Copyright ©ERS Journals Ltd 2001
Fluticasone and salmeterol downregulate in vitro, fibroblast proliferation and ICAM-1 or H-CAM expression1 Pulmonary Division, G. Gaslini Institute, Genoa, and 2 GlaxoWellcome S.p.A., Verona, Italy CORRESPONDENCE: G.A. Rossi, Dept. of Pulmonary Diseases, G. Gaslini Institute, Largo G. Gaslini 5, 16147, Genoa, Italy. Fax: 39 0103776590 Keywords: airway remodelling, ß2-agonists, bronchial asthma, corticosteroids
Received: July 27, 2000
ß2-adrenoreceptor agonists have pharmacological properties that may suggest an inhibitory effect on various aspects of the inflammatory and repair processes that characterize asthma. Since fibroblasts express ß2-adrenoreceptors, the effects of different concentrations (0.1100 nM) of fluticasone propionate (FP), salmeterol (S) and their combination (FP+S) on lung fibroblast proliferation and adhesion molecule expression were evaluated.
Stimulation of human foetal lung fibroblasts with a fibrogenic cytokine, basic fibroblast growth factor (bFGF), resulted in a [methyl-3H] thymidine ([3H]TdR) uptake, four-fold higher than that of control cultures (p=0.0001) and was significantly inhibited by S, at all the concentrations tested (0.1100 nM; p<0.05). No changes in bFGF-induced cell proliferation were observed in the presence of FP (0.1100 nM; p>0.05, all comparisons). In addition, the association FP+S did not improve the inhibitory activity of S alone (p>0.05, each comparison). An upregulation of intercellular adhesion molecule-1 (ICAM-1) expression was induced by tumour necrosis factor- These results show that in human foetal lung fibroblasts, fluticasone propionate and salmeterol are effective in modulating in vitro, different lung fibroblast biological functions that are likely to be involved in airway remodelling. Bronchial asthma is a chronic inflammatory disorder of the airways associated with reversible airway obstruction and bronchial hyperreactivity to a variety of stimuli 1. Although each of these components is recognized as an important part of the asthmatic phenotype, the primary underlying abnormality is thought to be the chronic airway inflammation that gives rise to reversible obstruction, hyperresponsiveness and tissue remodelling 2. The pathological findings of asthma include not only infiltration of mononuclear cells, mast cells and eosinophils in the airway tissues, but also sub-basement membrane fibrosis 3. Indeed, even in mild asthma, the inflammatory cells recruited in the airways release products that are able to mediate changes in airway patency and responsiveness 2, to damage bronchial epithelium and to stimulate myofibroblast proliferation and collagen depositing below the epithelial basement membrane 4. As a consequence, two distinct processes can be observed: 1) regeneration ad integrum, leaving no residual trace of the previous injury; or 2) replacement by connective tissue, through the depositing of increased amounts of collagen (especially types III and V), polysaccharides (such as hyaluronic acid) and fibronectin by activated interstitial fibroblasts 2, 5, 6. Besides being modulated by a variety of cytokines and mediators released by airway inflammatory and parenchymal cells, fibroblasts may also interact directly with inflammatory cells, as demonstrated by the frequent observation of membrane apposition of eosinophils and myofibroblasts in the bronchial mucosa of asthmatic patients 4, 68. This close contact that is probably due to adhesion molecule expression on the interacting cell surface, may allow the stimulation of inflammatory cells by both direct cell-to-cell interaction and high cytokine concentrations 7. Fibroblasts express a variety of surface molecules that include intercellular adhesion molecule-1 (ICAM-1, CD54) and hyaluronic-cellular adhesion molecule (H-CAM, CD44), which belongs to a family of transmembrane glycoproteins 7, 8. During inflammation or experimental exposure to cytokines, adhesion molecule expression is upregulated on both airway parenchymal cells and polymorphonuclear leukocytes and appears to be deeply involved in leukocyte transmigration and activation 9. Indeed, eosinophils have been shown to adhere to fibroblasts and to be activated through the binding of ICAM-1 to leukocyte integrin leukocyte function-associated antigen-1 (LFA-1, CD11a/CD18), which is expressed on the eosinophil surface 8, 10. As a receptor for glycan hyaluronate, as well as for collagen and fibronectin 11, the CD44 transmembrane glycoprotein appears to be involved in regulation of cell locomotion and modulation of cell-to-cell and cell-to-matrix interactions that occur during fibrogenesis 8, 12. Since asthma is characterized by airway inflammation and reversible bronchial obstruction, it is treated with anti-inflammatory drugs (i.e. corticosteroids) and bronchodilators, such as ß2-adrenoreceptor agonists 1. The therapeutic activity of corticosteroids results, at least in part, from inhibition of inflammatory mediator synthesis, release by parenchymal and inflammatory cells, and prevention of some of the histological changes that characterize airway remodelling in asthma, including subendothelial fibrosis 1, 13. Similarly, it has been recently shown that ß2-adrenoreceptor agonists have in vivo and in vitro pharmacological properties suggesting an inhibitory effect on various aspects of the inflammatory response related to asthma 14. With this background, a study was designed to evaluate in vitro, the activity of an inhaled glucocorticoid (fluticasone propionate (FP)) and of a long-acting ß2-adrenoreceptor agonist (salmeterol xinafoate (S)) on fibroblast proliferation and adhesion molecule expression.
Fibroblast culture GM 06114, a human foetal lung fibroblast cell line, was used in all the experiments performed. The GM 06114 cell line was selected because, in preliminary sets of experiments, it reacted to cytokine stimulation in a similar manner to human lung fibroblast primary cultures 7, 15. Fibroblasts were cultured until confluence, then collected and spread into 96-well plates (7,000 cells·well1) or into 24-well plates (60,000 cells·well1) for the evaluation of adhesion molecule expression. The cells were incubated with Dulbecco's Modified Eagle Medium (DMEM) (Euroclone Ltd, Paignton, Devon, UK) supplemented with 20% foetal calf serum (FCS; Euroclone Ltd) and penicillin/streptomycin (5,000 IU·mL1) (ICN Biomedicals srl, Costa Mesa, CA, USA) for 24 h at 37°C in 5% carbon dioxide (CO2). The medium was then removed and the cells resuspended in serum-free DMEM and incubated for an additional 48 h. The medium was then removed and replaced with fresh serum-free DMEM, in the presence of different concentrations of different stimuli: basic fibroblast growth factor (bFGF, 0.5100 ng·mL1) (SIGMA, Saint Louis, Missouri, USA), tumour necrosis factor- (TNF- , 0.520 ng·mL1) (Euroclone Ltd) or interleukin-4 (IL-4, 0.520 ng·mL1) (PeproTech EC Ltd, London, England). Unstimulated cells were used as negative control. After 24 h-incubation, cell proliferation and molecule adhesion expression were evaluated as described later.
Cell proliferation assay
Evaluation of adhesion molecule expression by human foetal lung fibroblast
To evaluate the effect of FP and/or S on adhesion molecule expression, fibroblasts were stimulated for 24 h with TNF-
Data and statistical analysis
Fibroblast proliferation and adhesion molecule expression following stimulation with basic fibroblast growth factor, tumour necrosis factor- or interleukin-4Among the stimuli tested, bFGF was highly effective in inducing a dose-dependent lung fibroblast proliferation, statistically significant at all the concentrations tested (p<0.01, each comparison with control cultures; fig. 1 (p>0.05, each comparison; fig. 1
Unstimulated fibroblasts grown in DMEM medium showed negligible basal expression of ICAM-1 but elevated levels of H-CAM (fig. 2 (5 ng·mL1) was able to induce an upregulation of ICAM-1 expression by lung fibroblasts (control: 6.88±0.44 mfc; TNF- 5 ng·mL1: 21.47±2.44 mfc; p=0.0004) (fig. 2
Neither TNF- nor IL-4 were able to significantly increase the expression of H-CAM, probably because of its high basal expression (control: 129.71±5.60 mfc) (fig. 2
Effect of fluticasone propionate and/or salmeterol on fibroblast proliferation
Effects of fluticasone propionate and/or salmeterol on adhesion molecule expression When evaluating changes in adhesion molecule expression induced by FP and S, it was found that both drugs were able to significantly inhibit ICAM-1 and H-CAM expression on human lung fibroblast surface. Exposure to FP induced a dose-dependent downregulation of TNF- -stimulated ICAM-1 expression, significant at all the concentrations tested (p<0.001, each comparison with TNF- -stimulated cultures; fig. 4 -stimulated ICAM-1 expression was also induced by S at all the concentrations tested (p<0.01, each comparison with TNF- -stimulated cultures). The inhibitory effect of S was lower than that observed in the presence of FP at the three highest concentrations tested (p<0.01; fig. 4
Since neither TNF- nor IL-4 modified H-CAM expression, after the preliminary experiments the effect of the two drugs on this adhesion molecule was evaluated on unstimulated fibroblasts. Each concentration of FP or S alone significantly inhibited H-CAM expression (fig. 4
Evaluating in vitro, a human foetal lung fibroblast cell line, it has been shown that long-acting ß2-agonist S is highly effective in inhibiting fibroblast proliferation and ICAM-1 and CD44 surface molecule expression. While no changes in cell proliferation were observed in the presence of FP, this drug was more effective than S in downregulating ICAM-1 and CD44 expression. Finally, associated S significantly increased the inhibitory activity of FP alone on ICAM-1 expression. Subepithelial layer thickening associated with an increased number of fibroblasts is characteristic of airway remodelling in asthma 4, 5, 1820. Evaluation of endobronchial biopsy specimens from asthmatics showed that the subepithelial layer consisted of dense fibrillary collagen types III and V of fibroblasts, rather than epithelial origin 5, 6. Although the effects of inflammatory processes on collagen deposits are still unclear, several inflammatory cell types, including mast cells, macrophages, eosinophils and epithelial cells, are able to produce and release factors that may stimulate fibroblasts 6. In addition to secreting extracellular matrix components, fibroblasts are able to release a variety of inflammatory cytokines and chemokines 21, 22 and to express surface molecules, such as ICAM-1 15 and CD44 8, 9. ICAM-1 acts as a ligand for surface molecules expressed on eosinophils, monocytes and lymphocytes 23. The expression of ICAM-1 is increased in tissues with inflammatory reactions 24 and it is associated with leukocyte transmigration over fibroblasts 25 and with increased eosinophil adhesion and activation 10. Therefore, an increased expression of ICAM-1 in response to cytokines may represent a mechanism by which fibroblasts adhere to and interact with inflammatory and immuno-effector cells in the respiratory mucosa of atopic patients 10, 23. The CD44 transmembrane glycoprotein represents a receptor for the extracellular matrix components and is thought to be involved in modulation of cell-to-cell and cell-to-matrix interaction that occurs during fibrogenesis 8, 12. In addition to providing mechanical support to the tissues, the extracellular matrix is increasingly recognized as a source of signals for cellular localization, migration, differentiation and activation 12, 26, 27. Recent data indicate that CD44 may be related to fibroblast migration and pulmonary repair after injury. CD44 is the surface receptor for matrix proteins supporting cell motility, including fibronectin and hyaluronan 28, and its expression is increased in hypertrophic scar fibroblasts 29. Interestingly, lung tissues from patients with acute alveolar fibrosis revealed CD44-expressing mesenchymal cells through newly formed fibrotic tissue and high concentrations of this surface protein were found along the motile structures (filopodia and lamellopodia) of fibroblasts 28. The most surprising finding of the present study was that S was more effective than FP in inhibiting bFGF-induced fibroblast proliferation. ß2-adrenoceptor agonists interact with specific cell membrane receptors, which activate signal transduction mechanisms leading to cellular responses that may be different in different cell types 14. Indeed, in addition to inhibiting or reversing the contractile response of bronchial smooth muscles, these compounds modulate a variety of inflammatory and immuno-effector cell functions, including T-lymphocyte proliferation and mononuclear cell surface receptor expression 14, 16, 30. Using a human foetal lung fibroblast cell line, this study has shown that S was highly effective in downregulating fibroblast proliferation and in reducing the expression of adhesion molecules that are likely to be involved in airway remodelling. In addition, at the two highest concentrations tested, S significantly increased the inhibitory activity of FP alone on ICAM-1 expression. These findings are in agreement with the results of a recent report, showing inhibition of ICAM-1 expression by human lung fibroblast primary cultures in the presence of budesonide and formoterol 31. Indeed, ß2-adrenergic receptor agonists may also activate glucocorticoid receptors in primary human lung fibroblasts 32. Although more effective than S in reducing surface molecule expression, FP did not have a direct inhibitory activity on bFGF-induced fibroblast proliferation. Glucocorticosteroids influence not only protein translation, but also post-transcriptional processing of protein synthesis 33. The complexity of the mechanisms involved in glucocorticoid activities may explain why cellular responses to these drugs are not only highly dependent on cell types, but also on cell functions 3335. Clinical studies demonstrate that in asthma, low-dose inhaled corticosteroids are not only able to downregulate the intensity of bronchial inflammation, but also to decrease subepithelial layer thickness 6, 14. Indeed, in vivo corticosteroids may be effective in controlling tissue remodelling by inhibiting several aspects of the inflammatory process, such as regulation of the transcription of genes related to the production of cytokines able to activate leukocytes and lung parenchymal cells 5, 6, 3133. In summary, these data support at biological level the hypothesis that the combination of the two types of drugs, long-acting ß2-adrenoceptor agonists and inhaled steroids, not only allows a better clinical control of asthma, but may also help to prevent long-term deterioration of lung function 1. Although the concentrations of drugs used in vitro in this paper are similar to those found in peripheral lung tissue after 50 µg inhaled dose of salmeterol 35, the clinical importance of the experimental observations reported here deserves further investigation.
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||