NLRP6 induces RIP1 kinase-dependent necroptosis via TAK1-mediated p38^MAPK/MK2 phosphorylation in S. typhimurium infection

Summary

Programmed cell death (PCD) is tightly orchestrated by molecularly defined executors and signaling pathways. NLRP6, a member of cytoplasmic pattern recognition receptors, has a multifaceted role in host resistance to bacterial infection. However, whether and how NLRP6 may contribute to regulate host PCD during Gram-negative bacterial infection remain to be illuminated. Here, we report that NLRP6 promotes RIP1 kinase-mediated necroptosis, a form of lytic PCD, in both an in vitro and in vivo model of Salmonella typhimurium infection. By downregulating TAK1-mediated p38^MAPK/MK2 phosphorylation, NLRP6 decreased RIP1 phosphorylation at residue S321 and subsequently increased RIP1 kinase-dependent MLKL phosphorylation. Suppression of p38^MAPK/MK2 cascade not only reduced the number of dead cells caused by NLRP6 but also decreased the systemic dissemination of S. typhimurium resulting from NLRP6. Taken together, our findings provide new insights into the role and regulatory mechanism of NLRP6-associated antimicrobial responses by revealing a function for NLRP6 in regulating necroptosis.

Graphical abstract

Highlights

• •NLRP6 promotes S. typhimurium -induced necroptosis in a RIP1 dependent manner
• •NLRP6 decreases phosphorylation of RIP1 at S321 via the TAK1-p38MAPK-MK2 pathway
• •NLRP6 deficiency is benefit for controlling S. typhimurium dissemination

Introduction

Necroptosis, a modality of programmed cell death (PCD) first described in 2005,¹ has important functions in host antimicrobial defense and inflammation.² The serine/threonine kinase RIP1 (receptor interacting protein kinase-1) is identified as a central mediator of necroptosis, and its activity is tightly controlled by phosphorylation modification. When subjected to either extrinsic or intrinsic driving factors, cells initiate the assembly of a cell death platform containing RIP1 and receptor interacting protein kinase-3 (RIP3), termed necrosome.³ RIP3 subsequently recruits and mediates phosphorylation of its substrate mixed lineage kinase domain-like pseudokinase (MLKL), leading to MLKL translocation from the cytoplasm to the plasma membrane and induction of membrane damage.⁴ MLKL serves as the necroptotic executioner to initiate lytic cell death, which not only increases the release of damage-associated molecular patterns (DAMPs) such as ATP and high-mobility group box 1 (HMGB1) but also triggers the inflammatory response and cell death mechanism in surrounding cells.⁵

NLRP6, a recently discovered member of the nucleotide-oligomerization domain-like receptors (NLRs), is present in the cytosol of both myeloid and intestinal epithelial cells, acting as a crucial regulator involved in maintaining intestinal homeostasis and orchestrating immune responses.^6,7 NLRP6 recruits the adaptor protein ASC to form an inflammasome complex with caspase-1 and caspase-11, resulting in processing pro-interleukin-1β (pro-IL-1β) and −18 (pro-IL-18) into their mature form.⁸ Unlike other NLRPs, NLRP6 is also capable of regulating inflammatory signaling and impeding bacterial clearance in an inflammasome-independent manner.⁹ An interesting study by Ghimire et al. reported that NLRP6-deficient macrophages showed reduced necroptosis following Gram-positive bacterial infection.¹⁰ However, detailed signaling mechanisms of how NLRP6 induced cell death by necroptosis have not been explored. In addition, whether NLRP6 is associated with necroptosis in response to Gram-negative bacterial infection remains elusive.

Salmonella enterica serovar typhimurium (S. typhimurium) is one of the most common Gram-negative facultative intracellular bacterium that has long been used as a model for investigating the interaction between enteropathogenic bacteria and the host.^11,12 Previous studies have shown that host PCD is a crucial immune defense mechanism in response to bacterial infection.¹³ For instance, the sacrifice of infected cells, such as the expulsion of epithelial cells containing S. typhimurium, is required for removing replicative niches for intracellular bacteria and preventing its negative influence on the remaining healthy cells.¹⁴ However, cell death is a double-edged sword, which also has benefits for bacterial pathogens. By manipulating the death of infected cells, S. typhimurium crosses the intestinal barrier and escapes from immune cells to maintain their nutrient requirement and promote their pathogenesis.¹⁵ Inhibition of necroptosis, therefore, is beneficial to ameliorating inflammatory damage, limiting pathogen spread, and promoting host survival during S. typhimurium infection.^16,17,18 Our recent study has reported that NLRP6 facilitated S. typhimurium intracellular replication by regulating host iron metabolism.¹⁹ In the current study, we further reveal a detrimental role of NLRP6 in promoting S. typhimurium systemic dissemination. Our results identify potential interconnection between NLRP6 and RIP1 kinase-dependent necroptosis during S. typhimurium infection, and suggest NLRP6 as a therapeutic target for limiting enteropathogenic bacteria systemic dissemination.

Results

NLRP6 enhances S. typhimurium -associated cell death via a pyroptosis-independent pathway

As a typical foodborne pathogen, S. typhimurium infection usually occurs through the oral route and first invades the intestinal epithelial cells (IECs). Once crossing the intestinal barrier, S. typhimurium is captured by phagocytes such as macrophages and further disseminates to systemic organs.²⁰ To explore whether NLRP6 is involved in host cell death during S. typhimurium infection, wild-type (WT) and NLRP6 knockout (NLRP6^−/−) human Caco-2 IEC lines were used to establish an in vitro model of S. typhimurium infection. There was no significant difference in the percentage of dead cells between S. typhimurium -infected WT and NLRP6^−/− Caco-2 cells at 2 h post-infection, whereas NLRP6^−/− Caco-2 cells displayed a significantly lower ratio of dead to normal cells when compared to WT Caco-2 cells at 4 and 16 h post-infection (Figures 1A and 1B). As a complementary approach, we also conducted studies detecting the cell viability. As shown in Figure 1C, the percentage of viable cells was significantly higher in NLRP6^−/− than in WT Caco-2 cells after infection. To further decipher the role of NLRP6 in phagocyte cell death during S. typhimurium infection, we conducted studies using bone marrow-derived macrophages (BMDMs) isolated from WT or Nlrp6^−/− mice. In line with the results obtained in IECs, the infected Nlrp6 ^−/− BMDMs exhibited a significant higher cell viability when compared with their WT counterparts (Figures 1D–1F). Given that mouse embryonic fibroblasts (MEFs) are an ideal tool for studying PCD and convenient for intervention experiments, we further performed in vitro infection experiments using MEFs isolated from WT or Nlrp6^−/− mice. Consistent with the above observation, NLRP6 deficiency decreased cell death in MEFs during infection (Figures 1G–1I).

Figure 1.

NLRP6 deficiency decreases S. typhimurium -induced cell death

(A) Cell death kinetics analyzed by LDH release in WT and NLRP6^−/− Caco-2 cells infected with S. typhimurium strain SL1344 at an MOI of ∼100 for the indicated time points.

(B) Cell death assayed by Sytox Green staining in WT and NLRP6^−/− Caco-2 cells infected with SL1344 at an MOI of ∼100 for 16 h.

(C) Cell viability of WT and NLRP6^−/− Caco-2 cells infected with SL1344 at an MOI of ∼100 for 16 h.

(D) Cell death kinetics analyzed by LDH release of WT and Nlrp6^−/− BMDMs infected with SL1344 at an MOI of 20 for the indicated time points.

(E) Cell death assayed by Sytox Green staining of WT and Nlrp6^−/− BMDMs infected with SL1344 at an MOI of 20 for 4 h.

(F) Cell viability of WT and Nlrp6^−/− BMDMs infected with SL1344 at an MOI of 20 for 4 h.

(G) Cell death kinetics analyzed by LDH release of WT and Nlrp6^−/− MEFs infected with SL1344 at an MOI of 10 for the indicated time points.

(H) Cell death assayed by Sytox Green staining of WT and Nlrp6^−/− MEFs infected with SL1344 at an MOI of 10 for 4 h.

(I) Cell viability of MEFs of WT and Nlrp6^−/− MEFs infected with SL1344 at an MOI of 10 for 4 h.

(J) Membrane permeability of WT and Nlrp6^−/− BMDMs infected with SL1344 at an MOI of 20 for 4 h. (n > 1000 cells per group).

(K) Western blot analysis of WT and Nlrp6^−/− MEFs infected with SL1344 at an MOI of 10 for 4 h.

(L) Cell death assayed by LDH release of WT and Nlrp6^−/− MEFs transfected with 50 nM either negative-control siRNA or Gsdmd-specific siRNA for 48 h, followed by infection with SL1344 at an MOI of 10 for 4 h. Data were compared with independent Student’s t test. Values are expressed as the mean ± SEM, and statistically significant differences are indicated. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ns, not significant.

A recent study reported that, during Gram-positive bacterial infection, NLRP6 promotes pyroptosis-associated cell lysis via assembling the NLRP6-ASC-caspase1/11 inflammasome complex and subsequent cleavage of the pyroptosis execution protein gasdermin D (GSDMD).¹⁰ Therefore, we wondered whether NLRP6 had a role in promoting pyroptosis during Gram-negative bacterial infection. Our data showed NLRP6 deficiency dramatically reduced membrane permeability in S. typhimurium -infected cells (Figure 1J), suggesting that NLRP6-associated cell death is accompanied by plasma membrane rupture. To our surprise, immunofluorescence analysis showed that S. typhimurium infection led to an increase of both NLRP6 and ASC within the cytoplasm of the Caco-2 cells, while we did not observe NLRP6 co-localize with ASC (Figure S1A). Importantly, loss of NLRP6 had no significant effect on S. typhimurium -induced cleavage of GSDMD in MEFs, Caco-2 cells and BMDMs (Figures 1K, S1B, and S2A). Knockdown of Gsdmd in MEFs using a small interfering RNA (siRNA) could not reverse the effect of NLRP6 deficiency on cell death (Figure 1L). In line with these findings, there was no significant difference in the cleavage-activation of either caspase-1 or caspase-4/11 and the release of IL-1β between S. typhimurium -infected WT and NLRP6 knockout cells (Figures S1C–S1I and S2B–S2F). These findings suggest that NLRP6 facilitates cell death during S. typhimurium infection. However, this phenomenon is independent of inflammasome-mediated pyroptosis.

NLRP6 promotes necroptosis in a RIP1 kinase-dependent manner during S. typhimurium infection

Necroptosis and pyroptosis are two of the most common types of PCD associated with plasma membrane pore formation.²¹ To investigate the contribution of NLRP6 to S. typhimurium -relevant necroptosis, the MLKL inhibitor GW806742X was used and the phosphorylation level of MLKL measured. Treatment with GW806742X reduced S. typhimurium -induced cell death and abolished the significant difference in the number of dead cells between WT and Nlrp6^−/− MEFs (Figure 2A). Of note, the level of MLKL phosphorylation was significantly lower in Nlrp6^−/− than in WT MEFs (Figure 2B), suggesting that NLRP6 mediates the induction of necroptosis during S. typhimurium infection. RIP1 is an upstream kinase that interacts with RIP3 and subsequently mediates the formation of RIP3-MLKL complexes, resulting in necroptosis induction. However, RIP3 can also induce necroptosis independently of RIP1.^22,23 Our data showed that treatment with the RIP3 inhibitor GSK’872 reduced S. typhimurium -induced cell death and reversed the significant difference in the number of dead cells between WT and Nlrp6^−/− MEFs (Figure 2C). HeLa cells, a RIP3-deficient cell line, were transfected with the HA-tagged RIP3 plasmid for establishing a non-pharmacological model.²⁴ We found a significantly higher percentage of cell death in HeLa cells co-expressing RIP3 and NLRP6 when compared to the control group during infection (Figure 2D). To further examine whether RIP1 is involved in NLRP6-mediated necroptosis, cells were treated with the RIP1 inhibitor necrostatin-1 (Nec-1) with or without the pan-caspase inhibitor zVAD as previously described.²⁵ Although the percentage of cell death was still lower in Nlrp6^−/− than in WT MEFs after administering zVAD, treatment with Nec-1 + zVAD not only reduced cell death but also reversed the apparent difference between the infected MEFs of these two genotypes (Figure 2E). Concomitantly, treatment of RIP3-expressed HeLa cells with either the MLKL inhibitor necrosulfonamide (NSA) or Nec-1 + zVAD abrogated the significant difference in the number of dead cells caused by NLRP6 (Figure 2F). We also found that phosphorylation of RIP1 at S166—a biomarker tightly associated with the activation of RIP1 and the initiation of necroptosis—was significantly lower in Nlrp6^−/− than in WT MEFs following S. typhimurium infection (Figure 2G). Meanwhile, a higher phosphorylated level of RIP1 at S321, a residue that inhibits RIP1 kinase activity, was observed in Nlrp6^−/− MEFs compared with WT MEFs (Figure 2G). Importantly, treatment of S. typhimurium -infected cells with the RIP1 inhibitor abolished the apparent difference in the phosphorylation level of RIP1(S166) and MLKL between these two genotypes during infection (Figure 2H). These observations collectively indicate that NLRP6 facilitates necroptosis during S. typhimurium infection through modulating RIP1 kinase activity and the subsequent cascade of RIP1-RIP3-MLKL.

Figure 2.

NLRP6 deficiency decreases S. typhimurium -induced necroptosis in a RIP1 kinase-dependent manner

(A) Cell death of WT and Nlrp6^−/− MEFs treated with either vehicle or the MLKL inhibitor GW806742X for 1 h, followed by infection with S. typhimurium strain SL1344 at an MOI of 10 for 4 h.

(B) Western blot analysis and relative intensity of WT and Nlrp6^−/− MEFs infected with SL1344 at an MOI of 10 for 4 h.

(C) Cell death of WT and Nlrp6^−/− MEFs treated with either vehicle or the RIP3 inhibitor GSK’872 for 1 h, followed by infection with SL1344 at an MOI of 10 for 4 h.

(D) Cell death of HeLa cells transfected with plasmids encoding His-NLRP6 together with HA-RIP3, followed by infection with SL1344 at an MOI of 100 for 4 h.

(E) Cell death of WT and Nlrp6^−/− MEFs treated with vehicle, the pan-caspase inhibitor zVAD or the RIP1 inhibitor Nec-1 along with zVAD for 1 h, followed by infection with SL1344 at an MOI of 10 for 4 h.

(F) Cell death of HeLa cells transfected with plasmids encoding His-NLRP6 together with HA-RIP3, after treated with vehicle, NSA or zVAD+Nec-1 for 1 h, followed by infection with SL1344 at an MOI of 100 for 4 h.

(G) Western blot analysis and relative intensity of WT and Nlrp6^−/− MEFs infected with SL1344 at an MOI of 10 for 4 h.

(H) Western blot analysis and relative intensity of WT and Nlrp6^−/− MEFs treated with either vehicle or the RIP1 inhibitor Nec-1 for 1 h, followed by infection with SL1344 at an MOI of 10 for 4 h. Data were compared with independent Student’s t test. Values are expressed as the mean ± SEM, and statistically significant differences are indicated. All data are representative of at least three independent experiments. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ns, not significant.

NLRP6 influences phosphorylation of RIP1 at S321 via the TAK1-p38^MAPK-MK2 signaling cascade

We next focused on elucidating the mechanisms underlying NLRP6-initiated necroptosis and the pathway whereby NLRP6 could regulate RIP1 phosphorylation at S321 in response to Gram-negative bacterial infection. It has been reported that NLRP6 dampens nuclear factor Kappa-B (NF-κB) activation.⁹ Therefore, we first sought to explore whether this pathway contributes to NLRP6-associated necroptosis. In line with the results obtained in the hematopoietic cells, our data showed that NLRP6 downregulated the kinetics of IκBα degradation in S. typhimurium -infected MEFs (Figure S3A), suggesting that NLRP6 could also inhibit NF-κB activity in non-hematopoietic cells. The activation of NF-κB can be triggered by various exogenous and endogenous stimuli such as interleukin-1β (IL1β), tumor necrosis factor (TNFα), or phorbol myristate acetate (PMA).²⁶ We found that either NF-κB activation or IκBα degradation induced by one of three agonists was suppressed by NLRP6 overexpression (Figures S3B–S3E). We further sought to identify where in the NF-κB signaling pathway NLRP6 might exert its inhibitory function. Overexpression of NLRP6 reduced NF-κB activity stimulated with both TNF-receptor associated factor 2 (TRAF2), the crucial downstream ligase of TNFα receptor, and TRAF6, the key downstream mediator of IL1β receptor. However, overexpression of NLRP6 had no inhibitory effect on the activation of NF-κB induced by TGF-β-activated kinase 1 (TAK1), inhibitor of nuclear factor-κB kinase β (IKKβ), and p65 (Figures S3F), suggesting that NLRP6 functioned at the level of the TAK1, which is an essential modulator that provides a vital link downstream of TRAF to upstream of IKK. We presume that TAK1 is involved not only in NLRP6-mediated NF-κB inhibition but also in NLRP6-associated necroptosis induction. Consistent with our hypothesis, treatment of S. typhimurium -infected MEFs with the inhibitor of TAK1 catalytic activity reversed the significant difference in cell death between WT and Nlrp6^−/− cells (Figure S3G).

We subsequently sought to determine the mechanism by which NLRP6 regulates TAK1-associated necroptosis. Our data showed that overexpression of NLRP6 had slight influence on the protein level of TAK1(Figure 3A). We also treated cells with the protein synthesis inhibitor cycloheximide (CHX) to assessed whether NLRP6 influences the stability of TAK1 protein and observed the degradation rate of TAK1 in WT cells was similar to that of Nlrp6^−/− cells (Figure 3B). These data suggested that NLRP6 may regulate the kinase activity of TAK1 but not the expression of TAK1. In line with our hypothesis, no apparent difference in the protein level of TAK1 was observed between S. typhimurium -infected WT and Nlrp6^−/− cells, while a significantly higher phosphorylated level of TAK1 was observed in Nlrp6^−/− than in WT MEFs (Figure 3C). TAK1 phosphorylation and activation were found to be closely related to its ubiquitination.^27,28 Our data supported these previous observations showing that NLRP6 interacted with TAK1 and Nlrp6^−/− cells displayed a dramatically higher level of ubiquitinated TAK1 when compared to WT cells during S. typhimurium infection (Figures 3D and 3E). The p38 Mitogen-activated protein kinase (p38^MAPK) is identified as one of the several downstream kinases that are regulated by TAK1 phosphorylation. Through phosphorylation of a direct downstream substrate MAPK-activated protein kinase 2 (MK2), the p38^MAPK-MK2 axis phosphorylates RIP1 at S321 to prevent necroptosis in inflammation and infection.²⁹ We presumed that the p38^MAPK/MK2 pathway has a potential role in NLRP6-induced TAK1-dependent necroptosis. Loss of NLRP6 had no significant effect on p38^MAPK and MK2 expression. However, treatment with the p38^MAPK inhibitor SB239063 increased S. typhimurium -induced cell death and abrogated the significant difference in the number of dead cells between WT and Nlrp6^−/− MEFs (Figure 3F). At 4 h post-infection, a dramatically higher accumulation of phosphorylated p38^MAPK was detected in Nlrp6^−/− than in WT MEFs (Figure 3G). Our data further showed that pharmacological inhibition of MK2 activity reversed the significant difference in NLRP6-caused cell death, and NLRP6 suppressed MK2 phosphorylation in S. typhimurium -infected MEFs (Figures 3H and 3I). Importantly, there were no apparent differences in the phosphorylation of RIP1 at S321 between these two groups of infected MEFs following administration with the MK2 inhibitor (Figure 3J). These findings demonstrate a role for the TAK1-p38^MAPK-MK2 kinase cascade in NLRP6-associated necroptosis-dependent enterocyte injury during S. typhimurium infection.

Figure 3.

NLRP6 deficiency increases RIP1 phosphorylation at S321 via the TAK1/p38^MAPK/MK2 cascade

(A) Western blot analysis and relative intensity of HEK293T cells transfected with plasmids encoding His-NLRP6 together with HA-RIP3 indicated concentration of Flag-TAK1.

(B) Western blot analysis and relative intensity of WT and Nlrp6^−/− MEFs treated with the protein synthesis inhibitor CHX for indicated time points.

(C) Western blot analysis and relative intensity of WT and Nlrp6^−/− MEFs infected with S. typhimurium strain SL1344 at an MOI of 10 for 4 h.

(D) Immunoblot analysis of whole-cell lysates of 293T cells transfected with plasmids encoding His-NLRP6 and Flag-TAK1, followed by immunoprecipitation with anti-Flag beads.

(E) Western blot analysis of protein extracts of WT and Nlrp6^−/− MEFs infected with SL1344 at an MOI of 10 for 4 h, followed by immunoprecipitation with anti-TAK1 antibody.

(F) Cell death of WT and Nlrp6^−/− MEFs treated with either vehicle or the p38^MAPK inhibitor SB239063 for 30 min, followed by infection with SL1344 at an MOI of 10 for 4 h.

(G) Western blot analysis and relative intensity of WT and Nlrp6^−/− MEFs infected with SL1344 at an MOI of 10 for 4 h.

(H) Cell death of WT and Nlrp6^−/− MEFs treated with either vehicle or the MK-2 inhibitor MK2-IN-1 for 30 min, followed by infection with SL1344 at an MOI of 10 for 4 h.

(I) Western blot analysis and relative intensity of WT and Nlrp6^−/− MEFs infected with SL1344 at an MOI of 10 for 4 h.

(J) Western blot analysis and relative intensity of WT and Nlrp6^−/− MEFs treated with vehicle or MK2-IN-1 for 30 min, followed by infection with SL1344 at an MOI of 10 for 4 h. Data were compared with independent Student’s t test. Values are expressed as the mean ± SEM, and statistically significant differences are indicated. All data are representative of at least three independent experiments. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ns, not significant.

NLRP6 modulates enterocyte and macrophage necroptosis during S. typhimurium infection

To investigate whether NLRP6-associated necroptosis occurs in IECs and macrophages, we performed in vitro infection experiments using Caco-2 cells and BMDMs. Our data show that diminishing the activity of MLKL, RIP3, or RIP1 resulted in the abrogation of the significant difference in cell death caused by NLRP6 (Figures 4A, 4C, S4A, and S4C). Western blot analyses show that S. typhimurium infection induced a significantly lower phosphorylated level of MLKL and RIP1 (S166) in NLRP6-deficient cells as compared to WT cells (Figures 4B–4D, S4B, and S4D). Importantly, S. typhimurium infected NLRP6-deficient cells displayed an apparent higher phosphorylated level of RIP1 at S320(human)/S321(mouse) when compared to WT cells (Figures 4E and S4E). These observations indicate that NLRP6 initiated necroptosis in both enterocytes and macrophages during S. typhimurium infection. We subsequently confirm the mechanism underlying the regulation of necroptosis by NLRP6. Our data showed that treatment with (5Z)-7-Oxozeaenol, SB239063, or MK2-IN-1 increased the number of dead cells in the infected NLRP6-deficient cells, which is similar to the WT cell results following inhibitor administration (Figures 4F, 4G, 4I, and S4F–S4H). Moreover, significantly higher levels of phosphorylated p38^MAPK and MK-2 were observed in NLRP6-deficient cells when compared to WT cells during infection (Figures 4H–4J and S4I). These observations demonstrate that NLRP6 enhances S. typhimurium associated necroptosis through a TAK1-p38^MAPK-MK2-dependent manner in both enterocytes and macrophages. Given that enterocyte cell death is one of the main reasons for the destruction of barrier function,³⁰ we further explored whether NLRP6-associated enterocyte necroptosis could influence intestinal permeability. As shown in Figure 4K, S. typhimurium infection resulted in a time-dependent decrease in transepithelial electrical resistance (TEER), and this decrease was greater in WT Caco-2 cell monolayers than NLRP6^−/− Caco-2 cell monolayers. Consistently, lower level of 4-kDa FITC-dextran (FD4) permeability and bacterial translocation were observed in NLRP6^−/− Caco-2 cell monolayers compared with WT Caco-2 cell monolayers after infection (Figures 4L and 4M), suggesting that NLRP6-initiated necroptosis contributed to the systemic dissemination of bacteria.

Figure 4.

NLRP6 deficiency decreases S. typhimurium -induced enterocyte necroptosis and intestinal barrier dysfunction

(A) Cell death of WT and NLRP6^−/− Caco-2 cells treated with vehicle, NSA or GSK’872 for 1 h, followed by infection with S. typhimurium strain SL1344 at an MOI of ∼100 for 4 h.

(B) Western blot analysis and relative intensity of WT and NLRP6^−/− Caco-2 cells infected with SL1344 at an MOI of ∼100 for 4 h.

(C) Cell death of WT and NLRP6−/− Caco-2 cells treated with vehicle, zVAD or zVAD + Nec-1 for 1 h, followed by infection with S. typhimurium strain SL1344 at an MOI of ∼100 for 4 h.

(D and E) Western blot analysis and relative intensity of WT and NLRP6^−/− Caco-2 cells infected with SL1344 at an MOI of ∼100 for 4 h.

(F) Cell death of WT and NLRP6^−/− Caco-2 cells treated with vehicle or (5Z)-7-Oxozeaenol for 1 h, followed by infection with SL1344 at an MOI of ∼100 for 4 h.

(G) Cell death of WT and NLRP6^−/− Caco-2 cells treated with vehicle or SB239063 for 30 min, followed by infection with SL1344 at an MOI of ∼100 for 4 h.

(H) Western blot analysis and relative intensity of WT and NLRP6^−/− Caco-2 cells infected with SL1344 at an MOI of ∼100 for 4 h.

(I) Cell death of WT and NLRP6^−/− Caco-2 cells treated with vehicle or MK2-IN-1 for 30 min, followed by infection with SL1344 at an MOI of ∼100 for 4 h.

(J) Western blot analysis and relative intensity of WT and NLRP6^−/− Caco-2 cells infected with SL1344 at an MOI of ∼100 for 4 h.

(K) TEER levels of WT and NLRP6^−/− Caco-2 cell monolayers infected with SL1344 at an MOI of ∼100.

(L) FD4 permeability of WT and NLRP6^−/− Caco-2 cell monolayers infected with SL1344 at an MOI of ∼100 for 3 h.

(M) Bacterial translocation of WT and NLRP6^−/− Caco-2 cell monolayers infected with SL1344 at an MOI of ∼100 for 3 h. Data were compared with independent Student’s t test. Values are expressed as the mean ± SEM, and statistically significant differences are indicated. All data are representative of at least three independent experiments. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001; ns, not significant.

NLRP6-mediated regulation of RIP1 kinase activity contributes to S. typhimurium -induced necroptosis in vivo

To investigate whether NLRP6 worsens the outcome of Gram-negative bacterial infection and induces necroptosis in vivo, a well-established infection model of mouse administered with S. typhimurium orally was used. We found that the intestinal permeability, determining by luminal mucosa-to-blood flux of the paracellular marker FD4, and bacterial burden in MLN and systemic organs were significantly higher in WT mice than in Nlrp6^−/− mice (Figures 5A–5D). These data indicate that NLRP6 facilitates bacterial dissemination. Our in vitro experiments demonstrated a role for NLRP6 in promoting necroptosis in both hematopoietic (macrophages) and non-hematopoietic cells (epithelial cells). Therefore, we further generated an infection model of bone marrow chimeras. Our data show that NLRP6-deficiency in either hematopoietic and non-hematopoietic cells provided protection against S. typhimurium infection and spread, suggesting that NLRP6 acts as a pro-pathogen factor in both cell types (Figures 5E–5G). To determine if observed differences were due to changes in microbiome composition caused by NLRP6 deficiency, WT and Nlrp6^−/− mice were co-housed for 4 weeks and then infected with S. typhimurium. Similarly, the WT mice had apparent higher bacterial loads in systemic organs than the Nlrp6^−/− mice (Figures 5H–5J).

Figure 5.

NLRP6 promotes S. typhimurium -induced intestinal barrier dysfunction and necroptosis in vivo

(A–D) Streptomycin-pretreated WT and Nlrp6^−/− mice were orally infected with 5 × 10⁷ CFU of S. typhimurium strain SL1344 and analyzed at 3 days post-infection. Intestinal permeability (A). Bacterial burden in MLN (B), liver (C) and spleen (D).

(E–G) Streptomycin-pretreated bone marrow chimeras were orally infected with 5 × 10⁷ CFU of SL1344 and analyzed at 3 days post-infection. Bacterial burden in MLN (E), liver (F) and spleen (G).

(H–J) WT and Nlrp6^−/− mice were co-housed together for 4 weeks, and then orally infected with 5 × 10⁷ CFU of SL1344 and analyzed at 3 days post-infection. Bacterial burden in MLN (H), liver (I) and spleen (J).

(K–M) Streptomycin-pretreated WT and Nlrp6^−/− mice were orally infected with 1 × 10⁸ CFU of S. typhimurium strain SL1344 and analyzed at 2 days post-infection. Immunohistochemical staining of colon sections. scale bars: 100 μm (K). Western blot analysis and relative intensity of colonic intestinal epithelial cells isolated from mice (L, M). Data were compared with independent Student’s t test. Values are expressed as the mean ± SEM, and statistically significant differences are indicated. All data are representative of at least three independent experiments. ∗p < 0.05; ∗∗p < 0.01; ns, not significant.

We subsequently assessed the effect of NLRP6 on necroptosis during S. typhimurium infection. Immunohistochemical analysis showed no significant difference in phosphorylated MLKL in the ileum between WT and Nlrp6^−/− mice (Figure S5A). However, colonic p-MLKL protein levels were significantly lower in S. typhimurium -infected Nlrp6^−/− mice than in wild-type mice (Figure 5K). To test whether this phenotype was present in enterocytes, either ileal or colonic IECs were isolated from WT and Nlrp6^−/− mice with or without infection. In line with data obtained in the immunohistochemical stain assay, similar phosphorylated levels of MLKL, RIP1(S321), and RIP1(S166) were observed in the ileum of WT and Nlrp6^−/− mice, whereas the infected Nlrp6^−/− mice had a significantly lower level of MLKL phosphorylation in the colon as compared with those in infected WT mice (Figures S5B, 5L, and 5M). Meanwhile, NLRP6 deficiency resulted in an increased phosphorylated RIP1 at S321 and a decreased phosphorylated RIP1 at S166 in colonic IECs (Figures 5L and 5M), suggesting that NLRP6 facilitated necroptosis in the colon through regulating the kinase activity of RIP1. In addition, although a higher level of p-RIP1(S321) was found in the spleen of Nlrp6^−/− mice compared with those of WT mice, no significant difference was observed between the expression of p-MLKL and p-RIP1(S166) of these two genotypes during infection (Figures S6A and S6B). To confirm the role of RIP1 in NLRP6-associated intestinal bacteria dissemination and colonic necroptosis, S. typhimurium -infected mice were administrated with the RIP1 inhibitor Nec-1s. Our data showed that the significant difference in intestinal permeability, tissue bacterial burden and colonic necroptosis between the wild-type and the Nlrp6^−/− mice was abrogated by Nec-1s (Figures 6A–6F). To better understand the observed mechanisms underlying NLRP6-relevant necroptosis, p38^MAPK and MK2 phosphorylation in colonic IECs were quantified. Consistent with our in vitro observations, S. typhimurium infection led to an increase of both p38^MAPK and MK2 phosphorylation, and p-p38^MAPK and p-MK2 protein levels were significantly higher in infected Nlrp6^−/− than in WT mice (Figure 6G). Treatment of S. typhimurium -infected mice with the MK2 inhibitor abolished the significant difference in necroptosis between WT and Nlrp6^−/− colonic IECs (Figures 6I and 6J). Importantly, the significant difference in tissue bacterial burden between the wild-type and the Nlrp6^−/− mice was also reversed by PF-3644022 (Figures 6K–6M). Taken together, these in vivo data further support our results obtained in vitro, showing that NLRP6 modulates RIP1-mediated MLKL phosphorylation and necroptosis during S. typhimurium infection.

Figure 6.

NLRP6 promotes the systemic dissemination of S. typhimurium by regulating MK2/RIP1-associated necroptosis in vivo

(A–F) Streptomycin-pretreated WT and Nlrp6^−/− mice were orally infected with 1 × 10⁸ CFU of S. typhimurium strain SL1344. Mice were administrated intraperitoneally with the RIP1 inhibitor Nec-1s at 1 h prior to infection and 24 h post-infection, followed by sample collection at 2 days post-infection. Western blot analysis and relative intensity of colonic intestinal epithelial cells isolated from mice (A, B). Bacterial burden in MLN (C), liver (D) and spleen (E). Intestinal permeability (F).

(G and H) Streptomycin-pretreated WT and Nlrp6^−/− mice were orally infected with 1 × 10⁸ CFU of SL1344 and analyzed at 2 days post-infection. Western blot analysis and relative intensity of colonic intestinal epithelial cells isolated from mice.

(I–M) Streptomycin-pretreated WT and Nlrp6^−/− mice were orally infected with 1 × 10⁸ CFU of SL1344 and administrated orally with the MK2 inhibitor PF-3644022 at 4 h post-infection, followed by sample collection at 2 days post-infection. Western blot analysis and relative intensity of colonic intestinal epithelial cells isolated from mice (I, J). Bacterial burden in MLN (K), liver (L) and spleen (M). Data were compared with independent Student’s t test. Values are expressed as the mean ± SEM, and statistically significant differences are indicated. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ns, not significant.

Discussion

In the current study, we aimed to elucidate the underlying interconnection between NLRP6 and cell death during Gram-negative bacterial infection. Using well-established models of S. typhimurium infection both in vitro and in vivo, we provide what is, to our knowledge, the first report that NLRP6 has a detrimental effect on host antimicrobial response through modulating the initiation of necroptosis. Our data uncovered a novel mechanism underlying NLRP6-mediated necroptosis during S. typhimurium infection. This process is orchestrated via activation of RIP1 kinase-dependent MLKL, which resulted from a decrease in RIP1 phosphorylation at S321 induced by TAK1-mediated p38^MAPK/MK2 phosphorylation. Therefore, NLRP6 may be utilized by S. typhimurium to disrupt the intestinal barrier and for subsequent systemic dissemination.

Emerging evidence demonstrates that NLRP6 plays a multifaceted role in pathogenic organism infection. By binding with the RNA helicase DEAH (Asp-Glu-Ala-His) box helicase 15 (Dhx15), NLRP6 detects viruses and initiates type I interferon (IFN) responses in an inflammasome-independent manner.³¹ However, NLRP6 could also protect hosts against RNA virus and the intestinal parasite Cryptosporidium in an inflammasome-dependent manner.^32,33 Similar responses have been described in Gram-positive bacterial infection. By recognizing Lipoteichoic acid, a product of Gram-positive bacteria, NLRP6 formed an inflammasome component and activated macrophage function that restricts Staphylococcus aureus-induced pneumonia.⁸ To explore the role of NLRP6 in host defense Gram-negative bacterial infection, our present study established a well-known model of S. typhimurium infection and reported that NLRP6 is associated with cell death both in hematopoietic cells (macrophages) and non-hematopoietic cells (IECs and MEFs). A dramatically increased number of dead cells, along with a significantly decreased number of viable cells, was observed in WT cells, whereas NLRP6-deficient cells were highly resistant to S. typhimurium -induced cell death.

Pyroptosis is type of PCD occurring in a wide range of microbial infections.³⁴ Previous research from other laboratories has reported that several members of the NLR family can trigger an inflammasome-associated downstream signaling cascade that initiates pyroptosis and forms GSDMD-mediated membrane pores.³⁵ A recent study reported that NLRP6 deficiency also led to a decrease of pyroptosis during Gram-positive bacterial infection.¹⁰ Interestingly, our data show that NLRP6 had little effect on S. typhimurium -induced cleavage of GSDMD. Blocking GSDMD processing could not reverse the significant difference in NLRP6-induced cell death. These results indicate that NLRP6 mediated cell death via a pyroptosis-independent pathway. Therefore, we will focus our next study on necroptosis, another common type of lytic PCD that is triggered following challenge with several pathogens.

Necroptosis is induced by MLKL-mediated formation of necroptotic pores and manifests as plasma membrane rupture. It has been demonstrated that the MLKL inhibitor NSA interferes with disulfide bond formation of MLKL in humans but not in murine MLKL.³⁶ Hence, we treated human Caco-2 and HeLa cells with NSA and administered murine MEFs and BMDMs with another MLKL inhibitor GW806742X. Our data show that suppression of MLKL significantly decreased NLRP6-mediated cell death among macrophages, IECs, and MEFs. The following data further support NLRP6 functions in amplifying necroptosis in S. typhimurium -infected cells: (1) Knockdown of NLRP6 resulted in decreased phosphorylated levels of MLKL; (2) a higher percentage of dead cells was observed in HeLa cells co-expressing RIP3 and NLRP6; and (3) downregulation of RIP1 activity reversed the apparent difference in cell death caused by NLRP6. Laxman Ghimire et al. recently reported that NLRP6-mediated necroptosis occurred during Gram-positive bacterial infection; however, the underlying mechanism has not yet been reported.¹⁰ In the present study, our observations expand the understanding of NLRP6 also triggering necroptosis during Gram-negative bacterial infection through a RIP1 kinase-dependent mechanism, and further demonstrated that the RIP1 inhibitor could effectively alleviate NLRP6-induced intestinal barrier dysregulation and the spread of enteropathogenic bacteria.

It has been reported that RIP1 kinase activity could be prevented via the NF-κB signaling pathway.³⁷ Anand et al. reported that NLRP6 alone was capable of inhibiting NF-κB activity, leading to the reduction of the inflammatory response and bacterial clearance in hematopoietic cells (macrophages and neutrophils).⁹ Our data show that NLRP6 also contributed to the downregulation of IκBα degradation in non-hematopoietic cells and suppressed IL1β, TNFα, or PMA-induced NF-κB activation. We subsequently investigated component functions as a potential NLRP6 target and observed that NLRP6 downregulated NF-κB activity in TRAF6 or TRAF2 overexpressed cells but not in TAK1/TAB1, IKKβ, or p65 overexpressed cells. These results suggest that NLRP6 regulated NF-κB activation by targeting the components of TAK1. Our results further showed that NLRP6 inhibits necroptosis through downregulating the phosphorylation of TAK1 rather than protein stability. Previous studies demonstrated that the p38^MAPK/MK2 module, a signaling cascade downstream of TAK1, has an essential role in controlling necroptosis, and MK2 is the unique molecular that directly phosphorylates RIP1 at S321, a residue that suppresses RIP1 S166 auto-phosphorylation.³⁸ Our data show that NLRP6 deficiency resulted in an apparent increased phosphorylation of p38^MAPK, MK2 and RIP1 S321 both in vitro and in vivo. Importantly, downregulation of either p38^MAPK or MK2 not only blocked the inhibitory effect of NLRP6 on RIP1(S321) phosphorylation but also abrogated the apparent difference in NLRP6-relevant cell death during infection.

In summary, our data report a novel detrimental role for NLRP6 in promoting RIP1 kinase-dependent necroptosis, but not pyroptosis, during S. typhimurium infection. We found that NLRP6 per se inhibits NF-κB pathway activity by downregulating TAK1 phosphorylation and subsequently contributes to a reduction in p38^MAPK/MK2 signaling cascade downstream of TAK1. Decreased MK2 phosphorylation resulted in a decrease in its phosphorylated target residue on S321 of RIP1, thus ultimately increasing RIP1-kinase-dependent phosphorylation and activation of the necroptosis executor MLKL. Therapeutic strategies targeting NLRP6 might be advantageous for decreasing Gram-negative bacteria-induced cell death and thus controlling infection.

Limitations of the study

This study uncovered the role of NLRP6 in regulating necroptotic cell death during S. typhimurium infection. Although our observations suggest NLRP6 as a therapeutic target for decreasing bacteria-induced host cell death and controlling the spread of S. typhimurium, it remains to be seen if NLRP6 is also required for orchestrating necroptosis induced by other enteropathogenic bacterial. In addition, it is worth figuring out how NLRP6 interacts with TAK1 and selectively modulates its ubiquitination modification in future.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Anti-mouse GSDMD	Abcam	Cat# ab219800; RRID: AB_2888940
Anti-human GSDMD	Abcam	Cat# ab210070; RRID: AB_2893325
Anti-mouse phospho-MLKL	Abcam	Cat# ab196436; RRID: AB_2687465
Anti-human phospho-MLKL	Abcam	Cat# ab187091; RRID: AB_2619685
Anti-MLKL	Proteintech	Cat# 66675-1-Ig; RRID: AB_2882029
Anti-mouse phospho-RIP1(Ser166)	Affinity	Cat# AF2398; RRID: AB_2845412
Anti-human phospho-RIP1(Ser166)	Immunoway	Cat# YP1467; RRID: AB_3094761
Anti-mouse phospho-RIP1(Ser321)	Cell Signaling Technology	Cat# 38662; RRID: AB_3094762
Anti-human phospho-RIP1(Ser320)	Cell Signaling Technology	Cat# 58274; RRID: AB_2933988
Anti-RIP1	Cell Signaling Technology	Cat# 3493; RRID: AB_2305314
Anti-IκBα	Proteintech	Cat# 10268-1-AP; RRID: AB_2151423
Anti-phospho-TAK1	Cell Signaling Technology	Cat# 4536; RRID: AB_330493
Anti-TAK1	Proteintech	Cat# 12330-2-AP; RRID: AB_2140101
Anti-phospho-p38MAPK	Cell Signaling Technology	Cat# 4511; RRID: AB_2139682
Anti-p38MAPK	Cell Signaling Technology	Cat# 9212; RRID: AB_330713
Anti-phospho-MK2	Cell Signaling Technology	Cat# 3007; RRID: AB_490936
Anti-MK2	Proteintech	Cat# 13949-1-AP; RRID: AB_2877995
Anti-Ubiquitin	Cell Signaling Technology	Cat# 43124; RRID: AB_2799235
Anti-Horseradish peroxidase (HRP) Flag	Sigma-Aldrich	Cat# A8592; RRID: AB_439702
Anti-6×His	Thermo Fisher Scientific	Cat# MA1-21315; RRID: AB_557403
Anti-β-actin	Bioss	Cat# bs-0061R; RRID: AB_10855480
Anti-GAPDH	Boster	Cat# BM1623; RRID: AB_2885058
Anti-α-Tubulin	Beyotime Biotechnology	Cat# AF0001; RRID: AB_2922414
Bacterial and virus strains
S. Typhimurium strain SL1344	Yang et al.39	N/A
Chemicals, peptides, and recombinant proteins
RPMI 1640 medium	HyClone Laboratories	Cat# SH30096.01
Fetal bovine serum (FBS)	Biological Industries	Cat# 04-121-1A
Macrophage Colony Stimulating Factor (M-CSF)	PeproTech	Cat# 315-02
1% (vol/vol) penicillin-streptomycin	Beyotime Biotechnology	Cat# C0222
Dulbecco’s Modified Eagle Medium (DMEM)	HyClone Laboratories	Cat# SH30243.01
PBS	Sangon Biotech	Cat# E607008
Gentamicin	Sangon Biotech	Cat# A506614
GW806742X	MCE	Cat# HY-112292
NSA	MCE	Cat# HY-100573
GSK’872	MCE	Cat# HY-101872
zVAD	Beyotime Biotechnology	Cat# C1202
Nec-1	MCE	Cat# HY-15760
(5Z)-7-Oxozeaenol	Ape×Bio	Cat# b7443
SB239063	MCE	Cat# HY-11068
MK2-IN-1	MCE	Cat# HY-12834
Ethidium Homodimer-I (EthD-I)	US Everbright	Cat# E4052
Lipofectamine 2000	Thermo Fisher Scientific	Cat# 11668019
SytoxGreen	Invitrogen	Cat# S34860
Cycloheximide (CHX)	MCE	Cat# HY-12320
NP-40 Lysis Buffer	Beyotime Biotechnology	Cat# P0013F
Protein A/G beads	Santa Cruz Biotechnology	Cat# sc-2003
FD4	MilliporeSigma	Cat# 46944
Ac-YVAD-pNA	Beyotime Biotechnology	Cat# P9701
Ac-LEVD-pNA	Beyotime Biotechnology	Cat# P9714
PMA	MilliporeSigma	Cat# 79346
TNF-α	PeproTech	Cat# 300-01A
IL-1β	PeproTech	Cat# 200-01B
Streptomycin	Sangon Biotech	Cat# A610494
Nec-1s	Selleck Chemicals	Cat# S8641
PF-3644022	Selleck Chemicals	Cat# S8224
Tergitol	MilliporeSigma	Cat# 15s9
Salmonella-Shigella agar plates	Hangwei	Cat# M0062
Paraformaldehyde	Servicebio	Cat# G1101
Critical commercial assays
LDH Assay Kit	Beyotime Biotechnology	Cat# C0017
Luminescent Cell Viability Assay Kit	Promega	Cat# G7570
Bradford protein assay kit	Beyotime Biotechnology	Cat# P0006
Human IL-1β kit	Boster	Cat# EK0392
Mouse IL-1β kit	Boster	Cat# EK0394
Dual Luciferase reporter assay kit	Promega	Cat# E1910
Deposited data
Original western data for figures	This paper; Mendeley Data	https://data.mendeley.com/ https://doi.org/10.17632/ffpxt24yzd.1
Experimental models: Cell lines
Caco-2	Deng et al.19	N/A
NLRP6−/− Caco-2	Deng et al.19	N/A
HeLa	ATCC	N/A
WT MEF	This paper	N/A
Nlrp6−/−MEF	This paper	N/A
WT BMDM	This paper	N/A
Nlrp6−/−BMDM	This paper	N/A
HEK293T	ATCC	N/A
Experimental models: Organisms/strains
Mouse: C57BL/6 (Wt) mice	The Laboratory Animal Center of Soochow University	N/A
C57BL/6 Nlrp6−/− mice	Cambridge-Suda Genome Resource Center	N/A
Oligonucleotides
siRNA targeting sequence: Gsdmd: GGUGAACAUCGGAAAGAUUTT	GenePharma	N/A
siRNA targeting sequence: Control: UUCUCCGAACGUGUCACGUTT	GenePharma	N/A
Recombinant DNA
Plasmid: pcDNA3.1-Flag for all transient expression vectors	This paper	N/A
Flag-RIP3	This paper	N/A
Flag-TRAF2	This paper	N/A
Flag-TRAF6	This paper	N/A
Flag-TAK1	This paper	N/A
Flag-TAB1	This paper	N/A
Flag-IKKβ	This paper	N/A
Flag-p65	This paper	N/A
Flag-NLRP6	Deng et al.19	N/A
Software and algorithms
IBM SPSS statistics 22	IBM	https://www.ibm.com/
GraphPad Prism 8	GraphPad software, inc.	http://www.graphpad.com
ImageJ	National Institutes of Health	https://imagej.nih.gov/ij

Resource availability

Lead contact

Further information and requests for reagents may be directed to and will be fulfilled by the lead contact Rui Huang (hruisdm@163.com).

Materials availability

This study did not generate new unique reagents. Plasmids and cell lines are listed in the key resources table and available for use upon request to the lead contact.

Data and code availability

• •All data is available within the paper or supplemental information. Original western blot images have been deposited at Mendeley and are publicly available as of the date of publication. The DOI is listed in the key resources table. Microscopy data reported in this paper will be shared by the lead contact upon request.
• •This paper does not report original code.
• •Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

Experimental model and study participant details

Animals

Mice were bred in ventilated cages under 12-hour light/dark cycles in specific pathogen-free (SPF) conditions and received sterile water and food ad libitum. All experiments were conducted with ten- to twelve-week-old male mice in accordance with the principles and procedures outlined in the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals (NIH Guidelines) and were approved by the Animal Experimental Committee of Soochow University (Grant 2111270). The mice were euthanized by CO₂ from compressed gas cylinders and we complied with all ethical regulations. Mouse strains used and generated in this study are listed in the key resources table.

Cell culture studies

The human colon carcinoma Caco-2 cell line, the human epithelial carcinoma HeLa cell line, the human embryonic kidney HEK293T cell line, the mouse bone marrow-derived macrophages (BMDMs) and the mouse embryonic fibroblasts (MEFs) were maintained in a 5% CO2 environment at 37°C as described in the method details section below.

Bacterial strains

S. Typhimurium strain SL1344 was a gift from Professor Qian Yang (Nanjing Agricultural University, Nanjing, China) and used for all experiments in this study.

Method details

Cell preparation

Bone marrow cells isolated from WT and Nlrp6^−/− mice were cultured in complete RPMI 1640 medium (HyClone Laboratories, Logan, UT, USA) supplemented with 10% (vol/vol) heat-inactivated fetal bovine serum (FBS; Biological Industries, Kibbutz Beit-Haemek, Israel), 10 ng/ml of Macrophage Colony Stimulating Factor (M-CSF; PeproTech, Rocky Hill, NJ, USA) and 1% (vol/vol) penicillin-streptomycin (Beyotime Biotechnology, Shanghai, China). Nonadherent cells were removed on day 3 and adherent cells became mature BMDMs on day 7. WT and Nlrp6^−/− MEFs isolated from E13.5 embryos were grown in Dulbecco’s Modified Eagle Medium (DMEM; HyClone Laboratories) containing 10% (v/v) heat-inactivated FBS (Biological Industries) and 1% (v/v) penicillin-streptomycin (Beyotime Biotechnology). WT Caco-2 cells (human colon carcinoma cell line) were a gift from Professor Weiqi He (Cambridge Suda Genome Resource Center, Soochow University, Suzhou, China). NLRP6^−/− Caco-2 cells were constructed in our previous work.¹⁹ HeLa cells (human epithelial carcinoma cell line) and HEK293T cells (human embryonic kidney cell line) purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA). Caco-2, HeLa and HEK293T cell lines were grown in DMEM medium (HyClone Laboratories) containing 10% (vol/vol) heat-inactivated FBS (Biological Industries).

Bacteria preparation

On the day of experiments, overnight S. Typhimurium cultures were diluted in a ratio of 1:100 and agitated for a further another 3 h until reaching the late-logarithmic phase in Luria-Bertani (LB) medium (Hangwei, Hangzhou, China). Upon washing 3 times with sterile PBS (Sangon Biotech, Shanghai, China), bacterial numbers were estimated and adjusted based on the optical density at 600 nm absorbance and ready for the further experiments.

S. Typhimurium infection in vitro

Bacterial suspension was added to mammalian cells at a multiplicity of infection (MOI) as described in the figure legends. BMDMs were washed 3 times with sterile PBS (Sangon Biotech) at 30 min after infection and other cells were washed at 1 h post-infection to remove extracellular bacteria. Infected cells were subsequently cultured in RPMI medium or DMEM medium containing 10% (vol/vol) heat-inactivated FBS (Biological Industries) and 50 μg/ml gentamicin (Sangon Biotech) for 2 h. Afterward, cell culture medium was replaced with those containing 10% (vol/vol) FBS (Biological Industries) and 10 μg/ml gentamicin (Sangon Biotech). Samples were collected for further experiments at the indicated time points as described in the figure legends. When appropriate, cells were pretreated or treated with GW806742X (HY-112292, MCE, New Jersey, USA), NSA (HY-100573, MCE), GSK’872 (HY-101872, MCE), zVAD (C1202, Beyotime), Nec-1 (HY-15760, MCE), (5Z)-7-Oxozeaenol (b7443, Ape×Bio, Houston, TX USA), SB239063 (HY-11068, MCE) and MK2-IN-1 (HY-12834, MCE) as described in the figure legends. To assess cell membrane permeability, Ethidium Homodimer-I (EthD-I; US Everbright, Suzhou, China) was used as previously described.²⁴

Plasmid, siRNA and Transfection experiments

Plasmid encoding RIP3, TRAF2, TRAF6, TAK1, TAB1, IKKβ or p65 was successfully constructed. Plasmid encoding NLRP6 was constructed in our previous work.¹⁹ The negative control siRNA (NC siRNA) and siRNA specific for Gsdmd were purchased from GenePharma (Shanghai, China). Cells were transiently transfected with plasmid or siRNA using Lipofectamine 2000 (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer’s instructions.

Cell death and viability assays

Cells were seeded in 96-well plates and incubated with the S. Typhimurium suspension as described above. For cell death assay, lactate dehydrogenase (LDH) released in supernatant was measured with a LDH Assay Kit (C0017, Beyotime Biotechnology), and the percentage of dead cells was calculated according to the manufacturer’s instructions. SytoxGreen (S34860, Invitrogen by Thermo Fisher Scientific) was used according to the manufacturer’s instructions and the intensity was measured as previously described.⁴⁰ For cell viability experiment, intracellular ATP levels were detected using a Luminescent Cell Viability Assay Kit (G7570, Promega, Madison, WI, USA) according to the manufacturer’s instructions.

Western blot analysis

Samples were prepared as previously described.³⁹ Briefly, the cells were collected in NP-40 Lysis Buffer (P0013F, Beyotime Biotechnology) containing phosphatase inhibitor (Roche) and EDTA-free protease inhibitor (Biosharp) and incubated on a rocker with ice for 30 min, then centrifuged at 12000 g/4°C for 15 min to obtain cell lysates. Cell lysates were boiled at 100°C for 5 min in 5×SDS loading buffer. Equivalent amounts of protein extracts were separated on SDS-PAGE and electroblotted onto PVDF membranes (MilliporeSigma, Burlington, MA, USA). The membrane was blocked and then probed with the corresponding primary antibodies at 4°C overnight. After incubating with appropriated secondary antibodies at room temperature for 1 h, membranes were visualized by chemiluminescence and the intensity was calculated by ImageJ Launcher broken symmetry software program (NIH, Bethesda, MD, USA). For western blot quantification, phospho blots and total blots were normalized to the housekeeping gene GAPDH, α-Tubulin or β-actin, then look at the phospho/total ratio using the normalized values.

Protein stability assay

Overexpression and endogenous TAK1 protein degradation was analyzed by a protein stability assay. In brief, 293T cells transfected with expression vector for Flag-TAK1 and increasing doses of expression vector for His-NLRP6. MEFs were incubated with 100 μg/ml the protein synthesis inhibitor cycloheximide (CHX, HY-12320). TAK1 protein levels were analyzed by Western blot analysis.

Immunoprecipitation and ubiquitination assay

For immunoprecipitation, 293T cells transfected with indicated plasmids were harvested and lysed with NP-40 Lysis Buffer (P0013F, Beyotime Biotechnology). Protein extracts were incubated with anti-Flag immunomagnetic beads (M8823, Sigma-Aldrich) at 4°C overnight. For ubiquitination assay, MEFs were harvested and lysed with NP-40 Lysis Buffer (P0013F, Beyotime Biotechnology). The supernatants were denatured at 95°C for 5 min in the presence of 1% SDS, then diluted with lysis buffer to reduce the concentration of SDS below 0.1% and incubated with TAK1 antibody (12330-2-AP, Proteintech) plus Protein A/G beads (sc-2003, Santa Cruz Biotechnology) at 4°C overnight. The immune complexes were then analyzed by Western blot analysis.

TEER and epithelial permeability assay

WT and NLRP6^−/− Caco-2 cells were seeded in Transwell inserts (3 μm pore size, Corning, NY, USA) for up to 14–21 days. The TEER was measured using a Millicell Electrical Resistance System (Millicells Voltmeter, MilliporeSigma) to monitor the monolayer integrity. A TEER value of at least 200 Ω/cm² was used for below experiments.⁴¹ The bacterial suspension was added to the apical side of the Transwell system at the MOI of ∼100. TEER was measured every 0.5 h. For bacterial translocation, the culture medium was collected from the basal well at 3 h post-infection, and then translocated bacteria were enumerated by plating. Epithelial permeability was measured by apical-to-basal transport of FD4 (46944, MilliporeSigma). Bacteria resuspended in DMEM containing 10% (vol/vol) FBS (Biological Industries) and 5 mg/ml FD4 were added to the apical side of the Transwell system at the MOI of ∼100. At 3 h post-infection, the culture medium was collected from the basal well. The fluorescence was measured at an excitation wavelength of 485 nm and an emission wavelength of 520 nm.

Caspase substrate cleavage assay

Cells were lysed with caspase assay buffer and mixed with Ac-YVAD-pNA (P9701, Beyotime Biotechnology) or Ac-LEVD-pNA (P9714, Beyotime Biotechnology) according to the manufacturer’s instructions. OD values were read at 405nm. Total protein content was measured using a Bradford protein assay kit (P0006, Beyotime Biotechnology) according to the manufacturer’s instructions. The cleavage activity was normalized to the protein concentration of each sample.

ELISA

The production of cytokines was detected with human IL-1β kit (EK0392, Boster) and mouse IL-1β kit (EK0394, Boster) according to the manufacturer’s protocols.

Luciferase reporter assays

HEK293T cells were seeded at a density of 5 x 10⁴ cells per well in 24-well plates overnight. Cells were then transfected with the experimental reporter plasmid pNF-κB-Luc (Firefly luciferase, Clontech, Mountain View, CA, USA), pIL8-Luc and the internal control plasmid pRL-TK (Renilla luciferase, Clontech), as well as other plasmids as described in the figure legend and our previous study.⁴² When appropriate, cells were treated with PMA (79346, MilliporeSigma), TNF-α (300-01A, PeproTech) or IL-1β (200-01B, PeproTech). Enzyme activity was measured using the Dual Luciferase reporter assay kit (E1910, Promega) according to the manufacturer’s instructions.

Mouse experiments and ethics statement

WT mice (C57BL/6) were purchased from the experimental animal center of Soochow University. Nlrp6 knockout mice on the C57BL/6 genetic background was a gift from Professor Ying Xu (Cambridge-Suda Genome Resource Center, Soochow University, Suzhou, China). To establish bone marrow chimeras, WT and Nlrp6^−/− mice were lethally irradiated with 8 Gray using a cesium source. Recipient mice were reconstituted intravenously with 8 million bone marrow cells derived from healthy donor mice at 3 h post irradiation, and then kept under 0.2% neomycin sulfate treatment via the drinking water, ad libitum, for 2 weeks. 8 weeks after transplantation, the chimeric mice were ready for the further experiments.

A model of oral S. Typhimurium infection was established as previously described.⁴³ In brief, mice orally treated with 20 mg/mouse streptomycin (Sangon Biotech) for 24 h and then orally administrated with S. Typhimurium strain SL1344. When appropriate, mice were treated with the RIP1 inhibitor Nec-1s (S8641, Selleck Chemicals, Houston, TX, USA) or the MK2 inhibitor PF-3644022 (S8224, Selleck Chemicals). To determine the viable counts of S. Typhimurium, MLN, liver and spleen were harvested aseptically, weighed, and homogenized in 4°C PBS containing 0.5% (vol/vol) Tergitol (15s9, MilliporeSigma) and 0.5% (vol/vol) FBS (Biological Industries). Bacterial burden was determined in appropriate dilutions by plating on Salmonella-Shigella agar plates (M0062, Hangwei). Ileum, colon and spleen samples were fixed in 4% paraformaldehyde (Servicebio, Wuhan, China), embedded in paraffin and cut into 5-μm thick sections. The sections were then incubated with p-MLKL antibody (ab196436, Abcam) according to the manufacturer’s instructions. Photomicrographs were taken with a Nikon Eclipse Ni-U fluorescence microscope (Nikon Corporation, Tokyo, Japan) with NIS-Elements F (Nikon Corporation). Ileal and colonic IECs were isolated as previously described and western blot analysis was performed as described above.⁴⁴ To assess intestinal permeability, mice were orally gavage with FD4 (12 mg per mouse; MilliporeSigma) 4 h prior to sacrifice. Serum samples were collected, and fluorescence was measured at an excitation wavelength of 485 nm and an emission wavelength of 520 nm.

Quantification and statistical analysis

Data are presented as mean ± SEM. Statistical analysis was performed using IBM SPSS statistics 22 (Chicago, IL, USA). Normality of numerical variables was assessed and in the case of normality. Comparisons of 2 groups were analyzed using an independent Student’s t test. Values of p < 0.05 were considered statistically significant.

Published: February 28, 2024