C57BL/6J mice were obtained from the Laboratory Animal Center of University of Science and Technology of China. Nlrp3^-/- mice and NLRP3^Y30E mutant (Nlrp3^Y30E/Y30E ) mice were described previously (33, 34). All animals were SPF, housed under a 12 light/12 dark cycle at 20-22°C, with water and food accessible at all times. All of the mouse experiments were approved by the Animal Care Committee of University of Science and Technology of China.

HDM was from Greer Laboratories. Anti-mouse IL-1β (p17) (AF-401-NA) was from R&D. Anti-mouse caspase-1 (p20) (AG-20B-0042) and anti-NLRP3 (AG-20B-0014) antibodies were from Adipogen. Anti-ASC (67824) antibody was from Cell Signaling Technology. Anti-β-actin (66009-1-Ig) was from Proteintech.

Anti-mouse antibodies used for flow cytometry were: CD3-FITC (BD, 553062, 145-2C11), CD19-FITC (eBioscience, 11-0193-82, 1D3), Ly-6G-PE (Biolegend, 127608, 1A8), CD45-PE (eBioscience, 12-0451-81, 30-F11), CD11c-PerCP-Cy5.5 (Biolegend, 117328, N418), CD11b-PerCP-Cy5.5 (BD, 550993, M1/70), CD11b-PE-Cy7 (eBioscience, 25-0112-82, M1/70), CD3e-PE-Cy7 (BD, 552774, 145-2C11), CD19-PE-Cy7 (Biolegend, 115520, 6D5), SiglecF-Alexa Fluor 647 (BD, 562680, E50-2440), MHCII-APC-eFluor 780 (eBioscience, 47-5321-82, M5/114.15.2), Ly-6G-BV421 (Biolegend, 127628, 1A8), CD45-BV510 (Biolegend, 103137, 30-F11), CD11c-BV510 (Biolegend, 117338, N418). Ultrapure LPS was obtained from Invitrogen. Nigericin was obtained from Sigma-Aldrich. RRx-001 (S8405) was from Selleck.

Wild type, Nlrp3^-/- , Nlrp3^Y30E/Y30E mice were anesthetized by intraperitoneal injection of pentobarbital sodium and instilled i.n. with 10μg HDM in 40μl saline on day 0. For days 7-11, mice were challenged daily with 10μg HDM in 40μl saline. Mice were euthanized for analysis on day 14. Mice in the control group were instilled with 40μL saline on day 0 and days 7-11.

Supernatants from BALF, serum, tissue lysates and cell culture were assayed for mouse IL-1β, IL-5, IL-13 or IL-6 (R&D Systems) according to the manufacturer’s instructions. Supernatants from BALF and cell culture were assayed for mouse IL-4 and IFN-γ (Novus) according to the manufacturer’s instructions. Mice serum was assayed for total IgE (BD) according to the manufacturer’s instructions. For HDM-specific IgE measurement, 5μg HDM was coated for each well overnight at 4°C. Then the plate was washed and conjugated with mice serum for 2h at room temperature. The remaining steps were the same as the manufacturer’s instructions of mouse IgE ELISA kit (BD).

Total RNA of mice lung tissue was isolated with TRIzol (Takara). RNA was applied to generate cDNA using the RevertAid H Minus First Strand cDNA Synthesis Kit (ThermoFisher) according to the manufacturer’s instructions. Quantitative PCR was performed using the Universal SYBR Green Fast qPCR Mix (Abclonal) in a LightCycler 96 Real-Time PCR System (Roche). Gapdh was used as a reference gene. The sequences of primers were as follows:

Mice were euthanized after HDM or saline treatment. Cut the skin on the neck and chest carefully with surgical scissors, cut open the chest cavity to connect to the atmosphere. Expose the trachea, cut a small opening in the trachea and insert the gavage needle to inject 1mL PBS containing 5mM EDTA into the lungs. After 1min filling, the BALF was absorbed by an 1mL syringe. Then repeat the perfusion and aspiration 3 times without waiting for filling. The supernatants of the first BALF were used to measure cytokines. All BALF cells were collected for FACS analyses.

After mice were euthanized, the MLN was isolated and ground into single cells suspensions. For one sample, 1x10⁶ MLN cells were cultured at 37°C under 5% CO2 in RPMI 1640 medium supplemented with 10% FBS, 1mM sodium pyruvate, 2mM L-glutamine and 50μg/mL HDM for 5 days. The supernatants were collected after 5-day-culture to detect the cytokines released by MLN cells.

After the 14-d HDM treatment, the mice were euthanized and the lung tissue was removed after perfusion with PBS, cut into small pieces using scissors and digested in 400U/mL collagenase I (Sigma-Aldrich) for 1h in a 37°C shaking incubator (200r.p.m.). The digestion was terminated by diluting with PBS. After lysing red blood cells, the suspension was filtered using a cell strainer (70μM), and then stained with fluorochrome-conjugated antibodies for 30 min. Washed antibody-marked cells were analyzed by BD Verse cytometers or sorted by a FACSAria flow cytometer (BD). BALF cells isolation was described before. After the 14-d HDM or saline treatment, the mice were euthanized. The BALF cells were isolated, stained with fluorochrome-conjugated antibodies and analyzed by BD Verse cytometers. For the caspase-1 activity assay, the cells were stained with FAM-FLICA fluorescent probe (Immunochemistry Technologies) for 40min in a 37°C incubator. And then the cells were washed, stained with antibodies and analyzed by flow cytometry. A description of the gating strategy is provided in Supplementary Figure 1 .

Lungs were excised and fixed in 4% formalin. Then the fixed lungs were paraffin-embedded. They were sectioned and stained with H&E or PAS. The photos were taken by TissueFAXS PLUS (TissueGnostic Gmbh).

BMDMs were derived from C57BL6 mice and cultured in 37°C in DMEM containing 10% FBS, 1mM sodium pyruvate, 2mM L-glutamine in the presence of 50ng/mL recombinant M-CSF (R&D Systems). AMs were derived from C57BL6 mice BALF and cultured in a complete medium.

All values were expressed as mean ± SEM. Unpaired t-test (GraphPad Prism) was used in all experiments. When P-value < 0.05, the data were considered significant.

To determine whether the inflammasome is activated in allergic asthma, we used an HDM-induced allergic asthma model ( Figure 1A ). Bronchoalveolar lavage fluid (BALF) cells were segregated as CD45^- cells, eosinophils, alveolar macrophages, neutrophils, T cells, B cells and dendritic cells ( Supplementary 1A ). BALF cells from HDM-challenged mice had more eosinophils, neutrophils and lymphocytes than saline-treated control mice after 14 days of HDM treatment ( Supplementary 2A ). In contrast, alveolar macrophage counts were decreased in HDM-treated asthmatic mice ( Supplementary 2A ). After HDM stimulation, mature IL-1β in the lung lysates of asthmatic mice was significantly increased compared with that in control mice ( Figures 1B, D ). However, the IL-1β concentration in serum was not affected by HDM challenge ( Figure 1C ), indicating that HDMs induced local IL-1β secretion in the lungs rather than systemic inflammation. The expression of mature caspase-1 was also elevated in the lungs of HDM-treated mice ( Figure 1D ). Taken together, these results suggest that HDMs stimulate the lung to produce the proinflammatory cytokine IL-1β.

Five receptor proteins have been confirmed to assemble inflammasomes, including NLRP1, NLRP3, NLRC4, AIM2 and pyrin (35). We found that only NLRP3 was highly expressed in mouse lungs, and the expression of NLRP3 was significantly increased after HDM challenge ( Figures 1E, F ). Therefore, we hypothesized that the maturation of IL-1β in the lungs was dependent on the activation of the NLRP3 inflammasome in HDM-induced asthma. Next, we used HDM to induce allergic asthma in Nlrp3^-/- mice and wild-type mice. The results showed that NLRP3 knockout attenuated IL-1β production in asthmatic lungs. HDM treatment induced caspase-1 cleavage and IL-1β maturation in wild-type mouse lungs, but not in Nlrp3^-/- mouse lungs ( Figures 1F, G ), indicating that inflammasome activation in asthmatic lungs is dependent on NLRP3. Collectively, HDM challenge promotes the activation of the NLRP3 inflammasome in the lungs of asthmatic mice.

Because we found that the activation of the NLRP3 inflammasome was induced by the HDM challenge, we next investigated which cell profile was involved in the activation of the NLRP3 inflammasome in the lungs. To detect NLRP3 expression in lung cell subsets, we used Nlrp3^-/- mice in which the eGFP fluorescence intensity correlated with NLRP3 expression (33, 36). We found that in asthmatic mice, NLRP3 was expressed in the myeloid cells, including alveolar macrophages, neutrophils and dendritic cells, but not in eosinophils, T cells, B cells or CD45^- cells ( Figures 2A, B ). The activation of caspase-1 in lung cells was evaluated by a fluorescently labeled inhibitor of caspase (FLICA) probe, and caspase-1 activation was detected in all three cell types expressing NLRP3 during the steady state. However, in asthmatic lungs, the activation of caspase-1 was upregulated only in alveolar macrophages, while in neutrophils and dendritic cells there was no significant difference in caspase-1 activation between control and asthmatic mice ( Figures 2C, D ). We next isolated AMs, neutrophils and DCs from single-cell suspensions of healthy and asthmatic mouse lung tissue to detect the expression of IL-1β and NLRP3. Similarly, the expression of IL-1β was increased in alveolar macrophages, while it remained unchanged in neutrophils and dendritic cells after asthma was induced ( Figure 2E ). The expression of NLRP3 was not changed in alveolar macrophages, neutrophils or dendritic cells ( Figure 2F ), but was increased in asthmatic lungs because the percentage of inflammatory cells was upregulated during asthma ( Figure 1E ,  F and Supplementary 2A ). Treatment with HDMs did not promote caspase-1 activation or IL-1β secretion in bone marrow-derived macrophages (BMDMs) ( Supplementary 3A ). In vitro HDM stimulation did not promote inflammasome activation in isolated alveolar macrophages from wild-type mice either ( Supplementary 3B ). The reason for this might be that alveolar macrophages do not recognize HDMs directly but respond to danger signals released by damaged cells or other signals caused by HDM stimulation in vivo. Taken together, our results demonstrated that HDM-induced NLRP3 inflammasome activation in the lungs was dependent on alveolar macrophages.

Because the NLRP3 inflammasome was activated by HDM stimulation in vivo, we next investigated whether NLRP3 affected the development of allergic asthma. We used an HDM-induced asthma model in Nlrp3^-/- mice and wild-type mice. Compared with wild-type mice, the counts of total BALF cells were reduced in Nlrp3^-/- mice exposed to HDM ( Figure 3A ). Compared with wild-type mice, BALF from Nlrp3^-/- mice contained lower counts of eosinophils and neutrophils ( Figure 3A ). Total IgE and HDM-specific IgE in the serum were also detected, revealing that Nlrp3^-/- mice had decreased concentrations of serum IgE, indicating a reduced T_H2 response and less HDM-specific B cell maturation in Nlrp3^-/- mice ( Figure 3B ). Lung histology studies also showed decreased peribronchial and perivascular leukocyte infiltration in the airways of Nlrp3^-/- mice after HDM challenge ( Figure 3C ). PAS staining showed that the mucus production and goblet cell hyperplasia induced by HDM exposure were diminished in Nlrp3^-/- mouse lungs ( Figure 3C ). To further investigate the HDM-induced T_H2 response, we measured cytokine secretion in BALF. BALF from HDM-treated Nlrp3^-/- mice exhibited significantly less IL-4, IL-5 and IL-13 production compared with BALF from HDM-treated wild-type mice ( Figure 3D ). IL-6 production was also reduced in Nlrp3^-/- BALF ( Figure 3D ). BALF from HDM-treated Nlrp3^-/- mice had a comparable number of lymphocytes and IFN-γ secretion to wild-type mice, and NLRP3 impaired only the T_H2 subset of helper T cells in HDM-induced allergic asthma. Next, we isolated cells from the mediastinal lymph nodes (MLNs) and restimulated these cells with HDMs in vitro. We also found reduced production of IL-4, IL-5, IL-6 and IL-13 in Nlrp3^-/- MLNs, while there was no difference in the concentration of IFN-γ between HDM-treated wild-type and Nlrp3^-/- mice ( Figure 3E ). Collectively, these results show that NLRP3 is required for the promotion of T_H2 responses and the pathological damage caused by asthma.

To explore whether activation of the NLRP3 inflammasome is involved in the progression of asthma, we applied the HDM model in Nlrp3^Y30E/Y30E mutant mice. The Nlrp3^Y30E/Y30E mutation does not affect NLRP3 expression but completely inhibits the NLRP3 inflammasome activation through blockade of NLRP3 dephosphorylation at Tyr30. We found that HDM-induced eosinophil infiltration was interrupted in Nlrp3^Y30E/Y30E mice, aligning with the phenotype in Nlrp3^-/- mice ( Figure 4A ). Nlrp3^Y30E/Y30E mice exposed to HDM showed decreased numbers of infiltrating inflammatory cells and mucus secretion compared with wild-type mice, as determined by H&E and PAS staining of airway cells ( Figure 4B ). The results demonstrated that NLRP3 promoted inflammation in asthma by forming inflammasomes. Thus, in response to HDM stimulation, NLRP3 promotes lung inflammation and pathological damage in an inflammasome-dependent manner.

Nlrp3^-/- mice were resistant to lung inflammation in HDM-induced allergic asthma. Owing to the development of NLRP3 inflammasome inhibitors, we investigated whether the inhibition of NLRP3 in vivo had a therapeutic effect on asthmatic mice. RRx-001 is a highly selective and potent NLRP3 inhibitor and has beneficial effects on NLRP3-driven inflammatory diseases (37). RRx-001 (1-bromoacetyl-3,3-dinitroazetidine) is an anticancer agent currently in phase III clinical trials and is a well-tolerated agent without clinically significant toxic effects (38). Thus, we explored the effect of this safe and effective NLRP3 inhibitor on asthma. HDM-treated mice were intraperitoneally injected with RRx-001 or vehicle on days 7, 9, and 11 ( Figure 5A ), and mice were sacrificed on day 14 to detect airway inflammation. The results showed that treatment with RRx-001 (10mg/kg) significantly inhibited HDM-induced infiltration of eosinophils, neutrophils and lymphocytes in the airway and suppressed eosinophilic airway inflammation ( Figure 5B ). In contrast to vehicle-treated mice, total IgE and HDM-specific IgE in the serum were decreased in RRx-001-treated mice ( Figure 5C ). RRx-001-treated mice displayed reduced peribronchial and perivascular infiltration of inflammatory cells and lower goblet-cell hyperplasia ( Figure 5D ). The production of the T_H2 response cytokines IL-4, IL-5 and IL-13 in the BALF and the supernatants of MLN lymphocytes restimulated with HDM in vitro were severely reduced in RRx-001-treated mice ( Figures 5E, F ). IL-6 production was also abrogated by RRx-001 treatment, while IFN-γ secretion was not altered ( Figures 5E, F ). These phenomena were consistent with Nlrp3^-/- mice. These results suggested that the NLRP3 inhibitor RRx-001 reduced the inflammation in the lungs and had therapeutic effects on HDM-induced allergic asthma.

To further investigate whether the therapeutic effects of RRx-001 on asthma are dependent on NLRP3, we treated HDM-challenged Nlrp3^-/- mice and wild-type mice with RRx-001 and vehicle. The experimental results revealed no significant differences in the number of eosinophils, neutrophils or lymphocytes in the BALF of Nlrp3^-/- mice between the RRx-001-treated group and the vehicle-treated group ( Figure 6A ). While RRx-001 inhibited BALF cells increases in wild-type mice, it had no effect on the inflammatory cell infiltration of Nlrp3^-/- mice in HDM-induced asthma. RRx-001 did not affect HDM-induced pathological lung damage or goblet cell mucus secretion in Nlrp3^-/- mice ( Figure 6B ). These results demonstrated that the therapeutic effects of RRx-001 on allergic asthma were dependent on NLRP3.