Most patients with cystic fibrosis (CF) develop progressive airflow obstruction. A subgroup of these patients also have airway hyperresponsiveness to inhaled bronchoconstricting agents^1,2 and reversibility of airflow obstruction in response to bronchodilators.³ Parallels have been drawn between these observations and other airway diseases manifesting with airway obstruction and bronchial hyperresponsiveness such as asthma and chronic obstructive pulmonary disease (COPD). This led to speculation that remodelling of the airway smooth muscle (ASM) layer may contribute to bronchial hyperresponsiveness in CF.

Studies in the past have been restricted to patients with severe CF. Pathological studies of the lungs of these patients obtained either at necropsy or following transplantation or lobectomy showed an increase in smooth muscle area compared with healthy controls or patients with COPD.^4,5 In this issue of Thorax Hays et al⁶ have, for the first time, studied the ASM layer in patients with mild to moderate CF using bronchoscopy and design based stereology. They found that the volume of smooth muscle in the airway submucosa in subjects with CF was higher than in normal controls and, furthermore, that this difference was attributable to smooth muscle cell hyperplasia rather than hypertrophy. This study raises several interesting questions. 1. Is this increase in the volume of the smooth muscle layer responsible for the airway hyperresponsiveness seen in many CF patients? 2. What factors are produced in the CF airway that could promote ASM hypertrophy? 3. Is there a relationship between the defective ion transport underlying CF pathophysiology and smooth muscle function in the CF airway?

Is the smooth muscle cell hyperplasia demonstrated by Hays et al sufficient to explain airway hyperresponsiveness in CF patients? Mathematical modelling approaches suggest that changes in airway dimensions in CF, including an increase in the smooth muscle area, probably contribute to airflow obstruction and bronchial hyperresponsiveness in these patients.⁴ Interestingly, tumour necrosis factor α (TNF-α), a cytokine found in increased amounts in the CF airway, has been shown to potentiate ASM contraction in response to cholinergic stimulation in vitro,⁷ and there is evidence that TNF-α induces a hypercontractile phenotype by enhancing agonist induced calcium signals, as well as agonist induced force generation.⁸ Thus, it may be that factors other than ASM hyperplasia alone contribute to bronchial hyperresponsiveness in CF. It would be interesting to investigate whether the presence and degree of airway hyperresponsiveness correlates with smooth muscle hyperplasia in these patients. The difficulty is that it is not possible to study the small airways—which collectively make the greater contribution to airflow obstruction—in the subgroup of patients with mild to moderate disease.

There are interesting similarities and contrasts between the inflammatory milieu found in the asthmatic and CF airways. Inflammation in CF is primarily neutrophil driven whereas, in asthma, T lymphocytes, eosinophils and mast cells are of greater importance although neutrophils may play a role in severe asthma. This has consequences for the range of cytokines and mediators that predominate in the inflamed airway. Airway inflammation and remodelling involves not just inflammatory cells but also structural cells such as fibroblasts, epithelial cells, and ASM cells. Inflammatory and structural cells produce cytokines, mediators, matrix modifying enzymes, chemokines, and growth factors that initiate and perpetuate inflammation and remodelling. The interplay between these cells and the multitude of biologically active molecules that they secrete is complex and not fully understood. However, those factors that promote proliferation of ASM cells in vitro include growth factors (such as basic fibroblast growth factor,⁹ transforming growth factor-β,⁹ platelet derived growth factor,¹⁰ epidermal growth factor¹¹ and insulin-like growth factor¹²), mediators including endothelin-1¹³ and cysteinyl leukotrienes,^12,14 proteolytic enzymes such as thrombin¹⁵ and mast cell derived tryptase,¹⁶ and cytokines including interleukin (IL)-1β.¹⁷ Many of these factors are found in increased amounts in the asthmatic airway and their influence on ASM function and airway remodelling has been the subject of intense study over the last decade. Some of these factors are also present in inflamed airways in CF.

It remains a matter of debate whether inflammation in the CF lung is driven by the opportunistic infections characteristic of this disease or whether it begins early in the disease, independent of lung infection. However, what is clear is that a chain of events occurs leading to a vicious cycle of infection, inflammation, lung tissue damage and further vulnerabililty to infection. Bacteria stimulate macrophages to produce IL-1β and TNF-α which, in turn, stimulate epithelial cells to produce chemokines such as IL-8, cytokines such as IL-1, and growth factors such as GM-CSF. IL-8 attracts neutrophils to the inflammatory site where they release LTB₄, reactive oxygen species, and proteases such as elastase which damage airway structural proteins and further stimulate cytokine production by epithelial cells. Since ASM mitogenesis can be stimulated in vitro by the actions of proteolytic enzymes including thrombin¹⁵ and tryptase,¹⁶ it is interesting to speculate whether proteases in the CF airway may have a similar effect. Other mediators and cytokines found in increased amounts in the CF airway which could promote ASM proliferation include the cysteinyl leukotrienes which augment growth factor induced ASM proliferation,¹⁴ and the potent smooth muscle mitogen and bronchoconstrictor endothelin-1.¹⁸

In CF a number of different mutations in the gene encoding the cystic fibrosis transmembrane conductance regulator (CFTR) lead to defective chloride secretion by epithelial cells and disordered ion transport. This alters the composition of the airway surface liquid, predisposing to pulmonary infection. In the intestine the myofibroblast layer contributes to chloride secretion by intestinal epithelial cells through the production of prostaglandin E₂ (PGE₂) mediated by cyclooxygenase in response to inflammatory cytokines¹⁹ and thrombin.²⁰ It is possible that ASM derived factors in the CF airway can modify airway epithelial ion transport. Indeed, it is known that a number of inflammatory cytokines and mediators increase PGE₂ production by ASM cells in vitro^21,22 and, furthermore, that PGE₂—the dominant prostanoid produced by ASM under inflammatory conditions—stimulates chloride secretion by airway epithelial cells.²³ It may be that ASM hyperplasia in CF is an adaptive response which compensates, in part, for defective chloride secretion.

Is it possible that the defect in the CFTR itself contributes to ASM hyperplasia in CF? Interestingly, recently published data show that inhibition of the Na–K–2Cl co-transporter by diuretics inhibits ASM proliferation in vitro,²⁴ suggesting that alterations in ion flux may modify hypertrophic or hyperplastic processes in these cells.

The findings of Hays et al add an important contribution to our knowledge about airway remodelling in CF—specifically that changes in the ASM layer are not confined to patients with severe disease but are present in those with mild to moderate disease. This is of particular importance as it suggests a possible mechanism underlying the airway obstruction and hyperresponsiveness observed in these patients in the clinic. The biological mechanisms underlying ASM hyperplasia in CF warrant further study which will enhance our understanding of the pathophysiology of this disease and may lead to novel approaches to treatment.