Published ahead of print on January 16, 2003, doi:10.1164/rccm.200209-1030OC
© 2003 American Thoracic Society Airway Structural Alterations Selectively Associated with Severe AsthmaInstitut National de la Santé et de la Recherche Médicale Unité 408, Faculté de Médecine Xavier Bichat and Service de Pneumologie, Hôpital Bichat, Paris, France Correspondence and requests for reprints should be addressed to Marina Pretolani, Ph.D., INSERM U408, Faculté de Médecine Xavier Bichat, 16 rue Henri Huchard, 75018 Paris, France. E-mail: mpretol{at}bichat.inserm.fr
To identify airway pathologic abnormalities selectively associated with severe asthma, we examined 10 control subjects, 10 patients with intermittent asthma, 15 patients with mild-to-moderate persistent asthma, 15 patients with severe persistent asthma, and 10 patients with chronic obstructive pulmonary disease. Bronchial biopsies were assessed for epithelial integrity; subepithelial basement membrane (SBM) thickness; collagen type III deposition; eosinophil, neutrophil, and fibroblast numbers; mucous gland and airway smooth muscle (ASM) areas; SBM-ASM distance; ASM hypertrophy (increased cell size); and the expression of the contractile proteins -actin, smooth muscle myosin heavy-chain isoforms, myosin light-chain kinase, and the phosphorylated form of the regulatory light chain of myosin. Neither mucosal eosinophilia nor neutrophilia, epithelial damage, or SBM thickness reflected asthma severity. In contrast, higher numbers of fibroblasts (p < 0.001), an increase in collagen type III deposition (p < 0.020), larger mucous gland (p < 0.040) and ASM (p < 0.001) areas, augmented ASM cell size (p < 0.001), and myosin light-chain kinase expression (p < 0.005) distinguished patients with severe persistent asthma from patients with milder disease or with chronic obstructive pulmonary disease. Stepwise multivariate regression analysis established that fibroblast numbers and ASM cell size were negatively associated with prebronchodilator and postbronchodilator FEV1 values in patients with asthma. We conclude that fibroblast accumulation and ASM hypertrophy in proximal airways are selective determinants of severe persistent asthma.
Key Words: airway smooth muscle hypertrophy myosin light chain kinase fibroblasts chronic obstructive pulmonary disease
Asthma is classically defined as reversible airflow obstruction and often remits in younger subjects with milder disease (1). However, various degrees of airflow obstruction may persist, and in the long term, asthma may become moderately to fully irreversible (23). Accordingly, a proportion of patients with asthma experiences chronic symptoms, episodic exacerbations, and severe persistent airway obstruction, despite the continuous use of ß2-agonists, associated with high doses of inhaled corticosteroids and pulses or regular oral doses of corticosteroids (3, 4). These patients represent a major clinical and pharmaeconomic problem, as they contribute to the majority of asthma costs through hospitalization, emergency visits, absence from work or school, and use of medication (4). Alternative antiinflammatory and immunomodulating drugs, such as methotrexate, gold, cyclosporin, intravenous Persistent airway inflammation and irreversible structural changes of the bronchial wall, defined as airway remodeling (68), are thought to play a prominent role in severe asthma. These alterations include sustained tissue eosinophilia, epithelial damage, subepithelial basement membrane (SBM) thickness, subepithelial fibrosis, as well as mucous gland and airway smooth muscle (ASM) hypertrophy and/or hyperplasia (69). In particular, the increase in ASM content is believed to explain the majority of airway luminal narrowing and the permanent reduction of the airway caliber in severe asthma (10, 11). Although several reports in the literature have related the clinical severity of asthma to some aspects of the inflammatory and remodeling response of the airways, particularly persistent eosinophilia, greater neutrophilia, SBM thickness, and subepithelial fibrosis (9, 12), findings of other studies failed to confirm this relationship (7, 8). Accordingly, the possibility that specific airway structural and molecular determinants would distinguish severe patients with asthma from patients with milder disease remains elusive. Using endobronchial biopsies from patients with asthma with different severity, we compared several features of airway inflammation and remodeling, such as epithelial damage, SBM thickening, collagen type III deposition, and an increase in mucous glands and in the number of eosinophils, neutrophils, and fibroblasts in the bronchial submucosa. In addition, because structural and functional alterations of the ASM are claimed to participate in airway functional abnormalities in asthma (13), ASM content, hypertrophy, proliferation, and changes in the expression of different contractile proteins were assessed. Finally, to determine whether the observed differences were specific of the asthma process, a group of patients with chronic obstructive pulmonary disease (COPD), another inflammatory disorder of the lung associated with persistent airflow limitation and with parenchyma and airway wall inflammation and remodeling (12, 14), was also examined. Some of the results of these studies have been previously reported in the form of an abstract (15).
Patients Forty patients with asthma fulfilling the criteria of the Guidelines for the Diagnosis and Management of Asthma of the National Heart, Lung, and Blood Institute/World Health Organization (NHLBI/WHO) (16) were recruited (Table 1) . Patients with asthma were classified as having intermittent, mild-to-moderate persistent, and severe persistent asthma, based on the frequency and the intensity of symptoms, the need and dose of medications, and the FEV1 values (Table 1). Ten COPD patients fulfilling the criteria of the NHLBI/WHO Global Initiative for COPD (17) and 10 healthy volunteers were also enrolled (Table 1). All subjects gave their written informed consent, and the protocol was approved by the Ethics Committee of the Hôpital Bichat (CP 02,819).
Fiberoptic Bronchoscopy and Biopsy Processing The bronchoscopy was performed by the same operator, according to the guidelines of the American Thoracic Society (18). Six biopsy specimens were taken from the subcarinae and were snap frozen.
Serial 5-µm sections were collected on silane-coated glass slides and fixed in acetone. Ki67 (proliferating cells), eosinophil cationic protein (eosinophils), elastase (neutrophils), propyl-4-hydroxylase (fibroblasts), collagen type III deposit, Morphometry was assessed by computer-assisted image analysis (Microvision Instruments, Histolab, Evry, France) on hematoxylin-stained sections. Total biopsy areas were determined in at least two noncontiguous tissue sections from the same biopsy, and values are expressed in mm2. SBM thickening (in µm) and epithelial integrity, defined as the percentage of length of the basement membrane with intact epithelium, were assessed (22, 23). The percentage area of the bronchial submucosa occupied by mucous glands and by ASM was determined. ASM content was also evaluated indirectly by measuring the distance (in µm) between the SBM and the ASM at regular intervals of 50 µm. A minimum of 20 measurements was assessed for each section. The diameter of individual ASM cells (in µm) was measured across the nucleus in a minimum of 50 cells in randomly selected fields, and at least two fields were examined for each tissue section. This measurement was repeated in bronchial tissue sections previously reacted with the anticollagen type III antibody. All of the parameters were examined in two to six serial sections from the same biopsy at three different depths for each biopsy and in sections from three to five different biopsies. The final figure was the mean of all the measurements obtained for each patient. For all parameters, the coefficients of variation between biopsies from the same patient and between areas of the same biopsy were of 611% and 47%, respectively. Part of the patients (7 out of 10 control subjects and 27 out of 40 patients with asthma) has been included in a previous study, and SBM thickness and Ki67 immunostaining were assessed in their bronchial biopsies (19).
Statistical Analysis
Assessment of Airway Inflammation and Remodeling When compared with control subjects (Figure 1A) , bronchial biopsies from patients with intermittent asthma showed large zones of epithelial damage and disruption with, in some cases, a complete loss of columnar cells and the SBM covered only by basal cells (Figure 1B). In addition, the SBM was thickened, and very few ASM bundles were seen (Figures 1A and 1B).
Quantification analysis showed higher numbers of eosinophils (p < 0.001) and neutrophils (p = 0.003), lower epithelial integrity (p = 0.005), and thickened SBM (p < 0.001) in patients with intermittent asthma as compared with control subjects (Table 2) . No significant changes in fibroblast numbers, in collagen type III deposition, and in the surface of the bronchial submucosa occupied by mucous glands and by the ASM were noted (Table 2). A moderate, but not significant, lower distance between the outer limit of the SBM and the ASM was observed in these patients (Table 2).
Patients with mild-to-moderate persistent asthma had lower eosinophil infiltration in the bronchial submucosa (p = 0.024) as compared with patients with intermittent asthma (Table 2). This value, however, remained significantly elevated, as compared with control subjects (p = 0.024; Table 2). In addition, patients with mild-to-moderate persistent asthma displayed similar neutrophil numbers, epithelial integrity, SBM thickness, collagen type III deposition, mucous gland area, and SBM-ASM distance than patients with intermittent asthma (Table 2), but they had higher fibroblast numbers (p = 0.036) and ASM areas (p = 0.003; Table 2). Patients with severe persistent asthma showed large zones of partial or complete epithelial restitution (Figure 1C), which was translated morphometrically into a higher epithelial integrity, as compared with patients with mild to moderate asthma (p = 0.010; Table 2). In these patients, eosinophil and neutrophil numbers, as well as SBM thickness, were similar to those of patients with mild-to-moderate asthma (Table 2). In contrast, higher numbers of fibroblasts and augmented collagen type III deposition in the stroma underlying the airway epithelium were seen (p < 0.001 and p = 0.016, respectively, as compared with patients with mild-to-moderate persistent asthma; Table 2). Finally, patients with severe persistent asthma displayed larger mucous gland and ASM areas than patients with mild-to-moderate persistent asthma (p = 0.040 and p = 0.001, respectively; Table 2 and Figures 1C and 1D), and their ASM was much closer to the airway epithelium than that of the other patient groups (p < 0.050; Table 2). Bronchial biopsies from COPD patients revealed epithelial metaplasia (Figure 1E) and higher numbers of neutrophils in the bronchial submucosa in relationship to control subjects (p < 0.001; Table 2). In contrast, these patients had lower SBM thickness and fibroblast numbers but similar eosinophil counts, epithelial integrity, collagen type III deposit, mucous gland and ASM area, and SBM-ASM distance as control subjects (Figure 1E and Table 2). Except for epithelial integrity and SBM-ASM distance, all of these values were significantly lower than those measured in patients with severe persistent asthma (p < 0.012; Table 2). Finally, to exclude potential artifacts related to the assessment of ASM and mucous gland surface, we evaluated morphometrically total biopsy areas in the five patient groups. Computer-assisted image analysis showed similar areas of bronchial tissue sections in control subjects (median, 0.45 mm2; range, 0.370.61), in patients with intermittent asthma (median, 0.48 mm2; range, 0.430.56), in patients with mild-to-moderate asthma (median, 0.55 mm2; range, 0.450.72), in patients with severe persistent asthma (median, 0.55 mm2; range, 0.480.65), and in patients with COPD (median, 0.52 mm2; range, 0.461.02).
ASM Cell Hypertrophy and Proliferation
To determine whether a confounding effect of extracellular matrix deposit in and around the ASM bundles would interfere with the assessment of cell size, this measurement was repeated in bronchial tissue sections previously immunostained with an anticollagen type III antibody. ASM cell size was comparable in Mayer's hematoxylin-stained and in collagen type III-stained preparations. In patients with mild-to-moderate persistent asthma, for instance, median values of ASM cell size were of 5.1 µm (range, 4.55.6) in collagen type III-stained tissue sections, whereas in patients with severe asthma, they were of 9.8 µm (range, 8.510.6). Furthermore, the overall intensity of collagen type III expression in the ASM was weak and did not vary significantly across the three groups of patients with asthma (median score of intensity ranging between 0.25 and 0.50). Finally, ASM cell proliferation was assessed by the immunodetection of the nuclear antigen, Ki67 (24). As previously reported (19), Ki67-positive cells were distributed to different degrees along the airway epithelium and the bronchial submucosa of control subjects, patients with asthma and patients with COPD, but ASM was consistently negative, irrespective of the patient group (Figure 1F, data not shown).
Contractile Protein Expression
Quantification analysis established that the expression of -actin and of the smooth muscle myosin heavy-chain isoforms, SM1 and SM2, was similar in control subjects, in patients with asthma from all groups, and in patients with COPD (Table 3). This contrasted with MLCK expression, which was significantly higher in patients with intermittent asthma (p = 0.002) and in patients with mild-to-moderate persistent asthma (p = 0.008) and in patients with COPD (p = 0.003) in relationship to control subjects (Figures 3A, 3B, and 3D, and Table 3). The intensity of MLCK expression was further amplified in patients with severe persistent asthma (p < 0.005, as compared with the other patient groups; Figure 3C and Table 3).
To determine whether augmented MLCK expression was accompanied by a greater MLCK activity, the immunohistochemical determination of the p-MLC was assessed. p-MLC was almost undetectable in ASM from control individuals and from COPD patients and became detectable in patients with asthma from all groups (Table 3). However, the intensity of the immunostaining was very weak, and no significant differences were noted when the asthma groups were compared with each other (Table 3). Serial tissue sections demonstrated strong positivity for SM1 and
Correlation Analyses
In this study, we showed that ASM hypertrophy and fibroblast accumulation, but neither epithelial damage nor granulocyte infiltration into the bronchial submucosa, or SBM thickness, are selectively associated with severe persistent asthma. The assessment of airway epithelium integrity by means of bronchial biopsies is confounded by cell desquamation, which may be a consequence of the disease, or an artifact secondary to tissue sampling (25). Using a previously validated morphometric technique (23), we found complete epithelium restoration in severe persistent, as compared with patients with milder asthma. This phenomenon was associated with reduced thickness of the SBM and lower numbers of eosinophils and neutrophils in the bronchial submucosa. In a subanalysis to account for drug treatment, we established that these phenomena were related to steroid use (data not shown), indicating that some components of airway inflammation and remodeling may be controlled by steroid treatment, even in patients with severe asthma. This hypothesis was further supported by univariate regression analysis, showing that individual values of prebronchodilator FEV1 positively correlated with eosinophil numbers and SBM thickness. These findings agree with previous studies, including ours, showing epithelial restoration (26), lower SBM thickness (19), and similar or reduced airway eosinophilia or neutrophilia in patients with severe steroid-dependent asthma (2730). Nevertheless, these data contradict other observations demonstrating persistent eosinophilia and increased numbers of neutrophils in biopsy, lavage, and induced sputum specimens of patients with severe asthma (27, 28, 3136). These conflicting results may originate from the wide range of disease severity of the patients included in those studies and from the fact that most of the patients with severe asthma had respiratory function, symptoms, and medication use comparable to patients with mild-to-moderate asthma in this investigation. A current belief relates subepithelial fibrosis, particularly SBM thickness, fibroblast and myofibroblast accumulation in the bronchial submucosa, transforming-growth factor-ß1 expression and extracellular matrix protein deposition, to asthma severity (3743). However, one report concerning patients with asthma as severe as ours in terms of symptoms, FEV1, and treatments demonstrated that collagen type I and III deposition, SBM thickness, and transforming-growth factor-ß1 expression would not predict the clinical severity of the disease (29). In contrast to these observations, we showed a slightly but significantly higher amount of collagen type III and increased numbers of fibroblasts in patients with severe persistent asthma. Limitations related to biopsy sampling and processing, fixation, and number of measurements taken may explain the disparate results concerning collagen III deposition (29). In contrast, data on fibroblast accumulation are not necessarily conflicting with the absence of correlation between transforming-growth factor-ß1 expression and asthma severity (29), as previous studies have identified eosinophils, and not fibroblasts, as the major cell source of this growth factor in mucosal bronchial biopsies from patients with asthma (41, 43). In addition, factors other than transforming-growth factor-ß1, including basic fibroblast-growth factor, platelet-derived growth factor, insulin-growth factor, endothelin-1, nitric oxide, and the T helper 2derived cytokines, interleukin-4 and interleukin-13, contribute to fibroblast proliferation and activation (44). This suggests that transforming-growth factor-ß1 expression in proximal airways may not necessarily represent the most valuable marker of subepithelial fibrosis, particularly in patients with severe asthma. The prominence of the submucosal glands is usually not mentioned in bronchial biopsy studies of mild and moderate asthma, but augmented submucosal gland areas have been reported in fatal cases of asthma (45). In this study, mucous glands represent a higher proportion of the submucosa in patients with severe persistent asthma, a phenomenon that may contribute to excessive mucous production and to airway narrowing in these patients. Univariate regression analysis showed a significant negative correlation between fibroblast numbers, collagen type III deposition, and mucous gland area and the extent of airflow limitation, as expressed by both prebronchodilator and postbronchodilator FEV1. Together, our findings suggest that an excessive fibrogenic process and mucous gland enlargement reflect asthma severity and occur despite high-dose, long-term steroid therapy. Of note, fibroblast accumulation was associated with both prebronchodilator and postbronchodilator FEV1 in the stepwise multiple regression analysis, indicating that this phenomenon represents a more reliable hallmark of severe asthma. Quantification of ASM in endobronchial biopsy samples is considered problematic mainly because of the impossibility to assess internal/external diameter of entire airways. Artifacts such as the presence of mechanical stresses imposed during bronchoscopy and the orientation of the ASM may also influence its measurement. In this study, we attempted at quantifying ASM mass both directly by determining the proportion of the bronchial submucosa occupied by the ASM and, indirectly, by measuring at regular intervals the distance between the outer limit of the SBM and the ASM. Higher ASM area and lower SBM-ASM distance were found in patients with severe persistent asthma, and they were correlated to both prebronchodilator and postbronchodilator FEV1, suggesting that ASM mass was increased in the peripheral airways in relationship to the degree of airflow obstruction.
A number of mechanisms may explain an augmented ASM content in asthma. For example, Thomson and Schellenberg (46) hypothesized that the presence of collagen deposition in and around ASM bundles may contribute to the overall increase in the ASM content. However, we excluded this possibility by showing weak collagen type III deposition in the ASM and no difference between the patient groups. A novel mechanism that may also explain the increase in ASM content in asthma involves the participation of fibromyocytes (12). These Both hyperplastic and hypertrophic changes contribute to ASM increase in severe asthma, as previously ascertained on autopsy specimens (49, 50), or using sophisticated three-dimensional morphometry (51, 52). In vitro ASM cell hyperplasia has been associated with an increase proliferation rate (53, 54). Using specific immunostaining for Ki67, a nuclear antigen expressed by proliferating cells (24), we showed a lack of ASM cell replication, irrespective of the patient group. These data are consistent with previous observations showing the absence of markers of proliferation in smooth muscle areas of restenosed coronary arteries, a pathologic condition characterized by tissue remodeling and fibrosis (55). In a recent study, however, Johnson and colleagues (56) reported increased proliferation rate in cultured ASM cells isolated from explanted lungs or from endobronchial biopsy samples of patients with asthma, as compared with control subjects. This discrepancy between these observations and our findings may arise from the striking differences existing between the in vivo and in vitro ASM cell environment. For example, it has been recently suggested that the interactions of ASM cells with matrix components present in the asthmatic airways, but not in in vitro cell culture-based systems, may result in a downregulation of different cell functions, including proliferation (57, 58). Nevertheless, the fact that no ASM cell proliferation was detected in our tissue specimens does not exclude that hyperplasia may have been induced over a long period of time before biopsy sampling, as previously demonstrated (4952). Measurement of individual smooth muscle cell size has already been validated to document cell hypertrophy in human tissue specimens (53). Here we showed that ASM cell size was augmented by 1.5-fold in patients with intermittent and with mild-to-moderate asthma, in relationship to control subjects and by threefold in patients with severe persistent disease. In addition, ASM cell size correlated negatively with prebronchodilator and postbronchodilator FEV1 values in univariate and, most importantly, in stepwise multiple regression analysis. These results suggest that ASM hypertrophy discriminates patients with severe persistent asthma from patients with milder disease and underline its major physiopathologic impact as a marker of disease activity. To exclude a confounding effect of changes in cell shape secondary to extracellular matrix deposition in and around ASM bundles, we compared ASM size in Mayer's hematoxylin- and in collagen type III-stained bronchial tissue sections, with very similar results. This suggests that augmented ASM cell size observed in patients with severe persistent asthma reflects true cell hypertrophy and not merely the presence of collagen deposit within ASM cells.
In vitro ASM hypertrophy is generally associated with increased amounts of proteins involved in the contractile machinery, including Finally, in an attempt to establish whether bronchial tissue abnormalities seen in patients with severe persistent asthma were specific of the asthma process, a group of patients with COPD, another chronic inflammatory lung disorder characterized by severe airflow limitation, was examined. As previously demonstrated (12, 14), bronchial biopsies from patients with COPD showed epithelial metaplasia and augmented neutrophil but not eosinophil numbers in their bronchial mucosa. Neither fibroblast accumulation nor mucous gland enlargement, thickening of the SBM, or changes in ASM content were noted in these patients. These findings corroborate previous observations showing that architectural alterations in COPD, when present, involve peripheral rather than proximal airways (8, 12, 62). Of note, ASM cell size and MLCK expression in COPD patients were significantly lower than those observed in patients with severe persistent asthma, supporting the hypothesis that ASM alterations seen in severe asthma do not extend to COPD. In conclusion, we provide evidence that airway structural abnormalities, but not features of mucosal inflammation, allow the discrimination of patients with severe persistent asthma from patients with milder disease and with COPD. Furthermore, this study is the first to document that fibroblast accumulation and ASM hypertrophy in proximal airways are the most reliable hallmarks of asthma severity. We believe that these findings have considerable clinical consequences, as they may help to identify predictive biomarkers for monitoring asthma progression and severity and, ultimately, for delivering more effective treatments.
The authors are indebted to patients who participated in the study, to the Service d'Explorations Fonctionnelles of the Hôpital Bichat, to the nurses of the Service de Pneumologie, to Mrs. Isabelle Poirier and Gabrielle Beuve for their expert technical assistance during fiberoptic bronchoscopies, to M. David Soussan (INSERM Unité 408) for help in performing statistical analyses, and to Mrs. Paule Loiseau (Institut Fédératif de Recherche 02, Faculté de Médecine Xavier Bichat) for assistance in initial morphometric analyses. The supporting sources had no involvement in the study design, the collection, the analysis and interpretation of the data, the writing of the report, and the decision to submit it for publication.
Supported by the Fonds de Recherche Hoechst Marion Roussel (grant no. 99,178) and by the Association Fédérative Nationale pour le Traitement à Domicile de l'Insuffisance Respiratoire," Paris (grant no. D0272 to L.B.). This article has an online supplement, which is accessible from this issue's table of contents online at www.atsjournals.org Received in original form September 11, 2002; accepted in final form January 13, 2002
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||