Spinocerebellar ataxia type 1 (SCA1) is an adult-onset, dominantly inherited neurodegenerative disease caused by expansion of a glutamine repeat tract in ataxin-1 (ATXN1). Although the precise function of ATXN1 remains elusive, it seems to be involved in transcriptional repression. We find that mutant ATXN1 represses transcription of the neurotrophic and angiogenic factor vascular endothelial growth factor (VEGF). Genetic overexpression or pharmacologic infusion of recombinant VEGF mitigates SCA1 pathogenesis, suggesting a new therapeutic strategy for this disease.
Transcriptional alteration is the earliest pathogenic signature in SCA1 mouse models, suggesting that dysregulated gene expression caused by mutant ATXN1 is central to the pathogenesis of SCA1,2,3,4,5. However, little is known about the direct targets of ATXN1-induced repression, particularly those genes that are dysregulated in the vulnerable Purkinje cell population. Moreover, it is unclear whether correcting the expression of aberrantly expressed genes can ameliorate SCA1.
Because Purkinje cells, the most susceptible neurons in SCA1, constitute less than 0.1% of cerebellar cells, we used laser-capture microdissection (LCM) to enrich for these neurons to identify dysregulated gene expression (Fig. 1a and Supplementary Fig. 1a; see Supplementary Methods for all methodology). We used tissue from SCA1 knock-in mice (henceforth referred to as SCA1 mice), a model that closely mirrors human SCA1 (ref. 6). This SCA1 mouse model expresses an expanded version of ATXN1 with 154 glutamine repeats, in contrast to a non–disease-causing repeat of 2 in mice and less than 40 in humans.
As the yield of RNA from LCM material was insufficient for microarray screening, we used a PCR-based approach to examine the expression of candidate genes involved in key neurodegenerative pathways. One of the genes that we found downregulated in SCA1 mice encodes VEGFA (henceforth called VEGF), an angiogenic and neurotrophic factor implicated in motor neuron disorders7,8,9 (Fig. 1b). In SCA1 mice the levels of Vegfa mRNA were decreased at postnatal day 30 (Supplementary Fig. 1b), well before any behavioral or pathological signs. The decrease in Vegfa mRNA was more pronounced than that of three other genes previously described as downregulated in SCA1: Gsbs, Homer3 and Slc1a6 (refs. 1,2). In the cerebellum, VEGF is widely expressed in neurons, glia and endothelial cells10,11, with robust expression in Purkinje neurons (Fig. 1 and Supplementary Fig. 1c,d).
Low Vegfa mRNA expression was confirmed at the protein level, as Vegf protein abundance was decreased by ∼30% in SCA1 cerebella as compared with wild-type cerebella (Fig. 1c,d). The decrease in VEGF was most prominent in Purkinje neurons, although granule cells also showed a trend towards lower Vegfa mRNA levels and a significant decrease in Vegf protein abundance (Supplementary Fig. 1e–h). We also found that Vegfa mRNA levels were lower in a transgenic SCA1 mouse model that expresses mutant ATXN1 specifically in Purkinje neurons12 (Supplementary Fig. 1i).
We next asked whether ATXN1 is directly responsible for downregulation of Vegfa mRNA expression. To test whether ATXN1 modulates Vegfa promoter activity, we used a Vegfa luciferase reporter assay13. Both expanded (ATXN1-84Q) and wild-type (ATXN1-2Q) ATXN1 repressed reporter activity in a dose-dependent manner (Fig. 1e). The ability of even wild-type ATXN1 to cause repression is in keeping with the observation that overexpression of wild-type ATXN1 can induce pathology in animal models and the notion that SCA1 can in part be caused by a gain of ATXN1 normal function14. Mutating a phosphorylation site crucial for ATXN1 toxicity (S776A)15 disrupted repression at the Vegfa promoter (Fig. 1f). Thus, ATXN1 repression of the Vegfa promoter correlates with its toxicity in vivo. Consistent with the repression of Vegfa promoter activity, primary cerebellar neurons from SCA1 mice secreted lower amounts of Vegf (Supplementary Fig. 2a). We next performed chromatin immunoprecipitation (ChIP) with an ATXN1-specific antibody to test for ATXN1 occupancy of the Vegfa promoter. The Vegfa promoter (but not the promoter of a closely related family member, Vegfc8) was indeed more abundant in the ChIP lysates (Fig. 1g and Supplementary Fig. 2b). The Vegfa promoter was also hypoacetylated (Fig. 1h), and inhibitors of histone deacetylases relieved ATXN-1 84Q–induced repression at the Vegfa promoter (Supplementary Fig. 2c), suggesting a role for histone acetylation in ATXN1-induced repression.
Given that VEGF is an angiogenic factor, diminished VEGF abundance could contribute to cerebellar dysfunction by limiting angiogenesis. We observed a significant decrease in cerebellar microvessel density and total vessel length in SCA1 mice (Fig. 1i and Supplementary Fig. 3a–d), as well as evidence of hypoxia in SCA1 cerebella using pimonidazole, a hypoxic cell marker (Supplementary Fig. 3e,f).
In addition to its proangiogenic function, VEGF is also a neurotrophic factor. Thus, inadequate VEGF levels could also be deleterious by limiting neurotrophic support to cerebellar neurons. We tested the effects of reduced Vegf amounts on mixed cerebellar neuronal cultures that express Vegf and its receptor Vegfr2 (the predominant VEGF receptor in neurons). Inhibition of Vegf signaling using Vegfr2 tyrosine kinase inhibitors or a neutralizing antibody to Vegf resulted in decreased neurite length and increased cell death (Supplementary Figs. 4–6). Thus, reduced Vegf signaling compromises the growth and survival of not only Purkinje neurons but also other cerebellar neurons through a reduction in neurotrophic support.
We then examined whether restoring VEGF expression improves the SCA1 phenotype in mice. We crossed SCA1 mice with transgenic mice that overexpresses human VEGF in neurons starting at embryonic day 14 (VEGFtg/–; Supplementary Fig. 7a,b and ref. 16). Genetic overexpression of VEGF enhanced motor performance of mice (as assessed by rotarod) at 13 weeks (Fig. 2a) and 6 months (Fig. 2b) of age and improved SCA1 cerebellar pathology (Fig. 2c–e). The thickness of the molecular layer (Fig. 2c), staining intensity of calbindin (which specifically labels Purkinje cells; Fig. 2d,e) and cerebellar microvessel density (Fig. 2f) were increased.
We next tested whether pharmacological delivery of recombinant Vegf is beneficial after disease onset. As Vegf cannot cross the blood-brain barrier, we continuously delivered mouse Vegf using an intracerebroventricular osmotic pump (Supplementary Fig. 7c,d). Intracerebroventricular infusion of Vegf improved motor performance and restored cerebellar pathology (Supplementary Fig. 8).
Our findings suggest a role for VEGF in SCA1 pathogenesis and indicate that restoring VEGF levels may be a potentially useful treatment in patients with SCA1. In addition, these findings advance our understanding of neurodegeneration in SCA1. First, there seems to be cross-talk between the degenerating nervous system and the vascular system, Second, our results may explain the alterations in energy metabolism and decreased oxygen consumption identifiable by functional magnetic resonance imaging in people with SCA1 (ref. 17). Third, alterations in VEGF itself or sequelae of VEGF signaling in blood or cerebrospinal fluid of affected individuals could prove to be a biomarker of disease progression. Finally, because ataxia pathways interact in pathogenic hubs18, our results could prove relevant to other ataxias and possibly even other neurodegenerative syndromes.
We thank H. Zoghbi (Baylor College of Medicine) and H. Orr (University of Minnesota) for generously providing ATXN1 constructs and the SCA1 mouse models, and S. Leibovich (University of Medicine and Dentistry of New Jersey) and P. D'Amore (Harvard Medical School) for VEGF luciferase reporter constructs. We also thank K. Gobeske for assistance with the intracerebroventricular delivery methods, A. Ma for help with pathological analyses and V. Brandt for editorial assistance. This work was funded by US National Institutes of Health grants K02 NS051340, R21 NS060080 and R01 NS062051 (P.O.); a US National Organization for Rare Disorders grant (P.O.), a US Brain Research Foundation Grant (P.O.) and a US National Ataxia Foundation grant (P.O.). M.C. received funding from the US National Institutes of Health training grant T32. The authors wish to dedicate this manuscript to fellow scientist T. Spann, who is currently fighting amyotrophic lateral sclerosis.
Supplementary Figures 1–8 and Supplementary Methods
About this article
Journal of Molecular Neuroscience (2016)