Effects of Dietary Fat Restriction on Endurance Training-induced Metabolic Adaptations in Rat Skeletal Muscle

endur-Abstract: Endurance exercise training enhances muscle fat oxidation while concomitantly reducing carbohydrate (glycogen) utilization during exercise, thereby delaying the onset of fatigue. This study examined the effects of dietary fat restriction on endurance training-induced metabolic adaptations in rat skeletal muscle. Male Sprague-Dawley rats were placed on either a control diet (CON: 19.2% protein, 21.6% fat, and 59.2% carbohydrate as a percentage of total energy) or a fat-restricted diet (FR: 21.5% protein, 2.4% fat, and 76.1% carbohydrate as a percentage of total energy) for 4 wks. Half the rats in each dietary group performed daily 6-h swimming exercise (two 3-h sessions separated by 45 min of rest) on 5 days each wk. Endurance training significantly increased the expression of β-hydroxyacyl CoA dehydrogenase (βHAD), a key enzyme of fat oxidation, and pyruvate dehydrogenase kinase 4 (PDK4), an inhibitory regulator of glycolytic flux, in the skeletal muscle of rats fed the CON diet. However, such endurance training-induced increases in muscle βHAD and PDK4 were partially suppressed by the FR diet, suggesting that a FR diet may diminish the endurance training-induced enhancement of fat oxidation and reduction in glycogen utilization during exercise. We then assessed the muscle glycogen utilization rate during an acute bout of swimming exercise in the trained rats fed either the CON or the FR diet and consequently found that rats fed the FR diet had a significantly higher muscle glycogen utilization rate during exercise compared with rats fed the CON diet. In conclusion, dietary fat restriction may attenuate the endurance training-induced metabolic adaptations in skeletal muscle.

aptations lead to enhanced muscle FAO and diminished glycogen utilization during acute bouts of exercise. This phenomenon is called the glycogen sparing effect, and it plays a key role in delaying the onset of fatigue during acute endurance exercise at a given intensity 4 .
To be successful in endurance sport events such as the marathon, race walking, and road cycling, it is advantageous not only to optimize the training-induced adaptations as mentioned above but also to reduce body weight and body fat mass. Some endurance athletes therefore tend to restrict their dietary intake of fat, which has higher energy density compared with carbohydrates and proteins 5 . However, such dietary fat restriction is likely to attenuate the muscle enzyme adaptations induced by endur-ance exercise training. This is because peroxisome proliferator-activated receptor PPAR β, which is suggested to be involved in the endurance training-induced increase in mitochondrial enzymes 6 and PDK4 7 , is also activated by free fatty acid FFA 8 . It is therefore hypothesized that dietary fat restriction, which is associated with a lower blood FFA concentration, attenuates muscle enzyme adaptations in response to endurance training. This hypothesis is further supported by the previous finding that the partial silencing of PPARβ expression by shRNA resulted in marked attenuation of the exercise training-induced mitochondrial adaptations in rat skeletal muscle 6 . Moreover, another study reported that pharmacologically lowering serum FFA resulted in a 30 decrease in the mRNA content of mitochondrial proteins in human skeletal muscle 9 . Based on these findings, it is more plausible that a substantial reduction in dietary fat intake has negative effects on endurance training-induced muscle enzyme adaptations, particularly the increase in mitochondrial FAO enzymes, and that the muscle glycogen sparing effect during acute exercise is blunted.
The purpose of this study was therefore to determine whether long-term intake of a fat-restricted diet 2.4 of total energies blunts endurance training-induced metabolic enzyme adaptations in rat skeletal muscle. If it does, we then evaluated muscle and liver glycogen utilizations during an acute bout of exercise in trained rats fed a fatrestricted or a control diet.

Experimental animals
Four-week-old male Sprague-Dawley rats CLEA Japan, Tokyo with body weights of 70-90 g were kept in individual cages. The environment was maintained at 22 1 with 50 5 humidity and illumination from 09:00 to 21:00. All animals were treated in accordance with national guidelines for the care and use of laboratory animals Notification of the Prime Minister s Office of Japan . The Animal Experimental Committee of the University of Tokyo approved this experimental protocol approval no. 29-10 .

Experiment 1
In Experiment 1, we evaluated the effects of dietary fat restriction on endurance training-induced metabolic enzyme adaptations in rat skeletal muscle. During an acclimation period of 5 days, the rats were allowed free access to water and a control diet 19.2 protein, 21.6 fat, and 59.2 carbohydrate as a percentage of total energy, 4.17 kcal/g , which was based on the AIN-93G formula 10 with the modification that the macronutrient composition was comparable to that in the diet consumed by typical endurance athletes 11 . All rats were acclimated to the swimming exercise for 10 min a day for 2 days before being divided into two dietary groups, matched for body weight and food efficiency: a group continued on the control diet CON; n 14 and a group fed a fat-restricted diet FR; n 15: 21.5 protein, 2.4 fat, and 76.1 carbohydrate as a percentage of total energy, 3.72 kcal/g , in which 90 of the fat in the CON diet was replaced with carbohydrates. The compositions of the diets are presented in Table 1.
Rats in each dietary group were further divided into either a sedentary group CON-SED or FR-SED, n 7 each or an exercise training group CON-EX, n 7 or FR-EX, n 8 . Rats in the CON-EX and FR-EX groups performed swimming exercise without a load for 6 h two 3-h exercise sessions separated by 45 min of rest on each of 5 days per wk for 4 wks. Seven or eight rats swam simultaneously in a barrel filled to a depth of 45 cm and with an average surface area of 200-230 cm 2 /rat. The temperature of the water was kept at 35 1 during the swimming exercise. This swimming exercise training has been considered the strongest stimulus for inducing metabolic enzyme adaptations in rat skeletal muscle 12,13 . All the rats were allowed ad libitum access to the experimental diets, and their food intake and body weight were recorded every other day.
Before the last training session started, blood samples were collected into capillary tubes from the tail vein. The capillary tubes were then centrifuged at 10,000 rpm for 5 min, and plasma samples were stored at 80 until analysis. At 18 h after the last training session, rats were sacrificed under anesthesia with isoflurane. The epitrochlearis and triceps muscles were dissected out, rapidly frozen in liquid N 2 , and stored at 80 until analysis. Intra-abdominal fat was removed and weighed. For muscle preparation, the epitrochlearis and triceps muscles were chosen because they have been shown to be mainly recruited during swimming exercise in rats. This is evidenced by the occurrence of glycogen depletion in response to a bout of swimming exercise 14 and adaptive increases in glucose transporter GLUT 4 and mitochondrial enzymes 12,13 . Biochemical analyses in the muscle and plasma samples were performed as described below.

Experiment 2
In Experiment 2, we assessed the muscle and liver glycogen utilizations during an acute bout of exercise in trained rats fed either a control diet or a fat-restricted diet. After a 6-day acclimation period, rats were fed either the CON diet n 16 or the FR diet n 16 for 4 wks. All rats performed the 4-wk swimming exercise training as described in Experiment 1. At 18 h after the last training session, the rats in each dietary group were further divided into two groups for the evaluation of the glycogen and triacylglycerol utilizations during an acute bout of exercise. One group was subjected to a 1-h period of swimming exercise with a weight equal to 5 of their body weight tied to their body. Four rats swam simultaneously in a barrel filled to a depth of 45 cm and with an average surface area of 400 cm 2 /rat. Immediately after the acute bout of swimming exercise, the rats were sacrificed under anesthesia with isoflurane. The epitrochlearis and triceps muscles and the liver were removed, rapidly frozen, and stored at 80 until analysis. The other group was sacrificed without performing an acute bout of swimming exercise to serve as a baseline control. The glycogen and triacylglycerol utilizations in the muscles and liver during the acute bout of exercise were estimated by subtracting the concentrations of each of these substrates in each rat that had exercised from the corresponding mean values in the baseline control rats in each dietary group.

Analytical procedures 2.4.1 Sample homogenization
The epitrochlearis muscles were homogenized in an icecold radio-immunoprecipitation assay lysis buffer Merck Millipore, Billerica, MA containing 50 mmol/L Tris-HCl pH 7.4, 150 mmol/L NaCl, 0.25 deoxycholic acid, 1 NP-40, and 1 mmol/L ethylenediaminetetraacetic acid EDTA with protease inhibitor cocktail Sigma-Aldrich, St. Louis, MO . The homogenates were subjected to three freezing-thawing cycles to disrupt intracellular organelles before being rotated end-over-end at 4 for 90 min to solubilize the protein. The homogenized samples were centrifuged at 700 g for 5 min at 4 , after which the supernatants were collected.

Western blot analysis
The protein concentration of the supernatants was determined using a BCA protein assay kit Thermo Fisher Scientific, Waltham, MA . Samples were prepared in a Laemmli buffer consisting of 277.8 mM Tris-HCl, 44.4 w/ v glycerol, 4.4 w/v lithium dodecyl sulfate, and 0.02 w/v bromophenol blue, at a pH of 6.8 Bio-Rad Laboratories, Hercules, CA with dithiothreitol Bio-Rad . The mixture was heated at 95 for 5 min in a heating block. The sample for GLUT4 measurement was prepared without using dithiothreitol or heating. Equal amounts of sample protein were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis 7.5 -10 resolving gels and then transferred to polyvinylidene difluoride membranes Merck Millipore at 200 mA for 90 min. Following this transfer, the membranes were blocked with 5 w/v skim milk in Tris-buffered saline TBS; 20 mmol/L Tris base, 137 mmol/L NaCl, pH 7.6 containing 0.1 Tween 20 TBS-T for 1 h at room temperature. The membranes were incubated overnight with the following primary antibodies at concentrations of 1:500-5,000 at 4 : anti-β-hydroxyacyl CoA dehydrogenase anti-βHAD, 1:500; Proteintech, Rosemont, IL , anti-PDK4 1:1,000; Proteintech , anti-GLUT4 1:1,000; from the laboratory of Dr. John O. Holloszy, Washington University, St. Louis, MO , and anti-δ-aminolevulinic acid synthetase ALAS, 1:5,000; from the laboratory of Dr. John O. Holloszy . The membranes were then incubated at room temperature for 1 h with secondary antibodies anti-mouse IgG or goat anti-rabbit IgG; Jackson ImmunoResearch, West Grove, PA at a dilution of 1:5,000 in TBS-T containing 1 skim milk. Bands visualization was performed using an enhanced chemiluminescence prime reagent GE Healthcare, Chicago, IL and quantified by Image Studio Digits Ver. 5.2; LI-COR Biosciences, Lincoln, NE . The membranes were stained with Ponceau Sigma-Aldrich to verify equal protein loading across lanes 15 . The intensities of immunobands were normalized to the total protein determined by quantifying all Ponceau red-stained bands in the relevant sample lane.

Citrate synthase activity measurement
The triceps muscles were homogenized in 175 mM KCl, 10 mM glutathione, and 2 mM EDTA, pH 7.4. After the homogenates were subjected to three freezing-thawing cycles, citrate synthase activity was measured using Srere s method 16 .

Glycogen measurement
The epitrochlearis muscles and liver were homogenized in 0.3 mol/L perchloric acid. After acid hydrolysis, the glycogen concentration was measured using the enzymatic method described by Lowry and Passonneau 17 .

Triacylglycerol measurement
The triceps muscle samples were homogenized in 0.9 w/v NaCl and extracted with chloroform-methanol 2:1 v/ v as described by Folch et al. 18 , separating the chloroform and methanol-water phases, and then further processed using the method of Denton and Randle 19 with modifications by Frayn and Maycock 20 . The triacylglycerol concentration was then measured using a Triglyceride E-test Wako Kit Fujifilm Wako Pure Chemical, Osaka, Japan .

Plasma substrates and insulin measurement
Plasma glucose, FFA, and triacylglycerol concentrations were measured using kits Glucose C2-Test Wako, NEFA C-Test Wako and Triglyceride E-Test Wako, respectively obtained from Fujifilm Wako Pure Chemical. Plasma insulin was determined with an enzyme-linked immunosorbent assay ELISA kit Rat insulin ELISA kit; Mercodia, Uppsala, Sweden .

Statistical analysis
Data were presented as the mean standard error of the mean SEM . Two-way analysis of variance ANOVA was performed either to determine the effects of exercise training with two levels: with and without swimming exercise training and diet with two levels: the CON diet and FR diet in Experiment 1 or time with two levels: before and after the acute bout of swimming exercise and diet in Experiment 2. The results observed in the Experiment 1 led us to posit a directional hypothesis that glycogen utilization is higher and triacylglycerol utilization is lower in the FR group than the CON group during the acute bout of exercise. In Experiment 2, we thus performed Student s one-tailed t-test to evaluate the directional hypotheses 21,22 . All statistical analyses were performed using BellCurve for Excel software Social Survey Research Information, Tokyo . Statistical significance was defined as p 0.05.

Results
3.1 Experiment 1 3.1.1 Final body weight, body weight gain, total energy intake, food efficiency, and intra-abdominal fat weight No significant interactions between exercise training and diet were observed for final body weight, body weight gain, and intra-abdominal fat weight Table 2 . Main effects of exercise training on these parameters were observed and resulted in significantly lower final body weight, body weight gain, and intra-abdominal fat weight in the EX group compared with the SED group. Moreover, the FR group had significantly lower final body weight, body weight gain, and intra-abdominal fat weight compared with the CON group.
There was no significant interaction between exercise training and diet for total energy intake or food efficiency Table 2 . The total energy intake and food efficiency were significantly lower in the EX group than the SED group. A significant main effect of diet on food efficiency but not on total energy intake was observed and resulted in significantly lower food efficiency in the FR group compared with the CON group. 3.1.2 CS activity and ALAS protein expression in skeletal muscle To clarify the effects of the FR diet on endurance exercise training-induced mitochondrial adaptations in skeletal muscle, we measured the CS activity and protein content of ALAS, which are frequently used as markers of mitochondrial adaptation to endurance exercise training 23,24 . There was no significant interaction between exercise training and diet for either CS activity in triceps or ALAS protein content in epitrochlearis muscle Figs. 1A and B . The CS activity and ALAS protein content in the muscle tissues were found to be significantly higher in the EX group than in the SED group. However, no significant main effect of diet on these enzymes was observed.

βHAD protein expression in epitrochlearis muscle
βHAD is a key enzyme in fatty acid β-oxidation, and the activity of this enzyme in muscle has been shown to be significantly correlated with the FAO rate during exercise 25 . In this study, we therefore measured the βHAD protein content as a marker of FAO capacity in skeletal muscle. No significant interaction between exercise training and diet was observed for βHAD protein content in the epitrochle- Table 2 Total energy intake, body weight, body weight gain, food efficiency, and intra-abdominal fat weight in rats. Values are means ± SEM; n=7 for all groups except FR-EX (n=8). *The main effects of swimming exercise training (exercise), diet, and their interaction (exercise × diet) were analyzed by two-way ANOVA. n.s., not significant.

CON-SED FR-SED CON-EX FR-EX
aris muscle Fig. 2 . The muscle βHAD protein content was significantly higher in the EX group than the SED group. On the other hand, a main effect of diet on βHAD protein expression was also observed, with the FR group having significantly lower βHAD protein expression compared with the CON group.

PDK4 protein expression in epitrochlearis muscle
Previous studies have shown that endurance exercise training decreases glycolytic flux and glucose oxidation in skeletal muscle during exercise via upregulation of the PDK4 content, which phosphorylates and inactivates the pyruvate dehydrogenase complex 3 . Therefore, we evaluated the PDK4 protein content in skeletal muscle, and the results showed that the EX group had significantly higher muscle PDK4 protein content compared with the SED group Fig. 3 . On the other hand, groups fed the FR diet had significantly lower PDK4 protein content compared with the groups fed the CON diet. The interaction between exercise training and diet for the PDK4 protein content in epitrochlearis muscle was not significant.

GLUT4 protein expression in epitrochlearis muscle
Endurance exercise training is the most potent stimulus in terms of enhancing the muscle GLUT4 content, which is highly correlated with glucose transport activity and glycogen concentration in skeletal muscle 26 . In our experiments, no significant interaction between exercise training and diet was observed for the GLUT4 protein content in the epitrochlearis muscle Fig. 4 . The EX group had significantly higher muscle GLUT4 protein content compared with the SED group. Moreover, a significant main effect of  diet on GLUT4 protein content in the epitrochlearis muscle was also observed and resulted in significantly higher GLUT4 protein expression in the FR group compared with the CON group. 3.1.6 Energy substrates and insulin concentrations in plasma The interactions between exercise training and diet for the plasma glucose, FFA, triacylglycerol, and insulin concentrations were not statistically significant Table 3 . The plasma glucose, triacylglycerol, and insulin concentrations but not the FFA concentration were significantly lower in the EX group than the CON group. On the other hand, a significant main effect of diet was observed for the plasma FFA concentration, with the FR group having a significantly lower plasma FFA concentration than the CON group.    Table 3 Plasma glucose, insulin, triacylglycerol, and FFA concentrations in rats.

CON-SED
FR-SED CON-EX FR-EX Values are means ± SEM; n=7 for all groups except FR-EX (n=8). *The main effects of swimming exercise training (exercise), diet, and their interaction (exercise × diet) were analyzed by two-way ANOVA. n.s., not significant; FFA, free fatty acid.
cogen concentration in epitrochlearis muscle was observed Fig. 5A . The muscle glycogen concentration after the acute bout of exercise was significantly lower than that before exercise. A main effect of diet on muscle glycogen was also observed, with the FR group having a significantly higher muscle glycogen concentration than the CON group. The estimated muscle glycogen utilization during the acute bout of swimming exercise was significantly greater in the FR group than the CON group Fig. 5B .
The interaction between time and diet for the glycogen concentration in liver was not significant Fig. 5C . The glycogen concentration in liver after the acute bout of exercise was significantly lower than that before exercise. However, no significant main effect of diet on liver glycogen concentration was observed. There was also no difference in the estimated liver glycogen utilization during the acute bout of exercise between the dietary groups Fig.  5D .
No significant main effect of time, diet, and their interaction were observed for triacylglycerol concentration in triceps muscle Fig. 5E . Although the estimated muscle triacylglycerol utilization during the acute bout of exercise tended to be smaller in the FR group than the CON group, the difference was not statistically significant because of greater variability Fig. 5F .

Discussion
Endurance athletes such as marathon runners, race walkers, and road cyclists strive to reduce their body weight and body fat mass by restricting their daily intake of fat, which has high energy density. In the present investigation, although the FR and CON groups had similar total energy intakes, the FR group showed significantly lower body weight and total intra-abdominal fat mass Table 2 . As a result, the food efficiency of the FR group was significantly lower than that of the CON group, which means that the mechanism by which dietary fat restriction prevents weight gain may be an increase in energy expenditure. A previous study demonstrated that the intake of a high-carbohydrate diet leads to an increase in the conversion of carbohydrates to fat 27 , which is an energetically costly process, converting 25 of the energy content of carbohydrates into heat 28 . Therefore, we speculate that the rats fed the FR diet, in which carbohydrates accounted for 78 of the total energy, had markedly increased fat synthesis from carbohydrates, thereby increasing their energy expenditure.
The main metabolic consequence of the adaptation of skeletal muscle to long-term endurance exercise training is the enhancement of fat oxidation and the concomitant reduction in carbohydrate utilization during exercise. This glycogen sparing effect in skeletal muscle may be attrib-uted to the endurance training-induced increase in muscle mitochondria, particularly FAO enzymes 4 . Consistent with previous findings, endurance swimming exercise training for 4 wks induced significant increases in CS activity and in the expressions of ALAS and βHAD in skeletal muscle Figs. 1 and 2 . It is thus plausible that the muscle glycogen sparing effect occurred during exercise in the EX groups. On the other hand, although no effect of diet was observed for CS activity and ALAS expression in skeletal muscle, the endurance training-induced increase in muscle βHAD, a key enzyme in fatty acid β-oxidation, was partially suppressed by feeding rats the FR diet Fig. 2 . The fact that the dietary fat restriction suppressed βHAD expression but not CS activity or ALAS expression may have been due to the distinct regulatory mechanisms of their expressions. While the exercise-induced expression of CS and ALAS is coordinately regulated by factors such as nuclear respiratory factor NRF -1 and -2 and PPARγ coactivator-1α PGC-1α 24 , βHAD expression is also regulated by the nuclear receptor, PPARβ, which is activated by fatty acids, as well as NRFs and PGC-1α 29 . A previous study demonstrated that increasing serum FFA in rats by feeding them a high-fat diet resulted in the upregulation of PPARβ activity and subsequent increases in FAO enzyme expressions in skeletal muscle 30 . In the present investigation, the FR group had significantly lower plasma FFA levels compared with the CON group Table 3 , which may have resulted in the diminished activation of PPARβ and the attenuation of the endurance training-induced increase in βHAD expression but not in CS activity or ALAS expression in skeletal muscle. Because the activity of βHAD has been shown to be significantly correlated with the FAO rate during exercise 25 , it is plausible that dietary fat restriction negated the endurance training-induced enhancement of the muscle FAO capacity. Previous studies have shown that endurance training induces an increase in the expression of PDK4, an inhibitory regulator of glycolytic flux, in skeletal muscle, which may also contribute to the endurance training-induced muscle glycogen sparing effect during exercise 3 . Whereas our results also showed that endurance training resulted in a significant increase in muscle PDK4 expression Fig. 3 , this increase was partially suppressed by feeding of the FR diet as well. This may also be attributed to the diminished activation of PPARβ by lower plasma FFA levels in the FR group, because PPARβ is also involved in the regulation of PDK4 as well as βHAD expression in skeletal muscle 7 . Our findings that FR diet feeding suppressed the endurance training-induced increases in the expression of βHAD and PDK4 in skeletal muscle led us to hypothesize that dietary fat restriction may negate the endurance training-induced muscle glycogen sparing effect during exercise.
To assess this hypothesis, the glycogen and triacylglycerol utilizations during an acute 1-h bout of swimming ex- The main effects of time, diet, and their interaction were analyzed by two-way ANOVA A, C, and E . Student s one-tailed t-test was used to evaluate the directional hypotheses that glycogen utilization is higher and triacylglycerol utilization is lower in the FR group than the CON group during the acute bout of exercise B, D, and F . CON, control diet group; FR, fat-restricted diet group.
ercise were evaluated in trained rats fed either the CON or the FR diet. As shown in Fig. 5F, although the difference was not statistically significant due to greater variability, the mean value of the estimated muscle triacylglycerol utilization during the endurance exercise seemed to be lower in the FR group than the CON group. Furthermore, the estimated muscle glycogen utilization during the acute bout of exercise was significantly higher in the FR group than the CON group Fig. 5B . These results may support our hypothesis that dietary fat restriction blunts the muscle glycogen sparing effect during exercise. Although the FR group showed a diminished muscle-glycogen sparing effect, the FR group had significantly higher muscle glycogen concentration than the CON group in the Experiment 2 Fig. 5A . It is well documented that the rate-limiting step in muscle glycogen synthesis is the glucose transport across the plasma membrane 31 . In addition, previous studies have shown that muscle glucose transport is highly and positively correlated with GLUT4 content 32 . In the present investigation, the FR group had significantly higher muscle GLUT4 content compared with the CON group Fig. 4 , suggesting that not only the higher carbohydrate intake, but also the elevated muscle GLUT4 expression, in the FR group compared with the CON group may have at least partially contributed to the increase in pre-exercise muscle glycogen concentration in the FR group.
While increased glycogen utilization during exercise is associated with an earlier onset of fatigue, carbohydrates are an energy-efficient fuel, in which the ATP yield per unit of oxygen consumption is 5.2 higher than fat 33 . Therefore, dietary fat restriction-induced adaptations that make it possible to store more glycogen and then utilize more glycogen during exercise might have potential benefits in high-intensity prolonged endurance exercise, where the available oxygen supply will limit the exercise capacity 34 . The fact that East African runners, who have dominated long-distance running with their superior performance, consume a very high-carbohydrate, low-fat diet 70 -80 carbohydrates and 20 fat as total energies , would be consistent with this hypothesis 35,36 .

Conclusion
In conclusion, the present investigation demonstrated that dietary fat restriction may attenuate the endurance training-induced enhancement of the expressions of muscle FAO enzyme and PDK4, thereby negating the glycogen sparing effect during exercise.