Endurance exercise affects skeletal muscle by reducing energy stores and increasing muscle protein breakdown. Although a small amount of glycogen is stored in the liver, the primary energy source during endurance exercise is glycogen stored in skeletal muscle 1. Exercise duration and intensity can be limited by glycogen availability, emphasizing the importance of replenishing these energy stores prior to subsequent exercise bouts 2. Exercise also increases muscle protein degradation. Muscle protein breakdown occurs continually, even at rest, releasing amino acids into the intracellular fluid and bloodstream to be used for protein synthesis or oxidized for energy 345. Protein synthesis is stimulated by exercise, but consumption of food must offset breakdown to create a positive net muscle protein balance 67.

Following exercise, acute physiological changes occur in the muscle that promote glucose uptake, glycogen accumulation and protein synthesis 689, but optimal replenishment of the energy stores and net protein balance are dependent on post exercise nutritional content and timing 101112. While glycogen synthesis requires glucose, protein synthesis requires amino acids. Combining carbohydrate with protein increases stimulation of the insulin-signaling and mTOR pathways, increasing both glycogen and protein synthesis 131415, suggesting that the ideal recovery food must contain both carbohydrate and protein to provide substrate for glycogen synthesis and achieve net protein balance.

In addition to the composition of the post-exercise food, exercise duration, intensity and training status influence glycogen and skeletal muscle protein status 116171819. While many exercise protocols used in research are designed to clearly observe post supplementation glycogen and muscle protein changes, these protocols are not typical training sessions for most individuals. For example, glycogen synthesis rate and amount are maximized when subjects exercise to exhaustion to deplete glycogen stores prior to supplementation 11819. Similarly, protein breakdown and subsequent synthesis is acutely higher after resistance exercise and supplementation in untrained compared to trained subjects 17. Protocols including a more realistic training scenario and foods such as cereal and nonfat milk may be equally effective in observing responses to post exercise supplementation as compared to using exhaustive protocols or untrained subjects.

Although muscle response during recovery to a carbohydrate-protein drink may be similar to that seen after whole-grain cereal and nonfat milk, we chose to compare a carbohydrate-only drink. Recreational athletes may be more familiar with carbohydrate drinks due to high product awareness and accessibility, and may not understand the benefit of added protein in post-exercise supplementation. Our goals were to use ordinary foods after moderate exercise to understand relative effects on glycogen repletion, and the phosphorylation state of proteins controlling protein synthesis for the average individual. Cereal and milk were selected since both are readily available, popular foods that are inexpensive and easily digested. Our hypothesis was that cereal and nonfat milk would be more effective than a popular carbohydrate-electrolyte sports drink in increasing muscle glycogen and the signaling activity of proteins controlling protein synthesis after moderate endurance exercise.

While both a 100% whole grain cereal and nonfat milk (Cereal) and 6% carbohydrate-electrolyte beverage (Drink) increased glycogen following moderate exercise, significant phosphorylation of mTOR and AKT only occurred after Cereal. Prior research has focused on comparing the effects of carbohydrate and carbohydrate-protein post-exercise supplementation on either glycogen 132829 or protein 714 synthesis after exercise. Our research examined the effects of readily available foods on glycogen synthesis and the phosphorylation state of proteins controlling protein synthesis after a typical cycling endurance workout.

After endurance exercise, glycogen is reduced and protein synthesis increased; however, the rate of protein degradation exceeds protein synthesis 17. Recovery foods that target either glycogen storage or protein synthesis can potentially affect future exercise performance by compromising muscle protein or energy stores, respectively. Reduction in glycogen, increased glycogen synthase activity, and increased insulin sensitivity prime the muscle for glycogen synthesis post exercise; however, glucose substrate must be available to support glycogen accretion 91930. Although protein synthesis also increases after resistance and endurance exercise, without substrate, net protein balance is not positive, only less negative 67. Food containing essential amino acids (EAAs) must be consumed to achieve a positive net protein balance 4 and insulin must also be present 313233. In our research, the carbohydrate in Drink supplied substrate for glycogen storage, but Cereal provided carbohydrate and EAAs necessary to support both glycogen and protein synthesis (Table 2).

As expected, insulin secretion during recovery was higher for Cereal compared to Drink, possibly due to the amino acids in the nonfat milk 1334. The plasma glucose AUC was lower after Cereal due to higher insulin and resultant increased glucose uptake by the exercised muscle, similar to other studies comparing carbohydrate-protein and carbohydrate recovery foods 132235. However, plasma lactate levels were significantly lower after Cereal compared to Drink. This drop in lactate is similar to that observed by Ivy et al. 29 after a carbohydrate-protein (80 g CHO, 28 g PRO, 6 g FAT) beverage, but not after isocarbohydrate (80 g CHO, 6 g FAT) or isocaloric (108 g CHO, 6 g FAT) carbohydrate beverages. Since plasma lactate is not a primary substrate for glycogen synthesis in the fed state 36, it is possible that a higher percentage of glucose was taken up by the muscle and stored as glycogen after Cereal rather than converted to lactate. While both treatments increased glycogen, we did not observe a difference between treatments, possibly due to the low sensitivity of the biopsy procedure or insufficient time to detect a difference.

Phosphorylation of Akt increased for Cereal but not for Drink, possibly coupled to the higher insulin levels after Cereal (Figure 6). In addition to increasing GLUT4 concentration at the cell membrane, Akt deactivates glycogen synthase kinase 3 (GSK-3), which allows activation, or dephosphorylation, of glycogen synthase 373839. Normally after exercise, glycogen synthase is activated to stimulate glycogen storage. As glycogen accretion occurs, glycogen synthase becomes phosphorylated, reducing glycogen synthase activity. Both Cereal and Drink increased glycogen, but compared to Drink, Cereal had lower glycogen synthase phosphorylation, suggesting that the greater Akt phosphorylation continued to stimulate glycogen synthase activity 60 minutes after Cereal despite elevated glycogen (Figure 5).

Akt also phosphorylates the mammalian target of rapamycin (mTOR), stimulating downstream phosphorylation of proteins controlling translation 40414243. In addition to Akt, mTOR is stimulated by amino acids, particularly leucine, either directly or indirectly 334445 but not aerobic exercise 154647. Unlike Drink, Cereal had a significant effect on mTOR and Akt phosphorylation (Figure 6), implying that mTOR was activated by Akt and also by the amino acids in the nonfat milk. The high correlation of Akt and mTOR for Drink but not for Cereal suggests that mTOR was directly stimulated by Akt for Drink and primarily through the alternate amino acid pathway for Cereal.

Activation of mTOR increases phosphorylation of p70^S6K, which activates ribosomal protein S6 (rpS6), a substrate of p70^S6K. rpS6 can also be activated by exercise through the extracellular signal-regulated kinase 1/2 (ERK1/2) through phosphorylation of p90RSK and p38 mitogen-activated protein kinase (MAPK) pathways 48495051. The significant increases in phosphorylation of rpS6 were almost identical between Cereal and Drink (Figure 6), unlike recent human and animal studies, suggesting an exercise effect. Karlsson et al. 43 observed a slight elevation in p70^S6K phosphorylation and corresponding rise in rpS6 phosphorylation in men 1 hour after resistance exercise followed immediately by a placebo beverage; however, the phosphorylation of both p70^S6K and rpS6 were significantly higher when a branched-chain amino acid (BCAA) drink was consumed after exercise. Similar to Karlsson, our lab has observed increased rpS6 phosphorylation 45 minutes after cycling exercise after both placebo and carbohydrate-protein beverages, although rpS6 phosphorylation was significantly higher after carbohydrate-protein compared to the placebo beverage 47.

Our lab has also observed timing of rpS6 phosphorylation in rats that was highly correlated to insulin 15. rpS6 phosphorylation was higher 30 minutes post exercise in animals given carbohydrate-protein post exercise compared to fasted, exercised controls. Interestingly, rpS6 phosphorylation was significantly increased at 90 minutes in animals that did not receive supplementation. At both time points, insulin was elevated in the respective animal groups compared to exercised controls. In the current study, we would expect the higher insulin and mTOR phosphorylation at 60 minutes after Cereal to result in higher rpS6 phosphorylation compared to Drink, but that did not occur, possibly due to the amount of supplementation provided or biopsy timing. The nearly identical increase in rpS6 phosphorylation for both Cereal and Drink suggest that these changes were due to exercise and independent of supplementation.

For translation initiation to occur, mTOR must increase phosphorylation of eukaryotic translation initiation factor 4E (eIF4E) binding protein 1 (4E-BP1), releasing eIF4E to bind to eIF4G, forming the eIF4F complex. Phosphorylation of eIF4E may be affected by phosphorylation of MAP kinase interacting serine/threonine kinase 1 and 2 (MNK1/MNK2) 52. Ueda et al. 52 established that changes in p38 MAPK phosphorylation of MNK1 directly influenced the levels of eIF4E phosphorylation while ERK1/2 activates both MNK1 and MNK2, but primarily affects the basal level of eIF4E phosphorylation. The role of phosphorylated eIF4E in protein synthesis is unclear; while some studies have concluded that phosphorylation of eIF4E is necessary for translation 53 others have not 525455. We observed a slight, insignificant decrease in phosphorylation of eIF4E after both Drink and Cereal, with no difference between treatments (Figure 6). This lack of change in phosphorylation of eIF4E between treatments agrees with the findings of Gautsch et al. 31, who observed no change in post-exercised rats that consumed saline, carbohydrate or a mixed meal. In addition, there was no difference in phosphorylation of eIF4E between fasted-rested rats and all exercise groups, suggesting that exercise did not affect eIF4E phosphorylation.

The form of our recovery foods did not seem to affect our results, although the rate of gastric emptying would be expected to be lower for solid food versus liquid food. Reed et al. 56 did not find a difference in liquid versus solid food for glycogen synthesis but provided a larger amount of carbohydrate (1.5 g•kg^-1 BW) in two feedings and studied the effects after 4 hours. Differences in gastric emptying rates between solid and liquid food may further change the respective appearance rates. Also, independent of the form, the splanchnic clearance rates of EAAs are not the same, so entry of amino acids into plasma will not match the ratio contained in the food 4. Liquid carbohydrate-protein and carbohydrate-free AA supplementation has been studied with respect to effects on protein synthesis, but direct comparisons between solid and liquid food are not as available 144657. The increase in Akt and mTOR phosphorylation, and increased glycogen in the current research, suggests that the solid whole grain cereal cleared the GI tract and was sufficiently available to the exercised muscle within 60 minutes after Cereal.

A possible limitation in our study design was the timing of the second muscle biopsy. Glycogen and protein synthesis occur at different rates, but prior research has not identified an optimal measurement strategy to detect concurrent changes. We considered 60 minutes post treatment to be sufficient to observe changes in both glycogen levels and proteins involved in translation initiation, the rate-limiting step in protein synthesis. Ivy, et al. 29 compared carbohydrate and carbohydrate-protein supplementation effects on glycogen levels after endurance exercise, testing glycogen at multiple time points using ¹³C-NMR. The glycogen accretion after a carbohydrate-protein and isocarbohydrate beverage differed between 20 and 60 minutes then converged at 2 hours. Their post exercise glycogen levels were lower and caloric content of the food higher compared to the current study, which can increase the synthesis rate during the first hour of recovery 355859. The rate of glycogen storage in the current study was suboptimal, even with supplementation, because the moderate cycling exercise did not deplete the glycogen level to support the maximal replenishment rate 58. However, with the higher amount of active glycogen synthase and phosphorylated Akt in Cereal, we may have seen a greater amount of glycogen storage with additional supplementation and subsequent muscle biopsies.

Increased phosphorylation of proteins involved in protein synthesis has been observed within 30 minutes of both solid and liquid supplementation. Vary and Lynch 60 biopsied rested rats at 30 and 60 minutes after feeding a mixed meal. Although phosphorylation of mTOR, Akt and p70^S6K remained elevated at 60 minutes compared to pre-feeding levels, phosphorylation was highest at 30 minutes. Research in our lab has shown significant increase in phosphorylation of mTOR and rpS6 in humans 45 minutes after post-exercise supplementation 47. Our results suggest that 60 minutes was sufficient to show a change in these proteins, but we may have not observed peak phosphorylation after supplementation. Further research is necessary to better understand relative timing of the phosphorylation of proteins controlling protein synthesis in humans with respect to exercise and supplementation.

Although the combination of protein and carbohydrate in Cereal affected the muscle differently than the carbohydrate in Drink, glycogen accretion and phosphorylation of proteins controlling the initiation of protein synthesis, except mTOR, were similar. This suggests that readily available foods such as cereal and nonfat milk can provide post-exercise supplementation and be used in lieu of a commercially-available sports drink after moderate exercise. Cereal and nonfat milk provide a less expensive whole food option as compared to sports drinks. It also provides easily digestible and quality protein in the milk, which could promote protein synthesis and training adaptations, unlike a carbohydrate sports drink. This is a potential option for individuals who refuel at home.

4E-BP1: eukaryotic translation initiation factor 4E binding protein 1; AA: amino acid; Akt: phosphatidylinositol 3-kinase, protein kinase B, aka PKB; ANOVA: analysis of variance; AUC: area under the curve; BCAA: branched-chain amino acids; bpm: beats per minute; ¹³C-NMR: Carbon Nuclear Magnetic Resonance Spectroscopy; Cereal: whole grain cereal with nonfat milk; CHO: carbohydrate; Drink: 6% carbohydrate-electrolyte drink; EAA: essential amino acid; EDTA: ethylenediamine tetraacetic acid; eIF4E: eukaryotic translation initiation factor 4E; End: at end of exercise; ERK: extracellular signal-regulated kinase; GSK: glycogen synthase kinase; MAP: mitogen-activated protein; MAPK: mitogen-activated protein kinase; MAPK: mitogen-activated protein kinase; MNK: mitogen-activated protein kinase interacting serine/threonine kinase; mTOR: mammalian target of rapamycin; PCA: perchloric acid; Post0: immediately prior to supplement; Post15: 15 minutes after supplement; Post30: 30 minutes after supplement; Post60: 60 minutes after supplement; Pre: before exercise; PRO: protein; RER: respiratory exchange ratio; RPE: rate of perceived exertion; rpS6: ribosomal protein S6; SDS-PAGE: sodium dodecyl sulphate polyacrylamide gel electrophoresis; Ser: serine; TBS: tris-buffered saline; TIFF: tagged image file format; TTBS: tween-20 tris-buffered saline

The authors declare that they have no competing interests.

LK recruited subjects, performed VO_2MAX tests, coordinated trial personnel, performed lactate assay, performed all statistical analysis and wrote document. ZD handled blood, assisted during VO_2MAX tests and trials, supervised assays, ran insulin assay, made reagents used in assays. BW handled blood, assisted during trials, performed glycogen assay. DH performed Western blots. YHL performed Western blots. JI defined the protocol, wrote and acquired grant, performed muscle biopsies, directed muscle tissue assays, reviewed and wrote portions of document. All authors read and approved the final manuscript.