Comparison of Airway Remodeling in Acute, Subacute ... - ATS Journals

Comparison of Airway Remodeling in Acute, Subacute, and Chronic Models of Allergic Airways Disease Natasha R. Locke*, Simon G. Royce*, Jacquetta S. Wainewright, Chrishan S. Samuel, and Mimi L. Tang Department of Allergy and Immunology, Murdoch Children’s Research Institute, the Royal Children’s Hospital; and Howard Florey Institute of Experimental Physiology and Medicine, the University of Melbourne, Parkville, Victoria, Australia

The relationship between airway inflammation and structural changes of airway remodeling, and their relative effects on airway function, are poorly understood. Remodeling is thought to result from chronic repetitive injury to the airway wall caused by airway inflammation; however, the mechanisms regulating remodeling changes have not been clearly defined. We examined the sequence of events in remodeling using three commonly used mouse models of allergic airways disease in which mice are exposed to nebulized ovalbumin for four consecutive days (acute), seven consecutive days (subacute), or three times a week for 6 wk (chronic). Surprisingly, we found that a very short period of exposure to ovalbumin was sufficient to elicit early changes of remodeling. Goblet cell hyperplasia and epithelial thickening were evident after just 4 d. In chronically challenged mice, these changes persisted and, in addition, subepithelial collagen deposition was significantly increased. This collagen deposition was associated with a failure to upregulate matrix metalloproteinase (MMP)-2, in conjunction with increased transforming growth factor-␤ and MMP-9 expression. The relationship between inflammation, remodeling changes, and airway hyperresponsiveness (AHR) were examined. The acute and subacute models exhibited marked airway inflammation, whereas the chronic model had very modest inflammation. Conversely, airway fibrosis was only evident in the chronic model. AHR was present in all three models; however, it was significantly higher in the chronic model compared with the acute (P ⬍ 0.05) and subacute (P ⬍ 0.05) models. These data demonstrate that both airway inflammation and airway fibrosis may contribute to AHR, with airway fibrosis leading to the greatest increases in AHR. Keywords: asthma; fibrosis; inflammation; mice

Numerous in vivo mouse models have been used to dissect the molecular mechanisms underlying allergic airways disease (AAD). However, the use of different strains of mice, a variety of sensitization and challenge protocols, and various methods of assessing responses makes it difficult to compare results from these models. Models employing short-term exposure to aerosolized antigen (acute and subacute models) are considered to replicate airway inflammatory events in asthma, whereas recurrent, long-term (chronic) exposure models are considered to replicate long-term changes of airway remodeling and airway hyperresponsiveness (AHR). Thus, acute models may be more indicative of the initiating events in asthma, whereas chronic models are considered to be more representative of human asthma with respect to cellular and structural changes within the airway wall, as well as the associated loss of airway function.

Airway remodeling is characterized by increased deposition of collagen (fibrosis) in the reticular basement membrane layer and peribronchial smooth muscle layer, fibroblast hyperplasia, goblet cell proliferation, mucous gland hyperplasia, epithelial disruption, smooth muscle hypertrophy/hyperplasia, and neovascularization (1). Although each of the components of airway remodeling is well characterized, the sequence in which these events develop is not known, and the factors and mechanisms regulating remodeling are not clearly delineated. Although eosinophilia is dependent on IL-5 (2), chronic epithelial and fibrotic changes occur independently of IL-5 (3). This indicates that separate mechanisms may control airway inflammation and airway wall remodeling. A role for IL-4 and IL-13 in promoting airway remodeling has been suggested in transgenic animal models (4–7). Both IL-4 and IL-13 stimulate fibroblast activation in vitro. These interleukins also stimulate the release of the profibrotic cytokine, transforming growth factor (TGF)-␤, from epithelial cells (8), which can contribute to subepithelial fibrosis in asthma (9, 10). These findings suggest that airway inflammation can directly promote remodeling. However, although acute inflammation plays an important role, it cannot be the only factor leading to remodeling, because optimal control of inflammation does not necessarily prevent remodeling. Conversely, the increased TGF-␤ expression in asthma (11) may exert an antiinflammatory effect, as TGF-␤1–null mice develop severe inflammation, and heterozygous mice subjected to a short-term ovalbumin (OVA) AAD model had increased eosinophilic inflammation, T helper cells (Th2) type 2 cytokine expression, but unaltered AHR (12). The matrix metalloproteases (MMPs) also play a key role in both inflammatory and remodeling pathogenesis. The gelatinases (MMP-2 and MMP-9) degrade basement membrane components and, therefore, facilitate egress of inflammatory cells, but also counteract airway wall remodeling caused by fibrosis. Further studies are required to clarify the relationship between inflammation and remodeling and to gain insight into the sequence of events that develop in the remodeling process. This article directly compares the functional, morphologic, and immunologic aspects of three major models of asthma, to shed light on the airway remodeling process, its relationship to airway inflammation, and their relative contributions to AHR.

MATERIALS AND METHODS Mice

(Received in original form February 23, 2006 and in final form December 13, 2006 ) *These authors contributed equally to this work. This work was supported by a Murdoch Children’s Research Institute Project Grant. Correspondence and requests for reprints should be addressed to Mimi L.K. Tang, M.D., Ph.D., Department of Allergy and Immunology, Royal Children’s Hospital, Parkville, Victoria 3052, Australia. E-mail: [email protected] Am J Respir Cell Mol Biol Vol 36. pp 625–632, 2007 Originally Published in Press as DOI: 10.1165/rcmb.2006-0083OC on January 19, 2007 Internet address: www.atsjournals.org

Six-week-old male BALB/c mice were purchased from the Walter and Eliza Hall Institute (Melbourne, Australia) and housed in a standard animal care facility (Department of Agriculture, University of Melbourne, VC, Australia).

Mouse Models of OVA-Induced Allergic Airway Disease Sensitization–challenge protocols for each of the three models used in this study are summarized in Figure 1. Briefly, mice were sensitized by intraperitoneal injection of 10 ␮g Grade V chicken egg OVA (Sigma Chemical Co., St. Louis, MO) and 1 mg aluminium potassium sulfate

626

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL 36

2007

100 ␮m were stained with hematoxylin and eosin (H&E), Masson’s trichrome, and Alcian blue–periodic acid Schiff (AB-PAS).

Morphometric Analysis of Inflammatory and Structural Changes

Figure 1. Sensitization and challenge protocols of acute, subacute, and chronic models of AAD. AAD was induced by intraperitoneal sensitization of BALB/c mice with 10 ␮g grade V OVA and 1 mg alum in saline. Sensitization was delivered on Day 0 (acute and subacute models) or on Days 0 and 14 (chronic model). Mice were then challenged by nebulization of 1% (wt/vol) OVA in saline for 30 min on four consecutive days (acute), seven consecutive days (subacute), or 3 d/wk for 6 wk (chronic). Control mice were sensitized with 1 mg of alum in 0.5 ml saline and challenged with nebulized saline.

Photomicrograph images were captured using a Spot Cooled Color Digital camera (Q Imaging, Burnaby, BC, Canada). Briefly, a minimum of 5 bronchi, measuring 150–350 ␮m in luminal diameter, were analyzed per mouse for the parameters described subsequently here, using Image Pro-Discovery software (Media Cybernetics, Silver Spring, MD), calibrated with a reference micrometer slide. The thickness of the bronchial epithelial layer, smooth muscle layer, and total wall was measured in H&E-stained sections. Subepithelial deposition of collagen was measured in sections stained with Masson’s trichrome. Hyperplasia of goblet cells within the bronchial epithelium was assessed by counting cells in AB-PAS–stained sections and expressed as the mean number of goblet cells per 100 ␮m of basement membrane. In addition, the full length of longitudinally orientated H&E-stained trachea was examined under 100⫻ oil immersion objective to enumerate intraepithelial eosinophils, expressed as the mean number of cells per 100 ␮m of basement membrane.

Immunohistochemistry for TGF-␤1, MMP-2, and MMP-9 (alum; Sigma Chemical) in 0.5 ml saline. Mice were then challenged for 30 min by nebulization of 1% (wt/vol) OVA in saline solution on Days 14–17 (acute model), Days 14–20 (subacute model), or 3 d/wk for 6 wk (chronic model) (n ⫽ 8 per group). Aerosol challenge was performed on groups of up to 16 mice in a whole-body inhalation exposure system attached to an ultrasonic nebulizer (NE-U07; Omron Corporation, Tokyo, Japan) with an output of 1 ml/min and 1- to 5-␮m particle size. Control mice were sensitized with 1 mg of alum in 0.5 ml saline and challenged with nebulized saline solution (n ⫽ 8 per group).

Airway AHR Methacholine-induced airway reactivity was assessed 24 h after the final aerosol challenge by direct plethysmography (Buxco Electronics, Troy, NY). Briefly, mice were anesthetized by intraperitoneal injection of ketamine and xylasine (200 ␮g/g and 10 ␮g/g, respectively), tracheotomized, and the jugular vein cannulated. Mice were ventilated with a small animal respirator (Harvard Apparatus, Holliston, MA) delivering 0.01 ml/g bodyweight at a rate of 120 strokes/min. Increasing methacholine doses were delivered intravenously, and airway resistance and dynamic compliance measured (Biosystem XA; Buxco Electronics) for 2 min after each dose. Results are expressed as the maximal resistance after each dose of methacholine minus baseline (PBS alone) resistance.

Determination of OVA-Specific Antibody Titres by ELISA Serum was obtained by lethal cardiac puncture of anaesthetized mice 24 h after the final OVA or saline challenge. Serum was stored in 100 ␮l aliquots at ⫺20⬚C for subsequent OVA-specific IgE measurement by ELISA, as previously described (13). Levels of OVA-specific IgE in serum samples were interpolated from a standard curve, generated from positive-control sera, and are expressed as units relative to the positive control (relative units).

Bronchoalveolar Lavage Immediately after cardiac puncture, bronchoalveolar lavage (BAL) was performed by infusion and extraction of 1 ml of ice-cold saline. This was repeated twice, and the lavages pooled (mean volume, 2.50 ⫾ 0.30 ml). Total viable cell counts were determined in a hemocytometer using trypan blue exclusion. Differential counts of eosinophils, neutrophils, lymphocytes, and monocytes were determined on cytospin smears of BAL samples (4 ⫻ 105 cells) from individual mice stained with DiffQuick (Life Technologies, Auckland, New Zealand) after counting 300 cells. Results are expressed as total cell number ⫻ 103.

Histopathology The right lung lobe and trachea were incubated in 10% formalin for 24 h and embedded in paraffin. Serial, 3-␮m sections taken every

Sections of lung tissue were stained immunohistochemically to detect and localize TGF-␤1, MMP-2, and MMP-9 protein expression. TGF-␤ and MMP-9 were identified using rabbit polyclonal antibodies, sc-146, specific for TGF-␤1 (Santa Cruz Biotechnology, Santa Cruz, CA), and ab16306, specific for latent and active forms of MMP-9 (Abcam, Cambridge, UK), respectively. Bound primary antibody was detected using anti-rabbit EnVision (Dako, Glostrup, Denmark). MMP-2 was identified by a biotinylated mouse monoclonal antibody against latent and active MMP-2 (clone MMP2/8B4; Abcam), and detected using streptavidin horseradish peroxidase (mouse on mouse Animal Research Kit; Dako). The chromagen 3,3-diaminobenzidine was used, and sections counterstained with haematoxylin.

Hydroxyproline Analysis of Lung Collagen A portion of each lung tissue was snap frozen for hydroxyproline determination, as previously described (14). Hydroxyproline content was used to estimate collagen content by multiplying by a factor of 6.94 (based on hydroxyproline representing ⵑ 14.4% of the amino acid composition of collagen in mammalian tissues) (15), and further expressed as a proportion of the lung dry weight (collagen concentration).

Statistical Analysis Results of morphometry, BAL differential cell counts, and OVAspecific IgE determination and hydroxyproline determination were expressed as mean (⫾ SE) for each experimental group. A nonparametric Mann–Whitney test was used to compare differences between groups. AHR measurements were compared using a two-way ANOVA and Bonferroni post test.

RESULTS OVA-specific serum IgE levels of each mouse were evaluated to ensure that adequate sensitization was achieved (Table 1). OVA-specific IgE titers were significantly elevated in all OVAsensitized and challenged mice compared with control mice. As expected, OVA-specific IgE titres were highest in the chronic model, where mice received two sensitizations and an extended period of aerosol challenge. Airway Inflammation in Acute, Subacute, and Chronic Models of Allergic Airway Disease

Inflammatory responses were assessed in situ by qualitative histologic evaluation of lung tissue sections and counts of tracheal intraepithelial eosinophils, and assessed ex vivo by differential cell counts of cells retrieved by BAL. Histologic analysis of lung tissue sections revealed negligible inflammation in saline-treated mice of acute, subacute, or chronic treatment groups (Figures 2A–2C).

Locke, Royce, Wainewright, et al.: Models of Allergic Airways Disease

627

TABLE 1. BRONCHOALVEOLAR LAVAGE DIFFERENTIAL CELL COUNTS AND OVALBUMIN-SPECIFIC IgE TITERS Acute Saline Total cells (⫻ 103) Eosinophils (⫻ 103) Neutrophils (⫻ 103) Lymphocytes (⫻ 103) Monocytes (⫻ 103) OVA-specific IgE*

57.71 0.21 0.51 0.26 56.94 0.027

⫾ ⫾ ⫾ ⫾ ⫾ ⫾

19.33 0.40 0.36 0.31 19.28 0.005

Acute OVA 366.14 111.39 6.14 6.42 208.33 1.73

⫾ ⫾ ⫾ ⫾ ⫾ ⫾

86.11† 96.03† 1.67‡ 2.55† 104.30† 1.15‡

Subacute Saline 180.38 0.23 2.51 3.31 174.32 0.000

⫾ ⫾ ⫾ ⫾ ⫾ ⫾

114.88 0.53 1.33 3.74 113.18 0.000

Subacute OVA 1726.3 1003.5 185.95 189.21 347.58 6.22

⫾ ⫾ ⫾ ⫾ ⫾ ⫾

921.60† 691.51† 113.48‡ 182.68† 86.84 3.17‡

Chronic Saline 137.81 0.05 6.96 1.32 129.48 0.000

⫾ ⫾ ⫾ ⫾ ⫾ ⫾

29.26 0.15 12.46 0.76 28.30 0.000

Chronic OVA 182.50 5.96 5.89 14.70 155.95 36.45

⫾ ⫾ ⫾ ⫾ ⫾ ⫾

79.27 7.75† 4.59 7.89† 67.37 13.92‡

Definition of abbreviation: OVA, ovalbumin. The lungs of OVA and saline-treated mice were lavaged with ice-cold PBS. Differential counts of eosinophils, neutrophils, lymphocytes, and monocytes were determined on cytospin smears of bronchoalveolar lavage (BAL) samples, from individual mice, stained with DiffQuick by counting 300 cells. Results are expressed as total cell number ⫻ 10⫺3 per ml BAL fluid retrieved. OVA-specific IgE titres were determined in individual serum samples by ELISA, and results are expressed as units relative to a standard curve generated from positive-control sera (relative units). * P ⬍ 0.05 † P ⬍ 0.01 ‡ P ⬍ 0.001

Inflammatory cell influx was very prominent in acute (Figure 2D) and subacute (Figure 2E) models, whereas a modest inflammatory infiltrate was present in the chronic model (Figure 2F). Importantly, inflammation was peribronchial and perivascular, but was not present within the lung parenchyma.

Similarly, analysis of cells retrieved from the airways by BAL (Table 1) revealed that inhalational OVA caused a significant increase in the total number of inflammatory cells present in the acute and subacute models, but did not result in significant inflammatory cell increase in the chronic model. These data

Figure 2. Inflammatory infiltrate in acute, subacute, and chronic models of AAD. Lung tissue sections were stained with H&E to assess inflammatory cell infiltrate. Shown are representative lung tissue sections from mice treated with saline (A–C ) or OVA (D–F ) in acute (A, B ), subacute (C, D ), and chronic (E, F ) models of AAD. High-power insets show changes to airway wall morphology. Bar ⫽ 50 ␮m.

628


correspond well with the small inflammatory responses observed in situ in chronic OVA–treated mice (Figure 2), and indicate that, in contrast to acute and subacute models, inflammation is not the predominant feature of the chronic mouse asthma model. Differential cell counts revealed significantly increased eosinophil numbers in all OVA groups. In acute and subacute models of AAD, high proportions of inflammatory cells present were eosinophils (30 and 58%, respectively), whereas, in the chronic model, monocytes were the predominant cell type, and only a small percentage of cells were eosinophils (3.3%). Lymphocytes were increased 3.9-, 5.9- and 8.4-fold in OVA-treated mice compared with saline control mice in acute, subacute, and chronic models, respectively. Neutrophil numbers were significantly increased only in acute and subacute models. Airway inflammation may also be assessed by enumeration of intraepithelial eosinophils within longitudinal tracheal sections. This is frequently used as a measure of airway inflammation in chronic models due to the reduced BAL cell counts and tissue inflammation observed in these longer-term models (3, 16). In contrast to the findings for BAL, numbers of tracheal intraepithelial eosinophils (Figure 3) were significantly elevated in all three models (acute, P ⬍ 0.05; subacute, P ⬍ 0.05; chronic,

2007

P ⬍ 0.001) as compared with the respective control mice, and there were no significant differences observed between the three models. Airway Remodeling in Acute, Subacute, and Chronic Models of Allergic Airway Disease

Early changes of remodeling, such as epithelial thickening (P ⬍ 0.001) and goblet cell hyperplasia (P ⬍ 0.001), were present in all three models (Figure 3). There was no significant difference in any of these parameters of remodeling between the three models of OVA-induced AAD. In contrast, a significant increase in subepithelial collagen deposition was only evident in the chronic model (P ⬍ 0.05). In addition, subepithelial collagen deposition in the chronic model of AAD was significantly greater (P ⬍ 0.05) than in the subacute and acute models. These findings were supported by quantitative analysis of total lung collagen, which also showed significantly greater collagen in the chronic model (P ⬍ 0.05) compared with the acute and subacute groups. Neither smooth muscle thickness nor basement membrane thickness were significantly altered by OVA treatment in acute, subacute, or chronic models.

Figure 3. Morphometric analysis of intraepithelial eosinophil counts and airway wall remodeling changes in acute, subacute, and chronic models of allergic airway disease. Serial sections of paraffin-embedded lung tissues stained with H&E, Masson’s trichrome, and ABPAS were analyzed using Image Pro-Discovery software for thickness of the smooth muscle layer, total wall, and subepithelial collagen. Total lung collagen was determined by hydroxyproline analysis. The number of goblet cells within the bronchial epithelium was assessed by counting mucin-positive cells in AB-PAS–stained sections. Intraepithelial eosinophils were counted in trachea sections stained with H&E under 100⫻ oil immersion objective. Data are expressed as the mean number of cells per 100 ␮m basement membrane. Error bars show the standard deviation. *P ⬍ 0.05, **P ⬍ 0.01.


Imunohistochemical Analysis of TGF-␤1, MMP-2, and MMP-9 Expression

The localization of three proteins, TGF-␤1, MMP-2, and MMP-9, known to regulate airway inflammation and airway remodeling changes in asthma, was investigated by immunohistochemistry in formalin-fixed, paraffin-embedded lung tissue sections from the three models of AAD (Figure 4). Little constitutive staining for TGF-␤1 was observed in the saline-sensitized and -challenged control mice. In the lungs of mice receiving acute, subacute, and chronic OVA sensitization and challenge, staining for TGF-␤1 was markedly increased. Strong cytoplasmic staining was localized to bronchial epithelial cells, connective tissue cells of the lamina propria and adventitia, and smooth muscle cells. Strong staining was also seen in inflammatory cells, especially in the acute and subacute models. Staining for MMP-2 revealed weak expression in the saline control mice. In the OVA-treated mice, MMP-2 expression was of high intensity and extent in the acute and subacute models, but was weak in the chronic model. Staining for MMP-9 revealed upregulated expression in acute, subacute, and chronic models of AAD compared with the saline control mice, with similar levels of expression in the three OVA models. Airway Function in Acute, Subacute, and Chronic Models of Allergic Airway Disease

To determine if the observed differences in lung structure between the three models of AAD were associated with physiologic differences in lung function, methacholine-induced AHR was measured by plethysmography 24 h after the final aerosol challenge. Results are expressed as the relative increase in resistance from baseline. All three models induced significant increases in airway reactivity to methacholine in comparison to saline-treated control animals (Figure 5). The dose–response curves of acute and subacute OVA-treated mice were not significantly different from each other. However, in the chronic model of AAD, mice exhibited higher airway AHR compared with acute and subacute models (P ⬍ 0.05). Importantly, the increase in AHR after OVA

629

exposure was highest in the group with significant collagen deposition; namely, the chronic OVA group.

DISCUSSION We have compared the changes evident in mouse models of acute, subacute, and chronic AAD to better understand the sequence of events occurring in airway remodeling and to clarify the relationship between airway inflammation, AHR, and airway remodeling. All three models received intraperitoneal sensitization with OVA; however, challenge regimens differed in the duration of OVA aerosol exposures. In acute and subacute models of asthma, mice typically receive a single sensitizing dose of OVA, followed by a short period of exposure to nebulized OVA solution (3–4 or 7–10 consecutive days, respectively) (4, 13, 17, 18). In chronic models of asthma, rodents are sensitized to OVA, often by two sensitizing doses, and then exposed to inhalational challenge with OVA for periods of 2–12 wk (10, 19–22). The development of AHR and airway remodeling are more marked when animals are challenged for longer periods (6–12 wk) (10, 19, 21–23). Mice exposed to low-mass OVA aerosol for 6 wk develop changes of airway inflammation, airway remodeling, and AHR that closely replicate the immunologic and pathophysiologic changes observed in human asthma (19). However, some features, such as recruitment of mast cells to the airway epithelium, eosinophils in the intrapulmonary airways, and increased smooth muscle mass, are not observed in this model (19, 24). Recent studies in humans suggest that remodeling occurs in early life, before the clinical manifestation of asthma, as a result of a primary defect of the epithelium and epithelial repair process (25, 26). Reticular basement membrane thickening can also precede clinical disease (27). However, the precise sequence of structural changes is not known. We investigated the sequence of events with respect to individual aspects of airway remodeling in mouse models.

Figure 4. Immunohistochemical staining for TGF-␤1, MMP-2, and MMP-9 in representative mouse lung tissues from acute, subacute, and chronic models of AAD. Representative lung tissue sections are shown for mice exposed to the acute, subacute, and chronic models of AAD, and for corresponding saline controls. Strong staining for TGF-␤1 was found in bronchial epithelium and peribronchial inflammatory cells in acute (A ), subacute (B), and chronic (C ) models, but only weak staining of occasional cells was present in saline control lungs (D ). A similar expression profile was observed for MMP-9 in acute (E ), subacute (F ), and chronic (G ) models, and saline controls (H ). In contrast, strong staining for MMP-2 was only observed in acute (I ) and subacute ( J ) models, with weak expression in the chronic AAD model (K ) and saline controls (L ). Bar ⫽ 100 ␮m.

630


2007

Figure 5. Airway hyperresponsiveness in acute, subacute, and chronic models of allergic airway disease. Methacholine-induced AHR was measured by plethysmography 24 h after the final aerosol challenge. The data are derived from three independent experiments, to give a total of n ⫽ 8 per experimental group (acute OVA, acute saline control, subacute OVA, subacute saline control, chronic OVA, and chronic saline control). Results are presented as mean ⫾ SD and expressed as the resistance change from baseline (PBS only) responses of individual mice (n ⫽ 8/group). Twoway ANOVA with Bonferroni post test was used to compare groups. *P ⬍ 0.05; **P ⬍ 0.01; ***P ⬍ 0.001.

A number of structural changes were observed in all three groups and, therefore, likely represent early changes in the course of airway remodeling. Goblet cell hyperplasia and epithelial thickening were evident after just 4 d of OVA inhalation. Persistence of changes in the airway epithelium may contribute to changes in the stroma that are an essential part of airway remodeling. As a response to microinjuries, the epithelium is not only responsible for releasing a cocktail of secretions (including mucins, healing peptides, and antibacterial enzymes) into the lumen, but has also been shown to be a major source of cytokines. Some of these are important in inflammation (granulocytemacrophage colony–stimulating factor and IL-8), TGF-␣ is a ligand for epidermal growth factor receptor, which promotes mucus secretion, and others are profibrotic (TGF-␤), acting directly on cells of the fibroblast/myofibroblast phenotype (28). This association between epithelium and fibroblast in promoting fibrosis is known as the epithelial mesenchymal trophic unit (29). Phenotypic changes in the epithelium and a subsequent oversecretion of mucus are well known features of severe asthma, and mucus plugs are a major contributor to asthma fatalities. Information on the role of goblet cell hyperplasia is less conclusive in mild and moderate asthma, probably due to the unavailability of well preserved human specimens for research; however, recent reports suggest goblet cell numbers are increased 2.5-fold (30). Certainly, goblet cell hyperplasia is a consistent feature of acute (31) and chronic (16) murine OVA sensitization models of AAD, and may occur in some animals between 24 and 36 h after a single exposure to antigen (32). We found that a very short period of exposure to OVA was sufficient to elicit goblet cell hyperplasia and epithelial thickening. Other airway remodeling changes were only noted after more prolonged exposure to OVA. The most striking change in the chronic model was an increase in subepithelial collagen, as determined by morphometric analysis of Masson’s trichrome–stained lung tissue sections, and total lung collagen, as determined by hydroxyproline analysis. Deposition of collagen in the airway is a key feature of airway remodeling. It is particularly significant in the lamina reticularis (33) and in the submucosa (34). We have used the thickness of Masson’s trichrome staining of the peribronchial collagen layer to detect changes in subepithelial collagen deposition. Our finding is consistent with previous reports of increased Masson’s trichrome staining (35) and increased hydroxyproline in the airway in asthma (36).

Major regulators of airway fibrosis, TGF-␤1, MMP-2, and MMP-9, were all upregulated in acute and subacute models of asthma compared with saline control animals. In acute asthmatic episodes, MMPs allow inflammatory cells to pass freely into lung tissues and contribute to airway inflammation. Although increased TGF-␤1 and MMP-9 expression were retained after prolonged antigen challenge, staining for MMP-2 in the chronic model returned to levels observed in saline control mice. This differential expression of the two MMPs may indicate different roles in the regulation of inflammation and airway remodeling. Reduced expression of MMP-2 may be important in chronic AAD by reducing leukocyte egress or by allowing the accumulation of extracellular matrix in the airway wall. The strong staining for MMPs in bronchial epithelial cells is consistent with the importance of the epithelial mesenchymal trophic unit in asthma pathogenesis. The relationship between airway inflammation and airway remodeling was found to be complex. Airway inflammation was marked in the acute and subacute models, but minimal in the chronic model of AAD. This is consistent with previous reports of chronic models of AAD (3, 16). It is important to note that mice in the chronic group received a second sensitizing dose of OVA, whereas mice in the acute and subacute models received a single sensitization. The effect of this second sensitizing dose on inflammatory responses has not been investigated here; however, it is likely that stronger T-cell responses are generated. The impact of stronger inflammatory responses on the induction of airway remodeling changes is unknown, but, given the lack of airway inflammation consistently observed in chronic models of AAD, enhanced immune responses alone are unlikely to explain the development of fibrosis. The induction of fibrosis consistently requires prolonged exposure to low levels of allergen. Interestingly, BAL fluid (BALF) eosinophils did not correlate with intraepithelial eosinophils in these models— intraepithelial eosinophils were increased in all three models, whereas BALF eosinophilia was only observed in the acute and subacute models. Furthermore, the various remodeling changes did not correspond well with eosinophil inflammation. Epithelial changes corresponded with intraepithelial eosinophil counts, but did not correlate with BALF eosinophil counts. In contrast, increased collagen deposition did not correspond with the number of BALF or intraepithelial eosinophils. Eosinophils may


contribute to epithelial damage and remodeling by production of eosinophil peroxidase, TNF-␣, and IFN-␣ (37). Our findings suggest that these effects are likely to be mediated primarily by tissue eosinophils rather than luminal eosinophils. Eosinophils are also thought to play an important role in the development of fibrosis via secretion of TGF-␤, IL-4, IL-13, and fibroblast growth factor 2. These factors are likely to act locally within the airway wall, acting directly upon fibroblasts to induce increased production of matrix proteins and differentiation of fibroblasts to myofibroblasts. It is therefore surprising that fibrosis was not related to tissue numbers of eosinophils and, instead, was only observed in the chronic model in which there was very modest airway inflammation. One possible explanation for this finding is that acute inflammatory cells within the airway may induce production of MMP-2 by epithelium, which acts to limit the progression of airway fibrosis. The regression of airway inflammation and epithelial production of MMP-2 in the chronic model would then allow TGF-␤–induced airway fibrosis to progress. Taken together, the findings in this study suggest that the contribution of eosinophils to airway remodeling and fibrosis is complex. Although eosinophils may contribute to remodeling changes in the epithelium, subepithelial fibrosis may progress in the absence of significant eosinophilic inflammation. It is possible that eosinophil migration may have occurred at an earlier time point, which was not identified on Day 64. Further studies using a monoclonal antibody to inhibit eosinophil activity would provide further information on the relative contribution of eosinophils to the remodeling process. What are the relative contributions of airway inflammation and airway remodeling to AHR? Airway wall thickening contributes to reduced compliance of the airway wall, reduction in elasticity, and bronchial narrowing (38, 39). However, its contribution to the major measure of lung function, AHR, has been questioned (40). Airway inflammation has been shown to contribute to AHR; however, AHR may persist despite optimal control of inflammation (41). We examined the relationships between AHR and both airway inflammation and remodeling to better understand how AHR might be influenced by each of these pathologies. We found that methacholine-induced AHR was raised in all models, but was highest in the chronic model. This suggests that both airway inflammation (prominent in the acute and subacute models) and airway remodeling (present only in the chronic model) can contribute to AHR, but raises the hypothesis that airway fibrosis may lead to the greatest increases in AHR. In the acute and subacute models of AAD, epithelial changes and/or airway inflammation may also contribute to increases in AHR. Hyperplasia of the mucus-secreting goblet cells will contribute to luminal narrowing. Mathematical models have predicted that an increase in the volume of the airway wall on the luminal side of the smooth muscle layer (thickened epithelium) will greatly increase the effect of smooth muscle contraction in narrowing the lumen of an airway (39, 42, 43). The evidence linking airway inflammation with AHR is less definite. The Th2 cytokines, IL-4, IL-5, and IL-13, may contribute to smooth muscle contraction; however, the importance of this is uncertain, as AHR can still occur in the absence of Th2 cytokines (44). We observed the greatest increase in AHR in the chronic model of AAD, which was associated with marked airway fibrosis and very modest inflammation, suggesting that airway fibrosis per se may be an important factor contributing to AHR. There are a number of considerations when using mouse models of AAD. We could not detect any increase in smooth muscle thickness in our study. Smooth muscle is thought to contribute to both acute and chronic asthma bronchoconstriction, and smooth muscle hyperplasia and hypertrophy accounts

631

for a thicker smooth muscle layer in the human and in some animal models, but was not observed in the chronic murine OVA sensitization model of AAD (19). This highlights the fact that murine models can only replicate some aspects of human asthma. One other drawback of the mouse model is the limited capacity to accurately measure lung function in a small animal. We have used the invasive plethysmography method because this yields data that most closely resembles that from human lung function tests. Given that a tracheostomy is performed, this approach can directly measure airway resistance changes in the pulmonary airways and eliminates confounding contributions from the upper airways. It also eliminates the interference associated with physical activity in unrestrained animals in chambers that contain high volumes of air. Genetic factors are important in AAD and its progression in mice (45, 46), just as it is in humans (47). We have established the models on the BALB/c background, as this strain is one of the best responders to OVA sensitization (45, 46). In summary, our findings demonstrate that the earliest changes of remodeling can occur after very brief periods of acute inflammation, and involve changes in the epithelium, including goblet cell hyperplasia and thickening and metaplasia of epithelial cells. These findings support the current understanding of airway remodeling being initiated by a defect of epithelial repair processes in asthma. We found that increased collagen deposition occurs later in the remodeling process, and may result in part from a failure to upregulate MMP-2 expression in the setting of increased TGF-␤1 and MMP-9. Furthermore, whereas eosinophils may contribute to epithelial remodeling changes, airway fibrosis may progress in the absence of significant airway eosinophilia. Finally, our data suggest that both airway inflammation and airway remodeling may contribute to AHR, with airway fibrosis intensifying AHR responses. These findings provide a better understanding of the processes regulating airway remodeling and AHR in asthma. Conflict of Interest Statement : None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript.

References 1. Chiappara G, Gagliardo R, Siena A, Bonsignore MR, Bousquet J, Bonsignore G, Vignola AM. Airway remodelling in the pathogenesis of asthma. Curr Opin Allergy Clin Immunol 2001;1:85–93. 2. Foster PS, Hogan SP, Ramsay AJ, Matthaei KI, Young IG. Interleukin 5 deficiency abolishes eosinophilia, airways hyperreactivity, and lung damage in a mouse asthma model. J Exp Med 1996;183:195–201. 3. Foster PS, Ming Y, Matthei KI, Young IG, Temelkovski J, Kumar RK. Dissociation of inflammatory and epithelial responses in a murine model of chronic asthma. Lab Invest 2000;80:655–662. 4. Rankin JA, Picarella DE, Geba GP, Temann UA, Prasad B, DiCosmo B, Tarallo A, Stripp B, Whitsett J, Flavell RA. Phenotypic and physiologic characterization of transgenic mice expressing interleukin 4 in the lung: lymphocytic and eosinophilic inflammation without airway hyperreactivity. Proc Natl Acad Sci USA 1996;93:7821–7825. 5. Shirakawa I, Deichmann KA, Izuhara I, Mao I, Adra CN, Hopkin JM. Atopy and asthma: genetic variants of IL-4 and IL-13 signalling. Immunol Today 2000;21:60–64. 6. Wills-Karp M, Luyimbazi J, Xu X, Schofield B, Neben TY, Karp CL, Donaldson DD. Interleukin-13: central mediator of allergic asthma. Science 1998;282:2258–2261. 7. Zhu Z, Homer RJ, Wang Z, Chen Q, Geba GP, Wang J, Zhang Y, Elias JA. Pulmonary expression of interleukin-13 causes inflammation, mucus hypersecretion, subepithelial fibrosis, physiologic abnormalities, and eotaxin production. J Clin Invest 1999;103:779–788. 8. Richter A, Puddicombe SM, Lordan JL, Bucchieri F, Wilson SJ, Djukanovic R, Dent G, Holgate ST, Davies DE. The contribution of interleukin (IL)-4 and IL-13 to the epithelial-mesenchymal trophic unit in asthma. Am J Respir Cell Mol Biol 2001;25:385–391.

632


9. Lee CG, Cho SJ, Kang MJ, Chapoval SP, Lee PJ, Noble PW, Yehualaeshet T, Lu B, Flavell RA, Milbrandt J, et al. Early growth response gene 1–mediated apoptosis is essential for transforming growth factor ␤1induced pulmonary fibrosis. J Exp Med 2004;200:377–389. 10. Kenyon NJ, Ward RW, Last JA, Airway fibrosis in a mouse model of airway inflammation. Toxicol Appl Pharmacol 2003;186:90–100. 11. Minshall EM, Leung DY, Martin RJ, Song YL, Cameron L, Ernst P, Hamid Q. Eosinophil-associated TGF-␤1 mRNA expression and airways fibrosis in bronchial asthma. Am J Respir Cell Mol Biol 1997; 17:326–333. 12. Scherf W, Burdach S, Hansen G. Reduced expression of transforming growth factor ␤ 1 exacerbates pathology in an experimental asthma model. Eur J Immunol 2005;35:198–206. 13. Keramidaris E, Merson TD, Steeber DA, Tedder TF, Tang ML. L-selectin and intercellular adhesion molecule 1 mediate lymphocyte migration to the inflamed airway/lung during an allergic inflammatory response in an animal model of asthma. J Allergy Clin Immunol 2001;107:734–738. 14. Samuel CS, Butkus A, Coglan JP, Bateman JF. The effect of relaxin on collagen metabolism in the non-pregnant rat pubic symphysis: the influence of estrogen and progesterone in regulating relaxin activity. Endocrinology 1996;137:3384–3390. 15. Gallop PM, Paz MA. Posttranslational protein modifications, with special attention to collagen and elastin. Physiol Rev 1975;55:418–487. 16. Temelkovski J, Hogan SP, Shepherd DP, Foster PS, Kumar RK. An improved murine model of asthma: selective airway inflammation, epithelial lesions and increased methacholine responsiveness following chronic exposure to aerosolised allergen. Thorax 1998;53:849–856. 17. Epstein MM. Do mouse models of allergic asthma mimic clinical disease? Int Arch Allergy Immunol 2004;133:84–100. 18. Wegmann M, Renz H. Animal models of experimental asthma. Ernst Schering Res Found Workshop 2005;50:69–87. 19. Kumar RK, Foster PS. Murine model of chronic human asthma. Immunol Cell Biol 2001;79:141–144. 20. Leung SY, Eynott P, Noble A, Nath P, Chung KF. Resolution of allergic airways inflammation but persistence of airway smooth muscle proliferation after repeated allergen exposures. Clin Exp Allergy 2004;34: 213–220. 21. Palmans E, Kips JC, Pauwels RA. Prolonged allergen exposure induces structural airway changes in sensitized rats. Am J Respir Crit Care Med 2000;161:627–635. 22. Sakai K, Yokoyama A, Kohno N, Hamada H, Hiwada K. Prolonged antigen exposure ameliorates airway inflammation but not remodeling in a mouse model of bronchial asthma. Int Arch Allergy Immunol 2001;126:126–134. 23. Ramos-Barbon D, Ludwig MS, Martin JG. Airway remodeling: lessons from animal models. Clin Rev Allergy Immunol 2004;27:3–21. 24. Kumar RK, Thomas PS, Seetoo DQ, Herbert C, McKenzie AN, Foster PS, Lloyd AR. Eotaxin expression by epithelial cells and plasma cells in chronic asthma. Lab Invest 2002;82:495–504. 25. Pohunek P, Roche WR, Turzikova J, Kudrmann J, Warner JO. Eosinophilic inflammation in the bronchial mucosa of children with bronchial asthma. Eur Respir J Suppl 1997;10:160S. (Abstr.). 26. Roche WR, Beasly R, Willliams JH, Holgate ST. Subepithelial fibrosis in the bronchi of asthmatics. Lancet 1989;1:520–524. 27. Barbato A, Turato G, Baraldo S, Bazzan E, Calabrese F, Tura M,

28.

29.

30. 31.

32.

33.

34.

35.

36.

37. 38. 39.

40. 41.

42. 43. 44. 45. 46.

47.

2007

Zuin R, Beghe B, Maestrelli P, Fabbri LM, et al. Airway inflammation in childhood asthma. Am J Respir Crit Care Med 2003;168:798–803. Davies DE, Wicks J, Powell RM, Puddicombe SM, Holgate ST. Airway remodelling in asthma: new insights. J Allergy Clin Immunol 2003; 111:215–225. Holgate ST, Davies DE, Lackie PM, Wilson SJ, Puddicombe SM, Lordan JL. Epithelial-mesenchymal interactions in the pathogenesis of asthma. J Allergy Clin Immunol 2000;105:193–204. Fahy JV. Remodeling of the airway epithelium in asthma. Am J Respir Crit Care Med 2001;164(10 Pt 2):S46–S51. Blyth DI, Pedrick MS, Savage TJ, Bright H, Beesley JE, Sanjar S. Induction, duration and resolution of airway goblet cell hyperplasia in a murine model of atopic asthma: effect of concurrent infection with respiratory syncytial virus and response to dexamethasone. Am J Respir Cell Mol Biol 1998;19:38–54. Blyth DI, Pedrick MS, Savage TJ, Hessel EM, Fattah D. Lung inflammation and epithelial changes in a murine model of atopic asthma. Am J Respir Cell Mol Biol 1996;14:425–438. Chetta A, Foresi A, Del Donno M, Bertorelli G, Pesci A, Olivieri D. Airways remodeling is a distinctive feature of asthma and is related to severity of disease. Chest 1997;111:852–857. Wilson JW, Li X. The measurement of reticular basement membrane and submucosal collagen in the asthmatic airway. Clin Exp Allergy 1997;27:363–371. Leigh R, Ellis R, Wattie J, Southam DS, de Hoogh M, Gauldie J, O’Byrne PM, Inman MD. Dysfunction and remodeling of the mouse airway persist after resolution of acute allergen-induced airway inflammation. Am J Respir Cell Mol Biol 2002;27:526–535. Nagao K, Tanaka H, Komai M, Masuda T, Narumiya S, Nagai H. Role of prostaglandin I2 in airway remodeling induced by repeated allergen challenge in mice. Am J Respir Cell Mol Biol 2003;29:314–320. Makino S, Fukuda T. Eosinophils and allergy in asthma. Allergy Proc 1995;16:13–21. Wiggs BR, Moreno R, Hogg JC, Hilliam C, Pare PD. A model of the mechanics of airway narrowing. J Appl Physiol 1990;69:849–860. Wiggs BR, Bosken C, Pare PD, James A, Hogg JC. A model of airway narrowing in asthma and in chronic obstructive pulmonary disease. Am Rev Respir Dis 1992;145:1249–1250. McParland BE, Macklem PT, Pare PD. Airway wall remodeling: friend or foe? J Appl Physiol 2003;95:426–435. van Essen-Zandvliet EE, Hughes MD, Waalkens HJ, Duiverman EJ, Kerrebijn KF. Remission of childhood asthma after long-term treatment with an inhaled corticosteroid (budesonide): can it be achieved? Dutch CNSLD Study Group. Eur Respir J 1994;7:63–68. Moreno R, Hogg J, Pare P. Mechanics of airway narrowing. Am Rev Respir Dis 1986;133:1171–1180. James A, Pare P, Hogg J. The mechanics of airway narrowing in asthma. Am Rev Respir Dis 1989;139:242–246. Cockcroft DW, Davis BE. Mechanisms of airway hyperresponsiveness. J Allergy Clin Immunol 2006;118:551–559. Shinagawa K, Kojima M. Mouse model of airway remodeling: strain differences. Am J Respir Crit Care Med 2003;168:959–967. Takeda K, Haczku A, Lee JJ, Irvin CG, Gelfand EW. Strain dependence of airway hyperresponsiveness reflects differences in eosinophil localization in the lung. Am J Physiol Lung Cell Mol Physiol 2001;281: L394–L402. Vercelli D. Genetic polymorphism in allergy and asthma. Curr Opin Immunol 2003;15:609–613.