BACKGROUND The microbiome-gut-brain axis, especially the microbial tryptophan biosynthesis and metabolism pathway (MiTBamp), is closely connected to bipolar disorder with current major depressive episode (BPD). METHODS We performed shotgun metagenomics sequencing (SMS) of faecal samples from 25 BPD patients and 28 healthy controls (HCs). Except for the microbiota taxa and MiTBamp analyses, we also built a classification model using the Random Forests (RF) and Boruta algorithm to find the microbial biomarkers for BPD. RESULTS Compared to HCs, the phylum Bacteroidetes abundance was significantly reduced, whereas that of the Actinobacteria and Firmicutes were significantly increased in BPD patients. We also identified 38 species increased and 6 species decreased significantly in the BPD group. In the MiTBamp, we identified that two Kyoto Encyclopedia of Genes and Genomes (KEGG) orthologies (KOs) (K00658 and K00837) were significantly lower in the BPD, and five KOs (K01696, K00382, K00626, K01667, and K03781) were significantly higher in the BPD group. We also identified significant genera and species which were closely related to these KOs. Finally, RF classification based on gut microbiota at the genus level can achieve an area under the receiver operating characteristic curve of 0.997. LIMITATIONS The features of cross-sectional design, limited sample size, the heterogeneity of bipolar disorders, and a lack of serum/plasma tryptophan concentration measurements. CONCLUSIONS The present findings enable a better understanding of changes in gastrointestinal microbiome and MiTBamp in BPD. Alterations of microbes may have potential as biomarkers for distinguishing the BPD patients form HCs.
Wen-tao Lai, Jie Zhao, Shu-xian Xu
Journal of affective disorders