Faecal microbiota composition and impulsivity in a cohort of older adults with metabolic syndrome

Abstract

Impulsivity is an important determinant of human behaviour, affecting self-control, reasonable thinking and food choices. Recent evidence suggests a role for gut microbiota in human behaviour, but the relationship between gut microbiota and impulsive behaviours remains largely unexplored. To address this knowledge gap, the present study aims to explore the associations between faecal microbiota composition with trait and behavioural impulsivity, in a subcohort of the PREDIMED-Plus trial, including older adults presenting overweight/obesity. Fecal samples (n = 231) were profiled for their microbiota composition using 16 S rRNA amplicon sequencing and impulsivity was determined through four different assessments. Adherence to different dietary patterns was estimated through questionnaires. Beta diversity analyses showed a significant association with the Conner's Performance Test (CPT) in multivariate-adjusted models, and, in total, 13 bacterial genera associated with CPT. Erysipelotrichaceae UCG 003 showed the highest association with CPT and known butyrate producers such as Butyricicoccus spp., Roseburia spp., and Eubacterium hallii were among the identified bacteria. The bacteria Lachnospiraceae UCG 001, Anaerostipes and Blautia were associated with CPT and also the adherence to healthy and unhealthy plant-based diets. In addition, functional analysis showed a significant negative association between the CPT and the glucuronate and galacturonate metabolic pathways. From the other impulsivity assessments, two more associations were identified, for the genus Phascolarctobacterium with the Stroop test, and the genus Lachnospiraceae GAG 54 with the positive urgency subscore of UPPS-P Impulsive Behaviour Scale. Overall, our findings suggest potential links between the faecal microbiota composition and function with behavioural impulsive inattention as determined by the CPT.

Fig. 1

Bacterial genera significantly associated with Conner’s Performance Test (CPT) commissions as determined with linear mixed models, adjusting for age (in years), sex, BMI (kg/m²), type 2 diabetes prevalence (no/yes), and recruitment centre as a random effect. The degree and direction of association of a specific taxon is indicated through the log2FoldChange, which represents the bias-corrected coefficients.

Fig. 2

Associations between bacterial genera with (A) healthy plant-based diet index (hBPD) and (B) unhealthy plant-based diet index (unhPBD), and (C) Mediterranean diet, as determined with linear mixed models, adjusting for age (in years), sex, BMI (kg/m²), type 2 diabetes prevalence (no/yes), and recruitment centre as s random effect. The degree and direction of association in the abundance of a specific taxon is indicated through the log2FoldChange, which represents the bias-corrected coefficients.

Fig. 3

Associations between microbial pathway abundances with Conner’s Performance Test (CPT) commissions, as determined with linear mixed models, adjusting for age (in years), sex, BMI (kg/m²), type 2 diabetes prevalence (no/yes), and recruitment centre as s random effect. The degree and direction of association in the abundance of a specific microbial pathway is indicated through the log2FoldChange, which represents the bias-corrected coefficients.