Glinical Investigator

Clin Investig (1992) 70:269 276

Originai Article

© Springer-Verlag 1992

-Proteinase inhibitor and mucus proteinase inhibitor in human lung emphysema G. Trefz, J. SchlieBer, B. Heck, V. Schulz, and W. Ebert Thoraxklinik Heidelberg-Rohrbach

Summary. The role of the antiproteases at-proteinase inhibitor (e~PI) and mucus proteinase inhibitor (MPI) in human lung emphysema was investigated by measuring their amount and functional activity against trypsin, leukocyte elastase, and pancreatic elastase in the bronchoalveolar lavage fluid (BALF). In addition, leukocyte elastase was quantified in the lavage samples by measuring the concentration of the elastase-elPI-complex. The study population consisted of 38 patients (5 nonsmokers, 8 former smokers, 25 smokers) with aquired emphysema (i.e., emphysema which is not caused by elPI deficiency), and 44 individuals (16 nonsmokers, 8 former smokers, 20 smokers) without emphysema. No differences were found between patients with and without emphysema in the activities of cqPI and MPI, or in the concentration of exPI. The concentration of MPI was significantly higher in the BALF of patients with emphysema than in that of patients without emphysema (p = 0.025). A significantly higher concentration of elastase-cqPI complex was found in patients with emphysema than in those without emphysema (p = 0.041). This finding could reflect the higher proteinase burden to which patients with emphysema are exposed. The increase of MPI in lavage fluid of patients with emphysema seems to be the result of increased production in emphysematous lungs. However, it remains unclear why patients develop emphysema while showing an increased content of MPI. Key words: el-Proteinase inhibitor- Proteinase inhibitor - Emphysema Abbreviations: cqPI=cfl-proteinase inhibitor; BALF=bron-

choalveolar lavage fluid; ELISA = enzyme-linked immunosorbent assay; LEIC=leukocyte elastase inhibitory capacity; MPI=mucus proteinase inhibitor; PEIC=pancreatic elastase inhibitory capacity; TIC = trypsin inhibitory capacity

The imbalance between proteinases and proteinase inhibitors in the lung is considered to be responsible for the development of pulmonary emphysema. An excess of proteinases and/or a reduction of antiproteolytic activity can cause destruction of alveolar walls. Animal studies have shown that only elastinolytic enzymes could cause formation of emphysema [35]. In 1963, Laurell and Eriksson [25] recognized that a hereditary deficiency of el-proteinase inhibitor (cqPI), which is an important inhibitor of neutrophil elastase, leads to the development of emphysema. However, only 1% of all human emphysema diseases are caused by cflPI deficiency. From a clinical point of view, cases of lung emphysema which are not caused by e~PI deficiency are more important and frequent. In this respect, it is generally agreed that cigarette smoke is the major reason for this kind of acquired emphysema. It was found at autopsy that more than 50% of smokers showed some evidence of emphysema whereas only 10-15% of nonsmokers showed signs of emphysema [38]. Ten to fifteen percent of smokers actually develop an emphysema which leads to clinical complications [38]. In vitro experiments have proven that components of cigarette smoke are able to diminish the inhibitory activity of cqPI by oxidizing a methionyl residue in the P~ position of the reactive center of the inhibitor [7, 20]. Consequently, leukocyte elastase, the physiological target enzyme of c~xPI, reacts too slowly with oxidized ~ I P I in vivo to prevent emphysema [31]. The oxidation of ~IPI could be measured in vitro by the loss of the inhibitory capacity against pancreatic elastase (PEIC), whereas the inhibition of trypsin (TIC) remains unaffected [7]. In the past, many efforts were made to understand the mechanisms underlying the development of emphysema. Proteinases and proteinase inhibitors were measured in serum and bronchoalveolar lavage fluid

270 (BALF) of different groups of patients with divergent results. Several groups reported that eaPI is less active in the BALF of smokers than in that of nonsmokers [8, 17], which is in accordance with the oxidation theory. However, these results could not be confirmed by others [4, 36, 37]. Determination of e~PI activity in serum also leads to different results. For example, in some investigations e~PI activity was found to be lower in smokers than in nonsmokers [2, 11], but others have reported that smoking does not affect cq-PI activity [26]. We have begun to investigate a group of patients who showed clinical signs of existing emphysema as proven by lung function tests, radiology, and histopathological findings. This group was compared with a control group of individuals who showed no clinical signs of emphysema. Our initial results have demonstrated that no significant differences in the functional activity of elPI are detectable in the serum of these two groups of patients [39]. When TIC/e~PI ratios were calculated in order to characterize active e~PI, unexpected ratios above unit were found in most BALF samples in contrast to those found in serum [39]. This turns our attention to other inhibitors present in lavage fluid besides e~PI. One candidate is the "mucus proteinase inhibitor" (MPI), also called antileukoprotease [16], which was first isolated from bronchial secretions in 1972 [19]. MPI is a reversible inhibitor of trypsin, cathepsin G, and chymotrypsin [34]. In addition, MPI was also found to be a fast-acting inhibitor of leukocyte elastase [3], but without inhibitory activity against pancreatic elastase. Like cqPI, MPI is sensitive to oxidizing agents. Oxidation results in a loss of the inhibitory activity against leukocyte elastase and trypsin [23]. It has been shown that MPI is localized in secretory granules of serous cells in the upper respiratory tract [21]. In the small peripheral airways of the lung, MPI occurs within Clara and goblet cells [14]. Extracellularly, MPI has been found to be associated with elastic fibers [41]. These results point to the importance of MPI in protecting the lung against elastolytic proteinases. In this paper, we present our data on the measurement of both activity and amount of cqPI as well as of MPI in BALF of patients suffering from emphysema, in comparison to individuals without emphysema.

Subjects Patients who underwent bronchoscopy because cancer was suspected were also examined for the

presence of emphysema. Emphysema was diagnosed as a result of data accumulated through lung function tests, X-ray, and in a few cases, histopathological examinations after surgery. The quotient of reduced forced expiratory volume in one second to vital capacity (FEV1/VC), the residual volume (RV), and the total resistance (Rt) were determined using a body plethysmograph (Bodytest, Jfiger, W/,irzburg, FRG). All values obtained were adjusted for body temperature and pressure saturated with vapor (BTPS). Reference values suggested by a commission of the European community were used. Emphysema was indicated by a RV value above the upper 2 s limit and a FEV1/VC quotient under 70%. For the radiological determination of emphysema, X-ray plates were screened according to different criteria to diagnose emphysema: destruction of lung spaces in the periphery, flattening out of the diaphragm, and enlargement of the retrosternal airspaces. Individuals were assigned to the group "without emphysema" when lung function parameters were in the predicted age-dependent standard ranges, and when no simultaneous radiological or histopathological signs of emphysema could be detected. Patients with emphysema showed pathological values in at least two parameters: in lung function parameters and radiological findings (30 patients), in radiological findings and histopathological results (5 patients), in lung function parameters and histopathological findings (1 patient), or in all three parameters (1 patient). The values of the different groups are shown in Table 1. The group of patients suffering from emphysema consisted of 38 subjects (5 nonsmokers, 8 former smokers, 25 smokers). The control group was composed of 44 subjects (16 nonsmokers, 8 former smokers, 20 smokers).

Methods

Bronchoalveolar lavage procedure Bronchoalveolar lavage was performed according to the recommendation given by the Deutsche Gesellschaft fiir Pneumologie und Tuberkulose [13]: 100 ml of physiological NaC1 solution (37°C instilled in 20 ml portions followed by gentle aspiration) was applied through the suction channel of a fiberoptic bronchoscope (Olympus BF type IT 10) wedged into a segmental bronchus of the right middle lobe or lingula. Because the patients also suffered from lung tumors, the radiologically normal, tumor-free part of the lung was lavaged. The first portion of an instilled 20 ml NaC1 solution

271 Table 1. Values of the lung function tests of the subjectswith and without emphysema with respect to their smoking behavior Number of patients Emphysema Nonsmokers

5

Former smokers

8

Smokers

25

No emphysema Non-smokers

16

Former smokers

8

Smokers

20

Age

RV % of the upper 2 s limit of the reference value

FEV1/VC %

Rt KPa/1/s

69.0 (63-77) 64.7 (50-73) 51.1 (21-70)

176.2+_75.0 (100-271) 134.04, 30.0 (106-175) 139.14, 31.8 (82--211)

68.8 +__5.2 (6~76) 52.7 4,14.3 (30-71) 57.6 4,13.8 (33-72)

0.33 _+0.13 (0.21-0.49) 0.50 4 0.36 (0.23-0.96) 0.37 4-0.20 (0.10-0.89)

52.3 (19-68) 57.5 (39-67) 46.8 (23-75)

89.0 4-24.8 (48-121) 90.0 4, 59.7 (27-180) 87,74,26.6 (44-113)

76.6 4, 6.5 (66-89) 75.6 +__6.7 (70-84) 75.8__8.2 (59-88)

0.40 4, 0.38 (0.22-1.47) 0.33 4, 0.18 (0.19 0,69) 0.22___0.11 (0.12 0.51)

recovered was discarded. The following portions were collected and immediately put on ice. The percentage of recovery was 5 1 + 6 % in the case of individuals without emphysema and 47 +_ 5% in patients with emphysema. The lavage fluid was passed through gauze to remove mucus, low-speed centrifuged (10min 200 × g; 4 ° C) to remove cells, and then ultracentrifuged (30 rain 15,000 x g; 4 ° C). Part of the supernatant was used immediately for measurement of inhibitory capacities. Another part was stored at - 2 0 ° C until the determinations of albumin, e~PI, and MPI could be done by immunological methods.

Inhibitory capacities Inhibitory capacities were assessed in unconcentrated B A L F after incubation of a constant amount of enzyme with increasing amounts of B A L F supernatant. The amount of B A L F necessary to inhibit the constant amount of enzyme in the test was calculated after determination of the equivalence point by linear regression using 6 points obtained from 6 different B A L F concentrations according to Green [18]. Concentrations of active enzymes were calculated by "active titration-site". The concentration of active trypsin was determined with p-nitrophenylguanidino-benzoate [10]. Elastases were titrated with azapeptides [32]. Leukocyte elastase was kindly provided by Dr. Lang (Merck, Darmstadt, FRG). Sensitive fluorogenic enzyme tests were used

for determination of the inhibitory capacities. Briefly, trypsin inhibitory capacity (TIC) was determined using Bz-Arg-AMC (concentration in the test: 0.3 mmol/l) as a substrate in 0.1 mol/1 Tris buffer, p H 7.8, containing 0.02 tool/1 CaCI2 and 2.75 nmol/1 of active trypsin [43]). Leukocyte elastase and porcine pancreatic elastase inhibitory capacities (LEIC and PEIC, respectively) were measured using M e O S u c - A l a - A l a - P r o - V a l A M C (concentration in the test: 0.2 retool/l) as a substrate [9] in 0.1 tool/1 Tris buffer, p H 7.5, containing 0.5 tool/1 NaC1, 0.05% (w/v) Triton X 100 and 0.25 nmol/1 of active leukocyte elastase or I nmol/1 porcine pancreatic elastase, respectively. Enzymes were incubated 30 min with B A L F in test buffer before the reaction was started by adding the substrate. Final volume in each test was 1 ml.

Quantification of ~IPI Concentrations of ~IPI in unconcentrated BALF were determined by radial immunodiffusion using commercially available agar plates (LC-Partigen, Behring-Werke, Marburg, FRG). Protein standard serum LC-V (Behring-Werke, Marburg, F R G ) was used for calibration of the assay. In a few cases it was necessary to concentrate the B A L F 20-fold by lyophilization in order to detect cqPI in these highly diluted probes.

Quantification of MPI The concentration of MPI in BALF was determined using an enzyme-linked immunosorbent as-

272

say (ELISA). Microtiter plates (Nunc Maxisorp, Denmark) were coated with anti-MPI rabbit IgG solution (5 txg/ml in 15 mmol/1 Na2CO3, 0.35 mol/1 NaHCO3, 0.2 g/1 NAN3, pH 9.6; 200 gl per well) at 4 ° C overnight. The coated plates were washed 5 times with wash buffer (10mmol/1 KH2PO4, 15 mmol/1 NaC1, 0.05 g/1 Tween 20, pH 7.4). MPI standard samples (calculated by amino acid analysis) and BALF samples were diluted in 2.7 mmol/1 KH2PO4 x H20, 6.5 mmol/1 Na2HPO,~, 0.14 mol/1 NaC1, 0.5 g/1 Tween 20, 0.2% (w/v) bovine serum albumin, pH 7.4. 200 gl of each sample were added to the wells. Then the plates were incubated for 3 h at 37 ° C. After the plates were washed 5 times in wash buffer, 200 gl of anti-MPI rabbit IgG peroxidase conjugate solution (dilution 1:200) were added to each well and incubated for 2 h at 37 ° C. The plates were washed again, and 100 gl of substrate solution (o-phenylendiamine) were added per well and incubated for 15 min at room temperature. The substrate reaction was stopped by adding 100 gl of 0.5 mol/1 H 2 8 0 4 , and the absorbance was read at 492 nm. MPI-antibodies conjugate and MPI standards isolated from seminal plasma according to [33], were kindly provided by Prof. R. Geiger, Institut ffir Klinische Chemie und Klinische Biochemie, Miinchen.

Quantification of elastase-cqPI-complex

TIC nmol enzyme / nmol inhibitors 1,00,8

-

0,6

1~:. 0,4 0,2

0

N I

I no emphysema

emphysema

Fig. 1. I n h i b i t o r y capacities o f cqPI and M P I in B A L F of patients w i t h (hatched columns) and w i t h o u t (open columns) emphysema against trypsin (TIC) and leukocyte elastase (LEIC)

as target enzymes. Median values are expressed as nmol of enzyme inhibited per nmol of cqPI plus MPI

PEIC nmol enzyme/ nmol inhibitor 1,0 0,8 0,6-

0,2 0 I

Elastase-cqPI complexes were determined with a commercially available enzyme immunoassay (Merck, Darmstadt, FRG). Unconcentrated lavage fluid was used.

LEIC

I no emphysema emphysema

Fig. 2. Inhibitory capacities of ~lPI in BALF of patients with (hatched column) and without (open column) emphysema against pancreatic elastase as target enzyme (PEIC). Median values are expressed as nmol of enzyme inhibited per nmol of elPI

Quantification of protein Total protein in BALF was determined according to Lowry et al. [27] (Protein assay kit, Sigma, Deisenhofen, FRG). Bovine serum albumin was used as a standard.

Statistical analysis Because the values observed in patients with or without emphysema were not normally distributed, results are given as medians with 95% confidence intervals. Statistical comparison between the groups was performed with Wilcoxon's non-parametric rank sum test [28]. Results

Figure 1 shows the median values of TIC and LEIC measured in the BALF of patients with and

without emphysema. The results are expressed as nmol of enzymes inhibited per nmol of cqPI and MPI, because these enzymes are inhibited by both inhibitors. In the case of PEIC (Fig. 2), the results are expressed in nmol enzyme inhibited per nmol cqPI, because MPI is not able to affect pancreatic elastase. No significant difference was observed between patients with and without emphysema or between smokers, ex-smokers, and non-smokers with regard to the three parameters. All values obtained were found to be below 1. LEIC, however showed a lower median value ( ~ 0.4 nmol enzyme/ nmol inhibitors) compared to TIC ( ~ 0.6 nmol enzyme/nmol inhibitors) (Fig. 1). The median value of PEIC was found to be in the range of 0.4 nmol enzyme/nmol cqPI. The total concentrations of inhibitors cqPI and MPI and the amount of leukocyte elastase measured as elastase-cqPI complex

273 MPI / protein

alpha1 - Pl / protein

elastase / protein

nmol / mg

nmol / mg

MPI / alpha 1 - Pl

pmol / mg protein

m01/ mol

0,6

0,6

60

0,5

0,5

50

0,4

0,4

4O

6-

0,3

30

5

20

4

~=~ ~

0,3 0,2 0,1

I

0,2 0,1

0

lO

0 F

0 I

I no emphysema

8

~

:,21N

3

l ~

emphysema

Fig. 3. Concentrations of cqPI, MPI, and elastase in BALF of patients with (hatched columns) and without (open columns) emphysema. The latter was assayed as elastase-cqPI-complex. elPI was measured by radial immunodiffusion, MPI by ELISA. Median values related to protein in order to correct different BALF dilutions. Protein concentrations were determined according to Lowry et al. [27] (*: P < 0.05; Wilcoxontest)

2

no emphysema emphysema

Fig. 4. Median values of the molar MPI/cqPI ratios in lavage of patients with (hatched column) and without (open column) emphysema. (* : p < 0.05 ; Wilcoxontest)

are given in Figure 3. Values are related to the protein content in order to consider the different dilutions of BALFs. The protein concentrations in the individual BALF samples showed variations over a wide range. However, no significant difference was observed between individuals with and without emphysema (120.4-t-103.2 gg/ml and 128.6-t102.5 gg/ml, respectively). The cqPI content was lower in the BALF of patients with emphysema than in that of individuals without emphysema (0.16 nmol/mg protein versus 0.20 nmol/mg protein). The difference did not reach statistical significance. In addition, no significant difference was found between smokers, ex-smokers, and nonsmokers, although smokers were found to have lower cqPI values than nonsmokers. In contrast, MPI concentrations were found to be significantly higher (p =0.025) in patients with emphysema (median: 0.34nmol/mg protein) than in those without emphysema (median: 0.16 nmol/mg protein). When smokers were compared to non-smokers, they showed higher MPI concentrations in their BALF, but not to a statistically significant degree. The results of the determination of elastasecqPI complex is also shown in Figure 3. Patients with emphysema showed higher leukocyte elastase concentrations in their BALF (6.78 pmol/mg protein) than patients without emphysema (4.52 pmol/ mg protein). The difference was statistically significant (p=0.041). No difference was observed between smokers, ex-smokers, and non-smokers with regard to elastase-e~PI complex.

Molar ratios between MPI and cqPI were calculated in order to estimate the importance of the two inhibitors in the study groups (Fig. 4). The MPI/cqPI ratio was found to be significantly higher (p=0.025) in the emphysema group (median: 2.11) than in those without emphysema (median:

1.17). Discussion In the past, many efforts have been made to investigate and understand the role of the protease-antiprotease imbalance as a leading cause for the development of emphysema. However, most of the preceding studies have been performed on lung secretions of healthy smokers and nonsmokers. Here we report on a study of the activities and concentrations of antiproteinases in the BALF of patients with acquired emphysema as compared to patients without clinical signs of this disease. Measurement of the inhibitory capacities was performed immediately after BALF samples were collected and centrifuged, because longer storage, even at - 2 0 ° C, was found to decrease considerably the inhibitory capacities. Furthermore, the application of very sensitive peptide substrates enabled us to use native, unconcentrated BALF. Thus, protein loss and inactivation of proteinase inhibitors during concentration of the lavage fluids as described in the literature [1] could be avoided. However, one must consider that the degree of inhibition observed in biological fluids is critically dependent on the substrates used [29]. The use of

274 other substrates, particularly natural substrates with higher molecular weights, can influence the extent of the inhibitory capacities. Trypsin, human leukocyte elastase, and porcine pancreatic elastase were used as target enzymes in order to calculate the inhibitory capacities in BALF fluids of patients with and without emphysema. The values reflecting the inhibition of the three enzymes were not significantly different between the two groups. LEIC and PEIC values were found to be lower than TIC values. In the case of PEIC this is understandable because pancreatic elastase is inhibited only by nonoxidized, active cqPI, whereas trypsin is inhibited by both oxidized and nonoxidized e~PI and MPI. In the case of LEIC one would expect similar values to those of TIC, because both enzymes, trypsin and leukocyte elastase, are identical in their behavior against cqPI and MPI. Several reasons might be responsible for the lower LEIC values. On the one hand, the time required for reaction with the inhibitors is different for the two enzymes. It is possible that 30 min are insufficient for the reaction of leukocyte elastase and its inhibitors, particularly MPI. On the other hand, it is also possible that the higher TIC values are caused by other trypsin-specific inhibitors present in bronchial secretions [24], which have no influence on leukocyte elastase activity. From the results it must be concluded that the classical, time-independent determination of the inhibitory capacities, which were used in this study, may not be useful to determine the presence and concentration of c¢~PI, which is restricted in its activity. In this context, measurement of the association rate constant for neutrophil elastase might be the more sensitive parameter as recently published by Wewers et al. [40]. Besides determination of the activity of the inhibitors, their concentrations were also measured using immunological methods. The values obtained were related to protein in order to compensate for different BALF dilutions. No differences were found in the cqPI concentrations between patients with and without emphysema as determined by radial immunodiffusion (Fig. 3). However, this is not astonishing, because patients who had developed emphysema as a result of cqPI deficiency were excluded from this study. Nonetheless, MPI concentrations determined by ELISA were found to be significantly higher in the group of patients suffering from emphysema (Fig. 3). The serum concentration of MPI is extremely low [15], but MPI could be detected in larger quantities in expectorated sputum and BALF [30]. Kramps et al. [22] reported that the MPI/cqPI ratio depends on

the part of the lung which is lavaged. They determined molar MPI/cclPI ratios in the range of 0.02 to 0.14 in the peripheral part of the lung of healthy volunteers, indicating that cqPI is the major inhibitor in this part of the lung. In BALF from larger airways, they measured a molar MPI/cqPI ratio of 1.42_+ 0.72, which was significantly higher than the ratios of the peripheral lavages. They concluded that MPI is the major inhibitor in the larger airways, whereas cqPI is predominant in the peripheral airways. In the present study, we found MPI/cqPI ratios higher than those reported by Kramps et al. (Fig. 4). Patients without emphysema had a median value of 1.17; patients with emphysema showed even higher ratios (median value: 2.11). The difference was found to be significant. One reason for this discrepancy might be the technique of performing bronchoalveolar lavage. Kramps et al. used a balloon-tipped catheter with a diameter of 2 mm inserted through the biopsy channel of the bronchoscope. We used the bronchoscope alone, which has a diameter of 6 mm. It is possible that we did not reach the peripheral airways to the same extent as Kramps et al. did. We rejected the first portion of recovered lavage fluid, which is believed to contain secretions from the larger airways, to minimize this technical discrepancy. The subsequent portions, which have reached the alveolar spaces, were used for the determination of the biochemical parameters. Furthermore, the different populations of patients investigated must be discussed in this context. Most of our patients (patients with and without emphysema) underwent bronchoscopy for the diagnosis of tumors. This might be an important point, because malignant diseases can stimulate bronchial secretion and therefore also the release of MPI. These two considerations might explain why the MPI concentrations obtained in these study were higher than those reported in literature. However, the significant increase of MPI concentrations in the group of patients with emphysema could not be explained solely by the existence of a tumor, because the group of patients without emphysema was also mainly composed of tumor patients. The finding that MPI is increased significantly in the lavage of emphysematous patients might therefore be an important indication of the stimulation of MPI production and secretion in the small airways of these patients. These results are in line with other recently published data obtained through immunomorphological studies by Willems et al. [42], who found that the destruction of the alveolar attachments is associated with a rise of MPI-containing

275

bronchiolar cells. Moreover, elastase-elPI complexes were measured by ELISA. Like MPI, leukocyte elastase bound in complex with elPI was found to be significantly higher in the emphysema group (Fig. 3). This increase in BALF of patients with emphysema reflects the higher protease burden with which they are confronted. These findings confirm the hypothesis that elastase plays an important role in the pathogenesis of lung emphysema. When smokers, ex-smokers, and nonsmokers were compared, none of the parameters measured were found to show a significant difference. Therefore, one must conclude that the observed differences reflect emphysema/nonemphysema rather than smoking/nonsmoking status. Finally, two important questions arise from our results: why do patients with emphysema have higher MPI concentrations in their BALF, and why is this inhibitor unable to prevent emphysema? One may speculate about the higher expression of MPI, which might be the response of the lung to the protease burden in the sense of an acute phase reaction against the destructive processes in the distal airways caused by proteases. It is known [12] that instillation of leukocyte or pancreatic elastase causes the discharge of secretory granules, deriving from serous cells of the bronchial glands, and Clara and goblet cells in the bronchioli as well, which cells could be shown to contain MPI [14]. MPI may thereby be released into the epithelial lining fluid. However, like ~IPI, only the native, nonoxidized form of MPI is able to inhibit leukocyte elastase in vivo. The ELISA used in this study for the quantification of MPI is not able to discriminate the active inhibitor from the inactive form. Therefore, it cannot be ruled out that a substantial amount of MPI is oxidized, e.g., by the myeloperoxidase system of granulocytes or by components of cigarette smoke. The oxidized MPI might be unable to prevent emphysema. Another important consideration seems to be the exact localization of the proteases or antiproteases involved in the destructive process. For instance, Campbell et al. [5, 6] found that only lowmolecular-weight inhibitors can completely inactivate leukocyte elastase in the presence of leukocytes. It appears that leukocytes may be able to release elastase into subcellular clefts that are protected from large inhibitors such as e~PI or MPI. In addition, the concentration of oxidative reagents in the vicinity of the cells seems to be an important factor. The development of emphysema might therefore be due to cell-mediated proteolysis in the microenvironment of leukocytes and macro-

phages, which could hardly be characterized by measuring inhibitor amounts and inhibitory capacities in BALF. New experimental concepts and systems, including cells as sources of proteases and inhibitors, must be evaluated to gain a better knowledge of the role of proteinase - proteinaseinhibitor balance in the pathogenesis of emphysema.

Acknowledgement. This work was supported

by the Forschungs-

rat Rauchen und Gesundheit.

References 1. Afford SC, Stockley RA, Kramps JA, Dijkman JH, Burnett D (1985) Concentration of bronchoalveolar lavage fluid by ultrafiltration: evidence of differential protein loss and functional inactivation of proteinase inhibitors. Anal Biochem 151:125-130 2. Beatty K, Robertie P, Senior RM, Travis J (1982) Determination of oxidised alpha-l-proteinase inhibitor in serum. J Lab Clin Med 100:186-192 3. Boudier C, Bieth JG (1989) Mucus proteinase inhibitor: a fastacting inhibitor of leukocyte elastase. Biochim Biophys Acta 995:36-41 4. Boudier C, Pelletier A, Pauli G, Bieth JG (1983) The functional activity of c~l-proteinase inhibitor in bronchoalveolar lavage fluids from healthy human smokers and nonsmokers. Clin Chim Acta 132:30%3/5 5. Campbell EJ, Campbell MA (1988) Pericellular proteolysis by neutrophils in the presence of proteinase inhibitors: effects of substrate opsonization. J Cell Biol 106:667-676 6. Campbell EJ, Senior RM, McDonald JA, Cox DL, Greco JM, Landis JA (1982) Proteolysis by neutrophils: relative importance of cell-substrate contact and oxidative inactivation of proteinase inhibitors in vitro. J Clin Invest 70: 845852 7. Carp H, Janoff A (1978) Possible mechanisms of emphysema in smokers. In vitro suppression of serum elastase-inhibitory capacity by fresh cigarette smoke and its prevention by antioxidants. Am Rev Respir Dis 118:617-621 8. Carp H, Miller F, Hoidal JR, Janoff A (1982) Potential mechanisms of emphysema: c~l-proteinase inhibitor recovered from lungs of cigarette smokers contains oxidised methionin and has decreased elastase inhibitory capacity. Proc Natl Acad Sci USA 79 : 2041-2045 9. Castillo M J, Nakajima K, Zimmerrnan M, Powers JC (1979) Sensitive substrates for human leukocyte and porcine pancreatic elastase: a study of the merits of various chromophoric and fluorogenic leaving groups in assays for serine proteases. Anal Biochem 99 : 53-64 10. Chase T, Shaw E (1970) Titration of trypsin, plasmin, and thrombin with p-nitrophenyl-p'-guanidinobenzoate HC1. Meth Enzymol 19:20-27 11. Chowdhury P, Bone RC, Louria DB, Rayford PL (1982) Effect of cigarette smoke on human serum trypsin inhibitory capacity and antitrypsin concentrations. Am Rev Respir Dis 126:177-179 12. Christensen TG, Korthy AL, Snider GL, Hayes JA (1977) Irreversible bronchial goblet cell metaplasia in hamsters with elastase-induced panacinar emphysema. J Clin Invest 59:379-404 13. Deutsche Gesellschaft fiir Pneumologie und Tuberkulose (1988) Empfehlungen zur diagnostischen bronchoalveol~iren Lavage. Prax Klin Pneumol 42:119-122

276 14. DeWater R, Willems LNA, Van Mujen GNP, Franken C, Fransen JAM, Dijkman JH, Kramps JA (1986) Ultrastructural localization of bronchial antileukoprotease in central and peripheral human airways by a gold-labeling technique using monoclonal antibodies. Am Rev Respir Dis 133:882890 15. Dijkman JH, Franken C, Kramps JA (1987) Antileukoprotease in sputum during bronchial infections. In: Taylor JC, Mittman C (eds) Pulmonary emphysema and proteolysis. Academic Press, New York London, pp 289-297 16. Fritz H (1988) Human mucus proteinase inhibitor (human MPI). Biol Chem Hoppe Seyler 369 [Suppl] : 79-82 17. Gadek JE, Fells GA, Crystal RG (1979) Cigarette smoke induces functional antiprotease deficiency in the lower respiratory tract of humans. Science 206:1315-1316 18. Green NM, Work W (1952) Pancreatic trypsin inhibitor. Biochem J 54:347-352 19. Hochstrasser K, Reichert R, Schwarz S, Werle E (1972) Isolierung und Charakterisierung eines Protease Inhibitors aus menschlichem Bronchialsekret. Hoppe-Seylers Z Physiol Chem 353 : 221-226 20. Johnson D, Travis J (1979) The oxidative inactivation of human ~-proteinase inhibitor: further evidence for methionin at the reactive center. J Biol Chem 254:4022 4024 21. Kramps JA, Franken C, Meijer CJLM, Dijkman JH (1981) Localization of a low-molecular-weight protease inhibitor in serous cells of the respiratory tract. J Histochem Cytochem 29:712-719 22. Kramps JA, Franken C, Dijkman JH (1988) Quantity of anti-leukoprotease relative to cq-proteinase inhibitor in peripheral airspaces of the human lung. Clin Sci 75:351-353 23. Kramps JA, Appelhans H, Nikiforov T, Dijkman JH (1989) Trypsin, chymotrypsin and leukocyte elastase are inhibited by the second COOH-terminal domaine of antileukoprotease. Am Rev Respi Dis 139 :A201 24. Kueppers F, Bromke BJ (1983) Protease inhibitors in tracheobronchial secretions. J Lab Clin Med 101:742757 25. Laurell CB, Eriksson S (1963) The electrophoretic e~-globulin pattern of serum in cq-antitrypsin deficiency. Scand J Clin Lab Invest 15:132-140 26. Lellouch J, Claude J-R, Martin J-P, Orssaud G, Zaoui D, Bieth JG (1985) Smoking does not reduce the functional activity of serum alpha-l-proteinase inhibitor. Am Rev Respir Dis 132 : 818-820 27. Lowry OH, Rosenbrough NJ, Farr AL, Randall RJ (1951) Protein measurement with the Folin phenol reagent. J Biol Chem 193:265-275 28. Mann HB, Whitney PR (1947) On a test of whether one of two random variables is stochastically larger than the other. Ann Math Stat 18:50-60 29. Morrison HM, Kramps JA, Burnett D, Stockley RA (1987) Lung lavage fluid from patients with cq-proteinase inhibitor deficiency or chronic obstructive bronchitis: anti-elastase function and cell profile. Clin Sci 72:373-381 30. Mullen JBM, Wright JL, Wiggs BR, Pare PD, Hogg JC (1985) Reassessment of inflammation of airways in chronic bronchitis. B M J 291 : 1235-1239 31. Padrines M, Schneider-Pozzer M, Bieth JG (1989) Inhibition of neutrophil elastase by alpha-l-proteinase inhibitor

oxidized by activated neutrophils. Am Rev Respir Dis 139:783-790 32. Powers JC, Boone R, Carroll DL, Gupton BF, Kam CM, Nishino N, Sakamoto M, Thy PM (1984) Reaction of azapeptides with human leukocyte elastase and porcine pancreatic elastase. J Biol Chem 259:4288-4294 33. Schiessler H, Fink E, Fritz H (1976) Acid-stable proteinase inhibitors from human seminal plasma. Methods Enzymol 45:842859 34. Schiessler H, Hochstrasser K, Ohlsson K (1978) Acid stable inhibitors of granulocyte neutral proteases in human mucous secretions: biochemistry and possible biological functions. In: Havemann K, Janoff A (eds) Neutral proteases of human polymorphonuclear leukocytes. Urban & Schwarzenberg, Mfinchen Wien Baltimore, pp 195-207 35. Snider GL, Hayes JA, Franzblau C, Kagam HM, Stone PJ, Korthy AL (1973) Relationship between elastinolytic activity and experimental emphysema-inducing properties of papain preparations. Am Rev Respir Dis 110:254-262 36. Stockley RA, Afford SC (1984) Qualitative studies of lung lavage cq-proteinase inhibitor. Hoppe Seylers Z Physiol Chem 365:503-510 37. Stone PJ, McGowan SE, Bernardo J, Snider GL, Franzblau C (1983) Functional cq-protease inhibitor in the lower respiratory tract of cigarette smokers is not decreased. Science 221:1187-1189 38. Thurlbeck WM, Ryder RC, Sternby N (1974) A comparative study of the severity of emphysema in necropsy populations in three different countries. Am Rev Respir Dis 109:23%248 39. Trefz G, Heck B, Schulz V, Ebert W (1989) Functional activity of the e~-proteinase inhibitor in the serum and bronchoalveolar lavage fluid in acquired pulmonary emphysema. Prax Klin Pneumol 43:446451 40. Wewers MD, Herzyk DJ, Gadek JE (1989) Comparison of smoker and nonsmoker lavage fluid for the range of the association with neutrophil elastase. Am J Respir Cell Mol Biol 1:423-429 41. Willems LNA, Otto-Verberne CJM, Kramps JA, ten HaveOpbroek AAW, Dij kman JH (1986) Detection of antileukoprotease in connective tissue of the lung. Histochemistry 86:165-168 42. Willems LNA, Kramps JA, Stijnen T, Sterk PJ, Weening JJ, Dijkman JH (1989) Antileukoprotease-containingbronchiolar cells. Am Rev Respir Dis 139:1244-1250 43. Zimmerman M, Ashe B, Yurewicz EC, Patel G (1977) Sensitive assays for trypsin, elastase, and chymotrypsin using new fluorogenic substrates. Anal Biochem 78:47-51 Received: July 15, 1991 Returned for revision: October 4, 1991 Accepted: November 6, 1991 Prof. Dr. W. Ebert Thoraxklinik Heidelberg-Rohrbach Amalienstrasse 5 W-6900 Heidelberg, FRG