We measured the content of long-chain fatty acids (LCFA) in biopsies obtained from the vastus lateralis muscle in humans at rest and after different exercise intensities. Nine volunteers exercised at 65% of maximal oxygen uptake (V˙o 2 max) for 40 min and at 90% of V˙o 2 max for another 15 min on a Krogh bicycle ergometer. LCFA measured in muscle tissue averaged 76 ± 5 nmol/g wet wt at rest, decreased significantly after exercise at 65% V˙o 2 max to 48 ± 4 nmol/g wet wt, and increased to 68 ± 5 nmol/g wet wt (P < 0.05) after high-intensity exercise. The calculated myocyte LCFA content at rest amounted to 69 ± 5 nmol/g wet wt, decreased by 43% (P < 0.05) after exercise at 65% ofV˙o 2 max, and subsequently increased by 54% after exercise at 90% ofV˙o 2 max(P < 0.05) compared with the values obtained at the lower workload. The blood plasma LCFA concentration during the low-intensity exercise (366 ± 23 nmol/ml) was similar to the values obtained at rest (372 ± 31 nmol/ml) but decreased significantly during the high-intensity workload (249 ± 49 nmol/ml). From these data it is proposed that 1) in human skeletal muscle, metabolism rather than cellular availability of LCFA governs the rate of LCFA utilization at rest and during exercise, and 2) consequently reduction in muscle LCFA oxidation during high-intensity exercise (e.g., 90%V˙o 2 max) is due primarily to a decrease in mitochondrial LCFA oxidation rate rather than an insufficient cellular availability of LCFA.
- skeletal muscle
during exercise at increasing intensities there is a shift in fuel utilization toward an increased combustion of carbohydrates at the expense of fat (for review see Ref. 37). With regard to the rate of long-chain fatty acid (LCFA) utilization in skeletal muscle during exercise, there are several regulatory steps. In addition to limitations or in the rate of lipolysis, muscular constraints on LCFA utilization could include transport through the endothelium lining of the microvascular compartment, across the interstitial space; transsarcolemmal transport; transport of the fatty acids in the cytosol of the muscle cell; and finally uptake and oxidation by the mitochondria. However, it is incompletely understood which of these steps is rate determining for muscular fatty acid utilization during exercise. Theoretically the main regulatory step may reside either in transport from the vascular to cytosolic compartment or in mitochondrial oxidation.
The first suggestion for the existence of a facilitated transport system for fatty acids in skeletal muscle originates from studies performed in the perfused rat hindlimb (34). Recently evidence has been presented that certain proteins, associated with the plasma membranes of both endothelial and muscle cells, contribute to the facilitated fatty acid uptake (1, 2, 5, 6a, 9, 24, 25, 28, 30). Thus, with the experimental evidence for a protein-mediated transport of fatty acids across the plasma membrane, the plasma membrane could be considered a subject of regulation for fatty acid utilization. Support for this notion is that fatty acid uptake and oxidation in human skeletal muscle during submaximal exercise did not increase linearly as a function of plasma fatty acid concentration (12, 35) but rather displayed saturation kinetics.
The oxidation of fatty acid by the muscle cell mitochondria is another potential site of control of overall muscular fatty acid utilization. The relative balance between supply and metabolism of fatty acids will be reflected in the cellular content of LCFA. Accordingly, we hypothesized that if the rate-determining step in LCFA utilization primarily lies in the supply of fatty acids originating either from extracellular sources requiring transendothelial and transsarcolemmal transport of fatty acids or possibly from the breakdown of intracellularly located triacylglycerols, a low and fairly constant cellular content of LCFA would be expected at rest and during exercise both at moderate and intense workloads. Alternatively, if the rate of muscular LCFA utilization is primarily governed by mitochondrial oxidation rate rather than by intracellular LCFA supply, a significant cellular content of LCFA would be expected, and its magnitude would depend on the metabolic rate of the muscle. Thus, when the metabolic rate of the skeletal muscle cell increases from resting conditions to a moderate exercise intensity, mitochondrial LCFA oxidation increases, which should result in a decline of cellular LCFA content, whereas during heavy exercise, cellular content of LCFA should rise as a consequence of the decline in mitochondrial LCFA utilization.
Thus, to evaluate the relative importance of the LCFA supply on the one hand and the mitochondrial LCFA oxidation on the other as determinants of the overall muscular LCFA utilization, in the present study we analyzed the blood plasma concentration and skeletal muscle content of LCFA in humans at rest and after two different levels of physical exercise.
MATERIALS AND METHODS
Nine healthy male volunteers participated in the study. The volunteers were 20–35 yr of age, body weight averaged 78 kg (70–100 kg), height 182 cm (175–188 cm), and maximal oxygen uptake (V˙o 2 max) 53 ml ⋅ min−1 ⋅ kg body wt−1 (50–55 ml ⋅ min−1 ⋅ kg body wt−1). All volunteers were fully informed about the nature of the study and the possible risks associated with it before they volunteered to participate, and written consent was given. The study was approved by the Copenhagen Ethics Committee and conforms with the code of ethics of the World Medical Association (Declaration of Helsinki). Volunteers were covered by state medical insurance and by the same insurance that covers hospitalized patients in case of complications.
The volunteers reported to the laboratory either by bus or car on the morning after an overnight fast. After 30 min of rest in the supine position resting blood was drawn from a catheter inserted in the antecubital vein. To obtain adequate muscle tissue for the lipid analysis (200 mg), two needle biopsies of the vastus lateralis muscle were taken with suction from the same skin incision under local anesthesia with lidocaine, with the needle pointing in two different directions. Exercise was then initiated on a Krogh bicycle ergometer. Exercise was performed at 65% ofV˙o 2 max for 40 min. At the end of the exercise period expired air was sampled in Douglas bags for measurement of V˙o 2 and CO2 excretion. From these measurements the respiratory exchange ratio (RER) was calculated. Immediately before the end of the exercise period blood was drawn from the antecubital vein, and volunteers were then quickly placed on a bed in the supine position. As rapidly as possible (within ∼5–15 s after termination of exercise) another pair of biopsies was obtained from the vastus lateralis muscle through another skin incision spaced 4–5 cm apart from the previous incision. Immediately thereafter, volunteers returned to the Krogh bicycle ergometer, continuing exercise for 15 min at 90% ofV˙o 2 max. Again blood was drawn and expired air was collected at the end of the exercise period, and at the termination of exercise another two biopsies were obtained from the vastus lateralis muscle, using the same procedure as after the first exercise period.
The needle biopsy samples were directly plunged into liquid nitrogen and stored at −80°C until analysis. All equipment used for the biopsies was cleaned in alcohol to avoid any contamination of lipids in the samples. Blood was sampled in dry syringes, immediately transferred to plastic test tubes containing EGTA, and centrifuged at 12,000g for 1 min. Plasma was immediately frozen at −20°C until further analysis.
LCFA in tissue and blood plasma.
The concentration of LCFA in blood plasma was analyzed as follows. To 300 μl of frozen blood plasma 2 ml ice-cold methanol were added, and then the mixture was subjected to gentle shaking at ambient temperature to thaw. Thereafter 4.0 ml of chloroform were added that included the internal recovery standard heptadecanoic acid (8 nmol). The methanol and chloroform contained 0.01% butylated hydroxytoluene to prevent autoxidation of unsaturated fatty-acyl moieties. One-dimensional thin-layer chromatography (TLC) on plates coated with silica gel 60 (Merck, Darmstadt, Germany) was applied to separate fatty acids (LCFA) from complexed lipids such as phospholipids and triacylglycerols. A solution of chloroform-methanol-water-acetic acid (10:10:1:1 vol/vol/vol/vol) was used to develop the plate until the lipid front had reached 1 cm above the application site of the lipid spots. Subsequently the plate was further developed with a solution of hexane:diethyl ether:acetic acid (24:5:0.3 vol/vol/vol). The lipid spots were made visible with rhodamine 6G and fluorescein in methanol (0.01%). The spots corresponding with LCFA were then scraped from the plate and methylated with 7% boron-trifluoride-methanol by the method of Morrison and Smith (16). A standard solution of fatty acid methyl esters was used to quantify the amount of LCFA in the blood plasma extract. The LCFA were methylated at 20°C for 15 min. Methylation was stopped through addition of an equal volume of double-distilled water. The methyl esters formed were extracted from the methylating mixture with pentane. The pentane was subsequently evaporated under a stream of nitrogen at 30°C followed by dissolution of methyl esters in 2,2,4-trimethylpentane containing 9 nmol methyl ester of pentadecanoic acid as recovery standard, to quantify the methyl esters by capillary gas chromatography (38). To assess the LCFA content in frozen tissue biopsies, aliquots of ∼200 mg wet wt of frozen skeletal muscle tissue (tissue obtained from both biopsies) were pulverized with the use of a previously cooled (liquid nitrogen) aluminum mortar and stainless steel pestle. The powdered tissue was extracted with methanol and chloroform as indicated above for the extraction of lipids from blood plasma. The analysis of LCFA in tissue with the use of TLC and capillary gas chromatography was also identical to that described above. The overall recovery of LCFA ranged from 90 to 105% after correction for the losses during the extraction and assay procedures with the use of the internal and recovery standards.
Blood plasma lactate was analyzed with a YSI 2700 select analyzer (Yellow Springs, OH).
Pulmonary oxygen uptake and RER during exercise were determined by collection of expired air in Douglas bags. The volume of air was measured in a Collins bell spirometer (Tissot principle), and the fractions of oxygen and carbon dioxide were determined with paramagnetic (Taylor Servomex, Crowborough, Sussex, UK) and infrared (LB-2; Beckman Instruments, Palo Alto, CA) systems, respectively. Two gas samples with known compositions were used to calibrate both systems regularly. Heart rate was recorded with a PE-3000 sports tester (Polar Electro, Helsinki, Finland).
Estimation of cellular LCFA content.
The LCFA content was measured in muscle biopsies (as described above). This includes LCFA present in blood (capillaries) and interstitial fluid trapped in the biopsy. The cellular content of LCFA (LCFAC) was estimated as follows where LCFAM and LCFAEC indicate the content of LCFA in whole muscle and in the extracellular space, respectively, and LCFACapand LCFAInt indicate the content of LCFA in the capillaries and the interstitial space, respectively where Hct is the hematocrit of the blood, VC is capillary volume, and [LCFA]P is plasma LCFA concentration, and The value 0.25 in the above calculation of LCFAInt is based on the assumption that the concentration of LCFA in the interstitial space is one-fourth of the value in the plasma (22). Furthermore, the capillary blood volume of the muscle is assumed to be 1.4 vol% (23). Oxygen saturation in forearm venous blood during exercise was found to be ∼90%. This high-oxygen saturation indicates major shunting of arterial blood through arteriovenous anastomoses to the veins. Accordingly, the LCFA concentration in forearm venous blood is most likely very close to the concentration in arterial blood. Consequently it was assumed that the forearm venous concentration of LCFA resembles the average concentration of LCFA in capillary blood. This assumption is supported by the fact that arteriovenous differences of LCFA across the leg are rather small both at rest and during exercise (on the order of 5–10% of arterial concentrations) (12). Furthermore, the total extracellular water compartment (capillary and interstitial volume) at rest and after exercise was calculated in accordance with the findings by Sjøgaard and Saltin (27), who measured a mean extracellular water content of 33 ml/100 g dry wt at rest in the vastus lateralis muscle, increasing to 60 ml/100 g dry wt after intense exercise. These values were transformed into 7.9, 10.6, and 13.2 ml/100 g wet wt, at rest, 65%, and 90% ofV˙o 2 max, respectively, assuming the water content of the biopsies to be 76% at rest and 78% during exercise (B. Kiens, unpublished observations).
Results are given as means ± SE. For each variable measured, a one-way ANOVA with repeated measures for the time factor was performed to test for changes during exercise. Differences between time points were detected with a pairwise multiple-comparison procedure (Student-Newman-Keuls method). In all cases, an α of 0.05 was used as level of significance.
At rest LCFA content in vastus lateralis muscle tissue amounted to 76 ± 5 nmol/g wet wt (Fig. 1). Exercise at an intensity corresponding to 65% ofV˙o 2 max resulted in a significant decrease in muscle LCFA content to 48 ± 4 nmol/g wet wt (Fig. 1). After exercise at 90% ofV˙o 2 max the content of muscle LCFA was subsequently significantly increased to 68 ± 5 nmol/g wet wt (Fig. 1). Calculation of the cellular LCFA content, i.e., after corrections for the contribution of LCFA present in the capillary and interstitial compartment, revealed a resting cellular content of 69 ± 5 nmol/g wet wt (Fig. 1). After exercise at 65% ofV˙o 2 max the calculated cellular LCFA content had decreased by 43% (P < 0.05) to 39 ± 3 nmol/g wet wt but was subsequently increased significantly by 54% to 60 ± 12 nmol/g wet wt after the heavy exercise bout compared with the values obtained after exercise at 65% ofV˙o 2 max (Fig. 1). Plasma concentrations of LCFA averaged 372 ± 31 nmol/ml at rest and remained similar after exercise at 65% ofV˙o 2 max (366 ± 23 nmol/ml). After exercise at 90% ofV˙o 2 max the concentration of plasma LCFA was significantly decreased to 249 ± 49 nmol/ml (Fig.2). From the blood plasma and estimated cell content data the mean blood-to-muscle cell ratios for LCFA at rest and at 65% and 90% ofV˙o 2 max can be estimated to be 5.4, 9.4, and 4.1, respectively.
The RER averaged 0.87 ± 0.02 at the end of exercise at 65% ofV˙o 2 max and increased significantly to 0.96 ± 0.02 when exercise was performed at 90% ofV˙o 2 max. Blood lactate concentrations averaged 0.67 ± 0.05 mmol/l at rest, remained at the same level after exercise at 65% ofV˙o 2 max (0.71 ± 0.06 mmol/l), but increased significantly to 5.0 ± 0.6 after exercise at 90% of V˙o 2 max. Oxygen uptake at 65% of V˙o 2 maxamounted to 2.63 ± 0.09 l/min and increased significantly to 3.62 ± 0.31 l/min at 90% ofV˙o 2 max.
In the present investigation we have used a novel approach to study regulation of fatty acid utilization in skeletal muscle at rest and during exercise. We have directly measured the total content of LCFA in biopsies from the vastus lateralis muscle in humans at rest and after different exercise intensities, and from these values we have also estimated the cellular content of LCFA. The findings demonstrate a substantial decrease in total and cellular LCFA content in muscle tissue after submaximal exercise (65% ofV˙o 2 max) compared with resting values, whereas after heavy exercise (90%V˙o 2 max) a significant increase in the value of both total and cellular LCFA content was seen compared with the values obtained after moderate submaximal exercise (65% of V˙o 2 max) (Fig. 1). From 65 to 90% ofV˙o 2 max the average RER value expectedly increased from 0.87 to 0.96. Even though RER values at 90% ofV˙o 2 max cannot be taken as representative for respiratory quotient values in the muscle, the RER value of 0.96 indicates that lipids only contribute little to the overall substrate oxidation at the highest exercise intensity. Our data strongly suggest that the decrease in lipid utilization during heavy exercise is not due to a decrease in cellular LCFA availability but rather to a decreased mitochondrial oxidation despite a high cellular fatty acid content.
The estimation of the cellular fatty acid content from values obtained in a muscle biopsy is based on several assumptions. For instance, the amount of blood in the vascular space is estimated from values of the capillary volume obtained in other studies of human muscle (23), and the concentration of fatty acids in the interstitial space is estimated using ratios between plasma and interstitial values obtained in the dog heart (22). The size of the extracellular compartment is based on measurements in humans at rest and during exercise done previously (27). Still, from the data in Fig. 1 it can be seen that the amount of fatty acids estimated to reside in the extracellular compartment is relatively small in comparison with the total amount measured in the tissue biopsy. Because of the small contribution of extracellular LCFA to the total tissue LCFA, the uncertainties in the above-mentioned assumptions only exert a limited effect on the amount of calculated cellular LCFA content. The fact that both the total directly measured content and the calculated cellular content show the same pattern of variation with exercise intensity strongly suggests that the changes in the estimated cellular content with exercise intensity are physiologically significant. Thus, after correcting for the fatty acids present in the capillaries and the interstitial compartment, it was found that LCFA content in the myocytes at rest amounted to 69 ± 5 nmol/g wet wt. Some of the LCFA measured in muscle biopsies may in fact stem from fat cells within the biopsy. However, the fact that we in the present study observed a decrease in the cellular LCFA content at moderate exercise strongly suggests that the fatty acids are mainly from muscle cells. If fat cells were the main source of LCFA in muscle biopsies, one would expect an increase during moderate-intensity exercise due to increased lipolysis going from rest to exercise. Our values are in line with the findings by Masoro (14), who reported a fatty acid level of 77 nmol/g wet wt in the rat gastrocnemius muscle, and are only slightly higher than previously reported in dog biceps femoris muscle (38). Our values are considerably lower than previously reported by others (3, 11, 33). It should be kept in mind that cellular values several times higher than the plasma concentration as found by others (3, 11, 33) are not compatible with the direction of flux of fatty acids from plasma to muscle and are therefore most likely not correct.
Although the subject of discussion for many years, which step(s) are rate determining in fatty acid oxidation during exercise remains unclear. The transport of blood-borne fatty acids from the vascular compartment to muscle mitochondria comprises several steps that could govern the overall rate of muscle fatty acid utilization. Recent studies have shown that LCFA uptake by different cell types exhibits all the kinetic properties corresponding with facilitated transport; hence, a concept of fatty acid trapping and transporting proteins at or in the plasma membrane for transmembrane fatty acid transport has been proposed (1, 2, 25, 28, 30, 31, 34). Three different proteins (at least), associated with the plasma membrane, have been identified as putative fatty acid receptors and/or transporters. These proteins are plasmalemmal fatty acid-binding protein (FABPpm; 40–43 kDa) (4, 5, 8, 13), fatty acid-transport protein (FATP; 63 kDa) (24), and fatty acid translocase (FAT or CD36; 88 kDa) (1), all found in skeletal muscle. The albumin-binding protein (ABP; 18 kDa), located at the luminal site of the capillary endothelium, may also play a role in the transport of fatty acids from the blood plasma to the cells (10, 17). These proteins are suggested to contribute to facilitated LCFA uptake in particular cell types (1, 2, 25, 28, 30, 34) and also recently in skeletal muscle plasma membranes (6a, 9). Thus these plasma-membrane associated proteins could be involved in determining the rate of overall fatty acid oxidation during exercise. Support for this notion is the findings of saturation in the uptake of fatty acids in skeletal muscle during exercise at relatively high plasma fatty acid concentrations (34), which was more pronounced in untrained than in trained muscle (12). Furthermore, training induced an increase in the expression of FABPpm in human skeletal muscle (13), a situation in which also the uptake and oxidation of LCFA are increased (12, 35). Finally, studies in isolated sarcolemmal vesicles from rat skeletal muscle also show saturation kinetics of palmitate transport (6a, 9). However, even though the transport of LCFA across the sarcolemma may be facilitated by transport proteins and does show saturation kinetics, this in itself does not prove that transport is rate determining for utilization at any particular physiological situation. In fact, the present findings support the notion that the rate of mitochondrial fatty acid oxidation has a dominant influence on the overall rate of LCFA utilization (i.e., transport and metabolic conversion) in skeletal muscle.
The findings of a significant cellular content of LCFA in myocytes at rest is in contrast to, e.g., glucose, for which intracellular concentrations in skeletal muscle are essentially zero (19). This has been taken as support for the notion that membrane transport, not metabolism, is rate limiting for glucose utilization. By analogy, the detection of a significant content of LCFA in skeletal muscle cells suggests that LCFA metabolism, not LCFA supply, is rate determining for LCFA utilization at rest. Furthermore, we propose that the increase in mitochondrial fatty acid oxidation going from rest to moderate exercise (65% V˙o 2 max) causes a decline in cellular LCFA content (Fig. 1), which in turn enhances the gradient of LCFA from blood plasma to skeletal muscle cells inasmuch as LCFA concentrations in plasma did not change from rest to moderate exercise. This steeper gradient, in combination with a possible increase in surface membrane LCFA transport capacity, favors the net flux of fatty acids from the vascular compartment to the muscular site of metabolic conversion. A further increase in the intensity of exercise (90% V˙o 2 max) is associated with a shift in fuel selection toward an increased combustion of carbohydrates. This decline in mitochondrial LCFA oxidation apparently results in an increase in the cellular level of LCFA (Fig. 1), implicating a less profound sink for LCFA in the exercising muscle cells. As a consequence the gradient of LCFA from the capillary compartment to the cytosol of the muscle cell becomes less steep, and net transport either by passive or facilitated diffusion declines. The decrease in blood plasma-tissue gradient of LCFA at 90%V˙o 2 max is even amplified by a decrease in the concentration of circulating LCFA. This line of reasoning underscores the notion that during acute changes in the intensity of exercise the rate of net transport of LCFA from the capillary compartment across the endothelium and sarcolemma to a large extent follows the rate of mitochondrial LCFA oxidation. This proposal is in line with recent findings by Sidossis and co-workers (26), who, making use of stable isotopes, reported that the proportion of extracted LCFA that is oxidized, as well as the total fat oxidation, was decreased when exercise intensity increases from 40 to 80% ofV˙o 2 max. Furthermore, by increasing the plasma LCFA concentration by infusion of Intralipid and heparin to 1–2 mmol/l during exercise of 85% ofV˙o 2 max, only a minor increase in fat oxidation was observed compared with the control experiment, in which the LCFA concentration in plasma was approximately 0.2–0.3 mmol/l (21). During Intralipid and heparin infusion at 85% of V˙o 2 max the rate of LCFA oxidation was also still markedly lower than the rate obtained during exercise at 65% ofV˙o 2 max (20). These findings give further support for the notion that factors other than LCFA availability are the main regulators of lipid metabolism during high-intensity exercise.
These considerations imply that at various intensities of exercise factors other than supply of LCFA are involved in determining the rate of LCFA oxidation in muscle mitochondria. Because the elucidation of possible mechanisms underlying the regulation of mitochondrial fatty acid oxidation was not the main goal of the present study, we may only speculate about the nature of these regulatory factors. A likely candidate is malonyl-CoA regulating the activity of carnitine acyltransferase I (CPT I) (7, 32), the enzyme responsible for the conversion of fatty acyl-CoA into fatty acylcarnitine before transfer of the acyl moieties to the intramitochondrial β-oxidation machinery. Because malonyl-CoA inhibits CPT I (15), it has been hypothetized that malonyl-CoA is also involved in the overall regulation of fatty acid oxidation (7, 15). Recent findings in humans, however, did not show any significant changes in malonyl-CoA in muscle during intense exercise (29). Moreover, malonyl-CoA content in skeletal muscle was decreased with increasing exercise intensities in the red quadriceps muscle in rats, whereas no changes were seen in the white quadriceps muscle (18). Thus these data do not support the hypothesis that the decrease in fatty acid utilization at high exercise intensities is due to an increased concentration of malonyl-CoA, inhibiting CPT I. Therefore other mechanisms have to be responsible for the decreased fatty acid utilization with increasing exercise intensities.
An alternative explanation is inhibition of muscle LCFA oxidation by ambiently increased lactate levels, as outlined by Van der Vusse and colleagues (36) in cardiac muscle cells. Enhanced concentrations of lactate, as observed in the present study in the volunteers exercising at 90% V˙o 2 max, may inhibit CPT I activity, which was suggested by Bielefeld and co-workers (6) to be an important regulating mechanism in cardiac tissue. It remains to be established whether this relationship between enhanced lactate level and decline in fatty acid oxidation also holds for skeletal muscle cells.
In summary, the main findings of this study indicate the presence of significant amounts of LCFA in human skeletal muscle both at rest and during exercise. At moderate exercise intensities cellular LCFA content decreases in the face of enhanced fatty acid oxidation rates. However, during intense exercise, when fat oxidation is low, an increase in cellular LCFA content was observed. It is proposed from the present data that at rest, as well as during exercise, the rate of muscle LCFA utilization is determined primarily by the rate of mitochondrial LCFA oxidation rather than the cellular availability of LCFA.
We thank professor E. A. Richter for performing the muscle biopsy procedures and I. B. Nielsen for technical assistance.
Address for reprint requests: B. Kiens, Copenhagen Muscle Research Centre, August Krogh Institute, Univ. of Copenhagen, DK-2100 Copenhagen, Denmark.
This study was supported by the Danish National Research Foundation (504-14).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.
- Copyright © 1999 the American Physiological Society