SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Objective

Transforming growth factor β (TGFβ) induces profibrotic responses in normal fibroblasts, and plays a fundamental role in the pathogenesis of fibrosis in scleroderma (systemic sclerosis [SSc]). The intensity of cellular responses elicited by cytokines is modulated by transcriptional coactivators such as the histone acetylase p300. The objective of these studies was to delineate the physiologic role of p300 in Smad-dependent profibrotic responses elicited by TGFβ.

Methods

Ectopic p300 was transiently expressed in normal dermal fibroblasts. Cellular p300 levels were suppressed using p300-specific ribozymes. The regulation of gene expression was examined by transient transfection assays, Northern blotting, and immunoblot analysis. The expression of p300 in normal and scleroderma fibroblasts was evaluated by confocal microscopy and immunoblotting, and p300 levels in skin from mice with experimental scleroderma were assessed by immunohistochemistry.

Results

In normal fibroblasts, TGFβ induced an increase in the levels of p300. Forced expression of ectopic p300 in these cells dramatically enhanced the magnitude of TGFβ responses, whereas selective depletion of p300 using ribozyme resulted in abrogation of TGFβ-induced collagen synthesis and promoter activity. Furthermore, TGFβ lost its ability to induce Smad-dependent transcription in p300-depleted fibroblasts. These responses could be fully rescued with ectopic p300. Abrogation of Smad-mediated TGFβ signaling was not due to alterations in the levels or the ligand-dependent phosphorylation or intracellular trafficking of endogenous Smads. Immunohistochemical analysis demonstrated substantially increased p300 expression in lesional skin from mice with chronic graft-versus-host disease, an animal model of scleroderma. Furthermore, levels of p300 were 2–3–fold higher in cultured fibroblasts derived from SSc patients than in fibroblasts from matched normal controls.

Conclusion

These results establish, for the first time, that the coactivator histone acetylase p300, itself a target of TGFβ regulation, is an essential component of the cellular TGFβ signal transduction pathways mediating stimulation of collagen synthesis in fibroblasts. Since the cellular abundance of p300 appears to govern the intensity of profibrotic responses elicited by TGFβ, elevated p300 expression in lesional tissue may contribute to the progression of skin fibrosis in scleroderma.

Fibrosis in scleroderma (systemic sclerosis [SSc]) results in part from excessive synthesis and accumulation of collagen (1). Transforming growth factor β (TGFβ), a multifunctional cytokine with potent stimulatory effects on collagen synthesis and myofibroblast differentiation in vivo and in vitro, plays a fundamental role in the pathogenesis of fibrosis (2). Although enhanced bioavailability, signaling, and cellular responsiveness to TGFβ have all been implicated as potential mechanisms underlying the pathogenesis of fibrosis, the precise molecular events leading to excessive collagen synthesis in SSc fibroblasts remain poorly understood.

Multiple transcription factors have been implicated in the regulation of collagen gene expression by TGFβ (3). The Smad family proteins are recently recognized intracellular signal transducers for TGFβ that are the only known direct substrates for activated TGFβ receptors (4). Based on their structure and function, 3 subgroups of Smads have been identified: receptor-associated Smads (R-Smads) (Smads 1, 2, 3, 5, and 8), common-mediator Smad (Smad4), and inhibitory Smads (Smad6 and Smad7). Recent studies have revealed that Smad3 plays a pivotal role in TGFβ stimulation of collagen gene expression in normal fibroblasts (5). Signaling by TGFβ is initiated when ligand binds to a transmembrane TGFβ receptor type II (TGFβRII), which phosphorylates and activates TGFβRI, represented in fibroblasts by activin-like receptor kinase (ALK-5). Activated ALK-5 then phosphorylates R-Smads, promoting their heterodimerization with Smad4 and translocation from the cytoplasm into the nucleus. Within the nucleus, the Smad hetero complex interacts with canonical Smad-binding elements (SBEs) of target genes to activate their transcription (6).

Transcriptional cofactors play vital roles in a variety of cellular processes. Although they do not directly bind to DNA, cofactors interact with sequence-specific DNA-binding protein factors and modulate the activation of multiple genes. One of the most extensively characterized transcriptional coactivators, p300, is a 300-kd nuclear protein present in limiting amounts in most cell types (7, 8). Cytokines, stress, and viral infections induce interactions between p300 and transcription factors or nuclear receptors in promoter-specific transcriptional complexes. Due to its intrinsic acetyltransferase activity, p300 can acetylate histone N-terminal lysine residues, resulting in the activation of chromatin-repressed promoters. In addition to histones, transcription factors such as p53, GATA-1, c-Myb, and NF-κB can also be acetylated by p300. Somatic mutations of p300 occur in a number of malignancies, and disruption of the p300 gene is associated with embryonic lethality in mice, indicating a unique and fundamental biologic role for p300 (9, 10).

We have shown previously that forced expression of ectopic p300 in skin fibroblasts resulted in enhanced COL1A2 gene promoter activity and collagen synthesis (11–13). Furthermore, TGFβ induced a direct physical interaction between Smad3 and endogenous p300 in these cells (13). However, the physiologic role of p300 in TGFβ regulation of collagen, and its involvement in pathologic fibrosis, remain unknown. In the present studies we investigated the functional significance of p300 in TGFβ-induced responses. The results indicated that ectopic expression of p300 in skin fibroblasts markedly enhanced the stimulation of collagen gene expression elicited by TGFβ. In contrast, selective lowering of p300 levels using ribozymes abrogated TGFβ stimulation of collagen synthesis and reduced Smad-dependent transcription. The levels of p300 were markedly elevated in SSc skin fibroblasts compared with matched controls. Furthermore, p300 expression was dramatically enhanced in lesional tissue in a mouse model of scleroderma. Importantly, TGFβ caused a substantial increase in p300 levels in normal fibroblasts. Together, these results indicate that the level of cellular p300 expression, itself subject to positive regulation by TGFβ, may determine the intensity of the fibrotic response elicited by TGFβ.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Cell cultures.

Primary cultures of neonatal foreskin fibroblasts were established as previously described (14). Cells were maintained in Eagle's minimal essential medium supplemented with 10% fetal bovine serum (FBS; Gibco BRL, Gaithersburg, MD), 1% vitamins, 1% penicillin/streptomycin, and 2 mML-glutamine, and studied between passages 4 and 8. All tissue culture reagents were obtained from BioWhittaker (Walkersville, MD). Skin biopsy specimens were obtained from the affected dorsal forearms of 5 patients with diffuse cutaneous SSc (2 women and 3 men, median age 45 years). The duration of skin thickening was 4–18 months. The patients fulfilled the American College of Rheumatology (formerly, the American Rheumatism Association) criteria for SSc (15) and were receiving no treatment for SSc at the time of biopsy. All skin biopsies were performed with informed consent and in compliance with the Institutional Review Board for Human Studies. Dermal fibroblast cultures were established from the biopsy specimens as previously described (16). Biopsy specimens from healthy volunteers matched for age, sex, and race were processed in parallel. Fibroblasts were maintained in Dulbecco's modified Eagle's medium supplemented with 10% FBS, and studied between passages 2 and 5.

Plasmid constructs and transient transfection assays.

The 772COL1A2-CAT reporter construct contains the sequences from −772 to +58 bp of the human COL1A2 promoter fused to the chloramphenicol acetyltransferase (CAT) reporter gene (17). The SBE4-TK-luc construct contains 4 copies of the consensus SBE in front of TK-luc (18). The 3TP-Lux construct contains both activator protein 1 (AP-1) consensus elements and an SBE derived from the plasminogen activator inhibitor 1 (PAI-1) promoter linked to the luciferase reporter gene (19). The renilla luciferase construct pRL-TK-luc was used as an internal control (12). The pCI p300 construct contains the full-length p300 complementary DNA (cDNA) in pCI vector (20). The p300-specific ribozyme constructs were made by inserting double-stranded oligonucleotides encoding hammerhead ribozymes and pol III termination sequence downstream of Val–transfer RNA promoter in pUC-tRVP (21). Inactive Rz p300 harbors a point mutation in the catalytic domain of the ribozyme, and cannot cleave target messenger RNA (mRNA).

Confluent foreskin fibroblasts were transiently transfected with reporter and expression vectors, or with empty vectors using SuperFect reagent (Qiagen, Valencia, CA) and incubated with TGFβ2 (12.5 ng/ml; Genzyme, Framingham, MA) for 48 hours. CAT and luciferase activities with the equal amounts of protein were determined by phase extraction method with 14C-chloramphenicol as substrate or using LARII reagent (Promega, Madison, WI). The results were normalized with renilla luciferase activities in each sample. Each experiment was performed in triplicate and repeated 2–3 times with consistent results.

Immunoblot analysis.

Fibroblasts were transiently transfected with plasmids encoding either the active or inactive p300-specific ribozyme in parallel, and incubated with TGFβ2 for 48 hours. At the end of the incubation period, cells were harvested and whole cell lysates were prepared using lysis buffer (Santa Cruz Biotechnology, Santa Cruz, CA). Equal aliquots (40 μg or 100 μg) were electrophoresed on 4–20% Tris-Glycine gradient gels (Bio-Rad, Hercules, CA), followed by immunoblot analysis. Primary antibodies (1:200 or 1:1,000 dilution) against p300 (C20), Smad4 (B8), Smad7 (N19), Smad1/2/3 (H2), or actin (C2) (all from Santa Cruz Biotechnology), β-actin (Sigma, St. Louis, MO), CREB-binding protein (CBP; Affinity BioReagents, Golden, CO), type I collagen (Southern Biotechnology, Birmingham, AL), Smad3 (Zymed, South San Francisco, CA), and α-smooth muscle actin (α-SMA; Sigma) were used. Blots were washed with Tris buffered saline–Tween, followed by incubation with appropriate horseradish peroxidase (HRP)–conjugated secondary antibodies for 1 hour. Antigen–antibody complexes were visualized by chemiluminescence (ECL Reagent; Amersham BioSciences, Piscataway, NJ) after 1 minute exposure to XAR5 film (Kodak, Rochester, NY). Signal intensities of bands were quantitated using Bio-Rad Molecular Analyst software or NIH image analysis software.

Northern blot analysis.

Total RNA from confluent fibroblasts transfected with p300 ribozyme vectors was prepared using TRIzol reagent (Life Technologies, Grand Island, NY), as described previously (12). Equal amounts of RNA were separated in 1% formaldehyde agarose gels and transferred to nitrocellulose filters, followed by hybridization with α32P-dCTP–labeled cDNA probes for human COL1A2. The levels of GAPDH mRNA were used for correcting for variations in loading. The hybridized membranes were autoradiographed for 24–48 hours at −70°C. The intensities of bands were measured and analyzed using Molecular Analyst software.

Immunocytochemistry.

Fibroblasts were seeded into 8-well Chamber glass slides (Lab-Tek Products, Naperville, IL) and transfected with active or inactive p300-specific ribozyme, followed by incubation with TGFβ2 (12.5 ng/ml) for 48 hours. At the end of the incubation, the cells were processed for immunocytochemistry with primary antibodies against p300, CBP, Smad4, or α-SMA (all at 1:200 dilution), and examined by confocal laser scanning microscopy (LSM 510, Zeiss, Jena, Germany) (22). The number of cells showing positive immunostaining was determined by scoring 100 individual fibroblasts from 5 microscopic fields.

Sclerodermatous graft-versus-host disease (GVHD) in mice.

To study p300 expression in fibrosis in vivo, a mouse model of scleroderma was used (23). Six-week-old female B10.D2(H-2) and BALB/c (H-2d) mice (The Jackson Laboratory, Bar Harbor, ME) were utilized as donors and recipients, respectively, for transplantation of bone marrow and spleen cells, as previously described (24). Briefly, recipient BALB/c mice lethally irradiated with 700 cGy from a Gammacel137Cs source were injected intravenously with donor mouse B10.D2 spleen cells (2 × 106/mouse) and bone marrow cells (106/mouse) suspended in RPMI 1640 with 10 units/ml heparin. Control BALB/c mice received BALB/c spleen and bone marrow cells in parallel. Transplanted animals were maintained in sterile Micro-Isolator cages and were supplied with autoclaved food and acidified water.

On day 22 following transplantation, mice (3–5/group) were killed by cervical dislocation, and dorsal skin tissues were processed for routine histology and immunohistochemistry. Skin was embedded in OCT and sectioned using a cryostat. After fixation with cold acetone, sections were blocked with serum-free blocking agent (Dako, Carpinteria, CA) for 15 minutes, followed by incubation with antibody to p300 for 2 hours at room temperature. For detection of immunocomplex, Envision+ was used according to the instructions of the manufacturer (Dako). Isotype controls were stained in parallel, and the specificity of the primary antibodies was confirmed. The number of p300-positive cells in the dermis was determined semiquantitatively by examining a minimum of 30 cells/field at 40× magnification in each slide. After counterstaining with hematoxylin, the sections were mounted with Permount (Fisher Scientific, Pittsburgh, PA), viewed under a microscope, and photographed. To prevent observer bias, histologic specimens were coded and examined without knowledge of experimental conditions. The ratio of positive cells to total number of cells counted in each field was then calculated. Paraffin-embedded skin sections were stained with hematoxylin and eosin, and dermal areas were determined from microscopic photographs using image analysis as described previously (23, 24).

Statistical analysis.

Data are presented as the mean ± SD. The statistical significance of the differences between experimental and control groups was determined by analysis of variance and Student's t-test. P values less than 0.05 were considered significant.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Enhanced stimulation of COL1A2 activity by forced expression of p300.

To explore the modulation of profibrotic responses elicited by TGFβ, we first examined the effect of forced expression of ectopic p300 on collagen gene expression. For this purpose, normal fibroblasts at confluence were transiently transfected with expression vector for p300 along with the 772COL1A2-CAT reporter construct, and incubated with TGFβ for 48 hours. The results from multiple transfection assays showed that while forced expression of ectopic p300 by itself reproducibly caused an ∼2-fold increase in COL1A2 promoter activity, increases in cellular p300 levels were associated with a dramatic dose-dependent enhancement of TGFβ-induced stimulation, with a >5-fold increase in fibroblasts transfected with 500 ng p300 (Figure 1A). Since p300 is present in limiting amounts in normal cells, these findings suggested that increasing the abundance of p300 sensitized the fibroblasts to TGFβ.

thumbnail image

Figure 1. Ribozyme-mediated selective depletion of p300 in dermal fibroblasts. A, Confluent fibroblasts were transiently transfected with the indicated concentrations of p300 expression vector or empty vector, along with 772COL1A2-CAT. After 48 hours of incubation with transforming growth factor β2 (TGFβ2) (12.5 ng/ml), cultures were harvested and chloramphenicol acetyltransferase (CAT) activities were determined. Open bars represent untreated fibroblasts; solid bars represent TGFβ-treated fibroblasts. Values are the mean and SD of triplicate determinations, normalized with renilla luciferase activities. ∗ = P < 0.05. B, Fibroblasts were transfected with inactive or active p300 ribozyme (Rz) in parallel, and incubated with TGFβ2 for 48 hours. Cultures were harvested and whole cell lysates were subjected to immunoblot analysis with antibodies against p300, CREB-binding protein (CBP), or actin. Representative immunoblots are shown (upper panel). Signal intensities for p300 were quantitated by densitometry, and the results (means from 2 separate experiments), corrected for small variations in loading efficiency, are shown as the mean and SD fold induction relative to controls (bars in lower panel). C, Fibroblasts were fixed, immunostained with antibodies against p300, and examined by confocal microscopy. Representative micrographs are shown (original magnification × 100).

Download figure to PowerPoint

Specific reduction of p300 levels in dermal fibroblasts by ribozyme.

To explore the hypothesis that the relative abundance of p300 governed the intensity of cellular response to TGFβ, ribozyme was used to reduce endogenous p300 levels in normal fibroblasts. First, the specificity of ribozyme was examined by immunoblot analysis. For this purpose, fibroblasts transiently transfected with expression vectors encoding an active ribozyme that specifically cleaves p300 mRNA, or its inactive mutant as control, were incubated with TGFβ for 48 hours. The results showed that endogenous p300 was detectable in whole cell lysates from nontransfected normal fibroblasts (data not shown), as well as from fibroblasts expressing inactive ribozyme (Figure 1B). Furthermore, TGFβ reproducibly induced a 3-fold increase in cellular levels of p300. A marked decrease in p300 levels (60–75% compared with fibroblasts transfected with inactive ribozyme) was observed in fibroblasts transfected with active p300 ribozyme, both in the absence and in the presence of TGFβ. No alterations in cellular viability were detected.

To further evaluate the substrate specificity of the active ribozyme, changes in the levels of CBP, a close homolog of p300, were determined in transfected fibroblasts. The results showed that in marked contrast to p300, cellular levels of CBP were completely unaffected by overexpression of p300-specific ribozyme (Figure 1B). These results indicated that in normal fibroblasts, the active ribozyme effectively targeted p300 mRNA, resulting in a marked selective decrease in cellular p300 levels.

Depletion of p300 by the ribozyme was examined in individual fibroblasts. For this purpose, transfected fibroblasts were incubated with TGFβ, immunostained with antibodies specific for p300 or CBP, and examined by confocal microscopy. In fibroblasts expressing inactive ribozyme, endogenous p300 was found to be localized predominantly in the nucleus (Figure 1C). Significantly, the levels of p300 were markedly elevated in fibroblasts incubated with TGFβ, consistent with results from immunoblot analysis. Overexpression of active p300 ribozyme was associated with a substantial and uniform decrease in cellular p300 levels in both unstimulated and TGFβ-stimulated fibroblasts (Figure 1C). In contrast, levels of CBP appeared to be unaltered in fibroblasts expressing either active and inactive p300 ribozyme (data not shown).

Prevention of TGFβ stimulation of collagen synthesis by cellular p300 depletion.

To investigate the physiologic role of cellular p300 in mediating the TGFβ response, dermal fibroblasts transiently transfected with active or inactive ribozyme were incubated with TGFβ for 48 hours, and the levels of cellular (Figure 2A) and secreted (data not shown) type I collagen were determined by immunoblot analysis. The results demonstrated that, whereas TGFβ significantly (>2-fold) enhanced the synthesis of collagen in fibroblasts expressing inactive ribozyme, little stimulation was seen in fibroblasts expressing the active ribozyme. The mechanistic basis underlying this phenomenon was investigated through Northern blot analysis using total RNA extracted from TGFβ-treated fibroblasts transfected with expression vectors encoding active or inactive p300-specific ribozymes. The results revealed that while TGFβ induced an ∼2-fold increase in COL1A2 mRNA expression in control fibroblasts, there was only a modest (∼25%) increase noted in fibroblasts expressing p300 ribozyme (Figure 2B).

thumbnail image

Figure 2. Selective reduction of collagen stimulation by p300 depletion. Confluent fibroblasts were transfected with inactive or active p300 ribozymes (Rz), followed by incubation with transforming growth factor β2 (TGFβ2) for 48 hours. A and C, Whole cell lysates were analyzed by immunoblotting using antibodies against type I collagen, α-smooth muscle actin (α-SMA), or actin. Representative immunoblots are shown. B, Total RNA was subjected to Northern blot analysis. GAPDH mRNA was used to normalize for variations in loading efficiencies. D, Upper panel, Fibroblasts were immunostained with antibodies to α-SMA (green), and examined by confocal microscopy. Blue color (4′,6-diamidino-2-phenylindole) indicates nuclei. Representative micrographs are shown (original magnification × 100). Lower panel, The proportion of fibroblasts displaying strong green staining was quantitated and is shown as the percentage (mean and SD) of the total number of fibroblasts.

Download figure to PowerPoint

To further investigate the functional role of p300 in mediating the stimulation of relevant proteins by TGFβ, the levels of cellular α-SMA, a marker of myofibroblast transdifferentiation, were determined. Immunoblot analysis showed that TGFβ incubation of fibroblasts transfected with inactive ribozyme resulted in increased expression of α-SMA (Figure 2C). Importantly, however, stimulation was not prevented by active p300 ribozyme. The regulation of α-SMA was further investigated by immunocytochemistry and confocal microscopy. After 48 hours, TGFβ induced strong α-SMA expression and incorporation into cytoskeletal stress fibers in fibroblasts expressing either inactive or active ribozyme (Figure 2D). Together, these results indicated that cellular p300 plays an essential role in regulation of collagen gene expression by TGFβ, whereas stimulation of other proteins functionally implicated in the pathogenesis of fibrosis, such as α-SMA, appears to be independent of p300.

Reduced stimulation of Smad-dependent transcription in p300-depleted fibroblasts.

To further investigate how p300 modulated TGFβ-induced transcriptional responses, the regulation of 772COL1A2-CAT in transiently transfected fibroblasts was investigated. The results from multiple transfection assays showed that while TGFβ caused an ∼3-fold increase in COL1A2 promoter-driven CAT activity in control fibroblasts, as expected, stimulation was reduced in fibroblasts expressing p300 ribozyme (Figure 3, left panel).

thumbnail image

Figure 3. Inhibition of Smad-dependent transcriptional responses by p300 depletion. Confluent fibroblasts were transiently transfected with 772COL1A2-CAT, 3TP-Lux, or SBE4-TK-luc, along with the indicated amounts of inactive or active p300 ribozyme (Rz). Following 48 hours of incubation with transforming growth factor β2 (TGFβ2) (12.5 ng/ml), fibroblasts were harvested and chloramphenicol acetyltransferase (CAT) or luciferase activities were determined. Transfection efficiency was monitored by cotransfection with renilla luciferase construct. Values are the mean and SD of triplicate determinations from multiple experiments. Open bars represent untreated fibroblasts; solid bars represent TGFβ-treated fibroblasts. ∗ = P < 0.05 versus TGFβ-treated fibroblasts transfected with inactive p300 ribozyme.

Download figure to PowerPoint

In multiple cell types, TGFβ induces direct physical interaction of p300 with activated R-Smads (25–28). To characterize the functional role of the ligand-induced p300–Smad interaction in TGFβ responses, the consequences of depleting cellular p300 were examined. For this purpose, fibroblasts transfected with either 3TP-Lux, a commonly used minimal TGFβ reporter construct containing both SBE and AP-1 binding elements derived from the PAI-1 promoter, or SBE4-TK-luc, containing 4 copies of the SBE along with active or inactive ribozyme, were incubated with TGFβ for 48 hours. The results of transfection assays revealed that while in control fibroblasts TGFβ enhanced the activities of both 3TP-Lux and SBE4-TK-luc as expected, expression of p300 ribozyme was associated with reduced stimulation of these reporters (Figure 3, middle and right panels). Collectively, these results indicate that cellular p300 plays an essential role in Smad-dependent TGFβ signaling in these fibroblasts.

Levels of endogenous Smads, and their activation, are unaltered in p300-depleted fibroblasts.

Repression of Smad-mediated TGFβ signaling in p300-specific ribozyme-transfected fibroblasts could have resulted from alteration in the expression, or ligand-induced intracellular trafficking or phosphorylation, of R-Smads, or changes in inhibitory Smad7 regulation. To explore these possibilities, fibroblasts transfected with p300 ribozymes were incubated with TGFβ for 48 hours followed first by immunoblot analysis of whole cell lysates. The results showed that neither basal nor TGFβ-regulated levels of Smad3, Smad4, or Smad7 were altered in fibroblasts expressing active p300 ribozyme, compared with control fibroblasts expressing inactive ribozyme (data not shown). Therefore, inhibition of Smad-dependent TGFβ signaling in p300-depleted fibroblasts could not be attributed to down-regulation of Smad3/4 or up-regulation of inhibitory Smad7 levels, suggesting instead alterations in R-Smad transcriptional activities.

The effect of ribozymes on the phosphorylation status of R-Smads in TGFβ-treated fibroblasts was next evaluated. Rapid phosphorylation of Smad2 was induced by TGFβ; this response was unaltered by p300 ribozyme exposure (data not shown). To examine the modulation of ligand-induced Smad trafficking, the distribution of endogenous Smad4 was examined by confocal microscopy in fibroblasts transiently transfected with active or inactive ribozyme. The results showed that TGFβ-induced rapid nuclear accumulation of Smad4 was unaffected by p300-specific ribozyme (data not shown). Together, these results indicate that ribozyme-mediated depletion of p300 is associated with inhibition of Smad-dependent TGFβ responses independent of changes in the expression levels, phosphorylation status, or subcellular localization of endogenous Smads.

Rescue of TGFβ-induced stimulation of collagen gene transcription by overexpressed p300 in the presence of ribozyme.

If depletion of cellular p300 was responsible for suppression of Smad-dependent transcription, then forced expression of ectopic p300 could be expected to rescue TGFβ stimulation even in the presence of active ribozyme. In order to evaluate this possibility, fibroblasts transiently transfected with p300 ribozyme and 772COL1A2-CAT reporter were cotransfected with increasing amounts of p300 expression vector, followed by incubation with TGFβ for 48 hours. The results showed that overexpression of p300 dose-dependently rescued the stimulatory response elicited by TGFβ in fibroblasts transfected with p300 ribozyme (Figure 4), consistent with the notion that cellular p300 depletion accounted for suppression of TGFβ responses in these fibroblasts.

thumbnail image

Figure 4. Rescue of collagen stimulation by ectopic p300 in the presence of p300 ribozyme (Rz). Fibroblasts transfected with 772COL1A2-CAT reporter construct along with inactive or active p300 ribozyme (2 μg) were cotransfected with the indicated concentrations of p300 expression vector. Following 48 hours of incubation with transforming growth factor β2 (TGFβ2) (12.5 ng/ml), chloramphenicol acetyltransferase (CAT) activities were determined. Values, normalized with renilla luciferase activity, are the mean and SD of triplicate determinations from 2 independent experiments. Open bars represent untreated fibroblasts; solid bars represent TGFβ-treated fibroblasts. ∗ = P < 0.05.

Download figure to PowerPoint

Elevated p300 in lesional tissue from mice with sclerodermatous GVHD.

To explore the role of p300 in fibrotic responses in vivo, a mouse model of chronic GVHD was used. Mice transplanted with minor histocompatibility complex–mismatched donor cells developed skin fibrosis, with a consistent 30–40% increase in dermal thickness by day 22 (Figure 5A). The expression of p300 in the lesional skin was examined by immunohistochemical analysis. The results showed markedly elevated p300 expression in dermal fibroblasts and infiltrating macrophages in lesional skin from mice with sclerodermatous GVHD compared with syngeneic transplant controls (Figure 5B). The p300-specific staining intensity in the epidermal region was also elevated compared with controls. Staining of similar histologic sections with isotype control antibodies indicated the specificity of the anti-p300 antibodies (data not shown).

thumbnail image

Figure 5. Elevated p300 levels in the skin of mice with sclerodermatous graft-versus-host disease (Scl-GVHD). B10.D2 or control spleen and bone marrow cells were transplanted into lethally irradiated BALB/c mice, and the mice were killed 22 days later. A, Hematoxylin and eosin–stained dorsal skin tissue (left panel, control; right panel, sclerodermatous GVHD). Arrows indicate extent of dermis. B, Frozen sections were examined by immunohistochemistry with anti-p300 antibody. Sections were counterstained with hematoxylin. Photographs were obtained in different randomly selected fields (original magnification × 40). Insets depict representative fibroblastic cells (original magnification × 100). C, Percentage of p300-positive fibroblastic cells was determined in 5 separate fields. Values are the mean and SD.

Download figure to PowerPoint

Elevated p300 in cultured SSc skin fibroblasts.

Under in vitro culture conditions, scleroderma fibroblasts display features of activation, with constitutive up-regulation of multiple TGFβ-inducible genes (29). In light of the crucial role of p300 in mediating Smad-dependent TGFβ responses, and the apparent link between cellular abundance of p300 and the intensity of TGFβ-induced signaling, the expression of p300 in skin fibroblasts from SSc patients was determined. For this purpose, lesional skin fibroblasts from 5 patients with untreated early diffuse disease and 5 healthy controls matched for age, race, and sex were grown to confluence in parallel, and examined by immunoblot analysis. The results showed that p300 levels were elevated in all 5 SSc fibroblast lines compared with healthy controls, with a 2.8-fold mean increase (Figure 6). Levels of type I collagen were significantly higher (1.8-fold) in lesional fibroblasts compared with healthy controls.

thumbnail image

Figure 6. Elevated p300 levels in fibroblasts from systemic sclerosis (SSc) patients. A, Confluent fibroblasts derived from lesional skin of 5 SSc patients and normal skin (NS) from 5 matched healthy controls were harvested, and equal amounts of whole cell lysates were subjected to immunoblot analysis using antibodies against p300, type I collagen, or β-actin. Representative immunoblots are shown. B, Signal intensities for p300 and type I collagen were quantitated by densitometry, and results, corrected for variations in β-actin band intensities, are shown as the mean and SD fold increase relative to controls (dotted line). ∗ = P < 0.05 versus matched healthy controls.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Transforming growth factor β stimulates the expression of extracellular matrix genes and multiple other genes that are important in the pathogenesis of fibrosis (2). Furthermore, TGFβ induces the transdifferentiation of normal mesenchymal cells into profibrotic myofibroblasts (30). Together, these biologic activities position TGFβ as the master regulator of physiologic and pathologic matrix remodeling. In normal fibroblasts, TGFβ activates multiple distinct signal transduction pathways. In addition to Smads, these include MAP kinase, phosphatidylinositol 3-kinase, and protein kinase C signaling cascades (for review, see ref. 31). Despite the potential contribution of these intracellular cascades to mediating the biologic activities of TGFβ, profibrotic TGFβ responses appear to be uniquely dependent on Smad signaling (5). The widely expressed coactivator and histone acetylase p300 is implicated in modulating cellular responses elicited by cytokines and growth factors (32). Previous studies from our laboratory and others have implicated p300 in Smad-dependent TGFβ responses (6, 12, 25–28). As a coactivator, p300 does not directly bind to DNA; rather, it influences transcription by bridging DNA-bound transcription factors and the basal transcriptional machinery, or by modifying transcription factors and chromatin through their acetylation. Because the role of p300 in physiologic regulation of TGFβ-inducible gene expression remains poorly understood, we undertook the present studies to investigate the modulation of collagen synthesis and myofibroblast transdifferentiation, 2 important TGFβ-induced profibrotic responses, by p300.

The results demonstrated that TGFβ induced a substantial increase in p300 levels in normal dermal fibroblasts. Despite the recognition that abundance of p300 is limiting for p300-dependent transcriptional responses, and clear evidence that gene dosage is a critical determinant of p300 function (9), little is currently known regarding the regulation of p300 expression. The present results provide the first evidence that p300 expression is enhanced by TGFβ, with significant consequences for TGFβ-induced transcriptional responses. At the present time, these results do not allow us to determine whether increased p300 levels in TGFβ-treated fibroblasts reflect increased stability of the protein due to its posttranslational modification, or increased transcription of the p300 gene.

Forced expression of p300 in normal fibroblasts substantially enhanced TGFβ-induced stimulation of collagen promoter activity. In contrast, depletion of cellular p300 resulted in abrogation of this response. Suppression of collagen stimulation in the low-p300 cellular context was specific, since α-SMA expression induced by TGFβ was unaltered in p300-depleted fibroblasts, suggesting that TGFβ-regulated α-SMA gene expression may be independent of Smads or p300. The results from transient transfection assays indicated that depletion of p300 specifically inhibited Smad-dependent TGFβ responses, suggesting that the ability of p300 to enhance the magnitude of TGFβ responses involved its interaction with R-Smads. This notion is supported by our previous demonstration that forced expression of ectopic p300 enhanced the TGFβ stimulation of a promoter driven by a minimal SBE (12). TGFβ failed to induce Smad-dependent transcriptional responses in p300-depleted fibroblasts despite the fact that levels of cellular R-Smads, and their TGFβ-induced activation, were unaltered, suggesting a role for p300 in modulating the activator function of Smads in TGFβ signaling. This effect appears not to involve R-Smad acetylation by p300, since we previously showed that TGFβ did not induce Smad2 or Smad3 acetylation in dermal fibroblasts (33).

The levels of p300 were significantly higher in dermis from mice with sclerodermatous GVHD than control mice. Elevated p300 levels were detected primarily in fibroblasts within the fibrotic lesion. The association of dermal fibrosis with p300 overexpression suggested a possible causal link between elevated p300 levels and cutaneous fibrosis. Furthermore, levels of p300 were consistently elevated in fibroblasts derived from SSc patients compared with fibroblasts from age-, sex-, and race-matched healthy controls. These observations, demonstrating altered expression of p300 in fibrosis, are highly significant and consistent with recent reports. Liu et al documented elevated p300 expression in fibrotic tissue from mice with bleomycin-induced lung fibrosis (34). Furthermore, increased interaction between cellular Smad3 and p300 was recently described in SSc fibroblasts (35). Because TGFβ up-regulated the expression of p300 in normal fibroblasts, increased p300 levels seen in fibrosis may reflect increased TGFβ signaling in fibrotic tissues.

Taken together, these results demonstrate that the p300 histone acetylase and coactivator is a target of stimulation by TGFβ in dermal fibroblasts. Cellular p300 has an indispensable functional role in mediating TGFβ-induced collagen gene stimulation. Whereas decreasing the levels of cellular p300 was associated with reduced magnitude of TGFβ responses, increasing p300 was associated with marked enhancement. The findings suggest that cellular abundance of p300 may determine the intensity of TGFβ responses by setting the maximal levels of transcriptional stimulation attainable upon activation by TGFβ. The finding of elevated cellular p300 expression in SSc skin fibroblasts, and in lesional tissue from mice with experimentally induced fibrosis, provides support for implicating p300 in the fibrotic process. Enhanced p300 expression may contribute to persistence and progression of the fibrotic response. Approaches aimed at selectively modulating cellular p300 expression or function may therefore represent an innovative therapeutic strategy to control tissue fibrosis.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

We are grateful to Drs. K. K. Yokoyama (Institute of Physical and Chemical Research, RIKEN, Koyadai, Japan) for the gift of active and inactive p300 ribozymes, J. Boyes (Institute of Cancer Research, London, UK) for p300 plasmid constructs, L. Zawel (Johns Hopkins University, Baltimore, MD) for SBE4-TK-luc plasmid, and J. Massague (Howard Hughes Medical Institute, New York, NY) for 3TP-Lux plasmid.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES