Ubp2 modulates DJ-1-mediated redox-dependent mitochondrial dynamics in Saccharomyces cerevisiae

Abstract

Mitochondrial integrity is a crucial determinant of overall cellular health. Mitochondrial dysfunction and impediments in regulating organellar homeostasis contribute majorly to the pathophysiological manifestation of several neurological disorders. Mutations in human DJ-1 (PARK7) have been implicated in the deregulation of mitochondrial homeostasis, a critical cellular etiology observed in Parkinson's disease progression. DJ-1 is a multifunctional protein belonging to the DJ-1/ThiJ/PfpI superfamily, conserved across the phylogeny. Although the pathophysiological significance of DJ-1 has been well-established, the underlying molecular mechanism(s) by which DJ-1 paralogs modulate mitochondrial maintenance and other cellular processes remains elusive. Using Saccharomyces cerevisiae as the model organism, we unravel the intricate mechanism by which yeast DJ-1 paralogs (collectively called Hsp31 paralogs) modulate mitochondrial homeostasis. Our study establishes a genetic synthetic interaction between Ubp2, a cysteine-dependent deubiquitinase, and DJ-1 paralogs. In the absence of DJ-1 paralogs, mitochondria adapt to a highly tubular network due to enhanced expression of Fzo1. Intriguingly, the loss of Ubp2 restores the mitochondrial integrity in the DJ-1 deletion background by modulating the ubiquitination status of Fzo1. Besides, the loss of Ubp2 in the absence of DJ-1 restores mitochondrial respiration and functionality by regulating the mitophagic flux. Further, Ubp2 deletion makes cells resistant to oxidative stress without DJ-1 paralogs. For the first time, our study deciphers functional crosstalk between Ubp2 and DJ-1 in regulating mitochondrial homeostasis and cellular health.

Fig 1. Ubp2 deletion restores the respiratory growth defect of Δ31Δ34 strain.

(A) Growth phenotype assessment. The indicated yeast strains were allowed to grow up to mid-log phase in SC dextrose broth at 30°C. Ten-fold serially diluted cells were spotted on the indicated media and incubated at permissive temperature (30°C) and non permissive temperature (24°C and 37°C). Images were captured at 36h and 72h for dextrose and glycerol, respectively. (B) Assessment of the Ubp2-HA levels in the absence of Hsp31 paralogs. Whole-cell lysate of the indicated strains was subjected to immunoblotting and probed for Ubp2-HA levels in late log phase(12h). (C) Densitometric analysis for probing the difference in expression of Ubp2-HA levels. (D) Growth phenotype assessment upon complementation with Ubp2. The indicated strains were allowed to grow up to the mid-log phase in SC Leu- media and subjected to grow at permissive and non-permissive temperatures in dextrose and glycerol. Images were captured for dextrose and glycerol after 36h and 72h, respectively. Unpaired Student t-test was performed for statistical analysis. Error bars represent the standard deviation in median values from 3 biological replicates. Asterisks indicate the p-value, *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001.

Fig 2. Ubp2 deletion reverts the defect in mitochondrial integrity in the absence of Hsp31 paralogs.

(A) Assessment of mitochondrial integrity. Yeast strains expressing MTS-mCherry grown in SC Leu^- dextrose till the mid-log phase were subjected to microscopic analysis to visualize mitochondria. Scale bar (10 μm). The three different mitochondrial morphologies scored have been represented (Fragmented, Intermediate, Tubular cells). (B) Quantification of cells exhibiting specific mitochondrial morphology. (C) Quantification of total mitochondrial mass by flow cytometric analysis. Cells grown until the mid-log phase were subjected to NAO staining and quantified using the BD FACS Verse instrument. (D) Evaluation of the functional mitochondrial mass by flow cytometry. Cells grown until the mid-log phase were subjected to JC-1 staining, a potentiometric dye, and acquired using BD FACS Verse. (E) Measurement of ATP levels. Mitochondria were isolated from the indicated strains, and ATP was measured using a fluorescence-based assay. One-way ANOVA with Tukey’s multiple comparison test was used for significance analysis. Error bars represent the standard deviation in median values from 3 biological replicates. Asterisks indicate the p-value, *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001.

Fig 3. Role of Ubp2 in regulating mitophagy.

(A, B) Mitophagy induction by Western analysis. Cells expressing OM45-GFP from the indicated strains were cultured in glycerol for the mentioned time points, and the lysates were subjected to Western blotting and quantified using Two-Way ANOVA for significance analysis.Error bars represent the standard deviation in median values from 3 biological replicates. Asterisks indicate the p-value, *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001 (B). ∆atg32 was used as a positive control to compare the defects in mitophagy. (C) Microscopic analysis of mitophagy induction. Cells expressing OM45-GFP were grown in YP-glycerol for 48h and subjected to fluorescence microscopy. The vacuole is stained with FM4-64 dye. Scale bars (10 µm). Images were zoomed up to 2X and represented.

Fig 4. Fzo1 levels are restored upon deletion of Ubp2 in the absence of Hsp31 paralogs.

(A, B) Evaluation of the relative steady-state levels of Fzo1 by western blotting. The cells from indicated strains lysed, and Fzo1 levels were probed in the whole cell lysate (A). The blots were subjected to quantification by densitometry, and the fold change in Fzo1 expression was represented. (C, D) Assessment of the turnover kinetics of Fzo1 using cycloheximide assay. The cells were subjected to treatment with cycloheximide for the indicated time points, and the lysates were subjected to western blotting (C). The blots were subjected to quantification densitometrically and represented the rate at which Fzo1 is degraded across the strains (D). Two-way ANOVA was performed using three biological replicates. (E, F) Evaluation of ubiquitination status of Fzo1. Genomically tagged Fzo1 at the C-terminus with HA-tag was subjected to pull-down analysis using HA-conjugated beads and probed with anti-HA and anti-ubiquitin specific antibodies (E). The Western blots were subjected to quantitation by densitometry, and fold change in the ubiquitination levels is represented (F). One-way ANOVA with Tukey’s multiple comparison tests was used for significance analysis. Error bars represent the standard deviation in median values from 3 biological replicates. Asterisks indicate the p-value, *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001.

Fig 5. Cell cycle is restored upon deletion of Ubp2 cells lacking Hsp31 paralogs.

(A) Cell cycle analysis. The restoration of cell cycle was analysed using PI staining after synchronizing the strains using alpha factor. Release after G1-arrest was compared at the indicated time points. (B,C) Assessment of cell morphology and size. The cell morphology and size in the indicated strains were analysed by microscopy (B). Scale bar (15 µm). The relative differences in cell size in the yeast strains (WT, Δ31Δ34, and Δ31Δ34∆ubp2 cells) were quantified using ImageJ (C). The values were plotted using GraphPad Prism 5.0. One-way ANOVA with Tukey’s multiple comparison test was used for significance analysis. Error bars represent the standard deviation in median values from 3 biological replicates. Asterisks indicate the p-value, *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001.

Fig 6. Basal ROS levels are restored upon restoration of mitochondrial dynamics.

(A,B) Evaluation of basal ROS levels by H₂DCFDA. Total ROS levels were measured in indicated strains by staining the cells with H₂DCFDA and subjected to flow cytometric analysis (A). Similarly, the cells from indicated strains were stained with H₂DCFDA and subjected to microscopic analysis (B). Scale bar (15 µm). (C,D) Evaluation of basal mitochondrial ROS. The mitochondrial ROS lelevls in indicated strains were determined by MitoSOX dye using flow cytometric analysis (C). Similarly, the mitochondrial ROS lelevls were probed by MitoSOX dye using microscopic analysis (D). Scale bar (10 µm). One-way ANOVA with Tukey’s multiple comparison test was used for significance analysis. Error bars represent the standard deviation in median values from 3 biological replicates. Asterisks indicate the p-value, *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001.

Fig 7. Loss of Ubp2 provides partial oxidative stress resistance in the absence of Hsp31 paralogs.

(A) Growth phenotype assessment. Yeast strains were grown till mid-log phase and were subjected to H₂O₂ stress. (B,C) Evaluation of Glutathione levels using flow by microscopy and cytometry. The GSH levels were determined by monochlorobiamine staining and subjected to microscopy (B), Scale bar (25 µm). The normalised GSH levels were probed using flow cytometry (C) by monochlorobiamine staining. (D, E) Assessment of relative GSH/GSSG ratio in the mitochondrial fraction (D) and cytoplasmic fraction (E) of the indicated strains, using Glutathione assay kit. One-way ANOVA with Tukey’s multiple comparison test was used for significance analysis. Error bars represent the standard deviation in median values from 3 biological replicates. Asterisks indicate the p-value, *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001.

Fig 8. Model depicting the role of Ubp2 in regulating mitochondrial dynamics in cells lacking Hsp31 paralogs.

(1) In healthy cells (left panel), DJ-1 homologs maintain optimum redox homeostasis by regulating balanced fusion and fission events. (2) The fusion-fission balance is crucial for maintaining healthy mitochondrial mass and function. (3) The ubiquitination status of Fzo1 dictates either fusion or degradation. Ubp2 mediates fusion by editing the ubiquitin chain on Fzo1, promoting mitochondrial fusion. Ubp12 opposes the function of Ubp2 by antagonizing fusion. (4) In cells lacking DJ-1 paralogs (right panel), the redox status is perturbed due to increased basal ROS levels. (5) The elevated ROS is due to an increase in the functional hyperfused mitochondrial structures, which in turn is used by the cells as an adaptive strategy. (6) The imbalance in fission-fusion dynamics in cells lacking DJ-1 paralogs is attributed to increased Fzo1 levels. The ubiquitination status of Fzo1 is altered due to increased expression of Ubp2, leading to enhanced fusion events and hence perturbation in the mitochondrial dynamics. The model was made using modified mitochondria icons from https://bioicons.com/ (simple_mitochondria_network icon by Marnie-Maddock https://github.com/MarnieMaddock is licensed under CC-BY 4.0 Unported https://creativecommons.org/licenses/by/4.0/);(mitochondria-grey icon by DBCLS https://togotv.dbcls.jp/en/pics.html is licensed under CC-BY 4.0 Unported https://creativecommons.org/licenses/by/4.0/), (mitochondrium-yellow icon by Servier https://smart.servier.com/ is licensed under CC-BY 3.0 Unported https://creativecommons.org/licenses/by/3.0/) and the rest of the figure was drawn.