A Pilot Longitudinal 16S rRNA Gene Sequencing Study Exploring the Relationship Between Gut Microbiota and Body Composition in Healthy Adults
Min-Yen Shih, Yu-Chen Yang, Yun-Ru LiuAbstract
The gut microbiome is linked to body composition, yet most studies involve probiotic or dietary interventions. This study explored relationships between changes in body composition and the fecal microbiota under natural lifestyle conditions. A repeated-measures design involved 15 adults completing four body composition assessments at 3-month intervals. Fecal samples from each time point underwent 16S rRNA gene sequencing. Participants were stratified by body composition parameters, and microbial profiles from initial and final measurements were compared to assess longitudinal patterns. Overweight participants showed lower alpha diversity. Linear mixed models revealed fecal microbiota remained stable across all four time points, with no statistically significant continuous trends observed longitudinally. Exploratory baseline-to-endpoint comparisons across stratified groups and Spearman correlation analyses suggested potential microbiota shifts, though these associations remained statistically non-significant. Preliminary observations exhibited that the OTU identified as Parasutterella excrementihominis tended to associate with higher body fat, whereas the putative species Akkermansia muciniphila showed a potential inverse association. Representative taxa, such as Dialister invisus , appeared enriched in individuals with higher skeletal muscle percentages, whereas the OTU assigned to Bifidobacterium pseudocatenulatum showed the opposite trend. Several associations differed by sex, suggesting modulation by host factors. These preliminary findings suggest possible fecal microbiota patterns associated with body composition, even without targeted interventions. While lacking robust linear associations in this small pilot cohort, the observed directional consistency across statistical approaches highlights the potential of fecal microbes as candidate indicators of metabolic health. These exploratory results require further validation in larger, longitudinal studies with sufficient statistical power.