Figure 1.
Cell-associated aggregates show key characteristics of biofilms.
(A) MDCK cells were infected with PAK-GFP (green), fixed, and stained for actin (blue) and extracellular DNA with DDAO (purple). (B) MDCK cells were infected with PAK-mCherry (red), fixed, and stained for actin (blue) and with FITC-HHA, which binds to Psl. (C) Bacterial viability after exposure to amikacin was evaluated by staining live bacteria with SYTO 9 (green) and counterstaining dead bacteria with propidium iodide (red). MDCK cells were also stained for actin (blue). Bacteria were incubated in liquid culture (MEM, top row) or with MDCK cells (bottom row). After 60 minutes of infection, bacteria in liquid culture or on MDCK cells were treated with amikacin (400 ug/ml) for 2 hours (top right and bottom right panels). Adherent single bacteria showed propidium iodide staining after amikacin treatment (white arrows). Representative confocal images from three independent experiments are shown. Scale bars, 10 µm.
Figure 2.
Cell-associated aggregation requires the T3SS translocon but does not require T3SS effectors.
MDCK cells stably transfected with PH-Akt-GFP (green) were infected with mCherry-expressing (A) PAK, (B) PAKΔSTY (lacks the known T3SS effectors), or (C) PAKΔpopB (has the functional needle apparatus but lacks the translocon) for 60 minutes, fixed, and stained for actin (blue). Cell-associated aggregates (red) were visible with PAK (A) and PAKΔSTY (B) and were accompanied by formation of membranous protrusions (white arrows) containing PH-Akt-GFP. PAKΔpopB (C) bound individually or as groups of 2 to 3 bacteria. Representative confocal images from three independent experiments are shown. Scale bars, 10 µm. (D) Cell-associated aggregation by PAK and T3SS mutants was quantified using spinning disk confocal microscopy. Shown is the number of aggregates (≥10 bacteria) normalized to PAK (n≥3 independent experiments). Data are mean ± SEM. ***p<0.001 compared to PAK. Statistics in Supplemental Statistical Analysis (Text S1).
Figure 3.
Bacterial aggregation in mouse pneumonia requires the T3SS translocon but does not require T3SS effectors.
BALB/c mice were injected intranasally with PAK, PAKΔSTY, or PAKΔpopB and lungs were isolated, sectioned, and stained at 3 hours post-infection. (A) Widefield epifluorescent imaging showed PAK (red) bound to lung epithelium (green). Representative image from 67 images is shown. Scale bar, 100 µm. (B) The amount of aggregation by PAK and T3SS mutants was quantified and plotted against bacterial density (n≥60 images for each strain). Linear regression lines were applied to each bacterial strain. (C) Comparisons of linear regression lines showed no differences in slope but significant differences in the intercept of PAKΔpopB compared to PAK or PAKΔSTY. Statistics in Supplemental Statistical Analysis (Text S1).
Figure 4.
Co-infection with T3SS+ bacteria restores cell-associated aggregation in PAKΔpopB.
MDCK cells were co-infected with equal amounts of mCherry-expressing (red) and GFP-expressing bacteria (green), fixed, and stained for actin (blue). Co-infection with (A) PAK-mCherry and PAK-GFP, (B) PAKΔpopB-mCherry and PAK-GFP, and (C) PAKΔpopB-mCherry and PAKΔSTY-GFP resulted in cell-associated aggregates composed equally of both strains of bacteria. Representative 3-D reconstructions from three independent experiments are shown. Scale bars, 10 µm. (D) Cell-associated aggregation after co-infection was quantified using spinning disk confocal microscopy. Shown is the number of aggregates (≥10 bacteria) normalized to PAK (n≥3 independent experiments). Data are mean ± SEM. ***p<0.001 compared to PAK. Statistics in Supplemental Statistical Analysis (Text S1).
Figure 5.
Supernatants from PAK-infected cells restore cell-associated aggregation and enhance abiotic biofilm formation.
(A) Cell-associated aggregation by PAKΔpopB in the presence of various supernatants was quantified using spinning disk confocal microscopy using the same method as in Figure 2D (n≥3 independent experiments). Data are mean ± SEM. ***p<0.001 compared to PAK. (B) PAK and isogenic T3SS mutants were inoculated into tissue-culture media (MEM) or supernatant from PAK-infected cells and assessed for biofilm formation on microtiter plates (n≥3 independent experiments). Data are mean ± SEM. *p<0.05 compared to control in MEM. Statistics in Supplemental Statistical Analysis (Text S1).
Table 1.
Strains and plasmids used in this study.