In vitro investigation of dipeptide repeat proteins in C9ORF72-associated motor neuron disease and frontotemporal dementia

© 2020 Mona Radwan Ahmed ElsayedMutations that expand a hexanucleotide (GGGGCC) tandem repeat sequence (HTRS) in intron-1 of the C9ORF72 gene is the major cause of familial motor neuron disease MND and frontotemporal dementia (FTD). The expanded HTRSs trigger abnormal bi-directional transcription, which generates both sense and antisense RNA transcripts that form foci. The subsequent translation of all reading frames of HTRSs generates five types of dipeptide repeat proteins (DRPs) via Repeat-Associated Non-ATG (RAN) translation. The sense (G4C2) RNA encodes poly-glycine-alanine (poly-GA), poly-glycine-proline (poly-GP), and poly-glycine-arginine (poly-GR), while the antisense (C4G2) RNA generates poly-GP, poly-proline-arginine (poly-PR), and poly-alanine-proline (poly-AP). The presumably toxic RAN proteins aggregate in neurons suggesting that they play a crucial role(s) in NMD and FTD pathogenesis. The mechanism underlying the DPR protein-induced neuronal toxicity have not been fully elucidated. The current thesis has employed a quantitative proteomics approach aimed at understanding how these DRPs, as putative “junk” proteins, confer damage to the cellular biochemistry via first identifying the protein interactome of each DRP and secondly by determining the impact of these DRP sequences on the global proteome. To this end, we have developed artificial cDNA constructs engineered by a randomized codon strategy to express short (10x) and long (101x) DPR proteins avoiding GC repeat sequences. When transiently expressed in mouse neuroblastoma cells, the soluble poly-PA and poly-GP were well tolerated at the lengths tested. Clear length-dependent neurotoxicity was seen for poly-GA that formed cytoplasmic aggregates only at the long length. When compared to 97Q-expanded huntingtin exon 1 aggregates, GA101 exhibited faster aggregation kinetics and stronger proteasomal recruitment as revealed in the aggregate proteome. Conversely, the arginine-rich DPRs (Poly-PR and Poly-GR) formed discrete nuclear aggregates and were profoundly toxic at both short and long repeat lengths. Data presented within this thesis has also provided evidence that the arginine valency in poly-PR and poly-GR triggers their pernicious binding to the proteome, compared to the sparse binding of the more inert poly-AP and poly-GA. The poly-PR and poly-GR interactome included proteins involved in ribosome biogenesis, translation and RNA splicing machinery. Using stalling reporters, long R-rich DPRs stalled translation, through a putative electrostatic jamming mechanism. These findings also indicate that Arg-rich dipeptide polymers are mimetics of endogenous arginine methylase substrates and recruit these methylase proteins. A consequence is hypomethylation of the proteome, including key proteins known to be involved in ALS mechanisms. The mechanism explains the previously reported abnormalities in arginine methylation patterns in the brains of ALS patients. Finally, this work has unravelled how Arg-rich dipeptide polymers impair the assembly of the actin cytoskeleton. This finding provides important mechanistic insight into the previously reported actin polymerization defects in multiple forms of amyotrophic lateral sclerosis