| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Division of Cardiovascular Research, The Hospital for Sick Children, Department of Laboratory Medicine and Pathobiology, and Medicine, University of Toronto, Toronto, Canada
1Correspondence: Division of Cardiovascular Research, The Hospital for Sick Children, 555 University Avenue, Toronto, Canada, M5G 1X8. E-mail:mr{at}sickkids.on.ca
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: transcription factors • proteinases • intracellular signaling • pulmonary hypertension
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). The changes observed in all these vascular pathologies
are characterized by smooth muscle cell proliferation, migration, and
accumulation of extracellular matrix glycoproteins such as collagen,
fibronectin, and tenascin (2
3
4
5)
. The beneficial
therapeutic effect of NO in patients with advanced pulmonary
hypertension appears to be related to its vasodilator properties as
well as to direct effects on the remodeling process, the nature of
which are obscure (6
, 7)
. The antiatherogenic effects of
NO have also been described as independent of its vasodilator function
(1)
and may be related to its role as an intracellular
signaling molecule. Administration of L-arginine, the substrate for NO,
suppresses monocrotaline-induced pulmonary vascular disease in rats
(7)
, and eNOS knockout mice have increased pulmonary
hypertension and vascular disease when subjected to chronic hypoxia
(8)
and more severe neointimal formation after cuff injury
of the femoral artery (9)
.
The features of vascular remodeling have been attributed to heightened
activity of proteolytic enzymes, both elastases and matrix
metalloproteinases (10
11
12
13
14
15
16
17
18)
. In experimental rabbits,
increased activity of a serine elastase has been demonstrated in
coronary arteries after cardiac transplant (14)
and in
vein grafts after arterial interposition (15)
. Increased
activity of a 20 kDa endogenous vascular serine elastase (EVE) appears
critical to the progression of experimental pulmonary vascular disease
(18)
. The activity of EVE is increased in the early period
after injection of the toxin monocrotaline (MCT) (10)
in
association with pulmonary endothelial injury (11)
.
Furthermore, administration of serine elastase inhibitors largely
prevents development or retard progression of hypoxia (12)
and monocrotaline-induced PVD in rats (13)
and reduces the
severity of the coronary arteriopathy after transplant
(14)
and the atherosclerotic degeneration of vein grafts
in rabbits (15)
. Inhibition of elastase activity not only
prevents the progression of vascular disease, but also appears to
induce regression (16
, 17)
.
The source of EVE appears to be smooth muscle cells (18)
,
and the sequelae of EVE activity have been shown in cultured cells to
be related to liberation of mitogenic growth factors from the
extracellular matrix (19)
and to induction of the
proliferative glycoprotein tenascin (16
, 17
, 20)
and of
fibronectin-dependent smooth muscle cell migration in response to
elastin peptides (21)
. EVE is produced in cultured smooth
muscle cells (SMCs) after serum stimulation and appears to function in
the microenvironment of the cell such that adhesion of elastin to the
cell surface by serum factors is critical in demonstrating the activity
of this enzyme (22
, 23)
. The induction of EVE by
serum-treated elastin (STE) is associated with tyrosine phosphorylation
of focal adhesion kinase (FAK) and ERK1 (24)
and by an
increase in the transcription factor AML1 (A and B isoforms)
(25)
. AML1 was initially described as a mutated gene in
acute myeloid leukemia. While it exits as multiple isoforms, the
products of alternative splicing of mRNA, the AML1B isoform is
consistently found in the nucleus where it is transcriptionally active.
Its expression is related to hematopoietic cell differentiation and
proliferation and to the transcription of genes such as leukocyte
elastase, myeloperoxidase, the receptor for macrophage
colony-stimulating factor, granulocyte-macrophage colony-stimulating
factor, osteocalcin, and the T cell receptor (26
27
28
29)
.
We speculated there might be a link between NO and the MAP kinase
signaling pathway that results in an increase in AML1B, the
transcription factor for EVE, since NO inhibits ERK activation by a
cyclic GMP-mediated activation of protein kinase G, which interferes
with RAS/raf interaction (30)
. We now demonstrate that NO
donors SNAP and DETA NONOate and a cGMP mimetic inhibit SMC elastase by
repressing ERK phosphorylation, and that the mechanism is consistent
with protein kinase G activity. We also report the novel finding that
the reduction in ERK phosphorylation shifts distribution of the
transcriptionally active isoform AML1B in the nucleus and the decrease
in AML1B in nuclear extracts is associated with reduced DNA binding.
These studies suggest a mechanism whereby NO might affect matrix
remodeling by suppressing the intracellular signaling cascade necessary
for the transcription of elastase.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
ATP, HRP-conjugated
anti-rabbit IgG antibody, and the enhanced chemiluminescence kit were
purchased from Amersham (Oakville, Ontario). PD 98059
(2'amino-3'-methoxyflavone) and SB202190 were obtained from Calbiochem
(San Diego, Calif.). Genistein, uric acid (UA),
4,5-dihydroxt-1,3-benzene-disulfonic acid (Tiron), diamide,
2',3'-dideoxycytidine (DDC), and diphenyleneiodonium (DPI) were from
Sigma, Mississauga, Ontario. Hydrogen peroxide was purchased from BDH
Inc. (Toronto, Ontario). DETA NONOate, 8-pCPT-cGMP, Rp-8-pCPT-cGMP,
peroxynitrite (ONOO-), and
1H-[1, 2, 4] oxadiazole [4,3-a]
quinoxalin-1-one (ODQ) were from Alexis Corp. (San Diego, Calif.).
S-nitro-N-acetylpenicillamine (SNAP) was obtained from BIOMOL Research
Laboratories (Plymouth Meeting, Pa.). Antibodies against
phospho-specific and -nonspecific ERK, or phospho-specific and
-nonspecific p38 were purchased from New England Biolabs (Beverly,
Mass.); an antibody against a 17 amino acid amino-terminal peptide
based on the human AML1 sequence was from Calbiochem (LaJolla, Calif.).
The Bradford reagent came from Bio-Rad Laboratories (Hercules, Calif.).
Tris-glycine gels were from Helix (Mississauga, Ontario) and
polyvinylidene difluoride (PVDF) membranes from Millipore (Bedford,
Mass.)
Isolation and culture of vascular SMCs
Pulmonary artery SMCs were harvested from central vessels of
juvenile pigs and cultured as in our previous studies
(22
23
24)
. Explants of the vascular media were cultured in
medium M199 supplemented with 10% FBS and 1% antibiotics-antimycotics
(100 units/ml penicillin, 100 µg/ml streptomycin, amphotericin B 250
ng/ml) in a humid 5% CO2 environment at 37°C.
The SMCs were used at passages 2–4.
Preparation of radiolabeled elastin and STE
Purified elastin from bovine neck ligament was radiolabeled
using [3H]-NaBH4, as
described previously (31)
. The
[3H]-elastin was reconstituted at 16 mg
elastin/ml (specific activity 2,000 cpm/µg elastin) in Tris assay
buffer (50 mmol/l Tris HCl, 150 mmol/l NaCl, 10 mmol/l
CaCl2 2H2O, 0.02% Brij, pH
8.0), boiled, and stored at -70°C until use. Before each assay,
[3H]-elastin was washed with Tris assay buffer
until the background counts became less than 100 cpm/100 µl
supernatant. The [3H]-elastin suspension was
then diluted to 105 cpm/20 µl. Pretreating the
[3H]-elastin with FBS has been shown to
stimulate elastin adhesion to SMC surfaces and induce endogenous
vascular elastase activity (22
23
24
25)
. Radiolabeled
insoluble elastin was incubated with FBS at a concentration of 10 mg/ml
and rotated overnight at 37°C, then washed four times with Tris assay
buffer to remove all unbound serum factors. The pretreated
[3H]-elastin suspension was then diluted to
105 cpm/20 µl with Tris assay buffer and added
to cultured SMCs for assay of elastase activity.
Activity assay of elastase in cultured SMCs
The assay has been previously described in detail
(22
23
24
25)
. Briefly, confluent SMCs were passaged to 24
wells of a 16 mm diameter multiwell tissue culture plate
(~5x104 cells per well). At confluence, SMCs
were serum starved for 24 h in serum-free M199 with 0.1% BSA and
1% antibiotics. Cell monolayers were incubated for another 24 h
with 20 µl [3H]-elastin either in serum-free
medium (control) or with serum-treated elastin prepared as described
above in a final volume of 1 ml in each well. To investigate the
effects of NO donors SNAP (0.01–1 mM) and DETA NONOate (0.1–1 mM),
and a cGMP mimetic 8-pCPT-cGMP (1 mM) on elastase activity, cells were
pretreated with each compound for 30 min. To evaluate the effect of
coadministration of a cGMP PKG inhibitor Rp-8-pCPT-cGMP (20 µM) on
SMCs treated with SNAP, cells were pretreated with Rp-8-pCPT-cGMP for
30 min. To check the effect of a specific p38 inhibitor on elastase
activity, cells were pretreated with SB202190 (20 µM) for 30 min;
control cells and STE-stimulated cells were similarly pretreated with
the vehicle 0.1% DMSO. SNAP, DETA NONOate, 8-pCPT-cGMP, and
Rp-8-pCPT-cGMP were freshly prepared in phosphate-buffered saline
(PBS), whereas SB202190 was dissolved in DMSO. The dose range of each
compound was based on previous studies (32
33
34
35
36
37
38)
. Eight
hundred µl of culture medium were harvested and microcentrifuged at
8160 g for 5 min. The amount of
[3H]-soluble elastin peptides in 600 µl of
supernatant was determined by scintillation counting (LKB Wallac 1219
Rackbeta counter, San Francisco, Calif.). To control for nonenzymatic
degradation of the elastin substrate, we incubated 20 µl of
[3H]-elastin with medium in cell-free wells in
the presence of each test compound and subtracted the counts obtained
as background. Compounds found to nonenzymatically degrade elastin
(background counts above 200 cpm/well) were not used in these assays.
To confirm the viability of cells in the presence of each test
compound, the trypan blue exclusion method was used after incubation
for 24 h.
Preparation of the total cell lysate
Total cell lysates were prepared to determine expression of
phospho-specific or -nonspecific ERK or p38. After 5 x
105 SMCs were plated and cultured for 36 h
in 60 mm dishes, confluent monolayers of SMCs were serum starved for
36 h. The SMCs were exposed to STE or an equal amount of elastin
(control) for 5, 10, or 30 min depending on the experiment after
pretreatment with DETA NONOate (0.1–1 mM), SNAP (0.01–1 mM), a cGMP
mimetic 8-pCPT-cGMP (1 mM), peroxynitrite ONOO-
(0.1 mM), and the inactivated decomposed control for
ONOO- for 30 min. To evaluate the effect of
coadministration of a cGMP PKG inhibitor Rp-8-pCPT-cGMP (20 µM) on
SMCs pretreated with SNAP, cells were pretreated with Rp-8-pCPT-cGMP
for 30 min, followed by treating cells with SNAP. After various
treatments, cells were lysed by addition of boiling lysis buffer
[sqb]1% sodium dodecyl sulfate (SDS), 1 mmol/l sodium vanadate, 10
mmol/l Tris pH 7.4]), scraped into a microcentrifuge tube, and boiled
for 5 min. Samples were centrifuged for 10 min to remove insoluble
material.
Preparation of subcellular fractions
Cytosolic fractions, nuclear extracts, and pellets were prepared
to determine the distribution and relative expression of AML1 isoforms
and whether there was a shift in molecular weight that would suggest an
alteration in phosphorylation status. After 2 x
106 cells were plated and cultured in 150 mm
dishes for 36 h, confluent monolayers of SMCs were serum starved
for 36 h Serum-starved cells pretreated with DETA NONOate (1 mM),
SNAP (1 mM), or 8-pCPT-cGMP (1 mM) were stimulated with STE or control
elastin for 30 min. To evaluate the effect of coadministration of a
cGMP and PKG inhibitor, Rp-8-pCPT-cGMP (20 µM), on SMCs pretreated
with SNAP, cells were pretreated with Rp-8-pCPT-cGMP for 30 min,
followed by treating the cells with SNAP. To check the effect of a
specific p38 inhibitor, SB202190, on AML1 isoform expression, cells
were pretreated with SB202190 (20 µM) for 30 min; control cells and
STE-stimulated cells were similarly pretreated with the vehicle, 0.1%
DMSO.
Subcellular fractions were prepared according to a standard protocol
(39)
. Briefly, after removing medium, cells were rinsed in
cold PBS, scraped into microcentrifuge tubes, and the pellet was washed
in PBS. Cells were lysed in cold hypotonic buffer [10 mmol/l
N-2-hydoxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES; pH 7.9), 1.5
mmol/l MgCl2, 19 mmol/l KCl, 0.2 mmol/l PMSF, 0.5
mmol/l DTT], followed by 15 passes with a type B Dounce homogenizer
(VWR Scientific Ltd., London, Ontario). Nuclei were pelleted at 3300
g for 15 min at 4°C. The supernatant was mixed with 0.11
vol of 10 X cytoplasmic extract buffer (0.3 M HEPES pH 7.9, 1.4 M NaCl,
0.03 M MgCl2) and centrifuged at 16,000
g for 30 min. The supernatant is designated as S100. The
nuclei pelleted were extracted in a high-salt buffer (20 mmol/l HEPES
pH 7.9, 25% glycerol, 1.5 mmol/l MgCl2, 0.8
mol/l NaCl, 0.2 mmol/l EDTA, 0.2 mmol/l PMSF, 0.5 mmol/l DTT) with
vortexing for 40 min, then microcentrifuged at 16,000 g for
30 min. The supernatants designated as nuclear extract were rapidly
frozen and stored at -70°C. The residual nuclear pellet was washed
with high-salt buffer once, boiled, and extracted by vortexing in SDS
buffer (1% SDS, 10 mmol/l Tris pH 7.4) for 60 min. After
microcentrifugation at 16,000 g for 30 min, the
salt-resistant supernatants designated as nuclear pellet were frozen at
-70°C until use.
Western immunoblotting
Aliquots of 10 µg of total protein from total cell lysate or
from different subcellular fractions, as determined by Bradford protein
assay were electrophoresed under reducing conditions by SDS-PAGE
(SDS-polyacrylamide gel electrophoresis) on an 8–16% polyacrylamide
Tris-glycine gel and transferred onto a PVDF membrane. Nonspecific
binding was blocked by incubating the blot in blocking buffer (5% dry
non-fat milk in 10 mmol/l Tris, pH 7.4, 50 mmol/l NaCl, and 0.5% Tween
20) for 1 h at room temperature.
The blot was then incubated with primary antibodies against phospho-specific or -nonspecific ERK, phospho-specific or -nonspecific p38 (1:1000), or AML1 (1:200). After the blots were washed with TBS-T (10 mmol/l Tris HCl, pH 7.4, 50 mmol/l NaCl and 0.5% Tween 20) for 30 min, they were incubated with HRP-conjugated anti-rabbit antibodies (1:5000) at room temperature for 1 h After the blots were washed with TBS-T for 30 min, protein bands were visualized by enhanced chemiluminescence. The intensity of specific bands was quantified by scanning soft-laser densitometry (Bio-Rad Gel Doc 1000) and values from three different harvests in each group were averaged. Equal loading and transfer of proteins were confirmed by visualizing proteins after staining the gel with Coomassie blue.
Electrophoretic mobility shift assay (EMSA)
The protein AML1:DNA binding site interactions were analyzed by
EMSA according to the protocol described by Meyers et al.
(26)
. The wild-type AML1 binding site double-stranded DNA
oligonucleotides (similar for all isoforms) were prepared by annealing
the complementary oligomers 5'-AATTCGAGTAT TGTGGTTAATACG-3'
and 5'-AATTCGTATTAACCACAATACTCG-3' (consensus binding site
is italicized) and labeling with [32P]dATP
in a Klenow reaction. The oligomers
5'-AATTCGAGTATTGTTAGTAATACG-3' and
5'-AATTCGTATTACTAACAATACTCG-3' were annealed to form an AML1
mutant oligonucleotide. Briefly, the AML1 binding site reaction
contained 10 µg total protein from nuclear extracts and 1 ng of
radiolabeled oligonucleotide in binding buffer (20 mmol/l HEPES, pH
7.8, 1 mmol/l MgCl2, 0.1 mmol/l EGTA, 0.4 mmol/l
DTT, 40 mmol/l KCl, 10% glycerol and 60 µg/ml salmon sperm DNA) and
proceeded for 30 min at room temperature. In competition studies, 50 ng
of unlabeled double-stranded wild type or mutant oligonucleotides or
0.5 µg of AML1 antibody were preincubated for 30 min with the
STE-treated nuclear extracts prior to the addition of radiolabeled
oligonucleotides. The reactions were resolved on a 5% nondenaturing
polyacrylamide gel in TBE buffer (50 mmol/l Tris borate and 1.0 mmol/l
EDTA, pH 8.0) and exposed to Kodak XAR film.
Statistical analysis
Statistical analyses were carried out on absolute values or
ratios as indicated. Differences between the treatment groups were
determined by a one-way ANOVA, followed by Student-Newman-Kuels test or
a Fisher’s protected LSD, if specified. A level of P <
0.05 was accepted as statistically significant. Data are presented as
mean ± SE. The number of experiments is
given in the figure legend.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). SNAP did not depress basal levels of elastase activity.
Since SNAP can also produce small amounts of superoxide
(35)
, we confirmed that the effect of SNAP was
specifically related to NO by showing that DETA NONOate, a pure NO
donor, similarly inhibited STE-induced elastase activity in a
dose-dependent manner (P<0.05) (Fig. 1B
). We
next determined that the mechanism of NO-mediated suppression of
elastase appeared to be related to cGMP-mediated PKG activity.
Pretreatment with a cGMP mimetic, 8-pCPT-cGMP (1 mM), abolished
elastase activity (P<0.05), and coadministration with the
PKG inhibitor Rp-8-pCTP-cGMP (20 µM) reversed the SNAP suppression of
elastase activity (P<0.05) (Fig. 1B
). There were
no significant effects of any of these compounds on the basal levels of
elastase activity, nor did they induce nonenzymatic degradation of
labeled elastin in cell-free wells (data not shown). These compounds
were not toxic to the cells, as evaluated by trypan blue exclusion
tests, and the lack of the significant effect of STE on DNA synthesis
had previously been confirmed.
|
To determine a direct link between NO suppression of elastase activity and soluble guanylate cyclase, we attempted to use the NO-sensitive soluble guanylate cyclase inhibitor ODQ (0.01 mM). This compound, however, caused nonenzymatic degradation of radiolabeled elastin. To negate the possibility that NO-mediated inhibition of elastase is related to the production of ONOO- that results from NO interaction with endogenous superoxide, several oxidants (ONOO-, hydrogen peroxide, diamide, and DDC) and antioxidants (UA, DPI, and Tiron) were tested. However, these compounds all induced nonenzymatic degradation of radiolabeled elastin.
NO suppresses ERK phosphorylation via cGMP
Our previous studies showed that STE-induced elastase activity was
mediated by ERK phosphorylation (24)
. We therefore
determined whether NO might suppress ERK phosphorylation. Protein
expression of ERK is not modulated by STE or any of the compounds used
(Fig. 2A, B
). STE pretreatment of SMCs induced a >10-fold increase
in phosphorylation of ERK (P<0.05). This induction was
reduced by > 60% with DETA NONOate and completely reversed by 1
mM of DETA NONOate as well as 0.1–1 mM of SNAP (P<0.05 for
all comparisons) (Fig. 2A
). Although for the technical
reasons described above we could not determine whether
ONOO- produced from NO and endogenous superoxide
suppressed elastase activity, we were able to show that when cells were
pretreated with ONOO- (0.1 mM) or its decomposed
inactive control, ERK phosphorylation was not inhibited (Fig. 2A
). Consistent with this, coadministration of a
ONOO- scavenger uric acid (0.1 mM) did not
reverse SNAP suppression of ERK phosphorylation in two independent
experiments (data not shown).
|
The effect of NO donors was consistent with PKG suppression of ERK
phosphorylation, since pretreatment of cells with the cGMP mimetic
8-pCPT-cGMP (1 mM) abolished STE-induced ERK phosphorylation and
administration of the cGMP PKG inhibitor Rp-8-pCTP-cGMP (20 µmol/l)
reversed SNAP suppression of STE-induced ERK phosphorylation
(P<0.05) (Fig. 2B
).
NO-cGMP mechanism inhibits nuclear expression and DNA binding of
AML1
Our previous studies showed that STE induction of ERK
phosphorylation is necessary for nuclear expression and DNA binding of
the transcription factor for elastase, AML1 (B isoform)
(24)
. We therefore determined whether we could correlate
the inhibitory effect of NO donors on ERK phosphorylation with a
reduction in nuclear expression of AML1B. STE induced a >
threefold increase in nuclear expression of AML1B (the 50 kDa
transcriptionally active isoform) (Fig. 3A
). Pretreatment of SMCs with DETA NONOate (1 mM), SNAP (1
mM), or 8-pCPT-cGMP (1 mM) abolished STE-induced nuclear expression of
AML1B (Fig. 3A
). Consistent with a cyclic GMP-mediated PKG
effect, coadministration of Rp-8-pCTP-cGMP (20 µmol/l) reversed SNAP
inhibition of STE-induced nuclear expression of AML1B (Fig. 3D
).
|
To exclude the possibility that STE-mediated induction of nuclear
expression of AML1B or its suppression by the NO-PKG pathway was
associated with a translocation from or to the cytoplasm, we also
monitored AML1B in cytoplasmic extracts but found no appreciable
changes with STE, NO donors, or the PKG stimulator or inhibitor at the
30 min time point (Fig. 3B, E
). AML1B has also been found in
the nuclear pellet, so it was of interest that STE decreases AML1B in
the nuclear pellet (P<0.05); this decrease was reversed by
DETA NONOate (1 mM), SNAP (mM), or 8-pCPT-cGMP (1 mM)
(P<0.05 by Fisher’s protected LSD) (Fig. 3C
).
The SNAP-induced reversal of AML1 expression in the nuclear pellet was
inhibited by coadministration of Rp-8-pCTP-cGMP (20 µmol/l;
P<0.05 by Fisher’s protected LSD) (Fig. 3F
). This suggests that NO-cGMP-PKG-mediated ERK
phosphorylation might redistribute AML1B in the nucleus.
To investigate whether this change might be associated with alteration in the phosphorylation status of AML1B, the nuclear extracts, cytoplasmic extracts and nuclear pellets from all the above conditions were analyzed by 8–16% SDS-PAGE run for a maximal time (>4 h) and immunoblotted with an antibody recognizing all AML1 isforms. We could not identify a band shift among different subcellular fractions or among different treatment groups, which would suggest an altered phosphorylation state, nor could we show differences upon phosphatase treatment of the various fractions (data not shown).
To confirm that the form of AML1B found in the nuclear extract could
bind DNA, EMSA was carried out as described in Materials and Methods
using a radiolabeled DNA oliogonucleotide encoding the AML1B binding
sequence (similar in all isoforms). The protein:DNA complexes formed
after stimulation with STE were inhibited by DETA NONOate (1 mM), SNAP
(1 mM), or 8-pCPT-cGMP (1 mM) (Fig. 4A
). Consistent with changes in AML1 protein expression, SNAP suppression
of STE-induced protein:DNA binding was reversed by coadministration of
Rp-8-pCTP-cGMP (20 µmol/l) (Fig. 4B
). Competition
experiments were also carried out to confirm the specificity of the
protein:DNA binding. A 50-fold increase in unlabeled wild-type AML1
oligonucleotide, but not the mutant oligonucleotide, inhibited the
formation of the protein:DNA complexes. The formation of these
complexes was also inhibited by the AML1 antibody, even though a
ternary AML1 antibody/protein/DNA complex (supershift) was not found.
The failure to produce a supershift might be due to the different
source of AML1 polyclonal antibody from that used in our previous
experiments where supershifts were demonstrated (25)
.
|
To negate the possibility that NO suppression of nuclear expression of AML1B is due to ONOO-, SMCs were pretreated with a ONOO- scavenger UA or a cell-permeant scavenger of superoxide, but SNAP suppression of STE-induced nuclear expression and DNA binding of AML1B was not reversed in two independent experiments (data not shown).
STE activation of ERK, not p38, causes NO suppression of AML1 and
elastase
To confirm the specificity of ERK among MAP kinases in inducing
AML1 expression and elastase activity, we investigated whether another
MAP kinase, p38, is phosphorylated by STE and whether SNAP inhibits
phosphorylation of p38. STE induced a > fivefold increase in
phosphorylation of p38 at 5 min (P<0.05) and values
returned to baseline by 30 min; protein expression of p38 was
unaffected (Fig. 5A
). Pretreatment of SMCs with SNAP (0.1 mM and 1 mM)
abolished the induction of p38 phosphorylation (P<0.05)
(Fig. 5B
). However, pretreatment of SMCs with a specific p38
inhibitor, SB202190, did not prevent the increase in AML1B in nuclear
extracts (not shown) or its DNA binding (Fig. 5C
) induced by
STE, and inhibited by the MEK inhibitor PD 98059. Consistent with this
finding, SB202190 did not inhibit STE-induced elastase activity. The
lack of effect of SB202190 on nonenzymatic degradation of elastin or
unstimulated control cells was confirmed (data not shown).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 14
, 15
, 49
, 41)
. Its pathophysiologic
significance has been documented in cultured smooth muscle cells, where
alterations in cell:extracellular matrix interactions induced by
heightened proteolytic activity are major determinants of increased
cell proliferation and migration (16
, 17
, 19
, 20)
.
Impaired endothelial NO production is also a feature of vascular
disease, and restoration of endothelial function or administration of
NO donors ameliorates these disorders (reviewed in ref 1
).
A variety of studies have suggested that the beneficial effect of NO
may be related to its role as an intracellular signaling molecule as
well as to its vasodilatory properties. NO can inhibit the mitogenic
response of SMCs to growth factors and the production of extracellular
matrix proteins such as proteoglycans, osteopontin, and thrombospondin,
the latter via protein kinase G (1
, 30
, 37
, 38)
. We now
report that NO can inhibit vascular remodeling by suppressing the
signaling mechanism required for the transcription of elastase. We
established that different NO donors inhibit elastase activity induced
in cultured vascular smooth muscle cells by STE and that the effect is
mediated by cGMP. Since elastase activity is dependent on
phosphorylation of ERK and on increased AML1B in nuclear extracts, we
documented that NO-mediated cGMP suppresses this pathway in association
with an alteration in the nuclear distribution of AML1B. The mechanism
is likely related to cGMP-mediated protein kinase G activity, which
interferes with upstream events, namely, the interaction between ras
and raf that is necessary for the activation of the MAP kinase kinase
(MEK) that phosphorylates ERK (30)
. The specificity of the
link between ERK and AML1B/elastase was further established when a MEK
inhibitor, but not a p38 inhibitor, reduced AML1B:DNA interaction in
nuclear extracts and elastase activity. These findings suggest not only
a novel link between NO and elastase-mediated matrix remodeling in
vascular disease, but also propose a new paradigm whereby
AML1B-mediated gene expression may be one of the downstream targets of
NO/cGMP-generating vasodilators in myeloid as well as nonmyeloid cells.
NO can modulate intracellular signaling in SMCs through cGMP-dependent
or -independent mechanisms (reviewed in ref 1
). Our
findings demonstrate that NO donor suppression of STE-induced elastase
activity was associated with the inhibition of ERK phosphorylation
mediated by cGMP. cGMP involvement in NO suppression of elastase
negates the possibility that NO directly inhibits elastase enzymatic
activity shown for cysteine proteinases (42)
. The NO donor
SNAP can produce ONOO-, which could be a weak
releaser of cGMP (36)
. However, DETA NONOate, a pure NO
donor that does not release superoxide or generate
ONOO-, had a similar inhibitory effect on ERK
phosphorylation. Intracellular generation of
ONOO- was also ruled out when we showed that
administration of exogenous ONOO- did not
inhibit ERK phosphorylation, and coadministration of a
ONOO- scavenger UA did not reverse SNAP
suppression of ERK phosphorylation.
We speculate that the mechanism whereby NO/cGMP interferes with
STE-induced ERK phosphorylation is related to induction of PKG
suppressing Ras:raf interaction and MEK activity, as was shown in
EGF-stimulated rat aortic smooth muscle cells (30)
. Other
possibilities also need to be considered, such as the effect of NO-cGMP
on integrin–SMC interactions required for the induction of elastase
activity (24)
. We showed that STE stimulation of vascular
cells results in phosphorylation of focal adhesion kinase (FAK) as well
as ERK and that pretreatment with RGD peptides to disrupt
integrin:matrix interactions inhibits elastase activation
(24)
. Recent studies in chondrocytes demonstrated that
NO/cGMP can inhibit 5ß1 integrin-mediated interaction with
fibronectin and subsequent actin polymerization, FAK translocation, MAP
kinase activation, and proteoglycan synthesis (43)
.
The mechanism whereby NO reduces expression of AML1B in nuclear
extracts in a cGMP-dependent manner is related to inhibition of ERK
phosphorylation. Although another member of the MAP kinase family, p38
could be involved in inducing expression of downstream genes (33
, 44)
, such as AML1B through cross talk with ERK, inhibition of
p38 does not alter nuclear expression of AML1B even though STE induces
p38 phosphorylation, which is inhibited by SNAP.
An interesting aspect of this work addresses how ERK phosphorylation
induces AML1B functional activity. The subnuclear localization of
transcriptionally active AML1B has been investigated in a number of
studies. Tanaka et al. (45)
reported that in COS cells
transfected with human AML1B, localization of the transcription factor
in nuclear extracts is associated with DNA binding and reporter
activity of a downstream gene, the T cell receptor. Studying Jurkat
cells transfected with human AML1B, Zeng et al. (46
, 47)
demonstrated that AML1B is found exclusively in nuclear pellets, where
it is transcriptionally active when colocalized with RNA polymerase ll.
Our studies in vascular smooth muscle cells are more comparable to
those of Tanaka’s in COS cells in that when STE stimulation shifts the
distribution of AML1B from the nuclear pellet to the extract,
expression and DNA binding property support its transcriptional
activity in terms of elastase activity. Recent studies suggest that
functional activity of certain transcription factors is related to
their ability to partition between the nuclear matrix and the extract,
i.e., mutated transcription factors remain bound to the matrix
(48
, 49)
. Thus, whereas ‘functional’ transcription
factors are bound to the matrix, ‘functioning’ transcription factors
may not be (48)
. Binding of transcription factors to the
nuclear matrix has also been associated with transcriptional inhibitory
activity (50)
. We have not, however, specifically
evaluated the AML1B in the pellet associated with RNA polymerase II;
this subfraction might be affected differently by STE and NO donors
when compared to the portion in the pellet as a whole.
The mechanism whereby ERK phosphorylation induces AML1B redistribution
in the nucleus and its functional activity remains to be determined.
Tanaka et al. (45)
demonstrated a band shift in AML1B on
immunoblots from COS cells transfected with human AML1B after
stimulation in 10% fetal calf serum. This phosphorylation was
associated with up-regulation not of DNA binding, but of
transcriptional activity, probably by modulating interaction with other
transcription factors.
In our studies, no AML1B band shift was evident under different
treatments or in the different subnuclear fractions associated with ERK
phosphorylation, making it unlikely that ERK directly phosphorylates
AML1B. Moreover, the site of ERK-mediated AML1B phosphorylation is
different from the nuclear matrix targeting signal, perhaps making it
less likely that ERK-mediated phosphorylation is responsible for the
shift in nuclear distribution (45
, 46)
. It is therefore
possible that phosphorylation of other transcription factors by ERK may
modulate AML1B nuclear localization. For example, Ets requires
phosphorylation and activation by ERK in order to interact with AML1B
through autoinhibitory domains, and this increases AML1B:DNA binding
and transcriptional activity (51)
. It would be interesting
to investigate whether binding to phosphorylated Ets also alters the
nuclear distribution of AML1B in smooth muscle cells.
Although the functional significance of AML1B as a transcription factor
in vascular smooth muscle cells has only recently been investigated by
our group, (25)
, AML1B regulates the transcription of
various genes that are important in hematopoiesis such as leukocyte
elastase, myeloperoxidase, granulocyte-macrophage-colony stimulating
factor, and the receptor for macrophage-stimulating factor
(27)
. In addition to hematopoiesis, AML1 isoforms are
involved in osteocalcin production, fibroblast transformation, and
skeletal muscle development (52
53
54)
. Our findings
therefore suggest that NO/cGMP-generating vasodilators could influence
AML1B-mediated gene expression not only in vascular remodeling, but
also in remodeling of bone and other tissues as well as in
hematopoeisis and neoplastic transformation. Thus, in myeloid or
nonmyeloid cells, AML1B could be a therapeutic target of a variety of
cGMP-generating vasodilators and vasculoprotective compounds such as
natriuretic peptides, carbon monoxide, and phosphodiesterase 5
inhibitors.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
enhancer. Mol. Biol. Cell. 15,3090-3099
This article has been cited by other articles:
|
|
 |
|
  S. Mishra, T. Fujita, V. N. Lama, D. Nam, H. Liao, M. Okada, K. Minamoto, Y. Yoshikawa, H. Harada, and D. J. Pinsky Carbon monoxide rescues ischemic lungs by interrupting MAPK-driven expression of early growth response 1 gene and its downstream target genes PNAS, March 28, 2006; 103(13): 5191 - 5196. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Humpl, J. T. Reyes, H. Holtby, D. Stephens, and I. Adatia Beneficial Effect of Oral Sildenafil Therapy on Childhood Pulmonary Arterial Hypertension: Twelve-Month Clinical Trial of a Single-Drug, Open-Label, Pilot Study Circulation, June 21, 2005; 111(24): 3274 - 3280. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. M. Hassoun Deciphering the "matrix" in pulmonary vascular remodelling Eur. Respir. J., May 1, 2005; 25(5): 778 - 779. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Humbert, N. W. Morrell, S. L. Archer, K. R. Stenmark, M. R. MacLean, I. M. Lang, B. W. Christman, E. K. Weir, O. Eickelberg, N. F. Voelkel, et al. Cellular and molecular pathobiology of pulmonary arterial hypertension J. Am. Coll. Cardiol., June 16, 2004; 43(12_Suppl_S): 13S - 24S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Florio, S. Arena, A. Pattarozzi, S. Thellung, A. Corsaro, V. Villa, A. Massa, F. Diana, G. Spoto, S. Forcella, et al. Basic Fibroblast Growth Factor Activates Endothelial Nitric-Oxide Synthase in CHO-K1 Cells via the Activation of Ceramide Synthesis Mol. Pharmacol., February 1, 2003; 63(2): 297 - 310. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. H. Campos, Y. Zhao, M. J. Pollman, and G. H. Gibbons DNA Microarray Profiling to Identify Angiotensin-Responsive Genes in Vascular Smooth Muscle Cells: Potential Mediators of Vascular Disease Circ. Res., January 10, 2003; 92(1): 111 - 118. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. H. Campos, W. Wang, M. J. Pollman, and G. H. Gibbons Determinants of Notch-3 Receptor Expression and Signaling in Vascular Smooth Muscle Cells: Implications in Cell-Cycle Regulation Circ. Res., November 29, 2002; 91(11): 999 - 1006. [Abstract] [Full Text] [PDF] |
P < 0.05 between designated groups.
