Neurodegeneration in a transgenic mouse model of multiple system atrophy is associated with altered expression of oligodendroglial-derived neurotrophic factors

Abstract

Multiple system atrophy (MSA) is a neurodegenerative disorder characterized by striatonigral degeneration and olivo-pontocerebellar atrophy. Neuronal degeneration is accompanied by primarily oligodendrocytic accumulation of alpha-synuclein (alphasyn) as opposed to the neuronal inclusions more commonly found in other alpha-synucleinopathies such as Parkinson's disease. It is unclear how alphasyn accumulation in oligodendrocytes may lead to the extensive neurodegeneration observed in MSA; we hypothesize that the altered expression of oligodendrocyte-derived neurotrophic factors by alphasyn may be involved. In this context, the expression of a number neurotrophic factors reportedly expressed by oligodendrocytes [glial-derived neurotrophic factor (GDNF), brain-derived neurotrophic factor (BDNF), and insulin-like growth factor 1 (IGF-1), as well as basic fibroblast growth factor 2 (bFGF2), reportedly astrocyte derived] were examined in transgenic mouse models expressing human alphasyn (halphasyn) under the control of either neuronal (PDGFbeta or mThy1) or oligodendrocytic (MBP) promoters. Although protein levels of BDNF and IGF-1 were altered in all the alphasyn transgenic mice regardless of promoter type, a specific decrease in GDNF protein expression was observed in the MBP-halphasyn transgenic mice. Intracerebroventricular infusion of GDNF improved behavioral deficits and ameliorated neurodegenerative pathology in the MBP-halphasyn transgenic mice. Consistent with the studies in the MBP-halphasyn transgenic mice, analysis of GDNF expression levels in human MSA samples demonstrated a decrease in the white frontal cortex and to a lesser degree in the cerebellum compared with controls. These results suggest a mechanism in which alphasyn expression in oligodendrocytes impacts on the trophic support provided by these cells for neurons, perhaps contributing to neurodegeneration.

Figure 1.

Altered expression of neurotrophic factors on αsyn overexpression in Tg mouse models of α-synucleinopathy. To examine the effect of αsyn on expression levels of neurotrophic factors, immunoblot analysis (A) was conducted on mouse brain homogenates from the posterior hemibrain, to detect the levels of αsyn, BDNF, IGF-1, bFGF, and GDNF in Tg mouse models expression αsyn under oligodendrocytic (MBP-hαsyn, n = 5; age, 8 months) or neuronal (PDGF-β-hαsyn, n = 5; age, 8 months; mThy1-hαsyn, n = 5; age, 8 months) promoters and αsynKO mice; representative blots from n = 3 of each group are shown. Quantitative analysis (B–G) of the levels of αsyn, BDNF, IGF-1, bFGF, and GDNF, respectively, is also presented. *Significant difference between NTg mice and the various αsyn Tg lines (p < 0.05, one-way ANOVA and post hoc Fisher). Error bars indicate SEM.

Figure 2.

Characterization of GDNF family members and receptors in MBP-hαsyn Tg mice. To examine the specificity of the αsyn effect on GDNF, immunoblot analysis (A) of GDNF-family members (Neurturin and Artemin) and GDNF receptors in MBP-hαsyn Tg mice (n = 5; age, 8 months) and NTG mice (n = 5; age, 8 months) was conducted on mouse brain homogenates from the posterior hemibrain; representative blots from n = 3 of each group are shown. Quantitative analysis of expression levels of Neurturin and Artemin (B) and of GDNF receptors (C) in MBP-hαsyn Tg mice is also presented. *Significant difference between MBP-hαsyn Tg and NTg mice (p < 0.05, one-way ANOVA and post hoc Fisher). Error bars indicate SEM.

Figure 3.

GDNF mRNA in MBP-hαsyn Tg mice. To examine the cell type-specific mRNA expression of GNDF, in situ hybridization with a probe against GDNF and colabeling with oligodendrocytic (MBP and GalC) and astroglial (GFAP) antibodies was conducted in NTg (A, C, E) and MBP-hαsyn Tg (B, D, F) mice in the frontal cortex and subjacent white matter. In situ hybridization with sense probe showed no signal in the NTg and MBP-hαsyn Tg mice (G and H, respectively). Quantitative analysis (I) of GDNF mRNA-positive oligodendrocytes and astrocytes in NTg and MBP-hαsyn Tg mice is also presented. Error bars indicate SEM. Scale bar, 50 μm.

Figure 4.

In vitro characterization of the effect of αsyn on GDNF expression. To examine the in vitro effect of αsyn on GDNF expression NPCs were differentiated to an oligodendrocytic phenotype, confirmed by MBP immunoreactivity (A, B). These cells were then infected with LV-control (C) or LV-αsyn (D); infection was confirmed by αsyn immunoreactivity. GDNF immunoreactivity was examined in the NPC-derived oligodendrocytes infected with LV-control vector (E) or LV-αsyn (F). To complement immunocytochemical studies, immunoblot analysis was also conduced to confirm differentiation, LV-αsyn expression, and the effect of LV-αsyn on levels of GDNF (G); conditioned media from these cells was used to determine levels of secreted GDNF by ELISA (H). KCl (55 mm) was used to enhance GDNF secretion as a positive control. Scale bar: A–F, 20 μm.

Figure 5.

Behavioral deficits in MBP-hαsyn Tg mice are attenuated by GDNF infusion. To assess the behavioral effects of GDNF infusion, motor behavior was assessed via the pole test; total time taken to descend the pole (A) and time taken to face downward on the pole (B) were recorded. Olfactory function was assessed using the buried pellet test, and time taken to find the pellet was recorded (C) (n = 6; age, 8 months in each group). *Significant difference between saline-infused NTg and MBP-hαsyn Tg mice (p < 0.05, one-way ANOVA and post hoc Fisher). ^#Significant difference between saline-infused and GDNF-infused MBP-hαsyn Tg mice (p < 0.05, one-way ANOVA and post hoc Fisher). Error bars indicate SEM.

Figure 6.

Immunohistochemical characterization of NTg and MBP-hαsyn Tg mice after GDNF infusion. To assess the effect of GNDF infusion on neuropathology, immunohistochemical analysis of GDNF immunoreactivity in saline-infused NTg mice (A), GDNF-infused NTg mice (B), saline-infused MBP-hαsyn Tg mice (C), and GDNF-infused MBP-hαsyn Tg mice (D) was performed and analyzed in E (n = 6; age, 8 months in each group). GFAP immunoreactivity in saline-infused NTg mice (F), GDNF-infused NTg mice (G), saline-infused MBP-hαsyn Tg mice (H), and GDNF-infused MBP-hαsyn Tg mice (I) was also measured and analyzed in J (n = 6; age, 8 months in each group). Scale bar, 200 μm (insets at higher magnification). *Significant difference between saline-infused NTg and MBP-hαsyn Tg mice (p < 0.05, one-way ANOVA and post hoc Fisher). ^#Significant difference between saline-infused and GDNF-infused NTg or MBP-hαsyn Tg mice (p < 0.05, one-way ANOVA and post hoc Fisher). Error bars indicate SEM.

Figure 7.

Immunohistochemical characterization of dendritic pathology and neuronal density in NTg and MBP-hαsyn Tg mice after GDNF infusion. Dendritic pathology was assessed via MAP-2 immunoreactivity in the frontoparietal cortex of saline-infused NTg mice (A), GDNF-infused NTg mice (B), saline-infused MBP-hαsyn Tg mice (C), and GDNF-infused MBP-hαsyn Tg mice (D), and analyzed in E (n = 6; age, 8 months in each group). Neurodegeneration was assessed via NeuN immunoreactivity in the frontoparietal cortex of saline-infused NTg mice (F), GDNF-infused NTg mice (G), saline-infused MBP-hαsyn Tg mice (H), and GDNF-infused MBP-hαsyn Tg mice (I), and analyzed in J (n = 6; age, 8 months in each group). Scale bar, 30 μm. *Significant difference between saline-infused NTg and MBP-hαsyn Tg mice (p < 0.05, one-way ANOVA and post hoc Fisher). ^#Significant difference between saline-infused and GDNF-infused MBP-hαsyn Tg mice (p < 0.05, one-way ANOVA and post hoc Fisher). Error bars indicate SEM.

Figure 8.

Dopamine marker immunoreactivity in motor regions of MBP-hsyn Tg mice after GDNF infusion. To examine the effects of GDNF infusion on dopaminergic cells, TH immunoreactivity in the caudoputamen of saline-infused NTg mice (A), GDNF-infused NTg mice (B), saline-infused MBP-hsyn Tg mice (C), and GDNF-infused MBP-hsyn Tg mice (D) was measured and analyzed in E (n = 6; age, 8 months in each group). TH immunoreactivity was also measured in the substantia nigra of saline-infused NTg mice (F), GDNF-infused NTg mice (G), saline-infused MBP-hsyn Tg mice (H), and GDNF-infused MBP-hsyn Tg mice (I), and analyzed in J (n = 6; age, 8 months in each group). Levels of DAT were also measured by immunohistochemistry in the caudoputamen of saline-infused NTg mice (K), GDNF-infused NTg mice (L), saline-infused MBP-hsyn Tg mice (M), and GDNF-infused MBP-hsyn Tg mice (N), and measured and analyzed in O (n = 6; age, 8 months in each group). Scale bars: A–D, K–N, 200 μm; F–I, 50 μm. *Significant difference between saline-treated MBP-hsyn Tg mice and NTg (p < 0.05, one-way ANOVA and post hoc Fisher). ^#Significant difference between saline-treated MBP-hsyn Tg mice and GDNF-treated MBP-hsyn (p < 0.05, one-way ANOVA and post hoc Fisher). Error bars indicate SEM.

Figure 9.

Immunohistochemical characterization of αsyn expression in NTg and MBP-hαsyn Tg mice after GDNF infusion. To determine the effect of GDNF infusion on αsyn levels, αsyn immunoreactivity in the cortex of saline-infused NTg mice (A, E), GDNF-infused NTg mice (B, F), saline-infused MBP-hαsyn Tg mice (C, G), and GDNF-infused MBP-hαsyn Tg mice (D, H) was measured and analyzed in I. Scale bars: A–D, 200 μm; E–H, 50 μm. *Significant difference between saline-infused NTg and MBP-hαsyn Tg mice (p < 0.05, one-way ANOVA and post hoc Fisher). Error bars indicate SEM.

Figure 10.

Immunoblot analysis of GDNF expression levels in human MSA brains. Immunoblot analysis was conducted to investigate the expression levels of GDNF in human MSA brain homogenates from the white matter of the frontal cortex and cerebellum (A) and quantified in B. GDNF levels were also quantified by ELISA to complement the immunoblot analysis (C). *Significant difference between human control and MSA samples (p < 0.05, one-way ANOVA and post hoc Fisher). Error bars indicate SEM.