Original Article

Excessive autophagy induces the failure of trophoblast invasion and vasculature: possible relevance to the pathogenesis of preeclampsia Li Gao a,b, Hong-Bo Qi a,b, Kamana KC a,b, Xue-Mei Zhang a,b, Hua Zhang a,b, and Philip N. Baker b,c

Introduction: Preeclampsia affects 5–7% of all healthy pregnancies and is characterized by hypertension and proteinuria. Although the pathogenesis of preeclampsia is still not fully understood, a failure of spiral artery transformation and aberrant placental vasculature are considered to be facets of this disease. Studies have also implicated increased autophagic activity. In this study, we investigated whether oxidative stress could increase autophagic activity and consequently affect trophoblast invasion and the placental vasculature. Methods: Placentas from 18 pregnancies complicated by preeclampsia and from 18 uncomplicated pregnancies, trophoblast HTR8/SVneo cell line (HTR8/SVneo) extravillous trophoblasts, and human umbilical vein endothelial cells (HUVECs) were employed. The levels of autophagy markers LC3, Beclin-1 and autophagosome were quantified by immunohistochemistry, Western blotting and RT-PCR in placental tissue, and in trophoblasts and endothelial cells that had been treated with an oxidative stress inducer glucose oxidase. Trophoblast invasion and endothelial cell tube formation were assessed in HTR8/SVneo cells or HUVECs that had been treated with glucose oxidase. Results: The expression of LC3, Beclin-1 and autophagosome was significantly increased in placentas from pregnancies complicated by early-onset preeclampsia and in HTR8/SVneo cells and HUVECs treated with glucose oxidase. In addition, trophoblast invasion and endothelial cell tube formation were significantly reduced in HTR8/ SVneo cells or HUVECs that had been treated with glucose oxidase. Conclusion: Our data suggest that oxidative stress induces increased autophagy in trophoblasts or endothelial cells which affects trophoblast invasion and the placental vasculature. Excessive autophagic activity may be involved in the development of preeclampsia. Keywords: autophagy, endothelial tube formation, oxidative stress, preeclampsia, trophoblast invasion Abbreviations: Atg, autophagy-related genes; EOPE, early-onset preeclampsia; EVT, extravillous trophoblast; HUVECs, human umbilical vein endothelial cells; IUGR, intrauterine growth restriction; MMP-9, matrix metallopeptidase 9; VEGFR2, vascular endothelial cell growth factor receipt 2

INTRODUCTION

P

reeclampsia is a pregnancy-specific disorder, characterized by hypertension and proteinuria. It affects 5–7% of all healthy pregnancies and causes approximately 50 000 maternal deaths annually [1,2]. Although the pathogenesis of preeclampsia is still not fully understood, the key contribution of the placenta is well recognized. Autophagy has been implicated in the pathogenesis of preeclampsia [3–5]. Autophagy involves cellular degradation of unnecessary or dysfunctional cellular components and recycling of cellular components through the lysosome [6]. Autophagy functions as an intracellular bulk self-degradation system and has been seen as an adaptive response to survival under oxidative stress and other environmental stressors [7]. However, recent studies indicated that excessive autophagy could promote cell dysfunction through excessive degradation of essential cellular constituents [8,9]. During autophagy, unique membrane-bound autophagosomes are formed to sequestrate damaged cellular components. The formation of autophagosomes is initiated by class III phosphoinositide 3-kinase and autophagy-related genes (Atgs). One Atg product is LC3, a light chain of the microtubule-associated protein 1; LC3 is a reliable marker of autophagosomes, thought to be involved in the regulation of assembly and disassembly of microtubules [10]. Beclin-1, a novel Bcl-2-homology (BH)-3 domain only protein [11], is another marker for autophagosome formation in the early stages of autophagy [12]. Beclin-1 coordinates the autophagy and membrane trafficking involved in several physiological and pathological processes; it is important for localization of autophagic proteins to a preautophagosomal structure [13].

Journal of Hypertension 2015, 33:106–117 a Department of Obstetrics and Gynecology, The First Affiliated Hospital of Chongqing Medical University, bCanada–China–New Zealand Joint Laboratory of Maternal and Fetal Medicine, Chongqing Medical University, Chongqing, People’s Republic of China and cLiggins Institute, University of Auckland, Auckland, New Zealand

Correspondence to Hua Zhang, MD, PhD, Department of Obstetrics and Gynecology, The First Affiliated Hospital of Chongqing Medical University, Canada–China–New Zealand Joint Laboratory of Maternal and Fetal Medicine, Chongqing Medical University, Chongqing 400016, People’s Republic of China. Tel: +86 23 89011102; fax: +86 23 89011082; e-mail: [email protected] Received 19 August 2013 Revised 30 July 2014 Accepted 30 July 2014 J Hypertens 33:106–117 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. DOI:10.1097/HJH.0000000000000366

106

www.jhypertension.com

Volume 33
Number 1
January 2015

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Excessive autophagy in preeclampsia

Levels of oxidative stress are increased in preeclampsia [14]. Although previous studies have implicated increased autophagic activity, consequent on placental hypoxia, in the pathogenesis of preeclampsia [15], the evidence for this is limited. In this study, we investigated our hypothesis that in preeclampsia, increased oxidative stress could induce increased autophagic activity and consequently affect both trophoblast invasion and placental vasculature.

collected from mid-way between the edge and centre from each placenta.

Reagents Glucose oxidase, the stimulator of oxidative stress, was purchased from Sigma Aldrich (G-7141; Amresco, Ohio, USA). 3-Methyladenine (3-MA), the inhibitor of autophagosome formation, was purchased from Santa Cruz (CAS 5142-23-4; Santa Cruz, California, USA).

MATERIALS AND METHODS The study was approved by the Ethics Committee of First Hospital of Chongqing Medical University, China. All patient-derived tissue samples were obtained with written informed consent.

Study populations Consent was obtained at diagnosis from 18 nulliparous women with early-onset preeclampsia (EOPE). Placentas from these pregnancies, and from 20 nulliparous normotensive pregnancies (delivered at term), were collected at delivery (caesarean section) between January 2010 and May 2012 at the Department of Obstetrics and Gynaecology of The First Affiliated Hospital of Chongqing Medical University. EOPE was defined as a maternal SBP at least 140 mmHg and/or DBP at least 90 mmHg measured on two occasions separated by at least 6 h, and proteinuria above 300 mg on a 24-h urinary collection or qualitatively, greater than 1þ, less than 34 weeks of gestation in accord with the guidelines of the American College of Obstetricians and Gynecologists [16]. Patients with chronic medical disorders such as diabetes mellitus, cardiovascular disease, collagen disorder, chronic renal disease, chronic hypertension, and metabolic diseases were excluded. Patient characteristics, including BMI at the beginning of pregnancy and at the end of pregnancy, are summarized in Table 1.

Placental tissue preparation Placental tissue was prepared within 1 h of delivery. For immunohistochemistry, chorionic and decidual surfaces were removed, and placental tissue (maternal side) was sampled from two placental lobules at either the edge of the placenta or mid-way between the edge and centre from each placenta and stored at 808C for further studies. For western blotting, placental tissues (maternal side) were TABLE 1. Clinical characteristics of study populations

Age (years old) BMI (at the end of pregnancy) BMI at the beginning of pregnancy SBP (mmHg) DBP (mmHg) Gestation weeks at diagnosis Gestational weeks of delivery Neonatal birth weight (g) Placental weight (g) 

Early-onset preeclampsia (N ¼ 18)

Healthy pregnant (N ¼ 20)

31.5  3.9 35.38  5.62 21.78  1.29

30.3  5.2 27.46  4.19 20.17  2.31

163.2  19.1 120.7  14.5 31.5  3.6 33.5  1.1 1963.7  235.3 396.7  57.4

112.4  20.2 71.6  8.4 N/A 39.0  1.2 2886.4  609.7 574.4  82.6

P < 0.01 compared to health pregnant women.

Journal of Hypertension

Cell cultures The trophoblast (HTR-8/SVneo) cell line (HTR8/SVneo) human-transformed primary EVT cell line was kindly provided by Dr C.H. Graham (Queen’s University, Kingston, Ontario, Canada). Human umbilical vein endothelial cells (HUVECs) were isolated from umbilical cords obtained from uncomplicated pregnancies by enzymatic digestion with 0.1% collagenase (17100-17; GIBCO, Invitrogen, New York, USA) as previously described [17]. HUVEC and HTR8/ SVneo cells were cultured in M200 or RPMI1640 (GIBCO, Invitrogen) containing 10% foetal bovine serum (FBS, 10099; GIBCO, Invitrogen).

Cell treatment HTR8/SVneo cells or HUVECs were treated with glucose oxidase (10 mU/ml) for 5 h or 6 h and 45 min. In some experiments, the HTR8/SVneo cells or HUVECs were pre-treated with the 3-MA (10 mmol/l) inhibitor of autophagosome formation for 2 h, followed by the treatment of glucose oxidase for 5 h or 6 h and 45 min. The concentrations of glucose oxidase or 3-MA and the culture time were based on previous studies [18].

Transmission electron microscopy

Placental tissues (1 mm3/piece) or cell pellets (106) were fixed with 4% glutaraldehyde overnight at 48C, rinsed with 0.1 mol/l phosphate buffer (pH 7.4) twice (15 min) and post-fixed with 1% osmium tetroxide for 1 h. Samples were then dehydrated in ethanol, saturated in epoxy resin618, embedded in epok812 and then incubated in an oven overnight at 378C. Satisfactory fields were chosen under an ordinary light microscope; the pellets were then sectioned (50–60 nm) and stained with uranyl acetate and lead citrate. Fields were examined using a Hitachi-7500 transmission electron microscope (Hitachi Limited, Japan).

Immunohistochemistry Placental expression of LC3 and Beclin-1 was determined by immunohistochemical staining of paraffin-embedded sections (5 mm). Antigen retrieval was performed by treating with citric acid buffer (pH 6.0). Non-specific antibody binding was blocked by incubation with 10% normal goat serum for 30 min. Rabbit antihuman LC3 monoclonal antibody (1 : 500, NB100-2220; NOVUS, Colorado, USA) or Beclin-1 (1 : 500, 2026-1; EPITOMICS, California, USA) was then added and incubated overnight at 48C. Sections were then washed with PBS and incubated with biotinconjugated goat antirabbit IgG for 30 min at 378C, followed by treatment with horseradish peroxidase-labelled streptavidin for 30 min at 378C. After washing, sections were www.jhypertension.com

107

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Gao et al.

incubated with 3, 30 -diaminobenzidine (DAB; chromogenic reagent; ZSGB-BIO, Beijing, China) and counterstained with haematoxylin. Negative controls were stained without primary antibody.

Co-localization of oxidative stress marker and autophagy markers The co-localization of oxidative marker nitrotyrosine and autophagy markers LC3 and Beclin-1 was measured by immunofluorescence double staining of frozen sections (5 mm). Sections were fixed by 4% paraformaldehyde at room temperature for 30 min and then blocked by 10% normal goat serum for 1 h at room temperature. Sections were then incubated with primary antibodies of LC3 (1 : 50, NB100-2220; NOVUS) or Beclin-1 (1 : 50, 2026-1; EPITOMICS), and nitrotyrosine (1 : 50, NB110-96877; NOVUS) overnight at 48C. After washing with PBS, sections were incubated with fluorescein isothiocyanate (FITC) conjugated with antimouse IgG or tetramethylrhodamine conjugated with antirabbit IgG. Images were observed using a fluorescence microscope (Leica, Germany).

Quantitative real-time PCR Total RNA was extracted from HTR8/SVneo cells by RNAiso Plus (D9108A; TaKaRa, Dalian, China) extraction, and the RNA concentration was measured by ultraviolet spectroscopy (Nanodrop 2000; Thermo, Massachusetts, USA). Total RNA (1 mg) was used for reverse transcription with PrimeScriptRT reagent kit with gDNA Eraser (DRR047A; TaKaRa). Primers were designed and synthesized by Shanghai Sangon (China); b-actin was used as an endogenous control for gene expression analysis. The sequences of the PCR primer pairs for each gene are shown in Table 2. For quantitative real-time PCR (qRT-PCR), we used a real-time PCR instrument (Thermal Cycler Dice Real Time System) with the SYBRPremix Ex Taq II (DRR820A; TaKaRa). PCR cycling conditions included pre-denaturing at 958C for 3 min, followed by 40 cycles (948C for 5 s, 588C for 15 s, 728C for 15 s) and by an extension at 728C for 30 s. The mean threshold cycle (Ct) values were normalized to b-actin, and the relative mRNA levels of LC3B and Beclin-1 were analysed by the 2DDCT method. Experiments were performed in triplicates and repeated three times.

Western blotting

Immunofluorescence The levels of LC3, matrix metallopeptidase 9 (MMP-9) and vascular endothelial cell growth factor receipt 2 (VEGFR2) expression in HTR8/SVneo cells or HUVECs were determined by immunofluorescence staining. After treatment cells were fixed for 10 min with 4% paraformaldehyde at 48C, and blocked with goat serum for 1 h at room temperature. Cells were then incubated with rabbit antihuman primary antibody LC3 (1 : 250, NB100-2220; NOVUS), MMP-9 (1 : 200, SC-6840; Santa Cruz) and VEGFR2 (1 : 200, 21097; Signalway Antibody, SAB, Maryland, USA) overnight at 48C, and finally incubated with FITC conjugated with antirabbit IgG or FITC conjugated with antigoat IgG; the nuclei were labelled by propidium iodide. Cells were then observed by confocal microscopy (Olympus FV10i, Japan). Autophagic puncta (green) were calculated in 100 cells using Image J software.

Determination of cell apoptosis by Hoechst 3328 fluorescence staining HTR8/SVneo cells or HUVECs after treatments were fixed with 4% paraformaldehyde for 10 min at 48C. After washing for three times with PBS, the cells were stained with Hoechst 33258 (C1018; Beyotime Institute of Biotechnology, Shanghai, China) for 4 min at room temperature to counter-stain nuclei and apoptotic morphology observed by using a fluorescence microscope (Nikon TE2000-s, Japan).

After treatments, HTR8/SVneo cells and HUVECs (1.0  106) were lysed with radioimmunoprecipitation assay buffer (P0013; Beyotime Institute of Biotechnology). Protein concentrations were determined by the use of BCA Protein Assay Kit (P0010S; Beyotime Institute of Biotechnology), according to the manufacturer’s instructions. Protein samples (30 mg) were loaded in 15% (for LC3) or 10% (for Beclin-1, caspase-3, b-actin) SDS-polyacrylamide gels, resolved by electrophoresis and transferred to polyvinylidene difluoride membranes (PVDF 0.22 mm; Merck Millipore, Billerica, Massachusetts, USA). Immunoblotting was performed using primary antibodies against LC3 (1 : 1000; NOVUS), Beclin-1 (1 : 1000; EPITOMICS) or caspase-3 (1 : 1000; Santa Cruz), MMP-9 (1 : 500, sc-21733; Santa Cruz), VEGFR2 (1 : 500, 11084, Signalway Antibody) and b-actin (1 : 1000; TA-09; ZSGB-BIO), respectively. Followed by incubation with the appropriate horseradish peroxidase-linked secondary antibodies (1 : 1000; Beyotime Institute of Biotechnology), the chemiluminescence of target proteins on the membranes was developed with BeyoECL Plus (P0018; Beyotime Institute of Biotechnology). The densitometric analyses of bands were performed by using a Chemi-doc imager (Bio-Rad, Hercules, California, USA). b-Actin is used as the loading control.

HTR8/SVneo invasion assay The invasion assay was performed in 24-well plates as the outer chambers, and polycarbonate filters (8 mm pores;

TABLE 2. Sequence of primers and TaqMan probes of human LC3, Beclin-1 and b-actin Sequence 50 !30

Gene LC3B (NM-022818.4) Beclin-1 (NM-003766.3) b-actin (BC002409)

108

www.jhypertension.com

0

Forward primer: 5 -CGATACAAGGGTGAGAAGCAG-3 Reverse primer: 50 -CTGAGATTGGTGTGGAGACG-30 Forward primer: 50 -AGGTTGAGAAAGGCGAGACA-30 Reverse primer: 50 -AATTGTGAGGACACCCAAGC-30 Forward primer: 50 -AGCGAGCATCCCCCAAAGTT-30 Reverse primer: 50 -GGGCACGAAGGCTCATCATT-30

Size PCR product 0

181 bp

196 bp 285 bp

Volume 33
Number 1
January 2015

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Excessive autophagy in preeclampsia

Transwell chamber, Millipore) as the inner chambers. The upper surface of the inner chamber was coated with 50 ml of diluted matrigel (356234; Becton Dickinson and Company, Franklin Lakes, New Jersey, USA, 1:9 in RPMI-1640). HTR8/SVneo cell suspension (1.0  105 cells) of 200 ml was seeded into the inner chamber. The outer chamber contained 600 ml of complete culture medium (RPMI-1640 containing 10% FBS). HTR8/SVneo cells were cultured for 17 h and then cultured with glucose oxidase (10 mU/ml) in the presence or absence of 3-MA (10 mmol/l) in the inner chambers. Untreated HTR8/SVneo cells were used as a basal control. After further culturing for 5 h, the cells on the filter were fixed by methanol for 10 min. Non-migrated cells on the upper surface of the filter were removed by gentle scraping with a cotton swab. Migrated cells on the lower surface of the filter were stained with crystal violet and then counted in 10 random fields at 200 magnifications using a light microscope (IX51; Olympus, Japan). The average percentages of invaded HTR8/SVneo cells were calculated as a ratio to HTR8/SVneo cells that had not been treated with either glucose oxidase or 3-MA.

Statistical analysis

Human umbilical vein endothelial cell tube formation assay

The expression of LC3 (Fig. 1a–d) and Beclin-1 (Fig. 1e–h) was significantly increased in both trophoblasts (Fig. 1b and f) and endothelial cells (Fig. 1d and h) in preeclampsia placentas compared to control placentas (Fig. 2, P < 0.05). The autophagosome in preeclampsia placentas was investigated by transmission electron microscopy (TEM). As shown in Fig. 3, autophagic vacuoles were seen in the syncytiotrophoblast (Fig. 3b) and endothelial cells (Fig. 3d), with typical double membranes, which are the characteristic markers of the early-stage autophagosome.

Data are presented as mean  SD. The statistical significance of the results was assessed by a Mann–Whitney U test or a Kruskal–Wallis test [analysis of variance (ANOVA)] using the Graphpad Prism software package (version 5.0; La Jolla, California, USA) with P value less than 0.05 being considered significant.

RESULTS Demographic data of the study population The clinical characteristics of 18 women with EOPE and 20 women with uncomplicated pregnancies (control group) are summarized in Table 1. BMI at the beginning of pregnancy was collected at first antenatal visit (around 11 weeks of gestation). BMI at the end of pregnancy was collected at the time of admission in the hospital for delivery, and was significantly higher in the EOPE group.

The expression of LC3 and Beclin-1 was significantly increased in placentas from pregnancies complicated by early-onset preeclampsia

Tube formation assays were performed in 96-well plates that were pre-coated with 50 ml/well diluted matrigel (1 : 1 in M200). To the 96 well plates, 200 ml of HUVEC suspension (2.5  104 cells) in the presence of glucose oxidase (10 mU/ml) were added, with or without 3-MA. HUVECs only were used as basal control. After incubation for 6 h and 45 min at 378C, tube formation was quantitatively measured by calculating the total tube length of tube-like structures that had been captured by a camera (Olympus) interfaced with Image-Pro Plus image analysis software (NIH Image). Tracks of endothelial cells organized into networks of cellular cords (tubes) were counted and averaged from five random fields (200 magnifications). The tube formation index is expressed as tube length (mm) per mm2area. Normal

Increased levels of nitrotyrosine co-localized with the increase in LC3 (Fig. 4b1–b3) or Beclin-1 (Fig. 4d1–d3) in preeclampsia placentas compared with control placentas (Fig. 4 a1–a3, Fig. 4 c1–c3).

EOPE

LC3

a

A

b

B

Beclin-1

e

E

Increased autophagy was associated with oxidative stress in preeclampsia placentas

EOPE

c

d

C

D

g

f

F

Normal

G

h

H

FIGURE 1 Immunohistochemistry images demonstrated increased expression of LC3 in trophoblasts (b) and endothelial cells (d) in placentas from women with EOPE compared to placentas from uncomplicated pregnancies (a and c). Immunohistochemistry images demonstrated the increased expression of Beclin-1 in trophoblasts (f) and endothelial cells (h) in placentas from preeclampsia pregnancies (e and g) (Aa–h: magnification 100; 1–8: magnification 400). EOPE, early-onset preeclampsia.

Journal of Hypertension

www.jhypertension.com

109

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Gao et al.

*

60 40 20

*

20 10 0

Be cl in

-1

LC 3

0

* 30

Be cl in

80

40

-1

EOPE

*

(b) Posistive endothelial cell (%)

Normal

100

LC 3

Positive trophoblast cell (%)

(a)

FIGURE 2 Semi-quantitative immunohistochemistry analysis showed that the percentage of LC3 or Beclin-1-positive trophoblasts (a) or endothelial cells (b) was significantly increased in placentas from cases of EOPE as compared to uncomplicated pregnancies (P < 0.05). EOPE, early-onset preeclampsia.

Autophagy was elevated by increasing oxidative stress in endothelial cells and trophoblasts Previous studies showed that glucose oxidase treatment induces the increase of oxidative stress in a number of cells, including endothelial cells [19,20]. To investigate whether treatment with glucose oxidase could enhance autophagy in trophoblasts or endothelial cells, HTR8/SVneo cells or HUVECs were treated with glucose oxidase. TEM images demonstrated that autophagic vacuoles (autophaosome) that contained degraded sub-cellular organelles were seen

in both HTR8/SVneo cells (Fig. 3f) and HUVECs (Fig. 3h) that had been treated with glucose oxidase. Immunofluorescence images demonstrated that autophagosomes (labelled as grey punta) were increased in the HTR8/SVneo cells (Fig. 5a) and HUVECs (Fig. 5b) that had been treated with glucose oxidase. However, the increase in autophagosomes (grey punta) was significantly inhibited by the treatment of 3-MA in both the HTR8/SVneo cells (Fig. 5c) and HUVECs (Fig. 5d,  P < 0.001, #P < 0.001). The mRNA (Fig. 6) and protein (Fig. 7) levels of LC3 and Beclin-1 were significantly increased in the HTR8/SVneo

Trophoblast cell EOPE

Normal (a)

Vascular endothelial cell EOPE

Normal (c)

(b)

(d)

Tissue Untreated (e)

GO (f)

Untreated (g)

GO (h)

Cell FIGURE 3 Transmission electron microscopy images demonstrated: the formation of early-stage autophagosomes with double membrane (arrow) in either trophoblasts (b) or endothelial cells (d) in placenta from EOPE pregnancies; the autolysosome [characteristic of later-stage autophagosome formation (arrow)] in HTR8/SVneo cells treated with GO (f) as compared to untreated HTR8/SVneo cells (e); the early stage of autophagosome (arrow) in HUVECs treated with GO (h) as compared to untreated HUVECs (g). EOPE, early-onset preeclampsia; GO, glucose oxidase; HUVECs, human umbilical vein endothelial cells.

110

www.jhypertension.com

Volume 33
Number 1
January 2015

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Excessive autophagy in preeclampsia NT

Merge

LC3 a1

A2

a2

A3

a3

B1

b1

B2

b2

B3

b3

C1

c1

C2

c2

C3

c3

D1

d1

D2

d2

D3

d3

Normal

A1

EOPE NT

Merge

Beclin-1

Normal EOPE FIGURE 4 Immunofluorescence double staining demonstrated that increased levels of NT (b1 or d1) co-localized with either LC3 (b2) or Beclin-1 (d2) in placentas from preeclampsia as compared to uncomplciated pregnancies (a1–a3 or c1–c3). a3, b3, c3, and d3 were merged images (a–d magnification 200; 1–12 original magnification 400). EOPE, early-onset preeclampsia; NT, nitrotyrosine.

cells (Fig. 6a and c, Fig. 7a and c) or HUVECs (Fig. 6b and d, Fig. 7b and d) that had been treated with GO, as compared to untreated cells. The increase in mRNA and protein levels of LC3 and Beclin-1 in HTR8/SVneo cells or HUVECs was significantly inhibited by treating with 3-MA, an inhibitor of autophagosome formation (Figs. 6 and 7, P < 0.01, # P < 0.001, P < 0.001).

Oxidative stress did not induce apoptosis in HTR8/SVneo cells or human umbilical vein endothelial cells To investigate whether glucose oxidase treatment induces apoptosis in HTR8/SVneo cells or in HUVECs, HTR8/SVneo cells or HUVECs were treated with glucose oxidase, and then the protein levels of caspase-3, which is an apoptotic Journal of Hypertension

marker, were measured. In addition, apoptotic morphology was also measured by Hoechst 33258, a fluorescent dye for nuclear staining. Hoechst 33258 staining results demonstrated that glucose oxidase treatment alone did not increase the number of apoptotic cells in either HTR8/ SVneo (Fig. 8a–d) or HUVECs (Fig. 8e–h). In addition, the expression levels of caspase-3 in HTR8/SVneo cells and HUVECs were not altered by glucose oxidase treatment either, as shown in western blotting results (Fig. 8i).

Trophoblast invasion was reduced by the treatment with glucose oxidase To investigate whether increased autophagy affects the function of trophoblasts, a trophoblast invasion assay was performed (Fig. 9a–d). The percentage of migrated www.jhypertension.com

111

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Gao et al. (a)

(b) HTR8 PI

LC3

Merge

HUVEC PI

LC3

Merge

Untreated GO 3-MA 3-MA+GO #

(c)

# 80

* Puncta/cell (HUVEC)

20

10

0

* 60

40

20

M A +G

O

A

O G

3M

3-

nt

re

at

ed

O A +G

3M

O G

A 3M

U nt

re a

te

d

0

U

Puncta/cell (HTRS)

30

(d)

FIGURE 5 Confocal microscopy images demonstrated that fluorescent densities of LC3 (grey punta) were increased in HTR8/SVneo cells (a) and HUVECs (b) after treatment with GO (10 mU/ml). However, the fluorescent densities of LC3 were reduced by the treatment with 3-MA (10 mmol/l) in HTR8/SVneo cells (a) and HUVECs (b). The numbers of grey fluorescence puncta in HTR8/SVneo cells (c) and HUVECs (d) were increased (P < 0.001) following treatment with GO; this increase was reduced by the treatment with 3-MA, as quantified by counting 100 cells (#P < 0.001). GO, glucose oxidase; HUVECs, human umbilical vein endothelial cells.

HTR8/Svneo cells following treatment of glucose oxidase was significantly reduced compared to untreated HTR8/ Svneo cells (Fig. 9e, P < 0.001). This reduction of HTR8/ Svneo cells invasion was partially prevented by the treatment of 3-MA (Fig. 9e, ##P < 0.001). To further investigate whether MMP-9 expression is affected by autophagy, the protein levels of MMP-9 were measured in HTR8/SVneo cells that had been treated with glucose oxidase. Both immunoflurescence (Fig. 9f) and Western blotting results (Fig. 9g, P < 0.001) show an 112

www.jhypertension.com

expression reduction of MMP-9 in glucose oxidase-treated HTR8/SVneo cells.

Endothelial cell tube formation was reduced by the treatment of glucose oxidase To investigate whether increased autophagy affects the function of endothelial cells, HUVECs were treated with glucose oxidase and an endothelial cell tube formation assay was performed (Fig. 10a–d). Endothelial cell tube formation index was significantly reduced in HUVECs that Volume 33
Number 1
January 2015

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Excessive autophagy in preeclampsia (b)

1

3-

0.5 0.0

3-

+G O

te d ea

+G

3-

3-

U

nt r

M A

3-

nt r U

O

3-

1.0

O

M A

0

1.5

M A

1

* 2.0

M A

2

2.5

O

Relative mRNA expression of Beclin-1 in HUVEC

*

G O

O

U nt r

3-

U nt r 3

te d

#

(d)

#

ea

G

ea

+G

O

te d

0

M A

3-

ea

G

O

M A

0

2

+G

1

* 3

M A

2

4

M A

3

(c) Relative mRNA expression of Beclin-1 in HTRS

#

Relative mRNA expression of LC3B in HUVEC

*

4

te d

Relative mRNA expression of LC3B in HTRS

# 5

G

(a)

FIGURE 6 mRNA expressions of LC3B (a, b) and Beclin-1 (c, d) were increased in HTR8/SVneo cells (a, c) and HUVECs (b, d) that were treated with GO (10 mU/ml) (P < 0.001). The increased levels of mRNA of LC3B and Beclin-1 were reduced by the treatment with 3-MA (10 mmol/l) in HTR8/SVneo cells (a, c) and HUVECs (b, d) (#P < 0.05). 3-MA treatment alone did not change the levels of mRNA of LC3B and Beclin-1 in either HTR8/SVneo cells or HUVECs. GO, glucose oxidase; HUVECs, human umbilical vein endothelial cells. (a)

(b)

1

2

3

4

1

2

3

Relative ratio of LC3II/LC3I

HUVEC

HTR8

4

LC3 I 19 kd LC3 II 17kd

Beclin-1 52kd

##

HTR8 2.5

HUVEC

2.0

**

1.5

*

1.0 0.5

O +G

M A

M A 3-

3-

G O

te d

O nt r U

3-

ea

+G

M A 3-

nt r U

M A

te d ea

G O

0.0 B-actin 43kd

# (c)

#

Relative expression level of Beclin-1

0.6

* * 0.4

0.2

+G O

M A

3M A

3-

G O

ea

te d

+G O

nt r U

G O

M A 3-

3M A

U

nt r

ea

te d

0.0

FIGURE 7 Western-blotting analysis demonstrated that the ratio of LC3-II/LC3-I expression (a, b) and the relative expression levels of Beclin-1 (c) in both HTR8/SVneo cells and HUVECs that were treated with GO (10 mU/ml) were increased compared to untreated ones (P < 0.01, P < 0.001). The increased ratios of LC3-II/LC3-I expression and Beclin-1 expression were significantly inhibited by treatment with 3-MA (10 mmol/l) (b, c; #P < 0.01, ##P < 0.001) (a: line 1, untreated; line 2, GO-treated; line 3, 3-MA alone; line 4, GO þ 3-MA). GO, glucose oxidase; HUVECs, human umbilical vein endothelial cells.

Journal of Hypertension

www.jhypertension.com

113

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Gao et al. Untreated

GO

3-MA

3-MA+GO

(b)

(c)

(d)

(e)

(f)

(g)

(h)

HTR8

(a)

HUVEC HUVEC

HTR8 Cleaved Caspase-3 17 kd β-actin 43 kd

1

2

3

4

1

2

3

4

(i) Relative expression level of cleaved caspase-3

0.20

0.15

0.10

0.05

M A +G O

3-

A M

3-

G O

O

M A +G U O nt re at ed A

3-

M

3-

G

U

nt

re

at

ed

0.00

FIGURE 8 Hoechst 33258 staining images demonstrated that apoptosis was not increased in HTR8/SVneo cells and HUVECs that were treated with GO (10 mU/ml) (a–h) compared with untreated cells. Western blotting data demonstrated that the expression level of cleaved caspase-3 (activation type) was not increased in either HTR8/SVneo cells or HUVECs (i; P > 0.05) by the treatment of GO. GO, glucose oxidase; HUVECs, human umbilical vein endothelial cells.

had been treated with glucose oxidase, compared to untreated HUVECs (Fig. 10e). This reduction of endothelial cell tube formation index was partially prevented by treating with 3-MA (Fig. 10e, P < 0.001, ##P < 0.001). The expression of a potential mediator of endothelial cell proliferation, VEGFR2, was measured in HUVECs that had been treated with glucose oxidase. Both immunofluorescence (Fig. 10f) and Western blotting (Fig. 10g, P < 0.01, # P < 0.001) results show that VEGFR2 expression in HUVECs was suppressed by glucose oxidase treatment.

DISCUSSION Autophagy is a biological process that cell degrades cellular components through the lysosomal pathway for survival in response to starvation. Induction of autophagy provides the cell with molecular building blocks and energy. Autophagy is thought to be involved in early human placentation [21]. 114

www.jhypertension.com

A number of studies have reported that typical markers of autophagy, LC3 and Beclin-1 expression, are significantly increased in placentas from pregnancies complicated by preeclampsia and intrauterine growth restriction (IUGR), suggesting that autophagy plays a key role in the pathogenesis of these diseases [21,22]. In this study, we also confirmed the increased expression of LC3 and Beclin-1 in trophoblasts and in endothelial cells in placentas from EOPE; this is partially consistent with a recent study which showed increased LC3, but not Beclin-1, expression in placentas from pregnancies complicated by preeclampsia [15]. Our TEM images also demonstrated the formation of autophagic vacuoles with the typical double membrane, which is the gold standard of determining the early stages of autophagy. The discrepancy between our findings and those of Oh et al. [15] could be due to the difference in study population. The present study only included EOPE cases, whereas the study by Oh et al. included both early Volume 33
Number 1
January 2015

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Excessive autophagy in preeclampsia Untreated (a)

GO

3-MA

(b)

(f) (e)

(c)

Untreated

GO

MMP9

3-MA+GO (d)

3-MA

3-MA+GO (g)

3-MA Untreated GO 3-MA +GO

0.1

O +G

A

A

M M 3-

3-

O

0.0 G

M A 3M A +G O

3-

G

U

nt r

ea

O

Merge

*

0.2

ed

0

0.3

at

*

0.4

re

50

0.5

nt

PI

100

β-actin 43kd

U

##

The realative expression of MMP9 in HTRS

150

te d

Invasion cell (%)

MMP9 92kd

FIGURE 9 The microscopy images showed that the numbers of invasive HTR8/SVneo cells were reduced by the treatment of GO (10 mU/ml) (b) compared to untreated cells (a) or cells treated with 3-MA alone (c). The reduced number of invasive HTR8/SVneo cells was partially prevented by the treatment with 3-MA (10 mmol/l) (d). Quantification analysis demonstrated that the number of invasive HTR8/SVneo cells was significantly reduced by treating with GO; this effect was prevented by the treatment with 3-MA (P < 0.001, #P < 0.01, e) (magnification 400). MMP-9 expression was significantly reduced in HTR8/Svneo cells that had been treated with GO (10 mU/ml); reduced levels of MMP-9 were partially reversed by 3-MA treatment (10 mmol/l) measured by immnuofluorescence (f) and western blotting (g) compared with those not treated with GO (P < 0.001; GO vs. 3-MA þ GO; P ¼ 0.872) (magnification 400). GO, glucose oxidase; HUVECs, human umbilical vein endothelial cells; MMP9, matrix metallopeptidase 9.

and late severe preeclampsia, and it is known that EOPE and late-onset preeclampsia have different causes, as well as clinical parameters and/or laboratory biomarkers [23]. In addition, compared to the study by Oh et al., the present study has a larger sample size. A recent study reported increased autophagy in extravillous trophoblast (EVT) cells under physiological or pathophysiological hypoxia, under both in-vitro and in-vivo conditions [21]. The expression of autophagy marker Beclin-1 was regulated by hypoxia condition [15]. Hypoxic microenvironments can cause oxidative stress which may be involved in the pathogenesis of preeclampsia. Whether increased oxidative stress induces autophagy in placental trophoblasts or endothelial cells, consequently affecting trophoblast functions and placental vasculature during placentation, has not been investigated. Our data showed that the autophagosome as well as the mRNA levels and protein levels of LC3 and Beclin-1 were significantly increased in both EVT and endothelial cells that had been treated with glucose oxidase, which is an oxidative stress inducer; this increase was inhibited by the treatment of 3-MA, which is an inhibitor of autophagosome formation [24,25]. We also showed the co-localization of nitrotyrosine, a well characterized oxidative stress marker, with both LC3 and Beclin-1 in preeclampsia placentas, suggesting oxidative stress, increases autophagy in preeclampsia. Treatment with the oxidative stress inducer, glucose oxidase, did not cause apoptosis of EVT and endothelial Journal of Hypertension

cells. Taken together, our data suggest that elevated autophagic activity in EVT and endothelial cells in preeclampsia is oxidative stress-dependent. The failure of EVT invasion and spiral artery remodelling may contribute to the pathogenesis of preeclampsia. We demonstrated that the invasive ability of EVT was reduced by treatment with glucose oxidase. This reduction was partially reversed by treating with 3-MA, an inhibitor of autophagosome formation. In addition, our data showed that the levels of MMP-9 were reduced in EVT that had been treated with glucose oxidase. The expression of MMP-9 is thought to enhance invasion [26]. Taken together, our data suggest that the reduction of EVT invasion induced by increased oxidative stress is autophagy-dependent. This is consistent with reports that inhibition of autophagic degradation of LC3 by transfecting ATG4B into EVT significantly reduced EVT invasion [21]. The placental vasculature plays an important role in placental development and is critical for nutrient, gas, and waste exchange on the foeto-maternal interface. Its development depends on the proper expression and interaction of angiogenetic signals and relevant growth factors. Impaired development of the placental vasculature may be involved in the pathogenesis of preeclampsia and increased oxidative stress has been implicated [27]. Our results showed that the endothelial cell tube formation was significantly decreased by glucose oxidase-induced oxidative stress. VEGFR2 has been suggested to participate in www.jhypertension.com

115

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Gao et al. Untreated

3-MA

GO

(a)

(b)

(f)

(e)

3-MA+GO (d)

(c)

Untreated

GO

3-MA

3-MA+GO

(g) 3-MA Untreated GO 3-MA +GO

VEGFR2 230 kd β-actin 43 kd

0.05 +G O A 3-

U

A

0.00 M

+G O

A

A

3-

M M 3-

U

G O

at ed

Merge

0.10

3-

0

*

0.15

M

5

0.20

G O

PI

**

10

at ed

15

# 0.25

nt re

The relative expression of VEGFR2 in HUVEC

## 20

nt re

Tube length (min/min2)

VEGFR2

FIGURE 10 Microscopy images showed that the endothelial cell tube length was reduced by the treatment of GO (10 mU/ml) (b) as compared to untreated cells (a) or treating with 3-MA alone (c). The reduction induced by GO was partially prevented by the presence of 3-MA (10 mmol/l) (d). Quantification analysis demonstrated that endothelial cell tube length was significantly reduced by the treatment with GO; this reduction was prevented by the presence of 3-MA (e, P < 0.001, ##P < 0.01, magnification 100). VEGFR2 and MMP-9 expression was significantly reduced in HUVECs that had been treated with GO (10 mU/ml); this was partially reversed by the treatment with 3-MA (10 mmol/l): immnuofluorescence (f, magnification 400) and western blotting (g) (P < 0.001, #P < 0.05). GO, glucose oxidase.

endothelial cell proliferation. Inhibition of VEGFR2 limits endothelial cell proliferation [28], and we found reduced VEGFR2 expression of endothelial cells that had been treated with glucose oxidase. Placental vascular dysfunction induced by increased oxidative stress may be through the induction of increased autophagy activity in preeclampsia. Elevated maternal BMI during pregnancy is a risk factor for developing preeclampsia. Recent studies suggest that maternal obesity may be associated with increased oxidative stress [29,30]. In this study, the BMI in preeclampsia pregnancies was greater than that in uncomplicated pregnancies. Accordingly, our data also showed that the oxidative stress marker, nitrotyrosine, was significantly increased in preeclamptic placentas. The increased oxidative stress in preeclamptic placentas may be associated with this higher BMI in preeclampsia. There are some limitations in this study in terms of the control placentas. Due to ethical issues, we were not able to collect healthy placentas at matched gestational ages. Therefore, our findings need to be confirmed using gestation-matched healthy placentas in the future. In addition, this study is also restricted by relatively small sample sizes and our findings need to be confirmed by larger sample size studies in the future. During pregnancy, enhancement of autophagy was found in EVT under physiological hypoxia in early gestation in order to achieve success in placental development. However, impaired autophagy in EVT in early placental tissue could contribute to the pathogenesis of preeclmpsia [21]. In conclusion, our data showed that oxidative stress induces increased autophagy in both trophoblasts and endothelial cells. The increased 116

www.jhypertension.com

autophagy activity in trophoblasts affected trophoblast invasion function, and increased autophagy activity in endothelial cells affected by placental vasculature. Our results suggest that excessive autophagy activity may be one of the factors involved in the development of preeclampsia. A better understanding of the mechanisms of autophagy activity may potentially provide a future interventional therapeutic target.

ACKNOWLEDGEMENTS The study was supported by National Natural Science Foundation of China (No. 81170585, 81100444, 81370732), Chongqing Science and Technology Commission of China (No. 2011BB5121) and Chongqing Municipal Health Bureau of China (No.2011-2-046). We are grateful to G. Liao and B. Liu for excellent technical assistance. Great thanks are given to Dr C.H. Graham for donating the HTR8/SVneo cell line. We also would like to thank Dr James Hearn, a native English speaker from the University of Auckland, for editing this manuscript.

Conflicts of interest There are no conflicts of interest.

REFERENCES 1. Redman CW, Sargent IL. Latest advances in understanding preeclampsia. Science 2005; 308:1592–1594. 2. Duley L. Preeclampsia and the hypertensive disorders of pregnancy. Br Med Bull 2003; 67:161–176. 3. Chen B, Longtine MS, Nelson DM. Hypoxia induces autophagy in primary human trophoblasts. Endocrinology 2012; 153:4946–4954. 4. Saito S, Nakashima A. Review: the role of autophagy in extravillous trophoblast function under hypoxia. Placenta 2013; 34 (Suppl):S79–S84.

Volume 33
Number 1
January 2015

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Excessive autophagy in preeclampsia 5. Signorelli P, Avagliano L, Virgili E, Gagliostro V, Doi P, Braidotti P, et al. Autophagy in term normal human placentas. Placenta 2011; 32:482– 485. 6. Lin NY, Beyer C, Giessl A, Kireva T, Scholtysek C, Uderhardt S, et al. Autophagy regulates TNFalpha-mediated joint destruction in experimental arthritis. Ann Rheum Dis 2013; 72:761–768. 7. Lee J, Giordano S, Zhang J. Autophagy, mitochondria and oxidative stress: cross-talk and redox signalling. Biochem J 2012; 441:523– 540. 8. Patel AS, Lin L, Geyer A, Haspel JA, An CH, Cao J, et al. Autophagy in idiopathic pulmonary fibrosis. PLoS One 2012; 7:e41394. 9. Choi J, Jo M, Lee E, Oh YK, Choi D. The role of autophagy in human endometrium. Biol Reprod 2012; 86:70; p. 1–10. 10. Matsushita M, Suzuki NN, Obara K, Fujioka Y, Ohsumi Y, Inagaki F. Structure of Atg5.Atg16, a complex essential for autophagy. J Biol Chem 2007; 282:6763–6772. 11. Oberstein A, Jeffrey PD, Shi YG. Crystal structure of the Bcl-X-L-beclin 1 peptide complex: Beclin 1 is a novel BH3-only protein. J Biol Chem 2007; 282:13123–13132. 12. Ferraro E, Cecconi F. Autophagic and apoptotic response to stress signals in mammalian cells. Arch Biochem Biophys 2007; 462:210–219. 13. Kang R, Zeh HJ, Lotze MT, Tang D. The Beclin 1 network regulates autophagy and apoptosis. Cell Death Differ 2011; 18:571–580. 14. Raijmakers MT, Dechend R, Poston L. Oxidative stress and preeclampsia: rationale for antioxidant clinical trials. Hypertension 2004; 44:374– 380. 15. Oh SY, Choi SJ, Kim KH, Cho EY, Kim JH, Roh CR. Autophagy-related proteins, LC3 and Beclin-1, in placentas from pregnancies complicated by preeclampsia. Reprod Sci 2008; 15:912–920. 16. Diagnosis and management of preeclampsia and eclampsia. Obstet Gynecol 2002; 99:159–166. 17. Jaffe EA, Nachman RL, Becker CG, Minick CR. Culture of human endothelial cells derived from umbilical veins. Identification by morphologic and immunologic criteria. J Clin Invest 1973; 52:2745– 2756. 18. Choe Y, Yu JY, Son YO, Park SM, Kim JG, Shi X, Lee JC. Continuously generated H2O2 stimulates the proliferation and osteoblastic differentiation of human periodontal ligament fibroblasts. J Cell Biochem 2012; 113:1426–1436.

19. Kaczara P, Sarna T, Burke JM. Dynamics of H2O2 availability to ARPE19 cultures in models of oxidative stress. Free Rad Biol Med 2010; 48:1064–1070. 20. Rost D, Welker A, Welker J, Millonig G, Berger I, Autschbach F, et al. Liver-homing of purified glucose oxidase: a novel in vivo model of physiological hepatic oxidative stress (H2O2). J Hepatol 2007; 46:482– 491. 21. Nakashima A, Yamanaka-Tatematsu M, Fujita N, Koizumi K, Shima T, Yoshida T, et al. Impaired autophagy by soluble endoglin, under physiological hypoxia in early pregnant period, is involved in poor placentation in preeclampsia. Autophagy 2013; 9:303–316. 22. Hung TH, Chen SF, Lo LM, Li MJ, Yeh YL, Hsieh TT. Increased autophagy in placentas of intrauterine growth-restricted pregnancies. PLoS One 2012; 7:e40957. 23. Lo JO, Mission JF, Caughey AB. Hypertensive disease of pregnancy and maternal mortality. Curr Opin Obstet Gynecol 2013; 25:124–132. 24. Petiot A, Ogier-Denis E, Blommaart EF, Meijer AJ, Codogno P. Distinct classes of phosphatidylinositol 30 -kinases are involved in signaling pathways that control macroautophagy in HT-29 cells. J Biol Chem 2000; 275:992–998. 25. Seglen PO, Gordon PB. 3-Methyladenine: specific inhibitor of autophagic/lysosomal protein degradation in isolated rat hepatocytes. Proc Natl Acad Sci U S A 1982; 79:1889–1892. 26. Xu DM, Mckee CM, Cao YH, Ding YC, Kessler BM, Muschel RJ. Matrix metalloproteinase-9 regulates tumor cell invasion through cleavage of protease nexin-1. Cancer Res 2010; 70:6988–6998. 27. Myatt L, Kossenjans W, Sahay R, Eis A, Brockman D. Oxidative stress causes vascular dysfunction in the placenta. J Matern Fetal Med 2000; 9:79–82. 28. Shimotake J, Derugin N, Wendland M, Vexler ZS, Ferriero DM. Vascular endothelial growth factor receptor-2 inhibition promotes cell death and limits endothelial cell proliferation in a neonatal rodent model of stroke. Stroke 2010; 41:343–349. 29. Malti N, Merzouk H, Merzouk SA, Loukidi B, Karaouzene N, Malti A, Narce M. Oxidative stress and maternal obesity: Feto-placental unit interaction. Placenta 2014; 35:411–416. 30. Saben J, Lindsey F, Zhong Y, Thakali K, Badger TM, Andres A, et al. Maternal obesity is associated with a lipotoxic placental environment. Placenta 2014; 35:171–177.

Reviewers’ Summary Evaluations

Referee 3

Referee 2 The authors report that oxidative stress induces increased autophagy in trophoblasts or endothelial cells with increased expression of LC3 and Beclin-1 in placentae from early-onset preeclamptic women by immunohistochemistry. This is in agreement with another recent study. Due to ethical reasons, the authors could not collect 2nd trimester placentae for comparison. This study is also limited by the relatively small sample size.

Journal of Hypertension

In this manuscript the authors clearly demonstrate that multiple markers of autophagy are increased in early onset preeclamptic placentas when compared to normal placentas. Overall, the manuscript conclusively demonstrates in both placentas and cell lines that there is a correlation between oxidative stress and excessive autophagy, which in turn induces failure of trophoblast and vasculature, thereby providing clues into the etiology of preeclampsia. The major drawbacks of the study are the small sample size and the fact that the conclusions stem from the comparison of preterm preeclamptic placentas with full-term control placentas.

www.jhypertension.com

117

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.