Altered gut bacterial-fungal interkingdom networks in patients with current depressive episode

Abstract

Introduction: Bacterial dysbiosis has been described in patients with current depressive episode (CDE); however, the fungal composition in the gut has not been investigated in these patients.

Methods: Here, we characterized the fungal gut mycobiota in patients with CDE. We systematically characterized the microbiota and mycobiota in fecal samples obtained from 24 patients with CDE and 16 healthy controls (HC) using 16S rRNA gene- and ITS1-based DNA sequencing, respectively.

Results: In patients with CDE, bacterial dysbiosis was characterized by an altered composition and reduced correlation network density, and the gut mycobiota was characterized by a relative reduction in alpha diversity and altered composition. Most notably, the CDE group had higher levels of Candida and lower level of Penicillium than the HC group. Compared with the HC group, the gut microbiota in patients with CDE displayed a significant disruption in the bacteria-fungi correlation network suggestive of altered interkingdom interactions. Furthermore, a gut microbial index based on the combination of eight genera (four bacterial and four fungal CDE-associated genera) distinguished CDE patients from controls with an area under the curve of approximately 0.84, suggesting that the gut microbiome signature is a promising tool for disease classification.

Conclusions: Our findings suggest that both bacteria and fungi contribute to gut dysbiosis in patients with CDE. Future studies involving larger cohorts and metagenomic or metabolomic analyses may clarify the structure and potential roles and functions of the gut mycobiome and its impact on the development of CDE.

Keywords: dysbiosis; fungi; gut microbiota-brain; mental.

FIGURE 1

Phylogenetic diversity of gut bacterial microbiomes among CDE and HC. Comparison of the ace (a), Chao (b), shannon (c), and simpson (d) index between two groups; (e) PCoA of Bray‐Curtis distance matrix analysis demonstrated that the bacterial microbiome composition of CDE clustered separately from HC; (f) Composition of fecal bacterial microbiome at the phylum level between the two groups; (g) Taxonomic cladogram obtained from LEfSe analysis of 16S sequences. (Blue) CDE taxa; (Green) taxa enriched in HCs. The brightness of each dot is proportional to its effect size. (h) HC‐enriched taxa are indicated with a positive LDA score (Green), and taxa enriched in CDE have a negative score (Blue). Only taxa meeting an LDA significant threshold > 2 are shown. * indicates p < .05, ns indicates p > .05

FIGURE 2

Phylogenetic diversity of gut fungal microbiomes among CDE and HC. Comparison of the ace (a), Chao (b), shannon (c), and simpson (d) index between two groups; (e) Composition of fecal fungal microbiome at the phylum level between the two groups; (f) Taxonomic cladogram obtained from LEfSe analysis of ITS sequences. (Blue) CDE taxa; (Green) taxa enriched in HCs. The brightness of each dot is proportional to its effect size. (g) HC‐enriched taxa are indicated with a positive LDA score (Green), and taxa enriched in CDE have a negative score (Blue). Only taxa meeting an LDA significant threshold >2 are shown. * p < .05, ** indicates p < .01, ns indicates p > .05

FIGURE 3

(a) Correlation networks between bacteria were built using Spearman's correlation test in the two study groups. Abundance correlation networks are shown, in which each circle (node) represents an operational taxonomic unit (OTU), its color, the bacterial phylum (blue, Firmicutes; green, Bacteroidetes; yellow, Actinobacteria; orange, Proteobacteria), and its size, the number of direct edges. Edges indicate the magnitude of distance correlation (positive in blue, negative in red). Only OTUs present in > 50% of the samples were taken into account, and only significant correlations (p < .05) are shown. The networks' parameters are presented in the table. The relative connectedness of the networks was calculated as the ratio between the number of significant interactions (edges) and the number of taxa (nodes) in the network. (b) Number of neighbors; means and standard error of mean (SEM) are indicated. **** indicates p < .0001

FIGURE 4

(a) Correlation networks at the family and genus levels were built in the two groups using distance correlation test. Abundance correlation networks are shown, in which each node represents a family or a genus, its shape, the kingdom (square, fungi; circle, bacteria), its color, and its size, the number of direct edges. Edges indicate the magnitude of distance correlation (positive in blue, negative in red). Only family and genus present in > 20% of the samples were taken into account, and only significant correlations (p < .05) are shown. The networks' parameters are presented in the table. The relative connectedness of the networks was calculated as the ratio between the number of significant interactions (edges) and the number of taxa (nodes) in the network. (b) Number of neighbors. Number of neighbors; means and SEM are indicated. **** indicates p < .0001. (c) Classification performance of multivariable logistic regression model using relative abundance of CDE‐associated genera was assessed by area under the receiving operational curve (ROC)