Abstract
The α1D-adrenergic receptor (α1D-AR) is a G protein-coupled receptor (GPCR) that is poorly trafficked to the cell surface and largely nonfunctional when heterologously expressed by itself in a variety of cell types. We screened a library of approximately 30 other group I GPCRs in a quantitative luminometer assay for the ability to promote α1D-AR cell surface expression. Strikingly, these screens revealed only two receptors capable of inducing robust increases in the amount of α1D-AR at the cell surface: α1B-AR and β2-AR. Confocal imaging confirmed that coexpression with β2-AR resulted in translocation of α1D-AR from intracellular sites to the plasma membrane. Additionally, coimmunoprecipitation studies demonstrated that α1D-AR and β2-AR specifically interact to form heterodimers when coexpressed in HEK-293 cells. Ligand binding studies revealed an increase in total α1D-AR binding sites upon coexpression with β2-AR, but no apparent effect on the pharmacological properties of the receptors. In functional studies, coexpression with β2-AR significantly enhanced the coupling of α1D-AR to norepinephrine-stimulated Ca2+ mobilization. Heterodimerization of β2-AR with α1D-AR also conferred the ability of α1D-AR to cointernalize upon β2-AR agonist stimulation, revealing a novel mechanism by which these different adrenergic receptor subtypes may regulate each other's activity. These findings demonstrate that the selective association of α1D-AR with other receptors is crucial for receptor surface expression and function and also shed light on a novel mechanism of cross talk between α1- and β2-ARs that is mediated through heterodimerization and cross-internalization.
Adrenergic receptors (ARs) mediate physiological responses to the catecholamines norepinephrine (NE) and epinephrine. ARs are subdivided into three major families (α1, α2, and β) based on their structure, pharmacology, and signaling mechanisms (Hieble et al., 1995). At least three closely related subtypes have been identified within each family, with each subtype differentially expressed in various tissues. ARs are members of the rhodopsin-like group I G protein-coupled receptor (GPCR) superfamily and exhibit the characteristic GPCR architecture featuring seven membrane-spanning domains.
Traditionally, GPCRs have been thought to function as monomers, but a significant amount of recent evidence suggests that GPCRs can also exist as dimers consisting of identical or distinct monomeric subunits. Dimerization of GPCRs may alter the receptors' functional, pharmacological, or regulatory properties and, in some cases, may be absolutely required for receptor function (Angers et al., 2002). For example, two nonfunctional GABAB receptors seem to form an obligate heterodimer that is necessary for cell surface expression and functional GABAB receptor activity (Marshall et al., 1999). Sweet and umami taste receptors also seem to form obligate heterodimers (Zhao et al., 2003). Many GPCRs are known to exhibit poor surface expression and functionality when expressed alone in heterologous cells (Tan et al., 2004), but beyond the examples of GABAB and taste receptors, the generality of GPCR heterodimerization in controlling receptor surface expression has not been widely explored.
The α1D-AR is a well documented example of a GPCR that is poorly expressed at the cell surface and largely nonfunctional when heterologously expressed alone in most cell types (Theroux et al., 1996; Chalothorn et al., 2002). Interestingly, we have recently found that heterodimerization with the closely related α1B-AR results in robust surface expression of the normally intracellular α1D-AR while also increasing α1D-AR responsiveness to NE and promoting agonist-induced α1D-AR internalization (Uberti et al., 2003; Hague et al., 2004b). These studies suggested the possibility that heterodimerization with α1B-AR might be required for α1D-AR function. However, other recent studies utilizing double α1A-/α1B-AR knockout mice have revealed the presence of functional α1D-AR, even in the absence of other α1-AR subtypes (Turnbull et al., 2003). Additionally, α1D-AR is found in some tissues that lack α1B-AR expression such as human bladder and specific regions of the spinal cord and brain, as well as certain rat and human blood vessels (Alonso-Llamazares et al., 1995; Michelotti et al., 2000; Tanoue et al., 2002; Sadalge et al., 2003). Thus, it is reasonable to speculate that trafficking of functional α1D-AR to the cell surface may involve the interaction of α1D-AR with other proteins beyond α1B-AR. In the present study, we screened a library of approximately 30 group I GPCRs for possible interacting partners that might be able to traffic α1D-AR to the cell surface in a functional manner.
Methods and Materials
Plasmids and Other Materials. Epitope-tagged (Flag- and HA-tagged) versions of human α1A-, α1B-, and α1D-AR cDNAs have been described previously (Vicentic et al., 2002; Uberti et al., 2003). Human α1D-AR C-terminally tagged GFP construct in pEGFP-N3 was kindly provided by Gozoh Tsujimoto (National Children's Hospital, Tokyo, Japan). HA-tagged β1- and β2-AR cDNAs were kindly provided by Hitoshi Kurose (Kyushu University, Japan). HA-tagged α2A-, α2B-, and α2C-AR cDNAs were kindly provided by Lee Limbird (Vanderbilt University Medical Center, Nashville, TN). Flag-tagged Dopamine1 and Dopamine2 receptor cDNAs were kindly provided by David Sibley (National Institutes of Health, Bethesda, MD). HA-tagged serotonin 5HT1A receptor cDNA was kindly provided by John Raymond (Medical University of South Carolina, Charleston, SC). Flag-tagged angiotensin AT1 and AT2 receptor cDNAs were kindly provided by Victor Dzau (Harvard Medical School, Boston, MA). HA-tagged muscarinic m1–5 acetylcholine receptor cDNAs were kindly provided by Allen Levey (Emory University School of Medicine, Atlanta, GA). HA-tagged opioid receptor cDNAs (μ and δ) were kindly provided by Ping-Yee Law (University of Minnesota School of Medicine, Duluth, MN). Flag-tagged lysophosphatidic acid lysophosphatidic acid-1 and -2 receptor cDNAs were kindly provided by Jerold Chun (University of California, San Diego, CA). Flag-tagged histamine H1–3 receptor cDNAs were kindly provided by Rob Leurs (Vrije Universiteit, The Netherlands). HA-tagged melatonin MT1 and Myc-tagged melatonin MT2 receptor cDNAs were kindly provided by Tarfa Kokkola (University of Kuopio, Finland). Flag-tagged melatonin-related receptor cDNA was kindly provided by Peter J. Morgan (Rowett Research Institute, Scotland), and HA-tagged purinergic P2Y1 receptor cDNA was kindly provided by Michael Salter (University of Toronto, Canada). For the screens examining the effects of receptor coexpression on α1D-AR surface expression, HA-α1D-AR was utilized in cases where the coexpressed receptor was Flag-tagged, whereas Flag-α1D-AR was utilized in cases where the coexpressed receptor was either HA- or Myc-tagged. Expression of the receptors was verified via Western blotting.
Other materials were obtained from the following sources: HEK-293 cells (American Type Culture Collection, Manassas, VA); fura-2/acetoxymethly ester and n-dodecyl-β-d-maltoside (DβM) (Calbiochem, San Diego, CA); (–)-norepinephrine bitartrate, BMY 7378 (8-[2-[4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-8-azaspiro[4.5]decane-7,9-dione dihydrochloride), prazosin, albuterol, Dulbecco's modified Eagle's medium (DMEM), penicillin, concanavalin A, streptomycin, bovine serum albumin (BSA), anti-FLAG M2 affinity resin, and HRP-conjugated anti-FLAG M2 antibody (Sigma-Aldrich, St. Louis, MO); anti-HA affinity matrix and 12CA5 anti-HA monoclonal antibody (Roche Diagnostics, Indianapolis, IN); and ECL reagent and enzyme-linked immunosorbent assay SuperSignal Pico ECL (Pierce Chemical, Rockford, IL). LipofectAMINE 2000 reagent and all electrophoresis reagents and precast 4 to 20% Tris-Glycine polyacrylamide gels were obtained from Invitrogen (Carlsbad, CA).
Cell Culture and Transfections. HEK-293 cells were maintained in DMEM, supplemented with 10% fetal bovine serum, 10 mg/ml streptomycin, and 100 U/ml penicillin at 37°C in a humidified atmosphere with 5% CO2. For heterologous expression of receptors, 2 to 4 μg of cDNA was mixed with LipofectAMINE 2000 (15 μl) and added to 5 ml of serum-free medium in 10-cm plates containing cells at 70 to 90% confluency for 16 h, followed by a change of medium. Cells were harvested 48 to 72 h after transfection.
Surface Expression Assay. HEK-293 cells were transiently transfected with the appropriate epitope-tagged constructs with LipofectAMINE 2000 as described above. After 24 h, cells were split into poly-d-lysine-coated 35-mm dishes and grown overnight at 37°C to 80 to 90% confluency. Cells were then rinsed three times with phosphate-buffered saline (PBS) with Ca2+, fixed with 4% paraformaldehyde in PBS/Ca2+ for 30 min, and rinsed three times with PBS/Ca2+. Cells were then incubated in blocking buffer (2% nonfat milk in PBS/Ca2+) for 30 min and incubated with the appropriate concentrations of HRP-conjugated anti-FLAG M2 or 12CA5 anti-HA monoclonal antibodies in blocking buffer for 1 h at room temperature. Following incubation with the HRP-conjugated anti-Flag antibody, cells were washed three times with blocking buffer, one time with PBS/Ca2+, and then incubated with enzyme-linked immunosorbent assay ECL reagent for 15 sec. Following incubation with the 12CA5 antibody, cells were washed three times with blocking buffer and incubated with the appropriate concentration of HRP-conjugated anti-mouse secondary antibody for 1 h, washed, and then incubated with ECL. The chemiluminescence of the whole 35-mm plate, which corresponds to the amount of receptor on the cell surface (Uberti et al., 2003), was quantified in a TD20/20 luminometer (Turner Designs, Sunnyvale, CA). Control experiments using antitubulin antibodies revealed no detectable luminescence, revealing no antibody penetration of the cells under the fixation conditions that were used. For each data point, three to five plates were averaged per experiment. The results were analyzed using unpaired Student's t tests where applicable (GraphPad Software Inc., San Diego, CA). For internalization assays, cells were first rinsed and then stimulated with or without 10 μM albuterol in DMEM for 30 min at 37°C before cell surface measurements described above. Mean values ± S.E.M. were calculated as percent absorbance in arbitrary units and statistically compared using the unpaired Student's t test, with a p value less than 0.05 considered significant.
Confocal Microscopy. Cells transiently transfected with HA- or GFP-tagged constructs were grown on sterile coverslips, fixed with 4% paraformaldehyde, and permeabilized with saponin buffer containing 2% BSA and 0.04% saponin in PBS for 30 min at room temperature. The cells were then incubated with 12CA5 anti-HA monoclonal antibody for 1 h at room temperature. After three washes with saponin buffer, cells were incubated with a rhodamine red-conjugated anti-mouse IgG at a 1:200 dilution for 1 h at room temperature. After three washes with saponin buffer and one wash with PBS, coverslips were mounted using Vectashield mounting medium (Vector Laboratories (Burlingame, CA). Cells were scanned with a Zeiss LSM 510 laser scanning confocal microscope (Carl Zeiss GmbH, Jena, Germany) as described previously (Uberti et al., 2003; Hague et al., 2004b). For detection of GFP, fluorescein isothiocyanate fluorescence was excited using an argon laser at a wavelength of 488 nm. The absorbed wavelength was detected for 510 to 520 nm for GFP. For detecting rhodamine red, rhodamine fluorescence was excited using a helium-neon laser at a wavelength of 522 nm. The pinhole size was maintained at 1 airy unit for all images.
Radioligand Binding. For radioligand binding, confluent 15-cm plates were washed with PBS and harvested by scraping. Cells were collected by centrifugation and homogenized with a Polytron. Cell membranes were collected by centrifugation at 30,000g for 20 min and resuspended by homogenization in 1× buffer (25 mM HEPES, 150 mM NaCl, pH 7.4, and 5 mM EDTA) with a protease inhibitor cocktail (1 mM benzamidine, 3 μM pepstatin, 3 μM phenylmethylsulfonylfluoride, 3 μM aprotinin, and 3 μM leupeptin). Radioligand binding sites were measured by saturation analysis of the specific binding of the α1-AR antagonist radioligand 125I-BE (20–800 pM) or the β-AR antagonist 125I-pindolol (40–2000 pM). Nonspecific binding was defined as binding in the presence of 10 μM phentolamine (for 125I-BE) or 100 μM isoproterenol (for 125I-pindolol). The pharmacological specificity of α1D-AR binding sites was determined by displacement of 125I-BE (50 pM) by NE, a nonselective adrenergic agonist, prazosin, a nonselective α1-AR antagonist, or BMY 7378, a selective α1D-AR antagonist. The pharmacological specificity of β2-AR binding sites was determined by displacement of 125I-pindolol (100 pM) by the β-AR-selective ligands CGP-12177 (4-[3-[(1,1-dimethyethyl)amino]-2-hydroxypropoxy]-1,3-dihydro-2H-benzimidazol-2-one), CGP-20712A [(±)-2-hydroxy-5-[2-[[2-hydroxy-3-[4-[1-methyl-4-(trifluoromethyl)-1H-imidazol-2-yl]phenoxy]propyl]amino]ethoxy]-benzamide methanesulfonate salt], and ICI 118,551 [(±)-1-[2,3-(dihydro-7-methyl-1H-inden-4-yl)oxy]-3-[(methylethyl)amino-2-butanol]. Data were analyzed by nonlinear regression analysis (Theroux et al., 1996).
Solubilization, Immunoprecipitation, and Western Blot Analysis. Membrane preparations (2–3 mg of protein) were prepared as described above and solubilized with 2% DβM in 1× buffer supplemented with a protease inhibitor cocktail for 2 h at 4°C with gentle agitation. Following solubilization, samples were centrifuged at 16,000g, and supernatants were diluted to 0.2% DβMin1× buffer supplemented with protease inhibitors. Soluble receptors were incubated with M2 anti-FLAG or anti-HA affinity matrix overnight at 4°C with gentle agitation. Resin was collected by centrifugation, washed with 1× buffer, and then eluted with 4× sample buffer (62.5 mM Tris-HCl, pH 6.8, 20% glycerol, 2% SDS, and 5% β-mercaptoethanol).
Immunoprecipitated samples were run on a 4 to 20% Tris-Glycine SDS-polyacrylamide gel electrophoresis, transferred to nitrocellulose, and the membranes were blocked with 5% nonfat dried milk in Tris-buffered saline containing 0.1% Tween 20 for 1 h at room temperature with gentle agitation. Membranes were then incubated with the appropriate concentration of HRP-conjugated M2-anti-FLAG antibody or 12CA5 anti-HA monoclonal antibody for 1 h at room temperature. Membranes were washed with Tris-buffered saline containing 0.1% Tween 20 and detected with ECL directly or alternatively incubated with the appropriate concentration of secondary IgG antibody and then detected with ECL.
Measurement of Intracellular Calcium Mobilization. Intracellular Ca2+ mobilization was measured using fura-2 as described previously (Theroux et al., 1996). In brief, confluent 15-cm plates of transiently transfected HEK-293 cells were washed one time with Ca2+-free Hanks solution and then detached using 0.25% trypsin. Cells were collected in 10 ml of Hanks buffer with Ca2+ and centrifuged for 2 min at 1000g at 4°C. Cells were resuspended in DMEM containing 0.05% BSA and incubated with 1 μM fura-2/AM for 15 min. Cells were then diluted, centrifuged and resuspended in biological salt solution (130 mM NaCl, 5 mM KCl, 1 mM MgCl2, 1.5 mM CaCl2, 20 mM HEPES, 10 mM glucose, and 0.1% BSA), and divided into 3-ml aliquots (2.0 × 106 cells/ml) and placed on ice. Prior to use, cells were warmed to 37°C, centrifuged, resuspended in 3 ml of biological salt solution, and transferred to a cuvette. Luminescence was measured by a PerkinElmer LS50 luminescence spectrofluorometer (PerkinElmer Life and Analytical Sciences (Boston, MA). NE (100 μM) was used to stimulate α1D-AR-induced Ca2+ mobilization. Calculation of [Ca2+]i was performed by equilibrating intra- and extracellular Ca2+ with 30 μM digitonin (Rmax) followed by 9 mM EGTA (Rmin) using a Kd of 225 nM for fura-2. Mean values ± S.E.M. were calculated and were statistically compared using the unpaired Student's t test, with a p value less than 0.05 considered significant. For desensitization assays, cells were pretreated with or without 10 μM albuterol in DMEM for 30 min at 37°C. Cells were then rinsed three times with Ca2+-free Hanks buffer, and Ca2+ mobilization was measured as described above.
Results
β2-AR Promotes α1D-AR Surface Expression. Previous studies have revealed that α1D-AR is primarily found in intracellular compartments when expressed in a variety of heterologous cells (Daly et al., 1998; McCune et al., 2000; Chalothorn et al., 2002; Hague et al., 2004b). Recently, we found that α1B-/α1D-AR heterodimerization can dramatically increase the surface expression and functional activity of α1D-AR (Uberti et al., 2003; Hague et al., 2004b). To screen for other potential interacting receptors that might traffic α1D-AR to the plasma membrane, we used a luminometer-based surface expression assay, which we have previously used to examine the surface expression of epitope-tagged GPCRs (Uberti et al., 2003; Hague et al., 2004b). In this assay, Flag- or HA-tagged α1D-AR was coexpressed with various group I GPCRs in HEK-293 cells, and the cell surface expression of the α1D-AR was quantified (Fig. 1). When expressed alone, α1D-AR was barely detectable on the cell surface, whereas coexpression with α1B-AR significantly increased α1D-AR surface expression, as previously reported (Uberti et al., 2003; Hague et al., 2004b). α1D-AR was also coexpressed in these screens with 28 other GPCRs. Strikingly, the only receptor other than α1B-AR to have an effect on α1D-AR surface expression was β2-AR, which produced nearly as dramatic an enhancement of α1D-AR trafficking as α1B-AR. In contrast, other members of the AR family and all of the other group 1 GPCRs examined did not have any detectable effect on trafficking of α1D-AR to the cell surface. These results reveal a novel and selective effect of β2-AR on facilitating efficient plasma membrane trafficking of α1D-AR, similar to the effect observed previously with α1B-AR.
To confirm these results using a different technique, we performed immunofluorescence confocal microscopy experiments. HEK-293 cells were cotransfected with GFP-tagged α1D-AR and HA-tagged β2-AR (Fig. 2, D–F) or HA-tagged β1-AR (Fig. 2, G–I). α1D-AR was found to be localized intracellularly when expressed alone (Fig. 2, A–C). However, as seen in Fig. 2, D to F, coexpression with β2-AR resulted in a dramatic translocation of α1D-AR from intracellular sites to the plasma membrane. In contrast, coexpression of β1-AR with α1D-AR did not result in any significant translocation of α1D-AR, which remained predominantly intracellular, whereas β1-AR immunostaining was found almost exclusively on the cell surface (Fig. 2, G–I). These data confirm that β2-AR, but not β1-AR, selectively promotes trafficking of α1D-AR to the plasma membrane.
β2-AR Physically Associates with α1D-AR in HEK-293 Cells. The strong enhancement of α1D-AR surface expression induced by coexpression with β2-AR suggested the possibility of a physical association between the two receptors. Coimmunoprecipitation experiments using differentially tagged receptors were performed to test this hypothesis. Flag-tagged α1D-AR was expressed either alone or in combination with HA-tagged β2- or β1-AR in HEK-293 cells (Fig. 3). As expected, Western blotting for Flag-tagged α1D-AR after immunoprecipitation with anti-Flag antibody revealed the presence of α1D-AR (data not shown). Interestingly, HA-tagged β2-AR was robustly coimmunoprecipitated with Flag-tagged α1D-AR (Fig. 3B, lane 2). In contrast, HA-tagged β1-AR did not detectably coimmunoprecipitate with α1D-AR (Fig. 3B, lane 3). The lack of α1D-/β1-AR coimmunoprecipitation was not due to inefficient receptor expression because parallel blots of cell lysates showed a comparable level of expression for both HA-tagged β1- and β2-AR (Fig. 3A). Furthermore, α1D-AR/β2-AR interactions were not detected when the receptors were expressed in separate populations of cells that were solubilized, sonicated, and mixed together prior to immunoprecipitation and Western blot analysis (data not shown). Taken together, these data indicate that α1D-AR and β2-AR exhibit selective heterodimerization in a cellular context.
Pharmacological Properties of α1D-/β2-AR Heterodimers. To determine whether the physical interaction of α1D-AR with β2-AR might result in altered α1D-AR pharmacological properties, ligand binding studies were performed using the α1-AR-selective antagonist 125I-BE. The density of binding sites (Bmax) and the affinity (Kd) for 125I-BE were determined via saturation analysis of specific 125I-BE binding to membranes expressing either α1D-AR alone or α1D-/β2-AR (Table 1). No significant difference in affinity for 125I-BE binding was observed upon coexpression of α1D-AR with β2-AR. However, the density of α1D-AR binding sites increased by nearly 2-fold upon coexpression of the two receptors (Table 1), similar to what has previously been observed for α1D-AR coexpression with α1B-AR (Uberti et al., 2003). These data are consistent with the observed effects of β2-AR and α1B-AR on promoting the surface expression of α1D-AR, since it is known that properly assembled multimeric proteins residing in the plasma membrane often exhibit much slower rates of turnover than unassembled proteins that are trapped in the ER/Golgi complex (Wanamaker et al., 2003). Further pharmacological studies were performed to assess the inhibition of specific 125I-BE binding by the endogenous ligand NE, the α1-AR-selective antagonist prazosin, and the α1D-selective antagonist BMY 7378. The Ki values for all three of these ligands were not significantly different for membranes expressing α1D-AR alone versus membranes expressing the α1D-AR/β2-AR combination (Table 1). Similarly, no changes in binding affinity were observed for several β-AR-selective ligands in displacing 125I-pindolol binding to membranes expressing β2-AR alone versus membranes expressing both α1D-AR and β2-AR (Table 2). Thus, heterodimerization of β2-AR with α1D-AR was not found in these studies to alter the limited number of receptor pharmacological properties that were examined.
Increased Functional Responses and Internalization of α1D-/β2-AR Heterodimers. Since coexpression with β2-AR enhanced α1D-AR surface expression and binding site density, we next determined if this physical interaction might increase α1D-AR functional responses. β2-AR and α1D-AR are known to be primarily coupled to different G proteins (Gs and Gq, respectively). We examined α1D-AR signaling by studying NE-induced intracellular Ca2+ mobilization. It has previously been reported that α1D-AR transfection into certain cell types results in a modest amount of constitutive receptor activity (Garcia-Sainz and Torres-Padilla, 1999; McCune et al., 2000), but we did not observe any evidence for agonist-independent receptor signaling upon α1D-AR transfection into HEK-293 cells. Furthermore, α1D-AR surface expression was not enhanced by treatment of the cells with prazosin in our experiments (data not shown). When cells expressing α1D-AR alone were stimulated with NE, a marginal amount of Ca2+ mobilization was observed, as previously reported (Hague et al., 2004b). However, there was a substantial (∼2.5-fold) increase in NE-stimulated Ca2+ mobilization when β2-AR was coexpressed α1D-AR (Fig. 4A). No significant Ca2+ mobilization was observed in cells transfected with β2-AR alone (data not shown). Since many GPCRs are known to undergo internalization from the cell surface in response to agonist stimulation (Claing et al., 2002), we also examined α1D-AR endocytosis in response to stimulation with the α1-AR-selective agonist phenylephrine (100 μM). When α1D-AR was expressed alone, stimulation with phenylephrine for 30 min had no significant effect on the small amount of α1D-AR on the cell surface. However, when α1D-AR was coexpressed with β2-AR and stimulated in the same fashion, more than 40% internalization of α1D-AR was observed. These results indicate that α1D-/β2-AR heterodimerization not only promotes α1D-AR surface expression, it also enhances α1D-AR-mediated signaling and allows for agonist-promoted internalization of α1D-AR.
Albuterol-Induced Cointernalization and Cross-Desensitization between α1D-/β2-AR Heterodimers. β2-AR is known to undergo rapid internalization from the cell surface upon stimulation with β-adrenergic agonists (Claing et al., 2002). Thus, to further assess the potential functional importance of the physical interaction between β2-AR and α1D-AR, we examined the possibility of cointernalization between these two receptors. As expected, a 30-min treatment with the β2-AR selective agonist albuterol (10 μM) resulted in more than 40% internalization of β2-AR, independent of whether the receptor was expressed alone or with α1D-AR (data not shown). When α1D-AR was expressed alone, no change in receptor surface expression was observed upon albuterol stimulation. Interestingly, however, α1D-AR underwent a robust cointernalization (∼35%) when coexpressed with β2-AR and stimulated with albuterol (Fig. 4C). These data suggest that heterodimerization with β2-AR allows for agonist-promoted cointernalization of α1D-AR.
GPCR internalization is known to play a role in regulating receptor desensitization and resensitization (Claing et al., 2002). Since we found that α1D-AR can undergo cointernalization with agonist-stimulated β2-AR, we next examined whether α1D-AR signaling might become desensitized upon agonist activation of coexpressed β2-AR. HEK-293 cells expressing α1D-AR or the α1D-/β2-AR combination were pretreated for 30 min with 10 μM albuterol, then stimulated with 100 μM NE to examine α1D-AR-induced Ca2+ mobilization. Strikingly, pretreatment with 10 μM albuterol resulted in a nearly complete attenuation of the aforementioned large increase in NE-stimulated Ca2+ mobilization in cells coexpressing α1D-/β2-ARs, whereas it had no effect at all in cells expressing α1D-AR alone (Fig. 5). To examine whether this effect might be due to β2-AR-induced increases in cellular cAMP, we pretreated matched plates of cells with 20 μM forskolin, which directly activates adenylyl cyclase, instead of albuterol. However, no effect of forskolin pretreatment on NE-stimulated Ca2+ mobilization was observed (Fig. 5). To assess whether the effects of the albuterol pretreatment might be due to α1D-/β2-AR cointernalization, we also performed experiments where cells were pretreated with concanavalin A, which prevents receptor internalization (Waldo et al., 1983), prior to albuterol pretreatment. Under these conditions, albuterol pretreatment had no detectable effect on NE-stimulated Ca2+ mobilization in cells coexpressing α1D-/β2-ARs. Taken together, these results suggest that α1D-AR signaling can be desensitized via cointernalization with β2-AR.
Discussion
The α1D-AR has been an enigma in the adrenergic field for many years due to the fact that it is inefficiently trafficked to the plasma membrane and, therefore, very difficult to study when heterologously expressed in most cell types (Theroux et al., 1996; Daly et al., 1998; McCune et al., 2000; Chalothorn et al., 2002; Hague et al., 2004b). We have previously shown that α1D-AR coexpression with α1B-AR results in markedly enhanced trafficking to the plasma membrane (Uberti et al., 2003; Hague et al., 2004b). Here, we show that coexpression with β2-AR, another AR family member, is also capable of dramatically increasing α1D-AR surface expression in heterologous cells. Importantly, the additional α1D-ARs that make it to the cell surface due to heterodimerization with β2-AR are functional since we observed significantly enhanced NE-stimulated Ca2+ mobilization in cells cotransfected with β2-AR. These findings provide an additional mechanism for the efficient trafficking of functional α1D-AR to the cell surface beyond heterodimerization with α1B-AR, which is important because mice lacking α1B-AR are known to retain at least some α1D-AR-mediated responses in certain tissues (Cavalli et al., 1997; Turnbull et al., 2003).
Our screen with approximately 30 different group I GPCRs demonstrated that α1D-AR heterodimerization is extremely selective. Notably, we found that β2-AR, but not the closely related β1-AR, facilitates α1D-AR surface expression and can be robustly coimmunoprecipitated with α1D-AR from cells. This subtype selectivity parallels our previous findings that α1B-AR, but not the closely related α1A-AR, promotes α1D-AR surface expression and exhibits coimmunoprecipitation with α1D-AR (Uberti et al., 2003; Hague et al., 2004b). Similar observations have recently been made for the specificity of GABABR1 heterodimerization, with extensive screens revealing that GABABR2, but not a number of other GPCRs, selectively promotes the surface expression of GABABR1 (Balasubramanian et al., 2004). Furthermore, subtype-selective heterodimerization has also been observed between certain combinations of other GPCRs (Angers et al., 2002). Such observations belie the notion that GPCR heterodimerization is a nonselective and/or artificial process and suggest instead that these interactions between receptors are quite specific, with this selectivity possibly providing important clues as to the physiological importance of the various associations.
We consistently observed in our studies a small amount of α1D-AR on the cell surface even in the absence of coexpression of either α1B-AR or β2-AR. This observation might seem to be inconsistent with the idea that α1D-AR requires heterodimerization with other receptors for trafficking to the plasma membrane. However, it is important to point out that HEK-293 cells are known to express a low level of endogenous β2-AR (Daaka et al., 1997). Thus, in HEK-293 cells transfected with α1D-AR alone, some of the receptor might be able to access the cell surface via heterodimerization with the low level of endogenous β2-AR, an effect that would presumably be greatly magnified upon β2-AR overexpression. Such a dependence on heterodimerization with endogenously expressed receptors may help to explain the variability in α1D-AR surface expression that has been observed in different cell types, with transfected α1D-AR in some cells being found almost completely inside the cell in a nonfunctional state (Theroux et al., 1996; McCune et al., 2000; Chalothorn et al., 2002; Hague et al., 2004b), while being found more significantly at the cell surface and more detectably functional in other cell types (Garcia-Sainz and Torres-Padilla, 1999; Garcia-Sainz et al., 2001; Waldrop et al., 2002). The relative levels of α1D-AR-interacting GPCRs such as α1B-AR and β2-AR expressed in these various cell types may be a central factor in determining the functional activity of transfected α1D-AR.
α1D-AR and β2-AR are both activated by the same endogenous ligands and are also known to be colocalized in many of the same cells in the cardiovascular, central nervous, and immune systems (Young et al., 1990; Nicholas et al., 1996; Guimaraes and Moura, 2001; Kavelaars, 2002). Moreover, there is an extensive literature on cross talk between α1- and β-ARs (Akhter et al., 1997; Michelotti et al., 2000; Dzimiri, 2002; Yue et al., 2004). The interaction that we observed in this study between α1D-AR and β2-AR could serve as a mechanism by which these receptors regulate each other's function in native tissues. Indeed, our studies in heterologous cells demonstrate that a β2-AR-selective agonist can promote robust cointernalization of α1D-AR. In addition, α1D-AR-mediated Ca2+ mobilization was greatly attenuated via pretreatment of cells with a β2-AR-selective agonist. This effect was probably due to cointernalization of α1D-AR with β2-AR since it was blocked by concanavalin A pretreatment and was not mimicked by forskolin pretreatment. Taken together, these studies indicate that β2-AR not only can regulate α1D-AR surface expression but can also control α1D-AR internalization and desensitization, thereby providing a novel mechanism of cross talk between α1- and β-ARs.
Like α1D-AR, many other GPCRs, such as α2C-AR (Daunt et al., 1997; Olli-Lahdesmaki et al., 1999; Hurt et al., 2000), GABABR1 (Marshall et al., 1999; Balasubramanian et al., 2004), trace amine receptors (Bunzow et al., 2001), and odorant receptors (McClintock et al., 1997; Lu et al., 2003; Hague et al., 2004a), among others, are retained in intracellular compartments and, therefore, mostly nonfunctional when expressed alone in heterologous cells. Thus, it is of tremendous physiological importance to understand the mechanisms involved in determining GPCR trafficking to the cell surface. Our studies reveal that subtype-selective heterodimerization is a critical determinant of α1D-AR cell surface expression. Moreover, these studies also provide novel insights into the mechanisms of cross talk between different subfamilies of adrenergic receptor.
Acknowledgments
We thank all of the investigators (mentioned under Materials and Methods) who supplied plasmids for the various receptors examined in our screens and Amanda Castleberry for technical assistance.
Footnotes
-
This work was supported by grants from the National Institutes of Health to R.A.H. and K.P.M., by grants from the American Heart Association to K.P.M and C.H., and by a Distinguished Young Scholar in Medical Research award from the W.M. Keck Foundation to R.A.H.
-
doi:10.1124/jpet.104.079541.
-
ABBREVIATIONS: AR, adrenergic receptor; NE, norepinephrine; GPCR, G protein-coupled receptor; HA, hemagglutinin; GFP, green fluorescent protein; HEK, human embryonic kidney; DβM, n-dodecyl-β-d-maltoside; BMY 7378, 8-[2-[4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-8-azaspiro[4.5]decane-7,9-dione dihydrochloride; DMEM, Dulbecco's modified Eagle's medium; BSA, bovine serum albumin; HRP, horseradish peroxidase; ECL, enzyme-linked chemiluminescence; PBS, phosphate-buffered saline; BE, 2-[β-(4-hydroxyphenyl)-ethylaminomethyl]-tetralone; CGP-12177, 4-[3-[(1,1-dimethyethyl)amino]-2-hydroxypropoxy]-1,3-dihydro-2H-benzimidazol-2-one; CGP-20712A, (±)-2-hydroxy-5-[2-[[2-hydroxy-3-[4-[1-methyl-4-(trifluoromethyl)-1H-imidazol-2-yl]phenoxy]propyl]amino]ethoxy]-benzamide methanesulfonate salt; ICI 118,551, (±)-1-[2,3-(dihydro-7-methyl-1H-inden-4-yl)oxy]-3-[(methylethyl)amino-2-butanol.
-
↵1 Current address: Synaptic Pharmaceutical Corp., Paramus, NJ.
- Received October 20, 2004.
- Accepted December 16, 2004.
- The American Society for Pharmacology and Experimental Therapeutics