p38-TFEB pathways promote microglia activation through inhibiting CMA-mediated NLRP3 degradation in Parkinson's disease

Background Parkinson’s disease (PD) is characterized by degeneration of dopaminergic neurons in the substantia nigra pars compacta (SNpc), accompanied by accumulation of α-synuclein, chronic neuroinflammation and autophagy dysfunction. Previous studies suggested that misfolded α-synuclein induces the inflammatory response and autophagy dysfunction in microglial cells. The NLRP3 inflammasome signaling pathway plays a crucial role in the neuroinflammatory process in the central nervous system. However, the relationship between autophagy deficiency and NLRP3 activation induced by α-synuclein accumulation is not well understood. Methods Through immunoblotting, immunocytochemistry, immunofluorescence, flow cytometry, ELISA and behavioral tests, we investigated the role of p38-TFEB-NLRP3 signaling pathways on neuroinflammation in the α-synuclein A53T PD models. Results Our results showed that increased protein levels of NLRP3, ASC, and caspase-1 in the α-synuclein A53T PD models. P38 is activated by overexpression of α-synuclein A53T mutant, which inhibited the master transcriptional activator of autophagy TFEB. And we found that NLRP3 was degraded by chaperone-mediated autophagy (CMA) in microglial cells. Furthermore, p38-TFEB pathways inhibited CMA-mediated NLRP3 degradation in Parkinson's disease. Inhibition of p38 had a protective effect on Parkinson's disease model via suppressing the activation of NLRP3 inflammasome pathway. Moreover, both p38 inhibitor SB203580 and NLRP3 inhibitor MCC950 not only prevented neurodegeneration in vivo, but also alleviated movement impairment in α-synuclein A53T-tg mice model of Parkinson’s disease. Conclusion Our research reveals p38-TFEB pathways promote microglia activation through inhibiting CMA-mediated NLRP3 degradation in Parkinson's disease, which could be a potential therapeutic strategy for PD. Graphical abstract p38-TFEB pathways promote microglia activation through inhibiting CMA-mediated NLRP3 degradation in Parkinson's disease. In this model, p38 activates NLRP3 inflammasome via inhibiting TFEB in microglia. TFEB signaling negatively regulates NLRP3 inflammasome through increasing LAMP2A expression, which binds to NLRP3 and promotes its degradation via chaperone-mediated autophagy (CMA). NLRP3-mediated microglial activation promotes the death of dopaminergic neurons. Supplementary Information The online version contains supplementary material available at 10.1186/s12974-021-02349-y.


Abstract Background
Parkinson's disease (PD) is characterized by the degeneration of dopaminergic neurons in the substantia nigra pars compacta (SNpc), accompanied by chronic neuroin ammation, autophagy dysfunction and αsynuclein accumulation in the form of Lewy bodies. Previous studies showed that misfolded α-synuclein upregulates the in ammatory and autophagy dysfunction in microglial cell. The NLRP3 in ammasome signaling pathway plays a crucial role in the neuroin ammatory process in the central nervous system. However, the inter-relationship between autophagy de ciency and neuroin ammation induced by αsynuclein accumulation is not well understood.

Methods
We investigated the impact of p38-TFEB-NLRP3 pathways on neuroin ammation in the α-synucleinA53T PD models, using a combination of immunoblotting, immuno uorescence, immunocytochemistry, ow cytometry, ELISA, and a series of behavioral texts.

Results
In the present study, we showed NLRP3 was degraded through chaperone-mediated autophagy (CMA) in microglia cell. Furthermore, p38-TFEB pathways inhibited CMA-mediated NLRP3 degradation in Parkinson's disease. Overexpress mice and BV2 cells with α-synuclein A53T mutant active P38, which inhibit the master transcriptional activator of autophagy, TEEB in BV2 cells. Notably, inhibition p38 had a protective effect on Parkinson's disease model, which depend on suppressing the activation of NLRP3 in mmasome pathway. Importantly, both p38 inhibitor SB203580 and NLRP3 inhibitor MCC950 not only prevent neurodegeneration in vitro, but also alleviates movement impairment in α-synuclein A53T -tg mice model of Parkinson's disease.

Conclusion
Our research reveals an endogenous regulatory mechanism of NLRP3 turnover and microgliadopaminergic neuron interaction, which may be a potential therapeutic strategy for Parkinson's disease.

Background
Parkinson's disease (PD) is characterized by the loss of dopaminergic neurons in the substantia nigra, resulting in neuromuscular dysfunction such as rest tremors. Previous studies shown that neuroin ammation may be the primary factor leading to the PD-related pathogenesis and causing the degeneration and loss of dopaminergic neurons. Proin ammatory cytokine such as tumor necrosis factor α (TNF-α) and IL-1β were found to be upregulated in the cerebrospinal uid and blood of PD patients [1].
Moreover, activated microglia releases proin ammatory cytokines which lead to neuronal death, in turn, dopaminergic neurons loss exacerbates neuroin ammation. This vicious circle is closely associated with the pathological process of PD. The NLRP3 in ammation complex contributes to activating the innate immune response and causes cell pyrotosis [2] , [3]. Recent studies demonstrated that NLRP3 in ammasome is involved in the progression of neurodegenerative disease and NLRP3 de ciency plays a protective role in animal models of Alzheimer's disease (AD) and PD [4,5]. Although the mechanisms that NLRP3 in ammasome participates in diverse in ammatory diseases have been extensively investigated, its regulatory networks in microglia are unclear.
Mitogen-activated protein kinases (MAPKs) regulate various cellular processes. The p38 mitogenactivated protein kinase (p38MAPK), a prominent member of the MAPKs, is not only involved in cell cycle, cell death, development, differentiation, senescence and tumorigenesis, but also functions as a speci c class of serine/threonine kinase regulating the in ammation responses. The important role of p38mediated in ammatory responses was originally con rmed as the target of the pyridinyl imidazole that inhibited in ammatory cytokines release in Lipopolysaccharides (LPS)-treated monocytes [6]. The MAPK pathways contributed to the increased production of proin ammatory cytokines in microglia that treated with toll-like receptor (TLR) ligands or beta-amyloid (Aβ) [7]. Recently, Mao and colleagues identi ed that p38 inhibits autophagy and promotes microglial in ammatory responses via ULK1 phosphorylation [8]. A few studies have demonstrated that pharmacological inhibition of p38 plays a protective role in several animal models of neurodegenerative disease [8]. Moreover, studies in vivo have also shown that p38 de ciency protected from aggravating AD progress via enhancing the level of autophagy [12,13]. The discovery of transcription factor EB (TFEB) has triggered more and more studies that try to explore the therapeutic potential of targeting TFEB to treat neurodegenerative diseases, because it works as a master regulator of autophagy [14]. Although the protective effect of p38 inhibition and its implication in neurodegenerative diseases are emerging, the mechanisms by which p38 regulates TFEB-mediated autophagy and in ammatory responses in microglia remains elusive.
CMA is involved in cytosolic protein substrate recognition by the HSC70(HSPA8), which targets lysosomes substrate binding via LAMP2A and the formation of a substrate translocation complex with membrane-bound LAMP2A is degraded by intra-lysosomal enzymes. About 30% of cytosolic proteins can be CMA substrates, containingthe CMA-speci c recognition motif (KFERQ). CMA activity to degrade proteins declines with age. This decline of CMA activity underpins the pathogenesis of neurodegeneration, including PD [15].
Here, we demonstrate that p38-TFEB pathway regulates NLRP3 in ammasome degradation via CMA. We not only identify that NLRP3 is a novel substrate for CMA, but also proved that p38 inhibits TFEB activity via phosphorylation which blocks NLRP3 degradation. Besides, our results reveal that both p38 inhibitor SB203580 and NLRP3 inhibitor MCC950 reduced microglia activation and dopaminergic neuronal loss induced by α-synuclein. All the results suggest the p38-TFEB pathway may be a potential treatment target for NLRP3 in ammasome-driven neurodegenerative diseases.

Page 4/25
Animal. The α-synuclein A53T -tg mice expressing mutant A53T α-synuclein (the 140 amino acid isoform) under the direction of the mouse prion protein promoter has been described previously [16]. The αsynuclein A53T -tg mice expressed mutant human A53T α-synuclein were purchased from Model Animal Research Center of Nanjing University. The structure, location and onset of the inclusions seen in the mutant mice resemble characteristics seen in human neuronal α-synucleinopathies and these transgenic mice are generally used in the study of Parkinson's disease. Animals were individually housed under light: dark (12:12 h) cycles and provided with food and water.
Balance beam test. Mice were placed on a narrow beam (2*3*60cm) suspended 20 cm above soft bedding, and their movement from one end towards the other end was recorded. The number of missteps (paw faults, or slips) was scored during the trip.
Pole test. This test was performed based on the method described by Ogawa [17] used as a measure for bradykinesia. Mice (n = 8 in each group) were placed the top of a pole (55 cm in height, 10 mm in diameter) with a rough surface. Both the time to turn and climb down was recorded.
Administration of the p38 inhibitor SB203580 and MCC950. All animal procedures were approved by the ethics committee of the Southern Medical University. All groups of animals received the inhibitor SB203580 (5 mg/kg) or MCC950 (10 mg/kg) through intraperitoneal injection. In each experimental cohort, male mice were randomly matched for group assignment. α-synuclein A53T -tg mice were injected with the drug starting from 5 months of age until the end of 9 months. The midbrain and cortex of half mice were dissected and used for Western blotting analysis. Meanwhile, the other half mice were perfused with 4% paraformaldehyde and para nized coronal sections were processed for immunohistochemistry assay. Immunohistochemistry. The para n sections of brain tissue were collected for routine immunohistochemistry staining for p-p38 (1:50), TH (1:100), IBA-1 (1:100), IL-1β (1:100), NLRP3(1:100), LAMP2A (1:100) and TFEB (1:100). The area of the midbrain dopaminergic neurons from brain sections according to the anatomical position and the location of TH-positive neurons has been described previously [18].
Transmission electron microscopy (TEM). SN4741 cells were washed once with PBS, collected with a cell scraper. The sections of midbrain tissue were collected and washed once with cold PBS. Cell and midbrain tissue pellets were re-suspended in the xative containing 2.5% glutaraldehyde in PBS at 4°C. Cell and midbrain tissue pellets were xed for 3 h. After rinsing with sodium cacodylate buffer, they were further xed in 1% OsO 4 in sodium cacodylate buffer on ice for 1 h and dehydrated with acetone, and then embedded in resin and polymerized at 60°C for 48 h. Ultrathin sections were obtained on an ultramicrotome and stained with uranyl acetate and lead citrate before observation under Hitachi H-7500 TEM.
Cell culture and transfection. SN4741 cells were cultured in DMEM medium containing 1% glucose, 100 U/ml penicillin-streptomycin, 1% L-glutamine and supplemented with 10% fetal bovine serum at 33°C. BV2 cells were cultured in DMEM medium supplemented with 10% fetal bovine serum at 37°C.Cells were transfected with Lipofectamine 3000 reagent (Invitrogen, Carlsbad, CA, USA) following the manufacturer's instructions.
Western blotting and co-immunoprecipitation. Cytosolic and nuclear fractions were collected using an isolation kit (KeyGEN, Nanjing, China) following the manufacturer's instructions. For Western blotting, cells were collected in RIPA buffer containing 20 mM Tris-HCl, pH 7.4, 150 mM NaCl and 1% Triton X-100 with protease inhibitor cocktail (Roche, Nutley, NJ, USA). The extracts were centrifuged for 10 min at 13,000 rpm at 4°C, and the supernatants were used for immunoblot analysis. For co-immunoprecipitation, cells were centrifuged at 13,000 g for 10 min at 4°C, and then the supernatant was transferred into a new tube. And then 5 µg of the corresponding rst antibody was added into the samples and incubating overnight at 4°C. The next day, 50 µl agarose beads were added into the samples and incubating 2 h at 4°C. The immunocomplex was collected with centrifugation at 1,500g for 1 min at 4°C. And washed 3 times with RIPA buffer. Proteins were eluted from beads with 2× SDS loading buffer and subjected to immunoblot analysis. The immunoreactive bands were detected by Odyssey Infrared Imaging System. The band intensity was analyzed with Image J analysis software. Antibodies used in Western blotting are diluted by 1000, except TH (1:100) and p-p38(1:100).
Apoptosis and cell death. Cell apoptosis of SN4741 cell was measured using a Cell Apoptosis Kit (Dojindo, Japan) according to the manufacturer's instructions. The percent of cell apoptosis was determined by staining with 2 µM Annexin V and 2 µM Propidium Iodide (PI). The number of cell apoptosis was counted by ow cytometry and images of apoptosis was taken by Laser Scanning Confocal Microscopy. Cell death of primary neuron was detected by staining with PI (1µg/ml) (KeyGEN) according to the manufacturer's instructions. Images were captured by uorescence microscope and analyzed by Image J analysis software.
Measurement of mitochondrial membrane potential. The mitochondrial membrane potential in cells was assessed using Tetramethylrhodamine methyl ester (TMRE; Thermo Fisher, T669). Cells were washed with 1×PBS and then incubated with TMRE (0.1µM) in the medium for 20 min in the dark. The intensity of uorescence was monitored using a uorescence microscope.
ELISA. Tissue specimen: After cutting the specimen, weigh 0.1 g of tissue, add 9 ml of PBS, and homogenize the specimen. Centrifuge for about 20 minutes (12000 rpm) and carefully collect the supernatant that used for ELISA detection following the manufacturer's instructions. The cell culture medium: Cell culture medium was centrifuged and used for ELISA detection following the manufacturer's instructions.
Immuno uorescence. Cells were grown on a confocal dish for 2 days, and then cells were washed with PBS three times, then xed with 4% paraformaldehyde for 15 min at room temperature. After that, cells were permeabilized with frozen methanol for 10 min at -20°C and blocked in 5% BSA for 30 min. And then samples were incubated with primary antibody (1:100 ~ 1:200) in 5% BSA overnight at 4°C, and then incubated with secondary antibody (1:100) in 5% BSA for 1 h at room temperature. An Olympus FV1000 confocal microscope with a 100× objective was used for image capture. The co-localization signal was analyzed with Image Pro Plus software.
Statistical analysis. All data of experiments were analyzed with GraphPad Prism 7 software (La Joya, Ca, USA). Data from the Western blot and immunohistochemistry analysis were performed with the t-test for two groups or ANOVA with Tukey's post-test (GraphPad Software) for multiple groups. Data from Balance beam test, Pole test, Transmission electron microscopy, Immuno uorescence, were analyzed by two-way ANOVA followed by Tukey's post-test. All values are expressed as the mean ± SEM, and p values < 0.05 were considered statistically signi cant. All experiments were repeated at least 3 times.

NLRP3 in ammasome is activated in α-synuclein A53T -tg mice
In order to investigate the effects of α-synuclein accumulation and related toxicity, we used α-synuclein A53T transgenic mice (α-synucleinA A53T -tg mice). The accumulation of α-synuclein was identi ed in both the SNpc and the cortex at 9 months of age (Fig.S1 A). To determine the effect of α-synuclein on NLRP3 in ammasome activation, we measured the expression of core components of this multiprotein complex of the canonical in ammasome in α-synuclein A53T -tg mice by IHC and Western blotting. In the brain sections of the SNpc and the cortex, NLRP3 in ammasome activation was remarkably intensi ed, as evidenced by increased NLRP3 expression as well as over-production of IL-1β (Fig. 1A-G).. Besides, microglial activation, as assessed morphologically by immunohistochemistry with the classic antibody speci c for Iba-1, microglia was activated in α-synuclein A53T -tg mice brain both of the SNpc and the cortex (Fig.S1 B). To further con rm this result, PD patients' serum was assayed by ELISA. Consistently, over-production of IL-1β and IL-18 were found in PD patients' serum ( Fig.S1 C-D). Collectively, these data suggest that the NLRP3 in ammasome is activated in the pathological process of PD.
2. NLRP3 in ammasome activation is suppressed by p38 inhibitor SB203580 To examine whether p38 involved in the NLRP3 in ammasome activation in PD, we rst investigated the levels of p38 phosphorylation by IHC. Results showed that compared with wild-type mice, p38 was activated in both the SNpc and the cortex of α-synuclein A53T -tg mice at 9 months of age (Fig. 1H, I).
To assess whether p38 mediates NLRP3 activation, α-synuclein A53T -tg mice were injected with p38 inhibitor SB203580 starting at the age of 5 months until the end of 9 months. Then, we analyzed NLRP3 and IL-1β by IHC. The results showed that SB203580 signi cantly reduced the levels of NLRP3 and IL-1β both in SNpc and cortex of α-synuclein A53T -tg mice ( Fig. 2A-D). Furthermore, the increased protein levels of NLRP3, ASC, cleaved CASP1, and cleaved IL-1β in α-synuclein A53T -tg mice (Fig. 2E-G). Previous work identi es a key role of microglia and NLRP3 in ammasome activation in the pathogenesis of neurodegeneration [19]. Therefore, we use the classic microglial cell line BV2 cells to explore the potential mechanisms of NLRP3 activity. The results are consistent with in vivo, the increased protein levels of NLRP3, ASC, cleaved CASP1, and cleaved IL-1β in BV2-A53T cells were abolished after SB203580 treatment ( Fig.S1 E, F). Also, we have detected the Caspase 1 activity as well as mature IL-1β using ELISA in the tissue homogenates, and results revealed that SB203580 could decrease the levels of Caspase 1 activity and mature IL-1β in α-synuclein A53T -tg mice (Fig.S1 G,H). Of note, SB203580 did not affect the transcription of α-synuclein (Fig.S1 I). Together, these data suggested that p38 inhibitor SB203580 reduced neuroin ammation caused by α-synuclein accumulation.

SB203580 induces NLRP3 degradation via chaperonemediated autophagy
The activation of NLRP3 in ammasome in cells is tightly regulated. Excessive activation of the NLRP3 in ammasome is involved in the pathological process of various diseases including PD. To test whether CMA degrades NLRP3, the interaction between NLRP3 and LAMP2A was examined by immunoprecipitation in brain tissues. The results showed that LAMP2A interacts with NLRP3 in vivo (Fig. 3A). Moreover, we found that SB203580 enhanced the NLRP3/LAMP2A and NLRP3/ HSPA8 interaction in BV2 cells (Fig. 3B, C).
In CMA, the substrate usually contains a KFERQ-like pentapeptide consensus sequence that is recognized by HSPA8 (HSC70). Inspection of the amino acid sequence of mouse NLRP3, four KFERQ-like motifs (355LEKLQ359, 603QIRLE607, 795QKLVE799 and 989EVLKQ993), were revealed (Fig.S2 A). To determine whether NLRP3 is a bona de substrate for CMA, the rst two amino acids of the mouse NLRP3 KFERQlike sequence were mutated to alanine (NLRP3AA). The result showed that mutation of the KFERQ-like sequence of NLRP3 abolished its interaction with HSPA8 (Fig. 3D, E).
To further assess whether CMA is involved in the regulation of NLRP3, we treated BV2 cells with serum deprivation, AR7 (10 µM, 24 h) or QX77 (10 µM, 24 h), the latter two chemicals are well-known CMA agonists. The results showed that both serum deprivation and CMA agonists increased the levels of LAMP2A and reduced NLRP3 (Fig. 3F-I). Moreover, the level of LAMP2A in the SNpc of mice was analyzed. Treatment with p38 inhibitor SB203580 elevated the level of LAMP2A, a key receptor for CMA (Fig. 3J, K).

SB203580 activates TFEB-mediated autophagy
Our data in Figs. 2 and 3 demonstrate that p38 mediates the NLRP3 in ammasome activation in BV2 cells and in α-synuclein A53T -tg mice, as evidenced by increased levels of NLRP3, cleaved CASP1, and IL-1β. Furthermore, in Fig. 4, we showed that p38 regulates NLRP3 turnover through CMA. All these prompt us to examine the role of p38 in autophagy regulation, which may subsequently contribute to the NLRP3 in ammasome activation [19]. TFEB is a master protein for lysosomal biogenesis. In SNpc sections from α-synuclein A53T -tg mice, the total levels of TFEB and LAMP1 were decreased, while the SQSTM1/p62 level was increased (Fig. 4A, B). To determine whether α-synuclein aggregation affects the TFEB nuclear translocation that regulating the transcription of autophagy-related genes, subcellular localization of TFEB was examined by sub-cellular fractionation. The result showed that the nuclear TFEB level decreased in SNpc sections of α-synuclein A53T -tg mice (Fig.S2 B, C).

p38 interacts with and phosphorylates TFEB at serine 211
Next, we aimed to elucidate the mechanism by which p38 regulates TFEB-mediated autophagy. To evaluate whether TFEB could be a substrate of p38, we rst detected the interaction between p38 and TFEB by co-immunoprecipitation. As expected, immunoprecipitation analysis showed that p38 interacted with TFEB both in BV2 cells and mice brain (Fig. 5A-E). Of note, SB203580 reduced the p38/TFEB interaction that was enhanced by α-synuclein A53T aggregation (Fig. 5B-E).
To identify the regulatory function of p38 on TFEB, we examined the role of p38 on the interaction between TFEB and14-3-3 proteins. As phosphorylation of TFEB at serine 211 by mTORC1 has been proved to promote TFEB binding to 14-3-3, we detected the level of TFEB phosphorylation with a phospho-Ser211 speci c antibody. It showed that SB203580 reduced the TFEB phosphorylation at serine 211 (Fig. 5F, G). To explore the mechanism by which p38 regulates TFEB's function, we analyzed the effect of overexpression kinase-dead p38 mutant on TFEB's nuclear translocation. Compared with p38 WT, kinase-dead p38 overexpression increased the nuclear translocation of TFEB (Fig. 5H, I) and enhanced the biogenesis of lysosome (Fig. 5J, K). Meanwhile, kinase-dead p38 overexpression declined the levels of NLRP3, ASC, C-CASP1 detected by WB (Fig. 5L, M). To further con rm the effects of p38alpha-MAPK (p38), we analyzed the effect of p38 siRNA. As gure S3 shown, p38 siRNA increased the biogenesis of lysosome (Fig. S3 A-B). Also, p38 siRNA decreased the levels of NLRP3, C-CASP1 detected by WB. (Fig. S3 C, D). These results supported that p38 activation suppresses TFEB function via phosphorylating TFEB at serine 211 and thus inhibiting TFEB nuclear translocation.
These ndings including the interaction between p38 and TFEB, the inhibitory effects of p38 inhibitor and the activation of NLRP3 in ammasome related to autophagy, raise the possibility that the p38 may curb TFEB to increase NLRP3 in ammasome activation. To test the hypothesis, we rstly overexpressed TFEB in α-synuclein-A53T BV2 cells. The results showed that TFEB overexpression reduced NLRP3 and cleaved CASP1 (Fig. S3 E, F). To examine whether p38 activates NLRP3 via regulating TFEB, we compared the effects of SB203580 treatment, TFEB knockdown, and both SB203580 treatment and TFEB knockdown in α-synuclein-A53T BV2 cells. Remarkably, the overactivation of NLRP3 in ammasome and cleaved CASP1 resulted from α-synuclein-A53T was almost completely reversed by SB203580, while TFEB knockdown eliminated the effect of SB203580 (Fig. S3 G, H). Similarly, the effect of SB203580 in IL-1β, IL-18 were abrogated in α-synuclein-A53T BV2 cells with TFEB knockdown (Fig. S3 I, J). To further test whether the autophagy/lysosome pathway is responsible for p38 inhibition-induced NLRP3 degradation, we treated α-synuclein-A53T BV2 cells with autophagy/lysosome inhibitor CQ (10 µM, 4 h). The results showed that CQ blocked SB203580-induced NLRP3 degradation (Fig. S3 K, L), which further supported that p38 activates NLRP3 in ammasome via disturbing TFEB-mediated autophagy.

α-synuclein-induced NLRP3 activation in microglia promoted dopaminergic neuronal loss
To examine whether microglial activation is responsible for the death of dopaminergic neurons, SN4741 cells were treated with conditioned medium from microglia. After appropriate treatments, cell death in SN4741 cells were determined by ow cytometry. As gure S4 A-B shown, conditioned medium formαsynuclein-A53T BV2 cells caused the dopaminergic neuronal loss, while NLRP3 inhibitor MCC950 (10 µM, 12h) and p38 inhibitor SB203580 can effectively eliminate the effect.
To address whether NLRP3 activation is involved in dopaminergic neuronal loss in vivo, α-synuclein A53Ttg mice were fed with NLRP3 inhibitor MCC950 starting at the age of 5 months until the end of 9 months.
Interestingly, MCC950 signi cantly inhibited the activation of microglia in α-synuclein A53T -tg mice, which suggests that NLRP3 plays a key role in the activation of microglia in PD (Fig. 6A-B). Then, we analyzed the expression of TH in SNpc. And results showed that MCC950 signi cantly reduced the levels of dopaminergic neuronal loss in SNpc of α-synuclein A53T -tg mice (Fig. 6C, D). Furthermore, the balance beam test and pole test showed that MCC950 treatment signi cantly improved the motor activity in αsynuclein A53T -tg mice (Fig. S4 C, D).
7. SB203580 plays a protective role in α-synuclein A53T -tg mice The above data showed that p38-TFEB pathway is involved in NLRP3 in ammasome activation, which prompts us to test whether p38 inhibitor SB203580 has a protective role in α-synuclein A53T -tg mice. Wild type and α-synuclein A53T -tg mice were treated with or without the p38 inhibitor SB203580. Balance beam test and Pole test showed that SB203580 treatment signi cantly improved the motor activity of αsynuclein A53T -tg mice (Fig. 6E, F). In addition, abnormal accumulation of α-synuclein could be alleviated after p38 inhibition (Fig. 6G, H). Synapse loss correlates with cognitive impairment in AD and PD [21,22]. Actually SB203580 rescues the reduced level of synapsin-1 (SYN1, the presynaptic protein) in αsynuclein A53T -tg mice (Fig. 4S E, F) that con rmed the protective role of p38 inhibition. Under TEM, αsynuclein A53T -tg mice have fewer synaptic vesicles, while SB203580 treatment partially restores synaptic vesicle loss (Fig. 6I. J). Taken together, these data indicated that p38 inhibition provides protection in αsynuclein A53T -tg mice.

Discussion
Parkinson's Disease (PD) is a common, age-related neurodegenerative disease a icting more than 2-3% of people over 65 and more than 4% of the population by the age of 85. The progressive loss of dopaminergic neurons in the substantia nigra eventually results in motor alterations of PD. The precise mechanism underlying the pathogenesis of PD is not yet been fully elucidated. Previous studies shown the defect of the autophagy-lysosomal pathway (ALP) and neuroin ammation are contributor in PD pathophysiology. Accumulating evidence suggests that immune signaling cascades are regulated by autophagy, while the mechanisms that modulate the in ammasome activation process is still unclear. In our study, NLRP3 was identi ed as a novel chaperone-mediated autophagy substrate and activation of CMA promotes NLRP3 degradation and thereby inhibits overproduction of proin ammatory cytokines in microglia. Moreover, we demonstrated that p38 activation caused by accumulated α-synuclein inhibits TFEB-mediated CMA, which promotes NLRP3 in ammasome activation. Furthermore, we found that both p38 and NLRP3 inhibitors can reduce aggregated α-synuclein-induced microglia activation and dopaminergic neuronal loss. Therefore, p38-TFEB pathways regulate NLRP3 degradation via CMA in aggregated α-synuclein-induced microglia activation promoting dopaminergic neuronal loss.
The chronic neuroin ammation in PD has been intensively investigated in the past decade, but the precise origin of the CNS in ammation response remains unclear. Evidences found that neuroin ammatory response is a key factor in the pathological process of PD [23]. Activated microglia and increased levels of pro-in ammatory factors were identi ed in postmortem analysis and PD models.
α-Synuclein abnormal aggregation can appear in neuron and microglia, which not only disrupt neuronal function, also causes activation of microglia that increase phagocytic activity and pro-in ammatory cytokines production [24]. It is still controversial whether α-synuclein accumulation can trigger microglia responses and pathological reactions. Our results not only identi ed that NLRP3 was a novel CMA substrate, but also revealed that p38 activation caused by pathologic α-synuclein inhibits TFEB activity that blocked CMA, and then decreased NLRP3 degradation in microglia. Studies found that α-synuclein abnormal aggregation activated microglia, which led to the death of dopaminergic neurons in the substantia nigra of the midbrain [25]. Toll-like receptor 2 (TLR2) of microglia was identi ed as the receptor for secreted α-synuclein, which transports α-synuclein into the microglial cytoplasm [26]. Overexpression of α-synuclein increases microglial activation in vivo. However, α-synuclein internalization still appears when TLR2 is de cient, suggesting α-synuclein uptake relies on multiple receptor systems [27].
The activation of NLRP3 in ammasome usually requires upregulation of the transcription of NLRP3 and pro-IL-1β by NF-κB. However, recent studies report that NLRP3 in ammasome can be regulated independently of transcription. For instance, lys-63-speci c deubiquitinase BRCC36 promotes NLRP3 deubiquitination and facilitates NLRP3 activation. Previous in vitro study showed that NLRP3 in ammasome is activated by pathologic α-synuclein. In this work, we revealed that the degeneration of NLRP3 through CMA is an essential event for the regulation of NLRP3 in ammasome. Our study demonstrated that p38-TFEB pathway regulates NLRP3 in ammasome in microglia which is involved in the pathological process of PD. Importantly our data emphasize the protective effect of p38 inhibition in anti-in ammation and autophagy enhancement which have been shown to help ameliorate the pathological symptoms of neurodegeneration diseases. Several studies revealed that p38 activation is closely connected to autophagy dysfunction. Mao and colleagues identi ed that p38 regulates macroautophagy and CMA via phosphorylating ULK and LAMP2A, respectively [ 28,29 . Besides, Chen et al. reported that p38 inhibition promotes mitophagy [11]. And recent studies have shown that autophagy plays a key role in in ammasome regulation [30]. Notably, autophagy dysfunction disrupts cell homeostasis thus induces excessive activation of in ammasomes. This study provided evidence showing that p38/TFEB-mediated CMA regulating the NLRP3 in ammasome.
Although we identi ed the role of p38 in the activation of NLRP3 in ammasome in α-synuclein A53T mice, the mechanism may be applicable in other neurodegenerative diseases.
In conclusion, we uncovered a novel NLRP3 degradation pathway by CMA in microglia, which is negative regulation by p38 through direct phosphorylating TFEB at ser211, inhibits the transcription of autophagy gene. Moreover, p38 inhibitor SB203580 provided protection, via enhancing TFEB-mediated CMA and reducing NLRP3 in ammasome activation both in vitro and vivo. Given that neuroin ammatory response is deeply involved in neurodegenerative diseases, the mechanisms we identi ed here shed new light on the neurodegenerative process. In addition, the p38-TFEB mediated CMA may be a potential therapeutic target for PD.

Consent for publication
All authors have approved the contents of this manuscript and provided consent for publication.

Availability of data and materials
All data in this study are included in this published article and its supplementary information les.

Competing interests
The authors declare that they have no competing interests. Statistical analysis of the scores of IL-1β staining between α-synucleinA53T-tg and wild type mice. *p<0.05. (E, F, G) Cell lysates from the cortex and SNpc of mice were immunoblotted using the indicated antibodies. The protein levels of NLRP3, ASC, cleaved CASP1 were statistically analyzed in F and G. *p < 0.05.   Immunohistochemistry (IHC) demonstrating the levels of TH in the SNpc of 9 months of wide type, α-