lable at ScienceDirect

Placenta 35 (2014) 109e116

Contents lists avai

Placenta

journal homepage: www.elsevier .com/locate/placenta

Increased NFAT5 expression stimulates transcription of Hsp70 inpreeclamptic placentas

J.K. Park a,b, T.G. Kang a, M.Y. Kang b, J.E. Park a, I.A. Cho b, J.K. Shin a,b, W.J. Choi a,b,S.A. Lee a,b, W.S. Choi b,c,1, H.M. Kwon d, J.H. Lee a,b, W.Y. Paik a,b,*,1

aDepartment of Obstetrics and Gynecology, School of Medicine, Gyeongsang National University, JinJu, South Koreab Institute of Health Sciences, School of Medicine, Gyeongsang National University, JinJu, South KoreacDepartment of Anatomy, School of Medicine, Gyeongsang National University, JinJu, South Koread School of Nano-Bioscience and Chemical Engineering, Ulsan National Institute and Science and Technology, Ulsan, South Korea

a r t i c l e i n f o

Article history:Accepted 13 December 2013

Keywords:Preeclamptic placentaHIF-1aNFAT5Hsp70

* Corresponding author. Department of ObstetricsMedicine, Gyeongsang National University, 90 ChilamKorea. Tel.: þ82 55 750 8152; fax: þ82 55 759 1118.

E-mail address: wypaik@gnu.ac.kr (W.Y. Paik).1 Both authors contributed equally to this work (Co

0143-4004/$ e see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.placenta.2013.12.005

a b s t r a c t

Objective: We investigated the expression of heat shock protein 70 (Hsp70), nuclear factor of activated Tcells 5 (NFAT5), and hypoxia-induced factor-1a (HIF-1a) in the placentas of normal and preeclampticpregnancies and in human placental hypoxia models in vitro to examine the regulatory mechanisms ofplacental Hsp70 expression.Methods: The expression levels of HIF-1a, NFAT5, and Hsp70 were examined in placental samples from10 females with preeclampsia and 10 normotensive control patients and in human choriocarcinomatrophoblast cells treated with 1 mM CoCl2 by western blotting. Using models of placental hypoxia,pharmacological inhibition of HIF-1a with chetomin and shRNA knockdown and overexpression ofNFAT5 were performed to investigate the roles of HIF-1a and NFAT5 in induction of Hsp70 by placentalhypoxia.Results: The levels of HIF-1a, NFAT5, and Hsp70 expression were significantly higher in the preeclampticcompared to normal placentas. In the placental hypoxia models, the expression of HIF-1a, NFAT5, andHsp70 were significantly higher after 3, 6, and 12 h of 1 mM CoCl2 treatment, respectively. Pharmaco-logical inhibition of HIF-1a suppressed the induction of NFAT5 and Hsp70 at the protein level. shRNAknockdown of NFAT5 suppressed the induction of Hsp70 protein and overexpression of NFAT5 stimu-lated the induction of Hsp70 mRNA and protein in models of human placental hypoxia in vitro.Conclusion: HIF-1a positively regulates the induction of NFAT5 and Hsp70 by placental hypoxia andNFAT5 stimulates transcription of Hsp70 in response to placental hypoxia in models of human placentalhypoxia in vitro.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Preeclampsia, a hypertensive, pregnancy-specific disorder, haslong been analyzed in terms of its association with cellular stress.However, preeclampsia, a multi-system disorder that affectsmaternal vascular function and fetal growth, remains one of themost serious complications of pregnancy. The pathophysiology ofpreeclampsia is unclear, but accumulating evidence suggests that anexcessive maternal systemic inflammatory response to pregnancy

and Gynecology, School of-dong, Jinju 660-702, South

-correspondence).

All rights reserved.

with systemic oxidative stress and resultant endothelial damageplays a crucial role in the pathogenesis of the preeclampsia [1,2].

Heat shock proteins (Hsps) are known primarily as intracellularproteins with molecular chaperone and cytoprotective functions[3]. However, a growing body of evidence suggests that extracel-lular Hsp70 may serve as a danger signal to the innate immunesystem and that it may contribute to the establishment of auto-immune diseases [4e6]. In preeclampsia, serum Hsp70 levels areincreased, and reflect systemic inflammation and oxidative stress[7,8]. Nevertheless, the regulation of Hsp70 in healthy individuals,as well as in patients with preeclampsia, has not been fully iden-tified. The increase in accumulation of Hsp70 under conditions ofstress is achieved by transcriptional activation of tonicity-enhancerbinding protein (TonEBP or nuclear factor of activated T cells,NFAT5) in the urinary concentrating mechanism [9]. TonEBP/NFAT5is a member of the Rel/NFkB/NFAT family of signal transcription

J.K. Park et al. / Placenta 35 (2014) 109e116110

factors [10]. A recent study suggested that activation of NFAT5 inthe innermedulla of the kidney is induced by hypoxia in addition tohypertonicity [11]. The function of NFAT5 is well-known in thekidney, but other functions are under investigation; e.g., animmunomodulatory function in T cells [12,13].

Although an excessive maternal systemic inflammatoryresponse to pregnancy with systemic oxidative stress is thought tobe involved in the pathogenesis of preeclampsia, it is unclear howhypoxia regulates placental expression of Hsps. We hypothesizedthat increased Hsp70 expression in preeclamptic placentas reflectsan excessive placental inflammatory response to placental oxidativestress and that NFAT5 stimulates transcription of Hsp70 in responseto placental hypoxia. To address our hypothesis, we investigated theexpression of Hsp70, NFAT5, and hypoxia-induced factor-1a (HIF-1a) in the placentas of normal and preeclamptic pregnancies and inhuman placental hypoxia models in vitro and investigated the rolesof HIF-1a and NFAT5 in induction of Hsp70 by placental hypoxia.

2. Materials and methods

2.1. Human placental tissue samples

The experimental protocol usedwas peer-reviewed and approved by the HumanSubject Research Committee of Gyeongsang National University Hospital, Jinju,Korea. Written informed consent, with permission to collect placental tissue, wasobtained from 20 pregnant females who were scheduled to undergo cesarean sec-tionwithout labor at 35e40 weeks of pregnancy. Of these pregnancies, 50% (n ¼ 10)were normal and 50% (n ¼ 10) were preeclamptic.

Preeclampsia was diagnosed based on the presence of hypertension (bloodpressure �140 mmHg systolic or �90 mmHg diastolic after 20 weeks of gestation inpreviously normotensive females) and proteinuria (urinary excretion �0.3 g ofprotein in a 24-h specimen, or �30 mg/dL or a reading �1þ on dipstick test in arandom urine specimen). Severe preeclampsia was defined as HELLP syndrome,eclampsia, or preeclampsia 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 dipstick test). All preeclamptic fe-males in this study had “severe” preeclampsia on the basis of the ACOG criteria [14].The control subjects had no evidence of chronic hypertension, preeclampsia, orgestational hypertension. No control or preeclamptic subjects had been diagnosedwith 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 full-thickness samples of w3 � 3 cm. The samples were then re-sectioned into piecesof w1 � 1 cm, frozen in liquid nitrogen, and stored at �70 �C. One 1 � 1-cm piecefrom each of the five different portions of each placenta was used for proteinextraction.

2.2. Double immunofluorescence staining

For histological studies, placental tissues were immunostained as describedpreviously, with some modifications. Sectioned placental tissues were incubated for1 h in blocking solution (5% normal goat serum, 1% bovine serum albumin, 0.5%Triton X-100, 0.05% sodium azide in 0.05 M PBS, pH 7.4), followed by 4 �C overnightincubation with mixed primary antibodies: monoclonal mouse anti-Hsp70 andpolyclonal rabbit antieHIFe1a or monoclonal mouse anti-Hsp70 and polyclonalrabbit anti-NFAT5 (1:200 dilution; Santa Cruz Biotechnology, Santa Cruz, CA). Afterseveral washes in 0.1 M PBS, the sections were incubated for 1 h with donkey anti-IgG conjugated to Alexa Fluor 488 (green; 1:1000 dilution; Invitrogen, Carlsbad, CA)or Alexa Fluor 594 (red; 1:1000 dilution; Invitrogen) as the secondary antibody.Sections were rinsed in PBS and wet-mounted using wet anti-fade reagent withDAPI (Invitrogen). Digital images using a BX51-DSU microscope (Olympus, Tokyo,Japan) were captured and documented. Omission of the primary antibodies servedas a negative control to estimate non-specific interactions (Supplementary Fig. 1).

2.3. Cell lines and cell culture

The choriocarcinoma cell line JEG-3 cells were cultured according to the protocolrecommended by the American Type Culture Collection (ATCC, Manassas, VA); i.e., inDulbecco’s modified Eagle’s medium supplemented with 10% heat-inactivated fetalbovine serum (Invitrogen), 2 mM L-glutamine, 100 U/mL penicillin, and 100 mg/mLstreptomycin (Invitrogen) at 37 �C in a humidified 5% CO2 incubator. When the cellsreached 80% confluence, as determined by eye, the cells were detached bytrypsinization.

2.4. Hypoxic treatment

In this study, CoCl2 was used as a hypoxia-mimicking agent as reported previ-ously [15]. To determine the appropriate amount of CoCl2 (Wako, Osaka, Japan),

2 � 105 JEG-3 cells/mL were seeded in dishes 10 cm in diameter 24 h before treat-ment with 0, 100, 200, 400, 600, 800, or 1000 mM CoCl2. For the experiments, theJEG-3 cells were cultured in the absence or presence of 600, 800, or 1000 mM CoCl2for 3, 6, 12, or 24 h.

Cell viability after exposure to hypoxic conditions was determined by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay. Briefly, cellswere seeded at 2 � 105/mL on a 96-well plate, and treated with or without CoCl2 for24 h. Then, 30 mL of 1 mg/mL MTT (Sigma Chemical, St. Louis, MO) were added toeach well, and the plate was incubated at 37 �C until a blue color appeared. Themedium was aspirated and replaced with dimethyl sulfoxide. The absorbance at560 nm was read using a microtiter plate reader.

2.5. Inhibition of HIF-1a by chetomin

Chetomin (Santa Cruz Biotechnology) was dissolved in dimethyl sulfoxide(DMSO) to prepare a 1 mM stock solution. Cells were seeded in 100-mm dishes 24 hbefore treatment with 75 nM chetomin. Cells were pretreated with 75 nM chetominor DMSO for 4 h at 37 �C under normoxic conditions. Then, the cells were exposed tohypoxic conditions (1 mM CoCl2) or maintained in a normoxic environment andincubated for 24 h.

2.6. shRNA knockdown and overexpression of NFAT5

Recombinant lentiviruses expressing shRNA against NFAT5 or eGFP-controlwereproduced as follows. Lentiviral pLKO.1-puro vectors encoding shRNA specific forNFAT5 or eGFP-control were purchased from Sigma TRC shRNA library. The NFAT5shRNA sequence (TRCN0000020019) used was 50-CCGGGCCCAGATTCAGTCA-GAGTTACTCGAGTAACTCTGACTGAATCTGGGCTTTTT-30 . The eGFP-control shRNAsequence (SHC005, MISSION pLKO.1-puro eGFP shRNA Control) used was 50-CCGGTACAACAGCCACAACGTCTATCTCGAGATAGACGTTGTGGCTGTTGTATTTTT-30 . Forlentiviral production, 293T cells were transfected with pLKO.1 vector together withpackaging plasmids encoding Gag/Pol, Rev, and VSV-G using Lipofectamine 2000(Invitrogen) according to the manufacturer’s instructions. Culture media containinglentiviral particles were collected 48 and 72 h after transfection and filtered. Viralsupernatants were pooled and stored at �80 �C. JEG-3 cells were infected with vi-ruses in media. Forty-eight hours after infection, cells were lysed and lysates wereprocessed for immunoblotting.

Cells were transfected with expression plasmids for pCMV-Tag (flag control),pCMV-Tag2C-Yc1 (Rel homology domain of NFAT5-c, NFAT5 dominant negativemutant) or pcDNA3.1 hNFAT5-c (accession no. NM_006599), using Lipofectamine2000 (Invitrogen). Cells were grown to w90% confluency in 60-mm2 plates, trans-fected with 1 mg of each plasmid and assayed after 24 h.

2.7. Western blotting

Western blotting for HIF-1a, NFAT5, and Hsp70 was performed using totalplacental lysates or JEG-3 cell lysates. For cultured JEG-3 cells, the medium wasaspirated, and the cells were scraped off into 1 mL of 1 � PBS. After centrifugation at130� g for 3 min, the supernatant was discarded, and the collected cells were lysed.Placental tissues and JEG-3 cells were suspended in lysis buffer containing 50 mMTris (pH 7.5), 150 mM NaCl, 5 mM EDTA, 1% Nonidet P-40, protease inhibitors (1 mMphenylmethylsulfonyl fluoride, 1 mg/mL aprotinin, and 1 mg/mL leupeptin), andphosphatase 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 bycentrifugation (10,000� g, 20 min, 4 �C), and the protein concentration in the su-pernatants was determined using a bicinchoninic acid protein assay kit (Pierce,Rockford, IL) with bovine serum albumin as the standard. The lysates were diluted insample buffer (0.15 M TriseHCl, pH 6.8, 2% SDS, 10% glycerol, and 5% 2-mercaptoethanol) and boiled for 5 min. Equal amounts of protein (30 mg) wereloaded into the lanes of 6e8% SDS-polyacrylamide gels, and the proteins wereseparated by electrophoresis. The separated proteins were transferred onto nitro-cellulose membranes. The membranes were washed in Tris-buffered saline con-taining 0.1% Tween 20 (TBST) and incubated with one of the following primaryantibodies: anti-HIF-1a, anti-NFAT5, or anti-Hsp70 (0.2 mg/mL; Santa CruzBiotechnology). The membranes were washed and incubated with secondary anti-body (1:10,000; Pierce). Immunoreactive proteins were detected by enhancedchemiluminescence using a western blotting system (Pierce).

2.8. Real-time quantitative reverse transcription PCR (qRT-PCR)

Total RNAwas extracted using the Trizol reagent (Invitrogen) and converted intocDNA with the Reversed Transcription System (Promega) according to the manu-facturer’s protocol. Quantitative PCR was performed on a LightCycler 480 Real-timePCR System with Power SYBR Green PCR Master Mix (Roche Applied Science).Relative values of mRNAwere normalized to the level of b-actin mRNA. The primersused in these experiments are as follows. Hsp70 (298 bp) forward, 50-GTGCAGTGGCCTACAGGATT-3’and Hsp70 reverse, 50-AGCGAGGAAGATCCTGCTTAT-30; b-actin (248bp) forward, 50-CTGGCACCACACCTTCTACAATG-3’and b-actin reverse 50-CCTCGTA-GATGGGCACAGTGTG-30 .

J.K. Park et al. / Placenta 35 (2014) 109e116 111

2.9. Statistical analysis

The clinical data are expressed as medians (range). Comparisons were per-formed using the ManneWhitney U test. The densities of immunoreactive bandswere analyzed using SigmaGel 1.0 (Jandel Scientific, Erkrath, Germany) and Sig-maPlot 7.0 (SPSS, Chicago, IL). The results of densitometric analyses are presented asarbitrary units (A.U.) normalized relative to the b-actin level. Data are presented asmeans � standard error (SE) of three independent experiments. Groups werecompared using Student’s unpaired t test. In all analyses, p < 0.05 was taken toindicate statistical significance.

3. Results

3.1. Subjects

Placental tissues were obtained from 20 pregnant females (10normal pregnancies and 10 preeclamptic pregnancies) who werescheduled to undergo cesarean section without labor at 35e40weeks of pregnancy, as reported previously [15]. There was a sig-nificant (p < 0.01) difference in parity between the females withpreeclampsia (0; 0e2) and those in the normal pregnancy controlgroup (2; 1e3), whowere scheduled for a repeat caesarean delivery(parity � 1). The preeclamptic and control groups were similar inmaternal age, maternal body mass index, and gestational age atdelivery, but the femaleswith preeclampsia had significantly highermean arterial pressure, significant proteinuria, and significantlylower neonatal birth weight (p< 0.01; Supplementary Table 1) [15].

3.2. Expression of HIF-1a, NFAT5, and Hsp70 in the preeclampticplacenta

Immunoreactivity for HIF-1a, NFAT5, and Hsp70 wereincreased in preeclamptic placentas and were prominent

Fig. 1. Double fluorescence immunostaining for Hsp70 and HIF-1a (A) and for Hsp70 and NFA(A) The levels of Hsp70 and HIF-1a immunoreactivity were increased in the preeclamptic plaimmunofluorescence labeling of Hsp70 and HIF-1a showed colocalization of the antigens. Dshowed colocalization of both antigens in mainly the cytosol. (B) The levels of Hsp70 andstaining, Hsp70 and NFAT5 images were merged (overlay). Double immunofluorescence lastaining was observed in the nucleus. The expression of Hsp70 and NFAT5 showed colocali

in syncytiotrophoblasts and villous endothelial cells(Supplementary Fig. 2). Double immunofluorescence labeling ofHsp70 and HIF-1a and of Hsp70 and NFAT5 showed colocaliza-tion of pairs of antigens mainly in the cytosol (Fig. 1). The levelsof HIF-1a, NFAT5, and Hsp70 expression were significantly higherin the preeclamptic compared to normal placentas (p < 0.05;Fig. 2).

3.3. Growth-inhibitory effects of CoCl2 on cell viability

The growth of JEG-3 cells was inhibited in a dose-dependentmanner by CoCl2 (Fig. 3A). The concentration of CoCl2 at whichw50% of the cells were viable (IC50) was 600 mM (Fig. 3A); however,the levels of Hsp70 expression in an in vitro model of humanplacental hypoxia after 3, 6, 12, and 24 h of 600 mM CoCl2 treatmentwere not as high as those in vivo (data not shown). Therefore, 1 mMCoCl2 was used in this work. The surviving fraction of JEG-3 cellswas roughly 35% after 24 h of treatment with 1mMCoCl2 (p< 0.05;Fig. 3A). The expression of cleaved poly(ADP-ribose) polymerase(PARP), a biochemical marker of cellular apoptosis, increased in atime-dependent manner after 6 h of treatment with 1 mM CoCl2(p < 0.05; Fig. 3B and C).

3.4. Expression of HIF-1a, NFAT5, and Hsp70 in an in vitro model ofhuman placental hypoxia

The levels of HIF-1a expression in an in vitro model of hu-man placental hypoxia were increased in a time-dependentmanner after treatment with 1 mM CoCl2 (p < 0.05; Fig. 4A).The levels of HIF-1a, NFAT5, and Hsp70 expression were

T5 (B) in placentas from normal pregnant (control, CTL) and preeclamptic females (PE).centa. After immunostaining, Hsp70 and HIF-1a images were merged (overlay). DoubleAPI immunostaining was observed in the nucleus. The expression of Hsp70 and HIF-1aNFAT5 immunoreactivity were increased in the preeclamptic placenta. After immuno-beling for Hsp70 and NFAT5 showed colocalization of both antigens. DAPI immuno-zation of both antigens in mainly the cytosol.

Fig. 2. Expression of HIF-1a, NFAT5, and Hsp70 in placentas from normal pregnant (control, CTL) and preeclamptic females (PE). (A) HIF-1a was measured as a single w120-kDaband. The expression of HIF-1a was significantly higher in preeclamptic (n ¼ 10) than in normal placentas (n ¼ 10). (B) NFAT5 was measured as a single w170-kDa band. Theexpression of NFAT5 was significantly higher in preeclamptic (n ¼ 10) than in normal (n ¼ 10) placentas. (C) Hsp70 was measured as a single w70-kDa band. The expression ofHsp70 was significantly higher in preeclamptic (n ¼ 10) than in normal (n ¼ 10) placentas. Each lane represents an individual placenta, and aliquots of 30 mg of total protein wereloaded. b-Actin was used as the positive control. The densities of the immunoreactive bands were analyzed using SigmaPlot 7.0. The results of densitometric analysis are presentedas arbitrary units (A.U.) normalized relative to the b-actin levels. Data are presented as means � SE; *p < 0.05.

J.K. Park et al. / Placenta 35 (2014) 109e116112

significantly higher after 3, 6, and 12 h of 1 mM CoCl2 treat-ment, respectively (p < 0.05; Fig. 4). Upon placental hypoxia,Hsp70 was induced at the total Hsp70 level. However, a newisoform was observed after 12 h of 1 mM CoCl2 treatment(Fig. 4C).

Fig. 3. Growth-inhibitory effects of CoCl2. (A) Human choriocarcinoma JEG-3 cells were treathe 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay and are shoJEG-3 cells was w35% after 24 h of 1000 mM CoCl2 treatment. (B) Biochemical marker of celCleaved poly(ADP-ribose) polymerase (PARP) was measured as a single w89-kDa band. (C1000 mM CoCl2 treatment. The lane loading, controls, and densitometric analyses were as d

3.5. Effects of HIF-1a inhibition on the expression of NFAT5 andHsp70 in an in vitro model of human placental hypoxia

In JEG-3 cells pretreated with 75 nM chetomin for 4 h beforetreatment with 1 mM CoCl2 for 24 h, the levels of HIF-1a, NFAT5,

ted with various concentrations of CoCl2 for 24 h. Viable cells were enumerated usingwn as percentages of the value of the respective control (CTL). The surviving fraction oflular apoptosis. Human choriocarcinoma JEG-3 cells were treated with 1000 mM CoCl2.) The expression of cleaved PARP increased in a time-dependent manner after 6 h ofescribed in Fig. 2. Data are presented as means � SE; *p < 0.05.

J.K. Park et al. / Placenta 35 (2014) 109e116 113

and Hsp70 expression were significantly decreased compared withthose in cells treatedwith 1mMCoCl2 for 24 h (p< 0.05; Fig. 5). Thelower-molecular-weight isoform of Hsp70 was highly inducibleupon placental hypoxia and was markedly inhibited upon inhibi-tion of HIF-1a (Fig. 5C).

3.6. Effects of shRNA knockdown and overexpression of NFAT5 onthe expression of Hsp70 in an in vitro model of human placentalhypoxia

The protein levels of NFAT5 and Hsp70 were significantly lowerin JEG-3 cells treated with 1 mM CoCl2 for 24 h after NFAT5 shRNApretreatment for 48 h than in cells treated with 1 mM CoCl2 aftercontrol shRNA pretreatment (p < 0.05; Fig. 6A and B). The lower-molecular-weight isoform of Hsp70 was highly inducible uponplacental hypoxia and was markedly inhibited upon inhibition ofNFAT5 (Fig. 6B). The protein levels of NFAT5 and Hsp70 weresignificantly higher in JEG-3 cells transfected with wild-type NFAT5for 24 h than in cells transfected with a NFAT5 dominant-negativemutant (p < 0.05; Fig. 6C and D). In addition, the mRNA level ofHsp70 was significantly higher in JEG-3 cells transfected for 24 h

Fig. 4. Expression of HIF-1a, NFAT5, and Hsp70 in controls (CTL) and models of human placen(A) HIF-1a was measured as a single w120-kDa band. The expression of HIF-1a increased inexpression of NFAT5 increased after 6 h of 1 mM CoCl2 treatment. (C) Hsp70 was measured aafter 12 h of 1 mM CoCl2 treatment. The total expression of Hsp70 increased after 12 h of 1 mtotal Hsp70 level. However, a new isoformwas observed after 12 h of 1 mM CoCl2 treatment.are presented as means � SE; *p < 0.05.

with wild-type NFAT5 than in cells transfected with a NFAT5dominant-negative mutant (p < 0.05; Fig. 6E).

4. Discussion

Hsp70 has immunoregulatory properties and exerts both pro-and anti-inflammatory effects in a cross-species manner. Elevatedsystemic levels of Hsp70 have been demonstrated in the serum offemales with preeclampsia [7,8], although Hsp70 expression in cellsdoes not necessarily lead to elevated Hsp70 levels in the systemiccirculation. Consistentwith the present study, higher Hsp70 proteinlevels were detected in placental tissues of patients with pre-eclampsia and/or fetal growth restriction than in normal pregnan-cies [16,17]. In the present study, a higher Hsp70 protein level wasfound in syncytiotrophoblasts and villous endothelial cells of pre-eclamptic placentas than in normal pregnancies. This suggests thatthe increased systemic levels of Hsp70 in the serum of females withpreeclampsia originate from syncytiotrophoblasts and villousendothelial cells of preeclamptic placentas. The double immuno-fluorescence labeling data suggest a cytosolic localization, and if theplacenta is the primary source of increasedHsp70, itmay depend onthe lysis of placental cells to liberate cytosolic proteins into the

tal hypoxia in vitro. Human choriocarcinoma JEG-3 cells were treated with 1 mM CoCl2.a time-dependent manner. (B) NFAT5 was measured as a single w170-kDa band. The

s a singlew70-kDa band until 6 h of 1 mM CoCl2 treatment and doublew70-kDa bandsM CoCl2 treatment. Upon placental hypoxia, the induction of Hsp70 was induced at theThe lane loading, controls, and densitometric analyses were as described in Fig. 2. Data

Fig. 5. Effects of HIF-1a inhibition on the expression of NFAT5 and Hsp70 in an in vitro model of human placental hypoxia. Human choriocarcinoma JEG-3 cells were treated with1 mM CoCl2 for 24 h with or without 75 nM chetomin pretreatment for 4 h (A) HIF-1awas measured as a singlew120-kDa band. The expression of HIF-1a increased significantly inJEG-3 cells treated with CoCl2 for 24 h. There was a significant decrease in the expression of HIF-1a when treated with 1 mM CoCl2 after 75 nM chetomin pretreatment for 4 h ascompared with 1 mM CoCl2 treatment alone. (B) NFAT5 was measured as a single w170-kDa band. The expression of NFAT5 increased significantly in JEG-3 cells treated with CoCl2for 24 h. There was a significant decrease in the expression of NFAT5 when treated with 1 mM CoCl2 after 75 nM chetomin pretreatment for 4 h, as compared with 1 mM CoCl2treatment alone. (C) Hsp70 was measured as a single w70-kDa band. Upon placental hypoxia, Hsp70 was induced at the total Hsp70 level. However, a new isoform was observedwith 24 h of 1 mM CoCl2 treatment. The expression of Hsp70 was significantly increased in JEG-3 cells treated with CoCl2 for 24 h. There was a significant decrease in the expressionof Hsp70 when treated with 1 mM CoCl2 after 75 nM chetomin pretreatment for 4 h as compared with 1 mM CoCl2 treatment alone. The lower-molecular-weight isoform of Hsp70was highly inducible upon placental hypoxia and was markedly inhibited upon inhibition of HIF-1a. The lane loading, controls, and densitometric analyses were as described inFig. 2. Data are presented as means � SE; *p < 0.05.

J.K. Park et al. / Placenta 35 (2014) 109e116114

circulation. Moreover, endothelial cells could be a source of Hsp70,as could circulating monocytes. In addition, while the expression ofHsp70 plays an initial protective role against oxidative stress, asmentioned in the literature, its overexpression may result in inter-villous endothelial dysfunction in preeclamptic placental villoustissues. Thus, increasedHsp70 expression in preeclamptic placentasmay play a role in the pathogenesis of preeclampsia.

The master regulator of the hypoxia response is HIF-1a. In thepresent study, hypoxia also induced NFAT5 and Hsp70 proteinlevels both in vivo and in models of human placental hypoxiain vitro. Moreover, HIF-1a and Hsp70 showed colocalization mainlyin the cytosol of syncytiotrophoblasts in vivo, and the induction ofHIF-1a was faster than those of NFAT5 and Hsp70 (3 h vs. 6 h and12 h of 1 mM CoCl2 treatment, respectively) in models of humanplacental hypoxia in vitro. Pharmacological inhibition of HIF-1awith 75 nM chetomin suppressed the induction of NFAT5 andHsp70 at the protein level in models of human placental hypoxia

in vitro. These results suggest that HIF-1a positively regulates theinduction of NFAT5 and Hsp70 by placental hypoxia. In contrast tothe results presented here, a recent study suggested that HIF-1ainterferes with TonEBP/NFAT5 function and suppresses the induc-tive effect of TonEBP/NFAT5 on the Hsp70 promoter in hypoxic andhyperosmolar nucleus pulposus cells [18]. Another study suggestedthat HIF-1a and NFAT5 are independently upregulated by hypoxiain human embryonic kidney 293 cells (HEK293 cells) [11].

NFAT5 is a transcription factor that regulates aldose reductase(AR), Hsp70, and other genes, allowing renal cells to adapt to highosmolality [9,19]. In the present study, placental hypoxia did notaffect the AR induction, despite the induction of NFAT5, in thein vitro model of human placental hypoxia (Supplementary Fig. 3).However, placental hypoxia affected the induction of NFAT5 andHsp70. Moreover, NFAT5 and Hsp70 colocalized in mainly thecytosol of syncytiotrophoblasts in vivo, and the induction of NFAT5was more rapid than that of Hsp70 in models of human placental

Fig. 6. Effects of shRNA knockdown and overexpression of NFAT5 on the expression of Hsp70 in an in vitro model of human placental hypoxia. (A and B) Human choriocarcinomaJEG-3 cells were treated with 1 mM CoCl2 for 24 h after NFAT5 shRNA or control shRNA pretreatment for 48 h (A) NFAT5 was measured as a singlew170-kDa band. The expression ofNFAT5 increased significantly in JEG-3 cells treated with CoCl2 for 24 h. There was a significant decrease in the expression of NFAT5 when treated with 1 mM CoCl2 after NFAT5shRNA pretreatment for 48 h as compared with 1 mM CoCl2 after control shRNA pretreatment for 48 h (B) Hsp70 was measured as a single w70-kDa band. Upon placental hypoxia,total Hsp70 was induced. However, a new isoform was observed with 24 h of 1 mM CoCl2 treatment. The expression of Hsp70 increased significantly in JEG-3 cells treated withCoCl2 for 24 h. There was a significant decrease in the expression of Hsp70 when treated with 1 mM CoCl2 after NFAT5 shRNA pretreatment for 48 h as compared with 1 mM CoCl2after control shRNA pretreatment for 48 h. The lower-molecular-weight isoform of Hsp70 was highly inducible upon placental hypoxia and was markedly inhibited upon inhibitionof NFAT5. (CeE) Human choriocarcinoma JEG-3 cells were transfected for 24 h with expression plasmids for flag control, wild-type NFAT5 (WT NFAT5) or NFAT5 dominant-negativemutant (DNM NFAT5). (C) NFAT5 was measured as a singlew170-kDa band. The expression of NFAT5 increased significantly in JEG-3 cells transfected with WT NFAT5 for 24 h. Therewas a significant decrease in the expression of NFAT5 when transfected with DNM NFAT5 for 24 h as compared with cells transfected with WT NFAT5. (D) Hsp70 was measured as asingle w70-kDa band. The expression of Hsp70 increased significantly in JEG-3 cells transfected with WT NFAT5. There was a significant decrease in the expression of Hsp70 whentransfected with the DNM NFAT5 for 24 h as compared with WT NFAT5. (E) The Hsp70 mRNA level increased significantly in JEG-3 cells transfected with WT NFAT5 for 24 h. Therewas a significant decrease in the Hsp70 mRNA level when transfected with the DNM NFAT5 for 24 h as compared with WT NFAT5. The lane loading, controls, and densitometricanalyses were as described in Fig. 2. Data are presented as means � SE; *p < 0.05.

J.K. Park et al. / Placenta 35 (2014) 109e116 115

hypoxia in vitro. shRNA knockdown of NFAT5 suppressed the in-duction of Hsp70 protein and overexpression of NFAT5 stimulatedthe induction of Hsp70 mRNA and protein in models of humanplacental hypoxia in vitro. These results indicate for the first timethat NFAT5 stimulates transcription of Hsp70 in response toplacental hypoxia.

In the present study, Hsp70was induced at the total Hsp70 levelupon placental hypoxia. However, a new isoformwas observed andthe lower-molecular-weight isoform of Hsp70 was highly inducibleupon placental hypoxia and was markedly inhibited upon inhibi-tion of HIF-1a or NFAT5. Further studies are needed to clarify thesignificance and mechanisms underlying the induction of a newisoform of Hsp70 upon placental hypoxia.

The present study showed that HIF-1a, NFAT5, and Hsp70 wereinduced at the protein level in vivo and in models of humanplacental hypoxia in vitro, suggesting that these proteins participatein the complex mechanisms of placental responses againstplacental hypoxia and form a regulatory loop. This regulatory loopprovides a potential target for the development of new treatmentstrategies in patients with preeclampsia.

Acknowledgments

This research was supported by a grant from Basic ScienceResearch Program of the National Research Foundation of Korea(2011-0020163, HMK).

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.placenta.2013.12.005.

References

[1] Redman CW, Sacks GP, Sargent IL. Preeclampsia: an excessive maternal in-flammatory response to pregnancy. Am J Obstet Gynecol 1999;180:499e506.

[2] Redman CW, Sargent IL. Latest advances in understanding preeclampsia.Science 2005;308:1592e4.

[3] Oyake J, Otaka M, Matsuhashi T, Jin M, Odashima M, Komatsu K, et al. Over-expression of 70-kDa heat shock protein confers protection againstmonochloramine-induced gastric mucosal cell injury. Life Sci 2006;79:300e5.

[4] Millar DG, Garza KM, Odermatt B, Elford AR, Ono N, Li Z, et al. Hsp70 promotesantigen-presenting cell function and converts T-cell tolerance to autoimmu-nity in vivo. Nat Med 2003;9:1469e76.

[5] Davies EL, Bacelar M, Marshall MJ, Johnson E, Wardle TD, Andrew SM, et al.Heat shock proteins form part of a danger signal cascade in response tolipopolysaccharide and GroEL. Clin Exp Immunol 2006;145:183e9.

[6] Luo X, Zuo X, Zhang B, Song L, Wei X, Zhou Y, et al. Release of heat shockprotein 70 and the effects of extracellular heat shock protein 70 on the pro-duction of IL-10 in fibroblast-like synoviocytes. Cell Stress Chaperones2008;13:365e73.

[7] Jirecek S, Hohlagschwandtner M, Tempfer C, Knöfler M, Husslein P, Zeisler H.Serum levels of heat shock protein 70 in patients with preeclampsia: a pilot-study. Wien Klin Wochenschr 2002;114:730e2.

[8] Fukushima A, Kawahara H, Isurugi C, Syoji T, Oyama R, Sugiyama T, et al.Changes in serum levels of heat shock protein 70 in preterm delivery and pre-eclampsia. J Obstet Gynaecol Res 2005;31:72e7.

[9] Fenton RA, Knepper MA. Mouse models and the urinary concentratingmechanism in the new millennium. Physiol Rev 2007;87:1083e112.

J.K. Park et al. / Placenta 35 (2014) 109e116116

[10] Miyakawa H, Woo SK, Dahl SC, Handler JS, Kwon HM. Tonicity-responsiveenhancer binding protein, a rel-like protein that stimulates transcription inresponse to hypertonicity. Proc Natl Acad Sci U S A 1999;96:2538e42.

[11] Villanueva S, Suazo C, Santapau D, Pérez F, Quiroz M, Carreño JE, et al. NFAT5is activated by hypoxia: role in ischemia and reperfusion in the rat kidney.PLoS One 2012;7:e39665.

[12] Trama J, GoWY,Ho SN. The osmoprotective function of the NFAT5 transcriptionfactor in T cell development and activation. J Immunol 2002;169:5477e88.

[13] Shin JA, Kwon HM, Han KH, Lee HY. TonEBP and SMIT expression in humanplacenta. Anat Cell Biol 2012;45:155e9.

[14] American College of Obstetricians and Gynecologists. ACOG practice bulletin.Diagnosis and management of preeclampsia and eclampsia. Int J GynaecolObstet 2002;77:67e75.

[15] Park JK, Jeong JW, Kang MY, Baek JC, Shin JK, Lee SA, et al. Inhibition of thePI3K-Akt pathway suppresses sFlt1 expression in human placental hypoxiamodels in vitro. Placenta 2010;31:621e9.

[16] Liu Y, Li N, You L, Liu X, Li H, Wang X. HSP70 is associated with endothelialactivation in placental vascular diseases. Mol Med 2008;14:561e6.

[17] Barut F, Barut A, Dogan Gun B, Kandemir NO, Aktunc E, Harma M, et al.Expression of heat shock protein 70 and endothelial nitric oxide synthase inplacental tissue of preeclamptic and intrauterine growth-restricted preg-nancies. Pathol Res Pract 2010;206:651e6.

[18] Gogate SS, Fujita N, Skubutyte R, Shapiro IM, Risbud MV. Tonicity enhancerbinding protein (TonEBP) and hypoxia-inducible factor (HIF) coordinateheat shock protein 70 (Hsp70) expression in hypoxic nucleus pulposuscells: role of Hsp70 in HIF-1a degradation. J Bone Miner Res 2012;27:1106e17.

[19] Ferraris JD, Williams CK, Persaud P, Zhang Z, Chen Y, Burg MB. Activity of theNFAT5 transactivation domain varies directly with extracellular NaCl con-centration. Proc Natl Acad Sci U S A 2002;99:739e44.