Dissociation Profile of Protonated Fucosyl Glycopeptides and Quantitation of Fucosylation Levels of Glycoproteins by Mass Spectrometry Michiko Tajiri,†,‡ Machiko Kadoya,† and Yoshinao Wada*,† Department of Molecular Medicine, Osaka Medical Center and Research Institute for Maternal and Child Health, 840 Murodo-cho Izumi, Osaka 594-1101, Japan, and CREST, Japan Science and Technology Agency, 4-1-8 Honcho Kawaguchi, Saitama 332-0012, Japan Received September 7, 2008
Mass spectrometry of glycopeptides is an efficient strategy for profiling glycans at specific sites in glycoproteins. To assess the reliability of this method for determining the fucosylation levels of glycoproteins, we conducted mass spectrometry of fucosylated glycopeptides from transferrin and haptoglobin. The biantennary glycans containing antenna R1,3/4 fucose or R1,6 core fucose showed different fragmentation behaviors in collision-induced dissociation of protonated glycopeptides. Stability was dependent on peptide backbone sequences. The major dissociation, cleavage of the GlcNAcβ1f2Man linkage of antenna, was evident at a slightly lower activation energy for the core fucosylated species, while the linkage of R1,6 core fucose was more stable than that of antenna R1,3/4 fucose. However, these fragmentations were induced only with sufficient loading of activation energy. The quantitation of fucosylated glycans by mass spectrometry of glycopeptides, without collisional activation, was thus justified. The fucosylation levels calculated from the signal intensities in electrospray (nanospray) ionization and ultraviolet matrix-assisted laser desorption/ionization mass spectra were essentially the same. The mass spectrometric profiling of glycopeptides from transferrin of congenital disorders of glycosylation (CDG-Ia and CDG-IIc) patients demonstrated that the elevation or reduction of fucosylation in pathological conditions can be reliably determined by MS of glycopeptides. Keywords: fucose • glycopeptide • mass spectrometry • oligosaccharide • N-linked
Introduction Glycoproteomics is an emerging topic bridging proteomics and glycomics. To support the development of this field, a rapid and sensitive method of elucidating glycan structures is essential for promoting studies on this class of post-translational modifications. Mass spectrometry (MS) of glycans is widely used for this purpose,1-3 and the fragmentation profiles obtained by collision-induced dissociation are used for discriminating different types of glycans including isomers.4-8 Also, profiling, which means relative quantitation of each oligosaccharide, is carried out based on the signal intensities observed in matrix-assisted laser desorption/ionization (MALDI) mass spectra of permethylated glycans or electrospray ionization (ESI) mass spectra of underivatized glycans, and this approach was demonstrated to be as reliable as conventional chromatography-based methods in a study by the Human Proteome Organization Initiatives.9 On the other hand, MS of glycopeptides has also attracted considerable attention, as it allows site-specific profiling of the * To whom correspondence should be addressed. Tel: +81-725-57-4105. E-mail:
[email protected]. † Osaka Medical Center and Research Institute for Maternal and Child Health. ‡ Japan Science and Technology Agency.
688 Journal of Proteome Research 2009, 8, 688–693 Published on Web 12/19/2008
glycans attached to glycoproteins.10-12 Glycopeptide samples for MS are obtained by proteolysis from glycoproteins and then subjected to LC-MS or directly to MS with loss of neither samples nor time. MS of glycopeptides is performed in positive ion mode detecting their protonated ions, [M + H]+. Because protonation occurs at the basic sites of peptides, the signal intensity of the ions is dependent on the proton affinity of the peptide portion but not on glycans, rationalizing the quantitation or profiling of attached glycoforms by means of their signal intensities.9 However, there remains a concern about the dissociation of specific glycoforms during ionization as observed for sialic acids. In addition, fucosylated species should be considered as well, since the fucose residue is a labile group in collision-induced dissociation (CID) of glycans derivatized by reductive amination,13 and migration and rearrangement of fucose residues during CID have been noted as an intriguing phenomenon in the MS of glycans.14,15 This migration occurs in the protonated but not alkali-cationized species of glycans or glycopeptides in the case of poor charge (proton) fixation,14 and charge-directed fragmentation is suggested to underlie this mechanism.16 For the N-glycans labeled with 2-aminobenzamide at their reducing ends, efficient transfer occurs between antennae and migration from core fucosylated glycans has also been reported.17,18 10.1021/pr800727w CCC: $40.75
2009 American Chemical Society
Dissociation Profile of Protonated Fucosyl Glycopeptides
research articles
Fucosylated glycans are involved in a variety of biological and pathological processes in eukaryotic organisms including tissue development, angiogenesis, fertilization, cell adhesion, inflammation, and tumor metastasis.19 In humans, the major site of fucosylation of N-glycans are antennae which undergo subterminal fucosylation in an R1,3 or R1,4 linkage to GlcNAc and core fucosylation in an R1,6 linkage to the innermost GlcNAc. The former constitutes the Lewis-related carbohydrate ligands which are involved in various biological processes.20 Core fucosylation has also attracted research attention in relation to the cytotoxicity of immunoglobulins and to cancer markers.21,22 Considering the biological implications of fucosylation, the reliability of profiling of fucosylated glycopeptides by MS requires confirmation. In contrast to a number of studies on fucose migration, the present investigation focuses on the stability of fucosylated glycopeptide ions in terms of quantitation or profiling but does not aim at structural characterization by CID. The dissociations of antenna R1,3/4 and R1,6 core fucosylated isomers of N-glycopeptides were studied as a function of collision energy. Although the charge state and the sequences of the peptide backbones of ions affected the cleavage of fucosyl and other glycosidic linkages, the overall results justified the quantitation of fucosylation levels based on the signal intensity of the mass spectrum. This method was applied to a typical example of fucosylation disorders, congenital disorders of glycosylation (CDG).
Experimental Section Preparation of Glycopeptides. Transferrin (TRFE), haptoglobin (HPT), and fetuin A were purified from serum of one of the authors (Y.W.) by affinity chromatography using rabbit immunoglobulins (Dako, Glostrup, Denmark) raised against each protein and an NHS-activated agarose column (GE Healthcare, Buckinghamshire, UK). Immunoglobulin-G (IgG) was purified by a Hi-Trap Protein G column (GE Healthcare). After reduction and carboxamidomethylation, TRFE and HPT were digested with a mixture of trypsin (Promega, Buckinghamshire, UK) and lysylendopeptidase (Wako, Osaka, Japan). Glycopeptides were extracted from the digest by a hydrophilic affinity method10 and was separated on a 1.0 mm × 150 mm C18 reversed phase column (Develosil, Nomura Kagaku, Seto, Japan). The purified glycopeptides from TRFE and HPT were desialylated with Arthrobacter neuraminidase (Nacalai tesque, Kyoto, Japan) in 50 mM sodium phosphate, pH 6.0, and divided into two parts. One part was loaded on an LCA (Lens culinaris) lectin-agarose column, and the flow-through fraction, which should contain antenna R1,3/4-fucosylated species, was collected for MS. The purity was confirmed by the resistance to β-galactosidase (from Jack bean)23 (Supplemental Figure 1, Supporting Information). Another part was treated with an R-Lfucosidase from bovine kidney (Sigma-Aldrich, St. Louis, MO) to remove the fucose residues and the core R1,6-fucosylated glycopeptides were then reconstructed using recombinant mammalian R1,6-fucosyltransferase (FUT8) produced in insect cells (a gift from Prof. Miyoshi)24 and GDP-β-L-fucose (Wako) in a 100 mM MES-NaOH buffer, pH 7.0. Mass Spectrometry. CID (MS/MS) experiments were carried out on an LCQ Deca XP ion trap mass spectrometer (ThermoFisher, San Jose, CA) equipped with a nanospray ion source (AMR, Tokyo, Japan). The glycopeptide sample was dissolved in a 0.1% formic acid/20% methanol solution at a concentration of 1-5 µM, and loaded in a PicoTip emitter (New Objective,
Figure 1. Collision-induced dissociation profiles of (a) triply or (b) doubly protonated glycopeptide TRFE-gp1 (CGLVPVLAENYNK; C for carboxamidomethylcysteine) containing a biantennary oligosaccharide. Missing groups are indicated in parentheses.
Woburn, MA) for infusion mode nanospray ionization MS. The nanospray source was given 1.2-1.4 kV of applied voltage, while the inlet capillary was held at 3.0-45.4 V and 200 °C. Other mass spectrometric parameters were as follows: tube lens offset at -60.0 to +55.0 V, multipole 1 offset at -5.13 to -2.34 V, multipole 2 offset at -13.2 to -8.0 V, intermultipole lens at -41.2 to -19.0 V, entrance lens at -95.6 to -36.4 V and trap DC offset at -10.12 to -9.87 V. Doubly or triply protonated glycopeptides containing specific glycoforms were selected as a precursor for the collisional activation. The collision gas was helium, and MS/MS scans were acquired using an activation qz of 0.250, activation time of 30 ms, and varying collision energies of up to 20% (normalized collision energy given by the instrument) were applied. The % intensities of precursor and product ions relative to the total ion intensity were plotted against the collision energy. Measurement of Fucosylation Levels by MS. TRFE was immunopurified from tested subjects for CDG screening, and a glycopeptide fraction from the tryptic digest of transferrin was prepared by the hydrophilic affinity method as described above. A part of the glycopeptides from IgG was defucosylated by R-L-fucosidase and then mixed with the untreated counterpart at different ratios. For MALDI MS, equal amounts of the glycopeptide solution and a MALDI sample matrix solution [10 mg/mL of recrystallized 2,5-dihydroxbenzoic acid (Wako) dissolved in 30% acetonitrile, 0.1% trifluoroacetic acid] were mixed on the sample Journal of Proteome Research • Vol. 8, No. 2, 2009 689
research articles
Tajiri et al.
Figure 2. MS/MS spectra of doubly protonated TRFE-gp1 containing (a) antenna R1,3 or (b) core R1,6 fucose at a collision energy of 18% and their dissociation profiles (c and d correspond to a and b, respectively). Missing groups are indicated in parentheses.
target. The mass spectra were acquired using a Voyager DE Pro time-of-flight (TOF) mass spectrometer (Applied Biosystems, Foster City, CA) equipped with a nitrogen 337 nm laser in the linear mode. The signals from about 500 laser shots were accumulated to form a spectrum, and relative quantitation was based on the signal intensities (heights) of the corresponding ions in the mass spectrum. For comparison, nanospray ionization MS was carried out on an LTQ XL (Thermo-Fisher). The sample solution was prepared for the analysis by LCQ described above.
Results and Discussion Dissociation from Different Charge State Precursors. Protonation of glycopeptides occurs mainly at the peptide portion, but the charge state-dependent dissociation of protonated glycopeptides has not been studied in detail. The major species generated by ESI of a glycopeptide TRFE-gp1 (CGLVPVLAENYNK; C for carboxamidomethylcysteine), which contains the N-glycosylation site Asn432 (linked to Lys) of TRFE and has two amino groups at the N-terminus and lysine, was doubly charged, but triply charged ions were also found in the mass spectrum (data not shown). In the tandem MS of the triply protonated [M + 3H]3+ precursor, the major dissociation occurred at the glycosidic linkage between the trimannosyl core and an antenna (Figure 1). In this case, one mobile proton was probably localized at the GlcNAc nitrogen, leading to bond cleavage via a “charge-directed” fragmentation mechanism. Indeed, the resulting Y-ions containing the peptide portion 690
Journal of Proteome Research • Vol. 8, No. 2, 2009
were doubly charged. On the other hand, from a doubly charged precursor in which the protons were, to a large extent, localized on the amino groups of the peptide, more diverse fragmentation occurred at the glycosidic bonds of the glycan. The charge state of Y-ions was unaltered in this “chargeremote” type of fragmentation. These results indicate the fragmentation pattern of glycopeptide ions to be dependent on the charge state of the precursor ions. In the following experiments, the glycopeptide ions having the same number of protons as the basic groups in peptides were studied. The major dissociation pathway in CID was found to be the cleavage of a GlcNAcβ1f2Man linkage in either of the isoforms. A preferential cleavage of this linkage in the ManR1f6 antenna as compared with the ManR1f3 antenna has been reported for the [M + Na]+ ions of the reductively aminated biantennary oligosaccharides by means of isotopic labeling of either antenna.25,26 This rule cannot be applied directly to glycopeptides, because the dissociation of protonated and alkalicationized ions occurs through different mechanisms each other. Dissociation of Fucosylated Isomers. The R1,3 Lewis fucose of TRFE-gp1 is located in the ManR1f6 antenna.27,28 The doubly protonated ions [M+2H]2+ of the core R1,6 and antenna R1,3 fucosylated isomers of TRFE-gp1 were analyzed by CID. As shown in Figures 2a and 2b, dissociation of the core fucosylated isomer occurred at a 15% collision energy mainly at the antenna GlcNAcβ1f2Man bond, while the antenna R1,3 fucosylated isomer was apparently unchanged with this level
Dissociation Profile of Protonated Fucosyl Glycopeptides
research articles
Figure 3. MS/MS spectra of triply protonated HPT-gp4 (VVLHPNYSQVDIGLIK) containing (a) antenna R1,3 or (b) core R1,6 fucose at a collision energy of 15% and their dissociation profiles (c and d correspond to a and b, respectively). Missing groups are indicated in parentheses. Symbols are the same with those indicated in Figure 2.
Figure 4. The relative ion intensities of precursor (a, b) and defucosylated (c, d) ions as a function of collision energy. TRFE-gp1 (9), TRFE-gp2 (0) and HPT-gp4 (2), each containing (a, c) antenna R1,3 or (b, d) core R1,6 fucose.
of activation. When higher energy was loaded, the antenna R1,3 fucosylated isomer generated two types of product ions derived from cleavage at either GlcNAc-Man in the ManR1f3 antenna or the Fuc-GlcNAc linkage (Figure 2c), and the core fucosylated isomer generated simple product ion mass spectra showing the predominant dissociation at the GlcNAc-Man linkage (Figure 2d). The intensity for the ion corresponding to [peptide +
3HexNAc4Hex] was much higher in the antenna fucosylated species compared with the core fucosylated one, because the latter requires cleavage of two bonds to give this product ion. To investigate whether the dissociation profile was dependent on the peptide sequence, other glycopeptides were analyzed. The triply protonated ions [M+3H]3+ of the glycopeptide HPT-gp4 (VVLHPNYSQVDIGLIK), containing an N-glycosylaJournal of Proteome Research • Vol. 8, No. 2, 2009 691
research articles
Figure 5. Correlation of the quantitation values for fucosylation obtained by different ionization methods. ESI-MS was performed using an ion trap mass spectrometer with nanospray ionization. MALDI MS was performed with UV-MALDI and linear TOF mode, and the measurements were repeated four times. (error bar, SD). For % fucosylation, the ratios of unfucosylated glycopeptides to fucosylated ones were calculated based on the signal intensities of the corresponding ions with a biantennary oligosaccharide. y ) 0.955x - 0.247 (r2 ) 0.977) TRFE-gp1 (9), TRFE-gp2 (0), fetuin A (1), IgG (b). Doubly (TRFE-gp1 and IgG) or triply (TRFE-gp2 and fetuin A) charged ions in the ESI mass spectra were used for calculation. Amino acid sequences of the glycopeptides from fetuin A and IgG were AALAAFNAQNNGSNFQLEEISR and EEQFNSTFR, respectively.
tion site at Asn241 of HPT, presented a distinct dissociation profile from TRFE-gp1 as shown in Figure 3. The product ion spectra from two different isomers of HTP-gp4 were similar to each other, both indicating predominant cleavage at the GlcNAc-Man linkage. This was most probably due to the fact that mobile protons were available for localization at the antenna GlcNAc in the [M + 3H]3+ ions of HPT-gp4 compared with the [M + 2H]2+ ions of TRFE-gp1. On the other hand, the dissociation profiles of the [M + 3H]3+ ions of TRFE-gp2 (QQQHLFGSNVTDCSGNFCLFR; C for carbamidomethylcysteine) containing Asn630 of TRFE were similar to those of TRFEgp1 (Supplemental Figure 2, Supporting Information), because the TRFE-gp2 sequence fixed a proton at Arg. In this way, the stability of protonated glycopeptide ions in CID may be dependent on the peptide backbone sequences. For example, the dissociation of triply charged species of antenna-fucosylated HPT-gp4 and core fucosylated TRFE-gp2 occurred at the same collision energy of 14%. It is noteworthy that the major dissociation pathway was the cleavage of the GlcNAc-Man bond, rather than the fucosyl linkage, in either case. Generation of defucosylated ions upon collisional activation is summarized in Figure 4c and d. For the glycopeptides analyzed herein, the generation of defucosylated ions with a biantennary oligosaccharide chain was more prominent in antenna fucose than core fucose. For all these glycopeptides, cleavage of the GlcNAcβ1f2Man linkage of core fucosylated species occurred at a lower collision energy than the antenna fucosylated isoform, indicating the latter species to be more stable in CID (Figure 4a and b). The reason for this difference is unclear, but the collision cross-section may be larger for the core fucosylated isomer. In fact, an effect of core fucosylation to restrict the flexibility of the ManR1-6 antenna has been suggested.29 692
Journal of Proteome Research • Vol. 8, No. 2, 2009
Tajiri et al.
Figure 6. MALDI MS of TRFE-gp2 from a healthy individual and a CDG-IIc patient. TRFE-gp2 was not desialylated before measurements, and MS was performed in linear TOF mode. Arrows indicate m/z4869 corresponding to fucosylated glycopeptide ions.
Overall, however, the dissociation of trapped ions was evident only when a significant level of activation energy was applied. These observations indicated that relative quantitation of glycoforms including fucosylated species by the MS of glycopeptides would be justified. Measurement of Fucosylation Levels by MALDI MS and Its Application. A concern remains about quantitation by MALDI MS. MALDI MS gives higher internal energy to the ions than ESI-MS, and the generated ions undergo two types of dissociation called in-source and postsource decays, both of which often occur in the glycan portion of glycopeptides ions. Parameters such as laser wavelength and power, extraction voltages, and “hot” or “cool” matrices directly influence the internal energy of generated ions. Despite these problems, regarding oligosaccharides, there are publications describing good quantitation based on the signal intensities observed in the MALDI mass spectrum.30 To justify the use of MALDI MS of glycopeptides with respect to fucosylation, measurements of fucosylation levels were carried out by ESI (nanospray ionization) and MALDI MS on four different glycopeptides from TRFE (gp1 and gp2), fetuin A and IgG. Fucosylation levels were calculated based on the signal intensities of unfucosylated and fucosylated glycopeptide ions. The values obtained from different ionization methods correlated quite well y ) 0.955x 0.247 (r2 ) 0.977) as shown in Figure 5. Finally, MALDI MS of glycopeptides was applied to the analysis of two different types of CDG, among which increased fucosylation has been reported for CDG-I as a secondary effect of the defective glycosylation.31 MALDI MS of TRFE-gp2 identified the elevation in the patients found in our screening study: 16.7, 22.6, and 32.4% for three CDG-Ia patients vs 9.4 ( 4.9% (mean ( SD) for 223 control individuals. The analysis was then applied to CDG-IIc, which is caused by reduced transport of GDP-fucose into Golgi.32-34 The decrease in fucosylated glycans of fibroblasts manifests as a deficiency of core fuco-
research articles
Dissociation Profile of Protonated Fucosyl Glycopeptides 35,36
sylation of N-linked oligosaccharides of glycoproteins, but analysis of serum proteins has not previously been clearly described. In the present study, MALDI MS of glycopeptides demonstrated no substantial fucosylation of serum TRFE from three CDG-IIc patients (Figure 6 and Supplemental Figure 3, Supporting Information).
Conclusion In the CID analysis of protonated glycopeptide ions, core fucosylated N-glycan isomers were more labile than the antenna fucosylated species, while dissociation of the fucosyl linkage of the Lewis element was more evident than that of the core fucose. The major dissociation pathway of the glycopeptides containing biantennary oligosaccharides was cleavage at the GlcNAcβ1-2Man linkage in the antenna. These fragmentations only occurred when sufficient activation energy was loaded, and thus the quantitation of fucosylated glycans by the MS (ESI and MALDI) of glycopeptides was justified. Mass spectrometric profiling of glycopeptides from CDG patient serum demonstrated that the method is able to reliably determine the elevation or reduction of fucosylation in pathological conditions.
Acknowledgment. We thank Prof. Miyoshi and Prof. Tonetti for generously providing recombinant mammalian R1,6-fucosyltransferase (FUT8) and the serum from three CDG-IIc patients.
Supporting Information Available: Supplemental Figures 1-3. This material is available free of charge via the Internet at http://pubs.acs.org.
(10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29)
References (1) (2) (3) (4) (5) (6) (7) (8) (9)
Harvey, D. J. Mass Spectrom. Rev. 1999, 18, 349–450. Dell, A.; Morris, H. R. Science 2001, 291, 2351–2356. Zaia, J. Mass Spectrom. Rev. 2004, 23, 161–227. Harvey, D. J.; Bateman, R. H.; Green, M. R. J. Mass Spectrom. 1997, 32, 167–187. Harvey, D. J. J. Am. Soc. Mass Spectrom. 2000, 11, 900–915. Desaire, H.; Leary, J. A. Anal. Chem. 1999, 71, 1997–2002. Takegawa, Y.; Ito, S.; Yoshioka, S.; Deguchi, K.; Nakagawa, H.; Monde, K.; Nishimura, S. Rapid Commun. Mass Spectrom. 2004, 18, 385–391. Kurimoto, A.; Kanie, O. Rapid Commun. Mass Spectrom. 2007, 21, 2770–2778. Wada, Y.; Azadi, P.; Costello, C. E.; Dell, A.; Dwek, R. A.; Geyer, H.; Geyer, R.; Kakehi, K.; Karlsson, N. G.; Kato, K.; Kawasaki, N.; Khoo,
(30) (31) (32) (33) (34) (35) (36)
K. H.; Kim, S.; Kondo, A.; Lattova, E.; Mechref, Y.; Miyoshi, E.; Nakamura, K.; Narimatsu, H.; Novotny, M. V.; Packer, N. H.; Perreault, H.; Peter-Katalinic, J.; Pohlentz, G.; Reinhold, V. N.; Rudd, P. M.; Suzuki, A.; Taniguchi, N. Glycobiology 2007, 17, 411– 422. Wada, Y.; Tajiri, M.; Yoshida, S. Anal. Chem. 2004, 76, 6560–6565. Tajiri, M.; Yoshida, S.; Wada, Y. Glycobiology 2005, 15, 1332–1340. Wuhrer, M.; Catalina, M. I.; Deelder, A. M.; Hokke, C. H. J. Chromatogr., B: Anal. Technol. Biomed. Life Sci. 2007, 849, 115–128. Harvey, D. J. Rapid Commun. Mass Spectrom. 2000, 14, 862–871. Ernst, B.; Mu ¨ ller, D. R.; Richter, W. J. Int. J. Mass Spectrom. Ion Processes 1997, 160, 283–290. Harvey, D. J.; Mattu, T. S.; Wormald, M. R.; Royle, L.; Dwek, R. A.; Rudd, P. M. Anal. Chem. 2002, 74, 734–740. Ma, Y. L.; Vedernikova, I.; Van den Heuvel, H.; Claeys, M. J. Am. Soc. Mass Spectrom. 2000, 11, 136–144. Wuhrer, M.; Koeleman, C. A.; Hokke, C. H.; Deelder, A. M. Rapid Commun. Mass Spectrom. 2006, 20, 1747–1754. Wuhrer, M.; Deelder, A. M. Anal. Chem. 2005, 77, 6954–6959. Ma, B.; Simala-Grant, J. L.; Taylor, D. E. Glycobiology 2006, 16, 158R–184R. Ley, K. Trends Mol. Med. 2003, 9, 263–268. Miyoshi, E.; Noda, K.; Yamaguchi, Y.; Inoue, S.; Ikeda, Y.; Wang, W.; Ko, J. H.; Uozumi, N.; Li, W.; Taniguchi, N. Biochim. Biophys. Acta 1999, 1473, 9–20. Taniguchi, N.; Miyoshi, E.; Gu, J.; Honke, K.; Matsumoto, A. Curr. Opin. Struct. Biol. 2006, 16, 561–566. Lucka, L.; Fernando, M.; Grunow, D.; Kannicht, C.; Horst, A. K.; Nollau, P.; Wagener, C. Glycobiology 2005, 15, 87–100. Ihara, H.; Ikeda, Y.; Taniguchi, N. Glycobiology 2006, 16, 333–342. Kato, K.; Yamaguchi, Y.; Takahashi, N.; Nishimura, M.; Iwamoto, S.; Sekiya, S.; Tanaka, K. J. Mass Spectrom. Soc. Jpn. 2004, 52, 284– 288. Kameyama, A.; Nakaya, S.; Ito, H.; Kikuchi, N.; Angata, T.; Nakamura, M.; Ishida, H. K.; Narimatsu, H. J. Proteome Res. 2006, 5, 808–814. van Rooijen, J. J.; Jeschke, U.; Kamerling, J. P.; Vliegenthart, J. F. Glycobiology 1998, 8, 1053–1064. Satomi, Y.; Shimonishi, Y.; Hase, T.; Takao, T. Rapid Commun. Mass Spectrom. 2004, 18, 2983–2988. Stubbs, H. J.; Lih, J. J.; Gustafson, T. L.; Rice, K. G. Biochemistry 1996, 35, 937–947. Harvey, D. J. Rapid Commun. Mass Spectrom. 1993, 7, 614–619. Callewaert, N.; Schollen, E.; Vanhecke, A.; Jaeken, J.; Matthijs, G.; Contreras, R. Glycobiology 2003, 13, 367–375. Lu ¨ bke, T.; Marquardt, T.; von Figura, K.; Ko¨rner, C. J. Biol. Chem. 1999, 274, 25986–25989. Lu ¨ bke, T.; Marquardt, T.; Etzioni, A.; Hartmann, E.; von Figura, K.; Ko¨rner, C. Nat. Genet. 2001, 28, 73–76. Lu ¨hn, K.; Wild, M. K.; Eckhardt, M.; Gerardy-Schahn, R.; Vestweber, D. Nat. Genet. 2001, 28, 69–72. Marquardt, T.; Lu ¨ hn, K.; Srikrishna, G.; Freeze, H. H.; Harms, E.; Vestweber, D. Blood 1999, 94, 3976–3985. Sturla, L.; Fruscione, F.; Noda, K.; Miyoshi, E.; Taniguchi, N.; Contini, P.; Tonetti, M. Glycobiology 2005, 15, 924–934.
PR800727W
Journal of Proteome Research • Vol. 8, No. 2, 2009 693
