EAAT3 impedes oligodendrocyte remyelination in chronic cerebral hypoperfusion-induced white matter injury

Abstract

Background: Chronic cerebral hypoperfusion-induced demyelination causes progressive white matter injury, although the pathogenic pathways are unknown.

Methods: The Single Cell Portal and PanglaoDB databases were used to analyze single-cell RNA sequencing experiments to determine the pattern of EAAT3 expression in CNS cells. Immunofluorescence (IF) was used to detect EAAT3 expression in oligodendrocytes and oligodendrocyte progenitor cells (OPCs). EAAT3 levels in mouse brains were measured using a western blot at various phases of development, as well as in traumatic brain injury (TBI) and intracerebral hemorrhage (ICH) mouse models. The mouse bilateral carotid artery stenosis (BCAS) model was used to create white matter injury. IF, Luxol Fast Blue staining, and electron microscopy were used to investigate the effect of remyelination. 5-Ethynyl-2-Deoxy Uridine staining, transwell chamber assays, and IF were used to examine the effects of OPCs' proliferation, migration, and differentiation in vivo and in vitro. The novel object recognition test, the Y-maze test, the rotarod test, and the grid walking test were used to examine the impact of behavioral modifications.

Results: A considerable amount of EAAT3 was expressed in OPCs and mature oligodendrocytes, according to single-cell RNA sequencing data. During multiple critical phases of mouse brain development, there were no substantial changes in EAAT3 levels in the hippocampus, cerebral cortex, or white matter. Furthermore, neither the TBI nor ICH models significantly affected the levels of EAAT3 in the aforementioned brain areas. The chronic white matter injury caused by BCAS, on the other hand, resulted in a strikingly high level of EAAT3 expression in the oligodendroglia and white matter. Correspondingly, blocking EAAT3 assisted in the recovery of cognitive and motor impairment as well as the restoration of cerebral blood flow following BCAS. Furthermore, EAAT3 suppression was connected to improved OPCs' survival and proliferation in vivo as well as faster OPCs' proliferation, migration, and differentiation in vitro. Furthermore, this study revealed that the mTOR pathway is implicated in EAAT3-mediated remyelination.

Conclusions: Our findings provide the first evidence that abnormally high levels of oligodendroglial EAAT3 in chronic cerebral hypoperfusion impair OPCs' pro-remyelination actions, hence impeding white matter repair and functional recovery. EAAT3 inhibitors could be useful in the treatment of ischemia demyelination.

Keywords: chronic cerebral hypoperfusion; excitatory amino acid transporter 3 (EAAT3); neuroprotection; oligodendrocyte progenitor cells (OPCs); white matter.

FIGURE 1

High expression of EAAT3 after white matter injury. UMAP plot showing clusters and cell type annotations in mouse cortex (A) and thalamus (B) in the Single Cell Portal database. (C) the EAAT3 expression in annotated cell clusters in the PanglaoDB database. (D) Representative immunofluorescence image and enlarged confocal image of A2B5 (red) and EAAT3 (green) in primary OPCs, as well as CNPase (red) and EAAT3 (green) in primary oligodendrocytes. (E–H) Representative immunoblotting images and quantification of relative protein level of EAAT3 in the hippocampus, cerebral cortex, and white matter at weeks 2, 4, 6, and 8 of mouse brain development, n = 3 in each group. (I–K) In a mouse model of traumatic brain injury and intracerebral hemorrhage, representative immunoblotting pictures and quantification of the relative protein level of EAAT3 in the hippocampus, cerebral cortex, and white matter were performed on Day 1, n = 3 in each group. (L–O) Representative western blot images and quantification of EAAT3 expression in the white matter, hippocampus, and cerebral cortex at different days after BCAS mice, n = 3 in each group. There was no difference in body weight between mice in each group. The data for each group conformed to a normal distribution. The p value was determined by ANOVA with Bonferroni's post‐hoc test. *p < 0.05, **p < 0.01, between the indicated groups, ns indicates non‐significance. Data are represented as means ± SEM.

FIGURE 2

White matter damage and demyelination in BCAS mice with chronic cerebral hypoperfusion can be rescued by injection of PDC. (A, B) Representative immunoblotting and quantification showed that the expression of EAAT3 decreased after injection of the inhibitor. The GAPDH protein served as a control. (C, D) Representative immunoblotting and quantification showed that the expression of CNPase can increase after injection of PDC. The GAPDH protein served as a control. (E) Representative images of CNPase (green) in CC of each group. (F) Results of quantitative analysis of the CNPase‐positive cells in each field of CC. n = 3 in each group, 2 or 3 images per animal. (G, H) Representative LFB‐stained images and quantification illustrating preserved white matter integrity due to decreased myelin rarefaction and white matter lesion formation in the corpus callosum (medial) on 14 days in PDC mice compared with the BCAS group. n = 3 in each group, 2 or 3 images per animal. (I–K) Representative electron microscopy images and quantification of axon diameter and G‐ratio on 14 days in indicated groups. (L) G‐ratio plotted as a function of axon diameter, n = 3 in each group. There was no difference in body weight between mice in each group. The data for H dose not exhibit a normal distribution. p Value was determined by ANOVA with the Kruskal–Wallis test. In addition, the data conformed to a normal distribution. p Value was determined by ANOVA with Bonferroni's post‐hoc test. *p < 0.05, **p < 0.01, between the indicated groups, ns indicates non‐significance, n = 3 in each group. Data are represented as means ± SEM.

FIGURE 3

PDC administration increases the survival, proliferation, and migration of OPCs. (A, B) Representative immunoblotting and quantification showed the expression of NG2 in white matter on Day 14 after BCAS and the expression can increase after injection of PDC. (C) Representative images of NG2 expression in CC in each group. (D) Results of quantitative analysis of the NG2‐positive cells in each field of CC in each group, n = 3, 2 or 3 images per animal. (E) Representative images of NG2 expression in SVZ in each group. (F) Results of immunofluorescent intensity of the NG2‐positive cells in each field of SVZ in each group, n = 3, 2 or 3 images per animal. (G) Representative immunofluorescence co‐staining images of EdU (red) and NG2 (green) in SVZ in each group. (H) Quantitative analysis of % EdU⁺/ NG2⁺ cells in total DAPI of SVZ, n = 3, 2 or 3 images per animal. (I) Representative immunofluorescence images of EdU immunoreactive primary OPCs in the indicated groups. (J) % EdU‐positive OPCs in total DAPI in the indicated groups, n = 3–5 images from at least three independent experiments. (K) Representative photomicrographs of migrated OPCs. (L) Quantitative analysis of the numbers of migrated OPCs, n = 3–5 images from at least three independent experiments. There was no difference in body weight between mice in each group. The data for each group conformed to a normal distribution. p Value was determined by ANOVA with Bonferroni's post‐hoc test. The data of H were analyzed using the Mann–Whitney U test. *p < 0.05, **p < 0.01, between the indicated groups, ns indicates non‐significance. Data are represented as means ± SEM.

FIGURE 4

PDC administration increases the differentiation of OPCs. (A) Representative immunofluorescence co‐staining images of MBP (red) and Olig2 (green) in primary oligodendrocytes. (B) Quantitative analysis of the number of MBP⁺/ Olig2⁺ cells in each field of the indicated groups. n = 3–5 images from at least three independent experiments. (C) The process extensions for primary oligodendrocytes were analyzed using Sholl analysis. By counting the intersections that processes make with concentric circles that are numbered 1–3 to represent increasing distance from the cell body, branching was measured, n = 3–5 images from at least three independent experiments. (D, E) Representative immunoblotting and quantification of MBP in each group, n = 3 mice. (F–I) Representative immunofluorescence images and quantification of MBP immunofluorescent intensity in media and paramedian of CC in each field of the indicated groups, n = 3, 3–5 images per animal. There was no difference in body weight between mice in each group. The data for each group conformed to a normal distribution. p Value was determined by ANOVA with Bonferroni's post‐hoc test. *p < 0.05, **P < 0.01, between the indicated groups, and ns indicates non‐significance. Data are represented as means ± SEM.

FIGURE 5

EAAT3 knockdown motivates OPCs differentiation for remyelination. (A) Representative immunofluorescence images of EAAT3 (red) after transfection. (B) Mean fluorescence intensity quantitative analysis of EAAT3 in each field of the indicated groups. n = 3–5 images from at least three independent experiments. (C) Representative images and quantitative analysis (D) of the percentage of SYTOX‐positive cells in all cells under each field, n = 3–5 images from at least three independent experiments. (E) Representative immunofluorescence co‐staining images of MBP (red) and Olig2 (green) in oligodendrocytes. (F) Quantitative analysis of the number of MBP⁺/Olig2⁺ cells in each field of the indicated groups. n = 3–5 images from at least three independent experiments. (G) The process extensions for primary oligodendrocytes were analyzed using Sholl analysis. By counting the intersections that processes make with concentric circles that are numbered 1–3 to represent increasing distance from the cell body, branching was measured. n = 3–5 images from at least three independent experiments. The data for each group conformed to a normal distribution. p Value was determined by ANOVA with Bonferroni's post‐hoc test. *p < 0.05, **p < 0.01, between the indicated groups, and ns indicates non‐significance. Data are represented as means ± SEM.

FIGURE 6

PDC administration activates the mTOR signaling pathway in vitro. (A) Representative immunofluorescence co‐staining images of Olig2 (red) and P‐mTOR (green) in primary OPCs. (B) Mean fluorescence intensity quantitative analysis of P‐mTOR in each field of the indicated groups. n = 3–5 images from at least three independent experiments. (C, D) Representative immunoblotting and quantification of P‐mTOR in indicated groups, n = 3 mice. There was no difference in body weight between mice in each group. The data for each group conformed to a normal distribution. p Value was determined by ANOVA with Bonferroni's post‐hoc test. *p < 0.05, **p < 0.01, between the indicated groups, ns indicates non‐significance. Data are represented as means ± SEM.

FIGURE 7

Inhibition of the mTOR signaling pathway can hinder the protective effect of PDC. (A, B) Representative immunoblotting and quantification of P‐mTOR and mTOR in indicated groups. The β‐Actin protein served as a control, n = 3 mice. (C, D) Representative immunoblotting and quantification of MAG in indicated groups. The β‐Actin protein served as a control, n = 3 mice. (E, G) Representative immunofluorescence images and quantification of MAG immunofluorescent intensity of CC in each field of the indicated groups. n = 3, 3–5 images per animal. (F, H) Representative LFB‐stained images and quantification illustrating preserved white matter integrity due to decreased myelin rarefaction and white matter lesion formation in CC in each group. n = 3, 3–5 images per animal. There was no difference in body weight between mice in each group. The data for H that do not exhibit a normal distribution. p Value was determined by ANOVA with the Kruskal–Wallis test. In addition, the data conformed to a normal distribution. p Value was determined by ANOVA with Bonferroni's post‐hoc test. *p < 0.05, **p < 0.01, between the indicated groups, ns indicates non‐significance. Data are represented as means ± SEM.

FIGURE 8

Inhibition of EAAT3 enhances primary OPCs survival and promotes the behavioral defects of BCAS mice. (A) Representative images and (B) quantitative analysis of the percentage of SYTOX‐positive cells in all cells under each field, n = 3–5 images from at least three independent experiments. (C) Representative immunofluorescence co‐staining images of A2B5 (red), NG2 (green), and DAPI (blue) in primary OPCs. (D) Quantitative analysis of A2B5⁺/ NG2⁺ cells in each field, n = 3–5 images from at least three independent experiments. (E, F) The pattern diagram and the exploratory preference for novel objects in the novel object recognition results 1‐month after BCAS, n = 10 mice. (G, H) The percentage of spontaneous alternations in the Y‐maze test results 1‐month after BCAS, n = 10 mice. (I) Grid walking test results 1‐month after BCAS, n = 10 mice. (J) Rotarod test results 1‐month after BCAS, n = 10 mice. There was no difference in body weight between mice in each group. The data for each group conformed to a normal distribution. p Value was determined by ANOVA with Bonferroni's post‐hoc test. *p < 0.05, **p < 0.01, between the indicated groups, and ns indicates non‐significance. Data are represented as means ± SEM.