Phosphorylation of NLRP6 at Serine 27 site is critical for NLRP6 inflammasome activation and neurological impairment in ischemic stroke

Abstract

Nucleotide-binding oligomerization domain-like receptor pyrin domain-containing 6 (NLRP6) is a pattern recognition receptor that is highly expressed in microglia and forms a cytosolic inflammasome to aggravate neuroinflammation and neuronal injury during ischemic stroke. However, the post-translational mechanisms regulating the NLRP6 inflammasome activation are not well characterized. Herein, we demonstrated that the phosphorylation of NLRP6 was markedly increased in response to cerebral ischemia and was closely associated with NLRP6 inflammasome function. Serine 27 (Ser27) was identified as a conserved phosphorylation site in NLRP6, and NLRP6 inflammasome activity was tightly regulated by phosphorylation at this site. Moreover, the oligomerization and subsequent inflammasome assembly and activation of NLRP6 were positively regulated by Ser27 phosphorylation. Phenotypically, dephosphorylation at the Ser27 site blocked NLRP6 inflammasome activation in ischemic stroke, drastically reducing inflammasome-related neuroinflammation and neurological injury in vitro and in vivo. Our findings demonstrate a pivotal mechanism regulating NLRP6 inflammasome activation in ischemic stroke and provide evidence for the therapeutic targeting of NLRP6 Ser27 phosphorylation.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12964-026-02683-9.

Introduction

Ischemic stroke is the leading cerebrovascular disorder worldwide, causing long-term disability, mortality, and significant economic burden [1]. Intravenous thrombolysis and endovascular thrombectomy can help recanalize ischemic vessels, but reperfusion often induces secondary brain injury [2, 3]. A major mediator of reperfusion-related damage is sterile innate inflammation, which is strongly associated with long-term motor and cognitive deficits [4–6]. In cerebral ischemia, neuronal death releases damage-associated molecular patterns (DAMPs) that rapidly drive microglial accumulation and subsequent macrophage infiltration in the peri-infarct border zone. By engaging pattern recognition receptors, DAMPs hyperactivate microglia and macrophages, initiating intracellular signaling cascades and inducing cytokine and chemokine production. These inflammatory cascades exacerbate brain injury and contribute to poor clinical outcomes. Thus, elucidating the mechanisms of innate inflammation may yield novel therapeutic strategies for ischemic stroke, with inflammasome signaling emerging as a central candidate pathway.

Intracellularly, inflammasomes assemble into multimeric protein complexes composed of NOD-like receptors (NLRs), the adaptor protein ASC, and the effector zymogen pro-caspase-1. These complexes are essential components of the innate immune system. Inflammasome activation is initiated by sensor stimulation, leading to autoproteolytic conversion of pro-caspase-1 to cleaved-caspase-1, which then cleaves pro-IL-1β and pro-IL-18 into their mature forms, driving rapid innate inflammatory responses [7]. Cleaved-caspase-1 also induces pyroptotic cell death via gasdermin D cleavage. Overactivated inflammasomes drive the onset and progression of various inflammatory diseases, such as neurodegenerative diseases [8–10]. NLRP6, a less-studied NLR, was mainly expressed in microglia, with its protein expression increasing markedly in response to ischemic stroke. NLRP6 associates with ASC and pro-caspase-1, promoting caspase-1–dependent maturation of IL-1β and IL-18, aggravating neuroinflammation, and resulting in neuronal death and neurological impairment, thereby mediating ischemic stroke injury [11]. Our previous study demonstrated that NLRP6 inflammasome activation mediates ischemic stroke injury. Nevertheless, the potential molecular mechanisms involved in NLRP6 inflammasome activation have not been completely defined, particularly the role of post-translational modifications in tuning NLRP6 activation in the brain.

Post-translational modifications are important regulators of NLRP6 inflammasome activation. Following infection with Citrobacter rodentium, K63-linked ubiquitination of NLRP6 enhances inflammasome assembly, inducing excessive maturation of IL-18 in the colonic mucosa, whereas the deubiquitinase CYLD counteracts this by deubiquitinating NLRP6, limiting IL-18 production and preventing severe colitis [12]. Notably, the post-translational regulatory mechanisms that modulate NLRP6 inflammasome activation in central nervous system disorders, particularly in ischemic stroke, remain unclear. Given that phosphorylation modulates several inflammasome sensors in other contexts, we hypothesized a phosphorylation-based control of NLRP6 inflammasome activation in ischemic brain injury. Here, we show that NLRP6 is hyperphosphorylated in experimental cerebral ischemia. Phosphorylation at Ser27 positively regulates NLRP6 inflammasome assembly and activity, thereby aggravating neuroinflammation and brain injury in ischemic stroke.

Materials and methods

Reagents and antibodies

Phos binding reagent (Phos-tag) (F4002) was obtained from APExBIO (Houston, TX, USA); Lambda phosphatase (P9614), lipoteichoic acid (LTA) (L2515), lipopolysaccharide (LPS) (Escherichia coli, O111:B4, L4130), Nigericin (N7143), and Flagellin (SRP8029) were acquired from Sigma-Aldrich (St. Louis, MO, USA); Coomassie Brilliant Blue R-250 (A610037) and Disuccinimidyl suberate (C100015) were obtained from Sangon Biotech (Shanghai, China); Blue Native polyacrylamide gel electrophoresis (PAGE) (P0545S), Fast Silver Stain Kit (P0017S), Calcein/Propidium Iodide (PI) Assay Kit (C2015M), Cell Counting Kit (CCK)-8 (C0037), and Lactate Dehydrogenase (LDH)Assay Kit (C0018) were obtained from Beyotime Biotechnology (Shanghai, China); Trichloroacetic Acid (TCA) Protein Precipitation Kit (C510011) was from Sangon Biotech; Mouse IL-1β ELISA Kit (E-EL-M0037) was from Elabscience (Houston, TX, USA); Mouse IL-18 Quantikine ELISA Kit (ab216165) was from Abcam (Cambridge, UK); Mouse IL-6 ELISA Kit (abs520004) was from Absin Bioscience (Shanghai, China); Mouse TNF-α ELISA Kit (EK0527) was from Boster Biological Technology (Pleasanton, CA, USA); Polyethylenimine (24765) was from Polysciences, Inc. (Warrington, PA, USA); 4’,6-diamidino-2-phenylindole (DAPI) (D4054) was from UElandy (Suzhou, China).

Phospho-(Ser/Thr) Phe antibody (9631) was obtained from Cell Signaling Technology (Danvers, MA, USA); Anti-pan Phospho-Serine/Threonine Mouse mAb (AP1067) was from ABclonal (Woburn, MA, USA); anti-FLAG M2 (F1804), and Anti-NLRP6 (SAB1302240) were from Sigma-Aldrich; anti-NLRP3 (AG-20B-0014-C100), Anti-caspase-1 (AG-20B-0042-C100), and anti-ASC (AL177) were from AdipoGen Life Sciences (San Diego, USA); anti-NLRC4 (PA5-88997), Anti-caspase-1 (PA5-99390), and Anti-caspase-1 (PA5-119528) were from Invitrogen, Thermo Fisher Scientific (Waltham, MA, USA); anti-His (2365), anti-Myc (2276), anti-NLRP3 (15101), and anti-ASC (67824) were from Cell Signaling Technology; anti-IL-6 (ab259341) and anti-TNF-α (ab183218) were from Abcam; anti-NLRC4 (A7382), anti-β-Actin (AC004), was from ABclonal; Mouse/Rabbit IgG (H + L) secondary antibodies (C31430/C3146) were from Thermo Fisher Scientific; Dylight 594 Goat Anti-Mouse IgG (A23410) was from Abbkine (Wuhan, China).

Animals

In our previous study, NLRP6^−/− mice were generated using CRISPR/Cas9 gene editing technology [11]. Wild-type mice were sourced from the Ensiweier Company. The mice were housed under conditions that were free of pathogens, and their environment was subject to a strict 12 h light cycle. The study was approved by the Ethics Committee of our university (Approval No. IACUC-CQMU-2024-0636). All procedures involving animals were conducted in full compliance with all relevant ethical guidelines and regulations and in line with the globally recognized standards for the care and use of animals in laboratories.

Stereotaxic injection, and Middle Cerebral Artery Occlusion and Reperfusion (MCAO/R) surgery

For stereotaxic injection, NLRP6^−/− mice weighing approximately 12 g were anesthetized with 1.4% isoflurane and positioned prone in a stereotaxic instrument. Subsequently a volume of 1.5 µL containing either wild-type pAAV-CMV-NLRP6-3xFLAG-P2A-mCherry-tWPA virus (titer: 5 × 10¹² vg/mL) or the mutant variant pAAV-CMV-NLRP6 (S27A)-3xFLAG-P2A-mCherry-tWPA virus (same titer) was administered via injection. The targets included the left lateral ventricle (coordinates: 1.0 mm anterior to bregma, 0.3 mm left of the sagittal suture, 2.2 mm depth from skull surface) and two additional cortical locations (site 1: 1.3 mm anterior to bregma, 0.5 mm left of sagittal suture, 2.0 mm depth; site 2: 2.1 mm anterior to bregma, 0.3 mm left of sagittal suture, 2.0 mm depth). Infusion occurred at 0.05 µL/min via a microsyringe pump. Post-injection, the needle remained in place for 8 min to allow diffusion before withdrawal. To facilitate anesthetic recovery, animals were transferred to a temperature-regulated platform. Four weeks post-injection, the animals were subjected to MCAO/R modeling.

In the operation for MCAO/R, the mice were anesthetized with 1.4% isoflurane and placed on a heating pad. The procedure involved meticulous dissection and permanent ligation of the left common carotid artery in each subject. A 60-minute occlusion was then induced by advancing a nylon monofilament through the artery to its origin at the left middle cerebral artery. Following this ischemic period, reperfusion was established by filament withdrawal and maintained for either 24 h or 7 days. The sham-operated controls underwent identical surgical manipulations without filament insertion.

Assessment of cerebral infarct volume and neurological deficits

Cerebral infarct volume was assessed using 2, 3, 5-triphenyltetrazolium chloride (TTC) staining. In brief, the mice were euthanized while under deep anesthesia. Their brains were then cut into five sections, each 1.5 mm thick. These sections were placed in a solution of 2% TTC and kept at 37 °C for 15 min. Photographs of the brain sections were taken using a digital camera. ImageJ (version 1.53q) software was used to assess the infarct sizes. Finally, the adjusted cerebral infarct volume was calculated. This was done using the following formula: (volume of the contralateral hemisphere – non-infarcted portion of the ipsilateral hemisphere) / contralateral hemisphere volume x 100%. Neurological deficits were evaluated using the mNSS, which ranges from 0 to 18, with 0 representing no deficit and 18 being the most severe.

Hematoxylin and eosin, nissl, and fluoro-jade B staining

The procedures for Hematoxylin and Eosin, Nissl, and Fluoro-Jade B staining are detailed in our previously published research paper [11].

Behavioral tests

Before the tests, the mice were acclimatized to the experimental room for 1 h. Baseline performance was established through pre-testing before the formal assessment. At 24 h or 7 days post-MCAO/R surgery, three behavioral tests (open field, rotarod, and limb grip strength) were administered sequentially per protocol, with 2-hour inter-test intervals.

Cell culture

HEK293T, HEK293FT, HT22, and BV2 cells were sourced from our laboratory. BV2 cell lines stably expressing C-terminally 3×FLAG-tagged NLRP6 (called NLRP6^F/F BV2 cells) or containing NLRP6 genomic knockout (called NLRP6^−/− BV2 cells) were established in our previous study through CRISPR/Cas9-mediated genome editing (11). The culture conditions for all cell lines were as follows: incubation at 37 °C with 5% CO₂ in Dulbecco’s Modified Eagle Medium (DMEM) containing 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin.

Oxygen-Glucose Deprivation and Reoxygenation (OGD/R) model

Our previous report detailed the protocol for inducing oxygen-glucose deprivation/reoxygenation (OGD/R) in both BV2 monocultures and BV2-HT22 co-culture systems [11].

Immunoprecipitation and Immunoblot analysis

To detect NLRP6 phosphorylation, tissues or cells were disrupted using a NP-40 lysis solution containing 1.0% Nonidet P-40, 50 mM Tris-HCl (pH 7.4), 50 mM EDTA, and 150 mM sodium chloride (NaCl). The lysis buffer contained 0.1 mM PMSF and phosphatase inhibitors. After lysis, the samples underwent sonication followed by centrifugation (12,000 × g, 10 min, 4 °C). Supernatants were incubated overnight at 4 °C with NLRP6 primary antibody-bound Protein A/G magnetic beads on a rocker. After washing the magnetic beads three times with PBS, proteins attached to the beads were released with 60 µL of 1× sodium dodecyl sulfate (SDS) buffer and denatured by boiling. The phosphorylation level of NLRP6 was detected by immunoblot analysis using a phospho-(Ser/Thr) Phe antibody.

To examine protein interactions, the cells or tissue specimens were lysed in Enzyme Immunoassay buffer (250 mM NaCl, 1 mM EDTA, 50 mM HEPES (pH 7.4), 1 mM DTT, protease inhibitors, 0.1% NP-40), followed by sonication and centrifugation (12,000 × g, 10 min, 4 °C). The supernatants were rotated overnight at 4 °C with IgG/specific antibodies. The immunoprecipitants were captured using Protein A/G magnetic beads and boiled in 60 µL of 1× SDS loading buffer. Protein interactions were detected by immunoblot analysis using specific antibodies.

For immunoblot analysis, samples underwent RIPA buffer lysis, with subsequent protein quantification via a BCA assay. Electrophoretic separation of the samples was performed using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), followed by transfer to a polyvinylidene difluoride membrane. After the membranes were incubated with primary antibodies, detection was performed using HRP-linked secondary antibodies.

TCA protein precipitation

Cell supernatants (400 µL) were mixed with 100 µL Precipitation Solution A and vortexed for 10 s. The mixture was incubated on ice for 1 h, then centrifuged at 12,000 × g for 15 min at 4 °C to pellet proteins. Pellets were washed with 600 µL Wash Solution B (10-s vortex, 10 min on ice), followed by another centrifugation under the same conditions. After 30 min air-drying at room temperature, pellets were reconstituted in 20 µL Dissolution Solution C with 10-s vortexing. pH was adjusted by adding 5 µL Neutralization Solution D dropwise until the solution turned blue. Finally, samples were heated at 95 °C for 5 min to denature proteins and subsequently analyzed by immunoblotting.

Identification of NLRP6 phosphorylation using mass spectrometry

NLRP6^F/F BV2 cells endogenously expressing NLRP6-3xFLAG were used in the OGD/R model. Cells were lysed in a protease/phosphatase inhibitor cocktail. Following centrifugation of the lysates at 12,000 × g for 10 min (4 °C), NLRP6-3xFLAG protein was immunoprecipitated from the supernatant using anti-FLAG M2 agarose beads and subsequently resuspended in 1x SDS sample buffer. Protein samples were separated using 10% SDS-PAGE, followed by fixation and silver staining for visualization. NLRP6-3xFLAG-containing gel slices were excised, trypsin-digested, and analyzed by LC-MS/MS at Shanghai Bioprofile Technology Co., Ltd.

Plasmid construction

The coding sequences of NLRP6 (NM_133946.3) and pro-IL-1β (NM_008361.4) were PCR-amplified using mouse microglial cDNA as a template. These fragments were then inserted into the XbaI and XhoI restriction sites of the pLenti6-FLAG vector, resulting in FLAG-tagged NLRP6 and pro-IL-1β constructs, respectively. Similarly, the pro-caspase-1 gene (NM_009807.2) was amplified from mouse microglial cDNA and subsequently inserted into the pcDNA3.1-His vector using BamHI and XhoI restriction sites, producing a His-tagged pro-caspase-1 expression plasmid. Additionally, the ASC gene (NM_023258.4) was amplified from mouse microglial cell cDNA and cloned into the pLVX-IRES-Puro vector (restriction enzyme sites: EcoRI and XhoI). Expression plasmids encoding FLAG-epitope-tagged NLRP6 point mutants with S/T/Y to A/D substitutions S27A, S27D, S47A, S112A, S225A, T291A, T409A, T541A, Y658A, Y658D, S688A, and S698A were constructed by overlap PCR using the pLenti6-NLRP6-FLAG plasmid as the template. All constructs were sequence verified.

NLRP6 inflammasome reconstitution in HEK293T cells

A previous study has provided a comprehensive overview of the reconstitution process of the NLRP6 inflammasome in HEK293T cells [13]. Briefly, HEK293T cells were seeded into 10-cm culture dishes and cultured overnight. Transient transfection was performed using linear polyethylenimine with the following plasmids: ASC (0.3 µg/dish), His-pro-caspase-1 (0.33 µg/dish), and FLAG-pro-IL-1β (4.6 µg/dish), in the presence or absence of FLAG-NLRP6 or its point mutants (1.2 µg/dish). After 48 h, cells were harvested and lysed in RIPA buffer supplemented with protease inhibitors. IL-1β maturation in lysates and secretion in the supernatant were detected using immunoblot analysis and ELISA, respectively.

Expression of NLRP6 in NLRP6^−/− BV2 cells

NLRP6-deficient BV2 cells were transduced with lentiviral vectors expressing wild-type, S27A, or S27D mutant NLRP6. Lentiviruses were produced in HEK293FT cells via transient co-transfection with pLenti6-NLRP6-FLAG (wild-type or mutant), psPAX2, and pMD2.G at a 4:3:1 ratio. We collected the viral supernatant, passed it through a 0.45-µm filter, and supplemented it with polybrene (8 µg/mL) before transduction of NLRP6^−/− BV2 cells. Stable cell lines expressing wild-type NLRP6, NLRP6-S27A, or NLRP6-S27D were selected using 2 µg/mL puromycin.

ELISA

The culture media were collected, and the concentrations of IL-1β, IL-18, IL-6, and TNF-α in the supernatants were measured using quantitative ELISA kits according to the manufacturer’s instructions.

Self-association and oligomerization assay of NLRP6

pcDNA3.1-NLRP6 (WT)-FLAG plasmid and pcDNA3.1-NLRP6 (S27A)-Myc plasmid were equivalently transfected into HEK293T cells to assess the self-association capability of NLRP6. Co-immunoprecipitation was performed to detect protein interactions between NLRP6 (WT)-FLAG and NLRP6 (S27A)-Myc.

To assess NLRP6 oligomerization, we employed blue native polyacrylamide gel electrophoresis (BN-PAGE) following a standard protocol [14]. The cells were harvested and lysed in native-PAGE lysis buffer (20 mM Bis-tris, 20 mM NaCl, 1 mM phenylmethanesulfonyl fluoride, 1 × EDTA-free protease inhibitors, 0.5 mM sodium fluoride, 0.5 mM sodium orthovanadate, 500 mM ε-aminocaproic acid, 10% glycerol, and 0.5% digitonin [pH 7.0]) for 30 min on ice. Following centrifugation of the lysate (20,000 × g, 30 min, 4 °C), the supernatant was collected. This supernatant was then mixed with Coomassie Brilliant Blue G-250 at a final concentration of 0.25%. Subsequently, BN-PAGE (Beyotime) was performed on a 4–13% gradient gel. The gel was incubated in 1% SDS solution at room temperature for 20 min, transferred to a PVDF membrane, and subjected to standard immunoblotting assay detection. For SDS-PAGE analysis, the gel bands from native electrophoresis were soaked in 1× SDS loading buffer for 10 min, microwaved for 30 s, and shaken for 30 min. Finally, the samples were loaded into the wells of an 8% SDS-PAGE gel and subjected to routine immunoblotting.

Immunostaining for NLRP6 oligomer

Coverslips were placed in culture dishes, followed by plating of BV2 cells expressing NLRP6-FLAG, NLRP6-S27A-FLAG or NLRP6-S27D-FLAG. The cells were then subjected to OGD/R. The cells were fixed for 10 min with 4% paraformaldehyde and permeabilized for 15 min with 0.5% Triton X-100. The cells were blocked with 5% bovine serum albumin in PBS. The formation of NLRP6 oligomer was determined by NLRP6 immunostaining using anti-FLAG antibody and Dylight 594 Goat Anti-Mouse IgG. Nuclear staining was performed using DAPI, followed by confocal microscopic visualization. Quantitative analysis was conducted by calculating the percentage of cells expressing the NLRP6 oligomer.

ASC oligomer cross-linking

The cells were harvested and lysed in Triton Buffer (0.5% Triton X-100, 50 mM Tris-HCl [pH 7.5], 150 mM NaCl, and EDTA-free protease inhibitor cocktail). The lysates were centrifuged at 6000 × g for 15 min at 4 °C. The resulting pellets were rinsed three times with PBS, suspended in 500 µL of Triton buffer, and subsequently incubated with 2 mM disuccinimidyl suberate at 37 °C for 30 min to induce cross-linking. After centrifugation, the cross-linked pellets were solubilized in 1× SDS loading buffer and heated to denature the proteins prior to immunoblotting.

ASC speck formation

The procedures for ASC speck detection using immunofluorescence staining are detailed in our previously published research paper [11].

Calcein/PI, CCK-8, and LDH assays

Following a co-culture of HT22 and BV2 cells exposed to OGD/R treatment, the Transwell inserts containing BV2 cells and the culture media of HT22 cells were removed, as previously described [11]. HT22 cells were stained with 300 µL of Calcein-AM/PI working solution for 15 min to assess viability. Viable cells fluoresced green (Calcein-AM), whereas necrotic cells fluoresced red (PI). For CCK-8 assay, HT22 cells were exposed to 300 µL of CCK-8 solution and incubated for 20 min. Absorbance was quantified at 450 nm using a microplate reader. For LDH assay, culture media of HT22 cells in the co-culture OGD/R model were collected, and the concentration of LDH released into the medium was measured using a lactate dehydrogenase activity assay kit following the manufacturer’s protocol.

Statistical analysis

Statistical analyses were carried out with GraphPad Prism 8.0 (GraphPad Software, San Diego, CA, USA). Data normality was assessed using the Shapiro–Wilk test. Results following a normal distribution are expressed as mean ± SD. For comparisons among multiple groups, one-way ANOVA followed by Tukey’s post-hoc analysis was applied. Between-group differences, assuming homoscedasticity, were evaluated using Student’s t-test. Non-parametric data were subjected to Kruskal–Wallis analysis. A P-value < 0.05 was considered statistically significant.

Results

Phosphorylation of NLRP6 was required for its inflammasome activation in ischemic stroke

To delineate the NLRP6 inflammasome activation mechanisms in ischemic stroke, we examined its assembly and activation processes. Ischemic stroke mice and OGD/R-exposed NLRP6^F/F BV2 cells showed enhanced interaction between NLRP6-ASC and activation of caspase-1 as well as maturation of the cytokines IL-1β and IL-18 (Figs. 1 A-1F and S1A). The BV2 knock-in cell line with an endogenously flag-tagged NLRP6 was established in our previous study [11] and endogenously expressed NLRP6-3xFLAG and other inflammasome components, similar to their wild-type counterparts (Figs. S2A-S2D). Phosphorylation plays a pivotal role in the regulation of protein activity. Therefore, we explored the involvement of phosphorylation in NLRP6 inflammasome activation. Data from Phos-tag SDS-PAGE and denaturing IP experiments showed that NLRP6 was markedly phosphorylated in response to ischemic stroke (Figs. 1G and H, S3A, and S3B). NLRP6 phosphorylation coincided with its inflammasome activation, suggesting that NLRP6 phosphorylation may regulate its inflammasome activation. To validate this hypothesis, NLRP6^F/F BV2 cells subjected to OGD/R were treated with lambda phosphatase to dephosphorylate NLRP6. Consequently, the assembly of the NLRP6 inflammasome and cleavage of the inflammatory cytokines caspase-1, IL-1β, and IL-18 were inhibited when NLRP6 phosphorylation was decreased (Figs. 1I-K, and S4A). Furthermore, purified NLRP6 was incubated with ASC, pro-caspase-1, and pro-IL-1β, resulting in NLRP6 inflammasome assembly and IL-1β cleavage. In contrast, NLRP6 inflammasome assembly and IL-1β cleavage were reduced in the in vitro assay when purified NLRP6 was dephosphorylated by lambda phosphatase (Fig. 1L). These findings indicate that NLRP6 phosphorylation is positively correlated with inflammasome activation in ischemic stroke and that changes in the phosphorylation level of NLRP6 could affect the assembly and activation of the NLRP6 inflammasome.

Fig. 1.

NLRP6 phosphorylation is needed for NLRP6 inflammasome activation in ischemic stroke. ACo-IP was performed to assess the interaction between ASC and NLRP6 in MCAO/R. B Cleavage of the inflammatory cytokines caspase-1, IL-1β, and IL-18 was evaluated by immunoblotting in MCAO/R. CStatistical analysis of the cleavage efficiencies of caspase-1, IL-1β, and IL-18 in Fig. 1B. n=3. **P < 0.01. D The interaction between ASC and NLRP6-3xFLAG was examined viaCo-IP in NLRP6^F/F knock-in BV2 cells in OGD/R. E Cleavage of caspase-1, IL-1β, and IL-18 was detected using immunoblotting in NLRP6^F/F knock-in BV2 cells in OGD/R. F Statistical analysis of the cleavage efficiencies of caspase-1, IL-1β, and IL-18 in Fig. 1E. n=3. **P < 0.01, ***P < 0.001. G Immunoblotting of NLRP6 phosphorylation and expression by Phos-tag SDS-PAGE and Normal SDS-PAGE assays in MCAO/R. H Immunoblotting of NLRP6 phosphorylation and expression by IP denaturation and SDS-PAGE in NLRP6^F/F knock-in BV2 cells in OGD/R. I NLRP6-3xFLAG phosphorylation and its interaction with ASC were investigated in NLRP6^F/F knock-in BV2 cells treated with OGD/R and λppase, using IP under denaturing conditions and Co-IP, respectively. J Maturation of caspase-1, IL-1β, and IL-18 was assessed by immunoblotting in NLRP6^F/F knock-in BV2 cells exposed to OGD/R and λppase. K Statistical analysis of the cleavage efficiencies of caspase-1, IL-1β, and IL-18 in Fig. 1J. n=3. *P < 0.05, **P < 0.01. L In vitro incubation of purified mouse His-NLRP6 with or without λppase treatment, plus recombinant His-ASC, His-pro-caspase-1, and His-pro-IL-1β. The samples were analyzed using a Co-IP assay, immunoblotting, and Coomassie blue staining. The yellow background indicates the phosphorylated protein band of NLRP6. The data represent at least three independent experiments

Identification and screening of the phosphorylation sites in NLRP6

To identify the phosphorylation sites within NLRP6 that influence NLRP6 inflammasome activation, endogenous NLRP6-3xFLAG was immunoprecipitated from OGD/R-stimulated NLRP6^F/F BV2 cells. Protein separation was achieved by SDS-PAGE with subsequent silver staining and LC-MS/MS analysis (Figs. S5A and S5B). We selected 10 high-ranking phosphorylation sites within NLRP6 (S27, S47, S112, S225, T291, T409, T541, Y658, S688, and S698) for further verification (Fig. S5C). These candidate phosphorylation sites were sequentially located in the PYD, NACHT, and LRR domains of NLRP6 (Fig. S5D). Almost all phosphorylation sites are evolutionarily conserved in mammals (Fig. S5E).

To elucidate how these phosphorylation sites regulate NLRP6 inflammasome activation, we reconstituted the NLRP6 inflammasome in HEK293T cells via co-transfection of NLRP6, ASC, pro-caspase-1, and pro-IL-1β plasmids. IL-1β maturation or secretion is a key indicator of NLRP6 inflammasome activation in the HEK293T reconstitution system (Figs. 2A and B). The phosphorylation level of NLRP6 increased significantly in the reconstituted NLRP6 inflammasome system (Fig. S6A). Next, 10 candidate phosphorylated residues within NLRP6 were replaced with alanine, a non-phosphorylatable amino acid. The biological function of NLRP6 mutants in IL-1β maturation and secretion was assessed using an NLRP6 inflammasome-reconstituted HEK293T cell model. Our findings showed that NLRP6-S27A and NLRP6-Y658A mutants exhibited an evident defect in promoting IL-1β maturation and secretion, whereas the other NLRP6 mutants had the same promoting effect as wild-type NLRP6 (Figs. 2C and D, S7A, and S7B). To further validate the functional importance of the S27 and Y658 residues in NLRP6 inflammasome activation, these two residues were replaced with aspartic acid (S27D and Y658D) to mimic the constitutively phosphorylated state of these residues. As expected, the S27A mutant did not promote IL-1β maturation and secretion in the reconstituted NLRP6 inflammasome system, whereas the S27D mutant restored IL-1β maturation and secretion (Figs. 2E-H). However, compared with the wild-type, both Y658A and Y658D mutants exhibited lower activity in IL-1β maturation and secretion (Figs. 2E and F). Therefore, data from the reconstitution system suggest that NLRP6 Ser27 phosphorylation plays a vital role in NLRP6 inflammasome activation.

Fig. 2.

NLRP6 phosphorylation at the Ser27 residue is essential for NLRP6 inflammasome activation. A, B The NLRP6 inflammasome was reconstituted in HEK293T cells by co-expressing the indicated doses of FLAG-tagged NLRP6 in addition to ASC, His-pro-caspase-1, and FLAG-pro-IL-1β. NLRP6 inflammasome activity was evaluated by detecting the maturation of caspase-1 and IL-1β in cell lysates (A) and the secretion of IL-1β in the supernatant (B). n=4–9. C, D The NLRP6 inflammasome was reconstituted in HEK293T cells by co-expressing different NLRP6 non-phosphorylatable mutants with ASC, His-pro-caspase-1, and FLAG-pro-IL-1β. NLRP6 inflammasome activity was assessed by analyzing IL-1β maturation in cell lysates (C) and its secretion into the supernatant (D). n=5–9. E, F The NLRP6 inflammasome was reconstituted in HEK293T cells by co-expressing wild-type, non-phosphorylatable, or phosphomimetic NLRP6 with ASC, His-pro-caspase-1, and FLAG-pro-IL-1β. NLRP6 inflammasome activity was assessed by analyzing IL-1β maturation in cell lysates (E) and its secretion into the supernatant (F). n=4-9. G, H The NLRP6 inflammasome was reconstituted in HEK293T cells by co-expressing the indicated doses of wild-type NLRP6, NLRP6-S27A, or NLRP6-S27D with ASC, His-pro-caspase-1, and FLAG-pro-IL-1β. NLRP6 inflammasome activity was assessed by analyzing IL-1β maturation in cell lysates (G) and its secretion into the supernatant (H). n=3–9. Asterisks indicate non-specific bands. NS, no statistical difference; ***P < 0.001, ****P < 0.0001

Ser27 phosphorylation promoted NLRP6 oligomerization and inflammasome activation

To investigate the biological significance of Ser27 phosphorylation in NLRP6 inflammasome activation under physiological conditions, LTA, a component of the surface of gram-positive bacteria, was used to stimulate NLRP6^F/F BV2 cells to specifically activate the NLRP6 inflammasome [15, 16]. LTA treatment of NLRP6^F/F BV2 cells enhanced NLRP6-3xFlag phosphorylation, promoted its interaction with ASC, and stimulated the maturation and release of both caspase-1 and IL-1β (Fig. 3A and B). However, the assembly of NLRP3 or NLRC4 inflammasomes in NLRP6^F/F BV2 cells was unaffected by LTA stimulation (Figs. S8A and S8B). Subsequently, wild-type NLRP6, NLRP6-S27A, or NLRP6-S27D was stably re-expressed in NLRP6^−/− BV2 cells generated using CRISPR/Cas9 technology in our previous study [11]. LTA treatment elicited robust NLRP6 phosphorylation and inflammasome activation in BV2 cells expressing NLRP6-WT or NLRP6-S27D, but not NLRP6-S27A, as reflected by the phosphorylation level of NLRP6-Flag (Fig. 3E), binding of NLRP6-Flag and ASC (Fig. 3C), and secretion of caspase-1 and IL-1β (Fig. 3D). Furthermore, LPS/nigericin-induced NLRP3 and LPS/flagellin-triggered NLRC4 inflammasomes activation remained intact in NLRP6-mutated BV2 cells (Figs. S9A and S9B). These results indicate that Ser27 phosphorylation of NLRP6 is essential for NLRP6 inflammasome activation.

Fig. 3.

Ser27 phosphorylation promotes NLRP6 oligomerization and inflammasome activation. A Immunoblotting was used to assess IL-1β maturation in cell lysates and the release of caspase-1 and IL-1β into the supernatant of NLRP6^F/F knock-in BV2 cells stimulated with LTA at the indicated concentrations. Equal loading of supernatants was verified by Coomassie blue staining of total protein. B Denaturing IP was performed to analyze NLRP6-3xFLAG phosphorylation, while Co-IP was carried out to examine the interaction between NLRP6-3xFLAG and ASC in NLRP6^F/F knock-in BV2 cells treated with 20 µg/mL LTA. C Co-IP was carried out to examine the interaction between NLRP6-FLAG and ASC in BV2 cells re-expressing wild-type NLRP6, NLRP6-S27A, or NLRP6-S27D following stimulation with 20 µg/mL LTA. D Immunoblotting was used to assess the levels of caspase-1 and IL-1β in the supernatants from BV2 cells re-expressing wild-type NLRP6, NLRP6-S27A, or NLRP6-S27D following stimulation with 20 µg/mL LTA. Equal loading of supernatants was verified by Coomassie blue staining of total protein. E IP denaturing experiment for NLRP6-FLAG phosphorylation in BV2 cells re-expressing wild-type NLRP6, NLRP6-S27A, or NLRP6-S27D following stimulation with 20 µg/mL LTA. F Co-IP analysis for self-association of NLRP6 from HEK293T cells transfected with NLRP6-Flag and NLRP6-Myc or the NLRP6-S27A-Myc mutant. G Representative fluorescence images of NLRP6 oligomerization in BV2 cells re-expressing wild-type NLRP6, NLRP6-S27A, or NLRP6-S27D following stimulation with 20 µg/mL LTA (scale bars = 50 μm). Red puncta indicate NLRP6 oligomerization. H Statistical analysis of the percentage of cells with NLRP6 puncta. n=6. ***P < 0.001, ****P < 0.0001. I Immunoblotting was used to assess NLRP6 oligomerization and expression by BN-PAGE and SDS-PAGE in BV2 cells re-expressing wild-type NLRP6, NLRP6-S27A, or NLRP6-S27D, following stimulation with 20 µg/mL LTA. Asterisks indicate non-specific bands. The data represent at least three independent experiments

Evidence from previous studies has indicated that the NLRP6 PYD domain self-oligomerizes into filamentous structures, forming an oligomeric assembly essential for NLRP6 inflammasome assembly and activation [17]. In the present study, the Ser27 site is located within the PYD domain of NLRP6 (Fig. S5D). Therefore, the impact of Ser27 phosphorylation on NLRP6 oligomerization was assessed. Co-immunoprecipitation results showed that the interaction between NLRP6-Flag and NLRP6-Myc was significantly impaired when the S27A mutant was introduced (Fig. 3F). Moreover, immunofluorescence imaging showed that S27A NLRP6 in LTA-stimulated BV2 cells exhibited diminished puncta formation compared to wild-type or S27D NLRP6 cells (Fig. 3G and H). Consistent with this observation, oligomer formation of wild-type NLRP6 and the S27D mutant was clearly detected by blue native-PAGE (a widely used method for detecting large protein complexes) in LTA-treated BV2 cells, whereas the S27A mutation effectively suppressed this oligomerization process (Fig. 3I). Collectively, these results indicate that NLRP6 Ser27 phosphorylation enhances NLRP6 oligomerization and subsequent inflammasome activation.

NLRP6 Ser27 phosphorylation aggravated neuroinflammation and neuronal injury in ischemic stroke in vitro

To evaluate the impact of NLRP6 Ser27 phosphorylation on neuroinflammation in ischemic stroke, a rescue model was created by re-expressing NLRP6 in NLRP6^−/− microglial BV2 cells, which were then subjected to OGD/R. Consistent with the data acquired from LTA stimulation, NLRP6 phosphorylation, NLRP6 oligomerization, assembly of the NLRP6 inflammasome, and secretion of caspase-1 and IL-1β induced by OGD/R treatment were distinctly inhibited in NLRP6-S27A-rescued microglial BV2 cells compared with those rescued with either NLRP6-WT or the NLRP6-S27D mutant (Figs. 4A-C and S10A). Furthermore, ASC oligomerization and speck formation, a critical intermediate step in NLRP6 inflammasome activation [18], were also significantly reduced in BV2 cells expressing NLRP6-S27A, compared with cells expressing WT or S27D NLRP6 (Fig. 4D-F). However, the secretion of inflammasome-independent cytokines IL-6 and TNF-α was not influenced by NLRP6 Ser27 phosphorylation in BV2 cells under OGD/R conditions (Figs. S11A and S11B). These findings indicate that Ser27 phosphorylation is critical for NLRP6 inflammasome activation and subsequent inflammasome-dependent neuroinflammation in ischemic stroke in vitro.

Fig. 4.

NLRP6 Ser27 phosphorylation aggravates neuroinflammation and neuronal injury in ischemic stroke in vitro. A IP denaturation for NLRP6-FLAG phosphorylation in BV2 cells re-expressing wild-type NLRP6, NLRP6-S27A, or NLRP6-S27D following stimulation with OGD/R. B Co-IP was carried out to examine the interaction between NLRP6-FLAG and ASC in BV2 cells re-expressing wild-type NLRP6, NLRP6-S27A, or NLRP6-S27D exposed to OGD/R. C Immunoblotting was used to assess the levels of caspase-1 and IL-1β in the supernatants from BV2 cells re-expressing wild-type NLRP6, NLRP6-S27A, or NLRP6-S27D exposed to OGD/R. Equal loading of supernatants was verified by Coomassie blue staining of total protein. D Representative images of ASC specks in BV2 cells re-expressing wild-type NLRP6, NLRP6-S27A, or NLRP6-S27D exposed to OGD/R (scale bars=100 μm). White arrows indicate ASC specks within the cell. E Statistical analysis of the proportion of ASC speck-positive cells in the ASC speck assay. n=3. *P < 0.05, ***P < 0.001. F Immunoblotting was used to assess ASC oligomerization in BV2 cells re-expressing wild-type NLRP6, NLRP6-S27A, or NLRP6-S27D exposed to OGD/R. Cells were cross-linked with DSS reagent, and the insoluble pellet was subjected to immunoblotting with an anti-ASC antibody (top). Immunoblotting using cell lysates is included to show ASC expression (bottom). G Schematic diagram illustrating the co-culture protocol of NLRP6-reconstituted BV2 microglial cells with HT22 neuronal cells. H Survival of HT22 cells co-cultured with BV2 cells reconstituted with wild-type NLRP6, NLRP6-S27A, or NLRP6-S27D under OGD/R conditions was evaluated by Calcein-AM/PI staining (scale bars = 100 μm). Red and green fluorescence indicate dead and surviving HT22 cells, respectively. I Statistical analysis of HT22 cell death rates from the Calcein-AM/PI assay. n=6. **P < 0.01, ***P < 0.001. J Viability of HT22 cells in co-culture with BV2 cells re-expressing wild-type or mutant NLRP6 under OGD/R conditions was measured using the CCK-8 assay. n = 9. NS, not significant; **P < 0.01. K Damage of HT22 cells in co-culture with BV2 cells re-expressing wild-type or mutant NLRP6 under OGD/R conditions was measured using the LDH assay. n = 9. ***P < 0.0001. The data represent at least three independent experiments

To further study the function of NLRP6 Ser27 phosphorylation in neuronal survival, NLRP6-rescued BV2 cells and HT22 neurons were co-cultured (Fig. 4G). The Calcein-AM/PI assay showed that HT22 cell death following OGD/R was reduced upon co-culture with NLRP6^−/− BV2 cells re-expressing the S27A NLRP6, compared with those cells re-expressing either wild-type or S27D NLRP6 (Fig. 4H and I). Consistently, CCK8 and LDH assays showed that co-culture with the NLRP6^−/− BV2 cells re-expressing the S27A NLRP6 conferred higher OGD/R resistance to HT22 cells than co-culture with those cells carrying wild-type or S27D NLRP6 (Fig. 4J and K). These findings indicate that phosphorylation of NLRP6 at Ser27 in BV2 cells enhances the cytotoxic effects exerted by BV2 microglial cells on HT22 neuronal cells. To investigate how NLRP6 Ser27 phosphorylation regulates microglial survival, we subjected NLRP6^−/− BV2 cells re-expressing wild-type NLRP6, NLRP6-S27A, or NLRP6-S27D to monoculture and OGD/R treatment. CCK8 analysis revealed that NLRP6^−/− BV2 cells re-expressing either wild-type NLRP6 or the S27D mutant exhibited lower viability under OGD/R conditions than cells expressed with the S27A mutant (Fig. S12A).

Thus, these results suggest that Ser27 phosphorylation of NLRP6 promotes its inflammasome activation and neuroinflammation and aggravates neuronal injury in ischemic stroke in vitro.

Ser27 phosphorylation of NLRP6 facilitated its inflammasome activation and exacerbated neurological deficit in ischemic stroke mouse model

Our previous study [11] generated NLRP6-deficient mice, and its knockout was reconfirmed through PCR and Western blot analyses (Fig. S13A-S13C). An adeno-associated virus (AAV) expressing wild-type NLRP6 or the NLRP6-S27A mutant was microinjected into the cerebral cortex of NLRP6-deficient mice to consolidate the physiological function of NLRP6 Ser27 phosphorylation in ischemic stroke in vivo (Fig. 5A). Infection efficiency was evaluated in brain sections by immunofluorescence detection of mCherry-tagged pAAV (Fig. S14A). NLRP6-deficient mice treated with pAAV showed expression levels of NLRP6, ASC, pro-caspase-1, and pro-IL-1β that were comparable to those in wild-type mice (Fig. S14B). Thus, supported by these results, we generated Nlrp6^−/− mice with AAV-mediated cortical re-expression of WT NLRP6 or the NLRP6-S27A mutant. Subsequently, an MCAO/R model was created. To confirm successful ischemia and ensure the absence of confounding vascular variations, regional cerebral blood flow was documented before ischemia, during occlusion, and after reperfusion (Fig. 5B). After ischemic attack, NLRP6 phosphorylation was reduced, the interaction between NLRP6 and ASC was inhibited, and the cleavage efficiencies of caspase-1, IL-1β, and IL-18 was decreased in the cortex of Nlrp6^−/− mice re-expressing NLRP6-S27A compared with those re-expressing WT NLRP6 (Figs. 5C and D and S18A). However, IL-6 and TNF-α production remained unaffected (Fig. S15A). These findings align with cellular observations, indicating that Ser27 phosphorylation of NLRP6 enhances its inflammasome activation and inflammasome-dependent neuroinflammation in mice with ischemic stroke.

Fig. 5.

Ser27 phosphorylation of NLRP6 facilitates its inflammasome activation and exacerbates neurological deficits in mice with ischemic stroke. A Diagram of the experimental strategy used for phenotypic validation of mice with ischemic stroke. B Regional cerebral blood flow before, during, and after cerebral ischemia. n=5. C IP denaturation for NLRP6-FLAG phosphorylation and Co-IP for the interaction between NLRP6-FLAG and ASC in ischemic stroke mice expressing the reconstituted wild-type NLRP6 or NLRP6-S27A mutan. The yellow background indicates the phosphorylated protein band of NLRP6. n=5. D Immunoblotting was used to assess the levels of caspase-1, IL-1β and IL-18 in ischemic stroke mice expressing reconstituted wild-type NLRP6 or NLRP6-S27A mutant. n=5. E, F Representative images and quantitative analysis of infarct volume TTC staining in ischemic stroke mice expressing reconstituted wild-type NLRP6 or the NLRP6-S27A mutant. n=6. G Neurological severity scores in ischemic stroke mice expressing reconstituted wild-type NLRP6 or NLRP6-S27A mutant. n=9. H Representative images of HE and Nissl staining of the morphology and structure in the peri-infarct cortex from ischemic stroke mice expressing a reconstituted wild-type NLRP6 or NLRP6-S27A mutant (scale bars = 50 μm). n=6. I, J Representative images and quantitative analysis of FJB immunostaining of denatured neurons in the peri-infarct cortex from ischemic stroke mice expressing a reconstituted wild-type NLRP6 or NLRP6-S27A mutant (scale bars = 50 μm). n=5. K-O Open field, rotarod, and grip strength tests in ischemic stroke mice expressing a reconstituted wild-type NLRP6 or NLRP6-S27A mutant. n=7-9. P, Q Representative images and quantitative analysis of TTC staining of infarct volume from mice expressing reconstituted wild-type NLRP6 or NLRP6-S27A mutant 7 days after ischemic stroke. n=6. R Neurological severity scores of mice expressing reconstituted wild-type NLRP6 or NLRP6-S27A mutant 7 days after ischemic stroke. n=9. S Survival percentage of mice expressing reconstituted wild-type NLRP6 or NLRP6-S27A mutant 7 days after ischemic stroke. n=13–15. NS, no statistical difference; **P < 0.01, ***P < 0.001

In terms of short-term brain injury, NLRP6-deficient mice re-expressing the NLRP6-S27A mutant displayed reduced cerebral infarct volume as indicated by TTC staining, ameliorated neurological functional deficits as indicated by lower scores in the mNSS test, better-preserved cortical morphology and structure as confirmed by HE and Nissl staining, and a reduced number of degenerating neurons in the peri-infarct region as demonstrated by FJB immunofluorescence after ischemia, compared with Nlrp6^−/− mice re-expressing WT NLRP6 (Fig. 5E-J). In short-term sensorimotor dysfunction, Nlrp6^−/− mice re-expressing NLRP6-S27A exhibited increased distance and speed of motion and non-differential number of bouts through the center area and percentage of duration in the center in the open field test, prolonged rotarod latency, and increased grip strength relative to mice reconstituted with wild-type NLRP6 following ischemic stroke in behavioral experiments (Figs. 5K-O, S16A, and S16B). For long-term functional outcomes, Nlrp6^−/− mice re-expressing NLRP6-S27A were more resistant to neurological deficits than the other groups at 7 days post-ischemia, as revealed by the lower ischemic volume, reduced neurological functional score, increased rotarod time, and enhanced muscle strength (Figs. 5P-R, S17A, and S17B). Moreover, the NLRP6-S27A re-expressing group demonstrated enhanced survival outcomes compared with the mice re-expressing wild-type NLRP6 (Fig. 5S).

Based on these findings, NLRP6 Ser27 phosphorylation contribute, at least in part, to inflammasome-related neuroinflammation and neurological injury after ischemic stroke.

Discussion

Accumulating evidence has implicated NLRP6 in the development of multiple diseases, such as inflammatory conditions [19–22] and neurodegenerative diseases [23]. Consistent with these broader roles, our previous study demonstrated that NLRP6 assembles an inflammasome with the adaptor ASC and pro-caspase-1 during ischemic stroke, resulting in proinflammatory cytokine release, neuronal death, and brain injury [11]. In this study, we demonstrated that NLRP6 is significantly phosphorylated during ischemic stroke. A phospho-deficient Ser27Ala (S27A) mutation hindered NLRP6 self-association and oligomerization, thereby preventing NLRP6 inflammasome assembly and activation. Consistently, Ser27 dephosphorylation mitigated cerebral ischemic injury in vitro and in vivo by dampening NLRP6 inflammasome-driven neuroinflammation. Thus, our findings show that the phosphorylation of NLRP6 at Ser27 functions as an accelerating mechanism for NLRP6 inflammasome activation and pathophysiological progression in ischemic stroke.

NLRP6, a less-studied NLR, serves as a cytosolic innate immune sensor that recognizes both exogenous PAMPs and endogenous DAMPs, modulating inflammatory responses in a manner specific to the cell type and context [24, 25]. NLRP6 comprises the N-terminal pyrin domain (PYD), NACHT domain, and C-terminal LRR domain. Under basal conditions, NLRP6 adopts an auto-inhibited state via LRR–NACHT interactions. Upon stimulation, LRR disengagement from NACHT permits an activating conformational change in NLRP6 [26]. Subsequently, NLRP6’s PYD domain forms an oligomeric filament that connects with ASC via homotypic PYD–PYD binding and recruits pro-caspase-1 to form the NLRP6 inflammasome [24, 26, 27]. NACHT-mediated oligomerization coupled to PYD filament assembly facilitates inflammasome formation [26]. Following the assembly of the NLRP6 inflammasome, pro-caspase-1 undergoes self-cleavage to become activated caspase-1. This active enzyme then processes inactive precursor cytokines, including pro-IL-1β and pro-IL-18, into their mature forms, ultimately inducing an inflammatory response. Alternatively, active caspase-1 also cleaves gasdermin D, releasing the N-terminal fragment that forms plasma-membrane pores, promoting cytokine release and pyroptosis [26–28].

Pathogenic ligands modulate NLRP6 inflammasome activation in response to microbial infection. For example, LTA, a surface component of gram-positive Listeria monocytogenes, binds directly to the LRR domain of the NLRP6 monomer with high affinity and induces conformational change and oligomerization, thus triggering the assembly of the NLRP6 inflammasome, which in turn leads to IL-18 maturation and enhances intestinal infection [15]. In hepatitis or rotavirus infection, NLRP6 recognizes viral double-stranded RNA to form liquid-like condensates; ASC recruitment then solidifies these condensates, promoting inflammasome activation and restricting intestinal viral replication [16]. In addition, taurine promotes the assembly of the NLRP6 inflammasome, whereas histamine and spermine inhibit its formation [29]. Collectively, these data define ligand- and metabolite-dependent control of NLRP6 assembly in infection. Beyond responding to microbial infections, aberrant activation of the NLRP6 inflammasome is implicated in the pathophysiology of various sterile inflammatory diseases, such as chronic obstructive pulmonary disease [30], alcoholic hepatitis [31], ulcerative colitis [32], and nicotine-induced nephrotoxicity [33]. In ischemic stroke, the deleterious role of the NLRP6 inflammasome in neuroinflammation and brain injury was demonstrated previously [11]. However, the mechanism regulating NLRP6 inflammasome assembly in sterile inflammatory diseases, particularly ischemic stroke, remains unclear. Given the sterility of the inflammatory cascades in cerebral ischemia, microbial ligands or metabolites that directly or indirectly regulate the assembly of NLRP6 inflammasomes are not suitable for regulating NLRP6 inflammasome activation in ischemic stroke. Further studies are required to explore the precise mechanisms underlying NLRP6 inflammasome activation during ischemic stroke.

Post-translational modifications covalently alter proteins to regulate conformation, localization, stability, and activity [34–36]. Previous research has demonstrated that ubiquitination, a widespread post-translational modification, is essential for the regulation of NLRP6 inflammasome activation. Specifically, K63-linked ubiquitination of the NLRP6 monomer promotes inflammasome formation in mice infected with Citrobacter rodentium by enhancing the interaction between NLRP6 and adaptor ASC. Deubiquitinase Cyld inhibits inflammasome activation by cleaving K63-linked ubiquitination of the NLRP6 monomer and prevents excessive IL-18 production, thus limiting severe intestinal inflammation [12]. These findings indicate that modulating NLRP6 deubiquitination by targeting Cyld may influence NLRP6 activation; however, the lysine acceptor site(s) for K63-linked ubiquitination remain undefined. Building on this principle in ischemic stroke, a previous study showed that the deubiquitinase BRCC3 promotes NLRP6 inflammasome assembly by binding to NLRP6 and removing its ubiquitin chains, thereby aggravating inflammation, pyroptosis, and neurological deficits [37]. The results of the BRCC3 study differ from those of the CYLD study, likely due to differences in the types of ubiquitin chains and lysine sites involved in NLRP6 ubiquitination. Previous studies have highlighted the importance of post-translational modifications in regulating NLRP6 inflammasome activation. Phosphorylation, a classical post-translational modification, strongly influences protein conformation and activity, and thus likely plays a role in NLRP6 inflammasome regulation in ischemic stroke. In this study, we observed marked phosphorylation of NLRP6 after ischemic stroke, and phosphorylation of the NLRP6 was essential for inflammasome assembly and activity. Molecular experiments revealed that NLRP6 Ser27 phosphorylation promotes inflammasome assembly and activation by increasing self-association and oligomerization of the NLRP6 monomer. Phenotypic experiments further revealed that NLRP6 Ser27 phosphorylation aggravates neuroinflammation, neuronal death, and neurological deficits by enhancing NLRP6 inflammasome activation. Together, these findings define the role and mechanism of NLRP6 Ser27 phosphorylation in NLRP6 inflammasome activation and neurological damage in ischemic stroke. Future studies should identify the kinases or phosphatases responsible for NLRP6 Ser27 phosphorylation in ischemic stroke.

The primary limitation of this study lies in the unavailability of a phosphorylation-specific antibody targeting the Ser27 residue of NLRP6. Although we endeavored to generate a custom phosphorylation-specific antibody against this site, our efforts ultimately proved unsuccessful. Second, the findings from the core cellular phenotypic experiments in this study would be substantially more reliable and generalizable if validated in primary neurons and microglia. Furthermore, given that sex dimorphism may exist in the mechanisms, phenotypes, and therapeutic responses of neurodegenerative diseases, and that this study exclusively used male mice for in vivo experiments, the applicability of our phenotypic and molecular findings to female mice remains unclear. Therefore, the absence of relevant experimental data from female mice constitutes another limitation of this study.

Conclusions

Our study reveals a novel mechanism by which NLRP6 inflammasome activation influences neuroinflammation and neurological damage in ischemic stroke. These findings indicate that modulating NLRP6 inflammasome activation by targeting NLRP6 Ser27 phosphorylation may represent a potential therapeutic strategy for treating ischemic stroke.

Abbreviations

DAMPs

Damage-associated molecular patterns

NLRP6

Nucleotide-binding oligomerization domain-like receptor pyrin domain-containing 6

NLRs

NOD-like receptors

ASC

Apoptosis-associated speck-like protein containing a CARD

LTA

Lipoteichoic acid

LPS

Lipopolysaccharide

PI

Propidium Iodide

TCA

Trichloroacetic acid

TTC

5-triphenyltetrazolium chloride

DMEM

Dulbecco’s Modified Eagle Medium

FBS

Fetal bovine serum

OGD/R

Oxygen-glucose deprivation/reoxygenation

NaCl

Sodium chloride

SDS

Sodium dodecyl sulfate

SDS-PAGE

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis

BN-PAGE

Blue native polyacrylamide gel electrophoresis

AAV

Adeno-associated virus

S27A

Ser27Ala

PYD

Pyrin domain

Authors’ contributions

JZ and QH designed and supervised the whole project; Tan performed the main experiments and analyzed the data; QH supplementally performed the experiments, wrote the manuscript and made the figures; MW, KH and Tang raised the animals, data analysis and exploration of results discussions; YZ and JL assisted in data analysis, editing, and revising manuscript; MX and XZ provided help for the revision experiments; Tan conducted all revision experiments and prepared the point-by-point response.

Funding

This work was supported by the National Natural Science Foundation of China (82472230) and the Youth Fund of the National Natural Science Foundation of China (82302474), the Chongqing Graduate Research and Innovation Project (CYS23318), and the Natural Science Foundation of Chongqing Science and Technology Committee (CSTB2024NSCQ-MSX0156 and CSTB2025NSCQ-GPX1243) .

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.