The tetrapeptide sequence of IL-18 and IL-1β regulates their recruitment and activation by inflammatory caspases

Abstract

Inflammasomes are multiprotein signaling complexes that activate the innate immune system. Canonical inflammasomes recruit and activate caspase-1, which then cleaves and activates IL-1β and IL-18, as well as gasdermin D (GSDMD) to induce pyroptosis. In contrast, non-canonical inflammasomes, caspases-4/-5 (CASP4/5) in humans and caspase-11 (CASP11) in mice, are known to cleave GSDMD, but their role in direct processing of other substrates besides GSDMD has remained unknown. Here, we show that CASP4/5 but not CASP11 can directly cleave and activate IL-18. However, CASP4/5/11 can all cleave IL-1β to generate a 27-kDa fragment that deactivates IL-1β signaling. Mechanistically, we demonstrate that the sequence identity of the tetrapeptide sequence adjacent to the caspase cleavage site regulates IL-18 and IL-1β recruitment and activation. Altogether, we have identified new substrates of the non-canonical inflammasomes and reveal key mechanistic details regulating inflammation that may aid in developing new therapeutics for immune-related disorders.

Keywords: CP: Immunology; IL-18; IL-1β; caspase-1; caspase-11; caspase-4; caspase-5; cytokines; inflammasomes; pyroptosis.

Figure 1.. Autoproteolyzed inflammatory caspase species differentially bind to inflammatory substrates

(A) Schematic of human CASP1 (top), CASP4 (middle), and CASP5 (bottom) depicting the catalytic cysteines and autoproteolytic sites. (B–G) Immunoprecipitation depicting binding of inflammatory caspase species to GSDMD and IL-1β (B, D, and F) or IL-18 (C, E, and G). Data are representative of three or more independent experiments.

Figure 2.. CASP4/5 cleave IL-1β and IL-18

(A and B) HEK293T cells expressing either GSDMD-V5 and IL-1β-Myc (A) or IL-18-V5 (B) were transiently transfected with the indicated constructs for 24 h before LDH analysis. (C and D) HEK293T cells expressing either GSDMD-V5 and IL-1β-Myc (C) or IL-18-V5 (D) were transiently transfected with the indicated constructs for 24 h and then supernatants and lysates were combined for immunoblotting. (E) Parent (wild-type) or CASP3 KO HEK293T cells were transiently transfected with plasmids coding for IL-18 and IL-1β or co-transfected with the indicated caspases. After 24 h, the caspase co-transfected samples were harvested, and the remaining samples were treated with DMSO, etoposide (100 μM), staurosporine (0.5 μM), or TRAIL (150 ng/mL) for an additional 24 h before immunoblotting. (F and G) Immunoprecipitation depicting binding of inflammatory caspase species to GSDMD and IL-1β (F) or IL-18 (G). Data are means ± SEM of three biological replicates. ***p < 0.001, **p < 0.01, and *p < 0.05 by two-sided Student’s t test. * = non-specific bands. Data are representative of three or more independent experiments.

Figure 3.. The P4–P1 tetrapeptide sequence of IL-18 and IL-1β regulates their recruitment and processing by inflammatory caspases

(A) Schematic depicting the sequence of wild-type IL-1β, IL-18, IL-1α, the caspases that were reported to cleave these substrates, and the tetrapeptide mutants. (B and C) HEK293T cells expressing IL-18-HA were transiently co-transfected with the indicated active caspase and IL-18 tetrapeptide mutants (B), or HEK cells were co-transfected with the indicated caspases and IL-1β tetrapeptide mutants (C) for 24 h prior to immunoblot analysis. (D–H) The indicated IL-18 (D) or IL-1β (E) constructs were purified from HEK293T cells and incubated with one activity unit (for IL-18) or three activity units (for IL-1β) of the indicated recombinant caspases for 1 h or 24 h prior to immunoblot analysis. HEK293T cells stably expressing GSDMD-V5 and IL-1β-Myc (F and G) or IL-18-V5 (H) were transiently co-transfected with the indicated catalytically inactive (C/A) caspase constructs and the IL-1β tetrapeptide mutants, IL-1β^LESD (F and H) or IL-1β^AAAD (E), for 48 h before immunoprecipitation and immunoblot analysis. * = non-specific bands. Data are representative of three or more independent experiments. High expo. = high exposure.

Figure 4.. Dimerization and LPS-activated CASP1/4/5 induce processing of inflammatory substrates

(A and B) HEK293T cells stably expressing ΔCARD DmrB-CASP1, -4, or -5 and GSDMD-V5 (A) or IL-18-V5 (B) were treated with AP20187 (1 μM) for 1 h or 24 h before LDH and immunoblot analysis. (C and D) HEK293T cells stably expressing ΔCARD DmrB-CASP1, -4, or -5 and GSDMD-V5 (C) or ΔCARD DmrB-CASP1, -4, or -5 and IL18-V5 (D) were transiently transfected with the indicated IL-18 (C) or IL-1β (D) constructs for 24 h before the addition of 1 μM AP20187 for 3 h (C) or 24 h (D). Lysates and supernatants were pooled and analyzed by immunoblotting. (E) HEK293T cells stably expressing IL-18-V5 and either CASP4 or CASP5 were transfected with LPS (25 μg/mL) for 24 h before samples were analyzed by immunoblotting. (F) HEK293T cells stably expressing the indicated IL-18 variants and either CASP4 or CASP5 were transfected with LPS (25 μg/mL) for 24 h and then analyzed by immunoblotting. Data are means ± SEM of three biological replicates. ***p < 0.001, **p < 0.01, and *p < 0.05 by two-sided Student’s t test. Data are representative of three or more independent experiments.

Figure 5.. Cytosolic LPS and pathogenic infections induce non-canonical inflammasome-mediated processing of inflammatory substrates in human macrophages and epithelial cells

(A) Caco-2 cells were primed with 100 ng/mL Pam3CSK4 for 3 h and then infected with Salmonella Typhimurium (multiplicity of infection [MOI] = 60) for 6 h. Cells and their supernatants were collected separately and analyzed by immunoblotting. (B and C) Parent (B) and CASP4 KO (C) HeLa cells were infected with Salmonella Typhimurium (MOI = 60) for 6 h. Cell lysates and supernatants were combined and analyzed by immunoblotting. (D and E) THP1 cells were differentiated into macrophages with phorbol 12-myristate 12-acetate (PMA) (50 ng/mL) for 24 h and primed with LPS (5 μg/mL) for another 24 h. Where indicated, THP1 macrophages were treated with ZVAD (40 μM) or VX765 (40 μM) 30 min before LPS transfections. Cells were transfected with LPS (25 μg/mL) for 24 h, and then samples were analyzed for LDH release (D) and immunoblotting (E). (F–H) THP1 cells were differentiated into macrophages with PMA (50 ng/mL) for 48 h and then transfected with LPS (25 μg/mL) for 24 h. Samples were then analyzed for LDH release (F), IL-18 secretion (G), and immunoblotting (H). Data are means ± SEM of three biological replicates. ***p < 0.001, **p < 0.01, and *p < 0.05 by two-sided Student’s t test. * = non-specific bands. Data are representative of three or more independent experiments.

Figure 6.. CASP4/5 preferentially cleave IL-1β at D27 to deactivate IL-1R signaling and IL-18 at D36 to activate cytokine signaling

(A and B) HEK293T cells were transiently co-transfected with the indicated constructs for 24 h and then analyzed by immunoblotting. (C) Schematic of inflammatory substrates and representative cleavage sites by inflammatory caspases. (D) HEK293T cells stably expressing IL-18-V5 and ΔCARD DmrB-CASP1 were transiently transfected with IL-1β^WT or IL-1β^Δ27 for 24 h before the addition of AP20187 (1 μM) for 1 h. Three independent experiments were then analyzed by immunoblotting, and the amount of cleaved IL-1β p17 was quantified by taking the ratio of the intensity of the cleaved IL-1β p17 to uncleaved IL-1β^WT or IL-1β^Δ27 respectively. (E and F) THP1 cells were differentiated into macrophages with PMA (50 ng/mL) for 48 h. Cells were then transfected with LPS (25 μg/mL) for 24 h, and then supernatants were analyzed by immunoblotting (E), or the amount of active cytokines was quantified by measuring IL-1R stimulation (F). (G) Parent THP1 cells were primed with LPS (5 μg/mL, 4 h, parent cells only) and then treated with nigericin (Nig., 5 μM) for 1 h before supernatants were analyzed for LDH release and IL-1 receptor activation by IL-18 and IL-1β. THP1 monocytes stably expressing dox-inducible IL-1β^WT, IL-1β^Δ27, or IL-1β^D27A were unprimed and were treated with doxycycline (Dox., 10 μg/mL, 24 h) and then treated with nigericin (Nig., 5 μM) for 1 h before analysis. Lysates and supernatants were then combined and analyzed by immunoblotting. Data are means ± SEM of three biological replicates. ***p < 0.001, **p < 0.01, and *p < 0.05 by two-sided Student’s t test. Data are representative of three or more independent experiments. * = non-specific bands.

Figure 7.. Caspase-11 does not cleave IL-18 but cleaves IL-1β to generate the p27 product

(A and B) HEK293T cells stably expressing ΔCARD Dmrb-CASP11 (A) or 2xFLAG-CASP11 (B) were transiently transfected with the indicated constructs. After 24 h, samples were treated with AP20187 (1 μM) (A) or LPS (25 μg/mL) (B) for 24 h and then analyzed for LDH release and immunoblotting. (C and E) Raw 264.7 macrophages were primed with LPS (5 μg/mL) for 4 h and then transfected with LPS (25 μg/mL) for 6 h before LDH release and immunoblot analysis (C) and IL-1 receptor activation by IL-1β (E). (D) Raw 264.7 macrophages were treated as in (C) and then analyzed for IL-1β release by ELISA. (F) HEK293T cell lysates expressing either human IL-18 or mouse IL-18 were incubated with one activity unit of the indicated recombinant caspases for 1 h or 24 h and then analyzed by immunoblotting. Data are means ± SEM of three biological replicates. ***p < 0.001, **p < 0.01, and *p < 0.05 by two-sided Student’s t test. Data are representative of three or more independent experiments. * = non-specific bands. The small m or h represents mouse or human proteins, respectively.