SIRT1 activation by 2,3,5,6-tetramethylpyrazine alleviates neuroinflammation via inhibiting M1 microglia polarization

Abstract

Background: Neuroinflammation has been reported as a potential contributing factor to brain diseases, and is characterized by activated microglia with release of multiple inflammatory mediators. 2,3,5,6-Tetramethylpyrazine (TMP) is an active alkaloid in Ligusticum chuanxiong Hort. and has various biological activities, including anti-inflammatory and neuroprotection properties. However, the anti-neuroinflammatory activity of TMP has been less studied and its potential molecular mechanisms in this field remain unclear. This study aimed to investigate the effects of TMP and its underlying mechanisms in neuroinflammation.

Methods: In vitro, lipopolysaccharide (LPS)-stimulated BV2 microglia were used to assess the effects of TMP on inflammatory cytokines as well as the components of the SIRT1/NF-κB signaling pathway, which were measured by using ELISA, western blotting, qRT-qPCR and immunofluorescence. Moreover, LPS-induced acute neuroinflammation model in mice was performed to detect whether TMP could exert anti-neuroinflammatory effects in vivo, and the EX527, a SIRT1 inhibitor, were given intraperitoneally every two days prior to TMP treatment. Serums and spinal trigeminal nucleus (Sp5) tissues were collected for ELISA assay, and the Sp5 tissues were used for HE staining, Nissl staining, immunofluorescence, qRT-PCR and western blotting.

Results: In vitro, TMP treatment significantly reduced the secretion of pro-inflammatory cytokines, including TNF-α and IL-6, promoted SIRT1 protein expression and inactivated NF-κB signaling pathway in LPS-induced neuroinflammation. Interestingly, pretreatment with EX527 blocked the therapeutic effects of TMP on neuroinflammation in vitro. Furthermore, TMP reduced the levels of pro-inflammatory cytokines and chemokines, and prevented microglia from polarizing towards a pro-inflammatory state through activating SIRT1 and inhibiting NF-κB activation in LPS-induced neuroinflammation in mice. And EX527 reversed the beneficial effects of TMP against LPS exposure in mice.

Conclusion: In summary, this study unravels that TMP could mitigate LPS-induced neuroinflammation via SIRT1/NF-κB signaling pathway.

Keywords: 2; 3; 5; 6-tetramethylpyrazine; NF-κB; SIRT1; microglia polarization; neuroinflammation.

Figure 1

TMP ameliorated neuroinflammatory responses in LPS-stimulated BV2 cells. (A) Structure of TMP. (B) BV2 cells were treated with TMP at different concentrations range from 1.56-200 μM for 24 h, and cell viability was determined using MTT assay (n=18). (C) BV2 cells were treated with LPS (0.039-10 μg/mL) for 24 h, and cell viability was determined using MTT assay (n=18). (D) BV2 cells were with TMP (1.56-200 μM) with LPS (0.5 μg/mL) for 24 h, and cell viability was measured (n=18). (E, F) BV2 cells were pretreated with 50, 100, and 200 μM TMP for 4 h and then stimulated by LPS (0.5 μg/mL) for 18 h The levels of TNF-α (E) and IL-6 (F) were determined by ELISA kits (n=3). Data are expressed as mean± SD. ^## P < 0.01 vs. Control, ^* P < 0.05 vs. LPS, ^** P < 0.01 vs. LPS.

Figure 2

TMP inhibited NF-κB signaling pathway in LPS-induced BV2 cells. Cells were pretreated with 50, 100, and 200 μM TMP for 4 h and then stimulated with LPS (0.5 μg/mL) for 1h. (A) Representative images of blot for p-IKKα/β, IKKα, IKKβ, p-IκBα, IκBα, p-p65 and p65. (B) Data are normalized to the mean value of LPS group, and quantitative analysis of p-IKKα/β/IKKα/IKKβ, p-IκBα/IκBα, p-p65/p65 were detected by Image J Data are expressed as mean± SD, n=3. ^## P < 0.01 vs. Control, ^* P < 0.05 vs. LPS, ^** P < 0.01 vs. LPS.

Figure 3

The anti-neuroinflammatory effects of TMP in LPS-induced BV2 cells were partially mediated by SIRT1. (A) BV2 cells were pretreated with 50, 100, and 200 μM TMP for 4 h and then stimulated with LPS (0.5 μg/mL) for 18 h. The protein expression of SIRT1 was determined by Western blotting (n=3). Data were normalized to the mean value of the LPS group. (B–E) SIRT1 inhibitor EX527 (10 μM) was given to BV2 cells 0.5 h prior to TMP (200 μM) treatment, 4 h later, BV2 cells were stimulated with LPS (0.5 μg/mL) for 12 h. The mRNA levels of TNF-α (B), IL-6 (C), COX-2 (D) and iNOS (E) were analyzed by qRT-PCR, and normalized to 18S (n=6). (F) BV2 cells were treated with TMP (200 μM) or EX527 (10 μM) + TMP (200 μM) for 4 h and subsequently induced with LPS (0.5 μg/mL) for 1 h. Representative immunofluorescence staining for p65 in the different groups. Nuclei were stained with DAPI. (G) The percentage of cells with p65 nuclear staining was evaluated from the immunostaining images and analyzed by Image J (n=6). Data are expressed as mean± SD. ^# P < 0.05 vs. Control, ^## P < 0.01 vs. Control, ^* P < 0.05 vs. LPS, ^** P < 0.01 vs. LPS, &P < 0.05 vs. TMP+LPS, &&P < 0.01 vs. TMP+LPS.

Figure 4

TMP ameliorated neuroinflammatory responses in LPS-induced mice. (A) Establishment of a mouse model of acute neuroinflammation. (B) Body weight in each group of mice (n=5). (C, D) The levels of TNF-α (C, E) and IL-6 (D, F) in serum (C, D) and Sp5 tissues (E, F) were determined by ELISA kits (n=5). (G, H) qRT-qPCR analysis of cytokines (G) and chemokines (H) in the Sp5 tissues of mice (n=5). Data are expressed as mean± SD. ^# P < 0.05 vs. Control, ^## P < 0.01 vs. Control, ^* P < 0.05 vs. LPS, ^** P < 0.01 vs. LPS, &P < 0.05 vs. TMP+LPS, &&P < 0.01 vs. TMP+LPS.

Figure 5

TMP mitigated LPS-induced lesion in the Sp5 of brain tissue in mice. (A) Representative HE staining images of Sp5. (B–F) Quantification of HE staining, including HE deep staining intensity (B), total area of cell nucleus (C), number of total cells (D), number of normal cells (E) and number of damaged cells (F). (G) Representative Nissl staining images of Sp5. (H–L) Quantification of Nissl staining, including Nissl body intensity (H), area of Nissl bodies (I), number of Nissl bodies (J), number of normal cells (K) and number of deep staining cells (L). Data are expressed as mean± SD, n=5. ^# P < 0.05 vs. Control, ^## P < 0.01 vs. Control, ^* P < 0.05 vs. LPS, ^** P < 0.01 vs. LPS, &P < 0.05 vs. TMP+LPS, &&P < 0.01 vs. TMP+LPS.

Figure 6

TMP modulated the microglial M1/M2 polarization in LPS-induced mice. (A) Representative immunofluorescence staining images of M1 microglia markers iNOS (green) and Iba-1 (red) in the Sp5 of LPS-induced mice (B, C) Quantification of iNOS intensity (B) and the number of iNOS⁺Iba-1⁺cells (C). (D) Representative immunofluorescence staining images of M1 microglia markers CD86 (green) and Iba-1 (red) in the Sp5 of LPS-induce mice. (E, F) Quantification of CD86 intensity (E) and the number of CD86⁺Iba-1⁺cells (F). (G) Representative immunofluorescence staining images of M2 microglia markers CD206 (green) and Iba-1 (red) in the Sp5 of LPS-induce mice. (H, I) Quantification of CD206 intensity (H) and the number of CD206⁺Iba-1⁺cells (I). Data are expressed as mean± SD, n=5. ^## P < 0.01 vs. Control, ^** P < 0.01 vs. LPS, &P < 0.05 vs. TMP+LPS, &&P < 0.01 vs. TMP+LPS.

Figure 7

TMP prevented neuroinflammation via SIRT1/NF-κB pathway in LPS-treated mice. (A) RT-qPCR analysis of M1 microglia markers (COX2, iNOS, and CD86) and M2 microglia markers (Arg-1 and CD206) in the Sp5 tissue of mice. (B) Representative blots of SIRT1, p-IKKα/β, IKKα, IKKβ, p-IκBα, IκBα, p-p65 and p65 in the Sp5 tissue of mice. α-Tubulin was used as an internal loading control. (C) Data are normalized to the mean value of LPS group, and quantitative analysis of SIRT1, p-IKKα/β, p-IκBα, p-p65 was measured by Image J. Data are expressed as mean± SD, n=5. ^## P < 0.01 vs. Control, ^* P < 0.05 vs. LPS, ^** P < 0.01 vs. LPS, &P < 0.05 vs. TMP+LPS, &&P < 0.05 vs. TMP+LPS.

Figure 8

Schematic models of molecular targets of TMP in neuroinflammation. Neuroinflammation is characterized by the secretion of inflammatory mediators and the over-activation of microglia. Therapeutically, TMP exert anti-neuroinflammatory effects through activating SIRT1 and inhibiting NF-κB signaling pathway.