&Adrenergic Receptor Sequestration

Transcription

1 THE JOURNAL OF BIOLOGICAL CHEMISTRY by The American Society for Biochemistry and Molecular Biology, Inc. Vol. 266, No. 1, Issue of January 5, pp ,1993 Printed in U. S. A. &Adrenergic Receptor Sequestration A POTENTIAL MECHANISM OF RECEPTOR RESENSITIZATION* (Received for publication, June 11, 1992) Steven S. Yu, Robert J. LefkowitzS, and William P. Hausdorffs From the Departments of Medicine (Cardiology) and Biochemistry, Howard Hughes Medical Institute, Duke University Medical Center, Durham, North Carolina Continuous exposure of cells to hormonal agonists regulatory proteins, or G proteins. In a process termed desenoften causes a rapid waning of the stimulated response. sitization, continuous agonist stimulation of the PzAR leads This desensitization effect has been extensively studied to a waning in receptor-mediated adenylyl cyclase response. in the &adrenergic receptor system, and attributed Two biochemical processes are temporally associated with largely to the rapid phosphorylation of the receptor by this rapid (seconds to minutes) functional desensitization. two kinases. Over a similar time frame (seconds to First, agonist occupancy leads to phosphorylation of PzAR by minutes), agonists also trigger a selective loss in the the p-adrenergic receptor kinase (PARK) and the CAMPcapacity of receptors to bind hydrophilic but not hy- dependent protein kinase (PKA) (reviewed in Ref. 1). These drophobic ligands, a phenomenon termed sequestracovalent modifications, along with the subsequent binding of tion. There is some evidence suggesting that sequestracytosolic factors, disrupt interaction of the receptor with the tion represents the rapid internalization of receptors, but the functional significance of sequestration has remained unclear. Upon the removal of agonist, both desensitization and sequestration are readily reversed with similar kinetics (ty-3 min for both). To investigate the possibility that receptor sequestration is involved in this resensitization of the adenylyl cyclase response, we applied two distinct approaches to block receptor sequestration: by pretreating cells with sucrose and by creating a sequestration-defective &-adrenergic receptor by site-specific mutagenesis. Both approaches effectively disabled receptor sequestration, with little effect on adenylyl cyclase stimulation or on desensitization. However, in both cases, no re- covery from desensitization was apparent even 20 min after the removal of agonist. Similarly, pretreating cells with concanavalin A almost completely blocked receptor sequestration and resensitization but only partially inhibited other receptor functions. Our results therefore suggest that sequestration of Bz-adrenergic receptors is a mechanism involved in reactivating and recycling desensitized receptors. The &-adrenergic receptor (&AR) is a member of a large family of hormone and neurotransmitter receptors that initiate their biological function by coupling to GTP-binding * This work was supported in part by National Institutes of Health Grant 4R37-HL The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked aduertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. $ TO whom correspondence should be addressed. Depts. of Medicine (Cardiology) and Biochemistry, Howard Hughes Medical Inst., Box 3821, Duke University Medical Ctr., Durham, NC Current address: Office of Health, Agency for International Development, SA-18, Rm. 1200, Washington, DC The abbreviations used are: P2AR, &adrenergic receptor; BARK, 0-adrenergic receptor kinase; PKA, CAMP-dependent protein kinase; G., stimulatory G protein; CHO, Chinese hamster ovary; Cod, concanavalin A; ISO, (-)-isoproterenol, a 0-adrenergic receptor agonist; WT, wild-type; CGP 12177, a hydrophilic ligand commonly used to assess cell surface receptor number; PBS, phosphate-buffered saline. 337 stimulatory G protein, G., thus reducing adenylyl cyclase activation. Second, agonists induce a similarly rapid process termed receptor sequestration, in which receptors remain detectable by lipophilic ligands, but become increasingly inaccessible to hydrophilic, membrane-impermeable ligands. Sequestration has been proposed to represent an internaliza- tion of receptors into an intracellular compartment (reviewed in Ref. 2). An alternative model of sequestration postulates that these receptors are not internalized, but rather, are sequestered in the plasma membrane in an altered conformation with different ligand-binding characteristics (3). Regardless of its exact nature, receptor sequestration is a process that is distinct from receptor down-regulation, as in the latter there is an actual decrease in the total cellular receptor number, and it is only apparent after prolonged (>1 h) exposure to agonist. Although it was earlier thought that sequestration of receptors played an important role in desensitization, there is now ample evidence contradicting this idea. For example, in experiments conducted by several groups, sequestration was blocked with agents such as concanavalin A (ConA) or phenylarsine oxide, with no apparent effect on desensitization (4-8). Conversely, normal receptor sequestration has been observed in cells in which rapid desensitization was almost abolished, either by mutagenesis of the receptor (9,lO) or by application of kinase inhibitors (11). Finally, the two processes have been temporally separated, with the loss of functional response apparent before sequestration (8). It has also been proposed that receptor sequestration is a prerequisite stage for down-regulation (4). However, we recently described a mutant receptor that was markedly impaired in its ability to undergo sequestration in response to agonist, but nevertheless, exhibited a wild-type pattern of down-regulation (12). Therefore, whether sequestration plays any role in the regulation ofp,ar function has remained controversial. In a previous study, Sibley et al. (13) reported that sequestered receptors exhibited decreased phosphorylation as compared to desensitized plasma membrane receptors. Additionally, vesicles containing the sequestered receptors were found to be rich in phosphatases. The present study was designed to further explore the possibility, as suggested by

2 Receptor Sequestration Resensitization and those findings, that sequestration may serve as a mechanism of receptor resensitization, the process by which the attenuated adenylyl cyclase response recovers after removal of the desensitizing agonist. EXPERIMENTAL PROCEDURES Mutagenesis and Cell Transfection-Mutant human &ARwere constructed by a general method for site-directed mutagenesis using the polymerase chain reaction (14). The purified DNA fragments were digested with appropriate restriction enzymes and ligated into pbc/nar (15). Following ligation into pbc/nar, authenticity of the mutated and wild-type portions of the polymerase chain reaction products was verified using dideoxy sequencing analysis. Human P2AR or mutant PzAR cdna-bearing plasmids were cotransfected with psv2-neo into CHO cells via coprecipitation with calcium phosphate. Geneticin-resistant clones stably expressing &AR or mutant &AR were screened by [12SI]cyanopindolol binding assay and subsequently maintained in Dulbecco s modified Eagle s medium with 10% fetal bovine serum plus 500 pg/ml geneticin. Expression levels in cells were fmol of receptor/mg of membrane protein. Desensitization and Resensitization of Cells-Cells expressing wildtype or mutant P2AR were preincubated for 20 min in media alone, or with 0.45 M sucrose, or 0.25 mg/ml ConA (CalBiochem). The cells were then exposed to 10 pm isoproterenol (ISO) and 0.1 mm ascorbic acid for 20 min to induce desensitization. After incubation, the cells that were to be assessed for desensitization were extensively washed with colddulbecco s phosphate-buffered saline (PBS; GIBCO) to remove the ISO, and then membrane fractions were prepared. Cells that were to be assessed for resensitization were washed with warm PBS and allowed to continue incubation in media alone for 3-60 min. This incubation was terminated by washing with cold PBS. Desensitization was expressed as percent loss of adenylyl cyclase response: % desensitization = 100 X [adenylyl cyclase stimulation in (control cells - desensitized cells)/(control cells)]. % Maximum adenylyl cyclase = 100 x (adenylyl cyclase stimulation)/(maximum adenylyl cyclase stimulation in control cells). The adenylyl cyclase response to 30 nm IS0 was generally used to quantitate the desensitization and resensitization since this concentration gave the largest. desensitization signal that could be consistently and reliably measured. Adenylyl Cyclose Assay-Cells were scraped into 5 mm Tris (ph 7.4), 2 mm EDTA, 1 mm sodium pyrophosphate, and 1 p~ microcystin (CalBiochem) (to inhibit phosphatase activity), rendering >90% of the cells permeable to trypan blue. Lysed cells were centrifuged at 40,000 X g for 30 min at 4 C, and the particulate fractions assayed for IS0 stimulated adenylyl cyclase activity as described in Ref. 15. Sequestration-Cells expressing wild-type or mutant &AR were preincubated for 20 min in media alone, or with 0.45 M sucrose, or 0.25 mg/ml of ConA. The media was then supplemented with 10 ym IS0 and 0.1 mm ascorbic acid for various lengths of time. One group of cells was washed with cold PBS to terminate reactions, while the other cells were washed with warm PBS and allowed to continue incubation in agonist-free media. Following incubations, all cells were treated with cold PBS containing 5 mm EDTA for min, and cells detached by vigorous shaking of the flasks. >90% of the cells remained impermeable to trypan blue following this treatment. Following low speed centrifugation (200 x g), cells were incubated at 13 C for 3 h with 200 p~ [1251]iodopindolol either alone (to define total binding), in the presence of 1.0 p~ (-)-propranolol (to define nonspecific binding), or in the presence of0.2 pm CGP (a hydrophilic ligand commonly used to assess cell surface receptor number (2)). The IS0 and CGP binding properties of both mutant forms of &AR examined in this study were same as those of wild-type p2ar (data not shown). Receptor sequestration was defined as the percentage of specific radioligand binding not displaced by CGP minus the basal level of sequestration as measured in cells before any exposure to agonist. The amount of Sequestration in naive (ie, before agonist exposure) cells was typically about lo%, a value similar to that obtained in previous studies by many groups, and considered basally sequestered receptors. RESULTS Kinetics of Reversal of Sequestration-Subcellular compartments containing the sequestered receptors are relatively deficient in G, (8, 16, 17) Therefore, if sequestration of desensitized receptors allows for their reactivation, adenylyl cyclase resensitization would not be apparent until receptors are reincorporated into the plasma membrane. In other words, the resensitization of adenylyl cyclase response cannot be any faster than the recycling of receptors back to the cell surface. To compare the kinetics of these processes, we first incubated CHO cells expressing the human PAR with IS0 for 20 min, resulting in the sequestration of approximately 25% of the total cell receptors. Incubation was then continued in agonistfree media for various lengths of time. Number of sequestered receptors was measured as described under Experimental Procedures. As shown in Fig. la, reversal of sequestration in transfected CHO cells occurs with a tlh of approximately 3 min. In a parallel experiment, Fig. lb shows that, upon the removal of agonist, the resensitization of adenylyl cyclase response also occurs with a tlh -3 min. The similar kinetics of the two processes demonstrate that the externalization of 100?, [;,*,>;&; 40 e 80 - j I *Tj 4 70, : s,. * 50.,:;.,,..,.. ;.~,~ Wl+sucrose S356G+S3&lG i Minutes IS0 added so FIG. 1. Time course of reversal of sequestration and resensitization. a, cells expressing wild-type &AR were exposed to 10 p~ IS0 for 20 min and then washed free of agonist and incubation continued in the absence of agonist for various lengths of time. Percentage of sequestered receptors was measured as described under Experimental Procedures. The dushed line represents a hypothetical time course of receptor sequestration (see Refs. 8, 9, 12 and Fig. 2 of this report for actual time courses) during the 20 min of IS0 incubation, while the time after 20 rnin depicts the percentage of receptors that remain sequestered following the removal of agonist. Basal sequestration is defined as the level of receptor sequestration before cells are exposed to agonist (-10%). Data are from a single experiment representative of four experiments. Error bars represent f S.E. of triplicate samples. b, cells expressing wild-type &AR were pretreated without ( WT, 0) or with ( WT + sucrose, 0) 0.45 M sucrose. These cells, along with cells expressing S356G -t S364G (0) were exposed to 10 p~ IS0 for 20 min, washed free of agonist and incubation continued in absence of agonist for various lengths of time as indicated. Membranes were prepared and adenylyl cyclase stimulation measured with 30 nm Is0 (100 nm for S356G + S364G). The dashed line represents a hypothetical time course of desensitization (see Refs. 30 and 31 for actual time courses) during the 20-min IS0 incubation, while the time after 20 min is the course of resensitization. Control adenylyl cyclase activity for these experiments is defined as the adenylyl cyclase activity stimulated by 30 nm IS0 (100 nm for S356G + S364G, see legend to Table I) for membranes prepared from cells with no prior agonist exposure. Data are means f ranges of two experiments for S356G + S364G, and means rtr S.E. of three to five experiments for WT and WT + sucrose.

3 ~~ ~ T I 25- n c 0 '= 10- E v) al 3 5- P-Adrenergic Resensitization Sequestration Receptor and 339 S356G+S364G 0' WT+sucrose s l -5 -I Minutes of incubation in IS0 FIG. 2. &Adrenergic receptor sequestration. Cells expressing human P2AR, S356G + S364G, or T360A + Q363K were exposed to 10,.LM IS0 for various lengths of time as indicated and sequestration assessed. Basal sequestration is defined as the level of sequestration in cells before agonist exposure (-10%). Data are means f S.E. of three to six experiments. Error bars for sucrose-treated and S356G + S364G cells were smaller than symbols. receptors is a sufficiently rapid mechanism to be involved in resensitization. Selective Inhibition of Sequestration by Pretreatment with Sucrose or Mutation of the Receptor-As shown in Fig. 2, when cells are exposed to agonist, receptors become sequestered rapidly with a tlh < 10 min and approximately 23% of the receptors are sequestered after 30 min of IS0 incubation. We note that there was no decrease in total receptor levels, as measured by ['251]iodopindolol binding, during this time. In our study, we wished to suppress this sequestration of receptors and observe how it affects resensitization. Specific inhibition of receptor sequestration was achieved using two different methods. Pretreating cells with a hypertonic concentration of sucrose has been demonstrated to be effective in blocking receptor endocytosis via clathrin-coated pits (2, 18). As this treatment has also been reported to inhibit P-adrenergic receptor sequestration in at least two cell types (2); we assessed the utility of sucrose in our transfected cell system. We therefore preincubated cells expressing the wildtype P2AR with 0.45 M sucrose for 20 min, and then exposed them to 10 p~ IS0 for various lengths of time. Even after 60 min of agonist incubation, almost no receptors became sequestered (Fig. 2). We next sought to establish the specificity of sucrose in inhibiting sequestration. Sucrose pretreatment did not markedly affect adenylyl cyclase stimulation or desensitization as shown in Table Ia, nor did it affect agonist binding affinity (data not shown). Inhibition of receptor sequestration was also achieved by mutagenesis of the receptor. Previously, our group reported a mutant P2AR in which 4 serine and threonine residues in the cytoplasmic C terminus tail of the receptor were substituted with alanine and glycine residues (12). Substitution of these residues prevented agonist-promoted phosphorylation, sequestration, and rapid desensitization but adenylyl not cyclase stimulation, thus providing evidence that this small region of the PZAR is involved in several regulatory processes. This finding prompted further mutagenesis studies in this region of the PZAR, leading to the construction of a number of mutant. receptors with different combinations of amino acid substitutions in this region? One mutant that was of particular J. P. Perkins, personal communication. J. Sung, J. Ostrowski, M. Hnatowich, H. E. Kendall, B. F. O'Dowd, R. J. Lefkowitz, and W. P. Hausdorff, manuscript in preparation. J TABLE I Effect of sucrose pretreatment and mutations on AC responsiveness and desensitization For AC stimulation, cells were preincubated with or without SUcrose as indicated, and particulate fractions were prepared. For desensitization measurements, after treatment with or without sucrose, cells were exposed to 10 PM IS0 for 20 min. The relative abilities of 30 nm IS0 to stimulate the AC response in membranes from naive and desensitized cells were then compared. In the case of cells expressing S356G + S364G, the relative stimulatory ability of 100 nm IS0 was examined because this dose produces equivalent AC stimulation in membranes from naive cells to that achieved with 30 nm IS0 in membranes from naive WT cells. Data shown in a and b were from experiments performed several months apart with different clones of cells and represent means f S.E. of four experiments for a, four to seven experiments for WT and S356G + S364G in b, and three experiments for T360A + Q363K in b. Adenylyl cyclase activity Basal ISO,.. ECm Desensitization pmol camplmgfmin nm % a WT 6.8 f ? k f 5.0 WT + sucrose 7.3 f C f 2.6 b WT 10.3 f f f f 2.4 S356G + S364G 6.7 f ? f f 4.8 T360A + Q363K 17.3 f t f 1.9 interest for the present study was comprised of substitutions of Ser356 and Ser3'j4 with glycines (S356G + S364G), and was totally impaired in its ability to undergo sequestration (Fig. 2). However, unlike the larger mutation (12), this mutant P2AR underwent desensitization much like the wild-type (Table Ib), thus making it suitable for this study. How Does Prevention of Sequestration Affect Resensitization?-We next examined the resensitization of adenylyl cyclase response in cells where receptor sequestration was blocked. Cells with the wild-type PzAR were pretreated with or without sucrose for 20 min. These cells, along with cells expressing S356G + S364G, were then exposed to 10 pm IS0 for 20 min. After washing the cells free of agonist and sucrose (when present), the cells were allowed to continue incubation in media alone for 20 min. As shown in Fig. 3, in all three cases, ISO-induced desensitization results in a rightward and downward shift of the agonist concentration-adenylyl cyclase response curve. In cells expressing wild-type &AR that were not treated with sucrose, this desensitization is substantially reversed (ie. resensitizes) after a 20-min incubation in agonist-free media as shown by the curve produced from resensitized cells in Fig. 3a. This reversal process occurs with a tlh of -3 min (Fig. 16). In the sucrose-treated cells, however, desensitization was not reversed after the same 20-min incubation, i.e. the cells remained desensitized (Fig. lb and Fig. 3b). Similarly, cells expressing S356G + S364G did not resen- sitize after 20 min of incubation in agonist-free media (Fig. 16 and Fig. 3c). While the extent of resensitization in the wild-type cells after 20 min varied somewhat (ranging from 60 to 80% complete), the virtually absolute failure of the sucrose-treated WT cells and S356G + S364G expressing cells to resensitize was a highly consistent finding. To further assess the effect of blocking sequestration on the resensitiza- tion process, we examined the functional properties of another mutant receptor, T360A + Q363K (substitutions of Thr3@' + Ala and Gln363 * Lys), which was only partially impaired in its ability to undergo sequestration (Fig. 2). While exhibiting normal adenylyl cyclase stimulation and substantial desensitization (Table Ib), T360A + Q363K resensitized at a rate that was approximately one-third to one-half of cells expressing wild-type P2AR (data not shown). As shown in many other cell lines (reviewed in Ref. Z),

4 Receptor Sequestration and Resensitization log [isoproterenol] M FIG. 3. Effect of sucrose on resensitization of adenylyl cyclase, Cells expressing wild-type p,ar that were pretreated without (a) or with (b) 0.45 M sucrose and cells expressing S356G + S364G (c) were incubated in 10 PM IS0 for 20 min. For each group, control cells were incubated in parallel in the absence of desensitizing agonist. All cells were washed, and for cells that were to be assessed for resensitization, incubation was continued in the absence of agonist for 20 min. Membranes were prepared from the cells, and adenylyl cyclase stimulation wasmeasured. Response was expressed as percentage of maximal response incontrolcells.data shown are fromsingleexperiments representative of four experiments. pretreatment of CHO cells with the agent ConA is also effective in preventing pzar sequestration, In two experiments, pretreatment of cells with ConA also prevented resensitization; 86 f 11.5% resensitization in untreated cells us. 6.8 f 16.8% resensitization in ConA-treated cells (mean -+ range; measured at 30 nm IS0 after 20 min of agonist-free incubation). However, ConA pretreatment also noticeably impaired adenylyl cyclase stimulation and desensitization (-30% impairment in both parameters, data not shown). RECYCLING ',, >I (RESENSITIZATIONI SEQUESTRATION DISCUSSION Although the sequestration receptors has been well characterized by many studies, by this group and others, no specific physiological function has yet been assigned to this process. Several observations reported in this study provide strong evidence that receptor sequestration may serve an important role in the resensitization of the P-adrenergic receptor. The results of our investigation are thus consistent with previous findings by Sibley et al. (13) in which sequestered,&adrenergic receptors, isolated in a light vesicle fraction from desensitized frog erythrocytes, were determined to be considerably less phosphorylated as compared to plasma membrane receptors. These authors had also demonstrated that these vesicular fractions were better able to promote dephosphorylation, in vitro, of exogenously added phosphorylated receptors when compared to other cell fractions. It was concluded that sequestered vesicles are enriched in phosphatases that dephosphorylate receptor (13). In experiments conducted by Waldo and colleagues (8), cells were treated with ConA 15 min after they were exposed to ISO, apparently with the intent to block the recycling of sequestered receptors back to the plasma membrane. They found that neither the recycling of sequestered receptors nor the resensitization of adenylyl cyclase response was affected under these conditions. However, this ConA treatment protocol would not have sufficiently blocked sequestration to test its correlation with resensitization. Thus, these results do not conflict with those of the present study. It is possible that sucrose has a direct influence on resensitization, rather than influencing resensitization via its effects on sequestration. Although we cannot exclude this possibility, we believe it to be highly unlikely for two reasons. First, the cells were not in the presence of sucrose during the resensitization incubation. They were treated with sucrose only during the desensitization incubation. Second, the specificity of sucrose in blocking sequestration without affecting G protein coupling and adenylyl cyclase stimulation, agonist binding (data not shown), and desensitization, argue against FIG. 4. Proposed mechanisms of &AR desensitization and resensitization. Upon agonist stimulation, the receptorbecomes phosphorylated by BARK and PKA, to bind, and thus leading to its uncoupling from G,. These uncoupled receptors are sequestered away from the cell surface into phosphatase-enriched vesicle compartments. In these compartments the phosphorylated receptors arepresumablydephosphorylatedandthusreactivated. Receptors are then recycled back to the plasma membrane where they can recouple to G,. any coincidental effect of sucrose on resensitization. Fig. 4 is a schematic illustration of our current model for the cellular and molecular events involved in the rapid desen- sitization and resensitization of the &AR. Upon agonist stimulation, the receptor becomes phosphorylated by PARK and PKA. Another protein, which is analogous to arrestin in the rhodopsin system (19), binds to the phosphorylated receptor and thereby further impairs interaction with G, (20). These uncoupled receptors are sequestered away from the cell surface into phosphatase-rich vesicle compart- ments. It is important to mention here that numerous studies have demonstrated that receptor phosphorylation itself is not the trigger for its internalization (9-11, 21, 22). The actual location of sequestered receptors has been a point of some controversy. Some groups have localized sequestered receptors in what they contend are intracellular vesicles (8, 13), whereas others propose that sequestered receptors adopt an alternate conformation that affects ligand binding, but remain associated with the plasma membrane (3, 23, 24). In a new report, a rapid, agonist-induced internalization of p2ar into intracellular vesicles was observed by immunofluorescence microscopy, occurring with roughly the same time course as sequestration (25). These results, while differing from those obtained in a previous immunofluores-

5 &Adrenergic Resensitization Receptor Sequestration and 341 cence study (3, 24), provide the strongest evidence to date supporting the notion of rapid agonist-induced receptor internalization. In any case, the hypothesis of a relationship between resensitization and sequestration put forth here is not dependent upon the biochemical nature of sequestration or the precise location of sequestered receptors. In the sequestered compartments, the phosphorylated receptors are presumably dephosphorylated, and thus reactivated. Alternatively, under the notion that receptors do not internalize, receptors on the cell surface adopt an altered binding conformation that renders it more susceptible to phosphatases. Subsequently, the receptors are recycled back to or reoriented within the plasma membrane where they recouple to G,. It is to be noted that sequestration and externalization has been demonstrated to occur continuously in the presence of agonist (26). Thus the steady state concentration of phosphorylated receptors in the plasma membrane during desensitization may reflect the competing processes of receptor phosphorylation versus receptor sequestration, dephosphorylation, and recycling. It might be expected that selectively blocking sequestration to prevent resensitization would result in an apparently enhanced desensitization. Throughout the course of our study, we did not observe any enhancement of desensitization when resensitization wasblocked.however, since the maximum extent of receptor sequestration in our cells was only 25%, by our estimate any such increase in desensitization would be relatively small and beyond accurate detection given the relatively high number of receptors in these cells combined with the imprecision of the desensitization assay. Such an increase in desensitization would perhaps be more apparent in cells with relatively low levels of P2AR compared to G,. In addition, it is also possible that sucrose or receptor mutations may have small inhibitory effects on desensitization, which would effectively negate or obscure any enhancement in desensitization due to the block of resensitization. Several other receptors, including those for insulin, epidermal growth factor, transferrin (27), and the muscarinic acetylcholine receptor (28,29), also undergo phosphorylation and internalization in response to the appropriate agonist. The results presented here suggest that internalization of receptors, such as the &AR, may play a fundamental role in the regulation of receptor function. Whether sequestration represents a general mechanism through which the phosphoryl- ation state and function of a variety of plasma membrane receptors are regulated remains to be determined. Acknowledgments-We thank Jacek Ostrowski for construction of the mutant receptors and Grace Irons and Kaye Harlow for their excellent cell culture assistance. REFERENCES 1. Hausdorff. W. P.. Caron. M. G.. and Lefkowitz. R. J. (1990) FASEB J. 4, Perkins, J. P., Hausdorff, W. P., and Lefkowitz, R. J. (1991) The p- Adrenergic Receptor (Perkins, J. P., ed) pp , Humana Press, Clifton, NJ 3. Wang, H., Berrios, M., and Malhon, C. C. (1989) Biochem. J. 263, 533-5RR 4. He%& C., Coulter, S. J., and Perkins, J. P. (1985) J. Biol. Chem. 260, Feldman, R. D., McArdle, W., and Lai, C. (1986) Mol. Pharmacol. 30, Kassis, S., Olasmaa, M., Sullivan, M., and Fishman, P. H. (19861 J. Bioi. Chem. 261, Box, R. J., and Staehlin, M. (1987) FEBS Lett. 214, Waldo. G. L.. NorthuD. J. K.. Perkins.. J. P... and Harden. T. K. (1983) J. Biol. Chem: 258, Hausdorff. W. P.. Bouvier. M.. ODowd. B. F.. Irons. G. P.. Caron. M. G.. and Le&owitz, R. J. (1989) J. Biol. Ckm. 264, Bouvier, M., Hausdorff, W. P., De Blasi, A,, O Dowd, B. F., Kobilka, B. K., Caron, M. G., and Lefkowitz, R. J. (1988) Nature 333, Lohse, M. J., Benovic, J. L., Caron, M. G., and Lefkowitz, R. J. (1990) J. Biol. Chem. 265, Hausdorff, W. P., Campbell, P. T., Ostrowski, d., Yu, S. S., Caron, M. G., and Lefkowitz, R. J. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, Sibley, D. R., Strasser, R. H., Benovic, J. L., Daniel, K., and Lefkowitz, R. J. (1986) Proc. Natl. Acad. Sci. U. S. A. 83,940& Nelson, R. M., and Long, G. L. (1989) Anal. Biochem. 180, Hausdorff, W. P., Hnatowich, M., O Dowd, B. F., Caron, M. G., and Lefkowitz, R. J. (1990) J. Bid. Chem. 265, Stadel, J. M., Strulovici, B., Nambi, P., Lavin, T. N., Briggs, M. M., Caron, M. G., and Lefkowitz, R. J. (1983) J. Biol. Chem. 268, , DeBlasi, A., Lipartiti, M., Motulsky, J., Insel, P. A,, and Fratelli, M. (1985) J. Clin. Endocrinol. Metab. 61, Daukas, G., and Zigmond, S. H. (1985) J. Cell Biol. 101, Wilden, U., Hall, S. W., and Kiihn, H. (1986) Proc. Natl. Acad. Sci. U. S. A. 83, Lohse, M. J., Benovic, J. L., Codina, J., Caron, M. G., and Lefkowitz, R. J. (1990) Science 248, Strader, C. D., Sigal I. S., Blake, A. D., Cheung, A. H., Register, R. B., Rands, E., Zemick: B. A,, Candelore, M. R., and Dixon, R. A. F. (1987) Cell 49, Campbell, P. T., Hnatowich, M., ODowd, B. F., Caron, M. G. Lefkowitz, R. J., and Hausdorff, W. P. (1991) Mol. Pharmacol. 39, Mahan, L. C., Motulsky, H. J., and Insel, P. A. (1985) Proc. Natl. Acad. SCL U. S. A Wang, H., BerGos, M.,and Malhon, C. C. (1989) Biochem. J. 263, von-zastrow, M., and Kobilka, B. K. (1992) J. Biol. Chem. 267, Kurz, J. B., and Perkins, J. P. (1992) Mol. Pharmacol. 41, Stahl, P., and Schwartz, A. L. (1986) J. Clin. Zrzoest. 77, Kwatra M. M., and Hosey, M. M. (1986) J. B~ol. Chem. 261, Eva. C.. Gamalero. S. R.. Genazzani., E.., and Costa. E. i *J. Pharmueol. I,,~ EX ;fher. 253; Su, f; Harden, T. K., and Perkins, J. P. (1980) J. Biol. Chem. 255, Lohsi: M. J., Lefkowitz, R. J., Caron, M. G., and Benovic, J. L. (1989) Proc. Natl. Acad. Sei. U. S. A. 86,