Age-dependent somatic expansion of the ATXN3 CAG repeat in the blood and buccal cell DNA of individuals with spinocerebellar ataxia type 3

Spinocerebellar ataxia type 3 (SCA3), a currently untreatable disorder, is caused by the expansion of a genetically unstable polyglutamine-encoding complex CAG repeat in the ATXN3 gene. Longer alleles are generally associated with earlier onset and frequent intergenerational expansions mediate the anticipation observed in this disorder. Somatic expansion of the repeat has also been implicated in disease progression and slowing the rate of somatic expansion in patients has recently been proposed as a therapeutic strategy. Here, we utilised high-throughput ultra-deep MiSeq amplicon sequencing of the ATXN3 repeat to precisely define the number of repeats, the exact sequence structure, the phased genotype of an adjacent single nucleotide polymorphism and to accurately quantify somatic expansion in blood and buccal cell DNA samples of a cohort of individuals with SCA3 from the Azores islands (Portugal). We revealed systematic mis-sizing of the ATXN3 repeat and high levels of inaccuracy of the traditional fragment length analysis approach that have important implications for attempts to identify modifiers of clinical and molecular phenotypes, including genetic instability. Quantification of somatic expansion in blood DNA revealed the expected effects of age and CAG repeat length, although the effect of repeat length was surprisingly modest with much stronger associations with age at sampling. We also observed an association of the downstream rs12895357 single nucleotide polymorphism with the rate of somatic expansion, and a higher level of somatic expansion in buccal cell DNA compared to blood. Although the levels of somatic expansion are much lower per repeat unit, the average level of somatic expansion at the ATXN3 locus in SCA3 patients is much higher than is typically observed at the HTT locus in Huntington disease patients. These data suggest that the ATXN3 locus in SCA3 patients in blood or buccal cell DNA might serve as a good biomarker for clinical trials testing suppressors of somatic expansion with peripheral exposure.