Figure 1.
InlK is a virulence factor of L. monocytogenes.
A. The inlK gene locus in L. monocytogenes compared with the same genomic region in the related non-pathogenic species L. innocua. The stem and circle represent transcription terminators. B. Kaplan-Meier curve represents the survival of BALB/c mice over time. Four BALB/c mice in each experimental group were infected i.v with 104 L. monocytogenes wild-type (EGD-e) or ΔinlK mutant. C. The L. monocytogenes EGD-e wild-type strain (WT), the ΔinlK mutant (ΔinlK) and the complemented strain (ΔinlK+pPL2 inlK) 104 CFU were inoculated i.v into BALB/c mice. Animals were euthanized 24 h, 48 h, 72 h or 96 h after infection and organs were recovered, homogenized, and homogenates serially plated on BHI. The number of bacteria able to colonize liver (left panel) and spleen (right panel) is expressed as log10 CFU. Four animals per bacterial strain, per time points and per experiment were used. Statistical analyses were performed on the results of 3 independent experiments using the Student t test. P values of <0.05 were considered statistically different and are labeled here as *.
Figure 2.
A. InlK amino acid sequence. The signal sequence is underlined and the different regions of leucine rich repeats (LRRs) are outlined. The consensus pentapeptide LPXTG at the C-terminal end is boxed. B. Detection by immunofluorescence microscopy of InlK over-expressing in L. monocytogenes EGD-e (WT), ΔinlK, WT+pADc-inlK, ΔinlK+pPRT-inlK and the ΔsrtA mutant over-expressing inlK (ΔsrtA+pPRT-inlK) grown in BHI medium using the rabbit polyclonal anti-InlK antibody. InlK was detected at the surface of InlK over-expressing bacteria (WT+pADc-inlK and ΔinlK+pPRT-inlK), whereas it was undetectable at the surface WT bacteria or at the surface of the ΔstrA mutant over-expressing inlK. C. Detection of InlK by Western blot on total lysates of L. monocytogenes EGD-e (WT), ΔinlK and ΔinlK+pPRT-inlK grown in BHI using the rabbit polyclonal anti-InlK antibody. Decreased concentrations of recombinant purified InlK were used as a positive control. D. Detection of secreted InlK in the supernatant of ΔsrtA mutants over-expressing InlK. Western blotting was carried out on trichloroacetic acid precipitates of ΔsrtA and ΔsrtA+pPRT-inlK culture (OD600 = 1) supernatants using the rabbit polyclonal anti-InlK antibody. E. Detection of purified recombinant InlK protein with rabbit polyclonal anti-live Listeria antibody, rabbit polyclonal anti-killed Listeria antibody, rabbit polyclonal anti-InlK and a rabbit pre-immune serum. InlK was detected only with the rabbit polyclonal anti-live Listeria antibody indicating that it is expressed during the in vivo infectious process. BSA was used as control protein. Two different amounts of proteins were tested (500 ng and 200 ng) to access signal specificity.
Figure 3.
InlK interacts with the Major Vault Protein.
A. Bacterial pull-down of purified GST-MVP with the L. monocytogenes strains WT+pPRT-empty, ΔinlK+pPRT-empty, ΔinlK+pPRT-inlK and WT+pPRT-inlJ. GST-MVP bound to InlK over-expressing bacteria (ΔinlK+pPRT-inlK) but not to other bacteria. GST-ScarA was used as control of the specificity of the MVP precipitation by InlK over-expressing bacteria. B. Bacterial pull-down of MVP-GFP from transfected HeLa cell lysates with the L. monocytogenes strains WT+pPRT-empty, ΔinlK+pPRT-empty, ΔinlK+pPRT-inlK and WT+pPRT-inlJ. MVP-GFP bound to InlK over-expressing bacteria (ΔinlK+pPRT-inlK) but not to other bacteria. C. Co-immunoprecipitation (Co-IP) of purified InlK (20 µg) with endogenous MVP of HT29 cells. The right panel shows anti-MVP co-IP performed on HT29 cell lysates. The left panel shows the control anti-MVP co-IP performed on lysis buffer. D. Detection of MVP recruitment at the surface of InlK over-expressing bacteria. HeLa cells were transfected with MVP-GFP (green), infected with L. monocytogenes EGD-e wild-type (WT), ΔinlK or ΔinlK+pPRT-inlK for 4 h, fixed for fluorescence light microscopy, and stained with anti-Listeria antibodies (red). Inset regions are magnified. The scale bar represents 1 µm.
Figure 4.
InlK/MVP interaction occurs in the cytosol, before actin polymerization.
A. Detection of MVP recruitment at the surface of intracellular InlK over-expressing bacteria. HeLa cells were transfected with MVP-GFP (red), infected with InlK expressing Listeria (ΔinlK+pPRT-inlK) for 4 h, fixed for fluorescence light microscopy. Intra- (only green) and extracellular (cyan = green+blue) bacteria were differentially stained with anti- Listeria antibody (cf Material and Methods). Inset regions are magnified. Arrows indicate another intracellular bacterium which recruit MVP-GFP. The scale bar represents 1 µm. The right panel represents the quantification of the intracellular bacteria that recruit MVP (mean%±SEM%) shown in the left panel. Statistical analyses were performed on the results of 3 independent experiments using the Student's t test. No significant difference was found between the 4 time points. B. Detection of MVP recruitment at the surface of intracytosolic InlK over-expressing bacteria. HeLa cells were transfected with MVP-tomato (red) and YFP-CBD (green), infected with InlK over-expressing Listeria (ΔinlK+pPRT-inlK) for 4 h, fixed for fluorescence light microscopy and stained with phalloidin (blue). MVP positive bacteria were also labeled with YFP-CBD revealing that MVP was recruited by intracytosolic bacteria after the lysis of the internalization vacuole. Inset regions are magnified. The scale bar represents 1 µm. C. Kinetics of MVP and actin recruitment at the surface of InlK over-expressing bacteria. HeLa cells were transfected with MVP-tomato (red) and actin-GFP (green), infected with InlK over-expressing Listeria (ΔinlK+pPRT-inlK) for 4 h, and prepared for real-time video microscopy. Image series were collected every 15 min for 2 h. The left part shows an MVP positive bacterium that never recruits actin. The right part shows MVP replacement by actin around the bacterium. No colocalization of MVP-Tomato and actin-GFP was detected. Time is indicated along the Y axis. The entire image sequence can be viewed as Video S1.
Figure 5.
MVP impairs the recruitment of autophagy markers.
A. Impaired recruitment of p62 to MVP positive Listeria. HeLa cells were transfected with MVP-GFP (green), infected with InlK over-expressing Listeria (ΔinlK+pPRT-inlK) for 4 h, fixed for fluorescence light microscopy, and stained with phalloidin (blue) and anti-p62 antibody (red). Inset regions are magnified. Arrows indicate independent bacteria The scale bar represents 1 µm. The vast majority of MVP-positive bacteria were completely devoid of anti-p62 labeling (95.1±2.0%; mean ± SEM from n = 3 experiments) but 4.9±2.0% (mean ± SEM from n = 3 experiments) were stained at one pole with MVP and at the other pole with p62. B. Impaired recruitment of GFP-LC3 on MVP positive Listeria. HeLa cells were transfected with MVP-tomato (red) and GFP-LC3 (green), infected with InlK over-expressing Listeria (ΔinlK+pPRT-inlK) for 4 h, fixed for fluorescence light microscopy, and stained with phalloidin (blue). Inset regions are magnified. The scale bar represents 1 µm. MVP and/or actin positive bacteria were never recognized by GFP-LC3. Arrows point to bacteria at different steps of the infection process: 1) InlK over-expressing bacterium is totally covered by MVP; 2) bacterium is partially labeled with MVP (at the poles) and actin (at the center); 3) bacterium is completely covered by actin; 4) bacterium is enclosed in an GFP-LC3 positive autophagosome. C. Kinetics of autophagy escape for MVP positive Listeria. Jeg3 cells were transfected with MVP-tomato (red) and GFP-LC3 (green), infected with InlK over-expressing Listeria (ΔinlK+pPRT-inlK) for 4 h, and prepared for real-time video microscopy. Image series were collected every 5 min for 2 h. Time is indicated along the Y axis. The left panel shows that the GFP-LC3 membranous aggregate detaches from the MVP positive Listeria. The entire image sequence can be viewed as Video S2. The right panel shows that the GFP-LC3 membranous aggregate on MVP positive bacteria does not lead to an autophagosome formation, whereas those bacteria efficiently divided. The entire image sequence can be viewed as Video S3. D. Impaired recruitment of GFP-LC3 and ubiquitin to MVP positive ΔactA Listeria. HeLa cells were transfected with MVP-tomato (red) and GFP-LC3 (green), infected with InlK over-expressing ΔactA (ΔactA+pADc-inlK) for 4 h, fixed for fluorescence light microscopy, and stained with anti-ubiquitin antibody (blue) and DAPI (white). Inset regions are magnified. The scale bar represents 1 µm. E. LC3 levels in infected RAW 267.4 macrophages. Left panel: RAW 267.4 macrophages were infected with L. monocytogenes EGD (WT), ΔactA or ΔactA+InlK for 6 h. Cell total lysates were immunobloted for LC3 and actin. Western blot is representative from 3 independent experiments. Right panel: Quantification of the relative LC3-II level (mean ± SEM) shown in the left panel. Statistical analyses were performed on the results of 3 independent experiments using the Student's t test. P values of <0.05 were considered statistically different.
Figure 6.
MVP-dependent escape from autophagy leads to increased Listeria survival.
A. InlK and ActA expression in Listeria strains used for survival assays. Total lysates of L. monocytogenes EGD-(pADc-GFP), EGD-(pADc-inlK), ΔactA-(pADc-GFP) and ΔactA-(pADc-inlK) grown in BHI were immunoblotted using anti-ActA and anti-InlK antibodies. B. Intracellular survival of EGD-(pADc-GFP), EGD-(pADc-inlK), ΔactA-(pADc-GFP) and ΔactA-(pADc-inlK) in RAW 267.4 macrophages. Statistical analyses were performed on the results of 3 independent experiments using the Student t test. P values of <0.05 were considered statistically different and are labeled here as *. C. Intracellular survival of ΔactA-(pADc-GFP) and ΔactA-(pADc-inlK) in MVP-GFP transfected Jeg3 cells. Statistical analyses were performed on the results of 3 independent experiments using the Student's t test. P values of <0.05 were considered statistically different and are labeled here as *. D. Intracellular survival of ΔactA-(pADc-GFP) and ΔactA-(pADc-inlK) in MVP-GFP transfected HeLa cells. Statistical analyses were performed on the results of 3 independent experiments using the Student's t test. P values of <0.05 were considered statistically different and are labeled here as *. E. MVP levels in RAW 267.4 macrophages treated with MVP-siRNA. Western blot is representative from 3 independent experiments. F. Intracellular survival of ΔactA-(pADc-GFP) and ΔactA-(pADc-inlK) in MVP knock-down RAW 267.4 macrophages. Statistical analyses were performed on the results of 3 independent experiments using the Student t test. P values of <0.05 were considered statistically different and are labeled here as *.
Figure 7.
Model for escape of autophagic recognition for L. monocytogenes expressing InlK.
During intracellular growth, cytoplasmic bacteria are able to escape from autophagy process using two independent virulence factors, ActA and InlK. On the one hand, the recruitment of VASP, Arp2/3 complex and actin via ActA masks the bacteria from ubiquitination and autophagic recognition. On the other hand, MVP recruitment via InlK is also able to protect bacteria from ubiquitination and autophagic recognition. In that way, depending on ActA and InlK expression, four possibilities could be distinguished: (1) When neither ActA nor InlK are expressed, the bacterial ubiquitination is followed by p62 and LC3 recruitment, leading to autophagosome formation around the bacterium. (2) When ActA is expressed (e.g. wild-type bacterium (WT) grown in BHI before cell infection) it is sufficient for Listeria to escape from autophagy. (3) In contrast, in the absence of ActA, InlK efficiently protects bacterium against autophagy recognition via MVP recruitment. (4) Finally, when ActA and InlK are co-expressed by the bacterium, InlK rapidly recruits MVP at the surface of the bacterium. Then, in some instance, ActA replaces InlK leading to a switch of the bacteria disguised from MVP to actin. The model is partially based on the results of Yoshikawa et al [11].