Published ahead of print on June 22, 2006, doi:10.1165/rcmb.2006-0135OC
© 2006 American Thoracic Society DOI: 10.1165/rcmb.2006-0135OC Airway Hyperreactivity in Exacerbation of Chronic Asthma Is Independent of Eosinophilic InflammationDepartment of Pathology, University of New South Wales, Sydney; School of Biomedical Sciences, University of Newcastle, Newcastle; and Division of Molecular Biosciences, John Curtin School of Medical Research, Australian National University, Canberra, Australia Correspondence and requests for reprints should be addressed to R. K. Kumar, Department of Pathology, University of New South Wales, Sydney, Australia 2052. E-mail: R.Kumar{at}unsw.edu.au
We have developed an animal model to investigate the mechanisms underlying an acute exacerbation of chronic asthma. Sensitized BALB/c mice were exposed to aerosolized ovalbumin, either as chronic low-level challenge (mass concentration 3 mg/m3) for 4 wk, a single moderate-level challenge ( 30 mg/m3), or chronic low-level followed by single moderate-level challenge (the acute exacerbation group). Compared with animals receiving chronic challenge alone, mice in the acute exacerbation group exhibited a more marked inflammatory response, with involvement of intrapulmonary airways and lung parenchyma, and increased numbers of lymphocytes and eosinophils in bronchoalveolar lavage fluid. They also developed airway hyperreactivity (AHR) to methacholine, demonstrable as increased transpulmonary resistance and decreased compliance. This pattern of AHR was absent in chronically challenged animals, but was also present in animals given single moderate-level challenge. However, compared with animals receiving a single moderate-level challenge, inflammation and AHR were induced more rapidly in the acute exacerbation group. Eosinophil-deficient GATA1 dbl mice exhibited undiminished AHR in the acute exacerbation model. We conclude that in mice with pre-existing airway lesions resembling mild chronic asthma, exposure to a moderately high concentration of inhaled antigen induces features of an acute exacerbation. The inflammatory response involves distal airways and is associated with a distinct pattern of AHR, which develops independent of the enhanced eosinophilic inflammation.
Key Words: airway inflammation bronchial hyperreactivity eosinophils severe asthma small airways disease
Acute exacerbations of asthma are a common clinical problem with major economic impact (1). Patients typically present with a variety of manifestations of worsening airflow obstruction and its consequences, which may be difficult to manage and can be life-threatening (1, 2). In any given year, over 10% of children with asthma develop at least one severe episode, often requiring attendance at a hospital emergency department (3). In adults, exacerbations are more common in those with severe, difficult-to-treat asthma (4). Viral infection and allergen exposure appear to be the most important triggers (1, 5). Viral infections characteristically elicit exacerbations of slow onset, whereas allergen exposure may be more likely to trigger exacerbations of sudden onset and greater severity (6, 7). Compared with stable asthma, an acute exacerbation is associated with exaggerated airway inflammation, including recruitment of increased numbers of eosinophils as well as of neutrophils (8), and more extensive involvement of smaller distal airways (9, 10). In parallel, there is increased airway resistance, to which distal airway lesions may contribute significantly (11). Various pathogenetic mechanisms have been invoked to explain the airflow obstruction, including exaggerated bronchoconstriction, airway wall edema, luminal obstruction as a consequence of mucus hypersecretion, and premature airway closure (2, 12). Further investigation of the pathophysiology of acute exacerbations is limited by the lack of a suitable animal model. Existing short-term models of allergic bronchopulmonary inflammation in mice do not mimic acute exacerbations, because the background lesions of chronic asthma are absent and the severity of inflammation, especially parenchymal, is far in excess of that in asthma (13). We have described a model of chronic asthma in mice, which are systemically sensitized to ovalbumin and subjected to inhalational challenge with controlled low levels of aerosolized antigen (14). After four or more weeks of exposure, this elicits airway lesions typical of mild human asthma, including intra-epithelial eosinophil accumulation, chronic inflammation in the lamina propria, mucous cell hyperplasia/metaplasia, and subepithelial fibrosis. Unlike short-term high-level challenge models, there is minimal parenchymal inflammation and the mice exhibit hyperreactivity to an inhaled cholinergic agonist that is demonstrably of airway rather than parenchymal origin (15). In the present study we have established a model in which mice previously challenged with low mass concentrations of antigen, which had developed changes of mild chronic asthma, exhibited airway inflammation resembling an acute exacerbation after they were exposed to a moderately high level of inhaled antigen. The animals also demonstrated a distinct pattern of airway hyperreactivity (AHR), which appeared to be related to the distal inflammatory lesions but was independent of the enhanced eosinophilic inflammation. This model provides a novel and appropriate experimental system in which to dissect pathogenetic mechanisms in an acute exacerbation of chronic asthma, and their relationship to the severity and progression of underlying chronic disease.
Experimental Model The protocols we employed for sensitization and inhalational challenge of mice have previously been described in detail (14, 16). Briefly, specific pathogenfree female BALB/c mice or eosinophil-deficient GATA1 dbl mice on a BALB/c background (17) (aged 8 wk at the commencement of experimental studies) received intraperitoneal injections of 50 µg of alum-precipitated chicken egg ovalbumin (Sigma Australia, Sydney, Australia) 21 and 7 d before inhalational exposure. They were maintained in a laminar flow holding unit (Gelman Sciences, Sydney, Australia) and housed in autoclaved cages on autoclaved bedding in an air-conditioned room on a 12 h light/dark cycle. Irradiated food and acidified water were provided ad libitum throughout. Mice were exposed to aerosolized ovalbumin in a whole-body inhalation exposure chamber (Unifab Corporation, Kalamazoo, MI). During the exposure, the animals were held in wire flow-through cage racks and filtered air was drawn through the 0.5-m3 inhalation chamber at a flow rate of 250 liters/min. A solution of 2.5% ovalbumin in normal saline was aerosolized by delivery of compressed air to a sidestream jet nebulizer (Trimed, Sydney, Australia) and injected into the airstream entering the chamber. Mice that received chronic low-level challenge were exposed to 3 mg/m3 aerosolized ovalbumin for 30 min/d on 3 d/wk for 4 wk. Moderate-level challenge consisted of a single exposure to 30 mg/m3 of ovalbumin for 30 min, while sham-challenged control animals were exposed to 30 mg/m3 of saline aerosol for the same period. Particle concentration within the breathing zone of the mice was continuously monitored using a DustTrak 8520 instrument (TSI, St. Paul, MN). Comparisons were made between experimental groups that either received chronic challenge alone, a single moderate-level challenge, or chronic challenge followed by single moderate-level challenge, which was designated the acute exacerbation group. Control groups included sham-challenged and naïve animals. Each group comprised six to eight animals, and replicate experimental groups were used for assessment of the inflammatory response and airway reactivity. All experimental procedures complied with the requirements of the Animal Care and Ethics Committee of the University of New South Wales (reference 04/06).
Bronchoalveolar Lavage and Histopathology The longitudinally orientated trachea and a horizontal slice from the mid-zone of the single-lobed left lung were embedded in paraffin. Eosinophils within the airway epithelial layer were quantified in hematoxylin and eosinstained sections of the trachea, as previously described (14). Peribronchiolar and perivascular inflammation in the lung parenchyma were semiquantitatively graded on a scale from 03, where no inflammation = 0, mild inflammation = 1, moderate inflammation = 2, and marked inflammation = 3. All counting/grading was performed by a single observer blinded to the identity of the samples, and slides were examined in a random order.
Enzyme Immunoassays
Airway Reactivity
Statistical Analysis
Inflammatory and Cytokine Response In the chronic challenge group of mice, there was a modest increase in the total number of cells recovered by lavage, with a significant increase in the proportion of lymphocytes (Table 1). In the acute exacerbation group, there was a larger increase in total cell numbers, which were significantly greater than in sham-exposed controls. This included numerous lymphocytes, which were evident as early as 4 h after the final moderate-level challenge, together with increased percentages of eosinophils and neutrophils, although the latter did not reach statistical significance (Table 1). Single moderate-level challenge elicited little or no inflammation at 4 h (not shown), but induced limited recruitment of lymphocytes and a significantly increased proportion of neutrophils at 18 h (Table 1).
In the major airways, intraepithelial accumulation of eosinophils was apparent in all antigen-challenged groups. This was maximal after 4 h in the acute exacerbation group, whereas it was only detected at 18 h in the single moderate-level challenge group (Table 1). In the lung, animals which had received prior chronic low-level exposure to aerosolized ovalbumin (i.e., both the chronic challenge and acute exacerbation groups) exhibited peribronchiolar lymphoid aggregates around the major intrapulmonary airways (Table 1). In the acute exacerbation group, accumulation of lymphocytes was also observed around distal intrapulmonary airways (Figure 1A), together with eosinophils and neutrophils, which were most noticeable in perivascular areas (Table 1, Figure 1B). These cells were absent in the chronic challenge group (Figure 1C). Perivascular eosinophils were particularly prominent in the single moderate-level challenge group (Table 1).
Low levels of IL-13 (129 ± 26 pg/ml) and IFN- (74 ± 16 pg/ml) were detected in BAL fluid from sham-challenged animals. Mean concentrations of these cytokines were not significantly increased in any of the antigen-challenged groups. Concentrations of IL-5 were below the detection limit of 10 pg/ml in all groups.
AHR
Effect of Eosinophil Deficiency In the acute exacerbation model, intraepithelial accumulation of eosinophils was absent in GATA1 dbl mice (0.3 ± 0.1 cells/mm, compared with wild-type mice [1.8 ± 0.2] and naïve mice [0.4 ± 0.1]). However, AHR was equivalent to that in wild-type mice, both in terms of increased RL and decreased Cdyn (Table 2).
In this study, we used our well-established chronic challenge model (14) as a substrate for development of a model of an acute exacerbation of allergic asthma. In the chronic challenge model, animals develop airway lesions within a period of 4 wk that are very similar to those observed in mild chronic human asthma. These include recruitment of eosinophils into the epithelial layer, chronic inflammation in the lamina propria, subepithelial fibrosis, and mucous cell hyperplasia/metaplasia. The changes are confined to the conducting airways, with no inflammation in the lung parenchyma, and relatively low numbers of inflammatory cells overall. When these mice were challenged on a single occasion with a moderately high level of aerosolized antigen, this elicited more marked inflammation, which involved distal airways to a greater extent and thus resembled an acute exacerbation of asthma. Furthermore, we found that the mice developed a pattern of AHR distinct from that of the chronic challenge model, apparently originating from the peripheral airways and lung.
With respect to the inflammatory response, the only changes observed in the lung tissue in the chronic low-level challenge model are peribronchiolar aggregates of chronic inflammatory cells (mostly lymphocytes) (15). However, the acute exacerbation model demonstrated recruitment of both eosinophils and neutrophils into the distal airways. Importantly, although the moderate-level challenge involves exposure to a greater mass concentration of aerosolized ovalbumin than in the chronic model, this is still at least 10-fold lower than in short-term experimental models used to study features of allergic bronchopulmonary inflammation. As pointed out in a recent editorial (13), whereas eosinophils typically account for only 13% of cells obtained by bronchoalveolar lavage in humans with asthma, they comprise 4080% of the recovered cells in short-term murine experimental models, which have severe perivascular inflammation. Our acute exacerbation model had a relatively low percentage of eosinophils in lavage fluid and much less parenchymal involvement. Presumably because the inflammation elicited by moderate-level challenge was still relatively mild, we were unable to detect increased levels of cytokines such as IL-5, IL-13, or IFN- The acute exacerbation model we have developed exhibited more rapid onset of recruitment of inflammatory cells than is seen in conventional models. Whereas the single moderate-level challenge elicited an inflammatory response that had a time course similar to that observed in most short-term models (i.e., with significant inflammation only apparent at 18 h after exposure to aerosolized antigen) airway inflammation was maximal by 4 h in the acute exacerbation group, both in terms of the cells recovered by lavage and the intra-epithelial accumulation of eosinophils, which were significantly increased compared with controls at 4 h but not at 18 h. This finding is consistent with our earlier report demonstrating that whereas a single high-level exposure triggered intra-epithelial accumulation of eosinophils from 6 h onwards, chronic challenge apparently primed the airways for more rapid recruitment of these cells, with large numbers of eosinophils recruited by 3 h after antigen exposure (20). Not only were eosinophils more rapidly recruited in the acute exacerbation group, but in addition there was greater and more rapid accumulation of lymphocytes in lavage fluid in these animals. Preliminary immunohistochemical studies (not shown) demonstrated the presence of substantially increased numbers of CD3-positive T cells in the lungs of the mice. These changes are consistent with evidence of increased numbers of activated T cells in lavage fluid and of recruitment of T cells to the small airways in patients with severe asthma (21). Moreover, increases in airway T cells have been shown to correlate with development of asthmatic exacerbations after glucocorticoid withdrawal (22). Phenotyping and functional assays of the lymphocytes recruited in the acute exacerbation model will therefore be of considerable interest. Animals in the acute exacerbation group exhibited a distinct pattern of AHR, which appeared to be related to the distal airway inflammatory lesions. In the chronic challenge group, in which AHR has been shown to primarily originate from the conducting airways using the forced oscillation technique (15), animals did not have significant AHR as assessed by an increase in RL or a decrease in Cdyn using whole body plethysmography. This finding probably reflects the observation that in BALB/c mice, over 70% of RL is accounted for by tissue resistance (23), while parenchymal tissue inflammation is minimal or absent in the chronic challenge model. In contrast, in the acute exacerbation model there was a significant increase in RL and a greater decrease in Cdyn in response to methacholine, consistent with the involvement of distal airways and lung tissue. Reflecting the significant inflammation in the lung tissue in the single moderate-level challenge group, these latter animals also had AHR in terms of RL and Cdyn. Thus it appeared that changes in RL and Cdyn were related to AHR originating from peripheral airways and lung and that, as we have reported in relation to the chronic model (15), the site of origin of AHR was indicative of the predominant site of inflammation. Correlating with the rapid recruitment of inflammatory cells noted above, induction of peripheral AHR in the acute exacerbation group was rapid, reaching a maximum at 4 h as compared with 24 h in the single moderate-level challenge group. Inflammation of the distal (peripheral) airways is well documented in humans with asthma, especially in fatal asthma (9, 24). This may be important in the development of clinical manifestations, because a significant component of airflow resistance in individuals with asthma originates from small airways, which appear to play a major role in hyperreactivity to methacholine and histamine (25, 26). Furthermore, peripheral airway obstruction is more marked in severe disease (27, 28) and may be associated with difficult-to-control asthma with more frequent development of exacerbations (11). The experimental model we have developed thus replicates characteristic features of an acute exacerbation of asthma, including distal airway inflammation and AHR originating from peripheral airways. Moreover, these lesions develop on an appropriate background of chronic inflammation and airway wall remodeling, induced as a result of the preceding chronic antigenic challenge. Compared with short-term models of allergic bronchopulmonary inflammation, this model involves realistic levels of exposure to antigen and realistic levels of inflammatory response. The speed with which the response is triggered resembles the rapidity of onset of allergen-induced exacerbations in the human disease.
In the context of the association between increased airway reactivity and the rapid recruitment and accumulation of eosinophils in the tissues, we investigated the role of eosinophils in the development of AHR in the acute exacerbation model, using mice unable to produce eosinophils due to a deletion of the double GATA-binding site in the GATA-1 promoter region (17). These animals have a selective deficiency of eosinophils without other significant phenotypic abnormalities. Unexpectedly, the GATA-1 Numerous investigators have argued in support of noncellular mechanisms involved in the development of AHR. These have not yet been defined, but impaired surfactant function, possibly resulting from interaction with proteins in inflammatory exudate, could lead to early and/or abnormal closure of small airways (30, 31). Early airway closure has been suggested to be a predisposing factor in humans with asthma who develop recurrent exacerbations (32). Recent studies in a short-term model of allergic airway inflammation in mice have related AHR to thickening of the airway mucosa and an increased propensity of the airways to close (33). These experiments have also provided direct evidence that fibrin accumulation may have an important role in the development of AHR (34). This new model offers significant opportunities for further study of the pathophysiology of acute exacerbations. In addition, the model may be valuable for identifying possible markers of disease severity. Few of the latter are currently recognized, although elevation of plasma levels of soluble ST2 protein (35) and increased serum arginase activity (36) have been suggested as useful candidates. In conclusion, we have shown that in mice chronically challenged with low mass concentrations of antigen, which have developed lesions resembling mild chronic asthma, exposure to a moderately high level of inhaled antigen induces features characteristic of an acute exacerbation of asthma. A rapid and exaggerated inflammatory response involving intrapulmonary airways and parenchyma is associated with AHR, apparently originating from the peripheral lung, which is in contrast to the predominantly central origin of AHR in the model of chronic asthma. This peripheral pattern of AHR appears to be independent of the recruitment of eosinophils.
The authors thank Dr. Alison Humbles and Dr. Craig Gerard for the GATA-1 dbl mice.
* These authors contributed equally to this work. This work was supported by NHMRC Australia. Originally Published in Press as DOI: 10.1165/rcmb.2006-0135OC on June 22, 2006
Conflict of Interest Statement: J.S.S. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. N.H. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. C.H. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. M.Y. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. P.S.F. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. R.K.K. received Received in original form April 5, 2006 Accepted in final form June 3, 2006
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||