| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Institut National de la Santé et de la Recherche Médicale Unité 506, Hospital Paul Brousse, Villejuif, France;
Department of Allergy and Respiratory Diseases, Department of Internal Medicine, University of Genoa, Genoa, Italy;
Oncogenesis, Differentiation, and Signal Transduction Laboratory, Institut de Recherche sur le Cancer-Centre Nationale de la Recherche Scientifique, Villejuif, France; and
Glaxo Wellcome, Medical Department, Verona, Italy
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-induced expression of 
-smooth muscle actin, the main marker
of myofibroblastic differentiation, both in very early and in mild
differentiated myofibroblasts. Finally, FP displays an additional
powerful anti-inflammatory effect, decreasing nuclear translocation
of NF-
B independent of the degree of myofibroblastic
differentiation. These data 1) suggest that myofibroblasts are priority
targets for ICS, which is able to revert them to a normal phenotype
even if they appear to be already engaged in their differentiation, and
2) may help to explain why asthma is improved by an early ICS
treatment, whereas advanced asthma is more resistant to these
drugs. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Fibroblast heterogeneity may also occur even among the same tissue, as
a consequence of an activation/differentiation process. The
differentiated cells, called myofibroblasts, transiently (normal tissue
repair) or permanently (pathological stromal reactions) express
contractile proteins, such as -smooth muscle actin
(-SMA).4 This
cytoskeletal element is the most important marker of myofibroblastic
differentiation because it causes a shift from a
migratory/proliferative phenotype to a contractile phenotype
(1) that has important consequences, for example, in lung
remodeling process. When cultured in vitro, lung fibroblasts should be
considered as permanently activated cells. Indeed, even resting
postconfluent fibroblasts maintain phenotypic characteristics that are
specific in vivo to activated myofibroblasts (1).
Therefore, it is likely that cultured normal fibroblasts represent the
initial step(s) of myofibroblastic differentiation before the induction
of -SMA.
Recently published data suggest that lung fibroblasts constitute a mixture of cellular subsets expressing different functional markers and probably have capacity to exert different immune functions (2, 3, 4). We have recently shown that Th2 cytokines, IL-4 and IL-13, specifically induce a proinflammatory-like signal transduction in cultured human lung myofibroblasts, leading to the expression of adhesion molecules and to the production of factors pivotal in the development of the inflammatory cascade (5, 6, 7).
These data strongly suggest that lung fibroblasts, representing 
30%
of the resident lung cells, when activated play a major role in the
remodeling of the lung microenvironment, acting not only as target
cells, but also as effector proinflammatory cells (8). To
date, airway fibroblasts were not considered as potential targets in
asthma therapy, even if extensive studies have been developed to
identify the direct inhibitory actions that corticosteroids display on
inflammatory and structural cells involved in pulmonary and airway
diseases (9).
Fluticasone propionate (FP) is a highly lipophilic inhaled corticosteroid (ICS) that possesses the highest receptor affinity among corticosteroids, implying fast association and slow dissociation from the receptor protein. This explains the higher FP potency and its faster clinical response onset when compared with other corticosteroids. The receptor affinity of FP and the rank orders of affinities of other corticosteroids are well correlated with the induction of anti-inflammatory responses at the cellular level as well as with the clinical efficacy of these compounds. Thus, because glucocorticosteroid responses are primarily receptor mediated, the magnitude of their clinical effects and efficacy can be explained by the physicochemical properties of the corticosteroids that determine the glucocorticosteroid receptor (GR) interactions.
It was already demonstrated, in asthma and rhinitis studies, that
regular topical FP therapy has a profound effect on an inflammatory
cascade underlying the clinical disease expression.
Glucocorticosteroids, such as FP, act intracellularly by binding to GR
to form GR-glucocorticosteroid complex, which then modifies expression
of certain genes. The GR may regulate gene expression directly by
binding to the DNA or indirectly by interacting with other
transcription factors. This may occur via various mechanisms, of which
the interaction with the transcription factors, AP-1 and NFB, has
been best characterized (10).
In this study, we have analyzed the in vitro effects of FP, an ICS commonly used in asthma therapy, on behavior of human lung fibroblasts at different stages of myofibroblastic differentiation.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Twenty primary cultures from human airways of normal donors
(embryonic fetal and adult) or from patients with Th2 lesions were
established in primary culture by enzymatic digestion of the tissues.
Phenotypic analysis of the fibroblastic cultures identified three types
of cells: 1) cultures expressing less than 5% of -SMA-positive
cells (essentially from normal donors), 2) cultures expressing between
25 and 40% -SMA-positive cells (from pathological samples), and 3)
fully differentiated cultures expressing 100% -SMA-positive cells
(from pathological samples). All these myofibroblastic primary cultures
expressed the fibroblastic marker ASO2 (5, 6). For this study, we have chosen four fibroblastic lung
primary cultures that represent the aforementioned different
groups of myofibroblastic differentiation.
The origin and characteristics of cultured fetal lung fibroblasts
(ICIG7) representing the very early step of myofibroblastic
differentiation (5% -SMA-positive cells) as well as of adult
primary lung fully differentiated myofibroblasts (FPA, 100%
-SMA-positive cells), derived from a stromal Th2 reaction, have been
previously described (11). Primary cultures BRONCO5 and
BRONCO3 were obtained from lung tissue neighboring a neoplastic lesion
and were adapted to in vitro culture after enzymatic digestion. These
cells can be classified as mildly differentiated myofibroblasts (30%
of -SMA-positive cells). Cells were cultured in DMEM medium
(Eurobio, Les Ulis, France) supplemented with 10% FCS (ATGC
Biotechnologie, Noisy-le-Grand, France), 2 mM glutamine, and 1%
antibiotics (Life Technologies, Cergy Pontoise, France), and incubated
at 37°C and 5% CO2 conditions.
Cytokines and Abs
Human rIL-4 and rIL-13 were a generous gift from Schering-Plough
(Dardilly, France) and Sanofi Recherche (Labège Innopole,
France), respectively. The anti-phospho-STAT1 (P-STAT1) and
anti-phospho-STAT6 (P-STAT6) Abs were obtained from New England
Biolabs (Beverly, MA); the anti-STAT3 Ab, the anti-p65 NFB
subunit, and the streptavidin FITC were purchased from Santa Cruz
Biotechnology (Santa Cruz, CA). The anti-STAT1,
anti-phospho-tyrosine kinase 2 (TYK2) (P-TYK2),
anti-TYK2, and antiactive mitogen-activated protein kinase (MAPK)
were obtained from Transduction Laboratories (Lexington, KY) and
Promega (Madison, WI). Antiphosphoserine I-B-specific polyclonal
IgG were obtained from Calbiochem (San Diego, CA). The
peroxidase-conjugated anti-rabbit IgG and the mAbs STAT6 were
obtained from Upstate Biotechnology (Lake Placid, NY). Biotinylated
goat anti-rabbit IgG and peroxidase-conjugated goat anti-mouse
(GAM) IgG were obtained from Immunotech (Marseille,
France).
Immunoblotting and immunoprecipitation
Immunoblotting and immunoprecipitation were performed as described previously (12). Briefly, confluent cell cultures were serum starved overnight before each experiment to reduce the basal phosphorylation. Cells were washed and stimulated by various concentrations of IL-4 or IL-13 at 37°C for 15 min. Sister cultures of each condition were pretreated with 10-6-10-12 M FP at 37°C for 1 h, to investigate the FP effects on the signal transduction induced by the Th2 cytokines. Cells were then resuspended in lysis buffer containing 0.5% or 1% Nonidet P-40. Proteins were incubated with different Abs and immune complexes, then captured with protein G-Sepharose beads (Pharmacia Biotech, Uppsala, Sweden) overnight at 4°C. Complexes were washed, boiled, solubilized with Laemmli buffer, and separated on 7.5% or 12% SDS-polyacrylamide gels before transfer onto polyvinylidene difluoride membranes (DuPont-NEN, Boston, MA). Membranes were probed with Abs, revealed with peroxidase-conjugated Abs, and then visualized by ECL (Amersham, Les Ulis, France).
Detection of p65 NFB nuclear translocation and analysis of
-SMA expression by confocal microscopy
For the detection of p65 NFB nuclear translocation, normal
fibroblasts, and myofibroblasts that had been cultured for 2 days were
preincubated in serum-free medium either containing or lacking
10-8 M FP for 1 h at 37°C, and then
stimulated for a further 15 min with or without TNF- (10 ng/ml).
To determine -SMA expression, 1-day cultured normal fibroblasts and
myofibroblasts were incubated in the absence or presence of
10-8 M FP and/or TGF- (5 ng/ml) for 72 h
at 37°C.
Subsequently, cells from both experiments were washed and permeabilized
with Ortho PermeaFix (Ortho Diagnostic Systems, Raritan, NJ) for 45
min, and indirect immunofluorescence was performed with rabbit Abs
recognizing the p65 subunit of the NFB complex or with mAb 1A4 that
recognizes human -SMA (DAKO, Glostrup, Denmark). Samples were then
incubated with Alexa Fluor488 GARa (p65) or with Alexa Fluor594
GAM (-SMA; Molecular Probes, Leiden, The Netherlands). To
determine the nuclear localization of the p65 subunit of the NFB
complex, nuclei were stained by incubating the cells for 3 min at room
temperature in the presence of propidium iodide (red nuclear staining).
After staining, cells were washed and centrifuged in a Cytospin 3
(Shandon, Pittsburgh, PA) and analyzed by laser scanning confocal
microscopy on a Leica TCS NT/SP Interactive Laser Cytometer equipped
with confocal optics (Leica Microsystems, Wetzlar or RFA).
RT-PCR analysis of -SMA expression
Oligonucleotide primers.
Primers used in PCR and the annealing conditions for -SMA
were upper, 5'-GTC CAC CGC AAA TGC TTC TAA-3', and lower, 5'-AAA ACA
CAT TAA CGA GTC AG-3', amplification product 141 bp (58°C, 30
cycles); and for GAPDH, upper, 5'-GGT GAA GGT CGG AGT CAA CGG A-3', and
lower, 5'-GAG GGA TCT CGC TCC TGG AAG A-3', amplification product 240
bp (60°C, 20 cycles).
mRNA isolation and RT-PCR analysis.
Total RNA was isolated from human lung fibroblasts by using
TRIzol reagent (Life Technologies), according to the manufacturer’s
specifications. cDNA was synthesized from total RNA using oligo(dT)
primer and Moloney murine leukemia virus reverse transcriptase (Life
Technologies), according to the manufacturer’s instructions. In each
PCR, 1 µCi/tube [-32P]dCTP was included to
label the PCR product. Aliquots of the RT-PCR were separated by PAGE.
The gel was dried, and the results were visualized by autoradiography,
as described (7).
For semiquantitative RT-PCR, amplified products of -SMA were
quantified by densitometric scanning of specific bands on negatives of
GAPDH transcripts in a PhosphorImager (Molecular Dynamics, Bondouffle,
France). Water was used as a negative control, and some samples were
amplified without reverse transcription to check for the presence of
DNA contaminants.
Nuclear protein extraction and EMSA
ICIG7 cells treated with TNF- and fluticasone, as described
in Confocal microscopy, were extracted following a
previously described protocol (13) Analysis of NF-B
DNA-binding activities in these extracts was performed by EMSA, as
reported previously (14). The quantification of retarded
oligonucleotides in the band corresponding to NF-B was monitored by
PhosphorImager quantification using a Storm apparatus.
The supershift EMSA was performed by incubating the protein extracts with 1 µl of RelA-, RelB-, c-Rel-, and p50-specific antisera from Santa Cruz Biotechnology for 1 h prior to the addition of the radiolabeled oligonucleotide.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
IL-4 and IL-13 specifically act on cultured human lung myofibroblasts, activating the JAK-STAT signaling with an inflammatory-like pattern (5). We analyzed the ability of FP to interfere with these signal transduction pathways.
We investigated the effects of FP treatment on low passaged (p13) and
older (p24) human lung fibroblasts at a very early stage of
myofibroblastic differentiation (ICIG7) signal transduction, activated
by IL-4 and IL-13 (Fig. 1
). Upper
panels are Western blots illustrating the phosphorylated proteins,
while lower panels show the expression of the native
proteins. Overall, our data show that Th2 cytokines induce the
phosphorylation of TYK2, STAT-1, STAT-3, STAT-6, and MAPK, and that FP
preferentially inhibits the action of IL-13 in young, but not in older
cultures.
|
A). IL-4 and IL-13 also induced the phosphorylation of
STAT1 (Fig. 1B) and STAT6 (Fig. 1C),
respectively. FP inhibited the action of both cytokines on STAT1 and
STAT6. We observed that STAT3 (Fig. 1D) is constitutively
phosphorylated, and that IL-4 and IL-13 increased its extent of
phosphorylation. FP decreased the level of phosphorylation only in the
cells stimulated with IL-13 (Fig. 1D). In these cells, we
also detected the appearance of a band of lower molecular
mass (80 kDa) that corresponds to the STAT3 isoform STAT3,
whose phosphorylation is increased by IL-13 and by the combined
treatment with IL-4 and FP. In contrast, the corticosteroid blocks
IL-13 stimulation, causing the disappearance of this phosphorylated
STAT3 isoform.
We also analyzed the activation of MAPK (extracellular signal-related
kinase (ERK) 1 and ERK2) that have been described as possible targets
for IL-4 and IL-13 (15, 16). In ICIG7 cells, ERK1 and ERK2
are constitutively phosphorylated, and both tested Th2 cytokines
up-regulated their level of phosphorylation. FP decreased both
constitutive and cytokine-induced phosphorylation of ERK1 and ERK2
(Fig. 1E).
When the same signaling pathways were studied in older ICIG7 cultures
(right panels), we found the same pattern of basal or
IL-4/IL-13-induced activation of JAK/STAT or MAPK. The exception were a
less efficient phosphorylation of TYK2 and the lack of expression of
STAT3. The major difference was the loss of the inhibitory effect
displayed by FP, which just inhibited the constitutive phosphorylation
of TYK2 and the Th2-induced phosphorylation of MAPK.
Analysis of lower panels (Fig. 1, A–E) shows
that in each sample the native proteins were homogeneously expressed.
In some instances, reblotting the membrane with anti-tubulin
mAbs confirmed the homogenous loading of the lysates in each
group (data not shown).
In Fig. 2, we analyzed the above signal
transduction pathways in the mildly differentiated human lung
fibroblasts BRONCO5 (right panels) and in fully
differentiated FPA myofibroblasts (left panels) at
the same passage level (p13) tested in ICIG7 cultures. Upper
panels are Western blots illustrating the phosphorylated proteins,
while the lower panels concern the expression of the native
proteins. These cells behaved similarly to older ICIG7 cells: Th2
cytokines activate most of the different pathways, but FP is less
efficient, decreasing only TYK2 phosphorylation.
|
A) and of MAPK (Fig. 2E) as well as the
stimulatory effects of IL-4 and IL-13. The only difference was that the
two Th2 cytokines did not modify the basal level of phosphorylation of
STAT3 in BRONCO5 and FPA cultures (Fig. 2D). Moreover, the
two myofibroblastic cell strains did not express STAT3. Concerning
STAT1 (Fig. 2B) and STAT6 (Fig. 2C), we also
confirmed that in BRONCO5 and FPA myofibroblasts, IL-4 and IL-13 induce
the phosphorylation of these STAT proteins. A major difference in these
myofibroblastic cultures was that FP did not interfere with these
signal transduction pathways, with the single exception of TYK2.
Indeed, FP partially inhibited the IL-13-induced phosphorylation
of TYK2.
Because glucocorticosteroids inhibited the activation of the
transcription factor NF-B (10), in Fig. 2F
we analyzed the extent of phosphorylation of the NF-B inhibitor
I-B, to understand whether FP was active on this step of the
NF-B activation pathway.
In these experiments, we used an anti-I-B mAb that recognizes
both the phosphorylated protein (upper band) and the native
one (lower band). I-B was already
phosphorylated in basal culture conditions, and treatment with Th2
cytokines and/or FP did not modify the intensity of phosphorylation of
I-B (upper band) nor induced the degradation of the
native protein (lower band).
Effects of FP on NFB activation in human lung fibroblasts and
myofibroblasts
Confocal microscopy.
Subsequently, we tried to confirm the constitutive activation of the
transcription factor NFB in early (ICIG7), mild (BRONCO5), and fully
differentiated (FPA) human lung myofibroblasts. By confocal microscopy,
we analyzed the localization of p65 subunit (RelA) of the NFB
complex (green staining) inside the nuclei (red staining).
Fig. 3 represents different overlay
pictures; cells in each sample displayed a cytoplasmic green staining
that demonstrated the presence of the native NFB p65 subunit. In
contrast, the nuclear yellow staining identifies the cells expressing
RelA at nuclear level. The three different lung myofibroblastic
cultures (ICIG7, BRONCO5, and FPA) were analyzed 48 h after
seeding during the proliferative phase. The percentage of cells
presenting a nuclear localization for NFB p65 subunit was 10% of
ICIG7 cultures (Fig. 3A1), but it increased up to 30–40%
in BRONCO5 and FPA myofibroblastic cultures (Fig. 3, B1 and
C1).
|
strongly increased the percentage of cells
expressing an activated NFB in the three myofibroblastic lung
cultures (Fig. 3, A2, B2, and C2), and
FP efficiently counteracted this induction, blocking almost totally
NFB activation in all samples (Fig. 3, A3, B3,
and C3). Fig. 3D represents the control samples
treated with a nonrelated isotype-matched mAb and propidium iodide. In
these cultures, nuclei were stained in red, whereas no background green
staining was detected.
Gel shift assay.
To confirm the inhibitory effect of fluticasone on basal and
TNF--induced NF-B, we performed gel retardation analysis of
nuclear proteins extracted from myofibroblastic cultures representing
an early (Fig. 4A) and an
intermediate stage (Fig. 4B) of differentiated cells, using
as a probe an oligonucleotide encoding the B sequence. As a control,
we used an extract from a T cell line that had been activated for
3 h with PMA plus ionomycin. The results show that both cell types
contained a constitutive B-binding activity, which comigrated with
the T cell NF-B dimer that has been previously identified as
RelA-p50 (17). TNF- treatment of cells increased the
binding of RelA-p50, whereas pretreatment with FP decreased it both in
control and TNF--treated cells. The efficiency of the inhibition was
quantified by PhosphorImager, and demonstrated that addition of FP
decreased the basal NF-B DNA binding by 15% in both cell strains
and the TNF--induced NF-B DNA binding by 30% and 70% in
ICIG7 and BRONCO cultures, respectively. These findings
confirmed that FP efficiently inhibits TNF--induced activation of
the prototypical RelA-p50 NF-B dimer both in early and mildly
differentiated lung myofibroblastic cells with a more powerful effect
on the latter cells. The fact that the basal NF-B activity was less
reduced by FP might be due to a distinct mechanism by which NF-B
activity is generated in the absence of TNF-. Gel retardation assays
could not be performed on fully differentiated myofibroblasts owing to
the too limited proliferative potential of these terminally
differentiated cells.
|
B proteins (Fig. 4C). Indeed, a supershift of the dimer was observed using an
Ab directed against RelA, and the use of an Ab against the p50 subunit
inhibited the binding.
Effects of FP on -SMA expression in human lung
fibroblasts and myofibroblasts
Confocal microscopy.
In these experiments, we explored the capacity of FP to interfere with
the myofibroblastic differentiation process, analyzing the effects of
the drug on the constitutive or TGF--induced expression of -SMA
(cytoplasmic red staining). Indeed, recent data suggest that expression
of -SMA decreases in glomeruli from patients with IgA
nephropathy treated with corticosteroids (18).
Constitutive expression of -SMA was low (5% positive cells) in
ICIG7 human lung myofibroblasts at an early stage of differentiation
(Fig. 5A1), intermediate (up
to 25–30% positive cells) in BRONCO5 mildly differentiated lung
fibroblasts (Fig. 5B1), and complete (100% positive cells)
in FPA fully differentiated lung myofibroblastic cultures (Fig. 5C1).
|
-SMA expression in ICIG7 cells (Fig. 5A2), whereas the inhibition was only partial in BRONCO
cells (Fig. 5B2) and absent in FPA cells (Fig. 5C2).
Treatment with TGF- strongly increased the percentage of
-SMA-positive cells, both in ICIG7 (Fig. 5A3) and BRONCO5
cultures (Fig. 5B3), but not in FPA cultures (Fig. 5C3). Finally, FP completely inhibited -SMA expression in
both early (Fig. 5A4) and mildly differentiated
myofibroblastic cultures (Fig. 5B4), treated with TGF-,
but was ineffective on totally differentiated lung myofibroblasts (Fig. 5C4).
RT-PCR analysis.
To confirm the results from confocal microscopy on -SMA modulation
by FP and TGF-, RT-PCR analysis was conducted using sets of primers
described in Materials and Methods. Transfer to an adhesion
surface displaying an increased mechanical tension may induce a rapid
and stable myofibroblastic differentiation of normal fibroblasts
(19). We adapted this technique to ICIG7 cells generating,
within 4 days, fully differentiated ICIG7 myofibroblasts. Thus, in
RT-PCR assays, we used ICIG7 cultures before and after their
differentiation and in the mildly differentiated lung myofibroblastic
cultures BRONCO5 and BRONCO3.
As shown in Fig. 6, the specific
transcript (141 bp) for -SMA, barely detectable in control ICIG7
cultures (Fig. 6A), was constitutively expressed in fully
differentiated ICIG7 myofibroblasts (Fig. 6B) and in the
mildly differentiated BRONCO5 (Fig. 6C) or BRONCO3 (Fig. 6D) cultures. TGF- increased -SMA expression in all
the samples. Treatment with FP efficiently inhibited the constitutive
-SMA expression in all the cultures, with the exception of BRONCO3
cells, whereas TGF--induced expression of -SMA was almost totally
inhibited in all the samples. The transcript (240 bp) for the
housekeeping gene GAPDH (used as internal standard) was expressed with
the same intensity in all the samples.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-induced expression of the marker of myofibroblastic
differentiation -SMA. Moreover, FP efficiently inhibits basal and
TNF--induced nuclear translocation of the p65 subunit of the
transcription factor NFB independently on the level of
myofibroblastic differentiation. The latter result could be very
important, because NFB plays a major role, triggering several genes
involved in the activation of the inflammatory process
(10).
The inhibitory action of FP on the JAK/STAT pathway induced by IL-4 and
IL-13 in lung fibroblasts is a new acquisition, even if recent studies
report the existence of a cross-talk between the JAK-STAT and the
glucocorticoid-nuclear receptor pathways. However, it must be stated
that these studies report a synergistic effect among GR, STAT3, and
STAT5, respectively (20, 21). Another recent study reports
that overnight treatment with dexamethasone inhibits IL-2 activation of
STAT5 in primary T cells. This inhibition is associated with a
decreased expression of the IL-2R
chain that also affects IL-4,
IL-7, and IL-15 signaling (22).
On the contrary, our results show that a short exposure to FP inhibits
the phosphorylation of STAT1, STAT3, STAT3, and STAT6 induced by Th2
cytokines, with no effects on the IL-4/IL-13R expression. Moreover,
lung fibroblasts at a very early stage of differentiation do not
express the -chain, and IL-4/IL-13 signaling is mediated by a type
II receptor (IL-4R/IL-13R1), whereas -SMA-positive lung
myofibroblasts constitutively express the c chain
(5, 6, 7), but once more its expression is not decreased
by FP.
The inhibition seems to be more specific on the action of IL-13, and is
also extended to the phosphorylation of TYK2 (the fourth member of the
JAK family) and of the MAPK members, ERK1 and ERK2. These inhibitory
effects seem to be more effective when myofibroblasts are treated early
both in their life span and differentiation pathway; however, the loss
of FP action on the JAK-STAT signaling in older lung myofibroblasts is
not due to the acquisition of a generalized resistance to
corticosteroids, because in these cells this drug is still able to
inhibit NFB. The therapeutic potential of corticosteroids may be
very important, acting at different levels and cutting at a very early
stage a cascade of inflammatory signals. In contrast, the residual
effect observed in older early and in mild differentiated
myofibroblasts (BRONCO5) on the phosphorylation of TYK2 may still have
discrete anti-inflammatory effects (23, 24).
Th2 cytokines activate in vivo and in vitro lung fibroblasts,
inducing matrix deposition, production of proinflammatory
mediators, and expression of -SMA (4, 5, 6, 7, 8, 25).
Similar results on matrix deposition have also been reported in skin
fibroblasts (26), even if the signal transduction
activated by IL-4 and IL-13 in these cells (27) is much
more restricted than that activated in lung fibroblasts
(5). Thus, it is likely that activation of JAK2/TYK2/STAT6
pathways may be sufficient to induce matrix deposition, and the broader
spectrum of signaling activated by Th2 cytokines in lung fibroblasts
could account for a larger panel of activation of inflammatory
mediators.
For instance, the different STAT members and MAPK control the
activation of several genes involved in apoptosis, cell proliferation,
and differentiation, and therefore in the inflammatory process. This
happens by direct binding to specific sequences present in the promoter
region of this gene or by formation of functional complex with other
transcription factors, amplifying their competence for gene activation.
The fact that FP inhibits the first step of their activation
(phosphorylation) process potentially guarantees a very large spectrum
on gene suppression. Our data also suggest that interplay between Th2
cytokines and FP on STATs activation may be much more complex. Indeed,
ICIG7 young cultures constitutively express STAT3, a splice variant
of STAT3 that behaves as a dominant-negative regulator of
transcription. This is a very intriguing finding, because this
competitor of STAT3 could theoretically interfere on the synergism
between STAT3 and GR (28). Moreover, its phosphorylation
is up-regulated by IL-13 and by the combined treatment with IL-4 and
FP, whereas the corticosteroid totally inhibits STAT3
phosphorylation induced by IL-13. This different behavior is most
likely due to the fact that IL-4 and IL-13, although sharing common
receptors, may induce different signaling in the same type of cells
(29). The fact that the two Th2 cytokines display an
opposite control on the phosphorylation of STAT3 suggests potential
differences in the pattern of gene activation controlled by STAT3 in
the presence of FP.
The transcription factor NFB is in the promoter region of several
genes such as adhesion molecules, inflammatory cytokines, chemokine
inflammatory enzymes, and receptors (30); thus, it exerts
a dominant role in the induction and stabilization of the inflammatory
process.
Therefore, inhibition of its activation by FP both in normal and
irreversibly differentiated lung myofibroblasts could constitute one of
the main anti-inflammatory effects in advanced asthma, considering
that it has been recently reported that in patients with mild asthma,
FP does not inhibit NFB-DNA binding in bronchoalveolar lavage
macrophages, and in airway epithelium of bronchial biopsies
(26). Our results also confirm the preferential
targeting of NFB by FP and the lack of I-B involvement
(10). The inhibition of -SMA expression by FP in early
and mild differentiated lung myofibroblasts is also a quite original
observation that may have very important consequences in the early
intervention of asthma treatment. Indeed, the expression of this
contractile protein causes on fibroblasts a shift from the migratory to
the contractile phenotype, with important consequences on the lung
function (1). A recent study showed that in patients with
IgA nephropathy, in vivo corticosteroid treatment reduces the number of
-SMA-positive cells (18). However, the authors of this
study (18) cannot determine whether this is a direct inhibitory
effect of the corticosteroid, and recognize that such a direct effect
has never been reported. Thus, our data represent the first direct
evidence of the capacity for a corticosteroid to inhibit -SMA
constitutive and TGF--induced expression. Interestingly, the -SMA
expression is totally suppressed when FP is associated with TGF-,
even if this cytokine is the most powerful inducer of this contractile
protein (31).
This approach allows a better definition of the early stages of
myofibroblastic differentiation. Indeed, we distinguish, within BRONCO5
cultures, at least three different myofibroblastic subsets. The first
one does not express -SMA, but can be induced by TGF- and is
sensitive to FP. The second one constitutively expresses -SMA and is
sensitive to FP, while the third expresses -SMA and is sensitive to
FP only in the presence of TGF-. It is possible that in mild
differentiated myofibroblastic cultures, TGF- also reactivates cells
constitutively expressing this marker, which become sensitive to the
inhibitory effects of FP. Interestingly, ICIG7 cultures, undergoing a
rapid and stable myofibroblastic differentiation by changing the
adhesion substratum, behave as the third subset identified in BRONCO5
cells. This result strengthens the existence of a step in the
myofibroblastic differentiation process characterized by a constitutive
expression of -SMA and by the sensitivity to inducing and inhibitory
agents.
This mechanism could be relevant in vivo. Indeed, an early event during
the induction of allergic airway inflammation is the production of the
chemokine monocyte chemoattractant protein-1 that polarizes the Th2
cells, but also acts on lung fibroblasts, inducing collagen deposition
and secretion of TGF- (4). Thus, it is possible that an
early treatment with FP may synergize with the TGF- produced by
fibroblasts, inhibiting the induction of -SMA and the
myofibroblastic differentiation.
Our in vitro data are in agreement with clinical studies that show that asthma could be improved by an early intervention with ICS, preventing the development of irreversible airway obstruction that occurs over time if the asthma is undertreated (32, 33, 34, 35). This suggests that part of the therapeutic potential exhibited by ICS in vivo could be due to a direct effect on lung structural cells such as myofibroblasts.
Moreover, the fact that FP inhibition of -SMA expression also occurs
in cells already expressing this contractile protein suggests the
possibility of an in vivo reversibility of the fibrosis by an early
corticosteroid treatment, as suggested in recent in vivo studies on a
different type of inflammatory/fibrotic disease (18).
In conclusion, we propose that FP may exert at least in part of its
anti-inflammatory activity through a direct action on lung
myofibroblasts. FP, in concentrations that probably occur in airway
lining fluid during inhalation therapy (36), rapidly
inhibits JAK-STAT signal transduction in lung myofibroblasts induced by
IL-4 and IL-13. FP is efficacious not only on mild differentiated lung
myofibroblasts by decreasing the -SMA expression, but also acts on
terminally committed myofibroblasts, decreasing the nuclear
translocation of NFB.
|
Footnotes |
|---|
2 E.C. and J.G.-M. contributed equally to this work.
3 Address correspondence and reprint requests to Dr. Bruno Azzarone, Institut National de la Santé et de la Recherche Médicale Unité 506, Hospital Paul Brousse, Villejuif, France. E-mail address: bazzarone{at}hotmail.com
4 Abbreviations used in this paper: -SMA, -smooth muscle actin; ERK, extracellular signal-related kinase; FP, fluticasone propionate; GAM, goat anti-mouse; GR, glucocorticosteroid receptor; ICS, inhaled corticosteroid; JAK, Janus kinase; TYK2, tyrosine kinase 2.
Received for publication May 21, 2001. Accepted for publication August 16, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B and lack of I-B involvement. Br. J. Pharmacol. 127:1003.[Medline]
and interleukin 1 stimulate the human immunodeficiency virus enhancer by activation of the nuclear factor B. Proc. Natl. Acad. Sci. USA 86:2336.B activation by tumor necrosis factor in the Jurkat T cell line is independent of protein kinase A, protein kinase C, and Ca2+-regulated kinases. Cytokine 3:257.[Medline]
in Hep3B cells. Biochem. Biophys. Res. Commun. 235:317.[Medline]
B. Biol. Chem. 273:592.
. J. Biol. Chem. 274:20818., a splice variant of transcription factor STAT3, is a dominant negative regulator of transcription. J. Biol. Chem. 27:13221.
B: a pivotal role in asthma and a new target for therapy. Trends Pharmacol. Sci. 18:46.[Medline]
1. J. Cell Biol. 142:873.This article has been cited by other articles:
|
|
 |
|
  G. W. Canonica Treating Asthma as an Inflammatory Disease Chest, July 1, 2006; 130(1_suppl): 21S - 28S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Birrell, S. Wong, E. L. Hardaker, M. C. Catley, K. McCluskie, M. Collins, S. Haj-Yahia, and M. G. Belvisi I{kappa}B Kinase-2-Independent and -Dependent Inflammation in Airway Disease Models: Relevance of IKK-2 Inhibition to the Clinic Mol. Pharmacol., June 1, 2006; 69(6): 1791 - 1800. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Baouz, J. Giron-Michel, B. Azzarone, M. Giuliani, F. Cagnoni, S. Olsson, R. Testi, G. Gabbiani, and G. W. Canonica Lung myofibroblasts as targets of salmeterol and fluticasone propionate: inhibition of {alpha}-SMA and NF-{kappa}B Int. Immunol., November 1, 2005; 17(11): 1473 - 1481. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
