Craig S. McIntosh

Over 40 diseases, primarily affecting the nervous system, are caused by expansion of simple repetitive sequences found throughout the human genome, termed repeat expansion diseases. Expansions can occur in coding and non-coding regions of the genome, leading to several proposed mechanisms of disease, accumulation of either toxic RNA or toxic protein, although gain-of-function mechanisms are suggested causes of pathogenesis. Currently, there is no cure nor effective treatment strategy for any repeat expansion diseases. However, for many of these expansion diseases, splice-switching antisense oligonucleotides (AOs) may offer promise as a therapeutic strategy, as these compounds have already demonstrated efficacy in the treatment of other types of genetic disorders. Antisense oligonucleotides are short synthetic nucleic acid analogues, designed to target specific pre-mRNA sequences by reverse-complementary Watson-Crick binding, thereby modifying processing and/or abundance of the transcript and the sequence of the encoded protein. While there are a number of applications for AOs, this study focuses on their utility for preventing translation of toxic protein isoforms, either by altering the target transcript to encode a truncated protein isoform, or by disrupting the reading frame to downregulate endogenous protein production. The first part of this study focused on ameliorating the toxic polyglutamine tract found in the ataxin-3 protein that causes spinocerebellar ataxia type 3 (SCA3). One of nine known polyglutamine disorders, SCA3 is a clinically heterogeneous disease, primarily exemplified by progressive ataxia impairing the speech, balance and gait of affected individuals. SCA3 is caused by expansion of a glutamine-encoding tract located at the 5′ end of the penultimate exon (exon 10) of the ATXN3 gene transcript, resulting in conformational changes in ataxin-3 and a toxic gain-of-function. Here, we describe highly efficient removal of the toxic polyglutamine tract of ataxin-3 in vitro by phosphorodiamidate morpholino oligomers (PMOs). Additionally, these PMOs induced a potentially beneficial downregulation of both the expanded iv and non-expanded protein isoforms. As SCA3 has a typical age of onset in the fourth decade, the observed downregulation could delay age of onset by reducing the amounts of the toxic aggregates. Although we induce downregulation of both isoforms, we believe that the proportion of the truncated protein may be sufficient for overall function of ataxin-3, as some studies have shown ataxin-3 protein to be partially dispensable. Recently, several in vitro and in vivo studies have found that targeted knockdown of transcription elongation factors SUPT4H1, and to a lesser extent SUPT5H, can reduce aggregation of expanded transcripts and protein, and alleviate the disease phenotype in animal models of various expansion diseases. We therefore sought to investigate in vitro, the potential of AO-mediated SUPT4H1 downregulation as a therapeutic strategy. We found that our AOs were able to significantly downregulate SUPT4H1, with minimal changes to the rest of the transcriptome. We then assessed whether this downregulation of SUPT4H1 lead to a reduction in expanded ATXN3 mRNA and/or ATXN3 protein expression, however, unfortunately in the models available and under the current study, no modification to the ATXN3 transcript or protein was observed. This lack of effect may be due to the relatively short, expanded repeat lengths in SCA3 cell lines, and we therefore recommend that future studies assess genes with larger expansions, such as the 100-1000s repeat tracts frequently observed in myotonic dystrophy type 1 (DMPK). In order to create an efficient screening process for finding clinic-ready AOs, it is important to have a detailed understanding of the principles of AO design. We therefore present a comprehensive rationale for efficiently design and in vitro delivery of splice modulating AOs. These approaches and recommendations provide a streamlined methodology for any researcher developing AO therapeutics. The results presented in this thesis indicate that morpholino oligomers will provide superior benefit for the treatment of spinocerebellar ataxia type 3, without the toxic effects that result from other antisense oligomer chemistries. Additionally, AO-induced SUPT4H1 knockdown may yet demonstrate therapeutic v application for a multitude of expansion diseases, pending further investigation into the whole transcriptome effects and in vivo efficacy of this strategy. Lastly, our guidelines for therapeutic AO development should aid other researchers in creating the most efficacious and safe AOs for clinical trials. The work presented in this thesis contributes to the greater body of knowledge about the applications of AOs, as well as the need for reliable and systematic protocols in AO research and interpretation. With ongoing collaboration from our industry partners, Sarepta Therapeutics, there is hope that the work presented here will provide a solid foundation for further research into AO therapeutics for the treatment of neurodegenerative expansion diseases.