INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Bronchiectasis is a chronic pulmonary disease of diverse etiology. Several pathogenetic components of the disease have been identified, including genetic and/or exogenous impairment of tracheobronchial mucus clearance, persistent infection by microorganisms, and chronic airway inflammation attributable to a vicious cycle of persistent host-defense mechanisms (1). Although bacteria infecting the airways initially generate factors that trigger the host defense system, the airway epithelium and recruited neutrophils may become activated to be sources of proinflammatory cytokines. Tumor necrosis factor (TNF)-, interleukin (IL)-6, IL-8, and intercellular adhesion molecule-1 have been shown to be constitutively synthesized and released by human bronchial epithelial cells in culture (2). Expression and release of these mediators were indeed upregulated when the cells were exposed to purified bacterial endotoxins (3). Such cells can also synthesize and modulate release of IL-1 in response to antigen challenge (4). Medium conditioned by similar cultures not only prolonged the survival of human neutrophils in culture (5), but also stimulated neutrophil chemotaxis and adhesion to human endothelial cells in vitro (2). Neutrophils recruited to bronchial sites may also be involved in cytokine-mediated autocrine loops that promote recruitment of additional neutrophils (6). The combination of chronic airway infection and inflammation thus leads to the persistent recruitment and accumulation of neutrophils in the inflamed airways (7), resulting in ongoing enzymatic damage to bronchial tissue.

Despite an earlier suggestion of involvement of neutrophil protease in the migration of neutrophils through the basement membrane (8), recent work has shown that chemotactic migration of neutrophils does not involve proteolysis of basement membrane components along these cells' path (9). Yet patients with bronchiectasis suffer from gradual and persistent damage to bronchial tissues and decrements in pulmonary function, and eventually to respiratory failure. We therefore postulate that the neutrophils recruited in bronchiectasis are locally activated to damage bronchial tissues in their vicinity. Although cytokines that induce and perpetuate neutrophil recruitment have been identified in sputum samples of patients with bronchiectasis (10), the key player(s) in the bronchial network in activating the pericellular proteolytic activity of neutrophils remains to be identified.

Proteoglycans normally serve a role in maintaining the integrity of the extracellular matrix (ECM) through interactions of the decorin core and the biglycan core with the matrix proteins elastin (11) and collagen (12). The regular positioning within the fibrillar protein meshwork of small, leucine-rich repeat proteoglycans with their interacting glycosaminoglycan chains offers one level of defense of the matrix components against invading proteases (13). Another level, against cell damage caused by enzymatically produced oxygen radicals, can be provided by hyaluronan and large aggregating proteoglycans (13, 14) in the pericellular environment and the spaces between bundles of collagen fibrils. It follows that the proteoglycans are among the first targets of pericellular tissue-degrading activities of neutrophils when these cells become activated in affected airways. We therefore sought to demonstrate the stimulatory effect of bronchial secretions on neutrophil-mediated proteoglycan degradation and to then identify the key component of the bronchial cytokine network that contributes to this effect. We chose to do this by challenging human peripheral blood neutrophils with bronchial secretions and to use bronchoalveolar proteoglycans entrapped in a model polyacrylamide matrix as a substrate for extracellular or pericellular degradation by the challenged neutrophils at physiologic pH.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects and Sputum Samples

Patients were recruited from the Bronchiectasis Clinic of The University of Hong Kong, Queen Mary Hospital. Inclusion criteria were bronchiectasis documented by high-resolution computed tomography (HRCT) of the chest, bronchiectasis of idiopathic etiology, chronic sputum production with a daily sputum volume of > 10 ml, absence of asthma defined according to American Thoracic Society guidelines (15) and of other major pulmonary diagnoses, and a steady clinical state as defined by lack of change of symptoms noted by the patient over the preceding 3 wk. Exclusion criteria were bronchiectasis of defined etiology (e.g., posttuberculous, primary ciliary dyskinesia, common variable immunodeficiency); maintenance use of oral or nebulized antibiotics; and use of antibiotics within the previous 3 wk. Workup for the etiology of bronchiectasis included taking of a history for tuberculosis, measles, whooping cough, and definite pneumonic episodes before the appearance of symptoms of bronchiectasis; sputum smear and culture for acid-fast bacilli; serum immunoglobulin measurement; and ciliary motility assessment in indicated cases. A sweat test was not performed because of the known rarity of cystic fibrosis among Chinese and the lack of suggestion of multisystem disease in any of the patients.

Sputum samples were collected in sterile pots over a maximum of 4 h. Patients taking inhaled bronchodilators or inhaled corticosteroids were advised to omit these drugs for at least 4 h before the initiation of sputum collection. Sputum samples were then immediately centrifuged at 50,000 × g for 1.5 h, and the sol phase of each sample was aliquoted and stored at 70° C until analysis. Another freshly obtained sputum specimen was saved for inspection, and was then sent to the microbiology laboratory for bacterial culture.

Isolation of Neutrophils

Heparanized venous blood samples were obtained from healthy volunteers. Neutrophils were recovered by centrifugation through a gradient of Histopaque 1077 and Histopaque 1119 (Sigma Chemical Co., St. Louis, MO), washed in phosphate-buffered saline (PBS), and resuspended in PBS (16). Trypan blue exclusion indicated 97% viability of the recovered cells

Recovery of Bronchoalveolar Proteoglycans

Proteoglycans were extracted from lungs (gas exchange tissue and bronchioles) of Sprague-Dawley rats for 48 h in a buffer (pH 5.8) containing guanidine-HCl (4 M) and protease inhibitors (0.1 M  -aminocaproic acid, 0.005 M benzamidine/HCl, and 0.01 M ethylenediamine tetraacetic acid [EDTA]). The cleared extract was fractionated by ultracentrifugation (100,000 × g for 48 h) in a cesium chloride gradient at an initial density of 1.33 g/ml (17). The bottom two-fifths of the resultant gradient was desalted, and the proteoglycans were recovered by targeting their glycosaminoglycan components with sequential precipitation in 0.5% cetylpyridinium chloride (CPC) in 0.025 M sodium acetate (pH 5.8), and then in sodium acetate-saturated ethanol. All operations were performed at 4° C. This yielded a proteoglycan preparation (33.2 µg hexuronate/g wet weight of tissue) with proteoglycans of a wide range of molecular sizes (Figures 1A and 1B; lane 1). The preparation was found susceptible to digestion by both chondroitinase ABC (Seikagaku, Tokyo, Japan) and nitrous acid (18), thus indicating the presence of glycosaminoglycans of the chondroitin and the heparan classes, respectively. The glycosaminoglycan composition of the preparation was similar to that reported in human lungs (19, 20).

View larger version (81K): [in this window] [in a new window]   Figure 1.   PAGE analysis of bronchoalveolar proteoglycans (lane 1) and products resulting from incubation of neutrophils and proteoglycan-gel beads (lanes 2 to 6). Gels stained with Alcian blue and Coomassie blue (A) were further stained with silver (B). Lanes 2 to 6: 20 µl of the 40-µl solution of sequential CPC/EtOH-precipitable products resulting from incubation of neutrophils and proteoglycan-gel beads mixed (1:1 [vol/vol]) with loading buffer and application of the resulting solutions in entirety in the indicated lanes. Lane 2: incubation of neutrophils and blank gel beads; lane 3: incubation of proteoglycan-gel beads; lane 4: incubation of neutrophils; lane 5: chondroitinase ABC-treated lane 6 material; lane 6: incubation of neutrophils and proteoglycan-gel beads.

Determination of Proteoglycan Degradation Index

The lung proteoglycan preparation (960 mg, dry weight) was mixed with acrylamide (4 g), bisacrylamide (81.3 mg), N,N,N',N'-tetramethylethylene diamine (84 µl), and ammonium persulfate (0.56%, 18 ml) in a total volume of 72 ml to yield gel beads of 5% gel strength and 2% crosslinking (21). The proteoglycan-gel bead preparation (10 mg, dry weight) was incubated (37° C, 2 h) with neutrophils (5 × 10⁵ cells) and/or sputum sol (100 µl) in a final volume of 1 ml of PBS (pH 7.4). We sought to recover and assay the glycosaminoglycan-containing products that were released into the incubation medium. These products were recovered by sequential precipitation with CPC and then with ethanol; the final precipitate was redissolved in 500 µl of water, and 200 µl of the solution was assayed (in duplicate) for the hexuronate content of the glycosaminoglycan components (22). Test incubation of the proteoglycan-gel beads with trypsin (2.5 to 10 µg/ml) indicated linear release of glycosaminoglycan-containing products (5 to 30 µg hexuronate/ml) into the medium. Each sample of sputum sol (S) was tested with five preparations of neutrophils (N). Parallel incubations with heat-inactivated neutrophils (N_in) and heat-inactivated sputum sols (S_in) served as controls. The percent increases in hexuronate content of the test incubations (N + S, N_in + S, and N + S_in) with reference to corresponding control incubations (N_in + S_in) were computed as proteoglycan degradation indices (PDIs).

Test incubations with N + S were also compared with those in which: (1) the neutralizing antibody, rabbit antihuman TNF- (Peprotech, London, UK); (2) recombinant human TNF- (rhTNF-); (3) Eglin C (Sigma); or (4) EDTA (Sigma) was included.

Gradient Polyacrylamide Gel Electrophoresis

Gradient (4% to 20%) polyacrylamide gel electrophoresis (PAGE) in sodium dodecyl sulfate (SDS) was done with a modified Laemmli system (23) in an E660 vertical slab gel electrophoresis system (Hoefer Pharmacia Biotech Inc., San Francisco, CA). The separating gel was topped with a stacking gel (4%). Products recovered by sequential CPC and ethanol precipitation were dried and dissolved to produce 40-µl solutions, of which 20-µl aliquots were mixed 1:1 (vol/vol) with the sample buffer of 0.125 M Tris-HCl (pH 6.8), 4% SDS, and 20% glycerol before being loaded into individual sample wells. After the 40-µl sample solutions were loaded, electrophoresis was performed at 36 mA (constant current) for 6 h at 4° C. The gel was rinsed free of SDS in a solution of 40% (vol/vol) methanol and 8% (vol/vol) acetic acid for 3 h and was then stained for glycosaminoglycans and proteoglycans with 1% Alcian blue in the rinse solution. The gel was then stained for proteins with 0.2% Coomassie blue in the rinse solution. With this technique, glycosaminoglycans and proteoglycans were stained blue and proteins were stained purple against a decolorized background. For enhanced staining with silver, the Alcian blue-Coomassie blue-stained gel was immersed in silver reagent (Bio-Rad Laboratories, Richmond, CA) and developed according to the manufacturer's instructions.

Assay of TNF- Level in Sputum Sol

The protein level of TNF- in the sputum sol phase was measured with a quantitative sandwich enzyme-linked immunosorbent assay (ELISA) kit (PeproTech, London, UK). Samples (100 µl) of diluted sputum sol (dilution range, 1:4 to 1:8, [vol/vol]) were dispensed into wells of microtiter plates that had been coated with monoclonal antihuman TNF-. After incubation for 3 h at room temperature, the wells were washed free of unbound proteins and were then incubated with an enzyme-linked polyclonal rabbit antibody directed against TNF- for another 45 min at room temperature. After further rinses to remove unbound antibody, a substrate solution was added to each well and the mixtures were incubated for 20 min at 37° C. The reaction was terminated by the addition of a "stop" solution. Absorbance was measured at 492 nm in an ELISA reader (Molecular Devices Corporation, Sunnyvale, CA). Assays were performed on duplicate samples. Serial dilutions of recombinant TNF- provided the assay standard curve. Standard additions of recombinant TNF- were made to defined dilutions (1:4 and 1:8 [vol/vol]) of a test sputum sol specimen, and the resultant parallel upward shifts in the standard curve (Figure 2) indicated little loss of recombinant TNF- in the assay.

View larger version (22K): [in this window] [in a new window]   Figure 2.   Standard curve for assay of tumor necrosis factor (TNF)-. Standard additions of rhTNF- to defined dilutions (1:4 and 1:8 [vol/ vol]) of a test sputum sol showed the expected upward shift in the standard curve, with little if any loss in rhTNF-.

Statistical Analysis

The Mann-Whitney U test was used to compare means. Two-way analysis of variance (ANOVA) was used to evaluate possible effects of interactions between sputum and neutrophils on measured PDIs. Statistical significance was accepted at p < 0.05. Correlation between PDIs in incubations with N + S and the corresponding sputum level of TNF- was determined with Pearson's product moment correlation coefficient, r, and the two-tailed p value. The INSTAT and PRISM statistical software packages (GraphPad, Inc., San Diego, CA) were used for the statistical analyses.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patient Profile

Ten patients fulfilling the set criteria were recruited. Their demographic and clinical features are shown in Table 1. All of the patients had diffuse bronchiectasis (involvement of more than one lobe of lung) documented on HRCT of the thorax, and all were nonsmokers.

Degradation of Gel-Entrapped Proteoglycans

The preparations of bronchoalveolar proteoglycans used for entrapment into gel beads were analyzed with PAGE. Double staining with Alcian blue and Coomassie blue revealed proteins smeared out as a result of substitution with electrophoretically polydisperse glycosaminoglycans (Figure 1A; lane 1). Further staining with silver intensified the smear and reinforced this observation (Figure 1B; lane 1). Incubations of the proteoglycan- gel beads or of neutrophils yielded no detectable product (Figures 1A and 1B; lanes 3 and 4, respectively), indicating that there was no significant leakage of proteoglycans from the gel beads or neutrophil degranulation due to handling. However, incubation of neutrophils with blank gel beads yielded products that showed up as an Alcian blue-enhanced, silver-stained smear in the molecular size range of  200 kD (Figure 1B; lane 2). This large-molecular-size smear, which inevitably constituted the background in test incubations of proteoglycan-gel beads with neutrophils, provided evidence that the products of neutrophil degranulation in response to exposure to the particulate gel beads included proteoglycans as a component. Indeed, products of the test incubation included not only  200 kD smear indicative of degranulation, but also extension of the smear into regions of lower molecular size, as well as some distinct protein bands also in the smaller molecular size range (Figure 1A; lane 6). These latter materials were evidently degradation products released from the proteoglycan-gel beads by the action of neutrophils; and the silver-stained electrophoretic pattern reinforced this observation (Figure 1B; lane 6). After treatment of the product of neutrophil-proteoglycan-gel bean incubation with chondroitinase ABC, the < 200-kD smear was apparently digested away, whereas the  200-kD smear remained (Figures 1A and 1B; lane 5 versus lane 6). However, the latter smear was no longer detectable after treatment of the chondroitinase-resistant material with nitrous acid (results not shown). These glycosaminoglycan-degrading treatments thus indicate that the hexuronate-containing components assayed for determination of the PDI were contributed by chondroitin sulfates (< 200-kD smear) and heparin sulfates ( 200-kD smear) in the product.

Quantitative analysis of hexuronate-containing fragments released into the incubation medium of test incubations with proteoglycan-gel beads depended on recovery of the polyanionic fragments by precipitation with CPC. Because leakage from the proteoglycan-gel beads was found to be insignificant, it was not unexpected that blank incubation of the proteoglycan-gel beads resulted in a null hexuronate reading. It follows that the hexuronate-containing macromolecules recovered by precipitation with CPC from the medium of control incubations with N_in + S_in were due to glycosaminoglycan-containing components native to the neutrophil and sputum samples (Table 2). Incubation of the proteoglycan-gel beads with either N + S_in or N_in + S resulted in release of hexuronate-containing macromolecules in excess of the release in the control incubations (p < 0.01 and p < 0.05; respectively) (Table 2). The PDIs of 55.3 ± 8.6 and 34.7 ± 8.1 (mean ± SD for 10 sputum samples), computed respectively for incubations with N + S_in and N_in + S, therefore reflect the independent actions of neutrophil and sputum samples on gel-entrapped proteoglycans.

Sputum Sols Stimulate Neutrophil-Mediated Degradation of Gel-Entrapped Bronchoalveolar Proteoglycans

Incubations of proteoglycan-gel beads with N + S resulted in significantly higher concentrations of CPC-recovered hexuronate-containing products than did incubations with either N + S_in or N_in + S (p < 0.001) (Table 2). The PDIs computed for the 10 sputum samples studied ranged from 126 to 210; these were invariable higher than the sum of the indices corresponding to the N + S_in and N_in + S effects. Two-way ANOVA indicated that the enhanced effect was due to significant interaction between the sputum and neutrophil samples. It is therefore possible that soluble factors in the sputum samples interacted with the neutrophils to stimulate neutrophil-mediated proteoglycan degradation and associated release of glycosaminoglycan-containing fragments into the incubation medium.

TNF- Is a Key Sputum Factor in Stimulating Neutrophil-Mediated Proteoglycan Degradation

The sputum level of TNF- was found to correlate with PDIs determined for proteoglycan-gel bead incubations with corresponding N + S mixtures (Figure 3). When neutralizing antibody against TNF- was included in the incubations with N + S, PDIs determined for the coincubations were dose-dependently reduced to approach values achieved in incubations with corresponding N + S_in mixtures (Figure 4). This was observed invariably in all of the 10 sputum samples tested. Anti- TNF- antibody effectively attenuated the sputum-stimulated effect, presumably by binding to the TNF- endogenous in the sputum samples. These observations reinforce the concept that TNF- in bronchial secretions is important in stimulating neutrophil-mediated proteoglycan degradation.

View larger version (13K): [in this window] [in a new window]   Figure 3.   Correlation between proteoglycan degradation indicies (PDIs) and protein levels of TNF- in sputum sols of patients with bronchiectasis (r = 0.8182, p < 0.01). Data for 10 samples of S are presented.

View larger version (31K): [in this window] [in a new window]   Figure 4.   Neutralization of the sputum-stimulated effect by anti-TNF- antibody. Antihuman TNF- antibody was included in incubations of proteoglycan-gel beads with sputum sol (S) and neutrophils (N). Values of PDIs are expressed as the mean ± SD of 10 samples of sputum sol, each tested on five different neutrophil preparations. p < 0.001 for all tested samples, PDI (N + S) versus PDI (N + S + anti-TNF-); p < 0.001, PDI (N + S) versus PDI (N + Sin); p = nonsignificant, PDI (N + S + anti-TNF- 100 ng/ml) versus PDI (N + Sin), except *p < 0.01.

Supplementation with rhTNF- of the proteoglycan-gel bead incubations with N + S was found to enhance the sputum-stimulated effect (Figure 5A). However, similar supplementation of incubations from which the sputum sol was omitted produced insignificant enhancement (Figure 5B). This suggests that TNF- can elicit a stimulatory effect on neutrophil-mediated proteoglycan degradation only when neutrophils are primed to respond and that the priming factor can be found in sputum sol.

View larger version (24K): [in this window] [in a new window]   View larger version (12K): [in this window] [in a new window]   Figure 5.   rhTNF- was incubated with proteoglycan-gel beads and neutrophils in the (A) presence or (B) absence of sputum sol. (A) Data for paired observations on nine sputum samples. Values of PDIs are expressed as mean ± SD. At 10 ng/ml of rhTNF-, the sputum-stimulated effect could be further significantly enhanced. *p < 0.05, # p < 0.01, **p < 0.001. (B) Values of PDIs are expressed as mean ± SD for six individual experiments, each performed in triplicate. No significant difference was found with and without rhTNF- when sputum sol was absent from the incubation medium.

Serine Protease is Responsible for the Sputum-Stimulated Degradation of Proteoglycans by Neutrophils

As much as 90% of the originally observed neutrophil-mediated proteoglycan degradation was inhibited when Eglin C was included in incubations of proteoglycan-gel beads with N + S (Figure 6A). Thus, the observed proteoglycan degradation and release of glycosaminoglycan-containing fragments were due to elastaselike activity. In the N + S coincubations that responded to EDTA treatment, about 20% of the neutrophil-mediated proteoglycan degradation was inhibitable (Figure 6B). It follows that only a minor proportion of the observed proteoglycan degradation was due to metalloproteinases.

View larger version (19K): [in this window] [in a new window]   View larger version (22K): [in this window] [in a new window]   Figure 6.   Inhibition of neutrophil-mediated PDI by (A) Eglin C or (B) EDTA. Data represent mean ± SD of PDIs from five paired sputum samples. As much as 90% of neutrophil-mediated proteoglycan degradation was inhibited by Eglin C. When inhibition by EDTA was significant, this amounted only to 20% of the PDI(N+S). # p < 0.001, *p < 0.01.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The major finding in this study was that bronchial secretions of patients with idiopathic bronchiectasis were active in stimulating neutrophil-mediated degradation of bronchoalveolar proteoglycans, and that TNF- is an important component of sputum sol that contributes to this effect. We postulated that the stimulated degradation of proteoglycans demonstrated in this study would play an important role in homeostasis of the ECM and in airway remodeling in bronchiectasis and possibly in other chronic inflammatory airway diseases as well.

Most work on lung tissue degradation has focused on elastin and collagen (24). However, the lung tissue matrix is a supramolecular assembly of matrix proteins with closely associated proteoglycans, some of which are specifically bound through their core proteins to the protein fibrils, and others of which fill the spaces between bundles of fibrils. Thus, proteoglycans may function as a "protective" barrier to tissue destruction. Together with dynamic repair processes, this may well explain the relatively slow decline of lung function in bronchiectasis despite the high levels of free proteolytic activity detected in airway secretions in bronchiectasis (25, 26). Serine proteases, collagenase (matrix metalloproteinase-[MMP]- 8 and gelatinase-B (MMP-9) derived from neutrophils were found to act on a variety of sites in the interglobular (G₁/G₂) domain of the core protein of aggrecan, a major member of the tissue matrix proteoglycans, and to thus release glycosaminoglycan-containing fragments into tissue fluids (27). The fragments were usually monitored in terms of their metabolically labeled sulfate contents or charge-related properties. Interpretation of the results of proteoglycan degradation can, however, be difficult if proteoglycan preparations with a low degree of sulfation of the glycosaminoglycan components are studied. We circumvented this by recovering glycosaminoglycan-containing fragments as insoluble complexes of CPC, and leaving any nonassociating glycoproteins in the supernatant. The percentage difference in hexuronate contents of the CPC-precipitated products between test and control preparations could then be taken as an index of proteoglycan degradation.

Although the glycosaminoglycan chains were apparently preserved in our incubation system with neutrophils (Figure 1; lane 6), neutrophil-derived activities that degrade glycosaminoglycan components have been reported. Neutrophil granules were found to contain exoglycosidase and sulfatase that acted on the nonreducing terminals of chondroitin sulfates, but the low optimum pH for these activities (28) suggests intracellular degradation in subcellular vesicles, rather than extracellular degradation. On the other hand, a heparan sulfate endoglycosidase activity (optimum pH of 6.3) was releasable upon incubation of human neutrophils (29) but showed minimal activity at neutral pH. It therefore appears that when pH is maintained at a physiologic level, as in our study, extracellular degradation of proteoglycans by neutrophils is confined to the core protein; degradation of the glycosaminoglycan components occurs only at acidic pH.

The glycosaminoglycans reported in sputum sol may in part be products of proteolytic actions of stimulated neutrophils on native bronchoalveolar proteoglycans in the inflamed environment. Among the glycosaminoglycans, chondroitin sulfates were observed in sputum samples from patients with chronic bronchitis and cystic fibrosis (30), and hyaluronate was observed in sputum samples from patients with asthma (31). It therefore remains to be determined whether sputum glycosaminoglycans can be monitored as an indicator of persistent neutrophil-mediated inflammatory activity in the bronchi of patients with bronchiectasis.

Neutrophil elastase is established as a potent enzyme that can degrade elastin, which is an important component of lung tissue, and it has been detected in variable quantities in the airway secretions of subjects with various inflammatory airway diseases, including emphysema, bronchiectasis, cystic fibrosis, and asthma (22, 32, 33). Excessive elastolytic activity has been incriminated in degradation of the elastin in lung tissue matrix, leading to lung function impairment (32, 33). Neutrophil elastase has also been shown to be the major source of neutrophil-degrading activity on proteoglycan secreted by rat aortic smooth-muscle cells or lung fibroblasts (34). We therefore anticipated that neutrophil elastase would play an important part in the degradation of proteoglycans in our lung ECM model, as we were able to confirm in the present study. Apart from elastase, metalloproteinases, which may be secreted by various inflammatory cells including neutrophils and macrophages, as well as bacteria, have also been shown to degrade proteoglycans, and there is rapidly growing interest in the role played by MMPs in matrix remodeling in airway diseases (35, 36). Neutrophil MMP-8 has been detected in the bronchoalveolar lavage fluid of patients with bronchiectasis at a level that correlated with disease severity (37). A recent study demonstrated that human neutrophils secrete MMP-9 in vitro and in vivo in response to endotoxin and proinflammatory mediators (38). Although most of the neutrophil-mediated proteoglycan degradation occurring upon sputum stimulation in our study came from neutrophil elastase, metalloproteinase activity, when significant, was found to contribute to about 20% of the proteoglycan-degrading activity. Hence, it would be of interest to further evaluate the role of MMPs in chronic bronchial infections.

High levels of IL-1, TNF-, and IL-8 have been consistently found in the expectorated bronchial secretions of patients with bronchiectasis and cystic fibrosis, (10, 39, 40), and this has been implicated in the neutrophil influx into and sustenance of an intense local inflammatory response in the affected bronchial tree. Cytokines have also been shown to increase in vitro solubilization of fibronectin by neutrophil elastase (41). Evidence is herein provided for a role of TNF- in mediating neutrophil degranulation. Furthermore, our findings demonstrate that rhTNF- requires accompanying sputum sol or mediators therein to stimulate neutrophil-mediated proteoglycan degradation. This is consistent with previous reports that stimulation of neutrophil degranulation may involve more than one mediator and that priming of neutrophils with recombinant TNF- enhanced both neutrophil activating peptide (NAP-2)- and recombinant IL-8-induced neutrophil degranulation (42). Admittedly, the medications used in our patients, especially inhaled corticosteroids or bronchodilators, may have modified in vivo some of the actions of their bronchial secretions (the omission of these medications for 4 h before sputum collection was done only to avoid sputum contamination by medication that might have affected the results of in vitro experiments), but we anticipate that any effect would tend to mitigate the activities that were demonstrated (40) and would not change the overall message of our findings.

This is the first report that bronchial secretions of patients with bronchiectasis are active in stimulating neutrophil-mediated degradation of bronchoalveolar proteoglycans. Attenuation of the sputum-stimulating effect observed in the study by a neutralizing anti-TNF- antibody confirms that functionally active TNF- in the samples mediated the degradative activity. Obviously, the in vivo situation would be much more complex and would most likely involve many other cells, mediators, enzymes, and bacteria, which operate as a network in stimulating the neutrophil response and the perpetuation of an inflammatory cycle. The relatively slow clinical course of progression of these patients' disease would suggest that there is significant in vivo remodeling of the airway matrix at the same time, such that the deterioration of lung function occurs slowly over years, although this aspect of bronchiectasis cannot be explored with the in vitro model of proteoglycan-gel beads used in the present study. The use of more dynamic investigative models would provide further insight into the process of airway damage in bronchiectasis and possibly other conditions in which there is chronic bronchial infection, such as cystic fibrosis, with which bronchiectasis shares many similarities despite the distinct difference in these diseases' underlying etiologies.