Abstract
Recent research on the gut-brain axis (GBA) indicates that the gut microbiome can significantly influence brain structural and functional connectivity. However, the specific causal relationships between the gut microbiome and brain imaging-derived phenotypes (IDPs) of functional/structural connectivity, as well as how the gut microbiota influences mood and cognition, remain unclear. This study utilizes data from large-scale genome-wide association studies (GWAS) and employs a bidirectional Mendelian randomization (MR) approach to evaluate the causal effects between the gut microbiome and brain connectivity. We obtained data on 196 gut microbiome taxa from the MiBioGen consortium and acquired IDPs for seven resting-state networks (RSNs) from the UK Biobank (UKB). Subsequently, we conducted bidirectional MR analyses to explore their causal relationships. In the forward MR analysis, Ruminococcus torques, Eubacterium fissicatena, and Coprobacter exerted positive effects on the default mode network (DMN), whereas Terrisporobacter influenced the structural connectivity of the dorsal attention network (DAN). Conversely, Gammaproteobacteria inhibited the functional connectivity of the ventral attention network (VAN). Additionally, reverse MR analysis revealed that increased functional connectivity of the DAN was positively associated with the abundance of Alloprevotella. The enhanced functional connectivity of the VAN negatively impacted Alloprevotella, Catenibacterium, and Methanobacteria. Furthermore, the structural connectivity of the frontoparietal network (FPN) and somatomotor network (SMN) significantly reduced the abundance of Bacilli and Intestinibacter, respectively. This study utilized a bidirectional MR approach to establish causal evidence for the relationship between the gut microbiome and brain network connectivity, and support the bidirectional regulatory pattern of the GBA. These findings provide new insights into the potential roles of gut microbiota in emotional regulation, cognitive function, and neurodevelopment, and offer a theoretical basis for microbiome-based intervention strategies.