| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
, in Placentas from Women with Preeclampsia1
a Departments of Obstetrics, Gynecology, and Reproductive Sciences and of Cell Biology and Physiology, University of Pittsburgh School of Medicine and Magee-Womens Research Institute, Pittsburgh, Pennsylvania 15213
ABSTRACT
Transcription factors orchestrate the development of extraembryonic tissues. Because placental hypoxia likely plays an important role in both normal and abnormal placentation, we have been investigating the hypoxia-inducible transcription factors (HIFs) in the human placenta. In this report, we focus on the placentas from women with preeclampsia. Because the placenta is a large, heterogeneous organ, we employed a systematic and unbiased approach to placental sampling, and our results are based on the analyses of eight biopsy sites per placenta. We observed no significant differences in HIF-1
or -2 mRNA expression between normal term and preeclamptic placentas. Nor was HIF protein expression significantly different, with the notable exception of HIF-2, which, on average, was increased by 1.7-fold in the preeclamptic placentas (P < 0.03 vs. normal term placentas). Considering all 48 paired placental biopsy sites (eight sites each for six normal term and six preeclamptic placentas), HIF-2 protein levels in the preeclamptic placentas exceeded those in the normal term placentas in 39, or 81%, of the paired sites (P < 0.0013). The HIF-2 immunoreactivity was mainly located in the nuclei of the syncytiotrophoblast and fetoplacental vascular endothelium in the preeclamptic villous placenta. To control for the earlier gestational age of the preeclamptic placentas, an additional group of placentas from preterm deliveries without preeclampsia were also evaluated. The HIF protein expression was comparable in these preterm specimens and the normal term placentas. We conclude that protein expression of HIF-2, but not of HIF-1 or -1ß, is selectively increased in the preeclamptic placenta. The molecular mechanism(s) of this abnormality as well as the genes affected downstream are currently under investigation. To our knowledge, this is the first report of abnormal HIF-2 expression in human disease other than cancer.
developmental biology, gene regulation, placenta, syncytiotrophoblast, trophoblast
INTRODUCTION
Preeclampsia is a common disease of human pregnancy and a leading cause of both maternal and neonatal morbidity and mortality [1]. Inadequate trophoblast invasion and deficient remodeling of uterine spiral arteries are associated with preeclampsia [2]. These deficiencies of placentation are postulated to cause focal regions of ischemia/hypoxia that, in turn, stimulate the overproduction of various placental products, such as proinflammatory cytokines, that spill over into the maternal circulation, thereby causing endothelial dysfunction and systemic disease [3]. The maternal manifestations of preeclampsia that occur during the third trimester include hypertension, proteinuria, excessive edema, organ hypoprofusion, as well as activation of platelets and coagulation.
The molecular consequences of cellular hypoxia have been extensively reviewed in recent years [4]. A variety of genes are modulated as an adaptive response to hypoxia (e.g., those involved in glucose transport, glycolysis, angiogenesis, and erythropoiesis) [4]. A prerequisite for many hypoxia-mediated responses is the basic-helix-loop-helix-PAS (Per-Arnt-Sim) family member transcription factors, hypoxia-inducible transcription factor (HIF)-1 [5] and -2 [6, 7], that heterodimerize with the aryl hydrocarbon nuclear translocator protein (ARNT or HIF-1ß) [4, 5] and then bind to the hypoxia response element (core consensus sequence: 5'-CGTG-3') in responsive genes. Although the identity of the oxygen sensor upstream of the HIF transcription factors remains elusive [8], the molecular mechanisms of HIF activation by hypoxia are becoming clearer and include protein stabilization [9, 10], thiol-redox regulation [11, 12], and interaction with CBP/p300 [12, 13]. Interestingly, agents besides hypoxia have been shown to activate HIF under normoxic conditions, including a variety of growth factors [14–16], v-SRC [17], phosphorylation by mitogen-activated protein kinase [18], as well as transition metals such as cobalt chloride and iron-chelating agents [19]. Finally, HIF transcription factors are critical for mouse development. Indeed, HIF-1 [20], -2 [21], and -1ß [22] homozygous knockout mice all have manifested embryonic lethality.
Recently, we [23] and other investigators [24–26] have begun to consider the potential contribution of HIF to normal development of the human placenta. A priori, these transcription factors are likely to be involved, particularly during the first 10 wk or so of placental development, when the organ resides in a relatively hypoxic environment (
20 torr) [27]. We found that both HIF-1 and -2 protein were increased during the early first trimester, which is consistent with the physiological hypoxia of the human placenta at that gestational stage, and regulation of HIF by hypoxia in the human placenta occurred at the level of protein and not mRNA [23]. In the present work, we extended our investigation of HIF into the abnormal placentation of preeclampsia. We hypothesized that HIF protein is increased in the preeclamptic placenta, either as a cause or a consequence of the ischemia/hypoxia that is postulated to afflict the placenta in this disease. Our objective, therefore, was to compare the abundance of HIF mRNA and protein in placentas from women with normal term deliveries and preterm deliveries, with and without preeclampsia, as well as to examine cellular localization of HIF protein. For the normal term and preeclamptic placentas, our results were based on a systematic and unbiased approach to sampling of each placenta involving eight biopsies per specimen [28].
MATERIALS AND METHODS
Placental Collection and Processing
All placental collections were approved by the Institutional Internal Review Board of Magee-Womens Hospital. Placentas were obtained from six nulliparous women with normal pregnancies undergoing cesarean section without labor at term (38–40 gestational wk). As an additional control group, six placentas were obtained from nulliparous women with preterm delivery and without preeclampsia (28–32 gestational wk). These patients with preterm deliveries were also delivered by cesarean section, five with and one without labor, and all neonates were appropriately grown for gestational age. Only one of these six had chorioamnionitis, which was reported as mild by the perinatal pathologist. Six placentas were obtained from nulliparous women with preeclampsia by cesarean section without labor (28–38 gestational wk). Preeclampsia was diagnosed according to standard criteria: onset of hypertension during late pregnancy with systolic and diastolic blood pressure 
140/90 on at least two occasions, and urinary protein of at least 1+ on dipstick or 300 mg/24 h [1]. All of the patients except one had hyperuricemia. Two patients also had HELLP syndrome (hemolysis, elevated liver function tests, and low platelets), and two had babies that were growth restricted.
The placentas were sampled immediately after extraction from the uterus. To ensure systematic and unbiased sampling of the entire placenta [28], we placed a plastic grid containing 16 holes over the maternal face of each placenta (Fig. 1A). Grids of various diameters were constructed, so that each placenta could be properly fitted. Sixteen placental biopsy specimens of approximately 0.5 g each were obtained through the holes in the fitted grid. Each biopsy specimen included decidua basalis and villous placenta, but not chorionic plate. After briefly rinsing each sample in three or four changes of saline, it was blotted and then snap frozen in liquid nitrogen. With two investigators obtaining the biopsy specimens, the entire procedure took less than 10 min.
|
For the analysis of HIF in the present work, we analyzed 8 of the 16 biopsy sites (alternate sites) obtained from the normal term and preeclamptic placentas. For the preterm placentas, only one biopsy per placenta was available. Each of these biopsies was compared with a pool of the eight biopsy sites from a normal term placenta (see below).
Northern Blot Analysis
Total RNA was isolated from placental tissues using RNAwiz (Ambion, Austin, TX). Placental tissues were first pulverized using a Bessman tissue pulverizer (Spectrum Medical Industries, Laguna Niguel, CA) and then homogenized in 10 volumes of RNAwiz using a Tissuemizer (Tekmar, Cincinnati, OH) at speed 60 for two 30-sec bursts. After a 10-min incubation at room temperature (RT), 0.2 volumes of chloroform were added to the homogenate, mixed thoroughly, and allowed to incubate for an additional 10 min. The mixture was next centrifuged at 10 000 x g for 10 min. The clear aqueous phase was collected in a clean tube and diluted with an equal volume of RNase-free water. The RNA was precipitated by the addition of one volume of isopropanol. After a 10-min incubation at RT, the RNA was recovered by centrifugation at 10 000 x g for 15 min at 4°C. The pellet was washed with 70% ethanol, allowed to air dry, and dissolved in minimal volume of RNase-free water. The amount of RNA was estimated by spectrophotometry (A260), and one optical density was taken to be 40 µg/ml. Ten micrograms of total RNA was separated on formaldehyde containing 1.5% agarose gels and transferred to nylon membranes (MSI, Westborough, MA). The RNA was cross-linked to the membrane using a Fisher Biotech UV-cross-linker (Fisher, Pittsburgh, PA).
For the preparation of the cDNA probes, plasmids pBS/HIF-13.2-3T7 (EcoRI digest of the complete cDNA produces three bands of sizes 2063, 1011, and 604 base pairs [bp]) was generously provided by Dr. Gregg Semenza. Plasmid hEPAS-pcDNA3 (HIF-2, a 2818-bp BamHI fragment) was a kind gift from Dr. Steven McKnight. Human ß-actin cDNA containing plasmid was purchased from ATCC (Manassas, VA). All restriction digestion fragments were purified from gel using the Jetsorb DNA purification kit (Genomed Inc., Research Triangle Park, NC).
The cDNA probes were made using 25–50 ng of the insert, 50 µCi of -32P-dCTP (3000 Ci/mmol; NEN-Dupont, Boston, MA), and 2 U of Klenow polymerase using the multi-prime DNA-labeling kit (Amersham Pharmacia Biotech, Piscataway, NJ). Unincorporated free 32P was removed using a spin column (Bio-Spin P30; Biorad, Hercules, CA).
The membranes were washed once briefly in 2x SSC (single strength: 0.15 M sodium chloride and 0.015 M sodium citrate) and processed further for hybridization. Prehybridization was carried out for 12 h in 6 ml of buffer containing 50% deionized formamide, 5x Denhardt's solution, 5x SSPE (single strength: 150 mM NaCl, 10 mM NaH2PO4, and 1 mM Na/EDTA), 1% SDS, 0.5 mg/ml of denatured salmon sperm DNA, and 0.1 mg/ml of tRNA at 42°C in a Hybaid oven (Hybaid Limited, Middlesex, UK) with roller bottles (250 x 35 mm). After decanting the prehybridization solution, labeled probe in 4 ml of prehybridization buffer was added to the membrane (2 x 106 cpm/ml), and the hybridization was carried out for 12–24 h. The membranes were washed in 2x SSC/0.1% SDS twice for 5 min at RT and then twice in the same solution at 37°C. The washing was carried out for another 15 min at 50°C with 0.1x SSC/0.1% SDS, when necessary. The membranes were exposed to Kodak-Bio-max-AR film (Eastman Kodak, Rochester, NY).
The membranes were stripped in 0.1% SDS at 60°C for 15–30 min and reprobed for another gene of interest or for the housekeeping gene, ß-actin, to correct for interlane loading variations. The signal on the film was quantitated by densitometry using the Videk Harmony program, version 4.03 (Videk Corporation, Rochester, NY).
Western Blot Analysis
Total proteins from placental tissues were extracted using 8 volumes of 1x Laemmli buffer (50 mM Tris-HCl [pH 6.8], 2% SDS, and 10% glycerol) containing 7 M urea, 5 mM dithiothreitol, 0.5 mM phenylmethyl sulfonyl fluoride, and 1 µg/ml each of protease inhibitors, leupeptin, aprotinin, and pepstatin. Tissues were homogenized with a Tekmar Tissuemizer for approximately 1 min at setting 6. The crude homogenate was centrifuged at 13 400 x g at 4°C, and the supernatant was subjected to mild sonication (microprobe, setting 4 for 5 sec; Sonifier-Cell disrupter, heat systems; Ultrasonics, Plainview, NY). The supernatant was then stored in aliquots at -80°C. Protein estimation was carried out using the Biorad reagent, and 150 µg of protein per lane was used for the Western blot analysis. Total protein samples were loaded in a 40-µl volume containing the desired amount of protein and 5% ß-mercaptoethanol and 0.005% pyronin (Sigma, St. Louis, MO). All preparations were finally boiled for 5 min and then briefly centrifuged. The proteins were separated on a 0.1% SDS containing 6% polyacrylamide gel (20 x 20 cm; Owl Separation System, Portsmouth, NH) at 100 V for 4 h. The proteins were next transferred to polyvinylidene fluoride membranes (Immobilon; Millipore, Bedford, MA) using a semidry transfer system (Panther semi-dry transfer system; Owl Separation System) at a constant current of 1.0 mA/cm2. Detection of proteins was carried out after blocking the membranes with a 1% solution of Blocking Reagent (Boehringer Mannheim, Indianapolis, IN) and 5% nonfat dry milk in 10 mM Tris (pH 7.4), 150 mM NaCl (TBS)-0.05% Tween-20 (TBS-T) for 8 h and incubation with the primary antibody overnight at 4°C. An anti-HIF-1 mouse monoclonal immunoglobulin (Ig) G2b and a rabbit polyclonal anti-HIF-2 were obtained from Novus Biologicals (Littleton, CO). For HIF-1, the antibody was diluted 1:200 v:v in TBS buffer containing 0.05% Tween-20 and 1% BSA (Sigma), and for HIF-2, the antibody was diluted 1:400 v:v in fresh blocking buffer. The HIF-1ß (rabbit polyclonal, Novus Biologicals) and ß-actin (mouse monoclonal IgG1, Clone AC-15; Sigma) antibodies were used at 1:1000 v:v dilution each. The membranes were washed in TBS-T buffer three times for 10 min each and then incubated with alkaline phosphatase-conjugated secondary antibody (1:5000 v:v dilution for HIF-1 and 1:7500 v:v for HIF-2, -1ß, and ß-actin) for 1 h. The membranes were next washed three times in TBS-T for 10 min each. They were further washed in TBS buffer without Tween-20 for 10 min and then allowed to equilibrate in alkaline phosphatase buffer (100 mM Tris [pH 9.5] and 150 mM NaCl) for 5 min. Chemiluminescent detection was carried out using the CDP-Star substrate (Boehringer Mannheim) diluted 1:200 v:v in the alkaline phosphatase buffer for 5 min. Membranes were exposed for different times to Kodak Bio-max–AR film.
The membranes were stripped with a buffer containing 62.5 mM Tris-HCl (pH 6.8), 2% SDS, and 100 mM ß-mercaptoethanol at 50°C for 15–30 min and reprobed with another antibody of interest. Because the amount of protein per lane was too great for accurate determination of ß-actin by densitometry, we loaded 15 µg from the same protein mix that was used for HIF detection on a separate gel. This gel was then probed with the monoclonal anti-ß-actin antibody.
Immunohistochemistry
A HIF-2 antibody-clone EP190b (a mouse IgG1
) directed against amino acids 535–631 of the molecule was kindly provided by Dr. Helen Turley [29]. No cross-reactivity of the HIF-2 antibody with HIF-1 was observed [30]. We used the mouse monoclonal rather than the rabbit polyclonal antibody, because the former produced less background on immunocytochemistry. A small piece of frozen placental biopsy specimen was placed in a cryomold containing cold OCT compound (Sakura Finetek Inc., Torrance, CA) and snap frozen in liquid nitrogen. Seven-micrometer sections were cut and mounted on Fisher Superfrost/Plus glass slides (Pittsburgh, PA) and fixed for 5 h at RT with 10% formalin. The sections were subjected to antigen retrieval using Antigen Unmasking Solution (Vector Laboratories, Burlingame, CA) and a microwave oven. After quenching of endogenous peroxidases using 1.0% hydrogen peroxide in methanol for 3 min and blocking with 1.5% normal horse serum for 20 min, the tissues were incubated with the monoclonal antibody overnight at 4°C. Preliminary studies showed that the optimal antibody concentration was 3–10 µg/ml. Because the antibody was provided as cell culture supernatant at 10 µg/ml, it was applied directly to the tissue sections or diluted accordingly for use. For the negative control, RPMI cell culture medium containing 10% fetal bovine serum, 0.002% sodium azide, and 3–10 µg/ml of IgG1 was used. The Dako Envision+ system (HRP/DAB, Carpinteria, CA) was employed for detection. After dehydration in ethanol and xylene solutions, a coverslip was applied using Cytoseal XYL (Stephens Scientific, Riverdale, NJ). The tissues were lightly counterstained with hematoxylin, Gill No. 2.
Data Analysis
For Figures 1B and 2A, the variance of the mean of the ratios was approximated based on a Taylor series expansion [31]. To test the hypothesis that the ratio is equal to 1.0, a Z-statistic was computed using the estimated SEM from the Taylor series expansion. For analyses involving the six matched pairs in Table 1, the variance was computed as the average of the individual variances, and the appropriate Z-statistic was used to test the hypothesis that the ratio is 1.0. Thus, if the ratio was not significantly different from unity (i.e., null hypothesis), then the HIF expression was comparable in normal term and preeclamptic placentas. In Figure 3, the mean values were compared by the unpaired t-test. For comparison of proportions, the chi-square test was applied. Finally, for comparison of gestational ages in the normal term and preterm deliveries with and without preeclampsia, one-factor randomized block design ANOVA was applied, and the Scheffe test was used for post-hoc comparisons of group means. A P value of 
0.05 was taken to be significant.
|
|
RESULTS
HIF Gene Expression in Normal Term and Preeclamptic Placentas
The total RNA extracted from the eight placental biopsy sites of one normal term and preeclamptic placenta each was loaded on one gel for Northern blot analysis. Representative Northern blots for HIF-1, -2, and ß-actin (from a total of six for each probe) are illustrated Figure 1B. The ratio of placental HIF mRNA expression (first normalized to ß-actin mRNA) in preeclampsia and normal pregnancy was calculated for each biopsy site. These ratios were then averaged according to placental biopsy site number for the six blots as depicted in the graph of Figure 1B. Except for HIF-1 site 5, none of the ratios were significantly different from 1.0 (stippled line), indicating comparable expression of HIF mRNA in normal term and preeclamptic placentas across the eight biopsy sites.
The ratios of HIF mRNA expression in the eight biopsy sites were also averaged for each placental pair or blot as depicted in Table 1. Again, the mean ratios did not differ significantly from unity, indicating comparable HIF gene expression in normal term and preeclamptic placentas.
The ranges for HIF/ß-actin mRNA expression among the eight placental biopsy sites between normal term and preeclamptic placentas were not significantly different. On average, the highest values for HIF-1 and -2, respectively, were 2.7- and 2.4-fold greater than the lowest values.
HIF Protein Expression in Normal Term and Preeclamptic Placentas
The total protein extracted from the eight placental biopsy sites of one normal term and preeclamptic placenta each was loaded on one gel in duplicate for Western blot analysis. After transfer, one of the membranes was probed for HIF-1 and the other for HIF-2. Then, one of the membranes was stripped and reprobed for HIF-1ß. By utilizing two membranes each per placental pair, only one stripping procedure was needed. Representative Western blots for HIF-1, -2, and -1ß are illustrated in Figure 2A (from a total of six for each probe). Because 150 µg of protein were loaded per lane, we could not use the same blots for ß-actin, because the signals were too intense for accurate quantification by densitometry. Therefore, one-tenth of the protein was loaded for eight biopsy specimens from one normal term and preeclamptic placenta for analysis of ß-actin as a confirmation of even protein loading. ß-Actin is not altered by hypoxia in human placental villous explants [23], and as illustrated in Figure 2A, the signal is comparable from lane to lane, thereby verifying even protein loading. The HIF-1ß protein may also serve as a confirmation of uniform protein loading in Figure 2A, because it is not affected by hypoxia in placental villous explants either [23].
|
The ratio of placental HIF protein expression in normal pregnancy and preeclampsia was calculated for each biopsy site. These ratios were then averaged according to placental biopsy site number for the six blots as depicted in the graph of Figure 2A. With the exception of placental biopsy site 11 for HIF-1ß, none of the other placental biopsy sites demonstrated ratios that were significantly different from unity (stippled line) for HIF-1 and -1ß. In contrast, the HIF-2 ratio significantly exceeded 1.0 in five of the eight placental biopsy sites.
The ratios of HIF protein expression in the eight biopsy sites were averaged for each placental pair or blot as depicted in Table 1. Again, the mean ratios did not differ significantly from unity for HIF-1 and -1ß, indicating comparable protein expression in normal term and preeclamptic placentas. However, the ratio for HIF-2 was 1.71 ± 0.27, which was significantly greater than 1.0, implying increased HIF-2 protein expression in the preeclamptic placentas. Considering all 48 paired placental biopsy sites (eight sites each for six normal term and six preeclamptic placentas), HIF-1 and -2 protein levels in the preeclamptic placentas exceeded those in the normal term placentas in 31 (65%, P = not significant) and 39 (81%, P < 0.0013) of the paired sites, respectively. In one of the placental pairs (;ns2, Table 1), the ratio of HIF-2 protein in the preeclamptic and normal placenta only reached 1.09; however, the levels in the former exceeded those in the latter in six of eight paired biopsy sites.
Figure 2B depicts the localization of HIF-2 immunoreactivity in two of the preeclamptic placentas with the greatest HIF-2 protein expression as documented by Western blot analysis (see above). The majority of the staining was in the syncytiotrophoblast, with some expression also evident in the fetoplacental vascular endothelium. The staining was both nuclear and cytoplasmic, with the former being particularly intense.
The ranges for HIF protein expression among the eight placental biopsy sites between normal term and preeclamptic placentas were not significantly different. On average, the highest values for HIF-1 and -2, respectively, were 2.7- and 2.8-fold greater than the lowest values.
HIF Protein Expression in Normal Term and Preterm Placentas
Our previous work showed no significant difference in either HIF-1 or -2 protein expression between placentas of 18–21 gestational wk and term; in both instances, the expression was low compared to that at 5–8 gestational wk [23]. Nevertheless, because the normal term and preeclamptic placentas were not gestationally aged matched in the current investigation (39.4 ± 0.3 vs. 33.7 ± 1.5 wk, respectively; P < 0.001), we obtained placentas from preterm deliveries without preeclampsia. The average gestational age for the preeclamptic group was not statistically different from the preterm group without preeclampsia (33.7 ± 1.5 vs. 30.6 ± 0.8 wk, respectively; P = 0.12). Because only one placental biopsy was available from each of the six preterm placentas, they were loaded onto one gel along with a pool of eight sites each for the six normal term placentas. As before, two identical gels were prepared, one for HIF-1 and the other for HIF-2. The latter was stripped and reprobed for HIF-1ß. One-tenth of the protein was loaded onto a third gel for ß-actin. The Western blots are depicted in Figure 3. The data are summarized in the graph: no significant differences were observed in the expression of HIF-1, -2, or -1ß protein between the placentas derived from normal term and preterm deliveries.
DISCUSSION
Our objective was to determine whether expression of HIF is increased in the preeclamptic placenta. The major findings were: 1) HIF-1 and -2 gene expression was not significantly different between normal term and preeclamptic placentas; 2) HIF-2, but not HIF-1 or -1ß, protein expression was significantly increased in preeclamptic relative to normal term placentas; 3) HIF-2 immunoreactivity was mainly observed in the syncytiotrophoblast and, to a lesser degree, in the fetoplacental vascular endothelium of preeclamptic placentas, and the intensity of nuclear staining exceeded that of the cytoplasm; and 4) HIF-1, -2, and -1ß protein expression was not significantly different in placentas from preterm deliveries without preeclampsia compared to normal term placentas.
The placenta is a large, heterogeneous organ. Moreover, the ischemia/hypoxia that is postulated to afflict the preeclamptic placenta is likely to be focal rather than global, depending on the number of uterine spiral arteries with absent or incomplete physiological remodeling [32]. Thus, conclusions about the placenta based on a single biopsy specimen may be erroneous [28]. Although the ranges of HIF mRNA and protein expression across the eight biopsy sites that we analyzed in each placenta were similar between normal term and preeclamptic specimens, they were considerable, with the highest value being, on average, 2.5-fold greater than the lowest, which is consistent with the heterogeneous nature of the human placenta. Therefore, the systematic and unbiased approach to placental sampling, as well as the multiple sites evaluated in the present work, strengthen our data about HIF mRNA and protein expression in the preeclamptic placenta. Note, however, that we only had one sample available from the placentas of preterm deliveries without preeclampsia, which does represent a weakness in the current investigation. Because these biopsy specimens were provided to us by our colleagues, at least no bias occurred on our part as to the region of the placenta from which they were taken (e.g., infarct or peri-infarct zone).
We also analyzed placentas from preterm deliveries without preeclampsia that were approximately the same gestational age as the preeclamptic placentas. We observed no significant difference in HIF-1, -1ß, and -2 protein expression between the normal term placentas and those from preterm deliveries without preeclampsia. These results are consistent with our earlier investigation of the ontogeny of HIF protein expression in the human placenta [23]. Although both HIF-1 and -2 were significantly increased in placentas of 5–8 gestational wk, presumably reflecting the physiological hypoxia of that gestational period, the levels were uniformly low throughout the rest of pregnancy. Thus, we did not anticipate any differences between the normal term and preterm placentas without preeclampsia (39 and 31 gestational wk, respectively).
The localization of HIF-2 protein was investigated in biopsy specimens of several preeclamptic placentas, with the highest expression observed by Western blot analysis. Comparable to our earlier work on normal term placentas [23], immunostaining of moderate intensity was observed in the syncytiotrophoblast layer and, to a lesser degree, in the fetoplacental vascular endothelium. When decidua basalis was present, the immunostaining was strong within extravillous trophoblasts in this tissue (data not shown). Because the intensity of immunoreactivity was particularly strong in the nuclei, the elevated HIF-2 level in the preeclamptic placenta is likely to have functional consequences.
The finding of increased HIF-2 protein expression in the preeclamptic placenta actually corroborates our preliminary work using other placentas (unpublished data), which prompted the present detailed investigation. However, the mechanism(s) of HIF-2 protein overexpression in preeclamptic placentas requires further study. An obvious possibility is that the elevated levels reflect the ischemia/hypoxia that is believed to affect many preeclamptic placentas [3, 32]. That HIF-1 was unchanged is not incompatible with this possibility, because HIF-2 protein may be more abundantly expressed under higher levels of oxygenation than HIF-1 (i.e., HIF-2 is a more sensitive marker of ischemia/hypoxia) [23, 30]. That both HIF-1 and -2 mRNA, as well as HIF-1ß protein, were not elevated in the preeclamptic placentas is consistent with our earlier work showing that they are not regulated by hypoxia in placental villous explants [23].
A surprising observation, however, was that HIF-2 protein was more abundant in preeclamptic than in normal term placentas in 39 of the 48 (81%) of the paired biopsy sites. In other words, HIF-2 was overexpressed in the preeclamptic placenta more on a global rather than on a focal basis, which is not easily reconcilable with the patchy distribution of absent or incomplete spiral artery remodeling and ischemia/hypoxia in the disease [2, 32]. Therefore, there may be additional reasons (other than ischemia/hypoxia) for the elevation in HIF-2 protein in preeclamptic placentas. Possibilities currently under investigation in our laboratory include the oxygen-sensing mechanism in the preeclamptic placenta, which may have higher sensitivity or gain; impaired degradation via the ubiquitin-proteosome pathway [9, 10]; and activation by growth factors [14–16].
As mentioned, we previously documented that the expression of both HIF-1 and -2 protein, but not mRNA, were significantly increased in placentas of 5–8 gestational wk [23]. Thereafter, they declined to low levels for the remainder of gestation, presumably reflecting the complete establishment of intervillous blood flow and adequate oxygenation [27]. An interesting, but difficult, question to investigate is whether the HIF-2 protein levels in the placentas of women destined to develop preeclampsia failed to decrease after 5–8 gestational wk. (This abnormality could arise as a consequence of defective ubiquitin-proteosome degradation, uterine spiral arteries that are resistant to physiological changes and result in persistent hypoxia, etc.) If so, then the biological consequence might be to program the villous stem cell cytotrophoblast toward a proliferative and away from an invasive phenotype, analogous to the influence of hypoxia on this cell type in vitro [33–36]. In other words, failure to down-regulate HIF-2 protein expression in the villous placenta during early pregnancy could prevent the villous cytotrophoblast from switching to an invasive phenotype, thereby precluding complete physiological remodeling of uterine spiral arteries and resulting in shallow placentation, placental ischemia/hypoxia, and ultimately, preeclampsia. Another biological consequence of this fundamental abnormality could be the hypercapillarity and exaggerated branching of the fetoplacental vasculature in the villous core [37]. All these derangements in the preeclamptic placenta are presumably secondary to altered expression of a variety of genes, some of which may be regulated by HIF-2. We are currently seeking to identify the genes downstream of HIF-2 in the human placenta.
In summary, the fundamental observation of this work is the exaggerated protein expression of HIF-2 in the preeclamptic placenta. To our knowledge, this is the first demonstration of abnormal HIF-2 expression in human disease other than cancer.
ACKNOWLEDGMENTS
We thank Dr. Helen Turley and colleagues for their generous contribution of the HIF-2 antibody for immunohistochemistry, as well as Drs. Gregg Semenza and Steven McKnight for their gifts of the HIF-1 and -2 cDNA probes, respectively. We also thank Drs. Phil Heine and Deb Draper for providing the placentas from the preterm deliveries without preeclampsia, as well as Robert Szmyd, Karen Bellisario, Cindy Schatzman, and the Prenatal Exposures Preeclampsia Prevention staff for their assistance in the procurement of normal term and preeclamptic placentas. We are grateful to Sue Kauffman for excellent clerical support.
FOOTNOTES
1 Supported by NIH RO1 HL56410 and PO1 HD30367. Portions of this work were published in abstract form (J Soc Gynecol Invest 2000; 7:287A). 
2 Correspondence: Kirk P. Conrad, Magee-Womens Research Institute, 204 Craft Ave., Pittsburgh PA 15213. FAX: 412 641 1503; rsikpc{at}mail.magee.edu
Accepted: August 29, 2000.
Received: May 16, 2000.
REFERENCES
regulates the VEGF expression and is potentially involved in lung and vascular development. Proc Natl Acad Sci U S A 1997; 94:4273–4278. is mediated by an O2-dependent degradation domain via the ubiquitin-proteasome pathway. Proc Natl Acad Sci U S A 1998; 95:7987–7992. (HIF-1) protein is rapidly degraded by the ubiquitin-proteasome system under normoxic conditions. J Biol Chem 1997; 272: 22642–22647. subunit. J Biol Chem 1996; 271:32253–32259. in response to hypoxia: their stabilization and redox signal-induced interaction with CBP/p300. EMBO J 1999; 18:1905–1914.[CrossRef][Medline]/ARNT. EMBO J 1998; 17:5085–5094.[CrossRef][Medline] and insulin-like growth factor 2. Cancer Res 1999; 59:3915–3918. (HIF-1) and enhance the transcriptional activity of HIF-1. J Biol Chem 1999; 274:32631–32637.. Genes Dev 1998; 12:149–162. and HIF-2 in normal human tissues, cancers and tumor associated macrophages. Am J Pathol 2000; 157:411–421.. Blood 1998; 92:2260–2268.This article has been cited by other articles:
|
|
 |
|
  F. Burbank History of Uterine Artery Occlusion and Subsequent Pregnancy Am. J. Roentgenol., June 1, 2009; 192(6): 1593 - 1600. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-H. Kim, H.-S. Hwang, Y.-T. Kim, H.-S. Kim, and Y.-W. Park Modulation of Matrix Metalloproteinase Secretion by Adenosine A3 Receptor in Preeclamptic Villous Explants Reproductive Sciences, November 1, 2008; 15(9): 939 - 949. [Abstract] [PDF]
|
|
|
|
 |
|
 C Pieh, H Agostini, C Buschbeck, M Kruger, J Schulte-Monting, U Zirrgiebel, J Drevs, and W A Lagreze VEGF-A, VEGFR-1, VEGFR-2 and Tie2 levels in plasma of premature infants: relationship to retinopathy of prematurity Br. J. Ophthalmol., May 1, 2008; 92(5): 689 - 693. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. T. Papageorghiou, F. Prefumo, K. Leslie, D. C. Gaze, P. O. Collinson, and B. Thilaganathan Defective endovascular trophoblast invasion in the first trimester is associated with increased maternal serum ischemia-modified albumin Hum. Reprod., April 1, 2008; 23(4): 803 - 806. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Rajakumar, A. Jeyabalan, N. Markovic, R. Ness, C. Gilmour, and K. P. Conrad Placental HIF-1{alpha}, HIF-2{alpha}, membrane and soluble VEGF receptor-1 proteins are not increased in normotensive pregnancies complicated by late-onset intrauterine growth restriction Am J Physiol Regulatory Integrative Comp Physiol, August 1, 2007; 293(2): R766 - R774. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Zamudio, Y. Wu, F. Ietta, A. Rolfo, A. Cross, T. Wheeler, M. Post, N. P. Illsley, and I. Caniggia Human Placental Hypoxia-Inducible Factor-1{alpha} Expression Correlates with Clinical Outcomes in Chronic Hypoxia in Vivo Am. J. Pathol., June 1, 2007; 170(6): 2171 - 2179. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. J. Lockwood, P. Toti, F. Arcuri, E. Norwitz, E. F. Funai, S.-T. J. Huang, L. F. Buchwalder, G. Krikun, and F. Schatz Thrombin Regulates Soluble fms-Like Tyrosine Kinase-1 (sFlt-1) Expression in First Trimester Decidua: Implications for Preeclampsia Am. J. Pathol., April 1, 2007; 170(4): 1398 - 1405. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Lam, K.-H. Lim, and S. A. Karumanchi Circulating Angiogenic Factors in the Pathogenesis and Prediction of Preeclampsia Hypertension, November 1, 2005; 46(5): 1077 - 1085. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. A. Karumanchi and Y. Bdolah Hypoxia and sFlt-1 in Preeclampsia: The "Chicken-and-Egg" Question Endocrinology, November 1, 2004; 145(11): 4835 - 4837. [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Tissot van Patot, J. Bendrick-Peart, V. E. Beckey, N. Serkova, and L. Zwerdlinger Greater vascularity, lowered HIF-1/DNA binding, and elevated GSH as markers of adaptation to in vivo chronic hypoxia Am J Physiol Lung Cell Mol Physiol, September 1, 2004; 287(3): L525 - L532. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. S. Torry, D. Mukherjea, J. Arroyo, and R. J. Torry Expression and Function of Placenta Growth Factor: Implications for Abnormal Placentation Reproductive Sciences, May 1, 2003; 10(4): 178 - 188. [Abstract] [PDF]
|
|
|
|
 |
|
 D. Casagrandi, C. Bearfield, J. Geary, C.W. Redman, and S. Muttukrishna Inhibin, activin, follistatin, activin receptors and {beta}-glycan gene expression in the placental tissue of patients with pre-eclampsia Mol. Hum. Reprod., April 1, 2003; 9(4): 199 - 203. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Stepan, W. Marqwardt, Y. Kuhn, M. Hockel, H.-P. Schultheiss, and T. Walther Structure and Regulation of the Murine Mash2 Gene Biol Reprod, January 1, 2003; 68(1): 40 - 44. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. A. Khalil and J. P. Granger Vascular mechanisms of increased arterial pressure in preeclampsia: lessons from animal models Am J Physiol Regulatory Integrative Comp Physiol, July 1, 2002; 283(1): R29 - R45. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Zhang and L.A. Salamonsen Expression of hypoxia-inducible factors in human endometrium and suppression of matrix metalloproteinases under hypoxic conditions do not support a major role for hypoxia in regulating tissue breakdown at menstruation Hum. Reprod., February 1, 2002; 17(2): 265 - 274. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
