Fig 1.
Comparative analysis of the role of SPI1-T3SS effector molecules SopA, SopB, SipA, and SopE2 using triple mutants.
1-day-old C57BL/6 mice were orally infected with 100 CFU wild type (WT) (filled circles), isogenic quadruple sopABE2sipA mutant (open circles), or ΔsopBE2sipA (open inverted triangles), ΔsopAE2sipA (open triangles), ΔsopABE2 (open squares), or ΔsopABsipA (open diamonds) S. Typhimurium. Viable counts in (A) isolated gentamicin-treated enterocytes and (B) total liver tissue homogenate at 4 days post infection (p.i.). (C) Quantitative RT-PCR for Cxcl2 mRNA in total RNA prepared from enterocytes isolated at 4 days p.i.. Values were normalized to uninfected age-matched control animals (crosses). Individual values and the mean from at least two independent experiments are shown (n = 4–6 animals per group). (D) Immunostaining for Salmonella (red) in small intestinal tissue sections at 4 days p.i. with 100 CFU WT S. Typhimurium (WT), ΔsopABE2sipA quadruple mutant, or ΔsopBE2sipA, ΔsopAE2sipA, ΔsopABE2, or ΔsopABsipA triple mutant bacteria. Counterstaining with E-cadherin (green), WGA (white) and DAPI (blue). Bar, 5 μm. (E) Co-immunostaining for Salmonella WT, ΔsopABE2, or ΔsopABsipA (green) and LAMP1 (red) in small intestinal tissue sections at day 4 p.i.. Counterstaining with E-cadherin (white) and DAPI (blue). Bar, 5 μm. (F) Quantitative evaluation of the percentage of intraepithelial S. Typhimurium associated with LAMP1 staining. Three tissue sections per infected neonate (n = 3) were analyzed at day 4 p.i.. Results represent the mean ± SD. (G) Co-immumostaining for Salmonella WT, ΔsopABE2 or ΔsopABsipA (red) and the GFP-expressing SPI2 reporter (pM973; green) in small intestinal tissue sections at day 4 p.i.. Counterstaining with E-cadherin (white) and DAPI (blue). Bar, 5μm. (H) Quantitative analysis of the percentage of intraepithelial S. Typhimurium expressing the SPI2 reporter. All microcolonies from three tissue sections per infected neonate were analyzed (n = 3) at day 4 p.i.. Results represent the mean ± SD.
Fig 2.
The influence of MyD88-dependent innate immune signaling on intraepithelial microcolony formation.
1-day-old MyD88+/+ and MyD88-/- mice were orally infected with 100 CFU S. Typhimurium WT. (A) 4 days after infection, small intestinal tissues were collected and analyzed by immunostaining. Three representative images showing S. Typhimurium (red) forming intraepithelial microcolonies in MyD88-/- mice. Counterstaining with E-cadherin (green), WGA (white) and DAPI (blue). Bar, 10 μm. For wild type controls see Fig 1D. (B) Quantitative evaluation of the number of intraepithelial microcolonies per villus in MyD88+/+ and MyD88-/- mice at 4 days p.i.. S. Typhimurium microcolonies were quantified in 20–30 villi per animal (n = 6–8). Results represent the mean ± SD. (C) Transmission electron microscopy (TEM) images of intraepithelial Salmonella in MyD88+/+ (left panel) and MyD88-/- mice (right panel). Asterisks highlight bacteria. Bar, 2 μm. (D) Co-immunostaining for Salmonella (green) and LAMP1 (red) in small intestinal tissue sections at day 4 p.i.. Counterstaining with E-cadherin (white) and DAPI (blue). Bar, 5 μm. (E) Quantitative evaluation of the percentage of intraepithelial S. Typhimurium associated with LAMP1 staining. Four neonates were analyzed at day 4 p.i.. Results represent the mean ± SD. (F) Co-immumostaining for Salmonella (red) and the GFP expressing SPI2 reporter (pM973, green) in small intestinal tissue sections at day 4 p.i.. Counterstaining with E-cadherin (white) and DAPI (blue). Bar, 5 μm. (G) Quantitative analysis of the percentage of intraepithelial S. Typhimurium expressing the SPI2 reporter. Microcolonies from tissue sections from four neonates were analyzed at day 4 p.i.. Results represent the mean ± SD.
Fig 3.
The redundant role of SipA and SopE2 for enterocyte invasion.
(A-C) 1-day-old C57BL/6 mice were orally infected with 100 CFU wild type (WT) (filled circles) or isogenic sopE2sipA mutant S. Typhimurium (filled triangles). Viable counts in (A) isolated gentamicin-treated enterocytes and (B) total liver tissue homogenate at 4 days p.i.. (C) Quantitative RT-PCR for Cxcl2 mRNA in total RNA prepared from enterocytes isolated at 4 days p.i.. Values were normalized to uninfected age-matched control animals (crosses). Individual values and the mean from at least two independent experiments are shown (n = 4–6 animals per group). The data for uninfected control animals and WT Salmonella infected mice are identical to Fig 1A–1C. (D-F) 1-day-old C57BL/6 mice were orally infected with 100 CFU WT (filled circles), isogenic sipA mutant (open squares), complemented ΔsipA psipA (filled squares), isogenic sopE2 mutant (open diamonds), or complemented ΔsopE2 psopE2 (filled diamonds) Salmonella. Viable counts in (D) isolated gentamicin-treated enterocytes and (E) total liver tissue homogenate at 4 days p.i.. (F) Quantitative RT-PCR for Cxcl2 mRNA in total RNA prepared from enterocytes isolated at 4 days p.i.. Values were normalized to uninfected age-matched control animals (crosses). Individual values and the mean from at least two independent experiments are shown (n = 3–5 animals per group). The data for uninfected control animals and Salmonella WT infected mice are identical to Fig 1A–1C. (G) Immunostaining for S. Typhimurium (red) in small intestinal tissue sections at 4 days p.i. with 100 CFU ΔsopE2sipA, ΔsipA, or ΔsopE2 S. Typhimurium Salmonella. Counterstaining with E-cadherin (green), WGA (white) and DAPI (blue). Bar, 5 μm. (H) Co-immunostaining for Salmonella ΔsipA, ΔsopE2 (green) and LAMP1 (red) in small intestinal tissue sections at day 4 p.i.. Counterstaining with E-cadherin (white) and DAPI (blue). Bar, 5 μm. (I) Quantitative evaluation of the percentage of intraepithelial S. Typhimurium associated with LAMP1 staining. All microcolonies from three tissue sections per infected neonate were analyzed (n = 5) at day 4 p.i.. Results represent the mean ± SD. (J) Co-immunostaining for Salmonella ΔsipA or ΔsopE2 (red) and the GFP-expressing SPI2 reporter (pM973; green) in small intestinal tissue sections at day 4 p.i.. Counterstaining with E-cadherin (white) and DAPI (blue). Bar, 5μm. (K) Quantitative analysis of the percentage of intraepithelial S. Typhimurium expressing the SPI2 reporter. All microcolonies from three tissue sections per infected neonate were analyzed (n = 3) at day 4 p.i.. Results represent the mean ± SD.
Fig 4.
Analysis of sopBE2 and sopAE2 double mutant S. Typhimurium.
(A-C) 1-day-old C57BL/6 mice were orally infected with 100 CFU wild type (WT) (filled circles), isogenic ΔsopBE2 (inverted open triangles), or ΔsopAE2 (open triangles) S. Typhimurium. Viable counts in (A) isolated gentamicin-treated enterocytes and (B) total liver tissue homogenate at 4 days p.i.. (C) Quantitative RT-PCR for Cxcl2 mRNA in total RNA prepared from enterocytes isolated at 4 days p.i.. Values were normalized to uninfected age-matched control animals (crosses). Individual values and the mean from at least two independent experiments are shown (n = 3–6 animals per group). The data for uninfected control animals and Salmonella WT infected mice are identical to Fig 1A–1C. (D) Immunostaining for Salmonella (red) in small intestinal tissue sections at 4 days p.i. with 100 CFU ΔsopE2, ΔsopBE2, or ΔsopAE2 S. Typhimurium. Counterstaining with E-cadherin (green), WGA (white) and DAPI (blue). Bar, 5 μm. (E) Percentage of epithelial cells positive for single bacteria or microcolonies (>1 intraepithelial bacterium) at 4 days p.i. with ΔsopBE2 or ΔsopAE2 S. Typhimurium. 30 Salmonella-positive epithelial cells per infected neonate (n = 8–13) were analyzed. Results represent the mean ± SD. (F) Co-immunostaining for Salmonella ΔsopBE2 and ΔsopAE2 (green) and LAMP1 (red) in small intestinal tissue sections at day 4 p.i.. Counterstaining with E-cadherin (white) and DAPI (blue). Bar, 5 μm. (G) Quantitative evaluation of the percentage of intraepithelial ΔsopBE2 and ΔsopAE2 S. Typhimurium associated with LAMP1 staining. All microcolonies from three tissue sections per infected neonate were analyzed (n = 3–4) at day 4 p.i.. Results represent the mean ± SD. (H) Co-immunostaining for ΔsopBE2 and ΔsopAE2 Salmonella (red) and the GFP-expressing SPI2 reporter (pM973; green) in small intestinal tissue sections at day 4 p.i.. Counterstaining with E-cadherin (white) and DAPI (blue). Bar, 5 μm. (I) Quantitative analysis of the percentage of intraepithelial S. Typhimurium expressing the SPI2 reporter. All microcolonies from three tissue sections per infected neonate were analyzed (n = 3–4) at day 4 p.i.. Results represent the mean ± SD.
Fig 5.
The role of SopB in the interaction between Salmonella and the epithelium.
1-day-old C57BL/6 mice were orally infected with 100 CFU wild type (WT) (filled circles), isogenic sopB mutant (filled triangles), or psopB-complemented ΔsopB (open triangles) S. Typhimurium. Viable counts in (A) isolated gentamicin-treated enterocytes, (B) total MLN and (C) total liver tissue homogenate at 2 days p.i.. (D) Quantitative RT-PCR for Cxcl2 mRNA in total RNA prepared from enterocytes isolated at 2 days p.i.. Values were normalized to uninfected age-matched control animals (crosses). Individual values and the mean from at least two independent experiments are shown (n = 3–5 animals per group). (E) Quantitative analysis of the number of cleaved caspase 3- and (F) cleaved caspase 8 positive cells per 200 times magnification image field. Positive cells from 20 image fields from one section were analyzed per infected neonate (n = 3–6) at day 3 p.i.. Results represent the mean ± SD. (G) Immunostaining for S. Typhimurium (red) in small intestinal tissue sections at 3 days p.i. with 100 CFU WT and ΔsopB S. Typhimurium. Counterstaining with E-cadherin (green), WGA (white) and DAPI (blue). Bar, 5 μm. (H) Co-immunostaining for ΔsopB S. Typhimurium (green) and LAMP1 (red) in small intestinal tissue sections at day 3 p.i.. Counterstaining with E-cadherin (white) and DAPI (blue). Bar, 5 μm. (I) Quantitative evaluation of the percentage of intraepithelial S. Typhimurium associated with LAMP1 staining. All microcolonies from three tissue sections per infected neonate were analyzed (n = 3) at day 3 p.i.. Results represent the mean ± SD. (J) Co-immumostaining for ΔsopB S. Typhimurium (red) and the GFP expressing SPI2 reporter (pM973; green) in small intestinal tissue sections at day 3 p.i.. Counterstaining with E-cadherin (white) and DAPI (blue). Bar, 5 μm. (K) Quantitative analysis of the percentage of intraepithelial S. Typhimurium expressing the SPI2 reporter. All microcolonies from three tissue sections per infected neonate were analyzed (n = 3) at day 3 p.i.. Results represent the mean ± SD.
Fig 6.
The role of SipA for intraepithelial microcolony formation.
(A) Mucosal barrier integrity tested by serum quantification 4 hours after oral administration of FITC labeled-4kDa dextran. 1-day-old C57BL/6 mice were left untreated (crosses) or infected with WT (filled circles) or ΔsopABE2 (open squares) S. Typhimurium. FITC labeled-4 kDa dextran was quantified in serum at day 3 p.i. as indicated. (B and C) Flow cytometric analysis of lamina propria immune cells. 4-day-old mice were left untreated or orally infected with 100 CFU WT or ΔsopABE2 S. Typhimurium and total small intestinal leukocytes were analyzed by flow cytometry at day 3 p.i.. (B) Monocytes (Ly6ChiLy6G-CD11b+ MHCIIlo/-CD45+DAPI-) and (C) neutrophils (Ly6G+Ly6CintCD11b+ MHCIIlo/-CD45+DAPI-) are depicted as percentage of CD45+ cells in non-infected (crosses), WT (filled circles) and ΔsopABE2 Salmonella (open squares). The results represent the mean values from at least two independent experiments (n = 4–6 per group). (D) Immunostaining for Salmonella in small intestinal tissue sections at 4 days after co-infection with 100 CFU GFP-expressing WT (yellow) and ΔsopABsipA (red) S. Typhimurium. WT Salmonella appear in yellow due to simultaneous staining for O4/O5 antigen (red) and GFP (green). ΔsopABsipA Salmonella appear in red due to simultaneous staining for O4/O5 antigen (red). Counterstaining with E-cadherin (white) and DAPI (blue). Bar, 10 μm. (E-J) 1-day-old C57BL/6 mice were orally infected with 100 CFU WT (filled circles), ΔsipA (open diamonds), ΔsipA complemented with psipAK635A E637W (half-filled diamonds), ΔsipA complemented with psipAD434A (filled squares), or ΔsipA complemented with psipAD434A K635A E637W (filled triangles) S. Typhimurium. Viable counts in (E) isolated gentamicin-treated enterocytes and (F) total liver tissue homogenate at 4 days p.i.. (G) Quantitative RT-PCR for Cxcl2 mRNA in total RNA prepared from enterocytes isolated at 4 days p.i.. Values were normalized to uninfected age-matched control animals (crosses). Individual values and the mean from at least two independent experiments are shown (n = 4–7 animals per group). The data for WT Salmonella infected mice and uninfected control animals are identical to Fig 1A–1C. (H) Immunostaining for Salmonella (red) in small intestinal tissue sections at 4 days p.i. with 100 CFU ΔsipA, ΔsipA complemented with psipAK635A E637W, ΔsipA complemented with psipAD434A, or ΔsipA complemented with psipAD434A K635A E637W S. Typhimurium. Counterstaining with E-cadherin (green), WGA (white) and DAPI (blue). Bar, 5 μm. (I) Co-immunostaining for LAMP1 (red) and ΔsipA complemented with psipAK635A E637W, ΔsipA complemented with psipAD434A, or ΔsipA complemented with psipAD434A K635A E637W S. Typhimurium (green) in small intestinal tissue sections at day 4 p.i.. Counterstaining with E-cadherin (white) and DAPI (blue). Bar, 5 μm. (J) Quantitative evaluation of the percentage of intraepithelial S. Typhimurium associated with LAMP1 staining. All microcolonies from three tissue sections per infected neonate were analyzed (n = 3) at day 4 p.i.. Results represent the mean ± SD.
Fig 7.
Graphical illustration of the role of the SPI1-T3SS effectors SopA, SopB, SopE2 and SipA during enterocyte invasion and intraepithelial proliferation in vivo.
(1) SipA (red dots) promotes the early recruitment of PMNs and causes barrier disruption and disease progression. This has been reported to occur via stimulation of the epithelial surface molecule p53-effector related to PMP-22 (PERP) and activation of the chemotactic eicosanoid hepoxillin A3 (HXA3) [68, 89]. This effect appears to be invasion-independent and strongly enhanced in the absence of SopA, SopB and SopE2 suggesting that these effectors exert regulatory functions. (2) Among the studied effector molecules, expression of SipA (red dots), SopE2 (green dots), or SopE (not shown here) alone is sufficient to facilitate enterocyte invasion. Intraepithelial Salmonella then reside within a LAMP1 negative endosomal compartment and fail to proliferate or express SPI2 encoded genes. (3a) SipA together with SopE2 or SipA together with SopB (blue dots) facilitate the recruitment of LAMP1 (yellow membrane) from the Golgi apparatus (GA) and the generation of a replicative compartment with intraepithelial bacterial proliferation and expression of SPI2 effector molecules. (3b) Enterocyte invasion or, alternatively, penetration of the epithelial barrier via innate stimulation and signaling through MyD88 induce expression of the chemokines Cxcl2 and Cxcl5 in the epithelium. (3c) SopB appears to directly or indirectly inhibit caspase 3 and caspase 8 mediated epithelial cell apoptosis. GA, golgi apparatus; HXA3, hepoxilin A3; LAMP1, lysosomal-associated membrane protein 1; PERP, p53-effector related to PMP-22; SPI2, Salmonella pathogenicity island 2.