INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Airway remodeling is a key feature of persistent asthma. Part of the remodeling process involves the laying down of extracellular matrix (ECM) proteins within the airways (1). The ECM is an intricate network of macromolecules that have the potential to influence a variety of cellular functions including migration, proliferation, and differentiation (2, 3). The ECM plays a crucial role in the maintenance of airway function and structure. It can also influence the distribution and adhesion of inflammatory cells, fluid balance, and elasticity, and can act as a reservoir of inflammatory mediators (4). The profile of ECM proteins in the airways of asthmatic subjects differs from that in nonasthmatic subjects. Reports to date have indicated alterations in the quantities of a variety of ECM proteins, including an increase in collagen 1, III, and V; fibronectin; tenascin; hyaluronan; versican; and laminin 2/2 (5) in asthmatic subjects; conversely, a decrease has been observed in collagen IV and elastin (9). Changes in ECM deposition within the airways of asthmatic individuals may lead to or contribute to the development of bronchial hyperresponsiveness (BHR). The causes of the change in matrix deposition are unknown, but plasma leakage from the microvasculature as part of the inflammatory process may play a role.

Because airway smooth-muscle (ASM) cells from asthmatic patients are difficult to acquire, many research groups have adopted the technique of passive sensitization to model hyperresponsiveness in vitro. The use of serum from allergic asthmatic individuals has permitted the demonstration of changes in the function of nonasthmatic ASM in terms of contractile responses, cytokine production, and proliferation that may reflect differences seen in asthmatic smooth muscle (10). Others have used bronchoalveolar lavage fluid from asthmatic patients to induce changes in nonasthmatic smooth-muscle cells (11). Because it has recently been reported that ASM cells also produce matrix proteins (12), which can themselves have profound effects on cell function, we wanted to assess the effects of exposure to asthmatic serum on this property of ASM.

Corticosteroids have been routinely used in the treatment of asthma to reduce BHR (13). The mechanism of action of corticosteroids is predominantly through the blocking of inflammatory-cell activation and migration into the airways (14). In vivo studies suggest that corticosteroid treatment has a limited effect in reducing the increase in components of the ECM (15). In the present study, we extended the use of our in vitro model of hyperresponsiveness (passive sensitization) (10) to examine the relationship between allergic inflammation and changes in ECM production by human ASM in culture. We investigated the effect of exposure of human ASM cells to serum from an atopic asthmatic or nonatopic, nonasthmatic individual on the production of human fibronectin; perlecan; elastin; laminin ₁, ₁, ₂, and ₁ chain; thrombospondin; chondroitin sulfate; collagen types I, III, IV, and V; versican; and decorin. We also investigated the effect of the corticosteroid beclomethasone (0.01 to 100 nM) on the production of ECM proteins by human ASM cells exposed to serum from atopic asthmatic and nonatopic, nonasthmatic individuals.

In this study, we compared the effect of serum from an asthmatic atopic patient and a nonatopic, nonasthmatic patient on matrix protein production by human ASM cells in culture. We also attempted to elucidate the mechanisms underlying the changes we observed.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Human ASM cells were obtained according to methods previously described (16). The protocol for accomplishing this was approved by the Human Ethical Review Committee of The University of Sydney. Briefly, human bronchial smooth muscle was obtained from the large bronchial airways (3- to 6-mm I.D.) of 10 patients (age: 44 ± 19 [mean ± SD] yr) undergoing resection for either lung transplantation or carcinoma. Smooth-muscle bundles were dissected free of the surrounding tissue and seeded into 25-cm² vented tissue culture flasks in 10% fetal bovine serum (FBS) in Dulbecco's modified Eagle's medium (DMEM) supplemented with 20 U/L penicillin, 20 µg/ml streptomycin, and 2.5 µg/ml amphotericin B, and were placed in a 5% CO₂ humidified incubator at 37° C (Sanyo, Sydney, Australia). Cultures were 99% pure, and cells grew to confluence over a period of 14 to 21 d and were passaged when confluent. Cell passaging was done with a solution of 0.5% trypsin in 1 mM disodium ethylenediamine tetraacetate (Na₂EDTA). Viable cell number was assessed with a hemocytometer and trypan blue when cells were to be used for an experiment; however, when the cells were to be passaged, a 1:3 split was performed. Smooth-muscle cell characteristics were routinely determined by immunofluorescence and light microscopy. Cells were stained with antibodies against smooth-muscle -actin (17), with the same process done without the primary antibody being used as a control. By light microscopy, the cells appeared spindle shaped, having central oval nuclei with prominent nucleoli, and displayed the typical "hill and valley" appearance of ASM cells.

ASM cells (Passages 4 to 8) from 10 patients were seeded at a density of 1 × 10⁴ cells/cm². Tissue culture plates were precoated with bovine fibronectin by incubating them with 1.56 µg fibronectin/cm² in phosphate buffered saline (PBS) for 2 to 3 h at 37° C. Cells were seeded into 48 wells of 96-well plates in DMEM supplemented with 10% atopic asthmatic serum or 10% nonatopic, nonasthmatic serum with or without beclomethasone (0.01 to 100 nM). Control wells (the remaining 48 wells of each plate), incubated in the absence of cells, were similarly processed and used for background absorbance measurements in enzyme-linked immunosorbent assays (ELISAs). Cells were grown for a total of 7 d, with replenishment of all culture components at Day 4. In separate six-well plates, cells were seeded in 10% atopic asthmatic serum or nonatopic, nonasthmatic serum in DMEM ± beclomethasone (0.01 to 100 nM). In a separate series of experiments, cells were treated with the leukotriene antagonist montelukast (0.01 to 100 nM) for a total of 7 d, with replenishment of all components at Day 4. After 7 d, ECM free of cells was prepared by treatment with sterile hypotonic ammonium hydroxide as described by Gospodarowicz and Lui (18). ECM proteins were washed with PBS and stored under sealed conditions at 70° C under minimal cover of PBS. ELISAs were performed on the 96-well plates, which had been thawed overnight at 4° C, using antibodies to the following matrix proteins: human fibronectin; perlecan; elastin; laminin ₁, ₁, ₂, and ₁ chains; thrombospondin; chondroitin sulfate; collagen types I, III, IV, and V; versican; and decorin. The presence of various ECM molecules was detected through ELISA, essentially as described by Underwood and colleagues (19), with the following modifications: (1) The concentration of primary antibodies was selected by prior titration to detect changing concentrations of ECM components with time; the antibody concentrations used varied from 5 to 10 µg/ml for purified antibody and from 1:250 to 1:10,000 dilutions for ascites fluids or serum. (2) Background absorbance for each primary antibody was obtained from bovine fibronectin-coated wells in the absence of cells, as described earlier. We found that this gave more reliable estimates of background signal than did the use of a subclass-matched, nonspecific antibody, although these were included for comparison. (3) The secondary antibodies were biotinylated antimouse or antirabbit IgG antibodies at a dilution of 1:1,000 + 10% FBS, incubated with the protein- primary antibody preparation for 1 h, after which peroxidase-conjugated streptavidin at 1:500 was added with incubation for 30 min. Reagents were diluted in blocking buffer, and intervening washes were done with PBS. 2,2'-Azino-bis-(3-ethylbenzo thiazoline)-6-sulfonic acid (ABTS), at 1.1 mg/ml in 0.05 M citrate, pH 4.5, was the substrate. Absorbance was read at 405 nm (490 nm reference) on an ELISA plate reader (Model 3550; Bio-Rad, Inc., Hercules, CA). Cell numbers from the six-well plates were assessed by manual cell counting. Nonspecific mouse immunoglobulins or rabbit serum were used as negative controls.

Affinity Purification

IgE was removed from allergic serum by batch mode affinity purification.

Agarose gel preparation. Antihuman IgE was coupled to CNBr- activated agarose according to the manufacturer's instructions. Ascites fluid (200 µl) containing > 400 µg anti-IgE antibody was buffered to pH 8.3 by mixing with 200 µl of 2× bicarbonate buffer (0.2 M NaHCO₃, 0.7 M NaCl), and the mixture was then added to 250 µl of prepared activated agarose. The gel was then mixed overnight at 4° C on an orbital mixer and washed five times with 250 µl of bicarbonate buffer (0.1 M NaHCO₃, 0.5 M NaCl), and the remaining active sites were blocked with 1 M ethanolamine, pH 8.0, for 2 h. The gel was then washed with 1 ml of 0.1 M acetate buffer, pH 4.0, with 0.5 M NaCl, followed by 1 ml of 0.1 M Tris-HCl, pH 8, with 0.5 M NaCl, in three alternating cycles.

IgE binding. Allergic serum with a high IgE titer was filtered through a 0.20-µm cellulose acetate filter. Serum (0.8 ml) was then buffered to pH 8.0 by mixing with 0.2 ml of modified Krebs-Henseleit solution (composition in mM: NaCl 111.6, KCl 4.7, CaCl₂ 2.5, MgSO₄ 1.2, KH₂PO₄ 1.2, NaHCO₃ 25.0, glucose 11.1) and 8.5 µl of 0.1 M NaOH. The affinity gel was rinsed with Krebs-Henseleit solution, pH 8.0, after which buffered serum was added and the mixture was left overnight at 4° C on an orbital mixer for IgE binding. Serum was removed from the agarose gel and the serum pH was immediately adjusted to 7.4 with 8.5 µl of 0.1 M HCl per ml of serum. The gel was washed 10 times with Krebs-Henseleit solution, pH 8, and the washes were kept for protein analysis.

IgE elution. IgE was eluted from the agarose gel with three washes of 500 µl of 0.05 M glycine, pH 2.8. Eluants (500 µl) were immediately adjusted to pH 7.4 by adding 50 µl of 10× Krebs-Henseleit solution and 25 µl of 0.1 M NaOH.

Assays. Serum, extracted serum, and eluted IgE fractions were analyzed for IgE titer, protein content (with the Bradford assay), and protein composition (by polyacrylamide gel electrophoresis).

Organ Bath Studies

Human lung tissue was obtained with approval from the Human Ethics Committee of the University of Sydney and the Central Sydney Area Health Service. Bronchial rings, measuring from 3 mm to 5 mm I.D. and 4 mm in length, were dissected free of surrounding tissue from one patient, and were tested by standard isometric organ bath techniques and found to be not sensitized to four standard allergens. Pairs of bronchial rings were held overnight at room temperature in atopic serum, nonatopic serum, atopic serum that had been incubated with the affinity gel, IgE eluted from the affinity gel, and eluted IgE mixed in a 1:1 ratio with atopic serum incubated with the affinity gel (reconstituted serum). Bronchial rings were then suspended in 5 ml organ baths containing Krebs-Henseleit solution that was bubbled with carbogen and maintained at 37° C. A load of 1 g to 2 g was placed on the bronchial ring tissues as determined by tissue size, and tissues were washed at 15-min intervals until a stable tone was established. Changes in tension were measured isometrically, using Grass FTO3 force transducers (Grass Instruments, Quincy, MA), and were recorded on a Maclab analogue-to-digital recorder (AD Instruments, Sydney, Australia). For each tissue, response to a maximal concentration of acetylcholine (1 mM) was elicited, and the tissue was washed repeatedly until baseline tension was reestablished. Tissues were finally washed and tested for response to 10 µl of a solution of Dermatophagoides pteronyssinus.

Materials

The following compounds were obtained from the sources given in parentheses: DMEM, Hanks balanced salt solution, Dulbecco's PBS, penicillin, streptomycin, amphotericin B, trypan blue, monoclonal antibody (mAb) clone III to laminin ₁ chain, rabbit antiserum to decorin, and specific antihuman fibronectin mAb clone 3E3 (GIBCO BRL; Life Technologies, Sydney, Australia); calcium chloride, magnesium sulfate, Na₂EDTA, HCl, methanol and glycerol (Ajax, Sydney, Australia); FBS (CSL, Australia); mAb to smooth-muscle -actin (mouse IgG₂ isotype), fluorescein isothiocyanate-conjugated goat antimouse lgG, trypsin MAb BA4 to elastin, mAb COL 94 to type IV collagen, mAb CS56 to chondroitin sulfate, and ABTS (Sigma, Australia); neutralizing chicken IgY to transforming growth factor-₁ (R&D Systems, Minneapolis, MN); mAb A76 to perlecan, mAb A65M to thrombospondin, and mAb A21 to laminin ₁ were prepared by Dr. P. Anne Underwood (20), and mAb D7 to laminin ₂ chain was a gift from Dr. J. Sanes of the Washington University School of Medicine, St Louis, MO. mAb 5D5 to versican was a gift from Dr. F. Rahemtulla of the University of Alabama, Birmingham, AL. MAbs 5D8/G9, 2G8/B1, and 1E2/E4 to collagen types I, III, and V, respectively, were gifts from Dr. J. Werkmerster of CSIRO Molecular Science, Melbourne Laboratory, Victoria, Australia. The leukotriene antagonist montelukast was a kind gift from Merck Sharp & Dohme Pharmaceutical (Sydney, Australia). Antigen extracts of D. pteronyssinus standardized mite DP 30,000 BAU/ml (Miles Laboratories, Inc., Elkhart, IN); antihuman IgE (mouse mAb clone HP6029; Calbiochem, Sydney, Australia); sepharose 4B (Pharmacia, Sydney, Australia).

Analysis of Results

Results of individual treatments for both the ELISA assay and the manual cell counts were obtained in triplicate. The ELISA results for each ECM protein were expressed as absorbance units at 405 nm per 1 × 10⁵ cells. A mean value was obtained for each treatment of ASM cells from each patient. The percent change from the fibronectin response in nonatopic serum for each individual result was then calculated, and all results with both nonatopic serum and atopic serum treatments were expressed relative to this value. An overall mean ± SEM was then calculated for all patients for each ECM protein examined. Analysis of variance with repeated measures and Fisher's product of least significant differences test was performed on the raw data (ELISA results and cell number). A value of p < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The human ASM cells cultured in the study produced a wide array of different matrix proteins, consisting of human fibronectin; perlecan; elastin; laminin ₁, ₁, ₂, and ₁ chains; thrombospondin; chondroitin sulfate; collagen types I, III, IV, and V; versican; and decorin (Figure 1). Atopic asthmatic serum induced a significant increase (p < 0.05, n = 7 to 10) in the production of fibronectin, laminin ₁ chain, perlecan, and chondroitin sulfate (Figure 1). In initial experiments, beclomethasone at 1 nM caused a significant reduction (p < 0.05, n = 7) in cell number both in cells exposed to atopic asthmatic serum (to 78.9 ± 9.5% of the control response) and in cells exposed to nonatopic, nonasthmatic serum (to 77.1 ± 2.4% of the control response) (Figure 2). However, beclomethasone at 1 nM did not reduce production of any of the ECM proteins examined, nor did it reverse the increase in fibronectin, laminin, perlecan, and chondroitin sulfate observed after exposure to atopic asthmatic serum (Figure 3). There was a general increase in ECM protein after exposure to beclomethasone 1 nM, but the individual ECM protein results were not significantly different from those without beclomethasone (Figure 3). Over a range of concentrations (0.01 nM to 100 nM), beclomethasone again reduced cell number (Figure 4). When this range of beclomethasone concentrations was examined against production of fibronectin, perlecan, and chondroitin sulfate, a significant increase (p < 0.05, n = 6) in production of all three ECM proteins was seen (Figures 5-7). Montelukast (0.01 nM to 100 nM) had no significant effect on the production of any of the ECM proteins examined. Neither atopic nor nonatopic serum affected cell viability in any of the 10 primary cultures studied.

View larger version (38K): [in this window] [in a new window]   Figure 1.   The relative amounts of ECM protein produced by human ASM cells in the presence of nonasthmatic, nonatopic serum (open bars) and atopic asthmatic serum (closed bars). Results are standardized for cell number and expressed as a percentage of response to fibronectin in the presence of nonatopic, nonasthmatic serum (mean ± SE). *Significant difference (p < 0.05, n = 10) between cells exposed to nonasthmatic, nonatopic serum and atopic asthmatic serum.

View larger version (14K): [in this window] [in a new window]   Figure 2.   The effect of 1 nM beclomethasone (beclo) on human ASM cell number after incubation in nonatopic, nonasthmatic serum or atopic asthmatic serum. Results are expressed as a percentage of the response in the absence of beclomethasone (control). *Significant difference (p < 0.05, n = 6).

View larger version (25K): [in this window] [in a new window]   Figure 3.   ECM production by human ASM cells after incubation in atopic asthmatic serum in the absence (open bars) and presence (closed bars) of beclomethasone (1 nM). Results are standardized for cell number and expressed as a percentage of response to fibronectin in the presence of nonatopic, nonasthmatic serum.

View larger version (16K): [in this window] [in a new window]   Figure 4.   Effect of beclomethasone (0.01 to 100 nM) on human ASM cell number after incubation in nonatopic, nonasthmatic serum (open bars) or atopic asthmatic serum (closed bars). Results are expressed as a percentage of the response in the absence of beclomethasone (control). *Significant difference from control sample in same serum (p < 0.05, n = 6).

View larger version (22K): [in this window] [in a new window]   Figure 5.   Fibronectin production by human ASM cells after incubation in asthmatic or nonasthmatic serum in the presence and absence of beclomethasone (0.01 to 100 nM). Results are standardized for cell number and expressed as a percentage of response to fibronectin in the absence of beclomethasone (control). *Significant difference from control sample in same serum (p < 0.05, n = 6).

View larger version (21K): [in this window] [in a new window]   Figure 6.   Perlecan production by human airway smooth muscle cells after incubation in asthmatic or nonasthmatic serum in the presence and absence of beclomethasone (0.01 nM to 100 nM). Results are standardized for cell number and expressed as a percentage of response to perlecan in the absence of beclomethasone (control). *Significant difference from control sample in same serum (p < 0.05, n = 6).

View larger version (19K): [in this window] [in a new window]   Figure 7.   Chondroitin sulfate production by human ASM cells after incubation in nonatopic, nonasthmatic (open bars) or atopic asthmatic (closed bars) serum in the presence and absence of beclomethasone (0.01 nM to 100 nM). Results are standardized for cell number and expressed as a percentage of response to chondroitin sulfate in the absence of beclomethasone (control). *Significant difference from control sample in same serum (p < 0.05, n = 3 to 6).

IgE Experiments

The allergic serum had a total IgE titer of 23,300 U/ml and a radioallergosorbent test (RAST) value of 4+ to D. pteronyssinus. The IgE titer of the extracted serum was 110 ± 64 U/ml after a single extraction and 64 ± 9.5 U/ml (mean ± SD, n = 2) after a second extraction. All serum samples were as effective as unfractionated serum in sensitizing bronchial rings to D. pteronyssinus. The contractile responses to 10 µl of D. pteronyssinus extract (expressed as a percentage of acetylcholine max) in tissues exposed to unfractionated serum, IgE-extracted serum, and doubly extracted serum were 102 ± 36%, 89 ± 25%, and 96 ± 27%, respectively (mean ± SD; n = 2).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The study showed that human ASM cells in culture have the capacity to produce a wide variety of ECM proteins, and that the profile of these proteins is altered by exposure to atopic asthmatic serum. In addition, beclomethasone at concentrations that induced a reduction in cell number did not decrease ECM protein concentrations from those without beclomethasone, but significantly increased ECM protein production. Furthermore, although there is evidence for an increase in the level of leukotrienes in the serum of asthmatic patients (21), the leukotriene antagonist montelukast did not inhibit the enhanced ECM protein production resulting from exposure to asthmatic serum.

The turnover of the ECM in the airways has been estimated at 15% daily (22). This turnover is regulated by de novo synthesis of ECM proteins, degradation of the ECM by the specific enzymes known as matrix metalloproteinases (MMPs), and inhibition of these MMPs by the specific inhibitors known as tissue inhibitors of matrix metalloproteinases (TIMPs). An increase in the overall content of any individual protein may result from its increased de novo synthesis, decreased production of MMPs, or increased TIMP production, or from combinations of these factors. Studies to date have shown an increase in the amount of ECM proteins in the airways of asthmatic individuals. These proteins include collagen 1, III, and V; fibronectin; tenascin; hyaluronan; versican; and laminin ₂/₂ (5), with a decrease in collagen type IV and elastin (9). The cells responsible for producing the ECM in the airways are unknown, but inflammatory cells, epithelial cells, and cells from the fibroblast lineage have been implicated (6, 23, 24). The present study is the first to show that the ASM has the potential to produce such a wide variety of ECM proteins, and the first to show an alteration in the production of fibronectin, perlecan, laminin, and chondroitin sulfate upon exposure of ASM cells, to human atopic asthmatic serum.

Others have examined the effect of serum on the production of individual ECM proteins by canine ASM cells in culture. In these experiments, Sheils and colleagues (25) found that exposure of ASM cells to homologous serum induced an increase in the production of fibronectin over that from cells grown in FBS.

The factors in serum that were responsible for the changes seen in ECM production in the present study are unknown. To date, a number of changes induced by atopic serum in isolated airway preparations have been observed. These include an increased contractile response to histamine, K⁺, tachykinins, and cholinergic agonists, and decreased relaxation responses to calcium-channel antagonists, potassium-channel-opening agents, vasoactive intestinal polypeptide, and -adrenoceptor agonists (10). Exposure of ASM to atopic asthmatic serum produces marked increases in the expression of interleukin (IL)-1 (26) as well as in messenger RNA (mRNA) for the low-affinity IgE receptor Fc RII (27). We have previously shown that when human ASM cells in culture are exposed to atopic serum and challenged with allergen, they exhibit increased expression of protein encoded by the protooncogene c-jun and an increase in cell growth as measured by incorporation of tritiated thymidine (10). Hakonarson and coworkers 1999 (28) recently reported an increase in mRNA expression for both (T-cell helper type 1) Th₁ and Th₂ cytokines, as well as in the proteins for the respective receptors, in human ASM cells after exposure to atopic asthmatic serum. Whether the mechanism underlying these changes in muscle growth and contractility also underlies the changes induced by allergic serum in the present study is unknown.

The physiologic sequelae of changes in ECM within the airways can only be the subject of speculation; however, individual ECM proteins can promote cell migration, differentiation, and survival. A shift in the profile of ECM proteins may have profound effects on the physical structure of the airways of asthmatic subjects, making them less distensible and more likely to retain the fluid resulting from vascular leakage. A characteristic feature of asthma is an increase in the amount of ASM in the airways (29). Much of the research so far done on the mechanisms responsible for this increase in smooth muscle has focused on mediators that can induce growth; however, a contributor to this increased growth that needs to be considered is an altered ECM bed surrounding the muscle. This concept is supported by studies in which proliferation of ASM cells was increased when they were grown on plastic tissue culture plates coated with specific ECM proteins (30).

The present study demonstrated increased production of a number of ECM proteins by human ASM cells in a model of an allergic environment. Studies of tissue, bronchoalveolar lavage fluid, and sputum samples from asthmatic subjects highlight increases in a number of ECM proteins, some of which are similar to those seen in our study. However, few of these studies have examined the changes in ECM proteins specifically in the vicinity of the ASM, with one report of increased hyaluronan and versican in this location (5). We observed changes in chondroitin sulfate a glycosaminoglycan related to hyaluronan. Like hyaluronan, chondroitin sulfate is a highly charged macromolecule with strong water-binding capacity. Any increase in such glycosaminoglycans could have marked effects on airway edema and contribute to altered airway function.

We attempted to determine whether IgE was the biologically active component of the allergic serum that was producing the changes in the profile of matrix proteins observed in our study. To do this, we stripped the serum of IgE through the use of affinity purification with an mAb to human IgE. However, even after repeated application of the eluted serum to the affinity column, the IgE content was reduced only to approximately 60 U/ml. This concentration of IgE still produced a 3+ RAST response to D. pteronyssinus, and when this serum fraction was incubated with nonsensitized bronchial ring tissue segments, subsequent application of antigen resulted in a contractile response. Many groups have observed changes in properties of ASM after passive sensitization with allergic serum, and the role of IgE in these changes is controversial. Whereas Hakonarson and associates (27) have reported convincing evidence for a role of IgE in such changes, others have produced just as strong evidence to the contrary (31). Thus, it was not possible in the present study to draw conclusions about the role of IgE in the changes observed in matrix protein production in response to asthmatic serum.

Corticosteroids, one of the most effective groups of drugs used in treating asthma, are known to reduce BHR (13). The mechanism of action of the corticosteroids is predominantly through blocking the recruitment of inflammatory cells into the airways and inhibiting of the release of proinflammatory cytokines from inflammatory cells (14). Corticosteroids have a variety of actions on ASM cells, including reducing their contractility (32), increasing their relaxation (33), inhibiting their proliferation (34), and decreasing their secretion of a variety of mediators, including prostaglandin E₂ (35), RANTES (36), and interleukin-8 (36). In vivo studies of the effect of corticosteroid treatment on ECM components within the airways suggest that it has limited effectiveness in reducing the increase in these components. Jeffrey and colleagues (37) showed that although 4 wk of treatment with budesonide reduced inflammation within the airways, prolonged treatment for up to 3.7 yr failed to reduce the increased thickening of the reticular basement membrane, and Laitinen and coworkers (15) found no change in collagen deposition in budesonide-treated, birch pollen-sensitive asthmatic individuals. The mechanism by which corticosteroids caused an increase in the production of ECM in the present study is unknown, but factors contributing to this effect could be increased production of ECM protein, increased production of TIMPs, or the suppression of MMP production. Inhibition of MMPs has been observed in fibroblasts, in which a decrease in the messages and proteins for MMP 2 and MMP 9 has been reported in response to corticosteroids (38). In the same study in which these effects were reported, epithelial cell MMP production was observed to be unaffected, suggesting that corticosteroids have cell-specific effects in their ability to inhibit MMP production.

The present study has shown increased production by human ASM cells of a variety of ECM components after exposure to atopic asthmatic serum. This suggests that after an acute allergic or inflammatory event, the accompanying vascular leakage (that is a hallmark of asthma) may contribute to the airway remodeling observed in asthma. Our results also suggest that although corticosteroids may address the issue of inflammation in asthmatic airways, they may not prevent the fibrosis induced by resident cells.