The influence of carbohydrate–protein co-ingestion following endurance exercise on myofibrillar and mitochondrial protein synthesis


L. Breen: McMaster University, Department of Kinesiology, 1280 Main Street West, Hamilton, ON, L8P 2P9, Canada. Email:


Non-technical summary  A single bout of exercise stimulates the production of new muscle proteins. Furthermore, ingesting protein in close proximity to exercise enhances the metabolic response. Long-term exercise training promotes muscle adaptation, and the mode of exercise performed determines the type of proteins that are made. To date, the types of proteins that are made when protein is ingested after endurance exercise are not known. We report that when well-trained male cyclists ingest protein with a carbohydrate drink after a high-intensity ride, production of proteins responsible for muscle contraction is increased. Proteins responsible for aerobic energy production are not responsive to protein feeding. Furthermore, specific signals within the muscle that control protein synthesis are responsive to protein ingestion, providing a potential mechanism to underpin our primary findings. These results suggest that protein feeding after intense endurance exercise may be important in maintaining the structural quality and power generating capacity of the muscle.


Abstract  The aim of the present study was to determine mitochondrial and myofibrillar muscle protein synthesis (MPS) when carbohydrate (CHO) or carbohydrate plus protein (C+P) beverages were ingested following prolonged cycling exercise. The intracellular mechanisms thought to regulate MPS were also investigated. In a single-blind, cross-over study, 10 trained cyclists (age 29 ± 6 years, inline image 66.5 ± 5.1 ml kg−1 min−1) completed two trials in a randomized order. Subjects cycled for 90 min at 77 ± 1%inline image before ingesting a CHO (25 g of carbohydrate) or C+P (25 g carbohydrate + 10 g whey protein) beverage immediately and 30 min post-exercise. A primed constant infusion of l-[ring-13C6]phenylalanine began 1.5h prior to exercise and continued until 4h post-exercise. Muscle biopsy samples were obtained to determine myofibrillar and mitochondrial MPS and the phosphorylation of intracellular signalling proteins. Arterialized blood samples were obtained throughout the protocol. Plasma amino acid and urea concentrations increased following ingestion of C+P only. Serum insulin concentration increased more for C+P than CHO. Myofibrillar MPS was ∼35% greater for C+P compared with CHO (0.087 ± 0.007 and 0.057 ± 0.006%h−1, respectively; P= 0.025). Mitochondrial MPS rates were similar for C+P and CHO (0.082 ± 0.011 and 0.086 ± 0.018%h−1, respectively). mTORSer2448 phosphorylation was greater for C+P compared with CHO at 4h post-exercise (P < 0.05). p70S6KThr389 phosphorylation increased at 4h post-exercise for C+P (P < 0.05), whilst eEF2Thr56 phosphorylation increased by ∼40% at 4h post-exercise for CHO only (P < 0.01). The present study demonstrates that the ingestion of protein in addition to carbohydrate stimulates an increase in myofibrillar, but not mitochondrial, MPS following prolonged cycling. These data indicate that the increase in myofibrillar MPS for C+P could, potentially, be mediated through p70S6K, downstream of mTOR, which in turn may suppress the rise in eEF2 on translation elongation.


AMP-activated protein kinase




carbohydrate plus protein


eukaryotic initiation factor 4E binding protein 1


endurance exercise


eukaryotic elongation factor 2


muscle protein synthesis


mammalian target of rapamycin

p38 MAPK

p38 mitogen-activated protein kinase


70 kDa S6 protein kinase


proline-rich Akt substrate 40 kDa


resistance exercise


Endurance (EE) and resistance exercise (RE) training regimens result in divergent phenotypic adaptations. Whereas RE promotes muscle hypertrophy and an increase in contractile force output (Jones & Rutherford, 1987; Hartman et al. 2007), EE training is characterized by an expansion of oxidative capacity, brought about through an increase in the size and density of mitochondria and accumulation of myosin heavy chain I (Holloszy, 1967; Harber et al. 2002; Tarnopolsky et al. 2007). At the metabolic level, the adaptation to exercise is determined by summing the acute transcriptional (Pilegaard et al. 2000; Hildebrandt et al. 2003) and translational responses (Wilkinson et al. 2008) to each exercise stimulus and the subsequent increase in synthesis of muscle proteins (Mahoney et al. 2005; Hawley et al. 2006; Tipton, 2008). Recently, Wilkinson and colleagues (2008) showed the response of myofibrillar and mitochondrial muscle protein synthesis (MPS) were dependent on the type of exercise performed. Furthermore, the interaction of RE and protein nutrition varies between different protein fractions (Moore et al. 2009). Thus, it is important to examine the response of different protein fractions in order to determine whether a particular nutritional intervention elicits the desired effect following varied types of exercise.

Whereas the response of mixed MPS to different types of exercise has been investigated (Tipton et al. 1996; Phillips et al. 1997; Harber et al. 2010), there is less information on the response of various proteins to exercise and nutrition. It is widely recognized that ingesting protein potentiates the anabolic effect of RE (Tang et al. 2007), seemingly due to the essential amino acid content (Tipton et al. 1999). Recently, Moore et al. (2009) showed that myofibrillar and sarcoplasmic proteins respond in a similar manner to protein feeding; however, rates of synthesis of only myofibrillar proteins were further increased when RE was combined with protein ingestion. Similarly, myofibrillar, but not mitochondrial protein synthesis increased in response to nutrient, including protein, ingestion following repeated sprints (Coffey et al. 2010). To date, no study has investigated the interactive effect of EE and protein ingestion on different muscle protein fractions.

There are many studies demonstrating the response of MPS to protein ingestion following RE (Phillips et al. 1997; Moore et al. 2009; Burd et al. 2010) but very few studies have sought to determine the effect of protein ingestion on MPS following EE. Levenhagen et al. (2002) reported that adding protein to a carbohydrate treatment increased post-EE leg and whole-body protein synthesis. This increase in synthesis was associated with increased net muscle protein balance. However, it was unclear whether the benefits observed were due to the protein per se, or an increase in total energy intake. To address this possibility, Howarth and colleagues (2009) showed that the addition of protein to carbohydrate increased mixed MPS compared with CHO treatments matched for total energy and carbohydrate content. Thus, it seems clear that the rate of mixed MPS responds to protein ingestion following EE, as well as RE. However, none of these studies (Levenhagen et al. 2002; Howarth et al. 2009) attempted to determine the specific protein fractions that contribute to changes in mixed MPS in response to protein ingestion or the mechanisms accounting for the response. Recent evidence from Coffey et al. (2010) suggests that nutrient provision prior to a high-intensity, repeat-sprint bout of cycling increases the rate of myofibrillar MPS in the post-exercise period. However, this finding may have been influenced by the unique overload stimulus of repeated sprint exercise, characterized by a greater rate of force production and large disturbances to ion homeostasis (Coffey et al. 2009). Thus, our primary aim was to investigate the impact of ingesting protein in addition to carbohydrate on the response of myofibrillar and mitochondrial protein synthesis rates following prolonged EE.

Acute changes in transcription and translation that occur following exercise regulate skeletal muscle protein turnover through a number of intracellular signalling proteins. The mammalian target of rapamycin (mTOR) is a key regulator of translational control, integrating environmental signals from nutrients and exercise to control cell growth (Fingar et al. 2004). The activation of mTOR signalling leads to the phosphorylation of downstream targets involved in mRNA translation initiation and elongation (Bolster et al. 2003), e.g. p70 ribosomal protein S6 kinase-1 (p70S6K). In concert with an increase in mitochondrial MPS following EE, Wilkinson et al. (2008) showed an increase in the phosphorylation of signalling proteins in the mTOR–p70S6K pathway. Furthermore, human (Ivy et al. 2008) and rat studies (Morrison et al. 2008) indicate that post-EE C+P ingestion increases the phosphorylation of intermediates in the mTOR–p70S6K pathway. To date, no study has characterized the response of signalling proteins to changes in myofibrillar and mitochondrial MPS following post-EE protein ingestion. Hence, the secondary aim of the study was to elucidate the potential intracellular signalling mechanisms (in the mTOR–p70S6K pathway) regulating the response of myofibrillar and mitochondrial MPS following post-EE protein ingestion.



Ten well-trained, male cyclists were recruited from local cycling clubs through advertisements. Their mean (±SD) age was 29 ± 6 years, body mass was 77.2 ± 6.5 kg, maximal oxygen uptake was 66.5 ± 5.1 ml kg−1 min−1 and maximal power output was 383 ± 25 W. Only cyclists who undertook two or more training sessions per week of 1–5h duration were eligible to participate. Participants had 7.5 ± 3.0 years of competitive cycling experience. All tests were completed within a 4 week period with both treatment trials separated by 14–21 days, with the exception of one participant who completed the second trial 32 days after the first. The purpose and methodology of the study were clearly explained to the participants. All participants gave their informed consent prior to taking part in the study and were deemed healthy based on their response to a general health questionnaire. The experimental protocol was approved by the Black Country Research Ethics Committee (Rec No: 08/H1202/130). The study conformed to the standards set by the Declaration of Helsinki.

Study design

Participants reported to the laboratory on three separate occasions. During the first visit maximal aerobic fitness was determined. Approximately 2 weeks later participants performed the first blinded trial in which they consumed a carbohydrate (CHO) or carbohydrate–protein beverage (C+P; Lucozade Sport Recovery Powder, GlaxoSmithKline, Brentford, UK). During each trial participants performed a 90 min high intensity, steady-state cycle before consuming CHO or C+P immediately and 30 min following the exercise bout. Mitochondrial and myofibrillar MPS were measured by combining isotopic tracer infusion and muscle biopsy techniques. Participants returned to the laboratory for the second blinded trial 14–21 days after the first trial, thereby serving as their own control. Trial order was randomized in a counter-balanced fashion.

Preliminary testing

Body mass Body mass was determined to the nearest 0.1 kg. Each participant was weighed in their cycling clothing without shoes on. Measurement of body mass was repeated prior to each of the two testing visits to ensure body mass remained constant throughout the study.

Maximal cycling test Maximum oxygen uptake (inline image) and maximal power output (Wmax) were determined using an incremental cycle test-to-exhaustion on an electrically braked cycle ergometer (Lode Excalibur Sport v. 2.0, Groningen, Netherlands). The test consisted of a 3 min warm-up cycling at a self-selected cadence at 95 W followed by an increase of 35 W every 3 min until volitional exhaustion. Breath-by-breath measurements were taken throughout exercise using an OxyCon Pro automated gas analysis system (Jaeger, Würzburg, Germany). The gas analysers were calibrated using a 4.99% CO2–15.01% O2 gas mixture (BOC Gases, Surrey, UK) and the volume transducer was calibrated with a 3 litre calibration syringe. Heart rate (HR) was measured continuously via telemetry using a HR monitor (Polar S625X; Polar Electro Oy, Kempele, Finland). inline image was considered maximal if two of the four following conditions were met: (1) a plateau in inline image with further increasing workloads (an increase of <2 ml min−1 kg−1); (2) a HR within 10 beats min−1 of the age predicted maximum (220 bpm – age); (3) a respiratory exchange ratio (RER) of >1.05; and (4) a rate of perceived exertion (RPE) greater than 17. The seat position, handlebar height and orientation used during baseline testing were recorded and replicated on subsequent visits to the laboratory. All exercise bouts were conducted in thermo-neutral conditions (21°C, 40% relative humidity).

Dietary analysis and control Participants’ diet was standardized for 48h prior to each treatment. During the preliminary testing phase, participants completed a 3 day food diary, representative of their average week (2 weekdays and 1 weekend day). A questionnaire of food preferences was also completed by participants. Using an on-line diet planner (Weight Loss Resources, Peterborough, UK), each of the 3 days was logged and energy and macronutrient intake was estimated. In our study cohort the average total daily energy intake was 2787 ± 164 kcal (5.3 ± 0.4 g (kg BM)−1 carbohydrate; 1.2 ± 0.1 g (kg BM)−1 fat; 1.5 ± 0.2 g (kg BM)−1 protein). The food parcels given to each participant matched their habitual energy and macronutrient intake. Participants were instructed to refrain from caffeine and alcohol and to consume only the food provided for them over the 2 days prior to arriving for each trial. Participants also were asked to consume their final meal no later than 22.00h to ensure a 10h fast prior to measuring myofibrillar and mitochondrial protein synthesis rate. An identical 2 day food parcel was provided to each participant prior to the second trial.

Physical activity control Participants were instructed to maintain their normal training volume and intensity throughout the course of the study but to refrain from training for 48h prior to each treatment trial. To monitor physical activity between trials, participants were asked to record all training 7 days prior to each trial. No differences were noted in training volume in the 7 days prior to each trial.

Experimental trial protocol

Each participant was instructed to arrive at the Human Performance Laboratory at ∼06.30h after an overnight fast, where standard measures of height and weight were taken. A cannula was placed into the forearm vein of one arm and a wrist vein of the other. The forearm cannula was used to infuse a stable isotopic tracer whilst the hand vein was heated for frequent arterialized blood sampling (Abumarad et al. 1981). After a resting blood sample had been obtained, participants then received a primed constant infusion of l-[ring-13C6] phenylalanine (prime: 2 μmol kg−1; infusion: 0.05 μmol kg−1 min−1; Cambridge Isotope Laboratories, Andover, MA, USA) to determine MPS (described below). Approximately 90 min after the start of the infusion, having rested in a supine position, participants were asked to complete a 90 min cycling exercise bout on a Lode Cycle Ergometer at a self-selected cadence ≥60 revolutions per minute (rpm). The exercise bout consisted of a warm-up cycle at 50%Wmax (189 ± 4 W) for 10 min and then 80 min, initially at 75%Wmax (283 ± 6 W). inline image, RER, HR and RPE were recorded over 25–30 min, 55–60 min and 85–90 min of the exercise bout (as described above). If participants indicated during exercise that the workload was too difficult and they were unlikely to complete the full 80 min, workload was lowered in 5% decrements to no less than 65%Wmax. Air conditioning and a fan were used when requested by participants. The exact settings of the air conditioning/fan and the time of any change in workload were recorded and replicated during the second trial. Throughout each trial, participants were allowed to drink water ad libitum. The amount of water consumed throughout the course of the each trial was found to be similar (1556 ± 249 ml for CHO and 1674 ± 233 ml for C+P).

Muscle biopsy and blood sampling Using a 5 mm Bergstrom biopsy needle, two muscle biopsies (∼100–150 mg of muscle tissue per biopsy) were obtained from the same leg during each trial. The order of biopsied leg was randomized and counterbalanced for each trial. Prior to the exercise bout (∼15 min) under local anaesthetic (1% lidocaine), the lateral portion of one thigh was prepared for the extraction of a needle biopsy sample from the vastus lateralis muscle. Biopsy incisions were made prior to exercise to allow the sample to be obtained as quickly as possible after exercise (5 ± 1 min post-exercise). Immediately after the post-exercise muscle biopsy was obtained, participants were asked to consume one of two treatment beverages described below. Four hours after consuming the treatment beverage, the second muscle biopsy was obtained (Fig. 1). The second biopsy was taken ∼2 cm proximal to the first biopsy. Biopsy samples were quickly rinsed in saline, blotted and divided into two to three aliquots, before being frozen in liquid nitrogen and stored at −80°C until later analysis. Arterialized blood samples from a heated wrist vein were collected at rest, immediately post-exercise and every 15 min following beverage consumption for 2h. Thereafter, blood samples were obtained at regular intervals for the remainder of the infusion. Blood was collected in ethylenediaminetetraacetic-containing, lithium heparin-containing and serum separator tubes and spun at 1,500 g for 15 min at 4°C. Aliquots of plasma and serum were the frozen at −80°C for subsequent analysis.

Figure 1.

Schematic diagram of the experimental protocol

Treatment beverages Immediately after the first muscle biopsy sample was obtained, subjects ingested either 25.2 g of carbohydrate (CHO) or 25.4 g of carbohydrate plus 10.2 g of whey protein isolate (C+P) dissolved in 250 ml of cold water (∼11 ± < 1 min post-exercise). A second identical beverage was consumed 30 min after the first beverage was finished. This dose regime provided a total carbohydrate and protein intake of 50.8 g and 20.4 g, respectively, in C+P and a total carbohydrate intake of 50.4 g in CHO. Participants were encouraged to consume the beverages within 2 min. Both CHO and C+P treatment beverages were matched for flavour (orange and passion fruit) and appearance. Beverages were administered to participants in a single-blinded manner, the order of which was randomized. Eight out of the 10 participants correctly identified the order of the treatments. The amino acid content of the whey protein was (as percentage content, w:w): Ala, 5.2; Arg, 2.2; Asp, 11.4; Cys, 2.3; Gln, 18.8; Gly, 1.5; His, 1.8; Ile, 6.7; Leu, 11; Lys, 10; Met, 2.3; Phe, 3.1; Pro, 5.7; Ser, 4.8; Thr, 7; Trp, 1.5; Tyr, 2.7; and Val, 6.1. A small amount of l-[ring-13C6] phenylalanine tracer was added to the C+P drink (6% of phenylalanine content) in order to minimize changes in blood phenylalanine enrichment after drink ingestion. The need to add isotopic tracer to the C+P beverage dictated that the investigators were not blinded to the treatment beverage order.


Blood analyses Plasma glucose, lactate and urea concentrations were analysed using an ILAB automated analyser (Instrumentation Laboratory, Warrington, UK). Serum insulin concentrations were analysed using a commercially available ELISA kit (IBL International, Hamburg, Germany), following the manufacturer's instructions. l-[ring-13C6]Phenylalanine tracer-to-tracee (t/T) enrichment was determined by gas chromatography, mass spectrometry (GCMS) (model 5973; Hewlett Packard, Palo Alto, CA, USA). Upon thawing, plasma samples were combined with diluted acetic acid and purified on cation-exchange columns (Dowex 50W-X8-200, Sigma-Aldrich, Poole, UK). The amino acids were then converted to their N-tert - butyldimethyl - silyl -N- methyltrifluoracetamide (MTBSTFA) derivative. Plasma 13C6 phenylalanine enrichment was determined by ion monitoring at masses 234/240. Appropriate corrections were made for overlapping spectra contributing to the t/T ratio. Phenylalanine, leucine and threonine concentrations were determined using an internal standard method (Tipton et al. 1999, 2001), based on the known volume of blood and internal standard added. The internal standards used were U-[13C915N]phenylalanine (50 μmol l−1), U-[13C6]leucine (120 μmol l−1) and U-[13C915N]threonine (182 μmol l−1) added in a ratio of 100 μl ml−1 of blood. Leucine, threonine and phenylalanine concentrations were determined by monitoring at ions 302/308, 404/409 and 336/346, respectively.

Muscle tissue analyses Muscle samples were analysed for enrichment of l-[ring-13C6] phenylalanine in the intracellular pool and bound myofibrillar and mitochondrial protein fractions. Intracellular amino acids were liberated from ∼20 mg of muscle. The tissue was powdered under liquid nitrogen using a mortar and pestle and 500 μl of 0.2 m perchloric acid (PCA) was added. The mixture was centrifuged at 10,000 g for 10 min. The pH of the supernatant was then adjusted to 5–7 with 2 m KOH and 0.2 m PCA and treated with 20 μl of urease for removal of urea. The free amino acids from the intracellular pool were purified on cation-exchange columns (described above). Intracellular amino acids were converted to their MTBSTFA derivative and 13C6 phenylalanine enrichment determined by monitoring at ions 234/240 (as described above) using GCMS.

Mitochondrial and myofibrillar protein isolation was achieved using a protocol adapted from Wilkinson et al. (2008). Approximately 70–100 mg of muscle tissue was homogenized in a 2 ml Eppendorf tube with a Teflon pestle in 10 μl mg−1 of ice-cold homogenizing buffer (0.1 mm KCl, 50 mm Tris, 5 mm MgCl, 1 mm EDTA, 10 mmβ-glycerophosphate, 50 mm NaF, 1.5% BSA, pH 7.5). The homogenate was spun at 1000 g for 10 min at 4°C. The supernatant was transferred to another Eppendorf tube and spun at 10,000 g for 10 min at 4°C to pellet the sarcoplasmic mitochondria (SMs). The supernatant was then removed and discarded. The pellet that remained from the original 1,000 g spin was washed twice with homogenization buffer. A glass Dounce homogenizer and tight fitting glass pestle were used to forcefully homogenize the pellet in homogenization buffer to liberate intermyofibrillar mitochondria (IMs). The resulting mixture of myofibrillar proteins (MYO) and IMs was spun at 1,000 g for 10 min at 4°C to pellet out the MYO. The supernatant was removed and spun at 10,000 g for 10 min at 4°C to pellet the IMs. The MYO, SM and IM pellets were washed twice with homogenizing buffer containing no BSA. The MYO fraction was separated from any collagen by dissolving in 0.3 m NaCl, removing the supernatant and precipitating the proteins with 1 m PCA. All samples were washed once with 95% ethanol. In order to determine 13C6 phenylalanine enrichment in the mitochondrial protein fraction, IM and SM fractions were combined as per Wilkinson et al. (2008). Mitochondrial and myofibrillar fractions were then hydrolysed overnight at 110°C in 0.1 m HCl/Dowex 50W-X8-200 (Sigma-Aldrich) and the constituent amino acids purified on cation-exchange columns (Dowex 50W-X8-200, Sigma-Aldrich). The amino acids were then converted to their N-acetyl-n-propyl ester derivative. Phenylalanine labelling was determined by gas-chromatography-combustion-isotope ratio mass spectrometry (GC-C-IRMS, Delta-plus XL, Thermofinnigan, Hemel Hempstead, UK) by monitoring at ions 44/45 for labelled and unlabelled CO2. Unfortunately, during processing two mitochondrial fraction samples were lost, therefore these data represent an n= 8.

Western blots The remaining muscle tissue (25–40 mg) was powdered on dry ice under liquid nitrogen using a mortar and pestle. Approximately 20 mg of powdered muscle was homegenized in 10 μl of lysis buffer per mg of powdered muscle (50 mm Tris pH 7.5; 250 mm sucrose; 1 mm EDTA; 1 mm EGTA; 1% Triton X-100; 1 mm NaVO4; 50 mm NaF; 0.50% PIC), using a hand-held homogenizer (PRO200, UK). Samples were shaken at 4°C for 30 min (12,000 rpm), centrifuged for 5 min at 6000 g and the supernatant removed for protein determination. Protein concentration was determined using the DC protein assay (Bio-Rad, Hertfordshire, UK). Equal aliquots of protein were boiled in Laemmli sample buffer (250 mm Tris-HCl, pH 6.8; 2% SDS; 10% glycerol; 0.01% bromophenol blue; 5%β-mercaptoethanol) and separated on SDS polyacrylamide gels (10–12.5%) for 1h at 58 mA. Following electrophoresis proteins were transferred to a Protran nitrocellulose membrane (Whatman, Dassel, Germany) at 100 V for 1h. The membranes were incubated overnight at 4°C with the appropriate primary antibody. The primary antibodies used were AMPKThr172 (cat. no. 15-115, Millipore, Billerica, MA, USA), mTORSer2448 (2976, Cell Signaling Technology, Inc., Danvers, MA, USA), S6KThr389 (Cell Signaling 9234), AktThr308 (Cell Signaling 4056), 4E-BP1Thr37 (SC6025, Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA), eEF2Thr56 (Cell Signaling 2332), PRAS40Thr246 (Cell Signaling 2610) and p38 MAPKThr180 (Cell Signaling 9212). The following morning the membrane was rinsed in wash buffer (Tris Buffered Saline with 0.1% Tween-20) three times for 5 min. The membrane was then incubated for 1h at room temperature within wash buffer containing the appropriate secondary antibody, either horseradish peroxidase (HRP)-linked anti-mouse IgG (7072, New England Biolabs, Inc., Ipswich, MA, USA; 1:1000) or anti-rabbit IgG (New England Biolabs 7074; 1:1000). The membrane was then cleared in wash buffer three times for 5 min. Antibody binding was detected using enhanced chemiluminescence (Millipore). Imaging and band quantification were carried out using a Chemi Genius Bioimaging Gel Doc System (Syngene, Cambridge, UK). Intracellular signalling targets were determined with n= 8 for CHO and C+P trials. Protein phosphorylation was expressed relative to the total protein.


The fractional synthetic rate (FSR) of mitochondrial and myofibrillar proteins were calculated using the standard precursor–product method:

display math(1)

Where ΔEb is the change in bound 13C6 phenylalanine enrichment between two biopsy samples, Ep is the precursor enrichment and t is the time between muscle biopsies.

The true precursor enrichment would be the labelled phenylalanine-tRNA (Baumann et al. 1994). However, measurement of this enrichment requires large amounts of tissue, which cannot typically be obtained from human volunteers. Thus, the intracellular (IC) free phenylalanine enrichment is commonly used in studies (Tipton et al. 1999; Howarth et al. 2009; Moore et al. 2009), primarily because it often is considered to be the best available correlate of muscle tRNA (Ljungqvist et al. 1997). Unfortunately, due to technical difficulties, the IC enrichment was available for only five participants. Thus, we chose to use the plasma precursor to estimate the IC enrichment. A comparison of arterialized plasma and IC enrichments from the samples available (n= 5) revealed the IC enrichment to be 70 ± 2% of the plasma (range 67–77%). Furthermore, a comprehensive examination of studies (e.g. Koopman et al. 2006; Beelen et al. 2008; Tang et al. 2009) in which phenylalanine tracers were used to determine fed-state FSR revealed the IC phenylalanine enrichment to be ∼70% of the plasma enrichment. We chose to use the IC enrichment estimated from the plasma enrichment as the precursor for ease of comparison of the results to other previously published studies (e.g. Tipton et al. 1996; Howarth et al. 2009). However, FSRs also were calculated with the unadjusted plasma precursor.


A within-subject repeated measures design was utilized for the current study. Exercise variables, blood analytes and Western blot data were analysed using a two-way ANOVA with repeated measures (treatment × time) to determine differences between each treatment beverage across time. When a significant main effect or interaction was identified, data were subsequently analysed using a Bonferroni post hoc test. Myofibrillar and mitochondrial FSR data were analysed using one-factor (treatment) repeated measures ANOVA. All statistical tests were analysed using statistical package for social sciences (SPSS) v. 18.0 (Chicago, IL, USA). Significance for all analyses was set at P < 0.05. All values are presented as means ± standard error of the mean (SEM).


Exercise variables

There was no between-trial difference in heart rate, cadence, inline image and RER measured at 25–30, 55–60 and 85–90 min of the steady state cycle (Table 1). Average HR over the 90 min cycle was 172 ± 3 and 172 ± 4 bpm for CHO and C+P, respectively. Average inline image over 90 min (51.3 ± 1.6 ml kg−1 min−1 for CHO and 51.6 ± 1.2 ml kg−1 min−1 for C+P) and RER over 90 min (0.88 ± 0.01 for CHO and C+P) were not different between treatments. Metabolic data indicated participants were cycling at 77 ± 1% of inline image during CHO and C+P trials. Average RPE over 90 min was similar for CHO and C+P (16 ± 4 for CHO and C+P).

Table 1. Exercise variables
Ex time (min)CHOC+P
  1. Values are the average recording over each 5 min phase and are presented as mean ± SEM. *Significant difference at 25–30 min compared with other time phases (P < 0.05).

HR (bpm)172 ± 5172 ± 3172 ± 1174 ± 4172 ± 1171 ± 2
Cadence (rpm)84 ± 5*79 ± 580 ± 384 ± 4*80 ± 379 ± 5
inline image (ml kg−1 min−1)53.2 ± 2.050.7 ± 2.050.9 ± 2.052.6 ± 1.950.5 ± 1.950.9 ± 2.6
%inline image80 ± 377 ± 276 ± 579 ± 276 ± 277 ± 3
RER0.89 ± 0.010.87 ± 0.020.87 ± 0.020.91 ± 0.010.87 ± 0.020.87 ± 0.01
RPE16 ± 117 ± 117 ± 116 ± 117 ± 118 ± 1

Blood analytes

Fasted blood glucose was 5.1 ± 0.3 mmol l−1 and 5.3 ± 0.3 mmol l−1 for CHO and C+P, respectively and remained similar immediately post-exercise. Approximately 30 min after consuming the first treatment beverage, plasma glucose concentration increased by ∼32% and ∼20% for CHO and C+P, respectively, with no significant difference between treatments. Plasma glucose concentration returned to basal values by 1.5h post-exercise for CHO and C+P and was constant for the remainder of the trial. Fasted serum insulin concentration was 6.0 ± 0.8 and 5.5 ± 0.5 μU ml−1 for CHO and C+P, respectively (Fig. 2A). Following drink ingestion, serum insulin concentration increased for both CHO and C+P, peaking at 30 min post-exercise (P < 0.001). Serum insulin increased to a greater extent for C+P (285 ± 32%) compared with CHO (60 ± 8%; P < 0.001). Serum insulin returned to basal values by 1.5h post-exercise for CHO and C+P and remained constant until the end of the trial. Following exercise, plasma lactate concentration increased by 150 ± 16% and 175 ± 19% compared with resting values for CHO and C+P, respectively (P < 0.001), with no difference between treatments. Lactate concentration returned to basal values by 3h post-exercise for CHO and C+P. Resting plasma urea concentration was similar for CHO and C+P and was stable immediately post-exercise (Fig. 2B). Following ingestion of C+P, plasma urea concentration increased by 20 ± 2% compared with resting values and remained elevated at 4h post-exercise (P < 0.05). Plasma urea concentration remained unchanged for CHO compared with resting values. From 15 min to 4h post-exercise, plasma urea concentration for C+P was significantly greater than CHO (P < 0.05).

Figure 2.

Serum insulin (A) and plasma urea (B) concentrations in CHO and C+P trials
Means within each trial with different subscripts are significantly different from each other (P < 0.05). *Significant difference between CHO and C+P at each respective time point. Values are means ± SEM; n= 10.

Plasma amino acid concentrations

Prior to and immediately post-exercise, plasma amino acid concentrations of phenylalanine, leucine and threonine were similar for CHO and C+P (Fig. 3). Following ingestion of C+P, plasma concentrations of phenylalanine, leucine and threonine increased by 37, 130 and 58%, respectively (P < 0.001) and peaked at 1h post-exercise (30 min after the second drink was ingested), after which, amino acid concentrations returned to basal levels such that there were no differences between CHO and C+P at 4h post-exercise. Following CHO ingestion, plasma phenylalanine, leucine and threonine concentrations were reduced at 30 min compared with immediate post-exercise values (P < 0.05). Phenylalanine and threonine concentrations remained lower for the remainder of the infusion. In contrast, plasma leucine concentration returned to pre-exercise values by 150 min post-exercise.

Figure 3.

Plasma concentrations of phenylalanine (A), leucine (B) and threonine (C)
Means within each trial with different subscripts are significantly different from each other (P < 0.05). *Significant difference between CHO and C+P (P < 0.05). Values are means ± SEM; n= 10.

Plasma and intracellular 13C6 phenylalanine enrichment

Plasma 13C6 phenylalanine enrichment increased from immediately pre- to post-exercise (P < 0.05) but was stable across the time of tracer incorporation, immediately post-exercise to 4h post-exercise for CHO and C+P (7.20 ± 0.03%t/T for both; Fig. 4). These data demonstrate that the additional tracer added to the C+P treatment beverage did not appear in the circulation more rapidly than the amino acids from the protein and that all measurements were made at isotopic equilibrium. Intracellular 13C6 phenylalanine enrichment for an available sub-set of participants (n= 5) was stable across the time of tracer incorporation for CHO and C+P (4.8 ± 0.5 and 4.5 ± 0.7%t/T, respectively; P > 0.05). However, due to the small sub-set of available intracellular samples (n= 5) our calculations revealed intracellular phenylalanine enrichments for eight participants would have been required to achieve sufficient statistical power (≥0.8). Based on a close correlation between available intracellular (n= 5) and plasma enrichments (r= 0.63), the mean predicted intracellular enrichment for the complete study cohort (n= 10) was 5.0 ± 0.1 and 5.1 ± 0.2%t/T for C+P and CHO, respectively. Trial order did not influence the tracer enrichment in plasma (7.0 ± 0.3 and 7.4 ± 0.2%t/T for trials 1 and 2, respectively; P= 0.2) or in the available intracellular samples (4.5 ± 0.4 and 4.8 ± 0.3%t/T for trials 1 and 2, respectively; P= 0.2).

Figure 4.

Enrichment of 13C6 phenylalanine in plasma
%t/T: percentage of tracer-to-tracee ratio. Values are means ± SEM; n= 10.

Post-exercise protein phosphorylation

Immediately post-exercise, mTORSer2448 phosphorylation was similar for CHO and C+P. At 4h post-exercise mTOR phosphorylation tended to increase for C+P (P= 0.1), whereas mTOR phosphorylation tended to decrease for CHO (P= 0.08). A group effect at 4h post-exercise revealed mTOR phosphorylation was greater for C+P compared with CHO (P= 0.02; Fig. 5A). However, there was no group by time interaction for mTOR phosphorylation. Immediately post-exercise, eEF2Thr56 phosphorylation was similar for CHO and C+P. At 4h post-exercise phosphorylation increased 1.4-fold for CHO compared with immediately post-exercise (P= 0.02). Furthermore, eEF2 phosphorylation for CHO was greater than C+P at 4h post-exercise (P= 0.04; Fig. 5B). Immediately post-exercise, p38 MAPKThr180 phosphorylation was similar for CHO and C+P. At 4h post-exercise p38 MAPK phosphorylation was unchanged for CHO and C+P (Table 2). Immediately post-exercise, p70S6KThr389 phosphorylation was similar for CHO and C+P. A time effect revealed p70S6K phosphorylation was increased at 4h for C+P compared with immediately post-exercise (P= 0.05); however, p70S6K phosphorylation was not significantly different from CHO at 4h (Fig. 5C). Immediately post-exercise, 4E-BP1Thr37 phosphorylation was similar for CHO and C+P. At 4h post-exercise 4E-BP1 phosphorylation showed a tendency to increase for CHO and C+P (P= 0.09) with no difference between treatments (Table 2). There was no difference in the phosphorylation of AMPKThr172, AktThr308 and PRAS40Ser246 immediately post-exercise between CHO and C+P. Furthermore, the phosphorylation of AMPK, Akt and PRAS40 remained unchanged from immediately post-exercise at 4h post-exercise for CHO and C+P (Table 2).

Figure 5.

Post-exercise protein phosphorylation of mTORSer2448 (A), eEF2Thr56 (B) and p70S6KThr389 (C)
*Significant time effect for protein phosphorylation at 4h compared with 0h post-exercise (P < 0.05). †Significant group effect for protein phosphorylation at 4h post-exercise between CHO and C+P (P < 0.05). Values are means ± SEM; n= 8.

Table 2. Post-exercise protein phosphorylation
Time post-exercise (h)0404Fold-change 0–4Fold-change 0–4
  1. Values are means ± SEM protein phosphorylation at 0 and 4 h post-exercise (arbitrary units) and fold-change over 4h post-exercise. Protein phosphorylation is expressed as a ratio of phospho to total protein. *Significant increase in phosphorylation at 4h compared with 0h post-exercise (P < 0.05). †Significant difference between CHO and C+P at 4h post-exercise (P < 0.05); n= 8.

AMPKThr1721.40 ± 0.041.36 ± 0.061.45 ± 0.061.33 ± 0.050.95 ± 0.030.89 ± 0.08
AktThr3080.80 ± 0.020.73 ± 0.020.74 ± 0.030.77 ± 0.050.92 ± 0.041.04 ± 0.04
4E-BP1Thr371.71 ± 0.071.77 ± 0.091.66 ± 0.081.77 ± 0.071.03 ± 0.021.04 ± 0.02
eEF2Thr560.65 ± 0.110.91 ± 0.11*0.71 ± 0.080.75 ± 0.07†1.40 ± 0.11*†1.05 ± 0.12
mTORSer24481.53 ± 0.211.40 ± 0.071.49 ± 0.121.58 ± 0.13†0.91 ± 0.091.05 ± 0.05†
P38 MAPKThr1800.65 ± 0.070.62 ± 0.100.61 ± 0.140.64 ± 0.081.15 ± 0.211.28 ± 0.08
p70S6KThr3891.13 ± 0.061.20 ± 0.041.17 ± 0.031.32 ± 0.06*1.06 ± 0.041.13 ± 0.05*
PRAS40Ser2460.83 ± 0.120.80 ± 0.130.70 ± 0.080.72 ± 0.060.97 ± 0.081.03 ± 0.08

Mitochondrial and myofibrillar FSR

Mitochondrial protein synthesis rates were similar for CHO and C+P (95% confidence interval (CI): 0.06–0.12 and 0.06–0.10%h−1, respectively; Fig. 6). Myofibrillar protein synthesis rates were ∼35% higher for C+P compared with CHO (CI: 0.07–0.11 and 0.05–0.07%h−1, respectively; P= 0.025). Rates of mitochondrial and myofibrillar protein synthesis were 30.2 ± 0.9% lower when the unadjusted plasma precursor was used in the calculation of FSR for C+P and CHO (0.061 ± 0.005 and 0.040 ± 0.004%h−1, for myofibrillar FSR, respectively). The difference in myofibrillar FSR between groups remained significant with the unadjusted precursor (P= 0.03).

Figure 6.

Myofibrillar (n= 10) and mitochondrial (n= 8) fractional synthetic rate
*Significant difference between CHO and C+P (P < 0.05). Values are means ± SEM.


This is the first study to investigate the response of various muscle protein fractions to protein ingestion after endurance exercise. The present study findings expand on those of previous investigations (Levenhagen et al. 2002; Howarth et al. 2009) to show that the addition of protein to carbohydrate (C+P) ingestion following 90 min of intense cycling by well-trained individuals stimulates an increase in rates of myofibrillar muscle protein synthesis (MPS) compared with carbohydrate alone (CHO). Interestingly, C+P did not increase mitochondrial MPS rates compared with CHO. The mechanism facilitating the adaptive response of myofibrillar MPS could potentially be due to enhanced mRNA translation as evidenced by differences (albeit marginal) in the phosphorylation of cell signalling intermediates with C+P compared with CHO.

The response of myofibrillar MPS to hyperaminoacidaemia with protein ingestion is not unique to post-exercise recovery from EE. Following RE, Holm et al. (2010) recently demonstrated that myofibrillar MPS increased in response to protein ingestion. No measurement of mitochondrial protein synthesis was made in that study. Additionally, an amino acid infusion increased myofibrillar MPS in resting muscle (Bohe et al. 2001, 2003). In addition, Moore et al. (2009) demonstrated that protein ingestion increased myofibrillar MPS at rest and after RE. Finally, increased myofibrillar protein synthesis was noted with protein ingestion following repeated sprints (Coffey et al. 2010). Taken together with our results, it seems clear that myofibrillar protein synthesis rates are particularly sensitive to hyperaminoacidaemia from ingested protein.

Mitochondrial protein synthesis rates are increased by EE with little response of myofibrillar proteins (Wilkinson et al. 2008). Thus, it may seem logical that others (Levenhagen et al. 2002; Howarth et al. 2009) have assumed that an increase in mixed MPS reported with post-EE protein ingestion would be due, primarily, to mitochondrial proteins. However, this notion is not supported by our data or others (Coffey et al. 2010). Instead, it seems that the increase in mixed muscle protein synthesis with protein ingestion immediately following EE (Levenhagen et al. 2002; Howarth et al. 2009) is attributed more aptly to the increase in the myofibrillar protein synthesis rates. However, a contribution of mitochondrial synthesis to the mixed muscle protein response cannot be completely dismissed. Whereas we studied mitochondrial protein synthesis in well-trained cyclists, previous investigations demonstrated increased mixed muscle protein synthesis in untrained subjects (Levenhagen et al. 2002; Howarth et al. 2009). It has been shown previously (Wilkinson et al. 2008) that training reduces the response of mitochondrial protein synthesis rates to exercise. Thus, it is possible that the exercise bout maximized the acute response in these well-trained cyclists causing a ‘ceiling’ effect of mitochondrial MPS. Therefore, although mitochondrial synthesis rates do not respond to protein ingestion in our trained subjects, it is possible that the increase in mixed muscle protein synthesis rates reported previously (Levenhagen et al. 2002; Howarth et al. 2009) may have included a mitochondrial component. Accordingly, whereas our data provide more information regarding the impact of protein nutrition on the acute stimulation of mitochondrial protein synthesis, clearly there is much more to be learned.

The reason for this lack of response of mitochondrial MPS to protein ingestion, in our hands, is not clear. Moore et al. (2009) demonstrated that the synthesis of the sarcoplasmic protein pool, composed mostly of mitochondrial proteins, was increased by ∼70% with protein ingestion, both at rest and following RE. Thus, these results suggest that rates of mitochondrial MPS increase in response to protein ingestion after RE. A response of mitochondrial protein synthesis to protein ingestion following RE, but not EE, seems somewhat counterintuitive, particularly given that mitochondrial protein synthesis responds to elevated amino acid levels at rest (Bohe et al. 2001, 2003).

Despite the evident responsiveness of mitochondrial protein synthesis rates to protein ingestion in other situations (Bohe et al. 2001, 2003; Moore et al. 2009), we and others (Coffey et al. 2010) were not able to detect an increase in response to protein following intense cycling. However, we cannot dismiss the possibility that the magnitude and/or duration of the response may have been smaller than that which is detectable with our current methods. In line with our data, a recent study by Coffey et al. (2010) showed that when a mixed macronutrient meal was ingested prior to a short bout of high-intensity, repeat-sprint cycling, myofibrillar proteins were preferentially synthesized in the post-exercise period, with no effect on mitochondrial proteins. The precise mechanism to explain the apparent synthesis of only myofibrillar proteins after nutrient ingestion in combination with exercise should be investigated further.

Finally, it is possible that the lack of response may be related to the temporal pattern of the response of mitochondrial proteins to the exercise. This supposition is supported by recent preliminary data suggesting that mitochondrial protein synthesis is much greater at 24h than 6h post-exercise (Burd, Phillips et al. personal communication). Further support comes from evidence that transcription of mitochondrial protein mRNA does not produce elevated mRNA levels for several hours after exercise – at least for some proteins (Yang et al. 2005). Therefore, the stimulation of mitochondrial MPS detected 24h after exercise (Burd, Phillips et al. personal communication) may be due primarily to translation of the new mRNA. Thus, the stimulation of mitochondrial protein synthesis may manifest itself far later than when we made our measurements. However, it should be noted that the 24h response was following RE and, to our knowledge, there are no data investigating the temporal pattern of mitochondrial protein synthesis following EE. Nevertheless, the acute increase in mixed muscle protein synthesis following post-EE protein ingestion noted in previous studies (Levenhagen et al. 2002; Howarth et al. 2009) may not be explained by an acute increase in mitochondrial FSR.

Since training adaptations result, ultimately, from the summation of the acute responses to exercise and nutrition (Hawley et al. 2006), protein consumption may potentiate this response. It is unclear from our results what impact protein ingestion following cycling may have on training adaptations. Increased myofibrillar MPS is normally associated with increased muscle mass and strength in the context of resistance training (Tipton, 2008; Burd et al. 2009). Thus, our data could be interpreted to suggest that protein ingestion following EE would lead to increased muscle mass (and overall body weight) with training, potentially decreasing the power-to-mass ratio. Alternatively, it is possible that muscle hypertrophy may increase power output, thus offsetting the increase in overall body weight. In support of this notion, Hickson et al. (1980) showed that gains in thigh mass after strength training, extended short-term endurance capacity. Alternatively, EE training increases myosin heavy chain I content, a key protein in type I fibres leading to more fatigue resistant fibres (Harber et al. 2004; Kohn et al. 2007). Thus, increased rates of myofibrillar protein synthesis with protein intake after EE may be a contributing factor to potentiate the adaptive response to EE training. Additionally, elevated rates of myofibrillar protein synthesis may simply reflect increased turnover due to increased degradation of myofibrillar proteins. Moderate intensity EE has been shown to stimulate muscle protein breakdown during (Blomstrand & Saltin, 1999; Van Hall et al. 1999) and immediately following exercise (Sheffield-Moore et al. 2004); this is thought to be due, at least in part, to increased breakdown of myofibrillar proteins (Carraro et al. 1990). Therefore, the increase in myofibrillar synthesis may be contributing to repair and remodelling of proteins damaged during the bout. Thus, enhanced myofibrillar protein synthesis with post-exercise ingestion of protein may impact training adaptations by one or more of several mechanisms. Clearly, these mechanisms require more study.

The increase in myofibrillar MPS with post-EE protein ingestion coincided with significant, albeit relatively modest, alterations in the phosphorylation of factors previously associated with increased mRNA translation (Fingar et al. 2004). These findings were surprising given that we were only able to determine signalling phosphorylation at 0 and 4h post-exercise and the phosphorylation of signalling intermediates with anabolic stimuli is relatively transient (Atherton et al. 2010; Camera et al. 2010). EE is known to activate proteins that regulate translation initiation (Mascher et al. 2007; Wilkinson et al. 2008; Camera et al. 2010) (i.e. mTOR) and elongation (Mascher et al. 2007) (i.e. eEF2) following EE. In addition, ingestion of protein increases rates of MPS via mTOR-p70S6K-mediated mechanism (Fujita et al. 2007). Previously, Ivy et al. (2008) demonstrated that mTOR and rpS6 phosphorylation was greater when C+P is ingested after EE as compared with a non-energetic placebo. However, due to the study design, it was unclear whether the effect of C+P was due to greater amino acid availability, elevated plasma insulin or a combination of the two. In contrast, an earlier experiment, in which rats were fed immediately after an exhaustive swim (Morrison et al. 2008) showed that the phosphorylation of translational proteins was sustained with C+P ingestion (Morrison et al. 2008). In line with these data, our results show that protein ingestion after EE modified the translational signalling response in the mTOR-p70S6K pathway and, potentially, contributed towards prolonging the effect of EE on protein synthesis. We also show that eEF2 phosphorylation increased by ∼40% at 4h post-EE in the CHO group, whereas there was no change with protein co-ingestion. Reduced phosphorylation of eEF2, due to eEF2 kinase inhibition, results in a more efficient translocation of the ribosome along the mRNA, thereby contributing to faster elongation and greater MPS (Carlberg et al. 1990). Our findings are further supported by prior human (Fujita et al. 2007) and cell culture (Wang et al. 2001) studies showing that mTOR regulation of p70S6K is inversely related to eEF2 phosphorylation. Together with an increase in initiation due to prolonged mTOR phosphorylation, the suppression of eEF2 phosphorylation (and improved elongation) may have contributed to the rise in myofibrillar MPS for C+P. Finally, we acknowledge that from our findings it is unclear whether alterations in translational signalling with C+P are causative, or merely associative with the increase in myofibrillar MPS.

In conclusion, we have shown that when protein is co-ingested with carbohydrates after cycling exercise myofibrillar, but not mitochondrial, protein synthesis is increased. It is possible that frequent post-endurance exercise protein ingestion may promote muscle hypertrophy over time. Thus, the implications of larger, more powerful muscles for cyclists should be carefully considered prior to making nutritional recommendations. On the other hand, increased myofibrillar MPS may enhance the synthesis of proteins associated with fatigue resistance or may serve to counteract a fasted-state rise in myofibrillar protein breakdown during and immediately following endurance exercise. Thus, the maintenance and structural integrity of contractile proteins may be enhanced. Thus, we posit that post-endurance exercise protein nutrition could have important implications for the adaptive response to endurance exercise and, potentially, the recovery of muscle function. Finally, the synthetic response of different protein fractions to endurance exercise and protein ingestion may be dependent on the individual training status, intensity and duration of the exercise bout, as well as the timing and quantity of post-EE nutrient strategies. Future studies should seek to address these issues in greater depth.


Author contributions

Data collection for the entire study was conducted at the School of Sport and Exercise Sciences, University of Birmingham. All analytical experiments were conducted at the School of Sport and Exercise Sciences at the University of Birmingham, the School of Graduate Entry Medicine & Health at the University of Nottingham and the Functional Molecular Biology Laboratory at he University of California, Davis. L.B., O.C.W., S.R.J., K.B. and K.D.T. contributed to the conception and design of the experiment. L.B., A.P., O.C.W., S.R.J., K.B. and K.D.T. contributed to the collection of the data. L.B., A.P., O.C.W., S.R.J., A.S., K.S., K.B. and K.D.T. contributed to the analysis and interpretation of the data. L.B., A.P., O.C.W., S.R.J., K.S., K.B. and K.D.T. contributed to drafting or revising the content of the manuscript. All authors approved the final version of the manuscript for publication. The authors declare no conflicts of interest.


We would like to thank Daniel Moore, Debbie Rankin and Robert Bagabo for cooperation during method development and data analyses and Ms Lorna Webb for assistance during data collection. We would also like to extend our appreciation to the participants for their time and effort. This study was funded by a research grant from GlaxoSmithKline Nutritional Healthcare, Brentford, UK to K.D.T.