Endocrinology and Metabolism

Myocellular triacylglycerol breakdown in females but not in males during exercise

Charlotte H. Steffensen, Carsten Roepstorff, Marianne Madsen, Bente Kiens

Abstract

The resting content and use of myocellular triacylglycerol (MCTG) during 90 min of submaximal exercise [60% of peak oxygen uptake (V˙o 2 peak)] were studied in 21 eumenorrheic female and 21 male subjects at different training levels [untrained (UT), moderately trained (MT), and endurance trained (END)]. Males and females were matched according to theirV˙o 2 peak expressed relative to lean body mass, physical activity level, and training history. All subjects ingested the same controlled diet for 8 days, and all females were tested in the midfollicular phase of the menstrual cycle. Resting MCTG, measured with the muscle biopsy technique, averaged 48.4 ± 4.2, 48.5 ± 8.4, and 52.2 ± 5.8 mmol/kg dry wt in UT, MT, and END females, respectively, and 34.1 ± 4.9, 31.6 ± 3.3, and 38.4 ± 3.0 mmol/kg dry wt in UT, MT, and END males, respectively (P < 0.001, females vs. males in all groups). Exercise decreased MCTG content in the female subjects by an average of 25%, regardless of training status, whereas in the male groups MCTG content was unaffected by exercise. The arterial plasma insulin concentration was higher (P < 0.05) and the arterial plasma epinephrine concentration was lower (P < 0.05) in the females than in the males at rest and during exercise. MCTG use was correlated to the resting concentration of MCTG (P < 0.001). We conclude that resting content and use of MCTG during exercise are related to gender and furthermore are independent of training status.

  • muscle substrate
  • training
  • triglycerides

it has previously been shown (25) that myocellular triacylglycerol (MCTG) is utilized during the postexercise period. MCTG stores also represent a potentially large energy source during exercise. However, the extent to which MCTG is used during exercise and the possible existence of differences in its use between trained and untrained (UT) subjects are still under debate. In studies (27, 37) using stable isotope techniques combined with indirect calorimetry, it was estimated that MCTG accounted for 20–25% of the oxidative metabolism during submaximal exercise. However, when direct measurements of MCTG concentration in muscle biopsies have been used, some studies (4,19, 34) have found a decrease in MCTG concentration during submaximal exercise, whereas others (1, 21, 22, 25, 44) have observed no change. In all of the above-mentioned studies, only male subjects have participated. Thus it is unknown whether gender differences exist in the utilization of MCTG during exercise. Some studies (18, 45, 46) have shown that females utilize lipids to a greater extent than males during submaximal exercise, but to our knowledge it has not been investigated whether this increased lipid utilization in females is primarily from MCTG or other lipid sources. Other studies (3, 6, 32) have not been able to find gender differences in lipid utilization during exercise. This could be due to differences in training status and exercise mode in the experimental designs. The aim of the present study was therefore to evaluate the contribution of MCTG during prolonged submaximal exercise, performed at the same relative workload, in female and male subjects at different training levels. The present study is part of a larger project evaluating metabolism in females and males during exercise at different training levels. The current study, however, focuses only on the effects of gender and training on MCTG levels and utilization.

MATERIALS AND METHODS

Twenty-one female and twenty-one male subjects were recruited to participate in the study. All subjects were young and healthy (although not screened for family history of type 2 diabetes) and were nonsmokers (Table 1). All subjects 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 conformed to the code of ethics of the World Medical Association (Declaration of Helsinki).

View this table:
Table 1.

Subject and testing characteristics

Preexperimental protocol.

All subjects initially performed an incremental exercise test on a Krogh bicycle ergometer to determine the peak oxygen uptake (V˙o 2 peak). In addition, all subjects filled out a questionnaire and a training log regarding habitual physical activity and training frequency, intensity, and duration as well as competition history to establish these variables. To measure lean body mass (LBM), a three-compartment model (9) was used. Thus total body composition was assessed by hydrostatic weighing (43), the pulmonary residual volume was determined by the oxygen-dilution method (31), and total bone mineral content was measured by dual-energy X-ray absorptiometry (DPX-IQ scanner, software version 4.6.6; Lunar, Madison, WI). These measurements were carried out under strictly controlled conditions. The subjects were studied 3–4 h after the last meal and had defecated and urinated beforehand. Furthermore, the subjects refrained from any physical activities on the day before the test, which was carried out in the midfollicular phase of the menstrual cycle in all the female subjects.

The female subjects were grouped into UT, moderately trained (MT), or endurance-trained (END) groups based on measuredV˙o 2 peak, and their level of physical activity was determined from their questionnaire and training log. Subjects in the UT group did not participate in any regular physical training, but some participated in leisure time physical activity once a week, and others occasionally biked for local transportation. TheirV˙o 2 peak measurements were <45 ml · kg body mass−1 · min−1. MT subjects participated in 2–4 h of regular exercise training per week and had a V˙o 2 peak of 45–55 ml · kg body mass−1 · min−1. Finally, END subjects participated in endurance-type physical training (cross-country skiing, running, rowing, cycling, and swimming) for at least 5–7 h per week, had trained for and participated in competition for at least 2 yr, and had aV˙o 2 peak >55 ml · kg body mass−1 · min−1 (Table 1). All female subjects were eumenorrheic with a normal cycle length of 28–35 days and did not take oral contraceptives.

The groups of male subjects were matched to the respective groups of female subjects according to theirV˙o 2 peak per kilogram of LBM as well as their habitual activity level and training history.

In all subjects, the habitual energy and nutrient intake were determined by a 5-day self-reported dietary record (Table2). All food and beverage intakes were weighed to the accuracy of 1 g and recorded. Subsequently, the energy intake and composition of the habitual diet were calculated by means of a computer database (Dankost 2000, Danish Catering Center, Copenhagen, Denmark). In addition, the expected individual energy intake was determined from the equation for calculation of energy needs provided by the World Health Organization (48). During the 8 days preceding the exercise experiment, each subject consumed a controlled, isoenergetic diet based on the individual food records consisting of 65% of energy (E%) carbohydrate, 20 E% fat and 15 E% protein (Table 2).

View this table:
Table 2.

Composition of habitual and experimental diets

Between three and seven days before the exercise experiment, each subject performed a pretest to determine the workload required to elicit 60% of V˙o 2 peak. The pretest was performed under conditions similar to those of the exercise experiment (after 12 h fasting and after having abstained from any physical activity the day before).

Experimental protocol.

All subjects reported to the laboratory by either bus or car at 8 AM after an overnight fast. All subjects abstained from any physical activity for 2 days before the exercise experiment. After the subjects rested for 30 min in the supine position, resting pulmonary oxygen uptake (V˙o 2) and CO2excretion (V˙co 2) were measured, and the respiratory exchange ratio (RER) was calculated. Thereafter, a catheter was inserted into the femoral artery under local anesthesia by use of aseptic techniques, and the tip was advanced proximally 2 cm above the inguinal ligamentum for blood sampling. After the catheter was inserted, the subjects rested for 60 min in the supine position before a second measurement of resting V˙o 2 andV˙co 2 and subsequent calculation of RER. Resting blood samples were drawn, and a muscle biopsy was obtained from the vastus lateralis muscle with suction under local anesthesia before the subjects started to exercise on a Krogh bicycle ergometer for 90 min. All subjects exercised at the same relative workload (60%V˙o 2 peak). Expired air was collected in Douglas bags every 15 min during exercise for measurement of V˙o 2 andV˙co 2 and calculation of RER. Blood samples were drawn at 15, 30, 60, 75, and 90 min of exercise. Heart rate was monitored throughout the experiment with a heart rate monitor (Vantage XL, Polar Electro). During exercise, subjects were allowed water ad libitum and were cooled by an electric fan. At the termination of exercise, another muscle biopsy was obtained through the same skin incision but with the needle pointing in a different direction.

All exercise experiments in females were carried out in the midfollicular phase of the menstrual cycle (between days 7and 11; mean day was 9.0 ± 0.2, 9.1 ± 0.4, and 9.4 ± 0.6 in UT, MT, and END groups, respectively).

Pulmonary oxygen uptake and RER.

Pulmonary oxygen uptake and RER were measured and calculated, respectively, during rest and exercise from the collection of expired air in the Douglas bags. The air volume in each Douglas bag was measured in a Collins bell spirometer (Tissot principle), and the O2 and CO2 content of the expired air was determined with a paramagnetic (Servomex) and an infrared (Beckman LB-2) system, respectively. Two gases of known composition were used to calibrate both systems regularly.

Muscle analysis.

In the present study, the biopsies were obtained from the same depth of the vastus lateralis muscle to prevent differences in fiber type composition between the biopsies obtained before and at the termination of exercise (30). The biopsy samples were divided in two. One part was frozen in liquid nitrogen within 10–15 s and stored at −80°C for subsequent biochemical analysis. The other part was mounted in embedding medium and frozen in precooled isopentane and then stored at −80°C for subsequent histochemical analysis. Serial transverse sections (10 μm) were cut and stained for myofibrillar ATPase to identify the different fiber types (2). Before biochemical analysis, ∼30 mg wet wt of muscle tissue were freeze dried and dissected free of all visible adipose tissue, connective tissue, and blood with the use of a stereo microscope, leaving the muscle fibers for further analysis. The muscle fibers were mixed, and ∼1 mg dry wt of the pooled fibers was used for measurement of the MCTG concentration according to the method of Kiens and Richter (24). Glycerol from the degraded triacylglycerol was assayed fluorometrically as described previously by Kiens and Richter (24). A coefficient of variation (CV) of 4% for the MCTG concentration was obtained between five samples from the mixed freeze-dried pool of fibers as described above. In contrast, when five small samples of wet muscle were dissected and analyzed separately, a CV of 31% resulted between the samples. As suggested by Wendling et al. (47), the relatively high variability in MCTG concentration between different locations in the muscle is a concern when detecting expected differences of 20–30%. To circumvent this potential problem, two or more biopsies might be obtained. However, the large number of subjects and the consistency in our findings regardless of training status indicate that our findings are not masked by methodological errors. To further avoid the impact of methodological errors as well as day-to-day variation in the analysis, we analyzed muscle tissue from both female and male subjects in one assay run.

Blood analysis.

Blood was sampled in heparinized syringes, immediately transferred to plastic test tubes containing EGTA, and centrifuged. Plasma was immediately frozen at −80°C until further analysis. Blood for the analysis of progesterone and estradiol was sampled in dry syringes and transferred to dry test tubes in which the blood coagulated for a few hours before it was centrifuged. Serum was frozen at −80°C until further analysis. The plasma concentration of insulin was determined using an RIA kit (insulin RIA 100, Pharmacia and Upjohn Diagnostics, Uppsala, Sweden). The plasma concentrations of epinephrine and norepinephrine were also determined using an RIA kit (KatCombi RIA, Immuno-Biological Laboratories, Hamburg, Germany) as were the serum concentrations of progesterone (progesterone 125I RIA, DGR Instruments) and estradiol (estradiol ultrasensitive RIA, DGR Instruments).

Statistical evaluation.

Results are given as means ± SE. A three-way ANOVA with repeated measures for the time factor was used to determine whether variables were influenced by gender, training status, or time as well as to test for a possible interaction between these three factors. For the variables independent of time, a two-way ANOVA was used to determine any influences of gender and training status and a possible interaction between these two factors. Because females and males were not pairwise matched, gender was not considered to be a repeated factor. In the case of a significant main effect of one or more factors, Tukey's post hoc test was used to detect pairwise differences between the means. Correlations were evaluated by means of linear regression analysis (Pearson product moment correlations). In all cases, an α of 0.05 was used as the level of significance.

RESULTS

Diet.

As we intended, the actual experimental diet averaged 65.5 E% carbohydrates, 19 E% fat, and 15.5 E% protein in all groups (Table2). There were no differences in energy intake in any groups for habitual vs. experimental diet. The energy intake was significantly higher in males than in females and in END vs. UT and MT subjects. In UT and MT females and males, the nutrient composition of the habitual diet was similar and slightly but significantly different from the experimental diet. In END females and males, the energy percentages of carbohydrates and fat in the habitual diet were similar to those of the experimental diet. Furthermore, the energy percentage from dietary carbohydrates was higher for END than for UT subjects.

Workload.

The average workloads during the bicycle exercise test were expressed relative to V˙o 2 peak and asV˙o 2 per kilogram of LBM (Table 1). All subjects completed the 90-min bicycle exercise test at a workload corresponding to 59% of V˙o 2 peakFurthermore, at each training level no gender differences were observed in V˙o 2 expressed relative to LBM. However, in females as well as males, the UT, MT, and END groups differed significantly in V˙o 2 expressed relative to LBM (P < 0.01).

RER.

At rest, RER was similar in all groups (0.80 ± 0.02, 0.81 ± 0.02, and 0.79 ± 0.02 in UT, MT, and END females, respectively, and 0.85 ± 0.05, 0.80 ± 0.03, and 0.79 ± 0.01 in UT, MT, and END males, respectively; Fig.1). RER remained constant throughout the exercise period in UT and MT subjects, averaging 0.87 ± 0.02 and 0.89 ± 0.02 in UT females and males, respectively, and 0.87 ± 0.02 and 0.89 ± 0.02 in MT females and males, respectively. However, RER remained constant during the first 60 min of exercise in END subjects (0.90 ± 0.02 and 0.91 ± 0.02 at 60 min in females and males, respectively) and subsequently decreased (P < 0.05) to 0.87 ± 0.02 and 0.88 ± 0.01 at 90 min in females and males, respectively. No gender differences or effects of training status were observed at any time point.

Fig. 1.

Respiratory exchange ratio (RER) at rest and during 90 min of exercise at 60% of peak oxygen uptake (V˙o 2 peak) in untrained (UT), moderately trained (MT), and endurance-trained (END) females (F) and males (M). Values are means ± SE; n = 7/group. * P < 0.001, significantly different from during exercise; † P < 0.05, significantly different from 15, 30, and 60 min in END females and males.

Pulmonary V˙o2.

o 2 remained constant throughout the exercise period, averaging 1.56 ± 0.07, 1.70 ± 0.05, and 2.19 ± 0.11 in UT, MT, and END females, respectively, and 2.17 ± 0.13, 2.31 ± 0.10, and 2.7 ± 0.08 in UT, MT, and END males, respectively. The averageV˙o 2 during exercise was higher in END subjects compared with UT and MT subjects (P < 0.001). Furthermore, the averageV˙o 2 during exercise was lower in females than in males within each training group (P < 0.001).

MCTG.

At rest, the MCTG content in the vastus lateralis muscle was significantly higher in the female subjects vs. the male subjects, regardless of training status (48.4 ± 4.2, 48.5 ± 8.4, and 52.2 ± 5.8 mmol/kg dry wt in UT, MT, and END females, respectively, and 34.1 ± 4.9, 31.6 ± 3.3, and 38.4 ± 3.0 mmol/kg dry wt in UT, MT, and END males, respectively;P < 0.001; Fig. 2). At the termination of exercise, a mean decrease (P < 0.001) of 25% in MCTG content was observed in the female subjects, regardless of training status. However, in the male subjects the MCTG content remained unchanged after vs. before exercise, regardless of training level. Thus a significant gender difference was observed in MCTG utilization during exercise. A positive correlation existed between MCTG content at rest and the degree of MCTG utilization during exercise (r = 0.61, P < 0.001, Fig. 3).

Fig. 2.

Myocellular triacylglycerol (MCTG) concentration in the vastus lateralis muscle before (filled bars) and after (gray bars) 90 min of exercise at 60% of V˙o 2 peak. Values are means ± SE; n = 7/group except for END females where n = 6. dw, Dry wt. * P< 0.001, significantly different from before exercise; † P < 0.01, significantly different from female pre- to postexercise differences within each training level; ‡ P < 0.001, significantly different from females within each training level.

Fig. 3.

The relationship between initial MCTG concentration and MCTG utilization during exercise in female and male groups (UT, MT, and END). Regr line, regression line.

Fiber type distribution.

The fiber type distribution in the vastus lateralis muscle was not different between UT and MT females (Table3). However, in END females the percentage of type I fibers was higher (P < 0.001) and the percentage of type IIB fibers was lower (P < 0.001) compared with UT and MT females. Differences were not observed in the percentage of type IIA fibers.

View this table:
Table 3.

Muscle fiber distribution in vastus lateralis muscle in females and males

In END males the percentage of type I fibers was higher (P < 0.001) and the percentage of type IIB fibers was lower (P < 0.001) compared with UT and MT males. The percentages of type I and IIB fibers did not differ between UT and MT males. No effect of training status was observed in the percentage of type IIA fibers in males.

Gender differences in fiber type distribution were not observed in MT or END groups, whereas UT females had a higher percentage of type I fibers than UT males (P < 0.01).

The area of all fiber types tended to be smaller in END males compared with both UT and MT males (Table 3). Furthermore, the area of the different fiber types was larger in all male groups than in the corresponding female groups (P < 0.01).

The calculated percentage of fiber types relative to fiber area was significantly larger for type I fibers in females compared with males in the UT and END groups but not in the MT group. The calculated percentages of type IIA and IIB fibers relative to area were significantly smaller in all female groups compared with the respective male groups (Table 3). The average percentages of the different fiber types did not differ significantly between the biopsies obtained before and at the termination of exercise (data not shown).

Circulating hormones.

At rest, the arterial plasma concentration of insulin averaged 8.3 ± 0.9, 8.3 ± 1.2, and 5.9 ± 0.9 μU/ml in UT, MT, and END females, respectively, and 7.1 ± 1.2, 6.2 ± 0.7, and 5.9 ± 0.8 μU/ml in UT, MT, and END males, respectively (Fig.4 A). The arterial plasma insulin concentration decreased continuously throughout exercise to 5.4 ± 0.6, 6.2 ± 2.1, and 4.3 ± 1.1 μU/ml at 90 min in UT, MT, and END females, respectively, and 3.2 ± 0.4, 3.1 ± 0.4, and 3.1 ± 0.5 μU/ml at 90 min in UT, MT, and END males, respectively (P < 0.001). The arterial plasma insulin concentration was significantly higher in females than in males at rest and during exercise regardless of training status (P < 0.05).

Fig. 4.

Arterial plasma concentrations of insulin, epinephrine, and norepinephrine at rest and during 90 min of exercise at 60% ofV˙o 2 peak in UT, MT, and END females and males. Values are means ± SE; n = 7/group.A: arterial insulin concentration. * P < 0.05, significant main effect of gender; ‡ P < 0.05, ‡‡‡ P < 0.001, significantly different from rest; † P < 0.05, ††† P< 0.001, significantly different from previous time point.B: arterial epinephrine concentration. * P < 0.05, significant main effect of gender; † P < 0.001, significantly different from rest and 30 and 60 min in UT and MT; ‡ P < 0.01, significantly different from rest in END; § P < 0.01, significantly different from rest and 30 min in END.C: arterial norepinephrine concentration. ‡ P < 0.001, significantly different from rest; † P < 0.05, significantly different from 30 and 60 min.

The resting arterial plasma concentration of epinephrine averaged 0.3 ± 0.1, 0.3 ± .01, and 0.4 ± 0.1 nmol/l in UT, MT, and END females, respectively, and 0.7 ± 0.2, 0.6 ± 0.2, and 0.6 ± 0.2 nmol/l in UT, MT, and END males, respectively (Fig.4 B). In UT and MT subjects, the arterial plasma epinephrine concentration did not change during the first 60 min of exercise but then increased throughout the last 30 min of exercise. In END subjects, the arterial plasma epinephrine concentration remained unchanged until 30 min of exercise but then increased throughout the rest of the exercise period. At 90 min, the arterial plasma epinephrine concentration averaged 1.9 ± 0.4, 2.9 ± 0.8, and 1.9 ± 0.6 nmol/l in UT, MT, and END females, respectively, and 3.6 ± 0.9, 3.9 ± 1.0, and 1.9 ± 0.4 nmol/l in UT, MT, and END males, respectively. The arterial plasma epinephrine concentration was significantly lower in females than in males at rest and during exercise regardless of training status (P < 0.05).

At rest, the arterial plasma concentration of norepinephrine was similar in females and males despite training status (Fig.4 C). The arterial plasma norepinephrine concentration increased significantly in all groups (P < 0.001) after the start of exercise and remained at a constant level until 60 min when it increased further during the last 30 min of exercise (P < 0.05). No gender differences or effects of training status were observed.

The serum concentration of progesterone averaged 0.19 ± 0.07, 0.12 ± 0.05, and 0.20 ± 0.14 ng/ml in UT, MT, and END females, respectively, and 0.23 ± 0.08, 0.26 ± 0.06, and 0.16 ± 0.06 in UT, MT, and END males, respectively. The serum concentration of progesterone was similar in females and males regardless of training status.

The serum concentration of estradiol averaged 49.1 ± 11.8, 57.6 ± 14.7, and 58.0 ± 10.6 pg/ml in UT, MT, and END females, respectively, and 34.8 ± 4.0, 35.7 ± 3.5, and 30.9 ± 2.8 in UT, MT, and END males, respectively. The serum estradiol concentration was significantly higher in females compared with males regardless of training status (P < 0.01).

DISCUSSION

The results of the present study demonstrate significant gender-based differences in resting content and utilization of MCTG during prolonged submaximal exercise at the same relative workload. At rest, the content of MCTG was significantly higher in females compared with males. During exercise, females utilized MCTG, whereas males did not. The observations of a higher resting content and MCTG use during exercise in females vs. males were made under conditions in which several parameters that could potentially affect substrate oxidation, including the pretest diet and the menstrual status and cycle phase of the female subjects, were standardized and carefully controlled.

MCTG content at rest.

It has previously been demonstrated (23) that diet influences the MCTG content in human skeletal muscle, i.e., the consumption of a fat-rich diet increases MCTG content at rest. The higher MCTG content at rest in the female subjects compared with the male subjects in the present study is presumably not ascribed to the diet as all subjects ingested the same carbohydrate-rich diet for 8 days preceding the exercise experiment. Actually, due to their larger energy intake, males consumed a larger absolute amount of fat compared with females. However, when expressed relative to LBM, the fat ingestion was similar in females and males.

To explain the finding of a higher resting MCTG content in females compared with males, we examined the muscle fiber composition, since it has previously been shown (7) in male subjects that type I fibers contain more MCTG than type II fibers. Furthermore, in a group of females and males, it has recently been shown (20) that MCTG content in soleus, tibialis anterior, and tibialis posterior muscles varied consistently with the expected fraction of type I fibers in these muscles. One might also consider the possibility that females have a higher content of MCTG in type I fibers and/or other fiber types than males. To our knowledge, this has not yet been investigated. END females and males had more type I fibers compared with both MT and UT females and males. There was no effect of gender in the percentage of type I fibers in MT and END subjects. However, UT females had a higher percentage of type I fibers than UT males. Previously, a similar fiber type composition in females and males has been found in UT subjects (5, 35, 39) as well as in physical education students (40). Other studies have found that in UT subjects (42), middle-distance runners (5), trained cyclists (11), and subjects representing a wide range of physical activity levels (33), females had a higher percentage of type I fibers than males. At all training levels in the present study, the area of all the fiber types was smaller in females than in males, especially for type II fibers, which is in accordance with previous observations (5, 35, 39, 40, 42). As a consequence, the calculated fiber composition expressed relative to fiber area revealed that type I fibers accounted for a relatively larger area in females than in males. Thus the higher percentage of type I fibers expressed relative to area might partly explain the higher resting content of MCTG in females. This is further supported by the findings in the present study of a modest but significant correlation between the percentage of type I fibers relative to area and the resting concentration of MCTG (P < 0.05,r = 0.39).

In the present study, no effect of training status on MCTG concentrations at rest was observed in females or males. Previous studies in males have reported inconsistent results, with some studies (22, 34) showing an increase in resting MCTG with training and others (19) showing no such increase. The strict dietary control may explain why an effect of training status on the resting MCTG concentrations was not observed in the present study. The experimental diet was a low-fat diet and differed from the diet in our (22) previous study. The possibility that the low-fat diet has concealed a training effect on MCTG concentration in the present study can therefore not be excluded. Another possible explanation of the increase in MCTG concentration with training in our (22) previous study might be that the subjects had exercised the day before the measurements were done. Because MCTG has been shown (25) to be utilized in the postexercise period, the difference in the reported resting concentration between trained and UT muscle (22) might be influenced by different degrees of postexercise MCTG use in trained and UT muscle.

MCTG utilization during exercise.

A major novel finding in the present study was that female subjects, regardless of training status, utilized a significant amount of MCTG during prolonged exercise, whereas male subjects did not. The fact that male subjects did not utilize MCTG during exercise to any measurable extent is in accord with our (22, 25) previous findings as well as other studies (1, 12, 21, 44) but does contrast with some studies (4, 19, 34) in which MCTG utilization was determined by applying the muscle biopsy technique. In recent studies (17, 38) in females in which a combination of isotope tracer technique and indirect calorimetry was used, it was suggested that the additional source of fatty acids oxidized during exercise at the same absolute workload after training (vs. before training) was provided by MCTG. Applying the same indirect methodology, several studies (27, 37) in males have indicated a similar significant utilization of MCTG during exercise. However, recent investigations in our laboratory revealed that combining isotope tracer technique and indirect calorimetry does not provide an accurate measure of MCTG use in males (13, 36) but does in females (36). This suggests that fat utilized during exercise is recruited from different sources in females and males (36).

To our knowledge, this study is the first to compare MCTG utilization in matched females and males at different training levels during prolonged exercise by applying the muscle biopsy technique. Recently, Guo et al. (12) evaluated the kinetics of intramuscular triacylglycerol fatty acids during exercise. Even though both female and male subjects participated in the study (12), a gender comparison was not made. Furthermore, net breakdown of MCTG in the vastus lateralis muscle was not observed during exercise, supporting our finding in male subjects. However, that study (12) revealed simultaneous esterification of plasma fatty acids to MCTG and MCTG hydrolysis when subjects exercised at 45% ofV˙o 2 peak. Whether these events take place concurrently at higher exercise intensities, as in the present study, is not known. If this is the case, however, interpretation of our data might be that the balance between the two processes is displaced toward a higher net breakdown of MCTG in females than in males.

The enzymatic regulation of triacylglycerol breakdown in skeletal muscle is poorly understood. It has been suggested (41) that a hormone-sensitive triacylglycerol lipase (HSL) similar to the adipose tissue HSL might regulate MCTG hydrolysis. Support for this hypothesis was provided when HSL protein and mRNA were detected in rat skeletal muscle by Western and Northern blotting, respectively (15, 16, 28, 29). Recently, Kjær et al. (26) also demonstrated the existence of HSL in human skeletal muscle, and evidence was provided that β-adrenergic stimulation increased the activity of HSL in skeletal muscle cells (26, 28, 29). However, in the present study, the arterial plasma concentration of epinephrine was significantly lower in females than in males during exercise, whereas the arterial plasma concentration of norepinephrine was similar in females and males (45). If HSL is responsible for the hydrolysis of MCTG, it might be speculated that females have a higher HSL activity compared with males due to a higher content of HSL and/or a higher sensitivity of HSL to potential stimulators. A gender difference in the content of HSL might exist, based on the indication of a higher content of HSL in type I fibers (28) and the fact that a greater percentage of type I fibers relative to area was found in females compared with males in the present study. Regarding lipolytic sensitivity to catecholamines, Hellström et al. (14) found, by means of the microdialysis technique, that during exercise, females have a higher release of glycerol and fatty acids from the abdominal adipose tissue than males despite the fact that both genders had similar concentrations of norepinephrine and that females had a lower concentration of epinephrine than males. The sensitivity of lipolytic activity toward these two catecholamines might be higher in females than in males in skeletal muscle tissue as well. It has previously been shown that the initial content of MCTG influences the use of MCTG during both exercise (8) and infusion of norepinephrine at rest (10). In the present study, the gender difference in MCTG utilization during exercise is at least partly explained by the gender difference in resting content of MCTG, because we observed a fair (r = 0.61) and significant (P < 0.001) correlation between these two parameters. Other factors such as sex hormones might, however, also influence MCTG hydrolysis. At rest, the circulating level of progesterone was identical in females and males, whereas the estradiol level was higher and displayed a greater variation in females than in males. Thus it might be speculated that estradiol exerts an influence on the resting MCTG content as well as on the degree of MCTG utilization during exercise in females. However, in the present study, the concentration of estradiol did not correlate with the MCTG content at rest or its utilization during exercise, rendering it unlikely that estradiol is a major regulatory factor acting directly on the storage and breakdown of MCTG in females.

It is controversial whether MCTG hydrolysis increases with training. In studies measuring the concentration of MCTG by the muscle biopsy technique, MCTG hydrolysis has been found to be similar in the trained and UT leg in males (22) and higher in END males (19, 34) than in UT males during exercise at the same absolute workload. However, in the present study, in which exercise was performed at the same relative workload, training status did not have any effect on the degree of MCTG breakdown. A similar finding was reported previously in males by Bergman et al. (1).

In the present study, we observed a gender difference in MCTG utilization at all training levels despite RER being similar during exercise in females and males. Even though we observed that RER decreased during the last 30 min of exercise in END subjects but not in UT and MT subjects, we were not able to observe any increased MCTG use in END vs. UT and MT subjects. The higher lipid oxidation during the last 30 min of exercise in END subjects seemed, however, to be covered primarily by plasma fatty acids, since oxidation of plasma fatty acids increased to a greater extent in END subjects than in MT and UT subjects during the last third of the exercise test (unpublished data).

A gender difference in MCTG utilization despite similar RER during exercise indicates that females and males obtain lipids from different sources during exercise. It has previously been observed (13,22) that fatty acids derived from plasma very low density lipoprotein-triacylglycerol might contribute to the oxidative metabolism. Furthermore, it has been suggested (22, 36) that intermyocellular lipids are possibly mobilized during exercise, and plasma fatty acids are also known to contribute to the oxidative metabolism. Thus a gender difference might also exist in the use of one or more of these three additional lipid sources.

In summary, the present study revealed a higher resting content of MCTG in females than in males despite training level. Furthermore, regardless of training status, females utilized significant amounts of MCTG during prolonged exercise whereas males did not. Training had no effect on the resting MCTG content or the use of MCTG during exercise in either females or males. Females had a lower concentration of epinephrine in plasma than males, so if HSL is responsible for MCTG hydrolysis it might be speculated that a gender difference exists in the concentration of HSL and/or the sensitivity of HSL toward epinephrine. Other factors such as the resting content of MCTG might influence the degree of MCTG hydrolysis as well.

Acknowledgments

We thank I. B. Nielsen, B. Bolmgren, and W. Taagerup for providing skilled technical assistance. We also thank Prof. Erik A. Richter for performing the invasive procedures.

Footnotes

  • This study was supported by Danish National Research Foundation Grant 504-12 and by the Danish Sports Research Council.

  • Address for reprint requests and other correspondence: B. Kiens, Copenhagen Muscle Research Centre, Dept. of Human Physiology, Universitetsparken 13, DK-2100 Copenhagen, Denmark (E-mail:Bkiens{at}aki.ku.dk).

  • 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.

  • 10.1152/ajpendo.00078.2001

REFERENCES

View Abstract