Turnover atlas of proteome and phosphoproteome across mouse tissues and brain regions.

Turnover Atlas of Proteome and Phosphoproteome Across Mouse Tissues and Brain Regions - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Cell . Author manuscript; available in PMC: 2026 Apr 17. Published in final edited form as: Cell. 2025 Mar 20;188(8):2267–2287.e21. doi: 10.1016/j.cell.2025.02.021 Search in PMC Search in PubMed View in NLM Catalog Add to search Turnover Atlas of Proteome and Phosphoproteome Across Mouse Tissues and Brain Regions Wenxue Li Wenxue Li 1 Department of Pharmacology, Yale University School of Medicine, New Haven, CT 06520, USA. 2 Cancer Biology Institute, Yale University School of Medicine, West Haven, CT 06516, USA. 11 These authors contributed equally. Find articles by Wenxue Li 1, 2, 11 , Abhijit Dasgupta Abhijit Dasgupta 3 Departments of Structural Biology and Developmental Neurobiology, St. Jude Children’s Research Hospital, Memphis, TN 38105, USA. 7 Present address: Department of Computer Science and Engineering, SRM University AP, Neerukonda, Guntur, Andhra Pradesh 522240, India. 11 These authors contributed equally. Find articles by Abhijit Dasgupta 3, 7, 11 , Ka Yang Ka Yang 3 Departments of Structural Biology and Developmental Neurobiology, St. Jude Children’s Research Hospital, Memphis, TN 38105, USA. 8 Present address: Department of Cell Biology, Harvard Medical School, Boston, MA 02115, USA 11 These authors contributed equally. Find articles by Ka Yang 3, 8, 11 , Shisheng Wang Shisheng Wang 4 Department of General Surgery and Liver Transplant Center, Proteomics-Metabolomics Analysis Platform, West China Hospital, Sichuan University, Chengdu 610041, China 11 These authors contributed equally. Find articles by Shisheng Wang 4, 11 , Nisha Hemandhar-Kumar Nisha Hemandhar-Kumar 5 Department of Neuro- and Sensory Physiology, University Medical Center Göttingen, 37073 Göttingen, Germany. Find articles by Nisha Hemandhar-Kumar 5 , Surendhar R Chepyala Surendhar R Chepyala 3 Departments of Structural Biology and Developmental Neurobiology, St. Jude Children’s Research Hospital, Memphis, TN 38105, USA. Find articles by Surendhar R Chepyala 3 , Jay M Yarbro Jay M Yarbro 3 Departments of Structural Biology and Developmental Neurobiology, St. Jude Children’s Research Hospital, Memphis, TN 38105, USA. Find articles by Jay M Yarbro 3 , Zhenyi Hu Zhenyi Hu 2 Cancer Biology Institute, Yale University School of Medicine, West Haven, CT 06516, USA. 9 Present address: Interdisciplinary Research center on Biology and chemistry, Shanghai institute of Organic chemistry, Chinese Academy of Sciences, Shanghai 201210, China Find articles by Zhenyi Hu 2, 9 , Barbora Salovska Barbora Salovska 1 Department of Pharmacology, Yale University School of Medicine, New Haven, CT 06520, USA. 2 Cancer Biology Institute, Yale University School of Medicine, West Haven, CT 06516, USA. Find articles by Barbora Salovska 1, 2 , Eugenio F Fornasiero Eugenio F Fornasiero 5 Department of Neuro- and Sensory Physiology, University Medical Center Göttingen, 37073 Göttingen, Germany. 6 Department of Life Sciences, University of Trieste, 34127 Trieste, Italy. Find articles by Eugenio F Fornasiero 5, 6, * , Junmin Peng Junmin Peng 3 Departments of Structural Biology and Developmental Neurobiology, St. Jude Children’s Research Hospital, Memphis, TN 38105, USA. Find articles by Junmin Peng 3, ** , Yansheng Liu Yansheng Liu 1 Department of Pharmacology, Yale University School of Medicine, New Haven, CT 06520, USA. 2 Cancer Biology Institute, Yale University School of Medicine, West Haven, CT 06516, USA. 10 Department of Biomedical Informatics & Data Science, Yale University School of Medicine, New Haven, CT 06510, USA. 12 Lead contact. Find articles by Yansheng Liu 1, 2, 10, 12, *** Author information Article notes Copyright and License information 1 Department of Pharmacology, Yale University School of Medicine, New Haven, CT 06520, USA. 2 Cancer Biology Institute, Yale University School of Medicine, West Haven, CT 06516, USA. 3 Departments of Structural Biology and Developmental Neurobiology, St. Jude Children’s Research Hospital, Memphis, TN 38105, USA. 4 Department of General Surgery and Liver Transplant Center, Proteomics-Metabolomics Analysis Platform, West China Hospital, Sichuan University, Chengdu 610041, China 5 Department of Neuro- and Sensory Physiology, University Medical Center Göttingen, 37073 Göttingen, Germany. 6 Department of Life Sciences, University of Trieste, 34127 Trieste, Italy. 7 Present address: Department of Computer Science and Engineering, SRM University AP, Neerukonda, Guntur, Andhra Pradesh 522240, India. 8 Present address: Department of Cell Biology, Harvard Medical School, Boston, MA 02115, USA 9 Present address: Interdisciplinary Research center on Biology and chemistry, Shanghai institute of Organic chemistry, Chinese Academy of Sciences, Shanghai 201210, China 10 Department of Biomedical Informatics & Data Science, Yale University School of Medicine, New Haven, CT 06510, USA. 11 These authors contributed equally. 12 Lead contact. AUTHOR CONTRIBUTIONS W.L. performed DIA-MS measurements, conducted phosphoproteomic experiments, prepared results, methods, figures and tables, and co-led the data analysis. A.D. processed the protein and phosphoproteomic lifetime data and conducted bioinformatic analyses with the help of S.R.C.. K.Y. prepared the proteomic samples and carried out TMT experiments with assistance from J.M.Y. S.W. developed the Tissue-PPT web portal. N.H.K. conducted verification experiments for phosphorylation-associated turnover under the supervision of E.F.F. Z.H. performed the PhosTAC experiments. B.S. contributed to data analysis. J.P., and Y.L. conceptualized the study. E.F.F., J.P., and Y.L. secured funding and co-supervised the entire project. Y.L. wrote the initial manuscript with input from all authors. All authors contributed to the writing of the final manuscript. * Correspondence: efornas@gwdg.de ** Correspondence: junmin.peng@stjude.org *** Correspondence: yansheng.liu@yale.edu Issue date 2025 Apr 17. PMC Copyright notice PMCID: PMC12033170  NIHMSID: NIHMS2066478  PMID: 40118046 The publisher's version of this article is available at Cell Previous version available: This article is based on a previously available preprint posted on bioRxiv on October 17, 2024: " An Extensive Atlas of Proteome and Phosphoproteome Turnover Across Mouse Tissues and Brain Regions ". SUMMARY Understanding how proteins in different mammalian tissues are regulated is central to biology. Protein abundance, turnover, and post-translational modifications like phosphorylation, are key factors that determine tissue-specific proteome properties. However, these properties are challenging to study across tissues and remain poorly understood. Here, we present Turnover-PPT , a comprehensive resource mapping the abundance and lifetime of 11,000 proteins and 40,000 phosphosites in eight mouse tissues and various brain regions, using advanced proteomics and stable isotope labeling. We revealed tissue-specific short- and long-lived proteins, strong correlations between interacting protein lifetimes, and distinct impacts of phosphorylation on protein turnover. Notably, we discovered a remarkable pattern of turnover changes for peroxisome proteins in specific tissues, and that phosphorylation regulates the stability of neurodegeneration-related proteins, such as Tau and α-synuclein. Thus, Turnover-PPT provides fundamental insights into protein stability, tissue dynamic proteotypes, and functional protein phosphorylation, and is accessible via an interactive web-based portal at https://yslproteomics.shinyapps.io/tissuePPT . Keywords: Protein turnover, Protein lifetime, Mouse tissues, Brain regions, Proteomics, Pulse SILAC, Protein phosphorylation, DIA-MS, TMT In brief: This study presents a comprehensive atlas of protein and phosphoprotein turnover across 16 mouse tissues and brain regions, using advanced proteomics and isotope labeling. It reveals tissue-specific protein lifetimes, coordinated turnover within protein-protein interaction networks, peroxisomal turnover diversity, and the regulatory role of phosphorylation in protein stability, offering valuable insights into proteome homeostasis and tissue-specific functions. Graphical Abstract INTRODUCTION Protein turnover is a fundamental process involving the continuous synthesis and degradation of proteins in all living organisms 1 . Protein turnover is critical for maintaining proteostasis, replacing damaged proteins, and ensuring the functional integrity of tissues 2 3 4 5 . Maintaining proper turnover in different tissues is particularly challenging. While housekeeping proteins involved in basic cellular functions must be expressed across tissues, different tissues have dissimilar needs and must respond in different ways to maintain homeostasis. Targeting a protein with rapid synthesis and degradation in a specific tissue may enhance treatment efficacy while minimizing effects on tissues where its turnover is slower. Furthermore, protein turnover is a critical molecular regulatory layer at the post-translational level buffering, tuning, or amplifying variability in proteomic abundance. High turnover rates for proteins with many copies become energetically costly for cells 6 and may signal a critical role for these proteins in tissue-specific functions. Thus, a comprehensive analysis of both protein abundance (PA) and protein lifetime (PT) in mammalian tissues can enhance our understanding of cell- and tissue-specific “economic” principles and provide significant biomedical insights 7 . Protein turnover rates in one or multiple mouse tissues were analyzed in a few previous studies, most of which utilized mass spectrometry (MS)-based shotgun proteomic approaches integrated with pulse stable isotope labeling by amino acids in cells (pSILAC) applied in animals 8 9 7 10 11 12 13 14 . The general underlying assumption is that, once animal growth and development have ceased and a "steady state" has been reached, the rate of protein synthesis equals the rate of protein removal 15 . Thus, the metabolic labeling of proteins with 15 N or stable heavy isotopic amino acids through dietary intake can be monitored to represent the newly synthesized protein molecules, which are then quantified by MS over time, to derive protein-specific turnover rates. Notably, due to the significant reuse of amino acids in multicellular organisms, specific mathematical modeling had to be developed to robustly determine in vivo protein lifetimes 9 11 16 . Together, these pioneering studies illustrated the turnover diversity among multiple tissues and protein families, but they have limitations. First, they achieved low proteome coverage, compared to deep analysis of protein abundance 17 18 , often due to the limited sensitivity and throughput of the MS methods available at the time. Moreover, they only examined fewer than 4–5 tissue types 9 11 13 , and did not assess the full diversity of protein turnover among multiple regions of a complex organ, such as the brain. Notably, most of these studies have focused solely on protein turnover rates without considering the absolute and relative mRNA and protein abundances among the tissues, prohibiting a systematic understanding of protein turnover regulation. In addition to the whole protein level, post-translational modifications (PTM) such as phosphorylation often determine protein activity 19 20 . The abundance of tens of thousands of phosphosites (P-sites) has been profiled across various mouse tissues 21 17 , the in vivo turnover of phosphorylated proteins in animals has yet to be measured. In cultured cancer cells, we and others have examined the effects of phosphorylation on protein degradation and clearance using pSILAC 15 combined with phosphoproteomic enrichment 22 23 24 25 . Nevertheless, how site-specific phosphorylation regulates in vivo protein lifetime and stability across different tissues and tissue regions remains unexplored. Such knowledge is critical as it may reveal new P-site nodes that can be targeted in human diseases such as neurodegeneration. Here, we harness reproducible proteomic strategies, namely data-independent acquisition (DIA) 26 27 28 and tandem mass tagging (TMTpro) 29 30 to extensively quantify protein and P-site turnover behaviors across eight tissues and nine brain regions in mice. Our datasets feature high coverage and the well-matched nature, where protein abundance and turnover, unmodified proteins and phosphorylated proteins, and tissue-specific profiles across multi-omics layers are closely aligned. Our findings reveal that proteostasis networks can be widely rewired by protein-protein interactions (PPI), organellar localizations, and site-specific phosphorylation such as critical P-sites on Tau and α-synuclein in the brain. Our Resource of tissue p roteome and p hosphoproteome t urnover atlas, or Tissue-PPT, is accessible via an interactive web portal ( https://yslproteomics.shinyapps.io/tissuePPT ). RESULTS In-depth quantitative protein turnover landscape of mouse tissues and brain regions We profiled the proteome-wide protein turnover kinetics in terms of protein half-lives (i.e., T 50 ; hereafter, protein lifetime or PT in short) of the heart, liver, spleen, lung, kidney, gut, plasma, and nine brain regions, including the cerebellum, frontal cortex, substantia nigra, thalamus, amygdala, entorhinal cortex, hippocampus, and olfactory bulb ( Figure 1 ). The PTs were measured in vivo using pSILAC labeling of Lysine-6 containing food fed to mice 31 over periods of 8 and 32 days. This approach was firstly validated using three biological replicates of whole brain tissues and five labeling time points (including days 0, Figure S1A – B ) and then applied to multi-tissue proteomic measurements employing both DIA-MS 26 27 28 and TMTpro 16-plex labeling 29 30 in two biological replicates, generating a total of 1200 MS raw files (Methods). Particularly, to enhance the precision of protein turnover quantification, we implemented the BoxCarmax-DIA multiplexing schema 32 for DIA-MS and performed extensive peptide-level fractionation (>80 fractions) for TMT 33 34 to effectively reduce the MS/MS data complexity for determining heavy-to-light (H/L) ratios. An ordinary differential equation-based (ODE) computational framework was subsequently employed to model the amino acid recycling and fit the Lysine-6 kinetics for DIA-MS and TMT datasets ( Figure 1A ) 16 . DIA-MS and TMT generated highly consistent and reproducible lifetimes across proteins (Pearson R =0.87, Figure 1B , Figure S1C – D ) and across tissues ( R =0.98, Figure 1C ). The protein-specific PT correlation between DIA-MS and TMT among tissues was also high (median R =0.80, Figure 1D ), while TMT overall derived ~15% more protein PT values than DIA-MS ( Figure S1E ). Figure 1. Generation of a high-quality protein turnover atlas across mouse tissues and brain regions. Open in a new tab (A) pSILAC-MS workflow used for cross-tissue protein turnover analysis in mice. BoxCarmax-DIA and TMTpro were employed to improve quantification accuracy. JUMPt with the ODE modeling considering free lysine decay was used for protein half-life calculation. Hippocampus not included for DIA-MS analysis and Tissue-PPT due to low sample yield. (B) Correlation analysis of PTs across the proteome derived by DIA and TMT methods based on the average of all the tissues. rho, Spearman correlation coefficient; R , Pearson correlation coefficient. (C) Correlation analysis of PT within each tissue based on median PT values per tissue quantified by DIA and TMT. (D) Density plot of Pearson R for PT correlation across mouse tissues (11 out of 14) between DIA and TMT. (E) Proteome coverage for both protein identification and protein lifetime profiling. (F) Venn diagram comparing mouse heart proteome coverage between this study and Hasper et al. (G) Spearman correlation of PTs between the two studies for the heart proteome. (H) Comparison of MS intensity and PT between this study and Hasper et al. (red dots). (I) Density plot of PTs across tissues. (J) Violin plots showing MS intensities and PTs of Tau and alpha-synuclein. The red dashed lines indicate the proteome-wide averaged values. By integrating all measurements (see Figure S1D and Methods), we quantified and assembled lifetimes for 11,171 unique protein groups in measured tissues and brain regions. On average, 9214 proteins were detected with their abundances quantified in non-plasma tissues; and turnover rates were measured for an average of 7075 proteins per tissue from Lysine-containing peptides from the same datasets ( Figure 1E ). We developed the Tissue-PPT , a Web-page App to support both individual protein- and protein list-level exploration of this dataset. The analytical depth of Tissue-PPT is comparable to recent large-scale proteome mapping efforts of mouse tissues 17 18 , but it triples the number of protein lifetimes measured across multiple tissues 11 13 14 . For instance, using Lysine-8 SILAC labeling, Rolfs et al. 11 measured protein turnover in five mouse tissue types, including liver (n =2004 proteins) and plasma (n =155). In comparison, Tissue-PPT measured lifetimes for 6077 liver proteins and 516 plasma proteins. Similarly, Hasper et al. 13 , utilizing 15 N labeling, analyzed four tissues, including liver (n =2099) and heart (n =1635), whereas Tissue-PPT determined turnover rates for 6492 proteins in heart. We found that Tissue-PPT covers 84.6% of heart proteins measured by Hasper et al. and reports highly correlated PTs ( R =0.79, Figure 1F – H ). Reassuringly, based on protein DIA-MS signals, Tissue-PPT effectively measured turnover rates for more low-abundance proteins ( P = 7.3e-137, Wilcoxon test, Figure 1H ). A similar observation was made when comparing the liver proteome results (P =1.01e-150, Figure S1F – I ), confirming the substantial analytical depth of Tissue-PPT . The overall proteome turnover patterns across tissues showed moderate similarity, with most of the proteome (66.7%-80.04%) having a PT of less than 10 days, indicative of the basic metabolism dynamics in mice ( Figure 1I ). The plasma proteome contains proteins produced and released from different tissues into the bloodstream. Our results allow us to uniquely address whether tissue proteins maintain their PTs in plasma. Based on more than 300 proteins that are detected in both plasma and each of the solid tissues, we found that blood proteins are generally longer-lived in plasma than in tissues and brain regions, possibly due to the lack of ubiquitin (Ub)-proteasome system (UPS) and cell proliferation in blood ( Figure S2A ). Due to the small number of proteins that can be detected by MS approaches in crude plasma samples 35 , we focused on PTs among tissues and brain regions in the analyses below. The median PT ranged from 3.27 days in the gut to 6.45 days in the cerebellum. The nine brain regions overall displayed significantly higher median PT (5.89 ± 0.42 days) than other tissues (3.27 days for the gut, 3.28 days for the liver, 3.62 days for the spleen, 3.82 days for the kidney, and 4.21 days for the lung), except for the heart (5.61 days). The longer protein lifetime in brain regions is consistent with previous reports 8 14 and can be mainly attributed to the brain’s lower regenerative capacity. Similarly, the heart has a very limited regenerative capacity, as cardiomyocytes rarely divide after birth 36 . Interestingly, the overall differences in protein turnover between tissues cannot be solely explained by cell proliferation. For example, despite the cell doubling time in the liver was determined to be 51 days 13 , its overall proteome lifetime is short, indicating additional in vivo factors such as cell-type diversity and tissue-specific network influencing protein turnover. Importantly, protein-specific turnover demonstrates considerable diversity: while the first 1% percentile have a PT of less than 1 day, the 99% percentile and above have a PT of greater than 100 days on average in all tissues ( Table S1 ). The mostly short-lived and long-lived proteins are diverse between tissues ( Figure S2B ). Only 49 proteins are consistently the top 5% most long-lived in each of the nine brain regions ( Figure S2C ), comprising those proteins enriched in TCA cycle and respiratory electron transport ( P =3.91E-12), myelin sheath ( P =2.43E-11), chromatin assembly ( P =8.59E-10), and collagen-containing extracellular matrix ( P =4.96E-13). Among them, only six structural proteins are commonly top 5% long-lived across all the other tissues, including Plp1 (PT, 87.54 days on average), Cldn11 (153.42 days), Mog (107.53 days), Nfasc (49.16 days), Col5a2 (114.16 days), and Ccdc177 (50.31 days). Together, these results underscore the importance of measuring and understanding individual protein turnover rates in various tissues. Protein abundance and lifetime profiling jointly define tissue proteome function and stability Protein turnover depicts a functional dimension that is largely independent of protein expression 6 32 . Our Tissue-PPT integrates matched PA and PT of the same proteins, which may enhance our understanding of protein essentiality in tissues. For example, by examining proteins Tau (MAPT) and α-synuclein (SNCA), we confirmed that both proteins exhibit significantly higher abundance in brain regions compared to other tissues. Furthermore, while α-synuclein shows comparable lifetimes in the brain and other tissues, the lifetime of Tau is significantly longer in the brain ( Figure 1J ), where it would be prone to accumulate changes that can lead to pathologies such as tauopathies 37 38 . To explore biological insights from PA and PT quantification, we firstly examined the total variance of PA and PT datasets using principal component analysis (PCA). As expected, all brain regions formed a distinct cluster separated from other tissues in the PCA of both PA and PT ( Figure 2A ), indicating a smaller biological variability between brain regions. The PT for the gut appeared as an outlier among solid tissues, possibly owing to its rapid cell proliferation. Overall, the cross-tissue correlation analysis of PA and PT showed a similar trend of PCA results ( Figure 2B ), suggesting substantial proteostasis diversity. Following, we annotated the functions of the 5% shortest- and longest-lived proteins based on their average PTs in brain regions and non-brain tissues ( Figure 2C and Figure S2D ). Intriguingly, distinct biological processes enriching either short- or long-lived proteins were identified ( Figure 2D and Figure S2E ). Just as examples, the “core matrisome” and “aerobic respiration” are associated with long-lived proteins, whereas “DNA damage response” and “protein polyubiquitination” are linked to rapid turnover, all consistent with previous reports 39 40 41 42 43 . To understand PT’s dependency on PA, we profiled the correlation between PA and PT across all tissues for different protein functional classes ( Figure 2E ). This analysis reveals that certain protein groups, such as the chaperonin complex, respiratory chain complex I, and proteins involved in organ formation, exhibited a positive correlation between PA and PT, indicating a coordination between protein expression and lifetime. In contrast, proteins from the ribosome and the 48S preinitiation complex show no or even negative PA-PT correlations, suggesting that cells use protein turnover to regulate the protein production machinery, thereby promoting tissue homeostasis. Figure 2. Concurrent protein abundance and lifetime profiling of the mouse tissue proteome. Open in a new tab (A) PCA of PA and PT in brain regions and solid tissues. (B) Hierarchical clustering heatmap showing Pearson correlation (after Spearman correction, see Methods) between tissues and brain regions based on PA and PT, respectively. (C) Hierarchical clustering heatmap of the top 5% of shortest-lived proteins (left panel) and the top 5% longest-lived proteins (right panel) with less than 50% missing values across all samples. The color bar denotes the Log 2 (T 50 ). (D) Biological processes enriched among the 5% shortest- and longest-lived proteins, based on their average PTs in brain regions and non-brain tissues. Enrichment p-values were reported by Metascape. All processes with “%InGO” (#GeneInGOAndHitList/ #GeneInHitList) >5% among the Top 100 list were shown. (E) Distribution of cross-tissue Spearman correlation between PA and PT for all proteins across tissues and regions. Upper panel: Density histogram of Spearman rho values for all proteins. Lower panel: Boxplots of protein-specific Spearman rho values for all proteins and selected GO terms. (F) Heat-circle (HC) plot visualization of PA and PT across tissues at different levels of cellular organization. Upper left panel: PT across all tissues, with proteome abundance normalized. Middle left panel: HC plot visualizing main cellular components across samples. Other panels: Examples for individual protein groups. The blue-to-red color gradient denotes PT from long to short in terms of Log 2 (T 50 days). The size of the HC plot circles is proportional to the Log 2 (riBAQ) in ppm value indicating PA. Triangle: Log 2 (riBAQ) < −1, Diamond: Log 2 (riBAQ) > 9. Next, to illustrate the synergistic profiling of PA and PT, we developed a heat-circle (HC) plot. In this plot, protein iBAQ values (a proxy for protein copy number) 44 45 are derived from DIA-MS readouts and then transformed to the relative iBAQ values (riBAQ 46 , expressed as Parts Per Million or ppm, see Methods), determining the relative size of the circles. The color gradient from red to blue indicates the lifetime of the proteins or protein sets, with red representing short-lived proteins and blue representing long-lived ones ( Figure 2F ). The HC plot essentially provides a view of proteome stability and cellular energy expenditure across various levels, including e.g., tissues, functional protein groups, organelles, sub-organelles, and individual proteins. At the organelle level, the HC plot reveals that the extracellular matrix (EM) consistently enriches long-lived proteins across tissues, indicating that EM proteins do not undergo rapid turnover compared to other cellular components 9 39 . A similar pattern is observed for components of the plasma membrane, such as solute carrier (SLC) proteins, which are likely critical for maintaining tissue integrity. Collagen proteins, although few in number, are highly abundant in non-brain tissues and are extremely long-lived in all tissues, reflecting their critical roles in maintaining tissue structure 47 48 . In contrast, mitochondrial and nuclear proteins exhibit higher cross-tissue variability, indicating tissue-specific dynamic regulation. At the sub-organelle level, the HC plot shows that respiratory chain complex I proteins in the brain have higher abundances and longer lifetimes compared to other tissues and other mitochondrial proteins, such as mitochondrial ribosomal subunits, emphasizing the importance of oxidative phosphorylation in the brain 12 , consistent to Figure 2D . At the individual protein level, SLCs exhibit a wide range of PAs but have similar PTs across tissues. Interestingly, certain proteins, such as Mrps24 in the mitochondrial ribosome, MTND1 in respiratory chain complex I, HSP90aa1 and Hsd17b12 in the EM, and Slc4a1 and SLC12a5 in the SLC, have lifetimes that differ from those measured within their complexes and functional classes, potentially pointing to moonlighting protein functions independent of their complexes. Together, the HC plot supported in Tissue-PPT App where PAs can be showcased in the scales of riBAQ, iBAQ, or DIA-MS intensity, effectively visualizes both PA and PT ( Figure S3A – B ), offering complementary insights into tissue functional diversity. Lifetime diversity of E3 ligases is critical for determining tissue-specific proteolysis To understand the proteolysis landscape across different tissues, we examined PA and PT profiles for the major cellular protein degradation machineries 49 : the proteasome, lysosome, E3 Ubiquitin Ligases (E3), E3 accessory proteins, deubiquitinating enzymes (DUBs), and protein folding chaperones ( Figure 3 ). Using HC plots, we found that both 19S and 20S proteasomes are tightly regulated by similar PA between subunits and correlated PT across tissues ( Figure 3A ). The kidney has an exceptionally fast turnover of the proteasome compared to the total proteome turnover ( Figure 2F ), which might be crucial for its function in degrading and reabsorbing the high load of proteins and small peptides filtered by the glomerulus. In comparison, lysosomal proteins display a much higher PA dynamic range and variability ( Figure 3B and 3D ). Individual enzymatic proteins in the lysosome, such as Man2b1, Pla2g15, and Capn1, are particularly short-lived in brain regions. Conversely, lysosomal proteins are long-lived in the spleen relatively ( Figure 3C ), likely due to the spleen’s role in immune surveillance and phagocytosis 50 . In addition to proteasome, DUBs maintain the most stable PA levels across tissues ( Figure 3D and Figure S4A – C ), suggesting the core activities mediated by DUBs are universally required across all cell types. Despite general high PA levels, specific molecular chaperones, like heat shock proteins (HSPs), exhibit diverse PT profiles ( Figure S4A ). Hspa12a is long-lived in all tissues, potentially due to its localization in extracellular exosomes. Again, Hsp90aa1, the stress-inducible isoform of the cytosolic chaperone protein HSP90 51 , has a short lifespan across tissues including brain, possibly reflecting its role in rapid proteostatic responses. Interestingly, E3 ligases and their accessory proteins, despite being the least abundant among the protein degradation machineries, are the most dynamic, exhibiting the highest PT variability across tissues ( Figure 3D – E ). Together, these findings underscore that the key steps of protein degradation such as the turnover of the E3 ligase pathway can vary in different tissues to maintain tissue-specific proteostasis. Figure 3. Characterization of protein removal processes across tissues. Open in a new tab (A) HC plot of proteasome subunits (19S and 20S), lysosomal proteins across tissues, and E3 ubiquitin ligases across tissues. The color and size are defined as in Figure 2F . E3 proteins with PA quantified across mouse tissues (11 out of 14) were shown. (B) Hierarchical clustering heatmap of PA profiles of the five proteins representing protein degradation machineries. The brown-to-green color bar indicates the increasing relative abundance in terms of Log 2 (riBAQ). (C) The same heatmap as (D) for PT profiles. The red-to-blue color bar indicates the increasing PT in terms of Log 2 (T 50 days). (D) Boxplot of standard deviation S.D. of [Log 2 (riBAQ of each protein)- Log 2 (riBAQ of averaged level)] for each protein list indicating PA variability across tissues (11 out of 14 tissues). The numbers above the boxes indicate protein numbers. (E) Boxplot of standard deviation S.D. of [Log 2 (T 50 of each protein)- Log 2 (T 50 of averaged level)] for each protein list indicating PT variability across tissues (11 out of 14 tissues). (F) Hierarchical clustering heatmap of ubiquitin (Ub) linkages K63, K48, K27, K11, and K6 across tissues. The color bar indicates the relative abundance of Ub chains, normalized using unmodified ubiquitin peptides (see Methods). (G) Heavy-to-light (H/L) ratios during pSILAC labeling for K48- and K63-linked Ubs, the unmodified Ub based on peptide TLSDYNIQK, and the averaged proteome. The dashed lines indicate brain regions, while the solid lines denote other tissues. The horizontal red dash line denotes H/L=4. Furthermore, we investigated the utilization of the ubiquitin resource across tissues. By re-searching our DIA-MS data for Ub modifications, we assessed signature peptides that distinguish Ub chains linked to different lysine (K) residues 52 53 49 . including K48, K63, K11, K27, and K6, which were detected in most tissues 54 55 ( Figure 3F ). Label-free quantification (Methods 49 ) estimated that K48-linked Ub, which plays a classic role in the UPS, is the predominant Ub chain. In line with the slow turnover in brain regions, we observed the lowest levels of K48-linked Ub in brain regions. Beyond relative abundance, our analysis inferred the diversity of recycling Ub resources. Strikingly, K48-linked Ub, once synthesized, appears to be retained significantly longer than K63-linked Ub and other proteins across most tissues ( Figure 3G ). This finding might indicate the coupling mechanism between substrate deubiquitination of K48-linked Ub and degradation 56 and differential recycling strategy of variable polyubiquitins 57 . Additionally, the liver demonstrated an exceptionally high turnover of K63-linked Ub which has the diverse signaling roles such as endocytosis and autophagy 58 . Collectively, our findings highlight tissue-specific Ub distribution, architectures 59 , and dynamics. Protein lifetime tightly correlates with protein-protein interaction in tissues Proteins rarely act alone in a living cell. When two proteins are engaged in a physical protein-protein interaction (PPI), it is tempting to hypothesize that they are synthesized and degraded in a coordinated manner. Previous studies have demonstrated that PTs of the same organelle, family, or complex are often correlated 60 61 9 . Recently, Skinnider et al. established a comprehensive PPI dataset across various mouse tissues, using protein correlation profiling mass spectrometry 62 . By integrating that dataset with Tissue-PPT , we were able to correlate PT and PPI across tissues in detail. First, we confirmed that the binary correlations of PA between PPI partners based on CORUM 63 , BioPlex 64 , and Skinnider et al. 62 are all significantly higher than those between random protein pairs that do not interact ( Figure 4A ), consistent with the established proteome organization mechanisms through PPIs 6 65 . Remarkably, PT profiles across tissues showed a similar correlating pattern ( Figure 4B ): e.g., Skinnider et al. identified 107,553 PPIs overlapping with our PT data, which are 2.69 and 14.76 times more than PPIs extracted from CORUM and BioPlex; and this PT data demonstrates the most dramatic aforementioned correlation difference (mean Pearson correlation, 0.53 for PPI partners vs. 0.07 for non-PPI pairs). The even more pronounced correlation of PTs between PPI partners, compared to PAs (right panels of Figure 4A – B ), demonstrates the importance of resolving tissue-specific PPI networks 62 66 , underscoring a previously underappreciated dependency of PT on PPI. We confirmed these results in an analysis comparing PPI pairs 62 and random protein-protein pairs filtered to the same KEGG pathways (see Methods and Figure S4D ). We additionally mapped the Pearson correlations based on PA and PT across tissues to PPI confidence levels in hu.MAP 67 . Consistently, this analysis reveals that PTs provided significantly better discrimination than PAs for those most confident PPIs (Level 5, Extremely High confidence, Figure 4C ). Figure 4. Association between protein lifetime and protein-protein interaction (PPI) across tissues. Open in a new tab (A) Boxplots of binary correlation coefficients (Pearson R) for PA between PPI partners, based on CORUM, Bioplex 3.0, and Tissue-PPI (by Skinnider et al.). P , Wilcoxon test. “In” and “Out” denote PPIs included or not listed. The red dashed lines denote R=0 and 0.5. (B) The same boxplot as in (A) for PT. (C) The same boxplot as in (A) for both PA and PT using PPI confidence levels in hu.MAP. Levels 1–5 indicate increasing PPI confidence. (D) ROC curves indicating PPI predictive power of PA, PT, and their logistic combination, alongside CORUM- and Bioplex-derived lists. The Extremely High and Very High confidence groups of PPIs from hu.MAP were used as true positives (TP). An equal number of randomly false pairs were used as false positives (FP) (Methods). AUC, Area Under the Curve. (E) Visualization of PT for PPI partners of PSMD1, LMNA, and AK2 proteins (the central nodes) in representative tissues. PTs of the central nodes are visualized using the yellow-to-green color gradient. The red-to-blue bar denotes the relative PT difference between PPI partners and central nodes (partner’s T 50 – central node protein’s T 50 ). Tissue unique PPI, PPI identified in this tissue, and PPI identified in any tissue are classified based on Tissue PPI (see Methods). Conversely, we asked whether PT correlation between protein pairs across tissues would suffice to predict a PPI, by using the receiver operating characteristic (ROC) analysis. We herein firstly assigned Level 4 (High) or 5 (Extremely High) PPIs in hu.MAP as “true positives” and the same number of random protein-protein pairs, which are not listed in any of the above resource, as “negatives”. We found that PT alone could effectively predict PPIs (AUC =0.82, Figure 4D ). Notably, combining PT and PA yielded an AUC of 0.89. And both PA and PT covariations outperformed CORUM and BioPlex in predicting hu.MAP PPIs, highlighting the potential of leveraging PT and PA across tissues to predict PPIs or refine PPI lists in animals. Note that this ROC analysis is based on cross-tissue correlation coefficients of PAs and PTs and therefore could not predict tissue-specific PPIs. Intriguingly, because of the additional evidence for tissue-specific PPI in Skinnider et al, examining the PT differences between a given protein and its PPI partners in one specific tissue or in any of the measured tissues leads to the recognition of PPI partners with constantly divergent turnover rates ( Figure 4E and Figure S3A ), such as Agrn for Psmd1, Lmnb2 for Lmna, and Map2 for Ak2 (in liver and kidney). These results might indicate the presence of additional partners and roles for these proteins which show peculiar “PT-deviating” profiles. Taken together, our findings demonstrate that cross-tissue PT is tightly constrained by PPIs, offering insights into turnover dynamics of individual proteins and protein networks. Multi-omic analysis reveals tissue-specific lifetime changes for peroxisome proteins How does proteome turnover contribute to gene expression and tissue diversity? Herein, we integrated Tissue-PPT with a recent dataset 68 describing both the transcriptome and translational profiles – measured using RNA sequencing and the sequencing of ribosome protected RNA fragments (hereafter, RPF), respectively – from six mouse tissues, five of which overlap with the ones in Tissue-PPT (liver, heart, lung, kidney, and brain). We found that the RPF data correlated better with the estimated proxy of protein synthesis rates than with the protein iBAQ values using our MS-based measurements in all the five tissues (see Methods and Figure S5A – B ), confirming the relevance of RPF data in inspecting global translational regulation. First, we performed multi-layered absolute correlation analysis across all proteins per tissue. The highest correlation is observed between mRNA and RPF data (Spearman rho =0.76–0.83). The correlation between RPF and the proteome ( rho =0.41–0.54) and the correlation between mRNA and the proteome ( rho =0.42–0.55) were both lower and comparable in this dataset ( Figure 5A ). Additionally, absolute PT exhibits weak or no correlation with absolute PA, but a slight yet notable negative correlation with levels of both mRNA and RNA being translated. This result aligns with the reported buffering role of protein turnover in globally modulating transcriptional 69 and translational regulations to fine-tune the functional proteome in -tissues 70 . Second, to discern which layer drives the specific tissue proteotype 71 , we compared multi-layered correlations across tissues, between proteins identified in all tissues (n =1919, Figure 5B ) and proteins exhibiting >4-fold higher PA levels in a particular tissue (n =418). We found that tissue-enrich proteins 72 are predominantly influenced by regulatory steps prior to protein synthesis rather than by protein turnover, because of the much stronger correlations between mRNA, RPF, and proteome than the correlations between mRNA or RPF and PT, as confirmed by both DIA- and TMT- derived PTs ( Figure S5C ). Third, to determine whether protein turnover of certain proteins and functions are particularly changed, we combined protein-specific correlations based on the protein’s GO cellular compartment ( Figure 5C ). This analysis revealed a remarkable anti-correlation between mRNA and PT, RPF and PT, and protein and PT ( rho = −0.42, −0.41 and −0.37, respectively) for peroxisomal proteins, as compared to other organellar proteins, indicating an intriguing opposed trend of PT to protein production for peroxisome. We then compared the averaged levels of mRNA, RPF, PA, and PT in each organelle ( Figure 5D ) and for individual peroxisome proteins among tissues, discovering protein-specific cross-layer correlation patterns ( Figure 5E – F and Figure S5D – E ). We found that, in general, 60% of peroxisomal proteins, despite having relatively lower transcript and translation levels, are long-lived, whereas the remaining 40% exhibit the opposite characteristics (i.e., higher levels of transcript/translation but short-lived, Figure 5F ). Comparing peroxisome proteins and the rest of the proteome revealed the significantly accelerated turnover of peroxisome proteins in the liver and their longer lifetime in heart ( Figure S5F ), demonstrating tissue-dependent lifetime regulation of peroxisome proteins is likely to exist. Figure 5. Cross-tissue multi-omic analysis and turnover control of peroxisome proteins. Open in a new tab (A) Proteome-wide absolute Spearman correlation between measurements of mRNA, RPF, PA, and PT in five tissues. The brain results were determined by averaging all brain regions. The red dashed lines denote Spearman rho =0 and 0.5. (B) Boxplots of protein-specific Spearman rho values between multi-omic layers for all measured proteins (upper panel) and tissue-enriched proteins (lower panel, protein abundance at least four times higher than the average of other tissues). (C) Heatmap visualizing the cross-tissue Spearman correlation between multi-omic layers. (D) Heatmap of quantitative results (column-scaled) for GO Cellular Components across multi-omic layers. The blue-to-red color bar represents the summed values of proteins associated with specific GO Cellular Components based on Perseus annotation. (E) Boxplots of mRNA abundance, PA, and PT levels for peroxisome proteins. (F) Heatmap of multi-omic quantitative results for individual peroxisome proteins in five tissues. Previous studies reported that peroxisomes are regulated by a selective autophagic degradation process called pexophagy 73 and that peroxisomal biochemical pathways are specialized in different organs 74 . In our data, we observed a significant correlation between the abundance of pexophagy-associated proteins and the overall lysosome levels across tissues (R =0.54, Figure S5G ). Additionally, both HC plots and individual PA~PT correlation plots, generated using Tissue-PPT , revealed negative PA vs. PT correlation for Pex3, an essential activator of pexophagy 75 and Atg12, a ubiquitin-like protein critical for the formation of autophagosome and autophagy 76 77 across tissues ( Figure S5H – I ). Together, this multi-omic analysis highlights the strong lifetime diversity of specific peroxisomal proteins in specific tissues, possibly enabling cells to rapidly adapt to cellular stress and metabolic demands. Site-specific phosphorylation functionally shapes protein lifetime across tissues In addition to whole protein turnover, Tissue-PPT presents tissue phosphoproteome turnover profiles. Overall, our DIA-MS and pSILAC-DIA measured the abundances of 67,169 P-sites and T 50 of 40,573 P-sites, delineating the dynamics of in vivo phosphorylation on a large scale. In the brain regions and non-brain tissues, respectively, we quantified 34,157 ± 2207 and 22,821 ± 4650 P-site-carrying peptides in terms of abundance and 12,861 ± 1441 and 7575 ± 2681 of them in terms of T 50 ( Table S2 , Figure 6A , and Figure S6A ). In contrast, only about 100 P-sites were quantified in the blood plasma. The PCA plots of phosphoproteomic abundance and turnover variances displayed the inter-tissue and inter-region patterns similar to total proteomic results ( Figure 6B ). P-site T 50 correlations between brain regions were markedly stronger than those between different tissues, similar to T 50 results of the total proteins ( Figure 6C and Figure S6 ). Figure 6. Profiling site-specific phosphorylation turnover and its impact across mouse tissues. Open in a new tab (A) Number of quantified P-sites with abundance and lifetime values across tissues (Hippocampus not included for phosphoproteomics due to low sample yield). (B) PCA plots of P-site abundance and lifetime across tissues. (C) Pearson correlation analysis of P-site abundance and lifetime between tissues, with the blue-to-red color bar indicating increasing correlation after Spearman correction. (D) Distribution of Spearman correlation between T 50 of the phosphorylated (phos_T 50 ) and non-phosphorylated peptides (nonPhos_T 50 ) for all specific P-sites across tissues, mapped to the kinase library based on kinase-substrate annotation. 106 kinases with 30 or more putative P-site substrates (Percentile >0.99) quantified with respective T 50 are shown. (E) Mapping of Spearman correlation between T 50 of the phosphorylated (phos_T 50 ) and non-phosphorylated peptides (nonPhos_T 50 ) across tissues onto a kinase phylogenetic tree. The size of the kinase nodes represents the number of P-site substrates, and the blue-to-red color bar indicates increasing Spearman rho . (F) Volcano plots showing P-sites linking to increased or delayed protein turnover (i.e., destabilizing or stabilizing the corresponding protein) in brain regions and non-brain tissues. The fold change was determined by comparing phos_T 50 to nonPhos_T 50 of the same peptide sequence. P , Student’s t-test. Blue and red dots denote P-sites (P <0.05) showing the |fold change| >1.5 (in brain) and >1.2 (in non-brain tissues). To explore how site-specific phosphorylation alters protein turnover across different tissues, we harnessed our previously developed DeltaSILAC method 22 by considering the non-phosphorylated protein forms 20 . DeltaSILAC essentially integrates pSILAC, phosphoproteomics, and a peptide-level matching strategy 22 20 25 and was initially applied to growing HeLa cells 22 23 24 . Herein, for a given P-site, we therefore compared the T 50 of a phosphorylated (p) peptide to the T 50 of its non-phosphorylated (np) counterpart within the same tissue’s whole proteomic results. We firstly distributed the correlation between T 50 values of p and np peptides for all P-sites across tissues ( Figure S6B ) to the recently established kinase library by the Cantley group and others 78 , 79 . A total of 106 kinases, each covering more than 30 P-site substrates quantified with respective T 50 pairs (Percentile >0.99), were evaluated ( Figure 6D ). We found the phosphorylation-induced T 50 alteration for the substrates of the same kinase can vary significantly. Mapping this result to kinase phylogenetic tree, we observed that substrates of CAMK family showed weak T 50 cross-tissue consistency between p and np peptides overall (e.g., rho =0.297–0.340 for MARK kinases, Figure 6E ). In contrast, the STE family presents strongest corresponding T 50 correlations, such as YSK1 ( rho =0.700), TAO1 ( rho =0.650), TAO2 ( rho =0.650), and MYO3A ( rho =0.646), much higher than CAMK’s results ( P <2.2e-16). Thus, after being phosphorylated by a particular kinase, a P-site lifetime and the corresponding protein-level lifetime might be regulated by independent in vivo mechanisms. Furthermore, leveraging the high data completeness of T 50 across brain regions, we also conducted a 2D functional enrichment analysis using protein-level functional annotation ( Figure S7 ). We found that the abundance differences of P-sites across brain regions are particularly relevant to biological processes such as Electron transport chain (P =0.00995), Endosome (P =0.00601), Response to ER stress (P =0.0317). On the other hand, P-site lifetime variability in the brain tends to be associated with different processes, such as Actin nucleation (P =0.0319) and Positive regulation of protein localization in plasma membrane (P =0.0466). This suggests that the distribution of P-site abundance and the duration of P-site presence across brain regions are critical to the brain’s complex functions. Next, we directly determined the real T 50 difference for p and np pairs (i.e., ΔT 50 ). In growing cells, more P-sites tended to associated with higher T 50 22 23 , a trend that was found not as evident in present tissue results ( Figure S6C ). Using volcano plots ( Figure 6F ), we identified 581 and 592 P-sites significantly extending or shortening T 50 across brain regions (P <0.05, |FC| >1.5), and 146 and 105 P-sites doing the same across other tissues (P <0.05, |FC| >1.2). Based on the kinase library annotation 78 , 79 , P-sites accelerating turnover in the brain were enriched as substrates for kinases TSSK2 (P =0.0160) and HUNK (P =0.0407), while those stabilizing proteins were enriched as substrates for SIK (P =0.0225), NEK11 (P =2.89E-4), and others ( Table S2 and Figure S6D ). Furthermore, for extremely in vivo long-lived proteins (ELLPs) such as nucleoporins, histone variants, and enzymes 47 80 , our data here identified specific P-sites that further extend or fine-tune their PT ( Figure S6E ). Together, our results profiled and prioritized P-sites based on their linkage to in vivo protein stability. And profiling Heavy/Light (H/L) ratios of p- and np- peptides for key P-sites and crucial proteins might enable new opportunities to target them (see examples Figure 7A ). Figure 7. Demonstration and verification of phosphorylation sites linked to protein turnover in mouse tissues. Open in a new tab (A) Heavy/Light (H/L) ratio curve examples during the labeling course for a phosphorylated (p) peptide and its non-phosphorylated (np) peptide counterpart of the same sequence. (B) Validation of phosphorylation’s impact on Tau protein using PhosTAC. Upper panel: The pSILAC experiment comparing Tau protein turnover after PhosTAC or DMSO treatments. Lower panel: H/L curves during treatment and pSILAC labeling. P , Student’s t-test. (C) Validation of Tau and alpha-synuclein P-sites associated with protein degradation in primary hippocampal cortical neurons. Left panel: Neurons were infected with FLAG-tagged alpha-synuclein or Tau, either as wild type (WT) or mimicking mutants dephosphorylated (T/S to A) or phosphorylated (T/S to D). After three days of expression, neurons were treated with cycloheximide, chased for different times, stained, automatically imaged, and FLAG fluorescence intensity was measured. Right panel: Fluorescence imaging results from three independent experiments for alpha-synuclein (T81) and Tau (MAPT, S522, T525). Bars represent SEM. Statistical test: ANOVA. *p < 0.05; ****p < 0.0001. Scale bar: 5 μm. As examples, we focused on microtubule-associated protein Tau and α-synuclein, both of which are widely recognized for their crucial roles in neurodegenerative diseases. Our DeltaSILAC analysis determined that phosphorylation at S522 and T525 of Tau significantly extended its PT by 7.24 and 9.20 days, respectively, across brain regions. Similarly, phosphorylation at T81 of α-synuclein markedly prolonged its PT by 13.20 days on average ( Figure 7A ). Consistently, hyperphosphorylated Tau was reported to promote aggregation and self-assembly into paired helical filaments tangles 81 , affecting protein degradation 82 . Promoting the removal of Tau phosphorylation might offer therapeutic potential. Similarly, the phosphorylation of α-synuclein affects its aggregation and neurotoxicity 83 . In the first validation approach, we employed a recently developed phosphorylation-targeting chimera (PhosTAC) technology to promote Tau dephosphorylation through induced proximity with the active PP2A holoenzyme 84 . Unlike proteolysis-targeting chimeras (PROTACs) that induce selective intracellular proteolysis, PhosTACs induce rapid and sustained protein dephosphorylation 84 85 . We firstly confirmed the downregulation of multiple Tau phosphorylation sites by PhosTAC treatment 84 . Then we measured the H/L ratios of Tau in an in vitro system, where the cell culture medium was replaced with heavy SILAC during PhosTAC administration ( Figure 7B ). We observed significantly accelerated Tau protein degradation (i.e., shorter PT) following PhosTAC treatment, indicating the potential to regulate Tau protein clearance through its dephosphorylation. As another independent verification, we overexpressed the wild-type and phosphomimetic mutants of Tau and α-synuclein in primary rodent cortical neurons. We then measured their lifetimes by imaging after cycloheximide blockade, as previously described 22 ( Figure 7C ). Our results indicate that for α-synuclein, the phosphomimetic mutation T81D significantly increases the protein’s lifetime by 3.81-fold compared to the wild-type. The Tau phosphomimetic mutations T522D and T525D exhibited analogous effects. In summary, the phosphoproteome turnover dataset included in Tissue-PPT provides promising opportunities for exploring the dynamics and turnover of individual P-sites. DISCUSSION We establish Tissue-PPT , a large inventory of proteome and phosphoproteome turnover across multiple tissues and brain regions, including more than 256,000 PT measurements in total. The Tissue-PPT Web App supports convenient navigation and discovery of PA and PT profiles for individual proteins and protein sets. Our datasets and analyses represent a significant step towards the molecular understanding of tissue phenotypic and functional diversity, elucidating not only the composition of a proteome and the quantity of its constituents, but also the lifespan and activity of individual proteins among tissues and brain regions. The high-coverage, precise, and reproducible determination of protein turnover kinetics reported in this study stems from the MS techniques we employed. Both DIA and TMTpro were devised to improve the quantitative accuracy via either the gas-phase or LC fractionations 28 32 30 , significantly ameliorating the missing value issue 86 in traditional shotgun proteomics. The in vivo SILAC strategy based on Lysine-6 was used to avoid the complexities associated with 15 N labeling which require sophisticated process algorithm 87 and might impair measurement precision 9 and proteome coverage 10 . Our results underscore the importance of integrating well-matched datasets across diverse biological contexts. First, matching PA with PT profiling 36 identifies positive PA-PT correlation indicating concerted post-transcriptional regulation. Proteins with the highest PAs and PTs can be considered housekeeping proteins essential for tissue function diversity. Conversely, maintaining shorter PTs for the most abundant proteins allows for rapid responses to environmental perturbations. Stabilizing low-abundance proteins, particularly those in protein complexes and rate-limiting enzymes, may promote proteome buffering. Our HT plots succinctly summarize these reciprocal PA-PT relationships. Second, matching phosphorylated and non-phosphorylated peptides ensures precise estimation of turnover effects due to site-specific phosphorylation 22 . This is essential, given that many proteoforms carrying other PTMs were not measured 20 . Using this approach, we discovered that in vivo phosphorylation can alter the stability of key proteins, such as Tau and α-synuclein. Our results emphasize that phosphorylation is not only crucial in initiating rapid cell signaling response, but also in regulating protein stability in steady-state tissues, possibly through kinase selectivity and variable mechanisms such as PPI 25 24 . The observed lifetime diversity of protein forms with different phosphorylation could be due to various factors, such as their participation into different protein complexes or aggregates, their different cellular locations, the order of in vivo PTM addition or their specific removal during protein maturation, and their relative abundance and stability in different cell types within the tissue. Concurrent turnover analysis of phosphorylation sites encoded by the same gene will significantly enhance our understanding of phosphoproteomics 88 and guide future biochemical and functional analyses of specific P-sites. Third, matching tissue-specific PTs with PPIs revealed a previously underestimated correlation. Prior inter-partner correlation was achieved from tissue-resolved PPIs rather than Bioplex and CORUM annotations. Accordingly, our findings indicate that most PPI partners have similar PTs, and deviations from this trend suggest additional moonlighting functions. These results may help inferring PPI networks and understanding proteome organization. Fourth, matching multi-layered omic profiles across tissues provides insights into how cells orchestrate organellar pathways to maintain tissue diversity, by incorporating transcriptome, RPF, proteome, proteome turnover 89 , and phosphoproteome quantifications. For instance, while peroxisome proteins were previously reported to have variable PTs 9 , our multi-omic analysis uncovers that they also exhibit significant variability at the mRNA and PA levels, and that strikingly, their PT cross-tissue profiles to be counteracting to PA variability. Meanwhile, our results underscored the preeminence of transcriptional and translational processes in shaping the global tissue-specific proteomes, in which turnover generally plays an insignificant role. We observed the significant PT regulation for specific pexophagy-associated proteins 73 . Compared to other proteins in the proteome, the peroxisome proteins seem to be shorter-lived in tissues of fast turnover while longer-lived in tissues of slow turnover. A causal link between peroxisome protein turnover and the total proteome turnover might exist. The corresponding molecular mechanism and how peroxisome turnover are functionally associated with tissue phenotypes remain to be established. Other concrete findings include an unexpected turnover trend that K48-linked ubiquitin is replaced slower across tissues compared to K63-linked ubiquitin, which might suggest that K48-linked Ub is directly recycled by proteasome-associated DUBs during degradation which are expressed stable across tissues 52 , whereas K63-linked Ub is degraded primarily by the lysosome 54 . Furthermore, the variable PT of E3-associated proteins may also point to the potential for developing tissue-specific protein degraders as therapeutic modules targeting underexplored E3 ligases 90 . It should be emphasized that the PT values in our study represent the de facto bulk protein turnover from all the cells in tissues within a live organism. While these values accurately describe protein dynamics and are relevant for in vivo experiments such as drug discovery where compounds are administered to whole-body animals, they differ from the cellular and in some cases subcellular protein turnover rates. Indeed, the protein clearance on a per-cell basis is influenced by the combined effects of degradation kinetics and cellular dilution due to cell division. For instance, the global PT is shortest in the gut, likely due to the rapid cell turnover in gut. While 15 N labeling presented technical challenges 10 91 , it was recently used to track both protein and DNA labeling in mouse tissues via a technique termed TRAIL 13 to account for tissue proliferation differences when comparing protein turnover. This means that our absolute PT data should be applied cautiously in experiments comparing proliferative tissues versus post-mitotic tissue types. We however note that neither our approach nor TRAIL can resolve protein turnover for different cell types within tissues, unless cell sorting is performed 9 . Different cell types could impact proteome readouts 92 – 94 and vary in cell doubling time difference, leading a major source of protein lifetime diversity between tissues, in addition to the effects of tissue-specific local network and environment 66 . To fully understand the biological diversity underlying tissue doubling, cell doubling, and PT, a single-cell proteomic turnover analysis 95 for all different cells, spatial proteomic turnover analysis within and across tissues, and new “cell-omics” strategies 96 might be needed. Limitations of the study In the analysis of T 50 values using our current ODE model, we assume that protein synthesis and degradation processes are overall balanced, maintaining stable proteome homeostasis in adult mouse tissues. This assumption is supported by our observation that adult mice did not exhibit significant weight changes during one month of pSILAC labeling. However, this equilibrium does not apply uniformly to all proteins. Certain subsets of proteins, such as those with cyclic expression patterns in circadian rhythms 97 , proteins synthesized during periods of animal growth, and proteins subject to excess degradation during tissue aging, could display distinct dynamics during the labeling and would be challenging to measure accurately. Furthermore, while we discovered overall delayed protein turnover in plasma proteome, the limited coverage of the plasma proteome and phosphoproteome by MS precluded a deeper investigation in the present study. While we verified several phosphorylation events linked to altered protein stability in brain cells, mechanistically defining the relationship between specific P-sites and protein stability and dissecting the dynamics of phosphorylation itself and phosphorylation-carrying proteins are beyond the scope of the study. Last, although this survey has provided an expansive view of in-vivo bulk protein lifetime across mouse tissues, it has not addressed many biological variables that may influence bulk protein turnover, such as the age, sex, strain, and diurnal cycles. Future high-throughput analyses are needed to explore and understand protein lifetime control in many physiological and pathological states. In conclusion, we established a high-quality comprehensive resource portraying the protein turnover dynamics in mammalian tissues, providing in-depth insights into the proteostasis regulation underlying tissue diversity. RESOURCE AVAILABILITY Lead contact Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead author, Yansheng Liu ( yansheng.liu@yale.edu ). Materials availability This study did not generate new unique reagents. Data and code availability The mass spectrometry raw data and searched results have been all deposited to the ProteomeXchange Consortium via the PRIDE 98 with the dataset identifier PXD055942. The source code for the Tissue Proteome and Phosphoproteome Turnover web application (Tissue-PPT) is available under the MIT license on GitHub ( https://github.com/yslproteomics/tissuePPT ) and archived on Zenodo ( https://zenodo.org/records/14671026 ). Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request. STAR★METHODS EXPERIMENTAL MODEL AND STUDY PARTICIPANT DETAILS The male B6SJL (C57BL/6 × SJL) mice were purchased from the Jackson Laboratory. Mice of approximately 9 months were maintained in the Animal Resources Center at St. Jude Children’s Research Hospital according to the Guidelines for the Care and Use of Laboratory Animals. All animal procedures were approved by Institutional Animal Care and Use Committee (IACUC) at St. Jude Children’s Research Hospital. Mice were maintained on a 12:12 h light/dark cycle in a temperature and humidity-controlled room with food and water ad libitum . METHOD DETAILS In vivo pSILAC Labeling and Tissue Dissection Male B6SJL mice of approximately 9 months were used for global protein turnover profiling. Each mouse was provided with 5 g SILAC food (Mouse Express L-LYSINE [13C6, 99%] MOUSE FEED kit, Cambridge Isotopes Laboratories) per day for metabolic labeling in vivo . Three days prior to metabolic labeling, mice were fed with the SILAC food composed of 100% light lysine to minimize the perturbation of protein homeostasis due to the switching from regular food to SILAC food. The mice were then fed with SILAC food for specified periods before being sacrificed by cervical dislocation. A total of five mouse biological replicates were labelled (N=5 different mice). In the pilot evaluation experiments, the time points of SILAC labeling were 0 days, 4 days, 8 days, 16 days, and 32 days for three of five mice from which the brain replicates were measured. For the experiments deriving the final tissue-PPT data, 0 days, 8 days, and 32 days of mice with multiple tissues and brain regions were analyzed in both BoxCarmax-DIA and TMT MS methods, using the two biological replicates (i.e., N=2 different mice). All the anatomical samples of body tissues (heart, liver, spleen, lung, kidney, gut, plasma) and brain regions (cerebellum, frontal cortex, substantia nigra, thalamus, amygdala, entorhinal cortex, hippocampus, and olfactory bulb) were dissected rapidly, frozen in liquid nitrogen, and stored at −80 °C. Tissue Protein Extraction and Digestion Protein extraction and MS analysis were performed based on an optimized protocol 99 . About 20 mg of the mouse tissue was weighed and lysed in ~200 μL lysis buffer (8 M urea, 50 mM HEPES, pH 8.5, 0.5% sodium deoxycholate, phosphatase inhibitor cocktail (phosphoSTOP, Roche) at 4°C in a bullet blender. Protein concentration was measured by the BCA assay. Lysate containing ~1 mg proteins were digested with Lys-C (Wako, 1:100 w/w) in lysis buffer at 21 °C for 3 h. The digested mixture was diluted 4 times with 50 mM HEPES (pH 8.5) to reduce urea to 2 M, and digested with trypsin (Promega, 1:50 w/w) overnight at 21 °C. The digestion condition was selected to ensure sufficient protein digestion while minimizing the potential for urea-derived protein carbamylation. The digested peptides were reduced by freshly prepared dithiothreitol (DTT, 1 mM) for 2 h, followed by alkylating with 10 mM iodoacetamide (IAA) in the dark for 30 min. The unreacted IAA was quenched by adding DTT to 30 mM and incubate for 30 min at RT. The samples were then acidified by addition of 1% trifluoroacetic acid (TFA). Acidification of peptides by trifluoroacetic acid was followed with desalting on Sep-Pak C18 column (Waters). Samples were split into two parts for DIA and TMT-DDA analyses and dried by SpeedVac. TMTpro Labeling and basic pH HPLC Fractionation About 100 μg of dried peptides were resuspended in 50 mM HEPES (pH 8.5) and labeled by 16-plex TMTpro reagent (Thermo Fisher Scientific, ~1:2 w/w) 100 . Peptides labeled with each channel of TMT was mixed equally and desalted using Sep-Pak C18 column (Waters). The TMT-labeled peptides were fractionated by offline basic pH reverse phase LC (RPLC). Injected peptides were separated using two tandem XBridge C18 columns (3.0 mm × 15 cm, 1.7 μm particle size, Waters) in a 3-h 10–45% gradient (buffer A: 10 mM ammonium formate, pH 8.0; buffer B: 95% acetonitrile, 10 mM ammonium formate, pH 8.0) to yield total of 80 or 96 concatenated fractions. Phosphopeptide Enrichment About 500 μg peptide per each sample was used for phosphoproteomic sample preparation 28 . The phosphopeptide enrichment was performed using High-Select ™ Fe-NTA kit (Thermo Scientific, #A32992) according to the kit instructions, as described previously 101 . Briefly, the resins of one spin column in the kit were divided into 5 equal aliquots, each used for one sample. The peptide-resin mixture was mixed and incubated for 30 min at 25 °C and gently shake per 10 min, and then transferred into the filter tip (TF-20-L-R-S, Axygen) to remove the supernatant by centrifugation. Then the resins adsorbed with phosphopeptides were washed sequentially with 200 μL × 3 washing buffer (80% ACN, 0.1% TFA) and 200 μL × 3 H 2 O to remove nonspecifically adsorbed peptides. The phosphopeptides were eluted off the resins by 100 μL × 2 elution buffer (50% ACN, 5% NH 3 •H 2 O) and dried with SpeedVac (Thermo Scientific). All centrifugation steps above were conducted at 500 g × 30 s. The eluates were dried immediately and resuspended with buffer A for mass spectrometry analysis. About 1.5 μg phosphopeptide was injected into mass spectrometry for phosphoproteome quantification. BoxCarmax-DIA and DIA Mass Spectrometry The total pSILAC proteome samples (digested peptides) were measured by the BoxCarmax-DIA method optimized for protein turnover analysis 102 . The Orbitrap Fusion Lumos Tribrid mass spectrometer (Thermo Scientific) coupled with a NanoFlex ion source was used for the data acquisition. The spray voltage at 2000 V and heating capillary at 275 °C. Briefly, one BoxCarmax is consist of four MS runs (1 st , 2 nd , 3 rd and 4 th injection) to reconstruct a full MS1 scan 102 . Each run cost 60 min. The MS1 AGC was set to be 2 × 10 6 and the maximum injection time was set at 256 ms. MS1 resolution is 120,000 at m/z 200 and the normalized HCD collision energy was 28%. The MS2 AGC was set to be 1.5 × 10 6 and the maximum injection time was 50 ms. The MS2 resolution was set to be 30 000 and the MS2 scan range was 200−1800 m/z. Both MS1 and MS2 spectra were recorded in profile mode. All the phosphoproteomic samples were measured by a traditional DIA method including a 150-min gradient 103 , 104 due to the limited phosphopeptide amounts on the same LC-MS. This DIA-MS consisted of one MS1 scan and 33 MS2 scans with variable windows 102 , 105 , Except for the MS1 scan range was set to cover 350 – 1650 m/z, other MS1 and MS2 settings remain identical to those in BoxCarmax. To match the phosphorylation data in DeltaSILAC analysis 22 , the total pSILAC proteomic samples (N=2) were measured repeatedly by the same DIA method using a 240-min LC gradient. LC separation was performed on EASY-nLC 1200 systems (Thermo Scientific, San Jose, CA) using a 75 μm × 50 cm length column (CoAnn Technology). For eluting peptides, buffer B (80% acetonitrile containing 0.1% formic acid) from 5% to 37% and corresponding buffer A (0.1% formic acid in H 2 O) were used in all the gradients. The flow rate was kept at 300 nL/min with the temperature controlled at 60 °C using a column oven (PRSO-V1, Sonation GmbH, Biberach, Germany). TMT Mass Spectrometry About 200 ng of peptides from every basic pH HPLC fraction were loaded on a reverse phase column (75 μm × 25 cm, 1.7 μm C18 resin, CoAnn Technology) interfaced with a Q Exactive HF mass spectrometer (Thermo Fisher Scientific) 106 . Peptides were eluted in a 90 min 10–35% gradient of buffer B (buffer A: 0.2% formic acid, 3% DMSO; buffer B: 67% acetonitrile, 0.2% formic acid, 3% DMSO). The mass spectrometer was operated in data-dependent mode with MS1 at settings of 60,000 resolution, 1 × 10 6 AGC target and 50 ms maximal ion time. The MS1 was followed by top 20 MS2 high resolution scans that were set as follows: 1.0 m/z isolation window, 0.2 m/z offset, 60,000 resolution, 110 ms maximal ion time, 1 × 10 5 AGC target, HCD, 32% normalized collision energy, and 15 s dynamic exclusion. Molecular Biology and Adeno-associated Virus (AAV) Production SNCA and MAPT sequences were retrieved from Ensembl ( https://www.ensembl.org/ ) and ordered in puc57-KANA from GenScript, flanked by restriction sites (AgeI/BshTI and SdaI/SbfI) for one-step insertion into AAV constructs. Appropriate mutations were synthesized. For detection and direct comparison in imaging experiments, a C-terminal 3X-FLAG sequence was included in the synthesized sequences. After insertion into the AAV backbone, a single cassette was generated under the control of a human synapsin 1 gene promoter 107 . After ligation, AAVs plasmids were amplified in SURE competent cells (Agilent) to avoid ITR loss. Final constructs were verified by sequencing and the absence of ITRs was confirmed by DNA restriction analysis. Sequences from gene synthesis are provided as supplementary files, and all plasmids are available from the authors upon reasonable request. Viruses were prepared as previously described 107 by cotransfection of helper plasmids with the target plasmid using Lipofectamine 2000 (Thermo Fisher). At 72-hour post-transfection, cells were harvested and lysed by 4 cycles of thawing and freezing followed by treatment with Benzoase nuclease (Millipore) and incubated at 37°C for ~30 min. After pelleting cell debris (14000 rpm, 30 min at 4°C), supernatants were filtered, aliquoted and snap frozen in liquid nitrogen. Viruses were stored at −80°C until use. Viruses were titrated by imaging to achieve comparable expression levels upon FLAG-staining (see below). Primary Cortical Neuron Preparation and Infection Primary cortical cultures were prepared from P2 neonatal rats ( Rattus norvegicus , Wistar) with minor adaptations to those previously described 108 , 109 . Cortical neurons were plated on 1 mg/ml poly-L-lysine-coated (Sigma) 96-well glass bottom plates optimized for imaging (Cellvis) at a concentration of 30,000 cells per well and maintained at 37°C in 5% CO 2 . On the second day in vitro (DIV), 5-fluoro-2’-deoxyuridine (FUdR; Sigma) was added to the culture at a final concentration of 5 μM to prevent glial proliferation. At 5 DIV, neurons were infected with AAVs containing the sequence for the gene of interest. At 8 DIV, cells were treated with cycloheximide (0.5 μg/ml final, Sigma) and followed for different times to determine the dynamics of protein turnover. Immunofluorescence, Imaging, and Analysis At the end of the experiments, neurons were fixed for 30 minutes in 4% paraformaldehyde (PFA) in phosphate-buffered saline (PBS) at room temperature (RT). After rinsing with PBS, the cells were quenched with 10 mM ammonium chloride in PBS at RT for 15 minutes. The cells were then washed three times for 5 minutes, blocked and permeabilized in permeabilization buffer (PB) containing 4% bovine serum albumin and 0.1% (v/v) Triton X-100 for 30 minutes at room temperature (RT). Primary antibody (anti-FLAG rabbit; Cell Signaling cat. 14793) was applied at a final concentration of 1:1000 in PB buffer for 1.5 hours at RT with gentle shaking. After four 15-minute washes with PBS, the secondary antibody (donkey; anti-rabbit AF488, Jackson Immunoresearch cat. 711–545-152) was applied in PB buffer for 1 hour at RT with gentle shaking. After four 15-minute washes with PBS, cells were imaged using a BioTek Cytation 5 Cell microscope with constant illumination and exposure. Images were analyzed as described previously 110 . Experiments were repeated on 3 independent cultures (N=3). In each experiment, at least 3 wells for each time point were analyzed amounting overall to ~3000 neurons per time point. Tau-PhosTAC Experiment and Measurement The Tau expressing cell line (HeLa tau/PP2A) was generated as previously described 111 . HeLa tau/PP2A expressing cells were treated with doxycycline (2 μg/mL) for 24 h to induce tau expression. The media was then replaced with fresh SILAC media supplemented with DMSO, PhosTAC (1 μM) for 1h, 4h or 12h before harvest. The cells were washed with precooled PBS twice and snap-frozen with the liquid nitrogen. Subsequently, a lysis buffer containing 10 M urea and the cOmplete ™ protease inhibitor cocktail (Roche, #11697498001) was added. All the crystal and liquid were transferred into 2 ml tube upon scraper and store at −80 °C for next sample preparation. Cells in this lysis buffer were thawed and a VialTweeter device (Hielscher-Ultrasound Technology) was used to sonicate the samples (4 °C; 1 min; two cycles). Upon sonication, the samples were centrifuged at 20,000 g for 1 hour to remove all the insoluble material. Protein concentration was measured using the Bio-Rad protein assay dye (Bio-Rad, cat. no. 5000006). Reduction and alkylation were carried out using 10 mM Dithiothreitol (DTT) for 1 hour at 56°C, followed by 20 mM iodoacetamide (IAA) in darkness for 45 minutes at room temperature. A precipitation-based digestion method was used here 112 – 114 . Briefly, five volumes of precooled precipitation solution (50% acetone, 50% ethanol, and 0.1% acetic acid) were added to the sample vortex 30 s. After overnight incubation at −20 °C, the samples were centrifuged (20,000 x g; 4 °C; 40 min). The precipitate was washed with precooled 100% acetone, centrifuged (20,000 × g; 4 °C; 40 min), and the remaining acetone was evaporated in a SpeedVac. For protein digestion, 300 μL of 100 mM NH4HCO3 with sequencing grade porcine trypsin (Promega) at a trypsin-to-protein ratio of 1: 20 were added and incubated overnight at 37 °C. The resulting peptide samples were acidified with formic acid and desalted using a C18 column (Macro Spin Columns, NEST Group INC.) according to the manufacturer’s instructions. The peptide concentration was assayed by nanodrop, 1 μg peptide was injected into mass spectrometry. The same DIA method as described above was used for the data acquisition. The Spectronaut software was used for data analysis (see below). QUANTIFICATION AND STATISTICAL ANALYSIS TMT Data Procession and Analysis For TMT data procession and analysis, the JUMP search engine 115 transformed peptide and protein identification by combining pattern-based scoring and de novo tag scoring, significantly improving the accuracy of peptide-spectrum matches (PSMs). This innovation utilized MS/MS raw data and a composite target/decoy database; a concept introduced to estimate false discovery rates (FDR) 116 . The protein database was generated by combining downloaded Swiss-Prot, TrEMBL, and UCSC databases and removing redundancy 117 . The database additionally included customized protein entries 118 containing common protein contaminants and missense SNP fragments of mouse from UCSC (using the translated sequence of 100-bp upstream and downstream nucleotide sequences), the identification of which were excluded during the generation of Tissue-PPT (see below). Generating a decoy database involved reversing target protein sequences and merging them with the accurate target database. FDR calculations assumed uniform mismatch distribution 115 . Mass tolerances for precursor and fragment ions were set to 15 ppm and 20 ppm, respectively. Up to two missed cleavage sites were permitted per peptide. TMTpro labeling at Lys or the N-terminus, along with Cys carbamidomethylation, were defined as static modifications, while Met oxidation was treated as a dynamic modification. Protein FDR was maintained below 1% by applying filters based on mass accuracy and JUMP-based matching scores (Jscore and ΔJn) 115 . Following the rule of parsimony, peptides shared by multiple proteins were generally assigned to the canonical protein form. In cases where no canonical form was defined in the database, the peptide was assigned to the protein with the highest peptide-spectrum match (PSM) count. The identified PSMs, peptides, and proteins were quantified using TMT reporter ions in the MS2 scans 34 . To correct for ratio compression in reporter-based quantification in the MS2 scans, we used fully SILAC-labeled mouse tissues, generated over two generations 55 , as negative controls to detect noise signals. For light Lys peptides, the noise detection process involved the following steps: (i) experimental TMT ions were extracted for each PSM; (ii) the abundance in the channel of the fully labeled SILAC tissues was considered noise; (iii) this noise was subtracted from all TMT channels; and (iv) PSM data were summarized into peptide and protein data. For heavy Lys peptides, we applied a similar approach, using the channel of unlabeled mouse tissues as a negative control to detect and remove noise. DIA Data Procession and Analysis The DIA-MS data analyses were performed using Spectronaut version 16 45 , 119 . All the raw DIA-MS datasets were firstly used for the library generation by the Pulsar search of Spectronaut 120 against the same mouse fasta sequence database used in TMT mass spectrometry. For the pSILAC DIA library generation, the labels were specified in the “Labeling” setting, the “Labeling Applied” option was enabled, the “Lys6” were specified as SILAC labeling in the second channel, and the “In-Silico Generate Missing Channels” and “Label” of Workflow setting was selected. Methionine oxidation was set as variable modification and cysteine carbamidomethylation was selected as fixed modification. For the phosphoproteomic data, the phosphorylation at S/T/Y was enabled as variable modification. For the targeted data extraction of pSILAC datasets and for subsequent identification and quantification, the Inverted Spike-In workflow (ISW) was used as described previously 120 . The “Qvalue” was selected for all data filtering. Both peptide and protein FDR cutoffs were controlled at 1%. For the phosphoproteomic data, the probability of PTM cutoff was strictly kept at >0.75 to ensure the phosphosites were localized 121 , similar to Class I confidence 122 , 123 . The PTM score >0.01 table was also exported from the Spectronaut 22 , 112 and then filtered by the PTM score >0.75 result for the following data analysis. The number and turnover of phosphosites were summarized based on the unique phosphopeptide level results (i.e., the phosphopeptides with multiple phosphorylation were counted multiple times). The pSILAC plasma dataset was analyzed separately due to the low number of plasma protein identities which may potentially impact the overall protein FDR control. The unlabeled tissues data (days 0) were analyzed by directDIA algorithm on Spectronaut 16. All the other Spectronaut settings were kept as Default. The quantitative results for heavy and light peptide precursors were exported by Spectronaut for following turnover calculations (see below). For the protein lifetime calculation in JUMPt 16 , the peptide heavy-to-light ratio was initially filtered based on a time-dependent increase (e.g., 32d > 8d), then summarized to the protein level. The protein iBAQ value 44 was directly exported from Spectronaut and then transformed into riBAQ values in the ppm scale 46 . For the phosphopeptide turnover calculation in JUMPt, both phosphopeptides and their corresponding non-phosphopeptides (with the same naked peptide sequence) were analyzed using the same free lysine turnover curve and modeling. Computational Procedure to Obtain Free Lysine Decay Curve For both BoxCarmax-DIA and TMTpro results, we used the JUMPt pipeline 16 to calculate protein and phosphopeptide half-lives, utilizing setting 2 in JUMPt, which incorporates the free Lys decay curve and protein turnover data to fit an ordinary differential equation (ODE) -based model to determine protein degradation rates. During the Lys-based pSILAC analysis of animals, the labeling process is significantly affected by free Lys recycled from protein degradation. Therefore, it is critical to obtain the free Lys decay curve during the labeling process, which is used as input in the JUMPt program 16 . In all BoxCarmax-DIA, DIA-MS, and TMTpro results, a small portion of the peptides contained two Lys residues (i.e., double-K-peptides). These double-K-peptides were used to derive the Lys decay curve. Briefly, the double-K-peptides should exhibit three peaks: light, mixed, and heavy. If each protein ( P ) has a synthesis rate ( S ) following zero-order kinetics, and H A % represents the average percentage of heavy amino acid (e.g. K ) over time ( t ) , we can derive the following: P M ( mixed ) = S × t × H A % × ( 1 - H A % ) × 2 ( as two Lys have an equal probability of being heavy ) P H ( heavy ) = S × t × H A % × H A % The ratio of the mixed peptide to the heavy peptide variable modification R ( = P M P H ) = ( 1 - H A % ) × 2 H A % was independent of synthesis rates. Thus, H A % = 2 2 + R L A % (the average percentage of light amino acid) = 1 - H A % Using these equations, we derived the average L A % during the pulse (e.g., eight days) from double-K-peptides. Our ODE model assumes that protein synthesis and degradation processes are overall balanced, ensuring stable proteome homeostasis in adult mouse tissues. This assumption is supported by the observation that adult mice maintained a stable body weight over the one-month pSILAC labeling period. However, as previous studies using the same labeling strategy, this equilibrium may not hold true for all proteins. Specific subsets, such as those with cyclic expression patterns driven by circadian rhythms, may exhibit distinct dynamics under these conditions, whose protein turnover would require much shorter labeling durations and enrichment strategies. Mouse Tissue Protein and Phosphopeptide Half-life Data Filtering and Assembling To generate the final PT data in the Tissue-PPT, both BoxCarmax-DIA and TMT data were further filtered and combined based on the following rules at the protein level: (i) For proteins with a half-life of at least 0.5 days, the multiplex-DIA BoxCarmax results were used as the primary data for reporting the final PT, supplemented by TMT method results. This means that protein PTs not successfully measured by DIA-MS but measured by the TMT data will be kept by taking the TMT-derived results to increase the proteomic PT coverage. This simple combination was performed due to the observed high consistency between DIA(x) and TMT(y) derived PTs (y=0.9585x, if we force the data to fit a y=kx curve, generating R 2 = 0.9729). Based on this high level of agreement and the purpose of analyzing PA and PT data together (PA datasets were measured by DIA-MS only), we also decided to directly keep the DIA-derived values for these proteins which were successfully measured by DIA-MS (instead of averaging DIA- and TMT-derived PTs), because DIA may provide more MS2-level fragmental ions for pSILAC quantification which might be helpful in improving quantitative single-to-noise ratios 102 ; all the raw datasets are made public (see below) for alternative processing strategies. (ii) For proteins with a half-life of less than 0.5 days (only 0.1–0.5% of the total results), which is more challenging in both DIA and TMT methods due to the low heavy signals and our labeling time course, we filtered the data with CV of less than 0.3 for both DIA and TMT data and then averaged them to generate the results. A total of 11,171 unique protein group entries were measured with PT values including protein variants (2.48%), plasma proteins, and hippocampus (TMT data only) and were all assembled in this report. For the phosphopeptide half-life analysis, only DIA-MS measurements were performed and the same JUMPt pipeline was used to calculate the half-life times. Replicates with a CV greater than 0.8 and half-life values of less than 0.5 days were excluded from the final analysis and excluded in Tissue-PPT App. To compare the turnover differences between phosphorylated and non-phosphorylated peptides, the total proteome data obtained using the 4-hour regular DIA method was utilized for peptide half-life calculations, employing the same pipeline. Data analysis was performed by precisely matching peptide backbones 22 . Protein Abundance and Synthesis Rate Estimation The protein iBAQ values were exported from Spectronaut using the default iBAQ export function based on the total proteome 4-hour DIA results for estimating absolute protein abundance. Data with Log 2 -transformed values less than 5 was filtered out. Relative iBAQ (riBAQ, in ppm) values 46 were calculated by dividing each value by the sum of tissue iBAQ values and then multiplying by one million. During visualization, outlier Log 2 (riBAQ) values less than −1 and greater than 9 were labeled as < −1 and > 9, respectively in the Tissue-PPT web app. The proxy of protein-specific protein synthesis rate ( K s y n ) was estimated as below 5 , 124 to qualify the RPF data: K s y n = l n ( 2 ) T 50 × i B A Q Where T 50 and i B A Q denote the protein-specific P T and i B A Q values. Kinase Substrate Mapping and Annotation For kinase-substrate mapping, the mouse phosphopeptide sequence was analyzed using the PTMoreR ( https://github.com/wangshisheng/PTMoreR ) 125 , which aligned the mouse phosphopeptide with human sequences with a 14-amino-acid window (Window similarity score > 14 i.e., the full match). Additionally, the resulting human sequences was used to retrieve kinase substrates from the kinase library 126 ( https://kinase-library.mit.edu/sites ) with the percentile threshold 99%. For the visualization of correlation between phosphopeptide and non-phosphopeptide for kinases, the number of kinase substrates more than 30 was used for visualization. Ubiquitin Linkage Identification and Turnover Quantification To search and determine the different type of ubiquitin chains via its lysine residue (i.e., the “ubiquitin code’), a “Gly-Gly” or diGly modification was set up as a variable modification in a separated search and summarized, as previously described 49 . The DIA raw data was directly searched by setting K-GlyGly as variable modification in Spectronaut with the PTM score > 0.75. The quantitative data from Spectronaut was reported as described above for the phosphoproteomics analysis and ISW workflow was also used 127 . The search results were further manually inspected. The quantities of K6-, K11-, K27-, K48-, and K63- were inferred based on most abundant peptide precursors for MQIFVK GG TLTGK (K6), TLTGK GG TITLEVEPSDTIENVK (K11), TITLEVEPSDTIENVK GG AK (K27), LIFAGK GG QLEDGR (K48), and TLSDYNIQK GG ESTLHLVLR (K63), reported in the peptide-level report from Spectronaut. To estimate the relative quantitative variability between K6-, K11-, K27-, K48-, and K63- linked chains, the above peptides carrying diGly were additionally normalized to the adjacent unmodified counterpart peptide with no miss-cleavage (TLSDYNIQK was used for K48 and K63, and TITLEVEPSDTIENVK was used for K11, K27 and K6). The heavy-to-light ratio of ubiquitin diGly peptide LIFAGK GG QLEDGR (K48), and TLSDYNIQK GG ESTLHLVLR (K63) and unmodified ubiquitin peptide TLSDYNIQK were also exported from Spectronaut to determine the respective turnover kinetics. Protein-protein Interaction (PPI) Mapping and Comparison We used different public databases to evaluate the PT to PPI correlation and PA to PPI correlation. The Tissue-PPI data was downloaded from Skinnider et al 62 which was based on protein abundance correlation profiling coupled with mass spectrometry (PCP-MS) in multiple mouse tissues. Databases of hu.MAP 67 ( http://humap2.proteincomplexes.org/ ), CORUM 63 ( https://mips.helmholtz-muenchen.de/corum/ ), and Bioplex 3.0 64 ( https://bioplex.hms.harvard.edu/interactions.php ) were downloaded and compiled separately for PPI mapping based on gene symbols. Pearson correlations between proteins participating in database-matched PPI pairs (PPI-in) were calculated based on PA and PT levels respectively measured across mouse tissues in our study. Proteins not associated with any PPIs (PPI-out, or no-PPIs) in these databases were used as the corresponding reference group, and their Pearson correlations were determined by random pairing of proteins. In addition to random protein pairs as negative controls, we further filtered these random protein pairs by requesting them to be participating in the same pathways, as an additional confirmatory analysis. In this analysis, the pathway enrichment was first performed using the DAVID Functional Analysis Tool for KEGG annotation. KEGG pathways with a “pathway” in their item names and a gene count number higher than 49 were selected for visualization. In the next step, for a particular pathway, we request at least one of the proteins in a PPI (in Tissue-PPI 62 ) belonging to this pathway for calculating PA and PT correlation between the PPI pairs (PPI-in), while both proteins forming a random-pair (not listed in Tissue-PPI) have to be annotated in the same pathway to be treated as “PPI out” as the corresponding control. The hu.MAP confidence levels (Level 5–1 indicating Extremely High, Very High, High, Medium High, and Medium) were additionally used to group the PPI pairs for comparing the Pearson correlations of PA and PT across tissues between PPI groups. To evaluate the predictive power of PPIs based on PA and PT’s cross-tissue correlations, Receiver Operating Characteristic (ROC) curves were generated with the corresponding Area Under the Curve (AUC) computed 128 . In this analysis, the “positive” PPI pairs were retrieved from the Extremely High and High Levels in hu.MAP database. The same number of “false” PPIs was randomly generated from any protein pairs excluding those pairs listed in any of hu.MAP, CORUM, Bioplex 3.0 and Tissue-PPI database. This ROC analysis is based on cross-tissue correlation coefficient of PAs and PTs and therefore was not used to predict tissue-specific PPIs. The logistic regression model was employed to evaluate the combined predictive power of PA and PT using Scikit-Learn 129 , employing default parameters. To investigate the PT difference between PPI pairs for protein examples in each tissue, the PPI network was generated by Cytoscape v3.10 based on Tissue-PPI. The relative PT value difference (delta T 50 in days) was visualized using partner proteins’ T50 subtracted by the central node protein’s T50 measured in the corresponding tissue. The whole brain PPI in Tissue-PPI was used as reference for our brain regions’ dataset. According to Tissue-PPI results 62 . The PPI types were classified as three groups in our analysis: (i) Tissue unique PPI: PPIs only be identified in one (this) specific tissue but not in other tissues, (ii) PPI identified in this tissue: PPIs identified in this specific tissue regardless of their identification in other tissues; (iii) PPI identified in any tissue: PPIs included in Tissue-PPI but may not be identified in this tissue. Protein Degradation Related Proteins To explore the protein degradation process, we mapped our PA and PT data to key protein removal cellular components, including the proteasome, lysosome, E3 ubiquitin ligases, E3 accessory proteins, deubiquitinating enzymes (DUBs), and chaperone proteins. The lists of E3 ubiquitin ligases, E3 ubiquitin ligase accessory proteins, and DUBs were downloaded from NIH ( https://esbl.nhlbi.nih.gov/Databases/KSBP2/Targets/Lists ). The list of molecular chaperons was downloaded from literature 130 . The variability of PA and PT of these proteins was calculated across tissues by standard deviation S.D. of [Log 2 (riBAQ of each protein)- Log 2 (riBAQ of averaged level)] for each protein list. Tissue Enriched Proteins Similar to Human Protein Atlas (HPA) 131 which defines tissue-enriched proteins as those with mRNA levels at least fourfold higher in a specific tissue compared to any other tissues, we defined tissue-enriched proteins to be those proteins with at least fourfold higher abundance in a specific tissue compared to any other tissues based on our DIA-MS derived protein intensities. mRNA-seq and Ribo-seq Data in Multiple Tissues The mRNA-seq and Ribo-seq data (RPF) of adult wild-type C57BL/6 mouse tissues were downloaded from a published paper via NCBI Gene Expression Omnibus under accession number GSE94982 ( https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE94982 : GSE94982_P42_RNA-seq_exon_level_tpm.txt.gz and GSE94982_P42_Ribo-seq_exon_level_tpm.txt.gz). All RPF and RNA-seq samples were combined and the transcripts per kilobase million (TPM) values >1 was filtered for mapping to the proteomic data. Spearman Correction In our study, correlation analyses were conducted using both Spearman and Pearson correlations while the cross-omics correlations were conducted using Spearman correlations. The Spearman correction, also known as the attenuation correction, is a statistical technique introduced by Spearman to adjust correlations for the effects of noise-induced measurement error 132 . As for the cases inferring respective PA and PT correlations between tissues for heatmap visualizations, the correlation coefficients ( R for Pearson correlation and rho for Spearman correlation) were further subsequently adjusted using the Spearman correction considering the experimental replicates 133 134 for the finally reported PA and PTs at proteome and phosphoproteome levels. The correlation matrixes are then visualized by pheatmap in R software. Other Bioinformatics Analysis Most data visualization was performed in R and GraphPad Prism version 10 (GraphPad Software, San Diego, California USA). The following R packages were used to visualize the data: ‘ggplot2’ (boxplots, density plots, volcano plots and histograms), ‘factoextra’ (principal component analysis), ‘pheatmap’ (heatmaps), ‘corrplot’ (correlation plots) and ‘UpsetR’ (UpSet plots). The Cytoscape v3.10.1 135 was used for the PPI plots. The Perseus software and its default GO annotation reference file 136 was used for the protein Gene Oncology Cellular Component (GO CC) and Gene Oncology Biological Process (GO BP) annotation with the mouse species. This annotation step allows proteins to be located in more than one cellular component or participating more than one biological processes. The 2D enrichment function in Perseus 137 was used to generate the data for the bubble plots. The protein-level functional annotation was also performed using Metascape ( https://metascape.org/ ) 138 . BioMart ( https://useast.ensembl.org/info/data/biomart/index.html ) was used to map the gene symbols between human and mouse species. Figure 1A , 2F , 3A and the Graphic Abstract were generated with the assistance of BioRender. The DAVID Functional Analysis 139 website was used for the KEGG pathway enrichment analysis. The GO enrichment was conducted on Metascape ( https://metascape.org/gp/index.html#/main/step1 ) with Express Analysis tool for summarizing the functions of short- and long- lived proteins. In Figure 2D , the “%InGO” (#GeneInGOAndHitList / #GeneInHitList) was used for selection and all the 35 functional classes with “%InGO” >5% among the Top 100 list generated by Metascape were presented 138 . We compared our tissue lifetime data with the findings from Hasper et al 13 and others based on mouse gene symbols. Tissue-PPT Website Inventory The website of Tissue-PPT ( https://yslproteomics.shinyapps.io/tissuePPT/ ) was created by Shiny framework (version 1.9.1) in R environment (version 4.4.1) and deployed on the shinyapps.io platform ( https://www.shinyapps.io/ ) to facilitate navigation of the database. Note the plasma proteomic datasets and the results of missense SNP peptide fragments (277 identifications) were included in the results and supplementary tables but excluded in Tissue-PPT website. The website of Tissue-PPT interactively provides queries about protein/phosphosite abundance and lifetime in various tissues, based on inquiries of matched gene symbols (human gene symbol supported while this is a mouse database) and protein Uniport accessions. It offers four major functions (a) including Heat-circle (HC) plot across tissues, (b) Heatmap analysis for protein sets, (c) Protein-specific bar plots, and (d) Correlation analysis between molecular layers for individual proteins or protein sets of interest, as well as convenient options to download all the resultant figures and tables, all with detailed user manual and instruction provided in the website. Particularly, for HC plot, the options of riBAQ (in ppm), iBAQ, and MS-intensity were enabled for the scale indicating protein expression. For the PT visualization, because the tissues may have basic turnover differences, Tissue-PPT further enables the option to visualize the rank-percentage (0–100% for the fast-to-slow turnover rates) in each tissue and a following-up option allowing the user to perform the row-scaled visualization on the current list of proteins. The website also supports local installation using the source code which is made public. Supplementary Material 1 Figure S1. Establishing a reproducible and in-depth proteome turnover atlas across mouse tissues and brain regions, related to Figure 1 (A) Coefficient of variation (CV) distribution of proteome-wide Heavy/Light (H/L) ratios in analyzing the mouse whole brain tissue with five SILAC labeling time points of 0, 4, 8, 16, and 32 days. Among the five labelled mice, three biological replicates were used for TMTpro-based quantification for this analysis. (B) The quantitative slopes of the log-transformed H/L ratios [i.e., ln (H+L)/L] growing over time are highly correlated between the experiments of 5 time points (0d, 4d, 8d, 16d, 32d) and 3 time points (0d, 8d, 32d). The slope primarily determines the protein turnover rates. Rho , Spearman correlation coefficient; R, Pearson R correlation coefficient. (C) Correlation (Spearman and Pearson) of protein lifetimes between two biological replicates measured by BoxCarmax-DIA. (D) Correlation plot of PTs across the proteome, as derived by DIA and TMT methods (same as Figure 1B but with the forced fitting to y= k x regression), R, Pearson R correlation coefficient. This plot demonstrates the strong correlation and comparable PTs determined by the two MS platforms after respective data procession (see Methods ). (E) Venn diagram comparing the protein coverage of DIA and TMT datasets. (F) Venn diagram comparing the turnover coverage of mouse liver proteome between this study and Hasper et al. (G) Scatterplot displaying the comparison of protein abundance (indicated by DIA-MS intensity in this study) and lifetime for the liver proteome between this study and Hasper et al. (red dots). (H) Correlation (Spearman and Pearson) of T 50 results between the two studies for the mouse liver proteome. (I) The boxplot displaying the proteome turnover coverage for the heart and liver proteomes measured by this study and Hasper et al. NIHMS2066478-supplement-1.pdf (1.1MB, pdf) 2 Figure S2. Short-lived and long-lived proteins and their associated functional processes across tissues, related to Figure 2 (A) The correlation of Log 2 (T 50 ) for the same proteins identified and measured in plasma and tissues. For each of the plasma proteins, its x axis denotes the PT measured in plasma while y axis represents PT in other tissues. The dashed line indicates the diagonal line of y=x (i.e., identical PT between tissue and plasma), which divides the proteins into two groups (i.e., PT lower or higher in plasma). The number and percentage of proteins in each of these two groups are shown, suggesting the PTs of most plasma proteins are generally higher than in plasma than other tissues (i.e., the majority of the proteins are below the diagonal line). For example, 76.8% of the liver proteins overlapping to the plasma proteome have a longer PT in blood plasma. (B) Upset plots of short-lived and long-lived proteins in each tissue. The top 5% shortest- and longest-lived proteins in each tissue were compared between tissues. (C) STRING Protein-protein interaction network (medium confidence) of 49 proteins that are the top 5% most long-lived in each of the nine brain regions. The Enrichment P values were reported by STRING analysis on the protein clusters. (D) Venn diagrams comparing short-lived and long-lived protein identities in brain regions and non-brain tissues. (E) Metascape ( https://metascape.org )-derived GO enrichment analysis (Top100 items) from short-lived and long-lived proteins in brain regions and non-brain tissues. The Enrichment P values were reported by Metascape analysis. NIHMS2066478-supplement-2.pdf (3MB, pdf) 3 Figure S3. Tissue-PPT App supports the functional discovery of proteins and protein sets by visualizing both protein abundance (PA) and protein lifetime (PT) and their relationships across tissues, related to Figure 2 and 4 (A) The Shiny Tissue-PPT App ( https://yslproteomics.shinyapps.io/tissuePPT/ ) was established for accessing and analyzing both PA and PT data with examples shown. The App includes the five major functions: Heat-circle (HC) plot across tissues, Heatmap analysis for protein sets, Protein-specific bar plots, Correlation analysis at the protein level, and Correlation analysis for protein sets. HC plot examples of protein-protein interaction partners of PSMD1, LMNA, and AK2 proteins in across tissues (see Figure 4E ). For the HC plot, the blue-to-red color gradient denotes PT from long to short. The size of the HC plot circles is proportional to the Log 2 (riBAQ) in ppm value indicating PA. Triangle: Log 2 (riBAQ) < −1, Diamond: Log 2 (riBAQ) > 9. (B) Heatmap examples of both PA and PT generated for proteasome subunits (19S and 20S) and Mitochondrial ribosomal subunits (small/large) across tissues. The scale of PA and PT are kept consist across proteome facilitating direct comparison between individual proteins and protein lists. In this case, PA of proteosome subunits is much higher than that of mitochondrial ribosomal subunits, while both complexes are long-lived. The orange-to-green color bar and red-to-blue color bar denote the small-to-large PA and PT values (Log 2 -transformed), respectively. NIHMS2066478-supplement-3.pdf (1.6MB, pdf) 4 Figure S4. HC plots for proteostasis-associated proteins and the correlation between PT and PPI resolved to biological pathways, related to Figure 3 and 4 (A-C) HC plots of protein folding chaperones, E3 accessory proteins, and deubiquitinating enzymes (DUBs) across tissues. (D) Heatmap of Pearson R for PA and PT between PPI partners based on KEGG pathways. In addition to random protein pairs as negative controls, here we further filtered these random protein pairs by requesting them to be participating in the same pathways (see Methods ). KEGG pathways with a “pathway” in their item names and a gene count number higher than 49 were shown. PPI list was derived from Tissue-PPI. Both proteins forming a random-pair (not listed in Tissue-PPI) have to be annotated in the same pathway to be treated as “PPI out” or “no PPI”. NIHMS2066478-supplement-4.pdf (765.4KB, pdf) 5 Figure S5. Multi-omic analysis of peroxisome and pexophagy proteins, related to Figure 5 (A) Spearman correlation between RPF (the sequencing of ribosome protected fragments) compared with MS-derived K syn (proxy of protein synthesis rate) and iBAQ values for five tissues. P-values were calculated using Wilcoxon matched-pairs signed rank test (One sided). Please refer to method section for K syn calculation. The RPF data was measured by ribosome sequencing. (B) Spearman correlation of Log 2 RFP to K syn (upper panel) compared with Log 2 iBAQ (lower panel) in Lung tissue as an example. (C) Boxplots of protein-specific Spearman correlation rho values between multi-omic layers for all measured proteins (upper panel) and tissue-enriched proteins (lower panel, see Methods ) based on DIA- (left panel) and TMT- (right panel) derived results. (D) Heatmap of individual peroxisome proteins among five tissues using the quantitative values from mRNA, RPF, PA, and PT layers. The brain results were determined by averaging all brain regions. The heatmap was created by the R pheatmap package with the row sale. Scale bar: red color is higher value per layer; blue color is lower value. (E) Heatmap visualizing the cross-tissue Spearman correlation of each peroxisome protein between two omics layers across tissues as indicated. The blue-to-red scale bar denotes the increasing Spearman correlation. (F) Boxplots exploring the relative PTs of peroxisome proteins with those of other proteins in the proteome in different tissues. The relative PT was calculated by dividing the PT by the average PT in five tissues. P , Wilcoxon test. (G) The negative correlation between the abundances of pexophagy-associated proteins and the lysosome protein levels. Plots were generated by the “Correlation analysis: protein sets” function in Tissue-PPT . Pearson correlation R and the significance of correlation (P value, based on cor.test in R) are shown. (H) Correlation between Log 2 (Protein abundance) and Log 2 (T 50 ) for protein examples of Pex3 and Atg12 involved in pexophagy. Plots were generated by the “Correlation analysis: proteins” function in Tissue-PPT . (I) HC plots of proteins involving in pexophagy-associated processes such as peroxisome designation and targeting, and autophagosome biogenesis. Plots were generated by the “Heat-circle (HC) plot across tissues” function in Tissue-PPT . NIHMS2066478-supplement-5.pdf (2.1MB, pdf) 6 Figure S6. Phosphoproteome turnover measurement across mouse tissues reveals the different T50 of the phosphorylated peptides and their non-phosphorylated counterparts, related to Figure 6 and 7 (A) Density plot of the quantified T 50 values of phosphorylation sites (P-sites) across mouse tissues based on all phosphopeptides measured. (B) Distribution of Spearman correlation between the T 50 of a phosphorylated (p) peptide and the T 50 of its non-phosphorylated (np) peptide counterpart across tissues and regions for all P-sites. The p and np peptides share the same amino acid sequence and only differ in the phosphorylation modification. (C) Density plot of DeltaT 50 (ΔT 50 ) that represents the difference between the T 50 of p peptide to the T 50 of its corresponding np peptide across tissues. (D) Venn diagram comparing the numbers of P-sites showing destabilization (T 50 of p peptide is shorter than T 50 of np peptide) or stabilization effects (T 50 of p peptide is shorter than T 50 of np peptide) in brain and non-brain areas. The significance is inferred by P <0.05 (Student’s t-test) and |fold change| >1.5 (in brain) and >1.2 (in non-brain tissues). (E) Heavy/light (H/L) ratio curves during the labeling process for a p peptide (green) and its np peptide counterpart (blue) of the same sequence as examples of long-lived proteins. NIHMS2066478-supplement-6.pdf (1.6MB, pdf) 7 Figure S7. Two dimensional (2D) functional enrichment analysis comparing proteome and phosphoproteome abundance and turnover variability across brain regions, related to Figure 6 (A) Bubble plot of abundance variability between proteome and phosphoproteome across brain regions. CV was calculated from phosphorylated protein abundance (y-axis, an average of all phosphosites) or protein abundance (x-axis) across all brain regions. 2D enrichment was performed using Perseus software and the bubble plot was visualized using ggplot2. Enriched GO terms were filtered with p-value <0.05 and number of proteins >5. The cycle size denotes the number of proteins in each GO term. The color scale represents the enriched GO P-value using Perseus analysis. (B) Bubble plot of lifetime variability between proteome and phosphoproteome across brain regions. CV was calculated from phosphorylated protein lifetime (y-axis, an average of all phosphosites) and protein lifetime (x-axis) across all brain regions. The cycle size denotes the number of proteins in each GO term. The color scale represents the enriched GO P-value. NIHMS2066478-supplement-7.pdf (699.3KB, pdf) 8 NIHMS2066478-supplement-8.xlsx (6.5MB, xlsx) 9 NIHMS2066478-supplement-9.xlsx (11.4MB, xlsx) Key resources table. REAGENT or RESOURCE SOURCE IDENTIFIER Antibodies Anti-FLAG rabbit Cell Signaling Cat# 14793 Donkey; anti-rabbit AF488 Jackson Immunoresearch Cat# 711-545-152 Biological samples C57BL/6J mice Jackson Laboratory N/A Chemicals, peptides, and recombinant proteins DMEM high glucose Gibco Cat# 10564011 Fetal bovine serum (FBS) Sigma Cat# F8318 DMEM for SILAC Thermo Scientific Cat# 88364 Dialyzed fetal bovine serum (FBS) Thermo Scientific Cat# 26400044 Penicillin/streptom ycin solution Gibco Cat# 15140122 Heavy L-Arginine-HCl (13C6 for SILAC) Cortecnet Cat# CCN250P1 Heavy L-Lysine-2HCl (13C6,15N2 for SILAC) Cortecnet Cat# CCN1800P1 Trypsin-EDTA (0.25%) Gibco Cat# 25200072 HPLC-grade water Thermo Scientific Cat# W64 Urea Sigma Cat# U5378 Halt ™ phosphatase inhibitor Thermo Scientific Cat# 78428 complete ™ protease inhibitor cocktail Roche Cat# 11697498001 Sequencing-grade modified trypsin Promega Cat# V5113 Lys-C Wako Cat# 12505061 LC-MS grade Acetonitrile Thermo Scientific Cat# 85188 Formic acid Thermo Scientific Cat# 85178 Trifluoroacetic acid, LC-MS grade Thermo Scientific Cat# 85183 Methanol J. T. Baker Cat# 9070-05 Ammonium hydroxide solution, 25% in H2O Sigma Cat# 05002-1L Acetic acid, glacial Thermo Scientific Cat# 9084-05 L-LYSINE [13C6, 99%] MOUSE FEED KIT Cambridge Isotopes Laboratories Cat# MLK-LYS-C 100% light lysine food Cambridge Isotopes Laboratories Cat# MLK-LYS-C Phosphatase inhibitor cocktail phosphoSTOP Roche Cat# 4906845001 Sep-Pak C18 column Waters Cat# WAT023590 High-Select ™ Fe-NTA kit Thermo Scientific Cat# A32992 Filter tip Axygen Cat# TF-20-L-R-S Bio-Rad protein assay kit Bio-Rad Cat# 5000002 Buffer A, 0.1% formic acid in water, LC-MS grade Thermo Scientific Cat# 85171 Buffer B, 0.1% formic acid in 80% acetonitrile, LC-MS grade Thermo Scientific Cat# LS122550 Ammonium bicarbonate Sigma Cat# A6141 TECP Thermo Scientific Cat# 20490 Iodoacetamide Sigma Cat# I1149 16 plex TMTpro Thermo Scientific Cat# A44520 Formic acid Thermo Scientific Cat# 28905 Autosampler MicroVials Thermo Scientific Cat# 03377299 XBridge C18 columns (3.0mmx 15 cm, 1.7 μm particle size) Waters N/A nanoLC column CoAnn Technologies, LLC Cat# HEB07505001718I ReproSil-Pur,120A, C18-AQ,1.9 um resin Dr. Maisch Cat# r119.aq PicoFrit LC-MS column New Objective Cat# PF360-75-10-N-5 FastDigest BshTI Thermo Scientific Cat# FD1464 SURE Competent Cells Agilent Cat# 200152 Lipofectamine 2000 Thermo Scientific Cat# 11668030 Benzoase nuclease Sigma Cat# 9025-65-4 Poly-L-lysine-coated Sigma Cat# 25988-63-0 5-fluoro-2’-deoxyuridine (FUdR) Sigma Cat# F0503 Cycloheximide Sigma Cat# 66-81-9 Ammonium chloride Sigma Cat# 12125-02-9 Bovine serum albumin Sigma Cat# 9048-46-8 Paraformaldehyde powder 95% Sigma Cat# 158127 Triton X-100 Sigma Cat# 9036-19-5 Anti-FLAG rabbit Cell Signaling Cat# 14793 Donkey; anti-rabbit AF488 Jackson Immunoresearch Cat# 711-545-152 96 Well glass bottom plates Cellvis Cat# P96-0-N Deposited data Raw mass spectrometry proteomics data This paper ProteomeXchange Consortium: PXD055942 Tissue-PPT This paper GitHub ( https://github.com/yslproteomics/tissuePPT ) Zenodo ( https://zenodo.org/records/14671026 ). Experimental models: Organisms/strains C57BL/6J mice Jackson Laboratory N/A Software and algorithms Spectronaut v16 Biognosys, Inc. N/A JUMP and JUMPt series Peng Lab https://github.com/JUMPSuite/JUMPt R (version 3.6.0) R Core Team https://www.r-project.org/ Perseus v2.1.1.0 Cox Lab https://www.maxauant.ora/Derseus/ Prism Graphpad v10 Graphpad Software, Inc. N/A ggplot2 package in R Wickham H. et al. https://ggplot2.tidvverse.org Biorender Biorender https://www.biorender.com/ LSD package in R Bjoern Schwalb https://www.rdocumentation.org/packages/LSD Cytoscape v3.10.1 Paul Shannon https://cvtoscape.org/ Pheatmap v1.0.12 Raivo Kolde https://www.rdocumentation.org/packages/pheatmap/versions/1.0.12/topics/pheatmap Corrplot v0.92 Taiyun Wei et al. https://www.rdocumentation.org/packages/corrplot/versions/0.92 factoextra v1.0.7 Alboukadel Kassambara et al. https://www.rdocumentation.org/packages/factoextra/versions/1.0.7/topics/fviz_pca UpsetR v1.4.0 Jake R Conway et al. https://www.rdocumentation.org/packages/UpSetR/versions/1.4.0/topics/upset Fiji v2.15.1 Johannes Schindelin at al. doi:10.1038/nmeth.2019 ; https://imagej.net/software/fiji/ PTMoreR Shisheng Wang et al. https://github.com/wangshisheng/PTMoreR Metascape Yingyao Zhou et al. https://metascape.org/gp/index.html#/main/step1 DAVID Functional Tool B.T. Sherman et al. https://davidbioinformatics.nih.gov/ BioMart Peter W Harrison et al. https://useast.ensembl.org/info/data/biomart/index.html Shiny 1.9.1 N/A https://www.shinvapps.io/ Other Orbitrap Tribrid Lumos Mass Spectrometer Thermo Scientific N/A Orbitrap Q Exactive HF Mass Spectrometer Thermo Scientific N/A Easy nLC 1200 system Thermo Scientific Cat# 1200 Fractionation HPLC Agilent 1260 N/A Column heater controller Sonation GmbH, Biberach Cat# PRSO-V1 BioTek Cytation 5 Cell microscope Agilent N/A SpeedVac Thermo Scientific Cat# SPD121 Nanodrop Thermo Scientific Cat# nanodrop 2000 Microplate Reader Biotek Epoch N/A ThermoMixer Thermo Scientific Cat# 5382000023 Open in a new tab Highlights: A high-quality, comprehensive resource of protein turnover across mouse tissues. Protein abundance and lifetime together define tissue proteome function and stability. Protein lifetime strongly correlates with protein-protein interactions in tissues. Site-specific phosphorylation functionally shapes in vivo protein lifetime. ACKNOWLEDGMENTS Y.L. thanks the support from the National Institute of General Medical Sciences (NIGMS), National Institutes of Health (NIH) through Grant R01GM137031 and RM1GM149406. J.P. is supported by RF1AG064909 and RF1AG068581 from NIH. EFF is supported by a CZI Collaborative Pairs Pilot Project Awards (Cycle 2; Phase 1). EFF also acknowledges the support of the SFB1286, Göttingen, Germany. We would like to thank Drs. Yun Jiao and Zhiping Wu for preparing the pSILAC mice, and Drs. Mandar Muzumdar and Ines Chen for their critical comments on the manuscript. Footnotes DECLARATION OF INTERESTS The authors declare no competing interests. SUPPLEMENTAL INFORMATION Figures S1 – S7 and Tables S1 and S2 Table S1. Protein abundance (PA, riBAQ), protein lifetime (PT) and MS-intensity (derived from BoxCarmax-DIA) determined for all protein entries across tissues and brain regions, related to Figure 1 Table S2. Phosphoproteomic abundance and lifetime results across tissues, related to Figure 6 Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. REFERENCES 1. Glass RD, and Doyle D (1972). On the measurement of protein turnover in animal cells. J Biol Chem 247, 5234–5242. [ PubMed ] [ Google Scholar ] 2. Pratt JM, Petty J, Riba-Garcia I, Robertson DH, Gaskell SJ, Oliver SG, and Beynon RJ (2002). Dynamics of protein turnover, a missing dimension in proteomics. Molecular & cellular proteomics : MCP 1, 579–591. [ DOI ] [ PubMed ] [ Google Scholar ] 3. Claydon AJ, and Beynon R (2012). Proteome dynamics: revisiting turnover with a global perspective. Molecular & cellular proteomics : MCP 11, 1551–1565. 10.1074/mcp.O112.022186. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 4. Ross AB, Langer JD, and Jovanovic M (2021). Proteome Turnover in the Spotlight: Approaches, Applications, and Perspectives. Molecular & cellular proteomics : MCP 20, 100016. 10.1074/mcp.R120.002190. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 5. Liu Y, Borel C, Li L, Muller T, Williams EG, Germain PL, Buljan M, Sajic T, Boersema PJ, Shao W, et al. (2017). Systematic proteome and proteostasis profiling in human Trisomy 21 fibroblast cells. Nature communications 8, 1212. 10.1038/s41467-017-01422-6. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 6. Liu Y, Beyer A, and Aebersold R (2016). On the Dependency of Cellular Protein Levels on mRNA Abundance. Cell 165, 535–550. 10.1016/j.cell.2016.03.014. [ DOI ] [ PubMed ] [ Google Scholar ] 7. Fornasiero EF, and Savas JN (2023). Determining and interpreting protein lifetimes in mammalian tissues. Trends Biochem Sci 48, 106–118. 10.1016/j.tibs.2022.08.011. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 8. McClatchy DB, Dong MQ, Wu CC, Venable JD, and Yates JR 3rd (2007). 15N metabolic labeling of mammalian tissue with slow protein turnover. Journal of proteome research 6, 2005–2010. 10.1021/pr060599n. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 9. Fornasiero EF, Mandad S, Wildhagen H, Alevra M, Rammner B, Keihani S, Opazo F, Urban I, Ischebeck T, Sakib MS, et al. (2018). Precisely measured protein lifetimes in the mouse brain reveal differences across tissues and subcellular fractions. Nature communications 9, 4230. 10.1038/s41467-018-06519-0. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 10. Alevra M, Mandad S, Ischebeck T, Urlaub H, Rizzoli SO, and Fornasiero EF (2019). A mass spectrometry workflow for measuring protein turnover rates in vivo. Nature protocols 14, 3333–3365. 10.1038/s41596-019-0222-y. [ DOI ] [ PubMed ] [ Google Scholar ] 11. Rolfs Z, Frey BL, Shi X, Kawai Y, Smith LM, and Welham NV (2021). An atlas of protein turnover rates in mouse tissues. Nature communications 12, 6778. 10.1038/s41467-021-26842-3. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 12. Kluever V, Russo B, Mandad S, Kumar NH, Alevra M, Ori A, Rizzoli SO, Urlaub H, Schneider A, and Fornasiero EF (2022). Protein lifetimes in aged brains reveal a proteostatic adaptation linking physiological aging to neurodegeneration. Sci Adv 8, eabn4437. 10.1126/sciadv.abn4437. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 13. Hasper J, Welle K, Hryhorenko J, Ghaemmaghami S, and Buchwalter A (2023). Turnover and replication analysis by isotope labeling (TRAIL) reveals the influence of tissue context on protein and organelle lifetimes. Molecular systems biology 19, e11393. 10.15252/msb.202211393. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 14. Rao NR, Upadhyay A, and Savas JN (2024). Derailed protein turnover in the aging mammalian brain. Molecular systems biology 20, 120–139. 10.1038/s44320-023-00009-2. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 15. Schwanhausser B, Busse D, Li N, Dittmar G, Schuchhardt J, Wolf J, Chen W, and Selbach M (2011). Global quantification of mammalian gene expression control. Nature 473, 337–342. 10.1038/nature10098. [ DOI ] [ PubMed ] [ Google Scholar ] 16. Chepyala SR, Liu X, Yang K, Wu Z, Breuer AM, Cho JH, Li Y, Mancieri A, Jiao Y, Zhang H, and Peng J (2021). JUMPt: Comprehensive Protein Turnover Modeling of In Vivo Pulse SILAC Data by Ordinary Differential Equations. Analytical chemistry 93, 13495–13504. 10.1021/acs.analchem.1c02309. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 17. Giansanti P, Samaras P, Bian Y, Meng C, Coluccio A, Frejno M, Jakubowsky H, Dobiasch S, Hazarika RR, Rechenberger J, et al. (2022). Mass spectrometry-based draft of the mouse proteome. Nature methods 19, 803–811. 10.1038/s41592-022-01526-y. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 18. Lu T, Qian L, Xie Y, Zhang Q, Liu W, Ge W, Zhu Y, Ma L, Zhang C, and Guo T (2022). Tissue-Characteristic Expression of Mouse Proteome. Molecular & cellular proteomics : MCP 21, 100408. 10.1016/j.mcpro.2022.100408. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 19. Aebersold R, and Mann M (2016). Mass-spectrometric exploration of proteome structure and function. Nature 537, 347–355. 10.1038/nature19949. [ DOI ] [ PubMed ] [ Google Scholar ] 20. Liu Y (2022). A peptidoform based proteomic strategy for studying functions of post-translational modifications. Proteomics 22, e2100316. 10.1002/pmic.202100316. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 21. Huttlin EL, Jedrychowski MP, Elias JE, Goswami T, Rad R, Beausoleil SA, Villen J, Haas W, Sowa ME, and Gygi SP (2010). A tissue-specific atlas of mouse protein phosphorylation and expression. Cell 143, 1174–1189. 10.1016/j.cell.2010.12.001. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. Wu C, Ba Q, Lu D, Li W, Salovska B, Hou P, Mueller T, Rosenberger G, Gao E, Di Y, et al. (2021). Global and Site-Specific Effect of Phosphorylation on Protein Turnover. Developmental cell 56, 111–124 e116. 10.1016/j.devcel.2020.10.025. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 23. Zecha J, Gabriel W, Spallek R, Chang YC, Mergner J, Wilhelm M, Bassermann F, and Kuster B (2022). Linking post-translational modifications and protein turnover by site-resolved protein turnover profiling. Nature communications 13, 165. 10.1038/s41467-021-27639-0. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 24. Hammaren HM, Geissen EM, Potel CM, Beck M, and Savitski MM (2022). Protein-Peptide Turnover Profiling reveals the order of PTM addition and removal during protein maturation. Nature communications 13, 7431. 10.1038/s41467-022-35054-2. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 25. Li W, Salovska B, Fornasiero EF, and Liu Y (2022). Toward a hypothesis-free understanding of how phosphorylation dynamically impacts protein turnover. Proteomics, e2100387. 10.1002/pmic.202100387. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 26. Venable JD, Dong MQ, Wohlschlegel J, Dillin A, and Yates JR (2004). Automated approach for quantitative analysis of complex peptide mixtures from tandem mass spectra. Nature methods 1, 39–45. 10.1038/nmeth705. [ DOI ] [ PubMed ] [ Google Scholar ] 27. Gillet LC, Navarro P, Tate S, Rost H, Selevsek N, Reiter L, Bonner R, and Aebersold R (2012). Targeted data extraction of the MS/MS spectra generated by data-independent acquisition: a new concept for consistent and accurate proteome analysis. Molecular & cellular proteomics : MCP 11, O111 016717. 10.1074/mcp.O111.016717. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 28. Gao E, Li W, Wu C, Shao W, Di Y, and Liu Y (2021). Data-independent acquisition-based proteome and phosphoproteome profiling across six melanoma cell lines reveals determinants of proteotypes. Mol Omics 17, 413–425. 10.1039/d0mo00188k. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 29. Li J, Van Vranken JG, Pontano Vaites L, Schweppe DK, Huttlin EL, Etienne C, Nandhikonda P, Viner R, Robitaille AM, Thompson AH, et al. (2020). TMTpro reagents: a set of isobaric labeling mass tags enables simultaneous proteome-wide measurements across 16 samples. Nature methods. 10.1038/s41592-020-0781-4. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 30. Liu D, Yang S, Kavdia K, Sifford JM, Wu Z, Xie B, Wang Z, Pagala VR, Wang H, Yu K, et al. (2021). Deep Profiling of Microgram-Scale Proteome by Tandem Mass Tag Mass Spectrometry. Journal of proteome research 20, 337–345. 10.1021/acs.jproteome.0c00426. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 31. Kruger M, Moser M, Ussar S, Thievessen I, Luber CA, Forner F, Schmidt S, Zanivan S, Fassler R, and Mann M (2008). SILAC mouse for quantitative proteomics uncovers kindlin-3 as an essential factor for red blood cell function. Cell 134, 353–364. 10.1016/j.cell.2008.05.033. [ DOI ] [ PubMed ] [ Google Scholar ] 32. Salovska B, Li W, Di Y, and Liu Y (2021). BoxCarmax: A High-Selectivity Data-Independent Acquisition Mass Spectrometry Method for the Analysis of Protein Turnover and Complex Samples. Analytical chemistry 93, 3103–3111. 10.1021/acs.analchem.0c04293. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 33. Yu K, Wang Z, Wu Z, Tan H, Mishra A, and Peng J (2021). High-Throughput Profiling of Proteome and Posttranslational Modifications by 16-Plex TMT Labeling and Mass Spectrometry. Methods Mol Biol 2228, 205–224. 10.1007/978-1-0716-1024-4_15. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 34. Niu M, Cho JH, Kodali K, Pagala V, High AA, Wang H, Wu Z, Li Y, Bi W, Zhang H, et al. (2017). Extensive Peptide Fractionation and y(1) Ion-Based Interference Detection Method for Enabling Accurate Quantification by Isobaric Labeling and Mass Spectrometry. Analytical chemistry 89, 2956–2963. 10.1021/acs.analchem.6b04415. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 35. Geyer PE, Holdt LM, Teupser D, and Mann M (2017). Revisiting biomarker discovery by plasma proteomics. Molecular systems biology 13, 942. 10.15252/msb.20156297. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 36. Lau E, Cao Q, Lam MPY, Wang J, Ng DCM, Bleakley BJ, Lee JM, Liem DA, Wang D, Hermjakob H, and Ping P (2018). Integrated omics dissection of proteome dynamics during cardiac remodeling. Nature communications 9, 120. 10.1038/s41467-017-02467-3. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 37. Sato C, Barthelemy NR, Mawuenyega KG, Patterson BW, Gordon BA, Jockel-Balsarotti J, Sullivan M, Crisp MJ, Kasten T, Kirmess KM, et al. (2018). Tau Kinetics in Neurons and the Human Central Nervous System. Neuron 97, 1284–1298 e1287. 10.1016/j.neuron.2018.02.015. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 38. Zhang Y, Wu KM, Yang L, Dong Q, and Yu JT (2022). Tauopathies: new perspectives and challenges. Mol Neurodegener 17, 28. 10.1186/s13024-022-00533-z. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 39. Li Q, Chang Z, Oliveira G, Xiong M, Smith LM, Frey BL, and Welham NV (2016). Protein turnover during in vitro tissue engineering. Biomaterials 81, 104–113. 10.1016/j.biomaterials.2015.12.004. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 40. Di Camillo B, Puricelli L, Iori E, Toffolo GM, Tessari P, and Arrigoni G (2023). Modeling SILAC Data to Assess Protein Turnover in a Cellular Model of Diabetic Nephropathy. Int J Mol Sci 24. 10.3390/ijms24032811. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 41. Polo SE, and Jackson SP (2011). Dynamics of DNA damage response proteins at DNA breaks: a focus on protein modifications. Genes Dev 25, 409–433. 10.1101/gad.2021311. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 42. Trulsson F, Akimov V, Robu M, van Overbeek N, Berrocal DAP, Shah RG, Cox J, Shah GM, Blagoev B, and Vertegaal ACO (2022). Deubiquitinating enzymes and the proteasome regulate preferential sets of ubiquitin substrates. Nature communications 13, 2736. 10.1038/s41467-022-30376-7. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 43. Prus G, Satpathy S, Weinert BT, Narita T, and Choudhary C (2024). Global, site-resolved analysis of ubiquitylation occupancy and turnover rate reveals systems properties. Cell 187, 2875–2892 e2821. 10.1016/j.cell.2024.03.024. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 44. Wilhelm M, Schlegl J, Hahne H, Gholami AM, Lieberenz M, Savitski MM, Ziegler E, Butzmann L, Gessulat S, Marx H, et al. (2014). Mass-spectrometry-based draft of the human proteome. Nature 509, 582–587. 10.1038/nature13319. [ DOI ] [ PubMed ] [ Google Scholar ] 45. Bruderer R, Bernhardt OM, Gandhi T, Xuan Y, Sondermann J, Schmidt M, Gomez-Varela D, and Reiter L (2017). Optimization of Experimental Parameters in Data-Independent Mass Spectrometry Significantly Increases Depth and Reproducibility of Results. Mol Cell Proteomics 16, 2296–2309. 10.1074/mcp.RA117.000314. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 46. Shin JB, Krey JF, Hassan A, Metlagel Z, Tauscher AN, Pagana JM, Sherman NE, Jeffery ED, Spinelli KJ, Zhao H, et al. (2013). Molecular architecture of the chick vestibular hair bundle. Nat Neurosci 16, 365–374. 10.1038/nn.3312. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 47. Toyama BH, Savas JN, Park SK, Harris MS, Ingolia NT, Yates JR 3rd, and Hetzer MW (2013). Identification of long-lived proteins reveals exceptional stability of essential cellular structures. Cell 154, 971–982. 10.1016/j.cell.2013.07.037. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 48. Verzijl N, DeGroot J, Thorpe SR, Bank RA, Shaw JN, Lyons TJ, Bijlsma JW, Lafeber FP, Baynes JW, and TeKoppele JM (2000). Effect of collagen turnover on the accumulation of advanced glycation end products. J Biol Chem 275, 39027–39031. 10.1074/jbc.M006700200. [ DOI ] [ PubMed ] [ Google Scholar ] 49. Ba Q, Hei Y, Dighe A, Li W, Maziarz J, Pak I, Wang S, Wagner GP, and Liu Y (2022). Proteotype coevolution and quantitative diversity across 11 mammalian species. Sci Adv 8, eabn0756. 10.1126/sciadv.abn0756. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 50. Nguyen JA, and Yates RM (2021). Better Together: Current Insights Into Phagosome-Lysosome Fusion. Front Immunol 12, 636078. 10.3389/fimmu.2021.636078. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 51. Zuehlke AD, Beebe K, Neckers L, and Prince T (2015). Regulation and function of the human HSP90AA1 gene. Gene 570, 8–16. 10.1016/j.gene.2015.06.018. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 52. Xu P, Duong DM, Seyfried NT, Cheng D, Xie Y, Robert J, Rush J, Hochstrasser M, Finley D, and Peng J (2009). Quantitative proteomics reveals the function of unconventional ubiquitin chains in proteasomal degradation. Cell 137, 133–145. 10.1016/j.cell.2009.01.041. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 53. Kim W, Bennett EJ, Huttlin EL, Guo A, Li J, Possemato A, Sowa ME, Rad R, Rush J, Comb MJ, et al. (2011). Systematic and quantitative assessment of the ubiquitin-modified proteome. Molecular cell 44, 325–340. 10.1016/j.molcel.2011.08.025. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 54. Tracz M, and Bialek W (2021). Beyond K48 and K63: non-canonical protein ubiquitination. Cell Mol Biol Lett 26, 1. 10.1186/s11658-020-00245-6. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 55. Dammer EB, Na CH, Xu P, Seyfried NT, Duong DM, Cheng D, Gearing M, Rees H, Lah JJ, Levey AI, et al. (2011). Polyubiquitin linkage profiles in three models of proteolytic stress suggest the etiology of Alzheimer disease. J Biol Chem 286, 10457–10465. 10.1074/jbc.M110.149633. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 56. Verma R, Aravind L, Oania R, McDonald WH, Yates JR 3rd, Koonin EV, and Deshaies RJ (2002). Role of Rpn11 metalloprotease in deubiquitination and degradation by the 26S proteasome. Science 298, 611–615. 10.1126/science.1075898. [ DOI ] [ PubMed ] [ Google Scholar ] 57. Kudriaeva AA, Livneh I, Baranov MS, Ziganshin RH, Tupikin AE, Zaitseva SO, Kabilov MR, Ciechanover A, and Belogurov AA Jr. (2021). In-depth characterization of ubiquitin turnover in mammalian cells by fluorescence tracking. Cell Chem Biol 28, 1192–1205 e1199. 10.1016/j.chembiol.2021.02.009. [ DOI ] [ PubMed ] [ Google Scholar ] 58. Erpapazoglou Z, Walker O, and Haguenauer-Tsapis R (2014). Versatile roles of k63-linked ubiquitin chains in trafficking. Cells 3, 1027–1088. 10.3390/cells3041027. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 59. Waltho A, Popp O, Lenz C, Pluska L, Lambert M, Dotsch V, Mertins P, and Sommer T (2024). K48- and K63-linked ubiquitin chain interactome reveals branch- and length-specific ubiquitin interactors. Life Sci Alliance 7. 10.26508/lsa.202402740. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 60. Currie J, Manda V, Robinson SK, Lai C, Agnihotri V, Hidalgo V, Ludwig RW, Zhang K, Pavelka J, Wang ZV, et al. (2024). Simultaneous proteome localization and turnover analysis reveals spatiotemporal features of protein homeostasis disruptions. bioRxiv. 10.1101/2023.01.04.521821. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 61. Mathieson T, Franken H, Kosinski J, Kurzawa N, Zinn N, Sweetman G, Poeckel D, Ratnu VS, Schramm M, Becher I, et al. (2018). Systematic analysis of protein turnover in primary cells. Nature communications 9, 689. 10.1038/s41467-018-03106-1. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 62. Skinnider MA, Scott NE, Prudova A, Kerr CH, Stoynov N, Stacey RG, Chan QWT, Rattray D, Gsponer J, and Foster LJ (2021). An atlas of protein-protein interactions across mouse tissues. Cell 184, 4073–4089 e4017. 10.1016/j.cell.2021.06.003. [ DOI ] [ PubMed ] [ Google Scholar ] 63. Giurgiu M, Reinhard J, Brauner B, Dunger-Kaltenbach I, Fobo G, Frishman G, Montrone C, and Ruepp A (2019). CORUM: the comprehensive resource of mammalian protein complexes2019. Nucleic Acids Res 47, D559–D563. 10.1093/nar/gky973. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 64. Huttlin EL, Bruckner RJ, Navarrete-Perea J, Cannon JR, Baltier K, Gebreab F, Gygi MP, Thornock A, Zarraga G, Tam S, et al. (2021). Dual proteome-scale networks reveal cell-specific remodeling of the human interactome. Cell 184, 3022–3040 e3028. 10.1016/j.cell.2021.04.011. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 65. Srivastava H, Lippincott MJ, Currie J, Canfield R, Lam MPY, and Lau E (2022). Protein prediction models support widespread post-transcriptional regulation of protein abundance by interacting partners. PLoS computational biology 18, e1010702. 10.1371/journal.pcbi.1010702. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 66. Sonawane AR, Platig J, Fagny M, Chen CY, Paulson JN, Lopes-Ramos CM, DeMeo DL, Quackenbush J, Glass K, and Kuijjer ML (2017). Understanding Tissue-Specific Gene Regulation. Cell reports 21, 1077–1088. 10.1016/j.celrep.2017.10.001. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 67. Drew K, Wallingford JB, and Marcotte EM (2021). hu.MAP 2.0: integration of over 15,000 proteomic experiments builds a global compendium of human multiprotein assemblies. Molecular systems biology 17, e10016. 10.15252/msb.202010016. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 68. Wang H, Wang Y, Yang J, Zhao Q, Tang N, Chen C, Li H, Cheng C, Xie M, Yang Y, and Xie Z (2021). Tissue- and stage-specific landscape of the mouse translatome. Nucleic Acids Res 49, 6165–6180. 10.1093/nar/gkab482. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 69. Salovska B, Zhu H, Gandhi T, Frank M, Li W, Rosenberger G, Wu C, Germain PL, Zhou H, Hodny Z, et al. (2020). Isoform-resolved correlation analysis between mRNA abundance regulation and protein level degradation. Molecular systems biology 16, e9170. 10.15252/msb.20199170. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 70. Wang D, Eraslan B, Wieland T, Hallstrom B, Hopf T, Zolg DP, Zecha J, Asplund A, Li LH, Meng C, et al. (2019). A deep proteome and transcriptome abundance atlas of 29 healthy human tissues. Molecular systems biology 15, e8503. 10.15252/msb.20188503. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 71. Romanov N, Kuhn M, Aebersold R, Ori A, Beck M, and Bork P (2019). Disentangling Genetic and Environmental Effects on the Proteotypes of Individuals. Cell 177, 1308–1318 e1310. 10.1016/j.cell.2019.03.015. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 72. Uhlen M, Fagerberg L, Hallstrom BM, Lindskog C, Oksvold P, Mardinoglu A, Sivertsson A, Kampf C, Sjostedt E, Asplund A, et al. (2015). Proteomics. Tissue-based map of the human proteome. Science 347, 1260419. 10.1126/science.1260419. [ DOI ] [ PubMed ] [ Google Scholar ] 73. Germain K, and Kim PK (2020). Pexophagy: A Model for Selective Autophagy. Int J Mol Sci 21. 10.3390/ijms21020578. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 74. Islinger M, Cardoso MJ, and Schrader M (2010). Be different--the diversity of peroxisomes in the animal kingdom. Biochim Biophys Acta 1803, 881–897. 10.1016/j.bbamcr.2010.03.013. [ DOI ] [ PubMed ] [ Google Scholar ] 75. Burnett SF, Farre JC, Nazarko TY, and Subramani S (2015). Peroxisomal Pex3 activates selective autophagy of peroxisomes via interaction with the pexophagy receptor Atg30. J Biol Chem 290, 8623–8631. 10.1074/jbc.M114.619338. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 76. Sakai Y, Oku M, van der Klei IJ, and Kiel JA (2006). Pexophagy: autophagic degradation of peroxisomes. Biochim Biophys Acta 1763, 1767–1775. 10.1016/j.bbamcr.2006.08.023. [ DOI ] [ PubMed ] [ Google Scholar ] 77. Mizushima N, Noda T, Yoshimori T, Tanaka Y, Ishii T, George MD, Klionsky DJ, Ohsumi M, and Ohsumi Y (1998). A protein conjugation system essential for autophagy. Nature 395, 395–398. 10.1038/26506. [ DOI ] [ PubMed ] [ Google Scholar ] 78. Yaron-Barir TM, Joughin BA, Huntsman EM, Kerelsky A, Cizin DM, Cohen BM, Regev A, Song J, Vasan N, Lin TY, et al. (2024). The intrinsic substrate specificity of the human tyrosine kinome. Nature 629, 1174–1181. 10.1038/s41586-024-07407-y. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 79. Johnson JL, Yaron TM, Huntsman EM, Kerelsky A, Song J, Regev A, Lin TY, Liberatore K, Cizin DM, Cohen BM, et al. (2023). An atlas of substrate specificities for the human serine/threonine kinome. Nature 613, 759–766. 10.1038/s41586-022-05575-3. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 80. Savas JN, Toyama BH, Xu T, Yates JR 3rd, and Hetzer MW (2012). Extremely long-lived nuclear pore proteins in the rat brain. Science 335, 942. 10.1126/science.1217421. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 81. Alonso A, Zaidi T, Novak M, Grundke-Iqbal I, and Iqbal K (2001). Hyperphosphorylation induces self-assembly of tau into tangles of paired helical filaments/straight filaments. Proceedings of the National Academy of Sciences of the United States of America 98, 6923–6928. 10.1073/pnas.121119298. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 82. Dickey CA, Kamal A, Lundgren K, Klosak N, Bailey RM, Dunmore J, Ash P, Shoraka S, Zlatkovic J, Eckman CB, et al. (2007). The high-affinity HSP90-CHIP complex recognizes and selectively degrades phosphorylated tau client proteins. J Clin Invest 117, 648–658. 10.1172/JCI29715. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 83. Kawahata I, Finkelstein DI, and Fukunaga K (2022). Pathogenic Impact of alpha-Synuclein Phosphorylation and Its Kinases in alpha-Synucleinopathies. Int J Mol Sci 23. 10.3390/ijms23116216. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 84. Hu Z, Chen PH, Li W, Douglas T, Hines J, Liu Y, and Crews CM (2023). Targeted Dephosphorylation of Tau by Phosphorylation Targeting Chimeras (PhosTACs) as a Therapeutic Modality. J Am Chem Soc. 10.1021/jacs.2c11706. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 85. Hu Z, Chen PH, Li W, Krone M, Zheng S, Saarbach J, Velasco IU, Hines J, Liu Y, and Crews CM (2024). EGFR targeting PhosTACs as a dual inhibitory approach reveals differential downstream signaling. Sci Adv 10, eadj7251. 10.1126/sciadv.adj7251. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 86. Wang S, Li W, Hu L, Cheng J, Yang H, and Liu Y (2020). NAguideR: performing and prioritizing missing value imputations for consistent bottom-up proteomic analyses. Nucleic Acids Res. 10.1093/nar/gkaa498. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 87. Guan S, Price JC, Ghaemmaghami S, Prusiner SB, and Burlingame AL (2012). Compartment modeling for mammalian protein turnover studies by stable isotope metabolic labeling. Analytical chemistry 84, 4014–4021. 10.1021/ac203330z. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 88. Ochoa D, Jarnuczak AF, Vieitez C, Gehre M, Soucheray M, Mateus A, Kleefeldt AA, Hill A, Garcia-Alonso L, Stein F, et al. (2020). The functional landscape of the human phosphoproteome. Nature biotechnology 38, 365–373. 10.1038/s41587-019-0344-3. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 89. Mihailovich M, Germain PL, Shyti R, Pozzi D, Noberini R, Liu Y, Aprile D, Tenderini E, Troglio F, Trattaro S, et al. (2024). Multiscale modeling uncovers 7q11.23 copy number variation-dependent changes in ribosomal biogenesis and neuronal maturation and excitability. J Clin Invest 134. 10.1172/JCI168982. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 90. Liu Y, Yang J, Wang T, Luo M, Chen Y, Chen C, Ronai Z, Zhou Y, Ruppin E, and Han L (2023). Expanding PROTACtable genome universe of E3 ligases. Nature communications 14, 6509. 10.1038/s41467-023-42233-2. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 91. Rahman M, Previs SF, Kasumov T, and Sadygov RG (2016). Gaussian Process Modeling of Protein Turnover. Journal of proteome research 15, 2115–2122. 10.1021/acs.jproteome.5b00990. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 92. Ding C, Li Y, Guo F, Jiang Y, Ying W, Li D, Yang D, Xia X, Liu W, Zhao Y, et al. (2016). A Cell-type-resolved Liver Proteome. Molecular & cellular proteomics : MCP 15, 3190–3202. 10.1074/mcp.M116.060145. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 93. Nusinow DP, Szpyt J, Ghandi M, Rose CM, McDonald ER 3rd, Kalocsay M, Jane-Valbuena J, Gelfand E, Schweppe DK, Jedrychowski M, et al. (2020). Quantitative Proteomics of the Cancer Cell Line Encyclopedia. Cell 180, 387–402 e316. 10.1016/j.cell.2019.12.023. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 94. Kim-Hellmuth S, Aguet F, Oliva M, Munoz-Aguirre M, Kasela S, Wucher V, Castel SE, Hamel AR, Vinuela A, Roberts AL, et al. (2020). Cell type-specific genetic regulation of gene expression across human tissues. Science 369. 10.1126/science.aaz8528. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 95. Sabatier P, Ye Z, Lechner M, Guzmán UH, Beusch CM, Izaguirre F, Seth A, Gritsenko O, Rodin S, Grinnemo K-H, and Olsen JV (2024). Global analysis of protein turnover dynamics in single cells. bioRxiv, 2024.2005.2030.596745. 10.1101/2024.05.30.596745. [ DOI ] [ PubMed ] [ Google Scholar ] 96. Zhang Z, Schaefer C, Jiang W, Lu Z, Lee J, Sziraki A, Abdulraouf A, Wick B, Haeussler M, Li Z, et al. (2024). A panoramic view of cell population dynamics in mammalian aging. Science, eadn3949. 10.1126/science.adn3949. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 97. Mauvoisin D, Wang J, Jouffe C, Martin E, Atger F, Waridel P, Quadroni M, Gachon F, and Naef F (2014). Circadian clock-dependent and -independent rhythmic proteomes implement distinct diurnal functions in mouse liver. Proceedings of the National Academy of Sciences of the United States of America 111, 167–172. 10.1073/pnas.1314066111. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 98. Perez-Riverol Y, Bai J, Bandla C, Garcia-Seisdedos D, Hewapathirana S, Kamatchinathan S, Kundu DJ, Prakash A, Frericks-Zipper A, Eisenacher M, et al. (2022). The PRIDE database resources in 2022: a hub for mass spectrometry-based proteomics evidences. Nucleic Acids Res 50, D543–D552. 10.1093/nar/gkab1038. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 99. Bai B, Tan H, Pagala VR, High AA, Ichhaporia VP, Hendershot L, and Peng J (2017). Deep Profiling of Proteome and Phosphoproteome by Isobaric Labeling, Extensive Liquid Chromatography, and Mass Spectrometry. Methods Enzymol 585, 377–395. 10.1016/bs.mie.2016.10.007. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 100. Wang Z, Yu K, Tan H, Wu Z, Cho JH, Han X, Sun H, Beach TG, and Peng J (2020). 27-Plex Tandem Mass Tag Mass Spectrometry for Profiling Brain Proteome in Alzheimer’s Disease. Anal Chem 92, 7162–7170. 10.1021/acs.analchem.0c00655. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 101. Gao Q, Zhu H, Dong L, Shi W, Chen R, Song Z, Huang C, Li J, Dong X, Zhou Y, et al. (2019). Integrated Proteogenomic Characterization of HBV-Related Hepatocellular Carcinoma. Cell 179, 561–577 e522. 10.1016/j.cell.2019.08.052. [ DOI ] [ PubMed ] [ Google Scholar ] 102. Salovska B, Li W, Di Y, and Liu Y (2021). BoxCarmax: a high-selectivity data-independent acquisition mass spectrometry method for the analysis of protein turnover and complex samples. Analytical chemistry 93, 3103–3111. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 103. Mehnert M, Li W, Wu C, Salovska B, and Liu Y (2019). Combining Rapid Data Independent Acquisition and CRISPR Gene Deletion for Studying Potential Protein Functions: A Case of HMGN1. Proteomics, e1800438. 10.1002/pmic.201800438. [ DOI ] [ PubMed ] [ Google Scholar ] 104. Li W, Chi H, Salovska B, Wu C, Sun L, Rosenberger G, and Liu Y (2019). Assessing the Relationship Between Mass Window Width and Retention Time Scheduling on Protein Coverage for Data-Independent Acquisition. J Am Soc Mass Spectrom. 10.1007/s13361-019-02243-1. [ DOI ] [ PubMed ] [ Google Scholar ] 105. Salovska B, Gao E, Muller-Dott S, Li W, Cordon CC, Wang S, Dugourd A, Rosenberger G, Saez-Rodriguez J, and Liu Y (2023). Phosphoproteomic analysis of metformin signaling in colorectal cancer cells elucidates mechanism of action and potential therapeutic opportunities. Clin Transl Med 13, e1179. 10.1002/ctm2.1179. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 106. Wang H, Yang Y, Li Y, Bai B, Wang X, Tan H, Liu T, Beach TG, Peng J, and Wu Z (2015). Systematic optimization of long gradient chromatography mass spectrometry for deep analysis of brain proteome. Journal of proteome research 14, 829–838. 10.1021/pr500882h. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 107. Keihani S, Kluever V, Mandad S, Bansal V, Rahman R, Fritsch E, Gomes LC, Gärtner A, Kügler S, Urlaub H, et al. (2019). The long noncoding RNA neuroLNC regulates presynaptic activity by interacting with the neurodegeneration-associated protein TDP-43. Sci Adv 5, eaay2670. 10.1126/sciadv.aay2670. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 108. Kaech S, and Banker G (2006). Culturing hippocampal neurons. Nat Protoc 1, 2406–2415. 10.1038/nprot.2006.356. [ DOI ] [ PubMed ] [ Google Scholar ] 109. Truckenbrodt S, Viplav A, Jähne S, Vogts A, Denker A, Wildhagen H, Fornasiero EF, and Rizzoli SO (2018). Newly produced synaptic vesicle proteins are preferentially used in synaptic transmission. Embo j 37. 10.15252/embj.201798044. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 110. Yousefi R, Jevdokimenko K, Kluever V, Pacheu-Grau D, and Fornasiero EF (2021). Influence of Subcellular Localization and Functional State on Protein Turnover. Cells 10. 10.3390/cells10071747. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 111. Hu Z, Chen P-H, Li W, Douglas T, Hines J, Liu Y, and Crews CM (2023). Targeted dephosphorylation of tau by phosphorylation targeting chimeras (PhosTACs) as a therapeutic modality. Journal of the American Chemical Society 145, 4045–4055. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 112. Zhou W, Li W, Wang S, Salovska B, Hu Z, Tao B, Di Y, Punyamurtula U, Turk BE, Sessa WC, and Liu Y (2023). An optogenetic-phosphoproteomic study reveals dynamic Akt1 signaling profiles in endothelial cells. Nature communications 14, 3803. 10.1038/s41467-023-39514-1. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 113. Salovska B, Gao E, Müller-Dott S, Li W, Cordon CC, Wang S, Dugourd A, Rosenberger G, Saez-Rodriguez J, and Liu Y (2023). Phosphoproteomic analysis of metformin signaling in colorectal cancer cells elucidates mechanism of action and potential therapeutic opportunities. Clinical and Translational Medicine 13, e1179. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 114. Collins BC, Hunter CL, Liu Y, Schilling B, Rosenberger G, Bader SL, Chan DW, Gibson BW, Gingras AC, Held JM, et al. (2017). Multi-laboratory assessment of reproducibility, qualitative and quantitative performance of SWATH-mass spectrometry. Nature communications 8, 291. 10.1038/s41467-017-00249-5. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 115. Wang X, Li Y, Wu Z, Wang H, Tan H, and Peng J (2014). JUMP: a tag-based database search tool for peptide identification with high sensitivity and accuracy. Mol Cell Proteomics 13, 3663–3673. 10.1074/mcp.O114.039586. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 116. Peng J, Elias JE, Thoreen CC, Licklider LJ, and Gygi SP (2003). Evaluation of multidimensional chromatography coupled with tandem mass spectrometry (LC/LC-MS/MS) for large-scale protein analysis: the yeast proteome. Journal of proteome research 2, 43–50. 10.1021/pr025556v. [ DOI ] [ PubMed ] [ Google Scholar ] 117. Bai B, Wang X, Li Y, Chen PC, Yu K, Dey KK, Yarbro JM, Han X, Lutz BM, Rao S, et al. (2020). Deep Multilayer Brain Proteomics Identifies Molecular Networks in Alzheimer’s Disease Progression. Neuron 106, 700. 10.1016/j.neuron.2020.04.031. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 118. Shrestha HK, Sun H, Wang J, and Peng J (2024). Profiling Mouse Brain Single-Cell-Type Proteomes Via Adeno-Associated Virus-Mediated Proximity Labeling and Mass Spectrometry. Methods Mol Biol 2817, 115–132. 10.1007/978-1-0716-3934-4_10. [ DOI ] [ PubMed ] [ Google Scholar ] 119. Bruderer R, Bernhardt OM, Gandhi T, Miladinovic SM, Cheng LY, Messner S, Ehrenberger T, Zanotelli V, Butscheid Y, Escher C, et al. (2015). Extending the limits of quantitative proteome profiling with data-independent acquisition and application to acetaminophen-treated three-dimensional liver microtissues. Molecular & cellular proteomics : MCP 14, 1400–1410. 10.1074/mcp.M114.044305. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 120. Salovska B, Zhu H, Gandhi T, Frank M, Li W, Rosenberger G, Wu C, Germain P-L, Zhou H, Hodny Z, et al. (2020). Isoform-resolved correlation analysis between mRNA abundance regulation and protein level degradation. Molecular Systems Biology 16, e9170. 10.15252/msb.20199170. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 121. Bekker-Jensen DB, Bernhardt OM, Hogrebe A, Martinez-Val A, Verbeke L, Gandhi T, Kelstrup CD, Reiter L, and Olsen JV (2020). Rapid and site-specific deep phosphoproteome profiling by data-independent acquisition without the need for spectral libraries. Nature communications 11, 787. 10.1038/s41467-020-14609-1. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 122. Cox J, and Mann M (2008). MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification. Nat Biotechnol 26, 1367–1372. 10.1038/nbt.1511. [ DOI ] [ PubMed ] [ Google Scholar ] 123. Olsen JV, Blagoev B, Gnad F, Macek B, Kumar C, Mortensen P, and Mann M (2006). Global, in vivo, and site-specific phosphorylation dynamics in signaling networks. Cell 127, 635–648. 10.1016/j.cell.2006.09.026. [ DOI ] [ PubMed ] [ Google Scholar ] 124. Liu Y, Mi Y, Mueller T, Kreibich S, Williams EG, Van Drogen A, Borel C, Frank M, Germain PL, Bludau I, et al. (2019). Multi-omic measurements of heterogeneity in HeLa cells across laboratories. Nature biotechnology 37, 314–322. 10.1038/s41587-019-0037-y. [ DOI ] [ PubMed ] [ Google Scholar ] 125. Wang S, Di Y, Yang Y, Salovska B, Li W, Hu L, Yin J, Shao W, Zhou D, Cheng J, et al. (2024). PTMoreR-enabled cross-species PTM mapping and comparative phosphoproteomics across mammals. Cell Rep Methods 4, 100859. 10.1016/j.crmeth.2024.100859. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 126. Johnson JL, Yaron TM, Huntsman EM, Kerelsky A, Song J, Regev A, Lin T-Y, Liberatore K, Cizin DM, Cohen BM, et al. (2023). An atlas of substrate specificities for the human serine/threonine kinome. Nature 613, 759–766. 10.1038/s41586-022-05575-3. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 127. Salovska B, Zhu H, Gandhi T, Frank M, Li W, Rosenberger G, Wu C, Germain PL, Zhou H, and Hodny Z (2020). Isoform-resolved correlation analysis between mRNA abundance regulation and protein level degradation. Molecular systems biology 16, e9170. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 128. Hajian-Tilaki K (2013). Receiver Operating Characteristic (ROC) Curve Analysis for Medical Diagnostic Test Evaluation. Caspian J Intern Med 4, 627–635. [ PMC free article ] [ PubMed ] [ Google Scholar ] 129. Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, Blondel M, Prettenhofer P, Weiss R, Dubourg V, et al. (2011). Scikit-learn: Machine Learning in Python. J. Mach. Learn. Res. 12, 2825–2830. [ Google Scholar ] 130. Shemesh N, Jubran J, Dror S, Simonovsky E, Basha O, Argov C, Hekselman I, Abu-Qarn M, Vinogradov E, Mauer O, et al. (2021). The landscape of molecular chaperones across human tissues reveals a layered architecture of core and variable chaperones. Nature communications 12, 2180. 10.1038/s41467-021-22369-9. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 131. Uhlén M, Fagerberg L, Hallström BM, Lindskog C, Oksvold P, Mardinoglu A, Sivertsson Å, Kampf C, Sjöstedt E, Asplund A, et al. (2015). Proteomics. Tissue-based map of the human proteome. Science 347, 1260419. 10.1126/science.1260419. [ DOI ] [ PubMed ] [ Google Scholar ] 132. Spearman C (1904). The Proof and Measurement of Association between Two Things. The American Journal of Psychology 15, 72–101. 10.2307/1412159. [ DOI ] [ PubMed ] [ Google Scholar ] 133. Jovanovic M, Rooney MS, Mertins P, Przybylski D, Chevrier N, Satija R, Rodriguez EH, Fields AP, Schwartz S, Raychowdhury R, et al. (2015). Immunogenetics. Dynamic profiling of the protein life cycle in response to pathogens. Science 347, 1259038. 10.1126/science.1259038. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 134. Csárdi G, Franks A, Choi DS, Airoldi EM, and Drummond DA (2015). Accounting for Experimental Noise Reveals That mRNA Levels, Amplified by Post-Transcriptional Processes, Largely Determine Steady-State Protein Levels in Yeast. PLOS Genetics 11, e1005206. 10.1371/journal.pgen.1005206. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 135. Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, Amin N, Schwikowski B, and Ideker T (2003). Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome research 13, 2498–2504. 10.1101/gr.1239303. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 136. Tyanova S, Temu T, Sinitcyn P, Carlson A, Hein MY, Geiger T, Mann M, and Cox J (2016). The Perseus computational platform for comprehensive analysis of (prote)omics data. Nature Methods 13, 731–740. 10.1038/nmeth.3901. [ DOI ] [ PubMed ] [ Google Scholar ] 137. Cox J, and Mann M (2012). 1D and 2D annotation enrichment: a statistical method integrating quantitative proteomics with complementary high-throughput data. BMC bioinformatics 13, S12. 10.1186/1471-2105-13-S16-S12. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 138. Zhou Y, Zhou B, Pache L, Chang M, Khodabakhshi AH, Tanaseichuk O, Benner C, and Chanda SK (2019). Metascape provides a biologist-oriented resource for the analysis of system-slevel datasets. Nature communications 10, 1523. 10.1038/s41467-019-09234-6. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 139. Sherman BT, Hao M, Qiu J, Jiao X, Baseler MW, Lane HC, Imamichi T, and Chang W (2022). DAVID: a web server for functional enrichment analysis and functional annotation of gene lists (2021 update). Nucleic Acids Res 50, W216–w221. 10.1093/nar/gkac194. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials 1 Figure S1. Establishing a reproducible and in-depth proteome turnover atlas across mouse tissues and brain regions, related to Figure 1 (A) Coefficient of variation (CV) distribution of proteome-wide Heavy/Light (H/L) ratios in analyzing the mouse whole brain tissue with five SILAC labeling time points of 0, 4, 8, 16, and 32 days. Among the five labelled mice, three biological replicates were used for TMTpro-based quantification for this analysis. (B) The quantitative slopes of the log-transformed H/L ratios [i.e., ln (H+L)/L] growing over time are highly correlated between the experiments of 5 time points (0d, 4d, 8d, 16d, 32d) and 3 time points (0d, 8d, 32d). The slope primarily determines the protein turnover rates. Rho , Spearman correlation coefficient; R, Pearson R correlation coefficient. (C) Correlation (Spearman and Pearson) of protein lifetimes between two biological replicates measured by BoxCarmax-DIA. (D) Correlation plot of PTs across the proteome, as derived by DIA and TMT methods (same as Figure 1B but with the forced fitting to y= k x regression), R, Pearson R correlation coefficient. This plot demonstrates the strong correlation and comparable PTs determined by the two MS platforms after respective data procession (see Methods ). (E) Venn diagram comparing the protein coverage of DIA and TMT datasets. (F) Venn diagram comparing the turnover coverage of mouse liver proteome between this study and Hasper et al. (G) Scatterplot displaying the comparison of protein abundance (indicated by DIA-MS intensity in this study) and lifetime for the liver proteome between this study and Hasper et al. (red dots). (H) Correlation (Spearman and Pearson) of T 50 results between the two studies for the mouse liver proteome. (I) The boxplot displaying the proteome turnover coverage for the heart and liver proteomes measured by this study and Hasper et al. NIHMS2066478-supplement-1.pdf (1.1MB, pdf) 2 Figure S2. Short-lived and long-lived proteins and their associated functional processes across tissues, related to Figure 2 (A) The correlation of Log 2 (T 50 ) for the same proteins identified and measured in plasma and tissues. For each of the plasma proteins, its x axis denotes the PT measured in plasma while y axis represents PT in other tissues. The dashed line indicates the diagonal line of y=x (i.e., identical PT between tissue and plasma), which divides the proteins into two groups (i.e., PT lower or higher in plasma). The number and percentage of proteins in each of these two groups are shown, suggesting the PTs of most plasma proteins are generally higher than in plasma than other tissues (i.e., the majority of the proteins are below the diagonal line). For example, 76.8% of the liver proteins overlapping to the plasma proteome have a longer PT in blood plasma. (B) Upset plots of short-lived and long-lived proteins in each tissue. The top 5% shortest- and longest-lived proteins in each tissue were compared between tissues. (C) STRING Protein-protein interaction network (medium confidence) of 49 proteins that are the top 5% most long-lived in each of the nine brain regions. The Enrichment P values were reported by STRING analysis on the protein clusters. (D) Venn diagrams comparing short-lived and long-lived protein identities in brain regions and non-brain tissues. (E) Metascape ( https://metascape.org )-derived GO enrichment analysis (Top100 items) from short-lived and long-lived proteins in brain regions and non-brain tissues. The Enrichment P values were reported by Metascape analysis. NIHMS2066478-supplement-2.pdf (3MB, pdf) 3 Figure S3. Tissue-PPT App supports the functional discovery of proteins and protein sets by visualizing both protein abundance (PA) and protein lifetime (PT) and their relationships across tissues, related to Figure 2 and 4 (A) The Shiny Tissue-PPT App ( https://yslproteomics.shinyapps.io/tissuePPT/ ) was established for accessing and analyzing both PA and PT data with examples shown. The App includes the five major functions: Heat-circle (HC) plot across tissues, Heatmap analysis for protein sets, Protein-specific bar plots, Correlation analysis at the protein level, and Correlation analysis for protein sets. HC plot examples of protein-protein interaction partners of PSMD1, LMNA, and AK2 proteins in across tissues (see Figure 4E ). For the HC plot, the blue-to-red color gradient denotes PT from long to short. The size of the HC plot circles is proportional to the Log 2 (riBAQ) in ppm value indicating PA. Triangle: Log 2 (riBAQ) < −1, Diamond: Log 2 (riBAQ) > 9. (B) Heatmap examples of both PA and PT generated for proteasome subunits (19S and 20S) and Mitochondrial ribosomal subunits (small/large) across tissues. The scale of PA and PT are kept consist across proteome facilitating direct comparison between individual proteins and protein lists. In this case, PA of proteosome subunits is much higher than that of mitochondrial ribosomal subunits, while both complexes are long-lived. The orange-to-green color bar and red-to-blue color bar denote the small-to-large PA and PT values (Log 2 -transformed), respectively. NIHMS2066478-supplement-3.pdf (1.6MB, pdf) 4 Figure S4. HC plots for proteostasis-associated proteins and the correlation between PT and PPI resolved to biological pathways, related to Figure 3 and 4 (A-C) HC plots of protein folding chaperones, E3 accessory proteins, and deubiquitinating enzymes (DUBs) across tissues. (D) Heatmap of Pearson R for PA and PT between PPI partners based on KEGG pathways. In addition to random protein pairs as negative controls, here we further filtered these random protein pairs by requesting them to be participating in the same pathways (see Methods ). KEGG pathways with a “pathway” in their item names and a gene count number higher than 49 were shown. PPI list was derived from Tissue-PPI. Both proteins forming a random-pair (not listed in Tissue-PPI) have to be annotated in the same pathway to be treated as “PPI out” or “no PPI”. NIHMS2066478-supplement-4.pdf (765.4KB, pdf) 5 Figure S5. Multi-omic analysis of peroxisome and pexophagy proteins, related to Figure 5 (A) Spearman correlation between RPF (the sequencing of ribosome protected fragments) compared with MS-derived K syn (proxy of protein synthesis rate) and iBAQ values for five tissues. P-values were calculated using Wilcoxon matched-pairs signed rank test (One sided). Please refer to method section for K syn calculation. The RPF data was measured by ribosome sequencing. (B) Spearman correlation of Log 2 RFP to K syn (upper panel) compared with Log 2 iBAQ (lower panel) in Lung tissue as an example. (C) Boxplots of protein-specific Spearman correlation rho values between multi-omic layers for all measured proteins (upper panel) and tissue-enriched proteins (lower panel, see Methods ) based on DIA- (left panel) and TMT- (right panel) derived results. (D) Heatmap of individual peroxisome proteins among five tissues using the quantitative values from mRNA, RPF, PA, and PT layers. The brain results were determined by averaging all brain regions. The heatmap was created by the R pheatmap package with the row sale. Scale bar: red color is higher value per layer; blue color is lower value. (E) Heatmap visualizing the cross-tissue Spearman correlation of each peroxisome protein between two omics layers across tissues as indicated. The blue-to-red scale bar denotes the increasing Spearman correlation. (F) Boxplots exploring the relative PTs of peroxisome proteins with those of other proteins in the proteome in different tissues. The relative PT was calculated by dividing the PT by the average PT in five tissues. P , Wilcoxon test. (G) The negative correlation between the abundances of pexophagy-associated proteins and the lysosome protein levels. Plots were generated by the “Correlation analysis: protein sets” function in Tissue-PPT . Pearson correlation R and the significance of correlation (P value, based on cor.test in R) are shown. (H) Correlation between Log 2 (Protein abundance) and Log 2 (T 50 ) for protein examples of Pex3 and Atg12 involved in pexophagy. Plots were generated by the “Correlation analysis: proteins” function in Tissue-PPT . (I) HC plots of proteins involving in pexophagy-associated processes such as peroxisome designation and targeting, and autophagosome biogenesis. Plots were generated by the “Heat-circle (HC) plot across tissues” function in Tissue-PPT . NIHMS2066478-supplement-5.pdf (2.1MB, pdf) 6 Figure S6. Phosphoproteome turnover measurement across mouse tissues reveals the different T50 of the phosphorylated peptides and their non-phosphorylated counterparts, related to Figure 6 and 7 (A) Density plot of the quantified T 50 values of phosphorylation sites (P-sites) across mouse tissues based on all phosphopeptides measured. (B) Distribution of Spearman correlation between the T 50 of a phosphorylated (p) peptide and the T 50 of its non-phosphorylated (np) peptide counterpart across tissues and regions for all P-sites. The p and np peptides share the same amino acid sequence and only differ in the phosphorylation modification. (C) Density plot of DeltaT 50 (ΔT 50 ) that represents the difference between the T 50 of p peptide to the T 50 of its corresponding np peptide across tissues. (D) Venn diagram comparing the numbers of P-sites showing destabilization (T 50 of p peptide is shorter than T 50 of np peptide) or stabilization effects (T 50 of p peptide is shorter than T 50 of np peptide) in brain and non-brain areas. The significance is inferred by P <0.05 (Student’s t-test) and |fold change| >1.5 (in brain) and >1.2 (in non-brain tissues). (E) Heavy/light (H/L) ratio curves during the labeling process for a p peptide (green) and its np peptide counterpart (blue) of the same sequence as examples of long-lived proteins. NIHMS2066478-supplement-6.pdf (1.6MB, pdf) 7 Figure S7. Two dimensional (2D) functional enrichment analysis comparing proteome and phosphoproteome abundance and turnover variability across brain regions, related to Figure 6 (A) Bubble plot of abundance variability between proteome and phosphoproteome across brain regions. CV was calculated from phosphorylated protein abundance (y-axis, an average of all phosphosites) or protein abundance (x-axis) across all brain regions. 2D enrichment was performed using Perseus software and the bubble plot was visualized using ggplot2. Enriched GO terms were filtered with p-value <0.05 and number of proteins >5. The cycle size denotes the number of proteins in each GO term. The color scale represents the enriched GO P-value using Perseus analysis. (B) Bubble plot of lifetime variability between proteome and phosphoproteome across brain regions. CV was calculated from phosphorylated protein lifetime (y-axis, an average of all phosphosites) and protein lifetime (x-axis) across all brain regions. The cycle size denotes the number of proteins in each GO term. The color scale represents the enriched GO P-value. NIHMS2066478-supplement-7.pdf (699.3KB, pdf) 8 NIHMS2066478-supplement-8.xlsx (6.5MB, xlsx) 9 NIHMS2066478-supplement-9.xlsx (11.4MB, xlsx) Data Availability Statement The mass spectrometry raw data and searched results have been all deposited to the ProteomeXchange Consortium via the PRIDE 98 with the dataset identifier PXD055942. The source code for the Tissue Proteome and Phosphoproteome Turnover web application (Tissue-PPT) is available under the MIT license on GitHub ( https://github.com/yslproteomics/tissuePPT ) and archived on Zenodo ( https://zenodo.org/records/14671026 ). Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request. ACTIONS View on publisher site PDF (4.8 MB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• Distinct Functional Signatures of Human Olfactory and Respiratory Mucus Revealed by Proteomics Combined With Machine Learning.#380755
• Expanding the human proteome with microproteins and peptideins.#295191
• Quantitation of Spatial Proteoforms in Alzheimer's Disease#422359
• Protein-protein interaction inhibitors in the human proteome: lessons from 117 targeted interactions.#457421
• Quantitative proteomics targeting classes of motif-containing peptides using immunoaffinity-based mass spectrometry.#452366