Placental glucose transporter (GLUT)-1 is down-regulated in preeclampsia

Accepted Manuscript Placental glucose transporter (GLUT)-1 is down-regulated in preeclampsia Benjamin P. Lüscher, PhD, Camilla Marini, Marianne Joerger-Messerli, Xiao Huang, Matthias A. Hediger, Christiane Albrecht, Daniel V. Surbek, Marc U. Baumann PII:

S0143-4004(17)30254-0

DOI:

10.1016/j.placenta.2017.04.023

Reference:

YPLAC 3641

To appear in:

Placenta

Received Date: 19 May 2016 Revised Date:

25 April 2017

Accepted Date: 26 April 2017

Please cite this article as: Lüscher BP, Marini C, Joerger-Messerli M, Huang X, Hediger MA, Albrecht C, Surbek DV, Baumann MU, Placental glucose transporter (GLUT)-1 is down-regulated in preeclampsia, Placenta (2017), doi: 10.1016/j.placenta.2017.04.023. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Original Article

Placental glucose transporter (GLUT)-1 is down-regulated in preeclampsia Benjamin P. Lüschera, b*+, Camilla Marinia, b, d+, Marianne Joerger-Messerlia, Xiao Huangb, c,

RI PT

Matthias A. Hedigerb,c, Christiane Albrechtb, c, Daniel V. Surbeka, b, §, Marc U. Baumanna, b, §

[email protected]; [email protected]; [email protected];

[email protected]; [email protected];[email protected]; [email protected]; [email protected] a

Hospital of Bern, University of Bern, Bern, Switzerland

b

M AN U

Inselspital, 3010 Bern, Switzerland

SC

Department of Obstetrics and Gynecology, and Department of Clinical Research, University

Swiss National Center of Competence in Research, NCCR TransCure, University of Bern, Bern,

Switzerland (www.nccr-transcure.ch) c

Institute of Biochemistry and Molecular Medicine, University of Bern, Bern, Switzerland

Bühlstrasse 28, CH-3012 Bern, +41 (0)31 631 41 11, +41 (0)31 631 37 37 d

Switzerland §

Co-senior authorship

+

TE D

Graduate School for Cellular and Biomedical Sciences (GCB), University of Bern, Bern,

EP

Equal contribution *Corresponding author: Benjamin P. Lüschera,b PhD, Department of Obstetrics and Department of Clinical Research, University Hospital of Bern, University of Bern, Effingerstrasse 102, Inselspital, 3010 Bern, Switzerland. Telephone: +41 (0)31 632 85 17,

AC C

E-mail: [email protected] Word count: Abstract: 213, Text: 2667 The study was funded in part by the Swiss National Science Foundation NCCR grant “TransCure” The authors report no conflict of interest.

ACCEPTED MANUSCRIPT

Abstract Introduction: Transplacental fetal glucose supply is predominantly regulated by glucose

RI PT

transporter-1 (GLUT1). Altered expression and/or function of GLUT1 may affect the intrauterine environment, which could compromise fetal development and may contribute to fetal

programming. To date it is unknown whether placental GLUT1 is affected by preeclampsia,

SC

which is often associated with intrauterine growth restriction (IUGR). We addressed the

hypothesis that preeclampsia leads to decreased expression and function of placental GLUT1.

M AN U

Methods: Placentae were obtained following normal pregnancy and from pregnancies affected by preeclampsia. Washed villous tissue fragments were used to prepare syncytial microvillous (MVM) and basal plasma membranes (BM) microvesicles. GLUT1 protein and mRNA expression was assessed by Western blot analysis and qPCR using Fast SYBR Green. Radio-labeled glucose

function.

TE D

up-take assay using placenta-derived syncytial microvesicles was used to analyze GLUT1

Results: GLUT1 protein expression was significantly down-regulated in (apical) MVM of the

EP

syncytiotrophoblast in preeclampsia (n=6) compared to controls (n=6) (0.40±0.04 versus

AC C

1.00±0.06, arbitrary units, P<0.001, Student’s t-test), while GLUT1 mRNA expression did not show a significant difference. In addition, the functional assay in syncytial microvesicles showed a significantly decreased in glucose transport activity in preeclampsia (61.78 ± 6.48 %, P<0.05) compared to controls. BM GLUT1 protein expression was unchanged and glucose up-take into BM microvesicles showed no differences between the preeclampsia group and controls. Discussion: Our study shows for the first time that in preeclampsia placental GLUT1 expression and function are down-regulated at the apical plasma membrane of the syncytiotrophoblast.

ACCEPTED MANUSCRIPT

Further studies are needed to assess whether these changes occur also in vivo and may contribute to the development of IUGR in preeclampsia.

AC C

EP

TE D

M AN U

SC

RI PT

Key words: Glucose transport; GLUT1; placenta; preeclampsia; syncytiotrophoblast.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Introduction Glucose is the most important energetic substrate for fetal wellbeing and growth. Fetal glucose

RI PT

availability is entirely dependent on maternal supply and subsequent transplacental transport since there is no significant de novo gluconeogenesis in the fetus. Thus, appropriate regulation of transplacental glucose transport is crucial to fetal development and survival [1].

SC

In the human term placenta the main barrier layer between the mother and her child is the syncytiotrophoblast, which represents a multi-nucleated, terminally differentiated syncytium

M AN U

maintained by fusion of underlying cytotrophoblast cells. The syncytiotrophoblast possess an epithelial structure comprising maternal-facing microvillous membrane (MVM) and fetal-facing basal membrane (BM). Glucose up-take and transport across the syncytiotrophoblast has been shown to be passive and transporter-mediated [2]. There is strong evidence that glucose

TE D

transfer is predominantly mediated by glucose transporter 1 (GLUT1 also known as SLC2A1). GLUT1, a member of the GLUT family, is a transmembrane protein which transports glucose by a sodium-independent facilitated diffusion mechanism [2, 3]. GLUT1 protein is the only relevant

EP

glucose transporter in the term placenta and is asymmetrically expressed in apical and basal

AC C

plasma membranes of the syncytiotrophoblast with substantially higher abundance in MVM than in BM [3]. It is possible, however, that other members of the glucose transporter family may contribute in transplacental glucose transport. Immunohistochemical studies revealed that glucose transporter 3 (GLUT3) is present in MVM but not in BM at all gestational ages [4]. Several factors regulate maternal-fetal glucose transfer: the amount of syncytial glucose transporters, glucose supply, which is determined by glucose serum levels and blood flow, as well as placental glucose metabolism. The transport rate depends on the transporter density on 1

ACCEPTED MANUSCRIPT

the apical and basal plasma membranes of the syncytiotrophoblast as well as the surface area of both membranes [5]. Several studies investigated the regulation of GLUT1 expression defining multiple regulatory mechanisms and a broad range of regulatory factors, such as change in

RI PT

transcription rate [6, 7], mRNA stabilization [8, 9] or protein stabilization [10], and recruitment of presynthesized GLUT1 from the internal stores to the cell surface [11, 12]. GLUT1 expression was shown to be modulated by a wide range of extracellular agents, such as cytokines, growth

SC

factors and steroids [7].

M AN U

However, only few studies have been carried out investigating the regulation of human placental GLUT1 in primary trophoblast cells. Hypoglycemic conditions were shown to lead to mRNA and protein up-regulation [13, 14], whereas hyperglycemic conditions have the opposite effect [13-15]. GLUT1 glucose transporter is one of the key proteins that respond to hypoxia in a

TE D

variety of cells and species. We have previously shown that the placental glucose transport system and the expression of its predominant transporter GLUT1 are modulated upon hypoxia. We have demonstrated that IGF-I increases the basal membrane content of GLUT1 and up-

EP

regulates basal membrane transport of glucose, leading to increased transepithelial glucose transport [15]. Further, GLUT1 and GLUT3 are upregulated upon hypoxia via a HIF-1-mediated

AC C

pathway in trophoblast cells suggesting that the GLUT response to hypoxia in vivo will be modulated not only by low oxygen tension but also by other factors that affects HIF-1 levels [16]. In an in vivo model of chronic placental hypoxia, i.e. pregnancies at high altitude, GLUT1 was found to be 40% lower in BM in high altitude than in low altitude, and BM GLUT1 correlated positively with birth weight at high, but not low altitude. We concluded that hypoxia acts to reduce fetal growth not simply by reducing oxygen delivery, but also by decreasing the density

2

ACCEPTED MANUSCRIPT

of nutrient transporters [17]. Placental hypoxia is associated with pregnancy complications such as fetal growth restriction and preeclampsia, which contributes substantially to perinatal morbidity.

RI PT

Although we and others previously published studies on glucose transport under normal and hypoxic conditions [7, 16-19], the regulation of the placental glucose transport system in

preeclampsia remains largely unclear. Intrauterine growth restriction (IUGR) is a common

SC

feature associated with preeclampsia. Thus we speculated that in preeclampsia the nutrient

M AN U

transporter systems were impaired. We hypothesized that in preeclampsia placental GLUT1 expression is altered. Therefore, we aimed to investigate the effects of preeclampsia on GLUT1

AC C

EP

TE D

protein expression and on glucose transport function across syncytial plasma membranes.

3

ACCEPTED MANUSCRIPT

Methods Placental tissues and membrane vesicles preparation

RI PT

Placentae from normal pregnancies and from pregnancies affected by preeclampsia were collected following elective caesarean sections. Exclusion criteria in normal pregnancies were fetal anomalies, intrauterine growth restriction (IUGR), diabetes, hypertension or preeclampsia,

SC

anemia, infectious disease, drug use or other medical or obstetric complications. The control group consisted of pregnancies without pathologies terminated at term by elective primary

M AN U

caesarean section due to breech presentation or upon patients’ request. Preeclampsia was defined according to the definitions of the International Society for the Study of Hypertension in Pregnancy (ISHHP) which require systolic blood pressure of ≥140mmHg and diastolic blood pressure of ≥90mmHg (repeated measurements with a six-hour interval) and proteinuria of

TE D

>300mg per day arising de novo after the 20th week of gestation and resolving completely by the 6th postpartum week. The study was approved by the ethical committee of the Canton of Bern, Switzerland (Kantonale Ethikkommission Bern, number 178/03), and all participating patients

EP

gave written informed consent. Total membrane isolation was obtained from washed and

AC C

homogenized villous tissue from placenta obtained immediately following elective caesarean section. Homogenates were centrifuged at 1’500 g for 15 minutes at 4°C to discard nuclear and cellular debris. Supernatant was centrifuged at 150’000 g for 60 minutes at 4°C to pellet down total membranes. Washed villous tissues were used to prepare syncytial microvesicles of (apical) microvillous and basal plasma membrane fractions using a previously described method [20]. Purity of the vesicle preparation was assessed by western blotting with specific antibodies

4

ACCEPTED MANUSCRIPT

against endothelial cells (CD31, ab28364, Abcam), fibroblasts (vimentin, v6630, Sigma) and

RI PT

macrophages (F4/80, ab96640, Abcam).

Assays of membrane composition

The protein content of the vesicles was determined by Bradford method [21] using a

bicinchoninic acid (BCA) protein assay kit (Sigma, Pierce) for the colorimetric detection and

SC

quantification of total protein. Bovine serum albumin (BSA) was used as a standard. Membrane

RNA isolation and quantitative PCR

M AN U

fractions were diluted between 10 and 100 times prior to being assayed.

Total RNA was isolated using Trizol reagent. Approximately 50 mg of frozen placental tissue

TE D

were subjected to homogenization in 1 ml Trizol solution for 30 s with a glass-Teflon homogenizer and left for 5 min on ice followed by a centrifugation step at 12000x g for 10 min at 4°C. Phase separation was performed with 0.2 ml chloroform followed by centrifugation at

EP

12000x g for 15 min at 4°C. RNA was precipitated from the water phase by adding 0.5 ml of isopropyl alcohol, incubated for 10 min at room temperature and centrifuged at 12000x g for 10

AC C

min at 4°C. After washing the pellet with 1 ml EtOH (75%) and centrifugation at 7500x g for 5 min at 4°C (twice), the RNA pellet was partially air dried and re-dissolved in DEPC-treated water by passing solution a few times through a pipette tip. Total RNA concentration was measured by Nanodrop 1000 (Thermo Scientific). All RNA samples included in the study had an OD260/280 ratio >1.8.

5

ACCEPTED MANUSCRIPT

Total RNA (2 µg) was reverse transcribed using SuperScript® III First-Strand Synthesis kit (Invitrogen) and subsequent reverse transcription PCR (RT-qPCR) analysis was performed in a volume of 10 µl which contained 0.05 µM primer, 1x fast SYBR green® Gene Expression Master

RI PT

Mix (Applied Biosystems) and 50 ng cDNA. Amplification reactions were performed in triplicates on MicroAmp optical 384-well reaction plate (Applied Biosystems) using default cycling

conditions for 45 cycles: activation of enzyme at 95°C/10 min, denaturation at 95°C/15 sec,

SC

annealing at 60°C/1 min, followed by the melting step. Tyrosine 3-monooxygenase/tryptophan

M AN U

5-monooxygenase activation protein zeta (YWHAZ) was used for normalization. The amount of GLUT1 relative to YWHAZ mRNA was calculated using the ΔCt method.

Immunoblotting

TE D

Total membrane isolation and plasma membrane extracts (3 µg protein) were loaded on 8% acrylamide SDS gels and separated using a running buffer containing 150 mmol/L glycine, 20 mmol/L Tris, 0.1% SDS. Proteins were transferred to the iBlot gel transfer stacks PDVF mini

EP

(Invitrogen) by the iBlot Gel Transfer Devices (Invitrogen). Transfer was confirmed by Ponceau staining (Fluka). Membranes were blocked with 5% milk in Tris-buffered saline containing 0.05%

AC C

Tween 20 (TBST) and then incubated with rabbit polyclonal anti-GLUT1 antibody (1:20,000; 071401 Millipore) in TBST containing 5% BSA. After washing the membranes with TBST, the secondary antibody was applied (horseradish peroxidase-coupled goat anti-rabbit, 1:10,000; NA9340V Amersham). Immunoreactive bands were visualized using the enhanced chemiluminescence reaction (Amersham Biotech), digitized on a ChemiDoc XRS+ Imager (Bio Rad) and quantified by densitometry using Image J 1.49i software (Wayne Rasband, National

6

ACCEPTED MANUSCRIPT

Institutes of Health, USA). The intensity of the signal obtained on the immunoblot was

RI PT

normalized using the immunochemical signal of β-actin as a control.

Radioactive Ligand Up-take Assay

Basal and apical plasma membranes were re-suspended in buffer D (250 mM Sucrose, 10 mM Hepes, pH 7.4) using a Glass/Teflon homogenizer. 75 µg of protein were incubated in a total

SC

volume of 40 μl with 20 μl of transport buffer (0.02 μCi/μl [3H] 3- O-methyl-D-glucose, 1 mM,

M AN U

buffer D) for 1–1800 s on ice. Briefly, we separately placed 20 μl vesicle homogenate and 20 μl up-take buffer into an ice cold 6.5 ml scintillation vials (S207, Simport). Reaction was started by vortexing the scintillation tube and was stopped after 1-1800 seconds. For the timing a digital metronome (Chord) was used. The up-take of methyl-D-glucose ([3H] 3-OG) was stopped by

TE D

adding 1.5 ml ice cold buffer D containing 2% EtOH and 250 μM of the inhibitor phloretin (P7912, Sigma). Plasma membranes were collected by a rapid vacuum filtration on glass fiber filters (MN-3, Macherey-Nagel) with a retention capacity of 600 nM, which were pre-washed

EP

with ice cold buffer D. After four washing steps, with 5 ml ice cold wash buffer (buffer D containing 100 μM phloretin, 2% EtOH and 4 mM 3-OG,) each, the filter-retained radioactivity

AC C

was determined by liquid scintillation counting. To account for non-mediated up-take and nonspecific binding we have performed baseline measurements by incubating the microvesicles with the tracer and in the presence of 250µM phloretin assuming that GLUT1-mediated transport is completely blocked under this condition. The obtained values were subtracted from the “total” up-take measurements for each time point to obtain “net” transporter-mediated transport.

7

ACCEPTED MANUSCRIPT

Statistical Analysis All data are presented as mean ± standard error of the mean (SEM). Statistical tests were

RI PT

carried out using Sigma Plot 11.0 software. When two conditions were compared and data were normally distributed, unpaired Student’s t-test was applied; otherwise Mann-Whitney U test

AC C

EP

TE D

M AN U

SC

was used. A P-value <0.05 was considered to reach significance.

8

ACCEPTED MANUSCRIPT

Results Placental GLUT1 mRNA expression

RI PT

To investigate the glucose transport system in normal and pathological conditions, GLUT1 mRNA levels were analyzed using quantitative RT-qPCR in a cohort of “in-house” placental tissue bank including normal term controls (n=18) and patients with preeclampsia (n=23). The patient’s

SC

maternal demographics and clinical data are shown in Table 1.The subsequent qRT-PCR

measurements demonstrated that GLUT1 was highly expressed in placental tissues. The

M AN U

comparison between the preeclamptic and control group revealed no significant differences in placental mRNA levels in whole placental tissue lysates (1.35±0.21 versus 0.96±0.18, respectively, arbitrary units, P=0.175, Student’s t-test) (Figure 1).

preparations

TE D

Placental GLUT1 protein expression in total membrane from placenta, MVM and BM fraction

To investigate placental GLUT1 protein expression, total membrane isolations from normal term

EP

(n=6) and preeclamptic (n=6) placentae, BM and MVM plasma membrane fractions from normal

AC C

term (n=6) and preeclamptic (n=6) placentae, were analyzed by Western blotting. The patient’s maternal demographics and clinical data are shown in Table 2. Total membrane isolation of preeclamptic placenta showed a significant up-regulation of GLUT1 compared to normal control (1.70±0.21 versus 1.00±0.17, respectively, arbitrary units, P=0.023, Student’s t-test) (Figure 2). In preeclampsia GLUT1 protein expression was decreased by 60 % in MVM vesicles compared to normal term placentae (preeclamptic versus normal, 0.40±0.04 versus 1.00±0.06, arbitrary units, P<0.001, Student’s t-test; Figure 3). No change in GLUT1 protein level was observed in BM 9

ACCEPTED MANUSCRIPT

when both groups were compared (preeclamptic versus normal: 1.38±0.23 versus 1.00±0.12,

Radioactive ligand up-take of GLUT1 in MVM and BM membrane

RI PT

respectively, P=0.097, Student’s t-test; Figure 4).

To measure glucose transport activity a vesicular radioactive up-take assay using an inhibitor stop technique was established. We found glass paper filters with a retention capacity of 600

SC

nM to be superior to nitrocellulose mixed ester membranes. A comparison between both filters

M AN U

and membranes showed equal values but much less nonspecific binding on glass paper (data not shown). We investigated the glucose up-take in MVM vesicles from normal term placentae (n=4) and placentae affected by preeclampsia (n=4) and the up-take in BM vesicles from normal term placentae (n=3) and placentae affected by preeclampsia (n=3). For each sample each time

TE D

point was measured in quadruplicates. The values were standardized to the 30 min time point. In general we observed a linear glucose transport activity until approximately 50 % of total glucose up-take, which was reached within the first six seconds in MVM vesicles from normal

EP

term placentae and within 60 seconds in preeclamptic MVM vesicles. Within the first six seconds glucose transport activity into MVM vesicles prepared from placentae compromised by

AC C

preeclampsia is decreased by 61.78 ± 6.48 % when compared to normal term (Figure 5A). When glucose transport activity in BM vesicles was analyzed, no difference in glucose up-take activity was observed (Figure 5B).

10

ACCEPTED MANUSCRIPT

Discussion Our data demonstrate a decrease in syncytial GLUT1 expression as well as an alteration on the

RI PT

functional level of glucose transport activity across the apical plama membrane of the syncytiotrophoblast in preeclampsia when compared with normal term placentae. The present study shows for the first time that syncytial GLUT1 expression and function is altered in

SC

preeclampsia as compared to normal pregnancies.

There was no change in mRNA level in whole placental tissue when normal term placentae and

M AN U

placentae following pregnancies complicated by preeclampsia were analyzed. Although no differences in GLUT1 mRNA levels were found, alterations of protein expression patterns may be observed on the subcellular levels. Thus total membranes fraction from placental tissues and syncytial (apical and basal) membrane fractions were prepared and analyzed by western

TE D

blotting. GLUT1 expression was significantly up-regulated in the total membrane fraction derived from placentae of patients compromised by preeclampsia. However, when GLUT1 protein expression was investigated at a subcellular levels of the syncytiotrophoblasts, i.e. MVM

EP

and BM, GLUT1 expression showed a 60% down-regulation in MVM in preeclampsia compared

AC C

to normal term placentae, whereas in BM no difference was found. These observations demonstrate that GLUT1 expression is regulated in a specific manner at different subcellular localizations.

Potential modifiers of placental GLUT1 expression might be both maternal are body mass index (BMI) and gestational age. No differences in preconceptional BMI were found between the two groups in the analyzed study population (Table 1 and 2). Furthermore we have investigated glucose serum levels prior to caesarean sections and found no significant differences between 11

ACCEPTED MANUSCRIPT

patients suffering from preeclampsia and control patients (Table 1 and 2). Further correlation analysis did not show a relationship between glucose serum levels and GLUT1 mRNA or protein expression (data not shown). These conditions indicate that the decrease in GLUT1 expression is

RI PT

not due to a metabolic derangement such as obesity with its sequelae. To exclude gestational age-dependent effects on placental nutrient transporter expression it would be ideal to

compare the pregnancies compromised by preeclampsia with age-matched healthy controls.

SC

Unfortunately the later cases do not exist. This is a common problem for all studies analysing

M AN U

pregnancies compromised by preterm delivery since all preterm deliveries bear per se a fetal and/or a maternal pathology. On the other hand, Jansson et al. have demonstrated that during the third trimester placental GLUT1 density does not change in both MVM and BM. Therefore we are confident that gestational age-dependent alteration in GLUT1 protein expression can be

TE D

ruled out in our study population [3].

Compared to normal term placentae, the glucose transport activity into MVM vesicles prepared from placentae affected by preeclampsia was decreased demonstrating that MVM glucose

EP

transport activity is impaired in preeclampsia. The question whether reduced microvillous transport has an impact on transplacental transport in vivo remains to be elucidated given that

AC C

the basal membrane glucose transport step is normally considered rate-limiting [7]. The exact mechanisms of GLUT1 expression regulation in preeclampsia remains to be elucidated. Several cellular mechanisms could cause the observed GLUT1 down-regulation in preeclampsia: Oxidative stress, a common feature in preeclampsia, is an important in vivo activator of placental nuclear factor-kB (NF-kB). In a recent work it has been shown that placental NF-kB is activated nearly 10-fold in preeclampsia [22]. Further oxidative stress leads to

12

ACCEPTED MANUSCRIPT

the opening of mitochondrial permeability transition pores [23] which precedes cell demise [24]. This leads to apoptosis [25] and remodeling of the plasma membrane [26, 27]. We have previously shown that placental GLUT1 expression and glucose transport capacity is up-

RI PT

regulated upon acute hypoxia in BeWo choriocarcinoma cells (BeWo) [17], which is frequently used as a model for the syncytiotrophoblast, as well as in the dual in vitro placenta perfusion model [16]. Interestingly, we have also observed that placental GLUT1 protein expression in vivo

SC

is down-regulated in chronic hypoxia, e.g. in placentae obtained following pregnancies in high-

M AN U

altitude [18]. These findings indicate that in vivo hypoxia is not the only factor regulating syncytial GLUT1expression. A variety of other signals such as extracellular glucose concentrations, insulin, insulin-like growth factors (IGF) and other effectors can override hypoxia-induced up-regulation of the glucose transporter system. It cannot be excluded that in

TE D

preeclampsia signals from the fetoplacental unit trigger other biochemical changes, such as for example altered expressions of IGF and IGF-binding proteins, in order to compensate for the reduced placental GLUT1 expression. Of note in IUGR transplacental glucose transport activity

EP

was shown not to be altered when compared to controls [28]. Our study has some limitations. First, the membrane-derived vesicles system is an in vitro-

AC C

experimental model only mimicking the natural transmembrane and transcellular process, which may underlie other microenvironmental influences in vivo. Second, we do not have information regarding expression and function of GLUT1 in normal ongoing pregnancies in which preeclampsia develops later, precluding an interpretation about the pathophysiological cascade in developing preeclampsia regarding GLUT1. Nevertheless, our results convincingly show that syncytial GLUT1 expression and function is decreased in fully established

13

ACCEPTED MANUSCRIPT

preeclampsia. Third, we cannot rule out the possibility that other members of the GLUT family may compensate for this decrease in GLUT1-mediated glucose transport across (apical) microvillous plasma membranes. GLUT3 was shown by Brown et al. to be present in the

RI PT

syncytial microvillous membrane early in gestation [4]. The same group demonstrated that placental GLUT3 decreases with gestational age supporting the idea that GLUT3 is of greater

SC

importance for placental glucose transport early in gestation.

M AN U

In conclusion, to the best of our knowledge, this is the first study showing that placental GLUT1 is down-regulated in preeclampsia, and that this decrease in GLUT1 expression is paralleled by reduced glucose transport activity across syncytial plasma membranes. Whether these preeclampsia-related changes occur also in vivo and play a role in the development of IUGR, a

AC C

EP

TE D

severe complication in preeclampsia, remains to be elucidated by further studies.

14

ACCEPTED MANUSCRIPT

Acknowledgements We would like to thank Ms. Ruth Sager, the lab technician, the members of the TransCure

RI PT

network for the scientific discussions as well as the midwives and physicians of the Department

AC C

EP

TE D

M AN U

SC

of Obstetrics of the University Hospital Bern for their help in collecting placentae.

15

ACCEPTED MANUSCRIPT

References

AC C

EP

TE D

M AN U

SC

RI PT

[1] G. Desoye, M. Gauster, C. Wadsack, Placental transport in pregnancy pathologies, Am J Clin Nutr 94(6 Suppl) (2011) 1896s-1902s. [2] M. Carstensen, H.P. Leichweiss, G. Molsen, H. Schroder, Evidence for a specific transport of D-hexoses across the human term placenta in vitro, Arch Gynakol 222(3) (1977) 187-96. [3] T. Jansson, M. Wennergren, N.P. Illsley, Glucose transporter protein expression in human placenta throughout gestation and in intrauterine growth retardation, The Journal of clinical endocrinology and metabolism 77(6) (1993) 1554-62. [4] K. Brown, D.S. Heller, S. Zamudio, N.P. Illsley, Glucose transporter 3 (GLUT3) protein expression in human placenta across gestation, Placenta 32(12) (2011) 1041-9. [5] N.P. Illsley, S. Hall, P. Penfold, T.E. Stacey, Diffusional permeability of the human placenta, Contributions to gynecology and obstetrics 13 (1985) 92-7. [6] A. Barthel, S.T. Okino, J. Liao, K. Nakatani, J. Li, J.P. Whitlock, Jr., R.A. Roth, Regulation of GLUT1 gene transcription by the serine/threonine kinase Akt1, The Journal of biological chemistry 274(29) (1999) 20281-6. [7] M.U. Baumann, S. Deborde, N.P. Illsley, Placental glucose transfer and fetal growth, Endocrine 19(1) (2002) 13-22. [8] F. Maher, L.C. Harrison, Stabilization of glucose transporter mRNA by insulin/IGF-1 and glucose deprivation, Biochemical and biophysical research communications 171(1) (1990) 210-5. [9] C. Taha, Z. Liu, J. Jin, H. Al-Hasani, N. Sonenberg, A. Klip, Opposite translational control of GLUT1 and GLUT4 glucose transporter mRNAs in response to insulin. Role of mammalian target of rapamycin, protein kinase b, and phosphatidylinositol 3-kinase in GLUT1 mRNA translation, The Journal of biological chemistry 274(46) (1999) 33085-91. [10] R.J. Boado, W.M. Pardridge, Glucose deprivation causes posttranscriptional enhancement of brain capillary endothelial glucose transporter gene expression via GLUT1 mRNA stabilization, Journal of neurochemistry 60(6) (1993) 2290-6. [11] Z.A. Khayat, A.L. McCall, A. Klip, Unique mechanism of GLUT3 glucose transporter regulation by prolonged energy demand: increased protein half-life, The Biochemical journal 333 ( Pt 3) (1998) 713-8. [12] S. Sasson, N. Kaiser, M. Dan-Goor, R. Oron, S. Koren, E. Wertheimer, K. Unluhizarci, E. Cerasi, Substrate autoregulation of glucose transport: hexose 6-phosphate mediates the cellular distribution of glucose transporters, Diabetologia 40(1) (1997) 30-9. [13] M. Fujishiro, Y. Gotoh, H. Katagiri, H. Sakoda, T. Ogihara, M. Anai, Y. Onishi, H. Ono, M. Funaki, K. Inukai, Y. Fukushima, M. Kikuchi, Y. Oka, T. Asano, MKK6/3 and p38 MAPK pathway activation is not necessary for insulin-induced glucose uptake but regulates glucose transporter expression, The Journal of biological chemistry 276(23) (2001) 19800-6. [14] N.P. Illsley, M.C. Sellers, R.L. Wright, Glycaemic regulation of glucose transporter expression and activity in the human placenta, Placenta 19(7) (1998) 517-24. [15] S. Hauguel-de Mouzon, A. Leturque, E. Alsat, M. Loizeau, D. Evain-Brion, J. Girard, Developmental expression of Glut1 glucose transporter and c-fos genes in human placental cells, Placenta 15(1) (1994) 35-46. [16] M.U. Baumann, H. Schneider, A. Malek, V. Palta, D.V. Surbek, R. Sager, S. Zamudio, N.P. Illsley, Regulation of human trophoblast GLUT1 glucose transporter by insulin-like growth factor I (IGF-I), PloS one 9(8) (2014) e106037. [17] M.U. Baumann, S. Zamudio, N.P. Illsley, Hypoxic upregulation of glucose transporters in BeWo choriocarcinoma cells is mediated by hypoxia-inducible factor-1, American journal of physiology. Cell physiology 293(1) (2007) C477-85. 16

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

[18] S. Zamudio, M.U. Baumann, N.P. Illsley, Effects of chronic hypoxia in vivo on the expression of human placental glucose transporters, Placenta 27(1) (2006) 49-55. [19] X. Huang, M. Luthi, E.C. Ontsouka, S. Kallol, M.U. Baumann, D.V. Surbek, C. Albrecht, Establishment of a confluent monolayer model with human primary trophoblast cells: novel insights into placental glucose transport, Molecular human reproduction (2016) 442-56. [20] N.P. Illsley, Z.Q. Wang, A. Gray, M.C. Sellers, M.M. Jacobs, Simultaneous preparation of paired, syncytial, microvillous and basal membranes from human placenta, Biochimica et biophysica acta 1029(2) (1990) 218-26. [21] M.M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Analytical biochemistry 72 (1976) 248-54. [22] J.E. Vaughan, S.W. Walsh, Activation of NF-kappaB in placentas of women with preeclampsia, Hypertension in pregnancy 31(2) (2012) 243-51. [23] H.D. Haworth RA, The Ca2+-induced membrane transition in mitochondria. II. Nature of the Ca2+ trigger site., Archives of biochemistry and biophysics 195 (1979) 460-7. [24] A.P. Halestrap, S.J. Clarke, S.A. Javadov, Mitochondrial permeability transition pore opening during myocardial reperfusion--a target for cardioprotection, Cardiovascular research 61(3) (2004) 372-85. [25] R.A. Gottlieb, Cell death pathways in acute ischemia/reperfusion injury, Journal of cardiovascular pharmacology and therapeutics 16(3-4) (2011) 233-8. [26] N.J. Faergeman, J. Knudsen, Role of long-chain fatty acyl-CoA esters in the regulation of metabolism and in cell signalling, The Biochemical journal 323 ( Pt 1) (1997) 1-12. [27] D.W. Hilgemann, M. Fine, M.E. Linder, B.C. Jennings, M.J. Lin, Massive endocytosis triggered by surface membrane palmitoylation under mitochondrial control in BHK fibroblasts, eLife 2 (2013) e01293. [28] Challis, D.E., et al., Glucose metabolism is elevated and vascular resistance and maternofetal transfer is normal in perfused placental cotyledons from severely growth-restricted fetuses. Pediatr Res, 47(3) (2000) 309-15.

17

ACCEPTED MANUSCRIPT

Figure Legends Figure 1 Quantification of GLUT1 RNA expression Term placentae (n=18) compared to placentae compromised by preeclampsia (n=23). YWHAZ was used as housekeeper. The amount of GLUT1 relative to YWHAZ mRNA was

RI PT

calculated using the ΔCt method. No significant difference was observed between both groups.

SC

Figure 2

Placental GLUT1 protein expression in total membranes of the placenta. Western blot analysis (A) and quantification (B) of GLUT1 (54 kDa) protein levels in total

M AN U

membrane isolation from term placenta (n=6) and PE placentae (n=6). GLUT 1 protein expression was normalized to β-actin. It is significantly increased in preeclampsia compared to controls (1.78 ± 0.22 to 1.00 ± 0.16, *P < 0.05).

Figure 3

TE D

Placental GLUT1 protein expression in MVM.

Western blot analysis (A) and quantification (B) of GLUT1 (54 kDa) levels in MVM isolated from normal term placentae (n=6) and preeclamptic (PE) placentae (n=6) normalized to βactin. In MVM GLUT1 expression is significantly lower in preeclamptic than in normal term

AC C

Figure 4

EP

placentae (arbitrary numbers, *P<0.05).

Placental GLUT1 protein expression in BM. Western blot analysis (A) and quantification (B) of GLUT1 (54 kDa) levels in BM isolated from normal term placentae (n=6) and preeclamptic placentae (n=6) normalized to β-actin. In BM GLUT1 protein expression is similar between placentae affected by preeclampsia and controls (arbitrary numbers, NS).

Figure 5 Radioactive ligand up-take of GLUT1 in MVM and BM vesicles.

1

ACCEPTED MANUSCRIPT (A) Radioactive ligand up-take of MVM vesicles from four term placentae (filled circles) and four preeclamptic placentae (empty circles). For each patients the experiment was repeated 4 times and each time point was performed in triplicates. Data are shown as mean ± sem of the analyzed patients. The data are normalized to the 1800s time point. The linear up-take takes place within the first six seconds. (**P<0.01; ***P<0.001) (B) Radioactive ligand up-

RI PT

take of BM vesicles from three term placentae (full circles) and three preeclamptic placentae (empty circles). There is no significant difference in the up-take of BM vesicles between

Table Legends Table 1

M AN U

Clinical parameters of patients for RNA expression

SC

normal term controls and preeclampsia.

BMI, preconceptional body mass index; NS, not significant; n.d., not determined *

including the actual pregnancy; **prior to caesarean section; Mann-Whitney U test; ++Fisher’s exact test

Table 2

TE D

+

Clinical parameters of patients for protein expression BMI, preconceptional body mass index; NS, not significant; n.d., not determined

EP

including the actual pregnancy; **prior to caesarean section; +Mann-Whitney U test

AC C

*

2

ACCEPTED MANUSCRIPT Table 1. Patients characteristics (GLUT1 mRNA measurements) p-value

Controls (n=18)

PE (n=23)

Age, y

32.4 ± 1.1

32.1 ± 1.3

0.91, NS

BMI, kg/m2

22.2 ± 1.0

23.8 ± 0.8

0.06, NS

Gravidity

2.6 ± 0.3

1.8 ± 0.2

0.99+, NS

Parity*

2.1 ± 0.2

1.4 ± 0.1

< 0.05+

Gestational age, weeks

38.8 ± 0.2

31.4 ± 0.8

< 0.001+

549 ± 27

Birth weight, g Systolic blood pressure, mmHg** Diastolic blood pressure, mmHg**

AC C

EP

Glucose serum level, mmol/L**

0.76++, NS

299 ± 40

< 0.001

3236 ± 71

1483 ± 148

< 0.001

114 ± 3

167 ± 4

< 0.001

67 ± 3

102 ± 2

< 0.001

< 0.15

1.89 ± 0.46

n.d.

5.01 ± 0.22

0.99+, NS

TE D

Proteinuria, g/24h**

9/14

M AN U

Placental weight, g

8/10

SC

Sex of the infant, female/male

RI PT

Mean ± sem

4.87 ± 0.18

ACCEPTED MANUSCRIPT Table 2. Patients characteristics (GLUT1 protein measurements) p-value

Controls (n=6)

PE (n=6)

Age, y

31.5 ± 1.9

33.7 ± 3.3

0.58, NS

BMI, kg/m2

22.0 ± 1.4

24.8 ± 1.3

0.19, NS

Gravidity

2.2 ± 0.2

2.2 ± 0.4

0.81+, NS

Parity*

1.8 ± 0.2

1.2 ± 0.2

0.07+, NS

Gestational age, weeks

38.6 ± 0.5

31.8 ± 0.9

< 0.001

530 ± 29

Birth weight, g Systolic blood pressure, mmHg** Diastolic blood pressure, mmHg**

AC C

EP

Glucose serum level, mmol/L**

n.d.

315 ± 76

< 0.05

3167 ± 166

1573 ± 209

< 0.001

120 ± 5

161 ± 3

< 0.001

75 ± 3

104 ± 4

< 0.001

< 0.15

2.37 ± 0.81

n.d.

5.33 ± 0.41

0.36, NS

TE D

Proteinuria, g/24h**

2/4

M AN U

Placental weight, g

2/4

SC

Sex of the infant, female/male

RI PT

Mean ± sem

4.80 ± 0.32

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Highlights: In preeclampsia placental GLUT1 protein expression is down-regulated

•

GLUT1-mediated glucose transport activity is decreased in preeclampsia

•

Preeclampsia has an impact on maternofetal glucose supply

AC C

EP

TE D

M AN U

SC

RI PT

•