Francisella tularensis Live Vaccine Strain (F. tularensis LVS) and Salmonella Typhimurium strains were obtained from BEI Resources Manassas, VA. The Francisella cultures were grown on Mueller-Hinton (MH) chocolate agar plates (BD Biosciences, San Jose, CA) supplemented with IsoVitaleX at 37°C with 5% CO₂ or in MH broth (BD Biosciences, San Jose, CA) supplemented with ferric pyrophosphate and IsoVitaleX (BD Biosciences, San Jose, CA) at 37°C with shaking (160rpm). Luria-Bertani broth or McConkeys agar was used for culturing S. Typhimurium. All bacterial culture stocks grown to the mid-log phase were stored at −80°C and aliquots were thawed for use in experiments. All bacteria were handled in a Bio-Safety Level-2 cabinet.

Six-to-eight-week-old wild-type and Nlrp3^−/− mice of either sex on a C57BL/6 background were purchased from Jackson Laboratory (Bar Harbor, ME). All mice were maintained in a specific pathogen-free environment in the Animal Resource Facility of New York Medical College. All animal work was conducted in accordance with the protocols approved by New York Medical College’s Institutional Animal Care and Use Committee.

Immortalized bone marrow-derived macrophages (BMDMs) from wild type, caspase1, Asc, Nlrp3, and Aim2 deficient mice on a C57BL/6 background were a kind gift from Dr. Katherine Fitzgerald, UMass Worcester, MA. These immortalized cell lines were generated using gene trap insertions and have been used in our previous studies (Dotson et al., 2013). Primary murine BMDMs were isolated from 6- to 8-week-old wild-type and Nlrp3^−/− mice on a C57BL/6 background purchased from Jackson Laboratory (Bar Harbor, ME) as previously described (Mahawar et al., 2012). These were differentiated into macrophages by culturing in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% heat-inactivated Fetal Bovine Serum (FBS), 2% l-glutamine, 1% sodium pyruvate, 1% HEPES, and 20% medium conditioned by L929 cells.

Bacterial replication in infected wild type, Asc^−/−, Caspase1^−/−, and Nlrp3^−/− macrophages was quantified by performing gentamicin protection assays as previously described (Rabadi et al., 2016; Alqahtani et al., 2018; Alharbi et al., 2019). The macrophages were infected with F. tularensis LVS at a multiplicity of infection (MOI) of 100:1 (Bacteria/cell ratio). In a similar experiment, wild-type and Nlrp3^−/− macrophages were infected with S. Typhimurium. Cells were lysed and intracellular bacteria were enumerated at 4 and 24h post-infection. Bacterial numbers were expressed as Log₁₀ CFU/ml.

For macrophage assays using MCC950, prior to treating primary BMDMs with MCC950, the metabolic state of the BMDMs was synchronized by starving the cells. To achieve this, the existing media was removed and replaced with 1ml of pre-warmed serum-free DMEM supplemented with 1μl of 10μM MCC950 dissolved in DMSO or 1μl of DMSO for vehicle control. BMDMs were incubated in this starvation media for 1h at 37°C, 5% CO₂. One hour post-incubation, the DMEM was replaced with complete BMDM media supplemented with DMSO or MCC950, and the gentamycin protection assays were performed as described above.

Western blotting analyses to detect levels of IκBα, ERK1/2, caspase1, and IL-1β were carried out as described previously (Mahawar et al., 2012; Dotson et al., 2013; Rabadi et al., 2016). Wild type, Aim2^−/−, or Nlrp3^−/− macrophages were infected with F. tularensis LVS at an MOI of 100 for 24h. Cells were lysed at this time and protein concentrations were normalized. Equal amounts of cell lysate were then resolved on 10% SDS-PAGE gels and transferred to polyvinylidene fluoride membranes and probed using antibodies specific to total and bio-active IκBα, ERK1/2, caspase 1, and IL-1β. The blots were developed and resulting bands quantified as previously described using Bio-rad Lab™ software (Dotson et al., 2013). Blots were then stripped and re-probed for β-actin to confirm the accuracy of protein loading.

Wild-type and Nlrp3^−/− mice were deeply anesthetized and infected intranasally (i.n.) with 0.6×10⁴CFU (sub-lethal dose), 1×10⁴CFU, or 1×10⁵CFU of F. tularensis LVS. Mice were observed for morbidity by weighing periodically and mortality for 21days. The results were expressed as Kaplan-Meier survival curves and statistical significance was established by the Log-rank test. To study bacterial clearance, wild-type and Nlrp3^−/− mice were infected i.n. with 1×10⁴CFU of F. tularensis LVS and killed on days three and seven post-infection. For studying in vivo role of MCC950, wild-type mice were injected IP with 200μl of MCC950 at a concentration of 10mg/kg body weight for 7days and were then challenged with 1.5×10⁴ CFUs of F. tularensis LVS intranasally. Lungs, livers, and spleens were isolated aseptically and homogenized using sterile zirconia beads, serially diluted, and plated onto MH chocolate agar plates to quantify bacterial burdens in these organs. Small portions of the organs were also fixed in 10% formalin, embedded in paraffin, sectioned, and stained with hematoxylin and eosin for histopathological evaluation and scoring as previously described (Bakshi et al., 2008). The resulting colonies were counted and bacterial burden was represented as Log₁₀ CFU/ml. Lung homogenates collected from wild-type and Nlrp3^−/− mice on days three and seven post-infection with 1×10⁴CFU of F. tularensis LVS were clarified by centrifugation and assayed for a panel of 23 cytokines using “Bio-plex Pro mouse cytokine 23-plex assay” (Biorad). Cytokine levels were expressed as picograms/ml (pg/ml).

The subcellular localization of p65 subunit of NF-κB was determined by immunofluorescence microscopy as previously described (Mahawar et al., 2012; Rabadi et al., 2016). Wild-type and Nlrp3^−/− macrophages were infected for 2h and processed for staining. At least 100 cells were counted at random to quantify the nuclear or cytosolic localization of p65.

Wild-type and Nlrp3^−/− mice infected with 1×10⁴ F. tularensis LVS were sacrificed on days 3 and 7 post-infection. Lungs, livers, and spleens were harvested aseptically from these mice and placed in excess of trizol. Organs were snap-frozen in liquid nitrogen and stored at −80°C until required. When ready, the organs frozen in trizol were thawed on ice and homogenized in 2ml of fresh trizol using a tissue-tearor. The homogenate was centrifuged at 15,000×g for 15min at 4°C to pellet the tissue debris and limit RNA degradation from tissue-resident nucleases. Total RNA was extracted from the clarified supernatants using the “Purelink RNA mini kit” (Invitrogen) following the manufacturer’s instructions. DNA was eliminated from the samples by DNAse treatment using “RNAse-Free DNAse” (Qiagen). Purified RNA was eluted in water and stored at −80°C.

Prior to RNA sequencing, total RNA extracted from mice had to be assayed for quantity and quality. RNA was quantified using Qubit RNA HS Assay kit on a Qubit fluorometer (Thermo Fisher). The quality of the RNA samples was measured in terms of RNA Integrity Number (RIN), which is a measure of sample degradation and strand breakage. RIN for samples was calculated using the 4200 Tapestation (Agilent) and samples with RIN>7 were considered intact and used for further RNA sequencing. Sequencing libraries were generated using TruSeq Stranded mRNA kit and murine RNA sequencing was performed on a Next Seq 550 (Illumina) and data were represented as transcripts per million (TPM). Transcript levels from wild-type mice were and Nlrp3^−/− mice were quantified and compared on days 3 and 7 post-infection using Salmon software (Patro et al., 2017). Salmon was also used to compare the change in transcript levels within the groups from day 3 to day 7 post-infection. Significance was established using the Wald log test and differences were considered significant at a q value <0.05. Functional annotation and clustering of genes were performed using DAVID Bioinformatics Resources 6.8 (Huang et al., 2009a,b).

All results were expressed as means±SD or mean±SEM. Statistical comparisons between the groups were made using unpaired t-tests or ANOVA with the Tukey-Kramer test. Survival results were plotted as Kaplan-Meier survival curves, and significance was calculated using the Log-rank test. Differences between the experimental groups were considered statistically significant at a p<0.05 level.

Activation of Aim2 and Nlrp3 inflammasomes results in the activation of caspase1 and cleavage of pro-IL-1β into its secreted mature form IL-1β. The cleaved bioactive forms of both caspase1 and IL-1β can be detected by western blotting. We first investigated the state of inflammasome activation in the wild type, Aim2^−/− and Nlrp3^−/− macrophages infected with F. tularensis LVS. Wild type, Aim2^−/− and Nlrp3^−/− macrophages were infected with 100 MOI of F. tularensis LVS (1×10⁷ CFUs of bacteria:1×10⁵ macrophages in a 12 well-plate). The infection was allowed to progress for 2h, and then the cells were treated with gentamicin for 1h to kill all extracellular and adherent bacteria. After 24h post-infection, the cell lysates were analyzed for bioactive levels of capsase1 and IL-1β by western blot analysis. In F. tularensis LVS-infected wild-type macrophages, consistent with previous reports (Dotson et al., 2013), we observed very low levels of bioactive caspase1 and IL-1β. The activation of both caspase1 and IL-1β was reduced significantly in the Aim2^−/− compared to the wild-type macrophages infected with F. tularensis LVS. However, significantly higher levels of bioactive caspase1 and IL-1β were observed in the Nlrp3^−/− compared to the wild-type and Aim2^−/− macrophages (Figures 1A,B and Supplementary Figure S1). We next investigated the levels of secreted IL-1β in the culture supernatants from the wild-type and Nlrp3^−/− macrophages infected with a varying multiplicity of F. tularensis LVS and treated with gentamicin to kill all the extracellular bacteria. The cell culture supernatants were collected 24h post-infection to determine the levels of IL-1β and TNF-α exclusively in response to the intracellular bacteria. A dose-dependent increase in the levels of both IL-1β and TNF-α were observed in the Nlrp3^−/− macrophages. These levels were significantly higher than those observed in culture supernatants from the wild-type macrophages (Figures 1C,D). These results indicate that activation of caspase1 and IL-1β in macrophages infected with F. tularensis LVS is partially dependent on Aim2. However, Nlrp3 is dispensable. Its loss, on the contrary, is associated with increased levels of bioactive forms of both the Caspase1 and IL-1β, and pro-inflammatory cytokines IL-1β and TNF-α in macrophages infected with F. tularensis LVS.

NF-κB and MAPK play important roles in the production of pro-inflammatory cytokines, especially TNF-α and pro-IL-1β. We next investigated if the elevated levels of both the IL-1β and TNF-α observed in the Nlrp3^−/− macrophages were due to enhanced activation of NF-κB and MAPK signaling. The wild-type and the Nlrp3^−/− macrophages were infected with 100 MOI of F. tularensis LVS and treated with gentamycin to determine the response to intracellular bacteria. The cell lysates were analyzed by western blotting for the total and phosphorylated levels of IκBα and ERK1/2 after 1, 2, and 3h of infection as a measure of NF-κB and MAPK activation. No differences in the phosphorylated-IκB (p-IκB) levels were observed between the wild-type and the Nlrp3^−/− macrophages after 1h of infection with F. tularensis LVS. However, the levels of p-IκB were significantly higher in the Nlrp3^−/− than those in the wild-type macrophages 2 and 3h post-infection. The ratio of p-IκBα to total IκBα was also higher in F. tularensis LVS-infected Nlrp3^−/− macrophages (Figures 2A,B). A similar trend was also observed for pERK1/2, with significantly elevated levels being observed at 2 and 3h post-infection in the Nlrp3^−/− compared to the F. tularensis LVS-infected wild-type macrophages (Figures 2C,D). We further confirmed the activation of NF-κB signaling by immunofluorescence microscopy to determine the translocation of p65 subunit of NF-κB to the nucleus in the wild-type and Nlrp3^−/− macrophages infected with F. tularensis LVS. Ninety-five percent of uninfected wild type or the Nlrp3^−/− macrophages revealed cytosolic localization of p65 subunit of NF-κB. About 35% of wild-type macrophages infected with the F. tularensis LVS had p65 localized to their nuclei 2h post-infection. Contrarily, in 90% of the Nlrp3^−/− macrophages infected with F. tularensis LVS, p65 was found to be localized to the cell nuclei (Figures 2E,F). Collectively, these results demonstrate that loss of Nlrp3 results in enhanced activation of NF-κβ and MAPK, which might, in turn, result in elevated levels of pro-inflammatory cytokines TNF-α and IL-1β.

We next investigated if the elevated levels of pro-inflammatory cytokines TNF-α and IL-1β in the Nlrp3^−/− macrophages infected with F. tularensis also result in an enhanced bacterial clearance. The ability of F. tularensis to survive and replicate in Nlrp3^−/− macrophages was determined by gentamicin protection assay. Wild-type and Nlrp3^−/− macrophages were infected with 100 MOI (bacteria:cell ratio) of F. tularensis LVS, and intracellular bacterial numbers were quantitated at 4- and 24-h post-infection. Nearly equal numbers of bacteria were recovered from wild-type and Nlrp3^−/− macrophages at 4h post-infection, indicating similar numbers of bacteria entered the cells. After 24h, the bacterial numbers in the Nlrp3^−/− macrophages did not increase from those observed at 4h post-infection. However, in the wild-type macrophages, the bacteria replicated and nearly 100-fold more bacteria were recovered than those from the Nlrp3^−/− macrophages (Figure 3A). We next performed LDH release assays to confirm that failure of F. tularensis LVS to replicate in the Nlrp3^−/− macrophages was not on account of enhanced cell death of the Nlrp3^−/− macrophages following infection. The extent of cell death did not differ between the wild-type and the Nlrp3^−/− macrophages infected with F. tularensis LVS when observed at 12 and 24h post-infection (Figure 3B). Wild-type and Nlrp3^−/− macrophages were infected with 100 MOI of Gram-negative intracellular pathogen, Salmonella Typhimurium, to confirm that the Nlrp3^−/− macrophages are not defective and are capable of supporting bacterial replication. S. Typhimurium replicated to near equal levels in both the wild-type and the Nlrp3^−/− macrophages (Figure 3C), indicating Nlrp3^−/− macrophages are capable of supporting bacterial replication. A similar assay was performed next using the wild type, Asc^−/−, and Caspase1^−/− macrophages to determine if these inflammasome components like Nlrp3 also impair the clearance of F. tularensis LVS. The numbers of bacteria recovered from Asc^−/− and Caspase1^−/− macrophages were significantly higher than those from the wild-type macrophages, indicating that, unlike Nlrp3, deficiency of both Asc and Caspase1 facilitate bacterial replication (Figure 3D). Like the macrophage cell lines (Figure 3A), F. tularensis LVS also failed to replicate in the primary BMDMs isolated from Nlrp3^−/− mice as compared to their wild-type counterparts (Figure 3E). Collectively, these results demonstrate that Nlrp3 specifically impairs the clearance of F. tularensis LVS, and Nlrp3-deficiency is associated with an enhanced bacterial clearance from the infected macrophages.

Our preceding in vitro assays demonstrated that Nlrp3-deficiency results in elevated levels of pro-inflammatory cytokines and enhanced bacterial clearance. We corroborated our in vitro findings by performing in vivo experiments in wild-type and Nlrp3^−/− mice. C57BL/6 wild-type and Nlrp3^−/− mice were infected intranasally with 1×10⁴CFU of F. tularensis LVS (1×LD₁₀₀). The cytokine levels were determined in the lung homogenates on days 3 and 7 post-infection. The lung homogenates from Nlrp3^−/− mice showed significantly higher levels of IL-1β, IL-6, IL-10, IL-12p70, G-CSF, GM-CSF, TNF-α, and RANTES than the wild-type mice on day 3 post-infection. The levels of IL-2, G-CSF, IFN-γ, and RANTES were significantly higher in the Nlrp3^−/− mice than the wild-type mice on day 7 post-infection (Figure 4). Levels of IL-1α, IL-4, IL-17, MCP-1, MIP-1α, or Eotaxin did not differ between wild-type and Nlrp3^−/− mice, while IL-3 and IL-5 remained below detection limits (data not shown). These results demonstrate that Nlrp3^−/− mice induce a potent cytokine response compared to the wild-type mice following F. tularensis infection, corroborating the results observed with Nlrp3^−/− macrophages.

The results thus far demonstrated that Nlrp3^−/− macrophages clear bacteria more efficiently than their wild-type counterparts and that the levels of pro-inflammatory cytokines are higher in both Nlrp3^−/− macrophages and mice following infection with F. tularensis LVS. Wild-type and Nlrp3^−/− mice were infected intranasally with 1×10⁴CFU of F. tularensis LVS. The infected mice were sacrificed on days 3 and 7 post-infection to determine the kinetics of bacterial clearance and evaluation of the histopathological lesions in the lungs, liver, and spleen. On day 3 post-infection, wild-type and Nlrp3^−/− mice had similar bacterial burdens in their lungs, livers, while the Nlrp3^−/− mice harbored significantly fewer bacteria in their spleens (Figure 5A). On day 7 post-infection, significantly fewer bacteria were recovered from the lungs and livers of the Nlrp3^−/− compared to the wild-type mice. These results corroborated findings from in vitro studies where Nlrp3^−/− macrophages cleared F. tularensis LVS better than their wild-type counterparts.

We next quantified histopathological lesions in these mice to determine the severity of tissue damage. Sections of the organs harvested from wild-type and Nlrp3^−/− mice infected with 1×10⁴CFU of F. tularensis LVS were stained with H&E and scored for histopathological lesions at 100× magnification in a blinded fashion. We observed that wild-type mice had significantly more inflammatory infiltrates, necrotic lesions, and consolidation of airspaces. In contrast, Nlrp3^−/− mice exhibited intact alveolar spaces (Figure 5B). In the livers of wild-type mice, numerous large necrotic granulomas were observed. On the contrary, the livers of Nlrp3^−/− mice displayed much smaller granulomas, evidenced by significantly lower histopathological scores (Figures 6B,C). In the spleen of wild-type mice, mild disruption of germinal centers and peri-lymphoid infiltration of red pulp was observed. The spleens of Nlrp3^−/− mice exhibited similar red pulp infiltration, but the germinal centers remained largely conserved (Figure 5B). Overall, the lesions in the lung, liver, and spleens of the Nlrp3^−/− mice were significantly less severe than wild-type mice (Figure 5C). These results suggest that Nlrp3^−/− mice clear bacteria more efficiently than the wild-type controls and exhibit less severe histopathological lesions in the lungs, liver, and spleen.

We corroborated our in vitro findings by performing in vivo experiments in C57BL/6 wild-type and Nlrp3^−/− mice. Mice were infected i.n. with 0.6×10⁴, 1×10⁴, and 1×10⁵ CFUs of F. tularensis and monitored for morbidity by measuring the body weights and for mortality by observing survival for 21days. One hundred percent of Nlrp3^−/− mice survived the infection dose of 0.6×10⁴ CFUs, while only 33% (2/6 mice) wild-type mice survived this infection dose (Figure 6A). At a higher infection dose of 1×10⁴ CFUs, all wild-type mice succumbed to infection by day 9 post-infection; however, 25% (2/8) of the Nlrp3^−/− mice survived this infection dose with an enhanced median survival time (MST) of 13days compared to 9days for the wild-type mice (Figure 6B). One hundred percent of both the wild-type and Nlrp3^−/− mice infected with 1×10⁵ CFUs of the F. tularensis LVS succumbed to infection. However, Nlrp3^−/− mice exhibited a significantly extended MST of 12days compared to 9days for the wild-type mice (Figure 6C). The infected Nlrp3^−/− mice lost body weight for up to 6days post-infection, and those that survived regained their initial body weight. A similar trend was also observed for the wild-type mice that survived the infection (Figures 6D–F). Collectively, these results demonstrate that at all infection doses of F. tularensis LVS tested, the Nlrp3^−/− mice survived better than the wild-type mice, further cementing the notion that Nlrp3 enhances the susceptibility to F. tularensis infection.

We next investigated if the detrimental effect of Nlrp3 in response to primary infection with F. tularensis is dependent on its role as an inflammasome activator. We used a small molecule inhibitor, MCC950, to selectively disrupt the Nlrp3 inflammasome without affecting its inflammasome-independent functions (Coll et al., 2019). Primary BMDMs isolated from C57BL/6 wild-type mice were seeded in DMEM containing 10μM MCC950 or an equal volume of DMSO as controls. The BMDMs were infected with 10 MOI of F. tularensis LVS and were lysed at 4 and 24h post-infection to count intracellular bacteria. No significant differences were observed in bacterial numbers recovered at 24h post-infection from MCC950 treated (7.28±0.88 Log₁₀ CFU/ml), vehicle control (6.94±0.73 Log₁₀ CFU/ml), or untreated (7.64±0.17 Log₁₀ CFU/ml) macrophages (Figure 7A). These results show that, unlike Nlrp3-deficient macrophages, the inhibition of Nlrp3 inflammasome in the wild-type macrophages does not affect bacterial clearance.

We next investigated if similar results could be replicated in vivo. Six-to-eight-week-old C57BL/6 mice were injected intraperitoneally with 10mg/kg of MCC950 for six consecutive days. These mice were then infected i.n. with 1.5×10⁴CFU of F. tularensis LVS on day 2 of the treatment. Mice injected with DMSO were used as vehicle controls and observed for morbidity and mortality. Both the MCC950 treated and control mice succumbed to the infection by day 11 post-infection. There were no differences in body weight loss patterns between the two groups of mice (Figures 7B,C). Collectively, these results demonstrate that the detrimental effects of Nlrp3 in response to F. tularensis LVS infection are independent of inflammasome-related functions.

Results from our preceding studies demonstrate a detrimental inflammasome-independent role of Nlrp3 in response to F. tularensis infection. Recent reports have suggested that in addition to the inflammasome-dependent function, Nlrp3 also serves as a transcriptional regulator (Bruchard et al., 2015; Wang et al., 2016a). We next investigated if the detrimental effects of Nlrp3 are due to its impact on the expression of various genes in mice infected with F. tularensis LVS. RNA-sequencing was performed to determine the transcript levels of genes using total RNA extracted from lungs of wild-type C57BL/6 and Nlrp3^−/− mice infected with 1×10⁴CFU of F. tularensis LVS on days 3 and 7 post-infection. Pairwise analysis of transcripts from C57BL/6 wild-type and Nlrp3^−/− mice on days 3 and 7 post-infection was performed to determine differentially expressed genes between these strains of mice.

A total of 10,512 gene transcripts were detected in wild-type and Nlrp3^−/− mice. A total of 1,426 transcripts were differentially expressed between wild-type and Nlrp3^−/− mice when analyzed based on two or more Log₂ Fold-change on days 3 and 7 post-infection (Figure 8A). Salmon analysis followed by Wald Test to establish the statistical significance (q<0.05) revealed that seven genes were upregulated in wild-type mice on day 3 post-infection, compared to Nlrp3^−/− mice. On the other hand, 12 genes were significantly upregulated in the Nlrp3^−/− mice compared to wild-type mice. On day 7 post-infection, 42 genes were significantly upregulated in the wild-type mice, while 12 genes were found to be upregulated in the Nlrp3^−/− mice (Figure 8B). Of the 19 differentially expressed genes on day 3, only five remained significantly differentially expressed on day 7 post-infection. These included Nlrp3 and Pttg1 in wild-type mice and three pseudogenes in Nlrp3^−/− mice.

We further analyzed the 42 genes that were significantly upregulated in the wild-type mice on day 7 post-infection using DAVID for functional clustering. The software correctly identified 39 of the 42 genes belonging to Mus musculus, while three genes were identified as pseudogenes. We constructed a functional annotation chart to ascribe broad functional categories to these 39 genes. The analysis was performed using a threshold count of 2 and 0.1 EASE to determine statistical significance. Thirty-three genes clustered into 19 functional annotation clusters, while six genes (Gypa, Mcmdc2, Gm4956, Gm15772, Smim4, and Slc25a34) could not be assigned to any functional cluster at the set level of stringency.

Based on UniProt Keywords, 13 genes (33.3%) fell into the cluster of signaling-related genes. Of these, 7 (54%) play a deleterious role in response to Francisella infection either by conferring anti-inflammatory phenotypes, preventing immune cell activation, or facilitating a silent entry of bacteria (Figure 8C). Nine genes clustered into the functional category of secreted proteins, and of these, four have been reported to have a deleterious role following infection with Francisella (Figure 8C). Based on the “Gene Ontology-cellular compartment-direct” (GO_CC_DIRECT) analysis, 19 genes fell into the cytosolic functional cluster, constituting 48.7% of the input genes. Among these, seven genes have detrimental outcomes to Francisella infection (Figure 8C). Cumulatively, these results indicate that Nlrp3 alone or in conjunction with unknown protein(s) affect the transcription of various genes in mice infected with F. tularensis LVS, which are known to enhance the host’s susceptibility.

Next, we analyzed the significantly upregulated genes in Nlrp3^−/− mice compared to wild-type mice. Twelve genes were found to be significantly upregulated in the Nlrp3^−/− mice on day 7 post-infection. When analyzed by DAVID, six of these were pseudogenes, and the DAVID database could not assign those to any functional clusters. The remaining six genes grouped into 12 functional clusters by a threshold count of 2 and an EASE of 0.1 (same as used for wild type mice). The remaining six genes were grouped together in the signaling cluster. Of these, four appear to have a beneficial role during F. tularensis infection, either by enhancing the innate immune signaling or by engaging in tissue remodeling to limit tissue damage in the lungs. These genes are implicated in positively affecting the JAK-STAT signaling and regulating IFN-γ production (Figure 8D). Thus, analysis of genes with increased transcript levels in Nlrp3^−/− mice reveals an increase in host-protective innate immune responses against F. tularensis. These results further establish the notion that Nlrp3 dampens the generation of a protective immune response during F. tularensis LVS infection.

Using the RNASeq data, we next determined the expression levels of C-type lectin receptors (Clrs), toll-like receptors (Tlrs), Nlrs, DNA sensors, and components of NF-κβ and MAPK-signaling pathways in wild-type and Nlrp3^−/− mice infected with 1×10⁴ CFUs of F. tularensis LVS on days 3 and 7 post-infection. The transcript levels of Clr, DEC-205 were significantly higher in the Nlrp3^−/− compared to the wild-type mice on day 7 post-infection. However, the reverse was true for mannose receptor 1, with significantly elevated transcript levels were observed in the wild type than Nlrp3^−/− mice on day 7 post-infection (Figure 9A). The transcripts of Tlr1, Tlr5, Tlr11, and Nod2, Nlrp1a, Naip2, and Nlrc4 were significantly higher in the Nlrp3^−/− than wild-type mice on day 7 post-infection (Figures 9B,C). The transcript levels of DNA sensors Aim2, cGas, and Sting were also significantly elevated in the Nlrp3^−/− mice on day 7 post-infection than those observed for F. tularensis LVS-infected wild-type mice (Figure 9D). A similarly significantly elevated transcript levels were also observed for CD40L, Rasgrp1, PI3cKG, the components of NF-κβ and MAPK-signaling pathways, respectively (Figure 9E). Collectively, these results indicate that the loss of Nlrp3 is associated with significant upregulation of several innate immune components following F. tularensis infection, which in turn allows the generation of a strong protective immune response in the Nlrp3^−/− mice than those observed for the wild-type mice.