HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (33) Citing Articles via Google Scholar Google Scholar Articles by Kiens, B. Search for Related Content PubMed PubMed Citation Articles by Kiens, B.

Skeletal Muscle Lipid Metabolism in Exercise and Insulin Resistance

Copenhagen Muscle Research Centre, Department of Human Physiology, Institute of Exercise and Sport Sciences, University of Copenhagen, Copenhagen, Denmark

ABSTRACT

I. PERSPECTIVE

II. LONG-CHAIN FATTY ACID UTILIZATION DURING EXERCISE

A. Arterial Long-Chain Fatty Acid Concentration

B. Utilization of Long-Chain Fatty Acids During Exercise

III. REGULATION OF LONG-CHAIN FATTY ACID UTILIZATION IN SKELETAL MUSCLE DURING EXERCISE

A. Supply of Long-Chain Fatty Acids

B. Transport From Circulation to Cytosol

1. FAT/CD36

2. FABPpm

3. FATP1

4. Caveolins

5. Influence of exercise

C. Transport in the Cytosol

D. Metabolism in Mitochondria

1. Regulation of fat oxidation by carnitine

2. Other potential regulators of fat oxidation

IV. CIRCULATING VERY-LOW-DENSITY LIPOPROTEIN-TRIACYLGLYCEROL

A. Skeletal Muscle Lipoprotein Lipase

V. INTRAMYOCELLULAR TRIACYLGLYCEROL

A. Factors Affecting Intramyocellular Triacylglycerol Content

1. Diet

2. Muscle fiber type

3. Gender

4. Physical training

B. Methodological Considerations

1. Biochemical determination

2. Morphological determination

3. 1H-MRS

4. Indirect measurements

C. Intramyocellular Triacylglycerol Utilization During Exercise

1. Studies where the muscle biopsy technique was applied

2. Studies where the 1H-MRS technique was applied

3. Indirect estimation of myocellular triacylglycerol utilization during exercise

D. Triacylglycerol Stored in Adipocytes Associated With Muscle

E. Is Intramyocellular Triacylglycerol Breakdown During Exercise Influenced by Training?

VI. REGULATION OF INTRAMYOCELLULAR TRIACYLGLYCEROL HYDROLYSIS

A. Hormone-Sensitive Lipase Activity During Exercise

B. Molecular Mechanisms

1. Phosphorylation

2. Allosteric regulation

VII. RELATION BETWEEN LIPIDS AND INSULIN RESISTANCE

A. Intramyocellular Triacylglycerols and Insulin Resistance

B. Lipid Intermediates and Insulin Resistance

C. Protein Kinase C

VIII. CONCLUSION AND FUTURE DIRECTIONS

GRANTS

ACKNOWLEDGMENTS

REFERENCES

	ABSTRACT

Top Next References

	I. PERSPECTIVE

Top Previous Next References

 
By the end of the 19th century it was suggested by Chauveau and Kaufmann (37) that glucose was the dominating energy substrate during exercise and that lipids were not directly oxidized but first converted into glucose and glycogen in the liver. In those years it was, however, of debate whether glucose was the essential energy substrate for the oxidative metabolism in muscle as Zuntz (345) argued that a mixture of substrates contributed as energy sources.

The carefully conducted experiments by Krogh and Lindhard resolved this debate (165). Krogh's idea was that the respiratory exchange ratio (RER) measured during identical exercise bouts for 2 h following the ingestion of different diets would indicate the preferential use of substrate for combustion. The main findings of Krogh and Lindhard were 1) that lipids were used as energy substrate and that subjects performed poorly during severe exercise when lipids were the preferential energy fuel, 2) that the preceding diet influenced metabolism during rest and in the postabsorptive state, and 3) that RER values increased with increasing exercise intensities, indicating a greater reliance on carbohydrate as energy fuel.

Subsequently, Christensen and Hansen (40, 41) by measuring oxygen uptake and respiratory quotient further described how diet, training, and exercise intensity and duration affected carbohydrate and lipid utilization.

Today it is well known that lipids, in addition to being necessary as fuel for the organism, have been found to have fundamental roles as messengers and regulators of transcription of genes involved in lipid metabolism. Furthermore, evidence is accumulating that lipids are implicated in the pathogenesis of several common human diseases including the metabolic syndrome, cardiovascular disease, and type 2 diabetes.

Lipids as fuel for energy provision originate from three different sources: albumin-bound long-chain fatty acids (LCFA) in the blood plasma, very-low-density lipoprotein-triacylglycerols (VLDL-TG), fatty acids from triacylglycerol located in the muscle cell (IMTG), and possibly fatty acids liberated from adipose tissue adhering to the muscle cells. This review focuses on the role of lipids in fuel provision in skeletal muscle during exercise and in insulin resistance of skeletal muscle.

	II. LONG-CHAIN FATTY ACID UTILIZATION DURING EXERCISE

Top Previous Next References

 
At the onset of exercise, there is a large increase in uptake and oxidation of LCFA in skeletal muscle. It is also well recognized that prolonged exercise for several hours at a low intensity induces a gradual decrease in the respiratory quotient and hence an enhanced lipid utilization at the expense of carbohydrates as energy fuel. Moreover, when the exercise intensity increases, a shift in fuel selection appears towards an increase in carbohydrate and decrease in fat utilization, and following endurance training a shift towards an enhanced lipid utilization is evident at least when exercise is performed at the same absolute work load in the untrained and trained state. It is interesting to note, however, that this training-induced shift in fuel selection is prevented when a carbohydrate-rich diet is consumed (116, 119). Even though a change in lipid utilization under different conditions is well described, the regulatory mechanisms controlling fatty acid uptake and oxidation are not yet completely known.

A. Arterial Long-Chain Fatty Acid Concentration

The level of the arterial concentration of LCFA during exercise is dependent on the preexercise diet, on the time elapsed since the last meal (73, 197), and on whether carbohydrates are consumed during exercise (194, 200). In general, the higher the fat content of the preexercise diet and the longer after the last meal, the higher the concentration of LCFA in plasma, whereas carbohydrate feeding inhibits the exercise-induced increase in arterial LCFA concentration due to an increase in arterial plasma insulin concentration (194) and consequently a decrease in adipose tissue lipolysis (129). During exercise an initial decline in the LCFA concentration is often observed followed by a slow increase. Therefore, if exercise is prolonged, the arterial plasma concentration of LCFA may increase to values markedly above resting levels, but at the most to resting levels during more intense exercise (3, 30, 35, 78, 79, 162, 239, 300). The initial drop in arterial LCFA concentration during exercise is most likely caused by an imbalance between slow mobilization of fatty acids from adipose tissue and rapidly increased extraction of LCFA by skeletal muscle.

B. Utilization of Long-Chain Fatty Acids During Exercise

Whole body LCFA uptake during exercise can be measured by tracer techniques, infusing either radioactive or stable isotopes. Combining tracer data and indirect calorimetry, findings generally indicate that 55–65% of total whole body fat utilization during moderate-intensity exercise is derived from plasma fatty acids (79, 117, 235, 239, 312). During prolonged submaximal exercise, the contribution of LCFA to energy provision increases with time (3, 314). If palmitate is labeled on the carbon molecules (¹⁴C or ¹³C), it is also possible to measure LCFA oxidation, provided that the bicarbonate pool has been prelabeled. Experiments have been conducted using infusion of ¹³C- or ¹⁴C-labeled acetate to correct for the amount of labeled carbon lost in side reactions in the tricarboxylic acid (TCA) cycle or fixed in the bicarbonate pool following oxidation of labeled LCFA. Different whole body acetate recovery values have appeared in the literature as this value obviously depends on exercise intensity and the location of the labeling on acetate. Accordingly, the whole body acetate recovery values obtained during exercise varied from 66 to 94% at exercise intensities ranging from 23 to 84% of maximal oxygen consumption when 1-¹⁴C-labeled acetate or [1-¹³C]acetate was infused (268, 296). When [1,2-¹³C]acetate was infused during bicycle exercise, whole body recovery ranged from 69 to 100% at exercise intensities from 40 to 75% of maximal oxygen consumption (237, 259, 313). The highest values were measured in the trials with the highest exercise intensity. During knee-extensor exercise during which only 2–3 kg muscle is engaged in the exercise, the whole body recovery of infused [1,2-¹³C]acetate was 85% (309). It is interesting to note that when the label is on the 2-position of acetate only, recovery is markedly less (296) and not representative for either 1-C or [U-¹³C]palmitate due to the higher tendency of the 2-carbon of acetate to participate in exchange reactions in the TCA cycle compared with the 1-carbon (339). When measurements are performed using arteriovenous leg balance methods during exercise, acetate recovery was close to 100% (237, 309).

When applying correction factors obtained by [1,2-¹³C]acetate, the available investigations utilizing infusion of [U-¹³C]palmitate demonstrate that 80–96% of whole body LCFA rate of disappearance (R_d) were oxidized during bicycle exercise at intensities ranging from 40 to 75% of maximal oxygen uptake (235, 312, 314). In one study there was a clear trend towards the highest percentage oxidation values at the highest work loads (312). It was also shown that when preexercise muscle glycogen content is low and fat oxidation during exercise consequently is high, the percentage of whole body LCFA R_d, which is oxidized is higher than when preexercise glycogen content is high, suggesting an influence of glycogen stores on oxidation of the LCFA taken up (237).

When [1-¹³C]palmitate was infused, 70% of the systemic plasma LCFA was oxidized during exercise at 68% of maximal oxygen uptake (117) when a mean acetate correction factor suitable for [1-¹³C]palmitate was applied. In studies in trained and untrained men, 58 and 76% of [1-¹³C]oleate were oxidized during exercise, representing 40 and 80% of peak oxygen uptake, respectively, also using a mean acetate correction factor suitable for [1-¹³C]oleate (269). The lower percentage of plasma LCFA oxidized in these studies obtained with 1-C- than with U-C-labeled tracers is probably due to loss of label inside reactions when using 1-C tracers, and all together, these observations tend to indicate that uniformly labeled tracers are preferable to 1-C-labeled tracers.

Concluding from the available evidence, it is evident that a major part of plasma LCFA taken up during exercise is oxidized in the body, and the fate of the remaining LCFA is likely to be esterification in noncontracting muscles or other tissues not directly involved in the exercise. In agreement with this assumption, it has been reported in humans that muscle contraction diverts leg uptake of LCFA to oxidation rather than to reesterification in IMTG (241) and that during leg exercise nonexercising muscle may in fact take up and reesterify LCFA to IMTG (262). Along these lines, it is also possible that esterification takes place in the nonactive motor units in a muscle working at a submaximal exercise intensity, whereas esterification of LCFA in the actual contracting fibers probably is less likely.

An important question is, however, the extent to which data obtained in whole body measurements reflect what is occurring locally in the exercising muscle. This question is obviously mostly relevant when large muscle groups are involved in exercise such as in bicycle ergometer exercise. To address this question, it is necessary to simultaneously measure whole body and leg uptake and oxidation of LCFA. Our own studies at 60–68% of peak pulmonary oxygen uptake show that only 32–45% of the systemic plasma LCFA total uptake (R_d) occurred in the two legs (117, 235, 237). Supplementing these data, the study by Burguera et al.(30) showed that during exercise at 45% of maximal oxygen consumption the uptake in the legs was 60% of whole body LCFA turnover during exercise. That such a relatively small fraction of systemic plasma LCFA total uptake is extracted in the active muscles implies that about one-half of the systemic plasma LCFA presumably is taken up in adipose tissue, the heart, liver, inactive or slightly active upper body muscles, and possibly other organs during bicycle exercise. However, looking at LCFA utilization at rest and during exercise, it appears that at rest there is a relatively high whole body lipid turnover, and the increase with submaximal bicycle exercise is about two- to threefold (30, 117, 235, 237). In contrast, leg LCFA uptake at rest is very low and the increase with exercise is on the order of 5- to 15-fold (30, 235). Thus another way to compare whole body and leg LCFA uptake is to compare the increase in utilization with exercise (Table 1). With an examination of data from the study by Roepstorff et al. (235), a gender comparison is also possible. It appears that 60–76% in endurance-trained females and males, respectively, of the exercise-induced increase in whole body LCFA R_d can be accounted for by uptake in the legs during submaximal bicycle exercise at 60% peak oxygen uptake, and no major gender effect was demonstrated in these endurance-trained individuals (Table 1). Thus 24–40% of LCFA R_d is taken up elsewhere than the exercising muscles. During one-legged exercise in which only a relatively small muscle mass (2–3 kg) is engaged in exercise, 55% of the increase in whole body palmitate oxidation was accounted for by increase in leg palmitate oxidation (310). Collectively the data on whole body and leg uptake of LCFA indicate that because of the relatively high whole body resting uptake (R_d) of LCFA, the R_d of LCFA during exercise only to a limited extent reflects uptake of LCFA in active muscle during exercise. It is worthwhile to note that due to simultaneous uptake and release of LCFA, true tracer-determined LCFA uptake is two to three times higher than net uptake (101, 117, 235, 300). In this respect, it might also be worth noting that while it has been suggested, based on experiments in three subjects, that net leg LCFA uptake is underestimated when the femoral venous catheter is placed in the normal antegrade (pointing towards the heart) direction rather than the retrograde direction (pointing towards the knee), tracer-derived leg LCFA uptake is not affected by catheter placement (310).

View this table: [in this window] [in a new window]   TABLE 1. Exercise-induced increase () in mean whole body and leg uptake of LCFA and percentage of increase in whole body uptake of LCFA that occurs across the legs

Summarizing the available data, it seems fair to conclude that during submaximal exercise of 90 min or less only a relatively minor part (35–60%) of the whole body LCFA turnover occurs in the exercising muscle. On the other hand, the exercise-induced increase in leg uptake of LCFA reflects the measured exercise-induced increase in whole body R_d fairly well, although the former underestimates the latter to some extent. The underestimation may be ascribed to LCFA uptake in accessory exercising muscle other than the legs. Moreover, it appears that the higher the exercise intensity, the closer to 100% oxidation of leg uptake of LCFA is achieved, whereas at low exercise intensities, an incorporation of the LCFA into IMTG may occur probably in inactive motor units.

	III. REGULATION OF LONG-CHAIN FATTY ACID UTILIZATION IN SKELETAL MUSCLE DURING EXERCISE

Top Previous Next References

 
A. Supply of Long-Chain Fatty Acids

There are several steps involved in the pathway from LCFA release either from adipose tissue or from the triacylglycerol-rich lipoproteins in the circulation into their final oxidation in the mitochondria which could play a significant role in regulation of LCFA oxidation (Fig. 1).

View larger version (47K): [in this window] [in a new window]   FIG. 1. Schematic representation of the likely routes taken by fatty acids from the capillary to the mitochondria. VLDL, very-low-density lipoprotein; ALBFA, albumin bound fatty acid; LPL, lipoprotein lipase; ALBR, albumin receptor; LBP, lipid binding protein; ACS, acyl CoA synthetase; FABPc, cytosolic fatty acid binding protein; ACBP, acyl CoA binding protein; TG, triacylglycerol; LCFA, long-chain fatty acids; CD36, fatty acid translocase; FABPpm, membrane fatty acid binding protein.

View larger version (25K): [in this window] [in a new window]   FIG. 2. Arterial plasma concentration (A), fractional uptake (B), uptake (C), and total oxidation (D) of free fatty acids (FFA = long-chain fatty acids) across the thigh during rest and 3 h of knee extension exercise in trained and untrained subjects. Values are means ± SE of 6 subjects in trained and untrained groups. *P < 0.05 compared with untrained; P < 0.05 compared with previous value. [From Turcotte et al. (300).]

From the circulation LCFA must pass the endothelium, the interstitial space, the plasma membrane, the cytosol, and the mitochondrial membranes for their final oxidation in the mitochondria (Fig. 1). The elaborate chain of reactions provides a number of possible points of regulating the supply of LCFA for oxidation. It has been a matter of debate whether the transendothelial and/or transsarcolemma transport of LCFA is a passive process dependent on the rate of cellular metabolism or occurs via plasma membrane protein-mediated transport. However, from recent studies it seems that both mechanisms are operating (22, 219).

Within recent years three putative fatty acid binding proteins located at the plasma membrane have been identified. These are 1) the plasma membrane-bound fatty acid binding protein (FABP_pm), 2) fatty acid translocase (FAT/CD36), and 3) the fatty acid transport protein (FATP). FABP_pm is a 43-kDa protein located peripherally on the plasma membrane, which in fact is identical to the mitochondrial enzyme aspartate aminotransferase (mAAT) (26, 288). FAT/CD36 is a 88-kDa integral membrane glycoprotein with two predicted transmembrane domains, and it is found to be 85% homologous to the glycoprotein IV or CD36 of human blood platelets and leukocytes (2). FATP is a 63-kDa integral protein with six predicted transmembrane domains (253), and recently a family of FATPs was identified (122). Both FABP_pm and FAT/CD36 are present in most metabolic tissues including human skeletal muscle (24, 31, 32, 149, 155, 234) (Fig. 3), and the transcript of FATP1 mRNA has also been detected in human muscle (24, 155). To date, the mechanism of action of these lipid binding proteins is not well known, but increasing evidence is emerging that they are involved in fatty acid uptake.

View larger version (19K): [in this window] [in a new window]   FIG. 3. Representative Western blots of fatty acid translocase (FAT)/CD36, plasma membrane-bound fatty acid binding protein (FABPpm), cytosolic fatty acid binding protein (FABPc), and acyl CoA binding protein (ACBP) in human skeletal muscle total crude membranes (FAT/CD36, FABPpm, and ACBP) or cytosolic (FABPc) fractions showing a single band at 88, 43, 15, and 10 kDa, respectively.

Recent evidence supports an important role for FAT/CD36 in uptake of LCFA in rodent skeletal muscle. For instance, overexpression of FAT/CD36 in mice reduced plasma triacylglycerol and LCFA concentrations and increased palmitate oxidation in contracting soleus muscle compared with wild-type muscles (135). Conversely, FA uptake was reduced in FAT/CD36 null mice (71). When plasma membranes were isolated, using the giant vesicle preparation, a higher FAT/CD36 protein expression was obtained in muscles from streptozotocin-induced diabetic rats (181) and in muscles from obese Zucker rats (182), which display increased fatty acid transport. Conversely, when rats were treated with leptin for 14 days, a reduction was obtained in sarcolemmal FAT/CD36 protein content in association with a decrease in FA transport (282). In humans, a fat-rich diet induced an increase in the expression of FAT/CD36 measured in homogenates from the vastus lateralis muscle (32, 238).

Both triacylglycerol content and FAT/CD36 protein and gene expression measured in the vastus lateralis muscle appeared to be higher in women than in men (155, 281) (Fig. 4). Furthermore, an increase in triacylglycerol storage was found in skeletal muscle (117, 148) in concert with an increase in FAT/CD36 protein induced by a high-fat diet (238). Moreover, in obese and type 2 diabetic men and women, the amount of FAT/CD36 protein, measured in giant vesicles prepared from the rectus abdominus muscle, was higher compared with lean controls, and the FAT/CD36 protein expression correlated with the IMTG content (25). On the basis of these findings of simultaneous high expression of FAT/CD36 and IMTG content, it could be speculated that an enhanced amount of FAT/CD36 would be of benefit for clearance of circulating LCFA. Once taken up in the skeletal muscle, the LCFA could be either oxidized in the mitochondria or directed towards resynthesis into triacylglycerol. If the need for energy is low compared with the uptake of LCFA, the latter will be directed towards resynthesis into IMTG. In accordance with this suggestion, overexpression of FAT/CD36 in C₂C₁₂ myotubes increased intracellular triacylglycerol content in the presence of excess (750 µM) palmitate but not when palmitate in the medium was low (80 µM) (11).

View larger version (16K): [in this window] [in a new window]   FIG. 4. Intramyocellular triacylglycerol (IMTG) (mmol/kg dry wt) and FAT/CD36 protein (arbitrary units) content in vastus lateralis muscle in male (n = 24) and female (n = 24) subjects. *P < 0.05 between gender. [Data from Kiens et al. (155) and Steffensen et al. (281).]

View larger version (99K): [in this window] [in a new window]   FIG. 5. Distribution of FAT/CD36 in human skeletal muscle. A: transverse cryosections from human vastus lateralis muscle were fixed and labeled with anti-FAT/CD36 mouse monoclonal antibody. B: colocalization of FAT/CD36 and caveolin-3 in human skeletal muscle. Longitudinal cryosections from the vastus lateralis muscle were fixed and double labeled with anti-FAT/CD36 mouse monoclonal antibody and anti-caveolin-3 rabbit polyclonal antibody. A–C: surface of a single muscle fiber. D–F: magnification of the central part of A–C. Arrows indicate a spot with a clear colocalization of FAT/CD36 and caveolin-3. Bars = 10 µm. [From Vistisen et al. (320).]

2. FABP_pm

With regard to FABP_pm, this protein may also play a role in the transport of LCFA across the sarcolemma. In rat muscle, fasting (48 h) increased FABP_pm protein content, but only in oxidative muscles which are highly dependent on fatty acid metabolism (301). A role for FABP_pm in transport of long-chain saturated and nonsaturated fatty acids has been suggested from use of antibodies against FABP_pm, which leads to inhibition of LCFA uptake in various cell types and of transport of LCFA into skeletal muscle giant vesicles in rats in a dose-dependent manner (263, 274, 287, 303, 344). Similarly, overexpression of FABP_pm by transfection in rat skeletal muscle increased fatty acid transport and metabolism in resting muscle (42). However, elucidation of the role of FABP_pm in LCFA transport was challenged by the findings that FABP_pm is identical to the mitochondrial isoform of the enzyme aspartate aminotransferase (mAspAT) (26, 288). In favor of a role of mAspAT/FABP_pm in fatty acid binding, molecular modeling studies of the crystal structure of mAspAT have identified a pocket, within the larger domain of the enzyme, which is of sufficient size to accommodate the typical LCFA (15). Whether in fact this pocket serves as a fatty acid binding site remains to be elucidated. Recently, a study using a polyclonal antibody against rat mAspAT in immunogold electron microscopy of rat tissue sections has reported a strong labeling of mitochondria in several cell types (36). Labeling was also observed in other locations such as endothelial cell surfaces, and it was concluded from these observations that mAspAT/FABP_pm is both a mitochondrial enzyme and a plasma membrane protein (36).

Dietary manipulations affect FABP_pm protein expression. A high-fat diet induced an increase in FABP_pm protein expression in vastus lateralis muscle in male volunteers, and a carbohydrate-rich diet decreased the FABP_pm content (238). The upregulation of FABP_pm protein was however only obtained after long-term dietary interventions as short-term dietary interventions did not affect FABP_pm protein expression (32, 238). In human obesity, a higher FABP_pm protein expression in homogenates from the vastus lateralis muscle was found compared with lean subjects, irrespective of gender (271). This is in contrast to findings by Bonen et al. (25), who showed a reduced or no difference in protein expression of FABP_pm measured in homogenates or plasma membranes (from giant vesicle preparations), respectively, from the rectus abdominis muscle in a gender mixture of obese compared with lean subjects. The discrepancy between these findings might be ascribed to the fact that different muscles have been investigated.

Exercise training is another condition where the capacity for lipid metabolism is changed. Accordingly, a higher FABP_pm protein expression in homogenates from the vastus lateralis muscle was also demonstrated with exercise training (149, 155). The training-induced upregulation of FABP_pm protein is obviously related to gender as changes in FAPB_pm protein were not obtained in women (155). Interestingly, gender differences in FABP_pm protein content in the vastus lateralis muscle were not obtained in nontrained subjects (155, 271) in contrast to findings of FAT/CD36 protein content, which is higher in females than in males (155).

3. FATP1

FATP1 is expressed in adipose tissue, heart, and skeletal muscle of mouse and rat (1, 122, 250, 253) and in skeletal muscle of humans (24, 155). Evidence for the importance of FATP in LCFA transport comes from experiments in cultured cells (122, 253) and yeast (67). A number of studies have reported that FATP1 has acyl CoA synthase activity (45, 105, 322), indicating a role in conversion of fatty acids to fatty acid acyl CoA. In growing 293 cells, fatty acids taken up through FATP1 were preferentially channeled into triacylglycerol synthesis, which has suggested a functional link between FATP1-mediated fatty acid uptake and lipid storage (106). Further support for this contention are the findings in FATP1 knock-out mice of a marked reduction in triacylglycerol and diacylglycerol content in the quadriceps muscle after 3 wk high-fat diet compared with wild-type mice (158). FATP1 protein content was markedly higher in homogenates from the soleus than the gastrocnemius muscle in rats (192), and high-fat diets caused an elevation in FATP1 protein content in the soleus muscle, but a reduction in the gastrocnemius muscle. In a group of matched females and males, vastus lateralis muscle FATP1 mRNA expression was not different despite the findings of significantly higher IMTG in the females than in the males (155, 281).

4. Caveolins

Within the recent years caveolae and lipid rafts have been suggested to be involved in fatty acid uptake. Caveolae are 50- to 100-nm flask-shaped invaginations of the plasma membrane. Caveolae have been well described in adipocytes, in endothelial cells, in type 1 pneumocytes of the lung, and in skeletal muscle cells. Caveolae are a morphological subclass of lipid rafts, specialized microdomains, composed of sphingolipids and cholesterol which contain caveolin, the protein which is essential for invagination of the plasma membrane (for review, see Ref. 46). Three caveolins have been identified of which caveolin-1 and caveolin-2 are expressed in most cell types, whereas caveolin-3 is restricted to skeletal muscle (254, 293).

Recent data in HepG2 and endothelial cells suggest that caveolae may play a significant role in uptake and intracellular trafficking of LCFA (217, 231). With regard to caveolin-1, it was shown that detergent-insoluble fractions from alveolar type II cells were enriched in FAT/CD36 and caveolin-1 (163) and furthermore that caveolin-1 binding of fatty acids was saturable (295). In the caveolin-1 null mice, it was found that serum triacylglycerol and fatty acid concentrations were markedly increased compared with the wild type, especially in the postprandial state (228). Taken together, these findings strengthen the view that caveolae and caveolin are involved in regulation of lipid transport. We recently demonstrated that FAT/CD36 was abundantly expressed in endothelial cells surrounding the capillaries in human skeletal muscle (Fig. 5) (320), and a close correlation was obtained between muscle content of FAT/CD36 and caveolin-1 (C. Roepstorff, K. Roepstorff, B. Vistisen, B. Kondrup, B. vanDeurs, and B. Kiens, unpublished data). Further support for a link between FAT/CD36 and caveolin-1 in human skeletal muscle was our recent findings of a coincidental increase in FAT/CD36 and caveolin-1 protein expression in human skeletal muscle, during recovery after intense, prolonged whole body exercise (237). These data suggest that caveolin-1 and FAT/CD36 may be regulated in a coordinated manner in conditions of increased LCFA utilization.

Besides the indications of an association of caveolin-1 and FAT/CD36 in human skeletal muscle endothelial cells, we also recently showed, in longitudinal muscle sections, in which part of a fiber was cut tangentially, a high degree of colocalization of caveolin-3 and FAT/CD36 in the sarcolemma (320) (Fig. 5). One putative mechanism whereby caveolin could be involved in cellular fatty acid uptake is that caveolae may regulate the function of fatty acid transporters such as FAT/CD36.

The endothelial cells lining the capillary are the first transport barrier for the circulating fatty acids before transport into muscle cell. Earlier studies suggested that a specific interaction of the albumin-FA complex with proteins associated with the endothelial membrane facilitated the dissociation of the FA-albumin complex and, hence, the transfer of FA from the vascular to the interstitial compartment (307). Because FAT/CD36 is present abundantly in the capillary endothelium as well as in the sarcolemma, FAT/CD36 may play a role in the transport both in the endothelial cells and in the plasma membrane. Because the protein has adhesion functions (72), it may be speculated that FAT/CD36 in the endothelium could be involved in adherence not only of albumin-bound fatty acids but also of fatty acids released from the hydrolysis of circulating VLDL-TG by lipoprotein lipase activity.

Summarizing the available data, it would seem that the fatty acid binding proteins expressed in the endothelium and plasma membrane are involved in fatty acid uptake and that caveolae and caveolins may be involved in this process as well. Yet, the precise functional role of and mechanisms by which the proteins are mediating LCFA uptake is not known, and the mechanisms of LCFA uptake into various mammalian cells may not be the same. It may seem redundant that muscle cells express several types of fatty acid transporters, but the possibility exists that they have distinct roles in the transport process, or that they function in a coordinated manner in the transport of LCFA. However, this is not known at present.

5. Influence of exercise

Then the question arises whether the transport proteins may play any regulatory role in LCFA uptake and oxidation in human skeletal muscle during exercise, a situation where LCFA uptake and oxidation can increase manyfold. In rat skeletal muscle, uptake of palmitate into sarcolemmal giant vesicles increased 50–75% after contractions (23) and correlated with expression of FAT/CD36 (187). The training-induced increase in LCFA utilization during exercise (147, 300) (Fig. 2) could theoretically be ascribed to an increased number and/or activity of the lipid-binding proteins in the trained compared with the nontrained state, which might facilitate LCFA transport into the myocyte. This view was supported by the findings of an increase in muscle FABP_pm protein expression with endurance exercise training in male subjects (149, 155). In addition, training of rats led to an increase in muscle FABP_pm together with a contraction-induced uptake and oxidation of palmitate (302). The training-induced increase in FABP_pm is apparently gender specific because the FABP_pm protein content was similar in untrained and well-trained female volunteers (155). Interestingly, when comparing endurance-trained and untrained female and male subjects, FAT/CD36 protein content in vastus lateralis muscle was not different (155). In contrast, short-term training for 9 days (297) as well as even a single exercise bout (237) increased FAT/CD36 protein content slightly (20–25%) in muscle, perhaps suggesting that increased FAT/CD36 expression is an early adaptation to increased muscle activity that may wane with sustained increased activity.

Even though there is evidence that the fatty acid binding proteins are involved in transport of LCFA across the membrane, this does not necessarily mean that the transport proteins or the transport process is rate limiting for fatty acid utilization during exercise. On the one hand, overexpression of FAT/CD36 increases fatty acid oxidation during electrically induced muscle contractions (135), supporting a regulatory role of FAT/CD36. On the other hand, an intracellular accumulation of fatty acids was demonstrated in human vastus lateralis muscle when exercise intensity was increased from 65 to 90% of VO_2peak and fat oxidation decreased markedly (154). This increase in LCFA content within the muscle cell during the intense exercise was even found in parallel with a significant decrease in blood plasma LCFA concentration (154). These findings give support to the notion that when exercise intensity is increased and fat oxidation decreases, fatty acid oxidation is limited by factors inside the muscle cell, rather than by limitations in transmembrane transport. In addition, by manipulating the preexercise muscle glycogen levels, Roepstorff et al. (237) achieved one situation with high fat oxidation and another with high carbohydrate oxidation in human skeletal muscle during a submaximal bicycle exercise bout. This resulted in a 100% higher leg plasma LCFA oxidation, measured by the tracer technique, during the high fat-oxidation trial compared with the high carbohydrate-oxidation trial, but the leg uptake, clearance, and fractional extraction of circulating LCFA were similar in the two trials. These findings further strengthen the view that there are several exercise conditions during which transmembrane transport during exercise may not be the important limiting step in plasma LCFA oxidation.

C. Transport in the Cytosol

LCFA taken into cells are activated in the cytosol by reaction with CoA and ATP to yield long-chain fatty acyl CoA (LCFA-CoA) catalyzed by long-chain acyl CoA synthetase (ACS) (Fig. 1). In addition to being substrates for -oxidation and triacylglycerol synthesis (48), it has also been proposed that LCFA CoA esters play a role in enzyme activation (68, 180), vesicular trafficking (273), and cell signaling (28). The active site of ACS has been located to the cytosolic surface of the peroxisomal endoplasmic reticulum and outer mitochondrial membranes (48). It was recently demonstrated in 3T3-L1 adipocytes that long-chain ACS is an integral membrane protein also located in the plasma membrane (84), and it was suggested that incoming LCFAs are immediately esterified at the plasma membrane. An efficient esterification maintains a low intracellular LCFA concentration and contributes to uptake of LCFA. ACS may therefore affect both the rate and directionality of LCFA movement across membranes and may be coupled to other proteins that participate in LCFA uptake such as FATP 1, which was reported to have ACS activity (45, 84, 105, 252, 322). There is now strong evidence that the intracellular transport of LCFA moieties is mediated by a cytoplasmic fatty acid binding protein (FABP_c) (17, 86, 88) and a cytoplasmic acyl CoA binding protein (ACBP) (226). Very little LCFA and LCFA-CoA actually exist as free or unbound molecules but are rather bound to cytosolic FABP_c and ACBP. LCFA-CoAs are amphipathic molecules and bind strongly to phospholipid membranes. Data indicate that the cytosolic binding proteins act in extracting LCFA-CoA from and prevent their binding to biological membranes and liposomes and donate the LCFA-CoAs for metabolic -oxidation (87, 226). Even though LCFA-CoAs are relatively water soluble and might move through the cytosol in an unbound fashion, their binding to FABP_c and ACBP is believed to be a major factor in controlling the free concentration of cytosolic LCFA-CoA. The importance of FABPc in LCFA uptake was revealed in mice, where the gene of the heart isoform of FABP_c was ablated. This resulted in a reduced uptake of LCFA in heart of 50% (199). In skeletal muscle, uptake of LCFA was decreased by 45% in homozygous mice but was unaffected in heterozygous mice (184).

Other conditions altering the content and regulation of the cytosolic fatty acid binding proteins are interventions leading to changes in lipid metabolism. Dietary manipulation is one such intervention. Available information is mainly on the cytoplasmic FABP_c. Moreover, most data are from rat studies, where animals have been fed a fat-rich diet, mainly composed of saturated fatty acids (44). Collectively, these data have shown that FABP_c in heart and skeletal muscle did not respond to an increase in saturated dietary fatty acids (44, 285). In contrast, a study in rat heart and skeletal muscle revealed that ingestion of a diet rich in omega-3 fatty acids markedly increased FABP_c content (43), suggesting that the length and degree of saturation of the carbon chain of the fatty acids are important for regulation of FABP_c. In agreement with these findings, in humans the ingestion of a fat-rich diet in which the ratio of polyunsaturated to saturated fatty acids was relatively high (0.6) induced a significant increase in FABP_c protein content in the vastus lateralis muscle, in contrast to the ingestion of carbohydrate-rich diet for 4 wk (238). It has been suggested that FABP_c cooperates with the membrane-associated lipid binding protein FAT/CD36 in uptake of LCFA in cardiac and skeletal muscle (186). Interestingly, in a dietary intervention study, a similar increase in FABP_c and FAT/CD36 protein content was observed, in the vastus lateralis muscle following a fat-rich diet in human volunteers in contrast to when a carbohydrate-rich diet was consumed (238).

Even though exercise training increases the potential for a higher lipid oxidation, neither FABP_c nor ACBP levels in vastus lateralis muscle in humans were changed by exercise training, suggesting that the amounts of these cytosolic proteins were abundant and sufficient in trafficking LCFA-CoA esters and LCFA within the cell during exercise (155). These data are supported by recent findings in rats showing that endurance training did not affect the protein level of ACBP in skeletal muscle (74) even though the lipid oxidation potential in muscle was increased.

D. Metabolism in Mitochondria

Long-chain fatty acyl CoA (LCFA-CoA) is transported across the mitochondrial membrane for final -oxidation to generate acetyl CoA for the tricarboxylic acid (TCA) cycle. Both transport and oxidation are potential sites of regulation (Fig. 6).

View larger version (59K): [in this window] [in a new window]   FIG. 6. Schematic representation of the likely routes taken for long-chain fatty acyl CoA (LCFA CoA) to enter the mitochondrion for subsequent -oxidation. CPT-1, carnitine palmitoyl transferase 1; CPT-2, carnitine palmitoyl transferase 2; ACS, acyl CoA synthetase; TCA, tricarboxylic acid cycle.

View larger version (15K): [in this window] [in a new window]   FIG. 7. Relationship between muscle activity of -hydroxyacylCoA dehydrogenase (HAD) and the highest rate of oxidation of fatty acids during 3 h of dynamic knee extensor exercise at 65% of knee extensor peak oxygen uptake. Each point represents values from a single individual. [From Kiens (146).]

The formation of malonyl CoA from acetyl CoA is catalyzed by the enzyme acetyl-CoA carboxylase (ACC), which exists in two isoforms (- and -form) of which ACC is dominating in skeletal muscle and heart (97)(Fig. 8). Two types of regulation of ACC have been described. The first is allosteric activation by the cytosolic concentration of citrate. In accordance, studies in rodents and humans revealed that high glucose availability at rest induced an elevation in cytosolic citrate concentration and in muscle malonyl CoA concentration (12, 224, 242, 243). The second type of regulation of ACC involves phosphorylation and inactivation by 5'-AMP-activated protein kinase (AMPK) (225, 242, 334).

View larger version (16K): [in this window] [in a new window]   FIG. 8. Regulation of malonyl CoA. Schematic illustration of how it is thought malonyl CoA content may be regulated in skeletal muscle. As discussed in text, it is controversial whether MCD is activated by AMPK. AMPK, 5'-AMP activated protein kinase; ACC, acetyl-CoA carboxylase; MCD, malonyl CoA decarboxylase; CL, citrate lyase.

By lowering the malonyl CoA content in skeletal muscle, the inhibitory effect on CPT1 is diminished, and the entry of acylated fatty acids into the mitochondria is facilitated. Relatively strong evidence has accumulated indicating that changes in ACC activity and malonyl CoA concentration regulate fatty acid entry into the mitochondria in resting muscle (reviewed in Ref. 195) and also during exercise in rodent muscle (225, 333, 334), but the question is whether the same regulatory mechanisms operate in human skeletal muscle during exercise. Earlier studies in human volunteers have failed to demonstrate decreases in malonyl CoA concentrations during prolonged submaximal exercise (202) and during short-term exercise at various submaximal intensities (203) despite marked increases in fatty acid oxidation compared with rest. However, recently we reported that during exercise at 60% of maximal oxygen uptake, malonyl CoA concentrations decreased compared with resting values, but the decrease was in fact similar when subjects were preconditioned with either low muscle glycogen levels and subsequently high AMPK activity or with high muscle glycogen levels and low AMPK activity (234), even though lipid oxidation was markedly different in the two conditions. These findings suggest that the decrease in malonyl CoA concentration from rest to exercise may contribute to the increase in absolute lipid oxidation when commencing exercise but plays a lesser role in fine tuning lipid oxidation during exercise. In another study exercise was performed with the knee-extensors at 60, 85, and 100% of the knee-extensor's maximal oxygen uptake (56). Going from rest to 60% one-legged exercise, an 50% reduction in ACC activity was associated with an increase in fatty acid oxidation, but no detectable change was noticed in the concentration of malonyl CoA. With increasing exercise intensities, at which the rate of fatty acid oxidation is diminished, the concentration of malonyl CoA was in fact modestly decreased and ACC activity was further reduced. From these human studies, it seems likely that other mechanisms than changes in ACC activity, AMPK activity, and malonyl CoA concentrations must be involved in the regulation of lipid utilization during exercise. In line with this suggestion are recent findings from studies performed in the perfused rat hindquarter in which low-intensity muscle contractions induced an increase in FA uptake and oxidation and a decrease in malonyl CoA muscle content without changes in total AMPK and ACC activities, suggesting that AMPK activation is not critical in the regulation of FA uptake and oxidation during low-intensity muscle contraction (223).

1. Regulation of fat oxidation by carnitine

Carnitine is substrate for CPT1 and is required for transport of the activated LC fatty acyl CoA across the inner mitochondrial membrane and is therefore essential for the -oxidation (Fig. 6). Carnitine could thus play a role in regulating lipid oxidation. Available data from studies at rest suggest that fat oxidation in skeletal muscle is not limited by carnitine (221, 234, 275). Carnitine could, however, be a molecule that has important regulatory roles in adjusting lipid oxidation during exercise. The classical findings that carnitine can be acetylated by acetyl CoA (80) makes carnitine a sink for acetyl groups during conditions where the rate of acetyl CoA formation from pyruvate exceeds the rate of utilization by the TCA cycle. Accordingly, it has been demonstrated that the concentration of acetyl carnitine within the muscle is enhanced with increasing exercise intensities (49, 121, 203, 246, 312) decreasing the availability of free carnitine (49, 121, 246, 312), resulting in low CPT1 activity due to low availability of its substrate free carnitine. In turn, this will lead to a diminished supply of LCFA-CoA to -oxidation, consequently limiting oxidation of fatty acids during exercise. This provides a potential mechanism whereby an increased availability of pyruvate and acetyl CoA formation can downregulate lipid oxidation (246). Supporting this view, an increase in muscle acetyl carnitine and decreased muscle free carnitine concentrations were observed concomitantly with a decrease in LCFA oxidation during increasing exercise intensity in male volunteers (312). Furthermore, we recently showed that when preexercise muscle glycogen concentrations were low, thereby providing few acetyl groups to acetylate carnitine, muscle free carnitine concentrations and rate of fat oxidation were markedly higher during submaximal exercise than when preexercise muscle glycogen concentrations were high and free carnitine concentrations in skeletal muscle were low. The findings indicate that the availability of free carnitine may limit fat oxidation during exercise (234). Interestingly, when collecting data from various published studies where acetyl carnitine concentrations were measured in skeletal muscle and substrate utilization was estimated from RER values during exercise, a close positive relationship between acetyl carnitine concentrations and RER values and a negative relationship between acetyl carnitine and fat oxidation is found (Fig. 9). Because total carnitine concentration in skeletal muscle is unaffected by exercise, this relationship suggests that high RER values and low fat oxidation may be related to the availability of free carnitine in human skeletal muscle during exercise. Putting it differently the positive relationship in Figure 9 suggests a relationship between free carnitine concentration in contracting muscle and muscle fat oxidation rate. Even though CPT1 no doubt is the central gatekeeper for entry of fatty acyl moieties into the mitochondria, it was recently shown in isolated mitochondria from rat muscle that FAT/CD36 is located in the mitochondrial membrane and immunoprecipitation with FAT/CD36 antibody coprecipitates CPT1 (33).This is interesting but surprising since morphological studies do not reveal such localization (143, 320). It was suggested that FAT/CD36 could play a role in LCFA transfer across the mitochondria membrane likely in combination with CPT1 (33), but further work is needed to investigate this possibility.

View larger version (19K): [in this window] [in a new window]   FIG. 9. Relation between acetyl carnitine and respiratory exchange ratio (RER) or fat oxidation. Relationship between the acetyl carnitine concentration in the human vastus lateralis muscle and RER (A) or fat oxidation (B) during exercise under different conditions. It should be noted that the sum of acetylcarnitine and free carnitine concentration is constant during exercise. Therefore, an increase in acetylcarnitine concentration coincides with a decrease in free carnitine concentration. [Data are extracted from several studies (203, 221, 234, 275, 312, 327) and reproduced from Roepstorff (233).]

View larger version (47K): [in this window] [in a new window]   FIG. 10. Metabolic interactions during exercise. Metabolic interactions discussed in the text that may determine the fat oxidation rate in human skeletal muscle during exercise. When exercise induces an increase in glycogenolysis and glycolysis, it will through the action of the PDH complex lead towards enhanced mitochondrial generation of acetyl CoA and secondly acetyl carnitine. In turn, this will result in a decrease in mitochondrial free carnitine and consequently in cytosolic carnitine concentrations. A reduction in cytosolic carnitine content results in a decrease in substrate availability for CPT1 and thereby the generation of LCFA CoA for -oxidation. When, on the other hand, the glycogenolytic and glycolytic rate is low, the mitochondrial acetyl CoA and acetyl carnitine generation appear to be lower, and a decrease in the carnitine level in the cell does not occur, and substrate for CPT1 is provided. Exercise also reduces the generation of malonyl CoA due to the increased phosphorylation (inactivation) of ACC. This reduction may be slightly more pronounced during lower carbohydrate oxidation, due to the lower concentration of cytosolic acetyl CoA. A reduction of malonyl CoA content with exercise may contribute to the increased fat oxidation appearing from rest to exercise. However, because a reduction in malonyl CoA content is apparently independent of fuel combusted during exercise, carnitine rather than malonyl CoA seems to be a likely candidate in fine tuning fat oxidation during exercise.

Other factors involved in the regulation of LCFA utilization during exercise could be considered. It has been shown in vitro in human skeletal muscle that a decline in pH from 7.1 to 6.8 resulted in a significant decrease of 34–40% in CPT1 activity both in sarcolemmal and interfibrillarly located mitochondria (16, 280). A link between pH and CPT1 makes sense, as during prolonged submaximal exercise when lipid utilization is high, only a small increase in muscle lactate and a small decrease in muscle pH is observed in contrast to exercise performed at high intensities, where high levels of muscle lactate have been obtained in parallel with reductions in pH (245) and lipid oxidation. Thus an exercise-induced decrease in pH causing a decrease in CPT1 activity could also seem to be a likely mechanism contributing to a decrease in lipid oxidation during high-intensity exercise.

Another step to consider is the -ketoacyl-CoA thiolase, the enzyme responsible for catalyzing the final reaction in -oxidation: -ketoacyl CoA to acetyl CoA and acyl CoA (Fig. 6). -Oxidation is inhibited mainly by feedback inhibition and acetyl CoA has been shown to be a potent inhibitor of -ketoacyl CoA thiolase. Therefore, when acetyl CoA is accumulating the thiolase is inhibited and accordingly so is -oxidation. In fact, inactivation of the thiolase causes the complete inhibition of palmitate -oxidation (177). Thus high concentrations of acetyl CoA, as observed during intense exercise in muscle (49, 62, 203) or when muscle glycogen levels are high (221, 234), may tend to slow -oxidation by inhibition of -keto acyl CoA thiolase.

Finally, changes in the membrane lipid composition may affect the kinetic properties of different membrane-associated enzymes and transporters, which could influence the LCFA uptake. Recently it was demonstrated that exercise training and diet induce modifications of the phospholipid composition of the membranes in human skeletal muscle (7, 8, 118). The functional role of a change in membrane phospholipid composition on transsarcolemmal transport of LCFA should be elucidated.

	IV. CIRCULATING VERY-LOW-DENSITY LIPOPROTEIN-TRIACYLGLYCEROL

Top Previous Next References

 
Endogenous triacylglycerol (TG) is secreted by the liver, wrapped up into VLDL. VLDLs are the main carriers of circulating TG in the postabsorptive state. The hydrolysis of core triacylglycerol in VLDLs is mediated by lipoprotein lipase.

Most authors have neglected the potential contribution of fatty acids derived from circulating VLDL-TG to energy substrates during exercise. This might be due to earlier findings where the arteriovenous (a-v) differences of total serum triacylglycerols (S-TG), both unlabeled and radiolabeled, were measured in subjects during forearm exercise (204). The authors concluded that a-v differences of S-TG did not exceed the level of error of the analytical methods used in the study. On the other hand, they also estimated that due to the accuracy of the methods, a-v differences of up to 20 µM could have been missed and such an a-v difference could, if fully oxidized, account for 25% of the oxygen extraction in the forearm due to the energy density of circulating TG. Thus the potential for a significant contribution of circulating S-TG was not ruled out. In addition, findings in the forearm may not be completely representative to findings in the leg. For example, Havel et al. (109) demonstrated a consistent 4–6% arterial-femoral venous difference of circulating triacylglycerol during exercise in two out of four subjects, whereas in the other two subjects the differences were small at <