FGF21 alleviates neuroinflammation following ischemic stroke by modulating the temporal and spatial dynamics of microglia/macrophages

Abstract

Background: Resident microglia and macrophages are the predominant contributors to neuroinflammation and immune reactions, which play a critical role in the pathogenesis of ischemic brain injury. Controlling inflammatory responses is considered a promising therapeutic approach for stroke. Recombinant human fibroblast growth factor 21 (rhFGF21) presents anti-inflammatory properties by modulating microglia and macrophages; however, our knowledge of the inflammatory modulation of rhFGF21 in focal cerebral ischemia is lacking. Therefore, we investigated whether rhFGF21 improves ischemic outcomes in experimental stroke by targeting microglia and macrophages.

Methods: C57BL/6 mice were subjected to middle cerebral artery occlusion (MCAO) and randomly divided into groups that received intraperitoneal rhFGF21 or vehicle daily starting at 6 h after reperfusion. Behavior assessments were monitored for 14 days after MCAO, and the gene expression levels of inflammatory cytokines were analyzed via qRT-PCR. The phenotypic variation of microglia/macrophages and the presence of infiltrated immune cells were examined by flow cytometry and immunostaining. Additionally, magnetic cell sorting (MACS) in combination with fluorescence-activated cell sorting (FACS) was used to purify microglia and macrophages.

Results: rhFGF21 administration ameliorated neurological deficits in behavioral tests by regulating the secretion of pro-inflammatory and anti-inflammatory cytokines. rhFGF21 also attenuated the polarization of microglia/macrophages toward the M1 phenotype and the accumulation of peripheral immune cells after stroke, accompanied by a temporal evolution of the phenotype of microglia/macrophages and infiltration of peripheral immune cells. Furthermore, rhFGF21 treatment inhibited M1 polarization of microglia and pro-inflammatory cytokine expression through its actions on FGF receptor 1 (FGFR1) by suppressing nuclear factor-kappa B (NF-κB) and upregulating peroxisome proliferator-activated receptor-γ (PPAR-γ).

Conclusions: rhFGF21 treatment promoted functional recovery in experimental stroke by modulating microglia/macrophage-mediated neuroinflammation via the NF-κB and PPAR-γ signaling pathways, making it a potential anti-inflammatory agent for stroke treatment.

Keywords: Microglia/macrophage; NF-κB; Neuroinflammation; PPAR-γ; Stroke; rhFGF21.

Fig. 1

Effects of rhFGF21 on infarct volumes and neurodeficits in the MCAO model. a Representative sample of brain slices with 2% TTC staining at 3 and 14 days after MCAO, and b quantification of the significant difference in the figure. n = 8. Values are mean ± SEM by unpaired t tests. c mNSS assessed at 1, 3, 7, and 14 days after MCAO. d–g Cumulative data illustrate the indicated neurobehavioral tests, including the time to contact (d) and the time to removal in the adhesive test (e) and results from the corner-turning test (f) and rotarod test (g) from day 1 to 14 after MCAO. n = 4 in the sham group, n = 10 in the MCAO and rhFGF21-treated group. Values are mean ± SEM by two-way ANOVA

Fig. 2

Effects of rhFGF21 on inflammatory cytokines after MCAO. mRNA expression levels of IL-1β (a), IL-6 (b), COX-2 (c), TNF-α (d), MCP-1 (e), CXCL1 (f), IL-10 (g), and TGF-β (h) in the cortex around the infarcted zone were detected via qRT-PCR at 6, 24, 48, and 72 h after MCAO. Values are mean ± SEM by one-way ANOVA, n = 6 per group. p value (red) of MCAO versus sham, p value (green) of MCAO versus MCAO+rhFGF21

Fig. 3

Effects of rhFGF21 on microglial polarization. a Representative pseudocolor and histograms of flow cytometry show the gating strategy for microglia (CD11b⁺CD45^int) and CD68⁺, CD86⁺, and CD206⁺ expressing microglia in cell suspensions from ischemic hemispheres. All gates were set using FMO control samples. b Bar graph summarizing the cell counts of microglia (CD11b⁺CD45^int) and CD68⁺, CD86⁺, and CD206⁺ expressing microglia in the brain 3 days after MCAO. c–f Quantification of flow cytometry shows the number of microglia (c) and their expression of CD68 (d), CD86 (e), and CD206 (f) at 1, 3, and 7 days. g Sketch picture indicating the area of the immunofluorescence pictures obtained from the penumbra of the infarct cortex. h Immunofluorescence staining shows CD16/32 expression by microglia (Iba-1) in the peri-infarct area at 3 days after MCAO under confocal observation. i Quantification of the counts of CD16/32⁺ microglia. n = 6 in sham the group, n = 12 in the MCAO and rhFGF21-treated group. Values are mean ± SEM, p value (red) of MCAO versus sham, p value (green) of MCAO versus MCAO+rhFGF21

Fig. 4

Effects of rhFGF21 on migrated peripheral immune cell infiltration in the CNS. a Representative flow cytometry analysis shows the gating strategy for NK cells (CD45^highCD3^-NK1.1⁺), CD4⁺T cells (CD45^highCD3⁺CD4⁺), CD8⁺T cells (CD45^high CD3⁺CD8⁺), macrophages (CD11b⁺CD45^highF4/80⁺), neutrophils (CD11b⁺CD45^highLy-6G⁺), and monocyte cells (CD11b⁺CD45^highLy-6G⁻Ly-6C⁺) using FMO. b Gating strategy of CD68, CD86, and CD206 from the subsets of macrophages using FMO control. c Quantification analysis shows the cumulative data for migrated peripheral immune cells at 3 days post-stroke. d Protein levels of CD68, CD86, and CD206 expressed in macrophages. e, f Line graphs summarize the flow cytometry data showing the dynamic distribution of infiltrated immune cells defined as CD45^high (e) and macrophages (f) in the brain at 1, 3, and 7 days after stroke. g–i Temporal presence of CD68⁺ (g), CD86⁺ (h), and CD206⁺ (i) macrophages at 1, 3, and 7 days after stroke among the three groups. n = 6 in the sham group, n = 12 in the MCAO and rhFGF21-treated group. Values are mean ± SEM by one-way ANOVA. p value (red) of MCAO versus sham, p value (green) of MCAO versus MCAO+rhFGF21

Fig. 5

Effects of rhFGF21 on peripheral immune cells in the spleen at day 3 after MCAO in mice. a Representative gating strategy of neutrophils (CD11b⁺Ly6G⁺), monocytes (CD11b⁺Ly6C⁺), CD8⁺ T cells, CD4⁺ cells, NK1.1⁺ cells, and macrophages (CD11b⁺F4/80⁺) in a single-cell suspension from the spleen using FMO control samples. b Cumulative data for quantifying the percentage of the above immune cell subsets. c Gating strategy of CD68, CD86, and CD206 expressed in macrophages from the spleen 3 days after stroke. d Bar graph shows the percentage of CD68⁺, CD86⁺, and CD206⁺ cells in macrophages. n = 6 in the sham group, n = 10 in the MCAO and rhFGF21-treated group. Values are mean ± SEM by one-way ANOVA

Fig. 6

Effect of rhFGF21 on immune cells from blood at 3 days after MCAO. a Representative dot plot showing the gating strategy of immune cell subsets from the blood. b Quantification analysis shows the percentage of NK cells, CD4⁺ T cells, CD8⁺ T cells, macrophages, neutrophils, and monocytes in the blood. c Gating strategy of CD68, CD86, and CD206 in macrophages. d Summarized flow cytometry data for quantifying CD68 and CD86 and CD206 expression in macrophages from the blood. n = 6 in the sham group, n = 10 in the MCAO and rhFGF21-treated group. Values are mean ± SEM by one-way ANOVA

Fig. 7

Effects of FGF21 on inflammatory cytokines (IL-1β, TNF-α, IL-6, and TGF-β) in sorted microglia and macrophages. a Microglia (CD11b⁺CD45^int) from the brain, and macrophages (CD11b⁺CD45^highF4/80⁺) from the brain or spleen sorted by FACS coupled with MACS, producing a purity of above 90%. b–d qRT-PCR shows the gene expression of IL-1β, TNF-α, IL-6, and TGF-β in microglia (b) and macrophages from the brain (c) and macrophages from the spleen (d). n = 8 per group. Values are mean ± SEM by one-way ANOVA

Fig. 8

Effect of rhFGF21 on the inflammatory response in primary microglia in vitro. a, b Primary cultured microglia were exposed to OGD (a) or LPS (b) plus vehicle or rhFGF21 and subsequently subjected to qRT-PCR to detect the gene expression of pro-inflammatory molecules iNOS, CD86, TNF-α, IL-1β, and IL-6 and anti-inflammatory molecules CD206, Arg-1, IL-10, and IGF-1. c, d Dot plots (c) and summarized graph (d) based on the flow cytometry results show the expression of CD86 in primary microglia evoked by LPS. e Colocalization of NF-κB (green) with nuclei (blue) in Iba1 (red)-marked microglia revealed that the nuclear translocation of NF-κB was blocked by rhFGF21 but that the influence of rhFGF21 was reversed by PD173074. f Statistical analysis shows the counts of microglia in which NF-κB translocate into the nuclei. n = 4 per group. Values are mean ± SEM by one-way ANOVA

Fig. 9

Effects of rhFGF21 on PPAR-γ and NF-κB signaling in LPS-stimulated BV2 cell line. a Representative band of p-FGFR1 and FGFR1 by western blot assay based on an internal control of β-actin and quantitated in the graph (c). b Amount of NF-κB and PPAR-γ in the nucleus of the control with H3 and quantitated in the graph (d NF-κB and e PPAR-γ). n = 4 per group. Values are mean ± SEM by one-way ANOVA