Figure 1.
Diet induced changes in the susceptibility to DSS.
WT or Rag KO mice were fed the experimental diets for 2% DSS treatment and throughout the experiment. Percent original BW of (A) WT (n = 4–6 mice/group) and (B) Rag KO mice following the start of DSS treatment (n = 3–5 mice/group). All three dietary groups were significantly different from each other (*P<0.05, ***P<0.001, two-way ANOVA). (C) Colonic blood scores in WT mice at d5 post DSS (n = 3 mice/group). TD was significantly different from PD (*P<0.05, one-way ANOVA with Tukey's post-tests). (D) Colonic lengths of the WT mice at d0 (n = 4 mice/group), d5 (n = 3 mice/group), and d14 (n = 4 mice/group except for TD n = 1 mouse) following the start of DSS treatment. There was a significant effect of DSS treatment over time and between diet groups as indicated (*P<0.05, two-way ANOVA with Bonferroni post-tests). (E) Histological scores of the distal colon of WT mice at d14 post DSS (n = 2–4 mice per group except for TD n = 1 mouse). CD was significantly different from PD (*P<0.05, one-way ANOVA with Tukey's post-tests). (F) BW change in mice fed CD, or PD the day before (−1) or 2d after (2) DSS treatment was initiated (n = 4–6 mice per group). All three dietary treatments were significantly different from each other (***P<0.001, two-way ANOVA). Data shown are the mean +/− SEM from one representative of three independent experiments. BW, body weight; CD, chow diet; DSS, dextran sodium sulfate; KO, knockout; PD, purified diet; Rag, recombinase-activating gene; TD, Teklad diet; WT, wild-type.
Figure 2.
The nutrient composition of the diets.
Figure 3.
Diet-mediated effects on fecal microflora.
(A) DGGE fingerprints of fecal DNA from the PD-fed (lanes 1–4) and TD-fed mice (lanes 5–8) (n = 4 mice/group). Data shown is one representative of three independent experiments. (B) Cluster analysis showing the similarity measurements of the DGGE banding patterns shown in panel A. (C) Metagenomic analysis showing the abundance of bacterial phyla present in feces from PD- and TD-fed mice (n = 2 mice/group). (D) Abundance of bacterial families in Firmicutes phylum present in feces from PD- and TD-fed mice (n = 2 mice/group). Significant difference was found in multiple bacterial phyla and families between PD and TD (***P<0.0001, Pearson Chi-Square Goodness of Fit test). DGGE, denaturing gradient gel electrophoresis; PD, purified diet; TD, Teklad diet.
Figure 4.
ABX treatments protected PD- and TD-fed mice from DSS-induced colitis.
WT mice were treated continuously with ABX and fed PD or TD for 2% DSS treatment. (A) Percent original BW of ABX treated PD- or TD-fed mice following the start of DSS treatment (n = 3–4 mice/group). Significant difference in BW change was found in PD versus TD, PD versus PD ABX, TD versus TD ABX, and PD ABX versus TD ABX (***P<0.001, two-way ANOVA). (B) Colonic length at d0 and d5 following the start of DSS (n = 3–4 mice/group and time point). There was a significant effect of DSS treatment over time and between groups as indicated (*P<0.05, **P<0.01, ***P<0.001, two-way ANOVA with Bonferroni post-tests). (C) Histological scores of distal colon at d5 post DSS (n = 3–4 mice/group). TD was significantly different from PD and TD ABX (**P<0.01, ***P<0.001, one-way ANOVA with Tukey's post-tests). Data is from one representative of three experiments. (D) Blood scores from PD- and TD-fed conventional or germ-free mice following 5d of DSS treatment (n = 3–6 mice/group). TD was significantly different from PD in conventionally raised mice (**P<0.01, two-way ANOVA with Bonferroni post-tests). Data shown are the mean +/− SEM. ABX, antibiotics; BW, body weight; DSS, dextran sodium sulfate; PD, purified diet; TD, Teklad diet; WT, wild-type.
Figure 5.
Diet-mediated effects on a primary and secondary infection with C. rodentium.
(A) Shedding of C. rodentium in the feces following a primary infection (n = 14 mice/group). (B) Shedding of C. rodentium in the feces following a secondary infection (n = 14 mice/group). Values are the mean +/− SEM. TD was significantly different from PD (*P<0.05, **P<0.01, ***P<0.001, two-way ANOVA with Bonferroni post-tests). (C) Fecal DNA was collected from PD-fed (lanes 1–4) and TD-fed mice (lanes 5–8) prior to secondary infection and (D) cluster analysis of the similarity of the DGGE banding patterns from panel C. CFU, colony forming units, DGGE, denaturing gradient gel electrophoresis; PD, purified diet; TD, Teklad diet.
Figure 6.
Diet-mediated effects on development of EAE in WT mice.
(A) Daily mean EAE scores of WT mice fed PD or TD (n = 11–13 mice/group). TD was significantly different from PD (**P<0.01, two-way ANOVA). (B) CDI of PD- or TD-fed mice. (C) The frequency of mice with individual EAE scores and the maximum EAE severity for each mouse. TD was significantly different from PD in CDI and maximum EAE scores (n = 11–13 mice/group, *P<0.05, **P<0.01, unpaired t test). (D) IFN-γ and IL-17 production (n = 4–10 mice/group). Values are the mean ± SEM from two independent experiments. CDI, Cumulative disease index; EAE, experimental autoimmune encephalomyelitis; ND, not detectable; PD, purified diet; TD, Teklad diet; WT, wild-type.