Activation of NLRP3 and AIM2 Inflammasomes by Porphyromonas gingivalis Infection

Abstract

Porphyromonas gingivalis, a major periodontopathogen, is involved in the pathogenesis of periodontitis. Interleukin-1β (IL-1β), a proinflammatory cytokine, regulates innate immune responses and is critical for the host defense against bacterial infection. However, excessive IL-1β is linked to periodontal destruction. IL-1β synthesis, maturation, and secretion are tightly regulated by Toll-like receptor (TLR) signaling and inflammasome activation. We found much higher levels of inflammasome components in the gingival tissues from patients with chronic periodontitis than in those from healthy controls. To investigate the molecular mechanisms by which P. gingivalis infection causes IL-1β secretion, we examined the characteristics of P. gingivalis-induced signaling in differentiated THP-1 cells. We found that P. gingivalis induces IL-1β secretion and inflammatory cell death via caspase-1 activation. We also found that P. gingivalis-induced IL-1β secretion and pyroptic cell death required both NLRP3 and AIM2 inflammasome activation. The activation of the NLRP3 inflammasome was mediated by ATP release, the P2X₇ receptor, and lysosomal damage. In addition, we found that the priming signal via TLR2 and TLR4 activation precedes P. gingivalis-induced IL-1β release. Our study provides novel insight into the innate immune response against P. gingivalis infection which could potentially be used for the prevention and therapy of periodontitis.

INTRODUCTION

Interleukin-1β (IL-1β), a proinflammatory cytokine, is critical in the host defense against many pathogens and regulates innate immune and inflammatory responses. IL-1β processing and secretion are tightly regulated by a two-step mechanism (1). The first event is the transcription of the pro-IL-1β gene, dependent on the activation of nuclear factor-κB (NF-κB) by Toll-like receptors (TLRs). TLRs sense invading pathogens outside the cell and in intracellular endosomes and lysosomes (2). Each TLR detects distinct pathogen-associated molecular patterns (PAMPs) derived from bacteria, mycobacteria, fungi, and viruses (3). These include lipoproteins (recognized by TLR1, TLR2, and TLR6), double-stranded RNA (TLR3), lipopolysaccharide (LPS; TLR4), flagellin (TLR5), single-stranded RNA (TLR7 and TLR8), and DNA (TLR9). Upon recognition of PAMPs, TLRs recruit a specific set of adaptor molecules that harbor a TIR domain, such as MyD88 and TRIF, and initiate downstream signaling events that lead to NF-κB activation, resulting in the upregulation of proinflammatory cytokines and chemokines. The second signal is the activation of inflammasome that converts pro-IL-1β to IL-1β. The inflammasome is composed of NLR or AIM2 family receptors and procaspase-1. An apoptosis-associated speck-like protein containing a caspase recruitment domain (ASC) is an adaptor protein with an N-terminal PYD and a C-terminal CARD. It links the PYD-containing NLR family member to procaspase-1, using its PYD to interact with the PYD of the NLRs and its CARD to interact with the CARD of procaspase-1. PYD-containing NLR family members assemble an inflammasome complex with ASC, which in turn recruits and activates caspase-1 (4–6). Several members of the NLR family proteins participate in the formation of distinct inflammasomes, including NLR family pyrin domain-containing 3 (NLRP3; also known as cyropyrin or NALP3), NLR family CARD domain-containing 4 (NLRC4; also known as IPAF), and NLRP1. Different inflammasomes are activated by various stimuli (7). For example, NLRP1 becomes activated by the lethal toxin produced by Bacillus anthracis, whereas NLRC4 responds to cytosolic flagellin in cells infected with Salmonella, Legionella, and Pseudomonas spp. The NLRP3 inflammasome is activated by a large variety of stimuli, including microbial products and endogenous signals, such as urate crystal, silica, amyloid fibrils, and ATP. Besides NLRs, AIM2 family members can activate inflammasomes. AIM2 is characterized by the presence of a pyrin domain and a DNA-binding HIN domain and activates caspase-1 by detecting cytosolic DNA (8). Upon activation, the NLR family members bind to adaptor protein ASC; in turn, the bound proteins recruit procaspase-1 for activation. Activated capsase-1 cleaves the proform of the cytokines IL-1β and IL-18 to their mature and secreted forms. Caspase-1 activation also induces a proinflammatory cell death called pyroptosis and thereby removes the replicative niche of intracellular pathogens (9). Assembly of the inflammasome requires a preceding priming signal via TLRs which is required to upregulate the expression of inflammasome receptors and the substrate pro-IL-1β, before the second signal can initiate inflammasome complex formation (10). Although IL-1β is required for host defense against pathogens, overreacted expression and secretion of this molecule can lead to tissue damage, and dysregulated inflammasome activation is related to the pathogenesis of a variety of inflammatory diseases (1, 11).

Periodontitis, one of the most common diseases, is an infection-driven chronic inflammatory disease of periodontium and the major cause of tooth loss. Periodontitis is induced by periodontopathogens, such as Porphyromonas gingivalis, Tannerella forsythia, and Treponema denticola. P. gingivalis, a Gram-negative short-rod anaerobe, is a predominant periodontal pathogen that expresses a number of virulence factors involved in the pathogenesis of periodontitis (12–14). Virulence components of P. gingivalis, such as LPS, lipoproteins, gingipains, and fimbriae, are important in the induction of immune responses, including production of cytokines and activation of inflammation-related signaling pathways (13, 15, 16). LPS, a strong inducer of IL-1β, induces caspase-1 activation and a pyroptosis-like cell death in human THP-1 cells (17). The patterns of cytokine production from human macrophages stimulated with purified bacterial components, LPS, or FimA differed from those observed when the cells were challenged with P. gingivalis (18). Thus, to investigate the host inflammatory responses evoked by P. gingivalis in vivo, we used live bacteria instead of bacterial components.

Since IL-1β has been linked with the destruction of periodontal tissue, examination of the interaction between live P. gingivalis and host cells leading to IL-1β release is necessary to understand the process of periodontal diseases and identify valuable targets for periodontal treatment (19). Therefore, the aim of this study was to elucidate the mechanism of P. gingivalis-induced IL-1β. We found that P. gingivalis-induced IL-1β secretion required both NLRP3 and AIM2 inflammasome activation in THP-1 cells. Moreover, the activation of the NLRP3 inflammasome was mediated by ATP release, the P2X₇ receptor (P2X₇R), and lysosomal damage. In addition, we also found that the priming signal provided via the TLR2 and TLR4 activation signal is a prerequisite for P. gingivalis-induced IL-1β release.

MATERIALS AND METHODS

Reagents.

We purchased chemicals and antibodies from the sources listed below. Phorbol 12-mystristate 13-acetate (PMA), hemin, vitamin K (3-phytyl-menadione), oxidized ATP (oxATP), KCl, and trichloroacetate (TCA) were from Sigma (St. Louis, MO). CA-074 methyl ester (CA-074 Me) and pyrrolidine dithiocarbamate (PDTC; a NF-κB inhibitor) were purchased from Calbiochem (San Diego, CA). Z-Trp-Glu(OMe)-His-Asp(OMe)-fluoromethylketone (Z-WEHD-FMK; caspase-1 inhibitor), benzyloxycarbonyl-V-A-D-O-methyl fluoromethyl ketone(Z-VAD-FMK; pancaspase inhibitor), anti-human IL-1β (catalog no. 2805), and anti-human NLRP3 (catalog no. AF6789) were from R&D Systems (Minneapolis, MN). Anti-human ASC (catalog no. 4628), anti-human AIM2 (catalog no. 8055), and anti-human caspase-1 (catalog no. 2225) antibodies and radioimmunoprecipitation assay buffer were from Cell Signaling Technology (Beverly, MA). Anti-human caspase-1 (catalog no. sc-56036), anti-human NLRC4 (catalog no. sc-99056), anti-human TLR2 (catalog no. sc-21760), anti-human TLR4 (catalog no. sc-10741), anti-human MyD88 (catalog no. sc-8196), anti-human TRIF (catalog no. sc-130902), anti-human TRAF6 (catalog no. sc-7221), and anti-β-actin (catalog no. sc-47778) antibodies were from Santa Cruz (Santa Cruz, CA).

GCF and gingival tissue sampling.

We obtained gingival crevicular fluid (GCF) and gingival tissue from patients who were scheduled to undergo treatment at the Department of Periodontology of the Pusan National Dental School. All donors gave written informed consent, and their rights were protected according to the protocol reviewed and approved by the Institutional Review Board of Pusan National University. GCF samples were collected from a periodontitis site (probing depth [PD] ≧ 5 mm, calculus ≧ 2 mm, and bleeding on probing [BOP]) as well as from a healthy site (PD ≦ 3 mm, no BOP). Sites for GCF collection were isolated with cotton rolls and gently air dried, and supragingival plaque was removed. A collection strip was inserted 1 to 2 mm into the sulcus for 1 min. Samples were eluted from the strip in 300 μl of phosphate-buffered saline (PBS) by centrifugation in an Eppendorf tube. Eluates were stored at −80°C until assays were performed. Healthy gingival tissue was sampled from healthy participants' tissue that showed no overt signs of gingival inflammation at a probing depth of ≦3 mm. Periodontitis tissue was obtained from sites with signs of gingival inflammation and a probing depth of ≧5 mm. Gingival samples were collected from both healthy and periodontitis sites after infiltration with an appropriate local anesthetic. Samples comprised the epithelial lining and a portion of the underlying connective tissue, and inflammation was confirmed by hematoxylin-eosin (H&E) staining. The gingival tissue specimens obtained were thoroughly rinsed with sterile normal saline solution, transferred into Eppendorf tubes, and stored at −80°C until use.

Bacterial culture.

P. gingivalis (strain 381) was grown in Gifu anaerobic medium broth (Nissui, Japan), which contained 5 mg/ml hemin and 0.5 mg/ml 3-phytyl-menadione (vitamin K) under anaerobic conditions at 37°C. An optical density at 650 nm of 1.0 was determined to correlate to 10⁹ CFU/ml. The bacteria were washed and resuspended in RPMI medium to infect the THP-1 cells at a multiplicity of infection (MOI) of 1:10, 1:50, or 1:100.

Cell treatment.

Cells of the THP-1 cell line, a human acute monocytic leukemia cell line, were differentiated to macrophage-like cells by treatment with 50 nM PMA overnight. The differentiated cells were infected with live P. gingivalis for 6 or 24 h. In some experiments, the cells were pretreated with Z-WEHD-FMK, Z-VAD-FMK, oxATP, KCl, CA-074 Me, and dimethyl sulfoxide (DMSO) at the indicated concentrations for 30 min before bacterial challenge.

Preparation of THP-1/ASC-GFP stable cell line.

For generation of retrovirus used for stable expression of green fluorescent protein-tagged ASC (ASC-GFP), HEK293T cells were transiently transfected with plasmids encoding gag-pol, vesicular stomatitis virus G protein, and pMSCVneo-ASC-GFP. At 34 and 50 h posttransfection, the culture supernatant containing retroviral particles was collected and added to THP-1 cells. Retrovirus-transduced THP-1 cells were selected with 500 μg/ml G418 for 5 days.

ASC pyroptosome quantitation in live cells.

THP-1/ASC-GFP cells were seeded in 8-well chambers and then primed with PMA (50 nM). THP-1/ASC-GFP cells were pretreated with various agents and further challenged with P. gingivalis. ASC pyroptosome formation was observed by use of a confocal laser scanning microscope (LSM 700; Carl Zeiss) at various time points. At least 400 cells were counted to enumerate the number of cells containing the ASC-GFP pyroptosome at the end of each time period. The percentage of cells with an ASC pyroptosome was calculated by dividing the number of cells with an ASC pyroptosome by the total number of cells counted.

Real-time qRT-PCR.

Total RNA was extracted with the TRIzol reagent (Invitrogen) following the manufacturer's instructions, and cDNA was synthesized with a reverse transcription system (Bioneer Co., Daejeon, South Korea). For quantitative reverse transcriptase (qRT-PCR), the cDNA was amplified using TaqMan Universal PCR master mix (Applied Biosystems) with gene-specific primers and fluorogenic TaqMan probes on an ABI 7500 real-time PCR system (Applied Biosystems) with the following programs: a 10-min preincubation at 95°C and 40 cycles of 15 s at 95°C and 1 min at 60°C. The data were normalized relative to those for GAPDH (glyceraldehyde-3-phosphate dehydrogenase) as an endogenous control. TaqMan assay primers and probes were purchased from Qiagen (Valencia, CA) (for GAPDH, catalog no. Hs99999905; for ASC, catalog no. Hs00203118; for NLRP3, catalog no. Hs00918082; for AIM2, catalog no. Hs00915710; for NLRC4, catalog no. Hs00892666).

RNA interference assay.

Human small interfering RNAs (siRNAs) for ASC, NLRP3, AIM2, NLRC4, TLR2, TLR4, MyD88, TRIF, TRAF6, and nontargeting control oligonucleotides were obtained from Qiagen (Valencia, CA). The sequences of each siRNA oligonucleotide in this pool are as follows: ASC siRNA, SI03086783; NLRP3 siRNA, SI03060323; NLRC4 siRNA, SI04374685; AIM2 siRNA, SI04261432; TLR2 siRNA, SI00050029; TLR4 siRNA, SI04951149; MyD88 siRNA, SI00300909; TRIF siRNA, RNF36; SI03027255; TRAF6 siRNA, SI03046043; and nontargeting control siRNA oligonucleotides, 1027281. For siRNA experiments in THP-1 cells, cells were seeded in 6-well plates at a density of 1 × 10⁶ cells per well in the presence of 50 nM PMA. The differentiated cells were then transfected with siRNA oligonucleotides (1,200 ng) for 48 h using the Attractene transfection reagent (Qiagen, Valencia, CA) in 1 ml of RPMI.

Cytokine analysis.

Tumor necrosis factor alpha (TNF-α) and IL-1β levels were measured by using enzyme-linked immunosorbent assay (ELISA) kits from eBioscience (San Diego, CA) following the manufacturer's instruction.

Immunoblot analysis.

Procaspase-1 and pro-IL-1β in the gingival tissue and cell extracts and their active forms precipitated by 10% TCA from the culture supernatant were analyzed by immunoblotting. The cells were washed in ice-cold PBS and lysed. A total of 50 μg proteins was separated by 15% SDS-PAGE, transferred to a polyvinylidene difluoride membrane, and probed with the indicated antibodies. Reactivity was determined by using horseradish peroxidase-conjugated anti-mouse, anti-rabbit, anti-sheep, or anti-goat IgG secondary antibody (diluted 1:5,000; Jackson ImmunoResearch Laboratories, Inc., PA), and the signals were visualized using SuperSignal West Femto maximum-sensitivity substrate (Pierce, Rockford, IL) with an LAS-4000 Fujifilm luminescent image analyzer. The band intensities of the immunoblot were quantified with NIH ImageJ software and are presented as a ratio relative to the intensity of β-actin.

Cell death assay.

Cell death was measured with a lactate dehydrogenase (LDH) cytotoxicity assay kit (CytoTox96 nonradioactive cytotoxicity assay; Promega). The percent cytotoxicity was calculated by the following formula: 100 × [(experimental LDH release − spontaneous LDH release)/(maximal LDH release − spontaneous LDH release)]. To determine the maximal LDH release, cells were treated with 1% Triton X-100.

ATP determination.

The extracellular ATP concentration was determined with an ATP determination kit (Invitrogen, Carlsbad, CA) following the manufacturer's instruction.

Statistics.

Statistically significant differences between samples were analyzed with an unpaired, one-tailed Student's t test. The data are shown as the mean ± standard deviation (SD). A P value of less than 0.05 was considered statistically significant.

RESULTS

The levels of IL-1β and inflammasome components are increased in periodontitis patients.

Periodontitis is a chronic inflammatory disease caused by major periodontopathogens. To describe the relationship between periodontitis and inflammasomes, we first investigated if the level of IL-1β was higher in the GCF of periodontitis patients. The level of IL-1β as well as that of TNF-α was about 4-fold higher in the GCF of periodontitis patients (Fig. 1A). Next, we examined the levels of inflammasome components in the gingival tissues from healthy subjects and subjects with chronic periodontitis. As shown in Fig. 1B, caspase-1 was highly expressed in the gingival tissue of periodontitis patients. The levels of expression of both NLRP3 and AIM2 were also higher in the periodontitis gingival tissues than the healthy ones. These finding suggest that NLRP3 and AIM2 inflammasomes are activated in periodontitis tissues, leading to caspase-1 activation and, in turn, IL-1β secretion.

FIG 1.

IL-1β and TNF-α secretion is increased in periodontitis patients. (A) GCF from healthy subjects and chronic periodontitis patients was analyzed for determining the concentrations of human IL-1β and TNF-α by ELISA. ***, P < 0.001. (B) The expression of inflammasome components in the gingival tissues of healthy subjects and chronic periodontitis patients was detected by immunoblotting. Each of nine gingival tissue specimens from healthy subjects and chronic periodontitis patients was examined. Results represent those from one of three individual experiments. The numbers below the lanes indicate average ratios normalized with the level of β-actin expression.

P. gingivalis induces caspase-1 activation, IL-1β secretion, and inflammatory cell death.

To determine whether P. gingivalis induces inflammation, we examined proinflammatory cytokine secretion. We differentiated cells of a human acute monocytic leukemia cell line (THP-1) to macrophage-like cells by treatment with PMA. The MOI- and time-dependent release of IL-1β and TNF-α after infection with P. gingivalis was detected in culture supernatants by ELISA (Fig. 2A). P. gingivalis activated caspase-1, and the amount of processed caspase-1 subunit in the culture supernatants was increased depending on time and the MOI of P. gingivalis (Fig. 2B and C). In addition to caspase-1 activation, P. gingivalis induced pro-IL-1β expression in the cells. These results suggest that P. gingivalis regulates caspase-1 activation, pro-IL-1β synthesis, and IL-1β secretion in THP-1 cells. Activation of caspase-1 not only leads to inflammation but also causes an inflammatory form of cell death called pyroptosis. Formation of the pyroptosome might be a mechanism by which dying macrophages release their cellular contents, such as lactate dehydrogenase (LDH), a marker of cell death. P. gingivalis dramatically induced LDH release from THP-1 cells. Cytotoxicity was increased in an MOI- and time-dependent manner upon P. gingivalis infection (Fig. 2D). ASC plays a critical role in Salmonella enterica serovar Typhimurium-induced pyroptic cell death and is required for the secretion of IL-1β induced by P. gingivalis (19, 20). To test whether P. gingivalis induces ASC-dependent pyroptosome formation, we generated a THP-1 cell line that stably expresses an ASC-GFP fusion protein. The level of ASC-GFP expression was comparable to that of the endogenous ASC protein (data not shown). ASC forms a single 1- to 2-mm supramolecular assembly in each macrophage cell in response to stimulation with proinflammatory stimuli (21). In unstimulated cells, ASC-GFP was evenly distributed in the cytoplasm and nucleus. After stimulation with P. gingivalis, the ASC-GFP fluorescence accumulated in bright large spots in the cytoplasm, indicating the formation of the ASC pyroptosome. Quantitative analysis revealed that the number of pyroptosome-containing cells increased in an MOI- and time-dependent manner after stimulation with P. gingivalis (Fig. 2E). These data suggest that P. gingivalis induces the formation of the pyroptosome, leading to pyroptosis (inflammatory cell death) in THP-1 cells.

FIG 2.

P. gingivalis enhances IL-1β secretion and caspase-1 activation, leading to inflammatory cell death. (A) PMA-primed THP-1 cells were infected with P. gingivalis (MOI, 100) for 6 or 24 h. Cell culture supernatants were assayed for human IL-1β and TNF-α by ELISA. The data represent mean values ± SDs (n ≧ 3). (B and C) The IL-1β and caspase-1 secreted into the culture supernatants (sup.) and the pro-IL-1β, procaspase-1, and β-actin in the cell lysates (cell) were detected by immunoblotting. (B) PMA-primed THP-1 cells were infected with P. gingivalis (P.g) (MOI, 100) for 3, 6, or 24 h. (C) PMA-primed THP-1 cells were infected with P. gingivalis (MOI, 10, 50, or 100) for 6 h. (D) The amount of the cytoplasmic enzyme LDH released into the culture supernatant was measured with an LDH cytotoxicity assay kit. The data represent mean values ± SDs (n ≧ 3). (E) Live PMA-primed ASC-GFP cells infected with P. gingivalis at each MOI for the indicated times were observed and photographed by fluorescence confocal microscopy. The graph on the right shows the percentages of cells containing the ASC pyroptosome. The percentages of cells containing the ASC pyroptosome were calculated as described in Materials and Methods. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

P. gingivalis-induced IL-1β secretion and proinflammatory cell death are dependent on caspase-1.

Caspase-1 is required for the processing and subsequent release of IL-1β. To confirm that P. gingivalis-induced IL-1β secretion was caused by caspase-1 activation, the cells were pretreated with the caspase inhibitor Z-WEHD-FMK or Z-VAD-FMK prior to P. gingivalis challenge. As shown in Fig. 3A and B, Z-WEHD-FMK or Z-VAD-FMK inhibited caspase-1 activation and IL-1β secretion, indicating that P. gingivalis-induced IL-1β secretion is dependent on caspase-1 activation. To determine whether P. gingivalis-induced pyroptosome formation and cell death are dependent on caspase-1 activation, cells were pretreated with the caspase inhibitor and then stimulated with P. gingivalis. Pretreatment with a caspase inhibitor blocked LDH release and the formation of an ASC pyroptosome in response to P. gingivalis infection (Fig. 3C and D). These data indicate that P. gingivalis-induced pyroptosis is dependent on caspase-1 activation.

FIG 3.

P. gingivalis-induced IL-1β secretion and proinflammatory cell death are dependent on caspase-1. (A to C) THP-1 cells were pretreated with Z-WEHD-FMK (30 μM) or Z-VAD-FMK (30 μM) for 30 min before P. gingivalis infection for 6 h. (A) The IL-1β and caspase-1 secreted into the culture supernatant (sup.) and the pro-IL-1β, procaspase-1, and β-actin in the cell lysates (cell) were detected by immunoblotting. (B and C) Cell culture supernatant was collected and assayed for IL-1β secretion by ELISA (B) and the cytoplasmic enzyme LDH (C). The data represent mean values ± SDs (n ≧ 3). (D) Live PMA-primed ASC-GFP cells were pretreated with Z-WEHD-FMK or Z-VAD-FMK for 30 min before P. gingivalis infection for 6 h or 24 h. The graph on the right shows the percentages of cells containing the ASC pyroptosome. **, P < 0.01; ***, P < 0.001.

P. gingivalis-induced IL-1β secretion requires both NLRP3 and AIM2 inflammasomes.

Next, we investigated whether the expression of inflammasome components is regulated by P. gingivalis. P. gingivalis infection increased the expression of NLRP3 and AIM2 in a time-dependent manner but did not affect the expression of ASC and NLRC4 (Fig. 4A and B). To determine which inflammasome components are involved in P. gingivalis-induced IL-1β secretion, we individually knocked down ASC, NLRP3, AIM2, and NLRC4 with specific siRNAs. Treatment with siRNAs almost completely blocked the expression of each protein (see Fig. S1A in the supplemental material). Using the knockdown cells, we found that ASC, NLRP3, or AIM2 knockdown significantly inhibited P. gingivalis-induced caspase-1 activation and subsequent IL-1β secretion without affecting the expression of procaspase-1 (Fig. 4C and D). In contrast, the silencing of NLRC4 did not inhibit caspase-1 activation or IL-1β secretion (Fig. 4C). The LDH release induced by P. gingivalis infection was also strongly inhibited in THP-1 cells transfected with siRNA for ASC, NLRP3, or AIM2 (Fig. 4E). To find out which inflammasome components P. gingivalis-induced pyroptosome formation is dependent on, ASC-GFP cells were transfected with siRNAs and then stimulated with P. gingivalis. Silencing of NLRP3 or AIM2 blocked the formation of an ASC pyroptosome in response to P. gingivalis infection, whereas the silencing of NLRC4 had a marginal effect (Fig. 4F). These results indicate that P. gingivalis-induced caspase-1 activation and subsequent IL-1β secretion are mediated by both NLRP3 and AIM2 inflammasomes.

FIG 4.

Caspase-1 activation, IL-1β secretion, cell cytotoxicity, and pyroptic cell death induced by P. gingivalis infection are mediated by NLRP3 and AIM2 inflammasome activation. (A) Real-time PCR measurement of ASC, NLRP3, AIM2, NLRPC4, and IL-1β mRNA expression in PMA-primed THP-1 cells at different times after P. gingivalis infection (MOI, 100). (B) Immunoblot analysis of ASC, NLRP3, AIM2, NLRPC4, and β-actin in P. gingivalis-infected cells. The numbers below the lanes indicate average ratios relative to controls normalized to the level of β-actin expression. (C to E) THP-1 cells were transfected with ASC, NLRP3, AIM2, or NLRC4 siRNA for 48 h. siRNA-transfected cells were infected with P. gingivalis (MOI, 100) for 6 h or 24 h. The IL-1β and caspase-1 secreted into the culture supernatants (sup.) and the procaspase-1 and β-actin in the cell lysates (cell) were detected by immunoblotting (C). Cell culture supernatant was collected and assayed for IL-1β secretion by ELISA (Con., control) (D) and the cytoplasmic enzyme LDH (E). The data represent mean values ± SDs (n ≧ 3). (F) Live PMA-primed ASC-GFP cells were transfected with ASC, NLRP3, AIM2, or NLRC4 siRNA for 48 h and infected with P. gingivalis (MOI, 100) for 6 h or 24 h and observed by fluorescence confocal microscopy. The graph on the right shows the percentages of cells containing ASC pyroptosome. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

P. gingivalis-induced caspase-1 activation is mediated by ATP, potassium depletion, and cathepsin B.

The NLRP3 inflammasomes are activated in response to danger signals, such as ATP, released from stressed cells (11, 22). Extracellular ATP binds to the P2X₇ receptor (P2X₇R), and the subsequent potassium efflux is thought to account for NLRP3 inflammasome activation (23). To examine whether P. gingivalis induces ATP release, THP-1 cells were infected with P. gingivalis and the ATP concentration in the culture supernatant was measured. P. gingivalis significantly induced ATP release (Fig. 5A; see Fig. S2A in the supplemental material). To evaluate whether the effects of P. gingivalis on caspase-1 activation and IL-1β secretion are mediated through P2X₇R, we pretreated the cells with oxATP, a P2X₇R antagonist that blocks the interaction of ATP with P2X₇R, for 30 min and then infected the cells with P. gingivalis for 6 or 24 h. The oxATP strongly blocked the caspase-1 activation and IL-1β secretion induced by P. gingivalis (Fig. 5B and C; see Fig. S2B and C in the supplemental material). Furthermore, oxATP treatment inhibited the induction of cytotoxicity and ASC-dependent pyroptosome formation by P. gingivalis infection (Fig. 5D and K; see Fig. S2D and K in the supplemental material). These results indicate that ATP has a critical role in P. gingivalis-induced NLRP3 inflammasome activation.

FIG 5.

ATP and potassium (K⁺) efflux are involved in P. gingivalis-induced NLRP3 activation. (A) THP-1 cells were infected with P. gingivalis for 6 h. The extracellular ATP concentration was determined using an ATP determination kit. The data are shown as mean values ± SDs (n ≧ 3). (B to J) THP-1 cells were pretreated with oxATP (300 μM), KCl (50 mM), CA-074 Me (10 μM), or DMSO for 30 min before P. gingivalis infection for 6 h. Secreted IL-1β and caspase-1 were detected by immunoblotting (B, E, and H). Cell culture supernatant was collected and assayed for IL-1β secretion by ELISA (C, F, and I) and the release of the cytoplasmic enzyme LDH (D, G, and J). (K) Live PMA-primed ASC-GFP cells were pretreated with oxATP (300 μM), KCl (50 mM), or CA-074 Me (10 μM) for 30 min prior to infection with P. gingivalis for 6 h and observed and photographed by fluorescence confocal microscopy. The graph on the right shows the percentages of cells containing ASC pyroptosome. DMSO was used as a negative control. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Potassium (K⁺) efflux is a common event induced by a broad range of inflammasome stimuli and is important in caspase-1 activation, IL-1β processing, and cell death. Consequently, activation can be specifically blocked by the addition of extracellular K⁺ (23, 24). To examine whether K⁺ efflux affects P. gingivalis-induced IL-1β secretion and caspase-1 activation, THP-1 cells were pretreated with extracellular KCl at a 30 mM concentration for 30 min and then infected with P. gingivalis for 6 or 24 h. The increased extracellular KCl concentration significantly reduced P. gingivalis-induced caspase-1 activation and IL-1β secretion (Fig. 5E and F; see Fig. S2E and F in the supplemental material). Given that depletion of intracellular potassium by bacterial toxins has been shown to induce pyroptic cell death of THP-1 cells, we asked whether intracellular K⁺ depletion also plays a role in P. gingivalis-induced LDH release and the formation of an ASC pyroptosome. To answer this question, we examined the effect of inhibiting K⁺ efflux by increasing the extracellular K⁺ concentration to 30 mM. Inhibition of the K⁺ efflux strongly reduced the cytotoxicity and ASC pyroptosome formation (Fig. 5G and K; see Fig. S2G and K in the supplemental material). Our results suggest that K⁺ efflux is needed for P. gingivalis-mediated NLRP3 inflammasome activation.

Phagosomal uptake of monosodium urate (MSU), silica crystals, aluminum salts, or fibrillar amyloid-β induces lysosomal damage and the leakage of lysosomal proteases, specifically, cathepsin B, into the cytosol, leading to the activation of the NLRP3 inflammasome and IL-1β secretion (25, 26). We examined whether P. gingivalis-induced caspase-1 activation and IL-1β secretion are dependent on cathepsin B. THP-1 cells were either untreated (control) or pretreated with a cathepsin B inhibitor (CA-074 Me, 10 μM) or DMSO for 30 min and subsequently infected with P. gingivalis for 6 or 24 h. The inhibition of cathepsin B with CA-074 Me almost completely inhibited P. gingivalis-induced caspase-1 activation and IL-1β secretion (Fig. 5H and I; see Fig. S2H and I in the supplemental material). To further examine the role of cathepsin B, we analyzed the LDH release and the formation of the ASC pyroptosome in the presence of CA-074 Me. The cathepsin B inhibitor significantly decreased the extent of P. gingivalis-induced cytotoxicity and ASC pyroptosome formation (Fig. 5J and K; see Fig. S2J and K in the supplemental material). These results indicate that the effect of P. gingivalis on NLRP3 inflammasome activation is also mediated by lysosomal disruption. Our data suggest that ATP, K⁺ efflux, and cathepsin B are critical in P. gingivalis-induced NLRP3 inflammasome activation and subsequent IL-1β secretion.

P. gingivalis-induced inflammasome activation requires the preceding TLR signaling.

The production of mature IL-1β requires the activation of two pathways: (i) pattern recognition receptors such as TLRs increase pro-IL-1β expression through NF-κB activation, and (ii) inflammasomes convert pro-IL-1β to IL-1β. The surface components of P. gingivalis, such as LPS, lipoproteins, and fimbriae, interact with host TLRs and activate the innate immune responses (13). We found that P. gingivalis infection increases pro-IL-1β gene expression and protein synthesis (Fig. 2A and B). Because TLR2 and TLR4 are the key molecules required for the cytokine production induced by Gram-negative periodontal bacteria (27, 28), we tested the roles of TLR2 and TLR4 in P. gingivalis-induced IL-1β secretion. Knockdown of TLR2 and/or TLR4 with specific siRNAs inhibited P. gingivalis-mediated pro-IL-1β induction as well as caspase-1 activation and IL-1β secretion (Fig. 6A and B; see Fig. S3A to C in the supplemental material). The silencing of TLR2 and/or TLR4 also decreased the extent of cytotoxicity and ASC pyroptosome formation induced by P. gingivalis (Fig. 6C and D; see Fig. S3D and E in the supplemental material).

FIG 6.

P. gingivalis-induced inflammasome activation requires the preceding TLR signaling. (A to C) THP-1 cells were transfected with TLR2 and/or TLR4 siRNA for 48 h. siRNA-transfected cells were infected with P. gingivalis (MOI, 100) for 6 h. The IL-1β and caspase-1 secreted into the culture supernatants (sup.) and the pro-IL-1β and β-actin in the cell lysates (cell) were detected by immunoblotting (A). Cell culture supernatant was assayed for IL-1β secretion by ELISA (B) and the cytoplasmic enzyme LDH (C). The data represent mean values ± SDs (n ≧ 3). (D) Live PMA-primed ASC-GFP cells were transfected with TLR2 and/or TLR4 siRNA for 48 h and then infected with P. gingivalis (MOI, 100) for 6 h. (E to J) THP-1 cells were transfected with MyD88, TRIF, and/or TRAF6 siRNA for 48 h. siRNA-transfected cells were infected with P. gingivalis (MOI, 100) for 6 h. The IL-1β and caspase-1 secreted into the culture supernatants and the pro-IL-1β, procaspase-1, and β-actin in the cell lysates were detected by immunoblotting (E and H). Cell culture supernatant was collected and evaluated for IL-1β secretion by ELISA (F and I) and release of the cytoplasmic enzyme LDH (G and J). The data represent mean values ± SDs (n ≧ 3). (K) Live PMA-primed ASC-GFP cells were transfected with MyD88, TRIF, and/or TRAF6 siRNA for 48 h and then infected with P. gingivalis (MOI 100) for 6 h. (L to N) THP-1 cells were pretreated with PDTC (300 μM) for 1 h prior to infection with P. gingivalis for 6 h. (L) The secreted IL-1β and caspase-1 were detected by immunoblotting. Cell culture supernatant was assayed for IL-1β secretion by ELISA (M) and the release of the cytoplasmic enzyme LDH (N). (O) Live PMA-primed ASC-GFP cells were pretreated with PDTC (300 μM) for 1 h prior to infection with P. gingivalis for 6 h and observed by fluorescence confocal microscopy. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Next, we examined the roles of downstream signaling molecules of the TLR2 and TLR4 pathways. By silencing of MyD88 and/or TRAF6 expression, we observed a significant downregulation of P. gingivalis-induced caspase-1 activation and IL-1β secretion (Fig. 6E and F; see Fig. S3F to H in the supplemental material). Similarly, the silencing of TRIF and/or TRAF6 downregulated the caspase-1 activation and the IL-1β release stimulated by P. gingivalis (Fig. 6H and I; see Fig. S3J and K in the supplemental material). These results show that MyD88 and TRIF, TLR adaptor molecules, are required for the activation of P. gingivalis-induced caspase-1 and IL-1β secretion. Knockdown of MyD88, TRIF, and/or TRAF6 strongly blocked LDH release and the formation of the ASC pyroptosome (Fig. 6G and J and K; see Fig. S3I, L, and M in the supplemental material). These results indicate that TLR signaling is the prerequisite for P. gingivalis-induced inflammasome activation leading to IL-1β secretion. We also used PDTC, an NF-κB inhibitor, to determine whether NF-κB activation is involved in pro-IL-1β induction. The induction of pro-IL-1β expression by P. gingivalis was inhibited by pretreatment of the cells with PDTC (Fig. 6L; see Fig. S3N in the supplemental material). PDTC pretreatment also inhibited P. gingivalis-induced caspase-1 activation, IL-1β secretion, LDH release, and the formation of the ASC pyroptosome (Fig. 6L to O; see Fig. S3N to Q in the supplemental material), suggesting a role of NF-κB for inflammasome activation.

DISCUSSION

Periodontal disease, the major cause of adult tooth loss, is characterized by a chronic inflammation caused by oral bacterial infection. Overgrowth of periodontal pathogens can result from a deficiency in the host defense system or modification of the subgingival environment. P. gingivalis, an oral black-pigmented Gram-negative bacterium, is primarily found in deep periodontal pockets, especially in sites with active disease (29). Upon P. gingivalis infection, a dense infiltration of inflammatory cells, including monocytes and macrophages, occurs in the periodontitis tissues. The inflammatory cells activate cytokine secretion as a defense against bacterial infection. Although cytokines are initially protective in the elimination of infected bacteria, overproduction of proinflammatory cytokines is related to periodontal destruction, including periodontal attachment loss, destruction of collagen, and alveolar bone resorption. The cytokine concentration in GCF and saliva is higher in patients with aggressive periodontitis than in healthy subjects, and these effects are associated with autoimmune diseases (30, 31). Thus, understanding the mechanisms of P. gingivalis-induced immune cell signaling could provide useful information for prevention and therapy of periodontitis. Little is known about the molecular mechanisms by which P. gingivalis infection causes IL-1β secretion in human immune cells. Therefore, we investigated the effect of P. gingivalis infection on inflammasome activation, which is important in the secretion of IL-1β during host innate immune responses.

First, we compared the levels of proinflammatory cytokines in GCF between healthy and periodontitis subjects. As expected, IL-1β and TNF-α production was significantly higher in GCF from periodontitis patients. Next, we measured the protein levels of inflammasome components in healthy and periodontitis gingival tissues and found that the NLRP3 and AIM2 proteins were upregulated in the gingival tissues from periodontitis patients. Moreover, the expression of caspase-1, which converts pro-IL-1β into mature IL-1β, was greatly enhanced in periodontitis gingival tissues. This is the first study reporting the upregulation of NLRP3, AIM2, and caspase-1 in periodontitis gingival tissues. Based on these results, we directly measured the TNF-α and IL-1β secretion from differentiated THP-1 cells after infection with P. gingivalis. P. gingivalis stimulated the secretion of IL-1β in an MOI- and time-dependent manner as well as enhanced the expression of pro-IL-1β. The amount of secreted caspase-1 increased depending on the time and MOI of P. gingivalis infection, without affecting the procaspase-1 levels in the cells. The inhibition of caspase-1 with Z-WEHD-FMK or Z-VAD-FMK blocked P. gingivalis-induced IL-1β secretion in THP-1 cells, suggesting that IL-1β maturation is mediated by active caspase-1.

Pyroptosis is an important event in inflammatory processes by which activated macrophages rapidly release large amounts of activated cytokines, such as IL-1β, into the extracellular space (21). Additionally, pyroptosis can restrict the replication of intracellular pathogens by removing the replicative niche. In Salmonella Typhimurium infection, the activation of the NLRC4 inflammasome and pyroptosis is temporally correlated with bacterial clearance (32). We found that P. gingivalis also induced pyroptic cell death via caspase-1 activation. It is tempting to speculate that pyroptic cell death plays a role in the clearance of P. gingivalis during inflammatory responses, but further studies would be needed to confirm its role.

Our study also showed that the expression of the components of NLRP3 and AIM2 inflammasomes, namely, ASC, NLRP3, AIM2, and caspase-1, was increased in response to P. gingivalis infection and each was required for IL-1β secretion, suggesting that P. gingivalis-induced caspase-1 activation and IL-1β secretion are dependent on the NLRP3 and AIM2 inflammasomes in THP-1 cells. In contrast to our findings, it was recently reported that NLRP3 expression is downregulated by subgingival biofilm in primary gingival fibroblast tissue, partly due to P. gingivalis (33). The discrepancy between studies may be attributed to the different cell types studied or to the difference between a live single-type bacterial infection and a biofilm infection.

The NLRP3 inflammasome is a cytosolic protein complex that is essential for processing and secretion of IL-1β via activation of caspase-1 (1). The diverse types of NLRP3 inflammasome activators, such as uric acid, asbestos, silica, and extracellular ATP, could induce reactive oxygen species synthesis and K⁺ efflux, leading to caspase-1 activation and IL-1β secretion (34). We found that P. gingivalis activates ATP release. Furthermore, the inhibition of K⁺ efflux by increasing the extracellular K⁺ concentration blocked the caspase-1 activation, IL-1β secretion, and pyroptic cell death induced by P. gingivalis. Inhibition of cathepsin B also reduced the caspase-1 activation and IL-1β production induced by P. gingivalis, suggesting a potential role of ATP, K⁺ efflux, and cathepsin B in P. gingivalis-mediated activation of the NLRP3 inflammasome.

AIM2 binding to cytosolic double-stranded DNA (dsDNA) via the HIN200 domain recruits the adaptor ASC through a PYD domain and activates caspase-1 (35). AIM2 controls inflammasome activation, IL-1β secretion, and cell death in response to bacterial dsDNA. Recent research found that AIM2 detects cytosolic DNA released from lysed Listeria monocytogenes or Francisella tularensis (36, 37). In this study, we showed for the first time that P. gingivalis infection induces AIM2 expression and promotes caspase-1 activation, IL-1β secretion, and pyroptic cell death through the AIM2 inflammasome. Moreover, we detected P. gingivalis and its DNA in the cytoplasm of THP-1 cells (data not shown), suggesting that AIM2 can detect P. gingivalis DNA. Therefore, P. gingivalis induces caspase-1 activation and IL-1β secretion via the AIM2 inflammasome as well as the NLRP3 inflammasome. Similarly, NLRP3, AIM2, and NLRC4 inflammasomes all contribute to caspase-1 activation in macrophages with Listeria monocytogenes (38).

The release of IL-1β is strictly controlled by a two-signal system (1), whereas a single signal is generally sufficient for the secretion of other cytokines. First, a transcriptional response must be activated via NF-κB activation, leading to synthesis of pro-IL-1β. Subsequently, the separate second signal pathway initiates the caspase-1-dependent maturation and secretion of IL-1β via inflammasome activation. We found that P. gingivalis infection activates both signals and results in IL-1β secretion. NF-κB-dependent TLR signaling was required for the synthesis of pro-IL-1β by P. gingivalis infection. It was reported that NF-κB-dependent TLR signaling upregulates the expression of NLRP3 (10). Consistent with this, our data showed that P. gingivalis infection stimulated the expression of NLRP3 as well as AIM2 at the mRNA and protein levels.

TLRs are expressed in multiple cell types at the site of periodontal infection, and it was previously shown that P. gingivalis signals inflammatory cytokine production from mouse peritoneal macrophages through both TLR2 and TLR4 (27, 28). Here, we found that P. gingivalis utilizes both TLR2 and TLR4 for stimulation of IL-1β secretion in THP-1 cells. The specific response initiated by individual TLRs depends on the recruitment of a single adaptor protein or combinations of adaptor proteins, such as MyD88, TIRAP, and TRIF. TLR2 and/or TLR4 recruits MyD88 and TRIF to activate NF-κB and mitogen-activated protein kinases for induction of inflammatory cytokines. We found that the adaptor proteins MyD88 and TRIF are required for the effects of P. gingivalis infection on caspase-1 activation and IL-1β secretion. These results imply that the transcription of NLRP3, AIM2, and pro-IL-1β via TLR signaling is a prerequisite for P. gingivalis-induced inflammasome activation and IL-1β secretion.

In summary, infection with P. gingivalis triggered the activation of NLRP3 and AIM2 inflammasomes via TLR2 and TLR4 signaling, leading to IL-1β secretion and pyroptic cell death. In addition, P. gingivalis-induced NLRP3 inflammasome activation was dependent on ATP release, K⁺ efflux, and cathepsin B (Fig. 7). Our study provides novel insight into the innate immune response against P. gingivalis infection which could potentially be used for the potential prevention and therapy of periodontitis.

FIG 7.

Infection with P. gingivalis triggers the activation of NLRP3 and AIM2 inflammasomes via TLR2 and TLR4 signaling, leading to IL-1β secretion and pyroptic cell death.