Low Glucose Enhances Na1/Glucose Transport in Bovine
Brain Artery Endothelial Cells
Tomoyuki Nishizaki, MD, PhD; Toshiyuki Matsuoka, MD
Downloaded from http://stroke.ahajournals.org/ by guest on June 17, 2017
Background and Purpose—Brain arteries are structurally characterized by the tight junctions of the endothelium and by
no vasa vasorum that feed arteries themselves. This raises the question of how brain arteries are provided with glucose.
A possible explanation is that glucose uptake into arteries may be mediated by both GLUT1, a facilitative glucose
transporter, and a Na1/glucose cotransporter (SGLT)-like glucose transporter. The functional role for the SGLT-like
glucose transporter, however, is unknown. In the present study we investigated SGLT-like glucose transporter– operated
glucose uptake into brain arterial endothelial cells by recording glucose-evoked Na1 currents and monitoring uptake of
[3H]-2-deoxy-D-glucose ([3H]-2-DOG).
Methods—Endothelial cells were cultured from bovine cerebral cortical arteries. Whole-cell patches were made to cells,
and glucose-evoked currents were recorded. Cells were incubated with [3H]-2-DOG, and the uptake was determined by
a liquid scintillation counter.
Results—Glucose and a-methyl-D-glucoside (aMeDG), a specific compound for the SGLTs, evoked Na1 currents in a
whole-cell voltage-clamp configuration, and the currents were enhanced in cells with over 30 minutes’ preincubation
in glucose-free media. Glucose-induced currents were inhibited by aMeDG, by the selective SGLT inhibitor phlorizin,
by dinitrophenol (DNP), an inhibitor of energy metabolism, or by deletion of Na1 from extracellular solution, which
indicates that glucose uptake into endothelial cells was mediated by a Na1- and energy-dependent glucose transporter.
Notably, the currents were desensitized, reduced in a glucose concentration– dependent manner, and markedly inhibited
by either a second application of glucose or the addition of glucose to the patch electrode filling solution; they were
potentiated, however, by treatment with cytochalasin B, a GLUT1 to GLUT5 inhibitor. Consistent with the results of
patch-clamp recordings, uptake of [3H]-2-DOG into endothelial cells was enhanced by glucose-free insult, and the
enhancement was mediated by an SGLT-like glucose transporter.
Conclusions— The results presented demonstrate that an SGLT-like glucose transporter takes part in glucose uptake into
brain artery endothelial cells and that the uptake is regulated by intracellular glucose concentrations; glucose-free insult
and the ensuing low cytosolic glucose enhance activity of the SGLT-like glucose transporter. The SGLT-like glucose
transporter in the brain arterial endothelium thus may be important in the maintenance of an adequate glucose
concentration in the arterial wall under conditions of stress, such as hypoglycemia. (Stroke. 1998;29:844-849.)
Key Words: sodium-glucose transport system n cerebral arteries n endothelium n glucose
B
rain arteries exhibit an unique structure distinct from body
arteries; they are lined with the tight junctions of the
endothelium and have no vasa vasorum to feed the arterial wall.1
One therefore assumes that a carrier protein for glucose, a
glucose transporter, may provide brain arteries with the major
energy source glucose. Indeed, GLUT1, a facilitative glucose
transporter, is shown to be expressed in the rat brain capillary
endothelium2,3 and in cultured endothelial cells from bovine
cerebral cortical arteries,4 thus supporting this hypothesis. In
addition, we earlier found that an SGLT-like glucose transporter
is expressed in cultured brain artery endothelial cells as well as
GLUT1.4 This suggests the possibility that this transporter may
be also involved in glucose uptake into endothelial cells. It is
unknown, however, whether glucose is actually taken up into
endothelial cells by this transporter, how this transporter is
regulated, or what functional role this transporter has.
See Editorial Comment, page 849
To address these questions, we monitored glucose-evoked Na1
currents in cultured endothelial cells from bovine cerebral cortical
arteries and assayed uptake of [3H]-2-DOG into cells. We show here
that low glucose enhances activity of the SGLT-like glucose
transporter and that this transporter may have a crucial role in the
maintenance of an adequate glucose concentration in the arterial
wall under conditions of stress, such as hypoglycemia.
Materials and Methods
Cell Culture
Bovine brain endothelial cells were cultured as described previously.4 Briefly, endothelial cell sheets were mechanically isolated from
bovine cerebral cortical arteries 300 to 500 mm in diameter. The
explants were plated on collagen-coated dishes and grown in
Dulbecco’s modified Eagle’s medium with 20% fetal bovine serum,
Received August 22, 1997; final revision received January 14, 1998; accepted January 15, 1998.
From the Department of Physiology, Kobe University School of Medicine, Kobe, Japan.
Correspondence to T. Nishizaki, MD, PhD, Department of Physiology, Kobe University School of Medicine, 7–5-1 Kusunoki-cho, Chuo-ku, Kobe 650,
Japan.
© 1998 American Heart Association, Inc.
844
Nishizaki and Matsuoka
Selected Abbreviations and Acronyms
BBB 5 blood-brain barrier
cyto B 5 cytochalasin B
DNP 5 dinitrophenol
2-DOG 5 2-deoxy-D-glucose
GLUT 5 facilitative glucose transporter
aMeDG 5 a-methyl-D-glucoside
SGLT 5 Na1/glucose cotransporter
1 ng/mL bovine pituitary fibroblast growth factor, 50 IU/mL penicillin, and 50 mg/mL streptomycin. The third-passage cells grown in
6-well, collagen-coated dishes and in coverslips were used for assay
of 2-DOG uptake and for patch-clamp recording, respectively.
Immunohistochemistry
Downloaded from http://stroke.ahajournals.org/ by guest on June 17, 2017
The brain arteries, endothelial cell sheets, and cultured cells were
fixed in 100% methanol, incubated in 0.3% H2O2/methanol, and
blocked in 10% normal goat serum before incubation with antihuman factor VIII/von Willebrand factor antibody (1:1000). The
cells were then stained with peroxidase-conjugated donkey antirabbit IgG and 3-amino-9-ethylcarbazole, followed by cell nucleus
staining with hematoxylin.
Electrophysiology
Cells were transferred to the recording chamber and continuously
superfused at room temperature (20°C to 22°C) with glucose-free
extracellular solution ([mmol/L] 145 NaCl, 5 KCl, 2.4 CaCl2, 10
HEPES, 0.331023 tetrodotoxin, pH 7.4) after treatment with
10 mmol/L glucose-containing or glucose-free extracellular solution
at 37°C for more than 30 minutes. For Na1-free extracellular
solution, 145 mmol/L NaCl was replaced with 145 mmol/L LiCl.
Whole-cell patches were made to the cells with use of patch
electrode filling solution consisted of (mmol/L) 150 KCl, 5 EGTA,
0.5 ATP, and 10 HEPES, pH 7.2. Membrane currents from wholecell voltage-clamp were recorded by an Axopatch-200A amplifier
(Axon Instrument Inc). After formation of whole-cell patches, series
resistance compensation was made up to '95%. Glucose was bath
applied to cells during recording. The currents were filtered at 5 kHz,
stored on magneto optical disk, and analyzed on a microcomputer
with pClamp software (version 6; Axon Instrument Inc).
Assay of [3H]-2-DOG Uptake
The confluent cells with or without glucose-free exposure at 37°C
were washed twice with glucose-free extracellular solution. Subsequently, the cells were incubated for 10 minutes in extracellular
solution containing [3H]-2-DOG (specific activity, 6.1 Ci/mL; Dupont) in the presence and absence of several kinds of inhibitors. In
some cases, [3H]-2-DOG uptake was assayed in Na1-free extracellular solution. After incubation with [3H]-2-DOG, the cells were
washed three times with glucose/ Na1-free solution and then lysed
with 0.1% SDS. The levels of [3H]-2-DOG uptake into endothelial
cells were detected by a liquid scintillation counter.
Results
Cultured Endothelial Cells
The inner layer of the brain artery was positive against the
marker for endothelial cells, anti-factor VIII antibody (Fig
1A). The endothelial cell sheet alone was isolated for cultures
(Fig 1B). Cultured cells displayed uniform cobblestone appearance typical for endothelial cells (Fig 1C), and the cells
were also positive against anti-factor VIII antibody (Fig 1D).
Effect of SGLT-like Glucose Transporter
Operation on Brain Endothelial Cells
Transport of glucose (D-isomer) mediated by the SGLTs can
be relatively estimated by monitoring Na1 currents because
April 1998
845
Na1 is cotransported with glucose into cells; otherwise,
L-glucose evoked no current. We therefore used D-glucose in
this and further experiments. Application of glucose
(0.1 mmol/L) produced inward currents at a holding potential
of -60 mV in the whole-cell voltage-clamp configuration (Fig
2). Notably, the currents were potentiated after more than 30
minutes’ pretreatment with glucose-free extracellular solution
(Fig 2). aMeDG (0.1 mmol/L), a specific compound for the
SGLTs,5 also produced currents to the same level as glucose
(0.1 mmol/L) (Fig 2). Currents evoked by glucose
(0.1 mmol/L) in cells with 1-hour glucose-free exposure were
inhibited by either the selective SGLT inhibitor phlorizin
(50 mmol/L) or the ATP uncoupler DNP (1 mmol/L) (Fig
3A). Glucose (0.1 mmol/L) never produced currents in the
presence of aMeDG (0.1 mmol/L) (Fig 3A) or in Na1-free
extracellular solution (Fig 3B). Voltage pulses from 2140 to
120 mV in 20-mV increments after application of glucose
(0.1 mmol/L) or aMeDG (0.1 mmol/L) generated voltagedependent inward currents that were inhibited by phlorizin
(50 mmol/L), although spontaneous currents were not observed (Fig 4A and 4B). These results indicate that a
Na1/energy-dependent glucose transporter (SGLT-like glucose transporter) was involved in glucose uptake into brain
endothelial cells and that the uptake was enhanced by
glucose-free insult.
Regulation of SGLT-like Glucose
Transporter–Operated Currents by
Cytosolic Glucose
Currents induced by glucose (0.1 mmol/L) or aMeDG
(0.1 mmol/L) were slowly desensitized, and a second application of glucose elicited a still lesser response after a
15-minute washing (Fig 3C). In addition, when glucose
(0.01 mmol/L) was added to patch electrode filling solution,
application of glucose (0.1 mmol/L) outside the patch pipette
evoked very small currents (Fig 3C). Glucose-induced currents reduced in a dose-dependent manner and .10 mmol/L
glucose produced no current (Fig 5). These results suggest
that low cytosolic glucose enhanced glucose uptake mediated
by the SGLT-like glucose transporter; otherwise, high cytosolic glucose inhibited the uptake.
Cytochalasin B (1 mmol/L), a facilitative glucose transporter (GLUT1 to GLUT5) inhibitor, markedly enhanced
currents induced by glucose (1 mmol/L), and 10 mmol/L
glucose produced currents in the presence of cyto B (Fig 3D)
although no current was evoked in the absence of cyto B (Fig
5), further supporting the idea that glucose uptake into
endothelial cells mediated by the SGLT-like glucose transporter is regulated by intracellular glucose concentrations.
Uptake of [3H]-2-DOG Into Brain
Endothelial Cells
Uptake of 2-DOG into endothelial cells was 37.161.8
pmol/mg protein per 10 minutes in cells without glucose-free
exposure (Table 1). The facilitative glucose transporter inhibitors, such as phloretin (50 mmol/L) and cyto B (1 mmol/L),
inhibited glucose uptake by 96% and 97%, respectively,
whereas phlorizin (50 mmol/L), the energy metabolism inhibitors DNP (1 mmol/L) and iodoacetate (1 mmol/L), the
846
SGLT in the Brain Artery Endothelium
Downloaded from http://stroke.ahajournals.org/ by guest on June 17, 2017
Figure 1. Cultured endothelial cells from brain cortical arteries. The coronal section of the brain artery (A) and the isolated endothelial
cell sheet (B) were stained with factor VIII antibody. Cultured endothelial cells are shown in panel C; the cells were positive for factor
VIII staining (D).
Na1-K1/ATPase pump inhibitor ouabain (100 mmol/L), or
deprivation of Na1 from extracellular solution blocked the
uptake by only 4% to 6% (Table 1). Consistent with the result
of glucose-induced currents, 2-DOG uptake was enhanced by
15.4 pmol/mg protein per 10 minutes in cells with 1-hour
incubation in glucose-free extracellular solution (D increase)
(Table 1). The enhancement was clearly inhibited by phlorizin, DNP, iodoacetate, ouabain, and deprivation of Na1, but,
in contrast, it was not affected by either phloretin or cyto B
(Table 1). This indicates that an enhancement in 2-DOG
uptake by glucose-free insult resulted from activation of the
SGLT-like glucose transporter.
Discussion
Figure 2. Glucose- and aMeDG-evoked currents in brain endothelial cells. Whole-cell patches were made to cells with and
without 1-hour exposure to glucose-free extracellular solution,
and glucose (0.1 mmol/L; n57) or aMeDG (0.1 mmol/L; n57)
was applied to the cells. The holding potential was 260 mV. In
this and following figures, inward currents correspond to downward deflections.
Plasma membranes are not permeable to a polar molecule
glucose, and glucose is therefore taken up into cells by
glucose transporters. Two families of glucose transporters are
presently identified: facilitative glucose transporters such as
GLUT1 (erythrocyte/HepG2), GLUT2 (liver), GLUT3
(brain), GLUT4 (muscle/fat), and GLUT5 (small intestine),6,7
and SGLTs (small intestine/kidney) such as SGLT1 and
SGLT2.8 –13 Extensive studies have been carried out to understand the distribution of these glucose transporters in a variety
of tissues. Very little is known, however, about glucose
transporters in brain arteries. Brain arteries are likely provided with glucose via glucose transporters, since they are not
fed by vasa vasorum and have a barrier that is formed by the
tight junctions of the endothelium. In an earlier study,4
Nishizaki and Matsuoka
Downloaded from http://stroke.ahajournals.org/ by guest on June 17, 2017
Figure 3. SGLT-like glucose transporter in brain endothelial
cells. Whole-cell patches were made to cells with 1-hour incubation in glucose-free extracellular solution. In panel A, glucose
(0.1 mmol/L) was applied to the cells in the presence of phlorizin (50 mmol/L; n57), DNP (1 mmol/L; n55), or aMeDG
(0.1 mmol/L; n55); in this case, glucose was applied when
aMeDG-induced currents reversed. B, glucose (0.1 mmol/L) was
applied to a single cell in Na1-free extracellular solution and, in
turn, Na1-containing extracellular solution (n55). C, glucose
(0.1 mmol/L) was repetitively applied to a single cell at
15-minute intervals (n57) or applied to cells 5 minutes after
patch formation with the patch electrode filling solution containing 0.01 mmol/L glucose (n57). D, glucose (1 mmol/L) was
applied to a single cell, and a second application was carried
out after treatment with 1 mmol/L cyto B (n57). In some cells,
10 mmol/L glucose was applied to cells in the presence of cyto
B (1 mmol/L; n57). The holding potential was 260 mV.
through use of immunohistochemical analysis and Western
immunoblot analysis, we found that the GLUT1 and SGLTlike glucose transporter are expressed in cultured endothelial
cells from bovine brain cerebral cortical arteries, which
suggests that glucose uptake into the endothelium is achieved
by these transporters. The present study provides further
evidence for the expression of the SGLT-like glucose transporter and makes clear its functional role in brain arteries.
The SGLTs, which are identified in the small intestine and
kidney, continuously carry glucose into cells against its
concentration gradient as far as extracellular glucose is
present; SGLT-operated currents are not desensitized and
enhanced in a glucose concentration– dependent manner.14 In
the present study, glucose-induced currents were inhibited by
the selective SGLT inhibitor phlorizin, by the ATP uncoupler
DNP, by deletion of Na1 from extracellular solution, or by
aMeDG, a specific agent for the SGLTs. aMeDG produced
currents in a fashion that mimics the effect of glucose, which
indicates that an SGLT-like glucose transporter actually
operates on the endothelium of brain arteries. Notably, the
currents, inconsistent with those via the SGLTs in brushborder membranes, were desensitized. More striking are the
observations that pretreatment with glucose-free media potentiated the currents and that otherwise, repetitive applica-
April 1998
847
Figure 4. Glucose- and aMeDG-induced current/voltage relations. Voltage pulses from 2140 to 20 mV in 20-mV increments
were applied to cells with 1 hour of glucose-free exposure
before and after application of glucose (0.1 mmol/L; n55) (A) or
aMeDG (0.1 mmol/L; n55) (B) in the presence and absence of
phlorizin (50 mmol/L).
tions of glucose, higher concentrations of extracellular glucose, or addition of glucose to the patch electrode filling
solution markedly reduced the currents. These findings suggest that the SGLT-like glucose transporter expressed in brain
endothelial cells has a characteristic distinct from the SGLTs
Figure 5. Dose-response effect of glucose on the evoked currents. Glucose was applied to cells with 1 hour of glucose-free
exposure at concentrations as indicated. The holding potential
was 260 mV. Typical currents are illustrated in the upper panel,
with results summarized in the lower panel. Each point represents the average percentage (6SD) of the current evoked by
0.1 mmol/L glucose (n55 to n57).
848
SGLT in the Brain Artery Endothelium
Uptake of [3H]-2-DOG Into Brain Endothelial Cells
Treatment
Uptake, pmol/mg
Protein per 10 min
D
Increase
n
1-h incubation in glucose-free solution (2)
Control
Phloretin (50 mM)
Cyto B (1 mM)
37.161.8
60
1.760.1
16
1.260.2
16
35.061.6
48
DNP (1 mM)
35.861.2
36
Iodoacetate (1 mM)
35.264.5
36
Ouabain (100 mM)
35.061.6
30
Na1-free
35.560.4
30
Phlorizin (50 mM)
1-h incubation in glucose-free solution (1)
Control
52.562.7
15.4
60
Phloretin (50 mM)
18.161.3
16.4
30
Cyto B (1 mM)
16.561.4
15.3
30
Downloaded from http://stroke.ahajournals.org/ by guest on June 17, 2017
Phlorizin (50 mM)
35.261.0
0.2
48
DNP (1 mM)
34.461.2
21.4
48
Iodoacetate (1 mM)
36.364.5
1.1
48
Ouabain (100 mM)
35.861.6
0.8
48
34.560.4
21.0
48
1
Na -free
in brush-border membranes and is regulated by cytosolic
glucose concentrations.
It is recognized that GLUT1 conveys glucose according to
glucose concentration gradient and is characterized by faster
glucose transport than the SGLTs. Interestingly, glucoseinduced currents in brain endothelial cells were potentiated in
the presence of cyto B, which suggests that cyto B produces
the same condition as glucose-free insult; cyto B predominantly blocks glucose entry via GLUT1, and the eventual
decrease in cytosolic glucose concentrations enhances activity of the SGLT-like glucose transporter.
Uptake of 2-DOG into brain endothelial cells was inhibited
by over 95% by the facilitative glucose transporter inhibitors
phloretin and cyto B in cells without glucose-free exposure,
which indicates that GLUT1 is responsible for glucose uptake
into the endothelium under normal conditions. On the other
hand, the uptake was enhanced by pretreatment with glucosefree media and the enhancement was inhibited by the SGLT
inhibitor, the energy metabolism inhibitors, and the Na1/K1ATPase pump inhibitor, and by deprivation of extracellular
Na1. This provides further evidence that glucose-free insult
followed by low cytosolic glucose enhances activity of the
SGLT-like glucose transporter, thus leading to an increase in
glucose uptake. It is presently unknown by what mechanism
glucose-free insult enhances activity of the SGLT-like glucose transporter. A study15 has demonstrated that high extracellular glucose inhibits insulin receptor kinase activity by
serine/threonine phosphorylation, suggesting that glucose
itself serves as a second messenger. The glucose signal may
stimulate the expression of the SGLT-glucose transporter or
the translocation of this transporter from cytosol to plasma
membrane, or it may directly activate this transporter. To
address this question, we are carrying out further
experiments.
The GLUT1 and SGLT-like glucose transporter in the
brain endothelium, thus, appear to supply brain arteries with
glucose; the GLUT1 operates under normal conditions and,
alternatively, the SGLT-like glucose transporter under conditions of stress such as hypoglycemia. The SGLT-like
glucose transporter seems to be important in the maintenance
of an adequate glucose concentration in the arterial wall.
Another functional role of the SGLT-like glucose transporter
may be its involvement in transport of glucose across the
BBB. GLUT1 in brain capillaries is proposed to be the major
glucose transporter at the BBB.2 GLUT1 is preferentially
localized on the abluminal membranes of the endothelium,3
and therefore transport of glucose across the BBB cannot be
explained by GLUT1 alone. Considering that the BBB is
composed of the endothelium, with the tight junctions and
foot processes of the astrocytes, cultured brain endothelial
cells here may not reflect the function at the BBB. Cultured
brain endothelial cells, however, exhibited g-glutamyl
transpeptidase activity (20567.9 nmol/mg protein per
minute), a BBB enzymatic marker16 whereas no activity was
obtained with cultured carotid artery endothelial cells, which
suggests that cultured brain endothelial cells possess characteristics of the BBB. In addition, the findings that glucose was
transported across the arterial wall by the SGLT-like glucose
transporter in inverted cerebral cortical arteries but not in
noninverted ones and that glucose transport was not observed
in inverted carotid arteries4 indicate that the SGLT-like
glucose transporter is selectively expressed in brain arteries
and conveys glucose from the luminal membrane toward the
abluminal side. The SGLT-like glucose transporter may thus
be responsible for glucose uptake into the endothelium at the
BBB and the GLUT1 for glucose transport toward astrocytes
just as glucose is absorbed by the SGLT1/2 in the brushborder membranes and exits the small intestine or renal
tubules across the basolateral membranes via GLUT2/5.14 To
obtain further evidence for this, we are currently testing
uptake of glucose using cocultures of astrocytes and endothelial cells.
In conclusion, the results presented here demonstrate that
an SGLT-like glucose transporter is involved in uptake into
brain endothelial cells as well as GLUT1 and that low glucose
enhances activity of the SGLT-like glucose transporter. The
SGLT-like glucose transporter seems to have a functional
role in the maintenance of an adequate glucose concentration
in the arterial wall under conditions of stress such as
hypoglycemia.
References
1. Edvinsson L, MacKenzie ET, McCulloch J. Cerebral Blood Flow and
Metabolism. New York, NY: Raven Press Publishers; 1993:40 –56.
2. Pardridge WM, Boad RJ, Farrell CR. Brain-type glucose transporter
(GLUT-1) is selectively localized to the blood-brain barrier. J Biol Chem.
1990;265:18035–18040.
3. Farrell CL, Pardridge WM. Blood-brain barrier glucose transporter is
asymmetrically distributed on brain capillary endothelial luminal and
abluminal membranes: an electron microscopic immunogold study. Proc
Natl Acad Sci U S A. 1991;88:5779 –5783.
4. Nishizaki T, Kammesheidt A, Sumikawa K, Asada, T, Okada Y. A sodiumand energy-dependent glucose transporter with similarities to SGLT1–2 is
expressed in bovine cortical vessels. Neurosci Res. 1995;22:13–22.
Nishizaki and Matsuoka
5. Birnir B, Donald DDF, Wright EM. Voltage-clamp studies of the Na1/
glucose cotransporter cloned from rabbit small intestine. Pflugers Arch.
1991;418:79 – 85.
6. Bell GI, Kayano T, Buse JB, Burant CF, Takeda J, Lin D, Fukumoto H,
Seino S. Molecular biology of mammalian glucose transporters. Diabetes
Care. 1990;13:198 –208.
7. Thorens B. Facilitated glucose transporters in epithelial cells. Annu Rev
Physiol. 1993;55:591– 608.
8. Hediger MA, Coady MJ, Ikeda TS, Wright EM. Expression cloning and
cDNA sequencing of the Na1/glucose co-transporter. Nature. 1987;330:
379 –381.
9. Hediger MA, Turk E, Wright EM. Homology of the human intestinal
Na1/glucose and Escherichia coli Na1/proline co-transporters. Proc Natl
Acad Sci U S A. 1989;86:5748 –5752.
10. Coady MJ, Pajor AM, Wright EM. Sequence homologies among
intestinal and renal Na1/glucose co-transporters. Am J Physiol. 1990;259:
C605– 610.
April 1998
849
11. Ohta T, Isselbacher KJ, Rhoads DB. Regulation of glucose transporters in
LLC-PK1 cells: effects of D-glucose and monosaccharides. Mol Cell
Biol. 1990;10:6491– 6499.
12. Morrison AI, Panayotova HM, Feigl G, Scholermann B, Kinne RK.
Sequence comparison of the sodium-D-glucose cotransport systems in
rabbit renal and intestinal epithelia. Biochim Biophys Acta. 1991;1089:
121–123.
13. Wells RG, Mohandas TK, Hediger MA. Localization of the Na1/glucose
co-transporter gene SGLT2 to human chromosome 16 close to the centromere. Genomics. 1993;17:787–789.
14. Wright EM. The intestinal Na1/glucose cotransporter. Annu Rev Physiol.
1993;55:575–589.
15. Pillay TS, Xiao S, Olefsky JM. Glucose induced phosphorylation of the
insulin receptor-functional effects and characterization of phosphorylation sites. J Clin Inves. 1996;97:613– 620.
16. Orlowski M, Sessa G, Green JP. g-glutamyl transpeptidase in brain
capillaries: possible site of a blood-brain barrier for amino acids. Science.
1974;184:66 – 68.
Downloaded from http://stroke.ahajournals.org/ by guest on June 17, 2017
Editorial Comment
Glucose is the source of metabolic energy in brain and
requires continuous transport from plasma to parenchyma.
Two classes of glucose carriers have been described in
mammalian cell membranes: facilitative glucose transporters,
which transport glucose down its concentration gradient, and
Na1/glucose cotransporters, which couple the downhill transport of Na1 with an uphill uptake of glucose. However, it is
unclear precisely which carriers are present in arterial endothelial cells. GLUT1, a facilitative glucose transporter, and
SGLT, a type of Na1/glucose cotransporter, are known to be
present in cultured brain artery endothelial cells (BAEC). The
present study provides electrophysiological and biochemical
evidence supporting a role for SGLT in cultured BAEC under
conditions of low glucose concentration.
The key finding is that SGLT activity is enhanced by low
extracellular glucose, which could serve as compensation for
decreased glucose uptake via GLUT1 under these conditions.
In a series of well-designed experiments, the authors demonstrate that SGLT activity is regulated by cytosolic glucose.
Addition of glucose to the pipette solution, which represents
the intracellular compartment, inhibited SGLT activity. A
dose-dependent inverse relationship was present between
extracellular glucose concentration and SGLT activity, an
effect that is presumably mediated by cytosolic glucose,
because activation of SGLT by low glucose is quickly
desensitized. Regulation of SGLT by cytosolic glucose is
further supported by the observation that inhibiting GLUT1-
mediated glucose uptake (potentially depriving the cell of
glucose) leads to potentiation of SGLT activity. One major
strength of the study is its confirmation of electrophysiological findings by a biochemical assay of glucose uptake with
use of radiolabeled glucose. The assay shows that depletion
of cytosolic glucose by preincubation in glucose-free solution
enhances glucose uptake, and the increment is abolished by
inhibitors of SGLT or Na1/K1-ATPase, which have little
effect on baseline uptake under normal glucose concentration.
The reader should be reminded that the main pathway for
glucose uptake into BAEC under normal glucose concentration is GLUT1-facilitated diffusion. During normoglycemia,
SGLT contributes very little to glucose uptake. It must also be
emphasized that the endothelial cells of this study were
isolated from large cerebral arteries, not from brain capillaries
that carry out the bulk of glucose transport across the
blood-brain barrier. The importance of SGLT in these cells
remains to be demonstrated. From a clinical perspective, it
will be important to compare these results in normal endothelial cells with those obtained from diabetic or atherosclerotic vessels.
Nabil J. Alkayed, MD, PhD
Patricia D. Hurn, PhD
Guest Editors
Anesthesiology/Critical Care Medicine
Johns Hopkins Medical Institutions
Baltimore, Maryland
Low Glucose Enhances Na+/Glucose Transport in Bovine Brain Artery Endothelial Cells
Tomoyuki Nishizaki and Toshiyuki Matsuoka
Downloaded from http://stroke.ahajournals.org/ by guest on June 17, 2017
Stroke. 1998;29:844-849
doi: 10.1161/01.STR.29.4.844
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1998 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/29/4/844
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click
Request Permissions in the middle column of the Web page under Services. Further information about this
process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Stroke is online at:
http://stroke.ahajournals.org//subscriptions/