| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Experimental Studies |
Centre de Recherche sur l’Endocrinologie, Moléculaire et le Développement, CNRS, 92 190 Meudon-Bellevue; Université Pierre et Marie Curie, Physiopathologie du Développement, Groupe Interactions Cellulaires (J.C.C., A.K.), 75252 Paris Cedex 05, France; and the Department of Obstetrics and Gynecology, University of Bern (A.M.), Bern, Switzerland
Address all correspondence and requests for reprints to: Dr. S. Hauguel-de Mouzon, Centre de Recherche sur l’Endocrinologie, Moléculaire et le Développement, CNRS, 9 rue Jules Hetzel, 92 190 Meudon-Bellevue, France. E-mail: shm{at}infobiogen.fr
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Glucose is transferred across the placenta by facilitative diffusion through specific glucose transporter proteins (reviewed in 8 . GLUT1 is the main isoform expressed in mammalian placentas (9, 10). It is found at high levels in human placenta throughout pregnancy (11, 12, 13). By contrast, the localization and pattern of expression of GLUT3 have not been clearly defined. GLUT3 messenger ribonucleic acid (mRNA) has been found at high levels in both rodent (14, 15) and human (16) placentas. We have recently identified GLUT3 protein in the labyrinthine zone of the rat placenta (14). In the human placenta, GLUT3 protein was either not detected (5, 12, 17) or was present at a low level (18), leading to the hypothesis that GLUT3 mRNA would be present but not translated into a readily detectable protein.
The present study was undertaken to characterize the expression of GLUT3 glucose transporter in the human placenta. GLUT3 mRNA and protein were detected by in situ hybridization as well as by immunodetection studies and compared to GLUT1 levels in various human placental cell types.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Freshly delivered human term placentas were aseptically collected after normal pregnancies (38–40 weeks postmenstruation) after cesarean section or vaginal delivery. After removing the chorionic and basal plates, tissue samples (5 x 5 mm) were excised within an intact placental lobule. Samples for immunohistochemistry and in situ hybridization were immediately fixed and processed as described in the relevant sections. Samples for immunoblotting and Northern blotting analysis were frozen in liquid nitrogen and stored at -80 C. The endothelial cells were prepared by a two-step collagenase-dispase and trypsin-deoxyribonuclease digestion of size-fractionated villi as previously described (19). Eight days after plating, the confluent endothelial cells were dissociated with trypsin and resuspended in DMEM for subsequent primary cultures. Protein and RNA analyses were performed on cells obtained after the second passage. Cytotrophoblast cells were prepared from term human placenta by the technique of Kliman et al. (20). After Percoll gradient, purified cytotrophoblasts (6 x 106 cells) were used to prepare whole lysate, and the remaining cells were plated at a density of 2 x 106/35-mm well in DMEM supplemented with 20% heat-inactivated FCS and antibiotics. Syncytiotrophoblast cells were obtained after 48-h culture and further processed for immunoblot analysis. The characteristics of these cultured cells have been described previously (21).
Antibodies and complementary DNA (cDNA) probes
The antibodies used were rabbit polyclonal antisera directed
against the intracellular C-terminus of rat brain GLUT1 (amino acids
479–492) purchased from East Acres Biologicals (Southbridge, MA) and
to the C-terminal 14-amino acid sequence of human brain GLUT3 (amino
acids 481–493), provided by Dr. G. W. Gould (University of Glasgow,
Glasgow, Scotland). Antiactin antibody raised against the C-terminal 11
residues of actin, epitope conserved in all actin isoforms was
purchased from Sigma Chemical Co. (St. Louis, MO). Human GLUT3 cDNA
probe was prepared from a 2.6-kilobase (kb)
SalI/BglI fragment subcloned into
pBSK+, and human GLUT1 cDNA was prepared from a 1.75-kb
SalI fragment subcloned into pGEM4Z. Vectors were provided
by Prof. G. I. Bell (University of Chicago, Chicago, IL). Rat ß-actin
cDNA probe was obtained as a 3.75-kb fragment subcloned into pBR322.
Probes were labeled with [
-32P]deoxy-ATP using the
multiprime labeling system (Amersham International, Aylesbury, UK). The
peptide corresponding to the C-terminal 13 amino acids of the human
GLUT3 sequence MNSIEPAKETTTNV (16) was purchased from Neosystem
(Strasbourg, France). Antisense synthetic 24-mer oligonucleotide
specific for rat 18S ribosomal RNA was end labeled with
[
-32P]ATP (Amersham International) using T4
polynucleotide kinase.
In situ hybridization
GLUT1 and GLUT3 cDNA probes were subcloned into PGEM-3Z vectors. Digoxigenin (DIG)-deoxy-UTP-labeled riboprobes were obtained using T7 and SP6 polymerases with an RNA labeling kit (Boehringer Mannheim, Germany). Sense probes were generated using restriction enzymes/RNA polymerases HindIII/SP6 for GLUT1 and HindIII/T7 for GLUT3, and antisense probes were generated using SacI/T7 for GLUT1 and SmaI/SP6 for GLUT3. Placental tissue samples (5 x 5 mm) were fixed by immersion in 4% paraformaldehyde in phosphate-buffered saline (PBS) for 3 h at 4 C, then soaked in increasing concentrations of saccharose (12%, 15%, and 18%). After embedding in Tissue-Tek (Miles, Naperville, IL) and freezing at -20 C, 5-µm cryostat sections were mounted on aminopropyltriethoxysilane (Sigma Chemical Co.)-coated slides and kept at -80 C with dessicant. After thawing, the mounted sections were rehydrated and postfixed with 4% paraformaldehyde for 20 min, then treated for 10 min with 0.25% acetic anydride in 0.1 mol/L triethanolamine. Prehybridization was carried out for 2 h at 50 C in 50% formamide, 5 x SSC (standard saline citrate), 5 x Denhart’s solution, 50 µg/mL yeast transfer RNA, 250 µg/mL salmon sperm DNA, 4 mmol/L ethylenediamine tetraacetate, and 2.5% dextran sulfate. Hybridization was carried out overnight at 50 C in a moist chamber using prehybridization buffer with sense or antisense RNA probes at concentrations of 1–2 ng/µL for GLUT1 and 5–10 ng/µL for GLUT3. Slides were washed with 50% formamide for 30 min in 2 x SSC. They were then treated twice for 30 min at 37 C with ribonuclease A (20 mg/mL) in 10 mmol/L Tris buffer, pH 8, containing 0.5 mol/L NaCl and 10 mmol/L ethylenediamine tetraacetate. The final wash was performed at room temperature in 0.2 and 0.1 x SSC. Slides were incubated for 30 min in 150 mmol/L NaCl and 100 mmol/L Tris-HCl, pH 7.5, containing 0.5% blocking agent (Boehringer Mannheim) and incubated overnight at room temperature with an alkaline phosphatase anti-DIG antibody diluted to 1:500 in the same buffer. After two 15-min washes in 150 mmol/L NaCl and 100 mmol/L Tris-HCl, pH 7.5, and a 3-min wash in 50 mmol/L MgCl2, 100 mmol/L NaCl and 100 mmol/L Tris-HCl, pH 9.5, sections were incubated for 2–3 h at 37 C with the substrates of alkaline phosphatase, 75 µg/mL nitro blue tetrazolium, and 50 µg/mL 5-bromo-4-chloro-3-indolyl-phosphate in the last wash containing 100 mmol/L levamisol (Boehringer). The reaction was stopped by rinsing with water, and slides were mounted in glycerol-gelatin.
Immunohistochemistry
Frozen cryostat sections (5 µm) of paraformaldehyde-fixed placental tissue were used for indirect immunofluorescence with anti-GLUT3 and anti-GLUT1 antibodies diluted 1:250 and 1:5000, respectively. Sections were air-dried, fixed in 2.5% paraformaldehyde for 20 min, washed in PBS, and incubated with 1% normal goat serum for 1 h to reduce nonspecific binding. After an overnight incubation at 4 C with 20 µL primary antibodies diluted in PBS containing 1% normal goat serum, sections were washed for 10 min with PBS. Incubation with the secondary antibody, fluorescein isothiocyanate-conjugated goat antirabbit IgG (Pasteur-Sanofi, Marnes-la-Coquette, France) at a 1:500 dilution was carried out in the dark for 1 h at 23 C. Sections were then washed for 30 min in PBS and mounted in mowiol (Sigma). Some sections were treated with 500 nmol/L tetramethylrhodamine B isothiocyanate-labeled phalloidin (Sigma) for 10 min, followed by three 10 min-PBS washes before incubation with primary antibody to visualize actin filaments. Controls were performed by omitting primary antibodies. For competition studies, the anti-GLUT3 antibody was preadsorbed with a peptide corresponding to the 13 C-terminal amino acids of human GLUT3, and all subsequent steps were carried out as described above.
RNA analysis
Total RNA was extracted from cultured cells according to Chomczynski’s procedure (22) and from whole placenta according to Chirgwin’s procedure (23). The RNA concentration was determined by measuring absorbance at 260 nm, and Northern blotting was performed as previously described (14). An antisense synthetic oligonucleotide (24-mer) specific for the 18S ribosomal subunit (24) was used to normalize for the relative differences in the amounts of RNA transferred to the filters. Quantitation of the signals was performed by scanning densitometry of the autoradiograms using a scanning analysis software (Scan Analysis, Biosoft, Cambridge, UK).
Membrane preparations
Placental membranes derived from four tissue fragments (1.5 g total wet weight) were obtained after differential centrifugations as described previously (14). Endothelial and syncytiotrophoblast cells were rinsed twice with ice-cold PBS, then directly scraped in 2 x electrophoresis sample buffer (Laemmli) containing 100 mmol/L dithiothreitol and sonicated for three 30-s periods at half-maximal speed (Sonics and Materials, Danbury, CT). Isolated cytotrophoblast cell pellets were lysed in 2 x electrophoresis sample buffer. Protein contents were measured using the standard Bio-Rad (Richmond, CA) and Pierce-BCA (Pierce Chemical Co., Rockford, IL) assays.
Immunoblotting
Samples containing equal amounts of membrane proteins solubilized in 2 x Laemmli buffer were applied to 10% SDS-polyacrylamide gels without heating and electrophoresed according to Laemmli (25). After electrophoresis, proteins were transferred electrically as described previously (14). Primary antibodies were used at 1:200 dilution for anti-GLUT3 and at 1:500 for anti-GLUT1 and anti-actin. The secondary antibody was a horseradish peroxidase-conjugated antirabbit IgG diluted 1:5000 (Amersham International). The immunoreactive signals were visualized by addition of ECL chemiluminescence reagent (Amersham International) according to the manufacturer’s instructions.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Northern blot analysis.
Concentrations of GLUT3 mRNA were
compared to GLUT1 and nonmuscle ß-actin in whole placental tissue and
in endothelial cells from intraplacental microvessels (Fig. 1
). Two GLUT3 transcripts of 4.1 and 2.7 kb were
detected in whole placenta and endothelial cells as previously reported
for other human tissues, but the smaller 2.7-kb transcript that is
known to be generated by alternative polyadenylation within the
3'-untranslated region (16) was predominantly expressed in endothelial
cells. Similar levels of the 4.1-kb GLUT3 mRNA were found in placenta
and endothelial cells (37 ± 8 vs. 44 ± 2
arbitrary densitometry units). GLUT1 mRNA, appearing as a single 2.8-kb
transcript, was strongly expressed in placenta and to a lesser extent
in endothelial cells from microvessels (234 ± 11 vs.
20 ± 1 arbitrary densitometry units). By contrast, the
concentration of nonmuscle ß-actin mRNA was 5-fold higher in
endothelial cells than in whole placenta (Fig. 1).
|
). Intense labeling was observed in the trophoblast
cells surrounding some terminal villi, whereas adjacent villi of the
same size remained unlabeled (Fig. 2A). A discontinuous distribution of
GLUT3 mRNA was observed in the trophoblast cell layer of intermediate
villi (Fig. 2B). Stromal cells of intermediate and terminal villi were
weakly stained with the GLUT3 probe, but cells from the vascular media
did not show staining (Fig. 2, A–C). Positive signal was also detected
in vascular endothelium of intermediate villi (Fig. 2C). No GLUT3
labeling was observed in placental sections incubated with a specific
sense probe (Fig. 2D).
|
, E–H). However, the signal appeared much stronger and
evenly distributed in trophoblast cells of all types of villi (Fig. 2, E–H). Endothelium of intraplacental vessels were clearly stained, but
at a lower intensity than the trophoblast cell layer (Fig. 2, F–H). In
addition, stromal cells regardless of the villous type and smooth
muscle cells in the vascular media were also lightly stained with the
GLUT1 RNA probe (Fig. 2, G and H). Immunological characterization of GLUT3 protein in human term placenta
A single prominent 49-kDa protein was detected in membrane
fractions from human term placenta using anti-GLUT3 antibody (Fig. 3, lanes 1 and 2), and no cross-reactivity was observed
with purified actin from rabbit muscle (Fig. 3, lane 3). By contrast,
an antiactin antibody detecting all actin isoforms recognized a
42.7-kDa band in both placental membranes (Fig. 3, lanes 4 and 5) and
in a purified actin sample (Fig. 3, lane 6). A protein of higher
molecular mass (
90 kDa) was also identified in the placenta with the
antiactin antibody (lanes 4 and 5). These results indicate that GLUT3
protein is present in term placenta and cannot be mistaken for actin,
which migrates at a lower molecular mass.
|
Immunocytochemical analysis.
GLUT3 immunoreactivity (green
staining) was found almost exclusively in the endothelium of fetal
vessels of intermediate villi (Fig. 4B). The fluorescent
staining was efficiently competed by preincubation with human GLUT3
peptide (Fig. 4C). Endothelial localization was confirmed by
immunostaining of an adjacent section with anti-CD31 and anti-von
Willebrand factor antibodies (not shown), two specific markers of
endothelial cells (19). Less intense staining intensity was noticed in
stromal cells of intermediate villi, whereas yellow fluorescence was
observed in the trophoblast layer regardless of the villous type (Fig. 4, A and B). The yellow staining detected in the trophoblast appeared
as a reaction due to binding of the secondary antibody, as control
sections showed the same coloration (Fig. 4C). By contrast, strong
GLUT1 immunoreactivity was found in the trophoblast layer of terminal
and intermediate villi (Fig. 5, A and B). Endothelium of
fetal vessels as well as fetal erythrocytes were also immunopositive
for GLUT1 (Fig. 5B). Phalloidin, which visualizes actin filaments,
counterstained both vascular media and perivascular myofibroblasts
(Fig. 5C).
|
|
, lane 4), although to a lesser extent
than membranes derived from whole placenta (Fig. 6, lane 1). GLUT3
protein had a higher apparent molecular mass in endothelial cells than
in placental membranes (52 vs. 49 kDa), suggesting that
GLUT3 is differently glycosylated in the two fractions. As expected,
GLUT3 protein was not detected in isolated cytotrophoblast cells and
differentiated syncytiotrophoblast cells (Fig. 6, lanes 2 and 3).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
GLUT3 mRNA exhibited a trophoblastic localization regardless of the villous type and was also present at significant levels in the endothelium of vessels from intermediate villi. This is at contrast with the intense homogeneous labeling detected in all cell types of intermediate villi with antisense as well as sense RNA probes (28). The heterogeneity of GLUT3 mRNA expression that we observed among villi of the same size cannot be related to any obvious structural or functional specificity at present. Because a heterogeneous pattern of staining was observed in sections derived from four placentas, we believe that in normal term pregnancies, GLUT3 is present in certain villi but not in others. From a metabolic standpoint, this result could reflect regional responses to different stimuli, as GLUT3 mRNA expression has been shown to be regulated under various metabolic and hormonal conditions including hyperglycemia, hypoxia, and cellular stress (14, 29, 30). A developmental regulation of GLUT3, decreasing as a function of gestational age (13) could also contribute to some extent to the heterogeneous mRNA distribution in term villi.
The discrepancy between GLUT3 mRNA and protein localization observed in the human placenta has been previously reported in other human tissues (5, 16, 27). The reason for these differences is not yet elucidated, but these findings indicate that the mRNA level cannot be taken as an index of the protein concentration. It could explain that GLUT3 protein was not detected in human placenta using membrane fractions of syncytial origin (5, 12, 17) that express GLUT3 mRNA but do not contain GLUT3 protein. Alternatively, this difference might result from an instability of GLUT3 transcripts or a block of the translation of GLUT3 mRNA as previously speculated (5) and could also reflect the heterogeneous mRNA distribution observed by in situ hybridization.
The present study demonstrates that in the human placenta, GLUT3
protein is predominantly localized around fetal vessels. By contrast,
GLUT1 is found in cells surrounding both maternal and fetal blood.
Based on the higher affinity for glucose of human GLUT3
(Km = 1 mmol/L) compared to GLUT1
(Km = 5 mmol/L), one may speculate that the
two isoforms will serve specific functions within the placenta. The
regional distribution of GLUT3 at the fetal interface may ensure
constant delivery of glucose to the fetal circulation even under
conditions of low maternal blood glucose concentrations. Such a
vectorial flow of glucose has been suggested to occur in neurons where
GLUT3 is predominantly localized (31). Besides placenta and brain,
GLUT3 protein is also strongly expressed in testis (7) and may provide
a high affinity transport system in tissues that should be protected
from fuel deprivation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received January 16, 1997.
Revised April 1, 1997.
Accepted May 6, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. R. Araujo, P. Goncalves, and F. Martel Modulation of Glucose Uptake in a Human Choriocarcinoma Cell Line (BeWo) by Dietary Bioactive Compounds and Drugs of Abuse J. Biochem., August 1, 2008; 144(2): 177 - 186. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Ericsson, B. Hamark, T.L. Powell, and T. Jansson Glucose transporter isoform 4 is expressed in the syncytiotrophoblast of first trimester human placenta Hum. Reprod., February 1, 2005; 20(2): 521 - 530. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. von Wolff, S. Ursel, U. Hahn, R. Steldinger, and T. Strowitzki Glucose Transporter Proteins (GLUT) in Human Endometrium: Expression, Regulation, and Function throughout the Menstrual Cycle and in Early Pregnancy J. Clin. Endocrinol. Metab., August 1, 2003; 88(8): 3885 - 3892. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. E. Mann, D. L. Yudilevich, and L. Sobrevia Regulation of Amino Acid and Glucose Transporters in Endothelial and Smooth Muscle Cells Physiol Rev, January 1, 2003; 83(1): 183 - 252. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Hahn, A. Blaschitz, E.T. Korgun, I. Lang, G. Desoye, G. Skofitsch, and G. Dohr From maternal glucose to fetal glycogen: expression of key regulators in the human placenta Mol. Hum. Reprod., December 1, 2001; 7(12): 1173 - 1178. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. G. Das, J. He, R. A. Ehrhardt, W. W. Hay Jr., and S. U. Devaskar Time-dependent physiological regulation of ovine placental GLUT-3 glucose transporter protein Am J Physiol Regulatory Integrative Comp Physiol, December 1, 2000; 279(6): R2252 - R2261. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Kamei, O. Tsutsumi, A. Yamakawa, Y. Oka, Y. Taketani, and J. Imaki Maternal Epidermal Growth Factor Deficiency Causes Fetal Hypoglycemia and Intrauterine Growth Retardation in Mice: Possible Involvement of Placental Glucose Transporter GLUT3 Expression Endocrinology, September 1, 1999; 140(9): 4236 - 4243. [Abstract] [Full Text]
|
|
|
|
|
|
T. Hahn, S. Barth, R. Graf, M. Engelmann, D. Beslagic, J. M. H. M. Reul, F. Holsboer, G. Dohr, and G. Desoye Placental Glucose Transporter Expression Is Regulated by Glucocorticoids J. Clin. Endocrinol. Metab., April 1, 1999; 84(4): 1445 - 1452. [Abstract] [Full Text]
|
|
|
|
|
|
K. Gaither, A. N. Quraishi, and N. P. Illsley Diabetes Alters the Expression and Activity of the Human Placental GLUT1 Glucose Transporter J. Clin. Endocrinol. Metab., February 1, 1999; 84(2): 695 - 701. [Abstract] [Full Text]
|
|
|
|
|
|
A. Y. Xing, J. C. Challier, J. Lepercq, M. Caüzac, M. J. Charron, J. Girard, and S. Hauguel-de Mouzon Unexpected Expression of Glucose Transporter 4 in Villous Stromal Cells of Human Placenta J. Clin. Endocrinol. Metab., November 1, 1998; 83(11): 4097 - 4101. [Abstract] [Full Text]
|
|
|
|
 |
|
 T. Hahn, S. Barth, U. Weiss, W. Mosgoeller, and G. Desoye Sustained hyperglycemia in vitro down-regulates the GLUT1 glucose transport system of cultured human term placental trophoblast: a mechanism to protect fetal development? FASEB J, September 1, 1998; 12(12): 1221 - 1231. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
