NLRP6 negatively regulates innate immunity and host defence against bacterial pathogens.

Members of the intracellular nucleotide-binding and oligomerization domain (NOD)-like receptor (NLR) family contribute to immune responses through activation of nuclear factor-κB (NF-κB), type I interferon and inflammasome signalling. Mice lacking the NLR family member NLRP6 were recently shown to be susceptible to colitis and colorectal tumorigenesis, but the role of NLRP6 in microbial infections and the nature of the inflammatory signalling pathways regulated by NLRP6 remain unclear. Here we show that Nlrp6-deficient mice are highly resistant to infection with the bacterial pathogens Listeria monocytogenes, Salmonella typhimurium and Escherichia coli. Infected Nlrp6-deficient mice had increased numbers of monocytes and neutrophils in circulation, and NLRP6 signalling in both haematopoietic and radioresistant cells contributed to increased susceptibility. Nlrp6 deficiency enhanced activation of mitogen-activated protein kinase (MAPK) and the canonical NF-κB pathway after Toll-like receptor ligation, but not cytosolic NOD1/2 ligation, in vitro. Consequently, infected Nlrp6-deficient cells produced increased levels of NF-κB- and MAPK-dependent cytokines and chemokines. Thus, our results reveal NLRP6 as a negative regulator of inflammatory signalling, and demonstrate a role for this NLR in impeding clearance of both Gram-positive and -negative bacterial pathogens.

Despite the availability of antibiotics, bacterial infections continue to threaten public health worldwide. According to CDC estimates, roughly 48 million people in the Unites States get sick every year due to foodborne illnesses resulting in 3000 deaths annually[5]. Listeria and Salmonella belong to the top three pathogens contributing to foodborne infections resulting in death[5,6]. Listeria monocytogenes and Salmonella typhimurium are facultative intracellular pathogens that have the ability to survive and replicate within macrophages and dendritic cells. Listeria rapidly escapes the phagosome to replicate in the cytosol, whereas Salmonella lacks a phagosomal escape mechanism and inhabits the salmonella-containing vacuole (SCV)[7-9]. A deeper understanding of the mechanisms by which these and other bacterial pathogens are sensed by the immune system may contribute to new approaches for developing antimicrobials. Cells of the innate immune system detect micro-organisms by means of a limited set of evolutionary conserved pattern recognition receptors (PRRs)[10]. NLR proteins represent a family of intracellular PRRs that survey the cytoplasmic compartment for infectious agents and cellular damage[1]. NLRs such as NOD1 and NOD2 contribute to host defense against microbial pathogens by inducing the production of pro-inflammatory cytokines through activation of NF-κB and MAPK signaling, whereas NLRs such as NLRP1, NLRP3 and NLRC4 facilitate the activation of inflammatory caspases in large multi-protein complexes termed inflammasomes[11,12]. Our knowledge of the former NLRs has markedly improved in recent years, but initial characterization of NLRP6 has only recently been reported[2-4,13]. These studies demonstrated NLRP6 to contribute to protection against colitis, colorectal tumorigenesis and non-alcoholic steatosis by regulating the integrity of the epithelial barrier and by altering the composition of the gut microflora[2-4,13]. However, its role in host defense against microbial pathogens and the signaling pathways regulated by NLRP6 remain unclear.

To characterize NLRP6 expression in immune cells, Nlrp6 transcript abundance in immune cells and epithelial cells was analyzed by real-time qPCR. As previously reported[14], neutrophils and T cells showed the highest expression levels for Nlrp6, followed by macrophages, epithelial cells and dendritic cells (), suggesting that NLRP6 may play an important role in these cell types. The domain architecture of NLRP6 resembles that of NLRP3 and consists of a N-terminal Pyrin domain, a centrally located nucleotide-binding domain and a stretch of C-terminally-located leucine rich repeats[1]. Nlrp6-deficient mice have been described[2-4,13], and animals appeared healthy and did not display gross abnormalities when housed in a specific-pathogen-free facility (data not shown). To characterize the role of NLRP6 during microbial infections, wildtype (WT) and Nlrp6 mice were infected intraperitoneally (i.p.) with a lethal dose of 106 CFU of Listeria monocytogenes (L. monocytogenes). Whereas the entire cohort of WT mice succumbed to infection within 5-6 days, 75% of Nlrp6 mice survived the infection and were still alive at day 20 post-infection (. In agreement, WT mice lost on average 20% of their initial body weight before succumbing to infection, whereas weight loss in Nlrp6 mice was initially limited (<10% by day 3) and body weight gradually returned to normal levels after day 4 (. To examine whether differential mortality in WT and Nlrp6 mice was associated with differences in bacterial dissemination, bacterial burdens in systemic organs was determined at days 1 and 3 post-infection. Notably, both liver and spleen of Nlrp6 mice contained significantly less bacteria than those of WT mice at day 1 (); and this difference further increased by day 3 post-infection (. Immunohistochemical analysis of liver sections confirmed L. monocytogenes to be nearly absent from liver capsules of Nlrp6 mice, whereas those of WT mice were morphologically distorted due to high pathogen counts (. L. monocytogenes infection is known to trigger histopathological lesions and the formation of inflammatory cell foci, the size of which correlates with disease severity[15]. To examine the extent of immune cell infiltration, hematoxylin and eosin (H&E)-stained liver sections of L. monocytogenes-infected WT and Nlrp6 mice were compared. In agreement with our previous results, inflammatory cell foci in Nlrp6 livers were significantly smaller than those found in WT mice at day 3 post-infection (. To examine the role of NLRP6 during alternative infection routes, cohorts of WT and Nlrp6 mice were infected with L. monocytogenes intravenously (i.v.). As with i.p.-infected animals (), bacterial counts in the liver and spleen of i.v.-infected Nlrp6 mice were significantly lower than the burdens measured in WT mice at day 3 post-infection (. Nlrp6 mice were recently shown to have an altered microflora composition that is transferable to co-housed WT mice[3]. In agreement, 16S rRNA analysis confirmed an increased abundance of Bacteroidetes (Prevotellaceae family) in the gastro-intestinal tract of Nlrp6 mice relative to the microbiota composition of separately housed WT mice (. Co-housing WT with Nlrp6 mice for 4 weeks equalized the prevalence of Prevotellaceae in WT and Nlrp6 mice (. However, co-housing did not alter the resistant phenotype of Nlrp6 mice to L. monocytogenes infection because Nlrp6 mice contained fewer bacteria in the spleen and liver compared to the levels of co-housed WT mice (). These results suggest that NLRP6 regulates L. monocytogenes infection independently of its microflora composition.

To determine whether resistance of Nlrp6 mice to infection is specific to L. monocytogenes or extends to other pathogens, cohorts of WT and Nlrp6 mice were infected i.p with Salmonella typhimurium (S. typhimurium). Resistance to S. typhimurium infection was increased in Nlrp6 mice as illustrated by the significantly lower bacterial burdens detected in the liver and spleen at days 1 () and 3 () post-infection. Inbred mice may carry distinct alleles of mouse Nramp1 (natural resistance-associated macrophage protein one), the gene products of which may differentially regulate host resistance to Salmonella and other bacterial pathogens[16]. Nramp1 transcripts of both Nlrp6 wildtype (C57BL/6J) mice encoded the NRAMP1 Gly/Asp169 polymorphism (data not shown), ruling out a possible impact on the increased resistance of Nlrp6 mice to Salmonella infection. Because both S. typhimurium and L. monocytogenes are facultative intracellular pathogens, we determined whether NLRP6 modulated resistance against extracellular pathogens such as Escherichia coli (E. coli) as well. To this end, we analyzed CFUs in systemic organs of wildtype and Nlrp6 mice 2 days after infection with this Gram-negative extracellular pathogen. Notably, bacterial burdens in the liver and spleen of Nlrp6 mice were significantly lower than in wildtype mice (). Together, these results suggest that NLRP6 may promote systemic dissemination and growth of both Gram-negative and -positive (S. typhimurium and L. monocytogenes, respectively) as well as intracellular and extracellular bacteria (S. typhimurium and E. coli, respectively) in infected hosts.

Myeloid cells play critical roles in early host defense responses against microbial pathogens. They develop in the bone marrow, but egress into circulation upon infection[17]. Because Nlrp6 is expressed in monocytes and neutrophils (), we initially made use of an automated haematology analyzer - which determines blood cell types based on morphology - to characterize the influx of monocytes and neutrophils in peripheral blood of WT and Nlrp6 mice. Prior to infection, monocyte and neutrophil precursors in the bone marrow were similar (), and no significant differences in counts of circulating monocytes and neutrophils were noted in naïve WT and Nlrp6 mice (), suggesting that basal differentiation of myeloid progenitor cells was not affected in these mice. However, both the fraction and total cell counts of circulating monocytes and neutrophils were markedly higher in peripheral blood of L. monocytogenes-infected Nlrp6 mice (). To confirm these results, immune cell infiltration in the peritoneal cavity of L. monocytogenes-infected animals was analyzed by flow cytometry. Basal monocyte and neutrophil populations in the peritoneal cavity of naïve WT and Nlrp6 mice did not differ (data not shown). In contrast, significantly higher infiltration of monocytes and neutrophils was observed in the peritoneal cavity of L. monocytogenes-infected Nlrp6 mice over WT mice (. Unlike granulocyte populations, the numbers of circulating lymphocytes and infiltrated T-cells in the peritoneal cavity were similar in L. monocytogenes-infected WT and Nlrp6 mice (. Concurrently, Nlrp6-deficiency did not alter the induction of T-cell responses because the frequency and number of OVA-specific CD8+ T-cells were similar in WT and Nlrp6 mice that were infected with OVA-expressing Listeria (Lm-OVA) (. Taken together, these results suggest NLRP6 to inhibit the influx of monocytes and neutrophils in circulation and into the peritoneum during bacterial infections.

To further examine the cell types responsible for NLRP6-mediated dissemination of bacterial infections, we created WT and Nlrp6 bone marrow chimeras. Bone marrow reconstitution was confirmed to reach a level of approximately 95% in irradiated mice (data not shown). As expected, Nlrp6 mice that were transplanted with Nlrp6 bone marrow were more resistant to L. monocytogenes infection than WT mice receiving WT bone marrow (). Notably, both WT mice transplanted with Nlrp6 bone marrow and Nlrp6 mice that received WT bone marrow showed an intermediate level of protection against L. monocytogenes (. Indeed, weight loss () and bacterial dissemination to the spleen and liver () of these two chimera groups was in between the levels seen for the two control groups (. These results suggest that both hematopoietic and non-hematopoietic cells contribute to NLRP6-mediated inhibition of bacterial clearance.

To gain understanding of the signaling mechanisms by which NLRP6 regulates bacterial infections, we first examined bacterial clearance by in vitro cultured Nlrp6 macrophages. Phagocytosis and bacterial replication of L. monocytogenes in Nlrp6 bone-marrow derived macrophages (BMDMs) were comparable to those of WT macrophages (). Unlike results from overexpression studies[14], but in agreement with a recent report that employed Nlrp6 siRNA knockdown in human PBMCs[18], we found NLRP6 to be dispensable for inflammasome activation by L. monocytogenes and S. typhimurium because neither caspase-1 processing nor maturation of IL-1β were affected in Nlrp6-deficient macrophages infected with these pathogens or stimulated with the bacterial ligands LPS and Pam3-CSK4 in the presence of ATP ().

We next assessed the potential role of NLRP6 in regulating activation of NF-κB and MAP kinase signaling in macrophages infected with L. monocytogenes. In WT BMDMs, phosphorylation of ERK1/2 and IκB peaked 60 min post-infection and gradually diminished thereafter (). Infected Nlrp6 macrophages had significantly increased phospo-ERK1/2 levels, and failed to down-regulate phospho-IκB (). In addition to macrophages, Nlrp6 neutrophils also responded to L. monocytogenes infection with elevated phospho-IκB and phospo-ERK1/2 levels (), suggesting that NLRP6 may negatively regulate NF-κB and ERK activation downstream of TLRs. In agreement, the TLR2 ligand Pam3CSK4 also induced increased phospo-ERK1/2 and phospho-IκB levels in Nlrp6 BMDMs (). Similar observations were made using the TLR4 ligand LPS (). Increased phospho-IkB and phospho-ERK1/2 levels were also noted in liver lysates of infected Nlrp6 mice (). Unlike with TLR agonists, phospho-IkB and phospho-ERK1/2 levels induced upon ligation of the cytosolic receptors NOD1 and NOD2 were similar in WT and Nlrp6 cells (), suggesting NLRP6 to be specifically involved in negatively regulating TLR-induced NF-κB and MAP kinase activation. Thus, in marked contrast to overexpression studies[14], our results in Nlrp6 cells suggest that NLRP6 may diminish rather than enhance TLR-induced activation of NF-κB and ERK signalling. To further confirm the role of NLRP6 in regulating these pathways, we measured the production of NF-κB- and MAP-kinase-dependent cytokine and chemokine levels at different time points. Abundance of TNFα, IL-6 and KC transcripts in cell lysates and secreted cytokines in culture supernatants were significantly higher in Nlrp6 macrophages infected with L. monocytogenes, or stimulated with Pam3CSK4 and LPS (). Interferon-β transcript levels nearly doubled in LPS-stimulated Nlrp6 macrophages after 2 h stimulation, but subsequently returned to levels of WT cells (. Cytotoxicity was limited to baseline levels in both WT and Nlrp6 macrophages, ruling out an important role for cell death induction in differential secretion of cytokines and chemokines (. Increased levels of IL-6 and KC were also measured in serum and the peritoneal cavity of infected Nlrp6 mice (. These results suggest that NLRP6 specifically suppresses TLR-induced NF-κB and MAP-kinase activation during L. monocytogenes infection in vitro and in vivo.

NF-κB is a family of nuclear transcriptional regulators that initiate pro-inflammatory gene expression[19]. To determine whether canonical or non-canonical NF-κB activation is altered in Nlrp6 deficient cells, we analyzed the phosphorylation status of the non-canonical NF-κB effector p100, whereas p105 phosphorylation status was monitored as a parameter of canonical NF-κB signaling. L. monocytogenes infection induced significantly higher levels of phosphorylated p105 in Nlrp6 macrophages, whereas those of phosphorylated p100 were similar in WT and Nlrp6-deficient macrophages (). Similar results were obtained when analyzing the isolated cytosolic fraction of L. monocytogenes-infected cells (). Additionally, this correlated with increased translocation of RelA in the nuclear fraction of infected Nlrp6-deficient cells, whereas the levels of the non-canonical effector RelB were similar in WT and Nlrp6 macrophages (). Confocal immunofluorescence microscopy of L. monocytogenes-infected WT and Nlrp6 cells confirmed the increased nuclear translocation of RelA in the latter genotype (). Increased p105 phosphorylation was also evident in total and cytosolic lysates of Pam3CSK4-stimulated Nlrp6 macrophages (). In addition, enhanced RelA translocation was confirmed by Western blotting in nuclear lysates and by confocal immunofluorescence staining (). qPCR analysis of IRAK-M and A20 failed to demonstrate differential expression of these negative regulators of TLR signaling in LPS-and Pam3CSK4-stimulated Nlrp6-deficient cells (. Although further analysis is required to determine the precise level at which NLRP6 negatively regulates TLR-induced responses, our results indicate that NLRP6 suppresses TLR-induced MAP kinase and canonical NF-κB signaling to dampen the production of pro-inflammatory cytokines and chemokines during bacterial infection. NLRP6 activation consequently leads to increased susceptibility to both intracellular (L. monocytogenes and S. typhimurium) and extracellular (E. coli) as well as Gram-positive (L. monocytogenes) and –negative (E. coli and S. typhimurium) bacterial pathogens. Our bone-marrow chimera experiments suggested both the hematopoietic and non-hematopoietic compartments to contribute to the increased resistance to infection of Nlrp6 mice. We further demonstrated here that NLRP6 plays a critical role in regulating host and immune responses in cells of the myeloid lineage, and further analysis is required to determine the specific non-hematopoietic cell type(s) contributing to NLRP6-mediated host defense signaling in vivo.

The detrimental role of NLRP6 during systemic infection with bacterial pathogens appears in sharp contrast with its protective role in gastro-intestinal tract, where it contributes to maintaining a healthy composition of the gut microbiota[20]. We hypothesize that these apparently contradictory roles of NLRP6 may be explained by the spatiotemporal context in which it operates. As NLRP6 appears critical for dampening the production of pro-inflammatory cytokines and chemokines downstream of TLRs, it may play a protective role under conditions where strong inflammatory responses might be detrimental to the host (as during colitis). On the other hand, NLRP6 may play an adverse role during bacterial infections and other situations in which potent inflammatory responses would be protective to the host.

Notably, previous studies characterizing the NLR family members NOD1, NOD2, NLRC4 and NLRP3 all pointed to a protective role for these NLRs in clearing microbial pathogens, and deficiency in these NLRs led to increased morbidity and mortality in infected animals[21-24]. In contrast, our data suggest NLRP6 to dampen inflammatory signaling, thereby promoting bacterial dissemination and colonization of systemic organs of the host. In light of these findings, we propose that one subclass of NLRs (including NOD1, NOD2 and NLRC4) may directly respond to microbial pathogens detected in intracellular compartments by initiating pro-inflammatory signaling that eventually contributes to bacterial clearance. A second subclass of NLR proteins, such as NLRP6, may act as a molecular switch to dampen or silence TLR–induced pathways triggered by extracellular recognition of bacterial ligands to prevent overt pathology. Regardless, our results highlight a critical role for NLRP6 in dampening host responses against several bacterial pathogens and suggest that therapeutic inhibition of NLRP6 activation may prove beneficial for treating infectious diseases.

Methods Summary

Nlrp6 knock-out mice were generated as described previously[2-4,13]. In vivo infection experiments were done on female mice that were 8-12 weeks old and were conducted under protocols approved by St. Jude Children's Research Hospital Committee on Use and Care of Animals. Lethal L. monocytogenes infection was established by infecting female mice with 1 × 106 CFU bacteria administered intraperitoneally (i.p). To study bacterial clearance, mice were infected with L. monocytogenes either at a dose of 3 × 105 CFU (i.p.). Immune cell infiltration in the circulation of infected mice was performed on blood collected by submandibular venipuncture and analyzed on a Forcyte hematology analyzer. Bone-marrow chimera experiment was done as described previously. Bone-marrow macrophages were prepared and infected as described previously. Samples were run on SDS-PAGE and immunoblotted with antibodies as reported previously. Data are expressed as mean ± s.e.m. Differences were analyzed by Student't t-test. P values ≤ 0.05 were considered significant.

METHODS

Bacterial infections and enumeration of bacterial burdens

Lethal L. monocytogenes infection was established by infecting female mice with 1 × 106 CFU bacteria administered intraperitoneally (i.p). Animals were weighed and monitored daily for mortality for up to 20 days. Differences in survival were analyzed by Mantel-Cox test. To study bacterial clearance, mice were infected with L. monocytogenes either at a dose of 3 × 105 CFU (i.p.) or at 3 × 104 CFU (i.v.)[7]. Alternatively, mice were infected with S. typhimurium at a dose of 1 × 104 CFU (i.p). E. Coli infection was performed in 6 weeks old female mice with 5 × 107 CFU (i.p.). Mice were sacrificed at days 1 and 3 post-infection (or day 2 in case of E. Coli infection), liver and spleen tissue was homogenized and bacterial burdens were enumerated by serial dilution on Brain-Heart-Infusion agar or LB agar plates as described previously[7].

Histopathology

Formalin-preserved liver sections were processed and embedded in paraffin by standard techniques. Longitudinal sections of 5 μm thick were stained with hematoxylin and eosin (H&E) and examined by a pathologist blinded to the experimental groups. For immunohistochemistry, formalin-fixed paraffin-embedded tissues were cut into 4 μm sections and slides were stained with anti-Listeria antibody.

Analysis of immune cell infiltration in the peritoneum and in circulation

WT and Nlrp6 female mice were infected with 3 × 105 L. monocytogenes (i.p.). At 3h post-infection, blood was collected by submandibular venipuncture and analyzed on a Forcyte hematology analyzer (Oxford Sciences) by a hematologist blinded to the experimental groups. In addition, cells were collected by peritoneal lavage in either 1m of IMDM (for cytokines analysis) or in 10 ml IMDM (for total cell recruitment to the cavity) on ice. Cells obtained from WT and Nlrp6 mice were washed twice with PBS and surface-stained for anti-CD11b (Biolegend; clone # M1/70), anti-Gr-1 (Biolegend; clone # RB6-8C5), anti-TCR β (Biolegend; clone # H57-597) and anti-Ly6G (BD Biosciences). Monocytes were characterized as CD11b+Gr-1-TCRb- cells, whereas CD11b+Gr1+ cells were taken as neutrophils.

WT and Nlrp6 mice were intravenously infected with 3×104 colony forming units (CFUs) of Listeria monocytogenes that was engineered to express the OVA antigen (LmOVA). At the peak of the T cell response (day 8), 100 μl of peripheral blood was obtained and peripheral blood leukocytes (PBLs) were isolated following red blood cell lysis. Cells were FACs stained with OVA-tetramers and T cell lineage markers (anti-CD4 and anti-CD8).

Bone-marrow chimeras

Bone marrow transfer was used to create Nlrp6 chimera mice wherein the genetic deficiency of Nlrp6 was contained to either the circulating cells (Nlrp6 > WT chimera) or non-hematopoietic tissue (WT > Nlrp6). Briefly, bone-marrow cells were collected from the femur and tibia of congenic WT (expressing CD45.1 leukocyte antigen) or Nlrp6 (expressing CD45.2 leukocyte antigen) donor mice by flushing with HBSS. After thorough washing, cells were resuspended in PBS and 100ul of this cell suspension containing 107 cells was injected retro-orbitally to the irradiated host mice. Four chimera groups were generated (WT > WT; WT > Nlrp6; Nlrp6 > WT and Nlrp6 > Nlrp6). The use of CD45.1 expressing congenic WT mice facilitated the verification of bone-marrow reconstitution at 6 weeks after irradiation. Cells in the blood were labelled with APC-conjugated CD45.1 and FITC-conjugated CD45.2 antibody. The chimera mice had ~95% reconstitution with the donor bone-marrow.

Macrophage culture and in vitro stimulation

BMDMs and peritoneal macrophages were prepared as described previously[7]. Cells were subsequently infected with L. monocytogenes or S. typhimurium as indicated; or stimulated with Pam3-CSK4 (10 μg/ml) or LPS (100 ng/ml). At the indicated time points, supernatant was collected and stored at -80°C for cytokine analysis at a later time and cells were lysed in RIPA buffer supplemented with complete protease inhibitor cocktail (Roche) and PhosSTOP (Roche). For analysis of caspase-1 activation and IL-1β secretion, macrophages were cultured with bacteria or TLR ligands for 3 h followed by addition of 5 mM ATP (Sigma) for 30 min.

In vivo lysates and serum cytokines

Liver was obtained from control and infected WT and Nlrp6 mice at 3h post-infection. The tissue was homogenized in RIPA buffer supplemented with complete protease inhibitor cocktail (Roche) and PhosSTOP (Roche). At this time, blood was also collected by submandibular venipuncture for serum cytokine analysis.

ELISA

Cytokine ELISA was performed according to manufacturer's instructions (Milliplex).

LDH assay

LDH assay was performed by using Promega cytotoxicity kit according to manufacturer's instructions.

Western blotting

Lysates were prepared in RIPA buffer supplemented with complete protease inhibitor cocktail (Roche) and PhosSTOP (Roche). Samples were resolved by SDS-PAGE and transferred to PVDF membranes by electroblotting. Membranes were blocked in 5% nonfat milk and incubated overnight with primary antibodies at 4°C and subsequently with HRP-tagged secondary antibodies at room temperature for 1h. Immunoreactive proteins were visualized with ECL method (Pierce). Semi-quantitative analysis of WB bands was performed by Image J program.

Confocal Immunofluorescence microscopy

Cells were grown overnight on glass coverslips. Next day, WT and Nlrp6 macrophages were either infected with GFP-labeled L. monocytogenes or stimulated with Pam3. The cells were fixed with 4% paraformaldehyde and processed for microscopy with anti-RelA antibody (Santa-Cruz) as described previously[7]. Cells, grown on coverslips, were mounted on glass slides with ProLong gold antifade reagent (Invitrogen) and visualized with a Nikon C1 confocal microscope using 40X objective lens. The images were processed and analyzed with Image J program.

Real-time PCR

Total RNA was isolated with Trizol (Invitrogen) according to the manufacturer's instructions. For Nlrp6 expression, macrophages and dendritic cells were differentiated from bone-marrow cells, neutrophils were obtained from the peritoneal cavity after 5h of thioglycollate injection, T-cells were purified from the spleen and epithelial cells were isolated from the intestine. 1 μg of RNA was reverse-transcribed to cDNA with poly dT primers using First-strand cDNA synthesis kit (Invitrogen). Transcript levels of different genes were analyzed using SYBR-Green (Applied Biosystems) on an ABI7500 real-time PCR instrument. Expression of GAPDH was used as an internal control. mRNA expression levels were normalized against a standard curve.

Co-housing experiments and analysis of fecal flora

Age and gender matched WT and Nlrp6 mice were coused together at 1:1 ratio for four weeks as described previously[3]. Fecal samples were obtained from the two groups of mice before and after co-housing. Frozen fecal samples were processed for DNA isolation using QiaAmp stool kit (Qiagen). 10ng of DNA was used in the PCR reaction and previously reported primer pairs were used for analysis of 16S rRNA analysis[3].

Intracellular growth curve

WT and Nlrp6 macrophages were infected with L. monocytogenes with a moi of 1:1 and intracellular growth curves were generated as described previously[25].

Nramp1 genotyping

A fragment of the Nramp1 gene was amplified by PCR using primers 5′-GCCATTGTGGGCTCAGATATGC-3′ and 5′-GCCATTATGGTAATGAGGAGACCG. The resulting PCR fragment was purifed and sequenced using the primer 5′-GCCATTGTGGGCTCAGATATGC-3′.

Statistics

Differences in the survival of animals were calculated by log-rank Mantel-Cox test using Graphpad Prism software. In all other experiments, p values were calculated using the two-tailed Student's t-test. p-values are denoted by *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001.