Fig. 7: Bacterial RfaD is a major virulence determinant in EHEC-induced microvillar effacement in C. elegans and human intestinal cells.

a The percentage of ACT-5 mislocalization in mCherry::ACT-5 worms fed with OP50 (n = 736, N = 20) and EHEC wild-type (n = 684, N = 20) and different EHEC isogenic gene deletion mutants, including the T3SS regulator ler (n = 99, N = 3, P = 0.6898), the T3SS injectisome components espA (n = 108, N = 3, P = 0.5503), espB (n = 158, N = 5, P = 0.6659), and espD (n = 100, N = 3, P = 0.2395), and escN (n = 91, N = 3, P = 0.9204), and the T3SS effectors eae (n = 98, N = 3, P = 0.164), espF (n = 105, N = 3, P = 0.0619), espFu (n = 100, N = 3, P = 0.528), espG (n = 97, N = 3, P = 0.4875), espH (n = 104, N = 3, P = 0.6639), espM (n = 112, N = 3, P = 0.0546), map (n = 96, N = 3, P = 0.9724), and tir (n = 99, N = 3, P = 0.173), and toxins, including hlyA (hemolysin A) (n = 92, N = 3, P = 0.8988), hlyE (hemolysin E) (n = 95, N = 3, P = 0.4307), stx2 (n = 95, N = 3, P = 0.6309), and stx1 (Shiga-like toxin 1) (n = 126, N = 4, P = 0.0005), and rfaD (functions in the biosynthesis of LPS) (n = 76, N = 3, P < 0.0001) for 4 days. b The qRT-PCR analysis of cyb-3 mRNA in N2 wild-type animals exposed to OP50, EHEC, the isogenic EHEC with stx1 deletion (∆stx1), and the isogenic EHEC with rfaD deletion (∆rfaD) at 20 °C for 4 days, respectively. c The percentage of ACT-5 mislocalization in the cyk-1 o/e worms fed with OP50, EHEC wild-type, or EHEC ∆rfaD bacteria for 3 and 4 days. 3 DPI (OP50: n = 91, N = 3, EHEC: n = 97, N = 3; EHEC ∆rfaD: n = 71, N = 3, P = 0.0002); 4 DPI (OP50: n = 91, N = 3, EHEC: n = 95, N = 3; EHEC ∆rfaD: n = 70, N = 3, P < 0.0001). d The representative confocal images of transgenic animal expressing mC::ACT-5 and CYK-1::GFP fed with OP50, EHEC, and the isogenic EHEC with rfaD deletion (∆rfaD) for 4 days. The DIC images and confocal images of the CYK-1::GFP signals, the mC::ACT-5 signals, the autofluorescence signals (UV), and the three fluorescence images overlaid over in each group (Merge) are presented. Scale bars represent 5 µm. e The quantification result of GFP intensity in the mC::ACT-5;CYK-1::GFP worms fed with OP50, EHEC, and the isogenic EHEC with rfaD deletion (∆rfaD) for 4 days in panel (d). *P < 0.05 compared to the EHEC wild-type group. f The representative SEM images of intestinal Caco-2 cell treated with Mock control, EHEC wild-type (EHEC), and the isogenic EHEC with rfaD deletion (∆rfaD) for 2 h. g The quantification results of EHEC-induced microvillar effacement in Caco-2 cell. EHEC significantly decreased the fraction of microvilli compared to mock control, and cell treated with ∆rfaD showed similar fraction of microvilli compared to the mock control. All experiment with ROI = 3 of each 425 μm² region. *P < 0.05, **P < 0.01 compared to the EHEC wild-type group. h The representative SEM images of the intestinal Caco-2 cell treated with mock control, EHEC wild-type (EHEC), and EHEC and the CDK-1 inhibitor RO3306 (10 nM). i The quantification results of EHEC-induced microvillar effacement in Caco-2 cells. RO3306 significantly reversed the microvillar effacement induced by EHEC. All experiment with ROI = 3 of each 425 μm² region. ***P < 0.001 compared to the EHEC wild-type group. j The representative SEM images of intestinal Caco-2 cell treated mock or EHEC with Scramble RNAi control or the shDIAPH1. k The quantification results of EHEC-induced microvillar effacement in Caco-2 cells. shDIAPH1 significantly abolished the microvillar effacement induced by EHEC. All experiment with ROI = 4 of each 425 μm² region. ***P < 0.001 compared to the RNAi mock group. Arrows indicate bacterial cells, and scale bars represent 5 µm in panels (f, h, and j). All quantitative data are presented as mean ± SEM, and each dot represented an independent result in the bar chart. All data statistics based on: *P < 0.05, **P < 0.01, and ***P < 0.001 by unpaired t test (two-tailed). n.s. Not significant. Source data are provided as a Source Data file.