Stress-induced phospho-ubiquitin formation causes parkin degradation

Lyudmila Kovalchuke; Eugene V. Mosharov; Oren A. Levy; Lloyd A. Greene

Next, we examined whether parkin's association with phospho-Ub is also central to its loss from the mitochondrial depolarizing agent CCCP. Mitochondrial depolarization has been shown to cause parkin loss in a PINK1-dependent manner [40] . As anticipated, CCCP led to a significant loss of wild-type overexpressed parkin after both 6 and 12 hours of treatment (33.5 ± 2.7% and 62.8 ± 1.7% parkin loss after 6 (N = 6) and 12 (N = 4) hours, respectively, p = 0.001 at both time points) (Fig. 5A,B) . By contrast, both H302A and K151E parkin mutants were fully The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. It . https://doi.org/10.1101/484857 doi: bioRxiv preprint 0.84 vs. no treatment) (Fig. 5A ). By 12 hours, K151E remained fully protected (100.7 ± 3.7% parkin remaining, N = 3, p = 0.87 vs. no treatment), but H302A showed a loss of 26.2 ± 4.4% (N = 4, p = 0.02 vs. no treatment) (Fig. 5B) . A likely explanation for the discrepancy between the behaviors of the K151E and H302A parkin mutants at 12 hours is that H302A has been shown to have greater residual phospho-Ub binding than K151E [47] . As with the L-DOPA and hydrogen peroxide models, the S65A parkin mutant was not at all protected from CCCP compared to wildtype parkin (65.9 ± 1.5% S65A (N = 5) vs. 66.5 ± 2.7% WT (N = 6) remaining at 6 hours, p = 0.9; 36.9 ± 3.2% S65A (N = 4) vs. 37.2 ± 1.7% WT (N = 4) remaining at 12 hours, p = 0.95), and we did not observe any protection of the H302A/S65A double mutant over the H302A single mutant after 12 hours of CCCP treatment (73.8 ± 4.4% H302A (N = 4) vs. 77.0 ± 0.4% H302A/S65A (N= 3) remaining, p = 0.75) (Fig. 5A,B) . Altogether, these data indicate that phospho-Ub generated by diverse stressors induces parkin degradation.