dAtaxin-2 mediates expanded Ataxin-1-induced neurodegeneration in a Drosophila model of SCA1

Abstract

Spinocerebellar ataxias (SCAs) are a genetically heterogeneous group of neurodegenerative disorders sharing atrophy of the cerebellum as a common feature. SCA1 and SCA2 are two ataxias caused by expansion of polyglutamine tracts in Ataxin-1 (ATXN1) and Ataxin-2 (ATXN2), respectively, two proteins that are otherwise unrelated. Here, we use a Drosophila model of SCA1 to unveil molecular mechanisms linking Ataxin-1 with Ataxin-2 during SCA1 pathogenesis. We show that wild-type Drosophila Ataxin-2 (dAtx2) is a major genetic modifier of human expanded Ataxin-1 (Ataxin-1[82Q]) toxicity. Increased dAtx2 levels enhance, and more importantly, decreased dAtx2 levels suppress Ataxin-1[82Q]-induced neurodegeneration, thereby ruling out a pathogenic mechanism by depletion of dAtx2. Although Ataxin-2 is normally cytoplasmic and Ataxin-1 nuclear, we show that both dAtx2 and hAtaxin-2 physically interact with Ataxin-1. Furthermore, we show that expanded Ataxin-1 induces intranuclear accumulation of dAtx2/hAtaxin-2 in both Drosophila and SCA1 postmortem neurons. These observations suggest that nuclear accumulation of Ataxin-2 contributes to expanded Ataxin-1-induced toxicity. We tested this hypothesis engineering dAtx2 transgenes with nuclear localization signal (NLS) and nuclear export signal (NES). We find that NLS-dAtx2, but not NES-dAtx2, mimics the neurodegenerative phenotypes caused by Ataxin-1[82Q], including repression of the proneural factor Senseless. Altogether, these findings reveal a previously unknown functional link between neurodegenerative disorders with common clinical features but different etiology.

Figure 1. dAtx2 Levels Modulate Ataxin-1[82Q]-Induced Eye Neurotoxicity

(A–E, A′–E′) Scanning electron microscopy (SEM) eye images and (F–J) retinal paraffin sections of eyes from the genotype combinations indicated on the top. Arrows indicate photoreceptor length. Transgenes are expressed form the gmr-GAL4 eye driver. (A, F) Control eyes show regularly arranged ommatidia, and evenly distributed interommatidial bristles (A). Control retinas have long, straight photoreceptors with no gaps (F). (B, G) SCA1^82Q eyes have disorganized ommatidia and interommatidial bristles are unevenly distributed or missing (compare B and A). SCA1^82Q retinas show fewer photoreceptors that are shortened and curved (compare G with F). (C, H) Suppression of external and retinal degeneration in SCA1^82Q flies with decreased dAtx2 levels (SCA1^82Q/dAtx2^X1). Note the regular arrangement of the ommatidia in SCA1^82Q/dAtx2^X1 animals when compared to SCA1^82Q eyes with normal levels of dAtx2 (compare C with B). Note also the improved retinal organization with longer straight photoreceptors (compare H with G). (D, I) Eyes co-expressing Ataxin-1[82Q] and wild type dAtx2 (SCA1^82Q/dAtx2^OE) have more disorganized ommatidia when compared to SCA1^82Q alone. Note the absence of bristles (compare D with B). In addition, photoreceptors are shorter and more degenerated (compare I with H). (E, J) Eyes overexpressing low levels of dAtx2 (dAtx2^OE) show mild ommatidial disorganization, with few missing bristles (compare E with A). Internally, photoreceptor cells are shortened (compare J with F). A-E flies were raised at 27°C and F-J at 25°C. (K) Semiquantitative RT-PCR revealing the levels of SCA1^82Q messenger RNA in flies with different levels of dAtx2. (L) Western blot analysis revealing the levels of Ataxin-1[82Q] protein in flies with different levels of dAtx2. Genotypes: (A–J): Control: yw/+; gmr-GAL4/UAS-LacZ. SCA1^82Q: UAS- SCA1^82Q [F7]/yw; gmr-GAL4/+. SCA1^82Q/dAtx2^X1: SCA1^82Q [F7]/yw; gmr-GAL4/+; dAtx2^X1/+. SCA1^82Q/ dAtx2^OE: UAS-SCA1^82Q [F7]/yw; gmr-GAL4/UAS- dAtx2 [4]. dAtx2^OE: gmr-GAL4/UAS- dAtx2 [4]. (A–E) SEM scale bar=100μm. (A′–E′) are 30μm X 30μm. (F–J) scale bar=10μm.

Figure 2. Specificity of the dAtx2/ Ataxin-1 Interaction

(A, B) Reduced levels of dAtx2 also suppress Ataxin-1[82Q]-induced motor dysfunction and shortened life span. (A) Motor performance quantification in flies of the indicated genotypes measured as climbing ability as a function of age. Control animals expressing just GFP perform well in the motor assay for 36 days (nrv2>GFP, black triangles), as do flies that are heterozygous for the dAtx2^X1 allele (nrv2>GFP/dAtx2^X1, green diamonds). Flies expressing Ataxin-1[82Q] in the nervous system show progressive motor dysfunction when compared with controls (nrv2> SCA1^82Q, blue circles). All nrv2> SCA1^82Q flies stop climbing after ∼23 days. Flies expressing Ataxin-1[82Q] in the nervous system but with decreased dAtx2 levels show improved motor performance (nrv2> SCA1^82Q/dAtx2^X1, red squares). Error bars represent standard deviation. Flies were raised at 27°C. (B) Survivorship in flies of indicated genotypes. Control flies (nrv2>GFP, black triangles) do not show significant mortality during the first 30 days (nrv2>GFP, black triangles). Expression of Ataxin-1 [82Q] in the nervous system leads to early mortality (nrv2> SCA1^82Q, blue circles). This phenotype is suppressed by decreasing the levels of dAtx2 (nrv2> SCA1^82Q/dAtx2^X1, red squares). (C–F) Reduced levels of dAtx2 do not suppress eye neurotoxicity or motor impairments caused by N-Htt^128Q. (C–E) Expression of N-Htt^128Q in the eye leads to retinal degeneration and loss of photoreceptors and other cell types (compare D with control shown in C). This eye phenotype is not altered in animals with reduced levels of dAtx2 (compare D and E). (F) Expression of N-Htt^128Q in the nervous system leads to progressive motor dysfunction (nrv2> N-Htt^128Q, red), and decreasing the levels of dAtx2 does not significantly suppress this phenotype (nrv2> N-Htt^128Q/dAtx2^X1, blue) Genotypes: (A–B): nrv2>GFP: nrv2-GAL4/UAS-eGFP. nrv2>GFP/dAtx2^X1: nrv2-GAL4/UAS-eGFP; dAtx2^X1/+. nrv2> SCA1^82Q: nrv2-GAL4/+; UAS-SCA1^82Q [M6]/+. nrv2> SCA1^82Q/dAtx2^X1: nrv2-GAL4/+; UAS-SCA1^82Q [M6]/dAtx2^X1. (C) yw; gmr-GAL4/UAS-GFP. (D) yw; gmr-GAL4/+; UAS-N-Htt^128Q/+. (E) yw; gmr-GAL4/+; UAS-N-Htt^128Q/dAtx2^X1. (F) nrv2>N-Htt^128Q: nrv2-GAL4/+; UAS-N-Htt^128Q/+. nrv2>N-Htt^128Q/dAtx2^X1: nrv2-GAL4/+; UAS-N-Htt^128Q/ dAtx2^X1.

Figure 3. dAtx2 Levels Modulate Ataxin-1[82Q]-Induced Loss of Mechanoreceptors

(A) Percentage of missing thoracic macrochaetae (large mechanoreceptor bristles) over a total of 26 in flies of the indicated genotypes raised at 25°C. Columns-1 and 2, no macrochaetae loss is detected in control flies or flies over-expressing dAtx2 (dAtx2^OE) in the thoracic sensory organ precursor (SOP) cells respectively with the scabrous-GAL4 driver. Column-3, mild decrease in thoracic macrochaetae number caused by expression of Ataxin-1[82Q] at low levels in the SOP cells (p<0.0001). Column-4, ∼80% decrease in the number of thoracic macrochaetae in animals co-expressing Ataxin-1[82Q] and dAtx2 (SCA1^82Q/dAtx2^OE) in the SOPs (p<0.0001). Error bars=s.e.m. (B–E) Images of the macrochaetae present in the posterior region of the notum and scutellum (white rectangle in B) of flies with same genotypes as (A). (B) Control thoraxes have eight macrochaetae in the selected area (white rectangle). (C-E) Over-expression of dAtx2 enhances Ataxin-1[82Q] induced macrochaetae loss. Black arrow indicates missing bristle in C and white arrow in E points to the only bristle present in the scutellum of an SCA1^82Q/dAtx2^OE animal. Flies grown at 25°C. (F) Anti-Sens immunofluorescence revealing the normal pattern of Sens distribution in the wing disc. White rectangle highlights the wing margin SOP and bristle precursor cells. (G–J) Close-up of the wing margin region from: control animals (G), animals expressing Ataxin-1[82Q] alone (H), wild-type dAtx2 (I) or coexpressing Ataxin-1[82Q] and dAtx2 (J). (K) Quantification of anti-Sens signal in the wing margin of animals of the genotypes indicated on the bottom. 20 wing discs were used per genotype and experiments carried out at 25°C (see materials and methods for more details). Bars=Std. Dev. (L) Percentage of lost macrochaetae in flies expressing Ataxin-1[82Q] with normal or decreased levels of dAtx2. Flies grown at 27°C to boost expression levels of the GAL4/UAS system. Column-1, control animals do not show macrochaetae loss. Column-2, ∼20% of thoracic macrochaetae are lost after expression of Ataxin-1[82Q] at high levels in the SOP cells (p<0.0001). Column-3, partial rescue of Ataxin-1[82Q]-induced macrochaetae loss in animals heterozygous for the dAtx2^X1 mutant (p<0.005). Bars= s.e.m. (M–N) Images of the macrochaetae present in the posterior notum and scutellum (white rectangle in B) of flies with same genotypes as L. Arrows indicate missing bristles. Flies grown at 27°C. Genotypes: control: sca^109–68-GAL4/UAS-GFP. dAtx2^OE:sca^109–68-GAL4/ UAS- dAtx2 [4]. SCA1^82Q: UAS- SCA1^82Q [F7]/+; sca^109–68-GAL4/+. SCA1^82Q/dAtx2^OE: UAS- SCA1^82Q [F7]/+; sca^109–68-GAL4/ UAS- dAtx2 [4]. SCA1^82Q/dAtx2^X1: UAS-SCA1^82Q [F7]/+; sca^109–68-GAL4/+; dAtx2^X1/+.

Figure 4. The Ataxin-1[82Q] and Ataxin-2 Proteins Physically Interact

(A) GST Co-AP pull-down experiments between dAtx2 and GST-constructs carrying human Ataxin-1[82Q] or Ataxin-1[82Q] with S776A mutation. (B) GST Co-AP experiments between human Ataxin-2 (Myc-hATXN2) and GST-Ataxin-1 with different polyglutamine lengths. (C) Interaction between GST-Ataxin-1[2Q] and endogenous hATXN2 in cell culture. (D) Co-AP pull down experiments between GST-Ataxin-1 and Flag-dAtx2 after nuclear/cytoplasmic fractionation. Tubulin is used as cytoplasmic marker and Mecp2 as nuclear marker. (E) Comparative analysis of the interaction of hAtaxin-2 with different domains of the Ataxin-1 protein. Lane-1, GST alone does not pull down Myc-hAtaxin2. Lane-2, GST-ATXN1^82Q pulls down Myc-hAtaxin2. Lane-3 expanded N-terminal Ataxin-1 (GST-ATXN1N-term^82Q) lacking the AXH domain (aa# 1–575) also pulls down Myc-hAtaxin-2 with high affinity. Lanes-4 and 5 show that both the C-terminus portion of Ataxin-1 containing the AXH domain (aa# 529–816; lane-4) or the AXH domain alone (aa# 558–700; lane-5) pull down Myc-hAtaxin2. These interactions are weaker than that of the N-terminus part (compare lanes 4 and 5 with lane 3). Cell lysates for co-AP were ran through Glutathione conjugated beads, and blots stained with anti-GST or anti-Myc. Immunoblots for dAtx2, Myc-hAtaxin2 and GFP carried out on cell lysates before the co-AP pull-down reveal similar levels of both proteins between samples.

Figure 5. Expanded Ataxin-1 Induces Nuclear Accumulation of Ataxin-2 in Drosophila and Human Neurons

(A–C) Longitudinal paraffin sections through adult fly retinas stained with anti-dAtx2 antiserum (green). Nuclei are visualized with anti-Lamin (white). Controls expressing only gmr-GAL4 driver (A) or gmr-GAL4 driver plus UAS-dAtx2 (C) show no detectable nuclear dAtx2 staining; arrowhead in (C) points to cytoplasmic accumulation of dAtx2. (B) dAtx2 localizes to the nucleus and accumulates in NIs (arrows) in retinal cells of SCA1^82Q animals. Also notice additional diffuse dAtx2 signal in the nuclei. (D) Pattern of ok107-GAL4 in the VNC, showing the area we analyzed (white square). (E–E′′) Confocal images of control Drosophila neurons in the larval ventral ganglion expressing CD8:GFP from the ok107-GAL4 neuronal driver and stained with anti-dAtx2 antibody. (E) No dAtx2 staining (red) is detected in the neuronal nucleus. Nuclear Lamin is shown in white. (E′) Same neurons as E showing CD8:GFP to reveal the neuronal cytoplasm. Images are merged in E′′. (F–F′′) Confocal images of ok107-GAL4 neurons expressing Ataxin-1[82Q] (SCA1^82Q) stained in parallel to E–E′′. (F) Shows dAtx2 signal (red) inside the nucleus (white) in the form of NIs (arrows), in addition to the cytoplasmic dAtx2 signal. (F′) Shows the same neurons visualized with CD8:GFP showing neuronal cytoplasm (green). Images are merged in F′′. (G–G′′) Triple immunofluorescence for Ubiquitin (blue), dAtx2 (red) and CD8:GFP (green) in ok107-GAL4 neurons expressing Ataxin-1[82Q]. Notice the overlapping of the dAtx2 (G) and Ubiquitin (G′) signals in the merged image in G′′ (arrows point to nuclear inclusions). (H, I) Human pontine neurons stained with anti-human Ataxin-2 antibody (brown). Hematoxylin (blue) was used for counterstaining. (H) No staining is detected in neuronal nuclei of non SCA1 postmortem tissue. (I) Ataxin-2 accumulates in NI (arrowhead) in a SCA1 neuron. Genotypes: (A) gmr-GAL4/+. (B) UAS-SCA1^82Q[F7]/+; gmr-GAL4/+. (C) gmr-GAL4/UAS-dAtx2[4]. (D, E-E′′) UAS-CD8:GFP/+; +; ok107-GAL4/+. (F-F′′ and G-G′′) UAS-CD8:GFP/+; UAS-SCA1^82Q [M6]/+; ok107-GAL4/+.

Figure 6. Pathogenic Ataxin-1[82Q], but Not Unexpanded Ataxin-1, Induces Nuclear Accumulation of dAtx2 In Vivo

(A–D) Anti-dAtx2 immunofluorescence (red) in ventral ganglion neurons expressing different Ataxin-1 transgenes and CD8-GFP (green) with the ok107-GAL4 driver. Nuclear Lamin (white) is also shown. (A) Control neurons show no nuclear dAtx2. (B) Neurons expressing Ataxin-1 with only 2Q (SCA1^2Q) present no nuclear dAtx2. (C) Neurons expressing wild type human Ataxin-1 with 30Q (SCA1^30Q) fail to show nuclear dAtx2 as well. (D) Neurons expressing expanded Ataxin-1[82Q] (SCA1^82Q) present nuclear dAtx2. Genotypes: (A) UAS-CD8:GFP/+; +; ok107-GAL4/+. (B) UAS-CD8:GFP/+; UAS-SCA1^2Q [F14]/+; ok107-GAL4/+ (C) UAS-SCA1^30Q [F1]/yw; UAS-CD8:GFP/+; ok107-GAL4/+. (D) UAS-CD8:GFP/+; UAS-SCA1^82Q [M6]/+; ok107-GAL4/+.

Figure 7. Nuclear Accumulation of dAtx2 Increases Its Toxicity In Vivo

(A–C) Immunofluorescence staining to reveal the localization of dAtx2 in neurons overexpressing CD8:GFP together with wild-type dAtx2 (dAtx2^OE) or dAtx2 with a nuclear export or a nuclear localization signal (dAtx2^NES or dAtx2^NLS, respectively). (A and A′) dAtx2 localizes to the cytoplasm even when overexpressed. (B and B′) dAtx2^NES distribution is similar to that of wild-type dAtx2. (C) nuclear localization of dAtx2^NLS. (D) Anti-dAtx2 immunoblot demonstrating similar expression levels of the different dAtx2 transgenes used in this work. The transgenes were driven by gmr-GAL4. Larval eye imaginal discs (ten per genotype) were used to avoid artifact potentially caused by cell degeneration or loss. dTubulin was blotted for loading control and larvae cultured at 25°C. Relative protein expression levels are indicated below normalized to the wild-type dAtx2 overexpressing line (dAtx2^OE, underlined) (E–H) SEM images showing the eye phenotypes caused by the different dAtx2 transgenes at 25°C. Transgenes were expressed with gmr-GAL4. (E) Control eyes. (F) Wild-type dAtx2 over-expression (dAtx2^OE) in the eye causes a mild phenotype with some ommatidial disorganization and bristle loss. (G) Expression of dAtx2^NES causes a similar phenotype to that caused by wild-type dAtx2. (H) Expression of dAtx2^NLS at similar levels as wild-type dAtx2 causes severe eye degeneration; the structure of the ommatidia is lost and interommatidial bristles are absent. In addition there is a strong decrease in eye size. Genotypes: (A, A′) UAS-CD8-GFP/UAS-dAtx2[4]; ok107-GAL4/+. (B, B′) UAS-CD8-GFP/+; UAS-dAtx2^NES/+; ok107-GAL4/+. (C) UAS-CD8-GFP/UAS-dAtx2^NLS; ok107-GAL4/+. (D–H) Control: gmr-GAL4/UAS-GFP. dAtx2^OE: gmr-GAL4/UAS-dAtx2[4]. dAtx2^NLS: gmr-GAL4/ UAS-dAtx2^NLS. dAtx2^NES: gmr-GAL4/+; UAS-dAtx2^NES /+.

Figure 8. Nuclear dAtx2 Induces Decreased Levels of Senseless (Sens) Protein and Loss of Mechanoreceptors

(A–D′) Immunofluorescence staining revealing Sens endogenous pattern in the wing margin SOPs on animals of genotypes indicated on top. (A and A′) Distribution of Sens in wing discs expressing a neutral NLS-DsRed protein (RedStinger) in the dpp territory (red in A′). Sens is detected as two parallel rows of cells (SOPs of the dorsal and ventral wing margins) uninterrupted in the area where they cross the dpp territory (arrowhead). (B, B′) Ataxin-1[82Q] eliminates Sens signal in the wing discs in a cell autonomous manner. Notice the gap in Sens signal when it crosses the dpp territory (arrowhead) expressing Ataxin-1[82Q] (red in B′). (C, C′) Expression of dAtx2^NES (red in C′) in the dpp territory does not affect Sens signal (arrowhead). (D, D′) Expression of dAtx2^NLS (red in D′) in the dpp territory reduces Sens signal in the SOPs, note the gap in Sens pattern (arrowhead). (E–P) Immunofluorescence staining revealing Sens (red, anti-Sens) and the dAtx2^NES or dAtx2^NLS proteins (green, anti-Flag), in salivary gland cells of the indicated genotypes. Nuclei are visualized using anti-Lamin (blue). (E–H) Control salivary gland cells showing robust Sens signal. Asterisks in G indicate the position of the nuclei. (I–L) dAtx2^NES-expressing salivary gland cells have normal Sens levels when compared to controls. (M–P) Expression of dAtx2^NLS in salivary gland cells causes a decrease of Sens protein. Sens signal (red) is dramatically decreased in dAtx2^NLS-expressing cells (compare N with F and J). Asterisks in N indicate the position of the nuclei. (Q–S) Effect of dAtx2^NES or dAtx2^NLS on adult thoracic macrochaetae formation after SOP specific expression. (Q) SOP-specific expression of dAtx2^NES causes no macrochaetae loss. (R) Expression of dAtx2^NLS in the SOP cells causes significant macrochaetae loss in the same conditions as Q. Arrows point out missing macrochaetae. (S) Quantification of lost macrochaetae in adult thoraxes of the same genotypes as Q and R. Column-1 shows no decrease in the number of macrochaetae per thorax in flies expressing dAtx2^NES. Column-2 reveals increased loss of macrochaetae following SOP specific expression of Atx2^NLS (p<0.0001). 20 animals were used per genotype; data was analyzed using Students t; error bars represent s.e.m. Experiments in Q–S were performed at 25°C. Genotypes: (A, A′) dpp-GAL4/ UAS-RedStinger. (B, B′) UAS-SCA1^82Q[F7]/+; +;dpp-GAL4/+. (C, C′, I–L) dpp-GAL4/UAS-dAtx2^NES. (D, D′, M–P) UAS-dAtx2^NLS /+; dpp-GAL4/+ (E–H) dpp-GAL4/+. (Q–S) dAtx2^NES: sca^109–68-GAL4/+; UAS-dAtx2^NES /+. dAtx2^NLS: sca^109–68-GAL4/ UAS-dAtx2^NLS.