F-box/LRR-repeat protein 12 reorchestrated microglia to inhibit scarring and achieve adult spinal cord injury repair

Abstract

Scarring is an insurmountable obstacle for axonal regeneration in recovery from spinal cord injury (SCI). It impedes the repair effects of therapeutic targets in cortical neurons, such as PTEN^-/- and hyper-IL-6, which cannot break through dense scar barriers to reconstruct neural circuits. However, methods for eliminating this process remain elusive. Here, we conducted a multiomics analysis of SCI and identified FBXL12 as an effective target for inhibiting scarring, further promoting spontaneous crossing of axons at the epicenter. We identified N6-Methyladenosine (m6A) modification as the predominant mRNA modification in SCI, with Fbxl12 being a major modification target. Furthermore, m6A modification specifically promoted FBXL12 synthesis in activated microglia. The overexpression of FBXL12 in microglia contributed to its homogeneous distribution and maintained a "scar-less healing" phenotype. Remarkably, FBXL12 therapy effectively reduced extracellular matrix deposition and decreased the scar area by ~70%. Importantly, axons grew through the epicenter and reached a length of more than 2.4 mm 56 days post-SCI, significantly improving motor function and reconstructing the neural circuit. Mechanistically, FBXL12 promoted cytoskeletal reorganization and migration in microglia by catalyzing the K63-linked ubiquitylation of Myosin heavy chain 14 (MYH14). Together, our results identify m6A-FBXL12-MYH14 axis as a novel cytoskeletal reorganization pathway in activated microglia and suggest FBXL12 as an effective target for a novel microglia-based approach to facilitate scarless functional recovery in SCI.

Fig. 1

m6A methylation of Fbxl12 is a candidate master process during SCI progression. a Schematic illustration of the anatomical location of the C57BL/6 mouse spinal cord crush injury model. b Normalized peak areas and modification proportions of various mRNAs in the sham and SCI groups. The normalized peak area was calculated as the number of sites detected for this type of modification, and the modification proportion was calculated as the ratio of the detected sites for one type of modification among all detected modification sites. c Immunoblotting of m6A writers and erasers in spinal cord tissue at different days postinjury as indicated. d Schematic diagram of the analysis of integrated transcriptional and epitranscriptomic profiling. e Gene cluster dendrogram of the WGCNA results. f Volcano plot of gene expression at 7 days postinjury compared with that in the sham group. g mRNA expression and m6A level of Fbxl12 at different days postinjury as indicated (one-way ANOVA, *P < 0.05, **P < 0.01). h Correlation between Fbxl12 mRNA expression and Fbxl12 m6A levels. i RT‒qPCR of Fbxl12 mRNA at different days postinjury as indicated (t test, mean ± SEM; all groups compared with the Sham group, *P < 0.05, **P < 0.01, ***P < 0.001). j Immunoblotting of FBXL12 in spinal cord tissue at different days postinjury, as indicated. The results are representative of three independent experiments

Fig. 2

m6A promotes FBXL12 synthesis in microglia. a Images of spinal cord lesions at different time points after injury; the lesions were stained with antibodies against FBXL12 (green) and IBA1 (red) or with DAPI (blue). b Quantification of IBA1- and IBA1/FBXL12-positive cells within the 500 µm range at different time points after injury. (one-way ANOVA, mean ± SEM; *, IBA1 positive, **P < 0.01, ****P < 0.0001, n = 5; #, IBA1/FBXL12 positive, ^#P < 0.05, ^####P < 0.0001, n = 5; all groups compared with the Sham group). c, d Immunoblotting of wild-type microglia treated with MBP and LPS as indicated. The graph below shows the blots normalized to β-actin (mean ± SEM, n = 3). e Immunoblotting of wild-type microglia treated with 200 ng/mL MBP for the indicated times. The graph below shows the blots normalized to β-actin (one-way ANOVA, mean ± SEM; ***P < 0.001, ****P < 0.0001, n = 3). f RT‒qPCR quantitation of Fbxl12 mRNA expression in SIM-A9 cells treated with 200 ng/mL MBP for the indicated times. mRNA expression is relative to that of the control group (mean ± SEM, n = 3). g MeRIP‒qPCR quantitation of Fbxl12 mRNA m6A levels in SIM-A9 cells treated with 200 ng/mL MBP for the indicated times. The m6A modification level is relative to that of the control group (t test, mean ± SEM; **P < 0.01, n = 3). h Immunoblotting of FBXL2 expression in microglia with METTL3, METTL14 and YTHDF1 knockdown as indicated. i Images of spinal sections at 28 dpi from different groups stained with antibodies against IBA1 and Mettl3 or with DAPI. j Images of spinal sections at 28 dpi from different groups stained with antibodies against IBA1 and Fbxl12 or with DAPI. k Quantification of IBA1 (left) and IBA1/Mettl3 (right) positive cells at 28 dpi (one-way ANOVA, mean ± SEM; ****P < 0.0001, n = 4–5). l Quantification of Fbxl12 immunoreactive intensity around the lesion site (left) and microglia (right) at 28 dpi (t test, mean ± SEM; ***P < 0.001, n = 6–8)

Fig. 3

FBXL12 regulates cytoskeletal reorganization, promotes migration and regulates the immune response of microglia. a Images of different microglia stained with crystal violet. Where indicated, the cells were wild-type (WT), Fbxl12-overexpressing (Fbxl12^WT-OE), Fbxl12-mutant (Fbxl12^Mut-OE) and Fbxl2-deficient (Fbxl12^-/-). b Quantification of migrated microglia in (a) (one-way ANOVA, ***P < 0.001, n = 3). c SEM images of different microglia. d Images of different microglia stained with F-actin and DAPI. Representative microglia are shown below the magnified window. Scale bar as indicated. e Quantification of filopodium numbers (top, n = 5) and lengths (bottom, n = 30) of microglia in (d) (one-way ANOVA, *P < 0.05, ***P < 0.001). f Chemokine secretion in microglia (Fbxl12 ^WT-OE vs WT). (t test, mean ± SEM; *P < 0.05, ***P < 0.001, n = 3). g Volcano plot of differentially expressed genes in microglia (Fbxl12^WT-OE vs WT). h Selected KEGG pathways differentially enriched in WT versus Fbxl12^WT-OE microglia. i Heatmap of differentially expressed genes associated with positive regulation of the migration pathway in microglia (Fbxl12 ^WT-OE vs WT). j Gene Ontology (GO) terms enriched in microglia (Fbxl12^WT-OE vs WT) were selected. k, l Heatmap of differentially expressed genes associated with inflammation and chemokine terms and complement and coagulation cascades (l) in microglia (Fbxl12^WT-OE vs WT)

Fig. 4

AAV-Fbxl12 delivery promotes microglial migration after SCI. a Schematic description of the intrathecal delivery of AAV-Fbxl12. b Immunofluorescence staining of spinal cord tissue for FBXL12 and IBA1 at 28 dpi. c Images of spinal cord lesions stained with antibodies against IBA1 and Fbxl12 or with DAPI (blue) and quantification of Fbxl12 immunoreactive intensity in microglia at 28 dpi (t test, mean ± SEM; ***P < 0.001, n = 3–6). d Images of the spinal cord stained with IBA1 (4 mm in length centered around the injury site). e Quantification of IBA1 immunoreactive intensity (n = 5). f Uniform manifold approximation and projection (UMAP) plot of spots from all sections visualized via the Seurat package. Through unsupervised clustering and marker gene expression, 7 cell types were annotated. The proportions of various cell types in the SCI and SCI^Fbxl12-OE groups are presented in pie charts (left). g–i Expression profiles of marker genes of neurons, microglia and fibroblasts are presented via UMAP, where each dot represents an individual spot and the color represents the expression level (dark red, high expression; dark blue, low). Spatial distribution of the expression profiles of marker genes in spinal cord sections, with colors representing expression levels (dark blue, high expression; dark blue, low expression). j UMAP plot of microglia with other cell types denoted by gray (top) and spatial distribution (bottom) of GSVA scores for positive regulation of migration. k Images of spinal sections at 28 dpi stained with antibodies against IBA1 and IFITM3 (n = 5). l Quantification of IBA1- and IBA1/IFITM3-positive cells within the 1500 µm range at different time points after injury. (t test, mean ± SEM; *, IBA1 positive, **P < 0.01, n = 5; #, IBA1/IFITM3 positive, ^#P < 0.05, n = 5). m UMAP plot of microglia with other cell types denoted by gray (up) and spatial distribution (down) of GSVA scores for scar-free wound healing promotion

Fig. 5

Intrathecal delivery of FBXL12 to microglia reduces scarring, ameliorates pathology and improves axon regeneration in mice. a, b Images of spinal sections at 28 and 56 dpi stained with antibodies against the indicated proteins. Quantification of the indicated immunoreactive area in the lesion site below the images (dashed area in images) (t test, mean ± SEM; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, n = 5–8). c Images of anti-5-HT-stained spinal sections from different groups of mice at 28 dpi and 56 dpi, showing serotonergic axons. Yellow stars indicate the lesion site. d Quantification of the density of serotonergic axons (normalized to the density proximal to the lesion site) in the spinal cord distal to the lesion site at 56 dpi (two-way ANOVA, mean ± SEM; *P < 0.05, ****P < 0.0001, n = 5–7). e Orthogonal projection image of the spinal cord with tissue clearing and 3D imaging. f Images of spinal sections at 28 dpi stained with antibodies against CD206, iNOS, S100A10, NF200, NESTIN and SOX2. g Amplitude and latency period of MEPs at 56 days postinjury (t test, mean ± SEM; *P < 0.05, n = 3). h Representative images of hindlimb movement in mice at 56 dpi with or without AAV-FBXL12 overexpression. i BMSs of mice with or without AAV-FBXL12 overexpression (two-way ANOVA, mean ± SEM; *P < 0.05, **P < 0.01, n = 10)

Fig. 6

FBXL12 mediates MYH14 K63 ubiquitination to orchestrate the cytoskeletal reorganization of microglia. a Schematic description of immunoprecipitation‒mass spectrometry (IP‒MS) of microglia treated with 200 ng/mL MBP for 4 days. b, c GO-molecular function and cellular component analyses of whole FBXL12-binding proteins. d KEGG pathway analysis of MBP-treated microglia. e Volcano plot of FBP changes in wild-type microglia and MBP-treated microglia. f Immunoblotting of MYH14 and MYH9 in different microglia. g Immunoblotting of MYH14 in WT and Fbxl12^−/− microglia treated with 200 ng/mL MBP for the indicated times. h Immunoprecipitation of FBXL12 with MYH14 in WT and Fbxl12^−/− microglia. i Images of different microglia stained with F-actin and antibodies against MYH14. Representative microglia are shown in the middle and below magnified windows. Scale bar as indicated. j Crystal violet staining of WT and Fbxl12^WT-OE microglia. k Quantification of migrated microglia in j (one-way ANOVA, mean ± SEM; ***P < 0.001, ****P < 0.0001, n = 3). l Images of WT and Fbxl12^WT-OE microglia stained with F-actin, with or without MYH14. m Quantification of the number (left, n = 6) and length (right, n = 16) of filopodium in l (one-way ANOVA, mean ± SEM; ****P < 0.0001)