ARTICLE pubs.acs.org/ac
Highly Sensitive MALDI Analyses of Glycans by a New Aminoquinoline-Labeling Method Using 3-Aminoquinoline/ r-Cyano-4-hydroxycinnamic Acid Liquid Matrix Kaoru Kaneshiro,* Yuko Fukuyama, Shinichi Iwamoto, Sadanori Sekiya, and Koichi Tanaka Koichi Tanaka Mass Spectrometry Research Laboratory and Koichi Tanaka Laboratory of Advanced Science and Technology, Shimadzu Corporation, 1, Nishinokyo-Kuwabaracho Nakagyo-ku, Kyoto 604-8511, Japan
bS Supporting Information ABSTRACT: In glycomics, mass spectrometry is an indispensable tool for high throughput analyses. Generally speaking, glycans contain many hydroxyl groups and are more difficult to ionize than peptides. Derivatization of glycans has been useful for increasing sensitivity. However, it takes time to purify and causes loss of sample. Here, we show a highly sensitive aminoquinoline (AQ)-labeling method of glycans on a matrix-assisted laser desorption/ionization (MALDI) target using a liquid matrix 3-aminoquinoline (3-AQ)/R-cyano-4-hydroxycinnamic acid (CHCA). It is a rapid procedure and reduces loss of sample material during the reaction process, especially in negative ion mode where 10 amol of monosialylated N-glycan were detected as AQ-labeled molecular ions. In addition, MS/MS of 10 amol of monosialylated N-glycan was achieved.
A
fter completing the euchromatic sequence of the human genome in 2003,1 postgenome analyses have become a key focus. With the advent of gene function analysis, proteome analysis has occupied an important position in bioscience research2,3 encompassing both post-translational investigation and sequence elucidation.47 Glycosylation is a common post-translational modification of proteins.8 Glycosylation pattern and amount of glycan relate to a variety of physiological processes such as intercellular signaling, immunity, and so on.911 Thus, glycan analysis is an indispensable tool to clarify vital activity at a molecular level. So far, many analytical methods of glycans have been proposed. Among these, mass spectrometry is an established tool for the structural analysis of glycans, since the development of matrix-assisted laser desorption/ionization (MALDI) and electrospray ionization (ESI) has enabled ionization of biological macromolecules.1214 As a result of its ability to achieve high sensitivity and rapid analysis, mass spectrometry is an indispensable analytical technique for glycoproteins.15,16 Additionally, glycans containing many hydroxyl groups are more difficult to ionize than proteins and peptides. In mass spectrometric analyses of glycoproteins, glycans are often enzymatically or chemically released from glycoproteins.16,17 In that case, derivatization of the glycan is carried out for higher sensitivity,16,17 as it increases the hydrophobicity/surface activity/charge to improve MS ionization responses.18 So far, the derivatization has been done by labeling the reducing ends of glycans with amine such as 2-aminopyridine (PA) or some other reagents.19,20 However, these labeling methods need purification steps which are complex, take time, and cause sample loss.19 To prevent this problem, Rohmer et al. reported a derivatization using a solid matrix 3-aminoquinoline (3-AQ) as a labeling reagent.21 This reaction takes place on a MALDI target, and 3-AQ plays the r 2011 American Chemical Society
Table 1. N-linked Glycan Standards
role not only of labeling reagent22 but also of MALDI matrix; hence, a purification step is no longer required. This derivatization is a dehydration reaction; thus, it is better to have lower water content. However, increasing the amount of organic solvent resulted Received: December 6, 2010 Accepted: April 20, 2011 Published: April 20, 2011 3663
dx.doi.org/10.1021/ac103203v | Anal. Chem. 2011, 83, 3663–3667
Analytical Chemistry
ARTICLE
Figure 1. Positive ion mass spectra of 100 fmol of NA2 glycan using DHB (A), 3-AQ/CHCA (B), and 3-AQ/CHCA (C) with heating at 60 °C for 60 min.
Figure 2. Negative ion mass spectra of 1 pmol of A1 glycan using DHB (A) and using 3-AQ/CHCA (B) and NA2 glycan using 3-AQ/CHCA (C). (B) and (C) were obtained after heating at 60 °C for 60 min.
in decreasing derivatization rates, probably because accelerated drying of the spots reduced the reaction time.21 Therefore, conditions such as solvent need to be optimized.21 On the other hand, liquid matrixes remain liquid at room temperature under atmospheric pressure and even under vacuum.23 Therefore, a liquid matrix 3-AQ/R-cyano-4-hydroxycinnamic acid (CHCA) reported by Kumar et al.24 can continue to provide a derivatization environment. In addition, using liquid matrixes significantly enhances the sample homogeneity, reducing the need for time-consuming searches for hot or sweet spots when solid matrixes are used.23 Here, we show a direct labeling method of glycans on a MALDI target using the liquid matrix 3-AQ/CHCA for high-sensitivity detection.
’ EXPERIMENTAL SECTION Materials and Reagents. The commercially available N-linked glycan standards NA2 glycan [(Gal-GlcNAc)2Man3(GlcNAc)2] and NA4 glycan [(Gal-GlcNAc)4Man3(GlcNAc)2] were purchased
from Sigma-Aldrich Japan K.K. (Tokyo, Japan). A1 glycan [Sia(GalGlcNAc)2Man3(GlcNAc)2] and A2 glycan [Sia2(Gal-GlcNAc)2Man3(GlcNAc)2] were purchased from Ludger Ltd (Table 1). MALDI-MS-grade 3-aminoquinoline (3-AQ) was purchased from Fluka Analytical. Ammonium dihydrogen phosphate, 99.99þ% (NH4H2PO4), was purchased from Sigma-Aldrich Japan K.K. (Tokyo, Japan). 2,5-Dihydroxybenzoic acid (DHB) and R-cyano-4hydroxycinnamic acid (CHCA) were purchased from LaserBio laboratories. Sodium chloride (NaCl) was purchased from Nacalai tesque, Inc. Milli-Q water was used in all preparations. HPLC-grade acetonitrile (ACN) was purchased from Wako Pure Chemical Industries, Ltd. A 1000 μfocus MALDI plate purchased from Hudson Surface Technology, Inc. was used as a MALDI target. Sample and Matrix Preparation. All standard glycans were dissolved in water at 1 amol to 1 pmol/0.5 μL. DHB matrix solution was prepared by dissolving 3 mg of DHB in 1 mL of ACN/water (1:1, v/v). Ten mM NaCl was prepared using water. A 0.5 μL aliquot of sample, 0.5 μL aliquot of DHB, and 0.5 μL 3664
dx.doi.org/10.1021/ac103203v |Anal. Chem. 2011, 83, 3663–3667
Analytical Chemistry
ARTICLE
aliquot of 10 mM NaCl were deposited sequentially onto a MALDI target (on-target mix preparation method). 3-AQ/CHCA Matrix Solution. A saturated CHCA solution was prepared by dissolving 10 mg of CHCA in 540 μL of ACN/water (1:1, v/v) and 60 μL of 100 mM NH4H2PO4 aqueous solution. 3-AQ/CHCA was prepared by dissolving 20 mg of 3-aminoquinoline in 150 μL of saturated CHCA solution. This is a bright yellow liquid matrix. Optimized 3-AQ/CHCA was diluted 10fold using ACN/water (1:1, v/v). AQ-Labeling Glycans Reaction. Glycans (0.5 μL) and 0.5 μL of 3-AQ/CHCA were deposited on MALDI target prepared by setting at 60 °C with a heating block (SCINICS CORPORATION; on-target mix preparation method). The MALDI target was heated for 60 min at 60 °C. After cooling to room temperature, the target was inserted in MALDI-quadrupole ion trap (QIT)-time of flight (TOF) MS to be measured. Mass Spectrometric Measurements. The mass spectra were acquired using a AXIMA-Resonance UV-MALDI mass spectrometer (Shimadzu/Kratos,UK), equipped with a nitrogen laser (337 nm wavelength) with 3 ns pulse width and a monochrome charge coupled device (CCD) camera system allowing sample monitoring. The maximum laser pulse rate is 10 Hz for the mass spectrometer. Collision-induced dissociation (CID) is executed within an ion trap, allowing multiple rounds of CID (MS/MS). The laser power was kept at threshold to obtain good intensity signal without generation of extensive noise in the mass spectra. Table 2. Detection Limits of N-Linked Glycan with DHB and 3-AQ/CHCAa positive ion mode
a
negative ion mode
DHB
3-AQ/CHCA
DHB
3-AQ/CHCA
NA2
10 fmol
100 amol
ND
50 amol
NA4 Al
5 fmol 100 fmol
50 amol 500 amol
ND 100 fmol
50 amol 10 amol
A2
100 fmol
500 amol
100 fmol
50 amol
1 amol1 pmol/well of analytes were analyzed. ND denotes that analyte molecular ions were not detected.
Using 3-AQ/CHCA or DHB matrixes, 100500 laser shots were accumulated at sweet spots. All these detection limits were calculated for molecular ions of signal-to-noise (S/N) ratio above 2.
’ RESULTS AND DISCUSSION On-Target AQ Derivatization of Glycans. 2,5-Dihydroxybenzoic acid (DHB) is a most widely used solid matrix for MALDIMS analyses of glycans. Using this matrix, ions are detected as cationized (Naþ or Kþ adducted) molecules. However, using a liquid matrix 3-AQ/CHCA, ions were detected not only as protonated and/or cationized molecules but also as 126 Da adducted molecules in positive ion mode (see Figure 1A,B). The 126 Da adducted molecules are reported as forming by nonreductive amination of 3-AQ at reducing end of glycans.17 The 126 Da adducted molecules are AQ-labeled glycan ions in 3-AQ/CHCA on MALDI target. Complete AQ-labeled glycans were achieved by heating at 60 °C for 60 min or more after depositing on a MALDI target (see Figure 1C). In negative ion mode, AQ-labeled sialoglycans were detected as a deprotonated ion [M þ 126 H] and AQ-labeled nonsialoglycans were detected as H2PO4 and the dehydration H2PO4H2O (PO3) attachment ion [M þ 126 þ H2PO4] and [M þ 126 þ PO3] (see Figure 2AC). Those H2PO4/PO3 are derived from ammonium dihydrogen phosphate. Negative ion mass spectra of disialylated glycan using 3-AQ/CHCA with heating at 60 °C for 60 min (AQ-labeled glycan) showed lower sialic acid loss than DHB (nonlabeled glycan; see Supporting Information Figure S-1A,C). In the detection of nonlabeled disialylated glycan, the use of 3-AQ/CHCA as a matrix obviously stabilized the possible loss of sialic acid from DHB (see Supporting Information Figure S-1A, B). This result indicates that AQ-labeled disialylated glycan using 3-AQ/CHCA stabilizes sialyl linkages in the effect of the matrix. Improvement of Detection Sensitivity of Glycans by AQ-Labeling with 3-AQ/CHCA. We compared detection limit of glycans using 3-AQ/CHCA (AQ-labeled glycan ions) with that using DHB (nonlabeled glycan ions). After AQ-labeling of glycans at 60 °C for 60 min, detection sensitivities of AQ-labeled glycans were dramatically improved (see Table 2). In positive ion mode,
Figure 3. Positive ion mass spectra of 50 amol of NA4 glycan using 3-AQ/CHCA.
Figure 4. Negative ion mass spectra of A1 glycan using 3-AQ/CHCA at 500 amol (A), at 100 amol (B), and at 10 amol (C). 3665
dx.doi.org/10.1021/ac103203v |Anal. Chem. 2011, 83, 3663–3667
Analytical Chemistry
ARTICLE
Figure 5. Positive MS/MS spectra of [AQ M þ Na]þ NA4 glycan at 1 pmol using DHB (A), at 1 pmol using 3-AQ/CHCA (B), and at 100 amol using 3-AQ/CHCA (C).
sensitivity of glycans using 3-AQ/CHCA were 100200 times those using DHB (see Table 2). For example, the detection limit of NA2 glycan with DHB was 10 fmol in positive ion mode while that with 3-AQ/CHCA was 100 amol. In negative ion mode, 1 pmol of neutral glycans (NA2, NA4 glycan) were not detected using DHB while 50 amol of neutral glycans could be detected as AQ-labeled molecular ions using 3-AQ/CHCA. Sensitivity of sialoglycans (A1 and A2 glycan) using 3-AQ/CHCA was improved by 200010 000 times compared to those using DHB (see Table 2). The highest sensitivity using 3-AQ/CHCA was confirmed for NA4 glycan in positive ion mode (50 amol; see Table 2 and Figure 3) and A1 glycan in negative ion mode (10 amol; see Table 2 and Figure 4). In order to determine the detection limit, we confirmed the rough linearity between average peak areas of [M þ 126 H] and concentrations in the spectra of Figure 4 (see Supporting Information Table S-1 and Figure S-2). Structural Analyses of Glycans by MS/MS. We compared MS/MS fragments of AQ-labeled glycans using 3-AQ/CHCA with that of nonlabeled glycans using DHB. As a result, almost all MS/MS fragment ions of AQ-labeled glycans were the same as that of nonlabeled glycans in both positive and negative ion mode (see Figure 5A,B). MS/MS of 100 amol of NA4 glycan in positive mode and 10 amol of A1 glycan in negative mode were successfully performed, respectively, and they showed a part of the sequence of AQ-labeled glycans (see Figure 5C and Supporting Information Figure S-3; fragmentation nomenclature by Domon and Costello25).
’ CONCLUSION In this study, we reported a new method for on-target labeling and high sensitivity detection of glycans using a liquid matrix 3-AQ/ CHCA. The liquid matrix 3-AQ/CHCA remains liquid at room temperature and continues to provide a derivatization environment. Therefore, it is able to act as a buffer for the chemical reaction.26 Our future aim is on-target enzymatic reactions for glycoproteins in 3-AQ/CHCA followed by a AQ-labeling of released glycans. This is expected to provide higher sensitivity and throughput for
the analysis of glycoproteins. It is likely to be suitable for applications such as determining changes in glycan structures or in levels that increase/decrease in association with certain human diseases or in cases where only a small amount of sample is available.
’ ASSOCIATED CONTENT
bS
Supporting Information. Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION Corresponding Author
*Tel:þ81 (0) 75 823-2897.
’ ACKNOWLEDGMENT This research is supported by the Japan Society for the Promotion of Science (JSPS) through its “Funding Program for WorldLeading Innovative R&D on Science and Technology (FIRST Program).” ’ REFERENCES (1) International Human Genome Sequencing Consortium Nature 2004, 431, 931945 (2) Madi, A; Pusztahelyi, T; Punyiczki, M; Fes€us, L. Acta Biol. Hung. 2003, 54, 1–14. (3) Yu, U; Lee, S. H.; Kim, Y. J.; Kim, S. J. Biochem. Mol. Biol. 2004, 37, 75–82. (4) Choudhary, C; Mann, M. Nat. Rev. Mol. Cell. Biol. 2010, 11, 427–439. (5) Salzano, A. M.; Crescenzi, M. Ann. Ist. Super. Sanita 2005, 41, 443–450. (6) Reinders, J; Lewandrowski, U; Moebius, J; Wagner, Y; Sickmann, A. Proteomics 2004, 4, 3686–3703. (7) Larsen, M. R.; Roepstorff, P. Fresenius Anal. Chem. 2000, 366, 677–690. 3666
dx.doi.org/10.1021/ac103203v |Anal. Chem. 2011, 83, 3663–3667
Analytical Chemistry
ARTICLE
(8) Ozcan, S; Andrali, S. S.; Cantrell, J. E. Biochim. Biophys. Acta 2010, 1799, 353–364. (9) Janik, M. E.; Litynska, A; Vereecken, P. Biochim. Biophys. Acta 2010, 1800, 545–555. (10) Gu, J; Taniguchi, N. Glycoconjugate J. 2004, 21, 9–15. (11) Schroeder, H. W., Jr.; Cavacini, L. J. Allergy Clin. Immunol. 2010, 125, S41–S52. (12) Tanaka, K.; Waki, H.; Ido, Y.; Akita, S.; Yoshida, Y.; Yoshida, T.. Rapid Commun. Mass Spectrom. 1988, 2, 151–153. (13) Fenn, J. B.; Mann, M.; Meng, C. K.; Won, S. F.; Whitehouse, C. M.. Science 1989, 246, 64–71. (14) Zaia, J. Chem Biol. 2008, 15, 881–892. (15) North, S. J.; Hitchen, P. G.; Haslam, S. M.; Dell, A. Curr. Opin. Struct. Biol. 2009, 19, 498–506. (16) Karas, M.; Hillenkamp, F. Anal. Chem. 1988, 60, 2299–2301. (17) Snovida, S. I; Chen, V. C; Perreault, H. Anal. Chem. 2006, 78, 8561–8568. (18) Anumula., K. R. Anal. Biochem. 2006, 350, 1–23. (19) Lamari, F. N.; Kuhn, R.; Karamanos, N. K. J. Chromatogr., B: Anal. Technol. Biomed. Life Sci. 2003, 793 (1), 15–36. (20) Harvey, D. J. J. Am. Soc. Mass Spectrom. 2000, 11, 900–915. (21) Rohmer, M.; Meyer, B.; Mank, M.; Stahl, B.; Bahr, U.; Karas, M. Anal. Chem. 2010, 82, 3719–3726. (22) Metzger, J. O; Woisch, R; Tuszynski, W; Angermann, R; Fresenius, J. Anal. Chem. 1994, 349, 473–474. (23) Tholey, A.; Heinzle, E. Anal. Bioanal. Chem. 2006, 386, 24–37. (24) Kumar Kolli, V. S.; Orlando, R Rapid Commun. Mass Spectrom. 1996, 10, 923–926. (25) Domon, B.; Costello, C. E. Biochemistry 1988, 27 (5), 1534–1543. (26) Mank, M.; Stahl, B.; Boehm, G. Anal. Chem. 2004, 76, 2938–2950.
3667
dx.doi.org/10.1021/ac103203v |Anal. Chem. 2011, 83, 3663–3667
