Antagonizing peroxisome proliferator‐activated receptor γ facilitates M1‐to‐M2 shift of microglia by enhancing autophagy via the LKB1–AMPK signaling pathway

Summary Microglia‐mediated neuroinflammation plays a dual role in various brain diseases due to distinct microglial phenotypes, including deleterious M1 and neuroprotective M2. There is growing evidence that the peroxisome proliferator‐activated receptor γ (PPARγ) agonist rosiglitazone prevents lipopolysaccharide (LPS)‐induced microglial activation. Here, we observed that antagonizing PPARγ promoted LPS‐stimulated changes in polarization from the M1 to the M2 phenotype in primary microglia. PPARγ antagonist T0070907 increased the expression of M2 markers, including CD206, IL‐4, IGF‐1, TGF‐β1, TGF‐β2, TGF‐β3, G‐CSF, and GM‐CSF, and reduced the expression of M1 markers, such as CD86, Cox‐2, iNOS, IL‐1β, IL‐6, TNF‐α, IFN‐γ, and CCL2, thereby inhibiting NFκB–IKKβ activation. Moreover, antagonizing PPARγ promoted microglial autophagy, as indicated by the downregulation of P62 and the upregulation of Beclin1, Atg5, and LC3‐II/LC3‐I, thereby enhancing the formation of autophagosomes and their degradation by lysosomes in microglia. Furthermore, we found that an increase in LKB1–STRAD–MO25 complex formation enhances autophagy. The LKB1 inhibitor radicicol or knocking down LKB1 prevented autophagy improvement and the M1‐to‐M2 phenotype shift by T0070907. Simultaneously, we found that knocking down PPARγ in BV2 microglial cells also activated LKB1–AMPK signaling and inhibited NFκB–IKKβ activation, which are similar to the effects of antagonizing PPARγ. Taken together, our findings demonstrate that antagonizing PPARγ promotes the M1‐to‐M2 phenotypic shift in LPS‐induced microglia, which might be due to improved autophagy via the activation of the LKB1–AMPK signaling pathway.

proliferating microglial cells, astrocytes, and other myeloid cells that ultimately produce pro-inflammatory cytokines, chemokines, and other inflammatory mediators, leading to neuronal damage (Ullah et al., 2017;White, Lawrence, Brough & Rivers-Auty, 2017). Microglial cells play critical roles in immune surveillance and host defense by acting as the prime resident innate-immune cells in the central nervous system (CNS; Perry & Holmes, 2014;Salter & Beggs, 2014).
Under normal conditions, microglial cells not only provide surveillance of the CNS environment but also respond to danger signals (Crotti & Ransohoff, 2016;Perry & Teeling, 2013). Activated microglial cells undergo morphological transformation (increase in the size of cell bodies and thickness of proximal processes and decreased ramification of distal branches; Plastira et al., 2016;Walker et al., 2014) and secrete pro-inflammatory cytokines, leading to self-perpetuating damage to the neurons, also known as the classically activated M1 phenotype. However, an alternatively activated M2 phenotype can be neuroprotective and neurosupportive and can promote recovery. The dichotomy of M1 and M2 is an oversimplified conceptual framework. The status of microglia may include a battery of different but overlapping functional phenotypes. However, the general M1 and M2 classification is nevertheless a useful concept to improve our understanding of microglial functional status during injury progression as well as to help us explore novel therapeutic strategies.
Peroxisome proliferator-activated receptor c (PPARc) is a ligandactivated transcription factor that belongs to the nuclear receptor family and a master modulator of glucose and lipid metabolism, organelle differentiation, and inflammation (Guo et al., 2017;Zhao et al., 2016). These form heterodimers with the retinoid X receptor and bind to peroxisome proliferator response elements (PPREs) in the promoter region of respective target genes (Hallenborg et al., 2016).
Growing evidence indicates that PPARc agonists efficiently display neuroprotective properties in response to harmful insults, particularly neuroinflammation. Activation of PPARc by troglitazone and pioglitazone reduces infarct volume by improving neurological function following middle cerebral artery occlusion in rats (Corona & Duchen, 2016;Culman, Zhao, Gohlke & Herdegen, 2007). Also, PPARc activation by rosiglitazone imparts antidepressant-and anxiolytic-like effects (Guo et al., 2017).
Dysfunction of autophagy and neuroinflammation have been implicated in the pathogenesis of neurodegenerative diseases (Deretic & Klionsky, 2018;Keller & Lunemann, 2018). Autophagic inhibition is involved in lipopolysaccharide (LPS)-induced microglial activation in cultured primary microglia and the mouse brain (He et al., 2018). Similarly, FAK-family interacting protein of 200 kDa (FIP200) ablation and autophagy inhibition in neural stem cells lead to their defective differentiation by increased infiltration and activation of microglia (Wang, Yeo, Haas & Guan, 2017). These findings indicate that autophagy deficiency might regulate microglial activation. Although there is growing evidence suggesting that PPARc is a master regulator of microglial M2 polarization in immune disease, such as multiple sclerosis (MS) and experimental allergic encephalomyelitis (EAE; Kaiser et al., 2009;Orihuela, McPherson & Harry, 2016;Xu & Paul, 2007), the precise mechanisms remain unclear. Therefore, this study aimed to investigate the effects of PPARc in regulating microglial polarization and to explore the involved mechanisms. Our results revealed that antagonizing PPARc facilitates the LPS-induced switch of microglial polarization from the M1 phenotype to the M2 phenotype, which may provide a potent therapeutic strategy for related neuroinflammatory diseases.
NFjB is a key transcription factor that upregulates various proinflammatory mediators, and it is essential for both M1-and M2-like microglial differentiation. After LPS stimulation, a significant increase in the phosphorylation of NFjB and IKKb was observed (Figure 1r

| Antagonizing PPARc reverses LPS-induced autophagy inhibition in primary cultured microglia
Numerous studies have shown the important roles of autophagy in microglial inflammation and phenotype shift (Liu et al., 2015;Yang et al., 2014;Zhang, Guo, Zhao, Shao & Zheng, 2016). However, whether PPARc alters LPS-induced microglial polarization by modulating autophagy remains unclear. As the conversion of nonlipidated LC3-I to lipidated LC3-II is a classical marker of autophagic activity (Mizushima & Yoshimori, 2007), the ratio of LC3-II/LC3-I was determined by Western blotting to demonstrate the effects of the PPARc antagonist on autophagy. LC3-II expression was enhanced in cells cotreated with T0070907 and LPS, and it was higher than that in the cells treated with LPS alone (Figure 2a). Double immunofluorescence further demonstrated an increase in the accumulation of immunostained LC3-II puncta (green), which colocalized with Iba1 (red) in the cytoplasm of T0070907 + LPS-treated cells (Figure 2e).
Complete autophagy requires the fusion of autophagosomes with lysosomes. To further evaluate the impact of T0070907 on autophagic flux, we transfected a tandem construct mCherry-EGFP-LC3B plasmid into microglial cells. In general, LC3 appears as a diffuse pattern in the cytoplasm. Upon activation of autophagy, LC3 undergoes aggregation, generating a punctate pattern. The EGFP signal is sensitive to the formation of low-pH autolysosomes where it is lost, whereas mCherry is more stable. Therefore, the appearance of red dots indicates activated autophagic flux, which is characterized by the successful fusion of autophagosomes with lysosomes. However, the yellow punctum, which is colocalized in both EGFP and mCherry fluorescence, indicates a compartment that has not fused with a lysosome. Figure

| Liver kinase B1 (LKB1) activation is necessary
for PPARc-mediated regulation of adenosine 5 0monophosphate (AMP)-activated protein kinase (AMPK) phosphorylation in primary cultured microglial cells As a regulator of autophagy (Kim et al., 2013;Mack, Zheng, Asara & Thomas, 2014;Shang & Wang, 2011), AMPK phosphorylation was found to be reduced in microglial cells after LPS exposure. Furthermore, the increase in AMPKa2 expression and reduction in p-AMPK protein were restrained by the PPARc antagonist T0070907 (Figure 3a,b). As phosphorylated mammalian target of rapamycin (mTOR) and unc-51-like autophagy activating kinase 1 (ULK1) are involved in the AMPK controlling autophagic process, we examined their phosphorylation in the presence of the PPARc antagonist. Figure 3a,c,d shows that T0070907 not only increases the phosphorylation of ULK1, but also decreases the phosphorylation of mTOR in LPS-treated microglia. LKB1, calmodulin-dependent protein kinase kinase b (CaMKKb), and transforming growth factor b-activated kinase (TAK1) are upstream kinases that can activate AMPK by phosphorylating Thr172, which is situated in the activation loop of the a-subunit (Jeon, 2016). In our study, we found that LPS stimulation inhibits LKB1 phosphorylation, but it does not affect CaMKKb and TAK1 phosphorylation. Accordingly, the application of the PPARc antagonist induced the phosphorylation of LKB1 (Figure 3e-h). LKB1 binds with pseudokinase Ste20-related adaptor (STRAD) and scaffoldinglike adaptor protein mouse protein 25 (MO25) to form a complex, and it subsequently achieves full activation (Lin et al., 2015;. We further demonstrated the impact of PPARc antagonist on LKB1 activation using a co-immunoprecipitation (Co-IP) assay. The results showed that LPS reduced the binding capacity of LKB1, STRAD, and MO25 in microglial cell lysates without affecting their protein expression levels. Conversely, PPARc antagonization facilitated the formation of the LKB1-STRAD-MO25 complex F I G U R E 2 The PPARc antagonist T0070907 improves autophagy in microglia. (a-d) The PPARc antagonist T0070907 prevented LPSinduced decreases in LC3-II/LC3-I, Beclin1, and Atg5 and LPS-induced increases in P62. Data are presented as mean AE SEM, n ≥ 4, *p < .05, **p < .01, compared to the Con group; # p < .05, ## p < .01, compared to the LPS group. (e) Representative confocal images of LC3 puncta formation; LC3 (green) was colocalized with Iba1 (red) in cells treated with the PPARc antagonist T0070907 and LPS. Nuclei counterstained with Hoechst 33342. Scale bar = 20 lm. (f) Representative TEM images. Scale bar = 2 lm. High magnification of the boxed areas is shown below. Scale bar = 500 nm. Autophagosomes (blue arrows), autolysosomes (red arrows), and multilamellar body (yellow arrow). (g) Confocal microscopy analysis was used to measure autophagosomes and autolysosomes by monitoring the distribution and alteration of mCherry and EGFP fluorescent signals from mCherry-EGFP-LC3B. Scale bar = 20 lm. Microglia were pretreated with 3-MA (500 lM) or vehicle for 15 min, followed by treatment with 0.01 lg/ml LPS and 0.1 lM PPARc antagonist T0070907 for 24 hr. (h) Immunofluorescence staining of LC3 (green), CD63 (red), and nuclei (blue) in LPS, LPS+T0070907, and LPS+T0070907 + 3-MA groups. Scale bar = 20 lm. The protein expressions of LC3 (i), Beclin1 (j), p62 (k), and Atg (l) were determined by Western blotting. Data are presented as mean AE SEM, n ≥ 4, *p < .05, **p < .01, ***p < .001, compared to the Con group; # p < .05, ## p < .01, ### p < .001, compared to the LPS group; † p < .05, † † p < .01, † † † p < .001, compared to the LPS+T0070907 group ( Figure 3i,j). These findings suggest that the phosphorylation of LKB1 is crucial for the regulatory effects of PPARc on autophagy.

| Antagonizing PPARc enhances autophagy in primary cultured microglia via LKB1 activation
Radicicol, an inhibitor of LKB1, was used to further elucidate the role of LKB1 in T0070907-induced AMPK activation and autophagy.
Our results showed that radicicol reduced the viability of microglial cells at the concentrations of 0.25 and 0.5 lM (Figure 4a F I G U R E 3 Antagonizing PPARc reverses LPS-mediated inhibition of autophagy in microglial cells by activating AMPK. Microglial cells were pretreated with 0.01 lg/ml LPS, followed by treatment with 0.1 lM PPARc antagonist T0070907 for 24 hr. The upstream regulatory protein levels of autophagy, AMPK (a, b), ULK1 (a, c), and mTOR (a, d) were analyzed by Western blotting. Data are presented as mean AE SEM, n ≥ 4, *p < .05, **p < .01, ***p < .001, compared to the Con group; # p < .05, ## p < .01, ### p < .001, compared to the LPS group; † p < .05, † † p < .01, † † † p < .001, compared to the LPS+T0070907 group. (e-h) The PPARc antagonist T0070907 increased the phosphorylation of LKB1, but it did not change the phosphorylation of CaMKKb and TAK1 in the microglial cells treated with LPS. (i) An anti-LKB1 antibody was used for Dynabeads Protein G immunoprecipitation, and it detected the immunoprecipitates of MO25 and STRAD by Western blotting. (j) An anti-MO25 antibody was used for Dynabeads Protein G immunoprecipitation, and it detected the immunoprecipitates of LKB1 and STRAD by Western blotting transfection with 50, 100, and 200 lM siRNA-FM were respectively measured. Compared to siRNA-LKB1 (1331), siRNA-LKB1 (1898) and siRNA-PPARc (1049) induced a significant decrease in LKB1 expression (60%) at a concentration of 200 nM (Figure 4i). Thus, LKB1 siRNA (1049) was selected for transfection in the subsequent experiments. Our results showed that LKB1 knockdown prevented the T0070907-induced increase in LC3-II/LC3-I, Beclin1/Atg6, and Atg5 and reduction in p62/SQSTM1 (Figure 4j-m). Collectively, our data indicate that PPARc antagonist-mediated autophagy is dependent on the LKB1-AMPK signaling pathway.

| Antagonizing PPARc facilitates primary microglial M1-to-M2 polarization via LKB1 activation
The expression of CD86 and CD206 was used to quantify M1 and M2 microglia by flow cytometry. There is increasing evidence that autophagy is important for the induction and function of M2 phenotype microglia. In the current study, the ratio of LC3-II/LC3-I and the expression levels of autophagic markers, including Beclin1 and Atg5, were enhanced by T0070907. Furthermore, we found that T0070907 treatment improved autophagosome formation and the autophagosome-lysosome fusion. Moreover, T0070907 induced the significant colocalization of LC3 puncta and CD63-positive compartments in microglial cells. 3-MA blocked the T0070907-induced enhanced effects of autophagy. Therefore, our results reveal that PPARc antagonism can enhance autophagy, which contributes to microglial M2 polarization.
AMPK activates autophagy (Hindupur, Gonz alez & Hall, 2015) by directly or indirectly activating ULK1 (Egan et al., 2011;Kim, Kundu, Viollet & Guan, 2011). AMPK not only directly phosphorylates and activates ULK1 to induce autophagy, but also indirectly activates ULK1 by inhibiting mTORC1, which phosphorylates and inhibits ULK1 to disrupt the AMPK-ULK1 interaction (Inoki, Kim & Guan, 2012). This coordinated regulation of AMPK, mTORC1, and ULK1 eliminates damaged organelles and maintains mitochondrial integrity by initiating autophagy. Herein, our results show that autophagy is enhanced by the PPARc antagonist by suppressing the levels of p-mTOR and increasing the expression of p-AMPK and p-ULK1.
PPARc knockdown also increases the phosphorylation of AMPK, which coincides with the effects of the PPARc antagonist T0070907.
AMPK activity is increased when its Thr172 residue in the activation loop is phosphorylated by upstream kinases (Hindupur et al., 2015). In mammals, there are three activating kinases for AMPK, namely LKB1, CaMKKb, and TAK1. We found that the PPARc antagonist only promotes LKB1 phosphorylation against LPS-mediated AMPK inhibition, and no effects on the phosphorylation of CaMKKb and TAK1 were observed. Additionally, the transfection of cells with PPARc siRNA also significantly improved p-LKB1 protein levels. Furthermore, we found that the PPARc antagonist accelerates the formation of the LKB1-STRAD-MO25 complex. To confirm the crucial roles of LKB1 in PPARc antagonist-mediated microglial polarization, radicicol was used to inhibit LKB1 activation. We found that radicicol could potentially inhibit AMPK activity and autophagy. Subsequently, radicicol reversed the M1-to-M2 shift and inhibited T0070907-induced phagocytosis. To rule out the effects of other targets besides LKB1, we used LKB1 siRNA to knock down LKB1 expression in primary microglia. LKB1 knockdown prevented T0070907-induced increases in LC3-II/LC3-I, Beclin1/Atg6, and Atg5 expression, as well as the reduction in p62/SQSTM1 expression. Similar to radicicol, LKB1 knockdown could significantly inhibit T0070907-mediated increases in the M2 microglial marker (CD206) and reductions in the M1 microglial marker (CD86). These results demonstrate that LKB1 activation is necessary for antagonizing PPARc-induced autophagy and the M1-to-M2 microglial phenotype shift.
To observe whether the PPARc antagonist T0070907 has offtarget effects, we used siRNA to knock down PPARc and then tested the effects of T0070907. We found that knocking down PPARc could reverse LPS-mediated activation of NFkB, inhibit LPSinduced microglial M1 polarization, and promote M2 polarization.
The PPARc antagonist T0070907 did not affect the above regulative effects of PPARc knockdown, suggesting that T0070907 acts on PPARc to regulate microglial polarization.

| CONCLUSION S
Our results reveal that antagonizing PPARc promotes an M1-to-M2 phenotypic shift in LPS-induced microglia, which might be due to improved autophagy via the activation of the LKB1-AMPK signaling pathway. The present study provides the first evidence for the critical role of the PPARc antagonist in microglial polarization, as well as providing a new perspective on microglia-mediated neuroinflammation.

| Primary microglial cell culture
Primary microglial cell cultures were performed as previously described and were isolated from 1-to 3-day-old postnatal Sprague-Dawley rats, which were purchased from Shanghai SIPPR-Bk Laboratory Animal Co. Ltd (Shanghai, China). All of the animal operational procedures were performed in accordance with the Institution for Animal Care and Use Committee and approved by Animal Core Facility of Nanjing Medical University. Briefly, primary cultures of glial cells were obtained from the cerebral cortices, which were F I G U R E 4 Phosphorylation of LKB1-AMPK is necessary for T0070907-mediated upregulation of autophagy. (a, b) Below the concentration of 0.05 lM, radicicol did not influence the viability of the microglia. After LPS stimulation for 15 min, the LKB1 inhibitor radicicol (0.025 and 0.05 lM) was applied for 15 min before treatment with PPARc antagonist T0070907 for 24 hr. (c-g) Radicicol turned over the protein expressions of LKB1, AMPK, LC3, Beclin1, and p62 in the microglia treated with LPS and T0070907. Data are presented as mean AE SEM, n ≥ 4, *p < .05, **p < .01, ***p < .001, compared to the Con group; # p < .05, ## p < .01, ### p < .001, compared to the LPS group; † p < .05, † † p < .01, compared to the LPS+T0070907 group. (h) Immunofluorescence staining with of LC3 (green), CD63 (red), and nuclei (blue) in the LPS, LPS+T0070907, and LPS+T0070907 + radicicol groups. Scale bar = 20 lm. (i) Quantitation of Western blotting data showing declines in LKB1, which was used to observe the efficiency of transfection. Microglia were pretreated with si-LKB1 (500 nM) or vehicle for 24 hr and then stimulated with LPS for 15 min and treated with PPARc antagonist T0070907 for 24 hr. LKB1 siRNA prevented T0070907-induced increases in LC3-II/LC3-I (j), Beclin1 (k), and Atg5 (m) and T0070907-induced decreases in P62 (l). Data are presented as mean AE SEM, n ≥ 4, **p < .01, ***p < .001 compared to the LPS group in the Ctrl; ## p < .01, ### p < .001, compared to the LPS+T0070907 group in the Ctrl earlier digested by 0. FBS, 100 U/ml penicillin, and 100 lg/ml streptomycin. Before the experiments, the percentage of the primary microglial cells was confirmed by Iba1 staining with over 97% purity.

| Drugs and treatment
All of the experiments were conducted 24 hr after the cells were seeded. Primary microglial cells were respectively treated with the PPARc antagonist rosiglitazone (Selleck Chemicals, Houston, TX, USA) or the PPARc antagonist T0070907 (Selleck Chemicals) diluted in dimethylsulfoxide (DMSO; Sigma) with a final concentration of 0.1 lM, which was applied after the treatment with LPS (0.01 lg/ml; Sigma-Aldrich, St. Louis, MO, USA). The BV2 cells were passaged before reaching confluence using 0.025% (v/v) trypsin/EDTA in PBS, and they were seeded on poly-D-lysine-coated dishes at a suitable density. Then, 10 lg/ml LPS was applied to differentiated BV2 cells 15 min before the treatment with 10 lM PPARc antagonist T0070907, and they were then cultured in a humidified 5% CO 2 -95% air environment at 37°C for 24 hr. Radicicol (0.025 lM and 0.05 lM; Selleck Chemicals) and 3-MA (500 lM; Sigma-Aldrich) were used as the LKB1 inhibitor and autophagy inhibitor, respectively.

| Transfection of primary microglial cells with plasmid
Microglial cells were cultured at the density of 15 9 10 4 cells per well in a 24-well plate at 37°C in a 5% CO 2 humidified atmosphere and grown to 60% confluency. Microglial cells were transfected with plasmid mCherry-EGFP-LC3 (Biogot Technology, Co., Ltd.) using Lipofectamine â MessengerMAX ™ reagent (Invitrogen, NY, USA) according to the manufacturer's instructions. F I G U R E 5 Blocking LKB1 reverses the T0070907-mediated transition between M1 and M2 phenotypes by suppressing NFjB. After LPS stimulation for 15 min, the LKB1 inhibitor radicicol (0.025 and 0.05 lM) was applied 15 min before treatment with PPARc antagonist T0070907 for 24 hr. (a) Fluorescence-activated cell sorting analysis of the microglia in the LPS, LPS+T0070907, and LPS+T0070907 + radicicol groups. Surface expression of CD86 and CD206 was detected in microglia by flow cytometry. The percentage of CD86 (b) and CD206 (c) cells in the microglia was determined. pHrodo ™ Green Zymosan BioParticles were added to the cells and imaged after 30, 60, 90, 120, and 150 min. The green staining in the microglial cells was due to Cell Tracker ™ Green. (d) The microglial cells showed the time course of red fluorescence increased, documenting the accumulation of pHrodo-conjugated zymosan bioparticles (1 lm in diameter) in the intracellular acidic environment corresponding to phagosomes. (e) The proportion between the red-stained cells and the total cells was calculated. Data are presented as mean AE SEM, n = 3, ### p < .001 compared to the LPS group in each time point; † p < .05, † † † p < .001, compared to the LPS+T0070907 group in each time point. Nonfluorescence appeared at a neutral pH outside of the cell. Scale bar = 20 lm. Radicicol (f) and knocking down LKB1 (i) reversed the PPARc antagonist T0070907-induced changes in the protein expression of the M1 markers (iNOS, green) and the M2 markers (CD206, red) by immunofluorescence staining. Scale bar = 20 lm. (g, h) Western blotting showed that radicicol prevented the decreases in iNOS and the increases in CD206 induced by T0070907. Data are presented as mean AE SEM, n = 4, ***p < .001, compared to the Con group; ## p < .01, ### p < .001, compared to the LPS group; † p < .05, † † † p < .001, compared to the LPS+T0070907 group. (j and k) Knocking down LKB reversed the PPARc antagonist T0070907-induced changes in the protein expressions of iNOS and CD206 by Western blotting. Data are presented as mean AE SEM, n = 4, ***p < .001, compared to Con group, respectively in Ctrl or in si-LKB1; ### p < .001, compared to LPS group, respectively in Ctrl or in si-LKB1 F I G U R E 6 Knocking down PPARc reduces the inflammatory response by promoting LKB-AMPK activation. (a) Labeled (FM) with green were used to observe the efficiency of transfection reagent in BV2 cells. Scale bar = 1,000 lm. (b) Images of dissociated microglial cells exposed to scrambled (Con, top panels) or PPARc siRNA, and labeled with antibodies to Iba1 (red) and PPARc (green), and colabeled with a nuclear stain (blue). Scale bar = 100 lm. (c) Quantitation of Western blotting data showing declines in PPARc. Statistical analysis was performed using one-way ANOVA followed by Bonferroni's post hoc test. Data are presented as mean AE SEM, n = 5, $$ p < .01 compared to the NC group. Microglia were pretreated with si-PPARc (500 nM) or vehicle for 24 hr and then stimulated with LPS for 15 min and treated with PPARc antagonist T0070907 for 24 hr. (d) PPARc antagonist T0070907 did not influence PPARc siRNA-mediated protein expressions of the M1 marker (iNOS, green) and the M2 marker (CD206, red). Scale bar = 20 lm. The experiment was repeated three times. (e, f) Western blotting was used to quantify the expressions of iNOS and CD206 in each group. Data are presented as mean AE SEM, n = 4, *p < .05, ***p < .001, compared to the Con group, respectively in Ctrl or in si-PPARc; ### p < .001, compared to the LPS group, respectively in Ctrl in Ctrl or in si-PPARc. PPARc siRNA inhibited the activation of NFjB and IKKb (e, f, and g). PPARc siRNA was also associated with a rise in p-LKB1 (e, h) and p-AMPK (e, i). The PPARc antagonist T0070907 did not affect the efficacy of PPARc siRNA. Data are presented as mean AE SEM, n ≥ 5, **p < .01, ***p < .001, compared to the LPS group in the control 10494-1-AP, RRID: AB_2263076, 1:5,000). The primary antibody was recycled and membranes were washed four times in Tris-buffered saline with Tween-20 (TBST) for 10 min each time. Goat antirabbit IgG (H + L) secondary antibody and HRP secondary antibodies (Thermo Fisher Scientific, Cat# AF3309, AB_228341, 1:10,000) were incubated for 1 hr at room temperature. After the removal of the secondary antibody, the membranes were washed thrice with TBST for 15 min each time. The relative proteins levels were detected by enhanced chemiluminescence reagent and ImageJ software.

| RNA extraction
Total RNA was extracted from the cells with different treatments using TRIzol reagent (Invitrogen Life Technologies, CA, USA), according to the protocol provided by the manufacturer. The concentration and purity of the total RNA were measured using a NanoDrop 2000 spectrophotometer (NanoDrop Technologies, Thermo Scientific, USA).
A reverse transcription kit was used to synthesize the complementary DNA, and the additional total RNA samples were stored at À80°C.

| Reverse transcription and real-time quantitative PCR
Total RNA (1 lg) was reverse-transcribed to synthesize the complementary DNA using a PrimeScript ™ RT Master Mix (TaKaRa, Japan).
According to the manufacturer's instructions, the reverse transcription reaction in the thermal cycler (Eppendorf, Germany) was incubated at 37°C for 15 min, activated at 85°C for 5 s, and then held at 4°C. Quantitative real-time PCR was performed using SYBR â Premix Ex Taq I (TaKaRa, Japan) to detect the mRNA levels on a QuantStudio 5 Real-Time PCR System (Applied Biosystems, USA). Each 10 lL reaction contained 19 SYBR Green Master Mix, forward and reverse primers (sequences are listed in Table 1) at 10 lM concentration, and a cDNA sample as the template. The PCR conditions were as follows: 95°C for 30 s, followed by 40 cycles at 95°C for 5 s and 58°C to 60°C for 34 s. The amplification specificity was validated by the presence of a single peak in the melting curves. GAPDH was used as the endogenous control, and the relative expression of the target genes was determined using the 2 (ÀDDct) method. To confirm the results, each sample was run in triplicate, and the experiments were repeated at least thrice.

| Immunofluorescence staining and immunofluorescence microscopy
Microglial cells were seeded at the density of 15 9 10 4 cells/well in a 24-well plate and treated with drugs in complete culture medium.
After 24 hr, the culture medium was removed, and after washing twice with PBS, the cells were fixed with 4% paraformaldehyde at room temperature for 30 min and rinsed thrice in PBS for 5 min each time. Then, the cells were blocked in 0.1% Triton X-100 and 3% BSA in PBS for 1 hr at room temperature and probed with primary antibodies overnight at 4°C. After they were washed, the cells were treated with FITC-labeled Alexa Fluor-488-and/or Alexa Fluor-T A B L E 1 Primer sequences using qPCR

| Laser scanning confocal microscopy
Confocal microscopy was used to measure autophagosomes and autolysosomes by monitoring the distribution and alteration of mCherry and EGFP fluorescent signals from mCherry-EGFP-LC3B in microglial cells treated with LPS and/or T0070907.

| Co-IP assay
Microglial cells were grown to 80%-90% confluency in a six-well plate and treated with LPS (0.01 lg/ml) and the PPARc antagonist Immunoprecipitates were collected by centrifugation at 547 g for 5 min at 4°C and washed with 1 ml of RIPA buffer containing PMSF, and this was repeated four times. The supernatant was aspirated and discarded, and the resuspended pellet was denatured with 59 loading buffer. After boiling, the immunoprecipitated complexes were separated by SDS-PAGE and analyzed by Western blotting using the following specific antibodies: anti-LKB1 (Proteintech; 1:1,000 dilution), anti-MO25 (CST, MA, USA; 1:1,000 dilution), and anti-STRAD (Abcam, Chicago, IL, USA; 1:2,000 dilution) as described previously.

| TEM analysis
To observe the autophagy, the ultrastructural analysis was performed as described previously (Liu et al., 2016

| Phagocytosis
Primary microglial cells were plated on a poly-(L-lysine)-coated confocal plate, incubated overnight in culture medium, and then processed according to the pHrodo ™ Red zymosan bioparticle procedure of Invitrogen (Carlsbad, CA, USA). After two washes with PBS, the PBS was added with 1 lM Cell Tracker ™ Green for 20 min. After two washes with PBS, these were incubated at 37°C in the Opti-MEM medium containing 10 ll/ml of pHrodo Red zymosan bioparticles with or without LPS (0.01 lg/ml) and/or T0070907 (0.1 lM) and/or radicicol (0.05 lM) for 15 min. The treated microglia were examined every 30 min by confocal microscopy (Nikon A1RSi, Tokyo, Japan).  Table 2). More than 90% knockdown of the targeted proteins was observed after 500 nM siRNA treatment. Scrambled siRNA and target gene-specific siRNAs were purchased from GenePharma (Shanghai, China).

| Flow cytometry
The microglia were collected and washed twice with PBS, and blocked by 0.1% Triton X-100 and 3% BSA in PBS. The microglia were then stained with CD86-FITC (rat) and CD206-PE (rat) antibodies (BD Biosciences, La Jolla, CA, USA), respectively. The microglia were examined with a BD FACSVerse flow cytometer (BD Biosciences), and all of the tests were controlled by the homologous isotype control antibodies.  Table 3). More than 90% knockdown of the targeted proteins was observed after 500 nM siRNA treatment. Scrambled siRNA and target gene-specific siRNAs were purchased from GenePharma (Shanghai, China).

| Statistical analysis
The obtained data are presented as mean AE SEM of at least three independent experiments. The relationship between two factors was analyzed using Pearson correlation analysis, and bootstrapping was used in paired-samples tests. Groups were compared using a twoway ANOVA with post hoc Bonferroni's multiple comparisons test and one-way ANOVA with post hoc Tukey's multiple comparisons test. All of the data were analyzed with GraphPad Prism 6.0 software. A value of p < .05 indicated that the difference was statistically significant.

DISCLOSURE
The authors declare that they have no conflict of interests to report.

ACKNOWLEDGMENTS
The