Application of Precision Genome Editing for Neurodegenerative Disorders

Abstract

The rapid development of genome editing technologies has allowed researchers to precisely and efficiently modify the mammalian genome. Remarkably, precision gene editing tools based on Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) have already entered the clinic and are being used to treat human patients. The focus on my doctoral thesis has been to apply precision gene editing tools in the development of potential treatments for neurodegenerative diseases. First, I describe the application of base editing as a therapeutic intervention for spinal muscular atrophy (SMA). SMA is the leading genetic cause of infant mortality. SMA is caused by homozygous loss or mutation in the essential survival motor neuron 1 (SMN1) gene that leads to SMN protein insufficiency, resulting in loss of motor neurons and paralysis. SMN2 is a nearly identical gene that partially compensates for the loss of SMN1. However, SMN1 and SMN2 differ by a single C:G-to-T:A substitution at nucleotide position 6 of exon 7 (C6T) that results in exon 7 skipping in mRNA transcript and production of rapidly degraded truncated SMNΔ7 protein. We developed genome editing approaches targeting SMN2 and 1) converted SMN2 T6>C by base editing or 2) modified five SMN2 regulatory regions with nucleases or base editors to upregulate SMN levels. We determined that base editing of exon 7 C6T resulted in the greatest upregulation of SMN protein, therefore, we selected this strategy for in vivo validation. We packaged the optimized base editor into AAV9 and delivered the strategy via intracerebroventricular (ICV) injection to neonatal Δ7SMA mice, a mouse model of severe SMA. We observed 87% average T6>C conversion in the cortex and over 40% average T6>C editing in the spinal cord of the treated animals. Base editing treatment resulted in improved motor function and extended average lifespan. We extended the therapeutic window for gene editing by co-administrating with an approved SMA drug (nusinersen). This one-time co-administration resulted in further improvements in motor function and extended survival to 111 days, compared to 17 days for untreated mice. These findings demonstrate the potential of a one-time base editing treatment for SMA. Secondly, I describe a novel application of base editors for modifying DNA repeat expansions in genes associated with trinucleotide repeat (TNR) disorders. TNR diseases are neurological movement disorders associated with somatic expansion of trinucleotide repeat sequences. TNR sequences become unstable in a length-dependent manner and are known to drive disease progression, inheritance, and anticipation. The most common pathogenic triplet sequence is ‘CAG/CTG’, which occurs in almost half of the known pathogenic TNR loci, including the HTT gene in Huntington’s Disease. The most prevalent hereditary ataxia in humans, Friedreich’s ataxia (FRDA), is caused by the expansion of ‘GAA’ repeats at the FXN locus. Small nucleotide changes within repeat tracts have been shown to reduce repeat instability in cell and animal models. In patients, benign TNR interruptions such as ‘CAA’ triplets in CAG repeats, or ‘GAG’ or ‘GGA’ in GAA repeats, are associated with reduced somatic instability. These interruptions also lead to delayed onset and progression of the disease and result in overall milder or absent clinical features compared to individuals with uninterrupted repeats of a similar length. We developed base editing strategies to introduce CAA interruptions at CAG repeats and A:T-to-G:C interruptions at GAA repeats and selected the most efficient editing approaches for in vivo validation. We administered the optimized base editors in vivo via ICV into neonatal mice and observed efficient base editing across disease-relevant tissues in mouse models of HD and FRDA. Lastly, we demonstrated that repeat interruptions significantly reduced somatic instability in the mouse brain. Finally, I describe the first application of prime editing to remove pathogenic trinucleotide repeat expansions both in patient cells and in vivo in mice. The GAA repeat expansions found in FRDA patients are located in the intron 1 of frataxin (FXN) gene. Pathogenic GAA repeats lead to transcriptional silencing of FXN and result in frataxin protein deficiency, which is responsible for disease progression. We developed prime editing strategies that precisely remove GAA repeat expansions from FXN alleles, with a minimal loss of the surrounding regulatory or coding sequence. We delivered the optimized prime editing strategy via neonatal ICV injection to mouse models of FRDA and observed efficient excision of GAA repeats across multiple disease-relevant tissues. Furthermore, we demonstrated that PE-mediated deletion of GAA repeats results in upregulation of FXN expression in FRDA patient-derived cells and in mice.