AJCN. First published ahead of print September 8, 2010 as doi: 10.3945/ajcn.2010.29566.
Fructose and glucose co-ingestion during prolonged exercise increases lactate and glucose fluxes and oxidation compared with an equimolar intake of glucose1–3 Virgile Lecoultre, Rachel Benoit, Guillaume Carrel, Yves Schutz, Gre´goire P Millet, Luc Tappy, and Philippe Schneiter ABSTRACT Background: When fructose is ingested together with glucose (GLUFRU) during exercise, plasma lactate and exogenous carbohydrate oxidation rates are higher than with glucose alone. Objective: The objective was to investigate to what extent GLUFRU increased lactate kinetics and oxidation rate and gluconeogenesis from lactate (GNGL) and from fructose (GNGF). Design: Seven endurance-trained men performed 120 min of exer_ 2max (maximal oxygen consumption) while incise at ’60% VO gesting 1.2 g glucose/min + 0.8 g of either glucose or fructose/min (GLUFRU). In 2 trials, the effects of glucose and GLUFRU on lactate and glucose kinetics were investigated with glucose and lactate tracers. In a third trial, labeled fructose was added to GLUFRU to assess fructose disposal. Results: In GLUFRU, lactate appearance (120 6 6 lmol kg21 min21), lactate disappearance (121 6 7 lmol kg21 min21), and oxidation (127 6 12 lmol kg21 min21) rates increased significantly (P , 0.001) in comparison with glucose alone (94 6 16, 95 6 16, and 97 6 16 lmol kg21 min21, respectively). GNGL was negligible in both conditions. In GLUFRU, GNGF and exogenous fructose oxidation increased with time and leveled off at 18.8 6 3.7 and 38 6 4 lmol kg21 min21, respectively, at 100 min. Plasma glucose appearance rate was significantly higher (P , 0.01) in GLUFRU (91 6 6 lmol kg21 min21) than in glucose alone (82 6 9 lmol kg21 min21). Carbohydrate oxidation rate was higher (P , 0.05) in GLUFRU. Conclusions: Fructose increased total carbohydrate oxidation, lactate production and oxidation, and GNGF. Fructose oxidation was explained equally by fructose-derived lactate and glucose oxidation, most likely in skeletal and cardiac muscle. This trial was registered at clinicaltrials. gov as NCT01128647. Am J Clin Nutr doi: 10.3945/ajcn.2010. 29566. INTRODUCTION
Regardless of the amount of exogenous glucose intake during exercise, there is a maximal exogenous oxidation rate of ’1 g/min. However, adding fructose to glucose during exercise has been shown to further increase exogenous carbohydrate oxidation rate (1–6) up to 1.75 g/min (7). Although the mechanism responsible for this increased exogenous carbohydrate oxidation is unknown, it has been suggested that the major limiting factor of exogenous carbohydrate oxidation is intestinal transport, and that it can be increased if fructose, which uses a molecular absorption pathway distinct from glucose, is co-ingested with glucose (1–3, 6, 8).
Fructose metabolism, however, differs markedly from glucose metabolism. Fructose is essentially taken up by intestinal and liver cells and is rapidly and almost completely converted into triose-phosphates (9). As a result, systemic fructose concentrations only increase marginally after fructose ingestion (10). Furthermore, because nonsplanchnic cells lack the enzyme fructokinase, and because hexokinase has a markedly lower affinity for fructose than for glucose, it is very unlikely that fructose is directly metabolized in skeletal muscle. Instead, it appears that fructose is essentially converted into glucose and lactate in liver cells to be subsequently oxidized in extrahepatic tissues (11, 12). It is well established that when fructose and glucose are ingested simultaneously during exercise, plasma lactate concentrations are significantly increased, compared with isoenergetic glucose or glucose polymers ingestion (2, 3, 6–8), which suggests that systemic lactate fluxes increase when mixtures of glucose and fructose are ingested. It was recently shown that when a lactate polymer (polylactate) was ingested simultaneously with glucose and fructose, both polylactate and fructose exhibited similar patterns of oxidation, suggesting rapid conversion of fructose into lactate before oxidation (13). As such, a beneficial effect of fructose and glucose co-ingestion on muscle performance would be expected because lactate is efficiently oxidized by active muscles during exercise (14, 15). Moreover, oral intakes of lactate polymers have been shown to be as efficient as oral glucose in maintaining glycemia during prolonged exercise (16) and to be oxidized without prior conversion into glucose (17). The fate of oral fructose administered with glucose during exercise, and more specifically the rate of lactate production elicited after fructose administration, has not been quantitatively assessed. The aim of this study was to assess the metabolic effects 1
From the Department of Physiology (VL, RB, GC, YS, GPM, LT, and PS), and the Institute of Sport Sciences (ISSUL) (GPM), Faculty of Biology and Medicine, University of Lausanne, Lausanne, Switzerland, and the Service of Endocrinology, Diabetes, and Metabolism, Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland (LT). 2 Supported by a grant from the Eidgeno¨ssische Sport Kommission, Magglingen, Switzerland. 3 Address correspondence to L Tappy, Rue du Bugnon 7, CH-1005 Lausanne, Switzerland. E-mail:
[email protected]. Received March 22, 2010. Accepted for publication August 17, 2010. doi: 10.3945/ajcn.2010.29566.
Am J Clin Nutr doi: 10.3945/ajcn.2010.29566. Printed in USA. Ó 2010 American Society for Nutrition
Copyright (C) 2010 by the American Society for Nutrition
1 of 9
2 of 9
LECOULTRE ET AL
of ingesting equimolar amounts of glucose or glucose + fructose mixtures in trained athletes during exercise.
Plasma hormones and substrate concentrations were measured immediately before and throughout the test at 20-min intervals.
SUBJECTS AND METHODS
Preliminary visit
Ethics This study was performed in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of the Faculty of Biology and Medicine, University of Lausanne (protocol #59/09). Participants Seven healthy, well-trained male endurance cyclists gave written informed consent to participate in this study. Inclusion criteria were age between 20 and 40 y, 3 y of cycling endurance training, body mass index (BMI; in kg/m2) between 18 and 25, and taking no medication 1 wk before and during the study. Subjects’ characteristics are shown in Table 1. Overall study design The overall study consisted of a preliminary session during _ 2max (maximal oxygen consumption) and maximal which VO aerobic power (MAP) output were assessed, followed by three 120-min submaximal exercise trials performed in a single-blind, randomized, crossover design. Visits were separated by 5–10 d. The 3 exercise trials aimed at comparing the metabolic effects of oral glucose (ingested at an average rate of 2 g/min) (GLU) or of an oral glucose + fructose (1.2 + 0.8 g/min) mixture (GLUFRU) during exercise. To evaluate glucose and lactate kinetics, total carbohydrate oxidation, and lactate oxidation, subjects were studied during a 2-h exercise protocol with oral administration of a glucose mixture (120 g every hour) or a glucose + fructose (72 + 48 g every hour) mixture. The 6,6-2H2 glucose and 13C3-lactate were infused to calculate glucose and lactate kinetics. Furthermore, indirect calorimetry was performed to measure total carbohydrate oxidation, and expired 13CO2 was monitored to calculate whole-body lactate oxidation. During the third exercise trial, the same procedures were performed in each subject with repeated administration of oral glucose + fructose labeled with 13C6-fructose to evaluate exogenous fructose metabolic fate and oxidation. Fructose conversion into lactate and glucose were evaluated by monitoring the systemic appearance of plasma 13C-labeled lactate and 13 C-labeled glucose. Total exogenous fructose oxidation was also measured by monitoring 13CO2 production. TABLE 1 Characteristics of study participants1 Mean 6 SD Age (y) Height (cm) Weight (kg) Fat mass (%) MAP (W) _ 2max (L/min) VO _ 2max/kg (mL min21 kg21) VO 1
tion.
30 179 69.5 12.0 365 5.1 73.4
6 6 6 6 6 6 6
5 6 3.4 2.3 25 0.3 5.3
_ 2max, maximal oxygen consumpMAP, maximal power output; VO
Within 2 wk of the beginning of the main trials, subjects reported to the laboratory for assessment of their anthropometric _ 2max. Volunteers’ characteristics and determination of their VO body weight and height were measured, and body fat was esti_ 2max and mated by skinfold thickness measurements (18). VO MAP were then determined during an incremental test to exhaustion performed on a cycle ergometer (Lode Excalibur; Lode, Groningen, Netherlands). First, subjects sat quietly on the ergometer for 5 min before starting pedaling at 60 W for 5 min. Power output was then increased by 35 W every 3 min until exhaustion. Throughout the incremental test, oxygen uptake _ 2) and carbon dioxide production (VCO _ (VO 2) were measured breath-by-breath (SensorMedics Vmax; Sensormedics Corp, Yorba Linda, CA), and heart rate was monitored continuously _ 2 was (Polar S810; Polar Electro Oy, Kempele, Finland). VO considered as maximal if 2 of the following criteria were met: _ 2 with increasing work rate, a respiratory a leveling off of VO exchange ratio above 1.05, and a maximal heart rate 10 beats of the predicted maximal heart rate. In all subjects, 2 criteria were met. MAP was calculated from the addition of the last complete step and the fraction of time spent in the following step multiplied by 35 W. Submaximal exercise trials For the 48 h before the test, subjects refrained from strenuous exercise. They were asked to record their diets during the 2 d before the first trial and to closely repeat the same dietary intakes before the 2 subsequent trials. It was thus assumed that glycogen concentrations were similar between trials. In addition, they received instructions to avoid foods naturally enriched in 13 C and to refrain from caffeine and alcohol. After fasting overnight, subjects reported at 0730 to the laboratory. After arrival, volunteers were asked to void and were then weighed. Next, a catheter was inserted into the antecubital vein (right arm), and an initial blood sample was drawn for baseline determination of plasma metabolites. Expired air samples were collected in quadruplicate for the assessment of baseline expired air 13CO2 isotopic enrichment. A second catheter was inserted into the left arm for stable-isotope tracer infusion. Ten minutes before the exercise trial started, a primed (11 lmol/kg) constant infusion of 0.11 lmol 6,6-2H2 glucose kg21 min21 _ 2 and VCO _ began. Subjects were asked to sit quietly with VO 2 being measured breath-by-breath (SensorMedics Vmax; Sensormedics Corp). After 9 min of rest, subjects took their position on the cycle ergometer. The 6,6-2H2 glucose infusion rate was increased to 0.44 lmol kg21 min21, and subjects started pedaling at an intensity of 50% of their individual MAP after 1 min. After starting exercise, the mouthpiece was removed, and a 600-mL carbohydrate drink was provided, which had to be consumed within 3 min. To provide a total of 2 L of the drink and 240 g of carbohydrate over the 120 min of exercise, 280 mL of the tested solution was provided thereafter, every 20 min during 100 min of exercise. The test drink contained either 120 g glucose/L (D-Glucose, Fluka Analytic; Sigma Aldrich, Buchs, Switzerland) (GLU) or
3 of 9
GLUCOSE AND FRUCTOSE CO-INGESTION DURING EXERCISE
72 and 48 g fructose/L (D-Fructose, Fluka Analytic; Sigma Aldrich) (GLUFRU), as well as 20 mmol NaCl/L and 20 mL lemon juice/L. _ 2 and For every 20 min of exercise, a measurement of VO _ VCO was performed for 5 min. At this time, a venous blood 2 sample was drawn, ratings of perceived exertion (RPE) were recorded, and a breath sample was obtained before subjects ingested 280 mL of the tested carbohydrate solution. These procedures were thus performed at 15, 35, 55, 75, 95 and 115 min. Participants did not receive any drink at the last time point. Lactate kinetics were assessed by using a primed (42.5 lmol/kg) constant infusion (2.25 lmol kg21 min21) of 13C3-lactate (Cambridge Isotopes, Andover, MA) during 40–120 min of exercise. Five subjects received the simultaneous primed constant infusion of labeled lactate and deuterated glucose, and 2 subjects underwent the exercise trial with only the infusion of deuterated glucose. The metabolic fate of fructose was assessed by using identical procedures, without labeled lactate infusion. However, the ingested fructose was labeled with artificially labeled U13Cfructose (Cambridge Isotopes). The enrichment of fructose in the drink was 1.44837 atom percent. All subjects underwent this trial. During the trial, heart rate was obtained beat-to-beat (Polar S810; Polar Electro Oy). At the end of the test volunteers were provided with a carbohydrate-rich snack and could rest. Analytic procedures Blood was collected on lithium heparin for measurement of glucose, fructose, lactate and tracers; with EDTA-coated tubes for measurement of insulin, cortisol, and free fatty acids (FFAs); or with EDTA-trasylol for measurement of glucagon. After blood collection, plasma was immediately separated from blood by centrifugation at 4°C for 10 min at 3600 g and stored at 220°C until analysis. Lactate concentration was measured enzymatically by using kits from Boehringer (Boehringer Mannheim, Mannheim, Germany). Plasma glucose concentration was measured by the glucose oxidase method with a Beckman glucose analyzer II (Beckman Instruments, Fullerton, CA). Colorimetric methods were used to assess plasma concentrations of FFAs (kit from Wako Chemicals, Freiburg, Germany). Commercial radioimmunoassay kits were used for the determination of plasma cortisol, insulin, and glucagon (Linco Research, St Charles, MO). Plasma 13Clactate and deuterated glucose abundance were measured by gas chromatography–mass spectrometry (GC-MS) (Hewlett-Packard Instruments, Palo Alto, CA), as described (19, 20). Plasma 13C-glucose and expired air 13CO2-isotopic enrichment were measured by isotope-ratio mass spectrometry (SerCon Ltd, Crewe, United Kingdom), as previously described (21, 22). Plasma fructose concentrations were measured by using the protocol from Petersen et al (23). Briefly, 0.1 mL mannitol (0.55 mmol/L) was added per 0.2 mL plasma. The plasma was deproteinized with 0.3 Newton (N) barium hydroxide and 0.3 N zinc sulfate and partially purified by sequential cation-anion exchange columns. After drying, the samples were derivatized with acetic anhydride and pyridine, and plasma fructose concentrations were measured by using GC-MS with electron impact ionization.
Calculations Plasma glucose and rates of lactate appearance and disappearance and metabolic clearance rate (MCR) were calculated from plasma 6,6-2H2 glucose and 13C3-lactate isotopic enrichment (IE), respectively, by using the Steeles’ equations for non–steady state conditions (24). The volume of distribution was considered to be 180 mL/kg for glucose and 100 mL/kg for lactate (25, 26). The rate of appearance of lactate from fructose (RaLF) (lmol kg21 min21) was calculated from the lactate Ra (RaL), 13Clactate, and 13C-fructose isotopic enrichment as follows: RaLF ¼
lactate IE RaL fructose IE
ð1Þ
where lactate IE and fructose IE are the enrichment of plasma lactate and ingested fructose, respectively. Gluconeogenesis from lactate was calculated from lactate incorporation into glucose (27, 28) GNGL ¼
glucose IE glucose Ra lactate IE k
ð2Þ
where GNGL (lmol kg21 min21) was the gluconeogenesis from lactate, and glucose IE and lactate IE are the isotopic enrichment of plasma glucose and lactate, respectively. k represents the correction factor for the loss of carbon labels in the tricarboxylic acid cycle during GNGL, which was assumed to be 0.67. Gluconeogenesis from fructose (GNGF) was calculated as GNGF ¼
glucose IE glucose Ra fructose IE k
ð3Þ
where GNGF (lmol kg21 min21) is the gluconeogenesis from fructose, and glucose IE and fructose IE are the isotopic enrichments of plasma glucose and fructose, respectively. k represents the correction factor for the loss of carbon labels in the tricarboxylic acid cycle during GNGL, which was assumed to be 0.67. Substrate oxidation Carbohydrate and fat oxidation rates (g/min) were calculated from respiratory gas exchange (29). Lactate oxidation rates were calculated from 13C abundance in expired CO2 and plasma lactate. Plasma lactate oxidation rate (lmol kg21 min21) was calculated without taking into account secondary labeling of glucose because gluconeogenesis was almost negligible (see Results) (25). _ Plasma lactate oxidation ¼ VCO 2
CO2 IE 1 ð4Þ lactate IE 22:4 3
where CO2 IE and lactate IE are the isotopic enrichment of expired CO2 and plasma lactate; 22.4 3 corresponds to the volume of CO2 per mol of lactate oxidized. Recovery of metabolic 13 CO2 in breath was considered to be complete. Exogenous fructose oxidation was calculated as follows (20): _ Exogenous fructose oxidation ðg=minÞ ¼ VCO 2 : CO2 IE VCO2 180 0:134 fructose IE 0:8 106
ð5Þ
where CO2 IE and fructose IE are the isotopic enrichment of breath CO2 and ingested fructose expressed in atom percent
4 of 9
LECOULTRE ET AL
_ excess (APE), VCO 2 is the total respiratory CO2 (mL/min), 0.8 is the recovery of 13C from fructose in breath CO2, 0.134 is the volume of CO2 (mL) produced by oxidation of 1 lmol fructose, 180 is the molar weight of fructose, and 106 represents the conversion from micrograms to grams. Total gluconeogenesis from fructose and lactate and lactate production and oxidation from fructose were calculated over the last 40 min of exercise, assuming that fructose absorption had reached steady state and was equal to the intake (ie, 0.8 g/min). Nonoxidative fructose disposal was considered to be the difference between fructose absorption and oxidation rates. Heart rate data were averaged every 5 s from beat-to-beat signal by using linear interpolation after removal of artifacts using “R” (30).
Student’s paired t tests were used to identify differences between single parameters for normally distributed data, and Wilcoxon’s signed rank tests were used for nonnormally distributed data. The respective change of exogenous fructose oxidation rate and gluconeogenesis from fructose was analyzed with a onefactor ANOVA. When applicable, multiple comparisons were performed with a Tukey post hoc test. All statistical computations were made by using “R,” an open source statistical software (30). P values , 0.05 were considered significant. All data are expressed as means 6 SDs. RESULTS
Tracers 13
Statistical analysis Data for the 2 GLUFRU trials were averaged; skewed distributions were log-transformed before statistical calculations. Analysis of the effect of treatment (GLU compared with GLUFRU) over time on metabolic variables and substrate kinetics was performed by using a 2-factor (time · condition) repeated-measures analysis of variance (ANOVA). When applicable, Tukey post hoc tests were performed.
C Enrichments of plasma lactate and glucose and expired CO2 are presented in Figure 1. Plasma lactate enrichment was significantly (P , 0.01) higher in GLU compared with GLUFRU during U13C-lactate infusion, reaching ’2.5 and 2 molar percent excess (MPE), respectively (Figure 1A). With 13 C-fructose ingestion, 13C-lactate enrichment reached steady state between 40 and 120 min at 0.39 6 0.06 MPE (Figure 1D). Glucose 13C enrichment remained low when labeled lactate was infused, reaching 0.03 and 0.02 APE in GLU and GLUFRU,
FIGURE 1. Mean (6SD) isotopic enrichment of plasma lactate (A and D), plasma glucose (B and E), and expired (Exp.) carbon dioxide (CO2) (C and F) during 13C-labeled lactate infusion (A, B, and C; n = 5) and during 13C-fructose ingestion (D, E, and F; n = 7). Enrichment of plasma lactate is expressed as molar percent excess (MPE). Enrichment of plasma glucose and expired CO2 are expressed as atom percent excess (APE). Open triangles represent glucosealone (GLU) and black squares represent glucose + fructose (GLUFRU) conditions during the lactate turnover trial. Black squares represent the GLUFRU condition with ingestion of 13C-labeled fructose. Note that the scales for the y axis for the 13C-lactate trials (panels A, C, and E) and for the 13C-fructose trials (panels B, D, and F) are different. All analyses were performed by 2-factor repeated-measures ANOVA in panels A–C. A: P , 0.01 for treatment, P , 0.001 for time, and P , 0.01 for treatment · time interaction. B: P = 0.01 for treatment, P , 0.001 for time, and P = 0.13 for treatment · time interaction. C: P = 0.34 for treatment, P , 0.001 for time, and P , 0.21 for treatment · time interaction. D–F: P , 0.001 for time (one-factor ANOVA with repeated-measures). *,**Significant difference between GLU and GLUFRU: *P , 0.05, **P , 0.01 (Tukey post hoc test). #Significant difference between 0 and 20 min (D) and between 0, 20, 40, 60, and 80 min (E and F) of exercise (P , 0.05, Tukey post hoc test).
GLUCOSE AND FRUCTOSE CO-INGESTION DURING EXERCISE
respectively (Figure 1B). Plasma glucose enrichment was significantly higher in GLU. In contrast, when 13C-fructose was ingested, glucose enrichment increased up to 0.2 APE (Figure 1E) at 100 min of exercise. With 13C-lactate infusion, expired 13 CO2 enrichment was 0.37 6 0.02 and 0.36 6 0.02 APE in GLU and GLUFRU, respectively (Figure 1C). With 13C-fructose ingestion, 13CO2 enrichment increased with time to reach a plateau of ’0.15 APE after 100 min (Figure 1F). Plasma metabolites At rest, plasma lactate concentration was 1.27 6 0.36 and 1.27 6 0.39 mmol/L in GLU and GLUFRU, respectively (Figure 2). It decreased slightly (1.21 6 0.26 mmol/L) and remained stable during exercise in GLU. As expected, plasma lactate concentration increased to 1.75 6 0.40 mmol/L in GLUFRU, which was significantly higher (2-factor repeatedmeasures ANOVA: treatment, P , 0.001; time, P , 0.001; treatment · time, P , 0.01) than in GLU (Figure 2A). Plasma glucose concentration increased at the onset of exercise in both conditions and was, on average, 6.2 6 0.9 mmol/L in both conditions (Figure 2B) with no difference between GLU and GLUFRU. Plasma fructose concentration was close to zero at the onset of exercise and increased significantly with time, reaching a plateau of 0.47 mmol/L after 80 min of exercise in GLUFRU (Figure 2D). Plasma FFAs were comparable in GLU and GLUFRU at rest (0.36 6 0.12 and 0.33 6 0.15 mmol/L, respectively). Plasma
5 of 9
FFAs decreased significantly with time during exercise (2-factor repeated-measures ANOVA: time effect, P , 0.001) and were on average significantly lower in GLUFRU (0.12 6 0.07 mmol/L) than in GLU (0.23 6 0.09 mmol/L) (2-factor repeated-measures ANOVA: treatment main effect, P , 0.01; Figure 2C). Plasma insulin concentrations (Figure 2E) were 6.1 6 1.9 and 6.1 6 1.2 lU/mL at rest in GLU and GLUFRU, respectively, and increased over time in response to carbohydrate ingestion (P , 0.001) with no difference between conditions. After 120 min, insulin concentrations were 8.1 6 1.9 and 9.3 6 2.6 lU/mL in GLU and GLUFRU, respectively. Plasma glucagon concentrations (Figure 2F) were comparable in GLU and GLUFRU at rest (44.2 6 11.4 and 45.9 6 11.6 pg/mL). During exercise, plasma glucagon concentrations increased slightly in both groups but were marginally lower in GLUFRU than in GLU (2-factor repeated-measures ANOVA: treatment, P , 0.01; time, P , 0.001; treatment · time, P = 0.31). Plasma cortisol concentration was similar between GLU and GLUFRU at rest (420 6 87 compared with 470 6 90 pmol/L) and during exercise (382 6 124 compared with 378 6 112 pmol/L). Substrate oxidation and kinetics In the GLU condition, total carbohydrate and fat oxidations were 3.1 6 0.4 and 0.35 6 0.16 g/min, respectively (not shown in tables or figures). Lactate rate of appearance (Ra) was 94 6 16 lmol kg21 min21, lactate rate of disappearance (Rd) was 95 6 16 lmol
FIGURE 2. Mean (6SD) plasma substrate concentrations in response to ingestion of glucose alone (GLU) or glucose + fructose (GLUFRU). For all variables, n = 7. All analyses were performed by 2-factor repeated-measures ANOVA in panels A–C, E, and F. A: P , 0.001 for treatment, P , 0.01 for time, and P , 0.01 for treatment · time interaction. B: P = 0.78 for treatment, P , 0.001 for time, and P = 0.87 for treatment · time interaction. C: P , 0.001 for treatment, P , 0.001 for time, and P , 0.28 for treatment · time interaction. D: P , 0.001 for time (one-factor repeated-measures ANOVA). E: P , 0.001 for treatment, P , 0.001 for time, and P = 0.068 for treatment · time interaction. F: P , 0.01 for treatment, P , 0.001 for time, and P = 0.31 for treatment · time interaction. *,**Significant difference between GLU and GLUFRU treatments: *P , 0.05, **P , 0.01 (Tukey post hoc test). #Significant difference between 0, 20, 40, and 60 min (P , 0.05, Tukey post hoc test). FFA, free fatty acids.
6 of 9
LECOULTRE ET AL
kg21 min21, and its oxidation was 97 6 16 lmol kg21 min21 (Figure 3). In GLU, lactate MCR was observed to be 75 6 13 mL kg21 min21. Oxidized Rd lactate was 102 6 12% in these conditions. Glucose Ra, Rd, and MCR were, on average, 75.9 6 9.5 lmol kg21 min21, 77.7 6 10.4 lmol kg21 min21, and 12.9 6 2.8 mL kg21 min21, respectively (Figure 4). Data obtained during the 13C3-lactate infusion indicated that liver gluconeogenesis from lactate (GNGL) was negligible, at most 2.0 lmol kg21 min21 (Figure 5). In the GLUFRU condition, and compared with GLU, total carbohydrate oxidation was significantly (P , 0.05) higher by ’7%, reaching 3.3 6 0.3 g/min, whereas fat oxidation rate was lower, on average 0.27 6 0.18 g/min (P , 0.05, not shown in tables or figures). Significantly higher lactate kinetics and oxidation rate (2-factor repeated-measures ANOVA: treatment main effect, P , 0.001) as well as glucose kinetics (P , 0.01) were observed in GLUFRU as compared with GLU. On average, lactate Ra, Rd, and oxidation rate were higher by ’30%, at 120 6 6, 121 6 7, and 127 6 12 lmol kg21 min21, respectively (Figure 3, A–C). In GLUFRU, lactate MCR (71 6 15 mL kg21 min21) was, however, not different from that in GLU (Figure 3D). Oxidized Rd lactate was 104 6 11% in GLUFRU and 102 6 12% in GLU. Glucose Ra and Rd were, on average, higher (2-factor repeatedmeasures ANOVA: treatment main effect, P , 0.01) in GLUFRU compared with GLU (Figure 4) by ’10%, at 83 6 13 and 84 6 15 lmol kg21 min21, respectively. In addition, glucose MCR was higher by 8%, at 13.7 6 2.6 mL kg21 min21. Despite the higher lactate kinetics and plasma lactate concentration in GLUFRU, GNGL remained negligible (Figure 5). With the use of labeled fructose, total exogenous fructose oxidation reached 38 6 4 lmol kg21 min21 or ’0.48 6 0.1 g/min during the later stages of the exercise trial (ie, after 100 min). Gluconeogenesis from fructose (GNGF) increased
FIGURE 3. Mean (6SD) lactate rate of appearance (Ra), rate of disappearance (Rd), metabolic clearance rate (MCR), and oxidation rate (ox.) during the lactate infusion trial. *,**Significant difference between glucose-alone (GLU) and glucose + fructose (GLUFRU) conditions: *P , 0.05, **P , 0.01 (Student’s paired t test). For all variables, n = 5.
with time and reached 19 lmol kg21 min21 after 100 min of exercise. Fructose conversion into plasma lactate (calculated with lactate Ra of the test with 13C3-lactate infusion, see section entitled “Calculations”) was 35.4 6 2.9 lmol lactate kg21 min21. Other variables _ 2, VCO _ Mean VO 2, and respiratory exchange ratio (RER) at rest and during exercise are shown in Table 2. RER was significantly higher in GLUFRU during exercise (P , 0.05, Wilcoxon’s signed-rank test). Mean heart rate was comparable between conditions, at 134 6 7 beats/min in GLU compared with 133 6 7 beats/min in GLUFRU (P = NS) (Table 2). After 20 min of exercise, RPE was 9.1 6 1.1 and 9.7 6 1.1 in GLU and GLUFRU, respectively. RPE increased significantly
FIGURE 4. Mean (6SD) glucose (Gluc.) rates of appearance (Ra; A), rates of disappearance (Rd; B), and metabolic clearance rate (MCR; C) in glucose-alone (GLU) and glucose + fructose (GLUFRU) conditions. All analyses were performed by 2-factor repeated-measures ANOVA. A: P , 0.001 for treatment, P , 0.01 for time, and P = 0.86 for treatment · time interaction. B: P = 0.01 for treatment, P = 0.01 for time, and P = 0.71 for treatment · time interaction. C: P = 0.08 for treatment, P , 0.01 for time, and P = 0.91 for treatment · time interaction. For all variables, n = 7.
7 of 9
GLUCOSE AND FRUCTOSE CO-INGESTION DURING EXERCISE
drolysis and hepatic glucose release (33, 34). Increased glucose utilization may be induced by fructose ingestion through an increase in glucokinase activity induced by fructose-1-P in hepatic cells (35). Although plasma glucose concentration did not change significantly in our study, this may be due to the relatively small increase in total glucose fluxes, which were insufficient to produce a significant increase in glycemia, or because fructose simultaneously increased glucose utilization to some extent. Such an increase in glucose utilization may indeed be induced by fructose ingestion through an increase in glucokinase activity induced by fructose-1-P in hepatic cells (35). Fructose ingestion led to an increase in total lactate Ra (+ ’30% in GLUFRU compared with GLU). This increased lactate production, together with the low plasma fructose concentrations, suggests that part of the ingested fructose was extracted at first pass by the splanchnic tissues and released into the bloodstream as lactate (11, 36). Assuming that the difference in lactate Ra between GLUFRU and GLU corresponds to fructose conversion into lactate, our results indicate that this pathway accounted for approximately half of the fructose oxidation. The other half could be accounted for by GNGF. Total lactate oxidation was also increased in GLUFRU compared with GLU. This most likely corresponds to lactate oxidation in active muscle fibers (37). Part of lactate oxidation may also take place in heart muscle, because lactate is efficiently used as an energy fuel in cardiomyocytes during exercise (38). Approximately 100% of lactate Rd was oxidized both with GLU and GLUFRU. Our figure is somewhat higher than previous reports (14, 25, 39, 40) showing that ’80% of lactate Ra was oxidized during exercise. It is likely that the high oxidative capacity of our highly trained subjects, reflected by their high _ 2max values, accounted for this difference (15). It is also VO possible that our data slightly overestimated the relative lactate oxidation, because lactate kinetics were measured in nonarterialized venous blood, which is known to lead to a slight underestimate of lactate kinetics (41). Nonetheless, our data clearly indicate that oxidation was the major fate of the lactate produced during exercise, whether from exogenous or endogenous glucose or from exogenous fructose. Hepatic lactate uptake and conversion into glucose contributes substantially to total lactate disappearance under fasting conditions. Increased lactate availability, produced by exogenous lactate infusion, increases fractional (42, 43) and total gluconeogenesis from lactate (GNGL) during prolonged exercise in fasted conditions (43). In contrast, our results show that GNGL
FIGURE 5. Mean (6SD) rates of gluconeogenesis from lactate (GNGLactate; n = 5) (A) and from fructose (GNGFructose; n = 7) in glucosealone (GLU) and glucose + fructose (GLUFRU) conditions. A: P = 0.31 for treatment, P = 0.10 for time, and P = 0.61 for treatment · time interaction (2-factor repeated-measures ANOVA). B: P , 0.001 for time (one-factor repeated-measures ANOVA). #Significant difference between 20, 40, 60, and 80 min of exercise (P , 0.05, Tukey post hoc test).
with time and reached 11.8 6 1.5 and 11.3 6 1.2 at the end of exercise in GLU and GLUFRU, respectively (2-factor repeatedmeasures ANOVA: treatment, P = 0.75; time, P , 0.001; interaction, P = 0.89). On average, RPE was 10.7 6 1.4 in GLU and 10.7 6 1.2 in GLUFRU during exercise (Table 2). DISCUSSION
To our knowledge, this study provides the first detailed assessment of fructose metabolism during high-dose glucose + fructose ingestion during exercise. We showed that oxidation of fructose co-ingested with glucose is explained equally by lactate oxidation produced from fructose cleavage and by glucose oxidation derived from fructose and direct fructose oxidation in splanchnic tissues. We observed that fructose co-ingested with glucose led to a slight, but significant, increase in the rate of glucose appearance. In contrast, previous observations have shown that in resting conditions, fructose administration did not change total glucose production due to an auto-regulation of glucose production, in which increased GNG is compensated for by decreased hepatic glycogenolysis (31). Differences between studies performed in resting conditions and exercise may be explained by the fact that exercise stimulates glucagon secretion (32) and that hyperglucagonemia impairs auto-regulation of glucose production, most likely by stimulating glucose-6-phosphate hy-
TABLE 2 _ _ 2), carbon dioxide production (VCO Average power output (PO), oxygen consumption (VO 2), respiratory exchange ratio (RER), heart rate (HR), and perceived exertion (RPE) at rest and during exercise1 Rest GLU PO (W) _ 2 (L/min) VO _ VCO 2 (L/min) RER HR (beats/min) RPE
— 0.374 6 0.308 6 0.83 6 57 6 —
0.09 0.07 0.04 7
Exercise GLUFRU — 0.403 6 0.331 6 0.83 6 59 6 —
0.06 0.05 0.04 7
GLU 182 3.015 2.803 0.93 134 10.7
6 6 6 6 6 6
13 0.17 0.15 0.03 7 1.4
P GLUFRU
Fructose
Exercise
F·E
6 6 6 6 6 6
— 0.92 0.09 ,0.05 0.75 0.97
— ,0.001 ,0.001 ,0.001 ,0.001 ,0.001
— 0.68 0.78 0.28 0.89 0.67
182 3.006 2.845 0.95 133 10.7
13 0.22 0.15 0.03 7 1.2
1 Values are means 6 SDs. GLU, glucose alone; GLUFRU, glucose + fructose; F · E, Fructose · Exercise interaction. P values were derived by 2-factor repeated-measures ANOVA.
8 of 9
LECOULTRE ET AL
TABLE 3 Summary of mean carbohydrate (CHO) fluxes and metabolic fate of fructose during ingestion of glucose alone (GLU) and glucose + fructose (GLUFRU) between 100 and 120 min of exercise when steady state of carbohydrate flux is assumed1 GLUFRU
Systemic lactate Ra (lmol kg21 min21) Systemic lactate oxidation (lmol kg21 min21) Glucose Ra (lmol kg21 min21) Total CHO oxidation (g/min) Nonoxidative fructose disposal (g/min) 1
Total GLU
Total
From fructose
Percentage of fructose ingestion
94 97 82 3.1 —
121 127 90 3.3 —
35 31 18.8 0.48 0.52
28 25 29 60 40
Ra, rate of appearance.
was very low, most likely as a consequence of carbohydrate intake and suppression of gluconeogenesis by hyperinsulinemia (44), although our subjects were well trained and had high GNG capacity (27). A reduction of liver blood flow, which occurs during strenuous exercise, may also contribute to lower hepatic gluconeogenesis (45). In agreement with previous studies, a peak oxidation rate of ’0.45 g exogenous fructose/min occurred after 100 min of exercise (2, 6, 7, 46). Because fructose conversion into lactate and oxidation was estimated to be 35 and 31 lmol kg21 min21, corresponding to 28% and 25% of the fructose load, respectively, and because fructose-derived glucose released into the systemic circulation was estimated to be 29% of the fructose load (Table 1), one can conclude that both the lactate and glucose released by the liver were essentially oxidized, presumably in active skeletal muscles (14, 15, 40) and the heart (38). Our data also indicate that, by the end of the exercise, ’40% of the labeled carbons of fructose had not been recovered in expired air (Table 3). Similar observations have been reported previously (3, 46), and it has been suggested that the liver may act as a carbon reservoir after fructose ingestion, which rapidly releases lactate while temporarily retaining part of the glucose synthesized as glycogen (46). Our study has several limitations. First, measurements of lactate kinetics and of fructose disposal were performed only during exercise but not at rest; furthermore, all measurements were performed in subjects after having received the glucose or glucose-fructose mixture, but measurements were not performed in nonfed subjects. The present data therefore do not evaluate how and to what extent exogenous fructose and glucose interact with endogenous lactate and glucose metabolism during exercise. However, our data do provide estimates of the metabolic pathways used for fructose disposal during exercise. Second, there are several limitations to the tracer techniques used for assessing the pathways of fructose disposal: although plasma lactate concentrations and isotopic enrichments had reached steady state in both the 13C-fructose and 13C-lactate experiments, lactate calculation of lactate appearance and oxidation is likely to be overestimated to some extent due to rapid equilibration of intracellular lactate and pyruvate and dependence on glycolytic rates (47); calculation of gluconeogenesis was made by using plasma 13C-lactate or oral 13C-fructose enrichments, which do not reflect true precursor enrichment; other approaches, such as mass isotopomer distribution analysis, may have provided more robust estimates of true gluconeogenic rates (48). Furthermore, gluconeogenesis was calculated with the use of a correction factor
to account for tracer loss into tricarboxylic acid cycle metabolites and to CO2 at the level of oxaloacetate (28, 49). This correction factor was used on the basis of the assumption that the bulk of fructose was converted into pyruvate or lactate before joining the gluconeogenic pathway, but it may be inappropriate if a substantial synthesis of glucose occurred directly from glyceraldehade-3 phosphate or under non–steady state conditions. Finally, both glucose and lactate kinetics were calculated by using a one-compartment model, which provide inaccurate estimates in non–steady state conditions compared with multicompartment models (50). The approach we used therefore allowed tracing the metabolic fate of orally administered 13C-fructose in a global way, but absolute rates may bear significant inaccuracies. In conclusion, our study shows that, when fructose is ingested simultaneously with glucose during exercise, both hepatic lactate and glucose increase significantly compared with ingestion of isocaloric amounts of glucose alone. Furthermore, total carbohydrate oxidation increased, essentially due to an enhanced oxidation of lactate. Because lactate is essentially oxidized by working skeletal muscle during exercise, these results show that co-ingestion of fructose with glucose not only allows the increase total carbohydrate oxidation beyond what may be obtained with increasing glucose doses but also provide an additional oxidative fuel to active skeletal muscles during exercise. We are indebted to all our volunteers for their commitment and effort spent in the laboratory. We thank Valentine Rey, Eunika Rossi, and Nathalie Stefanoni for all biological sample analyses. We also thank Charmaine Tam and Trent Stellingwerf for helpful discussions. This manuscript is dedicated to the memory of our friend Boris Chervaz. The authors’ responsibilities were as follows—VL, PS, and LT: designed the study; VL, RB, PS, GC, and GPM: performed the experiments; VL: performed the statistical analysis and wrote the manuscript; and LT, RB, YS, PS, GC, and GPM: edited the manuscript. None of the authors had any conflicts of interest.
REFERENCES 1. Adopo E, Peronnet F, Massicotte D, Brisson GR, Hillaire-Marcel C. Respective oxidation of exogenous glucose and fructose given in the same drink during exercise. J Appl Physiol 1994;76:1014–9. 2. Jentjens RL, Achten J, Jeukendrup AE. High oxidation rates from combined carbohydrates ingested during exercise. Med Sci Sports Exerc 2004;36:1551–8. 3. Jentjens RL, Moseley L, Waring RH, Harding LK, Jeukendrup AE. Oxidation of combined ingestion of glucose and fructose during exercise. J Appl Physiol 2004;96:1277–84. 4. Jentjens RL, Shaw C, Birtles T, Waring RH, Harding LK, Jeukendrup AE. Oxidation of combined ingestion of glucose and sucrose during exercise. Metabolism 2005;54:610–8.
GLUCOSE AND FRUCTOSE CO-INGESTION DURING EXERCISE 5. Jentjens RL, Underwood K, Achten J, Currell K, Mann CH, Jeukendrup AE. Exogenous carbohydrate oxidation rates are elevated after combined ingestion of glucose and fructose during exercise in the heat. J Appl Physiol 2006;100:807–16. 6. Jeukendrup AE, Moseley L, Mainwaring GI, Samuels S, Perry S, Mann CH. Exogenous carbohydrate oxidation during ultraendurance exercise. J Appl Physiol 2006;100:1134–41. 7. Jentjens RL, Jeukendrup AE. High rates of exogenous carbohydrate oxidation from a mixture of glucose and fructose ingested during prolonged cycling exercise. Br J Nutr 2005;93:485–92. 8. Currell K, Jeukendrup AE. Superior endurance performance with ingestion of multiple transportable carbohydrates. Med Sci Sports Exerc 2008;40:275–81. 9. Mayes PA. Intermediary metabolism of fructose. Am J Clin Nutr 1993; 58:754S–65S. 10. Decombaz J, Sartori D, Arnaud MJ, Thelin AL, Schurch P, Howald H. Oxidation and metabolic effects of fructose or glucose ingested before exercise. Int J Sports Med 1985;6:282–6. 11. Chandramouli V, Kumaran K, Ekberg K, Wahren J, Landau BR. Quantitation of the pathways followed in the conversion of fructose to glucose in liver. Metabolism 1993;42:1420–3. 12. Tappy L, Le KA. Metabolic effects of fructose and the worldwide increase in obesity. Physiol Rev 2010;90:23–46. 13. Azevedo JL, Tietz E, Two-Feathers T, Paull J, Chapman K. Lactate, fructose and glucose oxidation profiles in sports drinks and the effect on exercise performance. PLoS ONE 2007;2:e927. 14. Bergman BC, Wolfel EE, Butterfield GE, et al. Active muscle and whole body lactate kinetics after endurance training in men. J Appl Physiol 1999; 87:1684–96. 15. Van Hall G, Jensen-Urstad M, Rosdahl H, Holmberg HC, Saltin B, Calbet JA. Leg and arm lactate and substrate kinetics during exercise. Am J Physiol Endocrinol Metab 2003;284:E193–205. 16. Fahey TD, Larsen JD, Brooks GA, Colvin W, Henderson S, Lary D. The effects of ingesting polylactate or glucose polymer drinks during prolonged exercise. Int J Sport Nutr 1991;1:249–56. 17. Peronnet F, Burelle Y, Massicotte D, Lavoie C, Hillaire-Marcel C. Respective oxidation of 13C-labeled lactate and glucose ingested simultaneously during exercise. J Appl Physiol 1997;82:440–6. 18. Durnin JV, Womersley J. Body fat assessed from total body density and its estimation from skinfold thickness: measurements on 481 men and women aged from 16 to 72 years. Br J Nutr 1974;32:77–97. 19. Novel-Chate V, Rey V, Chiolero R, et al. Role of Na+-K+-ATPase in insulin-induced lactate release by skeletal muscle. Am J Physiol Endocrinol Metab 2001;280:E296–300. 20. Tounian P, Schneiter P, Henry S, Delarue J, Tappy L. Effects of dexamethasone on hepatic glucose production and fructose metabolism in healthy humans. Am J Physiol 1997;273:E315–20. 21. Schneiter P, Gillet M, Chiolero R, Jequier E, Tappy L. Hepatic nonoxidative disposal of an oral glucose meal in patients with liver cirrhosis. Metabolism 1999;48:1260–6. 22. Chiolero RL, Revelly JP, Leverve X, et al. Effects of cardiogenic shock on lactate and glucose metabolism after heart surgery. Crit Care Med 2000; 28:3784–91. 23. Petersen KF, Laurent D, Yu C, Cline GW, Shulman GI. Stimulating effects of low-dose fructose on insulin-stimulated hepatic glycogen synthesis in humans. Diabetes 2001;50:1263–8. 24. Wolfe RR. Radioactve and stable isotope tracers in biomedicine. New York, NY: Wiley-Liss, 2005. 25. MacRae HH, Noakes TD, Dennis SC. Effects of endurance training on lactate removal by oxidation and gluconeogenesis during exercise. Pflugers Arch 1995;430:964–70. 26. van Hall G, Stromstad M, Rasmussen P, et al. Blood lactate is an important energy source for the human brain. J Cereb Blood Flow Metab 2009;29: 1121–9. 27. Bergman BC, Horning MA, Casazza GA, Wolfel EE, Butterfield GE, Brooks GA. Endurance training increases gluconeogenesis during rest and exercise in men. Am J Physiol Endocrinol Metab 2000;278:E244–51. 28. Hetenyi G Jr. Correction for the metabolic exchange of 14C for 12C atoms in the pathway of gluconeogenesis in vivo. Fed Proc 1982;41:104–9.
9 of 9
29. Frayn KN. Calculation of substrate oxidation rates in vivo from gaseous exchange. J Appl Physiol 1983;55:628–34. 30. R Development Core Team. R: a language and environment for statistical computing. Vienna, Austria: Foundation for Statistical Computing, 2008. 31. Tounian P, Schneiter P, Henry S, Jequier E, Tappy L. Effects of infused fructose on endogenous glucose production, gluconeogenesis, and glycogen metabolism. Am J Physiol 1994;267:E710–7. 32. Wahren J, Felig P, Ahlborg G, Jorfeldt L. Glucose metabolism during leg exercise in man. J Clin Invest 1971;50:2715–25. 33. Paquot N, Schneiter P, Jequier E, et al. Effects of ingested fructose and infused glucagon on endogenous glucose production in obese NIDDM patients, obese non-diabetic subjects, and healthy subjects. Diabetologia 1996;39:580–6. 34. Surmely JF, Schneiter P, Henry S, Paquot N, Jequier E, Tappy L. Effects of glucagon in the control of endogenous glucose production in man. Nutrition 1999;15:267–73. 35. Donmoyer CM, Ejiofor J, Lacy DB, Chen SS, McGuinness OP. Fructose augments infection-impaired net hepatic glucose uptake during TPN administration. Am J Physiol Endocrinol Metab 2001;280:E703–11. 36. Coss-Bu JA, Sunehag AL, Haymond MW. Contribution of galactose and fructose to glucose homeostasis. Metabolism 2009;58:1050–8. 37. Brooks GA, Dubouchaud H, Brown M, Sicurello JP, Butz CE. Role of mitochondrial lactate dehydrogenase and lactate oxidation in the intracellular lactate shuttle. Proc Natl Acad Sci USA 1999;96:1129–34. 38. Gertz EW, Wisneski JA, Stanley WC, Neese RA. Myocardial substrate utilization during exercise in humans: dual carbon-labeled carbohydrate isotope experiments. J Clin Invest 1988;82:2017–25. 39. Donovan CM, Brooks GA. Endurance training affects lactate clearance, not lactate production. Am J Physiol 1983;244:E83–92. 40. Mazzeo RS, Brooks GA, Schoeller DA, Budinger TF. Disposal of blood [1-13C]lactate in humans during rest and exercise. J Appl Physiol 1986; 60:232–41. 41. Lehman SL, Brooks GA. Obtaining a representative blood sample in lactate tracer studies. Horm Metab Res 1990;22:470–7. 42. Miller BF, Fattor JA, Jacobs KA, et al. Lactate and glucose interactions during rest and exercise in men: effect of exogenous lactate infusion. J Physiol 2002;544:963–75. 43. Roef MJ, de Meer K, Kalhan SC, Straver H, Berger R, Reijngoud DJ. Gluconeogenesis in humans with induced hyperlactatemia during low-intensity exercise. Am J Physiol Endocrinol Metab 2003;284: E1162–71. 44. Camacho RC, Pencek RR, Lacy DB, James FD, Wasserman DH. Suppression of endogenous glucose production by mild hyperinsulinemia during exercise is determined predominantly by portal venous insulin. Diabetes 2004;53:285–93. 45. Fojt E, Ekelund LG, Hultman E. Enzyme activities in hepatic venous blood under strenuous physical exercise. Pflugers Arch 1976;361: 287–96. 46. Rowlands DS, Thorburn MS, Thorp RM, Broadbent S, Shi X. Effect of graded fructose coingestion with maltodextrin on exogenous 14Cfructose and 13C-glucose oxidation efficiency and high-intensity cycling performance. J Appl Physiol 2008;104:1709–19. 47. Avogaro A, Toffolo G, Miola M, et al. Intracellular lactate- and pyruvate-interconversion rates are increased in muscle tissue of noninsulin-dependent diabetic individuals. J Clin Invest 1996;98:108–15. 48. Trimmer JK, Schwarz JM, Casazza GA, Horning MA, Rodriguez N, Brooks GA. Measurement of gluconeogenesis in exercising men by mass isotopomer distribution analysis. J Appl Physiol 2002;93: 233–41. 49. McMahon M, Marsh HM, Rizza RA. Effects of basal insulin supplementation on disposition of mixed meal in obese patients with NIDDM. Diabetes 1989;38:291–303. 50. Hovorka R, Jayatillake H, Rogatsky E, Tomuta V, Hovorka T, Stein DT. Calculating glucose fluxes during meal tolerance test: a new computational approach. Am J Physiol Endocrinol Metab 2007;293:E610–9.
