Comparative analysis of gut microbiota in major depressive disorder and schizophrenia during hospitalisation - the case-control, post hoc study

Abstract

The aim of this study was to investigate the relationship between gut microbiota and major depressive disorder (MDD) and schizophrenia (SCZ) by comparing 36 inpatients with these conditions to 29 healthy controls (HC) matched for age, sex, and body mass index (BMI). Individuals with SCZ exhibited greater microbiota richness compared to HC (FDR P(Q)=0.028). Taxonomically, while no significant differences were observed between the microbiota of MDD and SCZ patients in a head-to-head comparison, both patient groups differed significantly when compared to HC. Interestingly, besides common patterns (such as a higher abundance of Erysipelotrichaceae UCG-003 and Streptococcus, and a lower abundance of Lachnospiraceae ND3007 group), unique patterns were exhibited only in MDD (with a higher abundance of Anaerostipes, Q=0.004) or SCZ (with a higher abundance of Sutterella, Q=0.001, and a lower abundance of Clostridium sensu stricto 1, Q=0.002). The Random Forest algorithm identified Ruminococcus torques group, Lachnospiraceae UCG-001, and Erysipelotrichaceae UCG-003 as highly discriminative features for both SCZ and MDD, while Suturella and Holdemania were unique features for SZC, and Lachnospiraceae genus CAG-56 and Anaerostipes for MDD. Additionally, between 50 % and 60 % of the differentially abundant taxa were found among the top 10 influential features in the RF models. In conclusion, while no significant differences were found between the microbiota of MDD and SCZ patients, distinct microbial patterns were found in each group when compared to HC. The study did not confirm universal microbial biomarkers reported in other studies but showed that the observed differences concern the bacteria associated with inflammation, the production of short chain fatty acids (SCFA), and the synthesis of metabolites linked to mental health (lactic acid, gamma-aminobutyric acid - GABA). The application of machine learning holds promise for further understanding the complex relationship between microbiota and these psychiatric disorders. The observed results should be treated with caution due to the limitations of this study (mainly sample size), therefore further researches under standardized environmental conditions with consistent analytical and bioinformatics approaches are warranted.