Fig 1.
Influence of different commensal E. coli strains on DSS-induced colitis in WT and Tlr5−/− mice.
(a + b) SPF C57BL/6 WT mice aged 6 to 8 weeks were administered 3.5% DSS in drinking water at day 0. Mice were additionally treated with EcN (DSS + EcN), MG1655 (DSS + MG1655), or MPK (DSS + MPK) resuspended in DSS-containing drinking water at 108 bacteria mL−1. (a) Change in body weight relative to start of DSS administration at day 0. (b) Left panel: HE-stained colonic sections at day 7 after start of DSS administration. Right panel: HCS at day 7. (c + d) SPF C57BL/6 WT mice aged 6 to 8 weeks were administered 3.5% DSS in drinking water at day 0. Mice were additionally treated with EcN (DSS + EcN) or an EcNΔfliC deletion mutant (DSS + EcNΔfliC) resuspended in DSS-containing drinking water at 108 bacteria mL−1. (c) Change in body weight relative to start of DSS administration at day 0. (d) Left panel: HE-stained colonic sections at day 7 after start of DSS administration. Right panel: HCS at day 7. (e) SPF C57BL/6 WT mice and SPF Tlr5−/− mice aged 6 to 8 weeks were administered 3.5% DSS in drinking water at day 0. Mice were additionally treated with EcN (DSS + EcN) or an EcNΔfliC deletion mutant (DSS + EcNΔfliC) resuspended in DSS-containing drinking water at 108 bacteria mL−1. Left panel: HE-stained colonic sections at day 7 after start of DSS administration. Right panel: HCS at day 7. Statistics: (a), (b), (c), (e), one-way ANOVA with Tukey multiple comparison test; (d) Kruskal–Wallis test with multiple comparisons. p-values < 0.05 are considered to represent statistical significance. (a–e) The data underlying this figure can be found in S1 Data. DSS, dextran sodium sulphate; EcN, E. coli Nissle 1917; fliC, flagellin; HCS, histological colitis score; HE, hematoxylin–eosin; MG1655, E. coli K12 MG1655; MPK, E. coli mpk; SPF, specific-pathogen–free; TLR, Toll-like receptor; WT, wild type.
Fig 2.
Flagella-dependent influence of different commensal E. coli strains on DSS-induced colitis in WT mice.
(a + b) SPF C57BL/6 WT mice aged 6 to 8 weeks were administered 3.5% DSS in drinking water at day 0. Mice were additionally treated with FEPs obtained from 107 (EcN FEP 107) and 1010 (EcN FEP 1010) EcN per 100 mL drinking water. (a) Change in body weight relative to start of DSS administration at day 0. (b) Left panel: HE-stained colonic sections at day 7 after start of DSS administration. Right panel: HCS at day 7. (c + d) SPF C57BL/6 WT mice aged 6 to 8 weeks were administered 3.5% DSS in drinking water at day 0. Mice were additionally treated with FEPs obtained from 1010 EcN, MG1655, or MPK per 100 mL drinking water. (c) Change in body weight relative to start of DSS administration at day 0. (d) Left panel: HE-stained colonic sections at day 7 after start of DSS administration. Right panel: HCS at day 7. (e) mTLR5-HEK293 cells were stimulated with FEP obtained from EcN, MG1655, and MPK for 24 h. FEPs were generated from the number of bacteria corresponding to a certain MOI (quasi-MOI). Resulting IL-8 secretion into cell supernatant as a result of TLR5 receptor activation was detected by ELISA. (f) mTLR5-HEK293 cells were stimulated with EcN and EcNΔfliC at different MOI for 24 h. Resulting IL-8 secretion into cell supernatant as a result of TLR5 receptor activation was detected by ELISA. (g + h) SPF C57BL/6 WT mice aged 6 to 8 weeks were administered 3.5% DSS in drinking water at day 0. Mice were additionally treated with 1010 viable bacteria of the indicated complementation mutants per 100 mL DSS-containing drinking water. (g) Change in body weight relative to start of DSS administration at day 0. (h) Left panel: HE-stained colonic sections at day 7 after start of DSS administration. Right panel: HCS at day 7. Statistics: (a), (c), (g), one-way ANOVA with Tukey multiple comparison test; (b), (d), (h), Kruskal–Wallis test with multiple comparisons. p-values < 0.05 are considered to represent statistical significance. The data underlying this figure can be found in S1 Data. DSS, dextran sodium sulphate; EcN, E. coli Nissle 1917; FEP, flagella-enriched preparation; fliC, flagellin; HCS, histological colitis score; HE, hematoxylin–eosin;; IL, interleukin; MG1655, E. coli K12 MG1655; MOI, multiplicity of infection; MPK, E. coli mpk; mTLR5-HEK293 cell, mouse-TLR5–expressing human embryonic kidney 293 cell; SPF, specific-pathogen–free; TLR, Toll-like receptor; WT, wild type.
Fig 3.
Detailed comparison of the flagellin amino-acid sequences of different E. coli strains.
Protein alignment of FliC proteins from EcN (CCQ05465.1), MPK, and MG1655 (NP_416433.1) genomes. (a) Schematic structure of flagellin according to Yonekura and colleagues [50,51]. D0 and D1 comprise conserved N- and C-termini of fliC, packed into α-helical structures in the filament core. The NTD2/CTD2- and D3-domain–containing HVR is attached adjacent to the D2 domains, located at the outer surface of the filament. (b) Quantification of amino-acid sequence similarities of all 6 flagellin domains. Computation was performed using similarity indices depicted in 3c. (c) The alignment was generated using MAFFT. Amino acids were colored by overall conservation (white: fully conserved, light blue to dark blue: high to low conservation). Consensus sequence and sequence conservation for pairwise comparisons of EcN to MPK and MG1655, respectively, are shown in yellow. Darker shades of yellow represent lower sequence conservation. (d) Schematic overview of the differences of the EcN flagellin HVR compared to the HVRs of both MG1655 and MPK. Sequences (insertions) that are only present in EcN HVR and not in MPK HVR or MG1655 HVR are highlighted as colored squares. AA, amino acid; CTD, C-terminal domain; EcN, E. coli Nissle 1917; fliC, flagellin; HVR, hypervariable region; MAFFT, Multiple Alignment using Fast Fourier Transform; MG1655, E. coli K12 MG1655; MPK, E. coli mpk; NTD, N-terminal domain.
Fig 4.
Partial deletion of the EcN HVR leads to loss of EcN probiotic effects.
(a) Schematic display of the deleted parts in the EcNΔ(fliC)HVR mutant. The orange line depicts the parts of the EcN HVR that were deleted. (b + c) SPF C57BL/6 WT mice aged 6 to 8 weeks were administered 3.5% DSS in drinking water at day 0. Mice were additionally treated with 1010 viable bacteria of EcN or the EcNΔ(fliC)HVR mutant resuspended in 100 mL DSS-containing drinking water. (b) Change in body weight relative to start of DSS administration at day 0. (c) HCS at day 7. Statistical analysis was performed using Student t test. Error bars represent SD. White dots in column bars represent each biological replicate. (d) Heat map of cytokine concentrations in serum from DSS-treated mice. Each column represents a different individual. (e) Detailed analysis of cytokine concentrations from serum depicted in (d). Asterisks indicate statistical significance. All shown cytokines provide the strongest differences between DSS-treated and DSS + EcNΔfliC(HVR)-treated as well as between DSS + EcN and DSS + EcNΔfliC(HVR)-treated groups. See S1 Table for detailed statistical analysis using Kruskal–Wallis test with multiple comparisons. p-values < 0.05 are considered to represent statistical significance. (b + c + e) The data underlying this figure can be found in S1 Data. CTD, C-terminal domain; DSS, dextran sodium sulphate; EcN, E. coli Nissle 1917; fliC, flagellin; GM-CSF, granulocyte-macrophage colony-stimulating factor; HCS, histological colitis score; HVR, hypervariable region; IFN, interferon; IL, interleukin; MCP, monocyte chemoattractant protein; MG1655, E. coli K12 MG1655; MPK, E. coli mpk; NTD, N-terminal domain; ns, not significant; SPF, specific-pathogen–free; TNF, tumor necrosis factor; WT, wild type.
Fig 5.
EcN-flagellin–induced protective effects are mediated by host TLR5+CD11c+ cells in the cLP.
(a) Schematic depiction of the generation of BMCM as described in the experimental procedures. Red dots represent the typical sites of TLR5 expression in the mouse intestine. (b) Different SPF BMCM were administered 3.5% DSS in drinking water at day 0. Mice were additionally treated with 1010 viable bacteria of EcN or the EcNΔ(fliC)HVR mutant resuspended in 100 mL DSS-containing drinking water. See text for further information on the nomenclature of distinct BMCM groups. Indicated p-values refer to the comparison of the respective data set with the WT → Tlr5−/− + DSS + EcN group. Upper panel: HCS at day 7. Lower panel: HE-stained colonic sections at day 7 after start of DSS administration. (d) Irradiated WT recipient mice were transplanted with bone marrow from Tlr5−/− and ΔCD11c donor mice in a 1:1 ratio. Mice were administered 3.5% DSS in drinking water at day 0. Mice were additionally treated with 1010 viable bacteria of EcN resuspended in 100 mL DSS-containing drinking water and compared to control groups. See text for further information on the BMCM groups nomenclature. Middle panel: HCS at day 7. Right panel: HE-stained colonic sections at day 7 after start of DSS administration. Statistics: (b) One-way ANOVA with Tukey multiple comparison test, (c) Mann–Whitney test, (d) Kruskal–Wallis test with multiple comparisons, p-values < 0.05 are considered to represent statistical significance. (b + c) The data underlying this figure can be found in S1 Data. BMCM, bone-marrow–chimeric mice; cLP, colonic LP; DSS, dextran sodium sulphate; EcN, E. coli Nissle 1917; fliC, flagellin; HCS, histological colitis score; HE, hematoxylin–eosin; HVR, hypervariable region; LP, lamina propria; SPF, specific-pathogen–free; TLR, Toll-like receptor; WT, wild type.
Fig 6.
Recombinant FliC protects against DSS-induced colitis.
(a) mTLR5-HEK293 cells were stimulated with rfliC(K12) or rfliC(EcN) for 24 h. Resulting IL-8 secretion into cell supernatant as a result of TLR5 receptor activation was detected by ELISA. (b) Experimental setup: SPF C57BL/6 WT mice aged 6 to 8 weeks were administered 2 μg recombinant flagellin daily via intragastral gavage. 3 days after start of flagellin administration, 3.5% DSS was added to the drinking water. Progress of DSS-induced colitis was monitored for additional 7 days. (c) HCS and representative HE-stained colonic sections at day 7. Statistical analysis was performed using the Kruskal–Wallis test. Error bars represent SD. White dots in column bars represent each biological replicate. (d) Heat map of cytokine concentrations in serum from DSS-treated mice as shown in (b). Each column represents a different individual. Right panel: IL-22 concentration in blood serum. Statistical analysis was performed using one-way ANOVA. Error bars represent SD. White dots in column bars represent each biological replicate. (e) Gating strategy to define the population of intDCs from the cLP. Lin = Ly6G/C, CD45R, CD64. (f + g) Proportion of IL-23+ (f) or IL-22+ (g) intDCs from the experiment shown in (b). Representative histograms or contour blots are shown. Statistical analysis was performed using one-way ANOVA. Error bars represent SD. White dots in column bars represent each biological replicate. p-values < 0.05 are considered to represent statistical significance. (a + c + d + f + g) The data underlying this figure can be found in S1 Data. cLP, colonic LP; DC, dendritic cell; DSS, dextran sodium sulphate; EcN, E. coli Nissle 1917; fliC, flagellin; FSC, forward scatter; HCS, histological colitis score; HE, hematoxylin–eosin; IL, interleukin; intDC, intestinal DC; LP, lamina propria; MG1655, E. coli K12 MG1655; mTLR5-HEK293 cell, mouse-TLR5–overexpressing human embryonic kidney 293 cell; ns, not significant; rfliC(EcN), recombinant flagellin from EcN; rfliC(K12), recombinant flagellin from MG1655; SPF, specific-pathogen–free; TLR, Toll-like receptor.