Exploring gut microbiota profile induced by antipsychotics in schizophrenic patients: insights from an Eastern European pilot study
摘要
Schizophrenia (SCZ) is an intricate multi-systemic mental illness which has been linked to alterations of the gastrointestinal (GI) microbiota via the bidirectional network known as the gut-brain axis (GBA). However, the specific microbial signatures characterizing SCZ remain unclear.
MethodsWe analyzed the gut microbiome composition of 57 Romanian adults, 30 SCZ, and 27 healthy controls (HCs), using amplicon sequencing on the V3-V4 region of the 16S rRNA gene isolated from stool samples. DNA was extracted with an optimized enzymatic and mechanical lysis protocol. Sequence profiling was performed using DADA2. Counts were variance‑stabilized (DESeq2 VST) for Bray–Curtis and weighted UniFrac distance calculation, principal coordinate analysis (PCoA), and permutational multivariate analysis of variance (PERMANOVA). To account for data compositionality, counts were also centered log‑ratio (CLR)–transformed for principal component analysis (PCA) and supervised sparse partial least square linear discriminant analysis (sPLS‑DA). Differential abundance was tested using Wilcoxon rank‑sum (false discovery rate - FDR corrected) and linear discriminant analysis (LDA) effect size (LEfSe) (LDA > 4, FDR < 0.01).
ResultsSCZ patients exhibited decreased Bifidobacterium (p = 0.0187), Blautia (p < 0.001), and Eubacterium (p = 0.00120) abundances compared to HCs. Additionally, SCZ patients showed higher Faith’s phylogenetic diversity both before (p = 0.049) and after rarefaction (p = 0.0022), while Shannon and Simpson indices were not significantly different between the two study groups. PCoA on Bray-Curtis and weighted UniFrac distances revealed clear SCZ-HC separation, with unadjusted PERMANOVA attributing 13–15% of variance to SCZ status alone (p = 0.001). After adjusting for smoking, diet, lifestyle, metabolic and comorbidity covariates, SCZ status remained the only significant term which contributed to variance in microbiome composition (R2 ~ 4.4%, p = 0.001). Within the SCZ group, only risperidone use (not age, sex, body mass index - BMI, glycaemia, lipids, or other medication use) was significantly associated with community structure (p < 0.01). PCA and sPLS-DA revealed that the principal drivers for group separation after taking into account the compositional nature of data were Erysipelotrichaceae UCG-003 and Anaerostipes in HCs and Holdemanella in SCZ (all p < 0.0001).
ConclusionsOur integrative analyses provide evidence of gut microbiome alterations in SCZ and reveal that risperidone use modulates community structure, underscoring the need to consider the GI effects of antipsychotics when choosing appropriate regimens for patients, while at the same time highlight the GBA as a promising target for future therapeutic strategies.