AJP - Renal Fuel your research with LabChart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Renal Physiol 293: F631-F640, 2007. First published May 9, 2007; doi:10.1152/ajprenal.00379.2006
0363-6127/07 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Figure and Video
Right arrow All Versions of this Article:
293/2/F631    most recent
00379.2006v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Xu, Q.
Right arrow Articles by Kopp, J. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Xu, Q.
Right arrow Articles by Kopp, J. B.

INNOVATIVE METHODOLOGY

In vitro models of TGF-beta-induced fibrosis suitable for high-throughput screening of antifibrotic agents

Qihe Xu,1,2 Jill T. Norman,2 Shashi Shrivastav,1 Javier Lucio-Cazana,3 and Jeffrey B. Kopp1

1Kidney Disease Section, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Department of Health and Human Services, Bethesda, Maryland; 2Division of Medicine, Royal Free and University College Medical School, University College London, London, United Kingdom; and 3Department of Physiology, University of Alcala, Alcala de Henares, Madrid, Spain

Submitted 22 September 2006 ; accepted in final form 7 May 2007


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Progressive fibrosis is a cause of progressive organ dysfunction. Lack of quantitative in vitro models of fibrosis accounts, at least partially, for the slow progress in developing effective antifibrotic drugs. Here, we report two complementary in vitro models of fibrosis suitable for high-throughput screening. We found that, in mesangial cells and renal fibroblasts grown in eight-well chamber slides, transforming growth factor-beta1 (TGF-beta1) disrupted the cell monolayer and induced cell migration into nodules in a dose-, time- and Smad3-dependent manner. The nodules contained increased interstitial collagens and showed an increased collagen I:IV ratio. Nodules are likely a biological consequence of TGF-beta1-induced matrix overexpression since they were mimicked by addition of collagen I to the cell culture medium. TGF-beta1-induced nodule formation was inhibited by vacuum ionized gas treatment of the plate surface. This blockage was further enhanced by precoating plates with matrix proteins but was prevented, at least in part, by poly-L-lysine (PLL). We have established two cell-based models of TGF-beta-induced fibrogenesis, using mesangial cells or fibroblasts cultured in matrix protein or PLL-coated 96-well plates, on which TGF-beta1-induced two-dimensional matrix accumulation, three-dimensional nodule formation, and monolayer disruption can be quantitated either spectrophotometrically or by using a colony counter, respectively. As a proof of principle, chemical inhibitors of Alk5 and the antifibrotic compound tranilast were shown to have inhibitory activities in both assays.

collagen; fibronectin; cell migration; Smad3; tranilast

RENAL FIBROSIS, INCLUDING glomerulosclerosis and tubulointerstitial fibrosis, remains largely incurable and irreversible, contributing to significant mortality (44). The molecular mechanisms of fibrosis are not fully understood, but multiple extracellular stimuli and intracellular signaling pathways have been implicated, and common pathways of fibrogenesis, such as an imbalance of extracellular matrix (ECM) synthesis and degradation, and cross-linking and stabilization of collagen proteins, have been identified (3, 10, 27). Despite this complexity, a central regulator of tissue fibrosis is the cytokine transforming growth factor-beta (TGF-beta), including three highly homologous mammalian isoforms, TGF-beta1, 2, and 3, which have similar fibrogenic effects in vitro (3, 10, 27). In vivo, the profibrotic effect of TGF-beta1 is well established, since it is overexpressed in most fibrotic tissues (3) and TGF-beta1 transgenic mice develop progressive fibrosis in multiple organs, especially the kidneys (8, 37). Thus inhibitors of TGF-beta have been sought as potential therapies to treat fibrotic diseases of the kidney (27), liver (4), lung (5), skin (19), heart, and blood vessels (18).

New drug development relies heavily on in vitro models (6, 20). However, in vitro models of fibrosis resembling in vivo fibrotic tissues and suitable for high-throughput quantification of global fibrogenesis have not been reported previously. This, at least in part, accounts for the limited progress in the development of antifibrotic drugs. In vitro models are especially important in the development of antifibrotic drugs, since fibrotic diseases in vivo are often complicated by inflammation, which can both serve as an initiating factor and contribute to disease progression. As a result, if a drug suppresses fibrotic lesions in patients or animal models, it is very difficult to dissect whether this effect is secondary to inhibition of inflammation or due to suppression of fibrosis per se. There is a clear need for validated in vitro models of inflammation-independent fibrosis.

Fibrosis is characterized by the net accumulation of ECM proteins and disruption of normal tissue architecture (3, 10). Mesenchymal cells, including fibroblasts in different organs, mesangial cells in renal glomeruli and smooth muscle cells in blood vessels, and epithelial cells, including tubular epithelial cells, are targets of the profibrotic effects of TGF-beta1 (3, 27). Mesenchymal cells are usually the major contributor of pathological ECM accumulation, and epithelial cells may transdifferentiate into mesenchymal cells through epithelial-mesenchymal transition (EMT), in which TGF-beta1 is believed to play an important role (27). In this study, we have tested the effects of TGF-beta1 on both mesenchymal cells and epithelial cells from kidneys and other organs cultured in a variety of culture systems and have established two in vitro models of fibrogenesis suitable for high-throughput assays.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Cell culture. Human mesangial cells (HMCs) were obtained from adult kidney tissue, as we have previously described, and were used between passages 8 and 12 (28). An immortalized human mesangial cell line (HMCL) was developed by Professor J. D. Sraer (Hôpital Tenon, Paris, France) (41) and is a kind gift from Dr. Xiongzhong Ruan (Royal Free and University College Medical School, London, UK) (35). Other cell lines, including NRK-49F rat renal fibroblasts, Hs 203 human spleen fibroblasts, LLC-PK1 renal tubular epithelial cells, Madin-Darby canine kidney renal tubular epithelial cells, and HK2 human renal tubular epithelial cells were purchased from American Type Culture Collection (Manassas, VA). A conditionally immortalized human podocyte cell line, bearing a temperature-sensitive SV40 T antigen, was a kind gift from Dr. Moin Saleem (University of Bristol, Bristol, UK) (36). Smad2–/–, Smad2+/+, Smad3–/– and Smad3+/+ mouse embryonic fibroblasts (MEFs) were kindly provided by Dr. Kathy Flanders (National Institutes of Health, Bethesda, MD) (11). All other cells in this study, including cardiac, hepatic, pulmonary, foreskin, embryonic fibroblasts, as well as pig aortic smooth muscle cells, human keratinocytes, and mouse tubular epithelial cells, were primary cultures from normal tissues generously provided by Dr. Ying Hong, Dr. Gisela Lindahl, and Dr. Robin McAnulty (University College London, London, UK), Dr. Carol Yee (National Institutes of Health, Bethesda, MD), and Dr. Esteban Mezey (Johns Hopkins Medical School, Baltimore, MD). Podocytes were differentiated by culture for 14 days at 37°C before being seeded in eight-well chamber slides for the experiments. All cell culture reagents were purchased from Invitrogen (Carlsbad, CA) unless stated otherwise. Cells were maintained in RPMI 1640 supplemented with 10% FBS and antibiotics (100 U/ml penicillin, 100 µg/ml streptomycin, and 0.25 µg/ml amphotericin B), except for human keratinocytes, which were grown in serum-free keratinocyte medium. Cells were trypsinized and seeded into 8-well chamber slides or 96-well plates in normal growth medium supplemented with 10% FBS and cultured overnight to 50–100% confluence. After a washing with serum-free medium, fresh medium supplemented with 1% FBS was added and cells were cultured for 24 h before any further experiments, if not stated otherwise. All experiments used 1% FBS, except when serum dose-response was tested. When experiments involved the use of inhibitors, these agents were dissolved in DMSO; vehicle control wells received equal volumes of DMSO.

Materials. Human TGF-beta1 was purchased from R&D Systems (Minneapolis, MN). Soluble collagens type I, III and IV (COL1, COL3, COL4), fibronectin (FN) and laminin (LN) were purchased from BD Biosciences (San Jose, CA). Calbiochem Alk5 inhibitor I ([3-(pyridin-2-yl)-4-(4-quinonyl)]-1H-pyrazole) and Calbiochem Alk5 inhibitor II (2-(3-(6-methylpyridin-2-yl)-1H-pyrazol-4-yl)-1, 5-naphthyridine) and p38 MAP kinase/Alk5 inhibitor SB203580 were purchased from EMD Biosciences (San Diego, CA). Tranilast was purchased from Sigma (St. Louis, MO). Cell culture systems used in this study included Lab-Tek 8-well chamber slides and Lab-Tek II 4-well chambered coverglass (Nalge Nunc, Rochester, NY); non-tissue culture (non-TC)-treated, tissue culture (TC)-treated plastic (by vacuum ionized gas), matrix protein-coated and poly-L-lysine (PLL)-coated TC-treated plastic 96-well plates; and COL1- and PLL-coated non-TC-treated plastic 96-well plates (not commercially available), were supplied by BD Biosciences.

Plasmid transfection. HMCs were transfected with pcDNA3 and pcDNA3-Smad7 (kind gifts from Dr. Peter ten Dijke, Leiden University Medical Center, Leiden, The Netherlands). Cells were transfected using the VPI-1004 Nucleofector Solution Kit (Amaxa, Cologne, Germany) and the A33 program, one of the Amaxa Basic Nucleofection Protocols for Primary Smooth Muscle Cells, according to the manufacturer's instructions. Transfection efficiency of ~80% was achieved as estimated by transfection of a control plasmid, pmaxGFP (Amaxa). Three additional Amaxa Basic Nucleofection Protocols for Primary Smooth Muscle Cells also led to high transfection efficiency (~80%), and another program had a transfection efficiency of ~30%.

Immunostaining. Cells were stained with 22 µM CellTracker Green CMFDA (5-chloromethylfluorescein diacetate, Invitrogen) at 37°C for 20 min and seeded into Lab-Tek eight-well glass chamber slides for experiments. At the end of the experiments, cells were fixed with methanol at –20°C, rinsed with PBS, and blocked with 1:5 normal donkey serum-PBS at room temperature for 20 min. Primary antibodies were added for 1 h at room temperature and included the following: non-immune rabbit and goat IgGs, rabbit anti-FN IgG (Sigma) and goat anti-COL1, anti-COL3, and anti-COL4 IgG (Southern Biotechnology Associates, Birmingham, AL). The final concentration of all rabbit IgG was 10 µg/ml and that for all goat IgGs was 27 µg/ml. After six 5-min washes with PBS, secondary antibody (1:500 Alexa Fluor 546-labeled donkey anti-rabbit IgG or anti-goat IgG, Invitrogen) was added at room temperature for 1 h. Coverslips were mounted in Mowiol (EMD Biosciences) containing 1 µg/ml 4'-6 diamidino-2-phenylindole (DAPI; Invitrogen), and the slides were stored at 4°C overnight before Z-stack immunofluorescence microscopic analysis.

Time-lapse microscopy. HMCs were seeded at 2.4 x 105 cells/well in a Lab-Tek II four-well chambered coverglass (Nunc) and cultured overnight. Medium was changed to fresh RPMI 1640 supplemented with 1% FBS, with or without TGF-beta1, and the wells were filled with sterile mineral oil to prevent evaporation. The cells were maintained at 37°C with a chamber heater (Bioptechs, Butler, PA) and visualized with a Zeiss Axiovert 200 inverted microscope equipped with a Hamamatsu Photonics (Hamamatsu City, Japan) digital charge-coupled camera. Images from the same field were captured every 4 min for up to 48 h using METAMORPH software (Universal Imaging, Downington, PA). Stacks of images for the entire time-lapse were analyzed to track the cells over time.

Picro-Sirius red staining and spectrophotometric analysis. Picro-Sirius red (0.1%) was purchased from Electronic Microscopy Sciences (Hatfield, PA). In brief, cells cultured in 8-well chamber slides or in 96-well plates were fixed in methanol overnight at –20°C, carefully washed once with PBS, and incubated in the staining solution at room temperature for 1 h. The staining solution was removed, and the cells were washed three times with 0.1% acetic acid (16). For photography, cells on chamber slides were dehydrated and clarified by three changes of 100% ethanol, 5 min each, followed by xylene, three changes, 10 min each, and coverslips were mounted with Permount (Electron Microscopy Sciences). For spectrophotometric analysis, picro-Sirius red was eluted in 0.1 N sodium hydroxide, 200 µl/well, the plates were placed on a rocking platform at room temperature for 1 h, and the optical density at 540 nm was determined using a Molecular Dynamics spectrophotometer (Sunnyvale, CA) (32).

Image analysis of nodule number and size. The air-dried 96-well plates with or without picro-Sirius red staining were subjected to Sorcerer colony counter (Perceptive Instruments, Suffolk, UK) analysis. The colony counter automatically calculated the number and average size of the nodules and generated a report of these data in Excel format (Microsoft, Redmond, WA).

Statistics. Data are presented as means ± SE. Group data were analyzed by ANOVA, with intergroup comparisons by Tukey's test, using Prism 4.0c software (GraphPad, San Diego, CA). A P value <0.05 was accepted as significant.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
To establish in vitro models of TGF-beta-induced fibrogenesis, we tested the effect of TGF-beta1 on both mesenchymal cells and epithelial cells cultured in eight-well chamber slides. These cells were either well-characterized primary cultures or established cell lines. We found that addition of TGF-beta1 for 24–48 h induced nodule formation in mesenchymal cells from the kidney (mesangial cells and fibroblasts), heart (fibroblasts), lung (fibroblasts), liver (stellate cells), spleen (fibroblasts), artery (smooth muscle cells), skin (fibroblasts), and mouse embryos (fibroblasts) on both glass and plastic surfaces (Fig. 1A). TGF-beta1-induced nodule formation in mesenchymal cells was generally rapid (occurring in <24 h), widespread throughout the culture, and characterized by an increase in nodule number and/or nodule size over time. Human and rat cardiac fibroblasts, for unknown reasons, were relatively refractory to TGF-beta1, with only a few nodules appearing after 48 h.


Figure 1
View larger version (77K):
[in this window]
[in a new window]

  		Fig. 1. Transforming growth factor (TGF)-beta1 induces nodule formation in mesenchymal and epithelial cells from kidney and other organs. A: TGF-beta1-induced nodule formation in human mesangial cells (HMCs) and mesenchymal cells from other organs. Mesenchymal cells were cultured in 8-well chamber slides, in the absence or presence of 10 ng/ml TGF-beta1 for 48 h. B: epithelial cells were seeded, pretreated, and exposed to 10 ng/ml TGF-beta1 as in A. C: TGF-beta1-induced nodule formation in HMCs was dose dependent.

 
Epithelial cells demonstrated considerable cell type-specific variability in their response to TGF-beta1, and the nodule formation in epithelial cells correlated well with 1) whether cells underwent morphological changes suggestive of EMT and 2) whether TGF-beta1 induced significant matrix accumulation. Cells that did not undergo significant EMT morphological changes and did not show significant matrix accumulation in response to TGF-beta1, such as keratinocytes, did not migrate into nodules; cells that underwent significant EMT morphological changes and had increased matrix accumulation in response to TGF-beta1, such as HK2 cells, form typical, widespread nodules; cells that underwent mild EMT morphological changes or expressed little matrix accumulation in response to TGF-beta1, such as podocytes, LLC-PK1 cells, and mouse tubular epithelial cells, only occasionally formed nodules or only formed loose cell aggregates. Madin-Darby canine kidney cells underwent cell morphological changes, including disappearance of domelike and tubule-like structures and enlargement of the cells before forming 3-dimensional (3D) nodule-like structures with morphological changes suggestive of increased cell death (Fig. 1B).

Since TGF-beta1-induced nodule formation was particularly robust and reproducible in HMCs and TGF-beta1 transgenic mice die of progressive glomerulosclerosis (23), in which mesangial cells play an important role, HMCs were selected for mechanistic studies and for refinement of a high-throughput screening assay. In HMCs, nodule formation was TGF-beta1 dose dependent, with effects seen at concentrations ≤1 ng/ml (Fig. 1C). It was also time dependent. Following the addition of TGF-beta1, we observed a striking increase in cell motility, beginning at ~6 h and progressing to nodule formation by 24–48 h (Supplementary video; all supplementary material can be found in the online version of this article on the AJP-Renal Physiology web site). Nodule formation is a highly dynamic process. Cells were seen to move from one nodule to another, and nodules themselves were also mobile and in some cases merged to form larger nodules.

TGF-beta1-induced fibrogenesis is mainly mediated by the TGF-beta type I receptor (Alk5)-Smad3 pathway (11, 25), and we found that TGF-beta1-induced nodule formation was mediated by a similar pathway. In HMCs, TGF-beta1-induced nodule formation was nearly completely blocked by transient transfection of Smad7, an endogenous inhibitor of Smad2/3 activity (Fig. 2A). Furthermore, in MEFs derived from Smad3 knockout mice, TGF-beta1-induced nodule formation was abrogated (Fig. 2B). In contrast, MEFs derived from Smad2 knockout mice developed nodules in response to TGF-beta1 in the same way as wild-type cells (Fig. 2B).


Figure 2
View larger version (87K):
[in this window]
[in a new window]

  		Fig. 2. TGF-beta1-induced nodule formation is Smad3 dependent. A: Smad7 transfection blocked TGF-beta1-induced nodule formation in HMCs. HMCs were transfected with pcDNA3 (control vector) or pcDNA3-Smad7 and exposed to TGF-beta1 for 48 h. B: knockout of Smad3, but not Smad2, completely blocked TGF-beta1-induced nodule formation in mouse embryonic fibroblasts (MEFs). Smad2+/+, Smad2–/–, Smad3+/+, and Smad–/– MEFs were treated with or without TGF-beta1 for 48 h.

 
We next characterized the composition of the TGF-beta1-induced nodules. Using both HMCs and NRK-49F cells cultured in eight-well chamber slides, we found that the nodules were composed of cells embedded in an ECM with abundant FN and collagen I (COL1). ECM components were visualized by picro-Sirius red staining under conventional light and polarized light, the latter specific for COL1 and COL3 (Fig. 3A) (16), and by immunofluorescence staining (Fig. 3B). Cells and their nuclei were visualized by CellTracker Green and DAPI staining, respectively (Fig. 3B). Immunofluorescence staining showed that the nodules contained abundant FN and COL1, with lesser amounts of COL3 and COL4.


Figure 3
View larger version (42K):
[in this window]
[in a new window]

  		Fig. 3. TGF-beta1-induced nodules contain abundant fibronectin (FN) and COL1 and are characterized by an increased COL1:4 ratio. A: picro-Sirius red staining of TGF-beta1-induced nodules. TGF-beta1-induced nodules in HMCs and NRK-49F rat renal fibroblasts were stained with picro-Sirius red and observed under an ordinary or polarized light microscope. Under normal light, picro-Sirius red stains all collagens, but visualizes only the interstitial collagens COL1 and COL3 under polarized light. B: immunostaining for extracellular matrix (ECM) proteins in TGF-beta1-induced nodules. CellTracker Green-stained HMCs were seeded in 8-well chamber slides and exposed to TGF-beta1 for 48 h. The nodules were immunostained with rabbit anti-FN IgG, goat anti-COL1, COL3, and COL4 IgGs, with rabbit IgG (normal control 1; NC1) and goat IgG (normal control 2; NC2) as negative controls, respectively. Shown are the expression of these ECM proteins (red), CellTracker Green-stained cytoplasm (green), 4',6-diamidino-2-phenylindole (DAPI)-stained nuclei (blue), and the merge of the triple-stained image. Merged images of triple-stained cells without TGF-beta1 treatment are shown as controls.

 
We hypothesized that TGF-beta1-dependent nodule formation involves increased cell migration and increased ECM production and furthermore that dynamic reciprocity between cells and ECM is critical to nodule formation. When HMCs were exposed to soluble COL1 (1.25–40 µg/ml) for 24 h in eight-well chamber slides, picro-Sirius red-positive collagen was visible on the cell layer and 3D cellular bundles or aggregates, and occasionally discrete nodules, were observed in the absence of exogenous TGF-beta1 (Fig. 4). In contrast, soluble COL3, COL4, FN, and LN (1.25–40 µg/ml) did not deposit on the cell monolayer, as detected by picro-Sirius red staining, and 3D structures were not observed (data not shown). However, when soluble COL1 was added together with equal concentrations of COL3, COL4, FN, and LN, deposition of visible 3D bundles and nodules was enhanced (data not shown). Combinations of these ECM proteins in concentrations as low as 0.25 µg/ml were also effective. These studies support a role for COL1 in nodule formation, and the effect of combinations of ECM proteins suggests that assembly of macromolecular ECM complexes may also contribute to nodule formation. The lack of effect of COL3, COL4, FN, or LN as individual components may be due to the use of denatured molecules or alternatively to limitations in matrix assembly in this in vitro system.


Figure 4
View larger version (109K):
[in this window]
[in a new window]

  		Fig. 4. Soluble COL1 induces nodule formation. Addition of 10 µg/mL soluble COL1 for 24 h to a monolayer of HMCs cultured on a 8-well chamber slide induced cell migration and formation of cellular aggregates and nodules in the absence of TGF-beta1. Phase contrast microscopy (top) and picro-Sirius red staining (bottom).

 
We next addressed the effect of anchoring HMCs to TC-treated plates, with or without individual ECM protein or PLL coating, on TGF-beta1-induced cell migration and nodule formation. In the negative control, non-TC-treated, uncoated plastic 96-well plates, cell attachment and cell spreading were limited, but TGF-beta1 induced nodules similar to those observed in eight-well chamber slides (Fig. 5A). This induction was largely prevented on TC-treated (by vacuum-ionized gas) surfaces (Fig. 5A) and completely blocked by further coating with individual ECM proteins (COL1, COL4, FN, or LN) (Fig. 5B). In contrast, inhibition of TGF-beta1-induced nodule formation by vacuum-ionized gas surface treatment was, at least in part, relieved by coating the TC-treated surface with PLL (Fig. 5A). The mechanisms underlying the inhibition and relief are not yet fully understood, but changes in electric charges on the plate surface and the different mechanisms of PLL, matrix proteins, and vacuum-ionized gas surface treatment to promote cell attachment (17) might play important roles. To further illustrate the impact of substrate changes on cell behaviors in different mesenchymal cells, we compared the HMC primary culture with HMCL, an immortalized human mesangial cell line, and NRK-49F, a rat renal fibroblast cell line, on non-TC and TC surfaces with or without COL1 and PLL coating. As shown in Supplementary Fig. 1, nodule formation was most pronounced in non-TC plates in all the three cell types and significant spontaneous nodules occurred in the HMCL cells on the non-TC plates. The nodule formation in all cell types was additively inhibited by plate COL1 coating and TC treatment. PLL coating also less potently suppressed nodule formation; however, its suppression is not additive to that of TC treatment. In contrast, TC treatment suppression of nodule formation was consistently relieved more or less by PLL coating in all cell types. Despite these common features in the three cell types, primary culture of HMCs and the HMCL cell line formed more nodules on PLL-coated TC-treated plates than on PLL-coated non-TC plates, whereas NRK-49F cells preferred PLL-coated non-TC, rather than PLL-coated TC plates, to form nodules after TGF-beta1 stimulation. These data suggest that different cell types may require similar but not exactly the same substrate conditions for TGF-beta1-induced nodule formation.


Figure 5
View larger version (90K):
[in this window]
[in a new window]

  		Fig. 5. Substrate effects on TGF-beta1-induced nodule formation in 96-well plates. HMCs were treated with or without 10 ng/ml TGF-beta1 for 48 h. A: negative control plates (untreated surface, coating "-"), tissue culture (TC)-treated plates (surface treated by vacuum ionized gas, coating "-"), poly-L-lysine (PLL)-coated TC-treated plates (surface treated by vacuum ionized gas, coated by PLL). B: matrix protein-coated TC-treated plates [surface treated by vacuum-ionized gas, coated by collagens type I and IV (COL1, COL4), laminin (LN), and FN, respectively]. Cells were fixed and stained with picro-Sirius red to detect TGF-beta1-induced collagen expression.

 
Importantly, the data shown in Fig. 5 and Supplementary Fig. 1 revealed two distinct but related models of TGF-beta1-induced ECM accumulation, the two-dimensional (2D) model on matrix-coated plates and the 3D model on non-tissue culture-treated or PLL-coated surfaces. As shown in Fig. 6A, due to the patchy accumulation of ECM in the cell monolayer in the 2D system, TGF-beta1-treated wells showed an increased picro-Sirius red staining (Fig. 6A), which can be quantitated spectrophotometrically after elution of the stain (Fig. 6B). Data are presented as either absolute optical density (540 nm) (Fig. 6B) or change in optical density compared with vehicle control (Fig. 7B). In contrast, due to disruption of the cell monolayer, TGF-beta1 stimulation of the cells in PLL-coated wells failed to increase the general staining density (Fig. 6C), but the nodule number (Figs. 6D and 7C), size (data not shown) can be analyzed automatically using a Sorcerer colony counter. We established that TGF-beta1-induced nodule formation did not occur in serum-free medium (Fig. 6D) and that 1–10% FBS supported nodule formation similarly (Fig. 6D). TGF-beta1-induced nodule formation was not dependent on a particular batch of FBS, as we tested eight different lots of FBS and they all supported nodule formation to a similar extent, despite a fivefold variation in their proliferative effect (Supplementary Table 1). To refine the nodule formation model for high-throughput assays, we compared TGF-beta1-induced nodule formation in different 96-well plates, including uncoated plastic, uncoated glass, tissue culture-treated plastic, ECM-coated plastic, and PLL-coated plastic, from different suppliers. PLL-coated 96-well plates were identified as optimal for the TGF-beta1-induced nodule formation, and FN-coated plates were shown to be the best for the TGF-beta1-induced 2D ECM accumulation.


Figure 6
View larger version (43K):
[in this window]
[in a new window]

  		Fig. 6. Quantification of TGF-beta1-induced ECM accumulation and nodule formation. A: macroscopic (left) and microscopic (right) images of picro-Sirius red staining of HMCs cultured in the presence and absence of TGF-beta1 in an ECM (FN)-coated TC-treated 96-well plate, as shown in Fig. 5B. B: spectrophotometric analysis of picro-Sirius red-stained cells in Fig. 5B; n = 3/group. C: macroscopic (left) and microscopic (right) images of picro-Sirius red staining of HMCs cultured in the presence and absence of TGF-beta1 in a PLL-coated TC-treated 96-well plate. D: quantification of TGF-beta1-induced nodules in a PLL-coated TC-treated 96-well plate. Subconfluent cells in a PLL-coated TC-treated 96-well plate were preincubated in 1% FBS medium for 24 h and then exposed to medium supplemented with 0, 1, 5 and 10% FBS, in the absence or presence of 10 ng/ml TGF-beta1 for 48 h. Images were analyzed with a Sorcerer Colony Counter; n = 3/group.

 

Figure 7
View larger version (18K):
[in this window]
[in a new window]

  		Fig. 7. In vitro model-based high-throughput screening of antifibrotic agents. A: proposed strategy for screening antifibrotic agents using PLL- and FN-coated TC-treated 96-well plates. B and C: tranilast, Calbiochem Alk5 inhibitor I, Calbiochem Alk5 inhibitor II, and the p38 MAP kinase/Alk5 inhibitor SB203580 showed antifibrotic activities in both assays; n = 8/group. ***P < 0.001 vs. the TGF-beta1-positive control.

 
Based on the above findings, we have proposed a strategy to screen compounds for antifibrotic activity by employing the two assays in parallel: the TGF-beta1-induced monolayer ECM accumulation assay in FN-coated plates and the TGF-beta1-induced nodule formation assay in PLL-coated plates (Fig. 7A). As a proof of principle, we tested four compounds known to alter TGF-beta signaling, including 2 Alk5 inhibitors (13, 38), which abrogated both nodule formation and monolayer ECM accumulation (Fig. 7, B and C). TGF-beta1-induced FN production is mediated by p38 MAP kinase independently of Smad2/3, whereas TGF-beta1-induced COL1, COL3, COL5, and COL6 production is Smad3 dependent (11, 25). SB203580 is reported to selectively suppress p38 MAP kinase at 10 µM and to also suppress Alk5 activity at higher concentrations (25). Interestingly, SB203580 suppressed TGF-beta1-induced nodule formation and monolayer ECM expression at both 10 and 25 µM (Fig. 7, B and C). Whether this is due to inhibition of Alk5 activity or due to suppression of TGF-beta1-stimulated p38 MAP kinase remains to be determined. Tranilast, a pharmaceutical compound that has been used in animal models of fibrotic disease (40) and in human subjects to treat keloids and to prevent vascular restenosis (33), abrogated both TGF-beta1-stimulated nodule formation and monolayer ECM accumulation (Fig. 7, B and C), at doses similar to those reported to have antiproliferative effects in leiomyoma cells (39).

The 2D and 3D model-based screening has been subjected to further refinement. For example, we have shown that 1) cells can be incubated in low-serum medium mixed with drugs and TGF-beta in certain orders, seeded in plates supporting 2D and 3D models of TGF-beta1-induced fibrogenesis, and then analyzed after a routine 48-h culture. Thus the preincubation in 10% FBS and subsequent wash and medium change to 1% FBS can be omitted (Supplementary Fig. 1); 2) for the colony analysis, if the cells are prestained with CellTracker Green, the nodules on PLL-coated plates can be analyzed using an Acumen Explorer (TTP Labtech, Cambridge, MA) that detects the green fluorescence of the cells and nodules (data not shown); 3) to further increase throughput, models can be established in 384-well plates (data not shown); 4) to address possible cytotoxic effects of any agents to be tested, additional experiments can be added to serve as controls for cell membrane integrity, by analyzing 50 µl of conditioned medium from each well using a CytoTox kit (Promega, Madison, WI) (data not shown). In both the 2D and 3D models, the cytotoxic assay does not affect subsequent Sirius red staining, 540-nm optical density measurement, and nodule analysis.


   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
We have developed two in vitro models of fibrogenesis induced by TGF-beta, featuring 2D (patchy) or 3D (nodular) accumulation of ECM. These models are distinct in morphology but are both characterized by discontinuous accumulation of ECM (Fig. 5), which is probably cell cycle dependent since quiescent cells in serum-free medium were refractory to TGF-beta1 (Fig. 6D). We have further shown that the 2D model is suitable to assess total ECM accumulation while the 3D model represents a biological consequence of TGF-beta1-induced ECM accumulation and mimics the disruption of normal tissue architecture in fibrosis. On the basis of these models, we have established complementary, fast, and robust assays suitable for high-throughput screening of new antifibrotic drugs. In contrast to existing molecular assays that can only be used to screen drugs targeting a single molecule or pathway (26, 45), these innovative assays can be used to identify antifibrotic drugs targeting known or novel TGF-beta-induced profibrotic signaling pathways or common pathways of fibrogenesis. It is possible to induce the TGF-beta response and then test compounds for the capacity to reverse the changes so that antifibrotic drugs that may reverse fibrosis can be identified (7).

By combining the 2D and 3D models for screening purposes, theoretically we can expect to identify three possible categories of agents: 1) those that prevent both ECM accumulation and nodule formation; 2) those that prevent cell migration and nodule formation without suppressing ECM expression; and 3) those that suppress ECM expression but fail to prevent cell migration and nodule formation. Although we expect that the first category will have the most potent antifibrotic actions, the second and third categories may also be of potential interest. If they exist, it may imply that it is not only the amount, but also the quality, of ECM that contributes to cell migration and nodule formation. It will be important to test these agents further and to establish whether inhibition of cell migration may serve as an independent strategy to preserve tissue structure in fibrosis.

Spontaneous nodule formation in prolonged in vitro culture has been previously reported in mesangial cells (15, 22, 42), smooth muscle cells (14, 30), and renal fibroblasts (31), and it has been reported that insulin (1), heparin (2), antioxidants (21), and macrophages (29) enhance nodule formation in mesangial cell cultures. It is controversial whether nodule formation is an in vitro model of physiological matrix assembly (22) or of pathological fibrogenesis (1, 14, 30, 31). Our report is the first to show that TGF-beta1 is a potent inducer of nodule formation. We propose that TGF-beta1-induced nodule formation is an in vitro model of TGF-beta1-induced fibrogenesis, probably representing a biological consequence of TGF-beta1-induced ECM synthesis and accumulation. Evidence supporting this interpretation of nodule formation includes the following: 1) it is characterized by increased COL1 and FN accumulation, similar to fibrotic tissues in vivo; 2) it is mediated by the profibrotic Alk5/Smad3 pathway; 3) disruption of the cell monolayer in this model resembles the disruption of tissue structure in fibrotic tissues in vivo; and 4) deposition of COL1 on the cell monolayer promoted nodule formation.

In general, mesenchymal cells are more sensitive to TGF-beta1 than epithelial cells in terms of nodule formation. This is consistent with the more direct contribution of mesenchymal cells to fibrotic and sclerotic diseases. A significant variability of nodule inducibility was also observed in both mesenchymal cells and epithelial cells from different organs. For example, of all the mesenchymal cells tested, glomerular mesangial cells and renal fibroblasts showed the strongest response to TGF-beta1 in terms of nodule formation while rat and human cardiac fibroblasts were most resistant. This is consistent with the in vivo sensitivity of the two organs to TGF-beta1 in Alb/TGF-beta1 transgenic mice. At 5 wk of age, around one-half transgenic mice had end-stage renal failure while no obvious heart fibrosis was observed (Xu Q, Lu H, Kopp JB, unpublished observation). It is intriguing to speculate that normal cardiac fibroblasts have a unique antifibrotic mechanism compared with other mesenchymal cells. Among the epithelial cells tested, HK2 human renal proximal tubular epithelial cells were most sensitive to TGF-beta1-induced nodule formation. This is consistent with an important role of renal proximal tubular epithelial cells in renal tubulointerstitial fibrosis (24, 34), although we cannot exclude the possibility that the HK-2 cells we used had developed to some degree of EMT compared with other tubular epithelial cells tested, including LLC-PK1 and mouse tubular epithelial cells.

In conclusion, we have developed two complementary in vitro models of TGF-beta1-induced fibrosis and have established novel assays to quantitate the models for high-throughput screening of inflammation-independent antifibrotic activity. Since mesenchymal cells from different organs and some epithelial cells showed a similar response in these models, antifibrotic drugs identified by these assays may have a broad application in combating fibrotic diseases. In support of this, Alk5 inhibitors and tranilast, which show excellent antifibrotic activities in the in vitro models, prevent not only renal fibrosis, but also fibrotic and sclerotic lesions of the liver, lung, and blood vessels (9, 12, 33, 43).


   GRANTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
Funding for this work was provided by the Intramural Research Program of the National Institute of Diabetes and Digestive and Kidney Diseases, and Q. Xu was supported by the British Heart Foundation, a Royal Free Peter Samuel Grant, a University College London (UCL) Bogue Fellowship, a UCL Value in People Award, and a Kidney Research UK Grant.


   ACKNOWLEDGMENTS
 
We gratefully acknowledge the assistance of Dr. Elizabeth Geras-Raaka with videomicroscopy and other laboratory assistance provided by Dr. Hiroshi Kajiyama and Dr. Takayuki Matsumoto. We thank the following individuals for providing cells: Dr. Kathy Flanders (wild-type, Smad2, and Smad3 knockout MEFs), Dr. Esteban Mezey (mouse hepatic stellate cells), Dr. Carol Yee (human foreskin fibroblasts and keratinocytes), Dr. Ying Hong (pig arterial smooth muscle cells), Dr. Robin McAnulty (human lung fibroblasts), Dr. Gisela Lindahl (human and rat cardiac fibroblasts), and Dr. Moin A Saleem (human podocytes). We also thank Dr. Peter ten Dijke for sharing the pcDNA3-Smad7 plasmid and Dr. George Martin for a critical review of the manuscript.


   FOOTNOTES
 

Address for reprint requests and other correspondence: J. Kopp, Rm. 3N116, NIDDK, NIH, 10 Center Dr., Bethesda, MD 20892-1268 (e-mail: jbkopp{at}nih.gov)

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
 GRANTS
 REFERENCES
 

  1. Abrass CK, Spicer D, Raugi GJ. Induction of nodular sclerosis by insulin in rat mesangial cells in vitro: studies of collagen. Kidney Int 47: 25–37, 1995.[ISI][Medline]
  2. Ayo SH, Kreisberg JI. Heparin increases hillock formation in mesangial cell cultures. J Am Soc Nephrol 2: 1153–1157, 1991.[Abstract]
  3. Branton MH, Kopp JB. TGF-beta and fibrosis. Microbes Infect 1: 1349–1365, 1999.[CrossRef][ISI][Medline]
  4. Breitkopf K, Haas S, Wiercinska E, Singer MV, Dooley S. Anti-TGF-beta strategies for the treatment of chronic liver disease. Alcohol Clin Exp Res 29: 121S-131S, 2005.[CrossRef][ISI][Medline]
  5. Brown KK, Raghu G. Medical treatment for pulmonary fibrosis: current trends, concepts, and prospects. Clin Chest Med 25: 759–772, 2004.[CrossRef][ISI][Medline]
  6. Butcher EC, Berg EL, Kunkel EJ. Systems biology in drug discovery. Nat Biotechnol 22: 1253–1259, 2004.[CrossRef][ISI][Medline]
  7. Chatziantoniou C, Dussaule JC. Insights into the mechanisms of renal fibrosis: is it possible to achieve regression? Am J Physiol Renal Physiol 289: F227–F234, 2005.[Abstract/Free Full Text]
  8. Clouthier DE, Comerford SA, Hammer RE. Hepatic fibrosis, glomerulosclerosis, and a lipodystrophy-like syndrome in PEPCK-TGF-beta1 transgenic mice. J Clin Invest 100: 2697–2713, 1997.[ISI][Medline]
  9. de Gouville AC, Boullay V, Krysa G, Pilot J, Brusq JM, Loriolle F, Gauthier JM, Papworth SA, Laroze A, Gellibert F, Huet S. Inhibition of TGF-beta signaling by an ALK5 inhibitor protects rats from dimethylnitrosamine-induced liver fibrosis. Br J Pharmacol 145: 166–177, 2005.[CrossRef][ISI][Medline]
  10. Eitner F, Floege J. Therapeutic targets for prevention and regression of progressive fibrosing renal diseases. Curr Opin Investig Drugs 6: 255–261, 2005.[Medline]
  11. Flanders KC. Smad3 as a mediator of the fibrotic response. Int J Exp Pathol 85: 47–64, 2004.[CrossRef][ISI][Medline]
  12. Gellibert F, de Gouville AC, Woolven J, Mathews N, Nguyen VL, Bertho-Ruault C, Patikis A, Grygielko ET, Laping NJ, Huet S. Discovery of 4-{4-[3-(pyridin-2-yl)-1H-pyrazol-4-yl]pyridin-2-yl}-N- (tetrahydro-2H- pyran-4-yl)benzamide (GW788388): a potent, selective, and orally active transforming growth factor-beta type I receptor inhibitor. J Med Chem 49: 2210–2221, 2006.[CrossRef][ISI][Medline]
  13. Gellibert F, Woolven J, Fouchet MH, Mathews N, Goodland H, Lovegrove V, Laroze A, Nguyen VL, Sautet S, Wang R, Janson C, Smith W, Krysa G, Boullay V, De Gouville AC, Huet S, Hartley D. Identification of 1,5-naphthyridine derivatives as a novel series of potent and selective TGF-beta type I receptor inhibitors. J Med Chem 47: 4494–4506, 2004.[CrossRef][ISI][Medline]
  14. Gimbrone MA Jr, Cotran RS. Human vascular smooth muscle in culture. Growth and ultrastructure. Lab Invest 33: 16–27, 1975.[ISI][Medline]
  15. Glass WF 2nd, Teng PR, Haney LB. Extracellular matrix distribution and hillock formation in human mesangial cells in culture without serum. J Am Soc Nephrol 7: 2230–2243, 1996.[Abstract]
  16. Grimm PC, Nickerson P, Gough J, McKenna R, Stern E, Jeffery J, Rush DN. Computerized image analysis of Sirius Red-stained renal allograft biopsies as a surrogate marker to predict long-term allograft function. J Am Soc Nephrol 14: 1662–1668. 2003.[Abstract/Free Full Text]
  17. Harper PA, Juliano RL. Isolation and characterization of Chinese hamster ovary cell variants defective in adhesion to fibronectin-coated collagen. J Cell Biol 87: 755–763, 1980.[Abstract/Free Full Text]
  18. Heeneman S, Cleutjens JP, Faber BC, Creemers EE, van Suylen RJ, Lutgens E, Cleutjens KB, Daemen MJ. The dynamic extracellular matrix: intervention strategies during heart failure and atherosclerosis. J Pathol 200: 516–525, 2003.[CrossRef][ISI][Medline]
  19. Ihn H. Scleroderma, fibroblasts, signaling, and excessive extracellular matrix. Curr Rheumatol Rep 7: 156–162, 2005.[CrossRef][Medline]
  20. Jackson PD, Harrington JJ. High-throughput target discovery using cell-based genetics. Drug Discov Today 10: 53–60, 2005.[CrossRef][ISI][Medline]
  21. Kitamura M. The antioxidant N-acetylcysteine induces mesangial cells to create three-dimensional cytoarchitecture that underlies cellular differentiation. J Am Soc Nephrol 10: 746–751, 1999.[Abstract/Free Full Text]
  22. Kitamura M, Mitarai T, Nagasawa R, Maruyama N. Differentiated phenotype of glomerular mesangial cells in nodular culture. Am J Physiol Renal Fluid Electrolyte Physiol 270: F614–F622, 1996.[Abstract/Free Full Text]
  23. Kopp JB, Factor VM, Mozes M, Nagy P, Sanderson N, Bottinger EP, Klotman PE, Thorgeirsson SS. Transgenic mice with increased plasma levels of TGF-beta 1 develop progressive renal disease. Lab Invest 74: 991–1003, 1996.[ISI][Medline]
  24. Lan HY. Tubular epithelial-myofibroblast transdifferentiation mechanisms in proximal tubule cells. Curr Opin Nephrol Hypertens 12: 25–29, 2003.[CrossRef][ISI][Medline]
  25. Laping NJ, Grygielko E, Mathur A, Butter S, Bomberger J, Tweed C, Martin W, Fornwald J, Lehr R, Harling J, Gaster L, Callahan JF, Olson BA. Inhibition of transforming growth factor (TGF)-beta1-induced extracellular matrix with a novel inhibitor of the TGF-beta type I receptor kinase activity: SB-431542. Mol Pharmacol 62: 58–64, 2002.[Abstract/Free Full Text]
  26. Liu XJ, Ruan CM, Gong XF, Li XZ, Wang HL, Wang MW, Yin JQ. Antagonism of transforming growth factor-Beta signaling inhibits fibrosis-related genes. Biotechnol Lett 27: 1609–1615, 2005.[CrossRef][ISI][Medline]
  27. Liu Y. Renal fibrosis: new insights into the pathogenesis and therapeutics. Kidney Int 69: 213–217, 2006.[CrossRef][ISI][Medline]
  28. Manzano VM, Munoz JC, Jimenez JR, Puyol MR, Puyol DR, Kitamura M, Cazana FJ. Human renal mesangial cells are a target for the anti-inflammatory action of 9-cis retinoic acid. Br J Pharmacol 131: 1673–1683, 2000.[CrossRef][ISI]
  29. Mattana J, Jaffer S, Singhal PC. Effects of macrophage supernatants on mesangial cell migration and hillock formation. Proc Soc Exp Biol Med 208: 271–276, 1995.[Abstract]
  30. May JF, Paule WJ, Rounds DE, Blankenhorn DH, Zemplenyi T. The induction of atherosclerotic plaque-like mounds in cultures of aortic smooth muscle cells. Virchows Arch B Cell Pathol 18: 205–211, 1975.[ISI][Medline]
  31. Muller GA, Rodemann HP. Characterization of human renal fibroblasts in health and disease. I. Immunophenotyping of cultured tubular epithelial cells and fibroblasts derived from kidneys with histologically proven interstitial fibrosis. Am J Kidney Dis 17: 680–683, 1991.[ISI][Medline]
  32. Pancre V, Delacre M, Herno J, Auriault C. Schistosomal egg antigen-responsive CD8 T-cell population in Schistosoma mansoni-infected BALB/c mice. Immunology 98: 525–534, 1999.[CrossRef][ISI][Medline]
  33. Pfab T, Hocher B. Tranilast and hypertensive heart disease: further insights into mechanisms of an anti-inflammatory and anti-fibrotic drug. J Hypertens 22: 883–886, 2004.[CrossRef][ISI][Medline]
  34. Phillips AO, Steadman R. Diabetic nephropathy: the central role of renal proximal tubular cells in tubulointerstitial injury. Histol Histopathol 17: 247–252, 2002.[ISI][Medline]
  35. Ruan XZ, Varghese Z, Powis SH, Moorhead JF. Dysregulation of LDL receptor under the influence of inflammatory cytokines: a new pathway for foam cell formation. Kidney Int 60: 1716–1725, 2001.[CrossRef][ISI][Medline]
  36. Saleem MA, O'Hare MJ, Reiser J, Coward RJ, Inward CD, Farren T, Xing CY, Ni L, Mathieson PW, Mundel P. A conditionally immortalized human podocyte cell line demonstrating nephrin and podocin expression. J Am Soc Nephrol 13: 630–638, 2002.[Abstract/Free Full Text]
  37. Sanderson N, Factor V, Nagy P, Kopp J, Kondaiah P, Wakefield L, Roberts AB, Sporn MB, Thorgeirsson SS. Hepatic expression of mature transforming growth factor beta 1 in transgenic mice results in multiple tissue lesions. Proc Natl Acad Sci USA 92: 2572–2576, 1995.[Abstract/Free Full Text]
  38. Sawyer JS, Anderson BD, Beight DW, Campbell RM, Jones ML, Herron DK, Lampe JW, McCowan JR, McMillen WT, Mort N, Parsons S, Smith EC, Vieth M, Weir LC, Yan L, Zhang F, Yingling JM. Synthesis and activity of new aryl- and heteroaryl-substituted pyrazole inhibitors of the transforming growth factor-beta type I receptor kinase domain. J Med Chem 46: 3953–3956, 2003.[CrossRef][ISI][Medline]
  39. Shime H, Kariya M, Orii A, Momma C, Kanamori T, Fukuhara K, Kusakari T, Tsuruta Y, Takakura K, Nikaido T, Fujii S. Tranilast inhibits the proliferation of uterine leiomyoma cells in vitro through G1 arrest associated with the induction of p21(waf1) and p53. J Clin Endocrinol Metab 87: 5610–5617, 2002.[Abstract/Free Full Text]
  40. Soma J, Sugawara T, Huang YD, Nakajima J, Kawamura M. Tranilast slows the progression of advanced diabetic nephropathy. Nephron 92: 693–698, 2002.[CrossRef][ISI][Medline]
  41. Sraer JD, Delarue F, Hagege J, Feunteun J, Pinet F, Nguyen G, Rondeau E. Stable cell lines of T-SV40 immortalized human glomerular mesangial cells. Kidney Int 49: 267–270, 1996.[ISI][Medline]
  42. Sterzel RB, Lovett DH, Foellmer HG, Perfetto M, Biemesderfer D, Kashgarian M. Mesangial cell hillocks. Nodular foci of exaggerated growth of cells and matrix in prolonged culture. Am J Pathol 125: 130–140, 1986.[Abstract]
  43. Tamai H, Katoh K, Yamaguchi T, Hayakawa H, Kanmatsuse K, Haze K, Aizawa T, Nakanishi S, Suzuki S, Suzuki T, Takase S, Nishikawa H, Katoh O. The impact of tranilast on restenosis after coronary angioplasty: the Second Tranilast Restenosis Following Angioplasty Trial (TREAT-2). Am Heart J 143: 506–513, 2002.[CrossRef][ISI][Medline]
  44. Wynn TA. Fibrotic disease and the T(H)1/T(H)2 paradigm. Nat Rev Immunol 4: 583–594, 2004.[CrossRef][ISI][Medline]
  45. Yingling JM, Blanchard KL, Sawyer JS. Development of TGF-beta signalling inhibitors for cancer therapy. Nat Rev Drug Discov 3: 1011–1022, 2004.[CrossRef][ISI][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Figure and Video
Right arrow All Versions of this Article:
293/2/F631    most recent
00379.2006v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Xu, Q.
Right arrow Articles by Kopp, J. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Xu, Q.
Right arrow Articles by Kopp, J. B.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2007 by the American Physiological Society.