In Vitro Characterization of Gut Microbiota-Derived Commensal Strains: Selection of Parabacteroides distasonis Strains Alleviating TNBS-Induced Colitis in Mice

Abstract

Alterations in the gut microbiota composition and diversity seem to play a role in the development of chronic diseases, including inflammatory bowel disease (IBD), leading to gut barrier disruption and induction of proinflammatory immune responses. This opens the door for the use of novel health-promoting bacteria. We selected five Parabacteroides distasonis strains isolated from human adult and neonates gut microbiota. We evaluated in vitro their immunomodulation capacities and their ability to reinforce the gut barrier and characterized in vivo their protective effects in an acute murine model of colitis. The in vitro beneficial activities were highly strain dependent: two strains exhibited a potent anti-inflammatory potential and restored the gut barrier while a third strain reinstated the epithelial barrier. While their survival to in vitro gastric conditions was variable, the levels of P. distasonis DNA were higher in the stools of bacteria-treated animals. The strains that were positively scored in vitro displayed a strong ability to rescue mice from colitis. We further showed that two strains primed dendritic cells to induce regulatory T lymphocytes from naïve CD4⁺ T cells. This study provides better insights on the functionality of commensal bacteria and crucial clues to design live biotherapeutics able to target inflammatory chronic diseases such as IBD.

Keywords: IBD; colitis; functional screening; holobiont; immune response; live biotherapeutic products (LBP); microbiota; probiotics.

Figure 1

Ability of the strains to restore the H₂O₂-induced disruption of the epithelial barrier. The Caco-2 confluent monolayers were pre-treated (or not) with the bacteria at the apical side (10:1 bacteria/cell ratio) for 30 min and sensitized with H₂O₂ (100 μM). Trans-epithelial electrical resistance (TEER) was measured at T0 and every 30 min. Data represent (A) the time-dependent means of relative changes (in %) of TEER ± SEM in comparison to TEER at T0 and (B) the final % TEER variations at 120 min. ^# and * refer to the comparison of the cells treated with H₂O₂ (H₂O₂ Control) and without H₂O₂ (N/S Control) or bacteria-treated cells with H₂O₂ versus H₂O₂ Control; * p < 0.05; ** or ^## p < 0.01.

Figure 2

In vitro immunomodulatory profiles of the strains. (A) IL-10, (B) IL-12p70 and (C) IFN-γ production was evaluated in the supernatants of peripheral blood mononuclear cells (PBMCs; n = 5 different donors) stimulated for 24 h by the tested strains or two control strains (L. acidophilus NCFM and B. animalis subsp. lactis BB12), in comparison to non-treated cells (N/S). (D) IL-10/IL-12 ratio was calculated. Data represent means ± SEM of the 5 independent donors. * refers to the comparison of bacteria-stimulated PBMCs versus untreated cells; * p < 0.05, ** p < 0.01, **** p < 0.0001.

Figure 3

Relative survival of selected strains to the simulated gastric fluid over 2 h. Results are expressed as the ratio of the CFU/mL at a given time point to the CFU/mL at time zero ± SEM with a semi-logarithmic scale. * indicates a default value corresponding to a number of CFU/mL below 100 (as the CFU/mL at T0 was slightly different among strains, the default values differ). Dotted lines are joining the last CFU measures and default values.

Figure 4

Ability of Parabacteroides distasonis strains to counteract the acute TNBS-induced colitis. (A) Body weight loss (as a percentage of the initial weight). (B) Macroscopic evaluation of colonic inflammation (Wallace score). Percentages of protection are indicated above each bar. (C) Histologic evaluation of colonic inflammation (Ameho score). (D) Representative histological sections (stained by H&E, 100× magnification) of mice treated with TNBS (TNBS) or not (healthy mice, ethanol-control mice) and orally treated with the selected strains. Data represent means of each group (n = 9 mice per group) ± SEM. ^# and * refer to the comparisons of TNBS versus healthy mice or bacteria-treated group versus TNBS control group, respectively; ** p < 0.01, *** p < 0.001, ^#### or **** p < 0.0001. (E) Abundance of P. distasonis specific DNA in the stools collected two days after colitis induction and evaluated by qPCR. Results are expressed as relative expression compared with values obtained from healthy mice. Data represent means of each group (n = 9 mice per group) ± SEM. * refer to the comparisons of each group versus healthy mice; * p < 0.05, *** p < 0.001.

Figure 5

Capacity of the strains to modulate (A) the plasmatic IL-6 concentration (pg/mL), (B) the fecal lipocalin-2 levels (μg/g feces), (C) the expression of genes encoding proinflammatory markers or (D) tight junction proteins during TNBS-induced colitis. IL-6 and fecal lipocalin-2 concentrations were measured by ELISA. Gene expression of Il1b, Il6, Tnfa, Cxcl2, Occludin and Zo1 was evaluated by qRT-PCR from colonic samples obtained two days after colitis induction. Results are expressed as Relative expression compared with values obtained from healthy mice. Data represent means of each group (n = 9 mice per group) ± SEM. ^# and * refer to the comparisons of the TNBS-treated control versus healthy mice or bacteria-treated group versus TNBS control group, respectively; ^# or * p < 0.05, ^## or ** p < 0.01, *** p < 0.001, ^#### or **** p < 0.0001.

Figure 6

Ability of the strains to modulate tight junction proteins in the colon. Representative immunofluorescent images of ZO-1, Claudin-3 and Claudin-2 of colon sections of the different groups of mice (A): healthy mice; (B): ethanol, (C): TNBS; (D): TNBS+PF-BaE7; (E): TNBS+PF-BaE11; (F): TNBS+AS93; (G): TNBS+PF-BaE5 and (H): TNBS+AS23, observed by immunofluorescence and confocal microscopy after labeling with specific primary antibodies and AF488-conjugated secondary antibody (green), with nuclei counterstained with DAPI (blue) stain. A red canal was added to identify autofluorescence tissue elements (red/yellow) and reinforce the specific labeling. Boxed areas represent magnified images (×2.5) as insets from the corresponding white box. Long arrows indicate normal TJ (ZO-1 and Claudin-3) distribution at the apical and lateral levels and labeling for Claudin-2; short arrows indicate discontinuities or diffuse cytoplasmic distribution of TJ. Scale bars 50 μm.

Figure 7

Ability of the strains (A) to activate murine BMDCs and to induce (B) the expression of Il33 in BMDC, (C) the induction of IL-10 producing CD4⁺ FoxP3⁺ T cells and (D) the gene expression of Ebi3 in CD4⁺ T cells. Data represent means ± SEM of 3 independent experiments. * refers to the comparison of bacteria-treated cells versus unstimulated cells (N/S); * p < 0.05, ** p < 0.01, **** p < 0.0001.