вход по аккаунту


Smad1 pathway is activated in systemic sclerosis fibroblasts and is targeted by imatinib mesylate.

код для вставкиСкачать
Vol. 58, No. 8, August 2008, pp 2528–2537
DOI 10.1002/art.23698
© 2008, American College of Rheumatology
Smad1 Pathway Is Activated in Systemic Sclerosis Fibroblasts
and Is Targeted by Imatinib Mesylate
Jaspreet Pannu,1 Yoshihide Asano,1 Sashidhar Nakerakanti,1 Edwin Smith,1
Stefania Jablonska,2 Maria Blaszczyk,2 Peter ten Dijke,3 and Maria Trojanowska1
normalized the production of CCN2 and collagen. Imatinib mesylate blocked activation of the Smad1 pathway
in transforming growth factor ␤–stimulated control
fibroblasts and reversed activation of this pathway in
SSc fibroblasts. Likewise, blockade of c-Abl abrogated
activation of the Smad1 pathway in SSc fibroblasts.
Conclusion. Our findings demonstrate that activation of Smad1 signaling occurs in a subset of SSc
patients and contributes to persistent activation of SSc
fibroblasts. Demonstration that the Smad1/CCN2 pathway is blocked by imatinib mesylate further clarifies the
mechanism of the antifibrotic effects of this compound.
This study suggests that SSc patients with activated
Smad1 signaling may benefit from imatinib mesylate
Objective. Activation of Smad1 signaling has recently been implicated in the development of fibrosis.
The goal of the present study was to gain further
insights into activation of the Smad1 pathway in fibrosis
in systemic sclerosis (SSc) and to determine whether
this pathway is targeted by the antifibrotic drug imatinib mesylate.
Methods. Levels of phosphorylated Smad1 and
total Smad1 were examined in SSc and control skin
biopsy samples by immunohistochemistry and in cultured fibroblasts by Western blotting. Activity of the
CCN2 promoter was examined by a luciferase reporter
gene assay. Interactions of Smad1 with the CCN2
promoter were examined by in vitro and in vivo DNA
binding assays. Expression of the nonreceptor tyrosine
kinase c-Abl and Smad1 was blocked using respective
small interfering RNA.
Results. Total and phosphorylated Smad1 levels
were significantly elevated in SSc skin biopsy samples
and in cultured SSc fibroblasts and correlated with
elevated CCN2 protein and CCN2 promoter activity.
DNA binding assays demonstrated that Smad1 was a
direct activator of the CCN2 gene. Small interfering
RNA–mediated depletion of Smad1 in SSc fibroblasts
Fibrosis, affecting skin, lungs, and other organs, is
the most prominent feature of systemic sclerosis (SSc).
While many factors are known to contribute to fibrosis,
activation of the transforming growth factor ␤ (TGF␤)
signaling pathway is considered to play a central role in
this process (1). TGF␤ is a pleiotropic cytokine with
diverse functions in many cell types (2). The Smad
pathway is a primary mediator of TGF␤ signaling (3).
The canonical Smad pathway is activated upon binding
of TGF␤ to the heteromeric serine/threonine kinase
receptors TGF␤ receptor type II (TGF␤RII) and
TGF␤RI (activin receptor–like kinase 5 [ALK-5]), which
leads to phosphorylation of Smads 2 and 3, oligomerization with Smad4, nuclear translocation, and activation of
various transcription programs (2). There is also increasing evidence for the activation of other (noncanonical)
signaling pathways downstream of TGF␤, but the specific mechanisms involved in activation of Smadindependent pathways have not been fully elucidated
(4). With respect to fibrosis, the nonreceptor tyrosine
kinase c-Abl is of special interest, because of the potent
Dr. Pannu’s work was supported by the National Scleroderma
Foundation (New Investigator grant). Dr. ten Dijke’s work was
supported by the Dutch Cancer Society. Dr. Trojanowska’s work was
supported by the NIH (grants AR-42334 and AR-44883).
Jaspreet Pannu, PhD (current address: University of Maryland School of Medicine, Baltimore), Yoshihide Asano, MD, PhD,
Sashidhar Nakerakanti, PhD, Edwin Smith, MD, Maria Trojanowska,
PhD: Medical University of South Carolina, Charleston; 2Stefania
Jablonska, MD, Maria Blaszczyk, MD: Warsaw Medical Academy,
Warsaw, Poland; 3Peter ten Dijke, PhD: Leiden University Medical
Center, Leiden, The Netherlands.
Address correspondence and reprint requests to Maria Trojanowska, PhD, Division of Rheumatology and Immunology, Medical
University of South Carolina, 96 Jonathan Lucas Street, Suite 912,
Mail Stop Code 637, Charleston, SC 29425-6370. E-mail: trojanme@
Submitted for publication August 8, 2007; accepted in revised
form April 18, 2008.
antifibrotic effects of the pharmacologic inhibitor of
c-Abl, imatinib mesylate. Imatinib mesylate has been
shown to inhibit profibrotic effects of TGF␤ in cultured
cells, including SSc fibroblasts (5), and to prevent lung
and kidney fibrosis in experimental mouse models (6,7).
TGF␤ activation of c-Abl occurs through phosphatidylinositol 3-kinase and p21-activated kinase 2 and is
independent of the Smad2/3 pathway (8).
The role of TGF␤ signaling in the process of
fibrosis in SSc has been mainly studied using fibroblasts
explanted from SSc skin. During early passages in culture, these cells demonstrate an “activated phenotype”
characterized by elevated synthesis of extracellular matrix (ECM) proteins (9). Accumulating evidence suggests that alterations of TGF␤ signaling may play an
essential role in the persistent activation of SSc fibroblasts. The documented changes include up-regulation
of ␣v␤5 and ␣v␤3 integrins, which were shown to
contribute to activation of latent TGF␤ and establishment of the autocrine TGF␤ loop (10,11). Additional
changes include alterations of the TGF␤ receptor ratio
(12) and the presence of persistently phosphorylated
Smad3 (13). Recent studies from our laboratory using an
in vitro model of SSc based on the altered ratio of TGF␤
receptors have also established that activation of the
Smad1 pathway may represent a novel aspect of profibrotic TGF␤ signaling that functions independently of
activation of the canonical Smad2/3 pathway (14). Activation of Smad1 downstream of TGF␤ was first described in endothelial cells, where TGF␤ signals through
2 distinct type I receptors, ALK-5 and ALK-1, and their
respective signal transducers (Smads 2 and 3 for ALK-5
and Smads 1 and 5 for ALK-1) (15). Based on our recent
study, it appears that this mode of signaling is also
operational in fibroblasts, but the contribution of Smad1
signaling to SSc fibrosis has not been examined.
The present study was undertaken to further
evaluate the role of Smad1 signaling in SSc fibrosis.
Given that the inhibition of the TGF␤-mediated fibrotic
response by imatinib mesylate is independent of Smads
2 and 3, we also sought to determine whether Smad1 is
a downstream target of imatinib mesylate. Our results
show that Smad1 signaling is activated in a subset of SSc
fibroblasts. We also demonstrate that imatinib mesylate
blocks TGF␤-induced activation of Smad1 and ERK-1/2
pathways in control fibroblasts and induces phosphorylation of Smad1 and ERK-1/2 in SSc fibroblasts. The
present study provides novel insights into the mechanism
of fibrosis in SSc.
Immunohistochemistry. The study group consisted of
8 patients with diffuse cutaneous SSc (dcSSc), 5 patients with
limited cutaneous SSc (lcSSc), and 8 healthy volunteers (Table
1). All patients fulfilled the criteria of the American College of
Rheumatology (ACR; formerly, the American Rheumatism
Association) for SSc (16).
Upon obtaining informed consent and in compliance
with the Institutional Review Board for Human Studies, skin
biopsy samples were obtained from the affected areas (dorsal
forearm) of the SSc patients. Skin biopsy samples were embedded in paraffin and used for immunohistochemistry. Immunohistochemical staining of paraffin-embedded sections
was performed using a Vectastain ABC kit (Vector, Burlingame, CA) according to the manufacturer’s instructions. Fourmicrometer–thick sections were mounted on silane-coated
slides and were then deparaffinized with Histoclear and rehydrated through a graded series of solutions of ethyl alcohol and
phosphate buffered saline (PBS). Skin sections were treated
with hydrogen peroxide for 30 minutes to block endogenous
peroxidase activity and then subjected to a 45-minute antigenretrieval treatment with antigen unmasking solution (Vector).
For immunostainings, we used primary antibodies against total
Smad1 (Abcam, Cambridge, MA) and phospho-Smad1 (17).
Incubations with the primary antibodies diluted in horse serum
in PBS to the indicated concentrations were performed overnight in a humidified chamber at 4°C as previously described
(18). After 3 rinses in PBS, binding sites of the primary
antibodies were detected with biotinylated IgG, and the sites of
peroxidase activity were visualized by using diaminobenzidine.
The sections were then counterstained with hematoxylin. Immunostaining was detected by light microscopy. Normal rabbit
IgG was used as a negative control (data not shown).
Cell culture. Skin biopsy samples were obtained from
the affected areas of the dorsal forearm of 7 patients with
dcSSc, all of whom fulfilled the ACR criteria for dcSSc (16).
Normal adult dermal fibroblasts derived from 5 healthy donors
matched for age, sex, and race with each SSc patient served as
controls. These fibroblasts were analyzed in parallel with the
respective SSc cell lines. Pairs of SSc and normal fibroblast
samples were processed as they were obtained from the donors
to avoid freezing the cell cultures and to ensure that all cells in
passages 2–5 were used. Dermal fibroblasts were obtained
from healthy donor skin biopsy samples by enzymatically
dissociating tissue specimens in 0.25% type I collagenase
(Sigma, St. Louis, MO) and 0.05% DNase (Sigma) in Dulbecco’s modified Eagle’s medium (DMEM) with 20% fetal bovine
serum (FBS). Fibroblasts were maintained in DMEM supplemented with 10% FBS for all experiments. Before treatment
with TGF␤ or imatinib mesylate and infection with adenoviruses, fibroblasts were incubated in serum-free medium
(DMEM/0.1% bovine serum albumin) for 24 hours.
Western blotting. Confluent healthy dermal fibroblasts
were lysed, and protein content was determined as described
previously (19). Protein (40–50 ␮g) was separated via sodium
dodecyl sulfate–polyacrylamide gel electrophoresis (SDSPAGE) and transferred to a nitrocellulose membrane (BioRad, Hercules, CA). The membrane was then blocked at room
temperature using 3% milk/Tris buffered saline–Tween
(TBST) for 1 hour. The blots were probed overnight with a
Table 1. Total Smad1 and phospho-Smad1 staining of fibroblasts in the skin of SSc patients and healthy
Patients with dcSSc
Patients with lcSSc
Healthy subjects
Rodnan skin
thickness score
(range 0–51)
No. of total
No. of phosphoSmad1–positive
18 years
5 years
2 years
1 year
21 years
* SSc ⫽ systemic sclerosis; dcSSc ⫽ diffuse cutaneous SSc; lcSSc ⫽ limited cutaneous SSc.
† At least 100 fibroblasts were counted in each specimen.
1:1,000 dilution of appropriate primary antibodies against
Smad1 and phospho-Smad1 (both from Cell Signaling Technology, Beverly, MA) and against TGF␤RI, TGF␤RII, and
CCN2 (all from Santa Cruz Biotechnology, Santa Cruz, CA).
Following washes with TBST, blots were incubated with an
appropriate horseradish peroxidase–conjugated secondary antibody. As a control for equal protein loading, membranes
were stripped and reprobed for ␤-actin using a monoclonal
antibody to ␤-actin (Sigma). Protein levels were visualized
using ECL reagent (Pierce, Rockford, IL) and quantitated
using densitometry software (NIH Image, National Institutes
of Health, Bethesda, MD; online at:
Transient transfection, luciferase assay, and RNA
interference. SSc and normal dermal fibroblasts were seeded
on 6-well plates (105 cells/well) and transfected 24 hours later.
Transient transfections with the previously described wild-type
(WT) CCN2 promoter construct (14) (a gift from Dr. Gary
Grotendorst) were performed in duplicate using FuGene 6
reagent (Roche, Indianapolis, IN) according to the manufacturer’s specifications. Two SSc and healthy control fibroblast
strains were transiently transfected with the CCN2-Luc promoter construct. Transfections were normalized using the
pSV-␤-galactosidase control vector. Luciferase activities in
aliquots normalized for equal protein concentrations were
determined 24 hours after transfection using the Luciferase
Assay System (Promega, Madison, WI).
Small interfering RNA (siRNA) directed against c-Abl
was purchased from Santa Cruz Biotechnology, along with the
specific primers for real-time quantitative polymerase chain
reaction (PCR). Nonsilencing siRNA used for this experiment
was purchased from Qiagen (Chatsworth, CA). Confluent
cultures of normal and SSc fibroblasts were transfected with
the indicated siRNA using RNAiFect reagent (Qiagen) for 48
hours. The cells were then lysed in radioimmunoprecipitation
assay (RIPA) buffer, and the amount of protein was estimated
in each sample as described above. Western blot analysis was
performed for phospho-Smad1 and total Smad1 as well as for
type I collagen protein levels. The experiments using siRNA
directed against Smad1 were performed as previously described (14). Quantitative real-time PCR was performed to
determine the levels of c-Abl and Smad1 depletion (14).
DNA affinity precipitation assay. Confluent dishes of
dermal fibroblasts were either left untreated or were treated
with TGF␤ (1 ng/ml) for 90 minutes. Cell pellets were lysed in
RIPA buffer, and 500 mg of whole cell extract was incubated
with 20 ␮g of poly(dI-dC) in 1⫻ DNA affinity precipitation
assay binding buffer (10 mM Tris HCl [pH 8.0], 40 mM KCl, 1
mM dithiothreitol, 6% glycerol, and 0.05% Nonidet P40) for 10
minutes at 4°C. The extracts were precleared with streptavidincoated agarose beads (Pierce), and the supernatants were
incubated with 1.5 ␮g of double-stranded WT connective tissue
growth factor (CTGF) (WT CCN2) or CCN2 oligonucleotide
mutated at the Sp-1 site (mutated Sp-1 CTGF) (both labeled
with biotin at the 5⬘ end) for 4 hours at 4°C with constant
rotation. The sequences of the forward oligonucleotides were
Figure 1. Immunodetection of Smad1 and phospho-Smad1 in skin samples from healthy individuals and from patients with systemic sclerosis (SSc).
A–D, Sections of normal skin (A and C) and SSc skin (B and D) were stained for total Smad1 (A and B) and phospho-Smad1 (C and D). Thick arrows
indicate positively stained fibroblasts; thin arrows indicate positively stained endothelial cells (original magnification ⫻ 400). E, Percentage of
Smad1-positive fibroblasts and phospho-Smad1–positive fibroblasts in SSc and normal healthy skin. Each symbol represents an individual subject.
At least 100 fibroblasts were counted for each specimen. The percentage of Smad1-positive fibroblasts and phospho-Smad1–positive fibroblasts was
higher in skin from patients with diffuse cutaneous SSc (dcSSc) than in skin from healthy individuals (P ⬍ 0.01 and P ⬍ 0.006, respectively). lcSSc ⫽
limited cutaneous SSc.
5⬘-GCC-TCG-GCC-GCC-CGC-CCC-AAA-CTC-ACA-CAAC-3⬘ for WT CTGF and 5⬘-GCC-TCG-GGC-CGC-TTC-CTCAAA-CTC-ACA-CAA-C-3⬘ for mutated Sp-1 CTGF. Poly(dIdC) (3 ␮g) was used as a negative control. Streptavidin-coated
agarose beads were added to each tube, and the samples were
rotated for an additional 2 hours at 4°C. Beads were recovered
by centrifugation and washed twice with Tris–EDTA followed
by 2 washes with 1⫻ DNA affinity precipitation assay binding
buffer and 2 washes with 1⫻ PBS. Proteins were eluted by
heating the beads with 2⫻ sample loading buffer at 95°C for
5 minutes. The supernatants were separated by 10% SDSPAGE and transferred onto nitrocellulose membrane. Western blotting was performed with antibody against Smad1
(Santa Cruz Biotechnology), and the bands were visualized
using enhanced chemiluminescence reagent.
Chromatin immunoprecipitation (ChIP). ChIP was
performed as described previously (20). Adult dermal fibroblasts were treated with either AdGFP or AdTGF␤RI
(AdALK-5) for 18 hours. Half of the AdGFP-treated cells
were also treated with TGF␤ (1 ng/ml) for 90 minutes. The
cells were crosslinked and lysed, and nuclei were isolated as
described previously (20). Chromatin was sheared by sonication to yield an average fragment size of 400 bp and microcentrifuged at 14,000 revolutions per minute, and supernatant
containing soluble chromatin was collected. The chromatin
solution was precleared with protein G–Sepharose for 30
minutes at 4°C and subsequently incubated with 4 ␮g of
antibody against Smad1 at 4°C overnight with constant rotation. Recovery of immunoprecipitated chromatin was done
using protein G–Sepharose beads for 2 hours at 4°C. The
chromatin–antibody–protein G–Sepharose complexes were recovered by centrifugation and washed, and the DNA was
eluted. DNA was purified using the QIAquick PCR purification kit (Qiagen). One microliter of the purified DNA was
used in each PCR. The PCR reaction was carried out for either
28 or 33 cycles, and the product was visualized by agarose gel
electrophoresis. A 213-bp region (–194 to ⫹19) of the CCN2
promoter encompassing the Sp-1 site was amplified using the
primers described previously (20).
Activation of the Smad1 pathway in SSc skin
biopsy samples. To investigate the involvement of the
Smad1 pathway in SSc fibrosis, we performed immunostaining for total Smad1 and phosphorylated Smad1 in
skin specimens from 8 patients with dcSSc, 5 patients
with lcSSc, and 8 healthy controls. Representative staining of SSc and healthy control skin samples is shown in
Figures 1A–D. Smad1- and phospho-Smad1–positive
fibroblasts were counted in each specimen, and the
summary of the results is shown in Figure 1E and Table
1. The analysis revealed heterogeneity of Smad1 and
phospho-Smad1 expression among SSc and control skin
sections; however, a significantly higher proportion of
positive fibroblasts was found in the skin of dcSSc
patients. The mean ⫾ SEM percentage of Smad1positive fibroblasts was 54.8 ⫾ 10% in dcSSc skin
compared with 22 ⫾ 5% in healthy skin (P ⬍ 0.01), and
the mean ⫾ SEM percentage of phospho-Smad1–
positive fibroblasts was 41 ⫾ 11% in dcSSc skin compared with 6 ⫾ 3% in healthy skin (P ⬍ 0.006) (Figure
1E). The expression of Smad1 and phospho-Smad1 in
patients with lcSSc was moderately elevated, but the
differences were not statistically significant. These find-
ings strongly suggest that activation of the Smad1 pathway contributes to the development of fibrosis in dcSSc.
Activation of the Smad1 pathway in cultured SSc
fibroblasts. Cultured SSc fibroblasts at early passages
retain some of the characteristics of the activated phenotype, including elevated levels of collagen and CCN2
genes (9). We examined the Smad1 phosphorylation
status and total Smad1 levels in 7 pairs of SSc and closely
matched control dermal fibroblast samples. Results
from 3 pairs of representative SSc and control fibroblast
samples are shown in Figure 2A. Phospho-Smad1 was not
detectable in control dermal fibroblasts, but was present
in 6 of 7 SSc fibroblast strains (results for the remaining
SSc and control strains are shown below). The level of
total Smad1 was markedly increased in all SSc strains
compared with that in control fibroblasts (summarized
in Figure 2B). The elevated levels of Smad1 correlated
with increased expression of CCN2 (Figure 2A). Consistent with a previous report (21), SSc fibroblasts exhibited elevated CCN2 promoter activity, which was resistant to further stimulation with TGF␤ (Figure 2C).
CTGF/CCN2 gene is a direct effector of the
Smad1 pathway. We have recently shown that overexpression of TGF␤RI (AdALK-5) in dermal fibroblasts
leads to up-regulation of CCN2 promoter activity
through the GC-rich motif, which was previously shown
to mediate Sp-1–dependent stimulation (14). Since this
response required Smad1, we wanted to determine
whether Smad1 binds directly to the proximal CCN2
promoter region. We first assessed Smad1 binding using
the DNA affinity precipitation assay. Fragments (30-bp)
of the CCN2 promoter corresponding to the WT and the
GC-rich motif promoter mutant were labeled with biotin
at the 5⬘ end and used in binding reactions with cell
extracts from untreated and TGF␤-treated fibroblasts
(Figure 3A). Smad1 bound to the CCN2 promoter in
untreated cells, and the binding was further increased
after TGF␤ treatment. Mutating the GC-rich motif almost
completely abrogated Smad1 binding. Since a previous
study using electrophoretic mobility shift assay analyses
suggested that Sp-1 interacts with the proximal GC-rich
motif (21), we also investigated the binding of Sp-1 using
the DNA affinity precipitation assay; however, binding of
Sp-1 to this promoter region was not detected under our
experimental condition (data not shown). The reason for
this discrepancy is currently not known.
To assess the interaction of Smad1 with the
endogenous CCN2 promoter, we used a ChIP assay. In
untreated cells, low levels of Smad1 were detectable at
the CCN2 promoter (Figure 3B); however, promoter
occupancy by Smad1 was increased after TGF␤ stimu-
pathways but is independent of the activation of canonical Smad2/3 signaling (14). Constitutive activation of
ERK-1/2 signaling is also observed in cultured SSc
fibroblasts (22,23). Since imatinib mesylate blocks fibrosis through a mechanism that is also independent of
Smad2/3, we wished to determine whether imatinib
mesylate affects activation of the Smad1 and/or ERK-1/2
The effects of imatinib mesylate were first tested
in control fibroblasts stimulated with TGF␤ and in our
Figure 2. Correlation of Smad1 and CCN2 expression in systemic
sclerosis (SSc) fibroblasts. A, Three pairs of fibroblast samples from
SSc patients and closely matched normal subjects (NS) were analyzed
for the presence of phospho-Smad1, total Smad1, and CCN2 by
Western blotting. Beta-actin was used as loading control. B, PhosphoSmad1 (pSmad1) and total Smad1 (TSmad1) protein levels were
determined in 7 pairs of normal and SSc fibroblast cultures. ⴱ ⫽ P ⬍
0.0018; ⴱⴱ ⫽ P ⬍ 0.0078, versus normal fibroblast cultures. C, CCN2
promoter activity in SSc fibroblasts was measured. Two SSc and
healthy control fibroblast strains were transiently transfected with the
CCN2-Luc promoter construct. Transfections were normalized using
the pSV-␤-galactosidase control vector. The activity of the CCN2
promoter in unstimulated control fibroblasts was arbitrarily set at 1.
The graph represents fold change in promoter activity in unstimulated
SSc fibroblasts and in response to stimulation with transforming
growth factor ␤ (TGF␤). Values in B and C are the mean and SEM.
lation. Constitutively elevated Smad1 binding was observed in cells transduced with AdALK-5, which is
consistent with the activation of the Smad1 pathway in
these cells (14). These data indicate that Smad1 is a
direct activator of the CCN2 gene in dermal fibroblasts.
ERK-1/2–Smad1 pathway is targeted by imatinib
mesylate. We have previously demonstrated that in our
experimental model of SSc based on an altered ratio of
TGF␤ receptors, the profibrotic response is mediated by
a persistent activation of Smad1 and ERK-1/2 signaling
Figure 3. Smad1 binds to the GC-rich motif in the proximal CCN2
promoter region in vitro and in vivo. A, DNA affinity precipitation
assay showing the binding of Smad1 to the 5⬘ end–labeled wild-type
CCN2 oligonucleotide probe (CCN2-oligo WT), but not the 5⬘ end–
labeled CCN2 oligonucleotide mutated at the Sp-1 site (CCN2-oligo
Mut.), in both the absence and presence of transforming growth factor
␤ (TGF␤) (top). C ⫽ control. Levels of total Smad1 in the samples
before the DNA affinity precipitation assay (input) are also shown.
Fifteen percent of the input samples were loaded. Mean and SEM
relative Smad1 levels are at the bottom. B, Chromatin immunoprecipitation assay of CCN2 promoter after stimulation for 90 minutes with
TGF␤ (1 ng/ml) of cells transduced with AdGFP or cells transduced
with AdALK-5 with no stimulation. A polyclonal antibody to Smad1
was used for immunoprecipitation. Shown are polymerase chain
reaction (PCR) analyses of the input (control), in the absence of
antibody (No Ab.), and after immunoprecipitation with antibody using
primers corresponding to the proximal region of the CCN2 promoter
encoding the Smad1 site. An increased number of PCR cycles (33) was
used to visualize Smad1 binding in control cells in the absence of
Figure 4. Imatinib mesylate blocks transforming growth factor ␤
(TGF␤)–induced phosphorylation of ERK-1/2–Smad1 and upregulation of the CCN2 gene in dermal fibroblasts. Confluent dishes of
adult fibroblasts were transduced with AdALK-5 (lanes 3 and 4) or
control adenovirus (AdGFP) (lanes 1 and 2) as previously described
(14). AdGFP-transduced cells were also stimulated with TGF␤ (1
ng/ml) for 45 minutes (lanes 5 and 6) or 24 hours (lanes 7 and 8).
Where indicated, imatinib mesylate (2 ␮g/ml) was added to cells 1 hour
prior to TGF␤ stimulation or 6 hours after transduction with
AdALK-5. Cell lysates were obtained as described in Patients and
Methods, and Western blot analysis was carried out for CCN2,
phospho-Smad1, total Smad1, phospho–ERK-1/2, and total ERK-1/2
protein levels. The blots were stripped and reprobed with antibody to
␤-actin for normalization. p44/42 MAPK ⫽ ERK-1/2.
ALK-5–based experimental model of SSc (Figure 4).
Consistent with the previous study, overexpression of
ALK-5 resulted in up-regulation of CCN2 protein levels,
increased phospho-Smad1 and total Smad1 levels, and
increased phospho–ERK-1/2 levels (compare lanes 1
and 3 in Figure 4). Treatment with imatinib mesylate
abrogated the up-regulation of CCN2, phospho-Smad1,
and phospho–ERK-1/2 (compare lanes 3 and 4), while
elevated total Smad1 levels were not affected by this
treatment. We next examined the effects of imatinib
mesylate on TGF␤-induced phosphorylation of Smad1
and ERK-1/2 and up-regulation of the CCN2 gene. Two
time points, 45 minutes and 24 hours, were examined in
order to evaluate both short- and long-term responses.
Treatment with imatinib mesylate blocked induction of
both phospho-Smad1 and phospho–ERK-1/2 (compare
lanes 5 and 6). Up-regulation of the CCN2 gene observed 24 hours after stimulation was also blocked by
imatinib mesylate (compare lanes 7 and 8).
Contribution of c-Abl to activation of the Smad1
pathway in SSc fibroblasts. The effects of imatinib
mesylate were next tested in 2 pairs of SSc and control
fibroblasts. Consistent with our previous reports, SSc
fibroblasts had elevated levels of TGF␤RI compared
with control fibroblasts (Figure 5A). Increased levels of
phosphorylated Smad1 and ERK-1/2, as well as increased total Smad1 and CCN2 levels, were present in
both SSc strains (Figure 5B). Treatment with imatinib
mesylate reversed the phosphorylation of Smad1 and
ERK-1/2 but did not normalize the elevated total Smad1
levels. Up-regulation of CCN2 was completely normalized in one of the pairs and partially inhibited in the
other pair. The heterogeneous response of SSc fibroblasts may reflect multiple mechanisms involved in upregulation of CCN2 in SSc fibroblasts.
Because imatinib mesylate has an inhibitory activity toward 3 kinases, c-Abl, c-Kit, and platelet-derived
growth factor receptor (PDGFR) and because both
c-Abl and PDGFR have been linked to SSc fibrosis, we
next determined whether blockade of c-Abl affects
Smad1 phosphorylation. As shown in Figure 6A, downregulation of c-Abl expression levels using specific
siRNA completely abolished Smad1 phosphorylation
and normalized collagen production in SSc fibroblasts.
These data demonstrate that c-Abl is required for the
activation of Smad1 in SSc fibroblasts. To confirm the
contribution of the Smad1 pathway to SSc fibrosis,
Smad1 was blocked in SSc and control fibroblasts using
Figure 5. Imatinib mesylate inhibits activation of the ERK-1/2–
Smad1 pathway in systemic sclerosis (SSc) fibroblasts. A, Expression of
transforming growth factor ␤ receptor I (TGF␤R I) and TGF␤RII was
analyzed by Western blotting in 2 pairs (pairs 4 and 5) of fibroblast
samples from SSc patients and normal subjects (NS). B, Cells were
treated with 2 ␮g/ml of imatinib mesylate 24 hours prior to cell lysis.
Western blot analysis was carried out for CCN2, phospho-Smad1, total
Smad1, phospho–ERK-1/2, and total ERK-1/2 protein levels. The
blots were stripped and reprobed with antibody to ␤-actin for normalization. p44/42 MAPK ⫽ ERK-1/2.
Figure 6. A, The nonreceptor tyrosine kinase c-Abl contributes to activation of the Smad1 pathway in systemic sclerosis (SSc) fibroblasts. Two pairs
of fibroblast samples (pairs 6 and 7) from SSc patients and normal subjects (NS) were treated either with control nonsilencing small interfering RNA
(siRNA) (Scr.) or with siRNA specifically directed against c-Abl (siAbl) for 48 hours prior to cell lysis. Western blot analysis was performed to
determine phospho-Smad1 (pSmad1) and total Smad1 (TSmad1) levels as well as type I collagen levels. The blots were stripped and reprobed with
antibody to ␤-actin for normalization (top). The down-regulation of c-Abl in fibroblasts in response to siAbl was quantified by real-time quantitative
polymerase chain reaction (PCR) (bottom) as described in Patients and Methods. B, Smad1 contributes directly to regulation of CCN2 and type I
collagen protein levels. Two pairs of normal and SSc fibroblasts were treated either with nonsilencing siRNA (Scr.) or with siRNA directed against
Smad1 (siS1) as described previously (14). Western blot analysis was performed to determine CCN2 and type I collagen levels. The blots were
stripped and reprobed with antibody to ␤-actin for normalization (top). Smad1 levels were quantified by real-time quantitative PCR (bottom) using
specific primers as described previously (14). Values in the lower panels in A and B are the mean and SEM.
previously developed Smad1 siRNA adenoviruses.
Blockade of Smad1 normalized the elevated CCN2
and collagen levels in SSc fibroblasts to levels observed
in the corresponding healthy controls (Figure 6B). Together, these data support the conclusion that activation
of the c-Abl/Smad1 signaling pathway contributes to SSc
TGF␤ is considered a central mediator of the
excessive collagen deposition in SSc, but the molecular
mechanisms underlying the chronic fibrotic response are
not fully elucidated. This study provides further support
for the contribution of a novel fibrogenic pathway
mediated by the activation of the ERK-1/2–Smad1–
CCN2 cascade to SSc fibrosis. Analyses of SSc skin
samples revealed elevated expression of total and
phospho-Smad1 protein in a subset of patients with
active disease. Fibroblasts cultured from SSc biopsy
samples demonstrated elevated levels of Smad1 protein,
and in 6 of 7 strains tested also showed the presence of
phosphorylated Smad1, indicating that activation of the
Smad1 pathway persists in cultured cells. Since several
previous studies have shown elevated levels of phosphoSmad3 in SSc fibroblasts (13,24), these data suggest that
distinct Smad pathways may play complementary roles
in the activation of cultured SSc fibroblasts. Surprisingly,
however, comparison of SSc and healthy skin tissues
showed similar levels of phosphorylated Smad3 (24).
The reason for the different expression of phosphorylated Smad3 in cells in vivo and in cultured cells remains
In vitro and in vivo DNA binding assays have
demonstrated that Smad1 binds to the GC-rich motif in
the proximal CCN2 promoter region. Although this site
was previously characterized as an Sp-1 binding site (21),
our data strongly suggest that this GC-rich motif represents a Smad1 binding site. This finding is also consistent
with the previous studies that showed binding of Smad1
to GC-rich motifs in several other promoters (25). Our
recent studies have shown that siRNA-mediated inhibition of Smad1 abrogated TGF␤-induced up-regulation
of CCN2 messenger RNA and protein levels as well as
AdTGF␤RI-induced CCN2 promoter activity (14). In
the present study, we extended these analyses to SSc
fibroblasts and demonstrated that the Smad1 pathway
also contributes to activation of profibrotic genes in
these cells. Together with the results of the DNA
binding assays, these studies indicate that the CCN2
gene is a direct effector of the Smad1 pathway. CCN2 is
required for the profibrotic effects of TGF␤ in cultured
cells and is considered to be a primary mediator of
chronic fibrotic processes in SSc skin (26). Thus, activation of the Smad1 pathway may contribute to maintaining constitutively elevated expression levels of CCN2 in
SSc fibroblasts in vivo and in vitro.
TGF␤-dependent activation of the Smad1 pathway was first described in endothelial cells (27). In these
cells, TGF␤ signals through a heteromeric receptor
complex that includes TGF␤RII, ALK-5, and ALK-1
(28). While the specific interactions between ALK-5 and
ALK-1 are still not fully elucidated, there is evidence
that the relative ratio of these 2 receptor subtypes
regulates the switch between proliferation and differentiation of endothelial cells (15). However, recent studies
have also shown that the ALK-1/Smad1 pathway may
play a role in kidney fibrosis (29–31). Furthermore,
TGF␤ signaling via activation of the ALK-1/Smad1
pathway and subsequent up-regulation of the Id1 gene
have been shown to contribute to transdifferentiation of
hepatic stellate cells into myofibroblasts (32). These
recent studies strongly suggest that activation of the
ALK-1/Smad1 pathway may represent a common feature of various fibrotic diseases and that this pathway
may play an important role in the development of organ
fibrosis, including SSc.
Imatinib mesylate is a tyrosine kinase inhibitor
that inhibits a wide range of kinases, including c-Abl (or,
Bcr-Abl), c-Kit, and PDGFR. Imatinib mesylate has
been effective in preventing the development of organ
fibrosis in the kidney, lung, liver, and skin in several
animal models (5–7); however, it was significantly less
effective in ameliorating an established fibrosis (33,34).
The antifibrotic effects of imatinib mesylate include
inhibition of fibroblast proliferation via blockade of
PDGFR kinase and inhibition of collagen production in
response to TGF␤. While it has been established that
the antifibrotic effects of imatinib mesylate are mediated
through inhibition of c-Abl, the role of this nonreceptor
tyrosine kinase in collagen regulation is currently unknown. Our findings suggest that Smad1 may represent
a downstream effector of the c-Abl signaling cascade in
dermal fibroblasts. Further studies are needed to clarify
the mechanism by which c-Abl exerts its profibrotic
Excessive deposition of collagen and other ECM
components is the hallmark of SSc, occurring in all
target organs and progressively interrupting circulation.
Effective antifibrotic treatments for SSc and other fibrotic diseases are not available at present. Due to its
prominent role in matrix regulation, TGF␤ signaling is
considered an attractive therapeutic target for controlling fibrosis. However, the complexity of the profibrotic
effects of TGF␤ that involve both Smad3-dependent and
-independent mechanisms presents a challenge in selecting appropriate therapeutic targets. For example, pharmacologic inhibitors of TGF␤RI kinase that result in
blockade of Smad3 signaling (ALK-5 kinase inhibitors)
did normalize elevated CCN2 levels in SSc fibroblasts
and, in general, ameliorated only selected aspects of SSc
fibrosis in vitro (35,36). Our study suggests that due to
the pleiotropic nature of TGF␤ signaling, targeting a
single component of the TGF␤ signaling pathway may
not be effective in ameliorating SSc fibrosis. The Smad1
pathway may represent, in addition to Smad3, a potential target for antifibrotic therapy in SSc and other
fibrotic diseases.
Dr. Trojanowska 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. Pannu, Smith, Jablonska, Blaszczyk, Trojanowska.
Acquisition of data. Pannu, Asano, Nakerakanti, Smith, Jablonska,
Analysis and interpretation of data. Pannu, Asano, Nakerakanti, ten
Dijke, Trojanowska.
Manuscript preparation. Pannu, ten Dijke, Trojanowska.
Statistical analysis. Pannu, Asano.
Supply and characterization of key reagents. Ten Dijke.
1. Pannu J, Trojanowska M. Recent advances in fibroblast signaling
and biology in scleroderma. Curr Opin Rheumatol 2004;16:
2. Massague J. TGF-␤ signal transduction. Annu Rev Biochem
3. Miyazono K, ten Dijke P, Heldin CH. TGF-␤ signaling by Smad
proteins. Adv Immunol 2000;75:115–57.
4. Moustakas A, Heldin CH. Non-Smad TGF-␤ signals. J Cell Sci
5. 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.
6. Daniels CE, Wilkes MC, Edens M, Kottom TJ, Murphy SJ,
Limper AH, et al. Imatinib mesylate inhibits the profibrogenic
activity of TGF-␤ and prevents bleomycin-mediated lung fibrosis.
J Clin Invest 2004;114:1308–16.
7. Wang S, Wilkes MC, Leof EB, Hirschberg R. Imatinib mesylate
blocks a non-Smad TGF-␤ pathway and reduces renal fibrogenesis
in vivo. FASEB J 2005;19:1–11.
8. Wilkes MC, Leof EB. Transforming growth factor ␤ activation of
c-Abl is independent of receptor internalization and regulated by
phosphatidylinositol 3-kinase and PAK2 in mesenchymal cultures.
J Biol Chem 2006;281:27846–54.
Trojanowska M. What did we learn by studying scleroderma
fibroblasts? Clin Exp Rheumatol 2004;22 Suppl 33:S59–63.
Asano Y, Ihn H, Yamane K, Jinnin M, Mimura Y, Tamaki K.
Increased expression of integrin ␣v␤3 contributes to the establishment of autocrine TGF-␤ signaling in scleroderma fibroblasts.
J Immunol 2005;175:7708–18.
Asano Y, Ihn H, Yamane K, Kubo M, Tamaki K. Increased
expression levels of integrin ␣v␤5 on scleroderma fibroblasts. Am J
Pathol 2004;164:1275–92.
Pannu J, Gardner H, Shearstone JR, Smith E, Trojanowska M.
Increased levels of transforming growth factor ␤ receptor type I
and up-regulation of matrix gene program: a model of scleroderma. Arthritis Rheum 2006;54:3011–21.
Mori Y, Chen SJ, Varga J. Expression and regulation of intracellular SMAD signaling in scleroderma skin fibroblasts. Arthritis
Rheum 2003;48:1964–78.
Pannu J, Nakerakanti S, Smith E, ten Dijke P, Trojanowska M.
Transforming growth factor-␤ receptor type I-dependent fibrogenic gene program is mediated via activation of Smad1 and
ERK1/2 pathways. J Biol Chem 2007;282:10405–13.
Goumans MJ, Lebrin F, Valdimarsdottir G. Controlling the
angiogenic switch: a balance between two distinct TGF-␤ receptor
signaling pathways. Trends Cardiovasc Med 2003;13:301–7.
Subcommittee for Scleroderma Criteria of the American Rheumatism Association Diagnostic and Therapeutic Criteria Committee. Preliminary criteria for the classification of systemic sclerosis
(scleroderma). Arthritis Rheum 1980;23:581–90.
Persson U, Izumi H, Souchelnytskyi S, Itoh S, Grimsby S, Engstrom U, et al. The L45 loop in type I receptors for TGF-␤ family
members is a critical determinant in specifying Smad isoform
activation. FEBS Lett 1998;434:83–7.
Kubo M, Czuwara-Ladykowska J, Moussa O, Markiewicz M,
Smith E, Silver RM, et al. Persistent down-regulation of Fli1, a
suppressor of collagen transcription, in fibrotic scleroderma skin.
Am J Pathol 2003;163:571–81.
Gore-Hyer E, Pannu J, Smith EA, Grotendorst G, Trojanowska
M. Selective stimulation of collagen synthesis in the presence of
costimulatory insulin signaling by connective tissue growth factor
in scleroderma fibroblasts. Arthritis Rheum 2003;48:798–806.
Nakerakanti SS, Kapanadze B, Yamasaki M, Markiewicz M,
Trojanowska M. Fli1 and Ets1 have distinct roles in connective
tissue growth factor/CCN2 gene regulation and induction of the
profibrotic gene program. J Biol Chem 2006;281:25259–69.
Holmes A, Abraham DJ, Chen Y, Denton C, Shi-wen X, Black
CM, et al. Constitutive connective tissue growth factor expression
in scleroderma fibroblasts is dependent on Sp1. J Biol Chem
Svegliati S, Cancello R, Sambo P, Luchetti M, Paroncini P,
Orlandini G, et al. Platelet-derived growth factor and reactive
oxygen species (ROS) regulate Ras protein levels in primary
human fibroblasts via ERK1/2: amplification of ROS and Ras in
systemic sclerosis fibroblasts. J Biol Chem 2005;280:36474–82.
Tourkina E, Gooz P, Pannu J, Bonner M, Scholz D, Hacker S, et
al. Opposing effects of protein kinase C␣ and protein kinase C⑀ on
collagen expression by human lung fibroblasts are mediated via
MEK/ERK and caveolin-1 signaling. J Biol Chem 2005;280:
Verrecchia F, Laboureau J, Verola O, Roos N, Porcher R,
Bruneval P, et al. Skin involvement in scleroderma—where histological and clinical scores meet. Rheumatology (Oxford) 2007;46:
Massague J, Seoane J, Wotton D. Smad transcription factors.
Genes Dev 2005;19:2783–810.
Shi-wen X, Pennington D, Holmes A, Leask A, Bradham D,
Beauchamp JR, et al. Autocrine overexpression of CTGF maintains fibrosis: RDA analysis of fibrosis genes in systemic sclerosis.
Exp Cell Res 2000;259:213–24.
Goumans MJ, Valdimarsdottir G, Itoh S, Rosendahl A, Sideras P,
ten Dijke P. Balancing the activation state of the endothelium via
two distinct TGF-␤ type I receptors. EMBO J 2002;21:1743–53.
Goumans MJ, Valdimarsdottir G, Itoh S, Lebrin F, Larsson J,
Mummery C, et al. Activin receptor-like kinase (ALK)1 is an
antagonistic mediator of lateral TGF␤/ALK5 signaling. Mol Cell
Abe H, Matsubara T, Iehara N, Nagai K, Takahashi T, Arai H, et
al. Type IV collagen is transcriptionally regulated by Smad1 under
advanced glycation end product (AGE) stimulation. J Biol Chem
Takahashi T, Abe H, Arai H, Matsubara T, Nagai K, Matsuura M,
et al. Activation of STAT3/Smad1 is a key signaling pathway for
progression to glomerulosclerosis in experimental glomerulonephritis. J Biol Chem 2005;280:7100–6.
Matsubara T, Abe H, Arai H, Nagai K, Mima A, Kanamori H, et
al. Expression of Smad1 is directly associated with mesangial
matrix expansion in rat diabetic nephropathy. Lab Invest 2006;86:
Wiercinska E, Wickert L, Denecke B, Said HM, Hamzavi J,
Gressner AM, et al. Id1 is a critical mediator in TGF-␤-induced
transdifferentiation of rat hepatic stellate cells. Hepatology 2006;
Neef M, Ledermann M, Saegesser H, Schneider V, Widmer N,
Decosterd LA, et al. Oral imatinib treatment reduces early
fibrogenesis but does not prevent progression in the long term.
J Hepatol 2006;44:167–75.
Vittal R, Zhang H, Han MK, Moore BB, Horowitz JC, Thannickal
VJ. Effects of the protein kinase inhibitor, imatinib mesylate, on
epithelial/mesenchymal phenotypes: implications for treatment of
fibrotic diseases. J Pharmacol Exp Ther 2007;321:35–44.
Chen Y, Shi-wen X, Eastwood M, Black CM, Denton CP, Leask
A, et al. Contribution of activin receptor–like kinase 5 (transforming growth factor ␤ receptor type I) signaling to the fibrotic
phenotype of scleroderma fibroblasts. Arthritis Rheum 2006;54:
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
Без категории
Размер файла
634 Кб
mesylate, systemic, smad, imatinib, targeted, sclerosis, pathways, activated, fibroblasts
Пожаловаться на содержимое документа