Biochimica et Biophysica Acta 1764 (2006) 1870 – 1880 www.elsevier.com/locate/bbapap
Review
Global methods for protein glycosylation analysis by mass spectrometry Bogdan A. Budnik a,1 , Richard S. Lee b,1 , Judith A.J. Steen c,⁎ a
Department of Pathology, Children’s Hospital Boston and Harvard Medical School, Boston, MA 02115, USA Department of Urology, Children’s Hospital Boston and Harvard Medical School, Boston, MA 02115, USA Department of Neurology, Children’s Hospital Boston and Harvard Medical School, Boston, MA 02115, USA
b c
Received 23 June 2006; received in revised form 1 October 2006; accepted 10 October 2006 Available online 18 October 2006
Abstract Mass spectrometry has been an analytical tool of choice for glycosylation analysis of individual proteins. Over the last 5 years several previously and newly developed mass spectrometry methods have been extended to global glycoprotein studies. In this review we discuss the importance of these global studies and the advances that have been made in enrichment analyses and fragmentation methods. We also briefly describe relevant sample preparation methods that have been used for the analysis of a single glycoprotein that could be extrapolated to global studies. Finally this review covers aspects of improvements and advances on the instrument front which are important to future global glycoproteomic studies. © 2006 Elsevier B.V. All rights reserved. Keywords: Global glycosylation analysis; Enrichment analysis; Chemical derivatization; Mass spectrometric methods of analysis; Precursor ion scanning; Electron capture dissociation; Electron transfer dissociation
1. Introduction A focus on global glycoprotein analysis has recently resulted in technical advances in the field. The application of these techniques could generate basic functional knowledge in glycobiology and yield many candidate biomarkers for several diseases including cancer. The major focus of these global glycosylation studies has been to find biomarkers [1] and to map glycosylation sites [2–7]. The mapping of glycosylation sites and the structures of the glycans may yield valuable information about glycosylation consensus sequences. Recent advances in other “omics” fields have created the potential to identify new candidate biomarkers of disease that may provide the needed information to improve diagnosis and staging/ grading of disease; discover new potential therapeutic targets; and provide more accurate prognostic information for patients [8]. Although there have been a high number of publications ⁎ Corresponding author. Children's Hospital Boston, Division of Neuroscience, Enders 307, 320 Longwood Ave, Boston, MA 02115, USA. Tel.: +1 617 919 2450; fax: +1 617 730 0243. E-mail address:
[email protected] (J.A.J. Steen). 1 Both authors contributed equally to this work. 1570-9639/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.bbapap.2006.10.005
about candidate markers, there has been a striking decrease in the number of US Food and Drug Administration (FDA) approved biomarkers since 2003 [8]. Refocusing the analysis of a global proteome to subproteomes may provide the necessary direction to produce more clinically relevant biomarkers of disease. One particular sub-proteome is the glycoproteome. Studying changes of global proteome may lead to the identification of clinically useful biomarkers and therapeutic targets of disease. Glycoproteins compromise approximately 25% of the currently available cancer biomarkers approved by the FDA [8]. A systems biology approach to the glycoproteome of various diseases or organisms may provide the necessary insight into the variations in glycosylation that are associated with varying degrees of pathogenesis. For instance, specific alterations in the glycans of currently available testicular tumor markers, such as αfetoprotein or human chorionic gonadotropin-β may result in improved clinical diagnostic and/or prognostic capability and lead to new insights into the pathogenesis of certain grades of the disease. Overall, glycosylated proteins represent the majority of cell surface markers and secreted proteins [9]. It is estimated that 50–60% of proteins in the human body are modified by
B.A. Budnik et al. / Biochimica et Biophysica Acta 1764 (2006) 1870–1880
1871
glycosylation [9–11]. Glycans are critical in determining a protein's stability, conformation, cellular signaling, and binding affinity for other molecules or glycan binding proteins [12,13]. Additionally, glycosylation has been implicated in numerous biological processes including cell growth and developmental biology, immune response, tumor growth, metastasis, anticoagulation, cell to cell communication, and microbial pathogenesis [14–25]. However, characterizing the glycoproteome is challenging because of the inherent heterogeneous and diverse nature of glycans and the complex nature of this modification [26]. The complete characterization of the glycoproteome requires the identification and analysis of the glycan structure; the protein expressing the glycan; and the protein that binds the glycan [27]. The glycans are not static, they may vary temporally with disease, and may be responsible for modulating as opposed to initiating a biological function as in a direct protein–protein interaction [26]. Interestingly, variations in the concentration of a single glycosyltransferase may change the degree of glycosylation of multiple proteins [28]. Global methods using mass spectrometry to study glycosylation may be a key technology in unraveling the biological function and intricacies of glycosylation. However, the study of protein glycosylation by mass spectrometry has its challenges. First, glycosylations can be highly labile, especially if the glycosylation is bound to threonine or serine residues, where the glycan is connected to the protein by an acetal bond. This lability of glycan complicates analyses and reproducibility. Second, glycosylated peptides have been described as exhibiting poor ionization efficiencies. Third, glycan modifications larger than a single sugar residue are extremely heterogeneous and are not particularly amenable to current bioinformatics platforms. The heterogeneity of glycans further complicates the analysis, as a single peptide may be modified by several forms of glycans, each having a different structure and mass, and to complicate things further, sometimes a different structure can also have the same mass. Although there are many obstacles to interrogating the glycoproteome, recent advances in sample preparation and mass spectrometry have demonstrated the ability to isolate and identify both the peptide backbone and glycan. Several groups in the last few years have used various analytical techniques in order to overcome some of the above described problems associated with the analysis of glycoproteins and glycopeptides. In this review we will cover the global approaches used for glycoprotein and glycopeptide analysis including enrichment analysis, chemical derivatization and mass spectrometry instrument based methods that aid the analysis of glycoproteins.
fluids span up to 10 to 12 orders of magnitude [29]. Some modified proteins are signaling proteins which are found in low abundance and as such are not selected for analysis in a complex mixture with a dynamic range greater than four orders of magnitude. These problems can be bypassed if the enrichment analysis for glycopeptides is performed. Enriching for glycoproteins by mass spectrometry has three significant advantages: (1) the most abundant unmodified proteins are excluded from the analyses such that low abundance glycoproteins are analyzed; (2) the glycopeptides do not have to compete with unmodified proteins for charge carriers during ionization thereby improving ionization efficiencies and increasing the likelihood of detecting the modified peptide; (3) instrument parameters can be optimized for glycopeptide analysis, such that labile glycans are not lost during ionization and the fragmentation methods and conditions are optimized for glycan analysis. For example by using low voltages in the interface region, glycans from O-glycosylated peptides will remain intact on the peptide backbone during analysis. In addition, if the goal of an experiment is to obtain peptide and glycan sequence information, CID (collision induced dissociation) fragmentation energies for glycopeptides are often at least 10 eV higher that that for unmodified peptides [30]. This is because the primary fragmentation process is glycan fragmentation when using CID for glycopeptides, whereas the peptide backbone fragmentation is a secondary process that has to be induced by increased collision energies. The large glycan structures are especially difficult to fragment as the branched structure has numerous degrees of freedom and much of the energy applied to fragment the glycopeptide will be absorbed by this glycan structure. However, it must be noted that if one attempts to analyze unmodified proteins along with the glycoproteins, these high energy conditions do not necessarily provide good sequence information for unmodified peptides.
2. Challenges of analyzing glycosylation by mass spectrometry
Lectin affinity chromatography for glycoprotein isolation has been used widely for several years [38]. Lectins are proteins that specifically interact with carbohydrates without modifying them. They generally interact with specific motifs in a glycan and the structural domain which interacts with the glycan varies between lectins. There are several different commercially available lectins, that can be used to selectively enrich for particular subsets of glycoproteins from a complex protein
The biggest challenge faced in the analysis of protein modifications is the dynamic range of protein concentrations. Most mass spectrometers can only analyze samples in which the concentrations of proteins cover three to four orders of magnitude, whereas the protein concentrations of biological
3. Enrichment methods Several recent studies highlight the success of enrichment analysis strategies using affinity chromatography methods to enrich proteins or peptides with a particular modification [31–33]. The glycan moiety can be used as a handle for affinity purification to enrich for glycosylated peptides as they have distinct properties, which can be used to specifically select the modified proteins/peptides using structural aspects (e.g. antibodies and lectins) [34,35] or chemical properties (e.g. derivatization methods) [36,37]. 3.1. Lectin affinity purification
1872
B.A. Budnik et al. / Biochimica et Biophysica Acta 1764 (2006) 1870–1880
mixture [39]. The enriched sample can then be preferentially analyzed. Lectin affinity purification has been applied to single proteins [40] and to global glycoproteomes. For example, in a human plasma sample, concanavalin A (ConA) lectin purification was followed by hydrophilic interaction liquid chromatography (HILIC) and partial deglycosylation to determine the sites of N-glycosylation [9]. Other recent large scale proteomic studies have used the specificity of lectins to study particular subsets of the glycoproteome [41–45]. The broad range of specificity found amongst lectins can be advantageous. A lectin with high specificity can be used to select a small subset of the glycoproteome, for example PNA — Arachis hypogaea agglutinin (peanut) is specific to glycans containing β-Gal, and DSA—Datura stramonium agglutinin (Jimson weed, thornapple) is specific to glycans containing GlcNAc residues. Conversely, less specific lectins i.e. general selectivity lectins can be used to explore larger portions of the glycoproteome using a “glyco-catch” approach. For example, the soybean lectin has been used for the identification of 22 periplasmic glycoproteins from Gram-negative bacteria [46]. In a similar manner, the glycoproteome of Caenorhabditis elegans was studied by enriching the proteolytic digest using both ConA and Galectin LEC6 (GaL6) lectin columns. The ConA column was used to capture high-mannose N-type glycans from digested sample. The flow-through from this column was then applied to a GaL6 column which selects for peptides with complex-type N-glycans. This type of approach was applied to study the extracellular domains from membrane proteins on mammalian cells [47]. This serial lectin affinity chromatography (SLAC) [48] strategy has been used to enrich for several glycan structures [49]. Coupling lectin affinity selection with recent advances in stable isotopic labeling, Regnier et al. successfully performed a comparative analysis of sialylated proteins [50]. This type of comparative study in glycoproteomics is a valuable tool for exploring the glycosylation sites of the whole proteome as well as a good tool for biomarker discovery, complementing the traditional approach using differential staining in 2D gel electrophoresis [51] and isotope dilution methods [52]. 3.2. Chemical derivatization methods Apart from affinity purification techniques using lectins, which do not change the structure of the modification and the peptide/proteins, several chemical/biochemical approaches for the detection and affinity purification of glycosylated proteins have been used. The reactions chosen for chemical derivatization must be specific to the group/modification of interest, in this case the glycan. Most of the strategies use two basic reactions: the Schiff-base reaction of aldehydes with a hydrazine [30,53– 55] or a Staudinger ligation between a phosphine and an azide [5,56]. Various combinations of these approaches afford specificity to the method, whereby subsets of the glycosylated proteins are enriched. Most of these derivatization methods tend to provide information about the peptide/protein identity that is modified without providing much information about the site of
glycosylation or the structure of glycosylation. This is because of inadequate search algorithms and because some of these methods greatly modify the glycan structure. 3.2.1. O-GlcNAc ketone enrichment One example of using the Schiff-base reaction to selectively enrich for particular carbohydrate moieties is O-GlcNAc ketone enrichment method developed by Hsieh-Wilson [3]. The OGlcNAc modification, an O-linked modification of serine and threonine amino acids of a single β-N-acetyl-glucosamine moiety has been implicated in different signaling pathways [57] and as such there is much interest in developing global proteomic methods for the analysis of O-GlcNAc modified proteins. Hsieh-Wilson et al. described a chemo-enzymatic method using an engineered β-1,4-galactosyl transferase (GalT) to transfer a ketone containing substrate onto proteins modified by the O-GlcNAc modification (see Fig. 1). The ketones were biotinylated with biotin-hydrazine (aminooxy biotin) in a Schiff-base reaction and subsequently the O-GlcNAcylated peptides/proteins were captured on a streptavidin affinity column. The biotinylated glycoconjugate protein can also be detected by chemiluminescence using streptavidin conjugated to horseradish peroxidase (HRP) [55]. The success of the strategy was validated by identifying the cAMP-responsive element binding factor (CREB), a low abundance protein with two known OGlcNAc sites, in a whole cell lysate [58]. In addition to identifying CREB, several other O-GlcNAc modified proteins were also identified. 3.2.2. Staudinger ligation A chemical reaction that is selective and specific to a single chemical moiety in the presence of complex biological mixture is powerful tool for biologically important modifications. An example of such reaction is the Staudinger reaction (a reaction of an azide with a phosphine). The standard Staudinger reaction was modified by the Bertozzi group [56]. Simply, in the modified reaction the intermediate aza-ylide formed in a standard Staudinger reaction reacts with an electrophilic trap to form an amide bond with a compound that is biotin tagged. The reaction is biologically unique as neither phosphines nor azides occur in any known biomolecules. Additional features, include the possibility of designing phosphines to incorporate a wide variety of tags including fluorescent probes and affinity tags such as biotin or FLAG tags [59,60]. The tagging method was also extending to other modifications such as farnesylation [61]. The strategy for detection and isolation of post-translationally modified proteins based on a tag attached to the modification substrate is referred to as tagging-via-substrate or the TAS technology [61]. TAS technology was later extended to the detection and identification of O-GlcNAc modification (see Fig. 2). Zhao et al. demonstrated that an unnatural peracetylated azido-GlcNAc substrate was incorporated into proteins with OGlcNAc modifications in several cell lines derived from diverse organisms, suggesting the enzymatic systems in these cell lines are compatible with the azido modification in the substrate [5].
B.A. Budnik et al. / Biochimica et Biophysica Acta 1764 (2006) 1870–1880
1873
Fig. 1. A chemo-enzymatic strategy for identifying O-GlcNAc-glycosylated proteins from cellular lysates. Several steps involved in the labeling process are indicated by the numbers.
O-GlcNAc modified proteins were enriched from cytosolic lysates, using the Staudinger capture reaction and the biotinylated phosphine capture reagent (bPPCR). This experiment was performed on metabolically labeled S2 cells (D. melanogaster cell line) and led to the identification of 51 OGlcNAc modified proteins, including 10 proteins that were previously identified [5]. Although this method was able to identify O-GlcNAc modified peptides, information pertaining to the site of the modification on the peptide backbone could not
always be provided because the modification is highly labile and CID fragmentation usually results in the loss of the modification without leaving a trace of the modification. Another potential disadvantage is that the effect of azido sugars on the metabolism of cells during growth has not been characterized. Despite the minor disadvantages the method poses, this approach permits the identification of several OGlcNAc modified proteins. Recently this approach was extended to a global study in HeLa cells where the identification
Fig. 2. A strategy for the global detection of O-GlcNAc proteins, showing a schematic representation of TAS technology. (A) In vivo labeling of the proteins with azido-GlcNAc, (B) a Staudinger ligation of an biotin phosphine tag to the azido-GlcNAc. (1) Peracetylated N-(2-azidoacetyl)glucosamine is metabolically converted to (2) UDP-azido-GlcNAc. The UDP-azido-GlcNAc is transferred to a protein by the enzyme O-GlcNAc transferase. (3) A phosphine biotin reagent reacts with the azido sugar on the glycosylated protein. Labeled proteins can be enriched using streptavidin.
1874
B.A. Budnik et al. / Biochimica et Biophysica Acta 1764 (2006) 1870–1880
of 199 putatively O-GlcNAcylated was documented [4]. Several function groups of proteins were found to be modified indicating the pervasiveness of this modification (see Table 1). In addition, the table shows the importance of these global glycoprotein studies, as previously this type of data was collected protein by protein and careful biochemistry experiments were used to determine that a protein was glycosylated. 3.2.3. Periodate-acid-Schiff coupled affinity (PAS) chromatography The periodate oxidation reaction utilizes the vicinal diol functionality, which is unique to glycans. A mild periodate oxidation reaction oxidizes these diols to aldehydes without affecting any other amino acid apart from methionine, which is oxidized under those conditions to the sulfoxide. The application of PAS to enrich for the glycoproteome was first reported using an iminobiotin hydrazide in the Schiff base reaction, the derivatized peptides/proteins were then affinity purified on a streptavidin column, and analyzed using mass spectrometry [53]. The periodate oxidation and coupling to a hydrazine via the Schiff-base reaction is an extremely versatile process, which can be used in a variety of ways using different coupling agents such as biotin hydrazides, digoxigenin hydrazides or even a simple hydrazide column. In addition the reaction may be performed on proteins or on peptides. Similar strategies were later used, whereby the glycopeptides were oxidized and captured directly on a hydrazine column [54]. The N-linked glycopeptides were eluted from the column by treating the column with PNGase F. This experiment introduced quantitation to the method by comparing two different samples using isotopically labeled water for the reaction. One sample was labeled with 18O and the other sample was labeled with 16O, the two samples were then mixed and analyzed. The major problem associated with the PAS strategy is that the glycan structures can be heterogeneously modified with an undefined number of hydrazide tags and the only way of sequencing the peptide backbone is by cleaving the glycan from the peptide using PNGase. Thus all information pertaining to the glycan structure is lost and only N-glycosylation sites can be determined in this fashion. Despite these downfalls these experiments do provide Table 1 Summary of the 199 proteins identified from HeLa cells by affinity purification, thought to carry the O-GlcNAc modification [55] Protein function
Percentage from identified
Protein synthesis/folding/turnover Cell growth/maintenance RNA processing/transcription Carbohydrate metabolism Unknown function Intracellular transport General metabolism Nucleotide metabolism Amino acid metabolism DNA damage response Muscle development
49 41 24 23 17 13 11 8 7 5 1
The table shows the pervasiveness of the O-GlcNAc modification in different cellular processes, which was hitherto unknown.
important information about which N-glycosylation sites are modified. The approach has been used for high throughput quantitative analyses of serumglycoproteins from cancerbearing mice relative to normal mice [62] and in a proteome screening platform to find relevant biomarkers from serum [1]. 3.2.4. β-elimination and Michael addition Another derivatization method that has been used to enrich for O-linked β-N-acetylglucosamine is β-elimination followed by Michael addition with dithiothreitol (BEMAD) [7]. This method was validated by mapping three previously known O-GlcNAc sites on Synapsin 1 and was used to map three new modification sites on the same protein and the method was also used to discover several sites on the nuclear pore complex. BEMAD has also been used in a quantitative approach, where two samples were compared by using deuterated (d6) DTT for one sample and unlabeled (d0) DTT for another sample [63]. This study by Vosseller et al. shows the utility of the method to quantitate both O-glycosylated and O-phosphorylated peptides. In these studies, to distinguish between the O-glycosylated vs. O-phosphorylated samples the peptides were (1) enzymatically dephosphorylated to quantitate the O-glycosylation sites or (2) subjected to chemical hydrolysis to remove the O-GlcNAc residues when quantitating the phosphorylation sites. Both these treatments were followed by the BEMAD derivatization procedure. 4. Mass spectrometry sample preparation methods for global glycosylation analysis In this section, we discuss various methods in sample preparation that could also be applied to sample preparation for the global analysis of glycopeptides and proteins by mass spectrometry. Most of these methods have been applied to single glycoproteins but these techniques in conjunction with current instrumentation may be considered as potential methods for a global approach. Analyzing the glycosylation of the proteins is challenging because it requires the determination of the carbohydrate structure (sequence, branching points, and heterogeneity) and the site of attachment to the protein backbone. Currently, two ionization procedures are mainly used for the characterization of carbohydrates by mass spectrometry, electrospray ionization (ESI) and matrix assisted laser desorption/ionization (MALDI) [64,65]. MALDI mass spectrometry has been commonly used to determine structures of glycoproteins and carbohydrates cleaved from glycoproteins. For example, glycoforms of small glycoproteins with single glycosylation sites, or with limited numbers of glycosylation sites can often be resolved by MALDI mass spectrometry coupled with TOF instruments [66,67]. To analyze larger proteins that have complex carbohydrate modifications by MALDI, the proteins must be cleaved into smaller fragments by enzymatic digestion and the glycans have to be treated with glycosidases to remove the carbohydrate modifications from the protein. The measurements of protein molecular weight before and after the removal of the attached glycan may provide information about the modification and the modified peptide; however, individual glycoforms are rarely resolved [68,69].
B.A. Budnik et al. / Biochimica et Biophysica Acta 1764 (2006) 1870–1880
1875
Fig. 3. An ECD spectrum of a triply O-GlcNAc glycosylated peptide, the number of glycans observed intact on the peptide backbone are denoted by arrows on the sequence. The spectrum provides information that facilitates the localization of the glycans to a single amino acid residue. Figure reproduced with permission from Ref. [83]. Copyright 1999 American Chemical Society.
N-glycans are cleaved from proteins using endoglycosidases such as Protein-N-glycosidase F (PNGase F) or endoglycosidase H (endo H). The enzymatic cleavage product of both enzymes results in a modified protein/peptide with characteristic signature or cleavage site that can be used to determine the site of glycosylation. The endo H cleaves the glycan structure between the two GlcNAc residues and core structure leaving reducing terminal GlcNAc residue attached to the peptide. PNGase F cleaves the glycan at the amide bond converting asparagine to aspartic acid and this change can be detected by a 1 Da mass increase [2]. Additionally, incubation in 18O/16O labeled water during the PNGase reaction incorporates labeled oxygen into the aspartic acid resulting in doublet peaks in the mass spectrum separated by two mass units. These doublets can be used (a) to localize the modification and (b) to perform relative quantitation of the glycan structure at that particular Nlinked glycosylation site [70,71]. The current methods of glycopeptide fragmentation provide excellent sequence coverage for both the glycan structure and peptide sequence and the use of glycosidases has proved to be very useful in characterizing glycans. A detailed study of the glycan structure of a protein involves the use of several enzymes that are specific for particular structural features of glycans. The general method entails incubating glycopeptides with highly specific exoglycosidases either sequentially or as an array [72], and monitoring the structure of the remaining carbohydrate per round of enzyme used. This method provides information about the structure of the complex carbohydrate, but can be slow and tedious. The method is also very dependent on availability of different enzyme libraries as well as on the purity of the enzyme, which is essential to ensure specific cleavage results. Unfortunately many of these enzymes are also very expensive as they are glycosylated and have to be purified from their native source in order to retain their specificity (changing the glycosylation patterns of these enzymes can change the specificity of the enzyme). Several commercially prepared enzymes were found to be contaminated with other nonspecific enzymes [73,74]. Usually methods using
serial enzymatic treatments are successful with conventional carbohydrates, but when samples contain chemical modifications or uncommon glycans, additional chemical degradation methods are necessary. This strategy, using tryptic digestion with subsequent exoglycosidase sequencing and analysis using MALDI was first performed by Cottrell and coworkers on recombinant human tissue inhibitor of metalloproteases (TIMP) [73]. It provided the site and structure of the modification. Unfortunately this strategy is limited to samples where the precise masses of the peptides and all other modifications are known. In order to overcome this shortcoming, the glycan can be released from the peptides and then sequenced with the serial glycosidase treatment thus allowing the sequencing of the glycans without having prior information about the peptide sequence [75]. The method was improved such that all enzyme digest steps could be performed on the MALDI target [76], but required that the matrix be removed by drop dialysis before each step of the digestion. In addition, SDS-PAGE gels were used to separate glycoproteins prior to glycan release using PNGase F enzyme. When analyzing glycopeptides and glycoproteins, it is important to desalt the sample and remove organic contaminants in order to obtain informative spectra. Removing salts avoids the formation of salt adducts, and cation and anion exchange materials have been used for desalting [77]. The most efficient method of desalting uses a microcolumn in a GELoader tip (Eppendorf) into which a mixed bed resin column of AG-3 (to remove anions) and AG-50 (to remove cations) and C18 (to remove organic materials) are packed [78]. HILIC chromatography and graphite columns are also useful for desalting [9,79]. 4.1. Fragmentation methods for the analysis of global glycosylation Understanding the behavior of glycopeptides in the mass spectrometer and utilizing the fragmentation characteristics of glycopeptides can greatly improve the quality of information
1876
B.A. Budnik et al. / Biochimica et Biophysica Acta 1764 (2006) 1870–1880
obtained by mass spectrometry. Designing experiments to answer a particular question requires understanding the limitations and advantages of fragmentation methods. In this section we describe CID and three fragmentation methods commonly used in Fourier Transform Ion Cyclotron Mass Spectrometers (FTICR) mass spectrometers: electron capture dissociation (ECD), infrared multiphoton dissociation (IRMPD), sustained off-resonance irradiation CID (SORI CID), and electron transfer dissociation (ETD) used in quadrupole ion traps. 4.1.1. Collision induced dissociation of glycopeptides The fragmentation patterns of both O-linked and N-linked glycopeptides have been described in the literature and there are differing opinions on how to get both peptide backbone information and glycan structural information from a particular peptide. In general the primary fragmentation pathway is glycan fragmentation whereas the peptide backbone fragmentation is secondary. The analysis of O-linked glycopeptides by high energy tandem mass spectrometry usually results in spectra with weak peptide back bone cleavage for peptides with attached carbohydrate groups and rather prominent signals for fragments resulting from the cleavage of carbohydrate side chains [80]. When low energy tandem mass spectrometry was applied to study peptides modified by single (NeuGc)2GalGalNAc moiety, the spectra showed dominant glycan fragmentation with NeuGc, Hex and HexNAc groups being lost from the precursor ion [81]. Mono- and di-fucosylated O-linked peptides were studied using a Q-ToF mass spectrometer with ESI under low energy fragmentation conditions [82]. The MS2 spectra in these studies display a significant oxonium ion signal for the hexose
loss from the precursor ion as well as few peptide bond cleavage fragments that contain the hexose groups, but these were very low abundance ion signals. These results show that for highly abundant species, the O-linked fucosylation can be assigned by using low energy collision conditions. Similar fragmentation studies on N-glycosylated peptides (high mannose) using a triple quadrupole instrument show MS2 spectra with the most abundant fragments corresponding to fragmentation of the glycan moiety [83–86]. Other fragmentation methods can be used in glycosylation studies to obtain both peptide sequence information and carbohydrate sequence information. Instruments capable of performing MSn experiments could be used in MS3 mode for specific glycan losses in the instruments like ion traps (Paul trap) or FTICR MS (Penning traps). In the ion trap, specific fragments from the MS2 experiments can be isolated and fragmented further i.e. the peptide fragment without the glycan attached could be fragmented further to obtain peptide sequence and the glycan ion could be fragmented to obtain the glycan structure. 4.1.2. FTICR methods of fragmentation for glycopeptides The invention of electron capture dissociation (ECD) in FTICR MS mass spectrometers for peptide sequencing [87] facilitated the single amino acid localization of labile modifications e.g. O-glycosylation [88] as well as N-glycosylation [89]. ECD produces odd-electron, free radical driven fragmentation. ECD produces cleavages specifically at N-αC backbone of the peptide leaving the modification intact on the peptide backbone (see Fig. 3). This feature of ECD is remarkable as it facilitates the localization of the modification and, with additional information from the CID spectra provides the mass of
Fig. 4. Various strategies for glycoprotein/glycopeptide isolation, detection and analysis are depicted in a flow chart. All of the strategies described here can be coupled to LC/MS, for example peptides eluted from a hydrazine column using PNGAse F can be collected, tippeddesalted/concentrated using C18 material and then subjected to LC/MS analysis.
B.A. Budnik et al. / Biochimica et Biophysica Acta 1764 (2006) 1870–1880
modification. Also IRMPD can be used to sequence the glycan structure of glycopeptides [90]. IRMPD fragmentation is used in FTICR cell and produces information rich spectra showing extensive fragmentation of the glycan with negligible or no peptide bond cleavage. ECD and IRMPD have been used in a complementary manner; ECD spectra provided 11 out of 15 possible back bone peptide cleavages; IRMPD independently provided exclusive fragmentation of the glycan producing spectra with enough information to identify three branching sites in the structure [89]. Sustained off-resonance irradiation (SORI) CID in combination with ECD was used to identify all glycan-specific modifications of a protein from a tryptic digest mixture [89]. A similar fragmentation method to ECD, electron transfer dissociation (ETD), has been recently introduced. ETD uses the same principle of unpaired electron attachment as ECD and can be performed in quadrupole ion trap mass spectrometers [91]. ETD fragmentation also results in the cleavage of N-αC bond of the peptide back bone preserving labile modifications. Use of ETD and MSn in the ion trap is a cheap and effective instrument/fragmentation choice for glycoproteomics. Recent published data shows great applicability of ETD to glycoproteomics approach [92]. 5. Instrument based methods for the analysis of global glycosylation In this section we describe methods that use mass spectrometry based experiments that use the labile nature of glycanprotein/peptide bond to identify glycosylated species. 5.1. Precursor ion scanning The development of ESI [64] resulted in the improved ionization of glycopeptides as compared with earlier studies performed using FAB [2,80,93]. The new ionization technique enabled the identification of glycan modified peptides by in source decay as well as precursor ion scanning for the oxonium ion. Precursor ion scanning experiments were used by Carr et al. to identify glycosylated proteins/peptides [94]. Precursor ion scanning is a method that takes advantage of characteristic ions. During a typical experiment on a triple quadrupole instrument, the first quadrupole scans a mass range such that peptide ions are transferred into the second quadrupole (the collision cell) where they undergo CID (collision induced dissociation). These ions are then transferred to the third quadrupole, which is set at a specific m/z i.e. for the characteristic fragment ion such that only the peptides that produce the characteristic ion that give rise to a signal at the detector. When glycopeptides are fragmented, glycospecific fragments are formed and these are usually found in the low mass region. These glycospecific fragments can be used as diagnostic markers indicating that the peptide being fragmented is a glycopeptide [94,95]. Since Carr et al. first used a triple quadrupole instrument for precursor ion scanning experiments to find glycopeptides, the carbohydrate specific detection scheme has been widely applied. In the positive ion mode, the
1877
signature ions commonly used are oxonium ion fragments such as Hex+ (m/z 163.06), [HexNAc]+ (m/z 204.09), and [HexHexNAc]+ (m/z 366.14). The last two ions are formed by either internal two bond cleavages within the glycan or can be derived from terminal sugars [96]. This method was improved by using an ion trap, where the characteristic ion chromatogram for the 204 [HexNAc] + ion, the 274 [NeuAc-H 2O] + ion, 292 [NeuAc] + ion and the 366 [Hex-HexNAc] + ion were concomitantly monitored during a LCMS experiment to determine which fraction of the chromatogram contains glycopeptides [97]. The observation of co-eluting ions at m/z 204 and 366 give greater confidence for the identification of a glycopeptide. Also the detection of 274 and 292 indicates the presence of sialic acid. Unfortunately precursor ion scanning is not always specific because low mass accuracy and resolution instrumentation is used i.e. the triple quadrupole and the ion trap mass spectrometers. The performance of triple quad instruments and Q-ToF instruments have been compared for precursor ion scanning experiments [98] and the results also show an advantage of the Q-TOF analyzer for these types of studies. The advent of hybrid Quadrupole-TOF instruments with higher mass accuracy helped to increase specificity of the method and discriminate between the oxonium ions and interfering peptide derived fragment ions [30]. 5.2. Neutral loss scanning Another method of glycopeptide detection is to look for a neutral loss of a sugar residue using a neutral loss scanning technique. This type of experiment is usually performed on triple quadrupole instruments but can be performed on Q-ToF and ion trap instruments as well [99]. Neutral loss scanning is a method similar to precursor ion scanning except that the method takes advantage of the loss of a neutral species during fragmentation. During such an experiment on a triple quadrupole instrument, the first quadrupole scans a mass range such that ions are transferred into the second quadrupole where they undergo CID. These ions are then transferred to the third quadrupole, which is set at an offset of a specific m/z i.e. the mass difference for the neutral loss (the loss of a neutral HexNAc-203) such that only the peptides that exhibit the loss of the neutral species produce a signal at the detector. Several manuscripts have been published using neutral loss scanning to detect glycosylation [42,100–103]. 6. Conclusion and perspective Recent studies of glycoproteins have demonstrated that glycoproteins undergo specific modifications that may alter cellular function and fate. Interrogating the specific changes in the glycoproteome in various models and biological systems may provide insights into basic biology and disease specific information that may lead to clinically relevant disease markers or therapeutic targets of disease. Although the analysis of the glycoproteome is challenging, we describe a number of tools for the characterization of glycoproteins and glycopeptides that may improve our ability to investigate and study the
1878
B.A. Budnik et al. / Biochimica et Biophysica Acta 1764 (2006) 1870–1880
glycoproteome. We expect that many of these techniques and methods can be modified and combined to improve the characterization of the global glycoproteome in large scale proteomics experiments. How can these new methods provide more information about the glycoproteome than what is currently known? For example, we imagine utilizing both O-GlcNAc enrichment strategies combined with ECD sequencing of the enriched peptides, such that the glycan can be localized to a single amino acid. Another example would be to utilize the high resolution and mass accuracy of the newer FTMS instruments to perform top down analysis of non-digested glycoproteins. We envision combining a lectin enrichment strategy to capture entire glycoproteins followed by direct analysis in combination with ECD and IRPMD to characterize both the peptide backbone and the glycan. Overall, the novel methods combining various enrichment strategies (Fig. 4) with rapidly improving mass spectrometry technology will improve the ability to interrogate the global glycoproteome. However, these improved technologies require improved bioinformatic platforms to better understanding of the glycoproteome and provide the foundation for high-throughput global glycoproteome analysis. References [1] S. Pan, H. Zhang, J. Rush, J. Eng, N. Zhang, D. Patterson, M.J. Comb, R. Aebersold, High throughput proteome screening for biomarker detection, Mol. Cell Proteomics 4 (2005) 182–190. [2] S.A. Carr, G.D. Roberts, Carbohydrate mapping by mass spectrometry: a novel method for identifying attachment sites of Asn-linked sugars in glycoproteins, Anal. Biochem. 157 (1986) 396–406. [3] N. Khidekel, S.B. Ficarro, E.C. Peters, L.C. Hsieh-Wilson, Exploring the O-GlcNAc proteome: direct identification of O-GlcNAc-modified proteins from the brain, Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 13132–13137. [4] A. Nandi, R. Sprung, D.K. Barma, Y. Zhao, S.C. Kim, J.R. Falck, Y. Zhao, Global identification of O-GlcNAc-modified proteins, Anal. Chem. 78 (2006) 452–458. [5] R. Sprung, A. Nandi, Y. Chen, S.C. Kim, D. Barma, J.R. Falck, Y. Zhao, Tagging-via-substrate strategy for probing O-GlcNAc modified proteins, J. Proteome Res. 4 (2005) 950–957. [6] H.C. Tai, N. Khidekel, S.B. Ficarro, E.C. Peters, L.C. Hsieh-Wilson, Parallel identification of O-GlcNAc-modified proteins from cell lysates, J. Am. Chem. Soc. 126 (2004) 10500–10501. [7] L. Wells, K. Vosseller, R.N. Cole, J.M. Cronshaw, M.J. Matunis, G.W. Hart, Mapping sites of O-GlcNAc modification using affinity tags for serine and threonine post-translational modifications, Mol. Cell Proteomics 1 (2002) 791–804. [8] J.A. Ludwig, J.N. Weinstein, Biomarkers in cancer staging, prognosis and treatment selection, Nat. Rev., Cancer 5 (2005) 845–856. [9] P. Hagglund, J. Bunkenborg, F. Elortza, O.N. Jensen, P. Roepstorff, A new strategy for identification of N-glycosylated proteins and unambiguous assignment of their glycosylation sites using HILIC enrichment and partial deglycosylation, J. Proteome Res. 3 (2004) 556–566. [10] R. Apweiler, H. Hermjakob, N. Sharon, On the frequency of protein glycosylation, as deduced from analysis of the SWISS-PROT database, Biochim. Biophys. Acta 1473 (1999) 4–8. [11] A. Kameyama, S. Nakaya, H. Ito, N. Kikuchi, T. Angata, M. Nakamura, H.K. Ishida, H. Narimatsu, Strategy for simulation of CID spectra of Nlinked oligosaccharides toward glycomics, J. Proteome Res. 5 (2006) 808–814.
[12] S. Roseman, Reflections on glycobiology, J. Biol. Chem. 276 (2001) 41527–41542. [13] P.M. Rudd, M.R. Wormald, R.A. Dwek, Sugar-mediated ligand-receptor interactions in the immune system, Trends Biotechnol. 22 (2004) 524–530. [14] J.B. Lowe, J.D. Marth, A genetic approach to mammalian glycan function, Annu. Rev. Biochem. 72 (2003) 643–691. [15] B. Casu, M. Guerrini, G. Torri, Structural and conformational aspects of the anticoagulant and anti-thrombotic activity of heparin and dermatan sulfate, Curr. Pharm. Des. 10 (2004) 939–949. [16] B.E. Collins, J.C. Paulson, Cell surface biology mediated by low affinity multivalent protein–glycan interactions, Curr. Opin. Chem. Biol. 8 (2004) 617–625. [17] D.H. Dube, C.R. Bertozzi, Glycans in cancer and inflammation— Potential for therapeutics and diagnostics, Nat. Rev., Drug Discov. 4 (2005) 477–488. [18] Y. Guo, H. Feinberg, E. Conroy, D.A. Mitchell, R. Alvarez, O. Blixt, M. E. Taylor, W.I. Weis, K. Drickamer, Structural basis for distinct ligandbinding and targeting properties of the receptors DC-SIGN and DCSIGNR, Nat. Struct. Mol. Biol. 11 (2004) 591–598. [19] H.Y. Hwang, S.K. Olson, J.D. Esko, H.R. Horvitz, Caenorhabditis elegans early embryogenesis and vulval morphogenesis require chondroitin biosynthesis, Nature 423 (2003) 439–443. [20] M. Inatani, F. Irie, A.S. Plump, M. Tessier-Lavigne, Y. Yamaguchi, Mammalian brain morphogenesis and midline axon guidance require heparan sulfate, Science 302 (2003) 1044–1046. [21] Y. Kinjo, D. Wu, G. Kim, G.W. Xing, M.A. Poles, D.D. Ho, M. Tsuji, K. Kawahara, C.H. Wong, M. Kronenberg, Recognition of bacterial glycosphingolipids by natural killer T cells, Nature 434 (2005) 520–525. [22] X. Lin, Functions of heparan sulfate proteoglycans in cell signaling during development, Development 131 (2004) 6009–6021. [23] D. Liu, Z. Shriver, G. Venkataraman, Y. El Shabrawi, R. Sasisekharan, Tumor cell surface heparan sulfate as cryptic promoters or inhibitors of tumor growth and metastasis, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 568–573. [24] S.I. Miller, R.K. Ernst, M.W. Bader, LPS, TLR4 and infectious disease diversity, Nat. Rev., Microbiol. 3 (2005) 36–46. [25] R. Sasisekharan, Z. Shriver, G. Venkataraman, U. Narayanasami, Roles of heparan-sulphate glycosaminoglycans in cancer, Nat. Rev., Cancer 2 (2002) 521–528. [26] R. Raman, S. Raguram, G. Venkataraman, J.C. Paulson, R. Sasisekharan, Glycomics: an integrated systems approach to structure–function relationships of glycans, Nat. Methods 2 (2005) 817–824. [27] G. Alvarez-Manilla, J. Atwood III, Y. Guo, N.L. Warren, R. Orlando, M. Pierce, Tools for glycoproteomic analysis: size exclusion chromatography facilitates identification of tryptic glycopeptides with N-linked glycosylation sites, J. Proteome Res. 5 (2006) 701–708. [28] N. Hashii, N. Kawasaki, S. Itoh, M. Hyuga, T. Kawanishi, T. Hayakawa, Glycomic/glycoproteomic analysis by liquid chromatography/mass spectrometry: analysis of glycan structural alteration in cells, Proteomics 5 (2005) 4665–4672. [29] N.L. Anderson, A. NG, The human plasma proteome—History, charachter, and diagnostic prospects, Mol. Cell Proteomics 1 (2003) 845–867. [30] J. Jebanathirajah, H. Steen, P. Roepstorff, Using optimized collision energies and high resolution, high accuracy fragment ion selection to improve glycopeptide detection by precursor ion scanning, J. Am. Soc. Mass Spectrom. 14 (2003) 777–784. [31] W. Zhang, G. Zhou, Y. Zhao, M.A. White, Y. Zhao, Affinity enrichment of plasma membrane for proteomics analysis, Electrophoresis 24 (2003) 2855–2863. [32] H. Zhang, E.C. Yi, X.J. Li, P. Mallick, K.S. Kelly-Spratt, C.D. Masselon, D.G. Camp II, R.D. Smith, C.J. Kemp, R. Aebersold, High throughput quantitative analysis of serum proteins using glycopeptide capture and liquid chromatography mass spectrometry, Mol. Cell Proteomics 4 (2005) 144–155. [33] Y. Zhao, W. Zhang, Y. Kho, Y. Zhao, Proteomic analysis of integral plasma membrane proteins, Anal. Chem. 76 (2004) 1817–1823.
B.A. Budnik et al. / Biochimica et Biophysica Acta 1764 (2006) 1870–1880 [34] J. Nawarak, S. Phutrakul, S.T. Chen, Analysis of lectin-bound glycoproteins in snake venom from the elapidae and viperidae families, J. Proteome Res. 3 (2004) 383–392. [35] R. Uematsu, J. Furukawa, H. Nakagawa, Y. Shinohara, K. Deguchi, K. Monde, S. Nishimura, High throughput quantitative glycomics and glycoform-focused proteomics of murine dermis and epidermis, Mol. Cell Proteomics 4 (2005) 1977–1989. [36] S.M. Brittain, S.B. Ficarro, A. Brock, E.C. Peters, Enrichment and analysis of peptide subsets using fluorous affinity tags and mass spectrometry, Nat. Biotechnol. 23 (2005) 463–468. [37] H. Mirzaei, F. Regnier, Affinity chromatographic selection of carbonylated proteins followed by identification of oxidation sites using tandem mass spectrometry, Anal. Chem. 77 (2005) 2386–2392. [38] H. Lis, N. Sharon, E. Katchalski, Identification of the carbohydrate– protein linking group in soybean hemagglutinin, Biochim. Biophys. Acta 192 (1969) 364–366. [39] J. Ji, A. Chakraborty, M. Geng, X. Zhang, A. Amini, M. Bina, F. Regnier, Strategy for qualitative and quantitative analysis in proteomics based on signature peptides, J. Chromatogr., B Biomed. Sci. Appl. 745 (2000) 197–210. [40] E. Mortz, T. Sareneva, S. Haebel, I. Julkunen, P. Roepstorff, Mass spectrometric characterization of glycosylated interferon-gamma variants separated by gel electrophoresis, Electrophoresis 17 (1996) 925–931. [41] F.M. Okanda, Z. El Rassi, Affinity monolithic capillary columns for glycomics/proteomics: 1. Polymethacrylate monoliths with immobilized lectins for glycoprotein separation by affinity capillary electrochromatography and affinity nano-liquid chromatography in either a single column or columns coupled in series, Electrophoresis 27 (2006) 1020–1030. [42] K. Vosseller, J.C. Trinidad, R.J. Chalkley, C.G. Specht, A. Thalhammer, A.J. Lynn, J.O. Snedecor, S. Guan, K.F. Medzihradszky, D.A. Maltby, R. Schoepfer, A.L. Burlingame, O-linked N-acetylglucosamine proteomics of postsynaptic density preparations using lectin weak affinity chromatography and mass spectrometry, Mol. Cell Proteomics 5 (2006) 923–934. [43] U. Lewandrowski, J. Moebius, U. Walter, A. Sickmann, Elucidation of Nglycosylation sites on human platelet proteins: a glycoproteomic approach, Mol. Cell Proteomics 5 (2006) 226–233. [44] Z. Yang, W.S. Hancock, T.R. Chew, L. Bonilla, A study of glycoproteins in human serum and plasma reference standards (HUPO) using multilectin affinity chromatography coupled with RPLC-MS/MS, Proteomics 5 (2005) 3353–3366. [45] R. Qiu, F.E. Regnier, Use of multidimensional lectin affinity chromatography in differential glycoproteomics, Anal. Chem. 77 (2005) 2802–2809. [46] N.M. Young, J.R. Brisson, J. Kelly, D.C. Watson, L. Tessier, P.H. Lanthier, H.C. Jarrell, N. Cadotte, F. St Michael, E. Aberg, C.M. Szymanski, Structure of the N-linked glycan present on multiple glycoproteins in the Gram-negative bacterium, campylobacter jejuni, J. Biol. Chem. 277 (2002) 42530–42539. [47] L. Guo, J.R. Eisenman, R.M. Mahimkar, J.J. Peschon, R.J. Paxton, R.A. Black, R.S. Johnson, A proteomic approach for the identification of cellsurface proteins shed by metalloproteases, Mol. Cell Proteomics 1 (2002) 30–36. [48] R.D. Cummings, S. Kornfeld, The distribution of repeating [Gal beta 1,4GlcNAc beta 1,3] sequences in asparagine-linked oligosaccharides of the mouse lymphoma cell lines BW5147 and PHAR 2.1, J. Biol. Chem. 259 (1984) 6253–6260. [49] A.M. Wu, J.H. Wu, W.M. Watkins, C.P. Chen, M.C. Tsai, Binding properties of a blood group Le(a+) active sialoglycoprotein, purified from human ovarian cyst, with applied lectins, Biochim. Biophys. Acta 1316 (1996) 139–144. [50] L. Xiong, F.E. Regnier, Use of a lectin affinity selector in the search for unusual glycosylation in proteomics, J. Chromatogr., B Anal. Technol. Biomed. Life Sci. 782 (2002) 405–418. [51] M.F. Lopez, K. Berggren, E. Chernokalskaya, A. Lazarev, M. Robinson, W.F. Patton, A comparison of silver stain and SYPRO ruby protein gel stain with respect to protein detection in two-dimensional gels and identification by peptide mass profiling, Electrophoresis 21 (2000) 3673–3683.
1879
[52] G. Zhang, T.A. Neubert, Automated comparative proteomics based on multiplex tandem mass spectrometry and stable isotope labeling, Mol. Cell Proteomics 5 (2006) 401–411. [53] J.S. Jebanathirajah, H. Stensballe, A. Jensen, O. Roepstorff, P (2002), Modification Specific Proteomics: integrated Strategy for Glyco and Phosphospecific Proteomics, presented at the ASMS 2002, Orlando, FL. [54] H. Zhang, X.J. Li, D.B. Martin, R. Aebersold, Identification and quantification of N-linked glycoproteins using hydrazide chemistry, stable isotope labeling and mass spectrometry, Nat. Biotechnol. 21 (2003) 660–666. [55] N. Khidekel, S. Arndt, N. Lamarre-Vincent, A. Lippert, K.G. PoulinKerstien, B. Ramakrishnan, P.K. Qasba, L.C. Hsieh-Wilson, A chemoenzymatic approach toward the rapid and sensitive detection of O-GlcNAc posttranslational modifications, J. Am. Chem. Soc. 125 (2003) 16162–16163. [56] E. Saxon, C.R. Bertozzi, Cell surface engineering by a modified Staudinger reaction, Science 287 (2000) 2007–2010. [57] N.E. Zachara, G.W. Hart, Cell signaling, the essential role of O-GlcNAc! Biochim. Biophys. Acta (2006) 599–617. [58] N. Lamarre-Vincent, L.C. Hsieh-Wilson, Dynamic glycosylation of the transcription factor CREB: a potential role in gene regulation, J. Am. Chem. Soc. 125 (2003) 6612–6613. [59] D.J. Vocadlo, H.C. Hang, E.J. Kim, J.A. Hanover, C.R. Bertozzi, A chemical approach for identifying O-GlcNAc-modified proteins in cells, Proc. Natl. Acad. Sci. U. S. A. 100 (2003) 9116–9121. [60] J.A. Prescher, D.H. Dube, C.R. Bertozzi, Chemical remodelling of cell surfaces in living animals, Nature 430 (2004) 873–877. [61] Y. Kho, S.C. Kim, C. Jiang, D. Barma, S.W. Kwon, J. Cheng, J. Jaunbergs, C. Weinbaum, F. Tamanoi, J. Falck, Y. Zhao, A tagging-viasubstrate technology for detection and proteomics of farnesylated proteins, Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 12479–12484. [62] Y. Zhang, A. Wolf-Yadlin, P.L. Ross, D.J. Pappin, J. Rush, D.A. Lauffenburger, F.M. White, Time-resolved mass spectrometry of tyrosine phosphorylation sites in the epidermal growth factor receptor signaling network reveals dynamic modules, Mol. Cell Proteomics 4 (2005) 1240–1250. [63] K. Vosseller, K.C. Hansen, R.J. Chalkley, J.C. Trinidad, L. Wells, G.W. Hart, A.L. Burlingame, Quantitative analysis of both protein expression and serine/threonine post-translational modifications through stable isotope labeling with dithiothreitol, Proteomics 5 (2005) 388–398. [64] J.B. Fenn, M. Mann, C.K. Meng, S.F. Wong, C.M. Whitehouse, Electrospray ionization for mass spectrometry of large biomolecules, Science 246 (1989) 64–71. [65] M. Karas, F. Hillenkamp, Laser desorption ionization of proteins with molecular masses exceeding 10,000 daltons, Anal. Chem. 60 (1988) 2299–2301. [66] B.U. Tsarbopoulos A, B.N. Pramanik, M. Karas, Glycoprotein analysis by delayed extraction and post-source decay MALDI-TOF-MS, Int. J. Mass Spectrom. (1997) 251–261. [67] Y. Wada, J. Gu, N. Okamoto, K. Inui, Diagnosis of carbohydrate-deficient glycoprotein syndrome by matrix-assisted laser desorption time-of-flight mass spectrometry, Biol. Mass Spectrom. 23 (1994) 108–109. [68] Y. Yang, R. Orlando, Identifying the glycosylation sites and site-specific carbohydrate heterogeneity of glycoproteins by matrix-assisted laser desorption/ionization mass spectrometry, Rapid Commun. Mass Spectrom. 10 (1996) 932–936. [69] J. Zaia, R.E. Boynton, A. McIntosh, D.R. Marshak, H. Olsson, D. Heinegard, F.P. Barry, Post-translational modifications in cartilage oligomeric matrix protein. Characterization of the N-linked oligosaccharides by matrix-assisted laser desorption ionization time-of-flight mass spectrometry, J. Biol. Chem. 272 (1997) 14120–14126. [70] J. Gonzalez, T. Takao, H. Hori, V. Besada, R. Rodriguez, G. Padron, Y. Shimonishi, A method for determination of N-glycosylation sites in glycoproteins by collision-induced dissociation analysis in fast atom bombardment mass spectrometry: identification of the positions of carbohydrate-linked asparagine in recombinant alpha-amylase by treatment with peptide-N-glycosidase F in 18O-labeled water, Anal. Biochem. 205 (1992) 151–158.
1880
B.A. Budnik et al. / Biochimica et Biophysica Acta 1764 (2006) 1870–1880
[71] B. Kuster, M. Mann, 18O-labeling of N-glycosylation sites to improve the identification of gel-separated glycoproteins using peptide mass mapping and database searching, Anal. Chem. 71 (1999) 1431–1440. [72] C.J. Edge, T.W. Rademacher, M.R. Wormald, R.B. Parekh, T.D. Butters, D.R. Wing, R.A. Dwek, Fast sequencing of oligosaccharides: the reagentarray analysis method, Proc. Natl. Acad. Sci. U. S. A. 89 (1992) 6338–6342. [73] C.W. Sutton, J.A. O'Neill, J.S. Cottrell, Site-specific characterization of glycoprotein carbohydrates by exoglycosidase digestion and laser desorption mass spectrometry, Anal. Biochem. 218 (1994) 34–46. [74] G.D. Roberts, W.P. Johnson, S. Burman, K.R. Anumula, S.A. Carr, An integrated strategy for structural characterization of the protein and carbohydrate components of monoclonal antibodies: application to antirespiratory syncytial virus MAb, Anal. Chem. 67 (1995) 3613–3625. [75] D.J. R.P. Harvey, R.H. Bateman, R.S. Bordoli, K. Howes, J.B. Hoyes, R.G. Vickerg, Examination of complex oligosaccharides by matrixassisted laser-desorption ionization mass-spectrometry on time-of-flight and magnetic-sector instruments, Org. Mass Spectrom. 753 (1994). [76] B. Kuster, T.J. Naven, D.J. Harvey, Rapid approach for sequencing neutral oligosaccharides by exoglycosidase digestion and matrix-assisted laser desorption/ionization time-of-flight mass spectrometry, J. Mass Spectrom. 31 (1996) 1131–1140. [77] V. Lattard, M. Fondeur-Gelinotte, S. Gulberti, J.C. Jacquinet, J. Boudrant, P. Netter, J. Magdalou, M. Ouzzine, S. Fournel-Gigleux, Purification and characterization of a soluble form of the recombinant human galactosebeta1,3-glucuronosyltransferase I expressed in the yeast Pichia pastoris, Protein Expr. Purif. 47 (2006) 137–143. [78] M. Kussmann, U. Lassing, C.A. Sturmer, M. Przybylski, P. Roepstorff, Matrix-assisted laser desorption/ionization mass spectrometric peptide mapping of the neural cell adhesion protein neurolin purified by sodium dodecyl sulfate polyacrylamide gel electrophoresis or acidic precipitation, J. Mass Spectrom. 32 (1997) 483–493. [79] M.R. Larsen, S.J. Cordwell, P. Roepstorff, Graphite powder as an alternative or supplement to reversed-phase material for desalting and concentration of peptide mixtures prior to matrix-assisted laser desorption/ionization-mass spectrometry, Proteomics 2 (2002) 1277–1287. [80] K.F. Medzihradszky, B.L. Gillece-Castro, C.A. Settineri, R.R. Townsend, F.R. Masiarz, A.L. Burlingame, Structure determination of O-linked glycopeptides by tandem mass spectrometry, Biomed. Environ. Mass Spectrom. 19 (1990) 777–781. [81] K. Hirayama, R. Yuji, N. Yamada, K. Kato, Y. Arata, I. Shimada, Complete and rapid peptide and glycopeptide mapping of mouse monoclonal antibody by LC/MS/MS using ion trap mass spectrometry, Anal. Chem. 70 (1998) 2718–2725. [82] B. Macek, J. Hofsteenge, J. Peter-Katalinic, Direct determination of glycosylation sites in O-fucosylated glycopeptides using nano-electrospray quadrupole time-of-flight mass spectrometry, Rapid Commun. Mass Spectrom. 15 (2001) 771–777. [83] K.P. Bateman, R.L. White, P. Thibault, Evaluation of adsorption preconcentration/capillary zone electrophoresis/nanoelectrospray mass spectrometry for peptide and glycoprotein analyses, J. Mass Spectrom. 33 (1998) 1109–1123. [84] R. Zeng, Q. Xu, X.X. Shao, K.Y. Wang, Q.C. Xia, Characterization and analysis of a novel glycoprotein from snake venom using liquid chromatography-electrospray mass spectrometry and Edman degradation, Eur. J. Biochem. 266 (1999) 352–358. [85] X. Zhu, C. Borchers, R.J. Bienstock, K.B. Tomer, Mass spectrometric characterization of the glycosylation pattern of HIV-gp120 expressed in CHO cells, Biochemistry 39 (2000) 11194–11204.
[86] J.P. Hui, T.C. White, P. Thibault, Identification of glycan structure and glycosylation sites in cellobiohydrolase II and endoglucanases I and II from Trichoderma reesei, Glycobiology 12 (2002) 837–849. [87] R.A. K.N. Zubarev, F.W. McLafferty, Electron capture dissociation of multiply charged protein cations. A nonergodic process, J. Am. Chem. Soc. 120 (1998) 3265–3266. [88] E. Mirgorodskaya, P. Roepstorff, R.A. Zubarev, Localization of Oglycosylation sites in peptides by electron capture dissociation in a Fourier transform mass spectrometer, Anal. Chem. 71 (1999) 4431–4436. [89] K. Hakansson, H.J. Cooper, M.R. Emmett, C.E. Costello, A.G. Marshall, C.L. Nilsson, Electron capture dissociation and infrared multiphoton dissociation MS/MS of an N-glycosylated tryptic peptic to yield complementary sequence information, Anal. Chem. 73 (2001) 4530–4536. [90] J.T. Adamson, K. Hakansson, Infrared multiphoton dissociation and electron capture dissociation of high-mannose type glycopeptides, J. Proteome Res. 5 (2006) 493–501. [91] J.E. Syka, J.J. Coon, M.J. Schroeder, J. Shabanowitz, D.F. Hunt, Peptide and protein sequence analysis by electron transfer dissociation mass spectrometry, Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 9528–9533. [92] J.M. Hogan, S.J. Pitteri, P.A. Chrisman, S.A. McLuckey, Complementary structural information from a tryptic N-linked glycopeptide via electron transfer ion/ion reactions and collision-induced dissociation, J. Proteome Res. 4 (2005) 628–632. [93] S.A. Carr, G.D. Roberts, A. Jurewicz, B. Frederick, Structural fingerprinting of Asn-linked carbohydrates from specific attachment sites in glycoproteins by mass spectrometry: application to tissue plasminogen activator, Biochimie 70 (1988) 1445–1454. [94] M.J. Huddleston, M.F. Bean, S.A. Carr, Collisional fragmentation of glycopeptides by electrospray ionization LC/MS and LC/MS/MS: methods for selective detection of glycopeptides in protein digests, Anal. Chem. 65 (1993) 877–884. [95] K.L. Duffin, J.K. Welply, E. Huang, J.D. Henion, Characterization of Nlinked oligosaccharides by electrospray and tandem mass spectrometry, Anal. Chem. 64 (1992) 1440–1448. [96] R.S. Annan, S.A. Carr, The essential role of mass spectrometry in characterizing protein structure: mapping posttranslational modifications, J. Protein Chem. 16 (1997) 391–402. [97] B. Sullivan, T.A. Addona, S.A. Carr, Selective detection of glycopeptides on ion trap mass spectrometers, Anal. Chem. 76 (2004) 3112–3118. [98] H. Steen, B. Kuster, M. Fernandez, A. Pandey, M. Mann, Detection of tyrosine phosphorylated peptides by precursor ion scanning quadrupole TOF mass spectrometry in positive ion mode, Anal. Chem. 73 (2001) 1440–1448. [99] H. Jiang, H. Desaire, V.Y. Butnev, G.R. Bousfield, Glycoprotein profiling by electrospray mass spectrometry, J. Am. Soc. Mass Spectrom. 15 (2004) 750–758. [100] L.E. Ball, M.N. Berkaw, M.G. Buse, Identification of the major site of O-linked beta-N-acetylglucosamine modification in the C terminus of insulin receptor substrate-1, Mol. Cell Proteomics 5 (2006) 313–323. [101] T.N. Corso, C.K. Van Pelt, J. Li, C. Ptak, X. Huang, Ultralow-volume fraction collection from nanoLC columns for mass spectrometric analysis of protein phosphorylation and glycosylation, Anal. Chem. 78 (2006) 2209–2219. [102] M.D. Hoffman, M.J. Sniatynski, J.C. Rogalski, J.C. Le Blanc, J. Kast, Multiple neutral loss monitoring (MNM): a multiplexed method for posttranslational modification screening, J. Am. Soc. Mass Spectrom. 17 (2006) 307–317. [103] D.B. Martin, J.K. Eng, A.I. Nesvizhskii, A. Gemmill, R. Aebersold, Investigation of neutral loss during collision-induced dissociation of peptide ions, Anal. Chem. 77 (2005) 4870–4882.