Motor center activity and reflexes from contracting muscle have been shown to be important for mobilization of free fatty acids (FFA) during exercise. We studied FFA metabolism in the absence of these mechanisms: during involuntary, electrically induced leg cycling in individuals with complete spinal cord injury (SCI). Healthy subjects performing voluntary cycling served as controls (C). Ten SCI (level of injury: C5-T7) and six C exercised for 30 min at comparable oxygen uptake rates (∼1 l/min), and [1-14C]palmitate was infused continuously to estimate FFA turnover. From femoral arteriovenous differences, blood flow, muscle biopsies, and indirect calorimetry, leg substrate balances as well as concentrations of intramuscular substrates were determined. Leg oxygen uptake was similar in the two groups during exercise. In SCI, but not in C, plasma FFA and FFA appearance rate fell during exercise, and plasma glycerol increased less than in C (P < 0.05). Fractional uptake of FFA across the working legs decreased from rest to exercise in all individuals (P < 0.05) but was always lower in SCI than in C (P < 0.05). From rest to exercise, leg FFA uptake increased less in SCI than in C subjects (14 ± 3 to 57 ± 20 vs. 41 ± 13 to 170 ± 57 μmol · min−1 · leg−1;P < 0.05). Muscle glycogen breakdown, leg glucose uptake, carbohydrate oxidation, and lactate release were higher (P < 0.05) in SCI than in C during exercise. Counterregulatory hormonal changes were more pronounced in SCI vs. C, whereas insulin decreased only in C. In conclusion, FFA mobilization, delivery, and fractional uptake are lower and muscle glycogen breakdown and glucose uptake are higher in SCI patients during electrically induced leg exercise compared with healthy subjects performing voluntary exercise. Apparently, blood-borne mechanisms are not sufficient to elicit a normal increase in fatty acid mobilization during exercise. Furthermore, in exercising muscle, FFA delivery enhances FFA uptake and inhibits carbohydrate metabolism, while carbohydrate metabolism inhibits FFA uptake.
- radiolabeled palmitate
- free fatty acids
- lipid metabolism
- growth hormone
- skeletal muscle
- physical activity
free fatty acids (FFA) are an important fuel source for metabolism in contracting muscle (4,24). It has previously been postulated that mobilization of FFAs is secondary to FFA uptake in skeletal muscle (23). In line with this, the arterial FFA concentration has been shown to inhibit FFA mobilization (21). In addition to such a direct negative feedback mechanism regulating FFA mobilization, the importance of other humoral mechanisms (2, 28) as well as of somatic neural involvement in the primary regulation of FFA turnover (15-17, 29) has been proposed. In regard to FFA uptake in muscle, it has often been emphasized that this depends on FFA delivery, an increase in plasma concentration, and availability of FFA, resulting in a rise in FFA uptake in contracting skeletal muscle (8). Other studies are compatible with the view that FFA uptake also depends on factors other than delivery and may be inversely coupled to carbohydrate metabolism (5, 26).
To clarify the role of blood-borne mechanisms for FFA mobilization during exercise, we have now studied individuals with spinal cord injury who have impaired neural mechanisms with respect to motor center activation of and feedback from skeletal muscle in the lower extremities. It was hypothesized that FFA mobilization will be reduced in the absence of somatic neural mechanisms. In addition, because metabolism in exercising muscle was also directly assessed, the study allowed evaluation of the interplay among FFA delivery, FFA uptake, and carbohydrate metabolism in muscle.
MATERIAL AND METHODS
Ten spinal cord-injured (SCI) individuals [age: 35 (27–45) yr (mean and range); level of injury: C5-T7; injury age: 12 (2–23) yr; weight: 78 (63–87) kg; 6 tetraplegic, 27–41 yr; C5-C6, 2–20 yr injury; and 4 paraplegic, 27–45 yr; T4-T7, 10–23 yr injury] and six healthy able-bodied control subjects [25 (20–32) yr; 76 (68–84) kg] gave their informed consent to participate in the study that was approved by the Municipal Ethical Committee of Copenhagen. Before the study, all subjects had their maximal oxygen uptake (V˙o 2 max) determined by the leveling off criterion of oxygen uptake during incremental work on a bicycle ergometer either during voluntary cycling [controls (C),V˙o 2 max: 4.4 ± 0.5 l/min, mean and SE] or during electrically induced cycling (SCIV˙o 2 max: 1.3 ± 0.2 l/min). All SCI individuals were neurologically stable with clinically complete motor lesion in the lower extremities (Frankel classification A). Tetraplegic subjects had motor control over a few muscles in the upper part of the body, whereas paraplegic individuals had full muscle control in the upper extremities. SCI subjects had participated in an ongoing training program for at least 4 mo, including regular performance of electrically induced cycling similar to the one used in this experiment.
[1-14C]palmitic acid (250 μCi) in ethanol (New England Nuclear, Boston, MA) was put into a sterile round-bottomed flask to which excess NaOH was added. After the mixture had been evaporated to dryness with ethanol, the residue of sodium salts was dissolved in 10 ml sterile saline and heated to 70°C while being stirred. When the odor of ethanol could no longer be detected, the mixture was cooled to 55°C, and 1 ml 20% sterile human serum albumin (Statens Serum Institute, Copenhagen, Denmark) was added. The albumin-palmitate solution was filter sterilized (Millipore, 0.22 um), checked for bacteria, and kept frozen by Isotopapoteket, Copenhagen, Denmark, until use.
Subjects arrived at the laboratory at 8:00 A.M. in a postabsorptive state (10-h fast). Teflon catheters were inserted below the inguinal ligament in one femoral artery and one femoral vein and advanced proximally so that the tips were located ∼2 cm proximal and distal to the inguinal ligament, respectively. For determination of leg blood flow by the thermodilution technique, a thermistor was inserted through the venous catheter and the tip advanced 10 cm proximally. A venous catheter was inserted in the antecubital vein of the contralateral arm for the constant infusion of albumin-bound [1-14C]palmitate (0.2 μCi/min) using a calibrated syringe pump. The exact infusion rate was determined for each individual by measuring the radioactivity in the infusate and weighing the syringe before and after the experiment. Infusion of palmitate was started after the catheterization procedure and was continued throughout the experiment. After catheterization, subjects rested supine for 60 min, after which they sat quietly on the bike for 15 min before beginning 30 min of exercise. A needle biopsy was taken from the vastus lateralis muscle using the Bergstrøm technique before and immediately after the end of exercise. The two biopsies were obtained through the same incision 15 cm above the knee, separating the biopsies by a 120° change in sampling angle. After the rest period, SCI subjects exercised for 30 min in the sitting position on a computer-controlled functional electrical stimulation exercise ergometer (REGYS I Clinical Rehabilitation System; Therapeutic Technology Tampa, FL). This system is composed of three primary subsystems: a lower extremity ergometer (Monark ergometer), a stimulus control unit, and a reclinable patient chair. The stimulus control unit controlled and monitored the electrical stimulation according to prescribed parameters (see below) entered into a microprocessor by a remote control keyboard. Before exercise, surface electrodes were placed at motor points, where stimulation threshold was lowest, of the quadriceps, hamstrings, and gluteal muscle groups of both legs. Three electrodes (2 active and 1 reference) were applied over each muscle group. Each electrode was coated with a buffered electrode gel that provided a conductive interface between the electrode and the skin. Six separate channels for sequential surface muscle stimulation were used during ergometry with a computer-controlled closed loop system. Each channel supplies monophasic rectangular pulses lasting 350 μs and delivered at 30 Hz to each of the two active electrodes over a given muscle. Stimulation intensities ranged from preset threshold levels determined for each individual muscle group to elicit a palpable contraction (18–40 mA) up to a maximum of 130 mA. The highest possible ergometer power output, based on prior studies, was chosen to enable subjects to work for 30 min. A pedal position sensor, allowing continuous calculation of velocity, was used by the computer to control the instantaneous stimulus amplitude required for each of the six muscle groups to result in a smooth motion and a constant cranking frequency of 50 rpm. The control subjects performed voluntary leg cycling on the same ergometer as SCI subjects and work load was adjusted to achieve a similar oxygen uptake rate as in the SCI subjects. All subjects had blood samples obtained at rest, after 15 and 30 min of exercise, and after 15 min of recovery for determination of hormones, metabolites, and hematocrit. Oxygen uptake, ventilation, and respiratory exchange ratio were determined with the use of an Ergo-Oxyscreen (Jaeger Instruments) apparatus. Heart rate and blood pressure were determined continuously throughout the experiment by electrocardiogram and intra-arterial pressure.
Blood and muscle sample analyses.
Plasma glucose was determined electrochemically with an automated glucose analyzer (Yellow Springs Instruments). Blood lactate, plasma glycerol, and free fatty acids were determined by enzymatic fluorometric methods. Catecholamine concentrations were determined by a single isotope radioenzymatic method evaluated previously (16). The concentrations of insulin, C-peptide, pancreatic glucagon, growth hormone (GH), ACTH, and cortisol were determined with radioimmunoassays. Hematocrit was determined in triplicate from microcapillary tubes. For determination of palmitate radioactivity, plasma lipids were extracted twice using Dole's extraction mixture (6). The upper phase was removed and evaporated to dryness under nitrogen. The residue was resuspended in chloroform-methanol (2:1 vol/vol) and applied together with appropriate lipid standards to a thin-layer chromatography glass plate (silica gel, 250-μm thickness, Sigma, St.Louis, MO) 3 cm from the bottom. Lipid separation was carried out using a mobile phase of petroleum ether-ether-acetic acid (70:30:1 vol/vol/vol), which was allowed to develop to a height of 16 cm (12). Lipids were separated into the following fractions: FFA, cholesterol, triacylglycerols, monoglycerides, and diglycerides. Each fraction was scraped off the plate, mixed with liquid scintillation fluid (Maxifluor, JT Baker, Deventer, Holland), and counted individually in a liquid scintillation counter. The recovery of radiolabeled palmitate was 92.2 ± 1.4% (n = 12). Content and tension of both oxygen and carbon dioxide in blood as well as pH were measured in a blood gas analyzer (OSM 4, Radiometer). Muscle biopsy samples were frozen in liquid nitrogen within 10 s after they had been taken and were stored at −80°C until further analysis. Before biochemical determinations, samples were freeze dried and dissected free of connective tissue, fat, and blood. Muscle glycogen was determined as glucose residues after hydrolysis in 1 M HCl at 100°C for 2 h, and muscle concentrations of lactate and creatine phosphate were determined with standard enzymatic methods (20).
Fractional uptake of FFA across the leg was calculated as the difference in FFA radioactivity between the arterial and venous plasma samples divided by the FFA radioactivity in the arterial sample (8). FFA delivery was calculated by multiplying plasma flow by the arterial plasma FFA concentration. Gross FFA uptake was calculated by multiplying the plasma delivery by the fractional uptake assuming that fractional uptake of palmitic acid is similar to fractional uptake of other FFAs (8). Appearance rate of FFA (Ra) was calculated assuming that the ratio between [14C]palmitate activity and [14C]palmitate infusion rate equals the ratio between FFA and FFA Ra(8). Net uptake or release of substrates across the leg were calculated by multiplying the blood or plasma flow by the arteriovenous difference in concentration.
Nonparametric statistical tests were used to test whether changes occurred with time (Friedman's test) or whether overall differences existed between groups (Kruskal-Wallis's test). Mann-Whitney's test for unpaired data was used to evaluate differences between experimental groups at specific time points (25). P < 0.05 (2-tailed testing) was considered significant.
During exercise, whole body oxygen uptake [1.02 ± 0.12 (SCI) vs. 1.08 ± 0.11 l/min (C)] as well as leg oxygen uptake [390 ± 31 (SCI) vs. 420 ± 35 ml · min−1 · leg−1 (C)] was similar in the two groups. Leg oxygen uptake (2 legged) accounted for 76 ± 8% (SCI) and 78 ± 9% (C), respectively, of whole body oxygen uptake during exercise, and heart rate rose in response to exercise in both groups [71 ± 6 (rest) to 118 ± 15 (30 min exercise) (SCI) and 68 ± 6 to 107 ± 5 beats/min (C)]. The respiratory exchange ratio was lower in C compared with SCI subjects during exercise (15 min: 0.83 ± 0.01 vs. 0.90 ± 0.02; 30 min: 0.85 ± 0.01 vs 0.93 ± 0.02). The arterial plasma FFA concentration remained unchanged in C throughout the exercise period, whereas it decreased in SCI subjects (Fig.1) and became significantly lower in SCI vs. C after 30 min of exercise (431 ± 54 vs. 740 ± 200 μmol/l) (Fig. 1). Glycerol concentrations in plasma increased from rest to 30 min of exercise in both groups but significantly more in C (52 ± 11 to 121 ± 27 μmol/l) than in SCI (65 ± 10 to 92 ± 12 μmol/l) subjects (Fig. 1). During exercise, the specific activity of FFA in plasma decreased insignificantly in C, whereas it increased in SCI (Fig. 2). FFA Ra increased in C but decreased with exercise in SCI (Fig.2). As leg blood flow both at rest [0.42 ± 0.05 (C) vs. 0.20 ± 0.02 (SCI) l · min−1 · leg−1] and during exercise [4.28 ± 0.48 (C) vs. 2.34 ± 0.16 (SCI) l · min−1 · leg−1] was higher in control individuals, FFA delivery to the legs was always two- to threefold higher in C [from 155 ± 42 (rest) to 1,073 ± 343 (15 min exercise) and 1,783 ± 625 (30 min exercise) μmol · min−1 · leg−1] than in SCI individuals [86 ± 10 (rest) to 647 ± 75 (15 min) and 553 ± 88 (30 min) μmol · min−1 · leg−1]. At rest, fractional uptake of FFA over the legs was significantly lower in SCI than it was in C subjects (14 ± 3 vs. 31 ± 8%). In both groups, fractional uptake decreased significantly with exercise, and it remained lower in SCI than in C subjects (7 ± 3 vs. 9 ± 2% at 15 min and 11 ± 3 vs. 15 ± 6% at 30 min). In C, gross leg FFA uptake increased progressively over the exercise period (from 40 ± 13 to 170 ± 57 μmol · min−1 · leg−1), whereas in SCI subjects a much smaller increase (14 ± 2 to 57 ± 20 μmol · min−1 · leg−1) was seen (Fig. 3).
The arterial plasma glucose concentration remained unchanged over the exercise period in C subjects but decreased significantly in SCI subjects (Table 1). Correspondingly, leg glucose uptake was markedly higher in SCI than in C (202 ± 35 vs. 71 ± 27 μmol · min−1 · leg−1) subjects during exercise (Fig. 3). Muscle glycogen breakdown also was significantly higher in SCI than in C (Table2), and muscle lactate concentration, leg lactate release, and plasma lactate all rose significantly more in SCI than in C subjects (Tables 1 and 2). Creatine phosphate depletion in exercising muscle was significant only in SCI subjects (Table 2).
Plasma insulin and C-peptide concentrations decreased during exercise in control subjects but remained constant in SCI subjects (Table3). Norepinephrine concentrations increased in both groups (Table 3), whereas epinephrine concentrations did not change significantly. The numeric increase in mean epinephrine value in SCI subjects reflected a significant increase in paraplegic but not in tetraplegic subjects (Table 3). The GH concentration in plasma increased significantly with exercise in SCI subjects, whereas there was no increase in C subjects (Table 3). Levels of glucagon, ACTH, and cortisol in plasma never changed during exercise (Table 3). Hematocrit increased similarly from rest to 30 min of exercise in SCI (44 ± 1 to 47 ± 1%) and in C (45 ± 1 to 47 ± 1%) subjects.
The major findings of the present study are that during involuntary electrically stimulated leg exercise, lipolysis (as indicated by rate of appearance of FFA in plasma and increase in plasma glycerol concentration) as well as leg fractional and overall uptake of FFA are markedly diminished in SCI individuals compared with healthy C individuals. These results add to previous studies with different protocols or simpler methodology, which also indicated that neural activity in motor centers and afferent nerves from working muscle is important for the increase in lipolysis during exercise (15-17, 29, 30). Apparently, blood-borne mechanisms are not sufficient to elicit a normal increase in FFA mobilization during exercise (14, 18,19). In addition, the study has further underlined the existence of an interplay between carbohydrate and lipid metabolism in exercising muscle. Thus in SCI compared with C subjects, a lower leg FFA delivery is accompanied by higher muscle glycogenolysis and glucose uptake, while the higher carbohydrate metabolism is associated with a reduced FFA fractional extraction.
An impaired sympathetic nervous activity both to muscle and skin and an impaired release of epinephrine from the adrenal medulla have been demonstrated in individuals with a high cervical spinal cord lesion (22, 27). Sympathoadrenal β-adrenergic stimulation is considered to be important for promoting lipolysis during exercise, a view supported by studies using the microdialysis technique in subcutaneous fat tissue (2). Furthermore, in both humans and dogs, local sympathetic nerve activity has been shown to directly enhance splanchnic lipolysis during exercise (13, 31). In the present study, however, the diminished lipolysis during exercise in SCI individuals could not be ascribed to a lower direct effect of sympathetic activity as judged from plasma catecholamine levels. Thus lipolysis was decreased both in SCI subjects with a cervical lesion and impaired catecholamine responses and in SCI subjects with a thoracic lesion and well-preserved catecholamine responses. In contrast to C individuals, however, during exercise neither subgroup of SCI individuals showed any decrease in levels of plasma insulin and C-peptide, which is cosecreted with insulin but not extracted in the liver. This finding is in line with the view that during exercise neural activity in motor centers and afferent nerves from working muscle promotes lipolysis partly by eliciting a decrease in insulin secretion (3, 11, 31).
The present study does not support the proposal that during exercise FFA mobilization from adipose tissue is secondary to FFA uptake in working muscle, being directly regulated by negative feedback from plasma FFA concentrations (21, 23). Thus, despite an increase in leg FFA uptake from rest to exercise that was accompanied by a decrease in plasma FFA concentration, the FFA Ranevertheless decreased in SCI individuals. Furthermore, the FFA Ra was lower in SCI than in C individuals in whom FFA appearance rose in the face of FFA concentrations that did not change from rest to exercise. The study also does not point to an important role of indirect feedback on lipolysis from the plasma glucose concentration. During exercise, a decrease in glucose concentration was accompanied in SCI individuals by a marked increase in GH concentration in plasma. These variables did not change in C individuals, and concentrations of other counterregulatory hormones did not change in any of the groups. Obviously, the marked GH response in SCI individuals did not compensate for the lacking effect on lipolysis of somatic neural mechanisms in these subjects.
The fact that during exercise gross leg FFA uptake was lower in SCI compared with C individuals probably reflects that leg muscle FFA uptake depends on leg FFA delivery (9), which was lower in the former than in the latter individuals. In turn, the lower FFA uptake in SCI individuals probably promoted glycogenolysis and glucose uptake in exercising muscle compared with C subjects (7). Although leg fractional FFA uptake would be expected to vary inversely with FFA delivery, it nevertheless was lower in SCI than in C individuals during exercise. Probably, the lower fractional FFA uptake in SCI individuals was in part secondary to the higher muscle carbohydrate oxidation in these compared with C subjects. An inhibitory effect of carbohydrate oxidation on fat metabolism in muscle has previously been inferred from studies showing that increasing carbohydrate availability increases carbohydrate utilization and reduces fat oxidation (5). Although the exact mechanism is not known, it has been suggested that the balance between two of the key regulatory enzymes responsible for the choice of substrate, pyruvate dehydrogenase and carnitine palmitoyl transferase, plays an important role (26). Another explanation for the lower leg FFA fractional uptake during exercise in SCI compared with C individuals is the fact that their leg muscles are less trained, contain more type II fibers (1), and, accordingly, have a lower capacity for fat oxidation. Furthermore, the fact that SCI individuals, due to muscle atrophy, exercised at a higher relative work load than C, probably favored carbohydrate relative to fat metabolism in SCI vs. C individuals.
In conclusion, electrically induced exercise results in a lower mobilization, leg delivery, and fractional uptake of FFA in SCI individuals compared with responses to voluntary exercise in healthy individuals. By contrast, carbohydrate oxidation, muscle glycogenolysis, and leg glucose uptake are higher in SCI than in healthy individuals. These findings indicate that blood-borne mechanisms alone, including direct feedback on lipolysis by decreasing plasma FFA levels, are not sufficient to elicit a normal increase in FFA mobilization during exercise. Furthermore, they are in line with the view that in exercising muscle, FFA delivery enhances FFA uptake and inhibits carbohydrate metabolism while, conversely, carbohydrate oxidation inhibits FFA uptake.
The demonstration of a reduced fat metabolism in the absence of intact neural mechanisms during exercise in humans underlines the importance of somatic afferent and efferent neural activity for regulation of lipolysis during physical activity. It is evident that blood-borne signaling by either circulating FFA or glucose levels is not sufficient for optimal metabolic control during exercise. The study also illustrates that substrate availability is an important determinant of fuel selection in exercise and that FFA and carbohydrate may substitute for each other as energy sources. However, the exact mechanisms regulating this interplay still remain to be elucidated.
From a more practical point of view, the study has shown that SCI individuals depend largely on carbohydrate for energy delivery during electrically induced exercise with their paralyzed muscles. This may potentially have implications with regard to limitations in their exercise endurance. The finding suggests a need for external carbohydrate supplementation for optimizing fuel delivery and endurance performance in SCI individuals treated with electrical stimulation. In contrast, it may be prudent to avoid extra lipid intake in these patients, because fat storage may be favored relative to combustion, resulting in obesity.
The authors thank L. Kall, P. T. Christensen, and R. Kraunsøe for excellent technical assistance.
Financial support from Danish Medical Research Council (9802626), Danish National Research Foundation (504), Team Denmark's Research Foundation, Danish Sports Research Council, and NOVO Nordisk foundation is gratefully acknowledged.
Address for reprint requests and other correspondence: M. Kjær, Sports Medicine Research Unit, Dept. of Rheumatology H, Bispebjerg Hospital, Bispebjerg Bakke 23, DK-2400 Copenhagen NV, Denmark (E-mail:).
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. Section 1734 solely to indicate this fact.
- Copyright © 2001 the American Physiological Society