Thioridazine reverts the phenotype in cellular and Drosophila models of amyotrophic lateral sclerosis by enhancing TDP-43 aggregate clearance.

Brain inclusions mainly composed of misfolded and aggregated TAR DNA binding protein 43 (TDP-43), are characteristic hallmarks of amyotrophic lateral sclerosis (ALS). Irrespective of the role played by the inclusions, their reduction represents an important therapeutic pathway that is worth exploring. Their removal can either lead to the recovery of TDP-43 function by removing the self-templating conformers that sequester the protein in the inclusions, and/or eliminate any potential intrinsic toxicity of the aggregates. The search for curative therapies has been hampered by the lack of ALS models for use in high-throughput screening. We adapted, optimised, and extensively characterised our previous ALS cellular model for such use. The model demonstrated efficient aggregation of endogenous TDP-43, and concomitant loss of its splicing regulation function. We provided a proof-of-principle for its eventual use in high-throughput screening using compounds of the tricyclic family and showed that recovery of TDP-43 function can be achieved by the enhanced removal of TDP-43 aggregates by these compounds. We observed that the degradation of the aggregates occurs independent of the autophagy pathway beyond autophagosome-lysosome fusion, but requires a functional proteasome pathway. The in vivo translational effect of the cellular model was tested with two of these compounds in a Drosophila model expressing a construct analogous to the cellular model, where thioridazine significantly improved the locomotive defect. Our findings have important implications as thioridazine cleared TDP-43 aggregates and recovered TDP-43 functionality. This study also highlights the importance of a two-stage, in vitro and in vivo model system to cross-check the search for small molecules that can clear TDP-43 aggregates in TDP-43 proteinopathies.

A feedback loop between dipeptide-repeat protein, TDP-43 and karyopherin-α mediates C9orf72-related neurodegeneration.

Accumulation and aggregation of TDP-43 is a major pathological hallmark of amyotrophic lateral sclerosis and frontotemporal dementia. TDP-43 inclusions also characterize patients with GGGGCC (G4C2) hexanucleotide repeat expansion in C9orf72 that causes the most common genetic form of amyotrophic lateral sclerosis and frontotemporal dementia (C9ALS/FTD). Functional studies in cell and animal models have identified pathogenic mechanisms including repeat-induced RNA toxicity and accumulation of G4C2-derived dipeptide-repeat proteins. The role of TDP-43 dysfunction in C9ALS/FTD, however, remains elusive. We found G4C2-derived dipeptide-repeat protein but not G4C2-RNA accumulation caused TDP-43 proteinopathy that triggered onset and progression of disease in Drosophila models of C9ALS/FTD. Timing and extent of TDP-43 dysfunction was dependent on levels and identity of dipeptide-repeat proteins produced, with poly-GR causing early and poly-GA/poly-GP causing late onset of disease. Accumulating cytosolic, but not insoluble aggregated TDP-43 caused karyopherin-α2/4 (KPNA2/4) pathology, increased levels of dipeptide-repeat proteins and enhanced G4C2-related toxicity. Comparable KPNA4 pathology was observed in both sporadic frontotemporal dementia and C9ALS/FTD patient brains characterized by its nuclear depletion and cytosolic accumulation, irrespective of TDP-43 or dipeptide-repeat protein aggregates. These findings identify a vicious feedback cycle for dipeptide-repeat protein-mediated TDP-43 and subsequent KPNA pathology, which becomes self-sufficient of the initiating trigger and causes C9-related neurodegeneration.