Gut microbiota is involved in the development of several chronic diseases, including diabetes, obesity, and cancer, through its interactions with the host organs. It has been suggested that the cross talk between gut microbiota and skeletal muscle plays a role in different pathological conditions, such as intestinal chronic inflammation and cachexia. However, it remains unclear whether gut microbiota directly influences skeletal muscle function. In this work, we studied the impact of gut microbiota modulation on mice skeletal muscle function and investigated the underlying mechanisms. We determined the consequences of gut microbiota depletion after treatment with a mixture of a broad spectrum of antibiotics for 21 days and after 10 days of natural reseeding. We found that, in gut microbiota-depleted mice, running endurance was decreased, as well as the extensor digitorum longus muscle fatigue index in an ex vivo contractile test. Importantly, the muscle endurance capacity was efficiently normalized by natural reseeding. These endurance changes were not related to variation in muscle mass, fiber typology, or mitochondrial function. However, several pertinent glucose metabolism markers, such as ileum gene expression of short fatty acid chain and glucose transporters G protein-coupled receptor 41 and sodium-glucose cotransporter 1 and muscle glycogen level, paralleled the muscle endurance changes observed after treatment with antibiotics for 21 days and reseeding. Because glycogen is a key energetic substrate for prolonged exercise, modulating its muscle availability via gut microbiota represents one potent mechanism that can contribute to the gut microbiota-skeletal muscle axis. Taken together, our results strongly support the hypothesis that gut bacteria are required for host optimal skeletal muscle function.

Gut microbiota produces a wide and diverse array of metabolites that are an integral part of the host metabolome. The emergence of the gut microbiome-brain axis concept has prompted investigations on the role of gut microbiota dysbioses in the pathophysiology of brain diseases. Specifically, the search for microbe-related metabolomic signatures in human patients and animal models of psychiatric disorders has pointed out the importance of the microbial metabolism of aromatic amino acids. Here, we investigated the effect of indole on brain and behavior in rats. Indole is produced by gut microbiota from tryptophan, through the tryptophanase enzyme encoded by the tnaA gene. First, we mimicked an acute and high overproduction of indole by injecting this compound in the cecum of conventional rats. This treatment led to a dramatic decrease of motor activity. The neurodepressant oxidized derivatives of indole, oxindole and isatin, accumulated in the brain. In addition, increase in eye blinking frequency and in c-Fos protein expression in the dorsal vagal complex denoted a vagus nerve activation. Second, we mimicked a chronic and moderate overproduction of indole by colonizing germ-free rats with the indole-producing bacterial species Escherichia coli. We compared emotional behaviors of these rats with those of germ-free rats colonized with a genetically-engineered counterpart strain unable to produce indole. Rats overproducing indole displayed higher helplessness in the tail suspension test, and enhanced anxiety-like behavior in the novelty, elevated plus maze and open-field tests. Vagus nerve activation was suggested by an increase in eye blinking frequency. However, unlike the conventional rats dosed with a high amount of indole, the motor activity was not altered and neither oxindole nor isatin could be detected in the brain. Further studies are required for a comprehensive understanding of the mechanisms supporting indole effects on emotional behaviors. As our findings suggest that people whose gut microbiota is highly prone to produce indole could be more likely to develop anxiety and mood disorders, we addressed the issue of the inter-individual variability of indole producing potential in humans. An in silico investigation of metagenomic data focused on the tnaA gene products definitively proved this inter-individual variability.