Nuclear translocation of PKM2 in astrocytes of EAE mice.

(A) Immunofluorescence staining of PKM2 with GFAP (astrocyte marker) in spinal cord of control mice (n=4) and MOG35–55-induced EAE mice. Disease onset (dpi 7–17, n=3), peak (dpi 14–24, n=4) and chronic (dpi 21–26, n=2) were defined dependent on the EAE course. Scale bar: 20 μm. While arrows indicated nuclear PKM2. (B) Immunofluorescence staining of PKM2 (green) with GFAP (red) in non-treated primary astrocytes (control) or primary astrocytes cultured with splenocytes supernatants of MOG35–55-induced EAE mice (MOGsup) for different time points (6 h, 12 h and 24 h). DAPI (blue) was used as a nuclear staining. Scale bar: 100 μm. (C) Mean fluorescence intensity of PKM2 in different groups of (B) was calculated by ImageJ. MFI, mean fluorescence intensity. (D) Nuclear PKM2 ratio in different groups of (B) were calculated. Fight fields of views per group were included in the analysis. The number of nuclear PKM2 was quantified by Image-Pro Plus software manually (eg. nuclear or cytoplasmic based on DAPI blue staining). The proportion of nuclear PKM2 is determined by normalizing the count of nuclear PKM2 to the count of nuclear DAPI, which represents the number of cell nuclei. Data are represented as mean ± SEM, one-way ANOVA. ***P<0.001. SEM, standard error of the mean.

Prevention of PKM2 nuclear transport reduced the glycolysis and proliferation of primary astrocytes.

(A) Verification of DASA-58 effect on the inhibition of PKM2 nuclear transport by immunofluorescence. Primary astrocytes were pretreated with 50 μM DASA-58 for 30 min and stimulated with MOGsup for 12h. Scale bar: 50 μm. Scale bar in enlarged image: 20 μm. (B) Nuclear ratio of PKM2 in each group was calculated. Five fields of views per group were included in the analysis. (C) Glycolysis level of astrocytes in each group was assessed by lactate production (N=5) and glucose consumption (N=4) assays. (D) Effect of DASA-58 on protein levels of glycolytic enzymes p-c-myc, LDHA and PKM2 were measured by western blotting. Right panel shows the quantification of phospho-c-myc expression normalized to total c-myc levels. (E) Proliferation of astrocytes were measured by CCK8. N=5. (F) Proliferation of astrocytes were measured by EdU assays. (G) EdU positive cells in each group was calculated from ten fields of views per group. Scale bar: 100 μm. The blot is representative of three independent experiments. Data are represented as mean ± SEM, one-way ANOVA. *P<0.05; **P<0.01; ***P<0.001.

PKM2 interacted with STAT3 and NF-κB and promoted their activation in astrocytes.

(A-B) Immunofluorescence staining of phospho-STAT3 (A) or phospho-p65 (B) with GFAP in astrocytes. Primary astrocytes were pretreated with 50 μM DASA-58 for 30 min and stimulated with MOGsup for 12h. Scale bar: 100 μm. Scale bar in enlarged image: 20 μm. (C) Western blotting analysis showed that DASA-58 inhibited the activation of NF-κB and STAT3 induced by MOGsup stimulation. (D) Nuclear-cytoplasmic protein extraction analysis showed cytoplasmic and nuclear protein levels STAT3 and p50/p65 upon DASA-58 treatment. (E) Immunoprecipitation demonstrated the interaction between PKM2 and STAT3, c-myc and p50/p65 subunits of NF-κB in primary astrocyte. Data are represented as mean ± SEM, one-way ANOVA. **P<0.01; ***P<0.001. SEM, standard error of the mean.

Identification of interaction between E3 ligase TRIM21 and PKM2 in astrocytes.

(A) Mass spectrometry (MS) showed the list of metabolic-related proteins that potentially interact with PKM2 in unstimulated (Ctl) and MOGsup-stimulated primary astrocytes. TRIM21 was identified to interact with PKM2. (B-D) Biological process of GO term (B), KEGG pathway (C) and Wikipathway (D) analysis of proteins identified by MS. (E) Interaction between PKM2 and TRIM21 was predicted with molecular docking and showed by PyMol. The hydrogen bonds were formed between Phe23, Thr87 of TRIM21 and Pro477, Lys 263 of PKM2. (F) Immunoprecipitation showed the interaction between endogenous PKM2 and TRIM21 in primary astrocyte. (G-H) Primary astrocytes were transfected with Myc-tagged TRIM21 and Flag-tagged PKM2, immunoprecipitation with anti-Flag (G) or anti-Myc (H) showed the exogenous binding between PKM2 and TRIM21 in astrocytes. (I) Full-length TRIM21 and a series of TRIM21 mutants with deletion (Δ) of various domains (top panel). 293[T cells were co-transfected with Flag-PKM2 and WT Myc-TRIM21 or their truncation mutants for 48 h. Immunoprecipitation was performed. (J) Full-length PKM2 and a series of PKM2 mutants with deletion (Δ) of various domains (top panel). 293[T cells were co-transfected with Myc-TRIM21 and WT Flag-PKM2 or their truncation mutants for 48 h. Immunoprecipitation was performed.

TRIM21 expression is upregulated in astrocytes of EAE mice.

(A-D) Single-cell RNA-seq profiles from naive and EAE mice (peak and chronic phase) CNS tissues. Naive (n=2); peak (dpi 14–24, n=3); chronic (dpi 21–26, n=2). (A) Violin plots displaying the expression of TRIM21 across the cell types identified. (B) Violin plots displaying the expression of TRIM21 in different phases of EAE and naive mice across the cell types identified. Expression of TRIM21 was shown to be elevated in EAE mice (peak and chronic) compared with naive mice. (C) UMAP representation of 12 clusters generated from sub-clustering of astrocytes. (D) Violin plots displaying the expression of TRIM21 at peak, chronic phases from EAE and naive mice in subclusters of astrocytes. (E) Analysis of TRIM21 mRNA expression in astrocytes from spinal cord during three stages (onset, peak, and chronic) of EAE and naive mice from GEO dataset GSE136358 (one-way ANOVA, *P<0.05; **P<0.01). (F) Primary astrocytes were treated with or without MOGsup for different time points. Analysis of TRIM21 expression by qPCR. (G) Western blotting analysis of TRIM21 protein expression in non-treated or MOGsup-treated astrocytes. The right panel shows the quantification of TRIM21 expression normalized to β-actin loading control (Paired t test, *P<0.05). (H) Immunofluorescence staining showed the upregulated expression of TRIM21 in astrocytes (marker: GFAP) of EAE mice (Unpaired t test, *P<0.05). Scale bar: 20 μm. Data are represented as mean ± SEM.

TRIM21-induced nuclear transport of PKM2 promoted glycolysis and proliferation of astrocytes.

(A) Overexpression of TRIM21 promoted nuclear translocation of PKM2. The right panel shows the quantification of nuclear PKM2 expression level normalized to nuclear lamin level (Paired t test, *P<0.05). (B) TRIM21 was silenced in primary astrocytes using two independent short hairpin RNAs. Nuclear-cytoplasmic fraction analysis showed that knockdown of TRIM21 decreased nuclear ratio of PKM2. The right panel shows the quantification of nuclear PKM2 expression level normalized to nuclear lamin level (one-way ANOVA, *P<0.05). (C) Immunoprecipitation showed that TRIM21 promoted the K63-linked ubiquitination of PKM2. (D) Western blotting analysis of STAT3 and NF-κB activation in control or TRIM21-overexpressed astrocytes. (E) Immunoprecipitation showed that TRIM21 promoted the interaction between PKM2 and its interacting proteins c-myc, STAT3 and p50. (F) Prevention of PKM2 nuclear import with DASA-58 (50 μM) reduced the nuclear retention of NF-κB subunits and STAT3 in TRIM21-overexpressed astrocytes. (G) EdU analysis of cell proliferation in TRIM21-overexpressed, DASA-58 treated TRIM21-overexpressed cells and control astrocytes (one-way ANOVA, **P<0.01). Scale bar: 100 μm. (H) Glycolysis of astrocytes were measured in TRIM21-overexpressed, DASA-58 treated TRIM21-OE cells and control astrocytes (one-way ANOVA, *P<0.05; ***P<0.001). EV: empty vector. Data are represented as mean ± SEM.

Intracerebroventricular injection of shTRIM21 ameliorates disease severity of Experimental Autoimmune Encephalomyelitis (EAE).

C57BL/6 mice were injected i.c.v with 1x107 IU shTRIM21 or control lentivirus (shVec) 15 days p.i. (onset). Mice were sacrificed at day 22 p.i. and spinal cords were harvested. (A) Disease was scored daily on a 0 to 5 scale. N=5 to 6 mice in each group. (B) Spinal cord sections were stained for markers of inflammation by hematoxylin and eosin (H&E) and demyelination by Luxol fast blue (LFB), respectively. (C) Scoring of inflammation (H&E) and demyelination (LFB) on a 0-3 scale. (D) TRIM21 expression in spinal cord of mice from shVec and shTRIM21 group was measured by immunofluorescence. (E) Demyelination in each group was assessed by MBP staining. MBP intensity was measured in the white matter of the spinal cord using Image-Pro. (F-G) Immunostaining of GFAP (F) and IBA1 (G) on spinal cord sections of shVec and shTRIM21-treated EAE mice. White matter and gray matter are shown as representative images. Quantification of GFAP positive cells/mm2, IBA1 positive cells/mm2 in both the white matter and gray matter. The measured areas included 3 to 5 fields per group. i.c.v., intracerebroventricular; p.i., postimmunization. Scale bar: 50 μm. Data are represented as mean ± SEM. *P<0.05; **P<0.01; ***P<0.001, as determined by two-way ANOVA analysis (A) or unpaired Student’s t test (C, F-G).

Schematic proposal of nuclear translocation of PKM2 in astrocytes of EAE.

In astrocyte of EAE mice, TRIM21 expression is upregulated. E3 ubiquitin ligase TRIM21 ubiquitylates PKM2 and promotes its nuclear translocation, nuclear PKM2 activated STAT3 and NF-κB pathways and interact with c-Myc to enhance glycolysis and proliferation in astrocytes. Thus, TRIM21-PKM2 pathway exerts a potential role in activating astrocytes and inducing EAE development.