INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Keywords: airway smooth muscle; asthma; cytokines; cytokine receptors; eotaxin

Airway wall inflammation is a central feature to the pathophysiology of asthma and other airway diseases such as chronic obstructive pulmonary disease. The induction and perpetuation of inflammation involves a complex and coordinate response involving multiple inflammatory cells, mediators, and cytokines. Evidence from our own studies (1-3) and from others (4-7) suggests that airway smooth muscle is a potential source of multiple proinflammatory cytokines (8, 9). Stimulation of cultured myocytes with either interleukin (IL)-1 or tumor necrosis factor- (TNF-) induces expression and release of chemokines including regulated upon activation, normal T-cell expressed and secreted (RANTES) (1), eotaxin (6, 7), IL-8 (5), and other cytokines such as granulocyte-macrophage colony-stimulating factor (GM-CSF) (2, 4), IL-6 (3), and IL-11 (10). GM-CSF, eotaxin, RANTES, and IL-8 are important cytokines for activation of eosinophils, one of the critical effector cells in the pathogenesis of asthma. RANTES is a potent chemoattractant for eosinophils as well as for other cell types observed in allergic inflammation, including monocytes and memory T-lymphocytes (11). IL-8, in addition to its action on neutrophils, is also a chemoattractant for activated eosinophils (12, 13). Eotaxin is a highly selective and potent local chemoattractant for eosinophils (14), and GM-CSF stimulates maturation, surface activation, and proliferation of several proinflammatory cells and is particularly important for the survival of eosinophils (15). Previously, we have demonstrated that human airway smooth muscle cells can enhance the survival of eosinophils through production of GM-CSF (2), and others have shown that elaboration of RANTES and eotaxin from these cells promotes eosinophil chemotaxis (1, 7). Production by airway smooth muscle of the aforementioned cytokines and chemokines implies a role for these structural cells of direct participation in the pathogenesis of airway inflammation through recruitment and activation of eosinophils and other infiltrating inflammatory cells.

Many forms of allergic asthma are characterized by eosinophil accumulation resulting from polarization of T-lymphocytes toward a Th2 phenotype and the coordinate expression of specific proinflammatory cytokines including IL-4 and IL-13, which are products of the gene cluster q31-33 on chromosome 5. IL-13 is a pleiotropic 12-kD protein that is produced in large quantities by activated CD4⁺ Th2 cells and induces several features of Th2-dominated inflammation, including IgE production (16), endothelial cell adhesion molecule expression (17), modulation of eosinophil apoptosis (18), eosinophil recruitment, and induction of airway hyperresponsiveness to inhaled spasmogens (19). Excessive IL-13 production in atopic and nonatopic asthma is now a well-documented feature of asthma (20-22). Transgene pulmonary overexpression of IL-13 and IL-4 in mice is associated with several key pathologic features of airways inflammation and remodeling in chronic severe asthma, including lymphocyte and eosinophil accumulation, mucus cell metaplasia, subepithelial fibrosis, and airways hyperresponsiveness (19, 23). An important additional feature of the study by Zhu and colleagues (23) was the selective induction of total lung eotaxin mRNA and increased protein levels in bronchoalveolar compartments in the transgene positive mice, which was not observed with other proinflammatory cytokines such as GM-CSF, IL-4, IL-5, and MCP-5. Selective eotaxin induction in IL-13 overexpressing mice establishes an important link between Th2-driven inflammation and eotaxin and implies that IL-13, through local eotaxin induction, may prime tissues in the lung to produce an exaggerated eosinophilic response (23).

The precise mechanism and cells involved in the induction of eotaxin release by IL-13 are unclear, and the direct effects of IL-13 on cytokine production by airway smooth muscle have not been investigated. The delay in the onset of airway hyperresponsiveness suggests that direct contraction of airway smooth muscle by IL-13 is unlikely, though it may activate other aberrant responses of airway smooth muscle such as accelerated proliferation and/or the production of proinflammatory cytokines. In the present study, experiments were performed to characterize the capacity of IL-13 and IL-4 to promote the release of eosinophil-activating cytokines from human airway smooth muscle cells and then to explore the receptors and intracellular mechanisms regulating cytokine release in these cells.

Our results demonstrate that airway smooth muscle responds to IL-13 by selectively inducing release of eotaxin and that this was dependent on the IL-4 receptor -chain (IL-4R) that was expressed constitutively on human airway smooth muscle. Additionally, IL-13 interacted synergistically with IL-1 to promote a selective and substantial release of eotaxin. These properties were also shared with IL-4. As with the action of IL-13 or IL-4 alone, synergy with IL-1 was dependent on the activation of IL-4R, and in turn this was linked to activation of multiple intracellular signaling pathways including signal transducer and activation of transcription-6 (STAT6) protein and mitogen- activated protein (MAP) kinases. Selective chemical inhibitors of MAP kinase pathways identified the importance of p42/p44 ERK and p38 MAP kinase activation in these responses.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Isolation and Culture of Human Airway Smooth Muscle Cells

In accordance with procedures approved by the Guy's and St. Thomas' Hospitals' Research Ethics Committee, macroscopically normal human bronchial smooth muscle was obtained from the lobar or main bronchus of 31 patients of either sex (mean age 63 ± 2 years; range 33- 76 years; 16 male, 15 female) undergoing lung resection for carcinoma of the bronchus. After removal of the epithelium and adherent connective and parenchymal tissue, smooth muscle cells were isolated by overnight enzymatic dispersion and placed into culture as previously described (2). Using fluorescent immunocytochemistry techniques, growth-arrested human airway smooth muscle cells (passages 1 and 2) stained (> 95%) for both smooth muscle-specific -actin and myosin heavy chain. When examined by light and electron microscopy, these cells displayed all the reported characteristics of viable smooth muscle cells in culture (24). Cells at passages 3-6 were used in all experiments.

Cell Stimulation and Collection of Cell-conditioned Medium

Near-confluent cells in 24-well plates were growth-arrested by incubation in RPMI 1640 containing 25 mM HEPES, 2 mM L-glutamine, 100 U/ml:100 µg/ml penicillin/streptomycin (supplemented RPMI) with the addition of 1 µM insulin, 5 µg/ml transferrin, 100 µM ascorbate, and 1 mg/ml bovine serum albumin (BSA). After 72 hours, cells were stimulated with recombinant human cytokines (R&D Systems, Abingdon, UK) in supplemented RPMI 1640 containing 1 mg/ml BSA. Where indicated, cells were pretreated for 30 minutes with blocking antibodies or istotype-matched controls that remained in the media throughout. In some experiments, cells were also pretreated for 30 minutes with inhibitors of MAP kinase (U 0126 1,4-diamino-2,3,-dicyano-1,4-bis [2-aminophenylthio] butadiene; SB 202190 4-[4-fluorophenyl]-2-[4-hydroxyphenyl]-5- [4-pyridyl] 1H-imidazole, purchased from Calbiochem, Nottingham, UK). These were dissolved in DMSO at 50 mM, and further dilutions were made in cell culture medium. The highest concentration of DMSO to which cells were exposed was 0.02% (corresponding to 10 µM of inhibitor). Inhibitors and their DMSO controls remained in the media throughout stimulation. Cell-conditioned medium was collected and cell-free supernatants were stored at -70° C until measurement of cytokine levels by enzyme-linked immunosorbent assays (ELISA).

Measurement of Cytokine Levels by ELISA

Cytokine levels in airway smooth muscle cell-conditioned culture medium were determined in duplicate by specific sandwich ELISA using matched monoclonal (antihuman) capture and biotinylated antihuman monoclonal (GM-CSF) or polyclonal (RANTES, eotaxin, IL-8) detection antibody pairs (R&D Systems) as described previously (25). The manufacturer's instructions were followed throughout. Levels of cytokines in cell-conditioned medium were initially expressed in nanograms per milliliter before normalization to nanograms per milliliter per million cells to correct for small differences in cell densities between patients. According to the manufacturer's guidelines none of the assays showed any cross- reactivity or interference with one another or with IL-1, IL-13, or IL-4.

Localization of Cell Surface IL-4 Receptors and IL-2 Receptor  by Fluorescent Staining

Localization of IL-4 receptors (IL-4R) was determined by specific cell surface binding of biotinylated recombinant human (rh) IL-4 (R&D Systems). Near-confluent growth-arrested cells, treated for 24 hours with 10 ng/ml IL-13 or IL-4 alone, or in combination with 10 ng/ml IL-1, and grown on Lab-Tek four-chamber microscope glass slides (NUNC; Life Technologies, Paisley, UK), were washed twice in ice-cold Ca²⁺/Mg²⁺-free phosphate-buffered saline (PBS). Cell monolayers were blocked with ice-cold PBS containing 3% fetal bovine serum (FBS) for 30 minutes before addition of an excess of biotinylated rhIL-4 (450 ng/ml) in PBS for 1 hour at 4° C. As a negative staining control, cells were treated with 1.5 µg/ ml of an irrelevant IL-4R ligand (soybean trypsin inhibitor) biotinylated to the same degree as the cytokine (R&D Systems). Other controls included preabsorbing 500 ng/ml biotinylated rhIL-4 with 1 mg/ml of a polyclonal goat IgG antihuman IL-4 antibody (R&D Systems) at room temperature for 30 minutes before incubating the mixture with blocked cells, as well as cross-competition of biotinylated rhIL-4 with nonbiotinylated rhIL-4. Specifically bound biotinylated rhIL-4 was detected by addition of 2.5 µg/ ml avidin conjugated with fluorescein isothiocyanate (avidin-FITC) at an f:p ratio of 5:1 (R&D Systems). After 30 minutes at 4° C in the dark, cells were washed three times in 1× RDF1 buffer (PBS/BSA-based proprietary reagent buffer formulated to minimize background staining and stabilize specific binding, R&D Systems). Finally, cells were fixed for 10 minutes in 2% paraformaldehyde (methanol-free, EM-grade, Polysciences, Warrington, PA), washed twice in water, counterstained in Hoescht 33342 (1 µg/ml), and mounted in PBS/glycerol (1:9) for viewing with an Olympus BX-50 microscope equipped with reflected UV illumination.

Expression of the -chain (c) was determined in FBS-blocked cells using an antihuman interleukin-2 receptor (IL-2R) -chain monoclonal (IgG₁) antibody (1 µg/ml in 1% FBS in PBS for one hour at room temperature, R&D Systems). Specific c localization was detected with a goat, antimouse IgG FITC conjugated secondary antibody (1:100 dilution in 1% FBS in PBS for one hour at 37° C, Sigma, Poole, UK). In control experiments, the primary and/or secondary antibody was omitted from the protocol and cells were incubated in either PBS containing 1% FBS alone or in the presence of a concentration-matched, isotype-matched control (Sigma). Cells were post-fixed in 2% paraformaldehyde, washed in water, and counterstained in Hoescht 33342 as previously described.

Flow Cytometric Analysis of IL-4R Cell Surface Expression

After stimulation for 24 hours with 10 ng/ml IL-13 or IL-4 alone, or in combination with 10 ng/ml IL-1, near-confluent, growth-arrested cells were removed from 25 cm² flasks using PBS-containing ethylenediamine-tetraacetic acid (EDTA; 0.5 mM for 20 min) and washed in 2 ml ice-cold PBS (200 g × 5 min). Suspensions were filtered through a disposable cell strainer (70 µm nylon mesh; Becton Dickinson, Oxford, UK) to remove cell aggregates. Aliquots of 1 ×10⁵ cells in 5-ml round-bottom polystyrene tubes and following centrifugation (200 g × 5 min) were resuspended in 1 ml ice-cold PBS-containing 3% FBS for 30 minutes. After recentrifugation, blocked cells were resuspended in 35 µl of an excess of biotinylated rhIL-4 (450 ng/ml) for one hour at 4° C. Irrelevant ligand (biotinylated soybean trypsin inhibitor) controls and preabsorbed rhIL-4 (antihuman IL-4 antibody) negative controls were included as in the fluorescent staining IL-4R localization studies (see above). Specifically bound biotinylated rhIL-4 was detected by addition of 10 µL avidin-FITC (2.5 µg/ml; R&D Systems). After 30 minutes in the dark at 4° C, cells were washed twice (200 g × 5 min) in 1× RDF1 buffer and finally resuspended in 0.5 ml ice-cold FACSflow (Becton-Dickenson, Oxford, UK) for flow cytometry. c expression was determined in FBS-blocked cells using an antihuman IL-2R -chain monoclonal antibody (R&D Systems). Specific c localization was detected with a goat, antimouse IgG FTIC-conjugated secondary antibody as described for the immunocytochemical localization studies. A concentration- and isotype-matched control (Sigma) was included for each treatment condition.

Analysis of specifically bound IL-4 fluorescent labeling and c expression was determined on a FACSCalibur flow cytometer (Becton Dickinson, Oxford, UK) using an argon laser (488 nm). For detection of FITC fluorescence, a 530-nm band-pass filter was placed in the light path. Forward versus 90° light scatter histograms were used to gate on intact live cells and eliminate debris. Fluorescence histograms of 1,024 channel resolution were collected for at least 10,000 events that satisfied the light scatter gating criteria at a flow rate of 60 µl/min. Geometric mean fluorescence intensities of labeled cells were determined using the CellQuest Pro analysis program (Becton Dickinson, Oxford, UK). Following analysis, cells were confirmed microscopically to be intact.

Western Immunoblot Detection of STAT6 and MAP Kinase Activation, and IL-4R and c Expression

Near-confluent, growth-arrested cells in 25-cm² flasks were stimulated with 10 ng/ml IL-13 or IL-4 alone or in combination with IL-1 (10 ng/ ml). Where indicated, cells were pretreated for 30 minutes with blocking antibodies or istotype-matched controls present throughout stimulation. Cytokine treatment was for 15 minutes for detection of STAT6 and MAP kinase activation and was 24 hours for IL-4R subunit expression. Peripheral blood human eosinophils were isolated and purified by anti-CD16 negative selection using the MACS system as previously described (2). Whole cell lysates were prepared as described elsewhere (26) and samples of total protein extracts (10 µg/lane) were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) on 10% acrylamide precast gels and transferred electrophoretically (Novex, Frankfurt, Germany) to nitrocellulose membranes. MAP kinase and STAT6 activation was detected using antibodies (for details see [26]) to the phosphorylated forms of p42/p44 ERK (cat. sc 7383; Santa Cruz Biotechnology Inc., Santa Cruz, CA), p38 (cat. 506124; Calbiochem), SAPK/JNK (cat. 9251; New England Biolabs, Beverly, MA), or STAT6 (1:1,000 dilution overnight at 4° C; cat. 9361; New England Biolabs). IL-4R expression was detected using a monoclonal antibody (1 µg/ml) to the IL-4R -subunit (cat. MAB230; R&D Systems), and the -chain subunit was detected with a polyclonal antibody (1:1,000 dilution) to the anti-IL-2R -chain (cat. MAB284; R&D Systems). Primary antibodies were detected with goat antirabbit IgG (c, p38, and STAT6) or goat antimouse (ERK, SAP/JNK, and IL-4R) horseradish peroxidase-conjugated secondary antibodies (Santa Cruz Biotechnology) and visualized by enhanced chemiluminescence (Pharmacia, Amersham, UK). Blots were stripped and reprobed with a mouse antiglyceraldehydes 3 phosphate dehydrogenase (GAPDH) monoclonal antibody (1:5,000, clone 6G5; Biogenesis, Poole, UK) to control for differences in loading. Signals were quantified using ImageQuant software (Molecular Dynamics, Sunnyvale, CA) on autoradiographs that depicted bands within a linear range of exposure. Densitometric data were normalized for GAPDH values and presented as -fold increases above basal optical density (OD) ratios.

Statistical Analysis

Data in the text and figure legends are expressed as mean ± SEM of observations obtained from cells cultured from n patient donors. EC₅₀/IC₅₀ values, and where necessary extrapolated maximum responses, were estimated for individual concentration-response curves using nonlinear least-squares regression analysis (SigmaPlot; SPSS Inc., Chicago, IL). EC₅₀/IC₅₀ values were converted to negative logarithmic values (pD₂) for all statistical comparisons, although for ease of comprehension, EC₅₀/IC₅₀ values are given in the text. Data were compared using one-way analysis of variance (ANOVA) followed by Bonferroni's t test post hoc to determine statistical differences (SigmaStat; SPSS Inc., Chicago, IL). A probability value (p) of less than 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

IL-13 Selectively Releases Eotaxin from Human Airway Smooth Muscle Cells and Synergizes with IL-1

Concentrations of eotaxin, GM-CSF, RANTES, and IL-8 were determined in the same supernatants. In all cultures examined, release of GM-CSF, RANTES, and IL-8 by unstimulated cells at 24 hours was less than could be detected by ELISA (< 2.8-10 pg/ml). Constitutive release of eotaxin was observed but did not exceed 25 ng/(ml million cells) (Figure 1A). The concentration of IL-1 that induced near-maximum generation of each of these cytokines was 1 ng/ml (Figures 1 and 2), consistent with previous observations (25).

View larger version (40K): [in this window] [in a new window]   Figure 1.   Effects of IL-13 and IL-1 on release of (A) eotaxin, (B) GM-CSF, (C) IL-8, and (D) RANTES from cultured human airway smooth muscle, determined by ELISA. Growth-arrested cells were left unstimulated (unstim) or were treated for 24 hours with increasing concentrations of IL-1 (circles) and IL-13 (squares) either alone or in combination (triangles). Points represent mean ± SEM of duplicate values from independent experiments using cells cultured from four different donors, cell passages 3-4. *p < 0.05, **p < 0.01, and ***p < 0.001 compared with IL-1-stimulated concentrations, §§p < 0.01 compared with unstimulated concentrations by Bonferroni's t test.

View larger version (19K): [in this window] [in a new window]   Figure 2.   Time course showing the relative kinetics of eotaxin release from unstimulated human airway smooth muscle cells (open circles) or after stimulation with 10 ng/ml IL-1 (closed circles) and IL-13 (squares) either alone or in combination (triangles). Points are mean ± SEM of duplicate values from independent experiments using cells cultured from three different donors, cell passages 3-5. *p < 0.05, **p < 0.01, and ***p < 0.001 compared with unstimulated cells at similar time points by Bonferroni's t test.

Addition of IL-13 (0.1 pg/ml-100 ng/ml) alone resulted in release of eotaxin but had no effect on detectable concentrations of GM-CSF, IL-8, or RANTES (Figure 1). Statistically significant increases in eotaxin release occurred with 10 ng/ml IL-13 (p < 0.05, n = 4), and maximum release occurred at 100 ng/ml and was approximately 2.5-fold higher than unstimulated concentrations and was similar to maximal concentrations found after stimulation with the same concentration of IL-1 (p > 0.05, n = 4). The concentration of IL-13 producing 50% of maximum (EC₅₀) eotaxin release was 4.27 ± 1 ng/ml.

When IL-13 was examined in combination with IL-1, a marked synergistic increase (EC₅₀ 2.73 ± 1 ng/ml) in eotaxin release was observed (Figure 1A) at 1, 10, and 100 ng/ml of each cytokine. At 100 ng/ml of IL-13 and IL-1 in combination, this increase was 16-fold above basal. In the same culture supernatants, this synergism was not observed for GM-CSF or IL-8 concentrations (Figures 1B and 1C). However, the presence of IL-13 in combination with IL-1 was found to reduce RANTES concentrations by approximately 50% compared with IL-1 alone (p < 0.01, n = 4) (Figure 1D). Examination of the time course for the increase in eotaxin release by 10 ng/ ml of IL-1 and IL-13 in combination revealed that detectable synergistic increases were present from 12 hours and maintained at least to 72 hours (Figure 2).

IL-4 Selectively Releases Eotaxin from Human Airway Smooth Muscle Cells and Synergizes with IL-1

Because IL-13 and IL-4 each binds the IL-4R -chain (IL-4R) (27), a common and essential component of the IL-4R, we next examined whether a similar profile of selective and synergistic eotaxin release occurred with IL-4 stimulation. Like IL-13, addition of IL-4 (1 pg/ml-100 ng/ml) increased eotaxin release, with no detectable effect on GM-CSF, IL-8, or RANTES (Figure 3). Statistically significant increases in eotaxin release occurred with 0.1 ng/ml IL-4 (p < 0.05, n = 3). Maximum eotaxin release occurred at 10-100 ng/ml and, like IL-13, was approximately 2.5-fold higher than unstimulated concentrations and reached similar concentrations compared with IL-1 (p > 0.05, n = 3). The EC₅₀ value for this effect was 59 ± 25 pg/ml. When stimulated by IL-13 and IL-4 (both at 10 ng/ml), concentrations of eotaxin were not different from that induced by either cytokine alone (not shown).

View larger version (39K): [in this window] [in a new window]   Figure 3.   Effects of IL-4 and IL-1 on release of (A) eotaxin, (B) GM-CSF, (C) IL-8, and (D) RANTES from cultured human airway smooth muscle, determined by ELISA. Growth-arrested cells were left unstimulated (unstim) or were treated for 24 hours with increasing concentrations of IL-1 (circles) and IL-4 (squares) either alone or in combination (triangles). Points are mean ± SEM of duplicate values from independent experiments using cells cultured from three different donors, cell passages 4-5. *p < 0.05, **p < 0.01, and ***p < 0.001 compared with IL-1-stimulated concentrations; §p < 0.05, §§p < 0.01 compared with unstimulated concentrations by Bonferroni's t test.

IL-4 in combination with IL-1 produced a marked synergistic increase (EC₅₀ 68 ± 21 ng/ml) in the concentration of eotaxin (Figure 3A). At 100 ng/ml of IL-4 and IL-1 together, this increase was more than 20-fold above basal. Like the IL-13 data, this synergism was not found with GM-CSF or IL-8 concentrations that were measured in the same culture supernatants (Figures 3B and 3C), and IL-4 and IL-1 in combination reduced RANTES concentrations by approximately 70% compared with IL-1 alone (p < 0.01, n = 3) (Figure 3D).

Involvement of the IL-4R in IL-13-stimulated Eotaxin Release

To determine whether IL-13-stimulated release of eotaxin was mediated by the IL-4R, the effect of a monoclonal neutralizing antihuman IL-4R antibody was examined. Preincubation for 30 minutes with the antibody (0.001-1 µg/ml) prevented IL-13- (p < 0.01, n = 3) but not IL-1-induced eotaxin release (Figure 4A). The concentration of antibody that inhibited IL-13-stimulated eotaxin release by 50% (IC₅₀) was 67.2 ± 7.6 ng/ml. Near-maximum inhibition occurred at 1 µg/ml. Similarly, the neutralizing anti-IL-4R antibody prevented synergistic increases in eotaxin release induced by IL-13 and IL-1 in combination (Figure 4B), returning eotaxin concentrations to those observed in the presence of IL-1 alone. The IC₅₀ for this effect was 25.03 ± 1.2 ng/ml. Maximum inhibition occurred at 1 µg/ml. Pretreatment of cells with an IgG_2a istotype-matched negative control antibody (0.001-1 µg/ml) had no effect on the concentrations of eotaxin induced by IL-13 or IL-4 alone (not shown) or in combination with IL-1 (Figure 4B). In two separate experiments, the neutralizing anti-IL-4R antibody (1 µg/ml) also prevented synergistic increases in eotaxin release induced by 10 ng/ml IL-4 and IL-1 in combination, returning eotaxin concentrations to those observed in the presence of IL-1 alone and abolished eotaxin release induced by IL-4 alone (not shown). Treatment with a monoclonal antibody neutralizing to c (1 µg/ml) had no effect on IL-13- or IL-4-stimulated eotaxin release (not shown).

View larger version (18K): [in this window] [in a new window]   Figure 4.   Effect of a monoclonal anti-IL-4R blocking antibody on (A) release of eotaxin from human airway smooth muscle after stimulation with 10 ng/ml IL-1 (circles) and IL-13 (squares) alone or (B) in combination (closed triangles). Open bars represent eotaxin concentrations in unstimulated cells, and filled, hatched, and gray bars represent eotaxin concentrations in cell-conditioned supernatants without the antibody after stimulation with IL-1, IL-13, and a combination, respectively. For comparison, control data (no antibody) for IL-1 and IL-13 alone are presented in both panels (note differences in scale between panels). A nonimmune isotype-matched negative control IgG2a antibody (open triangles) was also examined on eotaxin release induced by IL-1 and IL-13 in combination. Points are mean ± SEM of duplicate values from independent experiments using cells cultured from four different donors, cell passages 3-6. *p < 0.05, **p < 0.01, and ***p < 0.001 compared with stimulated concentrations in the absence of antibody by Bonferroni's t test.

IL-4R Localization and Expression

To investigate whether changes in IL-4R expression could account for synergistic increases in eotaxin secretion by human airway smooth muscle, the localization, relative receptor density, and fraction of cells bearing IL-4R were examined in cells treated for 24 hours with 10 ng/ml IL-1 and IL-13 either alone or in combination. In addition, because a functional IL-4R can consist of IL-4R, the common IL-4 and IL-13 binding element, and the -chain of the IL-2R (c), expression of both chains was examined in whole cell protein lysates.

In three separate experiments, constitutive specific cell surface binding of biotinylated IL-4 was detected on live nonpermeablized cultured myocytes, determined by avidin-FITC labeling of specifically bound biotinylated rhIL-4. Punctate, uniform staining was detected on almost all cells and to a lesser extent was also present on the surrounding extracellular matrix (Figure 5A), possibly reflecting the presence of secreted forms of the IL-4R (28). Cross-competition with 50-ng/ml rhIL-13 reduced detection of specifically bound biotinylated rhIL-4 (Figure 5B), and substitution of biotinylated rhIL-4 with an irrelevant IL-4R ligand (biotinylated soybean trypsin inhibitor) resulted in little or no detected fluorescence (Figure 5C). Similarly, preabsorbtion of IL-4 with an antihuman IL-4 blocking polyclonal antibody (Figure 5D) resulted in no detected staining. Treatment of cells for 24 hours with 10 ng/ml IL-13 or IL-4 alone or in combination with IL-1 (10 ng/ml) did not alter the observed degree or pattern of IL-4-FITC binding (not shown). Staining for c was not detected in unstimulated cells and was not induced by treatment with any of the cytokines examined (not shown).

View larger version (124K): [in this window] [in a new window]   Figure 5.   Constitutive localization of specific binding of IL-4 in cultures of human airway smooth muscle cells. Live, nonpermeablized cells were incubated with (A) biotinylated rhIL-4, (B) biotinylated rhIL-4 and 50 ng/ml rhIL-13 in combination, (C) biotinylated soybean trypsin inhibitor (an irrelevant IL-4R ligand), or (D) biotinylated rhIL-4 that had previously been preabsorbed with an antihuman IL-4 blocking polyclonal antibody. See METHODS for details. Bar represents 50 µm. Photomicrographs are representative of three independent experiments using cells cultured from three separate patients, cell passages 3-4.

In similar experiments using intact live cells cultured from five patients and harvested using 0.5 mM EDTA, the fraction of cells bearing IL-4R and the relative receptor density were examined by flow cytometric analysis. Unstimulated, positively labeled cells showed a uniform, sharp, single peak of fluorescence compared with the biotinylated irrelevant IL-4R ligand negative control, which showed relatively dull staining. More than 92.8 ± 2.5% of total gated live cells were positive for specific IL-4 binding (Figure 6A). Binding of IL-4 was completely prevented by preabsorbtion with an antihuman IL-4 blocking polyclonal antibody. Under these conditions, only 2% of cells bound IL-4. Similarly, in two separate experiments, the fraction of cells that specifically bound IL-4 fell to less than 5% of the total population after harvesting with trypsin (0.02%)-containing media, indicating the susceptibility of the IL-4R to trypsin digestion (data not shown). Cross-competition binding by 50 ng/ml rhIL-4 reduced both the percentage of cells binding biotinylated IL-4 (control 88.9 ± 4.9 versus treated 4.03 ± 1.1, p < 0.001, n = 3) and the geometric mean fluorescence intensity (control 80.0 ± 7.2 versus treated 19.1 ± 5.6, p < 0.05, n = 3). rhIL-13 (50 ng/ml) also cross-competed for binding with biotinylated IL-4, reducing the geometric mean fluorescence intensity (control 80.0 ± 7.2 versus treated 28.5 ± 5.1, p < 0.05, n = 3).

View larger version (26K): [in this window] [in a new window]   Figure 6.   Single labeling flow cytometric analysis of biotinylated rhIL-4 binding to human airway smooth muscle cells. In (A) data from a typical experiment are shown. Growth-arrested, live, nonpermeablized cells were incubated with biotinylated rhIL-4 (IL-4), biotinylated soybean trypsin inhibitor (irrelevant ligand), or biotinylated rhIL-4 that had previously been preabsorbed with an antihuman IL-4 blocking polyclonal antibody (IL-4 block) or were left unstained (no stain). Numbers in gates are mean ± SEM (n = 5) of the proportion of cells positive for specific IL-4 binding. In (B) changes in mean fluorescence intensities due to treatment of cells for 24 hours with 10 ng/ml of IL-13 or IL-4 in combination with IL-1 are also depicted. In both panels data are mean ± SEM from independent experiments using cells cultured from five separate patients, cell passages 2-4.

Treatment of the cells for 24 hours with IL-1 or IL-13 (10 ng/ml) alone, or in combination, did not significantly (p > 0.05, n = 5) alter the geometric mean fluorescence intensity of specific IL-4 binding, suggesting no change in IL-4R density following treatment with these cytokines. A small reduction in IL-4R receptor density was observed after combined treatment of the cells with IL-4 and IL-1, but this was not statistically significant (p > 0.05, n = 5; Figure 6B). The proportion of gated live cells binding IL-4 was always more than 90% and did not change with any of the cytokine treatments (p > 0.05, n = 5). As in the immunocytochemical localization studies, cell surface c expression was not detected by flow cytometry in unstimulated cells or in cells stimulated with any of the cytokine treatments (not shown).

In three experiments, Western immunoblot detection of the 140-kD IL-4R -chain revealed no change in expression from unstimulated cells after treatment with IL-13 or IL-4 alone or in combination with IL-1. In the same lysates, expression of the 64-kD c was not detected but was detected in lysates of unstimulated peripheral blood human eosinophils (Figure 7), indicating that the lack of expression in airway smooth muscles cell lysates was not due simply to a problem with the antibody or the detection conditions.

View larger version (30K): [in this window] [in a new window]   Figure 7.   Expression by Western immunoblot detection of IL-4R and c in cultured human airway smooth muscle and peripheral blood unstimulated human eosinophils (Eos) whole cell protein lysates. Growth-arrested cells were left unstimulated or were treated for 24 hours with 10 ng/ml of cytokines alone and in combination. Membranes were re-probed with a monoclonal antiGAPDH antibody. Autoradiographs are representative of independent experiments using cells cultured from three different donors, cell passages 4-5.

STAT6 and MAP Kinase Activation

In addition to possible changes in IL-4R expression, we tested the hypothesis that enhanced intracellular signaling through the IL-4R might also account for synergistic increases in eotaxin release from airway smooth muscle. In three separate experiments, exposure of cells to IL-13 (10 ng/ml) for 15 minutes induced a 16.6-fold increase in tyrosine phosphorylation (normalized for GAPDH concentrations) of STAT6 protein compared with unstimulated cells. In contrast, IL-1 (10 ng/ml) caused a 2.1-fold increase in STAT6 activation (p < 0.05). The extent of STAT6 activation when these cytokines were combined was no more than additive, reaching a 20.1-fold increase compared with unstimulated cells (Figure 8A). A similar additive increase in STAT6 phosphorylation was observed with IL-4 and IL-1 in combination (22.5-fold above basal), compared with IL-4 (19.2-fold) and IL-1 (2.1-fold) alone. In all cases, STAT6 activation induced by either IL-13 or IL-4 alone and in combination with IL-1 was inhibited (p < 0.001) by pretreatment of the cells for 30 minutes with 1 µg/ml of the neutralizing anti-IL-4R antibody (Figures 8A and 8B). Little or no effect of the blocking antibody could be detected on IL-1-induced STAT6 activation (Figure 8A).

View larger version (30K): [in this window] [in a new window]   Figure 8.   Effects of IL-13 or IL-4 in combination with IL-1 on STAT6 tyrosine phosphorylation in human airway smooth muscle. Growth- arrested cells were incubated with medium alone (unstimulated) or were treated with cytokines (10 ng/ml for 15 min) in the absence or presence of 30 minutes pretreatment with 1 µg/ml of a monoclonal anti-IL-4R blocking antibody (IL-4R). Proteins in cell lysates were separated by SDS-PAGE and immunoblotted using an antibody phospho-specificto STAT6. Membranes were reprobed with a monoclonal anti-GAPDH antibody. Bars in the graph (A) correspond to lanes in the autoradiograph above and represent mean ± SEM of -fold increases above basal STAT6 phosphorylation, normalized for GAPDH concentrations, from independent experiments using cells cultured from three separate patients, cell passages 3-5. In (B) an autoradiograph from a single experiment is shown, cell passage 6. ***p < 0.001 compared with stimulated concentrations in the absence of antibody by Bonferroni's t test.

In addition to STAT6 activation, activation of members of the MAP kinase superfamily was examined (Figure 9). Amounts of phosphorylated threonine and tyrosine of p42/p44 ERK (normalized for GAPDH concentrations) in cells stimulated with 10 ng/ml of IL-13 or IL-4 alone for 15 minutes were increased 10.4- and 12.2-fold above basal, respectively, compared with an 18.4-fold increase after stimulation with IL-1 alone. Increases in p38 MAP kinase activation were no more than 7.5-fold for IL-13 and 7.6-fold for IL-4, but were 17.5-fold for IL-1 alone. Less than 4-fold activation in SAP/JNK occurred with either IL-13 or IL-4, compared with 17.0-fold with IL-1. The extent of further increases in MAP kinase activation, detected after stimulation with IL-13 or IL-4 in combination with IL-1, was no more than additive (Figure 9). p42/p44 ERK, p38, and SAPK/JNK activation by either IL-13 or IL-4 alone were inhibited to control concentrations by pretreatment of the cells for 30 minutes with 1 µg/ml of the neutralizing anti-IL-4R antibody. No effect with this antibody was found on IL-1-stimulated activation of the p42/p44 ERK, p38, or SAPK/JNK MAP kinases (Figure 9). Furthermore, the antibody had very little effect on activation of p42/p44 ERK, p38, and SAPK/JNK when IL-1 was present with IL-4 or IL-13, compared with MAP kinase activation due to IL-1 alone (not shown).

View larger version (44K): [in this window] [in a new window]   Figure 9.   Effects of IL-13 or IL-4 in combination with IL-1 on MAP kinase activation in human airway smooth muscle. Growth-arrested cells were incubated with medium alone (unstimulated) or were treated with cytokines (10 ng/ml for 15 min) in the absence or presence of 30 minutes pretreatment with 1 µg/ml of a monoclonal anti-IL-4R blocking antibody (IL-4R). Proteins in cell lysates were separated by SDS-PAGE before immunoblot detection of phosphorylated p42/p44 ERK (upper panel ), p38 MAP kinase (middle panel ), and SAPK/JNK (p46 and p54 isoforms, lower panel ). Membranes were reprobed with a monoclonal antiGAPDH antibody. Autoradiographs are representative of independent experiments using cells cultured from three different donors, cell passages 4-5. Bars in the graph correspond to lanes in the autoradiographs above and represent mean -fold increases above basal phosphorylation levels for each MAP kinase, normalized for GAPDH concentrations. Note that the bars represent values that are superimposed and are not stacked on one another. Error bars are omitted for clarity but fell within 15% of stated means.

MAP Kinase Inhibitors Attenuate Synergistic Eotaxin Release

Finally, to determine whether MAP kinase activation was a necessary event in the synergy, U 0126 and SB 202190, potent and specific inhibitors of p42/p44 ERK and p38 MAP kinases, respectively (see [26] for references), were examined on the release of eotaxin induced by IL-13 or IL-4 alone and in combination with IL-1 (Figure 10). Preincubation for 30 minutes with either U 0126 (1 nM-10 µM) or SB 202190 (1 nM-10 µM) inhibited (p < 0.001, n = 4) synergistic increases in eotaxin release induced by IL-13 (10 ng/ml) and IL-1 (1 ng/ml) (Figure 10A). The maximum inhibitory effect of both compounds occurred above 10 µM. Interpolated IC₅₀ values for this effect were 497 ± 23 nM (U 0126) and 633 ± 87 nM (SB 202190). At 10 µM, both compounds inhibited (p < 0.05-0.001, n = 4) eotaxin release induced by IL-1 or IL-13 alone (Figure 10A). Similar data were obtained when IL-4 (1 ng/ml) was examined in place of IL-13 (Figure 10B). IC₅₀ values for U 0126 and SB 202190 against eotaxin release induced by IL-4 and IL-1 in combination were 173 ± 12 nM and 321 ± 41 nM, respectively. As with IL-13 alone, both compounds inhibited (p < 0.01, n = 4) IL-4-stimulated eotaxin release by up to 50% (Figure 10).

View larger version (21K): [in this window] [in a new window]   Figure 10.   Inhibitors of p42/ p44 ERK (U 0126, open circles) or p38 MAP kinase (SB 2020190, closed circles) activation attenuate release of eotaxin from human airway smooth muscle cells after stimulation with 10 ng/ml IL-13 (A) or 10 ng/ml IL-4 (B) in combination with 1 ng/ml IL-1 for 24 h. Bars represent eotaxin concentrations in cell-conditioned supernatants after stimulation with cytokines alone in the absence (open bars) or presence of 10 µM U 0126 (filled bars) or 10 µM SB 202190 (hatched bars). Data are expressed as a percentage of the control response to IL-1 alone and represent mean ± SEM of duplicate values from independent experiments using cells cultured from four different donors, cell passages 3-5. *p < 0.05, **p < 0.01, and ***p < 0.001 compared with stimulated concentrations in the absence of inhibitor by Bonferroni's t test.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

IL-13 and IL-4 are pleiotropic, proinflammatory cytokines whose biologic activities often overlap. Increasing evidence supports key roles for IL-13 in the pathogenesis of atopic disease and airways hyperresponsiveness. Evidence for IL-4R expression on airway smooth muscle has been circumstantial and based largely on observations such as inhibition of mitogen-stimulated proliferation by IL-4 (29) and inhibition of RANTES and monocyte chemotactic protein gene expression and release (1, 30). In the present study, we demonstrate that stimulation of cultured human airway smooth muscle cells with IL-13 or IL-4 induces release of eotaxin, with concentrations approaching those seen with IL-1 stimulation, but unlike IL-1, neither IL-13 or IL-4 induce release of GM-CSF, RANTES, or IL-8 protein measured in the same samples of cell-conditioned medium. In addition, marked synergy is found between IL-13 or IL-4 and IL-1 for eotaxin release that again does not extend to GM-CSF, RANTES, or IL-8. Concentrations of IL-- stimulated RANTES are reduced by both IL-13 and IL-4, consistent with an earlier study in human airway smooth muscle in which RANTES release was induced by a mixture of IL-1 and TNF (1). Eotaxin release induced by IL-13 or IL-4 is prevented by a monoclonal antibody neutralizing to the IL-4R but not to the IL-2R -chain (c). Likewise, synergy between IL-13 or IL-4 and IL-1 could be prevented by the anti-IL-4R antibody. In flow cytometric studies, stimulation with IL-13 or IL-4 alone and in combination with IL-1 did not change either the fraction of cells expressing IL-4R or the degree of specific IL-4 binding. Western immunoblot for IL-4R in whole cell lysates affirmed these findings. Expression of c was not detected in cultured airway smooth muscle cells. Collectively, these data suggest that selective release of eotaxin induced either by IL-13 or IL-4 alone or in combination with IL-1 is dependent on activation of a constitutive cell surface IL-4R. Furthermore, phosphorylation of p42/p44 ERK, p38, and SAPK/JNK MAP kinases or activation of STAT6 by either IL-13 or IL-4 is dependent on IL-4R and was found to be no more than additive when examined in combination with IL-1. Release of eotaxin after stimulation by IL-13 or IL-4 alone, or when combined with IL-1, is prevented by either U 0126 or SB 202190, indicating that eotaxin release is dependent on activation of both p42/p44 ERK and p38 MAP kinase pathways. Thus, we have characterized the biochemical expression and function of the IL-4R in cultured airway smooth muscle and provided a possible link to selective induction of eotaxin release and the activation of multiple intracellular activation pathways after stimulation with IL-4 and IL-13.

IL-4 and IL-1, can be found in elevated concentrations following allergen provocation in allergic asthma (20, 31-33), whereas significant elevations in IL-13 are found in the airways of patients with both allergic and nonallergic asthma (20, 34). It has been proposed that these cytokines are produced locally in the bronchoalveolar and bronchial submucosal compartments by several cell types following allergen provocation (32, 33, 35, 36). Thus, the simultaneous presence of these cytokines in tissues and bronchoalveolar fluid may contribute synergistically to the release of relatively large amounts of the C-C chemokine eotaxin from airway smooth muscle cells, resulting in selective recruitment of eosinophils, a key pathologic feature in nonimmunized mice following intratracheal administration of IL-13 (19) and of pulmonary transgene overexpression of IL-13 (23). Synergy between IL-13 and IL-1 has recently been shown for induction of the IL-1 receptor antagonist by human hepatoma HepG2 cells (37). A similar synergistic interaction is reported between IL-13 and TNF for upregulation of VCAM-1 on endothelial cells (38, 39), for activation of eosinophils (40), and, most recently, for the release of eotaxin from human keratocytes (41) and airway epithelial cells (42, 43). The amount of immunoreactive eotaxin produced by either IL-13 or IL-4 in combination with IL-1 from human airway smooth muscle cells is approximately 20,000 times greater than that induced by cytokine-stimulated airway epithelial cells, as reported by Lilly and colleagues (44). This observation, together with the finding that very low concentrations of IL-4 (EC₅₀ 60-70 ng/ml) are required to induce eotaxin release, suggests that under specific conditions airway smooth muscle could be a major source of this chemokine in the airway.

The effects of IL-4 on target cells are mediated following activation of IL-4R, which comprise dimers of IL-4R -chain (IL-4R) and -chain (c), the latter being common to several other cytokine receptors (IL-2R, IL-7R, IL-9R, and IL-15R [45]), and together are defined as the Type I receptor (46). Two components of the IL-13 receptor (IL-13R) have been identified, comprising an IL-13R1 and IL-13R2-chain. IL-13R1 binds IL-13 with low affinity, but a heterodimer of IL-13R1 and IL-4R acts as a high affinity functional receptor for both human IL-13 and IL-4, and this is known as the Type II IL-4R (46). IL-13R1 does not form dimers with c, and IL-13R2 does not dimerize with any other receptor moieties and does not appear to signal (28, 46). Thus, IL-4 and IL-13 have overlapping, but not identical, effector profiles (47) due in part to a shared and essential requirement of the 140-kD IL-4R subunit in multimeric IL-13 and IL-4R (47, 48). Consistent with the involvement of Type II IL-4R in mediating selective release of eotaxin from human airway smooth muscle, we found a near identical profile of activity between IL-13 and IL-4. This was true not only for their ability to amplify IL-1-stimulated eotaxin release but also for that to limit the release of RANTES in the same cell-conditioned medium samples. More importantly, eotaxin release induced by either IL-4 or IL-13, but not by IL-1, was abolished by an anti-IL-4R neutralizing monoclonal antibody and was inhibited to concentrations similar to those due to IL-1 alone when the antibody was examined in cells stimulated by IL-13 or IL-4 in combination with IL-1. Constitutive cell surface expression of IL-4R was demonstrated by indirect fluorescence cytochemical labeling and by flow cytometric analysis of biotinylated IL-4 specifically bound to more than 90% of airway smooth muscle cells. Evidence of specific Type II IL-4R binding was implied by the absence of binding of an irrelevant ligand and by displacement of biotinylated IL-4 by either IL-4 or IL-13. This was further corroborated by the finding that there was constitutive protein expression of the 140-kD IL-4R common binding element in airway smooth muscle whole cell lysates in the absence of any detectable cthe Type I IL-4R specific component (46). Further functional and immunologic-based investigations also failed to detect c expression, suggesting that the major receptor for IL-4 expressed on cultured human airway smooth muscle cells is the Type II IL-4R. Thus, our data agree with similar recent findings by Laporte and colleagues (49), who demonstrated constitutive expression for IL-4R, IL-13R1, and IL-13R2, but not c at the mRNA level, and suggest that in cultured human airway smooth muscle cells the sole dimerization partner for IL-4R is IL-13R1. Moreover, because IL-13R2 is not thought to be capable of signaling (28, 46), data from our study and from Laporte and colleagues (49) attest that the Type II IL-4R, the dimer composed of IL-13R1 and IL-4R, is the sole signaling receptor for both IL-4 and IL-13 in human airway smooth muscle cells. Expression of the nonsignaling protein product of IL-13R2 mRNA, detected in these cells (49), may allow a `sink' for bound IL-13 before bound ligand is donated to IL-13R1, and relative differences in expression of these IL-13R subunits may explain the observed greater potency of IL-4 over IL-13 in releasing eotaxin.

Several reports on T and B cells and other immune cells have demonstrated that IL-4R gene induction and surface protein expression can be upregulated following activation of the receptor by IL-4 (50, 51). Having determined biochemically and functionally that cultured human airway smooth muscle cells expressed the type II IL-4R, we reasoned that changes in IL-4R expression might account for synergistic increases in eotaxin secretion from these cells. However, this was not the case, as IL-4R density, as well as the fraction of cells bearing IL-4R, and expression of total cellular IL-4R-chain protein were unchanged by overnight treatment with cytokine combinations that caused synergistic eotaxin release. At present, changes in expression of the other component of the Type II IL-4R complex, IL-13R1, that could contribute to the synergy cannot be ruled out until appropriate antibodies become readily available.

IL-4R activation in many cell types triggers the function of several intracellular signaling molecules. STAT proteins are phosphorylated following ligand binding to IL-4R through receptor-associated Janus kinase (JAK) proteins. JAK-induced tyrosine phosphorylation of STAT proteins induces homo-dimerization and their translocation to the nucleus, where dimerized STAT proteins can transcriptionally modify expression of target genes. STAT6 has been shown to be essential for IL-4 and IL-13 signaling (52). In addition, induction of bronchial eosinophilic inflammation and airway hyperreactivity following repeated allergen challenge is abolished in STAT6-deficient mice (53, 54). Surprisingly, little is known of the role of STAT activation in airway smooth muscle. Here, we demonstrate that untreated human airway smooth muscle cells contain little detectable phosphorylated concentrations of the IL-4 and IL-13-specific transcription factor, STAT6. In contrast, phosphorylated STAT6 was readily detected at 15 minutes stimulation with IL-4 or IL-13, but not with IL-1. Another study has also demonstrated STAT6 activation in human airway smooth muscle cells after stimulation with IL-13 or IL-4 (49). We have expanded this observation and demonstrated an essential role for the Type II IL-4R mediating this response, because in the absence of detectable c, STAT6 activation by either IL-13 or IL-4 was abolished by treatment of the cells with a blocking antibody against IL-4R. Correlation with induction of eotaxin release by IL-13 or IL-4 with STAT6 phosphorylation suggests that IL-4R-mediated activation of the JAK/STAT6 pathway may be involved in the regulation of eotaxin release from human airway smooth muscle cells. This possibility is supported by recent observations made by Hoeck and Woisetscläger (55), wherein eotaxin production by human skin fibroblasts stimulated with IL-4 and TNF in combination was prevented after transfection with a trans-dominant negative STAT6 protein. Moreover, in the same study, two other TNF- and IL-1-inducible genes, MCP-1 and IL-8, were unaffected by STAT6 intervention. Matsukura and colleagues (56) have recently reported that in human airway epithelial cells STAT6 binds to the proximal region of the eotaxin promotor and contributes to eotaxin transcriptional regulation induced by IL-4. Indeed, a similar requirement of functional STAT6 protein for activation of the eotaxin gene but not GM-CSF, IL-8, or RANTES may explain the selective induction of eotaxin that we observed in human airway smooth muscle cells after stimulation by IL-4 or IL-13. However, confirmation of this and its importance in the synergy with IL-1 requires future direct STAT6 intervention strategies to be examined in airway smooth muscle.

On the basis of specific chemical inhibitors of MAP kinase signaling cascades, we have recently reported that IL-1-stimulated eotaxin release from human airway smooth muscle cells is dependent on activation of the p42- and p44-kD extracellular signal-regulated kinases (ERK) (collectively defined as p42/ p44 ERK) and activation of p38 MAP kinase (see Figure 10 and [26]). Few studies have investigated MAP kinase activation by either IL-13 or IL-4. Depending on the cell type, recent examples of stimulation (57, 58) or inhibition (59, 60) of MAP kinases can be found. Accordingly, we examined whether the basis of the observed synergism between IL-1 and IL-13 or IL-4 for eotaxin release involved MAP kinase activation. Consistent with recent findings by Laporte and colleagues (49), we observed that IL-13 and IL-4 induced the activation of p42/ p44 ERK. In addition, we found that this was IL-4R-dependent and that the p38 MAP kinase and, to a lesser extent, the p46- to p54-kD c-Jun amino-terminal kinase (JNK) or stress-activated protein kinase (SAPK) were also activated after ligation of IL-4R. Despite degrees of phosphorylation being no more than additive when IL-13 or IL-4 was combined with IL-1, the importance of p42/p44 and p38 MAP kinase activation in release of eotaxin after stimulation by IL-13 and IL-4 alone or when combined with IL-1 was confirmed by the finding that both the MEK inhibitor U 0126 and the p38 inhibitor SB 202190 attenuated eotaxin release by around 50% to 60%. Further combined intervention studies are required to determine the relative importance of these MAP kinase pathways and their relationship to STAT6 phosphorylation events in the overall response.

In conclusion, our data suggest that both IL-13 and IL-4 induce selective release of eotaxin from cultured human airway smooth muscle via a mechanism that is dependent on ligand-specific binding to constitutive cell surface IL-4R and activation of the common and essential IL-4R. Additionally, IL-4R activation in human airway smooth muscle cells is associated with activation of multiple intracellular pathways that include phosphorylation of STAT6 and key elements of the MAP kinase signaling cascade such as p42/p44 ERK and p38 MAP kinase and to a lesser extent SAPK/JNK. This is the first demonstration in smooth muscle cells that IL-13 and IL-4 also induce activation of the p38 and SAPK/JNK MAP kinase pathways and that as with p42/p44 ERK and STAT6, their activation was mediated by the IL-4R. We also demonstrate a marked IL-4R-dependent synergy between IL-13 or IL-4 with IL-1 for the release of eotaxin. The mechanism responsible for the synergy was not associated with either increased IL-4R expression or synergistic increases in phosphorylation of STAT6 or p42/p44 ERK, p38 MAP kinase, or SAPK/JNK. However, selective inhibitors of MAP kinase activation demonstrated that as with the effects of IL-13 or IL-4 alone, the synergy with IL-1 was dependent on activation of both p42/p44 ERK- and p38 MAP kinase-dependent pathways. Thus, synergy between IL-4 or IL-13 and IL-1 likely results from coordinate activation of these multiple signals at a locus that is downstream of the observed phosphorylation events, perhaps at the transcriptional or translational level (55, 56). Understanding the cellular mechanisms and intracellular pathways that modulate the release of cytokines from airway smooth muscle may offer new therapeutic strategies targeted at cellular recruitment and at the airway smooth muscle remodeling process in the airway wall of the diseased lung.