Gut microbiome signatures associated with depression and obesity.

Machine learning models based solely on taxonomic profiles classified subjects into depressed and non-depressed controls with a balanced accuracy of 0.90.

• Case-control study with 105 total subjects: 43 with major depressive disorder (MDD) and 62 non-depressed controls free from psychiatric comorbidities.
• Multiple machine learning methods were applied to shotgun metagenomic data following taxonomic annotation.
• Classification into depressed or non-depressed and normal weight or overweight achieved a balanced accuracy of 0.78 based solely on taxonomic profiles.

Novel bacterial taxa were identified as associated with depression, including reductions in Butyrivibrio hungatei and Anaerocolumna sedimenticola in MDD patients.

• These taxa represent newly reported associations with MDD not previously described in the literature.
• The study also replicated previously reported associations, such as decreased Faecalibacterium prausnitzii in patients with MDD.
• Shotgun metagenomics was used for taxonomic annotation to identify these differences.

Functional metagenomic analysis revealed an increase in tryptophan degradation and a decrease in queuosine synthesis pathways in MDD patients.

• Both tryptophan degradation and queuosine synthesis pathways are directly related to a decrease in monoaminergic neurotransmitter availability.
• These functional differences were identified through functional annotation of metagenomes.
• Additional differences were found in pathways linked to the synthesis of fundamental nutrients, associated with diet and inflammation.

Most metabolic functions found to be more abundant in controls compared to depressed subjects were encoded by Faecalibacterium prausnitzii.

• F. prausnitzii was consistently depleted in MDD patients, replicating prior reported associations.
• The functional analysis attributed the majority of control-enriched metabolic functions to this single species.
• This suggests a potential use of F. prausnitzii to improve depression symptoms as a gut-directed therapeutic target.

Shared microbiome features were identified between MDD and BMI, suggesting common underlying mechanisms linking depression and obesity.

• The study was designed to dissect the gut microbiome's relation to both MDD and body mass index (BMI), as well as lifestyle factors including diet and exercise.
• Depression and obesity are described as highly comorbid conditions that likely involve common risk factors and pathophysiological mechanisms.
• Classification models incorporating both depression status and weight category (normal weight or overweight) achieved a balanced accuracy of 0.78.

Microbial and functional signatures associated with depression were found to be independent of other covariates such as BMI and lifestyle.

• The study specifically examined the influence of diet and exercise as lifestyle covariates alongside BMI.
• Distinct microbial signatures remained associated with depression independent of these covariates.
• Shotgun metagenomics with both taxonomic and functional annotations was used to enable this dissection.