Significant oligodendrocyte progenitor and microglial cell death is a feature of remyelination following toxin-induced experimental demyelination

Abstract

The extent to which glial cell turnover features in successful remyelination is unclear. In this study, the rat caudal cerebellar peduncle-ethidium bromide lesion model was used to profile oligodendroglial and microglial/macrophage cell death and proliferation dynamics over the course of repair. Lesioned and control tissue was co-labelled with antibody markers for cell identity, proliferation, and apoptosis (TUNEL assay), then imaged at full thickness using confocal microscopy and quantified using custom CellProfiler pipelines. Early remyelination time points were marked by an increased density of total proliferating cells, including oligodendrocyte progenitor cells. Late remyelination time points featured increased TUNEL+ oligodendrocyte progenitor cells: however, most TUNEL+ cells within remyelinating lesions were Iba1+ microglia/macrophages. These results indicate that repairing lesions are characterized by a high degree of glial cell death and suggest that monitoring cell death-related by-products might have clinical value in the setting of remyelination.

Keywords: cell death; microglia; oligodendrocyte; oligodendrocyte progenitor cell; remyelination.

Graphical Abstract

Figure 1

Early remyelination time points are marked by an increased density of total proliferating cells, including OPCs. (A) Representative immunofluorescence images of control and 5-day-old lesion tissue. Dashed box indicates cell shown in (B). Hoechst (HO, blue); PDGFRα (red); Ki67 (yellow); composite (white). (B) Representative immunofluorescence images of a single Ki67+ OPC. (C) The density of total OPCs was quantified for each repair time point. (D) The density of total proliferating cells was quantified for each repair time point. (E) The density of proliferating OPCs (Ki67+/PDGFRα+ cells) was quantified for each repair time point. (F) The proportion of proliferating OPCs was quantified for each repair time point. Each individual data point represents the average across three regions of interest for a single animal. Bars represent mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 as determined by one-way ANOVA (C–F; control, n = 5; Day 2, n = 4; Day 5, n = 3; Day 10, n = 5; Day 14, n = 3; Day 21, n = 5), comparing each experimental group to the control group and applying Dunnett’s method for multiple comparisons. Abbreviation: dpl, days post-lesion.

Figure 2

Total TdT-mediated dUTP nick end labelling positive (TUNEL+) cell density remains elevated throughout the remyelination period. (A) Representative immunofluorescence images of control and 14-day-old lesion tissue. Hoechst (HO, blue); TUNEL (green); composite (white). (B) The density of total TUNEL+ cells was quantified for each repair time point. Each individual data point represents the average across three regions of interest for a single animal. Bars represent mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 as determined by one-way ANOVA (control, n = 5; Day 2, n = 3; Day 5, n = 3; Day 10, n = 5; Day 14, n = 3; Day 21, n = 5), comparing each experimental group to the control group and applying Dunnett’s method for multiple comparisons. Abbreviation: dpl, days post-lesion.

Figure 3

The densities of surviving and dying OLs remain stable throughout repair. (A) Representative immunofluorescence images of control and 2-day-old lesion tissue. Dashed box indicates cell shown in (B). Hoechst (HO, blue); Olig2 (yellow); PDGFRα (cyan); TUNEL (green); composite (white). (B) Representative immunofluorescence images of a single TUNEL+ OL. (C) The density of total OLs was quantified for each repair time point. (D) The density of TUNEL− OLs (‘surviving’ OLs) was quantified for each repair time point. (E) The density of TUNEL+ OLs was quantified for ‘early’ (Days 2/5) and ‘late’ (Days 10/14/21) repair. (F) Linear regression was used to model the density of TUNEL+ OLs over time. (G) The proportion of TUNEL+ OL was quantified for each repair time point. Each individual data point represents the average across three regions of interest for a single animal. Bars represent mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 as determined by one-way ANOVA (C–D, G; control, n = 5; Day 2, n = 4; Day 5, n = 3; Day 10, n = 5; Day 14, n = 3; Day 21, n = 5), comparing each experimental group to the control group and applying Dunnett’s method for multiple comparisons, or linear regression (F; control, n = 5; Day 2, n = 4; Day 5, n = 3; Day 10, n = 5; Day 14, n = 3; Day 21, n = 5). Abbreviation: dpl, days post-lesion.

Figure 4

Late remyelination time points are marked by an increased density of TUNEL+ OPC. (A) Representative immunofluorescence images of control and 10-day-old lesion tissue. Dashed box indicates cell shown in (B). Hoechst (HO, blue); Olig2 (yellow); PDGFRα (cyan); TUNEL (green); composite (white). (B) Representative immunofluorescence images of a single TUNEL+ OPC. (C) The proportion of TUNEL+ OPCs was quantified for each repair time point. (D) The density of TUNEL− OPCs was quantified for each repair time point. (E) The density of TUNEL+ OPCs was quantified for ‘early’ (Days 2/5) and ‘late’ (Days 10/14/21) repair. (F) Linear regression was used to model the density of TUNEL+ OPCs over time. Each individual data point represents the average across three regions of interest for a single animal. Bars represent mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 as determined by one-way ANOVA (C–E; control, n = 5; Day 2, n = 4; Day 5, n = 3; Day 10, n = 5; Day 14, n = 3; and Day 21, n = 5), comparing each experimental group to the control group and applying Dunnett’s method for multiple comparisons, or linear regression (F; control, n = 5; Day 2, n = 4; Day 5, n = 3; Day 10, n = 5; Day 14, n = 3; and Day 21, n = 5). Abbreviation: dpl, days post-lesion.

Figure 5

The majority of TUNEL+ cells are Iba1+ microglia/macrophages. (A) Representative immunofluorescence image of control and 5-day-old lesion tissue. Hoechst (HO, blue); Iba1 (yellow); TUNEL (green); composite (white). (B) Representative immunofluorescence image of TUNEL+ nuclei within Iba1+ microglia/macrophages from a 10-day-old lesion. (C) The density of total Iba1+ microglia/macrophages at each repair time point. (D) The density of total TUNEL+ microglia/macrophages at each repair time point. (E) The proportion of TUNEL+ microglia/macrophages (Iba1+ cells) at each repair time point. (F) The proportion of TUNEL+ cells that are Iba1+ microglia/macrophages at each repair time point. Each individual data point represents the average across three regions of interest for a single animal. Bars represent mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 as determined by one-way ANOVA (control, n = 5; Day 2, n = 3; Day 5, n = 3; Day 10, n = 5; Day 14, n = 3; Day 21, n = 5), comparing each experimental group to the control group and applying Dunnett’s method for multiple comparisons. Abbreviation: dpl, days post-lesion.

Figure 6

Oligodendroglial and microglia/macrophage cell densities during remyelination in a toxin-induced model of demyelination. TUNEL+ and Ki67+ subsets are shown. Datapoints represent averages across all animals at each time point (total OLs, n = 25; total OPCs, n = 25; total microglia/macrophages, n = 24; TUNEL+ OLs, n = 25; TUNEL+ OPCs, n = 25; TUNEL+ microglia/macrophages, n = 24; Ki67+ OPCs, n = 25). Standard errors are not shown to avoid figure cluttering but are available in Supplementary Table S2. Abbreviations: dpl, days post-lesion; OLs, oligodendrocytes; OPCs, oligodendrocyte progenitor cells.