Vol. 267, No . 22, Issue of August 5, pp. 15909-15915,1992 Printed in U.S.A.

THEJOURNALOF BIOLOGICAL CHEMISTRY 01992 by The American Society for Biochemistry and Molecular Biology, Inc

Inhibition by Dexamethasone of P3-Adrenergic Receptor Responsiveness in 3T3-F442AAdipocytes EVIDENCEFORA

TRANSCRIPTIONAL MECHANISM* (Received for publication, October 31, 1991)

Bruno FeveS,Beatrice BaudeQ, Stephane KriefQ, A. Donny StrosbergQ,Jacques PairaultS, and Laurent J. EmorineBY From the $U 282 Znstitut National de la Sand et de la Recherche Medicale, Centre National de la Recherche Scientifique, H6pital Henri Mondor, 94010-Creteil, SUPR0415 Centre National de la Recherche Scientifique, and Uniuersiti Paris VU, Znstitut Cochin de Genetique Mokculaire, 22 rue Mechain, 75014-Paris,France

Subclones of the murine 3T3 fibroblastic cell line which undergo adipose conversion in vitro (Green and Kehinde, 1974, 1976) have been extensively used as representative models for studying @-adrenergiccontrol of fat cells functions. A recent report (Guestet al., 1990) has suggested that differentiation of the 3T3-Ll subclone was accompanied by the induction of a principal @2ARcomponent and by the downregulation of the @lAR population expressed on preadipocytes. The presence of @2ARsin this clone has, however, previously been shown to depend on the addition tothe culture medium of glucocorticoids which are necessary for a complete adipocytic differentiation (Lai et al., 1982; Nakada et al., 1987). In cells of another clone, 3T3-F442A, where differentiation occurs without addition of corticoids, we have shown that @lARswere induced upon spontaneous adipose conversion, whereas they were down-regulated, and @2ARs up-regulated, in the presence of glucocorticoids (Fkve et al., 1990). The net effect of this corticoid treatment was mainly a reversal in the @1/@2ratio. In the mean time, the total number of 81 plus @2ARswas only slightly decreased, although the production of CAMPinduced by stimulation with @-adrenergicagonists was strongly depressed. Such a discrepancy between PAR density and @-adrenergic responsiveness is reminiscent of the situation encountered in rodent adipocytes (Muzzin et al., 1988; Arch, 1989; Zaagsma and Nahorski, 1990). From these and many other reports it and @2ARs,“atypiAdrenal corticoids play a major role in the regulation of was inferred that, beside the classical @l cal” BARS could be involved in the control by catecholamines lipid metabolism. It is therefore importantto study their of energy balance in adipose tissues and skeletal muscle (Arch, effects on the expression of PARS’ which are involved in the 1989; Hollenga et al., 1990; Challis et al., 1988). Direct evicontrol by catecholamines of fatty acid storage and mobilizadence for the existence of such receptors has come from the tion. characterization of a human and a murine gene (Emorine et * This work was supported by grants from the Centre National de al., 1989; Nahmias et al., 1991; Muzzin et al., 1991;Granneman la Recherche Scientifique, the Institut National de la Sant6 et de la et al., 1991) encoding a third PAR subtype, termed the @3AR. Recherche MBdicale, the MinistBre de la Recherche et de la Technol- The pharmacological profile and tissue distribution of the ogie, the Bristol-Myers-Squibb Company (Princeton, NJ), the Asso- latter have suggested its identity with the postulated atypical ciation pour le D6veloppement de la Recherche sur le Cancer, the receptors. This was substantiated by the direct characterizaLigue Nationale Frantaise Contre le Cancer, and theUniversity Paris tion in 3T3-F442A adipocytes of a major population of @3ARs VII. The costs of publication of this article were defrayed in part by which are functionally coupled to adenylate cyclase and lipolthe payment of page charges. This article must therefore be hereby

Modulation B3-adrenergic of receptor (B3AR) expression by dexamethasone was investigated in the murine 3T3-F442A adipocytic cell line. In untreated cells, a majorpopulationofbinding sites (62,000114,000 sites/cell) of low affinity for (-)-I3H1 CGP12177 and (-)-[‘261]iodocyanopindolo1 (corresponding to theB3AR subtype) was present along with a minor population (6,500-8,000 sites/cell) of sites of high affinity for the radioligands (correspondingto a mixture of theB1 and 82AR subtypes). Long-term exposure of the cells to 250 nM dexamethasone led to a sharp decreasein 83AR density (less than 5,000 sites1 cell) whichparalleled a diminishedpotency of the B3AR-selective agonists BRL37344 and CGP12177 to stimulate the production of intracellular CAMP.Analysis of RNA by polymerase chain reaction and nuclear run-on assays indicated that dexamethasone inhibited the synthesis of 83AR mRNA, resulting in 4-8-fold decrease in the steady-state levels of this mRNA. The down-regulation of 83AR protein andcellular mRNA appeared to be mediated by the receptor for glucocorticoids as assessed by the antagonistic action of the anti-glucocorticoidRU38486.

marked “uduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ethylamino]propyl)phenoxyacetate sesquihydrate (RR,SS distereThe nucleotide sequence(s1 reported in this paper has been submitted oisomer); (-)-[3H]CGP12177, (-)-4-(3-t-butylamino-2-hydroxyprototheGenBankTM/EMBLDataBankwith accession number(s) po~y)[5,7-~H]benzimidazole-2-one; (-)-[‘251]CYP, (-)-[‘251]iodo~yaX60438 and HSARB3A. nopindolo1; ICI118551, erythro-(+)-l-(7-methylindan-4-yloxy)-3-iso7 To whom correspondence should be addressed: Centre National propylaminobutan-2-01;IPR, (-)-isoproterenol; kb, kilobase(s); PCR, de la Recherche Scientifique UPR 0415, Laboratoire d’Immuno- polymerase chain reaction; SDS, sodium dodecyl sulfate; RU38486, Pharmacologie, MolBculaire, 22 rue Mbchain, 75014 Paris, France. 11@-(4-dimethylarninophenyl)- 170 - hydroxy - 17a-(propyl-l-ynyl)Tel.: 19-331-40-51-64-09;Fax: 19-331-40-51-72-10. estra-4,9-dien-3-one; Hepes, 4-(2-hydroxyethyl)-l-piperazineethane’ The abbreviations used are: OAR, @-adrenergicreceptor; bp, base sulfonic acid; GRE, glucocorticoid response elements; AP-1, activator pair(s); BRL37344, sodium 4-{2-[2-hydroxy-2-(3-chlorophenyl)- protein-1.

15909

15910

GlucocorticoidsAdipocyte Inhibit P3-Adrenergic

Receptors

DNA contamination. For PCR analysis, a sense (5"GCTCCGTysis activation (FBve et al., 1991). 83ARs exhibit a peculiar pharmacological profile character- GGCCTCACAG-3') andan anti-sense (5'-CTCGGCATCTGCCCCTA-3') primer were synthesized according to the sequence coding ized by a high efficiency of the lipolytic agonist BRL37344 for amino acid residues 2-7 and 177-182, respectively, of the murine and by a low affinity for classical &adrenergic antagonists, B3AR (Nahmias et d., 1991). For standardization purposes, a sense including the radioligands (-)-['251]CYP and (-)-[3H] (5'-GAGACCTTCAACACCCC-3') and anti-sense an (5'CGP12177.More strikingly,CGP12177which is a 81 and GTGGTGGTGAAGCTGTAGCC-3') primer were synthesized cor82AR antagonist, mediates an agonistic response upon bind- responding to nucleotides 453-469 and 669-688, respectively, of the sequence of the murine p-actin cDNA (Alonso et al., 1986). ing to 83ARs. First strandcDNA synthesis was performed on 5 pg of total RNA In the present work, we have used these pharmacological using an oligo(dT) primer and avian myeloblastosis virus reverse features to investigate the regulationbyglucocorticoidsof transcriptase (Stratagene). To verify that the amplified fragments 83AR expression in 3T3-F442A adipocytes. Based on quandid not proceed from residual DNA a control without reverse transcriptase was run in parallel for each sample. Following phenol and titative pharmacologicalexperimentsand on analysesof mRNA synthesis and intracellular levels, weconclude that chloroform extractions and ethanol precipitation, one fourth of the cDNA wassubjected to 30 cyclesof amplification (89 "C, 1min; 55 "C, dexamethasone,a potent synthetic glucocorticoidagonist, 1.5 min; and 69 "C, 1.5 min) using 2.5 units of Thermphylus aquaticus down-regulates P3AR expressionthroughatranscriptional (Taq) polymerase (Cetus) in 100 pl of a buffer consisting of 67 mM mechanism involvingthe glucocorticoid receptor. Tris-HC1, pH 8.4, 6.7 mM MgCl,, 100 pg/ml gelatin, 6.7 p~ EDTA, MATERIALS AND METHODS

Cell Culture-3T3-F442A cells (Green and Kehinde, 1976) were grown and differentiated in Dulbecco'smodified Eagle's medium supplementedwith 10% fetal calf serum (Gibco Laboratories) at 37 "C in an atmosphere of air CO, (9010, v:v). Upon reaching confluence, cells were continuously exposed to insulin (1 pg/ml). After acquisition of a mature adipocytic phenotype at day 8 after confluence, steroids (dexamethasone, RU38486, or progesterone) were added for the time indicated in figures and tables. Dexamethasone and progesterone were purchased from Sigma. RU38486 was a gift from Roussel-Uclaf (Romainville, France). Membrane Preparation and Adenylate Cyclase Assay-crude cell membrane were prepared as previously described (FBve et al., 1990) and stored at -80 "C. Protein content was assayed using BSA as a standard (Lowry et al., 1951). Adenylate cyclase (EC 4.6.1.1) activity was measured in triplicate for 10 min at 35 "C in a 50-p1 standard assay consisting of0.2mM [cx-~'P]ATP (Amersham, P B 171), 1mM CAMP,10 mM phosphocreatine, 0.6 unit creatine phosphokinase, 100 p M GTP, 5 mM MgCl,, 0.2 mM EDTA, and 50 mM Tris-HC1 (pH 7.9, with or without 8adrenergic effectors. The reaction was initiated by the addition of crude membranes (15-25 pg of protein) and terminated as described (Pairault et al., 1982). Binding Experiments-(-)-[3H]CGP12177(Amersham, TRK 835) binding studies were performed directly on intact adherent 3T3F442A adipocytes (FBve et al., 1990).Non specific binding, determined in the presence of 100 p~ (-)-isoproterenol (IPR), was usually 10-15 and 25-30%of the total binding a t 1 and 40 nMof radioligand, respectively. Under these conditions, the internalization of the radioligand was negligible. For binding assays to subcellular fractions, membrane aliquots (50 pg of protein) were incubated for 20 min at 37 "C with (-)-['251]CYP (Amersham, IM142), with or without competing ligand in the presence of 10 mM MgCl,, 1 mM ascorbic acid, and 50 mM Tris-HC1, pH 7.4.100 p~ GTP was added to shift allreceptors into thelow affinity conformation for agonists (Kent et al., 1980). Separation of bound from free ligand was achieved by filtration overglass fiber disks (Whatman GF/C).Saturation experiments were performed with (-)-['z51]cYP concentrationsranging from 5 to 3000 pM. Competition experiments were carried out at 300 pM (-)-['251]CYP. Nonspecific binding was determined in the presence of 100 p M IPR and usually represents 25% of the total binding a t 300 pM (-)-['251]CYP. IPR, salbutamol, and fenoterol were purchased from Sigma. CGP12177 was a gift from Ciba-Geigy AG (Basel, Switzerland). BRL37344 was kindly provided by SmithKline Beecham Pharmaceuticals (Epsom, United Kingdom), and IC1118551 by IC1 Pharma (France Division, Cergy-Pontoise). Data Analysis-Data generated from saturation and competition binding experiments were analyzed by the EBDA and LIGAND programs (Biosoft-Elsevier, Cambridge, England) (Munson and Rodbard, 1980; McPherson, 1985) and are presented as mean & S.E. of a t least three independent experiments performed in duplicate or triplicate. The statistical significance of the differences between the groups was assessed with a paired Student's t test. RNA Analysis-Total RNA was extracted from 3T3-F442A adipocytes (Cathala et al., 1983) and digested for 1 h at 37 "C with 5 units of RNase-free DNase I (Boehringer) per pg of RNA to minimize

10 mM @-mercaptoethanol,16 mM (NH4),S04,10% (v/v) dimethyl sulfoxide, 5% (v/v) formamide, 125 p~ of each deoxynucleotide triphosphate, and 125 nMof each primer. Samples were electrophoresed through a 2% agarose gel and blotted onto nylon membranes which where then hybridized simultaneously to @BARand @actin-specific probes. The j33AR probe was obtained by amplification of 5 ng of DNA from a X clone of the murine 03AR gene between a sense (5'-GCTCCGTGGCCTCACGAGAA-3')and an anti-sense (5'-CCCAACGGCCAGTGGCCAGTCAGCG-3')primer corresponding to amino acids 2-8 and 98-106 (nucleotides 41-60 and 330-354), respectively, of the human j33AR sequence. The actin probe was a 1-kb PstI fragment from the PAM91 actin plasmid (Minty et al., 1983). After hybridization, blots were subjected to a series of 15min washes; three in 2 X SSC (1 X ssc: 150 mM NaCl, 15 mM sodium citrate), 0.1% SDS (w/v) at 22 "C, one in the same buffer a t 45 "C and three in 0.1 X SSC, 0.05% SDS a t 45'C. For quantitation purposes, autoradiograms of the blots were analyzed with a scanning densitometer and peak areas of P3AR signals were normalized relative to those of p-actin signals. For transcription analysis, nuclei were prepared from 3T3-F442A adipocytes as described (Antraset al., 1991) except for the final centrifugation which was in transcription buffer (50 mM Tris-HC1, pH 8.0,140 mM KCl, 10 mM MgCl,, 0.5 mM EDTA, 1mM dithiothreitol, 0.1 mM phenylmethylsulfonyl fluoride) containing 40% glycerol. Nuclei isolated from about lo' cells wereused immediately after resuspension in a total volume of200 p1 of transcription buffer supplemented with final concentrations of 1 mM each ATP, CTP, and GTP,10 mM phosphocreatine, 10 pg/ml creatine phosphokinase, 1 mg/ml heparin sulfate, 4 mM MnCl,, 1500 units/ml placenta ribonuclease inhibitor, and 250 pCi of [ c x - ~ ~ P I U(400 T P Ci/mmol). Elongation of nascent transcripts was for 30 min at 30 "C. Following addition of NaCl to 250 mM, CaC12to 2 mM, and 300 units of RNasefree DNase I, the mixture was incubated for 10 min a t 37 "C and further incubated 30 min a t 37 "C after addition of EDTA to 25 mM, SDS to 1%(w/v), yeast transfer RNA to 50 pg/ml, and proteinase K to 100 pg/ml. The reaction was diluted to 400 pl with a solution of 250 mM NaC1, 1%SDS, and 20 mM EDTA and extracted three times by the hot phenol procedure. After ethanol precipitation, RNA was partially hydrolyzed by a treatment of 10 min on ice with 0.2 N NaOH. Following neutralization by addition of HC1to 0.2 N and Hepes buffer to 50 mM, RNA was precipitated twice with absolute ethanol. Quantitation of synthesized RNA was done by hybridizing (Antras et al., 1991) equal amounts (-10' cpm) of the purified RNA to nitrocellulose filters spotted with 5 pg ofthe PUC 18vector containing the p-actin cDNA (Spiegelman et al., 1983) or a 2-kb BarnHI-BglII fragment from the mouse p3AR gene (Nahmias et al., 1991). Filters were washed two times for 20 min in 2 X SSPE (1 X S S P E 0.18 M NaCl, 1mM EDTA, 10 mM sodium phsphate, pH7.7) a t 22 'c, treated for 30 rnin at 37 "C in the same buffer containing 10 pg/ml RNase A, rinsed in 0.1 X SSPE, 0.1% SDS, and submitted to three additional washes of 20 min in 0.1 X SSPE, 0.1% SDS at 55 "C.Autoradiograms of the filters were scanned densitometrically and 83AR signals normalized to those of actin. RESULTS

Glucocorticoids Down-regulate83AR-mediated Adenylate Cycluse Activity-Regulation of 83AR expression was studied

in 3T3-F442A adipocytes grown for 7 days after confluence

GlucocorticoidsAdipocyte Inhibit P3-Adrenergic Receptors and subsequently treated for various periods of time with the glucocorticoid analogue dexamethasone. Due to the presence of insulinin the culture medium, dexamethasone did not affect the differentiation stage of adipocytes (Moustaid et al., 1990). CGP12177 was used to specifically address the issue of 03AR coupling. It is a /3l and P2AR antagonist (Staehelin et al., 1983) which displays a P3AR agonistic activity as demonstrated by its effects in Chinese hamster ovary cells transfected with the human (F2.veet al.,1991) or murine (Nahmias et al., 1991) @BARgene. CGP12177-stimulated adenylate cyclase activity remained constant in control adipocytes but rapidly decreased in a time-dependent manner when cells were exposed to dexamethasone (Fig. 1). The modulation by glucocorticoids of P3AR expression was investigated further by comparing, in adipocytes treated by dexamethasone, the stimulation of adenylate cyclase activity induced byP3 (BRL37344, CGP12177)-selective or byP2 (fenoterol, salbutamo1)-selective agonists (Fig. 2). Exposure of the cells to dexamethasone resulted in a marked decrease of the CAMPproduction induced by any agonist. However, it strongly affected CGP12177- and BRL37344-stimulated adenylate cyclase activity but had a weaker effect on salbutamol and fenoterol responsiveness (Fig. 2 A ) . This differential action is more clearly visualized when expressing the results as the percentage of the maximal stimulation induced by an optimal dose of the nonselective agonist IPR. It thenappears (Fig. 2B) that the relative effect of P3AR-selective agonists was decreased, while that of P2AR-selective agonists was increased, in dexamethasone-treated versus control cells. Dose-response studies to these various agonists were performed to further analyze the modulation of PAR sensitivity by glucocorticoids. In control adipocytes, the apparentKactof BRL37344 and CGP12177 was much lower thanthat of salbutamol and fenoterol (Table I). In dexamethasone-treated cells, the K,,, values for the p2-selective agonists were largely reduced while that for BRL37344 was increased, and CGP12177 had nodetectable effects (Table I). Taken together, these results suggest that the predominant P3AR population of 3T3-F442A adipocytes is replaced by a major B2AR component upon exposure to glucocorticoids. Dexamethasone-induced Decrease in P3AR Sensitivity Is a Glucocorticoid Receptor-mediated Mechanism-The molecular mechanism of the dexamethasone-induced reduction in catecholamine responsiveness was analyzed by measuring IPR-, BRL37344-, and CGP12177-induced adenylate cyclase activity in adipocytes exposed to various concentrations of dexamethasone (Fig. 3). The maximum effect of the drug was

15911

IPR

Salbuiarnol Fenoteral

BRL37344 CGP12177

FIG. 2. Dexamethasone preferentially decreases adenylate cyclase activity stimulated by p3-adrenergic agonists. At day 7 afterconfluence adipocyte cultures were treated (closed bars) or not (open bars) with dexamethasone (250 nM) for 8 days. At day 15, membranes were prepared and adenylate cyclase activity stimulated by an optimal concentration (100 p ~of)each of the indicated agonist was determined. Results are expressed as agonist-stimulated over basal adenylate cyclase activity ( A ) or as the percentage of the IPR effect obtained for each selective agonist ( B ) .*, p < 0.05; **, p < 0.01; ***, p < 0.0001; ns, nonsignificant; dexamethasone-treated versus control adipocytes.

TABLEI Relative potency of various effectors for stimulating adenylate cyclase activity in control and dexamethasone-treated adipocytes Adipocytes were processed as described in legend to Fig. 2. Apparent activation constants (Kact)are the concentrations of each ligand required for a half-maximal stimulation of adenylate cyclase activity. Basal adenylate cyclase activity was 11.2 f 1.5 and 4.8 f 0.4 pmoles of cAMP/min/mg of protein in control and dexamethasone-treated adipocytes, respectively. No CGP12177-mediated adenylate cyclase activity could be detected in dexamethasone-treated adipocytes. I.

.

ugana

K4Ct adipocytes

Dexamethasone-treated adiDocvtes

nM

IPR BRL 37344 CGP 12177 Salbutamol Fenoterol

2,075 f 450 162 f 4 480 f 30 27,250 f 5,250 12,350 f 1,950

525 f 35" 775 f 105b

ND 2,325 f 875" 510 f 190"

" p< 0.05. b p < 0.01, dexamethasone-treated versus control cells.

ND, not detectable.

observed above 10 nM, and the concentration required for a half-maximal effect on IPR, BRL37344, and CGP12177 responses was between 1 and 2 nM, consistent with the known KO of the glucocorticoid receptor for dexamethasone in 3T3 I 10 12 14 16 adipocytes (Hainque et al., 1987; Nakada et al., 1987). Day of culture after confluence Moreover, the effect of dexamethasone on IPR-, FIG. 1. Down-regulation by dexamethasone of CGP12177- BRL37344-, and CGP12177-stimulated adenylate cyclase acstimulated adenylate cyclase activity in 3T3-F442A adipo- tivity was reversed by a 10-fold molar excess of the glucocorcytes. Dexamethasone (250 nM) was added to adipocyte cultures for ticoid antagonist RU38486(Fig. 4). Since RU38486is also 8 days starting from day 7 following confluence. CGP12177 (100 p ~ ) stimulated adenylate cyclase activity was measured in control (0) known for its antiprogestin activity (Phillibert, 1984), we and in dexamethasone-treated (0)adipocytes. Results represent ag- verified that a high concentration (250 nM) of progesterone onist-stimulated over basal adenylate cyclase activity. *, p < 0.02; **, had no significant effect on the PAR agonist-induced activip < 0.01; ***, p < 0.0001;dexamethasone-exposed versus control cells. ties. Thus, thisdown-regulation of P3AR coupling is likely to

15912

GlucocorticoidsAdipocyte Inhibit

D3-Adrenergic Receptors 300

A

Log[Dexamethasone]

3

(M)

(-)-[3H]CGP12177 (nM)

FIG. 3. Dose-dependent effect of dexamethasone on IPR-, BRL37344-, andCGP12177-stimulatedadenylate cyclase activity in 3T3-F442Aadipocytes. Mature adipocyteswere treated from day 7 to 15 after confluence with various concentrations (0-250 nM) of dexamethasone. Membrane adenylate cyclase activity induced by 100 p~ of each IPR (01,BRL37344 (O),and CGP12177 (0)was then measured. Results are from a typical experiment performed in triplicate.

Bl

Bound (frnoli106 cells)

FIG. 5. Characterization of (-)-[sH]CGP12177 binding sites in control and dexamethasone-treated adipocytes. Saturation curves( A )and Scatchard (1949) representation ( B )of a typical

experimentperformed on control (0) and dexamethasone-treated adipocytes (0).In the Scatchard representationdata were best-fitted for a two-site model in control adipocytes (dotted lines) and for a one-site model in dexamethasone-exposed cells. represented 5% of that expressed in control cells (5,500 f 6,500 sites/cell; p < 0.01, dexamethasone-treated uersm cona a trol cells). FIG. 4. Effect of dexamethasone, progesterone, and Saturation binding experiments were also carried out on RU38486 on BAR-stimulated adenylate cyclase activity in crude membrane fractionsusing the hydrophobic radioligand day 3T3-F442A adipocytes. Mature adipocytes were exposed from 7 to 15 after confluence either to 10 nM dexamethasone (Dex) or to (-)-['251]CYP. Here again, control adipocytesdisplayed a 100 nM RU38486 ( R U ) or to a mixture of dexamethasoneand major (90%) and a minor (10%) population of low and high RU38486 (Dez + RU), at the above concentrations, or to 250 nM affinity binding sites, respectively (Table 11). In dexamethaprogesterone. Membrane adenylate cyclase activity induced by 100 sone-treated cells only high affinity sites could be characterp~ of each IPR, BRL37344, and CGP12177 was then measured. Results are expressed as the increment of agonist-stimulated aden- ized. The KD values of the low affinitysites for(-)-[3H] ylate cyclase activity over basal. *, p < 0.05; **, p < 0.01;***, p < CGP12177 and for (-)-['251]CYP are in agreement with those of the PBAR for these ligands (Nahmias et al., 1991; FBve et 0.001; drug-treated versus control adipocytes. al., 1991; Emorine et al., 1989). The K D values for the high affinitysites expressed incontrolanddexamethasone-exbe mediated by the glucocorticoid receptor. posed adipocytescorrespond to that of classical P l and @2ARs, Dexamethasone Down-regulates 03AR Protein Expressionet al., 1990). respectively (Feve T o investigate whether the variations in agonist-stimulated Competition curves of P2- and B3AR-selective ligands adenylate cyclase activityarosefrommodulation of PAR also connumber, we measured BAR subtype densities on intact, con- against (-)-['251]CYP binding(datanotshown) a major P3AR firmed the glucocorticoid-induced switch from trol, or dexamethasone-treated adipocytes as well as in the corresponding membrane fractions. TheB3AR displays a low population toward a major p2AR component. In control adiaffinity for both(-)-[3H]CGP12177 and (-)-['251]CYP (Nah- pocytes, the B3AR-selective agonist BRL37344 was more efmias et al., 1991; Feve et al., 1991; Emorine et al., 1989) and ficient (ICso= 220 k 150 nM) than the P2AR-selective antagcan thus be distinguished from the classical P l and P2AR onist IC1118551 (IC50= 1840 f 250 nM) to displace binding of (-)-['251]CYP (300 p ~ )The . reverse situation was observed subtypes which have a high affinity for thesecompounds. Saturationexperiments were performedonintact cells in dexamethasone-treated cells (IC50= 5,200 f 1,500 nM and using a wide range of concentrations (0.1-150 nM)of the 43 k 32 nM for BRL37344 and ICI118551, respectively). Dexamethasone Down-regulates P3AR mRNA Expressionhydrophilic radioligand (-)-[3H]CGP12177 (Fig. 5). Control Thedexamethasone-induced decrease of adipocyte P3ARs adipocytes possessed a n heterogeneous PAR population resolved in a Scatchard (1949) plot into a major B3AR compo- revealedby our pharmacological studies was furtherconnent of low affinity for (-)-[3H]CGP12177 (113,800 +- 31,300 firmed by PCR analysisof mRNA steady-state levels (Fig. 6). sites/cell; KO = 23 f 5 nM) and a minor P l plus p2AR The B3AR- and P-actin-specific sequences amplified in this component of high affinity for the ligand (6,400 f 1,600 sites/ assaymainly proceeded fromthecorrespondingmRNAs, cell; KD = 0.6 f 0.1 nM). In contrast, dexamethasone-treated since only limited amplificationwas obtained when the samof reverse transcriptase. adipocytes exhibited a homogeneous high affinity BAR com- ples were not submitted to the action ponent (4950 f 250 sites/cell; KO = 0.6 f 0.1 nM). A low Such amplification could result from residual contaminating affinity component was occasionally resolved and, at most, DNA or from a low reverse transcriptase activity of the Taq

GlucocorticoidsAdipocyte Inhibit P3-Adrenergic Receptors

15913

TABLE I1 (-)-[12'I]CYP binding sites in membranes from control and dexamethasone-treated adipocytes Saturation experimentswere carried out on membranes from control and dexamethasone-exposed cells using (-)-["'I]CYP concentrations ranging from 5 to 3000 PM. Data were analyzed by the method of Scatchard (1949) with the EBDA/LIGAND program. Membrane protein content was used to normalize the density of binding sites and was 1.11 and 0.86 mg of protein/lO" cells in control and dexamethasonetreated adipocytes, respectively. (-)-['*sI]CYP binding sites High affinity

Control adipocytes Dexamethasone-treated adipocytes

07

09

615 492

369 246 615 492

Low affinity (BBAR)

KO f p M )

siteslcell

K D (PM)

43.2 f 9.4 45.3 f 8.9

7,850 f 1,000 4,600 2 2,350

1,92062,250 k 220

Dl6

013

T' 0 *Dex':o + - * - -

(81and 82AR)

Dex " 0

Dex

- + - * - +

'

' -

--

-WAR

siteslcell

016 +O

--

+px

DyRu

+Ru

f 7,200 None

'

"-

-

-. -

369

-

246

-4

.

".

L

*

-" .~. -

.

. )

a

-W A R

..""

m

*

FIG. 6. Down-regulation by glucocorticoids of p3AR mRNA expression in 3T3-F442A adipocytes. Left panels, 3T3-F442A 13 (013), and 16 adipocytes were exposed ( D e r ) or not (0)to 250 nM dexamethasone from day 7 ( 0 7 ) after confluence until days 9 (D9), ( D 2 6 ) .Cellular RNA prepared a t each time point was treated (+) or not (-) with reverse transcriptase to verify that the subsequent PCR amplification, performed by using a mixture of 83AR and 8-actin primers, did not proceed from contaminating DNA. The resulting products resolved on a 2% agarose gel were visualized by ethidium bromide staining (upper panel)and by hybridization to specific 83AR and 8-actin probes after blotting to nylon membranes (lower panel). Right panels,from day 7 to 16 of differentiation, adipocytes were exposed either to 10 nM dexamethasone or to 100 nM RU38486 ( R U )or toa combination of both (Dex + RU),and experiments were performed as above. The sizes (in base pairs) of molecular weight markers (123 DNA bp ladder from GIBCO-BRL) are indicated in the left margin, and the positions of the 83AR- and 8-actin-amplified fragments (expected a t 543 and 236 bp, respectively) are shown between the panels.

polymerase (Jones and Foulkes, 1989). Mature 3T3-F442A adipocytes maintained in culture from day 7 after confluence until day 16 expressed stable levels of P3AR mRNA. In contrast, exposure of the cells to 250 nM dexamethasone rapidly induceda 4-8-fold decline in thelevels of this mRNA species, although it had no action on &actin mRNA levels (Fig. 6, left panels). Thedown-regulation of the P3AR message induced by dexamethasone was likely mediated by the glucocorticoid receptor, since it was antagonized by RU38346 which had no effect by itself (Fig. 6, right panels). Since glucocorticoid receptors primarily act as transcription factors, we investigated the effects of dexamethasone onP3AR mRNA synthesis. Nuclear run-on assays showed that nuclei isolated from 3T3-F442A adipocytes treated for 30 min with 250 nM dexamethasone displayed an efficacy for in vitro elongation ofP3AR mRNAthat wasreduced by 50% as compared to thatof untreated cells. A longertreatment almost totally inhibited (80-90%) P3AR gene transcription (Fig. 7). DISCUSSION

The down-regulation of P-adrenergicresponsiveness induced by glucocorticoid in 3T3-F442A adipocytes was investigated. We show that the sensitivity to agonists and the membrane densityof the P3AR subtype, which is the predominant PAR of 3T3-F442Aadipocytes, isdramaticallydecreased upon exposure of the cells to dexamethasone. A parallel diminution of P3AR mRNA steady-state levels is observed which appears to be mediated by a transcriptional action of the glucocorticoid receptor. Dexamethasone Induces Down-regulation of P3AR Protein

4

Actin PUC 03AR

I

!

0 0.5 1

24

Time (hr) FIG.7. Effect of dexamethasone on the transcription rate of the B3AR gene in 3T3-F442A adipocytes. 3T3-F442A adipocytes were treated for the indicated times with 250 nM dexamethasone. Nuclei and RNA probes were processed as described under "Experimental Procedures." Results are representative of two independent experiments. and mRNA Levels-Exposure of 3T3 adipocytes to dexamethasone haspreviously been shownto increase expressionof the P2AR subtype while reducing that of the PlAR subtype (Lai et al., 1982; Nakada et al., 1987; FBve et al., 1990). Such a treatment only slightly decreasedthe total numberof P l plus P2ARs (see Scatchard representation in Fig. 5) but strongly depressed the production of CAMP induced by stimulation with P-adrenergic agonists (FBve et al., 1990). These analyses were pursued further by studying theeffects of glucocorticoids on P3AR coupling to adenylate cyclase and on its expression in 3T3-F442A adipocytes. In comparison with control adipocytes, the adenylate cyclase system of dexamethasone-treated cells displays a markedly decreased sensitivity to stimulation by P3AR-selective agonists. This correlateswith saturation binding experiments

15914

GlucocorticoidsAdipocyte Inhibit P3-Adrenergic Receptors

showing that adipocytes express a major population of sites of low affinity for either (-)-[3H]CGP12177 or (-)-[1251]CYP which is sharply decreased in cells exposed to glucocorticoids. Our previous studies identified these low affinity sites as belonging tothe P3AR subtype (FBve et al., 1991). PCR analysis of P3AR mRNA indeed indicates a paralleldecrease of steady-state levels of this message indexamethasonetreated uersus control adipocytes. Together, ourpharmacological experiments and RNA analysis converge to suggest that glucocorticoids down-regulate P-adrenergic responsiveness of 3T3-F442A adenylate cyclase by depressing cellular P3AR synthesis. The overall picture emerging from this and our previous work (Five et al., 1990, 1991) indicates that 3T3-F442A adipocytes express a heterogeneous PAR population with a major @3AR,a minor PlAR, and an accessory P2AR component. Exposure of either differentiating cells or mature adipocytes to glucocorticoids almosttotally depresses P l and P3ARs expression, whereas it stimulates that of the P2AR subtype which becomes predominant. Concomitantly, the total number ofBARS decreases 10-20-fold. In 3T3-F442A, this dramatic diminution of the major P3AR component may thus be the primary determinant of the down-regulation by glucocorticoids of P-adrenergic agonist-induced adenylate cyclase activity. Mechanisms for the Inhibition of P3AR Expression by Glucocorticoids-The dose-response curves of the dexamethasone effects on P3AR sensitivity of 3T3-F442A adipocytes, the antagonism of RU38486 toward this effect, and the lack of action of progesterone support the hypothesis that the glucocorticoid receptor is the major mediator of dexamethasone inhibition of P3AR expression. This assumption is strengthened by the parallel effects of dexamethasone and RU38486 on P3AR number and mRNA levels, in agreement with the well-known action of glucocorticoid receptors as transcriptional regulators, and is reinforced by the observation that dexamethasone inhibits nuclear synthesis of P3AR mRNA. Hormone-bound glucocorticoid receptorsenhance or decrease the rate of gene transcription by binding to short promoter DNA sequences termed glucocorticoid response elements(GRE). Enhancing GRE sequences (GRE(+)) have been found in the promoter region of the human and mouse P2AR genes (Emorine et al., 1987; Kobilka et al., 1987; Nakada et al., 1989), and one or more of these are likely to explain the effects of glucocorticoids on P2AR expression (Malbon and Hadcock, 1988). In contrast, no perfect matches with the consensus sequence proposed for negative GRE (GRE(-)) (Beato, 1989) have been found in the as-yet determined 5’flanking sequences (about 650 bp) of the mouse or human 83AR genes. These regions, however, must contain aproximal promoter as mRNA initiation sites have been mapped 50-80 nucleotides 3’ from recognition sites for CCAAT box binding factors.’ It is nevertheless possible that GRE(-)s exist in the P3AR gene but reside further upstream from the proximal promoter. Negative responses to glucocorticoids also occur when the glucocorticoid receptor interacts with other transcriptional regulators. This may (Sakai et al., 1988; Drouin et al., 1989; Diamond et al., 1990) or may not (Yang-Yen et al., 1990; Schule et al., 1990) require direct receptor-DNA contact. Thus, so-called composite GREs which contain no GRE(-) motifs but display sequences related to thebinding site of the activator protein-1 (AP-1) have been characterized (Diamond et al., 1990; Yang-Yen et al., 1990; Schule et al., 1990). In this

* B. FBve, B. Baude, S. Krief, A. D. Strosberg, J. Pairault, and L. J. Emorine, unpublished results.

respect, it is worth notingthat fat-specific elements, described in the promoters of several genes that participatein the adipose differentiation of 3T3-F442A cells (Hunt et al., 1986; Phillips et al., 1986), also contain AP-1 binding sites (Distel et al., 1987; Rausher et al., 1988; Ross et al., 1990). Thus, the presence in the proximal promoter regions of the murine and human P3AR genes of sequences related to fat-specific elebinding sites suggest ments (Emorine et al., 1991) and to AP-1 that, in 3T3-F442A adipocytes, the rapid and potent repression of P3ARgene transcription could result from the negative interaction of the glucocorticoid receptor with AP-1 and/or other adipose tissue-specific factors. Physiological Significance of the Glucocorticoid-induced P3AR Down-regulation in AdiposeTissue-Adrenal corticoids are known for their role in facilitating CAMP-mediated lipolysis in white adipose tissue. However, such permissive effects are often accompanied by diminished PAR-stimulated cAMP production (Exton et al., 1972; Braun and Hechter, 1970). At the same time, glucocorticoids suppress brown adipose tissue thermogenesis and sympathetic activity (Vander Tuig et al., 1984; York et al., 1985). These actions of corticoids may now be analyzed in view of the discovery of P3ARs in such tissues (Emorine et al., 1989; Nahmias et al., 1991). Brown fat thermogenesis is critically dependenton the uncoupling protein thermogenin and on lipolysis-liberated free fatty acids. It is thus likely that glucocorticoids suppress brown adipose tissue activity by down-regulating P3AR (and/ or PlAR) which has been suggested to control both thermogenin (Rehnmark et al., 1990) and lipolysis (Arch, 1989). Alternatively, glucocorticoids could suppress BARS involved at otherlevels of the afferent sympathetic pathway to brown adipose tissue. Both mechanisms are consistent with a glucocorticoid-induced inhibition of the sympatheticsystemstimulated brown adipose tissue activity (Vander Tuig et al., 1984; York et al., 1985), and may explain fat accretion in this tissue upon cortisone treatment (Skala and Hahn, 1971). In white adipose tissue, corticoid-induced diminution of 6adrenergic agonist-stimulated cAMP production might reflect down-regulation of P3ARs and/or PlARs numbers as is the case in 3T3-F442A adipocytes. This is not contradictory with facilitating effects of glucocorticoids on lipolysis since concomitant induction ofP2ARs maintainsa catecholaminesensitive adenylate cyclase activity producing limited cAMP levels sufficient for maximal lipolysis. Thus, although permissive effects of corticoids have been suggested to partly proceed from increased PAR levels or coupling (Malbon et al., 1988), our results suggest that this might not be true for lipolysis. Other causes for glucocorticoid-enhanced lipolysis could reside in the activation and/or inductionof proteins modulating the activity of hormone-sensitive lipase (e.g.kinases) as is the case in liver for enzymes of neoglucogenesis and glycogenolysis (reviewed by Exton (1987)). The neteffect of glucocorticoids on fatty acid mobilization would thus be to modulate the balance between a sympathetic control mediated by norepinephrine-activated P3AR and/or PlAR, and a humoral control mediated by epinephrine-activated P2ARs. Thus, P3ARs could possibly be involved in acute response to situations such as physical exercise or stress where food is not available but increased energy is required. The role of corticosteroids would be to attenuate thissympathetic energy recruitment which, if prolonged, could deplete energy stores. REFERENCES Alonso, S., Minty, A,, Bourlet, Y., and Buckingham, M. (1986)J. Mol. Euol. 23,ll-22 Antras, J., Lasnier, F., and Pairault, J. (1991)J. Bid. Chern. 266, 1157-1161 Arch, J. R. S. (1989)Proc. Nutr. SOC.48,215-223

GlucocorticoidsAdipocyte Inhibit P3-Adrenergic Receptors Beato, M. (1989)Cell 56,335-344 Braun, T.,and Hechter, 0.(1970)Proc. Natl. Acad. Sci. U. S. A . 66,995-1001 Cathala, C., Savouret, J. F., Mendez, B., Karin, M., Martial, J. A,, and Baxter, J. D. (1983)D N A 2,329-335 Challis, R. A. J., Leighton, S., Wilson, S., Thurlby, P. L., and Arch, J. R. S. (1988)Biochem. Pharmacol. 37,947-950 Diamond, M. I., Miner, J. N., Yoshinaga, S. K., and Yamamoto, K. R. (1990) Science 249, 1266-1272 Distel, R. J., R o , H.-S., Rosen, B. S., Groves, D.L., and Spiegelman, B. M. (1987)CeU 49,835-844 Drouin, J., Trifiro, M. A., Plante, R. K., Nemer, M., Eriksson, P., and Wrange, 0.(1989)Mol. Cell. Bwl. 9,5305-5314 Emorine, L. J., Marullo, S., Delavier-Klutchko. C., Kaveri, S. V., DurieuTrautman, O., and Strosberg, A. D. (1987)Proc. Natl. Acad. Sci. U. S. A . 84, 6995-6999 Emorine, L. J., Marullo, S., Briend-Sutren, M.-M., Patey, G., T a b , K.,DelavierKlutchko, C., and Strosberg, A. D. (1989)Science 246,1118-1121 Emorine, L. J.,Five, B., Pairault,J.,Briend-Sutren, M.-M., Marullo, S., Delavier-Klutchko, C., and Strosberg, A. D. (1991)Biochem. P h a r m o l . 41, 853-859 Exton, J. H. (1987)Diobetes Metabol. Rev. 3,163-183 Exton, J. H., Friedmann, N., Wong, E. H.-A,, Brineaux, J. P., Corbin, J. D., and Park, C. R. (1972)J. Biol. Chem. 247,3579-3588 Five, B., Emorine, L. J., Briend-Sutren, M. M., Lasnier, F., Strosberg, A. D., and Pairault, J. (1990)J. Biol. Chem. 266, 16343-16349 Five, B., Emorine, L. J., Lasnier, F., Blin, N., Baude, B., Nahmias, C., Strosberg, A. D., and Pairault, J. (1991)J. Biol. Chem. 266,20329-20336 Granneman, J. G., Lahners, K. N., and Chaudry, A. (1991)Mol. Phurmacol. 40,895-899 Green, H., and Kehinde, 0.(1974)Cell 1, 113-116 Green, H., and Kehinde, 0.(1976)Cell 7, 105-113 Guest, S. J., Hadcock, J. R., Watkins, D. C., and Malbon, C. C. (1990)J. Biol. Chem. 266.5370-5375 ",. . .. Hainque, B., ,Moustaid, N., Quignard-Boulang6, A., and Lavau, M. (1987) Bmhcm. Bwphys. Acta 931, 347-353 Hollenga, C., Haas, M., Deinum, J. T., and Zaagsma, J. (1990)Horm. Metabol. Res. 22, 17-21 Hunt, C. R., Ro, H.-S., Dobson, D., Min, H. Y., and Spiegelman, B. M. (1986) Proc. Natl. Acad. Sci. U. S. A. 83,3786-3790 Jones, M. D., and Foulkes, N. S. (1989)Nucleic Acids Res. 17,8387-8388 Kent, R. S., De Lean, A., and Lefkowitz, R. J . (1980)Mol. Pharmacol. 17,14~

~

~

33

Kibilka, B. K., Frielle, T., Dohlman, H. G., Bolanowski, M. A,, Dixon, R. A. F., Keller, P., Caron, M. G., and Leikowitz, R. J. (1987)J. Biol. Chem. 262, 7321-7327 Lai, E., Rosen, 0. M., and Rubin, C. S. (1982)J. Biol. Chem. 287,6691-6696 Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951)J. Biol. Chem. 193,265-275

15915

Malbon, C.C., and Hadcock, J. R. (1988)Biochem. Biophys. Res. Comrnun. 164,676-681 Malbon, C. C., Rapiejko, P. J., and Watkins, D. C. (1988)Trends Pharmacol. Sei. 9,33-36 McPherson, G. A. (1985)J. Pharmacol. Methods 14,213-228 Minty, A. J., Alonso, S., Guinet, J.-L., and Buckingham, M. (1983)J. Mol. Biol. 167,77-101 Moustaid, N., Lasnier, F., Hainque, B., Quignard-Boulange, A., and Pairault, J. (1990)J. Cell. Biochem. 42,243-254 Munson, P. J., and Rodbard, D.(1980)Anal. Biochem. 107,220-239 Muzzin, P., Seydoux, J., Giacobino, J.-P., Venter, J.-C., and Fraser, C. (1988) Biochem. Biophys. Res. Commun. 166,375-382 Muzzin, P., Revelli, J.-P., Khune, F., Gocayne, J., McCombie, W. R., Venter, J. C., Giacobino, J.-P., and Fraser, C. (1991)J. Biol. Chem. 266, 24053')AnF(R

N & & I , C., Blin, N., Elalouf, J.-M., Mattei, M.-G., Strosberg, A. D., and Emorlne L. J. (1991)EMBO J. 10,3721-3727 Nakada, hi. T., Stadel, J. M., Poksay, K. S., and Crooke, S. T. (1987)Mol. Phrmacol. 31,377-384 Nakada, M. T., Haskell, K. M., Ecker, D. J., Stadel, J. M., and Crooke, S. T. (1989)Biochem. J. 260, 53-59 Pairault, J., Lasnier, F., and Laudat, M. H. (1982)Eur. J. Biochem. 127,351358 Phil&, M., Djian, P., and Green, H. (1986)J. Biol. Chem. 261,10821-10827 Phillibert, D. (1984)in Adrenal Steroid Antagonism (Argawal, K. W., ed) pp. 77-101,Walter de Gmyter, Berlln Rauscher, F. J., 111, Sambucetti, L. C.,Curran, T., Distel, R. J., andSpiegelman, B. M. (1988)Cell 62,471-480 Rehnmark, S., Nbhad, M., Herron, D., Cannon, B., and Nedergaard, J. (1990) J. Bwl. Chem. 266,16464-16471 Ross, S. R., Graves, R. A., Greenstein, A., Platt, K. A., Shyu, H.-L., Mellovitz, B., and Spiegelman, B. M. (1990)Proc. Natl. Acad. Sei. U. 5. A . 87,95909594 Sakai, D. D Helms, S., Carlstedt-Duke J Gustafsson, J.-A., Rottman, F. M., and Yamimoto K. R. (1988)Genes &'&L 2, 1144-1154 Scatchard, G. (19k9)Ann. N. Y. Acad. Sci. 51,660-672 Schule, R., Rangarajan, P., Kliewer, S., Ransone, L. J., Bolado, J., Yang, N., Verma, I. M., and Evans, R. M. (1990)Cell 62,1217-1226 Skala, J and Hahn, P. (1971)Can. J. Physiol. Pharmacol. 49,501-505 Spiegelman, B. M., Frank, M., and Green, H. (1983)J. Biol. Chem. 268,100831nnm St&h&, M., Simmons, P., Jaeggi, K., and Wigger, N. (1983)J. Biol. Chem. 268,3496-3502 Vander Tuig J. G., Ohshima, K., Yoshida, T., and Bray, G. A. (1984)Life Sci. 34,1423-i432 Yang-Yen, H.-F., Chambard, J.-C., Sun, Y. L., Smeal, T., Schmidt, T. J., Drouin, J., and Karin, M. (1990)Cell 62,1205-1215 York, D. A,, Marchington, D., Holt, S. J., and Allars, J. (1985)Am. J. Physiol. 249, E299-E305 Zaagsma, J., and Nahorski, S. R. (1990)Trends Phurmacol. Sci. 11,3-7