Fig 1.
A functional screening reveals EspF as a novel virulence protein for CR virulence in Il22-/- mice.
A. Kaplan-Meier analysis of the survival rates in Il22-/- mice inoculated with wild-type (WT) CR or each strain in the sub-library C consisting of 96 mutants (Mut-1 to Mut-96). B. Kaplan-Meier analysis of the survival rate in Il22-/- mice inoculated with WT or Mut-19 CR. C. Schematics of normal or Tn5-interrupted espF genes and EspF protein expression in WT and Mut-19 CR, respectively. D-E. Weight loss (D) and clinical scores (E) of Il22-/- mice at indicated days post inoculation (dpi) with WT or Mut-19 CR. F. Il22-/- mice were inoculated with phosphate-buffered saline (PBS) or wild-type (WT) CR. At 7 dpi, WT CR was derived from the infected colon epithelia or LB culture and immunoblotted (IB) for EspF, with DnaK as a loading control. **** p < 0.0001 by Gehan-Breslow-Wilcoxon tests (B) and by Student’s t tests (D-E).
Fig 2.
EspF is critical for CR infection-induced morbidity and mortality in Il22-/- mice.
A. Upper, diagram shows CR espF gene and PCR products amplified using a common forward primer and pair 1 reverse primer (in red), or pair 2 reverse primer (in green). Bottom, representative image of amplified PCR products using the indicated primer pairs from the CR strains as indicated. B. Growth curves of WT and ΔespF CR in LB medium, at 1: 150 dilutions from the overnight cultures. C. Colon epithelial cells (CEC) derived Il22-/- mouse were infected in suspension with PBS, WT or ΔespF CR at 100 MOI for 3h, followed by immunofluorescence staining for CR-specific antisera, with nuclei counterstained by DAPI. The numbers of CR attached to CECs were quantified and normalized to the perimeter of individual CEC. D, F. Weight loss (D) and clinical scores (F) of Il22-/- mice at indicated periods post inoculation with WT or ΔespF CR. E. Kaplan-Meier analysis of the survival rate in Il22-/- mice inoculated with WT or ΔespF CR. G. Colonization kinetics of WT or ΔespF CR in Il22-/- mice, inoculated with 2 × 109 CFU of CR. H-I. Weight loss (H) and survival rate (I) of Il22-/- mice inoculated with WT, ΔespF, or ΔespF::EspF CR. ns, not significant; * p < 0.05, ** p < 0.01, and **** p < 0.0001 by Student’s t tests (C-D, F), by Gehan-Breslow-Wilcoxon tests (E), and by Long-rank test with Bonferroni’s multiple comparison adjustment (I).
Fig 3.
CR infection disrupts tight junction integrity in the colon in an EspF-dependent manner.
A. Colon length of Il22-/- mice, inoculated with phosphate-buffered saline (PBS) (n = 4), wild-type (WT, n = 3) or ΔespF (n = 5) CR, and euthanized at 7 days post inoculation (dpi). B. Hematoxylin and eosin staining of colons derived from Il22-/- mice infected as in (A). Scale bars, 200 μm. C. The histopathology scores of colon sections derived from Il22-/- mice infected as in (B). Shown are mean ± s.e.m of 10 random fields from two independent experiments. D. The CR burden in the liver (left) and the spleen (right) derived from Il22-/- mice, infected as in (A) at 7 dpi, were quantified. E. Serum levels of FITC-dextran, assessed at 4 h post oral administration, in Il22-/- mice infected with PBS (n = 5), WT CR (n = 9), or ΔespF CR (n = 4) at 7 dpi. F. Immunofluorescence micrographs of claudin-3 in the colons collected from Il22-/- mice, infected as in (A) at 7 dpi, with nuclei counterstained by DAPI. Scale bars, 100 μm. G. Il22-/- mice were infected as in (F), and colon epithelial cell lysates were derived and IB for indicated claudins, with β-actin as a loading control. H. The protein levels of indicated claudins, normalized to β-actin and PBS controls, were quantified by ImageJ software from three independent experiments. ns, not significant; * p < 0.05, ** p < 0.01, *** p < 0.001, and **** p < 0.0001 with one-way analysis of variance, followed by Bonferroni’s multiple comparison tests (A, D-E, H), and with Kruskal-Wallis test followed by Dunn’s multiple comparisons tests (C).
Fig 4.
EspF plays a specific role for CR infection-caused lethality in Il22-/- mice.
A. Quantitative PCR was used to determine mRNA levels of ler, espF, espG, map, and nleA relative to 16S rRNA in the CR collected from LB culture or Il22-/- mice at 7 days post inoculation (dpi) with wild-type Citrobacter rodentium (CR). Shown are mean ± SEM from 5 inoculated Il22-/- mice, with mRNA levels in CR derived from LB culture set a 1. B. Growth curves of indicated CR strains in LB medium, at 1:150 dilutions from the overnight cultures. C. The colony formation units (CFU) of live CR derived from fecal samples of Il22-/- mice, inoculated with 2 × 109 CFU of indicated CR strains, at 7 dpi. D-E. Weight loss (D) and survival rate (E) of Il22-/- mice inoculated with WT (n = 4), ΔespG (n = 5), Δmap (n = 5), or ΔespF (n = 5) CR. ns, not significant, * p < 0.05, ** p < 0.01, and **** p < 0.001 by Student’s t tests (A) and with one-way analysis of variance, followed by Bonferroni’s multiple comparison tests (C).