- Email: [email protected]
Inhibition of the PI3K-Akt Pathway Suppresses sFlt1 Expression in Human Placental Hypoxia Models In Vitro
Placenta 31 (2010) 621e629
Contents lists available at ScienceDirect
Placenta journal homepage: www.elsevier.com/locate/placenta
Inhibition of the PI3K-Akt Pathway Suppresses sFlt1 Expression in Human Placental Hypoxia Models In Vitro J.K. Park a, b, J.W. Jeong a, M.Y. Kang b, J.C. Baek a, J.K. Shin a, b, S.A. Lee a, b, W.S. Choi b, c, J.H. Lee a, b,1, W.Y. Paik a, b, *,1 a b c
Department of Obstetrics and Gynecology, School of Medicine, Gyeongsang National University, JinJu, Republic of Korea Institute of Health Sciences, School of Medicine, Gyeongsang National University, JinJu, Republic of Korea Department of Anatomy, School of Medicine, Gyeongsang National University, JinJu, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history: Accepted 14 April 2010
Objective: Although elevated expression of soluble fms-like tyrosine kinase 1 (sFlt1) plays a major role in the pathogenesis of pre-eclampsia, it is unclear how hypoxia regulates placental sFlt1 expression. Thus, we investigated sFlt1 expression in placentas from normal and preeclamptic pregnancies and in human placental hypoxia models in vitro to examine the role of the PI3K-Akt pathway in regulating the expression of this molecule. Methods: We examined the expression of VEGF, PlGF, sFlt1, PI3K, Akt, and HIF-1 in placental samples from ten women with pre-eclampsia and ten normotensive control patients and in human choriocarcinoma trophoblast cells treated with 600 mM CoCl2 by Western blotting. Using models of placental hypoxia, we also determined whether inhibition of the PI3K-Akt pathway plays a direct role in regulating the expression of sFlt1. Results: The VEGF, PlGF, sFlt1, PI3K, Akt, and HIF-1 levels were significantly higher in the preeclamptic placentas than the normal placentas. In the placental hypoxia models, the expression of VEGF and PlGF increased in a time-dependent manner, whereas the expression of sFlt1 plateaued after 3 h of CoCl2 treatment. The expression levels of p-Akt and PI3K were maximal after 6 and 12 h of CoCl2 treatment, respectively. The expression of HIF-1a increased in a time-dependent manner with CoCl2 treatment. Inhibition of the PI3K-Akt pathway with the PI3K-specific inhibitor LY294002 leads to decreased sFlt1 levels and unchanged or increased VEGF and PlGF levels. Conclusion: Inhibition of the PI3K-Akt pathway may be a useful therapeutic approach, if it were to decrease sFlt1 secretion without inhibiting VEGF or PlGF secretion. This pathway provides a potential target for a new treatment strategy in patients with pre-eclampsia. Ó 2010 Elsevier Ltd. All rights reserved.
Keywords: Placental hypoxia model sFlt1 PI3K inhibitor
1. Introduction Pre-eclampsia is a maternal syndrome defined in terms of its cardiovascular and renal features, i.e., new hypertension and proteinuria that resolve after delivery. The proteinuria is associated with a specific endothelial renal lesion called glomerular endotheliosis [1,2], and wide-ranging evidence indicates that the hypertension is secondary to diffuse endothelial dysfunction [3]. Thus, pre-eclampsia is now considered an endothelial disease [3].
* Corresponding author. Department of Obstetrics and Gynecology, College of Medicine, Gyeongsang National University, 90 Chilam-dong, Jinju, 660-702, Republic of Korea. Tel.: þ82 55 750 8152; fax: þ82 55 759 1118. E-mail address: [email protected] (W.Y. Paik). 1 Both authors contributed equally to this work. 0143-4004/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.placenta.2010.04.009
Syncytiotrophoblasts secrete many bioactive factors, including corticotropin releasing hormone (CRH), activin-A, inhibin-A, leptin, endoglin, sFlt1, placental growth factor (PlGF), and trophoblast microparticles. These are altered significantly in pre-eclampsia [3]. Of most interest is the soluble receptor for vascular endothelial growth factor (sVEGFR1), also called sFlt1. When present in excess, as in pre-eclampsia, it binds to and inactivates VEGF, a key survival factor for endothelium [4], thereby inducing systemic endothelial dysfunction. Poor placentation leads to poor uteroplacental perfusion and hypoxia, which stimulates sFlt1 production, possibly causing maternal syndrome [3]. The expression of angiogenic factors such as VEGF and its receptor VEGFR1 (also known as Flt1) can be stimulated by the phosphatidylinositol 3-kinase (PI3K) pathway. PI3K stimulates the phosphorylation and activity of Akt, which in turn increases the synthesis of transcription factors such as hypoxia-induced factor-1a (HIF-1a),
622
J.K. Park et al. / Placenta 31 (2010) 621e629
Table 1 Comparison of the study group characteristics. Placenta type
Caesarean section (n) Parity Maternal age (years) Maternal BMI Maternal MAP at delivery (mmHg) Maternal proteinuria (dipstick) Gestational age at delivery (weeks) Neonatal birth weight (g)
p
thereby upregulating VEGF [5] and VEGFR1 expression [6]. Thus, hypoxia not only induces apoptosis but also triggers survival responses such as HIF-1a upregulation via the PI3K/Akt pathway [7]. Placental HIF-1a is upregulated 2-fold in pre-eclampsia [8]. Although reduced placental oxygenation is thought to be involved in the pathogenesis of pre-eclampsia, it is unclear how hypoxia regulates placental sFlt1 expression. This study investigated sFlt1 expression in vivo and in models of human placental hypoxia in vitro to understand the role of the PI3K-Akt pathway in regulating the expression of sFlt1.
Preeclamptic (n ¼ 10)
Control (n ¼ 10)
10 0 (0e2) 26 (20e37) 28.5 (22.2e40.6) 133.3 (113.3e173.3)
10 2 (1e3) 28 (25e31) 27.3 (24.5e38.1) 93.3 (73.3e103.3)
e <0.01 0.086 0.880 <0.01
2.5 (1e4)
0
<0.01
36.3 (35.0e40.0)
38.1 (37.4e38.7)
0.102
2. Materials and methods
2435 (1220e3390)
3060 (2800e3420)
<0.01
2.1. Human placental tissue samples
Data are presented as medians (range) and analyzed using the ManneWhitney Utest. Values of p < 0.05 were considered statistically significant. BMI, body mass index; MAP, mean arterial pressure. MAP ¼ [(2 diastolic) þ systolic]/3.
The experimental protocol used was peer-reviewed and approved by the Human Subject Research Committee of Gyeongsang National University Hospital, Jinju, Korea. Written informed consent, with permission to collect placental tissue, was obtained from 20 pregnant women who were scheduled to undergo Caesarean section without labor at 35e40 weeks of pregnancy. Of these pregnancies, 50% (n ¼ 10) were normal and 50% (n ¼ 10) were preeclamptic.
Fig. 1. Angiogenic factors in preeclamptic placentas. (A) VEGF was measured as w20 kDa bands. (B) PlGF was measured as an w18 kDa band. (C) Flt1 was measured as a w180 kDa band. (D) sFlt1 was measured as a w100 kDa band. (E) The expression of VEGF, PlGF, Flt1, and sFlt1 was significantly higher in preeclamptic placentas (n ¼ 10) than in normal placentas (n ¼ 10). The sFlt1:VEGF ratio was significantly higher in preeclamptic placentas than in normal placentas, although the difference in the sFlt1:PlGF ratio between the two groups was not significant. Each lane represents an individual placenta; 30 mg of total protein were loaded. Alpha-tubulin was used as a positive control. The densities of the immunoreactive bands were analyzed using SigmaPlot 7.0 and are presented in arbitrary units (A.U.) normalized to a-tubulin levels. Data are presented as means SE; * p < 0.05.
J.K. Park et al. / Placenta 31 (2010) 621e629 Pre-eclampsia was diagnosed by the presence of hypertension (blood pressure 140 mmHg systolic or 90 mmHg diastolic after 20 weeks of gestation in previously normotensive women) and proteinuria (urinary excretion 0.3 g of protein in a 24-h specimen, or 30 mg/dL or a reading 1þ on a dipstick in a random urine determination). Severe pre-eclampsia was defined as HELLP syndrome, eclampsia, or pre-eclampsia with either severe hypertension (blood pressure 160 mmHg systolic or 110 mmHg diastolic) or severe proteinuria (proteinuria 2.0 g in a 24-h specimen, or a reading 2þ on a dipstick). All preeclamptic women in this study had “severe” pre-eclampsia on the basis of the ACOG criteria [9]. The control subjects had no evidence of chronic hypertension, pre-eclampsia, or gestational hypertension. No control or preeclamptic subject had been diagnosed with any chronic disease such as hypertension, diabetes mellitus, or renal disease before becoming pregnant. Tissues from five different portions of the placenta were sectioned into fullthickness samples of approximately 3 3 cm. The samples were then re-sectioned into pieces of approximately 1 1 cm, frozen in liquid nitrogen, and kept at 70 C. One 1 1-cm piece from each of the five different portions of each placenta was used for protein extraction. 2.2. Cell lines and cell culture Choriocarcinoma cell line JEG-3 cells were cultured according to the protocol recommended by the American Type Culture Collection (ATCC, Manassas, VA), i.e., in Dulbecco’s modified Eagle’s medium supplemented with 10% heat-inactivated fetal bovine serum (Invitrogen, Carlsbad, CA), 2 mM L-glutamine, 100 U/mL penicillin, and 100 mg/mL streptomycin (Invitrogen) at 37 C in a humidified 5% CO2 incubator. When the cells reached 80% confluence, as judged by eye, the cells were detached by trypsinization. 2.3. Hypoxic treatment Metal compounds such as cobalt chloride (CoCl2), nickel chloride, and desferrioxamine can produce conditions that mimic hypoxia [10]. In this study, we used CoCl2 as a hypoxia-mimicking agent. To determine the appropriate amount of CoCl2 (Wako, Osaka, Japan), 2 105 JEG-3 cells/mL were seeded in 10-cm diameter dishes 24 h before being treated with 0, 100, 200, 400, 600, 800, or 1000 mM CoCl2. For experiments, the JEG-3 cells were cultured in the absence or presence of 600 mM CoCl2 for 3, 6, 12, or 24 h.
623
CoCl2 for 24 h. Then, 30 mL of 1 mg/mL MTT (Sigma Chemical, St. Louis, MO) were added to each well, and the plate was incubated at 37 C until a blue color appeared. The medium was aspirated and replaced with dimethyl sulfoxide. The absorbance was read at 560 nm using a microtiter plate reader. 2.5. Inhibition of the PI3K-Akt pathway LY294002 (Cell Signaling Technology, Beverly, MA) acts as a highly selective inhibitor of phosphatidylinositol 3-kinase (PI3K) in vivo. Cells were preincubated with 50 mM LY294002 or with the solvent dimethyl sulfoxide for 1 h before CoCl2 treatment. 2.6. Western blot analysis Western blot analysis for sFlt1 and other proteins was performed using total placental lysates or JEG-3 cell lysates. For the cultured JEG-3 cells, the medium was aspirated, and the cells were scraped up in 1 mL of 1 PBS. After centrifugation at 130 g for 3 min, the supernatant was discarded, and the collected cells were lysed. The placental tissues and JEG-3 cells were suspended in a lysis buffer containing 50 mM Tris (pH 7.5), 150 mM NaCl, 5 mM EDTA, 1% Nonidet P-40, protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 1 mg/mL aprotinin, and 1 mg/mL leupeptin), and phosphatase inhibitors (1 mM sodium orthovanadate and 1 mM sodium fluoride) and lysed by repeated sonication with 2- to 5-s pulses. The lysates were clarified by centrifugation (10,000 g, 20 min, 4 C), and the protein concentration in the supernatants was determined using a bicinchoninic acid protein assay kit (Pierce, Rockford, IL) with bovine serum albumin as the standard. The lysates were diluted in sample buffer (0.15 M Tris-HCl, pH 6.8, 2% SDS, 10% glycerol, and 5% 2-mercaptoethanol) and boiled for 5 min. Equal amounts of protein (30 mg) were loaded onto the lanes of 12% SDS-polyacrylamide gels, and the proteins were separated by electrophoresis. The separated proteins were transferred to nitrocellulose membranes. The membranes were washed in Tris-buffered saline containing 0.1% Tween 20 (TBST) and incubated with one of the following primary antibodies: anti-HIF-1a, anti-Flt1, anti-VEGF, or anti-PlGF (0.2 mg/mL; Santa Cruz Biotechnology, Santa Cruz, CA); anti-cleaved caspase-3, anti-PI3K, or anti-p-Akt(ser473) (0.2 mg/mL; Cell Signaling Technology); or anti-sFlt1 (1 mg/mL, Zymed Laboratories, San Francisco, CA). The membranes were washed and incubated with secondary antibody (1:10,000; Pierce). Immunoreactive proteins were detected by enhanced chemiluminescence using a Western blot analysis system (Pierce).
2.4. MTT assay 2.7. Statistical analysis Cell viability after exposure to hypoxic conditions was determined by 3-(4,5dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay. Briefly, cells were seeded at 2 105 cells/mL on a 96-well plate, and treated with or without
The clinical data are expressed as the median (range). Comparisons were performed using the Mann-Whitney U-test. The densities of immunoreactive
Fig. 2. PI3K-Akt pathways in preeclamptic placentas. (A) HIF-1a and PI3Kp85a were measured as single w120 and w85 kDa bands, respectively. (B) p-Akt was measured as a w60 kDa band. (C) The expression of HIF-1a, PI3Kp85a, and p-Akt was significantly higher in preeclamptic placentas (n ¼ 10) than in normal placentas (n ¼ 10). The lane loading, controls, and densitometric analyses were as described in Fig. 1. Data are presented as means SE; *p < 0.05.
624
J.K. Park et al. / Placenta 31 (2010) 621e629
bands were analyzed using SigmaGel 1.0 (Jandel Scientific, Erkrath, Germany) and SigmaPlot 7.0 (SPSS, Chicago, IL). The densitometric analysis is presented in arbitrary units (A.U.) normalized to the a-tubulin level. Data are presented as the mean standard error (SE) of three independent experiments. Groups were compared using Student’s unpaired t-test. Values of p < 0.05 were considered significant.
decreased (p < 0.05; Fig. 6B, C), and CoCl2-induced Akt phosphorylation was inhibited for 24 h after LY294002 pretreatment. In addition, the expression of HIF-1 was delayed in cells treated with 600 mM CoCl2 after pretreatment with 50 mM LY294002 (Fig. 6D), compared with HIF-1 expression in cells treated with only 600 mM CoCl2 (Fig. 5B).
3. Results 3.1. Subjects We obtained placenta tissues from 20 pregnant women (10 normal pregnancies and 10 preeclamptic) who were scheduled to undergo Caesarean section without labor at 35e40 weeks of pregnancy. There was a significant (p < 0.01) difference in parity between the women with pre-eclampsia and those in the normal pregnancy control group, who were scheduled for a repeat caesarean delivery (parity 1) (Table 1). The preeclamptic and control groups were similar in maternal age, maternal body mass index, and gestational age at delivery, but the women with preeclampsia had significantly higher mean arterial pressure, significant proteinuria, and significantly lower neonatal birth weight (p < 0.01; Table 1). 3.2. Angiogenic factors in the preeclamptic placentas The expression of VEGF, PlGF, Flt1, and sFlt1 was significantly (p < 0.05) higher in the preeclamptic placentas than in the normal placentas. The sFlt1:VEGF ratio, an index of anti-angiogenic activity, was significantly (p < 0.05) higher in the preeclamptic placentas than in the normal placentas, although the difference in the sFlt1: PlGF ratio between the two groups was not significant (Fig. 1). The expression of HIF-1a, PI3Kp85a, and p-Akt was significantly (p < 0.05) higher in the preeclamptic placentas than in the normal placentas (Fig. 2). 3.3. Effects of CoCl2 on cell viability An MTT assay was performed to determine the level of CoCl2 required to create an appropriate hypoxic environment for the JEG3 cells. Cobalt chloride inhibited the viability of JEG-3 cells in a dosedependent manner (Fig. 3A). The concentration of CoCl2 at which approximately 50% of the cells were viable (IC50) was 600 mM (Fig. 3A); therefore, 600 mM CoCl2 was chosen for this experiment. The expression of cleaved caspase-3, a biochemical marker of cellular apoptosis, increased in a time-dependent manner after 6 h of treatment with 600 mM CoCl2 (p < 0.05; Fig. 3B, C). 3.4. Angiogenic factors in in vitro models of human placental hypoxia The expression of VEGF and PlGF increased in a time-dependent manner (p < 0.05). The expression of sFlt1 plateaued after 3 h of CoCl2 treatment. The sFlt1:VEGF and sFlt1:PlGF ratios were maximal after 3 h of treatment with 600 mM CoCl2 (p < 0.05; Fig. 4). The expression levels of p-Akt and PI3Kp85a were maximal after 6 and 12 h of CoCl2 treatment, respectively (p < 0.05). The expression of HIF-1a increased in a time-dependent manner with 600 mM CoCl2 treatment (p < 0.05; Fig. 5). 3.5. Inhibition of the PI3K-Akt pathway by LY294002 The phosphorylation of Akt was significantly increased in JEG-3 cells treated with CoCl2 for 12 h (p < 0.05; Fig. 6A). In JEG-3 cells pretreated with 20 or 50 mM LY294002 for 1 h before treatment with 600 mM CoCl2, the expression of p-Akt was significantly
Fig. 3. Effects of CoCl2 on cell viability. (A) Human choriocarcinoma JEG-3 cells were treated with different concentrations of CoCl2 for 24 h. Viable cells were measured using the MTT assay and are shown as a percentage of the value of the respective control. The viability of JEG-3 cells was inhibited by CoCl2 in a dose-dependent manner with an IC50 of 600 mM. (B) Investigation of a biochemical marker of cellular apoptosis. Human choriocarcinoma JEG-3 cells were treated with 600 mM CoCl2. Cleaved caspase3 was measured as w17 and w19 kDa bands. (C) The expression of cleaved caspase-3 increased in a time-dependent manner after 6 h of 600 mM CoCl2 treatment. The lane loading, controls, and densitometric analyses were as described in Fig. 1. Data are presented as means SE; *p < 0.05.
J.K. Park et al. / Placenta 31 (2010) 621e629
625
Fig. 4. Angiogenic factors in in vitro models of human placental hypoxia. Human choriocarcinoma JEG-3 cells were treated with 600 mM CoCl2. (A) VEGF was measured as w20 kDa bands. (B) PlGF was measured as an w18 kDa band. (C) Flt1 was measured as a w180 kDa band. (D) sFlt1 was measured as a w100 kDa band. (E) The expression of VEGF and PlGF increased in a time-dependent manner, whereas the expression of sFlt1 plateaued after 3 h of CoCl2 treatment. The sFlt1:VEGF and sFlt1:PlGF ratios were maximal at 3 h of 600 mM CoCl2 treatment. The lane loading, controls, and densitometric analyses were as described in Fig. 1. Data are presented as means SE; * p < 0.05.
3.6. Angiogenic factors in in vitro models of human placental hypoxia with inhibition of the PI3K-Akt pathway by LY294002 JEG-3 cells with or without 50 mM LY294002 pretreatment for 1 h were treated with 600 mM CoCl2 for 6, 12, or 24 h. After pretreatment with 50 mM LY294002, the expression levels of VEGF and PlGF were increased, and the levels remained unchanged or increased after subsequent treatment with 600 mM CoCl2, compared with the expression after 600 mM CoCl2 treatment without LY294002 pretreatment (Fig. 7AeD). The expression of sFlt1 after treatment with 600 mM CoCl2 was significantly lower in cells pretreated with 50 mM LY294002 than in cells treated with only CoCl2 (p < 0.05; Fig. 7E, F).
4. Discussion Vascular endothelial growth factor (VEGF) is an important regulator of vasculogenesis, angiogenesis, and vascular permeability. In contrast to its transient expression during the formation of new blood vessels, VEGF and its receptors are continuously expressed in cultured adult and fetal endothelial cells. Furthermore, VEGF mRNA transcripts accumulate in endothelial cells exposed to hypoxia. The proteinuria and hypertension that characterize pre-eclampsia are reminiscent of the side effects of antiVEGF therapies [11]. Experimental neutralization of free VEGF by the administration of sFlt1 in rats resulted in endothelial dysfunction in vivo [4]. This experimental evidence, as well as the
626
J.K. Park et al. / Placenta 31 (2010) 621e629
Fig. 5. PI3K-Akt pathways in in vitro models of human placental hypoxia. Human choriocarcinoma JEG-3 cells were treated with 600 mM CoCl2. (A) p-Akt, PI3Kp85a, and HIF-1a were measured as single w60, w85, and w120 kDa bands, respectively. (B) The expression of p-Akt and PI3Kp85a was maximal after 6 and 12 h of CoCl2 treatment, respectively. The expression of HIF-1a increased in a time-dependent manner with 600 mM CoCl2 treatment. The lane loading, controls, and densitometric analyses were as described in Fig. 1. Data are presented as means SE; * p < 0.05.
pathological findings in humans with pre-eclampsia, further supports a role of VEGF in maintaining normal blood vessels and provides clues regarding the tissues in which VEGF action is most important in the adult. Little is known of the role of placental growth factor (PlGF), a VEGF homolog. The loss of PlGF impairs angiogenesis, plasma extravasation, and collateral growth during ischemia, inflammation, wound healing, and cancer [12]. A report that the inhibition of both VEGF and PlGF is required to cause pre-eclampsia-like changes in pregnant rats implies that the inhibition of PlGF plays a critical role in sFlt1-induced endothelial dysfunction [4]. Our Western blot analysis indicated that the expression levels of VEGF, sFlt1, and PlGF were higher in the preeclamptic placentas than in the normal placentas. In in vitro studies, villous trophoblasts under hypoxia expressed higher levels of VEGF and sFlt1 mRNAs and secreted more VEGF and sFlt1 proteins, whereas PlGF mRNA and protein were decreased in hypoxia [13,14]. However, reports in preeclamptic placentas provide conflicting conclusions regarding
the expression of VEGF, sFlt1, and PlGF. In severe pre-eclampsia, immunostaining for cytotrophoblast VEGF and sFlt1 was decreased and immunostaining for cytotrophoblast PlGF was unaffected [15]; the PlGF mRNA level was increased in preeclamptic placenta compared with normal placentas [13]. Discrepancies in the findings between studies using preeclamptic placenta and in vitro models may result because a pre-eclampsia model with hypoxia alone is not sufficient for investigating the generation of placental oxidative stress in pre-eclampsia; an in vitro hypoxia/reoxygenation model may be a more appropriate model system [16]. Moreover, the unique characteristics of each placental compartment may explain the different findings in preeclamptic placentas, because anchoring villi, invasive cytotrophoblasts, and blood vessels are not distributed evenly throughout the placenta [15]. The sF1t1:PlGF ratio has been suggested as an index of antiangiogenic activity that reflects the changes in both markers [17] and has been reported to be a better indicator of pre-eclampsia than separate measurements of each marker [18]. In the present
J.K. Park et al. / Placenta 31 (2010) 621e629
627
Fig. 6. Inhibition of the PI3K-Akt pathway with LY294002, a known pharmacologic inhibitor of PI3K. (A) Human choriocarcinoma JEG-3 cells were treated with different concentrations of CoCl2 for 12 h; p-Akt was measured as a w60 kDa band. The expression of p-Akt increased significantly in JEG-3 cells treated with CoCl2 for 12 h. (B) Human choriocarcinoma JEG-3 cells were pretreated with different concentrations of LY294002 for 1 h; p-Akt was measured as a w60 kDa band. In JEG-3 cells pretreated with 20 or 50 mM LY294002 for 1 h before treatment with 600 mM CoCl2, the expression of p-Akt was decreased significantly. (C and D) Human choriocarcinoma JEG-3 cells were pretreated with 50 mM LY294002 for 1 h and then treated with 600 mM CoCl2. The CoCl2-induced Akt phosphorylation was inhibited for 24 h after LY294002 pretreatment. The expression of HIF-1 was delayed in cells treated with 600 mM CoCl2 after pretreatment with 50 mM LY294002 (Fig. 6D), compared with HIF-1 expression in cells treated with only 600 mM CoCl2 (Fig. 5B). The lane loading, controls, and densitometric analyses were as described in Fig. 1. Data are presented as means SE; *p < 0.05.
study, the index of anti-angiogenic activity was measured from placental tissue, and not from maternal serum as in other studies. As a result, the sF1t1:VEGF ratio differed more significantly than did the sF1t1:PlGF ratio. Although PlGF itself does not significantly stimulate angiogenesis in vivo or in vitro, this growth factor modulates the angiogenic activity of VEGF by forming PlGF/VEGF heterodimers [19]. Thus, the amount of VEGF also influences the complex placental etiology. The PI3K-Akt pathway regulates normal cellular functions such as survival, proliferation, and glucose metabolism. Depending on the cell type, hypoxia can either activate or inactivate the PI3K-Akt signaling pathway. For example, the pathway is activated by hypoxia in PC12, HT1080, and HeLa cells [20e22], but is inactivated
in HepG2, BeWo, BeWo31, and JAR placental cells and in primary cytotrophoblast cells [23e25] under hypoxic conditions. Although the mechanisms of activation or inhibition of the PI3KAkt signaling pathway by hypoxia are unknown, we demonstrated in the present study that this pathway is activated in preeclamptic placentas and in JEG cells in the presence of CoCl2, a hypoxiamimicking agent (within the initial 24 h). We also cultured JEG-3 cells in 1.5% O2 for 48 h to produce hypoxic conditions (data not shown) and observed that PI3K expression reached a peak at 6e12 h, but decreased by 48 h; p-Akt reached a peak at 3e6 h, but also decreased by 48 h; and HIF-1a increased continuously with time. These results were similar to our results using CoCl2 as a hypoxia-mimicking agent, whereas Chiang MH et al. [25] showed
628
J.K. Park et al. / Placenta 31 (2010) 621e629
Fig. 7. Angiogenic and anti-angiogenic factors in in vitro models of human placental hypoxia with inhibition of the PI3K-Akt pathway by LY294002. (A) Human choriocarcinoma JEG-3 cells were treated with 600 mM CoCl2 for 12 h with or without 50 mM LY294002 pretreatment for 1 h; VEGF was measured as w20 kDa bands. The level of VEGF expression increased with 50 mM LY294002 pretreatment compared with the control and increased when treated with 600 mM CoCl2 after 50 mM LY294002 pretreatment, as compared with 600 mM CoCl2 treatment alone. (B) Human choriocarcinoma JEG-3 cells were treated with 600 mM CoCl2 for 24 h with or without 50 mM LY294002 pretreatment for 1 h; VEGF was measured as w20 kDa bands. The level of VEGF expression increased with 50 mM LY294002 pretreatment compared with the control, and remained unchanged when treated with 600 mM CoCl2 after 50 mM LY294002 pretreatment as compared with 600 mM CoCl2 treatment alone. (C) Human choriocarcinoma JEG-3 cells were treated with 600 mM CoCl2 for 12 h with or without 50 mM LY294002 pretreatment for 1 h; PlGF was measured as an w18 kDa band. The PlGF expression increased with 50 mM LY294002 pretreatment compared with the control and remained unchanged when treated with 600 mM CoCl2 after 50 mM LY294002 pretreatment, as compared with 600 mM CoCl2 treatment alone. (D) Human choriocarcinoma JEG-3 cells were treated with 600 mM CoCl2 for 24 h with or without 50 mM LY294002 pretreatment for 1 h. PlGF was measured as a w18 kDa band. The PlGF expression increased with 50 mM LY294002 pretreatment compared with the control, and increased when treated with 600 mM CoCl2 after 50 mM LY294002 pretreatment as compared with 600 mM CoCl2 treatment alone. (E) Human choriocarcinoma JEG-3 cells were treated with 600 mM CoCl2 for 6 h with or without 50 mM LY294002 pretreatment for 1 h. There was a significant decrease in the expression of sFlt1 with 50 mM LY294002 pretreatment compared with the control, and a significant decrease when treated with 600 mM CoCl2 after 50 mM LY294002 pretreatment as compared with 600 mM CoCl2 treatment alone. (F) Human choriocarcinoma JEG-3 cells were treated with 600 mM CoCl2 for 24 h with or without 50 mM LY294002 pretreatment for 1 h. There was a significant decrease in the expression of sFlt1 with 600 mM CoCl2 treatment after 50 mM LY294002 pretreatment as compared with 600 mM CoCl2 treatment alone. The lane loading, controls, and densitometric analyses were as described in Fig. 1. Data are presented as means SE; *p < 0.05.
that hypoxia (1% O2 for 48 h) down-modulated both PI3K and p-Akt in cytotrophoblasts from normal placentas and in BeWo and JAR cells. There are two factors that may account for these apparently conflicting results. First, Chiang et al. [25] focused on the late responses, by which time apoptosis had advanced considerably; in contrast, the present study focused on early changes, which included sFlt1 activation. Second, Hung et al. [16] reported, an in vitro pre-eclampsia model with hypoxia alone, which was used by Chiang et al. [25], may not be sufficient for investigating placental oxidative stress in pre-eclampsia, and a model with hypoxia/reoxygenation may be more suitable. Given that the increased levels of PI3K, p-Akt, and HIF-1a that were among the early changes in JEG-3 cells cultured under CoCl2 are also seen in preeclamptic placenta, we consider our model to be more appropriate than the placental oxidative stress model of Chiang et al. [25] for investigating the generation of placental oxidative stress in pre-eclampsia. JEG-3 cells were chosen as an in vitro model system in the present study because according to Zhou et al. [15], among the choriocarcinoma cell lines JAR, JEG, and BeWo, the production of PlGF and VEGF in JEG cells was most similar to that in third-trimester cytotrophoblasts. Currently, the only treatment for pre-eclampsia is the delivery of the fetus and placenta. However, delivery of the fetus in the
late second or early third trimester, although necessary for the safety of the mother, can result in significant neonatal complications and possibly death owing to the immaturity of the premature neonate. The identification of sFlt1, which represents a critical connection between placental pathology and maternal endothelial dysfunction, may provide a target for effective treatment of pre-eclampsia. The ideal treatment would be to restore and maintain the balance between angiogenic and anti-angiogenic factors, i.e., to restore the biological activities of angiogenic factors such as VEGF and PlGF and overcome the effects of anti-angiogenic factors such as sFlt1. In a recent report, recombinant VEGF 121 treatment resulted in significant improvements in systolic blood pressure, glomerular endotheliosis, and proteinuria in a pregnant rat model of preeclampsia induced by sFlt1 overexpression, without apparent harm to the fetus [26]. Here, we presented the results of PI3K-Akt pathway inhibition in in vitro models of human placental hypoxia. We demonstrated that human placental hypoxia models have high levels of sFlt1 at baseline and that inhibition of the PI3K-Akt pathway by a PI3Kspecific inhibitor leads to a decreased sFlt1 level and unchanged or increased VEGF and PlGF levels.
J.K. Park et al. / Placenta 31 (2010) 621e629
Inhibition of the PI3K-Akt pathway may be a useful therapeutic approach, if it were to decrease sFlt1 secretion without inhibiting VEGF or PlGF secretion. This pathway provides a potential target for a new treatment strategy in patients with pre-eclampsia. Acknowledgements This research was supported by a grant from the National Research Foundation of Korea (NRF) funded by the Korean government (MEST) (2009-0067486) and a 2009 special clinical grant from Gyeongsang National University Hospital. Appendix. Supplementary data Supplementary data associated with this article can be found in the online version, at doi:10.1016/j.placenta.2010.04.009. References [1] Gaber LW, Spargo BH, Lindheimer MD. Renal pathology in pre-eclampsia. Baillière’s Clin Obstet Gynaecol 1994;8:443e68. [2] Roberts JM, Taylor RN, Musci TJ, Rodgers GM, Hubel CA, McLaughlin MK. Preeclampsia: an endothelial cell disorder. Am J Obstet Gynecol 1989;161: 1200e4. [3] Redman CWG, Sargent IL. Placental stress and pre-eclampsia: a revised view. Placenta 2009;30:S38e42. [4] Maynard SE, Min JY, Merchan J, Lim KH, Li J, Mondal S, et al. Excess placental soluble fms-like tyrosine kinase 1 (sFlt1) may contribute to endothelial dysfunction, hypertension, and proteinuria in preeclampsia. J Clin Invest 2003;111:649e58. [5] Tan C, Cruet-Hennequart S, Troussard A, Fazli L, Costello P, Sutton K, et al. Regulation of tumor angiogenesis by integrin-linked kinase (ILK). Cancer Cell 2004;5:79e80. [6] Okuyama H, Krishnamachary B, Zhou YF, Nagasawa H, Bosch-Marce M, Semenza GL. Expression of vascular endothelial growth factor receptor 1 in bone marrow-derived mesenchymal cells is dependent on hypoxia-inducible factor 1. J Biol Chem 2006;281:15554e63. [7] Martorell L, Gentile M, Rius J, Rodríguez C, Crespo J, Badimon L, et al. The hypoxia-inducible factor 1/NOR-1 axis regulates the survival response of endothelial cells to hypoxia. Mol Cell Biol 2009;29:5828e42. [8] Jeyabalan A, McGonigal S, Gilmour C, Hubel CA, Rajakumar A. Circulating and placental endoglin concentrations in pregnancies complicated by intrauterine growth restriction and preeclampsia. Placenta 2008;29:555e63. [9] American College of Obstetricians and Gynecologists. ACOG practice bulletin. Diagnosis and management of preeclampsia and eclampsia. Int J Gynaecol Obstet 2002;77:67e75. [10] Goldberg MA, Dunning SP, Bunn HF. Regulation of erythropoietin gene: evidence that the oxygen sensor is a heme protein. Science 1988;242:1412e4.
629
[11] Hurwitz H, Fehrenbacher L, Novotny W, Cartwright T, Hainsworth J, Heim W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004;350:2335e42. [12] Carmeliet P, Moons L, Luttun A, Vincenti V, Compernolle V, De Mol M, et al. Synergism between vascular endothelial growth factor and placental growth factor contributes to angiogenesis and plasma extravasation in pathological conditions. Nat Med 2001;7:575e83. [13] Ahmed A, Dunk C, Ahmad S, Khaliq A. Regulation of placental vascular endothelial growth factor (VEGF) and placenta growth factor (PlGF) and soluble Flt-1 by oxygenea review. Placenta 2000;21:S16e24. [14] Munaut C, Lorquet S, Pequeux C, Blacher S, Berndt S, Frankenne F, et al. Hypoxia is responsible for soluble vascular endothelial growth factor receptor-1 (VEGFR-1) but not for soluble endoglin induction in villous trophoblast. Hum Reprod 2008;23:1407e15. [15] Zhou Y, McMaster M, Woo K, Janatpour M, Perry J, Karpanen T, et al. Vascular endothelial growth factor ligands and receptors that regulate human cytotrophoblast survival are dysregulated in severe preeclampsia and hemolysis, elevated liver enzymes, and low platelets syndrome. Am J Pathol 2002;160:1405e23. [16] Hung TH, Skepper JN, Burton GJ. In vitro ischemia-reperfusion injury in term human placenta as a model for oxidative stress in pathological pregnancies. Am J Pathol 2001;159:1031e43. [17] Levine RJ, Thadhani R, Qian C, Lam C, Lim KH, Yu KF, et al. Urinary placental growth factor and risk of preeclampsia. JAMA 2005;293:77e85. [18] Levine RJ, Lam C, Qian C, Yu KF, Maynard SE, Sachs BP, et al. Soluble endoglin and other circulating antiangiogenic factors in preeclampsia. N Engl J Med 2006;355:992e1005. [19] Cao Y, Chen H, Zhou L, Chiang MK, Anand-Apte B, Weatherbee JA, et al. Heterodimers of placenta growth factor/vascular endothelial growth factor. Endothelial activity, tumor cell expression, and high affinity binding to Flk-1/ KDR. J Biol Chem 1996;271:3154e62. [20] Beitner-Johnson D, Rust RT, Hsieh TC, Millhorn DE. Hypoxia activates Akt and induces phosphorylation of GSK-3 in PC12 cells. Cell Signal 2001;13:23e7. [21] Chen EY, Mazure NM, Cooper JA, Giaccia AJ. Hypoxia activates a plateletderived growth factor receptor/phosphatidylinositol 3-kinase/Akt pathway that results in glycogen synthase kinase-3 inactivation. Cancer Res 2001;61:2429e33. [22] Alvarez-Tejado M, Alfranca A, Aragonés J, Vara A, Landázuri MO, del Peso L. Lack of evidence for the involvement of the phosphoinositide 3-kinase/Akt pathway in the activation of hypoxia-inducible factors by low oxygen tension. J Biol Chem 2002;277:13508e17. [23] Mottet D, Dumont V, Deccache Y, Demazy C, Ninane N, Raes M, et al. Regulation of hypoxia-inducible factor-1a protein level during hypoxic conditions by the phosphatidylinositol 3-kinase/Akt/glycogen synthase kinase 3b pathway in HepG2 Cells. J Biol Chem 2003;278:31277e85. [24] Flügel D, Görlach A, Michiels C, Kietzmann T. Glycogen synthase kinase 3 phosphorylates hypoxia-inducible factor 1 and mediates its destabilization in a VHL-independent manner. Mol Cell Biol 2007;27:3253e65. [25] Chiang MH, Liang FY, Chen CP, Chang CW, Cheong ML, Wang LJ, et al. Mechanism of hypoxia-induced GCM1 degradation: implications for the pathogenesis of preeclampsia. J Biol Chem 2009;284:17411e9. [26] Li Z, Zhang Y, Ying Ma J, Kapoun AM, Shao Q, Kerr I, et al. Recombinant vascular endothelial growth factor 121 attenuates hypertension and improves kidney damage in a rat model of preeclampsia. Hypertension 2007;50: 686e92.
Contents lists available at ScienceDirect
Placenta journal homepage: www.elsevier.com/locate/placenta
Inhibition of the PI3K-Akt Pathway Suppresses sFlt1 Expression in Human Placental Hypoxia Models In Vitro J.K. Park a, b, J.W. Jeong a, M.Y. Kang b, J.C. Baek a, J.K. Shin a, b, S.A. Lee a, b, W.S. Choi b, c, J.H. Lee a, b,1, W.Y. Paik a, b, *,1 a b c
Department of Obstetrics and Gynecology, School of Medicine, Gyeongsang National University, JinJu, Republic of Korea Institute of Health Sciences, School of Medicine, Gyeongsang National University, JinJu, Republic of Korea Department of Anatomy, School of Medicine, Gyeongsang National University, JinJu, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history: Accepted 14 April 2010
Objective: Although elevated expression of soluble fms-like tyrosine kinase 1 (sFlt1) plays a major role in the pathogenesis of pre-eclampsia, it is unclear how hypoxia regulates placental sFlt1 expression. Thus, we investigated sFlt1 expression in placentas from normal and preeclamptic pregnancies and in human placental hypoxia models in vitro to examine the role of the PI3K-Akt pathway in regulating the expression of this molecule. Methods: We examined the expression of VEGF, PlGF, sFlt1, PI3K, Akt, and HIF-1 in placental samples from ten women with pre-eclampsia and ten normotensive control patients and in human choriocarcinoma trophoblast cells treated with 600 mM CoCl2 by Western blotting. Using models of placental hypoxia, we also determined whether inhibition of the PI3K-Akt pathway plays a direct role in regulating the expression of sFlt1. Results: The VEGF, PlGF, sFlt1, PI3K, Akt, and HIF-1 levels were significantly higher in the preeclamptic placentas than the normal placentas. In the placental hypoxia models, the expression of VEGF and PlGF increased in a time-dependent manner, whereas the expression of sFlt1 plateaued after 3 h of CoCl2 treatment. The expression levels of p-Akt and PI3K were maximal after 6 and 12 h of CoCl2 treatment, respectively. The expression of HIF-1a increased in a time-dependent manner with CoCl2 treatment. Inhibition of the PI3K-Akt pathway with the PI3K-specific inhibitor LY294002 leads to decreased sFlt1 levels and unchanged or increased VEGF and PlGF levels. Conclusion: Inhibition of the PI3K-Akt pathway may be a useful therapeutic approach, if it were to decrease sFlt1 secretion without inhibiting VEGF or PlGF secretion. This pathway provides a potential target for a new treatment strategy in patients with pre-eclampsia. Ó 2010 Elsevier Ltd. All rights reserved.
Keywords: Placental hypoxia model sFlt1 PI3K inhibitor
1. Introduction Pre-eclampsia is a maternal syndrome defined in terms of its cardiovascular and renal features, i.e., new hypertension and proteinuria that resolve after delivery. The proteinuria is associated with a specific endothelial renal lesion called glomerular endotheliosis [1,2], and wide-ranging evidence indicates that the hypertension is secondary to diffuse endothelial dysfunction [3]. Thus, pre-eclampsia is now considered an endothelial disease [3].
* Corresponding author. Department of Obstetrics and Gynecology, College of Medicine, Gyeongsang National University, 90 Chilam-dong, Jinju, 660-702, Republic of Korea. Tel.: þ82 55 750 8152; fax: þ82 55 759 1118. E-mail address: [email protected] (W.Y. Paik). 1 Both authors contributed equally to this work. 0143-4004/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.placenta.2010.04.009
Syncytiotrophoblasts secrete many bioactive factors, including corticotropin releasing hormone (CRH), activin-A, inhibin-A, leptin, endoglin, sFlt1, placental growth factor (PlGF), and trophoblast microparticles. These are altered significantly in pre-eclampsia [3]. Of most interest is the soluble receptor for vascular endothelial growth factor (sVEGFR1), also called sFlt1. When present in excess, as in pre-eclampsia, it binds to and inactivates VEGF, a key survival factor for endothelium [4], thereby inducing systemic endothelial dysfunction. Poor placentation leads to poor uteroplacental perfusion and hypoxia, which stimulates sFlt1 production, possibly causing maternal syndrome [3]. The expression of angiogenic factors such as VEGF and its receptor VEGFR1 (also known as Flt1) can be stimulated by the phosphatidylinositol 3-kinase (PI3K) pathway. PI3K stimulates the phosphorylation and activity of Akt, which in turn increases the synthesis of transcription factors such as hypoxia-induced factor-1a (HIF-1a),
622
J.K. Park et al. / Placenta 31 (2010) 621e629
Table 1 Comparison of the study group characteristics. Placenta type
Caesarean section (n) Parity Maternal age (years) Maternal BMI Maternal MAP at delivery (mmHg) Maternal proteinuria (dipstick) Gestational age at delivery (weeks) Neonatal birth weight (g)
p
thereby upregulating VEGF [5] and VEGFR1 expression [6]. Thus, hypoxia not only induces apoptosis but also triggers survival responses such as HIF-1a upregulation via the PI3K/Akt pathway [7]. Placental HIF-1a is upregulated 2-fold in pre-eclampsia [8]. Although reduced placental oxygenation is thought to be involved in the pathogenesis of pre-eclampsia, it is unclear how hypoxia regulates placental sFlt1 expression. This study investigated sFlt1 expression in vivo and in models of human placental hypoxia in vitro to understand the role of the PI3K-Akt pathway in regulating the expression of sFlt1.
Preeclamptic (n ¼ 10)
Control (n ¼ 10)
10 0 (0e2) 26 (20e37) 28.5 (22.2e40.6) 133.3 (113.3e173.3)
10 2 (1e3) 28 (25e31) 27.3 (24.5e38.1) 93.3 (73.3e103.3)
e <0.01 0.086 0.880 <0.01
2.5 (1e4)
0
<0.01
36.3 (35.0e40.0)
38.1 (37.4e38.7)
0.102
2. Materials and methods
2435 (1220e3390)
3060 (2800e3420)
<0.01
2.1. Human placental tissue samples
Data are presented as medians (range) and analyzed using the ManneWhitney Utest. Values of p < 0.05 were considered statistically significant. BMI, body mass index; MAP, mean arterial pressure. MAP ¼ [(2 diastolic) þ systolic]/3.
The experimental protocol used was peer-reviewed and approved by the Human Subject Research Committee of Gyeongsang National University Hospital, Jinju, Korea. Written informed consent, with permission to collect placental tissue, was obtained from 20 pregnant women who were scheduled to undergo Caesarean section without labor at 35e40 weeks of pregnancy. Of these pregnancies, 50% (n ¼ 10) were normal and 50% (n ¼ 10) were preeclamptic.
Fig. 1. Angiogenic factors in preeclamptic placentas. (A) VEGF was measured as w20 kDa bands. (B) PlGF was measured as an w18 kDa band. (C) Flt1 was measured as a w180 kDa band. (D) sFlt1 was measured as a w100 kDa band. (E) The expression of VEGF, PlGF, Flt1, and sFlt1 was significantly higher in preeclamptic placentas (n ¼ 10) than in normal placentas (n ¼ 10). The sFlt1:VEGF ratio was significantly higher in preeclamptic placentas than in normal placentas, although the difference in the sFlt1:PlGF ratio between the two groups was not significant. Each lane represents an individual placenta; 30 mg of total protein were loaded. Alpha-tubulin was used as a positive control. The densities of the immunoreactive bands were analyzed using SigmaPlot 7.0 and are presented in arbitrary units (A.U.) normalized to a-tubulin levels. Data are presented as means SE; * p < 0.05.
J.K. Park et al. / Placenta 31 (2010) 621e629 Pre-eclampsia was diagnosed by the presence of hypertension (blood pressure 140 mmHg systolic or 90 mmHg diastolic after 20 weeks of gestation in previously normotensive women) and proteinuria (urinary excretion 0.3 g of protein in a 24-h specimen, or 30 mg/dL or a reading 1þ on a dipstick in a random urine determination). Severe pre-eclampsia was defined as HELLP syndrome, eclampsia, or pre-eclampsia with either severe hypertension (blood pressure 160 mmHg systolic or 110 mmHg diastolic) or severe proteinuria (proteinuria 2.0 g in a 24-h specimen, or a reading 2þ on a dipstick). All preeclamptic women in this study had “severe” pre-eclampsia on the basis of the ACOG criteria [9]. The control subjects had no evidence of chronic hypertension, pre-eclampsia, or gestational hypertension. No control or preeclamptic subject had been diagnosed with any chronic disease such as hypertension, diabetes mellitus, or renal disease before becoming pregnant. Tissues from five different portions of the placenta were sectioned into fullthickness samples of approximately 3 3 cm. The samples were then re-sectioned into pieces of approximately 1 1 cm, frozen in liquid nitrogen, and kept at 70 C. One 1 1-cm piece from each of the five different portions of each placenta was used for protein extraction. 2.2. Cell lines and cell culture Choriocarcinoma cell line JEG-3 cells were cultured according to the protocol recommended by the American Type Culture Collection (ATCC, Manassas, VA), i.e., in Dulbecco’s modified Eagle’s medium supplemented with 10% heat-inactivated fetal bovine serum (Invitrogen, Carlsbad, CA), 2 mM L-glutamine, 100 U/mL penicillin, and 100 mg/mL streptomycin (Invitrogen) at 37 C in a humidified 5% CO2 incubator. When the cells reached 80% confluence, as judged by eye, the cells were detached by trypsinization. 2.3. Hypoxic treatment Metal compounds such as cobalt chloride (CoCl2), nickel chloride, and desferrioxamine can produce conditions that mimic hypoxia [10]. In this study, we used CoCl2 as a hypoxia-mimicking agent. To determine the appropriate amount of CoCl2 (Wako, Osaka, Japan), 2 105 JEG-3 cells/mL were seeded in 10-cm diameter dishes 24 h before being treated with 0, 100, 200, 400, 600, 800, or 1000 mM CoCl2. For experiments, the JEG-3 cells were cultured in the absence or presence of 600 mM CoCl2 for 3, 6, 12, or 24 h.
623
CoCl2 for 24 h. Then, 30 mL of 1 mg/mL MTT (Sigma Chemical, St. Louis, MO) were added to each well, and the plate was incubated at 37 C until a blue color appeared. The medium was aspirated and replaced with dimethyl sulfoxide. The absorbance was read at 560 nm using a microtiter plate reader. 2.5. Inhibition of the PI3K-Akt pathway LY294002 (Cell Signaling Technology, Beverly, MA) acts as a highly selective inhibitor of phosphatidylinositol 3-kinase (PI3K) in vivo. Cells were preincubated with 50 mM LY294002 or with the solvent dimethyl sulfoxide for 1 h before CoCl2 treatment. 2.6. Western blot analysis Western blot analysis for sFlt1 and other proteins was performed using total placental lysates or JEG-3 cell lysates. For the cultured JEG-3 cells, the medium was aspirated, and the cells were scraped up in 1 mL of 1 PBS. After centrifugation at 130 g for 3 min, the supernatant was discarded, and the collected cells were lysed. The placental tissues and JEG-3 cells were suspended in a lysis buffer containing 50 mM Tris (pH 7.5), 150 mM NaCl, 5 mM EDTA, 1% Nonidet P-40, protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 1 mg/mL aprotinin, and 1 mg/mL leupeptin), and phosphatase inhibitors (1 mM sodium orthovanadate and 1 mM sodium fluoride) and lysed by repeated sonication with 2- to 5-s pulses. The lysates were clarified by centrifugation (10,000 g, 20 min, 4 C), and the protein concentration in the supernatants was determined using a bicinchoninic acid protein assay kit (Pierce, Rockford, IL) with bovine serum albumin as the standard. The lysates were diluted in sample buffer (0.15 M Tris-HCl, pH 6.8, 2% SDS, 10% glycerol, and 5% 2-mercaptoethanol) and boiled for 5 min. Equal amounts of protein (30 mg) were loaded onto the lanes of 12% SDS-polyacrylamide gels, and the proteins were separated by electrophoresis. The separated proteins were transferred to nitrocellulose membranes. The membranes were washed in Tris-buffered saline containing 0.1% Tween 20 (TBST) and incubated with one of the following primary antibodies: anti-HIF-1a, anti-Flt1, anti-VEGF, or anti-PlGF (0.2 mg/mL; Santa Cruz Biotechnology, Santa Cruz, CA); anti-cleaved caspase-3, anti-PI3K, or anti-p-Akt(ser473) (0.2 mg/mL; Cell Signaling Technology); or anti-sFlt1 (1 mg/mL, Zymed Laboratories, San Francisco, CA). The membranes were washed and incubated with secondary antibody (1:10,000; Pierce). Immunoreactive proteins were detected by enhanced chemiluminescence using a Western blot analysis system (Pierce).
2.4. MTT assay 2.7. Statistical analysis Cell viability after exposure to hypoxic conditions was determined by 3-(4,5dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay. Briefly, cells were seeded at 2 105 cells/mL on a 96-well plate, and treated with or without
The clinical data are expressed as the median (range). Comparisons were performed using the Mann-Whitney U-test. The densities of immunoreactive
Fig. 2. PI3K-Akt pathways in preeclamptic placentas. (A) HIF-1a and PI3Kp85a were measured as single w120 and w85 kDa bands, respectively. (B) p-Akt was measured as a w60 kDa band. (C) The expression of HIF-1a, PI3Kp85a, and p-Akt was significantly higher in preeclamptic placentas (n ¼ 10) than in normal placentas (n ¼ 10). The lane loading, controls, and densitometric analyses were as described in Fig. 1. Data are presented as means SE; *p < 0.05.
624
J.K. Park et al. / Placenta 31 (2010) 621e629
bands were analyzed using SigmaGel 1.0 (Jandel Scientific, Erkrath, Germany) and SigmaPlot 7.0 (SPSS, Chicago, IL). The densitometric analysis is presented in arbitrary units (A.U.) normalized to the a-tubulin level. Data are presented as the mean standard error (SE) of three independent experiments. Groups were compared using Student’s unpaired t-test. Values of p < 0.05 were considered significant.
decreased (p < 0.05; Fig. 6B, C), and CoCl2-induced Akt phosphorylation was inhibited for 24 h after LY294002 pretreatment. In addition, the expression of HIF-1 was delayed in cells treated with 600 mM CoCl2 after pretreatment with 50 mM LY294002 (Fig. 6D), compared with HIF-1 expression in cells treated with only 600 mM CoCl2 (Fig. 5B).
3. Results 3.1. Subjects We obtained placenta tissues from 20 pregnant women (10 normal pregnancies and 10 preeclamptic) who were scheduled to undergo Caesarean section without labor at 35e40 weeks of pregnancy. There was a significant (p < 0.01) difference in parity between the women with pre-eclampsia and those in the normal pregnancy control group, who were scheduled for a repeat caesarean delivery (parity 1) (Table 1). The preeclamptic and control groups were similar in maternal age, maternal body mass index, and gestational age at delivery, but the women with preeclampsia had significantly higher mean arterial pressure, significant proteinuria, and significantly lower neonatal birth weight (p < 0.01; Table 1). 3.2. Angiogenic factors in the preeclamptic placentas The expression of VEGF, PlGF, Flt1, and sFlt1 was significantly (p < 0.05) higher in the preeclamptic placentas than in the normal placentas. The sFlt1:VEGF ratio, an index of anti-angiogenic activity, was significantly (p < 0.05) higher in the preeclamptic placentas than in the normal placentas, although the difference in the sFlt1: PlGF ratio between the two groups was not significant (Fig. 1). The expression of HIF-1a, PI3Kp85a, and p-Akt was significantly (p < 0.05) higher in the preeclamptic placentas than in the normal placentas (Fig. 2). 3.3. Effects of CoCl2 on cell viability An MTT assay was performed to determine the level of CoCl2 required to create an appropriate hypoxic environment for the JEG3 cells. Cobalt chloride inhibited the viability of JEG-3 cells in a dosedependent manner (Fig. 3A). The concentration of CoCl2 at which approximately 50% of the cells were viable (IC50) was 600 mM (Fig. 3A); therefore, 600 mM CoCl2 was chosen for this experiment. The expression of cleaved caspase-3, a biochemical marker of cellular apoptosis, increased in a time-dependent manner after 6 h of treatment with 600 mM CoCl2 (p < 0.05; Fig. 3B, C). 3.4. Angiogenic factors in in vitro models of human placental hypoxia The expression of VEGF and PlGF increased in a time-dependent manner (p < 0.05). The expression of sFlt1 plateaued after 3 h of CoCl2 treatment. The sFlt1:VEGF and sFlt1:PlGF ratios were maximal after 3 h of treatment with 600 mM CoCl2 (p < 0.05; Fig. 4). The expression levels of p-Akt and PI3Kp85a were maximal after 6 and 12 h of CoCl2 treatment, respectively (p < 0.05). The expression of HIF-1a increased in a time-dependent manner with 600 mM CoCl2 treatment (p < 0.05; Fig. 5). 3.5. Inhibition of the PI3K-Akt pathway by LY294002 The phosphorylation of Akt was significantly increased in JEG-3 cells treated with CoCl2 for 12 h (p < 0.05; Fig. 6A). In JEG-3 cells pretreated with 20 or 50 mM LY294002 for 1 h before treatment with 600 mM CoCl2, the expression of p-Akt was significantly
Fig. 3. Effects of CoCl2 on cell viability. (A) Human choriocarcinoma JEG-3 cells were treated with different concentrations of CoCl2 for 24 h. Viable cells were measured using the MTT assay and are shown as a percentage of the value of the respective control. The viability of JEG-3 cells was inhibited by CoCl2 in a dose-dependent manner with an IC50 of 600 mM. (B) Investigation of a biochemical marker of cellular apoptosis. Human choriocarcinoma JEG-3 cells were treated with 600 mM CoCl2. Cleaved caspase3 was measured as w17 and w19 kDa bands. (C) The expression of cleaved caspase-3 increased in a time-dependent manner after 6 h of 600 mM CoCl2 treatment. The lane loading, controls, and densitometric analyses were as described in Fig. 1. Data are presented as means SE; *p < 0.05.
J.K. Park et al. / Placenta 31 (2010) 621e629
625
Fig. 4. Angiogenic factors in in vitro models of human placental hypoxia. Human choriocarcinoma JEG-3 cells were treated with 600 mM CoCl2. (A) VEGF was measured as w20 kDa bands. (B) PlGF was measured as an w18 kDa band. (C) Flt1 was measured as a w180 kDa band. (D) sFlt1 was measured as a w100 kDa band. (E) The expression of VEGF and PlGF increased in a time-dependent manner, whereas the expression of sFlt1 plateaued after 3 h of CoCl2 treatment. The sFlt1:VEGF and sFlt1:PlGF ratios were maximal at 3 h of 600 mM CoCl2 treatment. The lane loading, controls, and densitometric analyses were as described in Fig. 1. Data are presented as means SE; * p < 0.05.
3.6. Angiogenic factors in in vitro models of human placental hypoxia with inhibition of the PI3K-Akt pathway by LY294002 JEG-3 cells with or without 50 mM LY294002 pretreatment for 1 h were treated with 600 mM CoCl2 for 6, 12, or 24 h. After pretreatment with 50 mM LY294002, the expression levels of VEGF and PlGF were increased, and the levels remained unchanged or increased after subsequent treatment with 600 mM CoCl2, compared with the expression after 600 mM CoCl2 treatment without LY294002 pretreatment (Fig. 7AeD). The expression of sFlt1 after treatment with 600 mM CoCl2 was significantly lower in cells pretreated with 50 mM LY294002 than in cells treated with only CoCl2 (p < 0.05; Fig. 7E, F).
4. Discussion Vascular endothelial growth factor (VEGF) is an important regulator of vasculogenesis, angiogenesis, and vascular permeability. In contrast to its transient expression during the formation of new blood vessels, VEGF and its receptors are continuously expressed in cultured adult and fetal endothelial cells. Furthermore, VEGF mRNA transcripts accumulate in endothelial cells exposed to hypoxia. The proteinuria and hypertension that characterize pre-eclampsia are reminiscent of the side effects of antiVEGF therapies [11]. Experimental neutralization of free VEGF by the administration of sFlt1 in rats resulted in endothelial dysfunction in vivo [4]. This experimental evidence, as well as the
626
J.K. Park et al. / Placenta 31 (2010) 621e629
Fig. 5. PI3K-Akt pathways in in vitro models of human placental hypoxia. Human choriocarcinoma JEG-3 cells were treated with 600 mM CoCl2. (A) p-Akt, PI3Kp85a, and HIF-1a were measured as single w60, w85, and w120 kDa bands, respectively. (B) The expression of p-Akt and PI3Kp85a was maximal after 6 and 12 h of CoCl2 treatment, respectively. The expression of HIF-1a increased in a time-dependent manner with 600 mM CoCl2 treatment. The lane loading, controls, and densitometric analyses were as described in Fig. 1. Data are presented as means SE; * p < 0.05.
pathological findings in humans with pre-eclampsia, further supports a role of VEGF in maintaining normal blood vessels and provides clues regarding the tissues in which VEGF action is most important in the adult. Little is known of the role of placental growth factor (PlGF), a VEGF homolog. The loss of PlGF impairs angiogenesis, plasma extravasation, and collateral growth during ischemia, inflammation, wound healing, and cancer [12]. A report that the inhibition of both VEGF and PlGF is required to cause pre-eclampsia-like changes in pregnant rats implies that the inhibition of PlGF plays a critical role in sFlt1-induced endothelial dysfunction [4]. Our Western blot analysis indicated that the expression levels of VEGF, sFlt1, and PlGF were higher in the preeclamptic placentas than in the normal placentas. In in vitro studies, villous trophoblasts under hypoxia expressed higher levels of VEGF and sFlt1 mRNAs and secreted more VEGF and sFlt1 proteins, whereas PlGF mRNA and protein were decreased in hypoxia [13,14]. However, reports in preeclamptic placentas provide conflicting conclusions regarding
the expression of VEGF, sFlt1, and PlGF. In severe pre-eclampsia, immunostaining for cytotrophoblast VEGF and sFlt1 was decreased and immunostaining for cytotrophoblast PlGF was unaffected [15]; the PlGF mRNA level was increased in preeclamptic placenta compared with normal placentas [13]. Discrepancies in the findings between studies using preeclamptic placenta and in vitro models may result because a pre-eclampsia model with hypoxia alone is not sufficient for investigating the generation of placental oxidative stress in pre-eclampsia; an in vitro hypoxia/reoxygenation model may be a more appropriate model system [16]. Moreover, the unique characteristics of each placental compartment may explain the different findings in preeclamptic placentas, because anchoring villi, invasive cytotrophoblasts, and blood vessels are not distributed evenly throughout the placenta [15]. The sF1t1:PlGF ratio has been suggested as an index of antiangiogenic activity that reflects the changes in both markers [17] and has been reported to be a better indicator of pre-eclampsia than separate measurements of each marker [18]. In the present
J.K. Park et al. / Placenta 31 (2010) 621e629
627
Fig. 6. Inhibition of the PI3K-Akt pathway with LY294002, a known pharmacologic inhibitor of PI3K. (A) Human choriocarcinoma JEG-3 cells were treated with different concentrations of CoCl2 for 12 h; p-Akt was measured as a w60 kDa band. The expression of p-Akt increased significantly in JEG-3 cells treated with CoCl2 for 12 h. (B) Human choriocarcinoma JEG-3 cells were pretreated with different concentrations of LY294002 for 1 h; p-Akt was measured as a w60 kDa band. In JEG-3 cells pretreated with 20 or 50 mM LY294002 for 1 h before treatment with 600 mM CoCl2, the expression of p-Akt was decreased significantly. (C and D) Human choriocarcinoma JEG-3 cells were pretreated with 50 mM LY294002 for 1 h and then treated with 600 mM CoCl2. The CoCl2-induced Akt phosphorylation was inhibited for 24 h after LY294002 pretreatment. The expression of HIF-1 was delayed in cells treated with 600 mM CoCl2 after pretreatment with 50 mM LY294002 (Fig. 6D), compared with HIF-1 expression in cells treated with only 600 mM CoCl2 (Fig. 5B). The lane loading, controls, and densitometric analyses were as described in Fig. 1. Data are presented as means SE; *p < 0.05.
study, the index of anti-angiogenic activity was measured from placental tissue, and not from maternal serum as in other studies. As a result, the sF1t1:VEGF ratio differed more significantly than did the sF1t1:PlGF ratio. Although PlGF itself does not significantly stimulate angiogenesis in vivo or in vitro, this growth factor modulates the angiogenic activity of VEGF by forming PlGF/VEGF heterodimers [19]. Thus, the amount of VEGF also influences the complex placental etiology. The PI3K-Akt pathway regulates normal cellular functions such as survival, proliferation, and glucose metabolism. Depending on the cell type, hypoxia can either activate or inactivate the PI3K-Akt signaling pathway. For example, the pathway is activated by hypoxia in PC12, HT1080, and HeLa cells [20e22], but is inactivated
in HepG2, BeWo, BeWo31, and JAR placental cells and in primary cytotrophoblast cells [23e25] under hypoxic conditions. Although the mechanisms of activation or inhibition of the PI3KAkt signaling pathway by hypoxia are unknown, we demonstrated in the present study that this pathway is activated in preeclamptic placentas and in JEG cells in the presence of CoCl2, a hypoxiamimicking agent (within the initial 24 h). We also cultured JEG-3 cells in 1.5% O2 for 48 h to produce hypoxic conditions (data not shown) and observed that PI3K expression reached a peak at 6e12 h, but decreased by 48 h; p-Akt reached a peak at 3e6 h, but also decreased by 48 h; and HIF-1a increased continuously with time. These results were similar to our results using CoCl2 as a hypoxia-mimicking agent, whereas Chiang MH et al. [25] showed
628
J.K. Park et al. / Placenta 31 (2010) 621e629
Fig. 7. Angiogenic and anti-angiogenic factors in in vitro models of human placental hypoxia with inhibition of the PI3K-Akt pathway by LY294002. (A) Human choriocarcinoma JEG-3 cells were treated with 600 mM CoCl2 for 12 h with or without 50 mM LY294002 pretreatment for 1 h; VEGF was measured as w20 kDa bands. The level of VEGF expression increased with 50 mM LY294002 pretreatment compared with the control and increased when treated with 600 mM CoCl2 after 50 mM LY294002 pretreatment, as compared with 600 mM CoCl2 treatment alone. (B) Human choriocarcinoma JEG-3 cells were treated with 600 mM CoCl2 for 24 h with or without 50 mM LY294002 pretreatment for 1 h; VEGF was measured as w20 kDa bands. The level of VEGF expression increased with 50 mM LY294002 pretreatment compared with the control, and remained unchanged when treated with 600 mM CoCl2 after 50 mM LY294002 pretreatment as compared with 600 mM CoCl2 treatment alone. (C) Human choriocarcinoma JEG-3 cells were treated with 600 mM CoCl2 for 12 h with or without 50 mM LY294002 pretreatment for 1 h; PlGF was measured as an w18 kDa band. The PlGF expression increased with 50 mM LY294002 pretreatment compared with the control and remained unchanged when treated with 600 mM CoCl2 after 50 mM LY294002 pretreatment, as compared with 600 mM CoCl2 treatment alone. (D) Human choriocarcinoma JEG-3 cells were treated with 600 mM CoCl2 for 24 h with or without 50 mM LY294002 pretreatment for 1 h. PlGF was measured as a w18 kDa band. The PlGF expression increased with 50 mM LY294002 pretreatment compared with the control, and increased when treated with 600 mM CoCl2 after 50 mM LY294002 pretreatment as compared with 600 mM CoCl2 treatment alone. (E) Human choriocarcinoma JEG-3 cells were treated with 600 mM CoCl2 for 6 h with or without 50 mM LY294002 pretreatment for 1 h. There was a significant decrease in the expression of sFlt1 with 50 mM LY294002 pretreatment compared with the control, and a significant decrease when treated with 600 mM CoCl2 after 50 mM LY294002 pretreatment as compared with 600 mM CoCl2 treatment alone. (F) Human choriocarcinoma JEG-3 cells were treated with 600 mM CoCl2 for 24 h with or without 50 mM LY294002 pretreatment for 1 h. There was a significant decrease in the expression of sFlt1 with 600 mM CoCl2 treatment after 50 mM LY294002 pretreatment as compared with 600 mM CoCl2 treatment alone. The lane loading, controls, and densitometric analyses were as described in Fig. 1. Data are presented as means SE; *p < 0.05.
that hypoxia (1% O2 for 48 h) down-modulated both PI3K and p-Akt in cytotrophoblasts from normal placentas and in BeWo and JAR cells. There are two factors that may account for these apparently conflicting results. First, Chiang et al. [25] focused on the late responses, by which time apoptosis had advanced considerably; in contrast, the present study focused on early changes, which included sFlt1 activation. Second, Hung et al. [16] reported, an in vitro pre-eclampsia model with hypoxia alone, which was used by Chiang et al. [25], may not be sufficient for investigating placental oxidative stress in pre-eclampsia, and a model with hypoxia/reoxygenation may be more suitable. Given that the increased levels of PI3K, p-Akt, and HIF-1a that were among the early changes in JEG-3 cells cultured under CoCl2 are also seen in preeclamptic placenta, we consider our model to be more appropriate than the placental oxidative stress model of Chiang et al. [25] for investigating the generation of placental oxidative stress in pre-eclampsia. JEG-3 cells were chosen as an in vitro model system in the present study because according to Zhou et al. [15], among the choriocarcinoma cell lines JAR, JEG, and BeWo, the production of PlGF and VEGF in JEG cells was most similar to that in third-trimester cytotrophoblasts. Currently, the only treatment for pre-eclampsia is the delivery of the fetus and placenta. However, delivery of the fetus in the
late second or early third trimester, although necessary for the safety of the mother, can result in significant neonatal complications and possibly death owing to the immaturity of the premature neonate. The identification of sFlt1, which represents a critical connection between placental pathology and maternal endothelial dysfunction, may provide a target for effective treatment of pre-eclampsia. The ideal treatment would be to restore and maintain the balance between angiogenic and anti-angiogenic factors, i.e., to restore the biological activities of angiogenic factors such as VEGF and PlGF and overcome the effects of anti-angiogenic factors such as sFlt1. In a recent report, recombinant VEGF 121 treatment resulted in significant improvements in systolic blood pressure, glomerular endotheliosis, and proteinuria in a pregnant rat model of preeclampsia induced by sFlt1 overexpression, without apparent harm to the fetus [26]. Here, we presented the results of PI3K-Akt pathway inhibition in in vitro models of human placental hypoxia. We demonstrated that human placental hypoxia models have high levels of sFlt1 at baseline and that inhibition of the PI3K-Akt pathway by a PI3Kspecific inhibitor leads to a decreased sFlt1 level and unchanged or increased VEGF and PlGF levels.
J.K. Park et al. / Placenta 31 (2010) 621e629
Inhibition of the PI3K-Akt pathway may be a useful therapeutic approach, if it were to decrease sFlt1 secretion without inhibiting VEGF or PlGF secretion. This pathway provides a potential target for a new treatment strategy in patients with pre-eclampsia. Acknowledgements This research was supported by a grant from the National Research Foundation of Korea (NRF) funded by the Korean government (MEST) (2009-0067486) and a 2009 special clinical grant from Gyeongsang National University Hospital. Appendix. Supplementary data Supplementary data associated with this article can be found in the online version, at doi:10.1016/j.placenta.2010.04.009. References [1] Gaber LW, Spargo BH, Lindheimer MD. Renal pathology in pre-eclampsia. Baillière’s Clin Obstet Gynaecol 1994;8:443e68. [2] Roberts JM, Taylor RN, Musci TJ, Rodgers GM, Hubel CA, McLaughlin MK. Preeclampsia: an endothelial cell disorder. Am J Obstet Gynecol 1989;161: 1200e4. [3] Redman CWG, Sargent IL. Placental stress and pre-eclampsia: a revised view. Placenta 2009;30:S38e42. [4] Maynard SE, Min JY, Merchan J, Lim KH, Li J, Mondal S, et al. Excess placental soluble fms-like tyrosine kinase 1 (sFlt1) may contribute to endothelial dysfunction, hypertension, and proteinuria in preeclampsia. J Clin Invest 2003;111:649e58. [5] Tan C, Cruet-Hennequart S, Troussard A, Fazli L, Costello P, Sutton K, et al. Regulation of tumor angiogenesis by integrin-linked kinase (ILK). Cancer Cell 2004;5:79e80. [6] Okuyama H, Krishnamachary B, Zhou YF, Nagasawa H, Bosch-Marce M, Semenza GL. Expression of vascular endothelial growth factor receptor 1 in bone marrow-derived mesenchymal cells is dependent on hypoxia-inducible factor 1. J Biol Chem 2006;281:15554e63. [7] Martorell L, Gentile M, Rius J, Rodríguez C, Crespo J, Badimon L, et al. The hypoxia-inducible factor 1/NOR-1 axis regulates the survival response of endothelial cells to hypoxia. Mol Cell Biol 2009;29:5828e42. [8] Jeyabalan A, McGonigal S, Gilmour C, Hubel CA, Rajakumar A. Circulating and placental endoglin concentrations in pregnancies complicated by intrauterine growth restriction and preeclampsia. Placenta 2008;29:555e63. [9] American College of Obstetricians and Gynecologists. ACOG practice bulletin. Diagnosis and management of preeclampsia and eclampsia. Int J Gynaecol Obstet 2002;77:67e75. [10] Goldberg MA, Dunning SP, Bunn HF. Regulation of erythropoietin gene: evidence that the oxygen sensor is a heme protein. Science 1988;242:1412e4.
629
[11] Hurwitz H, Fehrenbacher L, Novotny W, Cartwright T, Hainsworth J, Heim W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004;350:2335e42. [12] Carmeliet P, Moons L, Luttun A, Vincenti V, Compernolle V, De Mol M, et al. Synergism between vascular endothelial growth factor and placental growth factor contributes to angiogenesis and plasma extravasation in pathological conditions. Nat Med 2001;7:575e83. [13] Ahmed A, Dunk C, Ahmad S, Khaliq A. Regulation of placental vascular endothelial growth factor (VEGF) and placenta growth factor (PlGF) and soluble Flt-1 by oxygenea review. Placenta 2000;21:S16e24. [14] Munaut C, Lorquet S, Pequeux C, Blacher S, Berndt S, Frankenne F, et al. Hypoxia is responsible for soluble vascular endothelial growth factor receptor-1 (VEGFR-1) but not for soluble endoglin induction in villous trophoblast. Hum Reprod 2008;23:1407e15. [15] Zhou Y, McMaster M, Woo K, Janatpour M, Perry J, Karpanen T, et al. Vascular endothelial growth factor ligands and receptors that regulate human cytotrophoblast survival are dysregulated in severe preeclampsia and hemolysis, elevated liver enzymes, and low platelets syndrome. Am J Pathol 2002;160:1405e23. [16] Hung TH, Skepper JN, Burton GJ. In vitro ischemia-reperfusion injury in term human placenta as a model for oxidative stress in pathological pregnancies. Am J Pathol 2001;159:1031e43. [17] Levine RJ, Thadhani R, Qian C, Lam C, Lim KH, Yu KF, et al. Urinary placental growth factor and risk of preeclampsia. JAMA 2005;293:77e85. [18] Levine RJ, Lam C, Qian C, Yu KF, Maynard SE, Sachs BP, et al. Soluble endoglin and other circulating antiangiogenic factors in preeclampsia. N Engl J Med 2006;355:992e1005. [19] Cao Y, Chen H, Zhou L, Chiang MK, Anand-Apte B, Weatherbee JA, et al. Heterodimers of placenta growth factor/vascular endothelial growth factor. Endothelial activity, tumor cell expression, and high affinity binding to Flk-1/ KDR. J Biol Chem 1996;271:3154e62. [20] Beitner-Johnson D, Rust RT, Hsieh TC, Millhorn DE. Hypoxia activates Akt and induces phosphorylation of GSK-3 in PC12 cells. Cell Signal 2001;13:23e7. [21] Chen EY, Mazure NM, Cooper JA, Giaccia AJ. Hypoxia activates a plateletderived growth factor receptor/phosphatidylinositol 3-kinase/Akt pathway that results in glycogen synthase kinase-3 inactivation. Cancer Res 2001;61:2429e33. [22] Alvarez-Tejado M, Alfranca A, Aragonés J, Vara A, Landázuri MO, del Peso L. Lack of evidence for the involvement of the phosphoinositide 3-kinase/Akt pathway in the activation of hypoxia-inducible factors by low oxygen tension. J Biol Chem 2002;277:13508e17. [23] Mottet D, Dumont V, Deccache Y, Demazy C, Ninane N, Raes M, et al. Regulation of hypoxia-inducible factor-1a protein level during hypoxic conditions by the phosphatidylinositol 3-kinase/Akt/glycogen synthase kinase 3b pathway in HepG2 Cells. J Biol Chem 2003;278:31277e85. [24] Flügel D, Görlach A, Michiels C, Kietzmann T. Glycogen synthase kinase 3 phosphorylates hypoxia-inducible factor 1 and mediates its destabilization in a VHL-independent manner. Mol Cell Biol 2007;27:3253e65. [25] Chiang MH, Liang FY, Chen CP, Chang CW, Cheong ML, Wang LJ, et al. Mechanism of hypoxia-induced GCM1 degradation: implications for the pathogenesis of preeclampsia. J Biol Chem 2009;284:17411e9. [26] Li Z, Zhang Y, Ying Ma J, Kapoun AM, Shao Q, Kerr I, et al. Recombinant vascular endothelial growth factor 121 attenuates hypertension and improves kidney damage in a rat model of preeclampsia. Hypertension 2007;50: 686e92.