Abstract
Evidence of preclinical interactions between the gut microbiome and brain health is accumulating. Studies of animal models and specific patient populations have suggested a relationship between gut microbiomes and brain volumes, but this association is understudied in apparently healthy humans. We conducted a population-based cross-sectional study of 623 Japanese men from the Shiga Epidemiological Study on Subclinical Atherosclerosis (SESSA). We performed 16S ribosomal RNA gene sequencing of stool samples collected during the follow-up stage (mean [SD] age, 68.0 [8.0] years; range, 46-83 years). All participants underwent brain magnetic resonance imaging and automated voxel-based morphometry. Principal coordinate analysis, linear discriminant, and multivariable linear regression analyses were performed. In multivariable linear regression analysis, after adjusting for age and total intracranial volume, only gray matter volume showed a positive association with alpha diversity (the Shannon index richness, q-value <0.01). However, no association was found after further adjustments for body mass index, physical activity, smoking, drinking, and hypertension. The weighted UniFrac distances (beta diversity) measured using principal coordinate analysis showed that lower and higher white and gray matter volumes formed distinct clusters (q < 0.01). In linear discriminant analysis and multivariable-adjusted linear regression analysis, several genera were significantly associated with gray and white matter volumes (q < 0.01); however, Lachnospiraceae, a butyrate-producing bacterium, was consistently related to a higher white matter volume in different statistical analysis models. Egarthellaceae, Bifidobacteraceae, and Selenomonadaceae showed a positive association with greater gray matter volume. Our findings support an association between gut microbiome diversity and brain volumes in middle-aged to older Japanese men. This study provides insight into the underlying effects of the gut microbiome on human brain volume.