Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

It has been suggested that neutrophils play a role in the pathogenesis of lung injury in many chronic respiratory diseases such as chronic bronchitis, diffuse panbronchiolitis, emphysema and cystic fibrosis (McCusker and Hoidal, 1988; Fick and Reynolds, 1983; Mikami, 1991). However, the mechanism by which neutrophils promote lung injury remains incompletely understood.

Elastase is a neutral serine protease that occurs in the azurophilic granules of the neutrophil. This enzyme can digest many components of the extracellular matrix, including elastin, collagen types I, III, and IV, fibronectin and proteoglycans, and it has been suggested to play important roles in the metabolism of connective tissue and in the resolution of foreign substances. It is likely that excessive release of elastase, accompanying neutrophil lysis and phagocytosis, induces many pathological events and results in functional disorders as a result of the broad specificity and proteolytic intensity of this enzyme. Indeed, it has been reported that neutrophil elastase is significantly increased in sputum and bronchoalveolar lavage fluids of patients with various chronic respiratory diseases (Mikami, 1991; Piccioni et al., 1992; Stockley, 1987; Goldstein and Döring, 1986). Therefore, attention has been focused on neutrophil elastase in lung injury. However, because neutrophils release a number of other toxic factors, such as lipid metabolites, oxidants and proteases, the participation of neutrophil elastase in the lung injury promoted by neutrophils remains unclear.

SLPI is a 11.7-kDa serine protease inhibitor of 107 amino acids. It inhibits a wide range of proteases, including neutrophil elastase, cathepsin G, chymotrypsin and trypsin (Kramps et al., 1981). It has been localized in the granules of serous cells of bronchial glands (Kramps et al., 1981) and in Clara cells and goblet cells of the bronchial epithelium (Mooren et al., 1982; Kramps et al., 1987). SLPI consists of two domains. Recently, Barbara et al. (1990) indicated that domain 2 (C-terminal domain) contains neutrophil elastase, chymotrypsin and trypsin inhibitory sites and suggested that domain 1 (N-terminal domain) might also be necessary to express trypsin inhibitory activity. Therefore, Masuda et al. developed truncated SLPIs, corresponding to the C-terminal domain (Asn⁵⁵-Ala¹⁰⁷, Arg⁵⁸-Ala¹⁰⁷, Arg⁵⁹-Ala¹⁰⁷), by recombinant DNA technology. As they expected, truncated SLPI was shown to be more specific for elastase than for trypsin, compared with native SLPI (Masuda et al., 1992; Masuda et al., 1994). We proposed that these truncated SLPIs might be promising as therapeutics for the treatment of lung diseases mediated by neutrophils.

Accordingly, we undertook this study to evaluate the potency of truncated SLPI, an inhibitor of serine proteases, in acute lung injury models in hamsters and to investigate the role of neutrophil proteases in lung injury. We showed that the combination of LPS and FMLP was capable of making a specific neutrophil-mediated lung injury model and that this lung injury was mediated mainly by neutrophil proteases.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Materials. The reagents used in this study and their sources were the following: FMLP, catalase, 1-PI, indomethacin and dexamethasone (Sigma Chemical Co., St. Louis, MO), CMK (Bachem Feinchemikalien AG, Budendorf, Switzerland), LPS (E. coli 004) (Difco Laboratories, Detroit, MI) and SOD (Wako Chemical Co., Osaka, Japan). FMLP was dissolved in dimethylsulfoxide and diluted to the appropriate concentration in saline.

Exogenous elastase-induced acute lung injury model in hamsters. Male Golden Syrian hamsters, 8 to 10 weeks old, were anesthetized with halothane and administered HSE (50 µg/animal) intratracheally and transorally. Three hours after HSE administration, BAL was performed, according to the method of Barry et al. (1990). Briefly, hamsters anesthetized with urethane (1 g/kg intraperitoneally). Then they were intubated endotracheally, and BAL was performed 3 times by the rewash lavage method (6 ml of PBS). As a marker of lung injury, the hemoglobin content of the BAL fluid was determined by the cyanmethemoglobin method (Matsubara and Shibata, 1969). Inhibition of HSE-induced lung hemorrhage was determined by dosing the hamster with truncated SLPI either i.v. 2 min before HSE treatment or intratracheally 1 hr before HSE treatment.

Combination of LPS and FMLP-induced acute lung injury model in hamsters. Hamsters anesthetized with halothane were administered LPS (0.42 mg/kg) intratracheally, and this treatment was followed 24 hr later by intratracheal administration of FMLP (0.25 mg/kg). Ninety minutes after FMLP administration, BAL was performed. Inhibition of lung hemorrhage was determined by dosing the hamster with truncated SLPI either i.v. 2 min before FMLP treatment or intratracheally at the same time as FMLP treatment.

Administration of drugs. In LPS and FMLP-induced lung injury, we investigated the effects of several drugs. CMK, a specific elastase inhibitor, and human plasma-derived 1-protease inhibitor were intratracheally administered simultaneously with FMLP treatment. Truncated SLPIs were administered i.v. 2 min before FMLP treatment. Dexamethasone and indomethacin, the latter a cyclooxygenase inhibitor, were administered p.o. 1 hr before FMLP treatment. SOD (50,000 U/kg) and catalase (90,000 U/kg) were given s.c. 2 hr before FMLP treatment and were followed by SOD infusion (6000 U/kg/30 min) from 10 min after FMLP treatment, under the conditions described by Yoshikawa et al. (1990).

Assay for elastase-like activity. Elastase-like activity in BAL fluid was measured by use of a synthetic substrate of elastase, methoxysuccinyl-alanyl-alanyl-prolyl-valine-7-amido-4-methylcoumarin, according to Mario et al. (1979). Briefly, 75 µl of the supernatant of centrifuged BAL fluid was added to 1.4 ml of 0.05 M Tris buffer (pH 7.5) containing 0.5 M NaCl and 1 mM CaCl₂. This solution was incubated with 25 µl of 6 mM substrate at 37°C for 60 min. The fluorescence intensity was measured at excitation 370 nm and emission at 460 nm. A standard curve was constructed to convert relative fluorescence intensity into concentration of 7-amido-4-methylcoumarin released by enzymatic hydrolysis. By reference to a standard curve constructed with known concentrations of HSE, elastase-like activity in BAL fluid was presented as the concentration of HSE producing the same level of fluorescence. The source of elastase-like protease recognized in LPS-induced inflammation was suspected to be the neutrophil or the macrophage. Neutrophil elastase is a serine protease that is inhibited by DFP. On the other hand, macrophage elastase is a metallo protease inhibited by EDTA. So, to identify the source of protease, we measured elastase-like activity in the presence of DFP (5 mM) or EDTA (10 mM) (Abrams et al., 1984).

Histopathology. For histopathological study, the lungs were inflated with 10% formaldehyde solution at a constant pressure of 25 cm H₂O for 48 hr. The lungs were then removed, and sections were prepared and stained with hematoxylin-eosin.

Production of recombinant truncated SLPI. The recombinant truncated SLPI was produced by Teijin Limited, Tokyo, Japan. Briefly, DNA fragments encoding the C-terminal domain (Asn⁵⁵-Ala¹⁰⁷) and (Arg⁵⁸-Ala¹⁰⁷) of SLPI were synthesized chemically by use of appropriate codons from E. coli. The human growth hormone gene was fused to this domain to optimize the expression size by linkage with the DNA sequence for Leu-Val-Pro-Arg, which is cleavable by thrombin. The expression vector was constructed and introduced into E. coli HB 101. After the transformed cells had been cultured, the fusion protein was obtained as an inclusion body. It was then dissolved and cleaved with thrombin. The active inhibitor was purified by chromatography after S-S refolding (Masuda et al., 1992).

Statistical analysis. The data were expressed as the mean ± S.E.M. Groups were compared by Student's standard two-tailed t test and by one-way analysis of variance in conjunction with Dunnett's multiple comparison t test when appropriate. Groups were considered different if the P value was less than .05.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Exogenous elastase-induced acute lung injury model in hamsters. First we evaluated the potency of truncated SLPI to inhibit HSE-induced acute lung injury when administered i.v. and intratracheally. The assessment of HSE-induced acute lung injury was performed by measurement of the hemoglobin content in BAL fluid (Durham et al., 1994; Fletcher et al., 1990; Gillissen et al., 1993; Herbert et al., 1992). Intratracheal administration of truncated SLPI (0.01, 0.03 and 0.1 mg/kg) 1 hr before HSE administration inhibited acute lung hemorrhage in a dose-dependent manner (fig. 1). Pulmonary hemorrhage was inhibited by 31.4, 52.1 and 61.8%, respectively, with an ED₅₀ of 34.7 µg/kg.

View larger version (49K): [in this window] [in a new window]   Fig. 1.   Effect of intratracheal treatment with truncated SLPI, (Arg58-Ala107)SLPI, on exogenous elastase-induced acute lung injury in hamsters. Each value is the mean ± S.E.M. (Arg58-Ala107)SLPI was administered intratracheally 1 hr before HSE administration. *P < .05 and ** P < .01, significantly different from HSE-treated group.

View larger version (42K): [in this window] [in a new window]   Fig. 2.   Effect of i.v. treatment with truncated SLPI, (Arg58-Ala107)SLPI, on exogenous elastase-induced acute lung injury in hamsters. Each value is the mean ± S.E.M. (Arg58-Ala107)SLPI was administered i.v. 2 min before HSE administration. ** P < .01, significantly different from HSE-treated group.

Combination of LPS and FMLP-induced acute lung injury model in hamsters. To make a specific neutrophil-dependent model, we tried intratracheal administration of FMLP, a neutrophil stimulator, 24 hr after LPS administration. In the LPS-treated animals, the number of cells in the BAL fluid was significantly increased compared with that of intact animals. But the elastase-like activity was not changed. From the differential cell count analysis, the cells found in BAL fluid 24 hr after LPS administration were mostly neutrophils. The cellular proportions of neutrophils, alveolar macrophages, lymphocytes and eosinophils in BAL fluid were 86.9, 7.8, 4.6 and 0.8%, respectively. The additional treatment with FMLP at 24 hr after LPS treatment caused a very significant increase in elastase-like activity (P < .01), as well as a significant increase in hemoglobin concentration, compared with those levels for LPS-treated animals (P < .01, fig. 3).

View larger version (41K): [in this window] [in a new window]   Fig. 3.   Effect of intratracheal treatment with truncated SLPI, (Arg58-Ala107)SLPI, on the LPS and FMLP-induced acute lung injury in hamsters. Each value is the mean ± S.E.M. (Arg58-Ala107)SLPI was administered intratracheally simultaneously with FMLP administration. Truncated SLPI inhibited dose-dependently the lung hemorrhage and attenuated the increase in elastase activity. * P < .05 and ** P < .01, significantly different from LPS and FMLP-treated group.

View larger version (38K): [in this window] [in a new window]   Fig. 4.   Effect of i.v. treatment with truncated SLPI, (Arg58-Ala107)SLPI, on the LPS and FMLP-induced acute lung injury in hamsters. Each value is the mean ± S.E.M. (Arg58-Ala107)SLPI was administered i.v. 2 min before FMLP administration. Truncated SLPI inhibited dose-dependently the lung hemorrhage and attenuated the increase in elastase activity. * P < .05 and **P < .01, significantly different from LPS and FMLP-treated group.

Histopathological study. LPS-treated animals showed a remarkable diffuse infiltration of neutrophils around the bronchium and in the alveolar spaces, as well as perivascular edema. Addition of FMLP induced severe lung hemorrhage in the alveolar spaces and distortion of alveolar septa. Truncated SLPI attenuated these effects of the additional FMLP treatment of LPS-treated animals (fig. 5).

View larger version (K): [in this window] [in a new window]   Fig. 5.   Light micrographs of the lung field around an airway in hamsters with LPS and FMLP-induced lung injury. A) Intact animal (×61.25). B) LPS-treated animal (×61.25). C) LPS and FMLP-treated animal (×61.25). D) LPS and FMLP-treated animal with truncated SLPI-treatment (×61.25, 30 mg/kg i.v.).

Identification of elastase-like protease in BAL fluid of LPS and FMLP-induced lung injury model. Elastase-like activity of BAL fluid in the LPS and FMLP-induced injury model was completely inhibited by the addition of DFP, a serine protease inhibitor, but was scarcely reduced by the addition of EDTA, a metallo protease inhibitor (fig. 6).

View larger version (22K): [in this window] [in a new window]   Fig. 6.   Identification of elastase-like activity in BAL fluid of hamsters with LPS and FMLP-induced lung injury. Addition of DFP, but not that of EDTA, completely inhibited elastase activity of BAL fluid in the LPS and FMLP-induced injury model.

Effects of several drugs on LPS and FMLP-induced lung injury. Finally, we investigated the effects of several drugs on the lung injury induced by LPS and FMLP. These drugs included truncated SLPIs, a protease inhibitor, an elastase inhibitor, anti-inflammatory agents and radical scavengers. Intratracheal administration of 1-PI at 20 mg/kg, that of the specific elastase inhibitor CMK at 0.2 mg/kg and that of truncated SLPIs all inhibited lung hemorrhage significantly (table 1). Dexamethasone inhibited lung hemorrhage only slightly, and indomethacin aggravated rather than inhibited it. The radical scavengers (SOD and catalase) were ineffective under our experimental conditions.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

A number of elastase inhibitors have been evaluated for their activity to inhibit HSE-induced lung hemorrhage (Fletcher et al., 1990; Herbert et al., 1992; Shah et al., 1992; Durham et al., 1994). However, this model alone appears to be inappropriate for evaluation of elastase inhibitor activity in vivo, because it is not surprising that an elastase inhibitor would inhibit elastase-induced lung injury, especially that assessed after intratracheal administration. Thus we undertook this study not only to evaluate the potency of truncated SLPI by the HSE-induced lung injury model but also to develop a specific neutrophil-mediated lung injury model for use in evaluating the potency of truncated SLPI.

The presence of bacterial endotoxin in combination with neutrophil infiltration is known to result in lung injury (Pierce et al., 1977). And it has been suggested that the neutrophil plays a role in the pathogenesis of endotoxin-associated lung injury (Repine et al., 1982). But the mechanism by which endotoxin results in lung dysfunction appears to be complex, because it is known that endotoxin injures lung tissue directly and stimulates monocytes to release tumor necrosis factor (Bachwich et al., 1986), interleukin-1 (Wewers et al., 1984) and numerous lipid mediators (Brown et al., 1988; Rylander and Beijer, 1987). Actually, although intratracheal administration of LPS produces lung injury, we could not find any correlation between the degree of lung injury and the number of neutrophils (data not shown). These findings imply that LPS-induced acute lung injury may be associated not only with neutrophils but also with many other factors.

Accordingly, to make a specific neutrophil-dependent model and to investigate the mechanism by which neutrophil promotes lung injury, we tried intratracheal administration of FMLP, which causes the release of neutrophil proteases and oxidants, 24 hr after LPS administration. We found that additional treatment with FMLP aggravated the LPS-induced lung injury, and we accepted this new model to be one of neutrophil-mediated, exudative and hemorrhagic lung injury based on histopathological observation.

The i.v. and intratracheal administration of truncated SLPI attenuated this lung injury, and the increase in elastase-like activity in LPS and FMLP-treated animals prompted us to consider that neutrophil protease might participate in the lung injury. Thus we focused our interest on elastase-like protease and investigated its possible participation in the injury. The source of elastase-like protease was suggested to be neutrophil from the results of DFP and EDTA treatment (fig. 6). Further, to clarify the participation of neutrophil protease in this injury model, we investigated the effects of 1-PI, of CMK, a synthetic neutrophil elastase inhibitor, of our recently developed truncated SLPIs and of several anti-inflammatory drugs (table 1). This lung injury was attenuated only by administration of elastase inhibitors such as 1-PI, CMK and the truncated SLPIs, and not by dexamethasone, indomethacin and radical scavengers. Weiss et al. (1986) described the synergy between neutrophil elastase and oxidants in neutrophil-induced tissue injury and indicated that the injury may be mediated in large part by the action of neutrophil elastase. Therefore, our results appear to support their contention. These results suggest that LPS and FMLP-induced lung injury is mediated mainly by neutrophil proteases.

Interestingly, our data show that a nearly 20-fold higher dose of SLPI (0.85 mg/kg vs. 47 µg/kg) was required to inhibit LPS and FMLP-induced lung hemorrhage than to inhibit the HSE-induced injury. In these two models, we used different administration timing. Specifically, in the HSE-induced model we evaluated the efficacy of SLPI as a pretreatment because it seems reasonable to assume that elastase-induced lung injury would be inhibited by co-administration with its inhibitor. In the LPS/FMLP-induced model, this consideration does not arise. We chose co-administration because we thought the physical irritation of successive intratracheal administrations might injure the respiratory tract of the animals and thus affect the results. However, we do not believe that the different experimental conditions caused the discrepancy in the results. It appears that LPS and FMLP-induced lung injury is mediated by hamster neutrophil proteases. Recently we found a species difference in the inhibitory activity of truncated SLPI toward elastase; i.e., its inhibition of hamster elastase was 30-fold weaker than that of human elastase (data not shown). We think that this difference explains the difference in effective dose between the two models. Gillissen et al. (1993) indicated that SLPI augments the glutathione level in lung epithelial lining fluids in sheep. The increase in the glutathione level in such fluids means an increased antioxidant screen on the respiratory epithelial surface. Moreover, truncated SLPI inhibits not only neutrophil elastase but also cathepsin G, chymotrypsin, trypsin, collagenase and chymase (Masuda et al., 1994). Therefore, this inhibitory activity against multiple proteolytic enzymes and oxidants might have added to the protective effects of the truncated SLPI on neutrophil-mediated acute lung injury. Moreover, in a recent study using an immunofluorescence technique, Willems et al. (1986) showed SLPI to be localized in association with elastic fibers in airway walls and alveolar septa. This evidence suggests that SLPI may play a role in protecting not only the epithelium but also the connective tissues from the enzymatic injury. We have preliminary data showing that the truncated SLPI administered i.v. becomes distributed to the lung tissue quickly. These results imply the efficacy not only of topical but also of systemic treatment with it.

In conclusion, our study has demonstrated that truncated SLPI attenuates the HSE-induced acute lung injury and the neutrophil-mediated acute lung injury induced by the combination of LPS and FMLP administration. Therefore, truncated SLPI appears to be a good candidate for the treatment of destructive lung diseases due to neutrophils, such as chronic bronchitis, emphysema and ARDS.