The RD1 Locus in the Mycobacterium tuberculosis Genome Contributes to Activation of Caspase-1 via Induction of Potassium Ion Efflux in Infected Macrophages

Abstract

A genomic locus called “region of difference 1” (RD1) in Mycobacterium tuberculosis has been shown to contribute to the generation of host protective immunity as well as to the virulence of the bacterium. To gain insight into the molecular mechanism, we investigated the difference in the cytokine-inducing ability between H37Rv and a mutant strain deficient for RD1 (ΔRD1). We found that RD1 is implicated in the production of caspase-1-dependent cytokines, interleukin-18 (IL-18) and IL-1β, from infected macrophages. The expression of these cytokines was similarly induced after infection with H37Rv and ΔRD1. However, the activation of caspase-1 was observed only in H37Rv-infected macrophages. The cytokine production and caspase-1 activation were induced independently of type I interferon receptor signaling events. We also found that the activation of caspase-1 was markedly inhibited with increasing concentrations of extracellular KCl. Furthermore, the production of IL-18 and IL-1β and caspase-1 activation were induced independently of a P2X7 purinergic receptor, and the inability of ΔRD1 in caspase-1 activation was compensated for by nigericin, an agent inducing the potassium ion efflux. Based on these results, we concluded that RD1 participates in caspase-1-dependent cytokine production via induction of the potassium ion efflux in infected macrophages.

Mycobacterium tuberculosis, an etiologic agent of human tuberculosis, is one of the leading threats to humans. It has been reported that M. tuberculosis still causes 9.2 million new cases of tuberculosis worldwide and 1.7 million deaths annually (49). The recent emergence of multidrug-resistant and extensively drug-resistant M. tuberculosis strains highlights the urgent need for extensive research unraveling the complex mechanism enabling the bacterium to be successfully parasitic in humans.

The protective immunity against M. tuberculosis is mediated mainly by Th1-type CD4⁺ T cells and CD8⁺ T cells. These T cells produce a large amount of cytokines, including gamma interferon (IFN-γ) and tumor necrosis factor alpha (TNF-α), resulting in the enhancement of macrophage bactericidal activity and the development of granulomas in which M. tuberculosis is killed and prevented from disseminating to the bloodstream and other tissues (12, 47). It has been demonstrated that T cells differentiate into Th1 cells in cooperation with several proinflammatory cytokines, such as interleukin-12 (IL-12), IL-18, and IFN-γ, that are produced by infiltrating dendritic cells, macrophages, and NK cells. A number of in vitro studies have shown that these cytokines are produced via recognition of pathogen-associated molecular patterns of M. tuberculosis by Toll-like receptors (TLRs) (43). However, the role of TLR-mediated signaling pathways in the protective immunity against M. tuberculosis is controversial. Abel et al. have shown that TLR4-deficient mice display reduced bacterial clearance during a long-term infection and develop chronic pneumonia (2). Drennan et al. have also reported that TLR2-deficient mice initially control an aerosol infection with M. tuberculosis, but develop increased bacterial burden and succumb to chronic pneumonia (8). On the other hand, recent studies have shown that TLRs are dispensable in the development of T-cell-mediated adaptive immunity, while myeloid differentiation protein 88 (MyD88) is required for restriction of the intracellular growth of M. tuberculosis (44). In addition, Fremond et al. (16) and Hölscher et al. (24) have shown that mice deficient for IL-1 receptor (IL-1R) succumbed to acute M. tuberculosis infection in a manner similar to mice deficient for MyD88, whereas mice deficient for TLR2, TLR4, TLR9, or Toll-IL-1R domain-containing adaptor protein could control acute M. tuberculosis infection to the same extent as wild-type mice. These findings suggest that MyD88 plays a much more prominent role in adaptive immunity than functioning as an adaptor molecule of TLRs, and the role of the MyD88-dependent IL-1R signaling pathway is necessary for induction of efficient protection against M. tuberculosis.

A genomic locus of M. tuberculosis called “region of difference 1” (RD1) was first discovered as a locus that is absent in a genome of Mycobacterium bovis BCG (30). RD1 is 9.5 kb in length and comprises nine genes, including the genes that encode the secretory proteins ESAT-6 (6-kDa early secreted antigen target) and CFP-10 (10-kDa culture filtrate protein). The other genes encode components of a secretion system that is called ESX-1 (ESAT-6 system 1). It is supposed that more than 14 proteins contribute to this secretion system (1). Although all components that are involved in ESX-1 have not yet been fully characterized, a number of proteins with known functional domains are identified. They include a putative chaperone with an AAA+ ATPase (Rv3868), a subtilisin-like serine protease (Rv3883c), and an FtsK/SpoIIIE-like ATPase (Rv3870, Rv3871). Rv3877 is predicted to be a membrane-spanning protein that could be part of the translocation pore in the cytoplasmic membrane. It has also been shown that RD1 is a critical region for bacterial virulence (6, 25, 31, 40). Furthermore, intensive studies have demonstrated the role of RD1 in the generation of protective immunity (5, 41) as well as necrosis induction (25, 27) and granuloma formation (48). To determine how RD1 contributes to the development of host immune response, we compared the cytokine-inducing ability between wild-type M. tuberculosis H37Rv and the mutant strain deficient for RD1. We found that RD1 was essential for activation of caspase-1 and subsequent secretion of IL-18 and IL-1β from macrophages infected with M. tuberculosis. The activation of caspase-1 was induced via a potassium ion efflux that is highly dependent on RD1 but independent of the P2X7 receptor. Moreover, we found that the type I interferon is not required for the activation of caspase-1 and cytokine production.

MATERIALS AND METHODS

Mice.

Female C57BL/6 and BALB/c mice were purchased from Japan SLC (Shizuoka, Japan). INF-α, INF-β, and INF-ω receptor 1 knockout mice (IFNAR1^−/− mice, on a C57BL/6 background) were kindly provided by Shigekazu Nagata (Kyoto University Graduate School of Medicine, Kyoto, Japan). P2X7 receptor knockout mice (P2X7R^−/− mice, on a C57BL/6 background) were obtained from Tatsuro Ishibashi (Kyushu University, Fukuoka, Japan). Mice were maintained under specific pathogen-free conditions and used at 7 to 9 weeks of age. All the experimental procedures were approved by the Animal Ethics and Research Committee of Kyoto University Graduate School of Medicine, Kyoto, Japan.

Bacterial strains.

M. tuberculosis H37Rv, an H37Rv mutant deficient for RD1 (ΔRD1) and an RD1-complemented strain (ΔRD1::RD1) (pYUB412::Rv3860-Rv3885c) were kindly provided by William R. Jacobs (Albert Einstein Institute, Bronx, NY) (25). These M. tuberculosis strains were grown at 37°C to the mid-log phase in Middlebrook 7H9 broth supplemented with 0.5% albumin, 0.2% dextrose, 3 μg ml⁻¹ catalase, and 0.2% glycerol. Bacteria were harvested, stirred vigorously with glass beads (3 mm in diameter), and centrifuged at 300 × g for 3 min to remove the bacterial clumps. The suspension was stored at −80°C in aliquots. After thawing, the viable bacteria were enumerated by counting the colonies after plating the diluted suspension on Middlebrook 7H10 agar plates containing 50 μg ml⁻¹ oleic acid, 0.5% albumin, 0.2% dextrose, 4 μg ml⁻¹ catalase, and 0.85 mg ml⁻¹ sodium chloride. In each experiment, bacteria were added to the macrophage culture based on the concentration after the absence of bacterial clumps was confirmed by Kinyoun staining.

Cells.

Peritoneal exudate cells of C57BL/6 and BALB/c mice were obtained by a peritoneal lavage 4 days after an intraperitoneal injection with 3 ml of thioglycolate medium (EIKEN Chemical, Osaka, Japan). Peritoneal exudate cells were washed and plated at 5.0 × 10⁵ cells well⁻¹ in 48-well plates and incubated for 3 h at 37°C in RPMI 1640 medium supplemented with 10% fetal calf serum. Nonadherent cells were removed by washes with warmed RPMI 1640 medium, and adherent cells were used as macrophages in the following experiments. Bone marrow cells were collected from tibiae of C57BL/6 mice and cultured with 100 ng ml⁻¹ mouse M-CSF (R&D Systems, Minneapolis, MN) for 5 days. After washes, adherent bone marrow-derived macrophages (BMDM) were collected and seeded at 5.0 × 10⁵ cells well⁻¹ in 48-well plates (21).

Quantitative real-time RT-PCR.

Total cellular RNA was extracted from peritoneal macrophages 9 h after infection with M. tuberculosis strains by using Nucleospin RNA II (Macherey-Nagel, Düren, Germany). RNA (0.2 μg) was treated with RNase-free DNase (Promega, Tokyo, Japan) to eliminate contaminating DNA and then subjected to reverse transcription (RT) using the SuperScript III first-strand synthesis system for RT-PCR (Invitrogen, Tokyo, Japan). Quantitative real-time RT-PCR was performed on ABI PRISM 7000 (Applied Biosystems, Tokyo, Japan) using Platinum Sybr green quantitative PCR SuperMix-uracil DNA glycosylase (Invitrogen) according to the manufacturer's instructions. The level of each cytokine mRNA expression was normalized on the basis of β-actin mRNA expression, and results were analyzed with ABI PRISM 7000 sodium dodecyl sulfate (SDS) software. The following DNA sequences were designed and used as PCR primers: Il-1β (5′-AAGCTCTCCACCTCAATGGACAG-3′, forward; and 5′-CTCAAACTCCACTTTGCTCTTGA-3′, reverse), Il-18 (5′-ACTGTACAACCGCAGTAATACGG-3′, forward; and 5′-AGTGAACATTACAGATTTATCCC-3′, reverse), and β-actin (5′-TGGAATCCTGTGGCATCCATGAAAC-3′, forward; and 5′-TAAAACGCAGCTCAGTAACAGTCCG-3′, reverse).

Cytokine production and detection of caspase-1.

Peritoneal macrophages were infected with M. tuberculosis strains at a multiplicity of infection (MOI) of 5 for 3 h. Cells were washed to remove extracellular bacteria and then incubated for 1, 9, and 21 h in the presence or absence of 500 to 2,000 NU/ml anti-IFN-β immunoglobulin G (IgG) (PBL Biomedical Laboratory, Piscataway, NJ), or 10 to 40 mM potassium chloride. Alternatively, macrophages were infected with H37RvΔRD1 at an MOI of 5 for 3 h. Cells were washed to remove extracellular bacteria and incubated for 21 h, and then nigericin (5 μM) and/or KCl (30 mM) was added and incubated for another 3 h. The culture supernatant was collected, and concentrations of cytokines were determined by enzyme-linked immunosorbent assay (ELISA) as reported previously (17, 19, 20). TNF-α, IL-6, and IL-1β were detected by using ELISA kits (eBioscience, San Diego, CA). IL-18 was detected using a pair of biotin-labeled and unlabeled monoclonal antibodies specific to IL-18 (MBL, Aichi, Japan).

In order to detect the activated form of caspase-1, 6 ml of the culture supernatants was incubated with 7 μg of rabbit anti-caspase-1 p10 IgG (Santa Cruz Biotechnology, Santa Cruz, CA) and protein G Sepharose (GE Healthcare Bio-Sciences AB, Uppsala, Sweden) to pull down caspase-1. Concurrently, infected macrophages were washed and lysed in phosphate-buffered saline containing 1% Nonidet P-40, 1 μg ml⁻¹ leupeptin, 1 μg ml⁻¹ pepstatin A, 1.5 μg ml⁻¹ aprotinin, and 2 mM dithiothreitol. The lysate was used for detection of procaspase-1. The samples were subjected to SDS-polyacrylamide gel electrophoresis and transferred to polyvinylidene fluoride membranes by electroblotting. The membrane was sequentially treated with rabbit anti-caspase-1 p10 IgG, anti-rabbit IgG conjugated with peroxidase, and ECL Plus (GE Healthcare). The bands representative of procaspase-1 and caspase-1 were detected by LAS-4000 Mini (Fujifilm, Tokyo, Japan). In addition, mature and proform types of IL-1β were detected by Western blotting using anti-IL-1β antibodies (R&D Systems) in the culture supernatant and the cell lysate, respectively. β-Actin was detected using anti-mouse β-actin monoclonal antibody (Sigma-Aldrich, Tokyo, Japan).

Statistical analysis.

Student's t test was used to determine the statistical significance of the values obtained, and a P value of <0.05 was considered to be statistically significant.

RESULTS

RD1 participates in the production of IL-18 and IL-1β, but not IL-6 or TNF-α, in macrophages infected with M. tuberculosis.

To investigate whether RD1 in the M. tuberculosis genome contributes to cytokine production, we analyzed the production of IL-1β, ΙL-18, IL-6, and TNF-α after infection with M. tuberculosis strains H37Rv, ΔRD1, and ΔRD1::RD. High levels of cytokine production were detected in peritoneal exudate macrophages of C57BL/6 mice in response to H37Rv infection at an MOI of 5 (Fig. 1A to D). The production of these cytokines was increased in a time-dependent manner, and the significant production was detected later than 12 h after infection. On the other hand, ΔRD1 did not induce such high levels of IL-1β and IL-18 production, whereas the production of TNF-α and IL-6 was comparable with that induced by H37Rv. In contrast with ΔRD1, the strain ΔRD1::RD1, an RD1-complemented strain, was capable of inducing the production of IL-1β and IL-18 as well as IL-6 and TNF-α, and the levels were almost similar to those induced by H37Rv. The results described above clearly indicate that RD1 is dispensable for TNF-α and IL-6 production but that it contributes to the production of IL-1β and IL-18 from infected macrophages.

FIG. 1.

RD1 participates in the production of IL-1β and IL-18, but not IL-6 or TNF-α, in M. tuberculosis-infected macrophages. Thioglycolate-induced peritoneal exudate macrophages of C57BL/6 mice were infected with H37Rv, ΔRD1, and ΔRD1::RD1 at an MOI of 5 for the indicated periods of time. The culture supernatant was collected, and the amounts of IL-1β (Α), IL-18 (B), TNF-α (C), and IL-6 (D) were measured by ELISA. Peritoneal exudate macrophages were infected with M. tuberculosis strains at an MOI of 1 for the indicated periods of time, and IL-1β production was measured (E). BMDM of C57BL/6 mice (F) and peritoneal exudate macrophages of BALB/c mice (G) were infected with M. tuberculosis strains at an MOI of 5 for 24 h. The culture supernatant was collected, and the amount of IL-1β was measured by ELISA. Data represent the mean ± standard deviations of triplicate assays and are representative of three independent experiments. *, a P value of <0.05 for ΔRD1-infected cells compared to either H37Rv-infected cells or ΔRD1::RD1-infected cells.

It has been shown that virulent M. tuberculosis induces a distinct response in macrophages if cells were infected with either a high or low dose (29). To rule out the possibility that the difference in cytokine production between H37Rv- and ΔRD1-infected macrophages is due to a high load of bacteria, we infected macrophages with M. tuberculosis strains at an MOI of 1 and measured the IL-1β production. Similar to the response at the high-dose infection, the response from H37Rv and ΔRD1::RD1 infection showed that cytokine production was induced as early as 12 h after infection (Fig. 1E), while ΔRD1 did not induce production, suggesting that the observed pattern of cytokine production had not resulted from an unreasonable load of bacteria. In addition, thioglycolate-induced macrophages are known to be in an activated state. We thus employed BMDM to determine whether RD1-dependent IL-1β production is observed in resting macrophages. As expected, infection with H37Rv and ΔRD1::RD1 induced a significant level of IL-1β production, but ΔRD1 infection did not (Fig. 1F). Furthermore, the requirement of RD1 in IL-1β production was observed for peritoneal macrophages of BALB/c mice, though the magnitude of the cytokine response was weaker than that of C57BL/6 macrophages (Fig. 1G).

RD1 contributes to the activation of caspase-1 in macrophages infected with M. tuberculosis.

We next measured the expression of IL-18 and IL-1β mRNAs by real-time RT-PCR after infection with M. tuberculosis strains. IL-18 mRNA was detected in unstimulated macrophages (Fig. 2A). The level was almost similar to that observed after infection with H37Rv and ΔRD1::RD1. Despite the absence of IL-18 secretion in macrophages infected with ΔRD1 (Fig. 1A), there was no difference in the level of IL-18 mRNA expression between H37Rv and this mutant. Compared to the control response (no infection), IL-1β mRNA expression was dramatically increased after infection with all three strains. This profile showed a significant contrast to the profile of secreted IL-1β (Fig. 1). These results clearly showed that though the proforms of IL-18 and IL-1β were generated after ΔRD1 infection, the mutant failed to induce the secretion of the mature forms of these cytokines.

FIG. 2.

RD1 contributes to secretion of IL-1β and IL-18 through the activation of caspase-1. Peritoneal macrophages were infected with M. tuberculosis strains at an MOI of 5 for 9 h. Total RNA was extracted and subjected to quantitative real-time RT-PCR to compare the expression levels of IL-18 (Α) and IL-1β (B). Peritoneal macrophages were infected with M. tuberculosis strains at an MOI of 5 for 24 h. The cell lysate was prepared, and the amounts of proIL-1β (C) and procaspase-1 (D) were determined by Western blotting. (C) To detect mature IL-1β, the culture supernatant was collected and Western blotting was done. As the direct detection of the activated form of caspase-1 in the supernatant was difficult, the culture supernatant was treated with anti-caspase-1 p10 antibodies plus protein G Sepharose beads to enrich caspase-1. (D) The sample was then subjected to SDS-polyacrylamide gel electrophoresis, and the relative amount of mature caspase-1 (p10) was determined by Western blotting. β-Actin was used as a loading control of the cell lysate.

Both IL-1β and IL-18 are members of the IL-1 family of cytokines and are produced as immature proteins. It has been shown that IL-1β and IL-18 are secreted after conversion into mature forms by activated caspase-1 (4, 32). Based on the level of these cytokine transcripts, it appeared that pro-IL-1β and pro-IL-18 were similarly induced after infection with M. tuberculosis strains. To determine the level of transcripts, we carried out Western blotting for IL-1β. As shown in Fig. 2C, the 35-kDa band corresponding to pro-IL-1β was similarly detected in the lysates of cells infected with three M. tuberculosis strains (Fig. 2C). We next determined whether the activation of caspase-1 was induced after infection with M. tuberculosis strains by analyzing the amount of procaspase-1 (p45) and a fragment of the activated form of caspase-1 (p10) (Fig. 2D). It has been shown that activated caspase-1 is secreted from cells along with mature IL-1β and IL-18 (36, 42). Concordantly, we detected caspase-1 in the culture supernatant but not in the cell lysate, suggesting that caspase-1 is mostly secreted after conversion from procaspase-1 to activated caspase-1 in this experimental system. Therefore, we measured the amount of procaspase-1 in the cell lysate and evaluated the activation of caspase-1 by measuring the amount of p10 in the culture supernatant. A large amount of procaspase-1 (p45) was detected in the lysate of uninfected macrophages. A similar amount of p45 was detected in macrophages infected with three M. tuberculosis strains. On the other hand, p10 was detected only in the culture supernatant of macrophages infected with H37Rv or ΔRD1::RD1 and was hardly detected in the culture supernatant of ΔRD1-infected macrophages. In proportion to the caspase-1 activation, the mature IL-1β was detected in the culture supernatant of cells infected with H37Rv and ΔRD1::RD1 (Fig. 2C). The results clearly showed that RD1 contributes to the activation of caspase-1, leading to the secretion of IL-1β and IL-18 from H37Rv-infected macrophages.

Endogenous IFN-β does not participate in the activation of caspase-1 in macrophages infected with M. tuberculosis.

Henry et al. have shown that IFN-β is necessary for the activation of caspase-1 in macrophages infected with Francisella tularensis and Listeria monocytogenes, whereas Salmonella enterica serovar Typhimurium, another intracellular bacterium, induces activation of caspase-1 independently of IFN-β (22). We examined whether IFN-β contributes to the activation of caspase-1 in macrophages infected with M. tuberculosis. We first infected macrophages with H37Rv in the presence of anti-IFN-β IgG and measured the production of IL-1β and IL-18. As shown in Fig. 3A and B, neutralization of IFN-β did not affect the production of these cytokines. The antibody employed in this study could block the secretion of IL-1β and IL-18 from macrophages infected with Listeria monocytogenes at the concentrations used in this experiment (data not shown). We also determined the effect of anti-IFN-β IgG on the activation of caspase-1. The Western blot clearly showed that the antibody did not affect the amount of p10 fragment of caspase-1 released after infection with H37Rv (Fig. 3C). The absence of an IFN-β contribution, as suggested by the findings described above, could be further confirmed by using type I IFN receptor knockout (IFNAR1^−/−) macrophages. As shown in Fig. 3D and E, the level of these cytokines produced from IFNAR1^−/− macrophages was comparable to that of wild-type macrophages. We also analyzed the activation of caspase-1 after infection with H37Rv. There was no difference in the amount of the fragment of activated caspase-1 (p10) between wild-type and IFNAR1^−/− macrophages (Fig. 3F). Therefore, we concluded that IFN-β was not necessary for caspase-1 activation in M. tuberculosis infection.

FIG. 3.

IFN-β does not contribute to the activation of caspase-1 in M. tuberculosis-infected macrophages. Peritoneal macrophages were infected with H37Rv at an MOI of 5 for 24 h in the presence or absence of anti-IFN-β IgG or control IgG. The amount of IL-1β (A) and IL-18 (B) in the culture supernatant was measured by ELISA. Data represent the mean ± standard deviations of triplicate assays and are representative of three independent experiments. Peritoneal macrophages were infected with H37Rv at an MOI of 5 for 24 h in the presence or absence of anti-IFN-β IgG (2 × 10³ NU/ml) or control IgG. Procaspase-1 (p45) in the culture supernatant and the activated form of caspase-1 (p10) in the cell lysate were detected by Western blotting (C). Peritoneal exudate macrophages from C57BL/6 and IFNAR1^−/− mice were infected with H37Rv at an MOI of 5 for 24 h. The production of IL-1β (D) and IL-18 (E) in the culture supernatant was measured by ELISA. In addition, the amounts of procaspase-1 (p45) in the cell lysate and the fragment (p10) of mature caspase-1, which was immunoprecipitated from the culture supernatant, were detected by Western blotting (F). β-Actin was used as a loading control of the cell lysate. WT, wild type; Ab, antibody.

Activation of caspase-1 in M. tuberculosis infection is dependent on potassium ion efflux.

It has been shown that nigericin, anthrax lethal toxin, monosodium urate, and ATP efficiently induce the activation of caspase-1. These reagents cause a temporal decrease in intracellular potassium concentration and then promote the activation of caspase-1 (38, 39). To know whether potassium efflux is involved in the activation of caspase-1, we examined the effect of increasing concentrations of extracellular KCl on the production of cytokines. The significant levels of IL-1β and IL-18 secretion induced after infection with H37Rv were clearly decreased with increasing concentrations of KCl (Fig. 4A and B). On the other hand, the production of TNF-α and IL-6 was not affected by even the highest concentrations of KCl (Fig. 4C and D). Using Western blot analysis, we determined the effect of extracellular KCl on the activation of caspase-1 in H37Rv-infected macrophages. As expected, the amount of p10 fragment was decreased with increasing concentrations of KCl and was diminished to the level of the noninfected control when cells were cultured with 40 mM KCl (Fig. 4E). These results suggested that the potassium ion efflux is essential for the secretion of IL-1β and IL-18 and that the inability of ΔRD1 to induce the production of these cytokines may be due to the inability of the potassium ion efflux to be induced.

FIG. 4.

Potassium ion efflux is essential for the secretion of IL-1β and IL-18 and activation of caspase-1 in M. tuberculosis-infected macrophages. Peritoneal macrophages were infected with H37Rv at an MOI of 5 for 24 h in the presence or absence of KCl at 0 mM (black columns), 10 mM (dark gray columns), 20 mM (light gray columns), and 40 mM (white columns). The amounts of IL-1β (Α), IL-18 (B), TNF-α (C), and IL-6 (D) were measured by ELISA. Data represent the mean ± standard deviations of triplicate assays. *, a P value of <0.05 for H37Rv-infected cells in the presence of 40 mM KCl compared to H37Rv-infected cells in the absence of KCl. Peritoneal macrophages were infected with H37Rv at an MOI of 5 for 24 h in the presence of increasing concentrations of KCl. The amounts of procaspase-1 (p45) in the cell lysate and the fragment (p10) of mature caspase-1, which was immunoprecipitated from the culture supernatant, were detected by Western blotting (E). β-Actin was used as a loading control of the cell lysate.

P2X7 receptor does not participate in the activation of caspase-1 in macrophages infected with M. tuberculosis.

The P2X7 receptor was identified as an important component for caspase-1 activation through promotion of potassium efflux (10, 26). Recently, it has been shown that in response to TLR agonists or infection with Staphylococcus aureus or Escherichia coli, caspase-1 activation is triggered by the addition of ATP, a signal that promotes caspase-1 activation through depletion of intracellular potassium caused by stimulation of the P2X7 receptor. On the other hand, caspase-1 activation induced by Salmonella or Listeria was not affected in macrophages deficient in the P2X7 receptor (15). In view of this reported finding, we analyzed whether the P2X7 receptor contributes to M. tuberculosis-induced caspase-1 activation by measuring the cytokine response in P2X7 receptor knockout (P2X7R^−/−) macrophages after H37Rv infection. As shown in Fig. 5, there was no difference in the secretion of IL-1β and IL-18 (Fig. 5A and B), and the activation of caspase-1 (Fig. 5C), between wild-type and P2X7R^−/− macrophages. The result indicated that the P2X7 receptor does not play any role in caspase-1 activation induced by M. tuberculosis.

FIG. 5.

P2X7 receptor does not contribute to the activation of caspase-1 in M. tuberculosis-infected macrophages. Peritoneal exudate macrophages from C57BL/6 and P2X7R^−/− mice were infected with H37Rv at an MOI of 5 for 24. The levels of IL-1β (A) and IL-18 (B) in culture supernatants were determined by ELISA. (C) The amounts of procaspase-1 (p45) in the cell lysate and the fragment (p10) of mature caspase-1, which was immunoprecipitated from the culture supernatant, were detected by Western blotting. β-Actin was used as a loading control of the cell lysate. WT, wild type.

Inability of ΔRD1 to induce production of IL-18 and IL-1β is compensated for by nigericin.

The findings described above implied that M. tuberculosis-induced caspase-1 activation is through the induction of the potassium ion efflux that is dependent on RD1 but not on the P2X7 receptor. If this is the case, the inability of ΔRD1 to induce cytokine maturation may be compensated for by the induction of the potassium ion efflux by some means. To test this possibility, macrophages were infected with ΔRD1 and then stimulated with nigericin, a potassium ionophore. The culture supernatant was collected, and the production of cytokines and the activation of caspase-1 were measured. The production of IL-1β and IL-18 was not induced by infection with only ΔRD1 or treatment with nigericin alone (Fig. 6A and B). However, a strong cytokine response was observed when ΔRD1-infected cells were stimulated with nigericin. Furthermore, the enhanced cytokine response was diminished mostly by the addition of 30 mM KCl. In proportion to the cytokine production, the amount of p10 was also increased by treatment with nigericin and was reduced by the addition of 30 mM KCl (Fig. 6C). Based on these data, we concluded that RD1 participates in the caspase-1-dependent cytokine production via induction of the potassium ion efflux in infected macrophages.

FIG. 6.

The inability of H37RvΔRD1 to induce production of IL-18 and IL-1β is compensated by the addition of potassium ionophore (nigericin). Peritoneal macrophages were infected with ΔRD1 at an MOI of 5 for 21 h. Nigericin (5 μM) and/or KCl (30 mM) was added, and the culture was continued for another 3 h. The culture supernatant was collected, and the production of IL-1β (A) and IL-18 (B) was measured by ELISA. Data represent the mean ± standard deviations of triplicate assays. *, a P value of <0.05 for ΔRD1-infected cells in the presence of nigericin compared to either ΔRD1-infected cells in the absence of nigericin or ΔRD1-infected cells in the presence of both nigericin and KCl. (C) The amounts of procaspase-1 (p45) in the cell lysate and the fragment (p10) of mature caspase-1, which was immunoprecipitated from the culture supernatant, were detected by Western blotting. β-Actin was used as a loading control of the cell lysate.

DISCUSSION

In the present study, the RD1 locus in the M. tuberculosis genome is implicated in the activation of caspase-1 via induction of the potassium ion efflux in infected macrophages. Koo et al. have shown recently that M. tuberculosis stimulates the secretion of IL-1β and IL-18, and the activity is closely related to the RD1 locus (28). Our results are consistent with their findings. In addition, we newly demonstrated in this study that M. tuberculosis-induced caspase-1 activation is not dependent on IFN-β. Furthermore, M. tuberculosis caused a potassium ion efflux independently of the P2X7 receptor. We also found that H37Rv has a higher ability to induce cell death of infected macrophages than ΔRD1 (data not shown). ESX-1-dependent cytolysis may be involved in the cytokine and lysosome secretion, as reported previously (28). However, the exact relationship between the intracellular molecular events and the cytokine secretion still remains to be elucidated.

It has been shown that caspase-1 is activated after infection with several bacteria, including Salmonella enterica serovar Typhimurium, Pseudomonas aeruginosa, Listeria monocytogenes, Shigella flexneri, and Francisella tularensis, and that various components or cascades are involved in the activation of caspase-1 (32, 50) Recently, Henry et al. have shown that IFN-β participates in the activation of caspase-1 in macrophages infected with Francisella tularensis and Listeria monocytogenes, whereas it is dispensable for the activation of caspase-1 in infections with Salmonella enterica serovar Typhimurium (22). In the case of M. tuberculosis infection, it has been shown that IFN-β is produced from macrophages infected with M. tuberculosis and that the production is dependent on ESX-1 (46). Our preliminary study also showed that H37Rv induced higher IFN-β production than ΔRD1 (data not shown). However, the present study clearly showed that IFN-β was not required for the activation of caspase-1 in M. tuberculosis infection. It has been shown that M. tuberculosis may induce the formation of the NALP3/ASC inflammasome (28). As reported previously, L. monocytogenes induces caspase-1 activation via the formation of the NALP3/ASC inflammasome. However, there was a difference in the requirement of IFN-β in the formation of inflammasome after infection with M. tuberculosis and L. monocytogenes. Although the nature of IFN-β-dependent signaling events is not known, it is probable that the inflammasome is completed via the formation of a multiprotein complex. There may be differences in the composition of inflammasome induced by infection with different bacterial species.

Franchi et al. have shown that the requirement for the P2X7 receptor and intracellular potassium in caspase-1 activation is different between infection with intracellular and extracellular parasitic bacterial species (15). According to their report, regarding infection with Staphylococcus aureus and Escherichia coli, caspase-1 activation was triggered by P2X7 receptor-mediated intracellular potassium depletion, which is induced by the addition of ATP. In contrast, infection with Salmonella enterica serovar Typhimurium and Listeria monocytogenes induced both caspase-1 activation and the cytokine secretion independently of the P2X7 receptor and potassium ion efflux. In M. tuberculosis infection, as shown here, the P2X7 receptor was not required for caspase-1 activation. However, a potassium ion efflux was necessary for caspase-1 activation, and RD1 was implicated in triggering the intracellular event. Several studies revealed that bacteria or bacterial components secreted in the cytosol induced caspase-1 activation (3, 19, 35) In this context, it has been shown that flagellin secreted by a type III secretion system of S. enterica serovar Typhimurium and listeriolysin O produced by L. monocytogenes are identified as effector molecules for inducing caspase-1 activation (14, 21). Recent evidence suggested that the ESX-1 secretory system of M. tuberculosis is capable of delivering several effector proteins to the host cytosol (1). Therefore, it is likely that some bacterial components induce caspase-1 activation during M. tuberculosis infection by intracellular potassium ion efflux.

It has been found that NALP3 and ASC are involved in the activation of caspase-1 in M. tuberculosis-infected macrophages (28). Pétrilli et al. have shown that the formation of the NALP inflammasome (NALP1 or NALP3) is induced under a low intracellular potassium concentration (39). Hentze et al. further demonstrated that cathepsin B contributes to the formation of the NALP3 inflammasome that is induced by potassium ion efflux (23). Taken together, it is probable that an ESX-1-dependent potassium ion efflux might cause the formation of the NALP3/ASC inflammasome through a release of cathepsin B from the lysosomal compartment. On the other hand, Fernandes-Alnemri et al. have shown that potassium depletion induces the formation of a pyroptosome, distinct from an inflammasome, which is largely composed of oligomerized ASC and can activate caspase-1 and release IL-1β (9). Although the formation of an inflammasome or pyroptosome may be involved in the activation of caspase-1 in infected macrophages, there was no information about the M. tuberculosis factors responsible for the potassium ion efflux followed by the activation of caspase-1. In this study, we demonstrated the close relationship between RD1 and the potassium ion efflux. M. tuberculosis and Mycobacterium marinum have been shown to induce permeation of the cell membrane (13, 18). Smith et al. have shown that the permeation is caused by pore formation (45). It has been suggested that ESAT-6 of M. tuberculosis has a membrane-lysing activity and that ESAT-6 of M. marinum could play a direct role in causing pore formation (7, 45). In addition, there are other M. tuberculosis components that are secreted through the ESX-1 secretion system, although their functions have not yet been identified (13, 34). Therefore, it is probable that one or more of the effector proteins secreted by the ESX-1 secretory system cause changes in the membrane integrity, leading to a decrease in the intracellular potassium level. We are attempting to identify the M. tuberculosis factors which may lead to such intracellular potassium perturbations in our future study.

The roles of IL-18 and IL-1β in the pathogenesis of tuberculosis still remain controversial. There is one recent report demonstrating that M. tuberculosis and Mycobacterium bovis BCG actively prevent inflammasome activation by use of a putative Zn²⁺ metalloprotease (33), while another report (28) and ours demonstrate that RD1-sufficient strains of mycobacteria can induce inflammasome activation. From our point of view, however, it can be suggested that IL-1β and IL-18 induced by M. tuberculosis are important for the protection (16, 44) and formation (37) of tuberculous granuloma. These cytokines, in concert with other cytokines or chemokines, may exert both beneficial and detrimental effects to the host, resulting in a complex pathology. Considering the fact that a RD1-deficient strain of M. tuberculosis cannot induce a strong activation of caspase-1 and secretion of IL-18 and IL-1β, it is tempting to assume that the limited efficacy of the BCG vaccine against adult pulmonary tuberculosis (11) is due, at least in part, to the absence of RD-1-dependent induction of mature IL-1β and IL-18. In fact, it has been shown that BCG or Mycobacterium microti strains that were transformed with the RD1 region show enhanced efficacy of vaccination in animal models (5, 41), although it is still unknown to what extent IL-1β and IL-18 play roles in this vaccination. Further studies are needed to obtain a comprehensive idea about the roles played by IL-1β and IL-18 in the pathogenesis of tuberculosis and to develop effective vaccines against tuberculosis.