| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
and NF-
B in
perinatal lung epithelium requires glutathione
biosynthesis
Oxygen Signalling Group, Tayside Institute of Child Health, Ninewells Hospital and Medical School, University of Dundee, Dundee DD1 9SY, United Kingdom
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
To test the genetic
capacity of the perinatal lung to respond to O2 shifts that
coincide with the first respiratory movements, rat fetal alveolar type
II (fATII) epithelial cells were cultured at fetal distal lung
PO2 (23 Torr) and then exposed to postnatal (23 
76 Torr; mild hyperoxic shift), moderate (23 152 Torr; moderate hyperoxic shift), or severe (23 722 Torr; severe hyperoxic shift) oxygenation. Nuclear abundance and
consensus binding characteristics of hypoxia-inducible factor
(HIF)-1
and nuclear factor (NF)-
B (Rel A/p65) plus glutathione
biosynthetic capacity were determined. Maximal HIF-1 activation at
23 Torr was sustained over the postnatal shift in (
)
PO2 and was elevated in vivo
throughout late gestation. NF-B was activated by the acute postnatal
PO2 in fATII cells, becoming maximal with moderate and severe oxygenation in vitro and within 6 h of
birth in vivo, declining thereafter. fATII cell and whole lung
glutathione and GSH-to-GSSG ratio increased fourfold with a postnatal
PO2 and were matched by threefold
activity increases in 
-glutamylcysteine synthetase and glutathione
synthase. GSH concentration depletion by
L-buthionine-(S,R)-sulfoximine abrogated both
HIF-1 and NF-B activation, with HIF-1 showing a heightened
sensitivity to GSH concentration. We conclude that O2-linked genetic regulation in perinatal lung epithelium
is responsive to developmental changes in glutathione biosynthetic capacity.
hypoxia-inducible factor-1; nuclear factor-B; lung
development; L-buthionine-(S,R)-sulfoximine; transcription factor; antioxidant; bronchopulmonary
dysplasia
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
PATHWAYS LINKED TO THE GENERATION of reactive oxygen
species (ROS) are believed to constitute a vital component of cellular O2 signaling mechanisms that integrate the expression of
genes involved in energy production, O2 transfer, cellular
differentiation, and free radical scavenging with prevailing
PO2 (3, 8, 23). Thus the intensity of
any form of O2-linked signal is governed by 1) the
direction of the shift in PO2
(PO2; i.e., a relative hypoxia or
hyperoxia), 2) the magnitude of the PO2, 3) the cellular
capacity for generating ROS, and 4) the expression of
antioxidants. It follows that both the form and magnitude of response
to any O2-based signal depends greatly on whether or not an
effective means for buffering the production of ROS is in place.
Lung maturation throughout gestation and in the postnatal period each occur within widely different PO2 values. In utero, saccularization of the lung and the functional differentiation of the epithelial lining proceed at a PO2 that lies continuously between 23 and 30 Torr, the O2 transfer capacity of the umbilical vein (13). During parturition, Na+-driven fluid absorption drains the fluid support within the lung, ventilation begins, and luminal end-tidal PO2 rapidly rises to stabilize at within a range of 70-100 Torr. The successful transition from placental to pulmonary gas exchange therefore incurs a four- to fivefold relative hyperoxic shift at the epithelial lining of the distal lung that superimposes on hormonally regulated perinatal developmental events such as the expression and production of surfactant (1, 2, 30), expression of ion-transport pathways and their components (24), expansion of gas-exchange surfaces, and alveolarization (45). Despite a late-gestational increase in expressed antioxidant enzymes, the full complement of antioxidant defenses in the fetal lung remains approximately threefold lower than that in the neonatal lung and sixfold lower than that in the adult lung (5, 16). It would therefore appear that the epithelial lining of the perinatal lung possesses a depressed capacity for buffering the production of ROS and, as such, may be acutely responsive to fluctuations in O2 availability. The foundation is laid for an important redox-linked signaling event unique to the period immediately after the first breath, which may modulate the pattern of gene expression in the epithelial lining of the lung.
The transduction of an O2 signal to the level of gene
expression requires the nuclear translocation and activation of
redox-responsive transcription factors over specific ranges of
PO2. Hypoxia-inducible factor-1
(HIF-1) and nuclear factor-B (NF-B) are activated by hypoxia
and oxidizing signals, respectively, and are therefore functionally
poised to coordinate the expression or suppression (i.e., by the loss
of transcription factor activity) of genes in response to
PO2 regimens in the lung
epithelium during the birth transition. The activation of HIF-1 is
consistent with its role in coordinating adaptive homeostatic responses
to hypoxia by regulating the expression of vascular, glycolytic, and
cell cycle regulatory genes in a wide variety of tissues (8, 11, 22),
whereas NF-B, first identified as a factor that regulates the
expression of the light chains in mouse B lymphocytes (39), is
central to the expression of stress response genes involved in
modulating the sensitivity of the cell to oxidative injury (12, 36).
Thus a molecular switching mechanism that may integrate gene expression
with the prevailing PO2 during the
transition at birth is in place in the fetal lung epithelium.
To test the concept that the fetal distal lung epithelium is
functionally responsive to shifts in O2 availability that
are representative of the birth transition and beyond, we derived the
following hypotheses: 1) re-creation of mild (23 76 Torr), moderate (23 152 Torr), and severe (23 722 Torr) hyperoxic shifts in isolated fetal alveolar type II (fATII)
epithelial cells will result in the differential activation of HIF-1
and NF-B, 2) this activation will be paralleled by an
altered redox potential as reflected in the glutathione (GSH-to-GSSG)
ratio and glutathione biosynthetic capacity, and 3) the
perinatal lung will experience parallel variations in glutathione
homeostasis and activation of genetic regulatory factors in response to
birth into an O2-rich environment. Our results provide
evidence in support of a functional O2-linked,
glutathione-buffered O2 signaling pathway that regulates the pattern of gene expression during the transition from placental to
pulmonary gas exchange in the distal lung.
| |
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
All experimental procedures involving the use of live animals were approved under the Animals (Scientific Procedures) Act, 1986 (United Kingdom).
Primary Culture of fATII Cells
Fetal rats were removed from pregnant Sprague-Dawley rats by cesarean section on day 19 of gestation (term = 22 days), and the lungs were excised, teased free from the heart and upper airway tissue, and finely minced, then washed free of erythrocytes with sterile, chilled Mg2+- and Ca2+-free Hank's balanced salt solution (0.5 ml/fetus). The cleaned lung tissue was resuspended in 1 ml/fetus of Hank's balanced salt solution containing trypsin (0.1 mg/ml), collagenase (0.06 mg/ml), and DNase I (0.012% wt/vol) and agitated at 37°C for 20 min. The solution was then centrifuged at 100 g for 2 min to remove undispersed tissue, the supernatant was saved to a fresh sterile tube, and an equal volume of Dulbecco's modified Eagle medium (DMEM) with 10% (vol/vol) fetal calf serum (FCS) was added to the supernatant. After the supernatant was passed through a 120-µm-pore sterile mesh, the filtrate was centrifuged at 420 g for 5 min, the pellet was resuspended in 20 ml of DMEM-FCS, and the cells were placed in a T-150 culture flask for 1 h at 37°C to enable fibroblasts and nonepithelial cells to adhere. Unattached cells were washed three times by centrifugation at 420 g for 5 min each and then seeded onto 24-mm-diameter Transwell clear permeable supports (0.4-µm pore size; Costar) at a density of 5 × 106 cells/filter and allowed to adhere overnight at a fetal distal lung PO2 (23 Torr,
3% O2-5% CO2). DMEM-FCS was
exchanged for 4 ml of serum-free PC-1 medium (BioWhittaker)
preequilibrated to a PO2 of 23 Torr
at 37°C 24 h later, and the cells were maintained at this
PO2 until the experiment. Cultures maintained this way remained viable for at least 96 h (adenylate energy
charge 
0.7; monolayer transepithelial resistance 250 
· cm2).
Mild, moderate, and severe PO2
regimens were re-created with four variable O2 incubators
(Biotech Galaxy) preset to fetal distal lung
PO2 (23 Torr, 3%
O2-5% CO2), early postnatal alveolar
PO2 (76 Torr, 10%
O2-5% CO2), mild hyperoxia (152 Torr, 21%
O2-5% CO2), and severe hyperoxia (722 Torr,
95% O2-5% CO2). After at least 24 h of
culture at 23 Torr, the cultures were placed into an equivalent volume
of PC-1 medium that had been previously equilibrated (24 h) to each
PO2 for 4 h before extraction as
detailed in Cell Harvesting and Nuclear Protein
Extraction.
Cell Harvesting and Nuclear Protein Extraction
Nuclear extracts were prepared from monolayers of fATII cells as detailed elsewhere (43), with minor modifications. The filters were washed twice in 5 ml of ice-cold O2-preequilibrated PBS, and the cells (1-2 × 107/PO2 regimen) were
collected and centrifuged at 420 g for 5 min at 4°C. Nuclei
were released by resuspending the pellet in 250 µl of buffer
A containing (in mM) 10 Tris · HCl (pH 7.8), 10 KCl, 2.5 NaH2PO4, 1.5 MgCl2, 1 Na3VO4, 0.5 dithiothreitol (DTT), and 0.4 4-(2-aminoethyl)benzenesulfonyl fluoride hydrochloride (AEBSF) and 2 µg/ml each of leupeptin, pepstatin A, and aprotinin. The suspension
was left in ice for 10 min followed by a 45-s homogenization at a
moderate speed. The nuclei were collected by centrifuging the slurry at
4,500 g for 5 min at 4°C and resuspending the nuclei in 100 µl of buffer B [buffer A adjusted to (in mM) 20 Tris · HCl (pH 7.8), 420 KCl, and 20% (vol/vol)
glycerol]. The nuclei were then lysed at 4°C for 30 min with
gentle agitation, the debris was cleared by centrifugation at 10,000 g for an additional 30 min at 4°C, and the supernatant was
frozen in liquid nitrogen and stored at 
70°C until used. In
all cases, protein content was determined by the Bradford method with
BSA as a standard.
Whole lung extraction. Lungs were excised from prenatal rats
(gestational days 19 and 21; n = 6 each),
postnatal rats (days 0, 1, 2, 3,
4, and 6; n = 4-6 each), and adults (8 wk), rinsed once in saline, and frozen in liquid N2
followed by storage at 70°C. Total protein was extracted by
homogenizing 50-100 mg of tissue in a suitable volume of a buffer
(1:40 wt/vol) containing (in mM) 20 HEPES (pH 7.5), 0.2 EDTA,
1.2 Na3VO4, 1.5 MgCl2, 2.5 NaH2PO4, 100 NaCl, 5 DTT, and 1 AEBSF and 10 µg/ml of leupeptin (Na3VO4, DTT, AEBSF, and
leupeptin were added before extraction). Tissue debris was formed into
pellets by centrifugation at 10,000 g for 30 min at 4°C,
and the supernatant was mixed with an equal volume of the same
extracting buffer supplemented with 40% (vol/vol) glycerol. For
glutathione and adenylate determinations, 50-100 mg of excised
tissue were homogenized in 10 volumes of 7% perchloric acid (PCA) at
4°C for 1 min. The homogenate was centrifuged at 10,000 g
for 30 min, and the supernatant was frozen in liquid nitrogen and
stored at 70°C. On the day of use, the samples were neutralized with a known volume of 3 M KHCO3, centrifuged
as above, and analyzed as detailed in Western Analysis and
Electrophoretic Mobility Shift Assays. Cytosolic
extracts for the assessment of antioxidant enzyme activities were
prepared by homogenizing the whole lung in 1 ml of buffer A (as
above) followed by centrifugation at 4,500 g for 5 min at
4°C. The supernatant was frozen in liquid nitrogen and kept at
70°C until used. The pellet was resuspended in 0.5 ml of
buffer B and processed for nuclear extraction as detailed above.
Western Analysis and Electrophoretic Mobility Shift Assays
Nuclear proteins (20-25 µg) were resolved by SDS-PAGE with a 7.5% separating phase at room temperature at 150 V for 1 h. After electrophoretic transfer onto nitrocellulose, each membrane was washed in Tris-buffered saline (TBS; 20 mM Tris · HCl, pH 7.6, and 500 mM NaCl) followed by blocking for 1 h at room temperature in TBS plus 0.1% (vol/vol) Tween 20 (TBS-T) with gentle agitation. After three washes in TBS-T, the membranes were incubated with either monoclonal anti-HIF-1 IgG (1:200; Novus Biologicals) or polyclonal
anti-NF-B p65 (Rel A) IgG (1:500; Santa Cruz Biotechnology) antibodies in TBS-T overnight at 4°C. Primary conjugates were visualized on film with an anti-rabbit IgG-biotinylated antibody coupled with streptavidin-horseradish peroxidase enhanced
chemiluminescence (ECL, Amersham Life Sciences). Western detection of
-actin was used as a semiquantitative internal control for lane loading.
Electrophoretic mobility shift assays (EMSAs) were conducted with the
following radiolabeled deoxyoligonucleotide sequences purchased from
Genosys: HIF-1 (consensus sequence underlined), W-18,
5'-GC
GTCTCA-3' (3-bp
missense control, M-18,
5'-GCCC
GCTGTCT CA-3');
and NF-B, W-22,
5'-AGTT
AGGC-3'
(1-bp missense control, M-22,
5'-AGTTGA
GACTTTCCCAGGC-3').
After end labeling with polynucleotide kinase (Boehringer Mannheim)
purifying and annealing probes, identical amounts of radioactivity (2 × 104 counts/min) were added to the binding reactions
containing 1-5 µg of fATII cell nuclear extracts in a final
volume of 40 µl in DNA binding buffer (20 mM HEPES, pH 7.9, 1 mM
MgCl2, and 4% Ficoll) containing 0.15 µg of poly(dI-dC)
(Boehringer Mannheim) as a nonspecific competitor. The mixtures were
incubated for 30 min at 25°C before separation on native
nondenaturing 4% polyacrylamide gels at room temperature by
electrophoresis in Tris-borate-EDTA buffer. Where indicated, nonlabeled
oligonucleotide competitor was added in 100-fold molar excess
immediately before the addition of a radiolabeled probe of the same
sequence. For supershift experiments, specific antibodies for HIF-1
(monoclonal anti-HIF-1 IgG; Novus Biologicals) or NF-B
[polyclonal rabbit p65 (Rel A) anti-NF-B IgG; Santa Cruz
Biotechnology] were used at 2 µg/reaction. The antibodies were
incubated with the nuclear extracts before addition of the probe for 1 h at 4°C and then processed as above. Distribution of the
32P label was visualized and quantitated on dried gels with
a Canberra-Packard Instant Imager.
L-Buthionine-(S,R)-Sulfoximine Pretreatment
L-Buthionine-(S,R)-sulfoximine (BSO) selectively inhibits the biosynthesis of reduced glutathione (GSH) by irreversibly blocking the activity of-glutamylcysteine synthetase
(-GCS) (18). A stock solution of BSO (1,125 µM) was prepared in
deionized water and stored at 4°C. fATII cells, cultured at 23 Torr
as in Primary Culture of fATII Cells, were
pretreated for 24 h with BSO [0 (control), 1, 10, or 50 µM] before the cells were shifted for 4 h to each new
PO2. In all cases, the final
concentration of BSO was adjusted with PC-1 medium preequilibrated to
the corresponding PO2. After each
treatment, the cells were processed for nuclear and cytosolic
extraction as in Cell Harvesting and Nuclear Protein
Extraction.
Metabolite and Enzyme Activity Determinations
Adenylate energy charge (4) was used as an index of cellular metabolic competence according to the formula [(ATP concentration + 0.5ADP concentration)/total adenylate concentration], with the concentrations of ATP, ADP, and AMP (total adenylate) determined spectrophotometrically (7). GSH concentrations were also determined spectrophotometrically in neutralized PCA extracts by following the glyoxylase-catalyzed production of S-lactyl-GSH at 240 nm in a 1-ml volume containing 790 µl of phosphate buffer (25 mM KH2PO4 and 25 mM K2HPO4, pH 6.8), 150 µl of 1% BSA, 10 µl of sample, 10 µl of glyoxylase-I (1 mg/ml), and 40 µl of methylglyoxal (0.1 M). Oxidized glutathione (GSSG) was determined in the same cuvette by the addition of 1 mg/ml of glutathione reductase (GSSG-RD) and 8 µl of 12 mM-NADPH and then by following the
change in absorbance at 340 nm (7). Control cuvettes contained the same content of buffer but with a matched sample volume of deionized water.
The same supernatant was used to determine the content of nicotinamide
(NADPH/H+ and NADP+) by glutamate dehydrogenase
and glucose-6-phosphate dehydrogenase reactions, respectively (7). The
protein content of each PCA precipitate was redissolved in 1 M NaOH and
determined as above, enabling the results to be expressed as micromoles
per milligram of protein.
The assay conditions for determining activities of glutathione biosynthetic enzymes are detailed below. In each case, specific activity is expressed as units per milligram of protein where 1 unit of enzyme activity is the amount that catalyzes the formation of 1 µmol product/min. All assays were conducted at 30°C.
Glutathione peroxidase. Glutathione peroxidase (GSH-PX; EC
1.11.1.9) was determined with the method previously described (29). Briefly, cytosolic extracts were incubated in PBS
buffer containing 5 mM EDTA, 10 mM NAD(P)H, 100 U/ml of GSSG-RD, 1.125 M NaN3 (a catalase inhibitor), and 150 mM GSH in a final
volume of 1 ml. The enzymatic reaction was initiated by the addition of
100 µl of 2 mM H2O2 (30%
H2O2, 10.15 M; the molar concentration of
H2O2 was calculated with the coefficient value
0.0394 cm
1 · mM1 at
240 nm), and the linear rate of conversion of NADPH/H+ to
NADP+ at 340 nm between 0 and 5 min after initiation of the
reaction was followed.
GSSG-RD. GSSG-RD (EC 1.6.4.2) was determined with the method previously described (31) with minor modifications. The rate of oxidation of NAD(P)H by GSSG at 30°C was used as a standard measure of enzymatic activity. The activity of GSSG-RD was measured by monitoring the rate of formation of NADP+ at 340 nm between 0 and 5 min after addition of the sample.
-GCS. -GCS (EC 6.3.2.2) was determined with
the method previously described (37). The reaction mixture (1 ml)
contained Tris · HCl (100 mM, pH 8.2), sodium
L-glutamate (10 mM), Na2ATP (5 mM), sodium
phospho(enol)pyruvate (2 mM), KCl (150 mM), NADH (0.2 mM),
pyruvate kinase (5 U, bovine heart type III), and lactate dehydrogenase
(10 U, rabbit heart type II). The reaction was initiated by adding the
sample, and the rate of NAD+ formation was followed at
340 nm.
Glutathione synthase. Glutathione synthase (GS; EC 6.3.2.3) was
assayed in a reaction mixture containing Tris · HCl
(100 mM, pH 8.2, at 30°C), KCl (50 mM),
L--glutamyl-L--aminobutyric acid (5 mM),
ATP (10 mM), glycine (5 mM), MgCl2 (20 mM), EDTA (2 mM),
and sample (added last) in a final volume of 0.1 ml. Added to this was
0.02 ml of 10% sulfosalicylic acid and 0.9 ml of a buffer containing
phospho(enol)pyruvate (0.5 mM), NADH (0.2 mM), pyruvate kinase
(1 U), MgCl2 (40 mM), KCl (50 mM), and
K2HPO4 (250 mM, pH 7.0). The reaction was
initiated with 1 U of lactate dehydrogenase, and the rate of
NAD+ formation was followed.
Statistical Analysis
Experimental results are expressed as means ± SE. Statistical analysis was performed with one-way analysis of variance with SigmaStat 2.0, followed by post hoc Tukey's test to determine significance among treatments. The a priori level of significance at 95% confidence was accepted at P < 0.05.| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Cellular Energy Charge in fATII Cells Exposed to Oxidative Stress
Cells remained metabolically competent throughout the period of preexperimental culture at 23 Torr and on exposure to eachPO2 (Table
1), with no notable change in
cellular energy charge. However, reoxygenation of hypoxic cultures
induced a modest increase in ATP concentration (not significant),
whereas the ADP and AMP concentrations remained static. Also shown in Table 1 is the static cellular energy charge of cell cultures pretreated with BSO (50 µM), which rules out any nonspecific toxicity of BSO. To further confirm this fact, the total protein content of
cells pretreated with 1, 10, and 50 µM BSO was determined to be not
significantly different (P > 0.05) from that in control cultures (data not shown). In addition, alveolar pretreatment with BSO
(50 µM) has been shown to intervene specifically at the level of the
cell cycle events (such as with p53) and the level of factors that are
key components of the signaling pathways governing apoptosis (such as
with Bax and Bcl-2 protooncogenes), suggesting specificity rather than necrotic toxicity (J. J. E. Haddad, R. E. Olver, and S. C. Land, unpublished observations).
|
Immunoblot Western Analysis for HIF-1 and
NF-B
and NF-B but in opposing directions.
HIF-1 showed a clear nuclear accumulation in cells maintained
without perturbation at 23 Torr over those maintained static at 152 Torr. When exposed to a PO2 of 23 76 Torr, nuclear abundance of this protein was reduced but
still remained significantly greater than that in oxygenated cultures
(Fig. 1A), and, indeed, HIF-1
was not detectable in the nuclei of cells exposed to a
PO2 greater than this range.
Figure 1B shows compiled densitometric data from at least four
separate experiments in reference to the abundance of -actin in each
lane. The nuclear abundance of NF-B increased with the elevation in
PO2 beyond 76 Torr, becoming maximal
on a shift toward moderate (23 152 Torr) and severe (23 722 Torr) hyperoxia (Fig. 1C). Compiled experimental
data showing densitometric analysis of NF-B abundance in reference to -actin are shown in Fig. 1D.
|
Analysis of HIF-1 and NF-B Consensus
Sequence DNA Binding Activity
and NF-B (Fig. 2, A and
C, respectively). HIF-1 binding activity was
determined with the probe W-18, which encodes a 5-base hypoxia response
element from the erythropoietin promoter. In parallel with the increase
in the nuclear abundance of HIF-1 in fATII cells maintained at 23 Torr or shifted from 23 to 76 Torr was a marked increase in HIF-1
DNA consensus binding activity, which diminished at higher
PO2 values (Fig. 2, A and
B). Similarly, NF-B binding activity was determined with the
probe W-22, which contained the NF-B consensus sequence from the
human immunodeficiency virus long-terminal repeat. Mild to severe
hyperoxia (PO2 of 23 76, 23 152, and 23 722 Torr) induced NF-B binding activity (Fig. 2, C and D), with maximal activation at
a PO2 of 23 722 Torr and
paralleled the appearance of this transcription factor in the nucleus.
|
In separate experiments, specificity of the transcription
factor-oligonucleotide complex formation noted in Fig.
3 was tested by 1) incubation of
nuclear samples with either M-18 or M-22, 2) addition of
unlabeled W-18 or W-22 at 100-fold molar excess to labeled
oligonucleotides, and 3) supershift analysis with nuclear extracts that had been preincubated with antibodies specific to HIF-1 or NF-B (p65) before addition of the appropriate probe. Figure 3A shows the maximal activation of HIF-1 at 23 Torr
and the retarded supershift complex with a shift from 23 to 76 Torr. The specific binding of HIF-1 is abolished with M-18 and the addition of 100-fold competitor. Figure 3B shows the activation of NF-B and the indicated supershift. Similarly, NF-B
binding was diminished with M-22 or 100-fold competitor.
|
Western Analysis of Cell Cultures Pretreated With BSO
fATII epithelial cells pretreated with BSO show a dose-dependent attenuation in the nuclear abundance of both transcription factors that correlates with the reduced GSH content (Table 2) within thePO2 regimens that stimulated their
maximal activity. Western immunoblotting revealed the effect of BSO (1, 10, and 50 µM) on HIF-1 (Fig.
4A) and NF-B (Fig. 4C)
nuclear abundance kinetics. Ratio analysis by densitometry relative to -actin is given for HIF-1 at 23 Torr and at a
PO2 of 23 76 Torr (Fig.
4B) and for NF-B at a
PO2 regimens of 23 76, 23 152, and 23 722 Torr (Fig. 4D).
|
|
Analysis of HIF-1 and NF-B
DNA Binding Activity With BSO Pretreatment by EMSA
and NF-B in Fig. 5,
A and C, respectively. Exponential decay
of the binding activity of either HIF-1 or NF-B is evident with
increasing concentrations of BSO. Figure 5, B (HIF-1) and
D (NF-B), shows histogram analysis of the dose-response curve.
|
Redox State in fATII Cells at Various PO2 Values
Oxidizing events in the cytosol are buffered, in part, by GSH-PX-catalyzed oxidation of 2GSH GSSG, with the generation
of NADP+ (Fig.
6A). Because this
reaction is readily reversible, the following criteria were used to
indirectly assess intracellular redox status in fATII cells and the
whole lung: 1) GSH reduction state [(GSH concentration GSSG concentration)/GSSG
concentration], 2) reduced (NADPH/H+)
and oxidized (NADP+) forms of nicotinamide, and 3)
assessment of glutathione biosynthetic enzyme activity.
Figure 6, B and C, shows the variations of
glutathione and nicotinamide, respectively, under each
PO2 regimen. GSH reaches an apparent
maxima over 23 76 and 23 152 Torr and,
despite a decline, is persistently elevated at 23 722 Torr tension over cultures that had not been exposed to a
PO2 (23 and 152 Torr). The ratio
of GSH to GSSG varies with treatments as follows: 5:1, 12:1, and 3:1
for a PO2 of 23 76, 23 152, and 23 722 Torr, respectively. With increasing abundance of GSH, NADPH/H+ levels dropped over the same
O2 range, which reflects the NADPH/H+
dependency of GSSG-RD (Fig. 6, A and C). Variations in
glutathione pool expressed as percentages of the control value (152 Torr) are illustrated in Fig. 6B. Glutathione homeostasis for
cell cultures pretreated with BSO is given in Table 2, where the
depletion degree is summarized for the dose-response curve at various
PO2 values.
|
Activity of Enzymes Involved in Glutathione Homeostasis
Glutathione enzyme activities were examined in control, fetal lung O2, postnatal lung O2, and mild and severely hyperoxic cultures (Table 3). fATII cells exposed to fetal distal lung PO2 (23 Torr) showed maximum activity of all enzymes investigated except for-GCS where the activation peak occurs between 23 76 and 23 152 Torr, although its activity is significant in comparison
to the control value (P < 0.05). GSH-PX, which catalyzes the
reduction of H2O2 and certain lipid peroxides,
had its activity significantly increased at 23 Torr (P < 0.01 compared with normoxia). The activity of this enzyme steadily declined
with increasing fetal to neonatal
PO2 regimens but remained
significant at 23 76 Torr (P < 0.05).
GSSG-RD activity increased maximally at 23 Torr (P < 0.05),
remained significant at 23 76 Torr (P < 0.05), and
steadily declined to a minimum at 23 152 and 23 722 Torr. -GCS and GS showed similar trends in their activities. Additionally, -GCS maintained a plateau between 76 and 152 Torr before declining in activity, yet remaining significant relative to the
baseline levels recorded in normoxia. GS activity was maximum at 23 Torr (P < 0.01), significant at 23 76 Torr
(P < 0.05), and declined at higher tensions. A significant
correlation by sigmoidal regression was found between GSH synthetic
activity of -GCS and the total GSH pool increase observed over fetal
to neonatal PO2 regimens (Fig.
7). There was no correlation with other enzymes involved in maintaining intracellular levels of
glutathione.
|
|
Glutathione Homeostasis and Enzyme Activities in Pre- and Postnatal Lung Development
Glutathione homeostasis was assessed in pre- and postnatal lungs obtained at various time intervals. GSH levels in prenatal lungs (gestation days 19 and 21) were significantly lower than those in postnatal lungs (2-6 h) where a maximum was observed relative to gestation day 19 (Fig. 8A). GSSG showed opposite kinetics to GSH variations in the prenatal lung (Fig. 8B), with a minimum observed in the late postnatal period (days 2-6). The content of GSSG at each gestation period was significantly lower than that of GSH (P < 0.01), with an average GSH-to-GSSG ratio of50:1 (Fig. 8, A and
B). GSSG concentration on gestation days 19 and 21 and 1-6 h postnatally was significantly higher than
that in the late postnatal period (days 1-6).
|
The activity of GSH-PX increases steadily to a plateau that is
maximally sustained between days 0 and 1 after birth
compared with those in the adult and on day 4 after birth
(Table 4). This activity was highly
significant on gestation day 19 (P < 0.001). The
activity of GSSG-RD showed opposite kinetics, with minimal activity
observed on day 0 (3 h) after birth. This activity regained a
modest increase on days 0 (6 h) and 1, although it was
still significantly lower compared with that in the adult. -GCS
activity started to increase on gestation day 19 and reached a
maximum on day 0 (3-6 h) after birth. The same
trend was observed with the activity of GS. The elevation in GSH
concentration in the postterm lung in the early postnatal period
compared with that in the preterm lung was significantly correlated
with the activities of -GCS (R = 0.85; P < 0.05;
Fig. 9A) and GS (R = 0.81;
P < 0.05; Fig. 9B).
|
|
Transcription Factor DNA Binding Activity in Pre- and Postnatal Lung Development
The DNA binding analysis of HIF-1 (Fig.
10A) indicated maximal activity
on gestation day 19, which declined during the early postnatal
period. DNA binding analysis of NF-B (Fig. 10B) indicated activation in the early postnatal period, reaching an apparent peak
3-6 h after birth, which then declined to a still detectable level
on postnatal days 1 and 4. The specificity of the
corresponding binding of HIF-1 or NF-B was determined by the
addition of mutant (M-18 and M-22), 100-fold competitor, and a
retarding band supershift formed by the specific immunoglobulin (data
not shown).
|
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Genetic responses to hypoxic or hyperoxic stresses are regulated, in part, by transcription factors in which redox-dependent activation, nuclear translocation, and DNA consensus binding are the rate-limiting determinants of the species and expression intensity of genes that sustain cellular functional integrity under either stress condition (8, 20). Because the O2 history of distal lung epithelial development is restricted to ~23 Torr in utero, we reasoned that the fourfold increase in PO2 that accompanies the first breath at birth would constitute a substantial O2 signaling event that would mediate a change from hypoxic to oxidative (aerobic) forms of genetic regulation.
In fATII epithelial monolayers maintained in the steady state at a
fetal distal lung PO2 (23 Torr),
HIF-1 showed a high level of nuclear translocation and activation
that persisted when the cultures were exposed to a sustained shift
toward early postnatal distal lung
PO2 (76 Torr). Neither nuclear translocation nor consensus sequence binding were detected in cells
cultured in the steady state at 152 Torr or exposed to a PO2 of 23 152 Torr or
beyond. Work conducted by others (47) on adult ferret lungs and lung
epithelial cell lines has determined that HIF-1 cellular abundance
is graded between 0.5 and ~30 Torr, becoming maximal at the lower end
of this range over 2-8 h. Our results suggest that the activity of
HIF-1 resides over a wider spectrum of
PO2 in fATII epithelial cells that
incorporates both fetal and early postnatal alveolar
PO2 values, an assertion that is
supported by our observations that HIF-1 nuclear abundance and
consensus binding activity are present in the whole lung preterm, is
sustained through the early postnatal period, and is lost thereafter.
Because the activation pathway for HIF-1 is favored by a reducing
environment (i.e., lowered ROS production) coupled with deoxyferroheme
conformation, muted ubiquitinization, proteasome degradation,
phosphorylation, and aryl hydrocarbon nuclear translocator-dependent
nuclear translocation (35), this finding presents the intriguing
possibility that one or more of these components serves to effect a
rightward shift in O2 dependency (i.e., raise the
Michaelis-Menten constant) of HIF-1 activation in the perinatal
lung. Although the functional significance of this remains to be
demonstrated, it seems that the sensitivity of HIF-1 activation
pathways to subambient PO2 values is
heightened during birth.
Nuclear abundance and consensus binding of NF-B were not detected in
fATII epithelial cells under steady-state culture at either 23 or 152 Torr; however, the activity of this transcription factor was powerfully
induced by shifts from fetal to early postnatal PO2 (76 Torr) and into moderate (152 Torr) and severe (722 Torr) hyperoxia, suggesting that this is
primarily responsive to acute changes in
PO2, which, at the lower end of the range, are coincident with early postnatal
PO2. As with HIF-1, the profile of
NF-B activation in the whole lung follows the expected change in
PO2 at birth, becoming maximally active in the initial 24 h of the postnatal period and falling thereafter. Although NF-B activation is well established as an early
response to oxidative stress in the lung (26, 36), a study (32) based
on fetal distal lung epithelial cultures indicated that there is a
strong association among moderate hyperoxia (i.e., PO2 > 152 Torr), ROS production,
and NF-B-regulated expression of epithelial Na+
channels, pointing to the importance of this pathway as a modulator of
developmental events in the lung at birth. When taken together with
HIF-1 data, the activity profiles of both transcription factors
appear sufficiently tuned to mediate the changeover from hypoxic to
oxidative forms of genetic regulation at
PO2 values that are within the range
of those expected with the first breaths at birth.
The signal transduction components that link the availability of
O2 to the activation of these transcription factors are
poorly defined but are broadly believed to hinge on the free abundance of oxidants (i.e., ROS) in the cytosol (15, 28). In the case of
HIF-1, for example, posttranslational stability, nuclear
translocation by the aryl hydrocarbon nuclear translocator, and
consensus DNA binding are coupled with O2-associated
changes in both conformation and activity of a ferroheme-containing
protein, believed to express peroxide generation via NADPH oxidase-type
activity (3, 23). Hypoxic cessation of peroxide production mediates
HIF-1 stabilization, nuclear translocation, and gene expression (21,
42). The activation pathway for NF-B, on the other hand, requires
the disassociation from its cytosolic inhibitory subunit IB, an
event that requires phosphorylation but which is favored by oxidizing
conditions (12, 28, 34). Clearly, the extent to which cells express
antioxidant defenses bears consequence for the efficiency with which an
O2- or redox-linked signal can be transmitted from
environment to effector protein.
To determine the redox-buffering capacities that accompany HIF-1 and
NF-B activation, intracellular levels of glutathione and its
cofactor nicotinamide were determined under each
PO2 regimen. The ROS scavenging
function of glutathione
(L--glutamyl-L-cysteinylglycine), the major
intracellular nucleophile in mammalian cells (27, 44), centers on the
oxidation of the cysteine thiol moiety of two molecules of GSH to
produce GSSG (Fig. 6A), a reaction catalyzed by GSH-PX. GSSG is
enzymatically recycled to 2GSH by GSSG-RD through an energy-consuming
reaction that is dependent on the availability of NADPH/H+
as an electron acceptor. Consequently, GSH-to-GSSG ratios serve as a
correlative index of the oxidative potential within the cytosol. Our
studies point toward an association between each
PO2 regimen and an increased total
glutathione pool in which the absolute concentration of GSH is elevated
four- to fivefold over that in cells maintained in steady-state culture
at either 23 or 152 Torr. The kinetics of GSH variation over fetal to
neonatal PO2 values showed maxima at
PO2 regimens of 23 76 and 23 152 Torr, which matched the minima
recorded for NADPH/H+ under the same conditions, and as
expected, an increase in NADP+ content was reported with
ascending PO2 regimens that are
significantly higher than the concentration of NADPH/H+.
Because GSH acts as a powerful intracellular redox buffer, changes in
its overall abundance may modulate redox-linked signaling events within
the cell (38). Interestingly, the observed elevation in GSH from 23 Torr was accompanied by a significant reduction in the nuclear
abundance and consensus binding of HIF-1 and an opposing elevation
in NF-B over fetal to neonatal
PO2 regimens and beyond. These
results suggest that the transcription factor activation profiles we
observed for HIF-1 and NF-B over the birth transition complement
the changes in the reducing potential of the glutathione pool and the
establishment of GSH as the predominant redox form. Experimental
depletion of the glutathione pool by dose-dependent inhibition of
-GCS with BSO resulted in the stepwise inactivation of both
transcription factors under each activating PO2 value. Intriguingly, although
HIF-1 appeared maximally active at 23 Torr in a low total
glutathione environment (145 µmol/mg protein; Table 2), its activity
was substantially more responsive to glutathione depletion compared
with all activated ranges of NF-B explored when the control
glutathione content was two- to threefold greater. Although we cannot
infer from these data alone that the effect is specifically ROS
dependent, it appears clear that maintenance of the glutathione pool
and, by inference, the shuttling of electrons between reductant and
oxidant components of this pathway are prerequisites for transcription
factor activation under any given O2 profile. We are
currently investigating the molecular mechanisms by which this
activation may be coupled to glutathione homeostasis and lung cellular
survival or death in the event of electrophilic or oxidative tissue injury.
The substantial changes in the GSH-to-GSSG ratio observed in response
to upward PO2 values in the fATII
monolayer culture model and the similarity of this response to changes
in the glutathione pool in whole lung over the perinatal period may occur via glutathione reduction-oxidation cycling or de novo synthesis. GSSG recycling to 2GSH is catalyzed by NADPH-dependent GSSG-RD, the
greatest detectable activity of which coincided with the fetal to
neonatal range of PO2 values in fATII
cells and the late-gestational phase of lung development. We have
observed that GSSG-RD blockade with
1,3-bis-(2-chloroethyl)-1-nitrosourea leads to a substantial
accumulation of GSSG at the expense of GSH in fATII monolayers exposed
to a PO2 of 23 76 Torr (Haddad et al., unpublished observations), which
would tend to suggest that this pathway is operative during moderate
PO2 values. However, the lowered
activity of this enzyme over hyperoxic PO2 regimens, coincident with
reduced GSH-PX activity, indicates that the importance of
glutathione recycling via this route is restricted to perinatal
PO2 values. The pathway through which
de novo synthesis of GSH from glutamate, glycine, and cysteine proceeds
is rate limited by the activities of -GCS and GS, both of which
exhibit elevated activities at fetal to early neonatal
PO2 values in fATII cells and
correlate positively with the increase in the GSH pool. Moreover, the
activities of both -GCS and GS are highly elevated in the whole
lung, particularly in the early postnatal period, again coincident with
the increase in postnatal PO2. De
novo synthetic pathways are clearly important routes for
establishing GSH as the dominant nucleophile in the newly ventilating
lung. Taken together, the marked elevations in GSH concentration and
enzyme activity involved in glutathione homeostasis and synthesis
(GSH-PX, GSSG-RD, -GCS, and GS), redox cycling (GSH-PX and GSSG-RD),
and antioxidant defense (GSH-PX) over the fetal to neonatal
PO2 and in the early perinatal
period underscore the importance of the glutathione biosynthetic
pathway as an adaptable component of respiratory antioxidant defenses
critical for surviving birth (16).
Functionally, the relationship among O2, glutathione
biosynthesis, and transcription factor activity bears important
implications for the pattern of cellular survivorship and
alveolarization on exposure to O2-linked stresses.
Perturbations in glutathione homeostasis in the lung epithelium have
been implicated in several pathological conditions such as idiopathic
pulmonary fibrosis (10), respiratory distress syndrome (9, 25), and
cystic fibrosis (33). In a number of cellular models, depletion of GSH
accelerates the onset of apoptosis on exposure to oxidants (6), an
effect that can be reversed by experimental maintenance of GSH content
by inhibition of its extrusion (17). Moreover, a recent study (11) in
HIF-1-knockout embryonic stem cells has demonstrated that hypoxic
induction of p53 and p21, suppression of Bcl-2 (an apoptosis suppressor
protein), and subsequent entry into apoptosis are dependent on the
presence of functional wild-type HIF-1 genes. Conversely, activation
of NF-B has been found to prevent entry into apoptosis after
treatment, including oxidative injury, which is disruptive to DNA and
is powerfully induced in A549 adenocarcinoma ATII cultures by lethal
doses of ROS (41, 46). Although there exists a broad argument in
support of a role for NF-B in mediating cytoprotection under
oxidative stress, the full implications of this suite of observations
remain controversial. Clearly, however, the relationship between cell
cycle events and the novel association between O2-linked transcription factor activation and glutathione biosynthesis in the
perinatal lung bears important implications for the treatment of
respiratory conditions in the newborn (compare Refs. 14 and 40).
In conclusion, we provide evidence that both abundance and activation
kinetics of HIF-1 and NF-B in the fetal distal lung epithelium
are differentially responsive to changes in O2 availability over the shifts in PO2 that occurs
during inhalation of the first breath. This is coincident with a
substantial O2-linked increase in the capacity of the
tissue to engage in glutathione biosynthesis and redox shuttling that
may effectively form a feedback mechanism governing redox-linked
signaling events. This O2-responsive characteristic of the
perinatal epithelium highlights PO2 as an important modulator of events crucial to the transition from
placental- to pulmonary-based modes of gas exchange and thus bears
consequence for the clinical treatment of pediatric respiratory disorders that require O2 therapy.
| |
ACKNOWLEDGEMENTS |
|---|
We thank Helen Murphy for expert technical assistance in isolating and preparing fetal alveolar type II epithelial cell cultures.
| |
FOOTNOTES |
|---|
This work was supported by Medical Research Council and Anonymous Trust grants to S. C. Land.
J. J. E. Haddad is recipient of a George J. Livanos Prize PhD scholarship.
This work was presented at the semiannual meeting of The Physiological Society in Manchester, UK (19).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Address for reprint requests and other correspondence: S. C. Land, Oxygen Signalling Group, Tayside Institute of Child Health, Ninewells Hospital and Medical School, Univ. of Dundee, Dundee DD1 9SY, UK (E-mail: s.c.land{at}dundee.ac.uk).
Received 12 August 1999; accepted in final form 26 October 1999.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
1. Acarregui, M. J., J. J. Brown, and R. K. Mallampalli. Oxygen modulates surfactant protein mRNA expression and phospholipid production in human fetal lung in vitro. Am. J. Physiol. Lung Cell. Mol. Physiol. 268: L818-L825, 1995
P < 0.05; 
