Abstract

The decline in fat oxidation at higher power outputs of exercise is a complex interaction between several mechanisms; however, the influence of mitochondrial bioenergetics in this process remains elusive. Therefore, using permeabilized muscle fibers from mouse skeletal muscle, we aimed to determine if acute exercise altered mitochondrial sensitivity to (1) adenosine diphosphate (ADP) and inorganic phosphate (Pi), or (2) carnitine palmitoyltransferase-I (CPT-I) independent (palmitoylcarnitine, PC) and dependent [palmitoyl-CoA (P-CoA), malonyl-CoA (M-CoA), and l-carnitine] substrates, in an intensity-dependent manner. As the apparent ADP Km increased to a similar extent following low (LI) and high (HI) intensity exercise compared with sedentary (SED) animals, and Pi sensitivity was unaltered by exercise, regulation of phosphate provision likely does not contribute to the well-established intensity-dependent shift in substrate utilization. Mitochondrial sensitivity to PC and P-CoA was not influenced by exercise, while M-CoA sensitivity was attenuated similarly following LI and HI. In contrast, CPT-I sensitivity to l-carnitine was only altered following HI, as HI exercise attenuated l-carnitine sensitivity by ∼40%. Moreover, modeling the in vivo concentrations of l-carnitine and P-CoA during exercise suggests that CPT-I flux is ∼25% lower following HI, attributed equally to reductions in l-carnitine content and l-carnitine sensitivity. Altogether, these data further implicate CPT-I flux as a key event influencing metabolic interactions during exercise, as a decline in l-carnitine sensitivity in addition to availability at higher power outputs could impair mitochondrial fatty acid oxidation.

Introduction

Skeletal muscle energetic demands increase profoundly during an acute bout of exercise, simultaneously activating many metabolic processes to sustain adenosine triphosphate (ATP) production [1]. While fat and carbohydrates represent the major substrates metabolized through oxidative phosphorylation, at higher power outputs beyond 65% of maximal aerobic capacity, a co-ordinated shift towards carbohydrate utilization occurs at the expense of fatty acid oxidation [2]. It is well established that numerous intracellular and extracellular regulatory points exist to mediate these reciprocal substrate interactions, mechanisms which are not mutually exclusive and occur in coordination [1]. Specifically, skeletal muscle fatty acid availability is influenced by adipose tissue lipolysis, blood flow delivery to working muscles, intramuscular lipid stores, and skeletal muscle lipolytic enzyme capacity, while the accumulation of metabolites can, in turn, influence glycolytic and carbohydrate flux [1]. Although these sites of regulation are important events influencing whole muscle substrate utilization, our mechanistic understanding of the intensity-dependent regulation of mitochondrial bioenergetics remains incomplete.

While classical literature focused on the importance of substrate availability for mitochondrial oxidative metabolism, it is now known that the regulation of enzyme sensitivity in addition to basic substrate-product concentrations are key events influencing mitochondrial ATP production [3]. For instance, moderate intensity continuous exercise has consistently been shown to attenuate mitochondrial ADP (adenosine diphosphate) sensitivity in both humans [4] and rodents [5,6]. While the functional consequences of this response remain unknown, the impairments in ADP sensitivity may be linked to the progressive rise in cytosolic free ADP during exercise [7,8]. As the accumulation of cellular metabolites occurs in an intensity-dependent manner [7], it is also possible that mitochondrial ADP sensitivity is influenced by power output, which could have implications on substrate selection. Furthermore, despite the role as a necessary co-substrate for mitochondrial ATP production [9], little is known regarding the regulation of inorganic phosphate (Pi) kinetics. Pi has traditionally been viewed as a minor regulator of oxidative phosphorylation [10,11], as concentrations required to drive half maximal respiration have been reported to be ∼30–50-fold lower [1113] than intramuscular free Pi content at rest [7,8]. However, similar to the kinetic properties of ADP [14], recent work in permeabilized muscle fibers has identified an apparent Km for Pi which may be more reflective of in vivo states [12]. Currently, it remains to be determined if mitochondrial Pi sensitivity is regulated following exercise in a co-ordinated manner with mitochondrial ADP sensitivity.

An additional consequence of the rise in free ADP at higher power outputs is the activation of glycolytic enzymes [1,7,15], promoting carbohydrate flux through pyruvate dehydrogenase (PDH), thus leading to the accumulation of mitochondrial acetyl-CoA. At the expense of free carnitine, matrix-derived acetyl-CoA is shuttled towards the formation of acetylcarnitine, buffering excess acetyl-CoA and preventing back inhibition of PDH. While this process is thought to maintain carbohydrate flux, a prominent theory proposes that the reduction in carnitine availability at higher power outputs of exercise limits carnitine palmitoyltransferase-I (CPT-I) flux, and therefore fatty acid oxidation (reviewed in [1618]). Despite this attractive theory implicating carnitine availability as a potential mechanism underlying reciprocal substrate selection, it is unknown if the magnitude of l-carnitine depletion with high intensity exercise is sufficient to limit CPT-I flux, given the sensitivity of this enzyme to various physiological l-carnitine concentrations [17,19,20].

Therefore, we determined the mitochondrial kinetic profiles of ADP, Pi, and l-carnitine in response to low (LI) and high (HI) intensity exercise in permeabilized muscle fibers. We provide evidence that alterations in mitochondrial ADP and Pi sensitivity may not contribute to the shift in substrate utilization at higher intensities of exercise. In contrast, l-carnitine sensitivity was only impaired following HI exercise, a response that appears required for l-carnitine concentrations to limit CPT-I flux. These data provide evidence that attenuated CPT-I flux could limit mitochondrial fatty acid oxidation at higher intensities of exercise, a process influenced by reductions in l-carnitine availability and CPT-I l-carnitine sensitivity.

Methods

Mice

C57Bl6 mice were bred on site at the University of Guelph. Animals were housed with a 12 : 12 h light–dark cycle and fed a standard chow diet ad libitum. All procedures were approved by the Animal Care Committee at the University of Guelph.

Acute steady-state exercise

Mice were acclimated to motorized rodent treadmills (Exer-3R treadmill, Columbus Instruments) for 10 min at 15 m/min, 5% grade, over a 3-day period prior to all exercise tests and steady-state exercise bouts. To determine an equivalent exercise duration matched for fatigue at each intensity, a subset of mice was run to volitional exhaustion at both a low intensity (LI, 15 m/min, 5% incline) and high intensity (HI, 20 m/min, 15% incline), separated by 1 week. Of the run-time reached on each maximal fatigue test (LI, 155 ± 11 min; HI, 48.5 ± 4.3 min), a 60% proportion was calculated to determine the duration of the acute steady-state exercise bouts, resulting in animals running for 90 min at low intensities or 30 min at high intensities.

Mice were subsequently randomly assigned to remain sedentary (SED) or perform a single bout of low (LI, 15 m/min, 5% grade, 90 min) or high (HI, 20 m/min, 15% grade, 30 min) intensity treadmill running. Immediately post-exercise, mice were anesthetized with an intraperitoneal injection of sodium pentobarbital (60 mg/kg) followed by liquid isoflurane inhalation. The red gastrocnemius muscles were removed, in which muscle from one limb were immediately snap frozen in liquid nitrogen and stored at −80°C until analyses, while the other was placed in ice-cold BIOPS preservation buffer for mitochondrial respiration experiments.

Preparation of permeabilized muscle fibers

The red gastrocnemius muscle placed in ice-cold BIOPS (50 mM MES, 7.23 mM K2EGTA, 2.77 mM CaK2EGTA, 20 mM imidazole, 0.5 mM dithiothreitol, 20 mM taurine, 5.77 mM ATP, 15 mM PCr, and 6.56 mM MgCl2·H2O; pH 7.1) was used to prepare permeabilized muscle fibers as previously described [21]. Briefly, muscles were trimmed of fat and connective tissue, and separated with fine-tipped forceps under a microscope. Fiber bundles were incubated on a rotor for 30 min at 4°C in 40 µg/ml saponin, a cholesterol-specific detergent that permeabilizes the plasma membrane. Permeabilized muscle fibers used for ADP and lipid-supported respiration experiments were then washed in MiR05 respiration buffer, containing EGTA (0.5 mM), MgCl2·6H20 (3 mM), potassium lactobionate (60 mM), KH2PO4 (10 mM), HEPES (20 mM), sucrose (110 mM), taurine (20 mM), and fatty acid free BSA (1 g/l); pH 7.1. Fibers used for Pi-supported respiration experiments were washed in respiration buffer in the absence of KH2PO4 for 30 min prior to performing respiration experiments.

Mitochondrial respiration

Mitochondrial respiration experiments in permeabilized muscle fibers were performed using high-resolution respirometry (Oroboros Oxygraph-2K: Oroboros Instruments, Innsbruck, Austria). All experiments were conducted in 2 ml of respiration buffer (with or without KH2PO4 as noted above) at 37°C and room saturating air, with constant stirring at 750 rpm. To prevent muscle fiber contraction, 5 µM blebbistatin was added prior to each experiment [14]. Following each experiment, fiber bundles were recovered, freeze-dried, and respiration data were normalized to fiber bundle weight.

ADP-supported respiration experiments (25, 100, 175, 250, 500, 1000, 2000, 4000, 6000, 8000, 10 000 µM ADP) were performed in the presence of 5 mM pyruvate and 1 mM malate (n = 9/group). 10 mM glutamate and 10 mM succinate were added following ADP titrations to determine maximal complex I and complex I/II-linked respiration. 10 µM cytochrome c was added to assess mitochondrial membrane integrity, and respiration did not significantly increase more than 10%. Respiratory control ratios (RCRs) were determined (state 3/state 4 respiration) to demonstrate the mitochondrial coupling. Pi titration experiments (25, 100, 175, 250, 500, 1000, 2000, 4000, 6000, 8000, 10 000 µM Pi) were conducted following the addition of 5 mM pyruvate and 1 mM malate, and 5 mM ADP (n = 9/group). 10 mM glutamate, 10 mM succinate, and 5 mM cytochrome C were added following Pi titrations.

Palmitoylcarnitine (PC) sensitivity was determined in the presence of 1 mM malate and 5 mM ADP, by performing titrations of 10, 20, 30, and 40 µM PC (n = 4–5/group). Submaximal palmitoyl-CoA (P-CoA) respiratory kinetics were determined in the presence of 1 mM malate, 5 mM ADP, and 1 mM l-carnitine. P-CoA titrations were performed (20, 40, 60 µM P-CoA), followed by the addition of 7 µM malonyl-CoA (M-CoA) to assess the inhibitory effects of M-CoA on CPT-I-mediated fatty acid transport (n = 9/group). To examine l-carnitine kinetic properties, titration experiments were performed (10, 25, 50, 100, 175, 250, 350, 500 µM l-carnitine) in the presence of 1 mM malate, 5 mM ADP, and 60 µM P-CoA (n = 9/group). To model in vitro concentrations, experiments in a separate subgroup of mice (n = 5–7/group) were performed in the presence of 1 mM malate and 5 mM ADP. For SED animals, 10 µM P-CoA and 250 µM l-carnitine were subsequently added to the respiration medium. LI conditions were simulated with the addition of 60 µM P-CoA and 175 µM l-carnitine, and HI with the addition of 60 µM P-CoA and 100 µM l-carnitine.

Western blotting

Whole red gastrocnemius muscle was homogenized in lysis buffer, centrifuged for 15 min at 1500g and 4°C, and diluted to 1 µg/µl protein content. Equal amounts of each sample were loaded for separation by SDS–polyacrylamide gel electrophoresis and transferred to a polyvinylidene difluoride (PVDF) membrane. Membranes were incubated in commercially available primary antibodies to detect OXPHOS (5 µl protein; Mitosciences; ab110413), ANT1 (10 µl protein; 1 : 1000; Abcam; ab110322), VDAC (10 µl protein; 1 : 1000 Abcam; ab14734), CPT-I (20 µl protein, 1 : 1000; CPT-I M11-A), AMPK total (10 µl protein, 1 : 1000; Cell Signaling; 2532), Phospho Thr172 AMPK (10 µl protein, 1 : 1000; Cell Signaling; 2535), β-HAD (10 µl protein, 1 : 1000; Abcam; ab68321). α-tubulin (10 µl protein, 1 : 1000; Abcam; ab7291) was used as a loading control. Western blots were quantified using FluorChem HD imaging chemiluminescence (Alpha Innotech, Santa Clara, USA). All samples for each protein were loaded and detected on the same membrane.

Predicted CPT-I flux

Calculated predictions of CPT-I flux were determined using the kinetic properties of l-carnitine-supported respiration depicted by our data, and previous reports of intracellular l-carnitine concentrations at rest and during exercise. V0, or predicted CPT-I flux, was determined using the Michaelis–Menten equation 
formula

Statistics

Statistical analyses were completed using GraphPad Prism 8 software (GraphPad Software, Ins, LA Jolla, USA). One-way ANOVAs were used to compare all parameters between SED, LI, and HI states with a Tukey's multiple comparisons post hoc analysis. ADP, Pi, and l-carnitine titrations were analyzed using Michaelis–Menten kinetics with a constraint of 100, and PC titrations using non-constrained Michaelis–Menten kinetics, with GraphPad Prism software as previously described [4]. Statistically significance was determined as P < 0.05. Data were expressed as mean ± SEM.

Results

Animal characterization

We first aimed to examine AMPK phosphorylation to ensure the red gastrocnemius muscle used in all experiments displayed cellular stress indicative of an acute exercise response. While both intensities of exercise increased AMPKThr172 phosphorylation, this effect occurred to a greater degree following HI compared with LI (Figure 1A), verifying the intensity-dependent increase in metabolic demand associated with the exercise intensities utilized in the present study. We next aimed to confirm the absence of changes in mitochondrial protein content and respiratory capacity between groups, as this could confound interpretations when examining potential mechanisms of action influenced by exercise intensity. To achieve this, we performed Western blotting on red gastrocnemius muscle and classical respiration experiments in permeabilized muscle fibers. Importantly, the content of electron transport chain subunits (CI, CII, CIII, CV) and proteins involved in facilitating fatty acid metabolism (CPT-I, β-HAD) and ADP transport (VDAC, ANT1) were not different between groups (Figure 1B,C). In addition, there were no differences in leak (absence of ADP and Pi) or maximal aerobic respiration (presence of ADP and Pi) between groups (Figure 1D,E). Combined, these data confirm that animal groups were similar with respect to mitochondrial capacity.

Exercise at high and low intensities does not influence mitochondrial protein content or respiratory function.

Figure 1.
Exercise at high and low intensities does not influence mitochondrial protein content or respiratory function.

Acute exercise increases AMPK phosphorylation in an intensity-dependent manner (A), but does not alter electron transport chain subunits, ADP transporters, or proteins involved in mitochondrial fatty acid transport and oxidation (B,C). α-tubulin was used as a loading control. Exercise did not alter maximal complex I/II-linked respiration and RCRs supported by ADP (D) or Pi (E). AMPK, AMP-activated protein kinase; ANT1, adenine nucleotide transporter 1; CPT-I, carnitine-palmitoyl transferase I; β-HAD, beta-hydroxyacyl-CoA dehydrogenase; VDAC, voltage-dependent ion channel; PM, pyruvate + malate; +D; PM + ADP; +G, PMD + glutamate; +S, PMDG + succinate; CytoC, Cytochrome C; RCR, respiratory control ratio. * indicates significant difference (P < 0.05) compared with SED. # indicates significant difference (P < 0.05) compared with LI (one-way ANOVA, Tukey post hoc analysis). Data expressed as mean ± SEM. n = 7–9/experiment.

Figure 1.
Exercise at high and low intensities does not influence mitochondrial protein content or respiratory function.

Acute exercise increases AMPK phosphorylation in an intensity-dependent manner (A), but does not alter electron transport chain subunits, ADP transporters, or proteins involved in mitochondrial fatty acid transport and oxidation (B,C). α-tubulin was used as a loading control. Exercise did not alter maximal complex I/II-linked respiration and RCRs supported by ADP (D) or Pi (E). AMPK, AMP-activated protein kinase; ANT1, adenine nucleotide transporter 1; CPT-I, carnitine-palmitoyl transferase I; β-HAD, beta-hydroxyacyl-CoA dehydrogenase; VDAC, voltage-dependent ion channel; PM, pyruvate + malate; +D; PM + ADP; +G, PMD + glutamate; +S, PMDG + succinate; CytoC, Cytochrome C; RCR, respiratory control ratio. * indicates significant difference (P < 0.05) compared with SED. # indicates significant difference (P < 0.05) compared with LI (one-way ANOVA, Tukey post hoc analysis). Data expressed as mean ± SEM. n = 7–9/experiment.

ADP and Pi-supported respiration

We next examined the effects of exercise intensity on mitochondrial ADP sensitivity. While exercise impaired ADP sensitivity, there were no differences in this response between exercise intensities, as both LI and HI increased the apparent ADP Km ∼ 30% (Figure 2A,C). In contrast with the regulation of ADP kinetics, exercise at either intensity did not alter mitochondrial respiratory sensitivity to Pi (Figure 2B,D). Altogether, these data suggest that the acute exercise-mediated regulation of ADP and Pi sensitivity is comparable regardless of exercise intensity, and therefore may not influence the intensity-dependent shift in substrate utilization.

Acute exercise attenuates ADP sensitivity but does not alter Pi sensitivity.

Figure 2.
Acute exercise attenuates ADP sensitivity but does not alter Pi sensitivity.

Exercise, at both intensities, impaired ADP sensitivity (A,C). Neither intensity of exercise altered Pi-supported respiration (B,D). Concentrations of 25, 100, 175, 250, 500, 1000, 2000, 4000, 6000, 8000, and 10 000 µM were used for both ADP and Pi titration experiments. * indicates significant difference (P < 0.05) compared with SED (one-way ANOVA, Tukey post hoc analysis). Data expressed as mean ± SEM. n = 7–9/experiment.

Figure 2.
Acute exercise attenuates ADP sensitivity but does not alter Pi sensitivity.

Exercise, at both intensities, impaired ADP sensitivity (A,C). Neither intensity of exercise altered Pi-supported respiration (B,D). Concentrations of 25, 100, 175, 250, 500, 1000, 2000, 4000, 6000, 8000, and 10 000 µM were used for both ADP and Pi titration experiments. * indicates significant difference (P < 0.05) compared with SED (one-way ANOVA, Tukey post hoc analysis). Data expressed as mean ± SEM. n = 7–9/experiment.

CPT-I-supported respiration

We next examined the ability of exercise intensity to alter mitochondrial lipid sensitivity. Mitochondrial respiratory sensitivity to PC was not influenced by exercise at either intensity (Figure 3A,B), suggesting an absence of changes in mitochondrial β-oxidation enzymes. We next performed titration experiments to determine the sensitivity of CPT-I to P-CoA and the effects of M-CoA inhibition. Consistent with previous work from our laboratory [6,22], acute exercise did not alter the respiratory sensitivity to P-CoA (Figure 3C), and while exercise significantly attenuated the inhibitory effects of M-CoA, this was not influenced by exercise intensity (Figure 3D). We next titrated l-carnitine to examine the sensitivity of CPT-I to this substrate. While we did not observe any differences in maximal l-carnitine-supported respiration in the presence of P-CoA (Figure 4A), titration experiments revealed that, in contrast with the kinetic properties of PC and P-CoA, the sensitivity of mitochondria to l-carnitine was influenced by exercise intensity (Figure 4B). More specifically, while l-carnitine sensitivity was similar between SED (apparent Km = 22.7 µM) and LI (apparent Km= 23.6 µM), HI exercise attenuated this response almost 40% (apparent Km = 35.0 µM, P < 0.03) (Figure 4C).

Acute exercise does not alter PC or P-CoA sensitivity; however, exercise reduced the sensitivity to M-CoA inhibition (7 µM M-CoA).

Figure 3.
Acute exercise does not alter PC or P-CoA sensitivity; however, exercise reduced the sensitivity to M-CoA inhibition (7 µM M-CoA).

CPT-I-independent sensitivity to PC was not influenced by exercise at either intensity (A,B). Concentrations of 10, 20, 30, and 40 µM were used for PC titration experiments. Exercise also did not alter P-CoA respiratory kinetics (C), while M-CoA inhibition of lipid-supported respiration was reduced with exercise at both intensities (D). MD, malate + ADP; LC, l-carnitine; M-CoA, malonyl-CoA; PC, palmitoylcarnitine; P-CoA, palmitoyl-CoA. * indicates significant difference (P < 0.05) compared with SED (one-way ANOVA, Tukey post hoc analysis). Data were expressed as mean ± SEM. n = 6/experiment due to the inability to recover some fibers for accurate dry weight determination following respiration experiments. n = 4–5/experiment for PC titrations.

Figure 3.
Acute exercise does not alter PC or P-CoA sensitivity; however, exercise reduced the sensitivity to M-CoA inhibition (7 µM M-CoA).

CPT-I-independent sensitivity to PC was not influenced by exercise at either intensity (A,B). Concentrations of 10, 20, 30, and 40 µM were used for PC titration experiments. Exercise also did not alter P-CoA respiratory kinetics (C), while M-CoA inhibition of lipid-supported respiration was reduced with exercise at both intensities (D). MD, malate + ADP; LC, l-carnitine; M-CoA, malonyl-CoA; PC, palmitoylcarnitine; P-CoA, palmitoyl-CoA. * indicates significant difference (P < 0.05) compared with SED (one-way ANOVA, Tukey post hoc analysis). Data were expressed as mean ± SEM. n = 6/experiment due to the inability to recover some fibers for accurate dry weight determination following respiration experiments. n = 4–5/experiment for PC titrations.

l-carnitine sensitivity is regulated in an intensity-dependent manner.

Figure 4.
l-carnitine sensitivity is regulated in an intensity-dependent manner.

In the absence of changes in l-carnitine Vmax (A), HI exercise attenuated l-carnitine sensitivity (B,C) compared with LI and SED. Concentrations of 10, 25, 50, 100, 175, 250, 350, and 500 µM were used for l-carnitine titration experiments. * indicates significant difference (P < 0.05) compared with SED. # indicates significant difference (P < 0.05) compared with LI (one-way ANOVA, Tukey post hoc analysis). Data were expressed as mean ± SEM. n = 5–7 in (A) due to the inability to recover some fibers for accurate dry weight determination following respiration experiments. n = 8–9/respiration experiment in (B,C).

Figure 4.
l-carnitine sensitivity is regulated in an intensity-dependent manner.

In the absence of changes in l-carnitine Vmax (A), HI exercise attenuated l-carnitine sensitivity (B,C) compared with LI and SED. Concentrations of 10, 25, 50, 100, 175, 250, 350, and 500 µM were used for l-carnitine titration experiments. * indicates significant difference (P < 0.05) compared with SED. # indicates significant difference (P < 0.05) compared with LI (one-way ANOVA, Tukey post hoc analysis). Data were expressed as mean ± SEM. n = 5–7 in (A) due to the inability to recover some fibers for accurate dry weight determination following respiration experiments. n = 8–9/respiration experiment in (B,C).

Predicted inhibition of CPT-I flux

The Michaelis–Menten l-carnitine kinetic curves depicted by our data spanned the published values for skeletal muscle l-carnitine content (Figure 5). Importantly, while resting concentrations appeared to represent saturating states, exercise is known to reduce muscle l-carnitine availability, which appears to reach a value that would limit CPT-I flux (Figure 5). Specifically, calculations performed using the kinetic properties and intramuscular l-carnitine concentrations associated with each metabolic condition allowed for an estimation of CPT-I flux (Table 1). In the absence of a change in l-carnitine sensitivity, the reduction in l-carnitine content associated with exercise would only be expected to predict a 3% and 11% decline in CPT-I flux in vitro, in response to LI and HI, respectively. However, when accounting for the attenuation in l-carnitine sensitivity, in addition to reduced content, a 20% reduction in CPT-I flux was predicted in vitro following HI (Table 1), in the absence of other intracellular regulatory points. These data indicate that independent reductions in l-carnitine availability or sensitivity are not sufficient to substantially decrease estimations of CPT-I flux at higher power outputs, however when combined, these effects may be important for the well characterized reduction in fat oxidation.

Comparison between l-carnitine kinetic properties and published values of l-carnitine content.

Figure 5.
Comparison between l-carnitine kinetic properties and published values of l-carnitine content.

l-carnitine-mediated respiratory kinetic curves span the published values of l-carnitine content within mouse skeletal muscle [25,31]. Concentrations of 10, 25, 50, 100, 175, 250, 350, and 500 µM were used for l-carnitine titration experiments. Minimal inhibition of l-carnitine-supported respiration (JO2 > 90% of Vmax) is observed at concentrations >175 µM l-carnitine.

Figure 5.
Comparison between l-carnitine kinetic properties and published values of l-carnitine content.

l-carnitine-mediated respiratory kinetic curves span the published values of l-carnitine content within mouse skeletal muscle [25,31]. Concentrations of 10, 25, 50, 100, 175, 250, 350, and 500 µM were used for l-carnitine titration experiments. Minimal inhibition of l-carnitine-supported respiration (JO2 > 90% of Vmax) is observed at concentrations >175 µM l-carnitine.

Table 1
Predicted inhibition of CPT-I flux at constant P-CoA (60 µM) as a result of reductions in l-carnitine availability and/or sensitivity

Data calculated using skeletal muscle l-carnitine content in mice [25,31] and ratios of intensity-dependent l-carnitine availability in humans [16,17,24,33,34,40].

 CPT-I-supported kinetic properties Δ l-carnitine content Δ l-carnitine content and sensitivity 
Vmax (pmol*s−1 *mg−1 dry wt) l-carnitine Km (µM) [l-carnitine] (µM) Predicted flux (pmol*s−1*mg−1 dry wt) % Inhibition Predicted flux (pmol/s/mg dry wt) % Inhibition 
SED 87.9 22.7 250 80.5 — 80.5 — 
LI 87.9 23.6 175 77.8 −3% 77.4 −4% 
HI 87.9 35.0 100 71.6 −11% 65.0 −20% 
 CPT-I-supported kinetic properties Δ l-carnitine content Δ l-carnitine content and sensitivity 
Vmax (pmol*s−1 *mg−1 dry wt) l-carnitine Km (µM) [l-carnitine] (µM) Predicted flux (pmol*s−1*mg−1 dry wt) % Inhibition Predicted flux (pmol/s/mg dry wt) % Inhibition 
SED 87.9 22.7 250 80.5 — 80.5 — 
LI 87.9 23.6 175 77.8 −3% 77.4 −4% 
HI 87.9 35.0 100 71.6 −11% 65.0 −20% 

CPT-I-supported respiratory flux

Our estimates of CPT-I flux obtained from l-carnitine titration experiments were performed in the presence of equal concentrations of P-CoA (60 µM); however, intramuscular levels of P-CoA are known to be regulated as a function of metabolic state [23]. Therefore, to further verify our predictions of CPT-I flux during exercise, we performed a subset of respiration experiments in permeabilized muscle fibers utilizing physiologically relevant substrate concentrations present at rest and during exercise (Figure 6). Specifically, animals remained SED or performed a single bout of either LI or HI exercise (n = 5–7/group). Following the addition of malate and ADP, respiration was determined in the presence of 10 µM P-CoA and 250 µM l-carnitine (SED), or higher P-CoA concentrations (60 µM for both LI and HI) and lower l-carnitine concentrations (175 µM or 100 µM for LI and HI, respectively) [2325]. Compared with SED, this approach revealed that fatty acid-supported respiration increased ∼75% when modeling LI, while despite the similarly higher P-CoA availability, mitochondrial fatty acid-supported respiration only increased ∼35% following HI, likely as a result of the reduction in l-carnitine content and sensitivity.

Lipid-supported respiration in the presence of physiological substrate concentrations associated with each metabolic state.

Figure 6.
Lipid-supported respiration in the presence of physiological substrate concentrations associated with each metabolic state.

While lipid-supported respiration was greater following both LI (60 µM P-CoA + 175 µM LC) and HI (60 µM P-CoA + 100 µM LC) compared with SED (10 µM P-CoA + 250 µM LC), the response to HI was 30% lower than LI. LC, l-carnitine; P-CoA, palmitoyl-CoA. * indicates significant difference (P < 0.05) compared with SED. # indicates significant difference (P < 0.05) compared with LI (one-way ANOVA, Tukey post hoc analysis). Data expressed as mean ± SEM. n = 5–7/experiment.

Figure 6.
Lipid-supported respiration in the presence of physiological substrate concentrations associated with each metabolic state.

While lipid-supported respiration was greater following both LI (60 µM P-CoA + 175 µM LC) and HI (60 µM P-CoA + 100 µM LC) compared with SED (10 µM P-CoA + 250 µM LC), the response to HI was 30% lower than LI. LC, l-carnitine; P-CoA, palmitoyl-CoA. * indicates significant difference (P < 0.05) compared with SED. # indicates significant difference (P < 0.05) compared with LI (one-way ANOVA, Tukey post hoc analysis). Data expressed as mean ± SEM. n = 5–7/experiment.

Discussion

We provide evidence that acute exercise does not alter mitochondrial Pi-supported respiratory kinetics, and that acute exercise-mediated attenuations in ADP sensitivity are not intensity dependent. In contrast, CPT-I-supported respiration is regulated in an intensity-dependent manner during exercise, in which impairments in l-carnitine sensitivity are only present following HI exercise. Altogether, these data further implicate l-carnitine-mediated CPT-I flux as a control point in intensity-dependent mitochondrial fatty acid oxidation.

ADP and Pi sensitivity do not contribute to intensity-dependent substrate selection

Since ATP supply, demand, and energetic mismatch, indicated by the accumulation of free ADP, differ as a function of exercise intensity, we speculated that the regulation of mitochondrial ADP kinetics may provide a potential mechanism implicated in reciprocal substrate selection. Consistent with previous work from our laboratory in both humans [4] and rodents [5], we confirm that acute low intensity exercise impaired ADP sensitivity in the absence of creatine. In support of a recent report determining high intensity exercise attenuated mitochondrial ADP sensitivity [26], we provide further evidence that this effect is of a similar magnitude to that of low intensities. It is unclear why a previous report showed an improvement in mitochondrial ADP sensitivity following HI exercise, although this discrepancy may reflect methodological differences of the microbiopsy technique [27]. In the present study, the kinetic properties of Pi were not altered with acute exercise at either intensity, yet similar to Scheibye-Knudsen et al. [12], the apparent Pi Km determined by our data (∼800–900 µM Pi) appears to be more representative of the cellular environment than traditional reports from isolated mitochondrial preparations within the 100–250 µM range [11,13]. Since intramuscular concentrations of free Pi (∼6 mM at rest) [7] are orders of magnitude greater than the apparent Pi Km, these data suggest that the acute control of mitochondrial respiration is mediated to a greater extent by ADP. In this respect, despite the attenuation in mitochondrial ADP sensitivity during exercise, respiration is stimulated as a result of cytosolic-free ADP increasing from ∼25 µM at rest to concentrations nearing 200 µM during intense exercise [7]. Therefore, a greater proportion of maximal ADP-supported respiration is maintained as a result of increased substrate availability (Figure 3A).

Exercise intensity-dependent regulation of CPT-I flux

While CPT-I is known to be a major control point in mitochondrial fatty acid oxidation [28,29], the present data support previous work [6,22] suggesting that, while CPT-I becomes less sensitive to the inhibitory effects of M-CoA during exercise, the sensitivity of CPT-I to P-CoA does not change. We also extend these findings to show that the regulation of PC, P-CoA, and M-CoA sensitivity are not intensity-dependent, findings in accordance with previous reports that M-CoA content is not altered by exercise intensity [30]. Therefore, we examined l-carnitine sensitivity as an alternative mechanism for the attenuation in fatty acid oxidation at higher intensities of exercise. While free l-carnitine is well established to decline in an intensity-dependent manner in both rodents [25,31] and humans [24], historical reports in isolated mitochondria from human skeletal muscle have determined an apparent l-carnitine Km of 200–500 µM [19,20], appearing to be 4–10-fold lower than intramuscular l-carnitine concentrations at high intensities of exercise. While these data would suggest that l-carnitine availability may not be limiting CPT-I flux, we aimed to determine l-carnitine sensitivity in permeabilized muscle fibers, which appear to be a more appropriate technique for predicting in vivo rates of mitochondrial substrate provision and respiratory kinetics [6,12,14,32]. Our data establish that the sensitivity of CPT-I-supported respiration to l-carnitine spans published values of intramuscular l-carnitine content in mouse skeletal muscle during varying metabolic states. Specifically, at rest, l-carnitine sensitivity (apparent Km = 22.7 µM) remains well below basal l-carnitine availability of ∼250 µM [25,31], suggesting a saturated process prior to any metabolic perturbations. During moderate intensity treadmill running, skeletal muscle l-carnitine availability has been shown to decline to ∼175 µM [25,31]; however, as fat oxidation is substantially elevated compared with resting states, this has remained an unexplained discrepancy in the carnitine hypothesis [1]. To reconcile this finding, our data suggest that, in the absence of changes in l-carnitine sensitivity, the reduction in l-carnitine availability at lower power outputs does not impair lipid-supported respiration. In contrast, while l-carnitine content is known to decline by up to 70% at higher power outputs in humans [24,33,34], thus nearing a content of 100 µM in mouse models, our data suggest l-carnitine-mediated CPT-I flux may become an important regulatory point. Combined with a reduction in l-carnitine sensitivity at high intensities, this would be predicted to attenuate CPT-I flux by ∼20% in vitro, in the absence of other regulation that may also decrease fatty acid provision to mitochondria, and thus further compromise fatty acid oxidation (e.g. decreased blood flow to white adipose tissue and reductions in arterial FFA concentrations [1]). As we found an ∼25% reduction in lipid-supported respiration when modeling in vivo substrate concentrations during in vitro permeabilized muscle fiber experiments, this provides further evidence that CPT-I l-carnitine sensitivity and availability may contribute to intensity-dependent changes in mitochondrial fatty acid oxidation.

While the current data highlight the importance of CPT-I flux in vitro, several other factors influence this relationship in the in vivo environment. Specifically, our experiments were modeled in the absence of other intracellular and cytosolic ligands, which could alter classical Michaelis–Menten enzyme kinetics. We used established substrate concentrations representative of the intramuscular environment to predict CPT-I flux; however, the precise subcellular localization of these metabolites in vivo remains a contentious issue and methodological challenge, which could influence the accessibility to CPT-I and thus mitochondrial substrate provision. Furthermore, key enzymes involved in adipose tissue and skeletal muscle lipolysis (hormone sensitive lipase, adipose tissue triglyceride lipase) and plasma membrane fatty acid transport (e.g. FAT/CD36) may be influenced by exercise intensity, which could alter fatty acid availability for CPT-I flux and subsequent rates of fatty acid oxidation [23]. Despite these limitations, the present study was designed to determine the potential regulatory points within mitochondria, and therefore permeabilized muscle fibers were used in our study to purposely remove these regulatory points. In addition, permeabilized fibers are known to be more representative of the in vivo environment than traditional isolated mitochondrial experiments, given the intact cellular and microtubular structure which is known to influence substrate sensitivity [6,32]. In addition, previous in vitro work has provided important insight into the predicted flux through key enzymes including phosphofructokinase [15] and CPT-I [35,36] when modeling changes in the intracellular environment associated with exercise, in the absence of extracellular in vivo conditions. Our data are consistent with these findings highlighting the importance of substrate sensitivity, and specifically, the potential implications this could have on CPT-I flux during exercise.

Perspectives and conclusion

The current data provide evidence that the intensity-dependent shift in substrate selection during exercise does not appear to be related to mitochondrial phosphate provision, as mitochondrial ADP and Pi sensitivity were regulated in a similar manner following exercise at either intensity. However, we demonstrate that reductions in l-carnitine sensitivity, in addition to availability, may be a mechanism contributing in the decline in mitochondrial fat oxidation at high intensities of exercise. Our data further support CPT-I as an attractive control point in this process, in accordance with previous findings suggesting pH-driven reductions in CPT-I activity [35], a decline in CoA availability [17,37], and attenuated breakdown of plasma and intramuscular lipid substrates required for CPT-I flux [38,39] at higher power outputs. While future work is required to provide insight into the physiological relationship between l-carnitine sensitivity and availability in human skeletal muscle, our data suggest that CPT-I flux is attenuated during higher intensities of exercise. Mediated by changes in l-carnitine content and sensitivity, this process could provide an additional mechanism which may be important in the reduction in mitochondrial fatty acid oxidation at higher power outputs.

Abbreviations

     
  • ADP

    adenosine diphosphate

  •  
  • AMPK

    5'AMP-activated protein kinase

  •  
  • ANT

    adenine nucleotide translocase

  •  
  • ATP

    adenosine triphosphate

  •  
  • AU

    arbitrary optical density units

  •  
  • β-HAD

    betahydroxyacyl-CoA dehydrogenase

  •  
  • CPT-I

    carnitine palmitoyltransferase-I

  •  
  • HI

    high intensity exercise

  •  
  • JO2

    mitochondrial O2 flux

  •  
  • Km

    Michaelis-Menten constant

  •  
  • LI

    low intensity exercise

  •  
  • M-CoA

    malonyl-CoA

  •  
  • PC

    palmitoylcarnitine

  •  
  • P-CoA

    palmitoyl-CoA

  •  
  • PDH

    pyruvate dehydrogenase

  •  
  • Pi

    inorganic phosphate

  •  
  • RCR

    respiratory control ratio

  •  
  • SED

    sedentary

  •  
  • VDAC

    voltage-dependent anion channel

  •  
  • Vmax

    maximal rate of respiration

Author Contribution

H.L.P. and G.P.H conceptualized the study rationale and designed experiments, whereas H.L.P conducted all experiments. Both authors interpreted data and wrote the manuscript.

Funding

This work was funded by the Natural Sciences and Engineering Research Council of Canada (NSERC) (G.P.H.) [NSERC grant number for G.P.H: 400362], and infrastructure was purchased with assistance from the Canadian Foundation for Innovation/Ontario Research Fund. H.L.P. is supported by an NSERC graduate scholarship.

Competing Interests

The Authors declare that there are no competing interests associated with the manuscript.

References

References
1
Spriet
,
L.L.
(
2014
)
New insights into the interaction of carbohydrate and fat metabolism during exercise
.
Sports Med.
44
,
S87
S96
2
Achten
,
J.
,
Gleeson
,
M.
and
Jeukendrup
,
A.E.
(
2002
)
Determination of the exercise intensity that elicits maximal fat oxidation
.
Med. Sci. Sports Exerc.
34
,
92
97
3
Holloway
,
G.P.
(
2017
)
Nutrition and training influences on the regulation of mitochondrial adenosine diphosphate sensitivity and bioenergetics
.
Sports. Med.
47
,
13
21
4
Perry
,
C.G.R.
,
Kane
,
D.A.
,
Herbst
,
E.A.F.
,
Mukai
,
K.
,
Lark
,
D.S.
,
Wright
,
D.C.
et al.  (
2012
)
Mitochondrial creatine kinase activity and phosphate shuttling are acutely regulated by exercise in human skeletal muscle
.
J. Physiol.
590
,
5475
5486
5
Miotto
,
P.M.
and
Holloway
,
G.P.
(
2016
)
In the absence of phosphate shuttling, exercise reveals the in vivo importance of creatine-independent mitochondrial ADP transport
.
Biochem. J.
473
,
2831
2843
6
Miotto
,
P.M.
,
Steinberg
,
G.R.
and
Holloway
,
G.P.
(
2017
)
Controlling skeletal muscle CPT-I malonyl-CoA sensitivity: the importance of AMPK-independent regulation of intermediate filaments during exercise
.
Biochem. J.
474
,
557
569
7
Howlett
,
R.A.
,
Parolin
,
M.L.
,
Dyck
,
D.J.
,
Hultman
,
E.
,
Jones
,
N.L.
,
Heigenhauser
,
G.J.F.
et al.  (
1998
)
Regulation of skeletal muscle glycogen phosphorylase and PDH at varying exercise power outputs
.
Am. J. Physiol. Regul. Integr. Comp. Physiol.
275
,
R418
R425
8
Phillips
,
S.M.
,
Green
,
H.
,
Tarnopolsky
,
M.
,
Heigenhauser
,
G.
and
Grant
,
S.
(
1996
)
Progressive effect of endurance training on metabolic adaptations in working skeletal muscle
.
Am. J. Physiol. Endocrinol. Metab.
270
,
E265
E272
9
Holian
,
A.
,
Owen
,
C.S.
and
Wilson
,
D.F.
(
1977
)
Control of respiration in isolated mitochondria: quantitative evaluation of the dependence of respiratory rates on [ATP], [ADP], and [Pi]
.
Arch. Biochem. Biophys.
181
,
164
171
10
Chance
,
B.
and
Williams
,
G.R.
(
1955
)
Respiratory enzymes in oxidative phosphorylation: kinetics of oxygen utilization
.
J. Biol. Chem.
217
,
383
394
PMID:
[PubMed]
11
Chance
,
B.
,
Ito
,
T.
,
Maitra
,
P.K.
and
Oshino
,
R.
(
1963
)
Control of endogenous adenosine triphosphatase activity of pigeon heart mitochondria by energy-linked reduction of diphosphopyridine nucleotide
.
J. Biol. Chem.
238
,
1516
1519
PMID:
[PubMed]
12
Scheibye-Knudsen
,
M.
and
Quistorff
,
B.
(
2009
)
Regulation of mitochondrial respiration by inorganic phosphate; comparing permeabilized muscle and isolated mitochondria prepared from type-1 and type-2 rat skeletal muscle
.
Eur. J. Appl. Physiol.
105
,
279
287
13
Tanaka
,
A.
,
Chance
,
B.
and
Quistorff
,
B.
(
1989
)
A possible role of inorganic phosphate as a regulator of oxidative phosphorylation in combined urea synthesis and gluconeogenesis in perfused rat liver
.
J. Biol. Chem.
264
,
10034
10040
PMID:
[PubMed]
14
Perry
,
C.G.
,
Kane
,
D.A.
,
Lin
,
C.-T.
,
Kozy
,
R.
,
Cathey
,
B.L.
,
Lark
,
D.S.
, et al.  (
2011
)
Inhibiting myosin-ATPase reveals a dynamic range of mitochondrial respiratory control in skeletal muscle
.
Biochem. J.
437
,
215
222
15
Peters
,
S.J.
and
Spriet
,
L.L.
(
1995
)
Skeletal muscle phosphofructokinase activity examined under physiological conditions in vitro
.
J. Appl. Physiol.
78
,
1853
1858
16
Stephens
,
F.B.
(
2018
)
Does skeletal muscle carnitine availability influence fuel selection during exercise?
Proc. Nutr. Soc.
77
,
11
19
17
Wall
,
B.T.
,
Stephens
,
F.B.
,
van Loon
,
L.J.
,
Constantin-Teodosiu
,
D.
,
Macdonald
,
I.A.
and
Greenhaff
,
P.L.
(
2013
)
Reduced fat oxidation during high intensity, submaximal exercise: is the availability of carnitine important?
Eur. J. Sport Sci.
13
,
191
199
18
Stephens
,
F.B.
,
Constantin-Teodosiu
,
D.
and
Greenhaff
,
P.L.
(
2007
)
New insights concerning the role of carnitine in the regulation of fuel metabolism in skeletal muscle
.
J. Physiol.
581
,
431
444
19
McGarry
,
J.D.
,
Mills
,
S.E.
,
Long
,
C.S.
and
Foster
,
D.W.
(
1983
)
Observations on the affinity for carnitine, and malonyl-CoA sensitivity, of carnitine palmitoyltransferase I in animal and human tissues
.
Biochem. J.
214
,
21
28
20
Long
,
C.S.
,
Haller
,
R.G.
,
Foster
,
D.W.
and
McGarry
,
J.D.
(
1982
)
Kinetics of carnitine-dependent fatty acid oxidation: implications for human carnitine deficiency
.
Neurology
32
,
663
666
21
Kuznetsov
,
A.V.
,
Veksler
,
V.
,
Gellerich
,
F.N.
,
Saks
,
V.
,
Margreiter
,
R.
and
Kunz
,
W.S.
(
2008
)
Analysis of mitochondrial function in situ in permeabilized muscle fibers, tissues, and cells
.
Nat. Protoc.
3
,
965
976
22
Holloway
,
G.P.
,
Bezaire
,
V.
,
Heigenhauser
,
G.J.F.
,
Tandon
,
N.N.
,
Glatz
,
J.F.C.
,
Luiken
,
J.J.F.P.
et al.  (
2006
)
Mitochondrial long chain fatty acid oxidation, fatty acid translocase/CD36 content and carnitine palmitoyltransferase I activity in human skeletal muscle during aerobic exercise
.
J. Physiol.
571
,
201
210
23
Watt
,
M.J.
,
Heigenhauser
,
G.J.F.
,
O'Neill
,
M.
and
Spriet
,
L.L.
(
2003
)
Hormone-sensitive lipase activity and fatty acyl-CoA content in human skeletal muscle during prolonged exercise
.
J. Appl. Physiol.
95
,
314
321
24
van Loon
,
L.J.C.
,
Greenhaff
,
P.L.
,
Constantin-Teodosiu
,
D.
,
Saris
,
W.H.M.
and
Wagenmakers
,
A.J.M.
(
2001
)
The effects of increasing exercise intensity on muscle fuel utilisation in humans
.
J. Physiol.
536
,
295
304
25
Primassin
,
S.
,
ter Veld
,
F.
,
Mayatepek
,
E.
and
Spiekerkoetter
,
U.
(
2008
)
Carnitine supplementation induces acylcarnitine production in tissues of very long-chain acyl-CoA dehydrogenase-deficient mice, without replenishing low free carnitine
.
Pediatr. Res.
63
,
632
637
26
Larsen
,
S.
,
Dohlmann
,
T.
,
Hindso
,
M.
,
Dela
,
F.
and
Helge
,
J.
(
2013
) High intensity training decreases mitochondrial ADP sensitivity in human adipose and skeletal muscle tissue. In
Comparative Mitochondrial Physiology Network 18.08
(
Laner
,
V.
,
Bidaurratzaga
,
E.
and
Gnaiger
,
E.
, eds), pp.
15
16
,
Obergurgl, Tyrol, Austria
27
Ydfors
,
M.
,
Hughes
,
M.C.
,
Laham
,
R.
,
Schlattner
,
U.
,
Norrbom
,
J.
and
Perry
,
C.G.R.
(
2016
)
Modelling in vivo creatine/phosphocreatine in vitro reveals divergent adaptations in human muscle mitochondrial respiratory control by ADP after acute and chronic exercise
.
J. Physiol.
594
,
3127
3140
28
Zammit
,
V.A.
,
Fraser
,
F.
and
Orstorphine
,
C.G.
(
1997
)
Regulation of mitochondrial outer-membrane carnitine palmitoyltransferase (CPT I): role of membrane topology
.
Adv. Enzyme Regul.
37
,
295
317
29
McGarry
,
J.D.
and
Brown
,
N.F.
(
1997
)
The mitochondrial carnitine palmitoyltransferase system: from concept to molecular analysis
.
Eur. J. Biochem.
244
,
1
14
30
Odland
,
L.M.
,
Howlett
,
R.A.
,
Heigenhauser
,
G.J.
,
Hultman
,
E.
and
Spriet
,
L.L.
(
1998
)
Skeletal muscle malonyl-CoA content at the onset of exercise at varying power outputs in humans
.
Am. J. Physiol.
274
,
E1080
E1085
PMID:
[PubMed]
31
ter Veld
,
F.
,
Primassin
,
S.
,
Hoffmann
,
L.
,
Mayatepek
,
E.
and
Spiekerkoetter
,
U.
(
2009
)
Corresponding increase in long-chain acyl-CoA and acylcarnitine after exercise in muscle from VLCAD mice
.
J. Lipid Res.
50
,
1556
1562
32
Smith
,
B.K.
,
Perry
,
C.G.
,
Koves
,
T.R.
,
Wright
,
D.C.
,
Smith
,
J.C.
,
Neufer
,
D.P.
et al.  (
2012
)
Identification of a novel malonyl-CoA IC50 for CPT-I: implications for predicting in vivo fatty acid oxidation rates
.
Biochem. J.
448
,
13
20
33
Constantin-Teodosiu
,
D.
,
Carlin
,
J.
,
Cederblad
,
G.
,
Harris
,
R.
and
Hultman
,
E.
(
1991
)
Acetyl group accumulation and pyruvate dehydrogenase activity in human muscle during incremental exercise
.
Acta. Physiol. Scand.
143
,
367
372
34
Sahlin
,
K.
(
1990
)
Muscle carnitine metabolism during incremental dynamic exercise in humans
.
Acta. Physiol. Scand.
138
,
259
262
35
Starritt
,
E.C.
,
Howlett
,
R.A.
,
Heigenhauser
,
G.J.F.
and
Spriet
,
L.L.
(
2000
)
Sensitivity of CPT I to malonyl-CoA in trained and untrained human skeletal muscle
.
Am. J. Physiol. Endocrinol. Metab.
278
,
E462
E468
36
Bezaire
,
V.
,
Heigenhauser
,
G.J.F.
and
Spriet
,
L.L.
(
2013
)
Regulation of CPT I activity in intermyofibrillar and subsarcolemmal mitochondria from human and rat skeletal muscle
.
Am. J. Physiol. Endocrinol. Metab.
286
,
E85
E91
37
Sahlin
,
K.
(
2009
)
Control of lipid oxidation at the mitochondrial level
.
Appl. Physiol. Nutr. Metab.
388
,
382
388
38
Dyck
,
D.J.
,
Putman
,
C.T.
,
Heigenhauser
,
G.
,
Hultman
,
E.
and
Spriet
,
L.
(
1993
)
Regulation of fat-carbohydrate interaction muscle during intense aerobic cycling
.
Am. J. Physiol. Endocrinol. Metab.
265
,
E825
E859
39
Romijn
,
J.A.
,
Coyle
,
E.F.
,
Sidossis
,
L.S.
,
Zhang
,
X.
and
Wolfe
,
R.R.
(
1995
)
Relationship between fatty acid delivery and oxidation during strenuous exercise
.
J. Appl. Physiol.
79
,
1939
1945
40
Hiatt
,
W.R.
,
Regensteiner
,
J.G.
,
Wolfel
,
E.E.
,
Ruff
,
L.
and
Brass
,
E.P.
(
1989
)
Carnitine and acylcarnitine metabolism during exercise in humans: dependence on skeletal muscle metabolic state
.
J. Clin. Invest.
84
,
1167
1173