MASS SPECTROMETRY ANALYSIS OF NITROTYROSINE-CONTAINING PROTEINS

Oxidative stress plays important roles in a wide range of diseases such as cancer, inﬂammatory disease, neurodegenerative disorders, etc. Tyrosine nitration in a protein is a chemically stable oxidative modiﬁcation, and a marker of oxidative injuries. Mass spectrometry (MS) is a key technique to identify nitro-tyrosine-containing proteins and nitrotyrosine sites in endogenous and synthetic nitroproteins and nitropeptides. However, in vivo nitrotyrosine-containing proteins occur with extreme low-abundance to severely challenge the use of MS to identify in vivo nitroproteins and nitrotyrosine sites. A preferential enrichment of nitroproteins and/or nitropeptides is necessary before MS analysis. Current enrichment methods include immuno-afﬁnity techniques, chemical derivation of the nitro group plus target isolations, followed with tandem mass spectrometry analysis. This article reviews the MS techniques and pertinent before-MS enrichment techniques for the identiﬁcation of nitrotyrosine-containing proteins. This article reviews future trends in the ﬁeld of nitroproteomics, including quantitative nitroproteomics, systems biological networks of nitroproteins, and structural biology study of tyrosine nitration to completely clarify the biological functions of tyrosine nitration. # 2013 Wiley Periodicals, Inc. Mass Spec Rev 34:423–448, 2015


I. INTRODUCTION
Oxidative/nitrative stress-mediated tyrosine nitration is involved in a wide range of physiological and pathological processes (Zhan & Desiderio, 2004, exists in normal physiology, and is enhanced in pathologies (Scaloni, 2006;Zhan & Desiderio, 2009a). Tyrosine nitration is not only a marker of oxidative injuries, but also alters the functions of proteins that participate in different pathophysiological processes such as cancer, inflammation disease, and neurodegenerative diseases (Zhan & Desiderio, 2004, 2006. The nitrotyrosine residue has unique physical and chemical properties (Yee et al., 2003;Ghesquiere et al., 2006;Zhan & Desiderio, 2006) such as pKa value (the pKa value of the phenolic hydroxyl group of nitrotyrosine (pKa ¼ $7.1) is significantly lower than that of tyrosine (pKa ¼ $10)), spectrophotometric properties, bulk, electron-density factors (the electron-density of the phenolic ring of nitrotyrosine is lower than that of tyrosine), and reducible to aminotyrosine. Protein nitration has extensive biological consequences such as modification of enzymatic activities, sensitivity to proteolytic degradation, impact on protein tyrosylphosphorylation, immunogenicity, and implication in disease (Abello et al., 2009). Identification of nitrotyrosine-containing proteins and accurate location of each nitrotyrosine site are key steps to understand functions and roles of tyrosine nitration (Zhan & Desiderio, 2009a). However, identification of a nitrotyrosine-containing protein is very challenging because its extreme in vivo low abundance (Haddad et al., 1994;Shigenaga et al., 1997) and various MS behaviors (Petersson et al., 2001;Sarver et al., 2001;Zhan & Desiderio, 2006Zhan, Wang, & Desiderio, 2013). MS, coupled with different chemical derivation (Zhan & Desiderio, 2009b) and enrichment techniques (Zhan & Desiderio, 2006Zhang et al., 2007), is necessary to identify a nitrotyrosine-containing protein (Zhan & Desiderio, 2009a). For mass spectrometry (MS) analyses, a characteristic photodecomposition pattern of a nitro group is present in UV-laser-based MALDI MS analysis of a nitrotyrosine-containing protein (Petersson et al., 2001;Sarver et al., 2001;Zhan & Desiderio, 2009b), but not for electrospray ionization (ESI)-MS (Kim et al., 2011;Lee et al., 2007Lee et al., , 2009bYeo et al., 2008;Zhan & Desiderio, 2009b). That photodecomposition pattern of a nitro group decreases signal intensities of a nitropeptide and complicates interpretation of a mass spectrum. However, that characteristic photodecomposition pattern can confirm the existence of a nitro group in an analyzed peptide (Zhan & Desiderio, 2009b).
Throughout this review, we clearly distinguish between endogenous in vivo nitroproteins/nitropeptides and synthetic in vitro nitroproteins/nitropeptides. Although many more publications relate to synthetic nitroproteins/nitropeptides, it is analytically much more challenging to analyze endogenous nitroproteins/nitropeptides because tyrosine nitration is an extremely low-abundance (1 in $10 6 tyrosines) oxidative-stress process. The goals of nitroprotein analysis include the identity of the nitroprotein and each site of modification. A protein usually contains several tyrosine residues, and in a population of to-be-nitrated proteins, a tyrosine site might not be stoichiometrically nitrated.
Because of the sensitivity requirement of MS analysis and the in vivo low abundance of nitrotyrosine sites, a chemical derivation and targeted enrichment prior to MS analysis is needed (Dekker et al., 2012;Freeney & Schoneich, 2013;Zhan, Wang, & Desiderio, 2013), including (i) nitrotyrosine antibodybased immunoaffinity enrichment of nitroproteins (Zhan & Desiderio, 2006;Sultana, Reed, & Butterfield, 2009) and of nitropeptides (Gusanu, Petre, & Przybylski, 2011), (ii) conversion of a nitro group to an amino group coupled with derivatization of the amino group (Tsumoto, Taguchi, & Kohda, 2010). Briefly, protection of alpha and epsilon-amino groups in a protein or peptide with 13 C 0 / 13 C 4 -or D 0 /D 6 -acetic anhydride, reduction of nitrotyrosine to aminotyrosine with sodium dithionate (also known as sodium hydrosulfite), and derivatization of aminotyrosine with 1-(6-methyl[D 0 /D 3 ] nicotinoyloxy) succinimide; (iii) conversion of a nitro group to an amino group coupled with target enrichment (Abello et al., 2010). Briefly, all amines are first acetylated, followed by conversion of nitrotyrosine to aminotyrosine, and biotinylation of aminotyrosine; (iv) reduction of the nitro group in a nitropeptide to an amino group and dansylated with dansyl chloride, followed with MS n analysis (Amoresano et al., 2007(Amoresano et al., , 2008; (v) the use of "light"-and "heavy"-labeled acetyl groups to block N-terminal and lysine residues of tryptic nitropeptides, followed with reduction of nitrotyrosine to aminotyrosine with sodium dithionite and derivatization of light-and heavy-labeled aminotyrosine peptides with either tandem mass tags (TMT) or isobaric tags for relative and absolute quantification (iTRAQ), respectively (Robinson & Evans, 2012); (vi) a new quantitative identification strategy used iTRAQ reagents to selectively label nitrotyrosine residues (not primary amines) coupled to MS analysis (Chiappetta et al., 2009); (vii) use of selective chemoprecipitation and subsequent release of tagged species (conversion of nitro group to a small 4-formylbenzylamido tag) for analysis of nitropeptides with liquid chromatography-tandem mass spectrometry (LC-MS/MS) (Prokai-Tatrai, Guo, & Prokai, 2011); and (viii) use of combined fractional diagonal chromatography (COFRADIC) (Ghesquiere et al., 2009;Larsen et al., 2011)peptide sorting is based on a hydrophilic shift after reduction of the nitro group to its amino counterpart, followed with ESI-MS (Ghesquiere et al., 2009) and MALDI-MS (Larsen et al., 2011) identification of a nitropeptide. Except for the proteomics method based on anti-nitrotyrosine antibodies and gel-based separation, multidimensional chromatography, precursor-ion scanning, and/or chemical derivatization have also emerged to identify and quantify nitroprotein and nitrotyrosine sites (Zhang, Yang, & Poschl, 2011;Freeney & Schoneich, 2013).
In-depth analyses of nitrotyrosine-containing proteins is needed to clarify biological functions and roles of tyrosine nitration, and several aspects are worth discussing here: (i) quantitative proteomics to quantify a nitrotyrosine-containing protein in a pathological condition and the degree of nitration (Robinson & Evans, 2012), (ii) use of bioinformatics to locate nitrotyrosine sites within important protein domains and motifs (Zhan & Desiderio, 2006, (iii) use of systems biology methods to clarify important protein system networks that are involved in nitroproteins (Zhan & Desiderio, 2010a;Zhan, Wang, & Desiderio, 2013), (iv) effects of local primary structure on tyrosine nitration (Seeley & Stevens, 2012), structural biology to reveal the three-dimensional crystal structure of nitrotyrosine-containing proteins to address influences of tyrosine nitration on protein functions towards development of a drug against tyrosine nitration (Palamalai & Miyagi, 2010;Zhan, Wang, & Desiderio, 2013), and (v) development of bodyfluid nitroproteomics and nitropeptidomics for discovery of body-fluid biomarkers for prediction, diagnosis, and prognosis of a disease (Zhan & Desiderio, 2010b;Zhan, Wang, & Desiderio, 2013).
The photodecomposition pattern of a nitro group was not found in an ESI-MS spectrum. TNM-nitrated angiotensin II [DRVY(-NO 2 )IHPF; MW ¼ 1090.76 Da] was analyzed with ESI-MS (Petersson et al., 2001). A mononitrated angiotension II ion ([M þ 2H] 2þ m/z 546.38, [NO 2 -Tyr]-angiotensin II) and dinitrated angiotension II ion ([M þ 2H] 2þ m/z 568.85, [(NO 2 ) 2 -Tyr]-angiotensin II) were found in the ESI-MS spectrum, but no decomposition pattern was found (Fig. 4). The doubly charged precursor ions for mononitrated angiotensin II at m/z 546.38 and for dinitrated angiotensin II at m/z 568.85 were further analyzed with MS/MS (Fig. 5); characteristic immonium ions at m/z 181.06 for mononitrated tyrosine and at m/z 226.0 for dinitrated tyrosine were found in the ESI-MS/MS spectrum (Fig. 5). The characteristic immonium ion-based precursor-ion scan spectra accurately identify a nitropeptide in complicated sample (Fig. 6). The ESI-MS behavior of a nitropeptide and the precursor-ion scans for an immonium ion at m/z 181.06 were further confirmed with ESI-MS analysis of TNM-nitrated bovine serum albumin (Petersson et al., 2001).
In the vMALDI-MS/MS analysis of synthetic peptides LE1 (Tyr-Gly-Gly-Phe-Leu; Y-G-G-F-L; 555.1818 Da), LE2 ((3-NO 2 )Tyr-Gly-Gly-Phe-Leu; (3-NO 2 )Y-G-G-F-L; 600.0909 Da), and LE3 ((3-NO 2 )Tyr-Gly-Gly-(d 5 )Phe-Leu; (3-NO 2 )Y-G-G-(d 5 )F-L; 605.1818 Da), b-and a-ions were the most-intense FIGURE 1. Generation of dityrosine and nitrotyrosine, and likely products from nitrotyrosine photochemical decomposition. Reproduced from Turko and Murad (2005), with permission from Elsevier, Inc., copyright 2005. fragment ions compared to y-ions ( Fig. 7) (Zhan & Desiderio, 2009b); those data were corroborated with MALDI-MS/MS analysis of nitrated angiotensin II (Petersson et al., 2001). Compared to the unmodified peptide (LE1), more collision energy optimized fragmentation of the nitropeptide ( Fig. 8A) but increased the intensity of the a 4 -ion and decreased the intensity of the b 4 -ion (a-ion ¼ loss of CO from a b-ion) (Fig. 8B). Furthermore, optimized laser fluence maximized fragmentation of the nitropeptide. Although MS 3 analysis confirmed the MS 2 -derived amino acid sequence, MS 3 analysis requires a higher amount of peptides relative to MS 2 (Zhan & Desiderio, 2009b). Thus, MS 3 analysis might not be suitable for routine analysis of endogenous low-abundance nitroproteins. Only when a target is determined can MS 3 be used for confirmation. To detect a nitropeptide, the amount of peptide must reach the sensitivity of a mass spectrometer; for synthetic nitropeptides, the sensitivity of vMALDI-LTQ was 1 fmol for MS detection, and 10 fmol for MS 2 detection (Zhan & Desiderio, 2009b).
The MS/MS (described above) of a nitrotyrosine-containing peptide was based on collision-induced dissociation (CID). However, the fragmentation behaviors of nitrotyrosine-containing peptides were different among CID-, electroncapture dissociation (ECD)-, electron-transfer dissociation (ETD)-, and metastable atom-activated dissociation (MAD)-MS Cook & Jackson, 2011). Those studies found that the presence of nitration did not affect the CID behavior of the peptides. For doubly charged peptides, production of ECD sequence fragments was severely inhibited with nitration; ECD of triply charged nitrotyrosine-containing peptides produced some singly charged sequence fragments. ECD of nitropeptides was characterized with multiple losses of small neutral species, including hydroxyl radicals, water, and ammonia. The origin of neutral losses was investigated with activated ion (AI) ECD.
Loss of ammonia appears to be the result of non-covalent interactions between a nitro group and protonated lysine sidechains . Further studies found that high kinetic energy helium MAD produced extensive backbone fragmentation with significant retention of post-translation modifications (PTMs). Although the high electron affinity of a nitrotyrosine moiety quenched radical chemistry and fragmentation in ECD and ETD, MAD does produce numerous backbone cleavages in the vicinity of the nitration. Compared to CID, MAD produced more fragment ions, and differentiated I/L residues in nitrated peptides. MAD induced radical-ion chemistry even in the presence of strong radical traps, and, therefore, offers unique advantages to ECD, ETD, and CID to determine nitrotyrosine-containing peptides (Cook & Jackson, 2011). Moreover, different types of CID-MS/MS have different abilities to identify nitrotyrosine-containing proteins . For the same samples, a QSTAR Elite (QTOF) with CID was used to identify 119 3NT peptides and 23 multiply nitrated 3NT peptides, whereas a dual-pressure iontrap mass spectrometer (LTQ Velos) with CID was used to identify 197 3NT peptides and 36 multiply nitrated 3NT peptides ( Fig. 9) . Therefore, the choice of an appropriate mass spectrometer is essential to analyze nitrotyrosine-containing peptides/proteins.
Several chemical derivation methods have been developed to analyze nitrotyrosine-containing peptide/protein prior to MS (Dekker et al., 2012). All the chemical derivation methods of nitrotyrosine that have been developed employ reduction of nitrotyrosine to aminotyrosine, and derivatization of the generated amino group with specific reagents. (i) Conversion of a nitrotyrosine residue to an aminotyrosine residue via reduction readily discerns aminotyrosine peptides in a background of non-nitrated peptides, and aminotyrosine peptides were more stable in a single MS mode and led to easy-to-interpret peptide mass maps (Ghesquiere et al., 2006). (ii) The use of dansyl chloride to label nitration sites followed with MS/MS plus a precursor-ion scan (Amoresano et al., 2007(Amoresano et al., , 2008. (iii) For MALDI-MS analysis of a nitropeptide, the optimum matrix was not 2,5-dihydroxybenzoic acid but sinapinic acid (Sheeley, Rubakhin, & Sweedler, 2005). (iv) A method was developed that specifically enriches nitropeptides to unambiguously identify nitrotyrosine peptides and nitration sites with LC-MS/ MS, and includes conversion of nitrotyrosine to N-thioacetylaminotyrosine, followed with high-efficiency enrichment of sulfhydryl-containing peptides with thiopropyl sepharose beads (Zhang et al., 2007). Derivation includes (a) acetylation with acetic anhydride to block all primary amines, (b) reduction of nitrotyrosine to aminotyrosine, (c) derivatization of aminotyrosine with N-succinimidyl S-acetylthioacetate, and (d) deprotection of S-acetyl on S-acetylthioacetate to form free sulfhydryl groups (Zhang et al., 2007). This method was used to study in vitro nitrated human histone H1.2, BSA, and mouse brain tissue samples (Zhang et al., 2007). (v) iTRAQ is an effective quantitative proteomics method, but is limited to primary amines. A new strategy was developed that was based on use of iTRAQ reagents coupled to MS analysis to selective label nitrotyrosine residues (Chiappetta et al., 2009) to simultaneously localize and quantify nitration sites in model proteins and biological systems (Chiappetta et al., 2009). (vi) A strategy that combined precursor isotopic labeling and isobaric tagging (cPILOT) increased the multiplexing capability to quantify a nitrotyrosine protein to 12 or 16 samples with TMT or iTRAQ, respectively. That method used light-and heavy-labeled acetyl groups to block N-termini and lysine residues of tryptic peptides. Nitrotyrosine was reduced to aminotyrosine with sodium dithionite, followed with derivatization of light-and heavy-labeled aminotyrosine peptides with either TMT or iTRAQ multiplex reagents (Robinson & Evans, 2012). This method demonstrated proof-of-principle in the analysis of in vitro nitrated BSA and mouse splenic proteins (Robinson & Evans, 2012). (vii) Improved chemical-labeling methods were designed to enrich nitrotyrosine-containing peptides independent of sequence context. In this procedure (Fig. 10), all amines were blocked with acetylation, followed by conversion of nitrotyrosine to aminotyrosine and biotinylation of aminotyrosine (Abello et al., 2010). Moreover, the entire reaction was carried out in a single buffer without any sample cleanup or pH changes to, thereby, reduce sample loss. Free biotin was removed with a strong-cation exchanger, labeled peptides were enriched with an immobilized avidin column, and enriched peptides were analyzed with LC-MS/MS (Abello et al., 2010). This method has been approved for in vitro nitrated samples (Abello et al., 2009(Abello et al., , 2010. (viii) Because of the photodecomposition of a nitro group with a MALDI UV-laser, a strategy was developed that includes (a) acetylation of N-terminal amines and epsilon-amines of lysine residues with acetic anhydride, (b) reduction of nitrotyrosine to aminotyrosine with sodium hydrosulfite, and (c) derivatization of aminotyrosine with 1-(6-methyl[d 0 /d 3 ] nicotinoyloxy) succinimide, followed with MALDI-TOF MS analysis (Tsumoto, Taguchi, & Kohda, 2010). (ix) The combined fractional diagonal chromatography (COFRADIC) approach (Ghesquiere et al., 2009;Larsen et al., 2011) was developed. Briefly, the basics of COFRADIC is reduction of nitrotyrosine to aminotyrosine with sodium dithionite; peptides are sorted with reverse-phase chromatography based on a hydrophilic shift from nitrotyrosine-containing peptide (more hydrophilic) to aminotyrosine-containing peptide (more hydrophobic) followed with ESI-MS (Ghesquiere et al., 2009) and MALDI-MS (Larsen et al., 2011) identification. COFRADIC has successfully been used to characterize tyrosine nitration in a TNM-nitrated BSA, peroxynitrite-nitrated proteome of human Jurkat cells (Ghesquiere et al., 2009;Larsen et al., 2011).
The interpretation of MS and MS/MS data of nitrotyrosinecontaining (and especially endogenous) peptides is very challenging. To avoid any risk of linking MS/MS spectra to an incorrect amino acid sequence, the combination of reduction of nitrotyrosine to aminotyrosine and use of the Peptizer algorithm to inspect MS/MS quality-related assumptions (Ghesquiere et al., 2011) has been developed. The optimal approach to determine the amino acid sequence of a nitropeptide is a manual approach (Zhan & Desiderio, 2006).

VI. BIOLOGICAL ROLES OF NITROTYROSINE-CONTAINING PROTEINS
Nitrotyrosine-containing proteins and nitrotyrosine sites that have been identified with MS/MS must be further analyzed to elucidate biological roles of tyrosine nitration. Until now, several approaches have been used to analyze roles of nitrotyrosine-containing proteins: (1) literature data-based rationalization of biological function, (2) protein domain and motif analyses, (3) systems pathway analysis, and (4) structural biology analysis. Here, nitroproteins from pituitary control and adenoma (Table 2) are used as an example to address those analyses.
1. Literature data-based rationalization of biological function: Nine nitroproteins and 10 nitrotyrosine sites, and 3 non-nitrated proteins from a human pituitary adenoma ( Table 2) were analyzed through a large number of literature data (Zhan & Desiderio, 2006). As a result, three non-nitrated proteins (glutamate receptor-interacting protein 2, ubiquitin, and interleukin 1 receptor-associated kinase-like 2) were recognized to interact with nitroproteins to form three nitroprotein-protein complexes (nitrated proteasome-ubiquitin complex, nitrated betasubunit of cAMP-dependent protein kinase (pKa) complex, and nitrated interleukin 1 family member 6-interleukin 1 receptor-interleukin 1 receptor-associated kinase-like 2 (IL1F6-IL1R-IRAK2)) (Zhan & Desiderio, 2006;Zhan, Wang, & Desiderio, 2013). Moreover, those nine nitroproteins and three nitroprotein-protein complexes were rationalized into a corresponding functional system (Fig. 13) (Zhan & Desiderio, 2006;Zhan, Wang, & Desiderio, 2013). The nitrated proteasome-ubiquitin  (2006) (3) Inflammation-related disease Lanone et al., 2002 Rectus abdominis muscle from the same control and septic patients   (Zhan & Desiderio, 2006), and control tissue (Zhan & Desiderio, 2004. Note: nY, nitrotyrosine. Modified from Zhan & Desiderio, 2004, 2006, with permission from Elsevier Science, copyright 2004, 2006 complex is an important enzymatic complex that is involved in the intracellular non-lysosomal proteolytic pathway (Zhan & Desiderio, 2006;Zhan, Wang, & Desiderio, 2013). Nitrated leukocyte immunoglobulin-like receptor subfamily A member 4 (LIRA4) might be involved in the immune system. Nitrated sphingosine-1phosphate lyase 1 (S1P lyase 1) participates in sphingolipid metabolism to regulate cell proliferation, survival, cell death, and immune system (Zhan & Desiderio, 2006;Zhan, Wang, & Desiderio, 2013). Nitrated centaurin beta 1 (CENT-b1) and nitrated cAMP-dependent protein kinase type I-beta regulatory subunit (PKAR1-b) are involved in the pKa signal pathway. IRAK-2 in the IL1-R complex and nitrated IL1-F6 are involved in the cytokine system. Nitrated zinc finger protein 432 (ZFP432) is involved in transcription regulatory systems. Nitrated Rho-GTPaseactivating protein 5 (RHOGAP5) and nitrated rhophilin 2 are involved in the GTPase signal pathway (Zhan & Desiderio, 2006;Zhan, Wang, & Desiderio, 2013). 2. Protein domain and motif analyses: Identification of protein domain/motif and location of nitrotyrosine sites into a protein domain/motif will assist in accurate elucidation of biological activities of tyrosine nitration (Zhan, Wang, & Desiderio, 2013 (Zhan & Desiderio, 2006;Zhan, Wang, & Desiderio, 2013). As a result, most nitrotyrosine sites occur within important protein domains and motifs (Zhan & Desiderio, 2006) (Fig. 14); that finding hints that tyrosine nitration alters protein functions. For example, sphingosine-1-phosphate lyase 1 (S1P lyase 1) (Fig. 14), nitrated in a human pituitary adenoma, is a key enzyme to catalyze decomposition of S1P. Two nitration sites (NO 2 -356 Y and NO 2 -366 Y) within the enzyme activity region could decrease the interaction intensity of enzyme: substrate (S1P lyase 1: S1P) to alter enzymatic activities of S1P lyase 1 (Zhan & Desiderio, 2006;Zhan, Wang, & Desiderio, 2013). 3. Systems pathway analysis: Each protein in a proteome does not work alone, but functions within a system of multiple, complex, and interacting systematic networks (Zhan & Desiderio, 2010a). It is necessary to determine effects of tyrosine nitration on those complex pathway systems networks. Ingenuity Pathway Analysis (IPA) (http://www.ingenuity.com/) and MetaCore Pathway Analysis programs (http://www.genego.com/metacore.php/) were used to elucidate signaling networks that involve nitroproteins from a human pituitary adenoma and control (Table 2) (Zhan & Desiderio, 2006;Zhan, Wang, & Desiderio, 2013). IPA pathway analyses (Zhan & Desiderio, 2010a) clearly indicated that those pituitary adenoma nitroproteins and their complexes are involved in the tumor necrosis factor (TNF) and interleukin 1 (IL1) signaling networks (Fig. 15A), which function in cancer, cell cycle, and reproductive system disease. Those control pituitary nitroproteins are involved in transforming growth factor beta 1 (TGFb1) and actin cellular skeleton signaling networks (Fig. 15B), which function in gene expression, cellular development, and connective tissue development. Both networks include a beta-estradiol signal pathway; that factor indicates that hormone metabolism is involved in a normal pituitary and pituitary adenoma. Moreover, pathway analysis showed that tyrosine nitration was involved in three important signaling pathway network systems (oxidative stress, cell-cycle dysregulation, and MAPK-signaling abnormality) in a pituitary adenoma (Zhan & Desiderio, 2010a;Zhan, Wang, & Desiderio, 2013). Those network data clarify biological roles of tyrosine nitration in pituitary tumorigenesis. 4. Structural biology analysis: Tyrosine nitration decreases the electron density of a phenolic ring of a tyrosine residue to diminish interaction intensity between enzyme and substrate or between receptor and ligand (Zhan & Desiderio, 2004). Therefore, the spatial position of a nitrotyrosine would obviously affect functions and biological roles of tyrosine nitration. The three-dimensional spatial structure of a protein determines its biological functions. If the three-dimensional spatial structure of a nitroprotein can be reconstructed from X-ray crystallography data, then it would be very easy to interpret the effect of tyrosine nitration on the 3D structure of a nitroprotein. Meanwhile, based on the 3D structure and tyrosine nitration site and domain, it is possible for one to design a small drug towards the 3D structure and domain that FIGURE 14. Nitration site and functional domains of sphingosine-1phosphate lyase 1. Reproduced from Zhan and Desiderio (2006), with permission from Elsevier Science, copyright 2006.
FIGURE 15. Significant signaling pathway networks mined from nitroproteomic dataset. A: Network was derived from adenoma nitroproteomic data, and function in cancer, cell cycle, reproductive system disease. A gray node denotes an identified nitroprotein or protein that interacts with nitroproteins. B: Network is derived from control nitroproteomic data and function in gene expression, cellular development, and connective tissue development and function. A gray node denotes an identified nitroprotein. An orange solid edge denotes a direct relationship between two nodes (molecules: proteins, genes). An orange non-solid edge denotes an indirect relationship between two nodes (molecules: proteins, genes contains tyrosine nitration (Zhan, Wang, & Desiderio, 2013). That study demonstrated that nitrated glyceraldehyde-3-phosphate dehydrogenase (GAPDH) could not bind nicotinamide adenine dinucleotide (NAD þ )-as shown with an NADþ binding assay (Palamalai & Miyagi, 2010). The X-ray crystal structure has been used to explain the effect of tyrosine nitration on the capability of NAD þ binding in GAPDH (Palamalai & Miyagi, 2010). MS analysis of nitrated GAPDH indicated that Tyr 311 and Tyr 317 were the only sites of nitration. The X-ray crystal structure revealed that the distances between Tyr 311 and Tyr 317 and the cofactor NAD þ were less than 7.2 and 3.7 A , respectively; those data imply that nitration of these two residues might affect NAD þ binding (Palamalai & Miyagi, 2010). Another example is that the X-ray crystal structure of mammalian succinate ubiquinone reductase (SQR or Complex II) is used to effectively explain association of protein tyrosine nitration (Tyr 142 ) of the Flavin subunit with the S-glutathionylated cysteine residue Cys 90 of mitochondrial complex II in a post-ischemic myocardium (Chen et al., 2008). Briefly, based on the X-ray crystal structure of SQR, the flavin subunit has a Rossman-type fold with four major domains. Tyr 142 is located in the major helix (residues 136-158) of a floating subdomain (residues 105-196). Specifically, Tyr 142 is highly surface-exposed and situated in the hydrophilic environment to suggest that this specific tyrosine is susceptible to nitration with OONO À . Moreover, Tyr 142 is $20 A away from the isoalloxazine ring of flavin adenine dinucleotide (FAD). Cys 90 is located within the part of the N-terminal beta barrel subdomain (residue 53-104) of the large FAD-binding domain. Cys 90 is near the AMP moiety of FAD ($7.7 A ), where major catalysis of electron transfer and O 2 •À production occurs. Therefore, S-glutathionylaton of Cys 90 seems likely to induce a conformational change near the floating subdomain (residues ; that change might increase the shielding effect on Tyr 142 to render Tyr 142 less accessible to OONO À oxidation (Chen et al., 2008). Therefore, the 3D structure of a protein can accurately explain effects of tyrosine nitration on that protein's structure and functions.

VII. CONCLUSIONS
Protein tyrosine nitration is an important oxidative/nitrativemediated modification that is associated with a wide range of different pathophysiological conditions (Zhan & Desiderio, 2004, 2006Zhan, Wang, & Desiderio, 2013). Also, evidence suggests the presence of a denitrase in mammalian tissues; however, a denitrase has not been isolated and its enzymatic activity not confirmed. Therefore, tyrosine nitration can be considered as reversible. Tyrosine nitration is not only a result from oxidative damage, but it also participates in pathophysiological processes (Smallwood et al., 2007). Nitration dynamically alters protein function (Mani & Moore, 2005), including activation or inactivation (Lin et al., 2007(Lin et al., , 2012Yamakura & Kawasaki, 2010). MS-based identification of nitrotyrosine-containing proteins and nitrotyrosine sites is essential to understand biological roles of this modification Spickett & Pitt, 2012;Tsikas, 2012). However, it is analytically very challenging to identify endogenous nitroproteins and nitrotyrosine sites due to nitration's low abundance in biological samples and its multiple mass spectrometric behaviors among MALDI UV-laser, ESI, CID, ECD, ETD, and MAD. Endogenous nitrotyrosine-containing proteins/peptides must be enriched prior to MS analysis. Several enrichment methods have been developed, and include immuno-affinity enrichment, biotin-affinity enrichment, and COFRADIC. Nitrotyrosine sites have been found in many different pathophysiological conditions. TMT-or iTRAQ-based quantitative nitroproteomics are needed to quantify disease-key nitroproteins/peptides. Protein domain/motif analysis, systems pathway analysis, and structural biological analysis of nitrotyrosine-containing proteins are significantly needed to elucidate the biological roles of tyrosine nitration.
Moreover, one must clearly realize that no highly reliable, high-throughput, high-sensitivity, and high-reproducibility method exists to analyze the extremely challenging endogenous tyrosine nitration in a proteome; therefore, different approaches are under development. Nitrotyrosine antibody-based immunoaffinity methods such as 2D-Westen blotting and NTAC succeeded to identify endogenous nitrotyrosine sites; however an overwhelming amount of non-nitrated tryptic peptides negatively affects characterization of nitropeptides. For that reason, we suggest development of immunoaffinity enrichment of tryptic nitropeptides-not nitroproteins-prior to MS analysis. Until now, most methods based on chemical derivatization (as described above) are used only for in vitro experiments and not for endogenous nitrotyrosine sites. Although the COFRADICbased characterization of nitropeptides succeeded in a serum proteome, throughput and sensitivity were very low, and it has not been used extensively in endogenous tissue nitroproteomes. Therefore, development of better nitrotyrosine analysis methods is necessary in the following aspects-alone or in combination: (i) derivatize a nitro to amino group to stabilize MS behaviors, (ii) develop specific amino group tags to enrich nitrotyrosine peptides, (iii) enrich nitrotyrosine-or aminotyrosine peptides is better than nitrotyrosine-or aminotyrosine proteins for sensitivity, (iv) improve liquid chromatography isolation, (v) develop super high-sensitivity mass spectrometers, (vi) choice of the appropriate ion source and collision model to fragment nitropeptide or aminopeptides, and (vii) develop reliable software for data analysis. The combined multiple aspects among items i-vii are recommended to maximize coverage of endogenous nitrotyrosine sites in a proteome.