Secondary Logo

Journal Logo

Increased Carbohydrate Oxidation after Ingesting Carbohydrate with Added Protein


Medicine & Science in Sports & Exercise: May 2008 - Volume 40 - Issue 5 - p 903-912
doi: 10.1249/MSS.0b013e318164cb52
BASIC SCIENCES: Original Investigations

Purpose: To examine the metabolic impact of including protein in a postexercise carbohydrate supplement when ingested between two bouts of prolonged running performed within the same day.

Methods: Six healthy men participated in two trials separated by 14 d, each involving a 90-min treadmill run at 70% V˙O2max followed by 4 h of recovery and a subsequent 60-min run at the same intensity. At 30-min intervals during recovery, participants ingested either a solution containing 0.8 g·kg−1·h−1 of carbohydrate (CHO) or the same solution plus an additional 0.3 g·kg−1·h−1 of whey protein isolate (CHO-PRO). Muscle biopsies were obtained from the vastus lateralis at the beginning and end of the recovery period, with a third muscle biopsy taken following the second treadmill run.

Results: Despite higher insulinemic responses to the CHO-PRO solution than to the CHO solution (P < 0.05), rates of muscle glycogen resynthesis during recovery were not different between treatments (CHO = 12.3 ± 2.2 and CHO-PRO = 12.1 ± 2.7 mmol glucosyl units per kilogram of dry mass per hour). Furthermore, there were no differences between treatments in muscle glycogen degradation during subsequent exercise (CHO = 2.2 ± 0.3 and CHO-PRO = 2.0 ± 0.1 mmol glucosyl units per kilogram of dry mass per minute). In contrast, whole-body carbohydrate oxidation during the second run was significantly greater with the CHO-PRO treatment than with the CHO treatment (48.4 ± 2.2 and 41.7 ± 2.6 mg·kg−1·min−1, respectively; P < 0.01).

Conclusion: These data show that the inclusion of protein in a carbohydrate-recovery supplement can increase the oxidation of extramuscular carbohydrate sources during subsequent exercise without altering the rate of muscle glycogen degradation.

1Sport and Exercise Science, University of Bath, Bath, UNITED KINGDOM; 2School of Sport and Exercise Sciences, Loughborough University, UNITED KINGDOM; 3Sunderland Royal Hospital, UNITED KINGDOM; and 4School of Biomedical Sciences, Queen's Medical Centre, Nottingham, UNITED KINGDOM

Address for correspondence: James Betts, Applied Physiology Research Group, Sport and Exercise Science, School for Health, University of Bath, United Kingdom, BA2 7AY; E-mail:

Submitted for publication June 2007.

Accepted for publication December 2007.

Carbohydrate ingestion is an effective means of accelerating muscle glycogen resynthesis following exercise, and a general positive correlation exists between the amount of carbohydrate ingested and the amount of muscle glycogen accumulated per hour of recovery (13). This relationship persists up to a carbohydrate intake of approximately 1.2 g·kg−1·h−1, at which point the rate of muscle glycogen synthesis plateaus slightly in excess of 40 mmol·kg−1 dry mass·h−1 (29).

Provided that the above carbohydrate ingestion rate is achieved using foods and/or supplements with a high glycemic index and in accordance with an effective feeding schedule (i.e., carbohydrate availability is adequate), it seems that the coingestion of other macronutrients will not further increase the rate of muscle glycogen storage (8,14,28). Conversely, several studies have reported that postexercise muscle glycogen storage can be facilitated when protein is added to lower quantities of carbohydrate (i.e., ≤ 0.8 g·kg−1·h−1), possibly due to the synergistic influence of carbohydrate and protein on insulin secretion (12,29,33). In fact, recent muscle biopsy evidence indicates that ingesting protein after exercise may be beneficial in terms of glycogen storage even when the presence of that protein actually displaces some carbohydrate from the diet (i.e., isoenergetic supplements (5,12)), although other evidence using 13C-MRS to measure muscle glycogen is not consistent with this effect (23). It should also be recognized that the glycogenic influence of protein when added to suboptimal quantities of carbohydrate has not been a consistent finding. However, the studies that are not consistent with the stated effect (8,25) have provided protein at a lower relative ingestion rate than the studies cited earlier (5,12,29,33) and have not, therefore, substantially augmented insulin concentrations during recovery.

Furthermore, given that substrate metabolism may vary between different types of exercise (27), it is notable that all the studies cited above have employed cycling as an exercise model prior to recovery, and only the study by Berardi et al. (2006) has measured muscle glycogen concentrations during a second exercise bout subsequent to recovery. In that study, muscle glycogen degradation rates were not significantly different between treatments during a cycling test following ingestion of energy matched supplements during recovery, although no measure of whole-body carbohydrate oxidation was reported (5). In the absence of any data pertaining to repeated bouts of treadmill running within the same day, our most recent investigation examined the effect of combined carbohydrate-protein ingestion during recovery from a 90-min run on the capacity for physical exercise 4 h later. Every participant in this study was able to exercise for longer following recovery when additional energy in the form of protein had been included in the postexercise diet, an improvement similar to that observed with an energy-matched increase in carbohydrate ingestion rate (6). It is therefore apparent that increasing the energy content of a postexercise supplement can facilitate recovery, irrespective of whether the additional energy originates from carbohydrate or protein. However, it remains to be established whether the ergogenic benefit of the added protein was due to (i) an acceleration of muscle glycogen resynthesis following the initial treadmill run, as has been reported following prolonged cycling (5,12,29,33); (ii) a reduced rate of muscle glycogen degradation during the subsequent exercise test; or (iii) some other mechanism not directly related to muscle glycogen availability (e.g., improved glycemia and/or increased blood glucose oxidation).

This study was therefore conducted in order to address these issues specifically in relation to two bouts of treadmill running performed within the same day. We hypothesized that ingestion of carbohydrate with added protein during a 4-h recovery from prolonged running would result in a greater rate of muscle glycogen storage than when ingesting a supplement of matched carbohydrate content. To provide additional novel insight, we also examined whether the additional protein would increase the rate of extramuscular carbohydrate oxidation during exercise subsequent to recovery, thus potentially sparing muscle glycogen.

Back to Top | Article Outline



Six healthy male participants volunteered to take part in this study and the characteristics of this cohort were as follows: age 22 ± 1 yr, body mass (BM) 73.8 ± 6.7 kg, and V˙O2max 61 ± 6 mL·kg−1·min−1 (means ± SD). These participants were healthy individuals whose habitual training involved a substantial cardiovascular component. Once fully briefed regarding the nature of the study, each participant provided written informed consent in keeping with the requirements of the Loughborough University ethical advisory committee, which approved this study.

Back to Top | Article Outline

Preliminary measurements.

Preliminary tests were administered to determine participants' submaximal and maximal oxygen uptakes (26) on a motorized treadmill (Technogym, Gambettola, Italy). A subsequent test was conducted within 14 d of main trials, both to familiarize participants with the study procedures and to confirm that the calculated running speeds were equivalent to the required intensity (i.e., 70% V˙O2max).

Back to Top | Article Outline

Experimental design.

Each participant performed two trials in a randomized order, separated by 14 d and applied in a double-blind manner. Over the 48 h leading up to the first of these trials, each individual accurately recorded their usual diet and subsequently adhered to this same dietary intake over the 48-h period prior to their second trial. Therefore, while there was a degree of variation between participants in terms of both energy intake and macronutrient profile (11,598 ± 4753 kJ·d−1, 58 ± 11% carbohydrate, 22 ± 9% fat, 19 ± 5% protein: mean ± SD), these variables were matched between the two trials, and all participants had ingested at least 200 g of dietary carbohydrate over the final 24 h prior to each main trial. Each trial involved a 90-min treadmill run at 70% V˙O2max (run 1) followed by 4 h of recovery, during which participants rested in the laboratory while ingesting either carbohydrate alone or a mixture of carbohydrate and protein. Muscle biopsies were obtained at the beginning and end of recovery to assess the rate of muscle glycogen resynthesis over the intervening period. A third muscle biopsy was taken following a second run (run 2: 60 min at 70% V˙O2max) in order to evaluate the rate of muscle glycogen degradation during exercise subsequent to recovery.

Back to Top | Article Outline

Experimental protocol.

Participants arrived in the laboratory following a 10-h overnight fast. After providing a urine sample, each participant's postvoid nude BM was recorded (Avery Ltd.) before a cannula was inserted into an antecubital vein and a 10-mL resting venous blood sample was obtained. The cannula was kept patent throughout each trial by frequent flushing with isotonic saline. Prior to exercise, the Douglas bag technique was used to collect a 5-min resting expired gas sample. Participants were required to stand for 15 min prior to the collection of all resting gas and blood samples to reduce the effect of body position on localized hemoconcentration and therefore facilitate meaningful comparisons between resting and exercise samples. A 5-min run at 60% V˙O2max was used as a standardized warm-up prior to running at a speed equivalent to 70% V˙O2max for 90 min (run 1). One-minute expired gas samples, heart rates (Polar 8810, Kempele, Finland), and ratings of perceived exertion (RPE) (7) followed by 10-mL venous blood samples were taken at 30-min intervals during run 1. Water intake was permitted ad libitum during run 1 in participants' first trials and was then matched in subsequent trials (0.79 ± 0.27 L; mean ± SD). Nude BM was recorded immediately following run 1 to assess hydration status through percent change in BM. Participants were then asked to rest in a supine position while a 3- to 5-mm skin incision was made to the anterior portion of the thigh under local anesthetic (2-3 mL of 1% lignocaine, Antigen Pharmaceuticals Ltd.) using a surgical blade. The percutaneous needle biopsy technique was then used to obtain muscle samples from the vastus lateralis.

The first volume of prescribed solution was ingested as soon as the first muscle biopsy had been removed from the thigh. The remaining seven volumes were then provided at 30-min intervals during the 4-h recovery, because the most rapid rates of muscle glycogen resynthesis have been reported when using this feeding schedule (29). The final aliquot was therefore provided 30 min prior to the second muscle biopsy, with participants being permitted 15 min to consume each volume. Expired gas samples and venous blood samples were taken during the recovery period every hour prior to feedings. Subjective ratings of gut fullness and thirst were recorded during gas sampling using adapted Borg scales (7) such that the anchor terms on each 6-20 scale ranged from "not full" to "very very full" and "not thirsty" to "very very thirsty," respectively. In addition, each participant's urine was collected throughout the 4-h recovery period. Approximately 3 h 45 min following the initial muscle biopsy, a further two skin incisions were made to the same leg at least 3 cm from one another and a second biopsy was sampled precisely 4 h following the first. Nude BM was then recorded, and, after the standard warm-up, participants were asked to run on the treadmill for 60 min at 70% V˙O2max (run 2). During this run, water intake was ad libitum in participants' first trials and then was matched in subsequent trials (0.39 ± 0.19 L; mean ± SD). All physiological measurements were obtained at 15 min intervals during run 2. Nude BM was recorded immediately after run 2, and the remaining incision site was used to obtain a final muscle biopsy. The three muscle biopsies taken in any given trial were sampled from the same leg, and the opposite leg was used in subsequent trials. The use of dominant/nondominant legs was, therefore, counterbalanced. Ambient temperature and humidity were recorded at 30-min intervals throughout all testing periods using a hygrometer (Zeal) and were not different between treatments; average values recorded were 20.5 ± 0.5°C and 55.6 ± 1.9% (mean ± SD), respectively.

Back to Top | Article Outline

Solution composition.

The carbohydrate (CHO) and carbohydrate-protein (CHO-PRO) solutions were provided in equal volumes for each treatment (590 ± 53 mL·h−1; mean ± SD) relative to each participant's BM such that the amount of carbohydrate (sucrose) ingested was 0.8 g·kg−1·h−1, which equates to a total carbohydrate intake of 236 ± 21 g (mean ± SD). The CHO-PRO mixture included an additional 0.3 g PRO·kg−1 BM·h−1 of whey protein isolate (amino acid profile: 20% glutamine, 11% leucine, 10% asparagine, 9% lysine, 7% proline, 6% threonine, 6% isoleucine, 5% valine, 5% alanine, all others 21%), thus providing a total of 89 ± 8 g of protein during recovery (mean ± SD). The estimated amount of energy that the CHO and CHO-PRO solutions made available for metabolism was therefore 13.4 and 18.0 kJ·kg−1 BM·h−1, respectively. Pretesting was conducted to ensure that these solutions were successfully matched for flavor (orange and passion fruit), consistency and odor. Successful double blinding of the treatments was further verified via an exit interview in which none of the participants was able to identify which solution they had ingested during the two trials.

Back to Top | Article Outline

Muscle analysis.

Once removed from the thigh, each muscle sample was immediately snap-frozen in liquid nitrogen, where it was later dissected to remove a 15- to 20-mg fragment of muscle, which was placed in a freeze-dryer (Pirani 10, Edwards) for 24 h at −50°C and between −10−1 and −10−2 mbar. The freeze-dried muscle fragments were then washed with 1 mL of petroleum ether to remove any excess lipid before each sample was reduced to a fine powder using an agate pestle and mortar. The precise quantity of powder obtained from each fragment was determined using an electronic balance scale (Mettler AE240, Switzerland) and these powdered samples were then stored at −80°C along with silica gel pending later analysis for glucose-6-phosphate (G-6-P), adenosine triphosphate (ATP), phosphocreatine (PCr), creatine (Cr), and both pro-and macroglycogen. The relative concentrations of all the metabolites cited above were determined according to the methods described by Harris et al. (1974) using a spectrophotometric plate reader (SpectraMax 190, Molecular Devices). All buffers, cofactors, standards, and enzymes were prepared immediately prior to each assay using grade I chemicals, and samples were analyzed at least in duplicate (both glycogen assays in triplicate).

In addition, to account for any contamination of the muscle samples with blood, lipid or connective tissue, all metabolites other than lactate were normalized in relation to each participant's mean total creatine concentration across both trials (i.e., PCr + Cr). All muscle metabolite data are reported as units per kilogram of dry mass, to avoid changes in concentration due to fluid shift during exercise. The contribution of muscle glycogen towards whole-body carbohydrate metabolism during run 2 was determined based on dual-energy x-ray absorptiometric analysis (DXA, Hologic Discovery W) of the lean tissue mass in the upper and lower parts of each leg for a typical 72.1-kg participant from this cohort (19.1% and 6.4% of BM, respectively). Given that glycogen is degraded approximately twice as rapidly in the triceps surae than in the vastus lateralis during treadmill running (17), overall rates of muscle glycogen degradation (g·min−1) were calculated as the product of the measured value in the vastus lateralis (g·kg−1 dry mass·min−1) and 7.5% of BM (i.e., the dry mass of all muscles in the upper leg plus twice the dry mass of lower leg muscles).

Back to Top | Article Outline

Blood analysis.

From each 10-mL whole-blood sample, 5 mL was dispensed into a nonanticoagulant tube where it was left to clot for 45 min at room temperature and then centrifuged at 2000g for 10 min at 4°C (Beckman-Coulter Allegra X-22R). The serum fraction was stored at −80°C, pending later analysis for insulin by radioimmunoassay (Coat-Count Insulin, MP Biomedicals Ltd.) using a gamma counter (Cobra 5000, Packard Instruments). The remaining 5 mL of whole blood was transferred into a tube containing the anticoagulant ethylenediaminetetraacetic acid (EDTA), from which triplicate 50- and 20-μL samples were taken in order to determine hematocrit (Hct Centrifuge, Hawksley) and hemoglobin concentrations, respectively. The latter was measured using a standard cyanomethemoglobin method (Boehringer Mannheim) and a spectrophotometer (Shimadzu 1240). The equations of Dill and Costill (11) were applied to these hematocrit and hemoglobin values to assess changes in plasma volume throughout the trials. Two further 20-μL samples of whole blood were deproteinized using 2.5% perchloric acid (200 μL), centrifuged at 7000g for 3 min (Eppendorf Centrifuge 5415c) and stored at −80°C for later determination of lactate concentration (19) using a flourometer (Locarte 8.9). The remaining whole blood was centrifuged at 2000g for 10 min at 4°C (Beckman-Coulter Allegra X-22R) before plasma was extracted, stored at −80°C, and later analyzed for glucose (Randox), free fatty acids (Wako NEFA C), glycerol (Randox), myoglobin (Randox), and urea (Randox), using an automatic spectrophotometric analyzer (Cobas-Mira plus, Roche).

Back to Top | Article Outline

Expired gas analysis.

Expired gas samples were collected using a Douglas bag, and the fractions of expired O2 and CO2 were assessed using paramagnetic and infrared analyzers, respectively (Servomex 1440). Total volumes expired were determined using a dry gas meter (Harvard Apparatus), and the temperatures of expired gases were measured with a digital thermometer (model C, Edale Instruments). These analyzers were calibrated prior to and at 2-h intervals during each test with gases of known composition and volume within the physiological range, as certified by prior gravimetric analysis (British Oxygen Company). Rates of oxygen uptake (V˙O2) and carbon dioxide production (V˙CO2) were then used to calculate carbohydrate and fatty acid oxidation rates (g·min−1) using the following formulae:

where N is the estimated rate of nitrogen excretion based on urinary/plasma urea (18).

Extramuscular carbohydrate oxidation was then calculated as the difference between whole-body carbohydrate oxidation (derived from indirect calorimetry) and the rate of intramuscular carbohydrate oxidation (i.e., the calculated rate of overall muscle glycogen degradation).

Back to Top | Article Outline

Urine analysis.

The osmolality of participants' pretrial urine samples was determined via freezing point depression using a cryoscopic osmometer (Gonometer 030, Gonotec), and adequate hydration was assumed for osmolality values below 900 mOsm·kg−1 (24). The urine collected during the 4-h recovery period was stored in a vessel containing 5 mL of 10% thymol-isopropanol as a preservative, and, once total volume had been recorded, a mixed 5-mL sample was taken and stored at −80°C. The urea concentration of this mixed sample was later determined as an estimate of total urine nitrogen excretion using procedures identical to those applied during plasma urea analysis. Plasma urea concentrations were then used to correct urinary urea excretion for changes in the total body urea pool during recovery (18), and nonprotein respiratory exchange ratios (NPRER) were calculated for this period according to the estimated rate of protein oxidation (15).

Back to Top | Article Outline

Statistical analyses.

Using similar supplements to those under investigation in the present study, van Loon et al. (2000) reported the inclusion of protein to accelerate muscle glycogen resynthesis by 18.8 mmol glucosyl units per kilogram of dry mass per hour, with a pooled standard deviation of 6.6 mmol glucosyl units per kilogram of dry mass per hour (N = 8). Based on these data, it was estimated that a sample size of 6 has a 99% power to detect such differences statistically. The incremental area-under-the-curve data that were calculated to assess insulinemic responses to each solution during recovery were found to be nonnormally distributed, and a Wilcoxon test was, therefore, applied to compare the median values between treatments for this data set, along with any other nonparametric variables (i.e., RPE, gut fullness, and thirst). Paired t-tests were used to analyze any other results involving a single comparison of two level means, while a two-way general linear model for repeated measures (treatment × time) was used to identify overall differences between experimental conditions. The Greenhouse-Geisser correction was used for epsilon < 0.75, while the Huynh-Feldt correction was adopted for less severe asphericity. Where significant F values were found, the Holm-Bonferroni stepwise correction was applied to determine the location of variance (4). Unless otherwise stated, all data are expressed as means ± standard errors of the mean (SEM). Specifically, descriptive statistics are presented as means ± standard deviations (SD), and nonnormally distributed data are expressed as medians and ranges.

Back to Top | Article Outline


A significant effect of time was established for total muscle glycogen concentrations (time: F = 83.3, P < 0.001; Fig. 1), but there were no differences between treatments at any time point, and rates of muscle glycogen resynthesis did not therefore differ between the CHO and CHO-PRO supplements (12.3 ± 2.2 and 12.1 ± 2.7 mmol glucosyl units per kilogram of dry mass per hour, respectively). Similarly, the rate of muscle glycogen degradation during exercise following recovery was no greater with the CHO treatment (2.2 ± 0.3 mmol glucosyl units per kilogram of dry mass per minute) than with the CHO-PRO treatment (2.0 ± 0.1 mmol glucosyl units per kilogram of dry mass per minute). Table 1 presents the two subglycogen pools separately along with concentrations of muscle ATP, PCr, Cr, G-6-P, and lactate. None of these variables responded any differently between treatments.





Serum insulin concentrations were significantly higher when the CHO-PRO mixture was ingested during recovery as opposed to the supplement containing the carbohydrate fraction alone (treatment: F = 13.1, P = 0.02; Fig. 2A). Accordingly, there were significant differences between treatments in the incremental area under the concentration curve that was calculated solely for the 4-h recovery period, with median insulinemic responses of 42.8 nmol (22.2-47.3) per 240 min·L−1 with the CHO treatment and 48.5 nmol (42.9-64.1) per 240 min·L−1 with the CHO-PRO treatment (P = 0.04). There were also notable differences in plasma glucose responses following ingestion of each supplement (treatment × time: F = 6.5, P = 0.01, Fig. 2B). Specifically, ingestion of the CHO-PRO solution did not substantially increase plasma glucose concentrations at any point during recovery, while ingestion of the CHO solution resulted in an initial glucose peak approximately 1 h after the commencement of feeding (P = 0.08 vs CHO-PRO). With the onset of exercise following recovery, plasma glucose concentrations decreased transiently with both treatments. However, this decrement was more pronounced with the CHO treatment, and the disparity between treatments was maintained thereafter, with significant differences noted after 30 min of exercise (P = 0.01).



Plasma urea concentrations began to increase immediately after ingestion of the first CHO-PRO solution and gradually increased further above the response to the CHO treatment as time progressed (treatment × time: F = 32.8, P < 0.001), with values significantly different between treatments throughout run 2 (P ≤ 0.02; Fig. 3A). The concentrations of blood lactate during each run and during the recovery period are illustrated in Figure 3B. Despite a significant interaction effect (treatment × time: F = 2.38, P = 0.015), the lactate concentrations in the latter stages of recovery were not statistically different between treatments (P = 0.08 at 4 h of recovery).



Irrespective of experimental treatment, plasma concentrations of free fatty acids and glycerol were elevated during the first exercise session before rapidly decreasing with the onset of feeding (Fig. 4). Both these variables were subsequently elevated in response to the second exercise session, and glycerol concentrations in particular were elevated significantly more so with the CHO treatment than with the CHO-PRO treatment (treatment × time: F = 5.8, P = 0.02), with a specific treatment difference identified after 45 min of exercise (P = 0.04).



At no time point was there any difference between treatments in the plasma myoglobin response to exercise. From basal levels of approximately 1.6 nM with both treatments, concentrations were elevated to peaks of 11.9 ± 5.2 nM with the CHO treatment and 10.1 ± 4.0 nM with the CHO-PRO treatment after 1 h of recovery, and they did not substantially decrease before the onset of the second run. Myoglobin release during the second run therefore supplemented the concentrations of this hemoprotein, culminating in eventual postexercise values of 27.5 ± 8.8 nM with the CHO treatment and 27.3 ± 9.1 nM with the CHO-PRO treatment.

Although only accounting for a small fraction of overall metabolism, the estimated rate of protein oxidation during recovery was greater with ingestion of the CHO-PRO solution than with ingestion of the CHO solution (0.96 ± 0.08 mg·kg−1·min−1 and 0.17 ± 0.15 mg·kg−1·min−1, respectively; P = 0.01). Given that the thermic effect of protein is higher than that of carbohydrate, metabolic rate was therefore higher with the CHO-PRO treatment during recovery (9.2 ± 0.5 vs 8.5 ± 0.6 kJ·min−1; P = 0.05). However, there were no treatment differences in either carbohydrate or fatty acid oxidation during recovery once corrected for the estimated protein oxidation rate. In contrast, overall rates of metabolism were more similar between treatments during the run following recovery (CHO = 64.1 ± 1.6 kJ·min−1; CHO-PRO = 66.6 ± 2.1 kJ·min−1). This was despite varied substrate selection, with CHO ingestion resulting in higher rates of fatty acid oxidation than CHO-PRO (4.71 ± 0.80 vs 2.56 ± 0.35 mg·kg−1·min−1; P = 0.01) and CHO-PRO ingestion producing higher rates of carbohydrate oxidation than ingestion of CHO alone (48.4 ± 2.2 vs 41.7 ± 2.6 mg·kg−1·min−1; P = 0.001). Importantly, this latter finding can be attributed entirely to the oxidation of extramuscular carbohydrate sources (i.e., blood glucose and lactate), given that there were no differences in the rate of muscle glycogen degradation between treatments (Fig. 5).



Preexercise urine osmolalities indicated an adequate degree of hydration for all participants across both trials (631 ± 350 mOsm·kg−1; mean ± SD). There were no treatment differences in terms of any other variable relating to hydration status (i.e., plasma volume, changes in BM, or participants' self-reported thirst ratings), and the total volume of urine produced during recovery was also similar between the CHO and CHO-PRO solutions (1204 ± 227 and 953 ± 52 mL, respectively). Relative exercise intensities were successfully standardized between treatments and averaged 74.0 ± 1.8% V˙O2max for run 1 and 73.7 ± 1.6% V˙O2max for run 2, as was reflected by the overall heart rates (174 ± 4 and 175 ± 4 bpm) and ratings of perceived exertion (13 ± 1 and 14 ± 1) that were recorded during these respective exercise sessions. All participants' subjective ratings of gut fullness were higher throughout the entire recovery period while ingesting the CHO-PRO solution (13 ± 1) than while ingesting the CHO solution (10 ± 1; P = 0.03).

Back to Top | Article Outline


The main aim of this investigation was to examine whether the ingestion of carbohydrate with added protein during recovery from prolonged running can accelerate muscle glycogen storage relative to a supplement of matched carbohydrate content. Converse to our hypothesis and contrary to existing evidence regarding prolonged cycling (5,12,29,33), the present study did not reveal any differences between treatments in the rate of muscle glycogen resynthesis during recovery. This was despite the fact that the supplement including whey protein isolate not only provided additional energy but also produced significantly higher serum insulin concentrations than when ingesting the carbohydrate fraction alone. In contrast, the secondary purpose of this investigation was to explore the metabolic responses to exercise subsequent to recovery with each treatment. In this regard, ingestion of the CHO-PRO mixture resulted in an improved maintenance of plasma glucose concentrations and increased oxidation of extramuscular carbohydrates during exercise, without altering the rate of muscle glycogen degradation. Although the second bout of exercise in the present study did not require participants to run to the point of volitional exhaustion, these findings may explain the ergogenic influence of carbohydrate-protein ingestion, which we reported previously when using the same supplements and protocol, but with an exercise capacity test following recovery (6).

Our observation that muscle glycogen resynthesis rates during the 4-h recovery period were similar between treatments is not consistent with previous studies in this area. Specifically, all other research on this topic has reported a more rapid accumulation of muscle glycogen during recovery when protein has been added to ≤ 0.8 g·kg−1·h−1 of carbohydrate to augment the insulinemic response (5,12,29,33). The precise reason for this discrepancy is not immediately apparent, but it may be related to the specific type of exercise that was performed prior to recovery (i.e., cycling vs running). Specifically, the glucose-transport capacity of previously exercised muscle may be compromised following running compared with cycling due to the increased eccentric muscle action and resultant myofibrillar damage associated with treadmill running (3). Studies using prolonged cycling to induce glycogen depletion indicate that including protein in the postexercise diet may be ineffective in accelerating muscle glycogen storage when sufficient carbohydrate is ingested; for example, ingesting ≥ 1 g·kg−1·h−1 appears to maximize the rate of muscle glucose uptake even without additional protein (8,14,28). A logical extension of this reasoning is that the lower maximal rate of muscle glycogen storage after exercise with a substantial eccentric component might be attained even when only 0.8 g·kg−1·h−1 of carbohydrate is ingested during recovery. With regard to the present findings, the elevated systemic concentrations of myoglobin provide evidence that at least some degree of exercise-induced muscle damage was sustained. Therefore, the detrimental effect of this tissue injury on muscle glucose transport capacity could potentially explain why rates of muscle glycogen resynthesis were both relatively low and virtually identical between treatments in the present study.

An alternative explanation for the similar rates of glycogen resynthesis between treatments in the present study is that the ingested whey protein isolate may not have been sufficiently effective in augmenting insulin secretion. While even modest increases in circulating insulin can increase glucose uptake, it has been proposed that a minimum insulin concentration of approximately 340 pM may be necessary to shift the primary fate of intracellular glucose from net oxidation to net storage (32). It was certainly the case in the present study that serum insulin concentrations only exceeded this figure when both carbohydrate and protein were ingested, but by no means was this increased insulin response comparable with the findings of some other studies in which a concomitant increase in muscle glycogen resynthesis has been reported (29,33). For example, the ingestion of 0.8 g CHO·kg−1·h−1 in the study of van Loon et al. (2000) resulted in peak insulin concentrations similar to those observed in the present study, but when these authors added 0.4 g·kg−1·h−1 of amino acids there was an 88% increase in insulinemic response during 5 h, and peak insulin concentrations exceeded 600 pM (29). It is not entirely clear why the protein provided by these authors proved to be such an effective insulin secretagogue relative to the supplement used in the present study, but this effect may be related to minor variations in the rate of protein ingestion and/or subtle differences in the precise amino acid composition of the protein fraction (most notably, the inclusion of free leucine and phenylalanine).

The differences between treatments in terms of blood glucose availability and substrate oxidation during run 2 in the present study also warrant discussion, particularly in view of our earlier examination of these exact supplements in which participants were all able to exercise for longer following recovery when protein had been included in their postexercise diet (6). While the current results clearly demonstrate that this ergogenic benefit cannot be attributed to differences in muscle glycogen availability, it has been suggested that fatigue during prolonged exercise can be postponed through an increased oxidation of extramuscular carbohydrate sources (10). Given that the rate of blood glucose oxidation is dictated primarily by the availability of blood glucose (31), it is likely that the increased oxidation rate of extramuscular carbohydrates following carbohydrate-protein ingestion in the present study is related to the elevated concentrations of blood glucose during exercise with this treatment. One possible mechanism through which the ingested protein may have influenced blood glucose availability is related to the rate of glucose appearance from the gastrointestinal tract with each respective treatment. Specifically, protein is known to empty from the gut slightly slower than carbohydrate (20), which may in turn have delayed intestinal glucose absorption such that exogenous carbohydrate was still appearing in the circulation at the onset of exercise following recovery. Findings consistent with this suggestion are the higher subjective ratings of gut fullness and lower peak glucose concentrations during recovery with the CHO-PRO treatment. Notably, with regard to this latter finding, recent evidence indicates that for trained athletes the attenuated glucose response with combined carbohydrate-protein ingestion during recovery is indeed more likely to reflect a reduced rate of glucose appearance into the circulation than an increased rate of glucose disposal (16). Such an interpretation would certainly be supported by the fact that neither carbohydrate oxidation nor carbohydrate storage (at least in muscle) were different between treatments during recovery in the present study.

Another potential source of the supplementary blood glucose following carbohydrate-protein ingestion may have been an increased availability of hepatic glycogen. Indeed, any glucose that could not be taken up by previously exercised muscle may have been directed towards an alternative site of storage within the liver during recovery from the initial treadmill run. In addition, the elevated concentrations of plasma urea with carbohydrate-protein ingestion are indicative of an increased availability of α-keto acids for gluconeogenesis, and it is likely that a substantial quantity of any newly generated glucose would be retained by the liver, particularly given the relative hyperinsulinemia during recovery (22). Evidence from studies using rodents certainly support the preferential resynthesis of liver glycogen following prolonged exercise, and it appears that dietary whey protein may stimulate this process (21). The rapid fall in serum insulin and plasma glucose concentrations at the start of run 2, combined with the increase in circulating free fatty acids, would then be expected to reduce the action of insulin on the liver and, therefore, permit an increase in hepatic glucose output (9). The treatment differences in blood glucose concentrations during run 2 may, therefore, reflect an increased availability of hepatic glycogen and gluconeogenic substrate with the CHO-PRO treatment, potentially further explaining our previous report of improved restoration of exercise capacity with this supplement (6). Nonetheless, the results of the present investigation only provide tentative support for such possibilities, and future research would provide additional insight through directly examining the influence of these supplements on liver glycogen availability and endogenous glucose production.

It was first identified by Adamo and Graham (1998) that the glycogen stored within human muscle is composed of two distinct subglycogen pools (i.e., proglycogen and macroglycogen), which vary both in size and protein content (1). These authors have subsequently demonstrated that, of the two glycogen pools, proglycogen may be more responsive to insulin and should, therefore, account for the majority of muscle glycogen synthesis following acute nutritional intervention (2). In view of this finding, it is interesting to note that the two forms of glycogen responded differently to each supplement in the present study. Specifically, a trend was apparent for more proglycogen to be stored when ingesting the carbohydrate-protein mixture (P = 0.15), while macroglycogen displayed an equal trend in the opposite direction (P = 0.15). It is therefore possible that macroglycogen synthesis was preferentially stimulated with the CHO treatment relative to the CHO-PRO treatment, either to compensate for suboptimal rates of proglycogen storage or as a consequence of the slightly lower macroglycogen concentration at the end of the first exercise bout with the CHO treatment. Nonetheless, this difference between treatments is negligible when considered as a fraction of overall muscle glycogen availability, as macroglycogen only accounted for a small proportion of total glycogen content in the current study. The acid soluble fraction of glycogen has previously been reported to account for approximately 15-25% of total muscle glycogen content at rest and is reduced to approximately 8-12% following prolonged exercise (2,30). Therefore, it seems reasonable to suggest that a substantial quantity of macroglycogen may have been degraded in order to sustain metabolism during the first exercise session in the present study. It is also noteworthy that both forms of glycogen were subsequently oxidized in relatively large quantities during the second bout of exercise, with total glycogen availability by the end of recovery being reduced by approximately 45% during run 2 (Fig. 1). In summary, the distinct responses of proglycogen and macroglycogen to mixed macronutrient ingestion remain an intriguing phenomenon, and further research is required to determine the specific physiological roles of these separate pools of muscle glycogen.

In conclusion, in contrast to the wealth of available literature that has employed cycling as the mode of exercise, adding 0.3 g·kg−1·h−1 of whey protein isolate to a recovery solution providing 0.8 g·kg−1·h−1 of carbohydrate did not accelerate the rate of muscle glycogen resynthesis in the 4 h following prolonged treadmill running. Furthermore, we have shown for the first time that addition of protein to a carbohydrate solution can increase the overall rate of carbohydrate oxidation during a second exercise bout subsequent to recovery without altering the rate of muscle glycogen degradation. We therefore conclude that extramuscular sources of carbohydrate were oxidized at a greater rate following ingestion of the mixed macronutrient supplement, possibly due to an elevated appearance of glucose from the liver and/or gastrointestinal tract during exercise. These findings may explain our previous observation of a delay in the onset of fatigue following recovery when carbohydrate and protein are ingested together following an initial bout of prolonged exercise (6).

This study received financial support from GlaxoSmithKline. The results of the present study do not constitute endorsement of any product by the authors or by ACSM.

Back to Top | Article Outline


1. Adamo KB, Graham TE. Comparison of traditional measurements with macroglycogen and proglycogen analysis of muscle glycogen. J Appl Physiol. 1998;84:908-13.
2. Adamo KB, Tarnopolsky MA, Graham TE. Dietary carbohydrate and postexercise synthesis of proglycogen and macroglycogen in human skeletal muscle. Am J Physiol. 1998;275:E229-34.
3. Asp S, Watkinson A, Oakes ND, Kraegen EW. Prior eccentric contractions impair maximal insulin action on muscle glucose uptake in the conscious rat. J Appl Physiol. 1997;82:1327-32.
4. Atkinson G. Analysis of repeated measurements in physical therapy research: multiple comparisons amongst level means and multifactorial designs. Phys Ther Sport. 2002;3:191-203.
5. Berardi JM, Price TB, Noreen EE, Lemon PWR. Postexercise muscle glycogen recovery enhanced with a carbohydrate-protein supplement. Med Sci Sports Exerc. 2006;38(6):1106-13.
6. Betts JA, Williams C, Duffy K, Gunner F. The influence of carbohydrate and protein ingestion during recovery from prolonged exercise on subsequent endurance performance. J Sports Sci. 2007;25:1449-60.
7. Borg G. Perceived exertion: a note on "history" and methods. Med Sci Sports Exerc. 1973;5:90-3.
8. Carrithers JA, Williamson DL, Gallagher PM, Godard MP, Schulze KE, Trappe SW. Effects of postexercise carbohydrate-protein feedings on muscle glycogen restoration. J Appl Physiol. 2000;88:1976-82.
9. Coker RH, Koyama Y, Denny JC, Camacho RC, Lacy DB, Wasserman DH. Prevention of overt hypoglycemia during exercise. Stimulation of endogenous glucose production independent of hepatic catecholamine action and changes in pacreatic hormone concentration. Diabetes. 2002;51:1310-8.
10. Coyle EF, Coggan AR, Hemmert MK, Ivy JL. Muscle glycogen utilization during prolonged strenuous exercise when fed carbohydrate. J Appl Physiol. 1986;61:165-72.
11. Dill DB, Costill DL. Calculation of percentage changes in volumes of blood, plasma and red cells in rehydration. J Appl Physiol. 1974;37:247-8.
12. Ivy JL, Goforth HW, Damon BM, McCauley TR, Parsons EC, Price TB. Early postexercise muscle glycogen recovery is enhanced with a carbohydrate-protein supplement. J Appl Physiol. 2002;93:1337-44.
13. Jentjens R, Jeukendrup AE. Determinants of post-exercise glycogen synthesis during short-term recovery. Sports Med. 2003;33:117-44.
14. Jentjens RLPG, Van Loon LJC, Mann CH, Wagenmakers AJM, Jeukendrup AE. Addition of protein and amino acids to carbohydrates does not enhance postexercise muscle glycogen synthesis. J Appl Physiol. 2001;91:839-46.
15. Jequier E, Acheson KJ, Schutz Y. Assessment of energy expenditure and fuel utilization in man. Annu Rev Nutr. 1987;7:187-208.
16. Kaastra B, Manders RJF, Van Breda E, et al. Effects of increasing insulin secretion on acute postexercise blood glucose disposal. Med Sci Sports Exerc. 2006;38(2):268-75.
17. Krssak M, Petersen KF, Bergeron R, et al. Intramuscular glycogen and intramyocellular lipid utilization during prolonged exercise and recovery in man: a 13C and 1H nuclear magnetic resonance specroscopy study. J Clin Endocrinol Metab. 2000;85:748-53.
18. Livesey G, Elia M. Estimation of energy expenditure, net carbohydrate utilization, and net fat oxidation and synthesis by indirect calorimetry: evaluation of errors with special reference to the detailed composition of fuels. Am J Clin Nutr. 1988;47:608-28.
19. Maughan RJ. A simple rapid method for the determination of glucose, lactate, pyruvate, alanine, 3-hydroxybutarate and acetoacetate on a single 20 ul blood sample. Clin Chim Acta. 1982;13:317-22.
20. Maughan RJ, Leiper JB, Vist GE. Gastric emptying and fluid availability after ingestion of glucose and soy protein hydrolysate solutions in man. Exp Physiol. 2004;89:101-8.
21. Morifuji M, Sakai K, Sanbongi C, Sugiura K. Dietary whey protein increases liver and skeletal muscle glycogen levels in exercise-trained rats. Br J Nutr. 2005;93:439-45.
22. Pencek RR, James FD, Lacy DB, et al. Exercise-induced changes in insulin and glucagon are not required for enhanced hepatic glucose uptake after exercise but influence the fate of glucose within the liver. Diabetes. 2004;53:3041-7.
23. Rotman S, Slotboom J, Kreis R, Boesch C, Jequier E. Muscle glycogen recovery after exercise measured by 13C-magnetic resonance spectroscopy in humans: effect of nutritional solutions. MAGMA. 2000;11:114-21.
24. Shirreffs SM, Maughan RJ. Urine osmolality and conductivity as indices of hydration status in athletes in the heat. Med Sci Sports Exerc. 1998;30(11):1598-602.
25. Tarnopolsky MA, Bosman M, Macdonald JR, Vandeputte D, Martin J, Roy BD. Postexercise protein-carbohydrate and carbohydrate supplements increase muscle glycogen in men and women. J Appl Physiol. 1997;83:1877-83.
26. Taylor H, Buskirk E, Henschel A. Maximal oxygen uptake as an objective measure of cardio-respiratory performance. J Appl Physiol. 1955;8:73-80.
27. Tsintzas K, Williams C. Human muscle glycogen metabolism during exercise: effect of carbohydrate supplementation. Sports Med. 1998;25:7-23.
28. Van Hall G, Shirreffs SM, Calbet JA. Muscle glycogen resynthesis during recovery from cycle exercise: no effect of additional protein ingestion. J Appl Physiol. 2000;88:1631-6.
29. Van Loon LJ, Saris WH, Kruijshoop M, Wagenmakers AJ. Maximizing postexercise muscle glycogen synthesis: carbohydrate supplementation and the application of amino acid or protein hydrolysate mixtures. Am J Clin Nutr. 2000;72:106-11.
30. Wee SL, Williams C, Tsintzas K, Boobis L. Ingestion of a high-glycemic index meal increases muscle glycogen storage at rest but augments its utilization during subsequent exercise. J Appl Physiol. 2005;99:707-14.
31. Weltan SM, Bosch AN, Dennis SC, Noakes TD. Influence of muscle glycogen content on metabolic regulation. Am J Physiol. 1998;274:E72-82.
32. Young AA, Bogardus C, Stone K, Mott DM. Insulin response of components of whole-body and muscle carbohydrate metabolism in humans. Am J Physiol. 1988;254:E231-6.
33. Zawadzki KM, Yaspelkis BB 3rd, Ivy JL. Carbohydrate-protein complex increases the rate of muscle glycogen storage after exercise. J Appl Physiol. 1992;72:1854-9.


©2008The American College of Sports Medicine