Neurodegenerative diseases caused by short expansive repeats like the (CAG) in Huntingtons disease (Orr 2012) or the (GGGGCC) repeat in C9orf72-associated Amyotrophic lateral sclerosis (ALS)/Frontotemporal dementia (FTD) (DeJesus-Hernandez et al. 2011) undergo an unusual type of translation called repeat associated non-AUG-dependent (RAN) translation (Cleary and Ranum 2014). Interestingly, RAN translation occurs without an AUG start codon (Cleary and Ranum 2014). This allows for the (GGGGCC) repeat mutation to be translated, even though it is located in the intron between exon 1 and exon 2 of the C9orf72 gene, which would normally be spliced out and degraded (DeJesus-Hernandez et al. 2011). Translation of the repeat occurs in all 3 reading frames, leading to the production of three distinct dipeptide repeat proteins (DPRs). RAN translation begins within the (GGGGCC) repeat, but the exact translation initiation site remains unclear. However, RAN translation does not stop at the end of the repeat and will continue to translate the intronic sequence until it reaches a stop codon. This means that each of the distinct DPRs will be fused to peptides encoded in the downstream intron sequence. Because the DPRs are derived from intron sequence that is spliced out of the mature C9orf72 mRNA, none of these intron-derived DPR fusion peptides are incorporated into the normal C9orf72 protein. While it is known that the DPR fusion peptides are made in patients, the precise sequences of the DPR fusion peptides that they produce is not currently known. Therefore, questions about where precisely RAN translation initiates, how many repeats are produced, and whether the number of repeats produced are uniform or heterogenous remain important but unresolved questions.There is also a C9orf72 antisense transcript, which contains the complementary repeat sequence (GGCCCC). This antisense transcript also undergoes RAN translation to produce another three DPRs (Zu et al. 2013). Therefore, a single DNA repeat expansion in one gene gives rise to six distinct DPRs. These DPRs form
p62 positive/pTDP-43 negative inclusions that are distinct hallmarks of C9orf72-associated ALS/FTD (Cleary and Ranum 2014). Our laboratory as well as others have shown that two of these DPRs, proline-arginine (PR) and glycine-arginine (GR) are highly toxic (Kwon et al. 2014; Wen et al. 2014; Rudich et al. 2017), however the mechanisms of toxicity are poorly defined.In order to study the mechanisms that cause C9orf72-associated ALS/FTD PR and GR toxicity, we utilized the Caenorhabditis elegans model system. With short lifespans (3-4 weeks), a conserved neuromuscular system, and a genome that encodes ~20,000 genes with many conserved human homologs, the C. elegans model system is highly relevant for the study of aging and age-related diseases like ALS (Olsen et al. 2006). To study how PR and GR cause toxicity in C. elegans, we created animals expressing codon-optimized (PR)50-GFP and (GR)50-GFP (Rudich et al. 2017). With this approach, we are able to observe the effects of a single DPR at a time, without additional contributions from either the loss of the C9orf72 gene expression, introduction of the G4C2 repeat containing RNA, or the other five RAN translated DPRs. Therefore, this is a pure DPR model. Our laboratory has previously shown (PR)50 and (GR)50 to be toxic by causing a decrease in motility (paralysis) and arrested growth, when expressed in muscle (Rudich et al. 2017). Nuclear localization of these two DPRs was also discovered to be necessary and sufficient for toxicity in C. elegans (Rudich et al. 2017)...