Item

Abstract

Anxiety disorders have a significant impact on global population health and have been found to influence the composition of the microbiome. However, the specific taxonomic and functional features of the microbiome associated with anxiety scores vary across studies, potentially due to limitations in taxonomic resolution, methodological differences, and confounding factors. We aimed to identify universal and cohort-specific species/strain-level taxonomic and functional features of the gut microbiome linked to anxiety. To achieve this, we collected stool samples from two large cohorts: Tulsa1000 (n=270, USA) and TwinsUK (n=359, United Kingdom), along with clinical information. The microbiome profiles were analyzed using shotgun metagenomics, and the pool of metabolites in the blood was additionally assessed through untargeted LC/MS metabolomics. We employed a compositionality-aware Nearest Balance method to examine the associations between taxonomic composition with anxiety and other factors. Our results revealed consistent taxonomic enrichments in high- or low-anxiety individuals across the cohorts, using both k-mer based metagenome profiling and metagenome-assembled genomes (MAGs). Among the highlights, several members of Enterocloster genus were positively linked with anxiety. Notably, enrichment often varied within genera and species, emphasizing the importance of fine taxonomic resolution to identify true taxon-anxiety associations. We further refined the correlates of anxiety by examining balances of microbial species. While we observed a significant association between alpha diversity and anxiety, the effect size was modest. Additionally, specific genomic features were significantly linked to anxiety, indicating that the microbiome of individuals with high anxiety may be conducive to pathogen growth and influenced by inflammation. In terms of functional alterations, we found that the gut microbiome in individuals with anxiety demonstrated an increased propensity for tryptophan synthesis, along with other aromatic amino acids. From a community structure standpoint, this manifested as a reduction in taxa predicted auxotrophic for tryptophan synthesis. However, our comparison with serum levels of tryptophan and related compounds did not reveal direct associations with the microbiome composition or anxiety. In conclusion, our study revealed taxonomic features consistently linked to anxiety across diverse cohorts, highlighting the significance of specific taxa that necessitate targeted experimental validation to elucidate their causal roles. Differences related to tryptophan metabolism suggest possible alterations of the pool of neuroactive compounds and their precursors in the gut (including kynurenine and indole). Further exploration of these mechanisms will enhance our understanding of how the microbiome contributes to the pathogenesis of anxiety and compensatory processes.