Characterizing the plasma metabolome during and following a maximal exercise cycling test

Document Type

Journal Article

Publication Title

Journal of Applied Physiology


The American Physiological Society


Centre for Integrative Metabolomics and Computational Biology




Manaf, F. A., Lawler, N. G., Peiffer, J. J., Maker, G. L., Boyce, M. C., Fairchild, T. J., & Broadhurst, D. (2018). Characterizing the plasma metabolome during and following a maximal exercise cycling test. Journal of Applied Physiology, 125(4). 1193-1203.

Available here.


Although complex in nature, a number of metabolites have been implicated in the onset of exercise-induced fatigue. The purpose of this study was to identify changes in the plasma metabolome and specifically, to identify candidate metabolites associated with the onset of fatigue during prolonged cycling. Eighteen healthy and recreationally active men (mean ± SD; age: 24.7 ± 4.8 yr; mass 67.1 ± 6.1 kg; body mass index: 22.8 ± 2.2; peak oxygen uptake: 40.9 ± 6.1 ml·kg−1·min−1) were recruited to this study. Participants performed a prolonged cycling time-to-exhaustion (TTE) test at an intensity corresponding to a fixed blood lactate concentration (3 mmol/l). Plasma samples collected at 10 min of exercise, before fatigue (last sample before fatigue <10 min before fatigue), immediately after fatigue (point of exhaustion), and 20 min after fatigue were assessed using a liquid chromatography-mass spectrometry-based metabolomic approach. Eighty metabolites were putatively identified, with 68 metabolites demonstrating a significant change during the cycling task (duration: ~80.9 ± 13.6 min). A clear multivariate structure in the data was revealed, with the first principal component (36% total variance) describing a continuous increase in metabolite concentration throughout the TTE trial and recovery, whereas the second principal component (14% total variance) showed an increase in metabolite concentration followed by a recovery trajectory, peaking at the point of fatigue. Six clusters of correlated metabolites demonstrating unique metabolite trajectories were identified, including significant separation in the metabolome between prefatigue and postfatigue time points. In accordance with our hypothesis, free-fatty acids and tryptophan contributed to differences in the plasma metabolome at fatigue.