Loss of DJ-1 alleviates microglia-mediated neuroinflammation in Parkinson’s disease via autophagy-lysosomal degradation of NLRP3

The following reagents were used: 3-Methyladenine (3-MA, M-9281), Lipopolysaccharide (LPS, L4516), 1-Methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP, M-0896), Nigericin (N-7143), and Poly-L-lysine (P0296) from Sigma-Aldrich; Fetal bovine serum (FBS, 10437028), B27 Supplement (17504044), Neurobasal Medium (21103049), and penicillin-streptomycin (15640055) from Gibco; Lipofectamine 3000 Transfection Reagent (L3000-015), Lipofectamine RNAiMAX Transfection Reagent (13778150), Pierce Protease and Phosphatase Inhibitor Mini Tablets, EDTA-free (A32961) from Thermo Fisher Scientific; Poly (dA:dT)/LyoVec™ (tlrl-patc), and Ultrapure Flagellin (FLA-ST, tlrl-epstfla-5) from Invivogen; DOTAP (HY-112754A) from MedChem Express; Adenosine triphosphate (ATP, A8270) and additional ATP (A8270) from Beijing Solarbio Life Sciences; Rapamycin (53-23-88-9) from Gene Operation.

The primary antibodies used in western blot and confocal microscopy in our present study are listed in Table 1.

Male C57BL/6J mice (2–3 months old, 25–30 g) were obtained from the Animal Core Facility of Nanjing Medical University (Nanjing, China). All animal procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of Nanjing Medical University. Mice were housed under controlled conditions (50-60% humidity, 12-h light/dark cycle) and allowed to acclimate for one week prior to experiments. All mice were specific pathogen-free (SPF).

Primary microglia were isolated from the brains of postnatal day 1-3 (P1-P3) C57BL/6J mice. Briefly, brains were dissected, meninges and large blood vessels were mechanically removed. Brain tissue was dissociated using 0.25% (w/v) trypsin for 5 min at room temperature, followed by filtration through a 100-µm strainer. After centrifugation (1500 × g, 10 min, RT), cells were plated in poly-L-lysine-precoated flasks containing DMEM/F-12 medium supplemented with 10% (v/v) FBS and 1% (v/v) penicillin-streptomycin. Cultures were maintained at 37°C in a humidified 5% CO₂ incubator, with medium changes every 5 days. After 12–14 days, microglia were detached by shaking, collected by centrifugation (1500 × g, 10 min, RT), and plated on poly-L-lysine-precoated 24-well plates at a density of 1.9 × 10⁵ cells/ml in complete medium.

Prior to treatment, primary microglia were transfected with DJ-1-specific siRNA or negative control siRNA. Cells were then stimulated with LPS (100 ng/ml) for 8 h followed by ATP (5 mM) for 0.5 h. Cell lysates were analyzed by Quantitative Real-Time PCR (qRT-PCR), western blotting, and co-immunoprecipitation (Co-IP). Cell culture supernatants were assessed for cytokine levels by ELISA.

BMDMs were isolated from the femurs and tibias of 2-3-month-old male C57BL/6J mice. Bone marrow was flushed with medium. Cells were plated on poly-L-lysine-precoated 24-well plates at 2 × 10⁶ cells/ml in DMEM medium supplemented with 10% (v/v) FBS, 1% (v/v) penicillin-streptomycin, and recombinant mouse M-CSF (0.1‰ w/v, Peprotech, AF-315-03). Cultures were maintained at 37°C in 5% CO₂. Medium was replaced after 24 h and subsequently every 3 days. BMDMs were used for experiments after 7 days of differentiation.

Before treatment, BMDMs were transfected with si-DJ-1 or control siRNA. Cells were stimulated with LPS (100 ng/ml) 7.5h plus ATP (5 mM) 0.5h, nigericin (10 µM), flagellin (100 ng/ml) 16h, or poly(dA:dT) (2 µg/ml) 14h. Flagellin was incubated with DOTAP at a 1:3 ratio in culture medium for 20 minutes prior to addition to the cells.

Cell lysates were analyzed by western blotting, and supernatants were analyzed by ELISA.

Adenoviral vectors expressing DJ-1 siRNA (si-DJ-1: sense 5'-ACCUUGCUAGUAGAAUAAACAGU-3', antisense 5'- ACUGUUUAUUCUACUAGCAAGGU -3') or control siRNA were obtained from GenePharma. After adherence, primary microglia and BMDMs were transfected with siRNA (200 nM) using Lipofectamine RNAiMAX in serum-free Opti-MEM medium according to the manufacturer's instructions. Following an 8-h transfection period, the medium was replaced with complete medium, and cells were incubated for a total of 48 h before subsequent treatments.

For NLRP3 plasmid transfection, HEK-293 cells and DJ-1 knockout HEK-293 cells were plated in 6-well plates at 6 × 10⁵ cells/well. Cells were transfected with NLRP3 plasmid using Lipofectamine 3000. After 8 h, the medium was replaced, and cells were treated as indicated. For pharmacological studies, the autophagy inducer rapamycin (10 nM) or the autophagy inhibitor 3-MA (5 mM) was added to the culture medium for 13 h after plasmid transfection.

Primary cortical neurons were isolated from C57BL/6J mouse embryos at embryonic day 13-16 (E13-E16). Brains were dissected, meninges and blood vessels removed, and tissue dissociated using 0.025% (w/v) trypsin for 30 min at 37°C. The suspension was filtered through a 40-µm strainer. Cells were plated on poly-L-lysine-precoated 24-well or 96-well plates in Neurobasal medium supplemented with 0.5 mM glutamine, 2% (v/v) B27, and 1% (v/v) penicillin-streptomycin. Cultures were maintained at 37°C in 5% CO₂. Half of the medium was replaced every 3 days. Neurons were used for experiments after 7 days in vitro.

Conditioned medium (CM) was collected from treated primary microglia, centrifuged (15,000 × g, 15 min), and the supernatant was mixed 1:1 with fresh Neurobasal medium. Primary neurons in 96-well plates were incubated with this mixture for 6 h prior to the tyramide signal amplification multiplex staining, western blotting and cell viability assay.

Following 6-h incubation with microglial CM, neuronal viability was assessed using the CCK-8 (Bimake, B34304). Briefly, 200 μL CCK-8 reagents diluted at 1:10 in DMEM were added to each well. Plates were incubated under light-protected conditions at 37°C for 2–4 h. Absorbance was measured at 450 nm using a microplate reader (Varioskan Flash, Thermo Fisher). Cell viability was calculated relative to the control wells.

Mice were anesthetized with sodium pentobarbital (40 mg/kg, i.p.) and secured in a stereotaxic frame. To evaluate the neuroprotective role of DJ-1, mice were randomly assigned to six groups (n=6-10/group): AAV-Vector + Saline, AAV-Vector + LPS, AAV-shDJ-1 + Saline, AAV-shDJ-1 + LPS; AAV-Vector + MPTP/p, AAV-shDJ-1 + MPTP/p. shDJ-1 target: sense 5'-ACCUUGCUAGUAGAAUAAACAGU-3', antisense 5'- ACUGUUUAUUCUACUAGCAAGGU-3'. AAV-shDJ-1-F4/80p-MCS-SV40 PolyA (GenePharma) virus was bilaterally injected (1.5 μL for each side) into the substantia nigra pars compacta (SNc; coordinates relative to bregma: AP: -3.0 mm, ML: ± 1.2 mm, DV: -4.3 mm). Three weeks after viral expression, freshly prepared LPS (0.5 µg) was stereotactically injected at the same site. The surgical site was disinfected with iodophor, and mice recovered on a heating pad. Mice were sacrificed 7 days after LPS injection for subsequent analyses.

For the subacute MPTP/p Parkinson's disease model, mice received subcutaneous injections of MPTP (25 mg/kg) three weeks after viral expression. MPTP/p was administered once daily for 5 consecutive days and were sacrificed 4 days after the last injection for subsequent analyses.

For the locomotor activity test, mice were placed in an activity monitor chamber (50 cm × 50 cm × 40 cm) to acclimatize for 15 min before the start of the test. The track of the mice was recorded for 10 min, and the speed was calculated using the TopScan Realtime Option Version 2.00 behavioral detection system and the supporting camera.

For the rotarod test, the mice were placed on a rod with a diameter of 30 mm and set the program to a low speed of 5 rpm/min, a high speed of 20 rpm/min, and an acceleration time of 2 min. The mice were trained once a day for three days before the formal test so that the mice learned to hold the rod. During the formal test, each mouse was tested three times. The latency to fall was recorded using the Rotarod Analysis System. If the mice could hold for 5 min at any one of the three opportunities, it was recorded as the maximum value of 5 min.

For the pole-climbing test, mice were placed on the top of the wooden pole (diameter 1 cm, height 50 cm, rough surface). The time taken by the mice to turn completely head downward was recorded as t-turn time, and the time taken by the mice to reach the floor with their four paws was recorded as t-total time. Each mouse was accustomed to the apparatus on the day before testing. The test was performed three times to ensure accuracy.

Brain tissues or cells were lysed in RIPA buffer (50 mM Tris-HCl pH 7.4, 150 mM NaCl, 1% NP-40) supplemented with protease and phosphatase inhibitors. Protein concentration was determined using a BCA assay kit (P0010, Beyotime Biotechnology). Protein samples (20 µg) were separated by 12% SDS-PAGE and transferred to PVDF membranes (Millipore). Membranes were blocked with 10% non-fat dry milk for 1 h at room temperature and incubated overnight at 4°C with primary antibodies. Membranes were then incubated with appropriate HRP-conjugated secondary antibodies for 1 h at room temperature. Protein bands were visualized using enhanced chemiluminescence (ECL) on an Image Quant LAS 4000 mini (GE Healthcare) and quantified using ImageJ software.

For Co-IP, cells were lysed in RIPA buffer and centrifuged (16,000 × g, 15 min, 4°C). Supernatants containing 500 µg protein were incubated overnight at 4°C with rotation with 10 μL of anti-DJ-1 or anti-NLRP3 antibody. 20 μL Protein A/G agarose beads (sc-2003, Santa Cruz) were added and incubated for 4 h at 4°C with rotation. Beads were washed three times with RIPA buffer. Immunoprecipitated proteins were eluted in SDS-PAGE sample buffer and analyzed by western blotting using the antibodies listed above.

After treatment, BMDMs or microglia were washed twice with PBS. Total RNA was extracted using Trizol reagent (15596026, Invitrogen) following the manufacturer's protocol. RNA concentration and purity (A₂₆₀/A₂₈₀ ratio) were determined spectrophotometrically. cDNA was synthesized from 1000 ng RNA using HiScript II Q Select RT SuperMix for qPCR (+gDNA wiper) (R233-01, Vazyme) under the following conditions: reverse transcription at 37°C for 15 min, and enzyme inactivation at 85°C for 5 s. cDNA was diluted 5-fold for qPCR.

qPCR reactions (10 μL) contained 2 μL diluted cDNA, 0.5 μL each of forward and reverse primers (10 µM), 5 μL SYBR Green qPCR Master Mix (P111-01, Vazyme), and 2 μL nuclease-free water. Reactions were performed using the following cycling parameters: 95°C for 10 min, 40 cycles of 95°C for 15 s and 60°C for 1 min, 95°C for 15 s, 60°C for 1 min, 95°C for 15 s.

Primers used were as follows:

Gapdh: Forward 5′-CCACTCCAGAAGAAAGACCTAAG-3′,

Reverse 5′-TGGCTGTTGAGGTGCTTATAG-3′.

Nlrp3: Forward 5′-CTCAACAGTCGCTACACGCAG-3′,

Reverse 5′-AGCTTAAGGGAACTCATGGGG-3′.

Relative mRNA expression was calculated using the 2^−ΔΔCt method with Gapdh as the reference gene.

Mice were deeply anesthetized with sodium pentobarbital (1%, 4.5 ml/kg, i.p.) and transcardially perfused with PBS followed by 4% paraformaldehyde (PFA). Brains were post-fixed in 4% PFA for 24–48 h at 4°C, dehydrated in 20% and 30% sucrose-PBS solutions (2–3 days each), embedded in OCT gel, and sectioned coronally at 30 µm thickness.

For IHC (TH staining): Typical brain slices were selected and immersed with 3% hydrogen peroxide (H₂O₂) for 15 min to remove endogenous peroxidase. Sections were permeabilized and blocked in PBS containing 0.3% Triton X-100 and 5% BSA for 1.5 h at room temperature. Sections were incubated with anti-TH antibody overnight at 4°C. After PBS washes, sections were incubated with secondary antibodyfor 1.5 h at room temperature, followed by streptavidin-HRP and development with DAB substrate (KGP1045, KeyGEN BioTECH). All the antibodies were diluted in PBS with 1% BSA. Sections were attached on slides, dehydrated through graded alcohols, transparented in xylene, coverslipped with neutral resin, and imaged using an Olympus BX51 microscope.

For TSA multiplex staining: Mounted sections were dried at 37°C, treated with 3% H₂O₂ for 10 min at room temperature to remove endogenous peroxidase, and blocked/permeabilized in blocking solution (NEON) for 1.5 h at 37°C. Sections were sequentially incubated with primary antibodies overnight at 4°C, followed by appropriate HRP-conjugated secondary antibodies for 1.5 h at room temperature. TSA fluorophore reagents were applied for 1 min, followed by quenching in distilled water. Antibody complexes were eluted using the provided elution buffer for 30 min at 37°C in the dark before proceeding to the next primary antibody. Nuclei were counterstained with DAPI and photographed by confocal microscope (Lecia SP8).

NLRP3⁺ cell counting and IBA-1⁺ proportion of perikaryon were analysed as previously described (26). NLRP3⁺ cells were quantified within a 0.3 mm² region of interest (ROI) consistently positioned in the SNc of all images. NLRP3⁺ cells, defined by co-localization of a DAPI-stained nucleus with surrounding NLRP3 immunoreactivity, were manually counted with counting software.

To evaluate morphological characteristics of IBA-1⁺ cells within the SNc, the defined SNc region was mapped onto corresponding IBA-1⁺ and DAPI channels. IBA-1⁺ cells exhibiting definitive DAPI⁺ nuclear labeling within the SNc were subjected to quantitative analysis. For each cell, somatic (IBA-1⁺ perikaryon) and nuclear (DAPI⁺) boundaries were manually traced with the ImageJ to generate paired ROIs. The relative somatic area was calculated as: IBA-1⁺ proportion of perikaryon = [(Perikaryon Area - Nuclear Area)/Perikaryon Area]. This metric served as a comparative indicator of perikaryal size alterations across experimental groups.

The Duolink^® PLA kit (DUO92101, Sigma-Aldrich) was used to detect protein-protein interactions. After treatment, cells were washed with PBS, fixed with 4% PFA for 20 min at room temperature, and permeabilized with 0.3% Triton X-100 in PBS for 30 min. Cells were blocked with Duolink^® blocking solution for 1 h at 37°C. Primary antibodies (rabbit anti-DJ-1 and mouse anti-NLRP3) diluted in Duolink^® antibody diluent were applied and incubated overnight at 4°C. After washing with PBST (PBS containing 0.05% Tween-20), the MINUS and PLUS PLA probes (mixed at a 1:1:3 ratio with antibody diluent) were added and incubated for 1 h at 37°C. The ligation reaction was performed for 30 min at 37°C using the ligation mix (1 μL ligase, 8 μL Duolink^® Ligation Buffer and 32 μL deionized water). Amplification was carried out with PLA amplification reagents (8 μL Amplification Buffer, 0.5 μL Polymerase and 32 μL deionized water) at 37°C for 100 min. After PBST washes, F-actin was stained with phalloidin-FITC (US Everbright) for 10–20 min, and nuclei were stained with DAPI for 30 s. Images were acquired using a confocal microscope (Lecia SP8).

Mice were anesthetized and perfused with PBS. Striatal tissues were rapidly dissected on ice, weighed, and homogenized in ice-cold lysis buffer (0.1 M HClO₄ 0.1 mM Na₂EDTA; 100 μL/mg tissue) using a pre-chilled homogenizer. Homogenates were centrifuged at 20,000 rpm for 30 min at 4°C. The supernatant was transferred and stored at -80°C until analysis.

Monoamine neurotransmitters and metabolites (dopamine, DA) were quantified using a BAS HPLC-ECD system equipped with an LC-4C electrochemical detector, a C18 reversed-phase column (BAS ODS, 1.0 × 150 mm, 5 µm), a DA-5 chromatography workstation and a fluidic delivery system (+500 mV). The mobile phase consisted of 1.0 mM sodium octanesulfonate, 50 mM NaH₂PO₄ 50 mM sodium citrate, 17 mM NaCl, 0.1 mM EDTA, and 4% (v/v) acetonitrile (pH 4.0, adjusted with phosphoric acid). Chromatographic separation was achieved at a flow rate of 0.08 ml/min with a 10 μL injection volume. Quantification was performed using the external standard method.

Cell culture supernatants were collected after stimulation. Brain tissues were homogenized in ice-cold PBS containing protease inhibitors (10 μL/mg tissue) using a homogenizer, vortexed 10 seconds every 10 min for 30 min on ice, and centrifuged at 16000 g 15 min at 4°C. Supernatants were collected. Concentrations of IL-1β (DY401, R&D Systems) in supernatants and brain homogenates were determined using commercial ELISA kits according to the manufacturers' protocols. Absorbance was measured at the appropriate wavelength using a microplate reader (Varioskan Flash, Thermo Fisher).

All data are presented as mean ± standard error of the mean (SEM). Statistical analyses were performed using GraphPad Prism 10 software. Comparisons between multiple groups were analyzed by Student's t-test, one-way or two-way analysis of variance (ANOVA), followed by Sidak's post-hoc test. P-value < 0.05 was considered statistically significant. For asymmetric n numbers, Type III sum of squares method were performed before two-way ANOVA.

To demonstrate the association between DJ-1 and inflammatory responses—specifically, that DJ-1 protein expressions are responsive to inflammatory stimuli—we conducted both in vivo and in vitro experiments. In vivo, we established PD models via stereotactic LPS injection into the SNc and via subacute MPTP administration. Immunofluorescence co-labeling of DJ-1 with IBA-1 and GFAP revealed that DJ-1 expression was markedly increased in microglia but not in reactive astrocytes following LPS injection (Figures 1A, B). A similar trend was observed in the MPTP model, albeit with a milder inflammatory response (Figures 1A, B). Given that NLRP3 inflammasome is the most abundantly expressed inflammasome in microglia, we stimulate primary microglia with the NLRP3 specific activator, i.e., LPS plus ATP. In line with the in vivo observations, a marked upregulation of DJ-1 expressions was observed after LPS or LPS plus ATP stimulation (Figures 1C, E). Bone marrow-derived macrophages (BMDMs) represent another cell type in which the NLRP3 inflammasome activation pathway is present. Similarly, we found that DJ-1 expressions were significantly upregulated after LPS or LPS plus ATP stimulation (Figures 1D, F). These findings suggest that DJ-1 is specifically upregulated in activated microglia under inflammatory conditions in NLRP3 related pathway.

To further elucidate the role of DJ-1 during NLRP3 inflammasome activation, we employed siRNA-mediated knockdown of DJ-1 in primary microglia followed by LPS or LPS plus ATP stimulation. DJ-1 expression was efficiently knocked down by siRNA in primary microglia (Supplementary Figures S1A, C) and BMDMs (Supplementary Figures S1B, D). We found that knockdown of DJ-1 in microglia significantly suppressed NLRP3-mediated inflammatory responses, as evidenced by reduced expression of caspase-1 (p20) and IL-1β (p17) (Figures 2A, C, D) as well as attenuated IL-1β secretion into the supernatant detected by Elisa (Figure 2G). Similar to primary microglia, knockdown of DJ-1 in BMDMs also significantly suppressed the NLRP3 inflammasome-mediated inflammatory pathway as indicated by decreased caspase-1 (p20) and IL-1β (p17) level (Figures 2B, E, F). ELISA of cell supernatants confirmed a significant reduction in IL-1β secretion (Figure 2H). Together these data indicate that DJ-1 deficiency dampens NLRP3 inflammasome activation and subsequent inflammatory responses in microglia and BMDMs.

In the brain, activated microglia and the inflammatory cytokines they produce can lead to neuronal apoptosis and axonal degeneration, while suppression of microglia-mediated inflammation can alleviate neuronal damage (27). To further investigate whether DJ-1 knockdown–mediated inhibition of microglial inflammation could alleviate neuronal injury, we treated primary cultured neurons with conditioned medium collected from stimulated control or DJ-1 knockdown microglia. Immunofluorescence staining of MAP2 showed preserved neuronal morphology and reduced neurite fragmentation under the stimuli from DJ-1 knockdown microglia compared to the negative control microglia supernatant (Figures 2I, J). Furthermore, CCK-8 assay results showed that primary neurons stimulated with conditioned medium from DJ-1 knockdown microglia exhibited improved cell viability (Figure 2K), accompanied by downregulation of the pro-apoptotic protein Bax and significant upregulation of the anti-apoptotic protein BCL-2 (Figures 2L–N). These results indicate that knockdown of DJ-1 suppresses microglial inflammatory responses, thereby reducing neuronal apoptosis and axonal degeneration.

In BMDMs, multiple inflammasome activation pathways exist beyond NLRP3, which can be activated by different stimulation combinations but ultimately lead to pro-caspase-1 cleavage and IL-1β secretion following ASC speck formation (28). To further investigate whether the inhibitory effect of DJ-1 on inflammasome activation is specific to the NLRP3 pathway, we stimulated DJ-1 knockdown BMDMs with combinations that specifically activate either the NLRC4 or AIM2 inflammasome. In BMDMs treated with ploy(dA:dT) which could activate the AIM2 inflammasome pathway, we found that no significant changes in neither caspase-1 (p20) nor IL-1β (p17) (Figures 3A–C, G). Co-stimulation with LPS and flagellin can activate the NLRC4 inflammasome. We found that DJ-1 knockdown in BMDM had no significant effect on the activation of the NLRC4 inflammasome, as evidenced by unchanged expression levels of caspase-1 and IL-1β (Figures 3D–F, H). Besides, immunofluorescence staining of ASC speck formation indicated that the regulatory effect of DJ-1 is specific to NLRP3 inflammasome activation and does not extend to other inflammasome pathways such as AIM2 or NLRC4 (Figures 3I, J). Collectively, these results underscore a specific and exclusive regulatory role of DJ-1 in the NLRP3 inflammasome pathway, with no detectable impact on the activation of other inflammasomes such as AIM2 and NLRC4.

Given that DJ-1 knockdown selectively impairs NLRP3 inflammasome activation, we next sought to determine whether this regulatory effect involves direct modulation of the NLRP3 protein expression levels. In both primary microglia and BMDMs, DJ-1 knockdown led to a significant reduction in NLRP3 protein levels after LPS or LPS plus ATP stimulation (Figures 4A, B, D, E). However, mRNA levels of NLRP3 remained unchanged (Figures 4C, F). These results suggest that DJ-1 knockdown does not affect the transcriptional activity of NLRP3 but instead regulates its protein stability through a post-transcriptional mechanism. Specifically, we hypothesize that DJ-1 may stabilize the NLRP3 protein structure and prevent its degradation. Therefore, we hypothesize that DJ-1 may function as a stabilizing factor for NLRP3 by directly interacting with the NLRP3 protein. Co-immunoprecipitation assays in primary microglia demonstrated a direct interaction between DJ-1 and NLRP3 proteins (Figure 4G). Furthermore, Proximity ligation assays (PLA) revealed that DJ-1 and NLRP3 are co-localized in the cytoplasm of microglial cells upon inflammatory stimulation, further supporting the possibility of a functional interaction (Figure 4H).

To further investigate the relationship between DJ-1 and NLRP3, we generated DJ-1 knockout HEK-293 cells using CRISPR-Cas9 technology. We then transfected equal amounts of NLRP3 plasmid into both DJ-1 knockout and control cells. The results showed that the protein level of NLRP3 was significantly reduced in DJ-1-deficient HEK-293 cells, consistent with the observations in primary microglia and BMDMs (Figures 4I, J). NLRP3 harbors multiple protein-interacting domains on its surface and has been reported to bind to several regulatory proteins such as NEK7 (29) or TRIM65 (16) which contribute to stabilizing its conformation. We hypothesize that DJ-1 may exert a similar stabilizing function as these proteins. Our results demonstrated that inhibition of autophagy with 3-MA led to a significant increase in NLRP3 protein levels in both wild-type and DJ-1 knockout HEK-293 cells, confirming that NLRP3 is degraded, at least in part, through the autophagy-lysosome pathway. Furthermore, due to the absence of DJ-1, rapamycin-induced autophagy caused a greater reduction in NLRP3 levels in DJ-1 knockout cells than in wild-type cells (Figures 4K, L). These findings support the notion that DJ-1 stabilizes NLRP3 by protecting it from autophagic degradation, likely through maintaining its proper conformational structure. In summary, we found that DJ-1 binds directly to the NLRP3 protein through a post-transcriptional mechanism, stabilizing its conformation and protecting it from autophagy–lysosome-mediated degradation. In the absence of DJ-1, this protective effect is lost, resulting in enhanced degradation of NLRP3 and subsequent downregulation of the inflammatory signaling pathway.

Given that microglial activation and NLRP3 inflammasome-mediated neuroinflammation are critical pathological mechanisms in Parkinson's disease (PD), we employed an in vivo model involving stereotactic injection of LPS into the substantia nigra to induce PD-like pathology (Figure 5A). To specifically knock down DJ-1 in microglia, we used an adeno-associated virus (AAV) vector expressing DJ-1 shRNA under the control of the microglia-specific promoter F4/80, and we confirmed that the AAV-mediated knockdown effectively reduced DJ-1 expression (Figures B, C). The AAV was stereotactically injected into the substantia nigra pars compacta (SNc), and after three weeks for viral expression, we performed a second stereotactic injection of LPS at the same site. Immunofluorescence analysis revealed that, following stereotactic injection of LPS into the SNc, mice with microglia-specific DJ-1 knockdown exhibited a markedly attenuated microglial inflammatory response, along with significantly reduced NLRP3 protein expression and reduced quantity of NLRP3⁺ cells (Figures 5B, D, E). IBA-1 integrated density level as well as proportion of perikaryon IBA-1 positive cells were attenuated after AAV-sh-DJ-1 injection (Figures B, F, G). Immunohistochemical staining for IBA-1 further confirmed that microglial activation was markedly reduced in mice with microglia-specific DJ-1 knockdown (Figure 5H). In addition, following LPS administration, the midbrain tissue of DJ-1-deficient mice exhibited significantly lower caspase-1 levels compared to control mice (Figures 5I, J). Consistently, ELISA analysis revealed a significant reduction in IL-1β levels in the midbrain (Figure 5K).

To further demonstrate whether knockdown of microglial DJ-1 not only suppresses neuroinflammation but also protects dopaminergic neurons from inflammation-induced degeneration, we evaluated striatal dopamine content by HPLC and tyrosine hydroxylase (TH) expression in the SNc by both immunohistochemistry and Western blot in the LPS stereotactic model and the sub-acute MPTP model (Figure 6A). Results from both models demonstrated that DJ-1 knockdown in microglia significantly preserved striatal dopamine content (Figure 6B). Furthermore, the number of TH-positive neurons in the SNc, the optical density of STR (as an indicator for the density of DA terminals) as well as the TH protein expression were consistently increased after AAV-sh-DJ-1 injection followed by LPS or MPTP treatment (Figures 6C–G). These findings suggest that microglial DJ-1 plays a pivotal role in mediating inflammation-induced dopaminergic neuron loss, and that its depletion can confer dual benefits by reducing neuroinflammation and protecting neuronal survival indicated by the preservation of TH⁺ cells in the SNc.

Moreover, we performed behavioral assessments on the LPS-induced PD model mice. The results showed that mice with microglia-specific DJ-1 knockdown exhibited significantly increased total distance and locomotor speed in the open field test compared to control mice (Figures 6H–J). Additionally, their performance in both the rotarod and pole test paradigms was markedly improved, indicating better motor coordination and balance (Figures 6K–M). These findings further support the notion that microglial DJ-1 knockdown alleviates neuroinflammation-related motor deficits in vivo.