Printed in U.S.A.. Regulation of Stably Transfected Platelet Activating Factor. Receptor in RBL-2H3 Cells. ROLE OF MULTIPLE G PROTEINS AND RECEPTOR ...
Val. 269,No. 40,Issue of October 7, pp. 24557-24563, 1994 Printed in U.S.A.
THEJOURNAL OF Browcrc~~ CHEMISTRY 0 1994 by The American Society for Biochemistry and Molecular Biology, Inc.
Regulation of Stably TransfectedPlatelet Activating Factor Receptor in RBL-2H3 Cells ROLE OF MULTIPLE G PROTEINS AND RECEPTOR PHOSPHORYLATION* (Received for publication, May 2, 1994, and in revised form, July 11, 1994) Hydar Ai$§,Ricardo M. Richardson$, Eric D. TomhaveS, Robert A. DuBoseS, Bodduluri HaribabuS, and Ralph Snydermanm From the Departments of $Medicine and lllmmunology, Duke University Medical Centeer, Durham, North Carolina 27710
Platelet activating factor (PAF)interacts with cellsur- released frommast cells, platelets, neutrophils,monocytes, and face receptors to mediate inflammatory responses. To macrophages (1).PAF is a potent chemoattractant for neutrodetermine the mechanisms of PAF receptor regulation, phils and elicits various otherresponses such as coronary vaspermeability (reviewed in we constructed epitope-tagged human PAF receptor oconstriction and increased vascular cDNA (ET-PAFR)and generated stable transfectants in a Refs. 2 and 3). The effects of PAF on neutrophils are mediated, rat basophilic cell line (RBL-2H3 cells). The expressed at least in part, via the activation of cell surfacereceptors receptors displayed ligand binding and functional prop- coupled to G proteins (4-6). Whereas a single classof receptors erties similar to the native receptors in neutrophils. for the chemoattractants fMLP and C5a activate both Ca2+ PAF-stimulated intracellular Ca2+ mobilization was not mobilization and exocytosis in a pertussis toxin (PTx)-sensitive inhibited by pertussis toxin (PTx), whereas phosphomanner, different receptor subtypes are thoughtt o be involved inositide hydrolysis and secretion wereblocked by in PAF-stimulated responses (7-9). This contention is based on the observations that in neutrophils, Ca2+mobilization requires "40%. The PTx-resistant secretion mediated byPAF was, however, inhibited by guanosine 5'-0-(2-thio- a 10-fold lower concentration of PAF for maximal responsethan diphosphate) in permeabilized RBL-2H3 cells, indicat- that requiredfor lysosomal enzyme release. Furthermore, Ca2+ ing a role for PTx-insensitiveG protein. In contrast to mobilization is resistant to inhibition by PTx, whereas exocythe PAF receptor, responses mediatedby formylpeptide tosis is substantially inhibitedby this toxin. These findings led and C5a chemoattractants were inhibited by PTx. PAF to the suggestion that PAF-induced Ca2+mobilization results stimulated a dose-and time-dependent phosphorylation from activation of high affinity receptors, while PAF-induced of its receptor. ET-PAFRwas also phosphorylated by exocytosis involves the activation of low affinity receptors(7,s). phorbol12-myristate13-acetate(PMA) and dibutyryl cy- cDNA encoding a single PAF receptor has been isolated and clic AMP. Staurosporine caused complete inhibition of expressed in COS-7, Chinese hamster ovary, and undifferentiET-PAFR phosphorylation by PMA but only partial inhi- ated HL-60 cells (10-13). In these systems, PAF-stimulated bition byPAF'. Receptor phosphorylation byPAF and Ca2+mobilization and phosphoinositide hydrolysis were not sigPMA correlated with desensitization as measured by a nificantly inhibited by PTx (10, 14, 15). Whether or not the decrease in both PAF-stimulatedGTPase activity in same receptor alsointeracts with a PTx-sensitive G protein and membranes and CaZ+ mobilization in intact cells. Phos- binds PAF with low affinity in neutrophils to stimulateexocyphorylation ofET-PAFRby dibutyryl cyclic AMP was tosis has not yetbeen reported. not,however,associatedwithdesensitization.These Another unanswered issue relatedt o the PAF receptor is the data demonstrate that a singlePAF receptor population mechanism by which this receptor is desensitized. PAF causes interacts with multipleG proteins to mediate its biological responses. Moreover, ET-PAFR, unlike the formylpep- stimulation of both GTPase activity in membranes and intratide orC5a receptors, is phosphorylatedby at least three cellular Ca2+ mobilization in intact platelets and neutrophils kinases (most likely protein kinases A and C and a recep- (16-19). These responses are inhibited by prior incubation of tor kinase). The functional consequences of cellular acti- cells with PAF or other stimulants (fMLP, C5a, interleukin-8, vation by various chemoattractants may depend uponthe and PMA) (16, 17, 19). This type of homologous and heterologous desensitization appears to be a general phenomenon by G protein to which their receptor is coupled. which functions of many G protein-coupled receptors are reguhomologous and lated (20). In &-adrenergic receptors, both Platelet activating factor (PAF,' 1-0-alkyl-2-acetyl-sn-glyc-heterologous desensitization involve several distinct but interero-3-phosphocholine) is a mediator of inflammation that is related processes including receptor phosphorylation, uncoupling of receptor from G protein, sequestration, and finally, * This work was supported by Grants DE-03738 and CA-29589 from down-regulation (20). Both G protein-coupled receptor kinase the National Institutes of Health. The costsof publication of this article and protein kinase A play important roles in homologous dewere defrayed in part by the payment of page charges. This article must sensitization of &-adrenergic receptors, whereas only protein therefore be hereby marked Yadvertisement" in accordance with 18 kinase A is involved in its heterologous desensitization (21). In U.S.C. Section 1734 solely to indicate this fact. 5 To whom all correspondence should be addressed: Dept. of Medicine, contrast, heterologous desensitization of the C5a receptor is Duke University MedicalCtr., Box 3680, Durham, NC 27710. Tel.: 919- mediated by protein kinase C (22). In the present study, we sought to determine ( a ) whether a 681-6756; Fax: 919-684-5653. The abbreviations used were: PAF, platelet activating factor (1-0-single PAF receptor populationis capable of mimicking the PAFalkyl-2-acetyl-sn-glycero-3-phosphocholine); ET-PAFR,epitope-tagged induced biological responses in neutrophils and(6) the mechaPAF receptor; Bt,cAMP, dibutyryl cyclicAMP; GDPPS, guanosine 5'-0(2-thiodiphosphate);fMLP, formylmethionylleucylphenylalanine; PTx, nisms by which PAF receptors are desensitized. To this end,we pertussis toxin;PMA, phorbol 12-myristate 13-acetate;ET-FR, epitope- expressed ET-PAFR cDNAs into a functionally responsive and tagged fMLP receptor; ET-C5aR, epitope-taggedC5a receptor. biologically relevant cell line (RBL-2H3 cells) and immunopre-
24557
24558
PAF Receptor Phosphorylation Desensitization and
cipitated ET-PAFR using an antibody (12CA5) specific to the epitope tag. Here we describe the first affinity purification of recombinant PAF receptors and reportunexpected findings regarding regulationof receptor functionsby multiple G proteins and receptor phosphorylation by a t least three kinases.
excitation wavelength of 355 nm and an emission wavelength of 410 nm. Maximumand minimumfluorescence were determined in the presence of 0.1% Triton X-100 and 20 mM Tris-HC1, pH 8.0, 5 mM EGTA, respectively. Intracellular calcium concentrations were calculated using -F)) (24). the following formula: [Ca"], = K,.((F - Fmin)/(Fmu Phosphoinositide Hydrolysis and Secretion of 6-HexosaminidaseRBL-2H3 cells (0.2 x 1O6/well)were seeded in a 24-well tissue culture EXPERIMENTAL. PROCEDURES plate and cultured overnight with 1 pCi/ml [3Hlinositolin inositol-free M~terials-[~~PlOrthophosphate (8500-9120 Cilmmol), ['251]iodine medium supplemented with 10% dialyzed fetal bovine serum. Intact 17.0 Cilmg, myo-[2-sHlinositol (24.4 CUmmol), [3H]PAF(10.0 CUmmol), cells were washed with HEPES-buffered saline supplemented wit.h 10 C3H1WEB2086 (10.5Cilmmol) and [y3'1GTP (6000 CUmmol) were pur- mM LiCl, and 0.1% bovine serum albumin and preincubated with 200 pl chased from DuPont NEN. Monoclonal 12CA5 antibody was obtained of the same buffer for 10 min at 37 "C. Cells werethen stimulated with different concentrations of PAF. The reaction was terminated 10 min from BabCo. PSV2neo was obtained from American Tissue Culture ice. Supernatant, orcell lysate Collection. Geniticin (G418) and all tissue culture reagents were pur- after stimulationby placing the plate on chased from Life Technologies, Inc. Protein G-agarose and protease (10 pl), was removed, and releaseof P-hexosaminidase was determined inhibitorswerepurchased from BoehringerMannheim.Indo-1 ace- as described (22). [3Hlinositol phosphates were extracted into aqueous phase by the addition of 750 pl of chloroform, methanol, 4 N HCl toxymethyl ester and pluronic acid were purchased from Molecular (100:200:2); 250 pl of 0.1 N HCl; and 250 p1of chloroform. Total L3H1 Probes. Streptolysin 0 was obtained from Burroughs Wellcome. Construction of Epitope-tagged PAF Receptor cDNA-A 9-amino acid inositol phosphates were separatedon columns of Dowex formate. Expeptide sequence (YPYDVPDYA)was inserted between the N-terminal periments with streptolysin 0-permeabilizedRBL-2H3 cells were perinitiator methionine and the second amino acid of human PAF receptor formed as described (25, 26). Labeling of ET-PAFR with [125Z]Iodine-RBL-2H3 cells (5 x lo6) exby polymerase chain reaction. The5'-oligonucleotide, in a 5' to 3' order, pressing ET-PAFR, preincubated or not with 100 nM PAF for 5 min, were contained 3 miscellaneous bases, 6 bases encoding a HindIII site, 3 an iodogen-coated bases from the 5'-untranslated sequence preceding the ATG codon of resuspended in200 pl of phosphate-buffered saline in glass tube (100 &tube). Carrier-free NalZ5I (200 pCi) was added to the the human PAF receptor,3basesencodingamethionine,27bases tube and incubated for 10 min at 4 "C. Cells were removed, washed, to encoding the 9-amino acid epitope tag, and 24 bases complementary lysed, immunoprecipitated, and analyzed as describedbelow. the cDNA sequence encoding amino acids 2-9of PAF receptor. The 3'-oligonucleotide was complementary to the C-terminal 7 amino acids Phosphorylation of ET-PAFR-Phosphorylation of ET-PAFRwas carried out exactly as described (22). Briefly, RBL-2H3 cells (1.5 x lo6) and stop codon with a 5'-BamHI site. Polymerase chain reaction was carried out with these primers using the human PAF receptor cDNA as expressing ET-PAFR were seeded i n 60-mm tissue culture dishes in normal growth medium and cultured overnightat 37 "C. The following template (13). The resulting polymerase chain reaction product was day, each dish was washed twice with 5 ml of phosphate-free Dulbecco's digested with HindIII and BamHI and ligated into the plasmidvector 150 pCi of [32Plorthophosmodified Eagle's medium and incubated with pBluescript. The entire coding region, including the epitope tag, was sequencedandtransferredinto pRK5 expression vector as a ClaI- phate for 90 min. Adherent cells were stimulated, and the reaction was stopped by placing the samples on ice. All subsequent procedures, unBamHI fragment. less otherwise stated, were carried outa t 4 "C. The cells were washed Cell Culture a n d Dansfection-RBL-2H3 cells were maintained as mllysis bufferof monolayer cultures in Earle's modified Eagle's medium supplemented with phosphate-buffered saline and lysed with 0.75 of with 15% fetal bovine serum, 2mM glutamine, penicillin (100 unitdml), the following composition: 150 mM NaCl, 50 mM Tris, pH 8.0, 1.0% and streptomycin (100pg/ml). For transfection, cells were detached by Nonidet P-40,0.5%deoxycholate, 0.1%SDS, 5 m~ EDTA, 10 mM sodium versene treatment and washedtwice in Eagle's medium supplemented fluoride, 10 mM sodium pyrophosphate, 10 pg/ml leupeptin, 1 pg/ml pepstatin, 100 pg/ml l-chloro-3-tosylamido-4-phenyl-2-butanone, 50 with 50 mM HEPES and 20% fetal bovine serum. Cells were resusand 10 pg/ml phenylpended in the same medium at a density of 5 x 107/ml;and 1 x lo7cells pg/ml l-chloro-3-tosylamido-7-amino-2-heptanone were electroporated (Gene Pulser,Bio-Rad) at 250 V and a capacitance methylsulfonyl fluoride. The lysate was transferred toa 1.7-ml Eppenof 500 microfarads in thepresence of selective G418-resistant marker- dorf tube and incubatedon a rotating wheel for 1h. Insoluble material was removed by centrifugation (10,000 x g for 5 mid. The supernatant containing plasmid PSV2neo (1 pg) and mammalian expressionvector in thepresence ofimmunoglobupRK 5 (19 pg) containingcDNA for ET-PAFR. The culture medium was was precleared with protein G-agarose 0.2%). The prerepIaced with fresh medium containing 1 mg/ml of geniticin (G418) 24 lin-free bovine serum albumin (final concentration, cleared cell lysate was incubated with 10 pgof 12CA5 antibody for 60 h after transfection.Ten days after electroporation, the antibiotic resistant clones were analyzedfor cell surface expressionof ET-PAFR by flow min and incubated with protein G-agarose for 60 min to immunopretwice with lysis cytometry using FITC-12CA5 antibody (see below). The top 3% of the cipitate epitope-tagged receptors. Samples were washed cells expressingET-PAFR were subjected to two roundsof sorting, and buffer supplemented with 0.1% ammonium sulfate and once with the same buffer in the absence of ammonium sulfate. SDS-sample buffer the resulting mixed cell population was used throughout this study. to 80 "C for 10 min. Proteins Binding Assays-For flow cytometric analysisof ET-PAFR, cells were (50 pl) was added to the pellet and heated preincubated withrat IgG (10pg/ml for 10 min) to saturate Fc receptors were separated on a 10% SDS-polyacrylamide gel and analyzed by present on the surface of RBL-2H3 cells (22). The binding was carried autoradiography. GTPase Assay-Membranes from untreated or desensitized RBLout ina total volume of 200 plof HEPES-buffered Hanks' balancedsalt solution for 90 min at 4 "C using FITC-12CA5 antibody (1pg/ml). The 2H3 cells expressingET-PAFR were prepared, and GTPase activity was cells were then washed and analyzed on a Beckton Dickinson FACS performed as described (22, 27). cytometer. Radioligand binding ofc3H1PAF to intact adherent RBL-2H3 cells RESULTS expressing ET-PAFR was carried out in a total volume of 400 pl of Ligand Binding to Expressed ET-PAFR in RBL-2H3 Cellsbinding buffer containing 150m~ choline chloride, 10 m~ Tris-HC1, 10 Membranes prepared from RBL-2H3 cells expressing ETmM MgCl,, pH 7.5, and 0.25% lipid-free bovine serum albumin (13). Samples were incubated for 4h a t 4 "C. Cells were washed four times PAFR, but not untransfectedcells, specifically bound the selecwith ice-cold buffer (500 pl), lysedwith0.1 N NaOH (500 pl), and tive PAF receptor antagonist [3HlWEB-2086 in a saturable analyzed for radioactivity. Binding ofL3H1WEB 2086, a PAF receptor manner. Scatchard analysisrevealed that theequilibrium disantagonist, to membranes(100 pg) was carried outa t 25 "C for 90 min sociation constant (K,) and B,, were 8.7 2 0.4 I"and 9.7 3 0.7 using the same buffer (500 111). Unlabeled PAF (10 p)was used to pmol/mg of protein, respectively. Attempts to perform saturadetermine nonspecific binding in both intact cells and membranes. [3HlPAFwere unsucBound and free radioligand were separated by rapid filtration through tion binding studies in membranes using Whatman GF/C filters presoaked with 1%bovine serum albuminfor 60 cessful due tohigh (>go%) nonspecific binding. However, intact min (23). Filters were then washed five times with 4 ml of ice-cold cells expressing ET-PAFR, but not untransfected cells, bound incubation buffer. [3H]PAF in a highly specific manner (nonspecific binding Calcium Measurernents-RBL-2H3 cells (3x lo6) were loaded with1 -10%) with a Kd of 5.3 3 0.3 m. The affinities for binding of p~ indo-1 acetoxymethyl ester in the presence of 1 p~ pluronic acid for both antagonist to membranes andagonist to intact RBL-2H3 30 min at room temperature. The cells were then washed and resusfor the native receptors in pended in 1.5 ml of buffer. Calcium measurements were carried out in cells were similar to the values a Perkin-Elmer fluorescence spectrophotometer (model 650-19) with a n neutrophils and platelets (4,13, 28). This indicates that the
24559
PAF Receptor Phosphoryla:tion and Desensitization I
1
T
L l
40
x
20 0 0-HEX. - P T X
0
0.01
0.1
10
10 [PAF] nM
100
FIG.1. Functional characteristics of ET-PAFR expressed in RBL-2H3 cells. For Ca2+mobilization, RBL-2H3 cells(2 x 106/sample) were preincubatedwith or without PTx (200 ng/ml for 3 h), loaded with indo-1, and stimulated with different concentrationsof PAF. Data are presented as percent of net maximum response in the absence of toxin (602 5 10 nM). Points are means of two experiments performed in duby plicate. Thebasal response (130 e 3 m) was not significantly altered €"x. Assays for the generation of inositol phosphates (PHIIF'S) and secretion of P-hexosaminidase (P-HEX) were performed in the same batch of cells. RBL-2H3 cells were cultured overnight inthe presence of [3H]inositol(1pCi/ml) and treated with or without PTx(200 ng/ml for 3 h). Cells were preincubated (10 min, 37 "C) with a HEPES-buffered saline containing10 mM LiCl in a total volume of 200 pl and stimulated with different concentrations of PAF for 10 min. Supernatant (10 plj was removed for the secretion assay, and the rest of the sample was used to determine the release of 13H]inositolphosphates. Maximum responses for [3Hlinositol phosphate (10,000 e 6 cpm) secretion(80* 3% of total p-HEX present in cell) were determinedafter basal responses (1000 z t 3 cpm for [3H1insoitol phosphate and 5.0 e 0.3% for &hexosaminidase) were subtracted. Data are presented as percent of net maximum stimulation in the absence of toxin. Points are mean e S.E. of a single representative experimentperformed in triplicate.
PAF
PAF
PMA/Ca2+
F'Ic. 2. Effect of GDPSS on PAFandPMA/Ca"-induced
responses in PTx-treated permeabilizedRBL-2H3 cells. RBL-2H3 cells were cultured overnight in the presence of [3H]inositoland were incubated with PTX (200 ng/ml, 3 h). Cells were then permeabilized with streptolysin 0, preincubated with GDPpS (300 VM,5 min), and stimulated witheither PAF (100 m) or PMA (100 nM) in the presence of free-Ca" concentration of 1 PM for 10 min. Generation of [3Hlinositol phosphates ( f H I I P s )and secretion of P-hexosaminidase (P-HEX)were determined as described in the legend to Fig. 1, Data are mean r S.E. of one of three similar experiments.
and GDPPS was added. As shown in Fig. 2, GDPpS (300 p ~ ) caused a substantial inhibition of PAF-induced responses but barely affected secretion induced by PMA and Ca". Immunoprecipitation of ET-PUR from RBL-2H3 Cells-We previously used the monoclonal 12CA5 antibody to selectively immunoprecipitate epitope-tagged M L P a n d C5a receptors (ET-FR and ET-C5aR) expressed inRBL-2H3 cells (22). We now tested whetherthis antibody could also immunoprecipitateETPAFR. AS shown in Fig. 3, in the absence of PAF stimulation, 12CA5 immunoprecipitated a broad lz5I-and a faint 32P-labeled 9-amino acid hydrophilicgroup at the N terminus of the trans- protein band a t molecular mass of 42 kDa and a sharp band at fected PAF receptor has no effect on its ligand bindingaffinity. 90 kDa. Upon ligand stimulation, the intensity of the 32P-laHowever, the tagged receptors in RBL-2H3 cells, as well as the beled but not the lz5I-1abeledbands increased substantially. wild type receptor inCOS-7 and Chinese hamster ovary cells, Following ligand stimulation, both thelower and upper bands were expressed at a much higher level than the native recep- displayed reduced mobility on SDS-PAGE. The 90 kDa band tors in neutrophils and platelets (12, 13). may representa receptor dimer composed of both slow and fast Effect of PTx a n d GDPPS on PAF-induced Responses in RBL- migrating forms. The epitope tagpeptide (YPYDVPDYA), 2H3 Cells Expressing ET-PAFR-The expressed tagged recep- against which the 12CA5 antibody was raised, inhibited the tors were functionally active and mediated many of the bio- immunoprecipitation of lZ5I-and 32P-labeledproteins (Fig. 3, A chemical and functional responses do as the native receptors in and B , lanes 3 and 4 ) . These data demonstrate that the immuneutrophils. In transfected RBL-2H3 cells, PAF stimulated the noprecipitated protein bands are ET-PAFR. Furthermore, PAF hydrolysis of inositol phospholipids, mobilization of Ca2+, and caused phosphorylation of ET-PAFR, which was associated exocytotic release of P-hexosaminidase in adose-dependent with a decrease in theelectrophoretic mobility of the receptor; manner. Thephosphoinositide hydrolysis and Ca2+ mobilization a phenomenon noted for other seven-transmembrane domain were detectable at a PAF concentration of 0.01 nM, whereas receptors (29-31). >0.3 n~ was required to induce detectable secretion (Fig. 1). Dose- and Time-dependent Phosphorylation of ET-PAFR by The EC,, values for Ca2+mobilization, phosphoinositide hydrol- PAF-To determine the dose and time dependence of phosphoysis, and secretion were0.15 z 0.02, 3.3 5 0.03, and 2.2 5 0.01 rylation of ET-PAFR, 32P-labeled RBL-2H3 cells were stimunM, respectively. lated with increasing concentrations of PAF or with a saturatWe next determined whether responses toPAF in RBL-2H3 ing concentrationof the ligandfor different times. The extent of cells were inhibited by PTX. As shown in Fig. 1, treatment of phosphorylation was quantitated by Cerenkov counting of exRBL-2H3 cells with PTx (200ng/ml, 3 h) had no effect on cised gel bands. As shown in Fig. 4, PAF stimulated dose-deintracellular Ca2+ mobilization. Under these conditions, Ca2+ pendent phosphorylation of both 42 and 90 kDa bands. Phosmobilization in response to M L P or C5a in RBL-2H3 cells phorylation of ET-PAFR was detectable a t a concentration of expressing ET-FR or ET-C5aR was completely inhibited (data 0.1 nM PAF and reached a maximum at 10 nM with an EC,, of cells with PTx value of 1 nM. Phosphorylation of ET-PAFR by PAF was exnotshown).Furthermore,preincubation (200 ng/ml) for up to 16 h had no effect on PAF-induced Ca2+ tremely rapid(Fig. 5). Incorporation of 32Pinto thereceptor was mobilization (data not shown). detectable 5 s after stimulation, reachedhalf-maximal at 10 s, PTx did, however,cause -40% decrease inboth inositol phos- and wascomplete within 1min. Incubation of cells with PAF for phate production and exocytosis without altering theEC,, Valup to 10 min resulted in no further change in the extent of ues (Fig. 1).To determine whether thePTx-insensitive compo- phosphorylation of either bands. nent of PAF-induced generation of [3H]inositol phosphates and Effect of Staurosporine on PAF a n d PMA-stimulated Phosexocytosis were mediated via PTx-insensitive G proteins, cells phorylation of ET-PAFR-To determine theinvolvement of prowere treated with PTx andpermeabilized with streptolysin 0 , tein kinaseC on PAF-induced phosphorylation of ET-PAFR, we
PAF Receptor Phosphorylation Desensitization and
24560
(6)32 P
(A)?
97 97"
FIG.3. Immunoprecipitation of ETPAFR by 12CA5 antibody. Intact RBL2H3 cells expressing ET-PAFR were labeled with 12,'1 ( A )and "P ( B )and exposed to vehicle or PAF (100 nM) for 5 min. Cell lysateswereimmunoprecipitatedwith 12CA5 antibody in the absence or presence of the epitope tag peptide (20 pg) a s indicated. Samples were separated on a 10% SDS-polyacrylamide gel underreducing conditions and autoradiographed for 2days for '"I and 14 h for R2Pa t -80 "C.
-
kDa
kDa
4 6 -
30 +
30
PAF(100nM) 1X A 5 Peptide
-
+
A
kDa
+
+ +
-
+
+
+ +
kDa
A
46-
464
5'
0
0
0.1 .1
3
15' 30'
10 33 100 1000
1' 2.5' 5'
10'
TIME
[PAF] nM
7500 O
5000
a
B
i
3000 0
0
ck" tl
2500 0
l
A
1 [PAF] nM
~
0
0
=
1000
~
~
5
t
.
.
n
10
TIME (min) FIG.5 . T i m e course of ET-PAFR phosphorylation.A, 32P-labeled
FIG.4. Dose-dependent phosphorylation of ET-PAFR A, 32PPAF (100 nM), labeled RBL-2H3 cells expressing ET-PAFR were stimulated with dif- RBL-2H3 cells expressing ET-PAFR were stimulated with a t different timesby placing the sampleson ferent concentrationsof PAF for 5 min, and the reactions were stopped and reactions were stopped ice. Cells were lysed with detergent buffer and immunoprecipitated by placing samples on ice. Cells were washed, lysed with detergent with 12CA5 antibody. Samples were then separated on a 10% SDSbuffer, and immunoprecipitated with 12CA5 antibody. Samples were then separatedon 10% SDS-polyacrylamidegel. The gel was autoradio- polyacrylamide gel. The gel was autoradiographed for 14 h. B , the graphed for 14 h. B , the amountof radioactivityflane was determinedby amount of radioactivityflane was determinedby counting excised phosphorylated bands. counting excised phosphorylated bands.
tested the effect of increasing concentrations (0.01-10 p ~ of) the protein kinase C inhibitor staurosporineon ligand-induced receptor phosphorylation. As shown in Fig. 6, staurosporine partially inhibited ET-PAFR phosphorylation in response to PAF. PMA also caused phosphorylation of ET-PAFR, and this response was totally inhibitedby staurosporine. TheIC,, value
for inhibition of PMA-stimulated phosphorylation of ET-PAFR by staurosporine was -200 nM. High concentration of staurosporine, (10PM) completely inhibited the basal, but not PAFinduced phosphorylation of ET-PAFR. Phosphorylation of ET-PMR, ET-CSaR, and ET-FR by BtFAMP-To determine whether ET-PAFR is susceptible to
a
*
PAF Receptor Desensitization Phosphoryl( and ztion A kDa
97
-
46
-
PAF(100nM) PMA(100nM) STAU. (uM)
- -
- - - - - + + + + + +
-
+
0
0 .01 0.1 1 3.3 10 10 0 .010.1
-
t
+
t
+ +
- - - - - -
1 3.3 10
24561
GTPase activity in membranes with an EC,, value at 8.1 2 2.0 nM. However, membranes prepared from cells pretreated with PAF (100 nM, 5 min) caused a shift in theEC,, value to30 2 1.6 nM and a decrease in themaximal responseby -50%. PMA, but not Bt,cAMP, caused desensitization of PAF-induced GTPase activity in a heterologous manner (Fig. 8B). Stimulation of indo-1 loaded RBL-2H3 cells with an EC,, dose of PAF (0.2 nM) caused a rapid and transient increase in Ca2+mobilization. This response was desensitized (by -50%) by the samedose of PAF (Fig. 9A). PMA almost completely inhibited PAF-induced Ca2+mobilization (Fig. 9B). In contrast,Bt,cAMP had no effect on Ca2+mobilization in response to PAF (Fig. 9C) or fMLP and C5a in cells expressing their respective receptors (data not shown). DISCUSSION
Y
'
$
I
x
50-
W
5
0
a
0
O
O
0.01
0.1
1
10
[STAU], U M FIG.6. Effect of staurosporineon PAF andPMA-induced phosphorylation of ET-PAFR "P-labeled RBL-2H3 cells expressing ET-PAFR were preincubated with or without different concentrations of staurosporine for 5 min and then stimulated with either PAF (100 nM; O), PMA (100 nM; O), or vehicle (0)for an additional 5 min. Cells were lysed, immunoprecipitated, and analyzedas described in thelegend to Fig. 4. Panel A shows autoradiogram from one of four representative experiments.Panel B shows the extentof phosphorylation a s determined by counting excised gel bands. Data are mean f S.E.of four separate experiments and are represented as percentage of maximum phosphorylation obtained with 100 nM PAF in the absence of staurosporine.
phosphorylation by protein kinase A, we tested the abilityof a membrane permeable CAMP analog, dibutyryl CAMP (Bt,cAMP) to phosphorylate ET-PAFR. Although Bt,cAMP ( 5 mM, 5 min) caused phosphorylation of ET-PAFR in a reproducible manner, the extentof this phosphorylation was small but consistently -30% over basal. This compares with a 2-3-fold increase for PMA and a 5-6-fold increase for PAF-induced responses (Figs. 4-61. Increasing the concentration of Bt,cAMP up to 10 mM or the incubation time to 30 min did not result in any further increase in ET-PAFR phosphorylation (data not shown). We utilized staurosporine to determine theselectivity of Bt,cAMP-induced phosphorylation of ET-PAFR. As shown in Fig. 7A, staurosporine completely inhibitedbasalbut not Bt,cAMP-induced receptorphosphorylation. We have previously demonstrated that PMA causes phosphorylation of ETC5aR but not ET-FR expressed in RBL-2H3 cells. We, therefore, tested whether these receptors also display differential sensitivity toBt,cAMP. As shown in Fig. 7, B and C, Bt,cAMP (10 mM) caused no phosphorylation of either receptor. Furthermore, such treatment had no effect on fMLP- or C5a-induced phosphorylation of their respective receptors. Desensitization of ET-PUR-To determine whetherreceptor phosphorylation is associated with desensitizationof ET-PAFR, we tested the effect of PAF, PMA, and Bt,cAMP pretreatment on ligand-induced responses in RBL-2H3 cells. Both agoniststimulated GTPase activity in membranes and intracellular Ca2+mobilization in intactcells were used as assays. As shown in Fig. 8A, PAF stimulated adose-dependentactivation of
The data presented herein clarify a number of issues related to the regulation of PAF receptors. In neutrophils, PAF stimulates intracellular Ca2+ mobilization and degranulation with different potencies as well as differential susceptibility to PTx. Based on this observation, it has been postulated that PAF mediates Ca2+mobilization and degranulation via activation of distinct PAF receptor subtypes (7, 8). Our data, usinga cloned receptor cDNA, clearly demonstratethat a single receptor population is capable of mediating both Ca2+mobilization and secretion. As in human neutrophils, the generation of inositol phosphates in response to PAF is partiallyinhibited by PTx in RBL-2H3 cells. The PTx-independent component of this response is, however, inhibited by GDPpS in permeabilized RBL2H3. This indicates that PAF can transduce signals via coupling to both PTx-sensitive and FTx-insensitive G proteins. The concentration of PAF (-1 nM) that causes maximum Ca2+mobilization is below the EC,, value (3.3nM) for the generation of inositol phosphates (Fig. 1).Interestingly, PAF-mediated phosphoinositide response, but not Ca2+mobilization, is partially inhibited by PTx in both RBL-2H3 cells (Fig. 1)and neutrophils (6). Since only a small increase in inositol-1,4,5-trisphosphate is required to elicit maximum Ca2+mobilization, it islikely that PTx-independent production of inositol phosphates by low concentrations of PAF are sufficient to inducemaximumCa2+ mobilization (32). Receptor phosphorylation has been shown to be a common mechanism by which many G protein-coupled receptors are desensitized(20). We have previously shown that epitopetagged fMLP and C5a chemoattractant receptors (ET-FR and ET-C5aR) are phosphorylated and desensitized following binding of their respective ligands. However, ET-C5aR, but not ETFR, is also phosphorylated by protein kinase C. This difference in phosphorylation is correlated with desensitizationby protein kinase C (22). The present studyprovides the first demonstration that thePAF receptor undergoes agonist-stimulatedphosphorylation. Phosphorylation of ET-PAFR appears tobe mediated by multiple kinases,including proteinkinase c , possibly a G protein-coupled receptor kinase, and protein kinase A. This conclusion is based on the observation that staurosporinecompletely inhibited PMA-induced receptor phosphorylation, whereas itonly partially inhibited the phosphorylation induced by PAF. Phosphorylation of ET-PAFR by PAF and PMA were correlated with desensitization, as measuredby attenuation of agonist-stimulated GTPase activity in membranes andinhibition of intracellular Ca2+mobilization in intact cells. In contrast, ET-PAFR phosphorylation by Bt,cAMP did not correlate with desensitization. This may be due to the low level of ETPAFR phosphorylation by Bt,cAMP. The lack of Bt,cAMP effect on functional responsesin RBL-2H3 cells is, however, inconsistent with previous findings in platelets where it inhibited the maximal GTPase activityin response to PAF without affecting
PAF Receptor Phosphorylation and Desensitization
24562
FIG.7. Phosphorylation of ETPAFR, ET-FR, and ET-C5aR by Bt,cA". A, ,'"P-labeled cells expressing
97
ET-PAFR were preincubated withor without different concentrations of staurosporine for 5 min and thenexposed to vehicle or Bt,cAMP (5 mM) for an additional 5 min. ( B and C ) ,RBL-2H3 cells expressing either ET-FR or ET-C5aR werelabeled with ?lP, incubated with vehicle or BQcAMP (10 mM, 5 min), and then exposed to fMLP, C5a, or no stimulantfor 5 min.Cellswerelysed,immunoprecipitated,andanalyzedasdescribedinthe legendtoFig.3.The gel wasautoradiographed for 40 h (A ), for 48 h ( B), and for h 72 ( C ) .
0 0
(A) ET-PAFR
-
(C)ET-C5aR
(B) ET-FR
kDa
46
-
c bCAMP (mM) Stau. ( p ~ )
0
0
0
5
5
5
0
1
5
0
1
5
0
0
+
" _
FMLP (1 P ~ )
"
"
"
C5a (100 nM)
-
"
"
"
"
UNTREATED PAF-TREATED
45(
A
12(
10
10
-
+
"
I
t
PAF (0.2 nM)
t
PAF (0.2 nM)
B [PAF] nM
L
O UNTREATED
E
A
2
-0
PMA-TREATED Bt2cAMP-TREATED
t
0 II)
m
a t a
PMA (100 nM)
t
PAF (0.2 nM)
50
C 0
0.1
1.0
10
100
10
[PAF] nM FIG.8. Desensitization of PAF-stimulated GTPase activity in membranes. Membranes were prepared from control (untreated) or cells treated for 5 min with the following: PAF (100 nM), PMA (100 nM), or BhcAMP (5 mM). These membranes were then assayed for GTPase activity a t different concentrations of PAF. Data are presented as percent of net maximal stimulation in untreatedcells (% of control). The basal and PAF-stimulated GTPase activity were 6.8 f 0.2 and 14.5 f 0.1 pmoVmg of protein, respectively. Data are mean? S.E. of three experiments performed in triplicate.
the EC,, value (17). Protein kinase A has been shown to phosphorylate many G protein-coupled receptors. However, the functional consequence of this phosphorylation appears to be receptor specific (33-35). Based on pharmacological and biochemical studies, neutrophils and platelets are thought to express different types of PAF receptors (4,28). This difference may explain the differential regulation byBt,cAMPof PAF receptors in the neutrophil-like RBL-2H3 cells and platelets. The human PAF receptor possesses 11 serines and threonines in thepredicted cytoplasmic tail. Two of these sites conform t o the consensus sequence (lUKYar(Sm)X(R/KVrfor potential phosphorylation by protein kinase C (36, 37). This compares with only one such site in the third cytoplasmic loop of the C5a receptor and none in anyof the cytoplasmic domains
t
Bt2cAMP (5 mM)
f
PAF (0.2 nM)
FIG.9.Desensitization of PAF-stimulated Ca2+ mobilization in intact cells. A, RBL-2H3 cells were loaded with indo-1 and stimulated with an EC,, dose of PAF (0.2 nM). After 5 min, cells were rechallenged with the same concentration of PAF. B, cells were preincubated with PMA (100 nM, 5 min) and stimulated with PAF (0.2 nM). C , cells were preincubated with Bt,cAMP (5 mM, 5 min) and stimulated with PAF. Intracellular Ca2+ concentrations were determineda s described under "Experimental Procedures."
of the fMLP receptor. PAF and C5a receptors, but not fMLP receptors, also possess numerous otherless favorablephosphorylation sites for protein kinase C. This difference appears to be correlated withthe susceptibility of PAF and C5a receptors, but not fMLP receptors, to undergo protein kinase C-mediated receptor phosphorylation and heterologous desensitization (22). ET-PAFR in RBG2H3 cells undergo moderate basal phosphorylation in theabsence of ligand stimulation (Fig. 6). The ability of staurosporine to completely inhibit basal and PMA but not PAF and Bt,cAMP-mediated ET-PAFR phosphorylation in-
PAF Receptor Phosphorylation dicates a role for protein kinase C in basal phosphorylation of this receptor. In contrast to numerous phosphorylation sites for protein kinase C, the cytoplasmic tail of the humanPAF receptor possesses only one consensus sequence(R(FUKXS)for phosphorylation by protein kinase A (37). This difference, in addition to the basal protein kinase C-mediated phosphorylation of ET-PAFR, may explain what appears tobe weak phosphorylation by Bt,cAMP. The physiological significance of protein kinase A mediated phosphorylation of the PAF receptor remains to be determined. Thedatapresentedhereindemonstratethat cloned ETPAFR cDNA expressed in RBL-2H3 cells mediate many of the biochemical and functional responsesas do the nativereceptors in neutrophils via interaction with PTx-sensitive and PTx-insensitive G proteins. ET-PAFR appears tobe phosphorylated by multiple kinases including, protein kinase C, protein kinase A, and presumably a G protein-coupled receptor kinase. This is in contrast toM L P receptor, which is phosphorylated only by a G protein-coupled receptor kinase, and C5a receptor, which is phosphorylated by bothprotein kinase C and a G proteincoupled receptor kinase (22, 38). Both of these receptors are resistant to phosphorylation byBt,cAMP. G protein-coupled receptor kinases are, for the most part, cytosolic enzymes that are translocated to the membrane upon ligand stimulation. cDNA encoding six G protein-coupled receptor kinases have thus farbeen identified, and five of them have been expressed indifferent cell linesand characterized (39-45).Substrate specificity has been established only for G protein-coupled receptor kinase 1, which phosphorylates light-activated rhodopsin. However, it has been shown that in human peripheral blood leukocytes, PAF but not C5astimulates translocation of a G protein-coupled receptor kinase (possibly G protein-coupled receptor kinase 2)from the cytosol to the membrane (46). It is, therefore, likely that different G protein-coupled receptor kinasesare involved inthe phosphorylation of different chemoattractant receptors. The data presented here further underscore the selectivity of chemoattractant receptor phosphorylation and indicates subtle differences in the mechanisms of their desensitization, which may have importantroles in the regulation of the inflammatory response.
and Desensitization
24563
Biochem. Biophys. Res. Commun. 130,677-684 8. Naccache, P. H., Molski, M. M., Volpi, M., Shefcyk, J.,Molski, T. F. P., Loew, L.,
Becker, E. L., and Sha’afi, R. I. (1986) J. Leukocyte Biol. 40,533-548 9. Kroegel, C., Yukawa, T., Westwick, J., and Barnes,P. J. (1989) Biochem. Biophys. Res. Commun. 162,511-521 10. Ye, R. D., Prossnitz, E. R., Zou, A., and Cochrane, C.G. (1991) Biochem. Biophys. Res. Commun. 180,105-111 11. Nakamura, M., Honda, Z., Izumi, T., Sakanaka, C., Mutoh, H., Minami, M., Bito, H., Seyama, Y., Matsumoto, T., Noma, M., and Shimizu, T. (1991) J. Biol. Chem. 266, 20400-20405 12. Honda, Z., Nakamura, M., Miki, I., Minami, M., Watanabe, T., Seyama, Y., Okado, H., Toh. H., Ito, K., Miyamoto, T., and Shimizu, T. (1991) Nature 349,342-346 13. Kunz, D., Gerard, N. P., and Gerard, C. (1992)J . Riol. Chem. 267,9101-9106 14. Murphy. P. M., Gallin, E. K., and Tiffany, H. L. (1990)J. Immunol. 145,22272234 15. Amatruda, T. T., 111, Gerard, N. P., Gerard, C., and Simon, M. I. (1993)J. Biol. Chem. 268, 10139-10144 16. Shukla, S. D., Morrison, W. J., Dhar,A. 11989)Mol. Pharmacof. 35, 409-413 17. Homma, H., and Hanahan, D. J. (1988)Arch. Biochem. Biophys. 262,32-39 18. OFlaherty, J. T., Jacobson, D. P., and Redman, J. F. 11992) Biochem. J. 288, 241-248 19. Tomhave, E. D., Richardson, R. M., Didsbury, J . R., Menard, L., Snyderman, R., and Ali, H. (1994) J. Zmmunol. 153,3267-3275 20. Dohlman, H. G., Thorner, J., Caron,M. G., and Lefiowitz, R.J. (1991)Annu. Reu. Biochem. 60,653-688 21. Hausdorff, W. P., Bouvier, M., ODowd, B. F., Irons, G. P., Caron, M. G., and Lefiowitz, R. J . (1989) J. Biol. Chem. 264, 12657-12665 22. Ali, H., Richardson, R. M., Tomhave, E. D., Didsbury, J . R.. and Snyderman,R. (1993)J . Biol. Chem. 268,24247-24254 23. Ukena, D., Dent, G., Birke, F. W.. Robaut, C., Sybrecht, G.W., and Barnes, P. J. (1988) FEES Lett. 228,285-289 24. Cobbold, P. H., and Rink, T.J . (1987) Biochem. J . 248, 313-328 25. Ali, H., Cunha-Melo,J. R., Saul, W. F., and Beaven, M. A. (1990)J. Biol. Chem. 265,745-753 26. Ozawa, K., Szallasi, Z., Kazanietz, M. G., Blumberg, l? M., Mischak, H., Mushinski, J . F.,and Beaven, M. A. (1993)J . Biol. Chem. 268, 1749-1756 27. Cassel, D., and Selinger,Z. (1977)Proc. Natf.Acad. Sci. U. S. A. 74,3307-3311 28. Hwang, S.-B. (1991) Ann. N.Y. Acad. Sci. 629, 217-226 29. Stadel, J. M., Nambi, P., Shorr, R. G. L., Sawyer, D. F., Caron, M. G., and Lefiowitz, R. J . (1983)Proc. Natl. Acad. Sei. U. S. A. 80, 3173-3177 30. Klein, P., Vaughan, R., Borleis, J., and Devreotes,P. 11987)J . B i d . Chem. 262, 358-364 31. Richardson, R. M., Kim, C., Benovic, J. L., and Hosey, M. M. (1993) J. Biol. Chem. 268,13650-13656 32. Krause, K. H., Schlegel. W., Wollheim, C. B., Anderson, T., Waldvogel, F. A,, and Lew, P. D. (1985)J. Clin. Znuest. 76, 1348-1354 33. Bates, M. D., Olsen, C. L., Becker, B. N., Albers. F. J., Middleton, J. P.,Mulheron, J . G., Jin. S.-L. C., Conti, M., and Raymond, J. R. (1993) J. Biol. Chem. 268,14757-14763 34. Bouvier, M., Collins, S., ODowd, B. F.,Campbell, P. T., deBlasi, A,, Kobilka, B. K., MacGregor, C., Irons, G. P., Caron, M. G., and Lefkowitz, R. J . (1989)J. Biol. Chem. 264, 1678616792 35. Chen, Y., and Yu, L. (1994) J . B i d . Chem. 269,7839-7842 36. Kemp, B. E., and Pearson, R. B. (1990) ‘Dends Biochem. Sci. 15, 342-346 37. Kennelly, P. J., and Krebs, E. G. (1991) J. Biol. Chem. 266,15555-15558 38. Tardif, M., Merey, L., Brouchon, L., and Boulay, F. (1993) J. Immunol. 150, 3534-3545 Acknowledgments-We thank Dr. Craig Gerard for the gift of cDNA 39. Kunapuli, P., and Benovic, J . L. (1993)Proc. Natl. Acad. Sci.U. S. A . 90,5588encoding the human PAF receptor. 5592 40. Haribabu, B., and Snyderman, R. (1993) Proc. Natf. Acad. Sci. U. S . A . 90, REFERENCES 9398- 9402 1. Camussi, G. M.. Aglietta, M., Coda, R., Bussolino, F., Piacibello, W., and Tetta, 41. Inglese, J., Freedman, N. J., Koch, W. J., and Lefkowitz, R. J. (1993)J. Bid. C. (1981) Immunology 42, 191-199 Chem. 268, 23735-23738 2. Snyderman, R., and Uhing, R.J. (1992) in Inflammation: Basic Principles and 42. Premont. R. T., Koch, W. J., Inglese, J., and Lefkowitz, R. J . (1994) J. B i d . Clinicel Correlates (Gallin, J. I., Goldstein, I. M.. and Snyderman, R., eds) Chem. 269,6832-6841 pp. 421-439, Raven Press Ltd., New York 43. Haga, K., and Haga, T. (1992) J. Biol. Chem. 267, 2222-2227 3. Hanahan. D. J. (1986)Annu. Reu. Biochem. 55,483-509 44. Koch, W. J . , Inglese, J.. Stone, W. C., andLefkowitz, R., J. (1993)J. Biol. Chem. 4. Hwang, S.-B (1988) J . Biol. Chem. 263, 3225-3233 268,8256-8260 5. Lad, P. M., Olson, C. V., andGrewal, I. S. (1985) Biochem. Biophys. Res. 45. Pitcher, J. A., Inglese, J., Higgins, J . B., Arriza, J. L., Casey, P. J., Kim, C., Commun. 129.632438 Benovic, J . L.. Kwatra, M.M., Caron, M. G., and Lefiowitz, R. J. (1992) 6. Verghese, M. W., Charles, L., Jakoi, L., Dillon, S. B., and Snyderman,R. (1987) Science. 257, 1264-1267 J . Zmmunol. 138,4374-4380 46. Chuang. T. T., Sallese. M., Ambrosini, G.. Parruti, G . , and deBlasi,A. (1992)J. 7. Naccache, P. H., Molski, M. M., Volpi, M., Becker, E. L., and Sha’afi,R. I. (1986) Biol. Chem. 267,68866892
