigms, respectively, were not due to confirmation of this hypothesis, d-agonists-induce dopamine non-specific effects on locomotion. release in the NAc in a manner that is reversed by either dOR or

Molecular Psychiatry (2014), 986 – 994 & 2014 Macmillan Publishers Limited m-d heteromer mediates antidepressant-like action N Kabli et al 991 150 150 150 immobility (s/5 min) Duration of 100 100 100

50 *** 50 50 Duration of Duration of *** *** *** *** immobility (s/5 min) immobility (s/5 min) 0 0 0

g) g) PEP saline saline NALTCTOPNALTCTOP UFP-512 PEP SCR UFP-512 saline (i.p.) FLUOX (1 UFP-512 + PEP FLUOX (10 IMI (10mg/kg,IMI (20mg/kg, i.p.) i.p.) UFP-512UFP-512 + + UFP-512 + PEP SCR IMI (20mg/kg + PEP) Figure 3. Effect of dissociating m- and d-ORs on the antidepressant-like effects of the d-agonist UFP-512 in the forced swim test. (a) Rats were microinjected bilaterally into the nucleus accumbens (NAc) with either saline or UFP-512, TAT-conjugated interfering peptide (PEP) or scrambled peptide (PEP SCR) followed by UFP-512, or TAT-conjugated interfering peptide (PEP) or scrambled peptide (PEP SCR) followed by saline. (b) Rats were injected intraperitoneally with saline or imipramine (IMI) or bilaterally into the NAc with ﬂuoxetine (FLUOX). A separate group of animals was microinjected bilaterally into the NAc with TAT-conjugated interfering peptide (PEP) followed by imipramine. (c) Rats were microinjected bilaterally into the NAc with either the d-opioid receptor (dOR) antagonist naltrindole (NALT) or the m-opioid receptor (mOR) antagonist CTOP (D-Phe-Cys-Tyr-D-Trp-Orn-Thr-Pen-Thr-NH2) followed by saline or UFP-512. Depressive-like behavior was assessed using the forced swim test following the last injection. Rat behavior was scored, whereby the predominant behavior for each interval was recorded according to the Lucki method.41 Immobility represented no activity except that which is required to keep the rat’s head above water. Statistical signiﬁcance was determined using one-way analysis of variance (ANOVA) followed by Dunnett post hoc analysis (***Po0.001 relative to saline-treated rats; n ¼ 6–12 animals per group).

50 50 50 to drink milk (sec) Average latency 40 40 40

30 30 30

20 20 20 * ** ** Average latency 10 10 Average latency 10 to drink milk (sec) to drink milk (sec) *** 0 0 0

PEP saline saline NALT CTOP UFP-512 UFP-512 PEP SCR saline, i.p.

UFP-512 + PEP DZP 1 mg/kg, i.p. UFP-512 UFP-512+ NALT + CTOP UFP-512 + PEP SCR Figure 4. Effect of dissociating m- and d-ORs on the anxiolytic-like and antidepressant-like effects of the d-agonist UFP-512 in the novelty- induced hypophagia paradigm. (a). Rats received bilateral intra-accumbens microinjections of saline or UFP-512, or injections of TAT- conjugated interfering peptide (PEP) or scrambled peptide (PEP SCR) followed by UFP-512 or saline. (b) A separate group of rats received intraperitoneal injections of saline or the anxiolytic diazepam (DZP). (c) A separate group of rats received intracerebroventricular microinjections of saline or UFP-512 or microinjections of the d-antagonist naltrindole (NALT) or m-antagonist CTOP (D-Phe-Cys-Tyr-D-Trp-Orn- Thr-Pen-Thr-NH2) followed by UFP-512 or saline. Anhedonia was assessed following the last injection. The latency to drink from palatable milk solution in a novel and stressful environment (a measure of anhedonia) was used as an index of anxiolytic-like antidepressant-like action. Statistical signiﬁcance was determined using one-way analysis of variance (ANOVA) followed by Dunnett post hoc analysis (*Po0.05, **Po0.01, ***Po0.001 relative to saline-treated rats; n ¼ 4–9 animals per group). mOR antagonists,61 suggesting involvement of the m-d heteromer. the NAc. Opioids, by modulating the excitability of the NAc and Furthermore, enhanced dopamine release represents the inducing dopamine release, may restore dopaminergic tone in the mechanism of action of buproprion in the NAc and accounts for NAc of depressed patients. Whereas the utility of m-agonists in the rapid onset of action of this and other antidepressant reversing depressive and anxiogenic behavior in animal models agents.62,63 Further, activation of the m-d heteromer modiﬁed and humans is limited, d-agonists produce potent and well- features of despair, anhedonia and anxiety, suggestive of a documented antidepressant-like and anxiolytic-like effects, pre- common mechanism, likely increased phasic dopamine release, sumably through the activation of the m-d receptor complex. In in alleviating depressive-like and anxiogenic behavior at the level contrast with mOR and dOR which desensitize, the m-d OR of the NAc. heteromer remains functional following prolonged exposure to Our data solidify the case for opioidergic interventions opioids,12,14 making this receptor target worthy of further improving depressive and anxiety-related symptoms through investigation. Further, although d-agonists—which activate the

& 2014 Macmillan Publishers Limited Molecular Psychiatry (2014), 986 – 994 m-d heteromer mediates antidepressant-like action N Kabli et al 992 100 100 100 Average time spent in *** ** 80 *** 80 open arms (sec) 80 * 60 60 60

40 40 40 open arms (sec)

open arms (sec) 20 20 20 Average time spent in Average Average time spent in Average 0 0 0

R T P P O PEP i.p. ALT saline saline N CTO UFP-512 g/kg, i.p. UFP-512 PEP SCR saline,

UFP-512 + PEP UFP-512 + NAL DZP 1 m UFP-512 + CT UFP-512 + PEP SC Figure 5. Effect of dissociating m- and d-ORs on the anxiolytic effects of the d-agonist UFP-512 in the elevated plus maze. (a) Rats received bilateral intra-accumbens microinjections of saline or UFP-512 or injections of TAT-conjugated interfering peptide (PEP) or scrambled peptide (PEP SCR) followed by UFP-512 or saline. (b) A separate group of rats received intraperitoneal injections of saline or the anxiolytic DZP (diazepam). (c) A separate group of rats received intracerebroventricular microinjections of saline or UFP-512 or injections of the d-antagonist naltrindole (NALT) or m-antagonist CTOP (D-Phe-Cys-Tyr-D-Trp-Orn-Thr-Pen-Thr-NH2) followed by UFP-512 or saline. Anxiety was assessed using the elevated plus maze following the last injection. Rat behavior was scored whereby total time spent in open arms was calculated. Statistical signiﬁcance was determined using one-way analysis of variance (ANOVA) followed by Dunnett post hoc analysis (*Po0.05, **Po0.01, ***Po0.001 relative to saline-treated rats; n ¼ 4–9 animals per group).

m-d heteromer—induce dopamine release in the NAc, they do not CONFLICT OF INTEREST induce intracranial self-stimulation and thus have low abuse The authors declare no conflict of interest. potential.64,65 Our findings also highlight the NAc as an important neural modulator of depressive-like and anxiogenic behavior that must ACKNOWLEDGEMENTS not be overlooked for its roles beyond addiction and motivation. This work was funded by a grant from the Canadian Institutes of Health Research and Traditionally, the frontal cortex and hippocampus as well as the an Ontario Mental Health Foundation Research Studentship (NK). Dr Susan R George amygdala have garnered the most attention in studies of holds a Tier I Canada Research Chair in Molecular Neuroscience. depression and anxiety, respectively.66,67 However, these three brain regions all send glutamatergic projections to modulate the function of the NAc.66 Thus, emotional processing through the AUTHOR CONTRIBUTIONS NAc would be shaped by cognitive, contextual and emotional NK and SRG designed the study and wrote the manuscript. NK performed radioligand memory-dependent inputs. Further, a critical component of binding, whole-cell binding, protein expression and coimmunoprecipitation, stereo- depression is anhedonia, or a blunted ability to experience taxic surgery and behavioral experiments. TN made the peptide minigene constructs. pleasurable or rewarding stimuli, which implicates the mesolimbic GB synthesized UFP-512. SRG and BFO supervised the project. reward circuitry in the pathophysiology of depression. Indeed, a number of neuroimaging studies have demonstrated pertur- REFERENCES bations in the mesolimbic reward system. For example, women with major depressive disorder display reduced NAc activity in 1 Trivedi MH, Rush AJ, Wisniewski SR, Nierenberg AA, Warden D, Ritz L et al. 30 Evaluation of outcomes with citalopram for depression using measurement- response to positive stimuli. Further, deep brain stimulation of based care in STAR*D: implications for clinical practice. Am J Psychiatry 2006; 163: the NAc has been has shown promise in reversing depressive 28–40. symptoms such as anhedonia in patients with major depression 2 Darko DF, Risch SC, Gillin JC, Golshan S. Association of beta-endorphin with 32,33 and anxiety. The NAc is also implicated in modulating specific clinical symptoms of depression. Am J Psychiatry 1992; 149: 1162–1167. depressive behavior in rodent models.63,67–70 Thus, modulating 3 Kennedy SE, Koeppe RA, Young EA, Zubieta JK. Dysregulation of endogenous neurotransmission in the NAc may prove to be a fruitful strategy in opioid emotion regulation circuitry in major depression in women. Arch Gen individuals with major depression. Psychiatry 2006; 63: 1199–1208. In conclusion, our data provide evidence of the role of the m-d 4 Zubieta JK, Ketter TA, Bueller JA, Xu Y, Kilbourn MR, Young EA et al. Regulation of opioid receptor heteromer in mediating antidepressant-like and human affective responses by anterior cingulate and limbic mu-opioid neurotransmission. Arch Gen Psychiatry 2003; 60: 1145–1153. anxiolytic-like behaviors in animal models. Future efforts should 5 Nyhuis PW, Gastpar M, Scherbaum N. Opiate treatment in depression refractory to be aimed at designing selective m-d heteromer ligands and antidepressants and electroconvulsive therapy. J Clin Psychopharmacol 2008; 28: mapping the role of the m-d heteromer in various brain loci in 593–595. order to allow for further exploration of this attractive therapeutic 6 Bodkin JA, Zornberg GL, Lukas SE, Cole JO. Buprenorphine treatment of refractory target for the management of treatment-resistant major depres- depression. J Clin Psychopharmacol 1995; 15: 49–57. sion. Available treatment strategies focusing on serotonergic and 7 Emrich HM, Vogt P, Herz A, Kissling W. Antidepressant effects of buprenorphine. monoaminergic neurotransmitter mechanisms only work for a Lancet 1982; 2:709. proportion of patients and only begin to exert their effects weeks 8 Emrich HM. A possible role of opioid substances in depression. Adv Biochem following their administration. Thus, faster-acting therapeutic Psychopharmacol 1982; 32: 77–84. 9 Emrich HM, Vogt P, Herz A. Possible antidepressive effects of opioids: action of agents with alternate mechanisms of action are needed. Coupling buprenorphine. Ann N Y Acad Sci 1982; 398: 108–112. to a distinct signal-transduction pathway and resisting the 10 Extein I, Pickar D, Gold MS, Gold PW, Pottash AL, Sweeney DR et al. desensitization that is characteristic of mORs render the m-d Methadone and morphine in depression [proceedings]. Psychopharmacol Bull heteromer a therapeutic target worthy of further investigation. 1981; 17: 29–33.

Molecular Psychiatry (2014), 986 – 994 & 2014 Macmillan Publishers Limited m-d heteromer mediates antidepressant-like action N Kabli et al 993 11 Hegadoren KM, O’Donnell T, Lanius R, Coupland NJ, Lacaze-Masmonteil N. 36 Schwarze SR, Ho A, Vocero-Akbani A, Dowdy SF. In vivo protein transduction: The role of beta-endorphin in the pathophysiology of major depression. delivery of a biologically active protein into the mouse. Science 1999; 285: Neuropeptides 2009; 43: 341–353. 1569–1572. 12 Kabli N, Martin N, Fan T, Nguyen T, Hasbi A, Balboni G et al. Agonists at the 37 Mao SC, Lin HC, Gean PW. Augmentation of fear extinction by infusion of glycine delta-opioid receptor modify the binding of micro-receptor agonists to the transporter blockers into the amygdala. Mol Pharmacol 2009; 76: 369–378. micro-delta receptor hetero-oligomer. Br J Pharmacol 2010; 161: 1122–1136. 38 Brebner K, Wong TP, Liu L, Liu Y, Campsall P, Gray S et al. Nucleus accumbens 13 George SR, Fan T, Xie Z, Tse R, Tam V, Varghese G et al. Oligomerization of long-term depression and the expression of behavioral sensitization. Science mu- and delta-opioid receptors. Generation of novel functional properties. J Biol 2005; 310: 1340–1343. Chem 2000; 275: 26128–26135. 39 Kim J, Lee S, Park H, Song B, Hong I, Geum D et al. Blockade of amygdala 14 Fan T, Varghese G, Nguyen T, Tse R, O’Dowd BF, George SR. A role for the distal metabotropic glutamate receptor subtype 1 impairs fear extinction. Biochem carboxyl tails in generating the novel pharmacology and G protein activation Biophys Res Commun 2007; 355: 188–193. profile of mu and delta opioid receptor hetero-oligomers. J Biol Chem 2005; 280: 40 Porsolt RD, Anton G, Blavet N, Jalfre M. Behavioural despair in rats: a new model 38478–38488. sensitive to antidepressant treatments. Eur J Pharmacol 1978; 47: 379–391. 15 Gomes I, Gupta A, Filipovska J, Szeto HH, Pintar JE, Devi LA. A role for hetero- 41 Lucki I, Singh A, Kreiss DS. Antidepressant-like behavioral effects of serotonin dimerization of mu and delta opiate receptors in enhancing morphine analgesia. receptor agonists. Neurosci Biobehav Rev 1994; 18: 85–95. Proc Natl Acad Sci USA 2004; 101: 5135–5139. 42 Randall-Thompson JF, Pescatore KA, Unterwald EM. A role for delta opioid 16 Milan-Lobo L, Whistler JL. Heteromerization of the mu- and delta-opioid receptors in the central nucleus of the amygdala in anxiety-like behaviors. receptors produces ligand-biased antagonism and alters mu-receptor trafficking. Psychopharmacology (Berl) 2010; 212: 585–595. J Pharmacol Exp Ther 2011; 337: 868–875. 43 Matsuzaki S, Ikeda H, Akiyama G, Sato M, Moribe S, Suzuki T et al. Role of mu- and 17 Torregrossa MM, Jutkiewicz EM, Mosberg HI, Balboni G, Watson SJ, Woods JH. delta-opioid receptors in the nucleus accumbens in turning behaviour of rats. Peptidic delta opioid receptor agonists produce antidepressant-like effects in the Neuropharmacology 2004; 46: 1089–1096. forced swim test and regulate BDNF mRNA expression in rats. Brain Res 2006; 44 Hawkins MF, Cubic B, Baumeister AA, Barton C. Microinjection of opioid 1069: 172–181. antagonists into the substantia nigra reduces stress-induced eating in rats. Brain 18 Broom DC, Jutkiewicz EM, Folk JE, Traynor JR, Rice KC, Woods JH. Nonpeptidic Res 1992; 584: 261–265. delta-opioid receptor agonists reduce immobility in the forced swim assay in rats. 45 Braida D, Paladini E, Gori E, Sala M. Naltrexone, naltrindole, and CTOP Neuropsychopharmacology 2002; 26: 744–755. block cocaine-induced sensitization to seizures and death. Peptides 1997; 18: 19 Perrine SA, Hoshaw BA, Unterwald EM. Delta opioid receptor ligands modulate 1189–1195. anxiety-like behaviors in the rat. Br J Pharmacol 2006; 147: 864–872. 46 Schmidt BL, Tambeli CH, Levine JD, Gear RW. mu/delta Cooperativity and 20 Jutkiewicz EM. The antidepressant -like effects of delta-opioid receptor agonists. opposing kappa-opioid effects in nucleus accumbens-mediated antinociception Mol Interv 2006; 6: 162–169. in the rat. Eur J Neurosci 2002; 15: 861–868. 21 Broom DC, Jutkiewicz EM, Rice KC, Traynor JR, Woods JH. Behavioral effects of 47 Gear RW, Levine JD. Nucleus accumbens facilitates nociception. Exp Neurol 2011; delta-opioid receptor agonists: potential antidepressants? Jpn J Pharmacol 2002; 229: 502–506. 90: 1–6. 48 Xiong W, Yu LC. Involvements of mu- and kappa-opioid receptors in morphine- 22 Kastin AJ, Scollan EL, Ehrensing RH, Schally AV, Coy DH. Enkephalin and other induced antinociception in the nucleus accumbens of rats. Neurosci Lett 2006; peptides reduce passiveness. Pharmacol Biochem Behav 1978; 9: 515–519. 399: 167–170. 23 Vergura R, Balboni G, Spagnolo B, Gavioli E, Lambert DG, McDonald J et al. 49 Li Y, Li JJ, Yu LC. Anti-nociceptive effect of neuropeptide Y in the nucleus Anxiolytic- and antidepressant-like activities of H-Dmt-Tic-NH-CH(CH2-COOH)- accumbens of rats: an involvement of opioid receptors in the effect. Brain Res Bid (UFP-512), a novel selective delta opioid receptor agonist. Peptides 2008; 29: 2002; 940: 69–78. 93–103. 50 Rogoz Z, Skuza G, Leskiewicz M, Budziszewska B. Effects of co-administration 24 Baamonde A, Dauge V, Ruiz-Gayo M, Fulga IG, Turcaud S, Fournie-Zaluski MC et al. of fluoxetine or tianeptine with metyrapone on immobility time and plasma Antidepressant-type effects of endogenous enkephalins protected by systemic RB corticosterone concentration in rats subjected to the forced swim test. Pharmacol 101 are mediated by opioid delta and dopamine D1 receptor stimulation. Eur J Rep 2008; 60: 880–888. Pharmacol 1992; 216: 157–166. 51 Tatarczynska E, Klodzinska A, Chojnacka-Wojcik E. Effects of combined adminis- 25 Pradhan AA, Befort K, Nozaki C, Gaveriaux-Ruff C, Kieffer BL. The delta tration of 5-HT1A and/or 5-HT1B receptor antagonists and paroxetine or fluox- opioid receptor: an evolving target for the treatment of brain disorders. Trends etine in the forced swimming test in rats. Pol J Pharmacol 2002; 54: 615–623. Pharmacol Sci 2011; 32: 581–590. 52 Berrocoso E, Mico JA. Cooperative opioid and serotonergic mechanisms 26 Scherrer G, Befort K, Contet C, Becker J, Matifas A, Kieffer BL. The delta agonists generate superior antidepressant-like effects in a mice model of depression. Int J DPDPE and deltorphin II recruit predominantly mu receptors to produce thermal Neuropsychopharmacol 2009; 12: 1033–1044. analgesia: a parallel study of mu, delta and combinatorial opioid receptor 53 Dulawa SC, Hen R. Recent advances in animal models of chronic antidepressant knockout mice. Eur J Neurosci 2004; 19: 2239–2248. effects: the novelty-induced hypophagia test. Neurosci Biobehav Rev 2005; 29: 27 Gaveriaux-Ruff C, Kieffer BL. Opioid receptor genes inactivated in mice: the 771–783. highlights. Neuropeptides 2002; 36:62–71. 54 Pellow S, Chopin P, File SE, Briley M. Validation of open:closed arm entries in an 28 Oleskevich S, Leck KJ, Matthaei K, Hendry IA. Enhanced serotonin response in the elevated plus-maze as a measure of anxiety in the rat. J Neurosci Methods 1985; hippocampus of Galphaz protein knock-out mice. Neuroreport 2005; 16: 921–925. 14: 149–167. 29 Gupta A, Mulder J, Gomes I, Rozenfeld R, Bushlin I, Ong E et al. Increased 55 Wang H, Pickel VM. Preferential cytoplasmic localization of delta-opioid receptors abundance of opioid receptor heteromers after chronic morphine administration. in rat striatal patches: comparison with plasmalemmal mu-opioid receptors. Sci Signal 2010; 3: ra54. J Neurosci 2001; 21: 3242–3250. 30 Epstein J, Pan H, Kocsis JH, Yang Y, Butler T, Chusid J et al. Lack of ventral striatal 56 Pickel VM, Towle AC, Joh TH, Chan J. Gamma-aminobutyric acid in the medial rat response to positive stimuli in depressed versus normal subjects. Am J Psychiatry nucleus accumbens: ultrastructural localization in neurons receiving mono- 2006; 163: 1784–1790. synaptic input from catecholaminergic afferents. J Comp Neurol 1988; 272: 1–14. 31 Forbes EE, Hariri AR, Martin SL, Silk JS, Moyles DL, Fisher PM et al. Altered striatal 57 Svingos AL, Moriwaki A, Wang JB, Uhl GR, Pickel VM. Ultrastructural activation predicting real-world positive affect in adolescent major depressive immunocytochemical localization of mu-opioid receptors in rat nucleus disorder. Am J Psychiatry 2009; 166: 64–73. accumbens: extrasynaptic plasmalemmal distribution and association with Leu5- 32 Bewernick BH, Hurlemann R, Matusch A, Kayser S, Grubert C, Hadrysiewicz B et al. enkephalin. J Neurosci 1996; 16: 4162–4173. Nucleus accumbens deep brain stimulation decreases ratings of depression and 58 Svingos AL, Clarke CL, Pickel VM. Cellular sites for activation of delta-opioid anxiety in treatment-resistant depression. Biol Psychiatry 2011; 67: 110–116. receptors in the rat nucleus accumbens shell: relationship with Met5-enkephalin. 33 Grubert C, Hurlemann R, Bewernick BH, Kayser S, Hadrysiewicz B, Axmacher N et J Neurosci 1998; 18: 1923–1933. al. Neuropsychological safety of nucleus accumbens deep brain stimulation for 59 Svingos AL, Moriwaki A, Wang JB, Uhl GR, Pickel VM. mu-Opioid receptors are major depression: effects of 12-month stimulation. World J Biol Psychiatry 2011; localized to extrasynaptic plasma membranes of GABAergic neurons and their 12: 516–527. targets in the rat nucleus accumbens. J Neurosci 1997; 17: 2585–2594. 34 Balboni G, Salvadori S, Guerrini R, Negri L, Giannini E, Jinsmaa Y et al. Potent 60 Hipolito L, Sanchez-Catalan MJ, Zanolini I, Polache A, Granero L. Shell/core delta-opioid receptor agonists containing the Dmt-Tic pharmacophore. JMed differences in mu- and delta-opioid receptor modulation of dopamine efflux in Chem 2002; 45: 5556–5563. nucleus accumbens. Neuropharmacology 2008; 55: 183–189. 35 Paxinos G, Watson CR. The Rat Brain in Stereotaxic Coordinates. Fourth ed. 61 Hirose N, Murakawa K, Takada K, Oi Y, Suzuki T, Nagase H et al. Interactions among (Academic Press: New York, NY, USA, 1998. mu- and delta-opioid receptors, especially putative delta1- and delta2-opioid

& 2014 Macmillan Publishers Limited Molecular Psychiatry (2014), 986 – 994 m-d heteromer mediates antidepressant-like action N Kabli et al 994 receptors, promote dopamine release in the nucleus accumbens. Neuroscience 66 Nestler EJ, Carlezon Jr. WA. The mesolimbic dopamine reward circuit in depres- 2005; 135: 213–225. sion. Biol Psychiatry 2006; 59: 1151–1159. 62 Willner P. The mesolimbic dopamine system as a target for rapid antidepressant 67 Tye KM, Mirzabekov JJ, Warden MR, Ferenczi EA, Tsai HC, Finkelstein J et al. action. Int Clin Psychopharmacol 1997; 12(Suppl 3): S7–14. Dopamine neurons modulate neural encoding and expression of depression- 63 Kitamura Y, Yagi T, Kitagawa K, Shinomiya K, Kawasaki H, Asanuma M et al. Effects related behaviour. Nature 2012; 493: 537–541. of bupropion on the forced swim test and release of dopamine in the nucleus 68 Alexander B, Warner-Schmidt J, Eriksson T, Tamminga C, Arango-Lievano M, accumbens in ACTH-treated rats. Naunyn Schmiedebergs Arch Pharmacol 2010; Ghose S et al. Reversal of depressed behaviors in mice by p11 gene therapy in the 382: 151–158. nucleus accumbens. Sci Transl Med 2010; 2: 54ra76. 64 Bosse KE, Jutkiewicz EM, Gnegy ME, Traynor JR. The selective delta opioid agonist 69 Chaudhury D, Walsh JJ, Friedman AK, Juarez B, Ku SM, Koo JW et al. Rapid SNC80 enhances amphetamine-mediated efﬂux of dopamine from rat striatum. regulation of depression-related behaviours by control of midbrain dopamine Neuropharmacology 2008; 55: 755–762. neurons. Nature 2012; 493: 532–536. 65 Do Carmo GP, Folk JE, Rice KC, Chartoff E, Carlezon Jr. WA, Negus SS. The selective 70 Christoffel DJ, Golden SA, Heshmati M, Graham A, Birnbaum S, Neve RL et al. non-peptidic delta opioid agonist SNC80 does not facilitate intracranial self- Effects of inhibitor of kappaB kinase activity in the nucleus accumbens on stimulation in rats. Eur J Pharmacol 2009; 604: 58–65. emotional behavior. Neuropsychopharmacology 2012; 37: 2615–2623.

Supplementary Information accompanies the paper on the Molecular Psychiatry website (http://www.nature.com/mp)

Molecular Psychiatry (2014), 986 – 994 & 2014 Macmillan Publishers Limited