Dopamine D₁ and corticotrophin-releasing hormone type-2α receptors assemble into functionally interacting complexes in living cells

Abstract

Background and Purpose

Dopamine and corticotrophin-releasing hormone (CRH; also known as corticotrophin-releasing factor) are key neurotransmitters in the interaction between stress and addiction. Repeated treatment with cocaine potentiates glutamatergic transmission in the rat basolateral amygdala/cortex pathway through a synergistic action of D₁-like dopamine receptors and CRH type-2α receptors (CRF_2α receptors). We hypothesized that this observed synergism could be instrumented by heteromers containing the dopamine D₁ receptor and CRF_2α receptor.

Experimental Approach

D₁/CRF_2α receptor heteromerization was demonstrated in HEK293T cells using co-immunoprecipitation, BRET and FRET assays, and by using the heteromer mobilization strategy. The ability of D₁ receptors to signal through calcium, when singly expressed or co-expressed with CRF_2α receptors, was evaluated by the calcium mobilization assay.

Key Results

D₁/CRF_2α receptor heteromers were observed in HEK293T cells. When singly expressed, D₁ receptors were mostly located at the cell surface whereas CRF_2α receptors accumulated intracellularly. Interestingly, co-expression of both receptors promoted D₁ receptor intracellular and CRF_2α receptor cell surface targeting. The heteromerization of D₁/CRF_2α receptors maintained the signalling through cAMP of both receptors but switched D₁ receptor signalling properties, as the heteromeric D₁ receptor was able to mobilize intracellular calcium upon stimulation with a D₁ receptor agonist.

Conclusions and Implications

D₁ and CRF_2α receptors are capable of heterodimerization in living cells. D₁/CRF_2α receptor heteromerization might account, at least in part, for the complex physiological interactions established between dopamine and CRH in normal and pathological conditions such as addiction, representing a new potential pharmacological target.

Tables of Links

TARGETS	LIGANDS
5-HT2A/C receptor	cAMP
α1B-adrenoceptor	CRH (CRF)
μ opioid receptor	Dopamine
CB1 receptor	Glutamate
CRF1 receptor	SCH23390
CRF2 receptor	Urocortin 1
D1 receptor
D2 receptor
GABAB2 receptor

This Table lists key protein targets and ligands in this document, which are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY (Pawson et al., 2014) and are permanently archived in the Concise Guide to PHARMACOLOGY 2013/14 (Alexander et al., 2013b).

Introduction

Stress-induced relapse to drug seeking is one of the main problems in drug addiction treatment (Koob, 2008), in part because of the lack of suitable pharmacological targets. It has been shown that the exposure to drugs of abuse and to stressful stimuli induce similar neuronal plastic changes strengthening excitatory inputs to midbrain dopaminergic neurons (Saal et al., 2003). However, the mechanisms of this interaction are still not fully understood. The available evidence supports a key role for the neurotransmitters corticotrophin-releasing hormone (CRH also known as corticotrophin-releasing factor) and dopamine driving the interaction between stress and addiction (Corominas et al., 2010; George et al., 2012; Gysling, 2012; Zorrilla et al., 2014). CRH signalling occurs through the activation of two class B1 GPRC (Alexander et al., 2013a,2014), CRF₁ and CRF₂ receptors. Several studies have shown that the CRF₁ receptor has a key role in plastic changes associated with stress and addiction (Shalev et al., 2010). However, more recently, the role of the CRF₂ receptor in plastic changes induced by addictive drugs has also been recognized (Wang et al., 2007; Hahn et al., 2009; Cadet et al., 2014; Guan et al., 2014). Wise and his group (Wang et al., 2005) have shown that a stressful stimulus induces the release of CRH in the ventral tegmental area (VTA) of both naive and cocaine-experienced rats, and that glutamate release becomes sensitized to CRH only in cocaine-experienced rats. They have shown that this sensitization of VTA glutamate is mediated by CRF₂ receptors (Wang et al., 2007). In addition, Orozco-Cabal et al. (2008), using an electrophysiological approach, showed a synergism between D₁-like dopamine receptors and CRF₂ receptors in basolateral amygdala to medial prefrontal cortex synaptic transmission in rats subjected to repeated cocaine treatment. Thus, the receptors involved in this synergism could be, in part, responsible for the link between addiction and stress.

Increasing evidence shows that the ability of GPCRs to form oligomers plays a key role in synaptic transmission (Ciruela et al., 2012). It has been documented that dopamine receptors are able to form heteromers between them and with other GPCRs (Perreault et al., 2014). Interestingly, it has been shown that the heteromerization of D₁ and D₂ dopamine receptors, a novel receptorial entity able to induce calcium mobilization through a switch to Gq-mediated activation of PLC (Lee et al., 2004; Rashid et al., 2007). More recent evidence suggests that the mechanisms of calcium mobilization induced by the heteromerization of D₁ and D₂ receptors are more complex, implying other intracellular mediators (Chun et al.,).

The possible heteromerization between D₁ and CRF₂ receptors could be responsible for the synergism between dopamine and CRH transmission (Orozco-Cabal et al., 2008). As the α-isoform of CRF₂ receptors is the main isoform expressed in the brain (Chalmers et al., 1995; Lovenberg et al., 1995; Dautzenberg and Hauger, 2002), we hypothesized that D₁ and CRF_2α receptors assemble into functional interacting complexes (i.e. heteromers). The existence of a D₁/CRF_2α receptor heteromer would constitute a novel potential target for drug addiction pharmacotherapy.

The present study aimed to determine the molecular and functional interactions between D₁ and CRF_2α receptors in living cells. Overall, we provide evidence of the existence of a novel and genuine D₁/CRF_2α receptor heteromer that behaves as a different receptorial entity to their individual constituents.

Methods

Plasmid constructs

The constructs encoding human D₁, CRF₁ and CRF_2α receptors were isolated by Touch-Down PCR using PfuUltra II Fusion HS DNA Polymerase (Agilent Technologies, Palo Alto, CA, USA). PCR products were digested with the respective restriction enzymes and cloned into the expression vector listed in Table 1. The D₂ receptor^YFP and GABA_B2 receptor^YFP constructs were prepared as previously described (Canals et al., 2003). The CRF_2α receptor^Flag and D₁ receptor-nuclear localization sequence (NLS)^YFP were generated by PCR overlapping extension as previously described (Heckman and Pease, 2007), using two coupled primers each. The primers used are listed in Table 1 and the restriction sites within the primers are underlined. All constructs were confirmed by DNA sequencing.

Table 1.

Constructs information

Construct	Vector	Restriction enzyme	Primers
D1 receptor	pcDNA 3.1	EcoRV and NotI	CCGGATATCGGCGCCAACATGAGGACTCTGAACACC
			CGAGCGGCCGCTCAGGTTGGGTGCTGACCGTT
CRF2α receptor	pcDNA 3.1	HindIII and XhoI	CCGAAGCTTGGCGCCACCATGGACGCGGCACTGCTC
			CGACTCGAGTCACACAGCGGCCGTCTGCTT
D1 receptorYFP	pEYFP-N1	EcoRI and BamHI	CTCGAATTCGCCACCATGAGGACTCTGAACACC
			CGTCGCCGTCCAGCTCGACCAG
CRF2α receptorCFP	pECFP-N1	EcoRI and KpnI	CCGGAATTCGCCACCATGGACGCGGCACTG
			CGTCGCCGTCCAGCTCGACCAG
CRF1 receptorYFP	pEYFP-N1	EcoRI and KpnI	GTGGAATTCACCATGGGAGGGCAC
			CCGCGGTACCCAGACTGCTGTGGA
D1 receptormyc/His	pcDNA 3.1 myc/His	EcoRV and BamHI	CCGGATATCGGCGCCAACATGAGGACTCTGAACACC
			CGAGGATCCGGTTGGGTGCTGACCGTT
CRF2α receptorRluc	pRluc-N1	EcoRI and KpnI	CTCGAATTCGCCACCATGGACGCGGCACTGCTC CGCGGTACCGCCACAGCGGCCGTCTGCTTG
D1 receptor-NLSYFP	pEYFP-N1	EcoRI and BamHI	CTCGAATTCGCCACCATGAGGACTCTGAACACC CCTAAGAGGGTTGAAAATCTTTTAAATTTTTTAGCATTAAAGGCATAAATG GCCTTTAATGCTAAAAAATTTAAAAGATTTTCAACCCTCTTAGGATGC CGTCGCCGTCCAGCTCGACCAG
CRF2α receptorFlag	pcDNA 3.1	HindIII and NotI	CCGAAGCTTGGCGCCACCATGGACGCGGCACTGCTC GTCGTCATCCTTGTAGTCCACAGCGGCCGTCTGCTT
			GACTACAAGGATGACGACGATAAGTGAGCG GAATAGAATGACACCTACTCAGACAATCGC

Each construct was generated by PCR amplification and the primers used are listed. The restriction enzymes and vectors into which they were sub-cloned are also indicated.

Cell culture and transfection

HEK293T cells were grown in 100 mm plates with DMEM (Gibco, Gaithersburg, MD, USA) supplemented with 10% FBS (Hyclone Labs, Logan, UT, USA), 1% penicillin/streptomycin 100× (Gibco), 2 mM GlutaMax (Gibco) and 1% non-essential amino acids 100× (Gibco) at 37°C in a 5% CO₂ humidified atmosphere. Transfection was performed 24 h post seeding, using either Lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA, USA) or linear polyethylenimine (Poly Sciences, Inc., Warrington, PA, USA).

Immunofluorescence

HEK293T cells (7 × 10⁵) were grown on 24-well plates on coverslips coated with poly-L-lysine (Sigma, Saint Louis, MO, USA). Between 0.4 and 0.5 μg of plasmid DNA was transfected. Forty-eight hours post-transfection, the cells were fixed with 4% paraformaldehyde (PFA). In some cases, cells transfected with fluorescent chimeras were incubated for 10 min at room temperature with 5 μg mL⁻¹ of WGA⁶⁴⁷ (Invitrogen) in PBS (Winkler Ltda., Santiago, Chile), washed and mounted with Dako mounting media (Dako, Carpenteria, CA, USA).

For immunodetection of wild-type receptors, cells were washed and permeabilized with 0.2% Triton-X100 (Sigma) for 10 min at room temperature, washed twice and blocked with PBS containing 1% BSA (Rockland Immunochemical, Gilbertsville, PA, USA) for 1 h at room temperature. Cells were incubated for 1 h at room temperature with primary antibodies: goat anti-CRF₂ receptor (dilution 1:200; sc-1826, Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) and rabbit anti-D₁ receptor (dilution 1:200; sc-14001, Santa Cruz Biotechnology, Inc.) in PBS–BSA solution. Then, cells were washed and incubated for 1 h with the following secondary antibodies: donkey anti-goat AlexaFluor⁴⁸⁸ and donkey anti-rabbit AlexaFluor⁵⁴⁶. Cells were washed and mounted with mounting media (Dako). In the case of immunofluorescence for D₁ receptor^myc/His, a rabbit anti-mouse antibody was used (dilution 1:200; ab18185, Abcam, Cambridge, MA, USA). For calnexin, a rabbit anti-calnexin was used (dilution 1:200, Sigma).

Confocal microscopy

Fluorescence images were captured with a confocal microscope (Fluoview 1000, Olympus, New York, NY, USA) and Fluoview v6.0 software (Olympus). The images were digitally obtained with a 100× objective (N.A 1.4 oil) and using a sequential mode of laser scanning. All the images were obtained without Gain and with Offset Zero. Captured images were processed with ImageJ software (Rasband, WS, ImageJ, NIH, http://rsbweb.nih.gov/ij/). The deconvolution analysis was performed as previously described (Blanco et al., 2011) with ‘Iterative Deconvolve’ plugins. The co-localization analysis was made with ‘JaCoP’ plugins. Pearson's coefficient was used to measure the overlap of the pixels in dual-channel images, reflecting the degree of dependency between two variables or fluorescent labels (Bolte and Cordelieres, 2006).

cAMP measurement

HEK293T cells (2 × 10⁶) were grown in 60 mm plates and transiently transfected with 2 μg of DNA. Forty-eight hours post-transfection, cells were rapidly washed twice in HBSS (14065-056, Gibco), detached and resuspended in the same buffer with 100 μM IBMX (Sigma). Four thousand cells per well were stimulated with SKF83959 (Tocris Bioscience, Ellisville, MO, USA) or urocortin I (Sigma) prepared in HBSS supplemented with 100 μM IBMX, 5 mM HEPES and 2% DMSO or 100 μM IBMX, 5 mM HEPES and 0.1% BSA respectively. cAMP measurement were performed on 384-well plate using a homogenous time-resolved fluorescence (HTRF) cAMP dynamic 2 Kit (Cisbio International, Bagnols-sur-Cèze, France), according to the manufacturer's recommendations.

Protein extraction and immunoprecipitation

HEK293T (8 × 10⁶) cells growing in a 100 mm dish were transiently transfected with D₁ receptor^myc/His, CRF_2α receptor^Flag, D₁ receptor^myc/His plus CRF_2α receptor^Flag and pcDNA3.1 using Lipofectamine 2000 (Invitrogen). Protein extracts were obtained by using the modified protocol of Burgueño et al. (2003). In brief, 48 h post-transfection, the cells were collected in PBS and washed twice with PBS. Then, the cells were resuspended in lysis buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 1 mM EDTA) with protease inhibitors Complete Mini (Roche Diagnostics, Indianapolis, IN, USA), incubated for 20 min on ice and centrifuged at 19 500× g for 30 min at 4°C. The pellet was resuspended in RIPA (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.25% deoxycholate, 1% NP-40, 1 mM EDTA, Millipore, Temecula, CA, USA) containing protease inhibitors, and homogenized through a piston sonicator (Cell Ultrasonic Disrrupter, Kontes, Vineland, NJ, USA) with two pulses of 5–10 s and then stood for 30 min on ice. Finally, the homogenate was centrifuged at 19 500× g for 30 min at 4°C. The soluble-rich membrane extracts were collected and the protein concentration determined with the Micro BCA™ Protein Assay Kit (Thermo Scientific, Rockford, IL, USA).

For inmunoprecipitation, the soluble-rich membrane extracts were pre-cleared with ‘TrueBlot Anti-Rabbit Ig IP Beads’ (eBioscience, San Diego, CA, USA). The samples were incubated with 0.8 μg rabbit anti-myc antibody (Ab9106, Abcam) according to the manufacturer's recommendations. Loading buffer 2× (8 M urea, 2% SDS, 100 mM DTT, 375 mM Tris, pH 6.8) was added to each sample. Immune complexes were dissociated by addition of DTT (to 25 mM) and heating to 37°C for 2 h and resolved by 8% SDS-PAGE with 8% urea. Proteins were transferred to nitrocellulose membranes and immunoblotted with mouse anti-Flag antibody (Stratagene, La Jolla, CA, USA), and then HRP-conjugated donkey anti-mouse IgG (dilution 1:5000, Jackson ImmnunoResearch Laboratories, Inc., West Grove, PA, USA). The immunoreactive bands were developed using a chemiluminescent detection kit (‘SuperSignal West Pico Chemiluminescent Substrate’, Thermo Scientific) according to the manufacturer's recommendations.

BRET assays

For BRET experiments, HEK293T cells transiently transfected with a constant amount (1 μg) of plasmid encoding CRF_2α receptor^Rluc and increasing amounts (0.25–3 μg) of plasmids encoding for D₁ receptor^YFP, D₂ receptor^YFP or GABA_B2 receptor^YFP, were rapidly washed twice in HBSS (137 mM NaCl, 5.4 mM KCl, 0.3 mM Na₂HPO₄, 0.4 mM KH₂PO₄, 4.2 mM NaHCO₃, 1.3 mM CaCl₂, 0.5 mM MgCl₂, 0.6 mM MgSO₄, 5.6 mM glucose, pH 7.4) containing 10 mM glucose, detached and resuspended in the same buffer. To control the cell number, protein concentration was determined using Micro BCA™ Protein Assay Kit (Thermo Scientific). Cell suspensions (20 μg protein) were distributed in triplicate into 96-well, black microplates with a clear bottom (Corning 3651, Corning, Stockholm, Sweden) for fluorescence measurement or white plates with a solid bottom (Corning 3600, Corning) for BRET determination. For BRET measurement, benzyl-coelenterazine (NanoLight Technology, Pinetop, AZ, USA) was added at a final concentration of 5 μM, and readings were performed 1 min after using the POLARstar Optima plate-reader (BMG Labtech, Durham, NC, USA) that allows the simultaneous integration of the signals detected with two filter settings (475 ± 30 and 535 ± 30 nm). The BRET ratio was defined as previously described (Canals et al., 2003; Ciruela et al., 2004). In brief, the BRET signal was determined by calculating the ratio of the light emitted by yellow fluorescent protein (YFP) (510–560 nm) over the light emitted by Rluc (440–500 nm). The net BRET values were obtained by subtracting the BRET background signal detected when the Rluc-tagged construct was expressed alone. Curves were fitted using a non-linear regression and one-phase exponential association fit equation using the GraphPad Prism software (GraphPad Prism, San Diego, CA, USA).

FRET experiments

FRET between CRF_2α receptor^CFP and D₁ receptor^YFP in living cells was determined by donor recovery after acceptor photobleaching. If FRET occurs, then the bleaching of the acceptor (i.e. YFP) yields a significant increase in fluorescence of the donor [i.e. cyan fluorescent protein (CFP)] (Vilardaga et al., 2008). In brief, cells expressing CFP plus D₁ receptor^YFP, YFP plus CRF_2α receptor^CFP, CRF_2α receptor^CFP alone, CRF_2α receptor^CFP plus D₁ receptor^YFP or CRF_2α receptor^CFP plus D₂ receptor^YFP were mounted in an Attofluor holder (Warner Instruments, Hamden, CT, USA) and placed on an inverted Axio Observer microscope (Zeiss Microimaging, Oberkochen, Germany) equipped with a 63× oil immersion objective and a dual-emission photometry system (TILL Photonics, Gräfelfing, Germany). A Polychrome V (Till Photonics) was used as the light source and signals detected by photodiodes were digitized using a Digidata 1440A analogue/digital converter (Molecular Devices, Sunnyvale, CA, USA). pCLAMP (Molecular Devices) and GraphPad Prism software were used for data collection and analysis respectively. Therefore, upon excitation at 436 ± 10 nm [beam splitter dichroic long-pass (DCLP) 460 nm] and an illumination time set to 10 ms at 10 Hz, the emission light intensities were determined at 535 ± 15 nm (YFP) and 480 ± 20 nm (CFP_pre) with a beam splitter DCLP of 505 nm. Subsequently, acceptor photobleaching was performed by direct illumination of YFP at 500 nm for 10 min. Finally, the emission intensities of YFP and CFP (CFP_post) were recorded again. FRET efficiency was calculated according to the equation: FRET_efficiency = 1 − (CFP_pre/CFP_post).

Calcium measurement

D₁ receptors alone or combined with CRF_2α receptors were transfected into HEK293T cells. Forty-eight hours after transfection, cells were loaded with 5 μM FURA-2 AM (Invitrogen) and 5 μM pluronic acid in HBSS (120 mM NaCl, 5.4 mM KCl, 0.8 mM MgCl₂, 15 mM glucose, 20 mM HEPES, pH 7.4) plus 1.8 mM CaCl₂ for 45 min at 37°C. Then, the medium was changed and allowed to de-esterify for 30 min at 37°C. Finally, the medium was changed to HCSS without calcium and fluorescence of FURA-2 AM was measured at 340 and 380 nm in a Spinning Disk confocal microscope (Olympus) (Leyton et al., 2014).

Statistical analysis

Results are presented as means ± SEM and analyses were carried out with GraphPad Prism v5.0 (GraphPad Software). Accordingly, Mann–Whitney test or one-way anova followed by Bonferroni post hoc test was used to determine significance.

Materials

SKF83959 [6-chloro-2,3,4,5-tetrahydro-3-methyl-1-(3-methylphenyl)-1H-3-benzazepine-7,8-diol hydrobomide] and SCH23390 [(R)-(+)-7-chloro-8-hydroxy-3-methyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine hydrochloride] were purchased from Tocris Biosciences (Bristol, UK). Urocortin I was purchased from Phoenix Peptides (Burlingame, CA, USA).

Results

Differential subcellular distribution of D₁ and CRF_2α receptors

The subcellular distribution of D₁ and CRF_2α receptors was evaluated. Accordingly, HEK293T cells were transiently transfected with D₁ or CRF_2α receptors tagged at their C-terminus with YFP and CFP (D₁ receptor^YFP and CRF_2α receptor^CFP) respectively. The fluorescence labelling of the individual cells was classified into surface (bright ring surrounding the cell) or intracellular (dense intracellular fluorescence) as previously done for α_1A/B- and α_1D-adrenoceptors respectively (Hague et al., 2004b).

As shown in Figure 1A (upper), D₁ receptor^YFP presented a mainly surface phenotype such as previously reported (O'Dowd et al., 2005). Quantitative analysis indicated that 77.4 ± 2.8% of the cells presented the surface phenotype. In contrast, CRF_2α receptor^CFP presented an intracellular phenotype (Figure 1A, bottom). Quantitative analysis indicated that 85.3 ± 2.7% of cells presented an intracellular phenotype. To remove the influence of the tagged fluorescence proteins in the subcellular distribution pattern of these receptors, immunofluorescence was performed in HEK293T cells transiently transfected with D₁ or CRF_2α receptors. As shown in Figure 1B, wild-type receptors showed the same phenotype as tagged receptors.

Figure 1.

Subcellular distribution pattern of D₁ and CRF_2α receptors expressed in HEK293T cells. (A) Fluorescence of chimeric proteins. D₁ receptor^YFP and CRF_2α receptor^CFP expressed in HEK293T-cells, fluorescence images of YFP and CFP were obtained by confocal microscopy. The upper panel shows the surface fluorescence of D₁ receptor^YFP, and the bottom panel shows the intracellular fluorescence of CRF_2α receptor^CFP. (B) Immunofluorescence of wild-type receptors. D₁ receptor (left) and CRF_2α receptor (right) were detected by immnunofluorescence. Six fields of cells per independent experiment (n = 3) were examined. Individual cells were classified as having either fluorescence almost exclusively in a bright ring surrounding the cell (surface), or dense intracellular fluorescence (intracellular). Data are expressed as mean ± SEM and represent results from about 200 cells. *P ≤ 0.05, Mann–Whitney test. Scale bar: 10 μm. (C–D) Concentration–response curves showing cAMP levels. Cells expressing D₁ receptors were stimulated with increasing concentration of SKF83959 (C); data are expressed as mean ± SEM (n = 3). Cells expressing CRF_2α receptors were stimulated with increasing concentration of urocortin (UCN)I (D), data are expressed as mean ± SEM (n = 6).

It has been reported that both D₁ and CRF_2α receptors signal through G_αs in HEK293T cells, increasing cAMP levels (Dautzenberg and Hauger, 2002; Neve et al., 2004; Gutknecht et al., 2008). In order to determine if the transfected receptors were functional in the conditions of the present study, we measured cAMP levels induced by respective agonists in HEK293 cells transfected with each receptor alone. We used the HTRF technique to measure cAMP levels induced by the presence of increasing concentrations of SKF83959, D₁R agonist and of urocortin I, a CRH₂R agonist. The concentration–response curves showed that activation of both receptors induced cAMP accumulation in a ligand concentration-dependent manner with a Log(EC₅₀) = −9.122 ± 0.15 for SKF83959 (Fig. 1C) and Log(EC₅₀) = −10.11 ± 0.29 for urocortin I (Figure 1D). To discount the possibility that cAMP accumulation is due to the activation of endogenous GPCRs present in HEK293T cells, we tested the effect of each ligand on HEK293T cells transfected only with the empty vector.

D₁ and CRF_2α receptors mutually modify their subcellular distribution

It has been reported that receptor heteromerization could change the dynamics and subcellular localization of the receptors involved (Terrillon and Bouvier, 2004). Thus, we evaluated the possibility that these receptors could be co-distributed when they are co-expressed in HEK293T cells. The co-expression of D₁ and CRF_2α receptors drastically changed their subcellular distribution pattern (Figure 2). D₁ receptors^YFP acquired a higher intracellular phenotype in cells co-expressing CRF_2α receptor^CFP showing a pattern similar to CRF_2α receptor^CFP yielding a high degree of superposition of both labels (Figure 2A). Consistently, a Pearson's coefficient of 0.86 ± 0.01 was obtained indicating that both receptors exhibited a high degree of co-localization. Interestingly, in some cells that expressed only D₁ receptor^YFP, in the same preparation, the surface phenotype of D₁ receptor^YFP was preserved (Figure 2A, asterisk). These data indicate that CRF_2α receptor^CFP drags D₁ receptor^YFP to the intracellular compartment, suggesting the formation of a heteromeric complex between D₁ and CRF_2α receptors. Figure 2B shows the quantification of the percentage of cells showing surface or intracellular D₁ receptor^YFP phenotype when expressed alone or with CRF_2α receptor^CFP. A significant decrease in the percentage of cells expressing D₁ receptor^YFP in the surface and a concomitant increase in the intracellular compartment was observed when D₁ receptor^YFP was co-expressed with CRF_2α receptor^CFP. In order to evaluate the specificity of the CRF_2α receptor^CFP induced-retention of D₁ receptor^YFP in the intracellular compartment, HEK293T cells, were co-transfected either with CRF₁ receptor^YFP or CRF₁ receptor^YFP plus CRF_2α receptor^CFP. CRF₁ receptor^YFP expressed alone presented a surface phenotype (data not shown) and its distribution was not changed when both CRF₁ receptor^YFP and CRF_2α receptor^CFP were co-expressed (Figure 2C), further proving that the effect of CRF_2α receptor over D₁ receptor localization was specific.

Figure 2.

D₁ receptor^YFP and CRF_2α receptor^CFP mutually modify their subcellular distribution when they are co-expressed. (A) Co-expression of chimeric protein D₁ receptor^YFP and CRF_2α receptor^CFP in HEK293T cells. Six fields of cells per independent experiment (n = 3) were examined. D₁ receptor^YFP acquires an intracellular phenotype in cells co-expressing CRF_2α receptor^CFP, with high degree of co-distribution in the merge. The cells that only expressed D₁ receptor^YFP maintained their surface phenotype (asterisk). (B) Quantification of surface and intracellular phenotypes when D₁ receptor^YFP is expressed alone or with CRF_2α receptor^CFP. The percentage of cells with receptors at the cell surface is greater when D₁ receptor^YFP was expressed alone, lower when CRF_2α receptor^CFP was expressed alone, and an intermediate level was obtained when the receptors where co-expressed. Data are expressed as mean ± SEM and represent results from about 200 cells of three independent experiments (n = 3) *P ≤ 0.05, Mann–Whitney test. (C) Co-expression of CRF₁ receptor^YFP and CRF_2α receptor^CFP. The subcellular localization of both receptors is maintained.

Heteromerization of D₁R and CRH_2αR

To confirm that D₁ and CRF_2α receptors physically interact, co-immuneprecipitation, BRET and FRET assays were performed in HEK293T cells. D₁ receptors and CRF_2α receptors were shown to physically interact.

Co-immunoprecipitation experiments were performed, as a classical biochemical approach. Interestingly, CRF_2α receptor^Flag (band of ∼70 kDa) was able to co-immunoprecipate together with D₁ receptor^myc/His using a rabbit anti-myc polyclonal antibody, only from extracts of cells transiently transfected with D₁ receptor^myc/His plus CRF_2α receptor^Flag (Figure 3A, IP: lane 3). In addition, a higher molecular weight band (∼130 kDa) was observed. Importantly, these bands did not appear in immunoprecipitates of proteins extracted from cells transfected with D₁ receptor^myc/His (lane 1), CRF_2α receptor^Flag (lane 2) or the empty pcDNA vector (lane 4). Finally, the presence of D₁ receptors in the crude extracts was confirmed by Western blot using a rabbit polyclonal antibody against c-myc (Figure 3A, right). Overall, these results showed that D₁ and CRF_2α receptors form a protein complex.

Figure 3.

Heteromerization of D₁ receptors and CRF_2α receptors. (A) Co-immunoprecipitation of D₁ receptor^myc/His and CRF_2α receptor^Flag from HEK293T cells. Solubilized extracts from cells expressing D₁ receptor^myc/His (lane 1), CRF_2α receptor^Flag (lane 2), D₁ receptor^myc/His plus CRF_2α receptor^Flag (lane 3) or empty pcDNA 3.1 vector (lane 4) were immunoprecipitated with a rabbit anti-myc polyclonal antibody. Extracts (crude) and immunoprecipitates (IP) were analysed using a mouse anti-Flag and a rabbit anti c-myc antibodies. CRF_2α receptor^Flag immunoprecipitated with D₁ receptor^myc/His (IP, lane 3). The presence of D₁ receptor^myc/His was corroborated (lanes 1 and 3). (B) BRET saturation curve. BRET was measured in HEK293T cells co-expressing CRF_2α receptor^Rluc plus D₁ receptor^YFP (blue squares), CRF_2α receptor^Rluc plus D₂ receptor^YFP (red circles) or CRF_2α receptor^Rluc plus GABA_B2 receptor^YFP (black circles). Co-transfections were performed with increasing amounts of the YFP-tagged vectors while the CRF_2α receptor^Rluc was maintained constant. Plotted on the X-axis is the fluorescence value obtained from the YFP, normalized to the luminescence value of CRF_2α receptor^Rluc 10 min after h-coelenterazine incubation. (C) Determination of the D₁ and CRF_2α receptor oligomerization by FRET experiments in living cells. D₁ receptor^YFP and CRF_2α receptor^CFP were expressed in HEK293T cells, and fluorescence images of CFP and YFP were recorded before (pre) and after (post) the YFP was photobleached by 5 min of exposure to light at 500 nm to corroborate the extent of acceptor photodestruction (left panel). Emission intensities of CRF_2α receptor^CFP (480 nm, blue) and D₁ receptor^YFP (535 nm, yellow) from single cells expressing both CRF_2α receptor^CFP and D₁ receptor^YFP were recorded before and after YFP photobleaching (right panel) to determine the FRET efficiency. Scale bar: 10 μm. (D) Quantification of the FRET efficiency of different FRET pairs: CFP plus D₁ receptor^YFP (n = 5), YFP plus CRH_2αR^CFP (n = 10), CRH_2αR^CFP (n = 5), D₁R^YFP plus CRF_2α receptor^CFP (n = 15) and D₂ receptor^YFP plus CRF_2α receptor^CFP (n = 10). The data indicate the mean ± SEM. An asterisk denotes the data that are significantly different from the control FRET pairs (i.e., CFP or YFP co-transfections). *P < 0.001, anova with a Bonferroni multiple comparison post hoc test.

Next, the ability of D₁ and CRF_2α receptors to heteromerize was evaluated using biophysical approaches. Accordingly, the degree of D₁/CRF_2α receptor interaction was analysed by means of BRET experiments (Figure 3B). Thus, a BRET saturation curve was constructed in HEK293T cells co-transfected with a constant amount of CRF_2α receptor^Rluc and increasing concentrations of D₁ receptor^YFP, D₂ receptor^YFP or GABA_B2 receptor^YFP. Interestingly, a positive BRET signal was obtained between CRF_2α receptor^Rluc and D₁ receptor^YFP. As shown in Figure 3B, the BRET ratio between CRF_2α receptor^Rluc and D₁ receptor^YFP increased as a hyperbolic function of the concentration of the D₁ receptor^YFP. Importantly, the pairing of CRF_2α receptor^Rluc with D₂ receptor^YFP led to a non-significant BRET signal, which was within the same range as that observed for the CRF_2α receptor^Rluc/GABA_B2 receptor^YFP pair (i.e. non-specific), thus supporting the specificity of the CRF_2α receptor^Rluc/D₁ receptor^YFP interaction.

Additionally, the formation of D₁/CRF_2α receptor heteromers was also assessed by means of FRET experiments. To this end, cells were transiently transfected with D₁ receptor^YFP and CRF_2α receptor^CFP, and their ability to heteromerize was determined by calculating the FRET efficiency using the donor recovery after acceptor bleaching approach (Fig. 3C). The formation of D₁/CRF_2α receptor heteromers is indicated by the resonance energy transfer between the fluorescent proteins, which was measured by the recovery of the CFP emission after photobleaching of YFP (Figure 3C). The application of the photobleaching protocol to cells expressing only CRF_2α receptor^CFP did not modify the emission intensity of CFP (Figure 3D). As additional controls, to ensure that the FRET signal was due to a specific interaction and not to a random proximity of the fluorophores, FRET efficiency was determined in cells co-expressing CRF_2α receptor^CFP plus YFP, D₁ receptor^YFP plus CFP or CRF_2α receptor^CFP plus D₂ receptor^YFP, at comparable fluorescence levels. Under these experimental conditions the FRET efficiency of the CRF_2α receptor^CFP/D₁ receptor^YFP pair was significantly higher (9.65 ± 1.07; P < 0.001) than that observed in the corresponding negative controls (i.e. CRF_2α receptor^CFP/YFP and CFP/D₁ receptor^YFP pairs) (Figure 3D). Interestingly, the FRET efficiency of the CRF_2α receptor^CFP and D₂ receptor^YFP pair was not significantly different from that observed for the other negative controls used (Figure 3D). Overall, these results strongly suggest that D₁ and CRF_2α receptors physically interact in living cells.

D₁ receptor and CRF_2α receptor form a stable heteromer

In order to further test the capacity of D₁ and CRF_2α receptors to physically interact and the stability of the heteromer, we used the heteromer mobilization strategy described by O'Dowd et al. (2005) associated with immunofluorescence for WGA⁶⁴⁷, a cell surface marker. Briefly, this strategy involves the incorporation of a NLS to a given GPCR such that the GPCR acquires the unique property of being retained in the intracellular compartment in the absence of antagonist, and in the cell surface in the presence of inverse agonist/antagonist. Thus, the heteromeric interaction and its stability are evaluated analysing the co-trafficking of both receptors to the different subcellular compartments. As expected, the incorporation of the NLS sequence reported by O'Dowd et al. (2005) to D₁ receptor^YFP (D₁ receptor-NLS^YFP) conferred a mainly intracellular phenotype for D₁ receptor-NLS^YFP (≈90% of the cells, data not shown).

The use of immunofluorescence with WGA⁶⁴⁷ as marker of the cell surface confirmed that D₁ receptor^YFP fluorescence is mainly present at the cell surface meanwhile CRF_2α receptor^CFP is present mainly intracellularly (Figure 4A). Figure 4B shows the results obtained with the strategy of O'Dowd et al. (2005) co-expressing D₁ receptor^YFP plus CRF_2α receptor^CFP (Fig. 4Ba–c), D₁ receptor-NLS^YFP plus CRF_2α receptor^CFP (Figure 4Bd–f) and of D₁ receptor-NLS^YFP plus CRF_2α receptor^CFP in the presence of 1 μM SCH23390 (Figure 4Bg–i). Colocalization analysis showed that the incorporation of NLS to D₁ receptor^YFP decreased Pearson's correlation coefficient between CFP and WGA⁶⁴⁷ from 0.37 ± 0.02 to 0.02 ± 0.01, and the presence of 1 μM SCH23390 restored the cell surface expression of CRF_2α receptor^CFP up to a Pearson's correlation coefficient of 0.27 ± 0.02 (Figure 4C).

Figure 4.

D₁ and CRF_2α receptors form a stable heteromeric complex in HEK293T cells. (A) Co-localization of CRF_2α receptor^CFP with the plasma membrane marker WGA⁶⁴⁷, when it was co-expressed with: D₁ receptor^YFP (a–c), D₁ receptor-NLS^YFP (d–f) or D₁ receptor-NLS^YFP treated with SCH23390 (1 μM) (g–i). The number of CRF_2α receptors in the plasma membrane was augmented when it was co-expressed with D₁ receptor^YFP (a–c). When co-transfected with D₁ receptor-NLS^YFP, the CRF_2α receptor^CFP co-trafficked with the D₁ receptor-NLS^YFP to the intracellular compartment, being excluded from the plasma membrane (d–f). The treatment with SCH23390 (1 μM) to the cells co-transfected with D₁ receptor-NLS^YFP and CRF_2α receptor^CFP allowed both receptors to be observed at the cell surface (g–i). (B) Pearson's correlation coefficient between CFP and WGA⁶⁴⁷. In each condition, the Pearson's coefficient was calculated from the different transfection conditions. Data are expressed as mean ± SEM and represent results from 18 to 30 cells of two independent experiments. (C) Quantification of the percentage of cells showing a surface phenotype for CRF_2α receptors. The percentage of cells was calculated in the different transfection conditions. Data are expressed as mean ± SEM and represent results from about 200 cells from three independent experiments (n = 3) *P ≤ 0.05, Mann–Whitney test.

The percentage of cells with CRF_2α receptor^CFP at the cell surface was estimated comparing the localization of CFP fluorescence with WGA⁶⁴⁷. When both receptors were co-expressed 41.8 ± 4.8% of cells presented CRF_2α receptor^CFP at the cell surface (Figure 4D). The incorporation of NLS to D₁ receptor^YFP (D₁ receptor-NLS^YFP) significantly decreased the percentage of cells with CRF_2α receptor^CFP at the cell surface indicating that D₁ receptor-NLS^YFP was physically interacting with CRF_2α receptor^CFP and was therefore able to drag CRF_2α receptor^CFP forcing the intracellular localization induced by NLS. The formation of a stable heteromer was further demonstrated in the presence of 1 μM SCH23390, a specific D₁ receptor antagonist/inverse agonist that is able to retain D₁ receptor-NLS^YFP at the cell surface. Indeed, in the presence of 1 μM SCH23390 a significant recovery in the percentage of cells expressing CRF_2α receptor^CFP at the cell surface was observed (Figure 4D). These data show that D₁ receptor^YFP and CRF_2α receptor^CFP form a stable heteromer.

As the intracellular fluorescence given by the D₁ /CRF_2α receptor heteromer looks like the cell reticular compartment, we analysed the colocalization of the receptors with calnexin, a marker of the cellular reticular compartment. As can be seen in Figure 5A, cells transfected with D₁ receptor^myc/His showed a poor colocalization with calnexin that significantly increased when D₁ receptor^myc/His and CRF_2α receptor were co-expressed (Figure 5B). In contrast, immunofluorescence against CRF_2α receptor showed high colocalization with calnexin (Figure 5A) that did not significantly change when both receptors were co-expressed (Figure 5C).

Figure 5.

Colocalization of D₁ and CRF_2α receptors with calnexin, a marker of the cellular reticular compartment. (A) Immunofluoresence of D₁ receptor^myc/His and CRF_2α receptors expressed alone or co-expressed in HEK293T cells and of calnexin. (B) Pearson's correlation coefficient between D₁ receptor^myc/His and calnexin in the presence or absence of CRF_2α receptors (P < 0.001; Mann–Whitney test). (C) Pearson's correlation coefficient between CRF_2α receptors and calnexin in the presence or absence of D₁ receptor^myc/His.

D₁/CRF_2α receptor heteromer and cAMP signalling

In order to test whether the heteromerization between D₁ and CRF_2α receptors switch their cAMP signalling mode, we measured cAMP induced by increasing concentrations of SKF83959 (D₁ receptor agonist; Figure 6A) and urocortin I (CRF₂ receptor agonist; Figure 6B) in HEK293T cells transfected with D₁ receptors, CRF_2α receptors and D₁ plus CRF_2α receptors. As expected, the presence of SKF83959 induced a concentration-dependent increase in cAMP in D₁ receptor, but not CRF_2α receptor, expressing cells (Figure 6A). The D₁ receptor specificity of the response to SKF83959 was confirmed because HEK293T cells transfected with the empty vector did not show a response. Interestingly, SKF83959 induced the same level of response when both receptors were co-expressed. The presence of 10 μM SCH23390 displaced the SKF83959 concentration–response curve to the right obtained with both, D₁ receptor and D₁ plus CRF_2α receptor expressing cells further confirming the specificity of the response (Figure 6A).

Figure 6.

Concentration–response curves for cAMP accumulation. HEK293T cells expressing D₁ receptors and CRF_2α receptor alone and cells co-expressing D₁ plus CRF_2α receptors were incubated in the presence of different concentrations of SKF83959 (A) or urocortin (UCN) I (B) with or without 10 μM SCH23390 (SCH). The results are the mean ± SEM of three independent experiments.

The presence of urocortin I induced a concentration-dependent increase in cAMP in CRF_2α receptor, but not D₁ receptor expressing cells (Figure 6B). The CRF_2α receptor specificity of the response to urocortin I was confirmed because HEK293T cells transfected with the empty vector did not show a response. Similar to what was observed with SKF83959, the presence of urocortin I induced the same level of response when both receptors were co-expressed (Figure 6B). The presence of 10 μM SCH23390 displaced the SKF83959 concentration–response curve to the right obtained with both D₁ receptor and D₁ plus CRF_2α receptor-expressing cells further confirming the specificity of the response (Figure 6A). To test a possible crosstalk between both receptors, we determined cAMP accumulation induced by urocortin I in the presence of SCH23390. As can be seen in Figure 6B, the presence of 10 μM SCH23390 did not modify the urocortin I concentration–response curve. These data indicate that the heteromerization between both receptors does not modify cAMP signalling from each individual receptor.

D₁/CRF_2α receptor heteromer activation by dopamine agonists triggers calcium mobilization

HEK293T cells transfected with only D₁ receptors or D₁ plus CRF_2α receptors were loaded with FURA-2 AM and intracellular calcium mobilization was measured in a calcium-free medium. The addition of 10 μM SKF83959, a D₁ receptor agonist, to cells expressing only D₁ receptors triggered calcium mobilization in just one of the tested cells (n = 67), representing 1.5% of them. However, the addition of 10 μM SKF83959 to cells co-transfected with D₁ plus CRF_2α receptors, triggered calcium mobilization in 36.7% of the tested cells (n = 71) (Figure 7A–B). Mobilization of intracellular calcium in response to the ionophore ionomycin was measured in each tested cell to determine cell responsiveness (data not shown).

Figure 7.

Calcium mobilization induced by the activation of D₁ receptors. HEK293T cells expressing D₁ receptors alone or D₁ plus CRF_2α receptors were loaded with FURA-2 AM. Intracellular calcium mobilization was measured in living cells at 340 and 380 nm in a confocal microscope and time lapse of the 340/380 ratio was plotted. The selective D₁ receptor agonist, SKF83959, generated calcium mobilization when the receptors were co-expressed (n = 71), but no effect was observed when D₁ receptors were expressed alone (n = 67). The data were obtained from four independent experiments.

Discussion and conclusions

In the present study, we show that D₁ receptors and CRF_2α receptors are capable of forming a stable heteromer in living cells, modifying their subcellular localization and signalling. Each receptor has a distinct subcellular localization, the D₁ receptor being mainly at the cell surface and CRF_2α receptor mostly intracellular. Interestingly, when both receptors were co-expressed, a balanced subcellular localization for both receptors was observed, thus suggesting that D₁ and CRF_2α receptors heteromerize. We demonstrated the existence of D₁/CRF_2α receptor heteromers by co-immunoprecipitation, BRET and FRET assays. Furthermore, we show that D₁ and CRF_2α receptors form a stable heteromer using the heteromer mobilization strategy described by O'Dowd et al. (2005). In addition, we showed that the heteromerization of D₁ receptors with CRF_2α receptors switches the signalling of D₁ receptors from a receptor that does not mobilize intracellular calcium to a receptor that mobilizes intracellular calcium. Thus, our results indicate that the heteromerization of D₁ receptors and CRF_2α receptors has at least two fundamental functional implications: (1) change in the subcellular localization of both receptors and (2) change in the signalling cascade associated with D₁ receptor stimulation.

Our results show that transfected CRF_2α receptors were mainly located intracellularly (≈85%); similar to what is has been observed in neurons of the rat dorsal raphe nucleus (Waselus et al., 2009). There are other GPCRs that are constitutively intracellular such as GABA_B1 receptors (Margeta-Mitrovic et al., 2000), CB₁ receptors (Andersson et al., 2003), α_1D-adrenoceptors (Hague et al., 2004a) and 5-HT_2A/C (Magalhaes et al., 2010) receptors. Interestingly, Markovic et al. (2008) have shown that the β isoform of the CRF₂ receptor is constitutively present in the cell surface, in contrast to what we observed with the CRF_2α receptor. The α and β isoforms of the CRF₂ receptor differ in their N-terminal sequence (Grigoriadis et al., 1996). It has been shown that the rat CRF_2α receptor has a non-cleavable pseudo signal peptide of 19-aminoacids in the N-terminal (Rutz et al., 2006), and that the lack of its cleavage affects the expression and traffic of CRF_2α receptors to the plasma membrane (Schulz et al., 2010; Teichmann et al., 2012). Interestingly, Teichmann et al. (2012) have shown that the rat CRF_2α receptor does not dimerize and have determined that this is due to its non-cleavable N-terminal pseudo signal peptide. Our data suggest that the presence of the pseudo signal peptide is not an impediment for the heteromerization of CRF_2α receptors with a class A GPCR such as the D₁ receptor. The 19-amino acid N-terminal segment of the CRF_2α receptor is highly conserved among several species. Therefore, it is not surprising that the human CRF_2α receptor studied herein is also found mainly intracellularly. Other receptors like α_1D-adrenoceptors (Hague et al., 2004a) and CB₁ receptors (Andersson et al., 2003) also have an intracellular localization that depends on their N-terminal. As stated before, the CRF₂ receptor was found predominantly in the cytoplasm in neurons of the rat dorsal raphe nucleus (Waselus et al., 2009). Even though the CRF₂ receptor antibody used by Waselus et al. (2009) might not discriminate between both CRF₂ receptor isoforms (Lukkes et al., 2011), their study relates to the CRF_2α receptor, as it has been shown the CRF_2α isoform is present in neurons and the CRF_2β receptor in non-neuronal cells in the CNS (Chalmers et al., 1995; Lovenberg et al., 1995; Dautzenberg and Hauger, 2002). In contrast to CRF_2α receptor, and as has been previously shown (Vickery and von Zastrow, 1999; O'Dowd et al., 2005; Sun et al., 2009), the transfected D₁ receptor presented a surface phenotype.

Our results show that the heteromerization between D₁ and CRF_2α receptors alters the subcellular localization of both receptors. The need for a GPCR partner to traffic to another subcellular compartment has been previously demonstrated for other GPCRs (Kaupmann et al., 1998; Margeta-Mitrovic et al., 2000). We observed that the CRF_2α receptor decreases the surface phenotype of the D₁ receptor to an intermediate surface/intracellular phenotype. There is evidence that when the D₁ receptor heteromerizes with the D₂ receptor, the D₁ receptor also acquires an intermediate phenotype (So et al., 2005). In this case, the heteromerization increases the presence of D₂ receptors and decreases the presence of D₁ receptors in the cell surface. We also showed that heteromerization between D₁ and CRF_2α receptors promotes the trafficking of CRF_2α receptors to the plasma membrane. Likewise, the D₁ receptor improves the trafficking of μ-opioid receptors (Juhasz et al., 2008) and α_1B-adrenoceptor improves the trafficking of α_1D-adrenoceptors to the cell surface (Hague et al., 2004a). Thus, the aforementioned findings, as well as the present study, suggest that the increased trafficking of CRF_2α receptors to the cell surface and of D₁ receptors to the intracellular compartment depends on the interaction between them. It is tempting to speculate that the changes in subcellular localization of CRF₂ receptors observed in the dorsal raphe after stress could be due to its heteromerization with another GPCR (Waselus et al., 2009).

The heteromer mobilization strategy described by O'Dowd et al. (2005) allowed us to further prove that D₁ and CRF_2α receptors are capable of forming a stable heteromer. In the presence of the NLS sequence in the D₁ receptor, co-expressed D₁ receptor-NLS and the CRF_2α receptor function as a unit present at the cell surface and internalizes as such, in the presence or absence of the D₁ receptor antagonist respectively.

Finally, we demonstrated that the heteromerization of D₁ with CRF_2α receptors conferred the property of mobilizing intracellular calcium. Cells expressing D₁ receptors alone are not capable of mobilizing calcium. In contrast, when the D₁ receptor is co-expressed with the CRF_2α receptor, intracellular calcium mobilization is induced by the presence of a D₁ receptor agonist. Interestingly, our results show a strong parallel between the heteromerization of D₁ with D₂ receptors and of D₁ with CRF_2α receptors, not only at the subcellular distribution level mentioned earlier, but also at the functional level. Similar to what we observed for D₁ and CRF_2α receptors, it has been shown that the capacity of D₁ receptors to mobilize intracellular calcium depends on the formation of a stable D₁/D₂ receptor heteromer (Lee et al., 2004; Hasbi et al., 2009; 2010,). It has been shown that the change in signalling pathway of the D₁ receptor in the D₁/D₂ receptor heteromer is due to a change in the GPCRs, from G_αs to G_αq/11 (Lee et al., 2004; Rashid et al., 2007; Hasbi et al., 2010; Verma et al., 2010). However, there is also evidence that the change in D₁ receptor signalling may also be due to downstream signalling pathways that do not depend on the heteromerization (Chun et al., 2013). There is also evidence of other GPCRs in which the heteromerization switches the GPCRs (George et al., 2000; Mellado et al., 2001; Charles et al., 2003; Ferrada et al., 2009). Thus, it is possible that the calcium mobilization associated with the activation of D₁ receptors when D₁ receptors are co-expressed with CRF_2α receptors is due to a change in the GPCRs. Further studies should address whether the change in intracellular calcium mobilization induced by the activation of D₁ receptors co-expressed with CRF_2α receptors is due to a change in its affinity for Gq (Hasbi et al., 2009) or is due to a downstream crosstalk (Chun et al., 2013), as has been shown for the D₁/D₂ receptor heteromer.

The functional implications of the existence of D₁/ CRF_2α receptor heteromers in vivo are presently unknown. However, there are several anatomical substrates such as the prefrontal cortex, amygdala, bed nucleus of the stria terminalis, lateral septum and VTA where these two receptors could be co-expressed in neurons and/or nerve terminals (Corominas et al., 2010; Gysling, 2012). It is tempting to speculate that the D₁/CRF_2α receptor heteromer could be responsible for the potentiation of glutamatergic transmission from the basolateral amygdala to medial prefrontal cortex (Orozco-Cabal et al., 2008) and for the switch from inhibition to facilitation of CRF_2α receptors observed in the lateral septum of rats repeatedly treated with cocaine (Liu et al., 2005).

In summary, we showed that D₁ and CRF_2αR receptors are capable of forming a stable heteromer with functional implications, such as changes in the subcellular localization of both receptors and in the signalling of D₁ receptors. We would like to suggest that the existence of a D₁/CRF_2α receptor heteromer might explain, at least in part, the complex physiological interactions established between dopamine and CRH in normal and pathological conditions such as addiction, representing a new potential pharmacological target.

Glossary

CFP

cyan fluorescent protein

CRH

corticotrophin-releasing hormone

CRF_2α receptor

corticotrophin-releasing hormone type-2_α receptor

YFP

yellow fluorescent protein

List of author contributions

J. F., P. G., K. A. A. and P. G. S. designed and performed experiments and analysed data. E. H. B. designed experiments and advised on the quantitative analysis of protein expression studies. J. M. C. advised on the design and analysis of calcium signalling studies. J. F., P. G. and P. G. S. wrote the manuscript. F. C. directed and coordinated the FRET and BRET experiments. K. G. directed and coordinated the transfection and signalling studies. All authors contributed on the editing of the manuscript.

Conflicts of interest

The authors declare no conflict of interest.