C57BL/6 mice were purchased from Orient Bio. Nlrp3^−/− mice were obtained from The Jackson Laboratory. Nlrc4^−/− mice were kindly provided by Dr. G. Nunez (University of Michigan, Ann Arbor, MI, USA). All mice were on C57BL/6 background and 8–10 weeks male mice were used for the experiments. All mice were maintained under specific pathogen-free conditions.

Protocols for the animal experiments were approved by the Institutional Ethical Committee, Yonsei University College of Medicine (2013-0164). All experiments were performed in accordance with the approved guidelines of the Institutional Ethical Committee, adhered to Guide for the Care and Use of Laboratory Animals of National Research Council (USA).

Mouse primary bone marrow-derived macrophages (BMDMs) were differentiated from bone marrow cells isolated from mouse femurs. Immortalized wild-type (WT), Asc^−/−, NLRP3-GFP-expressing BMDMs and ASC-GFP-expressing THP-1 cells were kindly provided by Dr. E. S. Alnemri (Thomas Jefferson University, Philadelphia, PA, USA). All BMDMs were maintained in L929 supplements-conditioned Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and antibiotics. THP-1-ASC-GFP cells were grown in RPMI-1640 supplemented with 10% FBS, 2 mM glutamine, 10 mM HEPES, 1 mM sodium pyruvate, and 0.05 mM 2-mercaptoethanol and antibiotics. THP-1-ASC-GFP cells were differentiated into macrophage-like cells using a phorbol 12-myristate 13-acetate treatment and then used for experiments. A549 cells were cultured in DMEM supplemented with 10% FBS and antibiotics. Cell death was determined by the extracellular release of lactate dehydrogenase using a CytoTox96 non-radioactive cytotoxicity assay kit (Promega).

Pseudomonas aeruginosa strain PAO1 was used. PAO1 deletion mutants (ΔlasR/rhlR, ΔlasR, ΔrhlR, ΔfliC, ΔlasR/rhlR/fliC, ΔlasR/fliC, and ΔlasB) were created by allelic replacement as previously described (22). For bacterial infection, all PAO1 strains were grown overnight in Luria-Bertani (LB) medium at 37°C with aeration, then diluted 1:50 with fresh LB medium, and grown for an additional 2 h. Bacteria were then pelleted by centrifugation and resuspended in Opti-MEM medium. The bacterial suspension was then added to the culture medium of BMDMs at the indicated multiplicity of infection. To prepare bacterial lysates, cultured PAO1 were lysed, sonicated, and centrifuged to remove insoluble pellets. The supernatants were normalized and used for immunoblotting.

For preparation of BS, PAO1 strains were grown overnight in LB medium and then diluted with fresh LB medium and grown for an additional 6 h. The bacterial culture was then normalized by the number of bacteria. The normalized bacterial suspension was centrifuged and filtered to remove bacteria, and the final filtrate was used as BS. For heat treatment, BSs were brought to heat at 98°C for 30 min to inactivate bacterial proteases activity. For protease treatment, BSs were incubated with proteinase K (100 μg/ml) at 37°C for 1 h and subsequently boiled at 98°C for 30 min to eliminate the proteinase K activity.

Flagellin purified from P. aeruginosa was purchased from InvivoGen. DOTAP liposomal transfection reagent was purchased from Sigma-Aldrich. Lipopolysaccharide (LPS), ATP, nigericin, pyocyanin, and proteinase K were purchased from Sigma-Aldrich. N-3-oxo-dodecanoyl-l-homoserine lactone (3-oxo-C12-HSL) and N-butyryl-l-homoserine lactone (C4-HSL) were obtained from Cayman. Antibodies were acquired to detect mouse caspase-1 (Adipogen, 20B-0042), mouse IL-1β (R&D, AF-401-NA), mouse IL-6 (Cell Signaling, 12912), mouse NLRP3 (Adipogen, 20B-0014), mouse ASC (Santa Cruz, SC-22514-R), and flagellin (InvivoGen, mabg-flapa).

Cells were harvested and then lysed in 20mM HEPES (pH 7.5) buffer containing 0.5% Non-idet P-40, 50mM KCl, 150mM NaCl, 1.5mM MgCl₂, 1mM EGTA, and protease inhibitors. After centrifugation to remove cell debris, soluble lysates were fractionated by SDS-polyacrylamide gel electrophoresis, transferred to PVDF membranes (Bio-Rad), and then immunoblotted by western blotting. For some experiments, BMDM culture supernatants were collected after inflammasome stimulations, and proteins were precipitated by the addition of a methanol/chloroform mixture as reported previously (23), and then immunoblotted. All blots shown are representative images of at least three independent experiments. Images have been cropped for presentation.

To trigger classical NLRP3 inflammasome activation, BMDMs were primed with LPS (0.25 μg/ml, 3 h) and then treated with ATP (2 mM, 45 min) or nigericin (5 μM, 45 min). To induce AIM2 inflammasome activation, BMDMs were transfected with poly dA:dT using lipofectamine according to the manufacturer’s protocol. Inflammasome activation was then determined by the detection of active caspase-1 p20 and active IL-1β in culture supernatants using immunoblots, and by quantification of extracellular IL-1β using the mouse IL-1β Quantikine ELISA kit (R&D systems). The levels of IL-6 in culture supernatants were quantified by mouse IL-6 Quantikine ELISA kit (R&D systems). For some experiments, BMDM supernatants were incubated with BS, subjected to protein precipitation by methanol/chloroform, and then immunoblotted. To deliver flagellin into the cytosol of BMDMs, flagellin (0.35 μg) was incubated with DOTAP liposomal transfection reagent (3.5 μl) in serum-free Opti-MEM medium (50 μl) at 25°C for 30 min. Then, 300 μl Opti-MEM medium was added, and the final mixture (250 μl total) was used to transfect BMDMs in a six-well plate.

To detect ASC oligomerization, a chemical cross-linking assay was performed as described previously (6). Briefly, cells were lysed and centrifuged at 6,000 rpm for 10 min. The pellets were washed and resuspended in phosphate-buffered saline. The resuspended pellets were cross-linked with 1 mM discuccinimidyl suberate [disuccinimidyl suberate (DSS); Pierce, Rockford, IL, USA] for 30 min. The cross-linked pellets and soluble lysates were fractionated by SDS-PAGE and immunoblotted with anti-ASC antibody. To visualize ASC speck-like aggregates, THP-1 cells stably expressing ASC-GFP were observed by confocal microscopy.

To measure mRNA levels, quantitative real-time PCR was performed. Briefly, total cellular RNA was isolated using the TRIzol reagent (Invitrogen) and reverse transcribed using PrimeScript RT Master Mix (Takara) according to the manufacturer’s instructions. Template DNA was amplified by quantitative real-time PCR using SYBR Premix Ex TaqII (Takara). Primers were as follows: 5′-GCC CAT CCT CTG TGA CTC AT-3′ and 5′-AGG CCA CAG GTA TTT TGT CG-3′ (Il-1b); 5′-AGT TGC CTT CTT GGG ACT GA-3′ and 5′-TCC ACG ATT TCC CAG AGA AC-3′ (Il-6); and 5′-CGC GGT TCT ATT TTG TTG GT-3′ and 5′-AGT CGG CAT CGT TTA TGG TC-3′ (Rn18s).

Selected solvents with different polarities (hexane, ethyl acetate, and acetonitrile) were used to extract metabolites from the samples. Each solvent was added to the same volume of sample, and the mixture was vortexed for 1 min. Samples were centrifuged (2,500 rpm, 5 min), the supernatant was removed and centrifuged again (15,000 rpm, 5 min), and 200 μl of the final supernatant was pipetted into sample vials with inserts.

The UFLC system was equipped with a DGU-20A5R degasser, LC-20AD XR binary pump system, SIL-20AC XR autosampler at 15°C, and CTO-20AC column oven at 35°C (Shimadzu). Chromatographic separation was conducted on a reversed-phase column (Imtakt Scherzo S-C18 column, 100 mm × 2.0 mm, i.d., 3 μm) at flow rate of 0.3 ml min⁻¹. The mobile phases used were (A) water and (B) methanol, both containing 0.1% formic acid. The gradient program was set as follows: 98% (A) and 2% (B), held for 1 min; linear increase from 2 to 95% (B) over 14 min, held for 3 min; return to initial conditions in 0.01 min and equilibration for 1.99 min.

High-resolution mass spectrometry detection was performed on both Exactive Plus Orbitrap and Q-Exactive Plus Quadrupole-Orbitrap mass spectrometers (Thermo) equipped with an electrospray ionization source and operated in positive mode. Screening and confirmation of the substances were performed using full scan mode in Exactive Plus and parallel reaction monitoring (PRM) mode in Q-Exactive Plus. For identification of selected molecules, the scan range and collision energy were applied to each substance at a resolution of 35,000 in PRM mode.

The raw data acquired by Xcalibur software (Thermo) were processed using MZmine2 software (24). After peak picking, mass detection, chromatogram building, and deconvolution, peaks were grouped depending on m/z and retention time value by RANSAC alignment and peak finder. The expected structure of candidates based on isotope rate and MS/MS (PRM mode) were identified by entering the formula and fragment data into mzCloud,¹ MetFrag,² and PubChem.³ The results for the most suitable candidates were then compared to standards by MS and MS/MS spectra.

All values were expressed as the mean and SE of three independent experiments unless otherwise indicated. Data were analyzed using one-way analysis of variance followed by Bonferroni post hoc test after checking the assumptions of normal distribution by Shapiro test and equal variance by Bartlett or Brown–Forsythe test. Otherwise, Welch’s t-tests were used for unequal variances. The p values ≤0.05 were considered significant. Statistical analyses were carried out using GraphPad Prism and R software.

To examine a potential role of P. aeruginosa QS in host inflammasome activation, we prepared a QS-defective ΔlasR/rhlR double mutant strain of P. aeruginosa PAO1 bacteria lacking LasR and RhlR, two principal transcriptional regulators of P. aeruginosa QS system. At 12 h postinfection, ΔlasR/rhlR mutant showed a reduced cytotoxicity to human lung epithelial A549 cells compared with WT P. aeruginosa (Figure 1A), suggesting that QS-deficient ΔlasR/rhlR mutant is less virulent. Then, we assessed inflammasome activation in BMDMs upon infection with WT and QS mutant P. aeruginosa. At 4 h postinfection, both WT and ΔlasR/rhlR mutant strain promoted robust inflammasome activation as determined by the secretion of active caspase-1 (p20) and IL-1β into BMDMs culture supernatant (Figure 1B). Of notice, prolonged infection with WT P. aeruginosa (6 or 8 h), but not with ΔlasR/rhlR, caused a manifest degradation of caspase-1 and IL-1β in the culture supernatants (Figure 1B). Consistently, levels of secreted pro-inflammatory cytokines IL-1β and IL-6 at 8 h postinfection were significantly lower in BMDMs infected with WT P. aeruginosa than in those with the ΔlasR/rhlR mutant (Figures 1C,D). This apparent QS-mediated impairment of inflammasome signaling is not likely a consequence of reduced growth of the WT strain, because the growth kinetics of WT and QS-defective strains were similar (Figure 1E). These observations indicate that prolonged infection with P. aeruginosa could impair host inflammasome-mediated responses in a QS-dependent manner.

To test whether bacterial secreted factors are responsible for the impairment of host inflammasome-associated molecules shown above, we collected bacteria-free supernatant (referred to as bacterial secretant (BS)) from P. aeruginosa cultures (Figure 2A). Simultaneously, we collected culture supernatant from BMDMs upon stimulation with NLRP3 inflammasome-activating LPS/ATP, and this LPS/ATP-stimulated BMDM supernatant was then incubated with BS from the WT P. aeruginosa culture (Figure 2A). Both procaspase-1 and active caspase-1 p20 in the BMDMs supernatant were clearly processed into fragments by BS, as was the IL-1β secreted from BMDMs (Figure 2B).

Pseudomonas aeruginosa produces QS-regulated proteases such as elastase (LasB) and alkaline protease (AprA) to degrade host defense proteins such as extracellular matrix (ECM) components and cytokines (11, 25). To examine whether P. aeruginosa proteases are involved in the degradation of inflammasome-associated molecules, BS was heated to eliminate protease activity. Consequently, heat inactivation of BS markedly abolished the degradation of inflammasome components such as caspase-1 and ASC that was observed after incubation with intact P. aeruginosa secretant (Figure S1A in Supplementary Material). Furthermore, the secretant of the ΔlasB strain, lacking the major P. aeruginosa QS-dependent protease LasB, failed to efficiently degrade inflammasome components (Figure S1B in Supplementary Material), indicating that LasB plays a role in this degradation.

LasR is an essential transcriptional regulator in P. aeruginosa that induces the expression of many QS-related genes including proteases (18). Unlike WT P. aeruginosa secretant, neither ΔlasR/rhlR nor ΔlasR secretant was able to process inflammasome components (Figure 2C), confirming that the QS system is necessary for the degradation of host inflammasome components by BS. Although both LasR and RhlR are the principal transcriptional regulators of P. aeruginosa QS, LasR is a prerequisite regulator for the expression of RhlR (26). Partial processing of caspase-1 was detected in ΔrhlR mutant secretant (Figure 2C), suggesting that the secretant contains a minimal base level of proteases. Indeed, previous studies showed that ΔrhlR mutant produces a kind of proteases including AprA (27, 28).

Caspase-1 and IL-1β secreted from P. aeruginosa-infected BMDMs were also degraded by WT P. aeruginosa secretant, but not by heat-inactivated or ΔlasR/rhlR secretant (Figures S1C,D in Supplementary Material). Along with caspase-1 and IL-1β, aggregates of inflammasome adaptor protein ASC are secreted from BMDMs upon inflammasome activation, which could propagate the inflammatory responses (29, 30). Monomeric ASC in the LPS/ATP-stimulated supernatant was also degraded by WT BS, but not by ΔlasR/rhlR secretant (Figure 2C). In ASC-GFP-expressing THP-1 cells, WT P. aeruginosa secretant clearly reduced nigericin-triggered formation of ASC aggregates, which appear as specks by confocal microscopy, while ΔlasR/rhlR secretant failed to decrease the number of ASC specks (Figure 2D; Figure S1E in Supplementary Material). Consistently, the level of secreted IL-1β from LPS/ATP-treated BMDMs was significantly impaired by the incubation with WT secretant, but not with heat-inactivated nor ΔlasR/rhlR secretant (Figure 2E). Pro-inflammatory cytokine IL-6 secreted by BMDMs was also degraded by incubation with WT secretant, but less with QS mutant secretant (Figures 2F,G). All of these findings clearly demonstrate that BS from P. aeruginosa can degrade host inflammasome components and cytokines, enabling QS-dependent proteases to disrupt innate immune defenses.

Given the host protein-processing effect of the P. aeruginosa QS system, we further examined a potential role of BS, which contains diverse QS-dependent virulence factors, in host inflammasome signaling pathways. Thus, we treated BMDMs with bacteria-free WT or QS-defective P. aeruginosa culture secretants (Figure 3A). To our surprise, BS from ΔlasR/rhlR or ΔlasR mutant strains promoted robust caspase-1 activation and IL-1β secretion, while WT secretant failed to trigger caspase-1 activation (Figure 3B). Meanwhile, ΔrhlR secretant isolated from 6- or 9-h preculture did not cause a robust caspase-1 processing. Because ΔrhlR mutant seems to still secrete some proteases, the deficiency of caspase-1 and IL-1β in the supernatant of BMDMs treated with ΔrhlR secretant could be explained by bacterial protease-mediated degradation.

Consistent with caspase-1 activation, ΔlasR/rhlR mutant secretant, but not that of WT bacteria, caused a robust production of IL-1β and IL-6 by BMDMs (Figures S2A,B in Supplementary Material), while secretants from both WT and ΔlasR/rhlR P. aeruginosa induced upregulation of Il-1β and Il-6 mRNA (Figures S2C,D in Supplementary Material). As for IL-6, it is likely that WT P. aeruginosa secretant could degrade IL-6 secreted from BMDMs via its proteases, because WT P. aeruginosa secretant triggered mRNA production of IL-6, subsequently leading to its secretion. On the other hand, the absence of caspase-1 and IL-1β in the supernatant of BMDMs treated with WT secretant could be explained by two possible scenarios. First, it is plausible that WT secretant may be unable to trigger inflammasome activation. The second hypothesis is that WT secretant may promote inflammasome activation, leading to the secretion of inflammasome proteins, but bacterial secreted proteases could then degrade caspase-1 and IL-1β in the extracellular supernatant. To test these possibilities, we performed similar experiments with heat-inactivated BS to avoid the involvement of proteases. As a result, WT secretant failed to induce caspase-1 activation and IL-1β secretion even after heat inactivation (Figures 3C,D). Instead, heat inactivation resulted in the appearance of procaspase-1 by WT secretant (Figure 3C), suggesting that secreted procaspase-1 may be degraded by bacterial proteases. These data demonstrate that WT secretant is unable to promote inflammasome activation regardless of protease activity. However, heat-treated WT secretant caused robust secretion of IL-6 (Figure 3E), suggesting that proteases from the WT secretant had been responsible for processing IL-6 in the extracellular space. To further clarify the involvement of proteases in the failure of inflammasome activation by WT secretant, BS from ΔlasB mutant, which is devoid of elastase expression, was examined. Treatment with ΔlasB secretant partially restored procaspase-1 in the supernatant but caused no caspase-1 processing (Figure 3F). This finding further supports that WT secretant is unable to promote inflammasome activation.

Next, we examined which sensor molecule is implicated in inflammasome activation by ΔlasR/rhlR secretant. In accordance with previous studies, P. aeruginosa infection in BMDMs resulted in NLRC4-dependent inflammasome activation (Figure 4A). Likewise, inflammasome activation by ΔlasR/rhlR secretant was abrogated in Nlrc4^−/− BMDMs (Figure 4B), indicating that ΔlasR/rhlR secretant triggered NLRC4-dependent inflammasome activation. A deficiency of ASC also diminished the caspase-1 activation and IL-1β secretion by ΔlasR/rhlR secretant (Figure 4C). By contrast, robust inflammasome activation was observed in Nlrp3-deficient BMDMs in response to ΔlasR/rhlR secretant (Figure 4D). Therefore, our data demonstrated that ΔlasR/rhlR secretant promotes NLRC4-mediated, but not NLRP3-mediated inflammasome activation.

Previously, it was shown that Salmonella infection triggered ASC oligomerization, considered as an essential event of inflammasome assembly (31, 32). Similarly, P. aeruginosa infection caused a robust oligomerization of ASC as determined by a DSS-mediated cross-linking assay (Figure S3 in Supplementary Material). The secretant of ΔlasR/rhlR P. aeruginosa, but not of WT, also triggered the formation of ASC dimeric and oligomeric structures only in the presence of NLRC4 (Figures 5A,B). The failure of WT secretant to induce ASC oligomerization further supports the hypothesis that it could not activate inflammasome signaling regardless of protease-mediated protein degradation. Moreover, even heat-inactivated WT secretant failed to promote the oligomerization of ASC inside cells (Figure 5C). Consistent with these data, BS from ΔlasR/rhlR mutant P. aeruginosa, but not from WT, triggered the formation of speck-like aggregates of ASC in human monocytic THP-1 cells expressing ASC-GFP (Figure 5D). These findings collectively demonstrate that the ΔlasR/rhlR mutant secretant, but not WT secretant, promotes the assembly and activation of NLRC4 inflammasome.

To further provide molecular insight into the inflammasome-activating capability of ΔlasR/rhlR secretant, we first examined whether specific protein components of BS drive inflammasome activation. Heat inactivation of ΔlasR/rhlR secretant significantly reduced caspase-1 activation and IL-1β secretion (Figure 6A; Figure S4A in Supplementary Material). This is consistent with the above results in Figures 3C,D, and these findings demonstrate that heat treatment significantly impairs BS-promoted inflammasome activation. Moreover, proteinase K treatment of ΔlasR/rhlR secretant completely abolished the inflammasome activation (Figure 6A; Figure S4A in Supplementary Material). These data strongly indicate that unidentified proteins in the secretant could be responsible for the secretant-triggered inflammasome activation. According to previous report, cytosolic presence of bacterial flagellin is critical to induce NLRC4 inflammasome activation upon P. aeruginosa infection (21). Related to this finding, we found that BS from the ΔfliC mutant lacking FliC flagellin was unable to promote inflammasome activation (Figure S4B in Supplementary Material). Intriguingly, the ablation of FliC in ΔlasR/rhlR or ΔlasR P. aeruginosa completely abolished the inflammasome activation by ΔlasR/rhlR or ΔlasR BS (Figure 6B), suggesting that flagellin in ΔlasR/rhlR secretant might be the main stimulus for inflammasome activation.

To verify this possibility that extracellular flagellin released from P. aeruginosa could trigger inflammasome activation, we treated BMDMs with purified flagellin from P. aeruginosa. Previous reports showed that intracellular delivery of flagellin using a liposomal agent such as DOTAP is essential for activation of the NLRC4 inflammasome (3, 33). Treatment of flagellin for 1–3 h failed to activate caspase-1 and IL-1β secretion in BMDMs (3, 21). However, in our study, 6-h incubation of flagellin in the extracellular medium was able to promote caspase-1 activation and IL-1β secretion (Figure 6C). This extracellular flagellin-triggered inflammasome activation depended on NLRC4 but not NLRP3 (Figures 6D,E). To explain WT secretant’s inability to activate the inflammasome, the presence of flagellin was assayed in the secretant from P. aeruginosa cultures. Of interest, intact flagellin was observed in the BS of the ΔlasR/rhlR mutant, whereas flagellin was severely impaired in the WT secretant (Figure 6F). Indeed, flagellin levels were markedly reduced in WT secretant, but not in QS-defective mutants secretant (Figure S4C in Supplementary Material). By contrast, no degradation of flagellin was observed in the bacterial lysate of WT P. aeruginosa (Figure 6G), indicating that secreted flagellin could be degraded by bacterial proteases in the secretant rather than within the cells. The degradation of flagellin was also observed, but slightly reduced in ΔlasB mutant, suggesting that both elastase and other protease are responsible for the impairment of flagellin (Figure 6H). These data demonstrate that QS-dependent production of protease from WT P. aeruginosa culture led to degradation of flagellin and thereby eliminated its ability to activate the NLRC4 inflammasome.

To further delineate the potential contribution of QS to inflammasome activation, metabolomic analysis was performed with BSs from WT and ΔlasR/rhlR P. aeruginosa using liquid chromatography–mass spectrometry. Unfortunately, under our experimental conditions, we did not identify any candidate molecule that was only present in ΔlasR/rhlR secretant, but not in WT secretant. However, pyocyanin was identified as the sole candidate molecule present in WT secretant but not in ΔlasR/rhlR secretant (Figure 7A; Figures S5A,B in Supplementary Material). Pyocyanin is one of the well-known QS-dependent major virulence factors of P. aeruginosa (17, 18) and was reported to damage host cells via its oxidative stress-based cytotoxicity (34, 35). Thus, we tested a potential role of pyocyanin in the inflammasome activation. Of note, pyocyanin remarkably suppressed the inflammasome activation and IL-1β secretion triggered by ΔlasR/rhlR secretant (Figures 7B,C). Furthermore, pyocyanin treatment resulted in a strong suppression of NLRP3-mediated caspase-1 activation and IL-1β secretion upon LPS/ATP or LPS/nigericin stimulation (Figures 7D,E). Pyocyanin also significantly attenuated LPS/nigericin-triggered cell death of BMDMs (Figure 7F). This might be due to reduced caspase-1 activation by pyocyanin. However, pyocyanin failed to affect poly dA:dT-triggered AIM2 inflammasome activation (Figure 7G). Supporting these observations, pyocyanin clearly reduced NLRC4- or NLRP3-dependent ASC oligomerization in response to ΔlasR/rhlR secretant or LPS/nigericin treatment (Figures 7H,I). These findings demonstrate that the QS-dependent virulence factor pyocyanin could suppress the activation of both the bacterial infection-promoted NLRC4 inflammasome and the host endogenous danger signals-triggered NLRP3 inflammasome.

Bacterial QS systems use acylated homoserine lactones as a QS signaling regulator, called an autoinducer. In the case of P. aeruginosa, two signaling molecules—3-oxo-C12-HSL, produced by LasI, and C4-HSL, produced by RhlI—are employed for bacterial communication to control the expression of many virulence factors (18). So far, these QS autoinducers have never been examined for their potential effect on inflammasome signaling. Interestingly, 3-oxo-C12-HSL, but not C4-HSL, significantly dampened inflammasome activation triggered by ΔlasR/rhlR secretant (Figure 8A). Moreover, 3-oxo-C12-HSL completely blocked the NLRP3 inflammasome activation in response to LPS/ATP stimulation (Figure 8B). This P. aeruginosa autoinducer 3-oxo-C12-HSL also strongly attenuated IL-1β secretion upon ΔlasR/rhlR secretant or LPS/ATP treatment (Figures 8C,D). Similar to pyocyanin, two P. aeruginosa autoinducers showed no effect on the AIM2-dependent caspase-1 activation triggered by poly dA:dT transfection (Figure 8E), indicating that 3-oxo-C12-HSL could suppress NLRC4 and NLRP3 inflammasome signaling. Supporting these findings, 3-oxo-C12-HSL diminished the NLRC4- and NLRP3-mediated ASC oligomerization by cross-linking experiments (Figures 8F,G). However, NLRP3 inflammasome-dependent cell death was not abolished, but rather increased by 3-oxo-C12-HSL (Figure 8H). Indeed, 3-oxo-C12-HSL showed a basal cytotoxicity to BMDMs regardless of inflammasome activity. Of notice, QS-associated factors, pyocyanin and 3-oxo-C12-HSL, significantly inhibited the oligomerization of NLRP3 in response to classical NLRP3 stimulators (Figure 8I; Figure S6 in Supplementary Material). These data suggest that the suppression of inflammasome signaling by two QS-related factors might be the upstream event of oligomerization of inflammasome sensor molecule such as NLRP3. These findings collectively demonstrate that two QS-related products, pyocyanin and 3-oxo-C12-HSL, are bona fide inhibitors of inflammasome activation, and that bacterial QS could provide an evasion mechanism to avoid the host inflammasome defense system.