Palmitoylation of NLRP3 Modulates Inflammasome Activation and Inflammatory Bowel Disease Development

Key Points

• NLRP3 is palmitoylated on Cys⁴¹⁹ during inflammasome activation.

• ZDHHC17 is the predominant enzyme mediating NLRP3 palmitoylation.

• 2-Bromopalmitate treatment can rescue mice from dextran sulfate sodium–induced colitis.

Abstract

Aberrant activity of NLRP3 has been shown associations with severe diseases. Palmitoylation is a kind of protein post-translational modification, which has been shown to regulate cancer development and the innate immune system. Here, we showed that NLRP3 is palmitoylated at Cys⁴¹⁹ and that palmitoyltransferase ZDHHC17 is the predominant enzyme that mediates NLRP3 palmitoylation and promotes NLRP3 activation by interacting with NLRP3 and facilitating NIMA-related kinase 7 (NEK7)–NLRP3 interactions. Blockade of NLRP3 palmitoylation by a palmitoylation inhibitor, 2-bromopalmitate, effectively inhibited NLRP3 activation in vitro. Also, in a dextran sulfate sodium–induced colitis model in mice, 2-bromopalmitate application could attenuate weight loss, improve the survival rate, and rescue pathological changes in the colon of mice. Overall, our study reveals that palmitoylation of NLPR3 modulates inflammasome activation and inflammatory bowel disease development. We propose that drugs targeting NLRP3 palmitoylation could be promising candidates in the treatment of NLRP3-mediated inflammatory diseases.

Introduction

Pattern recognition receptors constitute a powerful defense line in innate immunity. They recognize pathogen-associated molecular patterns and danger-associated molecular patterns and initiate the immune response (1). According to their structure and function, pattern recognition receptors are divided mainly into TLRs (1), RIG-I-like receptors (2), NOD-like receptors (NLRs) (3), and C-type lectin receptors (4). NACHT, LRR, and PYD domain-containing protein 3 (NLRP3) is the most important among NLRs because it “senses” various “danger signals” from exogenesis and endogenesis (5). NLRP3 activation comprises two steps. Priming through cytokines or TLRs and NLRs results in abundant transcription and translation of NLRP3 and pro–IL-1β. Under stimulation by toxins, ATP, or particulate matter, NLRP3 cooperates with an adaptor protein (apoptosis-associated speck-like protein [ASC]) and an effecter protein (procaspase-1) to form a large inflammasome complex. Assembly of the NLRP3 inflammasome involves cleavage of procaspase-1 into its active forms, which induce maturation of the proinflammatory cytokines IL-1β and IL-18 (6) and cleavage of gasdermin D to trigger cytokine release and pyroptosis (7, 8).

Aberrant activity of NLRP3 has been shown to be associated with neurodegenerative (9), autoimmune (10), autoinflammatory (11), and metabolic diseases (12) and cancers (12). Hence, exploring the specific mechanism of the activation and regulation of NLRP3 is important. The mechanism of NLRP3 activation is complex and indistinct, including K⁺ efflux (13), Ca²⁺ signaling (14), mitochondrial dysfunction (15), reactive oxygen species production (16), lysosomal leakage (17), and trans-Golgi disassembly (18). In addition, recent studies have emphasized the crucial role of NIMA-related kinase 7 (NEK7) in activation of the NLRP3 inflammasome (19–21). Moreover, multiple protein post-translational modifications have been shown to have important roles in regulation of the NLRP3 inflammasome, such as ubiquitylation (22, 23), phosphorylation (24–26), small ubiquitin-like modifier (SUMO)ylation (27, 28), and acetylation (29).

Protein palmitoylation is a type of reversible post-translational modification in which a 16-carbon fatty-acid palmitate attaches to the cysteine residue of a target protein (30). Palmitoylation catalyzed by a series of enzymes that contain a conserved zinc finger Asp–His–His–Cys (DHHC) motif contributes to the subcellular localization, stability, and functions of a protein (31). Protein palmitoylation has been shown to have a crucial role in immune regulation. For instance, palmitoylation of NOD-like receptor 1 (NOD1) and NOD2 by ZDHHC5 is essential for its membrane localization and bacterial sensing (32). Programmed-death ligand 1 (PD-L1) is palmitoylated by ZDHHC3 to prevent its degradation by lysosomes, and inhibiting PD-L1 palmitoylation serves as antitumor immunotherapy (33). Palmitoylation of stimulator of IFN genes (STING) at Cys^88/91 is crucial for downstream ligand binding and a type-I IFN response (34).

Inflammatory bowel disease (IBD) is an intestinal disorder disease that includes ulcerative colitis and Crohn’s disease and that is chronic, recurrent, and incurable. There are millions of patients suffering from IBD, and the number of cases increases dramatically worldwide (35). Although the causes of IBD are complex, including genetic factors, gut microbial factors, environment factors, and immunological abnormalities (36), aberrant activation of NLRP3 inflammasome has been shown to contribute to IBD pathogenesis (37–39).

Recent studies have reported that NLRP3 undergoes palmitoylation (40–42), and targeting NLRP3 palmitoylation seems to be a promising strategy. However, the process and effect of NLRP3 palmitoylation is controversial and unclear. In our study, we propose that NLRP3 palmitoylation mediated by ZDHHC17 modulates inflammasome activation. Inhibition of NLRP3 palmitoylation by the palmitoylation inhibitor 2-bromopalmitate (2-BP) effectively attenuates inflammasome activation in vitro and the severity of dextran sulfate sodium (DSS)–induced colitis in mice.

Materials and Methods

Cell lines and culture

Human embryonic kidney cells (HEK293T) were purchased from American Type Culture Collection (Manassas, VA). A human acute monocytic leukemia cell line (THP-1) was a gift from Dr. Jun Cui (MOE Key Laboratory of Gene Function and Regulation, School of Life Sciences, Sun Yat-sen University, Guangzhou, China). NLRP3-deficient THP-1 (THP-1–def) cells were a gift from Dr. Fushan Shi (Department of Veterinary Medicine, College of Animal Sciences, Zhejiang University, Hangzhou, China). HEK29T cells were cultured in DMEM supplemented with 10% FBS, penicillin (100 U/ml), and streptomycin sulfate (100 μg/ml). THP-1 and THP-1–def cells were cultured in RPMI 1640 medium supplemented with 10% inactive FBS, penicillin (100 U/ml), and streptomycin sulfate (100 μg/ml). Bone marrow–derived macrophages (BMDMs) were isolated from 7-wk-old C57BL/6 mice and cultured in RPMI 1640 medium with 10% inactive FBS, penicillin (100 U/ml), streptomycin sulfate (100 μg/ml), and 10% granulocyte macrophage CSF-conditioned medium from L929 cells for 6 d. The medium was changed every 2 d.

Reagents and antibodies

Phorbol 12-myristate 13-acetate (catalog number: S7791) was purchased from Selleck Chemicals (Houston, TX). LPS (tlrl-3pelps), nigericin (tlrl-nig), ATP (tlrl-atpl), alum crystals (tlrl-aloh), and monosodium urate (MSU) (tlrl-msu) were obtained from InvivoGen (San Diego, CA). N-Ethylmaleimide (E3876), hydroxylamine hydrochloride (255580), and 2-BP (21604) were sourced from MilliporeSigma (Burlington, MA). BMCC-biotin (21900) was purchased from Thermo Fisher Scientific (Waltham, MA). Streptavidin-HRP (M00091) was obtained from Genscript (Piscataway, NJ). Abs against hemagglutinin (HA) (H6908) and Flag (F3165) were purchased from MilliporeSigma. NLRP3 (D4D8T) rabbit mAb, IL-1β (D3U3E) rabbit mAb, IL-1β (3A6) mouse mAb, caspase-1 (E9R2D) rabbit mAb, cleaved caspase-1 (Asp²⁹⁷) (D57A2) rabbit mAb, and GM130 (D6B1) rabbit mAb were purchased from Cell Signaling Technology (Danvers, MA). Anti-NLRP3/NALP3 mAb (Cryo-2) (AG-20B-0014-C100) was obtained from AdipoGen Life Sciences (San Diego, CA). ASC mouse mAb (sc-271054) was sourced from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-ZDHHC17 Ab (ab154054) and anti-NEK7 Ab (ab133514) were purchased from Abcam (Cambridge, UK). Ab against GAPDH (GTX100118) was obtained from GeneTex (Irvine, CA). FITC-conjugated Affinipure goat anti-mouse Ig (Ig)G(H + L) (SA00003-1) and Cy3-conjugated Affinipure goat anti-rabbit IgG(H + L) (SA00009-2) were from Proteintech (Rosemont, IL).

Plasmid construction and primers

The expression plasmids of Flag-tagged ZDHHC1–ZDHHC24 were purchased from WZ Biosciences (Columbia, MD). ZDHHC17 was subcloned into the pcaggs-HA vector and pLenti CMV 3× Flag vector with the ClonExpress II one-step cloning kit (Vazyme, Nanjing, China) according to manufacturer instructions. Expression plasmids of NLRP3/m-NLRP3/ASC/caspase-1/IL-1β were constructed as stated previously by our research team. Small interfering RNA–targeting mouse zdhhc17 was designed and synthesized by HippoBio: si-zdhhc17-1, AAGGGUGAGUCAGCACUUGAU; si-zdhhc17-2, AAGCUGAAGGCUGACAAGGAA; and si-zdhhc17-3, UGGCCUCUUUCGUCCUGUUAU. The following are the main primers used in our study: HA-ZD17-F, 5′-catcattttggcaaagaattcATGCAGCGGGAGGAGGGA-3′; HA-ZD17-R, 5′-tgcatcgatgagctcgaattcCACCAGCTGGTACCCAGATCC-3′; HA-ZD18-F, 5′-catcattttggcaaagaattcATGAAGGACTGCGAGTACCAGC-3′; HA-ZD18-R, 5′-tgcatcgatgagctcgaattcGGGGTGGCCTCCTACCATG-3′; ZDHHS17-F, 5′-GATCATCATTCCCCATGGGTGGGTAACTGTGTAGG-3′; ZDHHS17-R, 5′-CACCCATGGGGAATGATGATCAAATTTTGCTATAC-3′; Lenti-ZDHHC/S17-F, 5′-ctagatatcttcgaaggatccATGCAGCGGGAGGAGGGA-3′; Lenti-ZDHHC/S17-R, 5′-tttgtagtcagcccgggatccCACCAGCTGGTACCCAGATCC-3′; N3-C8S-F, 5′-ACCCGCTCCAAGCTGGCCAGGTACCTGGAGGACCT-3′; N3-C8S-R, 5′-CTGGCCAGCTTGGAGCGGGTGCTTGCCATCTTCAT-3′; N3-C130S-F, 5′-ATCTCTATTTCTAAAATGAAGAAAGATTACCGTAA-3′; N3-C130S-R, 5′-TCTTCATTTTAGAAATAGAGATTCTCGAAAGGTAC-3′; N3-C150S-F, 5′-AGATTCCAGTCCATTGAAGACAGGAATGCCCGTCT-3′; N3-C150S-R, 5′-TGTCTTCAATGGACTGGAATCTGCTTCTCACGTAC-3′; N3-C279S-F, 5′-ATCATGAGCTCCTGCCCCGACCCAAACCCACCCAT-3′; N3-C279S-R, 5′-GGTCGGGGCAGGAGCTCATGATCAGGTCCCCCAGG-3′; N3-C419S-F, 5′-GGATCGTGTCCACTGGACTGAAACAGCAGATGGAG-3′; N3-C419S-R, 5′-CAGTCCAGTGGACACGATCCAGCAGACCAGGGGGA-3′; N3-C673S-F, 5′-ATTGAGAACTCTCATCGGGTGGAGTCACTGTCCCT-3′; N3-C673S-R, 5′-CACCCGATGAGAGTTCTCAATGCAAAAGGAAGAAA-3′; N3-C837S-F, 5′-TTGGTCAGCTCCTGCCTCACATCAGCATGTTGTCA-3′; N3-C837S-R, 5′-ATGTGAGGCAGGAGCTGACCAACCAGAGCTTCTTC-3′; N3-C838S-F, 5′-GTCAGCTGCTCCCTCACATCAGCATGTTGTCAGGATCTTG-3′; N3-C838S-R, 5′-CTGATGTGAGGGAGCAGCTGACCAACCAGAGCTTCTTCAG-3′; N3-C933S-F, 5′-AAACTACTCTCTGAGGGACTCTTGCACCCCGACTG-3′; and N3-C933S-R, 5′-AAGAGTCCCTCAGAGAGTAGTTTGATCCCCTTGTC-3′.

Immunoblotting and immunoprecipitation

The collected cells were lysed with lysis buffer (150 mM NaCl, 50 mM Tris-HCl, 5 mM EDTA, 10% glycerol, 1% Triton X-100, and 1% protease-inhibitor mixture) for 45 min at 4°C. Then, lysed cells were centrifuged at 13,000g for 10 min at room temperature. The supernatants were used for SDS-PAGE. For immunoprecipitation, the supernatants were incubated with IgG or the indicated Abs overnight at 4°C. Then, protein A/G–agarose (Pierce) was incubated with lysates for an additional 2 h at 4°C. Precipitates were washed thrice with lysis buffer, boiled with loading buffer, and used for SDS-PAGE.

Acyl–Biotin Exchange Assay for palmitoylation

The acyl–biotin exchange (ABE) assay for palmitoylation was undertaken as described previously (32). Briefly, the cells were lysed by lysis buffer with N-ethylmaleimide (50 μM) to block active sulfhydryl groups. Target proteins were immunoprecipitated with the indicated Ab. After that, beads were incubated with lysis buffer containing 1 M hydroxylamine hydrochloride for 1 h at room temperature and washed five times with lysis buffer (pH 6.2). Finally, the beads were incubated with BMCC-biotin–containing lysis buffer (5 μM; pH 6.2) for 2 h at room temperature and washed five times. The immunoprecipitates were analyzed by SDS-PAGE and detected by the indicated Ab. The palmitoylation intensity of NLRP3 was measured by ImageJ with three independent studies.

Click chemistry

The ABE assay for palmitoylation was undertaken as described previously (43). The cells were labeled with 50 μM palmitic acid analog (HY-W040304, MedChemExpress, China) for 6 h, collected, and lysed with lysis buffer. Target proteins were immunoprecipitated with the indicated Ab, and then the immunoprecipitation complex was incubated with click buffer (0.4 mM Biotin-azide, 0.5 mM Tris(benzyltriazolylmethyl)amine, 4 mM CuSO₄, 4 mM Tris(2-carboxyethyl)phosphine hydrochloride) for 2 h. After incubation, loading buffer was added to reaction buffer, and the samples were subjected for SDS-PAGE and detected by the indicated Ab.

Lentivirus production and infection of cells

The pLenti CMV 3× Flag-ZDHHC/S17 vector was transfected into HEK293T cells together with psPAX2 and pMD2.G using Lipofectamine 2000 (Thermo Fisher Scientific). At 36 h after transfection, the cell supernatants containing lentiviral particles were collected and used to infect THP-1 cells for 24 h along with supplementation with polybrene (4 μg/ml; Selleck Chemicals). The cells were cultured with puromycin (1.5 μg/ml; Selleck Chemicals) for 3 d to screen out positive cells and were detected by Western blotting.

ELISA

The IL-1β concentration in the supernatants of THP-1 cells was detected by the human IL-1β ELISA kit (557966; BD Biosciences, San Jose, CA). The IL-1β concentration in the supernatants of BMDMs and mice serum was measured with the mouse IL-1β ELISA kit (val601) according to manufacturer (R&D Systems, Minneapolis, MN) instructions.

ASC oligomerization

The indicated cells were lysed with lysis buffer and then centrifuged at 13,000g for 10 min. The cell lysates were subjected for Western blotting, whereas the pellets were washed thrice with PBS and then cross-linked with DSS (2 μM) for 30 min at 37°C. After that, the crosslinked pellets were centrifuged and mixed with SDS-loading buffer for Western blotting.

Confocal microscopy

HEK293T cells were transfected with the indicated plasmids for 48 h. Then, they were fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100 in PBS, and blocked with 5% BSA. After that, primary Abs were added to cells for overnight incubation at 4°C and then stained with Cy3- and FITC-conjugated secondary Abs. Finally, the cells were stained with 4′,6-diamidino-2-phenylindole for 5 min and then analyzed using a confocal laser scanning microscope (TCS, SP8; Leica, Wetzlar, Germany).

Induction of colitis

The protocol for animal experiments was approved (20220119001) by the Experimental Animal Ethics Committee of Guangzhou University of Chinese Medicine (Guangzhou, China). C57BL/6 mice were purchased from the experimental animal center of Guangzhou University of Chinese Medicine. Seven-week-old male mice were used in this study. For 7 d, 2.5% DSS dissolved in the drinking water was given to mice, and the mice were treated daily with 2-BP (0, 10, or 40 mg/kg) or MCC950 (20 mg/kg) by i.p. injection. 2-BP was dissolved in complete DMSO and then diluted with PEG300, Tween 80, and saline, and the concentration of DMSO in the final injection was less than 5%. Control animals were given equal amount of solvent. The mice were killed at day 12. The colons were harvested, measured, and used for histopathology and Western blotting.

Statistical analyses

All of the experiments were reproducible and repeated at least three times with similar results. Levene’s test for equality of variances was performed, which provided information for Student t tests to distinguish the equality of means. The means were illustrated using histograms with error bars representing the SD; a p value of < 0.05 was considered statistically significant.

Data availability statement

The data supporting the findings of this study are available within the article.

Results

NLRP3 is palmitoylated during its activation

Previously, we reported that multiple host and viral proteins modulate the NLRP3 inflammasome using various mechanisms (28, 44–48). Protein palmitoylation contributes to the trafficking and functions of proteins (31). In this study, we attempted to ascertain whether NLPR3 undergoes palmitoylation. First, in HEK293T cells expressing NLRP3 or ASC, we undertook an ABE assay. The results showed that NLRP3 (but not ASC) was palmitoylated (Fig. 1A). The click chemistry method also confirmed the palmitoylation of NLRP3 in HEK293T cells overexpressing NLRP3 (Fig. 1B). Activation of the NLRP3 inflammasome involves two stages (priming and activation (5)), so we then evaluated the palmitoylation of NLRP3 in the quiescent state and activated state. In 12-O-Tetradecanoylphorbol 13-acetate (TPA)–differentiated human leukemic monocyte (THP-1) macrophages, the palmitoylation of NLRP3 did not occur in the quiescent state (Fig. 1C) but was enhanced gradually upon stimulation with nigericin for 5 and 15 min or ATP for 15 and 30 min (Fig. 1C). Different moieties induce activation of the NLRP3 inflammasome in different ways, so we also treated two other inducers of the NLRP3 inflammasome, alum and MSU, in TPA-differentiated THP-1 cells and found that NLRP3 palmitoylation was also induced (Fig. 1D). Moreover, in LPS-primed mouse BMDMs, ABE assays also showed that NLRP3 was palmitoylated upon treatment with different inducers of the NLRP3 inflammasome: nigericin, ATP, alum, and MSU (Fig. 1E). Taken together, we demonstrated that NLRP3 is palmitoylated during its activation.

FIGURE 1.

NLRP3 is palmitoylated. (A) HEK29T cells were transfected with plasmid encoding NLRP3 or ASC for 48 h and lysed for the ABE assay, followed by Western blotting. Streptavidin-HRP was done to detect possible palmitoylated protein. The intensity of NLRP3 palmitoylation was measured by ImageJ (right). (B) HEK29T cells were transfected with plasmid encoding NLRP3 for 24 h and labeled with 50 μM palmitic acid analog for 6 h. The cells were collected and subjected to click chemistry assay. Streptavidin-HRP was done to detect possible palmitoylated protein. (C) TPA-differentiated THP-1 cells were stimulated with LPS (1 µg/ml), LPS plus nigericin (2 μM), or LPS plus ATP (2.5 mM) for the indicated time. The cells were lysed and subjected to the ABE assay. The level of NLRP3 palmitoylation was detected by streptavidin-HRP and measured by ImageJ (right). (D) TPA-differentiated THP-1 cells were stimulated with LPS (1 µg/ml), LPS plus nigericin (2 μM, 15 min), LPS plus ATP (2.5 mM, 30 min), LPS plus alum (200 μg/ml, 3 h), or LPS plus MSU (100 mg/ml, 3 h). The cells were lysed and subjected to the ABE assay. The level of NLRP3 palmitoylation was detected by streptavidin-HRP and measured by ImageJ (right). (E) LPS-primed BMDMs were stimulated with nigericin (2 μM, 15 min), ATP (2.5 mM, 30 min), alum (200 μg/ml, 3 h), or MSU (100 mg/ml, 3 h). The cells were lysed for the ABE assay, followed by Western blotting. The level of NLRP3 palmitoylation was detected by streptavidin-HRP and measured by ImageJ (right). *p < 0.05, **p < 0.01, ***p < 0.001. HAM, hydroxylamine; IB, immunoblotting; IP, immunoprecipitation; ns, no significance.

2-BP blocks the palmitoylation and activation of NLRP3

Next, we evaluated the effect of NLRP3 palmitoylation on regulation of activation of the NLRP3 inflammasome. 2-BP is a nonmetabolizable palmitate analog. It serves as an inhibitor of protein palmitoylation by irreversibly inhibiting the activity of all palmitoyltransferases (49). 2-BP has been used widely in the study of protein palmitoylation in vitro and in vivo (32–34). In TPA-differentiated THP-1 cells, the palmitoylation of NLRP3 induced by nigericin and ATP was blocked by 2-BP (Fig. 2A). Meanwhile, IL-1β maturation, IL-1β secretion, and caspase-1 cleavage induced by nigericin and ATP were reduced significantly upon 2-BP treatment (Fig. 2B, 2C). ASC oligomerization stimulated by nigericin and ATP was also attenuated by 2-BP in TPA-differentiated THP-1 cells (Fig. 2D). Furthermore, in LPS-primed BMDMs, 2-BP showed consistent blockade of NLRP3 palmitoylation (Fig. 2E), thereby reducing the maturation and secretion of IL-1β (Fig. 2F) and attenuating the ASC oligomerization (Fig. 2G) induced by nigericin and ATP. These results revealed NLRP3 palmitoylation to be a crucial part of activation of the NLRP3 inflammasome and that the palmitoylation inhibitor 2-BP could block NLRP3 palmitoylation effectively and then inhibit activation of the NLRP3 inflammasome.

FIGURE 2.

2-BP blocks the palmitoylation and activation of NLRP3. (A–D) TPA-differentiated THP-1 cells were stimulated with nigericin (2 μM) or ATP (2.5 mM) and treated with DMSO or 2-BP (50 μM, 6 h). The cells were lysed with lysis buffer. The cell lysates were subjected to the ABE assay. The intensity of NLRP3 palmitoylation was measured by ImageJ (right) (A), and pellets were used for analysis of ASC oligomerization (D). IL-1β content in supernatants was measured by Western blotting (B) and ELISA (C). (E and F) LPS-primed BMDMs were stimulated with nigericin (2 μM) or ATP (2.5 mM) and treated with DMSO or 2-BP (50 μM, 6 h). The cells were lysed with lysis buffer. The cell lysates were subjected to the ABE assay. The intensity of NLRP3 palmitoylation was measured by ImageJ (right) (E), and the pellets were used for analysis of ASC oligomerization (G). The IL-1β content in supernatants was measured by ELISA (F). **p < 0.01, ***p < 0.001. HAM, hydroxylamine; IB, immunoblotting; IP, immunoprecipitation.

NLRP3 is palmitoylated at Cys419

Typically, palmitate is attached to a cysteine residue via a thioester linkage (50), and there are 45 cysteine residues in the amino acid sequence of NLRP3 protein. To determine the specific modification site of NLRP3 palmitoylation, we first employed a predictor of palmitoylation sites, GPS-Palm (51), to analyze the possible palmitoylation sites in NLRP3 protein. Nine highlighted cysteine residues (Fig. 3A) were discovered. They were distributed in the PYRIN domain (Cys⁸), NATCH domain (Cys²⁷⁹ and Cys⁴¹⁹), LRR domain (Cys⁸³⁷, Cys⁸³⁸, and Cys⁹³³), the linkage between the PRYIN domain and NATCH domain (Cys¹³⁰ and Cys¹⁵⁰) and the linkage between the NATCH domain and the LRR domain (Cys⁶⁷³) (Fig. 3B). Second, we generated a mutation in these cysteine residues with a serine substitution respectively, and undertook ABE assays in HEK293T cells overexpressing wild-type (WT) or mutated NLRP3. Interestingly, the palmitoylation level of NLRP3 bearing the C419S mutation in the NATCH domain was reduced notably compared with that in WT NLRP3 (Fig. 3C), suggesting that NLRP3 was palmitoylated mainly at Cys⁴¹⁹.

FIGURE 3.

NLRP3 is palmitoylated at Cys⁴¹⁹. (A) Possible palmitoylation sites of NLRP3 as predicted by the GPS-Palm program. (B) Protein structure of NLRP3 showing all cysteine residues with the predicted cysteine molecules highlighted. (C) HEK29T cells were transfected with plasmid of encoding NLRP3 or NLRP3 mutant for 48 h and then lysed for the ABE assay. The level of NLRP3 palmitoylation was detected by streptavidin-HRP and measured by ImageJ (right). (D and E) HEK293T cells were transfected with plasmid encoding ASC, pro–IL-1β, procaspase-1, and NLRP3 or NLRP3 mutant. The IL-1β content in supernatants was measured by Western blotting (D) and ELISA (E). (F–H) NLRP3-deficient THP-1 cells were reconstituted with NLRP3 wild type (WT) or NLRP3 C419S by a lentivirus and treated with nigericin (2 μM) or ATP (2.5 mM). The IL-1β content in supernatants was measured by ELISA (F) and Western blotting (G). The cells were lysed for Western blotting, and the pellets were subjected to analysis for ASC oligomerization (H). ***p < 0.001. HAM, hydroxylamine; IB, immunoblotting; IP, immunoprecipitation.

We demonstrated above that blockade of NLRP3 palmitoylation resulted in attenuation of activation of the NLRP3 inflammasome. Hence, we further compared the role of WT NLRP3 and C419S NLRP3 in regulation of inflammasome activation. In HEK293T cells, a reconstructed NLRP3-inflammasome system was generated as described previously (46), and NLRP3 containing the C419S mutation induced a significant decrease in the maturation and release of IL-1β compared with that of WT NLRP3 (Fig. 3D, 3E). In addition, we stably reconstituted THP-1–def cells with WT NLRP3 and C419S NLRP3 through a lentivirus system, and as expected, the ability of THP-1–def cells reconstituted with C419S NLRP3 to induce IL-1β secretion (Fig. 3F, 3G) and ASC oligomerization (Fig. 3H) was reduced significantly compared with that of THP-1–def cells reconstituted with WT NLRP3. Collectively, these results suggested that NLRP3 was palmitoylated at Cys⁴¹⁹, and that the C419S mutation notably attenuated the NLRP3 palmitoylation level and served as a “brake” on activation of the NLRP3 inflammasome.

NLRP3 palmitoylation is mediated by ZDHHC17

Protein palmitoylation is catalyzed by palmitoyltransferases. There are 23 palmitoyltransferases in humans. They contain a conserved zinc finger DHHC motif (Asp–His–His–Cys) and are named ZDHHC1 to ZDHHC24 (ZDHHC10 is absent) (31). To identify the predominant palmitoyltransferases for NLRP3 palmitoylation, we constructed and expressed 18 palmitoyltransferases. In HEK293T cells transfected with plasmids expressing NLRP3 and ZDHHCs, ABE assays revealed that ZDHHC17 enhanced the palmitoylation level of NLRP3 markedly compared with that using other ZDHHCs (Fig. 4A). Additional assays confirmed that NLRP3 palmitoylation was enhanced significantly with an increase in ZDHHC17 expression but was not affected if ZDHHC18 was present (Fig. 4B). More importantly, a catalytically inactive form of ZDHHC17 (ZDHHS17) with the C467S mutation could not enhance NLRP3 palmitoylation in HEK293T cells (Fig. 4C). Furthermore, we generated THP-1 cell lines that stably overexpressed Flag-tagged ZDHHC17 and ZDHHS17 by a lentivirus system. Compared with control cells, the level of NLRP3 palmitoylation was enhanced obviously upon ATP treatment in THP-1 cells expressing ZDHHC17 but remained at a similar level in cells expressing ZDHHS17 (Fig. 4D). Conversely, in BMDMs, endogenous expression of zdhhc17 was downregulated by small interfering (si) RNA transfection, and nigericin-induced palmitoylation of NLRP3 was attenuated in si-zdhhc17 cells compared with that in control cells (Fig. 4E). Taken together, we demonstrated ZDHHC17 to be the predominant palmitoyltransferase for NLRP3 palmitoylation based on its enzyme activity.

FIGURE 4.

ZDHHC17 mediates NLRP3 palmitoylation. (A) HEK293T cells were cotransfected with plasmid encoding NLRP3 and plasmid encoding ZDHHC1–ZDHHC24 for 48 h, and the cells were lysed for the ABE assay, followed by Western blotting. The level of NLRP3 palmitoylation was detected by streptavidin-HRP and measured by ImageJ (right). (B) HEK293T cells were cotransfected with plasmid encoding NLRP3 and plasmid encoding ZDHHC17 or ZDHHC18 for 48 h. The cells were lysed for the ABE assay, followed by Western blotting. The level of NLRP3 palmitoylation was detected by streptavidin-HRP and measured by ImageJ (right). (C) HEK293T cells were cotransfected with plasmid encoding NLRP3 and plasmid encoding ZDHHC17 or ZDHHS17 for 48 h. The cells were lysed for the ABE assay, followed by Western blotting. The level of NLRP3 palmitoylation was detected by streptavidin-HRP and measured by ImageJ (right). (D) THP-1 cells stably expressing ZDHHC17 or ZDHHS17 were stimulated with LPS (1 µg/ml) or LPS plus ATP (2.5 mM). The cells were lysed for the ABE assay, followed by Western blotting. The level of NLRP3 palmitoylation was detected by streptavidin-HRP and measured by ImageJ (right). (E) BMDMs were transfected with si-NC or si-zdhhc17 for 48 h and stimulated with LPS (1 µg/ml) or LPS plus nigericin (2 μM). The cells were lysed for the ABE assay, followed by Western blotting. The level of NLRP3 palmitoylation was detected by streptavidin-HRP and measured by ImageJ (right). *p < 0.05, **p < 0.01, ***p < 0.001. HAM, hydroxylamine; IB, immunoblotting; IP, immunoprecipitation; ns, no significance.

ZDHHC17 promotes activation of the NLRP3 inflammasome

Next, we investigated the effect of ZDHHC17 in modulating activation of the NLRP3 inflammasome. In HEK293T cells reconstructed with an NLRP3-inflammasome system, IL-1β secretion was facilitated by ZDHHC17, but not ZDHHC18, which did not regulate NLRP3 palmitoylation (Fig. 5A, 5B). Meanwhile, in HEK293T cells transfected with plasmids expressing NLRP3 and ASC, ASC oligomerization was also stimulated in the presence of ZDHHC17, but not ZDHHC18 (Fig. 5C), which suggested that ZDHHC17 (but not ZDHHC18) could promote activation of the NLRP3 inflammasome. ZDHHC17 palmitoylated NLRP3 depending on its enzyme activity, so we explored whether ZDHHC17 promoted inflammasome activation depending on its enzyme activity. Interestingly, compared with control cells, maturation and secretion of IL-1β and ASC oligomerization stimulated by ATP were increased notably in TPA-differentiated THP-1 cells that stably expressed ZDHHC17 but not in cells that stably expressed ZDHHS17 (Fig. 5D–F), which suggested that the inflammasome promotion elicited by ZDHHC17 was also dependent upon its palmitoyltransferase activity. In addition, we investigated the regulation of zdhhc17 expression upon NLRP3 activation in BMDMs and found that zdhhc17 might be constitutively expressed in BMDMs (Supplemental Fig. 1A–D). Further, the expression of endogenous zdhhc17 was knocked down by small interfering RNA transfection, and the IL-1β secretion and NLRP3–ASC interactions induced by nigericin were attenuated compared with that in control cells (Fig. 5G, 5H), thereby indicating that ZDHHC17 was essential for the assembly and activation of the NLRP3 inflammasome. More importantly, overexpression of ZDHHC17 in THP-1–def cells did not enhance the secretion of IL-1β, further confirming that ZDHHC17 targets NLRP3 to promote inflammasome activation (Fig. 5I). Collectively, our results demonstrated that ZDHHC17-mediated palmitoylation of NLRP3 promoted inflammasome activation.

FIGURE 5.

ZDHHC17 promotes activation of the NLRP3 inflammasome. (A and B) HEK293T cells were transfected with plasmid encoding NLRP3, ASC, pro-IL-1β, procaspase-1, and ZDHHC17 or ZDHHC18. The IL-1β content in supernatants was measured by Western blotting (A) and ELISA (B). (C) HEK293T cells were cotransfected with plasmid encoding NLRP3, ASC, and ZDHHC17 or ZDHHC18 for 48 h. The cells were lysed with lysis buffer. Cell lysates were analyzed by SDS-PAGE, and pellets were used for analysis of ASC oligomerization. (D–F) THP-1 cells stably expressing ZDHHC17 or ZDHHS17 were stimulated with LPS (1 µg/ml) or LPS plus ATP (2.5 mM). The IL-1β content in supernatants was measured by Western blotting (D) and ELISA (E). The cells were lysed for Western blotting, and the pellets were subjected to analysis of ASC oligomerization (F). (G and H) BMDMs were transfected with si-NC or si-zdhhc17 for 48 h and stimulated with LPS (1 µg/ml) or LPS plus nigericin (2 μM). The IL-1β content in supernatants was measured by ELISA (G). The cell lysates were prepared using lysis buffer. Then, immunoprecipitation was undertaken with anti-NLRP3 Ab, followed by analyses by SDS-PAGE (H). (I) THP-1–def cells stably expressing ZDHHC17 were stimulated with LPS (1 µg/ml) or LPS plus ATP (2.5 mM). The IL-1β content in supernatants was measured by ELISA. **p < 0.01. ***p < 0.001. IB, immunoblotting; IP, immunoprecipitation; ns, no significance.

ZDHHC17 interacts with NLRP3 and promotes NEK7–NLRP3 interactions

To explore how ZDHHC17 modulates NLRP3 palmitoylation, we first undertook coimmunoprecipitation experiments in HEK293T cells transfected with plasmids expressing NLRP3 and ZDHHC17. NLRP3 interacted with ZDHHC17 (Fig. 6A), and ZDHHC17 associated with NLRP3 (Fig. 6B). Moreover, in TPA-differentiated THP-1 or LPS-primed BMDMs, NLRP3 showed no interaction with ZDHHC17 in the quiescent state but interacted with ZDHHC17 under nigericin treatment (Fig. 6C, 6D). Confocal laser scanning microscopy showed that NLRP3 and ZDHHC17 colocalized and formed large spots at Golgi (Fig. 6E, 6F). Further study revealed that ZDHHC17 was bound to the LRR domain of NLRP3 (Fig. 6G), which suggested that ZDHHC17 interacted with NLRP3 and then palmitoylated NLRP3 to facilitate activation of the NLRP3 inflammasome.

FIGURE 6.

ZDHHC17 interacts with NLRP3 and promotes NEK7–NLRP3 interactions. (A) HEK293T cells were cotransfected with plasmids encoding NLRP3 and ZDHHC17 for 48 h. The cell lysates were prepared using lysis buffer, then used for immunoprecipitation with anti-HA Ab or control IgG, and analyzed by SDS-PAGE. (B) HEK293T cells were cotransfected with plasmids encoding NLRP3 and ZDHHC17 for 48 h. The cell lysates were prepared using lysis buffer, then used for immunoprecipitation with anti-Flag Ab or control IgG, and analyzed by SDS-PAGE. (C and D) THP-1 cells or BMDMs cells were stimulated with LPS (1 µg/ml) or LPS plus nigericin (2 μM). The cell lysates were prepared using lysis buffer, then used for immunoprecipitation with anti-NLRP3 Ab or control IgG, and analyzed by SDS-PAGE. (E) HEK293T cells were transfected with pHA-NLRP3 and pFlag-vector or pFlag-ZDHHC17 for 48 h. Localization of HA-NLRP3 (red) and Flag-ZDHHC17 (green) was analyzed under confocal laser scanning microscopy. (F) HEK293T cells were transfected with pEGFP-NLRP3 or pFlag-ZDHHC17. Localization of EGFP-NLRP3 (green), Flag-ZDHHC17 (red), and GM130 (purple) was analyzed under confocal laser scanning microscopy. (G) HEK29T cells were transfected with pFlag-NLRP3 and pHA-vector, pHA-NLRP3, pHA-PYD, pHA-NACHT, or pHA-LRR. The cell lysates were prepared using lysis buffer, then used for flag immunoprecipitation with anti-HA Ab, and analyzed by SDS-PAGE. (H) NLRP3-deficient THP-1 cells were reconstituted with NLRP3 wild type (WT) or NLRP3 C419S by lentivirus and treated with LPS (1 µg/ml) or LPS plus nigericin (2 μM). The cell lysates were prepared using lysis buffer, then used for immunoprecipitation with anti-NEK7 Ab, and analyzed by SDS-PAGE. (I) THP-1 cells stably expressing ZDHHC17 or ZDHHS17 were stimulated with LPS (1 µg/ml) or LPS plus ATP (2.5 mM). The cell lysates were prepared using lysis buffer, then used for immunoprecipitation with anti-NEK7 Ab, and analyzed by SDS-PAGE. (J) BMDMs were transfected with si-NC or si-zdhhc17 for 48 h and stimulated with LPS (1 µg/ml) or LPS plus nigericin (2 μM). The cell lysates were prepared using lysis buffer, then used for immunoprecipitation with anti-NEK7 Ab, and analyzed by SDS-PAGE. IB, immunoblotting; IP, immunoprecipitation.

Recently, NEK7 was identified as an essential component of the NLRP3 inflammasome. It serves as the crucial mediator to modulate inflammasome activation by interacting with NLRP3 (19–21). We have shown that ZDHHC17 binds to the LRR domain and catalyzes NLRP3 palmitoylation at the NACHT domain. Interestingly, we realized that the LRR domain and NACHT domain are involved in NEK7 binding (19) and that a peptide array–based screening identified that ZDHHC17 could interact with NEK7 (52). Hence, we speculated whether ZDHHC17-mediated palmitoylation of NLRP3-modulated inflammasome activation by regulating NEK7–NLRP3 interactions. In THP-1–def cells reconstituted with WT NLRP3 and C419S NLRP3, NLRP3 with the C419S mutant failed to recruit NEK7 upon nigericin treatment (Fig. 6H), thereby indicating that NLRP3 palmitoylation plays an important part in NEK7–NLRP3 interactions. Compared with control cells, NEK7–NLRP3 interactions were enhanced significantly upon ATP treatment in THP-1 cells expressing ZDHHC17 but not in cells expressing ZDHHS17 (Fig. 6I). Meanwhile, Nlrp3 coimmunoprecipitation with Nek7 was attenuated in Zdhhc17 knockdown BMDMs activated with LPS and nigericin (Fig. 6J). To conclude, our results suggested that ZDHHC17 interacted with NLRP3 and palmitoylated NLPR3 to facilitate NEK7–NLRP3 interactions, which resulted in activation of the NLRP3 inflammasome.

2-BP suppresses NLRP3-mediated colitis in mice

We demonstrated above that NLRP3 palmitoylation is crucial for inflammasome activation and that a palmitoylation inhibitor, 2-BP, can block NLRP3 palmitoylation effectively and inhibit activation of the NLRP3 inflammasome in vitro. Next, we explored whether 2-BP could have effects in inflammatory diseases in vivo. Colitis induced in mice with DSS is mediated by NLRP3-based inflammation (38) and has been used extensively as a model to explore NLRP3 regulation. In a DSS-induced colitis model, the mice were given 2-BP (0, 10, or 40 mg/kg) and MCC950 (20 mg/kg; a specific inhibitor of NLRP3) (53) by i.p. injection. Interestingly, 2-BP and MCC950 attenuated weight loss significantly (Fig. 7A), increased the survival rate (Fig. 7B), and rescued the shortening of colon length (Fig. 7C, 7D) of mice. In colonic sections stained with H&E, infiltration of inflammatory cells induced by DSS was also attenuated by 2-BP and MCC950 (Fig. 7E, 7F). Moreover, IL-1β content in the serum (Fig. 7G) or colonic tissue (Fig. 7H) of mice was decreased notably upon treatment with 2-BP and MCC950. In summary, these data demonstrated that 2-BP could suppress NLRP3-mediated colitis in mice effectively.

FIGURE 7.

2-BP suppresses NLRP3-mediated colitis in mice. C57BL/6 mice were given 2.5% DSS for 7 d and then treated daily with 2-BP (0, 10, or 40 mg/kg) or 20 mg/kg MCC950 by i.p. injection. The bodyweights (A) and survival rates (B) of mice were evaluated. The mice were killed at day 12, and the colons were harvested, measured (C and D), and used for histopathology (E and F) and Western blotting (H). The IL-1β content in mice serum was measured by ELISA (G). *p < 0.05, **p < 0.01, ***p < 0.001.

Discussion

Here, we revealed a novel mechanism whereby palmitoylation of NLPR3 is required for inflammasome activation. Initially, through ABE assays, we identified that NLRP3 is palmitoylated in HEK293T cells, THP-1 cells and BMDMs and that blockade of NLRP3 palmitoylation by 2-BP markedly suppresses IL-1β secretion and ASC oligomerization in THP-1 cells and BMDMs, thereby indicating that NLRP3 palmitoylation is essential for its activation. Further computational and mutation analyses indicated that Cys⁴¹⁹ of NLRP3 is responsible for its palmitoylation. In addition, ZDHHC17 was identified to be the predominant palmitoyltransferase mediating NLRP3 palmitoylation. More importantly, application of 2-BP suppressed intestinal inflammation in mice. Taken together, we demonstrated that palmitoylation of NLRP3 modulates inflammasome activation and inflammatory bowel disease development (Fig. 8).

FIGURE 8.

A proposed model for the palmitoylation of NLRP3 modulates inflammasome activation and colitis development. In the quiescent condition, NLRP3 is inactive and not palmitoylated (left). However, in response to activating signals stimulating with nigericin, ATP, and MSU, ZDHHC17 interacts with NLPR3 and mediates NLPR3 palmitoylation, thereby promoting NLRP3-NEK7 interaction and inflammasome activation (right). Inhibiting NLRP3 palmitoylation using 2-BP attenuated inflammasome activation and colitis severity in mice (right).

Several post-translational modifications have been identified at NLRP3 and have important roles in regulating the NLRP3 inflammasome: ubiquitylation (22, 23), phosphorylation (24–26), and SUMOylation (27, 28). Novel post-translational modifications in NLRP3 are of great interest to researchers. During preparation of this article, several studies reported that NLRP3 undergoes palmitoylation (40–42). However, the process and role of NLRP3 palmitoylation are still controversial. Wang et al. (40) propose that palmitoylation prevents sustained inflammation by limiting NLRP3 inflammasome activation through chaperone-mediated autophagy. Two other recent studies hold the opposite view that palmitoylation of NLRP3 promotes NLRP3 inflammasome activation (41, 42). In addition, the specific enzymes meditating NLRP3 palmitoylation are different in different studies, including ZDHHC5, ZDHHC7, and ZDHHC12. Moreover, the specific palmitoylation sites in NLRP3 also differed from different studies, which might be due to different enzymes mediating different modification sites. Therefore, elucidating the process and role of NLRP3 palmitoylation is very important for the treatment of NLRP3-mediated inflammatory diseases. In this study, we demonstrated that NLRP3 is palmitoylated at Cys⁴¹⁹ and that the palmitoylation of NLRP3 facilitates NEK7 recruitment to license NLRP3 activation. Interestingly, it has been shown that the subcellular localization of NLRP3 is important for its activation, including migration to the plasma membrane (45) and localization at the endoplasmic reticulum (44), mitochondria (15), Golgi (26), and trans-Golgi (18). Although we demonstrated that NLRP3 was palmitoylated in Golgi, whether palmitoylation triggered to subsequent translocation is not known and merits further investigation.

ZDHHC17 is a palmitoyltransferase, is expressed ubiquitously in diverse tissues and cells (54), and is multifunctional. Depending on its enzyme activity, ZDHHC17 regulates progression of Huntington’s disease through palmitoylating the Huntington protein (55), modulates growth factor signaling by palmitoylating and stabilizing sprout-2 (56), and controls distal-axon integrity and somal responses to axonal damage by palmitoylating dual leucine zipper kinase and nicotinamide mononucleotide adenylyltransferase-2 (57). In this study, we revealed a novel function of ZDHHC17: it modulates the NLRP3 inflammasome depending on its palmitoylation activity. ZDHHC17 mediates NLRP3 palmitoylation and facilitates NEK7–NLRP3 interactions to license NLRP3 activation. In addition, our data confirmed that ZDHHC17 is Golgi-localized (58) and that NLRP3 undergoes Golgi localization (26) and showed that ZDHHC17 mediates NLRP3 palmitoylation in Golgi.

IBD includes ulcerative colitis and Crohn’s disease. IBD is a chronic immune-mediated disease affecting the gastrointestinal tract. The causes of IBD includes environmental, microbial, and immune-mediated factors (59). Often, IBD presents with abdominal pain, diarrhea, bloody stools, and weight loss. It is difficult to cure completely and can lead to bowel cancer (60). NLRP3 has been identified as an important regulator of intestinal inflammation that responds to microbial ligands and induces the secretion of IL-1β and IL-18 (61). Zhou et al. (39) revealed that NLRP3 with the R779C mutation induces a hyperinflammatory response and contributes to very early-onset IBD in infants and children younger than 6 y. In this study, we established a DSS-induced colitis model in mice that exhibited a distinctive phenotype similar to that of human IBD (62), and the application of the palmitoylation inhibitor 2-BP attenuated the severity of colitis in mice significantly. Hence, drugs targeting NLRP3 palmitoylation could be efficacious against IBD, as well as other NLRP3-mediated diseases.

The important role of protein palmitoylation in modulating protein function and disease development. Hence, several drugs targeting protein palmitoylation have been proposed and investigated. There are three main types of drugs that target the palmitoylation of proteins. First, drugs target specific palmitoylated cysteine residues to block palmitoylation. Such drugs include C-178 (a selective small-molecule antagonist of STING), which binds covalently to Cys⁹¹ of STING to block STING palmitoylation and inhibit type I IFN signaling (63). Second, drugs target ZDHHCs to inactivate their activity. Although there is no specific inhibitor for ZDHHCs, a nonselective palmitoylation inhibitor, 2-BP, has been used in multiple studies in vitro and in vivo (32–34, 64) to validate the effect of protein palmitoylation. Third, drugs target acyl-protein thioesterases to prevent depalmitoylation. For instance, Palm-B (a deacylating enzyme inhibitor) enhances melanocortin-1 receptor palmitoylation to maintain melanocortin-1 receptor signaling and suppress tumor progression (65). In this study, we demonstrated that NLRP3 is palmitoylated at Cys⁴¹⁹ and that ZDHHC17 is the predominant palmitoyltransferase for NLRP3 palmitoylation. 2-BP application inhibited NLRP3 activation effectively in vitro and in vivo. However, 2-BP is not suitable for clinical application due to its nonspecificity for ZDHHC enzymes and carries the risk of targeting nonrelevant proteins (66, 67). Importantly, Cys⁴¹⁹ has been shown to be a specific target site for acrylamide derivatives to suppress NLRP3 activity (68). Therefore, drugs targeting the Cys⁴¹⁹ of NLRP3 or ZDHHC17 activity would be more specific and efficacious and have more prospects for clinical application.

In conclusion, we revealed a distinct mechanism underlying NLRP3 activation in which NLRP3 is palmitoylated by ZDHHC17 to facilitate NEK7–NLRP3 interactions and license inflammasome activation. In addition, inhibiting NLRP3 palmitoylation using 2-BP attenuated inflammasome activation and colitis severity in mice. Our data suggest that drugs targeting NLRP3 palmitoylation could be efficacious in the treatment of NLRP3-mediated diseases.

Disclosures

The authors have no financial conflicts of interest.