Spinocerebellar ataxia type 7 cerebellar disease requires the coordinated action of mutant ataxin-7 in neurons and glia, and displays non-cell-autonomous bergmann glia degeneration

Abstract

Spinocerebellar ataxia type 7 (SCA7) is a dominantly inherited disorder characterized by cerebellum and brainstem neurodegeneration. SCA7 is caused by a CAG/polyglutamine (polyQ) repeat expansion in the ataxin-7 gene. We previously reported that directed expression of polyQ-ataxin-7 in Bergmann glia (BG) in transgenic mice leads to ataxia and non-cell-autonomous Purkinje cell (PC) degeneration. To further define the cellular basis of SCA7, we derived a conditional inactivation mouse model by inserting a loxP-flanked ataxin-7 cDNA with 92 repeats into the translational start site of the murine prion protein (PrP) gene in a bacterial artificial chromosome (BAC). The PrP-floxed-SCA7-92Q BAC mice developed neurological disease, and exhibited cerebellar degeneration and BG process loss. To inactivate polyQ-ataxin-7 expression in specific cerebellar cell types, we crossed PrP-floxed-SCA7-92Q BAC mice with Gfa2-Cre transgenic mice (to direct Cre to BG) or Pcp2-Cre transgenic mice (which yields Cre in PCs and inferior olive). Excision of ataxin-7 from BG partially rescued the behavioral phenotype, but did not prevent BG process loss or molecular layer thinning, while excision of ataxin-7 from PCs and inferior olive provided significantly greater rescue and prevented both pathological changes, revealing a non-cell-autonomous basis for BG pathology. When we prevented expression of mutant ataxin-7 in BG, PCs, and inferior olive by deriving Gfa2-Cre;Pcp2-Cre;PrP-floxed-SCA7-92Q BAC triple transgenic mice, we noted a dramatic improvement in SCA7 disease phenotypes. These findings indicate that SCA7 disease pathogenesis involves a convergence of alterations in a variety of different cell types to fully recapitulate the cerebellar degeneration.

Figure 1.

PrP-floxed-SCA7-92Q BAC mice exhibit Cre-dependent ataxin-7 expression and disease phenotype. A, Diagram of the PrP-floxed-SCA7 92Q BAC construct, indicating the location of the human ataxin-7 92Q cDNA flanked by loxP sites, a downstream EGFP reporter gene, and the first two exons and last exon from the murine prion protein gene (black boxes; not to scale). B, We performed RT-PCR analysis of cerebella with primers that coamplified murine and human ataxin-7, and then differentiated mouse endogenous ataxin-7 (m. atx7) from human transgenic ataxin-7 (h. atx7) by XmnI restriction digestion. PrP-floxed-SCA7-92Q transgene-positive mice yielded XmnI-cut bands, which were no longer present after crossing with CMV-Cre mice that ubiquitously express Cre recombinase. C, We performed quantitative RT-PCR analysis of cerebellar RNA with primers that coamplified murine and human ataxin-7, normalized to mouse 18S RNA levels, for nontransgenic and PrP-floxed-SCA7-92Q BAC mice (n = 3/group). We set the total ataxin-7 level to 1 in the nontransgenic control and found that the total ataxin-7 level in the PrP-floxed-SCA7-92Q BAC mice is ∼1.3, indicating that the BAC transgene is expressed at lower than endogenous levels. D, Immunofluorescent labeling of ataxin-7 (red) and calbindin (green) in cerebellar sections from 40-week-old PrP-floxed-SCA7-92Q BAC mice. Scale bar, 50 μm. E, We completed composite phenotype analysis for sets of nontransgenic (n = 14), PrP-floxed-SCA7-92Q BAC (n = 10), and CMV-Cre;PrP-floxed-SCA7-92Q BAC (n = 5) mice. PrP-floxed-SCA7-92Q BAC mice develop a significant phenotype, which is eliminated upon Cre-mediated excision (***p < 0.001, two-way ANOVA with Bonferroni's post hoc test). Error bars indicate SEM. F, We measured the mean latency to fall from the accelerating rotarod for sets of nontransgenic (n = 10), PrP-floxed-SCA7-92Q BAC (n = 10), and CMV-Cre;PrP-floxed-SCA7-92Q BAC (n = 5) mice. PrP-floxed-SCA7-92Q BAC mice performed significantly worse on the rotarod at 21 and 30 weeks of age compared with nontransgenics, and CMV-Cre;PrP-floxed-SCA7-92Q BAC mice performed much better than PrP-floxed-SCA7-92Q BAC mice at these time points (**p < 0.01, two-way ANOVA with Bonferroni's post hoc test). Error bars indicate SEM.

Figure 2.

Excision of polyQ-ataxin-7 from Bergmann glia partially ameliorates SCA7 neurological phenotypes. A, Immunofluorescence labeling of EGFP (green) and S100-β (red) in Gfa2-Cre;Z/EG cerebellar sections. B, Gfa2-Cre expression produced a 72% excision rate in Bergmann glia. C, We performed composite phenotype analysis for sets of nontransgenic (n = 16), PrP-floxed-SCA7-92Q (n = 13), Gfa2-Cre (n = 14), and Gfa2-Cre;PrP-floxed-SCA7-92Q BAC (n = 11) mice. Excision of polyQ-ataxin-7 from Bergmann glia in Gfa2-Cre;PrP-floxed-SCA7-92Q BAC mice yielded a markedly improved composite phenotype compared with PrP-floxed-SCA7-92Q BAC mice beginning at 20 weeks of age (*p < 0.05, ***p < 0.001, two-way ANOVA with Bonferroni's post hoc test). Error bars indicate SEM. D, When we analyzed these cohorts by accelerating rotarod testing, we found that PrP-floxed-SCA7-92Q BAC mice perform significantly worse than nontransgenic and Gfa2-Cre mice at 40 weeks of age (p < 0.001, two-way ANOVA with Bonferroni's post hoc test). Gfa2-Cre;PrP-floxed-SCA7-92Q BAC mice exhibited a trend toward improved rotarod performance, but this was not significant. Error bars indicate SEM.

Figure 3.

Excision of polyQ-ataxin-7 from Purkinje cells and inferior olive partially ameliorates SCA7 neurological phenotypes. A, Immunofluorescence labeling of EGFP (green) and calbindin (red) in Pcp2-Cre;PrP-floxed-SCA7-92Q BAC bigenic cerebellum and brainstem sections. Scale bars, 50 μm. B, Pcp2-driven Cre expression yielded a 52% excision rate in Purkinje cells and a 58% excision rate in inferior olive neurons. C, Composite phenotyping revealed that Pcp2-Cre;PrP-floxed-SCA7-92Q BAC mice (n = 9) are moderately improved compared with PrP-floxed-SCA7-92Q BAC mice (n = 8) from 20 to 30 weeks of age, but not compared with nontransgenic (n = 12) or Pcp2-Cre (n = 15) controls. Moreover, the neurological phenotype of Pcp2-Cre;PrP-floxed-SCA7-92Q BAC mice ceases to progress after 30 weeks of age, and thus becomes dramatically improved compared with their progressing PrP-floxed-SCA7-92Q BAC littermates at these time points (***p < 0.001, two-way ANOVA with Bonferroni's post hoc test). Error bars indicate SEM. D, Pcp2-Cre;PrP-floxed-SCA7-92Q BAC mice exhibit a trend toward improved rotarod performance at 22 weeks of age. By 40 weeks of age, however, Pcp2-Cre;PrP-floxed-SCA7-92Q BAC mice are markedly superior to their PrP-floxed-SCA7-92Q BAC littermates, and actually perform comparably to nontransgenic and Pcp2-Cre controls on the accelerating rotarod (**p < 0.01, two-way ANOVA with Bonferroni's post hoc test). Error bars indicate SEM.

Figure 4.

Excision of polyQ-ataxin-7 from Purkinje cells and inferior olive prevents cerebellar degeneration in PrP-floxed-SCA7-92Q BAC mice. A–D, Anti-calbindin antibody staining of cerebellar sections. Calbindin immunoreactivity in nontransgenic control cerebellum is intense (A), in contrast to the marked loss of calbindin immunoreactivity in the dorsal cerebellar folia of PrP-floxed-SCA7-92Q BAC mice (B); Gfa2-Cre;PrP-floxed-SCA7-92Q BAC mice (C) also display reduced calbindin immunoreactivity and molecular layer thinning, but cerebellar sections from Pcp2-Cre;PrP-floxed-SCA7-92Q BAC mice (D) exhibit intense calbindin staining comparable to nontransgenic controls. E, We quantified cerebellar molecular layer thickness by taking measurements at 100 μm intervals (perpendicular lines) along the outer region of three dorsal cerebellar folia per section. F, Results of cerebellar molecular layer quantifications (n = 6–7/group). PrP-floxed-SCA7-92Q BAC and Gfa2-Cre;PrP-floxed-SCA7-92Q BAC animals have significantly thinner molecular layers in their dorsal cerebellar folia (**p < 0.01 and ***p < 0.001, respectively, one-way ANOVA with Bonferroni's post hoc test). Cerebellar molecular layer thickness of Pcp2-Cre;PrP-floxed-SCA7-92Q BAC animals, however, was unchanged from nontransgenic controls. Error bars indicate SEM. Scale bars: A, 300 μm; E, 100 μm.

Figure 5.

Excision of polyQ-ataxin-7 from Purkinje cells and inferior olive, but not Bergmann glia, prevents loss of glial processes. A–D, Anti-GFAP antibody staining of dorsal cerebellar folia. Comparison of nontransgenic controls (A) to PrP-floxed-SCA7-92Q BAC mice (B) demonstrates an obvious reduction in Bergmann glia processes in the PrP-floxed-SCA7-92Q BAC mice; Gfa2-Cre;PrP-floxed-SCA7-92Q BAC cerebellar sections (C) resembled PrP-floxed-SCA7-92Q BAC sections, while Pcp2-Cre;PrP-floxed-SCA7-92Q BAC cerebellar sections (D) appeared comparable to nontransgenic controls. E, We quantified Bergmann glia processes by counting the number of GFAP-positive processes that cross a 150 μm line drawn parallel to, and ∼50 μm from, the pial surface. F, Results of quantifications of Bergmann glia processes (n = 6/group). PrP-floxed-SCA7-92Q BAC and Gfa2-Cre;PrP-floxed-SCA7-92Q BAC animals have significantly fewer Bergmann glia processes compared with nontransgenic mice (***p < 0.001, one-way ANOVA with Bonferroni's post hoc test). However, the number of Bergmann glia processes in Pcp2-Cre;PrP-floxed-SCA7-92Q BAC mice is intermediate between nontransgenic controls and PrP-floxed-SCA7-92Q BAC mice. Error bars indicate SEM. Scale bar, 30 μm.

Figure 6.

Excision of polyQ-ataxin-7 from Bergmann glia, Purkinje cells, and inferior olive greatly retards neurological disease onset in PrP-floxed-SCA7-92Q BAC mice. A, Composite phenotype scoring of Gfa2-Cre;Pcp2-Cre;PrP-floxed-SCA7-92Q BAC (trigenic) mice reveals markedly delayed onset of a progressive behavioral phenotype, and superior performance scores compared with PrP-floxed-SCA7-92Q BAC mice at all time points tested (**p < 0.01, ***p < 0.001, two-way ANOVA with Bonferroni's post hoc test; n = 6–8/group). Error bars indicate SEM. B, Trigenic mice consistently performed better on the accelerating rotarod than PrP-floxed-SCA7-92Q BAC mice at all time points; and rotarod performance was dramatically superior at 40 weeks of age (***p < 0.001, two-way ANOVA with Bonferroni's post hoc test; n = 6–10/group). Error bars indicate SEM.

Figure 7.

Interdependent connections between BG, PC, and IO account for the multi-cell type origin of SCA7 neurodegeneration. BG, PCs, and IO neurons each engage in bilateral interactions to yield an interdependent neural–glial network in which the function and survival of all cell types are assured together. During SCA7 disease pathogenesis, dysfunction of one cell type will inevitably impact the other connected cell types due to loss of trophic support and/or release of excitotoxic or inflammatory mediators. Altered function in one cell type thus culminates in all cell types failing to maintain appropriate signaling or other key functions, thereby exacerbating damage in the initial cell type while propagating abnormalities in the other connected cell types.