Muscle glycogen storage after prolonged exercise: effect of the frequency of carbohydrate feedings13

Muscle glycogen storage after prolonged exercise: effect of the frequency of carbohydrate feedings13 Louise M Burke, Gregory R Collier, Peter G Davis, Peter A Fricker, Andrew J Sanigorski, and Mark Hargreaves ABSTRACT We reported previously that intake of carbohydrate foods with a high glycemic index (GI) produced greater glycogen storage and greaten postprandial glucose and insulin responses during 24 h of postexercise recovery than did intake of low-gi carbohydrate foods. In the present study we examined the importance of the greaten incremental glucose and insulin concentrations on glycogen repletion by comparing intake of large carbohydrate meals ( gorging ) with a pattern of frequent, small, carbohydrate snacks ( nibbling ), which simulates the flattened glucose and insulin responses after low-gi carbohydrate meals. Eight well-trained triathletes [ ± SEM: 25.6 ± 1.5 y of age, weighing 70.2 ± 1.9 kg, and with a maximal oxygen uptake (VO2max) of 4.2 ± 0.2 L/min] undertook an exercise trial (2 h at 75% VO,max followed by four 30-s sprints) to deplete muscle glycogen on two occasions, 1 wk apart. For 24 h after each trial, subjects rested and consumed the same diet composed exclusively of high-gi carbohydrate foods, providing 10 g carbohydrate/kg body mass. The gorging trial provided the food as four large meals of equal carbohydrate content eaten at 0, 4, 8, and 20 h of recovery, whereas in the nibbling trial each of the meals was divided into four snacks and fed at hourly intervals (0-1 1, 20-23 h). However, there was no significant difference in muscle glycogen storage between the two groups over the 24 h (gorging: 74.1 ± 8.0 mmol/kg wet wt; nibbling: 94.5 ± 14.6 mmol/kg wet wt). The results of this study suggest that there is no difference in postexercise glycogen storage over 24 h when a high-carbohydrate diet is fed as small frequent snacks or as large meals, and that a mechanism other than lowered blood glucose and insulin concentrations needs to be sought to explain the reduced rate of glycogen storage after consumption of low-gi carbohydrate foods. Am J C/in Nutr l996;64:1 15-9. KEY WORDS Postexercise glycogen storage, glycemic index, meal frequency INTRODUCTION The rapid recovery of muscle glycogen stones after prolonged exercise is a critical issue for athletes. Many scientists have investigated dietary strategies for optimizing this process (1-8). In a recent study, our group reported on the effect of the glycemic index (GI) of carbohydrate feedings on glycogen storage over 24 h of recovery in muscles depleted of glycogen by prolonged cycling (9). Subjects consumed diets composed predominantly of carbohydrate foods, with one trial featuring foods with a high GI and the other providing foods with a low to moderate GI. The diets in this study were manipulated to provide equal amounts of carbohydrate, fat, protein, and fiber, with the GI difference between the diets being 37 GI units. We reported that high-gi carbohydrate foods produced greaten glycogen storage (106 ± 1 1.7 mmollkg wet wt) and greaten postprandial glucose and insulin responses during 24 h of postexercise recovery than did intake of low-gi carbohydrate foods (glycogen storage = 7 1.5 ± 6.5 mmol/kg wet wt, P = 0.02). Several mechanisms have been suggested to explain the metabolic effects of low-gi diets and should be examined in relation to our observations of lower glycogen storage. Delayed digestion and absorption is the most well-accepted explanation of the effects of low-gi diets ( 10, 1 1 ). It is possible that slower absorption of glucose will retard the presentation of substrate to muscle, and is therefore consistent with our findings of lower glycogen storage. A second mechanism has been reported from studies that have used indirect calorimetry to follow energy metabolism for 6 h after consumption of a low-gi diet. These studies reported greater glucose oxidation after the low-gi diet, possibly as a result of lower fatty acid and triacylglycerol concentrations (12). Finally, studies measuring the absorption of food by indirect methods (eg, breath-hydrogen measurements) or direct techniques (eg, ileostomy output in ileostomates) have reported that a considerable portion of the carbohydrate in low-gi foods is malabsonbed (13, 14). This has metabolic implications in terms of the production of shortchain fatty acids in the colon, and might also be expected to reduce the carbohydrate available for glycogen synthesis. Although this malabsorption is recognized, it has not generally been regarded as the most significant cause of the metabolic effects achieved by low-gi diets ( I 3, 14). Feeding high-gi foods in a series of small snacks mimics the pattern of delayed digestion and absorption expected with I From the Departments of Sports Science and Sports Medicine, Australian Institute of Sport, Canberra; the School of Nutrition and Public Health, Deakin University, Geelong; and the Department of Physiology, University of Melbourne, Parkville, Australia. 2 Supported by Mars Australia and the Australian Sports Commission. 3 Address reprint requests to LM Burke, Department of Medicine, Australian Institute of Sport, P0 Box 176, Belconnen, ACT 2616, Australia. E-mail: lburke@ausport.gov.au. Received August 30, 1995. Accepted for publication March 1 1, 1996. Am J C/in Nutr 1996:64:115-9. Printed in USA. 1996 American Society for Clinical Nutrition 115

116 BURKE ET AL low-gi foods while removing extraneous variables such as nutrient composition of the two meals or dietary treatments (10). The term gorging has been used to describe eating patterns composed of a large bolus, or series of infrequent large meals, whereas a nibbling diet provides the same food or meals as a series of small snacks (15). A nibbling feeding pattern has been shown to produce acute metabolic effects similar to those of low-gi diets: flattened blood glucose and insulin responses and a reduction in plasma fatty acids (15-17). Nibbling compared with gorging is therefore a suitable paradigm to examine the effect of delayed digestion and absorption (and presumably low-gi meals) on glycogen storage in glycogen-depleted athletes. From a practical viewpoint, this study could also provide useful information to athletes recovering from prolonged exercise sessions; is it better to consume large meals on frequent snacks to promote optimum glycogen storage? obtained from the vastus latenalis, 3 cm distal to the first biopsy site (18). Test diets The recovery diets provided a total of 10 g carbohydrate/kg body wt oven the 24 h and were composed almost entirely of high-carbohydrate foods with high GI. Small quantities of other foods (eg, tomato sauce or skim milk) were added to make the meals palatable. In the gorging trial, the diet was consumed in four large meals, each providing 2.5 g carbohydrate/kg body wt, and fed at 0 h (immediately after the muscle biopsy) and at 4, 8, and 20 h. In the nibbling trial, each meal was divided into snacks of equal carbohydrate content and fed at hourly intervals with a 9-h break for sleeping (ie, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 20, 21, 22, 23 h). A description of the gorging diet is provided in Table 1. SUBJECTS AND METHODS Subjects Eight well-trained triathletes [Li ± SEM: 25.6 ± 1.5 y of age, weighing 70.2 ± 1.9 kg, and with a maximal oxygen consumption (VO2max) of 4.2 ± 0.2 L/min] were recruited from a local pool of elite and subelite athletes. They were informed of the risks associated with participation in the study before they provided written consent. The experimental protocol was approved by the Human Ethics Committee of the Australian Sports Experimental Commission. design Subjects reported to the laboratory in the morning after an overnight fast on two occasions 1 wk apart. Subjects were given verbal and written instructions outlining a standardized diet and exercise protocol to be followed during the 48 h before each trial; a sample food plan providing 300-400 g carbohydrate/24 h was provided and no training on strenuous exercise was allowed for 36 h before each trial. Food and activity records were kept by each subject during both 48-h pretrial periods to promote compliance; the record kept before the first trial was checked by investigators and returned to subjects with instructions to duplicate meal plans in the subsequent trial. Once it was established that the pretrial preparation protocol had been met, an exercise bout was undertaken to lower muscle glycogen concentrations, with subjects riding their own bikes mounted on a windtrainer, for 2 h at a heart rate equivalent to =75% VO2max, followed by four 30-s all-out sprints with a 2-mm recovery period. Within 5-10 mm of cessation of exercise, a muscle sample was obtained from the vastus lateralis by percutaneous needle biopsy and quickly frozen in liquid nitrogen. A catheter was positioned in a forearm vein for blood sampling and was kept patent by periodic flushing with 0.9% saline containing a small amount of heparmn (10 000 UIL). Subjects were then monitored during 24 h of recovery while being fed a nibbling or gorging diet. Subjects were assigned to each diet in random order and were required to stay overnight in the laboratory during the recovery period so that diet and exercise could be fully controlled. No exercise or strenuous activity was permitted during this time. Twenty-four hours after starting the recovery diet, a second muscle sample was Measurements Venous blood samples were obtained at the same time points in both trials; corresponding to immediately before each meal and 30 and 60 mm after the meal in the gorging trial, and immediately before and 60 mm after each snack in the nibbling trial. Plasma glucose concentrations were measured with an automatic analyzer (model 705-0013; Hitachi, Japan) using an enzymatic method (Boehninger Mannheim, Mannheim, Germany). Plasma insulin concentrations were measured with a commercially available double-antibody nadioimmunoassay kit (Phadeseph Insulin RIA; Pharmacia Diagnostic AB, Uppsala, Sweden). Plasma triacylglycerol concentrations were measured by using an enzymatic hydrolysis technique with subsequent determination of glycerol by colonimetry (BMlHitachi System 705; Boehringen Mannheim). Muscle samples were analyzed for glycogen content by using an enzymatic, fluorometric technique (19). TABLE 1 Gorging diet fed to subjects during 24 h of recovery after prolonged exercise Meal period Food consumed Meals 1 (t = 0 h) and 4 (t = 20 h) 120 g Cornflakes 500 ml Skim milk 40 g Glucose polymer Meal 2 (t = 4 h) 360 g Whole-meal bread 60 g Tomato 40 g Glucose polym& Meal 3 (t = 8 h) 120 g Mashed instant potato (dry weight) 60 g Tomato sauce 40 g Glucose polymer Total (24 h) 636 g Carbohydrate 21 g Fat 97 g Protein 43 g Fiber 13.18 MI The sample diet is for a 64-kg subject. For the nibbling trial, each meal was divided into four equal parts and one part was eaten each hour. 2 Polycose; Ross Laboratories, Columbus, OH.

POSTEXERCISE GLYCOGEN STORAGE 117 Statistical analysis RESULTS Data from the two trials were compared by using analysis of variance for repeated measures, with significance accepted at the 0.05 level. Specific differences between means were located by using the Newman-Keuls post hoc test. Twenty-fourhour glycogen storage was compared between trials by using paired t tests. All data are reported as means ± SEMs. MINITAB (Minitab Inc, State College, PA) was used for analyses. Pretrial records indicated that all subjects completed the preparation protocol as instructed by refraining from training for 36 h and consuming 300-400 g carbohydnate/d for the 2 d before the trail. Although the same total amount of food was consumed in each trial, differences in metabolic responses were observed between the two dietary patterns. The daily profiles of plasma glucose, insulin, and triacylglycerols are shown in Figure 1. The gorging diet produced large increases 9 8-0-- Nibbling -. Gorging # 7. 6 J4 03060 120 180 03060 120 180 03060 120 180 240 03060 120 180 240 3b0 300. # I 240. 180 120 60. * * 0.. I U 0 30 60 120 180 0 30 60 120 180 0 30 60 120 180 240 0 30 60 120 180 240 1.5. 1.2. 1.0. 0.8. 0.5. 0.2 - - I -- - 03060 120 180 03060 120 180 03060 120 180 240 03060 120 180 240 FIGURE 1. Plasma glucose. insulin, and triacylglycerol profiles over 24 h with a gorging or a nibbling diet in eight well-trained subjects after glycogen-depleting exercise. < 0.01, #P < 0.05.

118 BURKE ET AL in plasma glucose and insulin concentrations after each meal, which returned to baseline values within 60-90 mm. The nibbling diet produced lower plasma glucose and insulin profiles over the day, a generally flatter profile. The plasma triacylglycerol profile over the 24 h was also lower during the nibbling trial. The degree of muscle glycogen depletion was not different in the two trials (Table 2). Glycogen storage after 24 h was not significantly different (Table 2), despite the differences in glucose and insulin profiles between trials (Figune 1). DISCUSSION groups. The results of this study suggest that the frequency of carbohydrate feedings does not influence total glycogen storage over 24 h of recovery from prolonged strenuous exercise. The 25% difference in mean glycogen storage between the diets raised the possibility of a type II statistical error. However, five of the eight subjects had greater glycogen storage with the nibbling regimen, whereas three had higher storage after the gorging diet. The difference between means essentially results from atypical values in two subjects; one had an abnormally high postexercise muscle glycogen concentration in one trial, and consequently low glycogen storage; whereas the other had an abnormally high postrecovery muscle glycogen concentration and a high glycogen storage result. If these subjects were removed from the analysis, mean muscle glycogen storage oven the 24-h period was 75.6 ± 7.0 mmol/kg wet wt for the gorging diet and 77.4 ± 6.6 mmol/kg wet wt for the nibbling diet (n = 6; P = 0.88). However, apart from the atypical muscle glycogen storage, there appears to be nothing that distinguishes these subjects from the others and we have no valid reason to exclude them from the study. Therefore, we accept these results as a reflection of the normal variation in muscle glycogen data resulting from intra- and interassay analytical variation, as well as variability of glycogen deposition within a muscle sample. Within our laboratory we have assessed the CV for the glycogen assay determined on different pieces of muscle from the same biopsy sample to be on the order of 15% (unpublished observations). The findings of this study agree with the work of Costill et al (3), who reported no difference in 24-h muscle glycogen storage when 525 g carbohydrate was consumed by glycogendepleted runners in two meals or seven meals. Skeletal muscle regulates the rate of glycogen synthesis through negative feedback, with a low muscle glycogen concentration itself providing a stimulus for glycogen storage (20). If there had been an increased rate of muscle glycogen storage in the early hours of recovery with one diet, subsequent glycogen storage may have TABLE 2 Muscle glycogen concentration immediately after prolonged exercise (0 h) and after 24 h of recovery with a nibbling or gorging diet Oh 24h mmol/kg wet wt Gorging 55.8 ± 9.0 129.9 ± 10.2 74.1 8.0 Nibbling 50.1 ± 9.6 144.5 ± 12.2 94.5 ± 14.6, i ± SEM; n - 8. There were no significant differences between slowed because of self-regulation, allowing the less replete muscle to catch-up in total storage during the later recovery period. In addition, the amount of carbohydrate provided to the muscle during recovery appears to be the most critical dietary factor in optimal glycogen synthesis (3, 5). The total carbohydrate consumed by the subjects in this study (10 g/kg body wt) is considered to be at or above the threshold for glycogen storage (3); it is possible that meal frequency may have an effect on muscle glycogen storage when total carbohydrate intake is inadequate or below the glycogen storage threshold. The nibbling diet in this study resulted in significantly reduced glucose, insulin, and triacylglycenol concentrations compared with the gorging trial. These acute metabolic effects are similar to those achieved by low-gi diets (15-17). However, the nibbling diet was not associated with a reduced rate of glycogen storage over 24 h, in contrast with the low-gi diet in our previous study (9). Therefore, the effects of low-gi diets on glycogen repletion cannot be explained by delayed absorption, and/or increased total carbohydrate oxidation. Furthermore, these data suggest that reduced blood glucose and insulin concentrations are not critical in glycogen storage, on at least did not reach limiting values in this study. Insulin is a known activator of muscle glycogen storage, both by stimulating uptake of glucose into the muscle cell and by activating glycogen synthase, the rate-limiting step in glycogen storage (20, 21). It has been tempting to explain the greater muscle glycogen storage seen with high-gi diets (9) on the addition of protein to carbohydrate feedings (22) to be due to associated increases in blood glucose and/or insulin concentrations. However, this study supports the findings of previous work (23) that manipulation of blood glucose and insulin concentrations within physiologic ranges is not associated with changes in muscle glycogen storage over longer recovery periods (ie, 24 h), at least when carbohydrate intake is adequate. The malabsonption of low-gi carbohydrate foods provides an alternative mechanism to explain reduced muscle glycogen storage. Estimates of malabsorption from breath-hydrogen (13) and ileostomy (13, 14) studies have reported that 15-20% of the available carbohydrate in low-gi foods such as lentils may escape digestion in the small intestine. These losses appear to be less than the reductions in blood glucose ( 14) or muscle glycogen storage (9) seen with low-gi diets; however, further investigation is warranted. Finally, although the frequency of meals during 24 h of recovery from strenuous exercise does not appear to directly influence muscle glycogen storage, there may be practical implications of meal timing and frequency that need to be considered. The amount of carbohydrate consumed remains a critical dietary factor in determining postexercise glycogen storage (3, 5, 23), and practical issues such as appetite and the availability of food may determine both how often food is consumed, and whether carbohydrate intake goals are met. It is generally considered that appetite is depressed after strenuous exercise and that a pattern of small snacks and/or high carbohydrate drinks may be more appealing than large meals in the immediate postexercise phase (24). Early intake of carbohydrate has been shown to produce a slightly increased rate of glycogen storage (7-8 mmol. kg. h ) in the first 2 h of recovery when compared with the normal nate of glycogen storage (5-6 mmol. kg. h )(4). However, it is considered that the major benefit of consuming carbohydrate in the im-

POSTEXERCISE GLYCOGEN STORAGE 119 mediate recovery phase is to promote earlier recovery rather than exploit a period of increased glycogen storage rate (24). Appetite and satiety factors may also determine total food intake over 24 h. Because many carbohydrate-rich foods are bulky, it can be difficult to achieve intakes of 500-600 g/d in a traditional Western eating style of three meals/d. Indeed, dietary surveys of endurance athletes have noted that high carbohydrate intakes are associated with a grazing style of frequent meals and snacks (25, 26). In summary, the results of this study suggest that large meals were as effective as small snacks in achieving glycogen storage during 24 h of recovery from prolonged exercise, when total carbohydrate intake was controlled and adequate. From a practical viewpoint, athletes are advised to plan a high-carbohydrate menu for postexencise recovery, and to plan the frequency of meals and snacks to suit considerations such as appetite and availability of food. In addition, the study suggests that delayed absorption, causing acute alterations to glucose, insulin, and blood fat profiles, does not adequately explain the reduced rate of glycogen storage seen with low-gi carbohydrate foods, and that an alternative mechanism needs to be investigated. U We thank Susan Beasley for her invaluable help with the biopsy procedures in this study. REFERENCES 1. Blom PSC, Hostmark AT, Vaage 0, Kardel KK, Maehlum S. Effect of different post-exercise sugar diets on the rate of muscle glycogen synthesis. Med Sci Sports Exerc l987;l9:491-6. 2. Brewer I, Williams C, Patton A. The influence of high carbohydrate diets on endurance running performance. Eur I Appi Physiol 1988:57: 698-706. 3. Costill DL, Sherman WM, Fink WJ, Maresh C, Witten M, Miller JM. The role of dietary carbohydrates in muscle glycogen resynthesis after strenuous running. Am I Clin Nuts 1981;34:1831-.6. 4. Ivy IL, Katz AL, Cutler CL, Sherman WM, Coyle EF. Muscle glycogen synthesis after exercise: effect of time of carbohydrate ingestion. I AppI Physiol l988;64: 1480-5. 5. Ivy IL, Lee MC, Brozinick IT, Reed MJ. Muscle glycogen storage after different amounts of carbohydrate ingestion. J AppI Physiol 1988:65:2018-23. 6. Keizer HA, Kuipers H, van Kranenburg G, Guerten P. Influence of liquid and solid meals on muscle glycogen resynthesis, plasma fuel hormone response, and maximal physical working capacity. Int I Sports Med 1986;8:99-l04. 7. Reed MI, Brozinick IT, Lee MC, Ivy IL. Muscle glycogen storage postexercise: effect of mode of carbohydrate administration. I AppI Physiol l989;66:720-6. 8. Roberts KM. Noble EG, Hayden DB, Taylor AW. Simple and complex carbohydrate-rich diets and muscle glycogen content of marathon runners. Eur J Appl Physiol l988;57:70-4. 9. Burke LM, Hargreaves M, Collier GR. Muscle glycogen storage after prolonged exercise: effect of the glycemic index of carbohydrate feedings. I Appl Physiol 1993;74:1019-23. 10. Jenkins DJA, Jenkins AL, Wolever TMS, et al. Low glycemic index: lente carbohydrates and physiological effects of altered food frequency. Am I Clin Nuts l994;59(suppl):706s-9s. 1 1. Wolever TMS. The glycemic index. In: Bourne GH, ed. Aspects of some vitamins, minerals and enzymes in health and disease. World Rev Nuts Diet 1990;62: 120-85. 12. Ritz P. Krempf M, Cloarec D, Champ M, Scharponnel B. Comparative continuous-indirect-calorimetry study of two carbohydrates with different glycemic indices. Am I Clin Nutr 1991 ;54:855-9. 13. Wolever TMS, Cohen Z, Thompson LU, et a!. heal loss of available carbohydrate in man: comparison of a breath hydrogen method with direct measurement using a human ileostomy model. Am I Gastroenterol l986;81:1 15-22. 14. Jenkins DIA, Cuff D, Wolever TMS, et al. Digestibility of carbohydrate foods in an ileostomate: relationship to dietary fibre, in vitro digestibility, and glycemic response. Am I Gastroenterol l987;82: 709-17. 15. Jenkins DJA, WoleverTMS, Vuksan V. et al. Nibbling versus gorging: metabolic advantages of increased meal frequency. N EngI I Med l989;321 :929-34. 16. Wolever TMS. Metabolic effects of continuous feeding. Metabolism 1990;39:947-5 I. 17. Jenkins DJA, Wolever TMS, Ocana AM, et al. Metabolic effects of reducing rate of glucose ingestion by single bolus versus continuous sipping. Diabetes 1990;39:775-8l. 18. Costill DL, Pearson DR. Fink WI. Impaired muscle glycogen storage after muscle biopsy. J Appl Physiol 1988;64:2245-8. 19. Passonneau IV, Lauderdale VR. A comparison of three methods of glycogen measurement in tissues. Anal Biochem l987;60:4o5-12. 20. Danforth WH. Glycogen synthetase activity in skeletal muscle: interconversion of two forms and control of glycogen synthesis. I Biol Chem l985;240:588-93. 21. Cohen P. Nimmo HG, Proud CG. How does insulin stimulate glycogen synthesis? Biochem Soc Symp l979;43:69-95. 22. Zawadzki KM, Yaspelkis BB, Ivy JL. Carbohydrate-protein complex increases the rate of muscle glycogen storage after exercise. I Appl Physiol 1992;72: 1854-9. 23. Burke LM, Collier GR, Beasley SK, et al. Effect of coingestion of fat and protein with carbohydrate feedings on muscle glycogen storage. I Appl Physiol 1995;78:2l87-92. 24. Coyle EF. Timing and method of increased carbohydrate intake to cope with heavy training, competition and recovery. I Sports Sci 1991 ;9(suppl):29-52. 25. Burke LM, Gollan RA, Read RSD. Dietary intakes and food use of groups of elite Australian male athletes. Int I Sport Nuts 1991 ;l :378-94. 26. Kirsch KA, von Ameln H. Feeding patterns of endurance athletes. Eur I AppI Physiol 1981 ;47: 197-208.