Base Editing-Mediated Suppression of a Frameshifted FXN mRNA Isoform Restores Canonical Frataxin Expression

Friedreich’s Ataxia (FRDA) is an autosomal recessive neurodegenerative and multisystem disorder characterized by the expansion of GAA trinucleotide repeats within the first intron of the frataxin (FXN) gene. While normal alleles typically contain between 5 and 33 repeats, pathogenic expansions exceeding 66 repeats trigger localized epigenetic silencing, drastically reducing the expression of frataxin. This mitochondrial protein is indispensable for iron-sulfur (Fe-S) cluster biogenesis and cellular energy metabolism. Affected individuals exhibit severely diminished FXN levels, whereas heterozygous carriers maintain approximately 50% of normal expression and remain asymptomatic, suggesting a viable therapeutic window for gene restoration. Mechanistically, these expanded GAA tracts facilitate heterochromatin formation and impede transcriptional elongation, leading to a deficit in functional frataxin and subsequent mitochondrial failure. This pathophysiology underlies the major clinical manifestations of FRDA, including progressive ataxia, cardiomyopathy, and metabolic impairment. Because disease progression is driven by frataxin deficiency, strategies aimed at restoring FXN expression represent a central objective in FRDA research. Genome editing strategies have primarily focused on targeting the pathogenic GAA repeats. Early studies established the feasibility of this approach by demonstrating that the excision or contraction of the repeat tract reactivates FXN expression across various models. Recent dvancements in precision tools, such as base and prime editing, have further refined these interventions. Collectively, these studies provide a strong therapeutic rationale for the direct excision or correction of the GAA repeats to restore FXN expression. However, achieving safe and reproducible expansion removal remains a significant challenge. Current genome editing platforms often suffer from unpredictable cleavage patterns within the repetitive tract, as the Cas nuclease may target multiple, poorly defined sites throughout the long GAA sequence. Consequently, while genome editing represents a rapidly evolving platform for disease-modifying therapies in FRDA, refining the precision and consistency of repeat deletion is essential for clinical translation. Given the identification of multiple FXN isoforms, we propose an alternative genome editing strategy designed to modulate the transcriptional landscape of FXN. This favors the expression of the canonical FXN1 isoform, thereby enhancing the residual transcriptional activity of the productive transcript element within the FXN locus. Although patients with FRDA exhibit markedly reduced canonical FXN expression, residual transcriptional activity provides an opportunity to enhance the proportion of productive transcripts. Using base editing directed to the duplicate splice site (dup-ss) element, we demonstrate an increase in FXN1 mRNA levels and corresponding frataxin protein expression in human induced Pluripotent Stem Cells (hiPSCs). The co-delivery of in vitro-transcribed Base Editor mRNA and guide RNAs facilitated transient, tightly regulated editing activity, thereby limiting nuclease exposure and safeguarding genomic integrity. These findings establish dup-ss modulation via DSB-free base editing as a feasible and scalable strategy to increase functional FXN expression. They further support the broader applicability of transcript-level genome editing approaches and provide a foundation for future studies in advanced 3D disease models, including cardiac organoids and dorsal root ganglion organoids, to evaluate molecular and functional rescue in clinically relevant contexts.