Aerosolized Administration of N-Acetylcysteine Attenuates Lung Fibrosis Induced by Bleomycin in Mice

Transcription

1 Aerosolized Administration of N-Acetylcysteine Attenuates Lung Fibrosis Induced by Bleomycin in Mice SHIN-ICHI HAGIWARA, YOSHIKI ISHII, and SATOSHI KITAMURA Department of Pulmonary Medicine, Jichi Medical School, Tochigi, Japan Reactive oxygen species (ROS) play an important role in the pathogenesis of pulmonary fibrosis. We previously demonstrated that N-acetylcysteine (NAC), an antioxidant, inhibited adhesion molecule expression and cytokine production in lung cells. When NAC is inhaled into the alveolar space, it is expected to directly interact with inflammatory cells and to elevate glutathione levels in the epithelial lining fluids. We therefore examined whether inhaled NAC inhibits lung fibrosis induced by bleomycin (BLM). Male ICR mice were given a single intravenous injection of BLM (150 mg/ kg). Thirty milliliters of NAC (70 mg/ml) or saline were inhaled twice a day for 28 d using an ultrasonic nebulizer. In the inflammatory phase (Day 7), NAC administration attenuated the cellular infiltration in both bronchoalveolar lavage fluid (BALF) and alveolar tissues. At Day 28, the fibrotic changes estimated by Aschroft s criteria and hydroxyproline content in the NAC inhalation group were significantly decreased compared with the BLM-only group (p 0.05). CXC chemokines, macrophage inflammatory protein-2 (MIP-2), cytokine-induced neutrophil chemoattractant (KC), and CC chemokines, macrophage inflammatory protein-1 (MIP-1 ), in BALF were mostly elevated on Day 7 in the BLM-only group; however, these elevations were significantly repressed by NAC inhalation (p 0.05). Lipid hydroperoxide (LPO) was also quantified in BALF. LPO was markedly increased on Day 3 in the BLM-only group, and this increase was significantly decreased by NAC inhalation (p 0.05). These results revealed that aerosolized NAC ameliorated acute pulmonary inflammation induced by BLM injection via the repression of chemokines and LPO production, resulting in the attenuation of subsequent lung fibrosis. These findings are limited to the BLM-induced lung fibrosis animal model. However, NAC inhalation will be expected to be a potential therapy for patients with other interstitial pneumonias because ROS are involved in the pathogenesis of lung injury in most interstitial pneumonia. Fibrosis is a reactive or reparative process characterized by the formation of excessive fibrous tissue. In the lung, inflammation and immune processes are the major pathogenic mechanisms that injure tissue and stimulate fibrosis (1). There is considerable evidence that oxygen-generated free radicals play a major role in inflammatory and immune-mediated tissue injury (2 4). Bleomycin (BLM) is an antineoplastic agent. The mechanism of the antineoplastic effect of BLM is that the BLM-iron complex reduces molecular oxygen to superoxide and hydroxy radicals that can then attack DNA and cause strand cleavage (5). The role of oxygen free radicals has been supported by studies showing that the addition of superoxide dismutase, an oxygen free radical scavenger, inhibits BLM-induced DNA breakage and cellular damage (Received in original form March 26, 1999 and in revised form September 21, 1999) Correspondence and requests for reprints should be addressed to Shin-ichi Hagiwara, M.D., Department of Pulmonary Medicine, Jichi Medical School, Minamikawachimachi Kawachigunn, Tochigi , Japan. kokyu2@ jichi.ac.jp Am J Respir Crit Care Med Vol 162. pp , 2000 Internet address: in vitro (6, 7). On the other hand, BLM induces pulmonary fibrosis as an adverse effect since the hydrolase that inactivates BLM is relatively scarce in lung tissue. Therefore, a BLM-induced pulmonary fibrosis model in mice is a helpful tool to examine the general mechanism of fibrosis, especially that mediated by oxygen free radicals. Oxidants produced by inflammatory cells are considered to play a role in lung injury in idiopathic pulmonary fibrosis (IPF). Antioxidants are decreased in bronchoalveolar lavage (BAL) from patients with IPF since the level of glutathione (GSH) was found to be approximately one third of the normal concentration (8). It is possible to raise the level of GSH in bronchoalveolar lavage fluid (BALF) in patients with IPF by aerosol administration of GSH, and the release of oxidants by cells isolated from the BAL was found to be decreased by aerosolized GSH (9). N-acetylcysteine (NAC) is not only a precursor of GSH but also shows a direct scavenging ability of oxygen free radicals (10, 11). Moreover, it regulates the production of some cytokines or the expression of adhesion molecules on endothelial cells and bronchial epithelial cells. On the basis of these findings, the effectiveness of NAC administration on animal models of lung fibrosis has been reported (12 14). Recently, Behr and colleagues (15) reported that oral high dose administration of NAC with low dose oral steroids to patients with IPF significantly improved the lung function index. However, NAC does not produce a sustained increase in GSH levels sufficient to increase the antioxidant capacity of the lungs, even when given in high oral doses (16). Furthermore, NAC itself is not detected in BALF nor in lung tissue when given orally (17). In contrast, aerosol administration of NAC directly act as an antioxidant in alveoli in addition to the GSH increasing effect. Therefore, aerosolized NAC may attenuate the inflammation and lung fibrosis more effectively. We examined here whether aerosol NAC administration inhibited lung inflammation and fibrosis induced by BLM injection in mice. Moreover, to evaluate the inhibitory mechanisms, the levels of chemokines and lipid hydroperoxide (LPO) in BALF were measured and compared among the study groups. METHODS Administration of BLM BLM hydrochloride (Nippon Kayaku, Tokyo, Japan) was injected intravenously into 7-wk-old ICR male mice once on Day 0 at a dose of 150 mg/kg body weight dissolved in 200 l saline. Control mice received an equivalent amount of saline. NAC Preparation and Administration N-acetyl-L-cysteine (Wako, Osaka, Japan) was dissolved in saline at a concentration of 70 mg/ml and neutralized at ph 7.0. Thirty milliliters of NAC solution or saline were inhaled twice a day in a chamber using an ultrasonic nebulizer. The inhalation was started 3 d before BLM or saline injection. The volume of the chamber was about 45 L maintained under normoxic and normocapnic conditions.

2 226 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL BAL Analysis BAL was performed through a tracheal cannula with 1 ml saline five times. In each mouse examined, approximately 4.5 ml (90%) of BALF were recovered. A 100- l aliquot was used for the total cell count, and the remainder was immediately centrifuged at 1,000 rpm for 10 min. The total cell count was performed using a hemocytometer, and cell differentiation was performed on cytocentrifuge slides with Wright-Giemsa staining of more than 500 cells. The supernatants of BALF were stored at 20 C to measure chemokines, GSH, and total protein. LPO was extracted from 1 ml of BALF using the extract solution and the extract was stored at 80 C. GSH levels in BALF were measured to determine the time-dependent changes in GSH concentration during 1 d using the GSH-400 kit (Oxis International Inc., Portland, OR). When NAC was inhaled twice a day for 2 wk, the GSH concentration 6 h after the last inhalation was measured. The concentration of total protein in BALF was measured by the BCA Protein Assay kit (Pierce Chemical Co., Rockford, IL). CXC chemokines such as macrophage inflammatory protein-2 (MIP-2), cytokine-induced neutrophil chemoattractant (KC), and CC chemokines such as macrophage inflammatory protein-1 (MIP-1 ) and monocyte chemoattractant protein-1 (MCP-1) were measured in BALF. MIP-2 was measured by the Mouse MIP-2 ELISA kit (R&D Systems, Minneapolis, MN). KC was measured by the Mouse KC sandwich EIA kit (Immuno-Biological Laboratories Co., Ltd., Gunma, Japan). MIP-1 was measured by the Mouse MIP- 1 ELISA kit (R&D Systems). MCP-1 was measured by the Mouse MCP-1 Immunoassay kit (Biosource International, Camarillo, CA). The amount of LPO in BALF was measured as an indicator of oxidative stress in the alveolar milieu by the Lipid Hydroperoxide Assay kit (Cayman Chemical, Ann Arbor, MI). Morphologic Evaluation and Measurement of Hydroxyproline Content Histopathologic evaluation was performed on animals that were not subjected to BAL. The left lungs were removed and inflated with 10% formaldehyde neutral buffer solution, and longitudinal tissue sections were stained with hematoxylin-eosin and Masson trichrome. Fibrotic findings in each section were scored using the criteria of Ashcroft and colleagues (18). Briefly, the grade of lung fibrosis was scored on a scale from 0 to 8 by examining more than 25 successive fields at a magnification of 100. The mean of all scores obtained from each field was employed as the fibrotic score of the specimen. To avoid observer bias, three observers, including a histopathologist, interpreted the images independently in a blinded fashion, and the mean of the three observers findings was considered to be the fibrotic score of the specimen. The right lungs were homogenized using a Polytron homogenizer (Kinematica, Steinhofhalde, Switzerland) at 3,500 rpm for 90 s and the hydroxyproline content was measured by Woessnor s method (19). Experimental Protocol We initially evaluated the cellular and morphologic changes in BALF and lung tissue after BLM injection. BAL was performed on Day 0 (before BLM or saline injection) and on Days 3, 7, 14, 21, and 28. In addition, histologic examinations were performed on these days. The concentrations of total protein and GSH in BALF were measured on the corresponding days. In the preliminary studies, no differences were found in histologic changes or hydroxyproline contents between the group inhaling saline after saline injection and the group inhaling NAC after saline injection. Therefore, we used the group inhaling NAC for 4 wk after saline injection as the control group. To evaluate the effect of NAC administration on BLM-induced lung injury, two experiments were designed. In Experiment 1, we examined lung fibrosis by measuring the hydroxyproline content and histopathologic change 28 d after BLM injection. The study groups were divided into five groups as shown: The S 4N group was a control group that received saline injection and NAC inhalation for 4 wk. The B 4S group received BLM injection and saline inhalation for 4 wk. The B 4N group received BLM injection and NAC inhalation for 4 wk. To examine the early and delayed effects of NAC administration on BLM-induced lung fibrosis, we compared the B 2N 2S and B 2S 2N groups. The former received NAC inhalation for 2 wk immediately after BLM injection. The latter received NAC inhalation for the last 2 wk after BLM injection. From 3 days before BLM injection, NAC inhalation was started in the B 4N and the B 2N 2S groups and saline inhalation in the other groups 3 d before BLM injection. In Experiment 2, we examined the effect of NAC inhalation on cellular infiltration in lungs induced by BLM, and also measured chemokines and LPO in BALF to determine whether NAC inhibits the chemokine production and has the scavenging ability of oxygen radicals. Statistical Analysis All data were exposed as means SEM. The data of chemokines and LPO were compared using the unpaired t test. To compare the hydroxyproline contents of each group, one-factor analysis of variance followed by Fisher s least significant difference test were used to detect differences between the groups. To compare the Ashcroft score of each group, the Kruskal-Wallis test followed by Dunn s test were used, and a probability value of less than 0.05 was considered statistically significant. RESULTS BAL Analysis after BLM Injection The recovery rates of BALF in all groups were greater than 90%. The time-dependent change of total GSH concentration in BALF after single inhalation of NAC is shown in Figure 1A. Four hours after NAC inhalation, GSH levels showed a peak and the elevated level of GSH was maintained for more than 12 h. The time course of the total GSH concentration in BALF after 2 wk of aerosolized NAC administration twice a day is shown in Figure 1B. The GSH concentration in BALF obtained 6 h after the last NAC inhalation reached a plateau 3 d Figure 1. (A) Changes of GSH concentration in BALF after the last inhalation. Elevated level of GSH was maintained for more than 12 h after single inhalation of NAC. (B) Time course of GSH concentration in BALF with 2 wk of NAC inhalation. GSH concentration in BALF peaked 3 d after the initiation of NAC inhalation (n 3 at each time examined).

3 Hagiwara, Ishii, and Kitamura: N-acetylcysteine Represses Bleomycin-induced Lung Injury 227 Figure 2. Time course of total cell counts and cell differentiation in BALF for 4 wk after BLM injection and the effect of NAC inhalation. Total cell counts significantly increased on Day 14 and remained unchanged until Day 28. Macrophage count (A) shows a similar course to the total cell count (B), whereas neutrophils (C) and lymphocytes (D) were significantly increased with the maximal levels on Days 7 and 14, respectively. Neutrophils and lymphocytes were significantly attenuated by NAC (n at each time examined). after the initiation of NAC administration. Therefore, in the following experiments, we started NAC administration 3 d before BLM injection. The changes in total cell counts and differential cell counts in BALF are shown in Figure 2. Total cell counts began to increase from Day 3 and reached a plateau on Day 14 (Figure 2A). The increased cellularity remained almost unchanged until Day 28. The macrophage count showed a similar course to the total cell count (Figure 2B). On the other hand, both neutrophils (Figure 2C) and lymphocytes (Figure 2D) showed peak levels between Days 7 and 14 and gradually decreased after Day 14. The concentration of total protein increased in accordance with the increase of neutrophils and lymphocytes in BALF and showed a gradual decrease until Day 28 (data not shown). Morphologic Changes after BLM Injection On Day 3, slight edematous thickening of alveolar septa and cellular infiltration were seen. On Day 7, alveolar thickening with infiltration of macrophages, a few lymphocytes, and neutrophils was observed. In the alveolar space, macrophages were also increased. On Day 14, alveolar thickening and cellular infiltrations had become more severe, and fibrosis had begun to develop in the thickened septa and intra-alveolar spaces. In the lung tissue, neutrophils were decreased and lymphocytes increased compared with Day 7. On Day 28, diffuse fibrosis was formed at the subpleural region (not shown). Effects of NAC Inhalation on Cellular Changes in BALF The maximal increase in cellularity in BALF was observed between Days 7 and 14. NAC inhalation tended to diminish the increase in total cell count (Figure 2A) and macrophage count (Figure 2B) in BALF, but this was not significant except for the total cell counts on Day 14. The neutrophil count was significantly decreased on Days 7 and 14 after NAC inhalation (p 0.01, p 0.05, respectively) (Figure 2C). The marked elevation of the lymphocyte count was also significantly repressed by NAC inhalation (p 0.01) (Figure 2D). Effects of NAC Inhalation on Histopathologic Changes The histopathologic changes showed a predilection for the subpleural parenchyma in all specimens treated with BLM. The severity of alveolitis was lower in the B 4N group than in the Figure 3. Histopathologic comparison between BLM injection plus saline inhalation (B 4S) and BLM injection plus NAC inhalation (B 4N) 7 d after BLM treatment. Alveolar thickenings with the cellular infiltration of monocytes, lymphocytes, and a few neutrophils are seen. In the alveolar space, macrophages are also seen (A). These features were slight in the B 4N group (B) (bar 10 m).

4 228 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL Figure 4. Histopathologic changes 28 d after BLM injection. Control tissue (S 4S) shows normal findings (A). Diffuse fibrosis with cellular infiltration and destruction of the alveolar structure were recognized in the B 4S group (B). In the B 2N 2S group (C), fibrous thickenings of alveolar septa were also recognized; however, the degrees of fibrosis and alveolar destruction were slight compared with those of the B 4S group. In the B 2S group (D), the degree of fibrosis was as severe as that of the B 4S group (bar 250 m). B 4S group (Figures 3A and 3B). As mentioned above, fibrotic change in the lung tissue was established by Day 28; therefore, the effect of NAC inhalation on fibrotic formation was evaluated on Day 28. On Day 28, the control group (S 4N) revealed a normal alveolar structure without cellular infiltration or fibrous thickening (Figure 4A). Diffuse fibrosis with destruction of the alveolar structure was observed mainly in the subpleural regions in the B 4S group (Figure 4B). In the B 4N and B 2N 2S groups, the degree of fibrosis was similar and lower than in the B 4S group (Figure 4C). In the B 2S 2N group, the fibrotic changes were as severe as those in the B 4S group (Figure 4D). To evaluate the histologic changes, we scored the microscopic findings according to the criteria of Ashcroft and colleagues (18). The Ashcroft score was highest in the B 4S group ( ). In the B 4N or B 2N 2S groups ( , , respectively), the Ashcroft score was significantly lower than that of the B 4S group (p 0.05). However, in the B 2S 2N group ( ), the score showed no significant difference (Figure 5A). Lung Hydroxyproline Analysis A comparison of the hydroxyproline contents among the five groups is presented in Figure 5B. The hydroxyproline contents in the B 4N and B 2N 2S groups ( , mg/ml, respectively) exhibited no significant differences compared with the S 4N group ( mg/ml), whereas they were significantly decreased compared with the B 4S group ( mg/ml) (p 0.05). The hydroxyproline contents in the B 4N and B 2N 2S groups were almost equal. No significant difference was observed between the B 4S and B 2S 2N groups. The hydroxyproline contents reflected the histopathologic changes of each group. Two weeks of early NAC administration was as effective as 4 wk of NAC administration in repressing lung fibrosis. Effects of NAC Inhalation on Chemokine Production and the Amount of Lipid Hydroperoxide in BALF The levels of MIP-2 (Figure 6A) and KC (Figure 6B), which are chemotaxins for neutrophils similar to human IL-8, peaked Figure 5. (A) Comparison of the Ashcroft score among the study groups. The Ashcroft score was highest in the B 4S group. In the B 4N or B 2N 2S groups, it was significantly lower than that of the B 4S group. However, in the B 2S 2N group, it showed no significant difference. (B) Hydroxyproline content in the right lung in five groups. Hydroxyproline content was significantly increased because of BLM injection. NAC inhalation for 4 wk or for the first 2 wk significantly attenuated the BLM-induced increase in hydroxyproline content; however, NAC inhalation for the last 2 wk showed no effect (n 5 in each group).

5 Hagiwara, Ishii, and Kitamura: N-acetylcysteine Represses Bleomycin-induced Lung Injury 229 Figure 6. Chemokines in BALF. CC chemokines, MIP-1, and CXC chemokines, MIP-2 and KC peaked on Day 7. These elevations were significantly repressed by NAC. However, another CC chemokine, MCP-1, gradually increased after Day 3, and this increase was not significantly repressed by NAC. The levels of NIP-1, MCP-1, and MIP-2 in the S 4N group were under the detectable limit. KC levels in the S 4S group were all about 40 rg/ml during the 4-wk period (n 3 at each time examined). on Day 7 and rapidly decreased thereafter. These elevations were significantly attenuated by NAC inhalation (p 0.05). MIP-1 (Figure 6C) and MCP-1 (Figure 6D) are chemotaxins for macrophages and lymphocytes. MIP-1 showed a course similar to CXC chemokines, that is, it peaked on Day 7, and the elevation was significantly attenuated by NAC inhalation (p 0.05). On the other hand, MCP-1 gradually increased from Day 3 and the elevated level remained until Day 28. NAC inhalation slightly attenuated the elevation of MCP-1 during this period (Figure 6D). LPO peaked on Day 3 and decreased soon after Day 14 (Figure 7). LPO production was also significantly lowered by NAC inhalation on Days 3, 7, and 14 (p 0.05). DISCUSSION The pathophysiology of BLM-induced pneumonitis is considered to consist of two phases. One is the early inflammatory phase characterized by cellular infiltration such as macrophage, neutrophil and lymphocyte infiltration into the interstitium and alveolar spaces. The other is the subsequent late fibrotic phase. Reactive oxygen species and proteases generated from infiltrated inflammatory cells are considered to injure the lung tissue, and excessive fibrosis occurs as a reparative process. Figure 7. The changes of the amount of LPO for 4 wk. LOP production in BALF was most increased on Day 3 and was significantly decreased by NAC (n 3 at each time examined). In the present study, aerosolized administration of NAC during all experimental periods showed a significant inhibitory effect on BLM-induced lung fibrosis. NAC administration during only the first 17 d showed a similar effect as that during the whole period, but administration during only the last 2 wk did not. Therefore, NAC is considered to affect BLM-induced lung fibrosis at the early inflammatory phase. Possible explanations for the antifibrosis effect of NAC are as follows: (1) inhibition of cellular infiltration, (2) scavenging the reactive oxygen species produced from inflammatory cells and lung cells, (3) direct protective detoxification of BLM-generated radicals before they damage tissue, and (4) inhibition of fibroblast proliferation. To determine the role of cellular infiltration inhibition, we examined the effect of NAC inhalation on BLM-induced cellular accumulation in lungs by BAL and histologic analysis. Aerosolized NAC administration significantly attenuated the BLM-induced increase in the total cell count and numbers of neutrophils and lymphocytes in BALF 7 and 14 d after BLM injection. The histologic examination supported this finding. Neutrophils and neutrophil products have been identified in increased amounts in the air spaces of patients with interstitial lung disease (ILD) and in animal models of lung fibrosis (20 23). These cells induce parenchymal injury by producing toxic radical oxygen species and by secreting a variety of proteolytic enzymes, including elastase, collagenase, and other neutral proteases. Recruited monocytes also contribute to pulmonary inflammation by the elaboration of oxygen radical species, proteolytic enzymes, and factors that attract additional inflammatory cells. T cells and B cells, like monocytes, are present in increased numbers in the lung interstitium in ILD (24), and the recruitment of lymphocytes to the lung has been shown to precede the development of pulmonary fibrosis in models of lung fibrosis (25). Various cytokines and adhesion molecules are involved in cellular accumulation in acute lung injury. Interstitial and alveolar macrophages from patients with ILD constitutively express MIP-1 and MCP-1, and cells other than lung macrophages contribute to the production of these leukocyte chemotaxins (26 28). These leukocyte chemoattractants belong to the supergene families of chemotactic and activating

6 230 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL cytokines, which include CXC, CC, and C chemokines (29 31). These chemokines may be critically involved in the migration and activation of leukocytes in ILD (32). We measured chemokines in BALF to evaluate the influence on cellular accumulation into the lung induced by BLM. The CXC chemokines, MIP-2 and KC, were markedly elevated at the acute inflammatory phase (Day 7), and NAC inhalation significantly repressed the production of those chemokines. We demonstrated that BLM stimulated IL-8 and MCP-1 production from pulmonary vascular endothelial cells and the expression of intercellular adhesion molecule-1 (ICAM-1) on the cells in vitro, and that NAC inhibited the effect of BLM (unpublished data). These chemokines and adhesion molecules are reported to be stimulated through the nuclear transfer factor, NF- B activation (33). Because NAC inhibits NF- B activation (34, 35), we speculate that NAC may repress the chemokine production and adhesion molecule expression through the inactivation of NF- B, thereby decreasing inflammatory cell accumulation into the lungs. The mechanism by which lung inflammation with cellular infiltrates causes the subsequent lung fibrosis has not yet been elucidated. However, a correlation was found between the severity of lung injury and subsequent lung fibrosis in BLM-injected rabbit models (36). The second and third possible explanation for the NAC effect is its antioxidant effect. We measured the amount of LPO in BALF, which was one of the indicators of oxidative stress, and our results showed the scavenging ability of oxygen radicals by direct NAC inhalation. Many findings suggest that reactive oxygen species are involved in the mechanism of BLM-induced lung injury. The mechanism of cell damage by BLM appears to act via the generation of free oxygen radicals, which participate in DNA scission (37), and lipid peroxidation of the cell membrane (38). It has been shown in vitro that superoxide dismutase inhibits BLM-dependent DNA chain breakage (37). NAC is a precursor of GSH and, in addition, a scavenger of reactive oxygen species such as hydrogen peroxide, hypochlorous acid, and hydroxiradical (39). Borok and colleagues (9) reported that aerosol GSH influenced cellular events since spontaneous release of oxidants by alveolar macrophages was significantly decreased. Moreover, Forman and Skelton (40) observed similar findings that rat alveolar macrophages utilize extracellular GSH to synthesize intracellular GSH and thus gain protection against oxidant damage. Norbert (14) reported that when both BLM and NAC were injected intratracheally into the rat lung, pathologic changes were minimal compared with those in the BLM-only group. The results of his study demonstrated almost complete protection from BLM lung toxicity. Although the route of BLM administration was different, our data showed similar results. NAC inhalation likely showed both a direct scavenging ability of BLM-generated oxygen radicals and the protective effect through GSH synthesis. Our data also indicated that the production of LPO preceded the elevations of CXC and CC chemokines. These findings suggest that reactive oxygen radicals were released earlier than chemokine production and might be related to stimulation of chemokine gene expression. GSH, an antioxidant, is decreased in the epithelial lining fluids (ELF) in patients with idiopathic pulmonary fibrosis (IPF), and GSH in ELF can be supplemented by the administration of aerosolized NAC. Therefore, it is also considered to be important that the enhanced scavenging ability of reactive oxygen species by increased GSH in the alveolar space inhibits lung tissue injury. We started NAC inhalation 3 d before BLM injection because it took 3 d for GSH levels in BALF to reach a plateau. There is a possibility that the 3 d of NAC preceding BLM was protective for the induction of pulmonary inflammation. To test this possibility, we compared the BALF findings and histopathology of the 3-d NAC B 4S group with the 3-d saline B 4S group on Days 7 and 28. However, the examination resulted in no significant differences in BALF cells and histopathology between the two groups (data not shown). These results suggested that the observed benefits were attributed to the 14-d treatment with NAC after BLM injection. From the findings that the peaks of cytokine production, LPO levels, and inflammatory cell accumulation were observed during the first 2 wk, NAC inhalation seems to be necessary in this early inflammatory phase. The last possible explanation for the NAC effect is that NAC inhibits fibroblast proliferation in the late fibrotic phase. It has been reported that NAC inhibits fibroblast proliferation in vitro (41). However, this explanation is not likely in vivo because NAC inhalation during the latter 2 wk showed no inhibitory effect on BLM-induced lung fibrosis in the present study. Concerning the correlation between lung fibrosis and MCP-1, MCP-1 enhances the procollagen gene expression of rat lung fibroblasts via the promotion of transforming growth factorbeta 1 (TFG- 1 ) gene expression (42). Our data did not show that NAC could inhibit the production of MCP-1 significantly; therefore, the production of growth factors such as TFG- 1 was not repressed by NAC, and this was considered to be one of the causative factors. Direct administration of GSH aerosol may augment the GSH levels in ELF; however, the elevated period of GSH occurred for only about 2 h after administration (43). To maintain a high level of GSH in ELF, inhalation of GSH might be required many times a day. We evaluated the effect of aerosolized NAC on BLM-induced pulmonary fibrosis in mice. Administration of aerosolized NAC at a concentration of 70 mg/ml twice a day maintained an elevated level of GSH all day. In addition, with aerosolized administration of NAC; not only GSH but also NAC itself may act on the inflammatory alveolar milieu. In conclusion, NAC inhalation showed an inhibitory effect on BLM-induced acute inflammation after pulmonary fibrosis. The effect was limited when NAC was administered in the acute inflammatory phase. This may be due to the inhibition of chemokine production or scavenging of reactive oxygen radicals by NAC itself. NAC is largely free of adverse effects and has been used for many years as a mucolytic agent by direct instillation or nebulization into the airway (44). Although our findings are limited to the BLM-induced animal model, NAC inhalation is expected to be a potential therapy for other interstitial pneumonia because ROS are involved in the development of almost all interstitial pneumonia. Acknowledgment : The writers thank M. Hironaka, M.D., for advice regarding the histopathologic examinations. References 1. Phan, S. M Diffuse interstitial fibrosis. In D. Massaro, editor. Lung Cell Biology. Marcel Dekker, New York Fridovich, I The biology of oxygen radicals. Science 201: Freeman, B. A., and J. D. Crapo Biology of disease: free radicals and tissue injury. Lab. Invest. 47: Schraufstatter, I. U., and C. C. Cochrane Oxidants; types, sources and mechanisms of injury. In R. G. Crystal and J. B. West, editors. The Lung: Scientific Foundations. Raven Press, New York Sausville, E. A., R. W. Stein, J. Peisach, and S. B. Horwitz Properties and products of the degradation of DNA by bleomycin and iron(ii). Biochemistry 17: Galvan, L., C. H. Huang, A. W. Prestayko, J. T. Stout, J. E. Evans, and S. T. Crooke Inhibition of bleomycin-induced DNA breakage by superoxide dismutase. Cancer Res. 41: Cunningham, M. L., P. S. Ringrose, and B. R. Lokesh Inhibition of the genotoxicity of bleomycin by superoxide dismutase. Mutat. Res. 135:

7 Hagiwara, Ishii, and Kitamura: N-acetylcysteine Represses Bleomycin-induced Lung Injury Cantin, A. M., R. C. Hubbard, and R. G. Crystal Glutathione deficiency in the epithelial lining fluid of the lower respiratory tract of patients with idiopathic pulmonary fibrosis. Am. Rev. Respir. Dis. 139: Borok, Z., R. Buhl, G. J. Grimes, A. D. Bokser, R. C. Hubbard, K. J. Holroyd, J. M. Roum, D. B. Czerski, A. M. Cantin, and R. G. Crystal Glutathione aerosol therapy to suppress the burden of oxidants on the alveolar epithelial surface in idiopathic pulmonary fibrosis. Lancet 338: Gillissen, A., and D. Nowak Characterization of N-acetylcysteine and ambroxol in anti-oxidant therapy. Respir. Med. 92: Henry, J. H., L. Rui-ming, and T. Li Glutathione cycling in oxidative stress. In L. B. Clerch and J. M. Donald, editors. Oxygen, Gene Expression, and Cellular Function. Marcel Dekker, New York Jamieson, D. D., D. R. Kerr, and I. Unsworth Interaction of N- acetylcysteine and bleomycin on hyperbaric oxygen-induced lung damage in mice. Lung 165: Shahzeid, S., B. Sarnstrand, P. K. Jeffery, R. J. McAnulty, and G. J. Laurent Oral N-acetylcysteine reduces bleomycin-induced collagen deposition in the lungs of mice. Eur. Respir. J. 4: Norbert, B Inhibition of bleomycin lung toxicity by N-acetyl cysteine in the rat. Pathology 17: Behr, J., K. Maier, B. Degenkolb, F. Krombach, and C. Vogelmeier Antioxidative and clinical effects of high-dose N-acetylcysteine in fibrosing alveolitis. Am. J. Respir. Crit. Care Med. 156: Bridgeman, M. M. E., M. Marsden, C. Selby, D. Morrison, and W. Macnee Effect of N-acetyl cysteine on the concentrations of thiols in plasma, bronchoalveolar lavage fluid, and lung tissue. Thorax 49: Cotgreave, I. A., A. Eklund, K. Larsson, and P. W. Moldeus No penetration of orally administered N-acetylcysteine into bronchoalveolar lavage fluid. Eur. J. Respir. Dis. 70: Ashcroft, T., J. M. Simpson, and V. Timbrell Simple method of severity of pulmonary fibrosis on a numerical scale. J. Clin. Pathol. 41: Woessner, J. F The determination of hydroxyproline in tissue and protein samples containing small proportions of this imino acid. Arch. Biochem Biophys. 93: The BAL Cooperative Group Steering Committee Bronchoalveolar lavage constituents in healthy individuals, idiopathic pulmonary fibrosis, and selected comparison group. Am. Rev. Respir. Dis. 141: Waters, L. C., M. I. Schwartz, R. M. Cherniak, J. A. Waldron, T. L. Dunn, R. E. Stanford, and T. E. King Idiopathic pulmonary fibrosis: pretreatment bronchoalveolar lavage cellular constituents and their relationships and lung histopathology and clinical response to therapy. Am. Rev. Respir. Dis. 135: Cailes, J. B., C. O Connor, P. Pantelidis, A. M. Southcott, M. X. Fitzgerald, C. M. Black, and R. M. du Bois Neutrophil activation in fibrosing alveolitis: a comparison of lone cryptogenic fibrosing alveolitis and systemic sclerosis. Eur. Respir. J. 9: Obayashi, Y., I. Yamadori, J. Fujita, T. Yoshinouchi, N. Ueda, and J. Takahara The role of neutrophils in the pathogenesis of idiopathic pulmonary fibrosis. Chest 112: Davis, G. S., A. R. Brody, and J. E. Craighead Analysis of airspace and interstitial lung disease. Am. Rev. Respir. Dis. 118: Kumar, R. K Quantitative immunohistologic assessment of lymphocyte populations in the pulmonary inflammatory response to intratracheal silica. Am. J. Pathol. 135: Antoniades, H. N., J. Neville-Golden, T. Galanopoulos, R. L. Kradin, A. J. Valente, and D. T. Graves Expression of monocyte chemoattractant protein-1 mrna in human idiopathic pulmonary fibrosis. Proc. Natl. Acad. Sci. U.S.A. 89: Standiford, T. J., M. W. Rolfe, S. L. Kunkel, J. P. Lynch, III, M. D. Burdick, A. R. Gilbert, M. B. Orringer, R. I. Whyte, and R. M. Strieter Macrophage inflammatory protein-1a expression in interstitial lung disease. J. Immunol. 151: Standiford, T. J., M. R. Rolfe, S. L. Kunkel, J. P. Lynch, III, F. S. Becker, M. B. Orringer, S. Phan, and R. M. Strieter Altered production and regulation of monocyte chemoattractant protein-1 from pulmonary fibroblasts isolated from patients with idiopathic pulmonary fibrosis. Chest 103:121S. 29. Strieter, R. M., and S. L. Kunkel Chemokines. In R. G. Crystal and J. B. West, editors. The Lung, 2nd ed. Lippincott-Raven Publishers, Philadelphia Walzt, A., S. L. Kunkel, and R. M. Strieter C-X-C chemokines: an overview. In A. E. Koch and R. M. Strieter, editors. Chemokines in Disease. R. G. Landes Co., Austin, TX Taub, D. D C-C chemokines: an overview. In A. E. Koch and R. M. Strieter, editors. Chemokines in Disease. R. G. Landes Co., Austin, TX Keane, M. P., T. J. Standiford, and R. M. Strieter Chemokines are important cytokines in the pathogenesis of interstitial lung disease. Eur. Respir. J. 10: Blackwell, T. S., and J. W. Christman The role of nuclear- B in cytokine gene regulation. Am. J. Respir. Cell Mol. Biol. 17: Schreck, R., P. Rieber, and P. A. Baeuerle Reactive oxygen intermediates as apparently widely used messengers in the activation of the NF- B transcription factor and HIV-1. EMBO J. 10: Tozawa, K., S. Sakurada, K. Kohri, and T. Okamoto Effects of anti-nuclear factor B reagents in blocking adhesion of human cancer cells to vascular endothelial cells. Cancer Res. 55: Shen, A. S., C. Haslett, D. C. Feldsin, P. M. Henson, and R. M. Cherniack The intensity of chronic lung inflammation and fibrosis after bleomycin is directly related to the severity of acute injury. Am. Rev. Respir. Dis. 137: Thrush, M. A., E. G. Mimnaugh, E. Ginsburgh, and T. E. Gram Studies on the interaction of bleomycin A with rat lung microsome II: involvement of adventitious iron and reactive oxygen in bleomycinmediated DNA chain breakage. J. Pharmacol. Exp. Ther. 221: Passero, M. A., J. K. Held, and P. N. Shearer Effects of bleomycin, O 2 concentration, and H 2 O 2 on peroxidation of arachidonic acid. Am. Rev. Respir. Dis. 127(Suppl.):S Aruoma, O. I., B. Halliwell, B. M. Hoey, and J. Butler The antioxidant action of N-acetylcysteine: its reaction with hydrogen peroxide, hydroxyl radical, superoxide, and hypochlorous acid. Free Radic. Biol. Med. 6: Forman, H. J., and D. C. Skelton Protection of alveolar macrophages from hyperoxia by r-glutamyl transpeptidase. Am. J. Physiol. 259:L Cantin, A. M., P. Larivee, and R. O. Begin Extracellular glutathione suppress human lung fibroblast proliferation. Am. J. Respir. Cell Mol. Biol. 3: Gharaee-Kermani, M., E. M. Denholm, and S. H. Phan Costimulation of fibroblast collagen and transforming growth factor 1 gene expression by monocyte chemoattractant protein-1 via specific receptors. J. Biol. Chem. 271: Buhl, R., C. Vogelmeier, M. Critenden, R. C. Hubbard, R. F. Hoyt R, E. M. Wilson, A. M. Cantin, and R. G. Crystal Augmentation of glutathione in the fluid lining the epithelium of the lower respiratory tract by directly administering glutathione aerosol. Proc. Natl. Acad. Sci. U.S.A. 87: Webb, W. R Clinical evaluation of a new effective mycolytic agent (N-acetyl-L-cysteine). Am. Rev. Respir. Dis. 86: