Modulation of IgE-dependent COX-2 gene expression by reactive oxygen species in human neutrophils

Modulation of IgE-dependent COX-2 gene expression byreactive oxygen species in human neutrophils

Antonio Vega,*,† Pedro Chacon,*,† Gonzalo Alba,* Rajaa El Bekay,* Javier Monteseirın,*,‡

Jose Martın-Nieto,§ and Francisco Sobrino*,1

*Departamento de Bioquımica Medica y Biologıa Molecular, Universidad de Sevilla, Sevilla, Spain; †Servicio deInmunologıa y Alergia, Hospital Universitario Virgen Macarena, Sevilla, Spain; ‡Clınica Sagrado Corazon, Sevilla,Spain; and §Departamento de Fisiologıa, Genetica y Microbiologıa, Universidad de Alicante, Alicante, Spain

Abstract: Cyclooxygenase (COX) is a key enzymein prostaglandin (PG) synthesis. Up-regulation of itsCOX-2 isoform is responsible for the increased PGrelease, taking place under inflammatory conditions,and also, is thought to be involved in allergic andinflammatory diseases. In the present work, we dem-onstrate that COX-2 expression becomes highly in-duced by anti-immunoglobulin E (IgE) antibodies andby antigens in human neutrophils from allergic pa-tients. This induction was detected at mRNA andprotein levels and was accompanied by a concom-itant PGE2 and thromboxane A2 release. We alsoshow evidence that inhibitors of reduced nicotin-amide adenine dinucleotide phosphate (NADPH)oxidase, such as 4-(2-aminoethyl)benzenesulphonylfluoride and 4-hydroxy-3-methoxyaceto-phenone,completely cancelled anti-IgE-induced COX-2 pro-tein up-regulation, suggesting that this process is me-diated by reactive oxygen species (ROS) derivedfrom NADPH oxidase activity. Moreover, the mito-gen-activated protein kinases (MAPKs), p38 and ex-tracellular signal-regulated kinase, and also, the tran-scription factor, nuclear factor (NF)-�B, are in-volved in the up-regulation of COX-2 expression, asspecific chemical inhibitors of these two kinases, suchas SB203580 and PD098059, and of the NF-�Bpathway, such as N(�)-benzyloxycarbonyl-l-leucyl-l-leucyl-l-leucinal, abolished IgE-dependent COX-2 in-duction. Evidence is also presented, using Fe2�/Cu2�

ions, that hydroxyl radicals generated from hydrogenperoxide through Fenton reactions could constitutecandidate modulators able to directly trigger anti-IgE-elicited COX-2 expression through MAPK andNF-�B pathways. Present results underscore a newrole for ROS as second messengers in the modulationof COX-2 expression by human neutrophils in aller-gic conditions. J. Leukoc. Biol. 80: 152–163; 2006.

Key Words: prostaglandins � allergy � signal transduction � ROS

INTRODUCTION

Prostaglandins (PGs) play important roles in many biologicalprocesses, including homeostatic and immune responses [1],

and an overproduction of arachidonic acid (AA) metabolites isassociated with chronic inflammation and allergic diseases [2,3]. PGH2 synthase [cyclooxygenase (COX)] is the rate-limitingenzyme in the conversion of AA into eicosanoids, i.e., PGs andthromboxanes (TXs) [1]. COX exists in two isoforms, COX-1and COX-2, of which COX-1 is constitutively expressed inmost tissues and appears responsible for the production of PGsunder normal, physiological conditions [2, 3], and COX-2expression is associated with inflammation [2, 3] and becomesinduced by various proinflammatory stimuli, including mito-gens, bacterial lipopolysaccharide (LPS), and cytokines [2–4].

Reactive oxygen species (ROS) are highly diffusible, ubiq-uitous ions and radicals generated from the reduction of mo-lecular oxygen. They include species such as superoxide an-ions (O2

��), hydrogen peroxide (H2O2), and hydroxyl radicals(·OH). ROS are produced during the respiratory burst of neu-trophils as a normal defense mechanism against pathogens.They modulate, however, multiple cellular functions such ascell growth and differentiation, proliferation, apoptosis, andgene expression, acting on transductional and transcriptionalregulatory pathways [5]. The most important source contribut-ing to ROS generation in neutrophils is the reduced nicotin-amide adenine dinucleotide phosphate (NADPH) oxidase sys-tem [6], which is composed of various subunits distributedbetween the plasma membrane (gp91phox and p22phox) and thecytosol (p47phox and p67phox) in resting phagocytic cells. Whenthese become activated, the NADPH oxidase cytosolic sub-units associate with membrane-bound, inactive componentsand assemble into a catalytically active enzyme [6].

The molecular mechanisms whereby COX-2 expression be-comes induced remain unclear. Recent studies have high-lighted a connection between up-regulation of COX-2 expres-sion and the generation of ROS during monocyte-to-macro-phage differentiation and the activation of mitogen-activatedprotein kinase (MAPK)-dependent pathways [7]. The interplaybetween MAPK-dependent pathways, increase of ROS levels,and activation of the transcription factor nuclear factor

1 Correspondence: Dpto. Bioquımica Medica y Biologıa Molecular, Facultadde Medicina, Universidad de Sevilla, E-41009 Sevilla, Spain. E-mail:[email protected]

Received July 25, 2005; revised February 22, 2006; accepted February 24,2006;

doi: 10.1189/jlb.0705411.

152 Journal of Leukocyte Biology Volume 80, July 2006 0741-5400/06/0080-152 © Society for Leukocyte Biology

(NF)-�B is largely known in neutrophils [8, 9]. In this context,COX-2 expression at the mRNA level is regulated positivelythrough the binding of NF-�B and other transcription factors,such as the nuclear factor of activated T cells, the CCAATenhancer-binding protein, or the cyclic adenosine monophos-phate-responsive element-binding protein, to cis-acting, regu-latory elements present in the COX-2 promoter [10–12].

The capacity of human neutrophils to induce COX-2 expres-sion under inflammatory conditions has been shown previously[4], and there is ample evidence of the participation of this celltype in allergic processes [13–15]. In this regard, we haveshown previously that a number of antigens are able to specif-ically activate the respiratory burst in neutrophils isolated fromallergic patients sensitized to the same such antigens [16].With this background, the present work was undertaken toassess whether in human neutrophils from allergic patients,COX-2 expression becomes induced upon their challenge withspecific antigens or with anti-immunoglobulin E (IgE) antibod-ies and also to analyze the potential involvement of ROS andNADPH oxidase in this process. We here provide evidencethat ROS-dependent signaling through the MAPK and NF-��pathways, driven by NADPH oxidase, represents a mechanismpreviously unrecognized for the regulation of COX-2 expres-sion upon neutrophil challenge with antigens or anti-IgE mol-ecules. This finding highlights the possibility that unquenchedROS could actively contribute to the development of allergicinflammation by acting as second messengers in the modula-tion of PGE2 and TXA2 synthesis through COX-2 up-regulationby human neutrophils.

MATERIALS AND METHODS

Reagents and antibodies

Antigens were commercially available extracts, including D1 (from Dermato-phagoides pteronyssinus), G3 (from Dactylis glomerata), T9 (from Olea euro-paea), E2 (from dog epithelium), and M6 (from Alternaria alternata), purchasedfrom Bial-Arıstegui (Bilbao, Spain). Goat anti-human IgE antibody, goat anti-human IgG antibody, and the Perm-cell kit were from Caltag Laboratories(Burlingame, CA). H2O2, Escherichia coli LPS, FeSO4, CuSO4, phorbol 12-myristate-13-acetate (PMA), goat IgG, N(�)-benzyloxycarbonyl-l-leucyl-l-leucyl-l-leucinal (MG-132), 4-hydroxy-3-methoxyaceto-phenone (HMAP),4-(2-aminoethyl)benzenesulfonyl fluoride (AEBSF), 2-amino-1,2,4-triazole,and 1,10-phenanthroline (PHE) were from Sigma-Aldrich (Madrid, Spain).PD098059 and SB203580 were products of Calbiochem (Madrid, Spain).[�-32P]Adenosine 5�-triphosphate (ATP; specific activity, 3000 Ci/mmol) andpolymerase chain reaction (PCR) primers were from Amersham-Pharmacia-Biotech (Barcelona, Spain). Random primers were obtained from Roche (Ma-drid, Spain). The double-stranded oligonucleotide 5�-AGTGAGGGGACTTTC-CCAGGC-3�, containing a consensus NF-�B site (underlined), and Moloneymurine leukemia virus (M-MLV) reverse transcriptase (RT) were obtained fromPromega (Madison, WI). Unconjugated mouse monoclonal antibodies (mAb) tohuman COX-2 were obtained from Cayman Chemical (Ann Arbor, MI). MousemAb against human �-actin (sc-8432), rabbit polyclonal antibody againsthuman inhibitor of �B (I-�B; � isoform; sc-371), mouse antiphosphospecificJun N-terminal kinase (JNK)1/2 (Thr183/Tyr185) antibody (sc-6254), and anti-bodies recognizing total (unphosphorylated plus phosphorylated) JNK1/2 (sc-7345) were purchased from Santa Cruz Biotechnology (CA). The rabbit poly-clonal antiphosphospecific p38MAPK (Thr180/Tyr182), mouse monoclonal an-tiphosphospecific extracellular signal-regulated kinase (ERK)1/2 (Thr202/Tyr204), and antibodies recognizing the total forms of these two MAPKs werefrom New England Biolabs (Beverly, MA). Mouse mAb against CD9 andCD203 were from Immunotech-IZASA (Barcelona, Spain). Goat anti-mouse

IgG-coated micromagnetic beads were from Miltenyi Biotec (Bergisch-Glad-bach, Germany). Antisera against p47phox and p67phox were kindly donated byThomas L. Leto (National Institutes of Health, Bethesda, MD). All culturereagents had endotoxin levels of �0.01 ng/ml, as tested by the Coatest Limuluslysate assay (Chromogenix, Molndal, Sweden).

Patients and controls

The groups studied included adult, atopic patients with bronchial asthma andhealthy, nonatopic volunteer controls. The asthmatic patients had given posi-tive results on skin-prick (Bial-Arıstegui) and specific IgE tests (HYTEC 288,Hycor Biomedical-IZASA, Barcelona, Spain) to at least one common allergenicantigen. These subjects had not received any treatment or specific hyposen-sitization and had not experienced episodes of asthma for at least 3 months orrespiratory-tract infections in the 4 weeks before blood sampling. The healthygroup had no history of allergy or bronchial symptoms and was negative for theskin-prick and specific IgE tests toward a battery of inhalant allergenicantigens (house dust mites, pollens, molds, and animal danders). The Univer-sidad de Sevilla Ethics Committee (Spain) approved this study previously, andeach subject had given prior informed consent.

Neutrophil isolation and cell culture conditions

Human peripheral neutrophils were isolated as described previously [17]. Forfurther purification, neutrophil preparations were incubated with mouse anti-human CD9 and CD203 antibodies and goat anti-mouse IgG-coated micro-magnetic beads. The purity of neutrophils was on average �99% (0.1%eosinophil and close to 0% basophil contamination) and was used immediatelyafter isolation. Neutrophils were cultured in RPMI-1640 medium supple-mented with 10% fetal bovine serum, 2 mM L-glutamine, 100 U/ml penicillin,and 100 g/ml streptomycin and maintained at 37°C in an atmosphere of 5%CO2 and 95% O2. None of the reagents used in this work significantly affectedthe viability of the cells at the concentrations used, as confirmed by the trypanblue dye exclusion test. Furthermore, after incubation with anti-IgE antibodiesfor 24 h, neutrophil viability was found to be �98% by measuring lactatedehydrogenase activity [18].

RT-PCR analysis of COX-2 mRNA levels

Total RNA from cultured neutrophils (107 cells) was isolated using the gua-nidine-phenol method [19], and 1–2 g RNA was reverse-transcribed intocDNA using M-MLV RT and random primers. cDNA was amplified by PCRusing the following specific primers for COX-2 (GenBank Accession NumberM90100) or glyceraldehyde 3-phosphaste dehydrogenase (GAPDH; GenBankAccession Number J04038) as a house-keeping gene control: 5�-TTCAA-ATGAGATTGTGGGAAAATTGCT-3� (COX-2 sense), 5�-AGATCATCTCT-GCCTGAGTATCTT-3� (COX-2 antisense), 5�-CCACCCATGGCAAATTC-CATGGCA-3� (GAPDH sense), 5�-TCTAGACGGCAGGTCAGGTCCACC-3�(GAPDH antisense). The reaction was performed by 35 cycles each of 94°C for1 min, 60°C for 1 min, and 72°C for 1 min. Amplified PCR products were of300 pb for COX-2 gene and of 605 pb for GAPDH gene.

Western blot analysis of COX-2protein expression

Neutrophils (107 cells) were washed with phosphate-buffered saline (PBS) andlysed by incubation for 30 min in a buffer composed of 20 mM Hepes, pH 7.9,5 mM KCl, 0.1% Nonidet P-40 (NP-40), 1 mM EDTA, 1 mM dithiothreitol(DTT), and 10 mM NaF, supplemented with the following protease-inhibitormixture: 10 g/ml aprotinin, 10 g/ml leupeptin, 10 g/ml soybean trypsininhibitor, 10 g/ml N-tosyl-L-phenylalanyl-chloromethyl ketone, 10 g/mlcaptopril, 1 mM phenylmethylsulfonyl fluoride (PMSF), 1 mM benzamidine,and 10 mM iodoacetamide. After centrifugation at 14,000 g, proteins (80 g)were separated on 7.5% sodium dodecyl sulfate-polyacrylamide gel electro-phoresis (SDS-PAGE) gels and electrophoretically transferred to polyvinyli-dene difluoride (PVDF) membranes as described [20]. These were probed for1 h without need of prior blocking with specific antibodies against humanCOX-2 or �-actin (1:2000 dilution) as a control of even protein loading.Primary antibody binding was detected by using horseradish peroxidase(HRP)-conjugated anti-rabbit or anti-mouse IgG antibodies (1:5000 dilution)followed by enhanced chemiluminescence (ECL) [21].

Vega et al. ROS-mediated IgE-dependent COX-2 up-regulation 153

p47phox and p67phox membrane translocation

Membrane and cytosolic fractions were extracted from neutrophils as describedpreviously [22]. Proteins were subjected to SDS-PAGE, electroblotted ontoPVDF membranes as previouly described [20], and incubated for 1 h with goatanti-p47phox or anti-p67phox IgG antibodies as above. Antibody binding wasdetected by incubation with HRP-conjugated anti-goat IgG antibodies (1:5000dilution) followed by ECL. Protein levels (membrane translocation and cytosoldisappearance) were determined by scanning densitometry analysis using theScion Image software (Scion Corp., Frederick, MD) and are presented inarbitrary units.

p38MAPK, ERK1/2, and JNK1/2 phosphorylation

Neutrophils (107 cells) were preincubated for 18 h in RPMI medium at 37°Cto reduce the p38MAPK, JNK1/2, and ERK1/2 basal phosphorylation levelsusually found in our preparations of human neutrophils. These were thensubjected to the treatments indicated in each case, washed with PBS, andimmediately lysed by incubation for 30 min in a buffer composed of 20 mMTris-HCl, pH 7.4, 50 mM �-mercaptoethanol, 2 mM EDTA, 25 mM NaF, 1%NP-40, 1% Triton X-100, 2 mM diisopropylfluorophosphate (DFP), and theprotease-inhibitor mixture above. Protein extracts (80 g) were resolved on10% SDS-PAGE gels and electrotransferred to PVDF membranes, which wereprobed for 1 h with antiphospho-p38MAPK, antiphospho-ERK1/2, or antiphos-pho-JNK1/2 antibodies (1:2000 dilution). Antibody binding was detected byusing HRP-conjugated anti-rabbit or anti-mouse IgG antibodies (1:5000 dilu-tion) followed by ECL.

PGE2 release

Neutrophils (107 cells) were incubated in 1.5 ml RPMI medium on 24-wellplates in the presence of antigen or anti-IgE antibody at the doses indicated ineach case. The concentrations of PGE2 released were measured spectropho-tometrically in the culture supernatants after neutrophil removal by centrifu-gation at 14,000 g for 5 min by using the PGE2 enzyme immunoassay kit(Cayman Chemical) according to the manufacturer’s protocol.

Measurement of TXA2 production

Neutrophils (107 cells) were incubated in 1.5 ml RPMI medium on 24-wellplates as described above for PGE2 release, and TXA2 was assayed afterneutrophil removal by centrifugation. As TXA2 has a short half-life (37 s) andis rapidly hydrolyzed nonenzymatically to its stable derivative TXB2, theThromboxane B2 enzyme immunoassay kit (Cayman Chemical) was used tomeasure free TXA2 indirectly.

ROS levels

The levels of ROS were analyzed by luminol-amplified chemiluminescence,measured in a BioOrbit 1250 luminometer (Turku, Finland) at 37°C, basicallyas described [16]. Briefly, neutrophils (106 cells) were incubated for 5 min in1 ml PBS supplemented with 10 mM glucose, 500 M CaCl2, 5 M luminol,and 100 M sodium azide for 5 min. Then, allergens or anti-IgE antibodies asstimulants were added at the doses indicated in each case. The chemilumi-nescent response was measured every 1 min in each reaction and was ex-pressed as the peak of luminescence recorded at 10 min.

I-�B cytosolic levels

For immunoblot analysis of I-�B-� levels, the cytoplasmic fraction was ob-tained from neutrophils as described previously [21] and subjected to SDS-PAGE, electroblotting onto PVDF membranes. These were probed with rabbitanti-human I-�B-� (1:2000 dilution) and thereafter, with HRP-conjugatedanti-rabbit IgG (1:5000 dilution).

Electrophoretic mobility shift assay (EMSA) ofNF-�B DNA-binding activity

Nuclear extracts were obtained as described previously [21] with minor mod-ifications. The nuclear pellet was resuspended in 25 l ice-cold relaxationbuffer, additionally containing 10% (v/v) glycerol and 380 mM NaCl, supple-mented with an antiprotease mixture composed of 2 mM DFP, 1 mM PMSF, 10mM iodoacetamide, 1 mM benzamidine, and 10 g/ml each aprotinin, leupep-

tin, and captopril. The double-stranded NF-�B oligonucleotide was end-labeled with [�-32P]ATP using T4 polynucleotide kinase. Nuclear proteins (10g) were incubated with radioactive probes for 30 min on ice in binding buffer(20 mM Hepes, pH 7.8, 6% glycerol, 1 mM EDTA, 50 mM KCl, 1 mM DTT,and 0.1% NP-40), supplemented with 1 g poly-(dI-dC) and 0.1 g poly-L-lysine. Protein-DNA complexes were resolved in native 6% polyacrylamidegels and visualized by autoradiography. For DNA-competition assays, a 100-fold molar excess of unlabeled oligonucleotide was included in the reactionsprior to addition of the radiolabeled probe.

RESULTS

Induction of COX-2 expression by specificantigens and anti-IgE antbody in humanneutrophils

To evaluate the potential role of antigens as inducers of COX-2expression, neutrophils, isolated from a patient specificallysensitized to the G3 antigen, were cultured in the presence ofthis antigen, and the expression of the COX-2 enzyme wasanalyzed by immunoblotting. Figure 1 shows that whereasCOX-2 was essentially undetectable in untreated neutrophils,after cell stimulation with the G3 antigen, expression of thisprotein was induced in a dose- and a time-dependent manner.Maximum COX-2 levels were attained at a 10-g/ml concen-tration of the G3 antigen (Fig. 1A), a dose at which the COX-2protein was detectable after 6 h of treatment, and the peak ofexpression was reached at 12–24 h (Fig. 1C). Neutrophilsamples from patients allergic to other antigens gave a similarresponse toward their specific sensitizing antigens (data notshown). Further studies were addressed to analyze whetheranti-human IgE antibodies could also induce COX-2 expres-sion in allergic patients. Results similar to those obtained forthe antigens above were obtained with anti-IgE antibodies usedas the stimulus in neutrophils from a G3-allergic patient (Fig.1, B and D). Figure 1E illustrates that in neutrophils from apatient exhibiting high serum levels of IgE specific for M6, T9,and G3, COX-2 synthesis became induced when they werecultured in the presence of either of these three antigens.Conversely, when neutrophils were cultured with antigens towhich the patient was not sensitized, such as E1, D1, or D2, theCOX-2 protein was not detected (Fig. 1E). The specificity ofthis response was assessed further in neutrophils from healthysubjects, in which COX-2 expression was not elicited by any ofthese allergens (Fig. 1F). Nevertheless, this enzyme was clearlydetected when neutrophils from healthy subjects were treatedwith LPS (Fig. 1F), a potent inducer of COX-2 expression inthese cells as described [4]. However, when neutrophils fromallergic patients were cultured with anti-IgG antibodies, noCOX-2 expression was detected, as illustrated in Figure 1G fora D1-allergic donor. Furthermore, in no case was COX-2 ex-pression found when neutrophils from allergic donors weretreated with nonspecific goat IgG antibodies (Fig. 1, A, B, andG). Further analysis using RT-PCR indicated that induction ofCOX-2 mRNA expression took place within 0.5 h of treatmentand was maintained for 24 h in neutrophils from allergicpatients after challenge with anti-IgE antibodies (Fig. 1H). Incontrast, undetectable levels of COX-2 mRNA were foundwhen these cells were treated with nonspecific goat IgG anti-bodies.

154 Journal of Leukocyte Biology Volume 80, July 2006 http://www.jleukbio.org

Antigen-promoted release of PGE2 and TXA2

To assess whether specific antigens were able to induce PGrelease, neutrophils from patients sensitized to G3 were cul-tured with different doses of this antigen or of anti-IgE anti-bodies for 24 h, and then, PGE2 and TXA2 levels were mea-sured in cell culture supernatants. When unstimulated cellswere exposed to the antigen, PGE2 (Fig. 2A) and TXA2 (Fig.2B) production was increased by six- to eightfold upon basallevels. This increase was G3 dose-dependent, and a maximalrelease of PGE2 and TXA2 was obtained at a 10-g/ml con-centration of G3 or anti-IgE antibodies at 24 h of treatment. Asalso shown, the COX-2 inducer LPS elicited a PGE2 and TXA2

release by neutrophils similar to that induced by anti-IgEantibodies (Fig. 2, A and B). Equivalent responses were ob-tained for other antigens, such as D1 or E1, in patients specif-ically sensitized to these antigens (data not shown).

Role of NADPH oxidase in the control of IgE-dependent COX-2 expression

ROS are recognized as second messengers modulating theexpression of several genes in a variety of immune-system celltypes [7, 15, 20]. To evaluate the possible implication of ROS,mainly derived from neutrophil NADPH oxidase activity, in theIgE-dependent COX-2 expression, we first assessed the possi-bility of IgE-elicited activation of the NADPH oxidase complextaking place upon translocation of its cytosolic subunits,p47phox and p67phox, to the plasma membrane. As shown inFigure 3A, the anti-IgE antibody induced a clear transloca-tion to the plasma membrane of p47phox and p67phox, and therewas a concomitant disappearance of these proteins from thecytosolic fraction. Second, as shown in Figure 3B, we foundthat after treatment with anti-IgE antibody or specific antigens,the enzyme complex was functionally active in the production

Fig. 1. Induction of COX-2 protein and mRNA expression by antigens and anti-IgE antibodies in human neutrophils, which (A and B) were from an allergic patientsensitized to G3 and cultured in the presence of the indicated concentrations of this antigen or anti-IgE antibodies (�-IgE) or of 10 g/ml IgG for 24 h. (C andD) Neutrophils from an allergic patient sensitized to G3 were cultured in the presence of this antigen or anti-IgE antibodies at 10 g/ml, LPS at 1 g/ml, or IgGat 10 g/ml for the times indicated, up to 24 h. Neutrophils from a patient sensitized to M6, T9, and G3 (E) or from a healthy subject (F) were challenged with10 g/ml M6, T9, G3, E1, D1, or D2 antigen or with 1 g/ml LPS for 24 h. (G) Neutrophils from a patient allergic to D1 were cultured for 24 h with anti-IgG (�-IgG)at the indicated doses or with 1 g/ml LPS or IgG for 24 h. In all cases, COX-2 protein levels were analyzed by Western blot analysis as described in Materialsand Methods. (H) Neutrophils from an allergic patient sensitized to T9 were left untreated or treated with 10 g/ml anti-IgE antibodies or IgG for the indicatedtimes, and then, COX-2 mRNA levels were analyzed by RT-PCR using COX-2- and GADPH-specific primers. All the experiments were performed at least intriplicate with similar results.


of ROS and that its activation was cancelled by specificNADPH oxidase inhibitors, such as HMAP, which competeswith NADPH for a binding site on the oxidase [23], andAEBSF, which blocks the assembly of the NADPH oxidasesubunits at the plasma membrane [24]. PMA was used in theseexperiments as a positive control for p47phox and p67phox

activation and for ROS generation [8, 22], with consistentresults (Fig. 3, A and B). To further confirm the implication ofROS arising from the NADPH oxidase system, we analyzed theeffect of HMAP and AEBSF on COX-2 expression. As shownin Figure 3C, these two inhibitors completely cancelled theIgE-dependent COX-2 protein synthesis.

H2O2 and specific ROS derived from Fentonreactions are implicated in the COX-2 expression

After passive diffusion through the plasma membrane, H2O2

can be converted into other ROS, such as the oxygen radicalsO2

�� and ·OH. The most likely mode of intracellular ·OH

radical production is via Fenton chemistry, which involves thereduction of H2O2 by ferrous or cuprous [25] ions according tothe reaction H2O2 � Fe2�/Cu2� 3 ·OH � OH� � Fe3�/Cu3�. Figure 4A shows that exogenously added FeSO4/CuSO4

elicited a strong enhancement of the IgE-dependent COX-2induction, although without effect by themselves, and that theFe2� and Cu2� chelator PHE inhibited IgE-promoted COX-2induction. Taken together, these data show an involvement ofNADPH oxidase-derived ROS, possibly with ·OH acting as amediator, in IgE-dependent COX-2 expression.

To further confirm the positive, modulatory role of ROS onCOX-2 expression, we analyzed the effect of H2O2 on thisprocess. This compound was found to induce by itself COX-2expression in a dose-dependent manner at the protein (Fig. 4B)and mRNA levels (Fig. 4C). Detection of this H2O2 stimulatoryeffect required, however, the treatment of cells simultaneouslywith AMT (Fig. 4, B and C), an inhibitor of catalase, a H2O2-removing enzyme [26]. Next, we assessed the possibility ofchemical species subsequently generated from H2O2 throughFenton reactions acting as modulators of COX-2 protein ex-pression. As shown in Figure 4D, the presence of Fe2�/Cu2�

ions notably enhanced the H2O2-induced COX-2 expression.In contrast, the Fe2� and Cu2� chelator PHE completelycancelled COX-2 protein accumulation elicited by H2O2

(Fig. 4D).

Role of MAPK in the control of IgE-dependentCOX-2 expression

MAPK family proteins, such as p38, ERK1/2, and JNK1/2,have been implicated in the regulation of COX-2 expression ina broad spectrum of cells [7, 27]. As shown in Figure 5,neutrophils from an allergic donor sensitized to D1 showed aclear phosphorylation and hence, activation of p38MAPK andERK1/2 in a time-dependent manner when cultured in thepresence of anti-IgE antibodies (Fig. 5A) or of the specificantigen D1 (Fig. 5B). No activation of these MAPKs wasobserved, however, in neutrophils from healthy donors whencultured in the presence of a set of antigens or in neutrophilsfrom allergic donors cultured with antigens to which the par-ticular donor was not sensitized (data not shown). In contrast tothe results obtained for p38MAPK and ERK1/2, the JNKactivation status did not change upon antigen treatment ofhuman neutrophils (data not shown). To study the implicationof these signaling pathways on the modulation of COX-2 ex-pression, we tested the effect on this process of PD098059, aspecific inhibitor of MAPK kinase [28], and of SB203580, aspecific inhibitor of p38MAPK [29]. Both compounds werefound to effectively block anti-IgE antibody-promoted induc-tion of COX-2 expression in neutrophils from allergic donors,whereas IgG had no such effect (Fig. 5, C and D).

MAPK pathway is associated with IgE-dependent ROS generation

Given the relationship between the activity of MAPKs and theintracellular redox status in neutrophils [8, 9], we investigatedwhether such interconnection was operative in the IgE-pro-moted COX-2 up-regulation. The NADPH oxidase inhibitors,HMAP and AEBSF, were found to completely cancel the

Fig. 2. Antigen- and anti-IgE antibody-promoted PGE2 and TXA2 release. (Aand B) Neutrophils were incubated for 24 h with different doses of the specificantigen (G3) to which the allergic donor was sensitized, with anti-IgE antibody(�-IgE) or with 1 g/ml LPS, and the levels of PGE2 (A) or TXA2 (B) weremeasured in the culture medium supernatants using an enzyme immunoassay.As TXA2 has a short half-life and spontaneously hydrolyzes to TXB2, TXA2

was measured as TXB2. The values shown are the mean � SE from threeindependent assays, in which each measurement was performed in triplicate.


IgE-dependent phosphorylation and hence, activation ofp38MAPK (Fig. 6A) and ERK1/2 (Fig. 6B). As ROS derivedfrom Fenton reactions appeared involved in the IgE-dependentmodulation of COX-2 expression, we studied whether thesespecies did contribute to p38MAPK and ERK1/2 activation.As shown, the addition of Fe2�/Cu2� ions strongly enhancedthe IgE-dependent phosphorylation of p38MAPK (Fig. 6C) andERK1/2 (Fig. 6D). Conversely, the chelator PHE completelycancelled the anti-IgE antibody-elicited, positive effect on theactivation of these two MAPKs (Fig. 6, C and D). It is notice-able that there was a robust phosphorylation of ERK1/2 in thepresence of Fe2�/Cu2� alone, i.e., in the absence of anti-IgEantibodies (Fig. 6D), whereas this was not the case forp38MAPK (Fig. 6C). These observations are similar to thosereported previously in vascular smooth muscle cells [30] andfibroblasts [31] incubated with different stimuli. These andpresent observations suggest that ERK1/2 proteins are moresensitive than p38MAPK to small intracellular redox changesin certain cell types. In agreement with these observations,

nonlethal concentrations of H2O2 have been demonstrated toactivate p38MAPK and ERK1/2 [32].

Involvement of transcription factor NF-�B onCOX-2 up-regulation by anti-IgE antibodies

The involvement of NF-�B in the activation of COX-2 geneexpression [10] and its role in inflammatory disorders [12] areknown. Thus, we next studied the implication of this transcrip-tion factor in the IgE-promoted COX-2 expression by usingnuclear extracts prepared from anti-IgE antibody- or D1 anti-gen-treated neutrophils isolated from allergic donors sensitizedto this antigen. We observed that the stimulation of these cellswith anti-IgE antibody (Fig. 7A) or D1 antigen (Fig. 7B)resulted in a weak activation of NF-�B after 30 min of treat-ment, which was maximal after 2–4 h of treatment and wasdecreased significantly at 24 h. These data on NF-�B activa-tion were consistent with previous results from Pouliot et al.[33]. Furthermore, the activation by anti-IgE antibody occurredin a dose-dependent manner (Fig. 7C). The specificity of this

Fig. 3. Involvement of NADPH oxidase in IgE-dependent COX-2 expression. (A) Neutrophils from an allergic patient sensitized to T9 were treated with 50 nMPMA for 10 min or with 10 g/ml anti-IgE antibody (�-IgE) for the times indicated. Then, membrane (left panel) and cytosolic (right panel) fractions were isolatedas described in Materials and Methods, resolved by SDS-PAGE, and electrotransferred to PVDF membranes. These were probed sequentially with specificantibodies against the p47 and after stripping p67 subunits of NADPH oxidase. Equal amounts of protein were loaded per lane. Histograms below each lanerepresent the mean � SE values quantitated from the blots from three separate experiments. (B) Neutrophils from an allergic patient sensitized to G3 were treatedor not with 500 M HMAP or AEBSF prior to the addition of 50 nM PMA, 10 g/ml anti-IgE antibody, or 10 g/ml G3 for 10 min. Then, O2

– production wasmeasured by luminol ECL. The values shown are the mean � SE from three independent assays in which each measurement was performed in triplicate. (C)Neutrophils from an allergic patient sensitized to G3 were left untreated or cultured with 500 M HMAP or AEBSF for 30 min prior to the addition of 10 g/mlanti-IgE antibody. After 24 h of treatment, COX-2 expression was analyzed by Western blotting. All the experiments were performed at least in triplicate.


DNA-binding activity was evidenced by competition with a100-fold molar excess of an unlabeled NF-�B probe (Fig. 7,A–C). As in unstimulated cells, NF-�B is anchored to thecytoplasm by a number of inhibitory proteins of the I-�Bfamily, which become degraded at the proteasome upon cellstimulation with concomitant translocation of NF-�B to thenucleus [34, 35], we next analyzed the cytoplasmic levels ofI-�B before and after anti-IgE antibody treatment. As shown inFigure 7D, anti-IgE antibody promoted, in a period of 1–4 h,an almost complete degradation of cytosolic I-�B in humanneutrophils, which correlated well in a timely manner with theenhancement of NF-�B DNA-binding activity (Fig. 7A). Fur-thermore, the use of MG-132, a proteasome inhibitor [34],strongly attenuated anti-IgE antibody-induced COX-2 proteinexpression (Fig. 7E).

It has been demonstrated that phosphorylation of theMAPKs p38 and ERK1/2 is a necessary step for NF-�Bactivation in some cell types [36]. In this context, the presenceof SB203580 or PD098059, specific inhibitors of p38MAPKand ERK1/2, respectively, completely inhibited the IgE-de-pendent NF-�B DNA-binding activity in human neutrophils(Fig. 7F). As well, a previous report has shown a positivecorrelation between H2O2-dependent COX-2 expression andNF-�B activation in cardiomyocytes [37]. We observed in thisstudy that exogenously added H2O2, in the presence of thecatalase inhibitor AMT, did enhance NF-�B DNA-bindingactivity in neutrophils (Fig. 8A). Finally, the potential partic-ipation of NADPH oxidase on the IgE-dependent NF-�B acti-vation was also assessed. We found that the two specific

NADPH oxidase inhibitors, HMAP and AEBSF, partially can-celled IgE-dependent NF-�B activation (Fig. 8B). Strengthen-ing these observations, the presence of Fe2�/Cu2� ions re-sulted in an enhancement of the DNA-binding activity ofNF-�B, a process that was cancelled completely by the pres-ence of PHE (Fig. 8B).

DISCUSSION

Although the capacity of human neutrophils to induce COX-2expression in response to inflammatory agonists has beenshown [4], no studies have been performed yet about themodulation of COX-2 expression by an IgE-dependent mech-anism in neutrophils or the molecular mechanisms implicatedin this process. The present study unravels that COX-2 expres-sion becomes up-regulated in human neutrophils at the mRNAand protein levels, with a concomitant PGE2 and TXA2 release,in response to challenge with specific antigens or anti-IgEantibodies. This phenomenon raises further questions aboutthe contribution of these cells to inflammatory processes, suchas allergy. In this context, the relationship between PGE2 andallergy remains to be well-defined and is currently the subjectof different interpretations. Given that increased levels of PGE2

and COX-2 expression have been found in asthmatic subjects[38], this finding, together with present data, could point out arole for COX-2 as inducer of the allergy state, and PGs couldset to exacerbate the injury on surrounding tissues. In this line,a relationship has been described between COX-2 expression

Fig. 4. Effect of H2O2 on COX-2 expression. Involvement of Fenton chemistry. (A) Neutrophils from an allergic patient sensitized to T9 were left untreated orcultured with 20 M FeSO4 plus 20 M CuSO4 (Fe2�/Cu2�) or with 20 M PHE for 30 min prior to the addition of 5 g/ml anti-IgE antibody (�-IgE) for 24 h.Then, COX-2 protein expression was analyzed by Western blotting. (B) Neutrophils from an allergic patient sensitized to T9 were cultured in the presence orabsence of 25 mM aminotriazole (AMT) and H2O2 at the indicated doses or of 1 g/ml LPS for 24 h, and COX-2 expression was then analyzed by Western blotting.(C) Neutrophils from an allergic patient sensitized to T9 were cultured with 1 g/ml LPS or with 25 mM AMT in the absence or presence of 5 M H2O2 for 5 h,and the levels of COX-2 mRNA were analyzed by RT-PCR. (D) Neutrophils from an allergic patient were preincubated with 25 mM AMT for 30 min. Then, 20M FeSO4 plus 20 M CuSO4 (Fe2�/Cu2�) or 20 M PHE was added to the culture for 30 min prior to the addition of LPS at 1 g/ml or of H2O2 at 0.5 or 5mM, as indicated, and COX-2 expression was analyzed by Western blotting. The results illustrated in each panel are representative of a total of three separateexperiments.


Fig. 5. Involvement of MAPKs in antigen/anti-IgE antibody-dependent COX-2 protein expression. Neutrophils from an allergic patient sensitized to D1 werecultured with 10 g/ml anti-IgE antibodies (�-IgE; A) or with 10 g/ml D1 antigen (B) for the times indicated, and phosphorylated p38MAPK (p38MAPK-P) andERK1/2 (ERK1/2-P) were subsequently detected. Total (phosphorylated plus unphosphorylated forms) p38MAPK and ERK1/2 are also shown. Neutrophils froman allergic patient sensitized to D1 were preincubated for 45 min with PD098059 (PD; C) or with SB203580 (SB; D) at the doses indicated and then, treated with10 g/ml anti-IgE antibody for 24 h. The expression of COX-2 was then analyzed by Western blotting. The results illustrated in each panels are representativeof a total of three separate experiments.

Fig. 6. Effect of NADPH oxidase inhibitors and Fenton-promoted cations on MAPKs triggering. Neutrophils from an allergic patient sensitized to T9 were culturedwith 500 M HMAP or AEBSF (A and B), with 20 M FeSO4 plus 20 M CuSO4 (Fe2�/Cu2�) or with 20 M PHE (C and D) for 30 min prior to the additionof 10 g/ml anti-IgE antibodies (�-IgE). After a further 30 min of incubation, the cells were lysed, and proteins were analyzed by Western blotting using specificantibodies against the phosphorylated forms of p38 (p38MAPK-P; A and C), ERK1/2 (ERK1/2-P; B and D), or the total phosphorylated plus unphosphorylatedforms of these proteins. Results illustrated in each panel are representative of a total of three separate experiments.


and NADPH oxidase activation and an attenuation for COX-2inhibition in microglial cells [39]. Recently, it has also beenevidenced that the PGE2-E-prostanoid receptor 2 (EP2) systemcontributes to the local production of granulocyte colony-stim-ulating factor during acute inflammation in mouse peritonealneutrophils [40]. However, another potential interpretation forthe association between PGE2 and COX-2 is that COX-2induction by antigens constitutes a physiological defensemechanism aimed at counteracting the effect of other proin-flammatory molecules. In support of the role of PGE2 as aprotective molecule in allergic processes, it has been shownthat COX-2 inhibition enhances the T helper cell type 2immune response (e.g., the allergic response) to cutaneoussensitization [41]. Also, it has been shown that PGs (e.g.,PGD2, PGE2, and PGF2-�) constitute potent inhibitors of O2

��

production by human neutrophils [42]. However, recent datahave brought into play new, potential signaling molecules, suchas PGE2 receptors. In this context, recent studies have high-lighted that mice lacking PGE2-EP receptors (EP3 subtype)develop a far more pronounced allergic inflammation than do

wild-type mice or mice deficient in other PGE2 receptor sub-types [43]. Thus, the PGE2 actions elicited through EP3 re-ceptors could be considered protective against allergic reac-tions. However, increased numbers of other EP receptors of theEP2 subtype can enhance the sensitivity of asthmatic airwaysmooth muscle cells to PGE2 [44]. Therefore, the balance ofEP3/EP2 receptor expression has been postulated as a keyfactor in the development of the allergic state. In contrast, apositive relationship between TXA2 and allergy seems well-accepted [45].

It is important to note that the three known forms of IgEreceptors, namely Fc receptor for IgE (FcεR)I, FcεRII/CD23,and galectin-3 [46], are present in neutrophils, thus supportingthe potential participation of these cells in allergic processes[13–15]. In this regard, we have shown that specific antigensare able to activate several functional responses by neutrophilsfrom allergic patients and highlighted the presence of specificIgE molecules bound to antigens on the surface of neutrophils[15, 16]. As well, we have reported the absence of IgG mole-cules specific for the antigens to which the patients were

Fig. 7. Role of NF-�B in IgE-dependent COX-2 induction. Neutrophils from a patient sensitized to D1 were treated with 10 g/ml anti-IgE antibody (�-IgE; A)or 10 g/ml D1 antigen (B) for the indicated times or with anti-IgE antibody at the indicated doses for 4 h (C). Then, nuclear extracts were prepared, and NF-�BDNA-binding activity was analyzed by EMSAs. Brackets show the retarded band. (D) Neutrophils from an allergic patient were treated with 10 g/ml anti-IgEantibody for the indicated times, and I-�B-� levels were analyzed by Western blotting in cytosolic extracts. (E) COX-2 expression was analyzed by Western blottingin neutrophils from an allergic patient sensitized to D1, pretreated or not with MG-132 for 2 h at the indicated doses, and then treated with 10 g/ml anti-IgEantibody for 24 h. (F) Neutrophils from an allergic patient sensitized to T9 were pretreated or not for 1 h with 1 M SB203580 or 10 M PD098059 prior to theaddition of 10 g/ml anti-IgE antibody. After 4 h of incubation, nuclear extracts were analyzed for NF-�B DNA-binding activity. Binding reactions marked withan asterisk (A–C, F) were performed in the presence of a 100-fold molar excess of unlabeled oligonucleotide probe on neutrophils treated with 10 g/ml anti-IgEantibody for 4 h. Two additional experiments were performed with results similar to those shown in each panel.


sensitized [15]. We have thus postulated that the mechanismswhereby antigens promote neutrophil activation could well berelated to the binding of antigens to specific IgE moleculesassociated with their specific receptors at the neutrophilplasma membrane. This mechanism would explain why func-tional responses elicited by antigens, such as induction ofCOX-2 expression, are not observed in neutrophils fromhealthy subjects or from allergic patients treated with antigensother than those to which they had previously become sensi-tized. In the latter case, antigens would not bind to pre-existentIgE/receptor complexes, and ensuing COX-2 induction wouldnot take place.

It is well known that stimulation of phagocytic cells inducesa set of phenomena, known collectively as the respiratory burst,characterized by an increase in the production of O2

�� anions[16, 47]. The major O2

��-generating enzyme in neutrophils isNADPH oxidase. Upon its reduction, O2 may sequentiallygenerate O2

��, which is rapidly converted into H2O2 by theaction of superoxide dismutase [48], and ·OH in the presenceof Fe2�/Cu2� cations, through the so-called Fenton reaction.Previously, we have demonstrated that in vitro challenge ofhuman neutrophils with antigens to which the patients weresensitized or with anti-IgE antibody elicited a greater release ofROS than that found in neutrophils from healthy subjects [16].Recent studies have highlighted a connection between thisup-regulation of COX-2 expression and the generation of ROSduring monocyte-to-macrophage differentiation [7]. It is inter-esting that neutrophils from allergic subjects have been shownto generate higher levels of O2

�� than those isolated fromhealthy individuals [49]. However, the molecular mechanismsmediating COX-2 gene induction in a wide range of cells,including neutrophils, remain unclear at present.

In agreement with our previous report [16], we now describethat specific antigens/anti-IgE antibodies were able to elicit a

clear activation of the NADPH oxidase complex in neutrophils,concomitant with the translocation to the plasma membrane ofits cytosolic p47phox and p67phox subunits and the subsequentrelease of ROS. This process was abrogated by HMAP andAEBSF, two specific NADPH oxidase inhibitors, which werealso found to cancel antigen/anti-IgE antibody-dependentCOX-2 expression. Moreover, as the addition of Fe2�/Cu2�

cations or of the Fe2�/Cu2� chelator PHE enhanced or inhib-ited, respectively, the induction of COX-2 expression promotedby antigens or anti-IgE antibodies, oxygen species derivedfrom Fenton reactions could constitute the causal effectors ofCOX-2 up-regulation.

Taken together, present data suggest that the induction ofCOX-2 expression promoted by specific antigens/anti-IgE an-tibodies is modulated by ROS, possibly acting through theactivation of MAPK-signaling pathways and the transcriptionfactor NF-�B. In this context, the interplay among MAPK-dependent pathways, NF-�B activation, and ROS is largelyknown in neutrophils [8, 9]. In this light, ERK1/2 andp38MAPK have been implicated previously in the induction ofCOX-2 expression in monocytes [7]. The complexity of theinteractions operative among these pathways was evidenced bythe fact that NADPH oxidase inhibitors cancelled the activa-tion of p38MAPK and ERK1/2 promoted by anti-IgE antibodytreatment and that SB203580 and PD098059, two inhibitors ofthese MAPKs, provoked a strong down-regulation of COX-2expression. Therefore, present data underscore that the posi-tive action of ROS production on IgE-dependent COX-2 in-duction is likely exerted via the activation of the MAPKs p38and ERK1/2.

In addition, the relevance of NF-�B-binding sites in theCOX-2 promoter and their role in development of the allergicstate are well-established [11, 12]. In the present work, weshow evidence that the stimulation of neutrophils with anti-IgE

Fig. 8. Effect of H2O2 and Fenton-derived ROS on IgE-dependent NF-�B activation. (A) Neutrophils from an allergic patient sensitized to D1 were treated with25 mM AMT for 30 min prior to the addition of H2O2 at the indicated doses after 4 h of incubation NF-�B. DNA-binding activity was analyzed in nuclear extracts.(B) Neutrophils from an allergic patient sensitized to D1 were cultured with or without 500 M HMAP or AEBSF, 20 M FeSO4 plus CuSO4 (Fe2�/Cu2�), or 20M PHE for 30 min prior to the addition of 5 g/ml anti-IgE antibodies. After 4 h of incubation, nuclear extracts were prepared, and NF-�B DNA-binding activitywas analyzed. Brackets show the retarded band. Binding reactions marked by an asterisk were performed in the presence of a 100-fold molar excess of unlabeledoligonucleotide in neutrophils treated with 25 mM AMT plus 10 M H2O2 (A) or with 5 g/ml anti-IgE antibody (B) for 4 h. Two additional experiments wereperformed with results similar to those shown.


antibodies or D1 antigen resulted in a weak activation ofNF-�B in a matter of minutes, followed by maximal activationafter 2–4 h of treatment. These data on NF-�B activation wereconsistent with previous results from Pouliot et al. [33]. Also,we have observed that degradation of cytoplasmic I-�B be-comes triggered upon anti-IgE antibody treatment with con-comitant activation of NF-�B in the nucleus. The relationshipsbetween NF-�B activation and the referred cytosolic signalingpathways are evidenced further by the fact that specific inhib-itors of MAPKs or of NADPH oxidase also negatively affectedNF-�B DNA-binding activity. In summary, present observa-tions support a role for intracellular oxidants as participants ina novel molecular mechanism, elicited by antigens or anti-IgEantibodies, leading to COX-2 induction in human neutrophilsduring the development of the allergic state. Our data establishthe existence of intracellular signal transduction mechanisms,such as O2

�� release, MAPK, and NF-�B activation, althoughthe exact hierarchical order of these pathways needs to beclarified still.

ACKNOWLEDGMENTS

A. V. and P. C. were supported by fellowships from theMinisterio de Educacion y Ciencia and G. A., by the Junta deAndalucıa (J.A.), Spain. This work was funded by Grant SAS-74/04 from the Consejerıa de Salud, J.A., awarded toF. S. A. V. and P. C. contributed equally to this work.

REFERENCES

1. Smith, W. L. (1992) Prostanoid biosynthesis and mechanisms of action.Am. J. Physiol. 263, F181–F191.

2. Vane, J. R., Bakhle, Y. S., Botting, R. M. (1998) Cyclooxygenases 1 and2. Annu. Rev. Pharmacol. Toxicol. 38, 97–120.

3. Smith, W. L., Dewitt, D. L., Garavito, R. M. (2000) Cyclooxygenases:structural, cellular and molecular biology. Annu. Rev. Biochem. 69,145–181.

4. Maloney, C. G., Kutchera, W. A., Albertine, K. H., McIntyre, T. M.,Prescott, S. M., Zimmerman, G. A. (1998) Inflammatory agonists inducecyclooxygenase type 2 expression by human neutrophils. J. Immunol.160, 1402–1410.

5. Irani, K. (2000) Oxidant signaling in vascular cell growth, death, andsurvival: a review of the roles of reactive oxygen species in smooth muscleand endothelial cell mitogenic and apoptotic signaling. Circ. Res. 87,179–183.

6. Babior, B. M. (1999) NADPH oxidase: an update. Blood 93, 1464–1476.7. Barbieri, S. S., Eligini, S., Brambilla, M., Tremoli, E., Colli, S. (2003)

Reactive oxygen species mediate cyclooxygenase-2 induction duringmonocyte to macrophage differentiation: critical role of NADPH oxidase.Cardiovasc. Res. 60, 187–197.

8. El Bekay, R., Alvarez, M., Monteseirın, J., Alba, G., Chacon, P., Vega, A.,Martın-Nieto, J., Jimenez, J., Pintado, E., Bedoya, F. J., Sobrino, F. (2003)Oxidative stress is a critical mediator of the angiotensin II signal in humanneutrophils: involvement of mitogen-activated protein kinase, calcineurin,and the transcription factor NF-�B. Blood 102, 662–671.

9. Detmers, P. A., Zhou, D., Polizzi, E., Thieringer, R., Hanlon, W. A.,Vaidya, S., Bansal, V. (1998) Role of stress-activated mitogen-activatedprotein kinase (p38) in �2-integrin-dependent neutrophil adhesion and theadhesion-dependent oxidative burst. J. Immunol. 161, 1921–1929.

10. Appleby, S. B., Ristimaki, A., Nelson, K., Narko, K., Hla, T. (1994)Structure of the human cyclooxygenase-2 gene. Biochem. J. 302, 723–727.

11. Iniguez, M. A., Martinez-Martinez, S., Punzon, C., Redondo, J. M., Fresno,M. (2000) An essential role of the nuclear factor of activated T cells in the

regulation of the expression of the cyclooxygenase-2 gene in human Tlymphocytes. J. Biol. Chem. 275, 23627–23635.

12. Barnes, P. J., Adcock, I. M. (1997) NF-�B: a pivotal role in asthma andnew target for therapy. Trends Pharmacol. Sci. 18, 46–50.

13. Lee, T. H., Nagakura, T., Wallport, M. J., Kay, A. B. (1982) Identificationand partial characterization of an exercise-induced neutrophil chemotacticfactor in bronchial asthma. J. Clin. Invest. 69, 889–899.

14. Nagy, L., Lee, T. H., Kay, A. B. (1982) Neutrophil chemotactic activity inantigen-induced late asthmatic reactions. N. Engl. J. Med. 306, 497–501.

15. Monteseirın, J., Chacon, P., Vega, A., El Bekay, R., Alvarez, M., Alba, G.,Conde, M., Jimenez, J., Asturias, J. A., Martınez, A., Conde, J., Pintado,E., Bedoya, F. J., Sobrino, F. (2004) Human neutrophils synthesize IL-8 inan IgE-mediated activation. J. Leukoc. Biol. 76, 692–700.

16. Monteseirın, J., Camacho, M. J., Montano, R., Llamas, E., Conde, M.,Carballo, M., Guardia, P., Conde, J., Sobrino, F. (1996) Enhancement ofantigen-specific functional responses by neutrophils from allergic patients.J. Exp. Med. 183, 2571–2579.

17. Boyum, A. (1968) Isolation of mononuclear cells and granulocytes fromhuman blood: isolation of monuclear cells by one centrifugation, and ofgranulocytes by combining centrifugation and sedimentation at 1 g. Scand.J. Clin. Lab. Invest. Suppl. 97, 77–89.

18. Gualberto, A., Marquez, G., Carballo, M., Youngblood, G. L., Hunt, S. W.,Baldwin, A. S., Sobrino, F. (1998) p53 transactivation of the HIV-1 longterminal repeat is blocked by PD 144795, a calcineurin-inhibitor withanti-HIV properties. J. Biol. Chem. 273, 7088–7093.

19. Chomczynski, P., Sacchi, N. (1987) Single-step method of RNA isolationby acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Bio-chem. 162, 156–157.

20. Vega, A., Chacon, P., Monteseirın, J., El Bekay, R., Alvarez, M., Alba, G.,Conde, J., Martın-Nieto, J., Bedoya, F. J., Pintado, E., Sobrino, F. (2004)A new role for monoamine oxidases in the modulation of macrophage-inducible nitric oxide synthase gene expression. J. Leukoc. Biol. 75,1093–1101.

21. Carballo, M., Marquez, G., Conde, M., Martın-Nieto, J., Monteseirın, J.,Conde, J., Pintado, E., Sobrino, F. (1999) Characterization of calcineurinin human neutrophils. Inhibitory effect of hydrogen peroxide on its enzymeactivity and on NF-�B DNA binding. J. Biol. Chem. 274, 93–100.

22. El Bekay, R., Alvarez, M., Carballo, M., Martın-Nieto, J., Monteseirın, J.,Pintado, E., Bedoya, F. J., Sobrino, F. (2002) Activation of phagocytic cellNADPH oxidase by norfloxacin: a potential mechanism to explain itsbactericidal action. J. Leukoc. Biol. 71, 255–261.

23. Satriano, J., Schlondorff, D. (1994) Activation and attenuation of tran-scription factor NF-�B in mouse glomerular mesangial cells in response totumor necrosis factor-�, immunoglobulin G, and adenosine 3�:5�-cyclicmonophosphate. Evidence for involvement of reactive oxygen species.J. Clin. Invest. 94, 1629–1636.

24. Diatchuk, V., Lotan, O., Koshkin, V., Wikstroem, P., Pick, E. (1997)Inhibition of NADPH oxidase activation by 4-(2-aminoethyl)-benzenesul-fonyl fluoride and related compounds. J. Biol. Chem. 272, 13292–13301.

25. Gunther, M. R., Hanna, P. M., Mason, R. P., Cohen, M. S. (1995) Hydroxylradical formation from cuprous ion and hydrogen peroxide: a spin-trappingstudy. Arch. Biochem. Biophys. 316, 515–522.

26. Prasad, T. K. (1997) Role of catalase in inducing chilling tolerance inpre-emergent maize seedlings. Plant Physiol. 114, 1369–1376.

27. Pouliot, M., Baillargeon, J., Lee, J. C., Cleland, L. G., James, M. J. (1997)Inhibition of prostaglandin endoperoxide synthase-2 expression in stimu-lated human monocytes by inhibitors of p38 mitogen-activated proteinkinase. J. Immunol. 158, 4930–4937.

28. Alessi, D. R., Cuenda, A., Cohen, P., Dudley, T., Saltiel, A. R. (1995) PD098059 is a specific inhibitor of the activation of mitogen-activated proteinkinase kinase in vitro and in vivo. J. Biol. Chem. 270, 27489–27494.

29. Cuenda, A., Rouse, Y., Doza, N., Meier, R., Cohen, P., Gallagher, T. F.,Young, P. R., Lee, J. C. (1995) SB 203580 is a specific inhibitor of a MAPkinase homologue which is stimulated by cellular stresses and interleu-kin-1. FEBS Lett. 364, 229–233.

30. Ushio-Fukai, M., Alexander, R. W., Akers, M., Griendling, K. K. (1998)p38 Mitogen-activated protein kinase is a critical component of theredox-sensitive signaling pathways activated by angiotensin II. Role invascular smooth muscle cell hypertrophy. J. Biol. Chem. 273, 15022–15029.

31. Yoshizumi, M., Abe, J., Haendeler, J., Huang, Q., Berk, B. C. (2000) Srcand Cas mediate JNK activation but not ERK1/2 and p38 kinases byreactive oxygen species. J. Biol. Chem. 275, 11706–11712.

32. Guyton, K. Z., Liu, Y., Gorospe, M., Xu, Q., Holbrook, N. J. (1996)Activation of mitogen-activated protein kinase by H2O2. Role in cellsurvival following oxidant injury. J. Biol. Chem. 271, 4138–4142.

33. Pouliot, M., Gilbert, C., Borgeat, P., Poubelle, P. E., Bourgoin, S., Cremi-non, C., Maclouf, J., McColl, S. R., Naccache, P. H. (1998) Expression and


activity of prostaglandin endoperoxide synthase-2 in agonist-activatedhuman neutrophils. FASEB J. 12, 1109–1123.

34. Ghosh, S., May, M. J., Kopp, E. B. (1998) NF-�B and Rel proteins:evolutionarily conserved mediators of immune responses. Annu. Rev.Immunol. 16, 225–260.

35. Palombella, V. J., Rando, O. J., Goldberg, A. L., Maniatis, T. (1994) Theubiquitin-proteasome pathway is required for processing the NF-�B1precursor protein and the activation of NF-�B. Cell 78, 773–785.

36. Lee, J. Y., Yu, B. P., Cheng, H. Y. (2005) Activation mechanisms ofendothelial NF-�B, IKK, and MAP kinase by tert-butyl hydroperoxide.Free Radic. Res. 39, 399–409.

37. Adderley, S. R., Fitzgerald, D. J. (1999) Oxidative damage of cardiomy-ocytes is limited by extracellular regulated kinases 1/2-mediated induc-tion of cyclooxygenase-2. J. Biol. Chem. 274, 5038–5046.

38. Profita, M., Sala, A., Bonanno, A., Riccobono, L., Siena, L., Melis, M. R.,Di Giorgi, R., Mirabella, F., Gjomarkaj, M., Bonsignore, G., Vignola, A. M.(2003) Increased prostaglandin E2 concentrations and cyclooxygenase-2expression in asthmatic subjects with sputum eosinophilia. J. Allergy Clin.Immunol. 112, 709–716.

39. Wang, T., Qin, L., Liu, B., Liu, Y., Wilson, B., Eling, T. E., Langenbach, R.,Taniura, S., Hong, J. S. (2004) Role of reactive oxygen species in LPS-inducedproduction of prostaglandin E2 in microglia. J. Neurochem. 88, 939–947.

40. Sugimoto, Y., Fukada, Y., Mori, D., Tanaka, S., Yamane, H., Okuno, Y.,Deai, K., Tsuchiya, S., Tsujimoto, G., Ichikawa, A. (2005) ProstaglandinE2 stimulates granulocyte colony-stimulating factor production via theprostanoid EP2 receptor in mouse peritoneal neutrophils. J. Immunol.175, 2606–2612.

41. Laouini, D., Elkhal, A., Yalcindag, A., Kawamoto, S., Oettgen, H., Geha,R. S. (2005) COX-2 inhibition enhances the TH2 immune response toepicutaneous sensitization. J. Allergy Clin. Immunol. 116, 390–396.

42. Simpkins, C. O., Alailima, S. T., Tate, E. A., Johnson, M. (1986) The effectof enkephalins and prostaglandins on O-2 release by neutrophils. J. Surg.Res. 41, 645–652.

43. Kunikata, T., Yamane, H., Segi, E., Matsuoka, T., Sugimoto, Y., Tanaka,S., Tanaka, H., Nagai, H., Ichikawa, A., Narumiya, S. (2005) Suppressionof allergic inflammation by the prostaglandin E receptor subtype EP3. Nat.Immunol. 6, 524–531.

44. Burgess, J. K., Ge, Q., Boustany, S., Black, J. L., Johnson, P. R. (2004)Increased sensitivity of asthmatic airway smooth muscle cells to prosta-glandin E2 might be mediated by increased numbers of E-prostanoidreceptors. J. Allergy Clin. Immunol. 113, 876–881.

45. Huszar, E., Szabo, Z., Jakab, A., Barta, I., Herjavecz, I., Horvath, I. (2005)Comparative measurement of thromboxane A2 metabolites in exhaledbreath condensate by different immunoassays. Inflamm. Res. 54, 350–355.

46. Gounni, A. S., Lamkhioued, B., Koussih, L., Ra, C., Renzi, P. M., Hamid,Q. (2001) Human neutrophils express the high-affinity receptor for immu-noglobulin E (FcεRI): role in asthma. FASEB J. 15, 940–949.

47. Thelen, M., Dewald, B., Baggiolini, M. (1993) Neutrophil signal transduc-tion and activation of the respiratory burst. Physiol. Rev. 73, 797–821.

48. Halliwell, B., Gutteridge, J. M. C. (1987) Free Radicals in Biology andMedicine, Oxford, UK, Clarendon, 18–19.

49. Teramoto, S., Shu, C. Y., Ouchi, Y., Fukuchi, Y. (1996) Increased spon-taneous production and generation of superoxide anion by blood neutro-phils in patients with asthma. J. Asthma 33, 149–155.


Modulation of IgE-dependent COX-2 gene expression by reactive oxygen species in human neutrophils

Documents

Transcript of Modulation of IgE-dependent COX-2 gene expression by reactive oxygen species in human neutrophils