Methods

Twenty healthy male volunteers (mean (SD) age 22.1 (4.1) years) agreed to be tested by signing an informed consent form. Their mean (SD) body mass index was 23.5 (1.6), their mean (SD) resting heart rate was 64 (7.8) beats/min, and every week they exercised for 2.6 (1.4) hours aerobically and 1.4 (1.3) hours anaerobically. The subjects refrained from exercise for 24 hours (day 0) before the experiment. Urine was then collected by the volunteers for a 24 hour control non-exercising period (day 1). On day 2, the participants ran on a treadmill (Powerjog JX100) at 70% of their maximal hear rate reserve (MHRR) for 30 minutes in the laboratory. This exercise intensity was selected because a recent literature review⁴ shows that changes in mood are commonly reported at 60–80% MHRR. On completion of the exercise and before recovery, subjects were asked to indicate the perceived intensity of their workout on a three point (light, moderate, hard) rating scale. After the laboratory session, the participants collected their urine for a further 24 hours (day 2). Once collected, the urine was kept at 4°C until transported to the laboratory where it was frozen. Phenylacetic acid levels in the samples were stable using this protocol. Urinary volumes were all in excess of 0.8 litres.⁵ As the total weight of phenylacetic acid in the 24 hour urine was measured (mg/24 hours), the subjects were free to consume water or other liquids ad libitum.

The concentrations of urinary phenylacetic acid were determined by the gas liquid chromatography method of Gusovsky et al.⁵. Each sample was run in the presence of a standard concentration of an internal control, phenylpropionic acid. Standards of phenylacetic acid in the range 10–40 μg/ml were used to calibrate the column. Peak areas of phenylacetic acid/phenylpropionic acid were calculated and used in the analyses.

Results

In 18 of the 20 subjects, the level of phenylacetic acid in the urine was higher after exercise, increasing by 14–572% compared with the values before exercising (fig 1). The mean (SD) value of phenylacetic acid before exercise was 99.4 (54.4) mg/24 hours and 176.0 (47.7) mg/24 hours after exercise. The difference between the two measurements was significant (F_1,19 = 26.6, p<0.0001; effect size (ES) = 1.2). The correlation between the two sets of scores, however, was not significant (r = 0.33, p = 0.09).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Percentage difference in urinary phenylacetic acid after exercise. Phenylacetic acid concentrations were measured as mg/24 hours and percentages are compared with values obtained before exercise.

Discussion

These results show substantial increases in urinary phenylacetic acid levels 24 hours after moderate to high intensity aerobic exercise. As phenylacetic acid reflects phenylethylamine levels³, and the latter has antidepressant effects, the antidepressant effects of exercise appear to be linked to increased phenylethylamine concentrations. Furthermore, considering the structural and pharmacological analogy between amphetamines and phenylethylamine, it is conceivable that phenylethylamine plays a role in the commonly reported “runners high” thought to be linked to cerebral β-endorphin activity. The substantial increase in phenylacetic acid excretion in this study implies that phenylethylamine levels are affected by exercise.

Although about 75% of subjects responded relatively homogeneously, there was considerable interindividual variability in the phenylacetic acid responses to exercise (fig 1). Interestingly, 17 of the subjects rated the exercise level as moderate, whereas three (11, 18, and 19 in fig 1) rated it as hard. Two of the latter (subjects 18 and 19) also showed the most noticeable increase in phenylacetic acid in the following 24 hours. (It should be noted that our statistical conclusions would not change if the outlier cases, 18 and 19, were disregarded.) The lack of significant correlation between phenylacetic acid levels before and after exercise indicates that the former only accounted for about 11% (r = 0.33; r² = 0.11) of the changes in the latter. Consequently, many factors may mediate phenylacetic acid responses to exercise, possibly including perceived and/or actual exercise intensity. Determination of these factors remains the object of future inquiries.

The present findings should serve as an incentive for further research into the mechanism(s) linking phenylethylamine to exercise. Such research should consider some important factors that were not addressed in this pilot study. Firstly, the inclusion of a passive activity control group is advised. Secondly, instead of relying on MHRR as here, future studies need to assess the actual Vo₂max of the participants. Thirdly, the changes in phenylacetic acid may be different in a sedentary sample in contrast with the relatively fit and physically active sample tested here. Therefore the influence of fitness on phenylacetic acid levels also needs to be examined. Finally, the effects reported here should also be examined in a clinically depressed population.