Cutting Edge: G Subunit β 1 Negatively Regulates NLRP3 Inflammasome Activation

This information is current as Tomohiko Murakami, Lerdluck Ruengsinpinya, Eriko of September 26, 2021. Nakamura, Yoshifumi Takahata, Kenji Hata, Hiroaki Okae, Shun'ichiro Taniguchi, Masafumi Takahashi and Riko Nishimura J Immunol 2019; 202:1942-1947; Prepublished online 18

February 2019; Downloaded from doi: 10.4049/jimmunol.1801388 http://www.jimmunol.org/content/202/7/1942

Supplementary http://www.jimmunol.org/content/suppl/2019/02/15/jimmunol.180138 http://www.jimmunol.org/ Material 8.DCSupplemental References This article cites 20 articles, 3 of which you can access for free at: http://www.jimmunol.org/content/202/7/1942.full#ref-list-1

Why The JI? Submit online. by guest on September 26, 2021 • Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2019 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Cutting Edge: Subunit b 1 Negatively Regulates NLRP3 Inflammasome Activation Tomohiko Murakami,* Lerdluck Ruengsinpinya,* Eriko Nakamura,* † ‡ Yoshifumi Takahata,*x Kenji Hata,* Hiroaki Okae, Shun’ichiro Taniguchi, Masafumi Takahashi, and Riko Nishimura* The NLRP3 inflammasome has important roles in inflammasome consists of a sensor molecule, NLRP3, an the pathogenesis of various inflammatory diseases. adaptor protein, ASC, and caspase-1. Activation of the NLRP3 However, the regulatory mechanisms of the NLRP3 inflammasome has important roles in the pathogenesis of var- inflammasome are not fully understood. In this study, ious inflammatory diseases including autoimmune disorders, we attempted to identify molecules that interact with gout, type 2 diabetes, obesity, and Alzheimer disease (2). NLRP3 upon its activation. We identified G protein The NLRP3 inflammasome senses various types of exoge- Downloaded from subunit b 1(GNB1),adownstreammoleculeof nous and endogenous danger signals, including environmental G protein–coupled receptors (GPCRs), which regulates irritants, endogenous damage-associated molecular patterns, the NLRP3 inflammasome activation. GNB1 was phys- and organelle damage such as lysosome rupture, mitochondria ically associated with NLRP3 via the pyrin domain of damage, and endoplasmic reticulum stress (3). The NLRP3 inflammasome is activated by a two-step process (1–3). Signal 1,

NLRP3. Activation of the NLRP3 inflammasome was http://www.jimmunol.org/ also known as the priming signal for the NLRP3 inflamma- enhanced in GNB1-knockdown or GNB1-deficient some, is mediated by microbial ligands recognized by TLR. murine macrophages, although a lack of GNB1 did Signal 1 activates the NF-kB pathway, resulting in the in- not affect activation of the AIM2 inflammasome. ASC creased expression of NLRP3 and pro–IL-1b. Signal 2, also oligomerization induced by NLRP3 was enhanced by known as the activation signal for the NLRP3 inflammasome, GNB1 deficiency. Conversely, NLRP3-dependent ASC is mediated by pathogen-associated molecular patterns or oligomerization was inhibited by the overexpression damage-associated molecular patterns, and promotes the as- of GNB1. This study indicates that GNB1 negatively sembly of ASC and pro–caspase-1, leading to activation of regulates NLRP3 inflammasome activation by suppress- the NLRP3 inflammasome. Several cellular signals such as by guest on September 26, 2021 ing NLRP3-dependent ASC oligomerization, and K+ efflux, Ca2+ signaling, mitochondria dysfunction, and re- it provides a regulatory mechanism of the NLRP3 active oxygen species have been proposed as the proximal triggers inflammasome. The Journal of Immunology, 2019, for NLRP3 inflammasome activation (2–5). These proximal 202: 1942–1947. triggers have been reported to be mediated by at least P2X7R, pore forming toxins, Ca2+ channels, and G protein– coupled receptors (GPCRs) (3, 6, 7) in response to NLRP3 nflammasomes are multiprotein complexes that are re- inflammasome activators. Additionally, recent studies iden- sponsible for the activation of inflammatory responses tified NEK7 as a new critical regulator of the NLRP3 I and which promote maturation of the proinflammatory inflammasome (8). However, the regulatory mechanisms of cytokine IL-1b and cell death by activating caspase-1 (1). the NLRP3 inflammasome are still elusive. Several nucleotide-binding domain and leucine-rich repeat In this study, to further understand the regulatory mecha- (LRR)–containing receptor (NLR) family form nisms of the NLRP3 inflammasome, we searched for molecules inflammasomes in response to their specific stimulators. The that interact with NLRP3 in response to NLRP3 inflammasome NLR family pyrin domain (PYD)–containing 3 (NLRP3) activators and identified G protein subunit b 1(GNB1),a

*Department of Molecular and Cellular Biochemistry, Osaka University Graduate Address correspondence and reprint requests to Dr. Tomohiko Murakami, Department of School of Dentistry, Osaka 565-0871, Japan; †Department of Informative Genetics, Molecular and Cellular Biochemistry, Osaka University Graduate School of Dentistry, 1-8 Environment and Genome Research Center, Tohoku University Graduate School of Yamada-Oka, Suita, Osaka 565-0871, Japan. E-mail address: [email protected] Medicine, Sendai 980-8575, Japan; ‡Department of Comprehensive Cancer Therapy, x The online version of this article contains supplemental material. Shinshu University School of Medicine, Nagano 390-8621, Japan; and Division of Inflammation Research, Center for Molecular Medicine, Jichi Medical University, Abbreviations used in this article: BMDM, bone marrow–derived macrophage; DSS, Tochigi 329-0498, Japan disuccinimidyl suberate; GNB1, G protein subunit b 1; GNB1 shRNA, GNB1 short hairpin RNA; GPCR, G protein–coupled receptor; IP3, inositol trisphosphate; KO, ORCIDs: 0000-0003-4367-9343 (K.H.); 0000-0003-3245-6670 (S.T.); 0000-0003- knockout; LRR, leucine-rich repeat; MSU, monosodium urate; NLRP3, NLR family 2716-7532 (M.T.). pyrin domain–containing 3; NP-40, Nonidet P-40; PLA, proximity ligation assay; PLC, Received for publication October 17, 2018. Accepted for publication January 30, 2019. ; PYD, pyrin domain; shRNA, short hairpin RNA; WT, wild-type.

This work was supported by Japan Society for the Promotion of Science grants-in-aid for Ó scientific research, the Nakajima Foundation, the Naito Foundation, the Kanae Foun- Copyright 2019 by The American Association of Immunologists, Inc. 0022-1767/19/$37.50 dation for the Promotion of Medical Science, the Senri Life Science Foundation, and the Astellas Foundation for Research on Metabolic Disorders. www.jimmunol.org/cgi/doi/10.4049/jimmunol.1801388 The Journal of Immunology 1943 downstream molecule of GPCRs (9). We showed that GNB1 Plasmids and transfection negatively regulated NLRP3 inflammasome activation by inhib- Expression vectors for mouse GNB1 were purchased from GenScript. Mouse iting ASC oligomerization, which is necessary for full activation NLRP3, ASC, and AIM2 expression vectors were used as described previously of the NLRP3 inflammasome. Our findings reveal a regulatory (12). Deletion mutants of NLRP3 were generated by PCR and standard ligation procedures and were subcloned into pcDNA4 Myc-His (Invitrogen). mechanism of the NLRP3 inflammasome by G protein subunit. Cells were transfected with the indicated plasmids using X-tremeGENE 9 (Sigma-Aldrich). Materials and Methods Lentivirus infection Mice 2 2 Lentiviral pLKO.1 plasmids targeting Gnb1 (GNB1 short hairpin RNA C57BL/6J (B6) mice were obtained from Japan SLC. ASC-deficient (Asc / , 2/2 [GNB1 shRNA]) (TRCN0000034431) were purchased from Sigma-Aldrich. B6 background) and GNB1-deficient (Gnb1 , B6 background) mice were GFP shRNA was used as a negative control. Lentiviral particle preparation generated and genotyped as described previously (10, 11). All animal exper- and infection were performed according to the manufacturer’s protocol iments were approved by the Osaka University Institute Animal Experiment (Sigma-Aldrich). BMDMs were infected with control or GNB1 shRNA and Committee and performed in accordance with the regulatory guidelines. selected by 4 mg/ml puromycin in culture. Chemicals, Abs, and other reagents Immunoblotting and ELISA LPS (L3024), ATP (A6419), and calf thymus DNA (D3664) were purchased Immunoblotting analyses were performed as previously described (4). Im- from Sigma-Aldrich. Nigericin (145-07263) was purchased from Wako Pure munoblotting with anti–b-actin Ab was performed to assess the amount of Chemical. Monosodium urate (MSU; AG-CR1-3950) was purchased from protein in each sample. The ASC oligomerization assay was performed as

AdipoGen. FliC (ALX-522-058) was purchased from Enzo Life Sciences. described previously (13). Pellets from whole-cell lysates were cross-linked Downloaded from For immunoblotting, anti-NLRP3 (Cryo-2, 1:5000) and anti-ASC (AL177, with disuccinimidyl suberate (DSS; 21655; Thermo Fisher Scientific) and 1:1000) Abs were purchased from AdipoGen. Anti–caspase-1 (M-20, 1:250) analyzed by immunoblotting with anti-ASC Ab. Immunoblotting results Ab was purchased from Santa Cruz Biotechnology. Anti–IL-1b (AF-401-NA, are representative of at least three independent experiments. Quantitative 1:2000) Ab was purchased from R&D Systems. Anti-ERK (4696, 1:2000), analyses of band intensities were performed using ImageJ (National Institutes anti–p-ERK (9101, 1:2000), anti-AKT (4691, 1:1000), and anti–p-AKT of Health). For ELISA, culture supernatants were analyzed by ELISA kits for (9271, 1:1000) Abs were purchased from Cell Signaling Technology. Anti- IL-1b (88-7013-88; eBioscience), TNF-a (88-7324-88; eBioscience), and FLAG (M185-3, PM020, 1:1000) and anti-HA (M180-3, 1:1000) Abs were IL-18 (7625; MBL). Intracellular cAMP was measured by a cAMP Select purchased from MBL. Anti-GNB1 (ab137635, 1:1000) and anti-Myc ELISA kit (501040; Cayman Chemical). Phospholipase C (PLC) activity was http://www.jimmunol.org/ (ab9132, 1:1000) Abs were purchased from Abcam. Anti–b-actin (017- measured by an IP-one ELISA kit (72IP1PEA; Cisbio). 24573, 1:5000) Ab was purchased from Wako Pure Chemical. The amount of lactate dehydrogenase released in supernatants was determined by lactate In vitro pulldown assay dehydrogenase release assay kit (04744926001; Roche). A Duolink in situ proximity ligation assay (PLA) kit (DUO92101; Sigma-Aldrich) was used for His-tagged NLRP3 (aa 1–200) (His-PYD), ASC, and GNB1-FLAG were PLA. Coverslips were visualized with a fluorescence microscope (DM4B; separately overexpressed in HEK293 T cells. Cells transfected with His-PYD Leica). Cells with more than 20 dots were considered positive for PLA. were lysed in lysis buffer (50 mM sodium phosphate, 300 mM NaCl, 1% NP-40, protease inhibitor mixture). Cells transfected with ASC and Cells and cell stimulation GNB1-FLAG were lysed in lysis buffer (3.25 mM sodium phosphate, 70 mM NaCl, 1% NP-40, protease inhibitor mixture). His-PYD proteins were pulled HEK293 and HEK293 T cells were cultured in DMEM (Wako Pure down with Dynabeads TALON (10104D; Thermo Fisher Scientific) according by guest on September 26, 2021 Chemical) supplemented with 10% FCS and penicillin/streptomycin (Wako to the manufacturer’s protocol and then incubated with lysates containing ASC Pure Chemical). Bone marrow–derived macrophages (BMDMs) were pre- and/or GNB1-FLAG in vitro for 30 min at 4˚C. The beads were washed five pared as previously described (4). Fetal liver–derived macrophages were times with phosphate buffer (50 mM sodium phosphate, 300 mM NaCl, prepared from the liver of embryonic day–14.5 fetuses. Isolated fetal liver cells 0.01% Tween 20). Bound proteins were eluted with 300 mM imidazole in were cultured on petri dishes in IMDM (Wako Pure Chemical) supple- phosphate buffer followed by immunoblotting analysis. mented with 10% FCS, L-glutamine, 2-ME, penicillin/streptomycin, and M-CSF for 4–5 d. To activate the NLRP3 inflammasome, macrophages were Statistical analysis primed with LPS (10 ng/ml, 4 h) before stimulation with ATP (5 mM, Comparisons of two groups were analyzed using an unpaired two-tailed 30 min), nigericin (20 mM, 30 min), or MSU (100 mg/ml, 4 h). To activate , the AIM2 inflammasome, LPS-primed macrophages were transfected with calf Student t test. The p values 0.05 were considered significant. thymus DNA (2 mg/ml) using Lipofectamine 2000 (Thermo Fisher Scientific) 2 h before harvest. To activate the NLRC4 inflammasome, LPS-primed mac- Results and Discussion rophages were transfected with FliC (100 ng/ml) using Lipofectamine 2000 4 h GNB1 interacts with NLRP3 before harvest. To activate the noncanonical inflammasome, LPS-primed macrophages were transfected with ultrapure LPS (tlrl-smlps; InvivoGen) To identify regulators of the NLRP3 inflammasome, we (1 mg/ml) using DOTAP (Sigma-Aldrich) 4 h before harvest. attempted to identify proteins that interacted with NLRP3 in response to NLRP3 inflammasome activators. Because NLRP3 Immunoprecipitation and purification of NLRP3-associated proteins inflammasome activation quickly induces cell death and the 2 2 LPS-primed Asc / BMDMs were stimulated with ATP or nigericin. Cells were release of intercellular proteins to the extracellular space, it lysed in ice-cold Nonidet P-40 (NP-40) buffer (1% NP-40, 50 mM Tris-HCl, would be difficult to isolate proteins bound to NLRP3 during pH 7.4, 150 mM NaCl, protease inhibitor mixture and phosphatase inhibitor mixture [Wako Pure Chemical]) for 30 min at 4˚C. The cell lysates were col- its activation. To avoid this, we performed experiments using 2/2 lected and centrifuged at 12,000 3 g for 10 min at 4˚C. The supernatants were Asc BMDMs, which are resistant to cell death induced by 2 2 preincubated with Dynabeads Protein G (Thermo Fisher Scientific) for 1 h at 4˚C inflammasome activation (11). LPS-primed Asc / BMDMs and the beads were discarded. Then, the supernatants were incubated with anti- NLRP3 Ab overnight at 4˚C, followed by incubation with Dynabeads Protein were stimulated with ATP or nigericin to induce NLRP3 G for 1 h at 4˚C. The beads were washed five times with NP-40 buffer. Bound activation, and then the cell lysates were coimmunoprecipi- proteins were eluted with SDS sample buffer followed by mass spectrometry. tated with anti-NLRP3 Ab. The immunoprecipitated proteins were then subjected to SDS-PAGE and analyzed by mass Mass spectrometry spectrometry. We identified four proteins, GNB1, FLNA, TheNLRP3bindingfractionswereseparatedbySDS-PAGE,andeachlanewassliced RL30, and TXTP, which interacted with NLRP3 in response into several pieces. The proteins in each gel slice were subjected to tryptic digestion. to ATP or nigericin (Fig. 1A). Of these, we focused on GNB1 The tryptic digested peptides were analyzed by liquid chromatography-electrospray ionization-tandem mass spectrometry (LTQ Orbitrap Velos analysis plus electron- because it is a downstream molecule of GPCRs, which are transfer dissociation; Thermo Scientific), followed by shotgun proteomics analysis. involved in NLRP3 inflammasome activation (5, 14–16). 1944 CUTTING EDGE: GNB1 NEGATIVELY REGULATES NLRP3 INFLAMMASOME ACTIVATION Downloaded from

2 2 FIGURE 1. GNB1 associates with NLRP3. (A) Shotgun proteomics analysis of NLRP3 binding proteins in Asc / BMDMs. The numbers indicate proteins 2 2 detected in each fraction. (B) LPS-primed Asc / BMDMs were unstimulated or stimulated with ATP. Immunoprecipitated proteins with anti-NLRP3 or anti- IgG Abs and cell lysates were immunoblotted (IB) with the indicated Abs. IP, immunoprecipitation. (C) PLA of GNB1 and NLRP3 in LPS-primed BMDMs unstimulated or stimulated with ATP. PLA signals indicate interactions between GNB1 and NLRP3 (left panels). Red, PLA signals; blue, nuclei. Scale bar, 30 mm. Right panel shows quantitative analysis of PLA-positive cells. Data show the mean 6 SE (n = 3). *p , 0.05. (D) HEK293 T cells were cotransfected with http://www.jimmunol.org/ the indicated expression vectors. Immunoprecipitated proteins with anti-FLAG or anti-NLRP3 Abs and cell lysates were analyzed by immunoblotting with the indicated Abs. (E) Schematic representation of NLRP3 and deletion mutants (upper panel). Numbers represent amino acid residues of NLRP3. HEK293 T cells were cotransfected with NLRP3 deletion mutants together with GNB1-FLAG or empty vector (2). Immunoprecipitated proteins with anti-FLAG Ab and cell lysates were analyzed by immunoblotting with the indicated Abs.

GNB1 is a component of heterotrimeric G proteins, which GNB1 knockdown enhances activation of the NLRP3 inflammasome are composed of a, b, and g subunits and have roles in signal We next evaluated the role of GNB1 in NLRP3 inflammasome transduction downstream of GPCRs (9). Mutations in GNB1 activation. First, we performed knockdown experiments in cause severe neurodevelopmental disability, hypotonia, and BMDMs using the shRNA system. Knockdown of GNB1 by guest on September 26, 2021 seizures (10, 17). Additionally, some mutations in GNB1 have (GNB1 shRNA) strongly reduced the expression of GNB1 but also been reported as oncogenic mutations (18). Although GNB1 did not affect the expressions of NLRP3, pro–caspase-1, and is involved in these diseases, the role of GNB1 on inflammasomes ASC compared with the controls (Fig. 2A). Additionally, has not been reported. knockdown of GNB1 had no effect on the expressions of The interaction of NLRP3 with GNB1 was confirmed by NLRP3 and pro–IL-1b induced by LPS treatment, indicating coimmunoprecipitation experiments (Fig. 1B). A PLA, which that the priming signal was not affected by GNB1 knock- detects the interactions of proteins, demonstrated the inter- down. Interestingly, caspase-1 activation and IL-1b release action of GNB1 with NLRP3 in situ (Fig. 1C). Notably, the by ATP treatment were significantly enhanced by GNB1 interaction of NLRP3 with GNB1 was moderately ob- knockdown (Fig. 2A, 2B, Supplemental Fig. 1A). By contrast, served in LPS treatment and was enhanced upon NLPR3 caspase-1 activation and IL-1b release in response to calf inflammasome activation by ATP treatment (Fig. 1B, 1C). To thymus DNA, which activates the AIM2 inflammasome, were further confirm the interaction of NLRP3 with GNB1, we not affected by GNB1 knockdown (Fig. 2A, 2B, Supplemental expressed FLAG-tagged GNB1 with NLRP3 in HEK293 Fig. 1A). Consistent with these results, GNB1 knockdown T cells and performed coimmunoprecipitation experiments enhanced cytotoxicity by ATP treatment but not by DNA (Fig. 1D). The experiments supported the data shown in (Supplemental Fig. 1B). Of note, LPS-induced TNF-a pro- Fig. 1B. These results indicated that GNB1 interacted with duction was not affected by GNB1 knockdown (Fig. 2C). NLRP3 in response to NLRP3 activation. Furthermore, GNB1 knockdown significantly enhanced b GNB1 associates with NLRP3 via its PYD nigericin-induced caspase-1 activation and IL-1 release, as well as enhanced NLRP3 inflammasome activation induced To determine which regions within NLRP3 associate with bythelysosomemembranedamagingagentMSU(Fig.2D,2E, GNB1,weexpressedFLAG-taggedGNB1withdeletion Supplemental Fig. 1C). These results suggest that GNB1 func- mutants of NLRP3 in HEK293 T cells and performed a tions as a negative regulator of NLRP3 inflammasome activation coimmunoprecipitation assay (Fig. 1E). GNB1 interacted but has no effect on AIM2 inflammasome activation. with mutant NLRP3 lacking the NACHT domain and the carboxy-terminal LRR domain (aa 1–200) and with mutant NLRP3 lacking the LRR domain (aa 1–600), but not with GNB1 deficiency enhances activation of the NLRP3 inflammasome mutant NLRP3 lacking the N-terminal PYD (aa 201–1033) To confirm the negative regulatory role of GNB1 in NLRP3 2 2 (Fig. 1E). These results indicated that the PYD of NLRP3 was activation, we performed experiments using Gnb1 / mac- involved in the interaction with GNB1. rophages obtained from Gnb1-deficient mice. Because GNB1 The Journal of Immunology 1945

FIGURE 2. GNB1 knockdown enhances activation of the NLRP3 inflammasome. (A and B) BMDMs were treated with GFP shRNA (control) or GNB1 shRNA (shGNB1). The cells were unstimulated or stimulated with the indicated stimuli. Supernatants (Sup) and cell lysates were analyzed by immunoblotting with the indicated Abs (A). Casp-1 p10 indicates activation of caspase-1. IL-1b release in supernatants was analyzed by ELISA (B). (C) BMDMs treated with control or shGNB1 were unstimulated or stimulated with LPS. TNF-a release in supernatants was analyzed by ELISA. (D and E) BMDMs were treated with Downloaded from control or shGNB1. The cells were unstimulated or stimulated with the indicated stimuli. IL-1b release in supernatants was analyzed by ELISA (D). Sup and cell lysates were analyzed by immunoblotting with the indicated Abs (E). ELISA data show the mean 6 SE (n = 4). *p , 0.05. ns, not significant. deficiency leads to either embryonic lethality or the death of induced by nigericin and MSU were significantly en- pups within 2 d after birth (10), we harvested fetal liver cells hanced (Fig. 3D, 3E, Supplemental Fig. 2D). Additionally, containing hematopoietic stem cells from wild-type (WT) or caspase-1 activation and IL-1b release by cytosolic LPS, which 2 2 http://www.jimmunol.org/ Gnb1 / (knockout; KO) littermate embryos and induced activates the noncanonical NLRP3 inflammasome via caspase-11, their differentiation into macrophages using M-CSF. GNB1 were also significantly enhanced by GNB1 deficiency KO macrophages expressed normal amounts of NLRP3, pro– (Supplemental Fig. 2E, 2F) but not by FliC, which activates caspase-1, and ASC compared with WT cells (Fig. 3A). Ad- the NLRC4 inflammasome (Supplemental Fig. 2G, 2H). ditionally, GNB1 deficiency had no effect on the expressions Taken together, these results demonstrated that GNB1 neg- of NLRP3 and pro–IL-1b induced by LPS, indicating that atively regulates activation of the NLRP3 inflammasome but GNB1 deficiency did not affect the priming signal. Consistent not the AIM2 or NLRC4 inflammasomes. with results of the GNB1 knockdown experiments, caspase-1 activation and IL-1b release by ATP treatment were signifi- GNB1 is dispensable for the phosphorylation of ERK and AKT by guest on September 26, 2021 cantly enhanced by GNB1 deficiency (Fig. 3A, 3B, in macrophages Supplemental Fig. 2A). By contrast, caspase-1 activation and We next investigated the GNB1-mediated regulatory mech- IL-1b release in response to DNA were not impaired in anisms of the NLRP3 inflammasome. A previous study GNB1 KO cells (Fig. 3A, 3B, Supplemental Fig. 2A). Con- reported that GNB1 deficiency reduced ERK phosphorylation sistent with these results, GNB1 deficiency enhanced cyto- in neural progenitor cells (10). Another report described that toxicity and IL-18 release by ATP treatment but not by DNA mutant GNB1 proteins activated the PI3K–AKT–mTOR (Supplemental Fig. 2B, 2C). LPS-induced TNF-a production and MAPK pathways (18). Furthermore, the phosphorylation appeared intact in GNB1 KO cells (Fig. 3C). Furthermore, of ERK and AKT was reported to be involved in NLRP3 in GNB1 KO cells, caspase-1 activation and IL-1b release inflammasome activation (19, 20). Therefore, we determined

2 2 FIGURE 3. NLRP3 inflammasome activation in GNB1-deficient macrophages. (A and B) WT and Gnb1 / (KO) BMDMs were unstimulated or stimulated with the indicated stimuli. Supernatants (Sup) and cell lysates were analyzed by immunoblotting with the indicated Abs (A). IL-1b release in Sup was analyzed by ELISA (B). (C) WT and KO BMDMs were unstimulated or stimulated with LPS. TNF-a release in supernatants was analyzed by ELISA. (D and E) WT and KO BMDMs were unstimulated or stimulated with the indicated stimuli. IL-1b release in supernatants was analyzed by ELISA (D). Sup and cell lysates were analyzed by immunoblotting with the indicated Abs (E). ELISA data show the mean 6 SE (n = 4). *p , 0.05. ns, not significant. 1946 CUTTING EDGE: GNB1 NEGATIVELY REGULATES NLRP3 INFLAMMASOME ACTIVATION the phosphorylation status of ERK and AKT in WT and effect of GNB1 on NLRP3 inflammasome-dependent ASC GNB1 KO BMDMs (Fig. 4A, Supplemental Fig. 3A, 3B). oligomerization, we overexpressed NLRP3 and ASC com- LPS induced ERK and AKT phosphorylation in WT and plexes in HEK293 cells that lack both components. ASC GNB1 KO BMDMs. Additionally, ATP stimulation reduced oligomerization was enhanced by the overexpression of ERK and AKT phosphorylation in LPS-primed WT and NLRP3, and NLRP3-induced ASC oligomerization was GNB1 KO BMDMs. However, significant differences in ERK blocked by GNB1 overexpression (Fig. 4C, Supplemental and AKT phosphorylation were not observed between WT and Fig. 4B). Although ASC oligomerization was also enhanced GNB1 KO BMDMs. These results suggest that GNB1 is not by AIM2 overexpression, AIM2 inflammasome-dependent involved in ERK or AKT signaling during LPS-priming and ASC oligomerization was not affected by GNB1 over- NLRP3 inflammasome activation, at least in macrophages. expression (Fig. 4C, Supplemental Fig. 4B). Additionally, an in vitro pulldown assay showed that GNB1 suppressed the PLC signals and cAMP are not affected by GNB1 knockdown interaction of NLRP3 PYD with ASC (Supplemental Fig. 4C). GNB1 is a downstream molecule of GPCRs (9). Previous Thus, we concluded that GNB1 suppresses ASC oligomeriza- studies showed that GPCRs such as calcium-sensing receptor tion mediated by NLRP3, which in turn inhibits NLRP3 (CASR), GPCR family C group 6 member A (GPRC6A), inflammasome activation. dopamine receptor D1 (DRD1), and transmembrane GPCR-5 Because GPCRs were reported to be involved in activation (TGR5) regulate the NLRP3 inflammasome, mediated by and regulation of the NLRP3 inflammasome (5, 14–16), PLC–inositol trisphosphate (IP3)–intracellular Ca2+ and/or GPCR signaling might be an important regulator of the Downloaded from cellular cAMP signaling cascades (5, 14–16). Therefore, we NLRP3 inflammasome (7). However, a detailed mechanism measured PLC activity and cellular cAMP levels in GNB1- of NLRP3 inflammasome activation mediated by GPCR knockdown BMDMs during NLRP3 inflammasome activa- signaling remains unclear. In this study, we identified GNB1 tion by performing inositol monophosphate (IP-One) ELISA as a NLRP3-binding protein. GNB1 associated with the PYD and cAMP ELISA assays, respectively (5). However, there was of NLRP3, which is essential for its interaction with ASC. We no significant change in the PLC and cAMP levels between further showed that GNB1 knockdown or GNB1 deficiency http://www.jimmunol.org/ controls and GNB1 knockdown BMDMs (Supplemental enhanced NLRP3 inflammasome activation, but not AIM2 Fig. 3C, 3D), indicating that PLC-IP3-Ca2+ and cAMP sig- and NLRC4 inflammasome activation, demonstrating GNB1 naling pathways are not involved in regulation of the NLRP3 is a specific negative regulator of NLRP3 inflammasome ac- inflammasome by GNB1. tivation. GNB1 knockdown or deficiency did not affect PLC– IP3–Ca2+, cAMP, or the phosphorylation of ERK and AKT GNB1 negatively regulates ASC oligomerization induced by NLRP3 signaling cascades in macrophages. However, NLRP3- To explore other possibilities for regulation of the NLRP3 induced ASC oligomerization was enhanced by GNB1 defi- inflammasome by GNB1, we considered the binding domain ciency and was blocked by the overexpression of GNB1. Taken by guest on September 26, 2021 of NLRP3 with GNB1. The PYD of NLRP3 is responsible for together, our data suggest that GNB1 suppresses ASC oligo- its interaction with ASC and subsequent ASC oligomeriza- merization, presumably through its interaction with NLRP3, tion, a critical step for caspase-1 activation (1). Because we and consequently negatively regulates NLRP3 inflammasome found that NLRP3 bound to GNB1 through the PYD activation. (Fig.1E),wehypothesizedthatGNB1isinvolvedin Our findings demonstrate a role for the G protein subunit in NLRP3-induced ASC oligomerization. We tested this the inhibition of the NLRP3 inflammasome. GPCRs are the by assessing ASC oligomerization levels in WT and GNB1 largest and most diverse group of receptors and transduce many KO macrophages. Importantly, in GNB1 KO cells, ATP- extracellular stimuli, such as neurotransmitters, chemokines, induced ASC oligomerization was clearly enhanced, whereas ions, peptides, lipids, nucleosides, light, and temperature, into exogenous DNA-induced ASC oligomerization was not intracellular signals (7, 9). Because most GPCR signals are affected (Fig. 4B, Supplemental Fig. 4A). To confirm the independent of NLRP3 inflammasome regulation, GNB1

2 2 FIGURE 4. GNB1 negatively modulates ASC oligomerization induced by NLRP3. (A) WT and Gnb1 / (KO) BMDMs were unstimulated or stimulated with the indicated stimuli. Cell lysates were analyzed by immunoblotting with the indicated Abs. (B) WT and KO BMDMs were unstimulated or stimulated with the indicated stimuli. Pellets from cell lysates were cross-linked with DSS and analyzed by immunoblotting with anti-ASC Ab (top panel). Total cell lysates were immunoblotted with the indicated Abs. (C) HEK293 cells were cotransfected with the indicated plasmids. Control indicates empty vector. Pellets from cell lysates were cross-linked with DSS and analyzed by immunoblotting with anti-ASC Ab (top panel). Total cell lysates were immunoblotted with the indicated Abs. The Journal of Immunology 1947 may contribute to the inhibition of unnecessary NLRP3 ac- 7. Tang, T., T. Gong, W. Jiang, and R. Zhou. 2018. GPCRs in NLRP3 inflamma- some activation, regulation, and therapeutics. Trends Pharmacol. Sci. 39: 798–811. tivation during GPCR activation. Further studies would re- 8. He, Y., M. Y. Zeng, D. Yang, B. Motro, and G. Nu´n˜ez. 2016. NEK7 is an es- veal the regulatory mechanism of the NLRP3 inflammasome sential mediator of NLRP3 activation downstream of potassium efflux. Nature 530: 354–357. mediated by GPCR signals. 9. Oldham, W. M., and H. E. Hamm. 2008. activation by In this study, we identified GNB1, FLNA, RL30, and TXTP G-protein-coupled receptors. Nat. Rev. Mol. Cell Biol. 9: 60–71. as proteins that interacted with NLRP3. However, we analyzed 10. Okae, H., and Y. Iwakura. 2010. Neural tube defects and impaired neural pro- genitor cell proliferation in Gbeta1-deficient mice. Dev. Dyn. 239: 1089–1101. only GNB1, not FLNA, RL30, or TXTP. Therefore, we 11. Yamamoto, M., K. Yaginuma, H. Tsutsui, J. Sagara, X. Guan, E. Seki, K. Yasuda, cannot rule out FLNA, RL30, or TXTP as potential regulators M. Yamamoto, S. Akira, K. Nakanishi, et al. 2004. ASC is essential for LPS-induced activation of procaspase-1 independently of TLR-associated signal adaptor molecules. of the NLRP3 inflammasome. Cells 9: 1055–1067. In conclusion, we identified GNB1 as a negative regulator 12. Yu, J., H. Nagasu, T. Murakami, H. Hoang, L. Broderick, H. M. Hoffman, and T. Horng. 2014. Inflammasome activation leads to Caspase-1-dependent mito- of the NLRP3 inflammasome. Our findings uncovered the role chondrial damage and block of mitophagy. Proc.Natl.Acad.Sci.USA111: of GNB1 in NLRP3 inflammasome activation and propose a 15514–15519. novel regulatory mechanism of the NLRP3 inflammasome. 13. Fernandes-Alnemri, T., J. Wu, J. W. Yu, P. Datta, B. Miller, W. Jankowski, S. Rosenberg, J. Zhang, and E. S. Alnemri. 2007. The pyroptosome: a supramo- lecular assembly of ASC dimers mediating inflammatory cell death via caspase-1 activation. Cell Death Differ. 14: 1590–1604. Acknowledgments 14. Rossol, M., M. Pierer, N. Raulien, D. Quandt, U. Meusch, K. Rothe, K. Schubert, We thank Dr. Tiffany Horng (Harvard T.H. Chan School of Public Health) T. Scho¨neberg, M. Schaefer, U. Kru¨gel, et al. 2012. Extracellular Ca2+ is a danger for providing inflammasome related materials, Dr. Kazunobu Saito (Research In- signal activating the NLRP3 inflammasome through G protein-coupled calcium sensing receptors. Nat. Commun. 3: 1329. stitute for Microbial Diseases, Osaka University) for mass spectrometry, and Downloaded from +/2 15. Yan, Y., W. Jiang, L. Liu, X. Wang, C. Ding, Z. Tian, and R. Zhou. 2015. Do- Dr. Yoichiro Iwakura (Tokyo University of Science) for providing Gnb1 mice. pamine controls systemic inflammation through inhibition of NLRP3 inflamma- some. Cell 160: 62–73. 16. Guo, C., S. Xie, Z. Chi, J. Zhang, Y. Liu, L. Zhang, M. Zheng, X. Zhang, D. Xia, Disclosures Y. Ke, et al. 2016. Bile acids control inflammation and metabolic disorder through The authors have no financial conflicts of interest. inhibition of NLRP3 inflammasome. [Published erratum appears in 2016 Immunity 45: 944.] Immunity 45: 802–816. 17. Petrovski, S., S. Ku¨ry, C. T. Myers, K. Anyane-Yeboa, B. Cogne´, M. Bialer, F. Xia, P. Hemati, J. Riviello, M. Mehaffey, et al; University of Washington Center for http://www.jimmunol.org/ References Mendelian Genomics. 2016. Germline de novo mutations in GNB1 cause severe 1. Schroder, K., and J. Tschopp. 2010. The inflammasomes. Cell 140: 821–832. neurodevelopmental disability, hypotonia, and seizures. Am. J. Hum. Genet. 98: 2. Menu, P., and J. E. Vince. 2011. The NLRP3 inflammasome in health and disease: 1001–1010. the good, the bad and the ugly. Clin. Exp. Immunol. 166: 1–15. 18.Yoda,A.,G.Adelmant,J.Tamburini,B.Chapuy,N.Shindoh,Y.Yoda, 3. Latz, E., T. S. Xiao, and A. Stutz. 2013. Activation and regulation of the inflam- O.Weigert,N.Kopp,S.C.Wu,S.S.Kim,etal.2015.MutationsinGproteinb masomes. Nat. Rev. Immunol. 13: 397–411. subunits promote transformation and kinase inhibitor resistance. Nat. Med. 21: 4. Murakami, T., J. Ockinger, J. Yu, V. Byles, A. McColl, A. M. Hofer, and T. Horng. 71–75. 2012. Critical role for calcium mobilization in activation of the NLRP3 inflam- 19. Ghonime, M. G., O. R. Shamaa, S. Das, R. A. Eldomany, T. Fernandes-Alnemri, masome. Proc. Natl. Acad. Sci. USA 109: 11282–11287. E. S. Alnemri, M. A. Gavrilin, and M. D. Wewers. 2014. Inflammasome priming 5. Lee, G. S., N. Subramanian, A. I. Kim, I. Aksentijevich, R. Goldbach-Mansky, by lipopolysaccharide is dependent upon ERK signaling and proteasome function. D. B. Sacks, R. N. Germain, D. L. Kastner, and J. J. Chae. 2012. The calcium- J. Immunol. 192: 3881–3888. by guest on September 26, 2021 sensing receptor regulates the NLRP3 inflammasome through Ca2+ and cAMP. 20. Liao, P. C., L. K. Chao, J. C. Chou, W. C. Dong, C. N. Lin, C. Y. Lin, A. Chen, Nature 492: 123–127. S. M. Ka, C. L. Ho, and K. F. Hua. 2013. Lipopolysaccharide/adenosine 6. Horng, T. 2014. Calcium signaling and mitochondrial destabilization in the trig- triphosphate-mediated in the regulation of NLRP3 protein ex- gering of the NLRP3 inflammasome. Trends Immunol. 35: 253–261. pression and caspase-1-mediated interleukin-1b secretion. Inflamm. Res. 62: 89–96. (n=3). *p<0.05. ns: not significant. not ns: *p<0.05. (n=3). Quantitative analysis of Casp- of analysis Quantitative LDH release in supernatants wasbyanalyzed LDH assay kit. release LDHindicates cytotoxicity. (C) after stimulation. collected Supernatants were stimuli. indicated the with stimulated or unstimulated byselected and (shGNB1) shRNA GNB1 or (control) shRNAGFP with Casp- of analysis Quantitative (A) BMDMs. knockdown caspase- of Quantitative analysis 1. Figure Supplemental A C

β Casp-1 (p10)/β-actin Casp-1 (p10)/ -actin (fold change) (fold change) 0 1 2 3 4 5 0 1 2 3

control control - - shGNB1 shGNB1 LPS

LPS+Nigericin control

control

1 (p10)/ shGNB1 * LPS+ATP

1 (p10)/ control

shGNB1 * β -

LPS+MSU shGNB1 actin in Fig. 2E using ImageJ. Data show the mean ± mean the showImageJ. Data using 2E Fig. actin in LPS+DNA β control - actin in Fig. 2A using ImageJ. (B) BMDMs were treated treated were BMDMs ImageJ. (B) using 2A Fig. actin in * control ns

shGNB1 shGNB1 B

LDH release (%) 1 and LDH release in GNB1 GNB1 in release LDH and 1 20 40 60 80 0

control ns - . The cells were cells The . puromycin shGNB1

LPS control ns

shGNB1 LPS+ATP control *

shGNB1 LPS+DNA SE

control ns

shGNB1 A B 4 80

- actin * * β 3 60 ns 2 40 ns ns ns 1 (p10)/ - 1 1 20 (fold change) (fold

0 LDH (%) release

Casp 0 WT KO WT KO WT KO WT KO WT KO WT KO WT KO WT KO - LPS LPS+ATP LPS+DNA - LPS LPS+ATP LPS+DNA

C D 4000 3 * *

3000 - actin β 2

pg /ml) ns 2000 *

- 18 ( 1 1000 IL 1 (p10)/ - 1 0 change) (fold 0

WT KO WT KO WT KO WT KO Casp WT KO WT KO WT KO - LPS LPS+ATP LPS+DNA - LPS+Nigericin LPS+MSU

E F 2500 * 2000 DOTAP- 1500 DOTAP LPS (pg/ml)

β β 1000 WT KO WT KO - 1 500

IL IB: α-Casp1 (p10) 0 Sup WT KO WT KO IB: α-IL-1β (mature) DOTAP DOTAP- Lysates IB: α-β-actin LPS

G H LPS+ 10000 ns - FliC 8000 WT KO WT KO 6000 IB: α-Casp1 (p10) (pg/ml) Sup β β 4000 IB: α-IL-1β (mature) - 1

IL 2000 α β 0 Lysates IB: - -actin WT KO WT KO - LPS+FliC

Supplemental Figure 2. Inflammasome activation in GNB1 deficient BMDMs. (A) Quantitative analysis of Casp-1 (p10)/β-actin in Fig. 3A using ImageJ. (B, C) WT and Gnb1−/− (KO) BMDMs were unstimulated or stimulated with the indicated stimuli. LDH release in supernatants was analyzed by LDH assay kit (B). IL-18 release in supernatants was analyzed by ELISA (C). (D) Quantitative analysis of Casp-1 (p10)/β-actin in Fig. 3E using ImageJ. (E, F) LPS-primed WT and GNB1 KO BMDMs were stimulated with the indicated stimuli. IL-1β release in supernatants was analyzed by ELISA (E). Supernatants (Sup) and cell lysates were analyzed by immunoblotting with the indicated antibodies (F). DOTAP-LPS indicates LPS transfection using DOTAP. (G, H) WT and GNB1 KO BMDMs were unstimulated or stimulated with the indicated stimuli. IL-1β release in supernatants was analyzed by ELISA (G). Supernatants (Sup) and cell lysates were analyzed by immunoblotting with the indicated antibodies (H). Data show the mean ± SE (n=3). *p<0.05. ns: not significant. A B

ns ns 4 4 3 3 /AKT /ERK 2 ns 2 ns ns ns pAKT 1

1 change ) (fold pERK (fold change) (fold 0 0 WT KO WT KO WT KO WT KO WT KO WT KO - LPS LPS+ATP - LPS LPS+ATP

C D 100 ns ns 20 ) 80 /ml) nM 15 60 ns ns ns

40 ( pmol 10 ns - one (

IP 20 5

0 cAMP 0 control control control shGNB1 shGNB1 shGNB1 control control control shGNB1 shGNB1 shGNB1 - LPS LPS+ATP - LPS LPS+ATP

Supplemental Figure 3. GNB1 is not involved in the phosphorylation of ERK and AKT, PLC signals, and cAMP production in macrophages. (A, B) Quantitative analysis of pERK/ERK (A) and pAKT/AKT (B) in Fig. 4A using ImageJ. Data show the mean ± SE (n=3). (C) BMDMs were treated with GFP shRNA (control) or GNB1 shRNA (shGNB1) and selected by puromycin. The cells were unstimulated or stimulated with the indicated stimuli. Cell lysates were collected after stimulation. IP-one levels in cell lysates were analyzed by an IP-one kit. IP-one level indicates PLC activity. Data show the mean ± SE (n=4). (D) BMDMs were treated with GFP shRNA (control) or GNB1 shRNA (shGNB1) and selected by puromycin. The cells were unstimulated or stimulated with the indicated stimuli. Cell lysates were collected after stimulation. Cyclic AMP (cAMP) levels in cell lysates were analyzed by a cAMP kit. Data show the mean ± SE (n=4). ns: not significant. A B ns 3 2 * * ns 1.5 2 ns 1 1 0.5 (fold change) (fold change) (fold 0 0 - GNB1 - GNB1 - GNB1 WT KO WT KO WT KO ASC ASC dimer/ASCin lysates ASC dimer/ASCin lysates control NLRP3 AIM2 - LPS+ATP LPS+DNA

C

His-PYD His α GNB1-FLAG - + - + + IB: -NLRP3 Input ASC - - + + + IB: α-β-actin IB: α-FLAG GNB1-FLAG - + - + Pulldown IB: α-ASC ASC - - + + IB: α-NLRP3 IB: α-FLAG Input IB: α-ASC IB: α-β-actin

Supplemental Figure 4. Quantitative analysis of ASC dimer and in vitro pulldown assay. (A, B) Quantitative analysis of ASC dimer/ASC in lysates of Fig. 4B (A) and Fig. 4C (B) using ImageJ. Data show the mean ± SE (n=3). *p<0.05. ns: not significant. (C) His tagged NLRP3 (aa 1-200) (His-PYD), ASC and GNB1-FLAG were separately overexpressed in HEK293T cells. Cell lysates were harvested 48 h after transfection. His-PYD proteins were pulled down with Dynabeads TALON and then incubated with ASC or/and GNB1-FLAG in vitro. Pulldown and input proteins were analyzed by immunoblotting with the indicated antibodies.