The effects of exercise training on the gut microbiota, microbial metabolites and microbial-host interactions in mice and humans

The gut microbiome (GM) consists of trillions of microbes that can modulate numerous physiological systems of its mammalian host. It is well established that behavioral practices and the environment encountered by the host can contribute significantly to composition and functional capacity of the GM. Research investigating the roles of the GM in modulating host physiology has also indicated an important role of the gut microbes in causing, exacerbating, or preventing disease. Considering these findings, it is vital to have an in-depth understanding of the effects that behaviors and environmental stimuli have on the composition and functional capacity of the gut microbiota. Of these potential mediators, exercise training has recently emerged as a candidate. Data relating exercise and the microbiome have pointed to intriguing associations between changes in the GM and host physiology, including alterations in behavior, immunity and metabolism. However, this research has been confined primarily to animal models and uses only correlations to address possible interactions between microbe and host. Therefore, the primary purpose of the work presented in this dissertation is to understand how exercise training can modulate the gut microbiota while concurrently unraveling the ways in which GM changes may directly contribute to changes in host physiology, in both mice and humans. This dissertation was divided into three major sections. First, we followed up on research conducted previously in our laboratory demonstrating that two six-week exercise training modalities (Forced Treadmill [FTR] vs. Voluntary Wheel Running [VWR]) led to differential inflammatory and clinical responses to a chemically-induced colitis insult in mice. To understand whether these observed outcomes may relate to differential changes in the gut microbiome, we performed an experiment examining the effects of these two training modalities on the GM composition. Interestingly, we observed that VWR and FTR differentially altered the composition of the mouse gut microbiota, indicating that the exercise-induced GM modifications may have contributed to the different colitis outcomes previously observed. In the second section of this dissertation, we investigated whether exercise-induced changes in the GM could contribute directly to host physiology. To accomplish this, we transplanted GMs from donor ‘exercised (VWR)’ or ‘sedentary’ mice into previously germ-free mice. First, we analyzed the composition of GM, concentrations of gut metabolites, and colon inflammation of mice that received GMs from donors. We found that the ‘exercised’ microbiota led to altered microbial communities and short chain fatty acid (SCFA) profiles, higher body weights, and reduced colon inflammation in the recipient, colonized mice. In a second cohort of GF mice, we used the same transplant and colonization design, with an additional administration of dextran-sodium sulfate (DSS) to induce acute colitis. Intriguingly, we found that the ‘exercised’ GM transplant led to an enhanced regenerative anti-inflammatory response in the colons of recipient mice after those mice received DSS for five days. In the last section of this dissertation, we wished to characterize the effects of exercise training on the human microbiome. To accomplish this, we conducted a longitudinal within-subjects design, whereby previously sedentary subjects participated in an endurance-based exercise intervention for six weeks (three days per week). This exercise period was followed by a return to sedentary activity for another six weeks. Fecal and blood samples were collected throughout the study, with dietary controls in place to avoid effects of diet in confounding any exercise-induced GM changes. We also examined whether obesity status had a role in exercise-induced regulation of the GM, and thus subjects were recruited on BMI status (lean vs. obese). After analysis, we found that the GM composition and metabolic capacity was significantly altered by exercise training. Moreover, we found that fecal microbial-derived SCFAs, which we had previously shown to be increased in exercised animals, also increased due to exercise training in humans. The exercise-induced changes in the GM were largely reverted after a return to sedentary activity, providing further evidence that physical activity status contributes significantly to GM composition. Lastly, we identified that obesity plays a role in modulating the GM response to exercise. In summary, this dissertation comprises evidence that indicates that exercise training can alter the gut microbiota and GM derived metabolites in mice and humans. Moreover, we show for the first time that the changes in the gut microbiota have direct, and likely critical, relationships to host physiology during exercise training. In turn, these data presented herein may be important for understanding physiological systems that beneficially impact human health