Introduction
Asthma is a chronic inflammatory airway disease that is traditionally characterized by reversible central airway obstruction and airway hyperreactivity [
1,
2]. Although treatment with bronchodilators and inhaled glucocorticosteroids (ICS) normally provide good control of the disease, a significant proportion of the asthmatic patients have persistent symptoms despite conventional therapy [
3]. This phenomenon, called uncontrolled asthma [
4,
5], represents a key challenge for increasing asthma control. Little is known about the inflammatory and remodeling processes causing persisting symptoms in this group of asthmatics. A likely cause of these symptoms is the presence of steroid-resistant components in both central and peripheral airways, as well as conventional inhalation therapy’s insufficient ability to reach the peripheral airways [
6]. The peripheral airways have an important but poorly studied role in asthma pathophysiology [
7,
8]. The few previous studies that have studied transbronchial biopsies from patients with asthma provide clear indications that both small airways and alveolar tissues may be subjected to cellular inflammation [
8‐
10].
In an inflammatory environment such as the asthmatic airway, fibroblasts are activated to myofibroblasts, which deposit molecules including collagens and proteoglycans into the surrounding extracellular matrix [
11‐
13]. Myofibroblasts can be induced by a variety of factors e.g. transforming growth factor (TGF)-β
1, the alternatively spliced domain A of fibronectin (EDA-fibronectin) as well as mechanical force [
14]. Proteoglycans have several important functions in the tissue including storing cytokines such as TGF-β
1, acting as structural components in tissue organization by forming large complexes with hyaluronan and collagens, as well as cell signaling affecting differentiation, migration and inflammation [
15‐
17]. The matrix metalloproteinases (MMPs) are a family of proteases, involved in the breakdown of extracellular matrix [
18]. A disruption in the balance between matrix production, MMPs and their regulators tissue inhibitors of metalloproteinases (TIMPs) can result in changes in extracellular matrix composition and fibrosis [
19].
We have previously shown that increased collagen deposition occurs in alveolar parenchyma already in patients with mild untreated asthma [
20]. However, how remodeling is linked to clinical control of asthma remains to be investigated. The aim of this study is to further advance this concept by a thorough investigation of extracellular matrix components and fibroblast densities in patients with controlled asthma and uncontrolled asthma on equivalent doses of ICS, compared to healthy controls. We hypothesized that patients with uncontrolled asthma have an altered tissue composition compared to both patients with controlled asthma and healthy controls, which could be one cause of the persistent symptoms in this group of patients.
Discussion
This study shows that the bronchial wall and alveolar parenchyma have an altered composition of matrix that differs between patients with controlled and uncontrolled asthma on equivalent doses of ICS, which could contribute to the symptoms in patients with uncontrolled asthma.
Due to the inaccessibility of tissue from alveolar parenchyma of asthmatic patients little is known regarding the remodeling processes in the peripheral lung and how these changes contribute to asthma pathology [
20]. The present study applied a unique study design of obtaining bronchial as well as transbronchial biopsies from patients with uncontrolled and controlled asthma on equivalent doses of ICS, as well as from healthy control subjects. Emerging new evidence show that distal airways are subjected to an inflammatory process in asthmatic patients [
8,
10]. TGF-β
1 plays a central role in airway remodeling and has been found to induce production of several types of collagens from myofibroblasts [
22,
23]. Although the patients with uncontrolled asthma had significantly thicker reticular basement membrane [
24] there was no difference in total collagen content in central airways when measuring the whole biopsy. However, expression of collagen in distal airways was increased in uncontrolled asthmatics but not in patients with controlled asthma, compared to controls. Thickening of alveolar walls due to increased amounts of collagen may lead to impaired gas exchange in the alveoli. We have previously reported on increased collagen in distal airways of patients with mild untreated asthma [
20], which raises the question if this accumulation of collagen could be decreased with ICS treatment and also why the patients with uncontrolled asthma have increased areas of collagen despite treatment. The patients were determined to be uncontrolled at the time when the biopsies were taken - could an increased dose of ICS reverse this accumulation of collagen?
Biglycan and decorin are two small leucine-rich proteoglycans involved in collagen fibrillogenesis and fibril stabilization in the tissue [
25]. Both decorin and biglycan have been found to bind TGF-β
1in vitro[
26]. However, a study by Kolb
et al. showed that,
in vivo, only decorin and not biglycan was found to interfere with TGF-β
1 activity [
27]. In addition, decorin has been found to act protective against liver fibrosis by attenuating TGF-β
1 signaling [
28]. In this aspect, one can speculate that the increase in decorin expression in patients with uncontrolled asthma could be a protective mechanism to mitigate remodeling. On the other hand, increased amounts of decorin could by regulating cross-linking and interfibrillar spacing create a stiffer collagen matrix, which could affect the overall elasticity of the lung tissue. Furthermore, decorin could also have a role in protecting collagen fibrils from cleavage by collagenases, as
in vitro studies of collagenases has shown [
29], which could contribute to the accumulation of collagen in the tissue. While the quantity of proteoglycan present is partly responsible for collagen fibril morphology, variations in the glycosaminoglycan chains have also been shown to play important roles in e.g. collagen fibril size control and interfibrillar spacing [
30‐
32]. Characterization of glycosaminoglycan chains from isolated biglycan and decorin from asthmatics and healthy individuals could provide valuable information on the subject.
Versican is a large proteoglycan that is highly interactive with several matrix components including hyaluronan and fibrillin [
33,
34]. Increased levels of versican have been reported in diseases such as atherosclerosis and cancers [
35,
36]. Accumulation of versican, as seen in the central airways of patients with uncontrolled asthma, could also lead to increased stiffness around cells, which in turn can influence their ability to migrate, proliferate, adhere and remodel the matrix [
37]. Versican could also increase stiffness in the lung by inhibiting elastin-binding proteins and interfering with the assembly of elastic fibers, thereby affecting lung function [
38]. We can only speculate regarding the decreased percentage areas of biglycan and versican in controlled asthmatics. In cultured fibroblasts serum induced production of proteoglycans is reduced by addition of corticosteroids [
39], since the fibroblasts used in this study are “healthy”, this could be in accordance with how fibroblasts in patients with controlled asthma respond to corticosteroids. It is possible that there is a difference in fibroblast response between the controlled and uncontrolled asthmatics, or that the inhaled corticosteroids do not reach the peripheral airways.
There was a decreased ratio between MMP-9 and TIMP-3, in central and distal airways of both patient groups, which indicates a proteolytic-antiproteolytic imbalance. Although, since the ratio is decreased in both patient groups it does not explain the differences in percentage areas of matrix molecules between patients with controlled and uncontrolled asthma. Additional analyses of other MMPs, TIMPs and their relationship with structural alterations in the lung tissue need to be further examined. Taken together, these structural alterations affect phenotypes of cells and the biomechanical properties of the lung.
The fibroblasts/myofibroblasts play a pivotal role in remodeling processes as the main extracellular matrix producing cells. Corticosteroids have been found to prevent myofibroblast accumulation and airway remodeling in mice [
40]. Interestingly, we found increased numbers of myofibroblasts in alveolar parenchyma in patients with uncontrolled asthma compared to healthy controls while the number of myofibroblasts was decreased in patients with controlled asthma compared to both controls and patients with uncontrolled asthma. Levels of TGF-β
1 mRNA as well as immunoreactivity have been found to be increased in the airways submucosa of asthmatic patients, with a direct correlation to the severity of the disorder [
41]. Interleukin-13 (IL-13) is another important mediator in asthma pathogenesis, which has been associated with tissue fibrosis, both by inducing proliferation of fibroblasts and collagen production but also by activation of TGF-β
1[
42‐
44].
EDA-fibronectin cooperates with TGF-β
1 in modulating fibroblasts into the activated myofibroblast phenotype [
14]. The increased amount of myofibroblasts in the alveolar parenchyma of patients with uncontrolled asthma could be linked to the increased expression of EDA-fibronectin, and is in accordance with our finding of an increased percentage area of several matrix components such as collagen and decorin.
The differences in myofibroblast numbers, in patients with controlled asthma compared to uncontrolled asthma, raises the question how the ICS treatment affects the tissue and why it varies between patient groups. Could it be due to differences in steroid-sensitivity, or does the inhalation therapy not reach the peripheral airways? Indeed, in patients with controlled asthma there was a negative correlation between numbers of myofibroblasts and daily dose of ICS, which was not found in the group of uncontrolled asthmatics. Although ICS has a well-documented effect on airway inflammation, little is known regarding long-term effects on remodeling and mesenchymal cell growth. Effects of steroids on extracellular matrix seem to depend on the nature of the study since in vitro and in vivo animal and human studies often show conflicting results. This matter is further complicated by the possibility that myofibroblasts respond differently to steroids in some patient-groups which is concordant with the heterogeneity of asthma.
Summary
Our study shows that tissue composition differs between patients with controlled and uncontrolled asthma on equivalent doses of ICS. Our data support the notion that patients who have remaining symptoms despite conventional ICS therapy have a pronounced matrix remodeling in both central and distal airways and should thus benefit from treatment strategies that target specific inflammatory and remodeling responses in the distal lung. The increased percentage areas of several matrix components and myofibroblasts could be a contributing factor to the persistent symptoms in uncontrolled asthma. Due to the heterogeneity of the disease, asthma medicine needs to be individualized, and new treatments for prevention of remodeling needs to be developed. Currently, several therapies directed at IL-13 are in development [
45]. How those therapies and other systemic treatments like oral corticosteroids, Omalizumab (Anti-IgE) and Mepolizumab (Anti-IL-5) etc. affect the inflammatory response, and structural changes in the peripheral lung, is an important future research area.
Acknowledgements
We thank Karin Jansner, Britt-Marie Nilsson, Lena Thiman and Marie Wildt, for skillful technical assistance with tissue-processing and ELISA.
Funding
This study was supported by the Swedish Medical Research Council (11550), Stockholm Sweden, the Evy and Gunnar Sandberg foundation, Lund, Sweden, the Heart-Lung Foundation, Stockholm, Sweden, Greta and John Kock, Trelleborg, Sweden, the Alfred Österlund Foundation, Malmö, Sweden, the Anna-Greta Crafoord Foundation, Stockholm, Sweden, the Konsul Bergh Foundation, Stockholm, Sweden, the Royal Physiographical Society in Lund, Sweden and the Medical Faculty of Lund University, Sweden.
Competing interest
None of the authors has a financial relationship with a commercial entity that has an interest in the subject of the presented manuscript or other competing interests to disclose.
Authors’ contributions
MW: Contributed to the study design, performed laboratory work, quantified immunostainings, performed statistical analysis, interpreted the data and wrote the manuscript. CA: Contributed to the study design, performed laboratory work, interpreted the data and critically revised the manuscript. AAS: Interpretation of data and revision of the manuscript. ET: Contributed with clinical characterization of the patients, performed MMP-9 ELISA on sputum and revised the manuscript. LB: Recruited patients, collected materials, supervised the study and revised the manuscript. JE and GWT: Contributed to study design, supervised the study, interpreted data and critically revised the manuscript. All authors approved the final version of the manuscript.