Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Pulmonary fibrosis is the end stage of a heterogeneous group of disorders of known and unknown etiology in which the appearance of excessive amounts of collagen deposited in a disorganized manner results in the loss of lung function and the premature death of affected individuals. In humans there is biochemical evidence of both increased collagen deposition (1) and synthesis (2) along with decreased collagenolytic activity (5). A number of animal models have been developed to investigate the pathogenesis of pulmonary fibrosis, the most common of which is that induced by the intratracheal instillation of bleomycin. In these models, increased collagen synthesis and decreased degradation contribute to enhanced deposition of collagen in the lung (6).

Current theories for the pathogenesis of pulmonary fibrosis are based largely on the hypothesis that in response to injury, inflammatory and immune cells enter the lung from the circulation and together with activated resident cells, release mediators that induce fibroblasts to proliferate or produce excess collagen (9). We have recently identified a potentially important role for endothelin-1 (Et-1) in the pathogenesis of pulmonary fibrosis. Et-1 is a 21-amino-acid peptide which is synthesized and released by many cell types, including endothelial cells (10, 11), epithelial cells (12), alveolar macrophages (16), and fibroblasts (17). Et-1 was originally identified as a potent vasoconstrictor (18, 19), but it is now recognized to be a pleuripotent mediator with both chemoattractant and mitogenic properties (20) as well as a stimulant of fibroblast collagen synthesis (21, 22). In patients with pulmonary fibrosis associated with systemic sclerosis (SSc) we have demonstrated that the lung epithelial lining fluid, sampled by bronchoalveolar lavage (BAL), contains polypeptide mediators capable of stimulating fibroblast proliferation; whereas fluid from normal volunteers had no effect (23). Using neutralizing antibodies and receptor antagonists we have shown that Et-1 is responsible for almost half of the lavage fluid-induced in vitro fibroblast proliferation (24). We have also demonstrated that levels of Et-1 in the lavage fluid of these patients are increased up to fivefold compared with those of controls, and that higher levels were observed in patients without overt symptoms of pulmonary fibrosis than those with clinical evidence of fibrosis (24). Others have localized Et-1 and Et-1 mRNA to macrophages, epithelial cells, and endothelial cells in the lungs of patients with pulmonary fibrosis (14, 25, 26). These studies provide strong evidence to suggest that Et-1 may play an important role in the pathogenesis of pulmonary fibrosis.

Little information is present on the sequential changes in Et-1 levels and localization during the development of pulmonary fibrosis. In this study we have characterized the temporal changes in lung Et-1 compared with those for collagen deposition during the development of bleomycin-induced pulmonary fibrosis in rats. Immunohistochemical techniques were also employed to examine the localization of Et-1 in normal and fibrotic lungs.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Animals

Lewis rats, aged 7-8 wk and weighing 190-220 g, were anesthetized by intramuscular injection of 0.75-1.0 ml/kg body weight Hypnorm (fentanyl citrate 0.315 mg/ml and fluanisone 10 mg/ml; Janssen Pharmaceutical, High Wycombe, UK). Doses of 1.5, 3.0, 4.5, and 6.0 mg/kg body weight of Bleomycin sulphate (Lundbeck, Luton, UK) were administered intratracheally in 0.3 ml sterile saline. Control animals received 0.3 ml saline alone. Briefly, the trachea was intubated with a 19-gauge canula and the solution introduced using a 1-ml syringe. A small volume of air (0.5 ml) was then quickly injected to flush the airways. Groups of eight rats, six for biochemical analysis and two for immunohistochemical studies, were killed 3, 7, 14, and 21 d after bleomycin instillation by an overdose of pentobarbitone. For immunohistochemical studies, lungs were fixed by intratracheal instillation of freshly prepared 4% paraformaldehyde in phosphate-buffered saline (PBS) containing 100 U/ml heparin at a pressure of 25 cm H₂O. The trachea was ligated and the thoracic contents removed en bloc. After 4 h immersion in fixative, tissues were transferred to 15% sucrose in PBS, prior to dehydration and embedding in paraffin wax. For measurement of collagen content, the vasculature was perfused with 5 ml PBS containing 100 U/ml heparin. Lungs were removed, weighed, and snap-frozen in liquid N₂.

Collagen Measurement

Lung collagen content was assessed by measuring hydroxyproline in proteins, using a high-pressure liquid chromatography (HPLC) method developed in this laboratory (27). Briefly, 100-mg aliquots of powdered lung tissue (obtained by crushing the tissue while frozen at 70°C) were hydrolyzed in 2 ml of 6 M HCl at 110°C for 16 h. Hydrolysates were mixed with 30 mg activated charcoal and filtered (Millipore, type DA, pore size 0.65 µm). A 100-µl aliquot of a 1-in-10 dilution of filtered hydrolysate was dried using a centrifugal vacuum concentrator.

Hydroxyproline was isolated and measured by reverse-phase HPLC after derivatization with 7-chloro-4-nitrobenz-2-oxa-1,3-diazole (NBD-Cl; Sigma, Poole, UK), as described previously (27). Briefly, dried hydrolysates were reconstituted in 100 µl of H₂O, buffered with 100 µl of 0.4 M potassium tetraborate (Sigma) and reacted with 100 µl of 12 mM NBD-Cl in methanol. Samples were incubated at 37°C for 20 min in the dark. The reaction was stopped by addition of 50 µl of 1.5 M hydrochloric acid; finally, 150 µl of 167 mM sodium acetate (Fisons, Loughborough, UK) in acetonitrile (26% vol/vol) was added. Samples were filtered (Millipore type GV, pore size 0.22 µm) and a 100-µl aliquot was loaded onto the column and eluted with an acetonitrile gradient as previously described (27). The hydroxyproline content in each sample was determined by comparing peak areas of samples from the chromatogram with those generated from standard solutions. The total amount of collagen in each lung was calculated, assuming that lung collagen contains 12.2% w/w hydroxyproline (28), and expressed as mg collagen/lung.

Measurement of Immunoreactive Et-1

Et-1 was extracted from lung tissue using the method described by Matsumoto and colleagues (29). Briefly, powdered lung tissue was homogenized in 10 vol of ice-cold 1 M acetic acid containing 10 µg/ml pepstatin, and boiled immediately for 10 min. The homogenates were centrifuged at 25,000 × g for 30 min at 4°C and Et-1 was extracted from the supernatant by acidification with an equal volume of 0.1% trifluoroacetic acid (TFA) and applied to a Sep Pak C₁₈ reverse phase minicolumn (Waters Chromatography Division, Millipore Corp., Milford, MA) prewashed with 2 ml of methanol, distilled water, and 0.1% TFA. The unbound material was eluted with 0.1% TFA and Et-1 eluted with 60% acetonitrile/0.1% TFA. The eluent was freeze-dried and reconstituted in 0.02 M borate buffer. Immunoreactive Et-1 was measured by radioimmunoassay (RIA) using recombinant human Et-1 as a standard (Amersham, Amersham, UK). Results were expressed as either pg Et-1/mg of lung tissue or as ng/lung.

The extraction of Et-1 from lung tissue was evaluated using 3-[¹²⁵I]iodotyrosyl-endothelin-1 (Amersham) and recoveries of approximately 95% were generally obtained (94.6 ± 3.8%, n = 4).

As well as Et-1, the antibodies used for both RIA and immunohistochemistry also cross-react with the Et-1 precursor, Big Et-1. Together these will be referred to as Et-1. The antibody used in the RIA also cross-reacts with endothelin-2 (Et-2). However, it has been shown that rat lung contains essentially no Et-2 (29).

Quantitative Histology

Tissue sections were prepared and stained with Harris's hematoxylin and eosin (H&E), and quantitative assessment of the total area of interstitial fibrosis was performed by image analysis using Optilab^TM software (ME Electronics Ltd, Reading, UK). In brief, sections were examined at a magnification of 10 using an Olympus BX50 microscope, and the field was divided into 1,000-µm² calibrated areas. Lung parenchyma comprised the whole of the image in each field, with less than 15% of the total area representing conducting airway lumen. The total area of interstitial fibrosis was computed as a function of the calibrated 1,000-µm² field area. In some instances, image analysis was facilitated by the use of false color imaging (blue/ green), which enhanced the H&E demonstration. Between 30-35 fields were examined per section for six sections.

Immunohistochemistry for Et-1

Paraffin sections (5 µm) were dewaxed in xylene, rehydrated through decreasing concentrations of ethanol, and washed in PBS. Tissue endogenous peroxidase activity was blocked by incubating the sections with 3% hydrogen peroxide for 3 min. Sections were washed 3 times for 5 min in PBS and incubated in a 1/50 solution of normal goat serum (DAKO, High Wycombe, UK) for 20 min. The serum was drained and blot-dried around the tissue. Sections were incubated overnight at 4°C with a 1/500 dilution in PBS of a high-affinity (K = 4.4 × 10¹⁰ l/M) mouse monoclonal antibody (obtained commercially from Department of Immunology, University College London, UK) with equal cross-reactivity to Et-1 and Big Et-1, as described previously (30). After three 5 min washes in PBS, sections were incubated with biotinylated goat antimouse IgG antiserum (1/100; DAKO) for 60 min at room temperature, washed in three 5 min changes of PBS, and incubated with a streptavidin/ peroxidase complex (1/100; DAKO) for a further 60 min. Labeled Et-1 was visualized by incubating sections in a solution of 600 µg/ml 3,3'-diaminobenzidine (DAB; Sigma) and 0.03% hydrogen peroxide (H₂O₂) for 3 min. Sections were washed in water, counterstained with hematoxylin, dehydrated, and mounted with DPX mountant (BDH Laboratory Supplies, Poole, UK). As controls, sections were incubated overnight at 4°C either with normal goat serum or murine IgG1 instead of primary antisera.

Quantitation of Et-1 Immunostaining

Numbers of Et-1 immunopositive cells within randomly selected areas of parenchymal fibrosis were determined using a 300-µm² graticule divided into 100-µm² fields. Because the fibrosis was focal and not uniform throughout the tissue, areas of fibrosis were selected and measured if more than 75% of the graticule field was occupied by a fibrotic lesion in the bleomycin-treated animals. Typically, 10-12 fields were examined per section for six sections. Comparative counts were performed on anatomically matched areas in tissues from saline-treated animals. Results were expressed as the mean number of immunopositive cells per 300-µm² tissue.

To quantify diffuse Et-1 immunoreactivity in parenchymal, microvascular, and conducting airway structures, computer-assisted image analysis densitometry was performed using Optilab^TM densitometry software. This technique is a modification of the methodology of Poston and Gall (31). In brief, the DAB reaction product was quantitated by hue, saturation, and intensity in terms of pixel number. A 30-50-µm² area of lung parenchyma and a 200- 300-µm² area of vascular and conducting airway structures (endothelium to adventitia for blood vessels, and luminal surface of the epithelium to the base of the submucosa for conducting airways) were selected in fibrotic areas and in anatomically matched controls; following image capture, background color was excluded by image thresholding. The number of pixels in the total stained area was measured and expressed as a percentage of the total area examined. Between five and six fields were examined from eight to ten anatomically matched structures per section for six sections. For all slides, the baseline was determined by analysis of control sections (without primary antibody).

Statistical Analysis

Statistical evaluation was performed using an unpaired t test or by one-way analysis of variance for multiple comparisons using the Newman-Keuls procedure on collagen and Et-1 measurements, and a Mann-Whitney U test on histological data. A P value of < 0.05 was considered significant. All data are shown as mean ± standard error of mean (SEM). For biochemical measurements the SEM represents animal-to-animal variability and for histologic data field-to-field variability.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

In initial experiments, bleomycin was instilled into the lungs of rats at four concentrations (1.5-6.0 mg/kg body weight) to determine the optimal dose for the development of pulmonary fibrosis. At 6.0 mg/kg there was high mortality and total lung collagen was not measured. All lower doses of bleomycin induced a significant increase in total lung collagen after 21 d compared with instillation of saline alone or in untreated animals (24.1 ± 3.0 to 25.3 ± 1.7, compared with control 15.8 ± 1.3; P < 0.025). There was no significant difference in total lung collagen between any of the bleomycin-treated groups. Due to increased mortality and loss of body weight in the higher-dose groups, 1.5 mg/kg was chosen for all further experiments.

Following intratracheal instillation of 1.5 mg/kg bleomycin, wet lung weight increased by 27% at 3 d (1.22 ± 0.04 g, compared with control 0.96 ± 0.02) and continued to rise to a maximum of almost double that of control animals by 7 d (1.72 ± 0.09, compared with control 0.90 ± 0.02). Thereafter, lung weight remained constant up to 21 d (1.70 ± 0.03, compared with control 0.94 ± 0.05). There was no significant change in lung weights of control animals at any time examined.

Collagen Measurement

Figure 1 shows the changes in total lung collagen 3, 7, 14, and 21 d after instillation of bleomycin. There was no change in collagen content at 3 d but after 7 d collagen content had increased by about 25% (22.9 ± 1.4 mg, compared with 18.5 ± 1.0 for controls; P < 0.05). The collagen content continued to increase, and by 21 d levels were almost double control values (38.1 ± 3.2 mg, compared with 21.3 ± 1.6 mg for controls; P < 0.001; Figure 1). There was no significant change in total lung collagen for controls at any of the times examined.

View larger version (24K): [in this window] [in a new window]   Figure 1.   Change in lung collagen content at various times following bleomycin instillation. Collagen content was measured at the times indicated after intratracheal instillation of bleomycin (1.5 mg/kg body weight) or saline alone. Each value represents the mean ± SEM for six animals. Significant differences between bleomycin-treated and untreated animals are indicated. *P < 0.05, **P < 0.005, #P < 0.001.

Endothelin Measurement

Total Et-1 levels were measured in lung tissue from control and bleomycin-treated animals (Figure 2). There was a significant increase in total Et-1 levels in bleomycin-treated animals by 3 d with values double those for control animals (9.40 ± 1.17 ng, compared with 4.48 ± 0.65 ng for controls; P < 0.01). Et-1 levels were further increased at 7 d, with values three times those of controls (16.58 ± 1.79 ng, compared with 4.82 ± 0.75 ng for controls; P < 0.001). At 3 and 7 d after bleomycin treatment, these values represent a 65% (7.68 ± 0.92 pg/mg tissue, compared with 4.65 ± 0.68 pg/mg for controls; P < 0.025) and 85% (9.89 ± 1.33 pg/mg, compared with 5.35 ± 0.82 pg/mg for controls; P < 0.025) increase in Et-1 concentration, respectively, above controls. The total Et-1 values for bleomycin-treated animals were then unchanged up to 21 d.

View larger version (20K): [in this window] [in a new window]   Figure 2.   Changes in Et-1 content at various times following instillation of bleomycin or saline alone. Each value represents the mean ± SEM for six animals. Significant differences between bleomycin-treated and untreated animals are indicated. *P < 0.01, **P < 0.001.

Histopathology

Control tissues from animals instilled with saline did not demonstrate any pathologic changes at any of the time points examined. Administration of bleomycin was associated with a diffuse inflammatory-cell influx at Day 3 with areas of microvascular leakage and early hyperplastic changes in Type II pneumocytes in regions of inflammatory-cell influx. Early alveolar septal edema was apparent in some areas. By Day 7, interstitial fibrosis with multiple inflammatory foci, containing predominantly macrophages, was evident. Edema was clearly present within alveolar septa, and a low-grade focal "honeycombing" reaction was evident. In addition, there were many examples of retractive edema, especially of the bronchiolar epithelium and microvascular adventitia. By Day 14, tissues showed a more mature regional interstitial fibrosis with a pronounced macrophage presence and focal lymphocytosis and lymphoid expansion. Focal alveolar reepithelialization was present with extensive remodeling of the alveolar unit. Early honeycombing was also observed. Microvascular leakage was noted around capillary beds, sometimes associated with a focal hemorrhagic reaction. By Day 21, there was evidence of focal condensation of extracellular matrix. Macrophage and lymphocyte margination was marked, particularly within the periphery of fibrotic areas. Regional re-epithelialization of the alveolar septa was pronounced (data not shown).

Computer-assisted image analysis on H&E sections of lung tissue from bleomycin-treated animals demonstrated a progressive increase in parenchymal fibrosis between Days 3 and 14 (Figure 3). There was no further increase at 21 d. This quantitative histologic assessment correlates well with the biochemical analysis (Figure 1).

View larger version (11K): [in this window] [in a new window]   Figure 3.   Percentage cross-sectional area of lung parenchyma containing fibrotic lesions from bleomycin-treated animals. Each value represents the mean ± SEM for n = 30-35.

Histological Assessment of Et-1-Immunostained Lung Sections

Et-1 immunoreactivity in tissues from control animals showed a diffuse bronchial epithelial staining, associated with both epithelial cells and occasional intra-epithelial leucocytes (Figure 4a). Diffuse immunoreactivity was also present in peribronchiolar and perivascular connective tissue as well as the microvascular endothelium. There were no differences in the staining patterns of control tissues at any time point examined.

View larger version (144K): [in this window] [in a new window]   Figure 4.   Representative sections of lung tissue from control and bleomycin-treated rats stained immunohistochemically for Et-1. (a) Normal section showing peribronchiolar region with staining of the bronchiolar epithelium (large arrow), Type II pneumocytes (small arrows), and occasional alveolar macrophages (arrowhead). (b) Section taken 7 d after instillation of bleomycin showing marked bronchiolar epithelial cell staining (large arrow). Within the alveoli, Et-1 imunoreactivity is largely confined to Type II pneumocytes (small arrows) and macrophages (arrowheads and inset). (c) Section taken 7 d after bleomycin treatment showing immunostaining of venule endothelium and diffuse staining of the media. Note strong Type II pneumocyte (arrows and inset) and inflammatory cell staining in the associated alveolar region. (d) Peribronchiolar alveolar fibrotic lesion at Day 14 demonstrating a marked inflammatory response. Note the Et-1-positive leukocyte infiltration of the bronchiolar epithelium (arrows). (e) Inflammatory cell activity (arrows) within a more mature fibrotic lesion at 14 d after bleomycin demonstrating diffuse Et-1 staining of extracellular matrix. (  f  ) Peri-arteriolar region 14 d after bleomycin with pronounced staining of arteriole adventitia and staining of the alveolar epithelial barrier (arrow). There is diffuse Et-1 staining of connective tissue. (g, h) No primary serum control of a 14 d post-bleomycin vascular or peribronchiolar alveolar fibrotic lesion, respectively. Scale bar = 70 µm.

Immunostaining of lung tissue from bleomycin-treated animals at Days 3 and 7 were essentially the same. There was marked immunostaining of the arteriolar endothelium and venular intima and media. Bronchiolar epithelium demonstrated a dense focal staining which was enhanced by the presence of Et-1 immunopositive intra-epithelial leukocytes, probably macrophages, located between the apical epithelial border and the basement membrane. Extensive Et-1 immunoreactivity was present in the alveolar bed, with marked staining of the alveolar epithelium including Type II pneumocytes. Strongly stained Et-1-positive macrophages were also evident, localized within alveolar spaces and associated with the septal epithelium. Peribronchiolar and perivascular connective tissue also demonstrated diffuse Et-1 immunostaining. Seven days after bleomycin treatment there were also multiple inflammatory cell foci, containing intensely stained macrophages and associated with hyperplastic Type II pneumocytes (Figures 4b, 4c). Et-1 immunoreactivity was also present in bronchiolar epithelium (Figure 4b) and in intra-epithelial leukocytes at 7 and 14 d (Figure 4d). Vascular staining was also evident (Figure 4c). At 14 d, fibrotic lesions demonstrated strong Et-1 immunoreactivity (Figure 5). In these regions there was pronounced staining of septal edema and many intensely stained macrophages, both central and peripheral to the area of fibrosis (Figures 4d, 4e and Figure 5). Et-1 immunostaining of microvascular structures was also clearly evident (Figure 4f), together with Et-1 staining of areas of microvascular leakage. A similar pattern of staining was observed at 21 d with the addition of Et-1 associated with alveolar re-epithelialization (data not shown). Omission of the primary antibody resulted in no staining (Figures 4g, 4h). The same was true when IgG1 was used as a control (not shown).

View larger version (183K): [in this window] [in a new window]   Figure 5.   Low-power micrograph of a 14 d post-bleomycin-treated lung showing intense Et-1 immunoreactivity in areas of fibrosis (F ). Scale bar = 300 µm.

The number of Et-1 immunopositive lung cells in bleomycin-treated animals was significantly elevated by Day 3 compared with saline controls, and continued to increase to Day 14 (Figure 6). There was no significant increase in cell number between 14 and 21 d.

View larger version (17K): [in this window] [in a new window]   Figure 6.   The number of Et-1 immunopositive lung cells in control and bleomycin-treated animals. Each bar represents the number of cells positively staining for Et-1 within a 300 µm2 area of lung tissue. Each value represents the mean ± SEM for n = 10-12. Statistical differences between bleomycin-treated and untreated animals are indicated. *P < 0.01.

Quantification of Et-1 staining demonstrated a continual involvement of the conducting airway epithelium and microvasculature in fibrotic areas throughout the time course of this study (Table 1). The proportion of Et-1 staining was increased 5- to 6-fold in bronchiolar epithelium at all times compared with their respective controls, although this was augmented by the presence of intra-epithelial leukocytes. For microvasculature, both arteriolar and venular staining was elevated 2- to 4-fold at Days 3 and 7. The fall in arteriolar staining by Day 21 was a consistent finding although many vessels at 21 d displayed disorganized connective tissue and compression of normal anatomy, often due to associated matrix hyperplasia, which may have biased the densitometric assessment.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

As far as we are aware, this is the first study to examine the sequential changes in Et-1 levels and localization during the development of bleomycin-induced pulmonary fibrosis in an animal model. Control lung Et-1 values were similar to levels previously obtained by Matsumoto and colleagues in rats (29). Et-1 levels increased rapidly following administration of bleomycin and reached values 3 times those of the controls by 7 d. Furthermore, this increase preceded the onset of fibrosis, as determined by a rise in lung collagen content, and co-localized with developing fibrotic lesions. This is consistent with human studies of pulmonary fibrosis which have demonstrated increased amounts of Et-1 and Et-1 mRNA in the lungs of patients with SSc (24, 32) and cryptogenic fibrosing alveolitis (26), as well as elevated levels in the plasma of these patients (26, 33). Et-1 is a potent stimulant of fibroblast chemotaxis (20), proliferation (20, 21, 24, 37), and collagen synthesis (21, 22). We have previously demonstrated that lung epithelial lining fluid obtained from patients with pulmonary fibrosis associated with SSc by BAL contains greatly increased levels of Et-1 compared with normal control subjects and is responsible for approximately half of the fibroblast mitogenic activity of this fluid (24). Together, these findings provide strong evidence for a role for Et-1 in the pathogenesis of pulmonary fibrosis.

The most likely source of the increased levels of Et-1 is an elevation of its synthesis by resident lung cells and by inflammatory cells entering the lung following administration of bleomycin. Et-1 is known to be produced by several cell types resident in the lung, including endothelial (10, 38, 39), epithelial (12, 14, 15), and endocrine cells (13); alveolar macrophages (16); and fibroblasts (17). Furthermore, endothelial cells have been induced to release increased amounts of Et-1 in response to injury (40), exposure to thrombin (41, 42), and stimulation with mediators, including transforming growth factor beta, tumor necrosis factor alpha, and interleukin (IL)-1 and IL-8, which are present in increased amounts in the lungs of patients with pulmonary fibrosis (43). In the present study, immunohistochemical analysis of normal rat lung demonstrated Et-1 immunoreactivity localized to rat bronchiolar epithelium, type II epithelial cells, macrophages, endothelial cells, and the intimal and medial region of vessel walls. This finding supports a previous study demonstrating immunoreactive Et-1 in rat airway epithelium (48), and is consistent with the findings of Et-1 localization in other species, including humans (14, 25, 26). In animals treated with bleomycin, there appeared to be an increase in staining intensity for Et-1 in all positively stained cell types, and quantitative analysis demonstrated an increase in the number of resident and inflammatory cells immunoreactive for this protein. The macrophage was the dominant cell type associated with Et-1 immunoreactivity, especially in developing fibrotic areas where macrophages formed foci within the lesion. Bronchioles, arterioles, and venules showed enhanced Et-1 immunoreactivity. Furthermore, re-epithelialization observed at 21 d was associated with Et-1 staining, which may suggest a possible role of Et-1 in epithelial-cell proliferation (49). The increased expression of Et-1 in lungs of bleomycin-treated animals is consistent with reports of increased expression of Et-1 in airway epithelia of human lungs from patients with pulmonary fibrosis (14). Et-1 mRNA expression has also been demonstrated in human fetal and adult lung, and has been shown in healthy and diseased lung to be expressed by the same cells that demonstrated Et-1 immunoreactivity (14). Furthermore, immunoreactive proEt-1 has been shown to be co-localized with immunoreactive mature Et-1 in the airway epithelia of the adult human lung (25). Together, these data provide evidence suggesting increased local synthesis of Et-1 by resident and inflammatory cells during the development of pulmonary fibrosis. Future studies of mRNA expression in cells within fibrotic lesions will help distinguish Et-1 synthesis from deposition and pinpoint the cellular sources.

Although the most likely cause of the increase in Et-1 is local synthesis, there are other possible mechanisms. For example, elevated Et-1 levels may occur as a result of decreased degradation by proteolytic enzymes (50). Alternatively, it has been proposed that normal lung may function to clear Et-1 from the circulation (51). This may occur by trapping of Et-1 on the surface of endothelial cells via the EtB receptor (52) or following microvascular leakage. Both subtypes of the Et-1 receptor (EtA and EtB) have been demonstrated in the airways and pulmonary vasculature of various species, including humans (52). Lung Et-1 could therefore increase if there was increased microvascular leakage and binding of Et-1 from the circulation. Edema and microvascular leakage were both associated with marked staining in our study following administration of bleomycin. However, our previous data obtained in patients with SSc demonstrated that Et-1 levels in the BAL fluid were elevated when expressed with respect to albumin, a recognized marker of vascular leakage, suggesting that an influx of Et-1 from the circulation is not the major contributor to the increased levels in the lung (24).

The increased levels of ET-1, as measured by RIA, were determined using an antibody which cross-reacts with Big Et-1 and Et-2. It has previously been shown that normal rat lung contains essentially no Et-2 (29). However, it is possible that this might change in fibrosis. Therefore, we cannot rule out the possibility that some of the increase we found in Et levels may have been due to Big Et-1 and Et-2. However, such possibilities do not detract from the principal findings regarding the involvement of Et in fibrosis.

In summary, we have characterized the temporal changes in lung Et-1 during the development of bleomycin-induced pulmonary fibrosis in rats. Total lung Et-1 levels increased 3-fold compared with untreated controls, and this increase occurred prior to the elevation in lung collagen content. Immunohistochemical staining of normal lung tissue demonstrated that Et-1 was associated predominantly with macrophages, bronchial epithelial cells, type II epithelial cells, and endothelial cells. Fibrotic regions contained large numbers of intensely stained macrophages and more- intensely stained alveolar epithelial cells. Unaffected areas of the lung showed a normal pattern of staining, which is consistent with a potential role for Et-1 in the development of pulmonary fibrosis. The association of increased Et-1 levels prior to the elevation in lung collagen content and increased staining for Et-1 with developing fibrotic regions provides strong evidence in support of a role for Et-1 in the pathogenesis of pulmonary fibrosis. From the observation of similar changes in the current model to those observed in patients with pulmonary fibrosis, we conclude that bleomycin-induced pulmonary fibrosis is a good model for investigating the role of Et-1 in the pathogenesis of pulmonary fibrosis. Future studies will utilize endothelin receptor antagonists to establish the importance of Et-1 in the pathogenesis of pulmonary fibrosis and to assess their potential efficacy in the treatment of this disease.