Journal of Applied Physiology

Effects of l-tyrosine and carbohydrate ingestion on endurance exercise performance

Troy D. Chinevere, Robert D. Sawyer, Andrew R. Creer, Robert K. Conlee, Allen C. Parcell


To test the effects of tyrosine ingestion with or without carbohydrate supplementation on endurance performance, nine competitive cyclists cycled at 70% peak oxygen uptake for 90 min under four different feeding conditions followed immediately by a time trial. At 30-min intervals, beginning 60 min before exercise, each subject consumed either 5 ml/kg body wt of water sweetened with aspartame [placebo (Pla)], polydextrose (70 g/l) (CHO), l-tyrosine (25 mg/kg body wt) (Tyr), or polydextrose (70 g/l) and l-tyrosine (25 mg/kg body wt) (CHO+Tyr). The experimental trials were given in random order and were carried out by using a counterbalanced double-blind design. No differences were found between treatments for oxygen uptake, heart rate, or rating of perceived exertion at any time during the 90-min ride. Plasma tyrosine rose significantly from 60 min before exercise to test termination (TT) in Tyr (means ± SE) (480 ± 26 μmol) and CHO+Tyr (463 ± 34 μmol) and was significantly higher in these groups from 30 min before exercise to TT vs. CHO (90 ± 3 μmol) and Pla (111 ± 7 μmol) (P < 0.05). Plasma free tryptophan was higher after 90 min of exercise, 15 min into the endurance time trial, and at TT in Tyr (10.1 ± 0.9, 10.4 ± 0.8, and 12.0 ± 0.9 μmol, respectively) and Pla (9.7 ± 0.5, 10.0 ± 0.3, and 11.7 ± 0.5 μmol, respectively) vs. CHO (7.8 ± 0.5, 8.6 ± 0.5, and 9.3 ± 0.6 μmol, respectively) and CHO+Tyr (7.8 ± 0.5, 8.5 ± 0.5, 9.4 ± 0.5 μmol, respectively) (P < 0.05). The plasma tyrosine-to-free tryptophan ratio was significantly higher in Tyr and CHO+Tyr vs. CHO and Pla from 30 min before exercise to TT (P < 0.05). CHO (27.1 ± 0.9 min) and CHO+Tyr (26.1 ± 1.1 min) treatments resulted in a reduced time to complete the endurance time trial compared with Pla (34.4 ± 2.9 min) and Tyr (32.6 ± 3.0 min) (P < 0.05). These findings demonstrate that tyrosine ingestion did not enhance performance during a cycling time trial after 90 min of steady-state exercise.

  • central fatigue
  • cycling
  • perceived exertion

recently, it has been hypothesized that, during prolonged exercise, an increased concentration of brain serotonin may be an important factor in the onset of central nervous system fatigue (2-4, 8, 10,27) and a high serotonin-to-dopamine ratio results in fatigue (17). Brain serotonin synthesis depends on the availability of free tryptophan, its amino acid precursor, and the activity of the rate-limiting enzyme, tryptophan hydroxylase (7,10). Similarly, tyrosine is the amino acid precursor to dopamine (32). These amino acid precursors compete for transport across the blood-brain barrier via the same carrier mechanism (17).

We speculated that, if tyrosine were elevated in the blood by ingestion during exercise and competed for transport across the blood brain barrier with tryptophan, increased uptake of tyrosine and a decreased uptake of tryptophan could result in a lower brain serotonin/dopamine ratio and improved endurance.

Limited research has been done on the effects of tyrosine ingestion on exercise endurance. Struder et al. (30) reported no beneficial effect of tyrosine supplementation when subjects cycled to exhaustion. On the other hand, Chaouloff et al. (10) reported that high doses of α-methyl-p-tyrosine improved exercise performance in rats, and the improved performance correlated with elevated dopamine concentration in the brain. Some investigations have shown that tyrosine administration increases dopamine synthesis and concentration in the brain (1, 19, 20), whereas others have shown that tyrosine administration leads to improvement of mood and well-being in human subjects under stress (5, 24). These results raise the possibility that tyrosine administration during exercise could offset feelings of fatigue and lead to improved performance.

The purpose of this study, therefore, was to test the effects of tyrosine ingestion on endurance under conditions of prolonged exercise. Because carbohydrate ingestion has also been shown to reduce the availability of free tryptophan and to improve endurance for prolonged exercise (18), we also determined whether combined tyrosine and carbohydrate supplementation has a greater beneficial effect on endurance than either one alone.


Nine male competitive cyclists from the local population participated in this study [25 ± 1 yr; 182 ± 2 cm; 73 ± 2 kg; peak oxygen consumption (V˙o 2 peak) of 4.5 ± 0.2 l/min, mean ± SE]. The Human Subjects Institutional Review Board at Brigham Young University approved this study, and all subjects were informed of the risks, stresses, and benefits of the investigation before signing an informed consent form.


To estimate submaximal workloads, subjects performed a continuous, progressive bicycle ergometer (Lode Excalibur, Lode, Groningen, Netherlands) protocol to determineV˙o 2 peak. Subjects began cycling at a resistance of 125 W. The resistance was increased by 25 W per min until volitional exhaustion. Expired volume was determined by a Fleisch pneumotach, and expired oxygen and carbon dioxide fractions were analyzed by a mass spectrometer (Marquette). Oxygen uptake (V˙o 2) and carbon dioxide production were calculated every 15 s by an on-line computer program (Consentius). The mass spectrometer was calibrated before testing by use of certified medical gases of known concentration. The heart rate was monitored continuously by radiotelemetry (Polar Electro, Port Washington, NY), and rating of perceived exertion (RPE, Borg 6–20 scale) was recorded at 1-min intervals.

Subjects also performed a familiarization trial to acquaint them with the time trial testing procedures and to minimize potential learning effects. During the familiarization trial, subjects pedaled for 90 min at a work rate demanding ∼70% ofV˙o 2 peak, followed immediately by a time trial performance test.

Experimental testing.

Experimental trials were given in random order, and the experiment was carried out with the use of a counterbalanced double-blind design. All trials were separated by 1 wk. In the evening before each trial, subjects underwent a 60-min ride at ∼70%V˙o 2 peak to normalize muscle glycogen. They then received a meal containing 1,300–1,330 kcal (73% carbohydrate, 13% fat, and 14% protein). During the 48 h preceding each trial, subjects refrained from vigorous activity with the exception of the 60-min ride at ∼70%V˙o 2 peak on the evening before each trial. A dietary record was kept, and the subjects were instructed to replicate food intake before each subsequent trial. Subjects reported to the laboratory in the morning after an overnight fast.

For blood sampling purposes, a Teflon intravenous catheter was inserted into a forearm vein under sterile conditions. The catheter remained in place during the test trials for sampling at 60 and 30 min before exercise (PRE60 and PRE30, respectively). Blood samples were also collected at the onset of exercise (E0), 30 (E30), 60 (E60) and 90 min (E90) during exercise, after 15 min (ETT15) into the endurance time trial, and immediately on test termination (TT). At PRE60, PRE30, E0, E30, E60, and E90, subjects consumed their randomly assigned drink supplement. The drink supplements consisted of either 5 ml/kg body wt of water sweetened with aspartame [placebo (Pla)], a 5 ml/kg solution with polydextrose (70 g/l; CHO), a 5 ml/kg body wt solution withl-tyrosine (25 mg/kg body wt; Tyr), or a 5 ml/kg solution with polydextrose (70 g/l) and l-tyrosine (25 mg/kg body wt) (CHO+Tyr). The drinks were matched in color and taste and delivered to the subjects in opaque water bottles. The drinks were developed and coded before delivery to the research center, and the codes were not broken until all of the data had been analyzed. After the 60-min rest period, the subjects exercised on the cycle ergometer for 90 min at a work rate demanding ∼70% V˙o 2 peak. During the test trials, gas samples were taken every 15 min to ensure that ∼70% V˙o 2 peak was maintained. Heart rate, respiratory exchange ratio (RER), and RPE (Borg 6–20) were recorded every 15 min. Table 1illustrates the experimental protocol during each trial.

View this table:
Table 1.

Time points during experimental procedures when subjects ingested drinks, blood samples were taken, andV˙o 2, HR, RER, and RPE were measured

Immediately after the 90-min cycling bout, subjects began a time trial performance test that required completion of a predetermined amount of work as rapidly as possible. The amount of work was equivalent to the amount of work completed while cycling at 70%V˙o 2 peak for 30 min. To calculate the total work to be performed during the time trial, a modification of a formula originally proposed by Jeukendrup et al. (23) was usedTotal amount of work for time trial(J)=0.70·Wmax·1,800 Subjects were aware of the amount of work accumulated but blinded to the elapsed time. The V˙o 2, heart rate, RER, and RPE were recorded at 15-min intervals during the time trial. Elapsed time was recorded at the end of the time trial.

Blood analyses.

All blood samples (5 ml) were drawn into a prechilled EDTA-containing (5 μl/ml whole blood) 12 × 75-mm tube and stored in ice water for 10 min and then centrifuged (Beckman model TJ-6R) at 1,520g for 10 min at 4°C. A 2-ml plasma aliquot was transferred to a separate tube and stored frozen (−20°C). Plasma lactate and glucose were analyzed in triplicate by use of an Analox Micro-Stat GM7 analyzer (Analox Instruments, Lunenburg, MA). Plasma free fatty acid (FFA) was determined enzymatically according to the method of Shimizu et al. (28). For separation of plasma free tryptophan and albumin-bound tryptophan, the remaining plasma (∼1 ml) was transferred to a Centrifree micropartition device (Millipore, Bedford, MA) and centrifuged at 1,500 g at 25°C for 20 min. The ultrafiltrate was then stored frozen (−80°C). Analyses of tyrosine and free tryptophan were performed directly from the ultrafiltrate by reverse-phase high-performance liquid chromatography (Coulochem II, ESA). The chromatogram was equipped with a Supelcosil LC-18-DB, 150-mm × 4.6-mm column. Column temperature was kept constant at 45°C. Flow rate of the isocratic mobile phase solution (0.14 M sodium acetate, 4% acetonitrile, pH 6.4) was kept constant at 1 ml/min. The ultraviolet detector (Waters 996) was set at 225 nm.

Statistical analysis.

Oxygen consumption, heart rate, RER, RPE, plasma glucose, lactate, FFA, tyrosine, free tryptophan, tyrosine-to-free tryptophan ratio, and time to complete the time trial were analyzed with a two-factor ANOVA with repeated measures. Differences between means for treatments at each time were ascertained by examining the 95% confidence intervals. The null hypothesis was rejected when P < 0.05. All data are reported as means ± SE.


Cardiorespiratory responses.

During the 90-min ride, mean V˙o 2(expressed as mean %V˙o 2 peak) across all time points was found to be 68.5 ± 1.3% for CHO, 68.9 ± 1.6% for CHO+Tyr, 70.7 ± 1.6% for Pla, and 70.0 ± 1.3% for Tyr. At ETT15 and TT for CHO+Tyr,V˙o 2 was significantly higher vs. any other time within the same trial, and TT was higher than ETT15 (P < 0.05) (Table 2). For CHO at TT, V˙o 2 was higher vs. all other times within this trial (P < 0.05). Although oxygen consumption trended in an upward fashion for both CHO trials during the time trial compared with Pla and Tyr, these values were not significantly different from one another.

View this table:
Table 2.

o 2, HR, RER, and RPE during 90 min of cycling followed immediately by a time trial

For all trials, RER declined steadily during the 90-min ride (Table 2). At ETT15, the subjects showed higher RER values during CHO+Tyr vs. Pla and Tyr, and the RER was significantly higher at TT during CHO+Tyr compared with Tyr and Pla (P < 0.05). For the same trial, the RER was higher at TT vs. all other time points within CHO+Tyr (P < 0.05). At TT during CHO, the RER was higher compared with Pla (P < 0.05). Within the same trial, the RER was also higher at TT for CHO compared with E45, E60, E75, E90, and E105 (P < 0.05).

Heart rate increased progressively throughout the 90-min ride (Table2). During the endurance time trial, heart rate increased abruptly compared with the 90-min ride at 70%V˙o 2 peak when subjects were fed CHO and CHO+Tyr and was significantly higher at TT compared with all other times within these trials, respectively (P < 0.05). CHO and CHO+Tyr heart rate values were higher at TT vs. Pla and Tyr (P < 0.05).


A progressive and similar rise in RPE was observed for the subjects during the 90-min cycling bout in all trials (Table 2). At TT, the RPE in Tyr was lower compared with Pla (P < 0.05).

Blood metabolites.

Plasma glucose was significantly higher at PRE30 and 0 in both CHO and CHO+Tyr (7.02 ± 0.20 and 6.25 ± 0.31 mmol/l, respectively) compared with the Pla and Tyr trials (4.41 ± 0.15 and 4.38 ± 0.12 mmol/l, respectively) (P < 0.05) (Fig.1). From 0 to E30, plasma glucose declined markedly from 6.01 ± 0.34 to 3.78 ± 0.24 mmol/l in CHO and from 5.97 ± 0.48 to 3.95 ± 0.14 mmol/l in CHO+Tyr (P < 0.05). At E60, plasma glucose was significantly higher in the CHO vs. Pla (5.12 ± 0.30 vs. 4.09 ± 0.12 mmol/l, respectively) (P < 0.05). From the onset of exercise to TT, plasma glucose decreased steadily to 3.12 ± 0.17 mmol/l in Pla and 3.27 ± 0.20 mmol/l in Tyr. These values were lower at TT vs. CHO (4.96 ± 0.67 mmol/l) and CHO+Tyr (4.72 ± 0.47 mmol/l) (P < 0.05). No significant differences were found in plasma glucose levels between CHO and CHO+Tyr at any time during exercise.

Fig. 1.

Plasma glucose concentrations during the four trials. TT, test termination; PRE60 and PRE30, 60 and 30 min before exercise, respectively; E0, onset of exercise; E30, E60, and E90, 30, 60, and 90 min during exercise, respectively; ETT15, 15 min into the endurance time trial. apolydextrose (CHO) different from placebo (Pla), l-tyrosine (Tyr), and CHO+Tyr (P < 0.05); bCHO+Tyr different from Pla and Tyr (P < 0.05); cCHO different from Pla and Tyr (P < 0.05); dCHO different from Pla (P < 0.05).

Blood lactate concentration rose from 0 to E30 in all groups and was maintained at near constant levels from E30 to E90 (Fig.2). At TT, the value for CHO+Tyr (7.47 ± 0.76 mmol/l) was higher than CHO (5.74 ± 0.96 mmol/l), Pla (4.21 ± 0.87 mmol/l), and Tyr (5.35 ± 0.88 mmol/l). At ETT15 and TT for CHO+Tyr, blood lactate levels were higher compared with all other times during this trial (P < 0.05). Blood lactate was higher at TT in CHO vs. LA (P< 0.05). Significantly higher lactate levels were observed at ETT15 and TT during CHO, Tyr, and Pla vs. baseline levels and E30, E60, and E90 (P < 0.05).

Fig. 2.

Blood lactate concentrations during the four trials.aCHO+Tyr different from Pla, Tyr, and CHO (P < 0.05); bCHO different from Pla (P < 0.05); cCHO+Tyr different from all other times within this trial (P < 0.05);dCHO different from PRE60, PRE30, E0, E30, E60, and E90 within this trial (P < 0.05); eTyr different from PRE60, PRE30, E0, E30, E60, and E90 within this trial (P < 0.05); fPla different from PRE60, PRE30, E0, E30, E60, and E90 within this trial (P < 0.05); gTyr different from PRE60, PRE30, E0, E30, and E60 within this trial (P < 0.05).

The pattern for the plasma FFA response is shown in Fig.3. In general, consumption of carbohydrate suppressed the levels of FFA compared with the noncarbohydrate trials over the duration of the exercise test.

Fig. 3.

Plasma free fatty acid concentrations during the 4 trials. aPla different from CHO and CHO+Tyr (P < 0.05); bTyr different from CHO and CHO+Tyr (P < 0.05); cCHO different from Pla and Tyr (P < 0.05).

In subjects who ingested tyrosine, plasma tyrosine concentration rose significantly from baseline values throughout exercise (Fig.4). From PRE30 to TT, plasma tyrosine levels were significantly increased in Tyr and CHO+Tyr vs. Pla and CHO (P < 0.05).

Fig. 4.

Plasma tyrosine concentrations during the 4 trials.aCHO +Tyr different from CHO and Pla (P < 0.05); bTyr different from CHO and Pla (P< 0.05).

Plasma free tryptophan levels declined from PRE60 to E30 in all groups. Compared with baseline values, plasma free tryptophan was significantly lower at E30 in CHO and CHO+Tyr (P < 0.05) (Fig.5). From E60 to TT, plasma free tryptophan levels rose in all groups. At each time point from E90 to TT, Pla and Tyr values were significantly higher than those for CHO and CHO+Tyr (P < 0.05). No significant differences were observed in plasma free tryptophan levels between CHO and CHO+Tyr.

Fig. 5.

Plasma free tryptophan concentrations during the 4 trials. aTyr different from CHO and CHO+Tyr (P < 0.05); bPla different from CHO and CHO+Tyr (P < 0.05).

The correlation between FFA and tryptophan was significant for Tyr (r = 0.72, P < 0.05), Pla (r = 0.74), and all treatments taken together (r = 0.70, P < 0.001). The highest correlation was seen with the Pla treatment. The plasma tyrosine-to-free tryptophan ratio (Fig.6) increased dramatically from PRE60 to TT in Tyr and CHO+Tyr. From PRE30, the plasma tyrosine-to-free tryptophan ratio was significantly higher for Tyr and CHO+Tyr vs. CHO and Pla (P < 0.05).

Fig. 6.

Tyrosine-to-free tryptophan ratios during the 4 trials.aTyr different from CHO and Pla (P < 0.05); bCHO+Tyr different from CHO and Pla (P < 0.05).

Endurance time trial.

Subjects fed CHO completed the endurance time trial in 27.17 ± 0.92 min. Those fed CHO+Tyr completed it in 26.11 ± 1.01 min. These times were significantly lower than performance times for Pla (34.44 ± 2.89 min) and Tyr (32.64 ± 3.05 min) (P < 0.05). For the entire F-test, the regular statistical power was observed to be 0.76. For CHO+Tyr, six of nine subjects completed the time trial faster than all other trials, and eight of nine subjects finished faster vs. Tyr and Pla. For the CHO trial, three of nine completed the time trial faster vs. all other trials, six of nine subjects finished faster than Tyr, and eight of nine subjects finished faster than Pla. Six of nine subjects finished the time trial faster during Tyr compared with the Pla trial. No significant differences were found in time to complete the endurance time trial for CHO vs. CHO+Tyr or Pla vs. Tyr. For specific comparisons between CHO vs. CHO+Tyr and Pla vs. Tyr, a SD of 1.5 would be required to detect differences at a statistical power value of ∼0.82 with nine subjects per group (25). The resultant effect size (ES) for CHO+Tyr compared with CHO is calculated as 0.37 with an estimated statistical power value at 0.11 (P = 0.05). For Tyr vs. Pla, the ES is calculated as 0.21, yielding a statistical power value of ∼0.08 (P = 0.05) (25). Despite no significant differences observed between Tyr and Pla and between CHO and CHO+Tyr, the ES and low statistical power suggest that caution be exercised when making conclusions about these latter comparisons.


The purpose of this study was to determine whether repeated doses of l-tyrosine, either with or without carbohydrate feedings, would improve cycling time trial performance after 90 min of submaximal cycling. The results showed that tyrosine ingestion, either alone or with carbohydrates, did not improve performance. The results did confirm that carbohydrate ingestion every 30 min during steady-state exercise significantly improved time trial performance. We conclude that under the experimental conditions of this study tyrosine was not ergogenic.

Only one other study has investigated the effects of tyrosine ingestion on exercise performance in humans (30). In a study by Struder et al. (30), subjects ingested 10 g ofl-tyrosine 15 min before and 60 min after beginning an exercise bout corresponding to an intensity of 2 mmol/l blood lactate. After 90 min of exercise, peak plasma tyrosine levels rose from ∼90 to ∼260 μmol/l and then declined slightly to ∼240 μmol/l at exhaustion (150 ± 42 min of exercise). Their results showed no effect of tyrosine on exercise time to exhaustion nor on the outcomes of tests of mental performance and self-perception performed immediately after the exercise to exhaustion. In their subjects who consumed tyrosine, they found high plasma prolactin levels that would suggest reduced dopamine levels in the brain (30). Animal studies have shown that tyrosine doses of 20 mg/kg resulted in marked increases in dopamine synthesis, but doses of 50 mg/kg resulted in dopamine levels less than baseline (1). It is possible that the doses used by Struder et al. were too high and led to an inhibition of dopamine synthesis rather than a stimulus. The doses used in the present study were about half of those used by Struder and associates but yielded plasma concentrations twice as high as those reported by them despite having similar baseline values as in this study (∼80–100 μmol/l). This might reflect the differences in mode and timing of administration and measurement between the two studies. Nevertheless, the lack of effect of tyrosine in our study may also be the result of excess tyrosine in the ingestate. Perhaps, in contrast to carbohydrate feeding, the continuous ingestion of tyrosine over time, as imposed under the present design, exceeds the amount that might be beneficial. Because we could not measure brain concentrations of any substances, all of this is conjecture, but the possibility is intriguing. The present results lay the foundation for further investigation.

It was our aim in the present study to increase the plasma tyrosine-to-free tryptophan ratio in the blood of our subjects, because a high tyrosine-to-free tryptophan ratio may favor tyrosine uptake into the brain and could subsequently augment brain dopamine synthesis and reduce serotonin synthesis and potentially enhance performance. The results show that plasma tyrosine levels increased approximately fivefold whereas plasma free tryptophan increased 19% and decreased ∼13% in Tyr and CHO+Tyr at TT compared with baseline values in these groups, respectively. The plasma tyrosine-to-free tryptophan ratio at TT was 43.4 in Tyr and 50.0 in CHO+Tyr compared with baseline values of ∼10.0. Although these increases in the plasma tyrosine-to-free tryptophan ratio may have resulted in reduced free tryptophan uptake and increased tyrosine uptake into the brain, they did not lead to significantly enhanced performance.

Some observations in the present results do suggest a possible enhancing effect of tyrosine. The V˙o 2, RER, and blood lactate reached significantly higher levels at TT in CHO+Tyr than in CHO alone. In addition, higherV˙o 2 and RER data were seen during CHO+Tyr at ETT15 with respect to all other trials and compared with previous times within the same trial, suggesting that subjects were exercising harder during this time period. Even though these metabolic responses did not result in a significant improvement on the time trial, they may suggest that had the time trial been designed differently, perhaps longer in length, the tyrosine in conjunction with carbohydrate may have resulted in improved performance. This suggestion is derived from the report of Banderet and Lieberman (5), who observed enhanced performance in numerous mood, cognitive, reaction time, and vigilance measures while subjects were exposed to 4.5 h of extreme environmental conditions. They suggested that decrements in performance resulting from central catecholamine depletion during prolonged exposure to stress could be attenuated by tyrosine ingestion. Further support for this supposition is found in our RPE data of Table 2. The fact that no differences were found in RPE for CHO+Tyr vs. CHO, even though the metabolic data indicate that subjects in CHO+Tyr were working harder during the time trial, suggests that enhanced central dopaminergic activity may have occurred and nullified the perception of fatigue. That this did not translate into improved performance time may once again be a reflection of the design. On the other hand, all of these observations taken together could mean that tyrosine ingestion somehow invoked an ergolytic response. For example, the higherV˙o 2 and RER in CHO+Tyr compared with Tyr with no difference in RPE and performance time might suggest that the abundance of plasma tyrosine might have resulted in reduced metabolic efficiency. Future studies must be designed to test these possibilities.

In the present study, the feeding of glucose within 60 min before exercise resulted in a significant elevation of blood glucose at the onset of exercise but led to a precipitous drop in blood glucose during the first 30 min of the exercise bout. Costill and co-workers (14) observed a similar response and attributed the decline of blood glucose to the combined effects of exercise and an elevated insulin concentration. In their study, endurance was reduced as a result of preexercise consumption of glucose. In contrast, in our study, blood glucose levels were restored in both CHO groups after E60 because of repeated carbohydrate consumption, and improvements were seen in time trial performance. The improved performance seen in this study concurs with previous studies that found that carbohydrate ingestion during exercise improves performance (13, 15, 21, 22,26). The improvement seen in the CHO groups is likely due to the maintenance of blood glucose homeostasis allowing a constant supply of glucose for oxidation in the working muscle (12). In addition, one could also speculate that the mechanism for a carbohydrate-induced increase in performance might also be related to the indirect effect on the brain resulting from glucose alteration of tryptophan. For instance, elevated blood glucose levels resulting from carbohydrate ingestion during exercise have been shown to suppress FFA levels (11, 26). This is clearly the case in the present study in which both CHO groups had lower FFA levels compared with the non-CHO groups (Fig. 3). Suppressed FFA levels have been shown to reduce plasma free tryptophan, because FFA and tryptophan compete for binding to albumin (16). The lower the FFA, the less tryptophan that would come off the albumin and the lower would be the level of free tryptophan. This response is also clearly demonstrated in Fig. 5, which shows that in the two CHO groups that have less FFA there is a reduction in free tryptophan. Previous research suggests that fatigue after prolonged exercise is associated with elevations of free tryptophan and serotonin in various regions of the brain and cerebrospinal fluid (6, 8-10) resulting from high concentrations of plasma free tryptophan. It is possible that, in the present study, the feeding of glucose resulted in a reduction of free tryptophan and a concomitant reduction of the tryptophan and serotonin in the brain and an improvement in performance. This suggestion is in harmony with that of Davis et al. (18) who observed similar effects of glucose on tryptophan during prolonged exercise.

In summary, this study was designed to test whether repeated ingestions of tyrosine during prolonged exercise could improve performance of human subjects during an endurance time trial. Despite evidence that tyrosine may have promoted some beneficial central effects that reduced perception of fatigue, it had no significant effect on performance time during the time trial. Carbohydrate feedings resulted in enhanced performance as observed in previous studies. The metabolic data suggest that the beneficial effects of carbohydrate ingestion may be related as much to its indirect effect on reducing central fatigue as it is to its well known peripheral effects on substrate metabolism.


The authors thank Dong Ho Han for excellent technical assistance.


  • This work was supported in part by the Gatorade Sports Science Institute and Natures Sunshine.

  • Address for reprint requests and other correspondence: A. C. Parcell, Human Performance Research Center, Brigham Young Univ., 120-E Richards Bldg., Provo, UT 84602 (E-mail:allen_parcell{at}

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

  • July 5, 2002;10.1152/japplphysiol.00625.2001


View Abstract