(surtout quand on prend des BCAA
Carbohydrate, Protein, and Fat Metabolism During Exercise After Oral Carnitine
Supplementation in Humans
Elizabeth M. Broad, Ronald J. Maughan,and Stuart D.R. Galloway
Twenty nonvegetarian active males were pair-matched and randomly assigned to
receive 2 g of L-carnitine L-tartrate (LC) or placebo per day for 2 wk. Participants
exercised for 90 min at 70% VO2max after 2 days of a prescribed diet (M ± SD: 13.6 ±
1.6 MJ, 57% carbohydrate, 15% protein, 26% fat, 2% alcohol) before and after supplementation.
Results indicated no change in carbohydrate oxidation, nitrogen excretion,
branched-chain amino acid oxidation, or plasma urea during exercise between
the beginning and end of supplementation in either group. After 2 wk of LC supplementation
the plasma ammonia response to exercise tended to be suppressed (0 vs.
2 wk at 60 min exercise, 97 ± 26 vs. 80 ± 9, and 90 min exercise, 116 ± 47 vs. 87 ±
25 μmol/L), with no change in the placebo group. The data indicate that 2 wk of LC
supplementation does not affect fat, carbohydrate, and protein contribution to metabolism
during prolonged moderate-intensity cycling exercise. The tendency toward
suppressed ammonia accumulation, however, indicates that oral LC supplementation
might have the potential to reduce the metabolic stress of exercise or alter ammonia
production or removal, which warrants further investigation.
Keywords: L-carnitine L-tartrate, cyclists, fat oxidation, carbohydrate oxidation,
protein utilization
The two primary, interrelated roles of L-carnitine (LC) in metabolism are to
transport long- and medium-chain fatty acids into mitochondria for -oxidation
(Fritz, 1963) and to buffer excess short-chain acyl groups, such as acetyl-CoA,
thereby maintaining optimum energy flux within mitochondria (Constantin-
Teodosiu, Cederblad, & Hultman, 1992). We have previously observed enhanced
carbohydrate
(CHO) oxidation during 60 min of cycling exercise in endurancetrained
males after supplementation with LC for 2 weeks (Abramowicz & Galloway,
2005), whereas the promoted benefits of carnitine supplementation include
increased fat oxidation. We have hypothesized that this shift in substrate utilization
after supplementation in trained athletes might be a result of the short-chain
Broad and Galloway are with the Dept. of Sports Studies, University of Stirling, FK9 4LA, Scotland
UK. Maughan is with the School of Sport and Exercise Sciences, Loughborough University,
Leicestershire, LE11 3TU, UK.
International Journal of Sport Nutrition and Exercise Metabolism, 2008, 18, 567-584
© 2008 Human Kinetics, Inc.
568 Broad, Maughan, and Galloway
acyl-group-buffering role of carnitine, but this action could also affect amino acid
oxidation in skeletal muscle. Furthermore, it might act in the same way in other
metabolically active tissues such as the liver, brain, or heart, thus influencing
whole-body substrate utilization.
To date, studies investigating the effect of LC supplementation on fuel utilization
during exercise have used gas-analysis techniques, with calculations for
CHO and fat oxidation based on a nonprotein respiratory quotient. Protein is generally
believed to contribute 5–10% of the total energy demand in prolonged exercise
(Graham & MacLean, 1992), with branched-chain amino acid (BCAA) oxidation
making the major contribution. Carnitine has been shown to facilitate the
metabolism of BCAAs in skeletal muscle by stimulating the conversion of
branched-chain keto acids (BCKAs) to carnitine esters (De Palo et al., 1993;
Veerkamp, Van Moerkerk, & Wagenmakers, 1985). In doing so, the inhibition of
BCKA dehydrogenase, one of the primary regulators of muscle amino acid metabolism,
is removed and free coenzyme-A is released for use in the many energyproducing
mitochondrial reactions. It has therefore been suggested that supplementation
with carnitine might further enhance the breakdown of BCAAs during
exercise by buffering the usual accumulation of BCKAs (Hoppel, 2003). Conversely,
if LC supplementation were to enhance fatty-acid uptake and metabolism
during exercise, it might reduce amino acid catabolism. If amino acid oxidation
during exercise were to change after LC supplementation, this would bring into
question the validity of using the nonprotein respiratory quotient to estimate fat
and CHO oxidation during exercise. To date, the effect of LC supplementation on
amino acids’ contribution to metabolism during exercise in humans has not been
investigated.
The aim of this study was to determine whether supplementation with
L-carnitine L-tartrate alters the fuel contribution to metabolism in endurancetrained
male athletes and specifically to examine any changes in protein contribution
to metabolism.
Methods
Twenty nonvegetarian male athletes actively involved in endurance training were
recruited. The participants’ characteristics are shown in Table 1. All participants
were fully informed about the study and underwent preparticipation screening
(medical history and physical activity questionnaires) before written informed
consent was obtained. The study was undertaken during the early preparation
phase of the cycling and triathlon competitive season to ensure that consistent
endurance-based training was being undertaken. No participant was suffering
from any metabolic disorder, and none was taking any medication or nutritional
supplements other than multivitamins/minerals or commercial sports drinks
during training. All experimental procedures were approved by the university
ethics of research committee, and all participants were free to withdraw from the
study at any time without obligation.
The study was undertaken using a double-blind, placebo-controlled, pairmatched
parallel design. Pair matching was undertaken primarily on the basis of
submaximal exercise workload and age. Participants came to the laboratory on
Carnitine Supplementation and Exercise Metabolism 569
four occasions over 4–5 weeks. The first visit was used to determine VO2max,
power output at 70% VO2max, and body composition using skinfolds (biceps, triceps,
subscapular, supraspinal, abdomen, midthigh, and calf; Norton & Olds,
2000). The maximal test was undertaken on an electrically braked cycle ergometer
(Lode Excalibur Sport V2.1, Lode BV, The Netherlands) in a laboratory where the
temperature was maintained at 20–21 °C.
All subsequent exercise trials, involving 90 min of steady-state exercise at
70% of VO2max, were undertaken on the same day of the week and at same time of
day. On the second visit participants undertook a familiarization trial to ensure
that the correct power output had been selected and to familiarize them with all
testing procedures. The final two visits were conducted before and after 2 weeks
of LC supplementation.
Supplementation
Supplementation consisted of two capsules taken twice daily with breakfast and
evening meals (i.e., four capsules/day total) for 14 days. The supplement capsule
consisted of 746 mg L-carnitine L-tartrate (L-Carnipure, Lonza Ltd., Basel, Switzerland),
thereby providing 2 g of LC per day. The placebo capsule (P) consisted
of a methylcellulose filler of the same weight as the carnitine. Participants’ compliance
to the supplementation was assessed by checking for any remaining capsules
at the end of 2 weeks and verbal questioning.
Dietary and Exercise Controls
Each participant was prescribed their dietary intake for 48 hr pre- and 24 hr posttrial,
based on attaining a minimum of 6 g carbohydrate · kg body mass (BM)–1
· day−1 and 1.5 g protein · kg BM–1 · day−1 and achieving estimated energy
requirements (Burke, 1996). These 2-day diets were designed around participants’
typical dietary intake taken from a 7-day food diary. Compliance was
assessed by using a checklist on which participants were asked to note any
changes to their prescribed diet. Along with their prescribed diet, participants
were asked to undertake the same exercise in the 48 hr before each trial to ensure
Table 1 Participant Characteristics (M ± SD), n = 10 in Each Group
Characteristic Placebo LC
Age (years) 32 ± 9 34 ± 10
Height (cm) 179 ± 7 178 ± 4
Body mass (kg) 75.7 ± 10.2 76.0 ± 9.5
Sum of skinfolds (mm) 62 ± 26 62 ± 27
VO2max (L/min) 4.92 ± 0.46 4.96 ± 0.64
Workload (W/kg) 3.1 ± 0.6 3.0 ± 0.6
Training history (years) 8.9 ± 5.3 9.0 ± 5.9
Current cycle training (hr/week) 6.5 ± 3.6 5.1 ± 2.4
570 Broad, Maughan, and Galloway
that any differences observed in nitrogen balance could be attributed to a carnitine
treatment effect and not to an effect of low glycogen stores (Lemon &
Mullin, 1980).
The Trials
Participants came to the laboratory for baseline measurements before treatment
commenced (0 weeks) and at the end of the 2-week supplementation period. A
24-hr urine collection was commenced 24 hr before the trial start time. Each trial
was undertaken 2 hr after a standardized meal consisting of 1 g/kg BM carbohydrate
(bread and jam). The last dose of the supplement was taken 3 hr before the
exercise trial, with a small snack that formed part of the prescribed diet. On participants’
arrival at the laboratory, we collected a pretrial urine sample before
assessing nude BM and supplied a heart-rate monitor. Participants then rested in a
supine position while a cannula (20 gauge, SSS Healthcare) was inserted into an
antecubital vein. After 5 min of seated rest a blood sample was drawn without
stasis, along with a free-flowing capillary sample from a preheated hand for analysis
of capillary pH, pCO2, and bicarbonate (Radiometer ABL 700, Copenhagen).
The cannula was kept patent at all times using a saline flush of 1 ml after sample
collections. Participants then began cycling for 5 min at 50% of their required
power output, followed by 85 min at a constant power output equivalent to
(M ± SD) 69.7% ± 4.4% VO2max, at a self-selected pedal cadence. Expired gas was
collected over 4 min at 15-min intervals (e.g., 13–17 min) from time zero using an
online gas-analysis system (Sensormedics Vmax 29, Holland) calibrated with
known gases before each test. Heart rate was recorded at 60-s intervals throughout
the trial, and a rating of perceived exertion (15-point Borg scale; Borg, 1982)
recorded every 10 min throughout exercise. Venous blood was drawn at rest and
15, 30, 60, and 90 min during exercise, and finger-prick capillary samples were
taken from a prewarmed hand at rest and at 30, 60, and 90 min of exercise. Water
was provided throughout the trial, with encouragement to achieve sufficient fluid
intake to prevent a reduction in body mass based on data collected in the familiarization
trials. Participants were cooled with a fan throughout all trials. After completing
the 90 min, they rested during the removal of the cannula, towel-dried, had
final nude BM recorded, and emptied their bladder again for sampling. For the
next 22 hr, participants maintained their prescribed dietary intake and undertook
another complete urine collection (thus completing 24 hr from the beginning of
their trial). If a participant needed to empty his bladder at any point during the
90 min he was allowed 2 min to attend to this, with a sample being drawn and the
volume included in calculations of fluid loss over the trial, and this did not alter
the total duration of activity conducted by the participants.
Blood and Urine Analysis
Before and at 15, 30, 60, and 90 min of exercise, duplicate 100-μl aliquots of
whole blood were immediately deproteinized in 1 ml of ice-cold 0.4-M perchloric
acid, shaken vigorously, and kept on ice until centrifugation at 10,000 rpm
for 3 min. Samples were subsequently frozen at –20 °C until analysis. Blood
lactate and glycerol were measured by fluorometric procedures (Jenway 6200
Carnitine Supplementation and Exercise Metabolism 571
fluorometer,
Jenway Ltd., Essex; Boobis & Maughan, 1983; Maughan, 1982).
The remaining blood was mixed well in EDTA tubes, and duplicate samples
were drawn into capillary tubes that were centrifuged at 10,000 rpm for microhematocrit
measurement. A further 1.5-ml portion of the blood sample was centrifuged
before duplicate aliquots of plasma were drawn off for glucose and
free-fatty-acid (FFA) analysis. Plasma glucose (Sigma Diagnostic), plasma FFA
(Wako Chemicals, Germany), and hemoglobin (cyanmethemoglobin method)
were assayed within 3 hr of blood draws using standard reagent kits (Hitachi
U2001, Hitachi Instruments Ltd., USA). Blood and plasma volume changes
were calculated from hematocrit and hemoglobin using standard equations (Dill
& Costill, 1974).
Additional blood was collected into lithium heparin tubes at rest and at 60
and 90 min exercise, centrifuged at 5,000 rpm at 4 °C for 10 min, and plasma
extracted into duplicate tubes and frozen at –60 °C until analysis. The rest and
90-min samples were used for noradrenaline analysis and adrenaline analysis by
high-perfomance liquid chromatography with electrochemical detection using the
methodology outlined by Goldstein, Feuerstein, Izzo, Kopin, and Keiser (1981),
and samples at rest and at 60 and 90 min were used to determine plasma carnitine
fractions by radiometric methods using liquid scintillation as outlined by McGarry
and Foster (1985).
Samples for amino acid assessment (BCAAs, alanine, and glutamate) were
prepared by mixing 80 μl of plasma (from the EDTA collection tube) with 20 μl of
1.375-mM internal standard 1 (L-methionine) and 10 μl of 3.3-M perchloric acid.
This mix was immediately vortexed, then centrifuged at 1,300 rpm for 10 min. The
supernatant was removed for analysis against a known standard by high-performance
liquid chromatography using fluorescence detection and precolumn
derivatization with 18 o-pthalaldehyde (Hypersel amino acid method, ThermoHypersil-
Keystone, Runcorn, UK) according to the method of Heinrikson and Meredith
(1984). In addition, duplicate 250-μl aliquots of plasma drawn from the lithium
heparin tube were immediately frozen at –20 °C until subsequent analysis
for
urea nitrogen and ammonia using Sigma Diagnostics kit 171-C for ammonia and
640-B for urea nitrogen (Sigma Diagnostics, St. Louis, MO, USA).
Urinary carnitine excretion was determined in each treatment period by
means of 24-hr urine collections before and after each exercise trial. A 5-ml
sample of mixed urine was collected and frozen at –60 °C until analysis, and the
total volume of urine excreted over the 24-hr period measured to the nearest milliliter.
Urinary carnitine fractions were subsequently analyzed (McGarry & Foster,
1985). An additional 5-ml sample was drawn from every urine collection before
volume measurement (including the immediate preexercise, immediate postexercise,
and any intervening collection) and was frozen at –20 °C until analysis
for urinary nitrogen determination via the total Kjeldahl nitrogen in water method
(Tecator application sub note ASN 3503) on a Tecator Kjeltec auto 1030 analyzer
(Foss, Denmark).
Nitrogen balance (assuming stable sweat and fecal losses) was estimated by
comparing the difference between 24-hr prescribed dietary protein intake
(divided by 6.25 to calculate nitrogen intake) and 24-hr urinary nitrogen excretion,
both before and after each exercise trial (Tarnopolsky, MacDougall, &
Atkinson, 1988).
572 Broad, Maughan, and Galloway
Statistics
All data were checked for normality of distribution and homogeneity of variance
before analysis. Within-group differences were assessed using repeated-measures
analysis of variance with time and trial as within-participant factors. Significant
main effects were then assessed using a paired t test with Bonferroni correction to
determine at which time points the differences lay. Changes between 0 and 2 weeks
were compared between groups using repeated-measures analysis of variance
with time as a within-participant factor and treatment group as a between-participants
factor. Differences between groups were then assessed using an independent-samples
t test with Bonferroni correction (SPSS version 11.0.0, SPSS Inc.). Significance
was accepted at p < .05 or Bonferroni-adjusted value. All data are expressed as
M ± SD unless otherwise specified.
Results
There was no difference between 0 and 2 weeks in the 2-day pretrial or the 24-hr
posttrial diet (Table 2). Participants were in apparent small positive nitrogen balance
throughout all exercise trials, with no difference between 0- and 2-week
trials or between treatment groups (Table 3).
Pretrial training and dietary controls were effective in ensuring that there
were no differences between trials or groups for preexercise plasma glucose (5.4 ±
1.0 and 5.5 ± 0.9 mmol/L for P at 0 weeks and P at 2 weeks, 5.3 ± 0.7 and 5.4 ±
0.6 mmol/L for LC at 0 weeks and LC at 2 weeks, respectively) or BM (75.0 ± 9.8
and 75.3 ± 9.8 kg for P at 0 weeks and P at 2 weeks, 75.7 ± 9.3 and 75.9 ± 9.4 kg
LC at 0 weeks and LC at 2 weeks, respectively). Blood and plasma volume fell by
the same degree (6–7% and 10–11%, respectively) in the first 15 min of steadystate
exercise (p < .01) and did not change further over the duration of exercise in
any trial. Furthermore, no differences were found between trials or groups for BM
change (–0.40 ± 0.27 vs. –0.50 ± 41 kg for P at 0 weeks and P at 2 weeks and
–0.50 ± 0.21 vs. –0.41 ± 0.37 kg for LC at 0 weeks and LC at 2 weeks, respectively)
or fluid intake over exercise (1.34 ± 0.28 vs. 1.34 ± 0.31 L for P at 0 weeks
and P at 2 weeks and 1.19 ± 0.31 vs. 1.22 ± 0.33 L for LC at 0 weeks and LC at
2 weeks, respectively); changes in hydration status over the exercise periods were
therefore small (0.5%) and the same in each trial. Exercise heart rate, cadence,
and rating of perceived exertion did not differ between the 0- and 2-week trials
Table 2 Composition of Prescribed Diets (M ± SD), n = 10 in Each
Group
2 Days Pretrial 24 hr Posttrial
Macronutrient Placebo LC Placebo LC
Energy (MJ) 13.5 ± 1.2 13.7 ± 1.9 14.0 ± 1.9 13.2 ± 2.0
CHO (g) 490 ± 59 499 ± 79 495 ± 68 500 ± 81
Protein (g) 123 ± 11 122 ± 14 119 ± 14 118 ± 19
Fat (g) 97 ± 10 91 ± 26 113 ± 26 83 ± 22
573
Table 3 Blood Urea Nitrogen, Plasma Amino Acids, Urinary Nitrogen Excretion, and Nitrogen Balance
Before and After 90 min of Exercise in Placebo- (P) and Carnitine-Supplemented (LC) Groups
Trial
Time
(min)
Urea N2
(mg/dL)
Total BCAA
(μmol/L)
Plasma
alanine
(μmol/L)
Plasma
glutamate
(μmol/L)
N2 excretion
(g in 24 hr) N2 balancea (g)
P 0 weeks 0 15.3 ± 2.0 418 ± 44 355 ± 80 63 ± 9 15 ± 5 4.3 ± 4.3
90 16.2 ± 1.9 415 ± 44 393 ± 58 50 ± 12b 16 ± 6 5.4 ± 5.3
P 2 weeks 0 15.5 ± 2.3 414 ± 82 382 ± 71 64 ± 18 16 ± 5 3.7 ± 4.9
90 16.3 ± 2.2 421 ± 72 413 ± 107 54 ± 14b 19 ± 6 2.6 ± 6.7
LC 0 weeks 0 15.0 ± 2.7 405 ± 59 386 ± 59 55 ± 13 17 ± 6 2.5 ± 4.7
90 15.7 ± 2.5 375 ± 40 412 ± 60 45 ± 17b 17 ± 6 5.1 ± 5.0
LC 2 weeks 0 14.5 ± 2.5 432 ± 109 407 ± 85 66 ± 26c 15 ± 6 4.3 ± 4.5
90 15.7 ± 2.5 445 ± 154 457 ± 113 52 ± 25b 16 ± 4 4.9 ± 5.0
Note. BCAA = branched-chain amino acids.
aNitrogen-balance data refer to 24 hr pre- and 24 hr postexercise, not 0 and 90 min. bSignificant change from resting value, p < .05. cGreater than LC 0-weeks resting
value, p < .05.
574 Broad, Maughan, and Galloway
within either group, although cadence was higher in the LC group (88 rpm) than
the P group (82 rpm; p < .05). Heart rate and rating of perceived exertion increased
over the duration of exercise (p < .01), whereas cadence fell (~5 rpm).
Hematological and Urinary Data
No differences were found for pH, pCO2, bicarbonate, glucose, or FFA responses
to supplementation between the P and LC groups (Table 4). There was no significant
change over 90 min of steady-state exercise for pH, pCO2, and plasma glucose,
whereas FFA and glycerol increased progressively throughout exercise in all
trials (p < .01). Mean blood lactate (Table 4) was below 2 mmol/L in both groups
at all times. There was no difference between groups for blood lactate at 0 weeks,
nor between 0 and 2 weeks. Blood lactate was elevated at 15 and 30 min in the LC
group at 2 weeks but did not quite reach statistical significance from 0 weeks.
There was no difference for glycerol between trials in the P group, whereas in the
LC group glycerol fell an average of 0.12 mmol/L from the 0- to 2-weeks trial
both at rest and during exercise (p = .07).
There was no difference in the exercise response of adrenaline (change over
exercise, 0 weeks P, 4.08 ± 3.10; 2 weeks P, 3.32 ± 2.10; 0 weeks LC, 3.21 ± 3.40;
and 2 weeks LC, 2.06 ± 1.62 nmol/L) or noradrenaline (0 weeks P, 8.76 ± 4.03;
2 weeks P, 9.41 ± 3.73; 0 weeks LC, 6.64 ± 2.37; and 2 weeks LC, 6.32 ± 1.66 nmol/L)
between trials within either group.
There were no between-trials changes in blood concentrations of urea nitrogen,
total BCAA, or alanine in either group (Table 3). There was also no difference
in urinary nitrogen excretion either over 24 hr (Table 3) or from immediately
before to after exercise (preexercise: 0.7 ± 0.5 g P 0 weeks, 0.9 ± 0.7 g P 2 weeks,
1.1 ± 1.1 g LC 0 weeks, and 1.2 ± 1.0 g LC 2 weeks; postexercise: 0.9 ± 0.6 g P
0 weeks, 0.9 ± 0.6 g P 2 weeks, 1.1 ± 0.3 g LC 0 weeks, and 1.2 ± 0.4 g LC
2 weeks). There was no change over the exercise period in blood BCAA or alanine
concentrations, but blood urea nitrogen increased progressively from 15 through
to 90 min of exercise (p < .01). Resting plasma glutamate was higher in LC after
2 weeks than at 0 weeks (p < .05), with no change between 0 and 2 weeks in P
(Table 3). Plasma glutamate concentrations fell over the duration of exercise in all
trials (p < .05).
Plasma ammonia increased over the exercise duration in all trials except for
the 2-weeks LC trial. Analysis revealed that plasma ammonia concentration was
suppressed toward the end of exercise at 2 weeks in the LC group compared with
0 weeks LC, but this did not quite reach statistical significance (Figure 1).
Substrate Metabolism
No significant difference was found in VO2, VCO2, VE, or respiratory-exchange
ratio (RER) during exercise between 0 and 2 weeks within either the P (mean
RER across the exercise period of 0.80 ± 0.03 and 0.80 ± 0.04 for P at 0 weeks and
P at 2 weeks, respectively) or the LC group (mean RER across the exercise period
of 0.80 ± 0.05 and 0.81 ± 0.04 for LC at 0 weeks and LC at 2 weeks, respectively),
and all except VCO2 changed across the exercise period, reflecting the expected
cardiovascular and ventilatory drift. Because of the absence of differences in
575
Table 4 Responses of Blood pH, pCO2 (kPa), Plasma Bicarbonate (mM, HCO3), Plasma Free Fatty Acids
(FFA; mM), Blood Glycerol (mM), Plasma Glucose (mM), and Blood Lactate (mM) to Exercise in Placebo- (P)
and Carnitine-Supplemented (LC) Groups
Variable Trial Rest 15 min 30 min 60 min 90 min
pH P 0 weeks 7.42 (0.01) — 7.38 (0.02) 7.39 (0.01) 7.41 (0.02)
2 weeks 7.41 (0.01) — 7.38 (0.02) 7.39 (0.03) 7.39 (0.02)
pH LC 0 weeks 7.41 (0.03) — 7.39 (0.02) 7.40 (0.03) 7.40 (0.03)
2 weeks 7.42 (0.02) — 7.38 (0.03) 7.39 (0.03) 7.40 (0.03)
pCO2 P 0 weeks 5.47 (0.26) — 5.52 (0.28) 5.46 (0.34) 5.34 (0.34)
2 weeks 5.48 (0.40) — 5.48 (0.44) 5.39 (0.41) 5.42 (0.33)
pCO2 LC 0 weeks 5.28 (0.35) — 5.27 (0.30) 5.28 (0.39) 5.20 (0.33)
2 weeks 5.38 (0.34) — 5.40 (0.37) 5.35 (0.40) 5.24 (0.31)
HCO3 P 0 weeks 25.7 (1.0) — 23.9 (1.3) 24.4 (1.1) 24.7 (1.2)
2 weeks 25.5 (0.
— 23.8 (1.2) 24.1 (1.6) 24.1 (1.2)
HCO3 LC 0 weeks 25.0 (1.4) — 23.5 (1.6) 24.2 (1.4) 24.0 (1.0)
2 weeks 25.7 (1.4) — 23.5 (1.3) 23.9 (1.2) 24.1 (1.3)
FFA P 0 weeks 0.28 (0.23) 0.17 (0.13) 0.28 (0.26) 0.42 (0.26) 0.61 (0.32)
2 weeks 0.26 (0.13) 0.18 (0.07) 0.26 (0.14) 0.42 (0.17) 0.54 (0.26)
FFA LC 0 weeks 0.33 (0.14) 0.24 (0.09) 0.37 (0.17) 0.60 (0.29) 0.76 (0.32)
2 weeks 0.23 (0.17) 0.18 (0.09) 0.32 (0.13) 0.52 (0.22) 0.75 (0.34)
(Continued)
576
Table 4 (Continued)
Variable Trial Rest 15 min 30 min 60 min 90 min
Glycerol P 0 weeks 0.07 (0.07) 0.12 (0.12) 0.16 (0.11) 0.21 (0.12) 0.28 (0.11)
2 weeks 0.10 (0.10) 0.14 (0.11) 0.15 (0.10) 0.21 (0.11) 0.29 (0.10)
Glycerol LC 0 weeks 0.21 (0.13) 0.25 (0.14) 0.26 (0.15) 0.34 (0.14) 0.42 (0.16)
2 weeks 0.12 (0.07) 0.14 (0.08) 0.17 (0.08) 0.21 (0.06) 0.28 (0.11)
Glucose P 0 weeks 5.41 (1.06) 4.44 (0.59) 4.45 (0.75) 4.29 (0.48) 3.93 (0.38)
2 weeks 5.55 (0.83) 4.41 (0.70) 4.57 (0.79) 4.52 (0.74) 4.20 (0.72)
Glucose LC 0 weeks 5.38 (0.68) 4.39 (0.46) 4.56 (0.79) 4.45 (0.55) 4.18 (0.51)
2 weeks 5.42 (0.56) 4.23 (0.76) 4.39 (0.60) 4.34 (0.50) 4.19 (0.46)
Lactate P 0 weeks 0.40 (0.33) 1.25 (0.46) 1.49 (0.55) 1.14 (0.39) 1.25 (0.30)
2 weeks 0.38 (0.24) 1.52 (0.54) 1.33 (0.56) 1.27 (0.69) 1.53 (0.32)
Lactate LC 0 weeks 0.27 (0.20) 1.49 (0.41) 1.52 (0.26) 1.34 (0.58) 1.36 (0.67)
2 weeks 0.26 (0.23) 1.80 (0.71) 1.96 (0.61) 1.43 (0.69) 1.25 (0.37)
Carnitine Supplementation and Exercise Metabolism 577
nitrogen balance or plasma amino acid concentrations during exercise as a consequence
of LC supplementation, CHO and fat utilization were estimated using the
nonprotein RER (Peronnet & Massicotte, 1991). There was no 0- to 2-weeks trial
difference in CHO oxidation between groups (Figure 2), although CHO oxidation
was higher at all time points during exercise in the 2-weeks trial than in the
0-weeks trial in the LC group. There was a trend toward a between-groups difference
in 0- to 2-weeks changes in fat oxidation during the 90 min of exercise
(p = .07; Figure 3).
Figure 1 — Plasma ammonia changes over exercise after 2 weeks of supplementation
with (a) placebo (P) and (b) L-carnitine L-tartrate (LC; M ± SEM). *p = .03 (not statistically
significant because of Bonferroni correction, p < .01), mean difference 17.1 μmol/L,
95% CI 0.52–33.73. †p = .09, mean difference 29.4 μmol/L, 95% CI –8.13 to 66.93.
578 Broad, Maughan, and Galloway
Total CHO oxidized over the exercise period was estimated to be 139 ± 33
and 132 ± 40 g in the P group and 137 ± 36 and 147 ± 32 g in the LC group for
0- and 2-weeks trials, respectively. For the same trials, total fat oxidized was estimated
to be 100 ± 16 and 105 ± 16 g in the P group and 105 ± 19 and 99 ± 21 g in
the LC group for 0- and 2-weeks trials, respectively.
Carnitine Status
Resting plasma total and acyl-carnitine fractions increased after 2 weeks of LC
supplementation (by 61% ± 42% and 152% ± 105%, respectively, p < .01), with
no change in free carnitine (17% ± 35% change, p = .27). There was no change in
any of these parameters at 2 weeks in P. Urinary carnitine excretion increased
Figure 2 — Rate of carbohydrate oxidation during 90 min of exercise in (a) the placebo
group (P) and (b) the L-carnitine L-tartrate group (LC; M ± SEM).
Carnitine Supplementation and Exercise Metabolism 579
after 2 weeks of LC supplementation (mean 6.4-fold, 25.2-fold, and 1.9-fold
increase for total, free, and acyl carnitine over 24 hr in urine, p < .01), with no
change in the P group.
Discussion
By prescribing diets for 2 days before and 24 hr after exercise and standardizing
pretrial exercise, we attempted to minimize the effect of factors that could influence
substrate metabolism during exercise, such as preexercise muscle glycogen
content (van Hall, Saltin, & Wagenmakers, 1999), plasma glucose and FFA concentrations
(Coyle, Coggan, Hemmert, Lowe, & Walters, 1985), and hydration
status (Shirreffs, Armstrong, & Cheuvront, 2004). Therefore, the absence of any
differences in blood urea nitrogen, urinary nitrogen excretion, nitrogen balance,
plasma BCAA, or alanine changes over exercise provides strong evidence that
Figure 3 — Rate of fat oxidation during 90 min of exercise in (a) the placebo group (P)
and (b) the L-carnitine L-tartrate group (LC; M ± SEM).
580 Broad, Maughan, and Galloway
there is no change in protein contribution to metabolism after 2 weeks of LC
supplementation, although this should be confirmed by an isotopic tracer study.
The fact that there were no differences in VCO2, pCO2, pH, and bicarbonate concentration
between trials and that VCO2, pCO2, and bicarbonate were stable across
the exercise period supports our assumption that the RER adequately reflects the
respiratory quotient and relative fat:CHO oxidation during steady-state exercise in
the current study (Peronnet & Massicotte, 1991).
The results of this study indicate no significant effect of LC supplementation
on CHO use during 90 min of steady-state exercise, but there was a tendency
toward a reduction in fat oxidation. In addition, plasma glycerol concentration
tended to be lower, and blood lactate higher, after 2 weeks of LC supplementation,
thereby supporting a tendency toward reduced mobilization and/or oxidation of
fatty acids. This is contrary to the promoted benefits of LC supplementation but
supports the trend shown by other studies in our laboratory after 2 weeks of LC
supplementation (Abramowicz & Galloway, 2005). There have been very few
well-controlled studies involving 2 weeks of LC supplementation that have measured
expired gas during exercise with which to compare these results. Marconi,
Sassi, Carpinelli, and Cerretelli (1985) found no difference in RER during 120 min
of walking at 65% VO2max in competitive walkers after 4 g LC/day, nor did Vukovich,
Costill, and Fink (1994) in participants performing 60 min of cycling exercise
at 70% VO2max after 6 g LC/day combined with a high fat preload. In contrast,
RER was decreased (indicating higher fat oxidation) in competitive runners (Williams,
Walker, Nute, Jackson, & Brooks, 1987) and untrained males (Wyss,
Ganzit, & Rienzi, 1990) after 3 weeks of LC supplementation. It is possible that
the duration of LC supplementation influences the effects observed on fuel metabolism
during exercise, and it has been suggested that periods of supplementation
of 8 weeks or longer might be required to observe effects on skeletal-muscle
metabolism because this is the usual procedure in animal studies (J. Harmeyer,
personal communication). Indeed, Arenas et al. (1991) observed that carnitine
ingestion (1 g twice daily over 6 months) prevented a training-induced decrease in
muscle free and total carnitine in trained athletes, but to date no studies using
shorter periods of supplementation have demonstrated any alteration in muscle
carnitine content with oral supplementation.
Responses of plasma BCAA, alanine, glutamate, and blood urea nitrogen
concentration to exercise were similar to those reported in other exercise trials in
humans (De Palo et al., 1993) and with LC supplementation (Angelini et al., 1986;
MacLean, Spriet, Hultman, & Graham, 1991). The increased plasma urea nitrogen
over the exercise bouts indicates that amino acids were catabolized during
exercise in this study (MacLean et al.). The lack of change in urinary nitrogen
excretion, which has been used to assess protein contribution to exercise in other
studies (Lemon & Mullin, 1980), either over the exercise period or over 24 hr after
exercise, indicates a low contribution of protein to exercise (<5% of total energy
expenditure). This might be because our participants were endurance trained and
because the prescribed diets ensured that they maintained energy balance, sufficient
CHO for training needs, and a positive nitrogen balance.
The novel finding of a tendency for blunting of ammonia (NH3) accumulation
toward the end of prolonged endurance exercise by LC in this study is consistent
with the findings that hyperammonemia is present in many cases of carnitine
Carnitine Supplementation and Exercise Metabolism 581
insufficiency (Llansola, Erceg, Hernandez-Viadel, Felipo, 2002). LC provision
has also previously been shown to reduce blood and brain ammonia and increase
glutamate concentrations, preventing the acute toxic effects of hyperammonemia
in mice (Grisolia, O’Connor, & Costell, 1984) and in epileptic children undergoing
valproate therapy (Gidal et al., 1997). Oyono-Enguelle et al. (1988), however,
found no difference in ammonia accumulation during exercise after 4 weeks
of supplementation with 2 g of LC per day, which might be related to the lower
exercise intensity (<50% VO2max) and/or shorter duration (60 min) not stimulating
the degree of ammonia production noted under our exercise conditions. The mean
resting NH3 concentrations in the current study are within the normal range of
20–60 μM (Graham, Turcotte, Kiens, & Richter, 1997), and the elevation over
exercise is similar to values reported during exercise of similar intensity and duration
(Bellinger, Bold, Wilson, Noakes, & Myburgh, 2000; MacLean et al., 1991;
Terjung & Tullson, 1992). This accumulation of plasma ammonia over exercise
correlates with muscle NH3 concentration and efflux (MacLean et al.). The primary
sources of increased NH3 are believed to be from deamination of AMP,
increased amino acid catabolism, or decreased removal, and NH3 might provide a
marker of muscle metabolic stress because its production increases toward the end
of endurance exercise and reflects the extent of the reliance of active muscle on
amino acid catabolism (Terjung & Tullson) or reflects low glycogen levels (Sahlin
& Broberg, 1990). Thus, NH3 accumulation has been linked with fatigue during
exercise (Ogino et al., 2000). In the absence of any change in estimated CHO
oxidation or nitrogen balance in the current study it would seem that glycogen
depletion or increased catabolism of amino acids cannot explain the apparent
blunting of ammonia accumulation during prolonged exercise after a period of
carnitine ingestion, and this effect could therefore be linked to increased removal
from the circulation.
Another mechanism for an attenuated NH3 accumulation could therefore be
through glutamate processing during exercise. Glutamate can accept an NH3
group to form glutamine, which is then released from muscle; it can also be
transaminated with pyruvate to form alanine or can be deaminated, producing
NH3 (Snow, Carey, Stathis, Febbraio, & Hargreaves, 2000). Because we also
observed no change in alanine or BCAA oxidation, it is possible that the lower
NH3 reflects an increased glutamine generation resulting from a more plentiful
supply of glutamate precursor before exercise, as was observed in this study.
Furthermore,
plasma NH3 and hypoxanthine concentrations have been shown to
be correlated (Ogino et al., 2000), and reduced hypoxanthine has been reported by
Volek et al. (2002) after LC supplementation, suggesting that carnitine can reduce
metabolic stress. Regardless of the mechanism, lowered NH3 concentrations
(especially toward the end of moderate- to high-intensity endurance exercise)
might reflect better maintenance of the ATP:AMP ratio in exercising muscle or
other metabolically active tissues and thus appear to be indicative of reduced metabolic
stress during exercise. If it is assumed, however, that muscle carnitine content
did not increase in our participant group, this raises the possibility that the
effects we observed on ammonia accumulation are the result of extramuscular
metabolic actions of carnitine in organs such as liver, kidney, heart, and brain
tissue that might affect ammonia production or removal and therefore deserve
further focused attention.
582 Broad, Maughan, and Galloway
Conclusion
This study indicates that LC supplementation does not appear to alter the proportional
contribution of protein, CHO, or fat to energy metabolism during prolonged
exercise in this sample of well-trained endurance athletes. LC supplementation
appears to blunt the accumulation of ammonia, which might reflect reduced metabolic
stress in the exercising muscle or increased ammonia removal from the circulation,
and this warrants further investigation.
Acknowledgments
The authors would like to thank Lonza Ltd., Basel, for their support of this research, and
Prof. Johein Harmeyer, Germany, for the analysis of carnitine fractions and his advice.
References
Abramowicz, W.N., & Galloway, S.D.R. (2005). Effects of acute versus chronic administration
of L-carnitine L-tartrate on the metabolic responses to submaximal exercise in
males and females. International Journal of Sport Nutrition and Exercise Metabolism,
15, 386–400.
Angelini, C., Vergani, L., Costa, L., Martinuzzi, A., Dunner, E., Marescotti, C., et al.
(1986). Use of carnitine in exercise physiology. Advances in Clinical Enzymology,
4, 103–110.
Arenas, J., Ricoy, J.R., Encinas, A.R., Pola, P., D’Iddio, S., Zeviani, M., et al. (1991).
Carnitine in muscle, serum, and urine of nonprofessional athletes: Effects of physical
exercise, training, and L-carnitine administration. Muscle & Nerve, 14, 598–604.
Bellinger, B.M., Bold, A., Wilson, G.R., Noakes, T.D., & Myburgh, K.H. (2000). Oral creatine
supplementation decreases plasma markers of adenine nucleotide degradation
during a 1-h cycle test. Acta Physiologica Scandinavica, 170, 217–224.
Boobis, L.H., & Maughan, R.J. (1983). A simple one-step enzymatic fluorometric method for
the determination of glycerol in 20 μl of plasma. Clinica Chimica Acta, 132, 173–179.
Borg, G.A.V. (1982). Psychophysical bases of perceived exertion. Medicine and Science in
Sports and Exercise, 14, 377–381.
Burke, L.M. (1996). The complete guide to food for sports performance. Sydney, Australia:
Allen & Unwin.
Constantin-Teodosiu, D., Cederblad, G., & Hultman, E. (1992). PDC activity and acetyl
group accumulation in skeletal muscle during prolonged exercise. Journal of Applied
Physiology, 73, 2403–2407.
Coyle, E.F., Coggan, A.R., Hemmert, M.K., Lowe, R.C., & Walters, T.J. (1985). Substrate
usage during prolonged exercise following a pre-exercise meal. Journal of Applied
Physiology, 59, 429–433.
De Palo, E.F., Metus, P., Gatti, R., Previti, O., Bigon, L., & De Palo, C.B. (1993). Branched
chain amino acids chronic treatment and muscular exercise performance in athletes: A
study through plasma acetyl-carnitine levels. Amino Acids, 4, 255–266.
Dill, D.B., & Costill, D.L. (1974). Calculation of percentages changes in volumes of blood,
plasma, and red cells in dehydration. Journal of Applied Physiology, 37, 247–248.
Fritz, I.B. (1963). Carnitine and its role in fatty acid metabolism. Advances in Lipid
Research, 1, 285–334.
Gidal, B.E., Inglese, C.M., Meyer, J.F., Pitterle, M.E., Antonopolous, J., & Rust, R.S.
(1997). Diet- and valproate-induced transient hyperammonemia: effect of L-carnitine.
Pediatric Neurology, 16, 301–305.
Carnitine Supplementation and Exercise Metabolism 583
Goldstein, D.S., Feuerstein, G., Izzo, J.L., Kopin, I.J., & Keiser, H.R. (1981). Validity
and reliability of liquid chromatography with electrochemical detection for measuring
plasma levels of norepinephrine and epinephrine in man. Life Sciences, 28,
467–475.
Graham, T.E., & MacLean, D.A. (1992). Ammonia and amino acid metabolism in human
skeletal muscle during exercise. Canadian Journal of Physiology and Pharmacology,
70, 132–141.
Graham, T.E., Turcotte, L.P., Kiens, B., & Richter, E.A. (1997). Effect of endurance training
on ammonia and amino acid metabolism in humans. Medicine and Science in
Sports and Exercise, 29, 646–653.
Grisolia, S., O’Connor, J.E., & Costell, M. (1984). L-Carnitine and ammonia toxicity. Bulletin
et Memoires de l’Academie Royale de Medecine de Belgique, 139, 289–297.
Heinrikson, R.L., & Meredith, S.C. (1984). Amino acid analysis by reverse-phase highperformance
liquid chromatography: Precolumn derivatization with phenylisothiocyanate.
Analytical Biochemistry, 136, 65–74.
Hoppel, C. (2003). The role of carnitine in normal and altered fatty acid metabolism. American
Journal of Kidney Diseases, 41, S4–S12.
Lemon, P.W.R., & Mullin, J.P. (1980). Effect of initial muscle glycogen levels on protein
catabolism during exercise. Journal of Applied Physiology, 48, 624–629.
Llansola, M., Erceg, S., Hernandez-Viadel, M., & Felipo, V. (2002). Prevention of ammonia
and glutamate neurotoxicity by carnitine: Molecular mechanisms. Metabolic
Brain Disease, 17, 389–397.
MacLean, D.A., Spriet, L.L., Hultman, E., & Graham, T.E. (1991). Plasma and muscle
amino acid and ammonia responses during prolonged exercise in humans. Journal of
Applied Physiology, 70, 2095–2103.
Marconi, C., Sassi, G., Carpinelli, A., & Cerretelli, P. (1985). Effects of L-carnitine loading
on the aerobic and anaerobic performance of endurance athletes. European Journal of
Applied Physiology, 54, 131–135.
Maughan, R.J. (1982). A simple, rapid method for the determination of glucose, lactate,
pyruvate, alanine, 3-hydroxybutyrate and acetoacetate on a single 20-μl blood sample.
Clinica Chimica Acta, 122, 231–240.
McGarry, J.D., & Foster, D.W. (1985). Free and esterified carnitine—Radiometric method.
In H.U. Bergmeyer (Ed.), Methods of enzymatic analysis (pp. 474–481). Germany:
VCH Verlagsgesellschaft mbH.
Norton, K., & Olds, T. (2000). Anthropometrica. Sydney, Australia: University of New
South Wales Press.
Ogino, K., Kinugawa, T., Osaki, S., Kato, M., Endoh, A., Furuse, Y., et al. (2000). Ammonia
response to constant exercise: differences to the lactate response. Clinical and
Experimental Pharmacology & Physiology, 27, 612–617.
Oyono-Enguelle, S., Freund, H., Ott, C., Gartner, M., Heitz, A., Marbach, J., et al. (1988).
Prolonged submaximal exercise and L-carnitine in humans. European Journal of
Applied Physiology, 58, 53–61.
Peronnet, F., & Massicotte, D. (1991). Table of nonprotein respiratory quotient: An update.
Canadian Journal of Sport Sciences, 16, 23–29.
Sahlin, K., & Broberg, S. (1990). Adenine nucleotide depletion in human muscle during
exercise: Causality and significance of AMP deamination. International Journal of
Sports Medicine, 11, S62–S67.
Shirreffs, S.M., Armstrong, L.E., & Cheuvront, S.N. (2004). Fluid and electrolyte needs for
preparation and recovery from training and competition. Journal of Sports Sciences,
22, 57–63.
Snow, R.J., Carey, M.F., Stathis, C.G., Febbraio, M.A., & Hargreaves, M. (2000). Effect of
carbohydrate ingestion on ammonia metabolism during exercise in humans. Journal
of Applied Physiology, 88, 1576–1580.
584 Broad, Maughan, and Galloway
Tarnopolsky, M.A., MacDougall, J.D., & Atkinson, S.A. (1988). Influence of protein intake
and training status on nitrogen balance and lean body mass. Journal of Applied Physiology,
64, 187–193.
Terjung, R.L., & Tullson, P.C. (1992). Ammonia metabolism during exercise. In D.R.
Lamb & C.V. Gisolfi (Eds.), Energy metabolism in exercise and sport (pp. 235–268).
Dubuque, IA: Brown & Benchmark.
van Hall, G., Saltin, B., & Wagenmakers, A.J.M. (1999). Muscle protein degradation and
amino acid metabolism during prolonged knee-extensor exercise in humans. Clinical
Science, 97, 557–567.
Veerkamp, J.H., Van Moerkerk, H.T.B., & Wagenmakers, A.J.M. (1985). Interaction of
short-chain and branched-chain fatty acids and their carnitine and CoA esters and of
various metabolites and agents with branched-chain 2-oxo acid oxidation in rat muscle
and liver mitochondria. The International Journal of Biochemistry, 17, 967–974.
Volek, J.S., Kraemer, W.J., Rubin, M.R., Gomez, A.L., Ratamess, N.A., & Gaynor, P.
(2002). L-Carnitine L-tartrate supplementation favorably affects markers of recovery
from exercise stress. American Journal of Physiology. Endocrinology and Metabolism,
282, E474–E482.
Vukovich, M.D., Costill, D.L., & Fink, W.J. (1994). Carnitine supplementation: Effect
on muscle carnitine and glycogen content during exercise. Medicine and Science in
Sports and Exercise, 26, 1122–1129.
Williams, C., Walker, M.P., Nute, M.G., Jackson, J., & Brooks, S. (1987). Fat metabolism
during prolonged exercise: Influence of carnitine supplementation. The Proceedings
of the Nutrition Society, 46, 138A.
Wyss, V., Ganzit, G.P., & Rienzi, A. (1990). Effects of L-carnitine administration on VO2max
and the aerobic–anaerobic threshold in normoxia and acute hypoxia. European Journal
of Applied Physiology, 60, 1–6.
