| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Cardiovascular Research Institute, and 2 Department of Medicine, University of California, San Francisco, California
Correspondence and requests for reprints should be addressed to John V. Fahy, M.D., M.Sc., University of California at San Francisco, Box 0130, 505 Parnassus Avenue, San Francisco, CA 94143. E-mail: john.fahy{at}ucsf.edu
ABSTRACT
Cellular inflammation of the airways with eosinophils and neutrophils is a characteristic feature of asthma and is considered relevant to the pathogenesis of the disease. Studies of large numbers of subjects with well-characterized asthma in recent years has resulted in new insights about the clinical and pathologic correlates of eosinophilic and neutrophilic inflammation in asthma. For example, eosinophilic asthma is a distinct phenotype of asthma that is associated pathologically by thickening of the basement membrane zone and pharmacologically by corticosteroid responsiveness. In contrast, noneosinophilic asthma, a sizeable subgroup of asthma that includes patients with severe disease, is not characterized by thickening of the basement membrane zone, and it appears to be relatively corticosteroid resistant. Eosinophilic and neutrophilic asthma are not mutually exclusive subtypes of asthma. Rather, neutrophils accumulate in the airways in patients with asthma with more severe airflow obstruction, where eosinophils may also be present in excess. In addition, neutrophils are prominent in airway secretions during acute severe asthma exacerbations, where it is possible that they have roles in both the initiation and resolution of attacks. These insights about the relationships between cellular inflammation and disease phenotypes of asthma support the concept that different subgroups of patients with asthma, despite clinically similar features, can be defined by specific cellular and molecular markers. The promise now is that these markers will ultimately guide personalized treatment programs.
Key Words: eosinophils • neutrophils • asthma
The link between asthma and airway inflammation is long established, but important advances have occurred in the past 25 years. For example, in 1986, it was shown that ozone-induced hyperresponsiveness to methacholine in healthy subjects without asthma is associated with airway neutrophilia and with changes in lung levels of cyclooxygenase metabolites of arachidonic acid (1). These findings provided more direct evidence for a link between airway inflammation and abnormal airway physiology, a link that was further suggested by the efficacy of corticosteroids in reducing airway inflammation while improving airway function in individuals with asthma (2). In addition, the late 1980s and first part of the 1990s saw the first large studies of markers of airway inflammation and airway remodeling in bronchial lavage and bronchial biopsies from subjects with mild and moderate asthma (3). These studies demonstrated that airway inflammation and airway remodeling are characteristics of asthma, even when it is clinically mild in severity. Coupled with these findings was increased awareness of the role of CD4+ T cells in allergic and parasitic diseases and new knowledge about Th1 and Th2 subsets of CD4+T cells (4). These insights prompted hypothesis-driven studies which showed that T cells in the airways of individuals with asthma had cytokine profiles characteristic of Th2 cells (5). With the discovery in the late 1990s of IL-13 as a key effector cytokine secreted by Th2 cells, and a potent inducer of an asthma phenotype in mouse models of asthma (6, 7), consensus began to emerge that asthma represents an adaptive allergic response to aeroallergens in which CD4+T cells of the Th2 subtype are key upstream orchestrators of typical asthmatic inflammation and pathology. Coincident with this consensus, however, were additional clinical data which showed that the inflammation found in many patients with asthma did not fit a typical Th2 profile, as well as a growing awareness of the importance of innate immune responses in the pathogenesis of airway inflammation in asthma (8). Among the best-studied innate responses in the lung are epithelial cell responses to aeroallergens and viruses. These studies clearly show that epithelial cells can respond to environmental stimuli in ways that affect the differentiation of dendritic cells and CD4+T cells and the accumulation of mast cells and eosinophils. Therefore, investigators studying airway inflammation in asthma are now considering the relative roles of innate and adaptive responses in the pathogenesis of inflammation and remodeling.
As larger groups of subjects with asthma have undergone detailed characterization studies in the past 10 years, including characterization studies of inflammation in airway secretions and tissues, the concept has evolved that asthma is a heterogeneous disease with several distinct pathologic phenotypes, each representing different pathophysiologic pathways that can ultimately lead to the clinical expression of asthma. This concept has placed the emphasis now on ensuring that all of the clinical and molecular phenotypes of asthma are comprehensively understood, because patients with asthma will then have the best chance of receiving treatments to which they will respond. Here I will examine cellular inflammatory phenotypes in asthma with a specific emphasis on the clinical correlates of eosinophilic and neutrophilic inflammation.
EOSINOPHILS IN ASTHMA
Increased numbers of eosinophils in peripheral blood and in airway secretions are a characteristic feature of asthma. Bousquet and coworkers (9) have shown that the number of eosinophils in peripheral blood and in bronchial lavage from subjects with asthma is associated with more severe disease. This association between eosinophilia and outcomes of asthma severity has been confirmed and extended in several other studies. For example, Louis and colleagues (10) showed that eosinophil numbers in induced sputum are highest in patients with more severe asthma. In addition, in work from my own group that included analyses of induced sputum and lung function from over 200 subjects with asthma, we found that the eosinophil percentage (which averaged 6% in this cohort) was independently associated with more severe airflow obstruction and heightened airway reactivity to methacholine (11).
Although patients with more severe asthma differ from patients with mild asthma in having more eosinophils in their peripheral blood and in their airways, Wenzel and coworkers (12) have pointed out that not all patients with severe asthma have airway eosinophilia. In fact, by examining eosinophil numbers in airway secretions and tissues from patients with severe asthma who clinically had a similar disease phenotype, these authors have been able to identify distinct subgroups of patients with and without eosinophilia. Interestingly, the numbers of tissue lymphocytes (CD3+, CD4+, CD8+), mast cells, and macrophages were higher in the subgroup of subjects with severe asthma with eosinophilia than in the subgroup without eosinophilia. In addition, the subepithelial basement membrane zone was thicker in the subgroup with eosinophilia than without. This noneosinophilic phenotype of asthma is not restricted to patients with severe disease. Berry and colleagues (13) have studied subjects with mild and moderate asthma stratified by the presence or absence of eosinophils in their induced sputum. They found that, compared with subjects with asthma with minimal airway eosinophilia, subjects with asthma with eosinophilia had increased thickness of the subepithelial basement membrane zone and a good short-term response to treatment with inhaled corticosteroids. Notably, the thickness of the subepithelial basement membrane zone in the noneosinophilic subgroup was similar to the thickness in the healthy control group. A thickened basement membrane zone therefore appears to occur only in individuals with asthma with eosinophilia, a link that is described in other airway diseases as well (14, 15). Together, these studies demonstrate that eosinophilic and noneosinophilic asthma represent distinct clinical and pathologic phenotypes. The importance of this finding is that these distinct subgroups may need to be treated differently. Indeed, the identification of eosinophilic airway inflammation has already been used to personalize antiinflammatory treatment (16, 17).
Th2 cytokines such as IL-4, IL-5, and IL-13 induce airway eosinophilia in animal models of asthma, and these cytokines are considered mediators of eosinophilia in subjects with asthma. T cells are an important source of Th2 cytokines, but there has been recent controversy about the T cell subtype responsible for Th2 cytokine secretion in asthma (18, 19). Specifically, the issue is whether the predominant T cell subtype in the asthmatic airway is the CD4+ T cell, which is class II MHC-restricted, or the CD4+ natural killer T cell expressing the invariant T cell receptor. The weight of current evidence favors the former. Conceptually, it is thought that individuals with atopic asthma, by far the largest subgroup of individuals with asthma, accumulate eosinophils in their airways through the sustained action of Th2 cytokines secreted by the Th2 subtype of CD4+ T cells. Notably, when our group recently examined the gene expression profile of airway epithelial brushings from subjects with asthma using a high-density microarray that had probe sets for all genes in the human genome, we found the most highly up-regulated genes were ones whose expression was up-regulated by IL-13 in cultured epithelial cells in vitro (20). Thus, using an unbiased experimental approach, we found evidence that IL-13 is an important activator of epithelial cells in asthma. Other studies that have measured IL-13 directly at the gene expression or protein level in subjects with asthma have found increased levels (21, 22). Analysis of data from these studies, however, clearly shows considerable variability in IL-13 levels among subjects with asthma, with a significant subgroup of subjects having levels in the normal range. Thus, some individuals with asthma have evidence of IL-13 up-regulation, and some do not. These data fit well with the idea that there are distinct molecular phenotypes of asthma, and the challenge now is to link specific molecular signatures with inflammatory, pathologic, and clinical phenotypes. Meeting this challenge will enable more targeted treatments for patients with asthma.
NEUTROPHILS IN ASTHMA
Neutrophils are the most abundant cell type in induced sputum from both healthy subjects and from subjects with asthma (Figure 1). Neutrophil numbers are not increased in airway secretions from patients with mild and moderate asthma, but Wenzel and coworkers (23) have shown that neutrophil numbers are higher than normal in airway lavage from patients with severe asthma. Interpretation of this finding has been confounded by the fact that patients with severe asthma take higher doses of inhaled or oral corticosteroids. However, in a study of 205 subjects with asthma in which we controlled for corticosteroid use in multivariate linear regression models, we found that neutrophil numbers in induced sputum from subjects with asthma across a wide severity range are associated with more severe airflow obstruction (11). In contrast, we found that neutrophil numbers are not associated with more severe airway reactivity to methacholine. This relationship between airway neutrophilia and airflow obstruction in asthma has been confirmed and extended recently in a study by Shaw and colleagues (24) in which sputum cell counts and lung function (FEV1 pre- and post-bronchodilator) were examined in over 1,100 patients with asthma. These authors found that both high neutrophil numbers and high eosinophil numbers are associated with a low prebronchodilator FEV1, but that only high neutrophil numbers are associated with a low post-bronchodilator FEV1. Interestingly, the authors calculated that a 10-fold increase in neutrophil count is associated with a 92-ml reduction in post-bronchodilator FEV1. Together, these studies show that airway neutrophilia is a characteristic of more severe asthma and suggest a possible mechanistic link between airway neutrophils and chronic airway narrowing in asthma.
|
Apart from the findings for airway neutrophils in chronic severe asthma, excess neutrophils have been found in airway secretions from individuals with asthma who are experiencing an acute exacerbation. For example, we have found that patients with asthma presenting to the emergency room with acute exacerbation have prominent neutrophilia in their sputum and high concentrations of neutrophil elastase (25). Many of these subjects gave a history of an upper respiratory tract infection as a precipitant for their asthma exacerbation, suggesting viral airway infection as an initiating mechanism for acute airway neutrophilia. Subsequently, Lamblin and coworkers (26) found high neutrophil numbers and high IL-8 levels in bronchial lavage from patients intubated for management of status asthmaticus. We have also analyzed airway secretions from patients with asthma intubated for management of acute asthma, but we did so by collecting tracheal aspirates from these patients, a strategy that allowed measurement of airway inflammation at multiple time points during intubation. We found that the number of neutrophils in tracheal aspirates collected from patients with asthma at the time of intubation was 10 times higher than normal (27). Although eosinophil numbers were also significantly higher than normal, the numbers of eosinophils were eightfold less than neutrophils. Surprisingly, the number of neutrophils increased significantly during the period of intubation in the subjects with asthma, probably because all patients had received intravenous corticosteroid treatment. Corticosteroids are known to prolong neutrophil survival by preventing apoptosis (28). Together, these studies in acute severe asthma show that neutrophils are usually the dominant inflammatory leukocyte accumulating in the airway. Although these findings raise questions about the role of these cells in the pathogenesis of acute and near fatal asthma exacerbations, the fact that neutrophilia in the airway persists and is even more prominent at the time of asthma recovery has prompted us to consider the possibility that neutrophils may be important for the resolution of asthma exacerbations. Specifically, we have hypothesized that neutrophil proteases are important for the digestion of airway mucus plugs, which occlude airways in acute asthma (29).
CONCLUSIONS
Eosinophilic asthma has emerged as a distinct phenotype of asthma that is associated pathologically by thickening of the basement membrane zone and pharmacologically by corticosteroid responsiveness. Noneosinophilic asthma is a sizeable subgroup of asthma, which includes patients with severe disease, and it may be relatively corticosteroid resistant. The mechanisms underlying noneosinophilic asthma remain poorly understood, but the identification of this important subgroup in recent years at least highlights the need for more intensive study of noneosinophilic disease mechanisms of asthma. Partitioning of asthma by markers of eosinophilia is hopefully only the initial step along a path that will ultimately result in our ability to identify specific subgroups of patients with asthma by defined cellular and molecular markers that will guide specific treatment approaches.
Neutrophils accumulate in the airway in more severe forms of chronic severe asthma, and neutrophil numbers are associated with chronic airway narrowing. In addition, neutrophils are prominent during acute severe asthma exacerbations, where it is possible that they have roles in both the initiation and resolution of attacks. Current understanding of the mechanisms of neutrophilia in acute and chronic asthma, and of the clinical consequences of decreasing airway neutrophils in asthma, is very limited. More needs to be known about the role of neutrophils in specific contexts in asthma, so that specific rationales can be developed for clinical trial designs that target neutrophils.
ACKNOWLEDGMENTS
The author acknowledges Prescott Woodruff, M.D., M.P.H (University of California, San Francisco), and Joe Arron M.D., Ph.D. (Genentech, Inc.), for helpful discussions about asthma phenotypes over the past 12 months. These discussions have informed many of the concepts outlined in this article.
FOOTNOTES
Conflict of Interest Statement: J.V.F. between 2006 and 2009 has served as a consultant to Aerovance, Arriva Pharmaceuticals, Biogen, Gilead, Roche, and Cytokinetics. In 2008 and 2009 he received research grants from Genentech for 
$450,000, from Boehringer Ingelheim for $100,000, and from Roche for $90,000 for clinical research related to preclinical and early phase drug discovery in asthma and COPD.
(Received in original form August 19, 2008; accepted in final form January 22, 2009)
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
