Publications in Life Science Alliance by NOMIS researchers

Intronic hexanucleotide repeat expansions (HREs) in C9ORF72 are the most frequent genetic cause of amyotrophic lateral sclerosis, a devastating, incurable motoneuron (MN) disease. The mechanism by which HREs trigger pathogenesis remains elusive. The discovery of repeat-Associated non-ATG (RAN) translation of dipeptide repeat proteins (DPRs) from HREs along with reduced exonic C9ORF72 expression suggests gain of toxic functions (GOFs) through DPRs versus loss of C9ORF72 functions (LOFs). Through multiparametric high-content (HC) live profiling in spinal MNs from induced pluripotent stem cells and comparison to mutant FUS and TDP43, we show that HRE C9ORF72 caused a distinct, later spatiotemporal appearance of mainly proximal axonal organelle motility deficits concomitant to augmented DNA double-strand breaks (DSBs), RNA foci, DPRs, and apoptosis. We show that both GOFs and LOFs were necessary to yield the overall C9ORF72 pathology. Increased RNA foci and DPRs concurred with onset of axon trafficking defects, DSBs, and cell death, although DSB induction itself did not phenocopy C9ORF72 mutants. Interestingly, the majority of LOF-specific DEGs were shared with HRE-mediated GOF DEGs. Finally, C9ORF72 LOF was sufficient albeit to a smaller extent to induce premature distal axonal trafficking deficits and increased DSBs.

FUS (fused in sarcoma) plays a key role in several steps of RNA metabolism, and dominant mutations in this protein are associated with neurodegenerative diseases. Here, we show that FUS is a component of the cellular response to topoisomerase I (TOP1)–induced DNA breakage; relocalising to the nucleolus in response to RNA polymerase II (Pol II) stalling at sites of TOP1-induced DNA breaks. This relocalisation is rapid and dynamic, reversing following the removal of TOP1-induced breaks and coinciding with the recovery of global transcription. Importantly, FUS relocalisation following TOP1-induced DNA breakage is associated with increased FUS binding at sites of RNA polymerase I transcription in ribosomal DNA and reduced FUS binding at sites of RNA Pol II transcription, suggesting that FUS relocates from sites of stalled RNA Pol II either to regulate pre-mRNA processing during transcriptional stress or to modulate ribosomal RNA biogenesis. Importantly, FUS-mutant patient fibroblasts are hypersensitive to TOP1-induced DNA breakage, highlighting the possible relevance of these findings to neurodegeneration.

Frontotemporal dementia and amyotrophic lateral sclerosis patients with C9orf72 mutation show cytoplasmic poly-GR and poly-PR aggregates. Short poly-(Gly-Arg) and poly-(Pro-Arg) (poly-GR/PR) repeats localizing to the nucleolus are toxic in various model systems, but no interactors have been validated in patients. Here, the neuronal interactomes of cytoplasmic GFP-(GR)149 and nucleolar (PR)175-GFP revealed overlapping RNA-binding proteins, including components of stress granules, nucleoli, and ribosomes. Over-expressing the poly-GR/PR interactors STAU1/2 and YBX1 caused cytoplasmic aggregation of poly-GR/PR in large stress granule–like structures, whereas NPM1 recruited poly-GR into the nucleolus. Poly-PR expression reduced ribosome levels and translation consistent with reduction of synaptic proteins detected by proteomics. Surprisingly, truncated GFP-(GR)53, but not GFP-(GR)149, localized to the nucleolus and reduced ribosome levels and translation similar to poly-PR, suggesting that impaired ribosome biogenesis may be driving the acute toxicity observed in vitro. In patients, only ribosomes and STAU2 co-aggregated with poly-GR/PR. Partial sequestration of ribosomes may chronically impair protein synthesis even in the absence of nucleolar localization and contribute to pathogenesis.