Effect of exercise on micronutrient status and stress and immune response

Exercise induces a range of physiological responses involving different organs, tissues and systems. The acute exercise response refers to the metabolic and mechanical effects directly following exercise, while the recovery after exercise concerns mechanisms to repair, refuel, replenish and return to homeostasis. Longer-term adaptations are important for growth and supercompensation. The role of nutrition or specific nutrients in these processes is substantial and demands further investigation. In this thesis, we investigated the exercise response during acute and repeated exercise and exercise training. In addition, a dietary intervention was conducted to investigate whether nutritional status, i.e. high vs low carbohydrate intake, modulated of the exercise induced stress and immune response.

First, we assessed the impact of exercise on the variation in blood magnesium levels before exercise and during 6hours post-exercise recovery in well-trained cyclists and triathletes. We showed that both ionized (iMg) and total magnesium (tMg) decreased directly after exercise and returned to pre-exercise levels within 3.5hours after exercise. The decrease in blood magnesium levels after exercise and subsequent increase a few hours later likely reflects re-distribution to muscles and to blood respectively. We concluded that exercise affects magnesium levels and timing of blood sampling to analyse magnesium status is important.

These observations in athletes highlighted the impact of an acute bout of exercise, but what would happen when older adults exercise for multiple days in a row remained unclear. Therefore, we examined changes in iMg and tMg levels during four consecutive days of prolonged walking exercise (~8hours) in a group of very old adults (>80 years). Blood samples were collected at baseline (1 or 2 days before the first walking day) and every walking day directly after finishing. Our results showed that iMg levels dropped directly after the first day of walking, while tMg showed no clear pattern. During subsequent days, iMg levels did not drop after exercise. After exercise, sub-optimal iMg and tMg levels were found in 88% and 16% of the participants, respectively. These sub-optimal levels were not associated with drop-out or health problems.

This previous study protocol, was also used to assess iron parameters in a group of healthy adults. We showed that plasma iron decreased across days, while ferritin increased. Haptoglobin showed a decrease after the first day and increased over subsequent days. Haemoglobin did not change after the first day but increased during subsequent days. These observations probably reflect increased iron losses via foot strike haemolysis, sweat and urine, but also the impact of exercise-induced inflammation on hepcidin and iron status.

The exercise-induced inflammatory response was established by the cytokine (IL-6, IL-8, IL-10, TNF-α and IL-1β) response. The first day of walking exercise caused an increase in cytokine levels, thereafter, levels decreased from day 1 to day 2 and remained rather stable during the following days. These results suggest that an acute inflammatory response occurred after the first day of walking and that individuals adapt rapidly to this type of repeated exercise.

Shifting our focus to prolonged training instead of acute exercise, we assessed whether salivary cortisol and testosterone could be used to assess training load in a group of elite swimmers. Previous research showed that cortisol and testosterone can be used to assess acute exercise stress in athletes, while their usefulness during a training periodization remained unclear. In our study, ten male elite swimmers were monitored during 10 consecutive weeks of training, ending with a competition. Although the training load decreased over the 10weeks, both salivary cortisol and testosterone levels remained unchanged. During competition, both salivary cortisol and testosterone increased directly after the first race and returned back to baseline levels within 2hours after the last race. This let us conclude that salivary cortisol and testosterone can be used to assess acute exercise stress but not prolonged training load over several weeks.

In chapter 7, we reviewed the literature regarding various research designs that were used to study overreaching and overtraining. We had noticed how the field is struggling with the question on how to stimulate, measure and/or follow up overtraining in athletes. Therefore, we discussed the available information on (non)functional overreaching and overtraining from the perspective of the researcher. This review can be seen as a kind of guideline on ‘how to perform an overreaching/overtraining study’ and what are the critical issues that a researcher should keep in mind.
We also investigated the effects of a short-term (2days) and prolonged (2weeks) low carbohydrate (LC) diet (<10En% carbohydrates) on the exercise induced stress and immune response in a cross-over study design with a high carbohydrate diet (HC) as a control diet. The results of this study showed that short-term adherence to a LC diet already led to metabolic changes, as reflected by lower respiratory exchange rates, lower glucose, higher free fatty acids and higher ketone levels. However, the exercise induced stress response was still higher after 2days on the LC diet, and attenuated after 2weeks on the LC diet, shown by very high plasma cortisol levels after 2days on the LC diet. We also showed that 2days adherence to the LC diet resulted in different immune cell counts after exercise compared 2days adherence to the HC diet, resulting in lower T cell, Th cell and B cell counts in the LC diet. These differences were diminished and not significant anymore after 2weeks adherence to both diets. T cell homing kinetics to the airways also showed differences between the LC and HC diet after 2days, but not after 2weeks on the diets, with lower Th cell airway homing on the LC diet 2hours after exercise. At baseline, immune cell count and homing were not different between diets, while differences between diets were detected directly after exercise and 2hours after exercise. A clear exercise response was evident for immune cell differential count and cell proliferation rate. We concluded that adaptation to a LC diet in terms of metabolic and stress response occurs within two weeks, but not after 2days; and that exercise, more than diet, differentially affects homing of immune cells in our study.