Immune cells mediated the causal relationship between the gut microbiota and anxiety disorders: A Mendelian randomization study

Background: Studies have demonstrated that the gut microbiome-immune system-brain axis plays an important role in neurological disorders. Furthermore, recent studies have shown that the gut microbiota influences the occurrence and progression of anxiety disorders, with potential involvement of immune cells. We aimed to investigate the causal impact of gut microbiota on anxiety disorders and identify potential immune cell mediators. Methods: We made use of the summary statistics of 196 gut microbiota (MiBioGen consortium), 731 immune cells, and anxiety disorders (Medical Research Council Integrative Epidemiology Unit consortium), from the extensive genome-wide association studies to date. To determine the causal links between gut microbiota and anxiety disorders, we employed bidirectional Mendelian randomization (MR) analyses, and further employed 2-step MR to confirm potential mediating roles of immune cells. Moreover, we conducted rigorous sensitivity analyses to assess the heterogeneity, robustness, and horizontal pleiotropy of our findings. Results: Bi-directional MR analysis revealed that 11 gut microbiota species can affect anxiety disorders, while the reversed causal relationship was not existed. Mediation analysis revealed that three immune cells mediated the causal relationships between two gut microbiota species and anxiety disorders. Specifically, "CD39+ resting Treg %resting Treg", "CD39+ resting Treg % CD4 Treg", and "BAFF-R on IgD+ CD38- naive B cell" mediated the effects of class Melainabacteria on anxiety disorders, with mediating impacts of 0.000075, 0.000096, and 0.000263, representing 5.98 %, 7.67 %, and 21.01 % of the total effects, respectively. Additionally, "BAFF-R on IgD+ CD38- naive B cell" also mediated the effects of order Gastranaerophilales on anxiety disorders, with a mediating impact of 0.000266, accounting for 19.06 % of the total effects. Limitations: The bacterial analysis was limited to genus level, overlooking species or strains. We used a lenient p-value threshold of p < 1.0 x 10(-5) for instrumental variables, instead of the typical p < 5 x 10(-8). Lastly, the GWAS focused on European participants, potentially limiting the generalizability of our findings to other ethnicities. Conclusion: The risk of anxiety disorders has been linked causally to gut microbiota, with three distinct immunophenotypes acting as potential mediators in this relationship. The role of gut microbiota in modulating immune cells, thereby influencing anxiety disorders, may offer new therapeutic strategies and management approaches.