Fig 1.
EBSS and wortmannin inversely affect the cytosolic population.
(A) Schematic of the experimental design employed to differentially modulate host cell autophagy. (B) Epithelial cells seeded on glass coverslips were pretreated with EBSS or 100 nM WTM as depicted in (A) and infected with S. Typhimurium wild-type bacteria harboring pFPV-mCherry for plasmid-borne, constitutive expression of mCherry. The proportion of bacteria accessible to anti-Salmonella LPS antibodies delivered to the mammalian cell cytosol was determined by digitonin permeabilization assay and fluorescence microscopy. n≥3 independent experiments. (C) Epithelial cells seeded on coverslips were pretreated as in (A) and infected with mCherry-S. Typhimurium (wild-type glmS::Ptrc-mCherryST::FRT bacteria constitutively expressing a chromosomal copy of mCherry) harboring the fluorescent reporter plasmid, PuhpT-gfpova (pNF101). At the indicated times, cells were fixed and the proportion of infected cells containing cytosolic bacteria (GFP-positive) was scored by fluorescence microscopy. The mean from each experiment is represented as a large dot (n = 3). (D) Cells seeded on coverslips were pretreated as in (A) and infected with wild-type bacteria harboring pFPV-mCherry. The number of bacteria in each infected cell was scored by fluorescence microscopy. Cells with ≥100 bacteria contain cytosolic S. Typhimurium. Each small dot represents one infected cell. n = 2 (1 h) or 3 (8 h) experiments. (E) Epithelial cells were infected with wild-type bacteria and the number of intracellular bacteria at 1 h and 8 h p.i. was determined by gentamicin protection assay. Fold-replication is CFUs at 8 h/1 h. The mean from each experiment is represented as a large dot (n = 3). For all panels, control = untreated cells; EBSS = Earle’s balanced salt solution treatment; WTM = wortmannin treatment. Asterisks indicate data significantly different from control (ANOVA with Dunnett’s post-hoc test, p<0.05).
Fig 2.
(A) RNA-seq results displayed as a volcano plot. Statistical significance is represented on the y-axis and magnitude of change on the x-axis. Statistically significant mRNAs and sRNAs are shown in blue dots for “up vacuole” and red dots for “up cytosol” categories. mRNAs/sRNAs that did not pass statistical significance are indicated by grey dots. Select genes of interest are indicated. (B) Pathway enrichment analysis. Top panels indicate regulons generated from transcriptional profiling of key regulatory mutants, while bottom panels display enriched gene sets generated from KEGG, GO, or from RNA-seq profiling of S. Typhimurium under 22 infection-relevant in vitro growth conditions. All displayed enrichments are significant (adjusted p-value <0.05) according to Fisher’s exact test with Benjamini-Hochberg correction for multiple comparisons. The number of genes in each category is shown in log2 scale. Only a selection of non-redundant pathways is presented; the complete list is available as S2 Dataset.
Fig 3.
Iron-associated genes are induced in the epithelial cytosol.
(A) Epithelial cells seeded on coverslips were infected wild-type mCherry-S. Typhimurium harboring gfpmut3 transcriptional reporters. At 8 h p.i., cells were fixed and stained with Hoechst 33342 to detect DNA. Representative confocal microscopy images show induction of iroB, nrdH, sitA, sufA, yjjZ and STnc3080 promoters in cytosolic bacteria. Green = transcriptional reporter, red = S. Typhimurium, blue = DNA. Scale bars are 10 μm. (B) Quantification of the mean fluorescence intensity (MFI) of GFP signal by fluorescence microscopy and ImageJ. Bacteria were designated as being cytosolic (Cyto) or vacuolar (Vac) if residing within cells with ≥100 bacteria or 2–40 bacteria, respectively. t0 represents the infection inoculum i.e. bacteria grown to late log phase in LB-Miller broth. Small dots represent individual bacteria; large dots indicate the mean of each experiment; horizontal bars indicate the average of two experiments. Acquisition parameters (exposure time and gain) were set up using PsitA-gfpmut3 (the highest GFP intensity) and these same parameters were applied throughout. Dashed lines indicate the range of background fluorescence in the GFP channel measured for mCherry-S. Typhimurium (no reporter plasmid).
Fig 4.
Manipulation of cellular iron levels primarily affects cytosolic bacteria.
(A) Epithelial cells seeded on coverslips were left untreated (control) or treated overnight in growth media supplemented with 10 μM, 30 μM, 100 μM or 300 μM ferric ammonium citrate (FAC). Cells were infected with wild-type mCherry-S. Typhimurium harboring PiroN-gfpmut3, PsitA-gfpmut3 or PyjjZ-gfpmut3 reporters, fixed at 8 h p.i. and the MFI of bacterial GFP signal was quantified by fluorescence microscopy and Image J. Bacteria were designated as being cytosolic (Cyto) or vacuolar (Vac) if residing within cells with ≥100 bacteria or 2–40 bacteria, respectively. Small dots represent individual bacteria; large dots indicate the mean of each experiment; horizontal bars indicate the average of two experiments. Acquisition parameters (exposure time and gain) were set-up using PiroN-gfpmut3 (the highest GFP intensity) and these same parameters were applied throughout. Dashed lines indicate the range of background fluorescence in the GFP channel measured for mCherry-S. Typhimurium (no reporter plasmid). Asterisks indicate data significantly different from control (Kruskal-Wallis test). (B) Epithelial cells seeded on coverslips were left untreated (control) or treated overnight with growth media containing 200 μM 2,2’-dipyridyl (DPI), a metal-chelating compound, 200 μM DPI and 200 μM ferric ammonium citrate (DPI+FAC), or 200 μM FAC alone (FAC). Infection strains and quantification of GFP signal were as described for (A). Asterisks indicate data significantly different from control (Kruskal-Wallis test). (C) Cells were treated as in (B) and infected with wild-type bacteria. The number of intracellular bacteria was quantified by gentamicin protection assay at 1 h and 8 h p.i. Fold-replication is CFUs at 8 h/1 h. The mean from each experiment is represented as a large dot (n = 3). Asterisks indicate data significantly different from control (ANOVA with Dunnett’s post-hoc test). (D) Cells were treated as in (B) and infected with wild-type bacteria. The proportion of cytosolic bacteria was quantified by CHQ resistance assay at 7 h p.i. The mean from each experiment is represented as a large dot (n≥3). Asterisk indicates data significantly different from control (ANOVA with Dunnett’s post-hoc test). (E) Cells were treated as in (B) and infected with mCherry-S. Typhimurium harboring a plasmid-borne reporter of cytosolic access, pNF101, and fixed at 8 h p.i. Left panel: the proportion of infected cells containing only cytosolic (all bacteria are GFP+, mCherry+), only vacuolar (all bacteria are GFP-, mCherry+) or a mixed population (Cyto&Vac) of bacteria was blindly scored by fluorescence microscopy. n≥5 independent experiments. Right panel: the number of cytosolic or vacuolar bacteria in each cell was blindly scored. Small dots represent individual bacteria; large dots indicate the mean of each experiment; horizontal bars indicate the average of two experiments.
Fig 5.
Zn2+ and Mg2+ are limiting in the vacuole, not the cytosol.
Upper panels: Epithelial cells seeded on coverslips were infected with mCherry-S. Typhimurium harboring PzinT-gfpmut3 or PmgtC-gfpmut3 transcriptional reporters. At 8 h p.i., cells were fixed and stained with Hoechst 33342 to detect DNA. Representative confocal microscopy images show induction of zinT (upper panel) and mgtC (lower panel) promoters in vacuolar bacteria. Green = transcriptional reporter, red = S. Typhimurium, blue = DNA. Scale bars are 10 μm. Lower panels: Quantification of the MFI of GFP signal by fluorescence microscopy and ImageJ. Bacteria were designated as being cytosolic (Cyto) or vacuolar (Vac) if residing within cells with ≥100 bacteria or 2–40 bacteria, respectively. t0 represents the infection inoculum i.e. bacteria grown to late log phase in LB-Miller broth. Small dots represent individual bacteria; large dots indicate the mean of each experiment; horizontal bars indicate the average of two experiments. Acquisition parameters (exposure time and gain) were set up using PzinT-gfpmut3 (the highest GFP intensity) and these same parameters were applied throughout. Dashed lines indicate the range of background fluorescence in the GFP channel measured for mCherry-S. Typhimurium (no reporter plasmid).
Fig 6.
SPI1-associated genes are up-regulated in the epithelial cytosol.
(A) Epithelial cells seeded on coverslips were infected with mCherry-S. Typhimurium harboring gfpmut3 transcriptional reporters. At 8 h p.i., cells were fixed and stained with Hoechst 33342 to detect DNA. Representative confocal microscopy images show induction of sicA, sopE2, sopF, siiA and lpxR promoters in cytosolic bacteria. PipB is a type III effector translocated by T3SS2 and the PpipB-gfpmut3 reporter served as a control for vacuole-specific gene induction. Green = transcriptional reporter, red = S. Typhimurium, blue = DNA. Scale bars are 10 μm. (B) Quantification of the MFI of GFP signal by fluorescence microscopy and ImageJ. Bacteria were designated as being cytosolic (Cyto) or vacuolar (Vac) if residing within cells with ≥100 bacteria or 2–40 bacteria, respectively. t0 represents the infection inoculum i.e. bacteria grown to late log phase in LB-Miller broth. Small dots represent individual bacteria; large dots indicate the mean of each experiment; horizontal bars indicate the average of 2–3 experiments. Acquisition parameters (exposure time and gain) were set up using PsicA-gfpmut3 (the highest GFP intensity) and these same parameters were applied throughout. Dashed lines indicate the range of background fluorescence in the GFP channel measured for mCherry-S. Typhimurium (no reporter plasmid).
Fig 7.
A subset of “up cytosol” genes is required for optimal proliferation in the epithelial cell cytosol.
(A) Epithelial cells were infected with wild-type (WT) bacteria or the indicated gene deletion mutants and the proportion of cytosolic bacteria at 7 h p.i. was quantified using the CHQ resistance assay. n≥4 independent experiments for each strain. Asterisks indicate data significantly different from WT (ANOVA with Dunnett’s post-hoc test). (B) Genetic complementation. Epithelial cells were infected with WT bacteria or deletion mutants harboring an empty vector (-), pWSK29 or pWKS30, or the respective vector encoding the indicated gene(s), except for the ΔfepB mutant, which was complemented with a chromosomal copy of fepB at the glmS site. The proportion of cytosolic bacteria at 7 h p.i. was quantified using the CHQ resistance assay. Large dots indicate the mean of each experiment (n≥3). Asterisks indicate data significantly different from WT (ANOVA with Dunnett’s post-hoc test). (C) Epithelial cells were infected with WT bacteria or the indicated gene deletion mutants and the proportion of cytosolic bacteria at 90 min p.i. was quantified using the CHQ resistance assay. Large dots indicate the mean of each experiment (n≥3). (D) Epithelial cells seeded on coverslips were infected with WT bacteria or gene deletion mutants harboring pCHAR-Duo(ASV), a plasmid-borne dual reporter–constitutive mCherry expression is driven by the synthetic ProB promoter and gfpmut3.1(ASV) (encoding for destabilized GFP) is under the control of the glucose-6-phosphate responsive uhpT promoter from S. Typhimurium. The number of cytosolic (GFP+, mCherry+) and vacuolar (GFP-, mCherry+) bacteria in each infected cell was blindly scored by fluorescence microscopy. Small dots represent individual bacteria; large dots indicate the mean of each experiment; horizontal bars indicate the average of 3 experiments. Asterisk indicates data significantly different from WT (Kruskal-Wallis test). (E) Data from (D) was reanalyzed to determine the percentage of infected cells containing >50 cytosolic bacteria. Large dots indicate the mean of each experiment. Asterisks indicate data significantly different from WT (ANOVA with Dunnett’s post-hoc test).
Fig 8.
Adaptation to the epithelial cytosol requires extensive transcriptional reprogramming by S. Typhimurium.
SCV-resident bacteria are translocating type III effectors via T3SS2, devoid of flagella and exposed to limiting zinc and magnesium concentrations. By contrast, cytosolic bacteria highly express genes implicated in iron uptake/storage, manganese and sugar transport suggesting that accumulation of these nutrients is important for bacterial proliferation in this compartment. Up-regulation of a subset of genes associated with oxidative stress resistance further indicates that bacteria are exposed to reactive oxygen species. Cytosolic bacteria are also T3SS1-active, flagellated and decorated with the MUC1-binding adhesin, SiiE, and therefore primed to adhere to, and enter, naïve cells upon their release.
Table 1.
Cytosol signature genes.