INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Idiopathic pulmonary fibrosis (IPF) is a chronic inflammatory disease characterized by an increase in fibroblast population and excessive accumulation of interstitial collagen in the lung (1). This disrupts gas exchange units and causes progressive respiratory failure (2). Systemic sclerosis (SSc) is a generalized disorder that may be characterized morphologically by the deposition of fibrous connective tissue in many organs (3). Pulmonary fibrosis, which exhibits histologic findings similar to those of IPF (4), is found in 46-81% of patients with SSc (5).

Antifibrinolytic activity has been reported to be reduced in the alveolar fluids of IPF patients (8), and this is partially explained by the observation of Kotani and colleagues (9) that plasminogen activator inhibitor (PAI)-1 antigen levels in bronchoalveolar lavage (BAL) supernatant fluids and PAI-2 antigen levels in BAL cell lysates were higher in patients than in normal subjects, whereas there were no differences in the antigenic levels of urokinase-type plasminogen activator between patients and control subjects. In addition, it has been shown that procoagulant activity is increased in the lungs of patients with IPF (8, 9), especially in patients with a progressive disease (9). Thus, it has been suggested that increased procoagulant activity in the lung implicates its involvement in the development of pulmonary fibrosis by causing excessive local deposition of fibrin, which is known to be important for fibroblast adherence and proliferation (10).

Tissue factor (TF) is a cell membrane-associated protein that serves as the receptor and the essential cofactor for factors VII and VIIa; TF is also the primary cellular initiator of the coagulation protease cascade (11). Therefore, to investigate further the relevance of fibrin turnover abnormalities in the development of lung fibrosis in IPF and interstitial pneumonia associated with SSc (IP-SSc), we examined the localization of TF and fibrin antigens using immunohistochemistry and of TF-mRNA using in situ hybridization. We also analyzed these data in patients with idiopathic bronchiolitis obliterans with organizing pneumonia (BOOP), which represents an acute lung injury pattern (12, 13).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patients

Subjects included ten patients with IPF, four with IP-SSc, and three with idiopathic BOOP. Four patients with SSc fulfilled the American Rheumatism Association preliminary criteria for the diagnosis of SSc (14). All of the patients underwent an open-lung, or video-thoracoscopic, lung biopsy to evaluate interstitial lung diseases. As controls, peripheral samples were taken from unaffected regions of resected lungs from 10 patients with primary lung tumors.

Tissue Samples

(1) Formaldehyde-fixed. One portion of each lung specimen was fixed immediately with 15% formaldehyde solution, then dehydrated and embedded in paraffin. Tissue sections 3 µm thick were mounted on poly-L-lysine-coated slides and then incubated overnight at 60° C before use.

(2) Freshly frozen. Another piece of the biopsy specimen was washed with 0.9% saline solution until the blood effluent was clear, then embedded in Optimal Cutting Temperature (OCT) compound (Tissue-Tek; Miles Laboratories Inc., Elkhat, IN), and stored at 80° C until required for use.

(3) Paraformaldehyde-fixed. Yet another portion of the specimen was fixed with 4% paraformaldehyde dissolved in phosphate-buffered saline (PBS) and immersed in 30% sucrose/PBS overnight at 4° C to reduce freezing artifacts. They were then embedded in OCT compound and stored at 80° C until use. The time from resection of the tissue to fixation was 10 min or less.

Antibodies

Murine monoclonal antibodies were used for the immunohistochemical studies. Anti-human TF antibody from American Diagnostica Inc. (Greenwich, CT) and anti-human surfactant apoprotein (SP-A) antibody from Teijin (Tokyo, Japan) were used at a dilution of 1:10,000. Antifibrin antibody from Cosmo Bio, Inc. (Tokyo, Japan) was used at a dilution of 1:1,000.

Immunohistochemistry

Formaldehyde-fixed tissues were used for the immunohistochemical localization of TF and SP-A. Immunohistochemistry was performed using the streptavidin-biotin method, with a SAB-PO kit (Nichirei Co. Ltd., Tokyo, Japan). Tissue sections were treated with 3% hydrogen peroxide in methanol to eliminate endogenous peroxidase activity. The antibodies on the samples were detected with 3,3'-diaminobenzidine tetrahydrochloride, and the specimens were counterstained with methylgreen (Merck, Darmstadt, Germany). For immunohistochemical controls, normal mouse immunoglobulin G (IgG) was used as a first antibody.

Immunohistochemistry with the anitfibrin antibody was performed using the same procedure (streptavidin-biotin method) on freshly frozen tissue sections. Sections 5 µm thick were cut using a cryostat microtome (Bright, Huntingdon, UK), and dried at room temperature for 30 min. After fixation in acetone for 10 min, the tissue sections were treated with 3% hydrogen peroxide in methanol to eliminate endogenous peroxidase activity.

Total RNA Extraction, TF-cDNA Amplification, and RNA Probe Preparation

Total RNA was purified from human placenta by the acid guanidinium-phenol-chloroform method (15). Ten micrograms of total RNA was reverse-transcribed with Moloney murine leukemia virus reverse transcriptase (SuperScript II; GIBCO BRL, Gaithersburg, MD). The specific TF-cDNA was amplified using the polymerase chain reaction (PCR), with sense primer 5'-CCGCTCGATCTCGCCGCCAACTG-3', and antisense primer 5'-GCTCTGCCCCACTCCTGCCTTTC-3', using published sequence information on human TF-cDNA (16). The reaction profile proceeded as follows: denaturation at 94° C for 30 s, annealing at 55° C for 30 s, and extension at 72° C for 60 s. These steps were carried out for 40 cycles. The PCR product of 755 base pair (bp) was subcloned into pBluescript at the site of EcoRI-EcoRV, and its nucleotide sequence was determined from both strands with T3 and T7 primer, using a dideoxy terminator cycle sequencing kit (Applied Biosystems, Foster City, CA). A digoxigenin-labeled cRNA probe was synthesized in the presence of ATP, GTP, CTP, UTP, and digoxygenin-labeled UTP with T7 or T3 polymerase to generate antisense (cRNA) or sense (mRNA) RNA probes, respectively.

In Situ Hybridization

We examined TF-mRNA expression using in situ hybridization of 4% paraformaldehyde-fixed lung tissue from four patients with IPF, two with IP-SSc, and four normal control subjects, as described previously (17). Briefly, 5-µm-thick sections were cut using a cryostat microtome, applied to poly-L-lysine-coated slides, and heated to 50° C overnight. Sections were permeated with 0.1% Triton-X 100/PBS, 0.2 N HCl, at room temperature, and then with proteinase K (1 µg/ml in PBS) at 37° C. The reaction was terminated by immersion of the slides into 4% paraformaldehyde, after which they were rinsed with glycine. The sections were prehybridized in 50% formamide and twofold standard saline citrate at 50° C. For the hybridization, 1.0 µg digoxigenin-labeled antisense or sense probe was diluted in 1 ml hybridization buffer (18), and 100 µl of the resulting solution was applied to each section. Hybridization was allowed to proceed for 16 h in a humidified chamber at 50° C. After hybridization, the sections were washed with 50% formamide and twofold standard saline citrate. The unhybridized, single-stranded RNA probe was removed using a solution containing RNAse A (20 µg/ml), and then the sections were washed with twofold standard saline citrate. The signals were detected using a RNA detection kit (Boehringer, Mannheim, Germany). At least 10 samples per specimen were analyzed.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Immunohistochemistry

(1) Localization of TF and SP-A. The distribution of TF expression is summarized in Table 1. Staining of TF in normal lungs gave positive results only in the basal layer of the bronchial epithelium (Figure 1). In contrast, lung tissues from patients with IPF exhibited positive staining for TF in the cuboidal cells lining the alveolar septa (Figure 2A). These cells were also stained positively with anti-SP-A antibody (Figure 2B), indicating that they were hyperplastic type II pneumocytes. Cuboidal epithelial cells lining the fibroblastic foci were also stained positively with anti-TF antibody (Figure 2C), but almost none of these cuboidal epithelial cells were stained with anti-SP-A antibody (Figure 2D). Some squamous metaplastic cells (Figure 2E) and some macrophages (Figure 2F) were also stained with anti-TF antibody. There were no fundamental differences between lung tissues from patients with IPF and IP-SSc with respect to the distribution of anti-TF and anti-SP-A labeling (Figure 3).

View larger version (104K): [in this window] [in a new window]   Figure 1.   Normal lung specimen stained with anti-TF antibody. Bronchial epithelial cells are positively stained (methylgreen; original magnification: ×100).

View larger version (111K): [in this window] [in a new window]   View larger version (103K): [in this window] [in a new window]   View larger version (159K): [in this window] [in a new window]   View larger version (117K): [in this window] [in a new window]   View larger version (126K): [in this window] [in a new window]   View larger version (87K): [in this window] [in a new window]   Figure 2.   Lungs from a patient with IPF, stained with anti-TF antibody or anti-SP-A antibody. (A) Cuboidal epithelial cells (arrowheads) lining the alveolar septa are stained with anti-TF antibody (methylgreen; original magnification: ×66). (B) Cuboidal epithelial cells lining the alveolar septa are stained with anti-SP-A antibody (methylgreen; original magnification: ×66). (C ) Cuboidal epithelial cells covering the fibroblastic foci (arrowheads) are also stained with anti-TF antibody (methylgreen; original magnification: ×66). (D) Cuboidal epithelial cells covering the fibroblastic foci (arrowheads) are not stained with anti-SP-A antibody (methylgreen; original magnification: ×66). (E ) Squamous metaplastic cells seen in advanced honeycomb lesions are also stained with anti-TF antibody (methylgreen; original magnification: ×40). (F  ) Some alveolar macrophages (arrows) are stained with anti-TF antibody (methylgreen; original magnification: ×80).

View larger version (104K): [in this window] [in a new window]   View larger version (113K): [in this window] [in a new window]   Figure 3.   Microphotographs of lung tissue from a patient with IP-SSc, stained with anti-TF antibody and anti-SP-A antibody. (A) Cuboidal epithelial cells lining the alveolar septa (arrowheads) and alveolar macrophages (larger arrow) are stained with anti-TF antibody (methylgreen; original magnification: ×100). (B) Cuboidal epithelial cells are also stained with anti-SP-A antibody (methylgreen; original magnification: ×80).

In patients with idiopathic BOOP, the cuboidal epithelial cells covering the thickened alveolar septa and intraluminal fibroblastic plugs were positively stained with the anti-TF antibody (Figure 4A). The cuboidal alveolar epithelial cells were positively stained with the anti-SP-A antibody, whereas almost all of the cells covering the intraluminal fibrotic lesions were negative (Figure 4B).

View larger version (115K): [in this window] [in a new window]   View larger version (104K): [in this window] [in a new window]   Figure 4.   Lung tissue from a patient with idiopathic BOOP labeled with anti-TF antibody or anti-SP-A antibody. (A) Cuboidal epithelial cells lining the septa (arrowheads) and the intraluminal granulation tissues (arrows) are positively stained with anti-TF antibody (methylgreen; original magnification ×66). (B) Cuboidal epithelial cells lining the septa are positively stained with anti-SP-A antibody, but those covering the intraluminal granulation tissues (arrows) are not (methylgreen; original magnification: ×66).

(2) Fibrin deposition. The distribution of fibrin deposition is summarized in Table 2. Fibrin was not detected in normal lung tissue. In contrast, in patients with IPF, IP-SSc, and idiopathic BOOP, fibrin was detected in the areas adjacent to the type II pneumocytes (Figure 5A). Fibrin was also detected in the intraluminal fibroblastic plugs seen in idiopathic BOOP patients (Figure 5B), which appeared to coincide with the distribution of TF.

View larger version (93K): [in this window] [in a new window]   View larger version (104K): [in this window] [in a new window]   Figure 5.   Lung tissues from patients with IPF or idiopathic BOOP, stained with antifibrin antibody. (A) The fibrin antigen is observed in the type II pneumocytes and the adjacent area in a patient with IPF (arrowheads) (methylgreen; original magnification: ×66). AS = alveolar space; AW-alveolar wall. (B) In a patient with idiopathic BOOP, the fibrin antigen is also observed in the cuboidal epithelial cells lining the granulation tissues (arrowheads) (methylgreen; original magnification: ×33).

TF-mRNA Detection by In Situ Hybridization

To confirm that TF is produced by the positively stained cells, we applied the in situ hybridization technique to study the localization of cells expressing mRNA for TF. As demonstrated in Figure 6A, there were no positive signals in sections from normal lungs. In contrast, TF-mRNA in lung tissue from patients with IPF was localized in hyperplastic type II pneumocytes and some of the alveolar macrophages (Figure 6B). The absence of nonspecific staining (sense probe) is shown in Figure 6C.

View larger version (87K): [in this window] [in a new window]   View larger version (113K): [in this window] [in a new window]   View larger version (86K): [in this window] [in a new window]   Figure 6.   In situ hybridization detection of TF-mRNA. (A) In the normal lung, TF-mRNA is not observed (Nitro blue tetrazolium; original magnification: ×50). (B ) In tissue from a patient with IPF, TF-mRNA is observed in type II pneumocytes (arrowheads) and some macrophages (arrows) (Nitro blue tetrazolium; original magnification: ×50). (C  ) Nonspecific staining is not observed in tissue hybridized with sense probe (Nitro blue tetrazolium; original magnification: ×50).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Stimulated alveolar macrophages are known to be able to produce and secrete TF (19); therefore, it has been suggested that macrophages are a source of TF in the lung (20). However, the study presented here demonstrated that TF antigens were expressed mainly in the type II pneumocytes covering the affected alveolar septa and the fibroblastic foci in patients with IPF and IP-SSc. We also showed that fibrin is deposited in the type II pneumocyte layer and the adjacent areas. These results suggest strongly that TF, as detected by immunohistochemistry, causes activation of the extrinsic coagulation pathway, resulting in local fibrin deposition in these patients. Our study also demonstrated, using in situ hybridization, that TF-mRNA was expressed mainly in the type II pneumocytes. These results indicate that not only macrophages, but also type II pneumocytes, are major sources of TF production in patients with IPF and IP-SSc.

Some differences between IPF and IP-SSc have been reported, providing evidence that these are two distinct conditions. For example, interleukin (IL)-8 levels in BAL fluids were significantly higher in patients with IPF than in those with IP-SSc (21). The outcome is also different between those with IPF and IP-SSc (22). We therefore compared the distribution of TF and fibrin deposition in the lung specimens from patients with IPF and IP-SSc, and we found no differences. We also examined the distribution of TF and fibrin deposition in the lungs of patients with idiopathic BOOP, an acute lung injury pattern (12, 13), and found that the TF antigen was detected mainly in hyperplastic type II pneumocytes. Taken together, these results indicate that production of TF is not a disease-specific phenomenon, but rather a common physiologic reaction that accompanies alveolar epithelial damage and contributes to the tissue repair of microinjuries in the alveolar septa.

In our study, cuboidal epithelial cells lining the fibroblastic foci and the granulation tissue were not stained with SP-A, although epithelial cells lining the affected alveolar septa showed positive staining with the antibody. Ultrastructural observations revealed that granulation tissue-lining cells in BOOP were lacking in cytoplasmic lamellar bodies, which are known to be a characteristic of typical type II pneumocytes. In contrast, alveolar-lining cells did exhibit the lamellar bodies (23). It remains to be learned whether the cell origin of the granulation tissue-lining and fibroblastic foci-lining cells differs from that of type II pneumocytes, or whether the difference in characteristics between these cells depends on the degree of cell maturation.

In our study, normal lung parenchyma was negative for TF staining of type II pneumocytes, which is consistent with the idea that TF is not normally expressed in endothelial cells or macrophages (24). A variety of stimulants, however, such as IL-1, tumor necrosis factor- (TNF-), bacterial endotoxin (25), and measles virus infection (26), have been reported to induce TF expression in these cells in vitro. Our results showing TF expression in type II pneumocytes in interstitial pneumonia, therefore, indicate that type II pneumocytes are activated under these conditions, although the precise mechanisms remain largely unknown.

Evidence has shown recently that pulmonary type II pneumocytes have the capacity to secrete several inflammatory cytokines, such as IL-6 and IL-8 (27), to express intercellular adhesion molecule-1 (28), and to produce monocyte chemoattractant protein-1 (29), transforming growth factor- (30), TNF- (31), and platelet-derived growth factor (32). Transforming growth factor- is known to stimulate fibroblasts to synthesize collagen, fibronectin, proteoglycans, and other proteins of the extracellular matrix (33). Platelet-derived growth factor is also known to be a potent chemoattractant and mitogen for fibroblasts and a stimulator of collagen synthesis by these cells (34). Furthermore, Matsui and colleagues reported that rat type II pneumocytes produced type I collagen after viral infection (35). Taken together, these findings suggest that type II pneumocytes are largely involved, not only in tissue repair, but also in the fibrotic process under certain pathological conditions via either a paracrine or an autocrine mechanism. Elucidating the mechanisms activating these cells is an important aspect that requires further investigation.