Unveiling the inhibition mechanism of Clostridioides difficile by Bifidobacterium longum via multiomics approach

Abstract

Antibiotic-induced gut microbiota disruption constitutes a major risk factor for Clostridioides difficile infection (CDI). Further, antibiotic therapy, which is the standard treatment option for CDI, exacerbates gut microbiota imbalance, thereby causing high recurrent CDI incidence. Consequently, probiotic-based CDI treatment has emerged as a long-term management and preventive option. However, the mechanisms underlying the therapeutic effects of probiotics for CDI remain uninvestigated, thereby creating a knowledge gap that needs to be addressed. To fill this gap, we used a multiomics approach to holistically investigate the mechanisms underlying the therapeutic effects of probiotics for CDI at a molecular level. We first screened Bifidobacterium longum owing to its inhibitory effect on C. difficile growth, then observed the physiological changes associated with the inhibition of C. difficile growth and toxin production via a multiomics approach. Regarding the mechanism underlying C. difficile growth inhibition, we detected a decrease in intracellular adenosine triphosphate (ATP) synthesis due to B. longum-produced lactate and a subsequent decrease in (deoxy)ribonucleoside triphosphate synthesis. Via the differential regulation of proteins involved in translation and protein quality control, we identified B. longum-induced proteinaceous stress. Finally, we found that B. longum suppressed the toxin production of C. difficile by replenishing proline consumed by it. Overall, the findings of the present study expand our understanding of the mechanisms by which probiotics inhibit C. difficile growth and contribute to the development of live biotherapeutic products based on molecular mechanisms for treating CDI.

Keywords: Bifidobacterium longum; Clostridioides difficile; microbe-microbe interaction; molecular mechanism; multiomics.

FIGURE 1

(A) Growth inhibition of C. difficile due to Bifidobacterium strains. Growth inhibition test was performed in triplicate, and the significance was compared with the C. difficile alone group. (B) Antimicrobial activity of B. longum against C. difficile was assessed using the spot-on-lawn method. Representative picture showing inoculum density–dependent antimicrobial activity of B. longum against C. difficile. C. difficile was used as a negative control (center spot of agar plate). Circles indicate inhibition zone. (a–d) Indicate the inoculated CFU of B. longum (10⁴, 10⁵, 10⁶, and 10⁷ CFU, respectively). (C) Width of the inhibition zone (mm) around the B. longum spot on the agar plate depends on the inoculated density of B. longum. These experiments were performed in triplicate. (D) Quantitative comparison of organic acids in culture media (n = 4). The significance was calculated via one-way analysis of variance and post-hoc (Tukey’s Honestly Significant Difference test). Error bar indicates the standard deviation. The symbols (*), (**), and (***) indicate p-value < 0.05, < 0.01, and < 0.001, respectively.

FIGURE 2

(A) Volcano plot of proteomics analysis. (B) Principal component analysis (PCA) plot of proteomics analysis. (C) Volcano plot of metabolomics analysis. (D) PCA plot of metabolomics analysis. In the volcano plot in panel (A), the red and blue colors indicate significantly upregulated and downregulated proteins, and in panel C the red and blue colors indicate metabolites that are at higher and lower levels, respectively, C. difficile cocultured with B. longum compared to C. difficile cultured alone. In the PCA plot, each experimental group was separately clustered. These results indicate that coculture with B. longum caused significant proteomic and metabolomic changes in C. difficile.

FIGURE 3

(A) Illustration of the effects of lactate reduction on ATP synthesis. This illustration was created using BioRender.com. (B) Alterations in intracellular protein levels of the LDH complex, ATP synthase, and Rnf complex. The symbol (*) indicates differentially expressed proteins (DEPs). (C) Intracellular protein levels of lactate and ATP. Error bar indicates the standard deviation. The symbols (*) and (**) indicate adjusted p-value < 0.05 and < 0.01, respectively.

FIGURE 4

(A) Illustration of the effects of proline metabolism upregulation on butyrate metabolism and toxin production. This illustration was created using BioRender.com. (B) Alterations in the intracellular protein levels of proline-dependent regulation. The symbol (*) indicates differentially expressed proteins. (C) Abundance levels of metabolites belonging to proline reduction and butyrate metabolism. (D) Increased levels of the intracellular NAD⁺/NADH ratio in C. difficile cocultured with B. longum. To measure the NAD⁺/NADH ratio, absolute quantification of NAD⁺ and NADH was performed, and then the NAD⁺ concentration was divided by the NADH concentration of each sample. (E) Increased extracellular proline level in B. longum monocultured media (n = 3). (F) Toxin A concentration in culture supernatant after 48 h of cultivation (n = 3). Error bar indicates the standard deviation. The symbols (*), (**) and (***) indicate p-value < 0.05, < 0.01, and < 0.001, respectively.

FIGURE 5

(A) Protein–protein interaction network of the gene ontology (GO) terms “ribosome” and “regulation of translation.” GOCC indicates gene ontology cellular Component and GOBP indicates the gene ontology biological process. Alteration of intracellular protein levels related to (B) ribosome, (C) regulation of translation, (D) molecular chaperones, and chaperonins.

FIGURE 6

Heatmap of nucleosides and nucleoside phosphates. The red and blue colors indicate relatively high and low abundances, respectively.