Toward allele-specific targeting therapy and pharmacodynamic marker for spinocerebellar ataxia type 3

Abstract

Spinocerebellar ataxia type 3 (SCA3), caused by a CAG repeat expansion in the ataxin-3 gene (ATXN3), is characterized by neuronal polyglutamine (polyQ) ATXN3 protein aggregates. Although there is no cure for SCA3, gene-silencing approaches to reduce toxic polyQ ATXN3 showed promise in preclinical models. However, a major limitation in translating putative treatments for this rare disease to the clinic is the lack of pharmacodynamic markers for use in clinical trials. Here, we developed an immunoassay that readily detects polyQ ATXN3 proteins in human biological fluids and discriminates patients with SCA3 from healthy controls and individuals with other ataxias. We show that polyQ ATXN3 serves as a marker of target engagement in human fibroblasts, which may bode well for its use in clinical trials. Last, we identified a single-nucleotide polymorphism that strongly associates with the expanded allele, thus providing an exciting drug target to abrogate detrimental events initiated by mutant ATXN3. Gene-silencing strategies for several repeat diseases are well under way, and our results are expected to improve clinical trial preparedness for SCA3 therapies.

Fig. 1.. PolyQ ATXN3 proteins accumulate in the CSF and plasma from individuals with SCA3 and distinguish patients with SCA3 from controls.

(A and B) PolyQ ATXN3 in CSF (A) and plasma (B) was measured using our immunoassay as described (see Materials and Methods). The number of cases per study group is included below the graphs. Graphs represent means ± SEM. Statistical differences represent unadjusted and adjusted analyses. Complete statistical analyses can be found in table S3. *P < 0.05 and ***P < 0.001; n.s., nonsignificant differences. (C and D) Area under the receiver operating characteristic curve (AUC) for polyQ ATXN3 between SCA3 and healthy controls (C) or presymptomatic SCA3 carriers (D) in both the CSF and plasma. Additional AUC values with other controls are included in table S3.

Fig. 2.. Neurofilament light can also discriminate patients with SCA3 from controls.

(A and B) NFL in the CSF (A) and plasma (B) was measured using the commercial NF-light kit on the Simoa HD-1 analyzer (Quanterix) (see Materials and Methods). The number of cases per study group is included below the graphs. Graphs represent means ± SEM. Statistical differences represent unadjusted and adjusted analyses. Complete statistical analyses can be found in table S4. *P < 0.05 and ***P < 0.001. (C and D) AUC for NFL between SCA3 and healthy controls (C) or presymptomatic SCA3 carriers (D) in both the CSF and plasma. Additional AUC values with other controls are included in table S4.

Fig. 3.. The rs7158733 SNP is highly prevalent in our SCA3 patient cohort and associates with the expanded allele.

(A) Schematic depicting the reference “TAC¹¹¹⁸” allele (top) in which the presence of C¹¹¹⁸ allows corresponding codon (TAC¹¹¹⁸) to encode for tyrosine at position 349 of the protein sequence (Y349). The TAA¹¹¹⁸ allele (bottom) is characterized by the single nucleotide change to A¹¹¹⁸, encoding for a TAA¹¹¹⁸ stop codon at position 349 and leading to generation of a truncated variant of ATXN3. (B) Bar graph indicating the percentage of cases with either the TAA¹¹¹⁸ (salmon) or TAC¹¹¹⁸ allele (white). The exact number of cases is shown within each section of the bar graph. The darker salmon color depicts the proportion of cases that have the TAA¹¹¹⁸ on the expanded allele.

Fig. 4.. PolyQ ATXN3 proteins accumulate in human fibroblasts derived from ATXN3 CAG expansion carriers, and treatment with ATXN3 siRNA leads to varying degrees of polyQ ATXN3 down-regulation.

(A) Fibroblast lines were grown from 17 patients with SCA3, 13 healthy controls, 14 non–SCA3 ataxia patients, and two asymptomatic ATXN3 CAG expansion carriers (presymptomatic SCA3). PolyQ ATXN3 proteins were measured using our developed immunoassay. Bars represent means ± SEM. ***P < 0.001. (B) Spearman’s correlation analyses of polyQ ATXN3 in fibroblasts and matching CSF from ATXN3 CAG expansion carriers (N = 11). The solid line represents the estimated regression line. (C) Four healthy control lines and five SCA3 lines were grown in parallel, and the same amount of cells were treated with control or ATXN3 siRNA for 5 days. Lysates were extracted, and polyQ ATXN3 was measured using our immunoassay. Bars represent means ± SEM of at least three replicates per line. Information on ATXN3 CAG repeat length and the presence of the rs7158733 SNP on each allele from matching blood samples as well as the percentage of polyQ ATXN3 knockdown (KD) from the different fibroblast lines are indicated underneath the graph. Statistical analyses were performed using a two-way analysis of variance (ANOVA) followed by Sidak’s multiple comparisons test. *P < 0.05, **P < 0.005, and ****P < 0.0001. Individual-level data are shown in tables S8 and S9.