| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Biochemistry, University of Wisconsin, Madison, Wisconsin 53706-1569
Address all correspondence and requests for reprints to: Dr. Jack Gorski, Department of Biochemistry, University of Wisconsin, 433 Babcock Drive, Madison, Wisconsin 53706. E-mail: gorski{at}biochem.wisc.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recent experiments have partially characterized the early (1–2 h) estradiol (E2) effects on glucose transport, and the results indicate that E2 is an important regulator of Glut1. A sc injection of 100 µg/kg E2 2 h before middle cerebral artery occlusion has been shown to reduce subsequent focal ischemic damage in rats (9). E2-treated rats also show higher Glut1 protein levels in the penumbral region of the ischemic lesions than control animals. These data represent more than a correlation, because rats transfected with Glut1 also show increased neuronal survival after middle cerebral arterial occlusion (10). Treatment with E2 also reduces the glucose transport inhibition caused by chemical insults related to Alzheimer’s disease; a 1-h E2 pretreatment of synaptosomes prepared from rat cerebral hemispheres results in a reduced level of glucose transport impairment caused by FeSO4 or amyloid B peptide (11). It is important to note that the prepared synaptosomes lack nuclei, suggesting that E2 may effect glucose transport without affecting glucose transporter transcription.
One of the most carefully studied estrogen-responsive model systems is the rat uterus (12). E2 treatment causes an entire series of growth-related responses, including an increase in the uptake rate of amino acids, nucleic acids, and glucose. The increase in the glucose transport rate after a single injection of microgram quantities of E2 was shown to be inhibited by the protein synthesis inhibitor cycloheximide (7). The degree of stimulation was similar to that of peptide hormones epidermal growth factor and fibroblast growth factor, and the kinetics [an increased maximum velocity (Vmax) and constant Km] suggest that E2 regulation of glucose transport, like the second response observed for peptide hormones, involves increased levels of glucose transporter protein synthesis. These data are consistent with a model in which stimulation of glucose transport by E2 in the immature rat uterus is the result of increased glucose transporter transcription and/or increased messenger RNA (mRNA) half-life, resulting in the associated accumulation of new glucose transporter protein.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RT-quantitative competitive-PCR (RT-qc-PCR) control
synthesis
Primers for RT-qc-PCR were chosen that amplified both 124- and
137-bp fragments of rat Glut1 and Glut4, respectively. Construction of
the PCR control fragment was performed by a modification of the method
of Siebert et al. (13). Briefly, oligonucleotides
complementary to primers used to amplify Glut1 and Glut4 were
synthesized with 3'-overhangs. Each overhang contains a half-site of a
different restriction enzyme. These complementary oligonucleotides were
annealed with the PCR primers and were then ligated onto a DNA
restriction enzyme fragment possessing ends complementary to those
restriction half-sites encoded by oligonucleotide 3'-overhangs.
Products of the ligation reaction were amplified using PCR; the
resulting fragment was gel purified and inserted into pGEM-3Zf+
(Promega Corp., Madison, WI). Transcripts of the inserts
were made, and the reaction mixture was subsequently treated with
deoxyribonuclease I (Promega Corp.) until no amplification
could be observed without first reverse transcribing the samples.
Primers to Glut1 and Glut4 contained a single degenerate base, so the
control RNA was isolated with both forms of primer, and equal amounts
of both were included in the RT-qc-PCR reaction.
RT-qc-PCR
Large preparations of RT reaction mixture were made, aliquoted
into tubes containing serial dilutions of control RNA, and incubated at
42 C for 1 h. Primers chosen for the PCR reaction were from Sivits
et al. (14). RT reactions were amplified by PCR for 27
cycles (95, 55, and 72 C) with Taq DNA polymerase
(Promega Corp.). Mineral oil overlay was removed, and an
aliquot was loaded onto a 3.5% high resolution agarose gel (Amresco,
Solon, OH; and FMC BioProducts, Rockland, ME) containing 0.5 µg/ml of
ethidium bromide, electrophoresed at approximately 9 V/cm, visualized
by UV light, and exposed to Polaroid film (Polaroid Corp.,
Cambridge, MA). Quantification was performed with NIH Image 1.6
software.
Western blots
Equal amounts of proteins were separated by SDS-PAGE and
transferred to nitrocellulose membrane. Antibodies to glucose
transporters were provided by the following: antibodies against Glut1,
Glut2, Glut3, and Glut4 were from Charles River Pharmservices (Southbridge, MA), and antibody against
Glut5 was a gift from Charles F. Burant (University of Chicago,
Chicago, IL). All Glut antibodies were made either against rat protein
(Glut1, -2, -4, and -5) or against mouse protein (Glut3), and this
antibody was tested and found to strongly cross-react with rat Glut3
protein. Antibody against
Na+,K+-adenosine triphosphatase
(Na+,K+-ATPase) was obtained from Upstate Biotechnology, Inc. (Lake Placid, NY). Nitrocellulose membranes
were blocked with 50% horse serum in PBS containing 0.5% (vol/vol)
Tween-20, and all antibody incubations were performed in 10% horse
serum in PBS containing 0.5% (vol/vol) Tween-20. Antibodies were
detected by chemiluminescence (NEN-DuPont, Boston, MA).
2-Deoxyglucose uptake assay
Rats were treated at various intervals with a single 1-µg ip
injection of E2 and were killed together (except for the 15
min point, which was performed separately after the 8 h point).
Two rats were pooled at each time point. Rats were killed, and their
uteri were removed and placed into HBSS (Sigma Chemical Co.) with 0.1 mM [3H]2-deoxyglucose
([3H]2dog; ICN Pharmaceuticals, Inc., Costa
Mesa, CA) for 30 min with constant shaking. After the incubation, uteri
were washed twice in HBSS without glucose, and then immediately frozen
on dry ice. Glucose uptake was measured by determining the amount of
[3H]2dog-6-phosphate ([3H]2dog-6-P) that
accumulated in each sample. [3H]2dog-6-P was separated
from [3H]2dog by anion exchange chromatography. Briefly,
frozen uteri were homogenized in 6% perchloric acid, and solids were
removed by centrifugation. The supernatant was titrated to pH 11 with 3
M KOH, and the resulting potassium perchlorate precipitate
was removed by centrifugation. The supernatant was placed once over a
column of Ag 1-X2, 200–400 mesh anion exchange resin (Bio-Rad Laboratories, Inc., Hercules, CA), and [3H]2dog
was eluted with five washes with 1–3 ml H2O.
[3H]2dog-6-P then was eluted with 1 M HCl.
[3H]2dog-6-P was measured with scintillation
counting.
Subcellular fractionation
Nineteen-day-old female Sprague Dawley rats (Harlan Sprague Dawley, Inc.) were injected with ethanolic saline or
E2 as previously described, or they were injected sc with
200 µl 150 mM sodium pyrophosphate plus 10 U human
recombinant insulin Novolin R (Novo Nordisk, Bagsvaerd, Denmark). Cell
membrane preparations were performed by a modification of the method
described by Hirshman et al. (15). Briefly, samples were
homogenized with a Tissumizer (model SDT-100N, Tekmar Co., Cincinnati,
OH) at 20,000 rpm for two to three bursts of 5 sec each in a buffer
containing 100 mM Tris (pH 7.5), 20 mM EDTA (pH
8.0), and 255 mM sucrose (pH 7.6). The homogenate was then
centrifuged at 1,000 x g for 5 min, and the resulting
supernatant was centrifuged again at 48,000 x g for 20
min. The pellet from this centrifugation was used for the preparation
of the membrane fraction, which was enriched in the membrane marker
Na+,K+-ATPase, and the supernatant was used for
the preparation of the microsomal fraction.
The membrane fraction was prepared by resuspending the pellet in 20 mM HEPES and 250 mM sucrose, pH 7.4 (buffer A). An equal volume of a solution containing 600 mM KCl and 50 mM sodium pyrophosphate was added, and the mixture was vortexed, incubated for 30 min on ice, and then centrifuged for 1 h at 227,000 x g. The resulting pellet was resuspended in buffer A and centrifuged for 1 h at 135,000 x g over a 36% sucrose cushion in buffer A. The resulting interface and all of the buffer above it were collected, diluted in an equal amount of buffer A, and centrifuged for 1 h at 227,000 x g. The resulting pellet was used as the membrane fraction.
The microsomal fraction was prepared from the supernatant of the initial 48,000 x g centrifugation. The supernatant was centrifuged for 1 h at 227,000 x g. The resulting pellet was used as the microsome fraction. Preparations were kept on ice during the protocol, and all centrifugations were performed at 4 C. Western blots were performed on the protein pellet from the sucrose cushion and initial homogenate for both Na+,K+-ATPase and glucose transporters to confirm that membrane markers were being selectively isolated and concentrated in the plasma membrane, and that no significant population of either glucose transporter was contained in any fraction except the plasma membrane and microsome fractions.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). The rate of glucose
uptake into uterine cells is determined by measuring how much
[3H]2dog is converted to [3H]2dog-6-P by
hexokinase, the first step in glycolysis. Increased levels of
[3H]2dog-6-P were detected when rats were killed 30 min
to 1 h after E2 injection, and this increased level of
[3H]2dog-6-P reached a plateau when the time between
E2 injection and death was between 4–8 h. Although some of
the experimental conditions differed from earlier studies (7, 16)
(i.e. the age of rats, the amount of E2
injected, the uterine incubation medium and temperature,
etc.), these observations are in general agreement with
these previous data in that a single injection of microgram quantities
of E2 causes a stimulation of glucose transport that is
detectable within 30 min and that peaks within several hours.
|
). Of the tissues used as controls for
glucose transporter expression, Glut1 expression was highest in brain;
however, Glut1 was found to be expressed in most tissues. Glut2 was
found in kidney and jejunum, and Glut3 was found predominantly in
brain, although it was also detected in skeletal muscle. Glut4
expression was highest in skeletal muscle, and Glut5 expression was
highest in kidney, although it also was expressed in jejunum and brain
(17, 18). Analysis of whole protein preparations in these tissues
revealed the appropriate expression patterns in the various glucose
transporter target tissues.
|
). Glut2 mRNA was not detected in
uterine complementary DNA, but Glut2 was detected in both liver and
kidney complementary DNA as expected. These data support the conclusion
that Glut2 is not expressed in the immature rat uterus.
|
Effect of E2 on Glut1 and Glut4 protein and
mRNA levels with respect to time
Having demonstrated that Glut1 and Glut4 were the glucose
transporter proteins detected in immature rat uterine tissue, uterine
RNA and protein samples were collected over a time course of
E2 treatment to determine whether either Glut1 or Glut4
expression levels change with E2 treatment. Western blots
measuring both Glut1 and Glut4 protein levels in immature rat uteri
were performed for a series of E2 treatments ranging from
1–12 h between injection and death (Fig. 4). Glut1 protein levels increased
between the 2 and 4 h E2 treatments, whereas Glut4
protein levels did not change during the time course. These results
suggest that increased levels of Glut1 protein contribute to the
observed increase in glucose uptake after E2 injection.
|
). Glut4 steady state mRNA levels
did not change in response to E2 over the time course of
the experiment. This result is in agreement with the constant levels of
Glut4 protein that were measured after the same E2 time
course. Glut1 mRNA levels increased approximately 4-fold relative to
control values within 4–6 h after E2 treatment, coinciding
with the increase in Glut1 protein levels and glucose uptake.
|
)
and the increase in mRNA levels of Glut 1 (Fig. 5) observed after an
injection of 1 µg E2 all correlate well with
respect to the percent increase, although the initial change in the
rate of glucose uptake suggests the existence of a second mechanism.
These data strongly suggest that E2 is up-regulating
glucose transport through the increased expression of Glut1. This
effect could be the result of E2 up-regulating
transcription of Glut1 mRNA, increasing mRNA half-life, or both types
of regulation.
Glucose transporters Glut1 and Glut4 in immature rat uterus do not
translocate to the plasma membrane in response to
E2 stimulation
Both Glut1 and Glut4 have been shown to translocate from
intracellular vesicles to the cell surface as a means of increasing the
rate of glucose transport (20). This type of up-regulation would result
in kinetics similar to those observed in E2-treated
immature rat uteri, an increased Vmax and a constant Km. To
determine whether the rapid increase in glucose uptake in response to
E2 could be due to translocation of either Glut1 or Glut4
to the plasma membrane, 19-day-old female Sprague Dawley rats were
injected with 1 µg E2 2 h before they were killed.
Within this time period there was a stimulation of glucose uptake (Fig. 1), but Glut1 mRNA and proteins levels did not increase (Figs. 4 and 5). Subcellular fractionations were performed on uterine tissue as
described in Materials and Methods. Western blots were
performed on equal amounts of proteins from membrane and microsomal
fractions of both E2-treated and control rats. Blots were
probed with antibodies specific for Glut1 and Glut4 to measure
translocation to the plasma membrane and, with
Na+,K+-ATPase as a membrane marker, to measure
the effectiveness of the fractionation.
Na+,K+-ATPase was detected exclusively in the
plasma membrane fractions (Fig. 6) and
was concentrated relative to whole protein samples (data not shown).
There was no significant change in Glut4 distribution between the
plasma membrane fraction and the microsomal fraction in control
vs. E2-stimulated uteri. Glut1 was detected
almost exclusively at the plasma membrane in both control and
E2-stimulated nuclei. Neither glucose transporter
translocated to the cell surface in response to E2. As a
control for the protocol, subcellular fractionation was performed on
skeletal muscle that had been treated with insulin, a peptide hormone
that is known to cause the translocation of Glut4 (6, 20).
Translocation of Glut4 was observed in response to insulin, indicating
the effectiveness of the protocol (Fig. 7).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cellular responses to estrogen have been grouped with respect to time as either early (rapid) or late, with early responses occurring only minutes after estrogen treatment. Late responses are assumed to be the result of estrogen altering target gene transcription. However, for the early responses, the mechanisms of action are more difficult to explain because of the time required for transcription to affect a cellular response. The rapid effects of E2 on immature rat uterine metabolism had been carefully measured by the early 1970s (12), and one of these effects was a stimulation in the rate of glucose uptake into the immature rat uterus after a single injection of E2. This stimulation was observed to begin in less than 1 h after E2 treatment (7), eventually reaching a plateau of approximately 4-fold greater than the control value, and then diminishing slightly (16). The stimulation was shown to exhibit kinetics consistent with an increase in transporter number, and the effect was sensitive to cycloheximide, indicating the need for new protein synthesis.
These characteristics, cycloheximide sensitivity and the observed kinetics, agree with the observed increase in Glut1 after E2 treatment. The increase may be caused by direct estrogen receptor (ER) activation of Glut1 transcription, or it may be a secondary response. The standard mechanism of action through which E2 can directly alter the transcription of a gene is well characterized. E2 freely passes through both the plasma and nuclear membranes and binds to the ER protein. ER then mediates the transcription of target genes through nearby sequences collectively referred to as ER elements (EREs). The consensus sequence to which ER binds is the Xenopus laevis vitellogenin A2 gene (21), a palindrome with the sequence AGGTCACAGTGACCT. A preliminary examination of 666 bp from the 5'-upstream region of rat Glut1 (22) did not reveal this sequence (data not shown), but rat Glut1 still may be transcriptionally regulated directly through ER. Other genes, such as progesterone receptor, are activated by ER through variations in the consensus ERE or through ERE half-sites (23), and the rat Glut1 5'-upstream sequence does contain half-sites (22).
The increase in Glut1 protein in response to E2 agrees with the previous cycloheximide and kinetic data, but the timing of the initial increase in the rate of glucose uptake, before Glut1 protein levels change, leaves open the possibility of a second mechanism. Stimulation of Glut1 by peptide hormones, such as insulin, has been shown to occur by at least two mechanisms, an increase in glucose transporter expression through the activation of transcription and/or protein synthesis, and glucose transporter translocation from intracellular vesicles to the plasma membrane. Estrogen has been shown to rapidly activate several components of signal transduction pathways that are also associated with glucose transporter translocation, such as cAMP (24), the release of intracellular calcium (25), and mitogen-activated protein kinase (26). Glut1 and Glut4, the two glucose transporters observed in the immature rat uterus, have both been shown to translocate, and translocation would satisfy the observed kinetics associated with E2 stimulation. Furthermore, the time required for translocation of glucose transporters fits within the same time window as glucose transport stimulation by E2. However, in this defined system, neither Glut1 nor Glut4 increased in plasma membrane fractions 2 h after a single 1-µg E2 injection. It therefore is unlikely that Glut1 or Glut4 translocation has a role in the initial uterine response to E2.
Although they are atypical, these data do not represent a unique set of conditions for the stimulation of glucose transport. Inhibitors of oxidative phosphorylation, such as azide, have been characterized in Clone 9 cells (a rat liver cell line that is not transformed and contains only Glut1) as causing a biphasic stimulation of glucose transport (27). The early phase is a rapid increase in glucose transport of 8- to 12-fold in 2 h and a later phase at 8–12 h that consists of another 1.6-fold increase above that obtained with the 2-h stimulation. Kinetics associated with the early increase indicate an elevated Vmax with no change in Km, but no increase in Glut1 protein or mRNA is observed until 8 h after azide treatment. In addition, no increase in Glut1 protein at the plasma membrane is observed (28), indicating that the increase is not due to the translocation of Glut1. The authors hypothesize that a fraction of the glucose transporter population at the cell surface is inactive and becomes activated in response to azide. In this way both the kinetics and the observed increase in glucose transport can be satisfied. The discussion of these similarities is not intended to suggest that E2 and azide stimulate glucose transport by the same mechanism. It is important to note that this is not the first description of an early glucose transport response that works predominantly through Glut1, has kinetics indicating an increase in transporter number, results in an eventual up-regulation of Glut1 protein, and cannot be explained by either an increase in Glut1 expression or translocation.
|
Footnotes |
|---|
Received September 2, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  D.-Y. Wang, R. Fulthorpe, S. N. Liss, and E. A. Edwards Identification of Estrogen-Responsive Genes by Complementary Deoxyribonucleic Acid Microarray and Characterization of a Novel Early Estrogen-Induced Gene: EEIG1 Mol. Endocrinol., February 1, 2004; 18(2): 402 - 411. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Daikoku, H. Matsumoto, R. A. Gupta, S. K. Das, M. Gassmann, R. N. DuBois, and S. K. Dey Expression of Hypoxia-inducible Factors in the Peri-implantation Mouse Uterus Is Regulated in a Cell-specific and Ovarian Steroid Hormone-dependent Manner. EVIDENCE FOR DIFFERENTIAL FUNCTION OF HIFs DURING EARLY PREGNANCY J. Biol. Chem., February 21, 2003; 278(9): 7683 - 7691. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Turkay Korgun, R. Demir, A. Hammer, G. Dohr, G. Desoye, G. Skofitsch, and T. Hahn Glucose Transporter Expression in Rat Embryo and Uterus During Decidualization, Implantation, and Early Postimplantation Biol Reprod, November 1, 2001; 65(5): 1364 - 1370. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. CHENG, M. COHEN, J. WANG, and C. A. BONDY Estrogen augments glucose transporter and IGF1 expression in primate cerebral cortex FASEB J, April 1, 2001; 15(6): 907 - 915. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
