the agonist-occupied form of the B-adrenergic receptor. We have utilized the .... (3-Adrenergic Receptor Phosphorylation by PAR Kinase-Reconsti- tuted PAR ...
Vol. 263, No. 8, Issue of March 15,pp. 3893-3897,1988 Printed in U.S.A.
THEJOURNALOF BIOLOGICAL CHEMISTRY 0 1988 hy The American Society for Biochemistry andMolecular Biology, Inc
@-AdrenergicReceptor Kinase ACTIVITY OF PARTIALAGONISTSFORSTIMULATIONOFADENYLATE ABILITY TOPROMOTERECEPTORPHOSPHORYLATION*
CYCLASE CORRELATESWITH
(Received for publication, April 20, 1987)
Jeffrey L. Benovic, Claudia Staniszewski, Federico Mayor, Jr., Marc G. Caron, and Robert J. Lefkowitz From the Departments of Biochemistry, Physiology, and Medicine, Howard Hughes Medical Institute, Duke University Medical Center, Durham, North Carolina27710
The B-adrenergic receptor (BAR) kinase is a recently discovered enzyme which specifically phosphorylates the agonist-occupied form of the B-adrenergicreceptor. We have utilized the agonist-dependent nature of this phosphorylation reaction to characterize the ability of partial agonists to interact with the receptor. Partial agonists were tested for their ability to: 1) stimulate adenylate cyclase activityin a three-component reconstituted system, and 2) promote phosphorylation of BAR by BAR kinase. There is an excellent correlation between the ability of partial agonists to stimulate adenylate cyclase activityand promote receptor phosphorylation by BAR kinase ( y = 1 . 0 2 ~ 0.01, r = 0.996, p < 0.001). Peptide maps of receptor phosphorylated by BAR kinase inthe presence of full or partial agonists are virtually identical with the partial agonist pattern reduced in intensity. Moreover, kinetic studies of BAR phosphorylation by BAR kinase suggest that partial agonistsalter the V,,, of the reaction with little, if any, effect on the K,. These results suggest that at steady state partial agonists transform a smaller portion of the receptor pool into the conformationally altered or activated form which serves as the substrate for BAR kinase, although they do not completely rule out the possibility that a partial conformational change is occurring.
Pharmacologists andbiochemists have long sought specific approaches to the measurement of agonist-induced conformational changes in receptors and other proteins. Such measurements have usually been indirect and reflective of events occurring distal to or as a result of the agonist-promoted conformational change. Since phosphorylation of the P-adrenergic receptor by PAR’ kinase seems to require agonist occupancy, and this presumablyreflects agonist-induced conformational changes, we wondered if this phosphorylation reaction could be used as a means of assessing the ability of agonists toinduce such conformational changes. In the@-adrenergic receptorsystem, as in a number of other pharmacological systems, numerous agonists and partial agonists are available which vary, not only in their affinity for the receptor, but in their maximumability to stimulate a biological process (9). It has been suggested previously that such partial agonists alsodisplayonlya partial ability to desensitize the adenylatecyclase-coupled system during short term incubations (10-12). Accordingly, in the present work, we sought to evaluatea series of P-adrenergic partial agonists for their ability to activate the receptors and convert them into a substrate for PAR kinase. We assessed biological activating function by reconstituting pure hamster lung @-adrenergic receptor intophospholipid vesicles with the isolated G. (stimulatory guanine nucleotide regulatory protein) and the resolved catalytic moiety of the adenylate cyclase system. In this completely reconstituted system, @-agonists are able to Mounting evidence indicates that the function of several stimulate adenylate cyclase activity. The ability to promote types of receptors may be modulatedby covalent modification receptor phosphorylationwas studied by measuring and cominvolving phosphorylation/dephosphorylation reactions (1). paring the ability of the different agonists to transform isoNotably, in the case of adenylate cyclase-coupled @-adrenergic lated hamster lungP-adrenergic receptors, reconstituted into receptors, both homologous and heterologous forms of desen- phospholipid vesicles, into a substrate for the P-adrenergic sitization appear to be associated with @-adrenergic receptor receptor kinase. phosphorylation (2-5). In the case of homologous desensitiEXPERIMENTALPROCEDURES zation, this reaction appears to be carried out by a specific Padrenergic receptor kinase (6). This unique cyclic AMP-inMaterials-Most chemicals were from sources previously described dependentkinaseisfoundinthecytoplasm of cells and (3, 13). (-)-Isoproterenol was from Sigma, while dobutamine, Lilly 46220, and dichloroisoproterenol were from Eli Lilly. Deoxyisoproappears to be translocated to the plasma membranes after agonist stimulation(7). The enzyme predominantly phospho- terenol was from Sterling-Winthrop Research Institute, and MJ 9184 rylates theagonist-occupied form of the receptor, presumably (zinterol) and (-)-sotereno1 were from Mead Johnson.Tertatolol (Servier, S2395) was generously provided by Dr. Antonio DeBlasi, reflecting the ability of agonist occupancy to produce confor- IRFMN, Milan, Italy. mational changes in the receptor whichexposepreviously Purification and Reconstitution of BAR-The B-adrenergic receptor inaccessible sites of phosphorylation. These siteslikely lie on from hamster lung was purified to >95% homogeneity by sequential cytoplasmic domains of thereceptorand possibly at the affinity and high performance liquid chromatography as described carboxyl terminus (8). (13). The purified receptor was reinserted into phosphatidylcholine
-
* The costs of publication
vesicles as previously described (6, 14). The protein-lipid pellets were
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.
‘The abbreviations used are: BAR, &adrenergic receptor; SDS, sodium dodecyl sulfate; EGTA, [ethylenebis(oxyethylenenitrilo)]tetraacetic acid.
3893
3894
Partial Agonist Stimulated Phosphorylation
resuspended in 20mM Tris-HC1, pH 7.2, 2 mM EDTA and used as a substrate for BAR kinase. Purification of G.-The stimulatory guanine nucleotide regulatory protein (G.) was purified from bovine cerebral cortex using methods adapted from Sternweis and Robishaw (15, 16). Briefly, membranes solubilized with 1% cholate were centrifuged, and the resulting supernatant was chromatographed on DEAE-Sephacel, Ultrogel AcA34, octyl-Sepharose, and hydroxylapatite with a final step on DEAESephacel. These procedures resulted in a preparation of G, 50-90% pure as judged by Coomassie Blue staining of polyacrylamide gels. G. was stored at -80 “C in a buffer consisting of20mM Tris-HC1, pH 8.0, 1 mM EDTA, 1 mM dithiothreitol, 1%sodium cholate, 150 mM NaCl. Preparation of Resolved Adenylate Cyche-The catalytic moiety of adenylate cyclase was solubilized from bovine caudate with sodium cholate and isolated from the other components of the system by Sepharose 6B chromatography, essentially as described by Strittmatter and Neer (17). In these preparations, adenylate cyclase appeared to be effectively separated from the other functional components of the system (18). Purification of the &Adrenergic Receptor Kinase (PAR Kinase)PAR kinase was purified from bovine cerebral cortex as described (19). Briefly, 100 g of bovine cortex washomogenized, andthe resulting high speed supernatant fraction was precipitated with 1326% ammonium sulfate. This material was then chromatographed on Ultrogel AcA34, DEAE-Sephacel, and CM-Fractogel. The preparations used were 10-20% pure as judged by Coomassie Blue staining of SDS-polyacrylamide gels. BAR kinase was stored at 4 “C in abuffer consisting of 20 mM Tris-HC1, pH 7.5, 2 mM EGTA, 10 pg/ml leupeptin, 10 pg/ml benzamidine, 5 pg/ml pepstatin, 0.1 mM phenylmethylsulfonyl fluoride, 60 mM NaCl. Co-reconstitution of the Purified Components into Phospholipid Vesicles-The co-reconstitution of purified hamster lung PAR, purified bovine cerebral cortex G,, and resolved bovine caudate nucleus adenylate cyclase was performed as previously described (20). The protein-lipid pellets were resuspended in 1.2 ml of 75 mM Tris-HC1, pH 7.8, 1 mM dithiothreitol before assaying for agonist-stimulated adenylate cyclase activity. (3-Adrenergic ReceptorPhosphorylation by PAR Kinase-Reconstituted PAR (typically 1-5 pmol/incubation) was incubated with PAR kinase (0.1-0.5 pg) in a buffer containing 20 mM Tris-HCI, pH 7.5, 2 mM EDTA, 20 mM NaCl, 6 mM MgCl,, 6 mM sodium phosphate, 0.5 mM ascorbic acid, 60 p~ [y-32P]ATP(1-3 cpm/fmol) at 30 “C for 30 min or the time indicated in the figure legends. The incubations also contained one of the following: buffer (control), (-)-isoproterenol (0.02-0.2mM), Lilly46220(0.3-1mM), zinterol (0.01mM), (-)soterenol (0.01 mM), dichloroisoproterenol (0.01-0.1mM), dobutamine (0.3-1 RIM), or deoxyisoproterenol (0.2-1 mM). These concentrations were chosen to saturate the receptor binding sites (9). The incubations were stopped by the addition of SDS sample buffer followed byelectrophoresis on 10% homogeneous polyacrylamide gels. Stoichiometries of phosphorylation were determined by cutting and counting the dried gel as previously described (6, 19). Peptide Mapping of Phosphorylated Recept~r-~~P-Labeled receptor was prepared by incubating 5pmol of reconstituted BAR with 10 pl of BAR kinase under phosphorylating conditions for 1 h at 30 “C. The incubations also contained 150 p M (-)-isoproterenol, 300 p M Lilly 46220,15 p~ dichloroisoproterenol, or no ligand. Reactions were stopped by the addition of 1 ml of cold 100 mM NaC1, 10 mM Tris, pH 7.2, followed by centrifugation at 400,000 X g for 30 min. The resulting pellets were resuspended in 500 pl of 50 mM NHIHC03 containing 0.4 mg ofStaphylococcusaureus V8 protease and incubated for 5 h at 22 “C. The samples were then recentrifuged at 400,000 X g for 30 min. The supernatants(which contained all of the proteolyzed 3ZP-labeledBAR as assessed by SDS-polyacrylamide gel electrophoresis) were lyophilized, resuspended in 100 pl of 0.1% trifluoroacetic acid, and injected on a W-PorexC-18 reverse phase high performance liquid chromatography column. The 32P-labeledpeptides were eluted with a linear gradient from 0 to 70% acetonitrile in 0.1% trifluoroacetic acid. Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis-Gel electrophoresis was performed by the method of Laemmli (21) using 10% homogeneous slab gels. Sample buffer contained 8% SDS, 10% glycerol, 5% 8-mercaptoethanol, 25 mM Tris-HC1, pH 6.5, and 0.003% bromphenol blue. After electrophoresis, gels were immediately dried prior to autoradiography. Adenylate Cyche Assays-Adenylate cyclase was assayed as previously described (20) using the method of Salomon et al. (22).
by PAR Kinase RESULTS
Seven adrenergic agents were investigated in thisstudy. As shown in Table I, their intrinsic activities in the different assays varied from a very poor partial agonist, such as dichloroisoproterenol, to a full agonist, such as isoproterenol. Fig. 1 demonstrates typical results obtained when several of the partial agonists were utilized to stimulate adenylate cyclase activity ina three-component reconstituted system consisting of purified PAR and G,and resolved adenylate cyclase (20). It can be observed that typical sigmoidal dose response curves were obtainedinall cases. Withthe full agonist, isoproterenol, anapproximate 4-fold stimulation of basal adenylate cyclase activity was achieved, while several different partial agonistswere considerably less effective in stimulating adenylatecyclase. The intrinsic activities of the various drugs determined in the adenylate cyclase activation assay are summarizedin Table I. We next investigated the ability of each agent to promote p-adrenergicreceptor phosphorylation by the PAR kinase. These experiments utilized the same pure hamster lung padrenergic receptor which had been used in the reconstituted TABLE I Intrinsic actiuities of several compounds as assessed in a reconstituted system Adenylate cyclase activity in the three-component reconstituted system was assayed as described under “Experimental Procedures.” Isoproterenol-stimulated activity was -40-50 pmol of cAMP/30 min above basal (see Fig. 1). Phosphorylation of reconstituted PARby PAR kinase was performed as described under “Experimental Procedures.” The values are reported as the mean S.D. ( N ) and are corrected for the basal (no ligand) level of phosphorylation.
*
Ade~~~2~~$’ase Phosphorylation reconstituted
Compound
of
reconstitutedsystem
(-)-Isoproterenol (t)-Zinterol (-)-Sotereno1 (f)-Deoxyisoproterenol Dobutamine Lilly 46220 (+)-Dichloroisoproterenol
1.oo
1.00 0.88 f 0.02 (2) 0.77 f 0.07 (3) 0.66 f 0.04 (2) 0.51 f 0.05 (5) 0.43 f 0.13 (4) 0.09 f 0.03 (4)
0.88 f 0.13 (3) 0.76 f 0.14 (7) 0.63 f 0.13 (10) 0.47 f 0.13 (7) 0.46 f 0.05 (10) 0.11 f 0.02 (10)
100-
> I-
u 4
ZINTEROL
-
80-
:$
5u fu
60-
L’
0
W
gs
40LlLLY 46220
2 -
>-
5
20-
n
Q
0
8
7
6 5 -LOG [COMPOUND]
4 (M)
3
FIG. 1. Agonist activation of adenylate cyclase in a reconstituted BAR-G.-adenylate cyclase system. Phospholipid vesicles containing purified PAR, purified G., and resolved adenylate cyclase were prepared as described under “Experimental Procedures.” The lipid vesicles (20 pl) were then assayed for adenylate cyclase activity in the presence of the appropriate ligand concentration. Each point represents the mean of duplicate determinations from a single experiment which was replicated several times with comparable results. In the experiment shown, the basal activity (16 pmol of cAMP/30 min) was subtracted from all the data and is represented as 0% on the y axis. Maximum activity (100%)was 45 pmol of cAMP/30 min above basal.
Partial Agonist Stimulated Phosphorylation by PAR Kinase three-component adenylatecyclase assays. Typical resultsare shown in Fig. 2. In the absence of agonist, very little @adrenergic receptor phosphorylation could be observed ( l a n e 1).In the presence of the full agonist isoproterenol, approximately 10-fold stimulation of phosphorylation was observed (hne 2). As compared with isoproterenol, the drugs Lilly 46220 and dichloroisoproterenol promoted much less phosphorylation of the receptor (lanes 3 and 4 ) . Summarized in Table I are the intrinsic activities of all seven agents for promoting @-adrenergicreceptor phosphorylation by BAR kinase. The @-adrenergicantagonists tertatolol and pindolol, which have been reported to promote @-receptordown regulationin some systems (23-25), do not promote receptor phosphorylation by BAR kinase (data notshown). Fig. 3 shows a correlation plot of the intrinsic activities of the different drugs as determined both for adenylate cyclase activation in the three-component reconstituted systemand for promotion of receptor phosphorylation. It can be observed that there is an excellent correlation with a linear regression of y = 1 . 0 2 ~ - 0.01, an r value of 0.996, and a p value
