Agonist-regulated Interaction between a 2 -Adrenergic Receptors and Spinophilin*

Previously, we demonstrated that the third intracellular (3i) loop of the heptahelical a 2A -adrenergic recep- tor ( a 2A AR) is critical for retention at the basolateral surface of polarized Madin-Darby canine kidney II (MDCKII) cells following their direct targeting to this surface. Findings that the 3i loops of the D 2 dopamine receptors interact with spinophilin (Smith, F. D., Ox-ford, G. S., and Milgram, S. L. (1999) J. Biol. Chem. 274, 19894–19900) and that spinophilin is enriched beneath the basolateral surface of polarized MDCK cells prompted us to assess whether a 2 AR subtypes might also interact with spinophilin. [ 35 S]Met-labeled 3i loops of the a 2A AR (Val 217 -Ala 377 ), a 2B AR (Lys 210 -Trp 354 ), and a 2C AR (Arg 248 -Val 363 ) subtypes interacted with glutathione S -transferase-spinophilin fusion proteins. These interactions could be refined to spinophilin amino acid residues 169–255, in a region between spinophilin’s F-actin binding and phosphatase 1 regulatory domains. Furthermore, these interactions occur in intact cells in an agonist-regulated fashion, because a 2A AR and spi- nophilin coimmunoprecipitation from cells is enhanced by prior treatment with agonist. These findings suggest that spinophilin may contribute MeOH fixation, a localization visualized fixation For co- localization used fixation side m g/ml), ; m g/ml). cells three g/ml) incubated with the h atroom rinsed three coverslip. of the a 2 AR subtypes interact with amino acids 169–255 in spinophilin. A , schematic diagram of the predicted spinophilin domain structure and of GST fusion proteins that were used in pull-down assays. Neurabin is a closely related protein to spinophilin (51% identical and 74% functionally similar at the amino acid level) with its region of least homology being within the region from amino acids 151–444 of spinophilin, especially 286–390 (the region against which the anti-spinophilin antibody used in these studies was developed (25)). B , GST fusion protein pull-down assays were performed as described under “Experimental Procedures.” The amount of 35 S-labeled Gen10–3i loop fusion protein retained in the GSH-agarose-GST fusion protein pellet was visualized by autoradiography and quantitated via scintillation counting. C , the a 2A AR-binding domain of spinophilin can be further refined to Sp169–255; virtually no interaction with Neurabin 146–453 can be observed. For these assays, the radiolabeled probe was [ 35 S]Met-(Met) 4 - a 2A 3i loop GST-Nb146–453. The “ input ” lane reflects 1/15 of the amount of [ 35 S]Met-(Met) 4 - a 2A 3i loop added to each of the GST fusion protein binding incubations.

The three ␣ 2 -adrenergic receptor (␣ 2 AR) 1 subtypes are members of the type II, biogenic amine-binding, G protein-coupled receptor family. These receptor subtypes all couple via the G i /G o family of GTP-binding proteins to the inhibition of ad-enylyl cyclase, inhibition of voltage-dependent calcium channels, potentiation of potassium currents via G protein-coupled, inwardly rectifying potassium channels, activation of phospholipase D, and activation of MAP kinase in native cells (1)(2)(3)(4). In heterologous cell systems, these receptors also couple to the activation of a variety of signaling molecules, including Ras (5-7), p70 S6 kinase (8), MAP kinase (9,10), and phospholipase D (11).
Although all three ␣ 2 ARs appear to activate similar signaling pathways, differences in the cellular trafficking of these subtypes have been reported, both in naive cells and following agonist activation. Subtype-selective differences in agonistelicited ␣ 2 AR redistribution have been noted in several experimental systems (12)(13)(14)(15)(16)(17)(18). The ␣ 2B AR subtype is readily internalized following agonist activation, whereas the ␣ 2A AR subtype typically is not (14,18). The ␣ 2C AR subtype has not been explored in as much detail with regard to agonist-elicited redistribution because of its considerable accumulation intracellularly (14). The ␣ 2 AR subtypes also manifest different trafficking itineraries in polarized Madin-Darby canine kidney II (MDCKII) cells, even in the absence of agonist treatment. The ␣ 2A AR subtype is targeted directly to the basolateral surface (19), whereas the ␣ 2B AR subtype is delivered randomly to both the apical and basolateral surfaces but is selectively retained on the basolateral surface (t1 ⁄2 ϭ 10 -12 h) in contrast to its rapid loss from the apical surface (t1 ⁄2 ϭ 5-15 min) (20). These findings suggest that there is a molecular mechanism responsible for the selective retention of the ␣ 2B AR on the basolateral sub-domain of MDCK cells, probably a retention mechanism shared by the basolaterally targeted ␣ 2A -and ␣ 2C AR subtypes (20). Although ␣ 2C ARs, like ␣ 2A ARs, are directly targeted to and retained on the basolateral subdomain, a significant proportion of these receptors is identifiable in an intracellular pool at steady state (14,18,20); the functional relevance of this intracellular ␣ 2C AR pool has yet to be clarified.
Receptor retention on the lateral subdomain of MDCKII cells likely involves the third intracellular loop of the ␣ 2 AR subtypes. For example, deletion of this loop in the ␣ 2A AR subtype (⌬3i ␣ 2A AR) results in accelerated basolateral receptor turnover (t1 ⁄2 Х 4.5 h) when compared with that for the wild-type receptor or with ␣ 2A AR structures that have been mutated in the N terminus or the C-terminal tail (all possessing a t1 ⁄2 of 10 -12 h) (21). Similarly, the ⌬3i ␣ 2B AR is not enriched at the basolateral surface of MDCKII cells at steady state (22).
Based on our findings that the ␣ 2B AR is rapidly removed from the apical surface following random delivery and that removal of the 3i loops of the ␣ 2A -and ␣ 2␤ AR subtypes accelerates surface turnover of these receptors, we hypothesize that ␣ 2 ARs interact, via their 3i loops, with protein(s) enriched beneath the basolateral surface of MDCKII cells to stabilize * This work was supported in part by National Institutes of Health Grants DK43879 (to L. E. L.) and NS37508 (to R. J. C.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18  their steady-state localization. Consequently, we were particularly intrigued by recent findings that the 3i loop of another G i /G o -coupled G protein-coupled receptor, the D 2 dopamine receptor, interacts with spinophilin (23)(24)(25), and that this protein is enriched beneath the basolateral surface of polarized MDCK cells (24). In addition, the multiple protein-interacting domains within spinophilin (24) suggest that its interaction with the receptor may facilitate the formation of a signaling complex to modulate signaling or recruitment of other proteins to a functional microdomain. The present studies were undertaken to identify whether spinophilin interacts with the 3i loops of the ␣ 2 AR subtypes and, if so, if these interactions are regulated by agents that modify ␣ 2 AR function.

EXPERIMENTAL PROCEDURES
Materials-The pGEMEX-2 vector and TnT in vitro translation kit were from Promega (Madison, WI). The [ 35 S]methionine (1000 Ci/mmol, at 10 mCi/ml) was purchased from PerkinElmer Life Sciences (Boston, MA). PVDF nylon membranes were from Millipore (Bedford, MA). The fast protein liquid chromatography and DEAE-Sephacel columns were from Amersham Pharmacia Biotech (Piscataway, NJ). Dodecyl-␤-maltoside and cholesteryl-hemisuccinate were purchased from Calbiochem (San Diego, CA) and Sigma Chemical Co. (St. Louis, MO), respectively. Antibodies against the HA epitope engineered into the ␣ 2 AR structures was obtained from BABCo (mouse) or from Roche Molecular Biochemicals (rat and mouse). Mouse anti-Myc antibodies were purchased from CLONTECH (Palo Alto, CA). Protein A-agarose was from Vector (Burlingame, CA). Centricon-10 concentrating filters were purchased from Amicon (Beverly, MA). Horseradish peroxidase-labeled anti-mouse and anti-rat antibodies were from Amersham Pharmacia Biotech. Horseradish peroxidase substrate for Western detection was Enhanced Chemiluminescence (ECL, Amersham Pharmacia Biotech). Cy3 and Alexa488 secondary antibodies were from Molecular Probes (Eugene, OR).
MDCKII Cell Culture and Polarization-MDCKII cells were plated at confluence (ϳ1-2.5 ϫ 10 5 cells) and grown on 12-mm Transwell filters (0.4-m pore size, Costar, Cambridge, MA) in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (Sigma) and 100 units/ml penicillin and 10 g/ml streptomycin at 37°C/5% CO 2 as described previously (19) except with daily media changes for 5-7 days. Under these conditions, cells form a monolayer and functionally polarize with distinct apical and basolateral surfaces separated by tight junctions. We routinely verify that tight junctions have formed and that the apical and basolateral compartments are functionally separated from one another using the nontransportable molecule [ 3 H]methoxy-inulin (19). For these leak assays, 2 Ci of [ 3 H]methoxy-inulin is added to the apical subcompartment and incubated for 1 h at 37°C/5% CO 2 followed by counting 100 l of the medium in each of the apical and basolateral subcompartments. Leaks range from 5-10%, and we discard from study any culture wells of Ͼ10% leak.

Immunofluorescent Labeling and Confocal Microscopy
Antibody Purification-Rabbit anti-spinophilin antibodies were generated by injection of purified glutathione S-transferase (GST) fusion proteins (fused to spinophilin amino acids 286 -390) as described previously by MacMillan et al. (26). Antibodies were purified from serum by affinity chromatography. Affinity matrices were generated by mixing 2 ml of Affi-Gel-15 and 1 ml of Affi-Gel-10 (Bio-Rad) equilibrated in 0.1 M HEPES, pH 7.0, in a 10 ml of a Poly Prep chromatography column (Bio-Rad). Purified GST-Sp286 -390 fusion protein (11.7 mg in 6.5 ml of PBS) was loaded onto the column and incubated with inversion for 4 h at 4°C. The resin was washed with 1ϫ PBS until free of unbound GST-Sp286 -390, as determined by A 280 . Unbound sites on the Affi-Gel matrix were blocked by incubation with 1 M ethanolamine for 1 h at 4°C with inversion. The column was equilibrated with 1ϫ PBS (0.05% NaN 3 ) and stored at 4°C. A GST "subtraction column" was prepared in the same manner, except GST alone was coupled to the Affi-Gel 10/15 mixed matrix.
Serum (2 ml) was added to the GST-Sp286 -390 affinity matrix and incubated with rotation for 2 h at room temperature. The column was washed three times with 1ϫ PBS, once with 333 mM NaCl in 1ϫ PBS, and then twice more with 1ϫ PBS. Antibody was eluted twice with 2 ml of 100 mM glycine, pH 2.5, and collected into 200 l of 1 M Tris-HCl, pH 9.0, to neutralize the sample. Eluted antibody was pooled, concentrated, and exchanged into 1ϫ PBS using an Amicon Stirred Cell with a YM30 filter (Amicon). To remove antibody directed against the GST portion of the GST-spinophilin fusion protein, concentrated antibody was incubated with the GST subtraction column, prepared as described above, by rotation for 30 min at room temperature. The pass-through from this column was collected and concentrated using an Amicon Stirred Cell as described above, and utilized as the anti-Sp286 -390 antibody. Antibody concentration was determined to be 1.44 mg/ml by protein assay (Bradford). Optimal working concentrations of antibody in Western and immunolocalization were derived empirically via Western blot analysis and immunofluorescence staining.
Fixation and Immunolabeling-Polarized MDCKII cells stably expressing the individual ␣ 2 AR subtypes were grown on Transwells, as described above, and then rinsed once with PBS-CM (phosphate-buffered saline with 1 mM MgCl 2 and 0.5 mM CaCl 2 ) and fixed for 15 min with either 100% methanol (MeOH) at Ϫ20°C or with 4% paraformaldehyde at room temperature (ϳ22°C) followed by quenching with two sequential 7.5-min incubations with 50 mM NH 4 Cl in PBS-CM. Spinophilin immunolocalization was best observed after MeOH fixation, whereas the ␣ 2 AR localization ("signal-to-background" ratio) was best visualized following paraformaldehyde fixation and quenching. For colocalization studies, we used MeOH for fixation of the polarized MDCKII cells.
After fixation, cells were rinsed two more times in PBS-CM, permeabilized in 0.2% Triton X-100 added to the cell surface of the excised Transwell for 20 min, and incubated in blocking buffer (0.1% Triton X-100 and 2% bovine serum albumin in PBS-CM) for 1 h. Primary antibody was added to the cell side of excised Transwells and incubated for either 1 h at room temperature or overnight (ϳ15 h) at 4°C. Mouse 12CA5 anti-HA antibodies were diluted at 1:250 (4 g/ml), and rabbit anti-spinophilin 286 -390 antibodies were used at a dilution of 1:100 (ϳ10 g/ml). MDCKII cells were washed three times for 15 min in PBS-CM at 22°C before adding secondary antibodies. The secondary antibodies were Alexa488-or Cy3-conjugated anti-rabbit or anti-mouse antibodies, diluted 1:1000 (2 g/ml) and were incubated with the cells for 1 h at room temperature. Cells were again rinsed three times for 15 min in PBS-CM and mounted cell-side-up onto a glass slide with Aqua-Polymount and sealed under a glass coverslip. Images were visualized on a Zeiss LSM 410, laser-scanning, confocal microscope in the Vanderbilt Cell Imaging Core Facility. Images were taken through a 40ϫ oil objective at 1.5ϫ magnification.

Generating [ 35 S]Met-labeled ␣ 2 AR 3i Loops as Ligands
The residues corresponding to the 3i loops of the ␣ 2A AR (amino acids 217-377 (27)), the ␣ 2B AR (amino acids 210 -354 (28)), and the ␣ 2C AR (amino acids 248 -363 (29)) were subcloned into the pGEMEX2 vector in-frame within the polylinker located downstream of the sequence encoding the methionine-rich viral coat protein Gene 10 (30). Alternatively, constructs were generated in which four methionines were inserted via polymerase chain reaction into the N-terminal region of the ␣ 2A AR 3i loop ((Met) 4 -␣ 2A 3i) and subcloned into the pGEMEX2 vector. All DNA constructs were verified by sequencing.
The Gen10 -3i loop fusion proteins and (Met) 4 -3i loops were transcribed, translated, and [ 35 S]Met-labeled using the Promega transcription and translation-coupled (TnT) rabbit reticulocyte lysate kit, as follows: 25 l of TnT reticulocyte lysate was added to 1 l of amino acid mix (1 mM, minus methionine), 2 l of reaction buffer, 1 l of TnT T7 RNA polymerase, 4 l of [ 35 S]methionine (1000 Ci/mmol, at 10 mCi/ml), and 1 l of RNasin ribonuclease inhibitor (40 units/l). Then, 1 g of the appropriate plasmid DNA template was added, and the volume was adjusted to 50 l with nuclease-free water. The mixture was incubated for 90 min at 30°C. Following each synthesis, products were analyzed and quantitated by 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and autoradiography. The band representing each probe was cut out of the dried gel and counted in scintillation mixture. GST-pull-down assays were performed such that each incubation contained an equivalent amount of [ 35 S]Met-labeled 3i loop as radioligand.

GST-spinophilin Fusion Protein Generation
GST-spinophilin fusion proteins were generated with spinophilin amino acid regions 151-444 and 169 -255 and expressed in DH5␣. Bacteria were grown at 37°C to an A 600 of 0.6. GST or GST fusion protein expression was initiated with the addition of 1 mM isopropyl-␤-D-thiogalactopyranoside and allowed to proceed for 2-6 h at 37°C. Bacteria were collected by centrifugation at 10,000 ϫ g and then lysed in 50 mM Tris-HCl, pH 7.4, 0.5% Triton X-100, 1 mg/ml lysozyme, 200 mM NaCl, 100 M PMSF, 1 g/ml soybean trypsin inhibitor, 1 g/ml leupeptin, 10 units/ml aprotinin (TT ϩ buffer) by one freeze-thaw cycle followed by probe sonication for three 30-s bursts on ice. GSH-agarose (1 ml of a 1:1 slurry equilibrated in TT ϩ buffer) was added to the supernatant of a 13,000 ϫ g centrifugation and incubated for 1 h at 4°C with inversion. This solution was transferred to a 0.8-ϫ 4-cm Poly-Prep column (Bio-Rad) and washed with 12 ml of TT ϩ buffer, 3 ml of 333 mM NaCl in TT ϩ buffer, and then with 6 ml of TT ϩ buffer. GST or GST fusion protein was eluted from the GSH-agarose by adding 3 ml of 10 mM free acid GSH in TT ϩ , pH 7.5. Eluted protein was concentrated and exchanged into PBS buffer using an Amicon Stirred Cell.

Binding of 3i Loops to GST-spinophilin
Equimolar concentrations of GST-spinophilin fusion protein were incubated with 300,000 cpm (estimated to represent ϳ40 pM) [ 35 S]Metlabeled ␣ 2A , ␣ 2B , or ␣ 2C 3i loop ligand (see above). GSH-agarose (1:1 slurry equilibrated with TT ϩ buffer) was then added to this incubation, rotated for 2 h at 4°C, and the resin collected by centrifugation. The resin was then exposed to four 1-ml TT ϩ washes. Interaction with GST-spinophilin versus GST (controls) was determined by elution of the 3i loop into 1ϫ Laemmli buffer (400 mM Tris, pH 6.8, 700 mM ␤-mercaptoethanol, 1% SDS, 10% glycerol) and separation of the eluates by 12% SDS-PAGE. The degree of interaction was quantitated by cutting and counting the bands corresponding to 3i loop (determined via autoradiography) in scintillation mixture.

Detergent Extraction and Coimmunoprecipitation of Full-length HA-tagged ␣ 2 AR Subtypes with Full-length Myc-tagged Spinophilin
CosM6 cells were plated at 1.75 ϫ 10 6 cells on 10-cm plates and maintained in DMEM supplemented with 10% fetal bovine serum and 100 units/ml penicillin and 10 g/ml streptomycin at 37°C/5% CO 2 . The following day, cells (at ϳ60 -80% confluence) were transfected using FuGENE 6 reagent (Roche Molecular Biochemicals), according to the manufacturer's specifications, with an empirically optimized ratio of 3 l FuGENE 6 reagent/1 g of plasmid DNA. The ␣ 2A , ␣ 2B , and ␣ 2C ARs (GenBank accession numbers A38316, X74400, and X57659, respectively) were encoded in pCMV4 and tagged at their 5Ј-end after the start ATG codon with the sequence corresponding to the hemagglutinin tag (HA; YPYDVPDYA), as described previously (19,20). Full-length spinophilin (GenBank accession AF016252) was expressed in pCMV4, and epitope-tagged with a Myc sequence inserted 5Ј after the start ATG codon (Myc; QKLISEEDLLRKR).
The supernatant of a 100,000 ϫ g centrifugation for 1 h at 4°C was defined as the D␤M/CHS-solubilized preparation. A 0.75-ml aliquot of this preparation was "precleared" by a 15-min incubation with 30 l of protein A-agarose equilibrated with D␤M/CHS buffer. HA-tagged ␣ 2A AR or Myc-tagged spinophilin was then immunoprecipitated following the addition of rat anti-HA monoclonal antibody or mouse anti-Myc monoclonal antibody, respectively, each at a 1:100 dilution, and incubation at 4°C for 1 h. The immune complex was isolated by centrifugation following a 1-h adsorption to protein A-agarose; nonspecifically adsorbed proteins were removed by washing the protein A resin three times in D␤M/CHS wash buffer (1 mg/ml D␤M, 0.2 mg/ml CHS, 20% glycerol, 25 mM glycylglycine, pH 7.6, 20 mM HEPES, pH 7.6, 100 mM NaCl, 5 mM EGTA, 1 g/ml soybean trypsin inhibitor, 1 g/ml leupeptin, 10 units/ml aprotinin, and 100 M PMSF) and centrifugation at 4°C. Proteins were eluted with the addition of 1ϫ Laemmli buffer and heating to 70°C for 5 min. Eluates were separated via 10% SDS-PAGE, transferred to an Immobilon P membrane (PVDF; Millipore) with a constant current of 1 amp for 72 min in CAPS transfer buffer (1 M cyclohexylamino-1-propane sulfonic acid (CAPS), pH 11, 10% methanol), and subjected to Western blot analysis.

Western Blot Analysis
PVDF membranes were blocked for 15 min in Tris-buffered saline (20 mM Tris, pH 7.6, 137 mM NaCl) with 0.1% Tween 20 (TBST) and 5% Carnation Instant powdered milk (w/v). The appropriate primary antibody was then added at a dilution of 1:1000 (Rat anti-HA) or 1:2000 (mouse anti-Myc monoclonal antibody) in blocking buffer and incubated at room temperature for 1.5-2 h. Blots were washed three times for 15 min with TBST and exposed to horseradish peroxidase-conjugated antirat or anti-mouse secondary antibodies, as appropriate, at a 1:2000 dilution in blocking buffer for 45 min at room temperature. Blots were washed again three times for 15 min in TBST, incubated with ECL Western blotting detection reagent (Amersham Pharmacia Biotech, Buckinghamshire, UK) for 1.5 min, and then exposed to x-ray film for variable times ranging from 5 s to 30 min. 1-153), a PP1 binding/regulatory region (amino acids 427-470 (26,31,32)), a single PDZ binding domain, and a C terminus that possesses a series of coiled-coil domains (see schematic in Fig. 2A). Satoh et al. (24) showed that spinophilin was localized to the lateral sub-domain in polarized MDCK cells. As shown previously, the ␣ 2A -adrenergic receptor also is enriched on the lateral sub-domain of these cells (19,20) and is revealed here using a Cy3 (red signal)conjugated secondary antibody directed against the 12CA5 antibody that recognizes the N-terminal HA epitope in the ␣ 2A AR (Fig. 1). A rabbit polyclonal antibody was raised against amino acids 286 -390 in spinophilin (26), a region that has virtually no sequence similarity to spinophilin's structural homolog, neurabin I ( Fig. 2A). As shown in Fig. 1, the affinity-purified polyclonal antibody against spinophilin, visualized here via Alexa488 (green signal)-conjugated secondary antibody, re- The immunolocalization of the receptor and spinophilin was performed as described under "Experimental Procedures." Secondary antibodies were Cy3-conjugated donkey anti-mouse (red) and Alexa488-conjugated goat anti-rabbit (green). The presence of yellow in the overlay image indicates colocalization of the fluorescent signal. Images were taken through a 40ϫ oil objective (NA ϭ 1.4) via a Leica-TCS laser-scanning confocal microscope at 1.5ϫ magnification in both the XY (top panels) and Z planes (lower panels corresponding to the blue line), as shown in the schematic diagram to the left of the confocal images. veals considerable enrichment of endogenous spinophilin at the lateral surface of these polarized cells, corroborating initial reports of Satoh et al. (24). The overlap of expression of the ␣ 2A AR and spinophilin in the lateral domain of MDCKII cells is demonstrated by the considerable amount of yellow signal present in the red/green overlay. Similar results were observed upon colocalization of the ␣ 2B AR subtype and spinophilin (data not shown). It should be noted, however, that some spinophilin also is detected intracellularly, including in a sub-apical compartment.

Coimmunolocalization of Endogenous Spinophilin with the ␣ 2 AR Subtypes in MDCKII Cells-Spinophilin is a ubiquitously expressed multidomain protein (25) composed of an F-actin binding domain (amino acids
The 3i Loops of All Three ␣ 2 AR Subtypes Interact with Spinophilin-Smith et al. (23) have demonstrated, via yeast twohybrid screens and gel overlay strategies, that the 3i loops of the D 2 dopamine receptor (short and long forms) interact with spinophilin in the region between the F-actin binding and PP1 domains ( Fig. 2A). Consequently, we created GST fusion proteins of spinophilin bounded between amino acids 151 and 444 (GST-Sp151-444).
As shown in Fig. 2B, GST pull-down assays revealed that the 3i loop of each of the ␣ 2 AR subtypes is able to interact with GST-Sp151-444. Radiolabeled [ 35 S]Met-Gen10 ␣ 2 AR 3i loops specifically interacted with GST-Sp151-444 but not with GST alone. To further refine the interacting regions of spinophilin, the 3i loop of the ␣ 2A AR was incubated with a fusion protein of GST-spinophilin amino acids 169 -255 (Fig. 2C). This region was selected because it represents the region of least homology with the spinophilin-related protein, neurabin I, a brain-specific protein (see Fig. 2A (33)). However, because there is a stretch of 14 amino acids identical between these protein fam-ily members within this region, it was of considerable importance to evaluate the ability of the corresponding region of neurabin I (Nb146 -453) to interact with the ␣ 2A AR 3i loop. As shown in Fig. 2C, little or no interaction was detected between the ␣ 2A AR 3i loop and GST-neurabin 146 -453 when compared with GST-spinophilin 151-444. Furthermore, binding to the 3i loop is demonstrated by the more restricted region of Sp169 -255; in fact, binding of the ␣ 2A -3i loop to this region is more readily detected to this region, than to GST-Sp151-444.
Interaction of the ␣ 2A AR and Spinophilin within the Cell Is Regulated by Agonist-It was of interest to determine whether or not these ␣ 2A AR-3i loop-spinophilin interactions, detected in vitro via GST fusion protein assays, could be detected in the context of a living cell. For these studies, we transiently coexpressed cDNAs encoding full-length HA-tagged ␣ 2A AR and Myc-tagged spinophilin in CosM6 cells. On the day of analysis, cells were incubated with or without 100 M epinephrine prior to extraction of the cell membranes and solubilization with D␤M/CHS, a detergent that extracts receptor in a functional conformation (30). As shown in Fig. 3 rine, to increase the amount of ␣ 2A AR seen in association with spinophilin, suggesting that regulated interactions with spinophilin could contribute both to receptor localization and coordination of signal transduction events mediated by ␣ 2 ARs. DISCUSSION Our studies have demonstrated an interaction between the third intracellular loops of the ␣ 2 AR subtypes and spinophilin and significantly extend previous findings of in vitro studies demonstrating that the 3i loops of the D 2 dopamine receptor short and long isoforms interact with spinophilin fusion proteins (23). Furthermore, our studies have refined this interaction to amino acids 169 -255 of spinophilin (neurabin II), a region that shares little homology with what is otherwise a very homologous protein, neurabin I. Finally, our studies are the first to document G protein-coupled receptor-spinophilin interactions in the context of a native cell and to demonstrate that these interactions are fostered by agonist binding to the receptor.
A variety of novel interactions are being reported for G protein-coupled receptors via their C terminus (34 -40) or 3i loops (30,40,41). In some cases, the interactions are fostered by agonist occupancy of the receptor such as for interaction of the ␤ 2 AR with NHERF/EBP50 (35) or the somatostatin receptor with cortactin binding protein 1 (38). In some cases, interactions appear to affect receptor signaling (37,39,42), whereas in others, the interactions may be critical for receptor trafficking (43).
Interactions between ␣ 2A ARs and spinophilin in the cell should be considered in the context of interactions between the 3i loop of ␣ 2 ARs and other protein partners. Regions of the third intracellular loop of the ␣ 2A AR have been shown to interact with 14-3-3 (30), ␤Ϫarrestin (41) and heterotrimeric G proteins (44). The ␣ 2B -and ␣ 2C AR 3i loops also have been demonstrated to interact with 14-3-3 (30). These interactions with 14-3-3 are competed for by a phosphorylated peptide of raf that blocks raf-14-3-3 interactions (45), suggesting that receptor activation of downstream signaling pathways, such as the raf-ras cascade, might disrupt pre-existing ␣ 2 AR-14-3-3 interactions, or vice versa. Interactions between ␣ 2A AR and ␤-arrestin are expected to occur following agonist-evoked G protein-coupled receptor kinase-mediated ␣ 2 AR phosphorylation (10). The 3i loop sequence employed for the ␣ 2A AR in these studies includes regions that have been proposed to contribute to interactions with heterotrimeric G proteins (44), but these amphipathic helical sequences are not present in the amino acids encoded by the ␣ 2B AR and ␣ 2C AR 3i loop ligands. The ability of all three 3i loop ligands to interact with spinophilin comparably (e.g. Fig. 2) suggests that the ␣ 2 AR-spinophilin interactions can occur independent of interactions with G proteins. It also is probable that ␣ 2 AR-spinophilin interactions do not prevent interactions with G proteins, because agonist occupancy of the ␣ 2A AR increases the amount of ␣ 2A AR that coimmunoprecipitates with spinophilin. Because agonist occupancy of ␣ 2A AR also favors receptor interactions with G proteins (46), it is likely that the ␣ 2A AR can interact simultaneously with spinophilin and its cognate G i protein. What remains to be established is whether this agonist-modulated interaction with spinophilin regulates acute or tonic receptormediated signaling, by analogy with findings for the D 1 dopamine receptor (39,47,48) or mediates retention at the basolateral surface of polarized epithelial cells (Fig. 1), previously demonstrated to require the 3i loop of the AR subtypes ␣ 2A (21) and ␣ 2B (22). Lysates were prepared and precleared as described under "Experimental Procedures." Top, mouse anti-Myc antibody adsorbed to protein A-agarose was used to immunoprecipitate Myc-spinophilin. Resulting precipitates were separated via SDS-PAGE, transferred to a PVDF membrane, and blotted with rat anti-HA antibody. Bottom, the blot was stripped and reprobed with the rabbit anti-spinophilin amino acids 286 -390 antibody. B, histogram representing the -fold increase over basal levels of coimmunoprecipitated receptor detected via Western blot following epinephrine treatment. *, two-tailed p Ͻ 0.005 determined using a paired t test comparing differences in band densities before and after epinephrine treatment (n ϭ 4 independent experiments with single or duplicate immunoprecipitates).