Associations of the Gut Microbiome with Neuroimaging Markers of Glymphatic Dysfunction and Cognitive Functions in Healthy Middle-Aged Chinese

Abstract

Introduction: Whilst gut microbiome (GM) dysregulation and glymphatic dysfunction have been identified as emerging risk factors of dementia in symptomatic individuals, it is unknown whether GM is associated with the glymphatic system, especially in the pre-symptomatic stage of dementia. We therefore aimed to identify the associations of GM with neuroimaging markers of glymphatic function, [diffusion tensor imaging-along the perivascular vascular space (DTI-ALPS)] and cognitive functions in a cohort of neurologically healthy middle-aged individuals. Methods: We recruited middle-aged Hong Kong Chinese without known cognitive impairment or symptomatic vascular disorders. We studied the associations of their GM characterized by shotgun metagenomic sequencing with cerebral DTI-ALPS index and fractional anisotropy (FA) as determined by brain magnetic resonance imaging. Cognitive performances were assessed by Victoria Stroop Test and Hong Kong Montreal Cognitive Assessment (HK-MoCA). Permutational multivariate analysis of covariance based on Bray-Curtis Dissimilarity was performed to calculate the β-diversity of the samples. Shannon alpha diversity was examined to measure the GM diversity within the subjects and partial least squares-discriminatory analysis was performed to discriminate characteristics of species-level microbiota with DTI-ALPS index and cognitive assessments. Data were analyzed under the full model of covariate adjustment including age, sex, body mass index, educational years, hypertension and smoking status. Results: In 241 Hong Kong Chinese (mean age 54±6 years, 47% men; HK-MoCA M= 27.9±2.25; Stroop Interference M= 11.69±7.37), we identified significant positive correlations between Firmicutes/Bacteroidetes ratio and the DTI-ALPS index (p<0.05). Species-level GM analyses indicated that Sanguibacter SGB15121 and Dorea sp AF24 7LB were positively associated with the DTI-ALPS index (p<0.01). Additionally, Roseburia intestinalis was positively associated with Stroop interference (p<0.01) while the decreased relative abundance in Barnesiella intestinihominis and Blautia luti were associated with increased Stroop interference (p<0.05). Although DTI-ALPS index was not significantly correlated with markers of cognition, we found a significant negative association between cerebral FA and Stroop interference (p<0.05). Discussion: In healthy middle-aged Chinese without dementia, GM dysregulation is significantly associated with neuroimaging surrogate markers of glymphatic dysfunction, suggesting a potential role of the GM in the subsequent development of dementia in asymptomatic individuals.