We used murine microglial BV2 cells to examine the molecular mechanism of autophagy and the activation of microglia. BV2 cells were cultured in high-glucose DMEM (Solarbio, China) supplemented with 10% FBS (Gibco, United States) and 1% penicillin-streptomycin liquid (Solarbio, China). BV2 cells were cultured at 37°C in a humidified incubator with 95% air and 5% CO2. Lipopolysaccharide (LPS) was obtained from sigma (L2630, United States). 3-MA was obtained from sigma (189490, United States).

The pEGFP-C1-Trem2-m (P9057) plasmid and the pEGFP-C1 (P0314) were obtained from Biolink (Miaoling, China). Plasmids were transfected using the Lipofectamine 3000 kit (Invitrogen, Carlsbad, CA, United States).

Six- to eight-week-old male C57BL/6 mice were purchased from Hunan Sja Laboratory Animal Co., Ltd. Male mice were selected to rule out the neuroprotective activities of estrogens, progesterone and certain other neuroactive steroids on mesencephalic DA neurons (Isenbrandt et al., 2021). The animals were fed in a controlled environment and provided standard rodent chow and water. The implementation of animal care and procedures was approved by the Ethics Committee of the Second Affiliated Hospital of Nanchang University and complied with the guidelines for laboratory animal care and use from the National Institutes of Health (NIH). The mice were divided into the following four groups: saline, 0, 7, and 21 days. Then, the mice in the 0, 7, and 21 days groups received 1 i.p. injection of MPTP-HCl (20 mg/kg free base, M0896, Sigma, United States) per day for five consecutive days (Yao et al., 2019). The mice in the saline group received 1 i.p. injection of saline per day for 5 days. The mice were decapitated at specific times, their brains were removed, and midbrain samples including the substantia nigra pars compacta (SNpc) were stored at −80°C.

The lentivirus vectors encoding the mouse TREM2 gene (NCBI ID: NM_031254.3) into the multicloning site of the lentivirus backbone plasmid of pHBLV-CMV-Zsgreen-puro. The constructs were cotransfected with packaging vectors into 293T cells for packaging followed by purification (Hanbio Co. LTD., Shanghai, China). Stereotaxic surgery was performed with a stereotaxic frame (Stoelting, Wood Dale, IL, United States) and a 5 μl Hamilton syringe fitted with a pulled glass capillary tube. After anesthesia, the head of the mouse was fixed in the flat-skull position for stereotactic operation. Lentiviruses (3 μl; 1 × 10¹⁰ IFU/μl per construct) were stereotaxically delivered into the right substantia nigra region (AP: 3.1 mm, ML: 1.2 mm, DV: 5.1 mm from bregma) as previously reported (Ye et al., 2018). One week later, the mice were treated with saline or MPTP according to the above methods. All mice were killed at a specified time, and the midbrain was stored at −80°C.

Total RNA was extracted from cells and mouse brain tissue samples using TriQuick Reagent (Solarbio, China). A NanoDrop ND-2000 Spectrophotometer (Thermo Fisher Scientific, Inc., Wilmington, DE, United States) was used to quantify the RNA concentration. For assessments of mRNA expression, 1 μg of total RNA was used to synthesize the complementary DNA using the PrimeScript^TM RT Reagent Kit with gDNA Eraser (Takara, Japan). cDNA products were then diluted 1:10 in ddH2O. RT-qPCR was performed using the FastFire qPCR PreMix (Tiangen, China). The primer sequences were as follows: TREM2 (forward): 5′-CG GGATCCATGGGACCTCTCCACCAGTTTC-3′, (reverse) 5′-CCGCTCGAGAGCAAAAGTAGCAGAAACAGA-3′; GAPDH (forward): 5′-GGGAAATTCAACGGCACAGT-3′, (reverse) 5′-AGATGGTGATGGGCTTCCC-3′; TNF-a (Forward): 5′-TATG GCTCAGGGTCCAACTC-3′, (reverse) 5′- GGAAAGCCCA TTTGAGTCCT-3′; IL-1β (forward): 5′-CTCACAAGCAGA GCACAAGC-3′, (reverse) 5′-CAGTCCAGCCCATACTTT AGG-3′. The expression level of each sample was compared with the expression of GAPDH as an internal control. Data are from three separate experiments, and each experiment was performed in triplicate.

The protein levels of TREM2, LC3, p38, p-p38, mTOR, and p-mTOR in cultured cells and mouse brain tissue samples were detected by Western blotting. Total protein was extracted using RIPA lysis buffer (P0013B; Beyotime, China) containing protease and phosphatase inhibitors (P1082; Beyotime, China). Protein concentration was measured by bicinchoninic acid protein assay (Thermo Fisher Scientific, Inc., United States). Equal amounts of protein were separated using SDS-PAGE (BOSTER, China) and then transferred to a PVDF membrane (Solarbio, China). Membranes were then blocked with 5% non-fat dry milk in TBS with 0.1% Tween and probed with the appropriate primary antibodies against TREM2 (1:1,000, ABclonal, A10482), LC3B (1:1000, Cell Signaling, #2775), p-p38 (1:1000, Cell Signaling, #9216), p38 (1:1000, Cell Signaling, #9212), p-mTOR (1:1000, Cell Signaling, #2971), mTOR (Cell Signaling, #2972), GAPDH (1:1000, Servicebio, GB12002), and β-actin (Cell Signaling, #3700) overnight at 4°C. HRP-conjugated secondary antibodies (goat anti-mouse/rabbit IgG, 1:5000) were used for antibody detection.

BV2 cells (1 × 10⁵) were seeded onto 24-well plates and incubated overnight. Cells were transfected with TREM2 plasmid or NC plasmid for 24 h and then stimulated by LPS (1000 ng/ml) for 24 h. The concentrations of proinflammatory cytokines TNF-a and IL-1β in culture supernatants were determined by ELISA kits (PT512, PI301, Beyotime, China) according to the manufacturer’s protocol.

The rotarod test and pole-climbing test were used to evaluate the coordination and balance of mouse movement as previously reported (Han et al., 2020). In the rotarod test, the mice were first trained on the instrument at a constant speed of 20 rpm for 5 min. Then, the mice were placed on the accelerator rod (4 to 40 rpm) for 300 s, and the time the mice stayed on the accelerator rod (the latency of decline) was recorded. In the pole-climbing test, a thick wooden pole with a diameter of 1 cm and a length of 50 cm and a thick wooden ball with a diameter of 2 cm at the top were fixed vertically on the base. The mice were placed on the ball and allowed to climb down. The time from the release of the ball to the mouse forelimb reaching the bottom of the club was recorded. Each mouse carried out three consecutive experiments with intervals of 30 min, and the average time was recorded in the process of data processing.

Animals were anesthetized with sodium pentobarbital at 7 days after MPTP administration. Brains were removed, post-fixed in 4% paraformaldehyde for 24 h, and cryopreserved in 30% sucrose for 48 h. The frozen brains were then coronally sectioned at a thickness of 20 μm on a cryomicrotome, and the sections were mounted on slides. Dopaminergic neurons were identified using a primary antibody against TH (25859-1-AP, Proteintech) and a biotinylated secondary antibody and streptavidin ABC solution. Immunostaining images were obtained with an Aperio GT450 (Leica, Germany). For immunofluorescence staining, sections were first incubated at 4°C overnight with the following antibodies: mouse anti-Iba1 (Servicebio, GB12105),rabbit anti -NeuN (Servicebio, GB11138). This was followed by incubation with Alexa Fluor-488 or Alexa Fluor-546-conjugated secondary antibody at room temperature for 2 h. Immunofluorescent staining images were obtained with a Pannoramic MIDI (3DHISTECH, Hungary).

All data are expressed as the mean ± standard error of the mean (SEM). SPSS version 19 (SPSS, Inc., Chicago, IL, United States) was used for statistical analysis. A double-tailed unpaired t-test was used for comparisons between the two groups, and the difference was statistically significant. Single factor analysis of variance (ANOVA) was used for comparison among groups, and the Bonferroni post-test was used. When P < 0.05, the difference was statistically significant.

In our study, we evaluated the expression of TREM2 in BV2 cell lines induced by lipopolysaccharide (LPS) and found that TREM2 expression increased briefly after the cells were stimulated by LPS for a short period of time, and the expression level of TREM2 mRNA decreased with time (Figure 1A) (p < 0.05). According to our results, the change in TREM2 expression level was more striking after 16-h treatment. We treated the cells with LPS (0.5 or 1 μg/ml) for 24 h and then detected the expression levels of TREM2 and LC3 protein (Figure 1C). The results showed that the TREM2 protein level decreased significantly under LPS treatment (Figure 1D) (p < 0.01), and so did the expression of mRNA (Figure 1B) (p < 0.05). Moreover, pretreatment with LPS (0.5 or 1 μg/ml) for 24 h suppressed autophagy as determined by the ratio of LC3II/I (Figure 1E) (P < 0.05).

Because LPS stimulated the expression of TREM2, we speculated that it could play a role in the activation of the BV2 cell line. After 24 h of LPS stimulation, the level of LC3-II in BV2 cells decreased, and autophagy was inhibited. Therefore, we then overexpressed TREM2 in the BV2 cell line and further verify the role of TREM2 overexpression in microglia. We verified the expression of TREM2 after TREM2 plasmid or control plasmid transfection by RT-qPCR (Figure 2A) (p < 0.01). In addition, we evaluated the TREM2 protein level by WB (Figures 2B,C) (p < 0.01). Compared to the NC group (transfected control plasmid), the cells transfected with TREM2 plasmid showed a higher ratio of LC3II/I following a 24 h incubation with LPS (Figures 2D,E) (all P < 0.05). Earlier studies have shown that autophagy is mainly regulated by the mTOR pathway. It was previously shown that LPS stimulation can activate the mTOR pathway (Sun et al., 2017). We have shown that TREM2 can enhance microglial autophagy induced by LPS, but it is not clear whether this process is carried out through the mTOR pathway. Then we evaluated the expression level of p-mTOR under LPS treatment after we overexpressed the TREM2. We found that, compared with the NC group, overexpressed the TREM2 inhibited the expression of p-mTOR and reduced the ratio of p-mTOR/mTOR induced by LPS (Figures 2F,G) (P < 0.01). To further elucidate the mechanism of TREM2-induced autophagy, we transfected TREM2 overexpression plasmid into BV2 cells and then we treated BV2 cells with 1 or 5 mM 3-MA for 24 h to inhibit autophagy. Overexpression of TREM2 could inhibit the expression of p-mTOR and promote autophagy, while 3-MA can inhibit this process (Figures 2H–K). This suggested that TREM2 regulate autophagy through mTOR pathway.

We demonstrated that the expression level of TREM2 decreased in the activated BV2 cells. The downregulation of TREM2 in response to LPS stimulation suggested that TREM2 might be involved in the regulation of the microglial response to LPS. We then determined whether the over-expression of TREM2 could attenuate LPS-induced BV2 microglial activation. To assess this issue, the over-expression of TREM2 in microglia was examined. We transfected TREM2 overexpression plasmid into BV2 cells and evaluated the expression level of p-p38 under LPS treatment (Figure 3A). Compared to the NC group, the cells transfected with TREM2 overexpression plasmid showed a lower ratio of p-p38/p38 following a 24 h incubation with LPS (Figure 3B) (P < 0.05). Over-activated microglia can mediate the detrimental effects of neurotoxicity and inflammation through the excess production of cytotoxic factors (Leng and Edison, 2020). We then investigated the expression of TNF-α and IL-1β mRNA and protein levels by RT-qPCR and ELISA following a 24 h incubation with LPS. We found that the expression of LPS-induced TNF-α and IL-1β in the TREM2 OE group was significantly lower than that in the NC group (Figures 3C,D) (P < 0.05). Taken together, the over-expression of TREM2 could effectively suppress LPS-induced microglial activation.

To further verify the role of TREM2 in Parkinson’s disease, we constructed an MPTP PD model. In this experiment, mice were injected with saline or MPTP at different times, and the changes in TREM2 expression were measured. First, we confirmed by immunohistochemistry the damage to dopaminergic neurons induced by MPTP (Figure 4A). Next, we manually count the number of TH-positive cells in the SNpc (the area between two red lines) (Figure 4B). Then, we observed by immunofluorescence that the activation level of microglia in the mouse brain increased after the MPTP injection (Figure 4C). In order to verify whether TREM2 changes in the brain of mouse induced by MPTP, we detected the expression level of TREM2 by WB. We found that the expression level of TREM2 increased on the 7th day and then returned to normal on the 21st day (Figures 4D,E), which is consistent with previous reports (Zhang et al., 2018). At the same time, we verified the damage of MPTP on the motor ability of mice through behavioral experiments, the results of pole-climbing test and rotarod test for bradykinesia showed that MPTP impaired the motor ability of mice (Figures 4F,G). We speculate that the expression of TREM2 could antagonize the activation of microglia.

Given that TREM2 regulates the expression of p-p38/p-mTOR, which is involved in the pathogenesis of inflammation, to inhibit inflammation in vitro, we next studied whether TREM2 can inhibit the activation of microglia in a PD mouse model. As reported in previous studies, we injected HBLV-TREM2 or HBLV-NC into the right SNpc (AP: 3.1 mm, ML: 1.2 mm, DV: 5.1 mm from bregma) (Ye et al., 2018). One week after the lentivirus injection, we detected the infection efficiency of the lentivirus by observing the fluorescence of the lentiviral GFP marker, we found that the lentivirus has different degrees of infection to microglia and neurons in SNpc of brain (Figure 5A). Concurrently, we examined the mRNA and protein expression of TREM2 in SNpc by RT-qPCR and WB (Figures 5B–D). One week after the lentivirus injection, we began to inject MPTP into the mice. Then, on the 7th day after the completion of the MPTP injection, we conducted a follow-up experiment. Through the immunohistochemical experiment, by manually counting the number of TH-positive cells in the SNpc, we found that the dopaminergic neurons of MPTP mice injected with TREM2 lentivirus were protected compared with those of the NC group (Figures 6A,B). Next, we observed by immunofluorescence that the activation level of microglia in the mouse brain (Figure 6C). Finally, we verified the protection of TREM2 on the motor ability of PD mice through behavioral experiments, the motor ability of MPTP-treated mice injected with HBLV-TREM2 was improved compared with the NC group (Figures 6D,E).

To further verify the mechanism of TREM2 in the PD model, we observed the expression level of LC3, p-p38, p38, p-mTOR, mTOR by WB (Figures 7A,B). Studies have shown that MPTP can promote the expression of p-p38, while TREM2 can inhibit this process (Figure 7C). Compared with the NC group, TREM2 inhibited the expression of p-mTOR, activated autophagy and promoted the expression of LC3 (Figures 7D,E).