| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
INSULIN-GLUCAGON-GI PEPTIDES-DIABETES MELLITUS |
Institute of Physiology, Medical Faculty (C.B., Y.F.), and Interdisciplinary Center of Clinical Research BIOMAT, Medical Faculty (C.B.), RWTH Aachen, D-52057 Aachen, Germany; Solvay Pharmaceuticals, Inc. (Y.F.), D-30173 Hannover, Germany; and Departament de Bioquímica i Biologia Molecular, Facultat de Biologia, Universitat de Barcelona (L.S., E.T., M.P., A.Z.), 08028 Barcelona, Spain
Address all correspondence and requests for reprints to: Prof. Yvan Fischer, Solvay Pharmaceuticals, Inc., Hans Böckler Allee 20, D-30173 Hannover, Germany. E-mail: yvan.fischer{at}solvay.com
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Rotenone recruited GLUT4 and GLUT1 to the cell surface from endosomal pool 2 without affecting pool 1. Kinetic experiments were consistent with rotenone acting on an intracellular compartment that is in close connection with the plasma membrane. In contrast, insulin caused rapid, complete depletion of GLUT4 from pool 1 and reduced the GLUT1 content of pool 2 by approximately 50%, whereas, surprisingly, no net decrease in GLUT4 occurred in this pool. Subsequent insulin withdrawal resulted in slow replenishment of pool 2 with GLUT1 and of pool 1 with GLUT4. When pool 1 was still largely depleted of GLUT4, a second insulin challenge did reduce GLUT4 in pool 2 and stimulated glucose transport to the same extent as the first insulin treatment.
In conclusion, the storage pool is the primary source of GLUT4 in response to insulin, but not to rotenone. In addition, the endosomal compartment is an important recruitment site of both GLUT1 and GLUT4 when the storage pool is either unaffected (rotenone) or depleted (by a previous insulin challenge). GLUT4 mobilized by insulin from the storage pool may pass through an intermediary (possibly endosomal) compartment on its way to the cell surface.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Glucose transport and utilization by cardiac myocytes is critical for the maintenance of normal morphology and function. Thus, GLUT4-lacking mice exhibit cardiac hypertrophy characterized by vascular sclerosis, interstitial fibrosis, and concentric hypertrophy (3, 4). Selective deletion of GLUT4 in the heart is also associated with modest hypertrophy, and cardiac dysfunction follows in response to ischemia (5). Moreover, a high rate of cardiac glucose metabolism may be crucial in conditions such as ischemia, as indicated by the beneficial effects of a selective increase in glucose utilization in animal and clinical studies (6, 7). Glucose uptake also reduces hypoxia-induced apoptosis in cultured neonatal rat cardiac myocytes (8). Conversely, the impairment of heart glucose metabolism in diabetes mellitus may contribute to the mechanical dysfunction and cardiomyopathy observed in this disease (9).
As in other peripheral insulin-sensitive tissues, glucose transport is regulated in heart through changes in the amount of glucose transporters (mainly GLUT4) present at the plasma membrane. These changes, in turn, result from a redistribution of transporters between an intracellular storage site(s) and the cell surface; thus, stimuli such as insulin trigger a translocation of GLUT4 and GLUT1 from intracellular vesicular structures to the plasma membrane in cardiac myocytes (1, 10, 11, 12). Because of its regulatory importance, the identification of the cellular traffic of glucose transporters and the internal GLUT-containing compartment is a crucial issue. In this regard, studies in cultured myocytes (13), heart (11, 14, 15), or isolated cardiac myocytes (1) suggest that in the unstimulated state, GLUT4 is distributed between at least two intracellular compartments. We have reported that in nonstimulated cardiomyocytes, intracellular GLUT4 is partitioned between two vesicle populations: one containing GLUT4, but little or no GLUT1 (pool 1), and the other containing, besides GLUT4, a substantial amount of GLUT1 and secretory carrier membrane proteins (SCAMPs; pool 2) (1). Here we examine the functional properties of these intracellular GLUT4 vesicle populations of cardiomyocytes by monitoring acute changes occurring in these vesicles in response to insulin or the mitochondrial inhibitor rotenone.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-globulin, protein
A-Sepharose-purified secondary antibodies for ECL detection (goat
antirabbit IgG and goat antimouse IgM, peroxidase-conjugated), and
rotenone were obtained from Sigma (Munich, Germany).
Aprotinin, pepstatin, and leupeptin were purchased from ICN Biochemicals, Inc. (Meckenheim, Germany); BSA (fraction V, fatty
acid free) was purchased from Roche (Mannheim, Germany).
Purified bovine insulin was a gift from Prof. Axel Wollmer (Aachen,
Germany). 2-Deoxy-D-[3H]glucose for
glucose transport measurements was obtained from Amersham Pharmacia Biotech (Braunschweig, Germany). All chemicals were
the highest grade available. Concentrated stock solutions of insulin
(in medium A, see below) or rotenone (in dimethylsulfoxide) were stored
at -20 C in appropriate aliquots and diluted just before addition to
the isolated cardiomyocytes. Immobilon polyvinylidene difluoride was
obtained from Millipore Corp. (Bedford, MA). All
electrophoresis reagents and mol wt markers were obtained from
Bio-Rad Laboratories, Inc. (Munich, Germany). Anti-GLUT4
antibody (OSCRX) was produced from rabbit after immunization with a
peptide corresponding to the last 15 amino acids of the GLUT4
carboxyl-terminus (1); monoclonal antibodies 1F8 (against
GLUT4) were supplied by Dr. Paul Pilch (Boston University, Boston, MA).
Antihuman GLUT1 glucose transporter was obtained from
Biogenesis (Bournemouth, UK).
Isolation of cardiomyocytes and glucose transport assays
Cardiomyocytes from adult female Sprague Dawley rats (180–220
g; fed ad libitum) were obtained as previously described
(16). The animals were acquired and used in compliance
with paragraph 6 of the German Animal Protection Law, and the
study was approved by the appropriate authority (Bezirksregierung
Cologne). Treatment of cardiomyocytes for all experiments (see figure
legends) was performed in medium A containing 6
mM KCl, 1 mM
Na2HPO4, 0.2
mM
NaH2PO4, 1.4
mM MgSO4, 128
mM NaCl, 10 mM HEPES, 1
mM CaCl2 and 2% BSA (fatty
acid free), pH 7.4, at 37 C, equilibrated with oxygen. The rate of
2-deoxy-D-glucose uptake was determined as
described previously (16) over the times indicated in the
figure legends. To calculate the half-life of the insulin effect on
blocking the insulin signal (Fig. 7
), transport data were fitted using
a computer program (PRISM) from GraphPad Software, Inc.
(San Diego, CA), according to the following equation: Y =
(Y0 - Ymin) x
e-Kt + Ymin (Eq I), where
Y0 is the initial, fully stimulated glucose
transport value at t = 0 (i.e. measured before
wortmannin addition), Ymin is the final transport
rate reached after reversal of the insulin effect (at 50 min), K is the
rate constant of the reversal process, and t is the time elapsed after
wortmannin addition. The corresponding half-life is equal to
0.69/K.
|
Protocol of vesicle immunoisolation
Protein A-purified monoclonal anti-GLUT4 antibody (1F8) or a
corresponding amount of nonspecific antibodies (-globulins) was
coupled to acrylamide beads (Reacti-gel GF 2000, Pierce Chemical Co., Rockford, IL) at a concentration of 1 mg antibody/ml resin
according to the manufacturer’s instructions. Before use, the beads
were saturated with 1% BSA in PBS (134 mM NaCl, 2.6
mM KCl, 6.4 mM
Na2HPO4, and 1.46
mM KH2PO4, pH
7.4) for at least 30 min (at room temperature) and washed in PBS.
Intracellular membranes (LDM; 50 µg/sample) were incubated overnight
at 4 C with different mixtures of two batches of beads (one batch with
1F8 and one with -globulin as nonspecific antibody), corresponding
to the varying amounts of 1F8 indicated in the figure legends, in a
constant total bead volume of 20 µl. After this incubation, the beads
were spun down, the supernatant was taken for later analysis, the beads
were washed five times in PBS, and the adsorbed material was eluted
with electrophoresis sample buffer according to Laemmli (0.1
M Tris-HCl, 20% glycerol, and 2% SDS, pH 6.8), incubated
for 5 min at 95 C, cooled, and microcentrifuged. The supernatant from
the vesicle immunoadsorption assay and the immunoadsorbed extract were
subjected to immunoblot analysis as previously reported
(1).
GLUT4 quantification by photoaffinity labeling with
[3H](2-N-[4(1-azi-2,2,2-trifluoroethyl)benzoyl]-1,3-bis-(D-mannos-4-yloxy)propyl-2-amine)
([3H]ATB-BMPA)
The labeling and quantification of cell surface GLUT4 was
performed using the nonpermeant photoreactive bismannose compound
[3H]-ATB-BMPA in a protocol previously
described and extensively validated (1). In brief,
cardiomyocytes (5 mg protein/sample) were treated for 30 min at 37 C in
the absence (basal) or presence of insulin (10 nM), before
they were incubated with [3H]ATB-BMPA, (300
µCi; 60 µM, final concentration), and irradiated for 3
min with UV light to covalently attach the label to GLUT4. The cells
were then solubilized, and the glucose transporter was immunopurified
and quantified as previously described (18). To quantify
the total cellular amount of the transporter, the cells (treated with
or without insulin) were first permeabilized (to allow access for the
nonpermeant label to intracellular GLUT4 stores) with the pore-forming
toxin 
-hemolysin, as described in Ref. 19 , before being
subjected to the labeling procedure.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
When we immunoadsorbed LDM from nonstimulated cardiomyocytes with a
saturating amount of 1F8, a large proportion of GLUT4 (80–100%) as
well as GLUT1 (
65%) and markers of the general endosomal recycling
pathway, SCAMPs (55%), were recovered in the adsorbed fraction (not
shown); this indicates that all of these proteins are at least
partially colocalized in LDM. On the other hand, less 1F8 was required
to reach a semimaximal degree of GLUT4 adsorption than in the case of
GLUT1 or SCAMPs (Fig. 1); thus, with a
relatively low amount of 1F8 (0.18 µg), with which a substantial part
of GLUT4 (>50%) was precipitated, little or no GLUT1 or SCAMP was
detected in the adsorbed fraction (Fig. 1). The immunoadsorption curve
of another endosomal protein, the IGF-II receptor, was virtually
identical to that found with SCAMPS (not shown).
|
Insulin-dependent changes in the composition of intracellular
GLUT4-containing vesicles
We next studied the time-dependent changes induced by insulin in
the two vesicle pools defined above. As shown in Fig. 2A, 2 min after insulin addition the rate
of glucose transport was about 4-fold over basal, which corresponds to
approximately 40% of the maximal effect of the hormone measured after
30 min of treatment (a maximal level of insulin-stimulated glucose
transport was reached approximately 10 min after insulin addition; not
shown). The amounts of GLUT4 and GLUT1 in the plasma membrane were also
increased after 2 min of insulin treatment (Fig. 2B).
|
, C and D). To determine which
of the two aforementioned vesicle pools (pool 1, pool 2, or both) is
affected by insulin, the data illustrated in Fig. 2D were used to
quantify the respective amounts of GLUT4 and GLUT1 in these pools.
Insulin produced a rapid and eventually nearly complete depletion of
GLUT4 in pool 1 (Table 1). Similarly, the
amount of GLUT1 in pool 2 decreased by half in response to insulin,
showing that this pool is also acutely mobilized (Table 1). In
contrast, the GLUT4 signal in pool 2 did not decrease with 2 or 30 min
of insulin treatment (and in some experiments even tended to increase;
Table 1).
|
90%) of GLUT4 in the
adsorbed fraction (not shown), indicating a similar immunoreactivity of
GLUT4 vesicles in all groups. Taken together, these results indicate that pool 1 is a specific, rapidly recruitable, insulin-sensitive GLUT4 compartment, whereas the presumably endosomal GLUT1-containing pool 2 is also acutely mobilized by the hormone.
To assess whether the amount of GLUT4 leaving pool 1 after stimulation
with insulin accounts for the increase observed in the plasma membrane,
we directly quantified the content of GLUT4 at the cell surface and
compared it to the total cellular amount of the transporter by using
the nonpermeant bismannose photolabel
[3H]ATB-BMPA, as described in Materials
and Methods. As shown in Fig. 3, insulin increased the GLUT4 content of the plasma membrane about
4-fold, in line with previous observations using the same method
(1). In parallel samples, the intracellular GLUT4 pools
were made accessible to the label by permeabilizing basal and
insulin-stimulated myocytes with -hemolysin (thus resulting in the
labeling of both cell surface and intracellular GLUT4, i.e.
of total GLUT4). Under these conditions, the total amount of labeled
GLUT4 in basal cells was about 6 times higher than that in intact
(nonpermeabilized) cells (Fig. 3); in other words, in the nonstimulated
state, only 16% of total GLUT4 was present at the cell surface.
Insulin treatment raised the level of cell surface GLUT4 to 60%, but,
as expected, it did not alter the total GLUT4 content of the cells
(Fig. 3). These percentages are in good agreement with values obtained
by others with another method (18% and 61% in basal and
insulin-stimulated hearts, respectively) (12). This means
that insulin translocates 52% of the intracellular GLUT4 to the plasma
membrane (i.e. 60% minus 16%/100% to 16%). On the other
hand, the immunoadsorption experiments summarized in Tables 1 and 2
have shown that about 50% of GLUT4 contained in LDM (and virtually all
GLUT4 in pool 1) is depleted by the hormone. As the LDM fraction
contains a large proportion of the GLUT4 found in intracellular
membrane fractions (our unpublished observations), it appears that most
GLUT4 leaving pool 1 eventually reaches the cell surface.
|
|
)
is that GLUT4 acutely mobilized from pool 1 to the cell surface may
undergo rapid reendocytosis and would therefore be found in pool 2 soon
after insulin addition. Furthermore, if pool 2 is indeed endosomal in
nature, and pool 1 a more specific intracellular GLUT4 storage
site, one would expect that in the course of GLUT4 reinternalization
from the plasma membrane to the intracellular compartment(s), the
transporter would first reach pool 2 before being segregated to pool
1.
To test these two assumptions, we monitored the redistribution of GLUTs
from the cell surface to intracellular membranes, starting under
conditions where the transporters’ content in the plasma membrane is
high. For this purpose, we first stimulated cardiomyocytes with insulin
(which increases the amounts of GLUT4 and GLUT1 in the plasma membrane;
Fig. 2B) (1), then removed insulin and examined the
changes occurring in pools 1 and 2 over time.
As expected, withdrawal of insulin resulted in a progressive decrease
in the rate of glucose transport from an initial highly stimulated
level toward basal, nonstimulated values (Fig. 4A). Preliminary experiments showed that
the half-life of the insulin effect after the insulin wash-out was
about 10 min (not shown), in good agreement with values reported in
isolated adipocytes (20, 21). A similar value was obtained
when the insulin signal was blocked by adding the PI3K inhibitor
wortmannin to fully insulin-stimulated cells (Fig. 7, lower
curve), which indicates that the washing procedure was efficient
in removing insulin.
|
) and a simultaneous increase in the intracellular
membranes immunoadsorbed with 1F8 (Fig. 4, C and D). The redistribution
of GLUT4 (and GLUT1) from the cell surface to intracellular vesicles
(Table 2) was slower than the
transporters’ recruitment at the onset of insulin stimulation (Table 1). Furthermore, 12 min after insulin removal a larger part of the
total increase in GLUT4 (seen at 7 µg 1F8, i.e. from 0.46
to 0.73) is accounted for by an increase in pool 2 (from 0.40 to 0.56)
compared with that contributed by GLUT4 repletion of pool 1 (from 0.06
to 0.17; Table 2). Likewise, at this time point the GLUT4 content of
pool 2 had already reached a new steady state, i.e. faster
than in the case of pool 1. These observations indicate that GLUT4
internalization from the cell surface to intracellular pool 2 is slower
than the transporter’s exocytosis induced by insulin addition, so
reendocytosis of freshly recruited GLUT4 cannot explain why insulin
fails to decrease the net GLUT4 content of pool 2. Moreover, the
recycling of GLUT4 from the plasma membrane to pool 2 occurs before the
sorting of the transporter to pool 1, which is consistent with the idea
that pool 2 is an endosomal compartment, and pool 1 is a more
specialized entity.
Effect of a second insulin challenge after previous stimulation and
partial reversal of the insulin effect
In view of the slow rate of GLUT4 retranslocation from the cell
surface to intracellular vesicles (Table 2), it appears unlikely that
insulin’s failure to decrease the net amount of GLUT4 in pool 2 (Table 1) is due to a fast reendocytosis of the transporter. We therefore
tested whether insulin mobilizes GLUT4 from pool 2.
The results of the reversal experiments described above (Fig. 4 and
Table 2) offer an opportunity to examine this possibility by using the
following rationale. As pool 1 remains largely depleted of GLUT4 for at
least 12 min after insulin is removed, whereas the GLUT4 content of
pool 2 is high (Table 2), a new addition of insulin at this time point
should only result in a substantial increase in glucose transport if
this latter GLUT4 pool is sensitive to the hormone.
To assess this, cardiomyocytes were subjected to a first insulin
treatment (for 20 min), then washed and incubated in insulin-free
medium for another 12 min to partially reverse the effect of the
hormone before fresh insulin was added again. Under these conditions,
the extent of glucose transport stimulation on the second insulin
treatment was similar to that observed after the first insulin
challenge induced at time zero (Fig. 5, upper panel, 
, compare C and D vs. A and B) or
to the effect of a single insulin addition to naive cells at the same
time point (i.e. at min 32; Fig. 5, upper panel,
). In a parallel set of experiments, we quantified GLUT4 in the
intracellular vesicles pools under the same conditions. As illustrated
in Fig. 5 (lower panel), the second insulin addition (at
time point C) resulted in a decrease in the GLUT4 content of
intracellular membranes (as measured at time point D), which was
largely due to a recruitment of the transporter from pool 2; thus, this
second insulin challenge caused a GLUT4 decrement in pool 2 from
0.50 at point C to 0.34 at point D, whereas the content of the pool 1
was only diminished from 0.36 to 0.28 over the same period. These
studies hence show that, at least under conditions where pool 1 is
still incompletely repleted, insulin mobilizes GLUT4 from pool 2.
|
Exposure of cardiomyocytes to rotenone caused a nearly 3-fold increase
in glucose transport (Fig. 6A) with a
concomitant enhancement of GLUT4 and GLUT1 in the plasma membrane (Fig. 6B) and a reduction in GLUT4 and GLUT1 in the vesicles immunoadsorbed
with 1F8 (Fig. 6C). In contrast to what had been found in vesicles from
insulin-stimulated cells, rotenone produced a clear decrease in the
GLUT4 content of pool 2, whereas only a small (nonsignificant) change
was observed in pool 1 (Table 3); in
addition, rotenone reduced the GLUT1 signal in pool 2 (Table 3). Thus,
mobilization of pool 2 appears to be the predominant effect by which
GLUTs are recruited to the plasma membrane in response to the
mitochondrial inhibitor.
|
|
, wortmannin decreased the rate of
glucose transport. In the absence of rotenone, the half-time of this
process was 12.9 ± 1.3 min. In the presence of rotenone, however,
the reversal of the insulin-dependent glucose transport stimulation was
significantly slowed (t1/2 = 35.0 ± 9.5
min; P < 0.05 vs. value determined in the
absence of rotenone; Fig. 7). Similar results were obtained when the
insulin signal was stopped by washing out the hormone (not shown).
These results are compatible with a possible effect of the
mitochondrial inhibitor at the level of GLUT endocytosis or, at the
very least, with its acting on a transporter pool that is in close
functional relation with the plasma membrane pool. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In nonstimulated cardiomyocytes, intracellular GLUT4 is distributed between at least two types of vesicle, one type (pool 1) that can be immunoadsorbed with a relatively small amount of anti-GLUT4 antibodies (1F8) and contains virtually no GLUT1 or SCAMPs (markers of the general endosomal recycling pathway), and another type (pool 2) that includes GLUT1 and SCAMPs (as well as the IGF-II receptor, another endosomal marker) and is only recovered with large amounts of 1F8. Pool 1 makes up about 60% of the GLUT4 present in LDM in the absence of insulin (see basal values in all tables). Insulin caused a rapid and complete depletion of pool 1, indicating that this pool is a major insulin-sensitive GLUT4 storage compartment, whereas pool 2 serves as a source of GLUT1 in response to the hormone (the apparent lack of effect of insulin on GLUT4 in pool 2 will be discussed below). Upon insulin withdrawal, there is a gradual (and relatively slow) replenishment of pool 1 with GLUT4 and of pool 2 with GLUT1.
In contrast to insulin, rotenone only recruited pool 2 (both GLUT4 and
GLUT1) with no effect on pool 1. These experiments thus demonstrate
that pool 2 can be mobilized independently of pool 1. Rotenone’s
recruitment of only one transporter pool is consistent with the fact
that the mitochondrial inhibitor increases the rate of glucose
transport only about 3-fold, i.e. less potently than insulin
(10-fold), which mobilizes both pools (at least GLUT4 from pool 1
and GLUT1 from pool 2). Moreover, rotenone slowed the reversal of
insulin’s effect on glucose transport (upon wortmannin addition). This
action of rotenone may be explained by either a decrease in the rate of
GLUT endocytosis or an enhancement of the reexocytosis of freshly
internalized transporters. Although we have not measured the rates of
glucose transporter endocytosis and/or exocytosis in this study, the
latter possibility seems unlikely because the reinternalization of
glucose transporters upon insulin withdrawal is slow. In any instance,
these experiments indicate that rotenone acts on an intracellular
transporter pool that is in close functional connection with the plasma
membrane; as pool 2 was the only intracellular membrane found to be
sensitive to the mitochondrial inhibitor [neither pool 1, nor high
density microsomes (data not shown) were affected], these results
further substantiate the idea that pool 2 is an endosomal
compartment.
The presence of two GLUT4 pools, and the action of insulin can be
explained in at least two ways. In the first model (Fig. 8, model A), GLUT4 present in pool 1 is
not only physically, but also functionally, separated from the general
endosomal recycling system, i.e. it can be independently
(and directly) recruited to the cell surface by insulin. Pool 1 would
therefore represent a specialized GLUT4 storage compartment, possibly
similar to synaptic vesicles in nerve endings. In a second model (model
B in Fig. 8), GLUT4 is prevented by some sort of retention signal from
mixing up with and/or being recycled to the cell surface along with the
proteins of the general recycling pathway such as the transferring
receptor, GLUT1, etc. In this case the action of insulin
could be, for instance, 1) to remove this retention signal so that the
segregated GLUT4 reenters the endosomal pathway, and, in addition, 2)
to activate exocytosis from this pathway.
|
, model C). On the
one hand, the fact that GLUT4-rich, GLUT1-poor vesicles (pool 1) can be
isolated by immunoadsorption suggests that these membranes represent a
physically separated entity that is largely depleted by insulin. On the
other hand, the finding that the GLUT4 content of pool 2 does not
decrease in response to insulin, whereas GLUT1 is clearly recruited
from this pool is compatible with the hypothesis that GLUT4 mobilized
from pool 1 may at least in part be first redistributed to pool 2 on
its way to the cell surface. Although there are alternative
explanations for this observation, they appear less likely for the
following reasons. A first alternative hypothesis would be that GLUT4
directly recruited from pool 1 to the plasma membrane by insulin (Fig. 8, A) is reendocytosed and recycled to pool 2 extremely rapidly.
However, our studies in cardiomyocytes indicate that the rate of GLUT4
reinternalization is very low. This concurs with observations in
adipocytes where the GLUT4 endocytosis was demonstrated to be very slow
in both the unstimulated and insulin-stimulated states (20, 25). A second possibility to explain the lack of decrease in
GLUT4 from pool 2 is that GLUT4 in this pool is not recruitable by
insulin (in contrast to GLUT1; which would be a variant of model A).
However, the experiments illustrated in Fig. 5B show that if pool 1 is
largely depleted of GLUT4 and pool 2 is enriched in this transporter,
the addition of fresh insulin causes GLUT4 mobilization from pool 2.
Moreover, this second insulin challenge brings about a similar
stimulation of glucose transport as when insulin is added to naive
nonstimulated cells (in which pool 1 represents the major intracellular
GLUT4 fraction; Table 2).
Therefore, it is conceivable that in response to (a first) insulin
treatment, at least part of GLUT4 stemming from pool 1 may be fed into
pool 2 before reaching the plasma membrane (Fig. 8C). Translocation of
GLUT4 from pool 1 to the cell surface would thus comprise an
intermediary step. Interestingly, insulin was reported to alter the
physico-chemical properties of intracellular GLUT4 vesicles in
adipocytes (26), suggesting that the hormone does more
than simply promote their translocation to and fusion with the plasma
membrane.
It should be stressed, though, that the possible transfer of recruited GLUT4 from pool 1 to the GLUT1-containing pool 2 does not necessarily mean that both transporters go from here to the cell surface via the same pathway. Thus, GLUT4 and GLUT1 may not be completely mixed up in pool 2 (e.g. they may be localized to different subcompartments); alternatively, there could be some kind of sorting after the GLUTs have left pool 2 (i.e. on their way to the cell surface). In other words, if in cardiomyocytes incorporation of GLUT4 and GLUT1 into the plasma membrane eventually involves different mechanisms, there must be a specific sorting for both isoforms either at the level of pool 2 or at a later stage after recruitment of this pool has been initiated. Further studies will be required to clarify this issue.
Whatever model (A, B, or C, or perhaps a combination of these) may be true, the amount of GLUT4 leaving pool 1 in response to insulin is sufficient to explain the increase in this transporter observed at the cell surface, as determined in the experiments with ATB-BMPA (see Results).
In conclusion, our study confirms the existence of at least two intracellular GLUT4-containing vesicle populations in cardiomyocytes. More importantly, beside the GLUT4- enriched storage pool, which is a major target of insulin’s action, the endosomal pool also represents a site from which glucose transporters can be mobilized to the cell surface. Finally, our results are compatible with a model in which GLUT4 recruited from the storage pool by insulin may pass through an intermediary compartment on its way to the cell surface. These data offer a novel view of the traffic of glucose transporters in cardiomyocytes and open new perspectives to pharmacologically modulate glucose uptake in heart.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: ATPase, Adenosine triphosphatase; GLUT4, glucose transporter 4; LDM, low density microsomes; PM, plasma membrane; SCAMP, secretory carrier membrane protein; ATB-BMPA, 2-N-[4(1- azi-2,2,2-trifluoroethyl)benzoyl]-1,3-bis-(D-mannos-4-yloxy)propyl-2-amine.
Received May 9, 2001.
Accepted for publication August 22, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  R. W. Schwenk, J. J.F.P. Luiken, A. Bonen, and J. F.C. Glatz Regulation of sarcolemmal glucose and fatty acid transporters in cardiac disease Cardiovasc Res, July 15, 2008; 79(2): 249 - 258. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Shen, M.-J. Lin, A. Yaradanakul, V. Lariccia, J. A. Hill, and D. W. Hilgemann Dual control of cardiac Na+ Ca2+ exchange by PIP2: analysis of the surface membrane fraction by extracellular cysteine PEGylation J. Physiol., August 1, 2007; 582(3): 1011 - 1026. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Matsui, T. Nagoshi, E.-G. Hong, I. Luptak, K. Hartil, L. Li, N. Gorovits, M. J. Charron, J. K. Kim, R. Tian, et al. Effects of chronic Akt activation on glucose uptake in the heart Am J Physiol Endocrinol Metab, May 1, 2006; 290(5): E789 - E797. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Wang and M. W. Quick Trafficking of the Plasma Membrane {gamma}-Aminobutyric Acid Transporter GAT1: SIZE AND RATES OF AN ACUTELY RECYCLING POOL J. Biol. Chem., May 13, 2005; 280(19): 18703 - 18709. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Yang and G. D. Holman Insulin and Contraction Stimulate Exocytosis, but Increased AMP-activated Protein Kinase Activity Resulting from Oxidative Metabolism Stress Slows Endocytosis of GLUT4 in Cardiomyocytes J. Biol. Chem., February 11, 2005; 280(6): 4070 - 4078. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Rachdaoui and L. E. Nagy Endothelin-1-stimulated glucose uptake is desensitized by tumor necrosis factor-{alpha} in 3T3-L1 adipocytes Am J Physiol Endocrinol Metab, September 1, 2003; 285(3): E545 - E551. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Gavete, M. Agote, M. A. Martin, C. Alvarez, and F. Escriva Effects of Chronic Undernutrition on Glucose Uptake and Glucose Transporter Proteins in Rat Heart Endocrinology, November 1, 2002; 143(11): 4295 - 4303. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Zeigerer, M. A. Lampson, O. Karylowski, D. D. Sabatini, M. Adesnik, M. Ren, and T. E. McGraw GLUT4 Retention in Adipocytes Requires Two Intracellular Insulin-regulated Transport Steps Mol. Biol. Cell, July 1, 2002; 13(7): 2421 - 2435. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
) or treated with one of the following protocols. In
protocol 1 (
