C57BL/6 (wild-type, WT) mice were purchased from Jackson Laboratories (Bar Harbor, ME) whereas NLRP6 KO mice were obtained from Millennium Pharmaceuticals (Cambridge, MA). All these mice were on a C57BL/6 background and used at 8-10 weeks old. Animals were kept in specific pathogen free environments with access to food and water. All experiments were performed in accordance with the Guide for the Care and Use of Laboratory Animals of the National Institute of Health. This study was approved by Institutional Animal Care and Use Committee (IACUC) at Louisiana State University and Agricultural and Mechanical College in Baton Rouge, LA.

De-identified human septic and non-septic spleen blocks (septic and non-septic) were kindly provided by Dr. Richard S Hotchkiss, Washington University School of Medicine at St. Louis, MO.

Polymicrobial sepsis was induced in mice using CLP as described in our previous publication (13). For bacterial sepsis, we injected E. coli at the dose rate of 5 X 10⁸ CFU/kg of mouse intra-peritoneally and observed survival for 12 days. Recombinant IL-18 was administered immediately after CLP (5 ug/mouse, IP) (24). CD4 or CD8 T cells were depleted by treating each mouse with 200 µg of anti-CD4 or anti-CD8 antibody (BioXcell, NH) 12 and 2 hours prior to CLP.

After specific time points, mice were euthanized. The peritoneal cavity was washed by injecting 7 ml of PBS containing heparin and dextrose. The peritoneal lavage fluid (PF) was collected in a sterile conical tube. A total of 6ml of lavage fluid was collected from each mouse. The total and differential leukocyte count was performed in the peritoneal fluid as described in our previous publication (14).

Flow cytometric staining was performed as recommended by the manufacturer’s protocol (BioLegend, San Diego, CA). The single cell suspensions obtained from septic WT and NLRP6 KO mice were stained with antibody against CD3ϵ (145-2C11), CD4 (GK1.5), CD8α (53-6.7), and PI (Thermo Fisher Scientific, Waltham, MA). The stained cells were fixed with 2% paraformaldehyde and analyzed using either a FACs Caliber or LSRFortessa X20 (BD). Appropriate isotype antibodies were used for each color. The data obtained were analyzed using FlowJo_V10 (Treestar).

Parafilm-embedded splenic tissue blocks from septic and non-septic patients were provided by Dr. Richard S. Hotchkiss, Washington University School of Medicine at St. Louis, MO. Immunofluorescence microscopy was performed in spleen tissue sections from human patients and mice as described in our previous publications (14, 25, 26). The following primary antibodies were used: human NLRP6, CD3, CD4, CD8 and mouse NLRP6, CD3, CD4, and CD8. After overnight incubation with these primary antibodies, the excess antibodies were washed off with cold PBS. The slides were then stained with appropriate secondary antibodies against human and mouse antigens for 30 minutes at room temperature. The image was taken under florescent microscope (Zeiss Axioskop 2 Plus) microscope.

Cytokines were measured using standard ELISA procedure as described in the manufacturer’s protocol (eBioscience or ThermoFisher scientific, MA).

Measurement of cell death was performed using Cytox-One Homogenous Membrane Integrity Assay Kit (Promega, WI) following the manufacturer’s protocol. LDH release was expressed in relative fluorescence units (RFU).

Data are presented as Mean ± SEM. Unpaired two-tailed student’s t test was used to compare the data between two groups. One-way ANOVA followed by Tukey’s multiple comparisons test was used whenever there were more than two groups. Statistical analysis was accomplished by GraphPad Prism 8.0 software (San Diego, CA). Survival analysis was performed using log-rank test. The experiments are representative of 2-3 independent experiments. A p value less than 0.05 was considered significant.

Our previous investigation demonstrated that NLRP6 serves as a negative regulator of host defense for Staphylococcus aureus-induced lung infection¹² whereas it acts as a positive regulator of host immunity for pulmonary K. pneumoniae infection (22). Next, to investigate the role of NLRP6 in sepsis, we induced moderate sepsis in NLRP6 KO and WT mice through CLP. The KO mice had better survival (80% vs 40%) compared to that of WT mice ( Figure 1A ). To examine whether the resistant phenotype in the KO mice is model specific, we infected KO and WT mice with a sub-lethal dose of E. coli intraperitoneally to induce bacterial sepsis. Consistent with the CLP model, the KO mice had improved survival compared to that of WT mice during E coli-induced sepsis ( Figure 1B ). Furthermore, we measured bacterial burden in multiple tissues, including spleen, kidney, and PF following sepsis. The KO mice had less bacterial burden in the PF, spleen, and kidney compared to that of WT mice at both 12- and 24-hours post-sepsis ( Figures 1C-E ). In previous studies, the NLRP6 inflammasome has been implicated in regulating gut microbiota composition (18, 20). To determine whether differences in gut microbiota can modulate the outcome of sepsis, we co-housed WT and KO mice for 4 weeks and induced sepsis. The co-housed KO mice still had less bacterial burden in the PF, spleen, and kidney compared to that of WT mice, indicating that the phenotype observed in the KO mice is not microbiota dependent ( Figures 1F-H ).

Sepsis is characterized by an initial hyper-inflammatory phase followed by a prolonged hypo-inflammatory or immunosuppressive phase (27, 28). Both of these phases are implicated in death of septic patients. Therefore, we determined whether NLRP6 initiates hyperinflammation to enhance mortality during sepsis. To this end, we induced sepsis through CLP and measured inflammatory cell recruitment in the PF at 24 hrs. Interestingly, WT mice had more total WBCs and neutrophils recruited in the peritoneal cavity compared to that of KO mice after sepsis ( Figure 2A ). In addition, we measured major proinflammatory cytokines in the cell free PF obtained from septic mice. The WT mice had increased levels of IL-1β, TNF-α, MCP-1, and CXCL1 compared to that of KO mice ( Figures 2B-E ). Reduction in IL-1β in the PF of KO mice compared to WT mice indicates that the NLRP6 inflammasome is activated during sepsis. Moreover, IL-10, which is considered predictive of fatal outcome in sepsis (29, 30), was also more abundant in WT compared to that of KO at 24 hours post-CLP ( Figure 2F ).

The spleen is an important organ which plays a critical role in combating infections during sepsis. To explore whether NLRP6 expression is upregulated in the spleen of septic patients, we obtained spleen tissue sections from septic and non-septic patients and performed immunofluorescence microscopy to determine the upregulation of NLRP6 in these tissue sections. NLRP6 (Red) was upregulated in CD3, CD4, and CD8 T cells (Green) of septic patients compared to that of non-septic patients ( Figure 3A ). Consistent with this finding, NLRP6 (Red) was upregulated in splenic CD3, CD4, and CD8 T cells (Green) of CLP-induced septic mice compared to sham-operated mice ( Figure 3B ).

Sepsis causes extensive loss of T-cells in the spleen that leads to immunosuppression and mortality (27, 31, 32). According to these observations, we examined whether the NLRP6 inflammasome regulates lymphocyte death during sepsis. For this, we first measured the number of T cells in the spleen after inducing sepsis and found that the WT mice had less CD3 positive lymphocytes compared to that of KO counterparts ( Figures 4A, B ). The normal CD4 to CD8 T cells ratio (2:1) observed in sham animals was reduced to 1:1 in both WT and KO mice after sepsis, suggesting that sepsis alters the proportion of CD4 and CD8 T cells in the spleen ( Figures 4C, D ). It is important to note that a similar observation was reported from the blood of septic patients (33). The number of CD4 T cells decreased after sepsis in both WT and KO mice; however, this reduction was less pronounced in the KO mice ( Figures 4E, F ). In contrast, the number of CD8 T cells increased in both WT and KO mice after sepsis; however, the increase was more pronounced in KO compared to that of WT counterparts ( Figures 4G, H ).

The reduced numbers of lymphocytes observed in the spleens of WT mice could be due to increased cell death. To this end, we induced sepsis in WT and KO mice and measured the extent of cell death in the spleen through flow cytometry. WT mice had more PI⁺ T cells compared to that of KO mice ( Figures 4I, J ). In addition, the PF obtained from septic KO mice displayed less LDH release compared to that of WT counterparts, indicating that NLRP6 enhances cell death during sepsis ( Figure 4K ).

It has been reported that the loss and dysfunction of T cell responses contribute to an immunosuppressive or hyporesponsive state of leukocytes during sepsis (27, 34). Because we found higher T cell depletion in WT compared to KO mice counterparts, we sought to determine their role in bacterial clearance in sepsis. In this context, we depleted T lymphocytes using specific antibodies against CD3, CD4, and CD8 cells. Administration of anti-CD3 antibody markedly increased the bacterial burden in NLRP6 KO mice ( Figures 5A, B ). Although depletion of CD4 T cells did not affect bacterial clearance during sepsis ( Figures 5C, D ), the depletion of CD8 T cells lead to increased bacterial burden in the PF and spleen in the KO mice ( Figures 5E, F ). In order to determine the efficiency of Ab depletion, 12 hours after antibody administration, we obtained blood samples for flow cytometry to determine the expression of CD4+, CD8+, and CD3e+ T cells. Our findings show that these antibodies are extremely efficient in depleting the specific T cell populations (data not shown).

Cytokines, such as IL-18 and IL-1β have been implicated in aggravating septic conditions (35–40). In addition, IL-1β and IL-18 are the two critical cytokines activated/cleaved by inflammasomes, including NLRP6 (15, 35). Therefore, we measured levels of IL-18 and IL-1β in PF of WT and KO mice and found that these cytokines were less abundant in KO mice compared to that of WT mice ( Figures 2B , 6A ). Moreover, we hypothesized that reduced IL-18 and/or IL-1β in the KO mice could account for higher survival in these mice. To test this hypothesis, we administered recombinant IL-1β or IL-18 to the KO mice immediately after CLP and monitored their survival. IL-18 administration reversed the survival advantage observed in the KO mice following sepsis. However, no difference in the survival was observed in KO mice upon IL-β treatment, indicating that NLRP6 increases susceptibility via IL-18, but not through IL-1β ( Figure 6B ). In addition, recombinant IL-18 administration immediately following CLP augmented bacterial burden in PF and spleen ( Figure 6C ) as well as enhanced WBC and neutrophil recruitment to PF ( Figure 6D ).

Finally, we evaluated the mechanisms by which IL-18 enhances mortality in NLRP6 KO mice following sepsis. Since NLRP6 KO mice displayed attenuated inflammation and less T cell death, we hypothesized that IL-18 could contribute to hyper-inflammation and high lymphocyte loss in the WT mice ultimately leading to mortality. To this end, we administered recombinant IL-18 (rIL-18) to the KO mice immediately after sepsis and determined the degree of inflammation. Mice that received rIL-18 had higher levels of inflammatory cytokines, such as IL-6, TNF-α, and MCP-1 compared to the groups that did not receive IL-18 ( Figures 7A-C ). In addition, we measured the extent of cell death by determining the levels of cell death alarmins (LDH and IL-1α) (14) in the PF after administration of rIL-18. Interestingly, the KO group that received rIL-18 showed higher levels of alarmins in the peritoneal lavage fluid suggesting higher cell death compared with the KO mice group that received PBS ( Figures 7D, E ). Since the KO mice had less cell death, we measured the T cell population in the spleen after induction of sepsis. Intriguingly, the KO mice that received rIL-18 had less CD3⁺ T cells in the spleen compared to the group that received PBS only ( Figures 7F, G ).