Redox Imbalance and Pulmonary Function in Bleomycin-Induced Fibrosis in C57BL/6,DBA/2,and BALB/c Mice

Abstract

The development of bleomycin-induced pulmonary fibrosis (BLEO-PF) has been associated with differences in genetic background and oxidative stress status. The authors’ aim was to investigate the crosstalk between the redox profile, lung histology, and respiratory function in BLEO-PF in C57BL/6, DBA/2, and BALB/c mice. BLEO-PF was induced with a single intratracheal dose of bleomycin (0.1 U/mouse). Twenty-one days after bleomycin administration, the mortality rate was over 50% in C57BL/6 and 20% in DBA/2 mice, and BLEO-PF was not observed in BALB/c. There was an increase in lung static elastance (p < .001), viscoelastic/inhomogeneous pressure (p < .05), total pressure drop after flow interruption (p < .01), and ΔE (p < .05) in C57BL/6 mice. The septa volume increased in C57BL/6 (p < .05) and DBA/2 (p < .001). The levels of IFN-γ were reduced in C57BL/6 mice (p < .01). OH-proline levels were increased in C57BL/6 and DBA/2 mice (p < .05). SOD activity and expression were reduced in C57BL/6 and DBA/2 mice (p < .001 and p < .001, respectively), whereas catalase was reduced in all strains 21 days following bleomycin administration compared with the saline groups (C57BL/6: p < .05; DBA/2: p < .01; BALB/c: p < .01). GPx activity and GPx1/2 expression decreased in C57BL/6 (p < .001). The authors conclude that BLEO-PF resistance may also be related to the activity and expression of SOD in BALB/c mice.

Keywords

pulmonary fibrosis redox imbalance pulmonary function mouse strains.

Introduction

Pulmonary fibrosis (PF) is a severe, chronic, and restrictive interstitial disease that results from inflammation-induced parenchymal injury followed by excessive extracellular matrix deposition (Cottin and Cordier 2008; Odajima et al. 2010). PF can be caused by different types of stimuli, such as persistent acute respiratory distress syndrome (Lamy et al. 1975), long-term exposure to silica (Driscoll et al. 1990), or cytotoxic treatment with bleomycin (Jones 1978). However, not all individuals exposed to the known stimuli indeed manifest PF (Fernandez Perez et al. 2010; Raghu et al. 2006).

Susceptibility to PF appears to be related to the interaction between multiple factors, such as genetic predisposition (Garcia 2011), oxidative stress (Homer et al. 2011; Kliment and Oury 2010), and type of inflammatory response (Homer et al. 2011). Several animal models have been widely used to investigate these interactions in the pathogenesis of PF, and many studies have used the bleomycin-induced PF (BLEO-PF) mouse model (Chung et al. 2003; Garcia-Prieto et al. 2010; Harrison and Lazo 1987; Lemay and Haston 2005; Martins et al. 2009; Moeller et al. 2008).

Bleomycin, which is widely used as an adjuvant in chemotherapy treatment, not only has cytotoxic anticancer properties but also presents oxidative properties mediated by copper, iron, and reactive oxygen species (Fyfe and McKay 2010; Muller and Zahn 1977). Studies have shown that patients undergoing bleomycin treatment who develop lung fibrosis present an imbalance between tissue inhibitors of metalloproteinase (TIMPs) and metalloproteinase (MMPs) activities (Manoury, Caulet-Maugendre, et al. 2006; Manoury, Nenan, et al. 2006) and upregulation of both Wnt (Konigshoff et al. 2009) and TGF-b-induced Notch signaling, which leads to the excessive deposition of collagen in the lung parenchyma (Aoyagi-Ikeda et al. 2010). However, little is known regarding the differences in bleomycin-related susceptibility in humans.

We have demonstrated that the differences in the effects of cigarette smoke depend on the mouse strain studied, suggesting that genetic background influences susceptibility to the effects of cigarette smoke on wound healing (Cardoso et al. 2009, 2007). Mouse strains and individual organs display a range of susceptibilities to cigarette smoke–induced oxidative stress. The BALB/c and C57BL/6 strains appear to be the best experimental models for studying the effects of cigarette smoke on the liver and lung, and the C3H and C57BL/6 strains are the best models for the effects of cigarette smoke on the brain (Rueff-Barroso et al. 2010). Studies have shown that C57BL/6 mice are more susceptible to PF and BALB/c mice are resistant, and these responses are mainly due to Th-mediated IFN-γ/IL-12 and IL-4/IL-5/IL13 cytokine release, respectively (Chung et al. 2003; Kikuchi et al. 2010).

Th1 cytokines are associated with the increased oxidant production that induces damage of key molecules, such as lipids, proteins, and DNA. The balance between oxidants and antioxidants can be maintained by the activity of antioxidant enzymes, such as catalase (CAT), superoxide dismutase (SOD), and glutathione peroxidase (GPx) (Russo et al. 2010). Oxidative stress occurs when a redox imbalance, defined by alterations in antioxidant enzyme activities, is followed by oxidative damage (Rahman and Kilty 2006). In this case, antioxidant treatments have been shown to decrease bleomycin-induced oxidant production (Chen and Stubbe 2005) and inflammation (Day 2008; Kikuchi et al. 2011; Russo et al. 2011; Tanaka et al. 2010).

Therefore, understanding the crosstalk between the redox profile, lung histology, and respiratory function in BLEO-PF in C57BL/6, DBA/2, and BALB/c mice may help to identify the missing links in the pathogenesis of PF.

Materials and Methods

Animals

Male 8-wk-old C57BL/6, DBA/2, and BALB/c mice (weight: 20–24 g) were purchased from the Division of Clinical Research, National Cancer Institute (Rio de Janeiro, Brazil). Mice were housed (10 per cage) in a controlled environment with a 12-h light/12-h dark cycle (lights on at 6 p.m.) and an ambient temperature of 25 ± 2ºC (humidity 40–60%). The animals had free access to water and food. The animals were acclimatized during the 2 weeks before the experimental procedures. All animal care and experimental protocols were conducted following the guidelines of the institutional care and use committee (Committee for Evaluation of Animal Use for Research from the Federal University of Rio de Janeiro, CAUAP-UFRJ) and the NIH Guide for the Care and Use of Laboratory Animals (ISBN 0-309-05377-3). The protocols were approved by CAUAP-UFRJ (under registry CEUA-IBCCF108).

Experimental Design

Intratracheal administration (i.t.) of bleomycin (Sigma-Aldrich, St. Louis, MO, USA) was performed on mice under intraperitoneal (i.p.) ketamine anesthesia as previously described (Manoury, Nenan, et al. 2006). After tracheostomy, bleomycin was instilled at a dose of 0.1 U in a final volume of 50 µL. The saline groups received saline by i.t. administration. This protocol was performed 3 times, and the number of animals and samples used in each experiment or assay is indicated in the Results section. Survival rates were analyzed throughout 21 days after bleomycin instillation. The remaining mice were subjected to a mechanical ventilation protocol and were sacrificed by exsanguination.

Pulmonary Mechanics

The animals were sedated with diazepam (1 mg i.p.), anesthetized with sodium pentobarbital (20 mg/kg body wt i.p.), paralyzed with pancuronium bromide (5 mg/kg body wt) intravenous (i.v), and mechanically ventilated (Samay VR15, Universidad de la Republica, Montevideo, Uruguay) with a frequency of 100 breaths/min, tidal volume of 0.2 mL, flow of 1 mL/s, and positive end-expiratory pressure of 2.0 cm H₂O. The anterior chest wall was surgically removed. A pneumotachograph (1.5-mm ID; length = 4.2 cm, distance between side ports = 2.1 cm) was connected to the tracheal cannula for the measurement of airflow (V=), and changes in lung volume were obtained by digital integration of the flow signal. The pressure gradient across the pneumotachograph was determined by means of a Validyne MP-45-2 differential pressure transducer (Validyne Engineering, Northridge, CA, USA). The equipment resistive pressure (= R_eq·V’) was subtracted from pulmonary resistive pressure to obtain the intrinsic values. The transpulmonary pressure was measured with the Validyne MP-45 differential pressure transducer. Lung resistive (ΔP1) and viscoelastic/inhomogeneous (ΔP2) pressures, total pressure drop after flow interruption (ΔPtot = ΔP1 + ΔP2), lung static (Est) and dynamic (E_dyn) elastance, and ΔE (E_dyn− Est) were computed by the end-inflation occlusion method. ΔP1 selectively reflects the airway resistance, and ΔP2 reflects stress relaxation or the viscoelastic properties of the lung (Faffe and Zin 2009). Lung mechanics were measured 10–15 times in each animal.

Bronchoalveolar Lavage Fluid (BALF)

The right ventricle was perfused with saline to remove blood, and the BALF was obtained from all mice by injecting phosphate buffered saline (PBS) 3 consecutive times to a final volume of 1.5 mL in the right lung. The fluid was withdrawn and stored on ice. The total protein content in the samples was determined by the method of Bradford (1976).

BALF Analysis for IFN-γ and IL-6

IFN-γ and IL-6 levels were determined using a standardized ELISA technique—enzyme-linked immunosorbent assay kits (Quantikine, R&D Systems, Inc., Minneapolis, MN, USA) following the instructions of the manufacturer. All samples were measured in duplicate.

BALF Determination of Total Collagen

The presence of hydroxyproline (OH-proline) was determined using a colorimetric method based on the reaction of chloramine-T, perchloric acid, and p-dimethylbenzaldehyde. The color that developed was measured at 557 nm (Woessner 1961).

Tissue Collection, Histology, and Stereology

After BAL, the right lung was ligated, and the left lung in all mice was inflated by instilling 4% formalin buffer at 25 cm H₂O pressure for 2 min, ligated, removed, and weighed. The inflated lungs were fixed for 48 h before embedding in paraffin. Serial sagittal sections were obtained for histological and morphometric analyses. To obtain uniform and proportionate lung samples, 18 fields (6 nonoverlapping fields in 3 different sections) were randomly analyzed using a video microscope (Zeiss-Axioplan—20× objective lens and JVC color video camera linked to a Sony Trinitron color video monitor; Carl Zeiss, Oberkochen, Germany) and a cycloid test-system superimposed on the monitor screen. The reference volume was estimated by point counting using the test points system (PT). The points that contacted the air spaces (PP) were counted to estimate the volume densities (Vv) of the structures (V = PP/PT) (Weibel et al. 2007). A total area of 1.94 mm² was analyzed to determine the volume density of air spaces (Vv alveoli and septum) in sections stained with hematoxylin and eosin (H&E). Two investigators that performed the measurement counted non-identified sections. The morphometric method was adapted from Vlahovic et al. (1999). Collagen organization was evaluated in tissue sections stained with Sirius red observed under polarized light where thick collagen fibers are strongly birefringent and yellow to red, whereas thin collagen fibers are weakly birefringent and greenish.

Oxidative Stress Markers—Antioxidant Enzymes Assay

The left lungs were immediately removed and homogenized on ice with 10% (w/v) PBS (pH 7.3) using an Ultra-Turrax® T8 homogenizer (Toronto, Canada) and centrifuged at 600 g for 5 min. The supernatants were kept at –20ºC until analysis of the antioxidant enzyme activities. The total protein content in the lung homogenate samples was determined by the method of Bradford (1976).

The CAT, SOD, and GPx activities were determined in the lung homogenates. CAT activity was measured by the rate of decrease in hydrogen peroxide concentration at 240 nm (Aebi 1984). SOD activity was assayed by measuring the inhibition of adrenaline auto-oxidation by absorbance at 480 nm (Bannister and Calabrese 1987). GPx activity was measured by monitoring the oxidation of NADPH at 340 nm in the presence of H₂O₂ (Flohe and Gunzler 1984).

Oxidative Damage Assay—Malondialdehyde Assay

As an index of lipid peroxidation, we used the thiobarbituric acid-reactive substances (TBARS) method for analyzing malondialdehyde products during an acid-heating reaction, as previously described by Draper and coworkers (1993). Briefly, samples from lung homogenates were mixed with 1 mL of 10% trichloroacetic acid and 1 mL of 0.67% thiobarbituric acid; the samples were then heated in a boiling water bath for 30 min. TBARS levels were determined by absorbance at 532 nm and are expressed as malondialdehyde equivalents (nm/mg protein).

Nitrite Assay

Nitrite, a subproduct of NO metabolism, was measured using the Griess reaction (Valenca et al. 2009). BAL samples were reacted with 50 µL of 1% sulfanilamide solution for 10 min and mixed with 50 µL of 0.1% naphthyl ethylenediamine solution. Formation of the stable azo compound with a purple color was measured spectrophotometrically by absorbance at 540 nm. The method was standardized with known concentrations of nitrite. Nitrite concentrations are expressed in µMol/mg protein.

Western Blotting

Twenty micrograms of lung homogenate protein were subjected to electrophoresis on 8% PAGE gels and transferred onto nitrocellulose membranes (Amersham, Arlington Heights, IL, USA). The nitrocellulose membrane was blocked with 5% BSA and subsequently incubated overnight with rabbit polyclonal IgG SOD-3, CAT, and GPx 1/2 antibodies (1:1,000 dilution; Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA). After 3 washing steps (10 min each), the levels of protein were detected using a goat anti-rabbit antibody (1:10,000 dilution) conjugated to horseradish peroxidase (Dako Cytomation, Santa Barbara, CA, USA), and bound complexes were detected using the enhanced chemiluminescence method (ECL; Amersham). Molecular weight standards were used to identify the target band (Precision Plus Protein All Blue Standards, Bio-Rad Laboratories, Hercules, CA, USA).

Statistical Analysis

Survival rates were expressed as percentages, and a Mantel-Cox log-rank test (χ² test) was used to detect differences in mice survival. Data are shown as the mean ± SEM. Following a normality test, parametric data were tested using one-way ANOVA and Tukey’s post-hoc test. Nonparametric data were tested using Kruskal-Wallis and Dunn’s post-hoc test. T-test was used for densitometric analysis of Western blotting bands. Statistical significance was defined as p < .05.

Results

Survival Rate

The survival curves of the groups are shown in Figure 1. There was no effect of bleomycin administration on the survival rate of BALB/c mice. Compared with BALB/c mice, higher mortality rates were found in both C57BL/6 and DBA/2 mice, which was represented by a gradual decline in the survival rates of the mice starting at 24 hours; there was a mice mortality rate of 50% in C57BL/6 and 20% in DBA/2 mice at the end of the experiment (n = 20 for each group).

Figure 1.

Survival curve of mice 21 days after intratracheal administration of bleomycin. The administration of bleomycin increased mortality in both C57BL/6 and DBA/2 mice. Survival rates are expressed as percentages, and the Mantel-Cox log-rank test (χ² test) was used to detect differences in survival rates among mouse strains (p < .05). Dark squares: BALB/c strain; dark triangles: DBA/2 strain; dark circles: BALB/c.

Pulmonary Mechanics

The values for lung mechanics during spontaneous breathing for each group are shown in Table 1. C57BL/6 mice exhibited increases of 53% in Est (p < .001), 35% in ΔP2 (p < .05), 26% in ΔPtot (p < .01), and 35% in ΔE (p < .05) 21 days following bleomycin administration. No statistical differences in the respiratory mechanics were observed in bleomycin-instilled BALB/c and DBA/2 mice compared with the paired saline controls (n = 10 for each group).

Table 1.

Mechanical ventilation of mice 21 days following bleomycin administration (0.1 U/mouse).

Groups		Elastance	▵P1	▵P2	▵Ptot	▵E
C57BL/6	Saline	33.8 ± 1.47	0.62 ± 0.03	0.95 ± 0.07	1.57 ± 0.06	4.55 ± 0.19
C57BL/6	Bleomycin	51.8 ± 4.89***	0.69 ± 0.03	1.28 ± 0.09*	1.98 ± 0.11**	6.16 ± 0.45*
DBA/2	Saline	21.6 ± 0.79	0.48 ± 0.01	0.68 ± 0.02	1.16 ± 0.02	3.47 ± 0.15
DBA/2	Bleomycin	20.6 ± 1.09	0.48 ± 0.01	0.73 ± 0.04	1.21 ± 0.04	3.64 ± 0.20
BALB/c	Saline	44.3 ± 2.34	0.72 ± 0.04	1.25 ± 0.06	1.98 ± 0.07	6.50 ± 0.38
BALB/c	Bleomycin	43.6 ± 3.14	0.67 ± 0.05	1.30 ± 0.07	1.96 ± 0.11	6.40 ± 0.34

Elastance—lung static elastance; ▵P1—this parameter reflects airway resistance; ▵P2—this parameter reflects stress relaxation (or viscoelastic properties); ▵Ptot—▵P1 + ▵P2; ▵E—viscoelastic component of elastance; data are expressed as the means ± SEM and statistically analyzed using ANOVA with Tukey post-test considering a significance level of 5%. All groups were compared with the saline control.

*p < .05. **p < .01. ***p < .001.

Inflammatory Response

The inflammatory cells and levels of IFN-γ, IL-6, and OH-proline are presented in Table 2. The number of inflammatory cells increased in C57BL/6, DBA/2, and BALB/c mice 1 day following bleomycin administration (p < .001). The levels of IFN-γ were reduced by 43% in C57BL/6 mice (p < .01). There was a 76% increase in the IL-6 levels in C57BL/6 mice (p < .01), whereas the IL-6 levels were decreased by 40% in DBA/2 mice (p < .05). The OH-proline levels were increased in C57BL/6 and DBA/2 mice compared with the respective saline groups (112%, p < .05 and 104%, p < .05, respectively) (n = 10 each group).

Table 2.

Inflammatory cells, pro-inflammatory cytokines, and OH-proline from bronchoalveolar lavage of mice treated with the bleomycin 0.1 U/mice.

Groups		Total cells	IL-6	INF-γ	OH-proline
C57BL/6	Saline	234.5 ± 29.0	0.55 ± 0.03	0.14 ± 0.01	3.166 ± 0.53
C57BL/6	Bleomycin	503.8 ± 50.7***	0.98 ± 0.08**	0.08 ± 0.01**	6.713 ± 0.23*
DBA/2	Saline	291.0 ± 20.4	1.04 ± 0.12	0.18 ± 0.01	4.257 ± 0.79
DBA/2	Bleomycin	515.6 ± 34.5***	0.62 ± 0.02*	0.15 ± 0.01	8.677 ± 0.33*
BALB/c	Saline	94.70 ± 6.22	0.99 ± 0.10	0.18 ± 0.01	4.240 ± 1.69
BALB/c	Bleomycin	251.2 ± 18.3***	0.80 ± 0.04	0.15 ± 0.01	7.559 ± 0.26

Total cells (cells/mL BALF). IL-6—interleukin 6 (ng/mL). INF-γ—interferon gamma (ng/mL). OH-proline—hydroxyproline (ng/mg protein). Data were expressed as mean ± SEM and statistically analyzed by ANOVA with Tukey post-test considering the significance level at 5%. All groups were compared with the control group.

*p < .05. **p < .01. ***p < .001.

There was an increase in the extent of fibrosis in lung histological samples from both C57BL/6 and DBA/2 mice 21 days following bleomycin administration (Figure 2). The volume density of septa was increased by 79% (p < .05) and 126% (p < .001) (Figure 3A), and the volume density of alveoli was decreased by 59% (p < .05) and 58% (p < .001) (Figure 3B) in C57BL/6 and DBA/2 mice, respectively, compared with the saline groups (n = 10 for each group).

Figure 2.

Photomicrographs of lung sections stained with hematoxylin and eosin from mice 21 days after intratracheal administration of 0.1 U of bleomycin in different strains (right column). Mice receiving saline were considered as controls (left column). A and B: C57BL/6; C and D: DBA/2; E and F: BALB/c. Extensive areas of fibrosis of the alveolar septa in both C57BL/6 and DBA/2 (B and D). There were no histological alterations in BALB/c following bleomycin administration. Bar 50 µm.

Figure 3.

Stereological analysis of lung parenchyma. Histological changes were demonstrated by H&E staining (original magnification, X20). (A) Vv[septum]—volume density of septa; (B) Vv[alveoli]—volume density of alveoli, alveolar sacs and ducts. Data are presented as the means ± SEM and were analyzed by Kruskal-Wallis followed by Dunns post-hoc tests (p < .05). The asterisks are representative of p < .05 (*) and p < .001 (***) (n = 10/strain). Bar 50 µm.

The hallmark characteristic of fibrosis is the excessive deposition of an extracellular matrix, such as collagen. In our experimental condition, bleomycin-induced excessive deposition of extracellular matrix proteins that were revealed by Red Picro Sirius in C57BL/6 (Figure 4B and 4D) mice and DBA/2 (Figure 4F and 4H). However, BALB/c mice showed no histological changes after administration of bleomycin (Figure 4J and 4M) with delicate collagens fibers in alveolar septa and in the wall of bronchiole.

Figure 4.

Lung photomicrographs of Sirius red stained sections (original magnification, ×20). Saline groups (a—C57BL/6; c—DBA/2, and e—BALB/c) with delicate collagens fibers in alveolar septa and in the wall of bronchiole and half view with polarized light. Lung sections 21 days after intratracheal administration of 0.1 U of bleomycin in different strains (b—C57BL/6 and d—DBA/2) with irregular collagen fibers in thickened septa, mainly in C57BL/6; half view with polarized light. There were no histological alterations in BALB/c following bleomycin administration (f). Bar 50 µm.

Activity of Oxidative Stress Markers

The activities of oxidative stress markers are shown in Table 3 (n = 10 for each group). A marked redox imbalance was observed in all strains studied. SOD activity was reduced in C57BL/6 and DBA/2 mice (76%, p < .001 and 24%, p < .001, respectively), whereas CAT activity was reduced in all strains 21 days following bleomycin administration compared with the saline groups (C57BL/6: 53%, p < .05; DBA/2: 33%, p < .01; BALB/c: 45%, p < .01). Although there was a reduction in GPx activity in C57BL/6 (31%, p < .05) and in BALB/c mice (26%, p < .01), there was an increase of 56% on the GPx activity in DBA/2 mice (p < .01) compared with the respective saline group. Nitrite levels were increased in C57BL/6 and BALB/c mice (129%, p < .001 and 150%, p < .001, respectively). Malondialdehyde (MDA) levels were increased in C57BL/6 (163%, p < .001) and were reduced in BALB/c mice (44%, p < .001) 21 days following bleomycin administration compared with the respective saline groups.

Table 3.

Biochemical analyses of oxidative stress and oxidative damage markers in lung tissue homogenates and bronchoalveolar lavage samples from mice 21 days following bleomycin administration (0.1 U/mouse).

Groups		SOD	CAT	GPx	Nitrite	MDA
C57BL/6	Saline	0.138 ± 0.014	7.295 ± 0.893	0.026 ± 0.002	3.900 ± 0.152	0.351 ± 0.037
C57BL/6	Bleomycin	0.033 ± 0.002***	3.420 ± 0.574*	0.018 ± 0.001*	8.933 ± 0.202***	0.924 ± 0.162***
DBA/2	Saline	0.193 ± 0.014	7.157 ± 0.487	0.016 ± 0.001	5.725 ± 1.647	0.377 ± 0.073
DBA/2	Bleomycin	0.146 ± 0.016***	4.824 ± 0.467**	0.025 ± 0.001**	6.817 ± 1.258	0.342 ± 0.031
BALB/c	Saline	0.095 ± 0.017	7.237 ± 0.652	0.019 ± 0.001	5.167 ± 0.260	1.399 ± 0.066
BALB/c	Bleomycin	0.085 ± 0.005	3.948 ± 0.324**	0.014 ± 0.001**	12.90 ± 0.447***	0.788 ± 0.022***

SOD—superoxide dismutase activity (U/mg protein); CAT—catalase activity (U/mg protein); GPx—glutathione peroxidase activity (mM/min/mg protein x 10^-3); Nitrite (mM/mg protein—bronchoalveolar lavage); MDA—malondialdehyde levels (nM/mg protein). Data are expressed as the means ± SEM and statistically analyzed by ANOVA with Tukey post-test considering a significance level of 5%. All groups were compared with the saline control.

*p < .05. **p < .01. ***p < .001.

Expression of Oxidative Stress Markers

SOD-3 expression was decreased in both the C57BL/6 and DAB/2 groups, but SOD-3 expression was increased in the BALB/c group (Figure 5A). CAT expression was increased in the C57BL/6 group but was deceased in the DBA/2 and BALB/c groups (Figure 5B). GPx expression was decreased in both the C57BL/6 and BALB/c groups 21 days following bleomycin administration (Figure 5C).

Figure 5.

Expression levels of antioxidant enzymes in protein extracts from lung homogenates (Western blot analysis). A: SOD-3—superoxide dismutase-3; B: CAT—Catalase; C: GPx 1/2—Glutathione peroxidase 1/2. Administration of bleomycin does not alter the expression of SOD-3 in lung homogenate samples from BALB/c. The Western blots were analyzed using optical densitometry (AU-arbitrary units) (n = 3/group). Data are presented as the means ± SEM and were analyzed by t-test (p < .05). The asterisks are representative of p < .05 (*) in comparison with the respective control group.

Discussion

Pulmonary fibrosis occurs following abnormal lung repair that results in pulmonary dysfunction (Pardo and Selman 2002). The mechanisms involved in the progression of pulmonary fibrosis include fibroblast activation and chronic inflammation (Meneghin and Hogaboam 2007). Oxidative stress has been associated with a number of diseases, including pulmonary fibrosis (Day 2008), as bleomycin, along with Fe2+, acts as is an exogenous source of oxidants due to Fenton’s reaction (Chen and Stubbe 2005). Studies have suggested that genetic background may play an important role in human PF susceptibility because only 10% of the patients undergoing chemotherapeutic regimens develop PF (Abid et al. 2001; du Bois 2002). Therefore, investigating the crosstalk between the redox profile, lung histology, and respiratory function in BLEO-PF in C57BL/6, DBA/2, and BALB/c mice may help to identify the missing links in the pathogenesis of PF.

Although C57BL/6 and BALB/c mice have been shown to be prone to oxidative damage in the lungs following long-term exposure to cigarette smoke (Rueff-Barroso et al. 2010), the oxidative parameters following bleomycin administration observed in the present study showed that BALB/c mice were resistant to oxidative damage, in contrast to the C57BL/6 and DBA/2 strains.

We showed that both SOD activity and expression are associated with the incidence of bleomycin-induced PF in mice. In the antioxidant enzyme system, SOD converts O²⁻ into H²O², whereas CAT and GPx convert H²O² into water. Along with this system, the GSH/GSSG pair is thought to be the major redox buffer in the cell, as it controls not only the cellular level of H²O² but also the levels of other organic hydroperoxides (Keller and Menzel 1989; Meister and Anderson 1983). Due to an impairment in the activity of antioxidant enzymes, increased oxidant production leads to lipid peroxidation, protein carbonylation, and DNA damage (Rahman and MacNee 1996). As lipid peroxidation products may result in dehydration and cell death, they have been considered to be important factors for the development of PF. Our results showed that MDA levels, which are representative of lipid peroxidation, were increased only in C57BL/6 mice. Increased lipid peroxidation has been described to be a direct oxidative product of hydroxyl radical attack in BLEO-PF (Sato et al. 2008). As previously described (Iraz et al. 2006; Kalayarasan et al. 2008), our results also show that the increased lipid peroxidation found in C57BL/6 mice was associated with reductions in the SOD, CAT, and GPx activities.

Three gene isoforms for SOD have been described in mammals: cytosolic SOD—SOD-1, mitochondrial SOD—SOD-2, and extracellular SOD—SOD-3 (Zelko et al. 2002). As SOD-3-deficient mice showed increased bleomycin-induced inflammatory responses in the lungs, extracellular matrix-located SOD seems to play an important role in protection against PF (Fattman et al. 2003). In the present study, C57BL/6 mice showed not only decreased SOD activity but also decreased SOD-3 expression in the lung. The decreased SOD activity observed in C57BL/6 mice 21 days after bleomycin administration could have resulted from enzyme inactivation by oxidants (De Raeve et al. 1997). As nitrite levels were found to be increased in C57BL/6 mice, O²⁻ might have first been converted into peroxonitrite (ONOO⁻), as the rate constant is over 3.5 times faster than the dismutation of O²⁻ by SOD (Dweik 2005). In addition, our data suggest that the increase in both SOD expression and activity might be associated with the resistance of BALB/c mice to PF.

Unlike SOD, CAT activities decreased in all groups following bleomycin administration; however, CAT expression was increased in C57BL/6 mice. The present data suggest that reduced SOD activity resulted in lower levels of hydrogen peroxide, which thus directly affected CAT activity, even though the expression of this enzyme was increased in C57BL/6 mice. In addition, reduced GPx activity impairs membrane phospholipid protection, which also results in lipid peroxidation, as observed in C57BL/6 mice.

Acute lung injury and normal repair of the alveolar-capillary membrane result in rapid restoration of tissue integrity and function following a variety of insults. Although acute inflammation and normal repair represent a complex interplay between humoral, cellular, and extracellular matrix networks, this process usually occurs in a sequential, yet overlapping, manner (Strieter 2008). The presence of myofibroblasts in patients with pulmonary fibrosis is documented in lung tissues and also in animal models of the disease (Phan 2002). It is established that fibroblasts transdifferentiate to myofibroblasts that express elevated levels of α-smooth muscle actin and, consequently, display a markedly enhanced ability to secrete extracellular matrix proteins (Phan 2002). Among several mediators, TGF-β is the major cytokine associated with pulmonary fibrosis; it is involved in the transition of fibroblasts into myofibroblasts (Strieter 2008), the synthesis of matrix proteins, and collagen degradation inhibition (Higashiyama et al. 2007; Moore and Hogaboam 2008). Studies have shown that a decrease in IFN-γ is associated with increased fibroblast activation and, therefore, increased extracellular matrix deposition, which leads to increased airway resistance in patients with idiopathic pulmonary fibrosis (Segel et al. 2003; Ziesche et al. 1999). In the present study, PF in C57BL/6 mice was associated with reduced levels of IFN-γ and increased levels of OH-proline. Moreover, C57BL/6 mice showed impaired pulmonary function and, for the most part, increased elastance and resistance, probably due to increased extracellular matrix deposition in mouse lungs (Faffe and Zin 2009).

In the present study, bleomycin administration induced PF and increased OH-proline in DBA/2 mice. Although the SOD and CAT activities were found to be reduced in these mice, no lipid peroxidation was observed, probably due to the elevation in GPx activity in these mice. In addition, DBA/2 mice showed no alterations in mechanical ventilation parameters, even though histological quantification showed clear indications of PF. Previous data have shown that a single intratracheal instillation of bleomycin resulted in reversible fibrosis in DBA/2 mice, whereas multiple doses resulted in persistent and prolonged fibrosis (Chung et al. 2003). The present data suggest that DBA/2 mice may be at an early stage of PF resolution or that the elastic components of the lung parenchyma have been preserved, as described by Manoury and coworkers, as pulmonary mechanics were unaltered 21 days following bleomycin administration (Manoury, Nenan, et al. 2006).

The present results also show that BALB/c mice, conversely, did not develop PF or show lipid peroxidation and showed no alterations in both IL-6 and IFN-γ levels and pulmonary mechanics parameters. Although the levels of TGF-β were not measured in this study, the lack of signs of fibrosis in BALB/c mice may be related to the maintenance of IFN-γ in these mice, as the production of IFN-y by Natural Killer T cells (NKT) and the subsequent production of TGF-β are pivotal for the development of BLEO-PF in mice (Kim et al. 2005).

BALB/c mice have been shown to present a significantly reduced number of myofibroblasts, which results in an inefficient production of collagen following wound repair processes (Cardoso et al. 2009). This may, in part, explain the observation that even following bleomycin administration, BALB/c mice did not develop PF.

As SOD acts as the first line of antioxidant defense in cells (Kaarteenaho-Wiik and Kinnula 2004), both CAT and GPx activities were probably reduced in BALB/c mice due to the absence of substrates (mainly H2O2). In addition, as the irreversible differentiation of myofibroblasts from fibroblasts is mediated by both ROS and TGF-β production through NADPH oxidase activation (Amara et al. 2010; Hecker et al. 2009), we suggest that this mechanism was active in BALB/c mice.

In addition, SOD has been described as a key enzyme in the process of BLEO-PF, as either intratracheal or inhaled administration of a hybrid phosphatidylcholine-SOD (PC-SOD: lecithinized superoxide dismutase) reduced collagen deposition in mouse lungs following bleomycin administration (Tanaka et al. 2010). Therefore, SOD seems to play a pivotal role in the development of BLEO-PF, as indicated by the maintenance of both SOD activity and expression in BALB/c mice.

The present results showed that MDA levels were reduced in BALB/c mice following bleomycin administration. As ROS, mainly H²O², have been implicated in increasing cell death in the lungs of mice with BLEO-PF (Kasper and Barth 2009; Tanaka et al. 2010), the present data seem to indicate that the low susceptibility of BALB/c mice to oxidative damage protected pulmonary parenchyma.

Taken together, the present results suggest that PF development is associated with both SOD activity and SOD-3 expression in C57BL/6, DBA/2, and BALB/c mice following bleomycin administration.

Footnotes

This work was supported by grants from FAPERJ and CNPq. A.C.N. and M.L. received bursaries from CAPES. M.A.S.S. received a bursary from CNPq. S.S.V. was supported by the Visiting Professor Program of UERJ.

Abbreviations

References

Abid

S. H.

Malhotra

Perry

M. C.

(2001). Radiation-induced and chemotherapy-induced pulmonary injury. Curr Opin Oncol 13, 242–8.

Aebi

(1984). Catalase in vitro. Methods Enzymol 105, 121–6.

Amara

Goven

Prost

Muloway

Crestani

Boczkowski

(2010). NOX4/NADPH oxidase expression is increased in pulmonary fibroblasts from patients with idiopathic pulmonary fibrosis and mediates TGFbeta1-induced fibroblast differentiation into myofibroblasts. Thorax 65, 733–8.

Aoyagi-Ikeda

Maeno

Matsui

Ueno

Hara

Aoki

Shimizu

Doi

Kawai-Kowase

Iso

Suga

Arai

Kurabayashi

. (2010). Notch induces myofibroblast differentiation of alveolar epithelial cells via TGF-ss/Smad3 pathway. Am J Respir Cell Mol Biol 45, 136–44.

Bannister

J. V.

Calabrese

(1987). Assays for superoxide dismutase. Methods Biochem Anal 32, 279–312.

Bradford

M. M.

(1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72, 248–54.

Cardoso

J. F.

Mendes

F. A.

Amadeu

T. P.

Romana-Souza

Valenca

S. S.

Porto

L. C.

Abreu

J. G.

Monte-Alto-Costa

(2009). Ccn2/Ctgf overexpression induced by cigarette smoke during cutaneous wound healing is strain dependent. Toxicol Pathol 37, 175–82.

Cardoso

J. F.

Souza

B. R.

Amadeu

T. P.

Valenca

S. S.

Porto

L. C.

Costa

A. M.

(2007). Effects of cigarette smoke in mice wound healing is strain dependent. Toxicol Pathol 35, 890–6.

Chen

Stubbe

(2005). Bleomycins: Towards better therapeutics. Nat Rev Cancer 5, 102–12.

10.

Chung

M. P.

Monick

M. M.

Hamzeh

N. Y.

Butler

N. S.

Powers

L. S.

Hunninghake

G. W.

(2003). Role of repeated lung injury and genetic background in bleomycin-induced fibrosis. Am J Respir Cell Mol Biol 29, 375–80.

11.

Cottin

Cordier

J. F.

(2008). Idiopathic pulmonary fibrosis. Presse Med 37, 1581–90.

12.

Day

B. J.

(2008). Antioxidants as potential therapeutics for lung fibrosis. Antioxid Redox Signal 10, 355–70.

13.

De Raeve

H. R.

Thunnissen

F. B.

Kaneko

F. T.

Guo

F. H.

Lewis

Kavuru

M. S.

Secic

Thomassen

M. J.

Erzurum

S. C.

(1997). Decreased Cu, Zn-SOD activity in asthmatic airway epithelium: Correction by inhaled corticosteroid in vivo. Am J Physiol 272, L148–54.

14.

Draper

H. H.

Squires

E. J.

Mahmoodi

Agarwal

Hadley

(1993). A comparative evaluation of thiobarbituric acid methods for the determination of malondialdehyde in biological materials. Free Radic Biol Med 15, 353–63.

15.

Driscoll

K. E.

Maurer

J. K.

Lindenschmidt

R. C.

Romberger

Rennard

S. I.

Crosby

(1990). Respiratory tract responses to dust: Relationships between dust burden, lung injury, alveolar macrophage fibronectin release, and the development of pulmonary fibrosis. Toxicol Appl Pharmacol 106, 88–101.

16.

du Bois

R. M

. (2002). The genetic predisposition to interstitial lung disease: Functional relevance. Chest 121, 14S–20S.

17.

Dweik

R. A.

(2005). Nitric oxide, hypoxia, and superoxide: the good, the bad, and the ugly! Thorax 60, 265–7.

18.

Faffe

D. S.

Zin

W. A.

(2009). Lung parenchymal mechanics in health and disease. Physiol Rev 89, 759–75.

19.

Fattman

C. L.

Chang

L. Y.

Termin

T. A.

Petersen

Enghild

J. J.

Oury

T. D.

(2003). Enhanced bleomycin-induced pulmonary damage in mice lacking extracellular superoxide dismutase. Free Radic Biol Med 35, 763–71.

20.

Fernandez Perez

E. R.

Daniels

C. E.

Schroeder

D. R.

St Sauver

Hartman

T. E.

Bartholmai

B. J.

E. S.

Ryu

J. H.

(2010). Incidence, prevalence, and clinical course of idiopathic pulmonary fibrosis: A population-based study. Chest 137, 129–37.

21.

Flohe

Gunzler

W. A.

(1984). Assays of glutathione peroxidase. Methods Enzymol 105, 114–21.

22.

Fyfe

A. J.

McKay

(2010). Toxicities associated with bleomycin. J R Coll Physicians Edinb 40, 213–5.

23.

Garcia

C. K.

(2011). Idiopathic pulmonary fibrosis: Update on genetic discoveries. Proc Am Thorac Soc 8, 158–62.

24.

Garcia-Prieto

Gonzalez-Lopez

Cabrera

Astudillo

Gutierrez-Fernandez

Fanjul-Fernandez

Batalla-Solis

Puente

X. S.

Fueyo

Lopez-Otin

Albaiceta

G. M.

(2010). Resistance to bleomycin-induced lung fibrosis in MMP-8 deficient mice is mediated by interleukin-10. PLoS One 5, e13242.

25.

Harrison

J. H.

Jr. Lazo

J. S.

(1987). High dose continuous infusion of bleomycin in mice: A new model for drug-induced pulmonary fibrosis. J Pharmacol Exp Ther 243, 1185–94.

26.

Hecker

Vittal

Jones

Jagirdar

Luckhardt

T. R.

Horowitz

J. C.

Pennathur

Martinez

F. J.

Thannickal

V. J.

(2009). NADPH oxidase-4 mediates myofibroblast activation and fibrogenic responses to lung injury. Nat Med 15, 1077–81.

27.

Higashiyama

Yoshimoto

Kaise

Matsubara

Fujiwara

Kikkawa

Asano

Kinoshita

(2007). Inhibition of activin receptor-like kinase 5 attenuates bleomycin-induced pulmonary fibrosis. Exp Mol Pathol 83, 39–46.

28.

Homer

R. J.

Elias

J. A.

Lee

C. G.

Herzog

(2011). Modern concepts on the role of inflammation in pulmonary fibrosis. Arch Pathol Lab Med 135, 780–8.

29.

Iraz

Erdogan

Kotuk

Yagmurca

Kilic

Ermis

Fadillioglu

Yildirim

(2006). Ginkgo biloba inhibits bleomycin-induced lung fibrosis in rats. Pharmacol Res 53, 310–6.

30.

Jones

A. W.

(1978). Bleomycin lung damage: The pathology and nature of the lesion. Br J Dis Chest 72, 321–6.

31.

Kaarteenaho-Wiik

Kinnula

V. L.

(2004). Distribution of antioxidant enzymes in developing human lung, respiratory distress syndrome, and bronchopulmonary dysplasia. J Histochem Cytochem 52, 1231–40.

32.

Kalayarasan

Sriram

Sudhandiran

(2008). Diallyl sulfide attenuates bleomycin-induced pulmonary fibrosis: Critical role of iNOS, NF-kappaB, TNF-alpha and IL-1beta. Life Sci 82, 1142–53.

33.

Kasper

Barth

(2009). Bleomycin and its role in inducing apoptosis and senescence in lung cells—modulating effects of caveolin-1. Curr Cancer Drug Targets 9, 341–53.

34.

Keller

D. A.

Menzel

D. B.

(1989). Effects of sulfite on glutathione S-sulfonate and the glutathione status of lung cells. Chem Biol Interact 70, 145–56.

35.

Kikuchi

Ishii

Morishima

Yageta

Haraguchi

Itoh

Yamamoto

Hizawa

(2010). Nrf2 protects against pulmonary fibrosis by regulating the lung oxidant level and Th1/Th2 balance. Respir Res 11, 31.

36.

Kikuchi

Ishii

Morishima

Yageta

Haraguchi

Yamadori

Masuko

Sakamoto

Yanagawa

Warabi

Ishii

Hizawa

. (2011). Aggravation of bleomycin-induced pulmonary inflammation and fibrosis in mice lacking peroxiredoxin I. Am J Respir Cell Mol Biol 45, 600–9.

37.

Kim

J. H.

Kim

H. Y.

Kim

Chung

J. H.

Park

W. S.

Chung

D. H.

(2005). Natural killer T (NKT) cells attenuate bleomycin-induced pulmonary fibrosis by producing interferon-gamma. Am J Pathol 167, 1231–41.

38.

Kliment

C. R.

Oury

T. D.

(2010). Oxidative stress, extracellular matrix targets, and idiopathic pulmonary fibrosis. Free Radic Biol Med 49, 707–17.

39.

Konigshoff

Kramer

Balsara

Wilhelm

Amarie

O. V.

Jahn

Rose

Fink

Seeger

Schaefer

Gunther

Eickelberg

(2009). WNT1-inducible signaling protein-1 mediates pulmonary fibrosis in mice and is upregulated in humans with idiopathic pulmonary fibrosis. J Clin Invest 119, 772–87.

40.

Lamy

M. L.

Fallat

R. J.

Koeniger

E. L.

Dietrich

H. P.

Kamm

Hill

J. D.

(1975). Pathophysiology of adult respiratory distress syndrome. Acta Anaesthesiol Belg 23 (Suppl.), 64–77.

41.

Lemay

A. M.

Haston

C. K.

(2005). Bleomycin-induced pulmonary fibrosis susceptibility genes in AcB/BcA recombinant congenic mice. Physiol Genomics 23, 54–61.

42.

Manoury

Caulet-Maugendre

Guenon

Lagente

Boichot

(2006). TIMP-1 is a key factor of fibrogenic response to bleomycin in mouse lung. Int J Immunopathol Pharmacol 19, 471–87.

43.

Manoury

Nenan

Guenon

Boichot

Planquois

J. M.

Bertrand

C. P.

Lagente

(2006). Macrophage metalloelastase (MMP-12) deficiency does not alter bleomycin-induced pulmonary fibrosis in mice. J Inflamm (Lond) 3, 2.

44.

Martins

Valenca

S. S.

Farias-Filho

F. A.

Molinaro

Simoes

R. L.

Ferreira

T. P.

Silva

P. M.

Hogaboam

C. M.

Kunkel

S. L.

Fierro

I. M.

Canetti

Benjamim

C. F.

(2009). ATLa, an aspirin-triggered lipoxin A4 synthetic analog, prevents the inflammatory and fibrotic effects of bleomycin-induced pulmonary fibrosis. J Immunol 182, 5374–81.

45.

Meister

Anderson

M. E.

(1983). Glutathione. Annu Rev Biochem 52, 711–60.

46.

Meneghin

Hogaboam

C. M.

(2007). Infectious disease, the innate immune response, and fibrosis. J Clin Invest 117, 530–8.

47.

Moeller

Ask

Warburton

Gauldie

Kolb

(2008). The bleomycin animal model: A useful tool to investigate treatment options for idiopathic pulmonary fibrosis? Int J Biochem Cell Biol 40, 362–82.

48.

Moore

B. B.

Hogaboam

C. M.

(2008). Murine models of pulmonary fibrosis. Am J Physiol Lung Cell Mol Physiol 294, L152–60.

49.

Muller

W. E.

Zahn

R. K.

(1977). Bleomycin, an antibiotic that removes thymine from double-stranded DNA. Prog Nucleic Acid Res Mol Biol 20, 21–57.

50.

Odajima

Betsuyaku

Nagai

Moriyama

Wang

D. H.

Takigawa

Ogino

Nishimura

(2010). The role of catalase in pulmonary fibrosis. Respir Res 11, 183.

51.

Pardo

Selman

(2002). Molecular mechanisms of pulmonary fibrosis. Front Biosci 7, d1743–61.

52.

Phan

S. H

. (2002). The myofibroblast in pulmonary fibrosis. Chest 122, 286S–9S.

53.

Raghu

Weycker

Edelsberg

Bradford

W. Z.

Oster

(2006). Incidence and prevalence of idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 174, 810–6.

54.

Rahman

Kilty

(2006). Antioxidant therapeutic targets in COPD. Curr Drug Targets 7, 707–20.

55.

Rahman

MacNee

(1996). Role of oxidants/antioxidants in smoking-induced lung diseases. Free Radic Biol Med 21, 669–81.

56.

Rueff-Barroso

C. R.

Trajano

E. T.

Alves

J. N.

Paiva

R. O.

Lanzetti

Pires

K. M.

Bezerra

F. S.

Pinho

R. A.

Valenca

S. S.

Porto

L. C

. (2010). Organ-related cigarette smoke-induced oxidative stress is strain-dependent. Med Sci Monit 16, BR218–26.

57.

Russo

R. C.

Garcia

C. C.

Barcelos

L. S.

Rachid

M. A.

Guabiraba

Roffe

Souza

A. L.

Sousa

L. P.

Mirolo

Doni

Cassali

G. D.

Pinho

Locati

Teixeira

M. M.

(2010). Phosphoinositide 3-kinase gamma plays a critical role in bleomycin-induced pulmonary inflammation and fibrosis in mice. J Leukoc Biol 89, 269–82.

58.

Russo

R. C.

Garcia

C. C.

Barcelos

L. S.

Rachid

M. A.

Guabiraba

Roffe

Souza

A. L.

Sousa

L. P.

Mirolo

Doni

Cassali

G. D.

Pinho

Locati

Teixeira

M. M.

(2011). Phosphoinositide 3-kinase gamma plays a critical role in bleomycin-induced pulmonary inflammation and fibrosis in mice. J Leukoc Biol 89, 269–82.

59.

Sato

Tashiro

Chibana

Yamashita

Karakawa

Kohrogi

(2008). Role of lipid-derived free radical in bleomycin-induced lung injury in mice: Availability for ESR spin trap method with organic phase extraction. Biol Pharm Bull 31, 1855–9.

60.

Segel

M. J.

Izbicki

Cohen

P. Y.

Christensen

T. G.

Wallach-Dayan

S. B.

Breuer

(2003). Role of interferon-gamma in the evolution of murine bleomycin lung fibrosis. Am J Physiol Lung Cell Mol Physiol 285, L1255–62.

61.

Strieter

R. M.

(2008). What differentiates normal lung repair and fibrosis? Inflammation, resolution of repair, and fibrosis. Proc Am Thorac Soc 5, 305–10.

62.

Tanaka

Ishihara

Azuma

Kudoh

Ebina

Nukiwa

Sugiyama

Tasaka

Namba

Sato

Mizushima

(2010). Therapeutic effect of lecithinized superoxide dismutase on bleomycin-induced pulmonary fibrosis. Am J Physiol Lung Cell Mol Physiol 298, L348–60.

63.

Valenca

S. S.

Pimenta

W. A.

Rueff-Barroso

C. R.

Ferreira

T. S.

Resende

A. C.

Moura

R. S.

Porto

L. C.

(2009). Involvement of nitric oxide in acute lung inflammation induced by cigarette smoke in the mouse. Nitric Oxide 20, 175–81.

64.

Vlahovic

Russell

M. L.

Mercer

R. R.

Crapo

J. D.

(1999). Cellular and connective tissue changes in alveolar septal walls in emphysema. Am J Respir Crit Care Med 160, 2086–92.

65.

Weibel

E. R.

Hsia

C. C.

Ochs

(2007). How much is there really? Why stereology is essential in lung morphometry. J Appl Physiol 102, 459–67.

66.

Woessner

J. F.,

Jr. (1961). The determination of hydroxyproline in tissue and protein samples containing small proportions of this amino acid. Arch Biochem Biophys 93, 440–7.

67.

Zelko

I. N.

Mariani

T. J.

Folz

R. J.

(2002). Superoxide dismutase multigene family: A comparison of the CuZn-SOD (SOD1), Mn-SOD (SOD2), and EC-SOD (SOD3) gene structures, evolution, and expression. Free Radic Biol Med 33, 337–49.

68.

Ziesche

Hofbauer

Wittmann

Petkov

Block

L. H.

(1999). A preliminary study of long-term treatment with interferon gamma-1b and low-dose prednisolone in patients with idiopathic pulmonary fibrosis. N Engl J Med 341, 1264–9.