RIPK1 distinctly regulates Yersinia-induced inflammatory cell death, PANoptosis

Abstract

Bacterial pathogens from the genus Yersinia cause fatal sepsis and gastritis in humans. Innate immune signaling and inflammatory cell death (pyroptosis, apoptosis, necroptosis [PANoptosis]) serve as a first line of antimicrobial host defense. The receptor-interacting protein kinase 1 (RIPK1) is essential for Yersinia-induced pyroptosis and apoptosis and an effective host response. However, it is not clear if RIPK1 assembles a multifaceted cell death complex capable of regulating caspase-dependent pyroptosis and apoptosis, or whether there is crosstalk with necroptosis under these conditions. Here, we report that Yersinia activates PANoptosis as evidenced by the concerted activation of proteins involved in pyroptosis, apoptosis, and necroptosis. Genetic deletion of RIPK1 abrogated the Yersinia-induced activation of the inflammasome/pyroptosis and apoptosis, but enhanced necroptosis. We also found that Yersinia induced assembly of a RIPK1 PANoptosome complex capable of regulating all three branches of PANoptosis. Overall, our results demonstrate a role for the RIPK1 PANoptosome in Yersinia-induced inflammatory cell death and host defense.

INTRODUCTION

Pathogenic Yersinia species induce severe sepsis and gastritis in humans and animals (1, 2). Regulated cell death is a key part of the host innate immune response and is a critical host defense mechanism in response to infection to curtail microbial dissemination and pathogenesis. Sensing of microbial infections by innate immune sensors triggers assembly of inflammasome, apoptosome, and necrosome complexes, which in turn activate inflammatory cell death (3). A plethora of studies including findings from our lab indicate extensive crosstalk among seemingly different cell death pathways (4–16), which led us to propose the concept of PANoptosis (17). PANoptosis is a unique, physiologically relevant, inflammatory programmed cell death pathway activated by specific triggers and regulated by the PANoptosome complex. The PANoptosome provides a molecular scaffold for contemporaneous engagement of key molecules from pyroptosis, apoptosis, and necroptosis, and it cannot be blocked by the terminal effectors of these individual pathways (11–13, 18). It is increasingly clear that PANoptosis is implicated in a wide range of infectious and inflammatory diseases and cancer (7–13, 15, 16, 19). However, the molecular composition of the PANoptosome and the phenotypic outcomes of the crosstalk and coregulation among pyroptosis, apoptosis, and necroptosis are dependent on the stimulus provided.

Yersinia infection induces robust inflammatory cell death in immune cells. Proteins produced by Yersinia inhibit the mitogen-activated protein kinase (MAPK) transforming growth factor-β activated kinase-1 (TAK1) function, which results in the loss of cellular homeostasis and activation of programmed cell death pathways (20–28). Previous studies have established that TAK1 restricts the spontaneous activation of the NLRP3 inflammasome and inflammatory PANoptotic cell death in macrophages (10). Subsequent studies demonstrated that both the pharmacological- and pathogen-mediated inhibition of TAK1 induce activation of caspase-1 (CASP1) and caspase-8 (CASP8), which play a key role in gasdermin D (GSDMD)-induced pyroptosis during Yersinia infection (11, 29, 30). In addition, our previous studies revealed a critical role for the kinase activity-dependent and -independent functions of RIPK1 in regulating TAK1 inhibition- and Yersinia-induced NLRP3 inflammasome, CASP8, and GSDMD activation (10, 11). In contrast, RIPK1 deletion results in the spontaneous activation of Z-DNA-binding protein 1 (ZBP1), an innate immune sensor of Z-nucleic acids, which in turn triggers RIPK3-mediated inflammatory cell death and embryonic lethality (31–37). Overall, the functional and molecular crosstalk among different RIPK1-mediated PANoptotic cell death components is only starting to emerge and remains poorly understood. Importantly, it is not clear whether Yersinia infection induces the assembly of a single RIPK1 PANoptosome cell death complex, and whether this complex is functional in driving the activation of PANoptosis.

In this study, we demonstrate that Yersinia infection activated pyroptosis, apoptosis, and necroptosis (PANoptosis) and that RIPK1 mediated the assembly of the PANoptosome to regulate Yersinia infection-induced PANoptosis and inflammatory responses. We also demonstrate a complex regulation of RIPK1 function associated with cellular homeostasis during Yersinia infection.

MATERIALS AND METHODS

Mice

Casp1/11^−/− (38), Ripk1^−/− (39); Ripk3^−⁄− (40); Ripk3^−⁄−Casp8^−⁄− (41), Ripk3^−⁄−Casp8^−⁄−Casp1/11^−⁄− (8) Zbp1^−⁄− (42), Casp3^–/– (43), Casp7^–/– (44), Gsdmd^–/– (45), and Mlkl^–/– (46), mouse strains were all described previously. C57BL/6 wild-type (WT) (The Jackson Laboratory) and littermate controls were bred at St. Jude Children’s Research Hospital (St. Jude). Animal studies were conducted under protocols approved by the St. Jude Animal Care and Use Committee.

Macrophage cultures

Total bone marrow cells from 6 – 10-week-old mice or fetal liver cells from day 13.5 embryos were harvested and cultured for 6 days in DMEM (Thermo Fisher Scientific, 11995–073) with 10% FBS (Biowest, S1620), 30% L929-conditioned medium, and 1% penicillin and streptomycin (Sigma Aldrich) to differentiate into primary bone marrow–derived macrophages (BMDMs) and fetal liver-derived macrophages (FLDMs), respectively. The macrophages were resuspended in antibiotic-free DMEM + FBS and seeded in 12-well plates at a density of 1 × 10⁶ cells per well. BMDMs and FLDMs were incubated overnight at 37°C followed by infection with Yersinia on the next day.

Yersinia infection

Yersinia pseudotuberculosis (Yp) was cultured on yeast extract tryptone (YT) agar plates at 37°C overnight under aerobic conditions. Single colonies of Yp were inoculated and cultured at 37°C overnight. Yp was subcultured further in fresh YT broth at 26°C, washed, and resuspended in PBS to infect BMDMs or FLDMs at a MOI of 5. After infection for 2 h, extracellular bacteria were killed by adding 50 μg/ml gentamicin.

Immunoblotting analysis

Whole cell lysates were prepared with cell culture supernatants for probing caspase activation and without supernatants for monitoring signaling activation as described previously (47), after Yersinia infection. SDS-PAGE sample buffer was added to cell lysates and boiled at 100°C for 10 min. After performing SDS-PAGE, separated proteins were transferred onto polyvinylidene difluoride membranes (Millipore). Proteins on the membrane were probed with primary antibodies against ASC (AG-25b-006–300, AdipoGen), caspase-1 (AG-20B-0042, AdipoGen), caspase-3 (9662, Cell Signaling Technology), cleaved caspase-3 (9661, Cell Signaling Technology), caspase-7 (9492, Cell Signaling Technology), cleaved caspase-7 (9491, Cell Signaling Technology), caspase-8 (AG-20T-0138-C100, AdipoGen), cleaved caspase-8 (8592, Cell Signaling Technology), FADD (05–486, Millipore), GSDMD (Ab209845, Abcam), RIPK1 (3493, Cell Signaling Technology), phospho-MLKL (37333, Cell Signaling Technology), RIPK3 (95702, Cell Signaling Technology), and β-actin (66009–1-IG, Proteintech). These membranes were further probed with HRP-conjugated secondary antibodies (anti-rabbit [111–035-047], anti-mouse [315–035-047], and anti-rat [112–035−−003], Jackson Immuno Research Laboratories).

Immunoprecipitation of RIPK1 PANoptosome

In vitro-differentiated primary WT C57BL/6 BMDMs were seeded 24 h prior to stimulations. BMDMs were infected with Yp (MOI, 5), Yp (MOI, 5) + zVAD (pan-caspase inhibitor, 30 μM), or mock infected for 6–8 h. BMDMs were subjected to lysis in NP-40 lysis buffer (50 mM HEPES, 150 mM NaCl, 1% NP-40, and protease inhibitor cocktail) containing 1 mM N-ethylmaleimide (Sigma-Aldrich). Whole cell lysates were harvested and incubated with 4 μg of indicated primary antibodies overnight at 4°C. Protein A/G plus agarose beads (Santa Cruz Biotechnology) were incubated in 3% BSA for 30 min, washed in PBS, and then incubated in NP-40 lysis buffer for 15 min. Agarose beads were then added to cell lysate samples and incubated at 4°C for another 2 h. Agarose beads were allowed to settle and washed in NP-40 lysis buffer at least three times. Immunoprecipitated proteins were eluted by boiling in SDS-PAGE sample buffer and were then subjected to immunoblotting analysis.

Real-time cell death analysis

In brief, macrophages were seeded at a density of 0.5 × 10⁶ cells/well in 24-well tissue culture plates and infected with Yp (MOI, 5). The nuclei of dying cells were stained using 20 nM SYTOX Green dye (Thermo Fisher Scientific, S7020). The image analysis, masking, and quantification of the dead cells were done using IncuCyte S3 software.

Cytokine analysis

The concentration of IL-18 was determined from the culture supernatants by classical ELISA for IL-18 (MBL International), as per the manufacturer’s instructions.

Statistics

For data analyses, GraphPad Prism version 5.0 software was used. Data are plotted and provided as mean ± SEM. The statistical significance for each of the experiments was calculated using t tests (two-tailed) for comparison between two groups and one-way ANOVA (with Dunnett’s or Tukey’s multiple comparisons tests) for comparison between more than two groups. Information about the number of experimental repeats is provided in the corresponding figure legends. The letter ’n’ in the figure legends represents the number of biological replicates used in the experiments. P < 0.05 was considered statistically significant.

RESULTS

Yersinia infection induces PANoptosis in macrophages

Yersinia has evolved to modulate host cell death responses through its immune evasion strategies. However, host immune cells sense this modulation to activate alternate forms of regulated cell death to restrict Yersinia infection. Recent studies showed that concerted activation of the multifaceted cell death mechanism PANoptosis is induced by microbial infections (13). Specifically, infection with the viral pathogens influenza A virus (IAV), vesicular stomatitis virus (VSV), coronavirus mouse hepatitis virus (MHV), the bacterial pathogens Listeria monocytogenes and Salmonella enterica serovar Typhimurium, and the fungal pathogens Candida albicans and Aspergillus fumigatus results in robust cell death with the hallmarks of PANoptosis (9, 12, 13, 16, 19, 48). A subsequent study confirmed the interconnection of pyroptosis, apoptosis, and necroptosis during Salmonella infection (49). However, whether Yersinia infection induces PANoptosis is not clear, and among pathogens that do induce PANoptosis, the molecular details and phenotypic outcomes of the crosstalk and coregulation among pyroptosis, apoptosis, and necroptosis can differ depending on the stimulus provided. Therefore, we assessed the ability of Yersinia to induce PANoptosis and systematically analyzed the results of these processes.

Using real-time imaging analysis, we found that Yersinia pseudotuberculosis infection induced robust cell death in wild-type (WT) bone marrow-derived macrophages (BMDMs) (Fig. 1A). We next sought to investigate whether this cell death in WT BMDMs was due to the activation of a single regulated cell death pathway or concerted activation of pyroptosis, apoptosis, and necroptosis (PANoptosis). Yersinia infection of Casp1/11^–/– BMDMs, which are defective in pyroptosis, resulted in cell death comparable to that observed in WT BMDMs (Fig. 1A). Similarly, Yersinia infection of Ripk3^–/– BMDMs, which are defective in necroptosis, also failed to show reduced cell death induction (Fig. 1A). Additionally, genetic deletion of pyroptosis (GSDMD), apoptosis (CASP3 or CASP7), or necroptosis (MLKL) individually failed to provide protection from Yersinia-induced cell death (Supp. Fig. 1). These observations suggest that Yersinia infection of BMDMs activates multiple regulated cell death pathways. Since CASP8 was reported to be essential for TAK1 inhibition- and Yersinia-induced apoptosis and GSDMD-mediated pyroptosis (11, 29, 30), we compared cell death kinetics between WT, Ripk3^–/–, Ripk3^–/–Casp8^–/–, and Ripk3^–/–Casp8^–/–Casp1/11^–/– BMDMs infected with Yersinia and found that Ripk3^–/–Casp8^–/– BMDMs showed substantially reduced cell death (Fig. 1A). Loss of CASP1/11 in addition to RIPK3 and CASP8 (Ripk3^–/–Casp8^–/–Casp1/11^–/– BMDMs) showed a trend toward a further reduction in cell death compared with Ripk3^–/–Casp8^–/– BMDMs, although this was not significant (Fig. 1A). These results indicate that combined loss of the key cell death activators of pyroptosis, apoptosis, and necroptosis leads to the abrogation of Yersinia-induced cell death.

FIGURE 1. Yersinia induces PANoptosis.

(A) Time course analysis of cell death in the indicated bone marrow-derived macrophages (BMDMs) during infection with Yersinia pseudotuberculosis (Yp) quantified by IncuCyte image analysis. (B) Measurement of IL-18 release from the culture supernatants of BMDMs collected at 12 h post-infection with Yp. (C) Immunoblot analysis of pro–caspase-1 (CASP1, p45) and the active CASP1 subunit (p20), pro-gasdermin D (GSDMD, p53), active GSDMD (p30), and inactive GSDMD (p22), pro–caspase-8 (CASP8, p55) and active CASP8 (p18), pro–caspase-7 (CASP7, p35) and active CASP7 (p20), pro–caspase-3 (CASP3, p35) and active CASP3 (p19/17), active phosphorylated MLKL (pMLKL, p54), and loading control β-Actin (p42) in the indicated BMDMs assessed in culture at 12 h after infection with Yp. All data are provided as mean ± SEM (A and B). “p” in Western blots denotes protein molecular weight. P < 0.05 is considered statistically significant. ns, not significant; **P < 0.01; ***P < 0.001; ****P < 0.0001 (one-way ANOVA [A], two-tailed t test [B]). Data are representative of three independent experiments with n = 2–3 (A–C).

In contrast to the cell death analysis, where loss of CASP1/11 had no impact on the phenotype, the release of the inflammasome-dependent proinflammatory cytokine IL-18 was significantly reduced in Casp1/11^–/– cells compared with WT BMDMs (Fig. 1B). IL-18 release was also significantly reduced in Ripk3^–/–Casp8^–/– BMDMs, and the observed defect was more pronounced in Ripk3^–/–Casp8^–/–Casp1/11^–/– BMDMs (Fig. 1B).

Further biochemical analyses using immunoblotting indicated that Yersinia infection of BMDMs induced activation of the pyroptotic markers CASP1 and GSDMD, the apoptotic initiator caspase CASP8 and executioners CASP3 and CASP7, and the phosphorylation of MLKL, which is required for necroptosis activation (Fig. 1C). Although Ripk3^–/–Casp8^–/– BMDMs showed a significant reduction in Yersinia-induced cell death (Fig. 1A), we detected low-level activation of CASP1 and GSDMD in these BMDMs, indicating pyroptosis was still occurring. In contrast, the activation of pyroptosis cell death markers was abolished in Yersinia-infected Ripk3^–/–Casp8^–/–Casp1/11^–/– BMDMs compared with Ripk3^–/–Casp8^–/– BMDMs (Fig. 1C). Collectively, the cell death and biochemical analyses indicate that Yersinia induces PANoptosis in BMDMs.

During IAV infection, the Z-nucleic acid sensor ZBP1 senses IAV and assembles the ZBP1-PANoptosome to trigger PANoptosis (9, 12, 48). Additionally, ZBP1 is required for fungi-induced PANoptosis (19). The established role of ZBP1 as a sensor in PANoptosis led us to ask whether ZBP1 is also involved in Yersinia-induced PANoptosis. However, genetic deletion of ZBP1 did not reduce PANoptosis in Yersinia-infected macrophages (Fig. 1). These results demonstrate that Yersinia infection activates a distinct PANoptosome to drive ZBP1-independent PANoptosis.

RIPK1 deficiency leads to reduced pyroptosis and apoptosis but enhanced necroptosis

Based on the above observations and the predominant role of the RIPK3 and CASP8 pathway in Yersinia infection-induced inflammatory cell death, we hypothesized that RIPK1 acts as a master regulator and upstream sensor to drive PANoptosis during Yersinia infection. Loss of RIPK1 expression causes embryonic lethality in mice, precluding the generation of Ripk1^–/– BMDMs (39). We therefore generated littermate WT and Ripk1^–/– fetal liver-derived macrophages (FLDMs) to study the role of RIPK1 in Yersinia-induced PANoptosis. We observed a substantial reduction in CASP1, GSDMD, CASP3, CASP7, and CASP8 activation in Yersinia-infected Ripk1^–/– FLDMs compared with littermate WT control cells (Fig. 2A). This indicated that the lack of RIPK1 expression failed to activate Yersinia-induced pyroptosis and apoptosis. Consistent with the reduced activation of CASP1, the release of the inflammasome-dependent proinflammatory cytokine IL-18 was significantly reduced in Ripk1^–/– FLDMs (Fig. 2B). In contrast to the reduced activation of pyroptotic and apoptotic markers, we observed spontaneous activation of MLKL, as measured by phosphorylation of MLKL (pMLKL), in Ripk1^–/– FLDMs, confirming that RIPK1 negatively regulates necroptosis to preserve cellular homeostasis (33, 35–37, 50, 51); the activation of MLKL was further amplified by Yersinia infection (Fig. 2A). In the context of cellular homeostasis, RIPK1 deficiency-induced necroptosis requires ZBP1 (50, 51), but our data show that ZBP1 is not required for Yersinia-induced necroptosis in RIPK1-sufficient cells as evidenced by normal MLKL activation in Zbp1^−/− cells (Fig. 1C); this suggests that the ZBP1-dependent necroptosis in the absence of RIPK1 and the necroptosis being activated by Yersinia use distinct upstream mechanisms. Real-time cell death analysis showed no significant reduction in cell death in the absence of RIPK1 (Fig. 2C), suggesting that the increase in necroptosis compensated for the loss of pyroptosis and apoptosis. Together, these findings support a positive role for RIPK1 in promoting Yersinia-induced PANoptosis in WT macrophages and confirm its deficiency as a causative factor responsible for sensitizing macrophages to undergo necroptotic cell death (Fig. 2D).

FIGURE 2. RIPK1 regulates Yersinia-induced PANoptosis.

(A) Immunoblot analysis of pro–caspase-1 (CASP1, p45) and the active CASP1 subunit (p20), active gasdermin D (GSDMD, p30) and inactive GSDMD (p22), pro–caspase-8 (CASP8, p55) and active CASP8 (p18), pro–caspase-7 (CASP7, p35) and active CASP7 (p20), pro–caspase-3 (CASP3, p35) and active CASP3 (p19/17), active phosphorylated MLKL (pMLKL, p54), and loading control β-Actin (p42) in RIPK1-deficient and littermate control fetal liver-derived macrophages (FLDMs) assessed in culture at 12 h after infection with Yersinia pseudotuberculosis (Yp). (B) Measurement of IL-18 release from the culture supernatants of FLDMs collected at 12 h post-infection with Yp. (C) Time course analysis of cell death in the indicated FLDMs quantified by IncuCyte image analysis. (D) Representation of the differential regulation of Yersinia-induced RIPK1-directed PANoptosis in wild-type (WT) macrophages compared with RIPK1-deficient cells. All data are provided as mean ± SEM (B and C). “p” in Western blots denotes protein molecular weight. P < 0.05 is considered statistically significant. ns, not significant; ****P < 0.0001 (two-tailed t test [B], one-way ANOVA [C]). Data are representative of three independent experiments with n = 3 (A–C).

Yersinia induces assembly of the RIPK1 PANoptosome complex

The essential role of RIPK1 in activating Yersinia-induced PANoptosis prompted us to test its role in assembly of the PANoptosome complex. We infected WT BMDMs with Yersinia or Yersinia in combination with the pan-caspase inhibitor zVAD. Western blot analysis of input samples and the immunoprecipitated proteins from the control IgG and RIPK1 antibody pull-downs showed that RIPK1 formed a molecular complex with CASP8, FADD, NLRP3, ASC, and RIPK3, demonstrating PANoptosome formation in response to Yersinia infection (Fig. 3A–B). Formation of this RIPK1 complex was further enhanced in the presence of zVAD (Fig. 3B), perhaps because treatment with zVAD prevents RIPK1 cleavage by CASP8. Overall, these observations indicate that Yersinia infection induces RIPK1 PANoptosome complex formation to drive PANoptosis (Fig. 3C).

FIGURE 3. Yersinia induces assembly of the RIPK1 PANoptosome complex.

(A-B) Immunoprecipitation (IP) of endogenous RIPK1 from lysates of wild-type (WT) bone marrow-derived macrophages (BMDMs) infected with Yersinia pseudotuberculosis (Yp) or Yp + zVAD followed by Western blot analysis of the PANoptosome components RIPK1, caspase-8 (CASP8), FADD, ASC, RIPK3, and NLRP3 from the input samples (A) and the corresponding immunoprecipitated products (B) (n = 2). Red asterisks indicate non-specific protein bands in the Western blots. (C) Depiction of the Yersinia-induced RIPK1 PANoptosome complex based on the experimental observations in the current study.

DISCUSSION

In this study we show that Yersinia induces PANoptosis and that RIPK1 assembles a PANoptosome protein complex in WT macrophages to drive this process. Genetic deletion of RIPK1 results in abrogation of Yersinia-induced apoptosis and pyroptosis, but not necroptosis. Instead, the loss of RIPK1 results in spontaneous activation of necroptosis, and this activation is further enhanced after Yersinia infection. These findings support the previously described role of RIPK1 in regulating RIPK3 quiescence to control the activation of necroptosis.

Our data demonstrate that hyperactivation of the RIPK3-MLKL axis is not sufficient to bypass the requirement for RIPK1 to drive Yersinia-induced inflammasome activation and IL-18 secretion, demonstrating an indispensable role of RIPK1 function for host defense against Yersinia infection. It is now well known that RIPK1 is directly phosphorylated by the MAP kinase TAK1, allowing it to act as a sensor of TAK1 activation status (52). Yersinia inhibits TAK1 function, resulting in the loss of cellular homeostasis and leading to RIPK1-dependent necroptosis (20–28). Moreover, we found that the upstream molecular events of Yersinia-induced necroptosis are mechanistically distinct from the RIPK1 deficiency-induced necroptosis, where ZBP1 triggers the activation of necroptosis (50, 51); our data conclusively show that ZBP1 is not required for Yersinia-driven MLKL activation and necroptosis. Furthermore, while ZBP1 is required for PANoptosis during IAV and fungal infections (9, 12, 19, 48), we observed that Yersinia induces a ZBP1-independent but RIPK1-dependent form of PANoptosis, highlighting the distinct PANoptosomes that can be formed under various conditions.

Overall, our findings reveal that RIPK1 regulates PANoptosis and is evolved such that both its activation and inactivation by microbial pathogens instigates robust inflammatory cell death and promotes host innate immune defense. PANoptosis is increasingly recognized as a common mechanism of inflammatory cell death that is particularly important in the case of microbial infections. Previous studies have shown that bacterial, viral, and fungal pathogens induce robust PANoptotic cell death (9, 13, 16, 19, 49). These studies indicate the existence of key innate immune sensors that act as master regulators of PANoptotic cell death. The increased understanding of the molecular mechanism of RIPK1 provided here may inform strategies for therapeutic modulation of RIPK1 to manage infectious and inflammatory disease conditions.