Structural Characterization of N-Glycopeptides by Matrix-Dependent

The fragments were further accelerated by 19 kV in the LIFT cell (LIFT means “lifting” the ... Figure 3 Structural analysis of sialooligosaccharid...
0 downloads 0 Views 300KB Size
Download PDF
Anal. Chem. 2004, 76, 6097-6101

Structural Characterization of N-Glycopeptides by Matrix-Dependent Selective Fragmentation of MALDI-TOF/TOF Tandem Mass Spectrometry Masaki Kurogochi and Shin-Ichiro Nishimura*

Division of Biological Sciences, Graduate School of Science, Frontier Research Center for the Post-Genomic Science and Technology, Hokkaido University, N21, W11, Sapporo 001-0021, Japan

The matrix-assisted laser desorption/ionization time-offlight (MALDI-TOF) MS technique described to date has proven to be a convenient and rapid method for identification and characterizations of proteins. However, the general MALDI-TOF MS analysis of complex carbohydrates and glycopeptides still entails special consideration of ionization and the fragmentation characteristics of labile carbohydrate moieties. In this study, an efficient and practical method we termed the matrix-dependent selective fragmentation (MDSF) technique of MALDI-TOF/TOF MS, which allows highly sensitive and reliable fragmentation of oligosaccharides and N-glycopeptides. Results from application of the MDSF technique to TOF/TOF MS analysis demonstrated that in comparison to the conventional postsource decay up to 170 times more sensitive product ion peaks could be obtained. It was also suggested that MDSF generates desired structural information based on the controlled cleavage of the singly charged precursor ion with different electronic excited states made by this method. Ideal product ion peaks observed by MDSF in TOF/TOF MS facilitated structural characterization of complex oligosaccharide derivatives including unstable Neu5Ac and Fuc residues and N-glycopeptides. Glycosylation is one of the most ubiquitous and crucial processes in posttranslational modifications of proteins observed in eukaryotic systems.1 It is estimated that over 90% of all mammalian proteins may exist in the glycosylated form at some point through their existence. During the past 10 years, mass spectrometry using electrospray ionization (ESI) and matrixassisted laser desorption/ionization (MALDI) has become a powerful and versatile tool for rapid identification and characterizations of proteins.2 In addition, a variety of methods that combine MALDI time-of-flight (TOF) MS or LC/ESI-quadrupole/TOF MS * To whom correspondence should be addressed. Tel: +81-11-706-9043. Fax: +81-11-706-9042. E-mail: [email protected]. (1) (a) Helenius, A.; Abei, M. Science 2001, 291, 2364-2369. (b) Roth, J.; Chem. Rev. 2002, 102, 285-303. (2) (a) Fenn, J. B.; Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse, C. M. Science 1989, 246, 64-71. (b) Tanaka, K.; Waki, H.; Ido, Y.; Akita, S.; Yoshida, Y.; Yoshida, T. Rapid Commun. Mass. Spectrom. 1988, 2, 151153. (c) Karas, M.; Hillenkanp, F. Anal. Chem. 1988, 60, 2299-2301. (d) Mann, M.; Hendrickson, R. C.; Pandey, A. Annu. Rev. Biochem. 2001, 70, 437-473. (e) Keough, Youngquist, R. S.; Lacey, M. P. Anal. Chem. 2003, 75, 156-165. 10.1021/ac049294n CCC: $27.50 Published on Web 09/15/2004

© 2004 American Chemical Society

or LC/ESI-ion trap/TOF MS with enzymatic degradations or fragmentations by collision-induced dissociation or postsource decay (PSD) have been employed for structural analysis of carbohydrates.3 It was also suggested that information of MSn fragmentation patterns of known oligosaccharides might facilitate establishment of some rules for the structural assignment of unknown oligosaccharide sequences.4 However, mass spectrometric analysis of more complex glycoconjugates such as glycopeptides, peptides having carbohydrate side chains, still entails careful consideration of the ionization and the fragmentation characteristics of the carbohydrate moieties of those molecules.3d Håkansson et al.5 reported that combined use of electron capture dissociation and infrared multiphoton dissociation allowed achieving selective fragmentation of peptide backbone and carbohydrate moieties. The advent of a much more efficient and general strategy for the structural analysis of glycopeptides is expected in proteomics research in terms of high-throughput analysis of the posttranslational modifications. In our previous work,6 it was demonstrated that MALDI-TOF/TOF spectrometry7 provided us with a nice fragment ion peaks generated from modified peptides precursor ions. They remarkably facilitated a procedure for the identification of the modification sites on those peptides bearing sugars (O-glycopeptides) or unstable fluorescent probes. In the present report, we describe an efficient and practical method for structural analysis of oligosaccharides and N-linked type glycopeptides on the basis of matrix-dependent selective fragmentation (MDSF) technique of MALDI-TOF mass spectrometry. (3) (a) Dwek, R. A.; Edge, C. J.; Harvey, D. J.; Wormald, M. R. Annu. Rev. Biochem. 1993, 62, 65-100. (b) Harvey, D. J. Mass Spectrom. Rev. 1999, 18, 349-450. (c) Dell, A.; Morris, H. R. Science 2001, 291, 2351-2356. (d) Zaia, J. Mass Spectrom. Rev. 2004, 23, 161-227. (4) (a) Mechref, Y.; Novotny, M. V. Chem. Rev. 2002, 102, 321-370. (b) Royle, L.; Mattu, T. S.; Hart, E.; Langridge, J. I.; Merry, A. H.; Murphy, N.; Harvey, D. J.; Dwek, R. A.; Rudd, P. M. Anal. Biochem. 2002, 304, 70-90. (c) Takegawa, Y.; Ito, S.; Yoshioka, S.; Deguchi, K.; Nakagawa, H.; Monde, K.; Nishimura, S.-I. Rapid Commun. Mass Spectrom. 2004, 18, 385-391. (5) (a) Håkansson, K.; Cooper, H. J.; Emmett, M. R.; Costello, C. E.; Marshall, A. G.; Nilsson, C. L. Anal. Chem. 2001, 73, 4530-4536. (b) Håkansson, K.; Chalmers, M. J.; Quinn, J. P.; McFarland, M. A.; Hendrickson, C. L.; Marshall, A. G. Anal. Chem. 2003, 75, 3256-3262. (6) (a) Kurogochi, M.; Nishimura, S.-I.; Lee, Y. C. J. Biol. Chem., in press. (b) Kurogochi, M.; Matsushita, T.; Nishimura, S.-I. Angew. Chem., Int. Ed. 2004, 43, 4071-4075. (7) Schnaible, V.; Wefing, S.; Resemann, A.; Suckau, D.; Bucker, A.; WolfKummeth, S.; Hoffmann, D. Anal. Chem. 2002, 74, 4980-4988. (b) Suckau, D.; Resemann, A.; Schuerenberg, M.; Hufnagel, P.; Franzen, J.; Holle, A. Anal. Bioanal. Chem. 2003, 376, 952-965.

Analytical Chemistry, Vol. 76, No. 20, October 15, 2004 6097

Figure 1. MALDI-TOF mass spectra of 1. Precursor ions observed in the presence of DHB (a) and CHCA (b).

EXPERIMENTAL SECTION Reagents and General Methods. Glycopeptide 3 (N-glycosylated hexapeptide) and asparaginyl oligosaccharide 4 were prepared from chicken egg white albumin according to the methods reported previously,8 and their detailed procedures are described in Supporting Information. Other pyridylaminated oligosaccharides used in this study were a gift from Dr. H. Nakagawa and were prepared from human IgG or chicken ovalbumin. Chicken egg white albumin and trypsin and R-chymotrypsin were purchased from Wako Pure Chemical Co. Ltd. All other chemical reagents were purchased from Wako Chemical Co. Ltd. Pronase was purchased from Calbiochem. Co. Ltd. All chromatographic analyses and separations were performed on a Hitachi L-6200 HPLC system equipped with a Superdex peptide 10/30 HR column (Amersham Pharmacia Biotech Inc.) or InertsilODS column (4.6 mm × 250 mm, GL Sciences Inc.) and Hitachi L-7405 UV detector. MALDI matrixes [2,5-dehydroxybenzoic acid (DHB) and R-cyano-4-hydroxycinnaminic acid (CHCA), sinapinic acid] and MALDI peptide calibration standard mixture containing angiotensin II, bombesin, ACTH (18-39), and somatostatin were purchased from Bruker Daltonics GmbsH.

Preparation of Matrix-Sample Crystals for MALDI. The matrix solutions were prepared as follows: DHB (10 mg) was dissolved in water (1 mL), and CHCA was prepared as a saturated solution in 3:1 (v/v) of acetonitrile/water. Stock solutions of oligosaccharides and glycopeptides were prepared by dissolving them in pure water. A 0.5 µL of matrix solution was applied on the target spot of an Anchorchip plate (Bruker Daltonics), which was equipped with 384 hydrophilic anchors on an otherwise hydrophobic surface, and 1 µL of the sample solution was added and then dried at room temperature. We employed these samples ( ∼1-10 pmol) both with MALDI-PSD and TOF/TOF modes using the above preparation procedure. MALDI-TOF Mass Spectrometry. All measurements were performed using an Ultraflex TOF/TOF mass spectrometer equipped with a reflector and controlled by the Flexcontrol 1.2 software package (Bruker Daltonics GmbsH, Bremen, Germany). In MALDI TOF-MS reflector mode, ions generated by a pulsed UV laser beam (nitrogen laser, λ )337 nm, 5 Hz) were accelerated to a kinetic energy of 23.5 kV. Metastable ions generated by laserinduced decomposition of the selected precursor ions were analyzed without any additional collision gas. In MALDI-TOF/TOF mode, precursor ions were accelerated to 8 kV and selected in a timed ion gate. The fragments were further accelerated by 19 kV in the LIFT cell (LIFT means “lifting” the potential energy for the second acceleration of ion source), and their masses were analyzed after the ion reflector passage. Masses were automatically annotated by using FlexAnalysis 2.0 software package. MALDI-PSD measurement was also performed using the Ultraflex instrument. In MALDI-PSD mode, precursor ions were accelerated to 25 kV, and metastable ions were focused on the reflectron detector by means of variable reflector voltages, which were stepped from 26.3 kV down to 1.65 kV in 17 segments. The spectra obtained at each reflector voltage setting were pasted together in FlexAnalysis 2.0 software package. External calibration of MALDI mass spectra was carried out using singly charged monoisotopic peaks of a mixture of human

Figure 2. Fragmentation of 1 by MALDI-PSD mode or MALDI-TOF/TOF mode in the presence of DHB (a) and (c), and in the presence of CHCA (b) and (d). * indicates protonated ion peaks. 6098

Analytical Chemistry, Vol. 76, No. 20, October 15, 2004

Figure 3. Structural analysis of sialooligosaccharide 2 by MALDI-TOF/TOF method by means of CHCA as a matrix. * indicates product ion peaks carrying labile sialic acid residue.

angiotensin II (m/z 1046.542), bombesin (m/z 1619.823), ACTH (18-39) (m/z 2465.199), and somatostatin 28 (m/z 3147.472). The mixture of these peptides was measured on the central spot of a 3 × 3 square by using external calibration. To achieve mass accuracy better than 60 ppm, internal calibration was carried out by doping the matrix solution with a mixture of the calibration peptides. Calibration of these mass spectra was performed (8) Ishihara, H.; Takahashi, N.; Ito, J.; Takeguchi, E.; Tejima, S. Biochim. Biophys. Acta 1981, 669, 221-221.

automatically by utilizing a customized macro command of the FlexControl 2.1 software package. The macro command was used for the calibration of the monoisotopic singly charged peaks of the above-mentioned peptides. PSD and TOF/TOF spectra were annotated with the BioTools 2.1 software package. RESULTS AND DISCUSSION To investigate the effect of matrix on the generation of the precursor ion peak, we first measured MALDI-TOF spectra of a pyridylaminated model oligosaccharide 1 in the presence of DHB Analytical Chemistry, Vol. 76, No. 20, October 15, 2004

6099

Figure 4. Fragmentations in the presence of CHCA of 3 (a), 4 (b), and 5 (c). Relative intensities of major product ion peaks of these three MS/MS spectra are represented in (d).

or CHCA, respectively. As shown in Figure 1, it was found that DHB matrix gave a mixture of the protonated precursor ion peak at m/z 1557.71 and the sodiated ion peak at m/z 1579.78, while CHCA matrix produced only simple precursor ion peak as a sodium adduct at m/z 1579.58, suggesting that oligosaccharides have much higher affinity toward sodium ion than proton in the presence of CHCA.9 Our attention was next focused on the feasibility of the stable sodiated precursor ion peak for the subsequent fragmentation experiments using the conventional PSD mode and TOF/TOF mode on the MALDI instrument. As indicated in Figure 2c and d, MALDI-TOF/TOF spectra of 1 gave much more sensitive and reliable fragment ion peaks (B and Y fragment ion series) generated from the precursor ion ([M + Na]+, m/z 1579) than those obtained by PSD mode (Figure 2a and b). It was clearly suggested that TOF/TOF spectrum measured in the presence of CHCA as a matrix afforded more reasonable distribution of highly sensitive and abundant fragment ion peaks than that obtained with DHB. It should also be noted that TOF/TOF of the CHCA-derived precursor ion at m/z 1579.54 yielded simple fragment ion peaks as sodium adduct without detectable desodiation and protonation (Figure 2d). The merit of the sodiated fragment ion signals in a relevant mass range produced by TOF/TOF spectra with CHCA is evident because these singly charged and sodiated fragment ion peaks enabled the rapid and accurate sequencing of the oligosaccharide derivatives. On the other hand, TOF/TOF spectrum of the same mass precursor ion ([M + Na]+, m/z 1579.78) derived from cocrystals with DHB matrix caused significant cation exchanging during the fragmentation process to afford a mixture of sodiated and protonated ions. Newly formed protonated ion peaks, [M + H]+ at m/z 1354, 1192, 528, 503, 366, and 204, and other remaining sodiated ions made subsequent structural analysis of this octasaccharide derivative difficult (Figure 2c). The use of CHCA as a matrix seemed to provide a completely different electronic excited state of this precursor ion from that made in the presence of DHB, suggesting that matrix absorbance energy makes specific distribution (localization) or delocalization of the excited electrons in the target molecules during desorption and ionization processes.9 As (9) Dreisewerd, K. Chem. Rev. 2003, 103, 395-425.

6100

Analytical Chemistry, Vol. 76, No. 20, October 15, 2004

a result, structures of the sodiated precursor ions observed as the same mass signal at m/z 1579 independently derived from CHCA or DHB might differ in terms of the electronic excited state that greatly influences subsequent fragmentation profiles of the TOF/TOF spectra. Versatility of the above CHCA-mediated TOF/ TOF method was demonstrated by characterizing a sialooligosaccharide derivative 2 as shown in Figure 3. MALDI-TOF mass spectrum showed a major precursor ion peak of 2 as a singly charged sodiated form (m/z 2038.35) together with an ion peak at m/z 1725.38 due to the significant degree of the desialylation under the ionization step (Figure 3a). However, highly sensitive and singly charged B/Y series of sodiated fragment ions produced by TOF/TOF of the precursor ion peak at m/z 2038.35 readily materialized a facile sequencing of the structurally complex decasaccharide 2 having both sialic acid (Neu5Ac) and fucose (Fuc) residues known as extremely labile species (Figure 3b). When DHB was employed for producing a precursor ion of N-glycosylated hexapeptide 3, it was suggested that TOF/TOF fragmentation of the protonated precursor ion, [M + H]+, gave favorable product ion peaks suited for systematic structural analysis of glycopeptides. It is of interest that fragmentation occurred predominantly at the glycoside bond between the two N-acetyl-D-glucosamine (GlcNAc) residues to the cleavage of the peptide backbone (Figure 4). As a result, two major product ions due to the monoglycosylated peptide ion (b*6) at m/z 849.45 and the hexasaccharide (Man5GlcNAc) ion at m/z 1014.35 facilitated analytical process. A series of b*- and y-ions were useful for identifying glycosylation site as well as peptide sequencing (Figure 4b). In addition, it is likely that fragmentation of sugars seemed to occur from the nonreducing end of the oligosaccharide (Figure 4a) and was sufficient to determine the sequence of the carbohydrate moiety. The finding that sodiated precursor ion of 3 derived in the presence of CHCA produced glycoside bond fragmentation with minimal cleavage of the peptide backbone is particularly significant (10) For example, see; (a) Hilaire, P. M. S.; Meldal, M. Angew. Chem., Int. Ed. 2000, 39, 1162-1179. (b) Nishimura, S.-I. Curr. Opin. Chem. Biol. 2001, 5, 325-335. (c) Coltart, D. M.; Royyuru, A. K.; Williams, L. J.; Glunz, P. W.; Sames, D.; Kuduk, S. D.; Schwarz, J. B.; Chen, X.-T.; Danishefsky, S. J.; Live, D. H. J. Am. Chem. Soc. 2002, 124, 9833-9844.

Figure 5. Structural analysis of glycopeptide 3 by MALDI-TOF/TOF method in the presence of DHB. * indicates monoglycosylated (GlcNAc) fragment ion peaks and ** means heptasaccharide (Man5GlcNAc2) carrying fragment ion peaks, respectively.

because it allows practical and general strategy for the determination of sugar chains (Figure 5a). Specific and exclusive cleavage at the core (GlcNAc)2 regions and quite similar fragmentation behavior (d) were observed in the TOF/TOF profiles (b, c) of asparaginyl- and pyridylaminated derivatives (4 and 5), providing data that would become useful for N-glycan database searching. The versatility of MDSF-based structural characterization was demonstrated by using two typical N-glycopeptides isolated from bovine pancreas ribonuclease B as described in Supporting Information. In conclusion, we found that the MDSF method in MALDITOF/TOF spectrometry gives highly sensitive and reliable product ion peaks for the efficient sequencing of both peptide backbones and oligosaccharides of N-glycopeptides at the same time. The results reported herein demonstrate the ability of MDSF to generate desired structural information based on the controlled cleavage of the singly charged precursor ion with different electronic excited states made by this method, although the mechanism of the formation of the different electronic excited states has not been clarified yet. The present method can be readily used for precise and reliable structure analyses such as identification of N-glycosylation sites of glycopeptides and sequencing of complex oligosaccharide derivatives including unstable Neu5Ac and Fuc residues. Ideal product ion peaks observed

on the basis of MDSF in TOF/TOF would be extremely useful for the posttranslational modifications study in terms of characterization of protein glycosylations and construction of compound libraries of glycopeptides.10 ACKNOWLEDGMENT This work was supported partly by a grant for “Research Project of the Functional Glycomaterials” from the Ministry of Education, Culture, Science, and Technology, Japan and also by a grant for “Research and Development on Glycocluster Controlling Biomolecules” from the New Energy and Industrial Technology Development Organization (NEDO). We thank Mr. T. Suenaga of Bruker Daltonics Japan for helpful assistance in MALDI-TOF/TOF mass spectrometry. We also appreciate to Dr. H. Nakagawa for his kind gift of PA sugar materials. SUPPORTING INFORMATION AVAILABLE Experimental procedures and spectral data (PDF). This material is available free of charge via the Internet at htpp:// pubs.acs.org. Received for review May 14, 2004. Accepted August 6, 2004. AC049294N

Analytical Chemistry, Vol. 76, No. 20, October 15, 2004

6101