Introduction

Abstract Introduction Materials and Methods Results Discussion References

Asthma is characterized by reversible airway obstruction and desquamation of the airway epithelium that is accompanied by infiltration of proinflammatory cells such as macrophages, lymphocytes, and eosinophils (1). Recent studies of the pathogenesis of asthma have centered on the chronic inflammatory process and the role it may play in the development of structural changes in the airway wall (4). Traditionally, the role of the smooth muscle cell in airway inflammation has been regarded as passive, contributing to the pathogenesis of asthma solely by its contractile properties. Several reports have shown, in addition to its contractile function, that in asthma airway smooth muscle can undergo hyperplasia and/or hypertrophy, leading to structural changes in the airway wall that contribute to the development of persistent airway obstruction and increased nonspecific airway hyperresponsiveness (5, 6). This apparent functional diversity of airway smooth muscle has prompted interest in the possibility that there is plasticity in its function that may be related to the severity of the tissue remodeling process during chronic inflammation of the airway wall (7, 8).

Additional reports from cell culture-based studies are emerging to suggest a further role for airway smooth muscle in airway inflammation by acting as an important source of proinflammatory and bronchoprotective mediators (8, 9). Human airway smooth muscle cells in culture, when treated with proinflammatory cytokines, have increased expression of cyclooxygenase (COX)-2 (10, 11), the inducible isoform of the enzyme, which is the major isoenzyme associated with inflammation. Other preliminary studies of human airway smooth muscle stimulated with proinflammatory cytokines have shown increased expression and release of regulated on activation, normal T cell expressed and secreted (RANTES) (12), interleukin (IL)-8 (13, 14), eotaxin (12, 15), and other cytokines such as IL-6 (16). RANTES, IL-8, and eotaxin are important chemokines for activation of eosinophils (17), critical effector cells in the pathogenesis of asthma (1). RANTES is a potent chemoattractant for eosinophils as well as for other cell types observed in allergic inflammation, including monocytes and memory T lymphocytes (17). IL-8, in addition to its action on neutrophils, is also a potent eosinophil chemoattractant (18). Eotaxin, however, is a highly selective chemoattractant for eosinophils (19). Production by airway smooth muscle of RANTES, IL-8, and eotaxin implies a role for these structural cells to participate directly in the inflammatory process through recruitment and activation of eosinophils. In addition to recruitment of eosinophils by chemoattractants, enhanced survival of infiltrating eosinophils is also thought to contribute to airway inflammation in asthma (20). As more evidence emerges that airway smooth muscle in vitro may be an important source of eosinophil-activating cytokines, we have sought to investigate whether human airway smooth muscle cells contribute to airway inflammation by also producing cytokines that prolong the survival of eosinophils.

The mechanisms underlying chronic inflammation of the airways are complex. Accumulating evidence suggests the involvement of cytokines in both the induction and perpetuation of these processes. Our own immunohistochemical studies of bronchial biopsies from asthmatic patients have shown increased infiltration of macrophages into the airways (2), as well as increased expression of IL-1 in the bronchial epithelium and marked increases in IL-1-producing cells in the asthmatic submucosa (21). Studies of alveolar macrophages from asthmatic subjects have also shown that IL-1 expression is upregulated (22). Similarly, in patients with symptomatic asthma there are increased levels of IL-1 in the bronchoalveolar lavage fluid compared with normal subjects or patients with asymptomatic asthma (23, 24). In the present study, we have chosen to investigate the effects of IL-1 on airway smooth muscle cells because of its potential importance in the pathology of asthma and its ability to stimulate production of eosinophil-activating cytokines such as RANTES, IL-8, and eotaxin. Furthermore, upregulation of IL-6 and IL-8 messenger RNA (mRNA) following exposure of human airway smooth muscle cells to atopic/asthmatic human serum was shown in a recent report to be blocked by the IL-1 receptor antagonist (IL-1ra), suggesting a role for serum IL-1 in mediating this effect (16).

To investigate the potential for airway wall smooth muscle to drive mucosal inflammation in asthma, conditioned medium from human airway smooth muscle cells stimulated with IL-1 was examined on eosinophil survival in vitro. Our results indicate that cultured human airway smooth muscle cells stimulated with IL-1 produce an eosinophil survival-enhancing activity that is functionally indistinguishable from granulocyte-macrophage colony stimulating factor (GM-CSF).

	Materials and Methods

Abstract Introduction Materials and Methods Results Discussion References

Materials

All chemicals were of analytical grade or higher. Recombinant human IL-1 and IL-1ra were purchased from R&D Systems (Abingdon, UK). Goat polyclonal immunoglobulin G (IgG) antibodies to human GM-CSF, IL-3, and IL-5 for the neutralization studies were also purchased from R&D Systems. Recombinant human GM-CSF was purchased from Boehringer-Mannheim (Lewes, UK). All cell culture media (Dulbecco's modified Eagle's medium [DMEM], minimal essential medium [MEM], and RPMI 1640), FCS, and cell culture reagents were obtained from Gibco Life Technologies (Paisley, UK). Collagenase (type CLS 1) and elastase (type 1) were obtained from Worthington Biochemical Corporation (Freehold, NJ). All cell culture plasticware was purchased from Falcon Labware (Becton Dickinson, Oxford, UK). Percoll was obtained from Pharmacia (St. Albans, UK). All other chemical reagents were obtained from Sigma (Poole, UK).

Airway Smooth Muscle Cell Isolation and Culture

Human bronchial smooth muscle was obtained from the lobar or main bronchus of 17 patients of either sex (mean age 63 ± 4 yr; range 21-77 yr) undergoing lung resection for carcinoma of the bronchus, as previously described in detail (25) with modifications (26). After removal of the epithelium, portions of the smooth muscle not invaded by the carcinoma were dissected free of adherent connective and parenchymal tissue under aseptic conditions in Hank's balanced salt solution. The smooth muscle was digested in 2 ml DMEM (supplemented with 1 mM sodium pyruvate, 2 mM L-glutamine, 1:100 nonessential amino acid mixture, 50 µg/ml gentamicin, and 1.5 µg/ml amphotericin B) containing 1 µM insulin, 5 µg/ml transferrin, 100 µM ascorbate, 1 mg/ml bovine serum albumin (BSA), and 3 mg/ml collagenase. After 30 min incubation at 37°C, the tissue was transferred into a similar enzyme mixture containing 3 U/ml elastase and further dissected under a microscope to remove any unwanted fibrous tissue from within the muscle itself. The tissue was chopped finely (approximately 1 mm³) and returned to the incubator for a further 18 h until fully digested. The resulting cell suspension was centrifuged (200 × g for 5 min) and the pellet was washed in supplemented DMEM containing 10% FCS. Cells were seeded at 5 × 10⁵ viable cells in 25-cm² culture flasks and maintained in a humidified atmosphere at 37°C in 5% CO₂/95% air. Fresh medium was replaced every 72 h.

After 14 to 18 d in culture, airway smooth muscle cells grew to confluence and were subcultured by incubating each flask with 3.5 ml trypsin/ethylenediaminetetraacetic acid (EDTA) (0.1 mg/ml in phosphate-buffered saline) solution for 5 min before adding an equal volume of supplemented DMEM containing 10% FCS and mechanically dispersing the cells by repeated gentle pipetting. Cells were centrifuged (200 × g for 5 min), resuspended in 12 ml supplemented DMEM containing 10% FCS, and plated in a 75-cm² culture flask. Cells were usually confluent in 10 to 12 d, after which further subculture allowed half the cells to be used for experimental work and the other half to be maintained in 75-cm² flasks to continue each cell line. Cells at passages 3 to 7 were used for all experiments, during which the proliferative response to FCS (27) and growth factors (25) was unchanged.

With the use of fluorescent immunocytochemistry techniques reported elsewhere (25, 28), growth-arrested cultured human airway smooth muscle cells stained (> 95%) for both smooth muscle -actin and smooth muscle-myosin heavy chain. When examined by light and electron microscopy (29), these cells displayed all of the reported characteristics of viable smooth muscle cells in culture (30).

Airway Smooth Muscle Cell Stimulation

Human airway smooth muscle cells were harvested from 75-cm² flasks by treatment with trypsin/EDTA during passages 3 to 7. To maximize compatibility with the eosinophil survival assay, airway smooth muscle cells were switched to an RPMI 1640-based growth medium. The growth characteristics of human airway smooth muscle cells in this medium were virtually identical to those of cells grown in the DMEM-based medium (M. P. Hallsworth and S. J. Hirst, unpublished observations). Harvested airway smooth muscle cells were washed in RPMI 1640 supplemented with 25 mM N-2-hydroxyethylpiperazine-N'-ethane sulfonic acid, 2 mM L-glutamine, and 100 U/ml:100 µg/ml penicillin/streptomycin (supplemented RPMI 1640) with the addition of 10% FCS, and seeded into 24-well plastic tissue-culture plates at a density of 2 × 10⁴ cells/well. When the cells approached confluence, growth was arrested by replacing the medium with 0.5 ml supplemented RPMI 1640 with the addition of 1 µM insulin, 5 µg/ml transferrin, 100 µM ascorbate, and 1 mg/ml BSA. After 72 h, airway smooth muscle cell monolayers were washed (2 × 0.5 ml) with supplemented RPMI 1640 containing 1 mg/ml BSA and then cultured in similar medium (0.5 ml) for a further period of up to 96 h in the absence or presence of varying concentrations of recombinant human (rh) IL-1 and any other agents under investigation. Cell-conditioned medium (0.5 ml) was collected, and cell-free supernatants were divided into aliquots and stored at 20°C until assayed for either eosinophil survival-enhancing activity or measurements of cytokine levels by enzyme-linked immunosorbent assay (ELISA). Airway smooth muscle cell viability was then determined by mitochondrial-dependent reduction of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) to formazan (31). Following collection of conditioned medium from airway smooth muscle cells in culture, the cell monolayers were washed (2 × 0.5 ml) in supplemented RPMI 1640 containing 10% FCS. MTT (100 µl of 5 mg/ml) was added to 0.5 ml RPMI 1640 containing 10% FCS and incubated at 37°C for 5 h. The resulting formazan product was solubilized by further addition of 0.5 ml 10% sodium dodecyl sulfate in 0.1 M HCl and quantified using a dual wavelength spectrophotometer (Anthos HTII; Salzburg, Austria). Cell viability was expressed as arbitrary optical density (OD) units.

Eosinophil Isolation and Purification

Eosinophils were isolated from 100 ml EDTA (0.2 M) anticoagulated peripheral blood from patients who had bronchial asthma or allergic rhinitis. After sedimentation of erythrocytes with 0.2 vol of 6% dextran in 0.9% saline for 45 min at room temperature, the leukocyte-rich plasma was aspirated and eosinophils were separated by depletion of neutrophils with anti-CD16-coated magnetic microbeads using the MACS isolation system (Miltenyi Biotech, Camberley, UK) according to the method of Hansel and colleagues (32). Briefly, leukocytes were washed by centrifugation at 400 × g for 10 min in MEM containing 2% FCS and were resuspended in 10 ml MEM/FCS. The cells were layered onto 20 ml of Percoll (density 1.088 g/ml) and centrifuged at 1,000 × g for 30 min. The mononuclear cell layer was carefully removed, along with the remaining Percoll, and the granulocyte pellet was resuspended in 1 ml MEM/FCS and counted in Kimura stain. Anti-CD16-coated microbeads were added to the cell suspension at a ratio of 50 µl per 5 × 10⁷ neutrophils and the cells were incubated for 30 min at 4°C. The cell suspension was then added to the top of the MACS cell separation column (Miltenyi Biotech) in a magnetic field, and the eosinophils were washed through with 30 ml MEM/FCS. The column eluate was centrifuged at 400 × g for 10 min and the cell pellet was resuspended in 1 ml MEM/FCS and counted in Kimura stain. Eosinophil preparations of greater than 95% purity were used in all experiments.

Eosinophil Survival Assay

Freshly isolated human peripheral blood eosinophils were resuspended at a concentration of 1 × 10⁶ cells/ml in supplemented RPMI 1640 containing 10% FCS. Cell suspensions (50 µl) were cultured in a 96-well plastic tissue-culture plate containing 50 µl airway smooth muscle cell-conditioned culture medium. Eosinophils were cultured for 7 d, after which viability was assessed in duplicate wells by trypan blue exclusion. In each experiment, eosinophils were cultured in the presence of 1 ng/ml rhGM-CSF alone as a positive control and in supplemented RPMI 1640 containing 10% FCS alone as a negative control.

Measurement of Cytokine Levels by ELISA

GM-CSF levels in airway smooth muscle cell-conditioned culture medium were determined in duplicate using a specific sandwich ELISA (R&D Systems). Samples were diluted until the level of GM-CSF was within the standard curve of the assay. The concentration of GM-CSF detected was initially expressed in nanograms per milliliter and then calculated as ng/ml/10⁶ cells to correct for small differences in cell densities between patients. The limit of detection was 2.8 pg GM-CSF/ml. This assay exhibited no cross-reactivity or interference with rhIL-1.

Statistical Analysis

Data in the text and figure legends are expressed as means ± SEM of observations obtained from airway smooth muscle cells cultured from n patients. Results were compared using Student's t test for unpaired observations. When parametric criteria were not met, data were analyzed using the Mann-Whitney U Rank Sum test (SigmaStat; Jandel Scientific, Erkrath, Germany). A probability value of less than 0.05 was considered significant.

	Results

Abstract Introduction Materials and Methods Results Discussion References

Conditioned Supernatants of IL-1-Stimulated Airway Smooth Muscle Cells Contain an Eosinophil Survival-Enhancing Activity

Freshly isolated eosinophils were more than 99% viable, as assessed by trypan blue exclusion, but only 1 ± 0.2% (n = 5) of the initial population survived after 7 d of culture in medium containing 10% FCS alone. In contrast, eosinophil survival was enhanced (P < 0.001, n = 7) by up to 90-fold when cultured with medium conditioned by human airway smooth muscle cells stimulated with IL-1 (1 pg-100 ng/ml) for 96 h (Figure 1). This enhancement of eosinophil survival by conditioned medium from airway smooth muscle cells occurred in an IL-1 concentration-dependent manner. Maximum production of the eosinophil survival-enhancing activity occurred at 1 to 3 ng/ml IL-1. The concentration that produced 50% of this effect (EC₅₀) was 0.027 ± 0.004 ng/ml IL-1 (Figure 1). Conditioned medium from unstimulated airway smooth muscle cells did not enhance eosinophil survival (Figure 1), and IL-1 itself did not have any direct effect on eosinophil survival (data not shown). Airway smooth muscle viability, assessed by the colorimetric MTT reduction assay (30), was also not affected (P > 0.05; n = 5) after stimulation with 1 ng/ml IL-1 (OD: 0.529 ± 0.057 versus 0.603 ± 0.07, unstimulated versus IL-1-stimulated).

View larger version (13K): [in this window] [in a new window]   Figure 1.   Production of an eosinophil survival-enhancing activity by IL-1-stimulated human airway smooth muscle cells. Near-confluent airway smooth muscle cells were growth-arrested for 72 h before stimulation for 96 h in the absence (open circle) or presence (filled circles) of IL-1. Samples of the cell-conditioned medium were cultured for 7 d with eosinophils. Eosinophils cultured in medium alone (open bar) or in the presence of 1 ng/ml rhGM-CSF (filled bar) were used as negative and positive controls, respectively, for eosinophil survival. Points represent means ± SEM of duplicate values from independent experiments using conditioned medium from airway smooth muscle cells cultured from seven patients. Airway smooth muscle cell-conditioned medium was assayed using eosinophils from four separate patients.

Time Course of Production of the Eosinophil Survival-Enhancing Activity

Eosinophil survival-enhancing activity could be detected in airway smooth muscle cell-conditioned medium after just 3 h of stimulation with IL-1 (1 ng/ml) and reached statistical significance at 6 h (P < 0.001). This activity continued to increase before reaching a plateau around 24 h, with no decrease in activity for up to 120 h of stimulation (Figure 2). Unstimulated airway smooth muscle cells did not produce any spontaneous survival-enhancing activity over the same time period.

View larger version (13K): [in this window] [in a new window]   Figure 2.   Time-dependent production of an eosinophil survival-enhancing activity by IL-1-stimulated human airway smooth muscle cells. Growth-arrested airway smooth muscle cells were stimulated in the absence (open circles) or presence (filled circles) of IL-1 for up to 120 h. Samples of the cell-conditioned medium were then cultured for 7 d with eosinophils. Eosinophils cultured in medium alone (open bar) or in the presence of 1 ng/ml rhGM-CSF (filled bar) were used as negative and positive controls, respectively, for eosinophil survival. Points represent means ± SEM of duplicate values from independent experiments using conditioned medium from airway smooth muscle cells cultured from three patients. Airway smooth muscle cell-conditioned medium was assayed using eosinophils from three separate patients.

Identification of the Eosinophil Survival-Enhancing Activity

The hemopoetic cytokines IL-3, IL-5, and GM-CSF are commonly associated with the augmentation of eosinophil survival. We attempted to prevent the eosinophil survival-enhancing activity produced by airway smooth muscle cells with specific polyclonal neutralizing antibodies to these cytokines. The survival-enhancing activity was inhibited by increasing concentrations of a polyclonal goat antihuman antibody to GM-CSF (0.3-100 µg/ml). The concentration of antibody which inhibited the survival-enhancing activity by 50% (IC₅₀) was 6.9 ± 0.54 µg/ml. Maximum inhibition occurred at 100 µg/ml. Similar antibodies to IL-3 and IL-5, and a normal goat IgG control had no effect on the eosinophil survival-enhancing activity (Figure 3).

View larger version (15K): [in this window] [in a new window]   Figure 3.   Neutralization of the eosinophil survival-enhancing activity in conditioned medium from IL-1-stimulated airway smooth muscle cells. Eosinophils were cultured in conditioned medium from IL-1 (1 ng/ml)-stimulated airway smooth muscle cells in the absence (filled square) or presence (filled circles) of specific polyclonal goat antibodies (30-min pre-incubation) to human GM-CSF, IL-3 (filled triangles), or IL-5 (filled inverted triangles). Normal goat IgG was used as an irrelevant antibody control (filled diamonds). Eosinophils cultured in medium alone (open bar) or in the presence of 1 ng/ml rhGM-CSF (filled bar) were used as negative and positive controls, respectively, for eosinophil survival. Points represent means ± SEM of duplicate values from independent experiments using conditioned medium from airway smooth muscle cells cultured from five patients. Airway smooth muscle cell-conditioned medium was assayed using eosinophils from four separate patients.

Production of GM-CSF by IL-1-Stimulated Airway Smooth Muscle Cells

Levels of GM-CSF were measured by ELISA in samples of the same cell-conditioned medium shown to contain an eosinophil survival-enhancing activity. The concentration of GM-CSF in conditioned medium from unstimulated airway smooth muscle cells was near the limit of detection (2.8 pg/ml) of the assay. Similarly, conditioned medium from cells stimulated with tumor necrosis factor- (TNF-) (0.003-30 ng/ml) also failed to produce detectable levels of GM-CSF (data not shown). In contrast, stimulation of airway smooth muscle cells with IL-1 (0.001-100 ng/ml) resulted in a concentration-dependent increase in production of GM-CSF. Maximum production of GM-CSF in the conditioned medium was stimulated by 30 ng/ml IL-1 (P < 0.001, n = 7) (EC₅₀ 0.141 ± 0.031 ng/ml of IL-1) (Figure 4a) and was 3.56 ± 0.83 ng/ml/10⁶ cells (equivalent to 0.367 ± 0.09 ng/ml GM-CSF). A similar concentration of rhGM-CSF was able to induce maximum eosinophil survival in our assay system (Figure 4a, inset). Production of GM-CSF from airway smooth muscle cells following stimulation with IL-1 (1 ng/ml) was time-dependent reaching statistical significance (P < 0.01) at 6 h and continuing to increase up to 120 h (Figure 4b). When stimulated with TNF- (10 ng/ml), significant levels of GM-CSF (P < 0.01, n = 4) were detected in conditioned medium only after 48 h (0.198 ± 0.052 ng/ml/10⁶ cells). At 120 h, production of GM-CSF by TNF- was less than 5% of that produced in response to IL-1. Unstimulated airway smooth muscle cells did not produce GM-CSF over the same time period.

View larger version (12K): [in this window] [in a new window]   Figure 4.   Production of GM-CSF by IL-1-stimulated human airway smooth muscle cells. (a) Near-confluent airway smooth muscle cells were growth-arrested for 72 h before stimulation for 96 h in the absence (open circle) or presence (filled circles) of IL-1. Points represent means ± SEM of duplicate values obtained from experiments using conditioned medium from airway smooth muscle cells cultured from seven patients. (Inset) The effect of rhGM-CSF is shown directly on eosinophil survival at 7 d. Points represent means ± SEM of duplicate values obtained from four independent experiments. (b) Time-dependent production of GM-CSF by IL-1 (1 ng/ml)- stimulated human airway smooth muscle cells. Airway smooth muscle cells were stimulated in the absence (open circles) or presence (filled circles) of IL-1 for up to 120 h. Points represent means ± SEM of duplicate values using conditioned medium from airway smooth muscle cells cultured from four patients.

Effect of IL-1ra on IL-1-Stimulated Airway Smooth Muscle Cells

To determine whether the production of GM-CSF by human airway smooth muscle cells stimulated by IL-1 was mediated directly by a specific IL-1 receptor, cells were cultured with an optimal concentration of IL-1 (1 ng/ml) and the effect of increasing concentrations of the rhIL-1ra was determined. Supernatants were tested using the eosinophil survival assay. Recombinant human IL-1ra (0.1-300 ng/ ml) inhibited the production of the eosinophil survival- enhancing activity induced by IL-1 in a concentration- dependent manner with an IC₅₀ of 53.7 ± 13.2 ng/ml (Figure 5a). Similarly, levels of GM-CSF measured in these supernatants by ELISA were found to decrease with increasing concentrations of IL-1ra. The IC₅₀ for this effect was 19.1 ± 6.4 ng/ml IL-1ra (Figure 5b). There was no statistical difference (P > 0.05, n = 4) between IC₅₀ values for the effects of IL-1ra against either the production of the eosinophil survival-enhancing activity or the production of GM-CSF, measured by ELISA.

View larger version (12K): [in this window] [in a new window]   Figure 5.   Effect of rhIL-1ra on production of eosinophil survival-enhancing activity and GM-CSF by human airway smooth muscle cells. (a) Eosinophils were cultured in conditioned medium from IL-1 (1 ng/ml)-stimulated airway smooth muscle cells in the absence (filled square) or presence (filled circles) of rhIL-1ra. Eosinophils cultured in medium alone (open bar) or in the presence of 1 ng/ml rhGM-CSF (filled bar) were used as negative and positive controls, respectively, for eosinophil survival. (b) GM-CSF production by IL-1 (1 ng/ml)- stimulated airway smooth muscle cells was determined in the absence (filled square) or presence (filled circles) of rhIL-1ra. In both panels, points represent means ± SEM of duplicate values from independent experiments using conditioned medium from airway smooth muscle cells cultured from four patients. Conditioned medium from airway smooth muscle cells was assayed using eosinophils from two separate patients.

Effect of Indomethacin on the Eosinophil Survival-Enhancing Activity Produced by Airway Smooth Muscle Cells

IL-1 is associated with COX-2 upregulation in airway smooth muscle cells (10, 11), so we examined the effect of the nonselective COX inhibitor indomethacin on production of the eosinophil survival-enhancing activity in airway smooth muscle cell-conditioned medium. Indomethacin (0.1-10 µM) had no effect on production of the activity by unstimulated cells or on cells stimulated with an optimal concentration (1 ng/ml) of IL-1 for 96 h (Figure 6). Similarly, in the same samples of cell-conditioned medium, production of GM-CSF by IL-1 was not found to be inhibited by indomethacin (0.1-10 µM) (data not shown). To ensure that any effect of indomethacin was not transient or overcome by 96 h, the same experiments were carried out with a 24-h stimulation period. There was no difference in the data obtained at 24 h compared with 96 h (data not shown).

View larger version (14K): [in this window] [in a new window]   Figure 6.   Effect of the COX inhibitor indomethacin on production of the eosinophil survival-enhancing activity from IL-1-stimulated human airway smooth muscle cells. Airway smooth muscle cells were growth-arrested for 72 h before addition of indomethacin in the absence (open circles) or presence (filled circles) of stimulation for 96 h with IL-1 (1 ng/ml). Samples of the cell-conditioned medium were cultured for 7 d with eosinophils. Eosinophils cultured in medium alone (open bar) or in the presence of 1 ng/ml rhGM-CSF (filled bar) were used as negative and positive controls, respectively, for eosinophil survival. Points represent means ± SEM of duplicate values using conditioned medium from airway smooth muscle cells cultured from five patients. Airway smooth muscle cell-conditioned medium was assayed using eosinophils from three separate patients.

Effect of Glucocorticosteroids on Production of GM-CSF by Airway Smooth Muscle Cells

Addition of dexamethasone (1 pM-10 µM) to airway smooth muscle cell cultures stimulated with IL-1 (1 ng/ ml) produced a concentration-dependent inhibition in the production of GM-CSF from these cells, measured by ELISA (Figure 7). Maximum inhibition of IL-1-stimulated GM-CSF production from human airway smooth muscle cells occurred at 1 µM dexamethasone. The IC₅₀ for this effect was 1.14 ± 0.36 nM.

View larger version (16K): [in this window] [in a new window]   Figure 7.   Effect of dexamethasone on production of GM-CSF by IL-1-stimulated human airway smooth muscle cells. GM-CSF production by IL-1 (1 ng/ml)-stimulated airway smooth muscle cells was determined in the absence (filled square) or presence (filled circles) of dexamethasone. Points represent means ± SEM of duplicate values from independent experiments using conditioned medium from airway smooth muscle cells cultured from five patients.

	Discussion

Abstract Introduction Materials and Methods Results Discussion References

The major finding of this study was that survival of human peripheral blood eosinophils in vitro was markedly prolonged by culture with conditioned growth medium from IL-1-stimulated, but not unstimulated, human airway smooth muscle cells. This effect occurred through the production of a soluble mediator from airway smooth muscle cells that was identified as GM-CSF.

To determine the functional relevance of any bioactivity produced following stimulation of human airway smooth muscle cells with IL-1, data were obtained in the first instance using a defined bioassay system. Production of an eosinophil survival-enhancing activity was characterized using this approach and then identified by neutralization studies using specific blocking polyclonal antibodies. This was followed by detection of secreted protein by ELISA. The specificity of the production of the bioactivity and the secreted protein was confirmed using a specific receptor antagonist.

Production of the eosinophil survival-enhancing activity by airway smooth muscle cells was dependent on stimulation with IL-1. No significant eosinophil survival- enhancing activity was detected in conditioned medium from unstimulated airway smooth muscle cells. Even prolonged conditioning of the medium by unstimulated cells failed to show activity, suggesting that an inducible rather than a constitutive activity was responsible. The concentrations of IL-1 (10-1,000 pg/ml) at which we observed significant eosinophil survival-enhancing activity derived from airway smooth muscle cells were within the range likely to be present during airway inflammation. In bronchoalveolar lavage fluid of symptomatic individuals with asthma, levels of IL-1 range from 20 to 900 pg/ml (23, 24, 33). Normal plasma levels of IL-1 are often below the limit of detection (< 40 pg/ml) (34).

In the blood, eosinophils are short-lived with a half-life of 18 h (35). Their survival in the extracellular space is determined by the local environment. In vitro studies have established that survival of peripheral blood human eosinophils is dependent on the continued presence of one of at least three hemopoetic cytokines. These are IL-3 (36), IL-5 (37), and GM-CSF (38). The reduction in eosinophil survival by a neutralizing antibody to GM-CSF, but not by neutralizing antibodies to IL-3 or IL-5, suggests that the survival-enhancing activity produced by human airway smooth muscle cells following stimulation with IL-1 was due specifically to GM-CSF. In later experiments (n = 13) we determined by ELISA that the mean concentration of GM-CSF produced by airway smooth muscle cells in response to 1 ng/ml IL-1 was 2.72 ± 0.45 ng/ml/10⁶ cells (equivalent to 0.376 ± 0.097 ng/ml in this study). The concentration of the neutralizing antibody to GM-CSF, reported by the manufacturer, that was required to inhibit 50% of the bioactivity as a result of 0.5 ng/ml rhGM-CSF was 4 to 6 µg/ml. Although there are obvious difficulties in drawing comparisons between the effects of native GM-CSF and recombinant GM-CSF, these values are in good agreement with our findings in which 6.9 µg/ml of the neutralizing antibody to GM-CSF was required to inhibit 50% of the eosinophil survival-enhancing activity produced by human airway smooth muscle cells. Furthermore, the finding that higher concentrations of the neutralizing antibody to GM-CSF could abolish eosinophil survival, whereas similar concentrations of neutralizing antibodies to IL-3 or IL-5 or an IgG control antibody had no effect, suggests that the major survival-enhancing activity produced by human airway smooth muscle cells was due to GM-CSF. Our previous studies have indicated that GM-CSF also represents the major eosinophil survival-activity produced by both alveolar macrophages (39) and peripheral blood mononuclear cells (40) obtained from asthmatic subjects. Increased levels of GM-CSF have been detected in both bronchoalveolar lavage fluid (23) and airway epithelial cells (41) of asthmatic subjects. Similarly, the numbers of cells expressing elevated mRNA levels for IL-3, IL-5, and GM-CSF are increased in bronchoalveolar lavage fluid of subjects with symptomatic asthma (42, 43). Together, these observations suggest that there is upregulation of cytokine expression in asthma and that several cell types, including airway smooth muscle, can influence eosinophil survival and function.

Production of GM-CSF by airway smooth muscle cells in response to IL-1 was confirmed in the cell-conditioned culture medium using ELISA. The magnitude of GM-CSF production from IL-1-stimulated airway smooth muscle cells was comparable to the concentrations of rhGM-CSF, reported in this study and elsewhere (44), found to stimulate maximum human eosinophil survival. In addition, the onset of production of GM-CSF in the cell-conditioned medium matched the initial time course of the eosinophil survival-enhancing activity. At later time points (> 24 h), however, detected levels of GM-CSF were in excess of the level that could induce maximum eosinophil survival in our assay system. This most likely accounts for the plateau response observed at 24 h for detection of the eosinophil survival-enhancing activity despite continued GM-CSF production measured by ELISA. Thus, the concentration and temporal characteristics of the data obtained accord with GM-CSF being produced by airway smooth muscle cells and subsequently mediating prolonged eosinophil survival. The failure of conditioned medium from unstimulated airway smooth muscle cells to prolong eosinophil survival is in agreement with a recent study, in which eosinophils showed prolonged survival when cultured with conditioned medium from TNF--stimulated, but not unstimulated, human bronchial myofibroblasts (44). In a previous study, however, conditioned medium from unstimulated human lung fibroblasts was found to prolong eosinophil survival (45). While such discrepancies may be explained by different experimental conditions, it is possible that human airway smooth muscle cells in culture have more in common with bronchial myofibroblasts than lung fibroblasts, and as such are able to regulate closely their production of GM-CSF to prevent inappropriate activation of the inflammatory process in the airway wall. Although changes in the function of airway smooth muscle cells (as a result of altered phenotype toward a myofibroblast-like state) have not yet been identified in chronic asthma, myofibroblasts are increasingly recognized to have morphologic features intermediate between those of fibroblasts and smooth muscle cells (46). Recent studies have suggested, despite their origin being undefined, that myofibroblasts are related to phenotypically altered smooth muscle cells (47, 48). This possible phenotype-dependent interrelationship between airway smooth muscle cells and myofibroblasts may explain many of the similarities found in our study and that of Zhang and colleagues (44). This, together with the increased content of airway smooth muscle in asthma, adds to the possibility that this tissue may produce pathophysiologically relevant levels of cytokines and other mediators. We are currently evaluating the phenotypic dependence of cytokine production from human airway smooth muscle.

To confirm that the effect of IL-1 on human airway smooth muscle cells was mediated by a specific IL-1 receptor, the effect of the IL-1ra was investigated both on the production of GM-CSF and on the activity enhancing eosinophil survival. IL-1ra is a naturally occurring 22- to 25-kD protein that competes with either IL-1 or IL-1 for cellular receptor binding and has no intrinsic efficacy (49). Recombinant human IL-1ra binds specifically to both IL-I receptor types (I and II) on human cells. In the present study rhIL-1ra inhibited both the eosinophil survival-enhancing activity and the production of GM-CSF from IL-1-stimulated human airway smooth muscle cells. In each case, the potency of IL-1ra against IL-1 was similar. IL-1ra had no direct effect on GM-CSF-induced eosinophil survival. These data suggest that both the release of GM-CSF and production of the eosinophil survival-enhancing activity by IL-1 on human airway smooth muscle cells were mediated by a specific IL-1 receptor population, and further supports the view that in this system GM-CSF release from these cells is directly responsible for enhanced eosinophil survival.

It has been reported that in different cell types, including vascular smooth muscle cells, IL-1 can stimulate production of prostaglandins. In human airway smooth muscle in culture this is associated with activation of COX activity as well as induction of COX-2, which was prevented by dexamethasone (10, 11). In the present study the nonselective COX inhibitor indomethacin did not inhibit either the IL-1-stimulated production of GM-CSF from airway smooth muscle cells or the IL-1-stimulated eosinophil survival-enhancing activity, suggesting that these effects are independent of the COX pathway. Similarly, prostaglandin E₂, one of the predominant COX metabolites produced following stimulation of human airway smooth muscle with IL-1 (10, 11), did not prolong eosinophil survival (M. P. Hallsworth and S. J. Hirst, unpublished observations). In contrast production of GM-CSF following stimulation of human airway smooth muscle cells with IL-1 was completely inhibited by dexamethasone. Dexamethasone is known to inhibit the production of several proinflammatory cytokines, including the production of RANTES (12) and IL-8 (14) from human airway smooth muscle cells in culture, and the induction of other enzyme systems (e.g., phospholipase A₂ and inducible nitric oxide synthase) that generate inflammatory mediators. The inhibitory effect of dexamethasone on GM-CSF production by IL-1 in our study is similar to that obtained with prednisolone in human myofibroblasts (44) and is consistent with a preliminary report in human airway smooth muscle cells showing inhibition of GM-CSF production by a single concentration of dexamethasone (1 µM) when cells were stimulated with a mixture of cytokines (50). This and other recent studies (12, 14, 50) suggest that airway smooth muscle, which was previously considered to have only a structural role in asthma, may be another important therapeutic target for the anti-inflammatory effects of steroids.

In conclusion, this study clearly shows that human airway smooth muscle cells in culture, when stimulated with IL-1, produce a soluble activity that prolongs the survival of human peripheral blood eosinophils in vitro. The biologic activity responsible for this effect was identified exclusively as GM-CSF using the bioassay system. This approach was used so that the functional relevance of our findings could first be assessed. The data raise the possibility, for the first time, that there is a direct interaction between airway smooth muscle and the eosinophil in which the smooth muscle cell is the effector. Indirect evidence to support this possibility in vivo is provided by a recent study of surgically resected tissue from asthmatic and nonasthmatic subjects, in which eosinophils were found to be present within and around the airway smooth muscle bundles of the small airways of asthmatics (51). Furthermore, the identity of the biologic activity accords with previous in vitro studies showing GM-CSF production by human vascular smooth muscle cells (52) and, more recently, a preliminary report by Saunders and associates (50) on human airway smooth muscle cells. In contrast to this preliminary study (50), we were unable to detect constitutive production of GM-CSF by either ELISA or the eosinophil survival bioassay, and we observed only negligible GM-CSF production after exposure to TNF- (< 5% of the total GM-CSF production induced by IL-1). Furthermore, because of the preliminary nature of the study by Saunders and associates (50), the biologic activity of GM-CSF produced by human airway smooth muscle cells was not investigated.

Apoptosis, the physiologic form of cell death, is believed to play a key role in the resolution of inflammation (53), and increased eosinophil apoptosis in vivo is associated with the resolution of airway inflammation and a clinical improvement in asthma (20). Our own (54) and other studies (38) have shown that cytokines such as GM-CSF delay eosinophil apoptosis in vitro. We can speculate that airway smooth muscle may be a pathophysiologically important cellular source of eosinophil-chemotactic (e.g., RANTES, IL-8, and eotaxin) and eosinophil survival-enhancing (e.g., GM-CSF) cytokines, and as a consequence has the potential to exacerbate eosinophilic inflammation by both recruitment of eosinophils and prevention of eosinophil apoptosis. Clearly, additional studies are needed to determine the relative importance of airway smooth muscle in this process in the diseased lung in vivo, particularly because the content of airway smooth muscle as a fraction of the total cells in the airway wall is increased in asthma. Studies such as this may redefine the role of airway smooth muscle in the pathogenesis of chronic inflammation of the airways.