Articles in PresS. Am J Physiol Lung Cell Mol Physiol (May 7, 2004). 10.1152/ajplung.00203.2003
Greater vascularity, lowered HIF-1/DNA binding and elevated
GSH as markers of adaptation to in vivo chronic hypoxia.
1
Tissot van Patot, M.C., 1Bendrick-Peart, J., 1Beckey, V.E., 1Serkova, N., 2Zwerdlinger,
L.
1
University of Colorado Health Sciences Center, Dept. of Anesthesiology, Denver,
Colorado, 80262, 2St. Vincent’s General Hospital, Leadville, Colorado, 80461.
Corresponding Author Information
Martha C. Tissot van Patot
Dept. Anesthesiology, B-113
University of Colorado Health Sciences Center
4200 E. 9th Avenue
Denver, CO 80262
Phone: 303 315-1869
Fax: 303 315-1899
email: martha.tissotvanpatot@uchsc.edu
Running Head: Adaptation to Chronic Hypoxia
Copyright © 2004 by the American Physiological Society.
Abstract
Vascularity is increased in placentae from high as compared to low altitude pregnancies.
An angiogenic response to hypoxia may protect an organ from further hypoxic insult by
increasing blood flow and oxygen delivery to the tissue. We hypothesized that increased
placental vascularity is sufficient to adapt to high altitude. Therefore, indices of hypoxic
stress would not be present in placentae from successful high altitude pregnancies.
Methods: Full-thickness placental biopsies were A) collected and frozen in liquid
nitrogen within 5 minutes of placental delivery and B) fixed in formalin for stereologic
analyses, at high (3100 m, n = 10) and low (1600 m, n = 10) altitude. HIF-1 activity was
analyzed by enzyme-linked immunoabsorbance assay (ELISA). Western blot analyses
were used to evaluate HIF-1!, HIF-1", HIF-2!, von Hippel-Lindau protein (pvHL),
VEGF, Flt-1, enolase, and GAPDH. Magnetic resonance spectroscopy (MRS) was used
to evaluate endogenous metabolism. Results: The ratio of placental capillary surface
density to villous surface density was 70% greater at high as compared to low altitude.
HIF-1 activity and HIF-1 associated proteins were unchanged in placentae from high vs.
low altitude pregnancies. Placental expression of HIF-1-mediated proteins VEGF, Flt-1,
enolase, and GAPDH were unchanged at high vs. low altitude. Succinate, GSH,
phosphomonoesters and ADP were elevated in placenta from high as compared to low
altitude. Conclusion: Placentae from uncomplicated high altitude pregnancies have
greater vascularity and no indication of significant hypoxic stress at term as compared to
placentae from low altitude.
Key Words: GSH, succinate, HIF-1-DNA binding, placenta, stereology
2
Introduction
Previously, we and others have reported increased vascularity in placentae that have
developed during healthy pregnancies at high as compared to low altitude (3, 17, 30). A
hypoxia-induced increase in vascularity is believed to assist in ‘rescuing’ the tissue from
hypoxia by increasing blood and thereby oxygen delivery to the area. Hypoxia can
activate hypoxia-inducible transcription factor (HIF-1) (6, 7). HIF-1 enhances
transcription of genes encoding hypoxia-sensitive proteins that are instrumental in
protecting the tissue from hypoxia, such as; erythropoietin, which increases the oxygen
carrying capacity of the blood, vascular endothelial growth factor (VEGF) and its
receptor Flt-1, which stimulate angiogenesis, and glycolytic proteins such as
glyceraldehydes 3-phosphate dehydrogenase (GAPDH) and enolase, which increase
energy production through anaerobic glycolysis (8, 16, 25, 31).
If the response to hypoxia is successful, for example vascularity (oxygen delivery) is
increased sufficiently to protect the tissue from hypoxia, the principle of negative
feedback dictates that HIF-1 activity, expression of hypoxia-sensitive proteins, oxidative
stress and increased glycolysis should be attenuated. Because placental vascularity is
established in early pregnancy, a hypoxia or altitude-induced increase in vascularity
probably occurs in early pregnancy as well (1). Therefore, establishing an increase in
oxygen delivery early in gestation should protect the placentae from hypoxia so that term
placentae of successful high altitude pregnancies should not be associated with markers
of hypoxic stress.
3
Thus, we hypothesized that greater vascularity in placentae from uncomplicated
pregnancies at high altitude would not be associated with markers of hypoxic metabolic
stress including enhanced HIF-DNA binding, expression of HIF-related and hypoxiasensitive proteins and metabolic hypoxic markers. Our approach was to examine
placentae from high- and low-altitude pregnancies in order to determine vascular
responses including; capillary and villous surface densities, HIF-DNA binding activity,
the presence of proteins associated with HIF-1 activation, including HIF-1!, HIF-2!,
HIF-1", and pvHL, and the expression of HIF-mediated proteins, VEGF, Flt-1. We also
determined metabolic stress markers including; GAPDH, enolase and metabolic
adaptation at each altitude. This strategy was designed to determine whether enhanced
placental vascularity in hypoxic placentae is associated with indicators of hypoxic stress.
This study is important because hypoxia is implicated in the pathogenesis of many
pregnancy complications including preeclampsia, intra-uterine growth restriction and
anemia, and poses a serious threat to the health of both fetus and mother (14). Fetal
growth and development are impaired in the presence of hypoxia. However, successful
pregnancy at high altitude represents successful adaptation to hypoxia. Therefore,
determining the mechanisms of successful adaptation to chronic hypoxia during
pregnancy and markers of adaptation failure may be a critical first step in determining
successful therapeutic intervention in hypoxia-mediated diseases of pregnancy.
Materials and Methods
Study Design. Approval from the Colorado Multiple Institutional Review Board
4
(COMIRB) at the University of Colorado Health Sciences Center was obtained to collect
term placentae from uncomplicated, singleton gestational women at University Hospital
in Denver, Colorado (1600 m). Permission was also obtained from St. Vincent’s Hospital
in Leadville, Colorado (3100 m). Ten subjects at low altitude and ten subjects at high
altitude were consented according to COMIRB guidelines and placentae were collected
immediately following placental delivery. Subjects were between the ages of 18-34 and
gestational age was 39-41 weeks. Two g of tissue (full thickness of the placentae,
including basal and chorionic plates) were collected from random locations and placed in
liquid nitrogen within 5 minutes of placental delivery, to stop the metabolic activity.
Previous data from our lab (26) indicates that placental samples must be placed into
liquid nitrogen within 9 minutes of placental delivery to avoid introducing artifact by
hypoxic/ischemic induction of glycolysis. The placenta was then dissected into 5 sections
from which 2 blocks extending from basal plate to chorionic plate were dissected, placed
into 10% formalin for 4 days and then paraffin embedded at less than 58oC for
stereologic analyses.
Immunhistochemistry. Placental sections were stained to label endothelium using a
polyclonal mouse anti-human CD34+ antibody (1:20; BioGenex, Napa, CA, USA),
followed by an ABC reagent (ABC, Vectastain Elite, Vector Labs, Burlingame, CA,
USA) labeled with peroxidase for which 3’ 3’-diaminobenzidine (Sigma, St. Louis, MO)
was used as a substrate. Negative controls were performed using mouse IgG in place of
primary antibody.
5
Stereologic Analysis. Stereology was performed as previously reported (30). Briefly, 4
slides from 4 sections (4 !m) of each placenta, and 16 fields per slide were evaluated. A
25-box microscope grid (25 X 25 !m) was used under 40X magnification; box size was
chosen to minimize the number of capillaries per box. The guidelines of a 125 !m2 (25
!m2 per box) grid were applied and capillary surface density of capillary luminal margins
(Vvcap) was calculated by 2*total capillary intersects/ number of points on villous tissue
* total test line length (= 2D, where D = distance between points on grid (25 !m)) (3).
The capillary (and villous) intersects were calculated by counting the villous tissue each
time it crossed a horizontal line within the square lattice parameters specified (3). Villous
surface density (Svvill) was calculated in a similar manner, using the villous intersects as
the numerator. A ratio of Svcap/Svvill is reported.
Western blot. For immunoblot assays, 30 µg total protein/lane or 50 µg nuclear
protein/lane were fractionated by electrophoresis using a NUPAGETM 4-12% bis-tris
gradient gel (Invitrogen, Carlsbad, CA). The proteins were then transferred to a
methanol-soaked polyvinylidene difluouride (PVDF) membrane using the semidry
immunoblot method (Owl Model HEP-1 Panther Semi-Dry Electroblotter – Nunc,
Rochester, NY). The membranes were immunoblotted using either HIF-1!, HIF-1", HIF2!, (Nuclear proteins, Novus Biologicals, Littleton. Colorado: NB100-105, 100-124,
100-132), pvHL, VEGF or Flt-1 (Santa Cruz Biotechnology, Santa Cruz, CA, Fl-181, C1, H-225), secondary antibodies conjugated with horseradish peroxidase (IgG-HRP) were
used for detection and visualization by Pierce-SuperSignalRcircle West Dura Extended
Duration Substrate (Pierce Biotechnology, Inc., Rockford, IL). Images were visualized
6
using the UVP BioChemi Imaging System and relative quantification by densitometry
was performed using LabWorks 4.0 software (UVP, Inc., Upland, CA). !-actin (Sigma,
St. Louis, MO, A-5441) was used as an internal control for protein loading and data are
expressed as a ratio of the protein of interest to !-actin.
Enzyme-linked immunoabsorbance assay (ELISA). An ELISA was used kit to assess HIF1 activity on all samples in a single experiment (BD Biosciences, Clontech, Palo Alto,
CA, K2077-1). Briefly, nuclear extract (20 !g) was added to a 96-well plate coated with
the DNA consensus binding sequence for HIF-1. Bound HIF-1 was detected by the
addition of mouse monoclonal primary antibody to HIF-1", followed by horseradish
peroxidase-conjugated secondary antibody. A microtiter plate reader (ThermoLab
Systems, Helsinki, Finland, Multiskan Ascent) was used to measure the enzymatic
product. A HIF-1 wild-type competitor oligonucleotide control was used to demonstrate
DNA-HIF-1 binding specificity. Samples were run in duplicate and CV values were less
than 10%.
Dual Perchloric Acid (PCA) lipid extraction of placental tissues.
To perform high-resolution magnetic resonance spectroscopy (MRS) on placental tissues,
we extracted the frozen placental samples using a dual perchloric acid (PCA)/lipid
extraction procedure developed in our laboratory (27). Snap frozen tissues were
powdered in a mortar grinder in the presence of liquid nitrogen. The powdered frozen
tissue was added to 6 ml of ice-cold 12% PCA and subsequently homogenized using
electrical homogenizer Poly Tron PT 2100 (Kinematica AG, Luzern, Switzerland). The
PCA homogenates were put into an ice-cold ultrasound bath for 5 min. Then, the
7
homogenates were centrifuged at 3000 x g and 4oC for 20 min. The aqueous phase was
collected, and the pellet was re-suspended with 2 ml of ice-cold PCA. The resuspended
homogenates were put in an ultrasound bath and centrifuged again in the same
conditions. The aqueous phase was added to the previously collected supernatant. The
supernatants, containing placental water-soluble metabolites, were then neutralized with
KOH, centrifuged for 20 min at 3000 x g and 4oC to remove potassium perchlorate, and
lyophilized overnight for PCA extracts. The tissue pellets, remaining after the first
centrifugations, were re-dissolved in 4 ml ice-cold water. The re-dissolved pellets,
containing placental lipids, were neutralized with KOH and lyophilized overnight for
lipid extracts. The lyophilized PCA extracts, containing water-soluble metabolites, were
reconstituted in 0.45 ml deuterium oxide (D2O, Cambridge Isotope Laboratories Inc.,
Andover, MA). The lyophilized lipid extracts were reconstituted in 1.5 ml of
deuterated chloroform/methanol mixture (CDCl3/CD3OD, 2 : 1 vol/vol). After
centrifugation, the supernatants were analyzed by MRS.
Magnetic Resonance Spectroscopy (MRS) on PCA and lipid extracts.
To calculate the absolute concentrations of water-soluble and lipid metabolites, one
dimensional MRS experiments were carried out using a 500 MHz Bruker NMR
spectrometer with an Avance console (Bruker, Karlsruhe, Germany). A dual QNP 5-mm
Bruker probe head was used for all experiments. For proton MRS, the operating
frequency was 500 MHz, and a standard pre-saturation pulse program was used for water
suppression. The other parameters were: 40 accumulations; 90o pulse angle; 0 dB power
level; 7.35 !s pulse width; 10 ppm spectral width; 12.85 s repetition time. Trimethylsilyl
8
propionic-2,2,3,3,-d4 acid (TMSP, 0.6 mmol/l for PCA extracts and 1.2 mmol/l for lipid
extract) was used as an external standard for the quantification of metabolites based on
1
H-MRS signals. 1H chemical shifts were referenced to TSP at 0 ppm. Before the 31P -
MRS experiments were recorded, 100 mmol/l EDTA was added to each PCA extract for
complexation of divalent ions. This resulted in 31P peaks with significant narrow line
width (especially important for ATP signals). The pH was adjusted again to 7. The
following NMR parameters with a composite pulse decoupling (CPD) program were
used: 202.1 MHz operating 31P frequency; 800 accumulations; 90o pulse angle; 12 dB
power level for 31P channel; 9 !s pulse width; 35 ppm spectral width; 2.0 s repetition
time. The absolute concentration of glycerophosphocholine (GPC), calculated from 1HMRS of the same extract, was used as an internal standard for quantification of
phosphorus metabolites in 31P -MR spectra. The chemical shifts of !-ATP at -10 ppm
were used as shift references. All MRS data were processed using the 1D WINNMR
programme (Bruker, Karlsruhe, Germany).
9
Statistical Analyses. Stereology, densitometry and ELISA data were analyzed using a
Student t-Test. MRS data were analyzed by ANOVA. Scheffe’s Post-hoc test to was used
to determine differences between variables. Significance for all statistical analyses was
accepted at p ! 0.05. Data are presented as representative immunoblots indicating subject
number in each lane and accompanied by densitometry analysis of immunoblots for the
entire study group (n = 10 per altitude).
Results
The characteristics of the subjects at1600 and 3100 M were similar with respect to
maternal age, gestational age of delivery, birth weight, placental weight, placental
volume and the ratio of placental weight to birth weight (Table 1).
Placental stereologic analyses indicated that the ratio of capillary surface density (Svcap)
to villous surface density (Svvill) was greater in high- vs. low-altitude pregnancies (low
altitude 5.65, high altitude 9.52, p = 0.03) (30) (Figure 1D). In order to determine if
greater vascularity protected the placentae from the hypoxic stress of high altitude, we
determined hypoxia-inducible transcription factor-1 (HIF-1)/DNA binding activity in
placentae collected at high and low altitude. ELISA results indicated that HIF-1/DNA
binding was 1.455 ng/ml (± 0.182) at low and 0.476 ng/ml (± 0.347) at high altitude, 3
fold less at high altitude (Figure 2) (p = 0.0001).
Because HIF-1 activity is primarily determined by the presence of the activated HIF-1!
subunit in the nucleus, we determined the presence of nuclear HIF-1! by Western blot
analyses. Nuclear HIF-1! was equivalent in placentae from low (n = 10) and high (n =
10
10) altitude pregnancies, as determined by densitometric analyses (Figure 3). To further
investigate reasons for less HIF-1 activity at high altitude, we determined nuclear HIF1!, to which HIF-1" must bind in order to create active HIF-1. Nuclear HIF-1! was
greater in placentae from high- (n = 10) as compared to low- (n = 10) altitude
pregnancies, as determined by densitometric analyses (Figure 4). However, individual
densitometry data indicated that placenta #3 expressed 2-3 fold more HIF-1! than other
placentae (Figure 4). When data were compared without placenta #3 there was no
significant difference in HIF-1! expression between low- and high-altitude placentae.
The von Hippel-Lindau protein (pvHL) binds to HIF-1" during normoxia, targeting HIF1" for ubiquitination and proteosomal degradation. In the presence of normoxia, as
suggested by unaltered HIF-1" at high altitude, there should be no difference in pvHL at
high as compared to low altitude. In Western blot analysis of total placental protein
extracts from 10 low- and 10 high-altitude pregnancies, there were, as expected, no
differences in pvHL expression (Figure 5). Thus, less HIF-1 activity in placentae from
high- vs. low-altitude pregnancies was not associated with any change in the presence of
HIF-1" or –1! subunits or expression of pvHL.
In some tissues, HIF-2 is activated under less severe hypoxic conditions than HIF-1 and
may be responsible for increased transcription of hypoxia-sensitive genes (32). To
determine if placental tissue at high altitude was responding to hypoxia via HIF-2 rather
than HIF-1, we analyzed placentae from 10 low- and 10 high-altitude pregnancies for
nuclear HIF-2" by Western blot analyses (Figure 6). HIF-2" was not different between
11
placental nuclear extracts from low and high altitude pregnancies, although there was a
trend toward greater expression at high altitude, which may prove significant in a larger
sample of placentae.
To further test whether placentae at high altitude are experiencing hypoxic stress
expression of hypoxia-sensitive angiogenic and glycolytic proteins was analyzed.
Expression of VEGF, a hypoxia-sensitive angiogenic protein, was not altered in placentae
from high (n = 10) as compared to low (n = 10) altitude as determined by Western blot
and densitometric analysis (Figure 7).
Because a hypoxia-stimulated increase in the VEGF receptor, Flt-1, could increase the
biological activity of VEGF, further Western blot analyses for Flt-1 expression in 10 lowand 10 high- altitude placentae were performed and densitometric analysis indicated that
expression did not differ between placentae from high- vs. low-altitude pregnancies
(Figure 7).
Due to the fact that cellular response to hypoxia initiates an increase in expression of
glycolytic enzymes and hence glycolysis, we determined the expression of glycolytic
enzymes enolase and GAPDH placentae from low (n = 10) and high (n = 10) altitude
pregnancies by Western blot and subsequent densitometric analyses. There was no
change in the expression of placental enolase and GAPDH between low and high altitude
pregnancies (Figure 8).
12
Because there was no evidence of HIF-1 activation or glycolytic stress, we sought to
determine if there were metabolic markers for hypoxia at high as compared to low
altitude, utilizing magnetic resonance spectroscopy (MRS). 1H and 31P- MRS analysis
indicated that metabolites succinate, glutathione-SH (GSH), phosphomonoesters (PME)
and ADP were increased in placentae collected at high altitude (Table 2). Thus the ratio
of PME to phosphodiesters (PDE) was also increased; however the ATP/ADP ratio did
not change, since ATP also showed a tendency to increase. No significant increase in
lactate, a marker for anaerobic glycolysis, was seen. There were no changes in polyunsaturated fatty acid (PUFA) concentrations in lipid spectra of high vs. low altitude
placentae, indicating no increase in lipid peroxidation (Table 3).
Discussion
In the present study, the main finding was that placentae from successful pregnancies
exposed to hypoxia through out gestation had greater placental vascularity and no
evidence of severe hypoxic stress. Placental HIF-1/DNA binding activity was actually
lower at altitude and there was no increase in the expression of hypoxia-sensitive
proteins. Interestingly, the reduction in HIF-1 activity was not associated with changes in
pvHL, nuclear HIF-1!, HIF-1", nor was HIF-2! increased in the nucleus. In regard to
metabolism, succinate was elevated as is often found during hypoxia, however GSH was
also elevated. Furthermore, there was no evidence of lipid peroxidation or glycolytic
activity.
The reduction in HIF-1 activity was not caused by introducing hypoxia during placental
collection, as all placental samples were minced and collected into liquid nitrogen within
13
5 minutes of vaginal delivery and stored at –80o C until nuclear proteins were extracted
for analyses. Further, MRS analysis did not indicate acute hypoxic insult in any of the
tissues since no increase in lactate nor decrease in the ATP/ADP ratio or glucose was
seen. Each Western blot was performed a minimum of 3 times, and each ELISA sample
was analyzed in triplicate (coefficient of variance less than 10%), producing consistent
results each time. The number of subjects within each group (n = 10 low-, n = 10 highaltitude placentae) was sufficient to achieve 98% power, showing less HIF-1 activity at
high altitude.
The investigators are aware that Denver is not sea level. Low-altitude placentae in this
study were collected at 1600 m, where the partial pressure of inspired oxygen (PiO2) is
122 mmHg as compared to 149 mmHg at sea level. Although, the change in PiO2 at 1600
m is not enough to cause altitude-induced illness (13), there may be metabolic, enzymatic
or protein changes in tissue without associated clinical complications. Therefore, future
studies examining sea level placental tissue in comparison to those at 1600 and 3100 m
are planned.
Because these placentae were collected following labor and delivery, it is possible that
changes in HIF-1 activity occurred as a result of labor and delivery and did not reflect in
vivo values prior to labor. Ideally, placentae from Cesarean sections should be used to
most closely assess in vivo values. This was a potential problem for our collections in
Leadville, Colorado, as all pre-planned Cesarean sections are performed in one of several
low-altitude facilities, making Cesarean section material impractical for collection in our
study design. However, in a large number of placentae from high altitude pregnancies all
had less HIF-1 activity suggesting that the results are reliable.
14
Because HIF-1 is important in promoting transcription of VEGF, Flt-1, GAPDH and
enolase and HIF-1 activity was not greater at high as compared to low altitude, it is not
surprising that there was no increase in the expression of hypoxia-sensitive proteins.
Although VEGF and Flt-1 are important for promoting vasculogenesis (2, 15, 21), it is
not surprising that there was no increased expression in the highly vascular placentae.
Placental vascular development occurs primarily during the first and early second
trimesters and is most likely no longer taking place at term. Furthermore, failure to
increase expression of glycolytic enzymes GAPDH and enolase at high altitude was
supported by MRS data indicating no change in glucose, lactate or ATP in high altitude
placentae. Although, with the glycolytic enzymes, protein expression may not change
whereas the activity of the enzyme might; changes in molecules such as ADP suggest this
may be the case.
Metabolic profiles of our high altitude placental tissue did not reveal changes which are
characteristic for acute short term hypoxia (Table 3) (4, 22, 28, 29). This suggests that
placentae from high altitude uncomplicated pregnancies are not exposed to hypoxic
conditions (possibly due to increased vascularity), but rather showed metabolic
adaptation to increased oxygen delivery. Phosphomonoestsers, precursors for membrane
phospholipids, were increased, indicating increased membrane synthesis, in accordance
with the data indicating greater capillary development in high altitude placentae. ADP
was elevated, but the ATP/ADP ratio indicating energy state was unchanged. ATP may
be converted to ADP more quickly to provide energy for enhanced membrane synthesis.
15
Previous studies on hypoxic tissue indicate that succinate is increased during hypoxia,
while GSH is reduced (4, 18, 28). During hypoxia, complex II in the mitochondrial
respiratory chain appears to switch from succinate dehydrogenase to fumerate reductase,
resulting in an accumulation of succinate (18). Because succinate is produced in the
mitochondrial respiratory chain, succinate concentrations are dependent on tissue PO2
(18). Glutathione-SH (GSH) is reduced during hypoxia by conversion to glutathione-S-Sglutatione (GSSG) and is more dependent on oxygen content than PO2 (4). Previous
studies in which succinate was greater and GSH lower during hypoxia were designed
such that PO2 and oxygen content were reduced. In contrast, in the current study
placentae from high altitude pregnancies most likely experienced lowered PO2 as a result
of hypobaric hypoxia, but greater oxygen content as a result of greater vascularity.
Therefore, we propose that the greater succinate concentration was due to the lower PO2
and the greater GSH concentration was due to greater oxygen content. Also, the equal
concentrations of PUFA in low and high altitude placentae indicated no evidence of
increased LPO (30), related to hypoxia.
Because GSH has been reported to reduce HIF-1 activity when exogenously administered
to hypoxic tissue (10, 11), greater GSH concentrations may have inhibited the activity of
HIF-1 in our high- as compared to low-altitude placentae. Our findings regarding HIF
activity do not dispute what others have found; that the placentae can increase HIF
activity when hypoxic and have greater HIF-1! early rather than late in pregnancy (5, 9,
23, 24). Canniggia et al. (5) reported that in normal, low-altitude pregnancies placental
16
HIF was elevated in early gestation (prior to 10 weeks) during the most severe placental
hypoxia, but decreased as the placenta invaded the uterus and was exposed to maternal
circulation. The altitude-induced increase in placental vascularity probably occurred
during the same early gestation time period; however, that could not be determined at this
time as sampling of early pregnancy placentae was not possible in our study.
Data from the literature report HIF consistently elevated during hypoxia in pathologic
conditions (25). Similarly, our current study reports no increase in HIF activity in
placentae from women who successfully completed pregnancy at high altitude. Therefore,
we consider that lack of HIF activity in placentae collected at high altitude does not
represent a pathologic condition but is rather a surrogate marker for successful adaptation
to high altitude. Our data suggesting that reduced HIF-1 activity represents adaptation to
high altitude are supported by a report from Hochachka and Rupert (12), in which
Andean natives had lower erythropoietin synthesis in response to hypoxia than
lowlanders, however the genetic sequences encoding erythropoietin and HIF-1! were
unchanged in the native population. Hochachka and Rupert hypothesized that their results
suggested ‘that the altered erthropoietic response in Andean natives reflects adaptations
in hypoxia sensing, rather than hypoxia response, mechanisms’ (12).
For example, failure to attain normal pregnancy at high altitude most often results in
preeclampsia, at rates 3-4 times those of low-altitude pregnancies (19) and HIF is
elevated 1.5 – 2.5 fold in placentae from preeclamptic pregnancies at low altitude (24).
Preeclampsia is characterized by greatly elevated blood pressure, poor placental
17
development and impaired placental blood flow (20). In contrast, the current study
indicates greater placental vascular development and less HIF activity in placentae from
normal pregnancies at high altitude.
Long-term hypoxia is a complication of many diseases including pregnancy-related
disorders, pulmonary and cardiovascular diseases. Determining the mechanisms by which
tissues successfully adapt to chronic hypoxia is crucial for survival of tissues challenged
by chronic hypoxia. Our data suggest that greater GSH concentration and less HIF
activity may be implicated in the mechanism of successful adaptation to chronic hypoxia.
However, our study design did not allow for establishing cause-effect relationships. It
remains to be evaluated, if the observed changes are evidence of adaptation or markers
for a yet unidentified mechanism.
Acknowledgments
The authors would like to thank Dr. John Reeves and Dr. Uwe Christians for their
invaluable mentorship and assistance in preparing this manuscript. We would also like to
acknowledge the invaluable assistance of the nursing staff on the labor and delivery
wards of University Hospital, Denver and St. Vincent’s General Hospital, Leadville,
Colorado without whom this study would not have been feasible. Also, we would like to
thank St. Vincent’s General Hospital for providing excellent facilities and a supportive
collaborative environment in which to work.
References
1.
Benirschke K and Kaufmann P. Pathology of the Human Placenta. New York:
Springer-Verlag, 1995.
2.
Brekken RA and Thorpe PE. Vascular endothelial growth factor and vascular
targeting of solid tumors. Anticancer Research 21: 4221-4229, 2001.
18
3.
Burton GJ, Reshetnikova OS, Milovanov AP, and Teleshova OV.
Stereological evaluation of vascular adaptations in human placental villi to differing
forms of hypoxic stress. Placenta 17: 49-55, 1996.
4.
Caceda R, Gamboa JL, Boero JA, Monge CC, and Arregui A. Energetic
metabolism in mouse cerebral cortex during chronic hypoxia. Neuroscience Letters 301:
171-174, 2001.
5.
Caniggia I, Mostachfi H, Winter J, Gassmann M, Lye SJ, Kuliszewski M,
and Post M. Hypoxia-inducible factor-1 mediates the biological effects of oxygen on
human trophoblast differentiation through TGFbeta(3). Journal of Clinical Investigation
105: 577-587, 2000.
6.
Chandel NS, McClintock DS, and Feliciano CE. Reactive oxygen species
generated at mitochondrial complex III stabilize hypoxia-inducible factor-1alpha during
hypoxia: a mechanism of O2 sensing. Journal of Biological Chemistry 275: 2513025138, 2000.
7.
Ehleben W, Bolling B, Merten E, Porwol T, Strohmaier AR, and Acker H.
Cytochromes and oxygen radicals as putative members of the oxygen sensing pathway.
Respiration Physiology 114: 25-36, 1998.
8.
Forsythe JA, Jiang BH, Iyer NV, Agani F, Leung SW, Koos RD, and
Semenze GL. Activation of vascular endothelial growth factor gene transcription by
hypoxia -inducible factor 1. Molecular and Cell Biology 16: 4604-4613, 1996.
9.
Genbacev O, Krtolica A, Kaelin W, and Fisher SJ. Human cytotrophoblast
expression of the von Hippel-Lindau protein is downregulated during uterine invasion in
situ and upregulated by hypoxia in vitro. Developmental Biology 233: 526-536, 2001.
10.
Haddad JJ and Land SC. O(2)-evoked regulation of HIF-1alpha and NFkappaB in perinatal lung epithelium requires glutathione biosynthesis. Am J Physiol Lung
Cell Mol Physiol 278: L492-503, 2000.
11.
Haddad JJ, Olver RE, and Land SC. Antioxidant/pro-oxidant equilibrium
regulates HIF-1alpha and NF-kappa B redox sensitivity. Evidence for inhibition by
glutathione oxidation in alveolar epithelial cells. Journal of Biological Chemistry 275:
21130-21139, 2000.
12.
Hochachka PW and Rupert JL. Fine tuning the HIF-1 'global' O2 sensor for
hypobaric hypoxia in Andean high-altitude natives. Bioessays 25: 515-519, 2003.
13.
Hultgren HN. High Altitude Medicine. Stanford, CA: Hultgren Publications,
1997.
14.
Kingdom JCP and Kaufmann P. Oxygen and placental villous development:
Origins of fetal hypoxia. Placenta 18: 613-621, 1997.
15.
Kumazaki K, Nakayama M, Suehara N, and Wada Y. Expression of vascular
endothelial growth factor, placental growth factor, and their receptors Flt-1 and KDR in
human placenta under pathologic conditions. Human Pathology 33: 1069-1077, 2002.
16.
Lu S, Gu X, Hoestje S, and Epner DE. Identification of an additional hypoxia
responsive element in the glyceraldehyde-3-phosphate dehydrogenase gene promoter.
Biochimica et Biophysica Acta 1574: 152-156, 2002.
17.
Mayhew TM. Changes in fetal capillaries during preplacental hypoxia: Growth,
shape remodelling and villous capillarization in placentae from high-altitude pregnancies.
Placenta 24: 191-198, 2003.
19
18.
Paddenberg R, Goldenberg A, Faulhammer P, Braun-Dullaeus RC, and
Kummer W. Mitochondrial complex II is essential for hypoxia-induced ROS generation
and vasoconstriction in the pulmonary vasculature. 163-169, 2003.
19.
Palmer SK, Moore LG, Young DZ, Cregger B, Berman JC, and Zamudio S.
Increased preeclampsia and altered blood pressure course during normal pregnancy at
high (3100 m) altitude in Colorado. American Journal of Obstetrics and Gynecology
18O: 1161-1168, 1999.
20.
Pridjian G and Puschett JB. Preeclampsia. Part 1: clinical and pathophysiologic
considerations. Obstetrical & Gynecological Survey 57: 598-618, 2002.
21.
Pufe T, Petersen W, Tillmann B, and Mentlein R. The angiogenic peptide
vascular endothelial growth factor is expressed in foetal and ruptured tendons. Virchows
Archiv 439: 579-585, 2001.
22.
Punkt K, Welt K, and Schaffranietz L. Changes of enzyme activities in the rat
myocardium caused by experimental hypoxia with and without ginkgo biloba extract
EGb 761 pretreatment. A cytophotometrical study. Acta Histochemica 97: 67-79, 1995.
23.
Rajakumar RA and Conrad KP. Expression, ontogeny and regulation of
hypoxia inducible transcription factors in the human placenta. Journal of the society for
Gynecologic Investigation 7: 184A, 2000.
24.
Rajakumar RA, Whitelock KA, Weissfeld LA, Daftary AR, Markovic N, and
Conrad KP. Selective overexpression of the hypoxia-inducible transcription factor, HIF2!, in placentas from women with preeclampsia. Biology of Reproduction: 499-506,
2001.
25.
Semenza GL. HIF-1: mediator of physiological and pathophysiological responses
to hypoxia. Journal of Applied Physiology 88: 1474-1480, 2000.
26.
Serkova N, Bendrick-Peart J, Alexander B, and Tissot van Patot MC.
Metabolite concentrations in human placentae and their changes due to the collection
time after delivery. Placenta 24: 227-235, 2003.
27.
Serkova N, Jacobsen W, Niemann CU, Litt L, Benet LZ, Leibfritz D, and
Christians U. Sirolimus, but not the structurally related RAD (everolimus), enhances the
negative effects of cyclosporine on mitochondrial metabolism in the rat brain. British
Journal of Pharmacology 133: 875-885, 2001.
28.
Singh SN, Vats P, Kumria MM, Ranganathan S, Shyam R, Arora MP, Jain
CL, and Sridharan K. Effect of high altitude (7,620 m) exposure on glutathione and
related metabolism in rats. European Journal of Applied Physiology 84(3): 233-237,
2001.
29.
Sumbayev VV, Budde A, Zhou J, and Brune B. HIF-1 alpha protein as a target
for S-nitrosation. FEBS Letters 535: 106-112, 2003.
30.
Tissot van Patot MC, Grilli A, Chapman P, Broad E, Tyson W, Heller DS,
Zwerdlinger L, and Zamudio S. Remodelling of Uteroplacental Arteries is Decreased
in High Altitude Placentae. Placenta 24: 326-335, 2003.
31.
Wenger RH. Cellular adaptation to hypoxia: O2-sensing protein hydroxylases,
hypoxia-inducible transcription factors, and O2-regulated gene expression. FASEB 16:
1151-1162, 2002.
32.
Wiesener MS, Jurgensen JS, Rosenberger C, Scholze CK, Horstrup JH,
Warnecke C, Mandriota S, Bechmann I, Frei UA, Pugh CW, Ratcliffe PJ,
Bachmann S, Maxwell PH, and Eckardt KU. Widespread hypoxia-inducible
20
expression of HIF-2alpha in distinct cell populations of different organs. FASEB Journal
17: 271-273, 2003.
21
Table 1. Subject Characteristics
Maternal Age (yrs)
Gestational Age (days)
Birth Weight (g)
Placental Weight (g)
Placental Volume (ml)
Placental Wt/Birth Wt
1600 M
24.89 (2.2)
274.8 (4.42)
3246.1 (192.38)
633.62 (58.1)
526.5 (27.25)
0.2 (0.027)
3100 M
25.5 (2.1)
275.17 (2.28)
3192.1 (108.84)
550 (31.7)
455 (37.12)
0.17 (0.013)
22
Table 2.
Metabolites
Val,Leu,Ile
Lactate
Alanine
Acetate
Glutamate
*Succinate
Glutamine
*GSH
Aspartate
PCr/ Cr
PC/GPC
Taurine
myo-Inositol
Glucose
Glycogen
*PME
PDE
*PME/PDE
ATP
*ADP
ATP/ADP
NAD
UDPG
Low Altitude
High Altitude p Value
0.754 (0.052)
3.838 (0.418)
0.503 (0.019)
0.141 (0.046)
1.205 (0.046)
0.888 (0.068)
4.728 (0.523)
0.518 (0.056)
0.054 (0.009)
1.308 (0.17)
0.178
0.243
0.828
0.076
0.617
0.140 (0.016)
0.643 (0.08)
0.218 (0.02)
0.508 (0.053)
0.023
0.188
0.215 (0.019)
0.513 (0.079)
0.401 (0.036)
0.585 (0.029)
2.769 (0.25)
0.926 (0.066)
1.47 (0.047)
0.366 (0.252)
0.762 (0.089)
0.418 (0.036)
0.456 (067)
0.798 (0.206)
2.264 (0.420)
0.774 (0.09)
1.364 (0.163)
0.014 (0.014)
0.001
0.276
0.528
0.397
0.368
0.236
0.592
0.157
2.342 (0.193)
2.378 (0.08)
4.366 (0.404)
2.758 (.405)
0.004
0.441
0.985 (0.075)
0.757 (0.029)
1.65 (0.145)
0.986 (0.139)
0.007
0.195
0.318 (0.016)
2.394 (0.117)
0.691 (0.056)
0.18 (0.038)
0.526 (0.048)
1.9 (0.245)
0.72 (0.90)
0.254 (0.086)
0.008
0.139
0.805
0.491
*Significantly different in high as compared to low altitude pregnancies (p ! 0.05).
Bold: classic markers for short-term hypoxia, which did not change in this study.
23
Table 3.
Lipids
PUFA
TAG
Total FA
Cholesterol
Low Altitude
High Altitude p-Value
23.282 (3.589) 27.193 (1.478)
6.155 (0.917) 6.924 (0.591)
27.466 (4.264) 31.639 (1.041)
3.594 (0.567) 4.154 (0.273)
24
0.4326
0.5120
0.3969
0.3896
Figure Legends.
Figure 1. Placental tissues from low- (1600 M) and high (3100 M) -altitude placentae
were fixed in 10% buffered formalin, paraffin embedded and immunohistochemically
studied using anti-CD34+ antibody and haematoxylin background stain, 200X. A. Low
altitude placenta negative, using mouse IgG in place of primary antibody. B. Low altitude
placenta (serial section from 1A). C. High altitude placenta. D. Ratio of capillary density
to villous surface density in placentae from pregnancies at low- (n = 10) and high (n =
10) altitude. C - capillaries, V- villi, *p = 0.038
Figure 2. Nuclear protein was extracted from low (1600 M) and high (3100 M) altitude
placentae and analyzed for HIF-1-DNA binding using an ELISA transcription factor
activity assay (n = 10 low altitude and n = 10 high altitude subjects). *p < 0.0001. Data
presented as mean ± the standard error of the mean.
Figure 3. Nuclear proteins, extracted from low (1600 M) and high (3100 M) altitude
placentae, were analyzed for HIF-1! and "-actin using Western blot analysis (Top). Data
are expressed as the ratio of relative densitometry units of the protein of interest to "actin for total protein (mean ± the standard error of the mean) (Bottom).
Figure 4. Nuclear protein, extracted from placentae from low- (1600 M) and high- (3100
M) altitude pregnancies were analyzed for HIF-1" and "-actin using Western blot
analysis (Top). Data are expressed as the ratio of relative densitometry units of the
protein of interest to "-actin for total protein (Bottom). The graph on the left depicts
25
individual data to demonstrate the variability of HIF-1! between subjects, while the
graph on the right depicts the mean ± the standard error of the mean at each altitude with
and without #3 at high altitude. *p = 0.033 greater than low altitude.
Figure 5. Total protein, extracted from low (1600 M) and high (3100 M) altitude
placentae, was analyzed for vHL and !-actin using Western blot analysis (Top). Data are
expressed as the ratio of relative densitometry units of the protein of interest to !-actin
for total protein (mean ± the standard error of the mean) (Bottom).
Figure 6. Nuclear protein, extracted from placentae from low- (1600 M) and high- (3100
M) altitude pregnancies were analyzed for HIF-2" and !-actin using Western blot
analysis (Top). Data are expressed as the ratio of relative densitometry units of the
protein of interest to !-actin for total protein (mean ± the standard error of the mean)
(Bottom).
Figure 7. Total protein, extracted from low (1600 M) and high (3100 M) altitude
placentae, was analyzed for VEGF, Flt-1 and !-actin using Western blot analysis (Top).
Data are expressed as the ratio of relative densitometry units of the protein of interest to
!-actin for total protein (mean ± the standard error of the mean) (Bottom).
Figure 8. Total protein, extracted from low (1600 M) and high (3100 M) altitude
placentae, was analyzed for GAPDH, enolase and !-actin using Western blot analysis
(Top). Data are expressed as the ratio of relative densitometry units of the protein of
interest to !-actin for total protein (mean ± the standard error of the mean) (Bottom).
26
27
V
12
*
10
A
V
EC
Svvcap/Svvill (cm3)
EC
8
6
4
2
0
B
1600 M
V
EC
C
D.
Figure 1.
3100 M
1.8
1.6
HIF Binding (ng
/ml)
(ng/ml)
1.4
1.2
1
*
0.8
0.6
0.4
0.2
0
1600 M
Figure 2.
3100 M
HIF-1!/"-actin
0.5
0.4
0.3
0.2
0.1
0
1600 m
Figure 3.
3100 m
Figure 4.
2.5
*
1.5
HIF-1!/!-actin
HIF-1!/!-actin
2
1
0.5
0
1
2
3
4
5
1600 m
6
7
8
9
1
Subjects
2
3
4
3
3100 m
4
5
6
7
*
1.4
1.2
1
0.8
0.6
0.4
0.2
0
1600 m
3100 m
3100 m
(without
#3)
1600 m
vHL/!-actin
3100 m
3100 m
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
1600 m
Figure 5.
1600 m
3100 m
HIF-2!/"-actin
1
0.8
0.6
0.4
0.2
0
Figure 6.
1600 m
3100 m
Figure 7.
1.2
1
Flt-1/!-actin
VEGF/!-actin
3.5
3
2.5
2
1.5
1
0.5
0
0.8
0.6
0.4
0.2
0
1600 m
3100 m
1600 m
3100 m
1.2
Enolase/!-actin
GAPDH/!-actin
2
1.5
1
0.5
0.8
0.6
0.4
0.2
0
0
1600 m
Figure 8.
1
3100 m
1600 m
3100 m