A free-choice high-fat high-sugar diet induces glucose intolerance and insulin unresponsiveness to a glucose load not explained by obesity.

Abstract

In diet-induced obesity, it is not clear whether impaired glucose metabolism is caused directly by the diet, or indirectly via obesity. This study examined the effects of different free-choice, high-caloric, obesity-inducing diets on glucose metabolism. In these free-choice diets, saturated fat and/or a 30% sugar solution are provided in an addition to normal chow pellets.
In the first experiment, male rats received a free-choice high-fat high-sugar (HFHS), free-choice high-fat (HF) or a chow diet. In a second experiment, male rats received a free-choice high-sugar (HS) diet or chow diet. For both experiments, after weeks 1 and 4, an intravenous glucose tolerance test was performed.
Both the HFHS and HF diets resulted in obesity with comparable plasma concentrations of free fatty acids. Interestingly, the HF diet did not affect glucose metabolism, whereas the HFHS diet resulted in hyperglycemia, hyperinsulinemia and in glucose intolerance because of a diminished insulin response. Moreover, adiposity in rats on the HF diet correlated positively with the insulin response to the glucose load, whereas adiposity in rats on the HFHS diet showed a negative correlation. In addition, total caloric intake did not explain differences in glucose tolerance. To test whether sugar itself was crucial, we next performed a similar experiment in rats on the HS diet. Rats consumed three times as much sugar when compared with rats on the HFHS diet, which resulted in obesity with basal hyperinsulinemia. Glucose tolerance, however, was not affected.
Together, these results suggest that not only obesity or total caloric intake, but the diet content also is crucial for the glucose intolerance that we observed in rats on the HFHS diet.

Full text

Page 1

ORIGINAL ARTICLE
A free-choice high-fat high-sugar diet induces
glucose intolerance and insulin unresponsiveness
to a glucose load not explained by obesity
SE la Fleur1,2, MCM Luijendijk1, AJ van Rozen1, A Kalsbeek2,3and RAH Adan1
1Department of Neuroscience and Pharmacology, Rudolf Magnus Institute of Neuroscience, University Medical Center
Utrecht, Utrecht, The Netherlands;2Department of Endocrinology and Metabolism, Academic Medical Center, University
of Amsterdam, Amsterdam, The Netherlands and3Netherlands Institute for Neurosciences, Amsterdam, The Netherlands
Objectives: In diet-induced obesity, it is not clear whether impaired glucose metabolism is caused directly by the diet, or
indirectly via obesity. This study examined the effects of different free-choice, high-caloric, obesity-inducing diets on glucose
metabolism. In these free-choice diets, saturated fat and/or a 30% sugar solution are provided in an addition to normal chow
pellets.
Method: In the first experiment, male rats received a free-choice high-fat high-sugar (HFHS), free-choice high-fat (HF) or a chow
diet. In a second experiment, male rats received a free-choice high-sugar (HS) diet or chow diet. For both experiments, after
weeks 1 and 4, an intravenous glucose tolerance test was performed.
Results: Both the HFHS and HF diets resulted in obesity with comparable plasma concentrations of free fatty acids. Interestingly,
the HF diet did not affect glucose metabolism, whereas the HFHS diet resulted in hyperglycemia, hyperinsulinemia and in
glucose intolerance because of a diminished insulin response. Moreover, adiposity in rats on the HF diet correlated positively
with the insulin response to the glucose load, whereas adiposity in rats on the HFHS diet showed a negative correlation.
In addition, total caloric intake did not explain differences in glucose tolerance. To test whether sugar itself was crucial, we next
performed a similar experiment in rats on the HS diet. Rats consumed three times as much sugar when compared with rats on
the HFHS diet, which resulted in obesity with basal hyperinsulinemia. Glucose tolerance, however, was not affected.
Conclusion: Together, these results suggest that not only obesity or total caloric intake, but the diet content also is crucial for
the glucose intolerance that we observed in rats on the HFHS diet.
International Journal of Obesity (2011) 35, 595–604; doi:10.1038/ijo.2010.164; published online 17 August 2010
Keywords: saturated fat; liquid sugar; diet-induced obesity; free fatty acids
Introduction
The worldwide prevalence of obesity is increasing dramati-
cally. Obesity is associated with insulin resistance, dyslipi-
demia and hypertension. This combination of symptoms,
referred to as metabolic syndrome, increases the risk of
developing type 2 diabetes mellitus.1The consumption of
sugar-sweetened beverages and saturated fat has been linked
to risks for obesity and diabetes. In recent decades, the intake
of sugar-sweetened beverages has clearly increased around
the world,2and from a recent survey among children,
consumption of both sugar-sweetened beverages and satu-
rated fat (especially from snack foods) exceed recommended
daily levels.3It is, therefore, important to understand how
consumption of dietary fat and sugar (in solution) con-
tributes to the development of obesity and type 2 diabetes
mellitus.
Several rodent models of obesity and insulin resistance
have been used to study the role of diet composition in
the development of obesity and insulin resistance. Rats that
are fed diets with added saturated fat and/or sugar for a
prolonged period of time develop obesity and insulin
resistance.4–6In rodents on a high-fat diet, insulin resistance
is followed by an insufficient compensation of the b-cell,
resulting in glucose intolerance.7,8The obesity in these
models is characterized by elevated concentrations of
circulating lipids, and it has been suggested that increased
circulating concentrations of free fatty acids (FFAs) and
Received 8 March 2010; revised 5 July 2010; accepted 7 July 2010; published
online 17 August 2010
Correspondence: Dr SE la Fleur, Department
Metabolism, F5-165, Academic Medical Center, University of Amsterdam,
Meibergdreef 9, Amsterdam 1105 AZ, The Netherlands.
E-mail: s.e.lafleur@amc.uva.nl
of Endocrinology and
International Journal of Obesity (2011) 35, 595–604
& 2011 Macmillan Publishers Limited All rights reserved 0307-0565/11
www.nature.com/ijo

Page 2

triglycerides mediate the development of insulin resistance
and glucose intolerance.9–11On the other hand, it has been
proposed that alterations in the brain and its projections to the
autonomic nervous system (ANS) could have an important
role in diet-induced alterations in glucose metabolism. For
example, Cruciani-Gluglielmacci12showed that a lard-based
high-fat diet in young Wistar rats resulted in rapid changes
in insulin secretion and action. These changes were not
explained by changes in the levels of plasma lipids, but instead
explained by changes in the sympathetic tone, and changes in
norepinephrine turnover in the hypothalamus, an area in the
brain important for the regulation of glucose metabolism.
Furthermore, Levin13showed reduced norepinephrine turn-
over in the brains of rats that were prone to develop obesity
before any changes in body weight. Thus, changes in the
hypothalamus could be an important mediator of diet-
induced alterations in glucose metabolism in addition to the
known effects of increased concentrations of plasma lipids.
Recently, we showed that rats consuming a diet consisting
of a choice between saturated fat, a 30% sugar solution or
regular pellet chow (a free-choice high-fat high-sugar (HFHS)
diet) become rapidly obese and, within a week, show
increased plasma glucose concentrations (without changes
in insulin), suggesting the development of glucose intoler-
ance and insulin resistance.14Moreover, rats on the HFHS
diet showed alterations in the level of the hypothalamus.
We observed that within the arcuate nucleus of the hypo-
thalamus, the expression of neuropeptide Y increased and
that of proopiomelanocortin decreased when rats were fed
a free-choice HFHS diet.15Because neuropeptide Y and
proopiomelanocortin are known to regulate glucose metabo-
lism,16,17we hypothesized that the HFHS diet would result in
glucose intolerance. Interestingly, rats on a free-choice high-
fat (HF) diet or on a free-choice high-sugar (HS) diet also
developed obesity. However, at the level of the hypothalamus,
changes were either opposite to that seen for rats on a free-
choice HFHS diet (in rats on HF diet) or not present (in rats on
HS diet).15Therefore, we hypothesize that although rats on all
different choice diets become obese, only rats on a free-choice
HFHS-choice diet become glucose intolerant.
To investigate this hypothesis, we again subjected rats to
the different free-choice diets, that is, HFHS, HF, HS or
regular chow. First, we compared intravenous glucose
tolerance in rats on a free-choice HFHS diet, on HF diet or
chow only. Because rats on HFHS diet showed clear glucose
intolerance, whereas rats on HF diet did not, we investigated
whether the consumption of the sugar solution explained
the differences between HFHSand HF diets, and measured
glucose tolerance in rats on the HS diet as well.
Materials and methods
Male rats (Wistar; Charles River, Sulzfeld, Germany) were
individually housed in Plexiglas cages in a temperature-
controlled(21–231C)andlight-controlled(lightson
between 0700 and 1900h) room. Rats were allowed to adapt
to their new environment for at least 5 days. All rats had ad
libitum access to pelleted rat chow (Special Diet Service (SDS),
Essex, UK) and tap water. All experiments were approved
by the Committee for Animal Experimentation of the
University Medical Center Utrecht, The Netherlands.
Experimental design
All rats were implanted with intra-arterial silicone catheters
through the right jugular vein, according to the method of
Steffens,18when they had reached a body weight of 4300g.
Experiment 1.
were switched to (1) a free-choice HFHS diet: a dish
of saturated fat (beef tallow (Ossewit/Blanc de Boeuf),
Vandermoortele, Belgium) and a bottle of 30% sugar water
(1.0 M sucrose mixed from commercial grade sugar and
water) were present in the cage, in addition to their standard
pellet chow and water bottle, or (2) a free-choice HF diet: a
dish of saturated fat was present in the cage in addition to
standard pellet chow and water bottle or (3) only standard
pellet chow and water.
At 2 weeks after surgery, rats (n¼6 per group)
Experiment 2.
free-choice HS diet: a bottle of 30% sugar water was present
in the cage, in addition to their standard pellet chow and
water bottle or (2) only standard pellet chow and water.
Rats (n¼5 per group) were switched to (1) a
For both experiments.
was measured 5 days per week over the first 2 weeks, and
thereafter twice a week. Food and water were refreshed twice
a week. An intravenous glucose tolerance test was performed
at 1 week and at 4 weeks after the start of the different diets:
animals were permanently connected to the blood-sampling
catheter, which was attached to a metal collar and kept out
of reach of the rats using a counterbalanced beam. This
allowed all manipulations to be performed outside the cages
without handling the animals. On the day of an experiment,
rats were fasted for 9h (from lights on at 0700h) before
the glucose infusion. At 1600h in the afternoon, a glucose
solution (1000mgkg–1body wt (in max 0.5ml saline)) was
injected as a bolus via the jugular vein catheter. A blood
sample (0.2ml) was collected (t¼0), immediately followed
by the glucose injection. Subsequently, blood samples
(0.2ml) were taken at 5, 15, 10, 20, 30 and 60min. Samples
were used to determine plasma concentrations of glucose
and insulin at these time points. The total amount of glucose
in plasma and the total amount of insulin released after a
glucose bolus injection was calculated from the area under
the curve (AUC) of every individual rat and averaged for the
experimental groups.
Two days after the last sampling day at week 4, all rats
were killed between 0900 and 1000h by decapitation within
10s after they had been taken from their home cages. Trunk
blood was collected to measure plasma leptin and FFA
Animals were weighted and food
Dietary fat, sugar and glucose metabolism
SE la Fleur et al
596
International Journal of Obesity

Page 3

concentrations. Individual mesenteric, epididymal, subcuta-
neous (inguinal) and perirenal white adipose tissues at the
left side were dissected, cleaned and weighed.
Analytical methods
Blood samples were immediately chilled on ice and centri-
fuged at 41C. The plasma was then stored at ?201C until
further analysis. Plasma glucose concentrations were deter-
mined using a Glucose/GOD-Perid method (Boehringer
Mannheim, Mannheim, Germany). Plasma immunoreactive
concentrations of leptin and insulin were determined using
a radioimmunoassay kit (Linco Research, St Charles, MI,
USA); samples were assayed in duplicate. As described in the
procedures of the assay, the amounts of sample, standards,
label, antibody and precipitating reagent were divided by
four. Plasma FFA was measured in triplicate using an
Acyl-CoA synthetase-acyl-CoA oxidase method (FFA; Roche
Diagnostics, Penzberg, Germany).
Statistical analysis
For differences in caloric intake and body weight gain,
repeated measures analysis of variance with multiple groups
were performed to determine effects over time and of
diet (chow, HF and HFHS or chow and HS) and interaction
effects (time?diet). If significant effects were detected,
post hoc analysis was performed to detect individual group
differences (Tukey), and individual differences at time points
(paired t-tests with Bonferonni correction). For fat mass,
substrate and hormone measures, and measures from the
glucose tolerance tests, analyses of variance were performed
for overall effects between groups, and if significant, post hoc
tests for individual group differences were performed
(Tukey). For chow vs HS effects, t-tests were used for
detecting differences. Po0.05 was considered significant.
Results
Free-choice HF and HFHS diets
Energy balance.
HFHS diets showed increased caloric intake compared
with rats on the chow diet (Figure 1). Rats on HFHS diet
maintained this high caloric intake for the remainder of the
study, whereas rats on HF diet decreased caloric intake to
chow levels (Figure 1a). Figures 1b–d show the intake of the
different foods that were presented to the rats. No differences
in chow intake were observed between rats on HFHS and HF
diets. Both rats on HFHS and HF diets consumed signi-
ficantly less chow (consuming B50kcalday–1from the chow
source) compared with the rats on the chow diet only
(B75kcalday–1). The overall intake of saturated fat in rats
on HF and HFHS diets was not significantly different. There
was, however, a small but significant decrease over time for
saturated fat intake in rats on HF diet, but not in rats on
Over the first week, both rats on HF and
week 1week 2 week 3week 4
0
40
80
120
chow
HF
HFHS
caloric intake (kcal/day)
week 1week 2week 3week 4
0
40
80
120
chow intake (kcal/day)
week 1week 2week 3week 4
0
20
40
60
80
saturated fat intake (kcal/d)
week 1week 2week 3week 4
0
10
20
30
40
sugar intake (kcal/d)
a
b
b
a
a
a
a
b
b
b
aa
b b
a
b b
a
bb
a
bb
a
b
a
b
c
ab
ab
ab
a
b
ab
ab
ab
Figure 1
HFHS diet (n¼6). (a) Rats on HFHS diet showed an increased caloric intake
during all 4 weeks, compared with rats on a chow diet. Rats on HF diet showed
increased caloric intake compared with chow controls only in the first week.
(b) Chow intake was decreased in rats on HFHS and HF diets compared with rats
on a chow diet. (c) Caloric intake from saturated fat was stable over 4 weeks in
rats on HFHS diet, whereas in rats on HF diet a decline was observed over the 4
weeks, with lower saturated fat intake in rats on HF diet compared with HFHS diet
in the fourth week. (d) Rats on HFHS diet consumed a stable amount of 30%
liquid sugar. Data are mean±s.e.m. White bars, rats on chow; striped bars, rats
on HF diet; and black bars, rats on HFHS diet. Different letters represent significant
differences between bars (Po0.05) as indicated by repeated measures analysis of
variance (ANOVA). Details on statistics are given in Supplementary Data 1.
Daily caloric intake in rats on a chow diet (n¼6), HF diet (n¼6) or
Dietary fat, sugar and glucose metabolism
SE la Fleur et al
597
International Journal of Obesity

Page 4

HFHS diet (Figure 1c). Intake of the sugar solution in rats on
HFHS diet was constant over time (Figure 1d).
Rats on HFHS diet gained more body weight (Figure 2a)
and accumulated more fat mass over 4 weeks, compared with
rats on HF diet or on a chow diet (Figures 2b and c). Plasma
concentrations of leptin and FFAs were increased signifi-
cantly in rats on HFHS and HF diets compared with rats on
a chow diet (Figures 2d and e). There were no significant
differences in plasma concentrations of leptin and FFAs
between rats on HFHS and HF diets (Figures 2d and e).
Details on statistics for data presented in Figures 1 and 2 are
provided in Supplementary Table 1.
Glucose tolerance.
glucose tolerance test was performed. After a 9-h fasting
After 1 week on the diet, an intravenous
period, basal plasma glucose concentrations were signifi-
cantly increased in rats on HFHS diet compared with those
on HF diet or on a chow diet (Table 1). Basal plasma insulin
concentrations tended to be increased in rats on HFHS diet
compared with the rats on a chow diet, but were not
different from rats on HF diet (Table 1). Furthermore, glucose
tolerance was impaired in rats on the HFHS-choice diet
compared with rats on HF diet or on a chow diet. The
amount of glucose in the circulation calculated after the
1gkg–1glucose bolus, that is, the AUC, was significantly
higher in rats on HFHS diet compared with that in rats on HF
diet or on a chow diet. The insulin response to the glucose
bolus was comparable between rats on HFHS, HF and chow
diets, that is, the AUC was not significantly different
(Table 1). As an estimation for the ability of the b-cell to
chow HF HFHS
0
5
10
15
20
total WAT (g)
chowHFHFHS
0
1
2
3
4
5
% total WAT (g/100g BW)
chowHFHFHS
0
5
10
15
leptin (ng/ml)
chow HF HFHS
0.0
0.5
1.0
1.5
2.0
2.5
FFA (mmol/l)
a
b
c
a
b
c
a
b
b
a
b
b
048121620 24 28
0
40
80
120
chow
HF
HFHS
*
*
*
: HFHS vs chow and HF p<0.05
time (days)
body weight gain (g)
Figure 2
chow or HF diets. Open circles: rats on chow; gray squares: rats on HF diet; black circles: rats on HFHS diet. Data are mean±s.e.m. After repeated measures analysis
of variance (ANOVA) revealed significant effects of Time and Time?Diet. *Po0.05. Details on statistics are given in Supplementary Data 1. Absolute body weight at
the beginning of diet switch was: chow: 352±5g, HF: 355±6g and HFHS: 347±8g, and at the end was: chow: 421±5g, HF: 430±12g, and HFHS: 443±9g.
(b) Total fat stores (mesenteric-omental, epididymal, perirenal and subcutaneous (inguinal)), (c) total fat stores corrected for body weight, and plasma
concentrations of (d) leptin and (e) free fatty acid in rats on chow, HF and HFHS diets. Data are mean±s.e.m. White bars: rats on chow; striped bars: rats on HF diet;
and black bars: rats on HFHS diet. Different letters represent significant differences between bars (Po0.05) after ANOVA indicated a significant effect of Groups
(Po0.05). Details on statistics are given in Supplementary Data 1.
(a) Body weight gain over 4 weeks in rats on chow, HF or HFHS diets (n¼6 for all groups). Rats on HFHS diet gained more weight compared with rats on
Dietary fat, sugar and glucose metabolism
SE la Fleur et al
598
International Journal of Obesity

Page 5

respond to the glucose challenge, we calculated insulin
secretion over the first 5min after injection per glucose unit
(I/G0–5: the difference between the insulin concentrations at
5 and 0min (DI5–0) divided by the difference between the
glucose concentrations at the same time (DG5–0)). The I/G0–5
was not different between the groups (Table 1).
Another glucose tolerance test was performed at 4 weeks.
After 9h of fasting during the light period, we observed
increased basal plasma concentrations of glucose in rats on
HFHS diet compared with rats on HF diet or on a chow diet;
concentrations between rats on HF and on a chow diet did
not differ (Figure 3a). Overall analysis revealed a trend
toward differences in basal plasma insulin concentrations
between rats on HFHS, HF and chow diets. The post hoc
analysis revealed a significant difference between basal
insulin concentrations in rats on HFHS diet compared with
rats on HF diet, and a trend towards a difference in plasma
insulin concentrations in rats on HFHS and on chow diets
(a detailed description of statistics are given in Supplementary
Data). Plasma insulin concentrations in rats on HF diet were
not different compared with rats on a chow diet (Figure 3b).
Although rats on HF diet for 4 weeks consumed a
considerable amount of saturated fat and became obese,
glucose tolerance was not affected (Figure 3c). On the other
hand, like we observed at 1 week, glucose tolerance again
was reduced in rats on HFHS diet for 4 weeks, compared with
rats on chow or HF diets. Plasma glucose concentrations
were significantly increased 5min after the injection of the
glucose bolus when compared with the glucose concentra-
tions in rats on HF and chow diets (Figure 3c). At 10 and
Table 1
on the diet
Glucose tolerance in rats on chow, HF and HFHS diets after 1 week
Chow (n¼6)HF (n¼6)HFHS (n¼6)
Means.e.m. Mean s.e.m.Mean s.e.m.
Glucose (AUC)
Insulin (AUC)
1010
36.8
94
5.9
872
53.8
187
15.1
1505*
74.3
251
21.2
Basal glucose (mg per 100ml) 121.9
Basal insulin (ngml–1)
3
0.3
117.2
3.0
2.3
0.4
135.9*
4.1**
8.4
0.6 2.3
I/Gt0–5
3.30.65.51.7 5.42.6
Abbreviations: AUC, area under the curve; HF, high fat; HFHS, high-fat high-
sugar. *Significant difference between HFHS and HF and between HFHS and
chow: Po0.05, after ANOVA detected a significant overall effect. **TREND
(P¼0.07) that there is a difference between chow and HFHS after ANOVA
detected an overall trend (P¼0.1).
10203040 5060
-50
50
150
250
*
*
#
time (min)
Δ Δ glucose (mg/dL)
-5
0
5
10
15
chow
HF
HFHS
#
#
time (min)
Δ Δ insulin (ng/ml)
chowHFHFHS
100
110
120
130
140
150
*
*
glucose (mg/dl)
chowHFHFHS
0
2
4
6
8
10
*
insulin (ng/ml)
chow HFHFHS
0
500
1000
1500
2000
*
*
AUC (glucose)
chowHFHFHS
0
25
50
75
100
*
AUC (insulin)
106050403020
Figure 3
groups n¼6). (a) Basal plasma glucose concentrations were significantly
higher in rats on HFHS diet when compared with rats on chow or HF diets. (b)
Basal plasma insulin concentrations were significantly higher in rats on HFHS
diet when compared with rats on HF diet; when compared with chow animals,
there was a trend. (c) Changes in plasma glucose concentrations and
(d) plasma insulin concentrations after a glucose bolus (1mgkg–1, intra-
venous) in rats on HFHS, HF or chow diets. Area under the curve for (e) plasma
glucose concentrations and (f) plasma insulin concentrations after a glucose
bolus in rats on HFHS, HF or chow diets. White bars or circles: rats on chow;
striped bars or squares: rats on HF; and black bars or circles: rats on HFHS.
Data are mean±s.e.m. For (a, b, e and f): *Po0.05 after analysis of variance
(ANOVA) revealed a significant effect of Group (Po0.05). For (c and d):
*Po0.05 for HFHS vs chow;#Po0.05 for HFHS vs HF. Details on statistics are
given in Supplementary Data 1.
Glucose tolerance at 4 weeks in rats on chow, HF or HFHS diets (all
Dietary fat, sugar and glucose metabolism
SE la Fleur et al
599
International Journal of Obesity

Page 6

15min, plasma glucose concentrations were still elevated
compared with those in rats on a chow diet (but not when
compared with the glucose concentrations in rats on HF
diet). Again, as was observed after 1 week, the AUC for
glucose was significantly higher in rats on HFHS diet
compared with the AUC for glucose in rats on HF and chow
diets (Figure 3e). There was also a significant difference
(using a repeated measures analysis of variance comparing all
groups) in the insulin response to the glucose load over time
between the rats on chow, HF or HFHS diets (Figure 3d, see
statistical details in Supplementary Data). The post hoc
analysis revealed a significant difference between the rats
on HF and HFHS diets. Furthermore, overall comparison of
the AUC for insulin revealed a trend toward differences
between the three diet groups (Figure 3f). The post hoc
analysis showed that the AUC for insulin after a glucose load
tended to be higher in rats on HF diet compared with rats on
HFHS diet (Figure 3f). Interestingly, the ability of the b-cell to
respond to the glucose bolus was clearly different between
the groups. The analysis of insulin secretion over the first
5min after the glucose bolus indicated that the early insulin
release to the glucose challenge was diminished in rats on
HFHS-choice diet for 4 weeks compared with rats on a chow
diet and compared with rats on HF diet (Figure 4a). No
differences in insulin secretion to a glucose load were
detected between rats on HF and chow diets (Figure 4a).
Thus, although both groups of rats become obese with
comparable plasma levels of FFA and leptin (Figures 2d and e),
only rats on HFHS diet show glucose intolerance. However,
there were differences in total amounts of fat mass after
4 weeks between the groups (Figure 2b). Therefore, we
correlated the amount of total fat mass with the insulin
response to a glucose load, that is, the I/G0–5,and with the
initial rise in plasma insulin concentrations over the first
5min (I0–5). In rats on HF diet, the insulin response correlated
positively with the amount of fat mass, whereas in rats on
HFHS diet, this correlation was negative (Figures 4b and c). In
addition, rats with similar amounts of white adipose tissue
released more insulin when on the HF diet than on the HFHS
diet. We also correlated the insulin response to the amount of
calories consumed; however, this only revealed a significant
positive correlation for insulin increases over the first 5min
after the glucose load in rats on HF diet (Figure 6).
Free-choice HS diet
Caloric intake over the total period of the experiment was
significantly increased in rats on HS diet compared with rats
on a chow diet. However, more detailed analyses by week
detected only a trend for weeks 1 and 4 (Figure 5a). Rats on
HS diet consumed significantly less chow than rats on a
chow diet only (Figure 5b). Chow intake in rats on HS diet
was stable over the course of the experiment (Figures 5b
and c), whereas a significant effect of time was detected for
sugar intake, although this was not reflected in differences
between the separate weeks (Figure 5c).
chowHFHFHS
0
2
4
6
8
*
*
I/G0-5m
0510
total WAT (g)
152025
0
3
6
9
12
HFHS
HF
chow
insulin increase (ng/ml)
05 10
total WAT (g)
152025
0
2
4
6
8
10
I/Gt0-5
Figure 4
(a) I/Gt0–5: the ratio of DI5–0 to DG5–0 (DI5–0/DG5–0 as a measure of the
insulin response to glucose in the first 5min in rats on HF, HFHS or chow diets.
Data are mean±s.e.m.; *Po0.05 after analysis of variance (ANOVA) revealed
a significant overall effect of Group. (b) For individual rats, absolute total
adiposity per group was correlated with the insulin response to a glucose load
(I/Gt0–5). In rats on HF diet, this correlation was positive (R2¼0.74; P¼0.03),
whereas in rats on HFHS diet this correlation was negative (R2¼0.77;
P¼0.02). (c) For individual rats, absolute total adiposity was correlated with
the insulin increments over the first 5min after the glucose bolus. This
correlation was positive for rats on chow diet (R2¼0.63; P¼0.06) and for rats
on HF diet (R2¼0.66; P¼0.05) and negative for rats on HFHS diet (R2¼0.69;
P¼0.04).
Glucose tolerance at 4 weeks in rats on chow, HF or HFHS diets.
Dietary fat, sugar and glucose metabolism
SE la Fleur et al
600
International Journal of Obesity

Page 7

The increase in caloric intake did not result in changes
in body weight. There were no differences in body
weight gain between rats on a chow diet or on HS diet
(Figure 5d). Although body weight gain was not affected
by the HS diet, abdominal fat pads (absolute or corrected
for body weight) were significantly heavier after 4 weeks
(Figures 5e and f). Furthermore, plasma concentrations
ofleptinandFFAswere
(Figures 5g and h).
At 1 week, an intravenous glucose tolerance test was
performed. Basal plasma concentrations of glucose and
insulin, and the glucose and insulin responses to the glucose
load, were not different between rats on HS or chow diets
(Table 2). After 4 weeks, we observed an increase in basal
plasma insulin concentrations; however, no differences
in basal plasma glucose concentrations or in glucose and
alsosignificantlyincreased
week 1 week 2week 3week 4
0
40
80
120
chow
HS
*
*
*
*
chow intake
week 1 week 2 week 3week 4
0
20
40
60
80
sugar intake
week 1 week 2week 3week 4
0
40
80
120
total caloric intake (kcal)
051015 202530
0
40
80
120
chow
HS
time (days)
body weight gain (g)
chow HS
0
5
10
15
20
*
TWAT (g)
chowHS
0
1
2
3
4
5
*
% total WAT (g/100g BW)
chow HS
0
5
10
15
*
leptin (ng/ml)
chowHS
0.0
0.5
1.0
1.5
2.0
2.5
*
FFA (mmol/l)
Figure 5
compared with rats on a chow diet; however, analysis on separate weeks only revealed a trend for weeks 1 and 4. (b) Chow intake was decreased in rats on HS diet
compared with rats on chow diet. (c) Liquid sugar intake in rats on HS diet. The analysis of variance (ANOVA) detected an effect of Time; however, post hoc analysis
only revealed a trend for a significant difference between weeks 3 and 4. (d) Body weight over time was not significantly different between rats on chow or HS diets.
Absolute body weight at the beginning: chow: 351±2g and HS: 343±6g; and at the end: chow: 430±7g and HS: 440±10g. Data are mean±s.e.m. White bars:
rats on chow; gray bars: rats on HS diet. Details on statistics are given in Supplementary Data 1. *Po0.05. (e) Total fat stores (mesenteric-omental, epididymal,
perirenal and subcutaneous (inguinal)), (f) total fat stores corrected for body weight, and plasma concentrations of (g) leptin and (h) free fatty acid in rats on chow
or HS diets. White bars: rats on chow; gray bars: rats on HS diet. Data are mean±s.e.m.; *Po0.05.
Daily caloric intake was calculated per week in rats on a chow diet (n¼5) or HS diet (n¼5). (a) Overall, rats on a HS diet consumed more calories
Table 2
1 and 4 weeks on the diet
Glucose tolerance in rats on chow (n¼5) and HS diets (n¼5) after
One weekFour weeks
ChowHS ChowHS
Mean s.e.m. Mean s.e.m. Mean s.e.m. Mean s.e.m.
Glucose (AUC)
Insulin (AUC)
1153
50.5
267
17.2
1447
53.3
136
3.2
1107
61.0
139
4.4
1242
50.8
99
3.2
Basal glucose
(mg per 100ml)
Basal insulin (ngml–1)
90.71.295.26.797.22.2100.22.5
1.60.6 2.40.1 2.20.123.9*0.2
I/Gt0–5
5.92.04.10.7 4.40.64.50.3
Abbreviations: AUC, area under the curve; HS, high sugar. *Significant
difference between HS and chow: Po0.05.
Dietary fat, sugar and glucose metabolism
SE la Fleur et al
601
International Journal of Obesity

Page 8

insulin responses to a glucose load between rats on chow and
HS diets were observed (Table 2).
No correlations were found between absolute fat mass and
the insulin response (neither the I/G0–5 nor the I0–5), or
between caloric intake and the insulin response in rats on HS
diet (data not shown).
Discussion
It is well known that high-calorie diets result in obesity with
elevated FFA concentrations and may result in insulin
resistance and glucose intolerance.7,8,19–23It is, however,
difficult to determine whether insulin resistance and glucose
intolerance are directly due to dietary changes, or are
indirectly due to the concomitant obesity.
In this study we present data that show that glucose
tolerance is directly affected by the composition of the diet
and not by obesity per se. Rats on both a free-choice HF
diet and a free-choice HFHS diet became obese with similar
plasma concentrations of FFAs, but only rats on a free-choice
HFHS diet developed hyperglycemia, hyperinsulinemia,
glucose intolerance and a diminished insulin response to a
glucose load. Rats on a free-choice HFHS diet that showed
similar levels of adiposity as rats on free-choice HF diet
released significantly less insulin to a glucose load. Thus, the
change in adiposity, because of the diet, did not explain
the difference in insulin responses to a glucose load between
rats on HFHS and HF diets. Moreover, although a clear
hyperinsulinemia and increased adiposity was observed in
rats on HS diet for 4 weeks, plasma glucose concentrations
and glucose tolerance were not different compared with rats
on a chow diet, suggesting that sugar alone is not responsible
for the observed glucose intolerance. Therefore, we conclude
that the combination of consuming saturated fat and liquid
sugar in our free-choice HFHS diet directly contributes to the
changes in glucose metabolism and is independent of
changes in adiposity or concentration of plasma FFA.
Several different mechanisms might underlie the changes
in glucose metabolism because of a free-choice HFHS diet.
In this study we show that obesity itself or the total amount
of caloric intake does not explain the changes in glucose
metabolism. As stated in the Introduction, elevated FFA
plasma concentrations may mediate obesity effects on
glucose metabolism. The brain and its projections to the
ANS, however, could also have an important role in diet-
induced insulin resistance and glucose intolerance.12,13One
of the changes in glucose metabolism in rats on HFHS diet
is the change in the insulin response to a glucose load.
It has been demonstrated that a defect in the acute insulin
response, or the early insulin response, occurs early in the
development of type 2 diabetes mellitus and it may contri-
bute to the conversion from normal to impaired glucose
tolerance and diabetes.24The early insulin response may well
depend on the sensitivity of the pancreas for glucose. The
pancreas is innervated by the ANS and it has been shown
that the hypothalamus may regulate this sensitivity via
this ANS innervation.25–28The latter hypothesis is supported
by our recent finding that centrally, neuropeptide Y and
proopiomelanocortin mRNA expression in the arcuate
nucleus were altered in HFHS-choice diet rats in such a
direction that it promotes glucose intolerance and insulin
resistance.15
Neuropeptide Y- and proopiomelanocortin-
expressing neurons in the arcuate nucleus affect glucose
metabolism via their projections to (pre-autonomic) hypo-
thalamic neurons that control the autonomic nervous input
to various peripheral organs, such as the pancreas.25Thus, it
could well be that alterations in the brain in rats on a free-
choice HFHS diet are mediating the effects on glucose
metabolism via the ANS.
Another explanation for the reduced insulin response to
a glucose load is an increased insulin clearance. However,
obesity and elevated concentrations of FFA have been asso-
ciated with decreased insulin clearance rather than increased
insulin clearance,11and therefore it does not seem likely that
the effects of a HFHS-choice diet on the insulin response to a
glucose load are because of alterations in insulin clearance.
In addition, we cannot exclude the possibility that altera-
tions in glucose metabolism because of a HFHS-choice diet
involve the ‘toxic’ effect of the combined increased levels
of glucose and fatty acids on the pancreas (the so-called
‘glucolipotoxicity’ theory).29However, as the glucose eleva-
tions in HFHS-choice diet rats are much less than glucose
concentrations necessary (twofold) to affect b-cell func-
tion,30it seems unlikely that the effect of the HFHS-choice
diet involves glucolipotoxicity.
Although rats on both HFHS and HF diets became obese,
there were differences in feeding behavior and the absolute
amount of fat mass between the groups that could affect
glucose tolerance and insulin responsiveness. As we have
previously shown,15rats on HFHS diet are hyperphagic,
whereas those on HF diet are only temporarily hyperphagic,
reducing their caloric intake after 1 week on the diet. In rats
on HFHS diet, after only 1 week we observed hyperglycemia,
glucose intolerance and a trend towards hyperinsulinemia
(which may point to reduced insulin sensitivity), whereas in
rats on HF diet, we did not observe changes in plasma
concentrations of glucose and insulin or glucose tolerance at
1 week. As total caloric intake was not different between the
two diet groups over the first week, it is unlikely that the
total amount of calories mediates the diet-induced glucose
intolerance in rats on HFHS diet.
Furthermore, caloric intake over 4 weeks and fat mass after
4 weeks were different between the rats on HF and HFHS
diets. The total caloric intake in all groups correlated
positively with the amount of fat mass (Figure 6). Further-
more, this fat mass showed a positive correlation with the
glucose-induced insulin response in rats on HF diet, which
corresponds to findings in obese subjects in which body fat
mass correlates positively with glucose-stimulated insulin
secretion.31–33However, in rats on HFHS diet, this correla-
tion was negative; thus, the more obese, the less insulin was
Dietary fat, sugar and glucose metabolism
SE la Fleur et al
602
International Journal of Obesity

Page 9

secreted after a glucose bolus. This suggests that a difference
in total caloric intake or obesity between the groups does not
explain the changes in glucose metabolism, and suggests
that dietary composition, and not obesity, is involved in
diet-induced glucose intolerance with diminished insulin
secretion.
As the HF-choice diet did not result in glucose intolerance,
whereas the HFHS-choice diet did, we examined whether
sucrose could be the important factor. It has been shown by
several studies that sucrose/fructose diets stimulate insulin
release and result in insulin resistance.5,34,35Like others, we
showed that rats on a HS-choice diet for 4 weeks exhibited
increased basal insulin concentrations, but did release
enough insulin to a glucose load to keep glucose concentra-
tions similar to that observed for controls.35Interestingly,
other studies do suggest that glucose intolerance occurs in
high sucrose-fed rats that are fed sucrose for a longer period
of time.36,37
We have to note that the method of presenting our diets
(using choice) is distinct from various studies in the
literature (using synthetic dry diets), making comparison
difficult. Early changes in glucose tolerance have been
described for rats on a 40%-fat diet,12which is not in
agreement with our results in rats on a free-choice HF diet.
This could be because of the difference in how the diets were
presented (that is, choice HF diet vs a synthetic HF diet);
however, caloric intake was also much higher in rats on a
synthetic HF diet compared with consumption in our study
and rats were much younger when experiments began. In
addition, several studies using a synthetic HFHS diet
(composed of similar percentages for fat/sugar as rats on a
free-choice HFHS diet choose) did find glucose intolerance
and insulin resistance; however, a synthetic HFHS diet also
increases adiposity and plasma FFA concentrations,4which
makes it difficult to distinguish between effects of the diet
and the obesity.
The early changes (that is, 1 week) in glucose tolerance
have also been reported when mice were fed a 40%-fat diet;
however, this diet was also low in protein (6%)7and this
could have affected outcome. As only chow contains
protein, a reduction in chow intake, as observed in our
study, will reduce the protein amount, which could affect
food intake and body weight.38,39However, rats on the
HF-choice diet and the HFHS-choice diet both reduced chow
intake by 29%, and thus this does not explain the develop-
ment of glucose intolerance in rats on a HFHS-choice diet.
Furthermore, as the chow diet contains 22% protein, a
reduction of B30% of this diet will still result in a sufficient
amount of protein for healthy growth (that is, comparable to
a 15% protein diet).40,41
In summary, the specific combination of a diet of saturated
fat and a sugar solution, but not fat or sugar alone, results in
rapid glucose intolerance and insulin unresponsiveness to a
glucose load, which is not explained by levels of circulating
lipids or fat mass accumulation. These data, together with
previous data showing effects of a HFHS-choice diet on
hypothalamic neuropeptide expression,15imply a role for
the central nervous system in these diet-induced alterations
in glucose metabolism, and in the early transition from
glucose intolerance to type 2 diabetes mellitus.
Conflict of interest
The authors declare no conflict of interest.
Acknowledgements
We thank Dr JM Chou-Green for English editing and
Drs MJM Serlie and M Soeters for critically reading the
manuscript. This research was supported by the Netherlands
Organization for Scientific Research (ZonMw 916.56.020).
References
1 Batsis JA, Nieto-Martinez RE, Lopez-Jimenez F. Metabolic syn-
drome: from global epidemiology to individualized medicine.
Clin Pharmacol Ther 2007; 82: 509–524.
2 Brownell KD, Farley T, Willett WC, Popkin BM, Chaloupka FJ,
Thompson JW et al. The public health and economic benefits
of taxing sugar-sweetened beverages. N Engl J Med 2009; 361:
1599–1605.
3 Macdiarmid J, Loe J, Craig LC, Masson LF, Holmes B, McNeill G.
Meal and snacking patterns of school-aged children in Scotland.
Eur J Clin Nutr 2009; 63: 1297–1304.
4 Buettner R, Scholmerich J, Bollheimer LC. High-fat diets:
modeling the metabolic disorders of human obesity in rodents.
Obesity (Silver Spring) 2007; 15: 798–808.
5 Pagliassotti MJ, Prach PA, Koppenhafer TA, Pan DA. Changes
in insulin action, triglycerides, and lipid composition during
sucrose feeding in rats. Am J Physiol 1996; 271: R1319–R1326.
6 Pagliassotti MJ, Prach PA. Quantity of sucrose alters the tissue
pattern and time course of insulin resistance in young rats.
Am J Physiol 1995; 269: R641–R646.
607080
total caloric intake
(kcal/d; over 4 wks)
90100 110 120
0
5
10
15
HF
HFHS
chow
r2=0.77
p<0.02
insulin increments (ng/ml)
Figure 6
4 weeks with first 5min insulin increases after the glucose load in rats on
HFHS, HF and chow diets.
Correlation between total caloric intake (kcal) measured over
Dietary fat, sugar and glucose metabolism
SE la Fleur et al
603
International Journal of Obesity

Page 10

7 Ahren B, Pacini G. Insufficient islet compensation to insulin
resistance vs reduced glucose effectiveness in glucose-intolerant
mice. Am J Physiol Endocrinol Metab 2002; 283: E738–E744.
8 Kaiyala KJ, Prigeon RL, Kahn SE, Woods SC, Porte Jr D, Schwartz MW.
Reduced beta-cell function contributes to impaired glucose tolerance
in dogs made obese by high-fat feeding. Am J Physiol 1999; 277:
E659–E667.
9 Boden G. Pathogenesis of type 2 diabetes. Insulin resistance.
Endocrinol Metab Clin North Am 2001; 30: 801–815 v.
10 Reaven GM. Banting lecture 1988. Role of insulin resistance in
human disease. Diabetes 1988; 37: 1595–1607.
11 Lewis GF, Carpentier A, Adeli K, Giacca A. Disordered fat storage
and mobilization in the pathogenesis of insulin resistance and
type 2 diabetes. Endocr Rev 2002; 23: 201–229.
12 Cruciani-Guglielmacci C, Vincent-Lamon M, Rouch C, Orosco M,
Ktorza A, Magnan C. Early changes in insulin secretion and
action induced by high-fat diet are related to a decreased
sympathetic tone. Am J Physiol Endocrinol Metab 2005; 288:
E148–E154.
13 Levin BE. Reduced norepinephrine turnover in organs and brains
of obesity-prone rats. Am J Physiol 1995; 268: R389–R394.
14 La Fleur SE, Vanderschuren LJMJ, Luijendijk MCM, Kloeze BM,
Tiesjema B, Adan RA. A reciprocal interaction between food-
motivated behavior and diet-induced obesity. Int J Obes (Lond)
2007; 31: 1286–1294.
15 La Fleur SE, van Rozen AJ, Luijendijk MC, Groeneweg F, Adan RA.
A free-choice high-fat high-sugar diet induces changes in arcuate
neuropeptide expression that support hyperphagia. Int J Obes
(Lond) 2010; 34: 537–546.
16 Obici S, Feng Z, Tan J, Liu L, Karkanias G, Rossetti L. Central
melanocortin receptors regulate insulin action. J Clin Invest 2001;
108: 1079–1085.
17 Van den Hoek AM, Voshol PJ, Karnekamp BN, Buijs RM, Romijn JA,
Havekes LM et al. Intracerebroventricular neuropeptide Y infusion
precludes inhibition of glucose and VLDL production by insulin.
Diabetes 2004; 53: 2529–2534.
18 Steffens AB. A method of frequent sampling blood and contin-
uous infusion of fluids in the rat without disturbing the animal.
Physiol Behav 1969; 4: 836.
19 Lichtenstein AH, Schwab US. Relationship of dietary fat to
glucose metabolism. Atherosclerosis 2000; 150: 227–243.
20 Storlien LH, Jenkins AB, Chisholm DJ, Pascoe WS, Khouri S,
Kraegen EW. Influence of dietary fat composition on develop-
ment of insulin resistance in rats. Relationship to muscle
triglyceride and omega-3 fatty acids in muscle phospholipid.
Diabetes 1991; 40: 280–289.
21 Wright DW, Hansen RI, Mondon CE, Reaven GM. Sucrose-
induced insulin resistance in the rat: modulation by exercise
and diet. Am J Clin Nutr 1983; 38: 879–883.
22 Paulsen SJ, Jelsing J, Madsen AN, Hansen G, Lykkegaard K, Larsen LK
et al. Characterization of beta-cell mass and insulin resistance in
diet-induced obese and diet-resistant rats. Obesity (Silver Spring) 2009;
18: 266–273.
23 Bergman RN. Minimal model: perspective from 2005. Horm Res
2005; 64 (Suppl 3): 8–15.
24 Bogardus C, Tataranni PA. Reduced early insulin secretion in the
etiology of type 2 diabetes mellitus in Pima Indians. Diabetes
2002; 51 (Suppl 1): S262–S264.
25 Buijs RM, Chun SJ, Niijima A, Romijn HJ, Nagai K. Para-
sympathetic and sympathetic control of the pancreas: a role for
the suprachiasmatic nucleus and other hypothalamic centers that
are involved in the regulation of food intake. J Comp Neurol 2001;
431: 405–423.
26 Kalsbeek A, la Fleur S, van Heijningen C, Buijs RM. Suprachias-
matic GABAergic inputs to the paraventricular nucleus control
plasma glucose concentrations in the rat via sympathetic
innervation of the liver. J Neurosci 2004; 24: 7604–7613.
27 Moore MC, Cherrington AD. Regulation of net hepatic glucose
uptake: interaction of neural and pancreatic mechanisms.
Reprod Nutr Dev 1996; 36: 399–406.
28 Van den Hoek AM, van Heijningen C, Schro-van der Elst JP,
Ouwens DM, Havekes LM, Romijn JA et al. Intracerebroventricular
administration of neuropeptide Y induces hepatic insulin resis-
tance via sympathetic innervation. Diabetes 2008; 57: 2304–2310.
29 Poitout V, Robertson RP. Glucolipotoxicity: fuel excess and {beta}-
cell dysfunction. Endocr Rev 2008; 29: 351–366.
30 Hagman DK, Latour MG, Chakrabarti SK, Fontes G, Amyot J,
Tremblay C et al. Cyclical and alternating infusions of glucose
and intralipid in rats inhibit insulin gene expression and Pdx-1
binding in islets. Diabetes 2008; 57: 424–431.
31 Ferrannini E, Natali A, Bell P, Cavallo-Perin P, Lalic N, Mingrone G.
Insulin resistance and hypersecretion in obesity. European Group
for the Study of Insulin Resistance (EGIR). J Clin Invest 1997; 100:
1166–1173.
32 Ferrannini E, Balkau B. Insulin: in search of a syndrome. Diabet
Med 2002; 19: 724–729.
33 Reaven GM, Moore J, Greenfield M. Quantification of insulin
secretion and in vivo insulin action in nonobese and moderately
obese individuals with normal glucose tolerance. Diabetes 1983;
32: 600–604.
34 Toida S, Takahashi M, Shimizu H, Sato N, Shimomura Y,
Kobayashi I. Effect of high sucrose feeding on fat accumulation
in the male Wistar rat. Obes Res 1996; 4: 561–568.
35 Ferrer-Martinez A, Marotta M, Turini M, Mace K, Gomez-Foix AM.
Effect of sucrose and saturated-fat diets on mRNA levels of genes
limiting muscle fatty acid and glucose supply in rats. Lipids 2006;
41: 55–62.
36 Havel PJ. Dietary fructose: implications for dysregulation of
energy homeostasis and lipid/carbohydrate metabolism. Nutr Rev
2005; 63: 133–157.
37 Hallfrisch J, Lazar F, Jorgensen C, Reiser S. Insulin and glucose
responses in rats fed sucrose or starch. Am J Clin Nutr 1979; 32:
787–793.
38 White BD, He B, Dean RG, Martin RJ. Low protein diets increase
neuropeptide Y gene expression in the basomedial hypothalamus
of rats. J Nutr 1994; 124: 1152–1160.
39 White BD, Dean RG, Martin RJ. An association between low
levels of dietary protein, elevated NPY gene expression in the
basomedial hypothalamus and increased food intake. Nutr
Neurosci 1998; 1: 173–182.
40 Du F, Higginbotham DA, White BD. Food intake, energy balance
and serum leptin concentrations in rats fed low-protein diets.
J Nutr 2000; 130: 514–521.
41 Goettsch M. Minimum protein requirement of the adult rat for
28-day periods of maintenance of body weight. J Nutr 1951; 45:
609–620.
Supplementary Information accompanies the paper on International Journal of Obesity website (http://www.nature.com/ijo)
Dietary fat, sugar and glucose metabolism
SE la Fleur et al
604
International Journal of Obesity