Activation of the NLRP1B inflammasome by caspase-8

Abstract

Cleavage of the innate immune receptor NLRP1B by various microbial proteases causes the proteasomal degradation of its N-terminal fragment and the subsequent release of a C-terminal fragment that forms an inflammasome. We reported previously that metabolic stress caused by intracellular bacteria triggers NLRP1B activation, but the mechanism by which this occurs was not elucidated. Here we demonstrate that TLR4 signaling in metabolically stressed macrophages promotes the formation of a TRIF/RIPK1/caspase-8 complex. Caspase-8 activity, induced downstream of this TLR4 pathway or through a distinct TNF receptor pathway, causes cleavage and activation of NLRP1B, which facilitates the maturation of both pro-caspase-1 and pro-caspase-8. Thus, our findings indicate that caspase-8 and NLRP1B generate a positive feedback loop that amplifies cell death processes and promotes a pro-inflammatory response through caspase-1. The ability of NLRP1B to detect caspase-8 activity suggests that this pattern recognition receptor may play a role in the defense against a variety of pathogens that induce apoptosis.

Fig. 1. LPS/2DG-mediated cell death is dependent on caspase-8 and not on the NLRP1B inflammasome.

a, b RAW264.7 cells (wild-type, Nlrp1b^–/–, or Casp1^–/–) were treated for 3 h with LPS and 2DG. Cells were stained with propidium iodide (PI) and analyzed by flow cytometry. c RAW264.7 cells were treated with LPS/2DG, a pan-caspase inhibitor (Z-VAD) or a caspase-8 inhibitor (Z-IETD) for 3 h and were stained with PI and analyzed by flow cytometry. d RAW264.7 cells were treated with LPS/2DG, a pan-caspase inhibitor (Z-VAD) or a caspase-8 inhibitor (Z-IETD) as indicated for 2 h and cell lysates were immunoblotted for caspase-1, caspase-8, and β-actin. e RAW264.7 cells were treated with LPS/2DG for indicated times. Cell lysates were immunoblotted for caspase-1, caspase-8, and β-actin. f–i RAW264.7 cells (wild-type, Casp8^–/– #1, Casp8^–/– #2 pool, Nlrp1b^–/–) were treated with LPS/2DG for 3 h (f) or 2 h (g–i). Cells were stained with PI and analyzed by flow cytometry or cell lysates were immunoblotted for caspase-1, caspase-8, and β-actin. j Schematic of NLRP1B activation by caspase-8. Made using Biorender.com. Blots are representative of three independent experiments. Error bars represent standard error of the mean of three independent experiments. Statistical significance was determined using a one-way (a, c) or (b, f) two-way ANOVA followed by a Sidak (a, b) or Tukey’s (c, f) post hoc test. *p < 0.05; **p < 0.01; ****p < 0.0001; NS nonsignificant.

Fig. 2. TLR4 and TRIF are required for LPS/2DG-mediated cell death.

a RAW264.7 cells were treated with LPS and Actinomycin D (ActD) for 2 h and analyzed for Il1b transcript levels using quantitative real-time PCR. b RAW264.7 cells were treated with LPS, 2DG and ActD for 2 h. Cell lysates were immunoblotted for caspase-1, caspase-8 and β-actin. c, d RAW264.7 cells (wild-type, Tlr4^–/–, Myd88^–/–, Ticam1^–/–) were treated with LPS/2DG for 3 h (c) or 2 h (d) (Ticam1 encodes TRIF). Cells were stained with propidium iodide (PI) and analyzed by flow cytometry; or cell lysates were immunoblotted for caspase-1, caspase-8, and β-actin. e RAW264.7 cells were treated with LPS, 2DG and dynasore for 2 h. Cell lysates were immunoblotted for caspase-1, caspase-8 and β-actin. f Bone marrow-derived macrophages (BMDMs) were treated with LPS, 2DG and dynasore for 2 h. Cell lysates were immunoblotted for caspase-1, caspase-8 and β-actin. Blots are representative of three independent experiments. Error bars represent standard error of the mean of three independent experiments. Statistical significance was determined by a one-way (a) or two-way ANOVA followed by a Tukey’s post hoc test. *p < 0.05; ****p < 0.0001; NS nonsignificant.

Fig. 3. Hypo-phosphorylated RIPK recruits pro-caspase-8 to TLR4/TRIF in LPS/2DG-treated cells.

a RAW264.7 cells (wild-type or Casp8^–/–) were treated for 1 h with LPS and 2DG. Endogenous RIPK1 was immunoprecipitated from cell lysates with an anti-RIPK1 antibody and then immunoblotted for pro-caspase-8 and RIPK1. Mouse IgG1 was used as a control. b, c RAW264.7 cells (wild-type or Ticam^-/-) were treated for 30 min with LPS, 2DG and the TAK1 inhibitor 5z-7-Oxozeaenol (5z-7) as indicated. Endogenous RIPK1 was immunoprecipitated from cell lysates and immunoblotted for TLR4 and RIPK1. d RAW264.7 cells were treated for 30 min with LPS, 2DG and 5z-7 and cell lysates were immunoblotted for RIPK1. e RAW264.7 cells were treated for 30 min with LPS. Cell lysates were incubated in Alkaline Phosphatase (AP) for 30 min and then immunoblotted for RIPK1. f RAW264.7 cells were treated for 2 h with LPS, 2DG and 5z-7. Cell lysates were immunoblotted for pro-caspase-8. g C57BL/6 BMDMs were treated as in (d). h RAW264.7 and (i) C57BL/6 BMDMs were treated with LPS, 2DG, for 2 h and TNFα/SMAC mimetic/Z-VAD (T/S/Z) for 3 h and immunoblotted for pro-caspase-8, phospho-MLKL S345, MLKL, and β-actin. Blots are representative of three independent experiments.

Fig. 4. Inhibition of IKKβ by 2DG prevents LPS-dependent phosphorylation of RIPK1 to drive caspase-8 activation.

a RAW264.7 cells were stimulated for 30 min with LPS, 2DG, 5z7 (TAK1 inhibitor), TPCA1 (IKKβ inhibitor), SB203580 (p38i; p38 inhibitor), and MRT67307 (TBK1/IKKε inhibitor) as indicated. Cell lysates were immunoblotted for RIPK1 and tubulin. b RAW264.7 cells were treated for 2 h with LPS, SB203580, 5z7, TPCA1, and 2DG. Cell lysates were immunoblotted for caspase-8 and β-actin. c RAW264.7 cells were treated for 15 and 30 min with LPS, TPCA1, and 2DG. Cell lysates were immunoblotted for phospho-Serine-536-p65 NFκB, total p65-NFκB, and β-actin. d RAW264.7 cells were treated for 30 and 60 min with LPS, MRT67307 and 2DG. Cell lysates were immunoblotted for phospho-Serine 396-IRF3, total IRF3, and β-actin. e Schematic model of the two-step phosphorylation cascade that leads to IKKβ activation. RAW264.7 cells were stimulated for 15 min with LPS, 5z7, TPCA1, and 2DG. Cell lysates were immunoblotted for phospho-Serine 176/177 IKKα/β, phospho-Serine 176/177 and Serine 180/181 IKKα/β, total IKKβ, and β-actin. f HEK293T cells were transfected with the indicated plasmids for 22 h followed by a 2 h treatment with 2DG. Cell lysates were immunoblotted for phospho-Serine 536 p65 NFκB, total p65 NFκB, and β-actin. All blots are representative of three independent experiments. Asterisk (*) indicates non-specific band. Schematic made with Biorender.

Fig. 5. Shigella flexneri induces caspase-8-dependent inflammasome activation.

a RAW264.7 cells were incubated with Shigella flexneri at indicated multiplicities of infection (MOI) for 1 h. Cell lysates were immunoblotted for caspase-1, pro-caspase-8, cleaved caspase-8, and β-actin. b, c Shigella flexneri (wild-type, M90T; and mutant, BS176) were incubated with RAW264.7 cells at an MOI of 5 for 1 h. Intracellular ATP concentrations were measured and cell lysates were immunoblotted for caspase-1, pro-caspase-8, cleaved caspase-8 and β-actin. d RAW264.7 cells (wild-type, Nlrp1b^–/–, Tlr4^–/–, Casp8^–/–) were incubated with Shigella flexneri at an MOI of 5 for 1 h. Cell lysates were immunoblotted for caspase-1, pro-caspase-8, cleaved caspase-8, and β-actin. e, f RAW264.7 cells were incubated with lethal toxin (LT) for 3 h, Shigella flexneri for 1 h, LPS/2DG, and MG132 for 2 h. Cell lysates were immunoblotted for caspase-1, GSDMD, and β-actin. Blots are representative of at least three independent experiments. Error bars represent standard error of the mean of three independent experiments. Statistical significance was determined using a one-way ANOVA followed by a Tukey’s post hoc test. ***p < 0.001; NS nonsignificant.

Fig. 6. Caspase-3, but not membrane damage, facilitates inflammasome activation.

a, b RAW264.7 cells and (c) C57BL/6 BMDMs were treated with LPS and 2DG for 2 h. Cell lysates were immunoblotted for GSDMD, GSDME, Pannexin-1 and β-actin. RAW264.7 cells (wild-type, Gsdmd^–/–Gsdme^–/–) were treated with LPS/2DG and trovafloxacin for 2 h (d) or 3 h (e). Cells were stained with PI and analyzed by flow cytometry, or cell lysates were immunoblotted for caspase-1 and β-actin. f, g RAW264.7 cells (wild-type, Casp3^–/–) were treated with LPS/2DG for 2 h (f) or 3 h (g). Cells were stained with PI and analyzed by flow cytometry, or cell lysates were immunoblotted for caspase-1, GSDMD, GSDME and β-actin. Blots are representative of three independent experiments. Error bars represent standard error of the mean of three independent experiments. Statistical significance was determine using a two-way ANOVA followed by a Tukey’s (e) or Sidak (f) post hoc test. ****p < 0.0001; NS nonsignificant.

Fig. 7. Caspase-8 induces proteasome-independent cleavage of NLRP1B.

a–c RAW264.7 cells (wild-type, Casp8^–/–, Nlrp1b^–/–) were treated as indicated for 2 h. Cell lysates were immunoblotted for pro-caspase-8, pro-caspase-1 and β-actin. d Schematic of caspase-1 activation by weak and strong stimuli of caspase-8. Made using Biorender.com. e–g RAW264.7 cells (wild-type, Casp3^–/–, Casp8^–/– #1, Casp8^–/– #2 pool, Nlrp1b^–/–) were treated as indicated for 2 h. Cell lysates were immunoblotted for pro-caspase-8, NLRP1B, cleaved caspase-3, and β-actin. h, i BALB/c BMDMs were treated as indicated for 2 h. Cell lysates were immunoblotted for pro-caspase-8, NLRP1B, and β-actin. Blots are representative of three independent experiments except (h) where n = 2. Asterisk (*) indicates NLRP1B cleavage products.

Fig. 8. Caspase-8-mediated cleavage of NLRP1B signals inflammasome formation.

a HEK293T cells were transfected with the indicated plasmids for 24 h. Cell lysates were immunoblotted for pro-caspase-8, NLRP1B, and beta-actin. (b, Top panel) Schematic of NLRP1B WT and the NLRP1B_Δ749-870 mutant lacking the linker between the LRR and FIIND domain. (b, Bottom panel) HEK293T cells were transfected with the indicated plasmids for 24 h. Cell lysates were immunoblotted for pro-caspase-8, NLRP1B, and β-actin. (c, Top panel) Schematic of murine NLRP1B allele 1 with the amino acid sequence between residues 785 and 849 with red indicating residues mutated to alanine. (c, Bottom panel) d HEK293T cells were transfected with the indicated plasmids for 24 h. Cell lysates were immunoblotted for pro-caspase-8, NLRP1B and β-actin. e Activated caspase-8 cleaves NLRP1B to form an inflammasome that activates caspase-8 and caspase-1. Asterisk (*) indicates cleavage products of NLRP1B. All immunoblots are representative of three independent experiments. Schematics made using Biorender.