| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
and Activation of Caspase-3 in Hypoxia-Reoxygenated Bone Marrow Stroma Is Negatively Regulated by the Delayed Production of Substance P1
Departments of
*
Medicine and
Surgery, Trauma Division, University of Medicine and Dentistry of New Jersey, Newark, NJ 07103
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(HIF-1) mRNA levels in BM stroma. Because long-term hypoxia induced
the expression of PPT-I in BM mononuclear cells, we used timeline
studies to determine whether PPT-I is central to the
biologic responses of BM stroma subjected to 30-min hypoxia
(pO2 = 35 mm Hg) followed by reoxygenation. HIF-1
mRNA and protein levels were increased up to 12 h. At this time,
-PPT-I mRNA was detected with the release of SP at 16 h. SP
release correlated with down-regulation of HIF-1 to baseline. A
direct role for SP in HIF-1 expression was demonstrated as follows:
1) transient knockout of -PPT-I showed an increase in HIF-1
expression up to 48 h of reoxygenation; and 2) HIF-1 expression
remained baseline during reoxygenation when stroma was subjected to
hypoxia in the presence of SP. Reoxygenation activated the PPT-I
promoter with concomitant nuclear translocation of HIF-1 that can
bind to the respective consensus sequences within the PPT-I promoter.
SP reversed active caspase-3, an indicator of apoptosis and
erythropoiesis, to homeostasis level after reoxygenation of hypoxic
stroma. The results show that during reoxgenation the
PPT-I gene acts as a negative regulator on the
expression of HIF-1 and active caspase-3 in BM stroma
subjected to reoxygenation. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The role of the stromal cells as cellular regulators of hemopoiesis shows their importance in maintaining homeostasis in the BM microenvironment (3, 4, 7, 8). The stromal cells are in close proximity to the hemopoietic stem cells so that cytokines and other soluble factors, released from the stroma, can interact with specific receptors on the BM stem cells and/or stroma (1, 7, 8). Cytokine interactions with the stroma and stem cells could trigger a cascade of biological responses to stimulate hemopoietic activity (7, 9). The induction of cytokines and other soluble factors in BM stroma (10, 11, 12) could be mediated by the following: inflammation, surgical trauma in which there is rapid blood loss, or any other physiologic disruption that requires hemopoietic activity.
This study addresses the roles of the preprotachykinin-I
(PPT-I) and hypoxia-inducible factor-1
(HIF-1) genes in hemopoietic homeostasis following
hypoxia and reoxygenation. In particular, we determined how each
regulates the expression of the other (9, 13). HIF-1 is
a transcription factor that is increased in conditions of reduced
cellular O2 (13, 14). DNA binding
site for HIF-1 is present in several genes that are sensitive to
changes in pO2 (15, 16). A recent
report (17) suggested that HIF-1 could also be induced
during inflammation. Peptides encoded by the PPT-I gene
exert pleiotropic functions with specific effects in the different
organs (18). In the BM and other lymphoid organs, PPT-I
peptides regulate hemopoiesis and immune functions (9, 19).
Prolonged exposure of BM cells to hypoxia induced the expression of
PPT-I (12). Substance P (SP), the major PPT-I
peptide, stimulates hemopoiesis (9). Thus, up-regulation
of PPT-I expression during reoxygenation of stroma exposed to hypoxia
(12) would be consistent with a need for hemopoietic
activity. Hypoxia reoxygenation is an indicator of hemorrhagic shock
(10, 20, 21), which requires replacement of immune and
other blood cells. In this report we examined differential gene
expression in SP-stimulated and unstimulated BM stroma using
microarray. HIF-1 mRNA represented one of the genes that was
significantly decreased in the stimulated cells. This observation was
unexpected because hypoxia induced PPT-I expression, which would result
in the production of SP (9, 12). To investigate this
paradox, we subjected BM stroma to hypoxia and then performed timeline
studies during reoxygenation to understand the molecular relationship
between HIF-1 and PPT-I genes. Justification
of this model is supported by reports that showed hypoxia reoxygenation
as an appropriate experimental system to study hemopoietic activity in
hemorrhagic shock (10, 20, 21). Blood gas levels of BM
aspirates from healthy subjects indicate pO2
ranging between 54 and 58 mm Hg (our unpublished data). This
suggests that during hemorrhagic shock, the pO2
of the BM might be much lower than indicated by the levels in the
arterial peripheral blood. Therefore, an understanding of hemopoietic
activity in a hypoxic microenvironment could not be extrapolated from
the reports in normoxic conditions, and such mechanisms remain
undefined.
The results showed that unstimulated stromal cells express high levels
of HIF-1 mRNA. Unlike other tissues in which HIF-1 mRNA is
constant during hypoxia, in BM stroma, its levels were increased
following reoxygenation with concomitant protein translation. In
timeline studies, the results showed that the delayed release of SP
during reoxygenation of hypoxic stroma hypoxia blunted the expression
of HIF-1 and inactivated caspase-3, an indicator of apoptosis
(22, 23). Reoxygenation activated the PPT-I promoter
subjected to hypoxia with concomitant translocation of two
transcription factors, NF-
B and HIF-1, capable of interacting
with the respective consensus sequences within the 5' regulatory region
of the PPT-I gene.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
SP, BSA, hydrocortisone, 2-ME, streptavidin, and Ficoll-Hypaque
were purchased from Sigma (St. Louis, MO). The neurokinin-1 (NK-1)
antagonist, CP-99,994, was obtained from Pfizer (Groton, CT). SP and
CP-99,994 were dissolved and stored as described (24).
Rabbit anti-SP, rabbit anti-active caspase-3, and rabbit
anti-RelA/p65 were purchased from Arnell Products (New York, NY),
Cell Signaling (Beverly, MA), and Santa Cruz Biotechnology (Santa Cruz,
CA), respectively. Two sources of HIF-1 mAb were purchased from BD
Transduction Laboratories (Lexington, KY) and Novus Biologicals
(Littleton, CO). Biotinylated SP was purchased from Chiron Mimotopes
(Emeryville, CA). Alkaline phosphatase-conjugated goat anti-rabbit
IgG was obtained from Kirkegaard & Perry Laboratories (Gaithersburg,
MD). Hoffman-LaRoche (Nutley, NJ) provided the recombinant human
IL-1.
Cells
BM stroma was prepared as described (25). Briefly,
BM aspirates were obtained from healthy donors, ages 20–35 years,
following guidelines from the Institutional Review Board of the
University of Medicine and Dentistry of New Jersey-New Jersey Medical
School. Unfractionated cells from BM aspirates were cultured at 33°C
in -MEM (Life Technologies, Grand Island, NY) with 12.5% FCS
(HyClone Laboratories, Logan, UT), 12.5% horse serum (HyClone
Laboratories), 0.1 µM hydrocortisone, 0.1 mM 2-ME, and 1.6 mM
glutamine. At day 3 of culture, the mononuclear cells were selected by
Ficoll-Hypaque density gradient and then replaced into the respective
tissue culture flask. Cultures were reincubated with weekly replacement
of 50% medium until confluence. CSH-HeLa cells were obtained from Dr.
S. Gonnery, Department of Biochemistry and Molecular Biology,
University of Medicine and Dentistry of New Jersey-New Jersey Medical
School. CSH-HeLa cells were cultured in DMEM with high glucose (Sigma),
1 mM glutamine (Sigma), and 10% FCS. All media and cell culture
reagents contained <0.015 EU/ml as determined by the
Limulus Amebocyte Lysate, E-TOXATE kit (Sigma).
Hypoxic exposure and preparation of cell extracts
BM stroma was subjected to hypoxia as described
(12). Five days before cells were subjected to hypoxia,
cultures that were 
80% confluent were transferred to a 37°C
incubator. Immediately before hypoxic exposure, nonadherent cells were
removed and discarded from the confluent stroma. After this, the
adherent cells were washed more than three times with serum-free
-MEM. Washing was terminated when there was no evidence of
nonadherent cells, which was determined microscopically. After
washing, 3 ml of serum-free -MEM supplemented with
insulin-transferrin-selenium-A (Life Technologies). Flasks were placed
in a hypoxic chamber and flushed with gas comprising 5%
CO2 and 95% N2 for 5 min
resulting in a pO2 of 30–45 mm Hg. Chambers were
sealed and then incubated at 37°C for 30 min. After this, flasks were
removed from the hypoxic chamber and then placed in a 37°C incubator
with a normoxic atmosphere (5% CO2 and
95% air). At different times after reoxygenation, stroma was used for
the following: 1) extraction of total RNA for Northern analyses; 2)
collection of cell-free supernatants, which were stored in
siliconized tubes at -80°C for ELISA and Western blots; or 3)
preparation of cell-free extracts from whole cells or nuclei. Extracts
were stored at -80°C for ELISA and Western blots.
Differential gene expression
The differences in gene expression between SP-stimulated and unstimulated BM stroma were examined using a commercial slide array with 2400 genes: Micromax Human cDNA Microarray System I (NEN Life Science, Boston, MA). The technique followed manufacturer’s instructions. Briefly, BM stromal cultures from nine healthy donors were divided into two groups: 1) unstimulated; or 2) stimulated with 10 nM SP for 16 h. Total RNA from each group was pooled and then used to prepare cDNA probes. Labeling of cDNA probes was performed with biotin and DNP for unstimulated and stimulated stroma, respectively. Slides were hybridized with a mixture of the labeled probes from the stimulated and unstimulated cells. Hybrids were detected with the Tyramide Signal Amplification system (NEN Life Science), which resulted in signals by cyanine-3 (unstimulated) and cyanine-5 (stimulated). Slides were scanned with ScanArray 3000 (GSI Lumonics, Watertown, MA), and the data were analyzed with NEN data analysis, version 2.0.
Transfection and reporter gene assay
Different parts of the PPT-I promoter were ligated in
pGL3-basic, upstream of the luciferase reporter (26) (see
Fig. 6
A). The following are constructs of the 5' regulatory
sequences of the PPT-I gene (26): 1) PPT-I-1.2
(1.2 kb), which includes the region upstream of the transcription site
(N0, 722 bp), Exon 1 (89 bp) and Intron 1 (409 bp); 2) N0; and 3) N0 in
tandem with Exon 1. Stromal cells, 80% confluent, were cotransfected
with pGL3 containing each insert and p-gal-Control (0.5 µg each)
using SuperFect (Qiagen, Valencia, CA). Transfectants were subjected to
30-min hypoxia and, after 16 h of reoxygenation, cells were lysed
in 30 µl of 250 mM Tris (pH 8.0) by repeated freeze-thaw cycles in a
dry ice/ethanol bath. Cell-free lysates (24 µl) were obtained by
centrifugation at 15,000 x g for 5 min at 4°C and
then diluted with 5x cell culture lysis buffer (Promega, Madison, WI).
Luciferase and -galactosidase (-gal) activities were quantitated
with 10 µl of lysates using a Luciferase assay system (Promega) and a
Luminescent -gal detection kit II (Clontech Laboratories, Palo Alto,
CA), respectively. The ratios of Luciferase/-gal in cells
transfected with vector alone ranged from 0.18 to 0.19 and was
normalized to 1. Hypoxia reoxygenation showed no change in -gal
activity (data not shown).
|
Competitive ELISA quantitated SP-IR as described (27). Briefly, 96-well plates were coated with a complex of streptavidin-biotinylated SP. Equal volumes (50 µl) of unknown samples and optimum rabbit anti-SP were added to quadruplicate wells. Each sample was assayed as undiluted and three serial dilutions. Bound anti-SP was detected with alkaline phosphatase-conjugated goat anti-rabbit IgG and Sigma 104 phosphatase substrate (Sigma). SP-IR levels were calculated from a standard curve developed with OD at 405 nm vs 12 serial dilutions of known SP concentrations.
Immunoprecipitation and Western blots
Cultures of BM stromal cells in a 25-cm2
flask were gently washed with PBS, pH 7.4, and then replaced with 1 ml
of PBS containing 1 mM PMSF and 5 µM leupeptin (both protease
inhibitors purchased from Sigma). Cell extracts were prepared by
repeated freeze-thaw cycles in a dry ice-ethanol bath. Total proteins
were quantitated in cell extracts from stromal cultures using a kit
from Bio-Rad (Hercules, CA), and 15 µg was analyzed in Western blots
for SP, HIF-1, or active caspase-3, as described (26, 27). HIF-1 and active caspase-3 were electrophoresed on 15%
SDS-PAGE and SP on 4–20% gradient gel (Invitrogen, Carlsbad, CA).
Proteins were transferred to polyvinylidene difluoride membrane (NEN
Life Science), which were incubated overnight with anti-HIF-1,
anti-SP, or anti-active caspase-3 at a final concentration of
1/1000. After this, membranes were washed and then incubated with
HRP-conjugated goat anti-rabbit IgG (1/5000) for 45 min. HRP was
developed with ECL Western blotting detection reagents (Amersham
Pharmacia Biotech, Piscataway, NJ). The
Mr values of developed bands were
compared with prestained protein standards (Diversified Biotech, Newton
Center, MA).
Nuclear extracts from BM stromal cells (26) were analyzed
for HIF-1 and RelA/p65 by immunoprecipitation as described
(31). Briefly, 500 µl of extracts was incubated
consecutively at 4°C with HIF-1 (250 µg/ml) overnight and
protein G-Sepharose (Sigma) for 2 h. Samples were centrifuged at
10,000 x g for 30 min at 4°C, and the pellets were
analyzed in Western blots for HIF-1 protein as described above.
Immunoprecipitation was repeated in the same supernatants with 500
µg/ml anti-RelA/p65.
Transient knockout of PPT-I in BM stromal cells
DNA sequences from exons 2 (nt: +101/+224), 3 (nt: +211/+245),
and 7 (nt: +431/+456) of -PPT-I (29) were synthesized
in tandem at the University of Medicine and Dentistry of New Jersey-New
Jersey Medical School Molecular Core Facility with HindIII
and Kpn.I linkers in the 5' or 3' ends. The linkers directed
the ligation of the sequence in the sense and anti-sense
orientations in the cloning site of pEGFP-N1 (Clontech Laboratories).
Stromal cells were transfected with pEGFP-N1. After this, cells were
examined at an excitation of 495 nm with an Olympus (Melville, NY)
Probis microscope and determined to be >90% positive for green
fluorescence. Transfectants were determined to be transiently deficient
of PPT-I expression by stimulating the transfected cells (sense and
anti-sense) or untransfected cells for 16 h with an inducer of
PPT-I (30): 25 ng/ml IL-1 in -MEM (Sigma) containing
2% FCS. Intracellular immunofluorescence with anti-SP and Western
blots indicated no evidence of PPT-I expression (data not
shown).
EMSA
EMSA assays were performed as described (26).
Table I shows the consensus
sequences for HIF-1 binding sites, which spanned DNA sequences
+643/+649 (HIF-1-1) and +1075/+1080 (HIF-1-2) of the reported
clone (GenBank accession no. AF252261). Table I also shows the
sequences of the oligonucleotide probes used in the EMSA for HIF-1.
The oligonucleotide probes for RelA/p65 were previously described
(28). Double-stranded probes, 20 ng (31),
were end labeled with [
-32P]ATP using T4
polynucleotide kinase. Labeled probe was incubated for 1 h with 2
µg of nuclear extracts from HeLa cells or BM stroma. Parallel
reactions were performed in the presence of excess cold competitor or
250 µg/ml anti-HIF-1. Reactions were separated on 4% PAGE.
Gels were dried and then developed by autoradiography after
12 h.
|
Northern analyses for steady-state -PPT-I and HIF-1
mRNA were performed as described (24). Stromal cells were
stimulated with 10 nM SP for 16 h in serum-free -MEM
supplemented with Insulin-Transferrin-Selenium-A and then immediately
subjected to hypoxia. Cultures were allowed to recover for different
times in normoxic conditions as described above. Control cultures
included unstimulated stroma placed in parallel experimental conditions
in serum-free medium. Total RNA (10 µg) and RNA ladders were
separated on 1.2% agarose. After this, RNA was transferred to nylon
membranes (S & S Nytran, Keene, NH), which were fixed by consecutive
exposure to UV cross-link and baked for 1 h at 80°C. Membranes
were hybridized with -PPT-I or HIF-1 cDNA probes, which were
randomly labeled with 3000 Ci/mM [-32P]dATP
(DuPont-NEN, Boston, MA) using the Prime-IT II random primer kit
(Stratagene, La Jolla, CA). After hybridization, membranes were washed
twice at room temperature with 2x SSC containing 0.1% SDS and then a
third wash for 20 min at 37°C with buffer of similar stringency.
Bands were developed by autoradiography. Band intensities from
rehybridization with cDNA for 18S rRNA were used for normalization.
Human cDNA for 18S rRNA and -PPT-I were previously described
(31). HIF-1 cDNA was prepared by RT-PCR with total RNA
from CSH-HeLa cells. The primer pair, 25 nt each, spanned sequences
+541/+1060 of the reported sequence for HIF-1 cDNA
(32). PCR products with the predicted size were ligated
into the cloning site of PCR2.1-TOPO (Invitrogen), and the DNA was
sequenced at the Molecular Resource Facility of New Jersey Medical
School in both orientations using T7 and SP6 primers. Analyses with the
GCG DNA/protein sequence analysis programs, version 10 (Genetics
Computer Group, Madison, WI) indicated >99% homology with the
reported clone (32).
Statistical analysis
Data were analyzed using Student’s t test to determine the significance (p value) between experimental values.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
mRNA in BM stromaThis study uses BM stromal cells, which comprise three major cell subsets. To ensure variability among human subjects, the following steps were followed in the preparation stage: 1) healthy donors were within a narrow range of age; 2) the same hematologist collected all of the BM aspirates using a constant volume of withdrawal: 3) the same laboratory personnel cultured the stromal cells. The phenotype of the adherent cell population was characterized from representative stromal flasks of seven different donors as described (24, 26). Based on the results of immunofluorescence, the mean percentages were 90 ± 5% fibroblasts, 5 ± 4% endothelial cells, and 3 ± 1% macrophages. Cytochemical staining for nonspecific esterase (24) indicated similar percentages of macrophage.
Microarray analysis was performed with total RNA extracted from
SP-stimulated (16 h) and unstimulated pooled BM stroma. Arrays
represented 2400 genes based on prosite motifs. Differences of 
2-fold
between stimulated and unstimulated cultures were considered
significant (Fig. 1). Based on the
biology of HIF-1 and its association with hypoxia and PPT-I, we
studied the mechanism that might partly explain the down-regulation of
HIF-1 mRNA by SP, shown in the boxed area of Fig. 1A. The
signal intensities for HIF-1 for unstimulated and stimulated stroma
were 2031 and 531, respectively, indicating a decrease of 4-fold by
SP stimulation.
|
, we used
Northern analyses to determine the differences in HIF-1 mRNA in BM
stroma stimulated with 10 nM SP for 16 h and unstimulated stroma.
Representative result for analyses performed with three different BM
donors is shown in Fig. 1
, B and C. There was a
significant decrease in HIF-1 mRNA (Fig. 1B, lane 2)
compared with unstimulated stroma (Fig. 1B, lane 1).
Normalization of band densities with 18S rRNA indicated a 10-fold
decrease in SP-stimulated stroma (Fig. 1C). The results
showed that SP mediated a decrease in the level of HIF-1 mRNA after
16 h.
Effects of hypoxia on HIF-1 in the presence or absence of SP
Hypoxia induced the expression of PPT-I in BM mononuclear cells
(12). Because HIF-1 is associated with reduced
O2, its down-regulation by SP (Fig. 1), the major
PPT-I peptide, was unexpected. The induction of PPT-I by hypoxia was
reported in experiments in which BM mononuclear cells were subjected to
24 h of hypoxia (12). Timeline studies were performed
to understand this previous report, and the observation is shown in
Fig. 1. BM stroma was subjected to hypoxia for 30 min in the presence
or absence of 10 nM SP. After this, stromal cultures were allowed to
recover for different times in normoxic conditions, and the stromal
cell extracts were analyzed in Western blots for HIF-1 protein.
Representative results of five experiments, each performed with a
different BM donor, showed a strong band in hypoxic cultures that were
recovered for 8 h (Fig. 2A,
lane 2) and 12 h (Fig. 2A, lane 3). A relatively
weak to almost undetectable band was shown for 16 h (Fig. 2A, lane 4), and at 24 h no band was detected (Fig. 2A, lane 5). Normoxic stroma showed no band (Fig. 2A,
lane l).
|
B shows the effects of SP on HIF-1 protein. This
question was addressed by adding 10 nM SP to the BM stromal cultures
during hypoxia and recovery times. In the presence of SP, no band was
detected for HIF-1 protein after recovery times of 8 and 12 h
(Fig. 2B, lanes 1 and 2). To determine the
specificity of this observation, we repeated similar experiment in the
presence of an antagonist (CP-99,994) to the high affinity SP receptor
NK-1 (19, 29). Stromal cells subjected to hypoxia and
recovered for 8 and 12 h in the presence of 10 nM SP and 100 nM
CP-99,994 resulted in the detection of HIF-1 protein (Fig. 2B,
lanes 3 and 4). These results indicate that NK-1
antagonist can reverse the blunting effect of SP on HIF-1
translation. CP-99,994 alone in normoxic stroma showed no band (data
not shown).
The next set of studies determined the level of HIF-1 mRNA at
different times during recovery from hypoxia. Fig. 3A shows representative
results of three experiments, each performed with a different BM donor.
The normalized ratios of HIF-1 mRNA over normoxic/unstimulated
stroma (fold change) are shown in Fig. 3B and were derived
by normalizing the densities of HIF-1 mRNA with those of 18S rRNA.
Consistent with detectable HIF-1 mRNA in the microarray studies and
the data shown in Fig. 1A, a band was detected for HIF-1
mRNA in the normoxic/unstimulated stroma (Fig. 3A, lane 1).
To determine the effect of SP on HIF-1 mRNA in the normoxic stroma,
cells were stimulated with 10 nM SP for 12 h. Compared with the
reduced level of HIF-1 mRNA in normoxic stroma stimulated with SP
for 16 h (Fig. 1, B and C), there was no
change in its level at 12 h (Fig. 3A, lane 7). When BM
stroma was subjected to hypoxia and then allowed to recover in a
normoxic atmosphere for 8 and 12 h, HIF-1 mRNA levels were
increased by 15- (Fig. 3A, lane 2, and B) and
10-fold (Fig. 3A, lane 3, and B), respectively.
At a 16-h recovery, there was a 9-fold decrease in HIF-1 mRNA
compared with an 8-h recovery (Fig. 3A, lanes 2 and
4, and B). Hypoxic exposure of stromal cells in
the presence of 10 nM SP showed no significant change in HIF-1 mRNA
levels compared with unstimulated/normoxic stroma (Fig. 3A, lane
1) after 8 h (Fig. 3A, lane 5) and
12 h (Fig. 3A, lane 6). The results described in this
section show that hypoxia mediates an increase in HIF-1 protein and
the respective mRNA, and that SP through an NK-1R blunted the induction
of HIF-1 over baseline/unstimulated BM stroma.
|
SP blunted the increase of HIF-1 protein in BM stroma subjected
to hypoxia (Fig. 2B). This negative effect of SP on HIF-1
expression was also observed for baseline HIF-1 mRNA in normoxic
stroma stimulated with SP for 16 h (Fig. 1). Long-term hypoxic
exposure of BM mononuclear cells induced PPT-I (12), and
HIF-1 level is increased during hypoxia (13).
Therefore, we studied the relationship between hypoxia-mediated
induction of PPT-I and HIF-1 with the goal to understand how SP
blunts HIF-1 expression in BM stroma during reoxygenation after
hypoxia.
SP-IR levels were analyzed by Western blots and ELISA from stromal
extracts obtained from normoxic and hypoxic stroma with different
recovery times. Representative Western blot for SP-IR in seven
experiments, each with a different BM donor, is shown in Fig. 4A. As expected
(9), SP-IR bands were not detected in normoxic cultures
(Fig. 4A, lane 1). Hypoxic BM stroma that was recovered for
8 h (Fig. 4A, lane 2) and 12 h (Fig. 4A,
lane 3) showed no band for SP-IR. However, strong bands were
detected in cultures recovered for 16 h (Fig. 4A, lane
4) and 24 h (Fig. 4A, lane 5). Using a more
sensitive method, we quantitated SP-IR levels by ELISA and showed
significant increase in SP-IR levels in the hypoxic stromal cultures
that were recovered for different times up to 16 h (Table II). SP-IR levels were analyzed in
the cell extracts (cell-associated) and supernatants (release). The
former analyses were necessary because "soluble factors" released
from BM stroma could anchor to the surrounding extracellular matrix
proteins. Total SP-IR levels (shown in Table II) are shown as the sum
of the mean values of cell-associated and release.
|
|
B. The data indicated light bands when cultures were
recovered for 4 h (lane 2) and 8 h
(lane 3). Strong bands were observed after 12-h
(lane 4) and 16-h (lane 5)
recoveries. The level of PPT-I mRNA was decreased by 24 h
(lane 6). Consistent with the literature
(26), no band was detected for normoxic stroma
(lane 1). The results showed that during
reoxygenation of hypoxic stroma, the PPT-I gene is
expressed, but at a later time than for HIF-1. Activity of PPT-I promoter to hypoxia
The regulatory sequences within the 5' regions of the
PPT-I gene consisted of consensus sequences that bind
to hypoxia-inducible transcription factors, HIF-1 and NF-B (Refs.
26 and 33 , and Fig. 5A). Therefore, we analyzed
the effects of hypoxia on the different DNA regions of PPT-I, shown in
Fig. 5A. BM stroma was cotransfected with pgal-Control
and/or pGL3 containing PPT-1-1.2 (N0-Exon 1-Intron 1) and subfragments,
N0 or N0-Exon 1. After 24 h, the transfectants were subjected to
hypoxia and, 16 h later, luciferase and -gal activities were
determined. The results, shown in Fig. 5B, indicate no
change in stromal cells transfected with pGL3/N0. There were
significant increases in the activities of PPT-1-1.2 and N0-Exon 1 when
the transfectants were subjected to hypoxia reoxygenation.
|
and one for NF-B (26)
(Fig. 5A). PPT-I activity was increased when the Exon 1
and/or Intron 1 were ligated downstream of N0 (Fig. 5B). To
this end, studies were performed to determine whether hypoxia mediated
nuclear translocation of the transcription factors, HIF-1 and
NF-B, and whether these proteins could bind to the respective
consensus sequences within PPT-I-1.2. Immunoprecipitation was performed
for anti-HIF-1 or anti-RelA/p65 in nuclear extracts. Because
the sample buffer contained reducing agent (2-ME), the precipitating
IgG was observed at 55 kDa in the blots for both HIF-1 and RelA/p65
(data not shown). The only other bands observed were those of the
predicted sizes for HIF-1 (Fig. 5C) and RelA/p65 (Fig. 5D). There was a weak band after 1-h recovery for HIF-1
(Fig. 5C, lane 1) and strong bands at 4- and 8-h recovery
(Fig. 5C, lanes 2 and 3, respectively). No band
was observed for normoxic stroma (Fig. 5C, lane 4). Western
blots for p65/RelA showed no detectable band after 5 min (Fig. 5D, lane 1). Bands for p65/RelA were detectable at 10 min
(Fig. 5D, lane 2) and 20 min (Fig. 5D, lane 3).
The optimum level of p65/RelA was detected after 30 min (Fig. 5D,
lane 4), and its intensity was significantly reduced by 2 h
(Fig. 5D, lane 5, 1 h; lane 6, 2 h).
Nuclear extracts from normoxic stroma showed no detectable band (Fig. 5D, lane 7).
EMSA showed that the immunoreactive transcription factors detected in
the nuclear extracts of hypoxic stroma for HIF-1 (Fig. 5C) could bind to both consensus sequences within the
PPT-I gene (26) (Fig. 6, A–C). Gel and supershift
assays with nuclear extracts from stroma with 1- and 2-h recoveries in
normoxic atmosphere showed binding with HIF-1-1 and HIF-1-2 (Fig. 6A). The bands disappeared when EMSA was repeated with
nuclear extracts and excess cold competitor from 1-h recovery (Fig. 6B), and shifted when the extracts were preincubated with
two different sources of mAb to HIF-1 (Fig. 6C).
Preincubation of HeLa extracts with the Ab mixture showed no shift of
the band (data not shown). The results show that the two consensus
sequences for HIF-1 within the 5' regulatory regions of PPT-I could
interact with HIF-1.
Role of SP in the regulation of HIF-1 expression
The timeline studies shown in
Figs. 2–4 and the down-regulation
by SP on HIF-1 mRNA (Fig. 1) strongly suggest that PPT-I induction
and SP release might lead to autocrine feedback on the stromal cells to
down-regulate HIF-1 expression. To address this further, BM stroma
was transiently transfected with anti-sense -PPT-I, which was
ligated upstream of green fluorescent protein. The transfected stromal
cells were subjected to 30 min of hypoxia, and after 16 h in
normoxic condition the cell extracts were assayed for HIF-1 or SP
protein by Western blots. In the presence of anti-sense PPT-I,
HIF-1 protein was detected up to 48 h (Fig. 7A). However, in stromal cells
transfected with sense sequences, the band intensity was reduced after
16 h of recovery in normoxic atmosphere (Fig. 7A, lane
5). Sense transfectants showed no band after a 24-h recovery (Fig. 7A, lane 6). Consistent with reports in the literature
(9, 26), no HIF-1 protein was observed in extracts from
normoxic cells (Fig. 7A, lane 1).
|
B shows the results of studies in which SP was
analyzed in stromal cells transfected with PPT-I in the sense or
anti-sense orientations. At 12, 24, and 48 h, SP was
undetectable (Fig. 7B, lanes 2–4). During the period that
SP was undetectable, bands for HIF-1 were present (Fig. 7A,
lanes 2–4). In sense-transfected stroma, SP was produced at 16
and 48 h (Fig. 7B, lanes 5 and 6). During
the times that SP was present, HIF-1 protein level was decreased
(Fig. 7A, lanes 5 and 6). The results
showed a delay in the down-regulation of HIF-1 in stromal
cells that had -PPT-I transiently knockout. In contrast, stromal
cells that could express PPT-I (sense-transfected) showed detectable
HIF-1A. Role of SP on the activation of caspase-3
Hypoxia to the BM could be associated with necrosis and apoptosis.
Cell viability of the hypoxic stroma showed no evidence of trypan blue
incorporation (data not shown). Therefore, we examined hypoxic stroma
during various recovery times for indication of apoptosis. Levels of
active caspase-3 were performed by Western blots using anti-active
caspase-3. No band was detected in normoxic stroma (Fig. 8A, lane 1). Hypoxic stroma
with recovery times of 4, 8, and 12 h showed strong bands (Fig. 8A, lanes 2–4). At 16 h, active caspase-3 was not
detected (Fig. 8A, lane 5).
|
A), we
next determined whether SP could prevent activation of caspase-3.
Indeed the results showed no detectable active caspase-3 when the
stromal cells were subjected to hypoxia in the presence of SP after 4
and 8 h (Fig. 8A, lanes 6 and 7). This is in
contrast to stromal cells subjected to hypoxia in the absence of SP
(Fig. 8A, lanes 2 and 3). To further study the
role of SP on the activation of caspase-3, stromal cells were
transfected with a green fluorescent protein expression vector
containing the sense and anti-sense sequences of -PPT-I, and
then the stroma were subjected to 30 min of hypoxia. At different times
following hypoxia, cell extracts were studied for active caspase-3 by
immunoprecipitation. The results shown in Fig. 8B indicate
the presence of active caspase-3 at 16 h in the PPT-I knockout
stromal cells/anti-sense transfectants (Fig. 8B). In
contrast, no band was detected in extracts from hypoxic stroma with
sense transfectants and 16-h recovery (Fig. 8B). Active
caspase-3 was detected in extracts in which PPT-I was transiently
knockout for up to 16 h. The recovery time at which SP is released
(16 h) in hypoxic stroma that was transfected with sense
oligonucleotides (16 h, Fig. 7B) correlated with the
disappearance of active caspase-3 (Fig. 8B). The results
indicated endogenous and exogenous PPT-I peptides blunted
hypoxia-mediated activation of caspase-3. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and SP (shown in this report)
indicate that HIF-1 expression could be important to neurogenic
inflammation. This would be an additional function by HIF-1 and
PPT-I peptides, which are associated with inflammation (9, 34). The mixture of venous and arterial blood in the vicinity of the BM sinus explained the range of pO2 of BM aspirate between 56 and 58 mm Hg (our unpublished data). This level of blood gas in the BM indicates that a relatively small decrease in peripheral blood pO2 could result in a significantly lower O2 level in the BM. Therefore, this study, which exposes BM stroma to short-term hypoxia, showed part of the changes that occur in the BM during fluctuation of physiologic O2 levels.
A summary of this report is shown in Fig. 9. Unlike other tissues in which HIF-1
mRNA is unchanged during hypoxia, in BM stromal cells its level is
increased at 4 h after hypoxia (data not shown) and up to 12
h (Fig. 9). The changes in HIF-1 expression in stromal cells are
intriguing. Preliminary studies suggest that the accessory macrophages
mediate the difference in HIF-1 expression in the fibroblast
subset. The role of the other BM subsets such as the progenitors and
differentiated nonadherent cells is yet to be determined. These are
ongoing studies to understand HIF-1 during the loss of blood and
inflammation. HIF-1 protein was detected following an increase in
its mRNA (Fig. 9). There was a delay in the induction of the
PPT-I gene with its mRNA detectable at 12 h and its
protein (SP) at 16 h (Fig. 9). The release of SP correlates with
down-regulation of HIF-1 expression (Fig. 9). A role for SP in the
regulation of HIF-1 expression was shown in experiments performed
with hypoxic BM stroma in which the PPT-I gene was
transiently knockout, and the use of specific SP receptor (NK-1)
antagonists. The experimental conditions that blunted the expression of
PPT-I showed detectable HIF-1 protein up to 48 h of
reoxygenation following hypoxia (Fig. 7). Furthermore, the results also
showed that exogenous SP blunted the expression of HIF-1 (Fig. 2).
Together, these findings led to a working model in which delayed
release of SP mediates autocrine stimulation of the BM stroma leading
to a negative effect on the expression of HIF-1 and the activation
of caspase-3 (Fig. 9).
|
and 8). Activated caspase-3 has
been shown to be key in the differentiation of the erythroid lineage
(35). The role of this protease in erythropoiesis could be
clinically important, as hypoxia could be caused by hemorrhagic shock.
Therefore, the presence of active caspase-3 could serve dual functions:
induce apoptosis in reduced cellular O2 and
mediate erythropoiesis during blood loss.
Although consensus sequences for HIF-1 and NF-B are present
within the 5' regulatory regions of the PPT-I gene
(26) (Fig. 5A), and despite the presence of the
respective transcription factors in the nuclei of hypoxic cells (Fig. 5, C and D), the results did not prove that these
transcription factors are involved in the induction of PPT-I. These are
ongoing studies by our research group, and the results are beyond the
scope of this report. Such studies will determine the additional
benefit to the BM and the immune system by the delayed induction of
PPT-I following hypoxia. At this time, we could speculate
that PPT-I gene products, which exert multiple functions within the
neuroendocrine-immune-hemopoietic axis (9), might benefit
distant organs following critical injuries associated with hemorrhagic
shock. Other gene products within this axis cannot be eliminated and
would require further studies.
Overall, this report, summarized in Fig. 9, brings an understanding of
PPT-I induction following long-term (24 h) hypoxia (12).
The results reinforce the ability of the "plastic" biological
feature of the stromal cells to regulate homeostasis in the BM.
Although this study shows PPT-I as a negative feedback on
HIF-1 and active-caspase-3, the presence of PPT-I peptides in the BM
could be important for the regulation of relevant hemopoietic
regulators such as cytokines (9, 30, 36). This would be
consistent for PPT-I peptides as hemopoietic regulators (9, 37, 38, 39) and the association of hypoxia with immune and blood
cell replacement (10, 17, 39). It is tempting to speculate
that these results might partly explain the mechanism in which the more
quiescent BM progenitors are preserved during hypoxia
(40). The role of proinflammatory cytokines on the
expression of HIF-1 to regulate hypoxia-sensitive genes is different
in normoxic and hypoxic conditions (41). Thus, the
findings in this study contribute to the physiology of cells following
relative hypoxic exposure to the BM microenvironment. These results
could not otherwise be deduced from reports with cells in normoxic
condition.
|
Footnotes |
|---|
2 K.R. and A.M. contributed equally to this work.
3 Address correspondence and reprint requests to Dr. Pranela Rameshwar, University of Medicine and Dentistry of New Jersey-New Jersey Medical School, 185 South Orange Avenue, MSB, Room E-579, Newark, NJ 07103. E-mail address: rameshwa{at}umdnj.edu
4 Abbreviations used in this paper: BM, bone marrow; SP, substance P; PPT-I, preprotachykinin-1; HIF-1, hypoxia-inducible factor-1; NK-1, neurokinin-1; -gal, -galactosidase; IR, immunoreactive.
Received for publication April 25, 2001. Accepted for publication August 14, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ubiquitination by a reconstituted von Hippel-Lindau (VHL) tumor suppressor complex. Proc. Natl. Acad. Sci. USA 97:10430. under the influence of nitric oxide. Blood 97:1009.B as a central mediator in the induction of TGF- in monocytes from patients with idiopathic myelofibrosis: an inflammatory response beyond the realm of homeostasis. J. Immunol. 165:2271.-preprotachykinin, the common precursor to substance P and neurokinin-A. FEBS Lett. 208:67.[Medline]
and tumor necrosis factor- stimulate DNA binding of hypoxia-inducible factor-1. Blood 94:1561.This article has been cited by other articles:
|
|
 |
|
  K. E Corcoran, A. Malhotra, C. A Molina, and P. Rameshwar Stromal-derived factor-1{alpha} induces a non-canonical pathway to activate the endocrine-linked Tac1 gene in non-tumorigenic breast cells J. Mol. Endocrinol., March 1, 2008; 40(3): 113 - 123. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Nowicki, D. Ostalska-Nowicka, B. Kondraciuk, and B. Miskowiak The significance of substance P in physiological and malignant haematopoiesis J. Clin. Pathol., July 1, 2007; 60(7): 749 - 755. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. E. Corcoran and P. Rameshwar Nuclear Factor-{kappa}B Accounts for the Repressor Effects of High Stromal Cell-Derived Factor-1{alpha} Levels on Tac1 Expression in Nontumorigenic Breast Cells Mol. Cancer Res., April 1, 2007; 5(4): 373 - 381. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Y. Cheung, L. Bogic, R. Gagnon, R. Harding, and R. A. Brace Morphologic Alterations in Ovine Placenta and Fetal Liver Following Induced Severe Placental Insufficiency Reproductive Sciences, December 1, 2004; 11(8): 521 - 528. [Abstract] [PDF]
|
|
|
|
 |
|
 O. Eickelberg, F. Seebach, M. Riordan, G. Thulin, A. Mann, K. H. Reidy, S. K. Van Why, M. Kashgarian, and N. Siegel Functional Activation of Heat Shock Factor and Hypoxia-Inducible Factor in the Kidney J. Am. Soc. Nephrol., August 1, 2002; 13(8): 2094 - 2101. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, Induction of
the particular gene with double arrows representing the relative
induction. 
, Down-regulation of the respective gene. At time 0,
HIF-1