Over-Expression of Meteorin Drives Gliogenesis Following Striatal Injury

Abstract

A number of studies have shown that damage to brain structures adjacent to neurogenic regions can result in migration of new neurons from neurogenic zones into the damaged tissue. The number of differentiated neurons that survive is low, however, and this has led to the idea that the introduction of extrinsic signaling factors, particularly neurotrophic proteins, may augment the neurogenic response to a level that would be therapeutically relevant. Here we report on the impact of the relatively newly described neurotrophic factor, Meteorin, when over-expressed in the striatum following excitotoxic injury. Birth-dating studies using bromo-deoxy-uridine (BrdU) showed that Meteorin did not enhance injury-induced striatal neurogenesis but significantly increased the proportion of new cells with astroglial and oligodendroglial features. As a basis for comparison we found under the same conditions, glial derived neurotrophic factor significantly enhanced neurogenesis but did not effect gliogenesis. The results highlight the specificity of action of different neurotrophic factors in modulating the proliferative response to injury. Meteorin may be an interesting candidate in pathological settings involving damage to white matter, for example after stroke or neonatal brain injury.

Keywords: GDNF; brain repair; forebrain injury; neurogenesis; oligodendrogenesis; striatum.

FIGURE 1

Assessment of lesioning, proliferation, and microglial response 6 weeks after intra-striatal injection of quinolinic acid (QA). Representative striatal sections of intact and lesioned brains 6 weeks post-lesion (arrow indicates injection site of QA lesion). The boxed area on each section corresponds to the adjacent 20× image detailing the immuno-labeling. (A) NeuN labeling in intact striatum and (B) 6 weeks post-lesion, detailing loss of neurons in adjacent 20× image. (C) BrdU labeling of an intact adult striatal section and (D) 6 weeks post lesion. (E) Iba1 staining in intact sections with adjacent 20× image detailing ramified microglia in the striatum and at (F) 6 weeks post-lesion with reactive microglia illustrated in the adjacent 20× image. Scale bar: 1 mm brain sections (A-F), 200 μm boxed images (A-F).

FIGURE 2

Validation of lentiviral expression 6 weeks post-lesion. Immunohistochemical analysis at 6 weeks post-injury in animals administered corresponding lentiviral over-expression constructs across three representative sections (arrow indicates injection site of lentivirus) (A,C,E) corresponding to 1.70, 1.34 and 0.98 mm from bregma, left to right. (A) lvGFP treated animals with GFP immuno-labeling with (B) 40× representative image of GFP⁺ astrocytic cells in the striatum. (C) lvMeteorin and (E) lvGDNF treated animals with Meteorin and GDNF immuno-labeling, respectively. (D) Tiled 40× image of lvMeteorin treated animals and (F) lvGDNF treated animals illustrate diffuse labeling within the striatum. Scale bar: 1 mm (A) analogous for (C,E), 0.1 mm (B,D,F).

FIGURE 3

Birth-dating of striatal neurons after quinolinic acid lesion to assess the effect of GDNF and Meteorin on these populations. (A) Quantification of BrdU⁺ cell density across treatment groups revealed no significant change in newborn cells in the striatum. (B) Quantification of BrdU⁺ cells co-expressing NeuN across treatment groups normalized to BrdU⁺ cell density revealed lvGDNF increased the number of new neurons in the ipsilateral striatum, with lvMeteorin having no effect compared to lvGFP controls. (C) Quantification of BrdU⁺/NeuN⁺/CalR⁺ cells normalized to BrdU⁺ cell density identified lvGDNF treatment trended toward an increase in this population. (D) CalR⁺ cells normalized to total BrdU⁺/NeuN⁺ cells indicated differentiation into this phenotype is altered across all newborn neurons. Overview image of BrdU/NeuN/CalR staining (E) and representative orthogonal z-stack confocal images of immuno-labeled (F) BrdU⁺/NeuN⁺/Darpp32^- neurons and (G) BrdU⁺/NeuN⁺/CalR⁺ interneurons (arrow). (H) NeuN negative interneurons were also observed as a primary subset of BrdU⁺/Calretinin⁺ cells (arrows). Data represents mean ± SEM. lvGFP n = 7; lvMeteorin n = 6; lvGDNF n = 5. Scale bars: 100 μm. cc, corpus callosum; lv, lateral ventricle; ac, anterior commissure; CalR, Calretinin.

FIGURE 4

Birth-dating of glial cells in the striatum after QA lesion and assessment of the effect of Meteorin on these populations. (A) Confocal orthogonal representation of cells in the striatum ipsilateral to lentiviral delivery and lesion confirm double labeling of BrdU⁺ cells co-stained for either s100β to identify new born astrocytes or (B) Olig2 and APC to identify newborn OPCs (BrdU⁺/Olig2⁺/APC^- cells) and mature oligodendrocytes (BrdU⁺/Olig2⁺/APC⁺ cells; boxed insets detail double and triple labeling of the cells). Newborn glial cells were quantified and represented as a percentage of total BrdU⁺ cell density. (C) Cell quantification of BrdU⁺/s100β⁺ cells normalized to BrdU⁺ cell density revealed lvMeteorin-treated animals had significantly more newborn s100β⁺ astrocytes in the ipsilateral striatum compared to lvGFP controls. For Oligodendroglial lineages, (D) BrdU⁺/Olig2⁺/APC^- (OPCs) revealed a significant increase in this population in lvMeteorin-treated animals compared to lvGFP-treated controls. (E) The percentage of BrdU⁺/Olig2⁺ cells that express APC (mature OLs) was assessed and was not significant between treatment groups. Oligodendrocyte lineage cell quantification was normalized to BrdU⁺ cell density. (F) Representative BrdU/Iba1 stain with example BrdU⁺/Iba1⁺ cell (boxed) and (G) quantification of BrdU+/Iba1+ cells as a percentage of total BrdU showed no significant difference between treatment groups. Data represents mean ± SEM, lvGFP n = 7; lvMeteorin n = 6; lvGDNF n = 5 for astrocyte and oligodendrocyte quantification and n = 3 for microglia cell counts across all groups. Scale bar: 100 μm. OPC, Oligodendrocyte precursor cell; OL, Oligodendrocyte.