A. COS-7 cells transfected with siRNA targeting SRF or a non-targeting siRNA. After 72 hours knockdown efficiency was assessed using real-time quantitative RT-PCR with SRF specific primers and normalized to GAPDH. Shown are means of three experiments ± standard deviation. B. Infection of SRF-knockdown cells with EPEC results in a significant reduction in nuclear accumulation of MAL-GFP, P = 0.0081 relative to the no siRNA control. Data are the means of at least 3 experiments where a minimum of 150 transfected cells were counted for each condition of each experiment. Values are means ± SEM. C. Representative images of cells counted in panel B. Scale Bar = 20 µm.

A. EPEC Δtir cannot induce the nuclear translocation of MAL-GFP. COS-7 cells were transfected with MAL-GFP, serum starved then infected with the indicated bacteria. The percentage of transfected cells that had MAL-GFP in the nucleus, cytoplasm or both nucleus and cytoplasm was determined. Data are the means of at least 3 experiments where a minimum of 150 transfected cells were counted for each condition of each experiment. Values are means ± standard deviation. Relative to the uninfected control *P = <0.45, ** P = 7.27⁻⁷. B. Representative images from A, MAL-GFP- (green), F-actin (red), DNA/bacteria (blue). C. Phosphorylation of Tir residues Y454 and Y474 is necessary for EPEC-induced MAL-GFP translocation. D. Representative images from C, F-actin (green), DNA/bacteria (Red). Scale Bar = 20 µm. E. Cells untreated or infected with EPEC for 3.5 hours were lysed and separated into 100,000 –g supernatant (S) or pellet (P) fractions. Equal amounts were separated by SDS-PAGE and actin in each fraction detected using an anti-actin antibody. E′ F-∶G-actin ratios from E were quantified as described in materials and methods. The mean % F-actin from at least 3 experiments is shown ± standard deviation.

A. The percentage of transfected cells displaying nuclear localization of MAL-GFP after serum starvation. COS-7 cells were cotransfected with MAL-GFP and cDNA expression constructs as indicated. After 18 hours they were serum starved for 24 hours then fixed and stained. The dotted red line represents the cut-off threshold for the percentage of cells displaying nuclear MAL-GFP required for a gene to be declared a hit. The cut-off (30%) was defined as a 2-fold increase over background (10% for 0.3% FCS or empty vector controls). Data represents the mean of 500 cells from 3 individual experiments ± standard deviation. B. COS-7 cells transfected with siRNA targeting the indicated genes or a non-targeting siRNA. After 72 hours knockdown efficiency was assessed using real-time quantitative RT-PCR with FLRT3, TESK1 or C22orf28 specific primers and normalized to GAPDH. Shown are means of three experiments, each using independent cDNA samples, ± standard deviation. C. EPEC-induced nuclear translocation of MAL-GFP is significantly reduced in ABRA (P = 0.0021) and FLRT (P = 0.000363) knockdown cells. Localization of MAL-GFP in serum starved COS-7 cells with respective protein knockdown after infection with EPEC. Data are the means of three experiments ± standard deviation. D. ABRA and FLRT3 induce transcription of SRE-luciferase. E. FLRT3 knockdown significantly reduces ABRA-induced nuclear translocation of MAL-GFP, P = 0.0017 relative to the no siRNA control. Data are the means of three experiments ± standard deviation.

During infection with the extracellular pathogen EPEC, Tir translocates to the host cell and inserts into the plasma membrane where it interacts with the bacterial surface protein intimin and anchors it to the cell. The C-terminus of Tir is phosphorylated by host cell kinases leading to the recruitment and binding of Nck. N-WASP and the Arp2/3 complex are recruited leading to the generation of actin filaments beneath the bacterium. ABRA can bind actin in the newly formed (or forming) pedestal and stabilise the structure. The change in G-actin to F-actin ratio induced by pedestal formation is “sensed” by MAL via direct or indirect actions from ABRA and FLRT3, whereupon it is freed from its inhibitory complex with G-actin, to enter the nucleus and interact with SRF, driving transcription of target genes.

A. EPEC induces nuclear accumulation of MAL-GFP. COS-7 cells were transfected with MAL-GFP, serum starved then stimulated with 15% foetal calf serum (FCS) or infected with the indicated bacteria. B. The percentage of transfected cells from panel A that had MAL-GFP in the nucleus, cytoplasm or both nucleus and cytoplasm was determined. Data are the means of at least 3 experiments where a minimum of 150 transfected cells were counted for each condition of each experiment. Values are means ± SEM. **P = 7.31336⁻⁰⁵, *P = 0.027. Scale Bar = 20 µm.

Transcription of SRF target genes measured by quantitative polymerase chain reaction (qRT-PCR). Data are the means of at least 3 experiments ± standard deviation.

A. Cellular distribution of MAL-GFP in serum-starved COS-7 cells infected with bacteria as indicated. Data are the means of at least 3 experiments where a minimum of 150 transfected cells were counted for each condition of each experiment. Values are means ± standard deviation. B. Representative images of pedestal formation induced by Tir^EHEC rescue strains. C. COS-7 cells transfected with siRNA targeting IRTKS. After 72 hours knockdown efficiency was assessed using real-time quantitative RT-PCR with IRTKS specific primers and normalized to GAPDH. Shown are means of three experiments ± standard deviation. D. Cellular distribution of MAL-GFP in IRTKS knockdown cells infected with KC12 pEspF_U. Data are means of three experiments ± standard deviation. Scale bar = 20 µm.

A. COS-7 cells transfected with empty pCMV-3xFlag vector, FLRT3-Flag or Flag-ABRA. Cells were infected with EPEC, fixed and stained with an anti-Flag antibody. FLRT3-Flag and Flag-ABRA are enriched at the EPEC pedestal (Arrowheads). B. Wild-type and ABRA knockdown cells were infected with EPEC for a total of 5 hours, fixed and stained with DAPI (red) and phalloidin (green). Pedestals in ABRA knockdown cells are disorganised (close-up). Scale bars = 20 µm and 5 µm in close-up panels.