Extracellular ATP Is an Autocrine/Paracrine Regulator of
Hypoxia-induced Adventitial Fibroblast Growth
SIGNALING THROUGH EXTRACELLULAR SIGNAL-REGULATED KINASE-1/2 AND
THE Egr-1 TRANSCRIPTION FACTOR*
Evgenia V.
Gerasimovskaya
§,
Shama
Ahmad¶,
Carl W.
White¶,
Peter L.
Jones,
Todd C.
Carpenter, and
Kurt R.
Stenmark
From the Developmental Lung Biology Research
Laboratory, Department of Pediatrics, University of Colorado Health
Sciences Center, Denver, Colorado 80262 and ¶ Department of
Pediatrics, National Jewish Medical and Research Center,
Denver, Colorado 80206
Received for publication, March 28, 2002, and in revised form, August 6, 2002
 |
ABSTRACT |
Important autocrine/paracrine
functions for the adenine nucleotides have been proposed in several
tissues. We addressed the possibility that extracellular ATP would
modulate/mediate hypoxia-induced adventitial fibroblast growth. Acute
hypoxia (3% O2, 10-60 min) increased extracellular
ATP concentrations in adventitial fibroblasts and in lung microvascular
endothelial cells, and chronic hypoxia (3% O2, 14-30
days) markedly attenuated the rate of extracellular ATP
hydrolysis by ecto-nucleotidase(s). Exogenous ATP stimulated [3H]thymidine incorporation in fibroblasts as did UTP,
ADP
, 2-methylthioadenosine triphosphate, adenosine
5'-(
,-methylene)triphosphate, and benzoylbenzoyl-ATP (2'-3'-O-(4-benzoylbenzoyl)-ATP), indicating that both P2Y and P2X
purinoceptors can mediate mitogenic responses. Suramin (100 µM), Cibacron blue 3GA (100 µM), and
pyridoxalphosphate-6-azophenyl-2',-4'-disulfonic acid (100 µM) as well as apyrase (5 units/ml) attenuated hypoxia- and ATP-induced and DNA synthesis, indicating activation and a functional role of purinoceptors under hypoxic conditions. ATP-induced DNA synthesis was augmented by hypoxia in an additive fashion, whereas
ATP and hypoxia synergistically increased growth factor-induced DNA
synthesis, again suggesting that ATP and hypoxia utilize similar signaling pathways to induce proliferation. Indeed, we found that ATP
(100 µM) and hypoxia (3% O2) induced
expression and activation of Egr-1 transcription factor, and both
stimuli acted, in part, through a
Gi/ERK1/2-dependent signaling
pathway. Suramin, Cibacron blue 3GA, and apyrase attenuated
hypoxia-induced ERK1/2 activation and Egr-1 expression. We conclude
that hypoxia induces ATP release from endothelial cells and fibroblasts
and that the activation of P2 purinoceptors is involved in the
regulation of DNA synthesis by fibroblasts under hypoxic conditions.
| |
INTRODUCTION |
Hypoxia has been shown to act as a direct proliferative stimulus
for fibroblasts in a variety of organs. This capability of fibroblasts
is unusual, at least among mesenchymally derived cells, and appears to
be important in normal development, wound healing, and fibrosis as well
as in the vascular changes that characterize hypoxic pulmonary
hypertension (1-3). With regard to pulmonary artery adventitial
fibroblasts, we have shown that among the resident vascular wall cells
they exhibit the earliest and most dramatic responses to hypoxic
exposure in vivo (4). In tissue culture we have also
demonstrated that hypoxia in the absence of exogenous mitogens induces
proliferation of pulmonary artery fibroblasts as well as some
subpopulations of aortic adventitial fibroblasts. This response is due
in large part to Gi/o-mediated activation of a complex
network of mitogen-activated protein kinases, whose specific
contributions to hypoxia-induced proliferation differ from those of
serum-induced growth signals (5). It remains unclear, however, whether
either activation or augmentation of the hypoxia-induced growth
response is due at least in part to autocrine/paracrine responses to
factors secreted by fibroblasts during hypoxia, which act through
G-protein-coupled pathways.
One factor that could contribute to such an autocrine loop is ATP.
Purines and pyrimidines (mainly ATP, ADP, adenosine, and UTP) have
widespread and specific extracellular signaling actions in the
regulation of a variety of functions in many tissues and appear to have
key roles in development, proliferation, differentiation, and release
of hormones, neurotransmitters, and cytokines (6-10). It is also
becoming evident that alterations in the physiology of purinergic
signaling may result in the development of a variety of pathologies
including disorders of the immune system, neurodegenerative, and
vascular diseases (7). Extracellular ATP can, in fact, stimulate the
growth and proliferation of vascular smooth muscle cells
(SMC),1 and this response may
play an important role in a variety of vascular diseases (10-12).
Importantly, in addition to nerves and circulating blood cells,
vascular cells themselves appear to be a potent source of ATP and other
adenine nucleotides as these products are known to be released into the
extracellular milieu in response to many vascular stress conditions
including ischemia/oxidative stress, flow, and mechanical stretch
(13-16). Because vascular cells have been shown to express
metabotropic (P2Y) and/or ionotropic (P2X) subtypes of purinergic
receptors and because these receptors have been shown to be implicated
in mitogenic signaling pathways (8, 9, 17), released ATP could induce
cell activation by an autocrine/paracrine mechanism. In fact, the
extracellular concentration of ATP and subsequent purinergic activation
appears to play a critical role in determining the intracellular
signaling set point of many key growth-regulating factors (18-20).
Furthermore, the concentration of extracellular ATP is rapidly
modulated by ecto-nucleotidases, which themselves may be regulated by
environmental factors. Therefore, the release of ATP and modulation of
its degradation by environmental or chemical stimuli may contribute to
the set point of cellular signaling pathways regulated by P2 receptors and raises the possibility that extracellular ATP is a critical modulator of signal transduction pathways operating to control proliferative responses. It remains unclear, however, whether hypoxia
actually increases extracellular concentrations of ATP in fibroblasts
and whether ATP is a component of the hypoxia-induced proliferative response.
It is known that hypoxia-induced cell proliferation is dependent on
expression of specific transcription factors. However, the
transcription factors involved in regulating the proliferative response
of non-transformed vascular fibroblasts to moderate levels of hypoxia
have not been elucidated. It is remarkable that many stress conditions
that can trigger cell proliferation, including hypoxia, mechanical
stress, and inflammation, can induce early growth response Egr-1
transcription factor (21-23), which has been shown to be critical for
proliferation and migration in many cell types (22, 24, 25). These
stress conditions have also been shown in independent experiments to
induce ATP release in a variety of cell types (7, 13, 26-28). Thus, it
seems possible that hypoxia itself and/or hypoxia-induced increases in
extracellular ATP could act to increase Egr-1 expression and activity,
which might be necessary for DNA synthesis. However, little information exists as to whether ATP or hypoxia directly regulates Egr-1 expression in adventitial fibroblasts.
The purpose of this study was to test the hypothesis that hypoxic
stress would increase extracellular ATP concentration, leading to the
creation of a "purinergic network" where local ATP release, degradation, and stimulation of purinergic receptors operate to control
fibroblast proliferation under hypoxic conditions. In addition, we
hypothesized that ATP and hypoxia would up-regulate expression and
activation of the Egr-1 transcription factor through Gi-initiated, ERK1/2-dependent pathways. We
used primary cultures of pulmonary artery adventitial fibroblasts to
examine the effects of hypoxia and ATP on growth responses, Egr-1
expression, and DNA binding activity. We evaluated the effects of
hypoxia on extracellular ATP concentration and on the enzymes
(ecto-ATPase(s)) that catalyze extracellular ATP degradation. Finally,
the role of ATP in hypoxia-induced growth, Egr-1 expression, and ERK1/2
phosphorylation was determined using pharmacological inhibitors. Our
findings strongly support the possibility that extracellular ATP plays
an important role in modulating hypoxic proliferative responses and
that it significantly modulates the response of the fibroblast to other
mitogens that may be present in the hypoxic environment.
| |
EXPERIMENTAL PROCEDURES |
Cell Culture--
Pulmonary artery adventitial fibroblasts were
isolated from tissue explants of 120-180-day gestational bovine
fetuses as previously described (29). Briefly, adventitia of lobar
pulmonary artery was mechanically separated from the media and
capillaries and chopped into small pieces (1 mm2) that were
then placed in 6-well plastic culture dishes (1 per well). When
numerous cells were observed around explants, tissue was removed, and
cells were cultured in Dulbecco's modified Eagle's medium
supplemented with 20 mM L-glutamine (Cellgro),
nonessential amino acids (1:100, v/v), 100 units/ml penicillin
and 100 µg/ml streptomycin (Sigma), and 10% fetal bovine serum (FBS)
(Gemini Bio-Products, Woodland, CA). Cell cultures were maintained in a
humidified atmosphere with 5% CO2 at 37 °C, and medium
was changed every 3 days. For expansion, fibroblasts were cultured to
confluence, harvested with trypsin-EDTA solution (0.2-0.5 g/liter),
and replaced at a 1:4 ratio. All studies were performed on cells
between passages 2 and 6. For Northern and Western blot analysis, cells
were plated at a density of 5 × 105
cells/cm2, cultured to 80% of confluence, growth-arrested
in 0.1% FBS/DMEM for 72 h, and then subjected to the experimental
conditions described below under "DNA Synthesis
Analysis."
Human lung microvascular endothelial cells were purchased as frozen
primary cultures from Clonetics Ltd. (Walkersville, MD). Cells were
cultured in endothelial cell basal medium (EBM-2) supplemented with
vascular endothelial growth factor, human fibroblast growth factor,
epidermal growth factor, hydrocortisone, ascorbic acid, insulin-like
growth factor, GA1000, and fetal bovine serum as per the
manufacturer's protocol. Hypoxic and normoxic bovine pulmonary artery
endothelial cells were generous gifts from Dr. H. Farber (Boston University).
DNA Synthesis Analysis--
DNA synthesis was determined by
[methyl-3H]thymidine incorporation. Cells were
plated in 24-well plates at a density of 15 × 10 3 cells/well in DMEM supplemented with 10% FBS. In 24 h cells were rinsed with phosphate-buffered saline (PBS) and incubated in
serum-deprived (0.1% FBS) DMEM for 72 h. Then cells were
stimulated with ATP (10
9-103
M) under either normoxic (21% O2, 5%
CO2 balance N2) or hypoxic (1, 3, or 10%
O2, 5% CO2 balance N2) conditions
at 37 °C in Plexiglas chambers (Bellco Glass, Vineland, NJ) in the
presence of 1.0 µCi of [methyl-3H]thymidine
(1 mCi/ml, 20 Ci/mmol, PerkinElmer Life Sciences) for 24 h. The
effects of UTP, ADPS, MeSATP, AMP-CPP, and BzATP (Sigma) on DNA
synthesis were tested under normoxic conditions in a similar manner.
When the effect of P2 receptor antagonists was tested, suramin,
Cibacron blue 3GA, or PPADS (Sigma) were added at the concentrations of
100 µM 30 or 60 min before stimulation with ATP and/or
hypoxia. For the experiments with apyrase, cells were preincubated for
2 h with a mixture of apyrases grade VI and VII (Sigma) at a final
concentration of 5 units/ml media. In the experiments aimed to prevent
ATP hydrolysis ARL67156 (300 µM) (30) or ATP-regeneration
system (0.1 mg/ml creatine kinase, 4 mM phosphocreatine,
and 0.025 mg/ml myokinase (31)) were added to the incubation media. At
the end of incubation cells were washed twice with PBS, incubated with
0.5 ml of 0.2 M perchloric acid for 3 min, washed with 1 ml
of PBS, and lysed in 0.3 ml of 1% SDS, 0.1 M NaOH. Samples
were harvested and mixed with liquid scintillation mixture (Ecoscint H,
National Diagnostics, Atlanta, GA), and incorporated radioactivity was
counted (cpm/min) in a liquid scintillation counter (Beckman LS 6500).
Northern Blot Analysis--
Total cellular RNA was extracted
with either TriReagent solution (Sigma) or with RNeasy kit (Qiagen
Inc., Santa Clarita, CA) according to the manufacturer's instructions.
RNA pellets were dissolved in diethyl pyrocarbonate-treated water and
heated for 10 min at 55 °C. RNA concentration was determined by
measurement of absorbency at 260 and 280 nm. Northern blot analysis was
carried out by using NorthernMax-GlyTM blotting kit
(Ambion, Austin, TX). 10-15 µg of total RNA were treated with
glyoxal loading buffer and separated by 1% agarose formaldehyde gel
electrophoresis at 85 V of constant voltage. After electrophoresis, RNA
was transferred to HybondTM N+ membrane (Amersham
Biosciences) by capillary transfer for 2-2.5 h. After transfer,
membrane filters were briefly rinsed in a transfer buffer and
UV-cross-linked (UV Stratalinker, Stratagene, La Jolla, CA). Membranes
were prehybridized in hybridization solution (UltraHyb, Ambion) for 45 min at 68 °C and hybridized with 50-100 ng/ml
digoxygenin-UTP-labeled RNA overnight at 68 °C. After washing, hybridized filters were probed with anti-digoxygenin antibodies (F(ab)2 fragments) conjugated to alkaline
phosphatase (1:1000 dilution, Roche Molecular Biochemicals). Bands
corresponding to Egr-1 mRNA were detected using disodium
3-(4-methoxyspiro{1,2-dioxetane-3,2-(5'-chloro)tricyclo[3.3.1.13.7]decan}-4-yl
phenyl phosphate (CSPD) chemiluminescent substrate (Roche Molecular
Biochemicals). The antisense RNA probe for bovine Egr-1 was produced by
in vitro transcription using a digoxygenin-labeling kit
(Roche Molecular Biochemicals). A 1.4-kilobase fragment of Egr-1
cDNA ligated into pBluecript (+) vector was obtained from American
Type Tissue collection (Manassas, VA). cDNA template was prepared
by linearization of the plasmid by KpnI digestion upstream
of T7 promoter, and digoxygenin-UTP-RNA probe was synthesized by
reverse polymerase reaction. To provide normalization for RNA loading
and transfer, after-UV-cross-linking blots were stained with 0.1%
methylene blue, 0.5% acetic acid, and bands corresponding to 18 S rRNA
were scanned. Alternatively, blots were double-probed with Egr-1 and
-actin digoxygenin-labeled RNA probes and normalized per -actin mRNA.
Preparation of Nuclear Extracts--
Fibroblasts were grown and
serum-starved as described above. Culture media was then replaced with
fresh serum-free DMEM, and cells were stimulated with either ATP (100 µM) or hypoxia (3% O2) for 2, 4, 6, 12, and
24 h. At the end of the incubation period, cells were washed twice
with ice-cold PBS, scraped off the dishes, and pelleted by
centrifugation at 500 × g for 3 min at 4 °C.
Nuclear extracts were prepared using NE-PERTM extraction
reagents (Piers, Rockford, IL) according to the manufacturer's recommendations. Nuclear extracts and the cytosolic fractions were
snap-frozen in liquid nitrogen in aliquots and stored at 
80 °C.
Protein concentration was determined by the Bradford method using the
Bio-Rad protein assay kit with bovine serum albumin as a standard.
Western Blot Analysis of Nuclear Extracts--
15 µg of
nuclear protein extract was separated by 8% SDS-polyacrylamide gel
electrophoresis and electroblotted to the polyvinylidene difluoride
membrane (Hybond P, Amersham Biosciences) in Tris-glycine buffer (25 mM Tris-base, 0.2 M glycine, 25% (v/v)
methanol, pH 8.3) at 350-400 V for 4 h at 4 °C. After
transfer, membrane filters were stained with Ponceau S (Sigma) to
verify equal protein loading and blocked with 5% nonfat dry milk in
TBS-Tween (50 mM Tris-HCl, pH 7.5, 150 mM NaCl.
0.05% Tween 20) for 1 h at room temperature. Membrane filters
were probed with Egr-1 rabbit polyclonal antibodies (C-19, Santa Cruz
Biotechnology, CA) at a dilution of 1:1000 in TBS-Tween with 5% milk
for 1 h at 4 °C. After intensive washing in TBS-Tween, membrane
filters were incubated with horseradish peroxidase-conjugated
anti-rabbit IgG (1:20,000 dilution, Amersham Biosciences), and
immunoreactive bands were detected by Renaissance ECL detection kit
(PerkinElmer Life Sciences) followed by exposure to Hyperfilm (Amersham Biosciences).
Evaluation of Phospho-ERK1/2--
Cells were plated in 6-well
cell culture plates at a density of 100 ×103 cells/well,
cultured to near confluence, and then serum-starved in 0.1% FBS, DMEM
for 72-96 h. To evaluate the involvement of Gi proteins in
ERK1/2 activation, cells were treated with pertussis toxin (PTx, 100 ng/ml) for 20-24 h. Cells were incubated under either normoxic (21%
O2) or hypoxic (3% O2) conditions in fresh serum-free DMEM in the presence or absence of ATP (100 µM) for 10 min. After incubation, cells were washed twice
with ice-cold PBS and lysed with cold Tris-HCl buffer (40 mM, pH 7.5) containing 0.25 M sucrose, 3 mM EGTA, 3 mM EDTA, 50 µM
-mercaptoethanol, 1 mM phenylmethylsulfonyl fluoride,
and complete protease inhibitors mixture (Calbiochem). Cell lysates
were centrifuged at 7500 × g for 10 min at 4 °C,
and supernatant fractions were collected and stored at 80 °C.
Protein concentration was determined by using the Bio-Rad protein assay
kit with bovine serum albumin as a standard. Samples of total cell
protein (5 µg) were separated by 10% sodium dodecyl
sulfate-polyacrylamide gel electrophoresis, transferred to
polyvinylidene difluoride membranes, and probed with rabbit polyclonal
antibodies against phospho-ERK1/2
(Tyr202/Thr204), 1:1000 dilution (New England
Biolabs, Beverly, MA) overnight at 4 °C. Blots were then washed with
TBS-Tween buffer and incubated with mouse anti-rabbit
peroxidase-conjugated IgG 1:10,000 dilution (Amersham Biosciences) for
1 h at room temperature. Immunoreactive bands were detected by an
ECL detection kit (Renaissance, PerkinElmer Life Sciences) followed by
exposure to Hyperfilm. For detection of the nonphosphorylated form of
ERK, membranes were stripped of the bound antibodies with buffer
containing 62.5 mM Tris-HCl, pH 6.8, 2% SDS, and 100 µM -mercaptoethanol, blocked with 5% nonfat dry milk
in TBS-Tween, and reprobed with rabbit polyclonal antibodies against
total ERK1/2 (1:1000) (New England Biolabs).
Electrophoretic Mobility Shift Assay--
To detect DNA-protein
complexes, nuclear extracts were incubated with 32P-labeled
double-stranded oligonucleotide probe containing two copies of
consensus (5'-GGA TCC AGC TAG GGC GAG CGG TAG CGA-3') or mutant
((5'-GGA TCC AGC TAG GGC GAG CGG TAG CGA-3') Egr-1 binding motifs
(Santa Cruz Biotechnology, CA). Oligonucleotides were 5'-end-labeled by
using [
-32P]ATP (3000Ci/mmol, 10 mCi/ml, PerkinElmer
Life Sciences) with T4 polynucleotide kinase according to the
manufacturer's instructions (New England Biolabs). Labeled
oligonucleotides were purified using Chroma Spin columns-10
(Clontech, Palo Alto, CA). The binding reaction was
carried out for 30 min at room temperature in 20 µl of mixture
containing 10 mM Tris, pH 8.0, 5% glycerol, 50 mM NaCl, 5 mM MgCl2, 10 µM ZnCl2, 2 mM dithiothreitol,
0.1 mM EDTA, 1 µg of poly(dI-dC) (Roche Molecular
Biochemicals), 0.5-1 ng of 32P-labeled DNA (8000 cpm), and
5 µg of nuclear protein. For the competition study, incubation
mixtures containing a 80-fold excess of unlabeled Egr-1
oligonucleotides were used. To test the specificity of Egr-1 DNA
binding complexes, antibody blocking experiments were carried out.
Nuclear extracts were preincubated with 2 µg of specific anti- Egr-1
polyclonal IgG (C-19X, Santa Cruz Biotechnology) for 45 min at room
temperature and then subjected to the binding reaction. Protein-DNA
binding complexes were separated from free DNA probe by native 4%
polyacrylamide gel electrophoresis in 0.5× TBE running buffer (0.045 M Tris borate, 0.001 M EDTA, pH 8.5) at 30 mA
constant current for 3-3.5 h. Gels were vacuum-dried and exposed to
Hyperfilm (Amersham Biosciences) in a cassette with two
intensifying screens at 80 °C for 6-20 h.
Determination of Extracellular ATP--
Total ATP content in the
extracellular media was detected with luciferase-luciferin kit
(Analytical Luminescence Laboratory, Sparks, MD) using a Monolight 3010 luminometer (Analytical Luminescence Laboratory). Cells in
60-cm2 Petri dishes (at a density of 500 × 103 cells/cm2) were exposed to either normoxic
(21% O2) or hypoxic (3% O2 and/or 1%
O2) conditions for the periods of time indicated in Fig. 1. At the beginning of the experiment, medium was drained from the cells
and fresh medium that had been pre-equilibrated with the appropriate
gas mixture was infused onto the cell monolayer. After incubation, 1 ml
of conditioned media was collected into chilled polypropylene tubes
(Sigma) and centrifuged at 12,000 × g for 10 min to
remove any cell debris. Individual 100-µl aliquots were taken and
heated at 95 °C for 1 min, and the luciferin-luciferase assay was
carried out. The sampled luminescence was compared with an ATP standard
curve performed in each individual experiment.
Determination of the Products of Extracellular ATP
Hydrolysis--
Cells were cultured under normoxic (21%
O2) or hypoxic (3% O2) conditions for 3-4
passages and then plated in 12-well culture dishes (40 × 103 cells/well). Confluent cells were growth-arrested for
72 h as described above and washed twice with 1 ml of PBS
(37 °C). To determine the products of [-32P]ATP
hydrolysis (specifically, [-32P]ADP,
[-32P]AMP, and 32Pi), culture
medium was replaced with 0.6 ml of serum-free DMEM (25 °C), and the
reaction was initiated by adding 0.1 mM ATP and 5 µCi of
[-32P]ATP (30 Ci/mmol, 2 mCi/ml, PerkinElmer Life
Sciences). After the indicated periods of time (Fig. 2), 2-µl
aliquots of culture medium were taken and spotted onto
polyethyleneimine-cellulose-coated TLC plates (Sigma) and
chromatographed in 1 M LiCl for 2.5-3 h. Radioactive spots
corresponding to adenine nucleotides and Pi were detected
by autoradiography using Hyperfilm (Amersham Biosciences) and verified
to be at the same position as those of unlabeled adenine nucleotide standards.
Data Analysis--
Density of bands of Western and Northern blot
film images was determined by using NIH Image 1.58 program. Data are
expressed as the average means ± S.E.; n equals the
number of replicates in one experiment or a number of observations in
independent experiments. To evaluate the significance of the obtained
data, analysis of variance between groups of data was performed by the
Student-Newman-Keuls test followed by one-way analysis of variance.
| |
RESULTS |
Hypoxia Increases Extracellular ATP Concentration--
We first
sought to determine whether hypoxia would stimulate ATP release from
fibroblasts (Fig. 1). Quiescent
growth-arrested adventitial fibroblasts were exposed to
pre-equilibrated normoxic or hypoxic media (21, 3, and 1%
O2) for up to 24 h. Zero-minute data represent the
concentration of ATP in the pre-equilibrated media before it was added
to cells. We found that ATP accumulated in the media of unstimulated
fibroblasts even under normoxic conditions (note the 24 h
concentration), consistent with constitutive release of ATP, as has
been observed in human skin fibroblasts and another cell types (32,
33). Exposure of fibroblasts to hypoxic conditions, which stimulated
cell growth (3 and 1% O2) resulted in increased accumulation of ATP in the media, with the highest concentration seen
at 10 min and with persistently increased concentrations observed even
at 24 h (Fig. 1A).
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 1.
Hypoxia increases extracellular ATP
concentration of fibroblasts and endothelial cells. ATP
concentration was measured in the media of cultured adventitial
fibroblasts (A) and endothelial cells (B) that
were exposed to normoxic (21% O2) and hypoxic (3%
O2 and 1% O2) conditions for varying periods
of time. After incubation, 1.5-ml aliquots of media were collected and
prepared for ATP assays with a luciferase/luciferin kit. The results
are expressed as means ± S.E. (*, p < 0.05 compared with normoxic (21% O2) controls). Data illustrate
one representative experiment for each cell type. Similar results have
been obtained in a minimum of three individual experiments.
|
|
Other cells in the vessel wall also may be a source of ATP for
adventitial fibroblasts. Because increased extracellular ATP concentrations have been observed in endothelial cells exposed to
hemodynamic stress and because in small vessels they can be the source
of many paracrine factors that influence fibroblast function (28), we
evaluated the effect of hypoxic exposure on endothelial extracellular
ATP levels. We found that hypoxia caused increases in the concentration
of ATP in the media of endothelial cells that were about 10-fold higher
than in fibroblasts (Fig. 1B). The time course of
accumulation was slightly different from that observed in fibroblasts,
with a peak at 30 min and no differences in extracellular ATP
concentration between normoxic and hypoxic cells at 24 h.
Chronic Hypoxia Decreases Ecto-nucleotidase(s)
Activity--
Extracellular ATP concentrations are known to be tightly
controlled by ecto-nucleotidase(s), which rapidly degrade purine and
pyrimidine nucleotides in the extracellular space. Thus, the concentrations of ATP available for stimulation of purinergic receptors
could be regulated by changes in the degradation of extracellular ATP.
We considered the possibility that hypoxia would decrease
ecto-ATPase(s) activity in either fibroblasts or endothelial cells,
thus slowing the rate of extracellular ATP degradation. Utilizing a
technique that allowed evaluation of products of ATP hydrolysis, we
evaluated the effects of acute and chronic hypoxic exposure on ATP
degradation in cultured cells. We found that brief hypoxic exposure
(<24 h) had no effect on ecto-ATPase activity in either fibroblasts or
endothelial cells (data not shown). However, long term hypoxic exposure
(3% for 
14 days), as might occur in various pathophysiologic states
including hypoxia-induced vascular remodeling, significantly decreased
ecto-ATP-ase(s) activity in both fibroblasts and endothelial cells. As
shown in Fig. 2A, the addition
of [-32P]ATP to adventitial fibroblasts maintained
under normoxic conditions resulted in rapid hydrolysis of
[-32P]ATP to [-32P]ADP,
[-32P]AMP, and 32P with
t1/2(ATP) = 10.6 ± 1.8 min, and
Vo = 25.6 ± 2.10 nmol/min/106
cells (panel A, lanes 1 and 4). In
cells exposed to chronic hypoxia, a delayed rate of ATP hydrolysis was
observed (t1/2(ATP) = 17.6 ± 2.8 min, and
Vo = 19.8 ± 1.95 nmol/min/106
cells in chronically hypoxic cells under hypoxic conditions
(panel A, lanes 2 and 5),
t1/2(ATP) = 22.0 ± 2.2 min, and
Vo = 16.0 ± 1.82 nmol/min/106
cells in chronically hypoxic cells under normoxic conditions (panel A, lanes 3 and 6)). A similar
effect of chronic hypoxia was observed in endothelial cells (normoxia
t1/2(ATP) = 1.75 min ± 0.25 min;
Vo = 77.45 ± 6.35 nmol/min/106
cells versus hypoxia t1/2(ATP) = 7.2 ± 1.6 min; Vo = 35.69 ± 2.93 nmol/min/106 cells (Fig. 2B)).
View larger version (29K):
[in this window]
[in a new window]
|
Fig. 2.
Chronic hypoxia delays the rate of
extracellular ATP hydrolysis by fibroblasts and endothelial cells.
Panel A, analysis of products of [-32P]ATP
hydrolysis by adventitial fibroblasts, chronically (14 days) cultured
under normoxic (minus lanes) or hypoxic (3% O2)
conditions (plus lanes). To determine the rate of
[-32P]ATP hydrolysis, assays were carried out under
both normoxic (21% O2) (lanes 1, 3,
4, 6) or under hypoxic conditions (3%
O2) (lanes 2 and 5) in serum-free
media containing 0.1 mM ATP, 10 mM
MgCl2, 1 mM EGTA, and 5 µCi of
[-32P]ATP. After the indicated periods of time, 2-µl
aliquots of extracellular media were spotted on
polyethyleneimine-cellulose plates, and nucleotides were separated by
thin layer chromatography in 1 M LiCl. Panel B,
analysis of [-32P]ATP hydrolysis by pulmonary artery
endothelial cells, chronically (14 days) cultured under normoxic
(minus lanes) or hypoxic (3% O2) conditions
(plus lanes). All assays were conducted under normoxic
conditions; reaction mixtures and chromatography conditions were
identical to those described for adventitial fibroblasts. Lanes
R on panels A and B represent
[-32P]ATP, incubated in DMEM without cells. Spots,
corresponding to [32-P]ATP,
[32-P]ADP, and [32-P]AMP were
verified using unlabeled adenine nucleotide standards.
|
|
Extracellular ATP Induces DNA Synthesis in Adventitial Fibroblasts
through Both P2Y and P2X Receptors; Hypoxia Augments the
Response--
Although extracellular ATP has been reported to
stimulate proliferation in a wide range of cell types including
endothelial cells, vascular SMC, and fibroblast cell lines, its effect
on vascular adventitial fibroblast proliferation either alone or in the
presence of hypoxia has not been examined. We found that ATP stimulated
increases in thymidine incorporation at doses between 106
and 103 M in quiescent, serum-starved,
adventitial fibroblasts under normoxic conditions (21%O2)
(Fig. 3A). Under hypoxic
conditions (3% O2), thymidine incorporation was increased
even in the absence of ATP, consistent with our previous observations
(5). We found that hypoxia augmented the proliferative effects of ATP
such that there was a shift in the dose-response curve to the left by
about one order of magnitude (Fig. 3A). The effect of ATP on
proliferation in the presence of the ecto-ATPase inhibitor ARL67156 and
an ATP-regenerating system (creatine kinase, phosphocreatine, and
myokinase) was also evaluated to more carefully delineate the
concentrations at which ATP stimulates [3H]thymidine
incorporation since considerable degradation of ATP likely takes place
over the 24-h time period required for the assessment of DNA synthesis.
In the presence of ARL67156, the concentration-dependent
curve was shifted significantly to the left and demonstrated that DNA
synthesis is initiated at concentrations as low as
107-106 M. A similar shift in
the concentration-dependent curve was also observed in the
presence of ATP-regenerating system (Fig. 3B).
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 3.
Effect of nucleotides, growth factors, and
hypoxia on DNA synthesis in adventitial fibroblasts. Quiescent
adventitial fibroblasts (72 h, 0.1% FBS-DMEM) were stimulated with the
agonist of interest in the presence of 1 µCi of
[3H]thymidine for 24 h, and incorporated
radioactivity was determined as described under "Experimental
Procedures." Panel A, effect of increasing concentrations
of ATP on [3H]thymidine incorporation under normoxic
(21% O2) or hypoxic (3% O2) conditions.
Panel B, ATP-induced [3H]thymidine incorporation in
the presence of ecto-ATPase inhibitor ARL67156 (300 µM)
or ATP-regeneration system (0.1 mg/ml creatine kinase, 4 mM
phosphocreatine, and 0.025 mg/ml myokinase). Panel C, effect
of hypoxia (3% O2), ATP, UTP, ADPS, BzATP (100 µM), MeSATP (300 µM), AMP-CPP
(,meATP) (100 and 1000 µM), or adenosine
(Ado, 100 µM) on [3H]thymidine
incorporation in quiescent fibroblasts. Panel D, effect of
purified peptide mitogens platelet-derived growth factor
(PDGF; 20 ng/ml), epidermal growth factor (EGF;
10 ng/ml), insulin-like growth factor-I (IGF-I; 100 ng/ml)
on [3H]thymidine incorporation under normoxic conditions
and in the presence of ATP (100 µM) or hypoxia (3%
O2). The data represent the means ± S.E. from three
to six independent experiments conducted on distinct cell populations.
*, p < 0.05 compared with nonstimulated control.
|
|
Because the concentration range of ATP found to stimulate DNA synthesis
could activate both P2Y and P2X receptors (11, 34, 35), we evaluated
the effects of selective P2Y and P2X receptor agonists on DNA
synthesis. We found that the P2Y receptor agonists UTP, ADPS, and
the P2Y/P2X agonist MeSATP stimulated increases in
thymidine incorporation which were nearly equivalent to those observed
with ATP at doses causing a maximal effect (Fig. 3C). In addition the
P2X7 selective agonist BzATP caused a less, but still significant
increase in DNA synthesis. The P2X1,3 receptor agonist AMP-CPP,
stimulated DNA synthesis only when used at 103
M. Adenosine, on the other hand, had no significant effect
on DNA synthesis, suggesting that P2Y and P2X purinoceptors, but not P1
purinoceptors, were likely involved in ATP-stimulated adventitial fibroblast proliferation. In contrast to the effects exerted by hypoxia and ATP (Fig. 3A), we observed synergistic
interactions between ATP and purified peptide mitogens and between
hypoxia and purified peptide mitogens (Fig. 3D).
Purinergic Receptor Antagonists and Apyrase Inhibit ATP- and
Hypoxia-induced Cell Proliferation--
Because extracellular ATP
concentrations were increased in response to hypoxia and an
autocrine/paracrine role of ATP has been demonstrated in another cell
types, we tested the possibility that ATP may mediate, at least in
part, the proliferative effect of hypoxia. We evaluated the effect of
P2 receptor antagonists as well as apyrase on hypoxia-induced DNA
synthesis. We found that preincubation with the P2 receptor antagonists
suramin (100 µM, 30 min), Cibacron blue 3GA (100 µM, 30 min), or PPADS (100 µM, 60 min)
attenuated [3H]thymidine incorporation in response to ATP
and hypoxia (3% O2) and in response to the combined
stimulation of cells with ATP and hypoxia (Fig.
4A). The inhibition of DNA
synthesis was not due to cytotoxic effects of suramin, Cibacron blue
3GA, or PPADS, because no change in lactate dehydrogenase levels or the
ability to exclude trypan blue was observed (data not shown). In
addition we found that apyrase, which specifically hydrolyzes ATP and
ADP in the incubation media, reduced [3H]thymidine
incorporation in response to ATP and hypoxia by ~50%, again
suggesting that extracellular ATP contributes to hypoxia-induced DNA
synthesis in adventitial fibroblasts (Fig. 4B).
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 4.
Purinergic antagonists and apyrase attenuate
ATP- and hypoxia-induced DNA synthesis in adventitial fibroblasts.
Cells were cultured under standard conditions and growth-arrested for
72 h. Quiescent cells were preincubated with suramin (100 µM), Cibacron blue 3GA (100 µM) for 30 min,
PPADS (50 µM) for 60 min (panel A), or with
apyrase (5 units/ml) for 2 h (panel B) followed by
stimulation with ATP (100 µM) or hypoxia (3%
O2) in the presence of 1µCi of
[3H]thymidine/well for 24 h. At the end of the
incubation, [3H]thymidine incorporation was assessed as
described "Experimental Procedures." The quantitative data are
expressed as percent increase over basal condition. The basal amount of
[3H]thymidine incorporation was 1225 ± 140 cpm.
Data are the means ± S.E. (S.E.) from three to six independent
experiments. *, p < 0.05 compared with nonstimulated
control (panel A); **, p < 0.05 compared
with either ATP- or hypoxia-induced response (panel
B).
|
|
ATP and Hypoxia Induce Egr-1 mRNA Expression in Adventitial
Fibroblasts--
Egr-1 is a transcription factor that is commonly
induced under stress conditions and that has been demonstrated to be
critical in stress-induced proliferative responses. Because hypoxia
increased extracellular ATP, we tested the possibility that Egr-1 would be a downstream signaling mediator induced in response to hypoxia and
ATP in adventitial fibroblasts. Using Northern blot analyses, we found
that both ATP (10 µM) and hypoxia (3% O2)
markedly induced Egr-1 mRNA and that the ATP-mediated response was
augmented under hypoxic conditions (Fig.
5). We also found that the pyrimidine nucleotide UTP, whose extracellular concentration can be elevated in
response to acute vascular stress, induced Egr-1 mRNA with similar
efficiency to ATP. Factors including platelet-derived growth factor
(PDGF), FBS, and phorbol 12-myristate 13-acetate (PMA), all previously shown to stimulate Egr-1, were used as
positive controls and for qualitative comparison purposes.
View larger version (46K):
[in this window]
[in a new window]
|
Fig. 5.
ATP and hypoxia induce Egr-1 mRNA
expression in adventitial fibroblasts. Cells were plated at a
density of 10 × 103/cm2, grown to 80%
confluence, and serum-starved in 0.1% FBS/DMEM for 72 h. 15 µg
of total cellular RNA, obtained from either control cells or cells
stimulated with either 10 µM ATP, 10 µM
UTP, hypoxia (3% O2), hypoxia (3% O2) + 10 µM ATP, 20 ng/ml platelet-derived growth factor
(PDGF), 10% FBS, or 10 ng/ml phorbol 12-myristate
13-acetate (PMA) for 30 min were analyzed by Northern blot
using digoxygenin-RNA probe for Egr-1 as described under
"Experimental Procedures." To verify equivalent RNA loading and
transfer, before hybridization membranes were stained with methylene
blue. Experiments with the same agonists were repeated at least six
times, always giving similar results.
|
|
Next, we compared time- and dose-dependent profiles for
both ATP and hypoxia on Egr-1 mRNA expression. ATP was shown to
stimulate Egr-1 mRNA in a time- and dose-dependent
fashion. A maximal induction of Egr-1 mRNA levels of ~8.5-fold
over basal was observed at 100 µM ATP (Fig.
6, panel A). At this
concentration Egr-1 mRNA was induced within 15 min, peaked at
1 h, and returned to basal level by 4 h. Hypoxic exposure
alone also stimulated an increase in Egr-1 mRNA expression in
quiescent serum-starved adventitial fibroblasts. A significant increase
was seen at O2 concentrations of 10% with a maximal effect
observed at 1% (Fig. 6, panel B). Using both 1 and 3%
oxygen concentrations, we found the time course and magnitude of
activation by hypoxia to be remarkably similar to that of ATP with a
7-fold induction observed within 30 min and a decline to base line by
6 h.
View larger version (47K):
[in this window]
[in a new window]
|
Fig. 6.
ATP and hypoxia induce Egr-1 mRNA
expression in adventitial fibroblasts in a time- and
dose-dependent manner. Panel A, effect of ATP on
Egr-1 mRNA expression. Growth-arrested cells were stimulated with
100 µM ATP for the indicated periods of time
(left) or stimulated with
109-103 M ATP for 1 h
(right). Panel B, effect of hypoxia on Egr-1 mRNA
expression. Growth-arrested cells were exposed to hypoxia (3%
O2) for 15 min to 6 h (left) or exposed to
the indicated oxygen concentrations for 1 h
(right). After treatment, total cellular RNA was
extracted, and 15 µg was analyzed by Northern blot using Egr-1
digoxygenin-RNA probe. Egr-1 mRNA expression was normalized to
-actin mRNA or to 18 S rRNA and expressed relative to the Egr-1
mRNA level in nonstimulated cells. Data represent the average ± S.E. of three independent experiments performed on separate cell
populations.
|
|
P2 Purinergic Receptors, Gi Proteins, and ERK1/2
Mediate Both ATP- and Hypoxia-induced Egr-1 mRNA
Expression--
Because we have previously shown Gi and
ERK1/2 to be important in hypoxia-induced proliferation and because P2
receptor antagonists attenuated hypoxia-induced proliferation
(above), we studied the effects of PTx (known to selectively
inhibit Gi/o-mediated signaling), a specific
ERK-activating kinase (MEK-1) inhibitor (PD98059), P2 receptor
antagonists (suramin, Cibacron blue, PPADS), and apyrase on hypoxia-
and ATP-induced Egr-1 mRNA levels. Northern blot analysis demonstrated that preincubation with pertussis toxin (100 ng/ml, 20-24
h) attenuated ATP- and hypoxia-induced Egr-1 mRNA expression by 40 and 48%, respectively (Fig. 7),
indicating that activation of Gi/o as well as
PTx-insensitive G proteins may underlie ATP- and hypoxia-induced
Egr-1 expression (panel A). The ERK-activating kinase
(MEK-1) inhibitor PD98059 (10 µM, 1 h) attenuated
ATP and hypoxia-induced responses by 63 and 58%, respectively.
Preincubation with suramin (panel A), Cibacron blue 3GA
(data not shown), and apyrase (panel C) inhibited both ATP-
and hypoxia-induced Egr-1 expression. PPADS, which is thought to act
preferentially on P2X receptors and only on some subtypes of P2Y
receptors, was less effective in blocking hypoxia-induced Egr-1
expression (panel B). Collectively, these data suggest that
P2 receptor activation by the release of endogenous ATP may occur under
hypoxic conditions.
View larger version (50K):
[in this window]
[in a new window]
|
Fig. 7.
Purinergic receptors, Gi proteins
and ERK1/2 mediate both ATP- and hypoxia-induced Egr-1 mRNA
expression. Growth-arrested adventitial fibroblasts were treated
with either PTx (100 ng/ml) for 20-24 h, PD 98059 (10 µM) for 1 h, suramin (100 µM) for 30 min at 37 °C, PPADS (100 µM) for 60 min, or with
apyrase (5 units/ml) for 2 h. After treatment, cells were
stimulated with ATP (100 µM) or hypoxia (3%
O2) for 1 h and then harvested for total RNA
extraction. 15 µg of RNA was subjected to Northern blot analysis, and
Egr-1 was identified using digoxygenin-RNA probe (panels
A-C). Before hybridization membranes were stained with
methylene blue to verify equality of RNA loading and transfer. Bands
corresponding to Egr-1 were normalized to 18 S rRNA, and results were
expressed relative to Egr-1 mRNA in nonstimulated cells. Data,
shown in panel D represent the average ± S.E. from
four to six independent experiments performed in four separate cell
populations.
|
|
To further examine the possibility that both ATP and hypoxia utilize
Gi-initiated ERK1/2 activation to control proliferation and Egr-1 expression, we evaluated the effects of pertussis toxin on
ATP- and hypoxia-induced ERK1/2 phosphorylation (Fig.
8, A and B). As
expected, we found that both ATP and hypoxia induced ERK1/2 activation,
and the combination of exogenous ATP and hypoxia enhanced ERK1/2
phosphorylation in an additive manner. Pertussis toxin partially
inhibited the hypoxia-induced response as well as the ATP-induced
response under normoxic and hypoxic conditions. In addition we found
that suramin and Cibacron blue 3GA significantly attenuated ERK1/2
phosphorylation in response to ATP and hypoxia, but PPADS had less
effect on ATP-induced ERK phosphorylation (Fig. 8C).
Finally, we found that apyrase almost completely attenuated ATP-induced
ERK1/2 phosphorylation and attenuated the hypoxic response at least by
30% (Fig. 8, E and D). Collectively these findings suggest that ATP and hypoxia activate Egr-1 expression and
proliferation through both Gi-dependent and
-independent activation of ERK1/2 and that ATP release and purinoceptor
activation contribute to hypoxia-initiated responses.
View larger version (42K):
[in this window]
[in a new window]
|
Fig. 8.
Hypoxia and ATP induce ERK1/2 phosphorylation
in Gi protein-dependent manner.
Adventitial fibroblasts were grown to 80% confluence and
serum-deprived for 72-94 h in DMEM, 0.1% FBS. Panel A,
after preincubation with PTx (100 ng/ml, 24h) cells were stimulated
with either ATP (100 µM) or with hypoxia (3%
O2) for 10 min. After stimulation, total cell lysates were
prepared. Panels C and D, quiescent cells were
pretreated with purinergic antagonists (suramin, Cibacron blue 3GA,
PPADS) or with apyrase as described in Fig. 7 and stimulated with
either ATP (100 µM) or with hypoxia (3% O2)
for 10 min. Five µg of total cell protein was subjected to 10%
SDS-polyacrylamide gel electrophoresis and Western blot analysis with
antibodies against phosphorylated-ERK1/2 (phospho-Tyr-202/Thr-204) and
nonphosphorylated ERK. Panels B and E,
phospho-ERK bands were normalized to nonphosphorylated ERK bands, and
the quantitative data are expressed as fold increases over basal
stimulation. Data represent average ± S.E. from four to six
independent experiments. Cont, control.
|
|
Hypoxia and ATP Increase Egr-1 Protein Level and DNA Binding
Activity--
To determine whether stimulation of fibroblasts with ATP
or hypoxia resulted in increased Egr-1 protein levels, Western blot analysis of nuclear extracts isolated from control cells as well as
cells exposed to either ATP or hypoxia for 0, 2, 4, 6, and 24 h
were performed (Fig. 9, panel
A). Immunoblotting of nuclear fractions with Egr-1 antibodies
revealed protein bands with an apparent molecular mass of 82 kDa, which
corresponds to Egr-1. In the cytosolic fraction only trace amounts of
Egr-1 protein were detected (data not shown). Responses varied over
time with hypoxia causing a more prolonged increase in Egr-1 protein
levels than ATP. In nuclear extracts obtained from ATP- and
hypoxia-treated cells, an additional protein band appeared after 2 h of stimulation, which became more visible at 4 and 6 h,
suggesting that along with increased protein expression,
posttranslational modification(s) of Egr-1 may occur.
View larger version (87K):
[in this window]
[in a new window]
|
Fig. 9.
ATP and hypoxia induce Egr-1
protein expression and enhance its DNA binding activity.
Panel A, Western blot analysis of Egr-1 protein in nuclear
extracts of adventitial fibroblasts. Preconfluent cells were
growth-arrested in 0.1% FBS-DMEM for 72 h and then exposed to 100 µM ATP or hypoxia (3% O2) in DMEM for the
indicated periods of time. At the end of each incubation, the nuclear
fraction was isolated, and 20 µg of nuclear protein was subjected to
immunoblotting analysis with Egr-1 (C-19) rabbit polyclonal antibodies
(see "Experimental Procedures"). Phorbol 12-myristate 13-acetate
(PMA; 10 nM) was used as a positive control.
Equality of protein loading was confirmed by staining of the membrane
with Ponseau solution for each experiment. Panels B-D,
identification of Egr-1 DNA binding activity in nuclear extracts of
adventitial fibroblasts stimulated with ATP or hypoxia. Growth-arrested
adventitial fibroblasts were exposed to ATP (100 µM,
2 h) or hypoxia (3% O2, 4 h) in fresh DMEM. At
the end of incubation, nuclear extracts were isolated, and 5 µg of
protein were subjected to EMSA. Panel B, EMSA of nuclear
extract with 32P-labeled DNA probe containing Egr-1 binding
consensus sequence in the absence () or presence (+) of Egr-1
antibodies (Egr-1 Ab). Panel C, the same nuclear
extracts were analyzed in the presence of excess of unlabelled Egr-1
binding DNA probe. Panel D, the same nuclear extracts were
investigated for the ability to interact with 32P-labeled
DNA probe containing mutant Egr-1 binding sequence. Presented data were
reproducible in minimum four individual experiments. *, indicates
unidentified specific binding complex. **, indicates nonspecific
binding. cont, control.
|
|
To demonstrate that increased Egr-1 protein expression in the nucleus
was associated with a specific interaction with its DNA recognition
site, we performed electrophoretic mobility supershift assays (EMSAs).
Based on the immunoblotting data (Fig. 9, panel A), we chose
the stimulation condition that had resulted in the highest Egr-1
protein levels (100 µM ATP for 2 h and hypoxia (3% O2) for 4 h). Again, nuclear extracts obtained from
phorbol 12-myristate 13-acetate-treated cells (PMA; 10 nM, 4 h) were used to provide positive controls for
the assay conditions. EMSAs of nuclear extracts isolated from ATP- and
hypoxia-stimulated cells revealed three visible DNA-protein complexes
(Fig. 9, panel B, minus lanes). Preincubation of
nuclear extracts with Egr-1 antibodies (Fig. 9 panel B,
plus lanes) demonstrated the capability of protein from the
upper band to form complexes with reduced electrophoretic mobility,
indicating that protein in these DNA binding complexes corresponds to
Egr-1. To further evaluate the specificity of the DNA-protein
interaction, we carried out binding reactions in the presence of an
80-fold excess of unlabeled oligonucleotides containing a consensus
Egr-1 binding sequence. We found that radioactivity in the Egr-1 band
as well as in one of the lower bands was markedly attenuated (Fig. 9,
panel C, Egr-1-bound and asterisk
(*)). Because the protein comprising one of the lower bands is observed
in equal amounts in control and in stimulated cells, it is possibly a
constitutively expressed nuclear factor enabling and/or specifically
interacting with Egr-1 DNA binding sites. Finally, we demonstrated that
32P-labeled oligonucleotides containing mutant Egr-1
binding sequence was ineffective in interacting with either Egr-1 or
with a constitutively expressed nuclear protein, suggesting that the
lower band (Fig. 9, panel D, two asterisks (**))
represents nonspecific DNA binding complexes.
| |
DISCUSSION |
Activation and proliferation of fibroblasts under
conditions of reduced oxygen tension in the local environment is
thought to play a critical role in a variety of fibrotic conditions
(1-4). However, it remains unclear as to how hypoxia exerts its
effects within the local environment and/or intracellularly to effect changes in proliferation. Previous investigations demonstrated that
G-protein activation was a critical upstream event in hypoxia-induced proliferation and that subsequent signaling through mitogen-activated protein kinase, phosphatidylinositol 3-kinase, and protein kinase C
pathways was involved in the proliferative response. Other studies demonstrated that extracellular ATP, acting through G protein-coupled receptors, could potentially modulate mitogen-activated protein kinase
signaling and proliferation in SMC (7, 10, 36-38). The present study
was therefore undertaken to address the hypothesis that hypoxia would
stimulate activation of a local purinergic signaling network whereby
increases in cellular ATP release, impaired degradation of released
ATP, and stimulation of P2 purinoceptors would operate to either
control or modulate fibroblast proliferation under hypoxic conditions
(Fig. 10). Our data demonstrate that
indeed hypoxia can increase extracellular ATP through enhanced release and impaired degradation (Figs. 1 and 2). Furthermore, ATP exerts effects on cell signaling and proliferation that are similar to those
effected by hypoxia, and the effects of hypoxia and ATP can be
abrogated by reducing extracellular ATP and/or blocking G-protein-coupled signaling (Fig. 4). Thus, the findings of the present
study suggest that ATP is a released by fibroblasts (and by endothelial
cells) under hypoxic conditions and that activation of P2 purinoceptors
is involved in the regulation of DNA synthesis under hypoxic
conditions.
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 10.
A scheme showing how ATP might act as
an autocrine/paracrine mediator of hypoxia-induced fibroblast
proliferation. Hypoxia increases extracellular ATP level by at
least two mechanisms; first, through the release of intracellular ATP
across the plasma membrane, and second, by down-regulating
ecto-nucleotidases (E1 (ecto-ATPase), E2 (ecto-ADPase), and E3
(ecto-5'-nucleotidase), thus decreasing the rate of ATP
hydrolysis. Both ATP and hypoxia stimulate adventitial fibroblast
proliferation through G protein-coupled P2Y receptors, ionotropic P2X
receptors, and subsequent ERK1/2 and Egr-1-dependent
signaling pathways. Additional possible mitogenic pathways are
accompanied by question marks. Ado, adenosine.
|
|
Investigations in other cell systems also demonstrate that
extracellular nucleotides can function as autocrine-paracrine mediators of a number of cell responses and that their appearance in the extracellular microenvironment can result from cell lysis, exocytosis of nucleotide-containing granules, or efflux through membrane transport
proteins (7, 36). Recent studies indicate that relatively large amounts
of ATP as well as UTP are released by mechanical stimulation
(e.g. shear stress, hypotonic swelling, or stretch) of
epithelial cells, endothelial cells, smooth muscle cells, glial cells,
fibroblasts, and hepatocytes and that the presence of these nucleotides
in the extracellular medium promotes activation of P2 purinoceptors
(13, 15, 27, 38-43). Our data demonstrate that hypoxia induces release
of ATP into the extracellular milieu from both adventitial fibroblasts
and endothelial cells and that the increase in ATP concentrations is
not the result of cell lysis. Although the measured ATP concentrations
in the conditioned media are similar to those previously reported in other stimulated cell systems, they seem to be lower than the threshold
for purinoceptor activation noted in this and other studies (32).
However, as has been mentioned by many others and elegantly shown in
recent studies, local ATP concentrations near the cell surface and,
accordingly, near the purinoceptors are much higher than those measured
in the conditioned media (44). In addition, our data suggest a
constitutive basal release of ATP by fibroblasts into the extracellular
space, as has been recently demonstrated for the other cell types, that
could act to modulate responses to other local stimuli even in the
absence of hypoxia (16, 18, 32). Thus, our data indicate that despite
ongoing ATP hydrolysis by ecto-nucleotidases, acute hypoxia induces
significant increases in extracellular ATP, supporting the possibility
of autocrine-induced changes in purinoceptor activation and subsequent cell signaling. In addition and in a broader physiologic context, they
also support the possibility that other sources of ATP such as might be
derived from the endothelial cells or from the sympathetic nerves in a
vascular adventitia (9, 11, 27, 45) could also contribute to high local
concentrations of ATP and, thus, cooperate in adventitial fibroblast
activation and proliferation under hypoxic conditions.
An intriguing question regarding ATP release in response to hypoxia is
whether ATP transport through the plasma membrane in vascular cells is
coupled to the mechanism of oxygen sensing. An essential role for ATP
acting as a sensory mediator has been proposed for several cell types.
It has been demonstrated that in the rat carotid body, co-release of
ATP and acetylcholine underlies the mechanism of chemoreception (46).
Autocrine ATP release is involved in shear stress-induced c-Jun
NH2-terminal kinase stimulation in SMC (14), in hydrostatic
control of rabbit urinary bladder tone (47), in mechanical
stress-mediated calcium mobilization in polarized airway epithelial
cells (15), and in cell volume regulation (48). Thus, these
observations together with our data raise the possibility that
extracellular nucleotides play an important role in transducing or
modulating a variety of hypoxia-induced responses in vascular cells.
The metabolism of extracellular ATP by
ecto-nucleotidase(s)/ecto-ATPases plays an important role in the
regulation of purinergic signaling. Recent observations of Moser
et al. (49) provide evidence that the antiproliferative
action of angiostatin is related to the inhibition of endothelial cell
surface, F1- F0-ATP synthase. In
addition, it has also been demonstrated that certain pathophysiologic conditions, especially those associated with endothelial cell activation, can actually decrease ATP-diphosphohydrolase activity (50).
In support of the possibility that hypoxia could effect this enzyme
system are observations that ATP-diphosphohydrolase activity in COS-7
cells is decreased after exposure to H2O2 and oxygen radicals (51). However, the effects of acute or chronic hypoxia
on ecto-nucleotidase activity have not been previously reported. Our
data provide new evidence that chronic hypoxic exposure of pulmonary
artery adventitial fibroblasts and endothelial cells can decrease the
rate of extracellular ATP hydrolysis, suggesting that regulation of
ecto-nucleotidase activity by hypoxia could be an additional mechanism
that contributes to the concentration of various extracellular
nucleotides levels in the vascular wall. Delayed ATP hydrolysis under
hypoxic conditions could cause changes in the ratio between
extracellular concentrations of purine nucleotides. Because these
different phosphorylation products (ATP, ADP, AMP, and adenosine)
activate different purinoceptor subtypes, our observations suggest that
hypoxia-induced dysregulation of nucleotidases could lead to diverse
changes in intracellular signaling and responses. Other extracellular
enzymes such as ecto-nucleoside diphospho-kinases also can contribute
to metabolism of extracellular nucleotides, and future studies need
to evaluate the effects of hypoxia on these enzymes.
Little is known of the receptors through which ATP stimulates
proliferation in fibroblasts (33). In vascular SMC ATP-induced proliferation is thought to be mediated largely through P2Y receptors acting through ERK1/2 and phosphatidylinositol 3-kinase signaling pathways (11). Our studies demonstrating that DNA synthesis occurs at
ATP concentrations of 107-106
M in the presence of ARL6715 and an ATP-regenerating
system, that UTP and ADPS stimulate proliferation, and that suramin
and Cibacron blue 3GA inhibit ATP-induced fibroblast proliferation also
suggest that P2Y receptors can initiate DNA synthesis in fibroblasts
(10, 11). However, both agonist and antagonist data also suggest that
P2X receptors can also mediate proliferative responses. The involvement
of P2X receptors in the mitogenic response is in contrast to vascular
SMC, however, consistent with stimulation of proliferation through P2X
receptors in osteoblasts (17). The data are also in agreement with
previous reports demonstrating the presence of different subtypes of
P2Y (P2Y1, P2Y2, P2Y4, P2Y6) and P2X (P2X3, P2X4, P2X7) receptors on
human and rat fibroblasts (33, 34). The relative contribution of P2Y
versus P2X receptor activation in hypoxia- and ATP-induced
fibroblasts proliferation probably will be dependent on local ATP
concentration and on relative expression level of P2Y and P2X
receptors, which likely vary depending on the activation state of the fibroblast.
Little information exists regarding the post-receptor signaling
pathways or transcription factors that are involved in ATP- or
hypoxia-mediated fibroblast proliferation. Because many of the same
conditions (oxidative stress, mechanical stretch, osmotic swelling)
previously shown to cause increases in extracellular ATP also induce
Egr-1 expression in cells and because Egr-1 has been linked with cell
proliferation, we evaluated the effect of ATP and hypoxia on Egr-1
expression in adventitial fibroblasts (24, 52, 53). ATP and hypoxia
were both potent inducers of Egr-1 mRNA expression and exerted
their effects over very similar time courses (Figs. 5 and 6). In
addition, both ATP and hypoxia increased the ability of Egr-1 to
interact with a specific DNA recognition site (Fig. 9). EMSAs revealed
two proteins, which specifically interact with consensus, but not with
mutant, Egr-1 DNA probe. Supershift with specific Egr-1 antibodies
revealed the inducible protein to be Egr-1. A second protein is not
inducible and may represent a constitutively expressed transcription
factor or co-regulator of Egr-1-DNA interaction. One candidate may be Sp-1, which has a DNA recognition site overlapping the site for Egr-1,
which can be expressed in nuclei under basal conditions (23). Our
observations are consistent with previous studies demonstrating that
ATP can induce c-Fos, c-Jun, Jun-B, and Egr-1 in rat mesangial cells
(54) and that hypoxia can induce Egr-1 in SMC and macrophages (55).
Importantly, we found that P2 purinoceptor antagonists attenuated
hypoxia-induced Egr-1 expression. Thus, Egr-1 appears to be an
important hypoxia-regulated transcription factor whose activity can be
regulated by P2 purinoceptors in vascular fibroblasts.
To further evaluate the role of ATP in hypoxia-induced growth, we
evaluated signaling events upstream of Egr-1 and proliferation induced
by ATP and hypoxia. We found that PTx inhibited 40-50% of both ATP-
and hypoxia-induced increases in Egr-1 mRNA expression and
proliferation, indicating that Gi/o proteins contribute
nearly equally to hypoxia and ATP-induced responses in adventitial
fibroblasts (Figs. 7 and 8). The role of Gi/o proteins
in hypoxia-induced proliferation previously has been established by us,
and the fact that purinergic receptors can mediate some of their
effects through PTx-sensitive Gi/o pathways has also
been established (5). The data reported here continue to support the
idea that Gi/o signaling is an important component of
hypoxia-induced signaling and proliferation in fibroblasts. Of interest
are observations in LLC-PK1 renal cells that Egr-1 up-regulates
transcription of the Gi gene (52), which suggests that
it will be important to determine whether hypoxia creates a positive
regulatory loop mediated through ATP to regulate Gi gene
expression in adventitial fibroblasts. P2 receptor antagonists and
apyrase, again supporting a role for ATP release and purinoceptor
activation in fibroblast growth, also attenuated hypoxia-induced
increases in Egr-1 expression and ERK1/2 phosphorylation. In addition,
our observations that ATP and hypoxia both act synergistically with
peptide growth factors operating through receptor tyrosine kinases also
suggest that they use similar signaling pathways and are consistent
with previous findings showing that G protein-coupled receptor and
receptor tyrosine kinases can act synergistically (56-58).
Collectively, these observations are consistent with the idea that
"constitutive" levels of ATP within the cellular environment might
act as key determinants of the set point of many signal transduction
pathways (18, 32).
Our data suggest that signaling pathways other than those mediated
directly by Gi may also contribute to Egr-1 mRNA
expression and cell proliferation in response to ATP and hypoxia in
adventitial fibroblasts. Other have demonstrated that P2Y receptor
activation can be associated with stimulation of PTx-insensitive
G-protein coupled signaling, resulting in phospholipase C activation
(7, 36, 59-61). That Gq and phospholipase C may be
important in hypoxia-induced proliferation is supported by observations
showing that phosphatidylinositol-specific phospholipase C is involved in hypoxia-induced DNA synthesis in bovine pulmonary artery adventitial fibroblasts (62). This would suggest that Gq-coupled
phospholipase C/protein kinase C/Ca2+-signaling cascades
may also be associated with hypoxia and ATP-induced Egr-1 expression
and cell proliferation. In support of this are observations by Yan
et al. (55) and by Lo et al. (63), who found that
protein kinase C2 and ERK1/2 are necessary for hypoxia-induced Egr-1
expression, although upstream signals were not defined. Phosphatidylinositol 3-kinase/Akt signaling pathways have also been
shown to be stimulated by hypoxia and ATP (64-66). It is also possible
that P2X receptor stimulation under hypoxic conditions leads to
increases in intracellular Ca2+ concentrations and
stimulation of signaling pathways distinct from ERK1/2 and Egr-1. In
support of this are our findings demonstrating that PPADS attenuates
hypoxia-induced proliferation but has little effect on ERK1/2 or Egr-1
activation. Future experiments will need to define the interactions
between distinct signaling pathways, which are activated by hypoxia,
and ATP that operate cooperatively to drive proliferative responses.
| |
ACKNOWLEDGEMENTS |
We are indebted to Viktoriya Marusyk for
excellent technical assistance and Dr. Aftab Ahmad for helpful advice
in the EMSA experiments. We thank Stephen Hofmeister and Marcia McGowan
for help in the preparation of the manuscript. We also thank Dr. R. Nemenoff for critical comments.
| |
FOOTNOTES |
*
This work was supported by Specialized Center for Research
in Atherosclerosis Grant HL 56481 and National Institutes of Health Grant HL 14985. Preliminary results were presented at the American Thoracic Society Annual Meeting, May 18-23, 2001, San Francisco and
the 41st American Society for Cell Biology Annual
TOP
ABSTRACT
INTRODUCTION
