Long somatic DNA-repeat expansion drives neurodegeneration in Huntington disease

Huntington Disease (HD) is a fatal genetic disease in which most striatal projection neurons (SPNs) degenerate. The central biological question about HD pathogenesis has been how the disease-causing DNA repeat expansion (CAGn) in the huntingtin (HTT) gene leads to neurodegeneration after decades of apparent latency. Inherited HTT alleles with a longer CAG repeat hasten disease onset; the length of this repeat also changes over time, generating somatic mosaicism, and genes that regulate DNA-repeat stability can influence HD age-at-onset. To understand the relationship between a cells CAG-repeat length and its biological state, we developed a single-cell method for measuring CAG-repeat length together with genome-wide RNA expression. We found that the HTT CAG repeat expands from 40-45 CAGs to 100-500+ CAGs in HD-vulnerable SPNs but not in other striatal cell types, with these long DNA-repeat expansions acquired at different times by individual SPNs. Surprisingly, somatic expansion from 40 to 150 CAGs had no apparent effect upon gene expression - but neurons with 150-500+ CAGs shared profound gene-expression changes. These expression changes involved hundreds of genes, escalated alongside further CAG-repeat expansion, eroded positive and then negative features of neuronal identity, and culminated in expression of senescence/apoptosis genes. Rates of striatal neuron loss across HD stages reflected the rates at which neurons entered this biologically distorted state. Our results suggest that HTT CAG repeats in striatal neurons undergo decades of biologically quiet expansion, then, as they asynchronously cross a high threshold, cause SPNs to degenerate quickly and asynchronously. We conclude that, at any moment in the course of HD, most neurons have an innocuous (but unstable) huntingtin gene, and that HD pathogenesis is a DNA process for almost all of a neurons life.