Next-generation sequencing of the athletic gut microbiota: a systematic review

Bibliographic analysis

A total of 16 original articles reported the applications of shotgun metagenomics or 16S amplicon sequencing for the analysis of fecal samples of athletes and individuals practicing sports [Table 3 and Supplementary Table 1]. These culture-independent techniques allow for characterising gut microbes without the need to culture them. In this regard, shotgun metagenomics offers some advantages compared to 16S sequencing, including a higher strain-level resolution, and the possibility of studying microbial genes and metabolic pathways. Selected articles were published between 2017 and 2022, highlighting the current relevance of this topic. To further investigate this emerging research trend, advanced mathematical models for text analysis were computed.

Network modelling highlighted the co-occurrence of several terms in these scientific publications: dietary intake, physical activity, and taxonomic and functional profiles of the microbiota. In this regard, previous authors suggested that training modulates microbial diversity and microbial metabolism^[4,5]. Moreover, a high degree of compliance with dietary recommendations for athletes greatly affects gut microbiota functionality and sports performance^[5,6]. Interestingly, this mathematical model suggests the association of specific taxonomic clades with sports performance indicators such as endurance in professional athletes. It has been reported that athletes show increased microbiome diversity estimators like α-diversity as well as microbial pathways and fecal metabolite production compared to controls^[2,3,14].

The importance of each term represented in the co-occurrence network was calculated considering the number of other terms with which it appears. Some relevant terms highlight the role of physical activity on the health status, taxonomic and functional profiles of the gut microbiota. It should be noted that the number of microbiome studies dealing with sports performance of elite athletes based on parameters like endurance has incremented in the last five years. In contrast, studies published before tend to focus on the relationship between gut microbes and general dietary and lifestyle factors in professional athletes.

To assess the potential impact of co-variables in the network analysis, Spearman correlation coefficients between terms were calculated using caret package^[30]. Then, terms showing correlation coefficients greater than 0.6 were determined. These terms were “intestinal microbiota”, “metabolic” and “specific”. It should be noted that these co-variables do not have a great impact on keyword relationships established by the co-occurrence network. On the other hand, synonyms and terms describing very similar concepts were excluded from the network to avoid bias.

The association of gut microbes with sports performance

Among the original articles included in this systematic review, nine and five microbiome studies involved shotgun metagenomics and amplicon 16S sequencing, respectively. Moreover, two studies involved both metagenomics and 16S sequencing. These next-generation sequencing experiments involve microbiota analysis of fecal samples of professional athletes and physically active individuals [Table 3]. Specifically, participant recruitment involved elite athletes from several disciplines, including: rugby players^[14], marathon runners (professional and non-professional)^[16,18,22], endurance athletes^[24,28], competitive cyclists^[19], rowers^[16,17,25], skiers^[22], football and cricket players^[27,29], and other athletes across 16 different sports^[20]. In addition, some studies characterised the microbiota of physically active senior orienteers^[21], as well as healthy but sedentary adults exposed to endurance and strength training^[23], and short-term exercise regime^[15]. It should be noted that most of these studies compared microbiota profiles of athletes and physically-active participants to sedentary controls [Table 3]. Most studies included both male and female participants, while four and one focused on male and female athletes, respectively [Table 3]. It should be noted that these authors did not determine potential taxonomic differences according to the sex of individuals. In general, the age of participants comprising young and elite athletes ranged from 18 to 40.

A total of 12 articles described the association of specific microbial clades with sports and exercise performance [Table 3]. Figure 2 provides a schematic representation of characteristic microbial clades found in the microbiota of athletes across different disciplines as well as sedentary adults exposed to exercise regime. As can be seen, Figure 2 summarizes characteristic taxonomic profiles reported in the selected studies and presented in Table 3. In this sense, Firmicutes phylum including Ruminococcaceae or Faecalibacterium was increased in endurance athletes compared to physically inactive individuals^[1,24]. Moreover, Firmicutes/Bacteroidetes ratio is correlated with obesity as well as controlling fungal occupancy in athletes^[24,31]. In this regard, bacteria from the phyla Firmicutes and Bacteroidetes represent 90% of the gut microbiota and Firmicutes/Bacteroidetes ratio has been associated with maintaining normal intestinal homeostasis. Increased or decreased Firmicutes/Bacteroidetes ratio may lead to various pathologies. Gut microbiota unbalance usually observed with obesity, including increases in the abundance of specific Firmicutes or Bacteroidetes species^[31]. On the other hand, high Akkermansia abundances were found in the microbiota of professional rugby players^[14] and competitive cyclists^[19]. Low-abundance Akkermansia species such as A. muciniphila comprise mucin-degrading microorganisms that are negatively correlated with obesity and metabolic syndrome^[19]. In contrast, Veillonella genus and Veillonella atypica might be slightly increased in marathon runners and rowers, although these increments are not statistically significant^[16,18]. It has been suggested that lactate is the primary source of energy for V. atypica. Blood lactate resulting from muscle activity is associated with fatigue, and its concentration and accumulation depend on various factors, including exercise intensity, load and density. Scheiman et al. demonstrated in a murine model that plasma lactate may reach the intestinal lumen, where it may also be metabolized by V. atypica^[18]. The relationship between physical activity levels and the prevalence of Veillonella involved in metabolic, protective, structural, and histological functions was also reported by Dorelli et al., and Manor et al.^[1,32]. The proposed mechanism of action has been proposed: Veillonella species metabolize lactate into acetate and propionate via the methylmalonyl-CoA pathway. Unlike V. atypica, many other microbes are capable of utilizing lactate through lactate dehydrogenase, but do not possess the full pathway to convert lactate into propionate. In this sense, systemic lactate resulting from muscle activity during exercise may enter the gastrointestinal lumen and is metabolized by Veillonella into propionate in the colon. Gut colonization of Veillonella promotes the Cori cycle by providing an alternative lactate-processing mechanism whereby systemic lactate is converted into SCFAs that re-enter the circulation. SCFAs are absorbed in the sigmoid and rectal region of the colon and enter circulation via the pelvic plexus, bypassing the liver and draining via the vena cava to reach the systemic circulation directly. As a consequence, microbial SCFAs derived from lactate improve athletic performance^[18].

Figure 2. Schematic representation of characteristic microbial clades and metabolic activities with a higher representation in the microbiota of athletes across different disciplines as well as sedentary adults exposed to exercise regime. *Microbial taxa showing significantly higher abundances in the microbiota of athletes or sedentary participants exposed to exercise regime compared to sedentary controls.

Similarly, Prevotella copri was increased in rowers who completed an east–west transatlantic rowing race^[17], and sedentary adults with a short-term exercise regime involving combined aerobic and resistance training^[15]. It has been reported that gene expression of metabolic pathways of P. copri involved in L-lysine is increased after ultra-endurance exercise. This essential amino acid plays a critical role in reducing muscular fatigue and contributes to muscular integrity. Furthermore, microbial derived lysine contributes to the human body protein pool^[17].

Other microbial clades that may be characteristic of the microbiota of athletes include unidentified taxa belonging to Erysipelotrichaceae in rugby players [Table 3]. These unclassified microorganisms are positively correlated to phenylacetylglutamine derived from phenylalanine. This compound is increased in athletes and associated with a lean phenotype^[14]. However, different rugby player positions require different physiological and physical demands. High abundances of several bacterial genera (Bacteroides, Eubacterium, Prevotella and Ruminococcus) were also found in competitive cyclists compared to sedentary adults [Table 3]. These genera are correlated with branched-chain amino acid and carbohydrate metabolism pathways. It has been suggested that upregulation of branched-chain amino acids biosynthesis leads to a decrease in exercise-induced muscle fatigue and promotes muscle-protein synthesis^[19]. Similarly, several clades like Dorea longicatena, Faecalibacterium prausnitzii, Roseburia hominis and unclassified species belonging to Subdoligranulum were increased in rowers throughout a race as well as physically active senior orienteers compared to the general older adult population^[17,21]. F. prausnitzii, R. hominis and Subdoligranulum are butyrate producers that contribute to reducing gut inflammation and oxidative stress. It should be noted that butyrate is the main nutrient for colonocytes that exert an anti-inflammatory activity^[21]. With regard to Dorea longicatena, this taxon is associated with insulin sensitivity^[17,33]. In addition, high abundances of Lachnospiraceae were determined in marathon runners and skiers^[22], while Coprococcus, Parasutterella genera and Ruminococcaceae family were increased in physically active individuals^[23].

Some studies reported characteristic differences in the microbiota profiles of young and adult elite athletes in order to monitor the potential of elite athletic candidates [Table 3]. In this sense, Han et al. determined the prevalence of Firmicutes, Bacteroidetes, Proteobacteria and Actinobacteria in the fecal microbiota of rowing athletes^[25]. However, unidentified species belonging to Clostridiales and Lachnospiraceae, as well as Faecalibacterium genus, showed higher abundances in metagenomes from elite adult participants than young elite athletes [Table 3]. These taxa are likely to produce short-chain fatty acids and may be associated with exercise-induced butyrate concentrations that may contribute to improving sports performance^[25]. It should be noted that Lachnospiraceae increments are strongly associated with fiber consumptions, although general dietary recommendations for athletes are characterised by low intake of non-amylaceous polysaccharides^[5,6]. In contrast, Bacteroides was enriched in the microbiota of young elite athletes [Table 3]. High abundances of oxidative stress tolerant clades may improve performance in competitive sports. The microbiota of elite athletes may also show increased proportions of microbial metabolic pathways involved in carbohydrate metabolism and multiple sugar transport systems compared to young non-elite athletes^[25].

Efforts to establish associations between individual microbial species and different types of physical activities have also been made [Table 3]. In this regard, Streptococcus suis, Clostridium bolteae, Lactobacillus phage LfeInf and Anaerostipes hadrus were associated with moderate dynamic components of exercise. It should be noted that dynamic components of exercise involve aerobic exercise, while static components of exercise are performed by increasing tension in a muscle while keeping its length constant. Therefore, sports disciplines can be classified into nine different categories according to the contributions (high, moderate or low) of both components^[34].

It has been reported that S. suis showed high proportions of metabolic pathways involved in the fermentation of sugar alcohols, while C. bolteae and A. hadrus may produce butyrate to promote gut homeostasis and anti-inflammatory effects^[20]. Other clades, including Bifidobacterium animalis, Lactobacillus acidophilus, Prevotella intermedia and butyrate-producing species such as Faecalibacterium prausnitzii, were associated with high dynamic and low static components of exercise. On the contrary, Bacteroides caccae was enriched in athletes from disciplines involving high dynamic and static components of exercise [Table 3].

Most articles selected in this systematic review do not involve intervention studies. In this regard, characteristic patterns in the fecal microbiota of athletes were elucidated at taxonomic and functional levels [Figure 2]. In addition, some studies report differences in the metabolomic profiles of participants. In this sense, differences in the composition and functional capacity of the gut microbiome of international-level athletes from several disciplines have been determined^[20]. It has been described that fecal microbiota samples of professional rugby players and sedentary controls show even greater separation at the functional metagenomic and metabolomic than at compositional levels^[14]. Similarly, microbial diversity, including butyrate-producing species, increased throughout the ultra-endurance events such as transoceanic rowing races. The functional potential of bacterial species involved in specific amino and fatty acid biosynthesis also increased^[17]. Additional studies dealing with rowing athletes report that ATP metabolism, multiple sugar transport systems and carbohydrate metabolism are enriched in the microbial community of these athletes^[25]. Differences in functional bile acid and histidine metabolism also differentiate power athletes from sedentary controls, while galactose metabolism markers differentiate endurance athletes from controls. Moreover, D-alanine and primary bile acid biosynthesis differentiate power and endurance athletes from controls. Scheiman et al. and Kostic reported that every gene in a major pathway metabolizing lactate to propionate is at higher relative abundance postexercise in marathon runners^[18,16]. On the other hand, an increment in the abundance of Prevotella was correlated with a number of amino acid and carbohydrate metabolism pathways in cyclists, including branched-chain amino acid metabolism^[19]. In contrast, some studies report characteristic microbiota patterns mainly at taxonomic level, including increased abundances of Faecalibacterium prausnitzii in physically active senior orienteers^[21].

Some studies report characteristic microbial-derived metabolic profiles of athletes from different disciplines. In this regard, cis-aconitate, succinic acid and lactate in urine samples and creatinine in faeces were found to be significantly different between groups of sports. These differences were evident despite the absence of significant differences in diet^[20]. In addition, microbial-derived SCFAs are enhanced within rugby players and their gut microbiota was predominantly correlated with creatine kinase (CK), total bilirubin and total energy intake^[14].

Dietary factors that may confound the association between exercise and the abundance of microbial taxa of interest were considered in several of these studies^[16,18]. Other studies report the assessment of macronutrient intake per day to enhance inter-individual comparison^[21] as well as quantitative diet evaluation^[22]. It has been reported that the versatility of the microbial community of athletes, which might affect their performance, is associated with dietary factors^[25]. However, other studies report the lack of in-depth dietary analysis and a matching non-athlete cohort (i.e., control group comprising non-athlete participants with similar demographic characteristics and exposed to the same intervention)^[19].

On the contrary, three of the selected articles comprised intervention studies. In this regard, Cronin et al. demonstrated modest changes in gut microbial composition and function in healthy but sedentary adults exposed to short-term exercise regime, with and without concurrent daily whey protein consumption^[15]. An association between whey protein intake and the β-diversity of the adult gut virome was found, while no major changes in the functional activity of the gut microbiota were determined, with the exception of urinary levels of trimethylamine N-oxide (TMAO) and phenylacetylglycine (PAG) excretion. Similarly, Moitinho-Silva et al. elucidated taxonomic differences, a significant increase in lymphocytes and a decrease in mean corpuscular haemoglobin concentration in healthy sedentary adults exposed to strength training^[23]. No change in dietary patterns towards the end of the intervention period and after it was observed. Finally, Fukuchi et al. investigated the effects of both official competition and a multi-strain lactic acid bacteria-fermented soymilk extract in taxonomy and urine metabolites in endurance athletes^[24]. Changes in urinary metabolites included a significant reduction in yeast and fungal markers, neurotransmitters, and mitochondrial metabolites including the tricarboxylic acid (TCA) cycle. Tricarballylic acid was positively correlated with the ratio of Firmicutes, while Parabacteroides distasonis was negatively correlated with several urinary metabolites, including 3-metylglutaconic, vanillylmandelic, quinolinic and kynurenic acids and thymine^[24].

Applications of next-generating sequencing technologies to assess the relationships between gut microbiota composition and sports performance have been summarised. Computational methods for text processing and bibliographic analysis revealed recent research trends focusing on the impact of microbial diversity and metabolic pathways on different components of physical activity. These studies highlight the role of butyrate-producing bacteria in promoting gut health and contributing to an optimal composition of the athletic microbiota.

Future trends

Future research trends may focus on the interactions of diet and gut microbial composition in the context of athletic performance as well as potential applications of pro- pre- and symbiotic supplements specifically developed for athletes. For this purpose, more studies dealing with metabolic activities and other mechanisms of action of specific clades associated with several indicators of sports performance are needed. In this regard, metagenomic sequencing enables the identification of microbial metabolic pathways showing higher abundances in the gut microbiota of athletes from several disciplines than in sedentary controls. This fact may contribute to a better understanding of the role of gut commensals in athlete physiology. In addition, further studies including a multi-omics approach (i.e. the combined analysis of metabolomic and metagenomic data) are needed to better elucidate the associations between physical activity, dietary patterns and gut microbiota. It should be noted that most articles published up to date comprise observational studies and the mechanisms of action of specific microbial clades contributing to sports performance remain unknown. In this regard, genomic sequences recovered from athlete metagenomes could be of great benefit to perform metabolic modelling in order to characterize those activities that may have a physiological impact. The integration of these techniques may lead to personalized nutrition interventions including specific nutrient recommendations aimed at improving performance by promoting certain taxa and enhancing certain metabolites of interest during exercise. Finally, more randomized controlled studies analyzing wider cohorts and controlling for potential confounding factors are needed to identify microbial species that could be used as biomarkers of sports performance.

Conclusions

This systematic review gives a general overview of the applications of next-generation sequencing, including shotgun metagenomics and 16S amplicon sequencing, to study the athletic gut microbiota. Bibliographic analysis using computational models revealed the growing research interest in elucidating the associations between specific microbial clades and sports performance. In this sense, butyrate producers such as A. hadrus, C. bolteae, F. prausnitzii, R. hominis and unidentified species belonging to Clostridiales, Lachnospiraceae and Subdoligranulum may contribute to maintaining gut homeostasis in athletes. On the other hand, species involved in branched amino acid and lactate metabolism, such as P. copri and V. atypica, may reduce muscular fatigue. In addition, athletic gut microbiota shows an increase in microbial metabolic pathways involved in carbohydrate metabolism. Future research will aim at developing personalised nutrition strategies and recommendations to optimise microbiota composition profiles associated with different components of exercise (high, moderate and low static and dynamic components). Future research trends may focus on the interactions of diet and gut microbial composition in the context of athletic performance. For this purpose, multi-omic studies dealing with metabolic activities and other mechanisms of action of specific clades associated with several indicators of sports performance are needed. As a consequence, these microorganisms could be used as biomarkers of sports performance to assess the potential of athletic candidates.