pulse and chase experiments

Pulse-chase analysis for studying protein synthesis and maturation

Affiliation.

• 1 Department of Life Sciences and Chemistry, Jacobs University Bremen, Bremen, Germany.
• PMID: 25367008
• DOI: 10.1002/0471140864.ps3003s78

Pulse-chase analysis is a well-established and highly adaptable tool for studying the life cycle of endogenous proteins, including their synthesis, folding, subunit assembly, intracellular transport, post-translational processing, and degradation. This unit describes the performance and analysis of a radiolabel pulse-chase experiment for following the folding and cell surface trafficking of a trimeric murine MHC class I glycoprotein. In particular, the unit focuses on the precise timing of pulse-chase experiments to evaluate early/short-time events in protein maturation in both suspended and strictly adherent cell lines. The advantages and limitations of radiolabel pulse-chase experiments are discussed, and a comprehensive section for troubleshooting is provided. Further, ways to quantitatively represent pulse-chase results are described, and feasible interpretations on protein maturation are suggested. The protocols can be adapted to investigate a variety of proteins that may mature in very different ways.

Keywords: glycan processing; immunoprecipitation; metabolic labeling; protein folding; protein transport; secretory pathway.

Copyright © 2014 John Wiley & Sons, Inc.

Publication types

• Histocompatibility Antigens Class I / biosynthesis*
• Isotope Labeling / methods*
• Protein Biosynthesis / physiology*
• Protein Folding*
• Histocompatibility Antigens Class I

METHODS article

Spaac pulse-chase: a novel click chemistry-based method to determine the half-life of cellular proteins.

• 1 Molecular Medicine Research Group, Robarts Research Institute, Western University, London, ON, Canada
• 2 Department of Physiology and Pharmacology, Western University, London, ON, Canada
• 3 Department of Anatomy and Cell Biology, Schulich School of Medicine & Dentistry, Western University, London, ON, Canada

Assessing the stability and degradation of proteins is central to the study of cellular biological processes. Here, we describe a novel pulse-chase method to determine the half-life of cellular proteins that overcomes the limitations of other commonly used approaches. This method takes advantage of pulse-labeling of nascent proteins in living cells with the bioorthogonal amino acid L-azidohomoalanine (AHA) that is compatible with click chemistry-based modifications. We validate this method in both mammalian and yeast cells by assessing both over-expressed and endogenous proteins using various fluorescent and chemiluminescent click chemistry-compatible probes. Importantly, while cellular stress responses are induced to a limited extent following live-cell AHA pulse-labeling, we also show that this response does not result in changes in cell viability and growth. Moreover, this method is not compromised by the cytotoxicity evident in other commonly used protein half-life measurement methods and it does not require the use of radioactive amino acids. This new method thus presents a versatile, customizable, and valuable addition to the toolbox available to cell biologists to determine the stability of cellular proteins.

Introduction

The stability and degradation of cellular proteins are critical parameters that influence and regulate most aspects of cellular physiology and many disease-associated processes. Targeted proteolysis of cellular proteins in a timely manner ensures that cells divide on cue, maintain their proper functions, or undergo apoptosis when appropriate ( Varshavsky, 2008 ; Knecht et al., 2009 ). Consequently, dysregulated proteolysis of key proteins is associated with many human diseases ( Hanna et al., 2019 ), for example the tumor suppressor p53 in cancer ( Hengstermann et al., 2001 ; Mantovani et al., 2019 ), α-synuclein, β-amyloid, and other misfolding-prone proteins in neurodegenerative diseases ( Baranello et al., 2015 ; Ciechanover and Kwon, 2015 ; Lehtonen et al., 2019 ), and folding-mutants of cystic fibrosis transmembrane conductance regulator in cystic fibrosis ( Sharma et al., 2004 ). Thus, quantification of the rate of degradation of cellular proteins, i.e., determination of protein half-lives, is an essential tool when studying physiological and disease-related cellular processes and can provide insight into the stability, regulation, mechanisms of degradation, and function of cellular proteins.

To date, three major methods are commonly used to determine protein half-life, each of which has specific advantages but also severe limitations ( Eldeeb et al., 2019 ). These include (1) treatment of cells with cycloheximide (CHX), a fast and effective, yet highly cytotoxic, inhibitor of protein synthesis ( Schneider-Poetsch et al., 2010 ), (2) pulse-chase experiments that involve labeling of newly synthesized proteins in living cells with radioisotopic amino acids such as 35 S-methionine ( Takahashi and Ono, 2003 ; Esposito and Kinzy, 2014 ), and (3) expression of a protein of interest as a GFP (green fluorescent protein) fusion, where GFP is either photoactivatable ( Zhang et al., 2007 ) or can be permanently photobleached ( Eden et al., 2011 ). Radiolabeling of nascent cellular proteins is often considered the gold standard for pulse-chase analysis and involves minimal disturbance to normal cellular conditions. However, a significant disadvantage of this method is the use of potentially biohazardous radioisotopes, the requirements for special permits, and the use of specific protocols and equipment in a containment environment. Furthermore, radiolabeling can induce DNA and cellular damage, lead to cell cycle arrest, alter cell morphology, and induce apoptosis ( Hu and Heikka, 2000 ; Hu et al., 2001 ). Alternatively, CHX can be used with standard laboratory equipment without the need for radioisotope precautions and is often preferred because of its simplicity. However, treatment of cells with CHX inhibits de novo protein synthesis and non-specifically affects a wide array of cellular processes, including kinase pathways and proteolytic machinery ( Hanna et al., 2003 ; Dai et al., 2013 ). Thus, the use of CHX treatment is not suitable when studying proteins with long half-lives, and furthermore may affect the protein half-life of cellular proteins in unintended ways. Protocols involving either 35 S-methionine pulse-chase labeling or CHX treatment require cell lysis, whereas GFP tags allow for monitoring protein half-life in living cells by microscopy. However, this latter method requires the heterologous expression of a GFP-tagged protein, and thus cannot be used to study endogenous proteins. Moreover, expression of GFP alone or as a fusion protein can induce proteome changes, alter kinase and ubiquitin signaling pathways, and cause cellular toxicity ( Baens et al., 2006 ; Coumans et al., 2014 ; Ansari et al., 2016 ).

Considering these methodological shortcomings, we identified a need for a more reliable and less constrained method to determine the half-life of cellular proteins. A new approach designed originally to measure global changes in proteome dynamics involves the labeling of living cells with L-azidohomoalanine (AHA), a bioorthogonal methionine analog containing a reactive azide moiety that is selectively incorporated into newly synthesized proteins ( Kiick et al., 2002 ; Dieterich et al., 2006 ). AHA-labeled proteins can then be reacted with an alkyne-containing molecule in a click chemistry reaction that allows for the isolation of AHA-labeled proteins for downstream proteomic analysis ( McShane et al., 2016 ). A major advantage of AHA labeling over 35 S-methionine labeling, CHX treatment, or the use of GFP fusion proteins is that AHA is non-toxic, non-radioactive, does not inhibit protein synthesis, and does not alter global protein ubiquitination or degradation ( Dieterich et al., 2006 ). While AHA-labeling has been used previously both in vitro and in vivo to measure global changes in protein synthesis without affecting cellular viability ( Dieterich et al., 2006 , 2010 ; Baskin et al., 2007 ; Roche et al., 2009 ; McShane et al., 2016 ), this technique has not yet been applied to studying the stability and degradation of individual proteins-of-interest.

Here, we describe a novel pulse-chase procedure for the non-toxic and non-radioactive determination of cellular protein half-life using click chemistry that can monitor specific proteins-of-interest ( Morey et al., 2016 ). This method is an adaptation of classical 35 S-methionine pulse-chase labeling that utilizes copper-free strain-promoted alkyne-azide cycloaddition (SPAAC) reactions to conjugate a fluorescent or biotin cyclooctyne probe onto newly synthesized AHA-labeled proteins. Following immunoprecipitation of a protein-of-interest, half-life of an AHA-labeled protein can be monitored by standard SDS-PAGE and immunoblotting. We provide examples for the application of this method to measure the half-life of both endogenous and over-expressed proteins in mouse and human cell lines, and in yeast cells. We also include a proof-of-principle example that provides novel insights into cholinergic neurobiology ( Morey et al., 2016 ). Additionally, we compare various commercially available fluorescent and biotin cyclooctyne probes for detection of AHA-labeled proteins.

Collectively, this method utilizes bioorthogonal click chemistry reactions in a manner that is compatible in a variety of cell systems and that allows user customization, thus establishing a versatile method with a wide applicability to molecular and cell biologists. To our knowledge, our study is the first to report on the use of click chemistry-based labeling to determine the protein half-life of specific proteins with high sensitivity. SPAAC pulse-chase thus presents a novel, reliable, and effective method to study protein stability in living cells.

Materials and Methods

Mammalian cell culture and cell lysis.

Mouse cholinergic SN56 neural cells (gift from Dr. J. K. Blusztajn, Boston University) ( Blusztajn et al., 1992 ) or human HEK293 or HeLa cells (ATCC) were grown as monolayers in DMEM supplemented with either 5% (SN56) or 10% FBS (HEK293 and HeLa; Invitrogen), and 1% Pen-Strep at 37°C with 5% CO 2 . Prior to experiments assessing the half-life of human 69-kDa choline acetyltransferase (ChAT) protein, cells were transiently transfected for 18–24 h at 37°C using either Lipofectamine 2000 or 3000 (Invitrogen) at ∼50% confluence with a plasmid encoding either wild-type or mutant proline-to-alanine P17A/P19A ChAT cDNA ligated to pcDNA3.1+ vector ( Dobransky et al., 2000 ; Morey et al., 2016 ). Following treatments, cells were collected and lysed on ice in RIPA buffer (50 mM Tris–HCl; pH 8.0, 150 mM NaCl, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1–0.5% SDS) supplemented with mammalian protease inhibitor cocktail (Sigma), phosphatase inhibitor cocktail (10 mM NaF, 1 mM Na 3 VO 4 , 20 mM Na 2 HPO 4 , 3 mM β-glycerolphosphate, 5 mM sodium pyrophosphate), 50 μM MG132, 10 mM N-ethylmaleimide (NEM; Calbiochem), and 800 U/ml DNase I (Invitrogen). Lysates were centrifuged for 10 min at 21,000 g at 4°C and protein concentrations were measured by BCA protein assay (Thermo). Aliquots of lysate supernatant were either used for immunoprecipitations or denatured in 1× Laemmli sample buffer (63 mM Tris–HCl; pH 6.8, 10% glycerol, 2% SDS, 0.005% bromophenol blue, 2.5% 2-mercaptoethanol) at 95°C for 10 min, then analyzed by SDS-PAGE and immunoblotting.

SPAAC Pulse-Chase in Mammalian Cells

SN56 cells, transiently expressing either wild-type or mutant P17A/P19A-ChAT, or wild-type HEK293 cells were grown to ∼70% confluence prior to the start of SPAAC pulse-chase. To begin, cells were washed twice with methionine-free (Met-) DMEM (Invitrogen) to remove excess methionine then live-labeled (pulsed) at 37°C for 4 h with 50 μM Click-iT L-azidohomoalanine (AHA; Invitrogen) in Met- DMEM supplemented with 5–10% dialyzed FBS (Invitrogen). Control cells were grown for 4 h in methionine-containing (Met+) DMEM with 5–10% FBS. AHA-labeled cells were then washed twice with Met+ DMEM, and subsequently incubated in Met+ DMEM with 5–10% FBS for 2–24 h at 37°C (chase). Additionally, for determination of ChAT protein half-life during proteasome inhibition, SN56 cells were treated with 5 μM MG132 throughout both the 4 h pulse and 24 h chase period (i.e., 28 h). Cells were collected on ice either immediately following the pulse (control and 0 h) or following the chase periods (2, 4, 10, and 24 h), lysed in supplemented RIPA buffer (above) containing 0.5% SDS and 10 mM of freshly prepared iodoacetamide ( Van Geel et al., 2012 ), then centrifuged for 10 min at 21,000 g at 4°C.

For fluorescent/biotin cyclooctyne labeling of AHA-labeled proteins from whole cell lysates, aliquots of cleared cell lysate were reacted with either 10 μM Click-iT TAMRA-DIBO (Invitrogen), 5 μM AFDye 488-DBCO (Click Chemistry Tools), or 10 μM Click-iT Biotin-DIBO (Invitrogen) for 1 h at 21°C followed by denaturation in 1× Laemmli sample buffer at 95°C for 10 min. For analysis, equal amounts of proteins (e.g., 25 μg) were resolved on SDS-PAGE gels and fluorescence was detected in-gel using a ChemiDoc MP system (Bio-Rad) at an excitation/emission of either 555/580 nm (TAMRA-DIBO) or 494/517 nm (488-DBCO). Alternatively, if reacted with Biotin-DIBO, protein samples were resolved on SDS-PAGE gels, transferred to PVDF membranes (Bio-Rad), then membranes were probed with Pierce High Sensitivity Streptavidin-HRP (Thermo) and Clarity Western ECL Substrate (Bio-Rad). As controls, AHA/cyclooctyne-labeled samples from whole cell lysates were transferred to PVDF membranes and standard immunoblotting was completed.

To determine the protein half-life of either ChAT or p53, anti-ChAT or anti-p53 IPs were prepared from cleared whole cell lysates of AHA-labeled cells as detailed below. Immunocaptured samples were washed twice with cold 0.5%-SDS RIPA buffer, twice with cold PBS to remove detergents, then subsequently reacted with strained cyclooctynes (10 μM TAMRA-DIBO, 5 μM 488-DBCO, or 10 μM Biotin-DIBO) in 1× PBS supplemented with mammalian protease inhibitors for 1 h at 21°C with gentle agitation. Samples were washed twice with cold 0.5%-SDS RIPA buffer and IP samples was eluted into 50 μl of 2× Laemmli sample buffer with 5% 2-mercaptoethanol at 85°C for 15 min with intermittent mixing. For analysis, equal volumes of immunoprecipitated AHA/cyclooctyne-labeled ChAT or p53 (e.g., 20 μl) were resolved on SDS-PAGE gels and either in-gel fluorescence (TAMRA-DIBO or 488-DBCO) or biotin (Biotin-DIBO) labeling was detected as above. Lastly, as controls, AHA/cyclooctyne-labeled IP samples were transferred to PVDF membranes and anti-ChAT or anti-p53 immunoblots were completed.

To calculate protein half-life, we assumed that the amount of AHA/cyclooctyne-labeled ChAT or p53, P ( t ), decays exponentially under first order kinetics according to the equation P ( t ) = P o e –αt , where P o is the fluorescence (TAMRA-DIBO or 488-DBCO) or biotin-ECL (Biotin-DIBO) intensity at t = 0. The slope of decay (α) was calculated by plotting the intensity of immunoprecipitated AHA/cyclooctyne-labeled ChAT or p53, corrected for the levels of total immunoprecipitated ChAT or p53 as measured by parallel immunoblotting, on a semi-logarithmic scale and performing linear regression. ChAT and p53 protein half-life, T 1/2 , was calculated according to first order kinetics where T 1/2 = ln(2)/α ( Eden et al., 2011 ).

Cycloheximide (CHX) Assay

SN56 cells were transfected and plated as for SPAAC pulse-chase to transiently express either wild-type or mutant P17A/P19A-ChAT, then treated with 100 μg/ml CHX for 2, 4, 6, or 8 h; control cells were treated with DMSO. Cells were collected and lysed on ice in supplemented 0.1%-SDS RIPA buffer, lysates were centrifuged for 10 min at 21,000 g at 4°C, then protein samples from cleared whole cell lysates were denatured in 1× Laemmli sample buffer at 95°C for 5 min. Protein samples were resolved on SDS-PAGE gels, transferred to PVDF membranes, then immunoblotting was completed. To determine the protein half-life of wild-type and P17A/P19A-ChAT, anti-ChAT immunoreactive bands were quantified by densitometry, plotted on a semi-logarithmic scale, and analyzed as for SPAAC pulse-chase by linear regression to determine a slope of decay.

Serial Dilution Assay for Detection Sensitivity of Strained Cyclooctynes

HEK293 cells were live-labeled in culture for 4 h with 50 μM AHA in methionine-free DMEM then collected immediately without a methionine chase. Control unlabeled cells were incubated in Met+ DMEM. Cells were collected and lysed on ice in supplemented 0.1%-SDS RIPA buffer, lysates were centrifuged for 10 min at 21,000 g at 4°C, and aliquots of cleared whole cell lysate were reacted with either 10 μM TAMRA-DIBO, 5 μM 488-DBCO, or 10 μM Biotin-DIBO for 1 h at 21°C. Protein samples were denatured in 1× Laemmli sample buffer at 95°C for 10 min then were serially diluted 1:1 with 1× Laemmli sample buffer a total of six times until reaching a final dilution of 1:64. For analysis, 25 μg of total protein initially, then an equal volume from each serially diluted samples (1:64 = 0.39 μg total protein), were resolved on SDS-PAGE gels and either in-gel fluorescence (TAMRA-DIBO or 488-DBCO) or biotin (Biotin-DIBO) labeling was detected as above. Anti-actin immunoblots were completed as a loading control.

Analysis of Cell Viability, Global Proteome Ubiquitination, Protein Solubility, and Heat Shock Response in Mammalian Cells Labeled With AHA

Mouse SN56 cells or human HEK293 or HeLa cells were live-labeled in culture for 4 h with 50 μM AHA and either collected immediately (0 h) or following 8 h of chase in Met+ DMEM. Control unlabeled cells were incubated in media with methionine (i.e., without AHA). To inhibit de novo protein synthesis SN56 cells were co-treated with 100 μg/ml CHX throughout both the 4 h AHA pulse and the 8 h chase periods (up to 12 h total); control unlabeled cells were treated with CHX for 4 h in Met+ DMEM, while control AHA-labeled cells were treated with DMSO. To induce apoptosis or protein misfolding, cells were treated in Met+ DMEM with either 200 nM staurosporine or 10 mM azetidine-2-caroxylic acid (AZC) for either 8 h or 24 h, respectively. Cells were lysed on ice in supplemented 0.1%-SDS RIPA buffer, lysates were centrifuged for 10 min at 21,000 g at 4°C, and aliquots of cleared whole cell lysate were reacted with 10 μM TAMRA-DIBO for 1 h at 21°C. Protein samples were then denatured in 1× Laemmli sample buffer at 95°C for 10 min, run on SDS-PAGE gels and AHA-labeled proteins were detected in-gel at an Ex/Em 555/580 nm. Subsequently, protein samples were transferred to PVDF membranes and immunoblotting was completed as indicated.

Alternatively, to determine if AHA may affect the overall solubility of cellular proteins, SN56 cells were live-labeled in culture for 4 h with 50 μM AHA and either collected immediately (0 h) or following 8 h of chase in Met+ DMEM as above. Additionally, as a positive control to induce protein misfolding cells were treated in Met+ DMEM with 10 mM AZC for 24 h. Cells were collected and lysed on ice in 0.1% Triton X-100 lysis buffer (50 mM Tris–HCl; pH 8.0, 150 mM NaCl, 0.1% Triton X-100) supplemented with protease/phosphatase inhibitors, 50 μM MG132, and 10 mM NEM. Lysates were centrifuged for 15 min at 15,000 g at 4°C and aliquots of Triton-soluble supernatant were prepared for immunoblotting by denaturing in 1× Laemmli sample buffer at 95°C for 10 min. To prepare Triton-insoluble proteins for immunoblotting the Triton-insoluble pellets were washed once with ice-cold PBS, then denatured in an equal volume of 2× Laemmli sample buffer with 5% 2-mercaptoethanol at 85°C for 15 min prior to separation on SDS-PAGE gels and immunoblotting.

Immunoprecipitation (IP)

For both anti-ChAT and anti-p53 IPs, cells were grown and treated on either 60 or 100 mm culture dishes to ∼90% confluence prior to collection. Cells were lysed in supplemented RIPA buffer (above) containing 0.5% SDS, then lysates were centrifuged for 10 min at 21,000 g at 4°C. Aliquots of cell lysate supernatants containing 1 mg protein were diluted to a final volume of 1 ml (1 mg/ml final) in supplemented RIPA buffer, then IP samples were incubated at 4°C for 18 h with either 2.5 μg of anti-ChAT primary antibody (CTab) ( Dobransky et al., 2000 ) or with 2 μg of anti-p53 primary antibody (DO-1; Santa Cruz) per mg protein. Immune complexes were captured onto 50 μl of protein-G Dynabeads (Invitrogen) for 1 h at 4°C, then washed with cold RIPA buffer and used for SPAAC pulse-chase as detailed above.

SDS-PAGE and Immunoblotting

Denatured protein samples from whole cell lysates and IPs were resolved on 7.5, 10, or 12% SDS-PAGE gels, then transferred to PVDF membranes by semi-dry electroblotting. For immunoblotting, membranes were blocked for 1 h at 21°C in 5% non-fat milk powder in PBS (137 mM NaCl, 2.7 mM KCl, 10 mM Na 2 HPO 4 , 1.8 mM KH 2 PO 4 ; pH 7.4) containing 0.15% Triton X-100 (PBST) followed by incubation overnight at 4°C with primary antibody. Probed membranes were washed with PBST, then primary antibodies were detected using 1:10,000 peroxidase-coupled secondary antibodies (Jackson ImmunoResearch) and Clarity Western ECL Substrate on a ChemiDoc MP system. The following primary antibodies were used: 1:1000 ChAT (CTab) ( Dobransky et al., 2000 ), 1:10,000 β-actin (Sigma), 1:1000 ubiquitin (Santa Cruz), 1:1000 p53 (DO-1; Santa Cruz), and 1:500 vinculin (Santa Cruz). For detection of biotinylated proteins following SDS-PAGE, membranes were blocked overnight at 4°C in PBST with 5% BSA, then incubated for 1 h at 21°C with 1:20,000 Pierce High Sensitivity Streptavidin-HRP, washed with PBST, and imaged using Clarity ECL as above.

Yeast Strains and Media

Yeast strains BY 4741 and BY Δ pdr5 were obtained from the Saccharomyces Genome Deletion Project. Standard yeast media were used 1 . Yeast transformations were performed according to a standard PEG/lithium acetate protocol as described before ( Kawai et al., 2010 ). For induction of gene expression driven by the GAL1 promotor, 2% galactose was used instead of glucose as a carbon source for liquid media.

SPAAC Pulse-Chase in Yeast

Liquid cultures of yeast cells (Δ pdr5 ) transformed with human ChAT expression plasmid were grown in non-inducing selective media (SD-leu) at 30°C for 24 h. Cells were pelleted, washed twice with sterile H 2 O, and cells were resuspended in 2% galactose containing media lacking methionine. SPAAC pulse-chase was initiated by the addition of 50 μM AHA and incubation of cells for 24 h at 30°C. Control cells were incubated for 24 h in methionine-containing media. Cells were pelleted and AHA labeling was subsequently terminated by first washing cells with sterile H 2 O then incubating them at 30°C for 2, 4, or 8 h in methionine-containing non-inducing SD-leu media. To inhibit the proteasome or lysosome, cells were co-treated with either 50 μM MG132 or 10 μM Bafilomycin A, respectively, throughout both the 24 h AHA-pulse and 8 h chase periods. Cells were collected, washed twice on ice with sterile H 2 O, and lysed in cold lysis buffer (50 mM HEPES; pH 7.5, 150 mM NaCl, 5 mM EDTA, 1% Triton X-100) supplemented with SIGMAFAST Protease inhibitor (Sigma), and 50 mM NEM. Sterile glass beads were added, cells were vortexed, and lysates were centrifuged for 10 min at 21,000 g at 4°C.

Aliquots of cleared protein lysates were reacted with 10 μM TAMRA-DIBO for 4 h at 21°C followed by denaturation in 1× Laemmli sample buffer at 95°C for 10 min. For analysis, equal amounts of proteins from each experimental sample were resolved on SDS-PAGE gels and fluorescence was detected in-gel using a ChemiDoc MP system at an excitation/emission of 555/580 nm. To determine ChAT protein half-life in yeast, anti-ChAT IPs were prepared from cleared whole cell lysates of AHA-labeled cells as detailed above. Immunocaptured samples were washed and subsequently reacted with 10 μM TAMRA-DIBO in 1× PBS for 1 h at 21°C with gentle agitation. IP samples were washed then eluted into 50 μl of 2× Laemmli sample buffer at 85°C for 15 min with intermittent mixing. For analysis, equal volumes of immunoprecipitated AHA/TAMRA-labeled ChAT were resolved on SDS-PAGE gels and in-gel fluorescence was detected as described above. As controls, AHA/TAMRA-labeled proteins from either whole cell lysates or ChAT IPs were transferred to PVDF membranes and standard immunoblotting was completed. ChAT protein half-life in yeast was calculated as detailed above.

Analysis of Cell Growth and the Heat Shock Response in Yeast Cells Labeled With AHA

Cell growth was assessed by liquid culture as previously described ( Duennwald, 2013 ). Briefly, liquid cell cultures were diluted to OD 600 0.15 and incubated at 30°C. OD 600 was measured every 15 min using a Bioscreen C plate reader (Growth Curves USA) for 24 h. Growth curves were generated and the statistical significance was determined using a two-tailed student t -test and GraphPad Prism. To measure the effect of AHA labeling on induction of the heat shock response in yeast, a fluorescent reporter system whereby expression of heterologous GFP is driven by binding of heat shock factor 1 (Hsf1) to a synthetic promoter containing four adjacent heat shock elements (HSE) was used ( Brandman et al., 2012 ). The cells were grown in inducing selective media (SD-leu) at 24°C for 24 h and/or 42°C for 1 h. Cells were pelleted, washed twice with sterile H 2 O, and cells were resuspended in media containing of 50 μM AHA and incubating cells for another 4 h at 24°C and/or 42°C. Cells were pelleted and AHA labeling was subsequently terminated by first washing cells with sterile H 2 O then cells were collected for imaging using Cytation 5 Cell Imaging Multi-Mode Reader (BioTek), and/or lysed on ice in 0.1% Triton X-100 lysis buffer (50 mM Tris–HCl; pH 8.0, 150 mM NaCl, 0.1% Triton X-100) supplemented with protease/phosphatase inhibitors. Lysates were centrifuged for 15 min at 15,000 g at 4°C, and immunoblotting. Immunoblotting was completed as described above. The following primary antibodies were used: 1:1000 Hsp104, Hsp42, and Hsp26 (gifts from J. Buchner), 1:1000 histone H3 (LSBio), and 1:1000 HSP70 (Santa Cruz).

Statistical Analysis

Steady-state proteins levels from immunoblots were measured by densitometry of immunoreactive bands using ImageLab 5.0 software (Bio-Rad), normalized to either β-actin, PGK1, or histone H3 and graphed as mean ± SEM from individual independent replicate experiments (n). Statistical analysis for experiments was completed by one-way ANOVA with either Dunnett’s or Tukey’s post hoc test using GraphPad Prism software. Statistical significance was set at p ≤ 0.05.

Overview of a Novel SPAAC Pulse-Chase Method

We developed a novel pulse-chase method to determine the half-life of cellular proteins based on SPAAC click chemistry reactions using non-radioactive labeling and detection reagents ( Figure 1 ). Briefly, in this method newly synthesized proteins are first live-labeled (pulsed) with AHA, a biorthogonal methionine analog that contains a reactive azide moiety, in cultured cells under methionine-free conditions and then chased with excess methionine. Cells are collected at specified times in chase media, lysed, and a protein-of-interest is immunoprecipitated (IP) according to established laboratory protocols. AHA-labeled proteins are then reacted via 1,3-dipolar cycloaddition to form stable 1,2,3-triazole conjugates with a strained cyclooctyne, such as 4-dibenzocyclooctynol (DIBO) or dibenzocyclooctyne (DBCO), that are modified with either a fluorescent (e.g., tetramethylrhodamine (TAMRA) or Alexa Fluor 488) or a biotin probe ( Sanders et al., 2011 ; Kim et al., 2012 ; Dommerholt et al., 2016 ). AHA/cyclooctyne-labeled proteins are then resolved on SDS-PAGE gels and protein half-life can be determined either in-gel (fluorescent) or following transfer to a PVDF membrane (biotin). Lastly, for analysis standard immunoblotting is performed on AHA/cyclooctyne-labeled protein samples, thus eliminating the need to prepare both AHA-labeled and non-labeled samples and reducing potential sources of error.

Figure 1. Schematic overview of biorthogonal strain-promoted alkyne-azide cycloaddition (SPAAC) pulse-chase for fluorescent/chemiluminescent determination of protein half-life. (1) Briefly, cultured cells are live-labeled (pulsed) with L-azidohomoalanine (AHA), a bioorthogonal methionine analog that contains a reactive azide moiety, under methionine-free conditions. (2) AHA-containing media is removed and labeled cells are first washed then chased with media containing methionine for desired times (e.g., up to 24 h). (3) Cells are collected, lysed, and protein/s of interest are immunoprecipitated. (4) Immunopurified AHA-labeled proteins are reacted with a strained cyclooctyne (e.g., 4-dibenzocyclooctynol; DIBO) that is modified with either a fluorescent (e.g., tetramethylrhodamine; TAMRA) or biotin probe to form stable triazole conjugates. (5) AHA/cyclooctyne-labeled proteins are resolved on SDS-PAGE gels and either (6a) fluorescence is detected directly in-gel (e.g., labeled with TAMRA-DIBO) or, (6b) if labeled with a biotin-cyclooctyne probe, proteins are then transferred onto PVDF membranes and detected by chemiluminescence using a HRP-conjugated streptavidin. (7) Lastly, immunoblotting is completed to measure steady-state and immunoprecipitated protein levels from AHA/cyclooctyne-labeled protein samples, and (8) subsequently for downstream data analysis. Adapted with permission from Morey et al. (2016) .

Determination of ChAT Protein Half-Life by SPAAC Pulse-Chase

We used this new method initially to address a previously unresolved question related to the cellular protein half-life of choline acetyltransferase (ChAT), the enzyme that catalyzes synthesis of the neurotransmitter acetylcholine (ACh) ( Oda, 1999 ; Abreu-Villaca et al., 2011 ). ChAT mutations are linked to congenital myasthenic syndrome (CMS), a rare neuromuscular disorder ( Engel et al., 2015 ). The CMS-related ChAT mutation V18M reduces enzyme activity and cellular protein levels ( Shen et al., 2011 ) and is located within a highly conserved proline-rich motif at residues 14 P KL P V PP 20 that shares homology with SH3-binding motifs. Work from our laboratory found that disruption of this proline-rich motif reduces ChAT protein levels and cellular enzymatic activity of mutant P17A/P19A-ChAT and V18M-ChAT in mouse cholinergic SN56 cells ( Morey et al., 2016 ). This reduction in cellular protein levels of mutant ChAT appeared to be due to enhanced ubiquitination, and thus we aimed to determine if the half-life of mutant ChAT protein is also reduced ( Figure 2 ). Using this new SPAAC pulse-chase method in ChAT-expressing SN56 cells ( Morey et al., 2016 ), we initially tested the method by first detecting the progressive global loss of fluorescent AHA/TAMRA-labeled proteins from whole cell lysates during the 0–24 h chase period in the absence of changes in total protein levels as measured in parallel by anti-actin immunoblotting ( Figure 2A ). Additionally, following anti-ChAT IPs and reaction of immunocaptured ChAT protein with the strained cyclooctyne TAMRA-DIBO, we observed progressive loss of fluorescent AHA/TAMRA-labeled ChAT protein during the 0–24 h chase period. As anticipated, the decay of P17A/P19A-ChAT protein appeared more rapid than that of wild-type ChAT. By quantifying the fluorescence intensities of AHA/TAMRA-labeled ChAT, we determined that the protein half-life of mutant P17A/P19A-ChAT (2.2 h) is significantly reduced by ∼10-fold compared to wild-type ChAT [19.7 h; F (1,41) = 110.043, p ≤ 0.0001; Figure 2B ].

Figure 2. SPAAC pulse-chase reveals that ChAT protein half-life is reduced by mutation of an N-terminal proline-rich motif. (A) Fluorescence detection of immunoprecipitated (IP) AHA/TAMRA-labeled wild-type (WT) and P17A/P19A-ChAT from transiently transfected SN56 cells following SPAAC pulse-chase (4 h AHA pulse, 0–24 h methionine chase) with the strained cyclooctyne TAMRA-DIBO. Control unlabeled cells were incubated in media with methionine (i.e., without AHA). Fluorescent AHA/TAMRA-labeled proteins from either whole cell lysates or anti-ChAT IPs were detected in resolved SDS-PAGE gels at an Ex/Em of 555/580 nm. Anti-ChAT and anti-actin immunoblots were completed on AHA/TAMRA-labeled protein samples for downstream data analysis and as loading controls. (B) Protein half-life of P17A/P17A-ChAT (2.2 h) is reduced compared to wild-type ChAT (19.7 h). ChAT fluorescence intensities from (A) were plotted on a semi-logarithmic scale and linear regression was completed to determine ChAT protein half-life. The slopes of decay for wild-type and P17A/P19A-ChAT were significantly different: F (1,41) = 110.043, p ≤ 0.0001, n = 5. A 24 h time point was not included for the quantification of P17A/P19A-ChAT half-life as the fluorescence intensity was indistinguishable from background. (C) SN56 cells, transiently expressing either wild-type or P17A/P19A-ChAT, were treated with 100 μg/ml cycloheximide (CHX) for the time points indicated and anti-ChAT immunoblots were completed. Anti-actin immunoblots were completed as a loading control. (D) Protein half-life of P17A/P17A-ChAT (4.1 h) is reduced as compared to wild-type ChAT (44.9 h). Steady-state ChAT protein levels from (C) were plotted on a semi-logarithmic scale and linear regression completed to determine wild-type ChAT protein half-life. The slopes of decay for wild-type and P17A/P19A-ChAT were significantly different: F (1,46) = 36.802, p ≤ 0.0001, n = 5. Importantly, both methods (B,D) demonstrated that the protein half-life of P17A/P19A-ChAT is ∼10% that of wild-type ChAT. Adapted with permission from Morey et al. (2016) .

To compare the results generated using our new protocol to an established method, we performed a CHX assay on SN56 cells transiently expressing wild-type or P17A/P19A-ChAT. Unfortunately, we could only treat SN56 cells for 8 h due to the potent toxicity of CHX on these cells. Thus, by anti-ChAT immunoblotting of lysates from ChAT-expressing SN56 cells treated with 100 μg/ml CHX for 2–8 h ( Figure 2C ), we determined that the cellular half-life of P17A/P19A-ChAT (4 h) is again ∼10-fold shorter than that of wild-type ChAT [45 h; F (1,46) = 36.802, p ≤ 0.0001; Figure 2D ]. Importantly, both methods revealed that the relative protein half-life of P17A/P19A-ChAT is ∼10% that of wild-type ChAT.

Our previously published results showed that inhibition of proteasome function by MG132 treatment, but not lysosomal function by chloroquine treatment, resulted in increased steady-state protein levels of both wild-type and P17A/P19A-ChAT ( Morey et al., 2016 , 2017 ). Furthermore, we have shown that MG132 treatment resulted in stabilization of ubiquitinated wild-type and P17A/P19A-ChAT, suggesting that ChAT protein degradation is regulated through the proteasome. Therefore, we tested whether proteasome inhibition by MG132 treatment correlates with an increase in ChAT protein half-life ( Figure 3 ). Thus, we performed SPAAC pulse-chase on SN56 cells transiently expressing either wild-type ( Figure 3A ) or mutant P17A/P19A-ChAT ( Figure 3B ) that were treated with 5 μM MG132 throughout both the 4 h pulse and 24 h chase periods. ChAT protein was recovered from AHA-labeled SN56 cells by IP and subsequently fluorescence intensities of AHA/TAMRA-label ChAT were quantified. We observed that MG132 treatment prevented the decay of wild-type ChAT protein when compared to DMSO-treated control cells [half-life = 22.5 h; F (1,46) = 4.79241, p ≤ 0.05; Figure 3C ]. Additionally, the half-life of P17A/P19A-ChAT protein was also increased during MG132 treatment (16.8 h) when compared to DMSO-control [2.2 h; F (1,41) = 18.9864, p ≤ 0.0001; Figure 3D ]. Overall, these initial experiments provided novel insight into the proteolytic regulation of human ChAT protein and demonstrated that SPAAC pulse-chase is a valid and effective method to determine cellular protein half-life that is also compatible with inhibitors of protein degradation.

Figure 3. Proteasome inhibition increases ChAT protein half-life. Fluorescence detection of immunoprecipitated (IP) AHA/TAMRA-labeled wild-type (WT; A ) or P17A/P19A-ChAT (B) from transiently transfected SN56 cells following SPAAC pulse-chase (4 h AHA pulse, 0–24 h methionine chase) with the strained cyclooctyne TAMRA-DIBO. Cells were treated with either DMSO-control or 5 μM MG132 throughout both the 4 h AHA pulse and the 24 h chase periods (up to 28 h total). Control unlabeled cells were incubated in media with methionine (i.e., without AHA). Fluorescent AHA/TAMRA-labeled proteins from either whole cell lysates or anti-ChAT IPs were detected in resolved SDS-PAGE gels at an Ex/Em of 555/580 nm. Anti-ChAT and anti-actin immunoblots were completed on AHA/TAMRA-labeled protein samples for downstream data analysis and as loading controls. (C) Proteasome inhibition by MG132 treatment increased the protein half-life of wild-type ChAT (no protein decay) as compared to DMSO-treated control cells (22.5 h). Wild-type ChAT fluorescence intensities from (A) were plotted on a semi-logarithmic scale and linear regression was completed to determine wild-type ChAT protein half-life. The slopes of decay for DMSO- and MG132-treated cells were significantly different from each other: F (1,46) = 4.79241, p ≤ 0.05, n = 5. (D) Proteasome inhibition by MG132 treatment increased the protein half-life of mutant P17A/P19A-ChAT (16.6 h) as compared to DMSO-control (2.2 h). ChAT fluorescence intensities from (B) were plotted on a semi-logarithmic scale and linear regression was completed to determine P17A/P19A-ChAT protein half-life. The slopes of decay for DMSO- and MG132-treated cells were significantly different from each other: F (1,41) = 18.9864, p ≤ 0.0001, n = 5. A 24 h time point was not included for the quantification of P17A/P19A-ChAT half-life in DMSO-treated cells as the fluorescence intensity was indistinguishable from background.

Analysis of Cell Viability, Global Proteome Ubiquitination, Protein Solubility, and Heat Shock Response in Cells Labeled With AHA

Labeling of cells with AHA has been widely used in the past and shown to be non-toxic, does not inhibit protein synthesis, and does not alter global protein ubiquitination or degradation ( Kiick et al., 2002 ; Dieterich et al., 2006 , 2010 ; Baskin et al., 2007 ; Roche et al., 2009 ; McShane et al., 2016 ). To confirm these findings in the context of our experiments, we incubated SN56 cells for 4 h with 5 μM AHA in Met- media (pulsed), then performed a chase in Met+ media for 8 h. As a control to prevent labeling of newly synthesized proteins with AHA, we co-treated cells with 100 μg/ml CHX as shown in previous studies ( Dieterich et al., 2006 ; Supplementary Figure 1A ). Importantly, incubation of cells with AHA alone failed to either induce apoptosis or alter global ubiquitination, whereas treatment with CHX throughout both the 4 h AHA pulse and the 8 h chase periods significantly induced apoptosis ( Supplementary Figure 1B ; p ≤ 0.05) and led to a depletion in global protein ubiquitination ( Supplementary Figure 1C ; p ≤ 0.01) when compared to DMSO-treated control cells.

While used previously both in vitro and in vivo without compromising cell or animal viability ( Kiick et al., 2002 ; Dieterich et al., 2006 , 2010 ; Baskin et al., 2007 ; Roche et al., 2009 ; Hinz et al., 2012 ; Calve et al., 2016 ; McShane et al., 2016 ), incorporation of AHA into nascent proteins may induce changes to protein folding, leading to induction of the heat shock response (HSR). This response can be assessed by immunoblot for elevated levels in the heat shock proteins HSP70, HSP90, and HSC70. When assayed in SN56 cells, we observed increased steady state levels of HSP70 ( Supplementary Figure 1A ) and HSP90 and HSC70 proteins ( Supplementary Figures 1D,E ; p ≤ 0.05) in AHA-labeled cells after 8 h of chase when compared to untreated cells. As a positive control, treatment of cells with 10 mM azetidine-2-caroxylic acid (AZC), a proline analog that induces protein misfolding ( Weids et al., 2016 ), for 24 h also promoted the synthesis of these HSPs ( Lee and Seo, 2002 ). Of note, HSP levels were significantly greater in AZC-treated cells compared to AHA-treated cells ( Supplementary Figures 1D–F , respectively; p ≤ 0.001).

Terminally misfolded proteins are often targeted for degradation through enhanced ubiquitination and proteasomal and/or lysosomal degradation ( Amm et al., 2013 ). Thus, to determine whether the observed elevation in HSPs correlated with AHA-induced misfolding and subsequent insolublization of proteins, we fractionated whole cell lysates from AHA-labeled SN56 cells into Triton-soluble and -insoluble fractions. Importantly, following anti-ubiquitin immunoblotting ( Supplementary Figure 2A ), we did not observe detectable changes in the abundance of total ubiquitinated proteins in either the Triton-soluble ( Supplementary Figure 2B ) or Triton–insoluble fraction ( Supplementary Figure 2C ) from AHA-labeled cells when compared to unlabeled cells. As a positive control, treatment of cells with 10 mM AZC for 24 h led to the accumulation of total ubiquitinated proteins in both the Triton-soluble ( Supplementary Figure 2B ; p ≤ 0.001) and Triton–insoluble fraction ( Supplementary Figure 2C ; p ≤ 0.001) when compared to control and AHA-treated cells.

We also measured the effect of AHA labeling on cellular toxicity and induction of the HSR in HEK293 and HeLa cells, two commonly used human cells lines. Similar to SN56 cells, we observed by immunoblotting ( Supplementary Figure 3A ) that incubation with 50 μM AHA failed to induce apoptosis or lead to changes in global protein ubiquitination in either HEK293 ( Supplementary Figures 3B,D ) or HeLa cells ( Supplementary Figures 3C,E ). Interestingly, in HEK293 cells AHA labeling had no effect on the steady-state levels of HSP90, HSP70, or HSC70 ( Supplementary Figures 3F,H,J , respectively), whereas in HeLa cells we observed an increase in HSP90 and HSP70 protein in AHA-labeled cells as compared to untreated cells ( Supplementary Figures 3G,I ; p ≤ 0.05). HSC70 protein levels were unchanged in HeLa cells following AHA labeling ( Supplementary Figure 3K ).

Taken together, these data suggest that AHA does not negatively affect cell viability or cellular protein degradation in the three mammalian cell lines tested here. While we do report that AHA treatment can induce the HSR, this was not universally observed and, importantly, does not correlate with either gross protein insolublization or cellular toxicity.

Comparison of Commercially Available Strained Cyclooctyne Probes

An advantage we propose for SPAAC pulse-chase is that various strained cyclooctyne reagents with unique probes can be conjugated to AHA-labeled proteins, thus increasing the versatility of this method. In addition to the TAMRA-DIBO label used in the experiments described above, we also prepared whole cell lysates from AHA-labeled HEK293 cells and compared conjugation of two other commercially available strained cyclooctynes, 488-DBCO and Biotin-DIBO, to that of TAMRA-DIBO. To test the sensitivity of these three strained cyclooctynes, we made serial dilutions of protein samples containing AHA/cyclooctyne-labeled proteins and ran them on SDS-PAGE gels. Signal intensities were then measured either in-gel when using fluorescent TAMRA-DIBO ( Figure 4A ) or 488-DBCO ( Figure 4B ), or the sensitivity of Biotin-DIBO ( Figure 4C ) was determined using HRP-conjugated streptavidin following transfer to PVDF membrane. Overall, we successfully conjugated these three different strained cyclooctyne probes to AHA-labeled proteins and observed little (Biotin-DIBO) or no (TAMRA-DIBO and 488-DBCO) background signal in protein samples from unlabeled cells. Furthermore, after lysates were serially diluted, we observed a similar signal sensitivity when using either TAMRA-DIBO, 488-DBCO, or Biotin-DIBO, and obtained robust fluorescent or chemiluminescent signals with as little as 0.39 μg of total protein (1:64 dilution from an initial 25 μg of protein).

Figure 4. Detection sensitivity comparison for three different strained cyclooctyne probes. HEK293 cells were live-labeled in culture for 4 h with 50 μM AHA then collected immediately without a methionine chase. Control unlabeled cells were incubated in media with methionine (i.e., without AHA). Cells were lysed and AHA-labeled protein samples from whole cell lysates were reacted with the strained cyclooctynes (A) TAMRA-DIBO, (B) 488-DBCO, or (C) Biotin-DIBO. AHA/cyclooctyne-labeled proteins were denatured in 1× Laemmli sample buffer, then serially diluted 1:1 with 1× Laemmli sample buffer a total of six times until reaching a final dilution of 1:64. Protein samples were run on SDS-PAGE gels and AHA-labeled proteins were detected in-gel at an Ex/Em of either 555/580 nm ( A ; TAMRA-DIBO) or 494/517 nm ( B ; 488-DBCO). Alternatively, AHA-labeled proteins reacted with Biotin-DIBO (C) were resolved on SDS-PAGE gels, then transferred to PVDF membranes and detected by chemiluminescence using HRP-conjugated streptavidin. Anti-actin immunoblots were completed as a loading control. Overall, similar detection sensitivity was observed between AHA-labeled protein samples reacted with TAMRA-DIBO, 488-DBCO, or Biotin-DIBO ( n = 4).

Determination of Endogenous p53 Protein Half-Life by SPAAC Pulse-Chase

We next tested whether this method could be used to determine the half-life of an endogenous protein-of-interest. To this end, we performed SPAAC pulse-chase analysis of the tumor suppressor protein p53, a protein that is essential for maintaining genomic stability and that is commonly mutated in many human cancers ( Mantovani et al., 2019 ). Following AHA-labeling of HEK293 cells and a 0–12 h chase period, AHA-labeled proteins were reacted with the strained cyclooctynes TAMRA-DIBO ( Figure 5A ), 488-DBCO ( Figure 5B ), or Biotin-DIBO ( Figure 5C ), and protein samples were resolved by SDS-PAGE. As anticipated, we detected the global progressive loss of fluorescent (TAMRA-DIBO or 488-DBCO) and biotin-DIBO-labeled proteins from whole cell lysates during the 0–12 h chase period in the absence of changes in total protein levels as measured in parallel by anti-actin immunoblotting. Additionally, following IP of endogenous p53 we were able to observe the progressive loss of AHA/cyclooctyne-labeled p53 protein during the 0–12 h chase period using all three of the strained cyclooctyne probes tested. The half-life of AHA-labeled endogenous p53 protein was determined to be 10.3 h ( Figure 5D ; TAMRA-DIBO), 12.7 h ( Figure 5E ; 488-DBCO), or 11.0 h ( Figure 5F ; Biotin-DIBO), respectively, thus demonstrating that these different cyclooctyne probes produce similar results that are comparable to previously published data ( Lukashchuk and Vousden, 2007 ; Dai et al., 2013 ). Together, these data suggest that not only can SPAAC pulse-chase be used to determine the half-life of endogenous proteins, but also that various commercially available strained cyclooctyne probes can be used interchangeably without dramatically altering the resulting protein half-lives.

Figure 5. Half-life determination of p53 by SPAAC pulse-chase. Detection of immunoprecipitated (IP) endogenous AHA-labeled p53 from HEK293 cells following SPAAC pulse-chase (4 h AHA pulse, 0–12 h methionine chase) with the strained cyclooctynes TAMRA-DIBO (A) , 488-DBCO (B) , or Biotin-DIBO (C) . Control unlabeled cells were incubated in media with methionine (i.e., without AHA). AHA/cyclooctyne-labeled proteins from either whole cell lysates or anti-p53 IPs were resolved on SDS-PAGE gels and AHA-labeled proteins were detected in-gel at an Ex/Em of either 555/580 nm ( A ; TAMRA-DIBO) or 494/517 nm ( B ; 488-DBCO). Alternatively, AHA-labeled proteins reacted with Biotin-DIBO (C) were resolved on SDS-PAGE gels, then transferred to PVDF membranes and detected by chemiluminescence using a HRP-conjugated streptavidin. Anti-p53 and anti-vinculin immunoblots were completed on AHA/cyclooctyne-labeled protein samples for downstream data analysis and as loading controls. The protein half-life of AHA-labeled p53 when reacted with three different strained cyclooctynes was determined to be 10.3 h ( D ; TAMRA-DIBO), 12.7 h ( E ; 488-DBCO), or 11.0 h ( F ; Biotin-DIBO). Fluorescent or chemiluminescent intensities from immunoprecipitated p53 were plotted on a linear scale and linear regression analysis was completed to determine p53 protein half-life ( n = 5).

We next sought to demonstrate the broader applicability of our method by expanding it beyond mammalian cells into the model organism yeast ( Saccharomyces cerevisiae ). One advantage to using yeast is that, unlike constitutive promoters used typically in many mammalian systems, a multitude of selectively inducible promoters exists for transgene expression ( Weinhandl et al., 2014 ). Thus, using human ChAT as a protein of interest for SPAAC pulse-chase in yeast, we generated expression vectors for either wild-type 69-kDa human ChAT or a yellow fluorescent protein (YFP)-tagged ChAT protein under the control of a galactose-inducible promoter. As yeast do not naturally express a ChAT ortholog, we first assessed whether the expression of human ChAT has an effect on the growth of yeast cultures under normal conditions. Spotting assays on agar plates ( Supplementary Figure 4A ) and growth curves in liquid media ( Supplementary Figure 4B ) showed no growth defect associated with the heterologous expression of human ChAT in yeast. Furthermore, in agreement with work in mammalian cells ( Resendes et al., 1999 ), YFP-tagged ChAT is diffusely localized throughout the yeast cytosol ( Supplementary Figure 4C ). Lastly, by anti-ChAT immunoblotting we observed that ChAT is stably expressed in transformed yeast after 10 h of galactose induction ( Supplementary Figure 4D ).

We next applied the SPAAC pulse-chase method to ChAT-expressing yeast cells treated with the proteasome inhibitor MG132 (50 μM) or Bafilomycin A (10 μM), an autophagy inhibitor ( Yoshimori et al., 1991 ), throughout both the 24 h AHA-pulse and 8 h chase periods. Similar to mammalian cells, we detected the global progressive loss of fluorescent AHA/TAMRA-labeled proteins from whole cell lysates during the 0–8 h chase period in the absence of changes in total protein levels as measured in parallel by anti-PGK1 immunoblotting as a loading control ( Figures 6A,B ). Co-treatment of cells with either MG132 or Bafilomycin A reduced the loss of AHA-labeled proteins from whole cell lysates. Importantly, following anti-ChAT IPs we observed progressive loss of fluorescent AHA/TAMRA-labeled ChAT protein during the 0–8 h chase period ( Figure 6B ). By quantifying the fluorescence intensities of AHA/TAMRA-labeled ChAT ( Figure 6C ) we determined that the protein half-life of human ChAT is 3.1 h in yeast. Furthermore, we observed that proteasome inhibition by MG132 co-treatment reduced the decay of ChAT protein as compared to DMSO-treated control cells [half-life = 18.7 h; F (1,46) = 5.6432, p ≤ 0.001; Figure 6C ], whereas Bafilomycin A co-treatment had no significant effect on the half-life of AHA/TAMRA-labeled ChAT (2.2 h). It is important to note that live-labeling of yeast cells with AHA did not result in any growth defect regardless of expression of human ChAT ( Supplementary Figure 4B ).

Figure 6. SPAAC pulse-chase analysis of ChAT protein half-life in yeast. (A) Fluorescence detection of total AHA-labeled proteins from BY Δ pdr5 yeast cells following SPAAC pulse-chase (24 h AHA pulse, 0–8 h methionine chase) and labeling with the strained cyclooctyne TAMRA-DIBO. Cells were co-treated with either 50 μM MG132 or 10 μM Bafilomycin A throughout both the 24 h AHA-pulse and 8 h chase periods to inhibit the proteasome or lysosome, respectively. Control unlabeled cells were incubated in media with methionine (i.e., without AHA). (B) Fluorescence detection of immunoprecipitated (IP) AHA/TAMRA-labeled human ChAT protein from ChAT-expressing BY Δ pdr5 yeast cells following SPAAC pulse-chase with the strained cyclooctyne TAMRA-DIBO. Fluorescent AHA/TAMRA-labeled proteins from either whole cell lysates or anti-ChAT IPs were detected in resolved SDS-PAGE gels at an Ex/Em of 555/580 nm. Anti-ChAT and anti-PGK1 immunoblots were completed on AHA/TAMRA-labeled protein samples for downstream data analysis and as loading controls. (C) Proteasome inhibition by MG132 treatment increased the protein half-life of human ChAT (18.7 h) as compared to DMSO-treated control yeast cells [3.1 h; F (1,46) = 5.6432, p ≤ 0.001]. Bafilomycin A co-treatment had no significant effect on the half-life of AHA/TAMRA-labeled ChAT (2.2 h). ChAT fluorescence intensities from (B) were plotted on a semi-logarithmic scale and linear regression was completed to determine human ChAT protein half-life ( n = 5).

Lastly, we analyzed the effect of AHA on the HSR in yeast. To accomplish this, we first utilized a fluorescent reporter system ( Supplementary Figure 5A ) whereby expression of heterologous GFP is driven by binding of heat shock factor 1 (Hsf1) to a synthetic promoter containing four adjacent heat shock elements (HSE) ( Sorger et al., 1987 ) in cells following exposure to hyperthermia or protein misfolding stress ( Brandman et al., 2012 ). Following treatment of yeast cell with AHA for 4 h at 24°C, we observed an increase in GFP-positive cells ( Supplementary Figure 5B ; p ≤ 0.001). As a positive control, we also observed an increase in GFP-positive cells following exposure to hyperthermic stress at 42°C for 1 h ( p ≤ 0.001), and that exposure of AHA-treated cells to 42°C led to a further two-fold increase in GFP positive cells as compared to hyperthermia alone ( p ≤ 0.001). Additionally, by immunoblotting ( Supplementary Figure 5C ) we show that treatment of yeast cells with AHA for 4 h at 24°C increased the steady-state protein levels of endogenous Hsp70, Hsp42, and Hsp104 ( p ≤ 0.001), while no changes were observed for HSP26 ( Supplementary Figure 5D ). Lastly, exposure of yeast cells to 42°C for 1 h led to an increase in these HSPs ( p ≤ 0.001) as compared to AHA-treated cells grown at 24°C. Taken together, these results demonstrate that our SPAAC pulse-chase method can be used successfully in yeast and, while AHA treatment did induce the heat shock response in yeast similar to some mammalian cells, this remained non-toxic and did not affect yeast viability.

Multiple methods are available to cell biologist to study the rate at which cellular proteins are degraded, but each of these approaches have severe limitations and experimental problems ( Table 1 ). Here, we establish a non-toxic and non-radioactive pulse-chase method for the determination of cellular protein half-life that utilizes SPAAC click chemistry reactions. We provide proof-of-principal examples for this method in multiple mammalian cells lines and in yeast using both heterologously expressed (wild-type and mutant human ChAT) and endogenous (tumor suppressor p53) proteins. Furthermore, by applying different commercially available fluorescent and biotin cyclooctyne probes, we demonstrate the versatility and flexibility of this novel method.

Table 1. Comparison of SPAAC pulse-chase to existing protein half-life methods.

The bioorthogonal amino acid AHA is a non-toxic methionine analog that can incorporate into newly synthesized proteins without altering global rates of protein degradation or ubiquitination ( Kiick et al., 2002 ; Dieterich et al., 2006 ). AHA has been used widely in both in vitro and in vivo studies to measure global changes in protein degradation ( Kiick et al., 2002 ; Dieterich et al., 2006 , 2010 ; Baskin et al., 2007 ; Roche et al., 2009 ; McShane et al., 2016 ) and is available commercially or can be synthesized in-house ( Link et al., 2007 ; Roth et al., 2010 ). While considered a non-obtrusive replacement for methionine in pulse-chase studies ( Ma and Yates, 2018 ; Steward et al., 2020 ), one potential caveat of using AHA is that its incorporation into nascent proteins may alter protein folding, thus leading to induction of the heat shock and/or misfolded protein response. Our results indicate that AHA labeling can induce the HSR in both mammalian and yeast cells. It is important to note, however, that in mammalian cells this induction was significantly less than that observed following treatment with AZC, a proline analog known to induce protein misfolding ( Fowden and Richmond, 1967 ; Trotter et al., 2001 ), or in yeast cells exposed to hyperthermic conditions. Additionally, AHA incorporation did not induce the HSR in human HEK293 cells, suggesting that this effect of AHA is not universal. Importantly, in support of the growing literature on the cellular safety of AHA, we provide further evidence that, in mammalian cells, AHA does not induce apoptosis nor lead to changes in global ubiquitination, does not promote accumulation of insoluble ubiquitinated proteins, and does not affect yeast cell growth. Furthermore, separate studies have used AHA to label proteins in live animals without negatively affecting animal behavior, growth and development, and physiology ( Hinz et al., 2012 ; Calve et al., 2016 ). Lastly, Lehner et al. (2017) reported that in in vitro studies AHA labeling of recombinant PDZ3 domain proteins results in only minor alterations to protein secondary structure while not affecting ligand binding and yielded a soluble, well-folded, and functional model protein. Thus, while AHA incorporation could introduce changes to protein folding, these effects appear to be minor and/or negligible, and our work together with the aforementioned studies suggests that AHA is a non-toxic methionine analog that is minimally invasive to cell physiology and is suitable for use in pulse-chase studies.

One limitation of SPAAC pulse-chase is that labeling of nascent proteins with AHA is directly proportional to the number of methionine residues in a given protein. Thus, if a protein-of-interest contains a small number of methionine molecules, or only contains the N-terminal methionine that is often excised during post-translational processing ( Giglione et al., 2004 ), detection of nascent proteins with AHA may be difficult. Fortunately, additional azide-containing bioorthogonal amino acids are available for labeling either phenylalanine (4-azido-L-phenylalanine) or tyrosine (4-propargyloxy-L-phenylalanine) residues, though use of these require specialized engineered cells expressing tRNAs that can accept these bioorthogonal amino acids ( Saleh et al., 2019 ). An alternative strategy to AHA involves the labeling of newly synthesized proteins in the presence of methionine with O-propargyl-puromycin (OPP), an alkyne-containing puromycin analog that forms covalent linkages with the C-terminus of nascent polypeptides ( Liu et al., 2012 ; Forester et al., 2018 ; Hidalgo San Jose and Signer, 2019 ). It is important to note though that incorporation of OPP into nascent proteins results in premature translation termination, release of C-terminally truncated peptides from the ribosome, and an overall inhibition of protein synthesis similar to that observed with CHX ( Liu et al., 2012 ; Forester et al., 2018 ).

One of the unique advantages to SPAAC pulse-chase is the ever-growing selection of commercially available reagents for SPAAC reactions, including various bioorthogonal amino acids as discussed above and complementary fluorescent or chemiluminescent probes. This offers a large degree of flexibility in experimental design and increased compatibility with commonly used SDS-PAGE and immunoblotting equipment. To demonstrate the flexibility of SPAAC pulse-chase, we used three different cyclooctyne probes, including fluorescent TAMRA-DIBO or 488-DBCO, or a biotinylated DIBO compound detected by chemiluminescence using streptavidin-HRP. Importantly, we observed little difference in the sensitivities of these probes and obtained similar results when measuring the protein half-life of endogenous p53, suggesting that these and potentially other cyclooctyne probes may be interchangeable. Taken together, by using click chemistry reagents this SPAAC pulse-chase method is modular and adaptable to a variety of experimental needs, limitations, and available resources.

In conclusion, we present a novel, non-toxic, and non-radioactive method as an alternative to classical 35 S-methionine live-labeling or CHX experiments for the determination of protein half-life. Importantly, this method utilizes bioorthogonal click chemistry reactions in a manner that is both compatible with different eukaryotic systems and that allows for end-user customization. We believe that this protocol will be of interest to and applicable to many researchers in the fields of molecular and cellular biology. Lastly, this method is an important example of the potential for click chemistry to improve existing methods and of the growing utility of click chemistry in studying biological systems.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Author Contributions

TMM, MAE, MLD, and RJR designed the experiments. TMM completed experimental work detailing SPAAC pulse-chase in mammalian cells, while MAE performed experiments in yeast. All authors assisted with data analysis and contributed to writing this manuscript through editing and revisions.

This work was supported by grants from the Canadian Institutes of Health Research to RJR (FRN-115135) and to MLD (MOP-136930, PJT-159781, and PJT-162431). TMM was a recipient of an Ontario Graduate Scholarship during this study.

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s Note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Acknowledgments

We thank Daniela Rotin (Cell Biology, Hospital for Sick Children) for providing laboratory equipment, reagents, and personnel (TMM) that contributed to the completion of some experiments included in this report.

Supplementary Material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fcell.2021.722560/full#supplementary-material

Supplementary Figure 1 | Analysis of cell viability, global proteome ubiquitination, and heat shock response in mouse cholinergic SN56 cells labeled with AHA. (A) SN56 cells were live-labeled in culture for 4 h with 50 μM AHA and either collected immediately (0 h) or following 8 h of chase in methionine-containing media. Control unlabeled cells were incubated in media with methionine (i.e., without AHA). Cells were lysed and AHA-labeled protein samples from whole cell lysates were reacted with the strained cyclooctyne TAMRA-DIBO. Protein samples were run on SDS-PAGE gels, AHA-labeled proteins were detected in-gel at an Ex/Em 555/580 nm, samples were transferred to PVDF membranes, and immunoblotting was completed as indicated. In support of previous studies ( Dieterich et al., 2006 ), incorporation of AHA into newly synthesized proteins was prevented when cells were co-treated with CHX, a potent inhibitor of protein synthesis. Additionally, incubation of cells with AHA alone failed to induce apoptosis (B) , nor did this lead to changes in total protein ubiquitination (C) as compared to unlabeled cells. Conversely, treatment of cells with 100 μg/ml of cycloheximide (CHX) throughout both the 4 h AHA pulse and the 8 h chase periods (up-to 12 h total) both induced apoptosis ( B ; ∗ p ≤ 0.05) and led to a depletion of total ubiquitinated proteins ( C ; ∗∗ p ≤ 0.01) as compared to control untreated cells. As positive controls, treatment of unlabeled cells with 200 nM staurosporine, a pan-kinase inhibitor ( Karaman et al., 2008 ), for 8 h induced apoptosis ( B ; ∗∗∗ p ≤ 0.01) while treatment with 10 mM azetidine-2-caroxylic acid (AZC), a proline analog that can induce protein misfolding ( Weids et al., 2016 ), for 24 h led to the enhanced accumulation of total ubiquitinated proteins ( C ; ∗∗∗ p ≤ 0.001) as compared to AHA-labeled cells. Induction of cellular apoptosis was measured by immunoblotting for the formation of a lower molecular mass cleaved form of poly (ADP-ribose) polymerase (PARP) ( Mullen, 2004 ), while levels of total ubiquitinated proteins was measured by anti-ubiquitin immunoblotting. To measure induction of the heat shock response following cell labeling with AHA, immunoblotting for the heat shock proteins HSP70, HSP90, and HSC70 was completed. Interestingly, enhanced synthesis of HSP70 (A) , HSP90 ( D ; ∗ p ≤ 0.05), and HSC70 ( E ; p = 0.09) was observed in cells incubated with AHA and chased for 8 h as compared to control untreated cells; this was prevented when AHA-labeled cells were co-treated with CHX. Furthermore, as a positive control treatment of unlabeled cells with 10 mM AZC for 24 h promoted the synthesis of HSP70 (A) , HSP90 ( D ; ∗∗∗ p ≤ 0.01), and HSC70 ( E ; ∗∗∗ p ≤ 0.01). The induction of HSP70 expression was greater in cells treated with AZC as compared to cells labeled with AHA ( F ; ∗∗∗ p ≤ 0.01). Statistical analysis was completed by one-way ANOVA with Dunnett’s post hoc test (mean ± SEM, n = 5).

Supplementary Figure 2 | Incubation of cells with AHA does not affect overall solubility of cellular proteins. (A) SN56 cells were live-labeled in culture for 4 h with 50 μM AHA and either collected immediately (0 h) or following 8 h of chase in methionine-containing media. Control unlabeled cells were incubated in media with methionine (i.e., without AHA). Cells were lysed in buffer containing 0.1% Triton X-100 and fractionated into soluble and insoluble proteins. Triton X-100 insoluble proteins were solubilized prior to SDS-PAGE by denaturing in 2× Laemmli sample buffer with 5% 2-mercaptoethanol. Following transfer to PVDF membranes, anti-ubiquitin immunoblotting was completed to detect levels of ubiquitinated cellular proteins in Triton-soluble and Triton-insoluble fractions. While incubation of cells with AHA led to the synthesis of the heat shock protein HSP70 as observed in Supplementary Figure 1 , this was only observable in the Triton-soluble fraction. Furthermore, there was no detectable change in the abundance of total ubiquitinated proteins following incubation with AHA in either the Triton-soluble (B) or Triton–insoluble fraction (C) as compared to unlabeled cells. As a positive control to induce protein misfolding, treatment of unlabeled cells with 10 mM AZC for 24 h promoted the synthesis of HSP70 (observable in both fractions) and led to the enhanced accumulation of total ubiquitinated proteins in both the Triton-soluble ( B ; ∗∗∗ p ≤ 0.001) and Triton–insoluble fraction ( C ; ∗∗∗ p ≤ 0.001) as compared to AHA-treated cells. Statistical analysis was completed by one-way ANOVA with Tukey’s post hoc test (mean ± SEM, n = 3).

Supplementary Figure 3 | Analysis of cell viability, global proteome ubiquitination, and heat shock response in human HEK293 and HeLa cells labeled with AHA. (A) HEK293 or HeLa cells were live-labeled in culture for 4 h with 50 μM AHA and either collected immediately (0 h) or following 8 h of chase in methionine-containing media. Control unlabeled cells were incubated in media with methionine (i.e., without AHA). As a positive control, unlabeled cells were treated with either 10 mM AZC for 24 h to induce protein misfolding. Cells were lysed and AHA-labeled protein samples from whole cell lysates were reacted with the strained cyclooctyne TAMRA-DIBO. Protein samples were run on SDS-PAGE gels, AHA-labeled proteins were detected in-gel at an Ex/Em 555/580 nm, samples were transferred to PVDF membranes, and immunoblotting was completed as indicated. Similar to SN56 cells ( Supplementary Figure 1 ), incubation with AHA failed to induce apoptosis ( B – HEK293; C – HeLa) or lead to changes in total protein ubiquitination in either HEK293 (D) or HeLa cells (E) as compared to unlabeled cells. Conversely, treatment of cells with AZC induced apoptosis ( B – HEK293; C – HeLa; ∗∗∗ p ≤ 0.001) and enhanced the accumulation of total ubiquitinated proteins in either HEK293 ( D ; ∗∗∗ p ≤ 0.001) or HeLa cells ( E ; ∗∗∗ p ≤ 0.001) as compared to AHA-labeled cells. Induction of cellular apoptosis was measured by immunoblotting for the formation of a lower molecular mass cleaved form of PARP, while levels of total ubiquitinated proteins was measured by anti-ubiquitin immunoblotting. To measure induction of the heat shock response following cell labeling with AHA, immunoblotting for the heat shock proteins HSP90, HSP70, and HSC70 was completed. In HEK293 cells, AHA labeling had no effect on the steady-state levels of HSP90 (F) , HSP70 (H) , or HSC70 (J) , while treatment with AZC did lead to increased protein levels of HSP90 ( F ; ∗∗ p ≤ 0.01) as compared to unlabeled and AHA-treated cells. In HeLa cells, labeling of cells with AHA and chasing them for 8 h led to increased steady-state protein levels of HSP90 ( G ; ∗ p ≤ 0.05) and HSP70 ( I ; ∗∗∗ p ≤ 0.001) as compared to control unlabeled cells; HSC70 protein levels were unchanged in HeLa cells following AHA labeling (K) . Lastly, treatment of HeLa cells with AZC increased the protein levels of HSP90 ( G ; ∗∗∗ p ≤ 0.001), HSP70 ( I ; ∗∗∗ p ≤ 0.001), and HSC70 ( K ; p = 0.06) as compared to control unlabeled cells. Statistical analysis was completed by one-way ANOVA with Tukey’s post hoc test (mean ± SEM, n = 4).

Supplementary Figure 4 | AHA labeling in yeast cells is non-toxic. (A) Growth assays on plates of yeast cells expressing ChAT and controls. Liquid cultures of yeast were spotting on selective inducing agar media and incubated at 30°C for 2–4 days before imaging. (B) Growth curves of Δpdr5 yeast cells expressing ChAT and controls grown in the presence or absence of 50 μM L-azidohomoalanine (AHA). (C) Fluorescence microcopy of yeast expressing ChAT-YFP. Cells were imaged using a Leica TCS SP5 II confocal microscope at 63× magnification. The scale bar represents 5 μm. (D) Immunoblots of protein lysates of yeast cells expressing ChAT and controls grown in the presence of AHA probed with anti-ChAT or anti-histone H3 antibodies.

Supplementary Figure 5 | AHA labeling induces a mild heat shock response in yeast. (A) Fluorescence microscopy of yeast cells expressing HSE-eGFP grown for 4 h in the presence or absence of AHA at 24 and 42°C and quantification ( B ; ∗ p ≤ 0.05, ∗∗ p ≤ 0.01; n = 5). (C) Anti-HSP immunoblots prepared with protein lysates from yeast cells grown in the presence or absence of AHA at 24 or 42°C and quantification (D) . Anti-histone H3 immunoblotting was completed as a loading control.

Abbreviations

ACh, acetylcholine; AHA, L-azidohomoalanine; AZC, azetidine-2-caroxylic acid; ChAT, choline acetyltransferase; CHX, cycloheximide; CMS, congenital myasthenic syndrome; CTab, anti-ChAT carboxyl-terminal peptide antibody; DBCO, dibenzocyclooctyne; DIBO, 4-dibenzocyclooctynol; GFP, green fluorescent protein; HSC70, heat shock cognate 71 kDa protein; HSE, heat shock element; HSF1, heat shock factor 1; HSP, heat shock protein; HSR, heat shock response; IP, immunoprecipitation; NEM, N-ethylmaleimide; OPP, O-propargyl -puromycin; p53, cellular tumor antigen p53; PARP, poly (ADP-ribose) polymerase; PGK1, phosphoglycerate kinase 1; PVDF, polyvinylidene fluoride; RIPA, radioimmune precipitation buffer; SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis; SPAAC, strain-promoted alkyne-azide cycloaddition; TAMRA, tetramethylrhodamine; WT, wild-type; YFP, yellow fluorescent protein.

• ^ http://cshprotocols.cshlp.org/site/recipes/nav_y.dtl

Abreu-Villaca, Y., Filgueiras, C. C., and Manhaes, A. C. (2011). Developmental aspects of the cholinergic system. Behav. Brain Res. 221, 367–378.

Google Scholar

Amm, I., Sommer, T., and Wolf, D. H. (2013). Protein quality control and elimination of protein waste: the role of the ubiquitin-proteasome system. Biochem. Biophys. Acta 1843, 182–196. doi: 10.1016/j.bbamcr.2013.06.031

PubMed Abstract | CrossRef Full Text | Google Scholar

Ansari, A. M., Ahmed, A. K., Matsangos, A. E., Lay, F., Born, L. J., Marti, G., et al. (2016). Cellular GFP toxicity and immunogenicity: potential confounders in in vivo cell tracking experiments. Stem Cell Rev. Rep. 12, 553–559. doi: 10.1007/s12015-016-9670-8

Baens, M., Noels, H., Broeckx, V., Hagens, S., Fevery, S., Billiau, A. D., et al. (2006). The dark side of EGFP: defective polyubiquitination. PLoS One. 1:e54. doi: 10.1371/journal.pone.0000054

Baranello, R. J., Bharani, K. L., Padmaraju, V., Chopra, N., Lahiri, D. K., Greig, N. H., et al. (2015). Amyloid-beta protein clearance and degradation (ABCD) pathways and their role in Alzheimer’s disease. Curr. Alzheimer Res. 12, 32–46. doi: 10.2174/1567205012666141218140953

Baskin, J. M., Prescher, J. A., Laughlin, S. T., Agard, N. J., Chang, P. V., Miller, I. A., et al. (2007). Copper-free click chemistry for dynamic in vivo imaging. Proc. Natl. Acad. Sci. U.S.A. 104, 16793–16797. doi: 10.1073/pnas.0707090104

Blusztajn, J. K., Venturini, A., Jackson, D. A., Lee, H. J., and Wainer, B. H. (1992). Acetylcholine synthesis and release is enhanced by dibutyryl cyclic AMP in a neuronal cell line derived from mouse septum. J. Neurosci. 12, 793–799. doi: 10.1523/jneurosci.12-03-00793.1992

Brandman, O., Stewart-Ornstein, J., Wong, D., Larson, A., Williams, C. C., Li, G. W., et al. (2012). A ribosome-bound quality control complex triggers degradation of nascent peptides and signals translation stress. Cell 151, 1042–1054. doi: 10.1016/j.cell.2012.10.044

Calve, S., Witten, A. J., Ocken, A. R., and Kinzer-Ursem, T. L. (2016). Incorporating of non-canonical amino acids into the developing murine proteome. Sci. Rep. 6:32377.

Ciechanover, A., and Kwon, Y. T. (2015). Degradation of misfolded proteins in neurodegenerative diseases” therapeutic targets and strategies. Exp. Mol. Med. 47:e147. doi: 10.1038/emm.2014.117

Coumans, J. V., Gau, D., Poljak, A., Wasinger, V., Roy, P., and Moens, P. (2014). Green fluorescent protein expression triggers proteome changes in breast cancer cells. Exp. Cell Res. 320, 33–45. doi: 10.1016/j.yexcr.2013.07.019

Dai, C. L., Shi, J., Chen, Y., Iqbal, K., Liu, F., and Gong, C. X. (2013). Inhibition of protein synthesis alters protein degradation through activation of protein kinase B (AKT). J. Biol. Chem. 288, 23875–23883. doi: 10.1074/jbc.m112.445148

Dieterich, D. C., Hodas, J. J., Gouzer, G., Shadrin, I. Y., Ngo, J. T., Triller, A., et al. (2010). In situ visualization and dynamics of newly synthesized proteins in rat hippocampal neurons. Nat. Neurosci. 13, 897–905. doi: 10.1038/nn.2580

Dieterich, D. C., Link, A. J., Graumann, J., Tirrell, D. A., and Schuman, E. M. (2006). Selective identification of newly synthesized proteins in mammalian cells using biorthogonal noncanonical amino acid tagging (BONCAT). Proc. Natl. Acad. Sci. U.S.A. 103, 9482–9487. doi: 10.1073/pnas.0601637103

Dobransky, T., Davis, W. L., Xiao, G. H., and Rylett, R. J. (2000). Expression, purification and characterization of recombinant human choline acetyltransferase: phosphorylation of the enzyme regulates catalytic activity. Biochem. J. 349, 141–151. doi: 10.1042/0264-6021:3490141

CrossRef Full Text | Google Scholar

Dommerholt, J., Rutjes, F. P. J. T., and van Delft, F. L. (2016). Strain-promoted 1,3-dipolar cycloaddition of cycloalkynes and organic azides. Top Curr. Chem. (Cham) 344:16.

Duennwald, M. L. (2013). Yeast as a platform to explore polyglutamine toxicity and aggregation. Methods Mol. Biol. 1017, 153–161. doi: 10.1007/978-1-62703-438-8_11

Eden, E., Geva-Zatorsky, N., Issaeva, I., Cohen, A., Dekel, E., Danon, T., et al. (2011). Proteome half-life dynamics in living human cells. Science 331, 764–768. doi: 10.1126/science.1199784

Eldeeb, M. A., Siva-Piragasam, R., Ragheb, M. A., Esmaili, M., Salla, M., and Fahlman, R. P. (2019). A molecular toolbox for studying protein degradation in mammalian cells. J. Neurochem. 151, 520–533. doi: 10.1111/jnc.14838

Engel, A. G., Shen, X. M., Selcen, D., and Sine, S. (2015). Congenital myasthenic syndromes: pathogenesis, diagnosis, and treatment. Lancet Neurol. 14, 420–434. doi: 10.1016/s1474-4422(14)70201-7

Esposito, A. M., and Kinzy, T. G. (2014). In vivo (35S)-methionine incorporation. Methods Enzymol. 536, 55–64. doi: 10.1016/B978-0-12-420070-8.00005-2

Forester, C. M., Zhao, Q., Phillips, N. J., Urisman, A., Chalkley, R. J., Oses-Prieto, J. A., et al. (2018). Revealing nascent proteomics in signaling pathways and cell differentiation. Proc. Natl. Acad. Sci. U.S.A. 115, 2353–2358. doi: 10.1073/pnas.1707514115

Fowden, L., and Richmond, M. H. (1967). Replacement of proline by azetidine-2-carboxylic acid during biosynthesis of protein. Biochim. Biophys. Acta 71, 459–461. doi: 10.1016/0006-3002(63)91104-1

Lehner, F., Santaniello, W., Bechstein, W. O., Zhuvarel, S. V., Isoniemi, H., Rummo, O. O., et al. (2017). Impact of azidohomoalanine incorporation on protein structure and ligand binding. Chembiochem 18, 2340–2350. doi: 10.1002/cbic.201700437

Giglione, C., Boularot, A., and Meinnel, T. (2004). Protein N-terminal methionine excision. Cell Mol. Life Sci. 61, 1455–1474.

Hanna, J., Guerra-Moreno, A., Ang, J., and Micoogullari, Y. (2019). Protein degradation and the pathological basis of disease. Am. J. Pathol. 189, 94–103.

Hanna, J., Leggett, D. S., and Finley, D. (2003). Ubiquitin depletion as a key mediator of toxicity by translational inhibitors. Mol. Cell Biol. 23, 9251–9261.

Hengstermann, A., Linares, L. K., Ciechanover, A., Whitaker, N. J., and Scheffner, M. (2001). Complete switch from Mdm3 to human papillomavirus E6-mediated degradation of p53 in cervical cancer cells. Proc. Natl. Acad. Sci. U.S.A. 98, 1218–1223. doi: 10.1073/pnas.98.3.1218

Hidalgo San Jose, L., and Signer, R. A. J. (2019). Cell-type-specific quantification of protein synthesis in vivo. Nat. Protoc. 14, 441–460. doi: 10.1038/s41596-018-0100-z

Hinz, F. I., Dieterich, D. C., Tirrel, D. A., and Schuman, E. M. (2012). Non-canonical amino acid labeling in vivo to visualize and affinity purify newly synthesized proteins in larval zebrafish. ACS Chem. Neurosci. 3, 40–49. doi: 10.1021/cn2000876

Hu, V. W., and Heikka, D. S. (2000). Radiolabeling revisited: metabolic labeling with (35)S-methionine inhibits cell cycle progression, proliferation, and survival. FASEB J. 14, 448–454. doi: 10.1096/fasebj.14.3.448

Hu, V. W., Heikka, D. S., Dieffenbach, P. B., and Ha, L. (2001). Metabolic radiolabeling: experimental tool or Trojan horse? (35)S-Methionine induces DNS fragmentation and p53-dependent ROS production. FASEB J. 15, 1562–1568. doi: 10.1096/fj.01-0102com

Karaman, M. W., Herrgard, S., Treiber, D. K., Gallant, P., Atteridge, C. E., Campbell, B. T., et al. (2008). A quantitative analysis of kinase inhibitor selectivity. Nat. Biotechnol. 26, 127–132.

Kawai, S., Hashimoto, W., and Murata, K. (2010). Transformation of Saccharomyces cerevisiae and other fungi: methods and possible underlying mechanism. Bioeng. Bugs. 1, 395–403. doi: 10.4161/bbug.1.6.13257

Kiick, K. L., Saxon, E., Tirrell, D. A., and Bertozzi, C. R. (2002). Incorporation of azides into recombinant proteins for chemoselective modification by the Staudinger ligation. Proc. Natl. Acad. Sci. U.S.A. 99, 18–24.

Kim, Y., Kim, S. H., Ferracane, D., Katzenellenbogen, J. A., and Schroeder, C. M. (2012). Specific labeling of zing finger proteins using noncanonical amino acids and copper-free click chemistry. Bioconjug. Chem. 23, 1891–1901. doi: 10.1021/bc300262h

Knecht, E., Aguado, C., Cárcel, J., Esteban, I., Esteve, J. M., Ghislat, G., et al. (2009). Intracellular protein degradation in mammalian cells: recent developments. Cell Mol. Life Sci. 66, 2427–2443. doi: 10.1007/s00018-009-0030-6

Lee, J., and Seo, J. (2002). Differential expression of two stress-inducible hsp70 genes by various stressors. Exp. Mol. Mol. 34, 131–136. doi: 10.1038/emm.2002.19

Lehtonen, S., Sonninen, T. M., Wojciechowski, S., Goldsteins, G., and Koistinaho, J. (2019). Dysfunction of cellular proteostasis in Parkinson’s disease. Front. Neurosci. 13:457.

Link, A. J., Vink, M. K. S., and Tirrell, D. A. (2007). Synthesis of the functionalizable methionine surrogate azidohomoalanine using Boc-homoserine as a precursor. Nat. Protoc. 2, 1884–1887. doi: 10.1038/nprot.2007.269

Liu, J., Xu, Y., Stoleru, D., and Salic, A. (2012). Imaging protein synthesis in cells and tissues with an alkyne analog of puromycin. Proc. Natl. Acad. Sci. U.S.A. 109, 413–418. doi: 10.1073/pnas.1111561108

Lukashchuk, N., and Vousden, K. H. (2007). Ubiquitination and degradation of mutant p53. Mol. Cell Biol. 27, 8284–8295. doi: 10.1128/mcb.00050-07

Ma, Y., and Yates, J. R. (2018). Proteomics and pulse azidohomoalanine labeling of newly synthesized proteins: what are the potential applications? Expert Rev. Proteomics. 15, 545–554. doi: 10.1080/14789450.2018.1500902

Mantovani, F., Collavin, L., and Del Sal, G. (2019). Mutant p53 as a guardian of the cancer cell. Cell Death Differ. 26, 199–212. doi: 10.1038/s41418-018-0246-9

McShane, E., Sin, C., Zauber, H., Wells, J. N., Donnelly, N., Wang, X., et al. (2016). Kinetic analysis of protein stability reveals age-dependent degradation. Cell 167, 803–815. doi: 10.1016/j.cell.2016.09.015

Morey, T. M., Albers, S., Shilton, B. H., and Rylett, R. J. (2016). Enhanced ubiquitination and proteasomal-degradation of catalytically-deficient human choline acetyltransferase mutants. J. Neurochem. 137, 630–646. doi: 10.1111/jnc.13574

Morey, T. M., Winick-Ng, W., Seah, C., and Rylett, R. J. (2017). Chaperone-Mediated Regulation of Choline Acetyltransferase Protein Stability and Activity by HSC/HSP70, HSP90, and p97/VCP. Front. Mol. Neurosci. 10:415. doi: 10.3389/fnmol.2017.00415

Mullen, P. (2004). PARP cleavage as a means of assessing apoptosis. Methods Mol. Med. 88, 171–181. doi: 10.1385/1-59259-406-9:171

Oda, Y. (1999). Choline acetyltransferase: the structure, distribution and pathological changes in the central nervous system. Pathol. Int. 49, 921–937.

Resendes, M. C., Dobransky, T., Ferguson, S. S., and Rylett, R. J. (1999). Nuclear localization of the 82-kDa form of human choline acetyltransferase. J. Biol. Chem. 274, 19417–19421. doi: 10.1074/jbc.274.27.19417

Roche, F. K., Marsick, B. M., and Letourneau, P. C. (2009). Protein synthesis in distal axons is not required for growth cone responses to guidance cues. J. Neurosci. 29, 638–652. doi: 10.1523/jneurosci.3845-08.2009

Roth, S., Drewe, W. C., and Thomas, N. R. (2010). A concise and scalable route to L-azidohomoalanine. Nat. Protoc. 5, 1967–1973. doi: 10.1038/nprot.2010.164

Saleh, A. M., Wilding, K. M., Calve, S., Bundy, B. C., and Kinzer-Ursem, T. L. (2019). Non-canonical amino acid labeling in proteomics and biotechnology. J. Biol. Eng. 13:43.

Sanders, B. C., Friscourt, F., Ledin, P. A., Mbua, N. E., Arumugam, S., Guo, J., et al. (2011). Metal-free sequential (3 + 2)-dipolar cycloadditions using cyclooctynes and 1,3-dipoles of different reactivity. J. Am. Chem. Soc. 133, 949–957. doi: 10.1021/ja1081519

Schneider-Poetsch, T., Ju, J., Eyler, D. E., Dang, Y., Bhat, S., Merrick, W. C., et al. (2010). Inhibition of eukaryotic translation elongation by cycloheximide and lactimidomycin. Nat. Chem. Biol. 6, 209–217. doi: 10.1038/nchembio.304

Sharma, M., Pampinella, F., Nemes, C., Benharouga, M., So, J., Du, K., et al. (2004). Misfolding diverts CFTR from recycling to degradation: quality control at early endosomes. J. Cell Biol. 164, 923–933. doi: 10.1083/jcb.200312018

Shen, X. M., Crawford, T. O., Brengman, J., Acsadi, G., Iannaconne, S., Karaca, E., et al. (2011). Functional consequences and structural interpretation of mutations of human choline acetyltransferase. Hum. Mutat. 32, 1259–1267. doi: 10.1002/humu.21560

Sorger, P. K., Lewis, M. J., and Pelham, H. R. (1987). Heat shock factor is regulated differently in yeast and HeLa cells. Nature 329, 81–84. doi: 10.1038/329081a0

Steward, K. F., Eilers, B., Tripet, B., Fuchs, A., Dorle, M., Rawle, R., et al. (2020). Metabolic implications of using bioorthogonal non-canonical amino acid tagging (BONCAT) for tracking protein synthesis. Front. Microbiol. 11:197.

Takahashi, M., and Ono, Y. (2003). Pulse-chase analysis of protein kinase C. Methods Mol. Biol. 233, 163–170. doi: 10.1385/1-59259-397-6:163

Trotter, E. W., Berenfeld, L., Krause, S. A., Petsko, G. A., and Gray, J. V. (2001). Protein misfolding and temperature up-shift cause G1 arrest via a common mechanism dependent on heat shock factor in saccharomyces cerevisiae. Proc. Natl. Acad. Sci. U.S.A. 98, 7313–7318. doi: 10.1073/pnas.121172998

Van Geel, R., Pruijin, G. J. M., van Delft, F. L., and Boelens, W. C. (2012). Preventing thiol-yne addition improves the specificity of strain-promoted azide-alkyne cycloaddition. Bioconjug. Chem. 23, 392–398. doi: 10.1021/bc200365k

Varshavsky, A. (2008). Discovery of cellular regulation by protein degradation. J. Biol. Chem. 283, 34469–34489. doi: 10.1074/jbc.x800009200

Weids, A. J., Ibstedt, S., Tamas, M. J., and Grant, C. M. (2016). Distinct stress conditions result in aggregation of proteins with similar properties. Sci. Rep. 6:24554.

Weinhandl, K., Winkler, M., Glieder, A., and Camattari, A. (2014). Carbon source dependent promoters in yeasts. Microb. Cell Fact. 13:5. doi: 10.1186/1475-2859-13-5

Yoshimori, T., Yamamoto, A., Moriyama, Y., Futai, M., and Tashiro, Y. (1991). Bafilomycin A1, a specific inhibitor of vacuolar-type H(+)-ATPase, inhibits acidification and protein degradation in lysosomes of cultured cells. J. Biol. Chem. 266, 17707–17712. doi: 10.1016/s0021-9258(19)47429-2

Zhang, L., Gurskaya, N. G., Merzlyak, E. M., Staroverov, D. B., Mudrik, N. N., Samarkina, O. N., et al. (2007). Method for real-time monitoring of protein degradation at the single cell level. Biotechniques. 42, 446–448. doi: 10.2144/000112453

Keywords : protein stability and degradation, protein half-life, pulse-chase analysis, click chemistry, SPAAC, mammalian cells, yeast

Citation: Morey TM, Esmaeili MA, Duennwald ML and Rylett RJ (2021) SPAAC Pulse-Chase: A Novel Click Chemistry-Based Method to Determine the Half-Life of Cellular Proteins. Front. Cell Dev. Biol. 9:722560. doi: 10.3389/fcell.2021.722560

Received: 09 June 2021; Accepted: 10 August 2021; Published: 07 September 2021.

Reviewed by:

Copyright © 2021 Morey, Esmaeili, Duennwald and Rylett. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Trevor M. Morey, [email protected] ; Martin L. Duennwald, [email protected] ; R. Jane Rylett, [email protected]

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

• Laboratory Techniques & Protocols , Science

Pulse Chase Analysis

Need Incubators for your lab?

As an Amazon Associate Conductscience Inc earns revenue from qualifying purchases

Introduction

The pulse-chase analysis is a powerful technique to study the synthesis, processing, and transport of proteins. The pulse-chase analysis is widely used in biochemistry and molecular biology for the examination of the cellular process by exposing the cells to a labeled compound (pulse). The labeled compound is introduced into the molecule or pathway being studied. In the chase phase, an unlabeled form of that compound replaces the labeled compound. The reaction is monitored to note the time taken by the unlabeled form of the compound to replace the labeled form. These experiments allow the assessment and observation of the evolution of a biological process over time.

Pulse-chase analysis is a specialized form of metabolic labeling which employs radioactive amino acids for observing cellular events over time. The radioactive amino acids are added in the cells for a few minutes (the “pulse”), washed away, and then the cells are exposed to nonradioactive amino acids (in excess). The cells are then taken for protein extraction. Pulse-chase analysis is used for the study of protein synthesis and processing, the examination of intracellular localization of nascent proteins over time, and for the assessment of their secretion, translocation, and degradation.

The pulse-chase experiment consists of three stages: the actual pulse-chase, immunoprecipitation, and sodium-dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). In the experiment, a fluorescently or radioactively tagged compound is used for time-based monitoring of the cellular events. After the pulse-chase, the proteins of interest are harvested by cell lysis through immunoprecipitation. The proteins are then run on SDS-PAGE for further analysis. The equipment needed for the experiment includes an incubator, Eppendorf tubes, a 26-gauge needle, a syringe, heat block, gel dryer, and Phosphor imaging screen.

Protocol (Magadán, 2017)

Radioactive pulse-chase

• Incubate the MDCK cells (grown and infected with 10 infectious doses of influenza A virus per cell for 5 hours at 37 o C) with 5 ml Trypsin-EDTA at 37 o C for 5 minutes.
• Transfer the cells to a 50 ml BD falcon tube and wash them twice with 10 ml pre-warmed Dulbecco’s phosphate-buffered saline (DPBS) and centrifuge for 1 minute at 2500 x g.
• Resuspend the cells in 1 ml DPBS and transfer in a 1.5 ml microcentrifuge tube.
• Centrifuge the cells at maximum speed for 15 seconds.
• Resuspend the cells in 200 µl of pre-warmed pulse medium ( Dulbecco Modified Eagle Medium‎ DMEM supplemented with 0.20 mM L-cysteine and 0.2 mCi/ml [35S]-methionine). Then, incubate the cells in a water bath at 37 o C for 2 minutes.
• Centrifuge for 15 seconds at 4 o C at maximum speed.
• Resuspend the pellet in 1.05 ml pre-warmed chase medium ( DMEM media supplemented with 7.5% fetal bovine serum and 67 mM L-methionine).
• Incubate the cells in a water bath at 37 o C for 20 minutes.
• Take 190 µl aliquots each at 0, 5, 10, 15, 20 minutes.
• Immediately transfer them to 1.5 ml centrifuge tubes containing 1 ml ice-cold DPBS.
• Centrifuge for 15 seconds at 4 o
• Resuspend the pellet in 1 ml non-denaturing lysis buffer (ice-cold)
• Incubate the cell lysate at 4 o C for 30 minutes with slow rotation.
• Centrifuge for 15 minutes at 4 o C at maximum speed.
• Discard the pellet and keep the supernatant.

Immunoprecipitation

• Wash 30 µl of protein A sepharose twice with 0.5 ml ice-cold Dulbecco’s phosphate-buffered saline DPBS and centrifuge for 1 minute at 3000 x g at 4 o
• Resuspend resin in 0.5 ml ice-cold DPBS supplemented with 0.001% Triton X-100 and anti-HA antibody of interest.
• Incubate for 2 hours at 4 o C under slow rotation.
• Wash the resin with 0.5 ml ice-cold non-denaturing lysis buffer (0.5% Triton X-100, 50 mM Tris-HCl (pH 7.5), 300 mM NaCl, 5 mM EDTA , and Complete mini, EDTA-free protease inhibitor cocktail) for two times.
• Add 10 µl bovine serum albumin .
• Add cleared cell lysate from each aliquot.
• Wash the resin twice with 0.5 ml ice-cold non-denaturing lysis buffer supplemented with 0.1% instead of 0.5% Triton X-100.
• Wash the resin with 0.5 ml ice-cold DPBS.
• Resuspend the resin in 20 µl of 4x LDS buffer.
• Boil the samples for 5 minutes.

SDS-PAGE and fluorography

• Load 15 µl of each sample on protein mini-gels.
• Run the gel for 3 hours at a constant 50 mA/gel.
• Fix the gels with 10 ml fixation solution (50% methanol and 10% acetic acid) for 30 minutes at room temperature under slow rocking.
• Dry gels in gel drier for 1.5 hours at 80 o
• Expose the films to the radioactive gels at room temperature overnight.

Single-cell pulse-chase for nuclear components ( Drozdz. & Vaux., 2016 )

Cell culture

• Culture the cells in cell culture medium containing Dulbecco’s modified Eagle’s medium, 10 % fetal calf serum , nonessential amino acids, and penicillin and streptomycin (100 U/mL).
• Grow the cells in an incubator at 37 °C with 5 % CO 2 supply and split them on 80 % confluence.
• Plate the cells on specific media and perform transfections according to the standard protocols.

Nano SIMS: preparation of stearic acid D35 stock solution

• Add 2 ml ethanol in 3.2 mg of stearic acid and dissolve.
• Mix it with 400 μL of 100 mM solution of NaOH in ethanol.
• Evaporate alcohol to get fatty acid soaps.
• Dissolve the fatty acid soaps in ultrapure water (0.5 ml) prewarmed to 55 °C and keep in a 55 °C water bath for 10 minutes.
• Dissolve 1 g of bovine serum albumin (fatty acid-free) in 9.5-mL Dulbecco’s Modified Eagle’s Medium (DMEM) and warm up to 55 °C.
• Add the dissolved fatty acid soaps to the bovine serum albumin solution (fatty acid-free) and then vortex for 10 seconds, and subsequently incubated for 10-minutes at 55 °C.
• Filter the stock solution under sterile conditions, aliquot, and keep at -20 °C to make a 1 mM stearic acid D35 stock solution.

Stearic Acid D35 cell labeling and embedding

• Plate mouse preadipocytes on 13-mm plastic coverslips in a 24-well plate at a 50 % confluence rate and leave them overnight.

Note: Coverslips pre-coated with poly-l-lysine, collagen, or more elaborate extracellular matrix preparations can be used depending on the cell type.

• Treat the cells with 10 μM stearic acid D35. Add 100 μL of 1 mM stearic acid stock solution to the 10-mL culture medium (pre-warmed) and vortex. Use it to replace the medium in the 24-well plate (500 μL per well). Leave for 6 hours in the incubator.
• Aspirate the medium and wash the coverslips with 1 mL phosphate buffered saline (PBS). Discard the PBS.
• Fix the cells by adding 500 μL of fixative warmed to 37 °C. Use 4% paraformaldehyde and 1 % glutaraldehyde solution in 100 mM PIPES pH 7.4. Leave for 20 minutes at room temperature.
• Replace the fixative with 500 μL of 2.5 % glutaraldehyde dissolved in 100 mM PIPES pH 7.4.
• Incubate for 1 hour at room temperature and transfer to 4 °C. Leave it overnight.
• Wash the samples thrice with 100 mM PIPES pH 7.4 for 10 minutes.
• Osmicate the cells with 1 % osmium tetroxide in 100 mM PIPES pH 7.4 for 1 hour.
• Wash the samples with deionized water for 20 minutes.
• Incubate the samples in a graded ethanol series, starting from 50 % ethanol for 15 minutes, then 70 % ethanol overnight at 4 °C, then 90 % ethanol for 15 minutes, then 95 % ethanol for 15 minutes, and 100 % ethanol for 2 hours with three solution changes during this time.
• Gradually infiltrate the samples with agar 100 epoxy resin by serial replacement of the resin solution in the coverslip-containing wells. Start with 25 % resin for 1 hour, then 50 % resin for 2 hours, then 75 % resin for 1 hour, and 100 % resin overnight.
• Transfer the samples to fresh 100 % resin for 3 hours and repeat this step once more.
• Resin-embed the layer of cells by inverting the coverslip onto an embedding capsule filled with fresh 100 % resin. Leave for 24 hours at 60 °C to allow polymerization.
• Submerge the blocks in liquid nitrogen and take off the coverslip and leave the cells embedded in the resin.
• Cut the resin block using a razor blade and glass knife to generate a trapezoid end containing the specimen.
• Cut the specimen into semi-thin sections of 0.5 μm using an ultramicrotome with a diamond knife, and mount them on 15-nm platinum-coated coverslips. Place them on a 60 °C heating block and allow to dry for a few minutes.

Backscattered electron imaging and NanoSIMS

• Record the points of interest by optical microscopy before starting backscattered electron (BSE) imaging.
• Transfer the sections to a scanning electron microscope, and acquire BSE images with a 2-kV incident beam with standard aperture (30 μm) and 5-mm working distance.
• Coat the sections with 5 nm of platinum in a high-resolution sputter coater to make the surface conductive for NanoSIMS imaging.
• A small region of the sample is revealed by uncoating to remove the material from the section for 2 H − and 1 H − mass analysis using the cesium ion beam.
• Use small apertures (D1 = 3 or D1 = 4) for single cells imaging to match the primary beam size with the pixel size. Tune the instrument for 2 H − and 1 H – for morphological information and calculate the 2 H/ 1 H ratio. Collect and process NanoSIMS images with a dwell time of 30,000 μs per pixel for a 256 × 256-pixel image.
• Manually align the BSE and NanoSIMS images for the identification of regions containing structures of biological interest.

Applications

Analysis of the in vivo assembly of bacteriophage T4 tail ( Ferguson. & Coombs., 2000 )

The pulse-chase analysis is a non-invasive tool to determine the assembly pathway of the complex tail of bacteriophage T4 virus. In the study, bacteriophage T4 mutants defective in the head assembly were used to infect the E. coli cultures to study tail assembly in isolation. The bacterial cultures were labeled with [3H] leucine on the onset of late protein synthesis. After the complete tail began to accumulate at a constant rate, the bacterial cultures were pulsed with [35S] methionine, and then chased. Completed tails were obtained and purified at one-minute intervals for the next 30 minutes. It was found that the closer the assembly point to the end of the pathway, the sooner the chase appears, presenting the assembly cascade. The results showed that the tail completion proteins such as gp18 (tail sheath) and 19 (tail tube) show the earliest inflection as compared to other tail proteins. The study validated the application of pulse-chase analysis to analyze and assess the assembly of viral components.

Assessment of protein maturation and degradation ( Jansens. & Braakman., 2003 )

The pulse-chase analysis is a powerful technique to analyze protein maturation and degradation. Using the short pulses, the whole process of protein synthesis to degradation can be determined in a natural environment. The technique has been widely used for endogenous and viral proteins including thyroglobulin, vesicular stomatitis virus, influenza hemagglutinin, envelope proteins, and immunoglobulins. This biochemical approach is a cutting-edge analysis method to study a variety of processes including protein folding, post-translational modifications, endogenous and intracellular transport, and the rate of protein degradation.

Assessment of MHC class II biosynthesis, maturation, and peptide loading ( Hou. et al. 2013 )

The pulse-chase analysis is a widely used technique for the exploration of the synthesis, processing, and transport of proteins. In the research, the pulse-chase experiment was used to study the major histocompatibility complex (MHC) class II synthesis, maturation, trafficking, and peptide loading in human Epstein-Barr virus-transformed B-lymphoblastoid cell lines (B-LCL). The MHC class II glycoproteins bind heterogeneous mixtures of peptides presented on the surface of antigen-presenting cells for the inspection of CD4+ T helper lymphocytes. The results obtained from these experiments can illustrate information to track changes in the molecular associations, an abundance of radiolabeled proteins, and the conformation-sensitive monoclonal antibodies (mAbs).

Biosynthesis, targeting, and processing of lysosomal proteins ( Pohl. & Hasilik., 2015 )

The radioactive amino acids and modifier groups are incorporated into proteins to study their life cycle and various modifications. Lysosomal enzymes can also be detected and characterized using pulse-chase analysis. The pulse-chase labeling provides a detailed insight into organelle-specific modifications of lysosomal proteins. For this, an antibody against lysosomal protease is used as a reference. In the study, cathepsin Z was synthesized as a larger proenzyme containing two N-linked oligosaccharides which mature to a shorter single-chain enzyme with oligosaccharides. The pulse-chase experiment demonstrates the conversion of the precursor into the mature form. The technique could also be used to study the deglycosylation of metabolically labeled proteins and the alterations in the apparent size of the attached glycopeptides.

Precautions

• Prepare fresh pulse-chase media before use.
• During lysis, it is essential to maintain the nucleus intact.
• If the protein expression is relatively low, then the cells can be transfected with a plasmid encoding the protein behind the promotor or a lipid mixture.
• Separate the folding process from translation if the effects of certain conditions such as ATP depletion are tested.
• Cycloheximide should be added in chase medium if the kinetics of the process has to be studied.

Strengths and Limitations

• The pulse-chase analysis is a powerful technique used for the assessment of protein synthesis, trafficking, and degradation.
• The technique could also be used to monitor the binding of proteins and the kinetics of the processes.
• The pulse-chase method is widely used to monitor the assembly cascade of viruses.
• The method should be readily applicable to the detection and assessment of multicomponent regulatory complexes and macromolecular machines accessible to the biochemical and biophysical investigation.
• The approach is limited to monitoring the kinetics of stable interactions.
• The labeling of the proteins could be challenging for low-abundant proteins.
• A. Jansens., & Braakman., I. (2003). Pulse-chase labeling techniques for the analysis of protein maturation and degradation . Methods Mol Biol, 232, 133-45.
• P. Ferguson., & Coombs., H. D. (2000). Pulse-chase analysis of the in vivo assembly of the bacteriophage T4 tail. J Mol Biol, 297(1), 99-117.
• Drozdz., & Vaux., J. D. (2016). Methods for Single-Cell Pulse-Chase Analysis of Nuclear Components. Methods Mol Biol, 1411, 159-76.
• Magadán, J. G. (2017). Radioactive Pulse-Chase Analysis and Immunoprecipitation . Bio Protoc, 4(8).
• Pohl., & Hasilik., A. (2015). Biosynthesis, targeting, and processing of lysosomal proteins: pulse-chase labeling and immune precipitation . Methods Cell Biol, 126, 63-83.
• Hou., C. H. Rinderknecht., A. V. Hadjinicolaou., R. Busch., & Mellins, E. (2013). Pulse-chase analysis for studies of MHC class II biosynthesis, maturation, and peptide loading . Methods Mol Biol, 960, 411-432.

Our Location

• 5250 Old Orchard Rd Suite 300 Skokie, IL 60077
• +1 847 983 3672

Monday – Friday 9 AM – 5 PM EST

Conduct Science

• Become a Partner
• Social Media
• Career/Academia
• Privacy Policy
• Shipping & Returns
• Request a quote

Customer service

• Account Details
• Lost Password

DISCLAIMER: ConductScience and affiliate products are NOT designed for human consumption, testing, or clinical utilization. They are designed for pre-clinical utilization only. Customers purchasing apparatus for the purposes of scientific research or veterinary care affirm adherence to applicable regulatory bodies for the country in which their research or care is conducted.

An official website of the United States government

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.

• Publications
• Account settings
• Advanced Search
• Journal List

Pulse Chase of Suspension Cells

Lai-yee wong, qiming liang, kevin brulois.

• Author information
• Copyright and License information

For correspondence: [email protected]

Pulse-chase method is a powerful technique used to follow the dynamics of proteins over a period of time. The expression level, processing, transport, secretion or half-life of proteins can be tracked by metabolically labeling the cells, such as with radiolabeled amino acids (pulse step). This protocol describes the condition used to study the folding and disulfide bond formation of immunoglobulin in suspension cells. With some minor modifications, this protocol can be adapted to study the degradation rate or the secretion of target proteins.

Materials and Reagents

A. pulse chase.

Cells growing in suspension

HBSS (Life Technologies, Invitrogen™, catalog number: 14175-095)

RPMI without methionine and cysteine (Sigma-Aldrich, catalog number: R7513)

Dialyzed FBS (Life Technologies, Invitrogen™, catalog number: 26400044)

N -Ethylmaleimide (NEM) (Sigma-Aldrich, catalog number: R3876)

Cyclohexamide (CHX) (Sigma-Aldrich, catalog number: C7698)

Express 35 S protein labelling mix (Perkin Elmer, catalog number: NEG072014MC)

Methionine (Sigma-Aldrich, catalog number: M5308)

Cysteine (Sigma-Aldrich, catalog number: C7352)

Labeling medium (see Recipes )

Chase medium (see Recipes )

2x stop buffer (see Recipes )

Labeling medium

RPMI lacking methionine or cysteine

1% Penicilin and Streptomycin

1% glutamine

5-10% dialyzed FBS (see Note 6 )

Chase medium (see Note 7 )

Labeling medium (see above) plus

5 mM cysteine

5 mM methionine

2x stop buffer

HBSS with 40 mM NEM (see Note 8 )

Lysis buffer

50 mM Tris (pH 7.4)

150 mM NaCl

0.5% Na deoxycholate

Protease inhibitors

Gel fixing solution

25% methanol

15% acetic acid

B. Cell lysis and immunoprecipitation

Antibody against protein of interest

Protein A/G beads (Thermo Fisher Scientific, catalog number: 20422)

Complete Protease Inhibitor Tablets (Roche Diagnostics, catalog number: 11836145001)

Lysis buffer (see Recipes )

C. SDS-PAGE

4-12% Bis-Tris protein gel (Life Technologies, Invitrogen™)

MOPS running buffer (Life Technologies, Invitrogen™, catalog number: NP0001)

Amplify solution (GE Healthcare, catalog number: NAMP100)

Gel drying solution (Life Technologies, Invitrogen™, catalog number: LC4025)

Gel fixing solution (see Recipes )

Eppendorf tube

26-gauge needle

1 ml syringe

Phosphor imaging screen

Cells are pulse-labeled and chased in a single tube and an aliquot of cells is removed from this tube for each time point of the chase.

After determining the numbers of chase time points (x), prepare enough cells for the experiment (x + 1, 2 × 10 6 per sample). See Note 1 .

Wash cells (2 × 10 6 per sample) with 2 ml of HBSS.

Pellet cells at 500 x g for 3 min at room temperature. Resuspend cells in 2 ml/sample of pre-warmed labeling medium. Mix gently.

Incubate the cells for 20 min at 37 °C incubator.

Pellet cells at 500 x g for 3 min at room temperature and resuspend cells in 100 μl/sample of pre-warmed labeling medium. Keep the cells at 37 °C, either in a water bath or an incubator during the labeling and chase periods.

Pulse for 2 min at 37 °C with [ 35 S] methionine (100 μCi/ml) (see Note 2 ).

Add 400 μl/sample of chase medium. Pipet up and down gently to ensure proper mixing.

Immediately take out 500 μl for the 0 min sample. Transfer to an eppendorf tube filled with 500 μl of 2x ice cold stop buffer (see Note 3 ).

Spin cells at 500 x g for 2 min at 4 °C and freeze pellet.

Repeat steps 7-8 for every time point.

Proceed to cell lysis or keep the pellet frozen in −80 °C.

Add 1ml cold lysis buffer to each cell pellet.

Apply mechanical shearing force to the cell lysate by passing it through a 26-gauge needle attached to a 1 ml syringe (repeat 5-10 times for thorough lysis). Incubate on ice for 15 min. Spin at 16,000 x g for 15 min at 4 °C to clarify the lysate. Transfer the clarified cell extracts to a new tube.

Add antibody against protein of interest to equal volume of cell extracts and rotate overnight at 4 °C (see Note 4 ).

Add 30 μl of Protein A/G beads and incubate for another 2 h at 4 °C.

Wash the immunoprecipitates twice with 1 ml of lysis buffer.

Add 50 μl of sample buffer without reducing agents (such as DTT or 2-mercaptoenthanol). See Note 5 .

Heat the samples 65 °C for 5 min. Centrifuge briefly before proceeding to non-reducing SDS-PAGE.

Load 25 μl of the immunoprecipitates on SDS-polyacrylamide gel.

Immerse the gel in fixing buffer for 15-30 min at room temperature.

Immerse the gel in amplify solution for 15 min at room temperature.

Immerse the gel in gel drying solution for another 15 min at room temperature.

Dry the gel on filter paper on top of a gel dryer of choice.

[ 35 S] methionine-labeled proteins can be visualized after exposure to a Phosphor imaging screen.

Representative data

The protocol described here was used to examine the maturation kinetics of pentameric IgM complexes during their passage through the secretory pathway. IgM assembly begins with the coupling of a heavy chain (H) and a light chain (L), resulting in a monomeric heavy and light chain intermediate (HL). A H2L2 unit is assembled and followed by large multimers of “H2L2”. At later points of the chase time, the signals for these IgM intermediates, especially the high molecular weight species, will decrease due to the successful assembly and secretion of mature IgM.

Figure 1. I.29 μ + mouse lymphoma cells were pulse labeled with 35 S-methoninie for 2 minutes and chased for the indicated times.

The cell extracts were immunoprecipitated with αIgM and the immunoprecipitates were resolved by non-reducing SDS/PAGE. IgM assembly intermediates are indicated.

Acknowledgments

The protocol presented here was adapted from van Anken et al. (2009). This work was partly supported by grants from the National Institutes of Health (CA082057, CA31363, CA115284, CA180779, AI105809, and AI073099); the Hastings Foundation; the Fletcher Jones Foundation and BaCaTec (J.U.J.). We also thank the Jung laboratory members for their support and discussions.

In order to ensure equal number of cells are used in each samples, it is necessary to determine the number of chase time points (x) and the volume of chase medium before starting the experiment. We normally prepare enough cells and medium for x (number of chase time points) plus 1 to account for any fluid loss during the experiment.

The incubation time with radioactive amino acids should be optimized depending on the target protein and the cells used. Subsequently, keeping the labeling step consistent, especially the number of cells and ‘pulse’ time with radioactive amino acids, will reduce variability in the end results.

Alternatively, the cells can be added directly to ice cold 2x lysis buffer and cell lysis can commence immediately.

The amount of antibody and cell extracts to be added and the incubation time for antibody-antigen binding should be determined and optimized according to each antibody and antigen. Incubation time can vary from 1 h to overnight at 4 °C.

Denaturing sample buffer with reducing agents can be used if disulfide bond formation is not being monitored.

Dialyzed FBS is used to prevent contamination of ‘cold’ methionine and cysteine. The amount added to the medium should be determined according to the cell line used.

Cyclohexamide (CHX) is added to reduce the level of newly translated protein. CHX as well as ‘cold’ methionine and cysteine should be added fresh to the medium.

NEM is an alkylating agent that covalently attaches to free SH-groups found on cysteines. It is added to prevent disulfide bonds from forming once the chase period has ended. In pulse-chase experiments that track the formation of disulfide bonds, NEM must be added to the stop and lysis buffers.

• 1. van Anken E, Pena F, Hafkemeijer N, Christis C, Romijn EP, Grauschopf U, Oorschot VM, Pertel T, Engels S, Ora A, Lastun V, Glockshuber R, Klumperman J, Heck AJ, Luban J and Braakman I (2009). Efficient IgM assembly and secretion require the plasma cell induced endoplasmic reticulum protein pERp1. Proc Natl Acad Sci U S A 106(40): 17019–17024. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ]
• 2. Wong LY, Brulois K, Toth Z, Inn KS, Lee SH, O’Brien K, Lee H, Gao SJ, Cesarman E, Ensser A and Jung JU (2013). The product of Kaposi’s sarcoma-associated herpesvirus immediate early gene K4.2 regulates immunoglobulin secretion and calcium homeostasis by interacting with and inhibiting pERP1. J Virol 87(22): 12069–12079. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ]
• View on publisher site
• PDF (398.8 KB)
• Collections

Similar articles

Cited by other articles, links to ncbi databases.

• Download .nbib .nbib
• Format: AMA APA MLA NLM

Add to Collections

Pulse-Chase Experiment for the Analysis of Protein Stability in Cultured Mammalian Cells by Covalent Fluorescent Labeling of Fusion Proteins

• First Online: 01 January 2009

Cite this protocol

• Kei Yamaguchi 2 ,
• Shinichi Inoue 2 ,
• Osamu Ohara 3 , 4 &
• Takahiro Nagase 2  

Part of the book series: Methods in Molecular Biology™ ((MIMB,volume 577))

3305 Accesses

21 Citations

We used HaloTag ® labeling technology for the pulse labeling of proteins in cultured mammalian cells. HaloTag ® technology allows a HaloTag-fusion protein to covalently bind to a specific small molecule fluorescent ligand. Thus specifically labeled HaloTag-fusion proteins can be chased in cells and observed in vitro after separation by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). The Fluorescent HaloTag ® ligand allows quantification of the labeled proteins by fluorescent image analysis. Herein, we demonstrated that the method allows analysis of the intracellular protein stability as regulated by protein-degradation signals or an exogenously expressed E3 ubiquitin ligase.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Subscribe and save.

• Get 10 units per month
• Download Article/Chapter or eBook
• 1 Unit = 1 Article or 1 Chapter
• Cancel anytime
• Available as PDF
• Read on any device
• Instant download
• Own it forever
• Available as EPUB and PDF
• Compact, lightweight edition
• Dispatched in 3 to 5 business days
• Free shipping worldwide - see info
• Durable hardcover edition

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Similar content being viewed by others

Nonradioactive quantification of autophagic protein degradation with L -azidohomoalanine labeling

Spatially resolved proteomic mapping in living cells with the engineered peroxidase APEX2

Determination of Protein Turnover Rates in the JAK/STAT Pathway Using a Radioactive Pulse-Chase Approach

Zhao, Z., Li, X., Hao, J., Winston, J. H., and Weinman, S. A. (2007) The ClC-3 chloride transport protein traffics through the plasma membrane via interaction of an N-terminal dileucine cluster with clathrin J Biol Chem 282 , 29022–29031.

Article   PubMed   CAS   Google Scholar  

Magee, A. I., Wootton, J., and de Bony, J. (1995) Detecting radiolabeled lipid-modified proteins in polyacrylamide gels Methods Enzymol 250 , 330–336.

Mosteller, R. D., Goldstein, R. V., and Nishimoto, K. R. (1980) Metabolism of individual proteins in exponentially growing Escherichia coli J Biol Chem 255 , 2524–2532.

CAS   Google Scholar  

Bonifacino, J. S. (1999) Metabolic labeling with amino acids P. 3.7.1–3.7.10. In J. E. Coligan, B. M. Dunn, D. W. Speicher, P. T. Wingfield, and G. P. Taylor (Eds.), Current Protocols in Protein Science 1 , Wiley, Hoboken, NJ.

Google Scholar  

Gronemeyer, T., Godin, G., and Johnsson, K. (2005) Adding value to fusion proteins through covalent labeling Curr Opin Biotechnol 16 , 453–458.

Keppler, A., Gendreizig, S., Gronemeyer, T., Pick, H., Vogel, H., and Johnsson, K. (2003) A general method for the covalent labeling of fusion proteins with small molecules In Vivo Nat Biotechnol 21 , 86–89.

Article   CAS   Google Scholar  

Los, V. G., and Wood, K. (2007) The HaloTag ™ p. 195–208. In D. L. Taylor, J. R. Haskins, and K. Giuliano (Eds.), Methods in Molecular Biology 356 . Human Press, Totowa, NJ.

Nagase, T., Yamakawa, H., Tadokoro, S., Nakajima, D., Inoue, S., Yamaguchi, K., Itokawa, Y., Kikuno, F. R., Koga, H., and Ohara, O. (2008) Exploration of human ORFeome: High throughput preparation of ORF clones and efficient characterization of their protein products DNA Res 15 , 137–149.

Yamaguchi, K., Nagase, T., Ando, A., and Ohara, O. (2008) Smurfi directly targets hPEM-2, a GEF for Cdc42, via a novel combination of protein interaction modules in ubiquitin–proteasome pathway Biol Chem In press.

Rogers, S., Wells, R., and Rechsteiner, M. (1986) Amino acid sequences common to rapidly degraded proteins: The PEST hypothesis Science 234 , 364–368.

Gilon, T., Chomsky, O., and Kulka, R. G. (1998) Degradation signals for ubiquitin system proteolysis in Saccharomyces cerevisiae EMBO J 17 , 2759–2766.

Zhu, H., Kavsak, P., Abdollah, S., Wrena, J. L., and Thomsen, G. H. (1999) A Smad ubiquitin ligase targets the BMP pathway and affects embryonic pattern formation Nature 400 , 687–693.

Rual, J. F., Hill, D. E., and Vidal, M. (2004) ORFeome projects: gateway between genomics and omics Curr Opin Chem Biol 8 , 20–25.

Nakajima, D., Saito, K., Yamakawa, H., et al. (2005) Preparation of a set of expression-ready clones of mammalian long cDNAs encoding large proteins by the ORF trap cloning method DNA Res 12 , 257–267.

Temple, G., Lamesch, P., Milstein, S., et al. (2006) From genome to proteome: developing expression clone resources for the human genome Hum Mol Genet 15 , R31–R43.

Download references

Acknowledgments

This project was supported by a grant from the Chiba prefectural government. We are grateful to B. Bulleit for his valuable comments. We thank K. Ozawa, T. Watanabe, and K. Yamada for their technical assistance.

Author information

Authors and affiliations.

Laboratory of Human Gene Research, Department of Human Genome Research, Kazusa DNA Research Institute, Chiba, Japan

Kei Yamaguchi, Shinichi Inoue & Takahiro Nagase

Department of Human Genome Research, Kazusa DNA Research Institute, Chiba, Japan

Osamu Ohara

Laboratory for Immunogenomics, RIKEN Research Institute for Allergy and Immunology, Yokohama City, Kanagawa, Japan

You can also search for this author in PubMed   Google Scholar

Editor information

Editors and affiliations.

Iwai Orthopaedic Medical Hospital, Minamikoiwa 8-17-5, Tokyo, 133-0055, Japan

Hisashi Koga

Rights and permissions

Reprints and permissions

Copyright information

© 2009 Humana Press, a part of Springer Science+Business Media, LLC

About this protocol

Yamaguchi, K., Inoue, S., Ohara, O., Nagase, T. (2009). Pulse-Chase Experiment for the Analysis of Protein Stability in Cultured Mammalian Cells by Covalent Fluorescent Labeling of Fusion Proteins. In: Koga, H. (eds) Reverse Chemical Genetics. Methods in Molecular Biology™, vol 577. Humana Press, Totowa, NJ. https://doi.org/10.1007/978-1-60761-232-2_10

Download citation

DOI : https://doi.org/10.1007/978-1-60761-232-2_10

Published : 03 August 2009

Publisher Name : Humana Press, Totowa, NJ

Print ISBN : 978-1-60761-231-5

Online ISBN : 978-1-60761-232-2

eBook Packages : Springer Protocols

• Publish with us

Policies and ethics

• Find a journal
• Track your research

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

Degradation Parameters from Pulse-Chase Experiments

Affiliation Department of Theory and Bio-Systems, Max Planck Institute of Colloids and Interfaces, D-14424 Potsdam, Germany

* E-mail: [email protected]

• Celine Sin, 
• Davide Chiarugi, 
• Angelo Valleriani

• Published: May 16, 2016
• https://doi.org/10.1371/journal.pone.0155028
• Reader Comments

Pulse-chase experiments are often used to study the degradation of macromolecules such as proteins or mRNA. Considerations for the choice of pulse length include the toxicity of the pulse to the cell and maximization of labeling. In the general case of non-exponential decay, varying the length of the pulse results in decay patterns that look different. Analysis of these patterns without consideration to pulse length would yield incorrect degradation parameters. Here we propose a method that constructively includes pulse length in the analysis of decay patterns and extracts the parameters of the underlying degradation process. We also show how to extract decay parameters reliably from measurements taken during the pulse phase.

Citation: Sin C, Chiarugi D, Valleriani A (2016) Degradation Parameters from Pulse-Chase Experiments. PLoS ONE 11(5): e0155028. https://doi.org/10.1371/journal.pone.0155028

Editor: Marie-Joelle Virolle, University Paris South, FRANCE

Received: December 20, 2015; Accepted: April 22, 2016; Published: May 16, 2016

Copyright: © 2016 Sin et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: The data analysed in this paper is either generated by means of computer programs available to anyone under request or extracted from a published plot in Ref [ 9 ]. The data extracted from Ref [ 9 ] is available in the supplementary file to the manuscript.

Funding: CS was financially supported by the FP7 Marie-Curie Initial Training Networks program “Network for Integrated Cellular Homeostasis (NICHE)” grant number PITN-GA-2011-289384 (to AV). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

The degradation of macromolecules such as mRNA and proteins is a complex biochemical process that is usually carried out by a series of subsequent biochemical reactions. Various experimental techniques can be used to study the decay of macromolecules over time; one of the most widespread methods to measure decay is pulse-chase experiments. A pulse-chase experiment consists of two phases: first, in the pulse phase, cells are exposed to a labeling compound that is integrated into newly synthesized macromolecules of interest. For example, if one wants to follow the fate of proteins or mRNA radio or isotope labeled amino acids [ 1 – 5 ] or nucleotides [ 6 , 7 ] can be introduced in the culture medium. In the second phase, the chase phase, the same compound in the unlabeled form is added in excess, replacing the labeled form. Any macromolecules synthesized during the chase phase will not be labeled. Throughout the chase, the amount of macromolecules that were synthesized during the pulse diminish due to degradation processes ( Fig 1a ). Tracking their amount over time delivers a curve that we call the decay pattern . Alternative methods, such as stopping synthesis of the macromolecule and measuring the amount left over time, also provide decay patterns. For example, chloramphenicol is often used to stop protein synthesis in bacterial cells. However, such methods may be very stressful for the cell, potentially resulting in measurements not indicative of normal cell conditions. Ultimately, the process of stopping synthesis corresponds in general to a pulse of infinite duration. From the point of view of data analysis, decay patterns obtained upon stopping the synthesis are a limit case of patterns generated with pulse-chase experiments.

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

(a) Depiction of molecules in a cell during a pulse chase experiment with pulse duration of 70 in arbitrary units (a.u.). For the purpose of illustration we show four snapshots of the experiment. In snapshot I (30 a.u.), pulse has just begun. The white dots depict the population of molecules already existing in the cell before the pulse. All newly synthesized molecules (red dots) are labeled by the pulse and measurable by the experimentalist. As the pulse continues in snapshot II (60 a.u.) we see more labeled molecules appear. Meanwhile, both labeled and unlabeled molecules degrade. In snapshot III (90 a.u.), the pulse has ended since some time. Newly synthesized molecules from this moment on are unlabeled. Again, both labeled and unlabeled molecules degrade. In snapshot IV (200 a.u.), all labeled molecules have degraded. Unlabeled molecules continue to be synthesized and degraded. (b) Age distribution of the labeled molecules, each curve corresponds to one phase in panel (a). In snapshot I, the pulse has just begun, and all molecules that are labeled by the pulse are no older than the time elapsed since the pulse has begun. In snapshot II, the pulse has been applied for some time; some labeled molecules may be quite old. In snapshot III the molecules cannot be younger than the time elapsed since pulse has been stopped. By snapshot IV, if there were molecules left, they would have that age distribution.

https://doi.org/10.1371/journal.pone.0155028.g001

Decay patterns are said to be exponential, if they look like straight lines in a plot linear in time and logarithmic in the (relative) abundance of the molecule. This makes fitting a single exponential to the decay pattern especially convenient—so much so that it is often used as the default procedure for analysis of decay curves. Unfortunately this procedure is often overused; one can tell when a quick look at the plot clearly indicates that the pattern is not as straight as it ought to be. As a way out of this cumbersome situation, many research studies use the half-time as a measure of the stability of the molecule because it can be estimated even without any fit of the data. For exponentially decaying molecules, there is a simple algebraic relationship between half-time and average lifetime of the molecules, but this relationship does not hold for non-exponentially decaying molecules. Thus, half-time is not a good measure of stability for non-exponential decay [ 8 ]. A further complication is that apparently, for some decay patterns, the pulse length affects the half-time [ 9 ]. This happens because the half-time is measured by simply extracting it from the data without taking pulse length into account. Thus, the half-time of the resultant population is actually not a quantity representative of the molecule’s decay alone, but also of the pulse length. The choice of the function used for the analysis of decay curves makes critical assumptions about the biochemistry of the degradation process at the single molecule level. As we shall see, the choice of the exponential function may lead to achieve incorrect information about the underlying biochemical processes and the stability of the macromolecules.

Another common approach to analyze non-exponential (sometimes called nonlinear) decay patterns is to fit them with a double exponential. Although this certainly improves the fit, it is not yet immediately obvious how the two characteristic timescales from the two exponentials depend on the length of the pulse or on other possible choices of the experimental design.

One fundamental question is: What is the process behind the decay pattern? We know that behind the decay pattern there is a degradation process. When we think about the degradation process, we think at the level of single molecules. Namely, we think at the processes (biochemical networks, perhaps competing against each other) that will eventually lead to the disappearance of the molecule. Taking into account that we are talking about biochemical interactions, the fate of each single molecule can be described quantitatively only in probabilistic terms, with the basic quantity of this description being the probability distribution of the molecule’s lifetime. The decay pattern, however, is not a single molecule measurement. Rather, it is the result of billions of molecules, each randomly decaying during and after the pulse.

When pulse-chase experiments are used to follow the macromolecules degradation, a bias may be introduced into the measurements through the pulse length. In this paper we will show that the length of the pulse affects the shape of the decay patterns. If this effect is not taken into consideration, the biochemical information we extract from the decay pattern would be biased. Unfortunately, this effect is usually not considered when choosing the pulse length. The connection between the single molecule perspective and the decay pattern requires an abstract view of the degradation process. Here we present an approach that allows us to correctly bridge the gap between the single molecule perspective and the population level for a wide range of decay patterns. Notably, our method provides a simple recipe that allows us to incorporate the pulse length in the fitting procedure when decay patterns are derived from pulse-chase experiments. The method does not rely on any a priori knowledge of the lifetime of the molecule and can be used independently of whether the pulse length is short or long compared to the lifetime of the molecules. First, we recapitulate the general theory providing a new derivation of the required mathematics and then we provide a simple recipe to incorporate the pulse length in the fitting procedure. Our results should be applied to refine the protocols for the design of decay assay experiments.

Aging of molecules

It is rather strange that an apparently simple concept like aging becomes so intriguing when applied to molecules. Molecules, at least complex molecules like proteins and mRNA, age in a fashion which is very similar to what we know based on our daily experience. Damage, shortening or lengthening, the attachment of other molecules (e.g. miRISC complex that attaches to the mRNA [ 10 , 11 ]) are common phenomena that make the target molecule older. In fact, some processes could even make the target molecule younger, e.g. when damage is repaired or a bound molecule detaches. What does it actually mean when a molecule becomes older, or ages? How can we characterize this phenomenon? Acutally, what best characterizes the effect of aging is not how much time has passed since birth or synthesis but the amount of time left until the end of life.

Let us look at this matter closely. Suppose for a moment that the lifetime distribution of an hypothetical molecule is exponential—as we shall see later, this implies that also the decay pattern is exponential—and we know that in this precise moment the age of the molecule is a , i.e. this molecule has been synthesized a time units ago. What is the distribution of its residual lifetime? As a consequence of the assumption of an exponential lifetime distribution, the answer is that the residual lifetime is independent of a and is the same as if the molecule had been synthesized right now. This answer means that the molecule does not age. However, if we know or suspect that molecules like mRNA and proteins must undergo a series of biochemical reactions [ 12 – 14 ] until we consider them as degraded, each step in the biochemical reaction network makes the molecule older, i.e. closer to its final end. When the molecule ages in the “complex” way just described, its residual lifetime becomes shorter. Therefore, its lifetime distribution cannot be exponential and so also the decay pattern cannot be an exponential.

Once we accept that molecules age— i.e. do not have an exponential lifetime distribution—there is another effect of aging related to the duration of the pulse. Imagine a pulse of a very short duration. The molecules synthesized during the pulse will all have quite the same residual lifetime, since they are likely to be all in exactly the same biochemical state. When the pulse becomes longer, some of the molecules synthesized at the beginning of the pulse may have been already degraded, some are definitively older but still there, and others will be just newly synthesized. In short, we have a mixture of ages in the population, and the composition of the population depends on the pulse length ( Fig 1b ). This is an important point for what follows: the initial condition at the time point of chase depends on the mixture of ages in the population of molecules, which depends on the length of the pulse. Therefore, fitting the decay pattern requires a knowledge of the age distribution at the beginning of the chase, which can be computed only if one knows the effect of the pulse on the age distribution. The effect of the pulse on the age distribution can only be known if one has a good model to describe how the molecules age, and has already calibrated the model. A priori, it is not possible to know if a pulse of a given length is long or short compared to the lifetime of the molecules, but as we shall see, this apparent circularity can be solved to provide a unique formula that takes pulse, chase and aging into account.

Degradation processes modeled with Markov chains

Macromolecules are degraded through a number of different pathways. While some can be approximated as exponentially decaying, many of these pathways are multi-step pathways with several complex biochemical stages [ 1 , 12 – 15 ]. The biochemical stages are connected to each other to form a network of biochemical states [ 8 , 16 – 18 ]. When we describe the dynamics of degradation of a single molecule, we think of this molecule as moving on this network of biochemical states in a stochastic fashion until degradation eventually happens. We will use the term “single step degradation” for pathways with only one measurable rate limiting step ( i.e. exponentially decaying), and the term “multi-step degradation” for more complex pathways, which necessarily includes the “single step” degradation as a limiting case [ 8 ]. The rates of the transitions between the biochemical states would depend on the concentration of the ligands and on their affinity to the target molecule, details rarely available on a large scale. It is therefore convenient to model this process as a Markov chain, which is a simple and mathematically treatable tool to describe stochastic process on discrete states [ 8 , 16 , 17 , 19 – 21 ]. From this we can model the decay as a single molecule stochastic process, and then solve the equations to derive the lifetime distribution, the age-dependent degradation rate, and the steady state distribution of the fraction of molecules in each state.

Decay after a pulse

Data analysis

Model calibration (parameter estimation)

We use several functions in MATLAB ® to calibrate the models. To minimize the objective function, we use fmincon (bounds: κ 10 , κ 12 , κ 20 ∈ [0.000001,1]). We also used GlobalSearch (with the default settings) and MultiStart (1000 start points) to better sample the available parameter space.

Results and Discussion

Most mRNA and protein decay patterns can be fit with one of two simple models: A two-stage model ( Fig 2a ) always results in a better fit than a one-stage (exponential decay) model, but many patterns are sufficiently described by the simpler one-stage model. The decision of whether a two-stage model or the exponential fit should be adopted depends on the balance between accuracy of the fit and number of parameters. The Akaike Information Criterion (AIC) is one such measure that can be used to select the better model [ 23 ]. In this paper, we concentrate on the use of the two-stage model. The motivation for this is two fold—firstly, for decay patterns from one-stage models (exponential decay), one does not need to take the pulse length into account. The decay curve of the system with exponential decay is independent of the pulse. Secondly, in the examples that we will later present, the one-stage model does not adequately capture the dynamics of the system ( Fig 2b ). Through the AIC, we find that one-stage models are too simplistic, and three stage models are not adequately identifiable. For other systems with degradation patterns not well described by a two-stage Markov state model, Eq (10) can be used to derive the appropriate expression. The two-stage model applied to the data of Ref. [ 9 ] has an RSS of 0.0011 ( Fig 2b ). To evaluate the goodness of fit, we apply the chi-squared test with three d.o.f. and find that we can not reject the null hypothesis which is that the two-stage model is a good fit (p-value < 10 − 5 ).

(a) Markov chain representation of a degradation process. Biochemical pathways (such as degradation) can be readily translated into Markov chain models: each biochemical entity is represented as a Markov state (circles) and the reaction speeds are represented as fluxes between the states (arrows). Here we show a possible Markov model of degradation containing two states. Newly synthesized molecules are in state 1. From state 1, there are two possible paths, either to state 2 with rate κ 12 or degraded with rate κ 10 . For those molecules that reach state 2, they are degraded with rate κ 20 . The rates κ 10 and κ 20 are necessarily different, otherwise the model collapses into a 1 state model. (b) 2-state model fit (black line) and exponential fit (red line) to sample data with 1 minute pulse (data from [ 9 ], blue spots). Note that in the log(abundance)-linear(time) scale, the data does not resemble a straight line, thus necessitating a model more complicated than a single exponential. The best fit using Eq (11) with C (Δ t ) from Eq (16) gives the following parameters: κ 10 = 0.0109 min − 1 , κ 20 = 0.002 min − 1 , κ 12 = 0.0189 min − 1 , and pulse = 1 min. κ exp = 0.0029. The decision in favor of the two-stage model is made on the basis of the AIC criterion thanks to its very small RSS.

https://doi.org/10.1371/journal.pone.0155028.g002

The general mathematical methodology to describe the decay (see Methods ) requires a specific form of the lifetime distribution in order to become practically useful. Ideas from biochemical networks indicate that a Markov chain is a useful and flexible framework to develop the basic models of molecular degradation [ 8 , 21 ]. In terms of a Markov chain the two-stage model consists of two states, say state 1 and state 2, and a degraded state called state 0 connected to each other ( Fig 2a ). The rates κ 12 , κ 10 , and κ 20 govern the transitions from state 1 to state 2, from state 1 to state 0, and from state 2 to state 0, respectively. In addition, the rates κ 10 , and κ 20 are composed of the degradation rates from each state and the basal dilution rate due to cell division and population growth; the basal dilution rate can have a negligible effect if the timescale of the experiment is shorter than the doubling time of the cell culture.

Using this network we model the life of a single molecule as follows. The molecule is synthesized to be in biochemical state 1 (or it moves very quickly through a series of biochemical steps until it reaches a biochemical state that we call state 1). The molecule dwells for a random amount of time in state 1. The average amount of time spent in state 1 is τ 1 = 1/( κ 10 + κ 12 ). The molecule then leaves state 1 and is either degraded, i.e. it jumps to state 0 with probability P 10 = τ 1 κ 10 , or it jumps into biochemical state 2 with probability P 12 = τ 1 κ 12 . If the molecule attains state 2, it will dwell in this state for a random amount of time with average τ 2 = 1/ κ 20 until it is eventually degraded ( i.e. P 20 = 1). The two-stage model does not mean that there are only two biochemical states for a molecule, it says that all other states are visited very quickly and that, at the end, two compound states are sufficient to describe the decay pattern for that molecule.

The lifetime of the molecule is now just the random time required to move from state 1 to state 0, taking into account the possibility to visit state 2 before degradation. Technically, this is the absorption time in 0 starting from state 1 and its probability density is the lifetime probability density f T ( t ). The equivalence of lifetime and absorption time is useful because there are plenty of mathematical tools to compute the probability density of the time to absorption. Once the probability density f T ( t ) is computed in terms of the (yet unknown) rates κ 10 , κ 12 and κ 20 , and of the pulse length t p , it can be used to compute the relative abundance C (Δ t ) ( Eq 10 ). Fitting the data will then finally deliver the appropriate values of the rates ( Fig 2b ).

(a) Verification of fitting procedure using simulated data separately. Using the parameters obtained from the best fit model to the data from [ 9 ] for pulse = 1 min, we fabricate sample data by calculating the abundance over time (dots) for different pulse lengths (1, 5, 30, 120, 1200 minutes) using the function that gives the decay pattern of the relative abundance C (Δ t ), Eq (16) , as function of the measurement time Δ t . We then fit resultant decay patterns with our fitting routine. We get back the same rates that were used to simulated the data for each experiment ( Table 1 ). This shows that if the system in the background is unchanging, we can reliably extract the parameters of the system by fitting the decay patterns individually. κ 10 = 0.0109 min − 1 , κ 20 = 0.002 min − 1 , and κ 12 = 0.0189 min − 1 . (b) Simultaneous fit of pooled simulated data. Here the simulated data is augmented with a small amount of multiplicative noise, Eq (17) . We fit the whole collection of data simultaneously (see Methods ). Values very close to our original simulation parameters are obtained ( Table 1 last row). This shows that under steady experimental conditions, we can reliably extract the parameters of the system by fitting the decay patterns simultaneously.

https://doi.org/10.1371/journal.pone.0155028.g003

Parameters from the best fit to the data simulated with different pulse lengths and the 2-state model as found by Multistart (MATLAB ® ) with 1000 start points. Minimization by fmincon (bounds κ 10 , κ 20 , κ 12 ∈ [0.000001,1]). Notice that the fits yield results identical to those used to generate the data. This proves the self-consistency of the procedure. Conversely in Table 2 , the parameter values are not stable; suggesting different degradation system dynamics in each experiment.

https://doi.org/10.1371/journal.pone.0155028.t001

As a next step, we have fit the individual experimental decay patterns as provided in Ref. [ 9 ] ( Fig 4a ) and compared the rates with each other ( Table 2 ). We find that the rates depart from each other much more than those found with the fabricated data. With the proviso that we have extracted the data from an old low-definition figure in semilogarithmic scale, the differences in the rates definitively increase with the duration of the pulse. Also the search for a unique set of rates that allow fitting the various pulsed curves all together gives a poor result ( Fig 4b ). The most likely explanation of the wrong fit is that the labeling has affected the cells and has thus contributed to change its internal environment so that protein degradation after a long pulse is different than protein degradation after a short pulse, an effect that may certainly depend also on the labeling technique.

(a) Decay patterns from Ref. [ 9 ] fit individually. Each decay pattern from Ref. [ 9 ] is fit with the 2-state model using Eq (11) with C (Δ t ) from Eq (16) . We find that for some of the decay patterns, the parameters obtained from the fitting are different from the others ( Table 2 ). This implies that the underlying system has changed in the different experiments. Possibly the labeling procedure has affected the cells and contributed to a change in the internal environment. (b) Simultaneous fit of pooled real data with Eq (12) . Here we pool the decay patterns and fit them simultaneously with the 2-state model. No good fits were found, despite using global and multistart techniques in the parameter search process. This implies that the underlying systems across the experiments can not be described by one unified model, at least not the two state model that we have considered. Possibly the labeling procedure has affected the cells and contributed to a change in the internal environment.

https://doi.org/10.1371/journal.pone.0155028.g004

Parameters from the best fit to the 2-state model as found by Multistart (MATLAB ® ) with 1000 start points. Minimization by fmincon (bounds κ 10 , κ 20 , κ 12 ∈ [0.000001,1] min − 1 ). Notice that each fit yields different parameters. This suggests that the degradation system dynamics in each experiment is different. In contrast Table 1 shows that the fits to simulated data with consistent parameters return the same rates. Data are reported in Table A in S1 Supporting Information

https://doi.org/10.1371/journal.pone.0155028.t002

Just pulse, no chase

If the steady state measurement of molecules is not available, any other time point can also be used. With other time points used as normalization, the expression for P is more complicated, but still solvable by hand for the two-stage model.

Parameters from the best fit to the 2-state model as found by Multistart (MATLAB ® ) with 100 start points. Minimization by fmincon (bounds κ 10 , κ 20 , κ 12 ∈ [0.000001,1] min − 1 ). The first row is the best fit for the data as if the experimentalist only took 3 measurements at t = 1,2,3 minutes. The 2nd row takes one more measurement ( t = 10 minutes) into consideration.

https://doi.org/10.1371/journal.pone.0155028.t003

Simulation of pulse no chase experiments, where the pulse is applied up until the time of measurement. Data is produced with rates κ 10 = 0.0109 min −1 , κ 20 = 0.0002 min −1 and κ 12 = 0.0189 min −1 using Eq (19) . Traces show the results of several fits, each fitting taking into consideration one additional data point ( Table 3 ).

https://doi.org/10.1371/journal.pone.0155028.g005

We discover that the output at late time points are insensitive to the parameters. Furthermore, the output is only sensitive to small values of κ 10 and κ 12 ( Fig 6 ). Note that here we have calculated the local sensitivities—the result is likely to differ for different ranges of κ ’s.

We plot the sensitivities of the measurements in a pulse no chase experiment assuming a 2-state model with κ 10 = 0.0109 min − 1 , κ 20 = 0.0002 min − 1 and κ 12 = 0.0189 min − 1 . (a) Output is only sensitive to small values of κ 10 . (b) Output is sensitive to a range of κ 20 . (c) Output is only sensitive to small values of κ 12 . For all parameters, taking more measurements at later timepoints does not help the parameter estimation because the output is not sensitive to deviations in the parameters at late times.

https://doi.org/10.1371/journal.pone.0155028.g006

Dynamical properties

Even if steady state expression levels of molecules depend only on their average lifetime, non-exponential decay has an effect concerning timing and dynamics of cellular response [ 18 , 19 , 26 ], and cell-to-cell variation in cellular content when cultures are subject to stochastic effects [ 27 ]. For this reason, it is an aspect of regulation that has to be taken into account if one wants to understand the reaction of cells to stress or to environmental changes. Furthermore, disentangling various hypotheses concerning the nature and the structure of biochemical pathways responsible for degradation [ 10 , 11 , 21 ] is possible only when models take the complexity of the pathways into account. Nevertheless, the derivation of the average lifetime required to compute the steady state properties (and the synthesis rate, if an independent measurement of the steady state abundance is available) requires the correct mathematics, which in most of the cases is not given by the exponential fit.

In this manuscript we focused on the two-stage model as a good approximation that describes most of the non-exponential decay patterns. This generalization is based on extensive experience from fitting thousands of mRNA and protein decay patterns. Nevertheless, there is no reason in principle to restrict the number of states to two, since the number of biochemical steps related to the destabilization of molecules is conceivably much larger. While it is not difficult to draw and mathematically describe larger networks with more states and alternative pathways, there is rarely enough data available to fix all the parameters [ 23 ]. A lucky exception is provided by sets of experiments where measurements are taken with different parts of the biochemical network deactivated [ 10 , 21 ].

Pulse-chase experimental techniques offer the advantage of a low impact on the metabolism and well-being of the cells. Yet, little attention has been devoted to the fact that the pulse has an effect on the decay pattern if the decay pattern is not exponential. By working through the mathematics of single molecule decay, we derive and demonstrate in a novel approach a series of equations that anyone can use to fit complex decay patterns. The values of the rates would then finally provide a valuable tool to compute other dynamic quantities.

When assessing the nature of the decay pattern we recommend to compare the fit of the two-stage model with the simpler exponential model to ensure that the data supports the more complex model. As a guideline, long pulses (and thus steady state measurements) tend to obfuscate short term dynamics and thus appear as exponential decay, as this catches the long term dynamics of the degradation process. We recommend the use of the shortest pulse as possible in order to detect short time effects. More pulses of different lengths increase the robustness and our confidence in the model and its calibration. As a byproduct, fitting curves generated from pulses of different lengths may allow discovery of perturbing effects from labeling, when the fits of the individual curves reveal very different rates beyond what one would expect from noise. One caveat to the current approach outlined here is the assumption that the quantities measured are reflective of synthesis and degradation only. However, in in vivo systems, each cell division results in a dilution effect of the molecules. Thus for such experiments it is imperative to choose experimental times well within the cell division time, or take cell division into account.

Supporting Information

S1 supporting information. supplementary note..

The file contains some supplementary calculations and the table with the experimental data extracted from Ref. [ 9 ].

https://doi.org/10.1371/journal.pone.0155028.s001

Acknowledgments

CS and AV thank M. Selbach and E. McShane for useful discussions.

Author Contributions

Conceived and designed the experiments: AV CS DC. Analyzed the data: CS AV DC. Wrote the paper: AV CS DC. Developed the in-house software used in analysis: CS AV. Mathematical derivations: AV CS. Sensitivity analysis: CS.

• View Article
• PubMed/NCBI
• Google Scholar
• 22. Taylor HM, Karlin S. An Introduction to Stochastic Modeling, Third Edition. Academic Press; 1998.
• 23. Burnham KP, Anderson DR. Model selection and multimodel inference: a practical information-theoretic approach. Springer Science & Business Media; 2003.