Download Glucose Metabolism Rather Than Insulin Is a Main Determinant of

0021-972X/00/$03.00/0
The Journal of Clinical Endocrinology & Metabolism
Copyright © 2000 by The Endocrine Society
Vol. 85, No. 3
Printed in U.S.A.
Glucose Metabolism Rather Than Insulin Is a Main
Determinant of Leptin Secretion in Humans
PETER WELLHOENER, BERND FRUEHWALD-SCHULTES, WERNER KERN,
DORLE DANTZ, WOLFGANG KERNER, JAN BORN, HORST L. FEHM, AND
ACHIM PETERS
Departments of Internal Medicine I (P.W., B.F.-S., W.K., D.D., H.L.F., A.P.) and Clinical
Neuroendocrinology (J.B.), University of Lubeck, 23538 Lubeck; and Department of Diabetes and
Metabolical Disorders, Klinikum Karlsburg (W.K.), Karlsburg 17495, Germany
ABSTRACT
Circulating plasma insulin and glucose levels are thought to be
major regulators of leptin secretion. There is evidence from in vitro
and animal experiments that glucose metabolism rather than insulin
alone is a main determinant of leptin expression. Here, we tested the
hypothesis that in humans also leptin secretion is primarily regulated
by glucose uptake and only secondarily by plasma insulin and glucose.
In 30 lean and healthy men we induced 4 experimental conditions by
using the blood glucose clamp technique. A total of 60 hypoglycemic
and euglycemic clamps, lasting 6 h each, were performed. During
these clamps insulin was infused at either high (15.0 mU/min䡠kg) or
low (1.5 mU/min䡠kg) rates, resulting in low-insulin-hypo, high-insulin-hypo, low-insulin-eu, and high-insulin-eu conditions. Serum lep-
L
EPTIN, A PEPTIDE secreted by adipose tissue, is considered an important hormone of energy balance. The
diurnal rhythm of plasma leptin is entrained to meal timing
(1), suggesting that the meal-related increase in plasma insulin or glucose could be a major stimulus for leptin secretion. In fact, leptin levels have been shown to increase during
euglycemic hyperinsulinemic clamp experiments (2). The
changes in leptin levels during euglycemic clamps depend
on the rate of insulin infusion, indicating a dose-dependent
stimulatory effect of insulin on leptin secretion (3). However,
using the same rate of insulin infusion, during hypoglycemic
clamps the rise of leptin was attenuated compared to that
during euglycemic clamps (4). In this context, it should be
pointed out that during such clamp experiments the rate of
insulin infusion and the target blood glucose level determine
the glucose disposal rate, reflecting whole body glucose
uptake.
There is evidence from in vitro and animal experiments
that glucose metabolism rather than insulin alone is a main
determinant for leptin expression (5–7). Mueller et al. demonstrated that a competitive inhibition of glucose transport
by 2 deoxy-d-glucose, phloretin, and cytochalasin dose dependently decreased leptin secretion and messenger ribonucleic acid (RNA) content in cultured rat adipocytes (7).
These researchers concluded that stimulation of leptin seReceived August 11, 1999. Revision received November 19, 1999.
Accepted December 6, 1999.
Address all correspondence and requests for reprints to: Peter Wellhoener, M.D., Medical University Lubeck, Department of Internal Medicine I, 23538 Lubeck, Germany.
tin increased from 0 –360 min by 20.5 ⫾ 4.1% in the low-insulin-hypo,
33.6 ⫾ 7.6% in the high-insulin-hypo, 39.6 ⫾ 6.0% in the low-insulineu, and 60.4 ⫾ 7.6% in the high-insulin-eu condition. Multiple regression analysis revealed a significant effect of circulating insulin
(low vs. high insulin; P ⫽ 0.001) and blood glucose (hypoglycemia vs.
euglycemia; P ⫽ 0.001) on the rise of serum leptin. However, when the
total amount of dextrose infused during the clamp (grams of dextrose
per kg BW) was included into the regression model, this variable was
significantly related to the changes in serum leptin (P ⫽ 0.001),
whereas circulating insulin and glucose had no additional effect.
These findings in humans support previous in vitro data that leptin
secretion is mainly related to glucose metabolism. (J Clin Endocrinol
Metab 85: 1267–1271, 2000)
cretion by insulin is probably not due to a direct effect of
insulin, but is secondary to its effect to stimulate glucose
uptake in adipocytes. Furthermore, Mizuno et al. were able
to show that plasma glucose significantly correlated with
leptin messenger RNA in lean mice and that glucose and
insulin enhanced leptin messenger RNA in lean animals (6).
On this background, we hypothesized that in humans also
primarily glucose uptake regulates leptin secretion and that
insulin only serves as a permissive factor facilitating glucose
uptake.
Subjects and Methods
Subjects
Experiments were carried out in 30 lean healthy men [mean ⫾ sem
age, 25.7 ⫾ 0.8 yr; range, 22–32 yr; body mass index (BMI), 22.9 ⫾ 0.4
kg/m2; range, 18.6 –26.0 kg/m2]. Exclusion criteria were chronic or acute
illness, current medication of any kind, smoking, alcohol or drug abuse,
and diabetes or hypertension in first degree relatives. Each volunteer
gave written informed consent, and the study was approved by the local
ethics committee.
Study design
To achieve 4 experimental conditions with different glucose disposal rates, i.e. whole body glucose uptake, we varied either the rate
of insulin infusion (high insulin vs. low insulin) or the target blood
glucose (euglycemic vs. hypoglycemic) in a total of 60 clamp experiments. Experiments lasting 6 h each were carried out in 30 healthy
men randomly assigned to 2 different groups (each of 15 persons).
Every subject participated in a series of 2 clamp sessions that differed
in glucose target levels (hypoglycemia and euglycemia), with the
order of sessions balanced across subjects. Both clamp sessions were
separated by at least 4 weeks. Insulin infusion rate during the clamps
was 1.5 mU/min䡠kg (low) in 1 group of 15 subjects and 15.0 mU/
1267
1268
JCE & M • 2000
Vol 85 • No 3
WELLHOENER ET AL.
min䡠kg (high) in another group of 15 subjects. Thus, the 4 conditions
included a hypoglycemic clamp with a low rate of insulin infusion
(low-insulin-hypo), a hypoglycemic clamp with a high rate of insulin
infusion (high-insulin-hypo), an euglycemic clamp with a low rate of
insulin infusion (low-insulin-eu), and an euglycemic clamp with a
high rate of insulin infusion (high-insulin-eu).
Clamp procedure
After a 10-h overnight fast subjects arrived at our medical research
unit at 0800 h. They were seated in a comfortable position, and a venous
cannula was inserted into a vein on the back of the right hand. This hand
was kept in a heated box (50 –55 C) to obtain arterialized blood. After
a 1-h baseline period, insulin (Hoechst, Frankfurt, Germany) was infused via a second cannula in the left antecubital vein at a continuous
flow of either 1.5 or 15.0 mU/min䡠kg depending on the experimental
condition. Simultaneously a 20% dextrose solution was infused at variable rates to control serum glucose. Arterialized blood was drawn every
5 min to measure blood glucose (Beckman glucose analyzer, Munich,
Germany). During euglycemic clamps serum glucose was held stable at
5.0 –5.5 mmol/L. During the stepwise hypoglycemic clamps we reduced
serum glucose to achieve plateaus of 4.2, 3.7, 3.0, and 2.3 mmol/L.
Plateaus were held for 45 min; the time to achieve the next lower plateau
was set at 45 min.
Measurements
Blood for leptin and insulin determinations was collected every 30
min. Serum was kept at ⫺20 C until analysis. In the high insulin group
potassium was controlled every 30 min and substituted orally when
below 4.0 mmol/L. The glucose infusion rate was integrated over time
to calculate the total amount of dextrose infused (in grams per kg BW).
RIAs were used to measure insulin [Pharmacia Biotech, Uppsala, Sweden; interassay coefficient of variation (CV), 7.5%; intraassay CV, 5.4%]
and leptin (Linco Research, Inc., St. Louis, MO; interassay CV, 6.1%;
intraassay CV, 5.4%).
Statistical analysis
Data are reported as the mean ⫾ sem. P ⬍ 0.05 was considered
significant. Because of greater interindividual variability in baseline
leptin levels, changes in serum leptin during the clamp were expressed
as ⌬ leptin (%). ⌬ leptin (%) was calculated as the difference between the
levels at 360 and 0 min divided by the level at 0 min. ANOVA for
repeated measures on the changes in serum leptin were performed
across all four conditions, including the factors insulin (high vs. low
insulin infusion rate) and blood glucose (euglycemic vs. hypoglycemic
condition). We also calculated the ratio between the changes in serum
leptin from 0 –360 min (percentage) and the total amount of dextrose
(grams per kg) that was infused during the whole clamp. ANOVA for
repeated measures performed on these ratios was also performed across
all four conditions. For further analysis we applied multiple linear
regression of serum leptin at 360 min to the total amount of infused
dextrose and covariates. Leptin at baseline (0 min) was included regardless of statistical significance. The variables insulin (high vs. low
insulin infusion rate), blood glucose (euglycemic vs. hypoglycemic condition), BMI, and insulin resistance were conditionally selected in a
forward stepwise procedure (inclusion criteria, P ⬍ 0.05). Insulin resistance was calculated using the homeostasis model assessment (HOMA)
of fasting insulin and glucose levels (8). Data were analyzed using the
SPSS, Inc. statistical program (version 6.1, SSPS, Inc., Chicago, IL).
Results
Figure 1 shows blood glucose and insulin levels during
hypoglycemic and euglycemic clamps. Blood glucose concentrations were equal in the low and high insulin conditions
throughout all sessions. In the high insulin condition, however, serum insulin concentrations 60 min after the clamps
were started were approximately 40-fold higher than those
in the low insulin condition. They were 622 ⫾ 32 pmol/L in
the low-insulin-hypo, 23,624 ⫾ 1,587 pmol/L in the high-
insulin-hypo, 543 ⫾ 34 pmol/L in the low-insulin-eu, and
24,029 ⫾ 1,595 pmol/L in the high-insulin-eu condition. The
total amount of dextrose per kg BW infused during the
clamps was 2.00 ⫾ 0.17 g/kg in the low-insulin-hypo, 3.05 ⫾
0.13 g/kg in the high-insulin-hypo, 3.66 ⫾ 0.17 g/kg in the
low-insulin-eu, and 4.61 ⫾ 0.12 g/kg in the high-insulin-eu
condition.
During the clamps, serum leptin concentrations increased
from 0 –360 min by 20.5 ⫾ 4.1% in the low-insulin-hypo,
33.6 ⫾ 7.6% in the high-insulin-hypo, 39.6 ⫾ 6.0% in the
low-insulin-eu, and 60.4 ⫾ 7.6% in the high-insulin-eu condition. Thus, the changes in serum leptin were distinctly
greater in high than low insulin conditions [effect of insulin:
F(1,28) ⫽ 6.48; P ⫽ 0.017] and also greater in the euglycemic
than hypoglycemic conditions [effect of blood glucose:
F(1,28) ⫽ 5.37; P ⫽ 0.028]. The ratio between changes in leptin
and the total amount of glucose infused was remarkably
constant across all four conditions: 10.6 ⫾ 2.1, 12.9 ⫾ 2.4,
11.8 ⫾ 1.9, and 12.9 ⫾ 2.1% g⫺1 kg [effect of insulin: F(1,28) ⫽
0.58; P ⫽ 0.452; effect of blood glucose: F(1,28) ⫽ 0.00; P ⫽
0.962], suggesting that changes in leptin during the clamp
were directly related to the total amount of dextrose infused
rather than to circulating insulin and glucose levels (Fig. 2).
We applied multiple regression analysis to further assess the
effect of each variable on the rise of serum leptin. The regression model on the rise in serum leptin showed a significant effect of the total amount of infused dextrose (P ⫽
0.001), whereas other covariates, such as insulin (high vs. low
insulin), blood glucose (hypoglycemia vs. euglycemia), BMI,
and HOMA insulin resistance, were nonsignificant (Table 1).
To confirm the effect of the total amount of infused dextrose,
we removed this variable from an alternate regression model
(Table 2); in this model the effects of both glucose (P ⫽ 0.001)
and insulin (P ⫽ 0.001) were significant.
Discussion
In agreement with results of previous studies, the present
data demonstrate a dose-dependent increase in serum leptin
levels during infusion of insulin at different rates (3, 9). Also,
as previously shown, the rise in serum leptin was blunted
during hypoglycemic compared to euglycemic clamp conditions despite identical rates of insulin infusion (4). Circulating insulin and glucose, therefore, appeared to have a
stimulatory effect on leptin secretion. However, both circulating insulin and glucose levels during the clamps also influenced the glucose disposal rate and hence the amount of
dextrose infused. When these effects on glucose disposal, i.e.
glucose uptake, were considered in the analysis of the data,
it became obvious that the effects of insulin and blood glucose on serum leptin concentrations could be explained by
their effect on glucose uptake. These findings suggest that the
effects of insulin and blood glucose on serum leptin concentrations might be reduced to their effects on glucose uptake,
which may be a single factor to modulate leptin secretion.
We designed our study to detect insulin effects on serum
leptin by using different levels of hyperinsulinemia. Increasing insulin levels to 500 pmol/L, as we did in the low insulin
group, will virtually eliminate hepatic glucose production
(10). Given an inhibited glucose production by the liver the
GLUCOSE METABOLISM AND LEPTIN SECRETION
1269
FIG. 1. Serum insulin levels and blood
glucose during hypoglycemic and euglycemic clamp experiments. The insulin infusion rate in the low insulin group was
1.5 mU/min䡠kg; that in the high insulin
group was 15.0 mU/min䡠kg. Open symbols represent clamps with low insulin
infusion rates; solid symbols represent
clamps with high insulin infusion rates.
Triangles are hypoglycemic clamp conditions; circles are euglycemic clamp conditions.
dextrose infusion rate in our experiments approximately reflects the whole body glucose uptake. Although at this level
of hyperinsulinemia the effect on glucose production is complete, glucose utilization is not saturated. It takes a high
arterial plasma insulin over 2500 pmol/L to saturate glucose
uptake under euglycemic conditions (11). To examine insulin
effects under conditions of a complete blockade of hepatic
glucose output, we compared two levels of hyperinsulinemia, i.e. approximately 500 and 25,000 pmol/L. To detect an
insulin dose-response relationship on receptor and postreceptor effects, Koltermann et al. used the hyperinsulinemic
clamp technique in humans with high doses of hyperinsu-
linemia (12, 13). Here we used the same method with similar
doses of insulin to study a dose-response relationship on
serum leptin concentrations. The dose-response relationship
between the serum insulin level and whole body glucose
uptake is a sigmoid curve. Thus, an effect of changes in
insulin in the lower range can be expected to be more pronounced than the effect we observed here with high concentrations. To be able to detect smaller effects, a larger
sample (including a total of 60 clamp sessions) of a homogeneous group was investigated.
During the hypoglycemic clamps, the counterregulatory
increase in cortisol and epinephrine levels could potentially
1270
JCE & M • 2000
Vol 85 • No 3
WELLHOENER ET AL.
FIG. 2. Scatterplot with bidirectional
error bars. The ⌬ leptin (%) at time t vs.
the total amount of infused dextrose
during the experiment until time t.
Open symbols represent clamps with
low insulin concentrations; solid symbols represent clamps with high insulin
concentrations. Triangles are hypoglycemic clamp conditions; circles are euglycemic clamp conditions. The relations of ⌬ leptin (%) vs. infused dextrose
are virtually identical despite different
insulin concentrations and different
levels of glycemia.
TABLE 1. Multiple linear regression of serum leptin at 360 min on the total amount of infused dextrose and covariates
Independent variables
1. Step
a
2. Step
Regression coefficient
SE
P value
Leptin at baseline
1.175
0.081
0.000
Leptin at baseline
Total amount of infused dextrose
1.205
7.757E-05
0.074
2.284E-05
0.000
0.001
Leptin at baseline was included regardless of statistical significance. The variable total amount of infused dextrose (at 360 min) and additional
covariates were conditionally selected in a forward stepwise procedure. The nonsignificant variables that were not included were blood glucose
(hypo- vs. euglycemia), insulin (low vs. high insulin infusion), BMI, and HOMA insulin resistance.
a
Multiple r2 ⫽ 0.822.
TABLE 2. Multiple linear regression of serum leptin after 360 min on covariates (without total amount of infused dextrose)
Independent variables
Regression coefficient
SE
P value
1. Step
Leptin at baseline
1.175
0.081
0.000
2. Step
Leptin at baseline
Blood glucose
1.171
0.783
0.074
0.247
0.000
0.002
3. Stepa
Leptin at baseline
Blood glucose
Serum insulin
1.191
0.782
0.779
0.069
0.227
0.228
0.000
0.001
0.001
Again, leptin at baseline was included regardless of statistical significance. The nonsignificant variables that were not included were BMI
and HOMA insulin resistance.
a
Multiple r2 ⫽ 0.849.
have influenced leptin secretion. For instance, epinephrine
has been shown to inhibit leptin secretion (14 –16), whereas
glucocorticoids, e.g. dexamethasone, have been shown to
stimulate leptin secretion in humans (17–19). However, compared to the euglycemic clamps, the increase in leptin levels
during the hypoglycemic clamps was already blunted after
120 min, i.e. when blood glucose was approximately 4.0
mmol/L. As this level of blood glucose was unlikely to have
stimulated epinephrine and cortisol secretion (20), an effect
of these counterregulatory hormones on the early differences
in leptin levels between the hypoglycemic and euglycemic
clamp conditions can be excluded.
Glucose uptake by skeletal muscles and adipose tissue
decreases by 60 –70% during insulin-induced hypoglycemia.
This decline is a counterregulatory mechanism that allows
shunting of glucose to more important organs, e.g. the brain
(21, 22). As shown in Fig. 2, the relation between serum leptin
and the amount of dextrose infused was virtually identical
during hypoglycemic and euglycemic clamp conditions.
Thus, it is unlikely that the counterregulatory hormones have
exerted a major effect on serum leptin that has not been
mediated by glucose uptake.
In interpreting the results one limitation of our study is
that two of the experimental conditions (low-insulin-hypo
and high-insulin-hypo) were not strictly at steady state, so
the glucose infusion rate may not have accurately reflected
whole body glucose disposal or glucose utilization. A second
limitation was that we did not directly measure glucose
uptake into adipose tissue, which is known to be the major
source of leptin production. It seems reasonable, however, to
GLUCOSE METABOLISM AND LEPTIN SECRETION
assume that during hyperinsulinemic clamps in healthy individuals glucose uptake into adipose tissue will parallel
whole body glucose uptake.
One potential mechanism by which glucose uptake could
regulate leptin secretion has been identified by Wang et al.,
showing that increased tissue concentrations of the end product of the hexosamine biosynthetic pathway, UDP-N-acetylglucosamine, results in rapid and marked increases in leptin
messenger RNA and protein levels (23). The hexosamine
biosynthetic pathway is a cellular sensor of energy availability and mediates the effect of glucose on the expression
of several gene products (24, 25). Thus, it seems possible that
in the present study infusion of insulin and glucose have
increased glucose uptake into the adipose tissue, causing an
accumulation of UDP-N-acetylglucosamine, which then
stimulated leptin secretion.
In summary, the amount of infused dextrose closely paralleled the rise of serum leptin in all experiments. Throughout all conditions, the increase in leptin per amount of infused dextrose was constant and independent of blood
glucose and insulin levels. Therefore, our findings in humans
support previous data from in vitro and animal experiments
that leptin secretion is mainly related to glucose metabolism.
Acknowledgments
We thank Christiane Zinke and Steffi Baxmann for their expert and
invaluable laboratory assistance, and Anja Otterbein for her organizational work. We gratefully thank Dr. Thomas Kohlmann for his methodological advice.
References
1. Schoeller DA, Cella LK, Sinha MK, Caro JF. 1997 Entrainment of the diurnal
rhythm of plasma leptin to meal timing. J Clin Invest. 100:1882–1887.
2. Boden G, Chen X, Kolaczynski JW, Polansky M. 1997 Effects of prolonged
hyperinsulinemia on serum leptin in normal human subjects. J Clin Invest.
100:1107–1113.
3. Saad MF, Kahn A, Sharma K, et al. 1998 Physiological insulinemia acutely
modulates plasma leptin. Diabetes. 47:544 –549.
4. Schmitz O, Fisker S, Orskov L, Hove KY, Nyholm B, Moller N. 1997 Effects
of hyperinsulinaemia and hypoglycaemia on circulating leptin levels in
healthy lean males. Diabetes Metab. 23:80 – 83.
5. Mizuno T, Bergen H, Kleopoulos S, Bauman WA, Mobbs CV. 1996 Effects
of nutritional status and aging on leptin gene expression in mice: importance
of glucose. Horm Metab Res. 28:679 – 684.
6. Mizuno TM, Bergen H, Funabashi T, et al. 1996 Obese gene expression:
reduction by fasting and stimulation by insulin and glucose in lean mice, and
persistent elevation in acquired (diet-induced) and genetic (yellow agouti)
obesity. Proc Natl Acad Sci USA. 93:3434 –3438.
1271
7. Mueller WM, Gregoire FM, Stanhope KL, et al. 1998 Evidence that glucose
metabolism regulates leptin secretion from cultured rat adipocytes. Endocrinology. 138:551–558.
8. Haffner SM, Miettinen H, Stern MP. 1997 The homeostasis model in the San
Antonio Heart Study. Diabetes Care. 20:1087–1092.
9. Tuominen JA, Ebeling P, Laquier FW, Heiman ML, Stephens T, Koivisto VA.
1997 Serum leptin concentration and fuel homeostasis in healthy man. Eur
J Clin Invest. 27:206 –211.
10. Cherrington AD. 1994 The acute regulation of hepatic glucose production. In:
Pagliassotti M, Davis S, Cherrington AD, eds. The role of the liver in maintaining glucose homeostasis. Austin: Landes; 19 –35.
11. Flakoll P, Carlson MG, Cherrington AD. 1996 Physiologic action of insulin.
In: LeRoith D, Taylor I, Olefsky J, eds. Diabetes mellitus. Philadelphia: Lippincott-Raven; 121–131.
12. Kolterman OG, Gray RS, Griffin J, et al. 1981 Receptor and postreceptor
defects contribute to the insulin resistance in noninsulin-dependent diabetes
mellitus. J Clin Invest. 68:957–969.
13. Kolterman OG, Insel J, Saekow M, Olefsky JM. 1980 Mechanism of insulin
resitance in human obesity. Evidence for a receptor and postreceptor defect.
J Clin Invest. 65:1272–1284.
14. Mantzoros CS, Qu D, Frederich RC, et al. 1996 Activation of ␤(3) adrenergic
receptors suppresses leptin expression and mediates a leptin-independent
inhibition of food intake in mice. Diabetes. 45:909 –914.
15. Gettys TW, Harkness PJ, Watson PM. 1996 The ␤3-adrenergic receptor inhibits
insulin-stimulated leptin secretion from isolated rat adipocytes. Endocrinology 137:4054 – 4057.
16. Li H, Matheny M, Scarpace PJ. 1997 ␤3-Adrenergic-mediated suppression of
leptin gene expression in rats. Am J Physiol. 272:E1031—E1036.
17. Papaspyrou Rao S, Schneider SH, Petersen RN, Fried SK. 1997 Dexamethasone increases leptin expression in humans in vivo. J Clin Endocrinol Metab.
82:1635–1637.
18. Miell JP, Englaro P, Blum WF. 1996 Dexamethasone induces an acute and
sustained rise in circulating leptin levels in normal human subjects. Horm
Metab Res. 28:704 –707.
19. Dagogo-Jack S, Selke G, Melson AK, Newcomer JW. 1997 Robust leptin
secretory responses to dexamethasone in obese subjects. J Clin Endocrinol
Metab. 82:3230 –3233.
20. Mitrakou A, Ryan C, Veneman T, et al. 1991 Hierarchy of glycemic thresholds
for counterregulatory hormone secretion, symptoms, and cerebral dysfunction. Am J Physiol. 260:E67–E74.
21. Capaldo B, Napoli R, Guida R, et al. 1995 Forearm muscle insulin resistance
during hypoglycemia: role of adrenergic mechanisms and hypoglycemia per
se. Am J Physiol. 268:E248 —E254.
22. Cohen N, Rossetti L, Shlimovich P, Halberstam M, Hu M, Shamoon H. 1995
Counterregulation of hypoglycemia. Skeletal muscle glycogen metabolism
during three hours of physiological hyperinsulinemia in humans. Diabetes.
44:423– 430.
23. Wang J, Liu R, Hawkins M, Barzilai N, Rossetti L. 1998 A nutrient-sensing
pathway regulates leptin gene expression in muscle and fat. Nature.
393:684 – 688.
24. Marshall S, Bacote V, Traxinger RR. 1991 Discovery of a metabolic pathway
mediating glucose-induced desensitization of the glucose transport system.
Role of hexosamine biosynthesis in the induction of insulin resistance. J Biol
Chem. 266:4706 – 4712.
25. Traxinger RR, Marshall S. 1991 Coordinated regulation of glutamine:fructose6-phosphate amidotransferase activity by insulin, glucose, and glutamine. Role
of hexosamine biosynthesis in enzyme regulation. J Biol Chem. 266:10148 –
10154.