Rapamycin reverts TDP-43 splicing defects and oxidative stress-induced alterations in a human in vitro model of TDP-43 proteinopathy

Aggregates of phosphorylated and ubiquitinated TDP-43 protein in the cytoplasm of neurons are an ALS neuropathological hallmark. Response to stress and formation of stress granules (SG) have been proposed as possible initiators of TDP-43 pathological aggregation. We recently showed that chronic oxidative stress by arsenite induces SG formation in fibroblasts and iPSC-motor neurons from TARDBP and C9orf72 patients. This insult also leads to the formation of phospho-TDP-43 aggregates, which are more abundant in C9orf72 cells and resemble those seen in ALS autoptic brains. Aim of our study was to generate a cell model of TDP-43 pathology for screening drugs able to prevent or reduce TDP-43 pathological inclusions. Given the high variability in the response to stress observed in ALS patients’ cells, we reproduced a chronic oxidative insult in human neuroblastoma SK-N-BE cells by exposure to low doses of arsenite for a time frame ranging from 9 to 24 hours. Our data showed TDP-43 mislocalization from the nucleus to the cytoplasm in a dose- and time-dependent manner and a block in the autophagic flux. Of interest, in this condition we also observed a defective splicing activity of TDP-43 towards selected RNA targets, including UNC13A, STMN2 and POLDIP3. Chronic arsenite treatment is therefore able to reproduce both TDP-43 nuclear loss-of-function and its cytoplasmic mislocalization and aggregation, the two neuropathological hallmarks of TDP-43 proteinopathy. Since autophagy impairment favors TDP-43 pathological aggregation, we tested two autophagy enhancers, Rapamycin and Lithium carbonate. We found that Rapamycin, but not Lithium, was capable of rescuing arsenite-induced loss of TDP-43 splicing activity on RNA targets, of reducing insoluble TDP-43 content and of re-establishing the autophagic flux. We already confirmed the efficacy of Rapamycin on TDP-43 splicing activity in C9orf72 patient-derived fibroblasts exposed to chronic oxidative insult and further studies are now in progress to validate findings also in C9orf72 iPSC-motor neurons. In conclusion, we have set up an experimental in vitro model of TDP-43 pathology in which Rapamycin proved to be beneficial, thus supporting the rationale for targeting autophagy in clinical trials. Moreover, this in vitro model can be exploited as a valuable platform for future drug screening approaches.