Akt Mediates Sequestration of the β2-Adrenergic Receptor in Response to Insulin*

The counterregulation of catecholamine action by insulin includes insulin-stimulated sequestration of the β2-adrenergic receptor. Herein we examined the signaling downstream of insulin receptor activation, focusing upon the role of 1-phosphatidylinositol 3-kinase and the serine-threonine protein kinase Akt (also known as protein kinase B) in the internalization of β2-adrenergic receptors. Inhibition of 1-phosphatidylinositol 3-kinase by LY294002 blocks insulin-induced sequestration of the β2-adrenergic receptor, implicating Akt in downstream signaling to the β2-adrenergic receptor. Phosphorylation studies of the C-terminal cytoplasmic domain of the β2-adrenergic receptor by Akt in vitroidentified Ser345 and Ser346 within a consensus motif for Akt phosphorylation. Double mutation (i.e.S345A/S346A) within this motif abolishes insulin counterregulation of β-adrenergic stimulation of cyclic AMP accumulation as well as insulin-stimulated sequestration. Furthermore, expression of constitutively activated Akt (T308D/S473D) mimics insulin action on cyclic AMP responses and β2-adrenergic receptor internalization. Expression of the dominant-negative version of Akt (K179A/T308A/S473A), in contrast, abolishes both insulin counterregulation of the cyclic AMP response as well as insulin-stimulated sequestration of the β2-adrenergic receptor. The action of the serine-threonine protein kinase Akt in insulin counterregulation mirrors the central role of protein kinase A in β-agonist-induced desensitization.

by the PI3-kinase reaction activate 3-phosphoinositide-dependent kinase (PDK1) (5), which in turn, activates a serine-threonine protein kinase Akt. Akt has been shown to be a key element in insulin signaling downstream of PI3-kinase and in the trafficking of the insulin-sensitive transporter GLUT4 (6,7). Mice lacking Akt2, for example, display insulin resistance and a diabetes mellitus-like state (8), reflecting interruption of insulin signaling to GLUT4 trafficking as well as to other downstream signaling.
Insulin counterregulates catecholamine action, a facet of insulin action that includes insulin-stimulated phosphorylation and trafficking of ␤ 2 -adrenergic receptors (␤2AR) (9 -14) Sequestration of ␤2AR occurs in response to counterregulation by insulin and also in response to chronic stimulation of ␤2AR by ␤-adrenergic agonist, a late phase of agonist-induced desensitization (15,16). For agonist-induced desensitization, activation of G-protein-coupled receptor kinases (GRK) and protein kinase A play critical roles in catalyzing phosphorylation of the ␤2AR required for eventual internalization (15). As counterregulation of ␤2AR by insulin involves changes in phosphorylation and the trafficking of the receptor, we were intrigued by the possibility that Akt may be mediating these effects of insulin on this well known member of the superfamily of Gprotein-coupled receptors (GPCRs). In the current work, we investigate the nature of the signaling downstream of insulin activation of PI3-kinase, focusing upon the role of Akt. We show that activation of PI3-kinase and Akt is essential for insulin counterregulation and sequestration of ␤2AR.

EXPERIMENTAL PROCEDURES
Cell Culture and Materials-Human epidermoid carcinoma (A431) and Chinese hamster ovary (CHO) cells were maintained in Dulbecco's modified Eagles medium supplemented with 5% fetal bovine serum (HyClone, Logan, UT), plus penicillin (60 g/ml) and streptomycin (100 g/ml), grown in a humidified atmosphere of 5% CO 2 and 95% air at 37°C (17). For epifluoresence imaging, the larger A431 cells proved to be superior for localization studies to the CHO cells, and so this analysis was confined to the A431 cells.
Epifluorescence Imaging-Microscopy of live cells was performed on the Eclipse TE300 (Nikon) inverted microscope equipped with ϫ40 oil objective. Images were acquired using MicroMAX Imaging System (Princeton Instruments Inc.) and WinView32 software (19).
Inhibitor Studies-Stably transfected clones were routinely challenged in cell medium following serum deprivation for 8 -12 h without or with 100 nM insulin for 15 min and the trafficking of the GFP-tagged ␤2AR monitored by epifluorescence microscopy. Cells were serum-deprived for 8 h prior to removing growth factors and catecholamines from the cell medium. PI3-kinase was inhibited using the LY294002 com-pound (10 M). For studies of the effects of inhibitor on the trafficking of the GFP-tagged receptor in response to insulin, the inhibitor was added 30 min in advance of the challenge with insulin.
Phosphorylation of the C-terminal Cytoplasmic Domain of the ␤AR by Akt-The fusion protein of GST with the C-terminal, cytoplasmic domain of the ␤2AR (BAC1-C1 protein) was phosphorylated at 37°C in a 50-l reaction mixture containing 0.05 Tris-HCl, 0.1% Nonidet P-40, 0.15 M NaCl, 5 mM dithiothreitol, 5 mM MgCl 2 , 0.1 mM ␥-[ 32 P]ATP (0.5-2.0 ϫ 10 15 cpm/mol), 0.02 unit/l Akt (Upstate Biotechnology) and BAC1 (1.0 M). One unit of Akt activity corresponds to 1 pmol of phosphate transfer per min at pH 7.4 at 30°C. Phosphorylation was initiated by the addition of the enzyme to the reaction mixture. After a 30-min incubation, the reaction mixture was treated with 0.1% SDS and 50 mM dithiothreitol for 5 min at 75°C and subjected to SDSpolyacrylamide gel electrophoresis. The resolved protein gels were fixed in 10% acetic acid, stained, and the radiolabeled proteins made visible by autoradiography.
Assay of Intracellular Accumulation of Cyclic AMP and Counterregulation by Insulin-For assay of cyclic AMP accumulation, stably transfected A431 cells were seeded in 24-well plates 48 h prior to determination, at a density of 1 ϫ 10 5 cells/well. On the day of experiment, cell culture medium was aspirated, the cells washed and replenished with Krebs-Ringer phosphate medium containing 10 M RO-201724 (cyclic AMP phosphodiesterase inhibitor), and then treated with the isoproterenol (10 M) for 5 min in a total assay volume of 300 l. The reaction was terminated by the addition of 100 l of 100% ethanol, and the cyclic To investigate the possible role of PI3-kinase in these responses, cells were exposed to 10 M LY294002 compound for 60 min prior to the addition of insulin. The time course for insulin-stimulated counterregulation of isoproterenol-stimulated cyclic AMP accumulation is displayed (panel E). Summation of experiments performed as outlined above is shown (panel F). Cells were incubated with 100 nM insulin in the presence or absence of 10 M LY294002, and the effects of insulin and this PI3-kinase inhibitor were measured on isoproterenol-stimulated cyclic AMP accumulation measured over 6 min. Experiments performed with LY294002 inhibitor alone were incubated for 60 min. The results presented are the mean values Ϯ S.E. of at least four separate experiments.

FIG. 2. Counterregulation of cyclic AMP signaling and ␤2AR
sequestration in response to insulin in human epidermoid carcinoma A431 cells. Time course of epifluorescence images of sequestration of GFP-tagged ␤2AR expressed in A431 cells (panel A). Cells were challenged with 100 nM insulin, and the sequestration of ␤2AR was examined at the times indicated. Cell membrane-associated ␤2AR are noted with white arrows; the sequestered ␤2AR are noted by the yellow arrowheads. The ability of insulin to counterregulated isoproterenol-stimulated cyclic AMP accumulation (panel B) was performed as described in the legend to Fig. 1. Insulin suppresses the isoproterenol-stimulated cyclic AMP response and the presence of the PI3-kinase inhibitor blocked the suppression. Parallel experiments were performed in A431 cells stably expressing GFP-tagged ␤2AR (panel C). Note that insulin stimulates ␤2AR sequestration and LY294002 blocks the insulin-stimulated sequestration. The results displayed are sample images, representative of more than four separate experiments. AMP content measured by the competitive binding assay, as described (11). To assay insulin counterregulation, cells were pretreated with or without LY294002 for 30 min and then challenged with 100 nM insulin for 15 min.

MALDI Time-of-Flight Mass Spectrometry for Analysis of Receptor
Phosphorylation-Confluent cultures of A431 cells were serum-starved for 12 h and then treated without and with 100 nM insulin for 15 min. In each case, cultures were pooled from five 100 mm Petri dishes. The FIG. 3. Phosphorylation of ␤2AR by insulin in vivo occurs at Ser, Thr, and Tyr residues: identification of sites of phosphorylation by Akt. Panel A, CHO-K cells were transiently transfected with HA-tagged ␤2AR adrenergic receptor and/or co-transfected with constitutively active mutant (CA-Akt) of Akt. After serum starvation for 12 h, cells were treated with 0.1 M insulin for 15 min. HA⅐␤2AR was isolated from cell lysate by immunoprecipitation with HA-specific antibodies, subjected to SDS-polyacrylamide gel electrophoresis, transferred to nitrocellulose blots, and stained with antibodies specific for phosphoserine (pS), phosphothreonine (pT), phosphotyrosine (pY) or ␤2AR (CM-2). The results displayed are sample images, representative of more than four separate experiments. Panel B, in vitro phosphorylation of a GST fusion protein with the C-terminal, cytoplasmic domain of ␤2AR (C1). GST fusion proteins with progressively truncated C-terminal cytoplasmic domains of ␤2AR (C4 -C6) or the GST itself were incubated with recombinant, activated Akt. The labeled products were separated in 10% SDS-polyacrylamide gels and visualized by autoradiography. Protein constructs included the full-length C-terminal domain of ␤2AR (C1, Arg 328 -Leu 413 fragment), Arg 328 -Tyr 366 fragment (C4), Arg 328 -Asn 352 fragment (C5), Arg 328 -Cys 341 fragment (C6) and GST itself. Arrows show positions of C-terminal domain phosphorylation by Akt, as deduced by the studies. The results displayed are sample images, representative of more than four separate experiments. Panels C and D, mass spectrometry of tryptic digests of ␤2AR obtained from A431 cells that were untreated or treated with 100 nM insulin for 15 min. The HA-tagged ␤2AR were isolated from the cells and subjected to tryptic digestion (9,10), followed by analysis of the peptides on a ABI voyager research-grade MALDI mass spectrometer operated in the linear mode (panel C) or on the linear versus reflectron mode (panel D).
cells were lysed and the ␤2AR subjected to immunoprecipitation with antibody CM04, using 25 g of antibody per 0.1 ml of A/G agarose. Immunoprecipitates were digested with 1 g of trypsin for 8 h at room temperature. The phosphate-containing peptides were isolated on Fe 3ϩ columns (0.1 ml volume) and then analyzed on an ABI Voyager DE-STR mass spectrometer using ␣-cyano-4-hydroxycinnamic acid as the matrix.

RESULTS AND DISCUSSION
We measured the ability of insulin to counterregulate ␤-adrenergic-stimulated cyclic AMP in CHO clones expressing ϳ30,000 ␤-adrenergic receptors per cell (Fig. 1). Isoproterenol (10 M) stimulated rapid accumulation of intracellular cyclic AMP (Fig. 1A). Treatment with insulin (100 nM) 15 min in advance of challenge with the ␤-adrenergic agonist inhibits the cyclic AMP response of the cells to stimulation by the ␤-adrenergic agonist isoproterenol, reducing the response by 50% (Fig. 1B). PI3-kinase has been shown to be an obligate, early downstream element in insulin signaling (3). We probed the possible role of PI3-kinase in insulin counterregulation of catecholamine-stimulated cyclic AMP accumulation using the PI3-kinase inhibitor LY294002. Treatment with LY294002 (10 M) alone had little effect on ␤-adrenergic stimulation of intracellular cyclic AMP accumulation (Fig. 1C); whereas it abolished the counterregulatory effect of insulin on cyclic AMP accumulation in response to isoproterenol (Fig. 1D), suggesting that the counterregulatory effects of insulin on catecholamine action share PI3-kinase activation upstream. The time course for insulin-stimulated counterregulation of ␤-adrenergic stimulation of cyclic AMP accumulation was found to be half-maximal within 5 min (Fig. 1E). Based upon the summation of the experiments, insulin is shown to counterregulate ␤-adrenergicstimulated cyclic AMP accumulation and in a manner that is sensitive to blockade by the LY294002 inhibitor of PI3-kinase (Fig. 1F).
Using a GFP-tagged version of the human ␤2AR, we examined the influence of insulin on receptor sequestration in human epidermoid carcinoma A431 cells in which endogenous ␤2AR expression is ϳ35,000 receptors per cell, Ͻ10% of which are GFP-tagged (Fig. 2). A431 cells have been employed for analysis of ␤2AR sequestration, providing an optimal cell type for epifluorescence microscopy (17,20). The time course for insulin-stimulated sequestration of ␤2AR was rapid, demonstrable intracellular accumulation (yellow arrowheads) and cell membrane thinning (white arrows) of ␤2AR observed within minutes of stimulation by insulin ( Fig. 2A). Insulin-stimulated counterregulation of ␤2AR was examined in the absence and the presence of the LY294002 inhibitor (Fig. 2B). The A431 cells display the same counterregulatory effects of insulin on ␤2AR signaling as was observed in the CHO clones. Epifluorescence studies of GFP-tagged ␤2AR in A431 cells revealed that in the presence of insulin, ␤2ARs underwent a prominent sequestration, internalizing from the cell membrane (white arrows) into perinuclear regions of the cells (Fig. 2C, yellow  arrowheads). The magnitude of the insulin-stimulated sequestration reaches Ͼ60% of the cellular complement of ␤2AR and is of greater magnitude than that stimulated by ␤-adrenergic agonist-induced sequestration (14). Treating the cells with LY294002, which abolishes the ability of insulin to inhibit catecholamine-stimulated cyclic AMP response, also abolished the ability of insulin to provoke the internalization of ␤2AR (Figs. 1 and 2). Treatment with the LY294002 compound alone produced no significant effect on the cyclic AMP response (Figs. 1F and 2B) or on the localization of ␤2AR (Fig. 2C).
Because activation of PI3-kinase results in the activation of downstream serine-threonine protein kinases, like Akt, we exmined the phosphorylation state of the ␤2AR (Fig. 3). An HA-tagged version of the human ␤2AR was expressed in CHO cells, and the cells were treated with insulin (Fig. 3A). Treatment with insulin (100 nM) resulted in increased phosphorylation of the ␤2AR, as detected with phosphoserine-, phosphothreonine-, and phosphotyrosine-specific antibodies. The ␤2AR has been shown both in vitro and in vivo to be a substrate for tyrosine phosphorylation catalyzed by the insulin receptor itself (9), confirmed herein (Fig. 3A) (10,11). Increases in phosphoserine and phosphothreonine content of the ␤2AR in response to insulin had been noted earlier (9), but the protein kinase(s) responsible for this phosphorylation remained elusive. To test further the nature of the serine phosphorylation of the ␤2AR, we investigated the effects of expression of a constitutively activated mutant of Akt (T308D/S473D; CA-Akt) on the phosphorylation state of the ␤2AR. Co-expression of CA-Akt resulted in increased phosphorylation of ␤2AR that mimicked that observed in cells stimulated by insulin (Fig. 3A). Thus, stimulation of cells with insulin or expression of CA-Akt leads to increased phosphoserine content of the ␤2AR.
The desensitization and sequestration of ␤2AR in response to the ␤-adrenergic agonist is mediated largely by phosphorylation of the receptor on its C-terminal, cytoplasmic domain (15,16). We hypothesized that Akt might be playing a role similar to GRK and/or protein kinase A on this domain downstream of insulin activation of PI3-kinase. Akt is a serine-threonine-specific protein kinase activated in response to the activation of PI3-kinase, so we examined the ␤2AR sequence for a motif found in other Akt substrates (21,22). Embedded in the Cterminal, cytoplasmic domain of the ␤2AR is a consensus site for Akt phosphorylation (21), namely Cys 341 -Asn 352 that contains RRSSLKAY (Fig. 3B).
We tested this hypothesis in vitro by preparing a series of GST-tagged proteins corresponding to the C-terminal domain of the ␤2AR progressively truncated from the C terminus (23). The full-length, C-terminal domain of the ␤2AR (C1) has been shown to be readily phosphorylated by protein kinase A, GRK2, and the insulin receptor tyrosine kinase (24). We used this in vitro approach with purified, activated Akt to establish the sites of ␤2AR phosphorylation and test if the canonical motif for Akt was, in fact, a substrate (Fig. 3B). The peptide corresponding to the full-length, C-terminal cytoplasmic domain (GST⅐Arg 328 -Leu 413 ) of the ␤2AR was a substrate for Aktcatalyzed phosphorylation. Deletion mutants of the C-terminal domain of the ␤2AR were prepared (C4, C5, and C6) and employed in the phosphorylation reaction. Peptides C4 (Arg 328 -Tyr 366 ) and C5 (Arg 328 -Asn 352 ), like the C1 protein itself, were phosphorylated by Akt. Deletion of Leu 342 -Leu 413 (C6) of the ␤2AR C-terminal domain, in contrast, resulted in the loss of phosphorylation by Akt. These results provide evidence that the RRSSLKAY motif in the C-terminal domain of the ␤2AR is phosphorylated by Akt. The phosphoamino acid analysis performed by immunoblotting (Fig. 3A), likewise confirms the results that Akt phosphorylates serine residues of the ␤2AR.
Having established the ability of the ␤2AR to act as a substrate for phosphorylation by Akt in vitro, we used this information to ascertain the phosphorylation of this site of the ␤2AR in vivo. To determine whether the canonical site for Akt phosphorylation indeed was phosphorylated in response to insulin, we stimulated A431 cells with insulin, isolated the ␤2AR, and determined the phosphorylation state of the receptor peptide fragment that harbors the canonical site. ␤2AR from A431 cells treated with and without insulin (100 nM, 30 min) were isolated, digested with trypsin, and subject to MALDI mass spec- The results displayed are sample images, representative of more than four separate experiments. Panel C, parallel experiments were performed using epifluorescence microscopy and clones expressing a GFP-tagged 5HT1c serotonin receptor for comparison. Cells expressing the 5HT2a receptor were untreated (Control) or stimulated with 100 nM insulin (ϩIns) or with 10 M isoproterenol (ϩIso). Note that unlike the ␤2AR, the 5HT2a receptors fail to demonstrate receptor sequestration in response to either insulin or the ␤2AR agonist. These results displayed are sample images, representing three independent experiments. Receptor phosphorylation was assayed following receptor immunoprecipitation from whole cell lysates using HA-specific antibodies. The immunoprecipitates were subjected to immunoblotting and stained with antibodies against phosphoserine (pS), phosphothreonine (pT), or phosphotyrosine (pY), as described in legend for Fig. 2. The results displayed are sample images, representative of more than four separate experiments. trometry (Fig. 3C). The analysis identified the phosphorylated peptide SSLK (495.6 m/z, denoted by asterisk). The amount of the phosphorylated peptide was increased in tryptic digests from ␤2AR isolated from the insulin-treated as compared with untreated cells (Fig. 3C). We employed mass spectrometry to test if the mass change was due to phosphorylation by comparing the spectrum of the mass analysis performed in the linear compared with the reflector mode. Serine-phosphorylated peptides are easily lost to in-source (25) and post-source (26) decay. This behavior is diagnostic for serine-phosphorylated peptides and can be detected by their differential loss in the reflectron mode as compared with the linear mode of detection of MALDI-TOF mass spectrometry (25)(26)(27)(28)(29). Subject the same sample to either mode revealed the loss of the 495.6 m/z species in the reflectron mode (peptide 1), indicating the lability of the phosphate group. The dehydroalanine derivative, displaying 401.3 m/z (Fig. 3D, peptide 2), is the product of the decay of the phosphorylated precursor peptide phospho-SSLK and was detected in both the linear and reflectron mode of MALDI mass spectrometry. Thus by phosphorylation studies in vitro as well as by phosphoserine analysis and by mass spectrometry in vivo we demonstrate the phosphorylation of a canonical Akt site in the ␤2AR in response to insulin stimulation.
It was important to ascertain if the Ser 345 , Ser 346 residues of the ␤2AR were critical to the ability of insulin to counterregulate the cyclic AMP response of the cells to isoproterenol stimulation. We prepared the S345A/S346A (AA) double substitutions in the ␤2AR and examined the ability of insulin to inhibit the cyclic AMP response and induce ␤2AR sequestration (Fig.  4). The S345A/S346A double mutant ␤2AR no longer demonstrates sensitivity to insulin with respect to isoproterenol-stimulated cyclic AMP accumulation (Fig. 4A). The AA mutant ␤2AR displayed normal activation in response to isoproterenol (data not shown). Similarly, the AA mutant ␤2AR was not sequestered in response to insulin (Fig. 4B). The sequestration observed in response to insulin stimulation for ␤2AR (Figs. 2A and 4B) was tested for another member of the superfamily of G-protein-coupled receptors, the 5HT2a serotonin receptor. The 5HT2a receptor was not sequestered either by 100 nM insulin (ϩIns) stimulation or stimulation with the ␤2AR agonist, isoproterenol (Fig. 4C, ϩIso). Thus, the ␤2AR receptor is selectively sequestered in response to insulin stimulation, and the serine residues that can act as substrates for Akt-catalyzed phosphorylation are essential for the counterregulatory effects of insulin on the ␤2AR. We gathered additional data on the phosphorylation of Ser 345 , Ser 346 sites of the ␤2AR by study of the receptors in vivo. Cells expressing the wild-type (WT) and the S345A/S346A (AA) double mutant form of the ␤2ARs were treated with insulin (100 nM), and the phosphorylation state of the receptors were established through immunoblotting with phosphoamino acid-specific antibodies, as outlined above. Whereas the wildtype receptor displayed increased protein phosphorylation, the AA mutant failed to show increased phosphoserine content in response to insulin (Fig. 5). Surprisingly, phosphotyrosine content of the AA double mutant receptor did not increase in response to insulin. These data suggest the possibility that the phosphorylation of the Ser 345 , Ser 346 sites and the Tyr 350 site may be hierarchical, linked in a manner whereby the phosphorylation of the Ser 345 , Ser 346 sites influences phosphorylation of the Tyr 350 residue.
We further probed whether Akt mediates the action of insulin on the counterregulation of the ␤2AR through the use of a constitutively active form of Akt (T308D/S473D; CA-Akt). If Akt is a mediator of insulin action in this signaling pathway, we might expect that activation of Akt would be insulinomimetic (Fig. 6). Immunoblotting of cell extracts from CHO and A431 clones transfected with empty vector or the CA-Akt mutant revealed enhanced immunoreactivity for the clones expressing a mutant form of Akt (Fig. 6A). Expression of the CA-Akt in these cells provoked a reduction in isoproterenolstimulated cyclic AMP accumulation, much like insulin treatment (Fig. 6B). The addition of insulin (100 nM) to the cells expressing the CA-Akt produced no further reduction in the isoproterenol-stimulated cyclic AMP response. Complementary studies of ␤2AR localization revealed marked internalization of ␤2AR in cells expressing the CA-Akt, resembling the situation noted when the cells were treated with insulin. Addition of insulin failed to alter significantly the marked internalization of ␤2AR observed in the cells expressing CA-Akt (Fig. 6C). These data suggest that with respect to the counterregulatory effects of insulin on ␤2AR action, CA-Akt is fully insulinomimetic.
We made use of a dominant-negative strategy to test further the role of Akt on the counterregulatory influence of insulin on ␤2AR (Fig. 7). The mutant form (K179A/T308A/ S473A) of Akt (DN-Akt) was expressed in the cells and the ability of insulin to inhibit ␤-adrenergic stimulation of cyclic AMP accumulation (Fig. 7A) and ␤2AR sequestration (Fig.  7B) measured. Expression of DN-Akt itself resulted in a small reduction in the cyclic AMP response to isoproterenol (Fig. 6A). For cells expressing the DN-Akt, the ability of insulin to counterregulate the ␤-adrenergic cyclic AMP response was abolished. Moreover, in the cells expressing DN-Akt, not only was the counterregulation by insulin abolished, but also treatment with insulin produced a significant increase in the cyclic AMP response. Analysis of ␤2AR localization in cells expressing DN-Akt revealed two novel features (Fig. 7B). In comparison with the control cells, the cells expressing DN-Akt displayed somewhat more intracellular GFP-tagged ␤2AR (arrowheads) in the unstimulated state. This observation is not at all unexpected, because Akt is known to influence the trafficking of membrane proteins, such as GLUT4 glucose transporter. The DN-Akt may well suppress the counter movement of ␤2AR to the cell membrane from intracellular compartment. Surface GFP-tagged ␤2AR were also readily observed (arrows) in the cells expressing DN-Akt, but unlike the control cells, these cells did not display marked sequestration of ␤2AR from the cell membrane in response to insulin. Thus, expression of DN-Akt abolished the ability of insulin to counterregulate the ␤-ad-renergic-stimulated cyclic AMP response as well as the ability of insulin to internalize ␤2AR.
The current studies illuminate a novel role of Akt mediating insulin counteraction of catecholamine action. Earlier observations demonstrated the following: insulin stimulates inactivation of the ␤2AR through tyrosyl phosphorylation (10); phosphorylation of ␤2AR Tyr 350 in response to insulin creates a SH2 binding site with which Grb2, dynamin, and the p85 subunit of PI3-kinase can interact (9); and insulin stimulates marked sequestration of the ␤2AR, which requires the integrity of Tyr 350 and activation of PI3-kinase (14). Herein we show that Akt plays an obligate role for insulin signaling to the ␤2AR. Elimination of an Akt phosphorylation motif renders the ␤2AR insensitive to counterregulation by insulin and unable to be sequestered in response to insulin. Expression of dominant-negative Akt blocks the ability of insulin to sequester ␤2AR, whereas expression of a constitutively active Akt mimics the effects of insulin on ␤2AR sequestration. Akt is well known to regulate endosomal trafficking, acting in concert with EEA1 and Rab5 (3). Akt activation results in a shuttling of the insulin-sensitive GLUT4 transporter-laden vesicles to the cell membrane. In a similar manner, Akt may be causing shuttling of newly formed, phospho-␤2AR-laden vesicles to intracellular locales. For agonist-stimulated sequestration of ␤2AR, both the serine-threonine kinases, protein kinase A and GRK, function to traffic receptors from the cell membrane to endosomes for re-cycling or degradation (15). For insulin-stimulated sequestration, Akt appears to play an analogous role trafficking ␤2AR to intracellular locales (Fig. 8). Insulin stimulates its receptor to autophosphorylate, phosphorylate the ␤2AR at Tyr 350 , Tyr 354 , and Tyr 360 , as well as IRS1,2, to thereby activate PI3-kinase activity. Activation of PI3-kinase leads to activation of PDK1 and downstream to activation of Akt. Akt, in turn, phosphorylates the ␤2AR on an Akt canonical site located in the cytoplasmic, C-terminal tail and provokes the sequestration and internalization of the ␤2AR. The interplay between PI3-kinase, Akt, and endosomal elements with the ␤2AR seems to be quite complex and will require further understanding of the intracellular trafficking pathways.