Role of the Cyclic AMP-dependent Protein Kinase in Homologous Resensitization of the (cid:1) 1 -Adrenergic Receptor*

A fundamental question in biology is how the various motifs in G protein-coupled receptors participate in the divergent functions orchestrated by these molecules. Here we describe a fundamental role for a serine residue at position 312 in the third intracellular loop of the human (cid:1) 1 -adrenergic receptor ( (cid:1) 1 -AR) in endocytic re- cycling of the agonist-internalized receptor. In receptor recycling experiments that were monitored by confocal microscopy, the agonist-internalized wild-type (WT) (cid:1) 1 -AR recycled with a t 0.5 of 14 (cid:2) 3 min. Mutagenesis of Ser 312 to alanine (Ser 312 3 Ala (cid:1) 1 -AR) or to the phosphoserine mimic aspartic acid (Ser 312 3 Asp (cid:1) 1 -AR) resulted in (cid:1) 1 -AR constructs that were pharmacologically indis- tinguishable from the WT (cid:1) 1 -AR. The internalized Ser 312 3 Asp (cid:1) 1 -AR recycled efficiently with a t 0.5 of 11 (cid:2) 3 min, whereas the internalized Ser 312 3 Ala (cid:1) 1 -AR was not recycled or functionally resensitized through the endo-somal pathway. Because this serine is a putative residue for phosphorylation by the cyclic AMP-dependent protein kinase (PKA), we examined the role of this kinase in recycling of the

The ␤ 1 -AR 1 mediates many of the cardiovascular actions of catecholamines such as the regulation of heart rate and the force of myocardial contraction (1,2). Activation of the ␤ 1 -AR and other GPCR in turn causes marked changes in the receptor protein and its associated signaling components (reviewed in Ref. 3). In the case of GPCR, the activated receptor becomes a substrate for modification by specific kinases that serve to uncouple the receptor from the G protein (4). One such example is homologous desensitization of the ␤ 2 -AR by ␤-agonists, whereby ligand-dependent phosphorylation of the ␤ 2 -AR by G protein-coupled receptor kinases (GRKs) is rapidly followed by receptor interaction with cytoplasmic ␤-arrestins that disrupt the interaction between the receptor and the G protein (5). This interaction presumably occurs in the cell membrane and causes rapid desensitization of the receptor and is intimately associated with endocytosis of the ligand-activated receptor via the clathrin-coated pit pathway (5)(6)(7)(8).
Initially, internalization of the GPCR was viewed as a means to uncouple the receptor from its signaling components, thereby dampening the overall response (9 -12). The results of many studies indicate that the itinerary of the internalized GPCR is receptor-and cell-specific (13). For example, in human embryonic kidney (HEK-293) cells, the -opioid receptor is internalized and then recycled, whereas the internalized Ѩ-opioid receptor is significantly degraded (14 -16). Cell-specific outcomes have been encountered with the ␤ 2 -AR, which is recycled in HEK-293 cells but degraded in A-431 cells (17,18). Intracellular trafficking for some GPCR, therefore, appears to promote their resensitization and recovery from desensitization (19 -23). However, little is known about the molecular mechanisms or the motifs within the GPCR that are involved in its various outcomes. The carboxyl-terminal tails of the ␤ 1 -AR and the ␤ 2 -AR have distinct PDZ-like domains. The PDZ-like domain of the ␤ 1 -AR interacts with PSD-95 (postsynaptic associated protein 90) and membrane-associated guanylate-inverted-2 (24,25), whereas the PDZ domain of the ␤ 2 -AR interacts with the Na ϩ /H ϩ exchanger regulator factor (NHERF; also known as EBP50 (ezrin-binding protein 50)) (26). Mutagenesis of the PDZ of the ␤ 2 -AR interferes with its efficient recycling and its functional properties (27). The motifs involved in regulating the recycling of the ␤ 1 -AR are more obscure. Our results demonstrate that a critical serine in the third intracellular loop of the human ␤ 1 -AR that is a substrate for reversible phosphorylation by PKA is involved in recycling and functional resensitization of the human ␤ 1 -AR.

Measurement of ␤ 1 -AR Cell Surface Expression by Radioligand Bind-
ing Assays-Membranous ␤ 1 -AR expression in HEK-293 cells stably transfected with FLAG-tagged wild-type or the Ser 312 point mutants was evaluated with 125 I-cyanopindolol binding studies performed at 20°C for 2 h. Cell lines expressing equivalent amounts of ␤ 1 -AR were used in these studies. The densities of the ␤ 1 -AR in these cell lines were 0.95 Ϯ 0.12 pmol of WT ␤ 1 -AR/mg, 1.1 Ϯ 0.2 pmol of Ser 312 3 Ala ␤ 1 -AR/mg, and 0.85 Ϯ 0.1 pmol of Ser 312 3 Asp ␤ 1 -AR/mg of protein.
Determination of ␤ 1 -AR Surface Expression by Biotinylation-HEK-293 cells expressing the WT ␤ 1 -AR or the Ser 312 3 Ala ␤ 1 -AR were exposed to buffer or to 10 M isoproterenol for 15 min at 37°C to induce receptor internalization. Then the medium containing isoproterenol was replaced with warm culture medium containing a 100 M concentration of the ␤-adrenergic receptor antagonist alprenolol to block further internalization in order to compare unambiguously the effect of mutagenesis of Ser 312 to Ala on recycling. After 3 h at 37°C, the medium was aspirated, and the cells were surface-biotinylated with 1.5 mg/ml sulfo-NHS-SS-biotin in Hanks' balanced salt solution with Ca 2ϩ and Mg 2ϩ (HBSS) at 4°C (30). The unreacted biotin was quenched with ice-cold 50 mM glycine in HBSS, and the cells were scraped into cold lysis buffer (50 mM Tris, pH 7.4, 2 mM EDTA, 2 mM EGTA, plus phosphatase inhibitors (50 mM NaF, 10 mM sodium pyrophosphate, 20 mM ␤-glycerophosphate, 1 mM p-nitrophenylphosphate, 1 mM Na 3 VO 4 , and 0.1 mM NH 4 MoO 3 ) and protease inhibitors (10 units/ml aprotinin, 0.1 mM phenylmethylsulfonyl fluoride, 0.1 mg/ml benzamidine plus 10 g/ml each of chymostatin, pepstatin, antipain, and leupeptin)). The homogenate was sonicated and then centrifuged at 100,000 ϫ g av for 20 min at 2°C. The membrane pellet was dissolved in lysis buffer (precipitation buffer supplemented with 100 mM NaCl, 0.2% SDS, and 1% Triton X-100) and recentrifuged at 100,000 ϫ g av for 20 min at 2°C. The resulting supernatant was collected, and equal amounts of protein from all samples were mixed with 50 l of bovine serum albumin-blocked ultralink-neutra avidin beads (Pierce) at 4°C overnight. The resin was collected by centrifugation, washed several times with lysis buffer without SDS, and then extracted with 10 l/100 g of input protein of 2ϫ Laemmli sample buffer with 20 mM dithiothreitol at 37°C for 40 min. The supernatant was subjected to electrophoresis on SDS-containing 4 -12% gels, transferred to nitrocellulose, and probed with the anti-FLAG antibody. The density of the M r 73,000 ␤ 1 -AR monomer from each condition was quantified using a Chemi DOC XRS densitometer (Bio-Rad) equipped with Quantity One software.
Biotinylation Assay of ␤ 1 -AR Recycling with Cleavable Biotin-␤ 1 -AR recycling was measured by the loss of internalized ␤ 1 -AR prelabeled with sulfo-NHS-SS-biotin. In this assay, cells expressing the WT ␤ 1 -AR or the Ser 312 3 Ala ␤ 1 -AR were pretreated with 100 g/ml of the lysosomal protease inhibitor leupeptin, beginning 1 h prior to biotinylation. The cells were biotinylated on ice for 20 min, and the excess biotin was quenched with glycine. Biotinylated cells were exposed to isoproterenol for 30 min and then cooled to 4°C to stop membrane trafficking, and the remaining surface biotin was quantitatively cleaved with glutathione cleavage buffer (50 mM glutathione in 75 mM NaCl and 10 mM EDTA containing 1% bovine serum albumin and 0.075 N NaOH) twice for 15 min at 4°C. Cultures were then returned to the culture medium at 37°C for 1, 12, and 24 h to allow internalized receptor to recycle before the cells were cooled to 4°C and incubated with glutathione cleavage buffer (twice for 15 min at 4°C) to ensure complete cleavage of any newly appearing surface biotin. Residual biotinylated (internalized) receptors were then isolated from cell lysates as described above.
Adenylyl Cyclase Assays for ␤-AR Desensitization and Resensitization-HEK 293 cells stably expressing the WT ␤ 1 -AR, Ser 312 3 Ala ␤ 1 -AR, and Ser 312 3 Asp ␤ 1 -AR were incubated for 1 h in serum-free DMEM supplemented with 10 mM HEPES prior to assay. Four identical sets of cells were set up; the first set was used as a control for desensitization, and the second set was used for resensitization assays. Cells for desensitization were exposed to 1 mM ascorbic acid (control) or 10 M isoproterenol for 10 min at 37°C and then washed with serum-free DMEM supplemented with 10 mM HEPES and processed for the preparation of membranes as described below. The third set was used as the control for resensitization, and the fourth set was used for resensitization assays. Cells for resensitization were exposed either to 1 mM ascorbic acid (control) or to 10 M isoproterenol for 3 h at 37°C and then washed with serum-free DMEM supplemented with 10 mM HEPES. The cells were then incubated with 100 M alprenolol for 1 h at 37°C and then washed with serum-free DMEM supplemented with 10 mM HEPES. Membranes were prepared from all cells by hypotonic lysis of the cells with 20 mM HEPES, pH 7.4, 2 mM MgCl 2 , 1 mM EDTA, and 1 mM 2-mercaptoethanol supplemented with 10 g/ml leupeptin and 10 g/ml aprotinin for 10 min on ice. The cells were transferred into a glass-glass homogenizer and lysed by 30 up-and-down strokes. Cell lysates were centrifuged at 2,500 ϫ g av for 5 min to pellet the nuclei, and the supernatant was centrifuged at 15,000 ϫ g av for 20 min to pellet the membranes. Then 50 g of membrane proteins were incubated at 30°C in a final volume of 0.1 ml containing 50 mM Tris-HCl, pH 7.4, 1 mM MgCl 2 , 10 mM phosphocreatine, 1 mM cyclic AMP, 2 mM 2-mercaptoethanol, 1 mg/ml bovine serum albumin, 0.4 mM EGTA, 2 mg/ml creatine kinase, 0.2 mM ATP containing 1 Ci of [␣-32 P]ATP, 1 mM GTP, and the various concentrations of isoproterenol. The assay was initiated by the addition of membranes and terminated after 10 min (31). The cyclic AMP that formed was isolated by column chromatography and quantified by liquid scintillation counting (31). Assays were performed in triplicate and replicated n ϭ 4 times. The K act Ϯ S.E. for each ␤ 1 -AR was calculated using the Graphpad Prism 4 program and statistical comparisons were analyzed using Graphpad Instat program.
Confocal Microscopy-HEK-293 cells stably transfected with FLAGtagged WT ␤ 1 -AR or its Ser 312 mutants were grown on poly-L-lysinecovered glass coverslips and serum-starved at 37°C for 1 h in DMEM supplemented with 25 mM HEPES, pH 7.4. The receptors were labeled with the M2 anti-FLAG fluorescein isothiocyanate-conjugated antibody (0.5 g/ml) for 1 h. Cells were treated with 10 M isoproterenol for 30 min, followed by an acid wash with (0.2 M NaCl, 0.5 M acetic acid) to strip off the excess of antibodies. Then 100 M alprenolol was added, and the coverslips were fixed with 4% paraformaldehyde in PBS, pH 7.4, for 10 min at room temperature at different time points to establish the recycle time. In those experiments where the slides were fixed after 1 h from the removal of isoproterenol, the slides were pretreated with 100 g/ml of the lysosomal inhibitor leupeptin for 1 h in serum-free medium.
In another series of experiments, a 1 M concentration of the specific PKA inhibitor H-89 (32) and 1 M concentration of the myristoylated PKA peptide inhibitor (PKI), TTYADFIASGRTGRRNAIHD (33), that inhibits phosphorylation of target proteins by binding to the proteinsubstrate site of the catalytic subunit of PKA were added to cells prior to isoproterenol.
Confocal fluorescence microscopy was performed using Zeiss Axiovert LSM 510 (100 ϫ 1.4 DIC oil immersion objective). Fluorescein isothiocyanate was excited with the 488-nm argon-krypton laser and imaged through the 520-nm long-pass emission filter. Z-stacks of images were exported as TIFF files, and individual sections were analyzed with Zeiss LSM 510 and NIH Image 1.6 software. A minimum of 10 images/panel was quantified. Confocal images were analyzed using the same threshold setting. The data are presented as the mean Ϯ S.E.
Generation of HEK-293 Cells Expressing Dominant Negative PKA-Two dominant negative PKA vectors, MT-REV AB and MT-REV AB -neo, were used in our studies (34). In both vectors, the metallothionine gene promoter drives the expression of a PKA RI␣ subunit mutant in which the two cyclic AMP-binding sites were mutagenized. MT-REV AB was used for transient expression, whereas MT-REV AB -neo was used to generate a stable construct in HEK-293 cells. The metallothionine promoter is induced by Zn 2ϩ ions and inhibited in the absence of Zn 2ϩ . The activity of PKA in these cells was determined by an in vitro kinase assay in which cyclic AMP-mediated phosphorylation of biotinylated Leu-Arg-Arg-Ala-Ser-Leu-Gly (Kemptide) was used to assess PKA activity (SignaTECT for PKA from Promega Corp.). An MT-REV AB -neoexpressing cell line was selected in which basal PKA activity was undetectable and cyclic AMP-mediated induction of the activity of PKA was 1.8 Ϯ 0.3-fold compared with about 20 Ϯ 4-fold in the cell line expressing the empty MT-neo vector.
The HEK-293 cells stably expressing the MT-REV AB -neo were grown on coverslips and transiently transfected with the WT ␤ 1 -AR for 2 days. The cells were then cultured with MEM supplemented with 10% dialyzed fetal bovine serum and 1 M ZnSO 4 overnight to induce MT-REV AB -neo expression or in the absence of zinc to repress MT-REV ABneo expression. Slides were processed for the recycling assay by confocal microscopy as described earlier.
Intact and in Vitro Cell Phosphorylation and MAPK Activity Assays-Intact cell phosphorylation of HEK-293 cells expressing the WT ␤ 1 -AR were performed at 37°C in phosphate-free DMEM supplemented with 25 mM HEPES, pH 7.4. The medium was supplemented with 350 Ci of 32 P i /ml (Amersham Biosciences) for 1 h with either 0.03% Me 2 SO or 1 M H-89. Stimulation with 10 M isoproterenol or with 20 M forskolin were for 10 min, followed by the addition of RIPA extraction buffer composed of 20 mM Tris-HCl, pH 8.0, 150 mM NaCl, 5 mM EDTA, 1% Triton X-100, 0.2% SDS, 1 mM phenylmethylsulfonyl fluoride, and 1 g/ml each of antipain, aprotinin, chymostatin, leupeptin, and pepstatin. Equivalent amounts of dissolved proteins were incubated with M2-FLAG-agarose beads (Sigma) at 4°C overnight. For in vitro cell phosphorylation, 50 g of membrane protein from HEK-293 cells stably expressing the WT ␤ 1 -AR or the Ser 312 3 Ala ␤1-AR were incubated in 200 l of PKA kinase buffer composed of 20 mM HEPES (pH 7.5), 10 mM MgCl 2 , 5 mM EGTA, 2 mg/ml sodium fluoride, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, and 1 g/ml each of antipain, aprotinin, chymostatin, leupeptin, and pepstatin. At the beginning of the experiment, the buffer was supplemented with 1 nM ATP and 0.3 mCi of [␥-32 P]ATP. To some samples, ascorbic acid or 10 M isoproterenol without or with 20 units of the catalytic subunit of PKA (Promega) were added and incubated for 10 min at 30°C. At the end of the incubation, the tubes were centrifuged, and the supernatant was discarded. The membrane pellets were dissolved in RIPA extraction buffer (35), and the cleared supernatant was collected. Equal amounts of proteins were mixed with 50 l of anti-M2-FLAG antibody coupled to agarose beads (Sigma) at 4°C overnight. The next day, the resins were washed in RIPA buffer, and the proteins were resolved by SDS-polyacrylamide gel electrophoresis in 10% gels. Dried gels were subjected to autoradiography and analyzed quantitatively with Packard Instantimager software.
HEK-293 cells were exposed to 50 ng/ml pertussis toxin for 24 h. Pertussis toxin-treated and untreated cells were grown in serum-free DMEM for 2 h and then exposed to 10 M isoproterenol for 5 min. Cell extracts were prepared by lysing the cells in MAPK lysis buffer composed of 20 mM Tris-HCl, pH 8, 137 mM NaCl, 0.75 mM MgCl 2 , 1 mM EGTA, 1% Triton X-100, 10% glycerol, 50 mM NaF, 1 mM sodium vanadate, 1 mM phenylmethylsulfonyl fluoride, 10 g/ml leupeptin, and 10 g/ml aprotinin. Equal amounts of proteins were subjected to SDSpolyacrylamide gel electrophoresis on 4 -12% gels, electroblotted to nitrocellulose, and probed with anti-phospho-Thr 202 /Tyr 204 P 42/44 MAPK (Cell Signaling Technology). The blots were stripped and reprobed with anti-P 42/44 MAPK antibody (Upstate Biotechnology Inc., Lake Placid, NY) to ascertain that equal amounts of protein were loaded onto the gels.

RESULTS AND DISCUSSION
Human WT ␤ 1 -AR expressed in HEK 293 cells that were exposed to isoproterenol for 30 min followed by agonist washout, exhibited rapid internalization and recycling (Fig. 1A). These data are consistent with the effect of isoproterenol on ␤ 2 -adrenergic receptors in this cell line (6). However, mutagenesis of the serine residue at position 312 in the third intracellular loop to alanine (Ser 312 3 Ala ␤ 1 -AR) prevented the recycling of the agonist-internalized receptor as determined by radioligand binding (Fig. 1A). Surface biotinylation of cells after 3 h from the removal of isoproterenol showed that the incorporation of biotin into the Ser 312 3 Ala ␤ 1 -AR was less than the WT ␤ 1 -AR (Fig. 1B). Functional assays were used to determine whether this mutation affected the coupling of the WT ␤ 1 -AR or the Ser 312 3 Ala ␤ 1 -AR to adenylyl cyclase in response to isoproterenol (Fig. 1C). Expression of the ␤ 1 -AR did not markedly increase basal activity of adenylyl cyclase when compared with its activity in cells expressing the empty expression pCDNA vector. Increasing concentrations of isoproterenol increased the activities of adenylyl cyclase by ϳ7-fold in each ␤ 1 -AR-expressing cell line with comparable coupling affinities ( Fig. 1C) (36).
To investigate the role of Ser 312 in the trafficking of the ␤ 1 -AR, we visualized membrane trafficking of antibody-labeled receptors using confocal microscopy. Both the WT ␤ 1 -AR and Ser 312 3 Ala ␤ 1 -AR underwent rapid endocytosis following their activation by isoproterenol, as indexed by the translocation of antibody labeled ␤ 1 -AR from the plasma membrane to endocytic vesicles (Fig. 2, b and j). Acid treatment of these cells removed the surface-exposed antibody and revealed that intracellular ␤ 1 -AR staining was in discrete punctate vesicular A, cells expressing either the WT ␤ 1 -AR or a mutant in which serine at position 312 in the third intracellular loop of the ␤ 1 -AR was converted to alanine were exposed to isoproterenol for 30 min to induce internalization. Then the cells were washed, and the distribution of recycled receptors over the next 8 h was assessed by radioligand binding using [ 3 H]CGP-12177. B, cells expressing the WT ␤ 1 -AR or the Ser 312 3 Ala ␤ 1 -AR were exposed to buffer or isoproterenol for 15 min to induce internalization and then washed extensively and exposed to alprenolol for 3 h. Cells were surface-biotinylated using sulfo-NHS-SS-biotin for 20 min at 4°C followed by immunoprecipitation and immunoblotting to quantify cell surface distribution of each ␤ 1 -AR. In C, isoproterenolmediated activation of adenylyl cyclase in membranes prepared from cells expressing the empty pCDNA 3.1 vector, WT ␤ 1 -AR, Ser 312 3 Ala ␤ 1 -AR, or Ser 312 3 Asp ␤ 1 -AR were compared. The K Act was 0.4 Ϯ 0.06 M for the WT ␤ 1 -AR, 0.12 Ϯ 0.05 M for the Ser 312 3 Ala ␤ 1 -AR and 0.8 Ϯ 0.06 for the Ser 312 3 Asp ␤ 1 -AR (p Ͼ 0.05, n ϭ 4). structures (Fig. 2, c and k). In a confocal recycling assay, involving initial exposure of the cells to isoproterenol, followed by an acid wash and then the addition of the ␤-adrenergic receptor antagonist alprenolol to prevent further ␤ 1 -AR internalization, it was revealed that the WT ␤ 1 -AR recycled rapidly and completely ( Fig. 2A, d-g). The t 0.5 for recycling of the WT ␤ 1 -AR was 14 Ϯ 3 min (Fig. 2B). Under the same experimental conditions, the agonist-internalized Ser 312 3 Ala ␤ 1 -AR did not recycle even after 24 h (Fig. 1A, l-p). The recycling rate of the phosphoserine mimic generated by substituting an aspartic acid for Ser (Ser 312 3 Asp ␤ 1 -AR) was comparable with that of the WT ␤ 1 -AR with a t 0.5 of 11 Ϯ 3 min (Fig. 2B).
To determine whether the recycled WT ␤ 1 -AR is reinserted into membranes and to verify the microscopy results in Fig. 2A, a biochemical recycling assay was performed (Fig. 3A). In this assay, cells expressing the WT ␤ 1 -AR or the Ser 312 3 Ala ␤ 1 -AR were surface-biotinylated with cleavable biotin and then exposed to isoproterenol or buffer for 30 min, followed by cleavage of the remaining cell surface biotin. The first cleavage step permits the analysis of the recycling kinetics of internalized biotinylated receptors without interference from de novo synthesized nonbiotinylated receptors. The data reveal complete cleavage of cell surface biotin from cells exposed to buffer only (compare lanes 1 and 2 in Fig. 3A). Exposure of the biotinylated cells to isoproterenol protected a significant amount of internal biotin from cleavage (compare lanes 3 and 4 in Fig. 3A). The biotinylated fraction of WT ␤ 1 -AR that was resistant to cleavage in lane 4 was less than the comparable fraction of Ser 312 3 312 3 Ala ␤ 1 -AR is impaired. Cells stably expressing either the WT ␤ 1 -AR, the Ser 312 3 Ala ␤ 1 -AR or the Ser 312 3 Asp ␤ 1 -AR were incubated with fluorescein isothiocyanate-anti-FLAG M2 antibody and then exposed to the appropriate agent, fixed, and visualized by confocal microscopy (A). In all of the cell lines, the ␤ 1 -AR was expressed on the cell surface of the cell (a, i, and q). Internalization into endocytic vesicles occurred in all cells that were exposed to 10 M isoproterenol for 15 min (b, j, and k), and this was confirmed by an acid wash that stripped the antibody from cell surface receptors (c, k, and s). Isoproterenol-treated cells were visualized after they were exposed to a 100 M concentration of the ␤-adrenergic antagonist alprenolol for up to 24 h. Receptor recycling was observed in cells expressing either the WT ␤ 1 -AR (d-g) or the Ser 312 3 Asp ␤ 1 -AR (t-w) but not the Ser 312 3 Ala ␤ 1 -AR even after 24 h (l-p). Each scale bar represents 5 m. B, quantification of ␤ 1 -AR recycling kinetics by confocal recycling assays. The LSM-510 software was used to determine the distribution of pixels between the membranous and intracellular compartments of acid washed cells. The ratios of membranous to intracellular pixels were determined for each time point after the washout of isoproterenol. The t 0.5 for recycling was calculated by fitting the relevant data to a single exponential function of time from y ϭ y o ϩ A(1 Ϫ e Ϫ1/ ), where y o and A are constants.

FIG. 2. Recycling and resensitization the Ser
Ala ␤ 1 -AR because some WT ␤ 1 -AR apparently recycled during the period when isoproterenol was present. After the first cleavage, the cells were returned to 37°C to allow the WT ␤ 1 -AR and the Ser 312 3 Ala ␤ 1 -AR to recycle for 1, 12, or 24 h. After each time period, the cells were cooled to 4°C and cleaved for the second time to ensure cleavage of any newly appearing surface biotin. The data reveal that by 1 h, no biotin was detected in the WT ␤ 1 -AR blots, reflecting membrane recycling of the WT ␤ 1 -AR and subsequent biotin cleavage (Fig. 3A, lanes  5-7). In contrast, the internalized (biotinylated) Ser 312 3 Ala ␤ 1 -AR was not changed after 1, 12, or 24 h from the addition of isoproterenol, reflecting their internal distribution (compare lanes 5-7 in Fig. 3A with lane 4). Importantly, the internalized (biotinylated) Ser 312 3 Ala ␤ 1 -AR experienced no detectable nonspecific loss of biotin or protein degradation under the experimental conditions used (i.e. with 100 g/ml leupeptin), indicating that the loss of biotin by the WT ␤ 1 -AR in lanes 5-7 of Fig. 3A was specifically due to membrane reinsertion.
Next we examined whether the differences in postendocytic trafficking between the WT ␤ 1 -AR or the Ser 312 3 Asp ␤ 1 -AR and the Ser 312 3 Ala ␤ 1 -AR were associated with significant effects on the ability of the Ser 312 3 Ala ␤ 1 -AR to be functionally resensitized after agonist stimulation of cells (Fig. 3B). In this assay, rapid desensitization of adenylyl cyclase in membranes expressing all three ␤ 1 -AR was observed after a 10-min exposure of cells to isoproterenol, indicating that mutagenesis of Ser at position 312 did not affect short term desensitization (37). In the next series of experiments, the cells were exposed to isoproterenol for 3 h to desensitize and internalize the ␤ 1 -AR, followed by agonist washout and recycling in the presence of alprenolol for 1 h. In this assay, we observed significant differences between the cell lines in their recovery from desensitization. WT ␤ 1 -AR and Ser 312 3 Asp ␤ 1 -AR cells recovered from desensitization and their activation of adenylyl cyclase in response to isoproterenol were comparable with control cells, whereas Ser 312 3 Ala ␤ 1 -AR-expressing cells were significantly desensitized under the same conditions (Fig. 3B). Thus, the modification of Ser at position 312 to alanine apparently disrupted the recycling itinerary involved in functional resensitization of this receptor.
The sequence around the serine 312 residue in the human ␤ 1 -AR is RRPS 312 , which corresponds to a putative site for phosphorylation by PKA (38). To investigate whether phosphorylation of the ␤ 1 -AR by PKA is involved in recycling of the internalized ␤ 1 -AR, we measured the recycling kinetics of the agonist-internalized WT ␤ 1 -AR under conditions in which PKA was inhibited. Inhibition of PKA chemically with its specific inhibitors H-89 (32) or with the cell-permeable myristoylated PKI (33) had no appreciable effect on agonist-induced endocytosis but prevented the recycling of the internalized ␤ 1 -AR (compare q and y to h in Fig. 4A). Because these agents might inadvertently inhibit other kinases (39), we used a dominant negative construct of the regulatory type 1 subunit (RI␣) of cyclic AMP-dependent protein kinase (34) to inhibit the activity of PKA in HEK-293 cells. This construct (MT-REV AB ) contains mutations in each of the two cyclic AMP binding sites in the RI␣ subunit to inactivate endogenous PKA or ␤ 1 -AR-stimulated PKA activity. Co-expression of the WT ␤ 1 -AR with MT-REV AB in the amounts described in Fig. 4C, completely inhibited basal PKA activity and reduced cyclic AMP-stimulated PKA activation in cell extracts by 96 Ϯ 3% (data not shown). Cell surface ␤ 1 -AR expression was assessed by [ 3 H]CGP-12177 binding to cells that were exposed to an initial 30-min pulse with isoproterenol followed by a 3-h washout (Fig. 4C). Coexpression of ␤ 1 -AR cDNA with the empty control vector resulted in 90 Ϯ 3% restoration of the original cell surface ␤ 1 -AR complement, whereas co-expression of ␤ 1 -AR cDNA with the MT-REV AB vector reduced cell surface expression by 62 Ϯ 5% (p Ͻ 0.01).
Recycling assays were carried out in a HEK-293 cell line stably expressing the MT-REV AB -neo construct in which the expression of the double mutant RI␣ subunit of PKA was under the control of the metallothionine promoter. Repression of MT-REV AB -neo expression in the absence of zinc ions, allowed the recycling of the internalized ␤ 1 -AR (dЈ-hЈ in Fig. 4A). Activation of MT-REV AB -neo expression by prior treatment of the cells with Zn 2ϩ ions prevented the efficient recycling of the agonist-internalized ␤ 1 -AR (kЈ-pЈ in Fig. 4A). Zn 2ϩ (1 M) had no effect on the recycling of the WT ␤ 1 -AR (data not shown). Two sets of culture dishes from each cell type were either cleaved or uncleaved to determine total biotinylation and cleavage efficiency for each cell line (lanes 1 and 2). The remaining cultures were exposed to isoproterenol for 30 min to induce internalization, and the remaining surface biotin was cleaved (first cleavage) with glutathione (lanes 3 and  4). The remaining cultures were returned to 37°C to allow each ␤ 1 -AR to recycle for 1, 12, and 24 h and then recleaved (second cleavage) to ensure cleavage of any newly appearing surface biotin. The loss of biotinylated WT ␤ 1 -AR after the second biotin cleavage in lanes 5-7 provides confirmation that the WT ␤ 1 -AR recycled rapidly, whereas retention of biotin in lanes 5-7 of the Ser 312 3 Ala ␤ 1 -AR samples indicates intracellular retention of this receptor. B, functional resensitization of the ␤ 1 -AR assessed by a membrane adenylyl cyclase assay. Cells expressing WT ␤ 1 -AR, Ser 312 3 Ala ␤ 1 -AR, and Ser 312 3 Asp ␤ 1 -AR were exposed to buffer or to 10 M isoproterenol for either 10 min or 3 h at 37°C. For the 3-h condition, the buffer or isoproterenol was replaced with 100 M of alprenolol for 1 h. Membranes were prepared from these cells by hypotonic lysis followed by differential centrifugation. Fifty g of membranes were exposed to 10 M isoproterenol or to 20 M forskolin in cyclase assay buffer for 10 min at 30°C. The ratio for the specific activity of adenylyl cyclase in response to isoproterenol to that for forskolin in each sample was determined to calculate the percentile of maximal adenylyl cyclase activity in each type of membrane. These experiments were replicated (n ϭ 6) each in triplicate to determine the S.E. Therefore, biochemical and genetic approaches indicate that PKA is involved in recycling of the internalized ␤ 1 -AR.
PKA-mediated phosphorylation of GPCR has been implicated in functional effects ranging from desensitization (40,41) or switching in G protein-coupling specificity from G s to G i (42). Therefore, we tested the effect of ablating or mimicking the phosphorylation state of Ser 312 on these parameters. Agonist-mediated phosphorylation of ERK1/2 in the three ␤ 1 -AR-expressing cell lines were comparable and were insensitive to pertussis toxin, suggesting that coupling of ␤ 1 -AR to G s was maintained (Fig. 5A). These data, when combined with those in Figs. 1-4 and (37) concerning the role of Ser 312 in desensitization and resensitization of the ␤ 1 -AR, indicate that the putative PKA phosphorylation site in the human ␤ 1 -AR is exclusively Confocal microscopy was used to visualize HEK-293 cells stably expressing the WT ␤ 1 -AR. Cells were serum-starved for 1 h and then incubated in the absence (Me 2 SO; DMSO) or presence of 1 M H-89 or myristoylated PKI for 1 h. Confocal recycling assays for the internalized ␤ 1 -AR were performed as described in the legend of Fig. 2. In Me 2 SO-treated cells, the WT ␤ 1 -AR recycled efficiently (e-h), but the ␤ 1 -AR did not recycle in cells exposed to either H-89 (n-q) or myristoylated PKI (v-y). B, the WT ␤ 1 -AR is recycled efficiently in Me 2 SO-treated cells (t 0.5 ϭ 14 Ϯ 6 min) but not in cells in which PKA was inhibited. C, the dominant negative PKA inhibitor (MT-REV AB ) was co-expressed with WT ␤ 1 -AR for 2 days. Cells in serum-free medium were exposed to 1 mM ascorbic acid or 10 M isoproterenol for 30 min at 37°C. The cells were washed and then returned to the incubator for 3 h. The distribution of the ␤ 1 -AR in intact cells was assessed by radioligand binding using [ 3 H]CGP-12177. The data represent the mean Ϯ S.E. of n ϭ 5 determinations, each in triplicate. In aЈ-pЈ of A, the WT ␤ 1 -AR cDNA was transfected into cells stably expressing the dominant negative PKA inhibitor (MT-REV AB -neo) for 24 h. Conditional activation of the mutant RI␣ subunit was accomplished by culturing the cells for 24 h in a zinc-free medium (aЈ-hЈ) to repress the expression of the mutant RI␣ subunit, or in a 1 M ZnSO 4 -containing medium to induce the expression of RI␣ subunit (iЈ-pЈ). Confocal recycling assays for the internalized ␤ 1 -AR were performed as described in the legend to Fig. 2. Conditional activation (ϩZn 2ϩ , mЈ-pЈ) of the dominant negative PKA inhibitor (MT-REV AB -neo) impaired endocytic recycling of the WT ␤ 1 -AR, whereas its repression (ϪZn 2ϩ , dЈ-hЈ) restored efficient recycling of these receptors. Each scale bar represents 5 m.
involved in trafficking of the internalized ␤ 1 -AR in this cell line.
If PKA is involved in recycling, its effect is achieved through phosphorylation of the ␤ 1 -AR in response to agonist activation. Robust phosphorylation of the ␤ 1 -AR was observed within 10 min of exposing the cells to isoproterenol (Fig. 5B). We have already determined that phosphorylation of the ␤ 1 -AR occurs exclusively on serine residues (35). Phosphorylation of the ␤ 1 -AR in response to forskolin, a general activator of all isoforms of adenylyl cyclase that activates PKA in a receptorindependent manner, increased the magnitude of receptor phosphorylation to about 15% of that attained by isoproterenol (Fig. 5C). H-89 pretreatment of cells abolished forskolin-induced ␤ 1 -AR phosphorylation, indicating that it was mostly due to PKA. H-89 reduced isoproterenol-mediated phosphorylation by about 13 Ϯ 4%, which was significant (p Ͻ 0.05). In the next series of experiments, we sought to determine whether the catalytic subunit of PKA (cPKA) phosphorylates the WT ␤ 1 -AR in membranes and whether mutagenesis of the putative PKAphosphorylated Ser 312 to Ala affects this parameter (Fig. 5D). The data in Fig. 5D reveal that cPKA markedly increased the phosphorylation of the WT ␤ 1 -AR and that the majority of the incorporated 32 P was in Ser 312 . Pretreatment of these membranes with 5 M of PKI inhibited cPKA-mediated phosphorylation of the ␤ 1 -AR, indicating that cPKA was the kinase involved (Fig. 5D, lane 9).
If a PKA-initiated mechanism is important for the resensitization of the ␤ 1 -AR, we should expect that the inhibition of ␤ 1 -AR phosphorylation by H-89 would be in line with the expected stoichiometric contributions of PKA versus GRK in phosphorylating this receptor. For the ␤ 2 -AR, it appears that six serine/threonine residues in the carboxyl terminus are phosphorylated by GRKs in response to agonist stimulation (43). The preferred residues for phosphorylation by GRK contain an acidic amino acid in their proximity (43). Within the FIG. 5. Regulation of ␤ 1 -AR signaling events by Ser 312 and PKA. A, mutagenesis of Ser 312 in the ␤ 1 -AR does not impair ERK1/2 phosphorylation in response to isoproterenol. HEK-293 cells stably expressing the various receptors outlined in A were exposed to vehicle or 50 ng/ml of pertussis toxin overnight. The cells were switched to serum-free DMEM for 2 h and then exposed to vehicle or 10 M isoproterenol for 5 min. Equal amounts of cell lysates were probed by immunoblotting with anti-phospho-P 42/44 MAPK antibody (1:000 dilution). The blots were then stripped and reprobed with anti-P 42/44 MAPK antibody (1:2,500 dilution). B, H-89 impairs forskolin-mediated phosphorylation of the ␤ 1 -AR. HEK-293 cells expressing the ␤ 1 -AR were metabolically labeled with 32 PiO 4 in phosphate-free DMEM and exposed to 10 M isoproterenol or 20 M forskolin Ϯ 1 M H-89 for 10 min. Cells were lysed in RIPA buffer, and equal amounts of protein were incubated overnight with anti-FLAG M2 IgG conjugated to agarose. Anti-FLAG immunoprecipitates were subjected to SDS-PAGE, and the gels were dried and subjected to autoradiography. Furthermore, the amount of 32 P incorporated into each band was counted by electronic autoradiography. The figure shows the phosphorylation of the␤ 1 -AR monomer (M r ϭ 73,000). C, the average counts/min incorporated into each ␤ 1 -AR monomer band from three separate experiments were plotted as the percentage of the counts/min incorporated by isoproterenol. Statistical analysis based on Student's t test revealed that H-89 significantly reduced the counts/min incorporated in the WT ␤ 1 -AR by isoproterenol (p Ͻ 0.05). D, comparison between in vitro phosphorylation by the catalytic subunit of PKA of membranes from cells expressing the WT ␤ 1 -AR or the Ser 312 3 Ala ␤ 1 -AR. Fifty g of membranes in kinase buffer containing either ascorbic acid (lanes 1, 3, 5, and 7) or 10 M isoproterenol (lanes 2, 4, 6, and 8) were incubated with 20 units of cPKA (lanes 3, 4, 7, and 8) for 10 min at 30°C. In lane 9, the membranes were incubated with 5 M of PKI peptide 5 min prior to the addition of cPKA. After phosphorylation, the ␤ 1 -AR was immunoprecipitated and subjected to SDS-PAGE, followed by autoradiography and electronic counting. carboxyl-terminal tail of the human ␤ 1 -AR, there are six serine residues with an acidic amino acid in their proximity that are potential sites for phosphorylation by GRK. Therefore, by analogy to the ␤ 2 -AR, the expected stoichiometry for GRK versus PKA in phosphorylating the ␤ 1 -AR would be about 6:1 (or 85:15%), which is in close agreement with the magnitude by which H-89 inhibited the phosphorylation of the ␤ 1 -AR (Fig.  5C). The data in Fig. 5D also indicate that pretreatment of the isolated membranes with isoproterenol, which promotes conformational changes in the receptor, had little effect on cPKAmediated phosphorylation of the ␤ 1 -AR. Therefore, it does not appear that steric hindrance of another phosphorylation site is caused by the Ser 312 3 Ala substitution; rather, cPKA is efficient in phosphorylating the agonist-unoccupied ␤ 1 -AR. These data indicate that PKA can phosphorylate the ␤ 1 -AR in cells in response to agonist activation and in isolated membranes, thereby providing a reversible signal that favors receptor recycling and recovery.
Many GPCR undergo agonist-induced endocytosis (8,14,22,23). Endocytosis seems to mediate opposite effects, namely resensitization (13-15, 19 -22) and down-regulation of the GPCR (12,14,22,23). The identification of PKA as the kinase involved in endocytic recycling of the ␤ 1 -AR identifies a novel role for PKA in receptor resensitization. Because the activation of PKA is but a consequence of catecholamine-mediated activation of the ␤ 1 -AR, this suggests that PKA is involved in homologous resensitization of the ␤ 1 -AR. Another facet concerning the relationship between the chronology and localization of PKA-mediated phosphorylation of the ␤ 1 -AR on the recycling itinerary of the agonist-occupied ␤ 1 -AR is revealed in the data of Fig. 5D. The data reveal that cPKA phosphorylation of the ␤ 1 -AR is independent of agonist occupancy and occurs rapidly in the membranes. Therefore, it is conceivable that PKA-mediated phosphorylation of Ser 312 preaddresses the trafficking itinerary of the agonist-internalized ␤ 1 -AR toward recycling and resensitization.
Persistent activation of PKA in cells that endogenously express the ␤ 1 -AR by chronic treatment with high levels of isoproterenol causes the destabilization of ␤ 1 -AR mRNA that ultimately results in down-regulation of the ␤ 1 -AR protein (44 -46). This effect is counteracted by PKA-mediated resensitization of the ␤ 1 -AR. These two apparently opposing effects of PKA, namely resensitization and mRNA destabilization of the ␤ 1 -AR, highlight an unexpected but a potentially important function for PKA as a regulator (rheostat) that strives to stabilize the flow of the signal (current) generated by the ␤ 1 -AR to a preset intensity (voltage). These findings also underscore an important physiological interplay between GRK-and PKAmediated phosphorylation of this GPCR. Since the phosphorylation by GRK and PKA are reversible, homologous desensitization through GRK and homologous resensitization through PKA are capable of generating several cycles of receptor activation, desensitization, and recovery, which potentially can maintain the signaling output from cells with low density of ␤ 1 -AR.
PKA is a plethoric regulator that is also involved in regulating the trafficking itinerary of many proteins such as the translocation of aquaporin-2 in response vasopressin in kidney cells and recycling of the N-methyl-D-aspartate receptor in cortical neurons (30,47). The effect of PKA on recycling of the Nmethyl-D-aspartate receptor highlights its potential role in heterologous resensitization of this receptor (30). Furthermore, PKA is involved in heterologous desensitization of the ␤ 2 -AR and others (40,41). Mutagenesis of the four serines that are putative sites for phosphorylation by PKA in the ␤ 2 -AR uncouples the receptor from G s and prevents ␤ 2 -AR-mediated activa-tion of ERK1/2 (42). However, mutagenesis of the corresponding serine in the human ␤ 1 -AR caused no adverse effects on its coupling to G s (37) (Fig. 1) or on ␤ 1 -AR-mediated activation of ERK1/2 (Fig. 5A). The motifs around the PKA-phosphorylation sites in each of these receptors are different but appear to play a major role in dictating the outcome for each of these modifications. A better understanding of the extended motifs surrounding PKA phosphorylation sites is fundamental in understanding their role in biochemical regulation of signal transduction.

VOLUME 282 (2007) PAGES 14626 -14634
MDM2 binding induces a conformational change in p53 that is opposed by heat-shock protein 90 and precedes p53 proteasomal degradation.
Mark Sasaki, Linghu Nie, and Carl G. Maki PAGE 14629: In Fig. 4 we reported that the p53-ubiquitin fusion protein has a mutant conformation (pAb1620Ϫ/pAb240ϩ). This p53-ubiquitin fusion protein has been used to mimic p53 mono-ubiquitinated in its C terminus. Subsequent to our manuscript being accepted, we discovered that the p53-ubiquitin fusion protein we used has a deletion of valine 218 near the pAb240 epitope. We have subsequently generated a new p53-ubiquitin fusion protein with valine 218 intact and compared its conformation to that of wild-type p53. This involved immunoprecipitation with the wild-type (pAb1620) and mutant (pAb240) conformation-specific antibodies followed by immunoblotting for p53. We did not detect an appreciable difference in conformation between wild-type p53 and the p53-ubiquitin fusion in these subsequent experiments. Therefore, the observation that the p53-ubiquitin fusion protein has an altered conformation compared with wild-type p53 was made in error. It remains possible that MDM2-mediated ubiquitination of p53, particularly at lysines within the p53 DNA-binding domain, could alter p53 conformation. However, the suggestion that C-terminal mono-ubiquitination might hold p53 in a mutant conformation is not supported.

VOLUME 279 (2004) PAGES 21135-21143
Role of the cyclic AMP-dependent protein kinase in homologous resensitization of the ß 1 -adrenergic receptor. The images shown in Figs. 2A (p. 21138) and 4A (p. 21140) were obtained from slides prepared simultaneously from the same colony of cells. Panel c in Fig. 2A and panel KЈ in Fig. 4A, due to an inadvertent error, were derived from slides that were not part of the set shown. The correct Fig. 2A, panel c,

ADDITIONS AND CORRECTIONS
This paper is available online at www.jbc.org We suggest that subscribers photocopy these corrections and insert the photocopies in the original publication at the location of the original article. Authors are urged to introduce these corrections into any reprints they distribute. Secondary (abstract) services are urged to carry notice of these corrections as prominently as they carried the original abstracts.