Recombinant AAVs Target Custom Frataxin Therapeutic Cassettes to Tissues of Pathophysiologic Relevance in Friedreich's Ataxia

Abstract

Friedreich's Ataxia (FRDA) is a genetic disease affecting multiple organ systems, in which an intronic triplet repeat expansion (GAA) suppresses expression of frataxin (FXN). This mitochondrial protein is involved in iron-sulfur clustering, and non-regenerating tissues become especially vulnerable to oxidative stress and cell damage. In FRDA patients, neuronal damage in the CNS and PNS impairs motor coordination, manifesting as progressive gait and limb ataxia. In addition, oxidative stress in the heart leads to hypertrophic cardiomyopathy in early adulthood, which is often lethal. Naturally occurring AAV serotypes are widely used in research settings and clinical trials to deliver genetic cargo throughout the body when administered systemically. However, broad transduction of current systemic vectors (e.g. AAV9) can have deleterious off-target effects[1], underscoring the need for vectors that limit efficient transduction to specific disease targets. In this study, we aimed to develop precision AAV-mediated FXN gene replacement therapy to replace physiologic levels of FXN in the deep cerebellar nuclei, dorsal root ganglia, and the heart in two mouse models of FRDA, while minimizing transduction of organ systems and cell types that are not major contributors to the disease process, including the liver. To this aim, we used a dual-engineering approach to optimize both the delivery system and therapeutic cargo for targeted replacement of physiologic levels of FXN. Three constructs of the FXN gene containing different putative regulatory elements were cloned and packaged into AAVs for CNS and PNS targeting. In early experiments, a cocktail of PHP.eB and PHP.S was used for broad CNS and PNS targeting of these constructs after intravenous delivery in a knock-in-knock-out (KIKO) mouse model of FRDA, which recapitulates the genetic etiology of the disease, but exhibits a mild motor phenotype. With this approach, we achieved broad FXN expression throughout the target regions. Notably, cassettes containing FXN regulatory elements yielded expression patterns and levels mimicking the healthy endogenous state. To evaluate whether restoring expression patterns is sufficient to rescue sensorimotor deficits of FRDA, we transitioned to an inducible shRNA knockdown mouse model of FRDA (FXNiKD) which exhibits a more severe motor and coordination deficit than the KIKO model, as well as a cardiac phenotype. For CNS targeting in these experiments, we instead utilized the next generation vector AAV.CAP-B10, selected for its equally high efficiency of crossing the rodent BBB as PHP.eB but with unique bias toward neurons, and superior detargeting of the liver, cerebellar Purkinje cells, astrocytes and oligodendrocytes, which are clinical off-targets in FRDA. Using this new combination of CAP-B10 and PHP.S, FXN was systemically delivered to a small pilot cohort of symptomatic FXNiKD mice and wild-type controls, and sensorimotor phenotype was evaluated. In addition, due to the degenerative nature of FRDA, we are investigating whether targeted exogenous FXN therapy is able to prevent motor and coordination deficit development in FXNiKD mice when prophylactically applied.1.Hinderer, C., et al., Severe Toxicity in Nonhuman Primates and Piglets Following High-Dose Intravenous Administration of an Adeno-Associated Virus Vector Expressing Human SMN. Human Gene Therapy, 2018. 29(3): p. 285-298.

Additional details