Don't mess with stress: vitamin supplementation and the role of oxidative stress for aerobic exercise adaptation

The effect of chronic antioxidant and anti-inflammatory supplementation on the adaptive process to exercise training is an emerging area of research, with training implications that have yet to be elucidated fully. The use of these agents has become more popular amongst athletes and recreationally active individuals, thus the long-term consequence of their chronic use for muscle and cardiovascular adaptation is an important consideration for this population. Usage of these supplements relates to the common belief that exercise-induced inflammation and oxidative stress play a crucial role in muscle soreness and weakness in the hours and days following a strenuous bout of exercise. Hence, anti-inflammatory and antioxidant supplements are used with the expectation that reducing exercise-induced stress in fatigued muscles will reduce soreness and promote quicker recovery.

More recently, the role of exercise-induced inflammation and oxidative stress in the adaptive cellular processes that underlie training effects of exercise has been highlighted. The reactive oxygen species produced in the muscle fibre during exercise by metabolic, mechanical and temperature stress pathways alter the redox status of the muscle fibre. Given that many transcription factors are redox sensitive, the shift in redox status that occurs during exercise is important for gene regulation that mediates cellular adaptation to exercise. Hence, by attenuating the acute oxidative stress experienced by muscle fibres during exercise, the use of antioxidant supplementation may attenuate adaptation to this stress (and associated training effects) as well.

Although blunted adaptation to exercise bouts with antioxidant supplementation has been observed in animal models (Gomez-Cabrera et al. 2008), the effects of antioxidant supplementation in humans with respect to cellular adaptation and training/performance effects is not well known. Paulsen et al. (2014) employed a randomized, double-blind study design to investigate the effects of vitamin C (1000 mg) and vitamin E (235 mg), supplemented daily for 11 weeks, on the effects of aerobic training, in comparison to a placebo supplement group. Oxidative stress, gene regulation and immunohistochemical and performance parameters were measured prior to and following the 11 week aerobic training intervention to evaluate the influence of antioxidants on training adaptations (cellular and performance), while simultaneously evaluating oxidative stress and gene regulation as a primary mechanism. While some differences were observed in markers of mitochondrial biogenesis and adaptive signalling in favour of the placebo group, no differences were observed between groups for aerobic performance.

The study by Paulsen et al. (2014) should be commended for several reasons, most notably in its initiative to clarify the previously unstudied interaction between aerobic exercise and vitamin supplementation in humans, building on the work done by Yfanti et al. (2009). The robust study design, e.g. the large sample size and the selection of commonly used daily dosages of the supplements, lend confidence to the results. It is noteworthy that the authors attempted to elucidate the underlying mechanistic physiology, quantifying both cellular adaptations (angiogenesis, mitochondrial content/biogenetic gene regulation) and aerobic training effects (performance on maximal and submaximal tests). The use of both interval and continuous exercise training contributed strong external validity to the study design, because both may be expected to be included in a typical aerobic training programme.

While this is an important and novel area of research, the rationale for potential effects of antioxidant supplementation on blunting of training effects needs to be more clearly communicated to convey to readers both the relevance and the strength of the study design. Likewise, a highly relevant finding from the study was not elaborated on, but certainly warrants further discussion. Both groups showed similar improvements in maximal oxygen uptake, despite inhibition of adaptive cell-signalling pathways observed in the antioxidant supplementation group. Compensatory adaptations probably occurred to account for the improvements in aerobic capacity, yet the authors do not comment on potential mechanisms for this finding. It is therefore unknown whether these potential mechanistic changes were due to the effects of supplementation.

Although the study design provided many strong elements, further design limitations may have precluded the researchers from observing true effects of antioxidant supplementation on aerobic training effects. For example, Ristow et al. (2009) measured the concentration of thiobarbituric acid reactive substances within skeletal muscle of previously untrained individuals following 3 days of exercise (in the presence or absence of vitamin C and E supplementation; Ristow et al. 2009). The tissue-specific measurement of lipid peroxidation performed by Ristow et al. (2009) may have provided a more accurate representation of the muscle fibre-specific redox environment and associated potential for antioxidant supplementation to effect muscle fibre adaptation. The study by Paulsen et al. (2014) used the plasma measurement of 8-isoprostane measured in the circulation, which may be more indicative of vascular stress, as opposed to muscle-specific oxidative stress, the fundamental physiological change expected to be affected by antioxidant supplementation.

The sampling time point selection of a pre- and postintervention construct may have missed differences in acute changes induced by an exercise bout. The premise for an effect of antioxidants on chronic training effects is related to the chronic blunting of the acute oxidative stress response to exercise, thereby attenuating associated gene regulation, which mediates training effects. The basal state of oxidative stress (measured systemically vs. tissue specifically) and subsequent gene regulation may not have been different between groups. However, the differences in the acute oxidative stress response and gene regulation associated with a bout of exercise may have differed, but may have been missed due to the sampling time point selection. Measurements of certain myogenic (MyoD) and metabolic (HKII, PDK4) genes have been shown to peak 8–12 h following exercise, but return to baseline expression levels 24 h postexercise (Yang et al. 2005). It would have been relevant to assess the regulation of these genes, in particular, in response to acute aerobic exercise in the study by Paulsen et al. (2014).

The influence of antioxidants on the acute inflammatory response to exercise has been shown by Fischer et al. (2004). Supplementation with vitamins C and E was observed to inhibit the release of interleukin 6 from contracting skeletal muscle in untrained humans following aerobic exercise (Fischer et al. 2004), although the response in untrained individuals is likely to differ from the response in those who are trained, such as the participant population in the study by Paulsen et al. (2014). While performance and muscle adaptation to aerobic training may be quantified adequately in a pre–post model, measuring the change in the acute response to an aerobic exercise stress may have been a more appropriate assessment of the fundamental mechanism investigated by Paulsen et al. (2014). The influence of supplementation on the acute response to exercise over the 11 week intervention could have been achieved using pre- and postsupplementation samples of an exercise bout prior to and following the aerobic intervention. However, it is acknowledged that this approach comes with certain limitations, because resource and logistical issues may restrict additional sampling. Furthermore, accurate sampling with acute exercise bouts is difficult to achieve, because the time courses of the relevant markers may differ from one another, thereby requiring multiple biopsies for adequate analyses.

Future research is warranted on perhaps the most critical issue that was not examined by Paulsen et al. (2014), namely the long-term impact of antioxidant supplementation and its effect on aerobic exercise performance. Subsequent studies could examine this effect over a period of time longer than 11 weeks and use more relevant performance measures, such as time trials or time to fatigue. The potential effects of antioxidant supplementation on gene regulation following acute aerobic exercise could prove to be irrelevant if these effects did not persist, or perhaps were even reversed in the long term.

The study by Paulsen et al. (2014) is well designed and contributes novel findings of antioxidant effects on aerobic training adaptations in a human training model. These novel findings need to be substantiated with future work involving measures of gene regulation and muscular oxidative stress in untrained participants, both before and after an acute bout of exercise, in order to understand more clearly the relationship between aerobic exercise training and vitamin C and E supplementation. As it stands currently, the true effects of chronic vitamin supplementation on aerobic training effects remain unclear.

Additional information

Competing interests

None declared.