Sensitive Analyses of Neutral N-Glycans using Anion-Doped Liquid

Jun 24, 2012 - Air Pollution and Industrial Hygiene · Apparatus and Plant Equipment .... Negative-ion MS2 spectra of 1 fmol N-glycans were successfull...
0 downloads 0 Views 954KB Size
Download PDF
Article pubs.acs.org/ac

Sensitive Analyses of Neutral N-Glycans using Anion-Doped Liquid Matrix G3CA by Negative-Ion Matrix-Assisted Laser Desorption/ Ionization Mass Spectrometry Takashi Nishikaze,* Yuko Fukuyama, Shin-ichirou Kawabata, and Koichi Tanaka Koichi Tanaka Laboratory of Advanced Science and Technology, Shimadzu Corporation, 1, Nishinokyo-Kuwabaracho, Nakagyo-ku, Kyoto 604-8511, Japan S Supporting Information *

ABSTRACT: Negative-ion fragmentation of N-glycans has been proven to be more informative than that of positive-ion. In particular, it defines structural features such as the specific composition of the two antennae and the location of fucose. However, negative-ion formation of neutral N-glycans by matrixassisted laser desorption/ionization mass spectrometry (MALDIMS) remains a challenging task, and the detection limit of Nglycans in negative-ion mode is merely at the subpicomole level. Thus, practical applications are limited. In this study, combinations of five liquid matrices and nine anions were used to ionize N-glycans as anionic adducts, and their performances for sensitive analyses were evaluated. The best results were obtained with anion-doped liquid matrix G3CA, which consists of p-coumaric acid and 1,1,3,3-tetramethylguanidine; the detection limits of anion adducted N-glycans were 1 fmol/well for NO3−, and 100 amol/ well for BF4−. Negative-ion MS2 spectra of 1 fmol N-glycans were successfully acquired with a sufficient signal-to-noise ratio and were quite useful for MS-based structural determination. The anion-doped G3CA matrix opens the way for sensitive and rapid analysis of neutral N-glycans in negative-ion MALDI at a low femtomole level.

P

Recently, negative-ion fragmentation of glycans has proven to be more informative than that of positive-ion fragmentation.6−15 In particular, negative-ion fragmentation defines structural features such as the specific composition of the antennae and the location of fucose residues.6−12 However, negative ionization of neutral glycans is much more difficult than positive ionization. The deprotonated form of a neutral glycan, especially a reducing neutral glycan, is relatively unstable and therefore undergoes vigorous fragmentation during or after ionization events. To form stable negative-ion species of the glycan, many researchers have attempted to ionize a neutral glycan as an anionic adduct.6,16−24 Under ESI conditions, glycans (e.g., small neutral oligosaccharides, human milk oligosaccharides, and N-glycans) have been successfully ionized as anionic adducts or deprotonated form.6,16,17 However, the negative-ion formation of glycans in MALDI is difficult, probably because of a lack of appropriate matrices. At present, a few matrices are available for negativeionization of neutral glycans. For example, β-carboline compounds have been used to ionize small neutral oligosaccharides as Cl− adducts in negative-ion MALDI.18−22 Wong et al. reported HSO4− and alkylsulphonates as dopants for matrices.23,24 They observed intense signals of relatively large neutral glycans containing N-glycans as anionic adducts,

rotein glycosylation is one of the most ubiquitous posttranslational modifications of proteins and plays an important role in biological functions such as cellular localization, turnover, and protein quality control.1−4 Typically, glycans are linked to serine, threonine, and asparagine residues. To date, various methodologies have been developed to characterize glycoproteins.5 In general, glycoproteins have microheterogeneity in their glycan moieties; therefore, glycans were often released from glycoproteins by enzymatic or chemical treatments for detailed glycan structural analysis. For N-glycan, which is bound to asparagine residues of glycoproteins, Peptide N-glycosidase F (PNGaseF) is most frequently used to release the glycan from glycoproteins. All Nglycans have a common pentasaccharide core structure, which has great structural diversity due to a variety of antennae structures and the presence or absence of core fucose residues. Mass spectrometry (MS) combined with electrospray ionization (ESI) and matrix-assisted laser desorption/ionization (MALDI) has become an indispensable tool for analyzing glycans due to its rapidity, high sensitivity, and usefulness in structural determination by tandem MS. An inherent feature of MALDI is that it ionizes analytes as a singly charged form; thus, it is a satisfactory technique for profiling glycan mixture. The structures of glycans based on tandem MS have generally been determined in positive-ion mode because glycans are usually ionized as [M + Na]+. Therefore, scant attention has been paid to negatively charged ions. © 2012 American Chemical Society

Received: April 12, 2012 Accepted: June 24, 2012 Published: June 24, 2012 6097

dx.doi.org/10.1021/ac3009803 | Anal. Chem. 2012, 84, 6097−6103

Analytical Chemistry

Article

while no structural information was obtained from collisioninduced dissociation (CID) of the adducts. To obtain informative MS2 spectra from N-glycans in negative-ion MALDI, Domann et al. recently optimized ionization conditions, including matrix selection and anion dopants.25 The best results were obtained with an NO3−doped 2,4,6trihydroxyacetophenone (THAP) matrix, but the detection limit was merely at the subpicomole level. In addition, they noted that the fine crystals of THAP, which yield good signals of glycans, degrade rapidly over several hours under high vacuum conditions, reducing the signal strength. The usefulness of an ionic liquid matrix, introduced by Armstrong et al.,26 has been demonstrated in the analysis of biomolecules by MALDI. Homogeneous liquid matrices are clearly advantageous over conventional solid matrices for automated and quantitative analyses. To date, various liquid matrices have been reported to be suitable for glycan analyses by MALDI-MS.27−33 Some studies have focused on analyzing acidic glycans such as sialylated and sulfated glycans in negativeion mode.28−30,33 However, none have focused on the use of liquid matrices for the negative-ion formation of neutral glycans. In this study, we demonstrate that the anion-doped liquid matrix G3CA, which consists of p-coumaric acid (CA) and 1,1,3,3-tetramethylguanidine (TMG),30 acts as an effective matrix for sensitive MALDI analysis of neutral N-glycans. An anion-adducted N-glycan was detected at the low- or subfemtomole level in negative-ion mode. Informative MS2 spectra were successfully obtained from 1 fmol N-glycan at a sufficient signal-to-noise ratio (s/n) for structural determination.

Figure 1. Structures of NA2 and A1 glycans.

0.25 mM to 250 mM were tested, and the concentration of 2.5 mM exhibited promising results for all anions. Evaluated anions and liquid matrices are listed in Table 1. Mass Spectrometry. First, 0.5 μL of sample solution was deposited on a μFocus MALDI plateTM 700 μm (Hudson Surface Technology, Inc., Fort Lee, NJ). A 0.5 μL matrix solution was then dropped onto the plate and left to dry. Mass spectra were acquired by using a MALDI-QIT-TOF-MS (AXIMA Resonance, Shimadzu/Kratos, UK). Samples were irradiated by a nitrogen UV laser (337 nm). Argon gas was used for CID fragmentation, and helium gas was used for ion cooling in the ion trap.

■

■

EXPERIMENTAL SECTION Reagents. All ammonium salts of anions were of special grade or greater purity. The MALDI matrix chemicals (2,5dihydroxybenzoic acid (DHB), THAP, α-cyano-4-hydroxycinnamic acid (CHCA), 2-(4-hydroxyphenylazo)benzoic acid (HABA), and CA) were purchased from Sigma-Aldrich (St. Louis, MO). The countercation butylamine (BA) was purchased from Sigma-Aldrich, and TMG was purchased from Tokyo Kasei (Tokyo, Japan). Methanol (MeOH) and acetonitrile (ACN) used in this study were LC-MS grade or greater purity. Butanol and ethanol were HPLC grade. Analytes. The biantennary complex glycan NA2, sialylated biantennary complex glycan A1, and Chicken ovalbumin (grade VII) were obtained from Sigma-Aldrich. The structures of NA2 and A1 are depicted in Figure 1. NA2 and A1 were reconstituted in water. N-glycans of ovalbumin were released by PNGaseF (Sigma-Aldrich). After overnight incubation at 37 °C, the released N-glycans were purified by a hydrophilic affinity method using cellulose fibrous medium (SigmaAldrich), as described previously.34,35 The purified N-glycans were further desalted using NuTip carbon (Glygen Corp., Columbia, MD) and then reconstituted in water. Matrix Preparation. Liquid matrices were prepared according to Fukuyama et al.30 with some modifications. Briefly, they were prepared by adding 1, 2, or 3 equivalents of counter cations to 1 equivalent of conventional matrix substances in MeOH; the solvent was then evaporated under vacuum overnight. The resulting solid or viscous liquid was dissolved in 50% ACN at a concentration of 100 mg/mL. The matrix solution was further diluted 5-fold with 50% ACN containing 2.5 mM anions. Different anion concentrations from

RESULTS AND DISCUSSION Evaluation of Anion-Doped Liquid Matrices for Sensitive Detection. In this study, combinations of five liquid matrices and nine anions were used to ionize N-glycan as anionic adducts. These matrices have been reported to be applicable to the ionization of glycans, including sialylated and sulfated glycans.27−33 Anions were selected based on the available literature that describes anion attachment to glycans.6,17,22,25 In addition, PF6−, BF4−, and SCN− were tested as novel anion sources for the negative ionization of glycans. Adduct species, composed of a small inorganic anion [A]− and a neutral analyte molecule [M], can be considered to exist as “proton-bound mixed dimers of anions ”, [M − H]−···H+···[A]−.36 Consequently, gas-phase basicities (GBs) of the deprotonated analyte molecule [M − H]− and the anion moiety [A]− play important roles in determining the stability of anionic adducts. In addition, GB of the deprotonated matrix molecule is an important parameter in MALDI, because if anions have higher GBs than matrix molecules, no anions and anionic adducts can exist in the matrix plume.18 Therefore, the matrix has two important roles in producing anionic adducts: (i) effective desorption of analyte molecules together with anions and (ii) preventing protonation of anions (i.e., not neutralizing anions). The results obtained here were thus interpreted based on these perspectives. All anion-doped liquid matrices used in this study formed a homogeneous spot of matrix/analyte/anion mixture on the MALDI plate, remaining stable for at least several hours under high vacuum. The increased anion concentration caused solidification of liquid matrices and degradation of mass 6098

dx.doi.org/10.1021/ac3009803 | Anal. Chem. 2012, 84, 6097−6103

Analytical Chemistry

Article

Table 1. Adduct Formation and Fragmentation of NA2 Glycan Using Anion-Doped Liquid Matrices DHBB (BA + DHB, 1:1)

anion

nominal massb

gas-phase basicity (kJ/mol)

[M + anion]−c

PF6− BF4− HSO4− I− NO3− Br− SCN− H2PO4− Cl−

145 87 97 127 62 79, 81 58 97 35, 37

1157a 1204a 1265 1294 1330 1332 1343 1351 1373

+++ ++++ ++ +++ + ++ n.d. + n.d.

DoFd − − − low high low high

GTHAP (TMG + THAP, 1:1)

G2CHCA (TMG + CHCA, 2:1)

G2HABA (TMG + HABA, 2:1)

[M + anion]−

[M + anion]−

[M + anion]−

DoF

[M + anion]−

DoF

++++ ++++ ++++ ++++ ++ ++++ + ++ +

− low − − high low high high high

++++ ++++ +++ ++++ +++ ++++ ++ +++ ++

− − − − low − middle middle middle

++ ++++ ++++ +++ n.d ++ n.d. + n.d.

DoF − low − − − −

++ +++ ++++ ++++ + ++ n.d. + n.d.

DoF − low − − high middle high

G3CA (TMG + CA, 3:1) CID spectra of [M + anion]−e √ √ √ √ √ √ √

a

Designates calculated values taken from ref 37. Residual values of gas-phase basicities were taken from ref 38. (NIST Chemisty WebBook). bMass of major isotope. c+ indicates s/n < 20, ++ indicates 20 < s/n < 100, +++ indicates 100 < s/n < 250, ++++ indicates 250 < s/n. dDoF means degree of fragmentation. The degree of fragmentation was evaluated using intensities of 2,4A6 fragment ion and [M + anion]− as follows; DoF = I[2,4A6]−/ (I[M + anion]− + I[2,4A6]−). − indicates no fragmentation, low indicates DoF < 10%, middle indicates 10% < DoF < 50%, high indicates 50% < DoF. e√ indicates an appearance of informative glycan-related fragment ions such as 2,4A and D ion in the CID spectrum of [M + anion]−.

Figure 2. Negative-ion mass spectra of NA2 glycan at 100 fmol using various liquid matrices and anions as dopants. (a) DHBB with BF4−. (b) GTHAP with BF4−. (c) G2CHCA with BF4−. (d) G2HABA with BF4−. (e) G3CA with BF4−. (f) DHBB with NO3−. (g) GTHAP with NO3−. (h) G2CHCA with NO3−. (i) G2HABA with NO3−. (i) G3CA with NO3−.

(e.g., PF6−, BF4−, I−, and HSO4−)37,38 gave strong signals of anion-adducted NA2 glycan in combinations with any liquid matrices used. However, NO3−, Br−, SCN−, H2PO4−, and Cl−, which have relatively high GB,38 had little or no signals of adducts in combinations with DHBB, GTHAP, or G2CHCA. These high-GB anions with G2HABA gave weak signals of [M + anion]− and extensive 2,4A-type fragment ions at m/z 1275.5 and 1478.5. Interestingly, the high-GB anions with G3CA produced strong signals of [M + anion]−, but their intensities were less than those of low-GB anion adducts. G3CA was found to produce all anion adducted species tested in this study with low fragmentation and superior signal intensity. The mass spectra of 100 fmol NA2 glycan using G3CA with various anions are indicated in Figure S-1 of Supporting Information.

spectral qualities. Anion concentrations of 2.5 mM exhibited promising results for all anion species. Figure 2 depicts the mass spectra of NA2 glycan at 100 fmol acquired using liquid matrices with BF4− or NO3− anions as typical examples, and Table 1 ranks the results of five tested liquid matrices in yielding various anion adducts of NA2 glycan. Other than the signals of anion-adducted NA2 glycan, [M + anion]−, some combinations exhibited unfavorable fragment ions at m/z 1275.5 and 1478.5 in the mass spectra (e.g., Figure 2h and i). The fragment ions correspond to 2,4A5 and 2,4A6 ions (fragmentation nomenclature by Domon and Costello39). The 2,4A-type cross-ring cleavage is one of the major fragmentation pathways in the negative-ion mode. The appearance and intensity of these fragment ions are indicated in the DoF columns of Table 1. Anions with relatively low GB 6099

dx.doi.org/10.1021/ac3009803 | Anal. Chem. 2012, 84, 6097−6103

Analytical Chemistry

Article

Figure 3. Negative-ion mass spectra of NA2 glycans at various concentrations. Spectra were acquired using (a) G3CA with BF4−, (b) G3CA with I−, and (c) G3CA with NO3−.

glycoprotein and has various types of neutral N-glycans.40−42 Negative-ion mode analyses of the N-glycans using NO3− or BF4−doped G3CA indicated spectral patterns similar to those obtained in positive-ion mode. As expected, BF4−doped G3CA yielded a mass spectrum with a better s/n ratio than NO3−doped G3CA. It should be noted that negative-ion mode analysis exhibited comparable or greater sensitivity than positive-ion mode analysis. Relationship between Ion Yields of [M + anion]− and GB of Anions. Following Domann et al.,25 we evaluated accurate ion yields of each anionic adduct of neutral NA2 glycans using sialylated N-glycan (A1) as an internal standard, because sialylated N-glycan ionizes as [M − H]−, not [M + anion]−. The ion yields of anion-adducted NA2 glycan estimated by relative intensities of [NA2 + anion]− and [A1 − H]− are plotted against the GBs of each anion in Figure 5. Interestingly,

The approximate detection limits of NA2 glycan were determined as the smallest amount on a single well at which the signal-to-noise ratio exceeded 3. When using G3CA, the detection limits of adducts were 1 fmol/well for NO3−, and 100 amol/well for BF4− and I− (see Figure 3). Improvement of two or more orders of magnitude over the previous best sensitivity, which is picomole or subpicomole using a combination of solid THAP matrix and NO3−,25 was observed. Figure 4 compares the mass spectra of N-glycans released from ovalbumin in both positive- and negative-ion modes using G3CA matrix with and without anion dopants (see section on “Evaluation of anionic species for MS2 experiments” regarding anion selection). Chicken ovalbumin is a well-characterized

Figure 5. Normalized ion yields of the anion-adducted NA2 glycans and GBs of anions. The ion yields of [NA2 + anion]− were normalized by dividing them by the ion yields of [A1 − H]−.

a nonlinear relationship was observed between the ion yields of anion-adducted NA2 glycan and the GBs of the attaching anions. This relationship can be rationalized by considering the relative relationship of GBs between the deprotonated analyte molecule [M − H]− and anions. It was found that the ion yield of analyte depends on whether the GB of the anion is greater or

Figure 4. Positive-ion and negative-ion mass spectra of N-glycans derived from ovalbumin. (a) Positive-ion mass spectrum using G3CA without anion dopant. Negative-ion mass spectra using (b) G3CA with NO3− and (c) G3CA with BF4−. Black circle denotes N-glycan signals. Asterisk denotes fragment ion signals. 6100

dx.doi.org/10.1021/ac3009803 | Anal. Chem. 2012, 84, 6097−6103

Analytical Chemistry

Article

Figure 6. Negative-ion MS2 spectra of the anion-adducted NA2 glycans at 100 fmol. (a) PF6− adduct. (b) BF4− adduct. (c) HSO4− adduct. (d) I− adduct. (e) NO3− adduct. (f) Br− adduct. (g) SCN− adduct. (h) H2PO4− adduct. (i) Cl− adduct. G3CA was used as a matrix. The main product ions are illustrated in the inset without distinguishing between the 6-antenna (upper) and 3-antenna (lower), but D (E) ions are only derived from the 6antenna (3-antenna).

It should be noted that the relative ion yields of the anionadducted NA2 glycans do not directly reflect their absolute intensities, which are determined by the s/n ratio in this study, because anion doping can affect the mass spectral noise level, total ion intensity, and laser threshold for ionization. Nevertheless, the column of G3CA in Table 1 and Figure 5 indicate that the combinations of G3CA and BF4−, HSO4−, or I− are most suitable for sensitive detection of N-glycans in negativeion MALDI. However, the critical disadvantage of HSO4− and I− adducts is their lack of applicability for sensitive structural analyses. A CID experiment of these anionic adducts does not provide informative MS2 spectra efficiently, as discussed below. Evaluation of Anionic Species for MS2 Experiments. Since fragmentation of anionic adducts was initiated by proton abstraction from N-glycan to anions, the GB of anions is a crucial parameter for successful MS2 analysis. Anions with low GB are unable to abstract protons from N-glycans, resulting in poor or no glycan fragment ions in the spectra. For example, CID of HSO4− or I− adducted N-glycans exhibits a dominant HSO4− or I− signal in the MS2 spectrum, indicating extensive anion loss during the CID process.6,23,25 Here, we performed MS2 experiments of all anion adducts tested in this study and compared their spectral patterns and fragmentation efficiencies under low-energy CID conditions using an ion trap instrument. Figure 6 presents the negative-ion MS 2 spectra of 100 fmol/well NA2 glycan. At this concentration, the intensity of all precursor ions seemed to be sufficient for MS2 experiments. MS2 of the PF6− adduct did not yield significant glycan fragment ions, in spite of the intense precursor ion, indicating a complete anion loss without proton

less than that of the analyte, and the differences are also important. In an adduct of a small anion and a neutral molecule, the adduct becomes more stable if the GBs of the two components in the mixed dimer are closely matched.36 In the present study, the two components correspond to the deprotonated NA2 glycan and the anion. We used a bracketing approach18,36 to roughly estimate the GB of deprotonated NA2 glycan, a value that is not available in the literature. The GB of deprotonated NA2 glycan was estimated to be between 1294 and 1332 kJ/ mol because low-energy CID of I− (GB; 1294 kJ/mol) adducted NA2 glycan gave I− as the dominant product ion, while Br− (GB; 1332 kJ/mol) adduct gave glycan fragment ions originating from the formation of [M − H]−.6 Our results indicated that for anions with GB values lower than 1300 kJ/ mol, the ion yields of adducts increased with increasing GBs of the attaching anions. This result could be due to the fact that GBs of anions are closer to that of the deprotonated NA2 glycan. The ion yield of HSO4− adduct was slightly higher than expected from its GB (1260 kJ/mol), because of hydrogen bonding between NA2 glycan and HSO4−, which may contribute to the extra stability of the adduct.36 For anions with GB values exceeding 1300 kJ/mol, the ion yields of adducts decreased rapidly with increasing GBs of the attaching anions. This result may be due to (i) the neutralization of anions in the matrix plume, (ii) the enhanced fragmentation of adducts, and/or (iii) the instability of adducts caused by mismatch of GB values between the deprotonated NA2 glycans and anions. 6101

dx.doi.org/10.1021/ac3009803 | Anal. Chem. 2012, 84, 6097−6103

Analytical Chemistry

Article

Figure 7. Negative-ion MS2 spectra of the anion-adducted NA2 glycans at 1 fmol. (a) BF4− adduct. (b) I− adduct. (c) NO3− adduct. G3CA was used as a matrix.

seems reasonable to suggest that the attaching anion is located near the reducing-terminal GlcNAc residue, not branch moieties, by forming multiple hydrogen bonding. In many cases, the multiple hydrogen bonding would likely involve the interaction between the 3-hydroxyl group of reducing-terminal GlcNAc and the attaching anion. Figure 7 presents negative-ion MS2 spectra obtained from 1 fmol NA2 glycan ionized as BF4−, I− or NO3− adducts. The spectra were acquired under the same conditions, including the same spectral accumulation and collision energy. The BF4− adduct produced MS2 spectra with an acceptable s/n ratio, whereas the I− adduct does not exhibit any informative signals in the MS2 spectrum. Each adduct has comparable intensities of precursor ions (Figure 3a and b). MS2 of the BF4− adduct formed product ions of glycans more efficiently than did the I− adduct. Using NO3−, the MS2 spectrum was successfully obtained with a sufficient s/n ratio (Figure 7c). Since the NO3− adduct is less sensitive than the I− or BF4− adduct in MS1 analyses (Figure 3), these results clearly indicate the effectiveness of NO3− adduct for sensitive MS2 measurement.

abstraction from NA2 glycans during the CID process (Figure 6a). Unfortunately, no signal of the anion itself was observable due to the inherent low-mass cutoff of the ion trap instrument. The HSO4− adduct did provide some fragments containing the anions, but their absolute signal intensities were quite low. The absolute signal intensities of the MS2 spectrum of the HSO4− adduct were less than one-fifteenth those of the NO3− adduct, indicating a preferential loss of anions during the CID process. In addition, the absence of D and E ions, which are important diagnostic fragments for less ambiguous structural determination,8 is a critical disadvantage. The D ion contains the intact 6-antenna and the branching mannose, and defines the composition of the antenna.7 Similarly, the E ion reflects the composition of the 3-antenna.8 Cl−, SCN−, H2PO4−, and NO3− adducts readily provided informative MS2 spectra, while Br− adducts exhibited complicated MS2 spectra, which can prevent correct structural determination. Pronounced low m/z fragment ions and various unfavorable signals at m/z 500 to 1200 were observed in the MS2 spectra of the Br− adduct (compare Figure 6e and f). The MS2 experiment of I− adduct needed extensive accumulation to obtain spectra, and they were complicated, as with the Br− adduct. The MS2 of I− or Br− adducts appeared to promote multiple cleavages (2,4A5/Y4 ion at m/z 910.3, 2,4A5/Y5/Y5 ion at m/z 951.5, and 2,4A5/Y5 ion at m/z 1113.3). Using BF4− glycan fragment, ions are produced with moderate efficiency, in spite of its quite low GB. The location of anion attachment is of fundamental interest. Several computer modeling studies have revealed that the multiple hydrogen bonding strengthens the interaction between a small neutral oligosaccharide and the attaching anion, and stabilizes the anion adduct.17,43 In other studies, the production of 2,4A6 ions has been reasonably explained by the initial proton abstraction from the 3-hydroxyl group of the reducing-terminal GlcNAc residue.7 Similarly, the production of D and E ions has been explained via a formation of corresponding C ions, which presumably originates from initial proton abstraction of the same position.7,8 Taking into account these considerations, it

■

CONCLUSIONS Combinations of five liquid matrices and nine anions were evaluated for sensitive analyses of neutral N-glycans in negativeion MALDI-MS. Among the liquid matrices, G3CA exhibited outstanding performances for all the anions. Low-femtomole or subfemtomole detection limits could be achieved using NO3− or BF4−doped G3CA. A homogeneous droplet of the liquid matrix enables us to avoid a time-consuming search for the socalled “sweet spot.” This is a great advantage over conventional solid matrices for rapid and efficient analyses, including MSn experiments. We can conclude that a combination of G3CA and NO3− is the most suitable for sensitive analyses, especially for MS2 measurement of neutral N-glycans in negative-ion MALDI-MSn. A combination of G3CA and BF4− is a possible alternative for superior sensitivity in MS1 detection and medium sensitivity in MS2 analysis. 6102

dx.doi.org/10.1021/ac3009803 | Anal. Chem. 2012, 84, 6097−6103

Analytical Chemistry

■

Article

(27) Mank, M.; Stahl, B.; Boehm, G. Anal. Chem. 2004, 76, 2938− 2950. (28) Laremore, T. N.; Murugesan, S.; Park, T.-J.; Avci, F. Y.; Zagorevski, D. V.; Linhardt, R. J. Anal. Chem. 2006, 78, 1774−1779. (29) Laremore, T. N.; Zhang, F.; Linhardt, R. J. Anal. Chem. 2007, 79, 1604−1610. (30) Fukuyama, Y.; Nakaya, S.; Yamazaki, Y.; Tanaka, K. Anal. Chem. 2008, 80, 2171−2179. (31) Ullmer, R.; Rizzi, A. M. J. Mass Spectrom. 2009, 44, 1596−1603. (32) Schnöll-Bitai, I.; Ullmer, R.; Hrebicek, T.; Rizzi, A.; Lacik, I. Rapid Commun. Mass Spectrom. 2008, 22, 2961−2970. (33) Przybylski, C.; Connet, F.; Bonnaffé, D.; Hersant, Y.; LortatJacob, H.; Daniel, R. Glycobiology 2010, 20, 224−234. (34) Wada, Y.; Tajiri, M.; Yoshida, S. Anal. Chem. 2007, 76, 6560− 6565. (35) Nishikaze, T.; Nakamura, T.; Jinmei, H.; Amano, J. J. Chromatogr. B 2011, 879, 1419−1428. (36) Cai, Y.; Cole, R. B. Anal. Chem. 2002, 74, 985−991. (37) Koppel, I. A.; Burk, P.; Koppel, I.; Leito, I.; Sonoda, T.; Mishima, M. J. Am. Chem. Soc. 2000, 122, 5114−5124. (38) Bartmess, J. E. Negative Ion Energetics Data. in NIST Chemistry WebBook, NIST Standard Reference Database Number 69; Linstrom, P. J., Mallard, W. G., Eds.; National Institute of Standards and Technology: Gaithersburg MD, 2012; http://webbook.nist.gov (retrieved June 5, 2012). (39) Domon, B.; Costello, C. E. Glycoconjugate J. 1988, 5, 397−409. (40) Mechref, Y.; Novotny, M. V. Anal. Chem. 1998, 70, 455−463. (41) Harvey, D. J.; Wing, D. R.; Küster, B.; Wilson, I. B. H. J. Am. Soc. Mass Spectrom. 2000, 11, 564−571. (42) Plasencia, M. D.; Isailovic, D.; Merenbloom, S. I.; Mechref, Y.; Clemmer, D. E. J. Am. Soc. Mass Spectrom. 2008, 19, 1706−1715. (43) Cai, Y.; Concha, M. C.; Murray, J. S.; Cole, R. B. J. Am. Soc. Mass Spectrom. 2002, 13, 1360−1369.

ASSOCIATED CONTENT

S Supporting Information *

Additional material described in the text. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Phone: +81 (0)75 823-2897. Fax: +81 (0)75 823-2900. Email: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research is supported by the Japan Society for the Promotion of Science (JSPS) through its Funding Program for World-Leading Innovative R&D on Science and Technology (FIRST Program).

■

REFERENCES

(1) Apweiler, R.; Hermjakob, H.; Sharon, N. Biochim. Biophys. Acta 1999, 1473, 4−8. (2) Parodi, A. J. Biochim. Biophys. Acta 1999, 1426, 287−295. (3) Fukuda, M.; Sasaki, H.; Fukuda, M. N. Adv. Exp. Med. Biol. 1989, 271, 53−67. (4) Arnold, J. N.; Wormald, M. R.; Sim, R. B.; Rubb, P. M.; Dwek, R. A. Annu. Rev. Immunol. 2007, 25, 21−50. (5) Mariño, K.; Bones, J.; Kattla, J. J.; Rudd, P. M. Nat. Chem. Biol. 2010, 6, 713−723. (6) Harvey, D. J. J. Am. Soc. Mass Spectrom. 2005, 16, 622−630. (7) Harvey, D. J. J. Am. Soc. Mass Spectrom. 2005, 16, 631−646. (8) Harvey, D. J. J. Am. Soc. Mass Spectrom. 2005, 16, 647−659. (9) Harvey, D. J.; Jaeken, J.; Butler, M.; Armitage, A. J.; Rudd, P. M.; Dwek, R. A. J. Mass Spectrom. 2010, 45, 528−535. (10) Harvey, D. J.; Rudd, P. M. Int. J. Mass Spectrom. 2011, 305, 120−130. (11) Amano, J.; Sugahara, D.; Osumi, K.; Tanaka, K. Glycobiology 2009, 19, 592−600. (12) Amano, J.; Osanai, M.; Orita, T.; Sugahara, D.; Osumi, K. Glycobiology 2009, 19, 601−614. (13) Takegawa, Y.; Deguchi, K.; Ito, S; Yoshioka, S.; Nakagawa, H.; Nishimura, S.-I. Rapid Commun. Mass Spectrom. 2005, 19, 937−946. (14) Deguchi, K.; Ito, H.; Takegawa, Y.; Nagai, S.; Nakagawa, H.; Nishimura, S.-I. Rapid Commun. Mass Spectrom. 2006, 20, 412−418. (15) Ito, H.; Yamada, K.; Deguchi, K.; Nakagawa, H.; Nishimura, S.-I. Rapid Commun. Mass Spectrom. 2007, 21, 212−218. (16) Pfenninger, A.; Karas, M.; Finke, B.; Stahl, B. J. Am. Soc. Mass Spectrom. 2002, 13, 1331−1340. (17) Jiang, Y.; Cole, R. B. J. Am. Soc. Mass Spectrom. 2005, 16, 60−70. (18) Cai, Y.; Jiang, Y.; Cole, R. B. Anal. Chem. 2003, 75, 1638−1644. (19) Yamagaki, T; Suzuki, H.; Tachibana, K. Anal. Chem. 2005, 77, 1701−1707. (20) Yamagaki, T.; Suzuki, H.; Tachibana, K. J. Am. Soc. Mass Spectrom. 2006, 17, 67−74. (21) Yamagaki, T.; Suzuki, H.; Tachibana, K. J. Am. Soc. Mass Spectrom. 2007, 18, 714−723. (22) Guan, B.; Cole, R. B. J. Am. Soc. Mass Spectrom. 2008, 19, 1119− 1131. (23) Wong, A. W.; Cancilla, M. T.; Voss, L. R.; Lebrilla, C. B. Anal. Chem. 1999, 71, 205−211. (24) Wong, A. W.; Wang, H.; Lebrilla, C. B. Anal. Chem. 2000, 72, 1419−1425. (25) Domann, P.; Spencer, D. I. R.; Harvey, D. J. Rapid Commun. Mass Spectrom. 2012, 26, 469−479. (26) Armstrong, D. W.; He, L.; Liu, Y.-S. Anal. Chem. 2001, 73, 3679−3686.

■

NOTE ADDED AFTER ASAP PUBLICATION This paper was published on the Web on July 3, 2012. Text corrections were added to Table 1 and the Figure 2 caption, and the corrected version was reposted on July 5, 2012.

6103

dx.doi.org/10.1021/ac3009803 | Anal. Chem. 2012, 84, 6097−6103