Table 1.
Genes with highest fold change in exponential compared to stationary phase. Genes with >10 fold enrichment are shown, ribosomal genes are highlighted with bolded text. rpsJ/S10 is highlighted with a black box.
Fig 1.
Exponential phase cells express high levels of ribosomal protein genes.
RNA-seq was performed on exponential and stationary phase WT Y. pseudotuberculosis cells. The DESeq2 method of pairwise comparisons was used to determine significant differences in transcript levels. KEGG pathway analyses were utilized to determine the biological pathways overrepresented in (A) exponential and (B) stationary phase cells.
Fig 2.
Ribosomal reporter expression is heightened in exponential phase cells.
(A) Growth curve of WT and PS10::gfp-ssrA Y. pseudotuberculosis strains at 26o C. Absorbance (A600nm) detected by plate reader at the indicated timepoints (hours, h). Four biological replicates shown with mean and standard deviation. (B) Single cell fluorescence microscopy used to quantify the fold change in fluorescence of the PS10::gfp-ssrA strain at the indicated timepoints, sampled from cultures in (A). Fold change in signal is relative to the mean GFP fluorescence value at 0h, which is shown as a dotted line at a value of 1. Horizontal lines: median values. Data represents 3 biological replicates for each condition. (C-D) qRT-PCR to detect S10, gfp, and 16S transcript levels during culture of the PS10::gfp-ssrA strain at (C) 26o C or (D) 37o C. RNA was isolated at the indicated timepoints. Data represents 6 biological replicates. (E) Representative images used to quantify single-cell fluorescence corresponding with 0h and 4h timepoints in panel B. Statistics: (B) Kruskal-Wallis one-way ANOVA with Dunn’s post-test, ****p < .0001.
Fig 3.
Multiple bacterial subpopulations are present within splenic microcolonies.
C57BL/6 mice were inoculated intravenously with the mCherry+ PS10::gfp-ssrA (constitutive mCherry), yopE::mCherry PS10::gfp-ssrA, or yopE::mCherry gfp-ssrA (constitutive GFP) strains. (A) CFU/spleen was quantified at days 2 and 3 post-inoculation. Dots: individual mice. (B) Microcolony areas from spleens were quantified by fluorescence microscopy at the indicated timepoints. Dots: individual microcolonies, representing all bacteria visualized within tissues (including single cells). (C) Representative images of microcolonies: day 2 and day 3 mCherry+ PS10::gfp-ssrA. (D) Reporter expression (ratio GFP/mCherry signal intensity) was quantified at the indicated timepoints and spatial locations (centroid: average of 4 measurements clustered at the geometric centroid of each microcolony, periphery: average of 4 measurements taken every 90o around the microcolony periphery). Horizontal dotted line represents a value of 1, equal amounts of GFP and mCherry signal. (E) Linear regression data indicating slope and R2 value of best fit lines for S10 centroid values compared to microcolony areas at the indicated timepoints. (F) Representative images of microcolonies: day 2 and day 3 yopE::mCherry PS10::gfp-ssrA. (G, H) Linear regression data indicating slope and R2 value of best fit lines for GFP compared to mCherry signal of microcolonies at the indicated spatial locations at (G) day 2 and (H) day 3. Horizontal dotted line at y = 250 separates low and high GFP signal. (H) Microcolonies with low levels of yopE::mCherry signal (<100) were analyzed separately (open circles). Dots represent individual mice (A) or individual microcolonies (all other panels). Statistics: (A-B) Mann-Whitney; (D) Wilcoxon matched-pairs; ***p < 0.001, *p < .05, ns: not-significant. (E, G, H) Significantly non-zero slope indicates correlation between values.
Fig 4.
T3SS induction slows growth, but cells can support high levels of S10 expression.
WT and yopE::mCherry PS10::gfp-ssrA strains were cultured at 37o C in the presence (+, T3SS-induced) or absence (-) of MgOx for the indicated timepoints (hours, h). (A) Growth curve of strains with and without MgOx. Absorbance (A600nm) detected by plate reader at the indicated timepoints. Statistics compare the yopE::mCherry PS10::gfp-ssrA strain in the presence (+) or absence (-) of MgOx. Mean and standard deviation are shown. (B) Representative fluorescence microscopy images of bacteria from (A) immobilized on 1% agarose pads. Fold change in (C) mCherry (yopE::mCherry) reporter signal and (D) GFP (PS10::gfp-ssrA) reporter signal in the absence (-) or presence (+) of MgOx. Values quantified in individual bacteria by fluorescence microscopy. Thresholding was used to select individual cell events, and single cell fluorescence was normalized to the average fluorescent value at 0h (value of 1, represented by dotted line). Each dot: individual cell, horizontal lines: median values. (E) Quantification of single cell bacterial areas (µm2) from samples in panels (C) and (D). Horizontal lines: median values. (F) Correlation plot of fold change in single cell mCherry and GFP fluorescence from bacteria in panels (C-E) cultured in the presence (+) of MgOx. Black arrows denote trends in fluorescence across timepoints. (G) Linear regression data indicating slope, R2, and significance of the lines of best fit shown in (F). All data represent 3 biological replicates for each strain and condition in this figure. Statistics: (A) Two-way ANOVA with Tukey’s multiple comparison test; (C-E) Kruskal Wallis one-way ANOVA with Dunn’s post-test; ****p < 0.0001, **p < .01, ns: not-significant. (F,G) Significantly non-zero slope indicates correlation between values.
Fig 5.
Lack of T3SS expression was associated with lower S10 expression.
WT PS10::gfp-ssrA, and P(-) PS10::gfp-ssrA (virulence plasmid-cured) strains were cultured at 37o C in the presence (+) of MgOx for the indicated timepoints (hours, h). (A) Growth curve of strains, absorbance (A600nm) was detected by plate reader at the indicated timepoints. Mean and standard deviation are shown. (B) Fluorescence microscopy to quantify single cell fold change in GFP (PS10::gfp-ssrA) reporter signal. Single cell fluorescence was normalized to the average fluorescent value of the WT strain at 0h (value of 1, represented by dotted line). Each dot: individual cell, horizontal lines: median values. (C) WT PS10::gfp-ssrA, ∆lcrF PS10::gfp-ssrA, and ∆iscR PS10::gfp-ssrA were cultured at 37o C in the presence (+) of MgOx. Growth curve of strains, absorbance (A600nm) was detected by plate reader at the indicated timepoints. Statistics compare the WT strain to other strains. Mean and standard deviation are shown. (D) Fluorescence microscopy to quantify single cell fold change in GFP reporter signal. Single cell fluorescence was normalized to the average fluorescent value of the WT strain at 0h (value of 1, represented by dotted line). Each dot: individual cell, horizontal lines: median values. All data represent 3 biological replicates for each strain and timepoint in this figure. Statistics: (A, C) Two-way ANOVA with Tukey’s multiple comparison test; (B, D) Kruskal Wallis one-way ANOVA with Dunn’s post-test; ****p < 0.0001, **p < .01, *p < .05, ns: not-significant.
Fig 6.
Droplet-based microfluidics can be used to model clustered bacterial growth.
The yopE::mCherry PS10::gfp-ssrA strain was encapsulated into droplets and cultured at the indicated temperatures in the presence (+) or absence (-) of MgOx. At the indicated times aliquots were taken and bacterial growth and reporter expression were quantified by fluorescence microscopy. Growth at 26o C was compared to 37o C, and (A) area (µm2), (B) fold change in mCherry signal (relative to 0h), and (C) fold change in GFP signal (relative to 0h) were quantified. Horizontal dotted lines represent a value of 1, the initial average value at time 0h. Growth at 37o C in the presence (+) or absence (-) of MgOx was compared, and (D) area (µm2), (E) fold change in mCherry signal (relative to 0h), and (F) fold change in GFP signal (relative to 0h) were quantified. (G, H) Linear regression data indicating slope and R2 value of best fit lines for GFP compared to mCherry signal of microcolonies grown (G) 4h or (H) 6h at 37o C, (+) or (-) MgOx. (I) mCherry/GFP signal ratios are shown at the indicated spatial locations for microcolonies grown -/ + MgOx. (J) Representative images of microcolonies grown within droplets under the indicated conditions, all images represent 6h timepoints. Differential Interference Contrast (DIC) shown alongside GFP, mCherry, and merged fluorescent channels. Dotted line outlines the periphery of each droplet. Scale bar: 20µm. All data represent 2 biological replicates for each condition in this figure. Mean and standard deviation are shown. Each dot represents an individual microcolony. Statistics: (A-F) Two-way ANOVA with Tukey’s multiple comparison test; (G, H) Linear regression, significantly non-zero slope indicates correlation between values; (I) Wilcoxon matched-pairs; ****p < 0.0001, ***p < .001, **p < .01, *p < .05, ns: not-significant.
Fig 7.
T3SS induction reduces gentamicin susceptibility, but S10 expression predicts survival.
Strains were grown in the presence (+) or absence (-) of MgOx for 2h at 37o C. Antibiotics were added (gentamicin: gent, ciprofloxacin: cipro) and strains were cultured an additional 4h. Samples were taken at 0h (addition of antibiotics), 2h, and 4h to quantify CFUs or quantify fluorescence by microscopy. (A) Experimental design. Image Created in BioRender. Cotten, K. (2025) https://BioRender.com/b41q462 (B) yopE::mCherry PS10::gfp-ssrA strain + /- MgOx, gent treatment. (C) yopE::mCherry PS10::gfp-ssrA strain + /- MgOx, cipro treatment. Fold change in CFUs is calculated by normalizing the number of viable bacteria at 2h and 4h to the number of viable bacteria at 0h. (D) Comparison between non-fluorescent WT and ∆lcrF strains, + /- MgOx, gent treatment. Raw CFU values are shown. LOD: limit of detection (horizontal dotted line). (E) Comparison between non-fluorescent WT and ∆lcrF strains, + /- MgOx, cipro treatment. Raw CFU values are shown. Fold change in (F) mCherry (yopE) and (G) GFP (S10) signal from cells represented in (B) was calculated by normalizing the single cell fluorescence to the average fluorescent untreated (unt) value at 0h (value of 1, dotted line). Untreated 0h, 2h, and gentamicin 2h are shown. Each dot: individual cell, horizontal lines: median values. Dead cells were excluded using the fixable blue dead cell stain. (H) Linear regression data indicating slope and R2 value of best fit lines for GFP compared to mCherry signal of bacteria in the presence of MgOx, either untreated (unt) 2h or gent treatment 2h. Statistics: (B-E) Two-way ANOVA with Tukey’s multiple comparison test, statistics shown represent comparisons + /- MgOx; (F, G) Mann-Whitney; (H) Linear regression, significantly non-zero slope indicates correlation between values; ****p < 0.0001, ***p < .001, *p < .05.
Fig 8.
Gentamicin treatment selects for surviving bacteria with reduced S10 expression in the mouse spleen.
C57BL/6 mice were inoculated intravenously with the yopE::mCherry PS10::gfp-ssrA or yopE::mCherry-ssrA PS10::gfp-ssrA strains. At 48h p.i., spleens were harvested from a subset of mice to represent the time of treatment (48h). Additional sets of mice were injected intraperitoneally with 40mg/kg gentamicin, and spleens were harvested 4h post-treatment (+gent 4h) (A) Representative images at 48h p.i. (untreated). (B) CFU/spleen at 48h compared to 4h gent treatment. Dots: individual mice. (C) Reporter expression (ratio mCherry/GFP signal intensity) was quantified at the indicated timepoints. Dots: individual microcolonies. (D) mCherry mean fluorescent intensity (MFI) across individual microcolonies. Dots: individual microcolonies. (E) GFP mean fluorescent intensity (MFI) across individual microcolonies. Dots: individual microcolonies. (F) Representative image at 4h post-treatment. Scale bars: 20µm. Statistics: (B-E) Mann-Whitney; **p < .01, *p < .05, ns: not-significant.