The House Dust Mite Allergen Der p 1, Unlike Der p 3, Stimulates the Expression of Interleukin-8 in Human Airway Epithelial Cells via a Proteinase-activated Receptor-2-independent Mechanism*

We investigated and compared the mechanisms by which two dust mite proteolytic allergens, Der p 1 and Der p 3, and a peptide agonist of proteinase-activated receptor 2 (PAR2AP) trigger interleukin (IL)-8 release from human pulmonary epithelial cells (A549). Although all three stimuli tested induced the up-regulation of IL-8 (mRNA and protein), the Der p 1-mediated signaling events did not exactly match those induced by PAR2AP and Der p 3. First, Der p 1 was less effective in stimulating IL-8 gene transcriptional activity than PAR2AP and Der p 3. Second, Der p 1-mediated IL-8 expression was mainly dependent on NF-κB, whereas Der p 3 and PAR2AP regulated IL-8 expression through the activation of both NF-κB and AP-1. Third, although all three MAP kinases, ERK1/2, p38, and JNK, were activated, Der p 1 induced IL-8 release exclusively via the ERK1/2 signaling pathway, whereas PAR2AP and Der p 3 also involved the other kinases. Fourth, in HeLa cells, Der p 1 was able to up-regulate IL-8 secretion independent of PAR2 expression, and in contrast with PAR2AP and Der p 3, Der p 1 was unable to affect calcium signaling via PAR2 in PAR2-expressing KNRK cells. Finally, cleavage by Der p 1 of a synthetic peptide representing the N-terminal activation-cleavage site of PAR2 did not release a high potency activator of PAR2 as does Der p 3. We conclude that Der p 1 (but not Der p 3)-induced IL-8 production in A549 epithelial cells is independent of PAR2 activation.

House dust mites are a major source of allergens that contribute to the rising incidence of allergic diseases such as bronchial asthma, perennial rhinitis, and atopic dermatitis (1)(2)(3). At least 17 allergen groups have been identified in the two predominant dust mite species, Dermatophagoides pteronyssinus and Dermatophagoides farinae (4). Allergens from group 1 (Der p 1 and Der f 1) display cysteine proteinase activity, whereas group 3, 6, and 9 allergens are serine proteinases that share sequence identity with trypsin, chymotrypsin, and collagenase, respectively. Although the precise pathophysiological role of the proteolytic activity of these house dust mite allergens has not been fully elucidated in vivo, recent data provide evidence that the proteinase activity amplifies allergen-induced bronchial asthma. Indeed, several reports have demonstrated that Der p 1 is capable of cleaving human proteins with potentially immunomodulatory effects, including A1-Pi (␣1-proteinase inhibitor) (5), CD23 (the human low affinity IgE receptor) (6,7), CD25 (the ␣ subunit of the human IL-2 receptor) (8), and CD40 (9). Moreover, incubation of confluent airway epithelial cells with Der p 1 or the serine proteinases Der p 3, Der p 6, or Der p 9 has been shown to lead to cleavage of the tight junction adhesion proteins, such as ZO-1, causing an increase in the epithelial permeability and access of allergens to dendritic cells (10 -13). This event appears to depend on the proteolytic activity of these allergens, since these changes in allergeninduced permeability can be prevented by cysteine or serine proteinase inhibitors. Finally, several studies indicate that Der p 1, as well as Der p 3 and Der p 9, can also induce the release of proinflammatory mediators, such as interleukin (IL) 2 -8, IL-6, eotaxin, granulocyte macrophage-colony stimulating factor (GM-CSF), or RANTES by airway epithelial cells (14 -17). These mediators trigger the accumulation of inflammatory cells, such as eosinophils and neutrophils, to perpetuate the chronic allergic inflammation of the airways. The expression of these cytokines and chemokines has been shown to depend on the enzymatic activity of these dust mite allergens, but the allergen-induced signal transduction mechanism(s) in human epithelial cells has not yet been studied in any depth. The relationship between the proteolytic activity of these allergens and their induction of chemokine production has been strengthened by data pointing to a potential role for proteinase-activated receptors (PARs) 1 and 2 in the action of Der p 1, 3, and 9 (15,17). PAR 1 and PAR 2 belong to a new subfamily of proteolytically activated G proteincoupled receptors with seven putative transmembrane domains. Activation of PARs is achieved when the extracellular amino terminus of the receptor is cleaved to expose a tethered ligand sequence that binds to the extracellular body of the receptor, leading to the G protein-coupled signal transduction (e.g. activation of phospholipase C, generation of inositol 1,4,5-triphosphate, increased intracellular Ca 2ϩ , and activation of protein kinase C (reviewed in Ref. 18)). Synthetic peptides based on the sequence of the tethered ligand can bind to and activate the receptor, bypassing the requirement for proteolysis (19).
The chemokine IL-8, which is a major chemoattractant and activator of neutrophils, appears to play a key role in the pathophysiology of asthma (20,21). Indeed, IL-8 expression is increased in the airway epithelium of patients with asthma (22). Patients with status asthmaticus also exhibit dramatic increases of this cytokine in their airways (23). Moreover, IL-8 has also been found to be an eosinophil chemoattractant relevant to the problem of asthma (21). IL-8 secretion and mRNA synthesis are strongly regulated by more than 100-fold by a plethora of external stimuli (24). A major part of this regulation occurs at the transcriptional level. A relatively small region of the 5Ј-flanking region of the IL-8 gene (nucleotides Ϫ1 to Ϫ133) is essential and sufficient for its transcriptional induction by most stimuli (24). This promoter contains an NF-B element that is required for activation in all cell types studied as well as AP-1 and CCAAT enhancer-binding protein (NF-IL6) binding sites. The latter two sites are not essential for induction but are required for maximal gene expression (24). To date, numerous lines of functional evidence indicate that the activity of transcription factors that can bind to NF-B, AP-1, and CCAAT/enhancer-binding protein (NF-IL-6) binding sites in the IL-8 promoter are regulated by the coordinated activation of the MAP kinases ERK, c-Jun NH 2 -terminal kinase (JNK), and p38 MAP kinase (p38 MAPK) in response to stimuli that induce IL-8 production (reviewed in Ref. 25). However, the precise molecular mechanisms by which all three types of MAP kinases regulate onset, duration, or extent of IL-8 mRNA synthesis are still not completely elucidated.
In this study, we have investigated and compared the mechanisms whereby Der p 1, Der p 3 and an agonist peptide-activator of PAR 2 (PAR 2 AP) induce the expression and the production of IL-8 in human airway epithelial cells. Our data demonstrate that, although Der p 1 and Der p 3 have both been described to trigger cytokine release from respiratory epithelial cells by activation of PAR 2 , the signaling events mediated by Der p 1 do not exactly match those induced by either Der p 3 or PAR 2 activation in terms of the transcription factors and MAP kinases mediating IL-8 expression. Moreover, we show that the ability of Der p 1 to up-regulate IL-8 secretion in HeLa cells, in contrast with Der p 3, cannot be attributed to the activation of PAR 2 . Finally, we find that cleavage by Der p 1 of a synthetic peptide sequence representing the N-terminal activation-cleavage site of PAR 2 does not release a high potency activator of PAR 2 , as does Der p 3. All of our data lead us to conclude that, in contrast with Der p 3, Der p 1-induced IL-8 production in human airway epithelial cells appears to be independent of PAR 2 activation.

EXPERIMENTAL PROCEDURES
Materials-The receptor-selective PAR 2 agonist (PAR 2 AP; SLIGRL-NH 2 ) and the standard PAR 2 -inactive control (␣-PAR 2 AP; LRGILS-NH 2 ) peptide, P20 (corresponding to rat PAR 2 residues 30 -45, 30 GPNSKGRSLIGRLDTP 45 yggc (sequence of PAR 2 in uppercase letters); yggc added for labeling and coupling) as well as a 27-mer P27 peptide (corresponding to human PAR 2 residues 28 -54, Ac- 28 GT-NRSSKGRSLIGKVDGTSHVTGKGVT 54 -NH 2 ) were synthesized by solid-phase methods (Peptide Synthesis Facility, University of Calgary, Faculty of Medicine). HPLC analysis, mass spectral analysis, and quan-titative amino acid analysis confirmed the composition and purity of all peptides. Stock solutions were prepared in 25 mM HEPES buffer, pH 7.4. Trypsin and TNF-␣ were purchased from Sigma and R&D, respectively. The MEK1/2 inhibitor U0126 was obtained from Cell Signaling. The p38 MAP kinase inhibitor, SB203580, the JNK inhibitor, SP 6001125, and actinomycin D were purchased from Calbiochem.
Der p Allergen Preparation-Natural Der p 1 was isolated from whole D. pteronyssinus culture extracts by immunoaffinity chromatography using immobilized 4C1 anti-Der p 1 monoclonal antibody (Indoor Biotechnologies). To obtain Der p 3, the flow-through of the Der p 1 affinity chromatography column was submitted to a 60% final saturation (NH 4 ) 2 SO 4 precipitation. After ultracentrifugation (45 min, 149,000 ϫ g), the supernatant was applied to a benzamidine-Sepharose 4 fast flow column (Amersham Biosciences) equilibrated with Tris buffer (50 mM), NaCl (0.5 M), pH 7.4. Der p 3 was eluted from the column with 50 mM glycine buffer, pH 2.5, and each 1-ml fraction was immediately equilibrated at pH 7.5 with 75 l of 1 M Tris buffer, pH 9.5. Fractions containing Der p 3 were pooled, concentrated (Amicon-Millipore regenerated cellulose ultrafiltration membranes, nominal molecular weight limit, 10 kDa), and applied on a Superdex 75 gel filtration chromatography column (Amersham Biosciences) equilibrated with PBS, pH 7.3. Purified Der p 3 was concentrated and stored at Ϫ20°C. The purity of the natural allergens was estimated by SDS-PAGE and immunoblot analyses. The proteins were detected after Coomassie Brilliant Blue staining, and murine polyclonal antibodies directed to recombinant ProDer p 1 and ProDer p 3 were used to verify the absence of cross-contamination (Fig.  1). The authenticity of Der p 1 and Der p 3 was confirmed by N-terminal sequencing and mass spectrometry.
Recombinant ProDer p 1 from CHO spent culture medium was purified to homogeneity as previously described (26). Before use, natural Der p 1 was activated for 15 min at room temperature in PBS, pH 7.3, containing 10 mM L-cysteine. In some experiments, Der p 1 was heatdenatured for 5 min at 100°C in the presence of 50 mM ␤-mercaptoethanol. The catalytic activity of Der p 1 was determined in a continuous rate (kinetic) assay using the fluorogenic peptide substrate N-tertbutoxycarbonyl-Gln-Ala-Arg-7-amido-4-methylcoumarin (Boc-Gln-Ala-Arg-7-amino-4-methylcoumarin (AMC)). Briefly, Der p 1 was activated for 15 min in 20 mM Hepes buffer, pH 7.4, containing 150 mM NaCl, 3 mM KCl, 1.5 mM CaCl 2 , 10 mM dextrose, 0.25 mM sulfinpyrazone, and 10 mM cysteine. The substrate Boc-Gln-Ala-Arg-AMC (20 l, 7.5 mM stock in Me 2 SO, 75 M final concentration; Bachem, Torrance, CA) was then added. The rate of the enzymatic activity was measured for 10 min using an AMINCO-Bowman Series 2 Luminescence spectrometer (Spectronic Unicam, Rochester, NY) with an excitation wavelength of 380 nm and an emission recorded at 460 nm. The amount of substrate cleaved was determined by measuring the fluorescence of known concentrations of AMC (0.75, 1.5, and 2.25 M; Sigma). The activity of the Der p 1 used for our work (house units/mg) ranged between 200 and 300 units/mg (nmol of Boc-Gln-Ala-Arg-AMC cleaved/min/mg of protein). Sample calculation was as follows: activity ϭ (0.00536697 fluorescence units/s ϫ (2.25 nmol/ml)/(2.41545 fluorescence units) ϫ 2 ml ϫ 60 s/min)/0.0021 mg of Der p 1; activity ϭ 282 nmol of Boc-Gln-Ala-Arg-AMC cleaved/min/mg of protein. A molecular weight of 30,000 was used to calculate the molar concentrations of Der p 1.
Cell Culture-A549 cells, a human alveolar type II pneumocyte cell line (a kind gift from Prof. J. Pestel, Institut Pasteur, Lille) and HeLa cells (obtained from the European Collection of Cell Culture) were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum, 2 mM L-glutamine, 100 units/ml penicillin, and 100 g/ml streptomycin. Cells were passed without the use of trypsin by scraping to minimize the proteolytic activation of the PARs. HeLa cells were transiently transfected with human PAR 2 using the pcDNA3 mammalian expression vector, as outlined below.
Activation of A549 and HeLa Cells-A549 and either vector-alone or PAR 2 -transfected HeLa cells were cultured to 80% confluence and then incubated in serum-free DMEM for a further 24 h. One hour before stimulation, the growth medium was replaced by fresh serum-free DMEM.
Cells were then incubated with different concentrations of Der p 1 (25-250 nM) or Der p 3 (5-60 nM) for 7 h. In some experiments, cells were then exposed to optimum concentrations of native Der p 1 or Der p 3 for different periods of time in serum-free DMEM. As controls, cells were also incubated with activating peptide (SLIGRL-NH 2 ) or the standard PAR 2 -inactive reverse sequence peptide (␣PAR 2 ; LRGILS-NH 2 ) (final concentration 25 M), trypsin (final concentration 10 nM), TNF-␣ (final concentration 10 ng/ml), heat-denatured Der p 1, recombinant ProDer p 1 (allergens used at a final concentration of 150 nM), or native Der p 3 treated with the serine proteinase inhibitor 4-(2-aminoethyl)benzenesulfonyl fluoride or aprotinin. When appropriate, A549 cells were treated for 1 h at 37°C with 25 M of the specific MEK inhibitor U0126 (Cell Signaling) or the specific p38 MAP kinase or JNK kinase inhibitors (SB203580 and SP6001125, respectively; Calbiochem) prior to the addition of Der p 1, Der p 3, or positive controls. Negative control cells were preincubated with equivalent amounts of Me 2 SO. Culture supernatants were collected, centrifuged for 5 min at 10,000 ϫ g, and stored at Ϫ80°C. In another set of experiments, a 1-h activation of A549 cells with Der p 1, Der p 3, or PAR 2 AP was followed by incubation with actinomycin D (10 g/ml) for 2 h at 37°C.
Measurement of IL-8 Secretion-After stimulation, cell-free supernatants were collected by centrifugation at 400 ϫ g for 5 min and assayed for IL-8 with enzyme-linked immunosorbent assay (ELISA) kits (Pharmingen) according to the manufacturer's protocol. IL-8 concen-trations were determined by interpolation from a standard curve performed with purified human IL-8.
Detection of IL-8 mRNA Expression by Reverse Transcription (RT)-PCR-Total RNA was extracted from cultured A549 cells using TriZol reagent (Invitrogen) followed by DNase treatment (RNase-free DNase; Invitrogen). cDNA was prepared with Superscript reverse transcriptase (Invitrogen). Amplification of IL-8 transcripts was accomplished with primers chosen according to the published sequence of human IL-8 cDNA (29). The housekeeping gene fragment of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used for verification of equal loading and of RT-PCR efficiency. The sequences of the primers (Invitrogen) were as follows: IL-8, sense 5Ј-TTGGCAGCCT-TCCTGATT-3Ј and antisense 5Ј-AACTTCTCCACAACCCTCTG-3Ј (PCR product, 203 bp); GAPDH, sense 5Ј-ACCACAGTCCATGC-CATCAC-3Ј and antisense 5Ј-ATGTCGTTGTCCCACCACCT-3Ј (PCR product, 405 bp). The PCRs were performed in a DNA thermal cycler (PerkinElmer Life Sciences 9600) with the following conditions: 25 cycles of denaturation (94°C for 1 min), annealing (55°C for 1 min), extension (72°C for 1 min), and final extension for 7 min at 72°C for GAPDH and IL-8. The intensity of the amplification bands was quantified with the NIH-Image program (version 1.62).
Transient Transfection and Luciferase Assay-The vectors encoding either the wild-type IL-8 promoter (Ϫ133/ϩ44)/luciferase reporter, or site mutations of one of the binding sites (AP-1, NF-IL-6, or NF-B) in the IL-8 promoter region were kindly provided by Dr. N. Mukaida (Division of Molecular Bioregulation, Cancer Research Institute, Kanazawa University, Japan). The constructs were described in detail in the publications from Mukaida's laboratory (30). The AP-1-luc construct, driven by 12-O-tetradecanoylphorbol-13-acetate-responsive elements within the 517-bp promoter of the human collagenase gene (MMP-1luc) (31), and the NF-B-luc reporter plasmid, containing the herpes simplex virus thymidine kinase minimal promoter and four copies of the IL-2 promoter B site, were kindly provided by Dr. P. Angel (German Cancer Research Center, Heidelberg, Germany) and Dr. C. Van Lint (Institut de Biologie et de Médecine Moléculaires, Université Libre de Bruxelles, Gosselies, Belgium), respectively. A549 cells were transfected using Fugene (Roche Applied Science) with one of these vectors for 4 h and then cultured in serum-free medium for a further 16 h before the Der p 1, Der p 3, or PAR 2 AP treatment for 60 min. The cells were then collected and lysed. Supernatants were evaluated for firefly luciferase activity with a luminometer. Luciferase activities were normalized with respect to protein concentration using the detergent-compatible protein assay (Bio-Rad).
Western Blotting Analysis-Treated A549 cells were washed in cold PBS, lysed with Laemmli sample buffer (60 mM Tris-HCl (pH 7.5), 2% SDS, 5% glycerol, 100 mM dithiothreitol, 0.001% bromphenol blue), and analyzed by SDS-PAGE and Western blotting with polyclonal antibodies specific for both phosphorylated and nonphosphorylated forms of ERK1/2, p38 MAP kinase, or JNK (1:1000 dilution; Cell Signaling) as described previously (41) as well as with a polyclonal antibody to IB␣ (C-21) and p50 NF-B subunits (Santa Cruz Biotechnology, Inc., Santa Cruz, CA). The intensities of the bands on the immunoblot were quantified with the NIH-Image program (version 1.62).
Nuclear Extract Preparation-Nuclear extracts from A549 cells were prepared as previously described (32). Briefly, treated A549 cells were scraped and washed in cold phosphate-buffered saline and then resuspended in 500 l of cold buffer A (10 mM HEPES (pH 7.9), 10 mM KCl, 1.5 mM MgCl 2 , 0.5 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride) supplemented with proteinase inhibitors (Complete; Roche Applied Science). The cells, swollen at 4°C for 10 min, were lysed in the presence of 0.4% Nonidet P-40. The homogenate was centrifuged for 5 min at 600 ϫ g, and the nuclear pellet was resuspended in 30 l of cold buffer C (20 mM HEPES (pH 7.9), 25% glycerol, 0.42 M NaCl, 1.5 mM MgCl 2 , 0.2 mM EDTA, 0.5 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride) supplemented with proteinase inhibitors (Complete; Roche Applied Science) and left for 30 min on ice. After centrifugation (20,000 ϫ g for 15 min at 4°C), the resulting supernatant was diluted by the addition of 150 l of cold buffer D (20 mM HEPES (pH 7.9), 20% glycerol, 50 mM KCl, 0.2 mM EDTA, 0.5 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride) supplemented with proteinase inhibitors (Complete; Roche Applied Science) and then stored in small aliquots at Ϫ80°C. Protein concentrations were determined by the Bradford method (Bio-Rad protein assay) with bovine serum albumin as a standard.
Electrophoretic Mobility Shift Assays-Electrophoretic mobility shift assays (EMSAs) were performed as previously described (41). Briefly, binding sites for double-stranded oligonucleotides, NF-B (5Ј-CGTG-GAATTTCCTCTG-3Ј; bp Ϫ83 to Ϫ68) and AP-1 (5Ј-GTGATGACT-CAGGTT-3Ј; bp Ϫ130 to Ϫ116) from the IL-8 promoter were endlabeled with [␥-32 P]ATP using T4 polynucleotide kinase (Invitrogen) and were used as probes. The binding reactions were performed at room temperature. Nuclear extract (10 g of protein) was first incubated for 20 min in the absence of probe in a 16-l reaction mixture containing 10 g of DNase-free bovine serum albumin (Amersham Biosciences), 1 g of poly(dI-dC) (Amersham Biosciences) as nonspecific competitor DNA, 1 mM dithiothreitol, 20 mM Tris-HCl (pH 7.5), 60 mM KCl, 1 mM MgCl 2 , 0.1 mM EDTA, and 10% (v/v) glycerol. Thirty thousand counts/ min of probe was then added to the reaction mixture, and the mixture was incubated for 20 min. Samples were subjected to electrophoresis at room temperature on 6% polyacrylamide gels at 150 V for 2-3 h in 1ϫ TGE buffer (25 mM Tris acetate (pH 8.3), 190 mM glycine, 1 mM EDTA). Gels were dried and autoradiographed at Ϫ70°C. The intensities of the bands on the immunoblot were quantified with the NIH-Image program (version 1.62).
PCR Cloning of the Human PAR 2 cDNA-Human PAR 2 cDNA was prepared by reverse transcription and PCR amplification of total RNA from A549 cells. PCR primer design was based on the published human PAR 2 sequence (33). The primer sequences were as follows: forward primer, 5Ј-AAGCTTCCACCATGCGGAGCCCCAGCGCGGCG-3Ј (containing a HindIII site and Kozak sequence shown in boldface type); reverse primer, 5Ј-GCGGCCGCTCAATAGGAGGTCTTAACAGT-GGT-3Ј (NotI shown in boldface type). The PCR product was digested with HindIII and NotI, cloned into pcDNA3 vector (Invitrogen), and sequenced.
Transfection of HeLa Cells-HeLa cells were seeded into 6-well plates at a density of 2.5 ϫ 10 5 cells/well and, on the following day, transfected with 500 ng of DNA using FuGENE TM -6 (Roche Applied Science) according to the manufacturer's protocol. At 24 h post-transfection, the cells were harvested using nonenzymatic cell dissociation solution and then seeded onto 12-well dishes or glass coverslips. On the next day, the medium of transfected cell monolayers was replaced with serum-free DMEM for 24 h before any treatment.
Immunofluorescence Confocal Microscopy-Transfected HeLa cells were grown on glass coverslips as described above. After treatments, cells were washed with PBS and fixed in 4% paraformaldehyde for 10 min at room temperature. The samples were saturated with PBS containing 0.5% gelatin, 0.25% bovine serum albumin, and 125 mM glycine for 1 h and stained for 1 h at room temperature with a 1:200 dilution of the anti-PAR 2 rabbit polyclonal antibody B5 (27). The samples were then washed three times with PBS containing 0.2% gelatin and incubated for 1 h with a 1:300 dilution of the secondary antibody: Alexa 546-coupled goat anti-rabbit immunoglobulin G (Molecular Probes, Inc., Eugene, OR). The samples were then washed three times in PBS with 0.2% gelatin and mounted for analysis on a Zeiss LSM510 laserscanning confocal microscope.
Calcium Signaling Assay-Changes in intracellular calcium concentration induced by Der p 1, Der p 3, trypsin, or PAR 2 AP in rat PAR 2 -or human PAR 2 N30A-transfected KNRK cells were measured with the intracellular calcium indicator fluo-3 acetoxymethyl ester (Molecular Probes) as previously described (34,35). Briefly, cells grown to 95% confluence were harvested without the use of proteinase, pelleted by centrifugation, resuspended in DMEM, and loaded with fluo-3 acetoxymethyl ester at a final concentration of 22 M (25 g/ml). Indicator uptake was allowed to proceed for 25 min at room temperature in the presence of 0.4 mM sulfinpyrazone, after which the cells were washed two times by centrifugation and resuspended in the following buffer: NaCl (150 mM), KCl (3 mM), CaCl 2 (1.5 mM), HEPES (20 mM), dextrose (10 mM), sulfinpyrazone (0.25 mM), and L-cysteine (10 mM), pH 7.4. Fluorescence measurements, reflecting elevations of intracellular calcium, were conducted at 24°C using an AMINCO-Bowman Series 2 Luminescence Spectrometer (Spectronic Unicam, Rochester, NY) with an excitation wavelength of 480 nm and an emission recorded at 530 nm. The fluorescence signals caused by the addition of test agonists (Der p 1, trypsin, or PAR 2 AP) were compared with the fluorescence peak height yielded by replicate cell suspensions treated with 2 M ionophore A23187 (Sigma). This concentration of A23187 was at the plateau of its concentration-response curve for a fluorescence response. To determine if the peptide(s) released from the PAR 2 cleavage/activation sequence (P20) of PAR 2 caused a calcium signal via the activation of PAR 2 , a cross-desensitizaton protocol was used (35), wherein a prior desensitization of PAR 2 by exposure of cells twice to a maximally active concentration of the PAR 2 -AP (SLIGRL-NH 2 , 50 M) abolishes the calcium signal caused by a subsequent addition of a putative PAR 2 -activating agonist but not by an agonist acting via a receptor other than PAR 2 . In this cross-desensitization protocol, at least a 5-min time period is allowed between the sequential addition of test compounds to the cell suspension to allow for the refilling of intracellular calcium stores (35).
Proteolysis by Der p 1, Der p 3, and Trypsin of Peptide Sequences Based on the Cleavage/Activation Domain of PAR 2 -Der p 1 (700 nM) was first activated as described for the assay of enzyme activity (above) and was then used to hydrolyze (25 min at room temperature) the peptide, P20 ( 30 GPNSKGRSLIGRLDTP 45 yggc; 50 M), representing the cleavage/ activation sequence (boldface type) of rat PAR 2 . The P20 hydrolysate was tested for the release of PAR 2 -activating sequences in two ways: (a) a PAR 2 cell reporter assay and (b) HPLC and mass spectral analysis. For the PAR 2 cell reporter assay, the P20 hydrolysate was added directly to an equal volume of a test sample of fluo-3-loaded rat PAR 2 -expressing KNRK cells in a calcium signaling assay (above). The presence of a calcium signal caused by the hydrolysate provided evidence for the release of a PAR 2 -activating peptide from P20. For the chemical analysis, the P20 hydrolysate was subjected to HPLC analysis using a trifluoroacetic acid gradient in acetonitrile, followed by mass spectral analysis of the separated peptides. In another set of experiments, a 27-mer P27 peptide representing the cleavage/activation sequence of human PAR 2 ( 28 GTNRSSKGRSLIGKVDGTSHVTGKGVT 54 -amide) was incubated at room temperature for 25 min with Der p 3 or trypsin (25 nM each) prior to HPLC analysis. Peak HPLC fractions, representing the peptides released from P20 by Der p 1 (50 M P20, 10 nM enzyme, 10 min at room temperature) or trypsin and from P27 by Der p 3 or trypsin were collected, freeze-dried, and subjected to mass spectral analysis (matrixassisted laser desorption ionization). Peptide sequences resulting from Der p 1, Der p 3, and trypsin hydrolysis were deconvoluted by analysis of the masses determined for each hydrolysis fragment isolated by HPLC. Since intact P20 alone does not yield a calcium signal via PAR 2 (28) and since in the P20 sequence, only peptides beginning with the SL . . . motif are capable of activating PAR 2 , any signal generated by the hydrolysis of P20 by Der p 1 must come from a cleavage of the Arg 36 -Ser 37 bond.
Statistical Analysis-Group data were compared using Student's t test. Differences were considered significant for p values of Ͻ0.05.

Der p 1 and Der p 3 Induce IL-8 Expression in A549
Cells-To determine whether Der p 1 and Der p 3 could enhance the synthesis of IL-8 protein in bronchoalveolar cells, A549 epithelial cells were cultured with various concentrations of Der p 1 and Der p 3 for 7 h, and IL-8 levels were measured in the supernatants by ELISA. Fig. 2 shows that activated Der p 1 and Der p 3 caused IL-8 secretion from A549 cells in a concentration-dependent manner when compared with control medium (p Ͻ 0.05). Der p 3 was shown to be more effective than Der p 1 at stimulating IL-8, since maximum responses were observed with 50 nM Der p 3 and 150 nM Der p 1, respectively (p Ͻ 0.05). Heat-inactivated Der p 1 and Der p 3 (65°C for 30 min), proteolytically inactive Der p 1 precursor (ProDer p 1), and Der p 3 preincubated with the irreversible serine proteinase inhibitor 4-(2-aminoethyl)benzenesulfonyl fluoride were unable to cause a significant up-regulation of IL-8 production, compared with untreated cells (data not shown), confirming that the allergen-induced IL-8 expression was dependent on the proteolytic activity of Der p 1 and Der p 3.
Given that chemokine expression is regulated primarily at the gene transcription level and to establish the time dependence of the IL-8 expression, we investigated the direct effect of both Der p 1 and Der p 3 on IL-8 mRNA induction. To this end, A549 cells were incubated for different periods of time with optimum concentrations of activated Der p 1 and Der p 3 (150 and 50 nM, respectively). These allergen amounts did not cause any shrinking or detachment of cells and did not reduce viability (trypan blue dye exclusion) (data not shown). Because a recent study suggested that both Der p 1 and Der p 3 might act via PAR 2 (15,17), experiments were also conducted with an agonist for PAR 2 , the activating peptide, SLIGRL-NH 2 . As shown in Fig. 3, a slight up-regulation in IL-8 mRNA levels was observed 30 min after Der p 1 treatment, reaching a plateau after 1 h and then gradually declining after 6 h. In contrast, both Der p 3 and the PAR 2 AP, SLIGRL-NH 2 , induced a more pronounced IL-8 mRNA expression that was maximal within 1 h of treatment and thereafter declined slowly to return to basal levels after 4 h (p Ͻ 0.05) (Fig. 3A). As expected, there was no difference in the mRNA level of a constitutively expressed GAPDH gene over the time period tested for all of these three stimuli. Concomitantly, we measured the IL-8 production in the culture supernatants of A549 after the application either of Der p 1, Der p 3, or PAR 2 AP. As shown in Fig. 3B, the release of IL-8 protein from Der p 1-, Der p 3-, or PAR 2 AP-treated A549 cells also increased in a time-dependent manner. Interestingly, Der p 3  To reveal whether the induction results from IL-8 gene transcription or post-transcriptional stabilization of IL-8 mRNA, we carried out transcriptional blockade studies. The RNA polymerase II inhibitor actinomycin D (10 g/ml) was added to A549 cell culture 1 h after Der p 1, Der p 3, or PAR 2 AP stimulation (i.e. the maximal IL-8 mRNA level induction time). After a 2-h actinomycin D treatment, the IL-8 mRNA levels were compared by RT-PCR. The IL-8 levels were also measured in the supernatants by ELISA after 17 h. The addition of actinomycin D to A549 cell cultures (after a 1-h treatment with Der p 1, Der p 3, or PAR 2 AP) reduced drastically each of the agonist-induced increases in IL-8 mRNA as well as the protein level (Fig. 4, A and B). These results thus suggest that Der p 1 and Der p 3 as well as PAR 2 AP stimulate expression of IL-8 mRNA mainly through transcriptional activation of the IL-8 promoter. However, this result does not exclude the possibility that post-transcriptional modification of IL-8 transcripts may also contribute to the effect of those stimuli on IL-8 mRNA levels.
AP-1 and NF-B Cooperate to Yield Maximal IL-8 Induction by Der p 1 or Der p 3-Many studies have revealed that a sequence spanning nucleotides Ϫ1 to Ϫ133 within the 5Ј-flanking region of the IL-8 gene is essential and sufficient for transcriptional regulation of the gene (36 -38). This region contains three cis elements important for the induction of IL-8 gene expression: AP-1 (bp Ϫ126 to 120), NF-IL6 (bp Ϫ94 to Ϫ81), and NF-B (bp Ϫ80 to Ϫ70) binding site (39,40). To confirm that Der p 1, Der p 3, and PAR 2 AP were acting transcriptionally and to identify the transcription binding sites that are important for IL-8 induction, luciferase reporter constructs containing the minimally essential wild-type IL-8 promoter region (bp Ϫ133 to ϩ44) or mutated at the level of the binding sites for either AP-1, NF-IL-6, or NF-B, were transfected into A549 cells (Fig. 5A). The point mutations made in each of these cis elements eliminate the ability to bind to their corresponding transcription factors (37). Transfected cells were assayed for luciferase activity in response to Der p 1, Der p 3, PAR 2 AP, and trypsin, a well known PAR 2 agonist (33). TNF-␣, a known potent inducer of IL-8 via the B sites was also used as a positive control of IL-8 induction via the B sites (41).
As shown in Fig. 5B, all of the stimuli tested increased the luciferase activity in cells transfected with the wild-type IL-8 promoter, suggesting that Der p 1 and Der p 3 as well as the enzymatic (trypsin) or the nonenzymatic (PAR 2 AP) activator of PAR 2 were acting transcriptionally. Importantly, mutation in the NF-B or AP-1 binding site (NF-B mut or AP-1 mut) resulted in a significant reduction of IL-8 gene induction by all of the tested activators (p Ͻ 0.05) (Fig. 5B). In contrast, mutation of NF-IL-6 site (NF-IL-6 mut) failed to reduce the promoter activation by Der p 1, Der p 3, PAR 2 AP, trypsin, or TNF-␣.
These data suggest that the presence of intact B and AP-1 binding sites are both necessary for maximal IL-8 induction by Der p 1 and Der p 3 as well as in response to PAR 2 activation.
Der p 3 and PAR 2 AP Increase NF-B and AP-1 Binding Complexes, whereas Der p 1 Induces Mainly the NF-B Complex-To evaluate the effects of Der p 1 and Der p 3 as well as PAR 2 AP on AP-1 and NF-B DNA binding activity, nuclear extracts were prepared from A549 cells either untreated or treated with Der p 1, Der p 3, or PAR 2 AP for different periods of time (10, 30, 60, and 120 min). A 30-min treatment of A549 cells with TNF-␣, a known activator of NF-B, was also used as a positive control for NF-B activation in A549 cells. All of these samples were subjected to EMSA analysis using a DNA probe containing either the AP-1 or the B binding site of the human IL-8 promoter (Fig. 6).
As previously reported (42), AP-1 was constitutively expressed in unstimulated A549 cells. Incubation of A549 cells with Der p 1 for various periods of time ranging from 10 to 120 min caused only a marginal increase in the DNA binding activity of AP-1 at 30 and 60 min that persisted at lower levels at 120 min (Fig. 6A). In contrast, Der p 3 rapidly and strongly affected the AP-1 binding activity, which was already increased at 10 min, gradually rose to a maximum at 1 h, and persisted for at least 2 h after cell stimulation (p Ͻ 0.05). Interestingly, a similar pattern of increase in AP-1 binding was demonstrated with A549 cells stimulated with the PAR 2 AP (Fig. 6A). The specificity of binding of AP-1 was confirmed by supershift assays. The data revealed that the AP-1 complex consists mainly of c-Fos and JunB and, to a lesser extent, JunD, c-Jun, and Fra-2 (Fig. 6B). The AP-1 binding complex was induced slightly by Der p 1 and more significantly by either Der p 3 or PAR 2 AP.
The kinetics of NF-B binding was quite different compared with AP-1 in A549 cells. Der p 1, Der p 3, and PAR 2 AP caused a comparable and significant NF-B nuclear translocation that was seen only after 30 min post-stimulation, peaked at 60 min, and persisted at lower levels at 120 min (p Ͼ 0.05) (Fig. 6A). However, TNF-␣ was consistently a more effective inducer of NF-B binding activity compared with the other three inducers. Supershift experiments using specific antibodies demonstrated that the NF-B complexes were mainly composed of p50/p65 and p52/p65 heterodimers (Fig. 6B). Since the proteolytic degradation of IB␣ subunit precedes the translocation of NF-B to the nucleus, we determined the effect of Der p 1, Der p 3, and PAR 2 AP on the protein levels of IB␣ in the cytoplasm of A549 cells treated with each of these three stimuli for times ranging from 10 to 120 min. Western blotting showed that Der p 1, Der p 3, and PAR 2 AP induced a slight decrease in the IB␣ level between 30 and 60 min and a complete restoration of the IB␣ levels after 120 min of cell stimulation according to their respective NF-B nuclear levels observed by EMSA analysis (Fig. 6, compare C and A). As expected, a complete degradation of IB␣ was induced by 30 min in response to TNF-␣.
Taken together, these results suggest that Der p 3 and PAR 2 AP activate NF-B and AP-1 binding to the IL-8 promoter, whereas Der p 1 induces mainly NF-B and very slightly AP-1 DNA binding activity.
Der p 3 and PAR 2 AP Induce NF-B and AP-1 Transcriptional Activity, whereas Der p 1 Induces Mainly NF-B Transcriptional Activity-Nuclear binding of transcription factors provides indirect evidence for their involvement in regulating IL-8 gene activation. Up to this point, we found that the presence of intact B and AP-1 binding sites is necessary for maximum IL-8 induction by Der p 1 and Der p 3 as well as for the response to PAR 2 activation. We next determined whether Der p 1, Der p 3, or PAR 2 AP can indeed stimulate AP-1 and NF-B-mediated transcriptional activity in A549 cells. To this end, A549 cells were transfected with AP-1-luc or NF-B-luc luciferase constructs. The NF-B-luc synthetic reporter construct contained four copies of the interleukin 2 promoter B sites in front of the minimal thymidine kinase promoter from the herpes simplex virus (32), whereas the AP-1-luc construct is driven by 12-O-tetradecanoylphorbol-13-acetate-responsive elements present within the 517-bp promoter of the human collagenase gene (31). Transfected cells were assayed for luciferase activity in response to Der p 1, Der p 3, PAR 2 AP, and trypsin. TNF-␣ was used as a positive control for NF-B activation.
As shown in Fig. 7, Der p 3, PAR 2 AP, and trypsin significantly induced luciferase activity in A549 cells transfected with NF-B-luc or AP-1 luc (p Ͻ 0.05). By contrast, Der p 1, like TNF-␣, was able to induce significantly NF-B-driven luciferase activity. These results corroborated our EMSA data, namely that Der p 3, trypsin, and PAR 2 AP increase the transcriptional activities of both NF-B and AP-1, whereas Der p 1 triggersmainlyNF-B-dependentgeneexpressionandveryslightlyAP-1dependent gene expression.
Taken together, our results suggest that Der p 3 and PAR 2 AP regulate the expression of the IL-8 gene through the induction and the activation of both of the transcription factors, NF-B and AP-1, whereas Der p 1-mediated IL-8 expression appears to depend mainly on the activation of NF-B in conjunction with some constitutive AP-1 activity.
Der p 1 and Der p 3 Stimulate ERK1/2, JNK, and p38 MAP Kinases in A549 Cells-Many studies indicate that IL-8 production results mainly from the activation of MAPK signaling cascades, including the ERKs, JNK, and p38 MAP kinases (reviewed in Ref. 25). To understand better the activation of NF-B and AP-1 and the subsequent production of IL-8, we next examined the effect of Der p 1, Der p 3, and PAR 2 AP on ERK, JNK, and p38 MAPK activation using a Western blot approach with phosphospecific antibodies that recognize the phosphorylated active forms of these kinases.
As shown in Fig. 8, Der p 1 induced a slightly delayed (not seen at 10 min after stimulation) but marked increase in the phosphorylation of both isoforms of ERK. The Der p 1-mediated activation of ERK1/2 was clearly evident at 30 min, was maximal at 60 min, and began to decline at 120 min. In contrast, treatment of A549 cells with Der p 3, PAR 2 AP, or trypsin (data not shown) caused a much more rapid activation of ERK1/2 than did Der p 1 (p Ͻ 0.05). ERK1/2 activation was already maximal at 10 min and persisted for up to 120 min (Fig. 8). TNF-␣ also induced a strong phosphorylation of ERK1/2 that was clearly evident as early as 10 min, was maximal at 60 min, and began to decline after 60 min. Pretreatment of cells with the specific MEK inhibitor U0126, which blocks ERK1/2 activation, fully inhibited the increased phosphorylation of ERK1/2 caused by all four agonists (data not shown). To determine whether this induction of MAP kinases was specific to the ERK pathway, all samples were also tested for the activation/phosphorylation of p38 MAPK and JNK. Interestingly, as shown in the bottom panels of Fig. 8, Der p 1, Der p 3, PAR 2 AP, and trypsin (data not shown) all induced a transient but very weak increase in the phosphorylation of p38 MAPK and JNK that could be visualized within 30 min of treatment but was no longer detected at 120 min. A much more pronounced activation of p38 MAPK and JNK was observed after TNF-␣ treatment. p38 MAPK and JNK activation induced by TNF-␣ was observed as early as 10 min and persisted up to 60 min (Fig. 8). These data indicated that, whereas Der p 1, Der p 3 and PAR 2 AP were able to activate ERK1/2, and to a lesser extent, p38 MAPK and JNK in A549 cells, a more rapid and persistent activation of ERK1/2 was induced by both Der p 3 and PAR 2 AP compared with Der p 1.
Inhibition of ERK1/2 Inhibits the Production of IL-8 in Response to Der p 1, Der p 3, and PAR 2 AP-Using specific pharmacological inhibitors, we assessed the role of ERK, p38 MAPK, and JNK activation on the production of IL-8 from A549 cells in response to Der p 1, Der p 3, or PAR 2 AP. Pretreatment of cells with U0126 (a MEK1/2 inhibitor) for 1 h

. The effect of NF-B, AP-1, and NF-IL-6 binding site mutation on the IL-8 promoter activity in response to Der p 1, Der p 3, or PAR 2 agonists.
A, A549 cells were transiently transfected with luciferase reporter constructs containing the Ϫ133 to ϩ44 fragment (WT) or its mutant forms with each of the three transcription factorbinding sites individually destroyed (NF-B mutation (NF-kB-mut), AP-1 mutation (AP-1-mut), and NF-IL-6 mutation (NF-IL6-mut)). B, cells were treated with medium alone or with Der p 1 (150 nM), Der p 3 (50 nM), PAR 2 AP (25 M), trypsin (20 nM), or TNF-␣ (10 ng/ml) for 6 h, and cellular lysates were tested for luciferase activity. RLU, relative luciferase units. Each data point is expressed as the mean Ϯ S.E. from at least three independent experiments performed in duplicate. markedly reduced the Der p 1-, Der p 3-, and PAR 2 AP-induced IL-8 production (Fig. 9). On the other hand, the blockade of p38 MAPK or JNK pathways by SB203580 (a p38 MAPK inhibitor) or SP600125 (a JNK inhibitor), respectively, failed to affect the secretion of IL-8 in response to Der p 1 but reduced the up-regulation of IL-8 caused by Der p 3 or PAR 2 AP, respectively. Importantly, pretreatment of A549 cells with each of these three inhibitors reduced the IL-8 secretion induced in response to TNF-␣ by more than 50% (data not shown). These data thus indicated that, among the three MAP kinases analyzed, ERK1/2 appears to be the only one involved in the enhanced production of IL-8 in response to Der p 1. However, for Der p 3 and PAR 2 AP, in addition to the important contribution of ERK1/2, p38 MAPK (for PAR 2 AP) and JNK (for Der p 3) also appear to play a role in the signaling pathways leading to the increased production of IL-8.
Der p 1-induced IL-8 Release May Be Independent of PAR 2 Activation-Previous work has suggested that both Der p 1 and Der p 3 trigger cytokine release from respiratory epithelial cells by activation of PAR 2 , but not PAR 1 (15,17). Although these studies demonstrated increases in intracellular calcium and cytokine secretion in response to stimulation with Der p 1 or Der p 3, a firm correlation between the cytokine secretion, the calcium flux, and the activation of PAR 2 by these two dust mite proteinases was not established. Thus, to assess unequivocally the dependence of Der p 1-and Der p 3-induced IL-8 production on PAR 2 activation, we evaluated the actions of these two proteinases on HeLa cells transfected (or not) with a PAR 2 receptor-expressing vector. It has been previously found that HeLa cells constitutively express PAR 1 but not PAR 2 (43). The expression of recombinant PAR 2 in the HeLa cells was verified by confocal microscopy using the B5 antibody specific for the cleavage/activation site of PAR 2 (27) (Fig. 10A).
As shown in Fig. 10B, in the absence of any stimulation, HeLa cells transfected with an empty vector or a PAR 2 -expressing vector expressed a comparable modest level of IL-8. Surprisingly, cell incubation with Der p 1 induced the same significant increase in the production of IL-8 irrespective of the presence or absence of PAR 2 at the cell surface. Interestingly, Der p 3 and trypsin induced the same IL-8 secretion levels in empty vector-transfected HeLa cells. However, this increase in IL-8 production was not observed in the PAR 2 -negative HeLa cells after treatment with the PAR 2 AP. In contrast, in the PAR 2 -expressing HeLa cells, there was a substantial increase in the production of IL-8 triggered by Der p 3, trypsin, and PAR 2 AP. This increase in IL-8 production was well above the levels yielded by Der p 1 treatment (Fig. 10B). Thus, the expressed PAR 2 receptors were indeed functional in terms of increased IL-8 production in response to nonenzymatic (PAR 2 AP) as well as enzymatic (Der p 3, trypsin) agonists. In sum, these results show that the dust mite proteinase Der p 3, like trypsin, triggers IL-8 release in airway epithelial cells through PAR 2 activation. On the other hand, our results also reveal that Der p 1-induced IL-8 production appears not to depend on PAR 2 activation.
Der p 1 Did Not Induce Changes in Intracellular Calcium in PAR 2expressing Cells-Previous studies have shown that PAR 2 activation causes an increase in intracellular calcium (35). Thus, monitoring changes in intracellular Ca 2ϩ provides a useful method to test the effectiveness and potency of different agonists to activate PAR 2 (27,28,35). Reliable PAR 2 calcium signaling assays have been previously developed using stably transfected PAR 2 -expressing KNRK cells (27). KNRK cells stably expressing PAR 2 were thus used to monitor changes in calcium signaling upon activation by Der p 1 and Der p 3, in comparison with trypsin and the PAR 2 AP (Fig. 11).  In the calcium mobilization assay shown in Fig. 11, Der p 1 failed to cause changes in intracellular Ca 2ϩ in stably transfected KNRK cell lines expressing either rat or human PAR 2 ( Fig. 11A and data not shown). In contrast, under the same experimental conditions, Der p 3 elicited a robust calcium signal and, as opposed to Der p 1, was able to desensitize PAR 2 (p Ͻ 0.05) (Fig. 11B) (Fig. 11, compare traces A and B). As expected, PAR 2 AP and trypsin caused increases in intracellular [Ca 2ϩ ] in PAR 2 -expressing KNRK cells (Fig. 11, A and B). In addition, PAR 2 cells exposed to receptor-desensitizing concentrations of the PAR 2 AP (50 M) before treatment with Der p 3 no longer responded to the enzyme, confirming that the calcium signal caused by Der p 3 resulted from the activation of PAR 2 (Fig. 11C).
To ascertain whether Der p 1 has the ability to "disarm" subsequent PAR 2 activation by the proteinases trypsin or Der p 3, cross-desensitization experiments were performed. It was found that pretreatment of PAR 2 -expressing KNRK cells with Der p 1 (533 nM) failed to decrease the calcium signal due to a subsequent exposure to trypsin (2 nM) or Der p 3 (5 nM) that acted via PAR 2 , as compared with cells treated only with trypsin (2 nM) or Der p 3 (5 nM) (n ϭ 3), but without prior exposure to Der p 1 (data not shown). In addition, in two independently conducted experiments, co-incubation of Der p 1 (533 nM) with Der p 3 (5 nM) for 25 min before adding the Der p 1-treated Der p 3 to PAR 2 -expressing KNRK cells did not diminish the calcium response due to Der p 3 (5 nM) activation of PAR 2 ; the signal was virtually the same as that for cells treated with Der p 3 that had not previously been exposed to Der p 1 (5 nM) (data not shown). Thus, not only was Der p 1 incapable of signaling via PAR 2 in these cells, but, in an intact cell system, this cysteine proteinase was not able either to disarm PAR 2 for Der p 3 activation or to inactivate Der p 3 enzymatic ability to trigger PAR 2 signaling in these cells.
To determine whether glycosylation of the N-terminal sequence of PAR 2 just prior to the cleavage/activation site might explain the inability of Der p 1 to activate PAR 2 , as previously shown for tryptase (28), calcium mobilization assays were performed using KNRK cells expressing a mutated glycosylation-deficient human PAR 2 receptor, where asparagine 30 at the N-terminal glycosylation site was substituted with alanine (hPAR 2 N30A KNRK cells) in order to disrupt the glycosylation sequon. Although both PAR 2 AP and trypsin caused a prompt increase in intracellular calcium in the hPAR 2 N30A cells, no calcium response was induced by Der p 1, indicating that the absence of PAR 2 glycosylation did not enable Der p 1 to activate PAR 2 (Fig. 11D).
Der p 1 and Der p 3 Cleavage Sites of PAR 2 -derived Cleavage/Activation Sequences-Since the calcium signaling experiments with PAR 2expressing cells failed to demonstrate any appreciable activation of PAR 2 with Der p 1, we wondered if the enzyme was indeed capable of hydrolyzing the N-terminal sequence of PAR 2 . Therefore, we used both a biochemical and a functional PAR 2 cell reporter calcium signaling assay to evaluate the ability of Der p 1 to cleave 20 (P20) (GPNSKGRS-LIGRLDTPyggc) and 27 (P27) (GTNRSSKGRSLIGKVDGTSHVT-GKGVT) amino acid peptides containing the sequence corresponding to the cleavage/activation site of rat and human PAR 2 , respectively. Upon enzymatic hydrolysis, the resulting cleavage products were characterized by HPLC and mass spectral analysis and compared with those peptides obtained after trypsin treatment.
Whereas trypsin (10 nM) cleaved Ͼ90% of rat P20 (50 M) during a 10-min incubation period at room temperature, Der p 1 (700 nM) cleaved ϳ60% of P20 upon incubation for 25 min at room temperature. The only detectable hydrolysis product made by trypsin corresponded to SLIGRLDTPYGGC, a sequence that can, on its own, activate PAR 2 . This relatively potent PAR 2 -activating hydrolysis product was not identified as one of the proteolysis products yielded by Der p 1. However, Der p 1 hydrolysis did yield minor amounts of the shorter peptide SLIGR (Fig. 12A). In previous work, we had determined that the  sequence SLIGR is capable of activating PAR 2 (44) but with very low potency. Thus, upon hydrolysis of P20, Der p 1 was not, in contrast with trypsin, capable of producing a peptide that can activate PAR 2 with high potency. Interestingly, the two most abundant hydrolysis products generated by Der p 1 (based on HPLC peak area at 210 nm) corresponded to GPNSKGRSLIGR (rat PAR 2 residues 30 -41) and LDTPyggc (rat PAR 2 residues 42-45 plus the added coupling sequence), neither of which would be able to activate PAR 2 . Therefore, in the synthetic tethered ligand peptide, Der p 1 preferentially cleaves after PAR 2 arginine residues 36 and 41 in a manner that would disarm the receptor rather than causing activation. Another hydrolysis product observed in small amounts, IGRLDTPYGGC, indicated that Der p1 can also cleave the P20 peptide after leucine residue 38. Thus, the data showed that Der p 1 was able to hydrolyze the rat PAR 2 sequence at its cleavage/activation site less efficiently than trypsin and that it would preferentially disarm the intact receptor, rendering it insensitive to trypsin activation. Of the peptides generated by Der p 1, only SLIGR would be capable of activating PAR 2 , albeit with very low potency.
To assess further the ability of Der p 1 to release a PAR 2 -activating peptide from P20, we used a functional assay, in which the enzyme hydrolysis products are first generated and the resulting peptides are tested on a PAR 2 -expressing cell line that responds by an increase in intracellular calcium. Since Der p 1 alone was not capable of causing an increase in intracellular calcium, any calcium signal resulting from the hydrolysis of the substrate P20 by Der p 1 must come from the peptides generated. As shown in Fig. 11A, Der p 1 alone failed to cause a calcium signal in the responsive cells that otherwise yielded a calcium signal when exposed to the PAR 2 AP, SLIGRL-NH 2 . However, upon incubation with Der p 1, the P20 hydrolysis product yielded a calcium signal that was 26 Ϯ 7% (n ϭ 3) of the maximal response to PAR 2 AP (Fig. 12B). This is equivalent to a response generated by 1.5 M PAR 2 AP. The magnitude of the response generated by the hydrolysis products of P20 was not sufficient, however, to desensitize all of the PAR 2 present on the cells. Notwithstanding, a prior desensitization of PAR 2 caused by two exposures to the PAR 2 AP (Fig. 12C) (35) was able to eliminate completely the signal generated by the Der p 1 hydrolysis products of P20, thereby indicating that the P20-derived peptide (most likely SLIGR) was acting via PAR 2 .
Under the same experimental conditions used for P20 cleavage, hydrolysis of the human PAR 2 P27 peptide by Der p 1 (1 M) yielded  Ac-GTNRSSKGRSLIGKVDGTSHVTGKG, which corresponds to the disarming of PAR 2 . Another peptide identified was TNRSSKGRS-LIGKVDGT. However, the presence of a PAR 2 -activating peptide sequence (like SLIGR) was not detected. In contrast, Der p 3 and trypsin were able to cleave the P27 synthetic peptide efficiently. In both cases, the only detectable hydrolysis product was SLIGKVDGTSHVT-GKGVT, which can on its own activate PAR 2 . On the whole, these results confirmed that, in contrast with Der p 3 and trypsin, Der p 1 was not able to cleave the human PAR 2 sequence efficiently at its cleavage/ activation site to yield a PAR 2 -activating peptide sequence.

DISCUSSION
We investigated and compared the molecular mechanisms by which the dust mite proteolytic allergens, Der p 1 and Der p 3, trigger the release of the proinflammatory cytokine IL-8 from human pulmonary epithelial cells (A549), with particular attention paid to a potential role for PAR 2 . The main finding of our study was that the signaling events mediated by Der p 3 to trigger IL-8 release in A549 cells and PAR 2expressing KNRK cells matched exactly those induced by peptide-or trypsin-activation of PAR 2 , whereas the signaling activated by Der p 1 to stimulate IL-8 release was distinct and, unlike Der p 3, was caused independently of PAR 2 activation. In the respiratory system, PAR 2 is expressed in the epithelium and is considered to play both a proinflammatory and a bronchodilator role (45,46). In the human alveolar epithelial A549 cells, activation of PAR 2 has been observed to trigger the secretion of several inflammatory mediators, such as IL-8, IL-6, prostaglandin E 2 , matrix metalloproteinase-9, and GM-CSF (48,49). A comparable effect of PAR 2 activation has also been observed in part in primary bronchial epithelial cells (17). However, the intracellular signaling pathways (other than the G q/11 ␣mediated activation of phospholipase C) that may regulate PAR 2 -induced release of these inflammatory mediators were not evaluated in any detail in those studies. Here, we confirmed that stimulating A549 cells with agonists of PAR 2 induced IL-8 secretion. We demonstrated further that the secretion results from the up-regulation of IL-8 mRNA (in a time-dependent manner) that was mainly dependent on new gene transcription, since the up-regulation of IL-8 mRNA was significantly reduced by actinomycin D. Consistent with this result, transfection studies with a wild-type IL-8 luciferase reporter construct showed that the IL-8 promoter was transcriptionally activated in response to PAR 2 agonists (PAR 2 AP, trypsin). Mutational analysis further identified the NF-B and AP-1 binding sites as the essential cis elements involved in IL-8 gene transcription in response to PAR 2 activation in A549 cells. Importantly, our data showed that PAR 2 stimulation led to activation of both NF-B and AP-1 transcription factors, as shown by EMSA and transfection experiments. NF-B binding to the IL-8 promoter peaked at 1 h, which paralleled the time course of degradation of IB␣. Supershift analysis identified the PAR 2 -induced translocated NF-B complexes as p65/p50 and p65/p52 heterodimers, p65 being the major NF-B subunit that regulates IL-8 transcription (24,49). Interestingly, this time course for PAR 2 induced NF-B activation has also been described in other cell types in response to both trypsin and PAR 2 agonist peptides (50,51). The kinetics of PAR 2 -triggered activation of AP-1 was quite different compared with NF-B. PAR 2 activation (PAR 2 AP, trypsin (data not shown)) rapidly and strongly increased the AP-1 binding activity, which was readily detected at 10 min, gradually increased to be maximal at 1 h, and persisted thereafter for at least 2 h. The AP-1 complex induced in response to PAR 2 activation consisted mainly of c-Fos, JunB, and Jun D and, to a lesser extent, c-Jun and Fra-2. Interestingly, c-Fos and Jun D are also the main components of the IL-8 promoter AP-1-DNA binding complex induced by the serine protease coagulation factor VIIa in keratinocytes (52,53). Thus, in airway epithelial cells, both NF-B and AP-1 transcription factors were activated by PAR 2 stimulation and apparently functioned together to result in IL-8 up-regulation. We believe this is the first report that documents a functional cooperative role for both NF-B and AP-1 transcription factors in IL-8 up-regulation in response to PAR 2 activation in this cell type. Hence, we suggest that the expression of other NF-B and AP-1 target proinflammatory genes (such as IL-6, GM-CSF, COX-2, and iNOS) may also be affected in a similar manner by PAR 2 stimulation. PAR 2 stimulation has been shown to activate members of the MAP kinase signal transduction family in a way that differs, depending on the cell type affected and the nature of the stimulus (54). In keeping with this diversity of activation of the MAP kinases, we found that PAR 2 agonists induced the phosphorylation/activation of all three MAP kinases FIGURE 12. Der p 1 and Der p 3 cleavage sites of PAR 2 -derived cleavage/activation sequences. A, cleavage of rat PAR 2 P20 by trypsin (upper arrow) and Der p 1 (lower arrows). The P20 peptide representing the cleavage/activation sequence of rat PAR 2 was subjected to proteolysis by either Der p 1 or trypsin as outlined under "Experimental Procedures," and the resulting peptides were isolated by HPLC and identified by mass spectral analysis. The boldface and dotted arrows indicate the major and minor cleavage sites, respectively, for Der p 1. Only one major hydrolysis site was detected for trypsin under the same conditions (upper arrow). B, the left-hand side of the trace shows the calcium response of wt-rPAR 2 cells treated with the hydrolysis products OE resulting from incubating Der p 1 (350 nM) with a peptide (25 M) containing residues 30 -45 of the rat PAR 2 sequence (P20; contains the PAR 2 cleavage/activation sequence); the PAR 2 AP SLI-GRL-NH 2 (E; 1 M) was then added to the same cell suspension. The right-hand trace shows a control cell sample treated with SLIGRL-NH 2 (E; 1 M). C, wt-rPAR 2 cells were exposed to receptor-desensitizing concentrations of the PAR 2 AP SLIGRL-NH 2 (E; 50 M) before treatment with the hydrolysis products (OE) resulting from preincubating Der p 1 (350 nM) with P20 (25 M).
(ERK1/2, p38 MAPK, and JNK) in A549 cells. However, the kinetics and extents of activation were quite different for the three kinases. Treatment of A549 cells with the PAR 2 AP (or trypsin; data not shown) caused a transient but modest increase in both p38 MAPK and JNK phopsphorylation/activation compared with the persistent and markedly increased levels of phosphorylation/activation of both kinases evoked by TNF-␣. In contrast, PAR 2 -mediated activation of ERK1/2 was strong, rapid in onset (as early as 10 min), and sustained (lasting up to 2 h). The timing of ERK1/2 activation was exactly in step with the appearance of AP-1 binding activity. Using selective MAP kinase inhibitors, we found that inhibition of ERK1/2 and p38 MAPK each decreased IL-8 induction in response to PAR 2 activation, whereas inhibition of JNK appeared to increase the induction of IL-8. The observed requirement of ERK1/2 activation for maximal PAR 2 -mediated IL-8 expression may relate to the role of ERK1/2 in AP-1 transactivation. Activation of ERK1/2 signaling has indeed been shown to induce phosphorylation and an increase in the trans-activating activity of the AP-1 transcription factors c-Fos, Jun D, and c-Jun (55,56). We found that these very same components were part of the PAR 2 -induced AP-1-DNA binding complex. Cross-talk between the activation of MAP kinases and NF-B may also occur in a number of settings. For example, in murine fibrosarcoma cells, p38 MAPK and ERK1/2 have both been reported to be required for NF-B p65 transactivation mediated by TNF-␣ (57). JNK has also been described as a potential upstream activator of NF-B (and AP-1) in the regulation of airway epithelial cell IL-8 expression (58). The MAP kinase family is also thought to regulate other steps in the overall process of gene expression, such as chromatin remodeling (59) or in regulating IL-8 mRNA stability (60,61). Additional studies are thus needed to elucidate the precise mechanism by which ERK1/2, p38, and JNK participate to regulate NF-B and AP-1 transcriptional activity in airway epithelial cells stimulated with PAR 2 agonists.
All of the signaling events described above, as triggered by trypsin and the specific PAR 2 AP, were reproduced exactly in timing and extent by treating A549 cells with the mite serine proteinase, Der p 3. As such, our data are entirely consistent with the study of Sun et al. (15), which reported that PAR 2 is one potential target for signaling by Der p 3. However, our study significantly extends these observations by demonstrating that the specific cellular signal transduction pathways linking Der p 3-mediated PAR 2 proteolysis/activation to the up-regulation of IL-8 production involve signaling via the MAP kinases ERK1/2, p38 MAPK, and JNK to activate the transcription factors NF-B and AP-1.
A key question to ask relating to the action of Der p 1, in contrast with Der p 3, is, "Does the mite-derived cysteine proteinase Der p 1 cause the elevation of IL-8 by stimulating PAR 2 , a receptor activated by various serine proteinases including Der p 3?" Our data are entirely consistent with previous work showing that Der p 1 triggers the release of IL-8 from A549 cells in a concentration-and time-dependent manner (14). However, in a number of respects, our results raise doubt as to whether Der p 1 acts via PAR 2 activation to increase IL-8 production. First, our data show that although Der p 1 stimulates the transcriptional activity of the IL-8 gene in A549 cells, the magnitude of stimulation by this allergen is much lower than that of either the PAR 2 AP or Der p 3 (which we believe does act via PAR 2 cleavage/activation). Further, the duration of the increase in IL-8 mRNA expression resulting from Der p 1 activation, although smaller in magnitude, was much more prolonged, compared with that triggered by the PAR 2 AP or Der p 3. Second, although the Der p 1-induced IL-8 mRNA up-regulation relies also primarily on binding sites for NF-B and AP-1, Der p 1 triggers mainly NF-B activation and only to a minor degree on AP-1 activation. Indeed, in agreement with the data reported by Stacey et al. (16), we showed that Der p 1 activates NF-B (consisting of p65/p50 and p65/p52 heterodimers) with a maximum obtained after 60 min. However, in contrast with PAR 2 agonists, Der p 1 induces only a weak delayed recruitment of AP-1 transcription factors (mainly Jun B and Jun D) to the IL-8 promoter. In view of the study realized by Hoffmann et al. (25) on IL-1-and epidermal growth factor-induced IL-8 expression, this result with Der p 1 may explain the low amounts of IL-8 release induced by Der p 1 compared with PAR 2 agonists. Indeed, according to these authors (25), the IL-8 gene behaves like an immediate early gene. In this respect, low, constitutively expressed amounts of AP-1 proteins in conjunction with the stimulusinduced nuclear translocation of p65 NF-B might be responsible for IL-8 transcription within the first 30 min of treatment. Thereafter, de novo synthesis of c-Fos and various Jun proteins might serve to maintain high levels of mRNA synthesis. In the context of Der p 1 treatment, the AP-1 site appears thus to function principally as a basal level enhancer (using the constitutive presence of AP-1 in the nucleus), since Der p 1 induces only a weak delayed recruitment and activation of AP-1 transcription factors (mainly Jun B and Jun D). This situation has already been described for TNF-␣ (review in Ref. 24). In contrast, PAR 2 activation induces a strong and rapid increase in the AP-1 DNA binding complex (in response to PAR 2 AP or Der p 3). The complex, composed mainly of c-Fos, Jun D, and Jun B, would contribute to the generation of maximal IL-8 amounts in response to PAR 2 activation. Third, as opposed to the actions of PAR 2 AP (or trypsin; data not shown), for Der p 1, ERK1/2 appears to be the principal MAP kinase involved in the up-regulation of IL-8. Further, the time course of ERK1/2 activation (over 30 min) is very distinct from the kinetics of PAR 2 -mediated ERK1/2 stimulation (markedly increased after 10 min). Finally, the data obtained with cells (HeLa or KNRK) that did or did not express PAR 2 revealed that Der p 1 was unable to induce either calcium mobilization or IL-8 up-regulation via PAR 2 activation, in clear contrast with the actions of either the PAR 2 AP or trypsin. That said, our data did show that in HeLa cells that do not express PAR 2 , Der p 1 was nonetheless able to increase IL-8 production (albeit modestly) ostensibly via a non-PAR 2 mechanism that has yet to be elucidated.
The experiments discussed above indicate that Der p 1 can stimulate IL-8 production via a mechanism that does not involve PAR 2 . However, we do not exclude the possibility that Der p 1 could cleave PAR 2 in certain circumstances. In this regard, we did find that the polypeptide sequence representing the cleavage/activation site of rat (but not human) PAR 2 can be cleaved by Der p 1 to yield a peptide sequence that in principle might be capable of activating PAR 2 (Fig. 12A), albeit with very low potency. Thus, when present, rat PAR 2 may be subject to a low level of proteolysis by Der p 1. However, when the cells are directly incubated with Der p 1, that low level of activation does not appear to be sufficient to cause an elevation of intracellular calcium. Moreover, it must be pointed out that, although Der p 1 can mainly cleave the synthetic P20 and P27 peptides in vitro, in a manner that in theory would predict a disarming of the receptor rather than causing activation, crossdesensitization experiments using intact PAR 2 -expressing KNRK cells indicated that a prior treatment of cells with Der p 1 does not impair the ability of the cells to respond subsequently to Der p 3, trypsin or PAR 2 AP. This lack of a disarming action of Der p 1 in intact cells is in keeping with the lack of ability of trypsin to cleave at a potential arginine target residue downstream from the cleavage/activation site (67). It is likely that the Arg 41 -Leu 42 bond in the tethered ligand sequence of rat PAR 2 , which is accessible to proteolytic cleavage in the soluble synthetic peptide in vitro, is rendered cryptic in the sequence of the intact receptor, as expressed at the cell surface. Further, it is evident that although Der p 1 can cleave synthetic rat PAR 2 -derived peptides at potential activation/disarming sites, this cleavage does not appear to happen when the receptor is present in intact cells.
Receptor glycosylation in the region of the N-terminal cleavage/activation site of the receptor that regulates its activation by tryptase (28) cannot be a factor accounting for the inability of Der p 1 to activate PAR 2 , since the receptor construct lacking the N-terminal glycosylation site was as resistant as was the wild type receptor in terms of the lack of ability of Der p 1 to cause a calcium signal in the cells. Trypsin, however was fully active in this regard in the glycosylation-deficient receptor, indicating that if sufficient proteolysis had been caused by Der p 1, a calcium signal should have been generated. Apart from a single previous study pointing to the activation of PAR 2 by Der p 1 (17), very few studies have demonstrated activation of PAR 2 by other cysteine proteinases, except for those like the arginine-specific (trypsin-like) cysteine proteinases (gingipains) produced by P. gingivalis (62,63). Interestingly, in a study showing that trypsin can activate eosinophils via PAR 2 , the effects of papain, a cysteine proteinase archetype that shares a structural motif including the catalytic triad with Der p 1 but is unlike the gingipains, appeared to act independently of PAR 2 (64). Thus, some cysteine proteinases, including Der p 1, appear to be able to activate cells via a PAR 2 -independent pathway.
We are not able to explain the differences between our results and those of Asokananthan et al. (17), who provided evidence for an activation of PAR 2 by Der p 1. Unfortunately, the precise specific activity and Western blot analysis of the Der p 1 protein used in that work were not recorded in that paper. A putative contamination of the cysteine protease Der p 1 preparations with serine proteases, such as Der p 3, 6, or 9, could be a contributing factor to explain the discrepancy between our current findings and those of earlier studies (14,17). Indeed, co-purification of serine proteinase activity with Der p 1 cysteine proteinase by both affinity purification using the monoclonal antibody-coupled column and by active site affinity purification has been reported by several studies dealing with Der p 1 isolated from natural sources (5, 6, 65) as well as for commercially available natural Der p 1 (66).
In our study, we ascertained the purity of our Der p 1 preparations by SDS-PAGE, Western blot analysis, and mass spectrometry in order to verify the absence of contaminating Der p 3 or other proteinases (Fig. 1). In other respects, some of the data previously reported for the action of Der p 1 via PAR 2 (14) appear somewhat contradictory. In one study by those investigators (12), it was shown that Der p 1 and Der p 9 induce the release of GM-CSF, IL-6, and IL-8 in bronchial epithelial cells. However, it was suggested in that study that these two mite allergens act presumably via different receptors (potentially PARs), since the calcium response generated in the BEAS-2B target cells by Der p 1 treatment (Fig. 6 in Ref. 12) did not cross-desensitize a subsequent calcium signal generated in the same cells with Der p 9 (12). It has recently been demonstrated that Der p 9 as well as Der p 3 induce phosphoinositide hydrolysis, transient cytosolic calcium mobilization, and release of GM-CSF and eotaxin in A549 cells by activating PAR 2 (15). Taken together, these two reports (12,14) would appear to be somewhat self-contradictory, with one study suggesting that Der p 1 causes its effects via a receptor distinct from PAR 2 in BEAS-2B cells (i.e. the Der p 9-activated receptor) (12), whereas the second study suggested that Der p 1 does indeed act via PAR 2 (14). We suggest that the data we report here support the hypothesis that Der p 1 can act via a receptor distinct from PAR 2 to induce an up-regulation of IL-8.
In summary, our study has delineated the transcription factors and the MAP kinases responsible for PAR 2 -activation-mediated IL-8 production in airway epithelial cells. We further confirm that the mite allergen Der p 3 induces IL-8 secretion via PAR 2 activation. Finally, we furnish evidence supporting the idea that, relative to trypsin and Der p 3, the cysteine proteinase Der p 1 is not an effective activator of PAR 2 and is unlikely to induce IL-8 release via PAR 2 cleavage. Although the Der p 1 target remains to be defined, we show that the regulation of IL-8 secretion in response to Der p 1 can be mediated at least in part through a PAR 2 -independent pathway that triggers the activation of both ERK1/2 MAP kinase and the IL-8 transcription factor, NF-B.