GGA3 Interacts with a G Protein-Coupled Receptor and Modulates Its Cell Surface Export

Abstract

Molecular mechanisms governing the anterograde trafficking of nascent G protein-coupled receptors (GPCRs) are poorly understood. Here, we have studied the regulation of cell surface transport of α₂-adrenergic receptors (α₂-ARs) by GGA3 (Golgi-localized, γ-adaptin ear domain homology, ADP ribosylation factor-binding protein 3), a multidomain clathrin adaptor protein that sorts cargo proteins at the trans-Golgi network (TGN) to the endosome/lysosome pathway. By using an inducible system, we demonstrated that GGA3 knockdown significantly inhibited the cell surface expression of newly synthesized α_2B-AR without altering overall receptor synthesis and internalization. The receptors were arrested in the TGN. Furthermore, GGA3 knockdown attenuated α_2B-AR-mediated signaling, including extracellular signal-regulated kinase 1/2 (ERK1/2) activation and cyclic AMP (cAMP) inhibition. More interestingly, GGA3 physically interacted with α_2B-AR, and the interaction sites were identified as the triple Arg motif in the third intracellular loop of the receptor and the acidic motif EDWE in the VHS domain of GGA3. In contrast, α_2A-AR did not interact with GGA3 and its cell surface export and signaling were not affected by GGA3 knockdown. These data reveal a novel function of GGA3 in export trafficking of a GPCR that is mediated via a specific interaction with the receptor.

INTRODUCTION

G protein-coupled receptors (GPCRs) are the largest superfamily of cell surface receptors and modulate a variety of cell functions under physiological and pathological conditions (1, 2). The functions of GPCRs are highly regulated by elaborately coordinated intracellular trafficking processes, including anterograde export of newly synthesized receptors to the cell surface, endocytosis of the cell surface receptors into the endosomal compartment in response to ligand activation, recycling of the internalized receptors from endosomes back to the plasma membrane, and transport of receptors to lysosomes for degradation. These trafficking processes determine the numbers of receptors at the cell surface available for binding to ligands and thus dictate the magnitude of ligand-elicited cellular responses. Indeed, it has become increasingly apparent that mistrafficking of GPCRs that leads to the dysfunction of the receptors directly links to the pathogenesis of human diseases, such as nephrogenic diabetes insipidus, retinitis pigmentosa, and male pseudohermaphroditism (3–5). However, compared with well-characterized internalization, recycling, and degradation (6, 7), the molecular mechanisms underlying the cell surface transport of nascent GPCRs from the endoplasmic reticulum (ER) through the Golgi apparatus are relatively less well understood.

The Golgi/trans-Golgi network (TGN) compartment is often referred to as the “sorting center” where newly synthesized proteins are sorted to be delivered to their final cellular destinations such as endosomes, lysosomes, and the plasma membrane. Post-Golgi transport can be mediated through clathrin-coated transport vesicles, which are composed of clathrin and various adaptor proteins. Adaptor protein complexes (APs), GGAs (Golgi-associated, γ-adaptin homologous, ARF-interacting proteins), and hepatocyte growth factor receptor substrate (Hrs) are well-known adaptor proteins for clathrin-coated vesicles. There are three GGA isoforms (GGA1, GGA2, and GGA3) in humans with similar trafficking functions. All three GGAs have identical domain organizations, containing the N-terminal Vps27, Hrs, Stam (VHS) domain followed by the GAT (GGAs and TOM1) domain, the hinge region, and the C-terminal GAE (γ-adaptin ear) domain. Each domain of the GGAs has been shown to interact with specific proteins to coordinate their trafficking functions. Specifically, the N-terminal VHS domain interacts with the DXXLL-type sorting motifs of cargo proteins that cycle between the TGN and the endosomal compartment (8–17). These highly coordinated VHS-DXXLL signal interactions specifically dictate cargo proteins into the TGN-to-endosome pathway.

It has been known that APs play an important role in the endocytic trafficking and lysosomal sorting of agonist-occupied GPCRs (7, 18–20) whereas Hrs modulates the recycling process of internalized GPCRs (21, 22). However, the function of GGAs in any trafficking processes of the GPCR superfamily has not been studied. To elucidate the molecular mechanisms underlying the cell surface targeting of newly synthesized GPCRs, here we have determined the role of GGA3 in the export trafficking of α₂-adrenergic receptors (α₂-ARs), prototypic GPCRs that have three subtypes (α_2A-AR, α_2B-AR, and α_2C-AR). We have demonstrated that GGA3 is required for the TGN-to-cell surface transport of α_2B-AR, which is mediated through a specific interaction. These data have identified a novel function for GGA3 in the plasma membrane receptor transport en route from the TGN.

MATERIALS AND METHODS

Plasmid constructions.

α_2B-ARs tagged with either green fluorescent protein (GFP) at the C terminus, encoded by the pEGFP-N1 vector (α_2B-AR-GFP), or three hemagglutinins (HA) (YPYDVPDYA) at the N terminus, encoded by the pcDNA3.1(−) vector (HA-α_2B-AR), were generated as described previously (23, 24). The glutathione S-transferase (GST) fusion protein constructs comprising the first intracellular loop (ICL1, residues 44 to 53), the second one (ICL2, residues 117 to 131), and the third one (ICL3, residues 205 to 369), different lengths of ICL3 (K205 to P284, R285 to E369, R285 to C326, N327 to E369, N327 to L348, L339 to Q359, and G349 to E369), and the C terminus (residues 430 to 453) of α_2B-AR were generated using the pGEX-4T-1 vector as described previously (23, 24), and a similar strategy was used to generate the GST fusion protein construct comprising ICL3 of α_2A-AR. Full-length GGA3 tagged with myc at its N terminus was generously provided by Juan S. Bonifacino (Eunice Kennedy Shriver National Institute of Child Health and Human Development, NIH). Arrestin-3 and its dominant negative mutant arrestin-3(201–409) were obtained from Jeffrey L. Benovic (Thomas Jefferson University). The GST fusion protein construct comprising the VHS domain of GGA3 was from Addgene (Cambridge, MA) as described previously (25). To generate GFP-tagged GGA3 and its domains (VHS, residues 1 to 146; GAT, residues 147 to 313; hinge, residues 314 to 493; and GAE, residues 595 to 723), the coding sequence for each was generated by PCR and then cloned into the pEGFP-C1 vector. α_2B-AR and GGA3 mutants were generated using the QuikChange site-directed mutagenesis kit (Agilent Technologies).

Cell culture and transient transfection.

HEK293, HeLa, MCF7, and HT29 cells were cultured as described previously (26, 27). Transient transfection of cells was carried out using Lipofectamine 2000 reagent (ThermoFisher Scientific) (28). The transfection efficiency was estimated to be greater than 70% based on the GFP fluorescence.

Generation of inducible cell lines expressing α_2B-AR.

In order to study newly synthesized α_2B-AR and to characterize its cell surface transport over the time, the Tet-On 3G tetracycline-inducible gene expression system (Clontech Laboratories, Inc.) was utilized to generate stable cell lines inducibly expressing HA-α_2B-AR in HEK293 cells. Briefly, HA-α_2B-AR was cloned into the pTRE3G-TRES vector at the BglII and ClaI restriction sites and cotransfected with the PLKO.1 vector. Single-cell colonies were selected by incubation with puromycin for 2 weeks and amplified. Intact cell ligand binding assays as described below were used to detect α_2B-AR expression at the cell surface. A total of 8 cell lines inducibly expressing HA-α_2B-AR ranging from 1.3 × 10⁵ to 8.5 × 10⁵ receptors per cell were generated. The cell line expressing HA-α_2B-AR at the highest level was utilized in the current study.

Depletion of GGA3.

Short hairpin RNA (shRNA) targeting GGA3 (356 AATTCCTGTGGATAGGACGCT 376) was kindly provided by Stuart Kornfeld (Washington University School of Medicine) as described previously (29). For shRNA-mediated knockdown of GGA3, cells cultured on 6-well plates were transiently transfected with 2.0 μg of control or GGA3 shRNA for 24 h. The cells were split into 12 wells at a density of 5 × 10⁵ cells per well and cultured for an additional 24 h. For small interfering RNA (siRNA)-mediated knockdown of GGA3 in cells, two Stealth RNAi duplexes (siRNA) targeting human GGA3 (1703 TGTGACAGCCTACGATAAA 1721) as described previously (30) were synthesized. To generate siRNA-resistant GGA3, two primers (5′-TGCCCTTCCTGTGACTGCGTATGACAAGAACGGCTTCCGCATC-3′ and 5′-GATGCGGAAGCCGTTCTTGTCATACGCAGTCACAGGAAGGGCA-3′) were used in the mutagenesis reactions using GFP-GGA3 as a template. A, C, C, T, and A in positions 1709, 1712, 1715, 1718, and 1721 in the nucleotide sequence of the construct GGA3 were mutated to T, G, T, C, and G, respectively, to achieve siRNA resistance without changing the encoded amino acid sequence.

Ligand binding of intact live cells.

The cell surface expression of α₂-ARs was measured by ligand binding of intact live cells using the cell-nonpermeable radioligand [³H]RX821002 (2-methoxyidazoxan) as described previously (23, 24). Briefly, cells were incubated with Dulbecco's modified Eagle medium (DMEM) plus [³H]RX821002 (20 nM) in a total volume of 400 μl for 90 min. The nonspecific binding of α₂-ARs was determined in the presence of rauwolscine (10 μM). The binding was terminated, and excess radioligands were eliminated by washing the cells with ice-cold DMEM. The retained radioligands were then extracted by digesting the cells in 1 M NaOH for 2 h. The radioactivity was counted by liquid scintillation spectrometry. To measure the maximum number of binding sites (B_max) and equilibrium dissociation constant (K_D) values of α_2B-AR, HEK293 cells inducibly expressing HA-α_2B-AR were incubated with different concentrations of [³H]RX821002 (0.5, 1, 2, 3, 4, 5, 7.5, 10, 20, and 30 nM) after induction with doxycycline at a concentration of 40 ng/ml for 24 h. In another set of experiments, the cells cultured and incubated with doxycycline under the same conditions were used for preparation of membrane proteins without incubation with the radioligand as described previously (31). For measurement of endogenous α₂-ARs, MCF7 and HT29 cells were cultured on 6-well dishes and transfected with GGA3 shRNA or siRNA for 48 h. For measurement of α_2B-AR internalization, HEK293 cells stably expressing α_2B-AR were cultured on 6-well dishes and transfected with control or GGA3 shRNA together with 1 μg of arrestin-3 for 24 h. After starvation for 3 h, the cells were stimulated with epinephrine (100 μM). The cells were washed 3 times, and α_2B-AR cell surface expression was measured by intact cell ligand binding at 4°C.

Flow cytometric analysis of receptor expression.

Total receptor expression was measured by flow cytometry as described previously (32). Briefly, HEK293 cells expressing HA-α_2B-AR were suspended in phosphate-buffered saline (PBS) containing 1% fetal calf serum at a density of 4 × 10⁶ cells/ml and permeabilized with 0.2% Triton X-100 in PBS for 5 min on ice. The cells were then incubated with high-affinity anti-HA-fluorescein (3F10) (2 μg/ml) at 4°C for 30 min. After two washings with 0.5 ml of PBS, the fluorescence was analyzed on a flow cytometer (Dickinson FACSCalibur).

Fluorescence microscopy.

HEK293 cells were grown on coverslips precoated with poly-l-lysine in 6-well plates and transfected with 50 ng of α_2B-AR-GFP together with 400 ng of control or GGA3 shRNA. The cells were permeabilized with PBS containing 0.2% Triton X-100 for 5 min and blocked with 5% normal donkey serum for 1 h. The cells were then incubated with antibodies against p230 (1:100 dilution) for 1 h. After washing with PBS (3 times for 5 min), the cells were incubated with Alexa Fluor 594-labeled secondary antibody (1:2,000 dilution) for 1 h. Images were captured using an LSM510 Meta Zeiss confocal microscope as described previously (24, 33). To detect the cell surface expression of inducibly expressed HA-α_2B-AR, the cells were stained with anti-HA antibodies (1:500 dilution) without permeabilization. To study the subcellular localization of GGA3, HeLa cells cultured in 6-well dishes were transfected with 200 ng of GFP-tagged GGA3 and then stained with antibodies against GM130, a Golgi body marker.

Measurement of ERK1/2 activation.

HEK293 cells inducibly expressing α_2B-AR were cultured in 6-well dishes and transfected with GGA3 siRNA or shRNA for 36 h. The cells were then incubated with tetracycline at 40 ng/ml for 24 h. After starvation for at least 3 h, the cells were stimulated with UK14304 for 5 min. In some experiments, the cells stimulated with phorbol-12-myristate-13-acetate (PMA) at a concentration of 1 μM for 5 min were used as a positive control. To measure extracellular signal-regulated kinase 1/2 (ERK1/2) activation by endogenous α₂-ARs, cells were transfected with GGA3 siRNA for 36 h and then stimulated with UK14304 as above. Stimulation was terminated by addition of 1× SDS-loading buffer. ERK1/2 activation was determined by immunoblotting using phospho-specific ERK1/2 antibodies (Santa Cruz) as described previously (28).

Measurement of cAMP production.

Cyclic AMP (cAMP) concentrations were measured by using the Cyclic AMP XP Assay kit (Cell Signaling Technology) as described previously (26). Briefly, HEK293 cells inducibly expressing α_2B-AR were cultured in 6-well dishes and transfected with GGA3 siRNA or shRNA for 36 h and then incubated with tetracycline at 40 ng/ml for 24 h. The cells were then split into 96-well plates at a density of 1 × 10⁴ cells/well. After starvation for 1 h, the cells were stimulated with forskolin (1 μM) with or without different concentrations of UK14304 (10 to 1,000 nM) in the presence of 0.1 mM isobutylmethylxanthine for 5 min at 37°C. The reactions were stopped by aspirating the medium, and the cells were lysed with 100 μl of lysis buffer. Fifty microliters of cell lysate was transferred into microtiter plates, and cAMP concentrations were measured according to the protocol provided by the kit.

Protein-protein interaction assays.

For coimmunoprecipitation (co-IP), HEK293 cells inducibly expressing HA-α_2B-AR were cultured on 100-mm dishes and transfected with 10 μg of GFP-GGA3 for 24 h. After incubation with doxycycline (40 ng/ml) for 24 h, the cells were washed twice with PBS, harvested, and lysed with 1 ml of lysis buffer (50 mM Tris-HCl [pH 7.4], 150 mM NaCl, 1% Nonidet P-40, 0.01% SDS, and Complete Mini protease inhibitor mixture). After gentle rotation for 1 h, samples were centrifuged for 15 min at 14,000 × g, and the supernatant was incubated with 50 μl of Dynabeads protein G for 1 h at 4°C to remove nonspecific bound proteins. Samples were then incubated with 2 μg of anti-α_2B-AR antibodies overnight at 4°C with gentle rotation followed by incubation with 50 μl of Dynabeads protein G for 4 h. The beads were washed 3 times with lysis buffer without SDS. Immunoprecipitated proteins were eluted with 30 μl of SDS-gel loading buffer and separated by SDS-PAGE. GFP-GGA3 and HA-α_2B-AR in the IP were detected by immunoblotting using GFP and α_2B-AR antibodies, respectively. For co-IP of α_2B-AR and endogenous GGA3, two 100-mm dishes of HEK293 cells were transfected with α_2B-AR or its mutant for 48 h and co-IP assays were carried out using α_2B-AR antibodies as described above.

For GST fusion protein pulldown assays, the GST fusion proteins were expressed in bacteria and purified as described previously (23, 24). GST fusion proteins immobilized on the glutathione resin were either used immediately or stored at 4°C for no longer than 3 days. GST fusion proteins tethered to the glutathione resin were incubated with total cell lysates in 500 μl of binding buffer (20 mM Tris-HCl [pH 7.5], 1% NP-40, 140 mM NaCl, and 1 mM MgCl₂) at 4°C for 4 to 6 h. The resin was washed 4 times with 0.5 ml of binding buffer, and the retained proteins were solubilized in SDS-gel loading buffer and separated by SDS-PAGE. Proteins bound to GST fusion proteins were detected by immunoblotting.

For GGA3 interaction with peptide-conjugated agarose beads, the peptide NH2-GKNVGVASGQWWRRRTQLSRE-COOH derived from the α_2B-AR ICL3 and its mutant NH2-GKNVGVASGQWWAAATQLSRE-COOH, in which RRR were mutated to AAA, were synthesized, purified by high-performance liquid chromatography (HPLC) to >75%, and directly conjugated to agarose beads by Biosynthesis Inc. (Lewisville, TX). The peptide-conjugated agarose beads (10 μl, approximately 13 μmol peptides) were incubated with 500 μg of cell lysate prepared from HEK293 cells transfected with the GFP-tagged GGA3 VHS domain in 500 μl of binding buffer (20 mM Tris-HCl [pH 7.4], 2% NP-40, 140 mM NaCl) at 4°C overnight. The resin was washed 4 times, and the retained proteins were solubilized in 1× SDS gel loading buffer and separated by SDS-PAGE. Bound proteins were detected by immunoblotting using anti-GFP antibodies. To study the direct interaction between the α_2B-AR ICL3 and the GGA3 VHS domain, the VHS domain and its mutant were generated as GST fusion proteins, eluted from the glutathione beads, and incubated with ICL3 peptide-conjugated agarose as described above. Bound proteins were detected by immunoblotting using anti-GST antibodies.

Statistical analysis.

Differences were evaluated using Student's t test, and P values of <0.05 were considered statistically significant. Data are expressed as the means ± standard errors (SE).

RESULTS

Characterization of cell lines inducibly expressing α_2B-AR.

The stable HEK293 cells generated by using the Tet-On 3G inducible expression system to drive the expression of HA-α_2B-AR were incubated with increasing concentrations of doxycycline for 24 h or incubated with doxycycline at a concentration of 40 ng/ml for different time periods, and the numbers of α_2B-AR at the cell surface were determined by intact cell ligand binding. Doxycycline dose-dependent and incubation time-dependent expression of α_2B-AR are shown in Fig. 1A and B, respectively. α_2B-AR expression at the cell surface was clearly detectable after 6 h induction with doxycycline at the saturating concentration of 40 ng/ml and reached a plateau after 20 h of induction. The time required to achieve 50% of the maximal receptor expression at the cell surface (t_1/2) was approximately 10.6 h after induction. Consistent with ligand binding data, robust expression of α_2B-AR at the cell surface was observed by confocal microscopy following staining with anti-HA antibodies in nonpermeabilized cells after treatment with doxycycline whereas α_2B-AR expression was undetectable in cells that had not been incubated with doxycycline (Fig. 1C). Doxycycline-induced expression of HA-α_2B-AR was further confirmed by immunoblotting using α_2B-AR antibodies (Fig. 1D).

FIG 1.

Inducible expression of α_2B-AR in HEK293 cells. (A) Doxycycline dose-dependent induction of cell surface α_2B-AR expression. The cells were incubated with different concentrations of doxycycline for 24 h, and the cell surface α_2B-AR expression was quantified by intact cell ligand binding using [³H]RX821002 at 20 nM. The data shown are percentages of specific binding obtained from cells treated with doxycycline (100 ng/ml), in which the mean value of specific [³H]RX821002 binding was 35,672 ± 797 cpm per well (n = 3). (B) Doxycycline time-dependent induction of cell surface α_2B-AR expression. HEK293 cells were incubated with doxycycline (40 ng/ml) for different time periods. The data shown are percentages of specific binding obtained from cells after induction for 36 h, in which the mean value of specific ligand binding was 36,123 ± 573 cpm per well (n = 3). (C) Detection of cell surface HA-α_2B-AR expression by confocal microscopy. HEK293 cells were incubated with or without doxycycline (40 ng/ml) for 24 h and stained with anti-HA antibodies in nonpermeabilized cells. Green, HA-α_2B-AR; blue, DNA staining by 4′,6-diamidino-2-phenylindole (DAPI). Bar, 10 μm. (D) Detection of HA-α_2B-AR expression by immunoblotting. HEK293 cells were incubated with or without doxycycline (40 ng/ml) for 24 h. Total cell lysates (50 μg) were separated by SDS-PAGE, and HA-α_2B-AR expression was measured by immunoblotting using α_2B-AR antibodies. Similar results were obtained in 3 experiments.

Effect of depleting GGA3 on the cell surface transport of α₂-ARs.

We then determined the effect of short hairpin RNA (shRNA)-mediated depletion of endogenous GGA3 on the cell surface expression of inducibly expressed α_2B-AR. The introduction of GGA3 shRNA markedly knocked down GGA3 (by 92%) compared with cells transfected with control shRNA (see Fig. S1A in the supplemental material). shRNA-mediated knockdown of GGA3 moderately but significantly attenuated the magnitude of α_2B-AR expression at the cell surface after doxycycline induction for more than 12 h (Fig. 2A). The maximal inhibition (approximately 30%) was observed after doxycycline induction for more than 20 h, and t_1/2 values were 10.8 and 10.9 h in cells transfected with control and GGA3 shRNA, respectively (Fig. 2A). The radioligand saturation binding curves showed that GGA3 knockdown significantly reduced the B_max but not K_D of α_2B-AR (Fig. 2B).

FIG 2.

Effects of GGA3 knockdown on the cell surface expression of α₂-ARs. (A) Effect of shRNA-mediated GGA3 knockdown on cell surface α_2B-AR expression. HEK293 cells inducibly expressing α_2B-AR were transfected with control or GGA3 shRNA and incubated with doxycycline (40 ng/ml) for different time periods. The cell surface α_2B-AR expression was determined by intact cell ligand binding using [³H]RX821002 at 20 nM. The data shown are percentages of specific binding obtained from cells transfected with control shRNA and treated with doxycycline for 24 h, in which the mean value of specific ligand binding was 34,408 ± 552 cpm per well (n = 3). (B) Effect of shRNA-mediated GGA3 knockdown on the B_max and K_D values of α_2B-AR. HEK293 cells transfected and treated with doxycycline for 24 h as described above were incubated with different concentrations of [³H]RX821002. In separate experiments, the cells treated under the same conditions were used for membrane protein preparation. The B_max values were 24.6 and 16.5 pmol/mg protein in control and GGA3 knockdown cells, respectively (P < 0.05, n = 3), whereas the K_D values in control and GGA3 knockdown cells were the same (4.1 nM). (C) Effects of siRNA-mediated GGA3 knockdown on the cell surface expression of different α₂-ARs. HEK293 cells inducibly expressing α_2B-AR were transfected with control siRNA, GGA3 siRNA, or GGA3 siRNA plus siRNA-resistant GFP-GGA3 (left bars). To determine the effect of GGA3 knockdown on the cell surface transport of α_2A-AR and α_2C-AR, HEK293 cells were cotransfected with GGA3 siRNA together with α_2A-AR (middle bars) or α_2C-AR (right bars) (n = 3 to 5). (D) Effect of shRNA- and siRNA-mediated knockdown of GGA3 on total α_2B-AR expression measured by flow cytometry following staining with HA antibodies in permeabilized cells (n = 4). (E) Effect of GGA3 knockdown on total HA-α_2B-AR expression measured by immunoblotting using HA antibodies. (F) Effect of depleting GGA3 on α_2B-AR internalization. HEK293 cells stably expressing α_2B-AR were transfected with GGA3 shRNA and then stimulated with epinephrine (100 μM) (n = 3). (G) Effect of GGA3 knockdown on the cell surface expression of endogenous α₂-ARs in MCF7 and HT29 cells measured by intact cell ligand binding. The mean values of specific ligand binding were 462 ± 56 cpm in MCF7 cells transfected with control shRNA, 478 ± 49 cpm in MCF7 cells transfected with control siRNA, and 452 ± 43 in HT29 cells transfected with control siRNA (n = 3). *, P < 0.05 versus the respective control.

GGA3 is a TGN-localized protein, and shRNA-mediated knockdown of GGA3 was shown to disrupt the localization of β-GalT, a trans-Golgi marker, in HeLa cells (29). To determine if the effect of shRNA-mediated GGA3 knockdown on the cell surface expression of α_2B-AR was induced by disruption of the TGN/Golgi body integrity, which may produce nonspecific effects on global protein transport, we determined the effect of GGA3 knockdown by siRNA that were shown to have no effect on the TGN structure (30). siRNA-mediated knockdown of GGA3 inhibited the cell surface expression of inducibly expressed α_2B-AR, and this inhibition was completely reversed by expression of an siRNA-resistant form of GGA3 (Fig. 2C; see also Fig. S1B in the supplemental material). Similar to α_2B-AR, the cell surface expression of α_2C-AR was also significantly attenuated, whereas the transport of α_2A-AR was not affected by GGA3 siRNA in HEK293 cells transiently transfected with the receptors (Fig. 2C). In contrast, GGA3 knockdown did not affect the overall synthesis of α_2B-AR as measured by flow cytometry (Fig. 2D) and by immunoblotting (Fig. 2E). Furthermore, GGA3 overexpression did not influence the cell surface expression of α_2B-AR (see Fig. S2A in the supplemental material), suggesting that expression of endogenous GGA3 is not a rate-limiting factor for the cell surface expression of α_2B-AR.

To eliminate the possibility that the reduction of the cell surface expression of α_2B-AR caused by GGA3 knockdown was induced by enhanced constitutive internalization of the receptor, we determined if expression of dominant negative mutant of arrestin-3, which has been shown to control the internalization process of α_2B-AR (34), could reverse the inhibitory effects. Expression of the dominant negative mutant arrestin-3(201–409) did not influence the cell surface expression of α_2B-AR in cells expressing control or GGA3 shRNA (see Fig. S2B in the supplemental material). Furthermore, GGA3 shRNA also did not influence the internalization of α_2B-AR in response to epinephrine stimulation (Fig. 2F). These data suggest that GGA3 does not play a major role in the internalization of α_2B-AR.

We next determined the role of GGA3 in the cell surface transport of endogenous α₂-ARs in breast cancer MCF7 cells, which express α_2B-AR and α_2C-AR (35), and in colon cancer HT29 cells, which express only α_2A-AR (36). Similar to their inhibitory effects on inducibly expressed α_2B-AR, shRNA- and siRNA-mediated knockdown of GGA3 decreased the cell surface expression of α₂-ARs in MCF7 cells by about 40% (Fig. 2G, left panel). In contrast, GGA3 siRNA did not alter the cell surface expression of α_2A-AR in HT29 cells (Fig. 2G, right panel). Altogether, these results demonstrate that the normal function of GGA3 is required for the cell surface export of exogenously transfected and endogenous α_2B-AR and α_2C-AR but not α_2A-AR.

Effect of GGA3 knockdown on α₂-AR-mediated signaling.

All three α₂-ARs couple to the Gi/Go family G proteins, and their activation has been well shown to stimulate the mitogen-activated protein kinases (MAPK) ERK1/2, inhibit adenylyl cyclases, and suppress voltage-gated calcium channels (24, 26, 37–39). To investigate if GGA3 knockdown-induced reduction of cell surface α_2B-AR expression could result in a concomitant defective signaling, the activation of ERK1/2 and the reduction of cAMP production were chosen as functional readouts. Consistent with their abilities to inhibit receptor cell surface transport, transient expression of siRNA targeting GGA3 significantly reduced α_2B-AR-mediated ERK1/2 activation in response to UK14304 stimulation in cells inducibly expressing α_2B-AR compared to cells transfected with control siRNA (Fig. 3A and B). Similarly, siRNA-mediated depletion of GGA3 (see Fig. S1D in the supplemental material) inhibited the activation of ERK1/2 by endogenous α₂-ARs by 48% in MCF7 cells (Fig. 3C and D). In contrast, GGA3 siRNA did not affect the activation of ERK1/2 by exogenously transfected α_2A-AR in HEK293 cells and by endogenous α_2A-AR in HT29 cells (see Fig. S1C and E and S3 in the supplemental material). Furthermore, siRNA-mediated GGA3 knockdown reduced the ability of α_2B-AR activation to inhibit the cAMP production in response to stimulation with forskolin (Fig. 3E). Similar to GGA3 knockdown by siRNA, shRNA-mediated GGA3 depletion significantly attenuated ERK1/2 activation and cAMP reduction in response to UK14304 stimulation in HEK293 cells inducibly expressing α_2B-AR (see Fig. S4 in the supplemental material). These data suggest that GGA3 modulates not only the cell surface transport but also the function of α_2B-AR.

FIG 3.

Inhibition of α_2B-AR-mediated signaling by depleting GGA3. (A) Effect of GGA3 knockdown on α_2B-AR-mediated ERK1/2 activation. HEK293 cells inducibly expressing α_2B-AR were transfected with control or GGA3 siRNA and incubated with doxycycline at 40 ng/ml for 24 h. After starvation for 3 h, the cells were stimulated with different concentrations of UK14304 (0.01, 0.1, and 1 μM) for 5 min. The simulation with PMA at 1 μM for 5 min was used as a positive control. (B) Quantitation of data shown in panel A (n = 4). (C) Effect of GGA3 knockdown on ERK1/2 activation by endogenous α₂-ARs. MCF7 cells were transfected with control or GGA3 siRNA for 48 h and stimulated with UK14304 at 1 μM for 5 min. (D) Quantitation of data shown in panel C (n = 3). (E) Effect of GGA3 knockdown on α_2B-AR-mediated inhibition of cAMP production. HEK293 cells inducibly expressing α_2B-AR were transfected, incubated with doxycycline, and stimulated with forskolin (1 μM) plus different concentrations of UK14304 for 5 min at 37°C (n = 3). The data shown are percentages of the mean values obtained from cells transfected with control siRNA and stimulated with forskolin alone. *, P < 0.05 versus the respective control.

GGA3 regulates α_2B-AR transport from the TGN.

We next determined the effect of GGA3 knockdown on the subcellular distribution of α_2B-AR in HEK293 cells by confocal microscopy. As expected, α_2B-AR was robustly expressed at the cell surface in cells transfected with control shRNA. In contrast, α_2B-AR was clearly arrested in the intracellular compartments, unable to transport to the cell surface, in GGA3 knockdown cells (Fig. 4A).

FIG 4.

Effect of GGA3 knockdown on subcellular localization of α_2B-AR. (A) HEK293 cells were transiently transfected with α_2B-AR-GFP together with control or GGA3 shRNA for 48 h. (B) Colocalization of α_2B-AR with p230. HEK293 cells were transfected with α_2B-AR-GFP together with control or GGA3 shRNA and then stained with anti-p230 antibodies. The images are representative of at least 3 independent experiments. Green, α_2B-AR-GFP; red, p230; blue, DNA staining by DAPI (nuclei); yellow, colocalization of α_2B-AR-GFP with p230. Bars, 10 μm. Similar results were obtained in 3 to 5 experiments.

To define the intracellular compartment in which GGA3 regulates α_2B-AR transport, α_2B-AR was colocalized with different intracellular organelle markers in cells expressing GGA3 shRNA. The intracellularly accumulated α_2B-AR was extensively colocalized with the TGN marker p230 (Fig. 4B) but not with the ER marker DsRed2-ER (see Fig. S5 in the supplemental material) in HEK293 cells transfected with GGA3 shRNA. These data suggest that GGA3 likely controls the cell surface transport of α_2B-AR en route from the TGN.

GGA3 interacts with the ICL3 of α_2B-AR.

It has been well defined that the trafficking function of GGA proteins is mediated through their direct interactions with cargo proteins (8–17). To elucidate the possible molecular mechanisms underlying the function of GGA3 in α_2B-AR export, we determined if GGA3 and α_2B-AR could physically associate to form a complex in co-IP assays using α_2B-AR antibodies. GGA3 was clearly detected in the IP of α_2B-AR antibodies (Fig. 5A).

FIG 5.

GGA3 interaction with α_2B-AR and its ICL3. (A) Interaction of α_2B-AR with GGA3 as measured by co-IP assays. HEK293 cells stably expressing HA-α_2B-AR were transfected with pEGFP-C1 (GFP) or GFP-tagged GGA3. Normal HEK293 cells transfected with GFP-tagged GGA3 were also used as a negative control (Ctrl). The receptors were immunoprecipitated with α_2B-AR antibodies. The amounts of GGA3 (upper panel) and α_2B-AR (lower panel) were determined by immunoblotting using GFP and α_2B-AR antibodies, respectively. Lysate, 1% of total input. (B) Sequences of ICL1, ICL2, ICL3, and C terminus (CT) of α_2B-AR. (C) Interaction of α_2B-AR intracellular domains with GGA3. Myc-GGA3 was expressed in HEK293 cells, and total cell homogenates were incubated with GST fusion proteins. Bound GGA3 was revealed by immunoblotting using myc antibodies (upper panel). Lysate, 5% of total input. (D) Interaction of the ICL3 of α_2A-AR and α_2B-AR with GGA3 in GST fusion protein pulldown assays as described in panel C. Lysate, 5% of total input. Lower images in panels C and D show Coomassie blue staining of purified GST fusion proteins. Similar results were obtained in at least 3 experiments.

All three α₂-ARs have similar structural features: the third intracellular loop (ICL3) is quite large, with more than 170 amino acid residues, whereas other loops and the termini are relatively short, with <25 residues (Fig. 5B). To identify the intracellular domain that mediated α_2B-AR interaction with GGA3, ICL1, ICL2, ICL3, and the C terminus of α_2B-AR were generated as GST fusion proteins and then incubated with total lysates prepared from cells transiently transfected with myc-GGA3. The GST fusion proteins of the ICL3 but not the ICL1, the ICL2, or the C terminus were capable of interacting with GGA3 (Fig. 5C).

As our preceding data have demonstrated that GGA3 modulates the cell surface transport of α_2B-AR but not α_2A-AR, we next compared the GGA3 interaction with the ICL3 of α_2A-AR and α_2B-AR in GST fusion protein pulldown assays. In contrast to α_2B-AR, the α_2A-AR ICL3 did not interact with GGA3 (Fig. 5D).

α_2B-AR interacts with the VHS domain of GGA3.

To search for the domain of GGA3 responsible for binding to α_2B-AR, the VHS, GAT, hinge, and GAE domains of GGA3 were generated as GFP fusion proteins (Fig. 6A). We first visualized their subcellular distribution by confocal microscopy. Consistent with other reports (14, 25), full-length GGA3 was extensively expressed in the Golgi network. Similarly, the GAT domain was also mainly localized to the Golgi network. The hinge, the VHS, and the GAE domains of GGA3 were largely expressed in the cytoplasm. In addition, the VHS and the GAE domains were seemingly also expressed in the nuclear compartment (Fig. 6B).

FIG 6.

Identification of the α_2B-AR-binding domain of GGA3. (A) Diagram showing the generation of GFP-tagged GGA3 and its domains. (B) Subcellular distribution of GGA3 domains. HeLa cells were transfected and stained with antibodies against the Golgi marker GM130. Green, GFP-GGA3; red, GM130; blue, DNA staining by DAPI (nuclei). Bar, 10 μm. (C) Interaction of the α_2B-AR ICL3 with different GGA3 domains. The GFP-tagged VHS, GAT, hinge, and GAE domains of GGA3 were expressed in HEK293 cells. Total cell lysates were incubated with GST-ICL3 fusion proteins. Bound GGA3 domains were revealed by immunoblotting using GFP antibodies. Total cell lysates expressing GFP alone were used as a control. (D) α_2B-AR interaction with the GGA3 VHS domain. The GGA3 VHS domain was generated as GST fusion proteins and incubated with total cell lysates prepared from HEK293 cells expressing HA-α_2B-AR. Bound HA-α_2B-AR was detected by immunoblotting using HA antibodies. Lysate, 5% of total input. Similar results were obtained in at least 3 experiments.

We next determined the interaction of individual domains of GGA3 with GST-ICL3 fusion proteins in GST fusion protein pulldown assays. The GGA3 VHS domain, but not the GAT, hinge, or GAE domain, bound to the ICL3 of α_2B-AR (Fig. 6C).

We then sought to define if the VHS domain of GGA3 was able to interact with full-length α_2B-AR. In this experiment, the VHS domain of GGA3 was generated as GST fusion proteins and its interaction with full-length HA-α_2B-AR was determined in GST fusion protein pulldown assays. GST-VHS fusion proteins, but not GST alone, strongly interacted with HA-α_2B-AR (Fig. 6D). These data demonstrate that α_2B-AR interacts with the VHS domain of GGA3.

Identification of the GGA3-binding site in the α_2B-AR ICL3.

The progressive deletion strategy was utilized to further identify the GGA3-binding domain in the α_2B-AR ICL3 in GST fusion protein pulldown assays. ICL3 was first split into the N-terminal half K205-P284 and the C-terminal half R285-E369. Similar to the entire ICL3, the C-terminal portion R285-E369 strongly interacted with the GGA3 VHS domain, whereas the N-terminal portion K205-P284 did not (Fig. 7A and B). Furthermore, the C-terminal half, but not the N-terminal half, was confirmed to interact with full-length GGA3 (Fig. 7C). A further split of the C-terminal portion revealed that N327-E369, but not R285-C326, interacted with the GGA3 VHS domain. Moreover, the GST fusion proteins with the fragment G349-E369, but not N327-L348 and L339-Q358, interacted with the GGA3 VHS domain (Fig. 7A and B). These data demonstrate that the GGA3-binding site localizes within the G349-E369 domain (Fig. 7A and B).

FIG 7.

Mapping of the GGA3-binding domain in the α_2B-AR ICL3 by progressive deletion. (A) Representative blot showing the interaction of different ICL3 fragments with the GGA3 VHS domain. (B) Summary of progressive deletion to identify the GGA3-binding domain in the ICL3 of α_2B-AR as shown in panel A. (C) Interaction of the N- and the C-terminal portions of the α_2B-AR ICL3 with full-length GGA3 as measured in GST fusion protein pulldown assays. Similar results were obtained in 3 experiments.

We have previously demonstrated that the triple Arg (3R) motif at positions 361 to 363 in the ICL3 mediates the α_2B-AR interaction with the COPII vesicle components Sec24C/D and modulates receptor export from the ER (23). To determine if the 3R motif also mediated the α_2B-AR interaction with GGA3, we determined the effect of mutating the 3R motif on the ICL3 interaction with the GGA3 VHS domain. The mutation of three Arg to three Ala (3A) significantly reduced (by 88%) the interaction of the ICL3 fragment G348-E369 with the GGA3 VHS domain (Fig. 8A and B). Similar to the mutation to 3A, the 3R mutation to three noncharged Gln (3Q) or three negatively charged Glu (3E) completely blocked the ICL3 interaction with the GGA3 VHS domain, whereas mutation of 3R to three positively charged Lys (3K) retained the interaction of ICL3 with GGA3 (Fig. 8C and D). These data demonstrate that the positively charged property of the 3R motif is a major determinant for interaction with GGA3.

FIG 8.

Identification of the triple Arg motif as the GGA3-binding site. (A) Effect of mutating 3R on the interaction of G349-E369 with the GGA3 VHS domain. G349-E369 and its 3R-3A mutant were generated as GST fusion proteins, and their interaction with the GGA3 VHS domain was determined in GST fusion protein pulldown assays. (B) Quantitation of data shown in panel A. (C) Effect of mutating 3R to 3A, 3K, 3Q, and 3E on the ICL3 interaction with the GGA3 VHS domain. (D) Quantitation of data shown in panel C. (E) Interaction of ICL3-conjugated agarose beads with the GGA3 VHS domain. G349-E369 and its mutant 3R-3A were synthesized and conjugated to agarose and then incubated with total cell homogenates prepared from cells expressing GFP-VHS. The bound VHS domain was revealed by immunoblotting using GFP antibodies. (F) Quantitation of data shown in panel E. (G) Interaction of α_2B-AR with endogenous GGA3 as measured by co-IP assays. HEK293 cells were transiently transfected with α_2B-AR or its mutant 3R-3A. HEK293 cells without transfection were used a negative control (Ctrl). The receptors were immunoprecipitated with α_2B-AR antibodies. The amounts of GGA3 (upper panel) and α_2B-AR (lower panel) were determined by immunoblotting using GGA3 and α_2B-AR antibodies, respectively. Lysate, 2% of total input. (H) Quantitation of data shown in panel G. *, P < 0.05 versus WT (n = 3 or 4).

To exclude the possible interference of GST on the ICL3 interaction with GGA3 and further characterize the interaction, a 21-residue peptide of the ICL3 (from G349 to E369) containing the 3R motif and a mutated peptide in which the 3R motif was mutated to 3A were synthesized and conjugated to agarose. Peptide-conjugated agarose beads were incubated with total cell lysates expressing GFP-VHS. Similar to the results obtained from GST fusion protein pulldown assays, the ICL3 peptide-conjugated agarose beads, but not control agarose beads, strongly bound to the VHS domain, whereas the mutated peptide-conjugated beads only very weakly bound to the VHS domain (Fig. 8E and F).

To determine if α_2B-AR is able to interact with endogenous GGA3 and if the interaction is thus mediated through the 3R motif, we compared the interaction of α_2B-AR and its 3R-3A mutant with endogenous GGA3 in co-IP assays. GGA3 was detected in the IP of α_2B-AR antibodies in cells expressing wild-type α_2B-AR but not in cells expressing the mutant 3R-3A (Fig. 8G). These data indicate that the 3R motif mediates α_2B-AR interaction with GGA3.

Identification of the α_2B-AR-binding motif in the VHS domain of GGA3.

To define the α_2B-AR-binding site in the GGA3 VHS domain, we focused on the highly negatively charged region EDWE in the VHS domain (Fig. 9A). We first determined the effect of mutating EDWE motif to AAWA on the interaction of the GGA3 VHS domain with the α_2B-AR ICL3 in GST fusion protein pulldown assays. Consistent with our previous data, mutation of the 3R motif markedly reduced the interaction of the ICL3 with the GGA3 VHS domain. Mutation of the motif EDWE to AAWA abolished the interaction of the GGA3 VHS domain with both wild-type ICL3 and the 3R-3A mutant (Fig. 9B and C). These data demonstrate that the acidic EDXE motif in the GGA3 VHS domain represents a specific α_2B-AR-binding site.

FIG 9.

Effect of mutating the acidic motif EDWE on GGA3 interaction with α_2B-AR. (A) Sequence of the GGA3 VHS domain. Acidic residues in the motif EDWE that were mutated to Ala (AAWA) are bolded. (B) Effect of mutating EDWE on the interaction of the GGA3 VHS domain with ICL3. ICL3 and its mutant 3R-3A were generated as GST fusion proteins, and the GGA3 VHS domain and its mutant AAWA were tagged with GFP. Their interactions were determined in GST fusion protein pulldown assays. The bound VHS domain was revealed by immunoblotting using GFP antibodies. Lysate, 5% of total input. (C) Quantitation of data shown in panel B (n = 3). (D) Motif-mediated direct interaction between α_2B-AR and GGA3. The GGA3 VHS domain and its mutant were generated as GST fusion proteins, eluted, and incubated with agarose beads conjugated with ICL3 or its mutant 3R-3A as described in the legend to Fig. 8E. Bound VHS domain was revealed by immunoblotting using GST antibodies. Input, 5% of total input. (E) Quantitation of data shown in panel D (n = 3). *, P < 0.05 versus the interaction between ICL3 and VHS. (F) Diagram showing the function of GGA3 in regulating α_2B-AR plasma membrane transport, which is likely mediated through an ionic interaction between the negatively charged EDWE motif in the VHS domain of GGA3 and the positively charged 3R motif in ICL3 of α_2B-AR.

To further determine if the interaction between α_2B-AR and GGA3 is direct, the GGA3 VHS domain and its mutant EDWE-AAWA were purified as GST fusion proteins and then incubated with the ICL3 peptide-conjugated agarose beads as described above (Fig. 8E). The wild-type VHS domain interacted with the wild-type ICL3 peptide, and mutation of either 3R or EDWE abolished the interaction (Fig. 9D and E). These data suggest that the interaction between α_2B-AR and GGA3 is direct and mediated by specific motifs.

DISCUSSION

The most important finding presented here is that we have identified a novel function for GGA3 as an important modulator in the cell surface transport of α_2B-AR, a prototypic GPCR, from the TGN and that the function of GGA3 in the trafficking of α_2B-AR is likely mediated through a specific interaction (Fig. 9F).

The GGA family proteins have been well defined to function as adaptors for clathrin-coated vesicles to specifically mediate the transport from the TGN to the endosomal compartment. Our studies have demonstrated that GGA3 plays an important role in the export trafficking of α_2B-AR. First, depleting GGA3 by shRNA and siRNA significantly reduced the cell surface expression of α_2B-AR as quantified by ligand binding of intact live cells. As ligand binding assays were carried out to measure the numbers of inducibly expressed α_2B-AR at the cell surface using the antagonist RX821002 and GGA3 knockdown did not alter the internalization of α_2B-AR, the reduction of cell surface receptors caused by GGA3 knockdown reflects the defective export of newly synthesized and fully maturated receptors rather than an augmented internalization of the receptors from the cell surface. In addition, the cell surface expression of endogenous α₂-ARs in MCF7 cells was also inhibited by GGA3 knockdown. However, the inhibitory effects caused by depleting GGA3 were moderate (inhibiting by less than 40%), which is well consistent with other reports demonstrating that depleting a specific component of the transport machinery produced only mild inhibition on the cell surface transport of GPCRs (23, 24). The simplest explanation for this could be that there are multiple pathways to direct α_2B-AR export from the Golgi body to the cell surface and GGA3 regulates only one of these pathways that mediate α_2B-AR forward transport. We have also shown that, similar to α_2B-AR, the cell surface transport of α_2C-AR was inhibited by GGA3 knockdown. Surprisingly, α_2A-AR cell surface expression was not affected by GGA3, suggesting that GGA3 has selectivity for different α₂-ARs.

Second, GGA3 knockdown induced an extensive accumulation of α_2B-AR in the TGN compartment. These data imply that GGA3 is involved in the plasma membrane receptor transport en route from the TGN. These data are also consistent with the Golgi body/TGN localization of GGA3 as well as its well-established function in post-Golgi transport. Our previous studies have shown that α_2B-AR exit from the Golgi body is dictated by the N-terminal YS motif and its transport from the Golgi body to the cell surface is regulated by Ras-like small GTPases, including Rab8 (33), Rab26 (24), and ARF1 (40), suggesting that multiple regulatory proteins are involved in the post-Golgi traffic of α_2B-AR. These results, together with other studies (41, 42), have demonstrated that GPCR export from the Golgi body to the cell surface is a highly regulated and dynamic process.

Third, GGA3 knockdown-induced reduction of α_2B-AR transport to the cell surface was in parallel with attenuated receptor signaling measured as ERK1/2 activation and cAMP reduction in response to stimulation with the agonist UK14304. Furthermore, we have shown that GGA3 knockdown, which did not influence α_2A-AR cell surface transport, did not affect α_2A-AR-mediated ERK1/2 activation. In addition, GGA3 knockdown had no effect on forskolin-stimulated cAMP production in the absence of UK14304. These data strongly suggest that the inhibition of α_2B-AR-mediated signaling is most likely caused by the decrease of receptor transport to the cell surface.

It has been well described that the function of GGAs in sorting proteins into the TGN-to-endosome pathway is tightly controlled by the interaction, via their VHS domains, with the DXXLL-type motifs presented in cargos. These cargo proteins include cation-dependent and cation-independent mannose 6-phosphate receptors (9, 11, 14, 15, 17), sortilin (13, 16), sorting-protein-related receptor (12, 43), low-density lipoprotein receptor-related proteins, and β-secretase (8, 10, 16). We have demonstrated that GGA3 and α_2B-AR physically associated in co-IP and GST fusion protein pulldown assays. Furthermore, consistent with its function in regulating protein transport, the GGA3 VHS domain is responsible for the interaction with α_2B-AR. We have further used the progressive deletion and mutagenesis strategies to successfully identify their interaction sites. We have found that mutation of the 3R motif profoundly reduced α_2B-AR interaction with GGA3, specifically the VHS domain. These data suggest that the 3R motif is a novel GGA3-binding signal that controls, at least in part, α_2B-AR interaction with GGA3. We have also identified the acidic motif EDWE in the GGA3 VHS domain, which mediates the GGA3 interaction with α_2B-AR, indicating that the interaction between α_2B-AR and GGA3 is ionic in nature. These data also suggest that regulation of α_2B-AR trafficking by GGA3 is highly specific. To the best of our knowledge, α_2B-AR is the only cargo molecule identified thus far that interacts with GGA3 through a charged motif. It is possible that GGA3 interaction with the 3R motif of α_2B-AR directs the receptor into the plasma membrane transport pathway (Fig. 9F), whereas its interaction with cargo proteins bearing the DXXLL-type signals controls the TGN-to-endosome transport pathway. This possibility is also supported by the fact that the cell surface transport of α_2C-AR, which has the RRK motif, is regulated by GGA3, whereas the cell surface transport of α_2A-AR, which does not have the triple basic motif (but has the sequence RWR instead), is not mediated through GGA3. It is also interesting that the triple basic motif is highly conserved in the ICL3 of many GPCRs, such as muscarinic receptors (subtypes 1, 2, 3, and 5) and serotonin receptors (subtypes 1A, 1B, 1D, and 2C), as described in our previous publication (23). Therefore, GGA3 may regulate the cell surface transport of these GPCRs.

We have previously demonstrated that the 3R motif in the ICL3 mediates α_2B-AR interaction with Sec24C/D, the components of ER-derived COPII transport vesicles, and that mutation of this motif reduces the ER export and the cell surface transport of the receptor (23). These data also imply that the cargo α_2B-AR may use the same motif to physically associate with distinct transport machineries to direct its export trafficking at different transport steps. It is possible that interaction of the 3R motif with Sec24 modulates the exit from the ER as well as the transport of newly synthesized α_2B-AR from the ER to the Golgi body, whereas the 3R motif interaction with GGA3 regulates the post-Golgi export trafficking. In agreement with this possibility, the same di-acidic motif has been demonstrated to interact with Sec24 and AP3 to control vesicular stomatitis virus protein G (VSVG) transport from the ER and the Golgi body, respectively (44, 45). Nevertheless, this study, together with many other reports, indicates that GPCRs may directly interact with the transport machineries to control their export trafficking (23, 24, 46).

Our data presented in this paper clearly reveal a role of GGA3 in the cell surface export of α_2B-AR and represent the first demonstration of the functional importance of the GGA family proteins in GPCR trafficking. However, there are many unanswered questions related to the regulation of GPCR transport by GGAs, such as the following. (i) Does GGA3 modulate the cell surface transport of many other GPCRs? (ii) Are other GGA family proteins (i.e., GGA1 and GGA2) involved in the transport of α_2B-AR or other GPCRs? (iii) If so, what molecular mechanisms are used? To address these interesting questions will significantly enhance our understanding of the GGA family protein function in vesicle-mediated GPCR trafficking. As great progress has been achieved over the past several years in defining the roles of transport machinery, regulatory proteins, and trafficking motifs in the cell surface movement of GPCRs en route from the ER and the Golgi body (47–56), further elucidation of the export mechanisms of GPCRs may reveal novel therapeutic targets for effective therapy of human diseases, involving abnormal trafficking of GPCRs.

Funding Statement

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.