Di-acidic Motifs in the Membrane-distal C Termini Modulate the Transport of Angiotensin II Receptors from the Endoplasmic Reticulum to the Cell Surface*

The molecular mechanisms underlying the endoplasmic reticulum (ER) export and cell surface transport of nascent G protein-coupled receptors (GPCRs) have just begun to be revealed and previous studies have shown that hydrophobic motifs in the putative amphipathic 8th α-helical region within the membrane-proximal C termini play an important role. In this study, we demonstrate that di-acidic motifs in the membrane-distal, nonstructural C-terminal portions are required for the exit from the ER and transport to the plasma membrane of angiotensin II receptors, but not adrenergic receptors. More interestingly, distinct di-acidic motifs dictate optimal export trafficking of different angiotensin II receptors and export ability of each acidic residue in the di-acidic motifs cannot be fully substituted by other acidic residue. Moreover, the function of the di-acidic motifs is likely mediated through facilitating the recruitment of the receptors onto the ER-derived COPII transport vesicles. Therefore, the di-acidic motifs located in the membrane-distal C termini may represent the first linear motifs which recruit selective GPCRs onto the COPII vesicles to control their export from the ER.

Recent studies have demonstrated that protein export from the endoplasmic reticulum (ER) 2 is a selective process that is mediated through COPII-coated transport vesicles and dictated by short, linear sequences called ER export motifs (1)(2)(3)(4)(5)(6)(7)(8). COPII vesicles contain a small GTPase Sar1 and the heterodimeric Sec23/24 and Sec13/31 complexes. Assembly of the COPII coat takes place on the ER membrane at discrete locations called ER exit sites and is regulated by the exchange of GDP for GTP and GTP hydrolysis by Sar1 GTPase. Of various ER export motifs identified, the di-acidic motifs have been found in the cytoplasmic C termini of several membrane proteins such as vesicular stomatitis virus glycoprotein (VSVG), cystic fibrosis transmembrane conductance regulator, and potassium channels (KAT1, TASK-3, and Kir2.1) (1, 2, 9 -12).
Interestingly, export function of the di-acidic motifs is mediated through their interaction with components of COPII transport vesicles, particularly Sec24 subunits (6,30,55,56). This interaction results in the concentration of cargoes in the ER exit sites and facilitates cargo recruitment onto the COPII vesicles (13).
G protein-coupled receptors (GPCRs) constitute the largest superfamily of cell surface receptors, which regulate a variety of cell functions (14 -16). Similar to many other plasma membrane proteins, GPCR export from the ER is a key event in the anterograde trafficking to the cell surface (17)(18)(19)(20). Indeed, ER export has been shown to be a rate-limiting step for the cell surface transport of the receptors (21) and a number of studies have recently identified highly conserved hydrophobic sequences, which are required for GPCR export from the ER to the cell surface (22)(23)(24)(25)(26)(27). Although the molecular mechanism underlying the function of these motifs in controlling GPCR export remains elusive, several studies suggest that they are likely involved in the regulation of proper receptor folding in the ER. First, these motifs are located in the 8 th ␣-helical region within the membrane-proximal C terminus which has been shown to interact with other intracellular domains. These intramolecular hydrophobic interactions may modulate proper receptor folding in the ER (25,28). Second, the defect in the transport of these motif mutants can be partially rescued by a number of treatments, such as pharmacologic chaperones, chemical chaperones and lowing temperature culture (20,25,27). Third, the receptor mutants have enhanced abilities to bind to chaperone proteins (25). Fourth, none of these motifs have been shown to directly link to the COPII vesicles, which exclusively mediate the export of newly synthesized cargoes, including GPCRs, from the ER (29 -34). Therefore, whether or not GPCR export from the ER is dictated by linear ER export motifs remains unknown.
The physiological function of angiotensin II (Ang II), an octapeptide hormone, is mediated through activating its cell surface receptors. There are two major subtypes of Ang II receptors, the Ang II type 1 (AT1R), and the Ang II type 2 receptors (AT2R), both of them belong to the superfamily of GPCRs (35)(36)(37)(38)(39). Similar to many other GPCRs, the precise functions of the Ang II receptors are crucially regulated by their intracellular trafficking including exocytosis, endocytosis, recycling, and degradation and indeed, mistrafficking and dysfunction of the Ang II receptors have been implicated in the pathogenesis of * This work was supported, in whole or in part, by National Institutes of Health Grant R01GM076167 (to G. W.). 1 To whom correspondence should be addressed: Dept. of Pharmacology and Toxicology, Georgia Health Sciences University, Augusta, GA 30912. E-mail: guwu@georgiahealth.edu. 2 The abbreviations used are: ER, endoplasmic reticulum; GPCR, G proteincoupled receptor; AR, adrenergic receptor; Ang II, angiotensin II; AT1R, angiotensin II type 1 receptor; AT2R, angiotensin II type 2 receptor; TGN, trans-Golgi network; VSVG, vesicular stomatitis virus glycoprotein.
certain diseases (36, 40 -42). We are interested in elucidating the molecular mechanism by which newly synthesized GPCR transport from the ER to the cell surface. We have demonstrated that the cell surface expression of AT1R is regulated by Ras-like small GTPases, such as Rab1, Rab2, Rab6, Sar1, and ARF1 in HEK293 cells and cardiac myocytes (29,(43)(44)(45)(46). We have also demonstrated that AT1R exit from the ER is mediated through the COPII vesicles and controlled by the F(X) 6 LL motif in the membrane-proximal C terminus and a single Leu residue in the center of the first intracellular loop (23,24,29,47). These studies indicate that, similar to endocytosis, cell surface export of AT1R is a highly regulated process. In contrast to AT1R, how AT2R exports from the ER and then transports to the cell surface remains largely unknown. We have recently demonstrated that AT2R transport from the ER to the Golgi depends on the normal function of Rab1 (48). In this report, we have studied the role of the C terminus in the cell surface expression of AT2R. We found that two di-acidic motifs, ExD and ExE, in the membrane-distal C-terminal regions of rat and human AT2R, respectively, play an obligatory role in their forward trafficking toward the cell surface and the function of the di-acidic motif in AT2R export is likely mediated through facilitating the recruitment of the cargo receptors onto the COPII vesicles. We have also demonstrated that ExE motif is also important for AT1R cell surface expression. These di-acidic sequences represent first linear motifs in the GPCR superfamily, which dictate receptor exit from the ER.
Plasmid Constructions-AT1R, AT2R, ␣ 2B -AR, ␤ 2 -AR, and ␣ 1B -AR tagged with GFP at their C termini or three HA at their N termini were generated as described previously (23,43,46,48,49). GFP and HA epitopes have been used to label GPCRs, resulting in the receptors with similar characteristics to the wild-type receptors (23,43,50). For generation of GFP-tagged AT2R mutant ⌬42 in which the C-terminal 42 amino acid res-idues was removed, the full-length AT2R was amplified by PCR (forward primer, 5Ј-GCGCAAGCTTATGAAGGACAACTT-CAGTTTTG-3Ј; reverse primer, 5Ј-GCTAGTCGACTTAAC-GAAACAATACAGGAAGG-3Ј), restricted with HindIII and SalI, and ligated into the pEGFP-N1 vector (Invitrogen). For generation of the HA-tagged AT2R mutant ⌬42, AT2R was amplified by PCR (forward primer, 5Ј-GCGCCTCGAGATGA-AGGACAACTTCAGTTTTG-3Ј; reverse primer, 5Ј-GCA-TAAGCTTTTA AACGAAACAATACAGGAAGGG-3Ј) in which AT2R was in-framed with HA, restricted with XhoI and HindIII and ligated into the HA-pcDNA3.1(Ϫ) vector which was generated as described previously (48). A similar strategy was used to generate GFP-and HA-tagged AT2R mutant ⌬7 in which the last 7 residues were deleted. Mutagenesis was carried out by using the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA). The structure of each construct used in the present study was verified by restriction mapping and nucleotide sequence analysis.
Cell Culture and Transient Transfection-Human embryonic kidney (HEK) 293 cells and PC12 cells derived from a pheochromocytoma of rat adrenal medulla were cultured in Dulbecco's modified Eagle's medium (DMEM) with 10% fetal bovine serum, 100 units/ml penicillin, and 100 units/ml streptomycin. The cells were transiently transfected by using Lipofectamine 2000 reagent (Invitrogen) as described previously (43). The expression levels of rat AT2R in HEK293 cells, human AT2R in HEK293 cells, rat AT2R in PC12 cells, and human AT2R in PC12 cells were expressed at 422, 366, 334, and 176 fmol/mg protein, respectively, measured by 125 I-Ang II binding in intact cells as described below.
For chemical rescue, HEK293 cells were incubated with DMSO at a concentration of 2% in a total of 2 ml of DMEM without fetal bovine serum for 24 h. For low temperature rescue, the cells were cultured at 30°C for 36 -40 h. To determine the effect of treatment with PD123319, a specific AT2R antagonist, on the cell surface expression of wild-type AT2R and its mutant ExD-AxA, HEK293 cells were transfected with AT2R for 6 h and incubated with PD123319 at a concentration of 10 M for 30 -36 h.
Flow Cytometric Analysis of AT2R and AT1R Expression-For measurement of receptor expression at the cell surface, cells were transfected with 1.0 g of HA-tagged receptor for 36 -48 h. After collecting the cells, the cells were suspended in PBS containing 1% fetal calf serum (FCS) at a density of 4 ϫ 10 6 cells/ml and incubated with high affinity anti-HA-fluorescein (3F10) at a final concentration of 2 g/ml for 30 min at 4°C. After washing twice with 0.5 ml of PBS/1% FCS, the cells were resuspended, and the fluorescence was analyzed as described above. Because the staining with anti-HA antibodies was carried out in the non-permeabilized cells and only those receptors expressed at the cell surface were accessible to anti-HA antibodies, the measured fluorescence reflected the amount of receptor expressed at the cell surface. To measure total receptor expression in cells transfected with HA-tagged receptors, the cells were permeabilized by incubation with PBS containing 0.2% Triton X-100 for 5 min before staining with anti-HA antibodies.
For measurement of total GFP-tagged receptor expression, cells were cultured on 6-well plates and transfected with 1.0 g of GFP-tagged receptor for 36 -48 h. The cells were collected and washed three times with phosphate-buffered saline (PBS). The cells were then resuspended at a density of 8 ϫ 10 6 cells/ml and the overall receptor expression was determined by measuring total GFP fluorescence on a flow cytometer (FACSCalibur; BD Biosciences, San Jose, CA) as described previously (23,48).
Measurement of Cell Surface Expression of AT2R, ␣ 2B -AR, ␤ 2 -AR, and ␣ 1B -AR by Intact Cell Ligand Binding-HEK293 cells cultured on 6-well dishes were transiently transfected with 1.0 g of ␣ 2B -AR, ␤ 2 -AR, or ␣ 1B -AR. After 6 h, the cells were split into 12-well dishes pre-coated with poly-L-lysine at a density of 5 ϫ 10 5 cells/well. 36 -48 h post-transfection, the cells were incubated with DMEM plus [ 3 H]RX821002, [ 3 H]CGP12177 at room temperature or [ 7 methoxyn-3 H]prazosin at 4°C at a concentration of 20 nM in a total of 400 l for 2 h to measure the cell surface expression of ␣ 2B -AR, ␤ 2 -AR, and ␣ 1B -AR, respectively. The nonspecific binding of ␣ 2B -AR, ␤ 2 -AR, and ␣ 1B -AR was determined in the presence of rauwolscine (10 M), alprenolol (20 M), and phentolamine (10 M), respectively, and accounted for less than 10% of the total binding. The binding was terminated and excess radioligand eliminated by washing the cells twice with ice-cold DMEM. All of the retained radioligand was then extracted by digesting the cells in 1 M NaOH for 2 h at room temperature. The liquid phase was collected and suspended in 5 ml of Ecoscint A scintillation fluid. The amount of radioactivity retained was measured by liquid scintillation spectrometry (23,46).
To measure the cell surface expression of AT2R, HEK293, and PC12 cells were transfected with HA-tagged AT2R as described above. The cells were incubated with DMEM plus 125 I-Ang II at room temperature at a concentration of 2.5 nM in a total of 200 l for 2 h. Nonspecific binding of AT2R was determined in the presence of PD123319 at a concentration of 2.5 M. To compare the ligand binding abilities of AT2R and its mutants, HEK293 cells expressing wild-type AT2R, E357A, or ExD-AxA were incubated with different concentrations of 125 I-Ang II (from 0.02 to 2.5 nM). The cells were washed twice with 1 ml of ice cold PBS and the cell-surface bound 125 I-Ang II was extracted by mild acid treatment (2 ϫ 5 min with 0.5 ml buffer containing 50 mM glycine, pH 3, and 125 mM NaCl). The radioactivity was counted in a gamma counter. All assays were performed in duplicate, and receptor density was expressed as fmol/mg membrane protein. Saturation binding curves were generated using Prism.
Fluorescence Microscopy-For fluorescence microscopic analysis of receptor subcellular localization, cells were grown on coverslips pre-coated with 50% poly-L-lysine in 6-well plates and transfected with 0.5 g of AT2R-GFP, and the cells were fixed with ice-cold methanol at Ϫ20°C for 10 min. For subcellular localization of HA-AT2R, HEK293 cells were permeabilized with PBS containing 0.2% Triton X-100 for 5 min and blocked with 5% normal donkey serum for 2 h. The cells were then incubated with TRITC-labeled antibodies against HA at a dilution of 1:30 for 1 h. After washing with PBS (3 ϫ 5 min), the coverslips were mounted and fluorescence was detected with a DMRA2 epifluorescent microscope (Leica Microsystems, Inc., Deerfield, IL) as described previously (23,43). Images were deconvolved using SlideBook software and the nearest neighbor deconvolution algorithm (Intelligent Imaging Innovations, Denver, CO).
Coimunoprecipitation of AT2R and Calnexin-HEK293 cells cultured on 100-mm dishes were transfected with 4 g of HAtagged AT2R for 36 h. The cells were washed twice with PBS, harvested, and lysed with 500 l of lysis buffer containing 50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, and Complete Mini protease inhibitor mixture. After gentle rotation for 1 h, samples were centrifuged for 15 min at 14,000 ϫ g, and the supernatant was incubated with 50 l of protein G-Sepharose for 1 h at 4°C to remove nonspecific bound proteins. Samples were then incubated with 3 g of anti-HA antibodies overnight at 4°C with gentle rotation followed by an incubation with 50 l of protein G-Sepharose 4B beads for 5 h. Resin was collected by centrifugation and washed three times each with 500 l of lysis buffer without SDS. Immunoprecipitated receptors were eluted with 100 l of 1 ϫ SDS gel loading buffer and separated by 8% SDS-PAGE. The amounts of AT2R and calnexin in the immunoprecipitates were determined by immunoblotting using anti-HA and calnexin antibodies, respectively, as described previously (23).
Statistical Analysis-Differences were evaluated using Student's t test and p Ͻ 0.05 was considered as statistically significant. Data are expressed as the mean Ϯ S.E.

RESULTS
AT2R Transport to the Cell Surface Depends on Its Membrane-distal C Terminus-A number of studies have demonstrated that the C termini play an important role in the cell surface transport of GPCRs. In an attempt to search for structural determinants for AT2R transport to the cell surface, we first measured the effect of deleting the entire C-terminal 42 residues (⌬42) (Fig. 1A) on the cell surface expression of rat AT2R in HEK293 cells. AT2R and its mutant ⌬42 were tagged with HA at their N termini and their cell surface expression levels measured by flow cytometry following staining with anti-HA antibodies in non-permeabilized cells. To exclude the possibility that C-terminal deletion may influence the total receptor expression, AT2R and its mutant ⌬42 were also tagged with GFP at their C termini, and the overall AT2R expression was measured by flow cytometry detecting the GFP signal. The cell surface expression of the AT2R mutant ⌬42 was markedly reduced as compared with AT2R, whereas total AT2R expression was not significantly different between wild type and mutated AT2R measured by flow cytometry (Fig. 1B). These data indicate that the C-terminal tail is involved in the regulation of cell surface targeting of AT2R.
We then used the progressive deletion strategy to look for the subdomain in the C terminus responsible for AT2R transport. Similar to the removal of the entire C terminus, deleting the last 7 amino acid residues (⌬7) markedly attenuated the cell-surface expression of AT2R (Fig. 1B). Deletion of the last 7 residues did not compromise the overall receptor expression (Fig. 1B). Similarly, deletion of the C-terminal 13, and 28 amino acid residues dramatically reduced the cell surface expression of AT2R without influencing total expression of the receptor (data not shown). These data suggest that the distal C-terminal 7 residues contain signals modulating AT2R transport to the cell surface.
To further confirm the effect of deleting the C terminus on the cell surface expression of AT2R, the subcellular distribution of AT2R and its deletion mutants was visualized at the steady state by fluorescent microscopy. As expected, wild-type AT2R tagged with either HA or GFP was nicely expressed at the cell surface, suggesting that epitope tagging does not alter AT2R transport to the cell surface. The AT2R mutants lacking the C-terminal 7 and 42 residues had a very similar distribution pattern and were extensively localized in the perinuclear region (Fig. 1C), paralleling the effect on the cell surface expression of the receptor.
Identification of Key Amino Acid Residues Required for the Export of AT2R to the Cell Surface-Our preceding data have demonstrated that the last 7 amino acid residues (EMDTFVS) are required for AT2R expression at the cell surface. This subdomain contains an ExD and an FV sequence which are very similar to the di-acidic and di-hydrophobic ER export motifs identified in several plasma membrane proteins (1-8), respectively. To identify key amino acid residues required for AT2R transport to the cell surface, each of these two sequences were substituted with Ala. Simultaneous mutation of two acidic residues in the ExD motif markedly blocked cell surface expression FIGURE 2. Site-directed mutagenesis identifies key residues in the membrane-distal C terminus required for the transport of AT2R from the ER to the cell surface. A, effect of mutating the C-terminal residues on the cell surface and total expression of AT2R. HEK293 cells were transfected with AT2R or its mutants in which the C-terminal sequences LR, ExD, and FV were mutated to Ala. Cell surface and total expression of AT2R and its mutants were measured as described in legend of Fig. 1B. The data shown are percentages of the mean value obtained from cells transfected with WT AT2R and are presented as the mean Ϯ S.E. of five experiments. *, p Ͻ 0.05 versus WT. B, subcellular distribution of AT2R and its mutant ExD-AxA. GFP-or HA-tagged AT2R and ExD-AxA were transiently expressed in HEK293 cells and their subcellular distribution was revealed by fluorescence microscopy detecting GFP fluorescence (upper panel) or following immunostaining with rhodamine-conjugated anti-HA antibodies (lower panel) as described in "Experimental Procedures." The data shown are representative images from at least five separate experiments. Scale bar, 10 m. FIGURE 1. Effect of deleting the C terminus on the cell surface expression and subcellular distribution of rat AT2R. A, sequences of the C terminus (CT) of AT2R and its ⌬7 and ⌬42 mutants, in which the C-terminal 7 and 42 residues were removed. B, cell surface and total expression of AT2R and its ⌬7 and ⌬42 mutants. For measurement of cell surface expression, AT2R or its individual mutant tagged with three HA at their N termini were transiently expressed in HEK293 cells and their cell surface expression was quantitated by flow cytometry following staining with anti-HA antibodies in non-permeabilized cells. For measurement of total expression, AT2R or its individual mutant tagged with GFP at their C termini were transiently expressed and total expression of the receptors was quantitated by flow cytometry measuring total GFP signal. The data shown are percentages of the mean value obtained from cells transfected with WT AT2R and are presented as the mean Ϯ S.E. of three experiments. *, p Ͻ 0.05 versus WT. C, subcellular distribution of AT2R and its ⌬7 and ⌬42 mutants. GFP-tagged (upper panel) or HA-tagged AT2R, ⌬7, and ⌬42 (lower panel) were transiently expressed in HEK293 cells. Subcellular distribution of the GFP-tagged receptors was revealed by detecting GFP fluorescence and the HA-tagged receptors by fluorescence microscopy following immunostaining with rhodamine-conjugated anti-HA antibodies (1:50) as described under "Experimental Procedures." The data shown are representative images from at least three separate experiments. Scale bar, 10 m. of AT2R by 88% (Fig. 2A). In contrast, mutation of FV moderately reduced AT2R expression at the cell surface by ϳ30%, whereas mutation of LR did not clearly influence the cell surface expression of AT2R ( Fig. 2A). Furthermore, mutation of LR, ExD, and FV did not alter the overall AT2R expression ( Fig.  2A).
Consistent with the cell surface expression of AT2R measured by flow cytometry, fluorescence microcopy analysis of the subcellular distribution of AT2R showed that, similar to the C-terminal truncation mutants, the ExD mutant tagged with GFP or HA was expressed in the perinuclear region unable to transport to the cell surface (Fig. 2B).
Co-localization of the ExD Motif Mutant with Different Organelle Markers-We next sought to define the intracellular compartment in which the ExD motif modulates AT2R transport. GFP-tagged AT2R ExD mutant was transfected into HEK293 cells and its co-localization with the ER marker DsRed2-ER, the ERGIC marker ERGIC-53, the cis-Golgi marker GM130 and the TGN marker p230 was revealed by microscopy. The intracellularly accumulated AT2R ExD mutant was extensively co-localized with DsRed2-ER (Fig. 3A), but not with ERGIC-53 (Fig. 3B), GM130 (Fig. 3C), and p230 (Fig. 3D). These data suggest that the di-acidic motif ExD in the distal C terminus plays a crucial role for AT2R export from the ER.
Role of Each Acidic Residue in the ExD Motif in AT2R Delivery to the Cell Surface-To further characterize the function of the ExD motif in AT2R transport, we determined the effect of mutating individual acidic residue in the ExD motif to Ala on AT2R expression at the cell surface. Mutation of Glu-357 to Ala significantly attenuated AT2R transport to the cell surface by ϳ50%, whereas mutation of Asp-359 to Ala reduced AT2R cell surface expression by about 90%, which is similar to the effect of simultaneous mutation of both acidic residues (Fig. 4A). These data suggest that the second acidic residue Asp-359 is more important than the first acidic residue Glu-357 in AT2R export to the cell surface.
We then determined the effect of mutating Glu-357 to Asp and/or Asp-359 to Glu on AT2R transport. Mutation of Glu-357 to Asp (ExD to DxD) produced a more detrimental effect on AT2R cell surface expression than mutation to Ala, whereas mutation of Asp-359 to Glu (ExD to ExE) partially but significantly restored AT2R cell surface expression compared with mutation to Ala (Fig. 4A). Simultaneous mutation of Glu-357 to Asp and Asp-359 to Glu (ExD to DxE) also dramatically blocked AT2R cell surface transport by 70% (Fig. 4A). In contrast, these mutations did not clearly influence the total receptor expression (Fig. 4A). Subcellular localization revealed that all mutations blocked AT2R transport to the cell surface, and the mutated receptors were extensively expressed in intracellular compartments (Fig. 4B). These data demonstrate that the transport function of the first acidic residue Glu-357 cannot be substituted with Asp, whereas the function of the second acidic residue Asp-359 can be partially replaced by Glu.
Role of the Di-acidic ExE Motif in the Cell Surface Transport of Human AT2R-The AT2R C-terminal sequences are highly conserved among different species. However, distinct di-acidic motifs are found. Rat and mouse AT2R have an ExD motif, whereas human AT2R contains an ExE motif in the membranedistal C termini (Fig. 5A). To further explore the role of the di-acidic motifs in AT2R transport, we determined if the ExE motif is required for human AT2R export using the same strategy as for rat AT2R. Mutation of two Glu residues in the ExE motif individually or in combination to Ala or Asp residues significantly reduced AT2R transport to the cell surface (Fig.  5B) and the mutated receptors were arrested inside the cell (Fig.  5C). Interestingly, mutation of Glu-359 to Ala produced more profound effect on AT2R transport than mutation of Glu-357 and the function of Glu-357 residue cannot be substituted with Asp, whereas the function of Glu-359 can be partially rescued by Asp (Fig. 5B). Furthermore, these mutations did not alter the total receptor expression (Fig. 5A). These data demonstrate that the di-acidic motif ExE is required for the cell surface transport of human AT2R.

Effect of Swapping the C Termini on the Cell Surface Export of
Rat and Human AT2R-The preceding data demonstrate that rat and human AT2R use the ExD and ExE motifs, respectively, to export from the ER and mutation of Asp-359 to Glu (ExD to ExE) in rat AT2R and mutation of Glu-359 to Asp (ExE to ExD) in human AT2R lead to a modulate reduction in their transport to the cell surface in HEK293 cells, suggesting that the functions of these two di-acidic motifs in the C termini in modulating the transport of rat and human AT2R are not fully exchangeable. To define if this is caused by other differences exist in their C termini, we determined the effect on the cell surface expression of swapping the C termini between rat and human AT2R, which was achieved by mutating Thr-346 to Ser and Asp-359 to Glu in rat AT2R (T346S/D359E) and mutating Ser-346 to Thr and Glu359 to Asp in human AT2R (S346T/E359D) (Fig. 6A). The cell surface expression of the rat AT2R mutant carrying the human AT2R C terminus (T346S/D359E) and the human AT2R mutant carrying the rat AT2R C terminus (S346S/E359D) was partially reduced as compared with their wild types (Fig. 6B). These data further confirm that the optimal export of rat and human AT2R is under control by their own C termini or di-acidic motif.
Effect of Mutation of the Di-acidic Motifs on the Cell Surface Transport of AT2R in PC12 Cells-To define if the function of the di-acidic motifs in modulating AT2R export is dependent on cell types examined, we determined the effect of mutating the di-acidic motifs on the cell surface expression and subcellular localization of both rat and human AT2R in PC12 cells, which endogenously express AT2R. Similar to the results observed in HEK293 cells, the cell surface expression of the rat AT2R mutant ExD-AxA and the human AT2R mutant EXE-AXA was dramatically decreased by about 80% as compared with their wild-type counterparts (Fig. 7A), whereas the total expression was about the same (data not shown). The mutated receptors were distributed in the perinuclear region (Fig. 7B). These data suggest that the export function of the di-acidic motifs in the C termini of AT2R is independent of cell types.
The Di-acidic ExD Motif Is Unlikely Involved in Proper AT2R Folding-To define the molecular mechanism underlying the function of the di-acidic motifs in AT2R export trafficking, we tested if the ExD motif is involved in proper folding of rat AT2R. In the first series of experiments, we measured if mutation of the ExD motif could alter AT2R ability to bind its ligand Ang II. AT2R and its E357A and ExD-AxA . Effect of mutating the ExD motif on AT2R transport to the cell surface. A, cell surface and total expression of AT2R and its ExD motif mutants. HEK293 cells were transiently transfected with HA-tagged AT2R and its mutants, in which the ExD residues were individually or simultaneously substituted by Ala or other acidic residues. The cell surface and total expression of the receptors was determined by flow cytometry after staining with anti-HA antibodies as described in "Experimental Procedures." The data shown are the percentages of the mean value obtained from cells transfected with wild-type AT2R (WT) and are presented as the mean Ϯ S.E. of four separate experiments. * and #, p Ͻ 0.05 versus cells transfected with WT AT2R and the mutant D359A, respectively. B, subcellular localization of AT2R and its mutants. HA-tagged AT2R and its mutants were transiently expressed in HEK293 cells and their subcellular distribution was revealed by fluorescence microscopy following staining with anti-HA antibodies. Scale bar, 10 m.

FIGURE 5. Effect of mutating the ExE motif in the membrane-distal C terminus on the cell surface transport of human AT2R.
A, sequences of the C terminus (CT) of human AT2R. B, cell surface and total expression of human AT2R and its ExE motif mutants. HEK293 cells were transiently transfected with HA-conjugated human AT2R and its mutants in which the ExE residues were individually or simultaneously substituted by Asp or Ala and the cell surface and total expression of the receptors was determined by flow cytometry after staining with anti-HA antibodies as described in "Experimental Procedures." The data shown are the percentages of the mean value obtained from cells transfected with wild-type AT2R (WT) and are presented as the mean Ϯ S.E. of four separate experiments. * and #, p Ͻ 0.05 versus cells transfected with WT and the mutant E359A, respectively. C, subcellular localization of human AT2R and its ExE motif mutants. HA-tagged AT2R and its mutants were transiently expressed in HEK293 cells and their subcellular distribution was revealed by fluorescence microscopy following staining with anti-HA antibodies. Scale bar, 10 m. mutants, which have differential impacts on the cell surface transport of the receptor, were transiently expressed in HEK293 cells and their abilities to bind 125 I-Ang II were measured in the intact cell ligand binding assay. It is apparent that AT2R and E357A were able to bind Ang II in an Ang II dose-dependent fashion (Fig. 8A). The B max values of AT2R, E357A, and ExD-AxA were 461, 202, and 36 fmol/mg protein, respectively (Fig. 8A). These data are strongly parallel with the effects of mutations on the cell surface expression of AT2R as measured by flow cytometry (Fig. 4A).
In the second series of experiments, we determined if lowtemperature culture and treatments with the chemical chaperone DMSO and the specific AT2R antagonist PD123319 were able to rescue cell surface expression of the di-acidic ExD motif mutant. These treatment conditions are known to promote the cell surface expression of certain misfolded proteins by virtue of either relaxing the quality control system or promoting tighter protein conformations within the ER. We have previously demonstrated that the residue Phe-436 modulates ␣ 2B -AR folding likely through intramolecular interactions with other hydrophobic residues such as Val-42 in the first transmembrane domain (25) and thus it was used as a positive control. HEK293 cells were transfected with ␣ 2B -AR, ␣ 2B -AR mutant F436A, AT2R, or AT2R mutant ExD-AxA and then either treated with DMSO, PD123319 or cultured at 30°C. Consistent with our published data, low temperature culture and DMSO treatment significantly increased the cell surface expression of ␣ 2B -AR mutant F436A as measured by intact cell ligand binding (Fig. 8,  B-E). In contrast, under the same conditions, reduced temperature culture and DMSO treatment did not clearly influence the cell surface expression of AT2R mutant ExD-AxA (Fig. 8, B-E). Interestingly, PD123319 treatment similarly promoted the cell surface expression of AT2R and its mutant ExD-AxA (Fig. 8, F  and G).
In the third series of experiments, we compared the interaction with the ER chaperone calnexin between AT2R and its mutant ExD-AxA. It has been well demonstrated that terminally misfolded GPCRs display strong and extended interactions with ER resident chaperones such as calnexin (51,52). HA-tagged AT2R and its mutant ExD-AxA were expressed in HEK293 cells and then immunoprecipitated by anti-HA antibodies. The amount of calnexin in the HA immunoprecipitates was then analyzed by Western blotting. The amounts of calnexin were similar in the HA immunoprecipitates from cells expressing AT2R and ExD-AxA (Fig. 8, H and  I). These data strongly suggest that the di-acidic ExD motif is unlikely involved in the regulation of correct AT2R folding.
Effect of Mutating the Di-acidic Motif ExD on the Recruitment of AT2R onto ER Exit Sites-Because di-acidic motifs have been described to mediate cargo interaction with Sec24 and to facilitate cargo recruitment onto COPII transport vesicles in some non-GPCR membrane proteins (1, 2, 6, 9 -12, 30, 55, 56), we determined if the ExD motif is also able to enhance the recruitment of AT2R onto the COPII vesicles. We first determined if AT2R export depends on the COPII vesicles. Transient expression of the GTP-bound mutant Sar1H79G to inhibit the formation of COPII vesicles significantly blocked the cell surface transport of AT2R, whereas expression of wild-type Sar1 did FIGURE 6. Effect of swapping the C termini on the cell surface transport of rat and human AT2R. A, an alignment of the C-terminal sequences of rat and human AT2R. The residues that are different in the rat and human AT2R are shown in bold. B, cell surface expression of human AT2R and its mutants. The double mutant T346S/D359E is a rat AT2R mutant carrying the human AT2R C terminus and the double mutant S346T/E359D is a human AT2R mutant carrying the rat AT2R C terminus. HEK293 cells were transiently transfected with HA-conjugated rat or human AT2R or their mutants and the cell surface expression of the receptors was determined by flow cytometry after staining with anti-HA antibodies. The data shown are the percentages of the mean value obtained from cells transfected with wild-type AT2R (WT) and are presented as the mean Ϯ S.E. of four experiments. *, p Ͻ 0.05 versus respective WT. not produce clear influence (Fig. 9A). These data suggest that the export of AT2R requires the COPII vesicles.
We next determined if the ExD motif influences the recruitment of AT2R onto the COPII vesicles. HEK293 cells were transfected with Sar1H79G together with GFP-tagged AT2R or its mutant ExD-AxA. AT2R subcellular localization and its co-  localization with Sec24D, a component of COPII transport vesicles, were visualized following staining with anti-Sec24D antibodies. Consistent with reduced cell surface expression, AT2R was largely expressed inside the cell in the presence of Sar1H79G and the intracellularly accumulated AT2R was strongly co-localized with Sec24D, an ER exit site marker (Fig.  9B). In contrast, the AT2R mutant ExD-AxA was unable to export to ER exit sites. These data suggest that mutation of the ExD motif disrupts the recruitment of the cargo AT2R onto ER exit sites or the COPII vesicles.
Differential Regulation of the Cell Surface Transport of Distinct GPCRs by Di-acidic Motifs-We next determined if diacidic motifs are required for the cell surface transport of other GPCRs, specifically AT1R, ␤ 2 -AR, and ␣ 1B -AR, which C termini contain the di-acidic sequences EVE, EQEKE, and EPE, respectively (Fig. 10A). Mutation of EVE to AVA in the distal C terminus moderately but significantly blocked the cell surface expression of AT1R by ϳ30%. In contrast, mutation of EQEKE to AQAKA in ␤ 2 -AR and EPE to APA in ␣ 1B -AR did not have clear effect on their cell surface transport. These data indicate the di-acidic motifs selectively modulate the cell surface export of AT1R and AT2R, but not ␤ 2 -AR and ␣ 1B -AR.

DISCUSSION
The most important finding presented in this report is that ER export and cell surface transport of AT1R and AT2R is directed by di-acidic motifs located in the membrane-distal, nonstructural C-terminal regions. These di-acidic motifs revealed here represent first linear signals in the GPCR superfamily, which control export trafficking, more specifically exit from the ER, of newly synthesized receptors.
The C-terminal tails of GPCRs consist of a putative amphipathic 8 th ␣-helix in the membrane-proximal region and a nonstructural membrane-distal region. The requirement of the membrane-proximal C-terminal portion for ER export has been described for a number of GPCRs, including AT1R, ␣ 2B -AR, rhodopsin, vasopressin V2 receptor, dopamine D1 receptor, adenosine A1 receptor, melanin-concentrating hormone receptor 1, and luteinizing hormone/choriogonadotropin receptor (24,53,54). Mutagenesis studies have also identified a number of highly conserved motifs in the membrane-proximal C termini essential for GPCR export from the ER, such as the E(x) 3 LL, FN(x) 2 LL(x) 3 L, and F(x) 3 F(x) 3 F motifs (22,26,27). We have recently identified the F(x) 6 LL motif in the 8 th ␣-helix, which modulates export of ␣ 2B -AR, ␤ 2 -AR, ␣ 1B -AR, and AT1R from the ER (24,25,47). In the present study, we have demonstrated that the membrane-distal C terminus is also absolutely required for AT2R anterograde transport to the cell surface as the AT2R mutant lacking the C-terminal 7 residues was unable to transport to the cell surface, which is similar to the mutant lacking the entire C-terminal 42 residues. More interestingly, mutagenesis studies identified two distinct di-acidic motifs, ExD and ExE, in the distal C-terminal regions of rat and human AT2R, respectively, which are required for receptor export from the ER and transport to the cell surface. Furthermore, mutation of the ExE motif in the membrane-proximal C terminus also attenuated AT1R transport to the cell surface. These data indicate that both the membrane proximal and distal C-terminal portions contain structural determinants directing GPCR export trafficking.
The di-acidic sequences have been demonstrated to function as ER export motifs in several membrane proteins (1, 2, 9 -12). Among the di-acidic ER export motifs identified, the sequence DxE is the best characterized in which two acidic residues are equally important in ER export and either acidic residue cannot be substituted by other acidic residue. We found that two acidic residues in the ExD or ExE motifs in AT2R differentially regulate receptor export with unique properties. First, the second acidic residue in the ExD and ExE motifs is more important than the first acidic residue in the export of AT2R. Second, the function of the second acidic residue can be partially replaced by conservative mutation with other acidic residue (i.e. Asp with Glu or Glu with Asp), whereas the first acidic residue cannot be functionally substituted by other acidic residue (i.e. Glu by Asp). These data suggested that the first acidic residues in the ExD and ExE motifs are invariant and the function of the second residue is determined at least in part by the negatively charged property.
Our data have demonstrated that ER export of rat AT2R is controlled by the ExD motif, whereas human AT2R by the ExE motif. These data indicate that AT2R from different species utilize distinct di-acidic motifs in the membrane-distal C termini as ER export codes. The fact that mutation of the ExD motif to DxE, ExE, or DxD in rat AT2R and mutation of the ExE motif to ExD, DxE, or DxD in human AT2R produced inhibi- tory effects on the cell surface transport of the receptors suggests that, among the 4 possible di-acidic combinations (ExD, DxE, ExE, and DxD), the ExD motif is the best for rat AT2R export, whereas the ExE motif is the best for human AT2R transport.
Furthermore, similar to mutation of the di-acidic motifs, swamping the C termini between rat and human AT2R produced moderate inhibitory effects on the export of both AT2R, further confirming different roles of the ExD and ExE sequences in their export. Moreover, mutation of the di-acidic motifs in rat and human AT2R produced almost same inhibition in HEK293 and PC12 cells, which were derived from human and rat, respectively. These data suggest that the function of these di-acidic motifs in modulating AT2R transport is not cell type-or organism-specific. Nevertheless, these data demonstrate that the transport function of the ExD motif in rat AT2R and the ExE motif in human AT2R are not fully exchangeable.
It has been demonstrated that di-acidic motifs are involved in rapid concentration of cargo into the COPIIcoated vesicles and are able to directly bind to the COPII component Sec24 (1,2,6,13,55,56). We have demonstrated here that the function of the ExD motif in modulating AT2R transport to the cell surface is likely mediated through enhancing the recruitment of the cargo onto ER export sites or the COPII transport vesicles. However, we failed to detect the interaction between Sec24 and AT2R either by co-immunoprecipitation from cells transiently expressing both AT2R and Sec24 or by AT2R C-terminal glutathione S-transferase fusion protein pull-down assay (data not shown).
It appears that di-acidic motifs may selectively or differentially modulate ER export of distinct GPCRs. Although the cell surface expression of AT1R and AT2R was significantly reduced by mutation of the di-acidic motifs in the C termini, the inhibitory effects on AT2R and AT1R were markedly different. Whereas mutating the ExD and ExE motifs almost abolished the cell surface transport of rat and human AT2R, respectively, mutation of the ExE motif at the very end of the C terminus of AT1R only produced a moderate inhibition, demonstrating that the C-terminal di-acidic motifs play a more important role in the transport of AT2R than AT1R. In contrast to AT1R and AT2R, the cell surface expression of ␤ 2 -AR and ␣ 1B -AR was not altered by mutating the di-acidic motifs in their C termini. These data suggest that the di-acidic motifs may specifically modulate ER export of Ang II receptors, but not adrenergic receptors.
Ang II plays an important role in modulating the physiological function of virtually all organs as well as in the development of a number of human diseases such as diabetes, hypertension, myocardial infarction, congestive heart failure, and stroke (35,36). As the function of Ang II is mainly mediated through the cell surface AT1R and AT2R, to further explore regulatory mechanism of the ER-to-cell surface targeting of both Ang II receptors may provide a novel foundation for designing drugs to treat human diseases involving Ang II and its receptors.