NLRP6 inflammasome regulates colonic microbial ecology and risk for colitis

Abstract

Inflammasomes are multiprotein complexes that function as sensors of endogenous or exogenous damage-associated molecular patterns. Here, we show that deficiency of NLRP6 in mouse colonic epithelial cells results in reduced IL-18 levels and altered fecal microbiota characterized by expanded representation of the bacterial phyla Bacteroidetes (Prevotellaceae) and TM7. NLRP6 inflammasome-deficient mice were characterized by spontaneous intestinal hyperplasia, inflammatory cell recruitment, and exacerbation of chemical colitis induced by exposure to dextran sodium sulfate (DSS). Cross-fostering and cohousing experiments revealed that the colitogenic activity of this microbiota is transferable to neonatal or adult wild-type mice, leading to exacerbation of DSS colitis via induction of the cytokine, CCL5. Antibiotic treatment and electron microscopy studies further supported the role of Prevotellaceae as a key representative of this microbiota-associated phenotype. Altogether, perturbations in this inflammasome pathway, including NLRP6, ASC, caspase-1, and IL-18, may constitute a predisposing or initiating event in some cases of human IBD.

Figure 1. The increased severity of colitis in ASC-deficient mice is transmissible to cohoused wild-type mice

To induce colitis, mice were given 2% DSS in their drinking water for 7 days (A) Weight loss of ASC^−/− mice and separately housed wild-type (WT) mice. (B) ASC^−/− mice and WT mice were cohoused for 4 weeks after which DSS colitis was induced. (C–F) Weight loss (C), colonoscopy severity score at day 7 (D), and survival (F) after induction of DSS colitis of WT mice that were cohoused with (i) in-house WT mice bred for several generations in our vivarium (IH-WT) or (ii) ASC^−/− mice (designated WT(IH-WT) and WT(ASC^−/−), respectively). (E) Representative images taken during colonoscopy of mice at day 7. (G, H) Representative H&E-stained sections of colons from WT(IH-WT), WT(ASC^−/−) and ASC^−/−(WT) mice sampled on day 6 (G) and day 12 (H) after the start of DSS exposure. Legend: Epithelial ulceration (arrowheads), severe edema/inflammation (*) with large lymphoid nodules (L), retention/regeneration of crypts (arrows), and evidence of re-epithelialization/repair of the epithelium (box). Scale bars=500 μm. Data are representative for three independent experiments. Error bars represent the SEM of samples within a group. *: p<0.05 by One-way ANOVA.

Figure 2. Maternal transmission of an exacerbated DSS colitis phenotype

Newborn ASC^−/− and WT mice were swapped between their respective mothers (cross-fostered), followed by induction of acute DSS colitis at 8 weeks of age. Body weight and colonoscopy severity score were measured in ASC^−/− mice and ASC^−/− mice cross-fostered with WT mothers (CF-ASC^−/−, A,B); WT mice and WT mice cross-fostered with ASC^−/− mothers (CF-WT, C,D); WT mice cohoused with ASC^−/− or cross-fostered ASC^−/− mice for 4 weeks (E,F). Data are representative of 3 independent experiments. Error bars represent the SEM of samples within a group. *: p<0.05 by One-way ANOVA.

Figure 3. Bacterial 16S rRNA-based analysis the fecal microbiota of WT and NLRP6^−/− mice

(A–D) Unweighted UniFrac PCoA of fecal microbiota harvested from WT mice single-housed or cohoused with ASC^−/− (A), IL-18^−/− (B), NLRP6^−/− (C) or all (D) mice. Samples from mice shown in (A) and (C) were taken just prior to cohousing and 28 days later. Dashed line illustrates separation of samples along PC1. (E) Distribution of family-level phylotypes in ASC-, IL-18-, NLRP6-deficient and cohoused WT mice, compared to single-housed WT mice. The horizontal axis shows the fold representation (defined as the ratio of the percentage of samples with genera present in knockout or cohoused mice versus single-housed WT mice); the left side of the axis indicates taxa whose representation is greater in single-housed WT mice; the right denotes taxa whose representation is greater in knockout or cohoused WT mice; the origin represents equivalent recovery of taxa in both groups. The vertical axis shows the calculated p value for each taxa as defined by G-test. Open diamonds represent taxa that were found only in KO/cohoused WT or single-housed WT mice, but where recovery was assumed to be 1 to calculate fold-representation. (F) Unweighted UniFrac PCoA demonstrating presence or absence of TM7 and Prevotellaceae in each sample. Dashed lines show separation of single-housed WT and cohoused WT and knockout mice on PC1. PC2 in panels (D) and (F) shows separation of communities based on host genotype/cohousing.

Figure 4. NLRP6-deficient mice harbor a transmissible colitogenic gut microbiota

(A) Analysis of NLRP6 expression in various organs, and (B) in colonic epithelial and hematopoietic (CD45+) cells. The purity of the sorted populations in panel B was analyzed using vil1 and ptprc as markers for epithelial and hematopoietic cells, respectively. (C) Bone-marrow chimeras were generated using WT and NLRP6^−/− mice as host and bone marrow donor. NLRP6 expression in the colon was analyzed 8 weeks after bone marrow transplantation. (D) Analysis of NLRP6 protein expression was performed by immunoprecipitation using an NLRP6 antibody and lysates of primary colonic epithelial cells isolated from WT and NLRP6^−/− mice. (E, F) Representative confocal images of colonic sections analyzed for expression of NLRP6 (red) and counterstained with DAPI. (E) 40×, (F) 100×, white dotted lines were drawn to illustrate the epithelial cell boundaries (G–J) Acute DSS colitis was induced in single-housed WT mice, in WT mice cohoused for 4 weeks with NLRP6^−/− mice (WT(NLRP6^−/−), the corresponding cohoused NLRP6^−/− mice (NLRP6^−/−(WT)) and single-housed NLRP6^−/− mice (NLRP6^−/−). Weight (G), colonoscopy severity score at day 8 (H), and survival (I) of single-housed versus cohoused WT and NLRP6^−/− mice. (J) Representative H&E stained sections of colons on day 8 after initiation of DSS exposure. Legend: Edema/inflammation (*), ulceration (arrowheads) and loss of crypts (arrow). Scale bars=500 μm. Data are representative of 3 independent experiments. Error bars represent the SEM of samples within a group. *: p<0.05 by One-way ANOVA.

Figure 5. Processing of IL-18 by NLRP6 inflammasome suppresses colitogenic microbiota

WT mice were cohoused with IL-1β^−/− mice or IL-18^−/− mice for 4 weeks and colitis was subsequently induced with DSS. Comparison of weight loss (A) in single-housed WT mice and in WT mice previously cohoused with IL-1β^−/− mice (WT(IL-1β^−/−)). Weight loss (B) and colonoscopy severity score at day 7 (C) for single-housed WT mice and WT mice previously cohoused with IL-18^−/− mice (WT(IL-18^−/−)). (D–F) Representative H&E-stained sections (D) and pathologic quantitation of disease severity (E, F) of colons from single-housed WT mice and WT mice cohoused with IL18^−/− mice sampled 6 days after the start of DSS administration. Scale bars=500 μm. (G, H) IL-18 levels measured in sera (G) and colon explants (H) obtained from WT and NLRP6-deficient mice without treatment. (I) Bone-marrow chimeras were generated using both WT and NLRP6^−/− mice as host and bone marrow donor. IL-18 production by colon explants was analyzed 8 weeks after bone marrow transplantation. (J) IL-18 concentrations in the serum 5 days after induction of DSS colitis. (K, L) Bone-marrow chimeras were generated using WT and IL-18^−/− mice as host and bone marrow donor: weight (K) and colonoscopy severity scores at day 7 (L) of mice with acute DSS colitis are shown. Data in panels A–E are representative of at least 3 experiments, data in panels I–L are representative of two experiments (n=6 mice/samples analyzed per group). *: p<0.05 by One-way ANOVA.

Figure 6. Microbiota induction of CCL5

(A) Representative H&E-stained sections of the colon, terminal ileum and Peyer's patches from WT, ASC^−/− and NLRP6^−/− mice not exposed to DSS. Legend: Mucosal hyperplasia in the colon (double arrows), increased crypt to villus ratio in the terminal ileum (*), and enlargement of Peyer's patches with formation of germinal centers (arrowheads). Scale bars = 500 μm. (B) Enumeration of subsets of hematopoietic cells harvested from the lamina propria of WT and NLRP6^−/− mice. (C, D) Analysis of CCL5 colonic mRNA expression (C), and protein expression in colonic explants (D) in WT, ASC^−/−, NLRP6^−/−, and IL-18^−/− mice. (E) CCL5 expression in epithelial cells from the colons of WT and NLRP6^−/− mice. (F,G) Analysis of CCL5 colonic mRNA expression (F) and protein expression in colonic explants (G) in single-housed WT mice and WT mice cohoused with NLRP6^−/− mice. (H, I) WT and CCL5^−/− mice were either single-housed or cohoused for 4 weeks with NLRP6^−/− mice followed by exposure to DSS. Weight loss (H) and colonoscopy severity score at day 7 (I) of mice after induction of acute DSS colitis. Data shown in panels A-G are representative of at least two experiments. Data presented in panels H and I are from three experiments (n=5–6 mice). Error bars represent the SEM of samples within a group. *: p<0.05 by One-way ANOVA.

Figure 7. Decreased abundance of Prevotella in antibiotic-treated NLRP6^−/− correlates with ameliorated colitogenic microbiota

(A–C) WT and NLRP6^−/− mice were treated with a combination of metronidazole and ciprofloxacin for 3 weeks. Prevotellaceae loads compared to total bacteria (A) were measured in fecal samples at the end of the antibiotic treatment period using qPCR analysis. DSS exposure was begun 3 days later. Weight loss (B) and colonoscopy score at day 7 (C). (D–F) NLRP6^−/− mice were treated with a combination of ampicillin, neomycin, vancomycin, and metronidazole for 3 weeks and then cohoused with WT mice for 4 weeks. In parallel, WT mice were cohoused with untreated NLRP6^−/− mice. Subsequently, DSS colitis was induced and weight (D) and colonoscopic assessment of mucosal damage at day 7 (E) were recorded. (F) qPCR assay for the abundance of Prevotella in fecal samples obtained after 4 weeks of cohousing. (G, H). WT mice were cohoused for four weeks with either NLRP6^−/− or NLRC4^−/− mice. (G) Unweighted UniFrac PCoA of fecal microbiota harvested after cohousing. (H) Unweighted UniFrac PCoA colored by relative abundance of Prevotellaceae as percent of total OTUs. (I) Quantification of Prevotellaceae in the crypt compartment, following extensive removal of stool content. (J–L) Representative transmission electron microscopy images taken from colonic sections of WT (J, ×4200), NLRC4^−/− (J, ×4200), NLRP6^−/−(J, ×2500), and ASC^−/− mice (J ×1700; K ×4200; L ×26,000).