Clinical Science (2009) 116, 137–145 (Printed in Great Britain) doi:10.1042/CS20080120 137
Fructose modifies the hormonal response and
modulates lipid metabolism during aerobic
exercise after glucose supplementation
Juan M. FERN´ANDEZ∗, Marzo E. DA SILVA-GRIGOLETTO∗†, Juan A. RUANO-RU´IZ∗,
Javier CABALLERO-VILLARRASO∗, Rafael MORENO-LUNA∗, Isaac T ´ UNEZ-FI ˜NANA‡,
Inmaculada TASSET-CUEVAS‡, Pablo P´EREZ-MART´INEZ∗, Jos´e L´ OPEZ-MIRANDA∗
and Francisco P´EREZ-JIM´ENEZ∗
∗Lipids and Atherosclerosis Research Unit, Reina Sofia University Hospital, CIBER Fisiopatolog´ıa de la Obesidad y Nutricion
(CIBEROBN), 14004 C´ordoba, Spain, †Andalusian Centre of Sports Medicine, 14003 C´ordoba, Spain, and ‡Department of
Biochemistry and Molecular Biology, School of Medicine, University of C´ordoba, 14071 C´ordoba, Spain
A B S T R A C T
The metabolic response when aerobic exercise is performed after the ingestion of glucose
plus fructose is unclear. In the present study, we administered two beverages containing GluF
(glucose+fructose) or Glu (glucose alone) in a randomized cross-over design to 20 healthy
aerobically trained volunteers to compare the hormonal and lipid responses provoked during
aerobic exercise and the recovery phase. After ingesting the beverages and a 15-min resting
period, volunteers performed 30 min of moderate aerobic exercise. Urinary and blood samples
were taken at baseline (t−15), during the exercise (t0, t15 and t30) and during the recovery phase
(t45, t75 and t105). Plasma insulin concentrations were higher halfway through the exercise period
and during acute recuperation (t15 and t75; P<0.05) following ingestion of GluF than after Glu
alone, without any differences between the effects of either intervention on plasma glucose
concentrations. Towards the end of the exercise period, urinary catecholamine concentrations
were lower following GluF (t45; P<0.05). Plasma triacylglycerol (triglyceride) concentrations were
higher after the ingestion of GluF compared with Glu (t15, t30, t45 and t105; P<0.05). Furthermore,
with GluF, we observed higher levels of lipoperoxides (t15, t30, t45 and t105; P<0.05) and oxidized
LDL (low-density lipoprotein; t30; P<0.05) compared with after the ingestion of Glu alone. In
conclusion, hormonal and lipid alterations are provoked during aerobic exercise and recovery by
the addition of a dose of fructose to the pre-exercise ingestion of glucose.
INTRODUCTION
The pre-exercise ingestion of carbohydrates is a common
dietary supplement that delays fatigue and this is
frequently consumed by sportsmen and women, and
recreational practitioners [1]. However, differences in the
mechanisms of absorption and the metabolic pathways
taken by the ingested carbohydrate provoke a range of
metabolic and hormonal responses that are still only
partially understood.
Fructose is a monosaccharide that has often been
studied as an ergogenic supplement in exercise conditions.
Key words: carbohydrate supplementation, exercise, fructose, glucose, hypertriacylglycerolaemia, lipid metabolism.
Abbreviations: Glu, glucose alone; GluF, glucose+fructose; HR, heart rate; NEFA, non-esterified fatty acid; LPO, lipoperoxide;
LDL, low-density lipoprotein; oxLDL, oxidized LDL; Petco2, end-tidal partial pressure of carbon dioxide; Peto2 , end-tidal
partial pressure of oxygen; RPE, rating of perceived exertion; TAG, triacylglycerol; ˙Vco2, carbon dioxide production; ˙V e, minute
ventilation; VLDL, very-low-density lipoprotein; ˙Vo2, oxygen consumption; VT1, ventilatory threshold 1; VT2, ventilatory
threshold 2.
Correspondence: Dr Juan M. Fern´andez (email
juf_nutryinves@yahoo.com).
C The Authors Journal compilation C 2009 Biochemical Society
www.clinsci.org
Clinical Science
138 J. M. Fern´andez and others
In comparison with glucose, pre-exercise ingestion of
fructose did not lead to reactive hypoglycaemia and improved
performance [2]. This improvement has also been
reported when fructose was combined with glucose and
ingested during an aerobic exercise [3,4].Nevertheless, in
these studies, the authors extensively studied mechanisms
related to improved performance; for example, exogenous
and endogenous rates of carbohydrate oxidation, without
directly analysing the metabolic effects of glucose
plus fructose on insulinaemic and glycaemic responses.
Moreover, at least during rest, a reduction has been
reported in the insulinaemic and glycaemic response
following the combination of different doses of fructose
with high glycaemic index carbohydrates [5–7]; a finding
that has not been studied under exercise conditions. Such
information would be of great importance, as one of the
key objectives in aerobic exercise is low insulinaemia and
the reduction in the vagal α-adrenergic stimulus to favour
the selective use of substrates [8–10] and β-adrenergic
stimulation for lipolysis [8].
Onthe other hand, there is growing interest in studying
the effects on lipid metabolism and insulin resistance
following the ingestion of fructose [11,12]. Stimulation
of de novo lipogenesis [13], a defect in the clearance of
VLDL (very-low-density lipoprotein) particles [14] or
even the esterification of non-oxidized NEFAs (nonesterified
fatty acids) in the liver [15] are among the
explanations for the hypertriacylglycerolaemia induced
by fructose in healthy subjects at rest. Meanwhile, the
reduction in insulin sensitivity appears to be dependent
on fructose-induced lipid modification [16]. Nevertheless,
a high variability has been observed in the studies
carried out to date, depending in part on the quantity of
fructose administered [17], and the type and quantity
of carbohydrate that accompanied the fructose [6]. Furthermore,
other factors that have been studied less, such
as the training status of the subjects [18] and exercise
performed during the post-absorption phase, might also
influence glycaemic and lipid metabolism after fructose
ingestion. For this reason, there is no consensus regarding
the ingestion of fructose either alone or in combination
with high glycaemic index carbohydrates; despite this,
its use as a low glycaemic index sweetener is still being
recommended, and fructose is even employed in sports
drinks [19].
To the best of our knowledge, no studies exist that
evaluate, in trained adults, the acute hormonal and metabolic
effects of the ingestion of a combination of GluF
(glucose+fructose) before aerobic exercise. The aim
of the present study was therefore to compare the acute
hormonal response induced by the pre-exercise ingestion
of Glu (glucose alone) or GluF and its effects on lipid
metabolism during the course of moderate aerobic
exercise and the acute recovery phase. The influence
of pre-exercise ingestion of these supplements on the
oxidative state of circulating lipids was also studied.
Table 1 Characteristics and training background of the
study subjects
Values are means+−
S.D. BMI, body mass index; HDL, high-density lipoprotein.
Variable Value
Biological
Age (years) 26+−
4.85
Weight (kg) 75+−
10.5
BMI (kg/m2) 23.45+−
1.93
Body fat (%) 16.96+−
3.44
Biochemical
Glycaemia (mmol/l) 4.19+−
0.43
Total cholesterol (mmol/l) 3.60+−
0.68
TAG (mmol/l) 0.70+−
0.21
HDL (mmol/l) 1.37+−
0.34
LDL (mmol/l) 1.88+−
0.48
Total training
Years 2.2+−
1.3
Days/week 5.7+−
0.5
Sessions/week 6.3+−
1.2
Minutes/week 349.7+−
69.2
Endurance training
Sessions/week 5.4+−
0.8
Minutes/week 246.3+−
60.6
MATERIALS AND METHODS
Subjects
A total of 20 healthy adultmen (mean age, 26+−
4.8 years)
volunteered to participate in the study. All subjects
were normoglycaemic and normolipaemic, and none
had undergone pharmacological treatment or had taken
vitamin or mineral supplements in the 2 months prior
to the study. The subjects were aerobically trained,
participating in physical exercise sessions more than three
times a week. Table 1 shows their anthropometric and
biological characteristics and training background. The
aims of the study and the possible risks involved were
explained and an informed consent form was signed by
each subject before the start of the study. The ethics
committee of the Reina Sof´ıa Hospital in C´ ordoba
approved all the procedures employed in the study.
Study design
As a preliminary analysis, the participants were subjected
to an anthropometric study, an evaluation of their
nutritional state and a pre-testing session in order to
determine suitable workloads during the experimental
exercise sessions { ˙Vo2max [maximal ˙V o2 (oxygen
consumption)] and 10RM (ten repetition maximum)
tests}. The subjects were thereafter randomly assigned
to perform two experimental trials in consecutive order,
which consisted of pre-exercise ingestion of Glu or GluF
and a session of moderate aerobic exercise (Figure 1). A
C The Authors Journal compilation C 2009 Biochemical Society
Metabolic effects of fructose during exercise 139
Figure 1 Schematic diagram of the experimental protocol
used in the Glu and GluF trial
The timing of consumption of the carbohydrate () in relation to the exercise
and recovery periods is given. Blood samples were taken at the times shown.
Urine production (grey dots and arrowhead) and rate of perceived exertion test
() are also shown.
1-week washout period followed each experimental trial,
and the subjects were also instructed to avoid moderateto-
severe physical exercise during the 24 h before each
intervention. This was controlled bymeans of an exercise
autoself recorder and by analysing basal concentrations
of Srm (standard reference material) creatine kinase, with
levels >200 units/l qualifying as an exclusion criterion.
The consumption of caffeine, alcohol and carbohydrates
with a high glycaemic index was also avoided during the
day before each trial.
Nutritional status and diet
A retrospective qualitative/quantitative assessment of the
frequency of food intake for the 4 weeks before the preliminary
evaluations was obtained from each subject by
a nutritionist. The composition of the normal diet and
consumption of alimentary antioxidants were calculated
with the aid of tables of the chemical composition of
food. This nutritional information was used to prescribe
an isocaloric diet, with a moderate glycaemic load and
glycaemic index [20,21]. The daily ration of vegetables
and fruit was set in accordance with the recommended
dietary intakes of ascorbic acid. This diet was followed
by the subjects for 2 weeks before the start of the
experimental trials and during the washout periods
between the interventions. A 24-h food consumption
diary was completed by the subjects during the week
before each experimental trial on days 3 (half week) and 6
(the day before the intervention). The composition of the
diet followed by the subjects was calculated and is shown
in Table 2.
Pre-testing and determination
of workload
All the participants performed a progressive endurance
test on an ergometer cycle (Ergometrics 800; Ergoline)
in order to calculate the load of endurance exercise at
the same time of day (10.00–13.00 hours) and under
identical ambient conditions (21–24 ◦C; 45–55% relative
humidity). The protocol comprised pedalling for 2 min
at 25 W and a further 2 min at 50 W as a warm-up;
afterwards, the load was increased at a rate of 25 W/min.
Table 2 Dietary composition during the week before the
Glu or GluF trial
Values are means+−
S.D. RDI, recommended daily intake.
Dietary variable Value
Caloric distribution
Total calories (kJ/day) 7492+−
1103
Carbohydrate (%) 44.56+−
5.89
Protein (%) 23.85+−
3.59
Fat (%) 31.25+−
5.54
Energy density (kJ/g) 5.69+−
0.88
Dietary carbohydrate
Total carbohydrate (g) 200.01+−
38.64
Dietary glycaemic load 124.41+−
16.43
Dietary glycaemic index 63.51+−
6.85
Total fibre (g) 23.24+−
5.51
Ascorbic acid
Intake (mg/day) 65.54+−
26.94
%RDI 109.2
Subjects were required to maintain a constant pedalling
rate of between 60 and 70 rev./min. The test came to an
end when: (i) the subject decided on his own to end it;
(ii) he could no longer maintain the minimum pedalling
rate of 60 rev./min; or (iii) the criteria for completion
of the exercise suggested by the American College of
Sports Medicine had been fulfilled [22]. HR (heart rate;
in beats/min) was recorded continuously throughout the
tests and for 3 min post-exercise by means of a 12-channel
ECG (ViasysTM; Pulse Biomedical). Gas exchange data
were obtained continuously by means of an automatic
breath-by-breath system (Oxycon Delta; Jaeger), which
was calibrated before each test to the appropriate
ambient conditions. The following parameters were
acquired (average of each 15-s interval): ˙V o2 and ˙Vco2
(carbon dioxide production) [in l/min at STP (standard
temperature and pressure)]; ˙Ve [minute ventilation; in
l/min BPTS (backward preferred transition speed)], ˙V e/
˙V
o2, ˙V e/ ˙Vco2, Peto2 (end-tidal partial pressure of
oxygen) and Petco2 (end-tidal partial pressure of carbon
dioxide). VT1 and VT2 (ventilatory thresholds 1 and 2)
and the workloads (in W) corresponding to them were
calculated using the methodology suggested by Davis
[23]. In brief,VT1 was determined bymeans of an increase
in both ˙Ve/ ˙Vo2 and Peto2 without a concomitant
increase in ˙Ve/ ˙Vco2. VT2 was determined by means
of an increase in both ˙V e/ ˙V o2 and Peto2 and a fall in
Petco2. VT1 and VT2 were calculated independently by
two observers. In the event of a failure to agree, a third
observer was consulted.
Supplement and exercise protocol
The subjects arrived at the laboratory between 08.00
and 09.00 hours after 10–12 h of nocturnal fasting.
C The Authors Journal compilation C 2009 Biochemical Society
140 J. M. Fern´andez and others
At 15 min before the start of the exercise period,
they consumed, in accordance with a cross-randomized
protocol, a water-based solution (400 ml) of 50 g of glucose
(glucose anhydride; C6H12O6 99.5%) or 50 g of
glucose plus 15 g of monosaccharide fructose. The 50 g
of glucose, which requires no hydrolysis for absorption,
has been established as the minimal amount of glycaemic
carbohydrate in the two trials, with the purpose that this
may provide the same threshold for detecting changes
in the glycaemic response when fructose was given
simultaneously in the GluF trial. For this reason, we
intentionally used two non-isocaloric supplementations,
as others studies have done previously [5,6]. The Glu
and GluF solutions had a total concentration of 12.5 and
16.5% respectively. The fructose provided 23.07% of
the carbohydrates in the combined supplement, and this
amount was determined on the basis of previous reports
of its digestibility and absorption in combination with
glucose under conditions of rest and exercise [24,25].
The specific concentration of fructose (3.75%) was lower
than the 10% maximum reported as being capable of
being absorbedwithout risk of gastrointestinal symptoms
[26]. Under both trial conditions (Glu or GluF), aerobic
exercise was performed on the ergometer cycle utilized
in the preliminary tests. Each subject exercised at
an intensity that corresponded to the workload value
attained at a point equidistant between VT1 and VT2
during the preliminary aerobic tests. For example, an
individual who had reached VT1 and VT2 with loads of
100 and 200 W respectively, would perform the exercise
session against a load of 150 W. In addition, for warmup
purposes, during 1 and 2 min of the exercise sessions,
we employed workloads of 25 and 50 W respectively;
after 3 min, the load calculated during the preliminary
test was used until a total of 30 min of exercise had been
accomplished. The pedalling rate was the same as used in
the preliminary tests.
Blood and urine samples, and
analytical assays
On the morning of each experimental trial, a 16-gauge
Vennflon cannulawas inserted into the antecubital vein in
order to extract blood samples. The samples were drawn
immediately before the ingestion of the carbohydrates
(t−15) and during the 2 h of the post-prandial period at
intervals of 15, 30 and 60 min, including the exercise
period (t0, t15 and t30) and during the recovery phase (t45,
t75 and t105). A very small volume of sterile saline was
infused immediately after the extraction of each sample in
order to keep the cannula clear. The blood samples
were collected into tubes containing 1 g/lEDTAandwere
always protected from the light. The tubeswere stored on
ice, and plasma was separated within 30 min of extraction
by low-speed centrifugation at 1500 g for 15 min at 4◦C.
Glucose concentrations were determined by spectrophotometric
methods using a modular analyser (ISE-
4-DDPPEEPP; Hoffmann-La Roche). Plasma insulin
levels were measured by CMIA (chemiluminescent
microparticle immunoassay) using an analyser (Architect
i-4000; Abbott). Plasma TAG (triacylglycerol; ‘triglyceride’)
levels were measured spectrophotometry using
a modular analyser (ISE-4-DDPPEEPP; Hoffman-La
Roche). Serum NEFA concentrations were determined
using an enzymatic colorimetric assay (NEFA kit;
Roche). The levels of LPO (lipoperoxide) were
determined using a commercially available kit (LPO-
586; Oxis International). Quantitative determination of
oxLDL [oxidized LDL (low-density lipoprotein)] in
plasma was performed by enzyme immunoassay using a
commercially available kit (Biomedica). Plasma lactic acid
was measured by enzymatic colorimetric assay using an
analyser (Cobas 400; Hoffman-La Roche).
Total urine production was also collected into sterile
containers before the ingestion of the carbohydrates
(t−15), at the end of the exercise period (t30) and
the end of the recovery period (t105). The subjects
were recommended to drink water ad libitum during
the exercise and recovery periods in order to
encourage urine production. Urinary concentrations of
adrenaline (epinephrine), noradrenaline (norepinephrine)
and creatinine were determined by means of HLPC,
using a chromatograph (Bio-Rad Laboratories) with
a reverse-phase column (flow rate of 1 ml/min) and
an electrochemical detector (at 500 mV and 10 nA).
Urinary concentrations of adrenaline and noradrenaline
are expressed relative to urinary creatinine (nmol/mmol
of creatinine). In addition, 1 day prior to collecting
the urine, we encouraged the subjects to abstain from
bananas, coffee, pineapples and walnuts.We also ensured
they did not take any of the following drugs: pheothiazin,
paracetamol, salsinol, isoproterenol or α-methyldopa.
Perceived exertion and HR
The RPE (rating of perceived exertion) was recorded
according to the Borg perceived exertion scale (CR10)
at three different times: mid-exercise, after completion
of the exercise [27] and 30 min after completion of the
exercise (see Figure 5). This last recording was used to
indicate the subject’s overall perception of the session
[28]. The scale had been used by the subjects during their
last training year, so there was a good reproducibility
and a period for such purpose was not deemed necessary.
In addition, HR was recorded continuously during the
study (Polar 810).
Statistical analysis
Traditional statistical methods were used to calculate
means and S.D. and S.E.M. The normality of the samples
was tested using the Shapiro–Wilk’s test. The effect of
the different interventions (Glu andGluF as independent
C The Authors Journal compilation C 2009 Biochemical Society
Metabolic effects of fructose during exercise 141
Figure 2 Plasma glucose (A) and insulin (B) concentrations
at rest, and throughout the exercise and recovery periods
following the pre-exercise ingestion of Glu or GluF
Values are means+−
S.E.M, n =20 subjects. There were significant carbohydrate×
time interactions for plasma insulin (P =0.003) as determined by repeatedmeasures
ANOVA. *P <0.05 compared with Glu.
variables) on glycaemia, insulinaemia, adrenaline, noradrenaline,
TAG, NEFA, oxLDL and LPO (as dependent
variables) was analysed by ANOVA with repeated
measures [2 (group)×7 (time)]. A Sidak correction was
used to adjust the P value in relation to the number of
comparisons that were performed. P<0.05 was adopted
for statistical significance. For all of the statistical tests,
we used the SPSS 11.5 package for MicrosoftWindows.
RESULTS
Plasma glucose, insulin and
urinary catecholamines
Plasma insulin concentrations were higher in the GluF
compared with the Glu trials during the exercise and
recovery phases (t15 and t75; P<0.05), without any
differences being found in the glucose concentrations
between the two interventions during these phases
(Figure 2). Lower urinary levels of adrenaline and
noradrenaline (Figure 3) were found at the end of the
exercise phase following GluF compared with Glu (t45;
P<0.05).
Plasma TAG and NEFA concentrations
Following the ingestion of GluF, TAG levels were
significantly higher during the exercise phase (t15 and t30;
Figure 3 Urinary adrenaline (A) and noradrenaline (B)
concentrations at baseline, and at the end of the exercise
and recovery periods following the pre-exercise ingestion of
Glu or GluF
Values are means+−
S.E.M., n =20 subjects. There were significant carbohydrate×time
interactions for urinary adrenaline and noradrenaline (P =0.001
and P =0.02 respectively) as determined by repeated-measures ANOVA.
*P <0.05 compared with GluF.
P<0.05) and acute recovery phase (t45 and t75; P<0.05)
compared with following Glu intake (Figure 4A). NEFA
levels after GluF (Figure 4B)were lower halfway through
the exercise phase (t15; P<0.05) and during the recovery
phase (t75; P<0.05), but were higher at the end of the
exercise phase (t30; P<0.05) compared with Glu intake.
LPO and oxLDL concentrations
Following the ingestion of GluF, plasma levels of LPO
were significantly higher during the exercise and recovery
periods (t15, t30, t45 and t105; P<0.05), as were those of
oxLDL at the end of exercise (t30; P<0.05), compared
with following consumption of Glu alone (Table 3).
Blood lactate concentrations, perceived
exertion and HR
Nodifferences were observed in the levels of blood lactate
during either the exercise or recovery periods following
treatment with the supplements (Figure 5, bottom panel).
RPE and HR were significantly less afterGluF compared
with Glu (t15, t30 and t60; P<0.05) (Figure 5, top and
middle panels respectively).
C The Authors Journal compilation C 2009 Biochemical Society
142 J. M. Fern´andez and others
Figure 4 Plasma TAG (A) and NEFA (B) concentrations at
rest, and during the exercise and recovery periods following
the pre-exercise ingestion of Glu or GluF
Values are means+−
S.E.M., n =20 subjects. There were significant carbohydrate×time
interactions for plasma TAG and NEFA concentrations (P =0.001
and P =0.03 respectively) as determined by repeated-measures ANOVA.
*P <0.05 compared with Glu.
DISCUSSION
To the best of our knowledge, this is the first study
in healthy sportsmen to demonstrate that the addition
of a dose of fructose to the pre-exercise consumption of
a glucose supplement produces a cluster of acute
metabolic changes during moderate aerobic exercise and
the subsequent recovery period. These fructose-induced
metabolic dysregulations consist of a higher level of
insulinaemia than following the ingestion of glucose
alone, a greater increase in TAG and markers of lipid
oxidation, and a lower urinary concentration of catecholamines
and plasma NEFAs during and after aerobic
exercise.
Higher levels of insulin are required for a given
glucose concentration when the metabolic clearance of
glucose has been reduced [29]. In our present study,
insulinaemia was 37.2 and 25.8% higher during the
exercise and recovery periods (t15 and t75 respectively)
after GluF compared with Glu alone, without this
being accompanied by differences in glycaemic behaviour
between the twointerventions.These findingsmay reflect
both an improved tolerance to a higher carbohydrate
intake, when an extra dose of carbohydrate was
administered, such as fructose, but also a lower systemic
insulin sensitivity following that fructose ingestion.
Dirlewanger et al. [30] have demonstrated that the
acute infusion of fructose in healthy volunteers induces
hepatic and extrahepatic insulin resistance by doubling
the need for insulin required to maintain a glucose
steady-state. It has been suggested that an overload
of lipid metabolites derived from fructose and the
subsequent over-production of ROS (reactive oxygen
species), able to spread outside the liver, could interfere
with insulin signalling [31]. Mainly in animal models of
fructose-induced insulin resistance, it has been observed
that an impaired oxidative state was accompanied by
an inflammatory gene response mediated by activated
NF-κB (nuclear factor κB) and the participation of
inflammatory cytokines, such as muscle TNF-α (tumour
necrosis factor-α) [32–34]. In the present study, both
post-prandial phenomena, a greater increase in TAG and
oxidative damage to lipids, as shown byLPOand oxLDL,
in the GluF trial would suggest a greater amount of
insulinaemia mediated by inflammatory mechanisms.
Anothermetabolic change that has been described after
fructose consumption is an increase inTAGlevels [35,36].
Elevated NEFA and TAG levels have been reported in
humans even at 45 min after the ingestion of 50 g of
fructose [37]. In our present study, at 30 and 45 min after
the ingestion of GluF (half-way through and at the end
of the exercise period) we observed TAG concentrations
Table 3 LPO and oxLDL measured immediately before Glu or GluF supplementation (t−15) and at their respective exercise
(t 0, t 15 and t 30) and recovery (t 45 and t 105) phases
Values are means+−
S.E.M. *P <0.05 compared with the Glu group. †Values adjusted for body mass index.
Time (min)
Variable t−15 t 0 t 15 t 30 t 45 t 105
LPO (nmol/dl)
Glu trial 0.36+−
0.01 0.36+−
0.01 0.36+−
0.02 0.35+−
0.02 0.35+−
0.02 0.39+−
0.01
GluF trial 0.36+−
0.01 0.38+−
0.02 0.40+−
0.01* 0.40+−
0.02* 0.40+−
0.02* 0.42+−
0.01*
OxLDL (ng/ml)†
Glu trial 83.21+−
13.38 − 89.24+−
14.57 96.29+−
15.96 94.46+−
14.06 89.59+−
13.93
GluF trial 82.35+−
15.12 − 90.00+−
15.98 105.26+−
16.15* 97.17+−
16.01 89.55+−
14.15
C The Authors Journal compilation C 2009 Biochemical Society
Metabolic effects of fructose during exercise 143
Figure 5 RPR, HR and blood lactate at rest, the middle
and end of exercise and its recovery period following the
pre-exercise ingestion of Glu or GluF
Values are means+−
S.E.M., n =20 subjects. There were significant carbohydrate×time
interactions for RPE (P =0.032) by repeated-measures ANOVA.
*P <0.05 compared with GluF. Lac, lactate.
that were 38.7 and 34.2%higher than afterGlu at the same
times. Moreover, TAG rapidly returned to basal levels
45 min after its peak level (t75) following GluF, discarding
the hypothesis of a delay in TAG clearance as one of the
possible causes of hypertriacylglycerolaemia, as has been
proposed by Chong et al. [15]. Furthermore, the higher
level of insulin found with GluF leads us to suggest
that the mechanism making the greatest contribution to
the increase in TAG was not a reduction in the activation
of lipoprotein lipase in adipose tissue and the
subsequent delay in lipid clearance [38], but rather de
novo lipogenesis, a mechanism widely explained after
fructose intake [13,39].
Along these lines, several studies have reported that
the fast arrival of fructose at the liver could cause an
overload of the pentose phosphate pathway leading to
the acute expression of lipogenic genes [40,41], and to the
rapid activation of hepatic lipogenesis and the secretion
of VLDL via SREBP-1c (sterol-regulatory-elementbinding
protein-1c) [42]. Under these conditions, the
mechanisms resulting in a decrease in the glycaemic
response (e.g. the hepatic uptake of glucose and
glycogen synthesis) are possibly inhibited, whereas the
gluconeogenic and lipogenic pathways are activated [40].
The present study was not designed to investigate the
molecular mechanisms responsible for the increase in
TAG after GluF ingestion; however, it is possible that
the performance of aerobic exercise (non-glycolytic)may
have accelerated the arrival of fructose to the portal
system and the activation of the lipogenic pathway, which
may be key in explaining the lipid behaviour observed.
On the other hand, it is known that the utilization
of substrates during aerobic exercise is modulated by
autonomic and endocrine control in response to the
type and amount of carbohydrate ingested [43].Although
the present study does not provide results describing
substrate oxidation, it has been reported previously that
fructose by itself [44] as well as an elevated insulinaemia
during exercise [45] increase carbohydrate oxidation to
the detriment of the contribution of lipids to energy
consumption. Additionally, in agreement withMacLaren
et al. [46], we have found that elevated insulinaemia
was accompanied by both a lower concentration of
catecholamines in urine from t−15 to t30 of the exercise
period and a low availability of NEFAs in the GluF
trial. If lower activation of lipolysis followed GluF, as
suggested by these findings, the addition of a dose
of fructose would negate an important part of the
health benefits derived from 30 min of aerobic exercise.
Conversely, ifGluF increased the carbohydrate oxidation
rate, as reported by other studies with GluF [3,4], the
use of this more rapidly available fuel would explain
the minor effect on RPE and HR, without a change in
blood lactate. Thus the improvement in these variables
during and after exercise with GluF intake could increase
the capacity to perform exercise more comfortably, but
would not favour the metabolic benefits aimed at during
aerobic exercise.
In conclusion, the addition of fructose to the preexercise
ingestion of a glucose supplement triggers greater
insulin secretion and lower adrenergic stimulation than
the ingestion of glucose alone. This combination of
carbohydrates also induces a greater increase in plasma
TAGs, a deterioration in the oxidative state of circulating
lipids and the suppression of circulating NEFAs during
moderate-intensity aerobic exercise and the post-exercise
recovery phase. Future studies in risk populations (e.g.
C The Authors Journal compilation C 2009 Biochemical Society
144 J. M. Fern´andez and others
patients with diabetes and obese subjects) are required
in order to analyse the metabolic effects of supplements
containing different doses of fructose in mixed preexercise
meals or beverages.
FUNDING
This work was supported by grants from the Consejer´ıa
de Turismo, Comercio y Deporte de la Junta de
Andaluc´ıa; the Fundaci ´on Hospital Reina Sof´ıa Caja Sur;
and the CIBER Fisiopatolog´ıa Obesidad y Nutrici ´ on, as
an initiative of the Instituto de Salud Carlos III (ISCIII).
REFERENCES
1 American College of Sports Medicine, American
Dietetic Association & Dieticians of Canada (2000)
Nutrition and athletic performance (Joint Position
Statement). Med. Sci. Sports Exercise 32, 2130–2145
2 Okano, G., Takeda, H., Morita, I., Katoh, M., Mu, Z. and
Miyake, S (1988) Effect of pre-exercise fructose ingestion
on endurance performance in fed men. Med. Sci. Sports
Exercise 20, 105–109
3 Riddell, M. C., Bar-Or, O., Wilk, B., Parolin, M. L. and
Heigenhauser, G. J. (2001) Substrate utilization during
exercise with glucose and glucose plus fructose ingestion in
boys ages 10–14 yr. J. Appl. Physiol. 90, 903–911
4 Jentjens, R. L., Moseley, L., Waring, R. H., Harding, L. K.
and Jeukendrup, A. E. (2004) Oxidation of combined
ingestion of glucose and fructose during exercise.
J. Appl. Physiol. 96, 1277–1284
5 Moore, M. C., Cherrington, A. D., Mann, S. L. and Davis,
S. N. (2000) Acute fructose administration decreases the
glycemic response to an oral glucose tolerance test in
normal adults. J. Clin. Endocrinol. Metab. 85, 4515–4519
6 Heacock, P. M., Hertzler, S. R. and Wolf, B. W. (2002)
Fructose prefeeding reduces the glycemic response to a
high-glycemic index, starchy food in humans. J. Nutr. 132,
2601–2604
7 Wolf, B. W., Humphrey, P. M., Hadley, C. W., Maharry,
K. S., Garleb, K. A. and Firkins, J. L. (2002) Supplemental
fructose attenuates postprandial glycemia in Zucker fatty
fa/fa rats. J. Nutr. 132, 1219–1223
8 Barbe, P., Mollet, L., Galitzky, J., Lafontan, M. and
Berlan, M. (1996) In situ assessment of the role of β-1,
β-2 and β-3 adrenoceptors in the control of lipolysis and
nutritive blood flow in human subcutaneous adipose
tissue. Br. J. Pharmacol. 117, 907–913
9 Campbell, P. J., Carlson, M. G., Hill, J. O. and Nurjhan, N.
(1992) Regulation of free fatty acid metabolism by insulin
in humans: role of lipolysis and reesterification.
Am. J. Physiol. 263, E1063–E1069
10 Arner, P., Kriegholm, E., Engfeldt, P. and Bolinder, J.
(1990) Adrenergic regulation of lipolysis in situ at rest and
during exercise. J. Clin. Invest. 85, 893–898
11 Elliott, S. S., Keim, N. L., Stern, J. S., Teff, K. and Havel,
P. J. (2002) Fructose, weight gain, and the insulin resistance
syndrome. Am. J. Clin. Nutr. 76, 911–922
12 Daly, M. (2003) Sugars, insulin sensitivity, and the
postprandial state. Am. J. Clin. Nutr. 78, 865S–872S
13 Schwarz, J. M., Neese, R., Shackleton, C. H. et al. (1993)
De novo lipogenesis during fasting and oral fructose in lean
and obese hyperinsulinemic subjects. Diabetes 42 (Suppl.),
39A
14 Frayn, K. N. and Kingman, S. M. (1995) Dietary sugars
and lipid metabolism in humans. Am. J. Clin. Nutr. 62,
250S–261S
15 Chong, M. F. F., Fielding, B. A. and Frayn, K. N. (2007)
Mechanisms for the acute effect of fructose on postprandial
lipemia. Am. J. Clin. Nutr. 85, 1511–1520
16 Lˆe, K. A. and Tappy, L. (2006) Metabolic effects of
fructose. Curr. Opin. Clin. Nutr. Metab. Care 9,
469–475
17 Truswell, A. S. (1994) Food carbohydrates and plasma
lipids-an update. Am. J. Clin. Nutr. 59, 710S–718S
18 Mettler, S., Lamprecht-Rusca, F., Stoffel-Kurt, N.,
Wenk, C. and Colombani, P. C. (2007) The influence
of the subjects’ training state on the glycemic index.
Eur. J. Clin. Nutr. 61, 19–24
19 Bantle, J. P. (2006) Is fructose the optimal low glycemic
index sweetener?In Nutritional Management of Diabetes
Mellitus and Dysmetabolic Syndrome, Nestl´e Nutr,
Workshop Ser. Clin. Perform. Program., vol. 11 (Bantle,
J.P. and Slama, G., eds), pp. 83–95, Vevey/Karger, Basel
20 Liu, S., Manson, J. E., Stampfer, M. J. et al. (2001) Dietary
glycemic load assessed by food-frequency questionnaire in
relation to plasma high-density-lipoprotein cholesterol and
fasting plasma triacylglycerols in postmenopausal women.
Am. J. Clin. Nutr. 73, 560–566
21 Jenkins, D. J. A., Wolever, T. M. S. and Taylor, R. H.
(1981) Glycemic index of foods: a physiological basis
for carbohydrate exchange. Am. J. Clin. Nutr. 34,
362–366
22 American College of Sports Medicine (2000) ACSM’s
Guidelines for Exercise Testing and Prescription, 6th edn.,
pp. 105–106,Williams & Wilkins, Baltimore
23 Davis, J. A. (1985) Anaerobic threshold: a review of the
concept and directions for future research. Med. Sci. Sports
Exercise 17, 6–18
24 Fujisawa, T., Mulligan, K., Wada, L., Schumacher, L.,
Riby, J. and Kretchmer, N. (1993) The effect of exercise
on fructose absorption. Am. J. Clin. Nutr. 58, 75–79
25 Riby, J. E., Fujisawa, T. and Kretchmer, N. (1993) Fructose
absorption. Am. J. Clin. Nutr. 58, 748S–753S
26 Skoog, S. M. and Bharucha, A. E. (2004) Dietary
fructose and gastrointentinal symptoms: a review.
Am. J. Gastroenterol. 99, 2046–2050
27 Borg, G. (1998) The Borg CR10 scale. Borg’s Perceived
Exertion and Pain Scales, pp. 137–138, Human Kinetics,
Champaign, IL
28 Day, M. L., McGuigan, M. R., Brice, G. and Foster, C.
(2004) Monitoring exercise intensity during resistance
training using the session RPE scale. J. Strength Cond. Res.
18, 353–358
29 Radziuk, J. (2000) Insulin sensitivity and its measurement:
structural commonalities among the methods. J. Clin.
Endocrinol. Metab. 85, 4426–4433
30 Dirlewanger, M., Schneiter, P., Jequier, E., Tappy, L. et al.
(2000) Effects of fructose on hepatic glucose metabolism in
humans. Am. J. Physiol. Endocrinol. Metab. 279,
E907–E911
31 Wei, Y. and Pagliassotti, M. J. (2004) Hepatospecific effects
of fructose on c-jun NH2-terminal kinase: implications for
hepatic insulin resistance. Am. J. Physiol. Endocrinol.
Metab. 287, E926–E933
32 Son, D., Hutchings, S. and Pang, C. C. (2005) Chronic
N-acetylcysteine prevents fructose induced insulin
resistance and hypertension in rats. Eur. J. Pharmacol. 508,
205–210
33 Armutcu, F., Coskun, O., Gurel, A. et al. (2005) Thymosin
α1 attenuates lipid peroxidation and improves
fructose-induced steatohepatitis in rats. Clin. Biochem. 38,
540–547
34 Togashi, N., Ura, N., Higashiura, K., Murakami, H. and
Shimamoto, K. (2000) The contribution of skeletal muscle
tumor necrosis factor-α to insulin resistance and
hypertension in fructose fed rats. J. Hypertens. 18,
1605–1610
35 Lˆe, K. A., Faeh, D., Stettler, R. et al. (2006) A 4-wk
high-fructose diet alters lipid metabolism without affecting
insulin sensitivity or ectopic lipids in healthy humans.
Am. J. Clin. Nutr. 84, 1374–1379
36 Park, O. J., Cesar, D., Faix, D., Wu, K., Shackleton, C. H.
and and Hellerstein, M. K. (1992) Mechanisms of
fructose-induced hypertriglyceridaemia in the rat:
activation of hepatic pyruvate dehydrogenase through
inhibition of pyruvate dehydrogenase kinase. Biochem. J.
282, 753–757
C The Authors Journal compilation C 2009 Biochemical Society
Metabolic effects of fructose during exercise 145
37 Nuttall, F. Q., Khan, M. A. and Gannon, M. C. (2000)
Peripheral glucose appearance rate following fructose
ingestion in normal subjects. Metab. Clin. Exp. 49,
1565–1571
38 Brunzell, J. D., Hazzard, W. R., Porte, Jr, D. and Bierman,
E. L. (1973) Evidence for a common, saturable, triglyceride
removal mechanism for chylomicrons and very low density
lipoproteins in man. J. Clin. Invest. 52, 1578–1585
39 Faeh, D., Minehira, K., Schwarz, J. M., Periasamy, R., Park,
S. and Tappy, L. (2005) Effect of fructose overfeeding and
fish oil administration on hepatic de novo lipogenesis and
insulin sensitivity in healthy males. Diabetes 54, 1907–1913
40 Wei, Y., Bizeau, M. E. and Pagliassotti, M. J. (2004) An
acute increase in fructose concentration increases hepatic
glucose-6-phosphatase mRNA via mechanisms that are
independent of glycogen synthase kinase-3 in rats. J. Nutr.
134, 545–551
41 Kabashima, T., Kawaguchi, T., Wadzinski, B. E. and
Uyeda, K. (2000) Xylulose-5-phosphate mediates
glucose-induced lipogenesis by xylulose-5-
phosphate-activated protein phosphatase in rat liver.
Proc. Natl. Acad. Sci. U.S.A. 100, 5107–5112
42 Miyazaki, M., Dobrzyn, A., Man, W. C. et al. (2004)
Stearoyl-CoA desaturase 1 gene expression is necessary for
fructose-mediated induction of lipogenic gene expression
by sterol regulatory element-binding protein-1cdependent
and -independent mechanisms. J. Biol. Chem.
279, 25164–25171
43 Borer, K. T. (2003) Hormonal regulation of fuel use in
exercise. Exercise Endocrinology, pp. 97–118, Human
Kinetics, Champaign, IL
44 Tappy, L., Randin, J. P., Felber, J. P. et al. (1986)
Comparison of thermogenic effect of fructose and glucose
in normal humans. Am. J. Physiol. 250, E718–E724
45 Mandarino, L. J., Wright, K. S., Verity, L. S. et al. (1987)
Effects of insulin infusion on human skeletal muscle
pyruvate dehydrogenase, phosphofructokinase, and
glycogen synthase. Evidence for their role in oxidative and
nonoxidative glucose metabolism. J. Clin. Invest. 80,
655–663
46 MacLaren, D. P. M., Reilly, T., Campbell, I. T. and
Hopkin, C. (1999) Hormonal and metabolic responses to
maintained hyperglycemia during prolonged exercise.
J. Appl. Physiol. 87, 124–131
Received 10 April 2008/28 May 2008; accepted 5 June 2008
Published as Immediate Publication 5 June 2008, doi:10.1042/CS20080120
C The Authors Journal compilation C 2009 Biochemical Society
). Est-ce que c'est une vérité un peu enjolivée pour forcer les ventes du complément de protéines en question ? puisqu'il les vend lui même.... je sais pas.
