This information is current as
of June 17, 2017.
Cyclooxygenase-2 Expression by Nonsteroidal
Anti-inflammatory Drugs in Human Airway
Smooth Muscle Cells: Role of Peroxisome
Proliferator-Activated Receptors
Linhua Pang, Mei Nie, Lisa Corbett and Alan J. Knox
J Immunol 2003; 170:1043-1051; ;
doi: 10.4049/jimmunol.170.2.1043
http://www.jimmunol.org/content/170/2/1043
Subscription
Permissions
Email Alerts
This article cites 49 articles, 28 of which you can access for free at:
http://www.jimmunol.org/content/170/2/1043.full#ref-list-1
Information about subscribing to The Journal of Immunology is online at:
http://jimmunol.org/subscription
Submit copyright permission requests at:
http://www.aai.org/About/Publications/JI/copyright.html
Receive free email-alerts when new articles cite this article. Sign up at:
http://jimmunol.org/alerts
The Journal of Immunology is published twice each month by
The American Association of Immunologists, Inc.,
1451 Rockville Pike, Suite 650, Rockville, MD 20852
Copyright © 2003 by The American Association of
Immunologists All rights reserved.
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
References
The Journal of Immunology
Cyclooxygenase-2 Expression by Nonsteroidal
Anti-inflammatory Drugs in Human Airway Smooth Muscle
Cells: Role of Peroxisome Proliferator-Activated Receptors1
Linhua Pang,2 Mei Nie, Lisa Corbett, and Alan J. Knox
C
yclooxygenase (COX),3 the rate-limiting enzyme in the
conversion of arachidonic acid (AA) to prostanoids, exists in two isoforms. COX-1 is constitutively expressed
in healthy cells, whereas COX-2 is associated with inflammation
and is induced by inflammatory stimuli. The expression of COX-2
is a key element in the pathophysiology of several inflammatory
disorders (1), and its regulation differs between cell types.
Nonsteroidal anti-inflammatory drugs (NSAIDs) are widely
used for their anti-inflammatory and analgesic properties and inhibition of COX activity, and the subsequent proinflammatory prostanoid generation is an important mechanism of their action.
However, NSAIDs have other pharmacological effects, including inhibition of the transcription factor NF-␬B that is critically involved in
the expression of several inflammatory genes (2, 3) and activation of
peroxisome proliferator-activated receptors (PPARs) (4 –7).
Division of Respiratory Medicine, City Hospital, University of Nottingham, Nottingham, United Kingdom
Received for publication August 20, 2002. Accepted for publication November
11, 2002.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1
This study was supported by the Wellcome Trust (Grant 05710) and the National
Asthma Campaign (Grant 01/029).
2
Address correspondence and reprint requests to Dr. Linhua Pang, Division of
Respiratory Medicine, Clinical Sciences Building, City Hospital, University
of Nottingham, Hucknall Road, Nottingham NG5 1PB, U.K. E-mail address:
[email protected]
3
Abbreviations used in this paper: COX, cyclooxygenase; HASM, human airway
smooth muscle; AA, arachidonic acid; NSAID, nonsteroidal anti-inflammatory drug;
Indo, indomethacin; Flur, flurbiprofen; Cig, ciglitizone; Dex, dexamethasone; PPAR,
peroxisome proliferator-activated receptor; PPRE, peroxisome proliferator response
element; 15d-PGJ2, 15-deoxy-⌬12,14-PGJ2; CRE, cAMP response element.
Copyright © 2003 by The American Association of Immunologists, Inc.
PPARs are members of the nuclear receptor superfamily of ligand-activated transcription factors and exist in three subtypes: ␣,
␤(␦), and ␥ (8). PPAR␣, predominantly localized to the liver, orchestrates ␤-oxidation of fatty acids. PPAR␥, predominantly expressed in adipose tissue, regulates adipocyte differentiation; the
physiological function of PPAR␤ is unclear (8). PPARs are also
expressed in other cells (5, 9), but their expression and functions in
airways is unknown. PPARs are activated by a heterogeneous
group of structurally dissimilar chemicals and the selectivity is
activator concentration- and cell type-dependent (8). The PGD2
metabolite 15-deoxy-⌬12,14PGJ2 (15d-PGJ2) is a direct-binding ligand for PPAR␥ (10). Other PPAR activators include WY-14643
(selective for PPAR␣), several NSAIDs, antidiabetic drugs (thiazolidinediones), and fatty acids (e.g., AA and other eicosanoids) (8).
PPARs have complex regulatory effects on inflammatory responses.
PPAR␣-deficient mice show a prolonged inflammatory response (11),
whereas PPAR␥ activation inhibits inducible NO synthase, TNF-␣,
IL-6, and IL-l␤ expression (9, 12) and prevents matrix metalloproteinase induction (13) in stimulated monocytes/macrophages.
The effect of NSAIDs on COX-2 expression and the role of
PPARs in this process is controversial. Meade et al. (4) demonstrated that COX inhibitors (NSAIDs), substrate (AA), and products (PGD2, 15d-PGJ2, and PGF2␣) induce COX-2 expression
transcriptionally in epithelial cells through PPARs, whereas Staels
et al. (5) found that PPAR␣ activation by WY-14643 suppresses
IL-1-induced COX-2 expression in human aortic smooth muscle
cells. It has also been shown that NSAIDs induce COX-2 expression but inhibit mitogen-induced COX-2 expression in the colon
cancer cell line (HT-29) and macrophage cell line (RAW 264.7)
(6), and that COX-2 expression is regulated by a negative feedback
loop mediated through PPAR␥ in the macrophage-like-differentiated U937 cells (7). Therefore, COX-2 expression is likely to be
0022-1767/03/$02.00
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
Nonsteroidal anti-inflammatory drugs (NSAIDs) have been shown to modulate cyclooxygenase (COX)-2 expression, but the mechanisms involved are controversial and may be cell specific. We show in this study that indomethacin (Indo), flurbiprofen (Flur),
and the selective COX-2 inhibitor NS-398 induced COX-2 expression and markedly enhanced IL-1␤-induced COX-2 expression
in human airway smooth muscle (HASM) cells. These effects were not reversed by exogenous PGE2, suggesting that they are
prostanoid-independent. Indeed, PGE2 also induced and enhanced IL-1␤-induced COX-2 expression. Peroxisome proliferatoractivated receptor (PPAR) ␣ and PPAR␥ (not PPAR␤) were expressed in HASM cells. PPAR␥ activators ciglitizone (Cig) and
15-Deoxy-⌬12,14-PGJ2 (15d-PGJ2), but not the PPAR␣ activator WY-14643, mimicked the effect of NSAIDs on COX-2 expression.
Treatment with Flur, NS-398, Cig, and 15d-PGJ2 alone, but not Indo and WY-14643, elevated COX activity; however, neither
enhanced IL-1␤-induced COX activity. Pretreatment with dexamethasone suppressed COX-2 expression, PGE2 release, and COX
activity induced by NS-398, Cig, IL-1␤, alone or in combination. Unlike IL-1␤, NS-398 and Cig did not cause NF-␬B (p65) nuclear
translocation, nor did they further enhance IL-1␤-induced NF-␬B translocation, but they stimulated PPAR␥ translocation. Indo,
NS-398, Flur, and 15d-PGJ2, but not WY-14643, induced transcriptional activity of a COX-2 reporter construct containing the
peroxisome proliferator response element (PPRE) on their own and enhanced the effect of IL-1␤, but had no effect on a COX-2
reporter construct lacking the PPRE. The results suggest that COX-2 expression by NSAIDs is biologically functional, prostanoidindependent, and involves PPAR␥ activation, and provide the first direct evidence that the PPRE in the promoter is required for
NSAID-induced COX-2 expression. The Journal of Immunology, 2003, 170: 1043–1051.
1044
Materials and Methods
Materials
DMEM, penicillin and streptomycin, L-glutamine, amphotericin B, PGE2,
SDS, PMSF, Triton X-100, glycerol, acrylamide/bis-acrylamide, Tris,
DTT, leupeptin, pepstatin, 2-ME, MTT, EGTA, polyclonal rabbit antimouse IgG coupled with HRP, polyclonal goat anti-rabbit IgG with HRP,
anti-PGE2 serum, ciglitizone (Cig), indomethacin (Indo), Flur, AA, dexamethasone (Dex), and other unspecified chemicals were all purchased from
Sigma-Aldrich (Poole, U.K.). FCS was purchased from Sera-Lab (Loughborough, U.K.). Recombinant human IL-1␤ was from R&D Systems
(Minneapolis, MN). NS-398, WY-14643, and 15d-PGJ2 were from Calbiochem-Novabiochem (Nottingham, U.K.). Anti-human COX-2 and
anti-COX-1 Abs were from Cayman Chemical (Ann Arbor, MI). Abs
against PPAR␣, ␤(␦), ␥, NF-␬B (p65), and I␬B␣ were from Santa Cruz
Biotechnology (Santa Cruz, CA); [5, 6, 8, 11, 12, 14, 15(n)-3H]PGE2,
Rainbow-colored protein m.w. markers, ECL Western blotting detection
reagent, and Hyperfilm-ECL were from Amersham Pharmacia Biotech
(Little Chalfont, U.K.). Pure nitrocellulose membrane was from Gelman
Sciences (Northampton, U.K.). The Bio-Rad protein assay reagent was
from Bio-Rad (Hemel Hempstead, U.K.). FuGene 6 transfection reagent
was from Roche (East Sussex, U.K.). The dual-luciferase reporter assay
system was from Promega (Southampton, U.K.).
Cell culture
Primary cultures of HASM cells were prepared from explants of HASM as
previously reported (19, 20). This protocol was approved by the Nottingham City Hospital Research Ethics Committee. Cells at passages 5 and 6
were used for all experiments. We have previously shown that the cells
grown in this manner depict the immunohistochemical and light microscopic characteristics of typical HASM cells (19). HASM cells were also
obtained from BioWhittaker (Wokingham, U.K.) and were used at
passage 6.
Experiment protocols
The cells were cultured to confluence in DMEM supplemented with 10%
FCS, the antibiotics penicillin (100 U/ml) and streptomycin (100 ␮g/ml),
the antifungal amphotericin B (2.5 ␮g/ml), and L-glutamine (4 mM) in
humidified 5% CO2/95% air at 37°C and growth-arrested in serum-free
medium for 24 h before experiments. Immediately before each experiment,
fresh serum-free medium was added. To test the effect of NSAIDs and
PPAR activators on their own, cells were treated with the drugs for 4 h in
most cases or for the times indicated in the time-course experiments. To
test the effect of NSAIDs and PPAR activators on IL-1␤-induced COX-2
expression, PGE2 release, and COX activity, they were added 30 min before the incubation with IL-1␤. The inhibitory effect of Dex was assessed
by preincubating the cells with Dex for 30 min before the incubation with
NSAIDs, PPAR activators, and IL-1␤, alone or in combination, for a further 4 h. All the reagents were dissolved in DMSO (final concentration
ⱕ0.4% v/v). In all the experiments, a group of control cells were incubated
with the drug vehicle for the same period of time as the experimental cells.
PGE2 assay and COX activity
After experiments, the culture media in 24-well plates were collected and
stored at ⫺20°C until the determination of PGE2 content by RIA as described previously (19, 20). The sensitivity for PGE2 was 75 pg/ml. The
anti-PGE2 antiserum had negligible cross-reactivity with other prostanoids
(19) except 15d-PGJ2. COX activity was assayed functionally by washing
the cells three times in PBS after the experiments and then incubating with
AA (15 ␮M) for a further 15 min. These samples were subjected to RIA for
PGE2 and the resulting PGE2 level was taken as an index of COX activity.
Preparation of whole-cell lysate
After treatment, cells in 24-well plates were washed twice with ice-cold
PBS and incubated for 5 min with an extraction buffer (0.9% NaCl, 20 mM
Tris-HCl, pH 7.6, 0.1% Triton X-100, 1 mM PMSF, 0.01% leupeptin) with
gentle shaking. The samples were collected and centrifuged, and the protein concentration in the supernatant was determined using the Bio-Rad
protein assay reagent.
Preparation of cytosolic and nuclear proteins
Nuclear and cytosolic extracts from the cells were prepared as described by
Eickelberg et al. (27) with minor changes. After treatment, cells in 90-mm
dishes were washed twice with ice-cold PBS and harvested in 1 ml of PBS
with a cell scraper. The samples were centrifuged for 2 min at 1000 ⫻ g,
and cell pellets were resuspended in 200 ␮l of low salt buffer (20 mM
HEPES, pH 7.9, 10 mM KCl, 0.1 mM NaVO4, 1 mM EDTA, 1 mM EGTA,
0.2% IGEPAL CA-630, 10% glycerol, 1 mM PMSF, and 0.01% leupeptin).
After 10 min of incubation on ice, the samples were centrifuged at 7000 ⫻
g for 30 min (4°C), and the supernatants were taken as cytosolic extracts.
Nuclei were then resuspended in 100 ␮l of high salt buffer (as low salt
buffer but with 420 mM NaCl and 20% glycerol), and nuclear proteins
were extracted by shaking on ice for 30 min. Samples were then centrifuged at 7000 ⫻ g for 30 min (4°C), and the supernatants were taken as
nuclear extracts.
Western blot analysis
Identification of the interested proteins were performed by Western blotting analysis as described before (19, 20). Briefly, protein samples (30
␮g/track) were subject to electrophoresis in 7.5% SDS-polyacrylamide gel.
Separated proteins were electroblotted to pure nitrocellulose membranes
and the blot was blocked for 2 h at room temperature with blocking buffer
(wash buffer with 8% fat-free dried milk powder). The blot was then incubated with monoclonal anti-COX-2 Ab (1/2000 dilution with blocking
buffer), or polyclonal Abs against PPAR␣, ␤(␦), ␥, NF-␬B (p65), or I␬B␣
(1/1000 dilution in blocking buffer) for 2 h, washed with wash buffer (PBS
pH 7.4 with 0.3% Tween 20), and incubated with HRP-conjugated secondary Abs (1/2000 dilution with blocking buffer) for 1 h. The blot was
washed again and then incubated with ECL Western blotting detection
reagent for 1 min and finally exposed to Hyperfilm-ECL.
Plasmids
The 3.9-kb COX-2 firefly luciferase reporter construct (3.9-kb COX-2Luc) in pGL3 vectors containing the putative PPRE sequence located between ⫺3721 to ⫺3707 (AGGCGACAGGTCA) upstream of the starting
point of the COX-2 gene and the 3.5-kb COX-2 firefly luciferase reporter
construct (3.5-kb COX-2-Luc) lacking the PPRE were kindly provided by
T. McIntyre (University of Utah, Salt Lake City, UT) and have been previously described in detail (4, 28). The internal Renilla luciferase control
vector pRL-SV40 was purchased from Promega.
Transfection of HASM cells and reporter assays
All transient transfections were performed using FuGene 6 (1 ␮g DNA:3
␮l FuGene 6) according to the manufacturer’s recommended protocol.
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
regulated in a cell-specific manner by different PPAR activators.
However, no studies have shown that the peroxisome proliferator
response element (PPRE) in the promoter region is required for
COX-2 expression by NSAIDs.
We and others have shown that human airway smooth muscle
(HASM) cells have important synthetic functions. In addition to
the synthesis of cytokines/chemokines (14 –16) and growth factors
(17, 18), HASM cells express COX-2 and release large quantities
of prostanoids (mainly PGE2) upon stimulation by proinflammatory cytokines and other mediators (19 –21). PGE2 produced from
COX-2 inhibits cell proliferation (22), mediates IL-1␤- and bradykinin-induced attenuation of cAMP generation in response to
␤2-adrenoceptor agonists (23, 24), and acts as an autocrine regulator of IL-8 and vascular endothelial growth factor release (15,
18), suggesting COX-2 induction modulates airway smooth muscle function (25). In terms of the effect of NSAIDs on COX-2
expression, Bonazzi et al. (26) showed that COX inhibitor flurbiprofen (Flur) reduced IL-1␤-induced COX-2 expression in HASM
cells but did not perform mechanistic studies.
To investigate the regulation of COX-2 expression and function
by NSAIDs in HASM cells and the role of PPAR activators in this
process, we studied the effect of NSAIDs on COX-2 expression in
the presence or absence of IL-1␤, a COX-2 inducer, and compared
the effect of NSAIDs with other PPAR activators. We show in this
study that NSAIDs induce COX-2 expression and enhance IL-1␤induced COX-2 expression via a prostanoid-independent mechanism involving PPAR␥, but not NF-␬B, activation. We also show
for the first time that deletion of the PPRE in the COX-2 promoter
abolishes the effect of NSAIDs. Paradoxically, PGE2, the major
COX product of these cells, also induces COX-2.
NSAIDs ENHANCE COX-2 EXPRESSION
The Journal of Immunology
1045
Results
NSAIDs enhance COX-2 expression by a prostanoid-independent
mechanism
PPAR␥ activators also enhance COX-2 expression
Because PPAR activators enhance COX-2 expression and a large
number of NSAIDs have recently been identified as PPAR activators (4, 11), we studied the expression of PPARs in HASM cells.
HASM cells were seeded at a density of 2 ⫻ 104 cells/well in 24-well
plates and grown to 50 – 60% confluence. COX-2-Luc (3.9 kb) or 3.5-kb
COX-2-Luc (0.4 ␮g) and pRL-SV40 (4 ng) were cotransfected into the
cells. Transfection was conducted in serum-free and antibiotic-free medium for 5 h and the medium was then replaced with fresh serum-free
medium and the cells were treated with test drugs for a further 4 h. The
cells were then washed with PBS and lysed in lysis buffer (Promega). The
activities of firefly and Renilla luciferase in the cellular extracts were measured using the dual-luciferase reporter assay system according to the manufacturer’s instructions (Promega). Relative luciferase activity was obtained by normalizing the firefly luciferase activity against the internal
control Renilla luciferase activity.
Cell viability
The toxicity of all the chemicals used in this study and their vehicle DMSO
to HASM cells was determined by MTT assay in a separate series of
experiments in 96-well plates, as described previously (19). Viability was
set as 100% in control cells.
Statistical analysis
Results were expressed as the mean ⫾ SEM of n determinations from
HASM cells obtained from two donors. Student two-tailed t tests were used
to determine the significant differences between the means. Values of p ⬍
0.05 were accepted as statistically significant.
FIGURE 2. Effect of NSAIDs and PGE2 on COX-2 expression. HASM
cells were treated with or without Indo or NS-398 for 4 h (A), with or
without 10 ␮M of either Indo or NS-398 for the times indicated (B), or with
or without PGE2, or Flur for 4 h (C). Cell lysates were prepared and COX-2
expression was analyzed by Western blotting as described under Materials
and Methods. These blots are representative of similar results obtained
three times.
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
FIGURE 1. Effect of NSAIDs on IL-1␤-induced COX-2 expression. A,
HASM cells were pretreated with or without increasing concentrations of
Indo or NS-398 or 10 ␮M of Flur for 30 min before the incubation with or
without IL-1␤ for a further 4 h. B, HASM cells were pretreated with or
without 10 ␮M of either Indo or NS-398 for 30 min before the incubation
with or without IL-1␤ (1 ng/ml) for the times indicated. C, HASM cells
were pretreated with or without exogenous PGE2, Indo, or NS-398, alone
or in combination, for 30 min before the incubation with or without IL-1␤
for a further 4 h. Cell lysates were prepared and COX-2 expression was
analyzed by Western blotting as described under Materials and Methods.
These blots are representative of similar results obtained three times.
Among various NSAIDs we tested before, the nonselective COX
inhibitor Indo and the selective COX-2 inhibitor NS-398 block
IL-1␤-induced COX-2 activity in HASM cells (19). Thus, we focused our study on these two NSAIDs. As previously demonstrated (19), IL-1␤ treatment (1 ng/ml, 4 h) induced a strong
COX-2 protein expression in HASM cells. This was markedly enhanced by pretreatment with Indo (1–100 ␮M), NS-398 (1–100
␮M), and another nonselective COX inhibitor Flur (10 ␮M) (Fig.
1A). Indo and NS-398 at 10 ␮M also time-dependently enhanced
IL-1␤-induced COX-2 expression (Fig. 1B), but COX-1 protein
expression was not altered (data not shown). Because the concentrations of 1 and 10 ␮M of Indo and NS-398 are those required to
block IL-1␤-induced PGE2 synthesis in these cells (19), it is possible that the enhanced COX-2 expression is a consequence of
PGE2 synthesis inhibition. If that is the case, the addition of exogenous PGE2 would reverse the enhancing effect of both Indo and
NS-398, as PGE2 is the dominant prostanoid product by these cells
(19). However, contrary to this hypothesis, PGE2 pretreatment enhanced IL-1␤-induced COX-2 expression in a concentration-dependent manner and did not reverse the effect of NSAIDs (Fig.
1C), suggesting that the effect of NSAIDs on IL-1␤-induced
COX-2 expression is independent of their inhibition of the enzyme
activity. We then examined if these NSAIDs could cause COX-2
expression on their own and we found that treatment with Indo and
NS-398 (1 and 10 ␮M) for 4 h induced COX-2 expression compared with the control (Fig. 2A), and 10 ␮M of both Indo and
NS-398 also caused COX-2 expression in a time-dependent manner, which appeared at 1 h, peaked at ⬃2– 8 h after incubation, and
declined thereafter (Fig. 2B). Flur and exogenous PGE2 (10 ␮M,
4 h) also induced COX-2 expression as shown in Fig. 2C.
1046
NSAIDs ENHANCE COX-2 EXPRESSION
We found that PPAR␣ and PPAR␥, but not PPAR␤, were constitutively expressed in HASM cells and that treatment with NS-398,
the selective PPAR␣ activator WY-14643, or the selective PPAR␥
activator Cig for 4 and 24 h did not alter the expression (Fig. 3A).
Similar results were also obtained with other NSAIDs, the selective PPAR␥ activator 15d-PGJ2, and IL-1␤ (data not shown). We
then examined if PPAR activators could mimic the effect of
NSAIDs on COX-2 expression and found that Cig and 15d-PGJ2
(1 and 10 ␮M, 4 h), but not WY-14643, caused COX-2 expression
on their own (Fig. 3B) and that pretreatment with Cig and 15dPGJ2, but not WY-14643, resulted in a similar enhancing effect on
IL-1␤-induced COX-2 expression (Fig. 3C) as that of NSAIDs.
COX-2 expression induced by NSAIDs and PPAR activators is
accompanied by increase in PGE2 synthesis and COX activity
FIGURE 4. Effect of NSAIDs and PPAR activators on PGE2 release (A)
and COX activity (B and C) in the absence or presence of IL-1␤. HASM
cells were treated with or without Indo, NS-398, WY-14643, Cig, 15dPGJ2, or Flur (all 10 ␮M) for 30 min before the incubation with or without
IL-1␤ (1 ng/ml) for a further 4 h. PGE2 was measured by RIA and COX
activity was determined by measuring PGE2 production from exogenous
AA as described under Materials and Methods. Each point represents the
mean ⫾ SEM of nine determinations from three independent experiments.
ⴱ, p ⬍ 0.05; ⴱⴱ, p ⬍ 0.01; ⴱⴱⴱ, p ⬍ 0.001 compared with control. ⫹⫹, p ⬍
0.01; ⫹⫹⫹, p ⬍ 0.001 compared with IL-1␤ alone.
FIGURE 3. Effect of NS-398 and PPAR activators on PPAR expression
(A), COX-2 expression (B), and IL-1␤-induced COX-2 expression (C).
HASM cells were treated with or without NS-398, WY-14643, or Cig for
4 and 24 h (A), or with or without WY-14643, Cig, or 15d-PGJ2 for 4 h (B),
or pretreated with or without WY-14643, Cig, or 15d-PGJ2 for 30 min
before the incubation with or without IL-1␤ for a further 4 h (C). Cell
lysates were prepared and PPAR and COX-2 expression was analyzed by
Western blotting as described under Materials and Methods. These blots
are representative of similar results obtained three times.
caused COX-2 induction, PGE2 release was not increased due to
their inhibition of the induced enzyme. The effect of another PPAR
activator, 15d-PGJ2, on PGE2 release could not be interpreted, as
15d-PGJ2 had cross-reaction with the anti-PGE2 serum we used for
RIA. Treatment with IL-1␤ (1 ng/ml, 4 h) caused a marked increase in PGE2 release, which was significantly reduced by Indo
( p ⬍ 0.001) and NS-398 ( p ⬍ 0.01), increased by Cig ( p ⬍ 0.01),
and not affected by WY-14643 (Fig. 4A), suggesting that the enhanced COX-2 expression by Cig is biologically functional.
Again, as would be expected, due to the inhibition of the enzyme
by the two NSAIDs, no increase on PGE2 release from the cells
pretreated with Indo and NS-398 was observed despite the fact that
both enhanced IL-1␤-induced COX-2 expression (Fig. 1, A and B).
To assess the function of induced COX-2 more accurately, after
treatment with test drugs the cells were washed with PBS and
incubated with 15 ␮M of the COX substrate AA for a further 15
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
We have previously demonstrated that COX-2 expression by
IL-1␤ is associated with increase in PGE2 synthesis and COX
activity in HASM cells (19); thus, we tested if COX-2 expression
induced by NSAIDs and PPAR activators was biologically functional. After 4 h of incubation, HASM cells released low levels of
PGE2 in unstimulated conditions, which was significantly inhibited by Indo (10 ␮M, p ⬍ 0.01), not affected by NS-398 and
WY-14643, but enhanced by Cig (10 ␮M, p ⬍ 0.01) (Fig. 4A),
suggesting that the basal level PGE2 release is the result of COX-1
activity, whereas enhanced PGE2 release from Cig-treated cells is
the result of COX-2 expression. Although both Indo and NS-398
The Journal of Immunology
1047
Dex inhibits the induced COX-2 expression, PGE2 release, and
COX activity
FIGURE 5. Effect of Dex on COX-2 expression (A), PGE2 release (B),
and COX activity (C) by NS-398, Cig, and IL-1␤. HASM cells were pretreated with or without Dex for 30 min before the incubation with or without NS-398, Cig, or IL-1␤, alone or in combination, for a further 4 h. Cell
lysates were prepared and COX-2 expression was analyzed by Western
blotting, PGE2 was measured by RIA and COX activity was determined by
measuring PGE2 production from exogenous AA as described under Materials and Methods. The blots in A are representative of similar results
obtained three times. Each point in B and C represents the mean ⫾ SEM
of six determinations from two independent experiments. ⴱ, p ⬍ 0.05; ⴱⴱ,
p ⬍ 0.01; ⫹⫹, p ⬍ 0.01; ⫹⫹⫹, p ⬍ 0.001 compared with corresponding
treatment without Dex.
min in the absence of NSAIDs, PPAR activators, and IL-1␤. PGE2
release from AA was regarded as the index of COX activity. As
shown in Fig. 4B, Indo markedly reduced ( p ⬍ 0.001), whereas
NS-398, Cig, and 15d-PGJ2 significantly increased ( p ⬍ 0.01, p ⬍
0.01, and p ⬍ 0.05, respectively) COX activity compared with
control cells, WY-14643 had no effect. IL-1␤ also increased COX
activity, which was significantly reduced by Indo ( p ⬍ 0.001) but
not affected by NS-398, WY-14643, Cig, and 15d-PGJ2 (Fig. 4B).
Flur also increased COX activity on its own ( p ⬍ 0.05) but had no
effect on IL-1␤-induced COX activity (Fig. 4C). These results indicate that the COX-2 induced by NS-398, Flur, and the PPAR␥
activators alone is biologically functional and that even though
NSAIDs and PPAR␥ activators enhance IL-1␤-induced COX-2
expression, the enzyme activity in the context of this study is not
increased accordingly.
We have shown before that IL-1␤-induced COX-2 expression,
PGE2 release, and COX activity are inhibited by Dex (19), an
COX-2 transcription inhibitor. In this study, we further investigated if the effects of NSAIDs and PPAR activators on COX-2
could be inhibited by Dex. As shown in Fig. 5A, COX-2 expression (4 h) by NS-398 (10 ␮M), Cig (10 ␮M), and IL-1␤ (1 ng/ml),
individually or in combination, was strongly suppressed by Dex (1
␮M). Dex also significantly inhibited the enhanced PGE2 release
caused by IL-1␤, Cig, and their combination (Fig. 5B, p ⬍ 0.01,
p ⬍ 0.05, p ⬍ 0.001, respectively) and the enhanced COX activity
caused by IL-1␤, NS-398, Cig, and their combinations (Fig. 5C,
p ⬍ 0.01 for all). Similar results were also observed with other
Indo and Flur (data not shown).
NS-398 and Cig do not cause or enhance NF-␬B translocation,
but cause PPAR␥ translocation
As the transcriptional factor NF-␬B had been shown to be critically involved in COX-2 expression (29), we went on to examine
whether NF-␬B activation was involved in the COX-2 expression
by NS-398 and Cig. After treatment with either drug for 30 min, a
slight reduction of I␬B␣ in the cytosol was observed but NF-␬B
levels in both cytosol and nucleus remained unchanged (Fig. 6A).
In contrast, IL-1␤ caused a clear reduction of both I␬B␣ and
NF-␬B in the cytosol and a marked increase of NF-␬B in the
nucleus. However, cotreatment of either NS-398 or Cig with IL-1␤
did not alter the effect of IL-1␤ (Fig. 6A). The results suggest that
NS-398 and Cig have no direct effect on NF-␬B activation in
HASM cells and that COX-2 expression by these two drugs is
likely via a mechanism different from that of IL-1␤. A possible
mechanism would be the activation of PPARs, which then interact
with other transcriptional factors, including NF-␬B, to regulate
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
FIGURE 6. Effect of NS-398, Cig, and IL-1␤ on I␬B␣ degradation,
NF-␬B translocation (A), and PPAR translocation (B). HASM cells were
treated with or without NS-398, Cig, or IL-1␤, alone or in combination, for
30 min. Cytosolic and nuclear proteins were prepared and the expression of
I␬B␣, NF-␬B, PPAR␥, and PPAR␣ was analyzed as described under Materials and Methods. These blots are representative of similar results obtained two to three times.
1048
NSAIDs ENHANCE COX-2 EXPRESSION
Cell viability
Cell viability after treatment with the chemicals used in this study
was consistently ⬎95% compared with cells treated with the
vehicle.
Discussion
FIGURE 7. Effect of NSAIDs and PPAR activators on the transcriptional activation of COX-2 reporter constructs. A total of 50 – 60% confluent HASM cells were cotransfected with 4 ng/well pRL-SV40 internal
control vector and 0.4 ␮g/well of either 3.9-kb COX-2-Luc or 3.5-kb COX2-Luc using FuGene 6 transfection reagent as described under Materials
and Methods, and incubated with or without Indo, NS-398, Flur, 15d-PGJ2,
or WY-14643 (all 10 ␮M) in the absence or presence of IL-1␤ (1 ng/ml)
for a further 4 h. Luciferase activities were determined on cell extracts and
relative luciferase activity was calculated as described under Materials and
Methods. Each point represents the mean ⫾ SEM of 7–10 determinations
from two or three independent experiments. ⴱ, p ⬍ 0.05; ⴱⴱ, p ⬍ 0.01; ⴱⴱⴱ,
p ⬍ 0.01 compared with control. ⫹⫹, p ⬍ 0.01; ⫹⫹⫹, p ⬍ 0.001 compared with IL-1␤ alone.
COX-2 expression. If so, NS-398, like the PPAR␥ activator Cig,
should be able to translocate PPARs from cytosol to the nucleus.
We found indeed that treatment with NS-398 or Cig, but not IL1␤, for 30 min resulted in a significant loss of PPAR␥ in the cytosol and a marked increase in the nucleus (Fig. 6B), and that all
three reagents caused a slight increase of PPAR␣ in the nucleus
even though the loss of PPAR␣ in the cytosol was not significant
(Fig. 6B). Because PPAR␣ activator WY-14643 did not exert any
effect on COX-2, these data suggest that PPAR␥ activation may be
involved in the regulation of COX-2 expression by NSAIDs in
HASM cells.
NSAIDs and 15d-PGJ2 induce transcriptional activation of
COX-2 promoter via PPRE
The 5⬘-regulatory region of human COX-2 gene contains a distal
PPRE (4, 28) that interacts with PPARs. To further assess the role
The major findings of this study are that NSAIDs induce the expression of biologically functional COX-2 and enhance IL-1␤-induced COX-2 expression in a prostanoid-independent manner and
that the effect is mimicked by PPAR␥, but not PPAR␣, activators.
Unlike IL-1␤, NSAIDs and PPAR␥ activators do not activate NF␬B. The fact that NSAIDs stimulated the activity of a COX-2
promoter construct which was not seen when the PPRE was deleted provides the strongest evidence yet that NSAIDs induce
COX-2 via PPAR activation. This study is the first to implicate
PPARs in regulating any function of HASM cells.
NSAIDs can regulate COX-2 expression, but studies are conflicting. For instance, meclofenamate, mefenamic acid, ibuprofen,
NS-398, and sulindac acid enhance COX-2 expression in mammary epithelial cells (4); aspirin and sodium salicylate suppress
IL-1␤-induced COX-2 expression in endothelial cells (32); flufenamic acid induces COX-2 expression but inhibits mitogen-induced COX-2 expression in a colon cancer cell line and macrophages (6). These data suggest that COX-2 expression is regulated
differently by NSAIDs in different cell types. In this study, we
report that Indo, NS-398, and Flur all induce COX-2 expression
and enhance IL-1␤-induced COX-2 expression independently of
prostanoid synthesis inhibition. Our results differ from those of
Bonazzi et al. (26) in HASM cells, who demonstrate that Flur
inhibits IL-1␤-induced COX-2 expression as a result of PGE2 synthesis inhibition; however, they did not study the effect of NSAIDs
alone on COX-2 expression. The reasons for the discrepancy are
unclear. One possible explanation could be different experimental
and culture conditions. In their experiments, the cells were treated
at the same time with Flur and IL-1␤, whereas in ours the cells
were treated with NSAIDs for 30 min before IL-1␤ stimulation,
allowing NSAIDs to exert their effects before IL-1␤. This is supported by the fact that COX-2 expression by NSAIDs occurs faster
than that by IL-1␤ (Figs. 1B and 2B). Alternatively, Flur may have
a different effect on COX-2 expression in different cells since the
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
of PPAR activation in COX-2 expression by IL-1␤, NSAIDs, and
PPAR␥ activators, two COX-2 firefly luciferase reporter constructs
were used to transfect the cells. The 3.9-kb COX-2 construct contains the PPRE (⫺3721 to ⫺3707) in addition to the proximal
regulatory elements required for induction by cytokines (30, 31)
such as NF-␬B (⫺223 to ⫺214), NF-IL-6 (⫺132 to ⫺124), and
cAMP response element (CRE) (⫺59 to ⫺53), whereas the 3.5-kb
COX-2 construct lacks the PPRE but retains the proximal regulatory elements (4, 28). After cotransfection with 3.9-kb COX-2-Luc
and the internal control vector pRL-SV40, relative luciferase activity was significantly induced by 10 ␮M of Indo ( p ⬍ 0.01),
NS-398 ( p ⬍ 0.01), Flur ( p ⬍ 0.05), and 15d-PGJ2 ( p ⬍ 0.001)
alone, but not by WY-14643 (Fig. 7A); IL-1␤ (1 ng/ml) alone also
markedly increased COX-2 promoter activity and the effect was
further enhanced by Indo, NS-398, Flur ( p ⬍ 0.01), and 15d-PGJ2
( p ⬍ 0.001), but not by WY-14643 (Fig. 7A). In contrast, the
3.5-kb COX-2-Luc lacking the PPRE was unresponsive to these
NSAIDs and 15d-PGJ2, but was still fully induced by IL-1␤ (Fig.
7B). These results suggest that NSAIDs and PPAR␥ activators
induce COX-2 expression via transcriptional regulation of the
COX-2 gene and that it is the presence of PPRE in the the distal
⫺3.5-kb to ⫺3.9-kb region that confers responsiveness to these
NSAIDs and 15d-PGJ2.
The Journal of Immunology
Like other nuclear receptors, PPARs are translocated from the cytosol to the nucleus upon binding with activators such as 15d-PGJ2
(38). Indeed, we found that PPAR␥ activators mimicked the effect
of NSAIDs on COX-2 expression and that PPAR␥, but not NF-␬B,
was translocated to the nucleus by PPAR␥ activator Cig as well as
NS-398. Although NSAIDs are also PPAR␣ activators, our results
suggest that they mainly activate PPAR␥ in these cells, as PPAR␣
translocation results were less impressive, and PPAR␣ activation
is not involved in COX-2 expression, as PPAR␣ activator WY14643 did not have any effect. This is supported by the findings in
the current study that NSAIDs and PPAR␥ activator 15d-PGJ2
induced COX-2 promoter activity on their own and enhanced IL1␤-induced COX-2 promoter activity only when PPRE was included in the construct (3.9-kb COX-2-Luc). Collectively our data
demonstrate that NSAID-induced COX-2 expression is likely to be
mediated by PPAR␥, however, do not exclude possible interaction
between PPARs and NF-␬B. The fact that the more sustained and
enhanced increase of COX-2 expression was observed when
NSAIDs were used together with IL-1␤ (Fig. 1B) compared with
NSAIDs alone (Fig. 2B) suggests that this may result from the
transactivation of two transcription factors, PPAR␥ by NSAIDs
and NF-␬B by IL-1␤ (29), or an interaction between PPAR␥ and
NF-␬B (Fig. 8). Recent studies have indeed shown that PPARs
have cross-talk with transcription factors NF-␬B (6, 39) and AP-1
(39, 40) and CREB-binding protein/p300 (40).
We also showed that PGE2 induces COX-2 expression and enhances IL-1␤-induced COX-2 expression in HASM cells. Because
PGE2 stimulates cAMP generation via either EP2 or EP4 adenylyl
cyclase-coupled receptors (41) in HASM cells (26) and there is a
CRE in the COX-2 gene promoter (42), it is likely that PGE2
up-regulates COX-2 expression by a cAMP-dependent positive
feedback loop (Fig. 8). This is supported by the findings that the
␤2-adrenoceptor agonists salbutamol and salmeterol and the direct
FIGURE 8. Schematic of proposed mechanism by which NSAIDs, COX products, and IL-1␤ regulate COX-2 expression in HASM cells. COX-2
expression is up-regulated by 1) NSAIDs, binding PPAR␥, which forms a heterodimer with retinoid X receptor (RXR) and binds to PPRE of COX-2
promoter; 2) IL-1␤, activating NF-␬B; 3) PGE2, stimulating cAMP and binding to CRE. COX-2 expression leads to the generation of prostanoids, of which
PGE2 and 15d-PGJ2, in an autocrine manner, enhance COX-2 expression by a cAMP- and a PPAR␥-dependent mechanism, respectively.
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
same authors also report that Flur, like Indo, significantly increases
COX-2 expression in corneal epithelial cells (33), and different
culture conditions may have an effect on HASM cell PPAR expression and consequently their response to NSAIDs.
The mechanisms of COX-2 regulation by NSAIDs vary between
NSAIDs and cell type. Because the main effect of NSAIDs is believed to be COX inhibition, it is reasonable to speculate that
COX-2 regulation by NSAIDs is via their inhibition on COX activity, whose end products exert a feedback on COX-2 expression.
For instance, Indo up-regulates endotoxin-induced COX-2 expression by removing the negative feedback of prostanoids in J774
macrophages (34). Paradoxically, we have found in this study that
PGE2, the main prostanoid produced by HASM cells, exerts a
positive feedback effect on COX-2 expression, which is in agreement with the report of Bonazzi et al. (33) in rabbit corneal epithelial cells. Meade et al. (4) also demonstrate that NSAIDs upregulate COX-2 expression independently of prostanoid inhibition
in mammary epithelial cells. In these cases, the increased COX-2
expression by NSAIDs is clearly not a feedback mechanism because 1) it is not reversed by exogenous PGE2; 2) COX-1 expression is not altered (4); 3) NSAIDs do not up-regulate COX-2 expression equally, even though they all block prostanoid generation
(4). Therefore, NSAID-induced COX-2 expression in this study is
a direct effect mediated by a prostanoid-independent mechanism.
Recent work has shown that NSAIDs, fatty acids, and prostanoids, compounds that are inhibitors, substrates, and products, respectively, of COX activity, regulate gene expression via PPAR
activation (6, 10, 35, 36). COX-2 expression is also induced by
PPAR activators (4, 33, 37), and it has been speculated that this
occurs via a PPRE in the COX-2 promoter region (4), although
studies deleting the PPRE and observing the effect of NSAIDs
have not been previously performed. Our study is the first to show
that PPAR␣ and PPAR␥, but not PPAR␤, are expressed constitutively in HASM cells, providing the basis for PPAR activation.
1049
1050
Acknowledgments
We thank Colin Clelland for providing us with specimens of human trachea. We also thank Prof. Thomas McIntyre for kindly providing us with
the COX-2 constructs.
References
1. DuBois, R. N., S. B. Abramson, L. Crofford, R. A. Cupta, L. S. Simon,
L. B. A. van de Putte, and P. E. Lipsky. 1998. Cyclooxygenase in biology and
disease. FASEB J. 12:1063.
2. Kopp, E., and S. Ghosh. 1994. Inhibition of NF-␬B by sodium salicylate and
aspirin. Science 265:956.
3. Mitchell, J. A., M. Saunders, P. J. Barnes, R. Newton, and M. Belvisi. 1997.
Sodium salicylate inhibits cyclo-oxygenase-2 activity independently of transcription factor (nuclear factor ␬B) activation: role of arachidonic acid. Mol. Pharmacol. 51:907.
4. Meade, E. A., T. M. McIntyre, G. A. Zimmerman, and S. M. Prescott. 1999.
Peroxisome proliferators enhance cyclooxygenase-2 expression in epithelial
cells. J. Biol. Chem. 274:8328.
5. Staels, B., W. Koening, A. Habib, R. Merval, M. Lebret, I. P. Torra, P. Delerive,
A. Fadel, G. Chinetti, J. C. Fruchart, et al. 1998. Activation of human aortic
smooth-muscle cells is inhibited by PPAR␣ but not by PPAR␥ activators. Nature
393:790.
6. Paik, J. H., J. H. Ju, J. Y. Lee, M. D. Boudreau, and D. H. Hwang. 2000. Two
opposing effects of non-steroidal anti-inflammatory drugs on the expression of
the inducible cyclooxygenase mediation through different signaling pathways.
J. Biol. Chem. 275:28173.
7. Inoue, H., T. Tanabe, and K. Umesono. 2000. Feedback control of cyclooxygenase-2 expression through PPAR␥. J. Biol. Chem. 275:28028.
8. Corton, J. C., S. P. Anderson, and A. Stauber. 2000. Central role of peroxisome
proliferator-activated receptors in the actions of peroxisome proliferators. Annu.
Rev. Pharmacol. Toxicol. 40:491.
9. Ricote, M., A. C. Li, T. M. Willson, C. J. Kelly, and C. K. Glass. 1998. The
peroxisome proliferator-activated receptor-␥ is a negative regulator of macrophage activation. Nature 391:79.
10. Forman, B. M., J. Chen, and R. M. Evans. 1997. Hypolipidemic drugs, polyunsaturated fatty acids, and eicosanoids are ligands for peroxisome proliferatoractivated receptors ␣ and ␦. Proc. Natl. Acad. Sci. USA 94:4312.
11. Devchand, P. R., H. Keller, J. M. Peters, M. Vazquez, F. J. Gonzalez, and
W. Wahli. 1996. The PPAR␣-leukotriene B4 pathway to inflammation control.
Nature 384:39.
12. Jiang, C., A. T. Ting, and B. Seed. 1998. PPAR-␥ agonists inhibit production of
monocyte inflammatory cytokines. Nature 391:82.
13. Marx, N., G. Sukhova, C. Murphy, P. Libby, and J. Plutzky. 1998. Macrophages
in human atheroma contain PPAR␥: differentiation-dependent peroxisomal proliferator-activated receptor ␥ (PPAR␥) expression and reduction of MMP-9 activity through PPAR␥ activation in mononuclear phagocytes in vitro.
Am. J. Pathol. 153:17.
14. Elias, J. A., Y. Wu, T. Zheng, and R. Panettieri. 1997. Cytokine- and virusstimulated airway smooth muscle cells produce IL-11 and other IL-6-type cytokines. Am. J. Physiol. 273:L648.
15. Pang, L., and A. J. Knox. 1998. Bradykinin stimulates IL-8 production in cultured
human airway smooth muscle cells: role of cyclooxygenase products. J. Immunol.
161:2509.
16. Pang, L., and A. J. Knox. 2001. Regulation of TNF-␣-induced eotaxin release
from cultured human airway smooth muscle cells by ␤2-agonists and corticosteroids. FASEB J. 15:261.
17. McKay, S., J. C. de Jongste, P. R. Saxena, and H. S. Sharma. 1998. Angiotensin
II induces hypertrophy of human airway smooth muscle cells: expression of
transcription factors and transforming growth factor-␤1. Am. J. Respir. Cell Mol.
Biol. 18:823.
18. Knox, A. J., L. Corbett, J. Stocks, E. Holland, Y. M. Zhu, and L. Pang. 2001.
Human airway smooth muscle cells secrete vascular endothelial growth factor:
up-regulation by bradykinin via a protein kinase C and prostanoid-dependent
mechanism. FASEB J. 15:2480.
19. Pang, L., and A. J. Knox. 1997. Effect of interleukin-1␤, tumour necrosis factor-␣
and interferon-␥ on the induction of cyclo-oxygenase-2 in cultured human airway
smooth muscle cells. Br. J. Pharmacol. 121:579.
20. Pang, L., and A. J. Knox. 1997. PGE2 release by bradykinin in human airway
smooth muscle cells: involvement of cyclooxygenase-2 induction. Am. J. Physiol.
273:L1132.
21. Belvisi, M. G., M. A. Saunders, E. B. Haddad, S. J. Hirst, M. H. Yacoub,
P. J. Barnes, and J. A. Mitchell. 1997. Induction of cyclo-oxygenase-2 by cytokines in human cultured airway smooth muscle cells: novel inflammatory role of
this cell type. Br. J. Pharmacol. 120:910.
22. Belvisi, M. G., M. Saunders, M. Yacoub, and J. A. Mitchell. 1998. Expression of
cyclo-oxygenase-2 in human airway smooth muscle is associated with profound
reductions in cell growth. Br. J. Pharmacol. 125:1102.
23. Pang, L., E. Holland, and A. J. Knox. 1998. Role of cyclo-oxygenase-2 induction
in interleukin-1␤ induced attenuation of cultured human airway smooth muscle
cell cyclic AMP generation in response to isoprenaline. Br. J. Pharmacol. 125:
1320.
24. Pang, L., E. Holland, and A. J. Knox. 1998. Impaired cAMP production in human
airway smooth muscle cells by bradykinin: role of cyclooxygenase products.
Am. J. Physiol. 275:L322.
25. Pang, L., A. Pitt, D. Petkova, and A. J. Knox. 1998. The COX-1/COX-2 balance
in asthma. Clin. Exp. Allergy 28:1050.
26. Bonazzi, A., M. Bolla, C. Buccellati, A. Hernandez, S. Zarini, T. Vigano,
F. Fumagalli, S. Viappiani, S. Ravasi, P. Zannini, et al. 2000. Effect of endogenous and exogenous prostaglandin E2 on interleukin-1␤-induced cyclooxygenase-2 expression in human airway smooth-muscle cells. Am. J. Respir. Crit. Care
Med. 162:2272.
27. Eickelberg, O., M. Roth, R. Lörx, V. Bruce, J. Rüdiger, M. Johnson, and L.-H.
Block. 1999. Ligand-independent activation of the glucocorticoid receptor by
␤2-adrenergic receptor agonists in primary human lung fibroblasts and vascular
smooth muscle cells. J. Biol. Chem. 274:1005.
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
adenylyl cyclase activator forskolin also induce COX-2 (our unpublished observation) and that PGE2 up-regulates COX-2 gene
expression (43, 44) and potentiates mitogen-induced COX-2 expression (9, 45, 46). Because PGE2 up-regulates COX-2 expression, NSAID inhibition of IL-1␤-induced PGE2 production would,
in theory, reduce COX-2 expression; however, because the effect
is compensated by NSAID activation of PPAR␥, COX-2 expression is, in fact, enhanced (Fig. 8). Moreover, 15d-PGJ2, the natural
ligand of PPAR␥, is a metabolite of PGD2 and we have shown that
IL-1␤ stimulates PGD2 synthesis in a COX-2-dependent manner in
HASM cells (19), it is possible that COX-2 expression is selfregulated in an autocrine manner by PGE2 and 15d-PGJ2 via a
cAMP- and PPAR␥-dependent mechanism, respectively (Fig. 8).
We also demonstrated that the PPAR␥ activator Cig significantly increases PGE2 accumulation either alone or in combination
with IL-1␤, in line with the increase in COX-2 expression. No
PGE2 increase could be detected with NSAIDs because its synthesis was blocked. Because one possibility that could account for
the low PGE2 levels seen in other studies (33, 37) is the lack of
substrate AA and because NSAIDs abolish PGE2 generation from
endogenous AA, the PGE2 levels could not reflect the increase in
COX-2 enzymatic activity under these conditions. To measure
COX activity more accurately, we examined COX activity by measuring the PGE2 generation from exogenous AA in the absence of
NSAIDs. Under these conditions, COX activity was significantly
increased in 15d-PGJ2-, Cig-, NS-398-, and Flur-treated cells compared with control cells. As far as we are aware, this is the first
evidence that COX-2 induced by NSAIDs is biologically functional when the drugs are no longer present. The reason that Indo
did not increase COX activity even though they did induce COX-2
expression can be explained by the fact that its inhibition on
COX-2 is time-dependent and irreversible (47, 48). However, despite the fact that both NS-398 and Flur are also time-dependent,
irreversible inhibitors of COX-2 (47, 49), our results clearly show
that their effect can be washed away, but the reason remains unknown. Increased COX activity was also observed with IL-1␤;
however, addition of NSAIDs and PPAR activators did not further
increase COX activity even though they did enhance COX-2 expression. This is probably because IL-1␤ is a stronger inducer of
COX-2 than NSAIDs and PPAR activators and the COX activity
has reached its maximum. Alternatively, NSAIDs and PPAR activators might indirectly inhibit cytokine-induced COX activity by
modulating other enzyme and signal transduction pathways that
regulate prostanoid metabolism (37).
In summary, we have demonstrated that NSAIDs induce COX-2
expression in HASM cells through PPAR␥ activation. The fact that
NSAIDs are not effective in treating airway inflammation in
asthma suggests that the inflammatory process is not mainly mediated by prostanoids and/or that NSAIDs may intensify the inflammatory response by activating PPARs and subsequently amplifying the response via COX-2 expression and proinflammatory
prostanoid production. Our observations may explain in part the
lack of efficacy of NSAIDs in the treatment of airway
inflammation.
NSAIDs ENHANCE COX-2 EXPRESSION
The Journal of Immunology
38. Bishop-Bailey, D., and T. Hla. 1999. Endothelial cell apoptosis induced by the
peroxisome proliferator-activated receptor (PPAR) ligand 15-deoxy-⌬12,14-prostaglandin J2. J. Biol. Chem. 274:17042.
39. Delerive, P., K. De Bosscher, S. Besnard, W. Vanden Berghe, J. M. Peters,
F. J. Gonzalez, J.-C. Fruchart, A. Tedgui, G. Haegeman, and B. Staels. 1999.
Peroxisome proliferator-activated receptor ␣ negatively regulates the vascular
inflammatory gene response by negative cross-talk with transcription factors
NF-␬B and AP-1. J. Biol. Chem. 274:32048.
40. Subbaramaiah, K., D. T. Lin, J. C. Hart, and A. J. Dannenberg. 2001. Peroxisome
proliferator-activated receptor ␥ ligands suppress the transcriptional activation of
cyclooxygenase-2: evidence for involvement of activator protein-1 and CREBbinding protein/p300. J. Biol. Chem. 276:12440.
41. Coleman, R. A., W. L. Smith, and S. Narumiya. 1994. International Union of
Pharmacology classification of prostanoid receptors: properties, distribution, and
structure of the receptors and their subtypes. Pharmacol. Rev. 46:205.
42. Inoue, H., T. Nanayama, S. Hara, C. Yokoyama, and T. Tanabe. 1994. The cyclic
AMP response element plays an essential role in the expression of the human
prostaglandin-endoperoxide synthase 2 gene in differentiated U937 monocytic
cells. FEBS Lett. 350:51.
43. Maldve, R. E., Y. Kim, S. J. Muga, and S. M. Fischer. 2000. Prostaglandin E2
regulation of cyclooxygenase expression in keratinocytes is mediated via cyclic
nucleotide-linked prostaglandin receptors. J. Lipid Res. 41:873.
44. Lo, C. J., M. Fu, F. R. Lo, and H. G. Cryer. 2000. Cyclooxygenase 2 (COX-2)
gene activation is regulated by cyclic adenosine monophosphate. Shock 13:41.
45. Hinz, B., K. Brune, and A. Pahl. 2000. Prostaglandin E2 upregulates cyclooxygenase-2 expression in lipopolysaccharide-stimulated RAW 264.7 macrophages.
Biochem. Biophys. Res. Commun. 272:744.
46. Hinz, B., K. Brune, and A. Pahl. 2000. Cyclooxygenase-2 expression in lipopolysaccharide-stimulated human monocytes is modulated by cyclic AMP, prostaglandin E2, and nonsteroidal anti-inflammatory drugs. Biochem. Biophys. Res.
Commun. 278:790.
47. Quellet, M., and M. D. Percival. 1995. Effect of inhibitor time-dependency on
selectivity towards cyclooxygenase isoforms. Biochem. J. 306:247.
48. Rosenstock, M., A. Danon, M. Rubin, and G. Rimon. 2001. Prostaglandin H
synthase-2 inhibitors interfere with prostaglandin H synthase-1 inhibition by nonsteroidal anti-inflammatory drugs. Eur. J. Pharmacol. 412:101.
49. Copeland, R. A., J. M. Williams, J. Giannaras, S. Nurnberg, M. Covinton,
D. Pinto, S. Pick, and J. M. Trzaskos. 1994. Mechanism of selective inhibition of
the inducible isoform of prostaglandin G/H synthase. Proc. Natl. Acad. Sci. USA
91:11202.
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
28. Pontsler, A. V., A. S. Hilaire, G. K. Marathe, G. A. Zimmerman, and
T. M. McIntyre. 2002. Cyclooxygenase-2 is induced in monocytes by peroxisome
proliferator activated receptor ␥ and oxidized alkyl phospholipids from oxidized
low density lipoprotein. J. Biol. Chem. 277:13029.
29. Newton, R., L. M. Kuitert, M. Bergmann, I. M. Adcock, and P. J. Barnes. 1997.
Evidence for involvement of NF-␬B in the transcriptional control of COX-2 gene
expression by IL-1␤. Biochem. Biophys. Res. Commun. 237:28.
30. Inoue, H., C. Yokoyama, S. Hara, Y. Tone, and T. Tanabe. 1995. Transcriptional
regulation of human prostaglandin-endoperoxide synthase-2 gene by lipopolysaccharide and phorbol ester in vascular endothelial cells: involvement of both
nuclear factor for interleukin-6 expression site and cAMP response element.
J. Biol. Chem. 270:24965.
31. Kim, Y., and S. M. Fischer. 1998. Transcriptional regulation of cyclooxygenase-2
in mouse skin carcinoma cells: regulatory role of CCAAT/enhancer-binding proteins in the differential expression of cyclooxygenase-2 in normal and neoplastic
tissues. J. Biol. Chem. 273:27686.
32. Xu, X. M., L. Sansores-Garcia, X. M. Chen, N. Matijevic-Aleksic, M. Du, and
K. K. Wu. 1999. Suppression of inducible cyclooxygenase 2 gene transcription
by aspirin and sodium salicylate. Proc. Natl. Acad. Sci. USA 96:5292.
33. Bonazzi, A., V. Mastyugin, P. A. Mieyal, M. W. Dunn, and
M. Laniado-Scheartzman. 2000. Regulation of cyclooxygenase-2 by hypoxia and
peroxisome proliferators in the corneal epithelium. J. Biol. Chem. 275:2837.
34. Pang, L., and J. R. S. Hoult. 1996. Induction of cyclooxygenase and nitric oxide
synthase in endotoxin-activated J774 macrophages is differentially regulated by
indomethacin: enhanced cyclooxygenase-2 protein expression but reduction of
inducible nitric oxide synthase. Eur. J. Pharmacol. 317:151.
35. Kliewer, S. A., S. S. Sundseth, S. A. Jones, P. J. Brown, G. B. Wisely,
C. S. Koble, P. Devchand, W. Wahli, T. M. Willson, J. M. Lenhard, and
J. M. Lehmann. 1997. Fatty acids and eicosanoids regulate gene expression
through direct interactions with peroxisome proliferator-activated receptors ␣ and
␥. Proc. Natl. Acad. Sci. USA 94:4318.
36. Yu, K., W. Bayona, C. B. Kallen, H. P. Harding, C. P. Ravera, G. McMahon,
M. Brown, and M. A. Lazar. 1995. Differential activation of peroxisome proliferator-activated receptors by eicosanoids. J. Biol. Chem. 270:23975.
37. Ledwith, B. J., C. J. Pauley, L. K. Wagner, C. L. Rokos, D. W. Alberts, and
S. Manam. 1997. Induction of cyclooxygenase-2 expression by peroxisome proliferators and non-tetradecanoylphorbol 12,13-myristate-type tumor promoters in
immortalized mouse liver cells. J. Biol. Chem. 272:3707.
1051