HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: 285/3/L527    most recent 00298.2002v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (38) Citing Articles via Google Scholar Google Scholar Articles by Xu, Y. D. Articles by Khalil, N. Search for Related Content PubMed PubMed Citation Articles by Xu, Y. D. Articles by Khalil, N.

EDITORIAL FOCUS

Release of biologically active TGF-1 by alveolar epithelial cells results in pulmonary fibrosis

¹Division of Respiratory Medicine, Department of Medicine, The University of British Columbia, and Vancouver Hospital; ²Department of Surgery and Vancouver Hospital, Vancouver V6H 3Z6; ³Department of Pathology, British Columbia Cancer Agency, Vancouver V5Z 4E6, British Columbia; and ⁴School of Medicine, University of Miami, Miami, Florida 33136

Submitted 29 August 2002 ; accepted in final form 14 February 2003

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION DISCLOSURES REFERENCES

 
Idiopathic pulmonary fibrosis (IPF) is a progressive fatal fibrotic lung disease. Transforming growth factor (TGF)-1 is present in a biologically active conformation in the epithelial cells lining lesions with advanced IPF. To determine the role of aberrant expression of biologically active TGF-1 by alveolar epithelial cells (AECs), the AECs of explanted normal rat lungs were transfected with the TGF-1 gene using the retrovirus pMX-L-s223,225-TGF-1. In situ hybridization using a digoxigenin-labeled cDNA of the puromycin resistance gene contained in the pMX demonstrated that pMX-L-s233,225-TGF-1 was selectively transfected into AECs of the explants. Conditioned media overlying explants obtained 7 days after being treated with pMX-L-s223,225-TGF-1 contained 14.5 ± 3.15 pg/ml of active TGF-1. With the use of Masson's trichrome staining of explant sections obtained 14 days after transfection, there were lesions similar to those in IPF, characterized by type II AEC hyperplasia, interstitial thickening, extensive increase in interstitial and subepithelial collagen, an increase in the number of fibroblasts, and areas resembling fibroblast buds. Collagens I, III, IV, and V and fibronectin were increased in explants treated with pMX-L-s223,225-TGF-1. The findings in the current study suggest that IPF may be a disorder of epithelial cells and not inflammatory cells.

transforming growth factor-1; idiopathic pulmonary fibrosis

In contrast to interstitial lung diseases, like sarcoidosis or hypersensitivity pneumonitis, lung biopsies from patients with IPF demonstrate minimal inflammation, and little evidence exists that lesions of IPF are preceded by inflammation (37, 38). Furthermore, IPF does not respond to the current standard therapy of immunosuppressive agents such as corticosteroids, cyclophosphamide, and azathioprine (37, 38). Collectively, these observations provide compelling evidence that inflammatory cells in IPF may not be important in the pathogenesis of IPF, but the overproduction of TGF-1 by epithelial cells may be critical to the fibrosis seen in IPF. The mechanism by which epithelial cells are induced to release TGF-1 is not known. In this paper, sliced lung explants from normal rats, which were relatively free of inflammatory cells, were cultured in the presence of the retrovirus pMX. The AECs of the explants were successfully transfected with the cDNA of TGF-1, carrying a site-directed mutation where cysteines at positions 223 and 225 were substituted with serines. This mutation results in the secretion of biologically active TGF-1 (8). This retrovirus was designated as pMX-L-s223,225-TGF-1, whereas the empty vector with no TGF-1 cDNA was designated as pMX. In lung explants cultured with pMX, normal lung architecture was observed. However, 7 days after treatment with pMX-L-s223,225-TGF-1, there were increased quantities of active TGF-1inthe serum-free conditioned media (CM) overlying the explants. In addition, 14 days after transfection, the lung explants histologically had evidence of fibrosis, fibroblast buds, and enlarged air spaces resembling remodeled lung, as seen in IPF. The same lung explants had increased synthesis of collagens I, III, IV, and V and fibronectin by Western blot analysis.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION DISCLOSURES REFERENCES

 
Materials. Keratinocyte growth factor (KGF) and TGF-1 ELISA kit were purchased from R&D (Minneapolis, MN). Agarose, hydrocortisone, retinol acetate, insulin-transferrin-selenium X, and fetuin were purchased from Sigma (St. Louis, MO). DMEM was purchased from GIBCO BRL (Burlington, ON, Canada).

Preparation of retroviral vectors. The plasmid pPK9A (gift from Dr. Lalage Wakefield, Laboratory of Chemoprevention, National Institutes of Health, Bethesda, MD) contained the entire length of the cDNA of LTGF-1 where cysteines in positions 223 and 225 were substituted with serines, resulting in TGF-1 protein in its biologically active form upon secretion. This cDNA was designated as L-s223,225-TGF-1. The 1.2-kb TGF-1 was isolated by BglII digestion and purified with a DNA gel extraction kit (Qiagen, Mississauga, ON, Canada). The L-s223,225-TGF-1 was subcloned into the retrovector pMX. The retrovirus containing the TGF-1 gene was designated as pMX-L-s-223,225-TGF-1, whereas the retrovirus, pMX with no TGF-1 cDNA, was designated as pMX. The retroviruses were produced using the packaging cell line Plat E (26), and viral titration was estimated based on the number of infected NIH/3T3 cells using the method as described by Kitamura (26).

Animal procedures. Female Sprague-Dawley rats free of respiratory disease and weighing 200-250 g were obtained from The University of British Columbia Vivarium. All procedures on the rats were approved by the Canadian Council of Animal Care. Rats were anesthetized with 0.4 ml of ketamine (Biomedia-MTC, Cambridge, ON, Canada) and 0.2 ml of rompun (Bayer, Etobicoke, ON, Canada), which were administered intraperitoneally. With blunt dissection, the trachea was exposed, and an 18-gauge catheter was inserted. Thoracic and abdominal cavities were exposed; the inferior vena cava and abdominal aorta were severed. To remove peripheral blood leukocytes in the pulmonary circulation, 10 ml of normal saline were injected into the right ventricle until the lungs turned white. The trachea, lungs, and heart were removed. To remove alveolar inflammatory cells, the lungs were lavaged with 50-60 ml of warm normal saline through the trachea. The lungs were then infused with 5 ml of 40°C 0.4% agarose DMEM [2x solution of serum-free DMEM and 0.8% agarose solution (GIBCO) at 1:1 concentration at 40°C, supplemented with hydrocortisone (0.2 µg/ml), retinol acetate (0.2 µg/ml), and 2% insulin-transferrin-selenium X]. The trachea was closed by tying a thread, and the lungs were placed into six-well plates. The plates were left on ice overnight to further solidify the lungs. The lungs were separated from the heart and gently sliced manually from each lobe with a sterilized scalpel. Each pair of lungs yielded 40 slices that were 1-2 mm in thickness.

Culture of lung slices. Sterile gelform was made by adding 1.5 ml of warm 0.4% agarose DMEM into six-well plates. After agarose DMEM was solidified, a sterilized scalpel was used to remove gelform from the well. Six lung slices were placed on the top of each gelform, and 1.5 ml of serum-free DMEM supplemented with hydrocortisone (0.1 µg/ml), retinol acetate (0.1 µg/ml), and 1% insulin-transferrin-selenium X were added on the bottom of the gelform. Six-well culture plates were incubated at 37°C, 5% CO₂. Medium was changed twice a week, and lung slices were turned every other day and were collected on days 7 and 14 after culture. Treatment of the lung slices consisted of media, KGF (25 ng/ml), pMX [10⁶ 50% tissue culture infective dose (TCID₅₀)], pMX (10⁶ TCID₅₀) plus KGF (25 ng/ml), and pMX-L-s223,225-TGF-1 (10⁶ TCID₅₀) plus KGF (25 ng/ml) in the absence or presence of anti-TGF-1 antibody (0.1 µg/ml) or fetuin (10 µM). The rationale for using fetuin is based on the observations that for TGF-1-mediated signal to occur, TGF-1 must first bind to the type II TGF- receptor (TR-II) (45). After binding to the TR-II, TR-II recruits and phosphorylates the type I TGF- receptor before signal transduction by TGF- (45). The major cytokine-binding domain in the extracellular component of TR-II is within a 19- to 20-amino acid disulfide-looped sequence designated as TGF- receptor homology domain 1 (TRH1). Fetuin is a glycoprotein synthesized by hepatocytes and found in serum (10). The TRH1 has significant homology with fetuin, and, therefore, fetuin can bind TGF-1 and prevent TGF-1 from associating with TR-II. The rationale for using KGF was based on the following observations. Normally, AECs have a very low proliferative index and are quiescent (44, 47). Retroviruses only transfect proliferating cells (44). KGF, also referred to as fibroblast growth factor-7 (FGF-7), is a heparin-binding growth factor (44, 47). The critical characteristic of KGF is that the biological activity of KGF is restricted to epithelial cells because the KGF receptor is expressed only by epithelial cells (44, 47). In vitro KGF has been found to be a potent mitogen for type II AECs, and the effects are maximal 2 days after exposure. Selection of AECs as target cells for retroviral gene transfection was achieved by using KGF to induce AEC proliferation.

For histology, lung slices were fixed in 4% paraformaldehyde for 24 h and then 70% ethanol before paraffin embedding, staining with hematoxylin and eosin (H&E) for histology, and Masson's trichrome for distribution of collagen and elastin (14). All slides were blinded, and two independent examinations were done and then collated. For the extent of lung involvement, the lung sections were examined under low power, revealing the entire lung section that contained normal looking, as well as fibrotic, lesions. The proportion of lung section stained green with Masson's trichrome was recorded as a percentage of the overall lung section. For grading the extent of staining with Masson's trichrome, 0 designated no staining, grade 1 designated detectable color (see Fig. 2C for an example of grade 1), and grade 2 designated more green staining and was between staining grades 1 and 3 (see Fig. 2C, changes seen in the right-hand corner, for an example of grade 2 staining). Grade 3 represented extensive staining with Masson's trichrome (see Fig. 2, D-G, for an example of grade 3 staining). The overall score was calculated by multiplying the percentage of lung involved with the grade. Immunohistochemistry using anti--smooth muscle actin antibody was done to identify interstitial fibroblasts or myofibroblasts (33). Some lung slices from each condition were used for in situ hybridization, whereas other slices were collected for protein extraction and Western analysis. CM overlying the explants were collected in the presence of protease inhibitors, leupeptin (0.5 µg/ml; Amersham, Buckinghamshire, UK), and aprotinin and pepstatin A (1 µg/ml each; both from Sigma, Oakville, ON, Canada), and frozen at -80°C until ready for TGF-1 quantitation by ELISA.

View larger version (72K): [in this window] [in a new window]   Fig. 2. Histological changes 14 days after culture of lung explants. Masson's trichrome staining was used to identify the presence of collagen. A: lung explant cultured in media alone. B: lung explant cultured in pMX. C: lung explant cultured with keratinocyte growth factor (KGF) demonstrates AEC hyperplasia (arrows). D: lung explant cultured with pMX-L-s223,225-transforming growth factor (TGF)-1/KGF demonstrates AEC hyperplasia (small arrow) and extensive subepithelial fibrosis (large arrow) and enlarged air space. E and F are higher magnifi-cations of areas in D, where small arrows identify hyperplastic AECs, large arrows point to connective tissue, and arrowheads identify enlarged air spaces. G: fibroblast bud where small arrow identifies hyperplastic AECs and large arrow points to connective tissue. The histology presented is representative of results obtained from 8 different experiments. A-C are at x400 magnification, D is at x100 magnification, and E-G are at x400 magnification.

In situ hybridization to localize L-s223,225-TGF-1. Tissue sections from paraffin-embedded lung explants were de-waxed in xylene, rehydrated in a series of graded ethanol concentrations, and treated with proteinase K solution (20 µg/ml; Boehringer Mannheim, Mannheim, Germany) in 50 mM Tris · HCl, pH 7.4, 10 mM EDTA, and 10 mM NaCl. After recovery of the puromycin resistance cDNA, the 1.756-kb insert was labeled with digoxigenin. After preparation of the probe cocktail, 30 µl were distributed over each section and covered with a coverslip. The negative control followed the same steps as above, but the labeled probe was replaced with unlabeled probe. After denaturation of the DNA by heating the slides at 95°C, DNA-DNA hybridization was performed by incubation of slides in a hybridization oven at 42°C overnight. The anti-digoxigenin-Ap antibody (Fab fragment from sheep) at 1:500 was applied to the section, which was then cultured in a humid chamber. Thirty microliters of color solution (4-nitro blue tetrazolium chloride/5-bromo-4-chloro-3-indolyl phosphate) were applied to the slides, which were left in the dark overnight and stained with 0.5% neutral red (1% acetic acid).

Quantitation of retrovirus-mediated gene transfection in lung explants. To determine the approximate transfection efficiency of the L-s223,225-TGF-1 construct, retrovirus-mediated enhanced green fluorescent protein (EGFP) gene transfection was performed on lung explants. Plat E cells (500,000) were cultured with DMEM containing 10% FCS in a 60-mm dish. Fifty to eighty percent subconfluent Plat E cells were transfected with pMX-EGFP by Lipofectin (GIBCO). The pMX-EGFP retroviral culture supernatants were collected after 48 h of culture. The estimated titers of retroviruses were 1-10 x 10⁶ colony-forming units/ml. Lung slices were cultured on top of agarose gel in the absence or presence of 1 x 10⁶ TCID₅₀ pMX-EGFP virus. KGF was used to direct the transfection of pMX-EGFP into AECs. Medium was changed twice a week, and the lung slices were harvested on day 14. The lung tissue sections placed on slides were examined by fluorescence microscopy. Five areas per section were randomly selected for the calculation of the percentage of EGFP expression cells in the lung explants.

Transfection of pMX retroviruses into alveolar inflammatory cells. Despite our efforts to remove inflammatory cells from the lungs, it is recognized that all inflammatory cells from the lungs cannot be totally eliminated in our model. It is, therefore, possible that inflammatory cells remaining in the explants could be transfected by the pMX-L-s223,225-TGF-1 gene in the conditions used. To determine whether alveolar cells could be transfected with the pMX-L-s223,225-TGF-1 gene, the cells in the BAL obtained earlier were cultured with KGF (25 ng/ml) and pMX-L-s223,225-TGF-1 (10⁶ TCID₅₀) in puromycin-selective media. The presence of pMX-L-s223,225-TGF-1/KGF are conditions hypothesized to be ideal for selective transfection of AECs. The plates were examined daily for 14 days for viable cells, and CM were collected 5, 7, and 14 days after culture for detection of TGF-1 by ELISA. A positive control for successful transfection with pMX-L-s223,225-TGF-1 was done by using L2 cells. L2 cells are a simian virus 40 transformed cell line of normal rat AECs (18) that were cultured in identical conditions as were the alveolar inflammatory cells.

Detection of TGF-1 in CM. TGF-1 protein quantitation was done using a commercially available ELISA kit that only detects biologically active TGF-1 (7). Neutral CM expected to contain biologically active TGF-1 were used (19, 24, 48). However, an aliquot of the same sample was acidified to remove the LAP-1 from any LTGF-1 present. The sample was neutralized, and the TGF-1 content of the previously acidified CM was quantitated in the same ELISA assay to determine the total TGF-1 present in each sample.

Western analysis to detect and quantitate connective tissue proteins, phosphorylated Smad2, and connective tissue growth factor. The lung explants were snap-frozen on dry ice with ethanol and stored at -80°C until protein extraction. Lung explant protein extraction was performed as described previously (48). Briefly, the frozen lungs were pulverized in a chilled mortar and placed in tissue lysis buffer containing 1 mM phenylmethylsulfonyl fluoride (Sigma). The samples were further homogenized in the presence of 0.5% Triton X-100 and then centrifuged at 15,800 g for 10 min at 4°C. The supernatants were collected, and protein levels were determined using a Bio-Rad protein assay (Bio-Rad Laboratories, Hercules, CA). The protein samples (25 µl) were electrophoresed on 10% SDS-PAGE in a Mini-Protean II Electrophoresis Cell (Bio-Rad). Protein molecular weight markers (Amersham) were run parallel to each blot as an indicator of the molecular weight. Equal loading of protein was evaluated using silver staining (not shown). It is of note that an additional method to silver staining was done to validate equal loading of protein on SDS-PAGE gels using Ponceau S staining solution (Sigma) or Coomassie brilliant blue (Sigma) staining. The separated proteins were transferred at 50 V overnight onto nitrocellulose membrane (GIBCO BRL) in a Mini Trans-Blot chamber with transfer buffer (25 mM Tris · HCl, 192 mM glycine, and 20% methanol). The nitrocellulose membrane was blocked for 1 h using 5% instant skim milk in Tris-buffered saline (TBS). For detection of procollagen I and III (Rockland, Gilbertsville, PA), a 1:3,000 dilution of antibody was used; for collagen IV (Rockland), a dilution of 1:2,500 antibody was used; for collagen V (Cedarlane, Hornby, ON, Canada), a dilution of 1:3,000 antibody was used. A 1:1,000 dilution was used for fibronectin, 1:750 for phosphorylated Smad2 (Upstate Biotechnology, Lake Placid, NY), and 1:500 dilution of connective tissue growth factor (CTGF). After being washed, the nitrocellulose membrane was incubated with horseradish peroxidase linked with the secondary antibody (anti-rabbit or anti-goat immunoglobulin G; Bio-Rad), as recommended by the manufacturer. Finally, the washed blots were exposed to an enhanced chemiluminescence (ECL) detection system (Amersham) and recorded on an autoradiograph (Kodak X-Omat film). Before being reprobed, the nitrocellulose membrane was incubated at 50°C for 30 min with a stripping buffer (100 mM 2-mercap-toethanol, 2% SDS, and 62.5 mM Tris · HCl, pH 6.7). The blots were rinsed twice with TBS. To ensure the removal of antibodies, membranes were incubated with the ECL detection reagents and exposed to film (Kodak). No band was detected, confirming that all antibodies were stripped off the membrane. The same nitrocellulose membrane was blocked using 5% instant skim milk in TBS for detection of the other collagens and fibronectin. Relative absorbance was determined using the Quantity I imaging system (Bio-Rad).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION DISCLOSURES REFERENCES

 
Transfection of AECs by L-s223,225 TGF-1. All pulmonary cells express TGF-1 (27) at the mRNA level. To detect cells transfected with the pMX-L-s223,225-TGF-1 gene, all vectors used contained the puromycin resistance gene. To identify the cells transfected with TGF-1, in situ hybridization using digoxigenin-labeled puromycin-resistant gene cDNA was used. Controls used for the in situ hybridization method in which the only change in the procedure was substitution of unlabeled cDNA probe for a labeled digoxigenin probe demonstrated no evidence of staining (Fig. 1A). In explants cultured with pMX-L-s223,225-TGF-1, AECs were observed to contain the puromycin resistance gene (Fig. 1, B and C, large arrows). Furthermore, in the areas of lung where transfection with the pMX-L-s223,225-TGF-1 gene was confirmed by the presence of the puromycin resistance gene, there was associated thickened interstitium, suggestive of fibrosis (Fig. 1, B and C, large arrows). In the adjacent lung where there was no evidence of transfection observed (Fig. 1C, small arrows), there was also no evidence of interstitial changes. Epithelial cells of airways (Fig. 1C; arrowhead), endothelial cells of vessels, smooth muscle cells of airways (Fig. 1C; arrowhead), and vessels had no detectable staining for the puromycin resistance gene. Although KGF can induce proliferation of bronchial epithelial cells (BECs) (44), BECs are not responsive to retroviral transfection (44). Retroviral infection is mediated through the receptor Pit 2, a sodium-dependent phosphate transporter (44). The resistance to infection may be based on inaccessibility or low numbers of the Pit 2 on the apical membrane of BECs. In explants used for control consisting of pMX without L-s223,225-TGF-1, there was staining present in AECs, but there was no associated thickening of the interstitium (data not shown). Lung sections receiving no treatment or treatment with KGF only were negative for the puromycin resistance gene (data not shown). We further confirmed that explants cultured in KGF were transfected with the retrovirus pMX by using pMX-EGFP. We demonstrated that explants cultured with media, pMX-EGFP, or KGF only had no fluorescent cells (Table 1). However, 16.58 ± 1.04% of cells per field in explants cultured with pMXEGFP in the presence of KGF were fluorescent.

View larger version (91K): [in this window] [in a new window]   Fig. 1. In situ hybridization using digoxigenin-labeled cDNA of the puromycin resistance gene. A: control using cDNA probe without digoxigenin labeling demonstrates no gene transfection. B: alveolar epithelial cells (AECs) contain the pMX vector in regions where there is interstitial thickening compatible with fibrosis (arrows). C: an adjacent lung section with normal-appearing lung (small arrows) and an airway (arrowhead) have no evidence of gene transfection, whereas areas with evidence of transfection have associated thickening of the interstitium (large arrow). The results presented are representative of experiments from 4 different experiments. A-C are at x100 magnification.

Release of TGF-1 by cells of explants. We next determined whether transfected cells of the explants release biologically active TGF-1. Seven days after explantation and treatment, serum-free CM overlying the explants were examined for the presence of active TGF-1 using an ELISA. The CM obtained from lung explants cultured with L-s223,225-TGF-1/KGF contained increased quantities of TGF-1. In the presence of antibody to TGF-1, all the TGF-1 activity in the CM was neutralized, confirming the specificity of the ELISA. Fetuin binds to TGF-1 and prevents the association of TGF-1 with TR-II and does not allow TGF- to interact with TR-II. For this reason, in the presence of fetuin, the quantity of TGF-1 in the CM was increased (Table 2). No TGF-1 was detected in the CM of explants of control conditions consisting of explants receiving no treatment, the retrovirus without the TGF-1 gene, pMX, KGF, or pMX plus KGF. These findings confirm that cells in the explants treated with pMX-L-s223,225-TGF-1/KGF release biologically active TGF-1. Total TGF-1 generated by the explants did not vary significantly among the various treatments. Fourteen days after explantation and treatment, the CM from pMX-L-s223,225-TGF-1/KGF in the absence or presence of fetuin contained small quantities of active TGF-1. Because all other evidence (see Figs. 1, 2, 3, 4, 5) indicated a biological response to TGF-1, the decreased detection of TGF-1, 14 days after transfection, does not minimize the significance of our findings.

View this table: [in this window] [in a new window]   Table 2. TGF-1 in conditioned media of rat lung explants

View larger version (74K): [in this window] [in a new window]   Fig. 3. -Smooth muscle actin (-SMA) staining of explants. A: lung explant cultured with pMX alone demonstrates no evidence of -SMA. B: lung explant cultured with pMX-L-s223,225-TGF-1/KGF demonstrates the presence of myofibroblasts (arrows). The histology presented is representative of 4 different experiments. A and B are at x400 magnification.

View larger version (37K): [in this window] [in a new window]   Fig. 4. A-E: connective tissue expression in lung explants. Seven days after explantation, procollagens I, III, and IV are constitutively expressed in untreated lungs (lane 1) with no significant increase in the presence of pMX, KGF, or pMX + KGF (lanes 2, 3, and 4, respectively). There is an induction of collagens I (A), III (B), IV (C), V (D), and fibronectin (E) observed after culture in pMX-L-s223,225-TGF-1/KGF (lane 5), whereas in the presence of neutralizing anti-TGF-1 antibody (lane 6) or the glycoprotein fetuin, a competitive ligand for type II TGF- receptor, there was a decrease in the collagens and fibronectin (lane 7). The blots presented are representative of results obtained from 4 separate experiments. There was no statistically significant difference in the comparisons of expression of collagens I, III, IV, V, and fibronectin among samples treated with media (no treatment), pMX, KGF, or pMX + KGF. For collagen I, *P = 0.01-0.04 compared with no treatment, pMX, KGF, or pMX + KGF, whereas in the presence of anti-TGF-1 antibody or fetuin, P < 0.05. For collagen III, *P = 0.01-0.02 compared with no treatment, pMX, KGF, or pMX + KGF, whereas in the presence of anti-TGF-1 antibody, P = 0.1 (not significant), but in the presence of fetuin, P < 0.05. For collagen IV, *P < 0.05 compared with no treatment, pMX, KGF, or pMX + KGF, whereas in the presence of anti-TGF-1 antibody or fetuin, P < 0.05. For collagen V, *P = 0.03-0.05 compared with no treatment, pMX, KGF, or pMX + KGF, whereas in the presence of anti-TGF-1 antibody or fetuin, P < 0.02. For fibronectin, *P = 0.01-0.04 compared with no treatment, pMX, KGF, or pMX + KGF, whereas in the presence of anti-TGF-1 antibody or fetuin, P = 0.01 and 0.04, respectively. In each instance, pMX or pMX-L-s223,225-TGF-1 used was 106 50% tissue culture infective dose (TCID50), anti-TGF-1 antibody was 0.1 µg/ml, and the KGF used was 25 ng/ml. All P values (2-tailed) were based on the Student's t-test.

View larger version (34K): [in this window] [in a new window]   Fig. 5. Expression of phosphorylated Smad2 (p-Smad2; A) and connective tissue growth factor (CTGF; B). A: constitutive expression of p-Smad2 was observed in lung explants receiving no treatment (lane 1), whereas the presence of pMX, KGF, or pMX + KGF led to a minor induction (lanes 2, 3, and 4, respectively; P = 0.12-0.22 compared with control). Explants cultured in pMX-L-s223,225-TGF-1/KGF had a marked expression of p-Smad2 (lane 5) where P < 0.05 compared with no treatment, and P = 0.03, 0.05, and 0.05 compared with pMX, KGF, and pMX + KGF, respectively. Neutralizing antibody and fetuin minimally decreased this induction (lanes 6 and 7; P = 0.3 and 0.1, respectively, compared with pMX-L-s223,225-TGF-1/KGF). B: expression of CTGF was increased in explants cultured in pMX-L-s223,225-TGF-1/KGF (lane 5) where P = 0.01-0.04 compared with no treatment, pMX, KGF, or pMX + KGF but was reduced when anti-TGF-1 antibody or fetuin were present (lanes 6 and 7; *P < 0.05 and 0.03, respectively, compared with pMX-L-s223,225-TGF-1/KGF). In each instance, pMX or pMX-L-s223,225-TGF-1 used was 106 TCID50, anti-TGF-1 antibody was 0.1 µg/ml, and the KGF used was 25 ng/ml. The blots presented are representative of results obtained from 4 separate experiments. All P values (2-tailed) were based on the Student's t-test.

Alveolar inflammatory cells are not transfected by L-s223,225-TGF-1. In the current model of pulmonary fibrosis, alveolar inflammatory cells were removed by extensive BAL. The pulmonary vasculature was flushed by forcefully injecting normal saline into the right ventricle while the inferior vena cava and aorta were severed. This method is likely to remove not all but a large number of inflammatory cells from the alveoli and vasculature of the lungs. To determine whether inflammatory cells remaining in the lungs could be transfected with pMX-L-s223,225-TGF-1, cells obtained by BAL were used. These inflammatory cells were cultured with pMXL-s223,225-TGF-1/KGF, conditions that have been demonstrated to be optimum for transfection of the TGF-1 gene into AECs of lung explants (Fig. 1; Tables 1 and 2). Because we had previously demonstrated that a cell line of AECs called L2 could be transfected by pMX-L-s223,225-TGF-1 (46), L2 cells were used as a positive control. It is of note that the pMX vectors contain the gene for puromycin resistance, and only cells successfully transfected with the vector will survive in media containing puromycin. Inflammatory cells did not survive in selected media containing puromycin (1 µg/ml; not shown), and there was no evidence of biologically active TGF-1 in the CM (Table 3). However, L2 cells treated with pMX-L-s223,225-TGF-1/KGF, previously demonstrated to secrete active TGF-1 in these conditions (46), remained viable in puromycin-selective media (not shown) and released active TGF-1 (Table 3). These findings, in addition to the absence of puromycin labeling in inflammatory cells (Fig. 1), confirm the selective nature of transfection of the L-s223,225-TGF-1 gene into AECs without transfecting any inflammatory cells that remain in the explants.

View this table: [in this window] [in a new window]   Table 3. TGF-1 in CM of rat alveolar macrophages or alveolar epithelial cells

Transfection of AECs with L-s223,225-TGF-1 leads to pulmonary fibrosis and remodeling. Because lung sections stained with Masson's trichrome demonstrated the histology as well as distribution of collagens, the histology presented is that observed with Masson's trichrome. H&E-stained sections are not shown. All slides were blinded, and two independent examinations were done and collected. All fields of all lung sections were examined. Lung explants cultured in media alone (Fig. 2A) or pMX (Fig. 2B) retained a normal alveolar architecture. Lung explants cultured with KGF demonstrated a previously described characteristic of micropapillary epithelial cell hyperplasia or "knobby proliferation" (Fig. 2C) (44, 47). The knobby appearance has been correlated with AEC proliferation (Fig. 2C) (44, 47). Unlike the normal lung in which type II AECs are 4% of the alveolar surface (44, 47), in these areas, all alveoli appeared to be hyperplastic type II AECs and resembled a cuboidal epithelial monolayer. The distribution of these changes was patchy, being present primarily in the periphery of the lung tissue. Lung explants cultured with media, pMX, or KGF did not demonstrate collagen. Lung explants cultured with L-s223,225-TGF-1 in the presence of KGF, designated as pMX-L-s223,225-TGF-1/KGF for brevity, demonstrated a patchy distribution of AEC hyperplasia and interstitial thickening on H&E (not shown) and significant distribution of collagen by Masson's trichrome (Fig. 2, D-G). The histological changes were observed throughout the explants treated with pMX-L-s223,225-TGF-1/KGF. However, the severity of changes was observed primarily on the surface, as deeper cuts into the paraffin-embedded lung section led to a loss of histological changes, resembling remodeled lung. These features suggest that the surface of the explants was most accessible to the retroviral vector and thus transfection. Masson's trichrome demonstrated extensive collagen deposition in areas of thickened interstitium (Fig. 2, D-G). In some sections, fibroblasts were observed (Fig. 2, D-G). Fibroblast buds were also found (Fig. 2G), which were lined with hyperplastic type II AECs (small arrow), and extensive fibrosis was observed subepithelially and within the bud (Fig. 2G, large arrow). In another section, evidence of enlarged air spaces reminiscent of honeycomb cysts was also seen (Fig. 2, E and F, arrowheads). The relative presence of collagen, as assessed by the extent of green coloration, was most evident in explants cultured with KGF and pMX-L-s223,225-TGF-1 14 days after transfection. The presence of anti-TGF-1 antibody seemed to have little effect on the expression of collagen (Table 4). The total absence of detectable collagen in explants treated with KGF plus pMX relative to the other control conditions is unclear. It is of interest that in regions where Masson's trichrome staining was observed, the overlying alveolar epithelium was hyperplastic in morphology, suggestive of proliferation (Fig. 2, D-G, small arrows). Induction of proliferation of AECs would make these AECs susceptible to pMX-mediated transfection. In the normal lung, interstitial fibroblasts are not visible. TGF- induces fibroblast proliferation, fibroblast recruitment, and differentiation to myofibroblasts, as well as connective tissue synthesis (17, 30). In lung sections receiving no treatment or cultured with pMX (Fig. 3A), KGF, or pMX plus KGF (not shown), there was no evidence of immunohistostaining with -smooth muscle actin. In explants cultured with pMX-L-s223,225-TGF-1/KGF, there were areas of substantial connective tissue with elongated cells that were likely to be fibroblasts. In these areas, -smooth muscle actin stained a number of cells within the thickened interstitium (Fig. 3B, arrows), indicating that the effects of TGF-1 on interstitial cells can lead to the development of myofibroblast, a cell phenotype that has more contractile properties than fibroblasts (14, 17, 33, 36). These changes were not observed in lung explants receiving no treatment or in explants treated with KGF, pMX, or KGF plus pMX. Lung sections obtained from explants cultured with pMX-L-s223,225-TGF-1/KGF and anti-TGF-1 antibody or fetuin had less fibrosis and the minimal number of cells that immunostained with -smooth muscle actin (not shown). It is of note that examination of explants on days 7 and 14 showed no evidence of tissue necrosis by light microscopy. Staining of explants with trypan blue also demonstrated no evidence of cell necrosis (data not shown).

Expression of connective tissue proteins in lung explants. In fibrotic lesions such as those seen in pulmonary fibrosis, there is increased synthesis of collagens I, III, and fibronectin, whereas collagen IV expression is associated with basement membrane synthesis, and collagen V expression is seen as part of provisional matrix (11). To verify and quantitate connective tissue synthesis, proteins from explants were immunoblotted for collagens I, III, IV, V, and fibronectin. Collagen I and fibronectin expression was observed in all experimental conditions, but the expression of these proteins was elevated in lung explants cultured with pMX-L-s233, 225-TGF-1/KGF (Fig. 4). In the presence of neutralizing antibody to TGF-1 or fetuin, there was a decrease in expression of connective tissue proteins (Fig. 4, lanes 6 and 7, respectively).

Cells of lung explants respond to TGF-1 by expressing phosphorylated Smad2 and CTGF. An index of target cell response to TGF- is the phosphorylation of Smad proteins that mediate intracellular signals of the TGF-1 superfamily (35), more specifically, the phosphorylation of the COOH-terminal SSXS motif of Smad2 and Smad3 (35). The phosphorylated Smad2 (p-Smad2) antibody is highly specific and sensitive in detecting TGF-1-dependent signaling (35). To confirm that the TGF-1 released in lung explants regulates signal transduction, proteins from the explants were immunoblotted with anti-phosphorylated Smad2 antibody. Explants receiving no treatment had detectable constitutive expression of p-Smad2 (Fig. 5, lane 1). The presence of pMX increased the expression of p-Smad2, suggesting that some nonspecific induction of a TGF-1-mediated signal occurred in the presence of pMX. However, there was a more marked increase in p-Smad2 when pMX-L-s233,225-TGF-1/KGF was in cultures of the lung explants. The presence of neutralizing antibody to TGF-1 did not significantly decrease the expression of p-Smad2, whereas the presence of fetuin resulted in some decrease of p-Smad2 expression.

The induction of connective tissue synthesis may occur in response to a number of cytokines, such as IL-1, PDGF, IGF-1, and FGF-2 (2, 32, 34, 43). Many of these cytokines are induced by TGF- (28). Induction of connective tissue synthesis by TGF-1, but not other fibrogenic cytokines, is mediated by CTGF (16, 30). CTGF is a 33- to 38-kDa cysteine-rich protein that also regulates fibroblast proliferation and chemotaxis (16, 30). To confirm that connective tissue synthesis observed in lung explants was mediated by TGF-1, the proteins were immunoblotted with anti-CTGF antibodies. Constitutive CTGF expression was observed in control samples (Fig. 5, lanes 1-4), but explants cultured with pMX-L-s233,225-TGF-1/KGF had a significant increase in CTGF (Fig. 5, lane 5). In the presence of neutralizing antibodies to TGF-1 or fetuin, CTGF expression was decreased (Fig. 5, lanes 6 and 7, respectively). These findings confirm that when the TGF-1 is released by the AECs of the explant, and it interacts with its TR-II, then there is induction of connective tissue synthesis by the cells in the explant. Lastly, the presence of insulin and KGF in the defined media may be considered components of inflammation. However, the presence of insulin and KGF without transfection with the retrovirus pMX-L-s223,225-TGF-1 had no effect on the structural integrity of the tissue by light microscopy, increase in the release of TGF-1 (Table 2), or connective tissue synthesis (Fig. 4). In the absence of pMX-L-s223,225-TGF-1, there was also no evidence of increase in responsiveness to TGF-1 by induction of p-Smad2 and CTGF (Fig. 5).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION DISCLOSURES REFERENCES

 
We had previously demonstrated that using an antibody to TGF-1 that immunoreacts with TGF-1 in its active conformation (22-24), TGF-1 was aberrantly present in the epithelial cells lining honeycomb cysts of patients with IPF (22, 23). AECs from normal controls or from an area of IPF lungs with no evidence of disease did not show immunoreactivity with this antibody of TGF-1 (22, 23). On the basis of these findings, the current study was done to determine the effects of biologically active TGF-1 from AECs. Our data demonstrate that the overproduction of a single cytokine, TGF-1, from AECs results in remodeling of the lung with interstitial fibrosis, fibroblast bud formation, and enlarged air spaces, which are lesions compatible with IPF. Although epithelial cells of IPF lungs have been observed to express other fibrogenic cytokines such as PDGF (2), TNF- (32, 34), IGF-I (43), and IL-1 (34), these cytokines are induced by TGF-1 (28, 43). For this reason, the importance of AEC-derived TGF-1 is even more relevant since the expression of these cytokines in IPF may be a consequence of the aberrant production of TGF-1 by AECs. Furthermore, substantial remodeling in the lung explants was observed after 14 days and suggests that chronic injury and inflammation may not be as important as previously thought. Rather, distortion of air spaces and extensive fibrosis may occur because of the intensity or the quantity of TGF-1 released.

The mechanism by which AECs overproduce TGF-1 in IPF is unclear. Early lesions of IPF have changes in the alveolar epithelium, suggestive of injury (12, 38). In addition, the presence of fibroblast foci is also speculated to be important as the initial lesion in the pathogenesis of IPF and is a criterion for the diagnosis of usual interstitial pneumonitis of IPF (12, 38). Fibroblast foci are small aggregates of actively proliferating myofibroblasts and fibroblasts and are hypothesized to be present at sites of AEC injury where there may be inadequate reepithelialization (15, 37). The lack of epithelial cells, in turn, leads to induction of the underlying fibroblast and myofibroblast proliferation and connective tissue synthesis (15, 37). Fibroblast foci were not observed in our lung sections, and this may be because, in this model, AECs were not injured by pMX or KGF. Although no fibroblast foci were observed, there were cells in the interstitium that resembled fibroblasts and protrusions of misshapen lung tissue that resembled lesions, referred to as fibroblast buds (22). Furthermore, the presence of KGF has been previously demonstrated to protect AECs from various injuries (42, 44, 47). The lack of injury in this model does not diminish the relevance of the observation that in the event of AEC release of active TGF-1, rapid pulmonary remodeling will follow. It should be noted that even though fibroblasts are the main source of connective tissue, epithelial cells can also synthesize a variety of connective tissue proteins (13). Because TGF-1 has been demonstrated to induce epithelial cells to synthesize collagens (13), it is then possible that the enhanced connective tissue synthesis observed in the explants may be due to the effects of TGF-1 on the AECs as well as underlying fibroblasts.

The findings from the current model do not support inflammatory cells as being important in the pathogenesis of pulmonary fibrosis. This is because inflammatory cells in the pulmonary vessels and alveoli were removed from the lung before slicing the lung for ex vivo culture. Maintaining viable lung tissue in culture also overcame the contribution of circulating leukocytes that could be recruited to the interstitium and alveoli. Furthermore, in situ hybridization demonstrated successful transfection of pMX-L-s223,225-TGF-1 into AECs and not inflammatory cells. Lastly, the lack of viability of rat alveolar inflammatory cells in puromycin-selective media demonstrates that even if alveolar inflammatory cells, such as macrophages, were present in the explants, they would not be transfected by the TGF-1 gene and contribute to the generation of TGF-1 in the explants. Although it is not likely that the inflammatory cells remaining in the lung were transfected by pMX-L-s225,225-TGF-1, these cells may respond to the AEC-derived TGF-1 by generating a variety of cytokines, some of which may be fibrogenic in nature. The effects of fibrogenic cytokines from residual pulmonary inflammatory cells cannot be excluded in the current model. Despite these considerations, there is a selective nature of AECs as a source of biologically active TGF-1 in this model, and it confirms the role of AECs as being important in the pathogenesis of pulmonary fibrosis and remodeling.

Observations of others also support the importance of AECs in the pathogenesis of pulmonary fibrosis. For instance, Munger et al. (31) demonstrated that the expression of the integrin v6 on AECs after lung injury could result in activation of LTGF-1 and consequent pulmonary fibrosis. Adamson et al. (1) observed that under hyperoxic conditions, there was fi-brosis in areas of AEC necrosis in lung explants that were free of peripheral blood. There is additional support for the importance of epithelial cells in the pathogenesis of pulmonary fibrosis in IPF. In transgenic mice, when human TGF- was overexpressed by pulmonary epithelial cells, there was pulmonary fibrosis without evidence of changes in the number of inflammatory cells in the lungs (29, 41). It is of interest that in another rat model of pulmonary fibrosis described by Sime et al. (40), when the adenovirus with L-s233,225-TGF-1 was administered intratracheally, there were increased quantities of TGF-1 in fluid retrieved by BAL. In addition, there was extensive pulmonary fi-brosis (40). Because adenoviruses can transfect many types of cells, it is possible that inflammatory as well as structural cells may have been induced to express TGF-1 in this study (40). The findings of Sime et al. are valuable in confirming that increased quantities of TGF-1 in the lung can lead to fibrosis (40) but do not identify the cellular source of TGF-1 release. Unlike the findings of Sime et al. (40), the findings from the current study demonstrate the role of TGF-1 release from AECs as being important in the pathogenesis of pulmonary fibrosis.

To confirm that the increase in connective tissue synthesis was mediated by TGF-1 generated by the cells of the explant, a number of means were undertaken. In the presence of neutralizing antibody to TGF-1 or fetuin, which is a competitive ligand for the TR-II (10), there was a decrease in connective tissue synthesis. The TGF-1 antibody or fetuin was not able to block connective tissue synthesis completely. This could be because the antibody or fetuin could not adequately reach all areas of the explants that may have been transfected with pMX-L-s223,225-TGF-1 or were responding to TGF-1. Alternatively, a significant amount of TGF-1 could already have been associated with the receptor by the time that the TGF-1 antibody or fetuin was added. Further confirmation that the connective tissue responses were due to TGF-1 was the expression of p-Smad2 and CTGF. Induction of p-Smad2, an intracellular enzyme, was increased in conditions where there was an induction of TGF-1. Under conditions meant to neutralize TGF-1 using TGF-1 antibody or fetuin, which would interfere with TGF-1 binding to its receptor, p-Smad2 was decreased only in the presence of fetuin but not anti-TGF-1 antibody. It is possible that the overproduction of TGF-1 could have resulted in the release of other TGF- isoforms, which may then have phosphorylated Smad2. However, fetuin, which mimics the TRH1, is not specific to inhibiting the effects of TGF-1 but may interfere with the binding of other TGF- isoforms to the TR-II and thus phosphorylation of Smad2. Lastly, CTGF is critical as an intermediary protein for the synthesis of connective tissue proteins by TGF-1 but not other fibrogenic cytokines (16, 30). Fibroblasts and epithelial cells respond to TGF-1 by expressing CTGF which, in turn, results in collagen synthesis (12, 36). Collagen synthesis mediated by cytokines other than TGF-1 does not induce CTGF (12, 36). Explants cultured with pMX-L-s223,225-TGF-1 had marked increases in CTGF, whereas the presence of TGF-1 antibody and fetuin suppressed CTGF expression. These findings confirm that TGF-1 generated by the cells of the explant results in connective tissue synthesis. Although the current model results in successful transfection of AECs with the TGF-1 gene followed by release of biologically active TGF-1 and a connective tissue response, there are some limitations in this model. These limitations are the patchy nature of successful transfection, the limited number of days the tissue can be maintained in culture, and the reduction of TGF-1 activity in conditioned media after 7 days. The reduction in TGF-1 activity may have been due to a deterioration of the TGF-1 secreted, internalization of TGF-1 by the cells of the explants, attachment of TGF-1 to the TGF--binding proteins present on the cells of the explant, or adherence of TGF-1 to the plastic of the culture dishes (48). It is possible that despite a stable transfection, there may have been a decrease in the synthesis of TGF-1bythe AECs. Alternatively, it is possible that although there was no histological evidence of a decrease in viability of AECs, there may have been AEC dysfunction after 14 days of transfection. This could lead to a decrease in the release of active TGF-1.

The clinical implications of these findings are highly significant. The previously held dogma described the pathogenesis of pulmonary fibrosis in IPF to start with a pulmonary injury, followed by recruitment and activation of inflammatory cells that release proinflammatory and fibrotic cytokines such as TGF-1, PDGF, IGF-I, FGF-2, IL-1, granulocyte/macrophage colony-stimulating factor, and TNF-, leading to chronic inflammation and fibrosis (15, 22-24, 37). On the basis of this premise, the treatment for IPF and many other progressive inflammatory and fibrotic diseases has been the use of anti-inflammatory agents such as corticosteroids, azathioprine, and cyclophosphamide (15, 22-24, 37). An example where injury and inflammation precede fibrosis is the most common animal model of pulmonary fibrosis induced by the anti-neoplastic antibiotic bleomycin (20, 21, 25, 48, 49). Bleomycin-induced lung toxicity is characterized by AEC injury followed by recruitment and activation of inflammatory cells before enhanced connective tissue synthesis (20, 21, 25, 48, 49). In addition, we have demonstrated that after bleomycin treatment, the administration of high doses of corticosteroids prevents the recruitment of inflammatory cells into the injured lung (25), and there is a decrease in collagen synthesis (39). However, the distribution of TGF-1 in its biologically active conformation in IPF lungs is not the same as in bleomycin-induced pulmonary fibrosis (BPF) (20, 22, 23, 48). We have demonstrated that after bleomycin administration, the biologically active form of TGF-1 is almost exclusively expressed by alveolar macrophages (20). The secretion of biologically active TGF-1 by alveolar macrophages is critical to the pathogenesis of pulmonary fibrosis after bleomycin administration (21, 25, 48, 49). Unlike BPF in lungs of patients with IPF, biologically active TGF-1 is expressed by AECs and epithelial cells lining honeycomb cysts and fibroblast buds (22-24). In addition, fluid lining alveolar spaces where the expression of TGF-1 in epithelial cells is highly expressed contained active TGF-1 (24). These findings suggest that in IPF lungs, epithelial cells release TGF-1. We have also demonstrated that high doses of corticosteroids do not alter the secretion of TGF-1 (25). Furthermore, in all the explant cultures, the media were supplemented with 0.2 µg/ml of hydrocortisone. The presence of corticosteroids had no effect on the generation of active TGF-1 or connective tissue synthesis, further confirming that in instances of increased TGF-1 production, the presence of corticosteroids is not of benefit. It must be recognized that epithelial cells are structural cells, and their numbers cannot be altered by any immunosuppressive agents, and if steroids do not decrease secretion of biologically active TGF-1, then this may explain the failure of standard immunosuppressive therapy for IPF and other progressive fibrotic diseases. It is then possible that progressive fibrotic diseases like IPF (15, 22-24, 37), cirrhosis of the liver (3), or glomerulonephritis (4) are disorders of epithelial cells and not inflammatory cells. If this is the case, as our study suggests, then treatment of progressive fibrotic diseases warrants development of therapeutic modalities that would interfere with the generation of TGF-1 or the effects of biologically active TGF-1.

	DISCLOSURES

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION DISCLOSURES REFERENCES

 
This work was supported by the Canadian Institutes of Health Research (CIHR), CIHR Research and Development, and Glaxo-SmithKline.

	ACKNOWLEDGMENTS

 
We thank Valerie Romanchuk for preparing the manuscript and Trina Simon for technical assistance.

	FOOTNOTES

 

Address for reprint requests and other correspondence: Dr. N. Khalil, Division of Respiratory Medicine, Jack Bell Research Centre, 2660 Oak St., Vancouver, BC V6H 3Z6, Canada (E-mail: nkhalil{at}interchange.ubc.ca).

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

	REFERENCES

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION DISCLOSURES REFERENCES

 

1. Adamson IY, Young L, and Bowden DH. Relationship of alveolar epithelial injury and repair to the induction of pulmonary fibrosis. Am J Pathol 130: 377-383, 1988.[Abstract]
2. Antoniades HN, Bravo MA, Avila RE, Galanopoulus T, Neville J, and Selman M. Platelet-derived growth factor in idiopathic pulmonary fibrosis. J Clin Invest 86: 1055-1064, 1990.[ISI][Medline]
3. Bauer M and Schuppan D. TGF-1 in liver fibrosis: time to change paradigms? FEBS Lett 502: 4-10, 2001.[ISI][Medline]
4. Bitzer M, Sterzel RB, and Bottinger EP. Transforming growth factor- in renal disease. Kidney Blood Press Res 21: 1-12, 1998.[ISI][Medline]
5. Blanchette F, Day R, Dong W, Lapris MH, and Dubois CM. TGF-1 regulates gene expression of its own converting enzyme furin. J Clin Invest 99: 1974-1983, 1997.[ISI][Medline]
6. Blobe GC, Schiemann WP, and Lodish HF. Role of transforming growth factor in human disease. N Engl J Med 342: 1350-1358, 2000.[Free Full Text]
7. Brenchley PE, Williams S, Coupes B, and Short CD. Strategies for meaningful measurement of circulating levels of transforming growth factor 1 in clinical samples. Transplantation 71: 1355-1356, 2001.[ISI][Medline]
8. Brunner AM, Marquardt H, Malacko AR, Lioubin MN, and Purchio AF. Site directed mutagenesis of cysteine residues in the pro region of the transforming growth factor 1 precursor: expression and characterization of mutant proteins. J Biol Chem 264: 13660-13664, 1989.[Abstract/Free Full Text]
9. Coultas DB, Zumwalt RE, Black WC, and Sobonya RE. The epidemiology of interstitial lung diseases. Am J Respir Crit Care Med 150: 967-972, 1994.[Abstract]
10. Demetriou M, Binkert C, Sukhu B, Tenenbaum HC, and Dennis JW. Fetuin/2-HS glycoprotein is a transforming growth factor- type II receptor mimic and cytokine antagonist. J Biol Chem 271: 12755-12761, 1996.[Abstract/Free Full Text]
11. Eckes B, Zigrino P, Kessler D, Holtkotter O, Shephard P, Mauch C, and Krieg T. Fibroblast-matrix interactions in wound healing and fibrosis. Matrix Biol 19: 325-332, 2000.[ISI][Medline]
12. Fang KC. Mesenchymal regulation of alveolar repair in pulmonary fibrosis. Am J Respir Cell Mol Biol 23: 142-145, 2002.
13. Funderburgh JL, Funderburgh ML, Mann MM, Corpuz LM, and Roth MR. Proteoglycan expression during transforming growth factor -induced keratocyte-myofibroblast transdifferentiation. J Biol Chem 276: 44173-44178, 2001.[Abstract/Free Full Text]
14. Grogan T, Reinhardt K, Jaramillo M, and Lee D. An update on "special stain" histochemistry with emphasis on automation. Adv Anat Pathol 7: 110-122, 2000.[ISI][Medline]
15. Gross TJ and Hunninghake GW. Idiopathic pulmonary fibrosis. N Engl J Med 345: 517-525, 2001.[Free Full Text]
16. Grotendorst GR. Connective tissue growth factor: a mediator of TGF-1 action on fibroblasts. Cytokine Growth Factor Rev 8: 171-179, 1997.[Medline]
17. Hashimoto S, Gon Y, Takeshita I, Matsumoto K, Maruoka S, and Horie T. Transforming growth factor-1 induces phenotypic modulation of human lung fibroblasts to myofibroblast through a c-Jun-NH₂-terminal kinase-dependent pathway. Am J Respir Crit Care Med 1163: 152-157, 2001.
18. Hoffmann G, Grote J, Friedrich F, Mutz N, and Schobersberger W. The pulmonary epithelial cell line L2 as a new model for an inducible nitric oxide synthetase expressing distal airway epithelial cell. Biochem Biophys Res Commun 217: 575-583, 1995.[ISI][Medl