It is well established that dietary protein plays an important role in both training adaptation and recovery (1). Dietary amino acids are required to replace endogenous oxidative losses and to provide the building blocks to repair and remodel body (including muscle) tissue during recovery (2,3). It is generally accepted that protein requirements are elevated in athletes (1), as reflected by current, generally nonspecific dietary protein consensus recommendations for athletes of 1.2–2.0 g·kg−1 body weight·d−1 (1). In the case of endurance athletes, these recommendations are largely based on research examining protein intakes required to achieve nitrogen balance (NBAL) (4,5), which as a method may underestimate true requirements (6). Nevertheless, Meredith et al. (4) demonstrated a mean protein requirement needed to maintain NBAL of 0.94 g·kg−1·d−1 with a recommended dietary allowance (RDA) suggested to cover 97.5% of the population at 1.26 g·kg−1·d−1, which is similar to other reports (5). However, identification of a protein intake that maintains NBAL without an appreciation for whether it can also optimize whole body protein metabolism or, more importantly from an athlete perspective, training quality and exercise performance may not provide the athlete with the most relevant nutritional advice (7). For example, a protein intake of 3.0 g·kg−1·d−1 was reported to attenuate the decline in exercise performance during an intensified training period compared with an apparently sufficient 1.5 g·kg−1·d−1 (8), which highlights that an ecologically valid protein intake for endurance athletes may be greater than current recommendations based on NBAL alone. Given that additional studies have attempted to clarify whether higher (i.e., >1.2 g·kg−1·d−1) protein intakes support the metabolic demand for dietary amino acids and maintain exercise performance with varying results (9–11), additional prospective controlled research is warranted to identify the nutritional intake that would enhance protein metabolism while concurrently supporting an endurance athletes’ training and performance goals.
We recently demonstrated using the indicator amino acid oxidation (IAAO) method that protein requirements in an endurance-trained population were as high as 1.83 g·kg−1·d−1 after a 20-km run as part of a 3-d, 35-km controlled training period (12). The benefit of IAAO is that it identifies the protein intake that maximizes whole body protein synthesis (6), which ostensibly would support greater postexercise recovery (7). In potential support, Rowlands et al. (9) demonstrated that sprint power was greater 60 h after an intense cycle training bout when consuming a controlled diet providing 1.9 g·kg−1·d−1 as compared with 1.2 g·kg−1·d−1 of protein. This performance benefit was also accompanied by a positive NBAL only on the higher protein diet (9). By contrast, in a subsequent study, these authors (13) demonstrated that postexercise performance recovery was not different between diets containing dietary protein at 1.5 or 1.9 g·kg−1·d−1 when NBAL is positive. Thus, dietary strategies targeted at enhancing protein metabolism and/or whole body anabolism may be optimal to support a high training quality. Moreover, higher protein diets (either through whole foods and/or with supplementation) may be ideally suited to maintain exercise performance within the context of a higher frequency (i.e., multiple as compared with single bouts) daily exercise stimuli (8,14), which would be characteristic of endurance athletes’ habitual training.
The purpose of this study was therefore to determine the effect of protein doses spanning the IAAO-derived safe intake in nonexercising adults to a safe intake for endurance athletes (i.e., 0.94 to 1.83 g·kg−1·d−1, respectively) (6,12) on protein metabolism and exercise performance during a period of controlled endurance training. Given that endurance athletes have been reported to consume on average ~1.5–1.6 g·kg−1·d−1 but also report moderate day-to-day variation in actual intakes (~11%–30% variation) (15–17), a majority of these athletes may consume suboptimal protein intakes (i.e., <1.83 g·kg−1·d−1), even if only briefly, during the course of their normal training. Thus, we determined both the metabolic and performance effects of short-term (i.e., 4 d) controlled diet and training as we believe that protein recommendations that consider both of these aspects are more relevant for athletes aiming to enhance recovery from and/or adaptation to training. We hypothesized that a higher protein intake (1.83 g·kg−1·d−1) would induce a positive whole body net protein balance and that this would translate into a greater exercise performance compared with a moderate (1.2 g·kg−1·d−1) and lower (0.94 g·kg−1·d−1) protein intake.
METHODS
Ethics statement
All participants were informed of the purpose of the study, the experimental procedures, and all the potential risks involved before written consent, which was obtained in their first session. This study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Health Sciences Research Ethics Board at the University of Toronto (REB#32895) and the institutional review board of Ajinomoto Co., Inc. This trial was registered at clinicaltrial.gov as NCT02801344.
Participants
Healthy endurance-trained males between the ages of 18 and 50 yr were recruited (Table 1). Completion of a Physical Activity Readiness Questionnaire (PAR-Q+) was obtained in the preliminary session. Participants were considered trained if they had a recent (over the past 4 wk) running history of ≥45 km·wk−1 or 4.5 h·wk−1. Inclusion in the study required their age-specific V˙O2max to be at least “very good” (18). To determine this, a ramp protocol was used where a constant running speed was maintained while treadmill incline (starting at 0%) was increased by 2% every 2 min. V˙O2max was defined as the oxygen consumption at which the subject’s RER was 1.15, HR (bpm) of the age-predicted maximum was reached, and/or the subject was unable to continue running. Participants were excluded if they reported use of tobacco or anabolic drugs (e.g., growth hormone, testosterone, etc.).
General anthropometric measurements (i.e., body weight, height, and composition) and resting energy expenditure (REE) were obtained in the morning after an overnight fast in the participants’ second visit. The participant’s REE was measured using open circuit indirect calorimetry (GA-300; iWorxSystems Inc., Dover, NH). Calculations were made from oxygen consumption (V˙O2) and carbon dioxide production (V˙CO2) using the abbreviated Weir equation (19). Body composition was measured via air displacement plethysmography with a correction for internal volumes (Bod Pod, COSMED USA Inc., Chicago, IL). This method measures body mass (weight) using a calibrated scale and volume by sitting inside the Bod Pod to determine body density. Body density can then be used to estimate fat mass and fat-free mass.
Study design
Each participant completed three separate intervention trials in a double-blind, randomized three-way crossover design separated by at least a 7-d washout period (see Fig. 1). There were three experimental blocks that each included performance testing before and after the 4-d period of controlled training. In the 2-d period (days −2 and −1) before each trial, participants consumed their habitual diet (HAB) before beginning 4 d (days 1–4) of intensified volume training (INT). During this training period, dietary intake was controlled with participants receiving a variable amount of protein (i.e., LOW, 0.94 g protein·kg−1·d−1; MOD, 1.20 g protein·kg−1·d−1; HIGH, 1.83 g protein·kg−1·d−1). The LOW diet represents the estimated average requirement for endurance athletes as determined by NBAL (4) and a safe intake for nonexercising adults based on the IAAO (6). The MOD diet represents the lower limit of the current recommended protein intake for endurance athletes by the American College of Sports Medicine (1) and a safe intake according to NBAL (4,5). The HIGH diet reflects the protein intake that was recently reassessed using the novel IAAO tracer method as being sufficient to maximize whole body protein synthesis (12). The energy intake was set to cover their daily energy expenditure and exercise-induced energy expenditure and with an appropriate target carbohydrate intake (i.e., 6.0–9.0 g·kg−1·d−1) (1). During INT, participants were required to run 20, 5, 10, and 20 km·d−1 over four consecutive days (days 1–4). This volume of training (i.e., 55 km over 4 d) was similar to their normal weekly training volume (see Table 1) but condensed over 4 d as compared with 7 d. These training volumes were completed at the runner’s convenience. A heart rate and GPS monitor (M400, Polar Electro, Kempele, Finland) was provided for participants to wear during training sessions to monitor the quantity and intensity of each training bout. There was no significant difference (P > 0.05) in heart rate (149 bpm; 79.3% HR max) or pace (5.33 min·km−1) of the participants during their 20-km training runs between trials, suggesting training quality was not influenced by the protein intake. Participants were asked to refrain from all other exercise during the day except normal daily activities (e.g., commuting, shopping, etc.) and were required to wear an accelerometer (wGT3X-BT; ActiGraph, Pensacola, FL) all day (except when bathing) during trials to monitor their habitual activity. Whole body protein metabolism for 24 h was assessed using the oral [15N]glycine method on days 1 and 4 of the controlled training block. Days 1 and 4 were selected to track changes across time and to align with the training volume (i.e., 20 km) that was used in our previous study to determine the safe protein intake that maximizes whole body protein synthesis (12).
;)
FIGURE 1: Study overview.
Diets and test drinks
Participants consumed whole food controlled diets providing 0.8 g·kg−1·d−1 of protein, 6–9 g·kg−1·d−1 of carbohydrate, and an energy intake of 1.6 × REE. To increase the daily protein intake to the target levels while maintaining the double-blind nature of the trial, participants were provided with three isocaloric test drinks in opaque bottles consisting of a complete profile of crystalline amino acids modeled after the amino acid composition of egg protein (12), which provided a total of 0.14 (LOW), 0.40 (MOD), or 1.03 g·kg−1·d−1 (HIGH) additional protein. The test drinks consisted of a crystallized amino acid mixture (Ajinomoto North America, Inc.); Tang flavor crystals (Kraft Canada ULC); carbohydrate as Polycal® (Danone Nutricia); a nutritionally complete, protein-free powder (PFD-1) (Mead Johnson & Company, LLC); grape seed oil (generic); Splenda® noncaloric sweetener (Heartland Consumer Products, LLC); and Poweraid zero® drops (Coca Cola, Ltd.). The experimental beverages were prepared by a separate investigator whose only responsibility within the trial was to prepare the beverages. Participants were instructed to consume each beverage at least 3 h apart with one immediately after exercise and the other two in-between main meals of the day (e.g., midafternoon and before bed). The remaining calorie content of the diet not provided by carbohydrates or amino acids was provided as fat. A supplemental multivitamin tablet (Centrum®, Pfizer Canada Inc.) was also provided to ensure daily vitamin requirements for a healthy adult were being met. Participants were instructed to not consume anything other than water outside of the prepared meals and snacks and were required to complete a dietary checklist to indicate that they had consumed the provided meals.
Protein metabolism
On the morning of days 1 and 4 of each trial, a baseline spot urine was collected. Subsequently, participants consumed 2 mg·kg−1 of [15N]glycine dissolved in water with all urine collected over the following 24 h used to noninvasively assess whole body protein kinetics (20). All urine produced after the initial morning spot urine up to the first urination of the following morning was collected, pooled, and stored at 4°C to obtain a daily protein turnover rate.
Urinary [15N] enrichments (i.e., tracer–trace ratio, t/Tr) of ammonia (at baseline and 24 h) and urea (at baseline and 24 h) were determined in duplicate by isotope ratio mass spectrometry by Metabolic Solutions Incorporated (Nashua, NH) to determine whole body nitrogen turnover and protein metabolism (20). Briefly, 24-h whole body nitrogen turnover (Q) kinetics were calculated using the harmonic mean of [15N]ammonia and [15N]urea end-product enrichment after correcting for baseline, pretracer enrichment. Q was calculated as g N·kg−1·h−1 = d/corrected t:Tr·t−1·BM−1, where d is the dose of oral [15N]glycine, corrected t/Tr is the baseline corrected [15N] enrichment, t is the time (i.e., 24 h), and BM is the participant’s body mass. Whole body protein synthesis (S) was calculated as S (g protein·kg−1·d−1) = (Q – E)/(t × BM) × 6.25, where E is total nitrogen excretion, which included urinary and miscellaneous losses. Measured nitrogen excretion was the sum of urinary urea nitrogen and creatinine nitrogen excretion, which represents >75% of all daily nitrogen excretion in an athletic population (21). Miscellaneous nitrogen excretion (e.g., sweat, fecal) was estimated according to previously published values in a trained endurance running population consuming 1.7 g·kg−1·d−1 of protein (21). Whole body protein breakdown (B) was calculated as B (g protein·kg BW−1·h−1) = (Q × IN)/(t × BM)/6.25, where IN is nitrogen intake provided by the controlled diet or habitual diet. S and B using the harmonic mean of Q were calculated using 24-h nitrogen excretion.
Performance testing
After familiarization with all tests, exercise performance before and after the 4-d controlled diet and training period included a 5-km time trial (5kmTT), maximal isometric voluntary contraction (MVC), and jump impulse (IMP) and peak force (PF).
Performance testing began with the MVC, which required participants to sit in a chairlike apparatus with their right leg secured to an immovable arm in a bent 90° position by straps on the lower leg. Participants were asked to perform a maximal isometric contraction of their knee extensors by attempting to straighten their leg against the lower pad of the machine for ~5 s. Participants were allowed three 5-s attempts with 1 min of rest in between. The single greatest maximal torque within the 5-s effort was determined by PowerLab with LabChart Pro v.8.0.5 (ADInstruments Inc., Colorado Springs, CO), and the attempt that attained the greatest force value was used in the inferential statistics. MVC was selected to provide an estimate of muscle fatigue and/or muscle damage. In our hands, the test coefficient of variation was 14.1% with an intraclass correlation of 0.885 as calculated from all pretrials (n = 30).
Jump metrics were then assessed via a vertical jump test using Advanced Mechanical Technology Inc. (AMTI) NetForce software on force platforms (AMTI Force Platform model BP600900, Watertown, MA). This test was used to measure differences in neuromuscular fatigue associated with the concentric power production of the leg extensors (22). To begin each test, participants stepped on the force plate and were asked to stand still briefly (~3 s), before directive from the investigator to jump. Participants were instructed to descend to a comfortable depth before jumping as high as possible (no pauses were required and arm swing was permitted). Force plate output signals were amplified and converted (AMTI GEN 5 Amplifier) to a digital ground reaction force for each millisecond over the 10-s collection period. Jump variables were analyzed from ground reaction force data in Excel, in a manual and blinded fashion, using previously described criteria/calculations (22). Participants were given three jumps to attain maximal jump height separated by 1 min of recovery. Jump IMP and PF exerted on the force platform during the takeoff phase of the jump were assessed. The single best jump was used in the inferential statistics.
Finally, the 5kmTT was run at a 0° slope on a powered treadmill (LifeFitness 9500HR, Brunswick and Co., Mettawa, IL). Participants were free to adjust the speed as desired using the buttons on the treadmill, but no information was given on heart rate, speed, or time; the participants were only able to see the distance they had covered. Water was provided for participants to consume ad libitum during the entire test. This test has a high specificity to the improvements in performance desired by this population and also has a high degree of reliability (23). In our hands, the test coefficient of variation was 5.1%, with an intraclass correlation of 0.955 as calculated from all pretrials (n = 30).
Muscle soreness
Corresponding subjective ratings of muscle soreness were collected from participants using a visual analog scale for the thigh, hamstring, and calf regions. Delayed onset muscle soreness (DOMS) ratings incorporated a 100-mm line, with 0 indicating no pain and 100 representing extremely painful. Participants were asked to mark their perceived soreness on the 100-mm line. Distance from the edge of the line (0) to the marked point was measured in millimeters, and this value was used for the analysis (24).
Statistics and data analysis
The sample size was based on differences in NBAL previously determined by Meredith et al. (4) in endurance-trained men consuming diets providing 0.9 and 1.2 g protein·kg−1·d−1 during habitual training, which are similar to our LOW and MOD intakes. Assuming a similar response in whole body protein balance as with whole body NBAL when consuming 0.9 and 1.2 g protein·kg−1·d−1 (i.e., ~2.3 and ~20.9 mg nitrogen·kg−1·d−1, respectively; common SD of 14 mg nitrogen·kg−1·d−1) with α = 0.05, β = 0.80, and accounting for ~10% dropout, n = 10 participants were determined to be sufficient to detect differences in whole body net protein balance between diets differing by ~0.3 g protein·kg−1·d−1.
Data were analyzed by generalized linear mixed effect model by IBM SPSS Statistics Software 24.0 with a two-way (time–condition) repeated-measures ANOVA. Standardized differences from pre- to postexercise were analyzed using a one-way (condition) repeated-measures ANOVA. Differences between means for significant main effects or interactions were determined using a Holm–Sidak post hoc test. To determine whether NB was significantly positive or negative, a paired samples t-test was used to compare differences from zero for all groups. Statistical significance was established at P < 0.05, and all metabolism data are expressed as means ± SD.
Performance variables were also analyzed using a one-way (condition) repeated-measures ANOVA, and the magnitude of change was examined through effect size (ES) calculation. Standardized difference thresholds were inferred as very small = 0.01–0.2 (25), small = 0.2–0.5, moderate = 0.5–0.8, and large = >0.8 (26). Performance data are expressed as ES (confidence interval).
RESULTS
Whole body protein metabolism
There were no group or time differences in S or B (P ≥ 0.33) (Figs. 2A and 2B). Whole body net protein balance demonstrated a main effect for condition (HIGH > MOD > LOW, P < 0.05) (Fig. 2C), and within-group differences were seen on days 1 (HIGH > MOD > LOW, P < 0.05) and 4 (HIGH > LOW, P < 0.05). HIGH was in positive net balance (P < 0.049), and LOW was in negative net balance (P < 0.05) over the duration of the controlled diet.
FIGURE 2: Whole body protein synthesis (A), protein breakdown (B), and net protein balance (C) on the first (day 1) and last (day 4) of the controlled diet and training period. Capital letters indicate main effect of condition (P < 0.05). *Different from LOW within day (P < 0.05). †Different from MOD within day (P < 0.05). Mean ± SD.
Effect sizes for posttraining period synthesis rates were large (ES = 1.04), small (ES = 0.21), and moderate (ES = 0.69) for HIGH over LOW, HIGH over MOD, and MOD over LOW, respectively. Effect sizes for posttraining period breakdown rates were large (ES = 0.88), very small (ES = 0.08), and moderate (ES = 0.57) for HIGH over LOW, HIGH over MOD, and MOD over LOW, respectively.
Performance
5kmTT was not different (P = 0.793) between conditions during the pretest (Fig. 3A). No effect of group (P = 0.415) or time (P = 0.660) was observed but there was a trend (P = 0.06) for an interaction for 5kmTT. Effect sizes are shown in Table 2. Mean pre- to posttest changes in 5kmTT time were +7.7 ± 19.2 s (~0.8% slower), +8.6 ± 38.4 s (~0.7% slower), and −7.5 ± 20.4 s (~0.7% faster) for LOW, MOD, and HIGH, respectively. The standard differences for HIGH over MOD and HIGH over LOW were −16.0 ± 28.91 s (P = 0.67) and −15.2 ± 36.21 s (P = 0.23), respectively (Fig. 3B). The standard difference for MOD over LOW was −0.84 ± 36.21 s (P = 1.00).
FIGURE 3: 5kmTT (A) before (Pre) and after (day 5) the controlled diet and training period and standardized differences (B). Maximal voluntary contraction torque of the knee extensors (C) before (Pre) and after (day 5) the controlled diet and training period and standardized differences (D). Mean ± SD (some error bars omitted for clarity).
TABLE 1: Participant characteristics.
TABLE 2: Inferential performance outcomes.
Performance outcomes for the MVC on days ‐3 and 5 are shown in Figure 3C. No effect of group (P = 0.63), time (P = 0.32), or interaction (P = 0.24) was demonstrated. Effect sizes are shown in Table 2. Mean pre- to posttest changes in MVC were −0.87 ± 0.77 N·kg−1 (~5.6% lower), −0.58 ± 1.60 N·kg−1 (~3.7% lower) and +0.40 ± 2.11 N·kg−1 (~2.6% higher) for LOW, MOD, and HIGH, respectively. Standardized differences are shown in Figure 3D. The standard differences for HIGH over MOD and HIGH over LOW were 0.98 ± 2.06 N·kg−1 (P = 0.51) and 1.27 ± 2.14 N·kg−1 (P = 0.29), respectively (Fig. 3D). The standard difference for MOD over LOW was 0.29 ± 2.01 N·kg−1 (P = 1.00).
A significant effect of time was shown for IMP (P = 0.01). No effect of group (P = 0.85) or interaction (P = 0.56) was demonstrated. Effect sizes are shown in Table 2. The mean pre- to posttest changes in IMP are −0.08 ± 0.15 N·s−1 (~2.4% lower), −0.28 ± 0.16 N·s−1(~8.3% lower), and −0.18 ± 0.23 N·s−1 (~5.3% lower) for LOW, MOD, and HIGH, respectively. The standard differences are −0.20 ± 0.23 kg·m−1·s−1 (P = 0.96), 0.10 ± 0.31 kg·m−1·s−1 (P = 1.00), and −0.10 ± 0.32 kg·m−1·s−1 (P = 0.89) for MOD over LOW, HIGH over MOD, and HIGH over LOW, respectively.
No effect of group (P = 0.39), time (P = 0.07), or interaction (P = 0.06) was demonstrated for PF. Effect sizes are shown in Table 2. The mean pre- to posttest changes in PF are −0.21 ± 0.42 N·kg−1 (~0.9% lower), −0.45 ± 0.64 N·kg−1 >(~2.0% lower), and −0.33 ± 0.33 N·kg−1 (~1.5% lower) for LOW, MOD, and HIGH, respectively. The standard differences are −0.25 ± 0.39 N·kg−1 (P = 1.00), 0.12 ± 0.70 N·kg−1 (P = 1.00), and −0.13 ± 0.51 N·kg−1 (P = 1.00) for MOD over LOW, HIGH over MOD, and HIGH over LOW, respectively.
Muscle soreness
There was a significant effect of time in perceived DOMS of the thigh (P = 0.02) and hamstrings (P = 0.04). There was no effect of group for the thigh (P = 0.77), hamstring (P = 0.26), or calf (P = 0.27). There was no effect of time for the calf (P = 0.12). There was no interaction for the thigh (P = 0.33), hamstring (P = 0.08), or calf (P = 0.40). There was a general trend for reductions in perceived DOMS for MOD and HIGH for the hamstrings and thigh relative to LOW.
DISCUSSION
Protein requirements have traditionally been assessed by determining the NBAL of a population in response to graded intakes, although the impact of these intakes on protein metabolism is arguably more relevant for athletes aiming to optimize their recovery (7). It is well established that dietary protein is important to support whole body and muscle anabolism after exercise (1), which generally occurs through an increase in rates of protein synthesis (3). We have previously demonstrated using the IAAO that the range of protein intakes in the present study stimulates whole body protein synthesis in a dose-dependent manner using a primed, constant oral [13C]-labeled tracer ingestion (12), which generally has a greater time resolution and precision compared with the single oral [15N]glycine bolus in the present study (20). However, the present tracer represents a practical, noninvasive method that allows for the characterization of free-living rates of whole body protein synthesis and breakdown in response to a controlled diet and training period. The lack of any statistical difference in protein synthesis or protein breakdown in the present study is not without precedent using our oral tracer model and end-product nitrogen kinetics (4). Our data qualitatively suggest that protein turnover may have been greatest in HIGH and constrained slightly in LOW. Nevertheless, subtle differences in synthesis and/or breakdown at the whole body level may still translate into meaningful differences in net balance, which may be further magnified at the muscle level (3). Although the present study is unable to determine the origin of the dose-dependent increase in whole body net balance, it may be related to the maintenance (MOD) or slight accrual (HIGH) of muscle (3) and/or plasma proteins (e.g., albumin) (27) that may translate into a greater performance directly via muscle repair or indirectly via plasma volume expansion, respectively.
It is generally accepted that endurance exercise increases daily protein requirements in trained populations (1). Notwithstanding the potential for NBAL to underestimate true requirements (6), the overall net balance with the LOW protein intake was the least positive despite our athletes consuming slightly higher (i.e., additional 0.14 g·kg−1) than the NBAL-derived RDA (28) or equivalent to the IAAO-derived safe intake suggested for the general population (6). This further highlights the inadequacy of this intake for athletic populations. Meredith et al. (4) demonstrated that intakes below the current RDA are associated with reductions in protein flux concomitant with a negative protein balance, indicating that suboptimal protein intakes in trained populations may be associated with maladaptive accommodations in protein metabolism. Although there were no statistical differences in S or B between groups or across time in the present study, LOW experienced a nonsignificant decrease of ~27% in synthesis and breakdown after just 4 d of consuming a suboptimal 0.94 g·kg−1·d−1 protein intake. These data could suggest that some level of accommodation may have been initiated in LOW as a means to maintain net protein balance (29), which overtime could constrain the athlete’s ability to effectively recover from and/or adapt to the stress of training, as will be discussed below. Based solely on the neutral protein balance and apparent maintenance of protein turnover, consuming 1.2 g·kg−1·d−1, which would represent the lower boundary of endurance athlete-specific recommendations (3) and general sports consensus positions (1), could be interpreted as representing a “sufficient” protein intake for our trained population.
Athletes and sports practitioners are arguably most interested in dietary intakes that maximize recovery and/or support high-quality training or performance outcomes. To this end, we performed a 5kmTT test to determine the impact of protein on the potential for a high-quality training bout given that it has a high degree of reliability (23,25) and is less variable for endurance performance testing than the often-used counterpart, time to exhaustion (30). Our observation that 5kmTT performance was trending toward being better maintained with HIGH could suggest that higher protein intakes that enhance whole body net protein balance may be associated with greater maintenance of overall training and/or performance capacity. The lack of statistical differences but small to large ES for HIGH to maintain exercise performance compared with MOD and LOW, respectively, is in general agreement with some (8), but not all studies (31,32). The greater cycling performance observed by Witard and colleagues (8) after 10 d of intensified training occurred in athletes consuming a high (~3 g·kg−1·d−1) but not moderate (~1.5 g·kg−1·d−1) protein intake. By contrast, similar cycling studies using 6–10 d of intensified training periods did not report any performance benefit for supplemental protein when athletes were consuming at least ~1.7 g·kg−1·d−1 (31,32). Thus, collectively the data could suggest that performance benefits from dietary protein during multiple days of endurance training could be saturable with the minimum intake to reach this threshold being closer to HIGH. We speculate that the performance benefit of higher protein intakes may be related to the ability to better replace any exercise-induced oxidative amino acid losses to support greater rates of protein remodeling and/or net deposition during recovery.
Weight-based running training may be influenced by neuromuscular function and subjective factors such as pain perception (33). In general, DOMS tended to be lowest in MOD and HIGH as compared with LOW in the present study, which would be consistent with the reported ability of dietary protein to attenuate muscle soreness (14). However, there was no difference in 5kmTT performance nor heart rate between MOD and LOW, suggesting the trend toward a greater 5kmTT with HIGH may not have been influenced (solely) by psychological effort or perceptual differences (34). Moreover, protein intake had no effect on PF or IMP jump performance metrics. Given that explosive power may influence performance in a weight-based performance test (22), the lack of difference between groups in the present study suggests that neuromuscular function did not influence our 5kmTT results.
Higher protein intakes during endurance exercise have been reported to maintain or enhance muscle strength (9,14). In addition, protein intakes that enhance NBAL have previously been associated with attenuated exercise-induced reductions in leg strength (13). Although there was a general trend toward a greater ES for MVC with higher protein intakes (i.e., HIGH>MOD>LOW), the lack of statistically significant differences between protein intakes may be related to the limited statistical power for this secondary outcome in the present study. Alternatively, a systematic review by Pasiakos et al. (14) suggested that the ergogenic effects associated with protein supplementation, such as the reduction in indirect markers of muscle damage (i.e., MVC), are most apparent over extended periods of training (e.g., days to weeks). Therefore, given the marked differences in net protein balance in the present study, a longer training period could have resulted in statistical differences between conditions for MVC. However, it should be noted that it is unclear if the maintenance of strength and/or reduction/resolution of muscle damage found in past studies with protein supplementation (13,14) is related to increases in muscle protein synthesis following the training that function to repair damaged proteins, especially of the functional myofibrillar protein (35), especially considering this process can be saturated with moderate acute protein intakes during postexercise recovery (36,37). Thus, additional studies are warranted to confirm whether higher daily protein intakes can attenuate muscle damage and/or enhance its recovery and, given its influence on time trial performance (38), whether this translates into increased running performance, as potentially suggested by the large ES in the 5kmTT of HIGH group.
The present study highlights the importance of protein to support protein metabolism and aid in the recovery from the stress of exercise to allow athletes to sustain a high training quality. Athletes may meet NBAL-derived protein recommendations (i.e., 1.2–1.4 g·kg−1·d−1) to sustain net protein balance provided their energy needs are met; however, a substantial portion of these athletes may not be meeting the intake of 1.83 g·kg−1·d−1 (15,16) that would optimize whole body protein synthesis (12) or potentially exercise performance (8). Moreover, it must be considered that almost half of athletes’ protein intake has been reported to be derived from plant-based sources (15), which may be limiting in some essential amino acids (e.g., branched chain amino acids) that are important for the repair and remodeling of body and muscle proteins (39). The supplemental dose used in the present study was provided in the form of crystalline amino acids modeled after the amino acid composition of egg protein, which is relatively enriched in essential amino acids and branched chain amino acids that are ultimately rate limiting in the stimulation of protein synthesis following endurance exercise (40). Thus, our high-quality egg protein intake may have been sufficient at 1.83 g·kg−1·d−1 to enhance metabolism and potentially exercise performance during training whereas athletes relying on a more plant-based protein diet may require a slightly greater intake.
In conclusion, our data demonstrate that a higher protein intake of 1.83 g·kg−1·d−1 enhanced whole body net protein balance and potentially provided small but meaningful benefits in exercise performance compared with protein intakes approximating previous nitrogen balance–determined estimates of the EAR (i.e., 0.94 g·kg−1·d−1) and RDA (i.e., 1.2 g·kg−1·d−1) for endurance athletes. These differences in metabolism and performance were observed with only a brief 4-d exposure to variations in dietary protein and despite consuming adequate carbohydrate. Thus, endurance exercisers who consume dietary protein toward the upper end (i.e., in our hands ≥1.8 g·kg−1·d−1) of the current recommendations of the American College of Sports Medicine (i.e., 1.2–2 g·kg−1·d−1) would help meet the metabolic demand for this macronutrient to support their training and/or exercise performance goals.
The authors thank the study participants for generously offering their time and effort. They also acknowledge their laboratory assistants, Mark Orlando and Sarkis Hannaian, for their help in the data collection of this study.
E. W., H. K., K. V., and D. M. conceived and designed the experiments; E. W., H. K., and K. V. performed the experiments; E. W. and H. K. analyzed the data; E. W. and H. K. wrote the paper. All authors have read and approved the final manuscript.
This work was supported by a research grant from Ajinomoto co., Inc. The authors declare no conflict of interest. Results of the present study do not constitute endorsement by the American College of Sports Medicine and are presented honestly without fabrication, falsification, or inappropriate manipulation of the data.
REFERENCES
1. Thomas DT, Erdman KA, Burke LM. Position of the Academy of Nutrition and Dietetics, Dietitians of Canada, and the American College of Sports Medicine: nutrition and athletic performance.
J Acad Nutr Diet. 2016;116(3):501–28.
2. Mazzulla M, Parel JT, Beals JW, et al. Endurance exercise attenuates postprandial whole-body leucine balance in trained men.
Med Sci Sports Exerc. 2017;49(12):2585–92.
3. Levenhagen DK, Carr C, Carlson MG, Maron DJ, Borel MJ, Flakoll PJ. Postexercise protein intake enhances whole-body and leg protein accretion in humans.
Med Sci Sports Exerc. 2002;34(5):828–37.
4. Meredith CN, Zackin MJ, Frontera WR, Evans WJ. Dietary protein requirements and body protein metabolism in endurance-trained men.
J Appl Physiol. 1989;66(6):2850–6.
5. Gaine PC, Pikosky MA, Martin WF, Bolster DR, Maresh CM, Rodriguez NR. Level of dietary protein impacts whole body protein turnover in trained males at rest.
Metabolism. 2006;55(4):501–7.
6. Humayun MA, Elango R, Ball RO, Pencharz PB. Reevaluation of the protein requirement in young men with the indicator amino acid oxidation technique.
Am J Clin Nutr. 2007;86(4):995–1002.
7. Phillips SM, van Loon LJ. Dietary protein for athletes: from requirements to optimum adaptation.
J Sports Sci. 2011;29(1 Suppl):S29–38.
8. Witard OC, Jackman SR, Kies AK, Jeukendrup AE, Tipton KD. Effect of increased dietary protein on tolerance to intensified training.
Med Sci Sports Exerc. 2011;43(4):598–607.
9. Rowlands DS, Rössler K, Thorp RM, et al. Effect of dietary protein content during recovery from high-intensity cycling on subsequent performance and markers of stress, inflammation, and muscle damage in well-trained men.
Appl Physiol Nutr Metab. 2008;33(1):39–51.
10. Rowlands DS, Thorp RM, Rossler K, Graham DF, Rockell MJ. Effect of protein-rich feeding on recovery after intense exercise.
Int J Sport Nutr Exerc Metab. 2007;17(6):521–43.
11. Rowlands DS, Wadsworth DP. Effect of high-protein feeding on performance and nitrogen balance in female cyclists.
Med Sci Sports Exerc. 2011;43(1):44–53.
12. Kato H, Suzuki K, Bannai M, Moore DR. Protein requirements are elevated in endurance athletes after exercise as determined by the indicator amino acid oxidation method.
PLoS One. 2016;11(6):e0157406.
13. Nelson AR, Phillips SM, Stellingwerff T, et al. A protein-leucine supplement increases branched-chain amino acid and nitrogen turnover but not performance.
Med Sci Sports Exerc. 2012;44(1):57–68.
14. Pasiakos SM, Lieberman HR, McLellan TM. Effects of protein supplements on muscle damage, soreness and recovery of muscle function and physical performance: a systematic review.
Sports Med. 2014;44(5):655–70.
15. Gillen JB, Trommelen J, Wardenaar FC, et al. Dietary protein intake and distribution patterns of well-trained dutch athletes.
Int J Sport Nutr Exerc Metab. 2017;27(2):105–14.
16. Burke LM, Slater G, Broad EM, Haukka J, Modulon S, Hopkins WG. Eating patterns and meal frequency of elite Australian athletes.
Int J Sport Nutr Exerc Metab. 2003;13(4):521–38.
17. Braakhuis AJ, Meredith K, Cox GR, Hopkins WG, Burke LM. Variability in estimation of self-reported dietary intake data from elite athletes resulting from coding by different sports dietitians.
Int J Sport Nutr Exerc Metab. 2003;13(2):152–65.
18. Shvartz E, Reibold RC. Aerobic fitness norms for males and females aged 6 to 75 years: a review.
Aviat Space Environ Med. 1990;61(1):3–11.
19. Weir JB. New methods for calculating metabolic rate with special reference to protein metabolism.
J Physiol. 1949;109(1–2):1–9.
20. Grove G, Jackson AA. Measurement of protein turnover in normal man using the end-product method with oral [15N]glycine: comparison of single-dose and intermittent-dose regimens.
Br J Nutr. 1995;74(4):491–507.
21. Tarnopolsky MA, MacDougall JD, Atkinson SA. Influence of protein intake and training status on nitrogen balance and lean body mass.
J Appl Physiol (1985). 1988;64(1):187–93.
22. Gathercole R, Sporer B, Stellingwerff T. Countermovement jump performance with increased training loads in elite female rugby athletes.
Int J Sports Med. 2015;36(9):722–8.
23. Russell RD, Redmann SM, Ravussin E, Hunter GR, Larson-Meyer DE. Reproducibility of endurance performance on a treadmill using a preloaded time trial.
Med Sci Sports Exerc. 2004;36(4):717–24.
24. Zainuddin Z, Newton M, Sacco P, Nosaka K. Effects of massage on delayed-onset muscle soreness, swelling, and recovery of muscle function.
J Athl Train. 2005;40(3):174–80.
25. Stevens CJ, Hacene J, Sculley DV, Taylor L, Callister R, Dascombe B. The reliability of running performance in a 5 km time trial on a non-motorized treadmill.
Int J Sports Med. 2015;36(9):705–9.
26. Sawilowsky SS. New effect size rules of thumb.
JAMSM. 2009;8(2):597–9.
27. Moore DR, Robinson MJ, Fry JL, et al. Ingested protein dose response of muscle and albumin protein synthesis after resistance exercise in young men.
Am J Clin Nutr. 2009;89(1):161–8.
28. Health Canada. Dietary reference intakes: the essential guide to nutrient requirements. National Academy of Sciences [Internet]; 2006. [cited 2018 Jan 26]. Available from:
https://fnic.nal.usda.gov/sites/fnic.nal.usda.gov/files/uploads/DRIEssentialGuideNutReq.pdf.
29. Young VR, Gucalp C, Rand WM, Matthews DE, Bier DM. Leucine kinetics during three weeks at submaintenance-to-maintenance intakes of leucine in men: adaptation and accommodation.
Hum Nutr Clin Nutr. 1987;41(1):1–18.
30. Laursen PB, Francis GT, Abbiss CR, Newton MJ, Nosaka K. Reliability of time-to-exhaustion versus time-trial running tests in runners.
Med Sci Sports Exerc. 2007;39(8):1374–9.
31. D’Lugos AC, Luden ND, Faller JM, Akers JD, McKenzie AI, Saunders MJ. Supplemental protein during heavy cycling training and recovery impacts skeletal muscle and heart rate responses but not performance.
Nutrients. 2016;8(9).
32. Hansen M, Bangsbo J, Jensen J, et al. Protein intake during training sessions has no effect on performance and recovery during a strenuous training camp for elite cyclists.
J Int Soc Sports Nutr. 2016;13(1):9.
33. Braun WA, Dutto DJ. The effects of a single bout of downhill running and ensuing delayed onset of muscle soreness on running economy performed 48 h later.
Eur J Appl Physiol. 2003;90(1–2):29–34.
34. Burnett D, Smith K, Smeltzer C, Young K, Burns S. Perceived muscle soreness in recreational female runners.
Int J Exerc Sci. 2010;3(3):108–16.
35. Wilkinson SB, Phillips SM, Atherton PJ, et al. Differential effects of resistance and endurance exercise in the fed state on signalling molecule phosphorylation and protein synthesis in human muscle.
J Physiol. 2008;586(15):3701–17.
36. Rowlands DS, Nelson AR, Phillips SM, et al. Protein-leucine fed dose effects on muscle protein synthesis after endurance exercise.
Med Sci Sports Exerc. 2014;47(3):547–55.
37. Witard OC, Jackman SR, Breen L, Smith K, Selby A, Tipton KD. Myofibrillar muscle protein synthesis rates subsequent to a meal in response to increasing doses of whey protein at rest and after resistance exercise.
Am J Clin Nutr. 2014;99(1):86–95.
38. Paavolainen LM, Nummela AT, Rusko HK. Neuromuscular characteristics and muscle power as determinants of 5-km running performance.
Med Sci Sports Exerc. 1999;31(1):124–30.
39. van Vliet S, Burd NA, van Loon LJ. The skeletal muscle anabolic response to plant- versus animal-based protein consumption.
J Nutr. 2015;145(9):1981–91.
40. Kato H, Suzuki K, Bannai M, Moore DR. Branched-chain amino acids are the primary limiting amino acids in the diets of endurance-trained men after a bout of prolonged exercise.
J Nutr. 2018;148(6):925–31.
