Distinctive subgingival microbial signatures in older adults with different levels of cognitive function.

Abstract

To examine association between subgingival microbial signatures and levels of cognitive impairment in older adults. We analysed subgingival plaque samples and 16S ribosomal RNA sequences for microbiota among 165 participants (normal controls [NCs]: 40, subjective cognitive decline [SCD]: 40, mild cognitive impairment [MCI]: 49 and dementia: 36). The bacterial richness was lower among individuals with worse cognitive function, and subgingival microbial communities differed significantly among the four groups. Declining cognitive function was associated with decreasing relative abundance of genera Capnocytophaga, Saccharibacteria_genera_incertae_sedis, Lautropia and Granulicatella, and increasing abundance of genus Porphyromonas. Moreover, there were differentially abundant genera among the groups. Random forest model based on subgingival microbiota could distinguish between cognitive impairment and NC (AUC = 0.933, 95% confidence interval 0.873-0.992). Significant correlations were observed between oral microbiota and sex, Montreal Cognitive Assessment (MoCA) score and Mini-Mental State Examination score. Partial correlation analysis showed that Leptotrichia and Burkholderia were closely negatively associated with the MoCA score after adjusting for multiple covariates. Gene function was not significantly different between SCD and NC groups, whereas three homozygous genes were altered in MCI patients and two in dementia patients. This is the first study to demonstrate an association between the composition, function and metabolic pathways of subgingival microbiota and different levels of cognitive function among older individuals. Future cohort studies should assess its diagnostic usefulness for cognitive impairment.