Fig 1.
Schematic of central players in actin polymerization in EPEC pedestals and podosomes.
A: Upon infection EPEC inserts its own receptor Tir in the host cell membrane. Tir binds intimin, which is inserted into the bacterial membrane. Within the host cell cytosol Tir is phosphorylated by Src and/or Abl/Arg family kinases leading to recruitment of Nck, N-WASp and Arp2/3, which leads to actin polymerization. Also, PI(3,4)P2 is synthesized in the membrane. Other proteins, like dynamin and cortactin are also involved in actin polymerization. It is the objective of this study to investigate if Tks5 localizes to the infection site. B: In podosomes signaling by integrins leads to synthesis of PI(3,4)P2 and recruitment and phosphorylation of Tks5 and cortactin. Upon binding to PI(3,4)P2 Tks5 binds a complex of Nck, Grb2 and N-WASp which again binds Arp2/3 to initiate actin polymerization. Recruitment to or involvement of Tks5 in EPEC pedestal formation is so far unknown. C: Tks5 is composed of one phox homology (PX) domain and five Src Homology 3 (SH3) domains.
Fig 2.
Tks5 localized with bundled actin in MDCK cells.
A: Western blot of lysates of Tks5-EGFP-MDCK cells and untransfected MDCK cells probed with anti-Tks5 (top panels) and α-tubulin (bottom panels) antibodies. Arrows indicate bands of Tks5-EGFP and Tks5 isoforms. α-tubulin was included as loading control. One splice variant of Tks5 was observed in WT MDCK cells. Tks5-EGFP-MDCK cells expressed various splice variants of endogenous Tks5 and/or Tks5-EGFP (125–140 kDa); Tks5-EGFP (150–175 kDa) was expressed at a higher level than endogenous Tks5 in the transfected cell line. B and C: Tks5-EGFP-MDCK cells were stained with Hoechst to visualize nuclei (blue) and phalloidin-rhodamine to visualize actin or α-tubulin antibody (red). Images are maximum projections of Z-stacks of 20 or 17 slices, respectively. White arrows in B indicate a site of colocalization of Tks5-EGFP and actin, whereas the yellow arrows in C point to Tks5-EGFP not co-localizing with α-tubulin, and the pink arrows point to α-tubulin. D: Tks5-EGFP-MDCK cells were transiently transfected with LifeAct-mRuby and treated with cytochalasin D to disrupt actin fibers and imaged live with fluorescence microscopy. Timepoints after cytochalasin D addition are indicated. Arrows indicate aggregates of actin, to which Tks5-EGFP also localized. Scale bars correspond to 10 μm.
Fig 3.
Tks5 localized to EPEC infection site and Tir is essential for the localization.
Tks5-EGFP-MDCK cells were infected with WT, Δtir, ΔescN, JPN15, or TirY474F EPEC for four hours, as indicated. Cells and bacteria were fixed and stained with phalloidin to visualize actin (red) and Hoechst and an antibody directed against lipid A in the bacterial membrane to visualize nuclei and EPEC bacteria (both shown in blue). Z-stacks were acquired on a confocal microscope, and the shown slices were selected to both show EPEC bacteria as well as Tks5-EGFP and actin localization. Tks5-EGFP and actin were recruited to the infection site of WT, TirY474F, and JPN15 EPEC, where they accumulated around the infecting bacteria. No recruitment was observed for Δtir and ΔescN EPEC. Arrows point to the positions of individual bacteria. Scale bars correspond to 5 μm.
Fig 4.
Tks5 was recruited simultaneously with actin and N-WASp to the site of EPEC infection and they colocalized in the pedestal.
A: Tks5-EGFP-MDCK cells were infected with WT EPEC for four hours. Cells and bacteria were then fixed and stained with phalloidin to visualize actin (red), Hoechst to visualize DNA and an antibody against lipid A in the bacteria membrane (both in blue). Tks5-EGFP co-localized with actin in the pedestal. B: Tks5-EGFP-MDCK cells transiently transfected with N-WASp-mCherry were infected with WT EPEC for four hours, then fixed and stained with Hoechst and lipid A antibody. Tks5-EGFP co-localized with N-WASp-mCherry at the infection site. Cells and bacteria in A and B were imaged on a confocal microscope, and the shown slices were selected to both show EPEC bacteria as well as protein localization. Arrows point to positions of individual bacteria. Scale bars correspond to 5 μm. C and D: Tks5-EGFP-MDCK cells transiently transfected with LifeAct-mRuby (C) or N-WASp-mCherry (D) were infected with low numbers of WT EPEC bacteria and imaged with widefield microscopy. Tks5-EGFP and LifeAct-mRuby or N-WASp-mCherry are shown as inverted contrast to ease the interpretation. A flash of Tks5-EGFP and LifeAct-mRuby or Tks5-EGFP and N-WASp were observed instantly upon adhesion (exemplified with Tks5-EGFP and LifeAct-mRuby in C). Upon adhesion, Tks5-EGFP was rearranged in the membrane simultaneously with LifeAct-mRuby or N-WASP-mCherry (exemplified with N-WASp-mCherry in D). Arrows point to positions of the initial rearrangements of Tks5 and LifeAct-mRuby or N-WASp-mCherry. Scale bars correspond to 3 μm.
Fig 5.
The PX domain of Tks5 is not required for EPEC infection.
Murine Embryonic Fibroblasts (MEFs) with wild-type Tks5 expression or expression of the deletion mutant ΔPX-Tks5 only were infected with WT EPEC for six hours. Cells and EPEC were then fixed and stained with Hoechst and phalloidin. A total of 60 cells of each type were imaged, and the number of bacteria infecting each cell was counted, and it was evaluated whether they formed pedestals. Outlier cells with more than 150 infecting EPEC were excluded from the analysis (six cells in total). A: Examples of MEFs infected by EPEC. Orange and white arrows indicate infecting bacteria with and without pedestal formation, respectively. Scale bars correspond to 20 μm for whole-cell images and 10 μm for inserts. Colors were adjusted to optimally visualize infecting bacteria. B: Number of infecting EPEC per cell with the average numbers indicated by a bar.