Inhibition of O‐GlcNAcylation protects from Shiga toxin‐mediated cell injury and lethality in host

Abstract Shiga toxins (Stxs) produced by enterohemorrhagic Escherichia coli (EHEC) are the major virulence factors responsible for hemorrhagic colitis, which can lead to life‐threatening systemic complications including acute renal failure (hemolytic uremic syndrome) and neuropathy. Here, we report that O‐GlcNAcylation, a type of post‐translational modification, was acutely increased upon induction of endoplasmic reticulum (ER) stress in host cells by Stxs. Suppression of the abnormal Stx‐mediated increase in O‐GlcNAcylation effectively inhibited apoptotic and inflammatory responses in Stx‐susceptible cells. The protective effect of O‐GlcNAc inhibition for Stx‐mediated pathogenic responses was also verified using three‐dimensional (3D)‐cultured spheroids or organoids mimicking the human kidney. Treatment with an O‐GlcNAcylation inhibitor remarkably improved the major disease symptoms and survival rate for mice intraperitoneally injected with a lethal dose of Stx. In conclusion, this study elucidates O‐GlcNAcylation‐dependent pathogenic mechanisms of Stxs and demonstrates that inhibition of aberrant O‐GlcNAcylation is a potential approach to treat Stx‐mediated diseases.


A B
C Figure EV2. Suppression of Stx2a-and Stx1a-induced O-GlcNAcylation rescues cells from apoptosis.
A Representative images showing TUNEL staining of Stx2a-treated THP-1 cells cultured in the presence of OSMI-1 or the vehicle control at 0, 3, 6, and 9 h time point each. FITC-dUTP staining (green fluorescence) in the TUNEL assay indicates active progression of apoptosis. Scale bars: 40 µm. B WST-1 dye-based cell viability assay of Stx2a-exposed THP-1 cells treated with or without OSMI-1 (10 µM final) at 0, 3, 6, and 9 h time point each (n = 3 biological replicates). C Representative flow cytometric plot showing apoptosis progression in THP-1 cells detected by TUNEL upon treatment with Stx1a (100 ng/ml) in the presence or absence of OSMI-1 (10 µM final), and quantification of the percentage of apoptotic cells at 9 h (n = 3 biological replicates). The effects of OSMI-1 treatment were compared with those of the vehicle (DMSO) control.
Data information: Error bars for bar graphs are presented as mean AE SD Statistical analysis was performed using two-tailed Student's t-test. *P < 0.05; **P < 0.01; and ***P < 0.001. Source data are available online for this figure. A B Figure EV3. Akt and p65 were directly O-GlcNAcylated in Stx2a-treated THP-1 cells.
A Representative western blot showing changes in phosphorylation status of Akt or Bad in THP-1 cells treated with Stx2a (10 ng/ml) for 3 h in the presence or absence of OSMI-1 (10 µM, final) or OGA inhibitor Thiamet G (2 µM, final). B Representative western blot images, before (left) and after (right) pull down using WGA-lectin conjugated to agarose beads or bead-only control, to determine O-GlcNAc attachment to Akt and p65 in lysates from Stx2a (10 ng/ml)-exposed THP-1 cells for 9 h in the presence or absence of OSMI-1 (10 µM, final).
Source data are available online for this figure. A Heatmaps representing the comparative expression levels for DEGs in the presence or absence of OSMI-1 treatment upon Stx2a intoxication related to the immune responses. B Heatmaps representing the comparative expression levels for DEGs in the presence or absence of OSMI-1 treatment without Stx2a exposure related to the inflammatory response (left) and apoptotic process (right).

Kyung-Soo Lee et al
Data information: The numbers within the tables are normalized gene expression level compared to HRPTEpi cells maintained in the absence of either Stx2a or OSMI-1. Expression values are represented with red (upregulation) or blue (downregulation) color using FPKM values by Cufflinks; the cutoffs used a fold change of at least 1.5 followed by pairwise comparison and Student's t-test with a Benjamini and Hochberg correction. The FPKM values were normalized using EdgeR within R and visualized using ExDEGA. Data are the means from two independent replicates. A B Figure EV5. Treatment with an OGT inhibitor improves severe loss of body weight due to the hemorrhagic symptoms in the intestines of mice challenged with Stx2a.
A Body weight changes in mice challenged with Stx2a in the presence or absence of OSMI-1 (300 or 1,000 µg/mouse). Data are presented as mean AE SD (n = 5 biological replicates per group, two-tailed Student's t-test). Comparisons between days 0 and 3 were indicated for each group. *P < 0.05; **P < 0.01; and ***P < 0.001. B Representative images of the intestines of mice at day 3 after Stx2a injection and combinatorial treatment with two doses of OSMI-1 or the vehicle alone.
Source data are available online for this figure.