Fig 1.
FAST-tagged proteins retain fluorescent properties after secretion into bacterial culture media.
(A) Diagram of constructs in the pAD vector for constitutive expression in Lm. (B) Lm strains expressing FAST-tagged proteins were cultured in LSM, then fluorescence intensities were measured on the filtered supernatants of each culture in presence of 5 μM HBR-3,5DM. Concentrations of FAST-labelled proteins were calculated by reference to a standard curve of purified FAST in LSM. Residual fluorescence measured for the strain producing non-secreted FAST represents bacterial lysis. (C) Diagram of constructs in the pSU2.1 vector for expression in Sf. (D) WT or ΔipaD Sf strains expressing FAST-tagged OspF or IpaB were cultured in M9 medium, then fluorescence intensities were measured on the filtered supernatants of each culture in presence of 5 μM HBR-3,5DM. Concentrations of FAST-labelled proteins were calculated by reference to a standard curve of purified FAST in M9 medium. Whereas ΔipaD strains secrete proteins constitutively, T3SS secretion is not activated in WT strains, thus the fluorescent signals measured for these strains (blue dots) reflect bacterial lysis and/or leakage of the T3SS. (B, D) Normalized values, means and standard deviations from three independent experiments were plotted. p-values represent the results of two-tailed Student’s t-tests with equal variance assumption. Source data are provided in S3 Table.
Fig 2.
Secreted FAST accumulates exponentially in the cytoplasm of infected cells.
(A) Spinning disk fluorescence microscopy images of LoVo cells infected with Lm secreting SP-FAST (cyan) at different time-points post-infection (h:min). The actin cytoskeleton (purple) was labelled with SiR-actin. The area where FAST fluorescence intensity was measured for graph (C) is boxed in orange. Scale bars, 5 μm. (B) Dispersion of fluorescence intensities. Fluorescence emission at 562 nm (FAST:HBR-3,5DM channel) was quantified over time within a region of fixed area in cells infected by Lm strains expressing either SP-FAST (in green, n = 127) or mCherry as a negative control (in blue, n = 35). As an indicator of the amplitude of fluorescence accumulation, the standard deviation of fluorescence intensity over time was plotted for each cell. A.U., arbitrary units. The p-value represents the result of a two-tailed Mann-Whitney non-parametric test. (C) Intensity of FAST signals measured over time in the region boxed in yellow in (A). The black line displays an exponential fit obtained over the ascending part of the curve (green dots). (D) Distribution of the doubling time of FAST fluorescence signals among the population of infected cells (n = 39). Source data are provided in S4 Table.
Fig 3.
Secreted FAST reveals the heterogeneity of Listeria residence time in internalisation vacuoles.
(A) Expected profile of fluorescence accumulation in internalisation vacuoles for Lm secreting SP-FAST. After bacterial adhesion, Lm enters epithelial cells via a zipper mechanism. Secreted FAST should start accumulating in vacuoles upon their closure, and then remain visible until vacuole rupture. In each vacuole, the level of fluorescence reflects the equilibrium between bacterial secretion of FAST and its leakage in case of membrane permeation. (B) Spinning disk microscopy images of LoVo cells infected with Lm ΔhlyA expressing SP-FAST (cyan) and mCherry (orange) for 35 min after entry. The actin cytoskeleton (purple) was labelled with SiR-actin. Scale bars, 5 μm; timescale, h:min. (C) Distribution of Lm residence times in internalisation vacuoles in LoVo cells. Green, WT strain carrying an integrated pAD-SP-FAST plasmid (n = 284); orange, ΔhlyA::SP-FAST strain carrying an integrated pHpPL3-mCherry plasmid (n = 306). The interpolated median lifetime of SP-FAST-labelled vacuoles, calculated from the raw distributions, are displayed in dark green and dark orange dashed lines for the WT and ΔhlyA strains, respectively. The p-value indicates the result of a two-tailed Student’s t-test on the distributions, assuming equal variance. Source data are provided in S5 Table.
Fig 4.
Listeria monocytogenes replicates inside long-term vacuoles decorated with LLO.
(A) Spinning disk microscopy images of LoVo cells infected with Lm expressing both LLO-FAST (in cyan) and mCherry (in orange) at several time-points post-infection. SiR-actin staining is shown in purple. eSLAPs are indicated with solid orange arrowheads; their past location is pointed with open arrowheads after their rupture. Scale bars, 5 μm; timescale, h:min. (B) Doubling times of Lm expressing mCherry in the cytoplasm (grey, n = 7) or in eSLAPs (green, n = 18) in infected LoVo cells. (C) Quantification of the increase in volume of eSLAPs, and thus of the growth of the bacteria they contain, for WT (green), prfA* (blue) or ΔhlyA (orange) Lm strains in infected LoVo cells. (D) Proportion of intracellular Lm that multiplied inside eSLAPs during a time-course of 8 h. Plotted values represent the ratio of the number of eSLAPs that had at least doubled in volume over the time course to the number of segmented mCherry objects (i.e. cytoplasmic or intravacuolar bacteria, isolated or in clusters) at the beginning of the observation (2 h p.i). Green, WT strain (n = 134); blue, prfA* strain (n = 33); orange (null), ΔhlyA strain (n = 113). Source data are provided in S6 Table.
Fig 5.
Listeria eSLAPs derive from internalisation vacuoles and display typical markers of LC3-associated phagocytosis.
(A) Differential labelling by YFP-CBD of the cytosolic versus intravacuolar populations of intracellular bacteria. LoVo cells were transfected with pEYFP-CBD (in cyan) 24 h before being infected with Lm expressing mCherry (in orange), then imaged at different time-points post infection. eSLAPs are indicated with solid orange arrowheads; their past location is pointed with open arrowheads after their rupture. The cell outline is indicated with purple dashed lines. Note that a strong non-specific YFP-CBD signal is also detected in the cell nucleus. Timescale, h:min. (B) Rab5, Rab7, LC3 and LAMP1 (in green) were detected by immunofluorescence in LoVo cells infected for 3 h with mCherry-expressing bacteria (in red). For acidity staining, LoVo cells infected for 2 h with eGFP-expressing bacteria (in red) were stained with LysoTracker Deep Red (in green), and observed 1 h afterwards on an inverted spinning disk microscope. Orange arrowheads point to representative eSLAPs. (A, B) Scale bars, 5 μm. Quantitative analyses of these experiments are provided as S11 Fig.
Fig 6.
Extended model of the intracellular life cycle of Listeria monocytogenes in colon adenocarcinoma epithelial cell lines.
(A) In the classical scenario, after receptor-mediated entry, Lm evades the vacuole thanks to the combined action of LLO and phospholipases. (B) Here we identified a population of Lm that can remain for several hours and multiply inside vacuoles in LoVo cells. These compartments (eSLAPs) are neutral, positive for Rab7, LC3 and LAMP1, and decorated with LLO. This second population of bacteria finally escapes into the cytoplasm at later time points.