Visual Abstract

Intestinal barrier is essential for dietary products and microbiota compartmentalization and therefore gut homeostasis. When this barrier is broken, cecal content overflows into the peritoneal cavity, leading to local and systemic robust inflammatory response, characterizing peritonitis and sepsis. It has been shown that IL-1β contributes with inflammatory storm during peritonitis and sepsis and its inhibition has beneficial effects to the host. Therefore, we investigated the mechanisms underlying IL-1β secretion using a widely adopted murine model of experimental peritonitis. The combined injection of sterile cecal content (SCC) and the gut commensal bacteria Bacteroides fragilis leads to IL-1β–dependent peritonitis, which was mitigated in mice deficient in NLRP3 (nucleotide-binding domain, leucine-rich-containing family, pyrin domain-containing-3) inflammasome components. Typically acting as a damage signal, SCC, but not B. fragilis, activates canonical pathway of NLRP3 promoting IL-1β secretion in vitro and in vivo. Strikingly, absence of fiber in the SCC drastically reduces IL-1β production, whereas high-fiber SCC conversely increases this response in an NLRP3-dependent manner. In addition, NLRP3 was also required for IL-1β production induced by purified dietary fiber in primed macrophages. Extending to the in vivo context, IL-1β–dependent peritonitis was worsened in mice injected with B. fragilis and high-fiber SCC, whereas zero-fiber SCC ameliorates the pathology. Corroborating with the proinflammatory role of dietary fiber, IL-1R–deficient mice were protected from peritonitis induced by B. fragilis and particulate bran. Overall, our study highlights a function, previously unknown, for dietary fibers in fueling peritonitis through NLRP3 activation and IL-1β secretion outside the gut.

Bacterial peritonitis and abdominal sepsis occur when intestinal integrity is disrupted, leading to escape of colonic content, which includes bacteria from microbiota and dietary products into the peritoneal cavity. If not properly treated, the infection can reach bloodstream, leading to bacteremia and sepsis. It is well known that resident peritoneal cells such as macrophages and mesothelial cells deflagrate immune response by recognizing microbiota products through innate receptors, such as TLR and NLR. The hallmark of the peritonitis is the release of many inflammatory mediators, including cytokines (e.g., IL-1β and TNF-α) and chemokines (e.g., IL-8 and MCP-1), which in turn recruit neutrophils and monocytes, fueling abscess development (1, 2). Although the contribution of the bacteria is evident, the role of dietary products in triggering the inflammation is unknown.

Despite being in low numbers in the human gut and present several metabolic and immunological benefits to the host, Bacteroides fragilis is the bacteria more frequently isolated from patients with anaerobic sepsis. Deficiency on IL-10 gene drastically increases mortality in murine model of B. fragilis–induced peritonitis as a result of exacerbated inflammation mediated by cytokines, such as IL-1β, not uncontrolled bacterial spread (3). Among inflammasomes already described, NLRP3 (nucleotide-binding domain, leucine-rich-containing family, pyrin domain-containing-3) is the best characterized and presents a wider range of different stimuli. Its activation results in molecular assembly of the adaptor molecule apoptosis-associated speck-like protein containing a caspase activation and recruitment domain (ASC) and caspase-1 to the NLRP3 oligomer, leading to caspase-1 proteolytic activation. In addition to IL-1β and IL-18 cleavage, assembly of the NLRP3 inflammasome leads to gasdermin D–mediated pyroptotic cell death. It is well established that NLRP3 is activated by a broad range of microbial, environmental, or self-derived molecules (4), including particulate materials, such as uric acid crystals, silica, asbestos fibers, cholesterol crystals, and fibrillar protein amyloid-β (5). Once activated by these insoluble components, NLRP3 initiates IL-1β–dependent inflammation, hence contributing to tissue damage and pathogenesis in gout, silicosis, asbestosis, atherosclerosis, and Alzheimer disease, respectively (611). Despite the importance of IL-1β in many inflammatory conditions, including colitis and sepsis, the contribution of inflammasome pathway in the peritonitis induced by B. fragilis, the gut commensal bacteria usually involved in clinical peritoneal infections, has not been explored until now.

Using a widely adopted murine model of peritonitis induced by inoculating B. fragilis in combination with sterile cecal content (SCC), a potentializing agent that acts as an adjuvant (1, 2), the current study mechanistically uncovers the involvement of the IL-1 pathway in the development of peritonitis. Our results reveal the key role of particulate dietary fiber, defined in this study as a second signal leading to IL-1β formation via activation of the NLRP3 inflammasome.

C57BL/6 (wild-type [WT]) and C57BL/6-derived deficient mice lineages Asc–/–, Casp1/11–/– Nlrp3–/–, Myd88–/–, ll-18r–/–, Il-1r–/–, Casp11–/–, and P2x 7–/– were maintained under specific pathogen–free conditions in the animal facilities of the Universidade Federal do Rio de Janeiro (UFRJ; Rio de Janeiro, Brazil). Asc–/–, Casp1/11–/–, and Nlrp3–/– mice were provided by Dr. D. S. Zamboni (Faculdade de Medicina de Ribeirão Preto, Universidade de São Paulo, Ribeirão Preto, Brazil). P2x7–/– and Casp11–/– mice were provided by Dr. R. C. Silva (Instituto de Biofísica Carlos Chagas Filho, UFRJ, Rio de Janeiro, Brazil). Il-18r–/–, Myd88–/– and Il-1r–/– mice were provided by Dr. M. Bellio (Instituto de Microbiologia Paulo de Góes, UFRJ, Rio de Janeiro, Brazil). Experiments using germ-free (GF) mice, Gpr43–/– and Gpr109a–/– mice were performed at Dr. Charles Mackay Lab (Monash University, Melbourne, VIC, Australia). Male mice 8–12 wk old were used. All mouse experiments were conducted according to the guidelines of the Animal Care and Use Committee of the Federal University of Rio de Janeiro (Comitê de Ética do Centro de Ciências da Saúde/UFRJ, license: IBCCF132).

B. fragilis 638R strain were routinely grown in brain heart infusion (BHI) agar supplemented with menadione (10 μg/ml) and hemin (0.5 μg/ml) and in prereduced and anaerobically sterilized brain heart infusion broth. Cultures were grown at 37°C in an anaerobic chamber with an atmosphere of 80% N2, 10% CO2, and 10% H2.

A murine model of peritoneal abscess formation was adapted from Nulsen et al. (12). In brief, mice were injected i.p. with B. fragilis (1 × 108 CFU/100 μl/animal) mixed with SCC (100 μl, 1:1 v/v). Alternatively, bran (10%, v/v) was used in place of SCC. Seven days later, animals were examined at necropsy and for the presence of one or more abscesses within the peritoneal cavity, and the abscess score was determined using number and size of intra-abdominal abscess as parameters. Alternatively, mice were euthanized at 4 or 24 h after injection, and peritoneal lavage was collected for cellularity determination and IL-1β quantification. At 24-h time point, peritoneal cells were stained with fluorescently conjugated mAbs against Ly-6G (clone 1A8; BioLegend) and CD11b (clone M1/70; BioLegend). The data were acquired using an FACSCalibur (BD Biosciences). Analysis of viable cells using a gate strategy in forward scatter × side scatter dot plot was adopted. The frequency of neutrophils (CD11b+Ly-6Ghi cells) was determined, and absolute number was measured using total cellularity values in the peritoneal cavity.

Peritoneal macrophages were obtained by peritoneal lavage with cold PBS. Peritoneal macrophages were seeded on tissue culture plates with RPMI 1640 media, 10% FBS, 100 U/ml penicillin–streptomycin, and 100 U/ml pyruvate and cultivated at 37°C, 5% CO2. For differentiated macrophages (bone marrow–derived macrophages [BMDM]), bone marrow cells were collected from femurs and tibiae. Both ends of the bones were cut, and bone marrow cells were extracted by flushing with DMEM containing 10% FBS using a syringe with a 27-gauge needle. Cell suspension was filtered through a 75-μm cell strainer and then centrifuged and resuspended in RBC lysis buffer (150 mM NH4Cl, 10 mM KCO3, and 0.1 mM EDTA [pH 7.4]) for 10 min on ice. Cells were washed in complete medium and finally resuspended at a density of 4 × 106 cells/ml in RPMI 1640, 20% FBS, 30% L-929 cell–conditioned media, 100 U/ml penicillin–streptomycin, and 100 U/ml pyruvate at 37°C, 5% CO2. At day 3 of culture, 10 ml of medium were added on plates. On seventh day of culture, BMDM were collected and seeded on tissue culture plates and kept in RPMI 1640 media 10% FBS, 5% L-929 cell–conditioned media, 100 U/ml penicillin–streptomycin, and 100 U/ml pyruvate. For dendritic cells differentiation, bone marrow cells were cultured with 10 ml of RPMI 1640, 10% FBS, 100 U/ml penicillin–streptomycin, 100 U/ml pyruvate, 0.05 mM 2-ME, and 25 ng/ml GM-CSF at 37°C, 5% CO2. At day 3 of culture, 10 ml of medium with 15 ng/ml GM-CSF was added on bottles. On seventh day of culture, dendritic cell–enriched cultures were collected and seeded on tissue culture plates and kept in RPMI 1640 media 10% FBS, 100 U/ml penicillin–streptomycin, and 100 U/ml pyruvate. All cells were used 24 h after.

Diets with different fiber contents were used. High fiber content (15%) diet (HF diet) (SF11-029), zero fiber content (0%) diet (ZF diet) (SF11-028), and normal chow (control diet) (AIN93G) were purchased from Specialty Feeds (Glen Forest, WA, Australia).

C57BL/6 mice were euthanized, and then the abdomen was opened, cecum was isolated, and the contents were squeezed into petri dish. Following peptone yeast glucose medium that was added in a proportion 1:1 (v/v), the material was filtered with gauze with two layers and sterilized in autoclave. The SCC was stored in freezer (–20°C) before use.

Cells were lysed in ice-cold cell-lytic solution containing 1% of a complete protease and phosphatase inhibitor mixture (Sigma-Aldrich). Mature IL-1β in the culture supernatant (SN) (300 μl) was precipitated by 20% TCA. Protein samples (20–50 μg of protein) were separated by 12% SDS-PAGE and transferred to Immobilon polyvinylidene difluoride membranes by electroblotting and then probed with Abs against IL-1β (no. I3767; Sigma-Aldrich, St. Louis, MO). Protein bands were visualized with the use of the West Pico Super Signal chemiluminescent substrate (Thermo Fisher Scientific) and exposed to a negative film, developed, and fixed. The same membranes were stripped and reprobed using β-actin as an internal control (ctr).

The levels of IL-1β and TNF-α in cell SN were quantified using ELISA kits (R&D Systems, Minneapolis, MN) according to the manufacturers’ protocols.

C57BL/6 specific pathogen–free mice were fed with either a conventional diet (control) or ZF diet for 2 wk. After this, four pellets of the stool were diluted in 3 ml of saline solution and vigorously vortexed for 5 min. The suspension was filtered through 70-μM filter, and 200 μl of this material was transferred by oral gavage to GF mice kept on a normal chow diet. This procedure was repeated 7 d later. At the end of 14 d, SCC were obtained from both groups of recipient reconstituted GF mice (SCC:GF reconstituted with microbiota from conventional diet–fed mice [ctr-GF] and SCC:GF reconstituted with microbiota from ZF-fed mice [ZF-GF]), as well as specific pathogen–free donor mice (SCC:ctr and SCC:ZF).

SCC was diluted 10 times in PBS. From this solution, 10 μl was placed on a microscope slide and analyzed in an inverted microscope (INV-100; BEL) with attached camera (IS500; BEL). The photos were taken in the ×20 objective.

BMDM or peritoneal macrophages from C57BL/6 mice were left untreated (medium) or stimulated with LPS (500 ng/ml) for 1 h. Commercially available purified dietary fiber was added at concentration of 2.5% and 5% (v/v) and after 6 h cells were stained with a caspase-1 FLICA probe (FAM-YVAD-FMK), as recommended by the manufacturer (Immunochemistry Technologies). The events were acquired on an FACSCalibur (BD Biosciences) (excites at 492 nm and emits at 520 nm). Viable cells using a gating strategy in forward scatter × side scatter dot plot was adopted.

The data were plotted and statistically analyzed using GraphPad Prism version 5 for Macintosh (GraphPad Software, San Diego, CA; https://www.graphpad.com). The statistical significance was calculated using a two-tailed Student t test or ANOVA with posttest Bonferroni. Differences were considered statistically significant when the p value <0.05.

Using a widely adopted mouse model of anaerobic peritonitis in which B. fragilis is injected i.p. in combination with SCC, we asked whether IL-1 signaling contributes to the development of peritoneal abscesses. Analysis at day 7 postinfection (p.i.) revealed that abscess formation (Fig. 1A) and, to a lesser extent, weight loss (Fig. 1C, 1E) was markedly reduced in mice deficient in IL-1R or MyD88 as compared with the C57BL/6 WT strain, but not in Il18r–/– mice (Fig. 1A, 1D). Extending this analysis to later stages (120 d p.i.), we verified that IL-1R–deficient mice recovered from the early lesions, similarly to WT mice, whether assessed by survival rates (Fig. 1G) or frequency of the remaining abscesses (data not shown). Because IL-1β secretion depends on inflammasome activation, we next evaluated the role of the adapter protein Asc and the caspase-1 in this model. Mirroring the resistant phenotype of MyD88 and IL-1R–deficient strains, we found that Asc–/– and double-knockout Caspase-1/11–/– mice preserved their body weight and displayed reduced abscess score at 7 d p.i. (Fig. 1B, 1F). Contrasting with the high inflammatory score of WT mice, Asc–/– and Caspase 1/11–/– mice showed reduced cellular infiltration in the peritoneal cavity at 24 h p.i. (Fig. 1H), suggesting that the protective phenotype in the acute phase of infection is linked to a reduced presence of neutrophils Ly-6Ghi (Fig. 1I). Based on these initial results, we concluded that a mechanism coupling MyD88 and the ASC/inflammasome/IL-1 pathway worsens acute peritonitis induced by B. fragilis and SCC while being, in the long-term, dispensable for host survival.

FIGURE 1.

Absence of axis ASC–caspase-1–IL-1R, but not IL-18R signaling, protects mice from peritonitis induced by B. fragilis plus SCC. (A–F) C57BL/6 (n = 17–20), Il1r1–/–(n = 14), Il18r–/– (n = 11), Myd88–/– (n = 8), Asc–/– (n = 8), Caspase-1/11–/– (n = 14), and Il1r1–/ (n = 12) mice were inoculated i.p. with B. fragilis + SCC (100 μl 1 × 108 CFU B. fragilis + 100 μl SCC; v/v). Inflammation score (abscess development) was determined after 7 d; each symbol represents one animal (A and B). In (C)–(F), monitoring of body weight; the results are shown as mean ± SD. Data are representative of the sum of three independent experiments. Significance was calculated with two-way ANOVA with Bonferroni posttest. #p < 0.05, ##p < 0.01, ###p < 0.001 compared with C57BL/6. *p < 0.05, as comparison between Il1r1–/– and Il18r–/– mice. (G) C57BL/6 (n = 8) and Il1r1–/–(n = 8) mice were inoculated i.p. with B. fragilis + SCC (100 μl 1 × 108 CFU B. fragilis + 100 μl SCC; v/v). Survival curve was determined after 120 d. Data are representative of two independent experiments. (H and I) C57BL/6 (n = 5), Asc–/– (n = 4), Caspase-1/11–/– (n = 4), and Il1r1–/ (n = 5) mice were inoculated i.p. with B. fragilis + SCC (100 μl 1 × 108 CFU B. fragilis + 100 μl SCC; v/v). After 24 h, total number of cells (H) and number of neutrophils (CD11b+Ly-6Ghi cells) (I) in the peritoneal lavage were determined by flow cytometry. Data are representative of two independent experiments; the results are shown as mean ± SD. Significance was calculated by two-way ANOVA with Bonferroni posttest. *p < 0.05, ***p < 0.001, compared with control (C57BL/6 sham). #p < 0.05, ##p < 0.01 compared with C57BL/6.

FIGURE 1.

Absence of axis ASC–caspase-1–IL-1R, but not IL-18R signaling, protects mice from peritonitis induced by B. fragilis plus SCC. (A–F) C57BL/6 (n = 17–20), Il1r1–/–(n = 14), Il18r–/– (n = 11), Myd88–/– (n = 8), Asc–/– (n = 8), Caspase-1/11–/– (n = 14), and Il1r1–/ (n = 12) mice were inoculated i.p. with B. fragilis + SCC (100 μl 1 × 108 CFU B. fragilis + 100 μl SCC; v/v). Inflammation score (abscess development) was determined after 7 d; each symbol represents one animal (A and B). In (C)–(F), monitoring of body weight; the results are shown as mean ± SD. Data are representative of the sum of three independent experiments. Significance was calculated with two-way ANOVA with Bonferroni posttest. #p < 0.05, ##p < 0.01, ###p < 0.001 compared with C57BL/6. *p < 0.05, as comparison between Il1r1–/– and Il18r–/– mice. (G) C57BL/6 (n = 8) and Il1r1–/–(n = 8) mice were inoculated i.p. with B. fragilis + SCC (100 μl 1 × 108 CFU B. fragilis + 100 μl SCC; v/v). Survival curve was determined after 120 d. Data are representative of two independent experiments. (H and I) C57BL/6 (n = 5), Asc–/– (n = 4), Caspase-1/11–/– (n = 4), and Il1r1–/ (n = 5) mice were inoculated i.p. with B. fragilis + SCC (100 μl 1 × 108 CFU B. fragilis + 100 μl SCC; v/v). After 24 h, total number of cells (H) and number of neutrophils (CD11b+Ly-6Ghi cells) (I) in the peritoneal lavage were determined by flow cytometry. Data are representative of two independent experiments; the results are shown as mean ± SD. Significance was calculated by two-way ANOVA with Bonferroni posttest. *p < 0.05, ***p < 0.001, compared with control (C57BL/6 sham). #p < 0.05, ##p < 0.01 compared with C57BL/6.

Close modal

Next, we sought to discriminate the contribution of bacteria versus SCC in the inflammasome-mediated pathogenic outcome. Although bacteria alone did not induce IL-1β in the peritoneal cavity, injection of SCC strongly stimulated IL-1β secretion after 4 (Fig. 2A) and 24 h (Fig. 2C). Accordingly, intense cellular infiltration was only observed in response to SCC (Fig. 2B, 2D). Consistent with our in vivo observations, SCC induced IL-1β secretion by peritoneal macrophages, BMDM, and bone marrow–derived dendritic cell–enriched culture (BMDC) in a dose-dependent manner (Fig. 2E, 2I–L). In contrast, the symbiont B. fragilis failed to induce mature IL-1β production (Fig. 2F). Of note, Western blotting revealed that B. fragilis upregulated the expression of pro–IL-1 (Fig. 2G) and promoted TNF-α secretion (Fig. 2H). The results suggest that proteolytic processing does not occur in the absence of SCC but does after addition of exogenous ATP (Fig. 2E–G), a well-known trigger of the NLRP3 inflammasome (13).

FIGURE 2.

SCC, but not B. fragilis, induces IL-1β secretion in vivo and in vitro. (AD) C57BL/6 mice were i.p. injected with PBS, B. fragilis (1 × 108 UFC, in 100 μl), SCC (100 μl), or the combination of B. fragilis + SCC (v/v). After 4 h (A and B) or 24 h (C and D), the peritoneal lavage was obtained for determination of total number of cells (B and D) and IL-1β in the SN by ELISA (A and C). Data are representative of two independent experiments; results are expressed as mean ± SD (n = 4 mice). Significance was calculated with unpaired t test. *p < 0.05, **p < 0.01, ***p < 0.001 compared with control. #p < 0.05, ##p < 0.01, ###p < 0.001 compared with B. fragilis alone. (E) Resident peritoneal macrophages from C57BL/6 mice were left untreated (medium) or stimulated with LPS (500 ng/ml), B. fragilis (multiplicity of infection [MOI]: 50), and/or SCC (5% v/v) for 6 h at 37°C. In the last 30 min of incubation, ATP (5 mM) was added when indicated, and the SN were collected for IL-1β quantification by ELISA. Data are representative of two independent experiments (n = 2); data are expressed as mean of triplicate wells. (FH) BMDM from C57BL/6 mice were left untreated (medium), stimulated with B. fragilis (MOI: 50) or with LPS (500 ng/ml) [in (H) only] for 6 h at 37°C. In the last 30 min of incubation, ATP (5mM) was added when indicated and the SN were collected for IL-1β (F) or TNF-α (H) quantification by ELISA. Data are representative of three independent experiments (n = 3); data are expressed as mean of triplicate wells. (IL) BMDM (I and J) and dendritic cell-enriched culture (BMDC) (K and L) from C57BL/6 mice were left untreated or stimulated with SCC (% v/v) for 6 h at 37°C and the SN were collected for IL-1β quantification by ELISA (I and K). Data are representative of at least three independent experiments; data are expressed as mean of triplicate wells. In parallel, Western blotting analysis of pro–IL-1β (31 kDa) in cell lysates and bioactive IL-1β (17 kDa) in SN of stimulated cells is shown (G, J, and L).

FIGURE 2.

SCC, but not B. fragilis, induces IL-1β secretion in vivo and in vitro. (AD) C57BL/6 mice were i.p. injected with PBS, B. fragilis (1 × 108 UFC, in 100 μl), SCC (100 μl), or the combination of B. fragilis + SCC (v/v). After 4 h (A and B) or 24 h (C and D), the peritoneal lavage was obtained for determination of total number of cells (B and D) and IL-1β in the SN by ELISA (A and C). Data are representative of two independent experiments; results are expressed as mean ± SD (n = 4 mice). Significance was calculated with unpaired t test. *p < 0.05, **p < 0.01, ***p < 0.001 compared with control. #p < 0.05, ##p < 0.01, ###p < 0.001 compared with B. fragilis alone. (E) Resident peritoneal macrophages from C57BL/6 mice were left untreated (medium) or stimulated with LPS (500 ng/ml), B. fragilis (multiplicity of infection [MOI]: 50), and/or SCC (5% v/v) for 6 h at 37°C. In the last 30 min of incubation, ATP (5 mM) was added when indicated, and the SN were collected for IL-1β quantification by ELISA. Data are representative of two independent experiments (n = 2); data are expressed as mean of triplicate wells. (FH) BMDM from C57BL/6 mice were left untreated (medium), stimulated with B. fragilis (MOI: 50) or with LPS (500 ng/ml) [in (H) only] for 6 h at 37°C. In the last 30 min of incubation, ATP (5mM) was added when indicated and the SN were collected for IL-1β (F) or TNF-α (H) quantification by ELISA. Data are representative of three independent experiments (n = 3); data are expressed as mean of triplicate wells. (IL) BMDM (I and J) and dendritic cell-enriched culture (BMDC) (K and L) from C57BL/6 mice were left untreated or stimulated with SCC (% v/v) for 6 h at 37°C and the SN were collected for IL-1β quantification by ELISA (I and K). Data are representative of at least three independent experiments; data are expressed as mean of triplicate wells. In parallel, Western blotting analysis of pro–IL-1β (31 kDa) in cell lysates and bioactive IL-1β (17 kDa) in SN of stimulated cells is shown (G, J, and L).

Close modal

To understand how SCC promotes IL-1β secretion and to establish the correlation with peritonitis development, we used ASC- and caspase-1/11–deficient cells. The absence of these components completely abolishes the IL-1β production induced by SCC (Fig. 3A), whereas TNF-α production is preserved (Fig. 3B). In the same line, caspase-1 inhibitor Z-YVAD abrogates IL-1β production but has only a slight effect on TNF-α levels induced by SCC (Fig. 3C, 3D).

FIGURE 3.

IL-1β production induced by SCC depends on canonical inflammasome pathway. (A and B) Macrophages (BMDM) from C57BL/6, Asc–/–, and Caspase-1/11–/– mice were left untreated or stimulated with SCC (% v/v) for 6 h at 37°C. After stimulation, the SN were collected for IL-1β (A) and TNF-α (B) quantification by ELISA. Data are representative of at least three independent experiments; data are expressed as mean ± SD of triplicate wells. (C and D) Macrophages (BMDM) from C57BL/6 mice were left untreated (medium) or pretreated for 1 h with caspase-1 inhibitor Z-YVAD (20 μM) and stimulated with SCC (5% v/v) for 6 h at 37°C. The SN were collected for IL-1β (C) and TNF-α (D) quantification by ELISA. Data are representative of two independent experiments; data are expressed as mean of triplicate wells. (E and F) Macrophages (BMDM) (E) and dendritic cell–enriched cultures (BMDC) (F) from WT and Caspase-11–/– mice were left untreated (medium) or stimulated with 5% v/v SCC. After 6 h, IL-1β was quantified in the SN. Data are representative of at least two independent experiments; the results are shown as mean of triplicate wells. Error bars indicate ± SD.

FIGURE 3.

IL-1β production induced by SCC depends on canonical inflammasome pathway. (A and B) Macrophages (BMDM) from C57BL/6, Asc–/–, and Caspase-1/11–/– mice were left untreated or stimulated with SCC (% v/v) for 6 h at 37°C. After stimulation, the SN were collected for IL-1β (A) and TNF-α (B) quantification by ELISA. Data are representative of at least three independent experiments; data are expressed as mean ± SD of triplicate wells. (C and D) Macrophages (BMDM) from C57BL/6 mice were left untreated (medium) or pretreated for 1 h with caspase-1 inhibitor Z-YVAD (20 μM) and stimulated with SCC (5% v/v) for 6 h at 37°C. The SN were collected for IL-1β (C) and TNF-α (D) quantification by ELISA. Data are representative of two independent experiments; data are expressed as mean of triplicate wells. (E and F) Macrophages (BMDM) (E) and dendritic cell–enriched cultures (BMDC) (F) from WT and Caspase-11–/– mice were left untreated (medium) or stimulated with 5% v/v SCC. After 6 h, IL-1β was quantified in the SN. Data are representative of at least two independent experiments; the results are shown as mean of triplicate wells. Error bars indicate ± SD.

Close modal

In this study, we used mice double deficient for caspase-1 and caspase-11. Caspase-11 is a proinflammatory caspase involved on pyroptosis and IL-1α production, as well as noncanonical inflammasome activation (14). To examine whether caspase-11 is involved in the IL-1β secretion induced by SCC, we used macrophages and dendritic cell–enriched culture from mice displaying single deficiency for this inflammatory caspase. Our results showed that the IL-1β response to side scatter was equivalent to those of WT-derived cells (Fig. 3E, 3F), indicating that caspase-11 is dispensable.

NLRP3 has been implicated in the pathogenesis of a wide array of inflammatory diseases. Next, we asked whether the NLRP3 inflammasome is required for IL-1β secretion induced by SCC. As shown in Fig. 4A, this response was marginal in Nlrp3–/– macrophages, as observed for Asc–/– and Caspase1/11–/– cells. These results are reminiscent of the phenotype of B. fragilis–primed macrophages stimulated by alum (Fig. 4B), a well characterized particulate activator of NLRP3 (15, 16). Similar results were obtained with dendritic cell–enriched cultures (i.e., deficiency of NLRP3 led to abrogation of IL-1β production), whereas the levels of TNF-α were preserved (Fig. 4C, 4D). Consistent with the in vitro phenotypes described above, in vivo studies performed with Nlrp3–/– mice confirmed that SCC contributes to B. fragilis–dependent abscess formation by activating inflammasome, thereby worsening the severity of peritonitis (Fig. 4E, 4F). Collectively, these results suggest that SCC, composed of dead bacteria from microbiota and diet products, can activate the canonical NLRP3 pathway, leading to IL-1β secretion and contributing with the inflammatory response that fuels abdominal sepsis and peritonitis.

FIGURE 4.

NLRP3 is required for IL-1β production and contributes to peritonitis development. BMDM (A and B) and BMDC (C and D) from C57BL/6, Nlrp3–/–, Asc–/–, Caspase-1/11–/–, and Il1r1-/- mice were left untreated or stimulated with different concentrations of SCC (% v/v) or with B. fragilis (multiplicity of infection [MOI]: 50) plus alum (500 μg/ml) (B) for 6 h at 37°C. The SN were collected for IL-1β (A–C) and TNF-α (D) quantification by ELISA. Data are representative of three independent experiments; the results are shown as mean of triplicate wells. (E and F) C57BL/6 (n = 15), Nlrp3–/– (n = 11), Caspase-1/11–/– (n = 5), and Il1r1–/ (n = 10) mice were inoculated i.p. with B. fragilis + SCC (100 μl 1 × 108 CFU B. fragilis + 100 μl SCC; v/v). (E) Inflammation score (abscess development) was determined after 7 d; each symbol represents one animal. Data are representative of the sum of three independent experiments. In (F), monitoring of body weight. The results are shown as mean ± SD. Data are representative of three independent experiments. Significance was calculated by two-way ANOVA with Bonferroni posttest. #p < 0.05, ##p < 0.01, ###p < 0.001 compared with C57BL/6 mice.

FIGURE 4.

NLRP3 is required for IL-1β production and contributes to peritonitis development. BMDM (A and B) and BMDC (C and D) from C57BL/6, Nlrp3–/–, Asc–/–, Caspase-1/11–/–, and Il1r1-/- mice were left untreated or stimulated with different concentrations of SCC (% v/v) or with B. fragilis (multiplicity of infection [MOI]: 50) plus alum (500 μg/ml) (B) for 6 h at 37°C. The SN were collected for IL-1β (A–C) and TNF-α (D) quantification by ELISA. Data are representative of three independent experiments; the results are shown as mean of triplicate wells. (E and F) C57BL/6 (n = 15), Nlrp3–/– (n = 11), Caspase-1/11–/– (n = 5), and Il1r1–/ (n = 10) mice were inoculated i.p. with B. fragilis + SCC (100 μl 1 × 108 CFU B. fragilis + 100 μl SCC; v/v). (E) Inflammation score (abscess development) was determined after 7 d; each symbol represents one animal. Data are representative of the sum of three independent experiments. In (F), monitoring of body weight. The results are shown as mean ± SD. Data are representative of three independent experiments. Significance was calculated by two-way ANOVA with Bonferroni posttest. #p < 0.05, ##p < 0.01, ###p < 0.001 compared with C57BL/6 mice.

Close modal

Extracellular ATP signaling through purinergic receptor P2X7 is a classical mechanism of NLRP3 inflammasome activation, leading to IL-1β secretion in many inflammatory settings, including bacterial infection (17, 18). We found that IL-1β production by P2X7-deficient macrophages or dendritic cell–enriched cultures stimulated with SCC was partially inhibited as compared with their WT counterparts (Fig. 5A, 5B). Noteworthy, however, the production of TNF-α was not impaired by P2X7–/– cells stimulated by SCC (Supplemental Fig. 1A, 1B). Adding weight to these data, in vivo experiments showed that P2x7–/– mice were not protected from acute peritonitis induced by B. fragilis and SCC (Fig. 5C, 5D).

FIGURE 5.

Mechanisms underlying IL-1β production induced by SCC. (A and B) BMDM (A) or BMDC (B) from C57BL/6 and P2x7–/– mice were left untreated (medium) or stimulated with SCC (% v/v), LPS (500 ng/ml), or LPS + alum (500 μg/ml) for 6 h at 37°C. When indicated, ATP (5 mM) was added in the last 30 min, and the SN were collected for IL-1β quantification by ELISA. Data are representative of three independent experiments; the results are shown as mean of triplicate wells. (C and D) C57BL/6 (n = 9) and P2x7–/– (n = 11) mice were inoculated i.p. with B. fragilis + SCC (100 μl 108 CFU B. fragilis + 100 μL SCC; v/v). (C) Monitoring of body weight and (D) inflammation score (abscess development) after 7 d. Data are representative of the sum of two independent experiments. In (C), the results are shown as mean ± SD. In (D), each symbol represents one animal. Significance was calculated by two-way ANOVA with Bonferroni posttest. (E and F) BMDM from C57BL/6 mice were left untreated or pretreated with cytochalasin D (20 μM), Ca-74Me (50 μM), MitoTEMPO (100 nM), or KCl (20 mM) for 1 h. After this time, macrophages were stimulated with SCC (2% v/v) for 6 h at 37°C. The SN were collected for IL-1β and TNF-α quantification by ELISA. Data are representative of two independent experiments; the results are shown as mean of triplicate wells.

FIGURE 5.

Mechanisms underlying IL-1β production induced by SCC. (A and B) BMDM (A) or BMDC (B) from C57BL/6 and P2x7–/– mice were left untreated (medium) or stimulated with SCC (% v/v), LPS (500 ng/ml), or LPS + alum (500 μg/ml) for 6 h at 37°C. When indicated, ATP (5 mM) was added in the last 30 min, and the SN were collected for IL-1β quantification by ELISA. Data are representative of three independent experiments; the results are shown as mean of triplicate wells. (C and D) C57BL/6 (n = 9) and P2x7–/– (n = 11) mice were inoculated i.p. with B. fragilis + SCC (100 μl 108 CFU B. fragilis + 100 μL SCC; v/v). (C) Monitoring of body weight and (D) inflammation score (abscess development) after 7 d. Data are representative of the sum of two independent experiments. In (C), the results are shown as mean ± SD. In (D), each symbol represents one animal. Significance was calculated by two-way ANOVA with Bonferroni posttest. (E and F) BMDM from C57BL/6 mice were left untreated or pretreated with cytochalasin D (20 μM), Ca-74Me (50 μM), MitoTEMPO (100 nM), or KCl (20 mM) for 1 h. After this time, macrophages were stimulated with SCC (2% v/v) for 6 h at 37°C. The SN were collected for IL-1β and TNF-α quantification by ELISA. Data are representative of two independent experiments; the results are shown as mean of triplicate wells.

Close modal

Lysosomal destabilization associated with potassium efflux, as well as generation of reactive oxygen species, are common cellular alteration associated with phagocytosis of sterile particulate matter, such as silica and alum, leading to NLRP3 activation (19, 20). We observed that IL-1β secretion was either abolished in macrophages pretreated with high concentration of KCl (to block potassium efflux); cytochalasin D, an inhibitor of actin polymerization; CA74-Me, an inhibitor of lysosomal (tiol) cathepsins (21); and MitoTEMPO, an inhibitor of mitochondrial reactive oxygen species (Fig. 5E). In contrast, TNF-α levels were preserved or only marginally reduced (Fig. 5F). Based on those results, we next evaluated whether a particulate component present in the colonic content could induce NLRP3 activation in macrophages. To this end, we submitted SCC to mild centrifugation with the aim to test the effects induced by soluble versus precipitated samples, either in vitro or in vivo. As shown in Fig. 6, soluble SCC (referred to in this article as SCC SN) failed to induce IL-1β production by macrophages and dendritic cell–enriched cultures (Fig. 6A, 6B). Consistent with the in vitro data, injection of B. fragilis combined to soluble SCC did not induce abscesses in the peritoneal cavity (Fig. 6C).

FIGURE 6.

Particulate matter present in SCC is essential for IL-1β secretion and abscess development. (A and B) BMDM (A) or BMDC (B) from C57BL/6 mice were left untreated (medium) or stimulated for 6 h at 37°C with SCC (1% v/v) or with SCC SN obtained after centrifugation (150 g; 1 min). The cytokine IL-1β was quantified by ELISA. Data are representative of two independent experiments; the results are shown as mean of triplicate wells. (C) C57BL/6 mice were inoculated i.p. with B. fragilis (100 μl 1 × 108 CFU) in combination with 100 μl SCC or SCC SN. Inflammation score was determined after 7 d. Data are representative of four independent experiments. Each symbol represents one animal; significance was calculated with unpaired t test. ***p < 0.001. (D) SCC from mice fed on control diet (SCC), ZF diet and HF diet were diluted 10 times in PBS and placed on a microscope slide and analyzed in the ×20 objective. Data are representative two independent experiments.

FIGURE 6.

Particulate matter present in SCC is essential for IL-1β secretion and abscess development. (A and B) BMDM (A) or BMDC (B) from C57BL/6 mice were left untreated (medium) or stimulated for 6 h at 37°C with SCC (1% v/v) or with SCC SN obtained after centrifugation (150 g; 1 min). The cytokine IL-1β was quantified by ELISA. Data are representative of two independent experiments; the results are shown as mean of triplicate wells. (C) C57BL/6 mice were inoculated i.p. with B. fragilis (100 μl 1 × 108 CFU) in combination with 100 μl SCC or SCC SN. Inflammation score was determined after 7 d. Data are representative of four independent experiments. Each symbol represents one animal; significance was calculated with unpaired t test. ***p < 0.001. (D) SCC from mice fed on control diet (SCC), ZF diet and HF diet were diluted 10 times in PBS and placed on a microscope slide and analyzed in the ×20 objective. Data are representative two independent experiments.

Close modal

SCC is composed of dead bacterial and diet products, including undigestible carbohydrates, which are turned into particulate fiber in the gut. In fact, we can visualize particulate fibers present in the SCC (×10 diluted) in the optical microscope (Fig. 6D). To better identification of these fibers, C57BL/6 mice were either fed on HF diet (three times more fiber than control; 15%) or absence of fiber. Two weeks later, SCC were obtained from each group, and fiber could be visualized in the first, but not in the last, sample, with sizes varying between 5 and 30 μM (Fig. 6D). Collectively, our data support the hypothesis that NLRP3/IL-1β–driven peritonitis depends on phagocytosis of a particulate component of SCC, possibly dietary fiber.

To verify the potential impact of dietary fiber on inflammasome activation, C57BL/6 mice were either fed on conventional (SCC ctr; 5%) diet, HF diet (SCC HF; 15%), or absence of fiber (ZF, SCC ZF; 0%) for 2 wk, and SCC were obtained. As shown in Fig. 7, SCC generated from mice subjected to a diet free of fiber induced reduced levels of IL-1β production by macrophages (Fig. 7A) or dendritic cell–enriched cultures (Fig. 7C). In contrast, IL-1β secretion was potentiated by SCC from mice that ingested higher amount of fiber (SCC HF) (Fig. 7A, 7C). In contrast, we verified that TNF-α production was upregulated by SCC independently of the fiber content (Fig. 7B, 7D). Strikingly, SCC HF was no longer able to elicit higher IL-1β production by macrophages in the absence of NLRP3 or caspase-1/11 expression (Fig. 7E). In contrast, we found that SCC HF, as well as conventional SCC, strongly stimulated TNF-α production by macrophages from these inflammasome-deficient mouse lineages (Fig. 7F).

FIGURE 7.

Dietary fiber accounts for NLRP3 inflammasome activation and IL-1β secretion in response to SCC. BMDM (A and B) or BMDC (C and D) from C57BL/6 mice were left untreated (medium) or stimulated for 6 h at 37°C with SCC (2% v/v) obtained from mice fed for 2 wk on control (SCC ctr; 4.5% fiber), ZF (SCC ZF; 0% fiber), or HF (SCC HF; 15% fiber) diets. The SN were collected for IL-1β and TNF-α quantification by ELISA. Data are representative of at least four independent experiments; the results are shown as mean of triplicate wells. Error bars indicate ± SD. (E and F) BMDM from C57BL/6, Nlrp3–/–, and Caspase-1/11–/– mice were left untreated (ctr) or stimulated with SCC (2% v/v) from control (4.5%) or HF diet–fed mice. After stimulation, the SN were collected for IL-1β and TNF-α quantification by ELISA. Data are representative of two independent experiments; the results are shown as mean of triplicate wells. (G and H) C57BL/6-derived BMDC were left untreated (medium) or stimulated with 1% v/v SCC control (SCC ctr), SCC ZF, or SCC from reconstituted GF mice with the gut microbiota from mice fed on control (SCC: ctr-GF) or ZF (SCC: ZF-GF) diet, as described in Materials and Methods. After 6 h, IL-1β and TNF-α was quantified in the SN. Data are representative of two independent experiments; the results are shown as mean of triplicate wells. Error bars indicate ± SD. (I) Peritoneal macrophages from C57BL/6 mice were left untreated (medium) or stimulated with SCC control (1% v/v), SCC from GF mice (SCC GF) (1% v/v), or B. fragilis (multiplicity of infection [MOI]: 50) for 6 h at 37°C. In the last 30 min of incubation, ATP (5 mM) was added. Western blotting analysis of pro–IL-1β p31 (31 kDa) in cell lysates was performed. (J and K) C57BL/6-derived BMDC were left untreated (medium) or stimulated with 1% v/v SCC ctr, ZF (SCC ZF), HF (SCC HF), or with SCC ZF pellet added of SCC ctr SN (SN ctr-ZF). After 6 h, IL-1β and TNF-α levels were quantified by ELISA. Data are representative of two independent experiments; the results are shown as mean of triplicate wells. Error bars indicate ± SD.

FIGURE 7.

Dietary fiber accounts for NLRP3 inflammasome activation and IL-1β secretion in response to SCC. BMDM (A and B) or BMDC (C and D) from C57BL/6 mice were left untreated (medium) or stimulated for 6 h at 37°C with SCC (2% v/v) obtained from mice fed for 2 wk on control (SCC ctr; 4.5% fiber), ZF (SCC ZF; 0% fiber), or HF (SCC HF; 15% fiber) diets. The SN were collected for IL-1β and TNF-α quantification by ELISA. Data are representative of at least four independent experiments; the results are shown as mean of triplicate wells. Error bars indicate ± SD. (E and F) BMDM from C57BL/6, Nlrp3–/–, and Caspase-1/11–/– mice were left untreated (ctr) or stimulated with SCC (2% v/v) from control (4.5%) or HF diet–fed mice. After stimulation, the SN were collected for IL-1β and TNF-α quantification by ELISA. Data are representative of two independent experiments; the results are shown as mean of triplicate wells. (G and H) C57BL/6-derived BMDC were left untreated (medium) or stimulated with 1% v/v SCC control (SCC ctr), SCC ZF, or SCC from reconstituted GF mice with the gut microbiota from mice fed on control (SCC: ctr-GF) or ZF (SCC: ZF-GF) diet, as described in Materials and Methods. After 6 h, IL-1β and TNF-α was quantified in the SN. Data are representative of two independent experiments; the results are shown as mean of triplicate wells. Error bars indicate ± SD. (I) Peritoneal macrophages from C57BL/6 mice were left untreated (medium) or stimulated with SCC control (1% v/v), SCC from GF mice (SCC GF) (1% v/v), or B. fragilis (multiplicity of infection [MOI]: 50) for 6 h at 37°C. In the last 30 min of incubation, ATP (5 mM) was added. Western blotting analysis of pro–IL-1β p31 (31 kDa) in cell lysates was performed. (J and K) C57BL/6-derived BMDC were left untreated (medium) or stimulated with 1% v/v SCC ctr, ZF (SCC ZF), HF (SCC HF), or with SCC ZF pellet added of SCC ctr SN (SN ctr-ZF). After 6 h, IL-1β and TNF-α levels were quantified by ELISA. Data are representative of two independent experiments; the results are shown as mean of triplicate wells. Error bars indicate ± SD.

Close modal

Changes in dietary fiber content shapes gut microbiota composition (22). Next, we asked whether alterations on intestinal microbiome could account for the differential influence of dietary fiber on SCC-driven IL-1β production. To this end, GF mice were reconstituted by oral gavage with donor feces originating from mice that were previously fed either with conventional diet or ZF diet. Two weeks after microbiota reconstitution of GF mice, SCC were recovered and analyzed. As shown in Fig. 7G, we found similar levels of IL-1β production after stimulation either by SCC ctr-GF or ZF-GF. These findings support the idea that the contrasting IL-1β responses observed with SCC presenting different contents of dietary fiber (Fig. 7A) were not due to secondary changes in microbiota composition. Control experiments revealed that SCC derived from GF mice (i.e., not reconstituted) failed to induce IL-1β secretion (Fig. 7G). Consistent with these results, SCC from GF mice only induced low levels of TNF-α in the cultures and reduced pro–IL-1 expression (Fig. 7H, 7I). To confirm our hypothesis, we centrifuged SCC control and mixed its SN fraction (presenting microbiota products) with pellets from SCC ZF (i.e., lacking particulate fiber; called SN control-ZF). As a proof of concept, our results revealed that IL-1β production was comparable to SCC ZF–stimulated cells (Fig. 7J), and again, the levels of TNF-α were similar between the samples (Fig. 7K). Collectively, our data suggest that inflammasome activation and IL-1β production induced by SCC is critically dependent on the presence of particulate fiber.

There is now awareness that increase of the fiber content in the diet not only impacts on intestinal microbiota composition, favoring the growth of beneficial nonpathogenic bacteria, but also increases bacterial metabolism, leading to higher production of short-chain fatty acids (SCFA) as acetate, butyrate, and propionate (23). We then asked whether SCC HF enhanced activation of NLRP3 via increased production of SCFA. Arguing against this hypothesis, we found similar levels of IL-1β and TNF-α production by WT cells as by GPR43- or GPR109A-deficient cells, the main G protein–coupled receptors (GPCRs) involved on SCFA signaling (24) (Supplemental Fig. 2).

As mentioned above, SCC is a heterogenous preparation composed of products from diet and heat-killed microbiota, posing difficulties to unequivocally demonstrate that fiber promotes inflammasome activation. To address this question, LPS-primed macrophages were stimulated with commercially available purified dietary fiber and caspase-1 activation was assessed by flow cytometry with FAM-YVAD-FMK (FLICA), a fluorescent probe that specifically binds the active form of caspase-1. Consistent with our hypothesis, purified fiber induces an increase in the frequency of FLICA+ cells (Fig. 8A), as well as strongly upregulated IL-1β production (Fig. 8B). As previously shown for the activation requirements of SCC, this response was dramatically reduced in NLRP3- or caspase-1/11–deficient cells (Fig. 8C, 8D). Taken together, these results suggest that purified fiber reproduces the effects of SCC by activating NLRP3 inflammasome and promoting IL-1β secretion.

FIGURE 8.

Dietary fiber induces caspase-1 activation and IL-1β secretion in a manner NLRP3 dependent contributing to peritonitis development. (A and B) BMDM from C57BL/6 mice were left untreated (medium) or stimulated with LPS (500 ng/ml) for 1 h, and purified fiber (FiberMais) was added at the indicated concentrations (% w/v; A) or 2.5% w/v (B). After 6 h, active caspase-1 was evaluated by flow cytometry using FAM-YVAD-FMK (FLICA) (excitation at 492 nm and emission at 520 nm) (A), and the SN were collected for IL-1β quantification by ELISA (B). Data are representative of two independent experiments; the results are shown as mean of triplicate wells. (C and D) BMDM (C) and BMDC (D) from C57BL/6, Nlrp3–/–, and Caspase 1/11–/– mice were left untreated (medium) or stimulated with LPS (500 ng/ml) for 1 h, and purified fiber (5% w/v) was added when indicated. After 6 h, the SN were collected for IL-1β quantification by ELISA. Data are representative of two independent experiments; the results are shown as mean of triplicate wells. (E and F) C57BL/6 mice were inoculated i.p. with PBS (Mock) or with B. fragilis (100 μl 1 × 108 CFU) in combination with 100 μl SCC from mice fed on control (Bf + SCC control) (n = 5), B. fragilis (100 μl 1 × 108 CFU) in combination with 100 μl SCC from mice fed ZF (Bf + SCC ZF) (n = 5), or B. fragilis (100 μl 1 × 108 CFU) in combination with 100 μl SCC from mice fed HF diets (Bf + SCC HF) (n = 5). The results show the (E) monitoring of body weight and (F) inflammation score after 7 d. Data are representative of two independent experiments. In (E), the results are shown as mean ± SD. In (F), each symbol represents one animal. Significance was calculated with two-way ANOVA with Bonferroni posttest. **p < 0.01, ***p < 0.001 compared with Mock; ##p < 0.01; ###p < 0.001 compared with Bf + SCC control. (G) C57BL/6 (n = 6) and Il1r–/– (n = 7) mice were inoculated i.p. with B. fragilis (100 μl 1 × 108 CFU) in combination with bran (10% v/v). Inflammation score was quantified after 7 d. Data are representative of two independent experiments. Each symbol represents one animal. Significance was calculated with unpaired t test. ***p < 0.001.

FIGURE 8.

Dietary fiber induces caspase-1 activation and IL-1β secretion in a manner NLRP3 dependent contributing to peritonitis development. (A and B) BMDM from C57BL/6 mice were left untreated (medium) or stimulated with LPS (500 ng/ml) for 1 h, and purified fiber (FiberMais) was added at the indicated concentrations (% w/v; A) or 2.5% w/v (B). After 6 h, active caspase-1 was evaluated by flow cytometry using FAM-YVAD-FMK (FLICA) (excitation at 492 nm and emission at 520 nm) (A), and the SN were collected for IL-1β quantification by ELISA (B). Data are representative of two independent experiments; the results are shown as mean of triplicate wells. (C and D) BMDM (C) and BMDC (D) from C57BL/6, Nlrp3–/–, and Caspase 1/11–/– mice were left untreated (medium) or stimulated with LPS (500 ng/ml) for 1 h, and purified fiber (5% w/v) was added when indicated. After 6 h, the SN were collected for IL-1β quantification by ELISA. Data are representative of two independent experiments; the results are shown as mean of triplicate wells. (E and F) C57BL/6 mice were inoculated i.p. with PBS (Mock) or with B. fragilis (100 μl 1 × 108 CFU) in combination with 100 μl SCC from mice fed on control (Bf + SCC control) (n = 5), B. fragilis (100 μl 1 × 108 CFU) in combination with 100 μl SCC from mice fed ZF (Bf + SCC ZF) (n = 5), or B. fragilis (100 μl 1 × 108 CFU) in combination with 100 μl SCC from mice fed HF diets (Bf + SCC HF) (n = 5). The results show the (E) monitoring of body weight and (F) inflammation score after 7 d. Data are representative of two independent experiments. In (E), the results are shown as mean ± SD. In (F), each symbol represents one animal. Significance was calculated with two-way ANOVA with Bonferroni posttest. **p < 0.01, ***p < 0.001 compared with Mock; ##p < 0.01; ###p < 0.001 compared with Bf + SCC control. (G) C57BL/6 (n = 6) and Il1r–/– (n = 7) mice were inoculated i.p. with B. fragilis (100 μl 1 × 108 CFU) in combination with bran (10% v/v). Inflammation score was quantified after 7 d. Data are representative of two independent experiments. Each symbol represents one animal. Significance was calculated with unpaired t test. ***p < 0.001.

Close modal

Finally, we asked whether differences on fiber content in the SCC impacts on peritonitis outcome (abscess formation) induced by B. fragilis. Mice injected with SCC ZF showed modest pathology, manifested by reduced weight loss and abscess score (Fig. 8E, 8F). In contrast, both disease indexes were worsened in mice infected with B. fragilis in combination with SCC HF, whether compared with mice inoculated with SCC ZF or control SCC. Finally, as proof of concept, we observed that as SCC, bran can promote peritonitis when combined with B. fragilis in a mechanism dependent on IL-1 pathway (Fig. 8G).

Collectively, our studies linked the colonic content of fiber to severity of peritonitis induced by B. fragilis. Although dietary fiber shapes healthy microbiota inside the gut and contributes to homeostasis, our mechanistic studies highlighted the dichotomic role of fiber. Predictably, inflammation is unleashed when the gut barrier is broken and fiber-rich intestinal content overflows to peritoneal cavity together with bacteria from microbiota. Opposite to commensal bacteria and acting as a typical damage signal, the extraintestinal fiber stimulates IL-1β–dependent inflammation by activating the NLRP3 inflammasome in innate immune cells contributing to pathology.

Although IL-1 family cytokines are implicated in host defense against several pathogens, excessive formation of IL-1β is deleterious in a broad range of inflammatory diseases, including cryopyrin-associated periodic syndromes, gout, Alzheimer, and type 2 diabetes (25). Congruent with knowledge emerging from clinical and experimental research on sepsis and peritonitis (26, 27), our studies in the B. fragilis/SCC model support the idea that IL-1β blockage or NLRP3 inhibition might attenuate inflammatory responses otherwise induced by dietary fibers that extravasate into the peritoneal cavity. Once activated by SCC, canonical NLRP3 pathway drives the formation of inflammatory abscesses in an IL-1β–dependent manner, dispensing the involvement of the IL-18/IL-18R pathway, probably by promoting chemokine production, vascular alterations, and leukocyte recruitment to the peritoneal cavity. Studies in multiple transgenic mice showed that targeting the NLRP3/IL-1β pathway blunted acute peritonitis without interfering on bacteria clearance and host resistance to disease.

Well known as a member of our gut microbiota, B. fragilis presents immunomodulatory properties, mostly mediated by its capsular polysaccharide A, with beneficial effects on immune homeostasis and inflammatory disorders. B. fragilis is an important symbiont protecting the host against colitis through induction of IL-10–secreting regulatory T cells (2830). In a recent in vitro study, Jiang et al. (31) have shown that B. fragilis (or its polysaccharide A alone) has anti-inflammatory properties, inhibiting IL-1β–induced inflammation in fetal intestinal epithelial cells. Despite the beneficial coexistence with its host, B. fragilis has pathogenic behavior upon escaping their habitat, becoming the most common cause of anaerobic infections in humans, such as hepatic bacteremia, peritonitis, and anaerobic abdominal sepsis (32). We and other groups have observed that mice inoculated with B. fragilis alone are refractory to development of acute peritonitis. In contrast, peritonitis was induced in B. fragilis–infected WT mice that were coinjected with SCC. Opposite to SCC and similar to the weak proinflammatory potential of other Bacteroides species (33), B. fragilis failed to induce IL-1β secretion in vitro and in vivo. Notwithstanding this, in this study, we showed that B. fragilis efficiently induces macrophage priming for inflammasome activation by cellular damage signals, such as ATP. Noteworthy in this context, B. fragilis resembles the phenotype of other members of symbiotic microbiota, most of which are unlikely to express exotoxins, type III secretion system proteins and to induce tissue damage in the gut. Although this is the scenario prevailing in the steady state, it has been shown that upon intestinal injury, selective members of the microbiota, such as Proteus miriabilis, induce colitis via activation of the NLRP3/IL-1β pathway in recruited inflammatory monocytes (33). More recently, in an in vitro study involving a different B. fragilis strain (NCTC9343), Chen et al. (34) reported that macrophages upregulated the expression of pro–IL-1β. Similar to the phenotype of our bacterial strain (638R), pattern recognition receptor sensing of NCTC9343 was not sufficient to induce a prompt secretion of IL-1β. However, appreciable levels of this cytokine were detected upon prolonged incubation with B. fragilis independently of bacterial viability (34), perhaps reflecting cellular damage and inflammasome activation by endogenously released signals. Alternatively, there is a precedent that macrophages activated by the B. fragilis strain ZY-312 increase their phagocytic activity and upregulate gene expression of inducible NO synthase, IL-12, and IL-1β in macrophages (35). Thus, admittedly, strain-dependent differences in enterotoxin expression and virulent potential of B. fragilis strains might influence the outcome of macrophage interaction with this symbiotic bacterium.

The concept that sterile particulate products trigger inflammatory responses through NLRP3 inflammasome has been supported by advances in the pathogenesis of several inflammatory diseases, such as silicosis, Alzheimer, and gout (6, 8, 9, 11). Experimental models of peritonitis, whether induced by the combined injection of LPS plus alum (36) or uric acid crystals (6), provided additional evidence that the NLRP3 inflammasome is activated by particulate matter and contributes to peritonitis. Against this general background, our study in the model of B. fragilis–induced peritonitis provides an example of the critical role that particulate dietary fibers exert on pathogenic outcome. Typically acting as a second signal of NRLP3 activation, the undigestible fiber stimulates IL-1β–dependent peritonitis.

Dietary fiber is a nutrient category that includes polysaccharides that are not enzymatically digestible in the human gut. In the large bowel, soluble fibers (such as guar gum and pectin) are fermented to SCFA by the local microbiota, whereas insoluble fibers, such as cellulose, add bulk to stool and affect the bowel movement (37). Although NLRP3 can be activated by broad range of exogenous and endogenous stimuli, the molecular mechanisms underlying ligand binding to sensor receptors and ensuing inflammasome activation have not been fully elucidated. There is also awareness that some inflammasome activators share the ability to induce NLRP3 oligomerization via pathways regulated by NEK7 and vimentin (3840). It is also known that K+ efflux induced by phagocytosis of particulate products, such as crystals formed in tissue environment or produced by endogenous metabolism, drive activation of the NLRP3 inflammasome via inducible release of lysosomal cathepsins (41). Congruent with this pathway, SCC-induced NLRP3 activation was abolished by treatment with cytochalasin D, Ca-74-Me, MitoTEMPO, and high extracellular concentrations of KCl, whereas the role of the ATP-P2X7 axis in the pathogenesis seems to be minor.

Albeit limited to the in vitro settings, we showed that purified fiber, in this study, used as a replacement for SCC induced NLRP3-dependent caspase-1 activation and IL-1β secretion in LPS-primed macrophages. Historically, the use of a dietary component as a potentiating agent of peritonitis induced by B. fragilis can be traced to the late 1970s. Using bran, a cereal grain component that is rich in dietary fiber and essential fatty acids, it was demonstrated that formation of experimental sterile abscesses is triggered by fiber particles present in feces, although its mechanism of action is not understood (42). As shown in this study, the finding that peritonitis induced by bran and B. fragilis is attenuated in IL-1R–deficient mice adds weight to the hypothesis that dietary fiber might worsen peritonitis by triggering inflammasome activation.

Fermentation of dietary soluble fibers by gut microbiota leads to SCFA production, including acetate, butyrate, and propionate. These metabolites can modulate cell functions either by histone deacetylase inhibition or through GPCRs such as GPR43 and GPR41 (24). Macia et al. (22) have shown NLRP3 activation in intestinal epithelial cells by SCFA, in particular acetate, through GPR43 and GPR109A receptors, leading to IL-18 production and protection against colitis. In this study, we showed that IL-β secretion induced by SCC is entirely independent of these SCFA receptors. This observation is not surprising because SCFA are extremely labile and probably do not resist the autoclaving process to which SCC is submitted during its preparation. Diet changes, such as the Western lifestyle involving reduced microbiota-accessible carbohydrates intake, are known to rapidly affect the composition of the gut microbiota in mammals (23, 43). Deprivation of dietary fiber promotes outgrowth of mucus-degrading bacteria species and higher susceptibility to enteropathogens (44). The absence of fiber favors the growth of Bacteroidiaceae family species and a reduction on frequency of Prevotellaceae family members (22). Studies using GF mice reconstitution protocols showed us that the altered gut microbiota induced by fiber deprivation do not account for the drastic reduction of the IL-1β production induced by SCC from ZF-fed mice, confirming the essential role of fiber content.

SCFA represent a group of immunomodulatory metabolites that blunts metabolic syndrome and inflammatory diseases, such as type 1 and type 2 diabetes, asthma, food allergies, cardiovascular disease, experimental diabetic nephropathy, and inflammatory bowel diseases, such as colitis (23, 24, 45). Despite its proinflammatory NLRP3-mediated effect described in this study, the immunomodulatory role of fiber, through the production of its metabolites SCFA, on sepsis is currently being investigated in our laboratory. Notwithstanding its benefits, the consumption of fermentable fiber can present deleterious role in specific scenarios. For example, Singh et al. (46) have shown that the increase of inulin intake, a soluble fiber that yields SCFA generation, induces icteric hepatocellular carcinoma in a microbiota-dependent manner. Notably, the depletion of fermenting bacteria markedly reduced systemic SCFA levels and prevented the disease, showing the deleterious effect of fiber through anti-inflammatory properties of its metabolites.

In conclusion, in this study we describe the involvement of the NLRP3–ASC–caspase-1–IL-1β axis in the outcome of peritonitis. In our B. fragilis–induced model, SCC injection has mimicked the impact of dietary fiber overflow from the gut into peritoneal cavity, resembling the natural breakdown of epithelial barrier compartmentalization. Our results point to an unprecedented pathological role of extraintestinal dietary fiber as a damage-associated molecular pattern, activating NLRP3 inflammasome in macrophages and dendritic cells. Future studies on the pathogenesis of peritonitis might clarify whether overflow of dietary fibers might synergize with B. fragilis and/or other symbiont microbiota to potentiate inflammation, ultimately worsening the outcome of bacterial infection.

We thank D. Faustino and R. Serra for technical assistance.

This work was supported by the Conselho Nacional de Desenvolvimento Científico e Tecnológico, the Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro, and the Fundação de Amparo à Pesquisa do Estado de São Paulo.

J.P.C., B.J.-A., L.A.L., and A.C.O. conceived and designed the experiments. J.P.C., B.J.-A., E.S.R.-J., and A.C.A.M.-S. performed the experiments. J.P.C., B.J.-A., and A.C.O. analyzed the data. E.d.O.F., R.M.C.P.D., J.E.-L., R.C.-S., E.M., C.R.M., D.S.Z., M.B., J.S., L.A.L., and A.C.O. contributed with reagents/cells/mice/materials. A.C.O. wrote the paper.

The online version of this article contains supplemental material.

Abbreviations used in this article

BMDC

bone marrow–derived dendritic cell–enriched culture

BMDM

bone marrow–derived macrophage

ctr

control

ctr-GF

GF reconstituted with microbiota from conventional diet–fed mice

GF

germ-free

HF diet

high fiber content (15%) diet

p.i.

postinfection

SCC

sterile cecal content

SCFA

short-chain fatty acid

SN

supernatant

UFRJ

Universidade Federal do Rio de Janeiro

WT

wild-type

ZF diet

zero fiber content (0%) diet

ZF-GF

GF reconstituted with microbiota from ZF-fed mice

1
Gibson
F. C.
 III
,
A. B.
Onderdonk
,
D. L.
Kasper
,
A. O.
Tzianabos
.
1998
.
Cellular mechanism of intraabdominal abscess formation by Bacteroides fragilis.
J. Immunol.
160
:
5000
5006
.
2
Onderdonk
A. B.
,
R. B.
Markham
,
D. F.
Zaleznik
,
R. L.
Cisneros
,
D. L.
Kasper
.
1982
.
Evidence for T cell-dependent immunity to Bacteroides fragilis in an intraabdominal abscess model.
J. Clin. Invest.
69
:
9
16
.
3
Cohen-Poradosu
R.
,
R. M.
McLoughlin
,
J. C.
Lee
,
D. L.
Kasper
.
2011
.
Bacteroides fragilis-stimulated interleukin-10 contains expanding disease.
J. Infect. Dis.
204
:
363
371
.
4
Próchnicki
T.
,
M. S.
Mangan
,
E.
Latz
.
2016
.
Recent insights into the molecular mechanisms of the NLRP3 inflammasome activation.
F1000 Res.
5
:
1469
.
5
Yang
Y.
,
H.
Wang
,
M.
Kouadir
,
H.
Song
,
F.
Shi
.
2019
.
Recent advances in the mechanisms of NLRP3 inflammasome activation and its inhibitors.
Cell Death Dis.
10
:
128
138
.
6
Martinon
F.
,
V.
Pétrilli
,
A.
Mayor
,
A.
Tardivel
,
J.
Tschopp
.
2006
.
Gout-associated uric acid crystals activate the NALP3 inflammasome.
Nature
440
:
237
241
.
7
Cassel
S. L.
,
S. C.
Eisenbarth
,
S. S.
Iyer
,
J. J.
Sadler
,
O. R.
Colegio
,
L. A.
Tephly
,
A. B.
Carter
,
P. B.
Rothman
,
R. A.
Flavell
,
F. S.
Sutterwala
.
2008
.
The Nalp3 inflammasome is essential for the development of silicosis.
Proc. Natl. Acad. Sci. USA
105
:
9035
9040
.
8
Dostert
C.
,
V.
Pétrilli
,
R.
Van Bruggen
,
C.
Steele
,
B. T.
Mossman
,
J.
Tschopp
.
2008
.
Innate immune activation through Nalp3 inflammasome sensing of asbestos and silica.
Science
320
:
674
677
.
9
Halle
A.
,
V.
Hornung
,
G. C.
Petzold
,
C. R.
Stewart
,
B. G.
Monks
,
T.
Reinheckel
,
K. A.
Fitzgerald
,
E.
Latz
,
K. J.
Moore
,
D. T.
Golenbock
.
2008
.
The NALP3 inflammasome is involved in the innate immune response to amyloid-beta.
Nat. Immunol.
9
:
857
865
.
10
Duewell
P.
,
H.
Kono
,
K. J.
Rayner
,
C. M.
Sirois
,
G.
Vladimer
,
F. G.
Bauernfeind
,
G. S.
Abela
,
L.
Franchi
,
G.
Nuñez
,
M.
Schnurr
, et al
.
2010
.
NLRP3 inflammasomes are required for atherogenesis and activated by cholesterol crystals. [Published erratum appears in 2010 Nature 466: 652.]
Nature
464
:
1357
1361
.
11
Heneka
M. T.
,
M. P.
Kummer
,
A.
Stutz
,
A.
Delekate
,
S.
Schwartz
,
A.
Vieira-Saecker
,
A.
Griep
,
D.
Axt
,
A.
Remus
,
T. C.
Tzeng
, et al
.
2013
.
NLRP3 is activated in Alzheimer’s disease and contributes to pathology in APP/PS1 mice.
Nature
493
:
674
678
.
12
Nulsen
M. F.
,
J. J.
Finlay-Jones
,
J. M.
Skinner
,
P. J.
McDonald
.
1983
.
Intra-abdominal abscess formation in mice: quantitative studies on bacteria and abscess-potentiating agents.
Br. J. Exp. Pathol.
64
:
345
353
.
13
Mariathasan
S.
,
D. S.
Weiss
,
K.
Newton
,
J.
McBride
,
K.
O’Rourke
,
M.
Roose-Girma
,
W. P.
Lee
,
Y.
Weinrauch
,
D. M.
Monack
,
V. M.
Dixit
.
2006
.
Cryopyrin activates the inflammasome in response to toxins and ATP.
Nature
440
:
228
232
.
14
Matikainen
S.
,
T. A.
Nyman
,
W.
Cypryk
.
2020
.
Function and regulation of noncanonical caspase-4/5/11 inflammasome.
J. Immunol.
204
:
3063
3069
.
15
Eisenbarth
S. C.
,
O. R.
Colegio
,
W.
O’Connor
,
F. S.
Sutterwala
,
R. A.
Flavell
.
2008
.
Crucial role for the Nalp3 inflammasome in the immunostimulatory properties of aluminium adjuvants.
Nature
453
:
1122
1126
.
16
Franchi
L.
,
G.
Núñez
.
2008
.
The Nlrp3 inflammasome is critical for aluminium hydroxide-mediated IL-1beta secretion but dispensable for adjuvant activity.
Eur. J. Immunol.
38
:
2085
2089
.
17
Ramos-Junior
E. S.
,
A. C.
Morandini
,
C. L.
Almeida-da-Silva
,
E. J.
Franco
,
J.
Potempa
,
K. A.
Nguyen
,
A. C.
Oliveira
,
D. S.
Zamboni
,
D. M.
Ojcius
,
J.
Scharfstein
,
R.
Coutinho-Silva
.
2015
.
A dual role for P2X7 receptor during Porphyromonas gingivalis infection.
J. Dent. Res.
94
:
1233
1242
.
18
Moreira-Souza
A. C. A.
,
C. L. C.
Almeida-da-Silva
,
T. P.
Rangel
,
G. D. C.
Rocha
,
M.
Bellio
,
D. S.
Zamboni
,
R. C.
Vommaro
,
R.
Coutinho-Silva
.
2017
.
The P2X7 receptor mediates Toxoplasma gondii control in macrophages through canonical NLRP3 inflammasome activation and reactive oxygen species production.
Front. Immunol.
8
:
1257
.
19
Jo
E.-K.
,
J. K.
Kim
,
D.-M.
Shin
,
C.
Sasakawa
.
2016
.
Molecular mechanisms regulating NLRP3 inflammasome activation.
Cell. Mol. Immunol.
13
:
148
159
.
20
Hornung
V.
,
F.
Bauernfeind
,
A.
Halle
,
E. O.
Samstad
,
H.
Kono
,
K. L.
Rock
,
K. A.
Fitzgerald
,
E.
Latz
.
2008
.
Silica crystals and aluminum salts activate the NALP3 inflammasome through phagosomal destabilization.
Nat. Immunol.
9
:
847
856
.
21
Orlowski
G. M.
,
J. D.
Colbert
,
S.
Sharma
,
M.
Bogyo
,
S. A.
Robertson
,
K. L.
Rock
.
2015
.
Multiple cathepsins promote pro-IL-1β synthesis and NLRP3-mediated IL-1β activation.
J. Immunol.
195
:
1685
1697
.
22
Macia
L.
,
J.
Tan
,
A. T.
Vieira
,
K.
Leach
,
D.
Stanley
,
S.
Luong
,
M.
Maruya
,
C.
Ian McKenzie
,
A.
Hijikata
,
C.
Wong
, et al
.
2015
.
Metabolite-sensing receptors GPR43 and GPR109A facilitate dietary fibre-induced gut homeostasis through regulation of the inflammasome.
Nat. Commun.
6
:
6734
6748
.
23
Tan
J.
,
C.
McKenzie
,
M.
Potamitis
,
A. N.
Thorburn
,
C. R.
Mackay
,
L.
Macia
.
2014
.
The role of short-chain fatty acids in health and disease.
Adv. Immunol.
121
:
91
119
.
24
Tan
J. K.
,
C.
McKenzie
,
E.
Mariño
,
L.
Macia
,
C. R.
Mackay
.
2017
.
Metabolite-sensing G protein-coupled receptors-facilitators of diet-related immune regulation.
Annu. Rev. Immunol.
35
:
371
402
.
25
Menu
P.
,
J. E.
Vince
.
2011
.
The NLRP3 inflammasome in health and disease: the good, the bad and the ugly.
Clin. Exp. Immunol.
166
:
1
15
.
26
Shakoory
B.
,
J. A.
Carcillo
,
W. W.
Chatham
,
R. L.
Amdur
,
H.
Zhao
,
C. A.
Dinarello
,
R. Q.
Cron
,
S. M.
Opal
.
2016
.
Interleukin-1 receptor blockade is associated with reduced mortality in sepsis patients with features of macrophage activation syndrome: reanalysis of a prior phase III trial.
Crit. Care Med.
44
:
275
281
.
27
Liu
Y.
,
Y. Y.
Jing
,
C. Y.
Zeng
,
C. G.
Li
,
L. H.
Xu
,
L.
Yan
,
W. J.
Bai
,
Q. B.
Zha
,
D. Y.
Ouyang
,
X. H.
He
.
2018
.
Scutellarin suppresses NLRP3 inflammasome activation in macrophages and protects mice against bacterial sepsis.
Front. Pharmacol.
8
:
975
991
.
28
Mazmanian
S. K.
,
J. L.
Round
,
D. L.
Kasper
.
2008
.
A microbial symbiosis factor prevents intestinal inflammatory disease.
Nature
453
:
620
625
.
29
Round
J. L.
,
S. K.
Mazmanian
.
2010
.
Inducible Foxp3+ regulatory T-cell development by a commensal bacterium of the intestinal microbiota.
Proc. Natl. Acad. Sci. USA
107
:
12204
12209
.
30
Lobo
L. A.
,
C. F.
Benjamim
,
A. C.
Oliveira
.
2016
.
The interplay between microbiota and inflammation: lessons from peritonitis and sepsis.
Clin. Transl. Immunology
5
:
e90
.
31
Jiang
F.
,
D.
Meng
,
M.
Weng
,
W.
Zhu
,
W.
Wu
,
D.
Kasper
,
W. A.
Walker
.
2017
.
The symbiotic bacterial surface factor polysaccharide A on Bacteroides fragilis inhibits IL-1β-induced inflammation in human fetal enterocytes via toll receptors 2 and 4.
PLoS One
12
:
e0172738
.
32
Brook
I.
2008
.
Microbiology and management of abdominal infections.
Dig. Dis. Sci.
53
:
2585
2591
.
33
Seo
S.-U.
,
N.
Kamada
,
R.
Muñoz-Planillo
,
Y. G.
Kim
,
D.
Kim
,
Y.
Koizumi
,
M.
Hasegawa
,
S. D.
Himpsl
,
H. P.
Browne
,
T. D.
Lawley
, et al
.
2015
.
Distinct commensals induce interleukin-1β via NLRP3 inflammasome in inflammatory monocytes to promote intestinal inflammation in response to injury.
Immunity
42
:
744
755
.
34
Chen
K.
,
N. K.
Shanmugam
,
M. A.
Pazos
,
B. P.
Hurley
,
B. J.
Cherayil
.
2016
.
Commensal bacteria-induced inflammasome activation in mouse and human macrophages is dependent on potassium efflux but does not require phagocytosis or bacterial viability.
PLoS One
11
:
e0160937
.
35
Deng
H.
,
Z.
Li
,
Y.
Tan
,
Z.
Guo
,
Y.
Liu
,
Y.
Wang
,
Y.
Yuan
,
R.
Yang
,
Y.
Bi
,
Y.
Bai
,
F.
Zhi
.
2016
.
A novel strain of Bacteroides fragilis enhances phagocytosis and polarises M1 macrophages.
Sci. Rep.
6
:
29401
29411
.
36
Giguère
P. M.
,
B. J.
Gall
,
E. A.
Ezekwe
Jr.
,
G.
Laroche
,
B. K.
Buckley
,
C.
Kebaier
,
J. E.
Wilson
,
J. P.
Ting
,
D. P.
Siderovski
,
J. A.
Duncan
.
2014
.
G Protein signaling modulator-3 inhibits the inflammasome activity of NLRP3.
J. Biol. Chem.
289
:
33245
33257
.
37
James
S. L.
,
J. G.
Muir
,
S. L.
Curtis
,
P. R.
Gibson
.
2003
.
Dietary fibre: a roughage guide.
Intern. Med. J.
33
:
291
296
.
38
dos Santos
G.
,
M. R.
Rogel
,
M. A.
Baker
,
J. R.
Troken
,
D.
Urich
,
L.
Morales-Nebreda
,
J. A.
Sennello
,
M. A.
Kutuzov
,
A.
Sitikov
,
J. M.
Davis
, et al
.
2015
.
Vimentin regulates activation of the NLRP3 inflammasome.
Nat. Commun.
6
:
6574
6586
.
39
Shi
H.
,
Y.
Wang
,
X.
Li
,
X.
Zhan
,
M.
Tang
,
M.
Fina
,
L.
Su
,
D.
Pratt
,
C. H.
Bu
,
S.
Hildebrand
, et al
.
2016
.
NLRP3 activation and mitosis are mutually exclusive events coordinated by NEK7, a new inflammasome component.
Nat. Immunol.
17
:
250
258
.
40
He
Y.
,
M. Y.
Zeng
,
D.
Yang
,
B.
Motro
,
G.
Núñez
.
2016
.
NEK7 is an essential mediator of NLRP3 activation downstream of potassium efflux.
Nature
530
:
354
357
.
41
Mangan
M. S. J.
,
E. J.
Olhava
,
W. R.
Roush
,
H. M.
Seidel
,
G. D.
Glick
,
E.
Latz
.
2018
.
Targeting the NLRP3 inflammasome in inflammatory diseases. [Published erratum appears in 2018 Nat. Rev. Drug Discov. 17: 688.]
Nat. Rev. Drug Discov.
17
:
588
606
.
42
Wilkins
T. D.
1979
.
Experimental anaerobic infections.
In
Microbiology.
D.
Schlessinger
, ed.
ASM
,
Washington, DC
, p.
215
.
43
Sonnenburg
E. D.
,
J. L.
Sonnenburg
.
2014
.
Starving our microbial self: the deleterious consequences of a diet deficient in microbiota-accessible carbohydrates.
Cell Metab.
20
:
779
786
.
44
Desai
M. S.
,
A. M.
Seekatz
,
N. M.
Koropatkin
,
N.
Kamada
,
C. A.
Hickey
,
M.
Wolter
,
N. A.
Pudlo
,
S.
Kitamoto
,
N.
Terrapon
,
A.
Muller
, et al
.
2016
.
A dietary fiber-deprived gut microbiota degrades the colonic mucus barrier and enhances pathogen susceptibility.
Cell
167
:
1339
1353.e21
.
45
Li
Y. J.
,
X.
Chen
,
T. K.
Kwan
,
Y. W.
Loh
,
J.
Singer
,
Y.
Liu
,
J.
Ma
,
J.
Tan
,
L.
Macia
,
C. R.
Mackay
, et al
.
2020
.
Dietary fiber protects against diabetic nephropathy through short-chain fatty acid-mediated activation of G protein-coupled receptors GPR43 and GPR109A.
J. Am. Soc. Nephrol.
31
:
1267
1281
.
46
Singh
V.
,
B. S.
Yeoh
,
B.
Chassaing
,
X.
Xiao
,
P.
Saha
,
R.
Aguilera Olvera
,
J. D.
Lapek
Jr.
,
L.
Zhang
,
W. B.
Wang
,
S.
Hao
, et al
.
2018
.
Dysregulated microbial fermentation of soluble fiber induces cholestatic liver cancer.
Cell
175
:
679
694.e22
.

The authors have no financial conflicts of interest.

Supplementary data