Activation of the nucleotide-binding oligomerization domain–like receptor family, pyrin domain–containing 3 (NLRP3) inflammasome initiates an inflammatory response, which is associated with host defense against pathogens and the progression of chronic inflammatory diseases such as gout and atherosclerosis. The NLRP3 inflammasome mediates caspase-1 activation and subsequent IL-1β processing in response to various stimuli, including extracellular ATP, although the roles of intracellular ATP (iATP) in NLRP3 activation remain unclear. In this study, we found that in activated macrophages artificial reduction of iATP by 2-deoxyglucose, a glycolysis inhibitor, caused mitochondrial membrane depolarization, leading to IL-1β secretion via NLRP3 and caspase-1 activation. Additionally, the NLRP3 activators nigericin and monosodium urate crystals lowered iATP through K+- and Ca2+-mediated mitochondrial dysfunction, suggesting a feedback loop between iATP loss and lowering of mitochondrial membrane potential. These results demonstrate the fundamental roles of iATP in the maintenance of mitochondrial function and regulation of IL-1β secretion, and they suggest that maintenance of the intracellular ATP pools could be a strategy for countering NLRP3-mediated inflammation.

Oligomerization of the nucleotide-binding oligomerization domain–like receptor family, pyrin domain–containing 3 (NLRP3) inflammasome complex triggers the initiation of an inflammatory response, which is associated with host defense against pathogens and the progression of chronic inflammatory diseases such as gout, type 2 diabetes, and atherosclerosis (1). The NLRP3 inflammasome consisting of NLRP3, the adaptor protein ASC, and caspase-1 can activate caspase-1, leading to the processing and secretion of the proinflammatory cytokines IL-1β and IL-18 (2).

A large variety of stimuli can activate the NLRP3 inflammasome. These pathogen-associated and damage-associated molecular patterns include particulate factors such as monosodium urate (MSU) crystals, basic calcium phosphate crystals, cholesterol crystals, and soluble factors such as extracellular ATP and nigericin (36). The diverse structures of pathogen-associated and damage-associated molecular patterns have led to the hypothesis that NLRP3 detects common intracellular changes that are shared by these stimuli, such as changes in redox states, ion concentration, and cellular stress. Inhibition of reactive oxygen species (ROS) production leads to suppression of NLRP3 inflammasome activation and IL-1β secretion (5, 6). Additionally, it seems that elevated intracellular or extracellular Ca2+ concentration causes IL-1β secretion through mitochondrial damage or calcium-sensing receptors, respectively, although the role of Ca2+ in NLRP3 activation is still controversial (79). Furthermore, reduction of intracellular K+ concentration has been reported to be the common step that is necessary and sufficient for caspase-1 activation (7). Therefore, these may be upstream events for NLRP3 activation and IL-1β secretion.

A change in cellular metabolic state might be a signal that activates the NLRP3 inflammasome because of the close relationship between the innate immunity, inflammation, and metabolic changes (8). In fact, it has been reported that NLRP3 activators deplete NAD (NAD+), leading to the limitation of the NAD+-dependent deacetylase SIRT2 activity, the polymerization of tubulin cytoskeleton, the translocation of NLRP3, and the subsequent secretion of IL-1β (9). Additionally, immune cells during infection and inflammation undergo a metabolic shift from oxidative phosphorylation to glycolysis to meet increased energy demand (10). A rapid increase in intracellular ATP (iATP) is required for cellular functions such as the biosynthesis of proinflammatory cytokines and the maintenance of mitochondrial membrane potential (Δψm) (11). Thus, decreased iATP levels might be a potent signal that activates the NLRP3 inflammasome.

We proposed that iATP, which plays a fundamental role in the maintenance of mitochondria function, is also implicated in regulation of IL-1β secretion. To test this, we modulated iATP levels produced by either glycolysis or oxidative phosphorylation in activated macrophages and investigated the consequences on Δψm and IL-1β secretion.

Bone marrow cells were isolated from the tibia and femurs of wild-type C57BL/6 mice. For differentiation into bone marrow–derived macrophages (BMDMs), the isolated cells were incubated for 6–7 d on petri dishes in DMEM (Life Technologies, Carlsbad, CA) with 30% L929 conditioned medium (source of M-CSF), 10% FBS (PAA Laboratories, Pasching, Austria), 1% HEPES (Life Technologies), and 1% penicillin-streptomycin (Life Technologies). After differentiation, cells were detached using cold PBS and plated for stimulation experiments in RPMI 1640 medium (Life Technologies) with 10% FBS, 1% HEPES, and 1% penicillin-streptomycin.

Human monocytes were purified from peripheral blood from healthy donors with the Ficoll-Paque gradient and subsequent MACS monocyte isolation kit (Miltenyi Biotec, Bergisch Gladbach, Germany). Human macrophages were obtained by culturing monocytes for 7 d in the presence of 50 ng/ml human M-CSF (Miltenyi Biotec). All experimental procedures were conducted in accordance with the Guiding Principles for the Care and Use of Human Tissues (Teijin Pharma, Tokyo, Japan), and each experimental protocol was approved by the Committee for the Experiments with Human Tissues in the Teijin Institute for Bio-Medical Research. All donors gave informed consent before collecting peripheral blood. THP-1 cells stably expressing short hairpin (sh)RNA against lamin (mock), NLRP3 (shNLRP3), caspase-1 (shCasp1), and ASC (shASC) were generated as described previously (12). The resulting THP-1 cells were cultured in RPMI 1640 medium with 10% FBS, 1% HEPES, 1% penicillin-streptomycin, and 1 μg/ml puromycin (InvivoGen, San Diego, CA).

BMDMs, THP-1 cells, or human macrophages were plated in 96-well, 96-well half area clear/bottom black, 48-well, or 6-well plates at a density of 1 × 105, 0.5 × 105, 1.5 × 105, or 1.5 × 106 cells/well, respectively. BMDMs and human macrophages were primed overnight with 100 ng/ml Pam3CSK4 (InvivoGen) and 200 ng/ml ultrapure LPS (InvivoGen), respectively. THP-1 cells were primed 72 h with 200 nM PMA. Then, primed cells were incubated 30 min with vehicle or the indicated concentrations of inhibitors in incomplete medium (without FBS), and then stimulated with nigericin (2.5 μM, AppliChem, Darmstadt, Germany), MSU crystal (0.25 mg/ml), octacalcium phosphate (OCP) crystal (0.1 mg/ml), or hydroxyapatite (HA) crystal (0.2 mg/ml) for the indicated time. For iATP depletion, cells were treated with vehicle or the indicated concentrations of inhibitors in glucose-free RPMI 1640 medium (Life Technologies) with 1% HEPES, 1% penicillin-streptomycin, and the indicated concentrations of 2-deoxyglucose (2DG; Sigma-Aldrich, St. Louis, MO). The inhibitors used in this study were the following: oligomycin (5 μM, Sigma-Aldrich), KCl (30 mM), z-VAD-fmk (20 μM, Enzo Life Sciences), EGTA (2.5 mM, AppliChem), E64d (50 μM, Enzo Life Sciences), and cytochalasin D (2 μM, Sigma-Aldrich).

Intracellular or extracellular ATP was measured using CellTiter-Glo luminescent cell viability assay (Promega, Madison, WI) according to the manufacturer’s instructions. Luminescence was measured with a luminometer, and ATP concentration was calculated by using an ATP standard curve. iATP content was expressed as nanomoles per milligram protein by protein concentration that was determined by the BCA method.

Mouse and human IL-1β, TNF-α, MCP-1, and IL-6 ELISA kits (eBioscience, San Diego, CA) and an IL-1α ELISA kit (BioLegend, San Diego, CA) were used to measure the corresponding chemokine/cytokine levels in supernatants according to the manufacturers’ instructions.

Lactate dehydrogenase (LDH) in supernatant was measured using CytoTox-ONE homogeneous membrane integrity assay (Promega) according to the manufacturer’s instructions. LDH release (%) was calculated by using the following formula: LDH release (%) = (value in sample − background)/(value in Triton X-100–treated sample − background) × 100.

Mitochondrial ROS level was measured with dihydrorhodamine 123 (DHR123, Life Technologies), which is a nonfluorescent ROS indicator that can passively diffuse across membranes where it is oxidized to fluorescent rhodamine 123, which localizes in the mitochondria. Briefly, BMDMs in 96-well half area clear/bottom black plates were primed and then incubated in glucose-free medium containing 10 mM 2DG for 90 min. After incubation, cells were loaded 30 min with 2 μM DHR123, and fluorescence intensity on 500 nm (excitation) and 536 nm (emission) was measured with a fluorescence plate reader.

The Δψm was measured with tetramethylrhodamine, ethyl ester (TMRE, Life Technologies), which is a fluorescent dye that is readily sequestered by active mitochondria depending on Δψm. Depolarized or inactive mitochondria have decreased membrane potential and then fail to sequester TMRE. Briefly, BMDMs in 96-well half area clear/bottom black plates were pretreated 30 min with vehicle or EGTA and then stimulated 90 min with medium, nigericin, or MSU. For iATP depletion, cells were incubated 90 min with vehicle or z-VAD-fmk in glucose-free medium containing 10 mM 2DG. Carbonyl cyanide 3-​chlorophenylhydrazone (CCCP, Sigma-Aldrich), an uncoupler of oxidative phosphorylation in mitochondria, was used to depolarize Δψm. After stimulation, cells were loaded 30 min with 500 nM TMRE, and fluorescence intensity on 549 nm (excitation) and 575 nm (emission) was measured with a fluorescence plate reader.

Lactate in the supernatant was measured by using Amplite colorimetric l-lactate assay kit (AAT Bioquest, Sunnyvale, CA) according to the manufacturer’s instructions.

Equal amounts of supernatants were loaded to SDS-PAGE, and immunoblot was performed by anti–caspase-1 Ab (Casper-1, AdipoGen, San Diego, CA) and by anti-IL-1β Ab (AF-401-NA, R&D Systems, Minneapolis, MN).

Cell viability was measured with double staining kit (Dojindo, Kumamoto, Japan) based on calcein-AM staining for live cells and propidium iodide (PI) staining for dead cells. Briefly, BMDMs plated in 96-well half area clear/bottom black plates were prime and incubated in glucose-free medium containing 10 mM 2DG for 6 h. For dead cell preparation for positive control, cells were treated 1 min with ethanol (70%). After incubation, cells were stained 15 min with 2 μM calcein-AM and 4 μM PI, and fluorescence intensities on 485 nm (excitation)/535 nm (emission) for calcein-AM and 530 nm (excitation)/620 nm (emission) for PI were measured with a fluorescence plate reader.

Intracellular cAMP was measured by using a cAMP direct immunoassay kit (BioVision, San Francisco, CA) according to the manufacturer’s instructions.

All data are expressed as mean ± SD. For two-group comparisons, a Student t test was used. For multiple comparisons, one-way ANOVA followed by a Dunnett test was used to compare each group versus a control group, and one-way ANOVA followed by a Tukey or Bonferroni test was used to compare each group with every other group. For human macrophage experiments, a paired t test was used. All data were statistically analyzed using GraphPad Prism software version 6.01 (GraphPad Software, La Jolla, CA). Differences with a p value <0.05 were considered significant.

We first determined the role of iATP in IL-1β secretion by primed murine BMDMs. To reduce iATP, cells were cultured in glucose-free medium in the presence of 2DG, an inhibitor of glycolysis. As shown in Fig. 1A, iATP content was reduced up to 4-fold by increasing 2DG concentration up to 10 mM. Paradoxically, glucose-free medium without or with low 2DG concentrations increased iATP compared with normal medium, which could be accounted for by reactive induction of oxidative phosphorylation in mitochondria. 2DG at the highest dose had no significant effect on LDH release (Supplemental Fig. 1A) and a small effect on cell death (Supplemental Fig. 1B) at least for the first 6 h. Interestingly, we observed an inverse correlation between iATP levels and secretion of IL-1β in primed BMDMs, with IL-1β starting to be observed in cells incubated with 0.1 mM 2DG (which caused a drop in iATP content by 10%), then further increasing as iATP levels were further reduced (Fig. 1A, 1B). Conversely, IL-1β secretion was decreased when iATP content was restored by glucose supplementation (Fig. 1C, 1D). IL-1β secretion was also induced by oligomycin treatment, suggesting that iATP loss due to the inhibition of mitochondrial ATP synthesis as well as glycolysis can induce IL-1β secretion (Supplemental Fig. 1C, 1D). Additionally, IL-1α was similarly induced as IL-1β when iATP was lowered (Fig. 1E). In contrast, TNF-α, IL-6, and MCP-1 were significantly inhibited by reduction of iATP (Fig. 1F, Supplemental Fig. 1E, 1F), suggesting that iATP is important for the synthesis of these cytokines/chemokines. Finally, human primary macrophages recapitulated the results obtained with murine BMDMs, as iATP depletion with 2DG also induced IL-1β secretion (Fig. 1G, 1H), suggesting that this phenomenon is highly conserved across different species.

FIGURE 1.

Reduction of iATP leads to secretion of IL-1β in human and murine macrophages. (A) BMDMs were incubated 6 h with 2DG (0.01, 0.1, 1, or 10 mM) in glucose-free medium. RPMI 1640 medium containing 10 mM glucose was used as normal medium. iATP was measured. (B) In the same conditions, IL-1β in the supernatant was analyzed by ELISA. (C) BMDMs were incubated 6 h with glucose (0.001, 0.01, 0.1, 1, or 10 mM) in glucose-free and 10 mM 2DG-containing medium. RPMI 1640 medium containing 10 mM glucose was used as normal medium. iATP was measured. (D) In the same conditions, IL-1β in the supernatant was analyzed by ELISA. (E) BMDMs were incubated 6 h with 2DG (0.01, 0.1, 1, or 10 mM) in glucose-free medium. RPMI 1640 medium containing 10 mM glucose was used as normal medium. IL-1α in the supernatant was analyzed by ELISA. (F) TNF-α in the supernatant was analyzed by ELISA. Results are shown as mean ± SD of triplicate and are representative of three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. (G) Human primary macrophages were incubated 6 h with 2DG (10 mM) in glucose-free medium. RPMI 1640 medium containing 10 mM glucose was used as normal medium. (H) In the same conditions, IL-1β in the supernatant was analyzed by ELISA. Values are expressed as the average of triplicate and results were obtained by using human macrophages from five donors. **p < 0.01, ****p < 0.0001. Med, normal medium (RPMI 1640 medium containing 10 mM glucose).

FIGURE 1.

Reduction of iATP leads to secretion of IL-1β in human and murine macrophages. (A) BMDMs were incubated 6 h with 2DG (0.01, 0.1, 1, or 10 mM) in glucose-free medium. RPMI 1640 medium containing 10 mM glucose was used as normal medium. iATP was measured. (B) In the same conditions, IL-1β in the supernatant was analyzed by ELISA. (C) BMDMs were incubated 6 h with glucose (0.001, 0.01, 0.1, 1, or 10 mM) in glucose-free and 10 mM 2DG-containing medium. RPMI 1640 medium containing 10 mM glucose was used as normal medium. iATP was measured. (D) In the same conditions, IL-1β in the supernatant was analyzed by ELISA. (E) BMDMs were incubated 6 h with 2DG (0.01, 0.1, 1, or 10 mM) in glucose-free medium. RPMI 1640 medium containing 10 mM glucose was used as normal medium. IL-1α in the supernatant was analyzed by ELISA. (F) TNF-α in the supernatant was analyzed by ELISA. Results are shown as mean ± SD of triplicate and are representative of three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. (G) Human primary macrophages were incubated 6 h with 2DG (10 mM) in glucose-free medium. RPMI 1640 medium containing 10 mM glucose was used as normal medium. (H) In the same conditions, IL-1β in the supernatant was analyzed by ELISA. Values are expressed as the average of triplicate and results were obtained by using human macrophages from five donors. **p < 0.01, ****p < 0.0001. Med, normal medium (RPMI 1640 medium containing 10 mM glucose).

Close modal

We next sought to determine which mechanisms are involved in iATP depletion-induced IL-1β secretion. Because several studies have reported that mitochondrial ROS mediates IL-1β secretion by various NLRP3 activators (5, 6), we first examined whether iATP depletion induced mitochondrial ROS production. As shown in Fig. 2A, iATP depletion did not affect mitochondrial ROS generation, whereas MSU, used as a positive control, significantly increased ROS production. These data strongly suggest that iATP depletion-induced IL-1β secretion is independent of mitochondrial ROS. Next, to examine whether IL-1β secretion was mediated by K+ efflux, calcium mobilization, cathepsins, or caspase-1 activation (1214), BMDMs were incubated with high extracellular K+ concentration, EGTA (Ca2+ chelator), E64d (cathepsins and calpain inhibitor), or z-VAD-fmk (caspase inhibitor) under iATP-depleting conditions (glucose-free medium containing 10 mM 2DG), and IL-1β in the supernatant was measured by ELISA. High K+, EGTA, and E64d did not affect iATP contents and IL-1β secretions (Fig. 2B, 2C). Additionally, these treatments had no effects on LDH release and IL-1α, although EGTA increased LDH release (Supplemental Fig. 2A, 2B). In contrast, z-VAD-fmk significantly inhibited iATP depletion-induced IL-1β secretion without affecting iATP levels (Fig. 2B, 2C), which suggests that iATP depletion induced caspase-1 and subsequent IL-1β secretion. Indeed, caspase-1 immunoblot showed that iATP depletion led to self-cleaved caspase-1 secretion (p20 form on Fig. 2D), which could be prevented by z-VAD-fmk (result not shown). Similarly, IL-1β processing was also observed (Fig. 2D). Taken together, these results strongly suggested that low iATP induced IL-1β secretion through caspase-1 activation, independently of mitochondrial ROS, K+ and Ca2+ ionic concentrations, and cathepsins. We next examined the involvement of NLRP3 inflammasome components in iATP depletion–induced IL-1β secretion by using THP-1 cells transfected with shRNA against lamin (mock), NLRP3 (shNALP3), caspase-1 (shCasp1), or ASC (shASC). As we previously observed with murine BMDMs, iATP depletion in primed mock THP-1 cells induced IL-1β secretion (Fig. 2E, 2F). In contrast, IL-1β secretion by shNLRP3, shCasp1, or shASC cells in response to iATP depletion was dramatically impaired (Fig. 2E, 2F). Collectively, these data clearly demonstrated that iATP depletion triggers NLRP3 inflammasome activation, leading to caspase-1 activation and subsequent IL-1β secretion.

FIGURE 2.

Reduction of iATP causes mitochondrial dysfunction, leading to IL-1β secretion through NLRP3 inflammasome activation. (A) BMDMs were incubated in glucose-free medium containing 10 mM 2DG (iATP depletion) for 90 min. RPMI 1640 medium containing 10 mM glucose was used as normal medium, and MSU (0.25 mg/ml) was used as a positive control. After incubation, cells were loaded with DHR123. (A–C) BMDMs were incubated 6 h with vehicle, K+ (30 mM), EGTA (2.5 mM), E64d (50 μM), or z-VAD-fmk (20 μM) in glucose-free medium containing 10 mM 2DG (iATP depletion). RPMI 1640 medium containing 10 mM glucose was used as normal medium. iATP was measured (B) and IL-1β in the supernatant was analyzed by ELISA (C). Equal volume of supernatant was electrophoresed on SDS-PAGE, and then caspase-1 and IL-1β were detected with anti–caspase-1 and anti–IL-1β Abs, respectively (D). (E and F) Mock, shNLRP3, shCasp1, or shASC THP-1 cells were incubated 6 h in the medium containing 10 mM glucose (Medium) or glucose-free medium containing 10 mM 2DG (iATP depletion). iATP was measured (E) and IL-1β in the supernatant was analyzed by ELISA (F). (G) BMDMs were incubated 90 min with z-VAD-fmk (20 μM) in medium containing 10 mM glucose (Med) or glucose-free medium containing 10 mM 2DG (iATP depletion), and then loaded 30 min with TMRE. CCCP (20 μM) in normal medium was used as positive control to deplete Δψm. Data are expressed as the percentage of control. Results are shown as mean ± SD of triplicate and are representative of three (A, B, C, and G) or two (D–F) independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Med, normal medium (RPMI 1640 medium containing 10 mM glucose).

FIGURE 2.

Reduction of iATP causes mitochondrial dysfunction, leading to IL-1β secretion through NLRP3 inflammasome activation. (A) BMDMs were incubated in glucose-free medium containing 10 mM 2DG (iATP depletion) for 90 min. RPMI 1640 medium containing 10 mM glucose was used as normal medium, and MSU (0.25 mg/ml) was used as a positive control. After incubation, cells were loaded with DHR123. (A–C) BMDMs were incubated 6 h with vehicle, K+ (30 mM), EGTA (2.5 mM), E64d (50 μM), or z-VAD-fmk (20 μM) in glucose-free medium containing 10 mM 2DG (iATP depletion). RPMI 1640 medium containing 10 mM glucose was used as normal medium. iATP was measured (B) and IL-1β in the supernatant was analyzed by ELISA (C). Equal volume of supernatant was electrophoresed on SDS-PAGE, and then caspase-1 and IL-1β were detected with anti–caspase-1 and anti–IL-1β Abs, respectively (D). (E and F) Mock, shNLRP3, shCasp1, or shASC THP-1 cells were incubated 6 h in the medium containing 10 mM glucose (Medium) or glucose-free medium containing 10 mM 2DG (iATP depletion). iATP was measured (E) and IL-1β in the supernatant was analyzed by ELISA (F). (G) BMDMs were incubated 90 min with z-VAD-fmk (20 μM) in medium containing 10 mM glucose (Med) or glucose-free medium containing 10 mM 2DG (iATP depletion), and then loaded 30 min with TMRE. CCCP (20 μM) in normal medium was used as positive control to deplete Δψm. Data are expressed as the percentage of control. Results are shown as mean ± SD of triplicate and are representative of three (A, B, C, and G) or two (D–F) independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Med, normal medium (RPMI 1640 medium containing 10 mM glucose).

Close modal

Because cAMP was reported to bind directly to NLRP3 and to inhibit NLRP3 activation (15), we next hypothesized that reduced iATP levels could lead to reduced cAMP levels and de-repression of NLRP3 activity. cAMP levels were not significantly changed by iATP depletion (Supplemental Fig. 2C), so a role of cAMP in IL-1β secretion induced by low iATP is unlikely. Finally, decreased iATP did not raise extracellular ATP levels, but had rather the opposite effect (Supplemental Fig. 2D), thus excluding a role for extracellular ATP signaling via the ATP/P2X7 receptor axis as an explanation of the effects of lowered iATP (16).

As several studies have demonstrated the important role of mitochondria in the regulation of IL-1β secretion (1720), we examined the change in Δψm, an indicator of mitochondrial function. Reduction of iATP levels significantly depolarized Δψm, although not reaching the maximal depolarization obtained with the positive control CCCP, an uncoupler of oxidative phosphorylation (Fig. 2G). Furthermore, pretreatment with z-VAD-fmk did not abrogate this effect (Fig. 2G). Taken together, these results demonstrated that reduced iATP concentration causes mitochondrial dysfunction and depolarization of Δψm, and concomitantly caspase-1 activation and subsequent IL-1β secretion in activated macrophages.

We next assessed whether iATP decrease is a common mechanism shared by different NLRP3 inflammasome activators under conditions where glucose was available as an energy source. We confirmed that nigericin decreased iATP as previously published (6, 7) and also showed that particulate NLRP3 activators such as MSU crystals or basic calcium phosphate crystals (such as OCP and HA) led to reduced iATP levels together with upmodulation of IL-1β secretion in primed BMDMs (Fig. 3A, 3B). As expected, cytochalasin D, an inhibitor of phagocytosis, inhibited MSU-induced iATP loss and IL-1β secretion, but it had little impact on nigericin-induced iATP depletion and IL-1β secretion (Supplemental Fig. 3A, 3B). In a time course analysis with MSU, reduction of iATP started at 15 min after stimulation whereas IL-1β in the supernatant was detected only after 30 min (Fig. 3C, 3D). Additionally, MSU also reduced iATP and induced IL-1β secretion in human primary macrophages (Fig. 3E, 3F). Taken together, these results suggest that iATP decrease is the trigger of IL-1β secretion by NLRP3 activators in murine and human macrophages.

FIGURE 3.

NLRP3 activators decrease iATP in human and murine macrophages. BMDMs were stimulated 2 h with medium, nigericin (2.5 μM), MSU (0.25 mg/ml), OCP (0.1 mg/ml), or HA (0.2 mg/ml). iATP was measured (A) and IL-1β in the supernatant was analyzed by ELISA (B). BMDMs were stimulated with medium or MSU (0.25 mg/ml) for the indicated time. iATP was measured (C) and IL-1β in the supernatant was analyzed by ELISA (D). Data are expressed as the percentage of iATP content at T = 0 min (C). Results are shown as mean ± SD of triplicate and are representative of two independent experiments (A–D). *p < 0.05, **p < 0.01, ****p < 0.0001. Human primary macrophages were stimulated 6 h with medium or MSU (0.25 mg/ml). iATP was measured (E) and IL-1β in the supernatant was analyzed by ELISA (F). Values are expressed as the average of triplicate and results were obtained by using human macrophages from three donors. **p < 0.01. Med, normal medium (RPMI 1640 medium containing 10 mM glucose).

FIGURE 3.

NLRP3 activators decrease iATP in human and murine macrophages. BMDMs were stimulated 2 h with medium, nigericin (2.5 μM), MSU (0.25 mg/ml), OCP (0.1 mg/ml), or HA (0.2 mg/ml). iATP was measured (A) and IL-1β in the supernatant was analyzed by ELISA (B). BMDMs were stimulated with medium or MSU (0.25 mg/ml) for the indicated time. iATP was measured (C) and IL-1β in the supernatant was analyzed by ELISA (D). Data are expressed as the percentage of iATP content at T = 0 min (C). Results are shown as mean ± SD of triplicate and are representative of two independent experiments (A–D). *p < 0.05, **p < 0.01, ****p < 0.0001. Human primary macrophages were stimulated 6 h with medium or MSU (0.25 mg/ml). iATP was measured (E) and IL-1β in the supernatant was analyzed by ELISA (F). Values are expressed as the average of triplicate and results were obtained by using human macrophages from three donors. **p < 0.01. Med, normal medium (RPMI 1640 medium containing 10 mM glucose).

Close modal

Finally, we examined the mechanisms involved in NLRP3 activator–induced iATP decrease. iATP is synthesized through the glycolysis pathway in the cytosol and oxidative phosphorylation in mitochondria. To evaluate the contribution of glycolysis in iATP decrease, we measured extracellular lactate, the end product of glycolysis, which is released into the extracellular space and proportionally correlated with intracellular glycolytic activity (21). As expected, 2DG significantly decreased extracellular lactate level (Fig. 4A). Nigericin decreased iATP levels essentially via inhibition of oxidative phosphorylation rather than inhibition of glycolysis. This is supported by the finding that nigericin treatment led to the collapse in Δψm (Fig. 4F) and no inhibition, but rather a stimulation, of lactate secretion (Fig. 4A). In contrast, MSU decreased lactate level (Fig. 4A) and depolarized Δψm (Fig. 4F), suggesting that iATP decrease by MSU is due to both inhibition of glycolysis and oxidative phosphorylation. As caspase-1 activation is a crucial step for IL-1β secretion, we next determined whether caspase-1 activation mediates iATP loss by NLRP3 activators. Upon nigericin or MSU stimulation, treatment with z-VAD-fmk failed to affect iATP levels (Fig. 4B) whereas it completely inhibited IL-1β secretion (Fig. 4C). These results suggest that iATP loss is an upstream event of caspase-1 activation upon NLRP3 inflammasome activation. We next determined whether Ca2+ mobilization mediates iATP loss by NLRP3 activators. BMDMs were treated with EGTA and then stimulated with nigericin or MSU. EGTA treatment significantly inhibited nigericin- or MSU-induced iATP loss and IL-1β secretion (Fig. 4D and 4E, respectively). Additionally, the collapse in Δψm induced by nigericin or MSU was significantly inhibited by EGTA treatment (Fig. 4F), demonstrating that Ca2+ influx–induced depolarization of Δψm leads to iATP loss. That iATP depletion also induced the depolarization of Δψm as shown in Fig. 2G suggests that the positive feedback loop between iATP loss and mitochondrial dysfunction leads to further induction of IL-1β, and therefore classical NLRP3 activators alter the balance between iATP and the Δψm.

FIGURE 4.

NLRP3 activators decrease iATP through Ca2+-mediated mitochondrial dysfunction. (AC) BMDMs were stimulated 2 h with medium, nigericin (2.5 μM), or MSU (0.25 mg/ml). Lactate level in the supernatant was measured. 2DG (10 mM) in the absence of glucose was used to inhibit glycolysis (A). iATP was measured (B) and IL-1β in the supernatant was analyzed by ELISA (C). (D and E) BMDMs were pretreated 30 min with vehicle or EGTA (2.5 mM) and then stimulated 2 h with medium, nigericin (2.5 μM), or MSU (0.25 mg/ml). iATP was measured (D) and IL-1β in the supernatant was analyzed by ELISA (E). (F) BMDMs were pretreated 30 min with vehicle or EGTA (2.5 mM) and then stimulated 90 min with medium, nigericin (2.5 μM), or MSU (0.25 mg/ml). After stimulation, cells were loaded with TMRE. Data are expressed as the percentage of control (unstimulated cell) (B, D, and F). Results are shown as mean ± SD of triplicate and representative of three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

FIGURE 4.

NLRP3 activators decrease iATP through Ca2+-mediated mitochondrial dysfunction. (AC) BMDMs were stimulated 2 h with medium, nigericin (2.5 μM), or MSU (0.25 mg/ml). Lactate level in the supernatant was measured. 2DG (10 mM) in the absence of glucose was used to inhibit glycolysis (A). iATP was measured (B) and IL-1β in the supernatant was analyzed by ELISA (C). (D and E) BMDMs were pretreated 30 min with vehicle or EGTA (2.5 mM) and then stimulated 2 h with medium, nigericin (2.5 μM), or MSU (0.25 mg/ml). iATP was measured (D) and IL-1β in the supernatant was analyzed by ELISA (E). (F) BMDMs were pretreated 30 min with vehicle or EGTA (2.5 mM) and then stimulated 90 min with medium, nigericin (2.5 μM), or MSU (0.25 mg/ml). After stimulation, cells were loaded with TMRE. Data are expressed as the percentage of control (unstimulated cell) (B, D, and F). Results are shown as mean ± SD of triplicate and representative of three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Close modal

Finally, as reduction of intracellular K+ concentration has been reported to be the common step for caspase-1 activation (7), we examined whether K+ efflux was also involved iATP loss. By blocking cellular K+ efflux, MSU- and nigericin-induced iATP loss and IL-1β secretion were attenuated (Supplemental Fig. 3C, 3D), meaning that K+ efflux upon NLRP3 activators is an upstream event to iATP downmodulation.

Because many different stimuli can activate the NLRP3 inflammasome in activated macrophages, it is likely that NLRP3 does not detect the different stimulants directly but rather detects shared cellular changes such as ion concentration and redox state. In the present study, we found that in nigericin- or crystal-stimulated macrophages, reduction of iATP is such a signal. Previous studies have demonstrated that iATP is necessary for the NLRP3 activation through its binding (22). Consistent with those results, we found that complete depletion of iATP (by 2DG and sodium azide under glucose starvation, blocking glycolysis and oxidative phosphorylation, respectively) led to total absence of IL-1β secretion (results not shown). Thus, reduction of iATP may represent a danger signal for NLRP3 activation, but complete loss of iATP no longer activates the NLRP3 inflammasome because of unavailability of fuels. Therefore, in our experiments reduction of iATP, which is linked to the reduction of the Δψm, might be the metabolic change for the NLRP3 activation. However, in some conditions (30 min of macrophage stimulation with gramicidin), NLRP3 activation occurred with no mitochondrial perturbation and no iATP decrease (7). This latter result suggests that iATP decrease may not represent a unified general mechanism accounting for NLRP3 activation by all stimuli.

K+ efflux has previously been reported to be upstream of calcium mobilization during NLRP3 activation (7, 13). As in our study, high K+ inhibits nigericin- or MSU-induced, but not low iATP-induced, IL-1β secretion, and thus reduction of iATP is logically the downstream event of K+ efflux and Ca2+ mobilization. Based on our findings, we propose the following model for the NLRP3 activation upon nigericin and crystal stimulation (see the summary schema in Fig. 5). In the iATP-dependent pathway, K+ efflux occurs first, followed by calcium mobilization triggering depolarization of Δψm, which causes iATP reduction. Reduction of iATP then induces IL-1β secretion through NLRP3 and caspase-1 activation. Conversely, iATP depletion can induce a collapse in Δψm (Fig. 2G), suggesting that there is a positive feedback loop between iATP reduction and depolarization of Δψm during NLRP3 activation. These findings are consistent with the theory that iATP is available to maintain Δψm and that Δψm is a driving force for ATP synthesis in mitochondria (11). Additionally, ROS-dependent and other K+-dependent pathways, but iATP-independent, are also involved in NLRP3 activation. Contribution of each event may vary depending on NLRP3 activators: MSU likely predominantly activates ROS-dependent and other K+-dependent IL-1β pathways rather than an iATP loss–dependent pathway. This is supported by the findings that high K+ completely inhibits MSU-induced IL-1β secretion but partially MSU-induced iATP loss (Supplemental Fig. 3C, 3D), and that MSU but not iATP depletion increases intracellular ROS (Fig. 2A). Alternatively, nigericin mainly activates an iATP loss–dependent pathway because iATP loss and IL-1β secretion by nigericin were completely inhibited by high K+ (Supplemental Fig. 3C, 3D). However, further studies are needed to understand the detailed mechanisms of NLRP3 activation by iATP reduction.

FIGURE 5.

Schema of NLRP3 activation sensed by decresed iATP. In the iATP-dependent pathway, K+ efflux occurs first, followed by calcium mobilization triggering depolarization of Δψm, which causes iATP reduction via the inhibition of oxidative phosphorylation (OxPhos) and glycolysis. Reduction of iATP then induces IL-1β secretion through NLRP3 and caspase-1 activation. Additionally, ROS-dependent and other K+-dependent IL-1β pathways are also involved in NLRP3 activation. Contribution of each event may be varied depending on NLRP3 activators: MSU likely activates ROS-dependent and other K+-dependent IL-1β pathways rather than an iATP loss–dependent pathway. Alternatively, nigericin mainly activates an iATP loss–dependent pathway. 2DG, a glycolysis inhibitor, can reduce iATP levels directly, without the need of the upstream events, K+ and Ca2+ ionic fluxes. The decrease in iATP concentration is the important step sensed by NLRP3 inflammasome, leading to caspase-1 activation and IL-1β secretion.

FIGURE 5.

Schema of NLRP3 activation sensed by decresed iATP. In the iATP-dependent pathway, K+ efflux occurs first, followed by calcium mobilization triggering depolarization of Δψm, which causes iATP reduction via the inhibition of oxidative phosphorylation (OxPhos) and glycolysis. Reduction of iATP then induces IL-1β secretion through NLRP3 and caspase-1 activation. Additionally, ROS-dependent and other K+-dependent IL-1β pathways are also involved in NLRP3 activation. Contribution of each event may be varied depending on NLRP3 activators: MSU likely activates ROS-dependent and other K+-dependent IL-1β pathways rather than an iATP loss–dependent pathway. Alternatively, nigericin mainly activates an iATP loss–dependent pathway. 2DG, a glycolysis inhibitor, can reduce iATP levels directly, without the need of the upstream events, K+ and Ca2+ ionic fluxes. The decrease in iATP concentration is the important step sensed by NLRP3 inflammasome, leading to caspase-1 activation and IL-1β secretion.

Close modal

iATP levels are decreased during infections and might well be a trigger for an efficient host defense. For example, macrophages react to some Mycobacterium tuberculosis strains by lowering the Δψm and iATP (23) and by inducing inflammasome activation and IL-1β production, ultimately promoting bacterial killing (24). Several intracellular bacteria or viruses reduce iATP levels (25). During hepatitis C virus replication, iATP is 5-fold higher in the places of virus replication than in the cytosol, where it is reduced to by 50% (26). Thus, a decrease in iATP following bacteria or viral infection could be a signal for inflammasome activation, tipping the balance between microorganism survival or in favor of the host cell. Decreased iATP levels are also observed in several inflammatory situations. Among these are ischemia–reperfusion and atherosclerosis (27, 28). Interestingly, in each of these experimental models, a parallel increase in IL-1β release was observed (4, 29). Taken together, a general pattern emerges where low iATP levels represent a danger signal for NLRP3 activation, leading to increased IL-1β release.

Different roles of iATP in regulating cytokines/chemokines were observed in this study. In activated macrophages, secretions of TNF-α, IL-6, and MCP-1 were decreased depending on iATP reduction by glycolysis inhibition, whereas IL-1β and IL-1α secretions were increased (Fig. 1, Supplemental Fig. 1). Conversely, it has been reported that IL-1β transcription induced by LPS, as well as TNF-α and IL-6, depend on glycolysis in macrophages (30, 31). In this study, we analyzed the roles of iATP in IL-1β secretion by using primed BMDMs, which already express IL-1β mRNA and pro–IL-1β protein. Thus, these discrepancies are likely due to differences between activated states of macrophages. Furthermore, we also demonstrated differential regulation of IL-1α and IL-1β secretion by z-VAD-fmk, confirming that NLRP3 is differentially involved in secretion of these two cytokines (32).

In conclusion, in the present study we show that in different specific settings (glycolysis inhibition by 2DG, oxidative phosphorylation inhibition by oligomycin, nigerin and crystal stimulation) decreased iATP is a danger signal for NLRP3 activation, leading to IL-1β secretion.

Additionally, our findings give new insights into mechanisms underlying NLRP3 activation and the regulation of cytokines/chemokines in activated macrophages, and they suggest that maintaining the iATP pool could be a strategic approach for treating NLRP3-related inflammatory diseases.

We thank Nathaliane Bagnoud for excellent technical support and Dr. Annette Ives and Prof. Hans Acha-Orbea for helpful discussions.

This work was supported by Fonds National Suisse de la Recherche Scientifique Grant 310030-130085/1 as well as by the Fondation Jean and Linette Warnery.

The online version of this article contains supplemental material.

Abbreviations used in this article:

BMDM

bone marrow–derived macrophage

CCCP

carbonyl cyanide 3-​chlorophenylhydrazone

2DG

2-deoxyglucose

DHR123

dihydrorhodamine 123

HA

hydroxyapatite

iATP

intracellular ATP

LDH

lactate dehydrogenase

Δψm

mitochondrial membrane potential

MSU

monosodium urate

NLRP3

nucleotide-binding oligomerization domain–like receptor family, pyrin domain–containing 3

OCP

octacalcium phosphate

PI

propidium iodide

ROS

reactive oxygen species

sh

short hairpin

TMRE

tetramethylrhodamine, ethyl ester.

1
McGettrick
A. F.
,
O’Neill
L. A.
.
2013
.
NLRP3 and IL-1β in macrophages as critical regulators of metabolic diseases.
Diabetes Obes. Metab.
15
(
Suppl 3
):
19
25
.
2
Wen
H.
,
Miao
E. A.
,
Ting
J. P.
.
2013
.
Mechanisms of NOD-like receptor-associated inflammasome activation.
Immunity
39
:
432
441
.
3
Martinon
F.
,
Pétrilli
V.
,
Mayor
A.
,
Tardivel
A.
,
Tschopp
J.
.
2006
.
Gout-associated uric acid crystals activate the NALP3 inflammasome.
Nature
440
:
237
241
.
4
Freigang
S.
,
Ampenberger
F.
,
Spohn
G.
,
Heer
S.
,
Shamshiev
A. T.
,
Kisielow
J.
,
Hersberger
M.
,
Yamamoto
M.
,
Bachmann
M. F.
,
Kopf
M.
.
2011
.
Nrf2 is essential for cholesterol crystal-induced inflammasome activation and exacerbation of atherosclerosis.
Eur. J. Immunol.
41
:
2040
2051
.
5
Cruz
C. M.
,
Rinna
A.
,
Forman
H. J.
,
Ventura
A. L.
,
Persechini
P. M.
,
Ojcius
D. M.
.
2007
.
ATP activates a reactive oxygen species-dependent oxidative stress response and secretion of proinflammatory cytokines in macrophages.
J. Biol. Chem.
282
:
2871
2879
.
6
Heid
M. E.
,
Keyel
P. A.
,
Kamga
C.
,
Shiva
S.
,
Watkins
S. C.
,
Salter
R. D.
.
2013
.
Mitochondrial reactive oxygen species induces NLRP3-dependent lysosomal damage and inflammasome activation.
J. Immunol.
191
:
5230
5238
.
7
Muñoz-Planillo
R.
,
Kuffa
P.
,
Martínez-Colón
G.
,
Smith
B. L.
,
Rajendiran
T. M.
,
Núñez
G.
.
2013
.
K+ efflux is the common trigger of NLRP3 inflammasome activation by bacterial toxins and particulate matter.
Immunity
38
:
1142
1153
.
8
McGettrick
A. F.
,
O’Neill
L. A.
.
2013
.
How metabolism generates signals during innate immunity and inflammation.
J. Biol. Chem.
288
:
22893
22898
.
9
Misawa
T.
,
Takahama
M.
,
Kozaki
T.
,
Lee
H.
,
Zou
J.
,
Saitoh
T.
,
Akira
S.
.
2013
.
Microtubule-driven spatial arrangement of mitochondria promotes activation of the NLRP3 inflammasome.
Nat. Immunol.
14
:
454
460
.
10
Krawczyk
C. M.
,
Holowka
T.
,
Sun
J.
,
Blagih
J.
,
Amiel
E.
,
DeBerardinis
R. J.
,
Cross
J. R.
,
Jung
E.
,
Thompson
C. B.
,
Jones
R. G.
,
Pearce
E. J.
.
2010
.
Toll-like receptor-induced changes in glycolytic metabolism regulate dendritic cell activation.
Blood
115
:
4742
4749
.
11
Beltrán
B.
,
Mathur
A.
,
Duchen
M. R.
,
Erusalimsky
J. D.
,
Moncada
S.
.
2000
.
The effect of nitric oxide on cell respiration: a key to understanding its role in cell survival or death.
Proc. Natl. Acad. Sci. USA
97
:
14602
14607
.
12
Pétrilli
V.
,
Papin
S.
,
Dostert
C.
,
Mayor
A.
,
Martinon
F.
,
Tschopp
J.
.
2007
.
Activation of the NALP3 inflammasome is triggered by low intracellular potassium concentration.
Cell Death Differ.
14
:
1583
1589
.
13
Murakami
T.
,
Ockinger
J.
,
Yu
J.
,
Byles
V.
,
McColl
A.
,
Hofer
A. M.
,
Horng
T.
.
2012
.
Critical role for calcium mobilization in activation of the NLRP3 inflammasome.
Proc. Natl. Acad. Sci. USA
109
:
11282
11287
.
14
Hornung
V.
,
Bauernfeind
F.
,
Halle
A.
,
Samstad
E. O.
,
Kono
H.
,
Rock
K. L.
,
Fitzgerald
K. A.
,
Latz
E.
.
2008
.
Silica crystals and aluminum salts activate the NALP3 inflammasome through phagosomal destabilization.
Nat. Immunol.
9
:
847
856
.
15
Lee
G. S.
,
Subramanian
N.
,
Kim
A. I.
,
Aksentijevich
I.
,
Goldbach-Mansky
R.
,
Sacks
D. B.
,
Germain
R. N.
,
Kastner
D. L.
,
Chae
J. J.
.
2012
.
The calcium-sensing receptor regulates the NLRP3 inflammasome through Ca2+ and cAMP.
Nature
492
:
123
127
.
16
Riteau
N.
,
Baron
L.
,
Villeret
B.
,
Guillou
N.
,
Savigny
F.
,
Ryffel
B.
,
Rassendren
F.
,
Le Bert
M.
,
Gombault
A.
,
Couillin
I.
.
2012
.
ATP release and purinergic signaling: a common pathway for particle-mediated inflammasome activation.
Cell Death Dis.
3
:
e403
.
17
Zhou
R.
,
Yazdi
A. S.
,
Menu
P.
,
Tschopp
J.
.
2011
.
A role for mitochondria in NLRP3 inflammasome activation.
Nature
469
:
221
225
.
18
Ichinohe
T.
,
Yamazaki
T.
,
Koshiba
T.
,
Yanagi
Y.
.
2013
.
Mitochondrial protein mitofusin 2 is required for NLRP3 inflammasome activation after RNA virus infection.
Proc. Natl. Acad. Sci. USA
110
:
17963
17968
.
19
Shimada
K.
,
Crother
T. R.
,
Karlin
J.
,
Dagvadorj
J.
,
Chiba
N.
,
Chen
S.
,
Ramanujan
V. K.
,
Wolf
A. J.
,
Vergnes
L.
,
Ojcius
D. M.
, et al
.
2012
.
Oxidized mitochondrial DNA activates the NLRP3 inflammasome during apoptosis.
Immunity
36
:
401
414
.
20
Freigang
S.
,
Ampenberger
F.
,
Weiss
A.
,
Kanneganti
T. D.
,
Iwakura
Y.
,
Hersberger
M.
,
Kopf
M.
.
2013
.
Fatty acid-induced mitochondrial uncoupling elicits inflammasome-independent IL-1α and sterile vascular inflammation in atherosclerosis.
Nat. Immunol.
14
:
1045
1053
.
21
Colussi
C.
,
Albertini
M. C.
,
Coppola
S.
,
Rovidati
S.
,
Galli
F.
,
Ghibelli
L.
.
2000
.
H2O2-induced block of glycolysis as an active ADP-ribosylation reaction protecting cells from apoptosis.
FASEB J.
14
:
2266
2276
.
22
Duncan
J. A.
,
Bergstralh
D. T.
,
Wang
Y.
,
Willingham
S. B.
,
Ye
Z.
,
Zimmermann
A. G.
,
Ting
J. P.
.
2007
.
Cryopyrin/NALP3 binds ATP/dATP, is an ATPase, and requires ATP binding to mediate inflammatory signaling.
Proc. Natl. Acad. Sci. USA
104
:
8041
8046
.
23
Jamwal
S.
,
Midha
M. K.
,
Verma
H. N.
,
Basu
A.
,
Rao
K. V.
,
Manivel
V.
.
2013
.
Characterizing virulence-specific perturbations in the mitochondrial function of macrophages infected with Mycobacterium tuberculosis.
Sci. Rep.
3
:
1328
.
24
Wong
K. W.
,
Jacobs
W. R.
 Jr.
2011
.
Critical role for NLRP3 in necrotic death triggered by Mycobacterium tuberculosis.
Cell. Microbiol.
13
:
1371
1384
.
25
Webster
S. J.
,
Daigneault
M.
,
Bewley
M. A.
,
Preston
J. A.
,
Marriott
H. M.
,
Walmsley
S. R.
,
Read
R. C.
,
Whyte
M. K.
,
Dockrell
D. H.
.
2010
.
Distinct cell death programs in monocytes regulate innate responses following challenge with common causes of invasive bacterial disease.
J. Immunol.
185
:
2968
2979
.
26
Ando
T.
,
Imamura
H.
,
Suzuki
R.
,
Aizaki
H.
,
Watanabe
T.
,
Wakita
T.
,
Suzuki
T.
.
2012
.
Visualization and measurement of ATP levels in living cells replicating hepatitis C virus genome RNA.
PLoS Pathog.
8
:
e1002561
.
27
Murphy
E.
,
Steenbergen
C.
.
2008
.
Mechanisms underlying acute protection from cardiac ischemia-reperfusion injury.
Physiol. Rev.
88
:
581
609
.
28
Leppänen
O.
,
Björnheden
T.
,
Evaldsson
M.
,
Borén
J.
,
Wiklund
O.
,
Levin
M.
.
2006
.
ATP depletion in macrophages in the core of advanced rabbit atherosclerotic plaques in vivo.
Atherosclerosis
188
:
323
330
.
29
Liu
Y.
,
Lian
K.
,
Zhang
L.
,
Wang
R.
,
Yi
F.
,
Gao
C.
,
Xin
C.
,
Zhu
D.
,
Li
Y.
,
Yan
W.
, et al
.
2014
.
TXNIP mediates NLRP3 inflammasome activation in cardiac microvascular endothelial cells as a novel mechanism in myocardial ischemia/reperfusion injury.
Basic Res. Cardiol.
109
:
415
.
30
Tannahill
G. M.
,
Curtis
A. M.
,
Adamik
J.
,
Palsson-McDermott
E. M.
,
McGettrick
A. F.
,
Goel
G.
,
Frezza
C.
,
Bernard
N. J.
,
Kelly
B.
,
Foley
N. H.
, et al
.
2013
.
Succinate is an inflammatory signal that induces IL-1β through HIF-1α.
Nature
496
:
238
242
.
31
Dietl
K.
,
Renner
K.
,
Dettmer
K.
,
Timischl
B.
,
Eberhart
K.
,
Dorn
C.
,
Hellerbrand
C.
,
Kastenberger
M.
,
Kunz-Schughart
L. A.
,
Oefner
P. J.
, et al
.
2010
.
Lactic acid and acidification inhibit TNF secretion and glycolysis of human monocytes.
J. Immunol.
184
:
1200
1209
.
32
Gross
O.
,
Yazdi
A. S.
,
Thomas
C. J.
,
Masin
M.
,
Heinz
L. X.
,
Guarda
G.
,
Quadroni
M.
,
Drexler
S. K.
,
Tschopp
J.
.
2012
.
Inflammasome activators induce interleukin-1α secretion via distinct pathways with differential requirement for the protease function of caspase-1.
Immunity
36
:
388
400
.

The authors have no financial conflicts of interest.

Supplementary data