Muscle glycogen concentration Table 1 As a result of the exercise and dietary manipulation, preexercise muscle glycogen concentration before both IExshort and IExlong was signi®cantly lo[r]
(1)Acta Physiol Scand 1999, 165, 337±345 High-intensity exercise and muscle glycogen availability in humans È DERLUND and B EKBLOM P.D BALSOM, G.C GAITANOS, K SO Department of Physiology and Pharmacology, Karolinska Institute and University College of Physical Education and Sports, Stockholm, Sweden ABSTRACT This study investigated the effects of muscle glycogen availability on performance and selected physiological and metabolic responses during high-intensity intermittent exercise Seven male subjects completed a regimen of exercise and dietary intake (48 h) to either lower and keep low (LOW-CHO) or lower and then increase (HIGH-CHO) muscle glycogen stores, on two separate occasions at least a week apart On each occasion the subjects completed a short-term (<10 min) and prolonged (>30 min) intermittent exercise (IEX) protocol, 24 h apart, which consisted of 6-s bouts of high-intensity exercise performed at 30-s intervals on a cycle ergometer Glycogen concentration (mean SEM) in m vastus lateralis before both IExshort and IExlong was signi®cantly lower following LOW-CHO [180 (14), 181 (17) mmol kg (dw)±1] compared with HIGH-CHO [397 (35), 540 (25) mmol kg (dw)±1] In both IExshort and IExlong, signi®cantly less work was performed following LOW-CHO compared with HIGH-CHO In IExlong, the number of exercise bouts that could be completed at a pre-determined target exercise intensity increased by 265% from 111 (14) following LOW-CHO to 294 (29) following HIGH-CHO (P < 0.05) At the point of fatigue in IExlong, glycogen concentration was signi®cantly lower with the LOW-CHO compared with HIGH-CHO [58 (25) vs 181 (46) mmol kg (dw)±1, respectively] The plasma concentrations of adrenaline and nor-adrenaline (in IExshort and IExlong), and FFA and glycerol (in IExlong), increased several-fold above resting values with both experimental conditions Oxygen uptake during the exercise periods in IExlong approached 70% of V o2max These results suggest that muscle glycogen availability can affect performance during both short-term and more prolonged high-intensity intermittent exercise and that with repeated exercise periods as short as s, there can be a relatively high aerobic contribution Keywords blood lactate, carbohydrates, catecholamines, diet, FFA, glycerol, glycogen, hypoxanthine, oxygen uptake, performance Received March 1998, accepted 17 November 1998 It is now well established that with prolonged continuous exercise, time to fatigue at a moderate submaximal exercise intensity is related to pre-exercise muscle glycogen concentration (BergstroÈm et al 1967) With high-intensity exercise the relation between the availability of muscle glycogen and performance is less clear This is due, at least in part, to differences in the type, intensity and duration of the exercise tests which have been used and differences in the exercise and dietary regimens which have been implemented to manipulate pre-exercise muscle glycogen concentrations (cf Maughan et al 1997) Furthermore, relatively few studies have included muscle biopsies to determine muscle glycogen concentrations During high-intensity exercise there is a rapid breakdown of muscle glycogen Gaitanos et al (1993) reported that glycogen concentration in m vastus lateralis decreased by 14% (43 mmol kg [dw]±1) after only one 6-s bout of `all-out' exercise performed on a cycle ergometer Therefore, during repeated bouts of short-duration highintensity exercise performed over a prolonged period of time, it could be expected that glycogen availability may become a limiting factor for the ability to sustain a highpower output Indeed, Bangsbo et al (1992b) reported that time to fatigue during an exercise protocol which included repeated 15-s bouts of high-intensity exercise, interspersed with 10-s recovery periods, was prolonged following a high carbohydrate intake Correspondence: Karin SoÈderlund, Karolinska Institute, Department of Physiology and Pharmacology, Box 5626, 114 86 Stockholm, Sweden Ó 1999 Scandinavian Physiological Society 337 (2) High-intensity exercise and muscle glycogen P D Balsom et al Acta Physiol Scand 1999, 165, 337±345 The aim of the current study was to investigate if performance during repeated short-duration bouts of high-intensity exercise, performed over both a short (<10 min) and prolonged (>30 min) duration, was in¯uenced by pre-exercise muscle glycogen concentration maximal exercise, voluntary stretching and two 6-s periods of high-intensity exercise at 50% of the experimental work load Subjects were instructed to refrain from strenuous physical activity during the 24-h period preceding each experiment and not to eat for h prior to performing each exercise protocol M A T E R I AL S AN D ME T HO D S Glycogen-depleting exercise (Exdeplete) Subjects performed continuous submaximal exercise on the cycle ergometer at a work load corresponding to 70% of VO2max for 90 This was followed by four 1-min exercise periods at »110% of VO2max, interspersed with 2-min rest periods, and ten 10-s bouts of high-intensity exercise interspersed with 50-s rest periods Subjects Seven highly motivated physically active male physical education students volunteered to participate in the study The mean (and SD) age, body mass and VO2max of the group was 24.0 (3.7) years, 72.9 (10.8) kg and 4.2 (0.3) L min±1 The study was approved by the Ethics Committee of the Karolinska Institute and the subjects were informed of the test procedures prior to giving their consent to participate Procedures and exercise protocols Subjects performed two high-intensity intermittent exercise protocols 24 h apart, on two separate occasions separated by at least week, on a specially adapted Cardionics Wingate friction-loaded cycle ergometer with rounded handle bars and toe-clips (Cardionics AB, Stockholm, Sweden) The ®rst exercise protocol (IExshort) consisted of ®fteen 6-s bouts of high-intensity exercise, interspersed with 30-s rest periods During each work period subjects were instructed to try to maintain a pedalling frequency of 140 r.p.m This was followed 24 h later by a prolonged intermittent exercise protocol (IExlong) where subjects performed repeated 6-s bouts of high-intensity exercise at a target pedalling frequency of 140 r.p.m., again interspersed with 30-s rest periods, until a prede®ned point of fatigue Twenty-four hours before completing IExshort subjects cycled for h (Exdeplete) to lower glycogen stores, i.e Exdeplete 24 h IExshort 24 h IExlong During the 48 h between Exdeplete and IExlong, a diet low in carbohydrate was consumed (LOW-CHO) on one occasion to maintain low glycogen stores and on the other occasion subjects consumed a high-carbohydrate diet (HIGH-CHO) to increase glycogen stores The order in which the two diets were administered was randomized Before completing Exdeplete for the ®rst time subjects visited the laboratory on at least six occasions to become familiarized with performing high-intensity intermittent exercise On one of these visits, maximal oxygen uptake and oxygen uptake during four 4-min submaximal work loads was measured using a graded continuous exercise test on the cycle ergometer IExshort and IExlong were preceded by a standardized 15-min warm up which consisted of continuous sub338 Short-term intermittent exercise protocol (IExshort) This protocol consisted of ®fteen 6-s bouts of high-intensity exercise, interspersed with 30-s rest periods At the start of each work period, subjects began pedalling with no friction applied to the ¯ywheel When a pedalling frequency of 120 r.p.m was reached (this took less than s) data collection was initiated and the resistance was instantaneously (and automatically via feedback from an on-line computer) added to the ¯ywheel by the lowering of a mechanical lever arm which was held in a raised position with an electro-magnet With the resistance in place, subjects were instructed to try and maintain a pedal frequency (target speed) of 140 r.p.m for s A visual feedback system, consisting of a series of lights on a graded scale, was used to guide subjects to maintain the target speed Pedalling frequency was measured via a photoelectric sensor which was connected on-line to a computer This method has been described in more detail previously (Balsom et al 1993, Balsom 1995) The work load was individually selected for each subject so that in the control condition »8 work periods could be completed (i.e for the entire s) at the target speed This was determined during the habituation visits The mean power output for the ®rst exercise period was »958 W which is more than three times greater than that which, during continuous exercise, would have elicited maximal oxygen uptake Muscle biopsy and venous blood samples were taken at rest and directly after the last work period In addition, a venous blood sample was taken 15 post-exercise to measure peak hypoxanthine accumulation (cf Hellsten 1993) Fingertip blood samples (25 lL) were taken at rest and post-exercise Prolonged intermittent exercise protocol (IExlong) This exercise protocol is represented schematically in Fig Subjects were instructed to complete as many 6-s bouts of high-intensity exercise as possible on the cycle ergometer, at the target speed of 140 r.p.m The exercise periods were performed as described for IExshort but with less friction applied to the ¯ywheel During the ®rst 30 exercise periods, the load was 80% of that used in Ó 1999 Scandinavian Physiological Society (3) Acta Physiol Scand 1999, 165, 337±345 P D Balsom et al High-intensity exercise and muscle glycogen Figure A schematic representation of the prolonged intermittent exercise protocol IExlong IExshort Thereafter, it alternated between 70 and 90% (see Fig 1) The test was terminated at a point of fatigue, de®ned as the point where either mean or end pedalling frequency (mean of last s) decreased to below 135 r.p.m for out of any consecutive work periods Subjects were allowed to drink water ad libitum during this exercise protocol Muscle biopsy samples were taken at rest and directly after the last exercise period Venous blood samples were taken at rest, after every 30th exercise period, and and 15 post-exercise Fingertip blood samples (25 lL) were taken at rest, after the 5th, 10th, 15th and 31st work periods, after every 30th work period thereafter, and post-exercise Oxygen uptake was measured continuously over six work and recovery periods, beginning at the start of the 20th, 50th and 70th work period Three Douglas bags were used on each occasion Bag to collect expired air from all six work periods, Bag to collect expired air from to 15 s of the ®rst three recovery periods and Bag to collect expired air from 15 to 30 s of the last three recovery periods Diet The low carbohydrate diet consisted of 4% CHO (% of total energy intake) and »3000 kcal day±1 The mean fat and protein content was 64 and 32%, respectively Subjects were supplied with a `food box' which contained all the food that they were to consume during the 48-h period between Exdeplete and IExlong They were instructed not to consume any additional food but encouraged to drink water ad libitum During the high-carbohydrate diet, CHO intake accounted for 67% of the total energy intake The total energy intake was also »3000 kcal day)1 and the mean fat and protein content was 20 and 13%, respectively Measurements Muscle biopsies Muscle biopsies were obtained from m vastus lateralis using a Weil Blakesly chonchotome (Wisex, MoÈlndal, Sweden) After local anaesthesia to desensitize the surrounding tissue, a 5±7-mm long incision was made in the skin and underlying muscle fascia Using the conchotome, a piece of muscle weighing »70 mg, was removed (»1 cm under the fascia) It was immediately frozen in liquid nitrogen, stored at ±70 °C and later freeze-dried, Ó 1999 Scandinavian Physiological Society powdered and analysed for glycogen using a method adapted from Bergmeyer (1970) as described by Harris et al (1974) In each experiment, biopsies were taken from different sites in the same leg Blood lactate Fingertip blood was haemolysed in a buffer solution (YSI 2357) of : dilution containing triton (5 g L±1) and stored at ±20 °C Whole blood lactate concentrations were measured enzymatically using a YSI 2300GL lactate analyser (Yellow Springs Instruments, Ohio, USA) as described by Foxdal et al (1992) Plasma hypoxanthine, adrenaline, nor-adrenaline, glycerol and FFA Venous blood samples (»5 mL) were drawn from a catheter, inserted in a super®cial forearm vein, using heparinized syringes EGTA (30 lL per 1.5 mL) was added to the samples which were to be analysed for FFA to ensure precipitation with Ca2+ Blood was directly chilled and centrifuged Plasma was stored at ±20 °C Plasma hypoxanthine concentration was analysed using high-performance liquid chromatography (Bioanalytical Systems, Indiana) with a method modi®ed from Wung & Howell (1980), as described by Hellsten (1993) Adrenaline and nor-adrenaline were analysed by high-performance liquid chromatography (Bioanalytical Systems, Indiana) with a method modi®ed from Hjemdahl et al (1979) Glycerol was analysed using a ¯uorometric method as described by Lowry & Passonneau (1973) FFA concentrations were analysed using an enzymatic method modi®ed from Shimizu et al (1979) as described by (Kiens et al 1993) Plasma hypoxanthine concentration was measured from the blood samples taken 15 post-exercise whereas the remaining measurements were made on the blood samples taken within after the cessation of exercise Oxygen uptake Oxygen uptake was measured using the Douglas bag technique The volume of expired air in each Douglas bag was measured using a Tissot spirometer (WE Collins, MA, USA) Fractions of oxygen and carbon dioxide were determined using a Beckman S-3A and LB-2 gas analyser, respectively (Beckman Instruments, Fullerton, USA) The gas analysers were calibrated with gases containing 16.04% oxygen and 3.85% carbon dioxide (as checked by AGA Gas AB, Sundbyberg, Sweden) 339 (4) P D Balsom et al Acta Physiol Scand 1999, 165, 337±345 Statistics The following results are presented as means and SEM unless otherwise stated Data between the two experimental conditions were compared using a one-way ANOVA for correlated means Statistical signi®cance was accepted at the 0.05 level fatigue, the resistance alternated between 70% (4.8 kg ± 672 W) and 90% (6.1 kg ± 854 W) The number of 6-s exercise bouts completed at the target speed in LOWCHO was 111 (14) This increased by 265% to 294 (29) with HIGH-CHO (P < 0.05) The corresponding overall performance times (i.e work + recovery) were 67 and 178 min, respectively R E S U LT S Blood lactate and plasma hypoxanthine, adrenaline, nor-adrenaline, FFA and glycerol (Table 1) Values for both IExshort and IExlong are presented in Table For each metabolite, the post- or peak-exercise concentrations were signi®cantly higher than the preexercise values (as the concentration of the catecholamines in plasma has been shown to fall rapidly on cessation of exercise (cf Kjaer 1992) peak values are presented in preference to post-exercise values) As can be seen in Table 1, no signi®cant differences were found in post-exercise FFA and glycerol values, between LOWand HIGH-CHO When comparisons were made from the measurements made at regular intervals during exercise, however, it was observed that at any given time point, FFA and glycerol values were higher (for all seven subjects) in the LOW-CHO experimental condition Also as can be seen in Table no signi®cant differences were found in post-exercise blood lactate concentrations When comparisons were made from the ®ngertip blood samples taken at regular intervals during exercise, individual but nonsigni®cant variations in blood lactate concentrations were observed High-intensity exercise and muscle glycogen Muscle glycogen concentration (Table 1) As a result of the exercise and dietary manipulation, preexercise muscle glycogen concentration before both IExshort and IExlong was signi®cantly lower following the exercise and diet regimen which included the low carbohydrate content (LOW-CHO) than following the exercise and diet regimen which included the high carbohydrate content (HIGH-CHO) During the 15 sprints in IExshort, the glycogen breakdown was 53 (14) mmol kg (dw)±1 (from 180 (14) to 127 (22) mmol kg (dw)±1 P < 0.05) The corresponding value following HIGH-CHO was 78 (29) mmol kg (dw)±1 (from 397 (35) to 319 (15) mmol kg (dw)±1, P < 0.05) The difference in glycogen breakdown between the two experimental conditions was not signi®cantly different In both IExshort and IExlong, postexercise glycogen concentration was signi®cantly lower with LOW-CHO compared with HIGH-CHO Performance IExshort (Fig 2) The resistance applied to the ¯ywheel was individually determined for each subject (see Methods) The mean of the seven subjects was 6.8 (0.6) kg, which at the target speed of 140 r.p.m corresponded to a power output of 958 (90) W For the ®rst s of each of the 15 exercise periods, there was no decrease in pedalling frequency in either of the experimental conditions The mean pedalling frequency for these s in LOW- vs HIGH-CHO was 139.4 (0.7) and 139.1 (1.3) r.p.m., respectively (P > 0.05) As can be seen in Fig 2, subjects were not able to maintain the target speed during the last s over all of the 15 exercise bouts The decline in pedalling frequency was, however, signi®cantly greater over the last four bouts in the LOW-CHO condition compared with HIGH-CHO (P < 0.05) IExlong (Fig 3) The resistance applied to the ¯ywheel for each subject was expressed as a percentage of the resistance used in IExshort which, for the ®rst 30 exercise periods, was 80% For the seven subjects, this represented a mean load of 5.6 kg (i.e 80% of 6.8 kg) which, at the target speed of 140 r.p.m corresponded to a power output of 784 (71) W For the remaining work periods until the point of 340 Oxygen uptake and RER Values measured during IExlong are shown in Table It can be seen that oxygen uptake during the 6-s exercise Table Muscle glycogen and blood metabolites for the intermittent exercise protocols IExshort and IExlong for the two experimental conditions (values are means and SEM, n = 7) LOW-CHO Post or Peakpk Pre Pre IExshort Glycogen b Lactate c Hypoxanthine d Adrenaline d Nor-adrenaline IExlong Glycogen Lactate Hypoxanthine Adrenaline Nor-adrenaline e Glycerol e FFA a 180 1.3 5.8 0.36 2.8 HIGH-CHO (14) (0.1) (0.5) (0.1) (0.6) 181 (17) 1.2 (0.2) 6.3 (0.8) 0.70 (0.3) 3.1 (0.7) 120 (17) 604 (62) 127 9.7 25.4 2.38 23.1 Post or Peakpk (22) (0.9) (4.0) (0.4)pk (5.3)pk 397 1.5 4.9 0.22 2.9 (35)* 319 (15)* (0.2) 10.5 (1.1) (0.7) 18.9 (3.8) (0.0) 1.59 (0.5)pk (0.6) 17.9 (4.1)pk 64 (22) 4.3 (0.6) 14.0 (1.4) 3.57 (0.5)pk 21.8 (0.9)pk 653 (69) 1669 (251) 540 1.7 4.5 0.26 2.5 83 215 (25)* 151 (48)* (0.2) 4.1 (0.4) (0.6) 13.7 (1.2) (0.0) 4.33 (0.5)pk (0.3) 20.9 (1.8)pk (12) 576 (58) (60)* 1852 (68) a mmol kg (dw))1 bmmol L)1 celmol L)1 dnmol L)1 *HIGH-CHO signi®cantly different from LOW-CHO Ó 1999 Scandinavian Physiological Society (5) Acta Physiol Scand 1999, 165, 337±345 P D Balsom et al High-intensity exercise and muscle glycogen Figure End pedalling frequency (mean and SEM of last s, n 7) for the LOW-CHO (j) and HIGH-CHO ( ) experimental conditions * signi®cantly different from HIGH-CHO periods was signi®cantly higher than during the subsequent recovery periods No signi®cant differences in any of the measurements were found between the two experimental conditions (i.e LOW- and HIGH-CHO) The mean maximal oxygen uptake of the group was 4.2 (0.1) L min±1 RER values measured at the ®xed time points during both exercise and recovery were signi®cantly higher during HIGH-CHO compared with during LOW-CHO DISCUSSION The main ®nding from the current study was that following the exercise and dietary regimen which included the high carbohydrate content (HIGH-CHO) subjects were, compared with the exercise and dietary regimen which included the low carbohydrate content (LOWCHO), better able to maintain a high-power output over the ®fteen 6-s bouts in IExshort and able to perform an Figure Relationship between pre-exercise muscle glycogen con- centrations (m vastus lateralis) and time to fatigue (sum of work and rest periods) for each of the seven subjects during the prolonged intermittent exercise protocol IExlong average of 183 more sprints in (IExlong) where standardized 6-s bouts of high-intensity exercise were repeated until fatigue It is well established that the concentration of glycogen in skeletal muscle can be manipulated by changes in the carbohydrate content of the diet and/or depleting exercise (BergstroÈm et al 1967) In the current study, the standardized exercise protocol used to lower muscle glycogen stores included a 2-h period of ±1 Table Oxygen uptake (L ) and RER values measured during IExlong (see Methods for sampling procedures) for the two experimental conditions (means and SEM, n = 7) LOW-CHO HIGH-CHO Exercise Recovery Exercise Recovery 0±6 s 0±15 s 15±30 s 0±6 s 0±15 s 15±30 s VO2 (bouts) 20±25 50±55 80±85 à 3.3 (0.2) 3.4 (0.1) 3.3 (0.2) 2.4 (0.2) 2.5 (0.1) 2.4 (0.1) 2.6 (0.1) 2.6 (0.1) 2.6 (0.1) 3.0 (0.2) 3.2 (0.3) 3.3 (0.3) 2.4 (0.1) 2.4 (0.1) 2.3 (0.2) 2.5 (0.1) 2.5 (0.1) 2.5 (0.1) RER 20±25 50±55 80±85 à 0.89 (0.01) 0.86 (0.02) 0.87 (0.01) 0.80 (0.01) 0.80 (0.01) 0.80 (0.01) 0.77 (0.02) 0.78 (0.02) 0.75 (0.01) 1.05 (0.02) * 1.03 (0.02) * 1.01 (0.02) * 0.91 (0.02) * 0.92 (0.01) * 0.91 (0.01) * 0.89 (0.02) * 0.90 (0.02) * 0.89 (0.02) * àn = *HIGH-CHO signi®cantly different from LOW-CHO signi®cantly different from to s Ó 1999 Scandinavian Physiological Society 341 (6) P D Balsom et al Acta Physiol Scand 1999, 165, 337±345 exercise, which consisted of both submaximal continuous exercise and repeated short-duration bouts of high intensity The aim of this protocol was to ensure that the glycogen concentration in both type I and II ®bres were lowered (Gollnick et al 1973) This was followed by, on one occasion, a diet with a low carbohydrate content to maintain low glycogen concentration and on the other occasion a high carbohydrate content to increase muscle glycogen stores From the results presented in Table 1, it can be seen that the exercise and dietary regimen used in the current study was successful in both lowering and keeping low (LOW-CHO) and lowering and then raising (HIGH-CHO) muscle glycogen concentrations When using exercise and dietary regimens to manipulate glycogen stores, it is important to avoid potential non-glycogen related `side-effects' which may impair subsequent performance during an exercise test protocol (cf Grisdale et al 1990, Maughan et al 1997) The seven subjects who participated in the current study were physical education students who on an average trained more than three times per week Therefore, it was not expected that the depleting exercise per se would have affected performance during the intermittent exercise protocols The two diets administered in the current study were isoenergetic Subjects were given food boxes and menus with clear instructions on the type and quantity of food to eat Therefore, the observed differences in performance could not be attributed to differences in the total energy intake between the two diets In addition, previous studies have shown that reducing the carbohydrate content in the diet to less than 10% of the total energy intake results in metabolic acidosis (Greenhaff et al 1987, 1988) In the current study, the carbohydrate content with LOWCHO was restricted to 4% of the total energy intake The fat and protein content of this diet was 64 and 32% of the total energy intake, respectively It has been suggested that when muscle glycogen concentrations are manipulated with such a low carbohydrate content and such high fat and protein contents, factors other than pre-exercise muscle glycogen concentration, for example, the body's acid±base status, can impair performance (cf Maughan et al 1997) In a recent study by Maughan et al (1997), a group of male subjects were ®rst administered a diet with a low carbohydrate content and then prior to performing a bout of high-intensity exercise administered sodium bicarbonate or sodium citrate to restore normal acid±base status Compared with a control situation, the time to fatigue did not change with the administration of the bicarbonate This led the authors to conclude that the metabolic acidosis that accompanies a low carbohydrate high-protein diet cannot be regarded as the primary cause of fatigue during high-intensity exercise The performance ®ndings in the current study are not easy to compare with the previous ®ndings reported in the literature It is well reported that with prolonged continuous exercise, time to fatigue at a moderate submaximal exercise intensity is affected by pre-exercise muscle glycogen concentration The in¯uence of the availability of muscle glycogen on the ability to sustain a high-power output during intense exercise, is however, not so well de®ned This may be explained, at least in part, by the numerous different exercise and dietary models that have been used in an attempt to investigate this issue (Jacobs et al 1981, Maughan & Poole 1981, Wootton & Williams 1984, Greenhaff et al 1987, Symons & Jacobs 1989, Bangsbo et al 1992a, Snyder et al 1992, Jenkins et al 1993, Nevill et al 1993) It should be noted that muscle glycogen concentrations were only directly measured in one (Symons & Jacobs 1989) of the above-mentioned studies The exercise pattern used in the current study consisted of repeated 6-s bouts of high-intensity exercise This exercise pattern has previously been associated with a rapid utilization of muscle glycogen Gaitanos et al (1993) reported that muscle glycogen concentration decreased from 317 to 201 mmol kg±1 dw after ten 6-s `all-out' sprints on a cycle ergometer It should be emphasized, however, that the exercise intensity used in the current study, although several-fold greater than that which would have elicited VO2max, was not a maximal `all-out' effort from a stationary start as used in the study by Gaitanos and co-workers This difference in exercise is re¯ected by the lower glycogen utilization rates found in the current study (Table 1) In IExshort, the subjects were not able to maintain the target power output for the entire 6-s duration of each exercise period over the 15 trials with either of the dietary conditions They were however, compared with the LOW-CHO, able to maintain a higher power output following the HIGH-CHO diet This suggests, that under the conditions employed in this study, high-intensity exercise performance was affected by the availability of muscle glycogen The mean post-exercise glycogen concentration after the 15 sprints in IExshort following LOW and HIGH-CHO were 128 (23) and 319 (15) mmol kg (dw)±1, respectively It is clear from these values that the muscle was not entirely depleted of glycogen It should be pointed out, however, that no measurements were made at the single ®bre level in the current study and thus information about variations in glycogen concentration within the different ®bre types was not available Almost 25 years ago, Gollnick et al (1973, 1974) reported that at high-exercise intensities glycogen depletion occurred initially in the type II ®bres Thus one may speculate, albeit tentatively and without direct evidence, that the earlier onset of fatigue High-intensity exercise and muscle glycogen 342 Ó 1999 Scandinavian Physiological Society (7) High-intensity exercise and muscle glycogen Acta Physiol Scand 1999, 165, 337±345 P D Balsom et al observed in the LOW-CHO condition can be partly explained by a selective glycogen depletion in type II ®bres (cf FrideÂn et al 1989) This hypothesis assumes, however, that there was a signi®cant contribution from anaerobic glycolysis throughout the 15 exercise periods Gaitanos et al (1993) reported that during the last of ten 6-s bouts of all-out exercise, glycolysis contributed to only 16.1% of the total ATP production As previously mentioned, although the exercise and recovery durations and the mode of exercise used in the current study was the same as that used by Gaitanos and coworkers there were differences in the exercise intensity In IExlong, the subjects were able to complete 111 bouts in LOW-CHO and 294 bouts in HIGH-CHO The resistance applied to the ¯ywheel in IExlong was, on an average only 20% less that used in IExshort where subjects were not able to maintain the target pedalling frequency over 15 exercise bouts in either of the experimental conditions Thus, the effect of decreasing the resistance by only 20% had a marked in¯uence on performance It should be noted that the mean power output in IExlong was 778 W, which is equivalent to »200% of VO2max In IExlong, the time to fatigue following the high-carbohydrate diet was a 2.5-fold greater than following LOWCHO This improvement in performance is in agreement with that reported by Bangsbo et al (1992a) who found that the time to fatigue during running on a motor-driven treadmill with high-intensity intermittent exercise (15-s work periods performed with 10-s intervals) was greater following a high-carbohydrate diet compared with following a low carbohydrate diet Previous reports have identi®ed an important role for lipid oxidation during high-intensity intermittent exercise (cf EsseÂn 1978) In the prolonged intermittent exercise protocol performed in the current study there was a gradual increase in the plasma concentration of FFA and glycerol over time, suggesting that the contribution from lipid oxidation to the total energy demand was also increasing (cf Hagenfeldt & Wahren 1971) This seems to agree with the ®ndings of EsseÂn et al (1977) who reported that during intermittent exercise regulatory mechanisms cause a gradual downregulation of glycolysis and an increase in fat oxidation Although no signi®cant differences were found in postexercise FFA and glycerol values between LOW- and HIGH-CHO, pre-exercise plasma FFA concentrations were signi®cantly higher with LOW-CHO, and at all of the sampling time points during exercise, FFA and glycerol values were higher (for all seven subjects) in LOW-CHO It has been previously reported that increased plasma FFA concentrations can impair glycogen utilization (Costill et al 1977) and glucose uptake (Hargreaves et al 1991) Furthermore, as can be seen in Table 2, RER values measured at the ®xed time points during both exercise and recovery were signi®cantly higher during HIGH-CHO compared with during LOW-CHO This is an indication that, as has been previously shown with moderate intensity continuous exercise (Christensen & Hansen 1939), during LOWCHO there was a greater contribution to the total energy expenditure from fat combustion than compared with during HIGH-CHO In apparent contrast to these ®ndings, however, was the fact that in IExlong no signi®cant differences between the two experimental conditions could be found in glycogen breakdown per exercise bout This latter ®nding is, however, only based on two glycogen measurements (i.e pre- and post-) and therefore is not sensitive enough to account for possible changes over time The mean oxygen uptake during the 6-s work periods was »70% of maximum oxygen uptake (Table 2) This is in agreement with previous ®ndings from our laboratory where we reported that during repeated 40m sprints, with 30-s recovery periods, oxygen uptake during the recovery periods increased to 66% of maximum oxygen uptake These results suggest that during this type of exercise energy produced aerobically is important not only to fuel recovery processes, but also to resynthesize ATP during contractile activity Indeed Gaitanos et al (1993) reported that over ten 6-s `all-out' sprints, performed at 30-s intervals, there was a decrease in the contribution from anaerobic metabolism to the total energy production and a subsequent shift towards aerobic metabolism The increase in plasma concentration of catecholamines observed in the current study is in agreement with the ®ndings of Brooks et al (1990) who also reported signi®cant increases during repeated 6-s running sprints In the current study no differences were found, between the two experimental conditions, in either peak adrenaline or peak nor-adrenaline concentrations In IExlong, with LOW-CHO however, concentrations of both adrenaline and nor-adrenaline were higher at the sampling time points, compared with HIGH-CHO It is evident that during high-intensity intermittent exercise the secretion of catecholamines plays an important role in both fat and carbohydrate metabolism (cf Kjaer 1992) In conclusion, the results of this study have shown that muscle glycogen availability can affect performance during both short-term and more prolonged high-intensity intermittent exercise with very short-duration exercise periods Furthermore, even with exercise periods as short as s, there can be relatively high demands on aerobic energy turnover when this type of exercise is performed over a prolonged period of time Ó 1999 Scandinavian Physiological Society This study was partly supported by a grant from the Karolinska Institute Research foundation Paul Balsom was supported by a grant from the University College of Physical Education and Sports, Stockholm and the Swedish National Centre for Research in Sports 343 (8) High-intensity exercise and muscle glycogen P D Balsom et al The authors' thanks are extended to all co-workers at Physiology III for their technical support; Kost och NaÈringsdata AB, Bromma, for help with the nutritional analysis programme and Cardionics AB & TenFour Sweden AB, for help with the cycle ergometer R E FE RE N CE S Balsom, P.D 1995 High Intensity Intermittent Exercise ± Performance and metabolic responses with very high intensity short duration work periods Medical Dissertation from the Karolinska Institute, Stockholm Sweden Balsom, P.D., Ekblom, B., SoÈderlund, K., SjoÈdin, B & Hultman, E 1993 Creatine supplementation and dynamic high-intensity intermittent exercise Scand J Med Sci Sports 3, 143±149 Bangsbo, J., Graham, T.E., Kiens, B & Saltin, B 1992a Elevated muscle glycogen and anaerobic energy production during exhaustive exercise in man J Physiol 451, 205±227 Bangsbo, J., Nùrregaard, L & Thorsùe, F 1992b The effect of carbohydrate diet on intermittent exercise performance Int J Sports Med 13, 152±157 Bergmeyer, U 1970 Methoden der Enzymatischen Analyse Verlag Chemie, Weinheim BergstroÈm, J., Hermansen, L., Hultman, E & Saltin, B 1967 Diet, muscle glycogen, and physical performance Acta Physiol Scand 71, 140±150 Brooks, S., Nevill, M.E., Meleagros, L et al 1990 The hormonal responses to repetitive brief maximal exercise in humans Eur J Appl Physiol 60, 144±148 Christensen, E.H & Hansen, O 1939 ArbeitsfaÈhigkeit und Ermudung Skand Arch Physiol 81, 161±171 Costill, D.L., Coyle, E., Dalsky, G., Evans, W., Fink, W & Hoopes, D 1977 Effects of elevated plasma FFA and insulin on muscle glycogen usage during exercise J Appl Physiol 43 (4), 695±699 EsseÂn, B 1978 Studies on the regulation of metabolism in human skeletal muscle using intermittent exercise as an experimental model Acta Physiol Scand 454 (Suppl), 7±32 EsseÂn, B., Hagenfeldt, L & Kaijser, L 1977 Utilization of blood-borne and intramuscular substrates during continuous and intermittent exercise in man J Physiol 265, 489±506 Foxdal, P., Bergqvist, Y., Eckerbom, S & Sundhagen, B 1992 Improving lactate analysis with the YSI 2300 GL: hemolizing blood samples makes results comparable with those for deproteinized whole blood Clin Chem 38, 2810± 2814 FrideÂn, J., Seger, J & Ekblom, B 1989 Topographical localization of muscle glycogen: an ultrahistochemical study in the human vastus lateralis Acta Physiol Scand 135, 381±391 Gaitanos, G.C., Williams, C., Boobis, L.H & Brooks, S 1993 Human muscle metabolism during intermittent maximal exercise J Appl Physiol 75, 712±719 Gollnick, P.D., Armstrong, A.B., Sembrowich, W.L., Shepherd, R.E & Saltin, B 1973 Glycogen depletion pattern in human skeletal muscle ®bres after heavy exercise J Appl Physiol 34 (5), 615±618 Gollnick, P.D., Piehl, K & Saltin, B 1974 Selective glycogen depletion pattern in human muscle ®bres after exercise of 344 Acta Physiol Scand 1999, 165, 337±345 varying intensity and at varying pedalling rates J Physiol 241, 45±57 Greenhaff, P.L., Gleeson, M & Maughan, R.J 1987 The effects of dietary manipulation on blood acid-base status and the performance of high intensity exercise Eur J Appl Physiol 56, 331±337 Greenhaff, P.L., Gleeson, M & Maughan, R.J 1988 The effects of diet on muscle pH and metabolism during high intensity exercise Eur J Appl Physiol 57, 531±539 Grisdale, R.K., Jacobs, I & Cafarelli, E 1990 Relative effects of glycogen depletion and previous exercise on muscle force and endurance capacity J Appl Physiol 69 (4), 1276± 1282 Hagenfeldt, L & Wahren, J 1971 Metabolism of free fatty acids and ketone bodies in skeletal muscle In: B Pernow & B Saltin (eds) Muscle Metabolism During Exercise, pp 153± 163 Plenum Press, New York Hargreaves, M., Kiens, B & Richter, E.A 1991 Effects of increased plasma free fatty acid concentrations on muscle metabolism in exercising men J Appl Physiol 70 (1), 194± 201 Harris, R.C., Hultman, E & NordesjoÈ, L.-O 1974 Glycogen, glycolytic intermediates and high-energy phosphates determined in biopsy samples of musculus quadriceps femoris of man at rest Methods and variance of values Scand J Clin Lab Invest 33, 109±120 Hellsten, Y 1993 Xanthine dehydrogenase and purine metabolism in man ± with special reference to exercise Acta Physiol Scand 151 (Suppl), 8±73 Hjemdahl, P., Daleskog, M & Kahan, T 1979 Determination of plasma catecholamines by high performance liquid chromatography with electrochemical detection: Comparison with a radioenzymatic method Life Sci 25, 131±138 Jacobs, I., Kaiser, P & Tesch, P 1981 Muscle strength and fatigue after selective glycogen depletion in human skeletal muscle ®bers Eur J Appl Physiol 46, 47±53 Jenkins, D.G., Palmer, J & Spillman, D 1993 The in¯uence of dietary carbohydrate on performance of supramaximal intermittent exercise Eur J Appl Physiol 67, 309±314 Kiens, B., EsseÂn-Gustavsson, B., Christensen, N J & Saltin, B 1993 Skeletal muscle substrate utilisation during submaximal exercise in man: Effect of endurance training J Physiol 469, 459±478 Kjaer, M 1992 Exercise effects on adrenergic regulation of energy metabolism In: D.R Lamb & C.V Gisol® (eds) Perspectives in Exercise Science and Sports Medicine, Vol 5: Energy Metabolism in Exercise and Sport, pp 345±381 Brown and Benchmark, Dubuque, IA Lowry, O.H & Passonneau, J.V 1973 A ¯exible system of enzymatic analysis Academic Press, New York Maughan, R.J., Greenhaff, P.L., Leiper, J.B., Ball, D., Lambert, C.P & Gleeson, M 1997 Diet composition and performance of high intensity exercise J Sports Sci 15, 265± 275 Maughan, R.J & Poole, D.C 1981 The effects of a glycogenloading regimen on the capacity to perform anaerobic exercise Eur J Appl Physiol 46, 211±219 Ó 1999 Scandinavian Physiological Society (9) High-intensity exercise and muscle glycogen Acta Physiol Scand 1999, 165, 337±345 P D Balsom et al Nevill, M.E., Williams, C., Roper, D., Slater, C & Nevill, A.M 1993 Effect of diet on performance during recovery from intermittent sprint exercise J Sports Sci 11, 119±126 Shimizu, S., Inoue, K., Tani, Y & Yamada, H 1979 Enzymatic microdetermination of serum free fatty acids Anal Biochem 98, 341±345 Snyder, A.C., Moorhead, K., Luedtke, J & Small, M 1992 Carbohydrate consumption prior to repeated bouts of high-intensity exercise Eur J Appl Physiol 66, 141±145 Symons, J.D & Jacobs, I 1989 High-intensity exercise performance is not impaired by low intramuscular glycogen Med Sci Sports Exerc 21, 550±557 Wootton, S.A & Williams, C 1984 In¯uence of carbohydrate status on performance during maximal exercise Int J Sports Med 5, 126±127 Wung, W.E & Howell, S.B 1980 Simultaneous liquid chromatography of 5-¯uoruracil, uridine, hypoxanthine, xanthine, uric acid, allopurinol and oxipurinol in plasma Clin Chem 26, 1704±1708 Ó 1999 Scandinavian Physiological Society 345 (10)