| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the MRC Group In Periodontal Physiology, Faculty of Dentistry, University of Toronto, Toronto, Ontario, Canada
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-smooth-muscle actin (SMA). During wound
maturation, these cells disappear. We have examined the
mechanisms that lead to myofibroblast deletion in a fibroblast culture
model. Transforming growth factor-ß (TGF-ß) was used to increase
SMA content in gingival fibroblasts (three- to sixfold). After
replating TGF-ß-induced cells at low density with serum,
there was a fivefold decrease in SMA protein content, SMA
protein synthesis, and SMA mRNA as cells proliferated. These
reductions were due to reduced SMA mRNA stability. For TGF-ß-induced
cells plated at high density without serum (ie, quiescent
conditions), protein content was reduced by only 20% over 12
days. TGF-ß protected SMA-positive cells against apoptosis in
serum-free cultures. Those cells that were protected against apoptosis
exhibited well-developed stress fibers enriched in SMA. We conclude
that, in quiescent myofibroblasts, SMA protein turnover
is slow, and cells are long-lived. In proliferative conditions
SMA protein and mRNA turn over quickly, and the myofibroblast
phenotype dissipates. The reduced apoptosis of myofibroblasts in
quiescent conditions is due in part to the organization of SMA into
stress fibers.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-Smooth-muscle actin
(SMA) is an actin isoform that is strongly expressed in
myofibroblasts3,4,5
and that may mark discrete
subpopulations of fibroblasts with constitutively high levels of
contractile capacity.6
SMA-containing myofibroblasts are
found in fibrocontractive lesions but are present only transiently
during healing of dermal wounds.7
In culture, fibroblasts
may acquire several phenotypic features of myofibroblasts including
stress fibers.8
These force-generating elements are often
enriched with SMA6
and may have in vivo
counterparts that are involved in wound contraction.9
In vivo, wound myofibroblasts are thought to arise locally
from quiescent fibroblasts without SMA.3
These SMA-null
cells proliferate and form myofibroblasts,
and then, after wound maturation, many of
these cells disappear by apoptosis.7,10
Although little is
known about the factors that regulate the differentiation and
subsequently the deletion of myofibroblasts, their transient
development from local fibroblastic cells suggests that, at least in
certain cases, cell differentiation in fibroblasts is not controlled by
genetically determined steps but instead may be a short-lived response
to locally available cytokines, such as transforming growth factor
(TGF)-ß, interferon (IFN)-
, platelet-derived growth factor (PDGF),
or heparin.11
Thus, although smooth-muscle cells
constitutively express high levels of SMA,12
fibroblasts
can be induced to express SMA by TGF-ß in vitro when the
cells are grown on rigid matrices.13,14
Further, the level
of SMA expression in cultured fibroblasts is modulated by their
proliferative status,14
substrate
flexibility,15
and cell density.16
Although a
great deal is known about the synthesis and stability of SMA in
quiescent and proliferating smooth-muscle cells,12,17
in
which SMA is a highly abundant and constitutively synthesized protein,
there is much less information on SMA metabolism in myofibroblasts,
which may be of central importance in phenotypic stability.
Little is known about the mechanisms that are involved in the formation and deletion of SMA-containing fibroblasts. In one possible regulatory system, the apparent lifetime of myofibroblasts may be regulated by the turnover rates of SMA mRNA and protein. For example, there is an apparently long-lived pool of SMA protein in gingival fibroblasts when grown on rigid but not compliant substrates.15 It is noteworthy that, on rigid substrates, incorporation of SMA into stress fibers is maximal, and the cells exhibit apparent high levels of intracellular tension. When cultured on certain types of matrix proteins, SMA expression may modulate cell shape15 and the migratory capacity of cells,18 suggesting that the SMA content of cells is influenced by intracellular tension and the substrate to which the cells are attached. In addition, the shape of cells and their relative degree of intracellular tension and spreading appear to be critical factors in determining whether cells in vitro live or die.19 In view of these reports, we have examined the importance of substrate-dependent SMA incorporation into stress fibers and the relative apoptotic susceptibility of cells on different substrates. We have also characterized the stability of SMA protein and mRNA in gingival fibroblasts to obtain insights into the plasticity of the myofibroblastic phenotype after TGF induction.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Primary cultures of human gingival fibroblasts were obtained from
biopsies of normal gingiva in patients aged 10 to 16 years, as
previously described.6
Cells at passages 3 to 12 were used
for all experiments. The cells were grown as monolayer cultures in
-minimal essential medium (MEM) plus 15% v/v fetal bovine serum on
plastic tissue culture dishes, fibronectin-coated plastic (10 µg/ml
of coating solution), or collagen-coated plastic as
described.15
Confluent cultures were serum-starved for 48
hours and treated with TGF-ß1 (R&D Systems, Minneapolis,
MN) or phosphate-buffered saline (PBS) vehicle for 72 hours in
serum-free medium. In preliminary experiments to optimize myofibroblast
formation, dose-response studies using 1, 5, 10, and 20 ng/ml
TGF-ß1 were conducted. Immunoblotting (see below) showed
that the optimal SMA response was 10 ng/ml TGF-ß, and all subsequent
experiments were performed at this dose. Others have used doses of
TGF-ß for SMA gene expression that are equivalent to levels of
TGF-ß1 found in wound fluid.20
After TGF-ß or vehicle treatment, the cells were replated at high
density in serum-free conditions or at low density in the presence of
serum (Figure 1)
. For our experiments,
cells were collected at 0, 3, 7, and 12 days. The initial time point
(ie, 0 day) indicates samples obtained after TGF-ß or vehicle
treatment but before replating, and time points 3, 7, and 12 days
indicate samples obtained after replating.
|
The phenotype of fibroblastic cells was studied after plating in 8-well multichamber slides (7.5 x 104 cells/cm2) coated with either 10 µg/ml fibronectin (Sigma Chemical Co., St. Louis, MO) or 0.5 mg/ml collagen (Vitrogen, Celltrix, Palo Alto, CA) coated as thin films or as a thick gel (~1 mm). These different thicknesses of gels were used because previous data15 and preliminary experiments had shown that the actin organization was influenced by the thickness of the collagen gel; in thick gels, the cells burrowed into the gels, and their morphology was not as well spread as in thin gels. Improved imaging of the cells burrowing into collagen gels was obtained by confocal microscopy (see below). Cells were washed with PBS, fixed with 1.5% paraformaldehyde, permeabilized with 0.3% v/v Triton-X 100 in PBS, and stained for SMA with a mouse anti-human SMA monoclonal antibody (1:50 dilution, clone 1A4; Sigma) for 1 hour at 37°C followed by either a fluorescein isothiocyanate (FITC)-conjugated or rhodamine-conjugated goat anti-mouse antibody (1:100, Sigma) for 1 hour at 37°C. Preimmune mouse and goat sera were used to block nonspecific staining before antibody incubation. For study of focal adhesion formation in cells on different substrates, cells were stained for talin (1:30 dilution, mouse monoclonal, clone 8d4; Sigma). Cells were also stained for desmin (1:50 dilution, clone DE-U-10; Sigma) and for vimentin (1:50 dilution, clone vim-13.2; Sigma). The cells were examined in a Leitz Orthoplan microscope, equipped with a specific filter for fluorescein or rhodamine, and were photographed. In some experiments quantitative immunofluorescence for SMA or for rhodamine phalloidin staining of stress fibers was performed with a confocal microscope (LCSM; Leica) on single cells. In this assay, single cells were imaged, and the fluorescence due to phalloidin binding of stress fibers within an area of known dimensions was measured. Data were normalized to the area and expressed as fluorescence per cell.
Western Blotting
We quantified TGF-ß1-induced modulation of SMA content by immunoblot analysis. Cell extracts were prepared in sodium dodecyl sulfate (SDS) sample buffer and boiled for 3 minutes at 95°C. Equal amounts of protein as determined by the BioRad assay were resolved by electrophoresis on 10% SDS gels, transferred to nitrocellulose filters, probed with the mouse monoclonal antibody for SMA followed by an HRP-conjugated second antibody, and developed with ECL reagents (Amersham, Oakville, Ontario, Canada). Subsequently, X-OMAT Kodak films were exposed to the blots, and the density of the bands was estimated by Scan Analysis (Biosoft, Cambridge, U.K.). Blots were stripped and probed with a monoclonal antibody for ß-actin (clone AC-15; Sigma) for comparison. The ratio of the density of SMA to ß-actin was calculated for each sample. In some experiments we prepared cytoskeletons that were enriched in stress fibers. Briefly, cells were permeabilized in 500 µl of buffer (1% Triton-X 100, 50 mmol/L KCl, 2 mmol/L MgCl2, 0.2 mmol/L adenosine 5'-triphosphate, 10 mmol/L Tris, pH 7.4, 2 mmol/L ethylene glycol bis(ß-aminoethyl ether)-N,N,N',N'-tetraacetic acid [ethylenebis(oxyethyleneitrilo)]tetraacetic acid, 0.5 mmol/L dithiothreitol) containing 1 µmol/L phalloidin (Sigma) to stabilize actin filaments. To obtain cytoskeletal preparations containing stress fibers, cell lysates were centrifuged at 100,000 x g for 1 hour. The pellet consisted of actin filaments and was dissolved in 50 µl of SDS sample buffer and analyzed by immunoblotting.
Flow Cytometry and Cell Counting
Single-cell suspensions were prepared (0.01% trypsin), fixed with 3.7% formaldehyde, permeabilized in 0.02% Triton with PBS, and stained for SMA as described.6 First (anti-SMA) and second (FITC-conjugated goat anti-mouse) antibody dilutions were 1:25 and 1:50, respectively. Cells were washed and resuspended in Mg2+ and Ca2+-free PBS. Samples were analyzed on a FACSTAR Plus flow cytometer (Becton-Dickinson, Mississauga, Ontario, Canada) with 488-nm excitation and 530/30-nm band pass filter for FITC. For all flow cytometry analyses, at least 1 x 104 cells were assessed in each sample, and only cells with forward and orthogonal light scatter characteristics similar to whole, intact fibroblasts were included in the analysis by electronic gates previously established for fibroblasts. For radioautography experiments of [3H]thymidine-labeled cells, flow cytometry was used to sort cells by fluorescence intensity due to SMA staining, and cytospin preparations were made. To estimate the fraction of S-phase cells in the whole population, cell suspensions were stained with 4',6-diamidino-2-phenylindole (1 µg/ml in 0.1% Nonidet P-40). LYSIS software (Becton Dickinson) was used for estimation of the S-phase population. For estimation of cell counts in cultures, 10-µl aliquots of cell suspensions were counted electronically (Coulter, Hialeah, FL).
[3H]Thymidine Labeling
For estimation of the fraction of S-phase cells, cultures were labeled with [3H]thymidine (20 Ci/mmol, 1 µCi/ml; Amersham) for 3 hours before fixation, and radioautographs of flow-sorted cells in cytospin preparations were prepared using liquid emulsion (Kodak, Rochester, NY). The percentage of labeled cells was estimated after light-microscopic examination.
Apoptosis Assay
To measure the proportion of apoptotic cells, breaks in DNA strands were detected by the terminal deoxynucleotidyltransferase-mediated UTP end-labeling (TUNEL) reaction. Cells were treated with vehicle or TGF-ß (72 hours, 10 ng/ml). Cells were replated and grown in serum-free or serum-containing conditions. Samples were collected at various time points after TGF-ß treatment, and cytospins were prepared. These samples were assayed for apoptotic cells with a TUNEL detection kit (Boehringer Mannheim, Montreal, Quebec, Canada). In this kit, terminal deoxynucleotidyl transferase catalyzes polymerization of biotinylated nucleotides to free 3'-OH DNA ends in a template-independent manner. Streptavidin-FITC was used for detection of biotinylated nucleotides, and the percentage of labeled cells was quantified by fluorescence microscopy.
Northern Analysis
Northern blotting was performed on RNA isolated from monolayer cultures grown on plastic tissue culture dishes after 3 days of treatment with TGF-ß1 or vehicle, followed by replating and incubation with or without serum as described above. RNA was isolated by the guanidium hydrochloride method. Total RNA was separated in denaturing 1.3% formaldehyde-agarose gels, transferred to a nylon membrane (Bio-Rad), and cross-linked by UV light. The McMolly Tetra program (SoftGene) was used to design 32-mer oligonucleotides (5'-TCCACAGGACATTCACAGTTGTGTGCTAGAGA-3' and 5'-CCATGCCAATCTCATCTTGTTTTCTGCGCAAG-3') complementary to specific sequences of the SMA and ß-actin mRNA 3'-untranslated regions, respectively. The oligonucleotides were labeled with [32P]ATP (Dupont, NEN, Markham, Ontario, Canada), using 3'-end labeling. Hybridization conditions, film exposure, and reprobing for glyceraldehyde phosphate dehydrogenase were performed as described previously.15
mRNA Stability
Northern analyses of SMA mRNA were conducted on cells that had been previously incubated with actinomycin D at 1 µg/ml for 24 hours. The concentration and duration of actinomycin D treatment were optimized in preliminary experiments in which it was found that 1 µg/ml actinomycin D inhibited 98% of [3H]uridine incorporation into RNA without affecting cell attachment and cell morphology over a 24-hour interval.
[35S]Methionine Labeling and Immunoprecipitation
Cells were treated with TGF-ß1 or vehicle, replated
in the presence or absence of serum, and harvested on days 3, 7, and 12
as described above (Figure 1)
. Before harvest, the cells were
metabolically labeled for 4 hours with [35S]methionine
(100 µCi/ml, ICN Biochemicals) in methionine-free MEM. Cells
were solubilized in 200 µl of lysis buffer (10 mmol/L Tris-HCl, pH
7.0, 50 mmol/L NaCl, 0.5% Triton X-100, 30 mmol/L
Na4P2O7, 50 mmol/L NaF, 100
µmol/L Na3VO4, 0.1% bovine serum albumin, 4
mmol/L MgCl2, 2 mmol/L phenylmethyl sulfonyl fluoride, 1%
aprotonin, 1 mmol/L benzamidine). Insoluble material was removed by
centrifugation at 10,000 x g for 5 minutes at 4°C.
The radioactivity in cell lysates was counted, and equal amounts of
radioactivity were used in immunoprecipitation assays. Supernatants
were immunoprecipitated with SMA antibody (4 µg/ml) overnight at
4°C. Immunocomplexes were recovered by binding to protein A-sepharose
(Zymed) and were washed four times with 25 mmol/L Tris-buffered saline
(pH 7.4), containing 0.5% Triton X-100 and 1 mg/ml bovine serum
albumin, and twice with 0.5 mmol/L NaCl and 25 mmol/L Tris-HCl (pH
7.4). The immunocomplexes were analyzed by electrophoresis on 10%
polyacrylamide gels followed by fluorography and were scanned to
quantify the density of the band.
Statistical Analysis
When appropriate, quantifications of blot or fluorograph densities and cell counts were collated, and the mean ± SEM was calculated. When direct comparisons between groups were made, an unpaired t-test was conducted, and statistical significance was computed. For all experiments, n = 3 per sample group or greater, and statistical significance was accepted at P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In culture, gingival fibroblasts constitutively express low but
detectable levels of SMA and are negative for keratin and factor
VIII,6
indicating that they are indeed fibroblastic and are
not epithelial or endothelial cells. Because proliferation of cultured
vascular smooth-muscle cells causes loss of SMA expression and
confluency augments SMA levels,21
we immunostained for
desmin to confirm that the gingival fibroblast cultures were not
contaminated by smooth-muscle cells. We found that these cells stained
positively for vimentin but were negative for desmin (Figure 2)
, which corresponds to the
V+A+D- fibroblasts described by
Skalli et al.22
|
. TGF-ß-treated cells from all
explants exhibited an upward, unimodal shift of staining for SMA,
indicating that all cells were responsive to TGF-ß and that the
SMA-positive populations, even after 7 days without TGF-ß, were
relatively uniform in their SMA content as indicated by the relatively
narrow distributions of SMA staining (Figure 3)
. Western blotting for
SMA and ß-actin showed increases of SMA compared with ß-actin and
total protein, which corresponded to the increases observed by flow
cytometric analyses (not shown). Because endogenous TGF-ß production
by cells from the different explant cultures could account for some of
the observed increase of SMA, we measured TGF-ß levels by an
enzyme-linked immunosorbent assay (ELISA) (R&D Systems). In
serum-containing and serum-free cultures, the ELISA showed that
endogenous levels of active TGF-ß in both types of culture were
always less than 5 pg/ml, which is 1 x 10-3 of the
dosage required to obtain induction of SMA by TGF-ß for these cells.
|
. Consequently we
next examined whether SMA expression depended on the density at which
the cells were plated (Figure 4B)
. Cells were treated with TGF-ß or
vehicle for 3 days in serum-free medium at a range of cell densities
from low (1.28 x 101
cells/cm2) to
moderate (1.28 x 103
cells/cm2) to high
(1.28 x 104
cells/cm2) and were analyzed
at 5 days after replating. When normalized for ß-actin content, the
SMA levels of TGF-ß-treated samples were severalfold higher for cells
plated at the higher densities (2.8-fold higher for cells plated at
1.28 x 104
cells/cm2, compared with
1.28 x 103
cells/cm2
and 40-fold higher
for cells plated at 1.28 x 104
cells/cm2,
compared with cells plated at 1.28 x 101
cells/cm2; Figure 4B
). Consequently, to optimize the
inductive effect of TGF-ß on SMA expression but also to allow cell
proliferation, cells were replated at 1.28 x 101
cells/cm2
for experiments performed in the presence of
serum.
|
Behavior of Replated Cells in Serum-Containing and Serum-Free Conditions
In rat dermal wounds, the relative abundance of myofibroblastic
cells expressing SMA varies at different stages of wound
healing;7
up to 70% of fibroblastic cells express SMA at
day 6, but they regress thereafter so that by day 30 myofibroblasts are
no longer detectable. To better understand the mechanisms involved in
the formation and deletion of these cells, we studied the fate of SMA
cells in expanding or static cell culture populations. We measured
changes in SMA content under serum-containing and serum-free conditions
for 12 days and assessed cell loss and cell growth in both conditions.
TGF-ß-induced and vehicle-treated controls were replated and grown in
the presence or absence of 10% serum to provide proliferating and
quiescent conditions that model wound healing or resting connective
tissues, respectively. In serum-free conditions when cells were plated
at high densities (1.28 x 104
cells/cm2),
the difference in cell numbers between day 0 (replated cells) and day
12 showed a 64 ± 4% and 57 ± 5% decrease in
TGF-ß-treated and vehicle-treated controls, respectively, which was
not significantly different between treatments (Figure 5
; P > 0.2). In
contrast, when cells were plated at low densities (1.28 x
101
cells/cm2) in serum-containing conditions,
cell numbers were increased 69-fold and 78-fold compared with initial
cell input in TGF-ß-treated and vehicle-treated samples, respectively
(P < 0.05). When cells were plated at higher
densities in the presence of serum, cell growth was also robust
although the increases were not as marked, because, at higher
densities, the cells became confluent after 12 days of growth.
There were no detectable S-phase cells in the serum-free cultures
over the 12-day time course. For serum-containing cultures plated at
low densities with or without previous TGF-ß treatment, the
percentage of S-phase cells was 19% at 3 days and 21% at 7 days. At
most, the difference in the percentage of S-phase cells between TGF-ß
pretreatment or vehicle was ~1% (P > 0.2).
By day 12, 5.6% of cells were in S-phase in vehicle-treated controls,
and 4.4% of cells were in S-phase after TGF-ß treatment. These data
indicated that, at the plating densities used here and in the presence
of serum, the cells proliferated, whereas in serum-depletion
conditions, proliferation was nil. Further, the effect of TGF-ß
pretreatment on cell proliferation appeared to be only small, a finding
that is consistent with earlier data showing cell type-dependent
variation of TGF-ß-induced proliferation.23
|
To better understand the mechanisms regulating the variability of
SMA expression in wound healing and in resting connective tissues,
cellular protein and mRNA were examined over the same 12-day time
course in serum-containing and serum-free conditions with vehicle or
TGF-ß-pretreated cells plated at 1.28 x 101
cells/cm2
for serum-containing conditions and 1.28 x
104
cells/cm2
for serum-free conditions.
Western analysis showed that, in serum-containing conditions, the
relative SMA protein content decreased over time, particularly between
days 0 and 3 of the time course; by day 12, SMA content had returned to
baseline levels. In serum-free conditions the SMA content was much
longer-lived and was reduced only 20% by day 12
(P < 0.01; Figure 6A
). These results were also confirmed by
flow cytometric analysis at 0 and 12 days in serum-containing cultures,
which showed a loss of SMA staining intensity in cells over time
(vehicle-treated cells at 0 and 12 days, respectively, 81 ± 1.9
and 73 ± 1.5 fluorescence channel; TGF-treated cells,
283 ± 4.2 and 89 ± 1.7 at days 0 and 12, respectively). It
is noteworthy that comparisons of SMA content by flow cytometry showed
no large reductions of SMA content between days 0 and 1. The
observed changes in the SMA content could be due to differences in the
rates of SMA synthesis as shown earlier for smooth-muscle
cells.21
We measured the incorporation of
[35S]methionine into nascent SMA by immunoprecipitation
with SMA antibody. The decrease of SMA content in serum-containing
conditions was parallel with the corresponding reduction in the
fractional synthesis of nascent SMA. Similarly, in serum-free
conditions, the SMA synthesis corresponded closely with the SMA content
(Figure 6B)
.
|
. These results suggested that
cessation of proliferation and not the absence of serum factors was
responsible for the maintenance of SMA content in high-SMA-containing
cells.
|
, the ratio of SMA mRNA to ß-actin
mRNA content showed a >twofold decrease between 0 and 3 days in cells
that were pretreated with TGF-ß and then replated in serum-containing
medium (P < 0.01). By day 12, there was no
detectable difference between controls and TGF-ß-treated samples
(P > 0.2). In serum-free conditions mRNA levels
were stable up to 7 days and showed a marked decrease of SMA mRNA only
by day 12.
|
. Untreated controls and
serum-free cultures showed no significant changes in mRNA turnover over
the actinomycin D time course. We examined whether the more rapid
turnover of SMA mRNA could be explained by a dilutional effect because
of more rapid proliferation of SMA-negative cells. Flow cytometry
showed that, over the experimental time course (0–1 days), there was
no significant downward shift of SMA content in the whole-cell
population; as a result, a large proportion of SMA-negative cells did
not appear in this time period. Further, we studied cell proliferation
in the SMA-positive and SMA-negative cell populations.
[3H]Thymidine labeled SMA-immunostained cells were sorted
by flow cytometry into high- and low-SMA-expressing cells based on
thresholds set for above background staining. Autoradiography of
cytospins showed no significant differences between the percents
labeled cells in either the high- or low-SMA-expressing cells and
between controls and TGF-ß-treated cultures under proliferating
conditions. For example on day 1, the labeling index (mean ± SEM)
for high- and low-SMA-expressing cells differed by only 1.7% (high,
16.9 ± 1.3%; low, 15.2 ± 0.5%; P > 0.2),
and by day 5 the difference was only 0.5% (P >
0.2). Thus the observed reduction of SMA mRNA stability was not due to
selective and more rapid overgrowth of SMA-negative cells during the
time period studied here. Effect of Substrate on Actin Organization
We examined whether extracellular matrix proteins such as collagen
and fibronectin may influence the stability of the SMA protein, perhaps
by regulating cell adhesion and consequently the supramolecular
organization of SMA into the actin cytoskeleton. We prepared
cytoskeletal proteins that were enriched with stress fibers, using a
Triton X-100 buffer and phalloidin stabilization to preserve
actin filaments and stress fibers. Immunoblotting of these preparations
showed that TGF-ß-pretreated cells on fibronectin-coated plates after
12 days exhibited ratios of SMA to ß-actin that were similar to those
of cells on tissue culture plastic (Figure 9)
. Cells on collagen-coated plates
showed ~50% reduction of the SMA-to-ß-actin ratio after 12 days
(P < 0.05). For control cells on fibronectin or
collagen that were not treated with TGF-ß, after 3 days of culture
there was a twofold increase in the ratio of SMA to ß-actin,
reflecting an increase of SMA incorporation into stress fibers.
In situ analysis of SMA and stress fiber organization by
quantitative confocal microscopy showed that TGF-ß treatment
increased SMA levels for cells on all types of substrates (Table 1)
. As shown by quantitative confocal
microscopy of rhodamine phalloidin-stained preparations,
TGF-ß-treated cells on fibronectin-coated substrates exhibited stable
levels of rhodamine phalloidin fluorescence throughout the culture
period of 12 days, whereas vehicle-treated cells and all cells on
collagen-coated substrates showed marked loss of staining over the time
course.
|
|
Apoptotic cells appear more frequently during the later stages of
wound healing,7,10
suggesting that apoptosis is an
important mechanism for the deletion of myofibroblasts. To monitor the
proportion of apoptotic cells in serum-containing or serum-free
conditions, TGF-ß-treated and untreated samples were immunostained
for SMA and sorted by flow cytometry into high- and low-SMA cells at 3,
7, and 12 days after replating on plastic. High- and low-SMA-expressing
populations were stained for TUNEL to determine the percentage of
apoptotic cells (Figure 10)
. The
sampling period of day 12 was chosen as the last sampling period
because, from preliminary experiments, it was coincident with a large
reduction in cell numbers that occurred in serum-free cultures after
this time (not shown). Consistent with this result, at day 12 in
serum-free cultures, there were, respectively, 67% and 66% apoptotic
cells in the high- and low-SMA populations treated with vehicle
(P > 0.4), whereas in TGF-ß-treated samples
there were 4% and 60% apoptotic cells in high- and low-SMA
populations, respectively (P < 0.01). In
cultures in which serum was added after TGF-ß pretreatment, there
were 10% and 1% TUNEL +ve cells for the high- and low-SMA
cells (P < 0.01), and in vehicle-treated
controls there were 4% and 13% for high- and low-SMA cells
(P < 0.05). These data show that, in
proliferating cultures, TGF-ß pretreatment may have contributed to
the selective deletion of the SMA-expressing cells. In contrast,
TGF-ß pretreatment in static cultures may protect high-SMA-expressing
cells from apoptosis.
|
. First, cells were
stained with TUNEL, talin, or rhodamine phalloidin to examine the
relationship between apoptosis, focal adhesions, and stress fiber
formation. Second, cells were stained for SMA and TUNEL to determine
whether SMA expression and the supramolecular organization of actin
protect cells from apoptosis. Cells were dichotomized into groups of
either poorly developed or well-developed stress fibers and examined
in situ (Figure 11)
. Cells
plated on thick collagen gels,15
collagen-coated plates, or
fibronectin-coated plates showed significant protection from apoptosis
if the cells were preincubated with TGF-ß. For all substrates, cells
with well-developed stress fibers were more resistant to apoptosis at
12 days. It is noteworthy that cells on fibronectin more frequently
showed prominent stress fibers and abundant focal adhesions, whereas
cells on collagen were more likely to show poor stress fiber
development and fewer focal adhesions. Consistent with the data of
cells plated on tissue culture plastic, TGF-ß reduced the percentage
of apoptotic cells on fibronectin or collagen by up to tenfold. Cells
that were not preincubated with TGF-ß showed very high percentages of
apoptosis when there were poorly developed stress fibers or when the
staining levels of SMA were low (Table 2)
. Indeed for cells on
fibronectin-coated substrates, there was a 35-fold-higher rate of
apoptosis in cells with poorly developed stress fibers than cells with
well-developed stress fibers.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Model Systems
The mechanisms that are involved in the expansion of specific fibroblast subpopulations during early wound-healing conditions are poorly understood. In these situations high levels of serum-derived proteins, cytokines, and provisional matrix proteins are present that profoundly affect fibroblast function.24 TGF-ß peaks at day 7 in rat dermal wounds and falls off dramatically thereafter.25 Our results are consistent with earlier findings showing that TGF-ß treatment at 10 ng/ml for 3 days induces high levels of SMA in fibroblasts14,26 and show that TGF-ß-induced cells replated on solid substrates demonstrate high-SMA expression over prolonged periods of time,15 long after the withdrawal of the inductive stimulus. This system generated high levels of SMA and a high proportion of SMA-expressing cells and is analogous to the types of conditions that would be anticipated in the first 4 to 8 days of wound healing when TGF-ß levels25 are those used in the present in vitro study (ie, ~10 ng/ml).
SMA Metabolism
In proliferating conditions SMA content decreased very rapidly, owing to reduction in the synthesis of nascent SMA as shown by [35S]methionine incorporation into the newly synthesized protein. This reduced SMA protein synthesis was evidently related to the pronounced decrease in SMA mRNA levels of TGF-induced myofibroblasts, which in turn indicated a rapid rate of turnover of the SMA mRNA, a similar finding to that reported earlier for smooth-muscle cells.12,17 The rapid turnover of SMA mRNA but not ß-actin mRNA was confirmed by blocking mRNA transcription with actinomycin D over 1 day of culture. It is noteworthy that the rapid turnover rate of SMA mRNA was not observed in untreated controls. This effect appeared selective for SMA in that there was no evidence of a similar change in ß-actin mRNA stability. Further, the more rapid loss of SMA mRNA was not due to a dilutional effect caused potentially by more rapid growth of SMA-negative cells. We conclude that the precipitous loss of SMA in proliferating cells is due to a rapid expansion in cell number that is simultaneous with an increase of SMA mRNA turnover and a concomitant loss of SMA mRNA.
Apoptosis
TGF-ß-induced myofibroblasts grown in serum-free medium
exhibited a low but steady decrease in cell numbers over time. There
was also a progressive increase in the percentages of cells staining
positive for TUNEL, which, by day 12, presumably resulted in the
observed large-scale loss of cells from the culture. Indeed, by day 12,
65% of the cells grown on plastic were apoptotic, which was
consistent with the decrease in SMA protein content that was seen in
immunoblots. Thus in contrast to the proliferating-cell model, it
appears that the main reasons for the loss of SMA protein content in
the nonproliferating model were related to reduced numbers of
myofibroblasts and increased cell death.
Different substrates conferred variable levels of protection against apoptosis in serum-free cultures. In particular, cells on fibronectin showed much lower incidence of apoptosis compared with cells on collagen. Because cells on fibronectin also exhibited more well-developed stress fiber systems enriched with SMA and abundant focal adhesions, our data indicate a measure of protection against apoptosis that is attributable to the supramolecular organization of actin and to the relative adhesive strength of the cells to the substrate. Previous work on endothelial cells has shown that well-spread, tightly adherent cells are more resistant to apoptosis than poorly adherent cells.19 Because high levels of SMA content may also inhibit motility in fibroblasts,18 we surmise that the organization of SMA into the stress fibers of well-spread, adherent cells confers protection against apoptosis for those cells that are more likely to contract the matrix and to remain in situ compared with cells that migrate. This effect appears to be mediated in part by TGF-ß, because this cytokine not only enhances SMA expression but does so more efficiently in cells that are attached to rigid substrates and with high levels of intracellular tension.15 Indeed if TGF-ß was present during myofibroblast induction, then TUNEL staining was greatly reduced. Because TGF-ß can be present at very high levels in chronic inflammatory lesions,27 we suggest that the persistence of myofibroblasts in nonproliferating conditions may be related to the persistence of high TGF-ß levels and the organization of SMA in the actin cytoskeletons. This finding has important implications for wound healing and scar tissue formation, because it suggests a mechanism by which the very cells that are most likely involved in scar formation (ie, the SMA-rich myofibroblast) are also those that are most protected against apoptosis and are strongly induced by TGF-ß. Perhaps by therapeutically altering the composition of the extracellular matrix or by affecting the physical properties of the matrix,15 significant blockage of this positive feedback loop could inhibit myofibroblast formation, because TGF-ß induction of myofibroblasts is very dependent on the compliance of the matrix. Notably, the apoptosis data that we provide here were determined from TUNEL assays that in addition to estimating apoptotic processes, can also reflect DNA repair28 and RNA synthesis.29 In view of our previous use of DNA-laddering methods,30 electron microscopy,30 and TUNEL31 to measure apoptosis in gingival fibroblasts, we think that nonapoptotic phenomena have probably not contributed significantly to errors in estimation of apoptosis.
Actin Organization
Previous data have indicated that the state of actin assembly may regulate actin expression by a feedback loop;32 thus, cells with incorporation of actin into heavily cross-linked filaments would be expected to show a slower rate of actin turnover. Because this prediction applies to the myofibroblasts described here, we speculate that the myofibroblasts with well-developed stress fibers and stable pools of SMA are more phenotypically stable because the actin filaments are more stable in these cells and less likely to convert to monomer. Exactly how the putative feedback system in cultured fibroblasts can convert the sensing of intracellular tension and actin organization into more stable mRNAs for SMA is unclear, but the data reported here are remarkably consistent with the concept proposed earlier.32 A particularly novel finding in this study is that the supramolecular organization of the actin cytoskeleton appears not only to stabilize SMA mRNA and protein but also to protect myofibroblasts from deletion by apoptosis. Thus the expression of SMA may not only be linked to wound contraction and retardation of motility in fibroblasts18 but may also enhance the lifetime of the cells in which it is most strongly expressed.
|
Footnotes |
|---|
Supported by MRC of Canada Grouop and maintenance grants (to C. A. G. M.) and by the Canadian Heart and Stroke Foundation.
Accepted for publication August 24, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-smooth muscle actin expression by fibroblasts. J Cell Physiol 1994, 159:161-175[Medline]
-Smooth muscle actin is transiently expressed by myofibroblasts during experimental wound healing. Lab Invest 1990, 63:21-29[Medline]
-smooth muscle actin expression in granulation tissue myofibroblasts and in quiescent and growing cultured fibroblasts. J Cell Biol 1993, 122:103-111-smooth muscle actin in fibroblasts. Am J Pathol 1999, 154:871-882-smooth muscle actin: retardation of motility in fibroblasts. J Cell Biol 1996, 134:67-80This article has been cited by other articles:
 |
|
 |
|
  Y. Lenga, A. Koh, A. S. Perera, C. A. McCulloch, J. Sodek, and R. Zohar Osteopontin Expression Is Required for Myofibroblast Differentiation Circ. Res., February 15, 2008; 102(3): 319 - 327. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Meyer-ter-Vehn, S. Sieprath, B. Katzenberger, S. Gebhardt, F. Grehn, and G. Schlunck Contractility as a Prerequisite for TGF-{beta}-Induced Myofibroblast Transdifferentiation in Human Tenon Fibroblasts Invest. Ophthalmol. Vis. Sci., November 1, 2006; 47(11): 4895 - 4904. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Laplante, M.-A. Raymond, A. Labelle, J.-I. Abe, R. V. Iozzo, and M.-J. Hebert Perlecan Proteolysis Induces an {alpha}2beta1 Integrin- and Src Family Kinase-dependent Anti-apoptotic Pathway in Fibroblasts in the Absence of Focal Adhesion Kinase Activation J. Biol. Chem., October 13, 2006; 281(41): 30383 - 30392. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.E. van Beurden, J.W. Von den Hoff, R. Torensma, J.C. Maltha, and A.M. Kuijpers-Jagtman Myofibroblasts in Palatal Wound Healing: Prospects for the Reduction of Wound Contraction after Cleft Palate Repair J. Dent. Res., October 1, 2005; 84(10): 871 - 880. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. D. Lovelock, A. H. Baker, F. Gao, J.-F. Dong, A. L. Bergeron, W. McPheat, N. Sivasubramanian, and D. L. Mann Heterogeneous effects of tissue inhibitors of matrix metalloproteinases on cardiac fibroblasts Am J Physiol Heart Circ Physiol, February 1, 2005; 288(2): H461 - H468. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Vittal, J. C. Horowitz, B. B. Moore, H. Zhang, F. J. Martinez, G. B. Toews, T. J. Standiford, and V. J. Thannickal Modulation of Prosurvival Signaling in Fibroblasts by a Protein Kinase Inhibitor Protects against Fibrotic Tissue Injury Am. J. Pathol., February 1, 2005; 166(2): 367 - 375. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. A. Walker, K. S. Masters, D. N. Shah, K. S. Anseth, and L. A. Leinwand Valvular Myofibroblast Activation by Transforming Growth Factor-{beta}: Implications for Pathological Extracellular Matrix Remodeling in Heart Valve Disease Circ. Res., August 6, 2004; 95(3): 253 - 260. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. A. Rice and L. A. Leinwand Skeletal myosin heavy chain function in cultured lung myofibroblasts J. Cell Biol., October 13, 2003; 163(1): 119 - 129. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Wang, M. Su, J. Fan, A. Seth, and C. A. McCulloch Transcriptional Regulation of a Contractile Gene by Mechanical Forces Applied through Integrins in Osteoblasts J. Biol. Chem., June 14, 2002; 277(25): 22889 - 22895. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Yamate, K. Sato, M. Ide, M. Nakanishi, M. Kuwamura, S. Sakuma, and S. Nakatsuji Participation of Different Macrophage Populations and Myofibroblastic Cells in Chronically Developed Renal Interstitial Fibrosis after Cisplatin-induced Renal Injury in Rats Vet. Pathol., May 1, 2002; 39(3): 322 - 333. [Abstract] [Full Text] |
