No abstract

The lifespans of proteins can range from moments to years within mammalian tissues. Protein lifespan is relevant to organismal aging, as long-lived proteins can accrue damage over time. It is unclear how protein lifetime is shaped by tissue context, where both cell division and proteolytic degradation contribute to protein turnover. Here, we develop turnover and replication analysis by 15N isotope labeling (TRAIL) for parallel quantification of protein and cell lifetimes. We deploy TRAIL over 32 days in 4 mouse tissues to quantify cell proliferation with high precision and no toxicity and determine that protein lifespan varies independently of cell lifespan. Variation in protein lifetime is non-random: multiprotein complexes such as the ribosome have consistent lifetimes across tissues, while mitochondria, peroxisomes, and lipid droplets have variable lifetimes across tissues. To model the effects of aging on tissue homeostasis, we apply TRAIL to progeroid mice and uncover fat-specific alterations in cell lifetime and proteome composition, as well as a broad decrease in protein turnover flux. These data indicate that environmental factors influence protein turnover in vivo and provide a framework to understand proteome aging in tissue context.

The lifespans of proteins range from minutes to years within mammalian tissues. Protein lifespan is relevant to organismal aging, as long‐lived proteins accrue damage over time. It is unclear how protein lifetime is shaped by tissue context, where both cell turnover and proteolytic degradation contribute to protein turnover. We develop t urnover and r eplication a nalysis by 15 N i sotope l abeling (TRAIL) to quantify protein and cell lifetimes with high precision and demonstrate that cell turnover, sequence‐encoded features, and environmental factors modulate protein lifespan across tissues . Cell and protein turnover flux are comparable in proliferative tissues, while protein turnover outpaces cell turnover in slowly proliferative tissues. Physicochemical features such as hydrophobicity, charge, and disorder influence protein turnover in slowly proliferative tissues, but protein turnover is much less sequence‐selective in highly proliferative tissues. Protein lifetimes vary nonrandomly across tissues after correcting for cell turnover. Multiprotein complexes such as the ribosome have consistent lifetimes across tissues, while mitochondria, peroxisomes, and lipid droplets have variable lifetimes. TRAIL can be used to explore how environment, aging, and disease affect tissue homeostasis.

Mutations to the LMNA gene cause laminopathies including Hutchinson-Gilford progeria syndrome (HGPS). The origins of tissue specificity in these diseases are unclear, as the A-type Lamins are abundant and broadly expressed proteins. We show that A-type Lamin protein and transcript levels are uncorrelated across tissues. As protein-transcript discordance can be caused by variations in protein lifetime, we applied quantitative proteomics to profile protein turnover rates in healthy and progeroid tissues. We discover that tissue context and disease mutation each influence A-type Lamin protein lifetime. Lamin A/C has a weeks-long lifetime in the aorta, heart, and fat, but a days-long lifetime in tissues spared from disease. Progerin is even more long-lived than Lamin A/C in the cardiovascular system and accumulates there over time. These proteins are insoluble and densely bundled in cardiovascular tissues, which may present an energetic barrier to degradation. We reveal that human disease alleles are significantly over-represented in the long-lived proteome. These findings indicate that gene therapy interventions will have significant latency and limited potency in disrupting long-lived disease-linked proteins such as Progerin.

<div>Abstract<p>Metabolic dysregulation is a known hallmark of cancer progression, yet the oncogenic signals that promote metabolic adaptations to drive metastatic cancer remain unclear. Here, we show that transcriptional repression of mitochondrial deacetylase sirtuin 3 (<i>SIRT3</i>) by androgen receptor (AR) and its coregulator steroid receptor coactivator-2 (SRC-2) enhances mitochondrial aconitase (ACO2) activity to favor aggressive prostate cancer. ACO2 promoted mitochondrial citrate synthesis to facilitate <i>de novo</i> lipogenesis, and genetic ablation of <i>ACO2</i> reduced total lipid content and severely repressed <i>in vivo</i> prostate cancer progression. A single acetylation mark lysine258 on ACO2 functioned as a regulatory motif, and the acetylation-deficient Lys258Arg mutant was enzymatically inactive and failed to rescue growth of ACO2-deficient cells. Acetylation of ACO2 was reversibly regulated by SIRT3, which was predominantly repressed in many tumors including prostate cancer. Mechanistically, SRC-2–bound AR formed a repressive complex by recruiting histone deacetylase 2 to the <i>SIRT3</i> promoter, and depletion of <i>SRC-2</i> enhanced <i>SIRT3</i> expression and simultaneously reduced acetylated ACO2. In human prostate tumors, ACO2 activity was significantly elevated, and increased expression of <i>SRC-2</i> with concomitant reduction of <i>SIRT3</i> was found to be a genetic hallmark enriched in prostate cancer metastatic lesions. In a mouse model of spontaneous bone metastasis, suppression of <i>SRC-2</i> reactivated <i>SIRT3</i> expression and was sufficient to abolish prostate cancer colonization in the bone microenvironment, implying this nuclear-mitochondrial regulatory axis is a determining factor for metastatic competence.</p>Significance:<p>This study highlights the importance of mitochondrial aconitase activity in the development of advanced metastatic prostate cancer and suggests that blocking SRC-2 to enhance <i>SIRT3</i> expression may be therapeutically valuable.</p></div>