Mutation of the bacterial transcription factor MxiE results in ubiquitylation of S. flexneri in IFNγ-primed epithelial cells.

(A – C) Cells were primed with 100 U/ml IFNγ or left unprimed and then infected with WT or ΔmxiE S. flexneri at an MOI of 50 –100. Cells were fixed 3 – 4 h post infection (hpi) and stained for linear ubiquitin (M1-Ub) with anti-M1 antibody. Percentage of M1-Ub-positive bacteria was quantified in infected A549 epithelial cells at 4 hpi (B) and HT29 epithelial cells at 3 hpi (C). Graphs show the average of three independent experiments and depict means ± SEM. Two-way ANOVA with Tukey’s multiple comparison tests were performed; all statistically significant comparisons are shown. ***p<0.001; **** p<0.0001; ns, not significant

S. flexneri ΔmxiE mutants become decorated with linear and lysine-linked ubiquitin in IFNγ-primed A549 cells.

(A) WT A549 cells expressing the indicated internally Strep-tagged Ubiquitin (INT-Ub) variants were primed with 100 U/ml IFNγ overnight and co-localization of INT-Ub with cytosolic S. flexneri (Sf) ΔmxiE was assessed at 4 hpi. Ubiquitin linkage-specific antibodies were used to determine the percentage of S. flexneri ΔmxiE staining positive for K27-linked ubiquitin (K27-Ub) (B) or K63-linked ubiquitin (C) in IFNγ-primed untransduced WT A549 cells. Co-staining of anti-M1 and anti-ubiquitin (FK2) on the surface of S. flexneri ΔmxiE in IFNγ-primed (100U/ml) A549 cells is shown in (D). Quantification of anti-M1-FK2 co-staining in untreated and IFNγ-primed A549 cells is depicted in (E). All infections were done at an MOI of 50 – 100. Data were generated from at least three independent experiments and shows mean ± SEM. One-way ANOVA followed by Dunnet’s multiple comparisons were performed of all groups against WT-ubiquitin group (A). Two-way ANOVA with Tukey’s multiple comparison tests were performed for (B) and (C). An unpaired t-test was performed between “both FK2+M1” groups (gray bars) (E). *p<0.05; **p<0.01; ***p<0.001; **** p<0.0001

Ubiquitylation of S.flexneri ΔmxiE is dependent on RNF213 but not LUBAC

(A) Immunoblotting for HOIP, HOIL-1, and RNF213 protein expression in untreated and IFNγ-primed WT and the corresponding gene deletion (KO) A549 cells. (B) Percentage ΔmxiE S.flexneri bacteria stained with anti-M1-linked ubiquitin in IFNγ-primed WT, HOIPKO, HOIL-1KO and RNF213KO A549 cells. (C) Percentage of WT and 7KR INT-Ub-positive ΔmxiE S. flexneri in IFNγ-primed WT, HOIPKO, and RNF213KO A549 cells. (D – G) Untreated and IFNγ-primed A549 and HT29 cells were infected with the indicated S. flexneri strains and immuno-stained for RNF213 and ubiquitin (FK2). Representative immunofluorescence microscopy images are shown for A549 infections in (F). RNF213-S. flexneri colocalization percentages were quantified in A549 (D) and HT29 (E) cells. (G) Quantification of ubiquitin and RNF213 colocalization with ΔmxiE S. flexneri in A549 cells. Percentages of ΔmxiE S. flexneri staining positive for antibodies specific for K27-linked (H) and K63-linked ubiquitin (I) are also provided. All data are represented by the mean ± SEM from at least three independent experiments. One-way ANOVA followed by Dunnet’s multiple comparisons were performed of all groups against WT A549 group in (A). Two-way ANOVA with Tukey’s multiple comparison tests were performed in (C, D and F). an unpaired t-test was performed between the “Ub+ & RNF213+” groups (gray bars) in (G). For (H and I), an unpaired-t test was performed. Comparisons not shown are non-significant. **p<0.01;; **** p<0.0001

S. flexneri virulence factors IpaH1.4 and IpaH2.5 induce proteasomal degradation of RNF213

(A-B) All S. flexneri strains express PilT to enhance adhesion and infection rates and infections were carried out at an MOI of 5 – 25 for 3 hours (A) Untreated and IFNγ-primed (100 U/ml) A549 cells were infected with indicated S. flexneri strains expressing GFP. Protein lysates were probed for expression of RNF213, HOIP, and GFP. GFP expression serves as a bacterial protein loading control. (B) Infected and uninfected A549 cells were cultured in the presence of different concentrations of the proteasomal inhibitor MG132 and RNF213 expression was monitored by immunoblotting. (C) HEK 293T cells stably expressing mCherry-RNF213 were transiently transfected with individual GFP-tagged IpaH effectors for 24 hours and RNF213 expression was assessed. (D) WT and catalytically inactive C368S mutants of IpaH1.4 and IpaH2.5 were transiently transfected into HEK 293T cells expressing mCherry-RNF213 and cell lysates were subjected to immunoblotting. (C-D) Denaturation of cell lysates at lower temperature (56°C) was required for RNF213 detection but also resulted in the formation of double bands for all IpaH-GFP constructs. Images are representative of three independent experiments UI: Uninfected.

Loss of IpaH1.4 is sufficient to render S. flexneri susceptible to RNF213-driven ubiquitylation

(A) IFNγ-primed WT and HOIPKO A549 cells were infected with the indicated PilT+, GFP+ S. flexneri strains for 3 hours at an MOI of 25 and RNF213 protein levels were assessed by Western blotting. Bacterial GFP and host actin are used as host protein loading controls. (B) Percentages of RNF213+ WT and ΔipaH1.4 S. flexneri in IFNγ-primed A549 cells are shown. (C) Representative microscopy images depicting RNF213 recruitment to cytosolic ΔIpaH1.4 bacteria complemented with empty vector, WT IpaH1.4, or catalytically inactive IpaH1.4 (C368S) in IFNγ-primed WT A549 cells. (D) Percentages of RNF213+ indicated S. flexneri strains in IFNγ-primed WT A549 cells are shown. (E) Percentages WT and ΔIpaH1.4 S. flexneri stained positive for anti-M1-linked antibody in IFNγ-primed WT and RNF213KO A549 cells are depicted. (F) The ilux operon was introduced into S. flexneri strains to use bioluminescence relative light units (RLU) as a proxy for bacterial growth. RLU was measured in IFNγ-primed WT A549 cells infected with the indicated S. flexneri strains at an MOI of 5. (B, D, E) Data represent the mean ± SEM from at least three independent experiments. An unpaired t-test was conducted for (B). One-way ANOVA followed by Dunnet’s multiple comparisons were performed between all groups against the group ΔipaH1.4 expressing WT IpaH1.4 (D) and WT A549 cells (E). *p<0.05; ***p<0.001; **** p<0.0001

Internally STREP-tagged Ubiquitin (INT-Ub) constructs.

Ubiquitin lysines are shown in red. Strep tag sequence in highlighted in green

Addition of an N-terminal HA-tag significantly reduces targeting of 7KR ubiquitin to S. flexneri ΔmxiE.

N-terminally untagged and N-terminally HA-tagged versions of the internally Strep-tagged Ubiquitin (INT-Ub) 7KR mutant variant were expressed in A549 cells. Cells were primed with 100 U/ml IFNγ overnight and co-localization of INT-Ub with cytosolic S. flexneri (Sf) ΔmxiE was assessed at 4 hpi. An unpaired t-test was conducted. **p<0.01

Ubiquitin-Activated Interaction Trap (UBAIT) identifies mouse Rnf213 as likely interaction partner of IpaH2.5.

(A) Diagram showing the experimental procedure for UBAIT detection of IpaH2.5 and IpaH3 (control) substrates from the colon and caecum of mice infected with Salmonella enterica Typhimurium (S. Tm). (B) Mass spectrometry of hits identified in the Strep-IpaH2.5 UBAIT (left) as compared to Strep-IpaH3 UBAIT (right). The top 12 proteins identified in the Strep-IpaH2.5 UBAIT samples as determined by peptide abundance are shown. The corresponding peptide abundances from Strep-IpaH3 UBAIT are shown. RNF213 and the previously identified IpaH2.5 substrate RNF31 (HOIP) are highlighted. PSM = peptide spectrum match. ND = Not Detected. NR = Not Reported.

Deletion of ipaH1.4 but not ipaH2.5 results in RNF213 recruitment to cytosolic S. flexneri.

(A) A549 cells were primed with IFNγ (100 U/ml) and infected with the indicated strains expressing GFP for 4 hours at an MOI of 100. RNF213 recruitment to cytosolic bacteria was quantified. (B) IFNγ-primed A549 RNF213KO cells were infected with the indicated PilT+ S. flexneri strains for 3 hours at an MOI of 25 and HOIP protein levels were assessed by Western blotting. (C) Representative microscopy images showing anti-M1-linked ubiquitin staining of ΔipaH1.4 S. flexneri bacteria in IFNγ-primed WT A549 cells. Quantified data are shown in Fig. 5E. Percentages of WT and ΔipaH1.4 S. flexneri bacteria stained with anti-K27-linked ubiquitin (D) or anti-K63-linked ubiquitin (E) in IFNγ-primed WT A549 cells. (F) Percentages of anti-M1-linked ubiquitin positive ΔipaH1.4 S. flexneri bacteria in IFNγ-primed WT, RNF213KO, HOIL-1KO and HOIPKO A549 cells. Data are represented by the mean ± SEM from three independent experiments. An unpaired t-test was conducted for (D, E). Data are represented by the mean ± SEM from three independent experiments. Two-way ANOVA with Tukey’s multiple comparison tests were performed for (A,F); all statistically significant comparisons are shown. **** p<0.0001; ***p<0.001; **p<0.01; ns, not significant

Mouse Rnf213 binds to ΔipaH1.4 S. flexneri and facilitates linear ubiquitylation of cytosolic bacteria

(A) Immunoblotting for mouse Rnf213 protein expression in untreated and IFNγ-primed WT and IFNγ-primed Rnf213KO MEF cells. Each lane represents MEFs from separate mouse embryos that were used for infections. (B and C) MEF cells were primed with mouse IFNγ (100 U/ml) and infected with the indicated strains expressing GFP and PilT for 2 hours at an MOI of 100. Rnf213-S. flexneri colocalization percentages were quantified and are shown in (B). (C) Percentage of M1-linked ubiquitin positive S. flexneri in IFNγ-primed WT and Rnf213KO MEF cells. All data are represented by the mean ± SEM from at least three independent experiments. An unpaired t-test was performed in (B). Two-way ANOVA with Tukey’s multiple comparison tests were performed in (C). **p<0.01; **** p<0.0001