INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In asthma an association is found between bronchial obstruction, inflammation, healing, and repair process (1). Indeed, pathological studies show alterations of the bronchial structures including the thickening of basement membrane by sub-basement membrane collagen deposition (2), smooth muscle hypertrophy/hyperplasia, epithelial damage, occlusion of airways by mucus secretions, and airway inflammatory cell infiltration (3). Clinical and radiological findings show that in some groups of asthmatics, the remodeling of the airways leads to permanent bronchial obstruction. There is evidence that these changes are driven by cellular mediator release in a situation of chronic airway inflammation. Eosinophils, mast cells, lymphocytes, and to a lesser extent macrophages are in increased number in allergic as well as nonallergic asthma. They were shown to secrete inflammatory and noninflammatory products that play a role in inflammation and healing (4, 5).

The hypothesis that the clinical heterogeneity of asthma in terms of disease severity is reflected in the degree of airway inflammation emerged as early as 1990 (6). However, limited information is available comparing the inflammatory response of severe asthma with milder forms of the disease. Recent studies of Synek and coworkers (7) pointed out that in fatal asthma, there is a redistribution of CD3⁺ T cells away from the epithelium and proximal enhancement of the eosinophil inflammatory infiltrate when compared with mild and moderate asthma. However, other investigators (8) proposed that sudden-onset fatal asthma is characterized by a relative paucity of eosinophils in the face of an excess of neutrophils in the airway submucosa. Also, Fahy and coworkers (9) demonstrated that neutrophils predominate more frequently than eosinophils in sputum from patients with acute exacerbation of asthma. They speculated that the neutrophilia may be related to respiratory tract infections, a frequent precipitant of acute asthma. Thus, although the eosinophil is the major effector cell in mild or moderate asthma, other inflammatory cells such as neutrophils may be critical in the perpetuation and aggravation of airway inflammation and possibly bronchial destruction during episodes of status asthmaticus (SA).

Indeed, a wide array of powerful agents (oxygen derivatives, serine proteinases, and zinc matrix metalloproteinases [MMPs]) stored in granules may be released by transmigrating neutrophils. More particularly, MMPs form a group of neutral proteinases able to degrade extracellular matrix components. Among MMPs secreted from neutrophils may be interstitial collagenase (MMP-8) initiating the cleavage of the triple helix of native collagen types I, II, III, and X, and 92 kDa gelatinase (MMP-9) (10) cleaving native types IV and V collagens, and to a lesser extent, fibronectin, laminin, entactin, and insoluble elastin. Thus, with regard to collagen type IV, laminin, entactin, and fibronectin, 92 kDa gelatinase has specific affinity for the subepithelial basal lamina, a specialized nonfibrillar connective tissue structure that anchors epithelial cells to parenchymal surfaces. Other inflammatory cells, all involved in asthma, such as macrophages (11), lymphocytes (12), mast cells, or eosinophils (13) are able to express 92 kDa gelatinase. We recently demonstrated that human bronchial epithelial cells were able to produce major constitutive 92 kDa gelatinase and upregulate its production in response to inflammatory injury (14, 15), thus playing a role in the physiopathological remodeling of airways.

The aim of this work was to investigate the local production of 92 kDa gelatinase in bronchial lavages of patients with SA characterized by prominent neutrophilic inflammation and bronchial cell desquamation, as compared with patients with mild asthma, in order to assess inflammatory features related to severe asthma. We also investigated the possible regulation of 92 kDa gelatinase both through activation process by stromelysin-1 and inhibition process by tissue inhibitor of metalloproteinase-1 (TIMP-1).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patients

Six patients with SA (age 32 ± 4 yr [mean ± SEM]) were included in this study. They all had previous history of asthma. They were admitted to the intensive care unit (ICU, Hôpital André Calmette, Lille, France) between January 1995 and May 1996, with respiratory failure resulting from severe asthma attack requiring intubation and the onset of mechanical ventilation (MV). Particular attention was focused to obtain precise information from the patient and/or from a witness, concerning the probable trigger of the final asthma attack (head or chest cold, exposure to aeroallergen or nonspecific irritants, exercise, psychological conflicts). One was a current smoker. Moreover, the time elapsed between the onset of the first symptoms and respiratory failure was estimated: three patients with SA exhibited sudden-onset SA (interval time < 3 h) and three progressive-onset SA (interval time > 3 h). Hospital treatment was conducted according to the current American Thoracic Society (ATS) recommendations (14) and consisted of corticosteroids (methylprednisolone 2 mg/kg/d), ₂-agonists (salbutamol 5 to 20 µg/kg/d), and continuous sedation (benzodiazepine and neuromuscular blocking agent when required). MV was carried out according to the procedure of controlled hypoventilation (low inspiratory time, low respiratory frequency, and reduced tidal volume) in order to obtain the largest ventilation for a maximal airway pressure of 50 cm H₂O. Fraction of inspired oxygen (FI_O2) was adapted to achieve Pa_O2 > 60 mm Hg.

Four control subjects (vC) without preexisting respiratory disease and requiring mechanical ventilation (age 41 ± 4 yr [mean ± SEM]) were also studied to evaluate the effects of MV on the bronchial cellular and molecular events. These patients were admitted to the ICU for drug-related suicide attempts (benzodiazepine) and required short-lasting MV because of severe alveolar hypoventilation. No additional treatment was administered.

Six patients with asthma (MA) (mean age 47 ± 3 yr) were also investigated. Asthma was defined according to the criteria of the ATS (15). All patients had either a reversible airway obstruction characterized by a 20% increase in forced expiratory volume in one second (FEV₁) after the inhalation of 200 µg of albuterol or bronchial hyperreactivity to methacholine (PC₂₀ < 8 mg/ml). Exclusion criteria were: oral or inhaled steroid use; history of an upper respiratory tract infection in the previous 6 wk; severe exacerbation of asthma requiring hospitalization in the 6 wk preceding the study; and tobacco use within the past year or greater than 10 pack-years total smoking history. Treatment with ₂-agonists was withheld for 8 h before bronchoscopy.

The control nonventilated group included 10 healthy volunteers (age 42 ± 5 yr) with pulmonary function within the normal range. They had no history of lung disease or cigarette smoking. The study was approved by the local committee of the hospital (No. 9307).

Fiberoptic Bronchoscopy and Bronchial Lavage (BL)

BL was performed in ventilated patients when it was expected to provide therapeutic benefit. Thus, BL were performed in case of characterized atelectasis (in groups SA and MV) and/or to remedy mucus impaction in patients with refractory SA despite optimal medical treatment. BL samples from patients with parenchymal opacities consistent with pneumonia on chest radiography were excluded. Fiberoptic bronchoschopy was performed through an adaptor on the endotracheal tube designed to minimize air leak (model 514900; Rfsch AG, Kernen, Germany). The FI_O2 was adjusted to 1.0 for 5 to 15 min before and throughout the procedure. The tip of the bronchoscope was wedged into different segmental bronchi to remove difuse mucus impaction in patients with atelectasis, or in the relevant segmental bronchus in patients with atelectasis. The lavage was performed by infusion of two or three 15-ml aliquots of sterile 0.9% saline solution at room temperature. Each aliquot was immediately gently aspired and the various fractions were pooled. The patient was monitored closely while receiving an FI_O2 of 1.0 for an additional 10 min. The FI_O2 was then adjusted to the prebronchoscopy level as tolerated, using continuous pulse oximetry and arterial blood gases. Finally, for each patient with SA, the various BL fractions related to different segmental bronchi were pooled for MMP investigation. An aliquot (5 ml) of BL was used for quantitative and qualitative bacterial cultures (the diagnosis of lower respiratory tract infection [LRTI] required 10⁶ colony-forming units per milliliter [cfu/ml]). LRTI was detected in two patients with SA (Proteus mirabilis 10⁷ cfu/ml for one patient, and Streptococcus pneumoniae 10⁶ cfu/ml for the other). All the other BL were sterile.

In healthy patients and asthmatics, fiberoptic bronchoscopy was performed after local anesthesia with lidocaine 2% applied to the upper respiratory tract. BL was performed into a segmental bronchus of the right middle lobe by slow infusion of two 15-ml aliquots of sterile 0.9% saline solution. Each aliquot was immediately aspired using a hand-held syringe and the different fractions were pooled. During bronchoscopy, oxygen was readily available. Nebulization with salbutamol was performed after the procedure if bronchospasm was noted.

BL were separated into their acellular component (BL fluid) and cellular component, by centrifugation at 400 × g for 10 min at 4° C. BL and BL fluids were aliquoted and frozen at 80° C until use. Total cell counts were performed on a small aliquot of BL using a hematocytometer. Cells were prepared with a cytocentrifuge and stained with May-Grünwald-Giemsa to allow the differential cell count. At least 400 cells on each slide were counted in each BL specimen to determine the differential count and were read by two investigators blinded to the clinical details of the patients. Average cell differentials of the two investigators were reported.

The lavage fluid volumes were normalized to the volume of epithelial lining fluid (ELF). ELF volumes were determined by the urea method (16) and calculated as:

Thus, the concentration of X component referenced to ELF was calculated as:

One milliliter of BL fluids from asthmatic and healthy patients was lyophilized to dryness, reconstituted in distilled water (1/100 of initial volume), aliquoted and stored at 80° C until albumin and urea assays. Other aliquots were immediately stored at 80° C until gelatinase and TIMP-1 analysis.

Urea, Albumin, and Myeloperoxidase Assays

Urea concentration was determined in BL by spectrophotometric absorbance at 600 nm using Berthelot reaction (diagnostic kit; Boehringer Mannheim). Albumin measurement in concentrated BL was determined by spectrophotometric absorbance at 628 nm, by using Bromocresol green (diagnostic kit; Sigma, St. Louis, MO).

Myeloperoxidase (MPO) content in lavage fluid samples was evaluated by immunoenzymatic dosage (MPO-EIA Diagnostic kit; Oxis International SA, Bonneuil sur Marne, France). Briefly, samples (BL or MPO standard solution) were incubated in wells of microplate recovered by anti-MPO monoclonal antibody. The presence of MPO was revealed by a second antibody (polyclonal of goat) marked with the biotin. The final stage was an amplification by a coupling biotin- avidin, in which the avidin is linked to the alkaline phosphatase. The concentration of MPO was determined after enzymatic revelation with para-nitrophenyl-phosphate (pNPP) as substrate, at 405 nm.

Zymography

Aliquots of BL fluids underwent electrophoresis in polyacrylamide gels containing 1 mg/ml gelatin or  casein, in the presence of sodium dodecyl sulfate (SDS-PAGE) under nonreducing conditions. After electrophoresis, gels were washed twice in 2.5% Triton X 100 for 1 h to remove SDS, rinsed briefly and incubated at 37° C for 24 h in reaction buffer (100 mM Tris-HCl, 10 mM CaCl₂, pH 7.4). After staining with Coomassie Brilliant Blue R-250, gelatin-degrading enzymes were identified as clear zones of lysis against a blue background. Molecular weights of gelatinolytic bands were estimated using prestained molecular weight markers.

Activities in the gel slabs were quantified using semiautomated image analysis (NIH Image 1.52), which quantifies both the surface and the intensity of lysis bands after scanning of the gels. Results are expressed as arbitrary units/24 h/10 µl ELF. To check that this method for measuring enzymatic activity on zymograms was linear over the range of activities in unknown samples, we evaluated activities for increasing volumes of biological fluids and found that arbitrary units obtained with the image analysis system increased linearly with the volume of the samples (r = 1.00) (17).

The pattern of proteinase inhibition was investigated by adding one of the following to the incubation buffer: 2 mM phenylmethylsulfonyl fluoride (PMSF) (final concentration) as a serine proteinase inhibitor, 2 mM N-ethylmaleimide (NEM) as a cysteine proteinase inhibitor, or 10 mM ethylenediaminetetraacetic acid (EDTA) as a metalloproteinase inhibitor.

Proteinase Assays on Radiolabeled Substrates

Free gelatinase and stromelysin activities were assayed using radiolabeled gelatin and  casein as respective substrates. Gelatin and  casein were radiolabeled with ³H-acetic anhydride according to Cawston and Barett (18). Specific activities were 880 and 920 kBq/mg respectively. To measure the free forms of proteinases in the presence of 50 µg of acetylated ³H-gelatin, aliquots of BL fluids were tested with or without 1 mM aminophenylmercuric acetate (APMA) (incubation at 37° C for 2 h). The proteolytic reaction was allowed to proceed for 48 h at 37° C and pH 7.4 in the presence of toluene to prevent bacterial contamination. Proteinase assays were performed in presence or not of EDTA as metal chelator, as described previously (19).

Reverse Zymography

Tissue inhibitors of metalloproteinases (TIMPs) secreted into BL fluids may be detected using reverse zymography (20). Briefly, aliquots of samples were resolved by 11.5% SDS-PAGE in the presence of 1 mg/ml gelatin. The standard zymographic method was modified after the removal of SDS from the gel, by incubating the gel for 1 h at 37° C in the conditioned medium issued from phorbol myristate acetate (PMA)-activated rabbit skin fibroblasts. This medium provides a source of activated gelatinases able to degrade extensively gelatin in the gel. Then the gel was incubated in the reaction buffer as previously described and stained/destained as in standard zymography. Protection of the gelatin in the gel by the presence of TIMPs led to the appearance of relatively dark bands against a lighter background. Recombinant TIMP-1 was used as reference.

Immunological Assays

TIMP-1 and 92 kDa gelatinase contents in some BL were quantified by competitive-binding, indirect, enzyme-linked immunosorbent assays (ELISA) as described previously (21). These assays had a sensitivity of  10 ng/ml and 1 ng/ml for gelatinase and TIMP-1, respectively. They measured (1) total enzyme present, whether free or bound to inhibitor or substrate, or whether in inactive or active form, and (2) total inhibitor. Standard curves were included in each assay using either purified 92 kDa gelatinase (Valbiotech, Paris, France) or recombinant TIMP-1 (generous gift from Synergen). Incubation with diluted 1:50 antiserum against human 92 kDa gelatinase (Valbiotech) was carried out for 1 h at room temperature. The wells were washed four times with Tris-buffered saline (TBS), 0.05% Tween 20, and incubated for 1 h with alkaline phosphatase goat anti-rabbit IgG 1:1,500 as the secondary antibody. The degradation of p-nitrophenyl phosphate disodium as substrate was measured at 405 nm.

Aliquots of dialyzed and 20-fold concentrated BL were separated by SDS-PAGE, and transferred to an Immobilon-P filter (polyvinylidene difluoride, 0.45 µm). Nonspecific staining was blocked by incubating the transfers for 90 min in TBS containing 5% nonfat dry milk. The transfers were then incubated overnight with rabbit polyclonal antiserum against human stromelysin-1 (MMP-3) (Valbiotech) diluted 1:500 in TBS. The blots were washed three times in TBS, 0.05% Tween 20 and incubated for 90 min with biotinylated goat anti-rabbit IgG diluted 1:1,000 as the secondary antibody. The blots were visualized using alkaline phosphatase and Fast red TR/naphthol AS-MX (from tablet sets; Sigma).

Degradation Products of Laminin

The determination of laminin degradation products in BL was performed by radioimmunoasay using a commercial kit (RIA-gnost laminin P1; Cis-Bio, France) with an antibody directed against the pepsin-resistant fragment P1 of laminin, as previously described (22). Briefly, the specific rabbit antibody was first incubated with the nonlabeled laminin fragment (standard or sample) for 24 h at 4° C. The ¹²⁵I-laminin fragment was then added for a 24 h incubation at 4° C. After the addition of the second antibody, goat anti-rabbit IgG, and incubation for 2 h at room temperature, the tubes were centrifuged and the radioactivity of the precipitates was measured in a gamma counter. Nonspecific binding was evaluated by replacing specific immune serum with normal rabbit serum. The reproducibility of the assays was below 10%. The observed value for healthy adults in serum was approximately 1.6 U/ml.

Statistical Analysis

Values were expressed as means ± SEM. Comparison of asthmatics and patients with SA was performed using the nonparametric Mann-Whitney U test. Linear regression using Spearman's rank test took into account BL samples from all patients in each group (thus corresponding to a total number of 26 subjects) and was carried out to evaluate relationships between BL gelatinase activity and albumin or cellularity.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Cellularity

Keeping in mind that the BL recovered fluid was similar in all groups (approximately 20 ± 2 ml), the total cell number recovered per ml BL from ventilated healthy patients exhibited a 1.8-fold increase of the total cell number compared with the group of nonventilated healthy patients (Table 1). By contrast, the total cell number recovered from patients with mild asthma remained unchanged compared with nonventilated healthy patients, whereas it was about 5-fold increased in patients with SA, as compared with ventilated healthy patients.

Cell differential counts showed that alveolar macrophages were the predominant cells (about 75%) in nonventilated healthy and asthmatic patients, whereas polymorphonuclear leukocytes (PMNs) became the predominant cells (82%) in patients with SA. In ventilated healthy patients, both alveolar macrophages and PMNs were the predominant cells. The percentage of shedded bronchial epithelial cells increased approximately 1.6-fold in patients with mild asthma or with SA versus nonventilated and ventilated healthy patients respectively. The number of eosinophils was increased approximately 7-fold in asthmatic versus all healthy patients and 2.4-fold in SA versus asthmatic patients.

Variations of ELF and Albumin

ELF volume was equivalent in asthmatic and all heathy patients (1.7 ± 0.8 ml versus 1.9 ± 0.7 ml/100 ml BL). In contrast, ELF volume increased about 5-fold in BL from SA patients compared with healthy or asthmatic patients (9.1 ± 1.5 ml/100 ml BL). High concentration of albumin in most of the BL from patients with SA was shown as compared with healthy and asthmatic patients (Figure 1).

View larger version (31K): [in this window] [in a new window]   Figure 1.   Variations of albumin concentration in BL. Albumin was evaluated by spectrophotometric absorbance at 628 nm, by using Bromocresol green (diagnostic kit; Sigma), in concentrated BL from nonventilated control healthy subjects (C), ventilated healthy subjects (vC), patients with mild asthma (MA), and patients with status asthmaticus (SA).

Zymography

Zymography on SDS-gelatin was used to determine the levels of gelatinase activity released in the airways of healthy patients, asthmatics, and patients with SA. The concentration of 92 kDa gelatinase in BL of patients with mild asthma was very low and did not differ from that currently observed in all healthy patients. In contrast, the level of 92 kDa gelatinase was acutely increased in BL of patients with SA (Figure 2A). Semi-automated analysis showed that its latent form (AU gelatinase/48 h/ELF) was enhanced 10- to 160-fold compared with healthy or asthmatic patients (Figure 2B).

View larger version (67K): [in this window] [in a new window]   View larger version (19K): [in this window] [in a new window]   Figure 2.   Gelatin zymography of BL from the different groups of patients. (A) Gelatin zymograms. Each aliquot was subjected to SDS-PAGE/gelatin electrophoresis. Lanes 1-3, undiluted BL from ventilated control patients (vC); lanes 4-5, undiluted BL from nonventilated control patients (C); lanes 6-7, undiluted BL from asthmatic patients with mild asthma (MA); lanes 8-12, BL (diluted 1:100) from patients with SA. Arrows indicate the positions of 200, 135, 92, and 72 kDa gelatinase activities. Prestained molecular mass markers (MW) are in the first lane from the right. (B) Quantitative evaluation of BL 92 kDa gelatinase activities on gelatin zymogram. Quantitative image analysis of the surface and intensity of lysis bands was carried out. Results are expressed as arbitrary units/48 h/100 µl ELF. BL were issued from control nonventilated healthy subjects (C), ventilated healthy subjects (vC), patients with mild asthma (MA), and patients with SA.

The pattern of secreted gelatinase was often closely distributed between the major latent form (92 kDa), the lipocalin-associated form (135 kDa), the probably dimeric form (200 kDa), and some active form (84 kDa) and thus appeared representative of neutrophil 92 kDa gelatinase pattern. Interestingly, no 72 kDa gelatinase activity was evidenced in any patient with SA. EDTA completely inhibited the activity of all forms of gelatinases whereas phenylmethylsulfonyl fluoride and N-ethylmaleimide did not (data not shown).

Zymography on SDS- casein was used to investigate the stromelysin activity released in the airways of healthy, asthmatic, and SA patients. Caseinolytic activity was never detected in 20-fold concentrated BL from healthy and asthmatic patients. In contrast, one minor 60 kDa and two major 46 and 26 kDa caseinolytic bands were clearly visualized in concentrated BL from patients with SA, possibly related to the proform and active forms of stromelysin-1 (MMP-3), respectively (Figure 3). EDTA completely inhibited the activity of all caseinase bands.

View larger version (55K): [in this window] [in a new window]   Figure 3.   Casein zymography of BL of patients with SA and identification of caseinases as metalloproteinases. Each 20-fold concentrated aliquot was subjected to SDS-PAGE/ casein electrophoresis. Casein substrate gels were then incubated in 100 mM Tris incubation buffer without inhibitor (lanes 1-3), or in the presence of 10 mM EDTA (lanes 4-6), 2 mM AEBSF (lanes 7-9) or 2 mM NEM (lanes 10-12). Arrows indicate the positions of 60, 46, and 26 kDa caseinase activities. Prestained molecular mass markers (MW) are in the left of the figure.

Proteinase Assays on Radiolabeled Substrates

Using ³H-gelatin or ³H- casein, no free gelatinolytic or caseinolytic activity was detected in BL from healthy and asthmatic patients either in absence or in presence of 1 mM APMA (Figure 4A). This result demonstrated that, under these conditions, the presence of metalloproteinase inhibitors such as TIMPs was sufficient to prevent free forms of activated gelatinases or stromelysins. However, when the gelatinase assays were performed in BL of patients with SA, low level of free metallogelatinolytic activity (1-5 µg of gelatin hydrolyzed per 48 h/ml ELF) was demonstrated in some samples. In addition, the presence of 1 mM APMA in BL of patients with SA amplified the free gelatinolytic response (2-25 µg of gelatin hydrolyzed per 48 h/ml ELF).

View larger version (30K): [in this window] [in a new window]   Figure 4.   Proteinase assays. BL from patients were investigated for free gelatinolytic activity (A) or for free caseinolytic activity (B). Results were expressed as 3H cpm gelatin or casein peptides released per 48 h/100 µl ELF (specific gelatin and casein activities = 880 and 920 kBq, respectively). Bars = SD (p < 0.01). Assays were performed with (black histogram) or without (white histogram) prior proteinase activation by 1 mM APMA.

Using ³H-casein, a net free metallocaseinolytic activity (2-50 µg of casein hydrolyzed per 48 h/ml ELF) was demonstrated in all investigated samples, whereas only 20% enhancement of this activity was observed in presence of APMA (Figure 4B).

Reverse Zymography

In accordance with the preceding suggestion, high content of TIMP-1 was readily demonstrated by reverse zymography in several BL from patients with SA whereas TIMP-1 was undetectable in BL of healthy and asthmatic patients (Figure 5). Indeed, such a high content of TIMP-1 may partially counterbalance excess of activated 92 kDa gelatinase or regulate somewhat 92 kDa gelatinase activation. TIMP-2 or 3 (21 and 24 kDa) was undetectable in BL of any group.

View larger version (35K): [in this window] [in a new window]   Figure 5.   Reverse zymography of BL. Aliquots of BL were subjected to 11.5% SDS-PAGE/gelatin electrophoresis. Lane 1, recombinant TIMP-1; lanes 2-3, BL from ventilated control patients (vC); lanes 4-5, BL from nonventilated control patients (nvC); lanes 6-9, BL from mild asthma (MA); lanes 10-13, BL from patients with SA. Arrows indicate position of molecular mass of recombinant TIMP-1 (21 kDa) and physiologic TIMP-1 (30 kDa).

Immunologic Studies

Immunologic studies performed by ELISA on 12 samples readily showed that the increase of TIMP-1 concentration in BL of patients with SA was less than the increase of 92 kDa gelatinase, thus resulting in a 5-fold elevated ratio between gelatinase and TIMP-1 in BL of patients with SA compared with healthy and asthmatic patients.

Immunoblotting analysis was performed to identify stromelysin subtype. Both purified human stromelysin-1 used as positive control and dialyzed concentrated BL from patients with SA were recognized by antibody against human stromelysin-1 (Figure 6). The 60 kDa proform as well as smaller bands (about 45 and 24 kDa) corresponding to autoactivation or degradation active products were recognized by the antibody. No response was observed with the same membrane using preimmune serum (negative control).

View larger version (59K): [in this window] [in a new window]   Figure 6.   Western blot analysis. Aliquots of 20-fold concentrated BL from patients were analyzed by immunoblotting under reduced conditions, with rabbit polyclonal antiserum against MMP-3, to define the matrix metalloproteinase property of 60-46-26 kDa caseinases. Lane 1, 200 ng of purified stromelysin-1 (MMP-3) as reference; lanes 2-5, 10 µl of dialyzed 20-fold concentrated BL from four patients with SA. The specificity of the 60, 46, and 26 kDa bands recognized by the immune serum was confirmed by the negative result obtained with preimmune serum; negative result was also obtained in BL from asthmatic and healthy control subjects (data not shown). Arrows indicate the migration of molecular mass standard.

Degradation Products of Laminin

Even if laminin represents a minor substrate for 92 kDa gelatinase, we investigated the putative destruction of basement membrane underlying the bronchial epithelium, by evaluating the degradation products of laminin. These products were undetectable in dialyzed 20-fold concentrated BL from healthy and asthmatic patients. In contrast, their presence (0.36 ± 0.10 U/ml BL) was clearly detected in the airspaces of 60% of patients with SA.

Correlation Studies

Levels of 92 kDa gelatinase activity evaluated by zymography were strongly correlated with albumin levels in bronchial airways (r = 0.86 and p < 0.0001) (Figure 7). Similar correlation was obtained with free metallogelatinolytic activity (r = 0.80, p < 0.0001). A weaker yet significant correlation (p = 0.042) was evidenced between free metallogelatinolytic activity and the level of laminin degradation products.

View larger version (11K): [in this window] [in a new window]   Figure 7.   Correlation of 92 kDa gelatinolytic activity in BL with albumin concentration. Linear regression was used for statistical analysis; r and p values were 0.95 and 0.0001, respectively.

A strong and significant correlation was also found between the 92 kDa gelatinase activity and the number of neutrophils (r = 0.89, p = 0.0016) (Figure 8A) and the number of shedded bronchial epithelial cells (r = 0.87, p < 0.0001) (Figure 8B).

View larger version (16K): [in this window] [in a new window]   Figure 8.   Correlation of 92 kDa gelatinase activity with: A, the total number of neutrophils, B, the total number of bronchial epithelial cells (BEC) in BL from all patients (n = 26). Linear regression was used for statistical analysis; r and p values were 0.89 and 0.001, respectively (A), and 0.87 and 0.001, respectively (B).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The indications, efficacy, and safety of BL obtained during bronchoscopy performed in patients under mechanichal ventilation for SA, were largely explained and discussed in a concomitant report (31). Briefly, bronchoscopy and BL sampling were performed only in case of refractory SA in spite of optimal medical treatment, in order to remedy possible mucus impaction or in case of atelectasis. No severe complication was noticed and clinical improvement was obtained in 43% of cases.

The presence of 92 kDa gelatinase in bronchoalveolar or bronchial lavages issued from healthy patients or patients with various lung diseases was previously shown in some studies (23). More precisely, the excessive release of this enzyme uncounterbalanced by specific inhibitor TIMP-1 in expectorations from children with cystic fibrosis was proposed to participate in airway damage (24). In contrast, 92 kDa gelatinase appeared unmodified or little modified during the course of adult respiratory distress syndrome, whereas 72 kDa gelatinase was activated, allowing us to propose a specific role for the latter in this severe inflammatory lung disease (25). To our knowledge, the possible involvement of gelatinases during acute severe asthma has not been investigated thus far.

Our results clearly demonstrate that BL issued from patients with SA contain acutely elevated levels of 92 kDa gelatinolytic enzyme, compared with healthy patients or patients with mild asthma. A number of findings confirm that this enzyme is a member of the MMP family: (1) it was secreted as a major 92 kDa proenzyme that was activated by organomercurials such as APMA, and (2), its activity was inhibited by chelators such as EDTA, but not by PMSF or NEM. However, the highly elevated levels of 92 kDa gelatinase activity in BL from patients with SA were associated with low level of free metallogelatinolytic activity. This latter was compatible with the high level of TIMP-1 detected by reverse zymography and with the relatively low level of activated forms (88/84 kDa) of 92 kDa gelatinase detected by zymography. Nevertheless, the approximately 5-fold enhanced ratio evaluated by ELISA between 92 kDa gelatinase and TIMP-1 in favor of gelatinase, as well as the net increase of free gelatinase after APMA activation strongly suggests that the amount of TIMP-1 is not sufficient to inhibit the large excess of enzyme in BL from patients with SA.

The elevated level of TIMP-1 may represent some control regulation of the excessive and acute release of 92 kDa gelatinase, whereas the low level of enzyme-activated forms does not preclude that activated forms may interact in situ with matrix components of injured subepithelial basement membrane. Indeed, as proposed by Fahy and coworkers (9) concerning the low level of free leukocyte elastase activity, the pertinent concentrations of free gelatinolytic activity for its matrix remodeling effects are those in bronchial connective tissues. The distance between these sites and the airway lumen could result in significant differences between the free gelatinase detectable in the airway luminal secretions and its concentration at its sites of action.

Concerning the highly probable enhanced 92 kDa gelatinase activation in situ, it is of interest to note that activated forms of stromelysin-1 were readily identified and characterized by zymography, inhibition pattern, immunoblotting, and caseinase assays against radiolabeled  casein, only in BL from patients with SA. The fact that stromelysin-1 was mainly found in airspaces as its activated forms, is in accordance with the previous concept that stromelysin-1 would not be secreted from cells as a proform but rather as activated forms (26). Because activated stromelysin-1 has been also proposed to be involved in the activation process of 92 kDa gelatinase (27), it is highly probable that this activation route is efficient in lung tissues during SA.

In any event, severe increase of 92 kDa gelatinase in bronchial airways during acute asthma may contribute to the excessive bronchial permeability which is readily demonstrated by the net increase of ELF and albumin in the BL of patients with SA. Indeed, the strong and significant correlation (r = 0.95, p < 0.0001) between 92 kDa gelatinase and albumin levels supports the hypothesis that acute enzyme activity may participate in the augmentation of bronchial permeability. A possible mechanism involved in the latter may be the shedding of bronchial epithelial cells from the basement membrane, by degrading both type IV and VII collagens. Type VII collagen is the major structural component of the anchoring fibrils that are critical for epidermal adhesion in the basement membrane zone. In addition, stimulated 92 kDa gelatinase production and/or secretion in situ may interfere with the degradation of type XVII collagen (28), a 18 kDa large extracellular and collagenous portion of transmembrane protein located in the hemidesmosomes of basal bronchial epithelial cells, thus promoting cell- matrix disruption and detachment of epithelial cells. Indeed, the latter is readily amplified in BL of patients with SA and the percentage of detached bronchial epithelial cells recovered in some BL may reach 40%, and even reached 74% in one patient.

Moreover, our results demonstrate that degradation products of laminin are readily detected only in airspaces of patients with SA, thus indicating some alteration of basement membrane underlying the bronchial epithelium. Even if laminin is not the preferential matrix substrate for 92 kDa gelatinase, the high and significant correlation (r = 0.70, p = 0.001) between the secretion of laminin degradation products and free 92 kDa gelatinase activity supports the hypothesis that 92 kDa gelatinase may contribute, among other proteinases such as stromelysin-1 and leukocyte elastase, to the active remodeling of bronchial subepithelial basement membrane.

Interestingly, among gelatinases, only increased production and/or secretion of 92 kDa gelatinase was demonstrated in BL of patients with SA. Level of 72 kDa gelatinase remained very low or undetectable, whatever the groups of patients. This result means that some specificity of gelatinase response in airways may reflect differences in major and preferential cells involved during status asthmaticus. Indeed, although the substrate specificities of 72 and 92 kDa gelatinases seem similar, the two enzymes are known to be synthesized by different cells in vitro or ex vivo. The 72 kDa form is synthesized by mesenchymal and endothelial cells and osteoblasts, whereas the 92 kDa form is produced mainly by inflammatory cells including PMNs, macrophages, eosinophils, and lymphocytes and by various tumor cells and normal cells as osteoclasts and keratinocytes. Moreover, our recent study demonstrated that human epithelial bronchial cells were able to express major constitutive 92 kDa gelatinase (29). Thus, the question was raised about the cellular origin of the 92 kDa acutely released in BL from patients with SA.

Neutrophil origin is supported by several arguments:

1. The 92 kDa gelatinase zymography data frequently resemble neutrophil 92 kDa gelatinase pattern, i.e., about 200 kDa (dimeric form of the latent form), 135 kDa (lipocalin-92 kDa gelatinase complex, 92 kDa (the latent form), and 88 kDa (the activated form).

2. The number of neutrophils is acutely increased in airways of patients with SA. Indeed, not only the cell differential count clearly demonstrates that neutrophils became the predominent cells (81 ± 5% versus 6 ± 4% in healthy and asthmatic patients), but the total number of airway cells is enhanced approximately 10-fold when reported to BL unit volume. Even if LRTI was suspected to be the trigger of the asthma attack in two patients, the elevated number of neutrophils observed in our study does not appear to be related to respiratory tract infection, because cellular components of BL with positive bacteriological cultures did not differ from all other sterile BL. Neutrophilia can be assigned partly to MV because the latter is accompanied by net attraction of neutrophils (44 ± 24%) in airways of ventilated healthy patients. However, in the latter, a smaller increase of total number of airway cells was observed. Also some attraction of neutrophils in airways (22 ± 13%) was recently observed in reference control patients for which BAL was performed for suspected ventilator-associated pneumonia (25). Nevertheless, our results clearly demonstrate that MV-induced neutrophil influx is not associated with enhancement of secreted 92 kDa gelatinase in airways, thus allowing us to exclude the contribution of MV-induced neutrophil recruitment to the acute secretion of 92 kDa gelatinase during SA. Other researchers have recently described airway neutrophilia either in sudden and fatal asthma (8) or during asthma exacerbation (9), and have proposed some explanations for the possible mechanism of recruitment of neutrophils. One of these might be the high concentrations of interleukin-8 (IL-8) (9) liable to be secreted from epithelial cells and to mediate chemoattraction of neutrophils to airways (30). Indeed, 4-fold augmentation of IL-8 was reported in our population of SA patients when compared with healthy subjects or patients with mild asthma (31).

3. The migration of neutrophils through basement membrane needs 92 kDa gelatinase. We recently demonstrated that 92 kDa gelatinase constitutes a major factor of neutrophil migration across basement membrane and that elastase may also contribute to this process by activating the proform 92 kDa gelatinase (32). A 100-fold increase of total neutrophil elastase was readily evidenced in BL from patients with SA when compared with healthy subjects or patients with mild asthma (31) and supports this concept. Some residual free elastase activity against synthetic substrate was also found only in BL from patients with SA, thus supporting the contribution of both 92 kDa gelatinase and leukocyte elastase during SA.

4. A strong and significant correlation (r = 0.89 with p = 0.0016) is obtained between the 92 kDa gelatinase content and the neutrophil number.

However, the highly probable neutrophil origin of 92 kDa gelatinase does not exclude the possible bronchial epithelial cell origin. Two arguments can be made in favor of bronchial epithelial cell origin. One is supported by the high and significant correlation (r = 0.87 with p < 0.0001) between the 92 kDa gelatinase content and the shedded bronchial epithelial cell number. The augmented 92 kDa gelatinase level would represent more the activation and/or the distress of bronchial epithelial cells in situ than the shedded bronchial cells by themselves. The second argument is supported by the fact that human bronchial epithelial cells, in primary cultures, are able to mainly express the 92 kDa gelatinase (29) and that 92 kDa gelatinase expression by human bronchial epithelial cells is upregulated in response to inflammatory cytokines such as IL-1 and tumor necrosis factor-alpha (TNF-) (33). It is well known that IL-1 and TNF- play a central and open synergistic role in the orchestration of pulmonary neutrophilic inflammatory diseases (34, 35). Thus, their release from actively chemoattracted neutrophils during SA would favor at least partially, the overregulation of 92 kDa by bronchial epithelial cells in situ.

The influence of systemic steroids given during the course of SA has been considered in our previous report (31). Taken together, our results and several other data suggest the hypothesis that dramatic increase of neutrophils in SA was related to the disease severity rather than steroid treatment. More specifically, it has been previously described (36) that steroids selectively and coordinately inhibit expression of 92 kDa gelatinase as well as TIMP-1 by inflammatory cells such as alveolar macrophages in vitro; steroids are also able to block the effects of one of the most potent signals (lipopolysaccharide [(LPS] endotoxin) for upregulation of 92 kDa gelatinase production. Moreover, our personal recent data (manuscript in preparation) clearly demonstrate the restrictive modulation of 92 kDa gelatinase production from alveolar macrophages during the course of human idiopathic pulmonary fibrosis. All these data together allowed us to suggest that a reduction in 92 kDa gelatinase level should be rather expected in BL from patients with SA in response to steroid treatment, thus eliminating a possible involvement of this treatment in the acute elevated 92 kDa gelatinase level observed.

In conclusion, on the basis of enzymatic and immunologic data, our study has demonstrated the presence of acutely elevated levels of 92 kDa gelatinase in the bronchial airways of patients with SA, partly counterbalanced by high TIMP-1 levels. The concomitant overproduction of stromelysin-1-activated forms supports the possibility that pro-92 kDa gelatinase is activated by stromelysin-1 during SA. In contrast, no or little 92 kDa gelatinase, nor stromelysin-1 were detected during mild asthma, indicating that the mechanism of airway inflammation in SA may be quite distinct from those in mild asthma. Several arguments allowed us to propose that cell origin of such elevated 92 kDa gelatinase levels may be shared between numerous activated chemoattracted neutrophils and activated bronchial epithelial cells in situ in response to lung injury. This acute enhanced secretion of 92 kDa gelatinase in bronchial airways during SA appears to be responsible for edema, increased bronchial lung permeability, and possibly some destruction of airways.