Itaconate confers tolerance to late NLRP3 inflammasome activation

Abstract

Itaconate is a unique regulatory metabolite that is induced upon Toll-like receptor (TLR) stimulation in myeloid cells. Here, we demonstrate major inflammatory tolerance and cell death phenotypes associated with itaconate production in activated macrophages. We show that endogenous itaconate is a key regulator of the signal 2 of NLR family pyrin domain containing 3 (NLRP3) inflammasome activation after long lipopolysaccharide (LPS) priming, which establishes tolerance to late NLRP3 inflammasome activation. We show that itaconate acts synergistically with inducible nitric oxide synthase (iNOS) and that the ability of various TLR ligands to establish NLRP3 inflammasome tolerance depends on the pattern of co-expression of IRG1 and iNOS. Mechanistically, itaconate accumulation upon prolonged inflammatory stimulation prevents full caspase-1 activation and processing of gasdermin D, which we demonstrate to be post-translationally modified by endogenous itaconate. Altogether, our data demonstrate that metabolic rewiring in inflammatory macrophages establishes tolerance to NLRP3 inflammasome activation that, if uncontrolled, can result in pyroptotic cell death and tissue damage.

Keywords: immunometabolism; inflammasome; innate immunity; itaconate; macrophages.

Figure 1.. Irg1 establishes LPS-mediated tolerance to signal 2 of NLRP3 inflammasome activation

(A) Design of BMDM pre-treatment with itaconate (ITA, 3 h) and classical NLRP3 inflammasome triggering: LPS priming for 3 h followed by ATP (45 min). (B and C) IL-1β protein detection in supernatants (B) or cell lysates (C) of BMDMs treated as in (A). No ATP was added in (C). (D) Overlay of the time course of IL-1β secretion in BMDMs primed with LPS for the indicated time followed by ATP, and intracellular ITA levels in BMDMs stimulated with LPS. ITA data from Swain et al. (2020). (E) Design of late inflammasome triggering. Cells are primed with LPS for 3 h (classical triggering) or 12 or 24 h (late triggering) followed by ATP. (F and G) IL-1β protein detection in supernatants (F) or cell lysates (G) of BMDMs treated as in (E). No ATP was added in (G). (H) Design of inflammasome tolerization. (I and J) IL-1β protein detection in supernatants (I) or cell lysates (J) of BMDMs treated as in (H). No ATP was added in (J). (K) Mice were injected with sodium itaconate (Na-ITA, 1.24 mg) i.p. 20 and 4 h before LPS. After LPS challenge (4 mg kg⁻¹, 6 h) IL-1β was analyzed in peritoneal fluid. Data are from n = 3 vehicle, Na-ITA mice and n = 5 LPS, Na-ITA + LPS mice. (L) Scheme of cutaneous IMQ administration and analysis of cell frequencies in mouse ear tissue determined by flow cytometry, with n = 4 mice per group. Western blots are representative of n = 3 experiments. GAPDH was used as loading control. T, tolerized; NT, non-tolerized; Unstim., unstimulated control. In (B), (F), and (I), n = 4 experiments. Data represent mean ± SEM. The p values were calculated using (B and K) 1-way ANOVA with Tukey’s test, (F and I) 2-way ANOVA with Sidak’s test, and (L) unpaired, two-tailed Student’s t test. See also Figure S1.

Figure 2.. Irg1 phenotype is associated with dysregulation of caspase-1 and GSDMD processing

(A) BMDMs were stimulated as in Figure 1H. Nigericin (Nig) for 1 h was used to trigger inflammasome. Fixed and permeabilized cells were analyzed for ASC scattering properties by flow cytometry. (B) Protein levels analyzed by western blot in supernatants of BMDMs stimulated as in Figure 1H, ATP was added for the indicated time. (C) IL-1β secretion in BMDMs treated as in Figure 1H. Caspase inhibitors (20, 50, and 100 μM) were added 30 min before ATP. Data are mean ± SEM, with n = 3 experiments. (D) GSDMD detection in cell lysates of BMDMs treated as in Figure 1H. VX-765 (caspase-1 inhibitor, 50 μM) was added 30 min before ATP. Western blots are representative of n = 3 experiments. GAPDH was used as loading control. See also Figure S2.

Figure 3.. ITA reconstitution rescues NLRP3 inflammasome tolerance in Irg1-deficient macrophages

(A) Design of inflammasome tolerization and ITA reconstitution (added at 4 h of LPS pre-stimulation). (B) IL-1β secretion in BMDMs stimulated and treated as in (A). ITA or other metabolites were added as indicated, with n = 3–4 experiments. The p values were calculated using 2-way ANOVA with Sidak’s test. (C) Pro-IL-1β detection in cell lysates of BMDMs treated and stimulated as in (A). No ATP was added. (D) PCA of global metabolic profiles of BMDMs stimulated with LPS for 24 h and reconstituted with ITA at 4 h of stimulation, with n = 4 cultures. (E) ITA quantification in samples from (D). (F and G) Protein detection in BMDM lysates treated and stimulated as in (A). No ATP was added in (F). Western blots are representative of n = 3 experiments. GAPDH was used as loading control. Data represent mean ± SEM. See also Figure S3 and Table S1.

Figure 4.. ITA synergizes with iNOS to tolerize NLRP3 inflammasome activation

(A) IL-1β secretion in BMDMs stimulated as in Figure 3A. SNAP (in micromolars) was added instead of ITA at 4h of LPS pre-treatment, with n = 3–6 experiments. (B) GSDMD detection in cell lysates of BMDMs treated as in Figure 1H. SEIT was added at 1 h of LPS pre-stimulation. (C) IL-1β secretion in BMDMs stimulated as in (B), with n = 3–4 experiments. (D) Irg1 mRNA levels in BMDMs stimulated with TLR ligands for 24 h, with n = 3 experiments. (E) iNOS protein detection in lysates of BMDMs stimulated as in (D). (F) Model of signaling pathways controlling expression of inflammasome tolerance. (G) IL-1β secretion in BMDMs stimulated as in Figure 1H. Pre-stimulation with various TLR ligands as indicated, with n = 3–4 experiments. (H) GSDMD detection in cell lysates of BMDMs stimulated as in (G). (I) IL-1β secretion in BMDMs treated and stimulated as in Figure 3A, with n = 3 experiments. (J) Protein detection in lysates of BMDMs stimulated with LPS for 24 h followed by ATP. ITA was added at 4 h of LPS stimulation. (K) IL-1β secretion in BMDMs treated as in (J), with n = 3 experiments. Western blots are representative of n = 3 experiments. GAPDH was used as loading control. Data represent mean ± SEM. The p values were calculated using 2-way ANOVA with Sidak’s test. See also Figure S4.

Figure 5.. ITA regulates pyroptosis after sustained LPS stimulation

(A) Design of classical (3 h of LPS priming) or late (24 h of LPS priming) inflammasome triggering by Nig. In some experiments, ITA was added during the 24 h priming period (ITA, 1 mM, added at 4 h of LPS priming). (B) Lactate dehydrogenase (LDH) activity in supernatants of BMDMs treated as in (A). (C and D) Flow cytometry analysis of pyroptosis in BMDMs treated as in (A). Cell frequencies are shown as percentages. (E) Western blot detection ofcaspase-1 (Casp1)and GSDMD forms in combined cell lysates and supernatants of BMDMs stimulated as in (A). Representative of n = 3 experiments. FLICA, caspase-1-specific fluorochrome inhibitor of caspases. Data are mean ± SEM. In (B) and (E), data are from n = 3–4 experiments. The p values were calculated using 2-way ANOVA with Sidak’s test. See also Figure S5.

Figure 6.. GSDMD is a gatekeeper of late pyroptosis and is itaconated in activated macrophages

(A) IL-1β secretion in BMDMs treated as in Figure 1H. EtGSH (5 mM) was added at 4 h of LPS pre-stimulation, with n = 3 experiments. (B and C) Changes in global protein (B) and itaconated-peptide (Ita-peptide; C) content in BMDMs treated as in Figure 1H (with no ATP addition) detected by mass spectrometry, with n = 6, 3 cultures from 2 experiments. The p values were determined with two-tailed, unpaired Student’s t test. (D) Proteins with detected cysteine modification by ITA (itaconation) from (C). (E) Mass spectra and schematics of the structure of the itaconated GSDMD peptide on Cys77. (F–I) Quantification of total GSDMD or itaconated GSDMD on Cys77 (Ita-GSDMD) detected by mass spectrometry (no ATP added). In (H), cells were treated as in Figure 3A (no ATP). In (I), cell were stimulated with LPS for 24 h. In (F) and (G), data are from n = 3 cultures; in (H), n = 6, 3 cultures from 2 experiments; and in (I), n = 2–3 cultures. (J and K) Flow cytometry analysis of pyroptosis in BMDMs treated as in Figure 5A, with n = 5–6 experiments. FC, fold change; pval, p value. Data represent mean ± SEM. pval determined using (A, F, G, and K) 2-way ANOVA with Sidak’s test and (H) 1-way ANOVA with Tukey’s test. See also Figure S6 and Table S2.