INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Pulmonary fibrosis is a potentially lethal disorder for which we have not yet developed any useful therapeutic tools. Pulmonary fibrosis occurs after various known or unknown lung injuries and is characterized by alveolar wall fibrosis with an accumulation of extracellular matrix, notably collagen, and by an increased number of activated alveolar macrophages (1, 2). Pulmonary fibrosis is commonly preceded by or associated with an inflammatory response. Although the relationship between inflammation and fibrogenesis is not entirely clear, many studies have demonstrated that monocytes/macrophages are a source of mediators capable of regulating fibroblast proliferation and other functions that are responsible for the fibrogenic outcome (2). Clinical and experimental findings have shown the primary role of various types of cytokine release from activated alveolar macrophages and monocytes, including tumor necrosis factor (TNF)- and interleukin (IL)-1 in the pathogenesis of pulmonary fibrosis. The release of IL-1 and TNF- from alveolar macrophages is known to be enhanced, and the gene expressions of these two cytokines are upregulated in alveolar macrophages and/or monocytes from patients with idiopathic pulmonary fibrosis (IPF) and those with asbestosis (3). Animal models and in vitro studies have shown that macrophage cytokines, namely, IL-1 and TNF-, play significant roles in the development of pulmonary fibrosis (4). In addition, IL-6 release from alveolar macrophages has also been implicated to play a role in the pathogenesis of pulmonary fibrosis in human and in experimentally induced pulmonary fibrosis (7). Experimentally, exposure to these cytokines in human lung fibroblasts has been shown to result in the upregulation of collagen types I and II and fibronectin gene expression (3). There have been reports of an abundant expression of transforming growth factor (TGF)- mRNA in alveolar macrophages (8) and a marked, consistent increase in TGF- production in epithelial cells and macrophages in the lung sections of patients with advanced IPF (9). Thus, there has been a general concept that TNF- and IL-1 (and IL-6) are involved in the initial inflammation, and the resultant expression of TGF- and platelet-derived growth factor (PDGF) are involved in the following fibrosis (2).

Bleomycin-induced pneumopathy in rodents has been used as an animal model of pulmonary fibrosis because it resembles that seen in humans. Similarly, in this model, the expression and release of IL-1 and TNF- from alveolar macrophages have also been shown to play a critical role in the early phase of inflammatory responses, and the resultant expression of TNF-, PDGF-A, and insulinlike growth factor (IGF)-I participate in the development of the bleomycin-induced pulmonary fibrosis (10). For example, bleomycin-induced pulmonary toxicity and fibrosis in mice is associated with a marked upregulation of the TNF- mRNA levels and can be prevented by anti-TNF- antibodies (4) and recombinant human soluble TNF- receptor (13). Anti-TNF- antibody caused a significant reduction in lung fibrosis, which was accompanied by the suppression of lung TGF- expression, in mice treated with bleomycin (10), and antibodies to TGF- prevented collagen accumulation in the lungs of mice treated with bleomycin (14).

Meanwhile, a crucial transcription factor that regulates the expression of TNF- and IL-1 genes in macrophages and monocytes is nuclear factor-B (NF-B) (15). Products of genes that are regulated by NF-B also cause the activation of NF-B. The proinflammatory cytokines TNF- and IL-1 both activate and are activated by NF-B. This type of positive regulatory loop may amplify and perpetuate the local inflammatory response (16). In fact, NF-B activation with an increased level of TNF- mRNA has been demonstrated in the lungs with bleomycin-induced pulmonary fibrosis (17).

Therefore, we performed the present experiment to determine if antisense oligonucleotides to NF-B may be a potential therapeutic tool for pulmonary fibrosis. The antisense oligonucleotides were incorporated into activated macrophages and monocytes through phagocytosis without using any specific vectors (virus, liposome, etc.) or gene transfection, and the resulting resolution of acute lung injury and pneumonitis/ fibrosis thereby improved the survival rate in mice of bleomycin-induced pulmonary pneumopathy.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals

Female C57BL/6CrSlc mice 8 wk of age and weighing 17 to 20 g were purchased from SLC Inc. of Japan (Tokyo, Japan). The animals were specific pathogen-free and arrived in filtered cages. Animals were kept in separate clean rooms and provided with food and water ad libitum, and the cages were placed in laminar flow hoods to minimize pulmonary infections. The animals were assigned to the following groups:

Control Groups

The control groups consisted of animals that received 200 µl of 0.9% NaCl solution by intravenous bolus via the tail vein on the following corresponding days (n = 10) and animals that received 150 mg/kg (n = 19) or 300 mg/kg (n = 10) bleomycin dissolved in 200 µl of 0.9% NaCl solution.

Antisense Group

Murine antisense phosphorothioate (S) oligonucleotide consisted of 19-mer analogues to the 5' end of the p65 subunit of NF-B, spanning the translation initiation site. The sequences of phosphorothioate oligonucleotides were as follows (18).

murine p65 antisense: 5'-GAAACAGATCGTCCATGGT-3'

murine p65 sense: 5'-ACCATGGACGATCTGTTTC-3'

We ordered and obtained these oligonucleotides from Oligo-Service Institute of Tsukuba (Ibaragi, Japan). These phosphorothioate (S) oligonucleotides are more stable in vivo or more resistant to intracellular enzymes (exonucleases and endonucleases) as compared with natural phosphodiesters (19). Further, when intravenously administered to the mouse, 85% or more of these S-oligonucleotides remain in the plasma 24 h after the injection, with a half-life of 5.1 d (19).

Nine hundred micrograms per mouse of antisense or sense phosphorothioate oligonucleotides to the p65 subunit of NF-B were dissolved in 200 µl of 0.9% NaCl solution and injected via the tail vein. The antisense or sense oligonucleotides were administered 6 h before and 5 d after 150 mg (n = 10 each) or 300 mg/kg (n = 10 each) bleomycin injection. Further, the antisense oligonucleotides were injected in an additional 10 animals only 5 d after the 150-mg/kg bleomycin injection. To determine the effects of the antisense oligonucleotides themselves, five animals were treated with the antisense oligonucleotides alone.

The experiments described above were performed to determine the effects of the antisense oligonucleotides on the mortality and body weight of mice. Additional animals were used for the following experiments.

Bronchoalveolar Lavage

Animals were pretreated with and without the antisense oligonucleotides (6 h before and/or 5 d after bleomycin injection), and the right lungs from animals killed before and at 12 h, 1, 2, 3, and 5 d, and 1, 2, and 4 wk after the 150-mg/kg bleomycin injection (n = 5 each) were used for BAL analysis. The additional animals were used to determine the effect of the antisense oligonucleotides on BAL cells (n = 5 each). The trachea was cannulated with a 20-gauge catheter that was secured with a ligature, and lavage of the right lung was performed using a 1-ml aliquot of saline (0.9% NaCl at room temperature) with a total of four lavages being performed. In all animals studied, the recovery of lung lavage fluid was 70% or greater. The recovered fluid was centrifuged at 800 × g for 10 min to sediment the cells. Total cell recovery was determined by Trypan blue exclusion. Two hundred cells of a Wright-Giemsa-stained cytocentrifuge slide preparation were counted for determination of the differential cell recovery.

Immunofluorescence Study

FITC-labeled p65 phosphorothioate (S)-oligonucleotides were synthesized with Expedite 8909 (PE Biosystems, Foster, CA) using 5'-fluorescein phosphoramidite (Glen Research Co., Sterling, VA) and the yield equaled the coupling efficiency of the phosphoramidite, showing virtually pure fluorescein oligonucleotides. The FITC-labeled antisense oligonucleotides were intravenously administered to animals according to the same procedure as in the actual experiments. The addition of the fluorescein did not produce any changes in the stability. Four animals without bleomycin injection were killed 12 and 24 h after the control saline injection with pretreatment with FITC-labeled oligonucleotides. The other four animals received the FITC-labeled oligonucleotides prior to the injection of 150 mg/kg bleomycin and were killed 12 and 24 h after the bleomycin injection. FITC-positive cells in both BALF and peripheral blood samples were visualized with a fluorescence microscope (ACASS 570; Meridian Instruments Inc., Okemos, MI) at an excitation wavelength of 490 nm.

TNF- Enzyme-linked Immunosorbent Assay

Blood serum and BALF samples were obtained by cardiac puncture and by lavage of the right lung, respectively, from animals killed before and at 6 and 12 h, and 1, 2, and 3 d after 150 mg/kg bleomycin injection (n = 5 each). Further, both samples were obtained from animals 6 and 12 h after 150 mg/kg bleomycin injection with pretreatment with the antisense oligonucleotides (n = 5 each). Samples were immediately placed on melting ice and subsequently centrifuged at 0° C and stored at 20° C until the assay. Immunoreactive TNF- in both blood serum and BALF samples was quantified by an ELSA kit with a minimum detectable dose of 5.1 pg/ml (Quantikine M mouse TNF-; R&D Systems, Minneapolis, MN). Each sample was assayed in triplicate according to the manufacturer's protocol.

Hydroxyproline Analysis

The right lungs from animals killed 14 and 28 d after bleomycin injection were frozen and kept at 70° C until use. The total right lung collagen content was estimated by an assay for hydroxyproline (20). After acid hydrolysis of the lung with 6 N HCl at 110° C for 16 h in a sealed tube, the hydroxyproline content was determined by high performance liquid chromatography (21) and expressed as nanomoles per lung.

Histologic Examination

Mice were killed 1, 2, 3, 4, 5, 6, 7, 14, and 28 d after bleomycin injection by exsanguination under ketamine anesthesia. The left lung was instilled with 10% formalin until the pleural surface became smooth and then immersed in the same fixative. Multiple paraffin-embedded 4-µm sections of the entire lung were prepared in the standard manner and stained with hematoxylin-eosin and Masson's trichrome.

The left lungs from animals killed 14 and 28 d after bleomycin injection were used for semiquantitative morphometric analysis using a numerical pathologic scale (22). Five sections of the entire lung stained with hematoxylin-eosin were chosen at random from each animal. More than 30 successive microscopic fields at a magnification ×100 from each animal were allotted a score from 0 (normal) to 6 (severest) using a predetermined scale of severity (22). The degree of cell infiltration and pneumonitis/fibrosis in the microscopic sections of the lung was assessed as the mean score for the observed fields in each animal. Grading was done in a blinded fashion by two observers, and the mean was used as the pathologic score.

Western Blot Analysis

The number of BALF cells (mainly alveolar macrophages) was 2 × 10⁵ per animal, which was too small to extract nuclear protein for Western blotting. Instead of alveolar macrophages, therefore, we attempted to use peritoneal macrophages for the analysis. Peritoneal lavages of animals were performed three times with 1.5 ml saline, and the number of recovered peritoneal cells (mainly macrophages) was 10⁷ or more per animal, from which we could isolate 20 µg or more of the nuclear protein. The peritoneal lavage was performed in animals 6, 12, and 24 h after the 150-mg/kg bleomycin injection with pretreatment with control saline, and in animals 6, 12, and 24 h after the bleomycin injection with the pretreatment with the antisense or sense oligonucleotides (n = 3 each).

Next, we tried to detect the change of NF-B activity of total lung tissue cells. The left lung was obtained from animals before and at 6, 12, and 24 h after the 150-mg/kg bleomycin injection after the bleomycin injection with the pretreatment with the antisense oligonucleotides (n = 3 each). After the washout of blood by saline perfusion via the left ventricle, the lung was homogenated with a polytron homoginator in a RIPA buffer with 0.25 M phenylmethyl sulfonylfluoride and protease inhibitors (50 µg/ml leupeptin, 1 mM dithiothreitol, 10 µg/ml soybean tryptase inhibitor and 100 µg/ml aprotinin). Nuclear protein was obtained by ultracentrifugation (105,000 × g for 60 min. Automatic Preparatine Ultracentrifuge Hitachi 55P-72; Hitachi, Japan) of the supernatant from the centrifugation of the lung tissue homogenates (600 × g for 10 min).

Nuclear protein (20 µg) from peritoneal macrophages or 50 µg of that from lung tissue homogenates was blotted onto nitrocellulose membranes and immunoblotted with rabbit anti-p65 (Upstate Biotechnology Inc., NY) or PU.1 antibody (Santa Cruz Biotechnologies, Santa Cruz, CA). Bound antibodies were detected with peroxidase-coupled antirabbit IgG and enhanced chemiluminescence (ECM) (Amersham International, Buckinghamshire, UK).

Reagents

Clinical-grade bleomycin hydrochloride and methylprednisolone sodium succinate were supplied by Nippon Kayaku Co. (Tokyo, Japan) and Upjohn Pharma. Japan Ltd. (Tokyo, Japan), respectively. The other reagents were all purchased from Sigma Chemical (St. Louis, MO), except when stated otherwise.

Statistical Analysis

All data are expressed as mean ± SEM. Statistical evaluation was performed using the unpaired t test or by one-way analysis of variance for multiple comparison using the Newman-Keuls procedure on data of the BALF cell count, body weight, and hydroxyproline content, and the Mann-Whitney U test on pathologic scores. Survival curves were generated as described by Kaplan and Meier (1958). A p value of < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Mortality

Mortality caused by bleomycin was dose-dependent, with most deaths occurring between Days 6 and 9. Ten of 19 animals (53%) and all of 10 animals (100%) died between 7 and 9 d and between 6 and 8 d after 150 and 300 mg/kg bleomycin injection, respectively. All 10 animals treated with the p65 antisense oligonucleotides 6 h before and 5 d after bleomycin injection survived after 150 mg/kg bleomycin injection, and four of 10 animals (40%) survived after 300 mg/kg bleomycin injection (Figure 1 and Table 1). The survival rate of animals treated with the antisense oligonucleotides was significantly improved (p < 0.01 for 150 mg/kg and p < 0.05 for 300 mg/kg bleomycin injection), as shown in Figure 1. However, animals treated with the antisense oligonucleotides 5 d after bleomycin injection and with the sense oligonucleotides 6 h before and 5 d after bleomycin injection showed a rate of mortality similar to that with bleomycin alone (Table 1). All of the saline-treated control animals and animals treated with the antisense oligonucleotides alone survived until 28 d (Table 1).

View larger version (12K): [in this window] [in a new window]   Figure 1.   The survival rate of animals treated with 150-mg/kg bleomycin injection alone (BLM alone, n = 19) and animals treated with the p65 antisense oligonucleotides 6 h before and 5 d after 150-mg/kg bleomycin injection (Antisense + BLM, n = 10). The survival rate of animals treated with the antisense oligonucleotides significantly improved.

Weight Loss

Animals abruptly lost body weight until 7 d and gradually recovered the body weight for 14 d after the 150 mg/kg bleomycin injection. Treatment with the p65 antisense oligonucleotides clearly reduced the loss of body weight, which recovered within 10 d after the 150 mg/kg bleomycin injection (Figure 2). Animals lost 34% of the body weight 6 d after 150 mg/kg bleomycin injection and the p65 antisense oligonucleotides significantly reduced the weight loss by bleomycin to 17% at the same day (p < 0.001) (Figure 2). However, the sense oligonucleotides did not significantly alter the loss of body weight by bleomycin. In contrast, saline-treated control animals and animals treated with the antisense oligonucleotides alone did not show any loss of body weight.

View larger version (18K): [in this window] [in a new window]   Figure 2.   The change in body weight of animals treated with saline alone (Saline alone, n = 10) and animals treated with 150-mg/kg bleomycin injection alone (BLM alone, n = 19), animals treated with the p65 antisense oligonucleotides 6 h before and 5 d after 150-mg/kg BLM injection (Antisense + BLM, n = 10). Animals abruptly lost body weight until 7 d after 150-mg/kg bleomycin injection and gradually recovered the body weight by 21 d after the injection. Treatment with the antisense oligonucleotides clearly reduced the loss of body weight which recovered within 10 d after the 150-mg/kg bleomycin injection. *p < 0.05, **p < 0.01, ***p < 0.001, compared with BLM alone.

BAL Findings

BAL cells consisted of 96% macrophages and 4% lymphocytes in control animals without bleomycin administration. This cell differentiation did not significantly change after bleomycin except for a 3% increase in neutrophils 28 d after bleomycin injection. Total cells (95% or more macrophages) of BALF significantly decreased 12 h to 3 d after bleomycin injection, and thereafter increased reaching a value above the previous value 28 d after bleomycin injection (Figure 3). Treatment with the antisense oligonucleotides prevented the change in the number of BAL cells by bleomycin, whereas the antisense oligonucleotides themselves did not induce any changes in BAL cells (Figure 3).

View larger version (13K): [in this window] [in a new window]   Figure 3.   Total cell number in BALF from animals treated with 150-mg/kg bleomycin injection (BLM alone, n = 5 each), animals treated with the p65 antisense oligonucleotides 6 h before (and 5d after) 150-mg/kg BLM injection (Antisense + BLM, n = 5 each) and animals treated with the p65 antisense oligonucleotides alone (Antisense alone, n = 5 each). Total cells of BALF significantly decreased 12 h to 3 d after bleomycin injection and thereafter increased, reaching a value above the previous value 28 d after bleomycin injection. The p65 antisense oligonucleotides inhibited the changes in BAL cells by bleomycin, whereas the antisense oligonucleotides themselves did not induce any changes. *p < 0.05, **p < 0.01, compared to that before the injection (Day 0). #p < 0.05, ##p < 0.01, compared with BLM alone.

Immunofluorescence Study

FITC-positive cells were observed in BALF cells and in peripheral blood mononuclear and polynuclear cells from animals before and 12 to 24 h after bleomycin injection. The number of FITC-positive cells increased as did the intensity of fluorescence localized at the center of cells (probably nucleus) by bleomycin injection, compared with those after control saline injection (Figure 4).

View larger version (91K): [in this window] [in a new window]   Figure 4.   FITC-positive cells of BALF from animals 24 h after saline injection that received FITC-labeled p65 phosphorothioate (S)-oligonucleotides injection 6 h before the saline injection (A) and those from animals 24 h after 150-mg/kg bleomycin injection that received the FITC-labeled oligonucleotide injection 6 h before the bleomycin injection (B). Bleomycin injection produced an increase in FITC-positive cells not only in the number but also in the intensity of fluorescence which was localized at the center of the cells.

TNF- Level in Blood Serum and BALF

The TNF- level in blood serum samples significantly increased at 6 h (573 ± 78 pg/ml, p < 0.01) and 12 h (307 ± 54 pg/ml, p < 0.05) after 150 mg/kg bleomycin injection compared with that before the injection (16 ± 11 pg/ml). The elevated level in TNF- gradually returned to that before the injection, i.e., 102 ± 82 pg/ml, 25 ± 17 pg/ml, and 18 ± 21 pg/ml of TNF- at 1, 2, and 3 d after the injection, respectively. TNF- in BALF samples was detectable at 6 h (61 ± 9 pg/ml) and at 12 h (23 ± 12 pg/ml of three samples; not detectable in two samples) after 150 mg/kg bleomycin injection, but it was not detectable before or at 1, 2, or 3 d after the injection. The pretreatment with the antisense oligonucleotides significantly reduced the elevation of TNF- by bleomycin in both blood serum samples (108 ± 31 pg/ml, p < 0.05) and BALF samples (22 ± 11 pg/ml, p < 0.05) at 6 h after the injection.

Hydroxyproline Content

The hydroxyproline content per lung is shown in Figure 5. The hydroxyproline content per lung significantly increased after bleomycin injection, and the increase was significantly inhibited by the treatment with the antisense oligonucleotides, which themselves did not induce any changes (Figure 5). In contrast, the sense oligonucleotides failed to inhibit the increase in hydroxyproline by bleomycin (Figure 5).

View larger version (37K): [in this window] [in a new window]   Figure 5.   Hydroxyproline content per lung from animals treated with saline alone (Saline alone, 14 d; n = 6, 28 d; n = 5), animals treated with the p65 antisense oligonucleotides alone (Antisense alone, n = 5), animals treated with 150-mg/kg bleomycin injection alone (BLM alone, 14 d; n = 5, 28 d; n = 5), animals treated with the p65 antisense oligonucleotides 6 h before and 5 d after 150-mg/kg BLM injection (Antisense + BLM, n = 5) and animals treated with the p65 sense oligonucleotides 6 h before and 5 d after 150-mg/kg BLM injection (Sense + BLM, n = 4). Hydroxyproline content per lung was significantly increased by bleomycin injection, and the increase was significantly inhibited by the antisense oligonucleotides, which themselves did not induce any changes. In contrast, the sense oligonucleotides failed to inhibit the increase in hydroxyproline by bleomycin. **p < 0.01, ***p < 0.001, compared to Saline alone. ##p < 0.01 compared with BLM alone.

Pathology

Necropsies taken between 6 and 9 d after bleomycin injection revealed acute alveolar injury with hemorrhagic edema (Figure 6A). Intravenous administration of bleomycin produced acute alveolar injury with hemorrhagic edema and acute pneumonitis 3 to 5 d after bleomycin injection that progressed to severe interstitial and intra-alveolar pneumonia and/or fibrosis 2 and 4 wk after bleomycin injection (Figure 6B). The lesions 2 to 4 wk after bleomycin injection showed consolidation of the lung parenchyma with loss of the alveolar architecture and increased cellularity, containing increased numbers of alveolar macrophages, lymphocytes, and scattered enlarged atypical, and/or degenerated epithelial cells. Such lesions varied from focal, confluent to diffuse, and stainable collagen (evaluated by Masson's trichrome staining) was clearly elevated in the lesions (Figure 6C). Lung sections from saline-treated control animals or animals treated with the antisense oligonucleotides alone, killed at 28 d, showed no significant pulmonary consolidation or fibrosis. The antisense oligonucleotides significantly inhibited not only the acute lung injury, but also the subsequent pneumonia/pneumonitis (Figure 6D).

View larger version (145K): [in this window] [in a new window]   View larger version (144K): [in this window] [in a new window]   View larger version (146K): [in this window] [in a new window]   View larger version (110K): [in this window] [in a new window]   Figure 6.   (A) A micrograph of lung from autopsy of animal that died 7 d after 150-mg/kg bleomycin injection, showing loss of vascular integrity with hemorrhagic pulmonary edema. Hematoxylin-eosin stain; original magnification: ×200. (B) A micrograph of lung from animal killed 14 d after 150-mg/kg bleomycin injection, showing diffuse consolidation of lung parenchyma with loss of alveolar architecture and increased cellularity. Hematoxylin-eosin stain; original magnification: ×75. (C ) A micrograph of lung from the same animal as in Figure B, showing an increase in the collagen content of the lung parenchyma. Masson's trichrome stain; original magnification: ×200. (D) A micrograph of lung from animal that was treated with the antisense oligonucleotides and killed 14 d after 150-mg-bleomycin injection, showing a significant improvement of the lung lesions by bleomycin. Hematoxylin-eosin stain; original magnification: ×75.

Bleomycin injection significantly increased the pathologic scores (Figure 7). The increase in pathologic scores by bleomycin was significantly inhibited by the antisense oligonucleotides, as shown in Figure 7.

View larger version (28K): [in this window] [in a new window]   Figure 7.   Pathologic scores from animals treated with 150-mg/kg injection alone (BLM alone, 14 d; n = 5, 28 d; n = 5) and animals treated with the p65 antisense oligonucleotides 6 h before and 5 d after 150-mg/kg BLM injection (Antisense + BLM, 14 d; n = 5, 28 d; n = 5). The antisense oligonucleotides significantly reduced the pathologic scores at both 14 and 28 d after bleomycin injection. **p < 0.01 compared with BLM alone.

NF-B Activity

Bleomycin injection increased the p65 levels of nuclear protein from peritoneal lavage cells, and the maximal expression occurred as early as 6 h after the bleomycin injection. The antisense oligonucleotides inhibited the increase by bleomycin, whereas the sense oligonucleotides did not (Figure 8). Bleomycin also increased the p65 levels of nuclear protein from lung tissue homogenates 6 to 24 h after the bleomycin injection. However, pretreatment with the antisense oligonucleotides failed to alter the increase in the lung tissue by bleomycin.

View larger version (56K): [in this window] [in a new window]   Figure 8.   The Western blot analysis of the p65 level of nuclear protein from peritoneal lavage cells of animals 24 h after control saline injection (lane 1), 24 h after 150-mg/kg bleomycin injection alone (lane 2), 24 h after 150-mg/kg bleomycin injection with the sense oligonucleotide treatment (lane 3), 6 h (lane 4 ) and 24 h after 150-mg/kg bleomycin injection with the antisense oligonucleotide treatment (lane 5 ). Bleomycin injection increased the p65 levels, which were inhibited by the antisense oligonucleotides and not by the sense oligonucleotides.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Autopsy of bleomycin-treated mice lost early in the present experiment revealed extensive pulmonary hemorrhagic edema, and lungs from mice killed within 7 d after bleomycin injection showed loss of vascular integrity with pulmonary edema as reported previously (23). The antisense oligonucleotides significantly improved the survival and the histologic changes, preventing the acute lung injury by bleomycin. The histologic and chemical (hydroxyproline) examinations showed that the sequential pneumonia and pulmonary fibrosis were also prevented by the antisense oligonucleotides. Because of the low yield of alveolar macrophages and the fact that the required oligonucleotides would be prohibitively expensive, in the present experiments we could not confirm the inhibition of NF-B activity in alveolar macrophages. However, Western blot analysis of the peritoneal macrophages showed the inhibition of NF-B activity by the antisense oligonucleotides. Further, the antisense oligonucleotides inhibited the elevation of TNF- in BAL cells that is known to be dependent on the NF-B activity of alveolar macrophages (15). Therefore, the present study has suggested that the inhibition of NF-B activity in monocytes/macrophages by the antisense oligonucleotides can induce the inhibition of both acute lung injury and pnuemonitis/ fibrosis by bleomycin.

Although, in the present experiment, bleomycin activated NF-B of lung tissue homogenates as previously reported (17), the antisense oligonucleotides failed to inhibit the activated NF-B. The antisense oligonucleotides are speculated to have been incorporated by alveolar macrophages, especially activated macrophages, and not by other lung tissue cells in the lung. Bleomycin may have activated NF-B in many other cells in the lung tissue, which does not necessarily induce lethal lung injuries and/or fibrosis. The population of alveolar macrophages relative to many other lung cells is too small to produce such changes by the antisense oligonucleotides. In fact, the antisense oligonucleotides did not produce any changes in the behavior, body weight, histology, or hydroxyproline content of control animals that did not receive the bleomycin injection.

It has been shown that there are mice that are sensitive or resistant to a single intravenously administered dose of bleomycin (24, 25), and single injection of bleomycin caused pulmonary fibrosis only in the sensitive strains (24) that we used for the present experiment. The single injection of bleomycin was chosen for the present experiments since clinical bleomycin-induced pulmonary fibrosis can be caused by intravenous injection and also has the advantage of showing more clearly the effectiveness of the antisense oligonucleotides than intra-tracheal or frequent peritoneal administration of bleomycin.

We initially determined that the maximal expression of NF-B in peritoneal macrophages occurred 6 h, and TNF- level in both BALF and serum samples reached a peak 6 h after bleomycin injection. Further, the antisense oligonucleotides used for the present experiment are very stable, and 85 to 90% of the oligo remains in the peripheral blood 24 h after the intravenous injection. Using the FITC-labeled oligonucleotides, we confirmed a large amount of the antisense oligonucleotides in BALF cells 24 h after the bleomycin injection. From these findings, we chose 6 h before bleomycin injection for the time of the antisense oligonucleotide injection. In addition, we chose 5 d after the injection for it since animals died 6 to 9 d after the bleomycin injection.

Cytokines, including TNF- from alveolar monocytes/macrophages, are known to be upregulated within 24 h after bleomycin injection in animal models of pulmonary fibrosis (12). In humans, the plasma level of TNF- was significantly increased as early as 3 h after bleomycin infusion (26). Harrison and colleagues (24), in mice that received intravenous injections of bleomycin (80 mg/kg), have shown that bleomycin treatment rapidly produces extensive pulmonary DNA damage in vivo and that changes in the levels of mRNA encoding pulmonary matrix proteins occur in vivo within 1 d after intravenous bleomycin treatment. For example, a marked increase in the pulmonary level of fibronectin mRNA in the lungs of sensitive mice was clearly evident as early as Day 1 after intravenous administration of bleomycin. In contrast, it has been reported that in hamsters an increase in total pulmonary fibronectin mRNA content began after 4 d and reached a peak 7 d after intratracheal instillation of bleomycin (27). The experiment of FITC-labeled oligonucleotides showed that the antisense oligonucleotides remained unchanged in alveolar and peripheral blood monocytes/macrophages until 24 h after the injection. Therefore, it is possible that the first injection of the antisense oligonucleotides (6 h prior to bleomycin injection) inhibited both the acute lung injury and the late fibrosis by bleomycin. In fact, the single injection of the antisense oligonucleotides 5 d after bleomycin failed to alter the survival or histologic changes by bleomycin.

In the present experiment, contrary to our expectation, the cell number of BALF (mainly macrophages) significantly decreased 0.5 to 7 d after the bleomycin injection and gradually increased thereafter. The late increase in the BALF cell number is consistent with results previously reported, most of which were from experiments with intratracheal administration of bleomycin (28). The antisense oligonucleotides inhibited the initial decrease in the BALF cell number. It is possible that rather than intraalveolar macrophages, the lung interstitial macrophages are strategically located to play an important role in fibrogenesis and other pathologic processes within injured lungs (29). Further, in the present experiments, the increase in the neutrophil number of BALF was not observed, although previous experiments have shown increased neutrophils in bleomycin-induced pulmonary fibrosis, and almost all of these experiments used intratracheal instillation of bleomycin (28). Although we have no evidence, the difference may have been due to the intravenous injection in our experiments. In addition, some reports have shown that neutrophils may not play a role in the development of pulmonary fibrosis because pulmonary fibrosis occurs in a neutrophil-depleted condition (30).

The current therapy for pulmonary fibrosis with corticosteroids and/or immunosuppressants has shown little benefit. Further, in animal models, various attempts designed to inhibit the initiation of bleomycin-induced pulmonary fibrosis have not led to a useful clinical trial because of the adverse effects on lungs and other tissue by the agents tested (13, 14, 31). Similarly, an anti-CD3 monoclonal antibody has been reported to inhibit bleomycin-induced pneumopathy of mouse (32). However, the role of T lymphocytes is uncertain (33), although their recruitment to the lung after bleomycin challenge is well documented (34). Further, Tran and colleagues (35) reported the effective prevention of bleomycin-induced pulmonary fibrosis after adenovirus-mediated transfer of the bacterial bleomycin resistance gene in mouse. However, this approach is limited to bleomycin-induced pneumopathy, and it is not useful for other types of interstitial lung fibrosis. Without using any specific vectors, the oligonucleotides in the present experiment were incorporated or phagocyted into activated alveolar macrophages, which are also observed in the lungs from patients with IPF (1, 2). Therefore, these oligonucleotides may be useful not only for bleomycin-induced pulmonary fibrosis, but also for other types of pulmonary fibrosis, including IPF.