Main

Inflammasomes are critical components of the innate immune response1. Following exposure to PAMPs and DAMPs, inflammasome sensors form a multi-protein complex that activates caspase-1, leading to cleavage of its downstream substrates, inflammatory signalling and inflammatory cell death. Individual inflammasome sensors detect and respond to specific PAMPs and DAMPs, but emerging evidence suggests that multiple inflammasome sensors can be activated in response to disease, particularly live pathogen infection, owing to the presence of multiple PAMPs and DAMPs2,3,4,5.

The AIM2 inflammasome is known to sense double-stranded DNA (dsDNA) and has essential roles in development, infectious diseases, inflammatory diseases and cancer6,7,8,9,10,11. However, several critical functions of AIM2 beyond its canonically described role in inflammasome formation and pyroptosis have been observed that cannot be explained by our current understanding of the AIM2 inflammasome2. Herpes simplex virus 1 (HSV1), a dsDNA virus that causes lifelong incurable, recurrent pathologies, and Francisella, a Gram-negative bacterium that can cause rapid lethality upon infection, are two diverse pathogens that are known to activate the AIM2 inflammasome and cell death6,10,11,12,13. To investigate whether cells exposed to numerous PAMPs and DAMPs during these viral and bacterial infections can simultaneously engage multiple inflammasome sensors, we infected wild-type bone marrow-derived macrophages (BMDMs) and BMDMs singly deficient in several major inflammasome sensors with HSV1 or F. novicida. Infection with HSV1 or F. novicida induced AIM2-dependent and NLRP3- and NLRC4-independent cleavage of caspase-1, release of the inflammasome-dependent cytokines interleukin (IL)-1β and IL-18, and cell death (Fig. 1a–o, Extended Data Fig. 1a–i). However, inflammasome activation and cell death were also partially reduced in Mefv–/– BMDMs during HSV1 and F. novicida infections (Fig. 1p–t, Extended Data Fig. 1j–l), suggesting that these infections activate pyrin to drive AIM2-mediated inflammatory signalling and cell death.

Fig. 1: HSV1 induces AIM2-, pyrin- and ZBP1-mediated caspase-1 activation, cytokine release and cell death.
figure 1

a, Immunoblot analysis of pro-caspase-1 (P45) and cleaved caspase-1 (CASP1) (P20) in HSV1-infected or poly(dA:dT)-transfected wild-type (WT) or Aim2−/− BMDMs. b, c, IL-1β (b) and IL-18 (c) release following HSV1 infection. d, Cell death in BMDMs after HSV1 infection for 16 h, measured by SYTOX Green uptake assay. Red indicates dead cells. e, Quantification of the cell death in d. f–j, Immunoblot analysis of CASP1 (f), release of IL-1β (g) and IL-18 (h), cell death images at 16 h after infection (i) and quantification of cell death (j) from WT or Nlrp3−/− BMDMs after HSV1 infection or lipopolysaccharide (LPS) plus nigericin (Ni) treatment. k–o, Immunoblot analysis of CASP1 (k), release of IL-1β (l) and IL-18 (m), cell death images at 16 h after infection (n) and quantification of cell death (o) from WT or Nlrc4−/− BMDMs after HSV1 or Salmonella Typhimurium infection. p–t, Immunoblot analysis of CASP1 (p), release of IL-1β (q) and IL-18 (r), cell death images at 16 h after infection (s) and quantification of cell death (t) from WT or Mefv−/− BMDMs after HSV1 infection or C. difficile toxin AB+ supernatant treatment. u–y, Immunoblot analysis of CASP1 (u), release of IL-1β (v) and IL-18 (w), cell death images at 16 h after infection (x), and quantification of cell death (y) from WT or Zbp1−/− BMDMs after HSV1 or influenza A virus (IAV) infection. a, f, k, p, u, Data are representative of at least three independent experiments. b, c, g, h, l, m, q, r, v, w, Data are mean ± s.e.m. NS, not significant; ****P < 0.0001 (two-tailed t-test; n = 6 from 3 biologically independent samples). d, i, n, s, x, Images are representative of at least three independent experiments. Scale bars, 50 μm. e, j, o, t, y, Data are mean ± s.e.m. NS, not significant; ****P < 0.0001 (two-tailed t-test; n = 8 from 4 biologically independent samples). Exact P values are presented in Supplementary Table 1. For gel source data, see Supplementary Figure 1.

Source data

Full size image

Because we observed residual AIM2-mediated caspase cleavage in Mefv–/– BMDMs after HSV1 and F. novicida infections (Fig. 1p, Extended Data Fig. 1j), we screened several innate immune sensors to determine their involvement. Loss of the sensor ZBP1, which regulates inflammatory cell death in infectious and sterile conditions14,15,16, resulted in reduced caspase-1 cleavage, release of IL-1β and IL-18 and cell death compared with wild-type BMDMs after HSV1 and F. novicida infections (Fig. 1u–y, Extended Data Fig. 1m–o), whereas loss of other sensors or adaptors had no effect (Extended Data Fig. 2), suggesting that ZBP1 is also involved in HSV1- and F. novicida-induced inflammatory signalling and cell death.

To determine whether pyrin and ZBP1 act cooperatively or redundantly to decrease caspase-1 activation and cell death, we next used colchicine to inhibit pyrin activation17,18 in Zbp1–/– BMDMs. We found that colchicine treatment in Zbp1–/– BMDMs further reduced cell death, cleavage of caspase-1 and release of IL-1β and IL-18 during HSV1 and F. novicida infections, mirroring the levels observed in Aim2–/– BMDMs (Fig. 2a, b, Extended Data Fig. 3), suggesting that ZBP1 cooperates with pyrin to drive the AIM2-dependent responses. We also generated Mefv–/–Zbp1–/– mice and found that cell death, cleavage of caspase-1 and release of IL-1β and IL-18 during HSV1 and F. novicida infections were further reduced in Mefv–/–Zbp1–/– BMDMs compared with Mefv–/– or Zbp1–/– BMDMs, mirroring the levels observed in Aim2–/– BMDMs (Fig. 2c–j, Extended Data Fig. 4a, b). These results suggest that pyrin and ZBP1 cooperate to induce AIM2-mediated inflammatory signalling and cell death during both viral infection with HSV1 and bacterial infection with F. novicida. Similarly, we also observed activation of cell death and caspase-1 in human THP-1 macrophages during HSV1 infection, and this activation was inhibited by short interfering RNA (siRNA) knockdown of AIM2 or combined knockdown of ZBP1 and MEFV (Extended Data Fig. 4c–e), suggesting that the AIM2- and pyrin and ZBP1-mediated cell death also occurs in human cells, the natural host of HSV1. Furthermore, the extent of cell death and caspase-1 cleavage was similar between wild-type and Mefv–/–, Zbp1–/– or Mefv–/–Zbp1–/– BMDMs treated with the synthetic AIM2 inflammasome ligand poly(dA:dT) (Extended Data Fig. 4f–h), suggesting that infection is a unique trigger causing pyrin and ZBP1 to cooperate to facilitate AIM2-mediated caspase-1 activation, cytokine release and cell death.

Fig. 2: ZBP1 cooperates with pyrin to drive AIM2-mediated caspase-1 activation, cytokine release and cell death.
figure 2

a, b, Immunoblot analysis of caspase-1 in BMDMs infected with HSV1 or F. novicida with or without colchicine (Col). Data are representative of at least three independent experiments. c, Cell death in wild-type, Aim2−/− or Mefv−/− Zbp1−/− BMDMs at 16 h after infection with HSV1 or F. novicida. Red indicates dead cells. Data are representative of at least three independent experiments. Scale bar, 50 μm. d, Quantification of the cell death from c. Data are mean ± s.e.m. (one-way ANOVA with Dunnett’s multiple comparisons test; n = 9 from 3 biologically independent samples). e, f, Immunoblot analysis of CASP1 in BMDMs infected with HSV1 (e) or F. novicida (f). Data are representative of at least three independent experiments. g–j, Release of IL-1β (g, i) or IL-18 (h, j) following HSV1 (g, h) or F. novicida (i, j) infection. Data are mean ± s.e.m. (one-way ANOVA with Dunnett’s multiple comparisons test; n = 6 from 3 biologically independent samples). Exact P values are presented in Supplementary Table 1. For gel source data, see Supplementary Figure 1.

Source data

Full size image

To understand how pyrin is activated during HSV1 and F. novicida infections, we first evaluated RhoA-GTP activity, which is known to be inhibited to activate pyrin in response to bacterial toxins such as the Clostridium difficile toxin TcdB17,18,19,20,21. We found that TcdB-mediated inhibition of RhoA-GTP activity occurred in an AIM2- and ZBP1-independent manner (Extended Data Fig. 5a), whereas HSV1-mediated inhibition was dependent on AIM2 but independent of ZBP1 (Extended Data Fig. 5a). We next examined RhoA-GTP levels in Aim2–/–, Asc–/– and Casp1–/– BMDMs during HSV1 and F. novicida infections. Whereas RhoA-GTP was absent in wild-type BMDMs upon infection, it was present in Aim2–/–, Asc–/– and Casp1–/– BMDMs (Extended Data Fig. 5b, c). By contrast, treatment with the synthetic AIM2 inflammasome ligand poly(dA:dT) did not inhibit RhoA-GTP activity in any of the genotypes tested (Extended Data Fig. 5d, e), suggesting that infection has a distinct ability to integrate the AIM2-mediated-signalling and RhoA-signalling cascades to influence pyrin activation.

We then investigated how ZBP1 was activated during HSV1 and F. novicida infections. In the context of influenza A virus infection, ZBP1 Zα domains sense nucleic acids to drive RIPK3-, caspase-8- and NLRP3-mediated inflammatory cell death14,15,22. Similarly, we found that HSV1- and F. novicida-induced cell death was reduced in Zbp1∆Za2/∆Za2 BMDMs compared with wild-type BMDMs and was similar to that of Zbp1–/– BMDMs (Extended Data Fig. 5f–i).

Our data show that HSV1 and F. novicida can activate multiple inflammasome sensors. Many pathogens can also induce multiple programmed cell death pathways, including pyroptosis, apoptosis and necroptosis14,23. Recent studies have found extensive crosstalk between programmed cell death pathways14,24,25,26,27, establishing the concept of PANoptosis14,28,29,30,31. Although inflammasome activation is primarily associated with gasdermin D (GSDMD)-mediated pyroptosis, there is emerging evidence of a contribution of inflammasome components in driving PANoptosis14, 15, 26, 27, 29, 30, 32. Therefore, we biochemically characterized the cell death induced by HSV1 and F. novicida infections. In wild-type cells, HSV1 and F. novicida activated key molecules involved in pyroptotic, apoptotic and necroptotic pathways (Fig. 3). However, we observed reduced activation of the pyroptotic molecules caspase-1, GSDMD and GSDME (Fig. 3a), the apoptotic molecules caspase-8, -3 and -7 (Fig. 3b) and the necroptotic molecules RIPK3 and MLKL (Fig. 3c) in Mefv–/– and Zbp1–/– BMDMs compared with wild-type BMDMs following HSV1 or F. novicida infection, and a complete loss of activation in Aim2–/– and Mefv–/–Zbp1–/– BMDMs. Collectively, these data suggest that HSV1 and F. novicida infections induce inflammatory cell death, PANoptosis, in a manner dependent on AIM2 and the coordinated activation of pyrin and ZBP1.

Fig. 3: AIM2, pyrin and ZBP1 promote inflammatory cell death in response to HSV1 and F. novicida infections.
figure 3

a–c, Immunoblot analysis of pro- (P45) and activated (P20) caspase-1 (CASP1), pro- (P53) and activated (P30) GSDMD, pro- (P53) and activated (P34) GSDME (a); pro- (P55) and cleaved (P18) caspase-8 (CASP8), pro- (P35) and cleaved (P17/P19) caspase-3 (CASP3), pro- (P35) and cleaved (P20) caspase-7 (CASP7) (b); phosphorylated MLKL (pMLKL), total MLKL (tMLKL), phosphorylated RIPK3 (pRIPK3) and total RIPK3 (tRIPK3) (c) in wild-type, Aim2−/−, Mefv−/−, Zbp1−/− or Mefv−/−Zbp1−/− BMDMs after HSV1 or F. novicida infection. Data are representative of at least three independent experiments. For gel source data, see Supplementary Figure 1.

Full size image

We next sought to understand the regulatory relationship between AIM2, pyrin and ZBP1 during infection with HSV1 or F. novicida. Loss of AIM2 completely abrogated the inflammatory cell death, whereas loss of pyrin or ZBP1 resulted in a partial reduction, suggesting that AIM2 functions upstream of pyrin and ZBP1. Protein expression of pyrin and ZBP1 was reduced in Aim2–/– BMDMs after HSV1 or F. novicida infection (Fig. 4a, b). Additionally, their expression was also reduced in Asc–/–, Casp1–/– and enzymatically inactive Casp1C284A/C284A BMDMs as well as in wild-type BMDMs treated with the caspase-1 inhibitor VX-765 (Fig. 4c, d), but not in Gsdmd–/–, Casp8DA/DA, Casp7–/–, Casp3–/– or Casp6–/– BMDMs after HSV1 and F. novicida infections (Extended Data Fig. 6a–c). Furthermore, expression of pyrin and ZBP1 was not affected during influenza A virus infection in Aim2–/–, Asc–/–, Casp1–/– and Nlrp3–/– BMDMs (Extended Data Fig. 6d). Therefore, AIM2-mediated signalling functions as an upstream regulator for the expression of pyrin and ZBP1 in response to HSV1 and F. novicida infections.

Fig. 4: AIM2-mediated signalling acts as an upstream regulator of pyrin and ZBP1, which are required to form the AIM2 PANoptosome.
figure 4

a, b, Immunoblot analysis of ZBP1, pyrin and AIM2 in wild-type, Aim2−/−, Mefv−/− and Zbp1−/− BMDMs infected with HSV1 (a) or F. novicida (b). c, Immunoblot analysis of caspase-1 (CASP1) activation and pyrin and ZBP1 expression in wild-type or Asc−/− BMDMs after HSV1 or F. novicida infection. d, Immunoblot analysis of CASP1 activation and pyrin and ZBP1 expression in wild-type, Casp1−/− and Casp1C284A/C284A (Casp1C284A) BMDMs and wild-type BMDMs treated with the CASP1 inhibitor VX-765 (20 μM; VX) after HSV1 or F. novicida infection. e, Immunoprecipitation (IP) in wild-type BMDMs with IgG control antibodies or anti-ASC antibodies after HSV1 or F. novicida infection or poly(dA:dT) transfection. a–e, Data are representative of at least three independent experiments. f, Immunofluorescence images of wild-type BMDMs at 12 h after HSV1 infection. Scale bars, 5 μm. Arrowheads indicate the ASC speck. Images are representative of three independent experiments. g, Quantification of the percentage of cells with ASC+AIM2+pyrin+ZBP1+specks among the ASC speck+ cells. Data are mean ± s.e.m. (one-way ANOVA with Dunnett’s multiple comparisons test; n = 6 from 3 biologically independent samples). h, Pulmonary viral titre at day 3 after infection with HSV1 (WT, n = 6; Aim2−/−, n = 5). Each symbol represents one mouse. Data are pooled from two independent experiments. Data are mean ± s.e.m. ***P < 0.001 (two-tailed t-test). i, Survival of wild-type, Aim2−/−, Mefv−/− and Zbp1−/− mice infected intranasally with 5 × 107 plaque-forming units (PFU) of HSV1. Survival data are pooled from three independent experiments. P values for survival of each genotype versus wild type are shown in the key. **P < 0.01; ***P < 0.001; ****P < 0.0001 (log-rank (Mantel–Cox) test). Exact P values are presented in Supplementary Table 1. For gel source data, see Supplementary Figure 1.

Source data

Full size image

To investigate the molecular mechanism underlying how AIM2, ASC and caspase-1 activity-dependent signals promote the expression of pyrin and ZBP1 in infected cells, we assessed transcription and observed that HSV1- or F. novicida-infected Aim2–/–, Asc–/– and Casp1–/– BMDMs had reduced transcription of Zbp1 and Mefv compared with the levels in wild-type cells (Extended Data Fig. 7a–d). It was previously shown that expression of pyrin and ZBP1 can be induced by interferon (IFN) signalling14,33. Therefore, we investigated whether AIM2 could induce IFN production to drive the expression of pyrin and ZBP1. We found that wild-type BMDMs produced IFN-β in response to HSV1 (multiplicity of infection (MOI) 10, 6 h after infection) or F. novicida infection (MOI 1, 3 h after infection), whereas Aim2–/–, Asc–/– and Casp1–/– BMDMs did not (Extended Data Fig. 7e, f). Additionally, the expression of pyrin and ZBP1 was restored upon complementation of IFN-β in the AIM2-, ASC- and caspase-1-deficient cells (Extended Data Fig. 7g, h). The differences in IFN production observed here are in contrast to previous studies with F. novicida, in which these knockouts have not been reported to have a marked loss of type I IFN production. These differences might be due to differing timepoints and MOIs: previous studies showed that Asc–/– BMDMs have slightly reduced IFN-β release 5 h after infection34 (MOI 100) and similar Ifnb mRNA levels 9 h after infection35 (MOI 100); Aim2–/– BMDMs have significantly increased IFN-β release 6 h after infection11 (MOI 250); and Asc–/– and Casp1/11–/– BMDMs have significantly increased IFN-β release 6 h after infection36 (MOI 50). These differences suggest that type I IFN release in response to F. novicida may be timepoint- and/or MOI-dependent.

Next, we sought to understand the molecular relationship between AIM2, pyrin and ZBP1 in inducing inflammatory cell death and PANoptosis during HSV1 and F. novicida infections. We hypothesized that these molecules would all be part of the molecular scaffold that enables contemporaneous engagement of key proteins from pyroptosis, apoptosis and necroptosis28,29,31,32. We observed interactions of ASC with AIM2, pyrin, ZBP1, caspase-1, caspase-8, RIPK3, RIPK1 and FADD by immunoprecipitation following HSV1 and F. novicida infections (Fig. 4e), and these interactions were not observed in Aim2–/–, Mefv–/–Zbp1–/–, Ripk3–/–Casp8–/– or Ripk3–/–Fadd–/– BMDMs (Extended Data Fig. 8). By contrast, treatment with poly(dA:dT) allowed interactions between endogenous ASC, AIM2 and caspase-1, but not pyrin, ZBP1, caspase-8, RIPK3, RIPK1 or FADD (Fig. 4e), suggesting that infection has a distinct ability to form this AIM2 multi-protein cell death-inducing complex. We have termed this complex the AIM2 PANoptosome. We also observed that ASC specks colocalized with AIM2, pyrin and ZBP1 collectively in the same cell at 12 h after infection with HSV1 or F. novicida (Fig. 4f, g, Extended Data Fig. 9a, b). Similarly, ASC specks also colocalized with caspase-8 and RIPK3 in the same cell (Extended Data Fig. 9c–f). Furthermore, the formation of this complex was required for cell death, as its disruption through the deletion of key components inhibited cell death in response to HSV1 and F. novicida infections (Fig. 2c, d, Extended Data Fig. 9g, h).

We then extended our findings to in vivo models of HSV1 and F. novicida infection. Lung lysates from infected wild-type mice showed activation of key pyroptotic, apoptotic and necroptotic molecules and increased expression of pyrin and ZBP1, but this activation and increased expression was not found in Aim2–/– mice (Extended Data Fig. 10a, b). The number of plaque- and colony-forming units for HSV1 and F. novicida, respectively, was increased in Aim2–/–, Mefv–/–, Zbp1–/– and Mefv–/–Zbp1–/– BMDMs and in the tissue of Aim2–/– mice (Fig. 4h, Extended Data Fig. 10c–e). Additionally, all mice lacking AIM2 succumbed to HSV1 infection within 8 days and mice lacking pyrin or ZBP1 succumbed within 13 days, whereas around 70% of the wild-type mice survived (Fig. 4i). Similarly, all mice lacking AIM2 succumbed to F. novicida infection within 4 days and mice lacking pyrin or ZBP1 succumbed within 9 or 11 days, respectively, whereas around 80% of the wild-type mice survived (Extended Data Fig. 10f). These results highlight an important role for the AIM2-regulated expression of pyrin and ZBP1, inflammatory cell death and PANoptosis in host defence against HSV1 and F. novicida infections.

In summary, AIM2, pyrin and ZBP1 form a multi-protein complex termed the AIM2 PANoptosome via interactions with ASC to induce PANoptosis during HSV1 and F. novicida infections. Loss of AIM2 reduces the expression of pyrin and ZBP1 during these infections, indicating that AIM2-mediated signalling functions as an upstream regulator of pyrin and ZBP1 to control assembly and activation of the AIM2 PANoptosome. The ability of HSV1 and F. novicida to activate AIM2 while also engaging the ZBP1 Zα domain and inhibiting Rho-GTP activity gives these pathogens the unique ability to form this AIM2-, pyrin- and ZBP1-containing PANoptosome complex. Other pathogens that can engage multiple sensors may also form similar complexes with shared central components.

Furthermore, we observed roles for ZBP1 in innate immunity. ZBP1 is known to sense influenza viral Z-RNA to promote binding of RIPK3, caspase-6 and the NLRP3 inflammasome to form the ZBP1 PANoptosome complex and drive PANoptosis14,15,22,31. The multi-protein complex identified here is mediated by AIM2 but may be similar to this ZBP1 PANoptosome, as it shares the same cell death effectors, including caspase-8 and RIPK3, and also leads to the induction of PANoptosis, characterized by activation of pyroptotic, apoptotic and necroptotic molecules. Additionally, we discovered that ZBP1 also interacts and localizes with ASC and acts as an inflammasome sensor to induce caspase-1 cleavage and inflammatory cell death in response to HSV1 and F. novicida infections. Furthermore, AIM2 regulated the expression of pyrin and ZBP1 during HSV1 and F. novicida infections, but not during influenza A virus infection.

Overall, our findings identify a critical interaction between AIM2, pyrin and ZBP1 that drives innate immune responses during pathogen infection. This regulatory mechanism defines the molecular basis of how infection with live pathogens that liberate numerous PAMPs or DAMPs can trigger the coordinated activation of innate immune and cell death signalling components to form a multi-protein complex and cause inflammatory cell death and cytokine release to shape the immune response and host defence.

Methods

Mice

C57BL/6 J (wild type), Mefv−/− (ref. 21), Aim2−/− (ref. 34), Nlrp3−/− (ref. 37), Nlrc4−/− (ref. 38), Zbp1−/− (ref. 39), Tlr3−/− (ref. 40), Trif−/− (ref. 41), Mda5−/− (ref. 42), Mavs−/− (ref. 43), Nlrp6−/− (ref. 44), Nlrp12−/− (ref. 45), Asc−/− (ref. 46), Casp1−/− (ref. 47), Casp1C284A/C284A (ref. 48), Gsdmd−/− (ref. 49), Casp8DA/DA (ref. 50), Casp7–/– (ref. 51), Casp3–/– (ref. 52), Casp6−/− (Jackson Laboratory, 006236; ref. 31), Ripk3–/– (ref. 53), Ripk3–/–Casp8–/– (ref. 54) and Ripk3–/–Fadd–/– (ref. 55) mice have been described previously. Mefv−/− mice were crossed with Zbp1−/− mice to generate Mefv−/− Zbp1−/− homozygous knockouts. All mice were bred and maintained in a specific pathogen-free facility at the Animal Resource Center at St Jude Children’s Research Hospital and were backcrossed to the C57BL/6 background (J substrain) for at least 10 generations. Both male and female mice were used in this study; age- and sex-matched 6- to 8-week-old mice were used for in vivo and 6- to 12-week-old mice were used for in vitro studies. Cohoused mice were used for in vivo analyses. Mice were maintained in 20–23.3 °C and 30–70% humidity with a 12 h light/dark cycle and were fed standard chow. Animal studies were conducted under protocols approved by the St Jude Children’s Research Hospital committee on the Use and Care of Animals.

Cell culture

Primary BMDMs were cultivated for 6 days in DMEM (Thermo Fisher Scientific, 12440-053) supplemented with 10% FBS (Biowest, S1620), 30% L929-conditioned medium, 1% non-essential amino acids (Thermo Fisher Scientific, 11140-050) and 1% penicillin and streptomycin (Thermo Fisher Scientific, 15070-063). BMDMs were then seeded into 12-well plates at a density of 1 million cells per well and incubated overnight before use. THP-1 cells (ATCC TIB-202) were grown in RPMI 1640 with 10% FBS and differentiated into macrophages in RPMI 1640 medium containing 20% FBS and 100 ng ml−1 phorbol 12-myristate 13-acetate (PMA) for 2 days. The THP-1 cell line was purchased directly from ATCC and was not further authenticated in our laboratory. Cells were tested for mycoplasma contamination using mycoplasma detection PCR and were found to be negative for mycoplasma contamination.

Virus and bacteria culture

Human herpes simplex virus 1 (HF strain) (ATCC; VR-260) was propagated in Vero cells, and the virus titre was measured by plaque assay in Vero cells. The influenza A virus (A/Puerto Rico/8/34, H1N1 (PR8)) was prepared as previously described31 and propagated from 11-day-old embryonated chicken eggs by allantoic inoculation. Influenza A virus titre was measured by plaque assay in MDCK cells. Salmonella Typhimurium strain SL1344 was inoculated into Luria-Bertani (LB) broth (MP Biomedicals, 3002-031) and incubated overnight under aerobic conditions at 37 °C. S. Typhimurium was then sub-cultured (1:10) for 3 h at 37 °C in fresh LB broth to generate bacteria grown to log phase. C. difficile strain r20291 AB- and AB+ strains (provided by N. Minton56) were streaked onto brain heart infusion agar (BD Biosciences, 211065) and incubated overnight at 37 °C in an anaerobic chamber. Single colonies were inoculated into tryptone-yeast extract medium and grown overnight at 37 °C anaerobically. F. novicida strain U112 was prepared as previously described6 and was grown in BBL Trypticase Soy Broth (TSB) (211768, BD Biosciences) supplemented with 0.2% l-cysteine (Fisher) overnight under aerobic conditions at 37 °C. Bacteria were subcultured (1:10) in fresh TSB supplemented with 0.2% l-cysteine for 3 h and resuspended in PBS. Gentamicin (50 μg ml−1, Gibco) was added 3 h after infection to kill extracellular F. novicida.

Cell stimulation and infection

For ligand-mediated AIM2 inflammasome activation, 2 μg of poly(dA:dT) (InvivoGen, tlrl-patn) was resuspended in PBS and mixed with 0.6 μl of Xfect polymer in Xfect reaction buffer (Clontech Laboratories, 631318). After 10 min, DNA complexes were added to BMDMs in Opti-MEM (ThermoFisher Scientific, 31985-070), followed by incubation for 5 h. For NLRP3 inflammasome activation, BMDMs were primed for 4 h with 500 ng ml−1 ultrapure lipopolysaccharide (LPS) from Escherichia coli (0111:B4) (Invivogen, tlrl-3pelps) and then stimulated for 45 min with 20 μM nigericin (Sigma, N7143). For NLRC4 inflammasome activation, S. Typhimurium was infected at an MOI of 1 for 2 h. For pyrin inflammasome activation, toxin AB+ C. difficile (2 × 107 colony-forming units (CFU) ml−1) were spun down, and the supernatant was sterilized using 0.22 μm filters and then added to BMDMs for 12 h. For HSV1 (6, 12 or 16 h; MOI 10) and IAV (16 h; MOI 10) infections, cells were infected in DMEM plain medium (Sigma, D6171). For F. novicida infection, cells were infected in DMEM (ThermoFisher Scientific, 11995-065) at a MOI of 50 (12 or 16 h) to determine caspase, GSDMD, and GSDME cleavage and for cell death analyses, at a MOI of 150 (24 h) to determine phosphorylated MLKL and RIPK3 levels or at a MOI of 1 (3 h) for IFN analyses. For inhibition of the pyrin inflammasome, 30 μM colchicine (Sigma, C9754) was added to BMDMs 1 h before infection. For inhibition of caspase-1, 20 μM VX-765 (Chemietek, CT-VX765) was added to BMDMs 1 h before the infection. For the IFN-β treatment, 50 ng ml−1 of IFN-β (PBL Assay, 12400-1) was added to BMDMs 3 h after infections. To measure in vitro viral titre, the indicated BMDMs were infected with HSV1 (MOI 10) for 12 h, and the supernatants were used for plaque assays in Vero cells. To measure in vitro F. novicida levels, cells were infected at an MOI of 100, and cells were collected with PBS containing 0.5% Triton X-100 at 6 or 12 h after infection. The F. novicida were then streaked onto TSB plates supplemented with 0.2% l-cysteine and grown overnight at 37 °C anaerobically.

siRNA-mediated gene silencing

The Accell human siRNA SMARTPools against AIM2 (E-011951-00-0010), MEFV (E-011081-00-0010) and ZBP1 (E-014650-00-0010) genes were purchased from Horizon. THP-1 macrophages were transfected with siRNA using Accell siRNA delivery media according to the manufacturer’s instructions (Horizon). As a negative control, non-targeting control siRNA (D-001910-01-50) was used.

Real-time PCR analysis

RNA was extracted using TRIzol according to the manufacturer’s instructions (Life Technologies). Isolated RNA was reverse transcribed to cDNA using the First-Strand cDNA Synthesis Kit (Life Technologies). Real-time PCR was performed on an ABI 7500 real-time PCR instrument with 2× SYBR Green (Applied Biosystems). Primer sequences used in this study were as follows: 5′-AAGAGTCCCCTGCGATTATTTG-3′ and 5′-TCTGGATGGCGTTTGAATTGG-3′ for Zbp1; 5′-TCATCTGCTAAACACCCTGGA-3′ and 5′-GGGATCTTAGAGTGGCCCTTC-3′ for Mefv; 5′-CGTCCCGTAGACAAAATGGT-3′ and 5′-TTGATGGCAACAATCTCCAC-3′ for Gapdh.

Measurement of active RhoA-GTP levels

BMDMs (2 × 106 cells) were seeded in 6-well plates and transfected with 0.5 μg ml−1 TcdB, 0.5 μg poly(dA:dT) or infected with HSV (MOI 10) or F. novicida (MOI 100). RhoA-GTP levels were measured using the RhoA G-LISA Activation Assay kit (BK121, Cytoskeleton) and normalized to total RhoA levels, which were measured using the Total RhoA ELISA kit (BK150, Cytoskeleton) according to the manufacturer’s instructions. For pull-down assays, a GST-tagged Rho binding domain (RBD) in the Rho effector protein was used, which has been shown to bind specifically to the GTP-bound form of RhoA17,19. RhoA-GTP in BMDMs (2 × 106 cells) was pull-down with Rhotekin–RBD beads using the RhoA activation Assay Biochem Kit (BK036, Cytoskeleton) according to the manufacturer’s instructions.

In vivo infection

Age- and sex-matched, 6- to 8-week-old wild-type wild-type and Aim2–/–, Mefv–/– and Zbp1–/– mice were used for infections. For HSV1 infection, mice were anaesthetized with 250 mg kg−1 avertin and then infected intranasally with HSV1 in 50 μl PBS containing around 5 × 107 PFU. Infected mice were monitored over a period of 14 days for survival. For the F. novicida infection, F. novicida strain U112 was grown in TSB supplemented with 0.2% l-cysteine overnight at 37 °C and then 1:10 subcultured for 4 h before infection. Mice were infected subcutaneously with 5 × 105 CFU of F. novicida in 200 μl PBS. Infected mice were monitored over a period of 12 days for survival. Lungs, livers and/or spleens collected at the indicated time points were homogenized in 1 ml PBS for viral titres to be enumerated by plaque assays or for bacterial titres to be streaked onto TSB supplemented with 0.2% l-cysteine plates and grown overnight at 37 °C anaerobically.

Immunoblot analysis

Immunoblotting was performed as described previously57. In brief, for caspase analysis, BMDMs were lysed along with the supernatant using 50 μl caspase lysis buffer (1× protease inhibitors, 1× phosphatase inhibitors, 10% NP-40 and 25 mM DTT) followed by the addition of 100 μl 4× SDS loading buffer. For signalling analysis, the BMDM supernatants were removed at the indicated time points, and cells were washed once with PBS, after which cells were lysed with RIPA buffer. Proteins from lung, spleen and liver tissues were extracted using RIPA buffer supplemented with protease and phosphatase inhibitors (Roche), and 30 μg per sample was loaded on the gel. Proteins were separated by electrophoresis through 8%–12% polyacrylamide gels. Following electrophoretic transfer of proteins onto PVDF membranes (Millipore, IPVH00010), non-specific binding was blocked by incubation with 5% skim milk; then membranes were incubated with the following primary antibodies: anti-caspase-1 (AdipoGen, AG-20B-0042, 1:1,000), anti-caspase-3 (CST, 9662, 1:1,000), anti-cleaved caspase-3 (CST, 9661, 1:1,000), anti-caspase-7 (CST, 9492, 1:1,000), anti-cleaved caspase-7 (CST, 9491, 1:1,000), anti-caspase-8 (CST, 4927, 1:1,000), anti-cleaved caspase-8 (CST, 8592, 1:1,000), anti-pRIPK3 (CST, 91702 S, 1:1,000), anti-RIPK3 (ProSci, 2283, 1:1,000), anti-pMLKL (CST, 37333, 1:1,000), anti-MLKL (Abgent, AP14272b, 1:1,000), anti-GSDMD (Abcam, ab209845, 1:1,000), anti-GSDME (Abcam, ab215191, 1:1,000), anti-pyrin (Abcam, ab195975, 1:1,000), anti-AIM2 (Abcam, ab119791, 1:1,000), anti-ZBP1 (AdipoGen, AG-20B-0010, 1:1,000), anti-ASC (Millipore, 04-147, 1:1,000 or AdipoGen, AG-25B-006-C100, 1:1,000), anti-RIPK1 (CST, 3493, 1:1,000), anti-FADD (Millipore, 05-486, 1:1,000), anti-β-actin (Proteintech, 66009-1-IG, 1:5,000), human anti-caspase-1 (R&D systems, MAB6215, 1:1,000), human anti-β-actin (CST, 4970, 1:1,000). Membranes were then washed and incubated with the appropriate horseradish peroxidase (HRP)-conjugated secondary antibodies (1:5,000 dilution; Jackson Immuno Research Laboratories, anti-rabbit (111-035-047), anti-mouse (315-035-047)) for 1 h. Proteins were visualized by using Luminata Forte Western HRP Substrate (Millipore, WBLUF0500), and membranes were developed with an Amersham imager; images were analysed with ImageJ (v1.53a).

Real-time cell death analysis

Real-time cell death assays were performed using an IncuCyte S3 imaging system (Essen Biosciences). BMDMs or THP-1 cells were seeded in 12-well plates (106 BMDMs or 5 × 105 THP-1 cells per well) and stimulated. After infection, 100 nM SYTOX Green (Thermo Fisher Scientific, S7020) was added. The images were acquired every 1 h at 37 °C and 5% CO2. The resulting images were analysed using the software package supplied with the IncuCyte imager, which counts the number of Sytox Green-positive BMDM nuclei (Sytox+ BMDM nuclei) present in each image.

Immunoprecipitation

Immunoprecipitation was performed as described previously58. In brief, after HSV1 or F. novicida infection or poly(dA:dT) treatment, wild-type BMDMs were lysed in a buffer containing 20 mM Tris-HCl (pH 7.4), 100 mM NaCl, 30 mM KCl and 0.1% NP-40. After centrifugation at 16,000g for 10 min, the lysates were incubated with either IgG control antibody (CST, 3900 S) or anti-ASC antibody (AdipoGen; AG-25B-006-C100) with protein A/G PLUS-Agarose (Santa Cruz Biotechnology) overnight at 4 °C. After washing with the above buffer, the immunoprecipitated proteins were collected by boiling in 1× SDS loading buffer at 100 °C for 5 min.

Immunofluorescence staining

Immunofluorescence staining was performed as described previously59. In brief, after infection, BMDMs were fixed in 4% paraformaldehyde (ChemCruz, sc-281692) for 10 min and permeabilized with PBS containing 0.5% Triton X-100 for 3 min. Cells were then incubated in PBS containing 1% skim milk for 1 h. Alexa Fluor 488 (ThermoFisher, A20181)-conjugated anti-ASC (Millipore, 04-147), Alexa Fluor 532 (ThermoFisher, A20182)-conjugated anti-AIM2 (Abcam, ab119791), Alexa Fluor 568 (ThermoFisher, A20184)-conjugated anti-pyrin (Abcam, ab195975), Alexa Fluor 647 (ThermoFisher, A20186)-conjugated anti-ZBP1 (AdipoGen, AG-20B-0010), Alexa Fluor 568 (ThermoFisher, A20184)-conjugated anti-RIPK3 (ProSci, 2283) or Alexa Fluor 647 (ThermoFisher, A20186)-conjugated anti-caspase-8 (Enzo, 1G12) were made according to the manufacturer’s instructions, and the coverslips were incubated with indicated antibodies (1:100) for 1 h. Cells were counterstained with DAPI mounting medium (Invitrogen, P36931). Images were acquired by confocal laser scanning microscopy (LSM780; Carl Zeiss) using 63× Apochromat objective.

Cytokine analysis

Cytokines were detected by using multiplex ELISA (Millipore, MCYTOMAG-70K), IL-18 ELISA (Invitrogen, BMS618-3) or IFN-β ELISA (BioLegend, 439408) according to the manufacturer’s instructions.

Statistical analysis

GraphPad Prism 8.0 software was used for data analysis. Data are presented as mean ± s.e.m. Statistical significance was determined by t-tests (two-tailed) for two groups or one-way ANOVA with Dunnett’s multiple comparisons test for more groups. Survival analysis was performed using the log-rank (Mantel–Cox) test. P values less than 0.05 were considered statistically significant where, *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this paper.