Caspase-1-Mediated Activation of Interleukin-1β (IL-1β) and IL-18 Contributes to Innate Immune Defenses against Salmonella enterica Serovar Typhimurium Infection

Abstract

Caspase-1 (Casp-1) mediates the processing of the proinflammatory cytokines interleukin-1β (IL-1β) and IL-18 to their mature forms. Casp-1-deficient mice succumb more rapidly to Salmonella challenge than do wild-type animals. Both Casp-1 substrates, IL-18 and IL-1β, are relevant for control of Salmonella enterica serovar Typhimurium. We used IL-18^−/− and IL-1β^−/− mice in addition to administration of recombinant IL-18 to Casp-1^−/− mice to demonstrate that IL-18 is important for resistance to the systemic infection but not for resistance to the intestinal phase of the infection. This suggests that IL-1β is critical for the intestinal phase of the disease. Thus, we show that Casp-1 is essential for host innate immune defense against S. enterica serovar Typhimurium and that Casp-1 substrates are required at distinct times and anatomical sites.

Caspases are evolutionarily conserved cysteine proteases that induce apoptosis. Although apoptosis is generally considered to be immunologically silent, caspase-1 (Casp-1) processes the proinflammatory cytokines interleukin-1β (IL-1β) and IL-18 to their mature forms (10). Casp-1 is, therefore, unique because it is both proapoptotic and proinflammatory (18). IL-1β and IL-18 are involved in immune protection, but they can also cause endotoxemia (10). Casp-1-deficient mice exhibit altered susceptibility to a variety of pathogens (17, 21, 30, 35), in addition to being resistant to endotoxic shock (15).

Salmonella enterica serovar Typhimurium is a gram-negative bacterium that causes a systemic typhoid-like disease and serves as an experimental model for human typhoid. Shigella flexneri is also an enteropathogen that causes acute localized inflammation in the colons of dysenteric patients. Both pathogens induce Casp-1-dependent cell death in macrophages and dendritic cells (4, 8, 16, 23, 37, 41).

In Salmonella infections, cell death is mediated primarily by the bacterial effector protein SipB, which is encoded in the chromosomal locus Salmonella pathogenicity island 1 (SPI-1). SPI-1 codes for both regulatory and structural components of a type III secretion system, as well as for secreted effectors and their chaperones (14). In addition to inducing host cell death, SPI-1 mediates host cell invasion, autophagy, and inflammation (2, 11, 13, 28). In Shigella infections, the bacterial virulence factor IpaB, a close relative of SipB, is directly linked to the initiation of inflammation through Casp-1 (30). In infected macrophages, Shigella activates Casp-1, which proteolytically processes IL-1β and IL-18 and stimulates an acute inflammatory response. Our previous work indicated that Salmonella infection in vivo induces a Casp-1-dependent proinflammatory response which may be dependent on the release of mature IL-1β and IL-18. In order to determine whether Casp-1-mediated inflammation through IL-1β and IL-18 is a common pathway in microbial infections, we investigated the link between the activation of Casp-1 and its substrates in S. enterica serovar Typhimurium pathogenesis.

In the present study, we compared the courses of S. enterica serovar Typhimurium infection in Casp-1-, IL-1β-, and IL-18-deficient mice. Mice were bred under specific-pathogen-free conditions, and in all experiments sex- and age-matched animals were used. Breeding pairs of homozygous Casp-1^−/−, IL-1β^−/−, and IL-18^−/− mice were kindly provided by BASF (15), D. Chaplin (31), and K. Takeda (34), respectively. Casp-1^−/− animals with a mixed C57BL/6 × 129Sv/J background were obtained and backcrossed 10 times into the C57BL/6 background. Mice that were both IL-1β deficient and IL-18 deficient were generated by intercrossing the single-knockout strains. Mice were genotyped by PCR to determine the presence of the knockout allele, as well as the wild-type or mutant Nramp1 allele (40). Depending on the Nramp1 status of the immunodeficient animals analyzed, C57BL/6 animals (Nramp1^S/S) or F1 C57BL/6 × 129Sv/J animals (Nramp^R/S) were used as controls. The statistical significance of the results of in vivo analyses was determined by using the Mann-Whitney U test for bacterial colonization and the chi-square test for mouse survival experiments.

Prior to in vivo analysis, we examined whether the Casp-1 processed cytokines are involved in Casp-1-dependent host cell death pathways. Salmonella-induced death of macrophages from wild-type, Casp-1-null, and IL-18-null animals was monitored by measuring lactate dehydrogenase release using a Cytotox cell death kit (Promega). The virulent S. enterica serovar Typhimurium strain SL1344 efficiently killed wild-type macrophages, while the noninvasive SPI-1⁻ mutant hilA::mTn5Km2 (12, 23) was not cytotoxic for any of the macrophages tested (Fig. 1). Macrophages from IL-18^−/− mice were susceptible to Salmonella-induced cell death; similar percentages of wild-type and IL-1β^−/− macrophages were killed in an SPI-1-dependent manner (20; data not shown). Thus, Casp-1 substrates are dispensable for Casp-1-dependent cell death pathways.

FIG. 1.

Casp-1-deficient macrophages are resistant to S. enterica serovar Typhimurium-induced macrophage death. Macrophages from wild-type, Casp-1-deficient, and IL-18-deficient mice were infected with SL1344 (solid bars) or the noninvasive SPI-1 mutant hilA::mTn5Km2 (open bars). The multiplicity of infection was 100:1. Macrophage death was recorded 3 h postinfection by monitoring the culture supernatants for lactate dehydrogenase activity, using a Cytotox96 cell death kit (Promega). Experiments were performed in triplicate. Means and standard deviations from a representative experiment are shown.

To determine the relationship between Casp-1 and its cytokine substrates during Salmonella infection, we monitored the survival of wild-type and gene-deficient animals following lethal oral challenge with wild-type S. enterica serovar Typhimurium. Casp-1^−/− mice succumbed to infection more rapidly than wild-type mice succumbed (Fig. 2A). The increased susceptibility was reflected in a reduced median survival time (6 days for Casp-1^−/− mice [P = 0.0001], compared to 8 days for wild-type animals). While Casp-1^−/− mice exhibited increased susceptibility to various pathogens, this result is surprising, since we previously reported that Casp-1^−/− mice are more resistant to oral infection with S. enterica serovar Typhimurium (21). We suggest that the observed differences in susceptibility resulted from the fact that independently generated Casp-1^−/− mouse lines with different genetic backgrounds were used in the two studies (15, 33).

FIG. 2.

Casp-1-deficient mice rapidly succumb to S. enterica serovar Typhimurium infection. Mice were orally inoculated with S. enterica serovar Typhimurium SL1344, and survival was monitored. (A) Mice with the Nramp1-susceptible C57BL/6 background were challenged with 10⁸ CFU bacteria. Casp-1^−/−, IL-1β^−/−, IL-18^−/−, IL-1β^−/−/IL-18^−/−, and wild-type mice were used. (B) Mice with the Nramp1-resistant 129Sv/J background were inoculated with 2 × 10¹⁰ CFU. ▪, wild-type mice; ○, Casp1^−/− mice. Each group contained 9 to 12 mice.

The gene locus Nramp1 (Slc11a1) mediates resistance in mice to several intracellular pathogens, including S. enterica serovar Typhimurium (38). Our knowledge about Salmonella pathogenesis is derived mainly from experimental infections in susceptible mouse strains (i.e., BALB/c or C57BL/6 mice), which carry a mutant Nramp1 allele. Indeed, some mouse strains that carry a wild-type Nramp1 allele can be up to 1,000-fold more resistant to S. enterica serovar Typhimurium infection than strains harboring the mutant allele are (39) and can actually become persistently infected with Salmonella (3, 22). To investigate whether the Nramp1 status influences the altered susceptibility of Casp-1^−/− animals to Salmonella infection, we challenged Casp-1^−/− and control mice carrying a Nramp1 wild-type allele with a high oral dose of S. enterica serovar Typhimurium SL1344 and monitored the survival. Consistent with the results obtained for the Nramp1 mutant C57BL/6 background (Fig. 2), Casp-1 deficiency also led to an increase in susceptibility in the context of the Nramp1 wild-type allele. While F1 C57BL/6 × 129Sv/J control animals survived an oral challenge with 2 × 10¹⁰ CFU, Casp-1-deficient mice succumbed to an infection with a median survival time of 20 days (P = 0.0004) (Fig. 2B). These data show that in the absence of Casp-1, the susceptibility to S. enterica serovar Typhimurium infection increases independent of the Nramp1 status of the infected animals.

To examine the role of the Casp-1 substrates IL-1β and IL-18 in the host response to Salmonella, the survival of Salmonella-infected cytokine-deficient mice was analyzed (Fig. 2A). Both IL-1β and IL-18 contributed to the control of Salmonella infection. IL-1β^−/− mice succumbed to infection more rapidly than did C57BL/6 mice (P = 0.0402), and IL-18^−/− mice died even more rapidly (P < 0.0001 for a comparison with C57BL/6 mice; P = 0.0113 for a comparison with IL-1β^−/− mice). In addition, no significant differences were observed when the mortalities of Casp-1^−/−, IL-18^−/−, and IL-1β/IL-18^−/− mice were compared, indicating that IL-18 is the Casp-1 substrate that dominates host resistance against Salmonella infection.

That IL-18 plays an important role in innate immunity against S. enterica serovar Typhimurium was also apparent when the bacterial burdens of infected organs in wild-type, IL-1β^−/−, IL-18^−/−, IL-1β^−/−/IL-18^−/− and Casp-1^−/− animals were quantified. A significant increase in the bacterial load (P ≤ 0.01) was evident in all organs from all gene-deficient mice tested compared to the organs from wild-type mice on day 5 postinfection (Fig. 3). The increased susceptibility of Casp-1^−/− mice was reflected in the average 30-fold-higher bacterial loads in infected Peyer's patches (PP), mesenteric lymph nodes (MLN), and spleens than in the infected organs of wild-type animals (Fig. 3). The organs of infected IL-1β^−/− mice contained lower numbers of bacteria than the spleens (P = 0.038) and MLN (P = 0.026) of IL-1β^−/−/IL-18^−/− and Casp-1^−/− animals, respectively, contained. In contrast, the bacterial burdens in organs of infected IL-18-deficient mice were not significantly different than the bacterial burdens in organs of infected IL-1β^−/−/IL-18^−/− and Casp-1^−/− animals. Together, these data show that, although IL-1β contributes to host defense against salmonellosis, IL-18 is the predominant Casp-1 substrate that mediates resistance to oral S. enterica serovar Typhimurium infection in the systemic phase of the disease.

FIG. 3.

Caspase-1-mediated activation of IL-1β and IL-18 increases the susceptibility of Casp-1-deficient mice to oral S. enterica serovar Typhimurium infection. Mice (seven mice per group) were orally inoculated with 5 × 10⁷ CFU S. enterica serovar Typhimurium SL1344, and the bacterial burdens in infected organs were determined on day 5 postinfection. The numbers of bacteria in infected organs were expressed per PP or per g of tissue for MLN and spleens. The bacterial counts for PP, MLN, and spleens of Casp-1^−/− mice were significantly higher than the bacterial counts for the organs of wild-type mice (P = 0.0041, P = 0.0012, and P = 0.0111, respectively).

To further examine the role of IL-18 in Casp-1-mediated host defense mechanisms, we treated Casp1^−/− mice with recombinant IL-18 (rIL-18) before S. enterica serovar Typhimurium infection and determined whether exogenous cytokine administration could rescue the deficiency in pathogen control (Fig. 4). The levels of viable bacteria in infected organs of Casp-1-deficient mice were returned to almost wild-type levels by adding rIL-18 during the initial phase of oral S. enterica serovar Typhimurium challenge (Fig. 4A). There was no significant difference in the bacterial organ load between rIL-18-treated Casp-1^−/− mice and wild-type or IL-1β-deficient animals (Fig. 4A and data not shown), emphasizing that the lack of IL-18 is primarily responsible for the increased susceptibility of Casp-1^−/− mice to Salmonella infection. Interestingly, exogenous IL-18 corrected S. enterica serovar Typhimurium susceptibility in Casp-1^−/− mice only at systemic sites (MLN and the spleen), whereas in infected PP of rIL-18-treated Casp-1^−/− mice there was no significant reduction in the bacterial burden. We suggest that the observed lack of mucosal protection against Salmonella challenge following treatment with rIL-18 may have been related to the finding that IL-1-induced inflammation is important in Salmonella pathogenesis at mucosal sites of infection. Thus, both Casp-1 substrates (IL-18 and IL-1β) are relevant for control of S. enterica serovar Typhimurium infection at early stages in the intestine, while IL-18 is the predominant molecule for host defense in the systemic phase of the infection.

FIG. 4.

Exogenous rIL-18 corrects S. enterica serovar Typhimurium susceptibility in Casp-1^−/− mice. Wild-type and gene-deficient animals were infected with 5 × 10⁷ CFU SL1344 orally (A) or with 250 CFU SL1344 intraperitoneally (B). For one group of Casp-1^−/− mice, 1.5 μg rIL-18 (Biosource) was injected intraperitoneally daily starting the day prior to Salmonella challenge. The geometric means and standard errors from representative experiments are shown. The number of bacteria in infected organs was expressed per PP or per g of tissue for MLN, spleens, and livers. (A) Bacterial burdens in infected organs (five mice per group) on day 4. The bacterial counts in rIL-18-treated Casp-1^−/− mice were significantly reduced in MLN and spleens (P = 0.0357). (B) Bacterial counts in livers and spleens on day 3 postinfection (four mice per group). Pretreatment of Casp-1^−/− mice with rIL-18 significantly decreased the bacterial burden (P = 0.0286).

The contribution of IL-18 to host resistance in systemic S. enterica serovar Typhimurium was also evident when we compared the bacterial burdens of infected organs in wild-type, IL-1β^−/−, IL-18^−/−, and Casp-1^−/− animals after intraperitoneal challenge with a lethal dose of S. enterica serovar Typhimurium SL1344 (Fig. 4B). On day 3 postinfection, the bacterial counts in the spleens and livers were similar for wild-type and IL-1β^−/− mice. In contrast, the bacterial burdens in the infected organs of IL-18- and Casp-1-deficient mice were significantly increased compared to the burdens in the wild-type controls. Pretreatment of Casp-1^−/− mice with rIL-18 before Salmonella challenge significantly reduced the levels of viable bacteria in infected organs (P = 0.0286) to levels that were close to wild-type levels (Fig. 4B). While these results emphasize the importance of Casp-1 and IL-18 in the innate immunity against S. enterica serovar Typhimurium independent of the route of inoculation, they also suggest that IL-1β does not contribute in a major way to the control of the systemic phase of an S. enterica serovar Typhimurium infection. Although IL-1 has been shown to increase host resistance to various microorganisms (5, 6), our data are in accordance with previous reports which showed that IL-1 is required for host resistance to S. enterica serovar Typhimurium challenge only in mouse strains with the wild-type Nramp1 background and not in animals carrying the mutant Nramp1 allele (24), such as the C57BL/6 strain used in this study.

While originally described as a gamma interferon-inducing factor (27), IL-18 has multiple biological activities that participate in both innate and acquired immune responses (1). As a member of the IL-1 family, IL-18 functions as a proinflammatory cytokine. However, it is also related to IL-12 with respect to its ability to drive Th1 responses and to enhance cell-mediated cytotoxicity. Previous studies to identify the role of IL-18 in Salmonella infections generated contrasting results. It was reported that mice treated with anti-IL-18 antibody were more susceptible to S. enterica serovar Typhimurium infection (7, 19). However, it was also reported that Salmonella caused a reduction in the IL-18 message in infected macrophages (9). The data obtained in this study clearly show that IL-18 plays a crucial role in the control of Salmonella infections. In addition, we demonstrated that IL-1β contributes to host resistance to oral S. enterica serovar Typhimurium infection and that IL-18 is the Casp-1 substrate responsible for the increased systemic susceptibility of Casp-1^−/− mice to infection.

IL-18 not only is a key factor in host resistance to Salmonella but also plays a role in endotoxemia (26). Casp-1-deficient mice are resistant to lethal endotoxemia induced by lipopolysaccharide (LPS) from Escherichia coli or S. enterica serovar Typhimurium (15, 25). Anti-IL-18-treated animals were also protected completely against E. coli LPS challenge or partially against S. enterica serovar Typhimurium LPS challenge (25). Lethal endotoxemia is widely used as an experimental model for gram-negative sepsis. However, different pathways of endotoxemia have been suggested depending on the source of LPS (26). To directly analyze the role of Casp-1 in Salmonella-induced septic shock, we challenged gene-deficient and control animals with live attenuated Salmonella. Immunocompetent animals can control oral infections with a high dose of the aroA-deficient strain S. enterica serovar Typhimurium SL7207 (29, 36), whereas systemic challenge causes septic shock (Fig. 5). Wild-type and IL-1β^−/− mice inoculated intraperitoneally with 10⁸ CFU S. enterica serovar Typhimurium SL7207 succumbed to gram-negative sepsis within 2 to 3 days after injection. In contrast, Casp-1^−/− mice, as well as IL-18-deficient mouse strains, were resistant to such a shock. Infected animals did not succumb until much later, and the mean survival times were significantly increased up to days 8 to 10 for Casp-1^−/−, IL-1β^−/−/IL-18^−/−, and IL-18^−/− mice (P < 0.0001). While previous studies suggested that endotoxemia induced after challenge with Salmonella LPS is due to a network of several proinflammatory cytokines (26), our data clearly indicate that IL-18 is a major player in Salmonella-induced gram-negative sepsis.

FIG. 5.

IL-18 deficiency protects against septic shock. Mice were injected intraperitoneally with 10⁸ CFU of live attenuated S. enterica serovar Typhimurium SL7207, and survival was monitored. IL-18-proficient mice (C57BL/6 and IL-1β^−/− mice) rapidly succumbed to septic shock, while IL-18-deficient animals (Casp-1^−/−, IL-18^−/−, and IL-1β^−/−/IL-18^−/− mice) were resistant. Each group contained 9 to 12 mice.

Our results demonstrate that Casp-1 is required for control of oral infection with wild-type Salmonella in mice, as well as for resistance to septic shock following systemic challenge with live attenuated S. enterica serovar Typhimurium. Furthermore, we found that in contrast to systemic S. enterica serovar Typhimurium infection, host defense against oral Salmonella challenge requires both Casp-1 substrates, IL-1β and IL-18. However, IL-18 is the predominant Casp-1 substrate that mediates control of Salmonella infection in the systemic phase. These data contrast with our previous findings and may have been due to a difference in mouse strain backgrounds. Our previous studies were conducted with independently generated Casp-1^−/− mice with a B10.RIII background. The B10.RIII mouse background strain has been shown to exhibit a more polarized Th1 response (32) and may confer higher levels of resistance to Salmonella. However, additional studies are required to elucidate the reasons for the different findings.

The data presented here suggest that the two Casp-1 substrates are relevant at different stages of infection. Since different roles for IL-1β and IL-18 were reported previously for Shigella-mediated activation of Casp-1 (30), we postulate that Casp-1-mediated activation of proinflammatory cytokines is a common pathway in microbial infections. Together, our results indicate the importance of Casp-1 substrates for innate immunity against S. enterica serovar Typhimurium.