INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Allergen exposure in atopic asthma leads to airflow obstruction, bronchial hyperresponsiveness, and tissue injury (1), which may be accompanied by a constitutive repair process that results in airway remodeling and persistent airflow obstruction (2, 3). Allergen-induced inflammation is a first step in airway injury and may contribute to processes leading to remodeling by inducing both the removal and synthesis of extracellular matrix (ECM) components.

We have previously demonstrated that segmental bronchoprovocation with antigen (AG-SBP) causes local generation of soluble fibronectin (4), which may be a factor in airway remodeling. Other factors can contribute to the injury/repair process, including an activation of airway inflammatory cells and release of matrix metalloproteinases (MMPs) that may play a role in matrix turnover and repair. Under normal circumstances, these enzymes degrade damaged matrix and maintain normal tissue homeostasis. However, under pathologic conditions, MMPs may be produced in excess and thus, may also contribute to tissue damage and activation of inappropriate repair mechanisms.

Of the metalloproteinases, MMP-9 is of particular relevance to asthma. Compared with normal subjects, increased concentrations of MMP-9 have been detected in the bronchoalveolar lavage fluid (BALF) (5, 6), sputum (7), and serum (8) of asthmatic subjects; increased MMP-9 messenger RNA (mRNA) (9) and immunoreactivity (9, 10) have been observed in bronchial biopsies of asthmatics, and enhanced enzymatic activity has been reported after antigen (AG) challenge (11). In a murine model, inhibition of MMP-9 by in vivo administration of tissue inhibitors of metalloproteinases (TIMPs) or a synthetic MMP inhibitor, decreased antigen-induced airway inflammation and hyperresponsiveness (12).

This study was designed to investigate the role of allergen in the induction of MMP-9 and TIMP-1 in allergic subjects. Bronchoscopy with saline (SAL)- and AG-SBP were done in 17 atopic subjects and bronchoalveolar lavage (BAL) was performed 5 min and 48 h after the challenge. BALF was evaluated for MMP-9 and TIMP-1 by ELISA. Gelatinase activity was evaluated by zymography and the source of MMP-9 was investigated by immunochemistry of BAL cells.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Seventeen allergic subjects were recruited for the study (Table 1). The subjects included 10 males and seven females ranging in age from 21 to 43 yr, who were in good health with the exception of allergic airway disease. Subjects had a positive skin prick test to one or more aeroallergens. Each subjects underwent a medical history, physical examination, and pulmonary function testing by spirometry. Informed consent was obtained from each subject prior to participation. The study was approved by the University of Wisconsin-Madison Center for Health Sciences Human Subjects Committee.

Segmental Bronchoprovocation (SBP) and BAL

To establish the airway response and determine allergen dose for SBP, a graded inhaled AG challenge was performed (13). The AG dose leading to a 20% decrease in FEV₁ (AG PD₂₀) was calculated from the dose-response curve. Allergens used in SBP included ragweed (GS Ragweed mix; Greer Labs, Lenioir, NC), cat AG (Fel D1; Bayer Allergy Products, Spokane, WA), and house dust mite (Dermatophagoides farinae; Miles Allergy Products, Spokane, WA). Bronchoscopy and SBP were performed as previously described (14). A volume of 10 ml of SAL (0.9% NaCl) was instilled into one segment and in a separate segment, AG-SBP was performed using an AG dose equal to 10% of the AG PD₂₀. BAL was performed in each segment 5 min and 48 h after challenge.

Analysis of BALF

BALF was kept on ice throughout processing. BAL cells were recovered from the lavage fluid by centrifugation at 200 × g for 10 min at 4° C. Total cell numbers were determined by hemacytometer, and differential cell counts were performed on cytospin preparations stained with a modified Giemsa-based Diff-Quik stain (Baxter Scientific Products, McGaw Park, IL). BALF were stored at 70° C until analyzed.

MMP-9 and TIMP-1 ELISAs

MMP-9 and TIMP-1 were measured using commercially available ELISA kits (RPN2614 and RPN2618, respectively; Amersham Life Sciences, Arlington Heights, IL). The MMP-9 ELISA detects free proMMP-9 and proMMP-9/TIMP complexes. It does not detect active MMP-9. For MMP-9 measurements, BALF were concentrated 20× at 4° C with a low protein-binding concentrator (Centriprep; Amicon, Beverly, MA) with a molecular weight cutoff limit of 3 kD; for TIMP-1 measurements, BALF were not concentrated. The assay sensitivities were 0.6 ng/ml for MMP-9 and 1.25 ng/ml for TIMP-1.

Zymography

The gelatinase activity of BALF was analyzed by zymography (15). BALF (20×) was diluted 1:5 in 6× concentrated nonreducing sample buffer, incubated at room temperature for 10 min, and separated by electrophoresis at 120 V on 7.5 or 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels containing 0.3% gelatin. The gels were incubated for 30 min in renaturing buffer containing 2.5% Triton ×100 to remove the SDS. Gels were then incubated overnight at 37° C in developing buffer containing 50 mM Tris-HCl, pH 7.5, 10 mM CaCl₂, and 0.02% polyoxyethylene 23 lauryl ether (Brij 35) (Sigma Chemical Co., St. Louis, MO). Substrate degradation was visualized by staining the gels with 0.5% Coomassie Brilliant Blue R250 and destaining with 7.5% acetic acid with 5% methanol. The protein substrate incorporated into the gel stained blue, leaving transparent bands where substrate-degrading enzymes were present. In selected experiments, the proform of the enzyme was activated by incubation of BALF with 1 mM paraaminophenylmercuric acetate (APMA) (Sigma, St. Louis, MO) for 1 h at 37° C (15). The protease class was confirmed by incubating gel slices with 5 mM 1,10-phenoanthroline, a selective metalloproteinase inhibitor, or 2 mM phenylmethylsulfonyl fluoride (PMSF), a serine protease inhibitor. Positive controls included recombinant proMMP-9 and proMMP-2, and dimeric MMP-9 or MMP-9/lipocalin complexes purified from activated neutrophils (all controls were obtained from Calbiochem Corp., San Diego, CA).

Immunocytochemistry

Cytospin preparations of fresh BAL cells were prepared on precoated slides (Cytoslide; Shandon, Inc., Pittsburgh, PA). Cells were fixed in periodate-lysine-paraformaldehyde fixative (16) and MMP-9 was detected by monoclonal anti-human MMP-9 antibody (clone 4H3; R&D Systems, Inc., Minneapolis, MN) using alkaline phosphatase-antialkaline phosphatase (APAAP) (D651; Dako, Carpinteria, CA) with Fuchsin substrate (Dako). Slides were counterstained with Mayer's hematoxylin (Dako).

Statistical Analysis

Data were expressed as medians with 25 and 75 interquartiles. A Wilcoxon signed rank test (or paired t test, for normally distributed data) was used to compare data obtained 5 min and 48 h after SBP with SAL or AG. Correlations were made using Spearman rank order correlation. A p value of < 0.05 was considered significant. Statistical analysis was performed using a SigmaStat software package (Jandel Scientific Software, San Rafael, CA).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Cellular Profile and Protein Analysis of BALF

At 5 min, the cellular profile was similar between the SAL and AG segments. Compared with values in the immediate lavage samples, there was an increase in the percentage of eosinophils and neutrophils and a corresponding decrease in the proportion of alveolar macrophages 48 h after SBP with either SAL or AG (Table 2). The AG-challenged segment had a marked rise in the total number of cells and proportions of eosinophils compared with the SAL-challenged segment. Levels of total protein and albumin were elevated at 48 h and were significantly greater in the AG-challenged compared with the SAL-challenged segment (Table 2).

Characterization of MMP-9 and TIMP-1 in BALF

At 5 min, MMP-9 concentrations in BALF were low but detectable by ELISA in 15 and 17 subjects and no differences in MMP-9 existed between AG- and SAL-challenged segments. Forty-eight hours after SBP, the concentrations of MMP-9 were significantly elevated in BALF from the AG-challenged segment (Figure 1A). Levels of MMP-9 also increased, but to a lesser extent, in the SAL-challenged segment. Compared with MMP-9, concentrations of the inhibitor, TIMP-1, were higher at baseline and augmented in lavage fluid from both the SAL- and AG-challenged segments 48 h after the initial SBP (Figure 1B). There were no significant changes in serum levels of MMP-9 and TIMP-1 after challenge (data not shown). When data are expressed as the molar ratio of enzyme to inhibitor, AG challenge caused a significant increase in this ratio in the BALF, but not in the serum (Figure 2).

View larger version (28K): [in this window] [in a new window]   Figure 1.   Concentrations of MMP-9 and TIMP-1 in BALF after AG-SBP. (A) MMP-9 was measured in BALF obtained 5 min (imm) and 48 h (late) after SBP with SAL or AG. MMP-9 was measured in 20× concentrated BALF; data are expressed as the calculated amount of MMP-9 in 1× BALF. A cutoff of 0.03 ng/ml was used for these calculated values. (B) TIMP-1 was measured in 1× BALF obtained 5 min (imm) and 48 h (late) after SBP with SAL or AG. A cutoff of 0.06 ng/ml was used for these calculated values. Bars represent medians with 25 and 75 interquartiles. *p < 0.05, versus imm SAL; p < 0.05, late AG versus late SAL; p < 0.05, late AG versus imm AG.

View larger version (17K): [in this window] [in a new window]   Figure 2.   Ratio of TIMP-1 to MMP-9. The molar concentrations of MMP-9 and TIMP-1 was calculated based on molecular weights of 92 and 28 kD, respectively. Data are expressed as a ratio of MMP-9 to TIMP-1. *p < 0.05, versus imm.

Zymography was performed to determine the form of MMP-9 present in BALF and serum (Figure 3A). Little gelatinolytic activity was detected in BALF from the SAL- or AG-challenged segment at 5 min. In BALF obtained 48 h after SAL challenge, faint bands of enzymatic activity were seen at approximately 92 and 72 kD. In serum and in BALF obtained 48 h after AG challenge, a major band of enzymatic activity was present at 92 kD and faint bands of activity could also be detected at approximately 200, 130, 85, and 72 kD. These bands coincided with control samples containing MMP-9 dimers, MMP-9/lipocalin, activated MMP-9, and proMMP-2, respectively. To demonstrate that the proMMP-9 present in BALF could be activated, treatment with APMA was performed on a BAL sample obtained 48 h after AG-BAL or a proMMP-9 standard. After activation, a band appeared at approximately 85 kD, which is indicative of the intermediate active form of MMP-9. The gelatinolytic activity in BALF from the AG-challenged segment was not affected when the zymogram was incubated in the presence of the serine protease inhibitor, PMSF (Figure 3B), but was abolished in the presence of the metalloenzyme inhibitor, o-phenanthroline (Figure 3C). No enzymatic activity was detected on casein-infused gels, indicating that BALF does not contain substantial amounts of stromolysins.

View larger version (60K): [in this window] [in a new window]   Figure 3.   Analysis of BALF by zymography. Zymogram of BALF from a representative subjects. Each lane contains 25 µl of sample + 5 µl 6× sample buffer. Panel A is from a 10% SDS-PAGE containing gelatin. Samples are from left to right: serum obtained 48 h after AG-SBP, BALF samples obtained 5 min after SAL-SBP or AG-SBP (IMM), BALF samples obtained 48 h after SAL-SBP or AG-SBP (LATE), the 48 h AG sample treated with APMA, commercial preparations containing proMMP-9 treated with APMA, proMMP-9 dimer, a mixture of proMMP-2 and active MMP-2, and MMP-9/lipocalin complex. Panels B and C are 7.5% SDS-PAGE containing gelatin. The gels were treated with PMSF ( panel B) or phenoanthroline ( panel C) during the development for enzymatic activity. Samples include a mixture of proMMP-2 and proMMP-9 standards or BALF obtained 48 h after AG-SB. Panel D is a 10% SDS-PAGE containing casein. Samples include a mixture of proMMP-2 and proMMP-9 standards or BALF obtained 48 h after AG-SB.

MMP-9 Is Associated with BAL Cells

On cytospin preparations of BAL cells obtained 48 h after AG challenge, MMP-9 immunoreactivity appeared to be primarily associated with neutrophils (Figure 4). Concentrations of MMP-9 in BALF were significantly correlated with numbers of BAL neutrophils in all segments (SAL imm, r_s = 0.591, p = 0.02; AG imm, r_s = 0.500, p = 0.05; SAL late, r_s = 0.769, p < 0.001; AG late, r_s = 0.663, p = 0.003). Forty-eight hours after challenge, concentrations of MMP-9 also correlated with numbers of alveolar macrophages (SAL late r_s = 0.545, p = 0.03, AG late, r_s = 0.500, p = 0.04).

View larger version (128K): [in this window] [in a new window]   View larger version (124K): [in this window] [in a new window]   Figure 4.   Immunocytochemistry. BAL cells obtained 48 h after AG challenge and stained with anti-MMP-9 antibody (A) or isotype-control antibody (B).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In this study we have demonstrated that allergen exposure induces the release of MMP-9 into the airway. MMP-9 immunoreactivity was detected in association with neutrophils by immunocytochemical analysis of cytospin preparations of BAL cells and concentrations of MMP-9 in BALF correlated with numbers of BAL neutrophils, and to a lesser extent, alveolar macrophages. These data provide evidence that MMP-9 is generated as a consequence of allergen challenge, and is produced locally by airway inflammatory cells, probably neutrophils. The enhanced generation of MMP-9 can be associated with the initiation of bronchial injury, which may then lead to remodeling in asthma.

Using zymography on gelatin- and casein-infused gels, we have determined that the predominant MMP in BALF after AG challenge is MMP-9. Five bands with gelatinase activity are present in BALF 48 h after Ag challenge, a prominent band that comigrates with purified latent MMP-9 (92 kD), faint bands that comigrate with purified active MMP-9 (85 kD), purified latent MMP-2 (72 kD), MMP-9 homodimers (200 kD) and complexes of MMP-9 and a 25 kD neutrophil-derived protein, lipocalin (135 kD). The higher molecular weight complexes are commonly observed in association with MMP-9 (17). The lower molecular weight band is most likely MMP-2, which has been reported after AG challenge in a mouse model of airway inflammation (12). However, it could represent MMP-1, which has been reported in moderately severe asthmatics (18) or MMP-3, which is present in epithelial lining fluid from patients with status asthmaticus requiring mechanical ventilation (6). The lack of enzymatic activity when casein was used as the zymogram substrate, suggests that these enzymes are not present in large quantities in BALF after AG challenge.

To determine the cell source of MMP-9 after allergen challenge, we used immunocytochemistry to examine the expression of MMP-9 in freshly isolated BAL cells. MMP-9 immunoreactivity was primarily associated with neutrophils (Figure 4). Furthermore, significant correlations were found between concentrations of MMP-9 in BALF and airway neutrophils. There was a correlation between MMP-9 and numbers of alveolar macrophages. These results are similar to what has been reported in moderate to severe asthma (5). Although in vitro studies have demonstrated MMP-9 release from eosinophils (9, 19, 20), we did not observe MMP-9 expression in association with eosinophils on cytospins containing freshly prepared BAL cells, or in cell lysates or cell culture supernatant fluids from purified eosinophils (data not shown). It is of interest that Ohno and colleagues were able to detect MMP-9 mRNA, but not MMP-9 protein associated with eosinophils in mucosal biopsies from asthmatic subjects (9). These investigators suggested that the lack of immunostaining for MMP-9 protein is probably due to rapid release of MMP-9 from eosinophils. Further study is necessary to establish in vivo expression and release of MMP-9 by mucosal and BAL eosinophils and to determine if resident airway cells such as epithelial cells, endothelial cells, and fibroblasts may also contribute to MMP-9 in the airway. Nonetheless, the highly significant correlation between MMP-9 and neutrophils in BALF and the immunocytochemistry demonstration of MMP-9 staining in association with neutrophils strongly suggest that airway neutrophils are a primary source of MMP-9 in these subjects. These observations raise the possibility that neutrophils contribute to airway injury in asthma, as has been suggested by studies showing persistent bronchial neutrophilia in severe asthma and status asthmaticus (21).

The functional significance of proMMP-9 in the BALF remains to be established; however, there is growing evidence that MMP-9 may be crucial for the migration of airway inflammatory cells and may play a role in airway remodeling. In a murine model, inhibition of MMP-9 by in vivo administration of TIMPs or a synthetic MMP inhibitor decreased antigen- induced airway inflammation and airway hyperresponsiveness (12). In vitro studies have shown that MMP-9 is required for neutrophil (24), eosinophil (19), and T-cell migration (25) through synthetic matrix.

In summary, we have used SBP with AG as a model to study events associated with the initiation of allergen-induced airway inflammation, injury, and possibly the transition to remodeling. We have demonstrated that an airway allergen challenge induced MMP-9 and TIMP-1 release. Moreover, the concentrations of MMP-9 correlate with numbers of airway neutrophils. Extrapolating this model of allergen-induced airway inflammation to known observations in active asthma, our findings support the hypothesis that antigen may contribute not only to inflammation, but possibly to eventual airway remodeling through the generation of MMPs and TIMPs.