Aging and diet alter the protein ubiquitylation landscape in the mouse brain

Abstract

Post-translational modifications (PTMs) regulate protein homeostasis, but how aging impacts PTMs remains unclear. Here, we used mass spectrometry to reveal changes in hundreds of protein ubiquitylation, acetylation, and phosphorylation sites in the mouse aging brain. We show that aging has a major impact on protein ubiquitylation. 29% of the quantified ubiquitylation sites were affected independently of protein abundance, indicating altered PTM stoichiometry. Using iPSC-derived neurons, we estimated that 35% of ubiquitylation changes observed in the aged brain can be attributed to reduced proteasome activity. Finally, we tested whether protein ubiquitylation in the brain can be influenced by dietary intervention. We found that one cycle of dietary restriction and re-feeding modifies the brain ubiquitylome, rescuing some but exacerbating other ubiquitylation changes observed in old brains. Our findings reveal an age-dependent ubiquitylation signature modifiable by dietary intervention, providing insights into mechanisms of protein homeostasis impairment and highlighting potential biomarkers of brain aging.

Fig. 1. Landscape of protein post-translational modifications in the mouse aging brain.

A Multi-omics approach scheme used to characterize mouse brain aging (3 or 4 vs. 33 months old, N = 5, biological replicates, males, C57BL/6J). Created in BioRender. Ori, A. (2025) https://BioRender.com/2947h3v. B Percentage of significantly affected transcripts, proteins, or PTMs (Adj.P < 0.05 for proteome and transcriptome, Q < 0.05 for PTMs; absolute log2 (FC) > 0.58). C PCA based on ubiquitylated peptide abundances from mouse brains. Ellipses represent 95% confidence intervals. The percentage of variance explained by each principal component is indicated. D Volcano plot for ubiquitin enrichment in mouse brain aging (N = 5, biological replicates, P values from Spectronaut differential abundance analysis). E Gene set enrichment analysis (GSEA) for ubiquitylated peptides affected by aging based on GO cellular component terms (Adj.P < 0.05, weighted Kolmogorov–Smirnov test). F Scatterplot illustrating the relationship between fold changes in protein abundance and corresponding ubiquitylated peptide levels with age. Proteins showing significant age-related changes at both the total protein (Adj. P < 0.05 from limma’s empirical Bayes moderated t-test) and ubiquitylated peptide level (Q < 0.05 from Spectronaut differential abundance analysis) are highlighted in orange. Proteins with significant changes in ubiquitylation only (Adj. P ≥ 0.05; Q < 0.01) are shown in purple. Two-sided Pearson’s correlation test. G, H Examples of proteins showing age-related changes of ubiquitylation (N = 5, biological replicates, Q values from Spectronaut for ubiquitylation and Adj.P from limma’s empirical Bayes moderated t-test for proteome, data shown as averages ± SD). I Correlation between ubiquitylation changes and protein half-life (upper panel) or changes in protein half-life during mouse brain aging (bottom panel) (Wilcoxon test). Boxplots show the median, 25th and 75th percentiles (box bounds), and whiskers extending to 1.5 times the Interquartile range; outliers not shown. *Q or Adj.P ≤ 0.05; **Q or Adj.P ≤ 0.01, ***Q or Adj.P ≤ 0.001, ****Q or Adj.P ≤ 0.0001. Source data are provided as a Source Data file. Specific P values are listed in Supplementary Data 8. Related to Figs. S1–S4, and Supplementary Data 1 and 2.

Fig. 2. Conservation of age-related ubiquitylation changes in killifish and mouse.

A Mapping of ubiquitylation sites between M. musculus and N. furzeri. The percentage of consistently up (red and light red for P < 0.05 or P < 0.25, respectively, in both datasets), down (blue and light blue for P < 0.05 or P < 0.25, respectively, on both datasets), and not consistently (gray; P < 0.25) regulated sites are shown in the pie plot (N = 5 for mouse; N = 4 for young, and N = 3 for old killifish; biological replicates). B Barplot showing correlation between age-related changes in ubiquitylated peptides (purple bars) and protein abundance (gray bars) in mouse and killifish. Two-sided Pearson’s correlation test. On the y-axis, different P thresholds are applied. C Heatmap of age-affected ubiquitylated sites conserved in mouse and killifish. Lysine position refers to the mouse protein sequence. Fold changes have been corrected for protein abundance (P < 0.05 in at least one species). D–G Examples of peptides that show age-related changes of ubiquitylation independently of protein abundance, both in mouse and killifish. The flanking sequence (FS) of the modified peptide residues is indicated under the protein name (N = 5 for mouse, N = 4 for young, and N = 3 for old killifish, biological replicates. Q values from Spectronaut differential abundance analysis, data shown as averages ± SD). *Q/Adj.P ≤ 0.05; **Q/Adj.P ≤ 0.01, ***Q/Adj.P ≤ 0.001, ****Q/Adj.P ≤ 0.0001. Source data are provided as a Source Data file. Related to Fig. S5 and Supplementary Data 3.

Fig. 3. Impact of proteasome inhibition and lysosome acidification impairment on protein ubiquitylation in iNeurons.

A Experimental scheme for iNeurons drug treatment and characterization (N = 4, biological replicates from two independent differentiation batches). Created in BioRender. Ori, A. (2025) https://BioRender.com/2947h3v. B PCA based on proteome data from iNeurons. The ellipses (drawn manually) highlight the differentiation between day groups. The percentage of variance explained by each principal component is indicated. C PCA based on ubiquitylated peptide abundances from iNeurons. The ellipses highlight the drug treatments. The percentage of variance explained by each principal component is indicated. D Barplots showing the number of identified ubiquitylated sites in different sample groups (N = 4 except N = 3 for 14 days DMSO and 7 days bortezomib, biological replicates, data shown as averages ± SD). Gene set enrichment analysis (GSEA) for ubiquitylated peptides affected by bortezomib (E) or bafilomycin (F) in 14-day iNeurons based on GO cellular component terms (Adj.P < 0.05, weighted Kolmogorov–Smirnov test). G Correlation between changes of ubiquitylation observed during mouse brain aging and 14-day iNeurons treated with bortezomib or bafilomycin (P < 0.05 in both datasets, two-sided Pearson’s correlation test). H Scatterplot comparing ubiquitylation changes observed in mouse brain aging and 14-day iNeurons treated with bortezomib (upper panel) or bafilomycin (lower panel) (P < 0.05 in both datasets, two-sided Pearson’s correlation test). I Heatmap of ubiquitylation changes occurring in neurodegeneration-associated proteins. Sites significant (P < 0.05, Spectronaut differential abundance test) in at least one of the compared datasets are shown. The lysine numbering refers to the mouse protein sequence. In this figure, mouse ubiquitylation changes were corrected for protein abundance changes. *P ≤ 0.05; **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001. Source data are provided as a Source Data file. Specific P values are listed in Supplementary Data 8. Related to Figs. S6–S8, and Supplementary Data 4 and 5.

Fig. 4. Quantification of ubiquitin-chain linkages in mouse brain and iNeurons.

A Experimental scheme for absolute quantification of ubiquitin chains and the total ubiquitin pool (N = 5 for mouse, N = 3 for iNeurons, biological replicates). Created in BioRender. Ori, A. (2025) https://BioRender.com/2947h3v. B Percentage of total and polyubiquitin in mouse brain aging and in iNeurons upon drug treatments measured via AQUA-PRM, * refers to unpaired t-test, data shown as averages ± SD. C Barplot of total ubiquitin, linear (M1), and branched lysine chains absolute quantification in young and old mice brain and in iNeurons (N = 5 for mouse, N = 3 for iNeurons, biological replicates, * refers to unpaired t-test, One-way ANOVA results are indicated above the barplots, data shown as averages ± SD. *P ≤ 0.05; **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001. Source data are provided as a Source Data file. Specific P values are listed in Supplementary Data 8. Related to Supplementary Data 6.

Fig. 5. Impact of dietary intervention on the brain proteome and ubiquitylome of old mice.

A Scheme of dietary intervention applied to old mice (N = 4 for ad libitum fed mice, N = 6 for re-fed mice, biological replicates, males, C57BL/6J, 26 months old, data shown as averages ± SD). Created in BioRender. Ori, A. (2025) https://BioRender.com/2947h3v. B PCA based on proteome and ubiquitylome data from ad libitum (AL) and re-fed mice (RF). Ellipses represent 95% confidence intervals. The percentage of variance explained by each principal component is indicated. C Percentage of significantly affected proteins, or ubiquitylated peptides (Q < 0.05; absolute log2 fold change (FC) > 0.58). D Volcano plot for protein abundance and ubiquitylated peptide changes in RF vs. AL mice (N = 4 for AL mice, N = 6 for RF mice, biological replicates). E Venn diagram displaying overlap between ubiquitylation significantly (Q < 0.05) affected by aging and dietary intervention (508 sites not significant in both datasets are not displayed, Hypergeometric test). F Scatterplot illustrating the relationship between ubiquitylated peptide changes observed with aging and those following dietary intervention in old mice. Only peptides with significant changes (Q < 0.05) in at least one dataset are shown. Peptides with congruent and significant changes in both datasets (Q < 0.05) are labeled as “Exacerbated” (dark gray). “Reverted-down” (yellow) denotes peptides that significantly decreased upon dietary intervention (Q < 0.05 in RF data) in the opposite direction to the age-related increase. “Reverted-up” (purple) indicates peptides that significantly increased upon intervention in the opposite direction to the age-related decrease. Peptides with consistent directional changes but significant in only one dataset are shown in light gray. Two-sided Pearson’s correlation test. G Heatmap comparing the normalized enrichment scores (NES) from gene set enrichment analyses (GSEA) performed on age-related and diet-induced changes of ubiquitylation. GO terms are sorted by their scaled similarity (1/Euclidean distance between NES), multiplied by the average −log10 (Adj.P) across the two datasets, weighted Kolmogorov–Smirnov test. H Heatmap highlighting age-related changes in ubiquitylation that are reverted by dietary intervention (Q < 0.05 in response to dietary intervention and of opposite log2 (FC) in response to aging). Source data are provided as a Source Data file. Related to Fig. S9 and Supplementary Data 7.