|Home | About | Journals | Submit | Contact Us | Français|
Crohn’s disease and ulcerative colitis are inflammatory bowel diseases with high prevalence in humans. Nlrp3 interacts with the adaptor protein ASC to activate caspase-1 in inflammasomes, protein complexes responsible for the maturation and secretion of IL-1β and IL-18. Decreased expression of the NOD-like receptor (NLR) Nlrp3 was recently associated with susceptibility to Crohn’s disease. However, the role of Nlrp3 in colitis has not been characterized. Here, we show that mice deficient for Nlrp3 or the inflammasome effectors ASC and caspase-1 are highly susceptible to dextran sodium sulfate (DSS)-induced colitis. Defective inflammasome activation leads to loss of epithelial integrity, resulting in systemic dispersion of commensal bacteria, massive leukocyte infiltration and increased chemokine production in the colon. As a consequence, significantly higher mortality rates were noted for mice lacking components of the Nlrp3 inflammasome. Therefore, the Nlrp3 inflammasome is critically involved in the maintenance of intestinal homeostasis and protection against colitis.
Human inflammatory bowel disease (IBD), comprising ulcerative colitis and Crohn’s disease, constitutes a major health problem in developed countries (Fiocchi, 1998). Ulcerative colitis exhibits a characteristic profile of chronic inflammation involving the distal colon and rectum, and is generally recognized as an immune-mediated disorder resulting from abnormal interaction between colonic microflora and mucosal immune cells (Goyette et al., 2007). Excessive inflammatory and immune responses in the intestine are thought to be due to a breach in the epithelial barrier in the gut that segregates commensal microflora from the host’s systemic organs (Strober et al., 2002). Indeed, deterioration of the mucus layer of the colon is prominent in patients with ulcerative colitis (Podolsky and Isselbacher, 1984; Rhodes, 1996). In addition, studies in rodents have linked tissue damage and disruption of the epithelial barrier in the gut to cytokine imbalances (Bouma and Strober, 2003). The production of these inflammatory mediators has been implicated in the pathogenesis of experimental colitis and IBD in humans (Podolsky, 2002).
The synthesis and secretion of pro-inflammatory cytokines is governed by germline-encoded receptors such as the toll-like receptor (TLR) and NOD-like receptor (NLR) family (Kanneganti et al., 2007; Kopp and Medzhitov, 2003). TLRs are membrane-bound receptors that detect pathogen-associated molecular patterns (PAMPs) in the extracellular milieu (Kawai and Akira, 2007). TLR activation results in the rapid transcriptional activation of effector genes, including cytokines and chemokines, that drive recruitment and/or activation of immune cells at mucosal surfaces. This immune cell recruitment is believed to play an important role in protecting against bacterial dissemination but may also underlie the clinical manifestations associated with inflammation as well as tissue damage therein. For instance, mice lacking the flagellin receptor TLR5 developed spontaneous colitis (Vijay-Kumar et al., 2007). Although mice deficient for the lipopolysaccharide (LPS) receptor TLR4, the lipoprotein receptor TLR2 or the TLR signaling adaptor MyD88 do not display an overt intestinal phenotype, they develop exacerbated injury upon exposure to dextran sodium sulfate (DSS) (Araki et al., 2005; Fukata et al., 2005; Rakoff-Nahoum et al., 2004).
In addition to TLRs, several members of the cytosolic NLR family have been identified as key regulators of cytokine production (Kanneganti et al., 2007). Notably, the NLR protein CARD15/NOD2 was the first gene to be associated with Crohn’s disease (Hugot et al., 2001; Ogura et al., 2001). NOD2 was subsequently shown to mediate activation of NF- B and MAP kinases (Girardin et al., 2003; Inohara et al., 2003), the NLR protein Nlrp3 (also referred to as Nalp3/CIAS1/Cryopyrin) is involved in activation of the cysteine protease caspase-1 (Lamkanfi et al., 2007). Homotypic interactions between the pyrin domain in the N-terminus of Nlrp3 and the bipartite adaptor protein ASC bridge the association of caspase-1 to Nlrp3 in a large protein complex referred to as the ‘inflammasome’ (Martinon et al., 2002). Activated caspase-1 processes the cytosolic precursors of the related cytokines interleukin (IL)-1β and IL-18, thus allowing secretion of the biologically active cytokines. Hence, mice lacking caspase-1 are defective in the maturation and secretion of IL-1β and IL-18 (Ghayur et al., 1997; Kuida et al., 1995; Li et al., 1995). IL-1β participates in the generation of systemic and local responses to infection, injury and immunological challenges by generating fever, activating lymphocytes, and by promoting leukocyte infiltration at sites of injury or infection (Dinarello, 1996). Although IL-18 lacks the pyrogenic activity of IL-1β, it is involved in the induction of several secondary pro-inflammatory cytokines, chemokines, cell adhesion molecules, and nitric oxide synthesis (Horwood et al., 1998; Olee et al., 1999).
Gain-of-function mutations within Nlrp3 have been associated with three autoinflammatory disorders characterized by skin rashes and prolonged episodes of fever in the absence of any apparent infection. These hereditary periodic-fever syndromes are Muckle-Wells syndrome (MWS), familial cold autoinflammatory syndrome (FACS) and neonatal-onset multisystem inflammatory disease (NOMID), and they are collectively referred to as the Cryopyrin-associated periodic syndromes (CAPS) (Agostini et al., 2004). Functional studies revealed that the disease-associated Nlrp3 mutations enhance caspase-1 activation and IL-1β secretion (Dowds et al., 2004). In addition, decreased Nlrp3 expression and IL-1β production was recently linked with increased susceptibility to Crohn’s disease in humans (Villani et al., 2009). However, the role of the Nlrp3 inflammasome in colitis has not been characterized. To understand the role of the Nlrp3 inflammasome in colitis, we studied the response of Nlrp3−/−, ASC−/− and Caspase-1−/− mice to DSS-induced colitis. Our results indicated a major role for the Nlrp3 inflammasome in protection against DSS-induced colitis and revealed its protective function in intestinal homeostasis.
Oral administration of DSS is directly toxic to the colonic epithelium (Kitajima et al., 1999) and triggers inflammation by disrupting the compartmentalization of commensal bacteria in the gut (Rakoff-Nahoum et al., 2004). To study the contribution of Nlrp3 to the development of colitis, we first assessed the mortality rate of age- and sex-matched wild-type and Nlrp3−/− mice after oral administration of 4% DSS in drinking water. Only 20% of wild-type mice died during the DSS administration period, but a mortality rate higher than 80% was noted for the Nlrp3−/− cohort (Fig. 1A). The experiment was repeated with a lower DSS concentration (3%) to study the phenotype of Nlrp3−/− mice under milder conditions. Nlrp3−/− mice suffered from more body weight loss from day 5 on (Fig. 1B). Simultaneously, stool consistency scores of Nlrp3−/− mice became significantly worse compared to those of DSS-fed wild-type mice (Fig. 1C). Differences in rectal bleeding were also apparent between the two groups, with Nlrp3−/− mice displaying significantly elevated scores relative to DSS-administered wild-type controls starting as early as day 2 (Fig. 1D). The evaluation of colon length is the parameter with the lowest variability in the model of DSS-induced colitis (Okayasu et al., 1990). To further assess the severity of colitis, colon length was measured in DSS-fed wild-type and Nlrp3−/− mice. Colons of Nlrp3−/− mice were on average 20% shorter than those of treated wild-type controls (Fig. 1E and F).
These clinical assessments were validated by histological examination of representative colon sections. In agreement with previous studies (Rakoff-Nahoum et al., 2004; Takagi et al., 2003), we observed marked histopathological changes in haematoxylin & eosin (H&E)-stained colons of DSS-treated wild-type mice characterized by crypt loss and infiltrating leukocytes (Fig. 1G). However, only minimal evidence of necrosis and ulceration was evident in colons of wild-type mice. In contrast, colonic sections of DSS-fed Nlrp3−/− mice displayed severe transmural inflammation with focal areas of extensive ulceration and necrotic lesions (Fig. 1G). Inflammatory infiltrates filled the lamina propria and submucosa in areas where the mucosa was intact, and often effaced the normal architecture of the tissue. Submucosal edema was often marked in areas of ulceration (Fig. 1G). Semi-quantitative scoring of these histological parameters confirmed that colitis severity in Nlrp3−/− mice was significantly higher than in wild-type mice (Fig. 1H). Wild-type mice were attributed an overall histological score of 1.625 ± 0.27, whereas Nlrp3−/− mice were assigned a score of 3.78 ± 0.15 (Fig. 1H). Consistent with the absence of disease in animals that were not fed DSS, no signs of inflammation or tissue damage were observed in colons of untreated wild-type and Nlrp3−/− mice (Supplementary Fig. 1A).
Intrarectal administration of 2,4,6-trinitrobenzenesulfonic acid (TNBS) represents an alternative model for the induction of acute colitis in mice through direct barrier destruction (Alex et al., 2009; Palmen et al., 1995). To assess whether Nlrp3 also exerts a protective role during acute TNBS-induced colitis, survival of wild-type and Nlrp3−/− mice was monitored for 5 days following intrarectal instillation of 150 mg/kg TNBS. As observed during acute DSS-induced colitis (Fig. 1), Nlrp3−/− mice were significantly more susceptible to acute TNBS-induced mortality than wild-type mice (Supplementary Fig. 1B). In addition, semi-quantitative scoring of inflammation in H&E-stained colon sections confirmed that colitis severity in Nlrp3−/− mice was significantly higher than in wild-type mice (Supplementary Fig. 1C). Collectively, these results demonstrate that Nlrp3-dependent signaling is critical for protection against acute DSS- and TNBS-induced mortality and morbidity.
Nlrp3 is expressed in a wide range of immune cells as well as in epithelial cells (Kummer et al., 2007). To determine the cell populations that are critical for Nlrp3-dependent protection against DSS-induced colitis, we generated 4 groups of Nlrp3 bone marrow chimeras (Fig. 2A). In agreement with our previous results (Fig. 1), Nlrp3−/− mice receiving Nlrp3−/− bone marrow presented with significantly worse symptoms of colitis relative to wild-type mice transplanted with wild-type bone marrow. Differences in clinical disease parameters between these groups such as body weight loss (Fig. 2B), stool consistency (Fig. 2C) and colonic bleeding (Fig. 2D) all reached statistical significance by day 7 post-DSS administration. Incidence and severity of colitis in Nlrp3−/− mice receiving wild-type bone marrow was comparable to that of Nlrp3−/− mice transplanted with Nlrp3−/− bone marrow (Fig. 2B–D), suggesting that Nlrp3 expression in non-hematopoietic cells is more important for protection against colitis than Nlrp3 expression in leukocytes. Indeed, wild-type mice transplanted with Nlrp3−/− bone marrow were less sensitive to DSS-induced colitis and presented with body weight changes, diarrhea and bleeding scores that were comparable to those of wild-type mice (Fig 2B–D). The marked improvement in the clinical manifestation of colitis in the latter groups was confirmed by less signs of severe histopathology in H&E-stained sections of the lamina propria of wild-type mice that received wild-type or Nlrp3−/− bone marrow (Fig. 2E, upper panels). In contrast, Nlrp3−/− mice presented with extensive crypt destruction and edema regardless of the Nlrp3 status of the transplanted bone marrow (Fig. 2E, lower panels). In agreement, colon homogenates of DSS-fed Nlrp3−/− recipients contained higher levels of inflammatory cytokines and chemokines relative to wild-type recipients (Supplementary Fig. 2). Overall, these results suggest that Nlrp3 expression in local cells of the colonic mucosa is critical for protection against DSS-induced colitis.
Nlrp3 recruits ASC and caspase-1 into a large protein complex termed the ‘inflammasome’ (Kanneganti et al., 2007; Lamkanfi and Dixit, 2009). To determine whether Nlrp3 inflammasome activation is implicated in protection against colitis, we assessed the response of mice lacking the downstream inflammasome components ASC and caspase-1. Similar to Nlrp3−/− mice (Fig. 1A), ASC−/− and caspase-1−/− mice were highly susceptible to DSS-induced colitis, with nearly all ASC−/− and caspase-1−/− mice dying within 2 weeks after administration of 4% DSS (Fig. 3A). As seen with Nlrp3−/− mice, ASC−/− and caspase-1−/− mice displayed significantly more body weight loss (Fig. 3B), higher stool consistency scores (Fig. 3C) and rectal bleeding (Fig. 3D) when fed on a milder regimen of 3% DSS. Moreover, the colon length of ASC−/− and caspase-1−/− mice was significantly reduced (Fig. 3E, F). Finally and as observed for Nlrp3−/− mice (Fig. 1G), H&E-stained colon sections of DSS-fed ASC−/− and caspase-1−/− mice displayed severe transmural inflammation with focal areas of extensive ulceration and necrotic lesions (Fig. 3G, H). The role of the Nlrp3 inflammasome in protection against DSS-induced colitis is not limited to the acute phase of disease because Nlrp3−/− and caspase-1−/− mice also suffered from increased body weight loss, diarrhea and reduced colon length during chronic disease (Supplementary Fig. 3). These results demonstrate that Nlrp3 inflammasome activation is critical for protection against DSS-induced colitis.
The Nlrp3 inflammasome is responsible for the maturation and secretion of the related cytokines IL-1β and IL-18 (Kanneganti et al., 2006; Mariathasan et al., 2006; Sutterwala et al., 2006). Notably, IL-18 has previously been associated with protection against DSS-induced colitis (Takagi et al., 2003). We therefore determined the amounts of IL-1β and IL-18 in serum of DSS-treated animals. IL-1β amounts in serum of wild-type, ASC−/− and caspase-1−/− mice barely rose above those of untreated animals at the three time points analyzed (days 1, 3 and 7; data not shown). Similarly, IL-1β amounts produced by colonic tissue from DSS-fed wild-type mice remained below 200 pg/ml, although caspase-1 deficient cells secreted even less IL-1β (Supplementary Fig. 4A). Unlike IL-1β, IL-18 was highly induced in the serum of DSS-treated wild-type mice, but not in ASC−/− and caspase-1−/− mice (Fig. 4A). Local IL-18 production in the colon was also induced in response to DSS-treatment as evidenced by the markedly increased IL-18 immunoreactivity (Fig. 4B). In agreement with an important role for IL-18 downstream of the Nlrp3 inflammasome, colons of caspase-1−/− mice contained significantly less mature IL-18 relative to DSS-fed wild-type mice (Fig. 4C). The results of the bone marrow chimera studies (Fig. 2) suggested that cells of the colonic mucosa represent a critical site of Nlrp3 inflammasome activation during DSS-induced colitis. To provide additional support for the colonic mucosa as an important site for Nlrp3 inflammasome activation, we determined the levels of mature IL-18 produced by isolated colonic epithelial cells. As in total colon extracts (Fig. 4C), colonic epithelial cells isolated from DSS-fed caspase-1−/− mice produced significantly less mature IL-18 than those of wild-type mice (Fig. 4D). Purity of the epithelial cells in the isolated population of colonic epithelia was confirmed by epithelial cell marker cytokeratin-18 as shown in supplementary fig 4B. Finally, we tested the role of IL-18 in protection against DSS-induced colitis. To this end, DSS-fed caspase-1−/− mice received a daily injection of saline or 1 μg recombinant IL-18 for 7 consecutive days. In agreement with an important role for IL-18 downstream of the Nlrp3 inflammasome, caspase-1−/− mice treated with recombinant IL-18 lost significantly less body weight when compared to those receiving PBS (Fig. 4E). Thus, Nlrp3 inflammasome signaling through IL-18 confers protection against DSS-induced colitis.
IL-18 has been linked to repair and restitution of ulcerated epithelium (Reuter and Pizarro, 2004), and colitis was previously shown to be more severe under conditions in which epithelial cell integrity is compromised (Rakoff-Nahoum et al., 2004). We therefore investigated the role of the Nlrp3 inflammasome in maintaining epithelial integrity in the gut. The intestinal barrier permeability in Nlrp3−/− and caspase-1−/− mice appeared normal prior to DSS treatment (Fig. 5A). However, the Nlrp3 inflammasome is important for regulation of gastrointestinal permeability after DSS-induced injury because significantly more FITC-dextran was recovered in serum of DSS-treated Nlrp3−/− and caspase-1−/− mice (Fig. 5A).
The decreased barrier function in the absence of Nlrp3 inflammasome signaling could be explained by increased apoptosis of epithelial cells and/or decreased cell proliferation. We first characterized the extent of apoptosis by terminal deoxynucleotidyl transferase mediated dUTP nick end labeling (TUNEL)-staining. The number of TUNEL-positive cells in colonic tissue of DSS-treated- Nlrp3−/− and caspase-1−/− mice was comparable to that of wild-type mice (data not shown), indicating that the absence of Nlrp3 inflammasome signaling does not affect apoptosis. 5′-bromo-2′-deoxy-uridine (BrdU)-staining was subsequently used to determine the role of the Nlrp3 inflammasome in epithelial cell proliferation. The epithelial crypts of DSS-treated Nlrp3−/− and caspase-1−/− mice presented with significantly less BrdU-positive cells (Fig. 5B, C). Untreated wild-type, Nlrp3−/− and caspase-1−/− mice all showed comparable amounts of BrdU staining in colonic crypts, suggesting that the Nlrp3 inflammasome is specifically required for epithelial cell proliferation after DSS-induced injury. Therefore, activation of the Nlrp3 inflammasome induces a compensatory proliferative response of epithelial cells in order to preserve the integrity of the epithelial layer during DSS-induced colitis.
It is well-established that commensal microflora in the lumen of the colon play an essential role during intestinal inflammation (Rembacken et al., 1999; Sutherland et al., 1991; Turunen et al., 1998). In addition, a functional Nlrp3 inflammasome may be required to mount a proper immune response to prevent commensal overgrowth. We therefore asked whether the profound disruption of the epithelial barrier in the colon of DSS-fed Nlrp3−/− and caspase-1−/− mice caused commensal overgrowth and bacteremia. To this end, mice were administered 3% DSS for 7 days and the number of colony-forming units (CFUs) in different tissues was determined at day 9. Significantly more bacteria were counted in the stool, liver, colon and mesenteric lymph nodes (MLN) of Nlrp3−/− and caspase-1−/− mice relative to DSS-fed wild-type mice (Fig. 6A). Increased bacteremia in Nlrp3−/− and caspase-1−/− mice was due to DSS-treatment because untreated mice showed similar bacterial counts in the stool and colon (Supplementary Fig. 5A), and their systemic organs were devoid of bacteria (data not shown). Systemic dissemination of bacteria and bacterial components triggers an exuberant cytokine and chemokine inflammatory response. To gain additional evidence of bacteremia, we measured a variety of cytokines and chemokines in serum of DSS-fed Nlrp3−/− and caspase-1−/− mice. In agreement with the increased bacterial dissemination in Nlrp3−/− and caspase-1−/− mice, the levels of the chemokines eotaxin, G-CSF, KC and MCP-1 were all significantly higher in serum of DSS-fed Nlrp3−/− and caspase-1−/− mice (Fig. 6B–E). In addition, serum levels of the pro-inflammatory cytokines IL-6 and TNF-α were also dramatically higher in Nlrp3−/− and caspase-1−/− mice when compared to wild-type mice (Fig. 6F, G).
We also assessed local cytokine and chemokine production in colon tissue and found these to be consistent with those in serum. The amounts of KC, eotaxin, G-CSF, MCP-1 and IL-6 were all higher in colons of Nlrp3−/− and caspase-1−/− mice relative to those of DSS-fed wild-type mice (Supplementary Fig. 5B). To characterize the immune cells responsible for the increased production of chemokines and cytokines in the colon, we examined the expression of cell surface markers on mononuclear cells that infiltrated the lamina propria and submucosa. Significantly increased numbers of neutrophils and macrophages (F4/80+ cells) were observed in the colon of DSS-fed Nlrp3−/− and caspase-1−/− mice (Supplementary Fig. 5C, left panels). In contrast, CD3 (T cell) and CD45R (B cell) staining were not significantly different in wild-type and inflammasome deficient mice (Supplementary Fig. 5C, right panels).
These results suggest that the increased DSS-induced morbidity and lethality in the absence of Nlrp3 inflammasome signaling may be caused by commensal overgrowth and bacteremia after the breach of the intestinal barrier. An exaggerated immune response to these commensal bacteria may further exacerbate disease severity. To address the role of commensal bacteria in the increased colitis severity in inflammasome deficient mice, we examined whether clinical parameters of DSS-induced colitis could be ameliorated with antibiotics. Nlrp3−/− mice were administered a 3% DSS-solution alongside a combination of the selective antibiotics metronidazole, neomycin and vancomycin from day 2 on. Disease severity was compared to Nlrp3−/− mice that were fed a 3% DSS-solution without antibiotics. A dramatic improvement in the clinical scores of the antibiotic-treated arm was observed over Nlrp3−/− mice that did not receive antibiotics (Supplementary Fig. 6). For instance, body weight loss in the antibiotics-treated arm was around 6%, whereas the group that was refused antibiotics presented with a loss of more than 20%. Prominent improvements in other clinical features including stool consistency and rectal bleeding were also noted for antibiotics-treated Nlrp3−/− mice. These marked improvements prompted us to examine the affect of antibiotics treatment on mortality after administration of a 4% DSS-solution. As before, ~ 80% of placebo (PBS)-treated Nlrp3−/− mice had died 2 weeks after DSS administration. In contrast, all Nlrp3−/− mice that were co-administered antibiotics remained alive by the end of the experiment (data not shown). These results indicate that overgrowth of colonic microflora contributed significantly to the increased DSS-induced morbidity and lethality of Nlrp3−/− mice.
The intestine is in permanent contact with billions of bacteria belonging to the normal intestinal flora. It is widely believed that the loss of mucosal tolerance may underlie the development of chronic intestinal inflammation. This loss of tolerance may be due to a breakdown in barrier function, triggering an abnormal immune response of local immune cells in response to components of commensal flora (Strober et al., 2002). The role of innate immune signaling in intestinal inflammation has only recently begun to be unraveled. TLR2−/− and TLR4−/− mice were significantly more susceptible to DSS-induced colitis (Fukata et al., 2005; Rakoff-Nahoum et al., 2004), suggesting a protective role for TLRs in DSS-induced colitis. In addition to TLRs, several members of the cytosolic NLR family have been identified as key regulators of cytokine production (Kanneganti et al., 2007). Notably, the NLR protein CARD15/NOD2 was the first gene to be associated with Crohn’s disease (Hugot et al., 2001; Ogura et al., 2001). Recently, decreased expression of Nlrp3 has been associated with Crohn’s disease (Villani et al., 2009). Whereas NOD2 mediates activation of NF- B and MAP kinases (Girardin et al., 2003; Inohara et al., 2003), the NLR protein Nlrp3 is involved in activation of the cysteine protease caspase-1 in a large cytosolic protein complex named the ‘inflammasome’ (Kanneganti et al., 2007; Lamkanfi et al., 2007).
We show here that Nlrp3−/− mice were significantly more susceptible to DSS-induced colitis. Similar to Nlrp3−/− mice, ASC−/− and caspase-1−/− mice were more sensitive to colitis-associated body weight loss, diarrhea, rectal bleeding and mortality during both the acute and chronic phase of disease, indicating a key role for the Nlrp3 inflammasome in protection against DSS-induced colitis. The role of the Nlrp3 inflammasome in protection against colitis is not limited to the DSS-induced model because Nlrp3−/− mice also suffered from increased body weight loss, diarrhea and reduced colon length in the acute TNBS-induced colitis model. Oral administration of DSS and TNBS is directly toxic to the gut and causes crypt destruction, mucosal erosion and ulceration. Epithelial damage induces a localized repair response characterized by increased division of stem cells at the base of crypts to replace damaged enterocytes (Radtke and Clevers, 2005). IL-18 production by the Nlrp3 inflammasome in colonic epithelial cells was identified as a crucial mediator of repair of the mucosal barrier and protection against DSS-induced colitis. Indeed, IL-18 has previously been associated to repair and restitution of ulcerated epithelium (Reuter and Pizarro, 2004). Mature IL-18 generated by the Nlrp3 inflammasome may subsequently bind to the IL-18R expressed on intestinal epithelial cells and local immune cells in the gut to exert its functions. Notably, also the TLR4/MyD88 signaling axis has been implicated in maintenance of epithelial cell homeostasis in the gut and protection against DSS-induced colitis (Fukata et al., 2005; Rakoff-Nahoum et al., 2004). This suggests that MyD88 contributes to epithelial cell homeostasis in the gut both at the level of TLR4 signaling and downstream of the IL-18R. In addition to IL-18, the cytokines IL-11 and IL-22 have been identified as important regulators of gastrointestinal mucosal biology (Keith et al., 1994; Zenewicz et al., 2008). It remains to be determined whether these cytokines operate in a hierarchical cascade or interact in a network of parallel pathways to confer protection against destruction of the mucosal barrier.
Earlier studies with caspase-1−/− mice and the caspase-1 inhibitor pralnacasan suggested a detrimental rather than a protective role for caspase-1 in DSS-induced colitis (Bauer et al., 2007; Loher et al., 2004; Siegmund et al., 2001b). However, our observation that caspase-1−/− mice are more susceptible to DSS-induced colitis is in agreement with a growing body of evidence suggesting a protective role for Nlrp3 inflammasome-mediated IL-18 production during colitis. Firstly, mice lacking the other inflammasome components Nlrp3 and ASC were also more susceptible to DSS-induced colitis. Secondly, both Il-18−/− and Il-18r−/− mice were shown to display increased susceptibility to DSS-induced colitis, which was associated with greater lethality and more severe histopathological changes (Takagi et al., 2003). Thirdly, also Il1r−/− mice present with increased intestinal damage and histopathology during DSS-induced colitis (Lebeis et al., 2009). Finally, several previous studies reported the development of more severe DSS-induced colitis in mice lacking the adaptor protein MyD88, which is required for the production of the caspase-1 substrates IL-1β and IL-18, as well as for signaling downstream of their respective receptors (Araki et al., 2005; Fukata et al., 2005; Rakoff-Nahoum et al., 2004). Noteworthy, the results from the knockout mouse models described above are sometimes in conflict with reports using (bio)chemical approaches for neutralization of caspase-1 and IL-18. For instance, experiments in IL-18 deficient mice suggested a beneficial role for IL-18 during DSS-induced colitis (Takagi et al., 2003), whereas IL-18 neutralization with recombinant IL-18 binding protein (Sivakumar et al., 2002) and IL-18 antibodies suggested a detrimental role for IL-18 (Siegmund et al., 2001a). In addition to differences in experimental design, characteristics inherent to (bio)chemical neutralization and knockout mouse models may have contributed to the different outcomes. On the one hand, chemical and biochemical inhibitors are most suited for therapeutic intervention in patients, although they are unlikely to achieve complete neutralization of the desired target and may suffer from pleiotropic effects that could interfere with disease outcome. On the other hand, gene-targeted deletion in knockout mice is a surer approach for complete removal of the protein under study. However, the possibility that gene deletion may trigger mild developmental defects that go unnoticed, but nevertheless may influence the disease phenotype cannot be completely excluded. Thus, (bio)chemical neutralization and gene-targeted deletion approaches each have particular advantages, and both should be considered to further our knowledge on the mechanisms underlying human disease.
Nlrp3−/−, ASC−/− and caspase-1−/− mice backcrossed to C57BL/6 background for at least 10 generations have been described before (Lamkanfi et al., 2008; Thomas et al., 2009). Mice were housed in a pathogen-free facility and the animal studies were conducted under protocols approved by St. Jude Children’s Research Hospital Committee on Use and Care of Animals. All mice were male 8–10 weeks old and maintained in an SPF facility. All experiments were conducted under protocols approved by the St. Jude Children’s research Hospital Committee on Use and Care of Animals.
For survival studies, acute colitis was induced with 4% (w/v) DSS (Molecular mass 36–40 kDa; MP Biologicals) dissolved in sterile, distilled water ad libitum for the experimental days 1–5 followed by normal drinking water until the end of the experiment (day 14). The DSS solutions were made fresh on day 3. For all other experimental readouts, DSS-induced colitis was induced by feeding mice 3% (w/v) DSS during 5 days, followed by normal drinking water until the end of the experiment on day 7. For bacterial count determination, mice continued to receive a 3% DSS-solution until day 7 and bacterial numbers were determined on day 9.
Body weight, stool consistency and the presence of occult blood were determined daily up to day 7. The baseline clinical score was determined on day 1. Scoring for stool consistency and occult blood was done as described previously (Wirtz et al., 2007). Briefly, stool scores were determined as follows: 0 = well-formed pellets, 1 = semiformed stools that did not adhere to the anus, 2 = semiformed stools that adhered to the anus, 3 = liquid stools that adhered to the anus. Bleeding scores were determined as follows: 0 = no blood by using hemoccult (Beckman Coulter), 1 = positive hemoccult, 2 = blood traces in stool visible, 3 = gross rectal bleeding.
After day 7, the entire colon was excised to measure the length of the colon and the weight of cecum. Colons were washed, fixed in 10% buffered formaldehyde and embedded in paraffin. Tissue sections were stained with hematoxylin & eosin (H&E). Histology was scored by a pathologist in a blinded fashion as a combination of inflammatory cell infiltration (score 0–3) and tissue damage (score 0–3). The presence of occasional inflammatory cells in the lamina propria was scored as 0, increased numbers of inflammatory cells in the lamina propria was assigned score 1, confluence of inflammatory cells extending into the submucosa was scored as 2, and transmural extension of the infiltrate was scored as 3. For tissue damage, no mucosal damage was scored as 0, lymphoepithelial lesions were scored as 1, surface mucosal erosion or focal ulceration was scored as 2, and extensive mucosal damage and extension into deeper structures of the bowel wall was scored as 3. The combined histological score ranged from 0 (no changes) to 6 (extensive infiltration and tissue damage).
For immunohistochemistry, formalin-fixed paraffin-embedded tissues were cut into 4 μm section and slides were stained for neutrophil, macrophage, T cell and B cell using the immunoperoxidase method with anti-neutrophil, anti-F4/80, anti-CD3, and anti-CD45R/B220 antibodies, respectively. IL-18 immunostaining was performed using a rat anti-mouse IL-18 antibody (MBL).
Serum was collected from blood drawn by cardiac puncture at the indicated time points. To measure the cytokine amounts in colon tissue, a part of colon was homogenized mechanically in PBS containing 1% NP-40 and complete protease inhibitor cocktail (Roche). Mouse cytokines and chemokines in serum and colon homogenate were measured with Luminex (Bio-Rad) and ELISA (R&D Systems) assays.
Colonic epithelial cells were isolated as described before (Greten et al., 2004). In brief, colons were dissected, washed with PBS and cut into small pieces. Colon segments were incubated in HBSS supplemented with 5 mM EDTA and 0.5 mM DTT for 30 min at 37°C with gentle shaking. Cells in the supernatants were filtered through a 70 μm cell strainer and washed twice. Enrichment for colonic epithelial cells was determined as the percentage of cells staining positive for the epithelial cell-specific marker cytokeratin-18. 85–90% of isolated cells stained positive for cytokeratin-18. (Supplementary Fig. 4B).
Samples of stool, colon and liver tissue were collected in 5 ml of a 3% thioglycolate solution and homogenized. Different dilutions of the obtained suspensions were plated on blood agar and BHI agar and incubated at 37°C for 48 hours. Bacterial counts were determined by colony forming assay.
To inhibit overgrowth of commensal bacteria during DSS administration, mice were treated with selective antibiotics: metronidazole (1g/L; Sigma) for killing anaerobic bacteria, neomycin (1g/L; Sigma) for killing gram negative bacteria and vancomycin (50 mg/Kg/day; Sigma) for inhibition of gram positive staphylococci and streptococci. Antibiotics treatment was started at day 2 after DSS administration and continued until day 9. Metronidazole and neomycin was added in drinking water, and vancomycin was given by oral gavage once daily.
Bone marrow transfer was used to create Nlrp3−/− chimera mice wherein the genetic deficiency of Nlrp3 was confined to either circulating cells (Nlrp3−/−> WT chimera) or non-hematopoietic tissue (WT > Nlrp3−/−). Briefly bone marrows were collected from femur and tibia of congenic WT (expressing CD45.1 leukocyte antigen) or Nlrp3−/− (expressing CD45.2 leukocyte antigen) donor mice by flushing with HBSS. After several washing steps, cells were resuspended in PBS at a concentration of 1×108/ml. 100 μl of this cell suspension was injected retro-orbitally in irradiated donor mice. 4 chimera groups were generated WT > WT (WT cells expressing CD45.1 into WT expressing CD45.2); WT > Nlrp3−/− (WT cells expressing CD45.1 into Nlrp3−/− expressing CD45.2); Nlrp3−/− > Nlrp3−/− (Nlrp3 expressing CD45.2 cells into Nlrp3−/− expressing CD45.2) and Nlrp3−/− > WT (Nlrp3−/− cells expressing CD45.2 into WT expressing CD45.1). The use of CD45.1-expressing congenic mice facilitated verification of proper reconstitution in the chimera mice. Bone marrow reconstitution was verified after 5 weeks by staining for CD45.1 and CD45.2 in blood cells using FITC-conjugated anti-CD45.1 and PE-conjugated anti-CD45.2. 7 weeks after bone marrow transfer, mice were fed with 3% DSS for 5 days. Body weight change, stool consistency and rectal bleeding were monitored daily. At day 7, mice were sacrificed to collect colon tissue for H&E staining.
In vivo assay to assess epithelial barrier permeability was performed using an FITC-labeled Dextran method as described (Furuta et al., 2001). Briefly, food and water were withdrawn and mice were gavaged with permeability tracer FITC-dextran (Mw 4000; Sigma-Aldrich) at a concentration 60 mg/100 g body weight. Blood was collected by heart puncture and FITC-dextran amount in serum was measured with a fluorescence spectrophotometer setup with emission and excitation wavelengths of respectively 490 nm and 520 nm. FITC-dextran concentration was determined from standard curves generated by serial dilution of FITC-dextran.
The number of proliferating cells in intestinal epithelium was detected by immunoperoxidase staining for thymidine analougue 5′-bromo-2′deoxyuridine (BrdU) as described (Rakoff-Nahoum et al., 2004). In brief, 1 mg/ml BrdU in PBS was injected intraperitoneally. 2 h later, colon tissue was collected and 4 cm of distal colon was fixed in 10% neutral buffered formalin and embedded in paraffin. Immunohistochemistry was performed using an in situ BrdU staining kit (BD Bioscience). Tissues were counterstained with hematoxylin. The number of BrdU-positive cells per intact and well-oriented crypt was determined.
Data are represented as mean ± SEM. Differences in group survival and bacteremia were analyzed with the Kaplan-Meier test using Prism5 (GraphPad Software). In all other cases, statistical significance was determined by Student’s t-test. P < 0.05 was considered statistically significant.
We thank Anthony Coyle, Ethan Grant, John Bertin (Millennium Pharmaceuticals), Gabriel Nuñez (University of Michigan) and Richard Flavell (Yale) for generous supply of mutant mice. We thank Deshani Perera and Gordon Johnson for technical assistance. We are very grateful to Jessica Woods and the staff in the St. Jude Animal Resource Center for mouse colony management. This work was supported by National Institute of Health Grant AR056296, a Cancer Center Support Grant (CCSG 2 P30 CA 21765) and the American Lebanese Syrian Associated Charities (ALSAC) to T-D.K. ML is supported by the Fonds voor Wetenschappelijk Onderzoek-Vlaanderen.
The authors declare that they have no competing financial interests.