Glucosamine inhibits IL-1β expression by preserving mitochondrial integrity and disrupting assembly of the NLRP3 inflammasome

The NLRP3 inflammasome promotes the pathogenesis of metabolic, neurodegenerative and infectious diseases. Increasing evidences show that the NLRP3 inflammasome is a promising therapeutic target in inflammatory diseases. Glucosamine is widely used as a dietary supplement to promote the health of cartilage tissue and is expected to exert anti-inflammatory activity in joint inflammation, which is a nucleotide-binding oligomerization domain-like receptor containing pyrin domain 3 (NLRP3) inflammasome-associated complication. Here, we investigated whether GlcN inhibits the NLRP3 inflammasome and dissected the underlying molecular mechanisms. We found that GlcN suppressed the NLRP3 inflammasome in mouse and human macrophages. A mechanistic study revealed that GlcN inhibited the expression of NLRP3 and IL-1β precursor by reducing reactive oxygen species generation and NF-κB activation in lipopolysaccharide-activated macrophages. GlcN also suppressed mitochondrial reactive oxygen species generation and mitochondrial integrity loss in NLRP3-activated macrophages. Additionally, GlcN disrupted NLRP3 inflammasome assembly by inhibiting NLRP3 binding to PKR, NEK7 and ASC. Furthermore, oral administration of GlcN reduced peritoneal neutrophils influx and lavage fluids concentrations of IL-1β, IL-6 MCP-1 and TNF-α in uric acid crystal-injected mice. These results indicated that GlcN might be a novel dietary supplement for the amelioration of NLRP3 inflammasome-associated complications.

and N-acetyl-D-glucosamine (GlcNAc) for 2 h, followed by incubation with ATP (5 mM) for 0.5 h. The IL-1β expression levels in the supernatants were measured by ELISA. (F) J774A.1 macrophages were incubated for 4 h with LPS (1 µg/ml) or PamsCSK4 (1 µg/ml; for non-canonical inflammasome) followed by incubation for 2 h with GlcN, followed by transfection with poly(dA/dT) (2 µg/ml) or LPS (2 µg/ml) for 6 h or by Salmonella infection (30 MOI) for 2 h. The IL-1β expression levels in the supernatants were measured by ELISA. The ELISA data are expressed as the mean ± SD of separate experiments as indicated. The Western blotting results Glucosamine is widely used as a dietary supplement to promote the health of joints and may reduce the pain caused by joint diseases. In addition to its joint-protective effects, glucosamine exerts anti-inflammatory activities and ameliorates inflammatory diseases. For example, glucosamine inhibits inducible nitric oxide synthase expression in lipopolysaccharide (LPS)-activated microglia 11 and reduces COX-2 expression in human breast cancer cells 12 . Glucosamine exerts in vivo anti-inflammatory effects because it ameliorates brain inflammation 11 , inflammatory bowel disease 13 , lung inflammation 14 , atherosclerosis 15 , experimental autoimmune encephalomyelitis 16 and renal fibrosis 17 . However, the effect of glucosamine on the NLRP3 inflammasome has not been explored yet.
Glucosamine is widely used as a dietary supplement to promote the health of osteoarthritis (OA) and rheumatoid arthritis (RA) 18,19 . OA and RA are the major joint diseases associated with elevated NLRP3 inflammasome 20,21 . Because glucosamine ameliorates NLRP3-associated inflammatory diseases, we hypothesize that the protective effects of glucosamine may be associated with its anti-NLRP3 inflammasome activity. In this report, we investigated whether GlcN inhibits the NLRP3 inflammasome and dissected the underlying molecular mechanisms in macrophages and in a NLRP3-associated mouse disease model.

Results
GlcN inhibits the activation of the NLRP3 inflammasome. We investigated the inhibition potential of D-(+)-glucosamine hydrochloride (GlcN) toward the NLRP3 inflammasome. First, LPS-primed mouse J774A.1 macrophages were incubated with GlcN for 2 h before stimulation with activators of the NLRP3 inflammasome. We found that GlcN (10 and 30 mM) significantly inhibited IL-1β secretion induced by ATP, nigericin, monosodium urate crystals (MSU) and E. coli infection (Fig. 1A). Under the same condition, GlcN did not significantly reduce TNF-α secretion induced by ATP and nigericin, indicating that GlcN specifically inhibited IL-1β secretion (Fig. 1B). GlcN also inhibited NLRP3 inflammasome-derived IL-1β secretion in LPS-primed human THP-1 macrophages (Fig. 1C). GlcN at lower concentration (1, 3 and 5 mM) also inhibited ATP-, MSUand palmitate-BSA-induced IL-1β secretion in LPS-primed human THP-1 macrophages (Fig. 1D). In addition, GlcN suppressed ATP-and MSU-induced IL-1β secretion in human primary peripheral blood mononuclear cells (PBMCs) (Fig. 1E). To investigate whether the inhibitory effect of GlcN specific to LPS priming only, we investigated the effect of GlcN on the nigericin-induced IL-1β secretion in Pam3CSK4 (toll-like receptor 2 ligand)primed J774A.1 macrophages. We found that GlcN significantly reduced IL-1β expression levels in the culture medium (Fig. 1F), indicating that the inhibitory effect of GlcN on IL-1β secretion was not specific to LPS priming only. Using Western blotting analysis, we found that GlcN reduced the expression levels of proIL-1β/IL-1β and IL-18 in the supernatants of ATP-activated or E. coli-infected J774A.1 macrophages ( Fig. 2A). GlcN also inhibited the expression levels of IL-1β in the cell lysates of ATP-activated or E. coli-infected J774A.1 macrophages (Fig. 2B). In addition, the inhibitory effect of GlcN on the NLRP3 inflammasome was confirmed by reduced expression levels of active caspase-1 (p10) in the supernatants of ATP-activated or E. coli-infected J774A.1 macrophages (Fig. 2C). It has been demonstrated that oligomeric ASC particles are released from macrophages into supernatants upon NLRP3 inflammasome activation and act as an extracellular danger signal that amplifies the inflammatory response 22 . By detection of the expression levels of ASC in the supernatants, we found that ASC particles were released from ATP-activated or E. coli-infected J774A.1 macrophages, and this effect was reduced by GlcN (Fig. 2D). Notably, 10 and 30 mM of D-(+)-glucose, D-glucosamine 3-sulphate, D-(+)-galactosamine hydrochloride and N-acetyl-D-glucosamine did not alter IL-1β secretion in ATP-activated macrophages (Fig. 2E). These findings indicated that the NLRP3 inflammasome was inhibited by GlcN, and this effect was not due to the osmotic effect. In addition, to investigate whether GlcN specifically inhibits the NLRP3 inflammasome, the effect of GlcN on the absent in melanoma 2 (AIM2)-, non-canonical-or NLR family CARD domain containing 4 (NLRC4)-inflammasome was tested. The results showed that GlcN not only inhibited NLRP3 inflammasome but also reduced IL-1β secretion in AIM2-, non-canonical-and NLRC4-inflammasome activated J774A.1 macrophages, which are stimulated by poly(dA:dT) and LPS transfection or by Salmonella infection, respectively (Fig. 2F).
GlcN inhibits mitochondrial damage. K + efflux is the early event of the activation signal of the NLRP3 inflammasome, which activates downstream signalling pathways, leading to caspase-1 activation 23 . To investigate whether GlcN inhibits the NLRP3 inflammasome by targeting the downstream signal of K + efflux, we replaced the culture medium with K + free medium to facilitate K + efflux to promote NLRP3 inflammasome activation. We found that GlcN inhibited IL-1β secretion induced by K + free medium and nigericin, a bacterial toxin that activates the NLRP3 inflammasome through K + efflux, in J774A.1 macrophages (Fig. 3A). These results indicated that GlcN reduced IL-1β secretion by inhibiting the downstream events of K + efflux during NLRP3 inflammasome activation. In addition, mitochondrial damage caused by K + efflux plays important roles in NLRP3 inflammasome activation 24 . We further investigate whether GlcN inhibited NLRP3 inflammasome activation by reducing mitochondrial damage. We found that ATP treatment caused mitochondria damage, as evidenced by reduced Mitotracker deep red staining (Fig. 3B). Treatment with GlcN or antioxidant N-acetyl-L-cysteine (NAC) reversed Mitotracker deep red staining, indicating reduced mitochondrial injury (Fig. 3B). To confirm the effect of GlcN on preserving mitochondrial integrity, DiOC 2 (3), a membrane-potential-sensitive cyanine dye, which penetrates are representative of three different experiments and the histogram shows the quantification expressed as the mean ± SD for these three experiments. *, *** and **** indicate a significant difference at the level of p < 0.05, p < 0.001 and p < 0.0001, respectively, compared to NLRP3 activator-treated cells. (One-way ANOVA with Dunnett's multiple comparisons test). The blots in (A-D) were cropped from different gels; full-length blots are included in the " Supplementary Information". www.nature.com/scientificreports www.nature.com/scientificreports/ the cytosol of eukaryotic cells and accumulates primarily in mitochondria with active membrane potentials, producing bright, red fluorescence, was used. We found that ATP treatment caused the reduced DiOC 2 (3) staining in LPS-primed macrophages, but this effect was inhibited by GlcN treatment (Fig. 3C). This result indicated that GlcN preserved mitochondrial integrity. Interestingly, GlcN alone increased the DiOC 2 (3) staining compared to the control cells, indicating that GlcN increased mitochondrial function (Fig. 3C). Furthermore, ATP treatment induced mitochondrial ROS production (Fig. 3D), leading to NLRP3 inflammasome assembly 24 . We found that ATP-mediated mitochondrial ROS production was reduced by GlcN treatment in J774 A.1 macrophages (Fig. 3D). These results demonstrated that GlcN inhibited NLRP3 inflammasome activation by reducing mitochondrial integrity loss.

GlcN inhibits NLRP3 inflammasome assembly.
It was demonstrated that activated double-stranded RNA-activated protein kinase (PKR) positively regulates the NLRP3 inflammasome by physically interacting with NLRP3 25 . These results prompted us to investigate the effect of GlcN on PKR activation and the interaction between PKR and NLRP3. We found that the phosphorylation level of PKR was increased by ATP treatment and that this effect was reduced by GlcN or 2-aminopurine (2-AP), a potent inhibitor of PKR (Fig. 4A). Notably, PKR physically interacted with NLRP3 in ATP-stimulated cells, and the interaction was disrupted by GlcN or 2-AP (Fig. 4B). Additionally, NIMA-related kinase 7 (NEK7), an NLRP3-binding protein, is an essential mediator of NLRP3 inflammasome activation downstream of K + efflux 26 . We found that NEK7 binds to NLRP3 upon ATP stimulation and that this effect was suppressed by GlcN and KCl, an inhibitor of K + efflux (Fig. 4C). It was demonstrated that ASC plays key roles in the regulation of the NLRP3 inflammasome via self-association and formation of a complex with NLRP3 27 . We asked whether GlcN affects NLRP3/ASC complex formation. We found that ASC formed a complex with NLRP3 upon ATP stimulation, and this effect was suppressed by GlcN . The data are expressed as the means ± SD of separate experiments as indicated. *, **, *** and **** indicate a significant difference at the level of p < 0.05, p < 0.01, p < 0.001 and p < 0.0001, respectively, compared to K + -free medium-, nigericin-or ATP-treated cells. (One-way ANOVA with Dunnett's multiple comparisons test).
www.nature.com/scientificreports www.nature.com/scientificreports/ and KCl (Fig. 4D). Furthermore, we found that ATP-induced ASC oligomerization was significantly reduced by GlcN (Fig. 4E). These results indicated that GlcN suppressed NLRP3 inflammasome by inhibiting NLRP3 inflammasome assembly. (1 µg/ml) followed by incubation for 2 h with GlcN. Cells were then incubated with ATP (5 mM) for 0.5 h. The cell lysates were crosslinked by disuccinimidyl suberate, and ASC oligomerization was measured by Western blotting. The Western blotting results are representative of three different experiments and the histogram shows the quantification expressed as the mean ± SD for these three experiments. *, **, *** and **** indicate a significant difference at the level of p < 0.05, p < 0.01, p < 0.001 and p < 0.0001, respectively, compared to ATPtreated cells. NS: not significant. (One-way ANOVA with Dunnett's multiple comparisons test). The blots in this figure were cropped from different gels; full-length blots are included in the "Supplementary Information". www.nature.com/scientificreports www.nature.com/scientificreports/ GlcN inhibits the priming signal of the NLRP3 inflammasome by reducing ROS production and NF-κB activation. In addition to the activation signal, full activation of the NLRP3 inflammasome required the priming signal from pathogen-associated molecular patterns, such as LPS, which transcriptional control the expression of NLRP3 and proIL-1β 28 . To investigate whether the priming signal of the NLRP3 inflammasome was affected by GlcN, LPS-and Pam3CSK4-mediated NLRP3 and proIL-1β expression in GlcN-treated J774A.1 macrophages were measured by Western blotting. We found that the NLRP3 and proIL-1β expression levels in both LPS-and Pam3CSK4-activated macrophages were significantly inhibited by GlcN (Fig. 5A,B), indicating that GlcN inhibited the priming signal of the NLRP3 inflammasome. ROS has been demonstrated to act as a critical event upstream of the LPS-mediated priming signal of the NLRP3 inflammasome 29 . We found that GlcN significantly inhibited ROS production in LPS-activated macrophages (Fig. 5C). In addition, NF-κB activation is also required for LPS-mediated NLRP3 and proIL-1β expression 30 . In this study, we found that LPS increased the transcriptional activity of NF-κB, and this effect was reduced by GlcN (Fig. 5D). To provide additional evidences for GlcN-mediated NF-κB inhibition, we investigated the effect of GlcN on LPS-induced phosphorylation of IκBα and IKKα/β in J774A.1 macrophages. We found that the phosphorylation levels of IκBα and IKKα/β were reduced by GlcN (Fig. 5E). However, the results showed that GlcN had no effect on the phosphorylation levels of www.nature.com/scientificreports www.nature.com/scientificreports/ ERK1/2, JNK1/2 and p38 (Fig. 5F), which are important signalling molecules for NLRP3 and proIL-1β expression in LPS-activated macrophages 31 .

GlcN inhibits the NLRP3 inflammasome and inflammation in MSU-injected mice. To investigate
whether GlcN exhibits anti-NLRP3 inflammasome activity in vivo, an NLRP3-associated inflammatory disease mouse model of MSU-mediated peritonitis was used 32 . As can be seen in Fig. 6A, MSU intraperitoneal injection (i.p.) caused significant peritoneal neutrophils influx; notably, oral administration of 250 mg/kg GlcN (three times daily before MSU injection) or i.p. of colchicine (once before MSU injection) significantly reduced peritoneal neutrophils influx. However, the total numbers of peritoneal lavage cells were no significant difference (Fig. 6A). MSU injection also increased the levels of IL-1β, IL-6, TNF-α and MCP-1 in the peritoneal fluid, and these effects were reduced by GlcN and colchicines (Fig. 6B). These findings indicated that GlcN inhibits the NLRP3 inflammasome and inflammation in vivo.

Discussion
The NLRP3 inflammasome is an important focus of biomedical research, and the number of NLRP3-related studies is rapidly growing. The NLRP3 inflammasome plays crucial roles in the pathogenesis of metabolic, autoimmune, neurodegenerative diseases as well as in host defence against invading pathogens 2 . Over the past decade, research efforts have focused on understanding the regulation of the NLRP3 inflammasome; currently, it has become a promising molecular target in the fight against inflammatory diseases 33 . We identified several small-molecule inhibitors of the NLRP3 inflammasome from natural products that ameliorate NLRP3-associated disease in vitro and in vivo. Citral and Epigallocatechin-3-gallate from Litsea cubeba and green tea, respectively, prevent lupus nephritis in mice 4,34 . Resveratrol, osthole and antroquinonol from the skin of red grapes, Cnidiummonnieri (L.) Cusson seeds and the Antrodia camphorata mycelium, respectively, mitigate IgA nephropathy in mice 9,10,35 . A cinnamaldehyde derivative from the essential oil of the leaves of Cinnamomum osmophloeum kaneh ameliorates LPS-induced renal inflammation in mice 36 . In addition, two small-molecule inhibitors of the NLRP3 inflammasome, MCC950 and β-hydroxybutyrate, were reported in Nature Medicine 2015 and showed promise in the clinic 7,37 . These results indicated that small molecules have the potential to be developed into NLRP3-targeting therapeutics against various NLRP3 inflammasome-associated diseases. www.nature.com/scientificreports www.nature.com/scientificreports/ We demonstrated that GlcN inhibited the NLRP3 inflammasome, machinery that plays important roles in the pathogenesis of diabetes and atherosclerosis 38,39 . These results suggested that GlcN may ameliorate diabetes and atherosclerosis by reducing NLRP3 inflammasome activation. However, there are little data regarding the side effects of GlcN supplementation on metabolic or vascular systems. It was reported that GlcN promotes endoplasmic reticulum stress in retinal Müller cells and leads to the resistance to insulin action, suggesting that GlcN potentially exacerbates diabetes-induced hyperglycaemia 40 . Another study also indicated that GlcN enhanced insulin resistance in mice fed a normal diet but ameliorated obesity and insulin resistance in mice fed a high-fat diet 41 . In addition, GlcN supplementation promotes endoplasmic reticulum stress, hepatic steatosis and accelerated atherosclerosis in apolipoprotein E-deficient mice, and these effects are independent of detectable changes in the levels of glucose, insulin and lipid in blood 42 . However, another report indicated that long-term GlcN supplementation was safe and no increase in atherosclerosis occurred, although increased hyperlipidaemia was observed 43 . These controversial findings suggest that the integrative function of GlcN may depend on the physiological or nutrient status.
Our current findings demonstrated that GlcN inhibited the priming signal of the NLRP3 inflammasome partially through reducing LPS-induced ROS production in macrophages. The anti-oxidative effect of GlcN was observed in a mouse model of cigarette smoke-induced lung inflammation 14 . Our previous study indicated that the LPS-mediated priming signal of the NLRP3 inflammasome was partially through MAPK 31 ; however, LPS-induced MAPK activation was not inhibited by GlcN. In the activation signal of the NLRP3 inflammasome, K + efflux plays important roles by triggering downstream signalling, leading to activation of the NLRP3 inflammasome 23,24 . Although we had no solid evidence showing the effect of GlcN on the P 2 X 7 receptor, a nonselective cationic channel abundantly expressed in macrophages that allows the efflux of K + 44 , we demonstrated that GlcN inhibited downstream signalling of K + efflux, thereby inhibiting IL-1β secretion (Fig. 3A). Mitochondrial damage is one of the downstream signalling events of K + efflux, which is characterized by mitochondrial ROS production and the loss of the mitochondrial inner transmembrane potential 24 . Mitochondrial ROS promotes the translocation of chloride intracellular channels to the plasma membrane for the induction of chloride efflux to promote NLRP3 inflammasome assembly, caspase-1 activation and IL-1β secretion 45 . We found that GlcN inhibited mitochondrial ROS production (Fig. 3D) and NLRP3 inflammasome assembly (Fig. 4); however, the effect of GlcN on chloride efflux needs further investigation. It should be noted that several events associated with the NLRP3 inflammasome activation were inhibited by GlcN. It is important to know whether GlcN had multiple targets or it just inhibited one major molecule and others are just a consequence. According to our previous study 31 , ROS is the key molecule controlling both priming and activating signals of the NLRP3 inflammasome. Our current findings demonstrated that GlcN inhibited the NLRP3 inflammasome by affecting both priming and activating events. We speculated that ROS may be the major molecule that inhibited by GlcN and other events such as reduced NF-κB activation and inflammasome assembly might be the consequence of ROS inhibition. However, more studies are needed to support the hypothesis.
Autophagy is cellular machinery that protects cells from damage. Recent reports have indicated that autophagy inhibits the NLRP3 inflammasome by reducing mitochondrial integrity loss and targeting inflammasome components for lysosomal degradation 46 . A previous study indicated that GlcN induced autophagy in chondrocytes in vitro and in articular cartilage in vivo 47 . Thus, we speculate that GlcN-mediated inhibition of the NLRP3 inflammasome may partially occur through autophagy induction in macrophages. Notably, LPS-induced TNF-α production was not significantly inhibited by GlcN in ATP-or nigericin-treated cells, but inhibited significantly in MSU-treated cells (Fig. 1B). The reduced TNF-α expression level in MSU-treated cells was due to the experimental condition. In the ATP-or nigericin-treated cells, cells exposed to LPS for total 6.5 h and exposed to GlcN only for the last 2.5 h. However, in the MSU-treated cells, cells exposed to LPS for total 30 h and exposed to GlcN for the last 26 h. We suggested that the reduced TNF-α in MSU-treated cells may result from (1) GlcN inhibited the LPS-mediated signal in the last 26 h; and (2) GlcN-induced autophagy that reduces TNF-α production.
Oral administration of GlcN reduced the peritoneal neutrophils influx and cytokine expression in the peritoneal fluid; however, the total numbers of peritoneal lavage cells were not significantly reduced by GlcN treatment (Fig. 6). The peritoneal lavage cells include not only neutrophils, but also epithelial cells, lymphocytes, plasma cells, monocytes, macrophages and mesothelial cells etc. In our current study we found that while peritoneal recruitment of neutrophils was reduced by GlcN in MSU-injected mice, total peritoneal lavage cells seem to be similar. This result indicated that GlcN specifically reduced the neutrophils influx, but not overall reduced the total number of peritoneal lavage cells. In the in vivo study, both GlcN and colchicine reduced the peritoneal neutrophils influx and cytokine expression in the peritoneal fluid, and GlcN was found to be more potent than colchicine in cytokine inhibition (Fig. 6). One of the possibilities is that mice orally administered with GlcN once daily for three days, but i.p. injection with colchicine only once before MSU injection. In previous studies, plasma concentrations of GlcN in rats or human treated with GlcN were increased to dozens to hundreds μM 48,49 . Although we demonstrated that GlcN at 1 mM can inhibit NLRP3 inflammasome in human THP-1 macrophages (Fig. 1C), the GlcN concentration is still higher than the physiological condition. However, many studies demonstrated that GlcN exerted benefit effects in vivo 11,[13][14][15][16][17] . We speculated that daily low dose GlcN exposure in vivo may inhibit the NLRP3 inflammasome. In addition, our current finding provided a possibility to use GlcN as a core structure to make derivatives with enhanced activity 50 . Additionally, as exciting finding is that GlcN extends the life span of nematodes and mice by activating AMP-activated protein kinase 51 . However, it should be noted that GlcN may induce renal toxicity in mice 52 . Therefore, although we provide evidence that GlcN exerts anti-NLRP3 inflammasome activity in vitro and in vivo, more studies should be performed before using GlcN as a therapeutic agent for NLRP3-associated diseases. Taken together, we found that GlcN might be a novel dietary supplement for the amelioration of NLRP3 inflammasome-associated complications by suppressing both priming and activation signals of the NLRP3 inflammasome (Fig. 7). Cell culture. The mouse macrophage cell line J774A.1 and human monocytic leukaemia cell line THP-1 were purchased from the American Type Culture Collection (Rockville, MD). We used whole blood from healthy volunteers recruited at the Tri-Service General Hospital in Taipei, Taiwan, after their approval according to the Institutional Review Board of the Tri-Service General Hospital, National Defense Medical Center and the patients' informed consent (TSGH-IRB-2-106-05-190 and TSGH-IRB-2-106-05-009). Human primary PBMCs were separated from whole blood from healthy volunteers by density gradient centrifugation using Histopaque-1077 within 24 h after blood was collected, and all experimental protocols were performed in accordance with the guidelines and regulations provided and accepted by the Institutional Review Board of the Tri-Service General Hospital, National Defense Medical Center. In the glucose-treated cells of Fig. 2E, the cells were propagated in glucose free RPMI-1640 medium (catalog number: 11879020, Thermo Fisher Scientific, Waltham, MA) supplemented with 10-or 30-mM D-(+)-glucose (catalog number: G8644-100ML, Sigma, St. Louis, MO). In other studies, the cells were propagated in RPMI-1640 Medium, HEPES (catalog number: 22400089, Thermo Fisher Scientific, Waltham, MA), which contains 2 g/L glucose. All cells were cultured in medium with 10% heat-inactivated foetal bovine serum at 37 °C in a 5% CO 2 incubator. THP-1 cells were cultured for 48 h in RPMI-1640 medium supplemented with 100 nM phorbol 12-myristate 13-acetate to induce monocyte-to-macrophage differentiation.
Activation of the inflammasomes. For NLRP3 inflammasome, cells were primed for 4 h with 1 µg/ml LPS followed by treatment for 2 h with GlcN. Cells were then treated with 5 mM ATP or 10 µM nigericin for 0.5 h, or 100 µg/ml MSU and 250 mM palmitate-BSA for 24 h. For AIM2 and non-canonical inflammasome activation, cells were primed for 4 h with 1 µg/ml LPS or 1 µg/ml Pam3CSK4 (for non-canonical inflammasome) followed by treatment for 2 h with GlcN. Cells were then transfection with 2 µg/ml poly(dA/dT) or 2 µg/ml LPS for 6 h, respectively, by GeneJammer transfection reagent. The IL-1β levels in the supernatants were analysed by ELISA, and the levels of proIL-1β/IL-1β, IL-18, caspase-1 (p10 and p45) and ASC in the cell lysates or supernatants were analysed by Western blotting.
Western blotting analysis of protein expression in the culture medium and lysates. For detection of protein expression in the culture medium, a mixture of 300 μl culture medium, 300 μl methanol, and 125 μl www.nature.com/scientificreports www.nature.com/scientificreports/ chloroform was prepared. After vortexing, 300 μl double-distilled water was added to the mixture and incubated for 10 min on ice before centrifugation for 10 min at 13,000 rpm at 4 °C. The supernatant was removed, and 500 μl methanol was added; after vortexing, the mixture was centrifuged for 10 min at 13,000 rpm at 4 °C, and then, the supernatant was removed again. The pellet was dried at 55 °C and dissolved in Western blotting loading buffer, followed by incubation in boiling water for 30 min. The samples were analysed by Western blotting. For detection of protein expression in the lysates, ice-cold phosphate-buffered saline (PBS)-washed cells were lysed with 100 μl ice-cold lysis buffer (20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% NP-40, 1% sodium deoxycholate, 2.5 mM sodium pyrophosphate, 1 mM β-glycerolphosphate, 1 mM Na 3 VO 4 , 1 μg/ml leupeptin, and 1 mM PMSF) on ice for 10 min. Insoluble material was removed by centrifuging at 4 °C for 15 min at 12000 g. The protein concentrations were determined using Bio-Rad protein assay dye. 50 μg proteins form each sample was analysed by Western blotting. The bands intensities from each blot were quantified by densitometric analysis using ImageJ software. The densitometry fold change of each group was calculated by comparing the results with the control group. The band density is normalized to actin or indicated protein before fold change is calculated. ASC oligomerization assay. J774A.1 macrophages were primed for 4 h with 1 µg/ml LPS followed by incubation for 2 h with GlcN. Cells were incubated with 5 mM ATP for additional 0.5 h, and then lysed with TBS buffer (50 mM Tris-HCl (pH 7.4) and 150 mM NaCl) containing 0.5% Triton X-100, EDTA-free protease inhibitors and phosphatase inhibitors. The lysates were centrifuged at 6,000 g for 15 min at 4 °C, and the pellets and supernatants were used as the Triton-insoluble fractions and Triton-soluble fractions, respectively. For the detection of ASC oligomerization, the Triton-insoluble pellets were washed twice with TBS buffer and then were re-suspended in 300 μl TBS buffer. The re-suspended pellets were crosslinking for 30 min with 2 mM disuccinimidyl suberate at 37 °C and then were centrifuged for 15 min at 6,000 g. The pellets were dissolved in SDS sample buffer and processed for SDS-PAGE. The ASC oligomerization were then analysed by Western blotting. NF-κB transcriptional assay. J774A.1 macrophages were stably transfected with the pNiFty2-SEAP plasmid, which is a NF-κB-inducible reporter plasmid used for NF-κB transcriptional assay. Plate 2 × 10 5 cells per well in 0.5 ml of culture medium and were grown overnight at 37 °C in a 5% CO 2 incubator. The cells were incubated with GlcN (10, 30 mM) for 2 h and then treated with 1 μg/ml LPS for 24 h. The NF-κB transcriptional activity was analysed using QUANTI-Blue medium as described previously 53 . palmitate-BsA complex preparation. Palmitic acid was dissolved in 95% ethanol at 60 °C at a concentration of 300 mM and then was mixed with 5% fatty acid-free bovine serum albumin in PBS to a final concentration of 15 mM.
Mouse model of uric acid crystal-mediated peritonitis. Seven-to-nine-week-old male C57BL/6JNal mice were purchased from the National Laboratory Animal Center (Taipei, Taiwan) and housed in a room under controlled temperature (23 ± 3 °C) and relative humidity (50 ± 5%). Animal experiments were performed with the approval of the Institutional Animal Care and Use Committee of the National Ilan University (approval num- Statistical significance was determined by t tests (two-tailed) for two groups or ANOVA (with Dunnett's multiple comparisons test) for three or more groups. P values less than 0.05 were considered to be statistically significant.