Discovery at the VICB
A Global View of S-Sulfenylation
By: Carol A. Rouzer, VICB Communications
Published: September 9, 2014
A comprehensive approach to identification of cysteine sulfenic acids in proteins provides insight into how this post-translational modification regulates cellular processes in response to oxidative stress.
Cysteine S-sulfenylation is a reversible oxidation reaction that converts the thiol group of protein cysteine residues to a sulfenic acid group (Figure 1A). Evidence suggests that this reaction may be a mechanism by which protein function is regulated in response to the redox balance within the cell. Previously recognized by a loss of reactivity to thiol-modifying agents, cysteine sulfenic acids are now detected through their ability to react with dicarbonyl compounds such as dimedone. In fact, the creation of a dimedone analog bearing a terminal alkyne group (DYn-2, Figure 1B) has enabled investigators to label cysteine sulfenic acid groups in proteins with a tag that can be linked, using click chemistry, to biotin-containing reagents. The attached biotin allows capture of the modified proteins or peptides using streptavidin chromatography. This approach has yielded considerable information about cysteine S-sulfenylation; however, comprehensive assessments of this process in intact cells responding to oxidative stress have not been carried out. This led Vanderbilt Institute of Chemical Biology member Dan Liebler and postdoctoral fellow Jing Yang to develop a robust approach for the quantification of cysteine S-sulfenylation across the entire cellular proteome (J. Yang, et al., (2014) Nat. Commun., published online September 1, DOI: 10.1038/ncomms5776).
Figure 1. Cysteine S-sulfenylation and tools used to detect it. (A) Cysteine (CSH) is mildly acidic and exists in an equilibrium with the conjugate thiolate anion (CS-). The anion reacts reversibly with hydrogen peroxide (or another oxidant) to yield cysteine sulfenic acid (CSOH). Reaction with a second cysteine thiolate irreversibly forms a disulfide bond. (B) Structure of DYn-2, which reacts with CSOH to provide an alkyne tag. Using click chemistry, the alkyne reacts with the azido group of the biotin reagent (C), providing a means to isolate the CSOH-containing peptide using streptavidin chromatography. The biotin reagent can be cleaved by exposure to UV light, which removes the biotin-containing portion of the reagent from the peptide.
Yang began by systematically optimizing conditions for the reaction of DYn-2 with cysteine sulfenic acids in proteins of intact cells. Similarly, he optimized the conditions for the click chemistry reaction to attach the DYn-2-labeled peptides to a photo-cleavable biotin-containing reagent (Figure 1C). He then applied these conditions to an overall protocol for the detection of cysteine S-sulfenylation (Figure 2). According to the protocol, DYn-2 treatment labels cellular protein S-sulfenyls. After cell lysis and digestion of the proteins into peptides, click chemistry attaches the biotin reagent, which is then used to capture the modified peptides using streptavidin affinity chromatography. Ultraviolet light cleaves the biotin reagent, releasing the peptides from the streptavidin column. Then high energy collisional dissociation tandem mass spectrometry provides the analytical platform for identification of the peptides and the sites of modification. The finding that DYn-2-tagged cysteines produce three characteristic diagnostic fragments during mass spectrometric analysis facilitates peptide identification.
Figure 2. Protocol for isolation and identification of cysteine sulfenic acids in intracellular proteins. The proteins are labeled using DYn-2 added to cultures of intact cells. The cells are lysed, and the proteins are digested. The photocleavable biotin reagent is attached to the DYn-2 tag using click chemistry, and the peptides are then captured using streptaviding chromatography. Following cleavage of the linker using UV light, the peptides are eluted and analyzed by liquid chromatography-tandem mass spectrometry. Image reproduced by permission from Macmillan Publishers Ltd, from J. Yang, et al., (2014) Nat. Commun., published online September 1, DOI: 10.1038/ncomms5776. Copyright 2014.
The researchers first applied their method to the evaluation of cysteine sulfenic acids present in the proteins of resting RKO human colon carcinoma cells. They detected 1,153 distinct peptides and were able to validate and identify the site of modification of 952 of these. The 952 modified cysteines represented sites on 680 distinct proteins, some of which had not previously been identified as subject to cysteine S-sulfenylation. Among the novel proteins were ACTN, ACLY, CFL1, EEF1H, MAPK1, HSP90, RPL18, SIRT6, TUBB, and YWAHE.
Next, the researchers asked how patterns of cysteine S-sulfenylation change during oxidative stress. To answer this question required a means of quantifying differences in levels of sulfenic acids between cells exposed to different treatments. The investigators accomplished this goal by using heavy and light DYn-2 isotopomers (Figure 3), which were synthesized by their collaborators Kate Carroll and Vinayak Gupta at the Scripps Research Institute in Florida. Control cells were labeled with “light” DYn-2, while cells exposed to oxidative stress were labeled with “heavy” (deuterated) DYn-2. The cells were then combined at a ratio of 1:1, and processed according to the protocol shown in Figure 2. The deuterium label in the heavy DYn-2 resulted in mass signals with a higher m/z value, enabling the investigators to distinguish the two cell treatments. The ratio of the intensity of the two signals provided a means to directly compare the levels of cysteine S-sulfenylation present in the two cell populations. With this technique in hand, Yang and Liebler investigated changes in cysteine S-sulfenylation in RKO cells responding to treatment with H2O2. They found that H2O2 exposure resulted in a >2-fold increase in >89% of S-sulfenylated protein sites. They also discovered that distinct sites on the same protein could respond quite differently to the H2O2 challenge. For example, Cys-91 of PRDX6 exhibited a marked increase (RH/L = 9.15) in S-sulfenylation, while the S-sulfenylation of Cys-47 dropped to an equally impressive degree (RH/L = 0.07). The decrease in S-sulfenylation of Cys-47 was explained by the susceptibility of sulfenic acids to further oxidation, yielding sulfinic and sulfonic acids, which do not react with DYn-2.
Figure 3. Quantification of changes in cysteine S-sulfenylation as a result of oxidative stress. Control cells are labeled with DYn-2 (red), while treated cells are labeled with deuterated DYn-2 (blue). The cells are combined, and samples are processed as shown in Figure 2. The signals from the treated cells will have a higher m/z value due to the presence of the deuterium, and the ratio of the “heavy” to the “light” signal provides a way to quantify changes in cysteine sulfenic acid levels as a result of the treatment. Image reproduced by permission from Macmillan Publishers Ltd, from J. Yang, et al., (2014) Nat. Commun., published online September 1, DOI: 10.1038/ncomms5776. Copyright 2014.
Bioinformatics analysis revealed that S-sulfenylated proteins in RKO cells are distributed across all cellular compartments and participate in a diverse array of biochemical processes and pathways. The investigators were surprised that they did not identify any S-sulfenylated transcription factors, since redox regulation of transcription factors has been reported. However, they note that these proteins may be present in such low amounts that they were not detected. Alternatively, their redox-dependent regulation might occur through interaction with other proteins that are S-sulfenylated.
In total, the investigators identified 1,105 sites of S-sulfenylation on 778 proteins in RKO cells. Most proteins (92%) contained only one or two sites, while a small number (1.3%) contained over 4 sites. Most of the sites (>60%) were solvent-exposed, suggesting that this is a major determinant of S-sulfenylation. However, the researchers acknowledge that sites buried deep in the center of the protein might not be accessible to their DYn-2 probe, and so might not be detected. Sequence analysis around S-sulfenylation sites indicated a high prevalence of glutamic acid or lysine residues and a near absence of cysteine. The investigators postulated that a deprotonated glutamate could hydrogen bond with the sulfhydryl of the nearby cysteine, lowering its pKa, and thereby facilitating its deprotonation and S-sulfenylation (Figure 1).
Having evaluated their method under conditions of severe oxidative stress, the researchers next asked how a physiological stimulus might affect cysteine S-sulfenylation. To explore this question, they chose epidermal growth factor (EGF) stimulation of A431 epidermoid carcinoma cells, as EGF triggers the generation of endogenous H2O2. The investigators found that EGF treatment produced a >2-fold increase in S-sulfenylation of about half of identified sites, as compared to an increase in 90% of sites observed in response to exogenous H2O2 in these cells. Although there was a general correlation between the sites modified by EGF and exogenous H2O2, the response to EGF was more selective and specific.
One S-sulfenylated protein that engendered particular interest was SIRT6 (sirtuin 6), a member of the NAD+-dependent deacetylase family of enzymes. SIRT6 plays a role in the regulation of metabolism, the response to stress, and tumorigenesis. SIRT6 suppresses the glycolytic pathway by acting as a corepressor for HIF1A (hypoxia inducible factor 1A), but the mechanism for this interaction is not fully understood. The investigators found that addition of H2O2 to mixtures of recombinant SIRT6 and an HIF1A protein fragment resulted in formation of a complex between the two proteins that was visible on non-reducing gels, but not reducing gels. This complex was not formed in the presence of dimedone. These results suggested that S-sulfenylation of a cysteine on SIRT6 in the presence of H2O2 preceded formation of a disulfide bond between the two proteins. This hypothesis was further supported by mass spectral analysis of the complex, which identified a disulfide bond between Cys-18 of SIRT6 and Cys-800 of HIF1A. These findings suggest an interesting new mechanism by which oxidative stress could lead to complex formation between the two proteins, resulting in repression of HIF1A activity by SIRT6 (Figure 4).
Figure 4. Mechanism for regulation of HIF1A by SIRT6. S-sulfenylation of Cys-18 of SIRT6 leads to disulfide bond formation between that cysteine and Cys-800 of HIF1A. The resulting complex is then inhibitory to HIF1A-dependent gene expression. Image reproduced by permission from Macmillan Publishers Ltd, from J. Yang, et al., (2014) Nat. Commun., published online September 1, DOI: 10.1038/ncomms5776. Copyright 2014.
Together the results illustrate the breadth of potential biochemical regulation through S-sulfenylation of cysteine residues. Clearly, cells possess extensive capacity for responding to the redox status of their environment. Yang and Liebler described a robust method for exploration of the implications of this common post-translational modification and, as in the case of SIRT6, the identification of novel mechanisms by which this modification can regulate protein function.