The combination of sport and sport-specific diet is associated with characteristics of gut microbiota: an observational study

Figures & data

Table 1 Subject characteristics

Table 2 Composition of the macronutrients and dietary fiber intake of controls, bodybuilders, and distance runners obtained from 3-d food diaries

Fig. 1 Comparison of the percentage of energy from each macronutrient category. The percentages were calculated by dividing the available energy from the macronutrients by the total energy. a The percentage of the energy obtained from carbohydrates; b The percentage of the energy obtained from protein; c The percentage of the energy obtained from total fat. The red line represents acceptable macronutrient distribution ranges (AMDR). The AMDR for carbohydrates is 55 to 65%, for total fat is 15 to 30%, and for protein is 7 to 20% of the energy intake for Korean adults. Controls versus bodybuilders: ⁺ p < 0.05 or ⁺⁺⁺ p < 0.001, bodybuilders versus distance runners: ^** p < 0.01 or ^*** p < 0.001

Fig. 2 Certain types of exercise training and athlete diet affected the relative abundance of some microorganisms. a Comparison of gut microbiota relative abundance at the genus level in the three groups. Relative abundance represent log of percentage as whole microbiota. For example, when the relative abundance of a particular genus is 1, it means 10% of whole microbiota, and when it is 0, it means 1% and when it is −1, it means 0.1%. Controls versus bodybuilders: ⁺ p < 0.05 or ⁺⁺ p < 0.01, bodybuilders versus distance runners: ^* p < 0.05 or ^** p < 0.01, distance runners versus controls: ^¥ p < 0.05 or ^¥¥ p < 0.01. b Total fat intake negatively correlated with relative abundance of Bifidobacterium in bodybuilders (Correlation coefficient: −0.52, p-value: 0.048). c Total fat intake positively correlated with relative abundance of Sutterella in bodybuilders (Correlation coefficient: 0.58, p-value: 0.023)

Table 3 Gut microbiota composition at species level

Fig. 3 There was no difference in the gut microbiota diversity between the controls, bodybuilders, and distance runners. a Comparison of observed species of controls, bodybuilders, and distance runners obtained from 30,000 sequences per sample. b Estimation of the abundance of unique operational taxonomic units (OTUs) using Chao1. Phylogenetic diversity was estimated using the average values for the Chao1 plot of the gut microbiota in the controls, bodybuilders, and distance runners. Data are based on 30,000 sequences per sample from the study subjects

Fig. 4 Protein intake negatively correlated with alpha diversity in distance runners. a Protein intake negatively correlated with OTUs (Correlation coefficient: − 0.53, p value: 0.04). b Protein intake negatively correlated with ACE (Correlation coefficient: − 0.51, p-value: 0.05). c Protein intake negatively correlated with Shannon index (Correlation coefficient: − 0.63, p-value: 0.012)

Fig. 5 The plot was generated using a generalized UniFrac principal coordinate analysis (PCoA) of fecal microbiota from 45 subjects. Generalized UniFrac PCoA analyzed genus rank level and included unclassified OTUs. Subject color coding: green, controls; blue, bodybuilders; and yellow, distance runners