Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 Aug 22.
Published in final edited form as: Immunity. 2013 Aug 15;39(2):311–323. doi: 10.1016/j.immuni.2013.08.001

Mitochondrial cardiolipin is required for Nlrp3 inflammasome activation

Shankar S Iyer 1,#, Qiong He 1,2,#, John R Janczy 1,2,#, Eric I Elliott 1,3, Zhenyu Zhong 6, Alicia K Olivier 4, Jeffrey J Sadler 1, Vickie Knepper-Adrian 1, Renzhi Han 7, Liang Qiao 6, Stephanie C Eisenbarth 8, William M Nauseef 1,2,5,9, Suzanne L Cassel 1,5,#, Fayyaz S Sutterwala 1,2,3,5,9,#
PMCID: PMC3779285  NIHMSID: NIHMS513034  PMID: 23954133

Summary

Nlrp3 inflammasome activation occurs in response to numerous agonists but the specific mechanism by which this takes place remains unclear. All previously evaluated activators of the Nlrp3 inflammasome induce the generation of mitochondrial reactive oxygen species (ROS), suggesting a model in which ROS is a required upstream mediator of Nlrp3 inflammasome activation. Here we have identified the oxazolidinone antibiotic, linezolid, as a Nlrp3 agonist that activates the Nlrp3 inflammasome independently of ROS. The pathways for ROS-dependent and ROS-independent Nlrp3 activation converged upon mitochondrial dysfunction and specifically the mitochondrial lipid cardiolipin. Cardiolipin bound to Nlrp3 directly and interference with cardiolipin synthesis specifically inhibited Nlrp3 inflammasome activation. Together these data suggest that mitochondria play a critical role in the activation of the Nlrp3 inflammasome through the direct binding of Nlrp3 to cardiolipin.

Introduction

The Nlrp3 inflammasome is a multiprotein complex consisting of the nucleotide-binding domain leucine-rich repeat containing (NLR) family member Nlrp3, the adaptor protein ASC and the cysteine protease caspase-1 (Agostini et al., 2004). The Nlrp3 inflammasome can activate caspase-1 in response to cellular danger resulting in the processing and secretion of the proinflammatory cytokines IL-1β and IL-18 (Kanneganti et al., 2006; Mariathasan et al., 2006; Martinon et al., 2006; Sutterwala et al., 2006). A diverse array of stimuli can activate the Nlrp3 inflammasome including both pathogen-associated molecular patterns (PAMP) and endogenous host-derived molecules indicative of cellular damage. The divergent qualities of the Nlrp3 inflammasome agonists have led to the supposition that the activators converge on a common pathway with a final endogenous ligand activating Nlrp3. An attribute that many activators of the Nlrp3 inflammasome share is their ability to generate reactive oxygen species (ROS). Interruption of ROS production with pharmacological inhibitors blocks activation of the Nlrp3 inflammasome, suggesting that the generation of ROS is a required upstream event for Nlrp3 activation (Cassel et al., 2008; Cruz et al., 2007; Dostert et al., 2008; Petrilli et al., 2007; Zhou et al., 2011). Recent studies have elucidated the cellular source of the responsible ROS to be of mitochondrial origin and independent of the NADPH oxidases (Meissner et al., 2008; Nakahira et al., 2011; van Bruggen et al., 2010; Zhou et al., 2011).

Linezolid belongs to the oxazolidinone class of antibiotics and has been employed in the treatment of infections caused by antibiotic-resistant bacteria, including vancomycin-resistant enterococcus, methicillin-resistant Staphylococcus aureus (Ament et al., 2002) and as part of multi-drug regimens in the treatment of mycobacterial infections (Sood et al., 2006). A number of infections associated with these pathogens, such as endocarditis and osteomyelitis, require antibiotic therapy for up to 8 weeks or more. This prolonged duration of therapy can be problematic, as linezolid has been associated with myelosuppression at rates as high as 32% of patients receiving this antibiotic (Attassi et al., 2002; Dawson et al., 2005). Although the mechanism underlying linezolid-induced myelosuppression remains undefined, mitochondrial toxicity has been implicated in this process (De Vriese et al., 2006; McKee et al., 2006; Soriano et al., 2005).

In this study we demonstrate that linezolid was capable of activating the Nlrp3 inflammasome. Activation of the Nlrp3 inflammasome by linezolid resulted in an in vivo inflammatory response with associated suppression of bone marrow erythroid precursors, consistent with the hematologic anomalies seen in patients that have been ascribed to direct effects of linezolid on mitochondria. In contrast to other activators of the Nlrp3 inflammasome, linezolid-induced activation occurred in a ROS-independent manner. However, activation of the Nlrp3 inflammasome by linezolid and all other Nlrp3 agonists tested was inhibited by mitochondrial stabilization with cyclosporine A. We further confirm the central role of the mitochondria in Nlrp3 activation by demonstrating that Nlrp3 binds to a mitochondrial phospholipid in a stimulus dependent manner, which we have identified as the mitochondrial specific phospholipid cardiolipin. In addition, the disruption of the cardiolipin synthesis pathway results in impaired Nlrp3 inflammasome activation and the addition of cardiolipin, but not other phospholipids, to a broken cell system triggers caspase-1 activation. The novel finding of an ROS-independent pathway to Nlrp3 inflammasome activation, the requirement for cardiolipin in this process and its direct binding to Nlrp3 sheds light on the mechanism by which the Nlrp3 inflammasome is activated and suggests that mitochondrial disruption and the resultant binding of cardiolipin to Nlrp3, and not ROS per se, is ultimately responsible for Nlrp3 inflammasome activation.

Results

Linezolid induces the secretion of IL-1β but not TNFα or IL-6 in vitro

The mechanism by which linezolid induces myelosuppression has not been elucidated. As inflammatory mediators have both positive and negative effects on myeloid and erythroid precursors, we asked if linezolid was capable of inducing an inflammatory response in vitro. As blood concentrations of linezolid in humans after oral administration of 600 mg of linezolid reach peak concentrations of approximately 16.3 to 21 μg/ml (MacGowan, 2003), we used doses of 1, 10, 100, and 200 μg/ml for our in vitro stimulation. In contrast to LPS, linezolid was a poor inducer of TNFα and IL-6 production by the murine macrophage J774A.1 cell line (Figure 1A). However, treatment of LPS-primed macrophages with linezolid at the clinically relevant dose of 10 μg/ml as well as the slightly higher dose of 100 μg/ml resulted in the secretion of IL-1β in a dose-dependent manner (Figure 1B). Linezolid alone did not induce IL-1β secretion (Figure S1A), consistent with the actions of other stimuli of the inflammasome such as silica and ATP that require prior macrophage priming in order to activate caspase-1 (Cassel et al., 2008; Mariathasan et al., 2006; Sutterwala et al., 2006). In addition, linezolid-induced secretion of IL-1β occurred rapidly with maximal secretion by 6 hrs post challenge (Figure S1B). Recent studies have demonstrated that necrotic cells are capable of inducing macrophage inflammasome activation (Ghiringhelli et al., 2009; Iyer et al., 2009). To ensure that the IL-1β secretion observed was not secondary to linezolid-induced cytotoxicity, we assessed the degree of cell death in macrophages following linezolid treatment by measuring lactate dehydrogenase (LDH) release as a marker of cellular damage. LDH release at 6 hrs after linezolid treatment suggested less than 10% cell death, by which time maximal IL-1β release had already been achieved (Figure 1C). This low induction of cell death with Nlrp3 inflammasome activation was confirmed by propidium iodide staining (Figure S1C). Linezolid was also capable of inducing IL-1β secretion from human peripheral blood mononuclear cells (PBMC) and the human monocytic THP-1 cell line via a mechanism that was inhibited by the caspase-1 inhibitor z-YVAD-fmk (Figure 1D and S1D). Collectively, these results demonstrate that linezolid was capable of inducing the caspase-1-dependent secretion of IL-1β from both human PBMC and mouse macrophages.

Figure 1. Linezolid induces caspase-1-dependent IL-1β secretion.

Figure 1

Open in a new tab

(A) J774A.1 mouse macrophages were stimulated with either linezolid or LPS; culture supernatants were collected 18 hrs later and TNFα and IL-6 release measured by ELISA. (B) LPS-primed J774A.1 macrophages were stimulated with either silica or linezolid; culture supernatants were collected 18 hrs later and IL-1β release measured by ELISA. (C) LPS-primed J774A.1 macrophages were stimulated with linezolid or silica; culture supernatants were collected at the indicated times and cytotoxicity measured by LDH release and expressed as a percentage of LDH release by Triton X-100 detergent. (D) LPS-primed human PBMC were incubated with 20 μM z-YVAD-fmk for 1 hr prior to the addition of silica or linezolid. 18 hrs after stimulation culture supernatants were collected and IL-1β release measured by ELISA. (E) LPS-primed J774A.1 macrophages were challenged with silica or 1, 10 or 100 μg/ml of the indicated antibiotic for 18 hrs; IL-1β release into the culture supernatant was measured by ELISA. (A-E) Determinations were performed in triplicate and expressed as the mean ± SD. Results are representative of two separate experiments. See also Figure S1.

Linezolid inhibits bacterial protein synthesis by binding to the bacterial 23S ribosomal RNA of the 50S subunit and preventing a functional 70S initiation complex from forming, which is essential for the bacterial translation process. To determine if other antibiotics were capable of inducing IL-1β secretion, we challenged J774.A1 macrophages with selected antibiotics from multiple different classes of antimicrobials, including those that target protein synthesis as their mechanism of action. Only gramicidin induced substantial IL-1β secretion from macrophages (Figure 1E and S1E), partially consistent with previous studies (Allam et al., 2011; Walev et al., 1995). Similar to linezolid the antibiotic chloramphenicol targets mitochondrial ribosomes but was incapable of activating the Nlrp3 inflammasome (Figure S1F). Based on these data, the capacity of linezolid to trigger IL-1β secretion was not shared by antibiotics with similar mechanisms of antibacterial action, but rather a property unique to linezolid.

Linezolid induces Nlrp3-dependent inflammatory responses in vitro and in vivo

To determine if the Nlrp3 inflammasome was involved in linezolid-induced IL-1β secretion, we challenged LPS-primed bone marrow-derived macrophages (BMM) from wild-type (WT), Nlrp3-, ASC-, caspase-1-, and Nlrc4-deficient mice with linezolid in vitro. Nlrp3-, ASC-, and caspase-1-deficient BMM displayed a marked defect in their ability to process and secrete IL-1β in response to linezolid compared to WT BMM (Figure 2A). In contrast, BMM deficient in Nlrc4, which is required for caspase-1 activation in response to cytosolic flagellin and PrgJ, a component of the bacterial type III secretion system (Sutterwala and Flavell, 2009), secreted IL-1β in response to linezolid (Figure 2A).

Figure 2. Linezolid induces Nlrp3-dependent inflammation in vitro and in vivo.

Figure 2

Open in a new tab

(A) LPS-primed BMM from WT, Nlrp3-, ASC-, caspase-1-, or Nlrc4-deficient mice were stimulated with either linezolid or silica. Culture supernatants were collected at 18 hrs after stimulation and IL-1β release measured by ELISA. (B) Lysates from LPS-primed WT, Nlrp3-, ASC-, or Nlrc4-deficient BMM stimulated with linezolid for 18 hrs were immunoblotted with antibodies against the p10 subunit of caspase-1. Determinations were performed in triplicate and expressed as the mean ± SD. Results are representative of four (A) and three (B) separate experiments. (C) WT, Nlrp3-, ASC- and caspase-1-deficient mice (n=5) were challenged with 200 mg/kg of linezolid intraperitoneally; 16 hrs later neutrophil influx into the peritoneum was determined. Control WT mice (n=3) were challenged intraperitoneally with D5W. Results are representative of at least three independent experiments. * p < 0.01 by two-tailed Mann-Whitney test. (D, E) WT and Nlrp3-deficient mice received daily intraperitoneal injections of linezolid (200 mg/kg) (WT, n=9; Nlrp3−/−, n=9) or an equal volume of control D5W (WT, n=10; Nlrp3−/−, n=6). 12 days later femurs were sectioned and stained with H&E. (D) Representative 600X images are shown. Open arrows indicate myeloid cells and black arrows represent erythroid cells. (E) The mature stages of myeloid and erythroid cells were counted in each section with two sections counted per femur. Results are pooled from two independent experiments and expressed as the mean ± SEM. * p < 0.01 by two-tailed Mann-Whitney test.

Caspase-1 activation involves autocatalytic processing of the 45 kDa pro-caspase-1 to generate two subunits, p20 and p10. Caspase-1 activation in LPS-primed WT BMM stimulated with linezolid was detected in immunoblots by the appearance of the p10 cleavage product (Figure 2B). Caspase-1 activation in response to linezolid in Nlrp3- or ASC-deficient LPS-primed BMM was not observed (Figure 2B). Taken together these findings suggest that linezolid was capable of inducing caspase-1-mediated IL-1β secretion in a manner dependent on the Nlrp3 inflammasome.

To examine if the Nlrp3 inflammasome was involved in the inflammatory response to linezolid in vivo we injected linezolid intraperitoneally into WT, Nlrp3-, ASC- or caspase-1-deficient mice (Figure 2C). Sixteen hours after challenge with linezolid, influx of neutrophils into the peritoneal cavity of WT mice was markedly greater than mice that received the control buffer D5W (5% dextrose in water). This neutrophil influx was significantly diminished in mice deficient in components of the Nlrp3 inflammasome, suggesting that the in vivo inflammatory response to linezolid was also dependent upon the Nlrp3 inflammasome (Figure 2C).

Linezolid therapy is associated with myelosuppression (Attassi et al., 2002; Dawson et al., 2005; Gerson et al., 2002; Taketani et al., 2009), and in particular with an increased myeloid/erythroid cell ratio (Dawson et al., 2005; Green et al., 2001; Taketani et al., 2009). We therefore examined myeloid and erythroid precursors in the bone marrow of mice treated with 200 mg/kg/day of linezolid for 12 days. Linezolid-treated WT mice had significantly decreased mature erythroid cells and increased myeloid cells compared to control animals (Figure 2D and E). In contrast, there was no difference observed in either erythroid or myeloid cells in linezolid-treated Nlrp3−/− mice compared to control animals (Figure 2D and E). These data demonstrate that linezolid induced inflammatory responses in vivo and its effect on the bone marrow depended on the Nlrp3 inflammasome. Hence the activation of the Nlrp3 inflammasome by linezolid may contribute to the in vivo toxicity associated with linezolid therapy.

Linezolid-mediated activation of the Nlrp3 inflammasome in vitro requires a potassium efflux

To gain mechanistic insights into how linezolid activated the Nlrp3 inflammasome, we examined pathways required for Nlrp3 inflammasome activation in response to other agonists. Particulate agonists such as monosodium urate (MSU), silica and alum require internalization by the macrophage in order to activate the Nlrp3 inflammasome (Cassel et al., 2008; Eisenbarth et al., 2008; Martinon et al., 2006). Given linezolid's lipophilic nature that allows it to cross the cell membrane, we predicted its activation of the Nlrp3 inflammasome would be independent of phagocytosis. Treatment of LPS-primed J774A.1 macrophages with cytochalasin D, which blocks actin polymerization, inhibited silica-induced, but not ATP-mediated, IL-1β secretion as expected (Cassel et al., 2008). Cytochalasin D did not inhibit linezolid from inducing macrophage IL-1β secretion (Figure 3A). These data demonstrate that, as predicted by its molecular structure, active phagocytosis of linezolid is not required for Nlrp3 inflammasome activation.

Figure 3. Linezolid-induced IL-1β secretion requires a potassium efflux.

Figure 3

Open in a new tab

(A) LPS-primed J774A.1 macrophages were incubated with 20 μM cytochalasin D for 30 min prior to the addition of linezolid, silica or ATP. 18 hrs later culture supernatants were collected and IL-1β release measured by ELISA. (B) LPS-primed WT and P2rx7−/− BMM were stimulated with linezolid, silica or ATP. 18 hrs after stimulation culture supernatants were collected and IL-1β release measured by ELISA. (C) LPS-primed J774A.1 macrophages were incubated in either high Na+ or high K+ containing media and then stimulated with either linezolid, silica or F. tularensis LVS (MOI 50:1). 18 hrs later culture supernatants were collected and IL-1β release measured by ELISA. (A-C) Determinations were performed in triplicate and expressed as the mean ± SD. Results are representative of three (A) and (B-C) two separate experiments.

The cell surface receptor P2RX7 has been implicated in Nlrp3 inflammasome activation in response to extracellular ATP (Solle et al., 2001). Linezolid-mediated IL-1β secretion remained intact in the absence of the P2RX7 (Figure 3B), suggesting that the P2RX7 was not required for linezolid-induced inflammasome activation. Additionally, this finding also suggests that linezolid-induced Nlrp3 inflammasome activation was not secondary to ATP release from damaged cells, as has been observed for the chemotherapeutic agent doxorubicin (Ghiringhelli et al., 2009).

A common step required by all known Nlrp3 agonists to induce caspase-1 activation is the efflux of cellular potassium (Petrilli et al., 2007). Preventing the potassium efflux by increasing extracellular potassium concentrations inhibits caspase-1 activation in response to stimuli that activate Nlrp3. Consistent with this; increased extracellular potassium significantly inhibited linezolid-induced IL-1β secretion from LPS-primed J774A.1 macrophages (Figure 3C). As expected, increased extracellular potassium also inhibited Nlrp3-dependent silica-induced, but not AIM2-dependent Francisella tularensis LVS-induced, IL-1β production (Figure 3C). Hence linezolid, like all other known Nlrp3 stimuli, required a cellular potassium efflux to induce macrophage IL-1β secretion.

Linezolid-induced Nlrp3 inflammasome activation is independent of ROS but inhibited by mitochondrial stabilization with cyclosporine A

ROS generation has been suggested to be an absolute requirement for activation of the Nlrp3 inflammasome (Dostert et al., 2008; Martinon, 2010; Tschopp and Schroder, 2010), and two recent studies have shown that this ROS is of mitochondrial origin (Nakahira et al., 2011; Zhou et al., 2011). To examine the role of ROS in linezolid-induced Nlrp3 inflammasome activation we utilized pharmacologic inhibitors of ROS. Macrophages were pretreated with the antioxidant N-acetylcysteine (NAC), the flavoprotein inhibitor diphenylene iodonium (DPI), or the metabotropic glutamate receptor 3 agonist (2R,4R)-4-aminopyrrolidine-2,4-dicarboxylate (APDC) prior to challenge with either linezolid, silica, nigericin, ATP, or alum. A recent study suggests that treatment with ROS inhibitors blocks priming of the Nlrp3 inflammasome (Bauernfeind et al., 2011); in order to specifically examine the effects of ROS on activation rather than priming of the Nlrp3 inflammasome we treated cells with DPI, NAC or APDC after LPS priming but prior to the addition of Nlrp3 agonists. As previously observed (Cassel et al., 2008; Cruz et al., 2007; Dostert et al., 2008; Petrilli et al., 2007; Zhou et al., 2011), the ROS inhibitors NAC, DPI and APDC blocked silica, nigericin, ATP, and alum-induced IL-1β secretion and caspase-1 activation when given after LPS priming (Figure 4A and B). Unexpectedly, NAC, DPI and APDC all failed to inhibit linezolid-induced IL-1β secretion (Figure 4A). Caspase-1 activation induced by linezolid was also not inhibited by DPI treatment (Figure 4B). To more closely examine the effect of inhibition of mitochondria-derived ROS, we treated macrophages with the mitochondria-targeted antioxidant Mito-TEMPO prior to exposure to agonists. As shown previously, Mito-TEMPO inhibited ATP-mediated IL-1β secretion and caspase-1 activation (Figure 4C and D) (Nakahira et al., 2011). Consistent with the data in Figure 4A, Mito-TEMPO failed to suppress linezolid-induced IL-1β secretion and caspase-1 activation (Figure 4C and D). Similarly, inhibition of the mitochondrial respiratory chain with antimycin A or rotenone blocked IL-1β induced by the ROS-dependent activators ATP and silica but had no effect on release of IL-1β induced by linezolid (Figure 4E). These data suggest that the generation of mitochondrial ROS is not required for linezolid to activate the Nlrp3 inflammasome. We next asked whether linezolid was capable of inducing mitochondrial ROS generation. As expected, the treatment of macrophages with ATP resulted in ROS generation demonstrated by a shift in MitoSOX fluorescence; however, treatment with linezolid had minimal effect on mitochondrial ROS generation in comparison to untreated macrophages (Figure 4F). Taken together, these data implicate a mechanism for linezolid-induced Nlrp3 inflammasome activation that is independent of ROS.

Figure 4. Linezolid-induced Nlrp3 inflammasome activation is independent of ROS generation.

Figure 4

Open in a new tab

(A) LPS-primed J774A.1 macrophages were pretreated with the ROS inhibitors DPI (20 μM), APDC (20 μM) or NAC (1 mM) for 30 min prior to stimulation with linezolid, silica, nigericin, ATP or alum. 18 hrs later supernatants were collected and IL-1β secretion measured by ELISA. (B) Lysates from LPS-primed J774A.1 macrophages challenged with linezolid or silica in the absence or presence of DPI were immunoblotted with antibodies against the p10 subunit of caspase-1. (C) LPS-primed J774A.1 macrophages were pretreated with the antioxidant Mito-TEMPO (500 μM) for 1 hr prior to stimulation with linezolid or ATP. 18 hrs later supernatants were collected and IL-1β secretion measured by ELISA. * p = 0.003 by two-tailed unpaired Student's t-test. (D) Lysates from LPS-primed J774A.1 macrophages challenged with linezolid or ATP in the absence or presence of Mito-TEMPO were immunoblotted with antibodies against the p10 subunit of caspase-1. (E) LPS primed J774A.1 macrophages were treated with antimycin A (5 μg/ml) or rotenone (4 μM) for 30 min prior to stimulation with ATP, linezolid or silica. Supernatants were collected after 12 hrs and IL-1β secretion measured by ELISA. (F) LPS-primed J774A.1 macrophages were treated with ATP (5 mM), Linezolid (25 μg/ml) or media for one hour and then stained with MitoSOX red (2.5 μM) for the final 15 min of the stimulation. (G) LPS-primed J774A.1 cells were pretreated with 20 μM CsA for 1 hr and then challenged with linezolid, silica or nigericin for 6 hrs. Supernatants were collected and cytokine secretions analyzed by ELISA. * p = 0.037, ** p = 0.003 by two-tailed unpaired Student's t-test. Determinations were performed in triplicate and expressed as the mean ± SD. Results are representative of four (A (left panel), C), three (G) and two (A (right panel), B, D-F) separate experiments. See also Figure S2.

Linezolid has been implicated in the inhibition of mitochondrial protein synthesis; both the amount and activity of mitochondrial respiratory-chain complexes are decreased in animals and patients receiving linezolid, thereby supporting the suspicion that linezolid has a deleterious effect on mitochondria (De Vriese et al., 2006; McKee et al., 2006; Soriano et al., 2005). Consistent with other Nlrp3 agonists that have been demonstrated to induce mitochondrial dysfunction (Misawa et al., 2013; Zhou et al., 2011) we found that linezolid also induced mitochondrial dysfunction as measured by diminished cellular ATP and NAD+ production (Figure S2A and B). We also observed diminished mitochondrial cytochrome c oxidase (complex IV) activity in macrophages challenged with linezolid (Figure S2C). To determine the effect of mitochondrial dysfunction on Nlrp3 inflammasome activation we pretreated J774A.1 macrophages with cyclosporine A, an inhibitor of mitochondrial membrane permeability transition (MPT), before stimulation with ROS-dependent and ROS-independent Nlrp3 activators. Inhibition of MPT prevented the release of IL-1β, but not TNFα, in response to both classes of activators (Figure 4G), suggesting that whereas ROS was not a prerequisite for activation of Nlrp3 inflammasome by all agonists, the generation of MPT was.

Mitochondrial cardiolipin binds to Nlrp3

Previous studies have shown Nlrp3 associates with the mitochondria upon activation (Misawa et al., 2013; Zhou et al., 2011). To determine what mitochondrial component binds to Nlrp3 we radiolabeled cellular phospholipids using 14C-linoleic acid and then treated macrophages with Nlrp3 agonists. Cells were fractionated into cytosolic and mitochondrial fractions and Nlrp3 was immunoprecipitated from the mitochondrial fraction. The immunoprecipitated fraction was then subjected to organic extraction and the amount of radioisotope within the organic phase, representing lipids pulled down with the Nlrp3 immunoprecipitation, was compared between control and stimulated groups. Increased radioisotope was found in the organic phase of cells in which the Nlrp3 inflammasome had been activated with either ATP or silica, suggesting increased mitochondrial lipid association with Nlrp3 (Figure 5A). These data suggest Nlrp3 associates with a mitochondrial phospholipid upon macrophage stimulation with an Nlrp3 agonist. To identify the phospholipid that was interacting with Nlrp3 we subjected the immunoprecipitates produced in Figure 5A to thin layer chromatography. A marked increase in the mitochondrial specific lipid cardiolipin was observed in Nlrp3 immunoprecipitates from macrophages that had been stimulated with either ATP or silica (Figure 5B).

Figure 5. Cardiolipin binds to Nlrp3.

Figure 5

Open in a new tab

(A, B) J774A.1 cells were radiolabeled for 24 hrs with 14C linoleic acid (10 μCi); cells were then LPS-primed and stimulated for 12 hrs with ATP or silica. The mitochondrial fraction was isolated, solubilized in 1% Triton X-100 containing lysis buffer and subjected to immunoprecipitation with anti-Nlrp3 antibody or isotype control antibody. Lipids associated with the immunoprecipitates were extracted with chloroform: methanol (2:1 v/v). Radioactivity associated with the organic fraction following immunoprecipitation was quantified (A) and subjected to 1-D HPTLC; the location of migration of cardiolipin (CL) was determined by running a cardiolipin standard in parallel (B). (A) Results are expressed as the mean ± SEM of three independent experiments; ** p < 0.01, *** p < 0.001 by two-tailed unpaired Student's t-test. (C, D) LPS-primed BMM were treated with staurosporine, silica or nigericin for 8 hrs. Cell lysates and supernatants were collected and lysates assessed for caspase-1 and caspase-3 via immunoblot (C) and supernatants for IL-1β via ELISA (D); determinations were performed in triplicate and expressed as the mean ± SD. (E) Lipid strips were incubated with lysates of RAW264.7 cells expressing HA-tagged mouse Nlrp3 or control buffer; strips were then washed and developed with a mouse anti-HA antibody followed by anti-mouse IgG-HRP. (F) Lipid strips were incubated with 10 μg/ml of His-tagged human Nlrp3, Nlrc4 or TLR4; strips were then washed and developed with a mouse anti-His antibody followed by anti-mouse IgG-HRP. (G) The indicated phospholipid-coated beads were incubated with 10 μg/ml of His-tagged Nlrp3; following washes, bound and unbound His tagged-Nlrp3 was determined by immunoblotting with an anti-Nlrp3 antibody. (H) Lipid strips were incubated with lysates of HEK293T cells expressing Flag-tagged full length Nlrp3.1-1036, or truncation mutants Nlrp3ΔLRR.1-719, PYD.1-95, NACHT.69-546 or LRR.535-719; strips were then washed and developed with a mouse anti-FLAG antibody followed by anti-mouse IgG-HRP. Results are representative of two (C, E, F, H), three (B, G) or four (D) separate experiments. See also Figure S3.

Cardiolipin is a non-bilayer forming phospholipid that in eukaryotic cells is exclusively found in the inner mitochondrial membrane. Its structure is unique as it consists of two diacylated phosphatidyl groups joined by a glycerol bridge (Osman et al., 2011). Cardiolipin has been shown to play a critical role in the activation of caspase-8 and the final downstream activation of caspase-3 in the extrinsic pathway of apoptosis. Additionally, a recent study suggested overlap between the activation of Nlrp3 and the induction of apoptotic cell death (Shimada et al., 2012). To evaluate whether these pathways were related we stimulated macrophages with Nlrp3 inflammasome agonists or staurosporine, an inducer of the extrinsic apoptotic pathway. We found that activation of the Nlrp3 inflammasome by silica, but not nigericin, in LPS-primed macrophages resulted in minimal activation of caspase-3 (Figure 5C). Apoptosis induced by staurosporine, shown by the cleavage of caspase-3, was not sufficient for activation of the Nlrp3 inflammasome in unprimed macrophages as shown by the absence of cleaved caspase-1 and lack of IL-1β secretion (Figure 5C and D). However, if macrophages were first primed with LPS, there was modest release of IL-1β and activation of caspase-1 induced by staurosporine (Figure 5C and D). These data suggest that the association of cardiolipin with Nlrp3 was not due to coincident activation of apoptotic pathways.

To confirm that Nlrp3 directly interacts with the mitochondrial lipid cardiolipin, we examined the interaction of Nlrp3 with phospholipids using a protein-lipid overlay assay. HA-tagged mouse Nlrp3 showed binding to cardiolipin and some weaker interaction with phosphatidic acid and 3-sulfogalactosylceramide (Figure 5E). In confirmation of the specificity of binding to Nlrp3, we found that purified His-tagged human Nlrp3 bound to cardiolipin in a protein-lipid overlay assay (Figure 5F), whereas neither His-tagged human TLR4 nor His-tagged human Nlrc4 bound to cardiolipin (Figure 5F). A recent study demonstrated that the phospholipid ceramide was capable of activating the Nlrp3 inflammasome (Vandanmagsar et al., 2011); to assess if ceramide and the related molecule sphingomyelin were capable of interacting with Nlrp3, we employed the use of lipid-coated agarose beads. Consistent with the results of the protein-lipid overlay assay, we observed binding of His-tagged Nlrp3 to cardiolipin-coated beads (Figure 5G). In contrast, Nlrp3 did not show any increase in binding to ceramide- or sphingomyelin-coated beads compared to control agarose beads (Figure 5G).

To characterize the Nlrp3 domain responsible for the interaction with cardiolipin we examined the binding of Flag-tagged Nlrp3 domain constructs (Mayor et al., 2007) to cardiolipin using the protein-lipid overlay assay. In addition to a full-length Nlrp3 (Nlrp3.1-1036), these constructs consisted of Nlrp3 with a truncated leucine-rich repeat (LRR) domain lacking the carboxyl aspect (Nlrp3ΔLRR.1-719), the Pyrin domain alone (PYD.1-95), the NACHT domain alone (NACHT.69-546) or a portion of the LRR domain (LRR.535-719). Cardiolipin interacted with both the full-length Nlrp3 construct (Nlrp3.1-1036), the LRR domain of Nlrp3 (LRR.535-719), and with Flag-tagged Nlrp3 with the truncated LRR domain construct (Nlrp3ΔLRR.1-719), but not with the NACHT (NACHT.69-546) or Pyrin domains (PYD.1-95) (Figure 5H). These data suggest that the LRR region shared by these three constructs was required for their interaction with cardiolipin (Figure 5H). Consistent with these results, we found that Flag-tagged full-length Nlrp3 (Nlrp3.1-1036), LRR domain (LRR.535-719) and Nlrp3 with a truncated LRR domain (Nlrp3ΔLRR.1-719) constructs, but not the NACHT (NACHT.69-546) or Pyrin domain (PYD.1-95), bound to cardiolipin-coated beads (Figure S3). Together these data demonstrate that through its LRR domain Nlrp3 can interact directly with the mitochondrial membrane lipid cardiolipin.

Cardiolipin is sufficient and required for inflammasome activation

Cardiolipin could be playing one of two roles in Nlrp3 inflammasome activation. It could be acting as a docking site to co-localize Nlrp3 to its activating ligand, and it could also be providing the activating signal itself. To assess whether the interaction between Nlrp3 and cardiolipin could result in inflammasome activation we utilized a broken cell system. Cells were primed with LPS then disrupted using nitrogen cavitation. Non-cardiolipin based liposomes had no effect on caspase-1 activation, but the addition of cardiolipin carrying liposomes resulted in the dose dependent activation of caspase-1 (Figure 6A).

Figure 6. Cardiolipin is required and sufficient for Nlrp3 inflammasome activation.

Figure 6

Open in a new tab

(A) LPS-primed J774A.1 cells were lysed by nitrogen cavitation in the presence of liposomes containing phosphatidylcholine (300 μM), phosphatidylserine (300 μM), phosphatidic acid (300 μM), or cardiolipin (300, 100, or 30 μM left panel; 300 μM right panel), incubated for one hour and analyzed for caspase-1 and GADPH by immunoblot. (B, C) J774A.1 macrophages grown in the presence or absence of 0.5 mM palmitate were LPS-primed followed by stimulation with silica, ATP or P. aeruginosa PAK strain (MOI 1:1). 12 hrs after stimulation culture supernatants were collected and IL-1β and TNFα release measured by ELISA. (D) J774A.1 macrophages grown in the presence or absence of 0.5 mM palmitate were LPS-primed followed by nitrogen cavitation lysis in the presence or absence of 300 μM of cardiolipin liposomes. Lysates were incubated for one hour and analyzed by immunoblot. (E) J774A.1 cells grown in the presence or absence of 0.5 mM palmitate were LPS-primed followed by stimulation with silica or linezolid. 12 hrs post stimulation cells were fractionated into mitochondrial and cytosolic fractions and subjected to SDS-PAGE followed by immunoblotting with antibodies to Nlrp3, COX IV (mitochondrial marker) and GAPDH (cytosolic marker). (F) THP-1 cells treated with the indicated siRNA were challenged with linezolid, silica, ATP, F. tularensis LVS (MOI 50:1) for 18 hrs or P. aeruginosa PAK strain (MOI 10:1) for 6 hrs. Culture supernatants were collected and IL-1β and TNFα release measured by ELISA (E). Lysates and supernatants were immunoblotted with antibodies against pro-caspase-1 and the p20 subunit of caspase-1 (F). Determinations were performed in triplicate and expressed as the mean ± SD. Results are representative of two (A-C, E, G) and three (D, F) separate experiments. * p < 0.05, ** p < 0.01, *** p < 0.005 by one-way ANOVA with Bonferroni's multiple comparison test. See also Figure S4.

To determine if the interaction between cardiolipin and Nlrp3 had biological relevance to inflammasome activation, we interfered with cellular cardiolipin synthesis using two independent methods. Treatment of cells with the saturated long chain fatty acid palmitate (C16:0) has been shown to diminish cardiolipin synthesis in cardiomyocytes (Ostrander et al., 2001). Using thin layer chromatography, we confirmed that J774A.1 macrophages grown in serum-free media in the presence of palmitate generated less cardiolipin than did control cells (Figure S4A and B). Consistent with a previous study (Wen et al., 2011), we observed that acutely challenging LPS-primed macrophages with palmitate resulted in the secretion of IL-1β (data not shown). However, LPS-primed macrophages grown in serum-free media in the presence of palmitate had markedly diminished IL-1β secretion in response to the Nlrp3 agonists silica and ATP (Figure 6B). In contrast, palmitate treatment did not compromise the production of TNFα in response to LPS-priming (Figure 6B), the secretion of IL-1β in response to Pseudomonas aeruginosa, an activator of the Nlrc4 inflammasome (Figure 6C), nor the ability of the mitochondria to generate ROS (Figure S4C). To determine if this inhibition was specific to palmitate, we cultured macrophages in serum-free media in the presence of oleate (C18:1), which does not interfere with cardiolipin synthesis (Ostrander et al., 2001). Oleate-treated macrophages retained their ability to secrete IL-1β in response to silica, ATP and linezolid (Figure S4D). Recent studies have implicated calcium flux and cAMP as critical regulators of Nlrp3 inflammasome activation (Lee et al., 2012; Murakami et al., 2012; Rossol et al., 2012; Zhong et al., 2013). Consistent with previous findings, the adenylate cyclase inhibitor KH7 induced the secretion of IL-1β from LPS-primed J774A.1 macrophages (Figure S4E) (Lee et al., 2012). However, palmitate treatment markedly diminished IL-1β secretion in response to KH7 suggesting that reduction of intracellular cAMP precedes the interaction of Nlrp3 with cardiolipin. The generation of cAMP and calcium flux in response to the Nlrp3 agonists silica and ATP remained intact in palmitate treated macrophages (Figure S4F and G).

To confirm the inhibition of Nlrp3 inflammasome activation associated with palmitate treatment was due to interference with cardiolipin synthesis, palmitate and control treated cells were LPS-primed and subjected to the broken cell system. Addition of cardiolipin liposomes to broken palmitate treated cells restored caspase-1 activation (Figure 6D). Together, these data suggest palmitate treatment inhibited IL-1β secretion in response to Nlrp3 agonists through diminished cardiolipin synthesis and thus that the ability of cardiolipin to directly activate Nlrp3 may be relevant biologically.

To determine if cardiolipin was necessary for the docking of Nlrp3 to the mitochondria, macrophages were incubated with palmitate or control and then stimulated with silica or linezolid. Macrophages were separated into mitochondrial and cytosolic fractions and the location of Nlrp3 was determined by immunoblot. Interference with cardiolipin synthesis through palmitate treatment inhibited the stimulus-dependent association of Nlrp3 with the mitochondria (Figure 6E), suggesting that in addition to activating Nlrp3, cardiolipin is necessary for the association of Nlrp3 with the mitochondria.

A critical step in the production of cardiolipin is the conversion of phosphatidylglycerol to cardiolipin by the mitochondrial enzyme cardiolipin synthase (CLS). In order to interfere with cardiolipin by a separate approach CLS was knocked down in THP-1 cells with two different siRNA constructs with a resultant 51% and 48% decrease in CLS mRNA (Figure S4H). As expected, diminished CLS mRNA resulted in decreased cellular cardiolipin as measured by thin layer chromatography (Figure S4I and J). Importantly, siRNA knockdown of CLS not only lessened cardiolipin levels but also impaired the release of IL-1β by THP-1 cells following stimulation with the ROS-dependent Nlrp3 activators silica and ATP as well as the ROS-independent activator linezolid (Figure 6F). Inhibition of CLS was specific to IL-1β and did not inhibit the release of TNFα (Figure 6F). In addition, while Nlrp3 inflammasome dependent IL-1β release was inhibited, activation of the AIM2 and Nlrc4 inflammasomes remained intact following treatment with F. tularensis LVS and P. aeruginosa, respectively (Figure 6F). Consistent with diminished IL-1β secretion in response to silica, ATP and linezolid after CLS siRNA treatment, we also observed a defect in caspase-1 activation in CLS siRNA treated cells specifically in response to Nlrp3 agonists (Figure 6G). Together, these data suggest the mitochondrial lipid cardiolipin is specifically required for Nlrp3 activation in response to both ROS-dependent and ROS-independent activators.

Discussion

Mitochondria play a crucial role in orchestrating the activation of the Nlrp3 inflammasome as evidenced by two studies showing that the generation of ROS by the mitochondrial respiratory chain is required for Nlrp3 inflammasome activation (Nakahira et al., 2011; Zhou et al., 2011). Our work identifies an Nlrp3 agonist that does not rely on ROS as an intermediary for inflammasome activation and suggests that Nlrp3 inflammasome activation is mediated by sensing mitochondrial dysfunction through the direct binding of Nlrp3 to the mitochondrial lipid cardiolipin.

The diverse molecules capable of activating the Nlrp3 inflammasome are structurally and biologically unrelated, making it unlikely that they directly interact with Nlrp3. It is more likely that these dissimilar agonists converge on a shared pathway leading to Nlrp3 inflammasome activation. Steps in Nlrp3 inflammasome activation that are shared by all known activators have until now included the efflux of potassium from the cell (Petrilli et al., 2007) and the generation of mitochondrial ROS (Nakahira et al., 2011; Zhou et al., 2011). However, we found that ROS was dispensable for Nlrp3 inflammasome activation in response to the oxazolidinone antibiotic linezolid, suggesting that ROS generation is not an absolute requirement for Nlrp3 inflammasome activation as has been previously suggested. We postulate that the events shared by ROS-dependent and ROS-independent activators of the Nlrp3 inflammasome may be mitochondrial destabilization and dysfunction, as stabilization with cyclosporine A blocked all Nlrp3 activators. In the case of ROS-dependent activators of the Nlrp3 inflammasome, such as silica, ATP and nigericin, it may be that ROS contribute to mitochondrial dysfunction and thereby indirectly promotes Nlrp3 inflammasome activation.

Given that mitochondria are phylogenetically bacterial symbionts of early eukaryotic cells and that cardiolipin is only found in mitochondria and bacteria, it is attractive to postulate that cardiolipin functions as an endogenous PAMP that is revealed upon mitochondrial dysfunction and detected by Nlrp3. Disruption of macrophage cardiolipin synthesis by palmitate exposure or siRNA knockdown of CLS resulted in specific defects in Nlrp3, but not Nlrc4 or AIM2, inflammasome activation, suggesting that cardiolipin plays a critical and selective role in Nlrp3 inflammasome activation. The requirement for cardiolipin in Nlrp3 inflammasome activation could be as a mitochondrial docking site for inflammasome assembly and subsequent activation or alternatively as the direct activating ligand for Nlrp3. Our studies suggest cardiolipin fulfills both roles, docking Nlrp3 at the mitochondria as well as directly activating the inflammasome.

Apaf-1 can form a large multiprotein complex called the apoptosome, which is responsible for the activation of caspase-9 in apoptosis. Parallels between the Nlrp3 inflammasome and the apoptosome have been highlighted previously when it was noted that low intracellular potassium is required for activation of the apoptosome as well as for activation of the Nlrp3 inflammasome (Cain et al., 2001; Tschopp, 2011). Additionally, interference with the important mitochondrial membrane channel VDAC not only inhibits apoptosis, but also reduces Nlrp3 inflammasome mediated IL-1β release (Zhou et al., 2011). Peroxidation of cardiolipin results in the release of bound cytochrome c into the cytosol where it binds to Apaf-1 (Garrido et al., 2006; Kagan et al., 2005). Apaf-1 oligomerization forms the apoptosome, which serves as a platform for caspase-9 activation (Wang, 2001). Cardiolipin also plays an important role in the extrinsic apoptotic pathway, as cardiolipin, following ROS-dependent translocation to the outer mitochondrial membrane, binds to caspase-8 and acts as a platform for its activation (Gonzalvez et al., 2008). Hence, it appears that while multiple apoptotic pathways converge on mitochondrial cardiolipin, cardiolipin is also critical in the pro-inflammatory pathway of Nlrp3 inflammasome activation. It remains unclear what factors are required to induce Nlrp3 inflammasome activation without additionally activating apoptosis, although our findings that staurosporine can activate caspase-1 only if cells are first primed suggests a preceding inflammatory stimulus may provide the signal for divergence. Alternatively, mitochondrial DNA has been suggested to play a role in Nlrp3 inflammasome activation (Nakahira et al., 2011; Shimada et al., 2012); it may be that mitochondrial DNA in concert with cardiolipin serves as a regulator in determining whether mitochondrial dysfunction will result in Nlrp3 inflammasome activation, necrosis or apoptosis.

Materials and Methods

Broken cell system

LPS-primed J774A.1 cells were resuspended at 1 × 107 cells/mL in Dulbecco's Phosphate-Buffered Saline (DPBS) (with Ca2+/Mg2+) supplemented with 25 mM KCl. For certain experiments, cells were pretreated with 0.5 mM palmitate for 8 hrs as above prior to priming. Cells were lysed by nitrogen cavitation as previously described (Iyer et al, 2009), after which lysates were incubated at 37°C for one hour in the presence of the indicated liposome and then analyzed by SDS-PAGE and immunoblot. Liposomes were produced as previously described (Iyer and Kusner, 1999). Briefly, stock concentrations of each lipid were made at 10 mM with cardiolipin (Avanti Polar Lipids) in ethanol, phosphatidylcholine (Avanti Polar Lipids) in chloroform:methanol (2:1), phosphatidic acid (Avanti Polar Lipids) in chloroform:methanol (2:1), or phosphatidylserine (Sigma) in chloroform. Organic solvents were evaporated under nitrogen, lipids were resuspended in DPBS and combined with phosphatidylcholine (1:1 molar ratio), which served as the vehicle for the liposomes. Liposomes were then sonicated with a probe sonicator for 10 min. Cardiolipin liposomes were added at concentrations of 300, 100 or 30 μM; phosphatidylcholine liposomes served as a negative control and were added at 300 μM.

Phospholipid analysis

THP-1 and J774A.1 cells were labeled with 6-10 μCi of 14C-linoleic acid (Perkin Elmer) in RPMI for 24 hrs. Unincorporated 14C-linoleic acid was washed away and where noted cells were treated with palmitate for 20 hrs. For phospholipid analysis of hCLS1 siRNA treated cells 14C-linoleic acid labeling was performed 48 hrs post-transfection with hCLS1 siRNA. Cells were harvested, washed, homogenized and lipids extracted by the Bligh-Dyer method (Bligh and Dyer, 1959) in the presence of 5 mM HCl. The organic phase was dried under nitrogen and equal amounts of radiolabeled lipids along with standard phospholipids were applied to a 10 × 10 cm HP-TLC silica gel 60 plate (Merck). Phospholipids were resolved by 1-D TLC using the solvent system chloroform:methanol:acetic acid:: 65:25:10 as described previously (Choi et al., 2007; Koprivnjak et al., 2011).

Mitochondrial isolation and immunoprecipitation

Mitochondria were isolated from J774A.1 cells as previously described (Iyer et al, 2009). For Nlrp3 immunoprecipitation, the mitochondrial fraction was solubilized in 1% Triton X-100 containing lysis buffer, pre cleared and subjected to immunoprecipitation with anti mouse Nlrp3 antibody (Enzo) or isotype control (IgG2b, Sigma). Lipids associated with the washed immunoprecipitates were extracted with chloroform: methanol (2:1 v/v) and subjected to 1-D HP-TLC as described above. Cytochrome c oxidase (COX) activity was assessed using a COX assay kit (Sigma), wherein COX activity is represented as decreased absorbance at 550 nM; inhibition of COX activity is represented as a reduction in the decrease in absorbance (ΔA550). Untreated mitochondria were treated with KCN (2.5 mM) as a positive control for COX inhibition.

Protein-lipid overlay assay and phospholipid coated bead assay

Binding of proteins to lipid strips (Echelon Biosciences) was performed according to the manufacturer's protocol. Strips were blocked in 3% fatty acid-free BSA in TBS (25 mM Tris, 150 mM NaCl pH7.2) followed by incubation with either column purified His-tagged constructs (10 μg/ml) or HA-tagged constructs (5 μg/ml), or transiently transfected HEK293T lysates expressing FLAG-tagged constructs in 1% fatty acid-free BSA in TBS for 8-16 hrs at 4°C. Strips were washed extensively in TTBS (0.05% Tween-20 in TBS) buffer to remove unbound proteins. Bound proteins were detected by immunoblotting and ECL detection (Pierce).

Binding of proteins to phospholipid-coated beads (Echelon Biosciences) was performed according to the manufacturer's protocol. Briefly, protein (10 μg Ni column eluates) or cell lysates were diluted in binding buffer (25 mM Tris, 150 mM NaCl pH 7.2, 0.5% Triton X-100), added to a 50 μl suspension of phospholipid-coated beads and incubated for 8 hr at 4°C. Beads were washed extensively with binding buffer to remove unbound protein; bound proteins were eluted by denaturing electrophoresis sample buffer and subjected to polyacrylamide gel electrophoresis followed by immunoblotting and ECL detection.

Supplementary Material

01

Highlights.

  • Nlrp3 inflammasome activation contributes to linezolid-induced in vivo toxicity;

  • Mitochondrial ROS is dispensable for linezolid-induced inflammasome activation;

  • Mitochondrial dysfunction is a prerequisite for Nlrp3 inflammasome activation;

  • Cardiolipin binds to Nlrp3 and is required for Nlrp3 inflammasome activation.

Acknowledgments

We thank Richard Flavell, John Bertin, and Millennium Pharmaceuticals for providing knockout mice, Fabio Martinon for providing Flag-tagged Nlrp3 constructs, Theresa Gioannini for providing His-tagged TLR4, and Paul Rothman and Rama Mallampalli for helpful discussion. NIH grants R01 AI087630 (F.S.S.), K08 AI067736 (S.L.C.), T32 AI007485 (J.R.J.), American Heart Association Scientist Development Grant 10SDG4140138 (R.H.), Muscular Dystrophy Association grant MDA171667 (R.H.), Asthma and Allergy Foundation of America fellowship (S.L.C.), and an Edward Mallinckrodt, Jr. Foundation scholarship (F.S.S.) supported this work. The Inflammation Program (W.M.N., S.L.C., and F.S.S) is supported by resources and use of facilities at the Veterans Affairs Medical Center, Iowa City, IA.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The authors have no conflicting financial interests.

References

  1. Agostini L, Martinon F, Burns K, McDermott MF, Hawkins PN, Tschopp J. NALP3 forms an IL-1beta-processing inflammasome with increased activity in Muckle-Wells autoinflammatory disorder. Immunity. 2004;20:319–325. doi: 10.1016/s1074-7613(04)00046-9. [DOI] [PubMed] [Google Scholar]
  2. Allam R, Darisipudi MN, Rupanagudi KV, Lichtnekert J, Tschopp J, Anders HJ. Cutting edge: cyclic polypeptide and aminoglycoside antibiotics trigger IL-1β secretion by activating the NLRP3 inflammasome. J Immunol. 2011;186:2714–2718. doi: 10.4049/jimmunol.1002657. [DOI] [PubMed] [Google Scholar]
  3. Ament PW, Jamshed N, Horne JP. Linezolid: its role in the treatment of gram-positive, drug-resistant bacterial infections. Am Fam Physician. 2002;65:663–670. [PubMed] [Google Scholar]
  4. Attassi K, Hershberger E, Alam R, Zervos MJ. Thrombocytopenia associated with linezolid therapy. Clin Infect Dis. 2002;34:695–698. doi: 10.1086/338403. [DOI] [PubMed] [Google Scholar]
  5. Bauernfeind F, Bartok E, Rieger A, Franchi L, Nunez G, Hornung V. Cutting edge: reactive oxygen species inhibitors block priming, but not activation, of the NLRP3 inflammasome. J Immunol. 2011;187:613–617. doi: 10.4049/jimmunol.1100613. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bligh EG, Dyer WJ. A rapid method of total lipid extraction and purification. Can J Biochem Physiol. 1959;37:911–917. doi: 10.1139/o59-099. [DOI] [PubMed] [Google Scholar]
  7. Cain K, Langlais C, Sun XM, Brown DG, Cohen GM. Physiological concentrations of K+ inhibit cytochrome c-dependent formation of the apoptosome. J Biol Chem. 2001;276:41985–41990. doi: 10.1074/jbc.M107419200. [DOI] [PubMed] [Google Scholar]
  8. Cassel SL, Eisenbarth SC, Iyer SS, Sadler JJ, Colegio OR, Tephly LA, Carter AB, Rothman PB, Flavell RA, Sutterwala FS. The Nalp3 inflammasome is essential for the development of silicosis. Proc Natl Acad Sci U S A. 2008;105:9035–9040. doi: 10.1073/pnas.0803933105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Choi SY, Gonzalvez F, Jenkins GM, Slomianny C, Chretien D, Arnoult D, Petit PX, Frohman MA. Cardiolipin deficiency releases cytochrome c from the inner mitochondrial membrane and accelerates stimuli-elicited apoptosis. Cell Death Differ. 2007;14:597–606. doi: 10.1038/sj.cdd.4402020. [DOI] [PubMed] [Google Scholar]
  10. Cruz CM, Rinna A, Forman HJ, Ventura AL, Persechini PM, Ojcius DM. ATP activates a reactive oxygen species-dependent oxidative stress response and secretion of proinflammatory cytokines in macrophages. J Biol Chem. 2007;282:2871–2879. doi: 10.1074/jbc.M608083200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Dawson MA, Davis A, Elliott P, Cole-Sinclair M. Linezolid-induced dyserythropoiesis: chloramphenicol toxicity revisited. Intern Med J. 2005;35:626–628. doi: 10.1111/j.1445-5994.2005.00912.x. [DOI] [PubMed] [Google Scholar]
  12. De Vriese AS, Coster RV, Smet J, Seneca S, Lovering A, Van Haute LL, Vanopdenbosch LJ, Martin JJ, Groote CC, Vandecasteele S, et al. Linezolid-induced inhibition of mitochondrial protein synthesis. Clin Infect Dis. 2006;42:1111–1117. doi: 10.1086/501356. [DOI] [PubMed] [Google Scholar]
  13. Dostert C, Petrilli V, Van Bruggen R, Steele C, Mossman BT, Tschopp J. Innate immune activation through Nalp3 inflammasome sensing of asbestos and silica. Science. 2008;320:674–677. doi: 10.1126/science.1156995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Eisenbarth SC, Colegio OR, O'Connor W, Sutterwala FS, Flavell RA. Crucial role for the Nalp3 inflammasome in the immunostimulatory properties of aluminium adjuvants. Nature. 2008;453:1122–1126. doi: 10.1038/nature06939. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Garrido C, Galluzzi L, Brunet M, Puig PE, Didelot C, Kroemer G. Mechanisms of cytochrome c release from mitochondria. Cell Death Differ. 2006;13:1423–1433. doi: 10.1038/sj.cdd.4401950. [DOI] [PubMed] [Google Scholar]
  16. Gerson SL, Kaplan SL, Bruss JB, Le V, Arellano FM, Hafkin B, Kuter DJ. Hematologic effects of linezolid: summary of clinical experience. Antimicrob Agents Chemother. 2002;46:2723–2726. doi: 10.1128/AAC.46.8.2723-2726.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Ghiringhelli F, Apetoh L, Tesniere A, Aymeric L, Ma Y, Ortiz C, Vermaelen K, Panaretakis T, Mignot G, Ullrich E, et al. Activation of the NLRP3 inflammasome in dendritic cells induces IL-1β-dependent adaptive immunity against tumors. Nat Med. 2009;15:1170–1178. doi: 10.1038/nm.2028. [DOI] [PubMed] [Google Scholar]
  18. Gonzalvez F, Schug ZT, Houtkooper RH, MacKenzie ED, Brooks DG, Wanders RJ, Petit PX, Vaz FM, Gottlieb E. Cardiolipin provides an essential activating platform for caspase-8 on mitochondria. J Cell Biol. 2008;183:681–696. doi: 10.1083/jcb.200803129. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Green SL, Maddox JC, Huttenbach ED. Linezolid and reversible myelosuppression. JAMA. 2001;285:1291. doi: 10.1001/jama.285.10.1291. [DOI] [PubMed] [Google Scholar]
  20. Iyer SS, Kusner DJ. Association of phospholipase D activity with the detergent-insoluble cytoskeleton of U937 promonocytic leukocytes. J Biol Chem. 1999;274:2350–2359. doi: 10.1074/jbc.274.4.2350. [DOI] [PubMed] [Google Scholar]
  21. Iyer SS, Pulskens WP, Sadler JJ, Butter LM, Teske GJ, Ulland TK, Eisenbarth SC, Florquin S, Flavell RA, Leemans JC, et al. Necrotic cells trigger a sterile inflammatory response through the Nlrp3 inflammasome. Proc Natl Acad Sci U S A. 2009;106:20388–20393. doi: 10.1073/pnas.0908698106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kagan VE, Tyurin VA, Jiang J, Tyurina YY, Ritov VB, Amoscato AA, Osipov AN, Belikova NA, Kapralov AA, Kini V, et al. Cytochrome c acts as a cardiolipin oxygenase required for release of proapoptotic factors. Nat Chem Biol. 2005;1:223–232. doi: 10.1038/nchembio727. [DOI] [PubMed] [Google Scholar]
  23. Kanneganti TD, Ozoren N, Body-Malapel M, Amer A, Park JH, Franchi L, Whitfield J, Barchet W, Colonna M, Vandenabeele P, et al. Bacterial RNA and small antiviral compounds activate caspase-1 through cryopyrin/Nalp3. Nature. 2006;440:233–236. doi: 10.1038/nature04517. [DOI] [PubMed] [Google Scholar]
  24. Koprivnjak T, Zhang D, Ernst CM, Peschel A, Nauseef WM, Weiss JP. Characterization of Staphylococcus aureus cardiolipin synthases 1 and 2 and their contribution to accumulation of cardiolipin in stationary phase and within phagocytes. J Bacteriol. 2011;193:4134–4142. doi: 10.1128/JB.00288-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Lee GS, Subramanian N, Kim AI, Aksentijevich I, Goldbach-Mansky R, Sacks DB, Germain RN, Kastner DL, Chae JJ. The calcium-sensing receptor regulates the NLRP3 inflammasome through Ca2+ and cAMP. Nature. 2012;492:123–127. doi: 10.1038/nature11588. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. MacGowan AP. Pharmacokinetic and pharmacodynamic profile of linezolid in healthy volunteers and patients with Gram-positive infections. J Antimicrob Chemother. 2003;51(Suppl 2):ii17–25. doi: 10.1093/jac/dkg248. [DOI] [PubMed] [Google Scholar]
  27. Mariathasan S, Weiss DS, Newton K, McBride J, O'Rourke K, Roose-Girma M, Lee WP, Weinrauch Y, Monack DM, Dixit VM. Cryopyrin activates the inflammasome in response to toxins and ATP. Nature. 2006;440:228–232. doi: 10.1038/nature04515. [DOI] [PubMed] [Google Scholar]
  28. Martinon F. Signaling by ROS drives inflammasome activation. Eur J Immunol. 2010;40:616–619. doi: 10.1002/eji.200940168. [DOI] [PubMed] [Google Scholar]
  29. Martinon F, Petrilli V, Mayor A, Tardivel A, Tschopp J. Gout-associated uric acid crystals activate the NALP3 inflammasome. Nature. 2006;440:237–241. doi: 10.1038/nature04516. [DOI] [PubMed] [Google Scholar]
  30. Mayor A, Martinon F, De Smedt T, Petrilli V, Tschopp J. A crucial function of SGT1 and HSP90 in inflammasome activity links mammalian and plant innate immune responses. Nat Immunol. 2007;8:497–503. doi: 10.1038/ni1459. [DOI] [PubMed] [Google Scholar]
  31. McKee EE, Ferguson M, Bentley AT, Marks TA. Inhibition of mammalian mitochondrial protein synthesis by oxazolidinones. Antimicrob Agents Chemother. 2006;50:2042–2049. doi: 10.1128/AAC.01411-05. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Meissner F, Molawi K, Zychlinsky A. Superoxide dismutase 1 regulates caspase-1 and endotoxic shock. Nat Immunol. 2008;9:866–872. doi: 10.1038/ni.1633. [DOI] [PubMed] [Google Scholar]
  33. Misawa T, Takahama M, Kozaki T, Lee H, Zou J, Saitoh T, Akira S. Microtubule-driven spatial arrangement of mitochondria promotes activation of the NLRP3 inflammasome. Nat Immunol. 2013 doi: 10.1038/ni.2550. [DOI] [PubMed] [Google Scholar]
  34. Murakami T, Ockinger J, Yu J, Byles V, McColl A, Hofer AM, Horng T. Critical role for calcium mobilization in activation of the NLRP3 inflammasome. Proc Natl Acad Sci U S A. 2012;109:11282–11287. doi: 10.1073/pnas.1117765109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Nakahira K, Haspel JA, Rathinam VA, Lee SJ, Dolinay T, Lam HC, Englert JA, Rabinovitch M, Cernadas M, Kim HP, et al. Autophagy proteins regulate innate immune responses by inhibiting the release of mitochondrial DNA mediated by the NALP3 inflammasome. Nat Immunol. 2011;12:222–230. doi: 10.1038/ni.1980. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Osman C, Voelker DR, Langer T. Making heads or tails of phospholipids in mitochondria. J Cell Biol. 2011;192:7–16. doi: 10.1083/jcb.201006159. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Ostrander DB, Sparagna GC, Amoscato AA, McMillin JB, Dowhan W. Decreased cardiolipin synthesis corresponds with cytochrome c release in palmitate-induced cardiomyocyte apoptosis. J Biol Chem. 2001;276:38061–38067. doi: 10.1074/jbc.M107067200. [DOI] [PubMed] [Google Scholar]
  38. Petrilli V, Papin S, Dostert C, Mayor A, Martinon F, Tschopp J. Activation of the NALP3 inflammasome is triggered by low intracellular potassium concentration. Cell Death Differ. 2007;14:1583–1589. doi: 10.1038/sj.cdd.4402195. [DOI] [PubMed] [Google Scholar]
  39. Rossol M, Pierer M, Raulien N, Quandt D, Meusch U, Rothe K, Schubert K, Schoneberg T, Schaefer M, Krugel U, et al. Extracellular Ca2+ is a danger signal activating the NLRP3 inflammasome through G protein-coupled calcium sensing receptors. Nat Commun. 2012;3:1329. doi: 10.1038/ncomms2339. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Shimada K, Crother TR, Karlin J, Dagvadorj J, Chiba N, Chen S, Ramanujan VK, Wolf AJ, Vergnes L, Ojcius DM, et al. Oxidized Mitochondrial DNA Activates the NLRP3 Inflammasome during Apoptosis. Immunity. 2012;36:401–414. doi: 10.1016/j.immuni.2012.01.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Solle M, Labasi J, Perregaux DG, Stam E, Petrushova N, Koller BH, Griffiths RJ, Gabel CA. Altered cytokine production in mice lacking P2X(7) receptors. J Biol Chem. 2001;276:125–132. doi: 10.1074/jbc.M006781200. [DOI] [PubMed] [Google Scholar]
  42. Sood R, Bhadauriya T, Rao M, Gautam R, Malhotra S, Barman TK, Upadhyay DJ, Rattan A. Antimycobacterial activities of oxazolidinones: a review. Infect Disord Drug Targets. 2006;6:343–354. doi: 10.2174/187152606779025860. [DOI] [PubMed] [Google Scholar]
  43. Soriano A, Miro O, Mensa J. Mitochondrial toxicity associated with linezolid. N Engl J Med. 2005;353:2305–2306. doi: 10.1056/NEJM200511243532123. [DOI] [PubMed] [Google Scholar]
  44. Sutterwala FS, Flavell RA. NLRC4/IPAF: a CARD carrying member of the NLR family. Clin Immunol. 2009;130:2–6. doi: 10.1016/j.clim.2008.08.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Sutterwala FS, Ogura Y, Szczepanik M, Lara-Tejero M, Lichtenberger GS, Grant EP, Bertin J, Coyle AJ, Galan JE, Askenase PW, et al. Critical role for NALP3/CIAS1/Cryopyrin in innate and adaptive immunity through its regulation of caspase-1. Immunity. 2006;24:317–327. doi: 10.1016/j.immuni.2006.02.004. [DOI] [PubMed] [Google Scholar]
  46. Taketani T, Kanai R, Fukuda S, Uchida Y, Yasuda K, Mishima S, Suyama T, Kodama R, Yoshino I, Kunishi H, et al. Pure red cell precursor toxicity by linezolid in a pediatric case. J Pediatr Hematol Oncol. 2009;31:684–686. doi: 10.1097/MPH.0b013e3181b1ed42. [DOI] [PubMed] [Google Scholar]
  47. Tschopp J. Mitochondria: Sovereign of inflammation? Eur J Immunol. 2011;41:1196–1202. doi: 10.1002/eji.201141436. [DOI] [PubMed] [Google Scholar]
  48. Tschopp J, Schroder K. NLRP3 inflammasome activation: The convergence of multiple signalling pathways on ROS production? Nat Rev Immunol. 2010;10:210–215. doi: 10.1038/nri2725. [DOI] [PubMed] [Google Scholar]
  49. van Bruggen R, Koker MY, Jansen M, van Houdt M, Roos D, Kuijpers TW, van den Berg TK. Human NLRP3 inflammasome activation is Nox1-4 independent. Blood. 2010;115:5398–5400. doi: 10.1182/blood-2009-10-250803. [DOI] [PubMed] [Google Scholar]
  50. Vandanmagsar B, Youm YH, Ravussin A, Galgani JE, Stadler K, Mynatt RL, Ravussin E, Stephens JM, Dixit VD. The NLRP3 inflammasome instigates obesity-induced inflammation and insulin resistance. Nat Med. 2011;17:179–188. doi: 10.1038/nm.2279. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Walev I, Reske K, Palmer M, Valeva A, Bhakdi S. Potassium-inhibited processing of IL-1 beta in human monocytes. EMBO J. 1995;14:1607–1614. doi: 10.1002/j.1460-2075.1995.tb07149.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Wang X. The expanding role of mitochondria in apoptosis. Genes Dev. 2001;15:2922–2933. [PubMed] [Google Scholar]
  53. Wen H, Gris D, Lei Y, Jha S, Zhang L, Huang MT, Brickey WJ, Ting JP. Fatty acid-induced NLRP3-ASC inflammasome activation interferes with insulin signaling. Nat Immunol. 2011;12:408–415. doi: 10.1038/ni.2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Zhong Z, Zhai Y, Liang S, Mori Y, Han R, Sutterwala FS, Qiao L. TRPM2 links oxidative stress to NLRP3 inflammasome activation. Nat Commun. 2013;4:1611. doi: 10.1038/ncomms2608. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Zhou R, Yazdi AS, Menu P, Tschopp J. A role for mitochondria in NLRP3 inflammasome activation. Nature. 2011;469:221–225. doi: 10.1038/nature09663. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01
NIHMS513034-supplement-01.docx (4.9MB, docx)

RESOURCES