Refining dietary protein recommendations for the athlete


  • Andrew M. Holwerda,

    1. Department of Human Movement Sciences, NUTRIM School for Nutrition, Toxicology and Metabolism, Maastricht University Medical Centre+, Maastricht, 6200 MD, The Netherlands
    Search for more papers by this author
  • Stephan van Vliet,

    1. Department of Human Movement Sciences, NUTRIM School for Nutrition, Toxicology and Metabolism, Maastricht University Medical Centre+, Maastricht, 6200 MD, The Netherlands
    Search for more papers by this author
  • Jorn Trommelen

    1. Department of Human Movement Sciences, NUTRIM School for Nutrition, Toxicology and Metabolism, Maastricht University Medical Centre+, Maastricht, 6200 MD, The Netherlands
    Search for more papers by this author


Accrual of skeletal muscle protein requires a sustained positive muscle protein balance, which is achieved when rates of muscle protein synthesis exceed rates of breakdown. A positive muscle protein balance is stimulated by the combination of a bout of exercise with the ingestion of protein during subsequent recovery. There is constant discussion amongst sports nutritionists and exercise scientists regarding the ‘optimal’ dietary protein requirements for strength- or endurance-trained athletes. A common recommendation of 1.2–1.7 g protein kg-1 day-1 presents a rather large range (102–144 g daily for an 85 kg person) and may be inherently inaccurate based on the range of energy demands in various sports (strength vs. endurance). There is also considerable uncertainty in these recommendations based on the inherently flawed nitrogen balance method and the continually adaptive nature of exercise training. It is also difficult to assess chronic protein ‘efficiency’ as habitually high (or low) protein intake may alter how protein is digested, absorbed and ultimately used for the synthesis of new muscle proteins.

While daily recommendations provide an easy benchmark for the athlete to follow, there may be inherent ‘problems’ with these recommendations. Most notably, aiming to fulfil a daily protein target can create an imbalance of meal-based protein quantities throughout the day. For example, a breakfast meal may contain much less protein than a dinnertime meal. Such an imbalance in the amount of protein consumed at each meal may lead to a suboptimal stimulation of muscle protein synthesis rates over the course of the day. Recent work has demonstrated that ingestion of ∼10 g essential amino acids (equivalent to ∼25 g of high quality protein) is required to maximally stimulate muscle protein synthesis at rest in young adults (Cuthbertson et al. 2005). It appears that ingesting slightly less dietary protein (∼20 g) is required to maximally stimulate muscle protein synthesis rates during the post-exercise recovery phase (Moore et al. 2009a). Given these findings, the evidence is building that a shift from daily to meal-based protein recommendations (e.g. ∼20–25 g protein per meal) may be a good starting point for athletes wishing to optimize skeletal muscle reconditioning following a programme of exercise training.

A recent article in The Journal of Physiology by Areta et al. (2013) provides the evidence base for the recommendation of dietary protein intake on a meal-by-meal basis for athletes. Specifically, the authors investigated the impact of timing and quantity of dietary protein provided in the 12 h recovery period on Akt–mTORC1 signalling and the stimulation of muscle protein synthesis rates in young men. A total of 80 g whey protein isolate was provided in a constant (8 × 10 g ‘pulse’; previously referred to as ‘spread’ by others), cyclical (2 × 40 g ‘bolus’; previously ‘pulse’) or intermediate (4 × 20 g ‘intermediate’) feeding pattern for 12 h following a single bout of resistance exercise. Skeletal muscle biopsies were collected during the primed constant stable isotope amino acid infusion for the calculation of myofibrillar protein fractional synthesis rates at 1–4 h, 4–6 h, 6–12 h and 1–12 h during recovery. The phosphorylation of intramuscular anabolic signalling molecules, which is proxy for their activation, was also measured.

The investigators reported that all feeding patterns stimulated an increase in muscle protein synthesis (88–148% from rest), but the 4 × 20 g (intermediate) feeding pattern effectively elevated muscle protein synthesis beyond that of both the 10 × 8 g (pulse) and 2 × 40 g (bolus) feeding patterns over the cumulative 12 h time period (31–48%). It is noteworthy that the study also addressed a novel, prolonged post-exercise recovery timeline of 12 h. Interestingly, a divergence in muscle protein synthesis rates beyond the 4 h post-exercise period was identified between the feeding patterns. This finding suggests that the ingested amino acids in the bolus and pulse groups were not used as effectively for the synthesis of muscle proteins as the intermediate group, and may have been oxidized or used for urea synthesis.

Previous work has established that the ingestion of 20 g whey protein in close temporal proximity to the exercise bout results in a prolonged stimulation of muscle protein synthesis for ≥5 h (Moore et al. 2009b). Based on the results in the intermediate condition (20 g every 3 h), it would appear that the second ingested dose of 20 g (provided at 3 h post-exercise) might have had little impact on further stimulating the post-exercise muscle protein synthetic response. There were no measurements of leucine oxidation in the study of Areta et al. and, as such, it is difficult to determine whether the dietary amino acids were being oxidized to an appreciable extent. Of course, it would not be overly surprising that skeletal muscle tissue is highly sensitive to dietary protein ingested in the morning following a bout of exercise, and loses amino acid sensitivity over the course of the day and subsequent meals.

Indeed, the investigators provided excellent insight into the influence of meal timing and protein quantity on muscle protein synthesis during a normal daily feeding period. However, in order to optimally increase skeletal muscle mass, the entire day including the overnight rest period should be considered. It has been demonstrated that the digestion and absorbance of casein protein provided during overnight rest is similar to that in waking hours (Groen et al. 2012) and, in fact, casein provided before sleep may be absorbed more rapidly than casein throughout the day. Furthermore, it has been demonstrated that rates of overnight mixed muscle protein synthesis are stimulated by ingesting 40 g of casein after resistance exercise and before sleep when compared to a placebo (∼22%) (Res et al. 2012). These data demonstrate that protein ingested after exercise, prior to overnight rest, may be a viable feeding strategy to optimize muscle protein synthesis during the otherwise fasting period of overnight rest. Thus, data from the present study can be applied in combination with data from newly emerging overnight studies to recommend optimal feeding strategies over an entire 24 h post-exercise period.

It is now becoming evident that athletes should focus on ingesting effective meal-based protein quantities as opposed to merely following daily recommendations. The findings from the current study reinforce the meal-based approach as post-exercise muscle protein synthesis rates were effectively sustained with the optimal, ‘intermediate’ feeding pattern. We congratulate Areta et al. on the novel post-exercise time period examined in this study, as they demonstrated a divergence in the stimulation of muscle protein synthesis beyond the immediate 4 h after exercise. The future direction of research in this area is very exciting as this study and others are beginning to examine acute muscle protein synthesis over an elongated period of exercise recovery. By studying the effects of post-exercise feeding patterns and the influence of meal-to-meal interactions on muscle protein synthesis, we can provide more accurate feeding recommendations to optimize muscle reconditioning and performance in athletes.


Additional information


We would like to thank Dr Nicholas Burd and Dr Luc van Loon for their helpful edits of this manuscript.

Competing interests