Inhibition of EED-mediated histone methylation alleviates neuroinflammation by suppressing WNT-mediated dendritic cell migration

Abstract

The epigenetic modification of histone H3 lysine 27 trimethylation (H3K27me3) by the embryonic ectoderm development (EED) protein is closely associated with the regulation of transcriptional programs and is implicated in autoimmune diseases. However, the efficacy of targeting H3K27me3 for the treatment of neuroinflammation remains unclear. In this study, we demonstrate that systemic administration of an EED inhibitor diminishes the inflammatory response mediated by dendritic cells (DCs), thereby alleviating experimental autoimmune encephalitis (EAE), a representative mouse model of autoimmune diseases in the central nervous system (CNS). Our findings indicate that EED inhibitors suppress DC migration by upregulating genes in the WNT signaling pathway that are epigenetically marked by H3K27me3. Conversely, inhibiting the WNT pathway partially reverses the impaired DC migration caused by EED inhibitors. Additionally, the genetic deletion of Eed inhibits DC migration and effectively mitigates autoimmune symptoms and inflammatory infiltration into the CNS in EAE. These results highlight EED as a critical regulator of DC migration and suggest its potential as a therapeutic target for autoimmune disorders.

Keywords: Autoimmune diseases; Dendritic cells; H3K27me3; Migration; Neuroinflammation; WNT.

Fig. 1

EED226 ameliorates symptoms and inflammation in EAE mice. A Scheme for the induction, assessment and drug administration of EAE mice. B Clinical scores of vehicle or EED226 treated EAE mice. n = 10 animals, 5 animals are sacrificed for the histopathological examination and flow cytometry analysis at day 22 post immunization (dpi 22). * indicates a statistically difference when compared with EAE + vehicle group. C Representative images of H&E staining (top) and LFB staining (bottom) for spinal cords from vehicle or EED226 treated EAE mice at dpi 22. Scale bar: 50 μm. n = 3. D Bar graph for LFB staining analysis of the demyelinated area in the spinal cords of vehicle or EED226 treated EAE mice at dpi 22. n = 3. E Representative images (left) and bar graph (right) for electron microscopy analysis of the demyelination in the spinal cords of vehicle or EED226 treated EAE mice at dpi 22. Scale bar: 2 μm. n = 3. F Representative images (left) and bar graph (right) for flow cytometry analysis of the frequency of naïve T cells (CD4⁺CD62L^hiCD44^lo) and central memory T cells (CD4⁺CD62L^loCD44^hi) in the spleen of vehicle or EED226 treated EAE mice at dpi 22. n = 3. G Representative images (left) and bar graph (right) for flow cytometry analysis of the frequency of Th1 (CD4⁺IFN-γ⁺) and Th17 (CD4⁺IL-17⁺) in the spleen of vehicle or EED226 treated EAE mice at dpi 22. n = 3. H Representative images (left) and bar graph (right) for flow cytometry analysis of the frequency of total DCs (CD11c⁺) in the spleen of vehicle or EED226 treated EAE mice at dpi 22. n = 3. I Representative images (left) and bar graph (right) for flow cytometry analysis of the frequency of mature DCs (CD11c⁺CD80⁺, CD11c⁺MHC-II⁺) in the spleen of vehicle or EED226 treated EAE mice at dpi 22. n = 3. J Bar graph for flow cytometry analysis of the percentage of CD80⁺ or MHC-II⁺ in CD11c⁺ DCs in the spleen of vehicle or EED226 treated EAE mice at dpi 22. n = 3. Data are presented as mean ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, not significant

Fig. 2

EED inhibitors impair the migration of DC2.4 cells. A SRB analysis for the survival rate of DC2.4 upon treatment with different EED inhibitors at various concentrations (0, 0.32, 0.63, 1.25, 2.5, 5, 10, 20 or 40 μM) for 24 h. n = 3. * indicates a statistically difference when compared with 0 μM. B Western blotting analysis (left) and corresponding quantification (right) for the expression of H3K27me3 and EED in DC2.4 after treatment with A395 (5 μM), MAK683 (5 μM), EEDi-5285 (5 μM) or EED226 (10 μM) for 24 h. n = 3. C Scheme for transwell analysis of chemokines-triggered migration of DC2.4 after treating with LPS or different EED inhibitors. D Representative images (left) and statistics (right) for transwell analysis together with crystal violet staining in DC2.4 after treating with A395 (5 μM), MAK683 (5 μM), EEDi-5285 (5 μM) or EED226 (10 μM) for 24 h and LPS (50 ng/mL) for 6 h. Scale bar: 200 μm. n = 3. E Immunostaining of F-Actin (green) in DC2.4 after treating with A395 (5 μM), MAK683 (5 μM), EEDi-5285 (5 μM) or EED226 (10 μM) for 24 h and LPS (50 ng/mL) for 6 h. Scale bar: 10 μm. n = 3. F Western blotting analysis (left) and the corresponding quantification (right) for the expression of H3K27me3 in DC2.4 after transfection with HA-EED plasmid and treatment with A395 (5 μM), MAK683 (5 μM), EEDi-5285 (5 μM) or EED226 (10 μM) for 24 h and LPS (50 ng/mL) for 6 h. n = 3. G Scheme for transwell analysis of chemokines-triggered migration of DC2.4 after transfection with HA-EED plasmid and treatment with LPS and EED inhibitors. H Representative images (left) and statistics (right) for transwell analysis together with crystal violet staining in DC2.4 cells after transfection with HA-EED plasmid and treatment with A395 (5 μM), MAK683 (5 μM), EEDi-5285 (5 μM) or EED226 (10 μM) for 24 h and LPS (50 ng/mL) for 6 h. Scale bar: 200 μm. n = 3. Data are presented as mean ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, not significant

Fig. 3

EED inhibitors mediate the transcriptional programs for DC migration. A PCA of the transcriptome profiles of DC2.4 after treating with EEDi-5285 (5 μM) for 24 h and LPS (50 ng/mL) for 6 h. n = 3 independent experiments. B Volcano plot of the transcriptome profiles from vehicle and EEDi-5285 group (fold-change > 2, P < 0.05). C Heatmap analyses of the differentially expressed genes vehicle and EEDi-5285 group. D GSEA analyses of genes enriched in vehicle or EEDi-5285 group. E The relative counts of indicated genes in DC2.4 from vehicle and EEDi-5285 group. F–H GSEA enrichment scores for indicated gene sets in vehicle or EEDi-5285 group. I The relative counts of indicated genes in DC2.4 from vehicle and EEDi-5285 group. Data are presented as mean ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001

Fig. 4

EED inhibitors block DC migration via promoting WNT signaling. A ChIP-seq analyses of H3K27me3 enrichment around TSS and TES regions in human DCs from GEO dataset GSE209566. Genes shown in rows were sorted in decreasing order by signal intensity. B Venn diagram of the overlapped genes between the H3K27me3-bound genes in DCs and differentially up-regulated genes in EEDi-5285 group. C KEGG analyses of the signaling pathways that enriched in the 431 overlapped genes from (B). D Heatmap analyses of the representative genes involved in different pathways of (C) in vehicle and EEDi-5285 groups. E Bland–Altman plot showing H3K27me3 ChIP-seq peaks in DCs with upregulated genes in EEDi-5285 group. F Representative H3K27me3 ChIP-seq peak tracks of indicated genes in DCs. G GSEA enrichment scores for WNT signaling in EEDi-5285group. H qRT-PCR analyses of the mRNA levels of WNT-related genes after treatment with A395 (5 μM), MAK683 (5 μM), EEDi-5285 (5 μM) or EED226 (10 μM) for 24 h and LPS (50 ng/mL) for 6 h. n = 3. I Representative images (left) and statistics (right) for transwell analysis together with crystal violet staining in DC2.4 after treating with Wnt-C59 (5 μM), A395 (5 μM), MAK683 (5 μM), EEDi-5285 (5 μM) or EED226 (10 μM) for 24 h and LPS (50 ng/mL) for 6 h. Scale bar: 200 μm. n = 3. Data are presented as mean ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, not significant

Fig. 5

EED deficiency impedes DC activation. A Diagram depicting generation of Eed cKO mice. B Western blotting analysis (left) and its corresponding quantification (right) for the expression of EED and H3K27me3 in WT or Eed cKO BMDCs. n = 3. C, D Representative images and bar graph for flow cytometry analysis of the percentage and number of CD11c⁺ cells in WT or Eed cKO BMDCs after treating with LPS (50 ng/mL) for 6 h. n = 3. E, F Representative images (E) and bar graph (F) for flow cytometry analysis of the percentages and numbers of CD80⁺, CD86⁺ or MHC-II⁺ cells in CD11c⁺ DCs of WT or Eed cKO BMDCs after treatment of LPS (50 ng/mL) for 6 h. n = 3. G Representative images (left) and quantification (right) for transwell analysis of chemokines-triggered migration of WT or Eed cKO BMDCs after treating with EEDi-5285 (5 μM) or EED226 (10 μM) for 24 h and LPS (50 ng/mL) for 6 h. n = 5. H Representative images (left) and bar graph (right) for flow cytometry analysis of the frequency of Th1 (CD4⁺IFN-γ⁺) and Th17 (CD4⁺IL-17⁺) cells after co-culture with MOG_35-55 (30 μg/mL, 24 h) and LPS (50 ng/mL, 6 h) stimulated WT or Eed cKO BMDCs. n = 3. I qRT-PCR analyses of the indicated WNT genes in WT or Eed cKO BMDCs after treatment with LPS (50 ng/mL) for 6 h. n = 3. J CUT&Tag-qPCR analyses of the enrichment of H3K27me3 in the promoters of WNT genes in WT or Eed cKO BMDCs after treatment with LPS (50 ng/mL) for 6 h. n = 3. K Quantification for transwell analysis of chemokines-triggered migration of WT or Eed cKO BMDCs after treating with Wnt-C59 (5 μM) for 24 h and LPS (50 ng/mL) for 6 h. n = 6. Data are presented as mean ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, not significant

Fig. 6

Eed deficiency in DC suppresses EAE progression. A Clinical score in WT and Eed cKO EAE mice. n = 14 animals, 9 animals are sacrificed for the histopathological examination and flow cytometry analysis at day 23 post immunization (dpi 23). * indicates a statistically difference when compared with WT group. B Representative images (left) and bar graph (right) of LFB staining for spinal cords from WT and Eed cKO EAE mice at dpi 23. Scale bar: 50 μm. n = 3. C Representative images (left) and bar graph (right) for electron microscopy analysis of the demyelination in the spinal cords from WT and Eed cKO EAE mice at dpi 23. Scale bar: 2 μm. n = 3. D–F Representative images (D) and bar graph (E, F) for flow cytometry analysis of the percentages and numbers of Th1 (CD4⁺IFN-γ⁺) and Th17 (CD4⁺IL-17⁺) in the brains from WT and Eed cKO EAE mice at dpi 23. n = 3. G Immunostaining of CD4 (green) and IL-4 (red) in the spinal cords from WT and Eed cKO EAE mice at dpi 23. Scale bar: 50 μm. n = 3. H Immunostaining of CD4 (red) and Foxp3 (green) in the spinal cords from WT and Eed cKO EAE mice at dpi 23. Scale bar: 50 μm. n = 3. I, J Representative images (I) and bar graph (J) for flow cytometry analysis of the frequency and numbers of CD11c⁺MHC-II⁺ DCs in the brains from WT and Eed cKO EAE mice at dpi 23. n = 3. K qRT-PCR analyses for the mRNA levels of pro-inflammatory cytokines Ifn-γ and anti-inflammatory cytokine Il-4 in the brains from WT and Eed cKO EAE mice at dpi 23. n = 3. Data are presented as mean ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001

Fig. 7

Eed loss alleviates periphery inflammation in EAE mice. A–E Representative images (A) and bar graph (B–E) for flow cytometry analysis of the percentages and numbers of Th1 (CD4⁺IFN-γ⁺), Th17 (CD4⁺IL-17⁺), Th2 (CD4⁺IL-4⁺) and Treg (CD4⁺Foxp3⁺) in the draining lymph nodes from WT and Eed cKO EAE mice at dpi 23. n = 3. F–H Representative images (F) and bar graph (G, H) for flow cytometry analysis of the percentages and numbers of total DCs (CD11c⁺) and mature DCs (CD11c⁺CD80⁺) in the spleen from WT and Eed cKO EAE mice at dpi 23. n = 3. I Bar graph for flow cytometry analysis of the percentage of CD80⁺ in CD11c⁺ DCs in the spleen from WT and Eed cKO EAE mice at dpi 23. n = 3. Data are presented as mean ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, not significant

Fig. 8

EED inhibitors hinder the chemotaxis of human moDCs. A Scheme of human moDCs culture and inflammation-associated phenotype detection. B The morphological changes from monocytes (Day0) to immature moDCs (Day 6) and eventually LPS-challenged mature moDCs (Day7). Scale bar: 20 μm. n = 3. C Representative images (left) and bar graph (right) for flow cytometry analysis of the frequency of CD209⁺ moDCs after cell culture for 7 days. n = 3. D Western blotting analysis (left) and its corresponding quantification (right) for the expression of EED and H3K27me3 in human moDCs treated with A395 (5 μM), MAK683 (5 μM), EEDi-5285 (5 μM), EED226 (10 μM) and LPS (300 ng/mL) for 24 h. n = 3. E qRT-PCR analyses of the mRNA levels of WNT-related genes in human moDCs treated with EEDi-5285 (5 μM), EED226 (10 μM) and LPS (300 ng/mL) for 24 h. n = 3. F, G Representative images (E) and statistics (F) for transwell analysis in human moDCs after treating with A395 (5 μM), MAK683 (5 μM), EEDi-5285 (5 μM), EED226 (10 μM) and LPS (300 ng/mL) for 24 h. Scale bar: 200 μm. n = 3. H, I Representative images (H) and bar graph (I) for flow cytometry analysis of the expression of CD83 and HLA-DR (MHCII) in CD209⁺ moDCs after treatment with A395 (5 μM), MAK683 (5 μM), EEDi-5285 (5 μM), EED226 (10 μM) and LPS (300 ng/mL) for 24 h. n = 3. J Elisa of TNF-α and IL-10 levels in the supernatant of human moDCs after treatment with A395 (5 μM), MAK683 (5 μM), EEDi-5285 (5 μM), EED226 (10 μM) and LPS (300 ng/mL) for 24 h. n = 3. Data are presented as mean ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001