From anti-fungal to potential neurotherapeutic: Posaconazole as an effective inhibitor of cellular TDP-43 pathology
Nathan Kochen, N.; Zafari, S.; Renaud, A.; Schneider, N.; Vunam, N.; Liao, E. E.; Dutton, J. R.; Braun, A. R.; Sachs, J. N.
Show abstract
Recently, we showed that ketoconazole, a known anti-fungal inhibitor of CYP51, stabilized TAR DNA-binding protein 43 (TDP-43) native self-interactions, reduced TDP-43 pathology and rescued TDP-43-induced SREBP2 downregulation. Despite its promising effects, ketoconazole is not viable for repurposing for ALS due to liver toxicity side effects that occur when orally delivered. To address this, we tested the activities of seven additional known azole-based CYP51 inhibitors in order identify a viable alternative to ketoconazole. Using our established TDP-43 mislocalization and aggregation assay in HEK293T cells, we identified posaconazole, an FDA-approved, CNS-penetrant and orally delivered anti-fungal, as the strongest inhibitor of TDP-43 pathology. Posaconazole was able to reduce insoluble TDP-43 and restore SREBP2 levels, outperforming ketoconazole. Mechanism of action (MOA) experiments suggest posaconazole is able to outperform ketoconazole by inducing a significantly stronger activation of autophagy and upregulation of heat shock proteins known to clear TDP-43. Further MOA experiments show that the effects of posaconazole on TDP-43 are dependent on its known ability to lower cellular cholesterol levels. By correlating our experimental results on the eight CYP51 inhibitors tested, we show that predicted affinity towards human CYP51 strongly correlates with the inhibitors ability to lower TDP-43 aggregation and mislocalization. Finally, we tested posaconazole in a low dose sodium arsenite ALS model in iPSC-derived motor neurons, showing that it is efficacious at inhibiting TDP-43 pathology in the nanomolar range. Altogether, these results support the repurposing of posaconazole for ALS/FTD as a means to prevent TDP-43 pathology.
Matching journals
The top 10 journals account for 50% of the predicted probability mass.