Metabolite-mediated interactions and direct contact between Fusobacterium varium and Faecalibacterium prausnitzii

Abstract

Background: The human gut harbors a diverse microbiota that is crucial for maintaining health but also contributes to several diseases. Understanding how microbial communities are assembled and maintained is critical for advancing gut health.

Results: We identified a unique interaction between the pathobiont Fusobacterium varium and the symbiont Faecalibacterium prausnitzii, both members of the gut microbial community; their interaction is driven by metabolites and direct cell-to-cell contact. Growth of F. varium was inhibited in the presence of F. prausnitzii because of a decrease in pH and an increase in β-hydroxybutyric acid. Conversely, the growth of F. prausnitzii was promoted in the presence of F. varium, likely via direct contact.

Conclusions: These findings highlight the importance of metabolite-driven interactions and direct contact in shaping gut microbial communities and emphasize the potential of interactions between F. prausnitzii and F. varium in influencing gut health. Video Abstract.

Keywords: Faecalibacterium; Fusobacterium; Gut ecosystem; Microbial interaction; Pathobiont; Symbiont.

Fig. 1

Inverse relationship between Fusobacterium and Faecalibacterium in Japanese adults. A Distribution of the relative abundance of the Fusobacterium genus in the feces of 236 participants (Supplementary Table 1). B Differences in bacterial taxonomy ranked by the linear discriminant analysis (LDA) effect size (P < 0.05) among participants with (n = 120) and without (n = 116) Fusobacterium in their feces. C Scatterplots of Pearson and Spearman correlation analysis between the relative abundances of the Fusobacterium and Faecalibacterium genera (n = 236). D, E Relative abundance of D Fusobacterium and E Faecalibacterium species in the 112 participants. F Heatmaps of Pearson and Spearman correlation analysis between the relative abundances of Fusobacterium and Faecalibacterium species (n = 112). G Scatterplots of Pearson and Spearman correlation analysis between the relative abundances of F. varium and F. prausnitzii (n = 112). The data were obtained by A, B, C 16S rRNA gene amplicon sequencing or D, E, F, G shotgun metagenomic sequencing

Fig. 2

Suppression of F. varium growth by F. prausnitzii via low pH and β-hydroxybutyric acid. A Effect of F. prausnitzii on F. varium growth. F. varium was cultured in the presence or absence of F. prausnitzii. *P < 0.05 (two-tailed Mann–Whitney U-test). B Effect of the supernatant from co-culture of F. varium and F. prausnitzii on F. varium growth. F. varium was cultured in the absence or presence of the supernatant at the indicated concentrations. **P < 0.01 (one-way ANOVA). C The pH of culture medium during bacterial growth. F. varium alone, F. prausnitzii alone, or both were cultured (n = 4, mean ± 1 SD). D Impact of pH on the inhibitory effect of co-culture supernatant. F. varium was cultured in the absence or presence of the supernatant from F. varium and F. prausnitzii co-culture adjusted or not to pH 6.7 (the initial pH of the medium). *P < 0.05, **P < 0.01 (one-way ANOVA). E Volcano plot of bacterial metabolites. F. varium and F. prausnitzii were co-cultured for 24 h, and the primary metabolites and short-chain fatty acids were measured by LC–MS/MS. Green dots indicate metabolites that increased by > 10 × in the co-culture supernatant compared with fresh medium (n = 4). Statistical significance was evaluated by using two-tailed unpaired t-test. F β-Hydroxybutyric acid concentration in bacterial cultures. F. prausnitzii alone, F. varium alone, or both were cultured for 24 h. The concentration of β-hydroxybutyric acid was measured by LC–MS/MS (n = 4, mean ± 1 SD). G Effects of pH and β-hydroxybutyric acid on F. varium growth. F. varium was cultured for 24 h in YCFA (pH 6.7 or 6.0) in the absence or presence of β-hydroxybutyric acid. *P < 0.05, ***P < 0.001; n.s., not significant (one-way ANOVA). A, B, D, G The number of F. varium cells was assessed by quantitative PCR (n = 4, mean ± 1 SD)

Fig. 3

Gene expression and metabolome analyses of F. varium cultured in low-pH and high-β-hydroxybutyric acid conditions. A Volcano plots of gene expression. F. varium was cultured for 24 h in YCFA, YCFA in the presence of F. prausnitzii, YCFA (pH 6.0), or YCFA (pH 6.0) supplemented with 2-mM β-hydroxybutyric acid (BHB). Gene expression was analyzed by RNA-seq (n = 4). Red dots indicate genes upregulated or downregulated by > 10 × (P < 0.01, two-tailed unpaired t-test). B Venn diagram of genes downregulated under the conditions used in A. C Transcript levels of the 12 genes shown in B that were consistently downregulated under all 3 conditions. TPM, transcripts per million. D Heatmap of amino acid contents. F. varium was cultured for 24 h in YCFA, YCFA (pH 6.0), YCFA supplemented with 2-mM β-hydroxybutyric acid (BHB), or YCFA (pH 6.0) supplemented with 2-mM BHB. Intracellular amino acid contents were measured by LC–MS/MS (n = 4) and normalized by bacterial cell number assessed by quantitative PCR. ***P < 0.001 (two-way ANOVA)

Fig. 4

Inhibition of F. varium growth by other intestinal bacteria. A Effects of intestinal bacterial species on F. varium growth. F. varium was cultured in the absence or presence of B. vulgatus (now named P. vulgatus), B. wexlerae, B. longum, or A. muciniphila. The number of F. varium cells was assessed by quantitative PCR (n = 4, mean ± 1 SD). *P < 0.05, **P < 0.01 (one-way ANOVA). B The pH of bacterial cultures. F. varium was cultured for 24 h in the absence or presence of the above species (n = 4, mean ± 1 SD). **P < 0.01 (one-way ANOVA). C Concentration of β-hydroxybutyric acid in bacterial cultures. F. varium was cultured as in B. The concentration of β-hydroxybutyric acid was measured by LC–MS/MS (n = 4, mean ± 1 SD). **P < 0.01 (one-way ANOVA). D Scatterplots for Pearson and Spearman correlation analysis between the relative abundances of Fusobacterium and Bacteroides or Blautia assessed by 16S rRNA gene amplicon sequencing analysis of Japanese adults (n = 236) (Supplementary Table 1)

Fig. 5

Promotion of F. prausnitzii growth by F. varium. A Volcano plot of gene expression in F. prausnitzii. F. prausnitzii was cultured for 24 h in YCFA in the presence or absence of F. varium. Gene expression was analyzed by RNA-seq (n = 4). Green circle indicates genes upregulated by > 80 × (P < 0.01, two-tailed unpaired t-test). B Transcript levels of the five genes upregulated > 80 × in A. TPM, transcripts per million. C Effect of F. varium on F. prausnitzii growth. *P < 0.05 (two-tailed Mann–Whitney U-test). D Effects of iron availability on F. prausnitzii growth. F. prausnitzii was cultured anaerobically in YCFA supplemented or not with hemin. *P < 0.05 (two-tailed Mann–Whitney U-test). E Effects of F. varium on growth of B. vulgatus and B. wexlerae cultured anaerobically in YCFA. F Scanning electron microscopy images of F. prausnitzii and F. varium. Bacteria were co-cultured anaerobically in YCFA for 24 h, fixed with glutaraldehyde, and observed under a scanning electron microscope (JSM-7500F, JEOL, Tokyo, Japan). Scale bars: 10 μm or 100 nm. C, D, E The number of bacterial cells was assessed by quantitative PCR (n = 4, mean ± 1 SD)