Anti-Inflammatory Effects of Dimethyl Fumarate in Microglia via an Autophagy Dependent Pathway

Abstract

Dimethyl fumarate (DMF), which has been approved by the Food and Drug Administration for the treatment of relapsing-remitting multiple sclerosis, is considered to exert anti-inflammatory and antioxidant effects. Microglia maintain homeostasis in the central nervous system and play a key role in neuroinflammation, while autophagy controls numerous fundamental biological processes, including pathogen removal, cytokine production, and clearance of toxic aggregates. However, the role of DMF in autophagy induction and the relationship of this effect with its anti-inflammatory functions in microglia are not well known. In the present study, we investigated whether DMF inhibited neuroinflammation and induced autophagy in microglia. First, we confirmed the anti-neuroinflammatory effect of DMF in mice with streptozotocin-induced diabetic neuropathy. Next, we used in vitro models including microglial cell lines and primary microglial cells to examine the anti-inflammatory and neuroprotective effects of DMF. We found that DMF significantly inhibited nitric oxide and proinflammatory cytokine production in lipopolysaccharide-stimulated microglia and induced the switch of microglia to the M2 state. In addition, DMF treatment increased the expression levels of autophagy markers including microtubule-associated protein light chain 3 (LC3) and autophagy-related protein 7 (ATG7) and the formation of LC3 puncta in microglia. The anti-inflammatory effect of DMF in microglia was significantly reduced by pretreatment with autophagy inhibitors. These data suggest that DMF leads to the induction of autophagy in microglia and that its anti-inflammatory effects are partially mediated through an autophagy-dependent pathway.

Keywords: DMF; anti-inflammation; autophagy; microglia; neuroinflammation.

FIGURE 1

Dimethyl fumarate reduces gliosis in the spinal cord of mice with STZ-treated diabetic neuropathy (A) C57BL/6 mice were injected with STZ (100 mg/kg, i.p.). After 3 days, the mice that developed diabetes were administered dimethyl fumarate (DMF, 45 mg/kg i.p.) once a day for four weeks. At the end of treatments, spinal cord sections were prepared and stained with anti-Iba-1 (B–E) or anti-glial fibrillary acidic protein (GFAP) antibody (F–H). White dotted lines represent the dorsal horn of the spinal cord. Representative Iba-1⁺ microglia are shown in the magnified images. The quantified relative fluorescence intensity for Iba-1⁺ cells (C), number of Iba-1⁺ cells (D), and the area of Iba-1⁺ cells in the spinal cord (E) were analyzed. Representative GFAP⁺ astrocytes are shown in the magnified images (F). The quantified relative fluorescence intensity for GFAP⁺ cells (G) and the area of GFAP⁺ cells in the spinal cord (H) were analyzed (n = 3 mice/group, 3 images/mouse). Data are presented as means ± standard error of the mean (SEM). *P < 0.05 compared with STZ-treated mice from ANOVA, Tukey’s multiple comparison test.

FIGURE 2

DMF reduces inflammation in microglia. HAPI cells were seeded at 4 × 10⁴ cells/well in a 96-well plate and incubated overnight. Cells were then treated with indicated concentration of DMF with or without LPS. After 24 h, nitric oxide (NO) production was measured by the Griess assay (A), and cell viability was measured by the MTT assay (B).(C, D) HAPI cells were seeded at 1 × 10⁶ cells/well in a 6-well plate and incubated overnight. Cells were then treated with 4 μM DMF with or without LPS (100 ng/ml). After 6 h, the mRNA expression levels of tumor necrosis factor-alpha (TNF-α) (C) and interleukin- 6 (IL-6) (D) was determined by real-time quantitative polymerase chain reaction (RT-qPCR). Fold changes were calculated as the ratio of expression level in the LPS-only treated group. n = 3 independent experiments/group. Data are presented as means ± SEM. *P < 0.05 compared with LPS-only treated group, #P < 0.05 compared with vehicle-only treated group from two-way ANOVA with multiple comparison test.

FIGURE 3

DMF induces the neuroprotective M2 phenotype and phagocytosis in microglia (A, B) HAPI cells were seeded at 2.5 × 10⁵ cells/well in a 6-well plate and incubated overnight. Cells were then treated with LPS (100 ng/ml) or DMF (4 μM) plus LPS (100 ng/ml) for 24 h. The expression of arginase-1 protein (Arg1) was determined by Western blot analysis (B) Histogram shows the densitometric analysis of Western blots. β-actin was used as the loading control and data were normalized to total β-actin. n = 3 independent experiments/group (C, D) BV-2 cells were seeded at 5 × 10⁴ cells/well in a 24-well plate and incubated overnight. BV-2 cells were then treated with DMF (4 μM) with or without LPS (100 ng/ml) for 16 h, followed by the addition of opsonized zymosan-red particles into BV-2 cells and then incubated for another 4 h. The phagocytosed zymosan particles were counted and expressed as the number of zymosan particles per cell. A total of 463 control cells, 1117 DMF-treated cells, 990 LPS-treated cells, and 636 DMF + LPS-treated cells were analyzed. Results are representative of two independent experiments. Data are presented as means ± SEM. *P < 0.05, **P < 0.01 compared with LPS-treated mice from ANOVA, Tukey’s multiple comparison test (E,F) Primary microglia were seeded at 5 × 10⁴ cells/well in a 24-well plate and incubated overnight. The primary microglia were then treated with DMF (4 μM) for 48 h, followed by the addition of opsonized zymosan-red particles into the primary microglia and then incubated for another 2 h. The primary microglia were stained with the anti-Iba-1 antibody. The phagocytosed zymosan particles were counted and expressed as the number of zymosan particles per Ib1-1 positive microglial cell. A total of 157 control cells and 154 DMF-treated cells were analyzed. n = 3 independent experiments/group. Data are presented as means ± SEM. *P < 0.05 from Student’s t-test between indicated two groups.

FIGURE 4

DMF increases autophagy in microglia. HAPI cells were seeded at 2.5 × 10⁵ cells/well in a 6-well plate and incubated overnight (A) Cells were then treated with DMF (4 μM) for 3 h (B) Cells were treated with LPS (100 ng/ml) or DMF (4 μM) plus LPS (100 ng/ml) for 24 h. The protein levels of LC3-I and LC3-II were determined by Western blot analysis. β-actin was used as the loading control. n = 3 independent experiments/group (C) Histogram shows the densitometric analysis of Western blots. β-actin was used as the loading control and data were normalized to total β-actin (D) HAPI cells were seeded at 2 × 10⁴ cells/well in a 24-well plate and incubated overnight. Cells were then treated with DMF (4 μM) for 24 h and stained with the anti-LC3 antibody (E) The punctate area was determined from at least 3 independent images (Vehicle: 29 cells and DMF: 26 cells) using ImageJ and was expressed as the ratio of vehicle controls (F) HAPI cells were seeded at 1 × 10⁶ cells/well in a 6-well plate and incubated overnight. Cells were then treated with DMF (4 μM) for 6 h. The mRNA expression levels of Atg7 was determined by RT-qPCR. Fold changes were calculated as the ratio of expression level in untreated control cells (G) Primary microglia isolated from MGC were seeded at 5 × 10⁴ cells/well in a 24-well plate and incubated overnight. Cells were then treated with DMF (10 μM) for 24 h, followed by staining with the anti-LC3 antibody (H) The number of LC3 puncta per cell was counted. A total of 27 control cells and 38 DMF-treated cells were analyzed (I) MGC were seeded at 2 × 10⁵cells/well in a 12-well plate and incubated overnight. Cells were then treated with DMF (10 μM) for 6 h. The mRNA expression levels of Atg7 was analyzed by RT-qPCR. Fold changes were calculated as the ratio of expression level in untreated control cells. n = 3 independent experiments/group. Data are presented as means ± SEM. *P < 0.05 from Student’s t-test for indicated two groups. **P < 0.01 compared with vehicle from ANOVA, Tukey’s multiple comparison test.

FIGURE 5

DMF induces anti-inflammatory responses in MGC via an autophagy-dependent pathway (A, B) MGC were seeded at 2 × 10⁵ cells/well in a 12-well plate and incubated overnight. Cells were then pretreated with bafilomycin (10 nM, Baf) or SBI-0206965 (5 μM, SBI) for 1 h, followed by treatment with DMF (4 μM) and LPS (1 μg/ml). After 6 h, the mRNA expression levels of TNF-α (A) and IL-6 mRNA (B) were measured by RT-qPCR. Fold changes were calculated as the ratio of expression level in LPS-treated cells (C, D) MGC were seeded at 3 × 10⁴ cells/well in a 96-well plate and incubated overnight. Cells were then pretreated with bafilomycin (5 (gray) or 10 (black)nM) for 1 h, followed by the addition of DMF (4 μM) and LPS (1 μg/ml). After 48 h, NO production was measured by the Griess assay (C) and cell viability was measured by the MTT assay (D) (E, F) MGC were seeded at 3 × 10⁴ cells/well in a 96-well plate and incubated overnight. Cells were then pretreated with SBI-0206965 (5 μM, gray) or Bay 11–7082 (2.5 μM, Bay, black) for 1 h, followed by treatment with DMF (4 μM) and LPS (1 μg/ml). After 48 h, NO production (E) and cell viability (F) were measured. n = 3 independent experiments/group. Data are presented as means ± SEM. *P < 0.05 compared with LPS treated group, n.s., not significant (P > 0.05).

FIGURE 6

DMF increases phagocytosis via an autophagy-dependent pathway in microglia (A, B) Primary microglia were seeded 5 × 10⁴ cells/well in a 24-well plate and incubated overnight. Cells were then pretreated with SBI-0206965 (5 μM) for 1 h, followed by the addition of DMF (10 μM). After 24 h, opsonized zymosan-red particles were added into the cultures which were incubated for 2 h. The primary microglia were stained with the anti-Iba-1 antibody. The phagocytosed zymosan particles were counted and expressed as the number of zymosan particles per cell. The analysis included 78 control cells, 96 cells treated with DMF, and 110 cells treated with DMF plus SBI-0206965. n = 3 independent experiments/group. Data are presented as means ± SEM. *P < 0.05 from ANOVA.