What causes airway remodelling in asthma?

Recently, the term ‘remodelling’ has been applied to structural changes seen in the asthmatic airway. Whether remodelling is a separate factor contributing to outcome, and what causes it, are fundamental questions for those planning treatment strategies in asthma care. Trends in asthma therapy have reflected contemporary concepts of airflow limitation for some time. The description of heightened bronchial reactivity has been an ideal ‘concept’, with sound physiological descriptors, leading to a redirection of aims of asthma therapy reflected in global guidelines [ 1, 2]. Elimination of factors that cause increased reactivity and the use of reactivity-modifying drugs has led to better symptomatic control and better long-term outcome. The ‘rediscovery’ of mucosal inflammation in asthma began a further revolution in thought. To gain control of asthma (symptoms, lung function, inflammation) and indeed improve outcome, high-dose inhaled corticosteroids have been incorporated into asthma management plans with objective benefit [ 3, 4]. The inflammatory process is complex, even in relatively mild, stable asthma. It reflects the acute changes of today's early asthmatic response, as well as yesterday's late asthmatic response. In practice, this does not occur on the clean ‘green field’ site of experimental models, but on the background of years of chronic, persistent inflammation. The short-term classic allergic reactions are accompanied by well-characterized cellular influx, epithelial injury and oedema, associated with the production of inflammatory cytokines and mediators [ 5, 6]. The continuous process of injury and healing is associated with specific changes now associated with remodelling [ 7]. The legacy of ongoing inflammation is a pattern of nonacute changes leading to thickening of the airway wall [ 8, 9], including smooth muscle hypertrophy/hyperplasia [ 10], subepithelial collagen deposition [ 11], submucosal collagen deposition [ 12] and increased vascularity [ 13]. These components of remodelling have an undetermined effect on symptom control, lung function, airway reactivity and outcome from acute attacks of asthma. Fibrosis within the airway wall (together with other features of remodelling) could lead to several adverse consequences. These include a poorer response to bronchodilator therapy during severe attacks, worsening of airflow obstruction because of mucosal inflammation and vascular congestion occurring internally and concentric to a remodelled zone with increased rigidity [ 14] and, possibly increased bronchial reactivity [ 15, 16]. The stiffness of the remodelled airway wall is increased by epithelial folding, enhanced further by mucosal thickening [ 17]. Circumstantial evidence links the use of inhaled corticosteroids with reversal of some structural changes (other than the more acute changes such as eosinophil infiltration and tissue oedema) with improved clinical indices. The 6 month study by Hoshino et al. in this edition of Clinical and Experimental Allergy [ 18] confirms the earlier work of Trigg et al. in which beclomethasone dipropionate was shown to cause an apparent reduction in thickness of subbasement membrane collagen [ 19]. As well, tenascin expression has been shown to be reduced in the mucosa of pollen-sensitive asthmatics treated with budesonide [ 20]. This reduction in airway wall matrix is in keeping with the evidence that corticosteroids are capable of reducing IGF-I dependent cultured fibroblast activity [ 21]. Inhaled corticosteroids have been shown to act systemically in asthmatics to reduce total skin collagen thickness [ 22], in keeping with the in vitro down-regulation of type I and III skin collagen production after relatively short-term exposure [ 23]. The evidence for true reversal of the remodelling process must be closely examined, before conclusions can be drawn regarding specific airway changes. Corticosteroids have been shown to reduce the airway cellular infiltrate in asthma when inhaled [ 24, 25], reduce airway smooth muscle cell proliferation [ 26], inhibit pro-collagen synthesis [ 27] and cause vasoconstriction [ 28]. The apparent reversal of subepithelial collagen remodelling reported by Hoshino et al. is accompanied by a suppression of expression of IGF-I by airway cells in almost all bronchial biopsies. There was apparently no suppression of TGF-β1 or PDGF. Given the known suppression of cytokine production by airway cells after treatment with inhaled corticosteroids, it is of interest that only one of three growth factors capable of causing pro-collagen production, fibroblast proliferation and possibly fibroblast chemotaxis [ 29] could be suppressed. It is also interesting to note that both TGF-β1 and PDGF have been identified previously in asthmatic airways [ 30, 31]. In these studies, both were shown to be increased in more severe asthma and expression was largely in eosinophils (over 99% for TGF-β1 in asthma). It would be reasonable to expect that many of the positive cells staining for these factors would be reduced after inhaled corticosteroids, either because of a reduction in eosinophil numbers or because of reduced activity. The lack of steroid sensitivity for TGF-β1 and PDGF expression may be unexpected, given the commonly observed suppression of eosinophil numbers after effective anti-inflammatory treatment. Rather than risk over-interpretation, it is clear that these factors are some of many shown to influence collagen production and fibroblast function [ 29]. The mechanism and regulation of action of growth factors is still speculative. Factors such as IGF-I are present in significant amounts in the human lung [ 32, 33]. Nuclear activation factors, such as NFκB and AP-1 which may be required to induce protein production in the remodelling process, are suppressible by corticosteroids [ 34, 35]. Cell replication, especially involving smooth muscle cells, may involve the action of specific mitogen-activated protein kinases (MAPK), which may not be inhibited by corticosteroids [ 26]. Growth factors, capable of mediating remodelling events in asthma, may therefore have differential responsiveness to inhaled corticosteroids, depending on the relevance of primary signal transduction pathways. The variability, and prolonged period of improvement in lung function after the initiation of treatment with inhaled corticosteroids [ 36] suggests that multiple processes are implicated in airway remodelling (and its reversal), and that some of these are relatively steroid resistant, such as TGF-β1 and PDGF expression as described by Hoshino et al. [ 18]. This variable response to the introduction of inhaled corticosteroids in asthma indicates a wide heterogeneity, at least at the functional level, in individual patient responsiveness [ 37]. Identification of the specific actions of growth factors and the subsequent elucidation of steroid-sensitive aspects of their function (expression, release, binding and cell-activation) will be essential to improve the potential role of anti-inflammatory medication in the reversal of airway wall remodelling in asthma.