Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Although apoptosis has been implicated as a homeostatic mechanism, it may play a role in human diseases in two different ways. First, diseases may be caused by a malfunction of the apoptosis mechanism. Repair after an acute lung injury requires the elimination of proliferating mesenchymal and inflammatory cells from the alveolar airspace or alveolar walls (1, 2). Failure to clear unwanted cells by apoptosis will prolong the inflammation because of the release of their toxic contents. Second, excessive apoptosis may cause disease. An injection of monoclonal anti-Fas antibody (Jo2) into adult mice caused hepatic failure and death (3), suggesting that acute fulminant hepatitis in humans may be caused by apoptosis mediated by Fas antigen (Fas).

Fas antigen is a cell surface protein that mediates apoptosis. It is expressed in various cells and tissues including the thymus, liver, ovary, heart, and lung. It shares structural homology with a number of cell surface receptors, including tumor necrosis factor receptor and nerve growth factor receptor (4). Mice carrying the lymphoproliferative (lpr) mutation have defects in the Fas antigen gene (5). The lpr mice develop lymphadenopathy and suffer from a systemic lupus erythematosus-like autoimmune disease, indicating an important role for Fas antigen in the negative selection of autoreactive T cells in the thymus (1).

Damage to and the loss of epithelial cells are commonly seen in acute lung injury and also in chronic fibrosing alveolitis. The acute pulmonary toxicity induced by bleomycin in vivo leads to DNA damage (6), which is known to induce apoptosis in vitro (7, 8). We previously found the incidence of apoptosis, Fas mRNA expression in alveolar epithelial cells, and Fas ligand (FasL) mRNA expression in infiltrating lymphocytes in a mouse model of bleomycin-induced pulmonary fibrosis (9). FasL in the physiologic cell death mechanism has a short half-life and may act only at a short distance when it is present at a sufficiently high concentration (10). In bleomycin-induced pulmonary fibrosis, FasL mRNA was continuously and excessively expressed in infiltrating lymphocytes (9). Therefore, we hypothesized that excessive apoptosis mediated by overexpression of FasL may induce pulmonary fibrosis. In this study, we investigated whether continuous inhalation of anti-Fas antibody can induce excessive and chronic apoptosis in bronchiolar and alveolar epithelial cells, which leads to pulmonary fibrosis in mice.

Tumor necrosis factor  (TNF-) promotes inflammation, cell migration, and proliferation. TNF- mRNA and protein have been detected in lungs from patients with idiopathic pulmonary fibrosis (11) and in lungs from mice with pulmonary fibrosis induced by bleomycin (12). Miyazaki and coworkers demonstrated that pulmonary fibrosis progressed in TNF- transgenic mice (13). Transforming growth factor  (TGF-) is unique in its widespread actions that enhance the deposition of extracellular matrix. In the model of pulmonary fibrosis induced by bleomycin, total lung TGF- mRNA (14) and protein content (15) were higher than those in normal mice, and the increased production of TGF- preceded the synthesis of collagens, fibronectin, and proteoglycans (16). In immunohistochemical study, TGF- was observed in bronchiolar epithelial cells of patients with advanced idiopathic pulmonary fibrosis (17). Therefore, we also assessed the expression of TNF- and TGF- mRNA in this model.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Mice and Anti-Fas Antibody Treatment

In these experiments, 6-week old ICR mice were divided into three groups. The first and second groups consisted of control mice, treated with saline only (n = 40) or with isotype-matched control hamster IgG antibody (n = 40) (Organon Teknika Co., Durham, NC), respectively. The third group (n = 80) was treated with anti-mouse Fas monoclonal antibody (JO-2) (Pharmingen, San Diego, CA). Following measurement of body weight, mice were put in a chamber and allowed to inhale 10 ml of an aerosolized anti-Fas antibody (1 or 10 µg/ml in saline), hamster IgG antibody (20 µg/ml), or saline for only 30 min. After inhalation, mice were returned to their cages and allowed food and water ad libitum. This treatment was continued for 2 weeks every other day. Mice (n = 8 per each time point) were anesthetized 6 h, or 1, 7, 14, or 28 d, after the inhalation and prepared for bronchoalveolar lavage (BAL) (n = 3), light microscopy and electron microscopy (n = 2), or RNA extraction and hydroxyproline measurement (n = 3).

Bronchoalveolar Lavage

The animals were anesthetized with an intraperitoneal injection of sodium pentobarbital (Dinabot Co., Osaka, Japan) either 6 h, or 1, 7, 14, or 28 d, after inhalation. After thoracotomy, the lungs were explored. Following insertion of a tracheal tube, BAL was performed through the cannulated tube with 5 ml of sterile saline at room temperature. The recovered fluid was filtered through a single layer of gauze to remove mucus. Cells in the lavage fluid were counted using a hemocytometer. Differential counts were performed on a total of 100 cells stained with Diff-Quick (Baxter Diagnostics, Düdingen, Switzerland).

Histology of the Lung

After thoracotomy, the pulmonary circulation was flushed with saline, and the lungs were explored. After sacrifice, the lung samples were inflated with 10% formalin solution instilled at 15 cm H₂O pressure through the trachea for 2 h and fixed with buffered 10% formalin solution for 24 h. After embedding in paraffin, the sections were prepared and stained with hematoxylin and eosin, and Elastica-Van Gieson staining was performed.

Electron Microscopy

For electron microscopy, the anti-Fas antibody-treated mice were killed on day 14 after the anti-Fas antibody inhalation, and lungs were fixed with 2.5% glutaraldehyde in 0.1 M phosphate buffer, pH 7.4, for 18 h. The lungs were dissected into small pieces and postfixed for 1.5 h in 1% OsO₄ dissolved in 0.1 M phosphate buffer (pH 7.4), then dehydrated through a series of graded ethanol solutions and embedded in Epon. Ultrathin sections were cut, stained with uranyl acetate and lead nitrate, and examined under a JEM-1200 EX transmission electron microscope (JEOL Co., Tokyo, Japan).

DNA Nick End Labeling of Tissue Sections

TdT-mediated dUTP-biotin nick end labeling (TUNEL) was performed according to the protocols described by Gavrieli and coworkers (18) with slight modifications (9).

Hydroxyproline Assay

The animals inhaled anti-Fas antibody (Jo2) at 1 or 10 µg/ ml, isotype-matched control hamster IgG at 20 µg/ml in saline, or saline alone for 30 min every other day for 2 weeks. The animals were anesthetized with an intraperitoneal injection of sodium pentobarbital on day 1, 7, 14, or 28. The lungs were frozen in liquid nitrogen, lyophilized (Lyph Lock 12; Labconco, Kansas City, MO), weighed, and minced to a fine homogeneous mixture. Lung tissue was hydrolyzed in 6 M HCl for 16 h at 120°C. The hydroxyproline content of each sample was determined according to the protocols of Woessner (19).

RNA Preparation and Analysis

Total RNA was prepared from lung tissues by use of an Isogen RNA extraction kit (Nippon Gene, Tokyo, Japan). For the polymerase chain reaction (PCR) analysis of RNA, cDNA was prepared by reverse transcription (RT) of 4 µg of each RNA sample in a 20-µl reaction volume containing 10 mM Tris-HCl (pH 8.8), 50 mM KCl, 4 mM MgCl₂, 0.1% Triton X-100, 1 mM dithiothreitol (DTT), 0.25 mM dNTPs, 5 µM random hexamer primers, 0.1 U/µl of ribonuclease inhibitor (Promega Corp., Madison, WI), and 10 U/µl of Moloney murine leukemia virus reverse transcriptase (Mo-MuLV-RT) (GIBCO-BRL, Gaithersburg, MD). The reaction mixture was incubated at 42°C for 1 h, and at 95°C for 5 min. The cDNAs were then diluted to 100 µl, and these cDNAs were used in all PCRs. The PCR amplifications were performed in a 50-µl reaction volume containing 5 µg of each cDNA, 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 2.5 mM MgCl₂, 0.1% Triton X-100, 0.2 mM dNTPs, and 1.25 U of Taq polymerase (Nippon Gene). The primers and probes used were as follows:

-actin:   Sense 5'-TCCTGTGGCATCCATGAAACT-3'

  Antisense 5'-CTTCGTGAACGCCACGTGCTA-3'

  Probe 5'-GGAGATTACTGCTCTGGCTC-3'

TGF-:   Sense 5'-ACCATCCATGACATGAACCG-3'

  Antisense 5'-TCCCAGACAGAAGTTGGCAT-3'

  Probe 5'-TTCAGCTCCACAGAGAAGAACTGC-3'

TNF-:   Sense 5'-GGCAGGTCTACTTTGGAGTCATTGC-3'

  Antisense 5'-ACATTCGAGGCTCCAGTGAA-3'

  Probe 5'-TATGGCTCAGAGTCCAACTC-3'

The conditions for amplification were as follows: -actin, TGF-, TNF-; 93°C for 3 min for 1 cycle, 93°C for 1 min, 63°C for 1 min, 72°C for 2 min for 35 cycles, and 72°C for 7 min for 1 cycle. Cycle curve studies confirmed that, for the amounts of cDNA being amplified, the reactions had not reached the plateau of the amplification curve at 35 cycles for any primer pair. Negative controls performed with no RT yielded no detectable fragments with either primer pair. PCR products for TGF-, TNF-, and -actin cDNA were transferred to Hybond-N hybridization transfer membranes (Amersham, Arlington Heights, IL). The membrane was hybridized with an oligonucleotide probe labeled with digoxigenin-ddUTP using the DIG oligonucleotide 3'-end labeling kit (Boehringer Mannheim Biochemicals, Indianapolis, IN). The hybridizations were performed according to manufacturer recommendations. The digoxigenin-labeled probe that hybridized with the PCR products was detected with the DIG nucleic acid detection kit (Boehringer Mannheim).

Statistical Analysis

To determine statistical significance, a Student's t test for nonpaired data was performed and values of P < 0.05 and P < 0.01 were considered significant.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

BAL Fluid Cell Analysis

Figure 1 shows that the number of total cells, macrophages, neutrophils, and lymphocytes in BAL fluid (BALF) were significantly increased on day 7 in anti-Fas antibody-treated mice compared to control IgG-treated mice (**P < 0.01, *P < 0.05). It should be stressed that the number of lymphocytes was still significantly increased on day 14, and even on day 28 compared to control mice, whereas neutrophils rapidly disappeared from BALF by day 14. There was no change in BALF cell population in saline-treated mice.

View larger version (29K): [in this window] [in a new window]   Figure 1.   The results of BAL cell number analysis. Data are shown as the mean ± SEM obtained from three mice. Significance (*P < 0.05, **P < 0.01) was determined by a Student's t test.

Microscopic Findings in Mouse Lung

Figure 2 shows representative histologic findings in this model. On day 1, inflammatory cells infiltrated only the area surrounding the bronchioles (Figure 2A). By day 7, alveolar walls were thickened with edema and infiltrated with neutrophils and mononuclear cells (Figure 2B). After 14 days, the alveolar septa were infiltrated with lymphocytes and plasma cells (Figure 2C). After 28 days, a large number of lymphocytes infiltrated into the interstitium, and proliferation of fibroblasts was observed (Figures 2D and 2E). There were only minimal changes in isotype-matched control antibody-treated mice on day 14 (Figure 2F). On day 28, Elastica-Van Gieson staining showed the deposition of microfibrils in thickened alveolar walls, and fibroblast accumulation in areas where alveolar spaces had collapsed (Figures 2G and 2H).

View larger version (152K): [in this window] [in a new window]   Figure 2.   Histologic assessment of mice lung. Lung tissues were prepared 1 h after inhalation, fixed in 10% formalin, routinely processed, and stained with hematoxylin and eosin. (A) On day 1, inflammatory cells were found around the bronchioles. (B) On day 7, alveolar walls were thickened with edema and neutrophils and mononuclear cells had infiltrated into the interstitium. (C) After 14 d, the alveolar septa were more thickened than on day 7, and infiltrated predominantly with lymphocytes and plasma cells. After 28 d (D and E), a large number of lymphocytes had infiltrated into the interstitium; thickening of alveolar septa, collapse of alveolar spaces, and proliferation of fibroblasts were observed. (F) There were only minimal changes in isotype-matched control antibody-treated mice on day 28. Elastica-Van Gieson staining shows marked staining of fibrils in interstitium on day 28 (G and H). (Original magnification: A-D and F, ×62.5; E and G, inset in B, ×125; H, ×250.)

DNA Fragmentation Analysis of Lung Tissue

Localization of DNA fragmentation was determined by TUNEL. TUNEL demonstrated positive signals in nuclei of bronchiolar epithelial cells on day 1 after anti-Fas antibody inhalation (Figure 3A). These signals were continued until day 28. In addition, positive signals were also found in nuclei of alveolar epithelial cells surrounding the bronchioles and signal distribution was gradually expanded to the whole lung field from day 1 to 14 (Figure 3B); these signals had resolved by day 28. 

View larger version (121K): [in this window] [in a new window]   Figure 3.   In situ DNA nick end labeling of tissue sections (TUNEL). TUNEL demonstrated positive signals mainly in bronchiolar epithelial cells on day 1 (A). On day 14, positive signals of alveolar epithelial cells were evident and diffuse (B). These positive signals were not found in lung specimens from isotype-matched control IgG-treated mice (C) or when TdT was eliminated (D) on day 14. (Original magnification: A, ×62.5; B-D, ×125, insets in A and B, ×250.)

Electron Microscopic Findings

Electron microscopic findings confirmed the presence of apototic nuclei in bronchiolar and alveolar epithelial cells. Bronchiolar epithelial cells, particularly Clara cells, contained condensed and fragmented nuclei. Chromatin condensation was also found in the nuclei of alveolar epithelial cells (Figure 4). These epithelial cells presented many vacuoles in their cytoplasm. Although typical morphological features of apoptosis were observed in only 12 of 240 (5%) epithelial cells, chromatin condensation and variously sized cytoplasmic vacuoles were detected in as many as 144 of 240 (60%) epithelial cells. Inhalation of control antibody or saline alone did not induce apoptosis or nuclear changes in epithelial cells.

View larger version (149K): [in this window] [in a new window]   Figure 4.   Electron microscopic findings. Bronchiolar epithelial cells on day 14, particularly Clara cells, contained condensed (B) and fragmented nuclei (arrow, A). Type II alveolar epithelial cells on day 14 contained multiple nuclei with margination and condensation of the chromatin and cell shrinkage, and also contained intact cytoplasmic organelles such as mitochondria and rough endoplasmic reticulum (C). Apoptotic cells (arrow) were engulfed in the cytoplasm of macrophages in the interstitium (D). (Bars: 1 µm.)

Hydroxyproline Assay

To quantitate the deposition of collagen, lung tissue was treated with acid and the hydroxyproline content was determined. There was a significant increase in lung hydroxyproline content in mice treated with 1 or 10 µg/ml anti-Fas antibody (day 14 [P < 0.05]; day 28 [1 µg/ml, P < 0.05; 10 µg/ml, P < 0.01]), compared with mice treated with control antibody or saline (Figure 5). There was also a significant difference in lung hydroxyproline content by day 28 between 1- and 10-µg/ml anti-Fas antibody-treated mice (P < 0.05).

View larger version (31K): [in this window] [in a new window]   Figure 5.   Effect of anti-Fas antibody on lung hydroxyproline content. Data are shown as the mean ± SEM from three mice. Significance (*P < 0.05, **P < 0.01) was determined by a Student's t test.

Cytokine Gene Expression

The expression of TGF- mRNA was upregulated in anti-Fas antibody-treated mice from day 1 to day 28, especially from day 7 to day 28, but not in mice treated with control antibody. TNF- mRNA were upregulated from day 1 to day 7 in both groups (Figure 6).

View larger version (39K): [in this window] [in a new window]   Figure 6.   RT-PCR analysis for the expression of TNF- and TGF- mRNA in lung tissues. Total RNA was extracted from lung tissues at 6 h, day 1, day 7, day 14, and day 28 after the inhalation of anti-Fas antibody (anti-Fas) and isotype-matched control IgG antibody (control IgG). RT-PCR amplification products were transferred to filters, and hybridized with an oligonucleotide probe labeled with digoxigenin-ddUTP.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The control of cell number is determined by a balance between cell proliferation and cell death. Although physiologic cell death occurs primarily through an evolutionarily conserved form of cell suicide termed apoptosis, excessive apoptosis may lead to a failure to clear the apoptotic cells, which results in the amplification of inflammation by released cell contents. Fas is abundantly expressed not only in the liver, but also in the heart and lungs. The primary cells from these tissues are sensitive to Fas-mediated apoptosis (20). It is possible that an abnormally activated Fas- FasL system plays a role in human disease. Hepatocytes transformed with human hepatitis C virus (4) express virus antigens and activate cytotoxic T cells to express FasL, which then bind to Fas on hepatocytes, inducing them to undergo apoptosis. This process normally occurs in the removal of virus-infected cells, but excessive apoptosis may lead to fulminant hepatitis.

We previously demonstrated that Fas mRNA was expressed in bronchiolar and alveolar epithelial cells and Fas ligand mRNA was upregulated in infiltrating lymphocytes in bleomycin-induced pulmonary fibrosis in mice (9). Therefore, we investigated whether Fas-FasL cross-linking can lead to pulmonary fibrosis. In this study, the repeated inhalation of anti-Fas antibody induced apoptosis of bronchiolar and alveolar epithelial cells and pulmonary fibrosis. On day 1, positive TUNEL signals were observed mainly in bronchiolar epithelial cells and in some alveolar epithelial cells around the bronchioles. On day 14, these signals of alveolar epithelial cells became evident and diffuse. It is possible that inhaled anti-Fas antibody was trapped mostly in the bronchioles, but that some antibodies reached the alveolar space. The repeated inhalation of antibody and infiltrating inflammatory cells caused DNA strand breaks in alveolar epithelial cells. The localization of TUNEL-positive signals was compatible with the localization of Fas mRNA in mice as previously described (9). The electron microscopic findings confirmed apoptotic changes in these cells (Figure 4). Analysis of hydroxyproline contents and microscopic findings demonstrated the proliferation of fibroblasts and the collagen deposition in lung tissues. These results demonstrated that excessive apoptosis caused by the Fas-FasL system can induce infiltration of inflammatory cells and pulmonary fibrosis in mice.

There is some evidence that the presence of an intact epithelial cell layer directly controls fibroblast proliferation (21, 22). Studies on the repopulation of denuded tracheal explants by epithelial cells showed that the denuded tracheal implants fill rapidly with fibroblasts, unless enough epithelial cells are introduced into the lumen to control fibroblast proliferation (21). Alternatively, epithelial cells may control the proliferation of fibroblasts by releasing cytokines that downregulate fibroblast activity. Therefore, it is possible that excessive apoptosis of epithelial cells may induce the proliferation of fibroblasts through the loss of inhibitory function against fibroblast proliferation. In polycystic human kidneys, it also has been demonstrated that apoptotic nuclei can be detected in noncystic tubular epithelial cells, but not in kidneys from normal humans (23). This suggests that apoptosis may be associated with the progressive loss of renal tissue in polycystic kidney disease, which may lead to secondary interstitial fibrosis.

In this study, we also demonstrated the kinetics of two key cytokines (TGF- and TNF-) that were detected in a number of animal models of fibrosis and in pulmonary fibrosis of human. TGF- mRNA was upregulated in anti-Fas antibody-treated mice, especially from day 7 to day 28. In contrast, TNF- mRNA was upregulated in the early phase (from day 1 to day 7) after treatment. The kinetic study on the appearance of TNF- and TGF- showed that there was an induction of TNF- in the early phase, presumably monocyte derived, and subsequently TGF- was increased in a rat model of bleomycin-induced pulmonary fibrosis (24). TGF- was observed using immunohistochemistry (17) in bronchiolar epithelial cells in lung tissues from patients with advanced idiopathic pulmonary fibrosis. The kinetics of these cytokines in the Fas antibody-induced fibrosis model seemed similar to those in the bleomycin-induced model. In addition, Abreu and coworkers demonstrated that Fas-Fas antibody ligation can lead to production of interleukin 8 by colonic epithelial cells in vitro, which represents another function mediated by Fas in addition to apoptosis (25). This result suggests that unknown functions of the anti-Fas antibody, other than causing apoptosis, may induce new gene expression. It should be determined whether anti-Fas antibody cross-linking can induce the expression of TGF- and other cytokines in bronchiolar and alveolar epithelial cells and in macrophages in vitro.

In conclusion, repeated inhalation of anti-Fas antibody mimicking Fas-FasL cross-linking resulted in excessive cell death of epithelial cells, which seemed to overwhelm the clearing mechanism necessary to maintain homeostasis. This condition may prolong the inflammation and interfere with re-epithelialization, which results in the overgrowth of mesenchymal cells. We have demonstrated here that the excessive apoptosis caused by anti-Fas antibody cross-linking leads to persistent inflammation and pulmonary fibrosis.