INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

It is generally accepted that pulmonary emphysema is a consequence of proteinase-antiproteinase imbalance in the lungs, which can be induced by cigarette smoking (1). Recognition that elastic fiber destruction is probably a central feature in the pathogenesis of smoking-induced emphysema has focused attention on elastolytic enzymes that might be involved. The lungs contain a number of elastolytic enzymes (2) that mainly originate from inflammatory cells such as neutrophils or macrophages that have emigrated into the lungs. Accordingly, it has long been debated which cell and/or which proteinase, if any, is mainly responsible for the pathogenesis of pulmonary emphysema associated with cigarette smoking. Although there is substantial evidence that neutrophils contribute to the progression of chronic obstructive pulmonary disease (COPD) in its advanced stage, the role of neutrophils in the early development of emphysema is less clear (3). Indeed, neutrophils are not prominently increased in emphysema (4), and the increase of alveolar macrophages (AM) is more evident than that of neutrophils in bronchoalveolar lavage fluid (BALF) from most smokers.

We have, however, shown in previous studies that the level of neutrophil elastase-₁-proteinase inhibitor complex is significantly increased in BALF from asymptomatic older smokers with computed tomographic evidence of emphysema as compared with BALF from those who had a similar smoking history but did not have any signs of emphysema (5), and this complex is positively correlated with elastin-derived peptides (EDP) in BALF (6). These data suggest a possible contribution of neutrophils to the development of subclinical emphysema.

In this study we extended our previous studies and focused on proteins considered to be specific for neutrophils. Upon stimulation, neutrophils can release large amounts of toxic oxygen radicals and a variety of granule proteins (7). Human neutrophil lipocalin (HNL) is a recently isolated 24-kD protein secreted from secondary granules of activated neutrophils, and thus can be a marker of neutrophil accumulation and/or activation (8, 9). We also evaluated two matrix metalloproteinases (MMPs) secreted from neutrophils: gelatinase B (MMP-9, a 92-kD type IV collagenase) and neutrophil collagenase (MMP-8). Recently, MMPs and their inhibitors have come under increasing scrutiny with regard to their participation in the parenchymal destruction and repair processes of different pulmonary diseases (10). Although MMP-9 is known to be expressed in several types of cells other than neutrophils, such as lymphocytes, eosinophils, and macrophages (13), neutrophil-originated MMP-9 partly binds with HNL, and the complex of HNL/MMP-9 is expressed as a 130-kD form of MMP-9 that, like HNL, is unique to neutrophils (14).

In this study we attempted to evaluate the neutrophil involvement in subclinical pulmonary emphysema by measuring HNL and the two matrix metalloproteinases MMP-9 and MMP-8 in BALF, from a large number of asymptomatic, community-based older volunteers.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Community-based asymptomatic older volunteers were recruited and were screened for emphysematous changes with high-resolution computed tomagraphy (CT) after giving informed consent for participation (15). Three pulmonary physicians, who were blind to any information about individual patients, independently evaluated the presence of emphysematous changes in all serial horizontal slices on CT scans. We defined subjects as having emphysema (+), regardless of the magnitude or severity of disease, on the basis of agreement of all three physicians. All other subjects were defined as emphysema (). Forty subjects aged 32 to 71 yr with emphysema confirmed by CT scans underwent bronchoalveolar lavage (BAL). Twenty-eight of the 40 subjects were current smokers and 12 were ex-smokers who had ceased smoking for a minimum of 1 yr. Twenty-five healthy subjects who had no evidence of emphysema on lung CT scans also volunteered for BAL (16 smokers and nine ex-smokers, aged 39 to 75 yr). None of these subjects was receiving regular medication or had a history of asthma, and none had symptoms of respiratory tract infection in the preceding 2 mo. These subjects were also given a physical examination, as well as pulmonary-function and blood tests, to confirm the absence of disease. The study was approved by the Ethics Committee of Hokkaido University School of Medicine.

BAL

The study design ensured that none of the subjects had been smoking cigarettes for at least 12 h prior to the BAL procedure, in order to remove the acute effect of smoking. Sequential BAL was done with a modification of the technique introduced by Rennard and coworkers (16) as previously described in our publication (5). In short, four separate 50-ml aliquots of sterile 0.9% saline were infused into the right middle lobe segment through a wedged flexible fiberoptic bronchoscope (Model BF-B3R; Olympus, Tokyo Japan), and gently suctioned out of the lung. The fluid returned from the first 50-ml aliquot was not used for the study because the fluid from first aliquots is known to contain cells and material that come more from large airways than from peripheral regions (16). The remaining lavage fluid was combined and used as the alveolar lavage fluid-rich fraction in the study.

BALF Processing

Recovered BALF was strained through several layers of gauze to remove excess mucus and debris, and then centrifuged at 1,500 rpm for 5 min. The cell-free supernatant was aliquoted and stored at 70 °C until analyzed. The cell pellet was washed twice and resuspended in 5 ml of phosphate-buffered saline (PBS). Cell viability was estimated by exclusion of trypan blue, and a differential cell count was performed on cytocentrifuge preparations by May-Giemsa staining. BALF samples were concentrated with Ultrafree-PI filter units (Nihon Millipore Ltd., Tokyo, Japan) for the assay of MMP-9, or with Centricon filters with a membrane having a molecular weight cutoff of 10,000 (Amicon Inc., Beverly, MA) for the assay of MMP-8.

Culture of Alveolar Macrophages

Alveolar macrophages (AM) were isolated by adherence onto poly-L-lysine-coated plastic wells (Corning, NY) for 1 h at 37° C in RPMI-1640 medium (GIBCO, Grand Island, NY) containing 100 U/ml of penicillin and 100 µg/ml of streptomycin, in a moist atmosphere of 95% air and 5% CO₂. After removal of the nonadherent cells by two washes in PBS, the remaining cells were cultured in RPMI-1640 containing antibiotics for an additional 24 h as previously described (15).

Measurement of HNL

A double-antibody radioimmunoassay was used for measurement of HNL as described elsewhere (17).

Immunoblot for HNL

Unconcentrated samples of BALF were analyzed with sodium dodecylsulfate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by immunoblot analysis, according to the manufacturer's instructions. Briefly, samples (20 µl) of BALF treated with sample buffer were subjected to electrophoresis on precast NuPAGE gel (Novex, CA) under nonreducing conditions. After electrophoresis, proteins were transferred from NuPAGE gel to a nitrocellulose membrane (0.2 µm) as described in the manufacture's (Novex) instructions. The blot was developed with an Immun-StarTM Chemiluminescent Protein Detection System (BIO-RAD, CA), with polyclonal rabbit anti-HNL as the primary antibody (1:2,000 dilution, overnight incubation at room temperature).

Measurement of MMP-9

Enzyme immunoassay (EIA) kits for MMP-9 were obtained from Fuji Chemical Industries (Toyama, Japan). Levels of MMP-9 in concentrated BALF and medium from AM culture were measured with a solid-phase EIA based on the recognition of the pro- and intermediate forms of MMP-9, as well as on their complexed forms with tissue inhibitor of metalloproteinase (TIMP)-1 as previously described (18). The detection limit of this assay was 3.1 ng/ml. This assay was done in duplicate, and the values were within the linear portion of the standard curve.

Gelatin Zymography

Gelatin zymography was done for unconcentrated samples of BALF and medium from AM culture as previously described (14). Neutrophil lysate was used as a positive standard. Four-microliter aliquots of cell-free supernatants of BALF mixed with sample buffer were subjected to electrophoresis on 8% (wt/vol) polyacrylamide gels containing 1 mg/ml gelatin in the presence of SDS under nonreducing conditions. After electrophoresis, gels were washed in 2.5% Triton X-100 for 30 min, rinsed briefly, and incubated at 37° C for 24 h in a buffer containing 50 mM Tris HCl, pH 8.0, and 10 mM CaCl₂. Following incubation, the gels were stained with Coomassie blue R-250 and destained in a solution of 5% acetic acid and 10% methanol. Zones of enzymatic activity appeared as unstained bands against a blue background.

Measurement of MMP-8

Kits for Solid-phase, peroxidase-labeled EIA of MMP-8 were obtained from Fuji Chemical Industries. Levels of MMP-8 in concentrated BALF were measured on the basis of recognition of the pro- and active forms of MMP-8, but not MMP-8 complexed with TIMP (19). The detection limit of this assay system was 1.9 ng/ml. The assay was done in duplicate, and the values were within the linear portion of the standard curve.

Statistical Aanalysis

Values are expressed as means ± SE unless otherwise stated. Statistical analyses were performed on the data through single-factor analysis of variance (ANOVA) among four groups, and with Student's unpaired t test for comparisons of two groups when appropriate. As to the data for both MMP-9 and MMP-8 in BALF, values were expressed as medians with absolute ranges, and analysis was done with the nonparametric Kruskal-Wallis test and the Mann-Whitney U test. Spearman's rank correlation was calculated to assess the correlation between data. In all tests, values of p < 0.05 were considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Characteristics of the Subjects

The subjects were classified into four groups according to smoking history and the presence of emphysema on CT scans. Emphysema detected on CT scans involved less than 25% of the total area of each lung in most subjects who were allocated to the emphysema groups. Clinical characteristics, pulmonary function data, and BALF findings are shown in Tables 1 and 2. There were no significant differences in age or pack-years of smoking among the four study groups. Pulmonary function data revealed that %FEV₁, FEV₁/FVC, and DL_CO/V_A were slightly but significantly lower in the two groups with emphysema than in the groups without it. The number of macrophages in BALF was not statistically different among the four groups. However, when the two groups of current smokers and the two groups of ex-smokers, respectively, were combined, the difference between the current smokers and the exsmokers reached statistical significance (26.2 ± 4.7 ng/ml versus 8.7 ± 1.4 × 10⁴/ml BALF, p < 0.01).

HNL in BALF

HNL was detected in the unconcentrated BALF from all of the subjects. The level of HNL in BALF was significantly increased in the subjects who had emphysema, regardless of their smoking status, as compared with the subjects who were currently smoking but had no evidence of emphysema (44.0 ± 8.0 ng/ml BALF and 47.8 ± 13.1 versus 12.6 ± 1.6 ng/ml BALF, p < 0.05 for each comparison) (Figure 1). The level of HNL in BALF appeared to be independent of the number of neutrophils in BALF because the two variables did not show any appreciable relationship (r = 0.1, p = NS).

View larger version (18K): [in this window] [in a new window]   Figure 1.   Concentration of HNL in BALF. Levels of HNL are significantly increased in BALF from current smokers with emphysema and ex-smokers with emphysema as compared with BALF from current smokers without emphysema (44.0 ± 8.0 ng/ml and 47.8 ± 13.1 ng/ml, versus 12.6 ± 1.6 ng/ml BALF, respectively, p < 0.05).

BALF from eight emphysema patients, four current and four ex-smokers) were subjected to immunoblot analysis to determine the molecular pattern of HNL. The samples were taken from the eight subjects who showed the highest concentrations of HNL. As shown in Figure 2, four major bands were identified, at the molecular masses of approximately 24, 45, 68, and 130 kD, which corresponded to monomeric, dimeric, and trimeric HNL, and to the molecular complex of HNL/MMP-9, respectively.

View larger version (87K): [in this window] [in a new window]   Figure 2.   Immunoblot of HNL in BALF. Representative BALF samples from four ex-smokers with emphysema (lanes 1 to 4) and four current smokers with emphysema (lanes 5 to 8). Four major bands were identified at the molecular masses of approximately 24, 45, 68 and 130 kD. Marker proteins used in this analysis were SeeBlue (Novex) Prestained standards including myosin, bovine serum albumin, glutamic dehydrogenase, alcohol dehydrogenase, carbonic anhydrase, and myogobin, which migrate on NuPAGE gel at molecular sizes of 188, 62, 49, 38, 28, and 18 kD, respectively.

MMP-9 in BALF

The levels of MMP-9 in concentrated BALF samples were in the detectable range in 22 of 32 subjects with emphysema (14 current and eight ex-smokers), but in only four of 22 subjects without emphysema (two current and two ex-smokers). For later analysis, values below the detection limit were assumed to be 1.95 ng/ml in concentrated BALF. Among both the current smokers and ex-smokers, the level of MMP-9 was significantly higher in subjects who had emphysema than in those who did not (n = 22; median: 0.79 ng/ml; range: 0.39 to 26.0 ng/ml, versus n = 14; median: 0.39 ng/ml; range: 0.39 to 1.91 ng/ml; p < 0.01, respectively; and n = 10; median: 1.08 ng/ml; range: 0.39 to 6.89 ng/ml, versus n = 8; median: 0.39 ng/ml; range: 0.39 to 1.66 ng/ml; p < 0.05; respectively) (Figure 3). The level of MMP-9 had no significant correlation with the number of neutrophils or macrophages in the BALF (r = 0.4, r = 0.1, p = NS for both), but instead had a significant correlation with the level of HNL in BALF (r = 0.76, p < 0.0001).

View larger version (15K): [in this window] [in a new window]   Figure 3.   Concentration of MMP-9 in BALF. Levels of MMP-9 were significantly increased in BALF from current smokers with emphysema as compared with BALF from current smokers without emphysema (p < 0.01). Furthermore, ex-smokers with emphysema also showed significantly higher levels of MMP-9 in their BALF than did ex-smokers without emphysema (p < 0.05).

The profile of gelatinolytic activities in BALF was analyzed through gelatin zymography (Figure 4). Representative examples of BALF from three current smokers who had emphysema (lanes 5 to 7) and three current smokers who did not (lanes 2 to 4) are shown. In lanes 2 to 4, a negligible amount of the 92-kD band was detected. By contrast, in lanes 5 and 6, which contained samples from the subjects who had increased levels of MMP-9 in their BALF (MMP-9: 17.0 ng/ml, 3.8 ng/ ml, respectively), three distinct gelatinolytic bands were observed with molecular masses of 92, 130, and 220 kD, although the 92-kD band was the most prominent. These bands were considered to correspond to latent MMP-9, HNL/MMP-9 complex, and the multiple form of MMP-9, respectively. This banding pattern was similar to that observed for the neutrophil lysate (lane 1), and was therefore hypothesized to be characteristic of MMP-9 released from neutrophils.

View larger version (52K): [in this window] [in a new window]   Figure 4.   Analysis of gelatinolytic activities in BALF. Representative BALF samples were taken from three smokers without emphysema (lanes 2 to 4) and three smokers with emphysema (lanes 5 to 7). The positions of molecular weight standards are marked on the left. Neutrophil lysate demonstrates three distinct bands of gelatinase present at over 200 kD, 130 kD, and 92 kD (lane 1). In lanes 2 to 4, a negligible band at 92 kD was detected. In lanes 5 and 6, containing samples from subjects who had increased levels of MMP-9 in their BALF (17.0 and 3.8 ng/ml, respectively), distinct gelatinolytic bands were observed at 92, 130, and 220 kD. The latent form of MMP-9 at 92 kD was most prominent, and no activated form of MMP-9 was detected.

MMP-9 Released from AM

In an attempt to search for another source of MMP-9 observed in BALF, we measured the level of MMP-9 in the supernatant of AM culture medium. AM were obtained from 15 subjects, consisting of 10 who had emphysema and five who did not. There was no significant difference between the two groups for the level of MMP-9 (n = 10; 4.8 ± 1.2 ng/ml, versus n = 5; 8.8 ± 1.6 ng/ml; p = NS) (Figure 5). The profile of gelatinolytic activities was analyzed with gelatin zymography. In contrast to the findings with zymography using BALF as a sample, the bands detectable in AM culture medium were at 92 kD, with a negligible level at 220 kD, but there was no indication of the 130-kD band, which should be specific for neutrophils (Figure 6). The bands from five ex-smokers and one current smoker are shown in lanes 2 to 7 of Figure 6. The density of the 92-kD band was not related to the presence of emphysema or to the smoking status of the subjects (data not shown).

View larger version (10K): [in this window] [in a new window]   Figure 5.   Concentration of MMP-9 in supernatant of AM culture medium. AM were cultured in serum-free medium for 24 h. There was no significant difference between subjects without emphysema and those with emphysema (NS) (open circle: ex-smoker, closed circle: current smoker).

View larger version (71K): [in this window] [in a new window]   Figure 6.   Analysis of gelatinase activity released by cultured AM. For each sample, 5 µl of culture medium supernatant was loaded onto the gelatin gels. The positions of molecular weight standards are marked on the left. Culture medium was harvested from two ex-smokers without emphysema (lanes 2 and 3), one current smoker with emphysema (lane 4), and three ex-smokers with emphysema (lanes 5, 6, and 7). Neutrophil lysate demonstrates three distinct gelatinolytic bands present at over 200 kD, 130 kD, and 92 kD (lane 1).

MMP-8 in BALF

MMP-8 was detected in concentrated BALF samples from 23 of 25 subjects who had emphysema and from eight of 14 subjects who did not. The values below the detection limit were assumed to be 0.95 ng/ml in the concentrated BALF for later statistical analysis. The current smokers who had emphysema demonstrated significantly higher levels of MMP-8 in their BALF (n = 16; median: 1.07 ng/ml; range: 0.06 to 16.67 ng/ml) than did the subjects who were currently smoking but did not have emphysema (n = 9; median: 0.06 ng/ml; range: 0.06 to 2.33 ng/ml) (p < 0.01) (Figure 7). The level of MMP-8 had no significant correlation with the number of neutrophils in BALF (r = 0.02, p = NS), but, like MMP-9, was weakly but significantly correlated with the level of HNL in BALF (r = 0.53, p = 0.005).

View larger version (16K): [in this window] [in a new window]   Figure 7.   Concentration of MMP-8 in BALF. Current smokers with emphysema showed significantly higher levels of MMP-8 in their BALF than did current smokers without emphysema (p < 0.01).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study demonstrated that the levels of HNL, MMP-9, and MMP-8 in BALF from subjects with CT-diagnosed mild emphysema were significantly higher than those from current smokers without emphysema. In addition, gelatin zymography profiles of BALF from current smokers with emphysema indicated the presence of HNL/MMP-9 complex and of multiple forms of MMP-9, a banding pattern characteristic of MMP-9 of neutrophil origin, as previously shown (8, 14). These data are in accord with those in our previous study, which demonstrated that smokers with subclinical emphysema had a significantly higher level of neutrophil elastase-₁-proteinase inhibitor complex in their BALF than did those who were comparably smoking but had no evidence of pulmonary emphysema (5), and provide further evidence of neutrophil involvement in the development of emphysema, since the HNL and MMPs measured in this study were all considered to be mainly derived from neutrophils. Interestingly, the subjects with subclinical emphysema in our study did not show any appreciable increase in the number of neutrophils in their BALF, and the main findings in the study could therefore not be considered simply to reflect neutrophil accumulation in airways or alveolar spaces in these subjects. Furthermore, the study also indicated that the increase in AM in BALF, despite the well-known feature of current smoking, did not discriminate the two groups of current smokers with and without emphysema from one another, because the two groups had a similarly increased number of macrophages in their BALF.

The unique strategy of our series of studies is to compare smoking subjects with emphysema as detected by CT scanning with subjects who have a comparable smoking history but no evidence of emphysema on CT scans (5, 15). Because the subjects of the present study were all community-based volunteers, they were generally asymptomatic, and the degree of their emphysema, if they had the disease, was mild to modest at most. By taking this strategy, we could eliminate possible concomitant bacterial infection of airways, which was occasionally superimposed as emphysema progressed, and discriminate the factors related to smoking but not necessarily linked to the pathogenesis of pulmonary emphysema. The appearance of very few neutrophils in BALF itself supports the idea that bacterial infections were not the reason for the elevated levels of neutrophil-derived proteins in subjects with emphysema.

Neutrophils are believed to have an important role in a variety of inflammatory reactions. However, the pathologic role of neutrophils in the development of pulmonary emphysema is still controversial, since neutrophils are not the predominant inflammatory cells in pulmonary emphysema and are poorly correlated with the severity of emphysema at the site of the disease (4). Indeed, recent studies seem to have focused more on the role of AM than on that of neutrophils as the inflammatory cells mainly responsible for the development of emphysema (20).

Activated neutrophils release their granule components upon stimulation. HNL has recently been purified and characterized as a possibly unique protein derived from neutrophils (17, 21), although messenger RNA (mRNA) for HNL has also recently been shown to be present in some other cells (22). The function of HNL has not been clarified at all, and the physiologic significance of complex formation with MMP-9 has not been understood either (23). HNL was reported to be increased in sputum from patients with clinically apparent COPD as compared with control subjects (24).

MMP-9 and MMP-8 are the two MMPs released from neutrophils that degrade virtually all known components of the extracellular matrix (ECM). Recently, the role of MMPs in the pathogenesis of pulmonary emphysema has attracted much attention (12). Finley and colleagues reported that the activities of MMP-9 and collagenase were increased in BALF from patients with emphysema as compared with smoking controls (25). In our study, we quantified the immunologic levels of MMP-9 and MMP-8 in BALF and characterized the molecular pattern of gelatinolytic enzymes. The inflammatory cells responsible for an increased level of MMP-9 in BALF may not be neutrophils, but rather macrophages, particularly in smokers, because there is considerable evidence that macrophages are another potent source of MMP-9. However, we could not find any significant differences between subjects with emphysema and control smokers in the amount of MMP-9 secreted from AM.

The appearance of HNL/MMP-9 complex in BALF from subjects with subclinical emphysema, confirmed by gelatin zymography and HNL immunobloting, suggests that neutrophil-derived MMP-9 is present in BALF. Another MMP of neutrophil origin is MMP-8, although this enzyme does not have elastolytic activity but rather the matrix-degrading activity on interstitial collagens types I, II, and III. MMP-8 has been labeled as neutrophil collagenase (26) because it is synthesized exclusively by neutrophils during their maturation in bone marrow, which is also the case with MMP-9. Thus, increased levels of this enzyme in BALF from subjects with emphysema are considered as evidence of neutrophil involvement in the development of emphysema. No source of MMP-8 other than neutrophils has been reported in the lungs, although a recent study showed that MMP-8 was expressed in other human cells such as fibroblasts and endothelial cells (27).

It is of note that there was dissociation between the number of neutrophils and the amount of degranulated proteins in BALF in the present study. Theoretically, the degranulation of neutrophils could occur at any time of neutrophil sequestration in the pulmonary microvasculature, in the process of emigration, or in the air spaces (28). Therefore, the likely explanation for this dissociation is that the degranulated proteins in BALF reflect the total number of neutrophils in the lungs, not only in alveolar spaces, but also in the interstinum and/or in airways in which neutrophils are firmly adherent and not lavaged out of the lungs. An alternative explanation is that neutrophils, if not increased in number in the lungs, may be more activated in subjects with subclinical emphysema than in smoking controls (29), and may thus release more granule proteins, such as HNL, MMP-9, or MMP-8. Indeed, several previous studies have shown that both sequestration and activation of neutrophils occur in the pulmonary microvessels during cigarette smoking, which supports the possibilities raised earlier (30, 31).

Additionally, it must be mentioned that the subjects in our study who had emphysema even after they had stopped smoking showed a sustained increase in the level of HNL in BALF. Taken together with our previous finding that neutrophil elastase-₁-proteinase inhibitor complex in BALF may predict the rate of further annual decline of lung function (32), it is tempting to speculate that persistent neutrophil-medicated inflammation could partly explain why in some subjects with emphysema the disease continues to progress after cessation of smoking. On the other hand, there were some current smokers with emphysema in this study who had no obvious increase in HNL or other neutrophil-derived proteinases. One possible explanation for this is the heterogeneity of emphysematous changes at an early stage of the disease in individual lungs. The fluid lavaged from the middle lobe of the lungs may not necessarily reflect the phenomena occurring at the site of emphysema.

In conclusion, this study showed increased levels of neutrophil-derived proteins such as HNL in BALF from subjects who had subclinical emphysema as compared with comparably smoking subjects without emphysema, thereby providing further evidence of neutrophil involvement in the degradation of elastin and other ECM proteins characteristic of pulmonary emphysema.