INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

IT IS CLEAR THAT DURING HIGH-INTENSITY aerobic exercise (>80% maximal oxygen consumption; O_{2 max}) plasma free fatty acid (FFA) mobilization is impaired and plasma FFA concentration remains low (21, 27). In turn, the low plasma FFA concentration is responsible for reducing fat oxidation and increasing muscle glycogen oxidation (8, 28, 35). In this situation, increasing plasma FFA concentration by intravenous infusion of long-chain triglycerides (LCT) and heparin increased fat oxidation and reduced muscle glycogen utilization by 15-40% (8, 28, 35). Obviously, triglyceride and heparin infusion is not a practical method to increase plasma FFA concentrations. Triglyceride ingestion is a much more practical means for providing exogenous fat. Unfortunately, typical dietary fat consists primarily of LCT, which are limited in their contribution to oxidation by slow rates of gastric emptying and transport into the systemic circulation (i.e., lymphatic transport), and hydrolysis of LCT is normally slow relative to the metabolic requirements of exercise. For example, Satabin et al. (30) reported that only 9% of a 44-g ¹³C-labeled LCT preexercise meal were oxidized during 2 h of exercise at 60% O_{2 max}. Presently, there is no practical way to increase long-chain fatty acid oxidation during exercise by ingestion of LCT alone.

Unlike LCT, medium-chain triglycerides (MCT), which contain fatty acids with 6-10 carbon atoms, are rapidly hydrolyzed and absorbed into blood within the intestinal lumen (1). Moreover, medium-chain fatty acids are not reesterified and are more easily transported into the mitochondria for subsequent oxidation compared with fatty acids from LCT (1, 29). Furthermore, MCT do not suppress gastric emptying (3). In fact, adding MCT to a carbohydrate meal increased gastric emptying compared with an equicaloric carbohydrate meal (3). Therefore, ingested MCT is potentially a readily available source of energy, and the addition of MCT to a carbohydrate meal may provide sufficient exogenous fuel to reduce the reliance on muscle glycogen during high-intensity exercise.

The principal purpose of the present investigation was to determine whether a tolerable amount of ingested MCT can, in fact, reduce muscle glycogenolysis and glycogen oxidation during intense exercise, a condition that appears to elicit inadequate endogenous fatty acid availability. An additional objective of this study was to determine whether the coingestion of MCT and carbohydrate increases plasma glucose availability and uptake before and during exercise. We added ~25 g of MCT (tricaprin; C10:0) to a preexercise carbohydrate meal (~50 g) ingested 1 h before 30 min of intense exercise (84% O_{2 max}). We fed our subjects carbohydrate together with MCT because 1) the coingestion of MCT and carbohydrate increases MCT oxidation compared with MCT ingestion alone (19), 2) the addition of carbohydrate improves palatability (16) and tends to increase gastrointestinal tolerance to MCT (20), and 3) preexercise carbohydrate ingestion will further impair endogenous fat oxidation (15). Therefore, the coingestion of carbohydrate and MCT increases the potential for MCT to reduce muscle glycogen utilization during exercise. Muscle glycogen use was assessed by using two independent methods. The change in glycogen concentration during exercise in the vastus lateralis was determined from biopsies. Additionally, glycogen oxidation during exercise was calculated as total carbohydrate oxidation minus the rate of glucose disappearance from plasma (R_d) (determined from dilution of a constant-rate intravenous infusion of [6,6-d₂]glucose).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects and general experimental design . Seven well-trained cyclists participated in this experiment. Their O_{2 max} was 4.6 ± 0.3 l/min (61.5 ± 2.5 ml · kg¹ · min¹), and their mean body weight was 75.1 ± 4.0 kg. Subjects were informed of the possible risks involved, and each signed a consent form, approved by the Internal Review Board of The University of Texas at Austin. On two occasions they cycled on an ergometer for 30 min at 84 ± 1% O_{2 max} 1 h after ingesting either carbohydrate alone or in combination with MCT. Two days before each trial the subjects performed the experimental exercise protocol (30 min at 84% O_{2 max}) to familiarize themselves with the procedure and to ensure homogeneity of the last exercise bout. During the day before both trials the subjects did not exercise and consumed a standardized diet.

Experimental procedures . On two occasions, separated by 5-7 days, subjects arrived at the laboratory in the morning after an overnight fast (12 h). On arrival, Teflon catheters were inserted into a forearm vein of both arms, one for isotope infusion, the other for blood sampling, and a heating pad was affixed to the hand and forearm of the sampling arm. Thereafter, they received a primed, constant-rate infusion of [6,6-d₂]glucose (0.39 µmol · kg¹ · min¹; prime = 33 µmol/kg) by using a calibrated syringe pump (Harvard Apparatus, South Natick, MA) while resting for at least 1 h. Exactly 1 h before exercise, they ingested one of two test meals: 1) carbohydrate (CHO; 0.72 g sucrose/kg body wt; ~50 g) or 2) MCT+CHO (0.36 g tricaprin/kg body wt; ~25 g; 0.72 g sucrose/kg). CHO was provided as a viscous paste, whereas MCT+CHO was combined into a chewable solid that dissolved at body temperature. Water was ingested with both meals (3.6 ml/kg body wt; ~250 ml). Exactly 1 h after ingestion of the test meal, the subjects cycled for 2 min at ~60% O_{2 max} and then 28 min at 84% O_{2 max} (total of 30 min). The order of the trials was counterbalanced. Muscle biopsies (~40-80 mg) were taken from the vastus lateralis (4) 30 min before and immediately after exercise. The samples were immediately frozen in liquid nitrogen and stored at 80°C for later determination of glycogen concentration. Blood samples were drawn every 10 min at rest after ingestion and at 5-min intervals throughout exercise. O₂ uptake (O₂) and CO₂ production (CO₂) were measured from 0-15 and 18-30 min via open-circuit spirometry.

Analytic procedures. Eight milliliters of blood were withdrawn for each sample. Plasma was separated by centrifugation, stored at 80°C, and later analyzed for concentration of glucose (YSI 23a glucose autoanalyzer; Yellow Springs Instruments, Yellow Springs, OH), glycerol (9), FFA (23), lactate (13), -hydroxybutyrate (36), and insulin (radioimmunoassay; ICN Biomedicals, Costa Mesa, CA). In addition, plasma isotopic enrichment of the aldonitrile acetate derivative of [6,6 d₂]glucose (33) was determined via gas chromotography-mass spectrophotometry (GCMS) for calculations of the rate of appearance (R_a) and R_d of glucose (see below).

Isotope enrichment sample preparation. Plasma samples (1 ml) were deproteinized by adding 1 ml 0.3 N Ba(OH)₂ and 1 ml 0.3 N Zn(SO)₄. Each tube was then vortexed and incubated in an ice bath for 20 min. After centrifugation (3,000 rpm for 15 min at 4°C), the supernatant was placed into a clean tube and the water was removed via vacuum centrifugation (Savant Instruments, Farmingdale, NY). The aldonitrile acetate derivative of glucose was prepared by adding 100 µl hydroxolamine-hydrochloride solution (20 mg/ml in pyridine) to the dried sample. After a 30-min incubation at 100°C, 75 µl of acetic anhydride (Supelco, Bellefonte, PA) were then added, and the samples remained incubating for 1 additional hour. Finally, the samples were evaporated under N₂. Before injection into the GCMS, the samples were reconstituted with ethyl acetate.

Muscle glycogen concentration . The frozen muscle sample was weighed, mechanically homogenized in a glycerol-Na₂PO₄ buffer, hydrolyzed in 2 N HCl (2 h), and neutralized with NaOH. Glucose concentration of the hydrolysate was determined enzymatically (24).

Measurement of gas exchange . Inspired air volume was measured continuously with a Parkinson-Cowan CD4 dry-gas meter (Rayfield Equipment, Waitsfield, VT). The expired gases were constantly sampled from a mixing chamber and analyzed for oxygen (model SA3, Applied Electrochemistry, Ametek, Pittsburgh, PA) and carbon dioxide (model LB-2, Beckman, Schiller Park, IL). These instruments were interfaced with a computer for calculations of the rate of O₂, rate of CO₂, and respiratory exchange ratio (RER).

Calculations . R_a glucose and R_d glucose were calculated by using the non-steady-state equation of Steele (32), modified for use with stable isotopes

where F is the isotope infusion rate, Vd is the glycerol volume of distribution (estimated to be 100 ml/kg), C₁ and C₂ are the plasma glycerol concentrations at times 1 and 2 (t₁ and t₂), respectively, and E₁ and E₂ are isotopic enrichment at t₁ and t₂, respectively. Carbohydrate and fat (i.e., triglyceride) oxidations were calculated from O₂ and CO₂ (10). It has been reported that the calculation of carbohydrate and fat oxidation from O₂ and CO₂ in trained cyclists exercising at 85% O_{2 max} (identical conditions as in the present study) was the same as that measured by using an alternative method (¹³C/¹²C ratio in breath) that does not rely on CO₂ for calculating substrate oxidation (26).

Statistical analysis . Data were analyzed by using a two-way ANOVA (treatment by time) for repeated measures with Tukey's post hoc analysis. Planned comparisons for mean values were made by using paired Student's t-tests, with P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The average power output was identical during both trials (286 ± 22 W; 84 ± 1% O_{2 max}). As shown in Table 1, the average (5- to 30-min) energy expenditure, O₂, RER, heart rate, and rating of perceived exertion were also similar during the two trials. Furthermore, carbohydrate oxidation was equally high throughout exercise during both CHO and MCT+CHO (360 ± 15 and 364 ± 17 µmol · kg¹ · min¹, respectively).

View this table: [in this window] [in a new window]   Table 1.   Average responses during 30 min of exercise at 84% O2 max

Plasma glucose kinetics and concentration . The mean rate of R_a glucose at rest before exercise was greater (P < 0.05) during MCT+CHO compared with CHO (25.4 ± 1.0 vs. 20.8 ± 0.8 µmol · kg¹ · min¹) (Fig. 1A). Similarly, the mean rate of R_d glucose from plasma was also 30% greater (P < 0.05) before exercise during MCT+CHO compared with CHO (26.9 ± 1.5 vs. 20.7 ± 0.7 µmol · kg¹ · min¹) (Fig. 1B). Plasma glucose concentration increased transiently during the 30-min period after ingestion of both meals, and then decreased during the 30- to 60-min period after ingestion as glucose uptake (R_d glucose) exceeded R_a glucose. Immediately before exercise, plasma glucose concentration was lower (P < 0.05) during MCT+CHO compared with CHO (4.4 ± 0.2 and 5.1 ± 0.1 mM, respectively).

View larger version (23K): [in this window] [in a new window]   Fig. 1.   A: rate of glucose appearance in plasma (Ra glucose). B: rate of glucose disappearance from plasma (Rd glucose). C: plasma glucose concentration during carbohydrate (CHO; ) and medium-chain trigylerides (MCT)+CHO () ingestion. * MCT+CHO significantly different from CHO, P < 0.05.

1

1

Plasma insulin concentration . In accordance with R_a glucose, the plasma insulin response, calculated as the integrated area under the insulin vs. time curve at rest, was also significantly greater (P < 0.05) during the hour after ingesting MCT+CHO compared with CHO (1,199 ± 276 and 895 ± 233 µU · min¹ · ml¹, respectively). During exercise, plasma insulin concentration declined to basal levels, and no differences were observed between MCT+CHO and CHO (6.2 ± 1.3 and 6.7 ± 1.1 µU/ml at 30 min of exercise, respectively).

Plasma substrate concentrations. Plasma FFA and glycerol concentrations decreased during the hour after the ingestion of both meals; however, only the reduction in plasma FFA was statistically significant (P < 0.05), and there were no differences in either plasma FFA or glycerol concentrations between trials (Fig. 2). During exercise, plasma glycerol increased (P < 0.05) above preexercise values, whereas plasma FFA concentration remained low (<150 M). Thirty minutes after ingestion, plasma -hydroxybutyrate concentration was greater (P < 0.05) during MCT+CHO compared with CHO (40 ± 8 vs. 18 ± 3 mM) (Fig. 3), and it remained more than twofold greater throughout exercise (P < 0.05). Because of the high-intensity exercise, plasma lactate concentration increased (P < 0.05) from ~2 mM at rest to 11.8 ± 1.3 and 11.8 ± 1.9 mM at the end of exercise during MCT+CHO and CHO, respectively.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   A: plasma free fatty acid (FFA) concentration. B: plasma glycerol concentration during CHO () and MCT+CHO ().  Significantly different from basal value (60 min), P < 0.05.  Significantly different from 0-min value, P < 0.05.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Plasma -hydroxybutyrate concentration during CHO () and MCT+CHO (). * MCT+CHO significantly different from CHO, P < 0.05.  Significantly greater than basal value (60 min), P < 0.05.

Muscle glycogen . The concentration of muscle glycogen within the vastus lateralis was similar before both trials (Table 2), and the reduction in glycogen concentration during exercise was not different for CHO and MCT+CHO [38.8 ± 4.0 and 42.0 ± 4.6 mmol/kg wet weight (ww), respectively]. In addition, because both R_d glucose and carbohydrate oxidation were similar during exercise in both trials, the calculated minimum rate of muscle glycogen oxidation was also similar during CHO and MCT+CHO (329 ± 15 and 331 ± 18 µmol · kg¹ · min¹, respectively). Therefore, two methods concur that muscle glycogen utilization was not different during MCT+CHO compared with CHO.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In previous studies, MCT have not been ingested under conditions where endogenous fat mobilization is very low, and exogenous fat supplementation is known to increase fat oxidation and reduce muscle glycogenolysis. The present study was designed to provide such optimal conditions to determine whether MCT ingestion can potentially reduce muscle glycogen oxidation. First, the combination of preexercise carbohydrate ingestion and high-intensity exercise reduced plasma FFA concentration to very low levels (<150 M). Second, during high-intensity exercise most energy is derived from muscle glycogen (27). Thus any energy derived from the exogenous MCT would likely be reflected as a reduction in muscle glycogen oxidation. Finally, a high glycolytic flux, such as that observed during high-intensity exercise or during exercise after a preexercise carbohydrate meal, has been shown to decrease LCT oxidation without impairing MCT oxidation (6, 31). The principal finding of this study, however, was that the addition of ~25 g of MCT to a carbohydrate meal did not reduce either net muscle glycogen utilization or calculated glycogen oxidation, even under these theoretically idealized conditions of high-intensity exercise after a carbohydrate feeding.

The addition of MCT to the carbohydrate meal did, however, significantly alter blood glucose kinetics at rest and during exercise. We found that adding MCT to a carbohydrate meal increased glucose availability in plasma throughout the study. It is known that coingesting MCT and carbohydrate increases gastric emptying compared with an equicaloric carbohydrate meal (3); however, the effect of adding MCT to a given amount of carbohydrate has not been studied. Our findings suggest that MCT may increase gastric emptying and intestinal absorption of the coingested carbohydrate. The resultant elevated plasma insulin response during MCT+CHO caused a greater R_d glucose at rest during MCT+CHO compared with CHO. During exercise, however, despite an elevated R_a glucose during MCT+CHO, R_d glucose and presumably plasma glucose oxidation were not different during the two trials. Similarly, it has been found that the addition of MCT to carbohydrate feedings during exercise did not alter exogenous carbohydrate oxidation (18). It is likely that the equally high rates of muscle glycogenolysis during MCT+CHO and CHO may have prevented further increase in glucose uptake during MCT+CHO, despite an elevated R_a glucose and a greater preexercise plasma insulin response. Therefore, neither the energy contribution of ingested MCT nor its effect on glucose kinetics altered substrate oxidation during high-intensity exercise.

Unlike fat supplementation via infusion of LCT and heparin, MCT ingestion does not increase plasma FFA concentration (16, 22, 30). However, the absence of a detectable increase in the concentration of plasma FFA does not indicate that the energy from the ingested MCT was not delivered to the systemic circulation. Massicotte et al. (22) reported that nearly 60% of a ¹³C-labeled MCT meal was oxidized during exercise without an elevation in plasma FFA concentration above control. This may be explained by a very rapid oxidation of medium-chain fatty acids, preventing their accumulation in plasma (1) and/or by the fact that fatty acids derived from MCT are rapidly absorbed within the portal circulation and metabolized to a large extent within the liver, increasing the production of ketone bodies (i.e., acetoacetate and -hydroxybutyrate) (11, 37). In the present study, MCT+CHO ingestion increased plasma -hydroxybutyrate concentration more than twofold, suggesting that plasma ketones served as an exogenous energy source in the systemic circulation. However, the extent to which an acute elevation in plasma ketone concentration contributes to energy production during exercise is not clear (2, 14). The elevation of plasma ketone concentration in the present study clearly did not reduce muscle glycogen oxidation during high-intensity exercise.

The oxidation of ketone bodies or fatty acids from MCT could potentially reduce the oxidation of a substrate other than muscle glycogen (i.e., endogenous fat or blood glucose). We have previously reported that indirect calorimetry provides a valid measurement of substrate oxidation during high-intensity exercise (80-85% O_{2 max}) (26), and presently we found that fat oxidation was very low during both CHO and MCT+CHO (2-2.5 µmol · kg¹ · min¹). Therefore, even if all of the fat oxidized during MCT+CHO was exogenous MCT, thus entirely sparing endogenous fat, the absolute reduction in endogenous fat oxidation would still be very small (<35 kcal) and would account for only a small portion of the ingested MCT (~15%). In addition, MCT ingestion did not alter R_d glucose during exercise, suggesting that blood glucose oxidation was not affected. Because muscle glycogen oxidation accounted for ~90% of the total energy production during the CHO trial, if an appreciable amount of energy was derived from the exogenous MCT during MCT+CHO, it should have been reflected as a reduction in muscle glycogen oxidation, but it did not decline.

The above observations suggest that not enough MCT was oxidized during exercise to result in a measurable sparing of muscle glycogen. Although, if all of the 25 g MCT (~200 kcal) had been oxidized in place of muscle glycogen, muscle glycogen oxidation and the change in muscle glycogen concentration could have been substantially reduced (~125 µmol · kg¹ · min¹ and 30 mmol/kg ww, respectively). However, the greatest rate of MCT oxidation previously reported after ingestion of a single dose of MCT was only 12 g/h (30); others have found the rate to be even lower (7-9 g/h) (19, 22). By using the highest reported MCT oxidation rate, this translates to only ~6 g MCT oxidized (~50 kcal) during the 30-min exercise bout in the present study and could account for a reduction in glycogen oxidation of only ~30 µmol · kg¹ · min¹ and a decrease in the change in muscle glycogen concentration of only ~7 mmol/kg ww. It is possible that ingesting a greater quantity of MCT may increase the absolute amount of energy derived from MCT. Unfortunately, as we presently observed, as well as had others (7, 16, 22) gastrointestinal distress (i.e., nausea and diarrhea) limited the acute dose of MCT that individuals could tolerate to ~25 g. Therefore, under the present conditions, the energy derived from this tolerable, acute dose of MCT was not enough to reduce muscle glycogen oxidation.

Ingestion of small aliquots of MCT throughout exercise can allow for a greater total amount of MCT to be tolerated. Two recent studies have reported that subjects were able to tolerate ~85 g of MCT by ingesting it over >2-h period of exercise (20, 34). Van Zyl et al. (34) found that, when compared with carbohydrate ingestion alone (65 g CHO/h), the coingestion of ~86 g MCT (~30 g MCT/h) and carbohydrate (65 g CHO/h) during 2 h of cycling at 60% O_{2 max} reduced the calculated rate of muscle glycogen oxidation (20-30 µmol · kg¹ · min¹) and slightly improved cycling performance (~3%) during a subsequent simulated 40-km time trial. In contrast, Jeukendrup et al. (20) found that the addition of 85 g of MCT to a 10% carbohydrate solution ingested while subjects cycled for 2 h at 60% O_{2 max} did not alter exogenous or endogenous carbohydrate utilization and did not improve cycling performance during a subsequent 15-min cycling time trial. These data suggest that ingesting a relatively large amount of MCT over a prolonged period (2 h) may slightly improve performance during a subsequent exercise bout lasting ~1 h but not during shorter duration exercise bouts (~15 min).

It is possible that several weeks of MCT ingestion and the subsequent chronic exposure to elevated plasma medium-chain fatty acid and ketone concentrations may increase the ability to oxidize medium-chain fatty acids and ketones and thus reduce the reliance on muscle glycogen oxidation during exercise. Mice fed a chronic diet containing MCT for 6 wk increased plasma -hydroxybutyrate concentration ~100% (12). This resulted in an increase of >20% in the activity of 3-ketoacyl-CoA transferase, an enzyme responsible for allowing muscle to oxidize plasma ketones (12). These adaptations were associated with a 20% greater muscle glycogen concentration after 30 min of forced swimming compared with that in control mice and a 10% longer swim time to exhaustion (12). Similarly, a 4-wk ketogenic diet in humans, resulting in a chronic resting plasma ketone concentration above 1 mM, has also been reported to reduce the reliance on muscle glycogen oxidation during exercise (25). Because the subjects in the present study were not chronically fed MCT or chronically exposed to high plasma ketone concentrations, their ability to oxidize them may not have been sufficiently enhanced, reducing the potential for MCT ingestion to spare muscle glycogen.

In summary, under conditions in which fat supplementation (i.e., intravenous infusion of LCT and heparin) is known to reduce muscle glycogen oxidation, a tolerable dose of ingested MCT (~25 g) did not reduce muscle glycogen oxidation or attenuate the decline in muscle glycogen concentration. MCT ingestion did, however, increase plasma glucose uptake at rest.