When human TDP-43 is overexpressed in yeast it is toxic and forms cytoplasmic aggregates. The mechanism of this toxicity is unknown. Genetic screens for TDP-43 toxicity modifiers in the yeast system previously identified proteins, including PBP1, that enhance TDP-43 toxicity. The determination in yeast that deletion of PBP1 reduces TDP-43 toxicity while overexpression enhances toxicity, led to the discovery that its human homolog, ATXN2, is associated with ALS risk. Thus, the yeast system has relevance to human disease. We now show that deletion of a new yeast gene, tip41Δ, likewise suppresses TDP-43 toxicity. We also found that TDP-43 overexpression and toxicity is associated with reduced autophagy. This is consistent with findings in other systems that increasing autophagy reduces TDP-43 toxicity and is in contrast to a report of enhanced autophagy when TDP-43 was overexpressed in yeast. Interestingly, we found that deletions of PBP1 and TIP41, which reduced TDP-43 toxicity, eliminated TDP-43′s inhibition of autophagy. This suggests that toxicity of TDP-43 expressed in yeast is in part due to its inhibition of autophagy and that deletions of PBP1 and TIP41 may reduce TDP-43 toxicity by preventing TDP-43 from inhibiting autophagy.
TDP-43 forms aggregates in the neurons of patients with several neurodegenerative diseases. Human TDP-43 also aggregates and is toxic in yeast. Here, we used a yeast model to investigate (1) the nature of TDP-43 aggregates and (2) the mechanism of TDP-43 toxicity. Thioflavin T, which stains amyloid but not wild-type TDP-43 aggregates, also did not stain mutant TDP-43 aggregates made from TDP-43 with intragenic mutations that increase or decrease its toxicity. However, 1,6-hexanediol, which dissolves liquid droplets, dissolved wild-type or mutant TDP-43 aggregates. To investigate the mechanism of TDP-43 toxicity, the effects of TDP-43 mutations on the autophagy of the GFP-ATG8 reporter were examined. Mutations in TDP-43 that enhance its toxicity, but not mutations that reduce its toxicity, caused a larger reduction in autophagy. TOROID formation, which enhances autophagy, was scored as GFP-TOR1 aggregation. TDP-43 inhibited TOROID formation. TORC1 bound to both toxic and non-toxic TDP-43, and to TDP-43, with reduced toxicity due to pbp1Δ. However, extragenic modifiers and TDP-43 mutants that reduced TDP-43 toxicity, but not TDP-43 mutants that enhanced toxicity, restored TOROID formation. This is consistent with the hypothesis that TDP-43 is toxic in yeast because it reduces TOROID formation, causing the inhibition of autophagy. Whether TDP-43 exerts a similar effect in higher cells remains to be determined.