Adoption of aerobic metabolism occurred some 2.2 billion years ago, facilitating more efficient energy metabolism and the development of more complex multicellular organisms Citation[1]. However, molecular oxygen can also be highly toxic to biological systems since it is reduced into chemically reactive derivatives, collectively called reactive oxygen species (ROS), which, in some cases, react at diffusion-limited rate constants with cell components, such as membrane lipids, DNA and proteins Citation[2,3]. Cells evolved extensive batteries of defenses against the threat posed by ROS, notably induction of antioxidant response genes, enzyme activities (such as catalase), the thioredoxin system and antioxidant molecules (such as glutathione). Biological systems normally exist in a state of redox homeostasis where antioxidant defenses cope efficiently with background ROS levels. However, if ROS levels increase so as to exceed available defenses or if the defenses decrease (as happens during aging), a state of oxidative stress results. This is a feature of aging and very important in human pathology Citation[4]. Moreover, it has increasingly come to be realized that ROS and related reactive nitrogen species (RNS) are potent biological signaling molecules, even in sub-stress scenarios Citation[5,6]. Proteins absorb the bulk of ROS, which have the potential to complicate the proteome both by altering levels of individual proteins through mechanisms, such as gene-induction and altered degradation of modified proteins, and by changing the covalent structure of proteins, resulting in targeted modification of protein function Citation[2]. Redox proteomics aims to detect and analyze redox-based changes within the proteome both in redox signaling scenarios and in oxidative stress.
Electrophoresis-based detection of redox lesions
There are many well-known limitations to 2DE – notably underselection of certain protein categories, limited dynamic range and comigration of multiple proteins Citation[7]. Notwithstanding these shortcomings, it has proved possible to detect changes in protein-expression signatures of cells experiencing oxidative stress Citation[8] and in cell signaling Citation[6,9]. Typically, these changes involve up- or down-regulation of a relatively small fraction of the visible protein spots, indicating that the proteome is remarkably robust to redox-based change Citation[8,9]. Since ROS/RNS can change the covalent structure of some proteins (as opposed to altering their individual abundance), it is often possible to reveal these changes by labeling modified proteins followed by detection.
A common strategy is to perform immunoblots of 2DE separations Citation[10]. These reveal formation of carbonyl groups on residue side chains, tyrosine nitration and glutathionylation Citation[11–13]. Oxidation of cysteine -SH groups (e.g., to form disulfides, nitrosothiols [-SNO] or sulfenic [SOH]/sulfinic [SO2H]/sulfonic acids [SO3H]), can have functional significance and lead to increased turnover of selected proteins. Protein disulfides, nitrosothiols and sulfenic acids are readily detectable in 2DE by selective reduction followed by labeling modified proteins with thiol-specific reagents Citation[14,15]. Detection of redox-based modifications in this manner is arguably more informative than mere changes in protein abundance.
A major difficulty presented by detection of effects of ROS/RNS on the proteome is the fact that while some modifications are irreversible and, therefore, amenable to procedures used in 2DE, important categories of modification are reversible (e.g., formation of methionine sulfoxide, cysteine sulfenic acid or disulfides, including glutathionylation) and it is technically difficult to ‘freeze’ proteomes to faithfully preserve this Citation[15,16]. This is highlighted by the growing realization that reversible oxidation of protein cysteines and methionines represent quantitatively important buffers to transient increases in ROS Citation[17,18]. When examining 2DE separations, it can be difficult to know what fraction of a particular protein is carrying a particular redox lesion. Some modifications (e.g., carbonylation) seem to be very quantitative with the majority of target proteins carrying the modification, while others (e.g., nitration at tyrosines) may be present with an abundance as low as one protein molecule in a million. Use of affinity selection methods (see later) to trap subproteomes separable by 2DE can sometimes overcome this limitation.
Trapping the proteome’s redox status
Protein thiol groups represent a technically challenging target for ROS as they can exist in a range of oxidized forms and yet represent the main point of interaction between redox status and cell signaling Citation[3,6]. There are two critical steps in analyzing the thiol proteome: blocking or ‘freezing’ of free thiols and reduction of reversibly modified thiols. Several methods and chemicals are available for thiol quenching, including alkylation with N-ethylmaleimide (NEM), iodoacetamide (IAM), methyl methanethiosulfonate (MMTS) or acidification. The method of choice depends on what further steps are required. For example, NEM reacts irreversibly, while acidification is reversible Citation[18]. The reducing step is of critical importance and there are several general thiol-reducing agents with their individual advantages and disadvantages. For instance, thiol-containing reducing agents can crossreact and compete with thiol-detection agents so their removal becomes necessary. At this point, a degree of specificity making mapping of particular thiol modifications becomes possible. Sodium arsenite selectively reduces -SOH Citation[19] while the ‘biotin switch’ used for the detection of nitrosylation uses sodium ascorbate for reducing -SNO Citation[20], although there has been much debate regarding the specificity of this reagent when not used under optimal conditions. Reduction of in vivo targets of the thioredoxin and glutaredoxin systems could be achieved by using the Trx/TrR system and Grx/glutathione/GR systems, respectively. Identification of target proteins of these systems is helpful in elucidating redox-regulated systems within cells.
Application of mass spectrometry technology
Emphasis has shifted in recent years as researchers seek technologies that are not as time- and labour-intensive as 2DE, but can allow for higher sample throughput. Increasingly, developments in mass spectrometry (MS) technology, including machine accuracy, and more freely available protein databases mean that there is an increasing emphasis on ‘gel-free’ proteomics. There have been two general approaches in which MS has been applied to proteomics; a discovery-proteomic and targeted-proteomic approach. Discovery- or shotgun-proteomics approach in its most general sense involves proteolytic digestion of a population of proteins and analyzing resulting peptides by MS/MS. It has significant limitations as the proteome is highly complex, the most abundant proteins can be identified multiple times and technical replicates can show limited overlap Citation[21]. If applied to redox proteomics it means that many of the proteins with reactive thiols that are modified to the greatest degree by an oxidant may be far less abundant than those modified to a lesser degree and, thus, not detected; however, it may be simplified by preselecting for only those peptides that contain redox-modified cysteines Citation[22]. Its main advantage is that, not only is the modified protein identified, but cysteines affected can also be mapped in a single analysis. Alternatively, a hypothesis-driven or targeted proteomic approach may be used as an analytical tool for structural and molecular studies of a specific protein, where specific peptides are selected for analysis by MS. This is known as selective reactive monitoring (SRM). This approach shows both high selectivity and sensitivity, but it is necessary to know exactly which peptide is subject to redox modification and its potential transition states, as well as potential modifications of the peptide (e.g., oxidation of thiol groups can increase peptide mass; e.g., -SOH: +16 Da, SO2H: +32 Da, -SO3H: +48 Da). The only drawback with this approach is that, when all potential peptide modifications are introduced, analysis can be rendered quite complicated. The new generation of MS instruments also facilitate analysis of intact proteins. Potential modifications can be identified by the difference in mass between modified and non-modified proteins in the spectrum. This application has been particularly used in MALDI TOF and electrospray ionization MS, where an increase in the mass of proteins by 305 Da indicates that the protein is glutathionylated at a cysteine residue Citation[23]. As the protein is not further fractionated by MS/MS it does not specify the cysteine residue that is glutathionylated, although once the modification is confirmed, cysteine-containing peptides may be targeted and tagged, as mentioned previously, to identify specific cysteines. Identification of redox modifications in MS is improved when the modification leaves a distinct, irreversible fingerprint on the peptide. For instance, analysis of nitrosylation is difficult as it is a modification that is extremely labile. Therefore, applying reagents that can irreversibly tag the site of modification can lead to more confident identifications of both protein and redox modification.
After the identification of the redox-modified peptides, and hence proteins, the next logical step is quantification of their modifications. Several reagents and techniques are available for relative quantification of proteins; isotope-coded affinity tags (ICAT) Citation[24], stable-isotope labeling of amino acids in cell culture (SILAC) Citation[25] and isobaric tags for relative and absolute quantification (iTRAQ) Citation[26]. However, measuring the relative quantity of a protein between paired samples does not tell us about the activity of the protein itself. This is especially important in reference to redox proteins containing thiol-based switches, which are highly susceptible to activation/inactivation. Therefore, quantifying the redox status of the protein may be more informative than merely quantifying the protein’s relative abundance. Indeed, calculation of individual cysteine pKa values within proteins have been determined with the aid of thiol-specific ICAT reagents by following chemical modification rates across a range of pH values Citation[27]. Similarly, ICAT reagents have been used, not only to identify ROS targets, but also to quantify oxidative thiol modifications of individual proteins. One of the first applications of this technology involved exposing proteins to either oxidative stress or normal conditions prior to labeling with either heavy or light ICAT reagents, respectively Citation[28,29].
A more recent method for relatively quantifying proteins, iTRAQ, has become especially popular as it allows up to eight samples to be analyzed simultaneously. A recent article incorporated both the selectivity and accuracy of ICAT and iTRAQ reagents, in identification of Trx targets in mice Citation[30]. This study used a cardiac-specific Trx1-overexpressing transgenic mouse model (Tg-Trx1). Trx1-reduced proteins were distinguished via comparison of the ICAT results from controls. The authors then compared ICAT data against iTRAQ protein expression results to reveal genuine Trx1 protein targets. These showed unchanged iTRAQ expression but altered ICAT ratios, as opposed to Trx1-induced proteins.
Analysis by SILAC has proved useful for the relative quantification of proteins in paired samples. This involves growing two cell populations, one in a medium containing a ‘light’ amino acid and the other in a medium containing a ‘heavy’ isotope of that amino acid. Incorporation of the SILAC isotope in proteomes of the two populations allows calculation of the relative abundance of each protein Citation[31]. However, SILAC should be used with caution in redox proteomics as, for complete incorporation of modified amino acids, auxotrophic strains for arginine and lysine are recommended. Considering that amino acid biosynthesis is potentially subject to redox control, using auxotrophic strains may result in an oversimplified picture of the redox proteome.
Conclusion
The next few years will be especially exciting for redox proteomics as we will see continued application of more advanced instrumentation and reagents in both selective tagging and quantification of the redox modifications of not only individual proteins but also of redox-sensitive residues within those proteins. This will yield greater insight to specific pathways subject to redox control within cells.
Financial & competing interests disclosure
David Sheehan’s laboratory is financially supported by the Higher Education Authority of Ireland, Irish Research Council for Education Science and Technology and the European Union. Brian McDonagh and José Antonio Bárcena are supported by the Andalusian Regional Government, Spain. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.
No writing assistance was utilized in the production of this manuscript.
References
- Owens T, Cess CD, Ramanathan A. Enhanced carbon dioxide greenhouse to compensate for reduced solar luminosity on early Earth. Nature277, 640–641 (1979).
- Davies MJ. The oxidative environment and protein damage. Biochim. Biophys. Acta1703, 93–109 (2005).
- Winterbourn CC. Reconciling the chemistry and biology of reactive oxygen species. Nat. Chem. Biol.4, 278–286 (2008).
- Halliwell B, Gutteridge J. Free Radicals in Biology and Medicine (4th Edition). Oxford University Press, Oxford, UK (2007).
- Oktyabrsky ON, Smirnova GV. Redox regulation of cellular function. Biochemistry72(2), 32–145 (2007).
- Chiarugi P, Buricchi F. Protein tyrosine phosphorylation and reversible oxidation: two cross-talking posttranslational modifications. Antioxid. Redox Signal.9, 1–24 (2007).
- Görg A, Weiss W, Dunn MJ. Current two-dimensional electrophoresis technology for proteomics. Proteomics4, 3665–3685 (2004).
- Rabilloud T, Heller M, Gasnier F et al. Proteomics analysis of cellular response to oxidative stress – evidence for in vivo overoxidation of peroxiredoxins at their active site. J. Biol. Chem.277, 19396–19401 (2002).
- Kanamoto T, Hellman U, Heldin CH et al. Functional proteomics of transforming growth factor-β 2-stimulated My1Lu epithelial cells: Rad51 as a target of TGFβ 1-dependent regulation of DNA repair. EMBO J.21, 1219–1230 (2002).
- Sheehan D. Detection of redox-based modification in two dimensional electrophoresis proteomic separations. Biochem. Biophys. Res. Commun.349, 455–462 (2006).
- Noda Y, Berlett BS, Stadtman ER et al. Identification of enzymes and regulatory proteins in Escherichia coli that are oxidized under nitrogen, carbon or phosphate starvation. Proc. Natl Acad. Sci. USA104, 18456–18460 (2007).
- Tyther R, Ahmeda A, Johns E et al. Proteomic identification of tyrosine nitration targets in kidney of spontaneously hypertensive rats. Proteomics7, 4555–4564 (2007).
- Newman SF, Sultana R, Perluigi M et al. An increase in S-glutathionylated proteins in the Alzheimer’s disease inferior parietal lobule, a proteomics approach. J. Neurosci. Res.85, 1506–1514 (2009).
- Lind C, Gerdes R, Hamnell Y et al. Identification of S-glutathionylated cellular proteins during oxidative stress and constitutive metabolism by affinity purification and proteomic analysis. Arch. Biochem. Biophys.406, 229–240 (2002).
- Baty JW, Hampton MB, Winterbourn CC. Detection of oxidant sensitive thiol proteins by fluorescence labelling and two dimensional electrophoresis. Proteomics2, 1261–1266 (2002).
- Leichert LI, Jakob U. Protein thiol modifications visualized in vivo. PLOS Biol.2, 1723–1737 (2004).
- Levine RL, Moskovitz J, Stadtman ER. Oxidation of methionines in proteins: roles in antioxidant defense and cellular regulation. IUBMB Life50, 301–307 (2000).
- Hansen RE, Roth JR, Winther JR. Quantifying the global cellular thiol-disulfide status. Proc. Natl Acad. Sci. USA106, 422–427 (2009).
- Saurin AT, Neubert H, Brennan JP et al. Widespread sulfenic acid formation in tissues in response to hydrogen peroxide. Proc. Natl Acad. Sci. USA101, 17982–17987 (2004).
- Jaffrey SR, Erdjument-Bromage H, Ferris CD et al. Protein S-nitrosylation: a physiological signal for neuronal nitric oxide. Nat. Cell. Biol.3, 193–197 (2001).
- Malmstrom J, Lee H, Aebersold R. Advances in proteomic workflows for systems biology. Curr. Opin. Biotechnol.18, 378–384 (2007).
- McDonagh B, Ogueta S, Lasarte G et al. Shotgun redox proteomics identifies specifically modified cysteines in key metabolic enzymes under oxidative stress in Saccharomyces cerevisiae. J. Proteomics72, 677–689 (2009).
- Hashemy SI, Johansson C, Berndt C et al. Oxidation and S-nitrosylation of cysteines in human cytosolic and mitochondrial glutaredoxins: effects on structure and activity. J. Biol. Chem.282, 14428–14436 (2007).
- Gygi SP, Rist B, Gerber SA et al. Quantitative analysis of complex protein mixtures using isotope-coded affinity tags. Nat. Biotechnol.17(10), 994–999 (1999).
- Ong SE, Blagoev B, Kratchmarova I et al. Stable isotope labeling by amino acids in cell culture, SILAC, as a simple and accurate approach to expression proteomics. Mol. Cell. Proteomics1, 376–386 (2002).
- Ross PL, Huang YN, Marchese JN et al. Multiplexed protein quantitation in Saccharomyces cerevisiae using amine-reactive isobaric tagging reagents. Mol. Cell. Proteomics.3(12), 1154–1169 (2004).
- Nelson KJ, Day AE, Zeng BB et al. Isotope-coded, iodoacetamide-based reagent to determine individual cysteine pK(a) values by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Anal. Biochem.375, 187–195 (2008).
- Sethuraman M, McComb ME, Huang H et al. Isotope-coded affinity tag (ICAT) approach to redox proteomics: identification and quantitation of oxidant-sensitive cysteine thiols in complex protein mixtures. J. Proteome Res.3, 1228–1233 (2004).
- Liechert LI, Gehrke F, Gudiseva HV et al. Quantifying changes in the thiol redox proteome upon oxidative stress in vivo. Proc. Natl Acad. Sci. USA105, 8197–8202 (2008).
- Fu C, Wu C, Liu T et al. Elucidation of thioredoxin target protein networks in mouse. Mol. Cell. Proteomics.8, 1674–1687 (2009).
- Mann M. Functional and quantitative proteomics using SILAC. Nat. Rev. Mol. Cell Biol.7, 952–958 (2006).