Using Targeted Base Editing to Interrupt Neurodegeneration

DNA sequences often repeat, a feature that allows for evolutionary plasticity.¹ However, excess repetition can result in disease. In particular, expanded trinucleotide repeat (TNR) sequences are linked to at least 40 severe disorders, including the neurodegenerative Huntington’s disease (HD) and Friedreich's ataxia (FRDA).² David R. Liu, a molecular biologist and biochemist from the Broad Institute, seeks a way to control TNRs. His recent study, published in Nature Genetics, showed that base editing can introduce interruptions within TNRs, stabilizing or even reducing their expansion.³

How Repeat Sequences Cause Disease

DNA repeats encompass about 30 percent of the human genome.¹ However, they become unstable once they grow beyond a certain length, overriding natural suppression mechanisms and yielding aberrant proteins that culminate in gain- or loss-of-function mutations.¹ Repeat sequence expansion is dynamic, so the actual number of repeats present differs between individuals and across generations, resulting in varied phenotypic presentation.^1,4 Typically, longer repeats are associated with worse prognoses.³

Repeat sequences linked with human diseases vary in length from three to eleven nucleotides. However, expanded TNR diseases were discovered first and are the most numerous.² In their study, Liu and his team focused on CAG and GAA repeats, which are linked to HD and FRDA, respectively.^2,3 Using base editing, a technique first pioneered by Liu nearly a decade ago,^4,5 the Broad Institute research team interrupted TNR sequences in mouse disease models and cells from human patients to see the effects on TNR expansion.³

Interrupting Repeat Sequences

Scientists have found that naturally occurring interruptions within repeat sequences reduce repeat instability and transgenerational transmission. They also produce milder disease phenotypes and delay disease onset and progression.^6,7 Liu and his team therefore sought to introduce CAA interruptions into CAG repeats because alternating CAG and CAA codons do not undergo somatic expansion in mouse models.⁶ To do this, they targeted CTG repeats in the opposite strand with cytosine deaminases to change the cytosine to thymine, thereby creating complementary CAA codons rather than CAG.

Liu and his colleagues first assessed the effectiveness of their strategy in fibroblast cell lines derived from HD patients, finding that while unedited cells showed progressive somatic CAG repeat expansion over time, edited cells did not. Instead, they presented a degree of pathogenic repeat sequence contraction. They found similar trends in a mouse model of HD, where introducing CAA interruptions into CAG repeats significantly reduced the average size of CAG repeats in multiple brain regions.

Looking at Other TNRs

Liu and his colleagues next applied their base editing strategy to other TNRs. FRDA stems from the suppression of frataxin production caused by GAA repeat expansion in the FXN gene.⁸ Similar to how CAA interruptions in CAG repeat sequences affect HD, GGA and GAG interruptions within GAA repeat regions are associated with less severe FRDA disease phenotypes and later disease onset.^7,8 As such, Liu’s team targeted GAA triplets using deoxyadenine deaminases to induce A•T to G•C interruptions. Edited patient-derived primary fibroblasts showed an approximately 1.5-fold increase in FXN mRNA expression, partially alleviating suppressed frataxin production.

Finally, to assess how base editing affects GAA repeat instability, the scientists used a mouse model of FRDA that undergoes progressive instability starting at 18 weeks of age. Introducing A•T to G•C interruptions reduced the average size of GAA repeats and the number of somatic repeat expansions in the cortex. Further, Liu and his colleagues found that the treatment effectively halted spontaneous cortical expansions and that longer FXN alleles were potentially more prone to contractions.³

Targeted Base Editing Opens New Doors

Few treatment options are currently available for TNR disorders, and those that do exist can only delay progression or mitigate symptoms.¹ As Liu and his team have capably demonstrated, targeted base editing offers a way to interact with TNR sequences for both research and clinical applications. More work remains, but this study represents a strong first step towards potentially arresting or reversing TNR-associated neurological decline.