Anal. Chem. 2007, 79, 5826-5837
Isolation of N-Linked Glycopeptides from Plasma Yong Zhou,† Ruedi Aebersold,‡ and Hui Zhang*,†,§
Institute for Systems Biology, Seattle, Washington 98103, Institute of Molecular Systems Biology, Swiss Federal Institute of Technology (ETH) Zurich and Faculty of Natural Sciences, University of Zurich, CH-8093, Switzerland, and Department of Pathology, Johns Hopkins University, Baltimore, Maryland 21287
Proteomic analysis of blood plasma can potentially identify biomarkers that are useful for classifying the physiological or pathological status of an individual and for monitoring the effects of therapy. However, the complexity of the plasma proteome, the large number of peptides generated per protein due to dynamic protein post-translational modifications of each protein, and sequence variations among individuals pose great challenges to current proteomic technologies. To overcome these challenges, we have recently developed a method for the high-throughput analysis of glycoproteins using solid-phase extraction of N-linked glycopeptides (SPEG). Here we describe a procedure for plasma analysis using SPEG in which each step of SPEG was optimized. The performance of optimization was monitored using mouse plasma spiked with radioactive-labeled human plasma glycoproteins. Our data show that a standard procedure for plasma proteome analysis can be developed using the SPEG technique, mainly due to the relatively constant protein content in plasma. Blood plasma can be considered a window into the state of an individual’s health. Since all proteins in plasma are secreted, shed, or otherwise leaked from different cells or tissues,1-3 the discovery of disease-specific biomarkers in plasma has been the focus of intense technology development in the field of proteomics.4 However, the peculiar properties of plasma pose specific challenges for all proteomic methods. First, the plasma proteome is assumed to consist of minimally tens of thousands of different protein species.1 Second, the plasma proteome is dominated by a few highly abundant proteins and the concentration range of plasma proteins is estimated to cover at least 10 orders of magnitude.1 Third, post-translational modifications, in particular glycosylation, introduce extensive heterogeneity for plasma proteins.1 Fourth, quantitative plasma proteomic methods are further complicated by the fact that the plasma proteome varies over time in each individual as well as between individuals in a population.5 * To whom correspondence should be addressed. E-mail:
[email protected]. † Institute for Systems Biology. ‡ University of Zurich. § Johns Hopkins University. (1) Anderson, N. L.; Anderson, N. G. Mol. Cell. Proteomics 2002, 1, 845-867. (2) Anderson, N. L.; Polanski, M.; Pieper, R.; Gatlin, T.; Tirumalai, R. S.; Conrads, T. P.; Veenstra, T. D.; Adkins, J. N.; Pounds, J. G.; Fagan, R.; Lobley, A. Mol. Cell. Proteomics 2004, 3, 311-326. (3) Lathrop, J. T.; Anderson, N. L.; Anderson, N. G.; Hammond, D. J. Curr. Opin. Mol. Ther. 2003, 5, 250-257. (4) Zhang, H. Expert Rev. Proteomics 2006, 3, 175-178.
5826 Analytical Chemistry, Vol. 79, No. 15, August 1, 2007
To overcome these challenges and to increase the likelihood of detecting low-abundance plasma proteins that are assumed to be most informative as biomarkers for a specific disease, plasma proteins or peptides have been extensively fractionated prior to their analysis.6,7 To date, three general strategies have been used for such complexity reduction. In the first method, the most abundant serum proteins are removed by affinity depletion.8-11 In the second approach, proteins or peptides are fractionated using their physicochemical properties such as size, charge, or hydropathy prior to mass spectrometric analysis. Specific implementations include two- and three-dimensional peptide chromatography9,12,13 or size fractionation of proteins prior to their digestion and analysis by LC-MS/MS.12 The third approach is based on the use of specific chemical probes that can selectively tag and facilitate subsequent isolation of a target peptide of protein. On average, tryptic digestion generates several dozen peptides per protein. Protein post-translational modifications and genetic variation among individuals further increase the number of peptides from each protein. Profiling of complex proteomes by LC-MS/MS is therefore complicated by the large number of redundant peptides from each protein. Theoretically, one unique peptide should be sufficient to unambiguously identify each parent protein. If such unique peptides could be isolated, the complexity of the samples for proteomic profiling would be reduced by 1 or (5) Nedelkov, D.; Kiernan, U. A.; Niederkofler, E. E.; Tubbs, K. A.; Nelson, R. W. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 10852-10857. (6) Omenn, G. S.; States, D. J.; Adamski, M.; Blackwell, T. W.; Menon, R.; Hermjakob, H.; Apweiler, R.; Haab, B. B.; Simpson, R. J.; Eddes, J. S.; Kapp, E. A.; Moritz, R. L.; Chan, D. W.; Rai, A. J.; Admon, A.; Aebersold, R.; Eng, J.; Hancock, W. S.; Hefta, S. A.; Meyer, H.; Paik, Y. K.; Yoo, J. S.; Ping, P.; Pounds, J.; Adkins, J.; Qian, X.; Wang, R.; Wasinger, V.; Wu, C. Y.; Zhao, X.; Zeng, R.; Archakov, A.; Tsugita, A.; Beer, I.; Pandey, A.; Pisano, M.; Andrews, P.; Tammen, H.; Speicher, D. W.; Hanash, S. M. Proteomics 2005, 5, 3226-3245. (7) States, D. J.; Omenn, G. S.; Blackwell, T. W.; Fermin, D.; Eng, J.; Speicher, D. W.; Hanash, S. M. Nat. Biotechnol. 2006, 24, 333-338. (8) Liu, T.; Qian, W. J.; Gritsenko, M. A.; Camp, D. G., 2nd; Monroe, M. E.; Moore, R. J.; Smith, R. D. J. Proteome Res. 2005, 4, 2070-2080. (9) Adkins, J. N.; Varnum, S. M.; Auberry, K. J.; Moore, R. J.; Angell, N. H.; Smith, R. D.; Springer, D. L.; Pounds, J. G. Mol. Cell. Proteomics 2002, 1, 947-955. (10) Pieper, R.; Gatlin, C. L.; Makusky, A. J.; Russo, P. S.; Schatz, C. R.; Miller, S. S.; Su, Q.; McGrath, A. M.; Estock, M. A.; Parmar, P. P.; Zhao, M.; Huang, S. T.; Zhou, J.; Wang, F.; Esquer-Blasco, R.; Anderson, N. L.; Taylor, J.; Steiner, S. Proteomics 2003, 3, 1345-1364. (11) Pieper, R.; Su, Q.; Gatlin, C. L.; Huang, S. T.; Anderson, N. L.; Steiner, S. Proteomics 2003, 3, 422-432. (12) Shen, Y.; Jacobs, J. M.; Camp, D. G., 2nd; Fang, R.; Moore, R. J.; Smith, R. D.; Xiao, W.; Davis, R. W.; Tompkins, R. G. Anal. Chem. 2004, 76, 11341144. (13) Tirumalai, R. S.; Chan, K. C.; Prieto, D. A.; Issaq, H. J.; Conrads, T. P.; Veenstra, T. D. Mol. Cell. Proteomics 2003, 2, 1096-1103. 10.1021/ac0623181 CCC: $37.00
© 2007 American Chemical Society Published on Web 06/26/2007
2 orders of magnitude. Therefore, several strategies have been developed that reduce the number of peptides from each protein by targeting peptides containing unique (N- or C-termini), rare (Cys, Met, Trp, His, etc.) amino acids, or post-translational modifications such as phosphorylated or glycosylated peptides.14 Protein glycosylation is one of the most common posttranslational modifications; it not only influences properties of the proteins it is attached to but also regulates diverse physiologic functions through specific protein-carbohydrate recognition.15-20 Protein glycosylation, in particular N-linked glycosylation, is prevalent in protein sorting and secretion into extracellular environments.21 In addition, the aberrant glycosylation of glycoproteins is a fundamental characteristic of oncogenesis and tumor progression.22-26 However, the most significant benefit of selective isolation of N-glycopeptides originates from the fact that the number of N-linked glycosylation sites (N-glycosites) in the human proteome is modest, known in principle, and identifiable with current proteomic technology. This is due to the fact that N-glycosites generally fall into the N-X-S/T sequence motif in which X denotes any amino acid except proline.27 The potential N-glycosites over the human proteome can therefore be computationally determined by scanning the sequences for the presence of the motif. Since N-glycosylation mainly occurs in extracellular proteins, the total number of N-glycosites that originate from human extracellular proteins and contain N-X-S/T sequence motif is ∼20 000, yet these represent the majority of extracellular proteins (over 70%).28 This indicates that the total number of N-glycosites can be resolved with current proteomic technology using high-performance liquid chromatography and high-massaccuracy mass spectrometry. This will significantly increase the throughput and sensitivity for biomarker discovery by reducing the number of peptides of each mass detected by the mass spectrometer. This in turn indicates that a very limited number of defined N-glycosites from the human proteome can be assigned to each observed peptide mass if the N-glycopeptides can be specifically isolated and analyzed. Therefore, a method that can specifically and efficiently isolate one N-glycopeptide from each N-glycosite of plasma protein would be of significant interest for protein biomarker discovery. Recently, we have developed a novel method for solid-phase extraction of N-linked glycopeptides from glycoproteins (SPEG).29 (14) Zhang, H.; Yan, W.; Aebersold, R. Curr. Opin. Chem. Biol. 2004, 8, 66-75. (15) Helenius, A.; Aebi, M. Science 2001, 291, 2364-2369. (16) Allahverdian, S.; Patchell, B. J.; Dorscheid, D. R. Curr. Drug Targets 2006, 7, 597-606. (17) Rudd, P. M.; Elliott, T.; Cresswell, P.; Wilson, I. A.; Dwek, R. A. Science 2001, 291, 2370-2376. (18) Bertozzi, C. R.; Kiessling, L. L. Science 2001, 291, 2357-2364. (19) O’Donnell, N. Biochim. Biophys. Acta 2002, 1573, 336-345. (20) Cloos, P. A.; Christgau, S. Biogerontology 2004, 5, 139-158. (21) Roth, J. Chem. Rev. 2002, 102, 285-303. (22) Glinsky, G. V. Crit. Rev. Oncol. Hematol. 1994, 17, 27-51. (23) Hakomori, S. Cancer Res. 1996, 56, 5309-5318. (24) Dennis, J. W.; Granovsky, M.; Warren, C. E. Biochim. Biophys. Acta 1999, 1473, 21-34. (25) Couldrey, C.; Green, J. E. Breast Cancer Res. 2000, 2, 321-323. (26) Hakomori, S. Adv. Exp. Med. Biol. 2001, 491, 369-402. (27) Bause, E. Biochem. J. 1983, 209, 331-336. (28) Zhang, H.; Loriaux, P.; Eng, J.; Campbell, D.; Keller, A.; Moss, P.; Bonneau, R.; Zhang, N.; Zhou, Y.; Wollscheid, B.; Cooke, K.; Yi, E. C.; Lee, H.; Peskind, E. R.; Zhang, J.; Smith, R. D.; Aebersold, R. Genome Biol. 2006, 7, R73. (29) Zhang, H.; Li, X. J.; Martin, D. B.; Aebersold, R. Nat. Biotechnol. 2003, 21, 660-666.
It is based on the conjugation of glycoproteins to a solid support using hydrazide chemistry,30,31 removal of nonglycosylated peptides by trypsin digestion, and the specific release of Nglycopeptides via peptide-N-glycosidase F (PNGase F). The recovered peptides are then identified and quantified by tandem mass spectrometry.29 Here, we describe an optimized standard procedure for blood plasma analysis using SPEG and mass spectrometry. Each step of the SPEG procedure was evaluated and optimized for isolation of plasma N-glycopeptides using mouse plasma spiked with 14C-labeled human plasma glycoproteins. This optimized SPEG procedure increases the specificity and efficiency of extracting plasma N-glycopeptides and therefore is expected to enhance biomarker discovery from plasma using SPEG. EXPERIMENTAL SECTION [14C]-Iodoacetamide Labeling of Human Proteins.32,33 Two human plasma glycoproteins, R-1-antitrypsin (AT) and R-1acid glycoprotein 1 (AGP), were selected for 14C radioactive labeling. Both human plasma proteins were obtained from Calbiochem (EMD Biosciences, San Diego, CA). Aliquots of 0.5 nmol of AT and AGP proteins were dissolved in 5 µL of denaturing buffer (8 M urea in 100 mM NH4HCO3, pH 8.3). Proteins were first reduced with 5 mM tris(2-carboxyethyl)phosphine (TCEP; Pierce, Rockford, IL) for 30 min at room temperature and then alkylated by the addition of 5 mM [14C]-iodoacetamide (50 mCi/ mmol; GE Healthcare, Buckinghamshire, England) and incubation for 1 h at 37 °C with constant, gentle agitation. Excess iodoacetamide (1 µL of 250 mM iodoacetamide in 100 mM NH4HCO3, pH 8.3) was added and allowed to stand for another 30 min. Labeled proteins were purified from unreacted iodoacetamide by passing samples through a P6 protein desalting spin column (Pierce). The molar ratio of 14C bound per AT and AGP was calculated using a Wallac 1409 liquid scintillation analyzer (PerkinElmer, Boston, MA). Purification of N-Glycopeptides from Mouse Plasma Spiked with 14C-Labeled Human AT and AGP. The mouse sample was from mouse strain NIHO1a kindly provided by Dr. Christopher J. Kemp (Fred Hutchinson Cancer Research Center, Seattle, WA). Isolation of N-glycopeptides from mouse plasma was performed based on the method previously described.29 Briefly, 20 µL of mouse plasma spiked with 5000 cpm 14C-labeled AT and AGP (equivalent to ∼0.5-1µg of protein) was first diluted to 40 µL with oxidation buffer (20 mM NaAc and 150 mM NaCl, pH 5.0) and exchanged to the oxidation buffer using a P6 protein desalting spin column, and then oxidized with NaIO4 at different concentrations and temperatures specified in the results section. After removing the oxidant with the same P6 desalting column, glycosylated proteins were coupled to 100 µL of hydrazide resin (Affi-prep, Bio-Rad, Hercules, CA) in coupling buffer (100 mM NaAc and 1.5 M NaCl, pH 4.5) overnight at room temperature. The unbound nonglycosylated proteins were then removed, and immobilized proteins on resin were denatured by washing the (30) Bayer, E. A.; Ben-Hur, H.; Wilchek, M. Anal. Biochem. 1988, 170, 271281. (31) Bobbitt, J. M. Adv. Carbohydr. Chem. 1956, 48, 1-41. (32) Luduena, R. F.; Roach, M. C. Biochemistry 1981, 20, 4437-4444. (33) Roach, M. C.; Bane, S.; Luduena, R. F. J. Biol. Chem. 1985, 260, 30153023.
Analytical Chemistry, Vol. 79, No. 15, August 1, 2007
5827
resin 3 times with 1 mL of 8 M urea in 100 mM NH4HCO3 (pH 8.3), then reduced with 5 mM TCEP for 30 min at room temperature, and alkylated with 10 mM iodoacetamide. Immobilized proteins were then washed 3 times with 1 mL of different denaturing buffers indicated in the Results and Discussion section for Figure 3 and digested with 20 µg of trypsin at 37 °C. Nonglycosylated peptides from immobilized glycoproteins were removed by washing three times each with 1.5 M NaCl, 80% acetonitrile, 100% methanol, water, and 50 mM NH4HCO3. NGlycopeptides were finally released from the resin by addition of 0.6 µL of PNGase F (500 units/µL; New England Biolabs, Beverly, MA) in 100 µL of 50 mM NH4HCO3 (pH 7.5) and incubation overnight at 37 °C. Isolation of N-glycopeptides from tryptic digests of mouse plasma was performed using a modified glycopeptide capture procedure. The 20 µL of mouse plasma was first diluted with an equivalent volume of 100 mM NH4HCO3, and proteins were denatured by slowly adding 2,2,2-trifluoroethanol (TFE; J.T. Baker, Philipsburg, NJ) to a 50% (v/v) final concentration. After cystine residues were reduced and alkylated as described above, each sample was further diluted 10-fold with 100 mM NH4HCO3 (pH 8.3) after spiking in [14C]-iodoacetamide-labeled AGP. Proteins were digested with trypsin overnight at 37 °C and cleaned using C18 spin columns (500 mg, Waters, Milford, MA). The 5 mL of eluate was then dried down to ∼500 µL by SpeedVac and mixed with another 500 µL of oxidation buffer (pH 5.0) before being oxidized in 10 mM NaIO4 for 1 h at room temperature in the dark. Peptides were cleaned up again with the C18 column and directly coupled to different amounts of hydrazide resin in elution buffer (0.1% TFA in 80% ACN) overnight at room temperature. Nonglycosylated peptides were removed by washing the resin three times each with 1 mL of 1.5 M NaCl, 80% ACN, water, and 50 mM NH4HCO3. N-Glycopeptides were then released with different amount of PNGase F as described in the Results and Discussion section for Figure 5D. Monitoring the Performance of Each Step of the Glycopeptide Capture Procedure. The radioactivity of standard proteins spiked into the mouse plasma at each step of the procedure was measured by a liquid scintillation analyzer with 4 mL of scintillation solution (National Diagnostics, Atlanta, GA). The data obtained were evaluated by one-way analysis of variance (ANOVA) and Newman-Keuls test using a commercially available computer software program (Prism 3.0, GraphPad, CA). At least three measurements were taken for each experiment, where p < 0.05 was accepted as significant. Stable Isotope Labeling of N-Glycopeptides with Succinic Anhydride. Before labeling, PNGase F-released N-glycopeptides were extracted from the resin, dried using a SpeedVac, and resuspended in DMF/pyridine/H2O (50%/10%/40% (v/v/v)). Light (succinic-d0-anhydride, Sigma) or heavy (succinic-d4-anhydride, C/D/N Isotopes, Pointe-Claire, Quebec, Canada) succinic anhydride solutions were added to a final concentration of 2 mg/mL. The samples were incubated at room temperature for 2 h, dried down using a SpeedVac, and resuspended in 0.1% trifluoroacetic acid (TFA). “Light” and “heavy” pairs were further mixed in a 1:1 ratio (v/v) and desalted using C18 columns (Waters) before being analyzed on an LCQ Classic ion trap mass spectrometer (ThermoFinnigan, San Jose, CA).34 5828
Analytical Chemistry, Vol. 79, No. 15, August 1, 2007
Peptide Identifications via MS/MS and Database Search. Peptides were analyzed using an ion trap mass spectrometer (LCQ Classic). Acquired MS/MS spectra were converted to mzXML files35 and searched against the International Protein Index (IPI) human protein database (version 3.16) using SEQUEST software.36 The database sequence tool was set to recognize the following modifications: carboxymethylated cysteines (mass +57 Da), oxidized methionines (mass +16 Da), and an enzyme-catalyzed conversion of Asn to Asp at the site of carbohydrate attachment (mass +0.984 Da). The mass tolerance of peptides is 3.0 Da between the measured mass and the calculated mass. Database search results were statistically analyzed using PeptideProphet, which effectively computes a probability for the likelihood of each identification being correct (on a scale of 0-1) in a data-dependent fashion.37 A minimum PeptideProphet probability score (P) filter of 0.9 was used to remove low-probability peptides. Since N-glycosylation occurs at a consensus N-X-S/T sequon, the remaining peptide sequences were additionally filtered to remove non-motif-containing peptides (Table 2). Relative peptide abundance in heavy compared to light was calculated using an algorithm termed the automated statistical analysis of protein abundance (ASAP) ratio.38 RESULTS AND DISCUSSION Establish a Method Using Protein Standards To Monitor the Performance of Each Step of the SPEG Procedure. If N-glycopeptides from each plasma protein can be specifically and efficiently isolated, the number of peptides from plasma proteins will be reduced, and the analysis of plasma proteins can be greatly simplified. For optimal and reproducible performance, each step in the SPEG procedure needs to be adapted to plasma samples and optimized to increase yield and specificity. To evaluate the procedure for plasma analysis and to optimize each step, we first established a method to determine the specificity and efficiency using standard proteins. These standards are also useful to monitor and analyze the sample preparations from multiple analyses performed at different times and by different laboratories to ensure their reproducibility. For this purpose, we selected two human plasma glycoproteins (AGP and AT), with [14C]-iodoacetamide-labeled cysteine residues, as standards to monitor each step of the procedure (Table 1). AT and AGP are classic plasma proteins, with concentrations in the range of 0.1-1 mg/mL, or 2-20 µM in normal human plasma. AGP (Swiss-Prot, P02763) is an acute phase plasma protein with 183 amino acids, a molecular weight of 41 000-43 000, 5 known N-glycosites, and 4 cysteine residues.39,40 Two cysteines (50%) are (34) Zhang, H.; Yi, E. C.; Li, X. J.; Mallick, P.; Kelly-Spratt, K. S.; Masselon, C. D.; Camp, D. G., 2nd; Smith, R. D.; Kemp, C. J.; Aebersold, R. Mol. Cell. Proteomics 2005, 4, 144-155. (35) Pedrioli, P. G.; Eng, J. K.; Hubley, R.; Vogelzang, M.; Deutsch, E. W.; Raught, B.; Pratt, B.; Nilsson, E.; Angeletti, R. H.; Apweiler, R.; Cheung, K.; Costello, C. E.; Hermjakob, H.; Huang, S.; Julian, R. K.; Kapp, E.; McComb, M. E.; Oliver, S. G.; Omenn, G.; Paton, N. W.; Simpson, R.; Smith, R.; Taylor, C. F.; Zhu, W.; Aebersold, R. Nat. Biotechnol. 2004, 22, 1459-1466. (36) Eng, J. M.; A. L.; Yates, J. R., 3rd. J. Am. Soc. Mass Spectrom. 1994, 5, 976-989. (37) Keller, A.; Nesvizhskii, A. I.; Kolker, E.; Aebersold, R. Anal. Chem. 2002, 74, 5383-5392. (38) Li, X. J.; Zhang, H.; Ranish, J. A.; Aebersold, R. Anal. Chem. 2003, 75, 6648-6657. (39) Hochepied, T.; Berger, F. G.; Baumann, H.; Libert, C. Cytokine Growth Factor Rev. 2003, 14, 25-34.
Table 1. Sequences of Two Human Blood N-Glycoprotein Standards: AGP (183 AA) and AT (394 AA)a glycoprotein
tryptic peptide sequences
cysteine
R-1-acid glycoprotein 1 (AGP)
QIPLCANLVPVPITNATLDQITGK NEEYNK.S SVQEIQATFFYFTPNK.T QDQCIYNTTYLNVQR ENGTISR EQLGEFYEALDCLRIPK DKCEPLEK QLAHQSNSTNIFFSPVSIATAFAMLSLGTK ADTHDEILEGLNFNLTEIPEAQIHEGFQELLR LGMFNIQHCK YLGNATAIFFLPDEGK VVNPTQK
x
R-1-antitrypsin (AT)
x x x x
N-X-S/T motif
glycan provedb
x x x x x
x x x x x
x x
x x
x x
x x
x x
x
x
identifiedc x
a AGP was used as a control for tryptic N-glycopeptides. Of the four cysteine residues in AGP, two are located within tryptic N-glycopeptides. AT contains only one cysteine residue located outside of any tryptic N-glycopeptides and serves as a control to monitor the specificity of SPEG. The sequences of tryptic glycopeptides are shown, while the N-X-S/T motifs are highlighted in italic type and cysteine residues in boldface type. b Data from UniProtKB/Swiss-Prot (http://au.expasy.org/). c N-Glycopeptides identified in this study.
located within tryptic N-glycopeptidessidentifiable using SPEG and tandem mass spectrometry.28 In addition, AGP normally has complex branched N-glycan structures, which are relatively difficult to be released from AGP by PNGase F. Therefore, AGP represents a type of N-linked glycoprotein whose N-glycopeptides are difficult to extract and serves as a good control to evaluate the yield of N-glycopeptides with PNGase F.41 AT (Swiss-Prot, P01009) is a 52 kDa glycoprotein in human plasma and functions as protease inhibitor to protect tissues from proteolytic enzymes.42 It contains 394 amino acids, 4 N-glycosylation sequons, and 1 cysteine residue. All but the C-terminal N-glycosylation sequon are occupied, i.e., glycosylated.42 Because the cysteine residue is located outside of the tryptic N-glycopeptides, AT can be used to monitor the completion of tryptic digestion and overall specificity of the glycopeptide isolation. Furthermore, as an inhibitor of trypsin in plasma, AT represents a protein that is difficult to digest. It is worth mentioning that, according to data currently available, there is no O-glycosylation site identified in either standard protein (data from UniProtKB/Swiss-Prot, http://au.expasy.org/). Standard proteins were labeled with [14C]-iodoacetamide and purified immediately from free [14C]-iodoacetamide as described in the Experimental Section. The labeling efficiency of cysteine residues was calculated by comparing the measured radioactivity to the theoretical radioactivity with all Cys being completely labeled. We determined the labeling efficiencies to be in the range of 3.80-4.05 [14C]-iodoacetamide per AGP molecule and 0.991.12 per AT molecule. This suggested that the labeling of Cys in AGP and AT was near completion. To determine whether the coupling efficiency measured by radioactivity of 14C-labeled standard proteins correlates with the yield of N-glycopeptides, we compared the coupling efficiencies measured by radioactivity with released glycopeptides determined by stable isotope labeling and tandem mass spectrometry. We found that, in different oxidation conditions, the two methods (40) Imre, T.; Schlosser, G.; Pocsfalvi, G.; Siciliano, R.; Molnar-Szollosi, E.; Kremmer, T.; Malorni, A.; Vekey, K. J. Mass Spectrom. 2005, 40, 14721483. (41) Chandrasekaran, E. V.; Davila, M.; Nixon, D.; Mendicino, J. Cancer Res. 1984, 44, 1557-1567. (42) Kolarich, D.; Weber, A.; Turecek, P. L.; Schwarz, H. P.; Altmann, F. Proteomics 2006, 6, 3369-3380.
produced a very similar pattern (Figure 1). However, the quantitative analysis using radioactivity produced higher sensitivity and accuracy. This allows us to monitor the performance of each step of SPEG using the 14C-radioactive labeling assay and to optimize the glycopeptide capture method for plasma analysis. It is worth mentioning that, compared to other commonly used protein labeling methods (e.g., fluorescence labeling and biotin labeling), radioactive labeling offers the highest sensitivity (
