Association of leisure time physical activity with gut microbiota composition in early adulthood

Abstract

Physical activity is associated with compositional changes in gut microbiota. Studies applying data from general populations exhibiting varying physical activity levels are scarce. This study investigated the associations between leisure time physical activity and gut microbiota composition in a general population of young adults. Faecal samples were analyzed with 16 S rRNA sequencing and leisure time physical activity were assessed in N = 302 (176 females, 126 males) 26-year-old individuals of European descent. Leisure time physical activity was assessed using a self-administrated questionnaire and expressed as metabolic equivalent (MET) h/wk. Participants were defined as inactive (N = 83) and active (N = 219) based on estimation to meet the physical activity guidelines. Diet was assessed using food diaries. First, associations of physical activity and gut microbiota were investigated and then the models were further adjusted for diet (N = 280). While gut microbiota diversity was similar, community composition differed between the active and inactive participants. The association diluted after adjustment for diet. Moreover, 16 genera, e.g., Lactobacillus, Veillonella, and Prevotella were more abundant in the active compared with the inactive participants. After adjustment for diet, these associations, except Barnesiella, were no longer detected, but other genera were found. In conclusion, physical activity in young adults is associated with gut microbiota community composition and an increase in the abundance of genera considered beneficial for overall health. The associations are partly confounded by diet.

Keywords: Cohort study; Diet; Gut microbiota; Health; Physical activity.

Fig. 1

Shannon diversity in the groups (blue = active, pink = inactive) based on Shannon index (active: mean = 3.17; inactive, mean = 3.37).

Fig. 2

(A) Beta diversity at the ASV-level amongst the physical activity groups based on Bray-Curtis dissimilarities (blue dots: active group, pink dots: inactive group). (B) The most dominant genus for each participant. The colours indicate the group each individual belongs to as detailed in the legend.

Fig. 3

Highest absolute coefficients of the linear model in Permanova. This describes the top genera driving the difference in Bray-Curtis dissimilarity between the physical activity groups (< 0 = inactive, > 0 = active group).

Fig. 4

Relative abundance of those genera that were identified as differentially abundant in DESeq2 analysis between active and inactive groups before adding diet to the model (black dots: abundance > 0, red dots: abundance = 0). The adjusted p values (adj.p) are obtained from the DESeq2 analysis.

Fig. 5

Relative abundance of those genera that were identified as differentially abundant in active and inactive groups after diet was added to the model (black dots: abundance > 0, red dots: abundance = 0). ). The adjusted p values (adj.p) are obtained from the DESeq2 analysis.