INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Signs of chronic inflammation can be discerned throughout the bronchial tree of asthmatic patients (1, 2), even in those with mild forms of the disease (3). Various morphologic changes can be observed in the bronchial tree that probably arise as a consequence of inflammatory processes. These alterations include thickening of the basement membrane, deposition of reticular collagen (4), and disruption of elastic fibers (7). An imbalance between the expression of proteases and antiproteases and particularly of matrix metalloproteinases (MMP) and their inhibitors (tissue inhibitors of metalloproteinase; TIMP) is believed to generate tissue destruction or abnormal tissue repair, such as fibrosis, in some pulmonary inflammatory diseases (8), and may also be important in asthma. MMP-9 is the most prominent MMP found in bronchoalveolar lavage fluid (BALF) from untreated asthmatic patients or those treated with glucocorticoids (GCs), and from healthy nonsmokers (12). Increased levels of MMP-9 have been found in bronchial biopsy specimens (13) and in BALF of untreated asthmatic patients as compared with control populations (12). However, the expression of other MMPs and of TIMPs had not previously been investigated.

TIMP-1 is the major member of a group of specific inhibitors that tightly control the activity of almost all MMPs except for membrane-type MMPs (MT-MMPs) (14). TIMP-1 inhibits gelatinase-type MMP, MMP-9 (92-kD gelatinase or 92-kD type IV collagenase), and MMP-2 (72-kD gelatinase or 72-kD type IV collagenase), as well as MMP-1 (interstitial collagenase). TIMP-1 also inhibits stromelysin-type MMPs, one of which, MMP-3, is involved in the activation of latent MMPs (15). In addition, TIMP-1 is secreted in association with MMP-9. TIMP-2 is secreted in association with MMP-2 and inhibits MMP-2 and MT-MMPs (14).

MMPs encompass a large family of zinc- and calcium-dependent endopeptidases with distinct substrate specificities so that they can degrade most components of the extracellular matrix (ECM). Therefore, the MMPs play an important role in physiologic and pathologic processes including ECM turnover, tissue degradation and repair, cell migration, and inflammation (14).

Inflammatory mediators, cytokines, and growth factors can modulate the production of both TIMPs and MMPs. TIMP-1 expression is regulated by interleukin (IL)-6 (17), whereas MMPs are modulated by IL-1 and tumor necrosis factor (TNF)- (20), and increased levels of these cytokines have been detected in BALF of asthmatic individuals after allergen challenge (21).

The primary aim of the present study was to determine whether TIMP-1 is present in BALF, and to compare TIMP-1 levels in stable, untreated asthmatic subjects with those of asthmatic subjects treated with glucocorticoids (GCs) and of healthy nonsmokers (controls). In order to determine whether there is an imbalance between MMPs and TIMPs in BALF fluid in asthma, the presence of TIMP-2 and of other MMPs was also investigated. These substances were characterized and quantified. Since a large increase of TIMP-1 was detected in untreated asthma patients, we also examined the relationship between TIMP-1 levels and the proinflammatory cytokines IL-6, TNF-, and IL-1.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

The clinical characteristics of the asthmatic subjects are detailed in Table 1. Thirty two stable asthmatic subjects were selected as previously described (2). All patients had reversible airways obstruction characterized by an increase of 12% in their FEV₁ value and an absolute FEV₁ value of 200 ml after inhalation of 200 µg of salbutamol. The clinical severity of chronic asthma was based on the step system of the Global Initiative for Asthma (GINA), which is used to grade chronic asthma from mild intermittent (Step 1) to severe persistent (Step 4) (22). Patients were considered as having stable asthma if the disease had been fully controlled for at least a month (GINA definition). The duration of asthma was established on the basis of the patient's history, followed by a careful clinical examination. The diagnosis of allergy was based on the clinical history and on skin-prick tests to common inhalant and food allergens found in the Montpellier area of France. None of the subjects was a current or previous smoker, and none had a history of allergic bronchopulmonary aspergillosis. Patients were excluded from the study if they had had a severe exacerbation of asthma requiring hospitalization during the month preceding the study.

Seventeen patients, allocated to the untreated asthma group, had not been treated with inhaled or oral corticosteroids for at least 3 mo before the beginning of the study, and none of the patients had used nedocromil, sodium cromoglycate, or theophylline during the previous 3 mo. However, these patients did receive short-acting inhaled ₂-agonists as required. Fifteen patients allocated to the GCs group were at the time being treated with inhaled and/or oral corticosteroids, and had been receiving this treatment for at least 3 mo. The corticosteroid treatment of these patients is presented in Table 1.

Nineteen healthy subjects, consisting of nine females and 10 males 25 to 60 yr of age, were used as a control group. None of these subjects had ever suffered from asthma or allergic diseases, and none presented any bronchial or respiratory tract infection during the month preceding the study. None of the controls was a current or previous smoker, and their pulmonary function was within the normal range.

Informed consent was obtained from all the participants prior to the study. The design of the study fulfilled all the criteria of the ethical committee of our hospital.

Recovery of BALF

Fiberoptic bronchoscopy was performed as previously described (2). Bronchoalveolar lavage (BAL) was done in one of subsegmental bronchi of the middle lobe by injecting three to five aliquots of 50 ml of sterile 0.9% NaCl solution at room temperature, which was reaspirated by gentle syringe suction. The different retrieved BALF fractions were pooled.

Processing of BALF

Cells were immediately separated from the fluid phase of the BALF by centrifugation at 400 × g for 10 min. An aliquot of cells resuspended in phosphate-buffered saline (PBS) was taken for cell counting, assessment of cell viability with the trypan blue dye exclusion test, and characterization of the different cell populations by May-Grünwald-Giemsa staining.

The fluid phase of the BALF was centrifuged at 5,000 × g for 20 min at 4° C to remove debris, and was stored as aliquots. For zymographic analysis and enzyme immunoassay (EIA) of MMP-1, MMP-3, and cytokines, Triton X-100 was added at a final concentration of 0.05% (wt/vol) to fresh aliquots, and the samples were concentrated 20 times through the use of Centricon filters with a 5,000 M.W. cutoff membrane (Amicon Inc., Beverly, MA) according to the manufacturer's instructions. Unconcentrated and concentrated aliquots were kept at 80° C for subsequent analyses.

Cell Culture

The human fibrosarcoma cell line HT1080 (CCl-121; American Type Culture Collection [ATCC], Rockville, MD) was cultured in Dulbecco's modified Eagle's medium, and the human monocytic cell line THP-1 (TIB-202; ATCC) was cultured in RPMI 1640 medium (Eurobio, Les Ullis, France) as described (12). THP-1 cells stimulated with phorbol myristyl acetate (PMA) (50 ng/ml) and HT1080 cells, both plated at a density of 10⁶ cells/ml, were cultured for 24 h in the absence of serum. Conditioned media were collected, centrifuged, and stored as described (12).

Measurements of TIMPs, MMPs, Albumin, and Cytokines in BALF

Measurements of TIMP-1, TIMP-2, MMP-2, MMP-3, and MMP-9 were made with one-step sandwich EIA methods. MMP-1 was measured with a two-step EIA (Biotrak; Amersham Life Science, Buckinghamshire, UK). The assays have 0.5, 3.0, 2.4, 0.4, 6.2, and 0.6 ng/ml sensitivity levels for TIMP-1, TIMP-2, MMP-1, MMP-2, MMP-3, and MMP-9, respectively, and measure the total amount of each protein whether it is present in a free or bound state. The MMP-1 assay recognizes total human MMP-1 (i.e., free MMP-1 and that complexed with TIMP-1). The MMP-3 assay recognizes pro-MMP-3, as well as active MMP-3 and TIMP-1 complexes. The MMP-9 and MMP-2 assays recognize human pro-MMP-9 and human pro-MMP-2 in both the free state and complexed with TIMP-1 or TIMP-2.

Levels of albumin were also measured with an EIA (23). The limit of detection of the assay was 10 ng/ml. The data on TIMP-1 and MMP concentration in BALF were normalized to the amount of albumin present, and were expressed as ng/mg albumin.

Measurements of IL-1, IL-6, and TNF- were made with a specific enzyme-linked immunosorbent assay (ELISA) (COMBO; BioSource Europe S.A., Fleurus, Belgium) in 20-fold concentrated BALF. The minimum detectable concentration was estimated to be 8 pg/ml for TNF-, 10 pg/ml for IL-1, and 10 pg/ml for IL-6.

Zymographic Analysis of MMPs

MMPs in BALF were analyzed with zymography, using 10% lithium dodecyl sulfate-polyacrylamide gels cast in 1-mm cassettes (Novex, San Diego, CA). Aliquots of concentrated BALF (as described for the processing of BALF) were analyzed using gels impregnated with collagen type I (0.5 mg/ml) from bovine skin (ICN Biomedicals Inc., OH) in an attempt to detect other MMPs that could have escaped detection with gelatin zymography (12), since this method has subpicogram sensitivity for MMP-1 (type I collagenase) (24). Latent and active MMPs can be distinguished according to their molecular weights, since upon activation the inhibitory propeptide region of the molecule is lost (15). The proportion of lysis zones (clear bands against a dark background) was estimated by densitometric scanning on the collagen gel as previously described (25).

Reverse Zymographic Analysis of TIMPs

The presence of TIMPs in BALF was investigated with reverse zymography. Concentrated samples were resolved on 15% lithium dodecyl sulfate-polyacrylamide gels copolymerized in the presence of both collagen, as described earlier, and 500 µl of conditioned medium from HT1080 cells as a source of collagenase. These cells secrete high levels of MMP-1 (300 ng/ml/24 h) (our unpublished data). The gels were processed as described by Oliver and coworkers (26). Areas of inhibition were detected as dark zones against a relatively clearer background. Native human TIMP-1 (Valbiotech, Paris, France) was used as a reference for the secretion of TIMP-1.

Western Blot Assay

Concentrated BALF was subjected to sodium dodecylsulfate-polyacrylamide gel electrophoresis (SDS-PAGE) on 4% to 12%-gradient gels (Novex) and blotted onto nitrocellulose. The blot was then probed with a mouse monoclonal antihuman MMP-1 antibody (2 µg/ml) (R&D Systems, Adbington, UK). Visualization was achieved with a peroxidase-conjugated antimouse IgG antibody (Sigma, St. Louis, MO) at a 1:3,000 dilution and an enhanced chemiluminescence system (NEN, Boston, MA). Conditioned medium from PMA-stimulated THP-1 cells was used as a reference for the secretion of latent and active MMP-1.

Immunoprecipitation

MMP-9 was identified by immunoprecipitation as previously described (12), using a mouse monoclonal antibody against human MMP-9 (2 µg/ml) (Oncogene Research, Cambridge, MA) that recognizes all forms of MMP-9 under nonreducing conditions.

Statistical Analysis

Statistical analyses were done with nonparametric tests. The Mann- Whitney U test was used for unpaired comparisons. Correlations were analyzed with Spearman's rank correlation test or with Kendall's t test.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Characteristics of Subjects

Demographic and functional characteristics of the asthmatic subjects are shown in Table 1. All patients were stable. Patients with untreated asthma (mean ± SD: 40 ± 14 yr) were significantly younger (p < 0.03, Mann-Whitney U test) than those treated with GCs (mean ± SD: 52 ± 16 yr). There were more males in the latter group than in the untreated one. There was no significant difference in the duration of disease in untreated asthmatic patients (mean ± SD: 20 ± 21 yr) and GC-treated patients (mean ± SD: 12 ± 8 yr).

The age of subjects in the control population (mean ± SD: 44 ± 15 yr) did not differ significantly from that of either asthmatic population.

Total and Differential Cell Counts in BALF

Total cell counts and percentages of different cell types are shown in Table 2. Total cell numbers recovered were similar in all groups. The number and percentage of eosinophils were significantly higher in untreated asthmatic subjects than in controls (p  0.0002, Mann-Whitney U test). No significant differences were observed for other cell types.

TIMP Levels in BALF

Analysis of BALF with reverse zymography revealed the presence of TIMP-1, visualized by an area of MMP inhibition at 28 kD (Figure 1A).

View larger version (83K): [in this window] [in a new window]   View larger version (16K): [in this window] [in a new window]   Figure 1.   Analysis of TIMPs in BALF. (A) Reverse zymography was performed as described in METHODS. Purified TIMP-1 (150 ng) (lane 1); 20-fold concentrated BALF from two asthmatic patients (lanes 2 and 3) (B and C ). Unconcentrated BALF were assayed for TIMP-1 with an EIA as described in METHODS. (B) TIMP-1 levels from 17 controls, 17 untreated subjects and 15 GC-treated subjects (C ). Data from untreated asthmatics and GCs were plotted as a function of the severity score as defined by GINA.

The concentration of TIMP-1 normalized to BALF albumin content was significantly higher in untreated asthmatic subjects than in the control and GC-treated groups (Figure 1B). It was approximately 13-fold higher (mean ± SD = 0.91 ± 0.73 ng/µg albumin) than in controls (mean ± SD = 0.07 ± 0.06 ng/µg albumin, p < 0.0001, Mann-Whitney U test) and  5-fold higher than in the GC-treated group (mean ± SD = 0.16 ± 0.11 ng/µg albumin, p < 0.0001, Mann-Whitney U test). TIMP-1 levels were also significantly higher in GC-treated than in control subjects (p < 0.005, Mann-Whitney U test). As shown in Figure 1C, TIMP-1 levels increased, although nonsignificantly (Kendall's t test), with the severity of asthma in untreated asthmatic subjects but not in GC-treated subjects. There was no correlation between TIMP-1 levels and FEV₁ values, atopy, or the duration of disease.

In all populations, TIMP-2 was below the assay detection limit in BALF either unconcentrated or concentrated 20-fold.

Characterization of MMPs in BALF

BALF from two subjects in each group, analyzed with zymography, displayed three to four lysis bands with varying intensities (Figure 2A). The two more intense bands, observed in the region of the 98-kD standard, reacted with an anti-MMP-9 antibody, and represent respectively the latent ( ) form of MMP-9 (Figure 2B) and MMP-9 complexed (c) with microglobulin (Figure 2B). The diffuse, faint band beneath latent MMP-9 was also recognized by the anti-MMP-9 antibody and represented an active (a) form of MMP-9 (Figure 2B). The proportion of active MMP-9, estimated by densitometric scanning on the collagen gel, represented < 10% of total MMP-9 (complexed + latent + active), and was not significantly different among the study groups. The thin lysis band (Figure 2A) observed near the 50-kD standard was identified as latent MMP-1, since it was recognized by an anti-MMP-1 antibody using Western blotting and immunodetection (Figure 2C, lanes 1 and 2). The proportion of the MMP-1 lysis zone relative to that of MMP-9 indicated that MMP-1 was at least 10- to 20-fold less abundant than latent MMP-9.

View larger version (52K): [in this window] [in a new window]   Figure 2.   Analysis of MMPs in BALF. BALF were concentrated and analyzed with zymography (A and B) or Western blotting (C ) as described in METHODS. (A) One percent of the original volume of BALF was analyzed from two control subjects, two untreated asthmatic subjects, and two GC-treated subjects. (B) Five hundred microliters of BALF from three control subjects was submitted to immunoprecipitation with an anti-MMP-9 antibody before zymographic analysis. (C  ) Immunoblotting was performed with 25 µl of 20-fold concentrated BALF from two control subjects (lanes 1 and 2) or with unconcentrated conditioned media harvested from THP-1 (lane 3), which served as a control for the presence of latent and active MMP-1; c = complexed,  = latent; a = active.

Thus, our data indicated the presence of MMP-1 and of active MMP-9 in the BALF of asthmatic subjects. They did not, however, allow us to rule out the presence of MMP-3 or MMP-2, which respectively have the same or a similar molecular weights as that of MMP-1 and active MMP-9.

MMP Levels in BALF

The presence and absolute values for concentrations of MMP-1, MMP-2, and MMP-3 were assessed with specific EIAs. None of the three was detectable in unconcentrated fluids, but they could be detected when BALF was concentrated 20-fold. Since TIMP-1 inhibits any MMP at the ratio of 1:1, the molar concentrations of MMP-1, MMP-2, and MMP-3 normalized to the BALF albumin content were compared with those of TIMP-1 and MMP-9. Figure 3 illustrates three important points. First, that mean levels of MMP-1, MMP-2, and MMP-3 were similar in all populations. Second, that the content of TIMP-1 was much higher (by 25- to 300-fold) than that of MMP-1, MMP-2, or MMP-3 in all of the study populations, and third, that molar concentrations of TIMP-1 were similar to those of all MMPs together with MMP-9 in controls, a little higher in GCs-treated subjects and far greater in untreated asthmatic subjects.

View larger version (31K): [in this window] [in a new window]   Figure 3.   Comparison of TIMP-1 and MMP molar content in BALF. BALF fluids were assayed for MMP-1, MMP-2, MMP-3, and MMP-9 with EIAs as described in METHODS. TIMP-1 values are taken from the data shown in Figure 1. Mean ± SEM.

Relationship between TIMP-1 Levels and BALF Cellularity or Proinflammatory Cytokine Levels in Untreated Asthmatic Subjects

Analysis of correlations between TIMP-1 levels and the number of cells in BALF demonstrated a significant linear relationship with the number of alveolar macrophages (AM) (Figure 4,  = 0.65, p < 0.005, Spearman rank test). In contrast, no correlation was observed with the number of lymphocytes, neutrophils, or eosinophils (Figure 4).

View larger version (28K): [in this window] [in a new window]   Figure 4.   Correlation between TIMP-1 and BALF cellularity in untreated asthmatic subjects. TIMP-1 values are taken from the data given in Figure 1. The numbers of different cell types were determined as indicated in METHODS. Both parameters are expressed per milliliter of BALF.

The proinflammatory cytokines TNF-, IL-1, and IL-6 in BALF were measured with ELISAs to investigate their potential influence on TIMP-1 levels. The levels of TIMP-1 and IL-6 were positively correlated (Figure 5,  = 0.84, p < 0.001, Spearman's rank correlation test). In contrast, there was no correlation between TIMP-1 levels and the amounts of IL-1 or TNF- (Figure 5).

View larger version (16K): [in this window] [in a new window]   Figure 5.   Correlation between TIMP-1 and proinflammatory cytokines in untreated asthmatic subjects. Measurement of cytokine content was done with ELISAs in 20-fold concentrated BALF, as described in METHODS, in cases in which sufficient lavage fluid remained. TIMP-1 values are taken from the data given in Figure 1.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In this study, we found that the levels of TIMP-1 in BALF were significantly higher in stable patients with untreated asthma than in GCs-treated subjects or in healthy nonsmoking controls. The levels were positively correlated with IL-6 concentrations and with the number of AM recovered in the BALF. Unlike TIMP-1, TIMP-2 was undetectable. The contents of MMP-1, MMP-2, and MMP-3 were much lower, and there was no difference in their contents between the different populations tested. In untreated asthmatic patients, the molar concentration of TIMP-1 exceeded the concentrations of all MMPs together.

TIMP-1 levels in untreated asthmatic subjects were not correlated with FEV₁, atopy, the duration of disease, or the age of the patients, but tended to increase with the severity of asthma. However, the lack of significant correlation may have been due to the selection of patients. First, the number of patients with asthma of Step 1 severity was very low because such patients do not generally require any treatment from an asthma specialist. In addition, we did not include Step 4 patients in our study, since it is difficult to enroll patients with severe persistent asthma who are not receiving any antiinflammatory treatment to control their disease. Our findings are therefore restricted to subjects with untreated, mild persistent to moderate persistent asthma who were receiving only ₂-agonists as required, and may not apply to patients with a more severe form of the disease or those with asthma exacerbations.

In GC-treated asthmatic subjects, TIMP-1 levels were significantly lower than in untreated patients. Thus, GCs seem to reduce TIMP-1 levels. In support of this hypothesis, GCs such as dexamethasone are known to be very potent inhibitors of MMP and TIMP transcription in vitro in AM (27) and in fibroblasts (28). Nevertheless, a direct in vivo effect has yet to be demonstrated. Alternatively, GCs could act indirectly by inhibiting inflammatory cytokines such as IL-6, which mediate TIMP-1 production (17).

TIMP-1 levels in untreated asthmatic subjects were significantly correlated with the number of AM, but not with that of eosinophils, lymphocytes, or neutrophils in BALF. This observation suggests that TIMP-1 originates from these cells. This argument is strengthened by the findings that TIMP-1 was expressed in AM isolated from BALF of the same groups as in the present study (25), and to a significantly greater degree at both the protein and mRNA levels in untreated asthmatic patients, than in GC-treated patients or controls.

A few possible explanations for the TIMP-1 increase in untreated asthma can be put forward. First, in untreated asthmatic subjects TIMP-1 levels were significantly correlated with IL-6 but not with IL-1 or TNF- levels, suggesting that TIMP-1 synthesis and release may be increased via IL-6. This possibility is compatible with the presence of an IL-6-responsive element in the TIMP-1 promoter (29), and with the stimulation of TIMP-1 synthesis by IL-6 in AM (19). IL-6 is mainly produced by monocytes/macrophages, and in asthma, AM have been shown to release increased levels of IL-6 both spontaneously (30) and after in vivo allergen challenge (21). More importantly, we found that this cytokine was at least partly responsible for the TIMP-1 increase in AM of patients with untreated asthma, via an autocrine mechanism (25). Other mediators may be implicated, such as transforming growth factor- (TGF-), which is increased in asthma (31) and modulates the expression of TIMP-1 in human fibroblasts (17). Second, the airway inflammation of asthma is characterized by an accumulation of inflammatory cells such as eosinophils, AM, lymphocytes and neutrophils. MMP 9 has been shown to be involved in the process of cell migration across basement membranes (16, 32, 33), and increased levels of MMP-9 have been found in the airways of asthma patients (12, 13). These observations suggest that the increase in TIMP-1 production may accompany or follow that of MMP-9, to reduce cell transmigration.

In our study, the molar concentration of TIMP-1 in untreated asthma greatly exceeded that of MMP, whereas similar levels of TIMP-1 and MMP-9 were found in control subjects. Because TIMP-1 inhibits most MMP, the presence of other MMPs in BALF was investigated. Our earlier results showed that all lytic activities released in BALF as detected with gelatin zymography were MMPs. Only latent MMP-9 was identified, and was the major MMP found in BALF in all groups of patients investigated (12). In the present study, we used collagen zymography, which is more sensitive than gelatin zymography, and this last observation was confirmed. Latent MMP-1 was identified immunologically, using western blotting. Active MMP-9 was detected at a much lower level than latent MMP-9, and was also identified immunologically. MMP-2 and MMP-3 were present as assessed through EIA, but their concentrations were too low to be detected with zymography. The concentrations of MMP-1, MMP-2, MMP-3, and MMP-9 combined were deemed to be too low to compensate for the excess of TIMP-1 found in the untreated asthmatic group. Moreover, the excess of TIMP-1 may explain why MMP-9 and MMP-1 were mainly in the latent form as shown on zymograms. Even if MMP-2 or MMP-3 was activated, there would be sufficient TIMP-1 to inhibit its catalytic activity. Together, these findings suggest an imbalance between enzyme and inhibitor in untreated asthma.

An imbalance between MMP and TIMP production could lead to tissue damage in inflammatory diseases of the lung, and is likely to be involved in the parenchymal destruction and repair processes in pulmonary diseases. The levels of MMP and TIMP in BALF, serum, or sputum have been investigated in a number of diseases including adult respiratory distress syndrome (10), bronchiectasis (11), cystic fibrosis (8) and idiopathic pulmonary fibrosis (9). An excess of protease over inhibitor was the usual finding, but an excess of TIMP was demonstrated in fibrotic diseases. Subepithelial fibrosis is a common feature of asthmatic airways (6). The excess of TIMP-1 production may be involved in the thickening of the basement membrane reticular layer, together with the deposition of matrix components, that is found in asthma. Because matrix metabolism is a balance between synthesis and lysis, an increase in inhibitor production would prevent proteolytic degradation and repair, and would hence result in the accumulation of collagen I, III, and V and of fibronectin (6), all of which are substrates of MMP (15). The origin of the deposited material has been attributed to myofibroblasts, and indeed, an increase in the number of myofibroblasts has been demonstrated (34). TIMP-1 may also act through its cell-growth-promoting activity by increasing myofibroblast proliferation (35), and TGF- production (17), and this in turn may amplify TIMP-1 production.

In conclusion, the new data provided by our study may have important clinical implications, since an excess of TIMP-1 could lead to myofibroblasts proliferation, accumulation of matrix components, and fibrosis, all of which are hallmarks of airway remodelling observed in patients with chronic asthma (6, 36). Early treatment of asthma with GCs may be useful to prevent or reduce thickening of the subepithelial basement layer (4, 5).