Skip Navigation LinksHome > January 2011 - Volume 43 - Issue 1 > Effect of High-Protein Feeding on Performance and Nitrogen B...
Medicine & Science in Sports & Exercise:
doi: 10.1249/MSS.0b013e3181e93316
Basic Sciences

Effect of High-Protein Feeding on Performance and Nitrogen Balance in Female Cyclists

ROWLANDS, DAVID S.; WADSWORTH, DANIEL P.

Free Access
Article Outline
Collapse Box

Author Information

Division of Exercise and Sport Science, Institute of Food, Nutrition, and Human Health, Massey University, Wellington, NEW ZEALAND

Address for correspondence: David S. Rowlands, Ph.D., Institute of Food, Nutrition, and Human Health, Massey University, PO Box 756, Wellington, New Zealand; E-mail: d.s.rowlands@massey.ac.nz.

Submitted for publication December 2009.

Accepted for publication May 2010.

Collapse Box

Abstract

Purpose: The effect of dietary protein ingested after exercise on recovery in women athletes is unknown. Therefore, we asked whether there is a meaningful difference between high- and low-protein recovery diets on the subsequent performance of well-trained female cyclists.

Methods: In a crossover, 12 female cyclists completed three high-intensity rides composed of 2.5-h intervals on day 1, followed by repeat-sprint performance tests on days 2 and 4, interspersed with a rest day. During the 4-h recovery on days 1 and 2, cyclists ingested 1.4/0.7/0.26 or 2.1/0.1/0.26 g·kg−1·h−1 of CHO/protein/fat in high-protein or isocaloric control conditions, respectively. At other times, cyclists ingested an isoenergetic high-CHO diet.

Results: No effect of protein dose on the mean power during repeat sprint was evident on day 2 (high-protein vs control = −1.1%; 95% confidence limits = ±4.6%) or on day 4 (1.7%; ±4.6%); furthermore, fatigue effects (slope) were unclear (day 2 = 1.4%; ±4.9%, day 4 = 0.5%; ±4.9%). Perceptions of leg tiredness and soreness were increased, and leg strength was reduced in the high-protein condition relative to control. In the high-protein condition, plasma glucose concentrations were lower during recovery, and plasma lactate concentrations were lower during the sprints. Effects on circulating creatine kinase activity were trivial. Net nitrogen balance during the experiment was positive in the high-protein condition (mean ± SD = 177 ± 140 mg of N·kg−1 fat-free mass) but negative in the control condition (−81 ± 73 mg of N·kg−1 fat-free mass); the estimated protein requirement was 1.28 g·kg−1·d−1 (±0.57 g·kg−1·d−1).

Conclusions: In contrast with the previous findings in males, we observed no clear influence of dietary protein quantity on the subsequent performance in females. The findings on nitrogen balance suggest that female cyclists training intensely have daily protein requirements approximately 1.6 times the recommended daily allowance but 0.65 times that of males.

Research into the role of dietary protein ingested after endurance exercise on recovery processes and subsequent performance has been conducted almost exclusively with male subjects. However, noteworthy differences exist in the circulating hormonal milieu (31) and amino acid, glucose, and lipid metabolism (16,29) between males and females. Females oxidize less leucine (16,21) and CHO and display a lipid-dominated transcriptional (31) and metabolic response with as much as 70% more lipid oxidized during submaximal exercise relative to men (29,30). Furthermore, there is some evidence that when compared with males, female athletes exhibit an attenuated metabolic and performance response to nutritional interventions such as CHO loading (18,32) and fat-supplemented diets (36). In review of the evidence from dietary survey and protein turnover, Tarnopolsky (30,31) concluded that female athletes seem to require approximately 15%-25% less dietary protein compared with matched male counterparts. However, no study has yet to specifically calculate the protein requirements for well-trained or elite-level women athletes; consequently, direct inference to females of the nature and effect of dietary protein interventions on physiological and performance responses seen in males may not be appropriate.

In men, there is some evidence that protein coingested with CHO in the form of a sports beverage after exercise can enhance 2-15 h of subsequent performance over CHO only (1,20,27,38). Although some authors report no clear effect during this period (2,26), the driving paradigm is that improved performance is due to the increased glycogen resynthesis (34,40). To investigate if the level of protein in the recovery diet of male endurance athletes undertaking several bouts of exercise simulating competition or intense training mattered during a longer assessment duration, we compared the effect of isocaloric high- versus low-protein recovery diets (1.4/0.7/0.26 vs 2.1/0.1/0.26 g·kg−1·h−1 of CHO/protein/lipid) ingested for 4 h after high-intensity cycling (energy expenditure = 10.5 MJ) on the 15- and 60-h subsequent performance in male cyclists (24). The high-protein diet substantially enhanced the 60-h subsequent performance. CHO ingestion during the 4-h recovery phase and at other times (8-10 g·kg−1·h−1) was provided to maximize muscle glycogen synthesis to diminish the influence of muscle glycogen content as a covariate (5,15). The performance outcome suggested to us that a protein dose-sensitive adaptive mechanism might account for the accrued benefit of high-dose protein feeding after an intensive endurance exercise in trained men. In support of this, Howarth et al. (14) showed that consuming a protein-rich CHO diet (1.2/0.4 g·kg−1·h−1 CHO/protein) during the 4-h recovery from 2 h of high-intensity interval cycling (∼7.8 MJ) led to a 50% elevation in mixed-muscle protein fractional synthetic rate in the exercised skeletal muscle relative to the appropriate control high-CHO diets. Further, there is evidence in men that protein ingestion after a strenuous endurance exercise is associated with attenuated muscle damage by way of small- to large-sized reductions in circulating creatine kinase concentration (7,24,26,27), up-regulation of components of the endurance exercise-specific transcriptome (25), and enhanced amino acid-stimulated signaling processes associated with protein synthesis (19).

Therefore, to investigate the effect of protein quantity ingested during recovery from high-intensity exercise on recovery and subsequent performance and to estimate dietary protein requirement during intensive training in well-trained women, we recruited a matched cohort of female cyclists (age, recent training volume, peak power per kilogram of fat-free mass (FFM)) and reproduced the experimental protocol used previously in males (24) with the exception of a 4-wk washout to control for effects of the menstrual cycle on protein and other substrate metabolism (29). We report noteworthy differences in the nature and magnitude of the physiological and performance response to protein dose between genders and provide novel empirical data on the protein requirement of well-trained female cyclists during a period of intensive training.

Back to Top | Article Outline

METHODS

Participants

Twelve endurance-trained female cyclists aged (mean ± SD) 30 ± 7 yr, standing 166 ± 6 cm, and having a body mass of 60.8 ± 3.4 kg completed the study. Estimated FFM was 49.5 ± 3.1 kg, and body fat mass was 19% ± 3% of total. Maximal oxygen uptake (V˙O2max) was 3.4 ± 0.4 L·min−1, and the corresponding peak power output (Wmax) was 260 ± 26 W. The cyclists had been training for 4.9 ± 4.1 yr, with a reported weekly average training of 12.8 ± 3.8 h during the 6 months before the study, which included high-intensity training and competition. Of the cyclists, six reported taking a form of contraceptive drug (oral or injection), whereas the remainder took none. Before beginning experimentation, all subjects read the study information sheet, were informed of their rights, screened for precluding health conditions, and signed a consent form. The study was approved by the Massey University's Human Ethics Committee.

Back to Top | Article Outline
General Design

Cyclists participated in a double-blind, randomized, crossover design, with a protein-rich recovery feeding intervention (high-protein) and an isocaloric low-protein control condition (control). The design consisted of preliminary fitness testing and a performance test familiarization trial and two 4-d experimental blocks (Fig. 1; see Experimental Protocol and Procedures for details). To control for any possible effect of menstrual cycle on outcomes, the two experimental blocks were spaced 28 d apart, with the first block beginning 3-7 d after the first day of menstruation, with the objective of all testing occurring during the midfollicular phase. No confirmatory hormonal measures were taken because the requirement to the control training and diet and to schedule commitments around participation prohibited any change in testing days on the outcome of a hormone test. The menstrual cycle phase does have a small influence on substrate selection during endurance exercise, but the differences are small when compared with the differences between the sexes (31). All cyclists reported regular menstruation with the exception of one reporting an irregular cycle during the study (4-wk washout after first menstruation was retained). Before the first testing block, cyclists recorded their training and diet in diaries provided for 10 and 2 d, respectively. Training was kept light for 3 d before the testing block. This training and dietary regimen was reproduced preceding the second block to standardize preconditioning.

FIGURE 1-Experimenta...
FIGURE 1-Experimenta...
Image Tools
Back to Top | Article Outline
Preliminary Testing

About 1-2 wk before the start of the first experimental block, cyclists reported to the laboratory for measurement of body composition, a test to determine V˙O2max and Wmax, and a familiarization trial. Fat and FFM were determined from skinfolds and regression equations (35). V˙O2max was determined on an electromagnetically braked cycle ergometer (Velotron, Seattle, WA) using a continuous graded exercise protocol as described previously (35). After a 5-min rest, the cyclists then performed a familiarization trial of the performance test protocol and of the perceptual sampling procedures. All testing occurred in a laboratory with air conditioning set at 19°C and maintaining 45%-50% relative humidity.

Back to Top | Article Outline
Experimental Protocol and Procedures

On day 1, cyclists consumed a standardized low-protein animal flesh-free diet, with the final meal taken 4 h before coming into the laboratory at ∼15:00 h. In the laboratory, cyclists completed a fatigue-inducing ride composed of 2.5-h intervals with an average estimated total work performed of 7.5 MJ. For each cyclist, the protocol consisted of a warm-up composed of 12 min at 30%, 5 min at 40%, and 5 min at 50% Wmax, 10 × 2-min intervals at 90% Wmax and 12 × 2-min intervals at 80% Wmax, alternated with 2-min recovery periods at 50% Wmax, and finished with three 5-min intervals at 70% interspersed with three 5-min recovery periods at 50%. Two or three cyclists completed the testing procedure at once, and the same group completed the second experimental block to repeat any effects of personality-induced psychostimuli.

Shortly after the exercise, cyclists voided, and then a 20-gauge cannula (Becton Dickinson, Medical Pte. Ltd., Singapore) was placed into an antecubital vein in a suitable position. A two-way stopcock valve (Connecta Plus 3; Becton Dickinson) was secured at the end of the cannula. A 10-mL blood sample was drawn at the time of placement (∼5 min after exercise). Additional samples were collected every 30 min for the following 120-min recovery period, with one final sample collected at the 180th minute. The cannula was kept patent with isotonic saline (Pharmacia & Upjohn, Peapack, NJ). Collected blood was immediately transferred into evacuated containers (Becton Dickinson & Co., Franklin Lakes, NJ) with EDTA or lithium heparin. These tubes were then immediately centrifuged at 2500g for 12 min (Centrifuge; Heraeus Sepatech Medifuge, Bruchsal, Germany). The plasma was aspirated into Eppendorf tubes and was then immediately frozen in liquid nitrogen and stored at −80°C for later analysis.

Immediately after the first draw of blood, cyclists consumed their first unit of the recovery food (see Recovery Feeding Intervention). Subsequent units were consumed at 30-min intervals immediately after blood collection. Cyclists remained in the laboratory for the first 3 h of the 4-h recovery feeding period during which time they showered and rested. After 3 h, cyclists were asked to consume the last unit within 1 h after leaving the laboratory, prepare for the following morning, and go to bed. Cyclists then fasted until the next morning.

On day 2 of the experimental protocol, cyclists reported to the laboratory between 06:30 and 07:30 h. Start time was standardized for each rider. On arrival, a cannula was placed as described, and a 10-mL blood sample was drawn. After blood collection, the cyclists consumed a small breakfast described previously, consisting of a cereal bar and 250 mL of water (24). Fifteen minutes afterward, a blood sample was taken, and cyclists started a warm-up composed of 12 min at 30%, 5 min at 40%, and 5 min at 50% Wmax, 6 × 2-min priming intervals at 80% Wmax interspaced with 2-min periods at 50% Wmax, and followed by the repeated sprint performance test as described in detail elsewhere (24,35,36). The average total internal work performed during the sprint test was 3.6 MJ (5.5 MJ including warm-up). The priming intervals were inserted into the protocol to (a) simulate high-intensity phases of a warm-up before a maximal-effort training session or competition, (b) provide some standardized high-intensity exercise blocks for psychometric measures under constant load as opposed to subject-selected pace as in the sprints, and (c) increase the duration of the high-intensity exercise session to ∼2 h. During all exercise procedures, the cyclists were cooled with fans on a standardized setting to minimize thermal distress.

After the performance test, the cyclists repeated the ingestion of the recovery food for 4 h in the same experimental condition as the previous day. This time, however, cyclists remained in the laboratory, and blood was collected every 30 min for only the first 90 min into recovery to accommodate work or study commitments. Cyclists ingested the remaining units at 30-min intervals at work or at home and then moved onto a standardized CHO-rich diet until the late evening of the following day (day 3). The diet was provided to the cyclists and consisted of CHO-rich foods (oats, rice, pasta, bread, vegetables, fruit, milk, yogurt, cheese, sports drink, fruit juice, or cereal bars). Participants were instructed to ingest a minimum of the food items to provide at least 10 and 8 g of CHO·kg−1 FFM on days 2 and 3, respectively; to accommodate unknown variation in hunger and metabolic rate, they could consume above the minimum amounts if still hungry but had to record all portions and replicate and record during the second block of the crossover, which was achieved through provision of any additional food item by the researchers and sign-off by the participants. The standardized high-CHO diet was designed to saturate intramuscular glycogen concentrations before exercise on day 4 and was consistent with the dietary macronutrient recommendations for endurance athletes for CHO when training hard (5,15). The diet was identical between conditions and, combined with the energy contribution from the exercise supplement and recovery food, provided total energy to approximate normal energy intake for female cyclists during recovery (11.1 MJ·d−1) (17).

On the morning of day 4, cyclists reported to the laboratory and were fitted with a cannula, after which time they repeated the performance test in exactly the same manner as on day 2. Urine was collected throughout each experimental block from the completion of exercise on day 1 until the beginning of exercise on day 4.

Back to Top | Article Outline
Recovery Feeding Intervention

The high-protein condition was a chocolate-coated bar and milk-like drink formulation (chocolate- or vanilla-flavored). The low-protein control condition was an isocaloric bar formulated by the same food technologist who made the high-protein bar and a milk-like drink formulated to resemble the intervention bar and drink as close as possible (similar taste, color, and texture; indistinguishable serving volume and size). The high-protein condition was designed to provide per kilogram of FFM per hour 1.6 g of CHO, 0.8 g of protein, and 0.29 g of fat, whereas the control provided 2.35 g of CHO, 0.12 g of protein, and 0.29 g of fat. One hundred grams of the bar used in the protein-enriched condition contained 27.4 g of CHO, 20.1 g of protein, and 4 g of fat. Ingredients were a protein blend (whey protein isolate, calcium caseinate, and soy protein isolate), glucose, honey, milk chocolate (14%; sugar, cocoa butter, cocoa liquor, whole milk powder, skim milk powder, and emulsifier), apple juice concentrate, crisped soy nuggets (isolated soy protein, rice flour, malt, and salt), glycerin, cocoa powder, emulsifier, canola oil, flavor, corn syrup solids, vitamins, and minerals. The control bar contained 5.1 g of protein, 41 g of CHO, and 7.2 g of fat per 100 g. Ingredients for the control bar were the same as above for the protein-enriched bar, except that the protein blend and the soy nuggets were replaced with waxy maize starch and rolled oats, respectively. The protein-enriched beverage contained 4.5% sucrose, 6.3% maltodextrin, 1.85% whey protein isolate, 3.7% milk protein concentrate, 3.2% cream powder, 0.1% salt, flavoring, and water. The control beverage was made of 4.5% sucrose, 12.2% maltodextrin, 2.45% cream powder, 0.12% salt, flavoring, and water. Bar and drink servings were given every 30 min, starting immediately after exercise on day 1 for 4 h totaling eight bars and drinks. The protein in the control bar and drink was contained within the chocolate coating, rolled oats, and in trace amounts in the cream powder used in the drinks to assist in blinding. In addition to the role in blinding, the fat content in the bar and drinks was included to help establish energy balance and to provide a quantity of fat normally present in the diet of athletes.

Back to Top | Article Outline
Fluid and CHO Replacement during Exercise

During all exercise procedures, the cyclists were provided with a 6.8% CHO electrolyte solution to provide CHO at the rate of 0.8 g·kg−1 FFM·h−1. On days 1 and 2, the cyclists were weighed immediately before and after exercise after they had toileted. If weight loss occurred, a mass of tap water equivalent to the deficit was provided during the 4-h recovery period to assist in rehydration.

Back to Top | Article Outline
Ratings of Fatigue and Exertion

On days 2 and 4, during exercise, cyclists were asked to rate the sensation perception on visual analog scales immediately after priming intervals 2 and 5 and sprints 1, 4, 7, and 10. The measured categories were sensations of tiredness, leg soreness, ability to sprint, level of exertion, and nausea.

Back to Top | Article Outline
Biochemistry

Plasma lactate and glucose concentrations were measured using an automated analyzer (Bayer Rapidlaboratory 800 System; Corning & Diagnostics Corp., Medfield, MA). Creatine kinase was measured by an enzymatic assay on a Cobas Fara II analyzer (Roche Diagnostics NZ Ltd., Auckland, New Zealand) in accordance with the manufacturer's recommendations. Urinary urea was measured by spectrophotometry (Jenway 6505; Jenway, Stone, UK) using enzymatic urea slow rate reagent kit (Thermo Electron; Medica Pacific, Auckland, New Zealand). Measurement of urinary creatine and creatinine was by high-performance liquid chromatography (1100 Series; Agilent-Hewlett Packard, Waldbronn, Germany).

Back to Top | Article Outline
Nitrogen Balance

Net nitrogen balance was calculated during three periods during the experimental block to total 60 h: 1) from the completion of cycling on day 1 to the final urine collection shortly before the preexercise meal on day 2, 2) during exercise on day 2, and 3) from the end of exercise on day 2 to the final urine collection on the morning of day 4 preceding the preexercise meal. Nitrogen intake was calculated from total dietary protein intake in grams divided by the amino acid-nitrogen constant 6.25. Nitrogen outputs were urinary urea, creatine, and creatinine (measured). Additional (estimated) nitrogen losses were from sweat at rest, sweat during exercise on day 2, and from feces and miscellaneous loss throughout the 60-h collection period based on the urine-to-sweat loss ratios reported previously (33).

Back to Top | Article Outline
Statistical Analysis
Sample size.

During study conception, the reliability of performance in an equivalently well-conditioned female cohort was unknown and was therefore assumed similar to that of the males. We predicted that the same 4.1% enhancement in performance would result in the female cohort as was observed in males (24). Therefore, we assumed equivalent power and decided to study an identical sample.

Back to Top | Article Outline
Data modeling.

The effect of protein dose on dependent outcomes was estimated from mixed modeling in the Statistical Analysis System (version 9.1; SAS Institute, Cary, NC). Most dependent variables were analyzed after log transformation to accommodate the heteroscedastic distribution of residuals and to express differences as percents; the exceptions were the perceptual and nitrogen balance data where log transformation was not appropriate. Outcomes were determined from the interaction between the following fixed effects: treatment, order of treatment (treatment 1 or 2), and day (day 2 or 4). For the analysis of sprint performance, lactate and glucose, and perceptual responses, sprint number or time was coded as a quantitative numeric predictor (as in linear regression) to model the effect of time on outcomes (slope and fatigue effects). For the analysis of creatine kinase and nitrogen balance, estimates were derived from repeated-measures models (as in traditional ANOVA).

Random effects were included in the models to account for individual responses to treatment and multiple-day outcomes. In the models for sprint performance, perceptual data, and glucose and lactate concentrations, the variation associated with moving between sprints (fatigue) or progression along a time series was also included. The within-subject SD was calculated as the root of the remaining residual variance.

Back to Top | Article Outline
Presentation of data.

Descriptive statistics are raw means and SD. Means derived from the analysis of log-transformed variables are back log-transformed (geometric) least squares means. Spread around these means is the coefficient of variation (CV). Data are rounded to two significant digits or, in some cases, three, where the first digit is "1."

Back to Top | Article Outline
Precision of estimation and statistical inference.

Adopting recommendations for progressive inferential statistics published in Medicine & Science in Sports & Exercise® (12) and elsewhere (9,28), statistical inference was by magnitude-based inference, that is, probabilistic determination that effects are greater than, less than, or equivalent to the defined practically or mechanistically quantitative thresholds. Estimate precision was by 95% confidence interval. After standardization, probabilistic inferences about the true value for outcomes except performance were qualified using the standardized difference (adopted from the Cohen effect size d: mean difference/appropriate SD). Standardized difference thresholds were as follows: trivial or negligible = 0.0-0.2, small = 0.2-0.6, moderate = 0.6-1.2, large = 1.2-2.0, very large = 2.0-4.0, enormous > 4.0). For performance, the threshold for a substantial effect was 1.1% (see, for derivation, Rowlands et al. (24)). Probability thresholds were obtained from the t distribution with likelihoods ordered into cutoffs and inferred as follows: almost certainly not <1%, very unlikely = 1%-5%, unlikely = 5%-25%, possible = 25%-75%, likely = 75%-95%, very likely = 95%-99%, and almost certain > 99% (12). In the case where most (>50%) of the uncertainty lies between the threshold for a substantial increase and decrease, the likelihood of the effect being trivial (negligible) is qualified. Effects were described as unclear or inconclusive if the likelihood overlapped into both positive and negative values by >5% each.

Back to Top | Article Outline

RESULTS

Performance

Performance outcome is shown in Figure 2. There was no clear effect of protein dose on the overall sprint mean power: day 2/4 mean power (CV), control = 237 W/245 W (21%) and high protein = 234 W/250 W (23%); mean difference (95% confidence limits (CL)) = −1.1%/1.7% (±4.6%). Furthermore, differences in the decline in mean power from sprint 1 to 10 (slope, fatigue effect) were also inconclusive on both days: day 2/4 slope (mean percentage decline in sprint mean power (CV)), control = 11.5%/9.5% (4.8%) and protein-enriched = 10.2%/9.1% (4.7%); mean difference = 1.4%/0.5% (95% CL = ±4.9%).

FIGURE 2-Mean sprint...
FIGURE 2-Mean sprint...
Image Tools

With the treatment effect factored out, overall sprint mean power on day 4 compared with day 2 was 4.2% higher (95% CL = ±3.2%) during the first and 4.9% higher (±3.3%) during the second experimental block, respectively. The second order of trial increased overall sprint mean power on days 2 and 4 relative to the first by an insubstantial 0.7% (±4.6%) and 1.4% (±4.6%), respectively.

Back to Top | Article Outline
Perceptual Ratings

A possible substantial increase in perceived effort of 0.4 U (±0.8 U) was observed in the high-protein condition during the preload priming intervals on day 2 (Fig. 3); all other effects at this time were unclear or trivial. During the sprints on day 2, cyclists demonstrated likely and very likely elevations in perception of tiredness and leg soreness, respectively, coupled with an almost-certain decrease in perceived leg strength in the high-protein condition relative to the control, but effects on effort and nausea were unclear (Table 1). During the preload priming intervals on day 4, feelings of tiredness (0.5 U; ±0.8 U) and soreness (0.5 U; ±0.7 U) were possibly elevated, and effort likely substantially elevated (0.6 U; ±0.8 U) in the high-protein condition; conversely, a likely substantial decrease in nausea (−0.3 U; ±0.3 U) was seen at this time (Fig. 3). During the sprints on day 4, likely and possible increases in leg soreness and tiredness were observed in the high-protein condition, respectively; effects on the other perceptions were unclear (Table 1 and Fig. 3).

Table 1
Table 1
Image Tools
FIGURE 3-Perceptual ...
FIGURE 3-Perceptual ...
Image Tools
Back to Top | Article Outline
Biochemical Parameters
Glucose and lactate.

Plasma glucose and lactate concentrations are shown in Figure 4. There were moderate (14%; ±5%) and large (22%; ±10%) increases in glucose concentration in the control condition during recovery on days 1 and 2, respectively. Effects on glucose concentration during the sprints were trivial, being 4% lower (±4%) and 1% higher (±6%) with the high-protein condition on days 2 and 4, respectively.

FIGURE 4-Plasma gluc...
FIGURE 4-Plasma gluc...
Image Tools

Plasma lactate concentrations were 24% lower (±12%) during recovery on day 2 in the high-protein condition; effects were unclear during recovery on day 1. During the sprints, there were moderate (28%; ±18%) and small (20%; ±22%) reductions in lactate concentration in the high-protein condition on days 2 and 4, respectively.

Back to Top | Article Outline
Creatine kinase.

Mean ± SD plasma creatine kinase activity in the reference (low-protein) condition before and after exercise on day 2 and before exercise on day 4 were 118 ± 85, 154 ± 128, and 85 ± 42 U·L−1, respectively. Effects of diet on plasma creatine kinase activity were unclear or trivial throughout. Specifically, there were trivial reductions in activity before exercise (−12%; 95% CL = ±17%) and on completion of the sprints on day 2 (−9%; ±18%) in the high-protein condition, relative to control; a trivial reduction of 5% (±19%) was also observed before exercise on day 4.

Back to Top | Article Outline
Nitrogen balance.

Nitrogen balance is shown in Figure 5A. From the completion of exercise on day 1 through the morning of day 2, nitrogen balance was positive in the high-protein condition but negative in the control condition; as a result, the net difference between conditions was 247 mg of N·kg−1 FFM (95% CL = ±47 mg of N·kg−1 FFM). In contrast, during exercise on day 2, cyclists were in neutral balance in the control but slightly negative in the high-protein condition (no clear difference), and from day 2 to the morning of day 4, cyclists were in slightly positive nitrogen balance in the high-protein condition and slightly negative in the control condition, with a net 34 mg of N·kg−1 FFM more positive balance (±48 mg of N·kg−1 FFM) in the high-protein condition. Overall, the net difference in balance was 260 mg of N·kg−1 FFM (±101 mg of N·kg−1 FFM). The nitrogen balance regression presented in Figure 5B produced an estimated average dietary protein requirement of 1.28 g·kg−1·d−1 (95% confidence interval = 0.54-2.67 g·kg−1·d−1) to achieve zero nitrogen balance.

FIGURE 5-A, Effect o...
FIGURE 5-A, Effect o...
Image Tools
Back to Top | Article Outline

DISCUSSION

The primary outcome of our study was that a high-protein-CHO diet ingested during recovery from high-intensity cycling offered no clear benefit to subsequent cycling performance relative to a low-protein control diet. Although also unclear at 15 h, the performance outcome at 60 h contrasts with our previously reported likely moderate enhancement with the high-protein recovery diet in equivalently well-trained male cyclists (24).

Our work is the first that we are aware of to investigate the effect of protein quantity ingested during recovery on the subsequent performance in well-trained female endurance athletes. We were prompted to study females from the findings in men (24) and other evidence that females might respond differently because of gender differences in lipid, CHO, and protein metabolism (16,21,29). The delayed (∼60 h) relative performance increase in males (5,15) suggests that the muscle of endurance-trained men may be more functionally responsive to the quantity of protein ingested in the recovery diet, with emerging evidence pointing to the effect of high-protein recovery feeding being driven by higher muscle protein synthesis and adaptive remodeling (14,25). Consequently, an interesting observation was that plasma creatine kinase activity was overall 25% lower in the low-protein condition (mean activity = 119 U·L−1 in women and 158 U·L−1 in men), suggesting less damage or faster repair to the sarcolemma in women in response to the exercise procedures. Recently, Enns and Tiidus (10) summarized a muscle-tissue protective role for estrogen that included antioxidant function and anti-inflammatory and membrane-stabilizing effects. With the high-protein feeding, creatine kinase activity was reduced (women = 98 U·L−1 and men = 126 U·L−1), and although there remains uncertainty about the true magnitude of the response, the observed protein dose-standardized difference was small for men and trivial for women. This difference could be secondary to less exercise-induced membrane damage in women or to an attenuated myocellular response to high-protein feeding. Further indirect evidence for lower sensitivity and attenuated mechanistic response to high-protein feeding in the current female cohort can be seen in the higher perception of general tiredness and leg soreness (males = decreased or no clear effect) and lower rating of leg strength (males = increase possible) in the high-protein condition.

Although estimate uncertainty leaves a substantial difference in performance in the order of that seen in males also a possibility, the lower sensitivity to protein dose in females might be linked to a metabolic mechanism. In response to other nutritional interventions, the response amplitude is also seen to be lower or absent in females compared with matched male counterparts. For example, females displayed an attenuated response to CHO loading and cycle time to exhaustion at 85% V˙O2max relative to men, as indicated by zero increase in muscle glycogen concentration compared with a 41% increase in males and a 5% compared with a 45% improvement in endurance time in males (32). However, when CHO intake was increased from 8.3 to 11.3 g of CHO·kg−1 FFM·d−1 in women and from 8.8 to 11.7 g of CHO·kg−1 FFM·d−1 in men, total muscle glycogen concentration in females increased by a similar magnitude (from 651 ± 188 to 738 ± 198 mmol·kg−1 dry mass in women and from 660 ± 120 to 741 ± 188 mmol·kg−1 dry mass in men), with the authors concluding that the additional CHO energy ameliorated the gender difference in CHO loading when ∼6.7 is compared with ∼8.5 g of CHO·kg−1 FFM·d−1 (32,34). In both the present and in our previous study in males, participants ingested at least 13 and 8 g of CHO·kg−1 FFM on days 2 and 3, respectively, to attain high muscle glycogen concentrations (6,15). These intakes are similar to those reported above (34). Therefore, if we assume a similar response in the current study, it is unlikely that differences in muscle glycogen concentration between conditions can account for the different magnitude of the performance outcome in response to the present dietary intervention, but muscle glycogen measurement would be required to confirm. Alternatively, in review, Tarnopolsky (29,31) found that females use proportionately more endogenous lipid and less CHO and protein (e.g., 46% less [21]) during endurance exercise, which may be due to the 17β-estradiol-mediated metabolism involving higher intramuscular lipid and use, increased expression of genes for fat metabolism, and increased adipocyte lipolysis (31). As such, the lipid-dominated less-flexible glycolytic metabolic response in females could be responsible for the observed inconsequential performance response to the level of protein in the recovery diet. Unfortunately, we were unable to obtain muscle biopsies, and we did not have the foresight to collect respiratory gas, but one interesting metabolic outcome was substantially lower blood lactate concentrations during the performance tests in the high-protein condition. In the previous study in males (24), lactate also trended lower in the high-protein condition on day 2 (−15%; ±19%) but, in contrast, was higher on day 4 (13%; ±9%). Although the higher day 4 concentrations could simply be a result of greater lactate production secondary to higher power output, the opposing lactate response in females could indicate reduced (sparing) CHO metabolism in females in response to the high-protein condition. Females display a lower glucose rate of appearance and disappearance during endurance exercise, which may relate to a lower sympathetic activation (13). Further research is required to better understand gender differences in metabolism and the response to protein feeding manipulation.

Gender differences in muscle fiber composition and physiology may account for the observed higher repeated-sprint mean power slope (fatigue rate) in the previous male cohort compared with females. Power and fatigue rate in multiple-sprint exercises are related to the contribution and capacity for anaerobic metabolism during the initial sprint (3). A higher percentage (39) and relative recruitment of type II muscle fibers, and a lower aerobic contribution during repeated sprints (3), coupled with lower lipid oxidation (31), suggests that males may experience faster muscle glycogen depletion, which could account for the higher FFM-adjusted relative work rates but greater decline (fatigue) in subsequent sprints, relative to females. Furthermore, because high-intensity exercise is an effective stimulus for adaptive response signaling (11), the higher relative exercise-induced protein turnover in men (21), combined with increased protein synthesis because of high-protein ingestion (14), could have supported greater or faster adaptive remodeling compared with females and, therefore, provide another explanation for the apparent insensitivity to protein dose in females. Some support for this idea come from the observation that the overall 8.3% increase (combined treatments) in sprint mean power on day 4 compared with day 2 (24) was 1.8-fold higher than the same relative increase in the current female cohort.

Finally, to the best of our knowledge, this was the first investigation to specifically calculate the protein requirement for well-trained/top sport female athletes undertaking high-intensity endurance exercise. The estimated average protein requirement was 1.6-fold higher than the recommended dietary allowance of 0.8 g·kg−1·d−1 (23) or 2.1-fold higher than the estimated average requirement (22), but within the range of 1.2-1.4 g·kg−1·d−1 suggested for female endurance athletes as estimated by others (22,30) from a dietary survey (30) and from nitrogen balance data in men (4,33), combined with inference fractional differences in protein turnover between genders (21). In the comparative male cohort (24), we estimated the requirement at 1.98 g·kg−1·d−1 (95% confidence interval = 0.69-5.89 g·kg−1·d−1). Although the nitrogen turnover might be transitory until a new protein equilibrium is established-with a 10-d adaptation period possibly required (33), these data suggest that the women required an average 0.65 times the protein of men to meet the nitrogen demands of the a multiday high-intensity exercise protocol. Nitrogen balance methods provide evidence for safe intake. However, we note that several reviewers (e.g., Phillips et al. [22] and Tipton and Wolfe [37]) have recently expressed concern about the validity of using nitrogen balance to make inferences to dietary protein consumption for athletes, whose goal is not only to avoid protein deficiency but also to provide appropriate intakes to optimize all aspects of muscle function, metabolic processes, and adaptation to strenuous training regimens. In fact, Phillips et al. (22) suggested that the primary goal of protein consumption for an endurance athlete is to be able to balance the protein oxidized during exercise and to support increased protein synthesis. Although the grand mean ± SD reported daily protein intake for endurance-trained female athletes were 1.2 ± 0.3 g·kg−1·d−1 (30), perhaps the reasons cited about optimizing adaptation during periods of intensive exercise could help explain the observed average daily protein intakes of 2.3 ± 0.6 g·kg−1·d−1 in stage race conditions or 2.7 ± 0.8 g·kg−1·d−1 during training in elite Australian National Squad female cyclists (17) or the 2.8-3.1 g·kg−1·d−1 observed in men during stage racing (6). Another pertinent point is that the uncertainty of the estimate of requirement and the estimate of how variable the requirement is between individuals (the high SD for nitrogen balance illustrated with the observation that one rider returned zero and another returned a negative balance (effect size > 0.2) on the high-protein condition in the female cohort in Figure 5B, whereas on the same condition, 3 of 12 returned negative balances in the male cohort [24]; not shown) suggests that the estimates of protein requirements in athlete cohorts are not especially precise and that individual demand for protein might vary considerably in athletes training hard; consequently, diligence is required in the athlete's dietary prescription and consensus statements.

The possible error in our estimation of nitrogen output falls mostly in the well-known and comparable methodological limitations, in particular the estimated rather than the measured fecal and miscellaneous obligatory losses of nitrogen (22,33). Noteworthy was that, in the absence of suitable measured or published data in female athletes training hard, we used the same proportional adjustments as used in the males for fecal and miscellaneous losses (24). Interestingly, only one group has measured all routes of nitrogen excretion, including feces in well-trained athletes (33), which is likely due to the displeasure and the risks associated with working with fecal samples. Therefore, more work is required to verify if these and the present estimates hold accurate to the true population mean. Another interesting methodological point that arose while analyzing nitrogen balance data is that-with the exception when first proposed by Calloway et al. (8)-to the best of our knowledge, few researchers in this area have considered nitrogen losses via experimental blood sampling. In the current multiday study, we estimated that approximately 5.15 g of nitrogen was lost from blood proteins and cellular components during the ∼60-h sampling period. The resulting elevation in hemopoiesis and albumin recovery could have had meaningful effects (elevation) on nitrogen balance through increased nitrogen retention. The addition of the estimated nitrogen loss from blood (32 mg of N·mL−1 [8]) changed the mean nitrogen balance intercept to 1.93 g·kg−1·d−1 in females and to 2.61 g·kg−1·d−1 in males. Isotope methods to capture turnover, in addition to a thorough assessment of nitrogen loss, are recommended in future work in dietary protein requirements and performance outcomes in female athletes.

In conclusion, the ingestion of a protein-enriched high-CHO diet for 2 d during the 4-h recovery period immediately after the high-intensity cycling did not clearly change subsequent performance in trained female cyclists, which contrasts with the likely moderate effect in similarly trained males at ∼60 h after the initial loading exercise. Further work is required to investigate whether gender differences in substrate oxidation and protein turnover, muscle characteristics, or attenuated muscle adaptive response could explain the differential response to dietary protein quantity ingested after high-intensity endurance exercise. Finally, the nitrogen balance results provide evidence to support the notion that female endurance athletes exercising intensely have body mass-adjusted daily protein requirements above that of the recommended daily allowance but below those of male endurance athletes.

The authors thank Karin Rossler, Rhys Thorp, Marjolein Ros, Mel Geluk, and Andy Hollings for assistance in the laboratory and David Brasford for creatine and creatinine analysis. The authors also thank Fonterra, New Zealand, Bronston and Jacobs, Auckland, and Nice and Natural, Auckland, for ingredient support.

This study was funded by income obtained from services provided by the Exercise Physiology and Metabolism Laboratory, Massey University, Wellington.

The results of the present study do not constitute endorsement by the American College of Sports Medicine.

Back to Top | Article Outline

REFERENCES

1. Berardi JM, Noreen EE, Lemon PWR. Recovery from a cycling time trial is enhanced with carbohydrate-protein supplementation vs isoenergetic carbohydrate supplementation. J Int Soc Sports Nutr. 2008;5:24.

2. Betts JA, Stevenson E, Williams C, Sheppard C, Grey E, Griffin J. Recovery of endurance running capacity: effect of carbohydrate-protein mixtures. Int J Sport Nutr Exerc Metab. 2005;15:590-609.

3. Billaut F, Bishop D. Muscle fatigue in males and females during multiple-sprint exercise. Sports Med. 2009;39:257-78.

4. Brouns F, Saris WH, Stroecken J, et al. Eating, drinking, and cycling: a controlled Tour de France simulation study. Part I. Int J Sports Med. 1989;10(1 suppl):S32-40.

5. Burke L, Kiens B, Ivy J. Carbohydrates and fat for training and recovery. J Sports Sci. 2004;22:15-30.

6. Burke LM. Nutritional practices of male and female endurance cyclists. Sports Med. 2001;31:521-32.

7. Cade JR, Reese RH, Privette RM, Hommen NM, Rogers JL, Fregly MJ. Dietary intervention and training in swimmers. Eur J Appl Physiol. 1991;63:210-5.

8. Calloway DH, Odell AC, Margen S. Sweat and miscellaneous nitrogen losses in human balance studies. J Nutr. 1971;101:775-86.

9. Cohen J. The Earth is round (P < 0.05). Am Psychol. 1994;49:997-1003.

10. Enns DL, Tiidus PM. The influence of estrogen on skeletal muscle: sex matters. Sports Med. 2010;40:41-58.

11. Gibala MJ, McGee SL, Garnham AP, Howlett KF, Snow RJ, Hargreaves M. Brief intense interval exercise activates AMPK and p38 MAPK signaling and increases the expression of PGC-1{alpha} in human skeletal muscle. J Appl Physiol. 2009;106:929-34.

12. Hopkins WG, Marshall SW, Batterham AM, Hanin J. Progressive statistics for studies in sports medicine and exercise science. Med Sci Sports Exerc. 2009;41(1):3-13.

13. Horton TJ, Grunwald GK, Lavely J, Donahoo WT. Glucose kinetics differ between women and men, during and after exercise. J Appl Physiol. 2006;100:1883-94.

14. Howarth KR, Moreau NA, Phillips SM, Gibala MJ. Coingestion of protein with carbohydrate during recovery from endurance exercise stimulates skeletal muscle protein synthesis in humans. J Appl Physiol. 2009;106:1394-402.

15. Ivy JL. Regulation of muscle glycogen repletion, muscle protein synthesis and repair following exercise. J Sports Sci Med. 2004;3:131-8.

16. Lamont LS, McCullough AJ, Kalhan SC. Gender differences in the regulation of amino acid metabolism. J Appl Physiol. 2003;95:1259-65.

17. Martin M, Martin D, Collier G, Burke L. Voluntary food intake by elite female cyclists during training and racing: influence of daily energy expenditure and body composition. Int J Sport Nutr Exerc Metab. 2002;12:249-67.

18. McLay RT, Thomson CD, Williams SM, Rehrer NJ. Carbohydrate loading and female endurance athletes: effect of menstrual-cycle phase. Int J Sport Nutr Exerc Metab. 2007;17:189-205.

19. Morrison PJ, Hara D, Ding Z, Ivy JL. Adding protein to a carbohydrate supplement provided after endurance exercise enhances 4E-BP1 and RPS6 signaling in skeletal muscle. J Appl Physiol. 2008;104:1029-36.

20. Niles ES, Lachowetz T, Garfi J, et al. Carbohydrate-protein drink improves time to exhaustion after recovery from endurance exercise. J Exerc Physiol. 2001;4:45-52.

21. Phillips SM, Atkinson SA, Tarnopolsky MA, MacDougall JD. Gender differences in leucine kinetics and nitrogen balance in endurance athletes. J Appl Physiol. 1993;75:2134-41.

22. Phillips SM, Moore DR, Tang JE. A critical examination of dietary protein requirements, benefits, and excesses in athletes. Int J Sport Nutr Exerc Metab. 2007;17:S58-S76.

23. Rodriguez NR, DiMarco NM, Langley S. Nutrition and athletic performance. Med Sci Sports Exerc. 2009;41(3):709-31.

24. 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 Metabol. 2008;33:39-51.

25. Rowlands DS, Thomson JS, Timmons BW, et al. Analysis of the effect of high protein-carbohydrate nutrition on global mRNA expression in skeletal muscle during recovery from high-intensity endurance exercise. Physiologist. 2008;51:66.

26. Rowlands DS, Thorp RM, Rossler K, Graham DF, Rockell MJ. Effect of protein-rich feeding on recovery following intense exercise. Int J Sport Nutr Exer Metab. 2007;17:521-43.

27. Saunders MJ, Kane MD, Todd MK. Effects of a carbohydrate-protein beverage on cycling endurance and muscle damage. Med Sci Sports Exerc. 2004;36(7):1233-8.

28. Schmidt FL, Hunter JE. Eight common but false objections to the discontinuation of significance testing in the analysis of research data. In: Harlow L, Mulaik S, Steiger J, editors. What If There Were No Significance Tests? Florence (KY): Routledge; 1997. p. 37-64.

29. Tarnopolsky MA. Gender differences in substrate metabolism during endurance exercise. Can J Appl Physiol. 2000;25:312-27.

30. Tarnopolsky MA. Protein requirements for endurance athletes. Nutrition. 2004;20:662-8.

31. Tarnopolsky MA. Sex differences in exercise metabolism and the role of 17-beta estradiol. Med Sci Sports Exerc. 2008;40(4):645-54.

32. Tarnopolsky MA, Atkinson SA, Phillips SM, MacDougall JD. Carbohydrate loading and metabolism during exercise in men and women. J Appl Physiol. 1995;78:1360-8.

33. Tarnopolsky MA, MacDougall JD, Atkinson SA. Influence of protein intake and training status on nitrogen balance and lean body mass. J Appl Physiol. 1988;64:187-93.

34. Tarnopolsky MA, Zawada C, Richmond LB, et al. Gender differences in carbohydrate loading are related to energy intake. J Appl Physiol. 2001;91:225-30.

35. Thorburn MS, Vistisen B, Thorp RM, et al. Attenuated gastric tolerance but no benefit to performance with adaptation to octanoate-rich esterified oils in well-trained male cyclists. J Appl Physiol. 2006;1001:1733-43.

36. Thorburn MS, Vistisen B, Thorp RM, et al. No attenuation of gastric distress or benefit to performance with adaptation to octanoate-rich esterified oils in female cyclists. Eur J Sport Sci. 2007;7:179-92.

37. Tipton KD, Wolfe RR. Protein and amino acids for athletes. J Sports Sci. 2004;22:65-79.

38. Williams MB, Rave PB, Fogt DL, Ivy JL. Effects of recovery beverages on glycogen restoration and endurance exercise performance. J Strength Cond Res. 2003;17:12-9.

39. Wüst RCI, Morse CI, de Haan A, Jones DA, Degens H. Sex differences in contractile properties and fatigue resistance of human skeletal muscle. Exp Physiol. 2008;93:843-50.

40. Zawadzki KM, Yaspelkis BBI, Ivy JL. Carbohydrate-protein complex increases the rate of muscle glycogen storage after exercise. J Appl Physiol. 1992;72:1854-9.

Keywords:

MUSCLE DAMAGE; GENDER DIFFERENCES; HIGH-INTENSITY ENDURANCE EXERCISE; RECOVERY

©2011The American College of Sports Medicine

Login

Article Tools

Images

Share

Search for Similar Articles
You may search for similar articles that contain these same keywords or you may modify the keyword list to augment your search.

Connect With Us