Trichostatin A prevents the accumulation of extracellular matrix in a mouse model of bleomycin-induced skin fibrosis.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 56, No. 8, August 2007, pp 2755–2764 DOI 10.1002/art.22759 © 2007, American College of Rheumatology Trichostatin A Prevents the Accumulation of Extracellular Matrix in a Mouse Model of Bleomycin-Induced Skin Fibrosis Lars C. Huber,1 Jörg H. W. Distler,2 Falk Moritz,1 Hossein Hemmatazad,1 Thomas Hauser,3 Beat A. Michel,1 Renate E. Gay,1 Marco Matucci-Cerinic,4 Steffen Gay,1 Oliver Distler,1 and Astrid Jüngel1 Objective. Tissue fibrosis is a hallmark compromising feature of many disorders. In this study, we investigated the antifibrogenic effects of the histone deacetylase inhibitor trichostatin A (TSA) on cytokinedriven fibrotic responses in vitro and in vivo. Methods. Skin fibroblasts from patients with systemic sclerosis (SSc) and normal healthy control subjects were stimulated with profibrotic cytokines in combination with TSA. Human Col␣1(I) and fibronectin were measured using real-time polymerase chain reaction, and levels of soluble collagen were estimated using the SirCol collagen assay. Electromobilty shift assay and confocal fluorescence microscopy were used to investigate the intracellular distribution of Smad transcription factors. For in vivo analysis, skin fibrosis was quantified by histologic assessment of mouse skin tissue in a model of bleomycin-induced fibrosis. Results. Reductions in the cytokine-induced transcription of Col␣1(I) and fibronectin were observed in both normal and SSc skin fibroblasts following the addition of TSA. Similarly, the expression of total collagen protein in TSA-stimulated SSc skin fibroblasts was reduced to basal levels. The mechanism of action of TSA included inhibition of the nuclear translocation and DNA binding of profibrotic Smad transcription factors. Western blot analysis revealed an up-regulation of the cell cycle inhibitor p21 by TSA, leading to reduced proliferation of fibroblasts. In addition, in bleomycininduced fibrosis in mice, TSA prevented dermal accumulation of extracellular matrix in vivo. Conclusion. These findings provide novel insights into the epigenetic regulation of fibrosis. TSA and similar inhibitory compounds appear to represent early therapeutic strategies for achieving reversal of the cytokine-driven induction of matrix synthesis that leads to fibrosis. Systemic sclerosis (SSc) is characterized by severe fibrosis of the skin and various internal organs. SSc skin fibroblasts show distinct features of a dysregulated phenotype in vitro and in vivo, leading ultimately to the production of high levels of collagen and other proteins of the extracellular matrix as well as a reduced expression of matrix-degrading enzymes. However, the mechanisms responsible for the activation of fibroblasts, and thus the development of chronic and progressive fibrotic disease, remain unclear (for review, see ref. 1). Several profibrotic cytokines have been shown to be strongly associated with the pathogenesis of SSc, including transforming growth factor ␤ (TGF␤) (2), interleukin-4 (IL-4) (3), and platelet-derived growth factor (PDGF) (4). Moreover, alterations in the downstream signaling pathways of TGF␤ have been observed, in particular, an increase in the expression of phosphorylated Smad transcription factors (5,6). These cellular alterations have been reported to be stable over multiple generations of SSc fibroblasts in vitro. Genetic studies, in contrast, could not identify clear associations between specific mutations of the genome and the Supported in part by the Swiss National Science Foundation (SNF 3200B0-103691). 1 Lars C. Huber, MD, Falk Moritz, MD, Hossein Hemmatazad, MD, Beat A. Michel, MD, Renate E. Gay, MD, Steffen Gay, MD, Oliver Distler, MD, Astrid Jüngel, PhD: University Hospital Zurich, and Zurich Center for Integrative Human Physiology, Zurich, Switzerland; 2Jörg H. W. Distler, MD: University of ErlangenNuremberg, Erlangen, Germany, and University Hospital Zurich, Zurich, Switzerland; 3Thomas Hauser, MD, University Hospital Zurich, Zurich, Switzerland; 4Marco Matucci-Cerinic, MD, PhD: University of Florence, Florence, Italy. Address correspondence and reprint requests to Lars C. Huber, MD, Center of Experimental Rheumatism, University Hospital Zurich, Gloriastrasse 25, Zurich 8091, Switzerland. E-mail: Lars.Huber@usz.ch. Submitted for publication December 21, 2006; accepted in revised form April 20, 2007. 2755 2756 excessive accumulation of extracellular matrix (7). The profibrotic phenotype of SSc fibroblasts could therefore be imprinted by a combination of epigenetic alterations, such as methylation and acetylation, as has been recently suggested (8). Epigenetic modifications comprise heritable alterations in the DNA itself without any changes in the nucleotide sequence. Unlike alterations of the genome, epigenetic changes are reversible and offer the potential opportunity to reverse the epigenetic pattern through therapeutic strategies (9). Modifications of histones, in particular, histone (de)acetylation, are among the principal mechanisms that have been described as epigenetic changes. In normal resting cells, DNA is highly organized within nucleosomes, in an octomeric core unit of chromatin and in DNA-binding nucleoproteins (histones) (10). Thereby, the dynamic packaging of chromatin regulates the extent of gene transcription. Histone hyperacetylation is generally associated with a loosened state of chromatin and increased rates of gene transcription. Conversely, deacetylation of histones causes tighter wrapping of the DNA around the nucleosome and prevents transcription factors and RNA polymerase II from binding (11). Targeted deacetylation is induced by histone deacetylases (HDACs), which are multisubunit enzyme complexes that remove the acetyl group from the histones via a charge-relay system, using zinc ions as prosthetic groups (for review, see ref. 12). HDAC inhibitors function by dislodging the zinc ion, thus turning off the charge-relay system. The most potent reversible HDAC inhibitor known is trichostatin A (TSA), which fits exactly into the catalytic site of the HDAC enzymes (13). During recent years, TSA has been used in clinical trials as a novel therapeutic strategy for various cancers, and these studies have shown that TSA induces cell cycle arrest, cell differentiation, and apoptotic cell death (14–16). Of interest, TSA was also postulated as a lead compound in the development of antifibrogenic drugs (17), and was tested as a promising therapeutic agent in hepatic fibrosis (18) and in the prevention of cutaneous radiation syndrome (19). Fibrotic diseases comprise a wide array of clinical entities. In this regard, novel molecules blocking the excessive accumulation of extracellular matrix are of great interest in the development of therapeutic strategies. In the present study, we investigated the antifibrogenic effects of TSA, both in human SSc skin fibroblasts in vitro and in an animal model of skin fibrosis in vivo. HUBER ET AL PATIENTS AND METHODS Patients and fibroblast cultures. Normal and SSc fibroblast cultures were prepared from biopsy specimens that were obtained from the skin of patients with SSc (n ⫽ 6) and healthy control subjects (n ⫽ 3) at the University of Florence. Biopsy samples from the SSc patients were from areas of affected skin. All patients fulfilled the criteria for SSc as suggested by LeRoy et al (20), and all subjects signed a consent form that was approved by the institutional review board of the University of Florence. After enzymatic digestion of the skin biopsy specimens with dispase II (Boehringer-Mannheim, Rotkreuz, Switzerland), cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) containing 10% heat-inactivated fetal calf serum (FCS), 25 mM HEPES, 100 units/ml penicillin, 100 g/ml streptomycin, 2 mM L-glutamine, and 2.5 g/ml amphotericin B (all from Gibco BRL, Basel, Switzerland). Fibroblasts from passages 4–8 were used for these experiments. Cells were treated with TSA (Sigma, St. Louis, MO) in a concentration of 2 M (21). A stock solution of TSA was prepared in ethanol and stored at ⫺80°C. The final concentration of ethanol in the medium was 0.06%. Real-time reverse transcription–polymerase chain reaction (RT-PCR). Total RNA was isolated from SSc fibroblasts using the RNeasy kit (Qiagen, Basel, Switzerland) according to the manufacturer’s instructions, which included DNase treatment. To generate complementary DNA (cDNA), total RNA (300–500 ng) was subjected to RT using Moloney murine leukemia virus reverse transcriptase (2.5 units/l), random hexamers (2.5 M), dNTPs (2 mM each), and RNase inhibitor (1 unit/l) (all from Applied Biosystems, Rotkreuz, Switzerland). The RT reaction was performed in a total volume of 20 l in a GeneAmp PCR cycler (Applied Biosystems) at 25°C for 10 minutes, followed by 30 minutes at 48°C and then 5 minutes at 95°C. The cDNA was stored at ⫺20°C until analyzed further. Samples without enzyme were used as negative controls in the RT reaction, to exclude genomic contamination. Quantification of specific messenger RNA (mRNA) was performed by TaqMan single-reporter and SYBR Green real-time PCRs, using the ABI Prism 7700 Sequence Detection System (Applied Biosystems) as described previously (22). SYBR Green real-time PCR was performed for human Col␣1(I) (forward primer 5⬘-TCAAGAGAAGGCTCACGATGG-3⬘, reverse primer 5⬘-TCACGGTCACGAACCACATT3⬘) and human fibronectin 1 (forward primer 5⬘-TTCTAAGATTTGGTTTGGGATCAAT-3⬘, reverse primer 5⬘-TCTTGGTTGGCTGCATATGC-3⬘). To confirm specific amplification by the SYBR Green PCR, a dissociation curve analysis was performed for each primer pair, and both non-RT negative controls and water controls were used for these analyses. The amounts of loaded cDNA were normalized using a predeveloped 18S ribosomal RNA control kit (Applied Biosystems) as an endogenous control. Differential gene expression was calculated with the threshold cycle (Ct), and relative quantification was calculated with the comparative Ct method. Only samples with a difference of at least 4 cycles between the signals in cDNA samples and negative controls (corresponding to a 24 [16-fold] difference in expression) were considered for the calculations. All experiments were performed at least in duplicate. TRICHOSTATIN A IN SKIN FIBROSIS Measurement of collagen. Total soluble collagen in cell culture supernatants was quantified using the SirCol collagen assay (Biocolor, Belfast, Northern Ireland). For these experiments, confluent cells were incubated for 24 hours with 40 l DMEM/5% FCS per cm2 of culture dish surface. One milliliter of Sirius red dye, an anionic dye that reacts specifically with basic side-chain groups of collagens under assay conditions, was added to 400 l supernatant, followed by incubation under gentle rotation for 30 minutes at room temperature. After centrifugation at 12,000g for 10 minutes, the collagen-bound dye was redissolved with 1 ml of 0.5M NaOH, and the absorbance was measured at 540 nm in an MRX enzymelinked immunosorbent assay reader (Dynex Technologies, Chantilly, VA). Microtiter tetrazolium (MTS) assay. SSc and normal dermal fibroblasts were incubated with TSA (2 M) in 96-well plates for 48–72 hours. Cell proliferation was analyzed using an MTS assay (CellTiter96 AQueous Cell Proliferation Assay; Promega, Wallisellen, Switzerland) according to the manufacturer’s instructions. For this analysis, an MRX microplate reader (Dynex Technologies) was used, at a test wavelength of 490 nm. Untreated fibroblasts were used as controls. Western blot analysis for p21. Induction of the cell cycle inhibitor p21WAF1/Cip1 by TSA was analyzed by Western blotting. SSc skin fibroblasts (2 ⫻ 106/well) were incubated in the absence or presence of TSA (2 M) for 48 hours and lysed in Laemmli buffer. Protein extracts from SSc fibroblasts were analyzed by electrophoresis on 10% sodium dodecyl sulfate– polyacrylamide gels, and transferred onto a nitrocellulose membrane. The membrane was blocked for 1 hour with Tris buffered saline (TBS) containing 0.1% Tween 20 (TBST) and 5% dehydrated skim milk. Blots were then incubated overnight at 4°C in the presence of monoclonal antibodies against p21WAF1/Cip1 (Cell Signaling Technology, Beverly, MA) or ␣-tubulin (Sigma). The blots were washed 3 times with TBST for 15 minutes each, and incubated with horseradish peroxidase–conjugated goat anti-mouse IgG secondary antibodies at room temperature for 1 hour. Signals were detected with enhanced chemiluminescence Western blot detection reagents (Amersham Bioscience, Freiburg, Germany) and exposed to radiographic films (SuperRX; Fujifilm Medical Systems, Stamford, CT). Electrophoretic mobility shift assay (EMSA). SSc skin fibroblasts were cultured to confluence in cell culture flasks (225 cm2) and incubated with medium only or with TSA (2 M) for 48 hours. TGF␤ (5 ng/ml; R&D Systems, Abingdon, UK) was then added for 60 minutes. Cells were collected by scratching in ice-cold phosphate buffered saline (PBS). Nuclear extracts were prepared according to the protocol described by Andrews and Faller (23). The concentration of nuclear protein was determined using the bicinchoninic acid protein assay reagent kit (Pierce, Rockford, IL), with normalization for the amount of protein within each experiment. Nonradioactive EMSA was performed using an EMSA kit (Panomics, Redwood City, CA) according to the manufacturer’s instructions. Three micrograms of nuclear protein was incubated for 30 minutes at 21°C with biotinylated oligonucleotides containing the Smad3/4 binding site (Panomics). As a negative control, the binding reaction was performed in the presence of an excess of unlabeled double-stranded oligonucleotide. The samples were electrophoretically separated 2757 (120V for 1.5–2 hours) in a nondenaturing polyacrylamide gel (6% with 2.5% glycerol) and blotted (300 mA for 30–40 minutes) on a Biodyne B (0.45 m) positively charged nylon membrane (Pall, Basel, Switzerland). The transfer buffer contained 20% methanol, 0.27M Tris, and 2M glycine. After transfer, the membrane was ultraviolet (UV)– crosslinked at 254 nm for 3 minutes, using a Stratalinker UV crosslinker (Stratagene, La Jolla, CA). Biotin was labeled with alkaline phosphatase–conjugated streptavidin (1:1,000; Dako, Glostrup, Denmark), and streptavidin was detected with CDP-Star substrate (Applied Biosystems) according to the manufacturer’s instructions. Chemiluminescence signals were visualized by exposing the membrane to an Agfa Curix Ortho HT-A film (Agfa-Gevaert, Kontich, Belgium) for 2 minutes (24). Confocal microscopy for Smad3. For confocal fluorescence microscopy, SSc skin fibroblasts (5,000/well) were grown overnight in 24-well plates containing round coverslips (Hecht Assistant, Altnau, Switzerland). Fresh medium with 10% FCS, containing TGF␤ (5 ng/ml) and/or TSA (2 M), was then added. After incubation at 37°C for 1–2 hours, SSc skin fibroblasts were fixed with methanol and the nuclei were stained with DAPI. The intracellular localization of Smad transcription factors was investigated by confocal fluorescence microscopy. Cells were incubated with polyclonal goat antibodies against Smad3 (Santa Cruz Biotechnology, Santa Cruz, CA) or with control antibodies, followed by incubation with fluorescein isothiocyanate (FITC)–labeled secondary rabbit anti-goat antibodies (Dako). Confocal fluorescence microscopy was performed using a Leica SP2 inverted microscope (Leica, Wetzlar, Germany). Bleomycin-induced dermal fibrosis. Skin fibrosis was induced in 6–8-week-old, pathogen-free female C3H/HeJ mice (Sankyo, Tokyo, Japan) by local injection of bleomycin for 24–28 days (25,26). One hundred microliters of bleomycin dissolved in PBS at a concentration of 0.5 mg/ml was administered every other day by subcutaneous (SC) injection in defined areas of the upper back. SC injection of 100 l PBS was used as the control. Two subgroups of mice were treated with bleomycin followed by TSA in different concentrations. TSA was dissolved in PBS/1% DMSO and administered every other day by intraperitoneal (IP) injection in a total volume of 0.1 ml. The concentrations of TSA were 0.5 g/gm/day for low-dose therapy and 1 g/gm/day for high-dose therapy. After 4 weeks, the animals were killed by CO2 asphyxiation. The injected skin was removed and processed for histologic analysis. Experiments were performed in 2 independent series. Histologic analysis. The injected sections of skin were fixed in 4% formalin and embedded in paraffin. Fivemicrometer sections were stained with hematoxylin and eosin for the determination of dermal thickness. For analysis of connective tissue, Masson’s trichrome staining was performed. Dermal thickness of the injected sections was analyzed with an Imager1 microscope (Carl-Zeiss, Jena, Germany) at 200-fold magnification, by measuring the distance between the epidermal–dermal junction and the dermal–subcutaneous fat junction at sites of induration in 3 consecutive skin sections from each animal. In each series of experiments, the dermal thickness was calculated as the fold increase compared with that in controls. The analysis was performed by 2 independent 2758 HUBER ET AL examiners (AJ and LCH) who were blinded to the treatment groups. Statistical analysis. Results are expressed as the mean ⫾ SEM. Statistical analysis was performed using GraphPad Prism software, version 4.03 (GraphPad Software, San Diego, CA). For analysis between different groups, Wilcoxon’s test was used. P values less than or equal to 0.05 were considered significant. RESULTS Effects of TSA on the expression of profibrotic molecules at the mRNA and protein levels. TSA was described as a lead compound in the development of antifibrogenic drugs for hepatic fibrosis (17). In this context, we investigated the antifibrogenic effects of TSA by exposing human SSc skin fibroblasts to TGF␤ (5 ng/ml) alone or in combination with different concentrations of TSA. As shown in Figure 1, levels of mRNA for both Col␣1(I) and fibronectin were strongly induced by TGF␤ in SSc skin fibroblasts, as expected. In particular, following treatment with TGF␤, mRNA for Col␣1(I) was induced a mean ⫾ SEM 2.7 ⫾ 0.5 fold and mRNA for fibronectin was induced 5.7 ⫾ 1.3 fold. These effects were abolished when TSA (2 M) was added simultaneously with TGF␤ to cultures of SSc skin fibroblasts. In this regard, addition of TSA reduced the TGF␤-stimulated production of Col␣1(I) mRNA from a mean ⫾ SEM 2.7 ⫾ 0.5 fold to 0.9 ⫾ 0.1 fold (P ⬍ Figure 1. Effects of trichostatin A (TSA) on the transcription of extracellular matrix genes. Systemic sclerosis (SSc) and normal dermal fibroblasts were stimulated with transforming growth factor ␤ (TGF␤) (5 ng/ml) (solid bars) or with TGF␤ (5 ng/ml) plus TSA (2 M) (shaded bars), and compared with unstimulated fibroblasts (open bar). Results are the mean and SEM fold change in levels of mRNA for Col␣1(I) and fibronectin relative to that in unstimulated cells, from 4 independent experiments; values are normalized to the expression of 18S ribosomal RNA. ⴱ ⫽ P ⬍ 0.05 versus SSc cells treated with TGF␤ alone. Figure 2. Inhibitory effects of TSA on the release of collagen protein in SSc dermal fibroblasts. TGF␤, interleukin-4 (IL-4), and plateletderived growth factor (PDGF) up-regulated the de novo synthesis of collagen by SSc fibroblasts. When TSA (2 M) was added to the cultures, both the basal and cytokine-induced production of collagen were inhibited. Results are the mean and SEM from 4 independent experiments. See Figure 1 for other definitions. 0.05) and the TGF␤-stimulated induction of fibronectin mRNA from 5.7 ⫾ 1.3 fold to 2.4 ⫾ 1.3 fold (P ⬍ 0.05) (n ⫽ 4 for each) (Figure 1). In experiments with skin fibroblasts from healthy controls, TGF␤ induced the production of Col␣1(I) mRNA a mean ⫾ SEM 2.3 ⫾ 0.8 fold and the production of fibronectin mRNA 3.5 ⫾ 0.6 fold, whereas the addition of TSA reduced the amount of TGF␤-induced mRNA production to 1.1 ⫾ 0.5 fold for Col␣1(I) and to 2.4 ⫾ 1.3 fold for fibronectin (Figure 1). We next tested the effects of TSA on skin fibroblasts at the protein level, by analyzing the supernatants of TGF ␤ - and TSA-treated SSc skin fibroblasts (Figure 2). Similar to the data obtained from the assessments of mRNA, TGF␤ (5 ng/ml) strongly increased the de novo synthesis of collagen from a mean ⫾ SEM 190 ⫾ 29 pg/ml to 455 ⫾ 43 pg/ml. The combination of TGF␤ and TSA down-regulated the synthesis of collagen to 170 ⫾ 55 pg/ml (P ⬍ 0.05), which was below the basal levels found in cultures of unstimulated cells. Addition of TSA to unstimulated SSc skin fibroblasts reduced the synthesis of collagen from 190 ⫾ 29 pg/ml to 115 ⫾ 49 pg/ml. These results clearly indicate that TSA inhibits the synthesis of collagen by SSc skin fibroblasts. To investigate whether TSA also blocks the action of other profibrotic cytokines, we exposed SSc skin fibroblasts to IL-4 (5 ng/ml) or PDGF (20 ng/ml) alone TRICHOSTATIN A IN SKIN FIBROSIS or in combination with TSA (2 M) for 48 hours. The stimulatory effects of IL-4 and PDGF on SSc skin fibroblasts were similar to the effects observed with TGF␤. The collagen production by SSc fibroblasts was increased to 288 ⫾ 73 pg/ml following stimulation with IL-4, and increased to 235 ⫾ 16 pg/ml following stimulation with PDGF. Strikingly, addition of TSA (2 M) reversed both the IL-4–induced collagen synthesis (to 103 ⫾ 25 pg/ml) and the PDGF-stimulated collagen synthesis (to 103 ⫾ 22 pg/ml), representing a return to normal levels, suggesting that TSA functions as an inhibitor of the final common pathway in the production of extracellular matrix proteins. Kinetics of TSA. We found that the observed effects of TSA were both dose- and time-dependent. Untreated cells and TGF␤-stimulated cells were exposed to TSA in concentrations of 100 nM, 500 nM, 1 M, and 2 M. Quantification of collagen protein within 48 hours after incubation showed that TSA acted in a clear dose-dependent manner, with the highest reduction in collagen synthesis observed following addition of 2 M TSA and the lowest reduction observed with 100 nM TSA. These doses are within the physiologic range of concentrations of TSA that have been applied in vitro in other studies (18,27). Similarly, TSA acted in a time-dependent manner. Protein levels were measured following a single addition of TSA and TGF␤ to the cultures, with the levels determined at 48 hours, 96 hours, and 144 hours after incubation. The maximal effects of TSA were observed after incubation for 48 hours (results not shown). These results confirm that TSA functions as a reversible HDAC inhibitor. Previous studies have shown that TSA might lead to inhibition of the cell cycle as well as apoptotic cell death (28,29). With the doses of TSA used in the present study, no increase in the rate of apoptosis could be observed, as determined by annexin V/propidium iodide double-staining and fluorescence-activated cell sorter analysis (results not shown). However, an effect of TSA was observed on the cell cycle. When SSc skin fibroblasts were exposed to TSA (2 M), proliferation of fibroblasts was reduced by 29 ⫾ 3% after 72 hours, as measured by MTS assay. To investigate the molecular mechanisms of this TSA-induced cell cycle inhibition, Western blotting for the cell cycle inhibitor p21 was performed. When SSc skin fibroblasts were analyzed after exposure to TSA for 48 hours, p21 was clearly induced (results not shown). Levels of Smad transcription factors. The downstream signaling action of TGF␤ is mainly mediated by 2759 Figure 3. Representative results from electrophoretic mobility shift assay, demonstrating the inhibition of nuclear Smad3/4 by TSA in SSc dermal fibroblasts. Lane 1, Positive control; lane 2, free probe without nuclear extracts; lane 3, nuclear extract from unstimulated SSc fibroblasts; lane 4, nuclear extract from TGF␤-stimulated SSc fibroblasts; lane 5, nuclear extract from unstimulated SSc fibroblasts treated with TSA; lane 6, nuclear extract from TGF␤-stimulated SSc fibroblasts treated with TSA; lane 7, nuclear extract from unstimulated SSc fibroblasts with an excess of unlabeled cold probe; control lanes were ran on a separate gel. Signals for Smad3/4–DNA complexes provide evidence of a strong activation of Smad3/4 by TGF␤ (5 ng/ml) (lane 4). This activation could be inhibited when TSA (2 M) was added (lane 6). See Figure 1 for definitions. Smad transcription factors. To determine whether TSA alters the expression pattern of these molecules, EMSA was performed. SSc skin fibroblasts were exposed to TGF␤, alone or in combination with TSA, for 2 hours. As shown in Figure 3, TGF␤ strongly induced the DNA-binding fraction of the profibrotic Smad3/4. This effect, however, was almost completely abolished when the cells were coincubated with TSA. These findings were also confirmed on the morphologic level, by confocal fluorescence microscopy of skin fibroblasts using FITC-labeled anti-Smad3 antibodies (Figure 4). In untreated control SSc fibroblasts (Figure 4a), green fluorescence within the nuclei, representing the presence of Smad3 molecules, could be observed only sporadically. After exposure of SSc skin 2760 HUBER ET AL Figure 4. Representative images from confocal fluorescence microscopy of human SSc dermal fibroblasts. SSc fibroblasts were cultured on chamber slides with TSA and/or TGF␤ and stained with DAPI for localization of nuclei (red) and fluorescein isothiocyanate–labeled anti-Smad3 antibodies (green). Intranuclear Smad3 molecules could be observed only sporadically in untreated SSc fibroblasts (a) and in TSA-treated SSc fibroblasts (b). Stimulation of the fibroblasts with TGF␤ resulted in cytoplasmic and nuclear distribution of Smad3 molecules (c). Simultaneous application of TGF␤ in conjunction with TSA completely abolished this TGF␤-induced nuclear translocation in SSc fibroblasts (d). See Figure 1 for definitions. fibroblasts to TSA alone (Figure 4b), a similar pattern of scant Smad3 distribution was seen. When TGF␤ was added, however, Smad3 molecules were observed to be distributed over the cytoplasm and within the nucleus (Figure 4c). The addition of TSA completely abolished this TGF␤-induced nuclear translocation (Figure 4d). Effects of TSA on bleomycin-induced dermal fibrosis in vivo. Following our in vitro studies, we tested the effects of TSA in vivo using a previously described animal model of bleomycin-induced skin fibrosis (25). No obvious toxic effects were observed in TSA-treated mice. Repeated SC application of bleomycin caused marked thickening of the mouse skin at the sites of injection. When analyzed histologically, the skin from bleomycin-exposed animals showed clear signs of tissue fibrosis, including the accumulation of extracellular matrix within the dermis and the adjacent subcutis, thus leading to replacement of subcutaneous fat tissue with fibrillar collagen bundles (Figures 5a and b). When TSA was administered by IP injection after SC injections of bleomycin, the fibrotic effects of bleomycin were clearly reduced, as shown in Figures 5c and d. These observations were quantified by the measurement of dermal thickness. In particular, injection of bleomycin induced an increase in dermal thickness by a mean ⫾ SEM 71 ⫾ 14% as compared with that in untreated control mice (P ⱕ 0.001). Injections of TSA TRICHOSTATIN A IN SKIN FIBROSIS 2761 Figure 5. Prevention of the accumulation of extracellular matrix by trichostatin A (TSA) in experimental dermal fibrosis. Skin fibrosis was induced in mice by repeated subcutaneous injections of bleomycin (b–d). Groups of mice were either left untreated (b) or treated with intraperitoneal injections of TSA (c and d) at a concentration of 0.5 g/gm/day (c) or 1.0 g/gm/day (d). Phosphate buffered saline–treated mice served as controls (a). Injections of TSA prevented an increase in dermal thickness, as analyzed by computer-assisted measurement of the dermo–epidermal distance (red bars), and also prevented the accumulation of dense collagen bundles (arrows in b) that would replace subcutaneous fat tissue (asterisks in a and b versus c and d). Representative tissue sections in a–d were examined at the same magnification (original magnification ⫻ 200). prevented thickening of the skin. At a TSA dosage of 0.5 g/gm/day, the level of bleomycin-induced dermal fibrosis was significantly reduced (P ⬍ 0.003), to 16 ⫾ 4% of that in PBS-treated control animals. Similar levels of dermal thickness were observed when TSA was applied IP at a higher dosage (Figure 6), indicating that therapeutic effects could be achieved even with submicromolar doses of TSA. Moreover, morphologic analysis of the mouse skin with Masson’s trichrome staining, used for detection of collagen fibers on paraffin-embedded tissue sections, further revealed the dense accumulation of 2762 HUBER ET AL Figure 6. Change in dermal thickness in groups of mice with bleomycin-induced skin fibrosis treated with TSA at either 1 g/gm/ day (TSA high) or 0.5 g/gm/day (TSA low) or left untreated, as compared with a control group of phosphate buffered saline (PBS)– treated mice. Results are the mean and SEM fold change in dermal thickness relative to controls. ⴱ ⫽ P ⬍ 0.05. See Figure 5 for other definitions. collagen fibers induced by repeated bleomycin injections, whereas TSA partly prevented this process. In particular, dermal collagen bundles were less densely packed, and the subcutaneous fat was still present. These data are consistent with the results obtained in vitro and confirm the role of TSA as a promising lead compound in the development of antifibrogenic drugs. DISCUSSION Fibrosis is a clinical hallmark of many disorders, and excessive accumulation of extracellular matrix within internal organs ultimately leads to severe complications, such as respiratory distress, liver failure, and death. At the moment, no effective treatment is known to inhibit fibrotic processes. In the context of SSc, numerous studies have identified a profibrotic phenotype of fibroblasts in which the cells are stimulated by elevated levels of various cytokines at the early stages of the disease (4,30), ultimately leading to the development and progression of tissue fibrosis. Of interest, the profibrotic cellular features characterizing this phenotype remain unchanged during cell division (8). In addition to genetic mutations and clonal selection, which have been suggested as possible mechanisms, there has been growing interest in epigenetic modifications. Most of the evidence regarding the role of epigenetic alterations in disease has been gained from the use of epigenetic drugs, such as HDAC inhibitors. In the present study, we tested the effect of TSA, the strongest HDAC inhibitor currently known (12), on collagen production by skin fibroblasts, using in vitro and in vivo experiments with SSc skin fibroblasts and a mouse model of bleomycin-induced skin fibrosis. The addition of TSA normalized the levels of mRNA transcripts for Col␣1(I) (procollagen) and fibronectin, as well as the levels of collagen protein in cytokinestimulated SSc skin fibroblasts. Consistent with the data from other studies, our findings indicate the involvement of epigenetic histone modifications in the expression of fibrogenic molecules. The use of HDAC inhibitors prevents the removal of acetyl groups from core histones, and a state of histone hyperacteylation is usually associated with increased rates of gene transcription. Since TSA inhibited the growth factor–induced expression of profibrotic molecules in our experiments, several possible mechanisms of action can be discussed. One possible mechanism is the induction of genes essential for the suppression of collagen expression. Conversely, TSA or gene products induced by TSA might interfere with the signaling cascade involved in tissue fibrosis. Furthermore, TSA might affect the production of cellular products indirectly by, for example, inducing cell cycle arrest, apoptosis, or cell death. Binding of TGF␤ to the serine/theronine kinase receptor (TGF␤RII) activates TGF␤RI and initiates a downstream signaling cascade through Smad transcription factors. In particular, Smad proteins become activated by phosphorylation. Phosphorylated Smad2 and Smad3 then form a complex with Smad4, resulting in a heteromeric structure that translocates to the nucleus (6,31). Together with other DNA-binding factors, the Smad complexes regulate the expression of several profibrotic genes. Smad6 and Smad7, in contrast, have been linked to antifibrotic activities. Of the factors investigated in the present study, the DNA-binding activities of Smad3 and Smad4 were clearly affected, and nuclear translocation was inhibited by TSA. Whether this effect is attributable to epigenetic modifications of histones or whether TSA directly inhibits the phosphorylation of Smads has to be addressed in further studies. Moreover, TSA appears to have other functional properties in addition to the inhibition of histone deacetylation. Recent reports have described a novel, histone acetylation–independent mechanism by which HDAC inhibitors cause dephosphorylation of intracellular proteins, which challenges the view of the role of TSA as a sole HDAC inhibitor (32). Collagen repressors, such FLI1 and p53, were not investigated in the present study, but previous findings suggest that FLI1 is affected by the use of epigenetic TRICHOSTATIN A IN SKIN FIBROSIS drugs (8). Alternatively, the fact that TSA provokes cell cycle arrest could be another explanation for the reduced production of collagen proteins in SSc skin fibroblasts. Several studies performed in cancer cells have shown that TSA induces the expression of the cyclin-dependent kinase inhibitor p21, thus leading to cell cycle inhibition in melanoma and glioma cells (28,33). Consistent with these findings, TSA strongly induced the expression of p21 in SSc skin fibroblasts, and the population of proliferating cells was clearly reduced. Taken together, the findings from our experiments indicate that TSA prevents the accumulation of extracellular matrix by several mechanisms, including epigenetic alterations of the expression pattern of Smad transcription factors, and interference with regulation of the cell cycle. Our data are consistent with the results from previous studies in which the antifibrotic properties of TSA have been demonstrated (18,27). However, our study amends the data obtained in vitro, with experiments performed in vivo using a common animal model of skin fibrosis. In the bleomycin-induced fibrosis mouse model, micromolecular concentrations of TSA prevented the development of skin fibrosis without the occurrence of obvious toxic side effects. After application of 0.5 g/gm of TSA per day, the bleomycin-induced collagen bundles within the dermis and the replacement of subcutaneous fat tissue were reduced to levels comparable with those in the skin of untreated animals. However, increasing the applied concentration of TSA to 1.0 g/gm/day did not result in further improvement. These findings indicate that the therapeutic effects of TSA could be achieved even with doses in the submicromolar range. In conclusion, we have shown that the HDAC inhibitor TSA has potent antifibrogenic effects on SSc skin fibroblasts in vitro. With respect to the pivotal role of profibrotic cytokines such as TGF␤, PDGF, and IL-4 in the pathogenesis of fibrotic diseases, we were able to show that TSA abrogates the stimulating effects of these factors on extracellular matrix production. Moreover, in a mouse model of bleomycin-induced fibrosis, TSA prevented the development of skin fibrosis as quantified by changes to the dermal thickness within the bleomycin injection sites. With regard to the strong antifibrogenic activities of TSA, we propose that TSA and related inhibitory compounds should be considered for use as early pharmacologic strategies to conquer the development and progression of fibrosis in SSc and other cytokine-driven fibrotic entities. 2763 ACKNOWLEDGMENTS We thank all staff members of the laboratory of electron microscopy at the University of Zurich for their excellent support, in particular, Dr. M. Hoechli for providing assistance with the confocal microscopy. AUTHOR CONTRIBUTIONS Dr. Huber had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study design. Huber, Distler, Hauser, R. E. Gay, Matucci-Cerinic, S. Gay, Distler, Jüngel. Acquisition of data. Huber, Moritz, Hemmatazad, Hauser, Jüngel. Analysis and interpretation of data. Huber, Distler, S. Gay, Jüngel. Manuscript preparation. Huber, Moritz, Michel, R. E. Gay, S. Gay, Distler, Jüngel. Statistical analysis. Huber. REFERENCES 1. Derk CT, Jimenez SA. Systemic sclerosis: current views of its pathogenesis [review]. Autoimmun Rev 2003;2:181–91. 2. Blobe GC, Schiemann WP, Lodish HF. Role of transforming growth factor ␤ in human disease [review]. N Engl J Med 2000;342:1350–8. 3. Distler JH, Jungel A, Caretto D, Schulze-Horsel U, KowalBielecka O, Gay RE, et al. Monocyte chemoattractant protein 1 released from glycosaminoglycans mediates its profibrotic effects in systemic sclerosis via the release of interleukin-4 from T cells. Arthritis Rheum 2006;54:214–25. 4. Gay S, Jones RE Jr, Huang GQ, Gay RE. Immunohistologic demonstration of platelet-derived growth factor (PDGF) and sis-oncogene expression in scleroderma. J Invest Dermatol 1989; 92:301–3. 5. Ishida W, Mori Y, Lakos G, Sun L, Shan F, Bowes S, et al. Intracellular TGF-␤ receptor blockade abrogates Smad-dependent fibroblast activation in vitro and in vivo. J Invest Dermatol 2006;126:1733–44. 6. Mori Y, Chen SJ, Varga J. Expression and regulation of intracellular SMAD signaling in scleroderma skin fibroblasts. Arthritis Rheum 2003;48:1964–78. 7. Johnson RW, Tew MB, Arnett FC. The genetics of systemic sclerosis [review]. Curr Rheumatol Rep 2002;4:99–107. 8. Wang Y, Fan PS, Kahaleh B. Association between enhanced type I collagen expression and epigenetic repression of the FLI1 gene in scleroderma fibroblasts. Arthritis Rheum 2006;54:2271–9. 9. Ballestar E, Esteller M, Richardson BC. The epigenetic face of systemic lupus erythematosus. J Immunol 2006;176:7143–7. 10. Barnes PJ, Adcock IM, Ito K. Histone acetylation and deacetylation: importance in inflammatory lung disease [review]. Eur Respir J 2005;25:552–63. 11. Wade PA. Transcriptional control at regulatory checkpoints by histone deacetylases: molecular connections between cancer and chromatin. Hum Mol Genet 2001;10:693–8. 12. De Ruijter AJ, van Gennip AH, Caron HN, Kemp S, van Kuilenburg AB. Histone deacetylases (HDACs): characterization of the classical HDAC family [review]. Biochem J 2003;370: 737–49. 13. Finnin MS, Donigian JR, Cohen A, Richon VM, Rifkind RA, Marks PA, et al. Structures of a histone deacetylase homologue bound to the TSA and SAHA inhibitors. Nature 1999;401:188–93. 14. Chung YL, Wang AJ, Yao LF. Antitumor histone deacetylase inhibitors suppress cutaneous radiation syndrome: implications for 2764 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. increasing therapeutic gain in cancer radiotherapy. Mol Cancer Ther 2004;3:317–25. Williams RJ. Trichostatin A, an inhibitor of histone deacetylase, inhibits hypoxia-induced angiogenesis. Expert Opin Investig Drugs 2001;10:1571–3. Marks PA, Richon VM, Breslow R, Rifkind RA. Histone deacetylase inhibitors as new cancer drugs [review]. Curr Opin Oncol 2001;13:477–83. Rombouts K, Niki T, Wielant A, Hellemans K, Geerts A. Trichostatin A, lead compound for development of antifibrogenic drugs. Acta Gastroenterol Belg 2001;64:239–46. Niki T, Rombouts K, De Bleser P, De Smet K, Rogiers V, Schuppan D, et al. A histone deacetylase inhibitor, trichostatin A, suppresses myofibroblastic differentiation of rat hepatic stellate cells in primary culture. Hepatology 1999;29:858–67. Chung YL, Lee MY, Wang AJ, Yao LF. A therapeutic strategy uses histone deacetylase inhibitors to modulate the expression of genes involved in the pathogenesis of rheumatoid arthritis. Mol Ther 2003;8:707–17. LeRoy EC, Black C, Fleischmajer R, Jablonska S, Krieg T, Medsger TA Jr, et al. Scleroderma (systemic sclerosis): classification, subsets and pathogenesis. J Rheumatol 1988;15:202–5. Mishra N, Reilly CM, Brown DR, Ruiz P, Gilkeson GS. Histone deacetylase inhibitors modulate renal disease in the MRL-lpr/lpr mouse. J Clin Invest 2003;111:539–52. Jungel A, Distler JH, Kurowska-Stolarska M, Seemayer CA, Seibl R, Forster A, et al. Expression of interleukin-21 receptor, but not interleukin-21, in synovial fibroblasts and synovial macrophages of patients with rheumatoid arthritis. Arthritis Rheum 2004;50: 1468–76. Andrews NC, Faller DV. A rapid micropreparation technique for extraction of DNA-binding proteins from limiting numbers of mammalian cells. Nucleic Acids Res 1991;19:2499. Distler JH, Jungel A, Huber LC, Seemayer CA, Reich CF III, Gay RE, et al. The induction of matrix metalloproteinase and cytokine HUBER ET AL 25. 26. 27. 28. 29. 30. 31. 32. 33. expression in synovial fibroblasts stimulated with immune cell microparticles. Proc Natl Acad Sci U S A 2005;102:2892–7. Distler JH, Jungel A, Huber LC, Schulze-Horsel U, Zwerina J, Gay RE, et al. Imatinib mesylate reduces production of extracellular matrix and prevents development of experimental dermal fibrosis. Arthritis Rheum 2007;56:311–22. Lakos G, Takagawa S, Varga J. Animal models of scleroderma. Methods Mol Med 2004;102:377–93. Rombouts K, Niki T, Greenwel P, Vandermonde A, Wielant A, Hellemans K, et al. Trichostatin A, a histone deacetylase inhibitor, suppresses collagen synthesis and prevents TGF-␤1-induced fibrogenesis in skin fibroblasts. Exp Cell Res 2002;278:184–97. Wetzel M, Premkumar DR, Arnold B, Pollack IF. Effect of trichostatin A, a histone deacetylase inhibitor, on glioma proliferation in vitro by inducing cell cycle arrest and apoptosis. J Neurosurg 2005;103:549–56. Jungel A, Baresova V, Ospelt C, Simmen BR, Michel BA, Gay RE, et al. Trichostatin A sensitises rheumatoid arthritis synovial fibroblasts for TRAIL-induced apoptosis. Ann Rheum Dis 2006; 65:910–2. Higley H, Persichitte K, Chu S, Waegell W, Vancheeswaran R, Black C. Immunocytochemical localization and serologic detection of transforming growth factor ␤1: association with type I procollagen and inflammatory cell markers in diffuse and limited systemic sclerosis, morphea, and Raynaud’s phenomenon. Arthritis Rheum 1994;37:278–88. Massague J. How cells read TGF-␤ signals [review]. Nat Rev Mol Cell Biol 2000;1:169–78. Chen CS, Weng SC, Tseng PH, Lin HP. Histone acetylationindependent effect of histone deacetylase inhibitors on Akt through the reshuffling of protein phosphatase 1 complexes. J Biol Chem 2005;280:38879–87. Florenes VA, Skrede M, Jorgensen K, Nesland JM. Deacetylase inhibition in malignant melanomas: impact on cell cycle regulation and survival. Melanoma Res 2004;14:173–81.