Human ZBP1 induces cell death-independent inflammatory signaling via RIPK3 and RIPK1

Abstract

ZBP1 is an interferon-induced cytosolic nucleic acid sensor that facilitates antiviral responses via RIPK3. Although ZBP1-mediated programmed cell death is widely described, whether and how it promotes inflammatory signaling is unclear. Here, we report a ZBP1-induced inflammatory signaling pathway mediated by K63- and M1-linked ubiquitin chains, which depends on RIPK1 and RIPK3 as scaffolds independently of cell death. In human HT29 cells, ZBP1 associated with RIPK1 and RIPK3 as well as ubiquitin ligases cIAP1 and LUBAC. ZBP1-induced K63- and M1-linked ubiquitination of RIPK1 and ZBP1 to promote TAK1- and IKK-mediated inflammatory signaling and cytokine production. Inhibition of caspase activity suppressed ZBP1-induced cell death but enhanced cytokine production in a RIPK1- and RIPK3 kinase activity-dependent manner. Lastly, we provide evidence that ZBP1 signaling contributes to SARS-CoV-2-induced cytokine production. Taken together, we describe a ZBP1-RIPK3-RIPK1-mediated inflammatory signaling pathway relayed by the scaffolding role of RIPKs and regulated by caspases, which may induce inflammation when ZBP1 is activated below the threshold needed to trigger a cell death response.

Keywords: RIPK1; RIPK3; SARS-CoV-2; ZBP1; inflammatory signaling.

Figure 1. ZBP1 stimulates inflammatory signaling independently of cell death

1. A

   Schematic illustration of WT and Zα1α2‐mutant (Zα1α2mut) human ZBP1 inducibly expressed in HT29/TO‐ZBP1 cells.

2. B

   Western blot analysis of the dose‐dependent expression of ZBP1 in HT29/TO‐ZBP1^WT and HT29/TO‐ZBP1^Zα1α2mut cells treated with indicated concentrations of Dox for 24 h. Blots are representative of three biological repeats.

3. C

   Flow cytometry analysis of ZBP1 expression in HT29/TO‐ZBP1 cells treated with Dox for 24 h. FLAG antibody staining was used to determine ZBP1‐expressing cells. Data are presented as mean with S.E.M (n = 5 biological replicates). Two‐way ANOVA and Tukey's multiple comparisons tests were used to test for statistical differences between indicated condition and DMSO‐treated condition within each cell line. ****P < 0.0001 for all conditions.

4. D, E

   Cell death analysis of HT29 and HT29/TO‐ZBP1 cells as indicated following 24 h treatment with Dox using SytoxGreen to stain dead cells. Data are presented as mean ± S.E.M ((D) n = 5 biological replicates, (E) n = 3 biological replicates). Two‐way ANOVA and Sidak's multiple comparisons tests were used to test for the statistical differences between indicated condition and untreated condition within each cell line. n.s., not significant (P ≥ 0.05); ***P = 0.0004; ****P < 0.0001.

5. F

   Relative viability of HT29/TO‐ZBP1^WT cells treated with 500 ng/ml Dox for up to 3 days in combination with 20 μM zVAD, 10 μM Nec1s, 10 μM GSK′872, and/or 1 μM NSA as indicated was determined by the CellTitre‐Glo assay. Values are normalized to untreated wells. Data are plotted as mean with S.E.M. (n = 4 biological replicates). Two‐way ANOVA and Dunnet's multiple comparisons tests were used to test for statistical differences between indicated condition and DMSO‐treated condition. n.s., not significant (P > 0.99); ***P = 0.0006 for zVAD, P = 0.0007 for zVAD+NSA; ****P < 0.0001.

6. G

   Cell death analysis of HT29/TO‐ZBP1^WT Clone B9 cells following treatment as indicated using SytoxGreen to stain dead cells. Cells were pretreated with 100 nM LCL161, 20 μM zVAD combined with10 μM GSK′840, 10 μM GSK′872, or 10 μM Nec1s for 1 h followed by treatment with 10 ng/ml TNF. Cell death analysis was carried out at the end of 24 h incubation with all compounds. Data are plotted as mean with S.E.M. (n = 3 biological replicates). Repeated measures one‐way ANOVA and Sidak's multiple comparisons tests were used to test for statistical differences between indicated condition and untreated control. ****P < 0.0001; n.s., not significant (P > 0.4).

7. H

   Cytokine concentration in the culture media of HT29/TO‐ZBP1 cells treated for 24 h with the indicated Dox concentrations or vehicle. Data are presented as mean with S.E.M (n = 4 biological replicates). Two‐way ANOVA and Sidak's multiple comparisons t‐test were used to test for the statistical differences between the two cell lines at each concentration. n.s., not significant (P > 0.5); **P = 0.0035 for CXCL8 at 20 ng/ml, P = 0.0067 for CXCL1 at 20 ng/ml, P = 0.0020 at 50 ng/ml, P = 0.0013 at 40 ng/ml; ***P = 0.0004 for CXCL8 at 25 ng/ml, P = 0.0008 at 30 ng/ml; ****P < 0.0001.

8. I

   Western blot analysis of cells from the same wells as used in (H) for ZBP1 expression levels, representative of four biological replicates.

9. J

   CXCL8 concentration in the culture media from the experiment described in (E). Data are presented as mean with S.E.M. (n = 3 biological replicates). Two‐way ANOVA tests and Sidak's multiple comparisons test were used to test for statistical differences between indicated conditions. **P = 0.0079; ****P < 0.0001.

10. K

    Transwell migration of THP1 cells toward conditioned media from HT29/TO‐ZBP1^WT cells treated with DMSO or 500 ng/ml Dox for 24 h. Data are presented as mean with S.E.M. (n = 3 biological replicates of conditioned media). An unpaired t‐test was used to test for statistical differences between indicated conditions. ****P < 0.0001.

11. L

    Transwell migration of primary neutrophils toward conditioned media from HT29/TO‐ZBP1^WT cells treated with DMSO or 500 ng/ml Dox for 24 h, or HT29 chemotaxis buffer (control buffer) containing equal volume and amount of DMSO or Dox as the conditioned media. Data are presented as individual values with grand mean of the migrated percentage induced by control buffer containing DMSO or Dox (n = 10 biological replicates), or conditioned media from DMSO‐treated cells (n = 11 biological replicates) or Dox‐treated cells (n = 12 biological replicates), where the number of biological replicates is defined by the total number of independent cell cultures tested on primary neutrophils from four donors. An unpaired t‐test was used to test for statistical differences between indicated conditions. ****P < 0.0001.

12. M

    Transwell migration of primary neutrophils towards conditioned media from HT29/TO‐ZBP1^WT cells treated with 0 or 500 ng/ml Dox in combination with DMSO or 20 μM zVAD +10 μM Nec1s for 24 h. Data are presented as individual values with mean and S.E.M. (n = 3 biological replicates of conditioned media). One‐way ANOVA and Sidak's multiple comparisons test were used to test for statistical differences between indicated conditions. n.s. = not significant (P = 0.4129); ***P = 0.0004.

Source data are available online for this figure.

Figure EV1. ZBP1 induces inflammatory signaling independently of cell death

1. A

   Western blot analysis of ZBP1 levels in HT29, HT29/TO‐ZBP1^WT, and HT29/TO‐ZBP1^Zα1α2mut cells treated or not with Dox or IFNβ (10 ng/ml) for 24 h. Blots are representative of two biological replicates.

2. B

   Cell death analysis of HT29 and HT29/TO‐ZBP1 cells as indicated following 48 h treatment with Dox using SytoxGreen to stain dead cells. Data are presented as mean ± S.E.M (n = 5 biological replicates). One‐way ANOVA and Sidak's multiple comparisons tests were used to test for the statistical differences between indicated conditions and untreated condition of the same cell line. *P = 0.0283; ****P < 0.0001.

3. C

   Relative viability of HT29/TO‐ZBP1^WT and HT29/TO‐ZBP1^Zα1α2mut cells treated with Dox for up to 72 h was determined by the CellTitre‐Glo assay. Values were normalized to that of 0 h for each cell line. Data are presented as mean with S.E.M (n = 3 biological replicates). Two‐way ANOVA and Tukey's multiple comparisons tests were used to test for the statistical differences between different cell lines. n.s. = not significant (P = 0.9510); ****P < 0.0001.

4. D

   Cell death analysis of HT29/TO‐ZBP1^WT cells by SytoxGreen staining following treatment with TSZ in combination with the indicated inhibitors. Cells were pretreated with 100 nM LCL161, 20 μM zVAD combined with10 μM GSK′840, 10 μM GSK′872, or 10 μM Nec1s for 1 h followed by treatment with 10 ng/ml TNF. Cell death analysis was carried out at the end of 24 h incubation with all compounds. Data are presented as mean (n = 2 biological replicates) with individual data points indicated.

5. E, F

   Cytokine concentrations in the culture media of HT29/TO‐ZBP1 cells treated with Dox for 24 h. Cells from the same wells were lysed for Western blot to determine ZBP1 expression levels. (E) Data are presented as mean with S.E.M (n = 6 biological replicates). Brown‐Forsythe and Welch ANOVA tests and Dunnet's T3 multiple comparisons test were used to test for statistical significances between indicated conditions. ***P = 0.0002; *P = 0.0447; ****P < 0.0001; n.s., not significant (P = 0.2153). Western blots are representative of six biological replicates. (F) Data are presented as mean with S.E.M (n = 4 biological replicates). Unpaired t‐tests were used to test for the statistical differences between the indicated conditions. *P = 0.0257; **P = 0.0034. Cell lysates from one biological replicate were analyzed by Western blotting.

6. G

   CXCL1 concentration in the culture media from the experiment described in (Fig 1E). Data are presented as mean with S.E.M. (n = 3 biological replicates). Two‐way ANOVA tests and Sidak's multiple comparisons test were used to test for statistical differences between indicated conditions. n.s., not significant (P > 0.09); ****P < 0.0001.

7. H

   CXCL8 concentration in the culture media of HT29/TO‐ZBP1^WT cells treated with 0 or 50 ng/ml Dox in combination with DMSO, 10 μM Nec1s, 10 μM GSK′872, or 1 μM NSA for 24 h. Data are plotted as mean with S.E.M. (n = 4 biological replicates). One‐way ANOVA and Sidak's multiple comparisons tests were used to test for statistical differences between indicated conditions. n.s. = not significant (P > 0.39); **P = 0.0012; ****P < 0.0001; *P = 0.0297. Cells from the same wells were analyzed by Western blotting for ZBP1 and RIPK3 levels. Blots are representative of three biological replicates.

Figure EV2. ZBP1 expression‐induced cytokine secretome promotes the chemotaxis of neutrophils

1. A, B

   Chemokine (A) and cytokine (B) arrays of the conditioned media from HT29/TO‐ZBP1^WT cells treated with DMSO or 500 ng/ml Dox for 24 h. Normalized pixel intensity values are presented relative to values from DMSO‐treated samples (n = 1 biological replicate).

2. C

   Transwell migration of primary neutrophils from Donor A towards conditioned media from HT29/TO‐ZBP1^WT cells treated with DMSO or 500 ng/ml Dox for 24 h, or control buffer containing equal volume and amount of DMSO or Dox as the conditioned media. Data are presented as mean with S.E.M. (n = 3 biological replicates). Two‐way ANOVA and Dunnet's multiple comparison tests were used to test for statistical differences between the migration of neutrophils toward conditioned media from DMSO‐treated cells and that from Dox‐treated cells at each time point. n.s. = not significant (P > 0.05); *P = 0.0113 for the 2 h time point, P = 0.0133 for the 3 h time point.

3. D

   Transwell migration of differentiated or undifferentiated HL60 cells toward control buffer (NT) or conditioned media from HT29/TO‐ZBP1^WT cells treated with DMSO or 500 ng/ml Dox for 24 h. Data from each biological replicate of differentiated HL60 cells and conditioned media are connected with solid lines (n = 4 biological replicates). A paired t‐test were used to determine the statistical difference between indicated conditions. *P = 0.0447.

Figure 2. ZBP1‐induced inflammatory signaling requires RIPK3 and RIPK1 but not their kinase activity

1. A

   Cytokine concentration in the culture media of HT29/TO‐ZBP1^WT cells treated with 0 or 50 ng/ml Dox in combination with indicated inhibitors for 24 h. TAK1i, 1 μM 5z‐7‐oxozeaenol. IKKi, 1 μM IKK Inhibitor VII + 5 μM IKK Inhibitor XII. Nec1s, 10 μM. GSK′872, 10 μM. NSA, 1 μM. Data are presented as mean with S.E.M. (n = 3 biological replicates). One‐way ANOVA and Sidak's multiple comparisons tests were used to test for statistical differences between Dox‐induced samples pretreated with DMSO and corresponding inhibitor. n.s. = not significant (P > 0.2); *P = 0.0104; **P = 0.0041; ****P < 0.0001.

2. B

   Relative viability of HT29/TO‐ZBP1^WT cells treated with the same conditions as in (A). Data are plotted as mean with S.E.M. (n = 4 biological replicates). One‐way ANOVA and Sidak's multiple comparisons tests were used to test for statistical differences between all conditions and DMSO‐treated condition, with significantly different conditions indicated. *P = 0.023; ***P = 0.0002.

3. C

   Transwell migration of primary neutrophils toward conditioned media from HT29/TO‐ZBP1^WT cells treated with 0 or 500 ng/ml Dox in combination with indicated inhibitors for 24 h. Data are presented as individual values with grand mean (n = 3 biological replicates of conditioned media from GSK′872 and NSA‐treated wells, n = 6 biological replicates of conditioned media from other conditions). Two‐way ANOVA and Sidak's multiple comparisons test were used to test for statistical differences between indicated conditions. ****P < 0.0001; n.s. = not significant (P > 0.96).

4. D

   CXCL8 concentration in the culture media of HT29/TO‐ZBP1^WT cells transfected with siRNA targeting mismatch sequence (MM) or RIPK3 and treated with Dox at indicated concentrations for 24 h. Data are presented as mean with S.E.M. (n = 4 biological replicates). Unpaired t‐tests were used to test for statistical differences between indicated conditions. **P = 0.0079. Cell lysates from one biological replicate were analyzed by Western blotting.

5. E

   Cytokine concentration in the culture media of HT29/TO‐ZBP1^WT or HT29/RIPK1‐KO clone aA3/TO‐ZBP1^WT cells treated with 0, 50 or 500 ng/ml Dox. data is plotted as mean with S.E.M. (n = 6 biological replicates). Brown‐Forsythe and Welch ANOVA tests and Dunnet's T3 multiple comparisons test were used to test for statistical differences between indicated conditions. Cell lysates were analyzed by Western blotting for ZBP1 levels. Blots are representative of six biological replicates. n.s. = not significant (P > 0.2); ***P = 0.0003 for CXCL8, P = 0.0008 for CXCL1; ****P < 0.0001.

6. F

   Relative viability of HT29/TO‐ZBP1^WT and HT29/RIPK1‐KO clone aA3/TO‐ZBP1^WT cells treated with 500 ng/ml Dox for up to 3 days. Values are normalized to day 0 of treatment within each cell line. Data are plotted as mean with S.E.M. (n = 3 biological replicates). Two‐way ANOVA and Sidak's multiple comparison tests were used to test for statistical differences between WT and RIPK1‐KO cells at the indicated time points after Dox treatment. ***P = 0.0003; ****P < 0.0001.

7. G, H

   NF‐κB activity in HEK293FT cells with and without stable expression of full‐length (FL) RIPK3‐2xFV or RIPK3ΔC‐2xFV (ΔC) transfected with dual luciferase reporters and ZBP1 or empty vector (EV) as indicated. In (H), inhibitors were added immediately after transfection: Nec1s, 10 μM. GSK′872, 10 μM. NSA, 1 μM. Reporter activities were measured 24 h after transfection. Data are presented as mean with S.E.M (n = 3 biological replicates). One‐way ANOVA and Sidak's multiple comparisons test were used to test for statistical differences between indicated conditions. n.s. = not significant (P > 0.05); ****P < 0.0001. (G) Cell lysates were analyzed by Western blotting. Blots are representative of three biological replicates. Asterisk indicates background signals of the antibody.

Source data are available online for this figure.

Figure EV3. RIPK1 mediates ZBP1‐induced inflammatory signaling and cell death

1. CXCL8 and CXCL1 concentration in the culture media of HT29/TO‐ZBP1^WT Clone B9 cells following 24 h treatment with 0 or 500 ng/ml Dox in combination with 10 μM GSK′840 or 10 μM GSK′872 as indicated. Data are presented as mean with S.E.M. (n = 3 biological replicates). Two‐way ANOVA tests and Tukey's multiple comparisons test were used to test for statistical differences between indicated conditions. n.s., not significant (P ≥ 0.05); *P = 0.0174; ***P = 0.0006; ****P < 0.0001.

2. Cytokine concentration in the culture media of HT29/TO‐ZBP1^WT or two independent clones of HT29/RIPK1‐KO/TO‐ZBP1^WT cells treated with Dox as indicated. N.D., not detected. Data are presented as mean with S.E.M. (n = 3 biological replicates). One‐way ANOVA and Sidak's multiple comparisons test were used to test for statistical differences between indicated conditions. n.s. = not significant (P > 0.05); **P = 0.0079 for CXCL1, P = 0.0012 for CXCL8; ***P = 0.0008; ****P < 0.0001. Cell lysates from the same wells were analyzed by Western blotting for ZBP1 and RIPK1 levels. Blots are representative of three biological replicates.

3. Relative viability of HT29/TO‐ZBP1^WT cells or two independent clones of HT29/RIPK1‐KO/TO‐ZBP1^WT cells after treatment with or without 500 ng/ml Dox for up to 3 days as indicated. Viability is normalized to no Dox‐treatment within the same cell line. Data are presented as mean with S.E.M. (n = 3 biological replicates). Two‐way ANOVA and Sidak's multiple comparison tests were used to test for statistical differences between WT and RIPK1‐KO cells after Dox treatment. ****P < 0.0001.

4. Two biological replicates of Ub‐conjugate enrichment analysis shown in Fig 3C.

Figure 3. K63‐Ub and M1‐Ub facilitate ZBP1 inflammatory signaling

1. A

   Western blot analysis of anti‐FLAG (ZBP1) immunoprecipitation from HT29/TO‐ZBP1^WT cells treated with 0 or 500 ng/ml Dox for 16 h. Blots are representative of three biological replicates. Immunoprecipitated material was treated or not with USP21 (1 μM) for deubiquitination. Input loaded was 5% for ZBP1 and 1% for co‐immunoprecipitants. Asterisk indicates antibody heavy chain signal in IP samples.

2. B

   Enrichment of Ub‐conjugates by GST‐1xUBA for analysis of ubiquitination status of RIPK1, ZBP1, and RIPK3 in HT29/TO‐ZBP1^WT cells treated with 0 or 500 ng/ml Dox for 16 h. Blots are representative of at least three biological replicates. After enrichment, samples were treated or not with USP21 (1 μM) for deubiquitination. Asterisk indicates unspecific bands. Arrows indicate RIPK3 signal.

3. C

   Enrichment of Ub conjugates for analysis of the ubiquitination status of RIPK1 and ZBP1 using GST‐1xUBA or linkage‐specific SUBs in HT29/TO‐ZBP1^WT cells treated with 0 or 500 ng/ml Dox for 16 h. See Fig EV3D for two biological replicates.

4. D

   Enrichment of Ub conjugates by GST‐1xUBA for analysis of ubiquitination status of RIPK1 and ZBP1 in HT29, HT29/TO‐ZBP1^WT, and HT29/RIPK1‐KO aA3/TO‐ZBP1^WT cells treated with 500 ng/ml Dox for 16 h. Blots are representative of four biological replicates. Asterisk indicates antibody background signal.

5. E

   NF‐κB activity in HEK293FT/RIPK3‐2xFV cells transfected with dual luciferase reporters, ZBP1 or EV, and GFP (Ctrl) or GFP‐tagged SUBs as indicated. Luciferase reporter activities were measured 24 h after transfection and normalized to GFP + EV‐transfected condition. Data are plotted as mean with S.E.M. (n = 4 biological replicates). One‐way ANOVA and Sidak's multiple comparisons tests were used to test for statistical differences between the indicated condition and ZBP1/GFP‐transfected condition. ****P < 0.0001. Cell lysates were analyzed by Western blotting to determine expression of ZBP1, GFP, or GFP‐tagged SUBs and Actin. Blots are representative of three biological replicates.

6. F, G

   NF‐κB activity in HEK293FT/RIPK3‐2xFV cells transfected with dual luciferase reporters, ZBP1 or EV, and variants of CYLD or OTULIN. Luciferase reporter activities were measured 24 h after transfection and normalized to ZBP1/EV‐transfected cells. Data are plotted as mean with S.E.M. Multiple Welch t‐tests were used to test for statistical differences between the indicated condition and the ZBP1/EV‐transfected condition. (F) n = 4 biological replicates. **P = 0.0012, n.s. = not significant (P = 0.2342). (G) n = 3 biological replicates. *P = 0.0137; n.s. = not significant (P = 0.8374). Cell lysates were analyzed by Western blotting to determine expression of ZBP1, CYLD or OTULIN, and Actin. Blots are representative of (F) four or (G) three biological replicates.

Source data are available online for this figure.

Figure 4. cIAP1 and LUBAC facilitate ZBP1‐induced inflammatory signaling

1. A

   NF‐κB activity in HEK293FT/RIPK3‐2xFV cells transfected with siRNA targeting HOIP (siHOIP) or a mismatch sequence (siMM), and 48 h later transfected with dual luciferase reporters and ZBP1 or GFP (control). Luciferase reporter activities were measured 24 h after transfection and normalized to the siMM+ZBP1 samples. Data are plotted as mean with S.E.M. (n = 4 biological replicates). A Welch's t‐test was used to test for the statistical difference between indicated conditions. ****P < 0.0001. HOIP knockdown and ZBP1 expression levels were analyzed by Western blotting. Blots are representative of three biological replicates.

2. B

   NF‐κB activity in HEK293FT/RIPK3‐2xFV cells treated with 0 or 100 nM CpA and transfected with dual luciferase reporters and ZBP1 or EV. Luciferase reporter activities were measured 24 h after transfection and normalized to the 0 nM CpA + ZBP1 sample. Data are plotted as mean with S.E.M. (n = 3 biological replicates). A Welch's t‐test was used to test for the statistical difference between indicated conditions. ****P < 0.0001.

3. C

   Cytokine concentration in the culture media of HT29/TO‐ZBP1^WT cells treated with 0 or 100 nM CpA and 0 or 50 ng/ml Dox as indicated for 24 h. Data are plotted as mean with S.E.M. (n = 3 biological replicates). Unpaired t‐tests were used to test for the statistical difference between indicated conditions. **P = 0.0049; *P = 0.0394.

4. D–F

   Western blot analysis of anti‐FLAG (ZBP1) immunoprecipitation from HT29 or HT29/TO‐ZBP1^WT or HT29/RIPK1‐KO clone aA3/TO‐ZBP1^WT cells treated with 0 or 500 ng/ml Dox for 16 h. Asterisk indicates antibody heavy chain signal in IP samples. Blots are representative of at least (D) two and (E–F) three biological replicates.

Source data are available online for this figure.

Figure 5. RIPK3 oligomerization induces ubiquitination‐mediated inflammatory signaling

1. A

   Schematic illustration of the chemical‐inducible RIPK3 oligomerization system.

2. B

   Western blot analysis of HCT116/RIPK3‐2xFV cells (FL and ΔC) treated with 100 nM dimerizer. Blots are representative of two biological replicates.

3. C

   Time courses of mRNA levels of CXCL8 and CXCL1 in HCT116/RIPK3‐2xFV cells (FL and ΔC) in response to treatment with 100 nM dimerizer. Data are plotted as mean with S.E.M. (n = 3 biological replicates). Two‐way ANOVA and Tukey's multiple comparisons test were used to test for statistical differences between 0 h and indicated time points in FL cells. n.s. = not significant (P = 0.1360); **P = 0.0046 for FL 2 h, P = 0.0012 for FL 3 h; ****P < 0.0001; ***P = 0.0008.

4. D

   CXCL8 concentration in the culture media of HCT116/RIPK3‐2xFV cells (WT and RHIM^mut) treated with 100 nM dimerizer for 24 h. In RIPK3‐RHIM^mut three key residues, VQV, of the RHIM region was changed to AAAA. Data are plotted as mean with S.E.M. (n = 5 biological replicates). One‐way ANOVA and Sidak's multiple comparisons tests were used to test for statistical differences between indicated time points. **P = 0.0011; n.s. = not significant (P = 0.9301). Inset: Western blot analysis to determine RIPK3 expression levels in one biological replicate. Line indicates that image was cut and spliced to remove nonrelevant lanes from the scanned blot.

5. E

   Relative viability of HCT116/RIPK3‐2xFV cells treated with 0 or 100 nM dimerizer for 24 h. Data are presented as mean with S.E.M. (n = 3 biological replicates). A Welch's t‐test was used to test for statistical differences between indicated conditions. n.s. = not significant (P = 0.2294).

6. F

   Cell death analysis of HCT116/RIPK3‐2xFV cells treated with 0 or 100 nM dimerizer or TSZ (100 nM CpA, 20 μM zVAD and 1 ng/ml TNF) for 24 h using SytoxGreen to stain dead cells. Data are plotted as mean with S.E.M. (n = 3 biological replicates). One‐way ANOVA and Sidak's multiple comparisons test were used to test for statistical difference between indicated conditions. n.s., not significant (P = 0.1032); **P = 0.0014.

7. G, H

   Relative mRNA levels of CXCL8 and CXCL1 in HCT116/RIPK3‐2xFV cells pretreated with DMSO, TAK1 inhibitor, or IKK inhibitors for 1 h or 100 nM CpA for 0.5 h before treatment with 0 or 100 nM dimerizer for 3 h. Data are presented as mean with S.E.M. (n = 3 biological replicates). One‐way ANOVA and Sidak's multiple comparisons tests were used to test for statistical differences between indicated conditions and DMSO+dimerizer–treated condition. ****P < 0.0001. (H) Cell lysates were analyzed by Western blotting for cIAP1 levels. Blots are representative of two biological replicates.

8. I

   CXCL8 concentration in the culture media of HCT116/RIPK3‐2xFV cells pretreated with 100 nM CpA for 1 h before treatment with 0 or 100 nM dimerizer for 24 h. Data are presented as mean with S.E.M. (n = 3 biological replicates). An unpaired t‐test was used to test for the statistical difference between indicated conditions. **P = 0.0042.

9. J

   Relative CXCL8 mRNA levels in WT or HOIP‐knockout HCT116/RIPK3‐2xFV cells treated with 0 or 100 nM dimerizer for 3 h. Data are plotted as mean with S.E.M. (n = 3 biological replicates). A Welch's t‐test was used to test for the statistical difference as indicated. ***P = 0.0010. Cell lysates were loaded for Western blot analysis.

10. K

    Relative CXCL8 mRNA levels in HCT116/Tet‐On‐GFP‐K63‐SUB/RIPK3‐2xFV, HCT116/Tet‐On‐GFP‐M1‐SUB/RIPK3‐2xFV and HCT116/Tet‐On‐GFP/RIPK3‐2xFV cells treated with 0 or 100 ng/ml Dox for 48 h before stimulated with 0 or 100 nM dimerizer for 3 h. Data are presented as mean with S.E.M. (n = 4 biological replicates). Brown‐Forsythe and Welch ANOVA tests and Dunnet's T3 multiple comparisons test were used to test for statistical differences between indicated conditions. n.s. = not significant (P = 0.9997); **P = 0.0030; *P = 0.0366. Cells were analyzed by Western blotting for the inducible‐expression levels of GFP (control) and GFP‐SUBs. Blots are representative of three biological replicates.

11. L

    Relative chemokine mRNA levels in HCT116/RIPK3‐2xFV cells pretreated with 0 or 20 μM zVAD in combination with DMSO, 10 μM Nec1s, 10 μM GSK′872, or 1 μM NSA for 1 h, followed by treatment with 0 or 100 nM dimerizer for 3 h. Data are plotted as mean with S.E.M. (n = 3 biological replicates). ****P < 0.0001. One‐way ANOVA and Sidak's multiple comparison test were used to test for statistical differences between indicated conditions.

12. M

    Relative CXCL8 mRNA levels in HCT116/RIPK3‐2xFV cells stably knocked down against mismatch (shMM) or two different sites of caspase‐8 (shCASP8‐A and shCASP8‐B), pretreated with DMSO, 10 μM Nec1s, 10 μM GSK′872, or 1 μM NSA for 1 h followed by treatment with 0 or 100 nM dimerizer for 3 h. Data are plotted as mean with S.E.M. (n = 3 biological replicates). Two‐way ANOVA and Tukey's multiple comparison tests were used to test for statistical differences between indicated conditions. ***P = 0.0001; **P = 0.0035. Cell lysates from one biological replicate were analyzed by Western blotting to determine caspase‐8 levels.

Source data are available online for this figure.

Figure EV4. RIPK3 oligomerization induces inflammatory signaling

1. A

   Relative viability of HT29 or HT29/RIPK3‐2xFV cells pretreated with combinations of 100 nM CpA, 20 μM zVAD, and 10 μM Nec1s or DMSO (−) as indicated for 1 h, and then treated with or without 2 ng/ml TNF for 24 h. Viability is normalized to the DMSO‐only condition for each cell line. Data are presented as mean with S.E.M. (n = 3 biological replicates). Two‐way ANOVA and Sidak's multiple comparison test were used to test for statistical differences between indicated conditions and DMSO‐treated condition of each cell line. **P = 0.0044; ****P < 0.0001.

2. B

   Western blot analysis of necroptosis markers in HT29 cells pretreated with 100 nM CpA and 20 μM zVAD for 1 h before stimulating with 20 ng/ml TNF as indicated. Blots are representative of four biological replicates.

3. C

   Western blot analyses of inflammatory marker proteins in HT29/RIPK3‐2xFV cells treated with 100 nM dimerizer. Blots are representative of two biological replicates.

4. D, E

   Relative CXCL8 expression (D) and CXCL8 concentration in culture media (E) of HT29/RIPK3‐2xFV cells treated with 0 or 100 nM dimerizer as indicated. Data are presented as mean with S.E.M. (n = 3 biological replicates). A Welch's t‐test was used to test for statistical differences between indicated conditions. **P = 0.0016 (D); **P = 0.0093 (E).

5. F

   Relative viability of HT29/RIPK3‐2xFV cells treated with 0 or 100 nM dimerizer for 24 h in the presence of 0.5 μg/ml mouse IgG. Data are presented as mean with S.E.M. (n = 3 biological replicates). A Welch's t‐test was used to test for statistical differences between indicated conditions. n.s. = not significant (P = 0.3268).

6. G, H

   Western blot analysis of (G) HEK293FT/RIPK3‐2xFV cells (FL or ΔC) and (H) U2OS/NOD2 cells stably expressing RIPK3 variants treated with 100 nM dimerizer as indicated. Blots are representative of two biological replicates.

7. I

   Time course of relative CXCL8 and TNF expression in U2OS/NOD2/RIPK3‐2xFV cells (FL or ΔC) treated with 100 nM dimerizer as indicated. Data are plotted as mean with S.E.M. (n = 3 biological replicates). Two‐way ANOVA and Tukey's multiple comparisons test were used to test for statistical differences between 0 h and indicated time points in FL cells. **P = 0.0021 for 2 h, P = 0.0010 for 3 h; n.s. = not significant (P = 0.6323); ****P < 0.0001.

8. J, K

   Relative CXCL8 and TNF expression (J) and CXCL8 concentration in the culture media (K) of HT29/RIPK3‐2xFV cells pretreated or not with 2 μM CpA for 1 h before treated or not with 100 nM dimerizer for 3 h. Data are plotted as mean with S.E.M. (n = 3 biological replicates). Unpaired t‐tests were used to test for statistical differences between indicated conditions. ****P < 0.0001; *P = 0.0145 (J). A Welch's t‐test was used to test for statistical differences between indicated conditions. **P = 0.0097 (K).

9. L

   Relative CXCL8 expression in WT or XIAP‐knockout HCT116/RIPK3‐2xFV cells treated with or without 100 nM dimerizer for 3 h. Data are plotted as mean with S.E.M. (n = 3 biological replicates). A Welch's t‐test were used to test for statistical differences between indicated conditions. n.s. = not significant (P = 0.2057). Cell lysates were analyzed by Western blotting. Blots are representative of two biological replicates.

Figure EV5. ZBP1 mediates SARS‐CoV‐2‐induced inflammation

1. A

   Relative expression of ZBP1 in post‐mortem lung samples of two COVID‐19 patients compared with healthy lung biopsies (Blanco‐Melo et al, , Data ref: Blanco‐Melo et al, 2020b).

2. B

   Patient‐averaged single cell Transcript Per Million (TPM) values of ZBP1 in lung and peripheral blood of COVID‐19 patients in progressive or convalescent stage, compared to healthy controls (Ren et al, , Data ref: Ren et al, 2021b). Data are plotted as individual values per patient with mean and S.E.M. (n = 25 for healthy, n = 77 for progression, n = 102 for convalescence patients). Kruskai–Wallis test and Dunn's multiple comparisons test were used to test for statistical differences between indicated conditions. ****P < 0.0001; *P = 0.0375; **P = 0.0036.

3. C

   Pearson's correlation of individual single cell TPM values of indicated genes with virus load (Ren et al, , Data ref: Ren et al, 2021b). Data are presented as individual values from each cell, with linear regression line and its 95% confidence bands. ZBP1, n = 720, IL‐6, n = 94, TNF, n = 211, CXCL10, n = 1,053, CXCL8, n = 1,463, ACTB, n = 2,683 (cells). The positive correlation is significant (P < 0.0001) between TPM (virus) and TPM (ZBP1), TPM (IL‐6), TPM (TNF), TPM (CXCL10), or TPM (CXCL8). Correlation between TPM (virus) and TPM (ACTB) is not significant (P = 0.7862).

4. D

   Fold of mRNA levels of SARS‐CoV‐2‐encoded N1 and N2 (C) over β‐Actin (ACTB, control), at indicated time after infection in Calu‐3 cells, normalized over mock 16 h. Data is presented as mean with S.E.M (n = 3 biological replicates). Two‐way ANOVA and Tukey's multiple comparison tests were used to test for the statistical differences between each time point with 16 h within the same MOI. n.s., not significant (P > 0.4); *P = 0.0376; ****P < 0.0001.

5. E

   Western blot analysis of SARS‐CoV‐2‐infected Calu‐3 cells at MOI = 2 for the indicated time. Analysis was performed on one biological replicate.

6. F, G

   Expression of IFNB1 and IFIT1 (F) or SARS‐CoV‐2‐encoded N1 and N2 (G) relative to β‐Actin in Calu‐3 cells with stable knockdown of ZBP1 (shZBP1‐50, shZBP1‐52 and shZBP1‐53) and control cells (shMM) infected with mock or SARS‐CoV‐2 virus at MOI = 2 for 72 h. Data are presented as mean with S.E.M (n = 3 biological replicates). One‐way ANOVA and Sidak's multiple comparisons test were used to test for statistical differences between indicated conditions. n.s. not significant (P = 0.3831); ***P = 0.0006 for IFNB1, P = 0.0003 for IFIT1; *P = 0.0143; ****P < 0.0001 (F) and to test for statistical differences between shMM MOI = 2 and indicated conditions. *P = 0.0294 for shZBP1‐50, P = 0.0259 for shZBP1‐53; ***P = 0.0008; n.s., not significant (P > 0.05) (G).

Figure 6. ZBP1 mediates SARS‐CoV‐2‐induced cytokine production

1. A

   Western blot analysis of ZBP1, RIPK1, RIPK3, and MLKL levels in Calu‐3 cell lines with stable expression of shRNAs targeting ZBP1 or a mismatch sequence (shMM), untreated (NT) or treated with 10 ng/ml IFNβ for 48 h. Blots are representative of two biological replicates.

2. B, C

   Relative mRNA levels (B) or cytokine levels in the culture media (C) of indicated genes in Calu‐3 cells infected with mock or SARS‐CoV‐2 viruses at indicated MOIs for 16, 24, 48 or 72 h. Data are plotted as mean with S.E.M. (n = 3 biological replicates). Two‐way ANOVA and Tukey's (B) or Dunnet's (C) multiple comparison test were used to test for statistical differences between each infected sample and mock‐treated sample and between the two different MOIs at 72 h time point. For (B): ****P < 0.0001; ***P = 0.0006, **P = 0.0022 for ZBP1, P = 0.0072 for CXCL10; n.s. = not significant (P > 0.05). For (C): ****P < 0.0001; ***P = 0.0002; *P = 0.0244; n.s. = not significant (P > 0.05).

3. D, E

   Relative mRNA levels (D) or secreted cytokine levels (E) of indicated genes in Calu‐3 cells knocked down against mismatch sequence (shMM) and ZBP1 at three different sites (shZBP1‐50, shZBP1‐52 and shZBP1‐53) infected with mock or SARS‐CoV‐2 virus at MOI = 2 for 72 h. Data are presented as mean with S.E.M. (n = 3 biological replicates). One‐way ANOVA and Sidak's multiple comparisons test were used to test for statistical differences between indicated conditions. n.s. = not significant, *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

4. F

   Schematic model of the proposed ZBP1‐induced signaling pathways for inflammatory responses and cell death in human cells.

Source data are available online for this figure.