Resin-Assisted Capture Coupled with Isobaric Tandem Mass Tag Labeling for Multiplexed Quantification of Protein Thiol Oxidation

Summary

Protein thiol oxidation has significant implications under normal physiological and pathophysiological conditions. We describe the details of a quantitative redox proteomics method, which utilizes resin-assisted capture, isobaric labeling, and mass spectrometry, enabling site-specific identification and quantification of reversibly oxidized cysteine residues of proteins.

Abstract

Reversible oxidative modifications on protein thiols have recently emerged as important mediators of cellular function. Herein we describe the detailed procedure of a quantitative redox proteomics method that utilizes resin-assisted capture (RAC) in combination with tandem mass tag (TMT) isobaric labeling and liquid chromatography-tandem mass spectrometry (LC-MS/MS) to allow multiplexed stochiometric quantification of oxidized protein thiols at the proteome level. The site-specific quantitative information on oxidized cysteine residues provides additional insight into the functional impacts of such modifications.

The workflow is adaptable across many sample types, including cultured cells (e.g., mammalian, prokaryotic) and whole tissues (e.g., heart, lung, muscle), which are initially lysed/homogenized and with free thiols being alkylated to prevent artificial oxidation. The oxidized protein thiols are then reduced and captured by a thiol-affinity resin, which streamlines and simplifies the workflow steps by allowing the proceeding digestion, labeling, and washing procedures to be performed without additional transfer of proteins/peptides. Finally, the labeled peptides are eluted and analyzed by LC-MS/MS to reveal comprehensive stoichiometric changes related to thiol oxidation across the entire proteome. This method greatly improves the understanding of the role of redox-dependent regulation under physiological and pathophysiological states related to protein thiol oxidation.

Introduction

Under homeostatic conditions, cells generate reactive oxygen, nitrogen, or sulfur species that help to facilitate processes, such as metabolism and signaling^1,2,3, extending to both prokaryotes and eukaryotes. Physiological levels of these reactive species are necessary for proper cellular function, also known as 'eustress'^1,4. In contrast, an increase in oxidants that leads to an imbalance between oxidants and antioxidants can cause oxidative stress, or 'distress'¹, which leads to cellular damage. Oxidants transduce signals to biological pathways by modifying different biomolecules, including protein, DNA, RNA, and lipids. In particular, cysteine residues of proteins are highly reactive sites prone to oxidation due to the thiol group on cysteine, which is reactive towards different types of oxidants⁵. This gives rise to a diverse range of reversible redox-based posttranslational modifications (PTMs) for cysteine, including nitrosylation (SNO), glutathionylation (SSG), sulfenylation (SOH), persulfidation (SSH), polysulfidation (SS_nH), acylation, and disulfides. Irreversible forms of cysteine oxidation include sulfinylation (SO₂H) and sulfonylation (SO₃H).

Reversible oxidative modifications of cysteine residues may serve protective roles preventing further irreversible oxidation or serve as signaling molecules for downstream cellular pathways^6,7. The reversibility of some thiol redox PTMs allows cysteine sites to function as "redox switches"^8,9, wherein changes in the redox state of these sites alter protein function to regulate their role in transient processes. The modulatory effects of redox PTMs¹⁰ have been observed in many aspects of protein function¹¹, including catalysis¹², protein-protein interactions¹³, conformation change¹⁴, metal ion coordination¹⁵, or pharmacological inhibitor binding¹⁶. Additionally, redox PTMs are involved in cysteine sites of proteins that regulate pathways such as transcription¹⁷, translation¹⁸, or metabolism¹⁹. Given the impact that redox PTMs have on protein function and biological processes, it is important to quantify the extent of oxidation that a cysteine site undergoes in response to a perturbation of the redox state.

The identification of cysteine sites with altered redox states is focused on the comparison of the oxidation state at the site-specific level between normal and perturbed conditions. Fold change measurements are often utilized to determine what sites are significantly altered as this helps users interpret what cysteine sites may be physiologically significant to the study. Alternatively, stoichiometric measurements of reversible thiol oxidation across a specific sample type give a general picture of the physiological state with respect to cellular oxidation, an important measurement that is often overlooked and underutilized. Modification stoichiometry is based on quantifying the percentage of modified thiol as a ratio to total protein thiol (modified and unmodified)^20,21. As a result, stoichiometric measurements offer a more precise measurement than fold change, especially when using mass spectrometry. The significance of the increase in oxidation can be more readily ascertained by using stoichiometry to determine the PTM occupancy of a particular cysteine site. For example, a 3-fold increase in thiol oxidation could result from a transition of as little as 1% to 3% or as big as 30% to 90%. A 3-fold increase in oxidation for a site that is only at 1% occupancy may have little impact on a protein's function; however, a 3-fold increase for a site with 30% occupancy at resting state may be more substantially affected. Stoichiometric measurements, when performed between total oxidized thiols and specific oxidative modifications, including protein glutathionylation (SSG) and nitrosylation (SNO), can reveal ratios and quantitative information with respect to specific modification types.

Because reversible thiol oxidation is typically a low-abundance posttranslational modification, multiple approaches have been developed for the enrichment of proteins containing these modifications out of biological samples. An early approach devised by Jaffrey and others, named the biotin switch technique (BST)²², involves multiple steps wherein unmodified thiols are blocked through alkylation, reversibly modified thiols are reduced to nascent free thiols, nascent free thiols are labeled with biotin, and the labeled proteins are enriched by streptavidin affinity pulldown. This technique has been used to profile SNO and SSG in many studies and can be adapted to probe for other forms of reversible thiol oxidation^23,24. While BST has been utilized to probe for different forms of reversible thiol oxidation, one concern with this approach is that enrichment is impacted by the non-specific binding of unbiotinylated proteins to streptavidin. An alternate approach developed in our laboratory, named resin-assisted capture (RAC)^25,26 (Figure 1), circumvents the issue of enrichment of thiol groups via the biotin-streptavidin system.

Following the reduction of reversibly oxidized thiols, proteins with nascent free thiols are enriched by the thiol-affinity resin, which covalently captures free thiol groups, allowing for more specific enrichment of cysteine-containing proteins than BST. Coupling RAC with the multiplexing power of the recent advances in isobaric labeling and mass spectrometry creates a robust and sensitive workflow for the enrichment, identification, and quantification of reversibly oxidized cysteine residues at the proteome-wide level. Recent advances in mass spectrometry have enabled much deeper profiling of the thiol redox proteome, increasing the understanding of both the cause and effect of protein thiol oxidation²⁷. The information gained from site-specific quantitative data allows for further studies of the mechanistic impacts and downstream effects of reversible oxidative modifications²⁸. Utilizing this workflow has provided insight into the physiological impacts of reversible cysteine oxidation with respect to normal physiological events such as aging, wherein levels of SSG differed with respect to age. The aging effects on SSG were partially reversed using SS-31 (elamipretide), a novel peptide that enhances mitochondrial function and reduces SSG levels in aged mice, causing them to have an SSG profile more similar to young mice²⁹.

Pathophysiological conditions attributed to nanoparticle exposure have been shown to involve SSG in a mouse macrophage model. Using RAC coupled with mass spectrometry, the authors showed that SSG levels were directly correlated to the degree of oxidative stress and impairment of macrophage phagocytic function. The data also revealed pathway-specific differences in response to different engineered nanomaterials that induce different degrees of oxidative stress³⁰. The method has also proven its utility in prokaryotic species, where it was applied to study the effects of diurnal cycles in photosynthetic cyanobacteria with respect to thiol oxidation. Broad changes in thiol oxidation across several key biological processes were observed, including electron transport, carbon fixation, and glycolysis. Furthermore, through orthogonal validation, several key functional sites were confirmed to be modified, suggesting regulatory roles of these oxidative modifications⁶.

Herein, we describe the details of a standardized workflow (Figure 1), demonstrating the utility of the RAC approach for the enrichment of total oxidized cysteine thiols of proteins and their subsequent labeling and stoichiometric quantification. This workflow has been implemented in studies of the redox state in different sample types, including cell cultures^27,30 and whole tissues (e.g., skeletal muscle, heart, lung)^29,31,32,33. While not included here, the RAC protocol is also easily adapted for the investigation of specific forms of reversible redox modifications, including SSG, SNO, and S-acylation, as previously mentioned^25,29,34.

Protocol

All procedures described in the protocol related to animal or human samples/tissues were approved by and followed the institutional guidelines of the human and animal research ethics committee.

1. Sample homogenization/lysis

1. Frozen tissue samples
   1. Mince frozen tissue (~30 mg) on a glass microscope slide on dry ice using a prechilled razor blade and forceps. Transfer the minced tissue to a prechilled 5 mL round-bottom polystyrene tube containing 700 µL of buffer A (see Table 1) and incubate on ice for 30 min, protected from light.
   2. Disrupt the tissue for 30 s or until completely homogenized with a hand-held tissue homogenizer. Place the samples on ice and allow the foam to subside for another 10 min.
      NOTE: An aluminum baking sheet placed on dry ice provides a stable working surface and platform for the initial processing/mincing of the tissue.
2. Alternatively, use adherent cell cultures in 100 mm culture dishes as the starting material.
   1. Keep the cells on ice and use a serological pipette to rinse the cells twice with 10 mL of ice-cold PBS containing 100 mM NEM.
   2. Lyse the cells by adding 1 mL of cold homogenization/lysis buffer and scraping vigorously with a rigid cell scraper. Transfer the lysate to a 2 mL centrifuge tube using a micropipette.
      NOTE: Rinsing buffer and lysis buffer may be scaled accordingly to different size culture vessels. Typically, 2-5 million cells are required; however, this varies depending on the lysis efficiency and protein yield for specific cell types. Homogenization buffer may be prepared without NEM for samples being analyzed for total thiols.
3. Transfer the resulting homogenate (step 1.1.2 or 1.2.2) to a 2 mL centrifuge tube using a micropipette, and centrifuge at full speed (≥16,000 × g) at 4 °C for 10 min.
4. Transfer the supernatant (~700 µL or ~1 mL for cell culture) using a micropipette to a 5 mL conical microcentrifuge tube and incubate for 30 min at 55 °C in the dark with shaking at 850 rpm.
5. Using a glass serological pipette, add 4 mL of ice-cold acetone to the samples and incubate at -20 °C overnight for precipitation of protein and removal of excess N-ethylmaleimide.

2. Resin-assisted capture

1. Wash the precipitated protein pellets twice with acetone by centrifuging at 20,500 × g at 4 °C for 10 min, decanting the acetone, removing any remaining acetone using a micropipette, and adding 3 mL of fresh, ice-cold acetone using a glass serological pipette. Invert several times to mix. After the second wash, allow the pellets to air-dry for 1-2 min, being careful not to over-dry as resuspension may become difficult.
2. Using a micropipette, add 1 mL of buffer B (see Table 1) and solubilize the protein using repeated sonication for 15-30 s at a time using a bath sonicator with an output of 250 W and brief vortexing. Measure the protein concentration using the bicinchoninic acid (BCA) assay according to the manufacturer's protocol.
3. To standardize the protein concentrations across samples for further processing and ensure complete removal of NEM, transfer 500 µg of protein to a 0.5 mL 10 kDa centrifugal filter using a micropipette and adjust to a final volume of 500 µL with resuspension buffer.
4. Centrifuge at 14,000 × g at room temperature until the volume in the centrifugal filter is less than 100 µL. Collect the samples by inverting the filter in a collection tube. Centrifuge at 1,000 × g for 2 min and adjust to a final volume of 500 µL using buffer C (see Table 1).
5. Reduce the protein thiols by adding 20 µL of 500 mM dithiothreitol (DTT) using a micropipette to a final concentration of 20 mM and incubating the samples for 30 min at 37 °C while shaking at 850 rpm.
6. After reduction, transfer the samples using a micropipette to 0.5 mL 10 kDa centrifugal filters and centrifuge for 15 min at 14,000 × g at room temperature or until the volume in the centrifugal filter is less than 100 µL. Add buffer D (see Table 1) to make up the volume in the centrifugal filter to 500 µL.
   1. Repeat the centrifuging and addition to 500 µL in step 2.6 three times, and after the fourth centrifugation, collect the samples by inverting the filter in a collection tube and centrifuging at 1,000 × g for 2 min.
7. Measure the protein concentration using the BCA assay according to the manufacturer's protocol.
8. During this buffer exchange, prepare the thiol-affinity resin by weighing the appropriate amount of resin (30 mg/sample) using a microbalance and transferring it to a 50 mL centrifuge tube. Then, using a serological pipette, add water for a final concentration of 30 mg/mL resin and incubate at room temperature for 1 h with agitation for proper hydration of the resin.
   NOTE: The thiol-affinity resin mentioned above has been discontinued by the manufacturer. A possible replacement for this thiol-affinity resin is commercially available. However, this replacement has a nearly 5-fold less binding capacity (see Supplemental Information). Alternatively, the thiol-affinity resin can be synthesized using 2-(pyridyldithio) ethylamine hydrochloride and N-hydroxysuccinimide-activated resin (see Supplemental Information).
   1. After hydration of the resin, place the spin columns on a vacuum manifold and transfer 500 µL of the resin slurry using a micropipette to each column. Apply vacuum for removal of water; repeat this step once to obtain a total of 30 mg of resin per column. Alternatively, centrifuge at 1,000 × g for 2 min instead of using the vacuum manifold for this and all the resin washing and elution steps.
      NOTE: Cutting the end of a 1000 µL pipette tip to increase the bore size helps with the transfer of the resin. It is important to triturate between pipetting to ensure that the resin remains suspended and homogeneous and equal amounts of resin are transferred to each column.
   2. Wash the resin by adding 500 µL of ultrapure water with a micropipette and applying vacuum for removal of the water; repeat this 5 times. Then, wash the resin 5 times with 500 µL of buffer E (see Table 1).
      NOTE: Alternatively, centrifugation at 1,000 x g for 2 min may be used in place of a vacuum manifold for all subsequent wash steps. All the proceeding wash steps are performed with a volume of 500 µL. When adding wash buffers to the column, carefully add with enough force to fully resuspend the resin while avoiding splashing and loss of resin; this allows for complete and efficient washing of resin.
9. Using a micropipette, transfer 150 µg of protein from each reduced sample to a new tube and adjust to a final volume of 120 µL of buffer C (see Table 1). Transfer the protein solution using a micropipette to a plugged spin column containing the resin, place the cap on the column, and incubate for 2 h at room temperature with shaking at 850 rpm.
10. Wash the resin five times with 25 mM HEPES, pH 7.0; 8 M urea; followed by five times with 2 M NaCl; followed by five times with 80% acetonitrile (ACN) with 0.1% trifluoroacetic acid (TFA); and finally five times with 25 mM HEPES, pH 7.7, as described in step 2.8.2 and replace the plug.
    NOTE: Samples may be eluted here for analysis at the protein level (e.g., SDS-polyacrylamide gel electrophoresis (SDS-PAGE), western blot) as described in step 4.1.

3. On-resin tryptic digestion and TMT labeling

1. Prepare enough sequencing-grade modified trypsin solution for 6-8 µg per sample by solubilizing it at a concentration of 0.5 µg/µL in buffer C (see Table 1) so that the final volume allows for at least 120 µL per sample. Using a micropipette, add 120 µL of this trypsin solution to the samples and incubate overnight at 37 °C with shaking at 850 rpm.
   NOTE: To increase the digestion efficiency, an additional digestion step can be included the next day by removing the trypsin solution and replacing it with fresh solution, and continue the digestion for 2 h.
2. Wash the resin five times with 25 mM HEPES, pH 7.0; followed by five times 2 M NaCl; followed by five times with 80% ACN with 0.1% TFA; followed by three times with 25 mM HEPES, pH 7.7. Finally, wash the resin two times with 50 mM triethyl ammonium bicarbonate buffer (TEAB) and replace the plug.
3. Prepare TMT labeling reagents by first allowing them to warm to room temperature before spinning down briefly using a centrifuge at 16,000 × g. Add 150 µL of anhydrous ACN to each vial of TMT labeling reagent using a micropipette. Incubate the vials at room temperature on a thermomixer set to 850 rpm for 5 min to solubilize the reagent completely. Briefly vortex and spin down at 16,000 × g to collect the reagent.
4. Using a micropipette, add 40 µL of 100 mM TEAB to the washed resin, then add 70 µL of the dissolved TMT reagent and incubate for 1 h at room temperature with shaking at 850 rpm. Store any remaining TMT reagent at -80 °C.
   NOTE: Take note of the individual TMT labels assigned to each biological sample (Figure 1).
5. Wash the resin five times with 80% ACN with 0.1% TFA, three times with 100 mM ammonium bicarbonate buffer (ABC), pH 8.0, and twice with water as previously described and replace the plug.

4. Peptide elution

1. Elute the labeled peptides by adding 100 µL of 20 mM DTT in 100 mM ABC, pH 8.0, to each column using a micropipette and incubate at room temperature for 30 min on a thermomixer set to 850 rpm.
   NOTE: After the addition of DTT, the resin will clump. The resin can be disrupted with a pipette tip to break up clumps and ensure the complete elution of the peptides.
2. After this incubation, place the column on a vacuum manifold intended for solid-phase extraction (SPE), apply vacuum, and elute the samples into a 5 mL microcentrifuge tube. Repeat this step once.
3. Finally, add 100 µL of 80% ACN with 0.1% TFA, incubate for 10 min at room temperature, and elute into the same 5 mL centrifuge tube. Collect all the fractions in the same 5 mL microcentrifuge tube.
   NOTE: To prevent sample loss, low binding tubes must be used for elution, and volumes must be kept at or below a volume of 4.0 mL for a single 5 mL tube.
4. Place the eluted samples in a vacuum concentrator until dry. Store the dry peptides at -80 °C and resuspend them later.
   ​NOTE: Samples may also be eluted separately, and an aliquot may be removed and analyzed by SDS-PAGE for analysis at the peptide level before combining the samples.

5. Peptide alkylation and desalting/clean-up

1. Resuspend the dried peptides by adding a small volume of 100 mM ABC buffer, pH 8.0 (no greater than 500 µL), using a micropipette. Use repeated sonication for 15-30 s at a time using a bath sonicator with an output of 250 W and vortex to solubilize and transfer to a 2 mL tube.
   NOTE: The volume of 100 mM ABC, pH 8.0 to be added is based on the volume needed to resuspend the DTT at a molarity of 150 mM. Users will need to determine the amount of DTT present in their sample based on what was added originally in step 4.1.
2. Add enough concentrated stock solution (600 mM) of iodoacetamide (IAA) dissolved in ABC using a micropipette to achieve a 1:4 molar ratio of DTT:IAA and incubate the samples at RT for 1 h with shaking at 850 rpm.
3. Acidify the samples to pH < 3 by adding concentrated TFA (10%) using a micropipette and perform sample desalting using reverse-phase clean-up according to the manufacturer's instructions.
4. Place the clean peptides in a vacuum concentrator until dry. Store the dry peptides at -80 °C until further analysis.

6. Liquid chromatography-tandem mass spectrometry

1. Resuspend the dried peptides by repeated sonication for 15-30 s at a time using a bath sonicator with an output of 250 W and vortexing in 20-40 µL of water containing 3% ACN. Determine the peptide concentration by performing a BCA assay according to the manufacturer's protocol.
2. Separate the samples by reversed-phase LC and MS/MS as previously described⁶ and record the MS1 spectra over the m/z range of 400-2000. Ensure high-energy collisional dissociation (HCD) is utilized to obtain reporter ion intensity information for the analysis of isobarically labeled peptides. See the methods sections of previous reports for more details about instrument run conditions^27,30 and the analysis of MS data^27,31.
   NOTE: Different LC-MS/MS systems or settings can be used to analyze the peptide samples. The coverage and sensitivity of peptide identification will depend on the particular system and settings used.

Representative Results

Completion of the protocol will result in highly specific enrichment of formerly oxidized cysteine-containing peptides, often with >95% specificity^27,35,36. However, several key steps of the protocol require special attention, e.g., the initial blocking of free thiols prior to sample lysis/homogenization, which prohibits artificial oxidation and non-specific enrichment of artificially oxidized thiols²⁵. Samples may be analyzed at several stages of the protocol and by different methods, including SDS-PAGE analysis of both proteins and peptides. SDS-PAGE allows for qualitative analysis of samples wherein total-thiol samples enable ratio-metric comparisons between samples for determining different levels of oxidation due to treatments/stimuli (Figure 2A). To further investigate the oxidation levels of individual proteins, SDS-PAGE gels may be subjected to western blotting³⁶ (Figure 2B). This enables the model system to be analyzed in greater detail, generating supporting data and further hypotheses about networks and biological pathways. These methods/supporting data can also be utilized as quality control to confirm the expected responses before further in-depth analysis such as LC-MS/MS. Reporter ion intensities of cysteine-containing peptides analyzed by LC-MS/MS can be used to quantify thiol oxidation stoichiometry at individual Cys site levels (Figure 2C,D).

Figure 1: Sample processing workflow. The sample processing workflow is adaptable for investigating thiol oxidation in various sample types and biological systems. The workflow allows for investigation of oxidation at both the protein and peptide levels (e.g., SDS-PAGE, western blot) as well as deep coverage for quantitative, site-specific identification of individual cysteine sites using HPLC coupled with mass spectrometry. Sample processing can be completed in as little as three days, including the completion of several critical steps for the generation of quality, consistent data. Sample multiplexing via TMT labeling allows for the processing of multiple samples in parallel at the same time. The representative 10-plex TMT labeling scheme illustrates how samples can be arranged considering the potential crosstalk from the total-thiol channel. With the isotopic impurities of the TMT reagents, the signal intensity of one channel with high intensity (such as total-thiol) can contribute to another channel with low signal intensity and influence its quantification³⁷. In the scheme, a pooled total-thiol channel (a combination of control and experimental samples) is expected to contain high levels of Cys-peptides and is labeled with 131N, which will have a signal in channel 130N. Thus, channel 130N is not used in the experiment. The amount of channel crosstalk created by TMT labels can be found in the manufacturer's certificates of analysis for a corresponding batch of the reagent. This figure has been adapted from Guo et al., Nature Protocols, 2014²⁵. Abbreviations: NEM = N-ethylmaleimide; DTT = dithiothreitol; SDS-PAGE = sodium dodecylsulfate polyacrylamide gel electrophoresis; SPE = solid-phase extraction; LC-MS/MS = liquid chromatography-tandem mass spectrometry; TMT = tandem mass tag. Please click here to view a larger version of this figure.

Figure 2: Analysis of peptides from RAC enrichment. (A) SDS-PAGE analysis of oxidized peptides from RAW 264.7 cells treated with the chemical oxidant, diamide, for 30 min at increasing concentrations (0.1 and 0.5 mM) and total peptide thiols. This sub-figure and sub-figure D are adapted from Guo et al., Nature Protocols, 2014 ²⁵. Peptides were visualized by silver staining. (B) RAW 264.7 cells were treated with exogenous oxidants (hydrogen peroxide and diamide) at increasing concentrations. The resulting SSG-enriched protein eluate was separated by SDS-PAGE and subsequently probed by western blot for individual proteins (GAPDH, TXN, PRDX3, and ANXA1). This sub-figure is adapted from Su et al., Free Radical Biology and Medicine, 2014³⁶. (C) Representative MS/MS spectrum data of a cysteine-containing peptide viewed in Xcalibur software. The inset MS/MS image shows the corresponding reporter ion intensities for the same peptide in each TMT channel. In this experiment, the total-thiol sample was assigned to the TMT label 131N, which has the highest intensity of all channels used in the experiment. (D) Stoichiometry of iTRAQ-labeled, enriched, oxidized peptides as measured by LC-MS/MS. The total-thiol channel was used as a reference to calculate the stoichiometry of oxidation based on the ratio of reporter ion intensity of each sample compared to that of the total-thiol channel. Abbreviations: RAC = resin-assisted capture; SDS-PAGE = sodium dodecylsulfate polyacrylamide gel electrophoresis; Ctrl = control; GAPDH = glyceraldehyde 3-phosphate dehydrogenase; TXN = thioredoxin; PRDX3 = thioredoxin-dependent peroxide reductase; ANXA1 = annexin A1; LC-MS/MS = liquid chromatography-tandem mass spectrometry; MS/MS = tandem mass spectrometry; TMT = tandem mass tag; iTRAQ = isobaric tag for relative and absolute quantitation. Please click here to view a larger version of this figure.

Buffer name	Purpose	Contents
Buffer A	Lysis/homogenization	250 mM 2-(N-morpholino)ethanesulfonic acid (MES), pH 6.0; 1% sodium dodecylsulfate (SDS); 1% Triton X-100; and 100 mM N-ethylmaleimide (NEM)
Buffer B	Resuspension following protein precipitation and first buffer exchange	250 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), pH 7.0; 8 M urea; 0.1% SDS
Buffer C	Reduction/ Enrichment/ Digestion of Cysteine-containing proteins	25 mM HEPES, pH 7.7; 1 M urea; 0.1% SDS
Buffer D	Second buffer exchange following reduction	25 mM HEPES, pH 7.0, 8 M urea; 0.1% SDS
Buffer E	Washing the resin after hydration	25 mM HEPES, pH 7.7

Table 1.  List of buffers

Supplemental Information. Please click here to download this File.

Discussion

Resin-assisted capture has been utilized across a variety of sample types and biological systems for the investigation of oxidative modifications of cysteine residues^25,29,30. This method allows for the evaluation of samples at multiple levels and readouts, including proteins and peptides using SDS-PAGE and western blot analysis, as well as individual cysteine sites using mass spectrometry. Regardless of the sample type or the final endpoint, the method ultimately allows for highly efficient and specific enrichment of cysteine-containing proteins and peptides³⁸. Using RAC, we have identified changes in the oxidation state of up to several thousands of cysteine sites in different model systems following a perturbation.

In mice subjected to fatiguing muscle contractions, 2,200 S-glutathionylation sites were identified, with more than half having significantly altered levels of S-glutathionylation³². RAC has also been used to profile reversible thiol oxidation of >2,100 sites in cyanobacteria following exposure to a photosynthesis inhibitor or different light conditions⁶. Recently, we profiled total reversible thiol oxidation and S-glutathionylation of >4,000 cysteine sites in RAW 264.7 macrophage cells under resting conditions²⁷. Similarly, Behring et al. quantified the oxidation of ~4,200 cysteine sites in A431 cells following epidermal growth factor stimulation³⁹. These studies demonstrate the robustness of RAC to identify many cysteine sites (at least several thousand) that undergo reversible thiol oxidation. Additionally, the fractionation of samples can enhance the coverage of peptides recovered from an experiment.

Outside of experimental controls, wherein either positive or negative control samples may be employed to confirm the biological responses of the model system, total-thiol enrichment may be performed in parallel with oxidation of thiols. This total-thiol sample provides both stoichiometric comparisons and a baseline from which experimental or treated samples may be compared. In short, this total-thiol sample provides a measurement of the total number of cysteine thiols for a given Cys site in a given sample. This concept was also adopted by the OxiTMT method, which generates samples that contain totally reduced thiols that represent "total cysteine content" for comparison against oxidized cysteines⁴⁰.

In contrast to OxiTMT, RAC is not constrained by the plex number of iodoTMT and thus can incorporate more total thiol channels to better represent the thiol content of multiple sample types used in a study. Additionally, preparing a global sample (not subject to enrichment) in parallel with the thiol redox proteomics workflow may be needed to check whether protein abundances are altered in different sample types. As the method is adaptable to multiple types of redox modifications, proper controls for the specific modification of interest must be considered. For example, ultraviolet light and mercury chloride are both effective at cleaving SNO from proteins, creating an effective negative control for the measurement of SNO^7,25. An effective control for investigating SSG-modified proteins is the omission of the glutaredoxin enzyme from the reduction cocktail during the reduction step. Owing to the relatively high specificity of glutaredoxin, its omission eliminates the reduction of SSG-modified proteins and prevents them from undergoing disulfide exchange and ultimately from being enriched in the final analysis³⁶.

There are several steps within the workflow for RAC that are fundamental for the generation of quality, reproducible data. One of the first crucial steps, and the most important, is the alkylation/blocking of free thiols using the membrane-permeable alkylating agent, N-ethylmaleimide (NEM), which reacts rapidly across a broad pH range^41,42. This step prohibits nascent thiols from being oxidized during sample processing and mitigates non-specific enrichment of these artificially oxidized thiols during disulfide exchange enrichment. Inadequate alkylation will result in an increased background and false-positive signal, as was identified in a previous report⁶.

However, because of its high reactivity and ability to block free thiols, caution must also be taken to ensure its complete removal from samples prior to enrichment, where any remaining NEM can bind and interfere with resin coupling, ultimately resulting in the loss of signal due to a decrease in protein binding. This is accomplished by performing acetone precipitation and several rounds of buffer exchange using molecular weight cut-off filters. Monitoring and maintaining proper pH throughout the protocol is also crucial. A pH of 6.0 is maintained to mitigate disulfide shuffling and the formation of mixed disulfides prior to enrichment. Other steps where extra care must be taken include the enrichment and elution steps, wherein pH values of 7.7 and 8.0 are required for proper enrichment and elution, respectively. Erroneous pH values during these steps will result in a decrease or loss of signal.

To date, there are many chemistry-based methods aside from RAC that are widely used for studying cysteine thiol oxidation, including the most used method, the biotin switch technique^43,44. One thing that all these methods have in common, including RAC, is that they are based on indirect methods for the detection of oxidized cysteines. They rely on chemically modified intermediates of the original oxidized thiol for detection. However, a key attribute that sets RAC apart from other methods is the ability to collect multiplexed, quantitative data on specific cysteine sites of oxidized thiols.

The method is performed with proteins/peptides covalently bound to the resin, which allows proceeding steps (e.g., reduction, labeling, washing, digestion) to be carried out without further handling. By performing LC-MS/MS on multiplexed samples, datasets are generated that enable proteome-wide discoveries. The effects of a specific treatment or stimuli across multiple sample groups are observed at a global level, which enables the discovery of novel mechanisms and pathways. The fundamental workflow is highly adaptable to the end-users' specific needs and areas of interest. Orthogonal validation of findings observed in mass spectrometry data remains a challenge. Site-directed mutagenesis of a specific site and utilization of assays that investigate the consequent effects is a common but labor-intensive approach.

To screen candidate sites that may be biologically significant, bioinformatic studies may be used to learn more about the characteristics of a site with a high level of oxidation, such as a site's proximity to an active site or secondary structure²⁷. Molecular dynamics simulations may prove to be of great value in future studies as they can model the effects of redox modifications on protein structure and provide insight into how a protein's function may be affected^13,45. By implementing this robust strategy, we hope the scientific community will benefit by adapting this method to their own unique model system and expand the current knowledge of redox biology across many different models and biological systems.

Materials

Name	Company	Catalog Number	Comments
2-(Pyridyldithio)ethylamine hydrochloride	Med Chem Express	HY-101794	Reagent for in-house resin synthesis
2.0 mL LoBind centrifuge tubes	Eppendorf	22431048
5.0 mL LoBind centrifuge tubes	Eppendorf	30108310
5.0 mL round bottom tubes	Falcon	352054
Acetone	Fisher Scientific	A949-1
Acetonitrile	Sigma Aldrich	34998
Activated Thiol–Sepharose 4B	Sigma Aldrich	T8512	Potential replacement for thiol-affinity resin
Amicon Ultra 0.5 mL centrifugal filter	Millipore Sigma	UFC5010BK
Ammonium bicarbonate	Sigma Aldrich	09830
Bicinchonicic acid (BCA)	Thermo Scientific	23227	Protein Assay Reagent
Centrifuge	Eppendorf	5810R
Centrifuge	Eppendorf	5415R
Dithiothreitol (DTT)	Thermo Scientific	20291
EDTA	Sigma Aldrich	E5134
HEPES buffer	Sigma Aldrich	H4034
Homogenizer	BioSpec Products	985370
Iodoacetimide (IAA)	Sigma Aldrich	I1149
N-ethylmaleimide	Sigma Aldrich	4259
NHS-Activated Sepharose 4 Fast Flow	Cytiva	17-0906-01	Reagent for in-house resin synthesis
QIAvac 24 Plus vacuum manifold	Qiagen	19413
Sodium chloride	Sigma Aldrich	S3014
Sodium dodecyl sulfate (SDS)	Sigma Aldrich	L6026
Sonicator	Branson	1510R-MT
Spin columns	Thermo Scientific	69705
Strata C18-E reverse phase columns	Phenomenex	8B-S001-DAK	Peptide desalting
Thermomixer	Eppendorf	5355
Thiopropyl Sepharose 6B	GE Healthcare	17-0420-01	Thiol-affinity resin; *Production of Thiopropyl Sepharose 6B resin has been discontinued by the manufacturer (see protocol for details).
TMT isobaric labels (16 plex)	Thermo Scientific	A44522	Peptide labeling reagent; available in multiple formats
Triethylammonium bicarbonate buffer (TEAB)	Sigma Aldrich	T7408
Trifluoroacetic acid (TFA)	Sigma Aldrich	T6508
Triton X-100	Sigma Aldrich	T8787
Trypsin	Promega	V5820
Urea	Sigma Aldrich	U5378
Vacufuge Plus speedvac	Eppendorf	22820001	vacuum concentrator
Vortex mixer	Scientific Industries	SI-0236