of September 15, 2018.This information is current as -Independent Process

γProliferator-Activated Receptor Macrophages through a PeroxisomeGlucocorticoid Binding and Signaling in

Inhibits2-Prostaglandin J12,14∆15-Deoxy-

Bruno Fouqueray and Laurent BaudAdeline Cheron, Julie Peltier, Joëlle Perez, Agnès Bellocq,

http://www.jimmunol.org/content/172/12/7677doi: 10.4049/jimmunol.172.12.7677

2004; 172:7677-7683; ;J Immunol 

Referenceshttp://www.jimmunol.org/content/172/12/7677.full#ref-list-1

, 20 of which you can access for free at: cites 35 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists All rights reserved.Copyright © 2004 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on September 15, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on September 15, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

15-Deoxy-�12,14-Prostaglandin J2 Inhibits GlucocorticoidBinding and Signaling in Macrophages through a PeroxisomeProliferator-Activated Receptor �-Independent Process1

Adeline Cheron, Julie Peltier, Joelle Perez, Agnes Bellocq, Bruno Fouqueray, andLaurent Baud2

15-Deoxy-�12,14-PGJ2 (15d-PGJ2) is involved in the control of inflammatory reaction. We tested the hypothesis that 15d-PGJ2would exert this control in part by modulating the sensitivity of inflammatory cells to glucocorticoids. Human U937cells and mouseRAW 264.7 cells were exposed to 15d-PGJ2, and binding experiments were performed with [3H]dexamethasone as a glucocorticoidreceptor (GR) ligand. 15d-PGJ2 caused a transient and concentration-dependent decrease in [3H]dexamethasone-specific bindingto either cells through a decrease in the number of GR per cell without significant modification of theKd value. These changes wererelated to functional alteration of the GR rather than to a decrease in GR protein. They did not require the engagement ofperoxisome proliferator-activated receptor � (PPAR�), because the response to 15d-PGJ2 was neither mimicked by the PPAR�agonist ciglitazone nor prevented by the PPAR� antagonist bisphenol A diglycidyl ether. 15d-PGJ2 altered GR possibly throughthe interaction of its cyclopentenone ring with GR cysteine residues because the cyclopentenone ring per se could mimic the effectof 15d-PGJ2, and modification of GR cysteine residues with methyl methanethiosulfonate suppressed the response to 15d-PGJ2.Finally, 15d-PGJ2-induced decreases in glucocorticoid binding to GR resulted in parallel decreases in the ability of GR to activatethe transcription of a glucocorticoid-inducible reporter gene and to reduce the expression of monocyte chemoattractant protein-1.Together these data suggest that 15d-PGJ2 limits glucocorticoid binding and signaling in monocytes/macrophages through aPPAR�-independent and cyclopentenone-dependent mechanism. It provides a way in which 15d-PGJ2 would exert proinflam-matory activities in addition to its known anti-inflammatory activities. The Journal of Immunology, 2004, 172: 7677–7683.

A rachidonate is converted through the cyclooxygenase(COX)3 pathway to PGD2 and by nonenzymatic dehy-dration of PGD2 to 15-deoxy-�12,14-PGJ2 (15d-PGJ2)

during both the initial and the resolution phase of inflammatoryprocess (1). Administration of COX inhibitors during the firstphase inhibits inflammation, whereas their administration duringthe later phase exacerbates it (1). This observation raises the pos-sibility that 15d-PGJ2 could exert sequentially pro- and anti-inflammatory activities.

The proinflammatory properties of 15d-PGJ2 remain poorly de-fined. In vitro studies have only shown that 15d-PGJ2 increasesmonocyte/macrophage expression of the neutrophil-selective che-mokine IL-8 (2). By contrast, anti-inflammatory activities of 15d-PGJ2 have been well documented; they are mainly related to itsability to limit inflammatory cell accumulation and activation (3).15d-PGJ2 inhibits endothelial cell expression of adhesion mole-

cules (e.g., VCAM1 and ICAM1) and promotes leukocyte apopto-sis (3). In addition, 15d-PGJ2 inhibits the transcription of genescoding for proinflammatory proteins, including IL-1�, TNF-�,COX2, NO synthase-2, and matrix metalloproteinases (4, 5). Thiscontrol is associated with the inactivation of transcription factorssuch as NF-�B, AP-1, and STATs (4). Two molecular mechanismsare potentially involved. First, 15d-PGJ2 binds to the peroxisomeproliferator-activated receptor � (PPAR�), a member of the nu-clear receptor superfamily that decreases the availability of coac-tivators, and thereby prevents the activity of transcription factors(6). Alternatively, 15d-PGJ2 inhibits multiple steps in the NF-�Bsignaling pathway by PPAR�-independent mechanisms (7). Thesemechanisms involve direct interactions of a highly reactive elec-trophilic carbon atom in the cyclopentenone ring of 15d-PGJ2 withcritical cysteine residues in target proteins (8).

Glucocorticoids play also a crucial role in the resolution of in-flammation (9). They exert this role through intracellular glucocor-ticoid receptors (GR) that regulate gene transcription in two dif-ferent ways (10). First, GR may form a homodimer that bindsglucocorticoid-responsive elements (GREs) in the 5� upstreampromoter or enhancer region of glucocorticoid-responsive genes toincrease or repress the rate of their transcription. Alternatively, GRmonomer may interact directly with other transcription factors,such as AP-1 and NF-�B, to suppress their efficacy. Thus, theavailability of GR is a major factor limiting glucocorticoid action(11). During the inflammatory response, this availability is con-trolled by glucocorticoids themselves and by a great diversity ofpro- and anti-inflammatory mediators, including IL-4, IL-10, IL-13, somatostatin, and TGF-�1 (12).

Whether such a control is exerted by 15d-PGJ2 as well is notknown. Therefore, the present study was undertaken to explore the

Institut National de la Sante et de la Recherche Medicale, Unite 489, Serviced’Explorations Fonctionnelles Multidisciplinaires, AP-HP Hopital Tenon, Paris,France

Received for publication October 23, 2003. Accepted for publication April 9, 2004.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked advertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported by Institut National de la Sante et de la Recherche Medi-cale and Faculte de Medecine Saint-Antoine.2 Address correspondence and reprint requests to Dr. Laurent Baud, Institut Nationalde la Sante et de la Recherche Medicale, Unite 489, Hopital Tenon, 4 rue de la Chine,75020 Paris, France. E-mail address: laurent.baud@tnn.ap-hop-paris.fr3 Abbreviations used in this paper: COX, cyclooxygenase; BADGE, bisphenol Adiglycidyl ether; 15d-PGJ2, 15-deoxy-�12,14-PGJ2; GR, glucocorticoid receptor;GRE, glucocorticoid-responsive element; MCP-1, monocyte chemoattractant pro-tein-1; MMTS, methyl methanethiosulfonate; NAC, N-acetylcysteine; PPAR�, per-oxisome proliferator-activated receptor �; TK-Luc, thymidine-luciferase.

The Journal of Immunology

Copyright © 2004 by The American Association of Immunologists, Inc. 0022-1767/04/$02.00

by guest on September 15, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

effects of 15d-PGJ2 on glucocorticoid binding and signaling. Ev-idence was obtained that 15d-PGJ2 limits glucocorticoid bindingand signaling in human and murine monocytes/macrophagesthrough a PPAR�-independent mechanism. It provides a way inwhich 15d-PGJ2 could exert proinflammatory activities during theinflammatory process.

Materials and MethodsCells

The promonocytic U937cell line was obtained from American Type Cul-ture Collection (Manassas, VA) and maintained at 37°C in a culture me-dium (RPMI 1640 with 10 mM HEPES, 100 U/ml penicillin, and 100�g/ml streptomycin) supplemented with 10% FCS under a 5% CO2-95%air atmosphere. These cells were differentiated to a mature macrophage-like phenotype, as previously described (13): 0.5 � 106 cells/ml culturemedium supplemented with 2.5% FCS were pretreated for 24 h with 1.2%DMSO, washed, and resuspended in 0.5 ml/well culture medium (24-wellplate). They were treated with PMA (50 ng/ml; Sigma-Aldrich, St. Louis,MO) for 3 h at 37°C, washed, and then incubated with the indicated con-centrations of 15d-PGJ2 (Biomol, Plymouth Meeting, PA). The mouse cellline RAW 264.7 was obtained from American Type Culture Collection.Cells were grown to confluence in DMEM supplemented with 10% FCS,100 U/ml penicillin, and 100 �g/ml streptomycin under a 5% CO2-95% airatmosphere (14).

Cell viability analysis

Cells were trypsinized, washed, and resuspended in 0.2% trypan blue in PBS.Nonviable cells taking up blue dye were counted using a hemocytometer.

GR assay

A whole cell binding assay was used to determine GR number and affinityin untreated and 15d-PGJ2-treated cells (14). PMA-differentiated

U937cells or RAW 264.7 cells were incubated in culture medium supple-mented with 2.5% FCS and containing [3H]dexamethasone (85.0 Ci/mmol;Amersham Pharmacia Biotech Europe, Saclay, France) with or withoutunlabeled dexamethasone (Sigma-Aldrich). For Scatchard analysis, theconcentration of [3H]dexamethasone ranged from 0.5 to 64 nM; for one-point binding assays, 10 nM [3H]dexamethasone was used. Unlabeleddexamethasone was used at a concentration of 20 �M. After incubation at37°C for 3 h, monolayers were washed six times with cold PBS, and cellswere lysed in 1 N NaOH. Lysates were harvested and counted in a betaspectrometer.

Supplementary binding assays were performed using a cytosol fraction,as previously described (15, 16). RAW 264.7 cells were harvested byscraping into culture medium. The pellet was suspended in 1 ml of buffer(10 mM HEPES, 1 mM EDTA, and 20 mM sodium molybdate, pH 7.4)and ruptured with a plastic homogenizer. The homogenate was centrifugedfor 1 h to 18,500 � g at 4°C, and the supernatant referred to as cytosol wasfrozen until further analysis. For binding assays, cytosol fraction (0.25 ml)was incubated for 4 h at 4°C with 10 nM [3H]dexamethasone together withor without unlabeled dexamethasone. Bound and unbound [3H]dexametha-sone were separated by the charcoal-dextran technique (16).

Western blotting

Culture medium was removed, and RAW 264.7 cells were resuspended in1.5 ml of PBS using a cell scraper. The cells were then pelleted by cen-trifugation (600 � g, 4°C, for 10 min), frozen at �80°C for 4 h, thawed,and resuspended in 100 �l of ice-cold protease inhibitory buffer (1 mMPMSF, 20 �g/ml leupeptin, 20 �g/ml aprotinin in 50 mM Tris (pH 7.4),100 mM NaCl, 2 mM EDTA, and 1% Nonidet P-40). The lysate wascentrifuged (4,000 � g, 4°C, for 30 min). A portion of the supernatants wasreserved for protein determination, and the protein concentration in super-natant samples was adjusted with the protease inhibitory buffer. Thesesamples were boiled for 3 min in Laemmli loading buffer and subjected toelectrophoresis on a 7.5% polyacrylamide-SDS gel. Proteins were electro-eluted onto a nitrocellulose membrane (Immobilon-P; Millipore, Bedford,MA) that was blocked for 18 h at 4°C in 10% nonfat dry milk solution in

FIGURE 1. Effect of 15d-PGJ2 on [3H]dexamethasone-specific binding. RAW 264.7 cells (A) and PMA-differentiated U937cells (B) were exposed tothe indicated concentrations of 15d-PGJ2 for 3 h before the whole cell binding assay was performed with 10 nmol/L [3H]dexamethasone. C, RAW 264.7cells were exposed to the indicated concentrations of 15d-PGJ2 for 3 h before the percentage of dead cells was assessed by measuring trypan blue uptake.D, PMA-differentiated U937cells were exposed to 2.5 �M 15d-PGJ2 for the indicated periods before the whole cell binding assay was performed with 10nmol/L [3H]dexamethasone. Values are the mean � SD of three to six determinations. �, p � 0.05 compared with the15d-PGJ2-untreated control.

7678 15d-PGJ2 DECREASES GLUCOCORTICOID BINDING AND SIGNALING

by guest on September 15, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

PBS and 0.1% Tween. GR was detected using a polyclonal Ab (1/100dilution; Affinity Bioreagents, Golden, CO) and a peroxidase-labeled anti-IgG secondary Ab (1/10,000 dilution). Thereafter, the membrane was de-veloped with the ECL detection reagent (Amersham Pharmacia BiotechEurope).

Transient transfection and luciferase assay

U937 cells were transiently transfected with a glucocorticoid-inducible re-porter plasmid containing a minimal thymidine kinase promoter with two con-sensus GRE upstream of the gene for luciferase ([GRE]2 TK-Luc, a gift fromDr. M.-G. Catelli, Centre National de la Recherche Scientifique, Paris, France),using the DEAE-dextran procedure (14). Thereafter, these cells were dif-ferentiated with PMA and exposed to 15d-PGJ2 and/or dexamethasonebefore luciferase activity was assayed as previously described (17).

ELISA for monocyte chemoattractant protein-1 (MCP-1)

RAW 264.7 cells were incubated with or without 2.5 �M 15d-PGJ2 for 2 hand then with increasing concentrations of dexamethasone (10–100 nM)for 1 h, before being challenged with LPS (10 ng/ml). The cell supernatantswere harvested after 6 h, centrifuged to remove cell debris, and stored at�80°C until they were analyzed for MCP-1. Concentrations of MCP-1were measured by ELISA (Quantikine mouse JE/MCP-1; R&D Systems,Minneapolis, MN). The minimum detectable concentration was 2 pg/ml.

PMA-differentiated U937cells were preincubated with or without 5 �M15d-PGJ2 for 3 h. Thereafter, these cells were cultured for 6 h with dexa-methasone (10–100 nM), and MCP-1 concentrations were measured in thesupernatants by ELISA (Quantikine human MCP-1; R&D Systems).

Statistics

Results are presented as the mean � SD. Comparisons between groups ofvalues were made with Student’s t test. A difference between groups of p �0.05 was considered significant.

Results15d-PGJ2 limits glucocorticoid-specific binding tomonocytes/macrophages

To assess initially the sensitivity of monocytes/macrophages to15d-PGJ2, RAW 264.7 cells and PMA-differentiated U937cellswere exposed for 3 h to varying concentrations of 15d-PGJ2 beforebinding experiments were performed. The addition of 15d-PGJ2

caused a concentration-dependent decrease in [3H]dexamethasone-specific binding to either strain of cells (Fig. 1, A and B). Themaximal effect was seen at 10 �M. These changes could not berelated to differences in cell viability because cell exposure to 15d-PGJ2 for 3 h did not affect their capacity to exclude trypan blue(Fig. 1C). In PMA-differentiated U937cells, the above changes in[3H]dexamethasone-specific binding were detectable after 0.5 h,peaked after 3 h, and thereafter decreased progressively (Fig. 1D).

To determine whether 15d-PGJ2 promoted a decrease in GRnumber or affinity, saturation binding and Scatchard analysis wereperformed (Fig. 2). Exposure of PMA-differentiated U937cells to2.5 �M 15d-PGJ2 for 3 h yielded a 50% decrease in the averagenumber of GR sites per cell (control cells, 106,366 � 30,984;15d-PGJ2-treated cells, 54,336 � 4,340; p � 0.005; n � 3) withoutsignificant modification of the Kd value (control cells, 14.3 � 7.7nM; 15d-PGJ2-treated cells, 17.3 � 7.6 nM).

These results raised the question of whether 15d-PGJ2 might de-crease the steady state levels of GR protein. Western blotting exper-iments were therefore performed. Exposure of RAW 264.7 cells (Fig.3) or PMA-differentiated U937cells (data not shown) to 15d-PGJ2 for3 h did not modify GR levels, indicating that 15d-PGJ2-inducedchanges in [3H]dexamethasone-specific binding were related to func-tional alteration of the GR rather than to adecrease in GR protein.

15d-PGJ2 limits glucocorticoid-specific binding through aPPAR�- and membrane receptor-independent process

15d-PGJ2 binds to PPAR� and affects the inflammatory processthrough both PPAR�-dependent and -independent mechanisms. Inaddition, U937cells and RAW 264.7 cells express PPAR�, at highand low levels, respectively (4). Thus, we assessed the involve-ment of PPAR� in the response to 15d-PGJ2, by examining theeffects of PPAR� antagonist and agonist. Exposure of RAW 264.7cells or PMA-differentiated U937cells to the PPAR� antagonistbisphenol A diglycidyl ether (BADGE) (18, 19) augmented[3H]dexamethasone-specific binding, but did not interfere with the

FIGURE 2. Effect of 15d-PGJ2 on the number and the affinity of [3H]dexamethasone binding sites in PMA-differentiated U937cells. A and B, Saturationbinding (A) and Scatchard analysis (B) of [3H]dexamethasone binding data obtained from untreated cells (E) and cells exposed to 2.5 �M 15d-PGJ2 for3 h (F). One representative experiment of three is shown.

FIGURE 3. Western blot analysis of the effect of 15d-PGJ2 on GR pro-tein level. RAW 264.7 cells were exposed to the indicated concentrationsof 15d-PGJ2 for 3 h. One representative experiment of four is shown.

7679The Journal of Immunology

by guest on September 15, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

15d-PGJ2 response (Fig. 4, A and B). Similarly, exposure of RAW264.7 cells or PMA-differentiated U937cells to the potent PPAR�agonist ciglitazone did not affect [3H]dexamethasone-specificbinding (Fig. 4, C and D), suggesting that PPAR� is not involvedin the response to 15d-PGJ2.

At a lower concentration than that required for PPAR� activa-tion, 15d-PGJ2 can act through binding to a cell membrane receptor(20). Thus, we also determined whether 15d-PGJ2 could affect[3H]dexamethasone-specific binding in the absence of cell membrane.Cytosol fraction was isolated from RAW 264.7 cells and used in abinding assay. After 1-h cytosol exposure to 5 �M 15d-PGJ2,[3H]dexamethasone-specific binding decreased from 170.6 � 3.0 to118.6 � 14.5 fmol/mg protein ( p � 0.05; n � 3). Together theseresults suggest that 15d-PGJ2 limited glucocorticoid-specific bindingthrough a PPAR�- and membrane receptor-independent process.

15d-PGJ2 limits glucocorticoid-specific binding through theinteraction of its cyclopentenone ring with GR

15d-PGJ2 is characterized by the presence of a cyclopentenonering that contains an electrophilic carbon (7, 8). This carbon canreact covalently by means of the Michael addition reaction withnucleophiles such as the free sulfhydryls of cysteine residues inproteins. The hormone binding domain of GR contains such cys-teine residues, and their oxidation or chemical modification affectsligand binding (21). Thus, we first hypothesized that 15d-PGJ2 wouldinhibit glucocorticoid binding through a direct modification of thehormone binding domain of GR by the cyclopentenone ring system.Consistent with this hypothesis, cyclopentenone (2-cyclopenten-1-

one), a compound that contains also an unsaturated carbonyl group,caused a concentration-dependent decrease in [3H]dexamethasone-specific binding to RAW 264.7 cells (Fig. 5A). Efficient concentra-tions were �125–500 �M, as previously described in other cell mod-els (7, 22). In contrast, cyclopentanone and cyclopentene, whichcontain a saturated carbonyl and a double bond without carbonyl,respectively, had no effect (data not shown). As a further confirmationof this hypothesis, methyl methanethiosulfonate (MMTS), a mem-brane-permeant, cysteine-specific, modifying reagent (16), decreased[3H]dexamethasone-specific binding to RAW 264.7 cells and com-pletely prevented the response to 15d-PGJ2 (Fig. 5B).

Cyclopentenone PGs, including 15d-PGJ2, induce an intracel-lular oxidative stress that may underlie part of their biologicaleffects (23). As oxidative stress limits the ligand binding activity ofthe GR (24), the question arises as to whether 15d-PGJ2 wouldinhibit glucocorticoid binding indirectly, through the oxidation ofcritical cysteine residues in the GR. However, exposure of RAW264.7 cells to N-acetylcysteine (NAC), an antioxidant, did not af-fect the response to 15d-PGJ2 (Fig. 5C), indicating that such amechanism was not required.

15d-PGJ2 limits glucocorticoid signaling and anti-inflammatoryefficiency

We assessed whether 15d-PGJ2-induced changes in glucocorticoidbinding were associated with changes in glucocorticoid signaling.Given that glucocorticoids regulate gene transcription in part viathe binding of GR to GRE (10), we used a reporter gene under the

FIGURE 4. 15d-PGJ2 limits glucocorticoid-specific binding through a PPAR�-independent process. 1) The influence of BADGE, a PPAR� antagonist,on the response to 15d-PGJ2 was tested. RAW 264.7 cells (A) or PMA-differentiated U937cells (B) were preincubated for 10 min with or without 100 �MBADGE and incubated for 3 h with or without 5 �M 15d-PGJ2. Thereafter, the whole cell binding assay was performed with 10 nmol/L [3H]dexamethasone.Values are the mean � SD of triplicate determinations. 2) Effect of ciglitazone, a PPAR� agonist, on glucocorticoid binding. RAW 264.7 cells (C) or PMA-differentiated U937cells (D) were exposed to the indicated concentrations of ciglitazone for 3 h. Thereafter, the whole cell binding assay was performed with 10nmol/L [3H]dexamethasone. Values are the mean � SD of triplicate determinations. �, p � 0.05 compared with the15d-PGJ2-untreated control.

7680 15d-PGJ2 DECREASES GLUCOCORTICOID BINDING AND SIGNALING

by guest on September 15, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

control of a promoter including two copies of the GRE (Fig. 6).Exposure of PMA-differentiated U937cells transfected with[GRE]2 TK-Luc to dexamethasone for 4 h resulted in a 5-foldincrease in luciferase activity. The addition of 15d-PGJ2 caused aconcentration-dependent decrease in this response.

We also evaluated the pathophysiologic relevance of the 15d-PGJ2-induced changes in glucocorticoid binding and signaling. Asa marker of glucocorticoid efficacy, we examined the regulation ofMCP-1 expression. Dexamethasone decreased in a concentration-de-pendent manner the expression of MCP-1 in both LPS-stimulatedRAW 264.7 cells (Fig. 7A) and PMA-differentiated U937cells (Fig.7B). The addition of 15d-PGJ2 significantly limited this response. Fi-nally, consistent with the binding studies, 15d-PGJ2-induced changesin glucocorticoid anti-inflammatory efficiency were mimicked by cy-clopentenone (Fig. 7C) and MMTS (Fig 7D).

DiscussionThe present study demonstrates that 15d-PGJ2 transiently inhibitsglucocorticoid binding and signaling in monocytes/macrophages.The mechanism for this effect is likely to be through the interactionof the 15d-PGJ2 cyclopentenone ring with cysteine residues inconstituents of the GR activation pathway.

The lowest 15d-PGJ2 concentration required to inhibit GR bind-ing and signaling in monocytes/macrophages was �100 nM. Asthe amounts of 15d-PGJ2 released in vitro into the medium of

RAW 264.7 cells stimulated with LPS reach �80 nM (25), theamounts of 15d-PGJ2 generated in vivo at the inflammatory siteare presumably sufficient for it to play a role in modulating GRefficiency, although this may not be the case in all human tissues(26). Clearly, this role cannot be explained by a reduction in cellviability. Indeed, 15d-PGJ2 concentrations up to 5 �M did notaffect RAW 264.7 cell viability, and the 15d-PGJ2 effect on GRbinding was mostly reversible. These results are consistent withprevious studies showing that 15d-PGJ2 induces an apoptotic re-sponse only at concentrations 5 �M (27).

After 15d-PGJ2 addition, decreases in GR binding were fast,being detectable after 30 min. This suggests that the response to15d-PGJ2 is related to rapid processes such as GR degradation orGR post-translational modifications that impede ligand binding.An alternative would be that 15d-PGJ2 decreases GR gene tran-scription and, hence, GR expression. This latter possibility is sup-ported by the observation that 15d-PGJ2 can significantly reduceCOX2 gene transcription and COX2 expression after �3 and 4 h,respectively (28). However, this possibility is unlikely becauseGR protein amounts were not significantly modified by effectiveconcentrations of 15d-PGJ2. In addition, decreased GR genetranscription would have no repercussions on GR protein avail-ability after a �3-h period, because the GR turnover rate is slow(half-life of GR protein reaches 19 h) (29, 30).

FIGURE 5. 15d-PGJ2 limits glucocorticoid-specific binding through the interaction of its cyclopentenone ring with GR. A, Effect of cyclopentenone onglucocorticoid binding. RAW 264.7 cells were exposed to the indicated concentrations of cyclopentenone for 3 h. Thereafter, the whole cell binding assaywas performed with 10 nmol/L [3H]dexamethasone. B, Influence of MMTS, a cysteine-specific modifying reagent, on the response to 15d-PGJ2. RAW264.7 cells were preincubated for 30 min with (F) or without (E) 500 �M MMTS and exposed to the indicated concentrations of 15d-PGJ2 for 3 h.Thereafter, the whole cell binding assay was performed with 10 nmol/L [3H]dexamethasone. C, Influence of NAC, a scavenger of reactive oxygen species,on the response to 15d-PGJ2. RAW 264.7 cells were preincubated for 10 min with (F) or without (E) 5 mM NAC and exposed to the indicatedconcentrations of 15d-PGJ2 for 3 h. Thereafter, the whole cell binding assay was performed with 10 nmol/L [3H]dexamethasone. Values are the mean �SD of three or four determinations. �, p � 0.05 compared with the15d-PGJ2-untreated control.

7681The Journal of Immunology

by guest on September 15, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Interestingly, 15d-PGJ2-induced modifications in GR bindingwere independent of PPAR� activation, as suggested by two ob-servations. First, although U937cells express high levels and RAW264.7 cells express negligible levels of PPAR� (4, 31), 15d-PGJ2

exhibited a similar effectiveness in the two cell lines. Second, theresponse to 15d-PGJ2 was neither mimicked by the specific 15d-PGJ2 agonist ciglitazone nor modified by the 15d-PGJ2 antagonistBADGE. In agreement with this observation, 15d-PGJ2 has beenshown previously to control several functions of monocytes/mac-rophages through PPAR�-independent mechanisms (32).

15d-PGJ2 is characterized by the presence of a cyclopentenonering with an electrophilic carbon that is able to form Michael ad-ducts with nucleophiles such as free sulfhydryls in cysteine resi-dues (7, 8). By this mechanism, 15d-PGJ2 has been shown to di-rectly affect the activity of proteins, including I�B and NF-�B (7,8). In our study several lines of evidence suggest that the responseto 15d-PGJ2 depends on an interaction between the 15d-PGJ2 cy-clopentenone ring and cysteine residues in constituents of the GRactivation pathway. First, the response to 15d-PGJ2 was mimickedby 2-cyclopenten-1-one. That levels of 2-cyclopenten-1-one re-quired to modify GR binding were 100-fold higher than levels of15d-PGJ2 could be explained by the fact that 2-cyclopenten-1-oneand 15d-PGJ2 contain one and two electrophilic carbons, respec-tively. Second, cell exposure to MMTS, a membrane-permeant,cysteine-specific, modifying reagent, decreased GR binding andcompletely suppressed the response to 15d-PGJ2. The five cysteineresidues contained in the hormone binding domain of GR (21) arepresumably the main target for the two electrophilic carbons of15d-PGJ2. Consistent with this mechanism, modifications of cys-teine residues in the hormone binding domain by oxidation, ni-trosylation, or chemical alteration have been shown previously tolimit the GR binding capacity (15, 24, 33). An alternative expla-nation would be that 15d-PGJ2 modifies cysteine residues in con-stituents of the GR activation pathway indirectly by promoting anoxidative stress. Indeed, 15d-PGJ2 may decrease antioxidant pro-teins such as glutathione and glutathione peroxidase, and, hence,increase the availability of reactive oxygen species that react withNO to form peroxinitrites (23, 26). Nevertheless, this indirectmechanism is unlikely, because NAC, an antioxidant, did not pre-vent the response to 15d-PGJ2.

FIGURE 6. Effect of 15d-PGJ2 on glucocorticoid signaling. U937cellswere transiently transfected with the reporter plasmid [GRE]2 TK-Luc,differentiated with PMA, and exposed to the indicated concentrations of15d-PGJ2 for 3 h. Thereafter, they were incubated with (F) or without (E)5 nM dexamethasone for 4 h before luciferase levels were measured. Val-ues are the mean � SD of triplicate determinations. �, p � 0.05 comparedwith the15d-PGJ2-untreated control.

FIGURE 7. Effect of 15d-PGJ2 on glucocorticoid anti-inflammatory efficiency. A, RAW 264.7 cells were incubated with (F) or without (E) 2.5 �M15d-PGJ2 for 2 h and then with the indicated concentrations of dexamethasone for 1 h before being challenged with LPS (10 ng/ml). MCP-1 concentrationswere measured in the supernatants by ELISA after 6 h. B, PMA-differentiated U937cells were preincubated with (F) or without (E) 5 �M 15d-PGJ2 for3 h. Thereafter, these cells were cultured for 6 h with the indicated concentrations of dexamethasone, and MCP-1 concentrations were measured in thesupernatants by ELISA. C, RAW 264.7 cells were incubated with or without 250 �M cyclopentenone for 2 h and then with 50 nM dexamethasone for 1 hbefore being challenged with LPS (10 ng/ml). MCP-1 concentrations were measured in the supernatants by ELISA after 6 h. D, RAW 264.7 cells wereincubated with or without 500 �M MMTS for 30 min and then with 50 nM dexamethasone for 1 h before being challenged with LPS (10 ng/ml). MCP-1concentrations were measured in the supernatants by ELISA after 6 h. Values are the mean � SD of values obtained in three or four independentexperiments. �, p � 0.05 compared with the untreated control.

7682 15d-PGJ2 DECREASES GLUCOCORTICOID BINDING AND SIGNALING

by guest on September 15, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

In addition to its effect on the binding capacity of GR, 15d-PGJ2

promoted a decrease in the ability of GR to control the expressionof MCP-1, a proinflammatory chemokine. This response is pre-sumably due to a decrease in the transcriptional activity of GR, assuggested by experiments using [GRE]2 TK-Luc-transfected cells.Alternatively, this response may be caused by a decrease in themolecular interaction of GR monomer with other transcription fac-tors, such as AP-1 and NF-�B. Thus, at the onset of the inflam-matory process, when high levels of 15d-PGJ2 are produced (1),15d-PGJ2 would participate in the development of inflammationby limiting both the binding and the anti-inflammatory efficiencyof corticoids. The availability of corticoids depends partly on theactivity of 11�-hydroxysteroid dehydrogenase-1, a reductase thatconverts inactive cortisone to the active GR agonist cortisol (34).Interestingly, PPAR� ligands have been shown to down-regulatethe expression of 11�-hydroxysteroid dehydrogenase-1 (35). Thus,in vivo, 15d-PGJ2 would blunt glucocorticoid efficiency by twodifferent ways, by limiting the availability of cortisol through aPPAR�-dependent mechanism and by impeding its binding to GRthrough a PPAR�-independent mechanism.

In summary, we have identified a novel molecular mechanismby which 15d-PGJ2 may exert proinflammatory effects in additionto its numerous anti-inflammatory effects. This finding highlightsthe necessity to precise the timing of either pro- or anti-inflamma-tory effects of 15d-PGJ2 throughout the inflammatory process be-fore identifying 15d-PGJ2 as a potential target for therapeuticintervention.

AcknowledgmentsWe thank M.-G. Catelli for the [GRE]2 TK-Luc construct, andN. Sabirhoussen for secretarial assistance.

References1. Gilroy, D. W., P. R. Colville-Nash, D. Willis, J. Chivers, M. J. Paul-Clark, and

D. A. Willoughby. 1999. Inducible cyclooxygenase may have anti-inflammatoryproperties. Nat. Med. 5:698.

2. Zhang, X., J. M. Wang, W. H. Gong, N. Mukaida, and H. A. Young. 2001.Differential regulation of chemokine gene expression by 15-deoxy-�12,14 pros-taglandin J2. J. Immunol. 166:7104.

3. Lawrence, T., D. A. Willoughby, and D. W. Gilroy. 2002. Anti-inflammatorylipid mediators and insights into the resolution of inflammation. Nat. Rev. Im-munol. 2:787.

4. Ricote, M., A. C. Li, T. M. Willson, C. J. Kelly, and C. K. Glass. 1998. Theperoxisome proliferator-activated receptor-� is a negative regulator of macro-phage activation. Nature 391:79.

5. Jiang, C., A. T. Ting, and B. Seed. 2000. PPAR-� agonists inhibit production ofmonocyte inflammatory cytokines. Nature 391:82.

6. Li, M., G. Pascual, and C. K. Glass. 2000. Peroxisome proliferator-activatedreceptor �-dependent repression of the inducible nitric oxide synthase gene. Mol.Cell. Biol. 20:4699.

7. Straus, D. S., G. Pascual, M. Li, J. S. Welch, M. Ricote, C.-H. Hsiang,L. L. Sengchanthalangsy, G. Ghosh, and C. K. Glass. 2000. 15-Deoxy-�12,14-prostaglandin J2 inhibits multiple steps in the NF-�B signaling pathway. Proc.Natl. Acad. Sci. USA 97:4844.

8. Rossi, A., P. Kapahi, G. Natoli, T. Takahashi, Y. Chen, M. Karin, andM. G. Santoro. 2000. Anti-inflammatory cyclopentenone prostaglandins are di-rect inhibitors of I�B kinase. Nature 403:103.

9. Barnes, P. J. 1998. Anti-inflammatory actions of glucocorticoids: molecularmechanisms. Clin. Sci. 94:557.

10. Karin, M. 1998. New twists in gene regulation by glucocorticoid receptor: isDNA binding dispensable? Cell 93:487.

11. Bronnegard, M. 1996. Steroid receptor number. Individual variation and down-regulation by treatment. Am. J. Respir. Crit. Care Med. 154:S28.

12. Baud, L., B. Fouqueray, A. Bellocq, S. Doublier, and J. Peltier. 2000. Inflam-matory mediators modulate glucocorticoid binding and signalling. Curr. TrendsImmunol. 3:177.

13. Brigino, E., S. Haraguchi, A. Koutsonikolis, G. J. Cianciolo, U. Owens,R. A. Good, and N. K. Day. 1997. Interleukin 10 is induced by recombinantHIV-1 Nef protein involving the calcium/calmodulin-dependent phosphodiester-ase signal transduction pathway. Proc. Natl. Acad. Sci. USA 94:3178.

14. Bellocq, A., S. Doublier, S. Suberville, J. Perez, B. Escoubet, B. Fouqueray,D. R. Puyol, and L. Baud. 1999. Somatostatin increases glucocorticoid bindingand signaling in macrophages by blocking the calpain-specific cleavage of Hsp90. J. Biol. Chem. 274:36891.

15. Galigniana, M. D., G. Piwien-Pilipuk, and J. Assreuy. 1999. Inhibition of glu-cocorticoid receptor binding by nitric oxide. Mol. Pharmacol. 55:317.

16. Souque, A., J. Fagart, B. Couette, and M.-E. Rafestin-Oblin. 1996. Sulfhydrylgroups are involved in the binding of agonists and antagonists to the humanmineralocorticoid receptor. J. Steroid Biochem. Mol. Biol. 57:315.

17. Bellocq, A., S. Suberville, C. Philippe, F. Bertrand, J. Perez, B. Fouqueray,G. Cherqui, and L. Baud. 1998. Low environmental pH is responsible for theinduction of nitric-oxide synthase in macrophages: evidence for involvement ofnuclear factor-�B. J. Biol. Chem. 273:5086.

18. Wright, H. M., C. B. Clish, T. Mikami, S. Hauser, K. Yanagi, R. Hiramatsu,C. N. Serhan, and B. M. Spiegelman. 2000. A synthetic antagonist for the per-oxisome proliferator-activated receptor � inhibits adipocyte differentiation.J. Biol. Chem. 275:1873.

19. Hinz, B., K. Brune, and A. Pahl. 2003. 15-Deoxy-�12,14-prostaglandin J2 inhibitsthe expression of proinflammatory genes in human blood monocytes via a PPAR-�-independent mechanism. Biochem. Biophys. Res. Commun. 302:415.

20. Monneret, G., H. Li, J. Vasilescu, J. Rokach, and W. S. Powell. 2002. 15-Deoxy-�12,14-prostaglandin D2 and J2 are potent activators of human eosinophils. J. Im-munol. 168:3563.

21. Simons, S. S., and W. B. Pratt. 1995. Glucocorticoid receptor thiols and steroid-binding activity. Methods Enzymol. 251:406.

22. Cippitelli, M., C. Fionda, D. Di Bona, A. Lupo, M. Piccoli, L. Frati, andA. Santoni. 2003. The cyclopentenone-type prostaglandin 15-deoxy-�12,14-pros-taglandin J2 inhibits CD95 ligand gene expression in T lymphocytes: interferencewith promoter activation via peroxisome proliferator-activated receptor-�-inde-pendent mechanisms. J. Immunol. 170:4578.

23. Kondo, M., T. Oya-Ito, T. Kumagai, T. Osawa, and K. Uchida. 2001. Cyclopen-tenone prostaglandins as potential inducers of intracellular oxidative stress.J. Biol. Chem. 276:12076.

24. Makino, Y., K. Okamoto, N. Yoshikawa, M. Aoshima, K. Hirota, J. Yodoi,K. Umesono, I. Makino, and H. Tanaka. 1996. Thioredoxin: a redox-regulatingcellular cofactor for glucocorticoid hormone action: cross talk between endocrinecontrol of stress response and cellular antioxidant defense system. J. Clin. Invest.98:2469.

25. Shibata, T., M. Kondo, T. Osawa, N. Shibata, M. Kobayashi, and K. Uchida.2002. 15-Deoxy-�12,14-prostaglandin J2. A prostaglandin D2 metabolite gener-ated during inflammatory processes. J. Biol. Chem. 277:10459.

26. Bell-Parikh, L. C., T. Ide, J. A. Lawson, P. McNamara, M. Reilly, andG. A. FitzGerald. 2003. Biosynthesis of 15-deoxy-�12,14-PGJ2 and the ligation ofPPAR�. J. Clin. Invest. 112:945.

27. Hortelano, S., A. Castrillo, A. M. Alvarez, and L. Bosca. 2000. Contribution ofcyclopentenone prostaglandins to the resolution of inflammation through the po-tentiation of apoptosis in activated macrophages. J. Immunol. 165:6525.

28. Subbaramaiah, K., D. T. Lin, J. C. Hart, and A. J. Dannenberg. 2001. Peroxisomeproliferator-activated receptor-� ligands suppress the transcriptional activation ofcyclooxygenase-2. J. Biol. Chem. 276:12440.

29. Mc Intyre, W. R., and H. H. Samuels. 1985. Triamcinolone acetonide regulatesglucocorticoid-receptor levels by decreasing the half-life of the activated nuclear-receptor form. J. Biol. Chem. 260:418.

30. Webster, J. C., C. M. Jewell, J. E. Bodwell, A. Munck, M. Sar, andJ. A. Cidlowski. 1997. Mouse glucocorticoid receptor phosphorylation status in-fluences multiple functions of the receptor protein. J. Biol. Chem. 272:9287.

31. Castrillo, A., M. Mojena, S. Hortelano, and L. Bosca. 2001. Peroxisome prolif-erator-activated receptor-�-independent inhibition of macrophage activation bythe non-thiazolidinedione agonist L-796,449: comparison with the effects of 15-deoxy-�12,14-prostaglandin J2. J. Biol. Chem. 276:34082.

32. Harris, S. G., J. Padilla, L. Koumas, D. Ray, and R. P. Phipps. 2002. Prostaglan-dins as modulators of immunity. Trends Immunol. 23:144.

33. Okamoto, K., H. Tanaka, H. Ogawa, Y. Makino, H. Eguchi, S.-I. Hayashi,N. Yoshikawa, L. Poellinger, K. Umesono, and I. Makino. 1999. Redox-depen-dent regulation of nuclear import of the glucocorticoid receptor. J. Biol. Chem.274:10363.

34. Escher, G., I. Galli, B. S. Vishwanath, B. M. Frey, and F. J. Frey. 1997. Tumornecrosis factor � and interleukin 1� enhance the cortisone/cortisol shuttle. J. Exp.Med. 186:189.

35. Berger, J., M. Tanen, A. Elbrecht, A. Hermanowski-Vosatka, D. E. Moller,S. D. Wright, and R. Thieringer. 2001. Peroxisome proliferator-activated recep-tor-� ligands inhibit adipocyte 11�-hydroxysteroid dehydrogenase type 1 expres-sion and activity. J. Biol. Chem. 276:12629.

7683The Journal of Immunology

by guest on September 15, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from