NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
Alzate O, editor. Neuroproteomics. Boca Raton (FL): CRC Press/Taylor & Francis; 2010.
Carol E. Parker , Viorel Mocanu , Mihaela Mocanu , Nedyalka Dicheva , and Maria R. Warren .
Post-translational modification of proteins is important for the regulation of cellular processes, including the cellular localization of protein, the regulation of protein function, and protein complex formation. Post-translational modification of proteins is part of what makes proteomics so much more challenging than genomics. Not only does the proteomic “alphabet” contain more letters than the genome (21 common amino acids as opposed to four nucleotides), but these amino acids can also be modified by literally hundreds of modifications that change their molecular weights, the fundamental physical property measured by mass spectrometry. Of these modifications, the most common and naturally occurring are cleavage, acetylation, formylation, methionine oxidation, phosphorylation, ubiquitination, and glycosylation (which is a whole set of modifications rather than a single modification). A more comprehensive list of protein modifications can be found elsewhere (1). In addition, other non-post-translational modifications, such as crosslinking, fluorescent labels, and spin labels, can be used to probe protein structure and function.
Regardless of the modification there are some general points to consider. Analysis of post-translational modifications (PTMs) by mass spectrometry can be difficult and the level of difficulty is dependent on (i) the mass shift in the peptide molecular weight, (ii) the overall abundance of the modified peptide, (iii) the stability of the modification during mass spectrometry (MS) and MS/MS analysis, and (iv) the effect of the modification on the peptide’s ionization efficiency (and therefore the sensitivity).
The modified peptide must be detected. While it is possible to identify a protein (or a family of homologous proteins) from MS/MS spectra of only a few of its peptides, locating the exact site of a modification requires detection and in most cases MS/MS sequencing of the specific modified peptide. This can be difficult since not all peptides are equally detected. The “detectability” of a peptide depends on the peptide abundance and its proton affinity, which is a function of its sequence and modifications. This is illustrated in the mass spectrum of a bovine serum albumin (BSA) tryptic digest in Figure 5.6 in the Chapter 5. Although each peptide is present in the same molar abundances, there is high variability in the ion abundances, with some peptides being completely absent.
The purification method is an important factor to consider. Proteomic samples typically are analyzed from Polyacrylamide gel with sodium dodecyl sulfate-polyacrylamide gel electrophesis (SDS-PAGE) gels. Although it is possible to get high sequence coverage of highly abundant proteins from a gel, it is more common to have incomplete sequence coverage, partially due to losses during peptide extraction (2). For this reason, we prefer to use non-gel protein purification methods for modification-site determination by electrospray ionization-liquid chromatography (ESI-LC)/MS/MS, although PTM analysis from gels is sometimes the only option, and gel-based separation combined with LC-matrix-assisted laser desorption ionization (MALDI) has been successful for certain projects.
Peptide identification from a protein or translated genomic database is probability based. This is done by comparing the observed MS/MS spectrum with the theoretical spectra for the predicted proteolytic peptides (see Figure 6.1) of all proteins in the database. If more than a few modifications at a time are considered, the search time increases exponentially, and the probability score is decreased if more modifications are used in order to achieve a “match.”
Schematic of MS spectrum of a tryptic-digested protein.
Another factor critical to database searching is that the modification must be specific to only certain amino acids. Non-specific modifications, for example, modifications that can react with any amino acid, cannot be searched with some common database search engines, such as Mascot™ (3).
With some modifications attachment sites can be determined directly from the MS/ MS spectrum. This requires that the modification be stable enough to withstand the energetics of MS and/or MS/MS analysis. Also, the added mass of the modification should not shift the total mass of the peptide outside the mass range suitable for MS/ MS sequencing (~800–2500 Da). For stable modifications, not only is the peptide’s molecular weight shifted, but all fragment ions containing the modified amino acid are mass-shifted as well due to the modification (Figure 6.2).
Schematic of MS/MS fragmentation, showing shift in m/z due to peptide modification. All fragments containing this modification (m) show an increase in m/z corresponding to the mass of the modification.
The types of molecules that can modify a protein encompass a wide range of compound classes—with different physical and chemical properties. It is difficult to discuss all of even the most common modifications thoroughly so we focus here on a few that represent increasing levels of difficulty in mass spectrometric analysis. This mass shift is characteristic of the particular type of post-translational modification (e.g., acetylation, 42 Da; phosphorylation, 80 Da, etc.).
N-terminal acetylation is a common modification, and is one of the groups that block Edman sequencing of a protein. Acetylation is also common on the amino groups of lysine and arginine, and is one of the easiest post-translational modification sites to identify due to the 42 Da mass shift (resulting from the replacement of one of the hydrogens from the amine group with COCH3) on the modified amino acid. Formylation is similar to acetylation (but with a smaller, 28 Da mass shift). N-formylation is a common modification in bacterial proteins. Formylation from formic acid can also occur as an artifact during protein purification if formic acid is used.
The 42-Da mass shift due to acetylation is shown in Figure 6.3. In the corresponding MS/MS spectra, all of the fragment ions containing the acetyl group are also shifted by 42 Da. In this example, all of the “b” ions in the acetylated peptide are shifted in mass (b′), while the “y” ion series shows no change in mass. This indicates that the acetylation is on the N terminus.
A: MALDI-MS of Des-Arg bradykinin (top trace) and acetylated bradykinin (bottom trace). B: MALDI-MS/MS spectra of Des-Arg bradykinin (top trace) and acetylated bradykinin (bottom trace).
Phosphorylation is an important regulatory process involved in cell signaling and cancer. Phosphorylation site analysis presents numerous analytical challenges. First, phosphopeptides are present in low abundance, making them more susceptible to suppression by other components in the source. Second, the presence of the phosphoryl group decreases the sensitivity of the peptide—the more phosphoryl groups, the lower the sensitivity. Although this loss of sensitivity has been reported by many research groups (4–6) in both MALDI and ESI, new work by Steen et al. (7) claims that this is “mythology,” and the reason that phosphopeptides are so difficult to detect is simply due to their low copy number. Whatever the reason, the proposed approach to phosphopeptide analysis is the same—enrich the sample in phosphopeptides, and, as much as possible, prevent loss of the phosphoryl group during ionization.
To answer this question, we collected MS spectra from a set of three peptides from the kinase domain of the insulin receptor, which had zero, one, or three phosphotyrosines. As can be seen from the spectra in Figure 6.4, which were obtained from equimolar solutions with the same number of MALDI shots or the same number of nanospray spectra obtained from each sample, the MALDI spectra clearly show a decrease in the signal intensity with MALDI ionization as a function of increasing phosphorylation. There is also a decrease in sensitivity of these peptides with increasing phosphorylation in electrospray, although it is not as pronounced (Figure 6.4).
Sensitivity as a function of the number of phosphoryl groups. A: Positive ion electrospray (deconvoluted spectrum). B: Positive ion MALDI-MS (DHB matrix).
Because the phosphoryl group is somewhat labile, some modified peptides will lose their phosphoryl groups during the ionization process. Phosphotyrosine can lose HPO3, giving the unmodified amino acid; phosphoserine and phosphothreonine often undergo loss of 80 Da and/or H3P04 (98 Da), although we have observed the loss of 98 Da from phosphotyrosine-containing peptides as well. To reduce these decomposition reactions during MALDI-MS/MS analysis of phosphopeptides, a “cold” matrix, 2,5-dihydroxybenzoic, is usually used for phosphopeptide analysis, because less energy is transmitted to the analyte than when alpha-cyano-4-hydroxycinnamic acid (“alpha-cyano”), a “hot” matrix, is used (Figure 6.5).
Comparison of peak intensities in the MS spectra of a phosphopeptide mixture analyzed by MALDI-MS using (A) alpha-cyano-4-hydroxycinnamic acid and (B) dihydroxybenzoic acid (DHB) as the matrix showing the overall increase in signal intensities with DHB. (more. )
Because of the low abundance of most phosphopeptides, and the requirement for high sequence coverage, we normally use LC/MS/MS techniques to separate the peptides before analysis and to reduce suppression effects. In phosphorylation site determination by LC/MS, ion signals that are 80-Da higher than the calculated mass of a peptide indicate the addition of a phosphoryl group. MassLynx (Q-tof) and Analyst (Q-trap) software allow several different types of scanning techniques, including data-dependent triggering of MS/MS spectra and “neutral loss” scanning, which uses the loss of 98 Da (corresponding to the elements of H3P04 from phosphoserine- or phosphothreonine-containing peptides) to trigger MS/MS analysis. MS/ MS analysis can then reveal where the phosphorylation/dephosphorylation sites are located on the peptides (Figure 6.6).
MS/MS spectrum of a phosphorylated peptide showing fragments resulting from loss of 80 and 98 Da corresponding to a loss of HPO3 and H3P04.
Analytical “challenges” make phosphorylation site determination a “project” rather than a routine analysis. Our current strategy is to first perform LC/MS/MS with the ThermoFisher Orbitrap because of its high sensitivity. If the LC/MS/MS of the entire digest mixture is unsuccessful in finding the phosphorylated peptides, the second step is to use a variety of affinity purification techniques to try to increase the concentration of the phosphopeptide in the sample. These affinity techniques include the use of IMAC beads (8), anti-phosphotyrosine beads (9), or Perkin-Elmer’s “Phos-trap” titanium oxide beads, or zirconium oxide beads, followed by direct MALDI analysis of the captured phosphopeptides, or elution followed by nanoelectrospray ionization. In “direct MALDI analysis,” the beads are placed directly on the MALDI target, rather than eluting the trapped peptides from the beads prior to analysis (9).
Although there are new enrichment techniques developed each year, phosphorylation is still a challenging area of research. Unfortunately, so far, there is no “magic bullet,” and there seem to be advantages and disadvantages of each new method—some enrichment techniques seem to work better for multiply phosphorylated peptides, some for singly phosphorylated peptides, etc., with different degrees of non-specific binding of nonphosphorylated peptides. Often the same methods that trap the phosphoryl group also will trap acidic peptides.
Another strategy is to compare the spectra of phosphorylated peptides before and after treatment with phosphatase (10). Peptides which previously contained phosphoryl groups will show a decrease in molecular weight of 80 or 98 Da. Often, peptides will “appear” only after phosphatase treatment—this is especially true for multiply phosphorylated peptides, either because of their low initial sensitivities or because of their strong binding to the affinity media (Figure 6.7).
Peptide mass spectra before and after phosphatase treatment.
In conclusion, often the most critical factor in studying phosphorylation is getting enough of the purified protein in order to have enough of the low-abundance phosphoprotein. Even then, phosphorylation site determination is very dependent on the particular peptide sequence.
Sulfation of tyrosine residues is a fairly common modification, but is even more difficult to determine than phosphorylation [for a recent excellent review of tyrosine sulfation determination, see Monigatti et al. (11)]. Most of the work done on tyrosine sulfation has been through the use of radioactive 35 S, analogous to the way 32 P is used in phosphorylation studies. The sulfate modification is even more labile to ionization than phosphorylation—easily losing S03. MALDI-MS of sulfotyrosine-containing peptides can provide molecular weight information—although only in the negative ion mode. Molecular weight information can be provided by both positive and negative ion ESI or by negative ion MALDI (Figure 6.8), but localization of the sulforyl group can be difficult because this group is so labile (11). The neutral loss of SO3, even with low collision energies, leaves the peptide in its original unmodified state, making the localization of the modification difficult (11).
Negative ion MALDI MS/MS spectra of two synthetic sulfotyrosine-containing peptides. Fragment ions containing the sulfotyrosine residue are marked with an asterisk (*). (Upper boxes: full spectra; lower boxes: expanded region). Collaborator: David Klapper. (more. )
Two model tyrosine-sulforylated peptides (synthesized by the University of North Carolina Peptide Synthesis Facility as C-terminal amides) were examined with a series of MALDI matrices in the positive and negative ion modes. As expected, the negative ion mode produced “cleaner” spectra (less sensitivity to contaminants) with more abundant molecular ions compared to ions resulting by loss of the sulforyl group. With DHB as the matrix, one of the synthetic peptides (KESDsYLKNT) produced an abundant (M+H) + ion, even in the positive ion mode, but the positive ion MS/MS spectra in all matrices showed only loss of 80 Da. When “cool” matrices, such as DHB and THAP, were used in the negative ion mode, the sulforyl groups in both peptides could be localized, as is shown in Figure 6.8. Newer “gentle” ionization techniques such as ETD, IRMPD, and ECD (described in Chapter 5) may also be useful for the MS/MS analysis of sulfotyrosine-containing peptides (11,12).
Another analytical challenge is that the mass shift due to sulforylation (a shift of 80 Da due to the addition of S03) is nearly isobaric with that due to phosphorylation on all but the highest resolution mass spectrometers (monoisotopic masses: SO3 79.9568; HPO3 79.9663). The Leary group has developed a strategy using alkaline phosphatase to remove phosphorylated peptides, leaving only the sulforylated peptides (13). This research group has also pioneered the use of anti-sulfotyrosine antibodies for selective enrichment of sulfotyrosine-containing peptides (14).
Because of the strong interaction between biotin and avidin (and the even stronger interaction with streptavidin), biotinylation is often used to aid in protein purification. Although we have done “on-target” analysis of biotinylated peptides (9), the avidin-biotin interaction is so strong that in some protein purification studies, we usually perform a tryptic digestion on the affinity-bound protein. The biotinylated peptide may remain on the bead, but the protein can be identified from the other tryptic peptides. If the goal of the study is the localization of the biotinylation site, it may be removed from the avidin beads by an acidic solution containing acetonitrile and formic acid, or by “competing” it off with biotin (Figure 6.9).
MALDI-MS (A) and MALDI-MS/MS (B) mass spectrum of a biotinylated peptide. Source: Reprinted from Raska, C. S., Parker, C. E., Sunnarborg, S. W. et al. (2003) Rapid and sensitive identification of epitope-containing peptides by direct matrix-assisted laser (more. )
Nitric oxide (NO) can modify the redox state of cysteine, and NO-modified cysteine can be considered an intermediate between reduced cysteine (Cys-S-H) and oxidized cysteine (Cys-S-S-Cys). Thus NO can affect protein folding, regulating protein activity by inducing the formation of disulfide bonds (15). S-nitrosylation of cysteine is usually not studied directly, as are other modifications. Instead, the NO moiety is replaced with a His tag (15) or a biotin tag (16) to facilitate purification and detection (Figure 6.10). Using these techniques, S-nitrosylation of cysteines has been found to be important for signal transduction, including initiation of apoptosis (17). It is now thought that S-nitrosylation may be as important as phosphorylation for signaling.
MS/MS spectrum of a nitrosylated peptide after a biotin switch and the characteristic biotin fragment ions at m/z 429 and 430. Source: Reprinted from Whalen, E. J., Foster, M. W., Matsumoto, A., Ozawa, K., Violin, J. D., Que, L. G., Nelson, C. D., Benhar, (more. )
Ubiquitin is an 8800 Da protein that attaches to lysine residues in the target protein, and may target the protein for degradation. Ubiquitination plays a role in cell division, cell death, and signal transduction, and has been found to be an important factor in synaptic remodeling (18). From an analytical perspective, digestion of a ubiquitinated protein with trypsin cleaves the ubiquitin chain, leaving a residual GG tag on the modified lysine (complete cleavage) or an LRGG tag (incomplete cleavage), as in Figure 6.11. The resulting mass shifts can be used to identify the modified lysine residue. Because of the transient nature of ubiquitinated proteins, ubiquitination is very difficult to detect and identify unless you have large amounts of material. For a more complete discussion of the mass spectrometric determination of ubiquitination, we refer the readers to reference (19) and the references cited therein.
MS/MS fragmentation of a model (A) GG- and (B) LRGG-tagged peptide.
Most of the above examples are modifications that remain intact and remain on the amino acid while the peptide backbone fragments, or are labile and fall off completely during ionization or during collisional-induced dissociation (CID). Figure 6.12 shows an example where the modification and the backbone both fragment under MS/MS conditions, making interpretation of the spectrum much more difficult.
(A) MALDI-MS of a tryptic digest of unmodified synuclein. (B)MALDI-MS of a tryptic digest of dye-modified synuclein. (C) MS/MS of dye-modified tryptic peptide. Collaborators: Gary Pielak and Rebecca Ruf, unpublished results.
Of all of the post-translational modifications, glycosylation is one of the most challenging because of the variability in the attached glycans, and the isobaric nature of many of these glycans. Often, the glycans are removed by glycosidases and their branch structure is determined separately from the protein. For proteomics, the opposite approach is often used—the glycans (which can shift the peptide molecular weights out of the range for peptide sequencing) are enzymatically removed, and the deglycosylated protein is studied separately.
Glycosylation is an extreme example of a case where the modification undergoes fragmentation. Adding to the analytical challenge is that glycosylation is not merely a single modification, but is often a set of modifications in which a single glycosylated site on a peptide may have a number of glycan isoforms with different chain lengths and branches attached. Because of this heterogeneity, there is no single mass shift associated with glycosylation, unlike other modifications. Glycosylation is variable, with different arrangements of glycans and branch structures (high mannose, hybrid, complex) (Figure 6.13). In addition, not only the branch structure but the extent of glycosylation at a particular site can vary as well.
Types of N-glycan structures (sialic acid residues are not shown).
What adds to the analytical challenge is that intact glycans also are high-molecular weight modifications (as much as several thousand Da), which can shift the molecular weights of the peptides out of the mass region where they can be sequenced. Also, because of the heterogeneity of glycan structures, these modified peptides can appear as “humps” on the baseline rather than as discrete peaks. Often, glycosylated peptides simply do not appear in the peptide digest spectra at all. Glycosylated peptides are also difficult to extract from PAGE gels, a standard purification technique used in proteomics. The consensus sequence for N-glycosylation is NxS or NxT (where x ≠ P), so potential glycosylation sites can be predicted from the sequence. There is no corresponding consensus sequence for O-glycosylation.
There are three basic approaches to the study of protein glycosylation. The first is to leave the glycosylation on the protein. This approach can define the site of attachment and determine the type of glycosylation at each site (i.e., high mannose, hybrid, or complex), even though the exact branch structure cannot be determined. The analytical approach to determine the class of N-linked glycosylation is shown in Figure 6.14 (20). In this approach, the terminal sialic acid groups are first removed by neuramidase. After determination of the type of branch structure by MALDI-MS, the N-linked glycans can be enzymatically removed from glycopeptides with PNGaseF, and the amino acid sequence can be determined by MALDI-MS/MS or by nanoESI. The PNGase cleaves the glycan from the asparagine, releasing the glycan and converting the asparagine to aspartic acid, resulting in a shift of +1 Da from the Mw of the native peptide.
Analytical scheme for the analysis of the type of glycan attached to particular NxS or NxT sites. Source: Reprinted from Zhu, X., Borchers, C, Bienstock, R. J. and Tomer, K. B. (2000) Mass spectrometric characterization of the glycosylation pattern of (more. )
The type of glycan structure on a particular peptide (site-specific determination of glycosylation) can be inferred from the masses observed during MS/MS analysis. A series of fragment ions 162 Da apart in the MS/MS spectra indicates the presence of mannose moieties. Similarly, loss of an N-acetyl glucosamine residue ("GlcNac") leads to a mass difference of 203.08 Da, while the loss of a fucose causes a Mw shift of –146 Da, loss of a sialic acid (–291 Da), and loss of GlcNac-Gal (–365 Da). Obviously, this can get complicated, especially since more than a single type of glycan or branch structure can be present on a particular peptide, so that these patterns of losses overlap each other. For a fairly simple example of a peptide that contains a high-mannose N-linked glycan, see Figure 6.15.
ESI-MS/MS spectrum of a glycosylated peptide of at (M+3H)+3 = 1103.4 Da, showing losses of mannose and GlcNac moeties. MaxEnt software was used to transform the multiply-charged peptides to their corresponding +1 forms. Source: Reprinted from Zhu, X., (more. )
The second approach to the study of protein glycosylation is to remove the glycans, and to determine their structures separately. This results in an average picture of the glycosylation on the protein, but not the details of where each type of the glycosylation was located on the protein. An example of this type of study is shown in Figure 6.16.
MALDI-MS of permethylated human serum N-glycans, analyzed after removing them from the proteins. Source: Reprinted from Morelle, W., Canis, K., Chirat, R, Faid, V. A. and Michalski, J.-C. (2006) The use of mass spectrometry for the proteomic analysis (more. )
Several mass spectrometrists have specialized in the area of determination of the glycan structure on particular peptides—notably the research groups of Orlando, Harvey, and Morelle—and the reader is referred to their research and review articles for a more detailed description of the analysis of glycoproteins (21–28).
In this book chapter we have described mass spectrometric approaches to the study of protein modification. This book chapter is intended only as a broad overview of the use of mass spectrometry for the study of post-translational modifications— for more detailed studies, the reader is referred to review articles and current literature on specific modification types. New developments in instrumentation (as discussed in the Chapter 5) will have a profound effect on this rapidly changing area of research.
Many of the examples shown here were acquired at the UNC-Duke Proteomics Center, which was partially funded by an anonymous gift in honor of Michael Hooker. We would also like to acknowledge support from the Specialized Cooperative Center for Reproductive and Infertility Research (2 U54 HD35041–11), and from the UNC Lineberger Cancer Center (5 P30 CA16806–31).
Krishna R. G., Wold F. Identification of common post-translational modifications. In: Creighton T. E., editor. Protein structure: A practical approach. Oxford: Oxford University Press; 1997. pp. 91–116.
Larsen M. R., Sorenson G. L., Fey S. J., Larsen P. M., Roepstorff P. Phospho-proteomics: Evaluation of the use of enzymatic de-phosphorylation and differential mass spectrometric peptide mass mapping for site specific phosphorylation assignment in proteins separated by gel electrophoresis. Proteomics. 2001; 1 :223–8. [PubMed : 11680869 ]
Bonewald L. F., Bibbs L., Kates S. A., et al. Study on the synthesis and characterization of peptides containing phosphorylated tyrosine. Journal of Peptide Research. 1999; 53 :161–9. [PubMed : 10195453 ]
Janek K., Wenschuh H., Bienert M., Krause E. Phosphopeptide analysis by positive and negative ion matrix-assisted laser desorption/ionization mass spectrometry. Rapid Communications in Mass Spectrometry. 2001; 15 :1593–9. [PubMed : 11544598 ]
Lee K. A., Craven K. B., Niemi G. A., Hurley J. B. Mass spectrometric analysis of the kinetics of in vivo rhodopsin phosphorylation. Protein Science: A Publication of the Protein Society. 2002; 11 :862–74. [PMC free article : PMC2373533 ] [PubMed : 11910029 ]
Steen H., Jebanathirajah J. A., Rush J., Morrice N., Kirschner M. W. Phosphorylation analysis by mass spectrometry. Myths, facts, and the consequences for qualitative and quantitative measurements. Molecular and Cellular Proteomics. 2006; 5 :172–81. [PubMed : 16204703 ]
Raska C. S., Parker C. E., Dominski Z., et al. Direct MALDI-MS/MS of phosphopeptides affinity-bound to immobilized metal ion affinity chromatography beads. Anal Chem. 2002; 74 :3429–33. [PubMed : 12139050 ]
Raska C. S., Parker C. E., Sunnarborg S. W., et al. Rapid and sensitive identification of epitope-containing peptides by direct matrix-assisted laser desorption/ionization tandem mass spectrometry of peptides affinity-bound to antibody beads. J Am Soc Mass Spectrom. 2003; 14 :1076–85. [PubMed : 14530088 ]
Torres M. P., Thapar R., Marzluff W. F., Borchers C. H. Phosphatasedirected phosphorylation-site determination: a synthesis of methods for the detection and identification of phosphopeptides. Journal of Proteome Research. 2005; 4 :1628–35. [PubMed : 16212415 ]
Monigatti F., Hekking B., Steen H. Protein sulfation analysis—A primer. Biochimica et Biophysica Acta, Proteins and Proteomics. 2006; 1764 :1904–13. [PubMed : 16952486 ]
Mikesh L. M., Ueberheide B., Chi A., et al. The utility of ETD mass spectrometry in proteomic analysis. Biochimica et Biophysica Acta. 2006; 1764 :1811–22. [PMC free article : PMC1853258 ] [PubMed : 17118725 ]
Yu Y., Hoffhines A. J., Moore K. L., Leary J. A. Determination of the sites of tyrosine O-sulfation in peptides and proteins. Nature Methods. 2007; 4 :583–88. [PubMed : 17558413 ]
Hoffhines A. J., Damoc E., Bridges K. G., Leary J. A., Moore K. L. Detection and purification of tyrosine-sulfated proteins using a novel anti-sulfotyrosine monoclonal antibody. Journal of Biological Chemistry. 2006; 281 :37877–87. [PMC free article : PMC1764208 ] [PubMed : 17046811 ]
Camerini S., Polci M. L., Restuccia U., et al. A novel approach to identify proteins modified by nitric oxide: The HIS-TAG switch method. Journal of Proteome Research. 2007; 6 :3224–31. [PubMed : 17629318 ]
Forrester M. T., Foster M. W., Stamler J. S. Assessment and application of the biotin switch technique for examining protein S-nitrosylation under conditions of pharmacologically induced oxidative stress. Journal of Biological Chemistry. 2007; 282 :13977–83. [PubMed : 17376775 ]
Jaffrey S. R., Erdjument-Bromage H., Ferris C. D., Tempst P., Snyder S. H. Protein S-nitrosylation: A physiological signal for neuronal nitric oxide. Nat Cell Biol. 2001; 3 :193–7. [PubMed : 11175752 ]
Ehlers M. D. Activity level controls postsynaptic composition and signaling via the ubiquitin-proteasome system. Nat Neurosci. 2003; 6 :231–42. [PubMed : 12577062 ]
Parker C. E., Mocanu V., Warren M. R., Greer S. F., Borchers C. H. Mass spectrometric determination of protein ubiquitination. In: Patterson W. C., Cyr D. M., editors. Ubiquitin-proteosome protocols. Totowa, NJ: Humana Press; 2005. pp. 153–73. [PubMed : 15917631 ]
Zhu X., Borchers C., Bienstock R. J., Tomer K. B. Mass spectrometric characterization of the glycosylation pattern of HIV-gp120 expressed in CHO cells. Biochemistry. 2000; 39 :11194–204. [PubMed : 10985765 ]
Atwood J. A., Cheng L., Alvarez-Manilla G., et al. Quantitation by isobaric labeling: Applications to glycomics. Journal of Proteome Research. 2008; 7 :367–74. [PubMed : 18047270 ]
Colangelo J., Orlando R. On-target exoglycosidase digestions/MALDI-MS for determining the primary structures of carbohydrate chains. Analytical Chemistry. 1999; 71 :1479–82. [PubMed : 10204046 ]
Harvey D. J. Proteomic analysis of glycosylation: Structural determination of N- and O-linked glycans by mass spectrometry. Expert Review of Proteomics. 2005; 2 :87–101. [PubMed : 15966855 ]
Harvey D. J. Structural determination of N-linked glycans by matrix-assisted laser desorption/ionization and electrospray ionization mass spectrometry. Proteomics. 2005; 5 :1774–86. [PubMed : 15832364 ]
Harvey D. J., Royle L., Radcliffe C. M., Rudd P. M., Dwek R. A. Structural and quantitative analysis of N-linked glycans by matrix-assisted laser desorption ionization and negative ion nanospray mass spectrometry. Analytical Biochemistry. 2008; 376 :44–60. [PubMed : 18294950 ]
Morelle W., Canis K., Chirat F., Faid V. A., Michalski J.-C. The use of mass spectrometry for the proteomic analysis of glycosylation. Proteomics. 2006; 6 :3993–4015. [PubMed : 16786490 ]
Morelle W., Michalski J.-C. Analysis of protein glycosylation by mass spectrometry. Nature Protocols. 2007; 2 :1585–1602. [PubMed : 17585300 ]
Yang Y., Orlando R. Identifying the glycosylation sites and site-specific carbohydrate heterogeneity of glycoproteins by matrix-assisted laser desorption/ionization mass spectrometry. Rapid Communications in Mass Spectrometry. 1996; 10 :932–36. [PubMed : 8777327 ]
