Stoichiometric Thiol Redox Proteomics for Quantifying Cellular Responses to Perturbations

Abstract

Post-translational modifications regulate the structure and function of proteins that can result in changes to the activity of different pathways. These include modifications altering the redox state of thiol groups on protein cysteine residues, which are sensitive to oxidative environments. While mass spectrometry has advanced the identification of protein thiol modifications and expanded our knowledge of redox-sensitive pathways, the quantitative aspect of this technique is critical for the field of redox proteomics. In this review, we describe how mass spectrometry-based redox proteomics has enabled researchers to accurately quantify the stoichiometry of reversible oxidative modifications on specific cysteine residues of proteins. We will describe advancements in the methodology that allow for the absolute quantitation of thiol modifications, as well as recent reports that have implemented this approach. We will also highlight the significance and application of such measurements and why they are informative for the field of redox biology.

Keywords: cysteine; redox post-translational modifications; redox proteomics; thiol proteome.

Figure 1

Cysteine thiol modifications. Reduced thiols can be subjected to multiple types of modifications. In this cartoon, disulfide formation includes those that form intramolecularly, or within the same protein, or intermolecularly through exchange with a sulfur group on another protein or molecule. Sulfinylation and sulfonylation occur as subsequent, irreversible oxidative modifications of cysteine thiols that are sulfenylated. Note that persulfidation is also referred to as sulfhydration. Palmitoylation is shown here as a representative modification of acylation, which is the covalent linking of an acyl chain to a cysteine; however, acylation can include other molecules besides 16C palmitate. With the exception of sulfinylation and sulfonylation, all modifications illustrated here are considered reversible.

Figure 2

Thiol redox modifications at the molecular level. Two examples where thiol redox modifications alter the structural and functional properties of proteins are presented. (A) Human GAPDH monomer in complex with the nicotinamide adenine dinucleotide (NAD⁺) cofactor (PDB: 1u8f) [30]. Cysteines 152 and 156 are shown and labeled in magenta and highlight the close proximity of Cys152 to the NAD⁺ binding pocket. Note that the catalytic cysteine in GAPDH is Cys152 in humans and Cys150 in mice, rats, or rabbits. Simplified cartoons of findings from rabbit GAPDH Cys150 [27] when reduced, sulfenylated, and glutathionylated are shown to the right of the structure. Glutathionylation of GAPDH active site cysteine can also lead to the formation of an intramolecular disulfide that causes GAPDH to inactivate and aggregate but protects against irreversible oxidation. (B) (i) Representative bar plot of EGFR (epidermal growth factor receptor) activity when the Cys797 in the active site is reduced, sulfenylated, or glutathionylated. The plot represents trends as fold change based on the data observed in [31]. (ii) Wild-type human EGFR kinase domain (PDB: 3vjo) [32] with Cys797 shown and labeled in magenta. Adenosine 5′-(β,γ-imino)triphosphate (AMP-PNP), a non-hydrolysable analog of ATP, is also shown in the molecule to highlight Cys797’s close proximity to the catalytic site of EGFR. (iii) T790M mutant human EGFR in complex with the inhibitor naquotinib (PDB: 5y9t) [33], which is covalently linked to Cys797 and is used to treat cancer. The efficacy of inhibitors that covalently bind to Cys797 can be impacted by the oxidation of Cys797. Molecular images were generated with ICM-Browser (http://www.molsoft.com/icm_browser.html (accessed on 9 December 2020)).

Figure 3

Modeling redox networks and redox state. (A) Simplified cartoon representing the redox circuits of a redox network present in the extracellular domain (e.g., receptors) and different intracellular compartments of a cell. The uniquely colored shapes represent different pathways, or circuits, that function in specific subcellular compartments. Interconnected pathways are denoted by solid lines that link the shapes, while dashed lines represent transient interactions among circuits. (B) Plot illustrating the microscopic effect of thiol redox signaling on the global redox state. For a given protein (protein X) or redox-sensitive cysteine site (Cys site X), the distribution of oxoforms can vary between steady state and perturbed conditions. This plot shows that the majority of oxoforms of a protein or cysteine site are reduced under steady state conditions; however, in the presence of a perturbing agent, the distribution transitions to a majority of oxidized oxoforms. As noted by the heatmap to the right, the transition of the protein or Cys site to become more oxidized following a perturbation has a negligible effect on the overall redox state of the cell. Under homeostatic conditions, the transition of reversibly oxidized proteins and Cys sites to become more or less oxidized within a specific circuit may also have little impact on the global redox state. (C) Plot illustrating the macroscopic effects of thiol redox signaling. In this scenario, the ratio of reduced versus oxidized glutathione (GSH/GSSG) is monitored as a readout for the global redox state. Under steady state conditions, the global redox state is under a more reduced environment (more GSH than GSSG); however, a severe perturbation causes the global redox state to become more oxidized (more GSSG than GSH). This scenario demonstrates an oxidative stress; however, the same effect could be observed from a perturbation that causes reductive stress. In the context of the model in (A), a macroscopic effect may be observed across multiple circuits, as opposed to a microscopic effect observed in one circuit.

Figure 4

Thiol redox state and stoichiometry concept. For any given protein, the occupancy of oxidized versus reduced forms is determined by the ratio of oxidized over total abundance of protein analyzed (total sum of oxidized and reduced forms) and is used to interpret modification stoichiometry. Multiple scenarios of increasing thiol oxidation are illustrated to represent how the stoichiometry of oxidative modifications can change. Depending on the type of modification and protein, thiol redox modifications can activate or inactivate a protein’s function. As an example, the heatmap gauge on the left side of the figure illustrates how changes in thiol oxidation stoichiometry can correlate with the amount of activated protein, where more oxidation leads to less activated protein.

Figure 5

Simplified schematics of different quantitative redox proteomics approaches. Each schematic uses two proteins per scenario to represent the reduced (S-H) and oxidized (S-Ox) forms. Oxidized thiols represent any variety of reversible thiol modifications shown in Figure 1. Blue reduced thiol groups represent the formation of nascent free thiols following the reduction of oxidized thiols. In the cysteine-reactive phosphate tag and RAC (resin assisted capture) methods, total thiol samples are included as a separate reference channel for stoichiometric measurements (i.e., total thiol, representing all detectable thiols). Measurements are based on reporter ion intensities. (A) Labeling reduced and oxidized cysteines with isotope coded affinity tags (OxICAT). Free thiols are irreversibly labeled with isotopically light ICAT, followed by the reduction of oxidized thiols to free thiols that are labeled with isotopically heavy ICAT. Labeled proteins are then digested and enriched by streptavidin beads via the biotin group on the ICAT reagents. The enriched peptides are released by the cleavage of the biotin moiety and analyzed by mass spectrometry. The modification stoichiometry is determined by the ratio of a peptide’s corresponding heavy and light peptide intensities. (B) Free thiols are initially alkylated, and then oxidized thiols are totally or selectively reduced (if probing for a specific modification). Endogenous phosphorylation is removed to avoid the recovery of phosphorylated peptides during immobilized metal affinity chromatography (IMAC). Nascent free thiols are labeled with cysteine-reactive phosphate tags (CPTs), the proteins are digested, and the peptides are labeled with tandem mass tags (TMT). The CPT tags allow the peptides to be enriched by IMAC and are eluted off the column prior to analysis by mass spectrometry. The oxidation occupancy is determined by taking the ratio of reporter ion intensities in the oxidized and total thiol channels to evaluate the modification stoichiometry. It is ideal that each sample type (i.e., control, experiment, treated, untreated) should have its own total thiol channel for the appropriate measurement and comparison of redox state. (C) Similar to the approach in (B), free thiols are alkylated, and then oxidized thiols are reduced totally or selectively. Resin assisted capture (RAC) of nascent free thiols using a thiopropyl sepharose resin is followed by on-resin digestion and TMT labeling. Finally, peptides are eluted off the resin prior to their analysis by mass spectrometry. The oxidation occupancy is determined by taking the ratio of reporter ion intensities in the oxidized and total thiol channels to evaluate the modification stoichiometry. It is ideal that each sample type should have its own total thiol channel for the appropriate measurement and comparison of redox state. (D) Like OxICAT, thiol reactivity profiling uses isotopically light and heavy thiol-directed probes; however, these probes are designed to only label free thiols to identify and quantify reactive thiol groups in different types of samples (control or treated). Therefore, this method does not represent the oxidized cysteines, but rather cysteines that are reactive. The procedure begins by labeling free thiols with a probe called 2-iodo-N-(prop-2-yn-1-yl)acetamide) (IPM), which is followed by digestion. Thiol groups labeled by the IPM reagent are then tagged with isotopically light or heavy variants of an azido-UV-biotin tag and enriched by streptavidin-coated beads. The enriched peptides are eluted by UV due to the photocleavable linker in the azido-UV-biotin tag and analyzed by mass spectrometry. The reactivity is determined by the ratio of a peptide’s corresponding heavy and light peptide intensities.