A quantitative spatial proteomics analysis of proteome turnover in human cells

Measuring the properties of endogenous cell proteins, such as expression level, subcellular localization, and turnover rates, on a whole proteome level remains a major challenge in the postgenome era. Quantitative methods for measuring mRNA expression do not reliably predict corresponding protein levels and provide little or no information on other protein properties. Here we describe a combined pulse-labeling, spatial proteomics and data analysis strategy to characterize the expression, localization, synthesis, degradation, and turnover rates of endogenously expressed, untagged human proteins in different subcellular compartments. Using quantitative mass spectrometry and stable isotope labeling with amino acids in cell culture, a total of 80,098 peptides from 8,041 HeLa proteins were quantified, and their spatial distribution between the cytoplasm, nucleus and nucleolus determined and visualized using specialized software tools developed in PepTracker. Using information from ion intensities and rates of change in isotope ratios, protein abundance levels and protein synthesis, degradation and turnover rates were calculated for the whole cell and for the respective cytoplasmic, nuclear, and nucleolar compartments. Expression levels of endogenous HeLa proteins varied by up to seven orders of magnitude. The average turnover rate for HeLa proteins was ~20 h. Turnover rate did not correlate with either molecular weight or net charge, but did correlate with abundance, with highly abundant proteins showing longer than average half-lives. Fast turnover proteins had overall a higher frequency of PEST motifs than slow turnover proteins but no general correlation was observed between amino or carboxyl terminal amino acid identities and turnover rates. A subset of proteins was identified that exist in pools with different turnover rates depending on their subcellular localization. This strongly correlated with subunits of large, multiprotein complexes, suggesting a general mechanism whereby their assembly is controlled in a different subcellular location to their main site of function.     

Harmonizing Labeling and Analytical Strategies to Obtain Protein Turnover Rates in Intact Adult Animals

Changes in the abundance of individual proteins in the proteome can be elicited by modulation of protein synthesis (the rate of input of newly synthesized proteins into the protein pool) or degradation (the rate of removal of protein molecules from the pool). A full understanding of proteome changes therefore requires a definition of the roles of these two processes in proteostasis, collectively known as protein turnover. Because protein turnover occurs even in the absence of overt changes in pool abundance, turnover measurements necessitate monitoring the flux of stable isotope–labeled precursors through the protein pool such as labeled amino acids or metabolic precursors such as ammonium chloride or heavy water. In cells in culture, the ability to manipulate precursor pools by rapid medium changes is simple, but for more complex systems such as intact animals, the approach becomes more convoluted. Individual methods bring specific complications, and the suitability of different methods has not been comprehensively explored. In this study, we compare the turnover rates of proteins across four mouse tissues, obtained from the same inbred mouse strain maintained under identical husbandry conditions, measured using either [¹³C₆]lysine or [²H₂]O as the labeling precursor. We show that for long-lived proteins, the two approaches yield essentially identical measures of the first-order rate constant for degradation. For short-lived proteins, there is a need to compensate for the slower equilibration of lysine through the precursor pools. We evaluate different approaches to provide that compensation. We conclude that both labels are suitable, but careful determination of precursor enrichment kinetics in amino acid labeling is critical and has a considerable influence on the numerical values of the derived protein turnover rates.     

A proteomics strategy for determining the synthesis and degradation rates of individual proteins in fish

In order to study the protein dynamics in the tissues of fish we have developed a proteomics-based strategy to determine the rates of synthesis and degradation of individual proteins. We have demonstrated the feasibility of this approach by measuring the turnover of multiple isoforms of parvalbumin (β1-7) in the skeletal muscle of common carp (Cyprinus carpio). A stable isotope-labelled amino acid ([(2)H(7)] l-leucine) was administered to the carp via the diet and its incorporation into the isoforms of parvalbumin in muscle over time was monitored by LC-MS analysis of signature peptides. The relative isotope abundance was calculated and used to deconvolute the data. The β7 parvalbumin isoform had a rate of synthesis that was greater than the rate of degradation. In contrast the rate of degradation of the β5 isoform exceeded its rate of synthesis, whilst the analysis revealed that the other parvalbumin β-isoforms (β1, β2, β3, β4 and β6) had a rate of synthesis that was equal to the rate of degradation. This work has addressed a number of technical challenges and represents the first study to use proteomic approaches to measure the turnover of individual proteins in fish.     

The diversity of protein turnover and abundance under nitrogen-limited steady-state conditions in Saccharomyces cerevisiae

To establish more advanced models of molecular dynamics within cells, protein characteristics such as turnover rate and absolute instead of relative abundance have to be analyzed. We applied a proteomics strategy to analyze protein degradation and abundance in Saccharomyces cerevisiae. We used steady-state chemostat cultures to ascertain well-defined growth conditions and nitrogen limited media, which allowed us to rapidly switch from (14)N to (15)N-isotope containing media and to monitor the decay of the (14)N mono-isotope signals in time. We acquired both protein abundance information and degradation rates of 641 proteins. Half-lives of individual proteins were very diverse under nitrogen-limited steady-state conditions, ranging from less than 30 min to over 20 h. Proteins that act as single physical complexes do not always show alike half-lives. For example the chaperonin-containing TCP-1 complex showed similar intermediate half-lives ranging from 7 to 20 h. In contrast, the ribosome exhibited a wide diversity of half-lives ranging from 2.5 to over 20 h, although their cellular abundances were rather similar. The stabilities of proteins involved in the central sugar metabolism were found to be intermediary, except for the glycolytic enzymes Hxk1p and Fba1p and the TCA-cycle proteins Lsc2p and Kgd1p, which showed half-lives of over 20 h. These data stress the need for inclusion of quantitative data of protein turn-over rates in yeast systems biology.     

Protein degradation in skeletal muscle: implications of a first order reaction for the degradative process

The rate of protein degradation is usually thought to be first order, i.e. determined by the nature of the protein as a substrate. It is not immediately apparent if this is the case for the overall process in the cell since rates of turnover of individual proteins may vary between tissues. In muscle the characteristics of protein turnover in relation to DNA-unit size have led to the development of a model for protein turnover in which degradation rates are determined by the rate of dissociation of protein subunits from the myofibrillar matrix. This is a necessary step if heterogeneous turnover occurs and if degradation and resynthesis of myofibrillar proteins occurs peripherally to the myofibril. As a result a first order rate can be envisaged so that during muscle growth the protein mass per unit DNA increases to a characteristic amount thus determining the specific activity of the degrading system. Such a mechanism may apply to all cells.     

Accumulation of “Old Proteins” and the Critical Need for MS‐based Protein Turnover Measurements in Aging and Longevity

Aging and age‐related diseases are accompanied by proteome remodeling and progressive declines in cellular machinery required to maintain protein homeostasis (proteostasis), such as autophagy, ubiquitin‐mediated degradation, and protein synthesis. While many studies have focused on capturing changes in proteostasis, the identification of proteins that evade these cellular processes has recently emerged as an approach to studying the aging proteome. With advances in proteomic technology, it is possible to monitor protein half‐lives and protein turnover at the level of individual proteins in vivo. For large‐scale studies, these technologies typically include the use of stable isotope labeling coupled with MS and comprehensive assessment of protein turnover rates. Protein turnover studies have revealed groups of highly relevant long‐lived proteins (LLPs), such as the nuclear pore complexes, extracellular matrix proteins, and protein aggregates. Here, the role of LLPs during aging and age‐related diseases and the methods used to identify and quantify their changes are reviewed. The methods available to conduct studies of protein turnover, used in combination with traditional proteomic methods, will enable the field to perform studies in a systems biology context, as changes in proteostasis may not be revealed in studies that solely measure differential protein abundances.     

Impact of oxidative stress on the function,abundance, and turnover of the Arabidopsis 80S cytosolic ribosome

Abiotic stress in plants causes accumulation of reactive oxygen species (ROS) leading to the need for new protein synthesis to defend against ROS and to replace existing proteins that are damaged by oxidation. Functional plant ribosomes are critical for these activities, however we know little about the impact of oxidative stress on plant ribosome abundance, turnover, and function. Using Arabidopsis cell culture as a model system, we induced oxidative stress using 1 µm of H₂O₂ or 5 µm menadione to more than halve cell growth rate and limit total protein content. We show that ribosome content on a total cell protein basis decreased in oxidatively stressed cells. However, overall protein synthesis rates on a ribosome abundance basis showed the resident ribosomes retained their function in oxidatively stressed cells. ¹⁵N progressive labelling was used to calculate the rate of ribosome synthesis and degradation to track the fate of 62 r‐proteins. The degradation rates and the synthesis rates of most r‐proteins slowed following oxidative stress leading to an ageing population of ribosomes in stressed cells. However, there were exceptions to this trend; r‐protein RPS14C doubled its degradation rate in both oxidative treatments. Overall, we show that ribosome abundance decreases and their age increases with oxidative stress in line with loss of cell growth rate and total cellular protein amount, but ribosome function of the ageing ribosomes appeared to be maintained concomittently with differences in the turnover rate and abundance of specific ribosomal proteins. Data are available via ProteomeXchange with identifier PXD012840.     

Growth rate correlates negatively with protein turnover in Arabidopsis accessions

Previous studies with Arabidopsis accessions revealed that biomass correlates negatively to dusk starch content and total protein, and positively to the maximum activities of enzymes in photosynthesis. We hypothesized that large accessions have lower ribosome abundance and lower rates of protein synthesis, and that this is compensated by lower rates of protein degradation. This would increase growth efficiency and allow more investment in photosynthetic machinery. We analysed ribosome abundance and polysome loading in 19 accessions, modelled the rates of protein synthesis and compared them with the observed rate of growth. Large accessions contained less ribosomes than small accessions, due mainly to cytosolic ribosome abundance falling at night in large accessions. The modelled rates of protein synthesis resembled those required for growth in large accessions, but were up to 30% in excess in small accessions. We then employed ¹³CO₂ pulse‐chase labelling to measure the rates of protein synthesis and degradation in 13 accessions. Small accessions had a slightly higher rate of protein synthesis and much higher rates of protein degradation than large accessions. Protein turnover was negligible in large accessions but equivalent to up to 30% of synthesised protein day^?1 in small accessions. We discuss to what extent the decrease in growth in small accessions can be quantitatively explained by known costs of protein turnover and what factors may lead to the altered diurnal dynamics and increase of ribosome abundance in small accessions, and propose that there is a trade‐off between protein turnover and maximisation of growth rate.     

Calculation of half-lives of proteins in vivo. Heterogeneity in the rate of degradation of yeast proteins

Summary A method is given for the calculation of half-lives of proteins in vivo from the measurement of the decrease of radioactivity in pulse-labelled proteins with time. This method could be particularly useful for the study of the degradation of proteins in cells that have a low growth rate.The method applied to growing yeast indicates that there are two major classes of protein. The class with low turnover constitutes the bulk of yeast protein and has a half-life of 160 h in a medium with glucose or galactose and of 50 h in a medium with ethanol. The class of proteins with high turnover (half-life between 0.8 and 2.4 hours) represents from 1% of total protein in yeast growing on glucose to 7% in yeast growing on ethanol.It is shown that some proteins which are derepressed during growth on ethanol or induced during growth on galactose are particularly susceptible to degradation in a medium which contains glucose.It is proposed that protein degradation is regulated by a coarse control at the level of protease activity and a fine control on the susceptibility of individual proteins to proteases.     

Proteome dynamics in complex organisms: using stable isotopes to monitor individual protein turnover rates

The complete definition of changes in a proteome requires information about dynamics and specifically the rate at which the individual proteins are turned over intracellularly. Whilst this can be achieved in single-cell culture using stable isotope precursors, it is more challenging to develop methods for intact animals. In this study, we show how dietary administration of stable isotope-labelled amino acids can obtain information on the relative rates of synthesis and degradation of individual proteins in a proteome. The pattern of stable isotope-labelling in tryptic peptides can be deconstructed to yield a highly reliable measure of the isotope abundance of the precursor pool, a parameter that is often difficult to acquire. We demonstrate this approach using chickens fed a semisynthetic diet containing [(2)H(8)]valine at a calculated relative isotope abundance (RIA) of 0.5. When the labelling pattern of gel-resolved muscle proteins was analyzed, the intracellular precursor isotope abundance was 0.35, consistent with dilution of the amino acid precursor pool with unlabelled amino acids derived from degradation of pre-existing proteins. However, the RIA was stable over an extended labelling window, and permitted calculation of the rates of synthesis and degradation of individual proteins isolated by gel electrophoresis. For the first time, it is feasible to contemplate the analysis of turnover of individual proteins in intact animals.     

Protein turnover: measurement of proteome dynamics by whole animal metabolic labelling with stable isotope labelled amino acids

The measurement of protein turnover in tissues of intact animals is obtained by whole animal dynamic labelling studies, requiring dietary administration of precursor label. It is difficult to obtain full labelling of precursor amino acids in the diet and if partial labelling is used, calculation of the rate of turnover of each protein requires knowledge of the precursor relative isotope abundance (RIA). We describe an approach to dynamic labelling of proteins in the mouse with a commercial diet supplemented with a pure, deuterated essential amino acid. The pattern of isotopomer labelling can be used to recover the precursor RIA, and sampling of urinary secreted proteins can monitor the development of liver precursor RIA non-invasively. Time-series analysis of the labelling trajectories for individual proteins allows accurate determination of the first order rate constant for degradation. The acquisition of this parameter over multiple proteins permits turnover profiling of cellular proteins and comparisons of different tissues. The median rate of degradation of muscle protein is considerably lower than liver or kidney, with heart occupying an intermediate position.     

Cardiac Myosin Filaments are Maintained by Stochastic Protein Replacement

Myosin and myosin-binding protein C are exquisitely organized into giant filamentous macromolecular complexes within cardiac muscle sarcomeres, yet these proteins must be continually replaced to maintain contractile fidelity. The overall hypothesis that myosin filament structure is dynamic and allows for the stochastic replacement of individual components was tested in vivo, using a combination of mass spectrometry– and fluorescence-based proteomic techniques. Adult mice were fed a diet that marked all newly synthesized proteins with a stable isotope-labeled amino acid. The abundance of unlabeled and labeled proteins was quantified by high-resolution mass spectrometry over an 8-week period. The rates of change in the abundance of these proteins were well described by analytical models in which protein synthesis defined stoichiometry and protein degradation was governed by the stochastic selection of individual molecules. To test whether the whole myosin filaments or the individual components were selected for replacement, cardiac muscle was chemically skinned to remove the cellular membrane and myosin filaments were solubilized with ionic solutions. The composition of the filamentous and soluble fractions was quantified by mass spectrometry, and filament depolymerization was visualized by real-time fluorescence microscopy. Myosin molecules were preferentially extracted from ends of the filaments in the presence of the ionic solutions, and there was only a slight bias in the abundance of unlabeled molecules toward the innermost region on the myosin filaments. These data demonstrate for the first time that the newly synthesized myosin and myosin-binding protein C molecules are randomly mixed into preexisting thick filaments in vivo and the rate of mixing may not be equivalent along the length of the thick filament. These data collectively support a new model of cardiac myosin filament structure, with the filaments being dynamic macromolecular assemblies that allow for replacement of their components, rather than rigid bodies.     

A Sampling of the Yeast Proteome

In this study, we examined yeast proteins by two-dimensional (2D) gel electrophoresis and gathered quantitative information from about 1,400 spots. We found that there is an enormous range of protein abundance and, for identified spots, a good correlation between protein abundance, mRNA abundance, and codon bias. For each molecule of well-translated mRNA, there were about 4,000 molecules of protein. The relative abundance of proteins was measured in glucose and ethanol media. Protein turnover was examined and found to be insignificant for abundant proteins. Some phosphoproteins were identified. The behavior of proteins in differential centrifugation experiments was examined. Such experiments with 2D gels can give a global view of the yeast proteome.     

Determining degradation and synthesis rates of arabidopsis proteins using the kinetics of progressive 15N labeling of two-dimensional gel-separated protein spots

The growth and development of plant tissues is associated with an ordered succession of cellular processes that are reflected in the appearance and disappearance of proteins. The control of the kinetics of protein turnover is central to how plants can rapidly and specifically alter protein abundance and thus molecular function in response to environmental or developmental cues. However, the processes of turnover are largely hidden during periods of apparent steady-state protein abundance, and even when proteins accumulate it is unclear whether enhanced synthesis or decreased degradation is responsible. We have used a (15)N labeling strategy with inorganic nitrogen sources coupled to a two-dimensional fluorescence difference gel electrophoresis and mass spectrometry analysis of two-dimensional IEF/SDS-PAGE gel spots to define the rate of protein synthesis (K(S)) and degradation (K(D)) of Arabidopsis cell culture proteins. Through analysis of MALDI-TOF/TOF mass spectra from 120 protein spots, we were able to quantify K(S) and K(D) for 84 proteins across six functional groups and observe over 65-fold variation in protein degradation rates. K(S) and K(D) correlate with functional roles of the proteins in the cell and the time in the cell culture cycle. This approach is based on progressive (15)N labeling that is innocuous for the plant cells and, because it can be used to target analysis of proteins through the use of specific gel spots, it has broad applicability.     

Avian proteomics: advances, challenges and new technologies

Proteomics is defined as an analysis of the full complement of proteins of a cell or tissue under given conditions. Avian proteomics, or more specifically chicken proteomics, has focussed on the study of individual tissues and organs of interest to specific researchers. Researchers have looked at skeletal muscle and growth, and embryonic development and have performed initial studies in avian disease. Traditional proteomics involves identifying and cataloguing proteins in a cell and identifying relative changes in populations between two or more states, be that physiological or disease-induced states. Recent advances in proteomic technologies have included absolute quantification, proteome simplification and the ability to determine the turnover of individual proteins in a global context. This review discusses the current developments in this relatively new field, new technologies and how they may be applied to biological questions, and the challenges faced by researchers in this ever-expanding and exciting field.     

Protein Turnover in Retina

Abstract: Rabbit retinas were exposed in vitro to 0.5-h pulses of [³H]leucine or [¹⁴C]Ieucine. Some retinas were harvested promptly after labeling to measure synthesis. These were combined, in double-labeling experiments, with retinas that had been returned to unlabeled medium for a subsequent 1 h or 3.75 h to measure degradation. All of the proteins were solubilized, and separated according to size by gel electrophoresis. The gels were cut into 95 slices, and each slice was differentially counted. The amount of protein in the slice was estimated from the Coomassie blue staining, and its molecular weight from the distribution of molecular weight (MW) standards. Turnover rates of the various sizes of proteins were calculated from these data using certain well-defined assumptions. Retinal protein contained about 32 ± 10³ nmol of polypeptide per g, with a median MW of 27,000. Total synthesis was at the rate of 103 nmol/g of protein/h, with the most rapid synthesis in the 33,000–43,000 MW range, at 2 nmol/g/h for every 1000 increment in MW. Protein renewal averaged 0.52%/h, but varied directly (p < 0.0001) with MW, so that proteins of 10,000 MW were being renewed at about 0.1%/h and proteins of 140,000 MW at about 1.4%/h. Taken together, the measurements of fractional renewal and the measurements of degradation of the newly synthesized proteins demonstrated that each slice contained proteins with markedly different breakdown coefficients, and provided enough information to characterize the proteins in the slice in terms of a fast and a slow subgroup. This analysis indicated that: breakdown coefficients varied much more than rates of synthesis and were therefore the prime determinant of the amount of each protein that was present; as MW increased, breakdown coefficients of the long-lived proteins increased (p < 0.0001), accounting in major part for the correlation between size and turnover; most staining bands were due to proteins with peculiarly long lifespans; the proteins with the slowest turnover of all appeared to be histones: there was an unusually rapid synthesis of a 138,000 MW polypeptide with a moderately short half-life (about 3 h).     

Role of the calpain system in muscle growth.

Muscle protein degradation has an important role in rate of muscle growth. It has been difficult to develop procedures for measuring rate of muscle protein degradation in living animals, and most studies have used in vitro systems and muscle strips to determine rate of protein degradation. The relationship between results obtained by using muscle strips and rate of muscle protein turnover in living animals is unclear because these strips are in negative nitrogen balance and often develop hypoxic cores. Also, rate of protein degradation is usually estimated by release of labeled amino acids, which reflects an average rate of degradation of all cellular proteins and does not distinguish between rates of degradation of different groups of proteins such as the sarcoplasmic and the myofibrillar proteins in muscle. A number of studies have suggested that the calpain system initiates turnover of myofibrillar proteins, which are the major group of proteins in striated muscle, by making specific cleavages that release thick and thin filaments from the surface of the myofibril and large polypeptide fragments from some of the other myofibrillar proteins. The calpains do not degrade myofibrillar proteins to small peptides or to amino acids, and they cause no bulk degradation of sarcoplasmic proteins. Hence, the calpains are not directly responsible for release of amino acids during muscle protein turnover. Activity of the calpains in living cells is regulated by calpastatin and Ca2+, but the nature of this regulation is still unclear.     

Automated live cell imaging of green fluorescent protein degradation in individual fibroblasts.

To accurately interpret the data from fluorescent proteins as reporters of gene activation within living cells, it is important to understand the kinetics of the degradation of the reporter proteins. We examined the degradation kinetics over a large number (>1,000) of single, living cells from a clonal population of NIH3T3 fibroblasts that were stably transfected with a destabilized, enhanced green fluorescent protein (eGFP) reporter driven by the tenascin-C promoter. Data collection and quantification of the fluorescence protein within a statistically significant number of individual cells over long times (14 h) by automated microscopy was facilitated by culturing cells on micropatterned arrays that confined their migration and allowed them to be segmented using phase contrast images. To measure GFP degradation rates unambiguously, protein synthesis was inhibited with cycloheximide. Results from automated live cell microscopy and image analysis indicated a wide range of cell-to-cell variability in the GFP fluorescence within individual cells. Degradation for this reporter was analyzed as a first order rate process with a degradation half-life of 2.8 h. We found that GFP degradation rates were independent of the initial intensity of GFP fluorescence within cells. This result indicates that higher GFP abundance in some cells is likely due to higher rates of gene expression, because it is not due to systematically lower rates of protein degradation. The approach described in this study will assist the quantification and understanding of gene activity within live cells using fluorescent protein reporters.