BV2 microglia cells from Procell (China) were grown in MEM (Procell, China) with 10% FBS (Gibco, USA) at 37 °C and 5% CO₂ in a humidified air environment.

All in vivo experiments were approved by the Animal Ethics Committee of Chengdu University of Traditional Chinese Medicine (No. 2021-30). 20 male C57BL/6 mice (18-22g), provided by SPF Biotechnology Co., Ltd (No. SCXK 2019-0010, Beijing, China), were maintained at 22 ± 2°C temperatures and 55 ± 5% relative humidity under a 12 h light/dark cycle with free access to food and water ad libitum. A mouse model of acute neuroinflammation was established by a single injection of LPS as described previously (16). Upon adaptive feeding for one week, the mice were randomly divided into four groups (n=5): Control group; LPS group; TMP + LPS group; and TMP+EX527+LPS group. TMP+LPS and TMP+EX527+LPS group were injected intraperitoneally (i.p.) with 50 mg/kg TMP (Chengdu Must Bio-technology CO., Ltd, China) that was dissolved in saline with 20% PEG 300. Control and LPS group were treated with (i.p.) saline containing 20% PEG 300 once a day for consecutive 14 days. Starting from the 8th day, mice in the TMP+EX527+LPS group were i.p. with 5 mg/kg EX527 (Selleck, USA, dissolved in DMSO: PEG 300: saline, 5:30:65) every two days for a total of four times. On the 15th day, the LPS, TMP+LPS and TMP+EX527+LPS group were injected i.p. with 5 mg/kg LPS (Escherichia coli 0127: B8, Sigma-Aldrich, USA), and Control group were treated (i.p.) with saline. After 6 hours of LPS injection, the mice were anesthetized by inhalation with isoflurane and the blood and brains were collected.

BV2 cells were inoculated into 96-well plates (5×10³ cells/well). After 24 h, TMP with or without LPS at different concentrations was used to treat the cells for 24 h. MTT assay was used to determine cell viability as described previously (17).

Concentrations of TNF-α and IL-6 in the cell medium and the serum and Sp5 tissues from LPS-induced mice were measured by ELISA kits (Neobioscience, China).

Total RNA from BV2 cells and Sp5 tissues were derived with TRIzol Reagent (Invitrogen, USA) and reversed to cDNAs using RevertAid Master Mix with DNase I (Thermo Fisher Scientific, USA). cDNA samples, gene specific primers ( Table 1 ) and SYBR™ Green Master Mix (Thermo Fisher Scientific) were used to perform qPCR experiments according to the protocol, which were collected by ABI StepOnePlus PCR system (ABI-7500, Thermo Fisher Scientific). The relative mRNA expressions of genes were calculated by the 2^-△△CT method as described previously (18).

As described previously (19), proteins in BV2 cells and Sp5 tissues were extracted using RIPA buffer with protease inhibitor and PMSF (Beyotime, China). BCA protein assay kit (Thermo Fisher Scientific) was used to measure the protein concentration. Equal amounts of proteins sample (10-20 µg) were fractionated on SDS-PAGE and shifted onto PVDF membranes (Bio-Rad, USA). The membranes were blocked with 5% skimmed milk for 1 h and then incubated with primary antibodies ( Table 2 ) overnight at 4°C. After that, the blots were incubated with secondary antibody for 2 h at room temperature. The immune-blotting signals were visualized by chemiluminescence (Thermo Fisher Scientific) using Tanon 5200 (China). Quantitative analysis was analyzed by Image J software.

BV2 cells were fixed in 4% paraformaldehyde for 10 min, infiltrated with 0.5% Triton X-100 (Bio-Rad) for 20 min after washing 3 times with PBS and then incubated with anti-p65 antibody (1:400) at 4°C overnight, followed by Alexa Fluor 488 goat anti-rabbit IgG (1:1000) for 1 h and DAPI (1:1000, Beyotime, China) for 20 min. Images were acquired with a confocal microscope (FV-OSR, Olympus, Japan) and quantitated using ImageJ.

After the brain tissue was fixed with 4% paraformaldehyde and embedded in paraffin, 5 μm sections were cut for HE and Nissl staining (20). The images were obtained by an optical microscope (Ti2, Nikon, Japan) and quantitated using ImageJ.

As described previously (21), after deparaffinizing and rehydrating, the 5 μm sections of paraffin-embedded brain tissue were soaked with 10 mM citric acid (pH 6.0) and heated in a water bath to recover antigenicity, and then blocked with 10% BSA at room temperature for 1h. After that, sections were stained with primary anti-Iba-1 (1:200) and anti-CD206 (1:200) antibodies, anti-iNOS (1:200) antibodies, and anti-CD86 (1:200) antibodies at 4°C overnight, followed by fluorescently labeled secondary antibodies for 1 h and DAPI for another 20 min. A confocal microscope (FV-OSR, Olympus, Japan) was used to acquire images of the stained sections. The images were quantitated using ImageJ.

All data were analyzed using GraphPad Prism 9.4.0 software and were presented as mean ± SD. The differences between groups were performed by one-way ANOVA, and P < 0.05 were considered statistically significant.

Firstly, the cytotoxicity of different concentrations of TMP (structure in Figure 1A ) on BV2 cells was assessed by MTT assay. Results showed that TMP at 0-200 μM had no significant observable effect on cell viability with or without the presence of 0.5 μg/mL LPS ( Figures 1B–D ). Next, we assessed whether TMP exerted inhibitory effect on inflammatory cytokines in LPS-stimulated BV2 cells. TMP significantly restrained the production of TNF-α and IL-6 according to ELISA results ( Figures 1E, F ). Thus, TMP alleviates the LPS-induced neuroinflammatory response in BV2 cells.

According to recent studies, NF-κB signaling plays crucial roles in modulating neuroinflammation (22). LPS stimulation promoted phosphorylation of IKKα/β, p65 and IκBα, and TMP inhibited the phosphorylation of the above proteins ( Figures 2A, B ). Thus, TMP could suppressed the activation of NF-κB pathway.

SIRT1 negatively regulate the activity of NF-κB signaling pathway in the model of neuroinflammation (23). The expression of SIRT1 protein was decreased after LPS stimulation, and TMP significantly increased SIRT1 expression in LPS-induced BV2 cells ( Figure 3A ). In order to further detect whether the anti-neuroinflammatory effect of TMP is mediated by SIRT1, the SIRT1 inhibitor EX527 was used. As shown in Figures 3B–E , in LPS-stimulated BV2, TMP reduced the transcription of TNF-α, IL-6, COX-2 and iNOS, while EX527 reversed the effects of TMP on these mRNA expressions. Since p65 is a key nuclear transcription factor in the NF-κB pathway and its nuclear translocation is critical to the activation of the NF-κB pathway (24), subsequently we explored whether TMP affects the translocation of p65. In LPS-induced BV2 cells, the proportion of cells with p65 translocated to the nucleus was increased, TMP treatment reduced nuclear accumulation of p65, while EX527 reversed this change ( Figures 3F, G ). These data suggest that SIRT1 mediates, at least to some extent, the anti-neuroinflammatory effects of TMP in LPS-induced BV2 cells.

Next, to investigate the effects of TMP on neuroinflammation in vivo, LPS induced acute neuroinflammation model was established ( Figure 4A ). TMP had no significant effect on the body weight of mice ( Figure 4B ). Following LPS stimulation, the contents of TNF-α and IL-6 in serum and Sp5 tissues were significantly increased, whereas TMP administration reduced these cytokines levels ( Figures 4C–F ). Meanwhile, the results of qRT-PCR indicated that TMP treatment markedly decreased the expressions of cytokines, like TNF-α, IL-6 and IL-1β, as well as chemokines, including CCL2, CCL3, CCL5, and CXCL10 in the Sp5 of brain tissue from LPS-induced mice ( Figures 4G, H ). Additionally, data of HE staining and Nissl staining of Sp5 tissues showed that LPS administration caused obvious neuronal loss and morphological changes, including nuclear condensation, hyperchromatism and obvious eosinophilic cytoplasmic degeneration, while TMP ameliorated these pathological changes ( Figure 5 ). Notably, SIRT1 inhibitor EX527 reversed these effects of TMP ( Figures 4 , 5 ).

Activation and polarization of microglia is one of the characteristics of neuroinflammation (25). To identify whether the anti-neuroinflammatory effects of TMP was related to microglia polarization, immunofluorescence staining and qRT-qPCR were performed to evaluate microglia markers in the Sp5 tissue. Results indicated that compared with LPS group, TMP treatment suppressed the expressions of M1 markers iNOS ( Figures 6A–C ) and CD86 ( Figures 6D–F ), but facilitated the expression of M2 marker CD206 ( Figures 6G–I ). Additionally, the mRNA expression of M1 marker COX-2, iNOS and CD86 ( Figure 7A ) were significantly increased after LPS injection, while TMP inhibited the expression of M1 markers and increased the transcription of the M2 markers, including Arg-1 and CD206 ( Figure 7A ) in Sp5 tissue. And EX527 partially counteracted the effects of TMP on microglia polarization ( Figures 6 , 7A ). These data suggests that TMP could modulated the microglial M1/M2 polarization in acute neuroinflammation, partly mediated by SIRT1.

SIRT1 expression in the Sp5 was observably decreased after LPS injection, and TMP markedly reversed the reduction of SIRT1 ( Figures 7B, C ). Besides, after LPS injection, the expressions of p-IKKα/β, p-p65and p-IκBα were elevated, and TMP treatment notably prevented the changes. Consistently, the effects of TMP were partially blocked by EX527 ( Figures 7B, C ). These data suggest that TMP ameliorates the neuroinflammatory response induced by LPS through SIRT1/NF-κB signaling pathways.