Abstract
BACKGROUND: The gut microbiota may influence cognitive function via the gut-brain axis. This study aimed to investigate the gut microbiota profiles of 346 older Korean individuals with subjective cognitive decline but no symptoms (SCD), mild cognitive impairment (MCI), or Alzheimer’s disease (AD). METHODS: Participants aged an average of 72.3 years underwent the profiling of cognitive function, amyloid-β (Aβ) deposition, apolipoprotein E (APOE) genetic variants, depression status, nutrition, and lifestyles. Human fecal bacterial FASTA/Q data (SCD, n = 24; MCI, n = 246; AD, n = 76) were processed using Quantitative Insights Into Microbial Ecology 2 (QIIME2) tools. Operational taxonomic units (OTUs) and their counts were assigned with the National Center for Biotechnology Information Basic Local Alignment Search Tool (BLAST). Machine learning models (random forest and XGBoost) identified key bacterial taxa differentiating groups. RESULTS: Redundancy analysis revealed associations between gut microbiota composition and cognitive function, age, gender, nutritional status, and body mass index. All three groups shared 71 common bacterial genera with distinct taxonomic profiles across cognitive states. The AD group uniquely harbored Hominisplanchenecus and Lentihominibacter, while the SCD group exclusively contained Anaerosacchariphilus and Anaerobutyricum. Phascolarctobacterium was shared between the AD and MCI groups, and Anaerostipes between the MCI and SCD groups. The SCD group showed significantly elevated Bifidobacterium catenulatum, Anaerobutyricum hallii, and Anaerostipes hadrus. Network analysis demonstrated greater microbial community complexity in the SCD group compared to the MCI and AD groups. Gut bacteria correlated with depression, Aβ deposition, APOE status, and cognitive scores. CONCLUSIONS: This study identified distinct gut microbiota profiles associated with different stages of cognitive impairment in older Korean adults. The observed associations between gut bacterial composition and cognitive function, neurodegeneration biomarkers, and related clinical factors suggest potential relationships that warrant further investigation. These findings contribute to the growing understanding of gut-brain interactions in cognitive aging. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s13195-025-01820-9.