nLuc: nanoluciferase; UTR: untranslated region. levels and ameliorates disease, suggesting poly-GA is pathogenic. Importantly, loss-of-function mutations in the eukaryotic translation initiation factor 2D (models. Our in vitro studies in mammalian cells yield similar results. Here, we show a conserved role for in DPR expression. is the most common monogenic cause of inherited amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD)1,2, and also causes up to 10% of what appears Prostaglandin E1 (PGE1) to be sporadic ALS. The G4C2 repeat expands in patients to hundreds or thousands of copies that vary in number in different cells of the same individual. This genetic insult is thought to cause ALS/FTD via three non-mutually exclusive mechanisms: (1) loss of function due to decreased expression of C9ORF72 protein, (2) RNA toxicity from bidirectionally transcribed sense (GGGGCC) and antisense (CCCCGG) transcripts, and (3) proteotoxicity from dipeptide repeat (DPR) proteins produced from the expanded nucleotide repeats3. This study focuses on the molecular mechanisms underlying DPR toxicity. Strong evidence suggests that DPRs are toxic in both cell culture and animal models of disease4,5. Despite the presence of the expanded G4C2 repeat in the non-coding region of the RNA and SPP1 the absence of an AUG initiating codon, DPRs are translated in all three reading frames from both sense and antisense transcripts through a process called repeat-associated non-AUG (RAN) translation4. Poly-glycine-alanine (poly-GA), poly-glycine-proline (poly-GP), and poly-glycine-arginine (poly-GR) are produced from sense transcripts, whereas poly-proline-arginine (poly-PR), poly-proline-alanine (poly-PA), and poly-GP are generated from antisense transcripts6C8. DPRs are present in neural cells of patients with models for provides a powerful system for the study of molecular and cellular mechanisms underlying neurodegenerative disorders due to: (1) its short lifespan (~3C4 weeks), (2) its compact nervous system (302 neurons in total), (3) the ability to study neuronal morphology and function with single-cell resolution, (4) the fact that most genes have human orthologs, and (5) the ease of genomic engineering and transgenesis, which enables the rapid generation of worms that harbor human gene mutations, permitting in vivo modeling of neurodegenerative disorders. In this study, we initially developed transgenic animals Prostaglandin E1 (PGE1) that carry, under the control of a ubiquitous promoter, 75 copies of the G4C2 repeat flanked by intronic sequences. Compared to controls, these animals produce DPRs (poly-GA, poly-GP, poly-GR), and display evidence of neurodegeneration, as well as locomotor and lifespan defects. A second set of transgenic worms expressing the 75 G4C2 repeats and the flanking intronic sequences exclusively in neurons displayed similar phenotypes, suggesting that DPR production in this cell type is sufficient to cause disease phenotypes. These models provide an opportunity to study in vivo the molecular mechanisms underlying DPR production. Through a candidate approach, we identified a role for the Prostaglandin E1 (PGE1) non-canonical translation initiation factor (ortholog of human did not affect the formation of G4C2 RNA foci, but prominently decreased poly-GA steady-state levels, mildly affected poly-GP levels, and improved locomotor activity and lifespan in both models. Supporting the phylogenetic conservation of our model for genome contains a ortholog does not contain G4C2 repeats. Therefore, to study the molecular mechanisms underlying DPR production from G4C2 repeats, we generated worms that carry a transgene encoding 75 copies of the G4C2 sequence under the control of a ubiquitous (gene (Fig.?1a). To monitor the expression of poly-GA, the most amyloidogenic DPR27,28, a nanoluciferase (nLuc) reporter was placed in the poly-GA reading frame. Hereafter, we will refer to these transgenic animals as (Fig.?1a). In parallel, we generated four control strains: (a) worms carry an identical sequence to the animals, but the upstream translation initiation codon CUG, which is required for translation of poly-GA in vitro20,29C31, is mutated to UAG (Fig.?1a), (b) worms lack the G4C2 repeats and the intronic sequences flanking.