Microbiota-metabolome interplay in depression: Metabolic insights and diagnostic potential.

Significant disruptions were identified in 53 gut microbial species, 12 microbiota-related metabolic pathways, and 34 metabolites in depressive individuals compared to controls.

• Analysis was performed on fecal and serum metabolomes in first-episode depression and matched controls (n = 186).
• Findings were validated in three independent cohorts (n = 223, 85, 52) including a drug intervention cohort.
• Both fecal and serum metabolomes were analyzed to characterize metabolic disruptions.

Sixteen metabolites exhibited reversal after drug administration in depressed patients.

• Drug intervention validation was performed in one of the three independent cohorts (n = 52).
• These 16 metabolites were among the 34 metabolites identified as significantly disrupted in depression.
• Reversal after drug administration suggests these metabolites may reflect treatment-responsive biological changes in depression.

Partial Spearman analysis identified 271 species-metabolite correlations in depressive individuals.

• Partial Spearman correlation analysis was used to control for potential confounders.
• These correlations link specific gut microbial species to metabolic changes associated with depression.
• Mediation analysis further unveiled 61 metabolite-mediated species-depression correlations.

Key features associated with depression included Bifidobacterium longum, Parasutterella excrementihominis, tyrosine, serotonin, and homovanillic acid.

• Both microbial species and metabolites were highlighted as key features.
• Serotonin and homovanillic acid are neurotransmitter-related metabolites, implicating monoaminergic pathways.
• Tyrosine is a precursor to catecholamines and was among the highlighted metabolites.
• These features were identified through integration of microbiota and metabolomics data.

A machine learning model using 34 metabolites achieved AUC values of 0.82 and 0.80 in discriminating depression from controls in test and validation sets, respectively.

• The model was evaluated using area under the receiver operating characteristic (ROC) curve.
• AUC of 0.82 was achieved in the test set and 0.80 in the validation set.
• The model was built using 34 metabolites identified as significantly disrupted in depressive individuals.
• Performance was validated in independent cohorts to confirm diagnostic potential.

Mediation analysis revealed that metabolites act as key mediators linking gut microbiota to depression.

• 61 metabolite-mediated species-depression correlations were identified through mediation analysis.
• This finding supports the role of metabolites as functional intermediaries between gut microbiota and depressive pathophysiology.
• The mediation framework was used to dissect direct versus metabolite-mediated effects of microbial species on depression.

The study design incorporated first-episode depression patients and matched controls with validation across multiple independent cohorts.

• The primary discovery cohort included n = 186 first-episode depression patients and matched controls.
• Three independent validation cohorts included n = 223, n = 85, and n = 52 participants.
• One validation cohort included a drug intervention arm to assess treatment-related metabolite changes.
• Matching of cases and controls was employed to reduce confounding.