Methylamidation for Isomeric Profiling of Sialylated Glycans by

Jul 14, 2014 - ... conducted using an ion source gas of 3 psi, a curtain gas of 35 psi, ..... approach toward de novo glycan characterization: A Man-5...
4 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/ac

Methylamidation for Isomeric Profiling of Sialylated Glycans by NanoLC-MS Qiwei Zhang,†,§ Xiaojun Feng,†,§ Henghui Li,† Bi-Feng Liu,† Yawei Lin,*,‡ and Xin Liu*,† †

Britton Chance Center for Biomedical Photonics at Wuhan National Laboratory for Optoelectronics−Hubei Bioinformatics & Molecular Imaging Key Laboratory, Systems Biology Theme, Department of Biomedical Engineering, College of Life Science and Technology, Huazhong University of Science and Technology, Wuhan 430074, China ‡ Department of Chemistry, School of Science, Wuhan University of Technology, Wuhan 430070, China S Supporting Information *

ABSTRACT: The analysis of isomeric glycans is a challenging task. In this work, a new strategy was developed for isomerspecific glycan profiling using nanoLC-MS with PGC as the stationary phase. Native glycans were derivatized in the presence of methylamine and trispyrrolidinophosphonium hexafluorophosphate and reduced by the ammonia−borane complex. Methylamidation stabilized the retention time and peak width and improved the detection sensitivity of sialylated glycans to 2−80-fold in comparison to previous ESI-MS methods using the positive-ion mode. Up to 19 tetrasialylated glycan species were identified in the derivatized human serum sample, which were difficult to detect in the sample without derivatization. Furthermore, due to high detection sensitivity and chromatographic resolution, more isomeric glycans could be identified from the model glycoprotein Fetuin and the human serum sample. As a result, up to seven isomers were observed for the disialylated biantennary glycan released from Fetuin, and three of them were identified for the first time in this study. Using the developed analytical strategy, a total of 293 glycan species were obtained from the human serum sample, representing an increase of over 100 peaks in comparison to the underivatized sample. The strategy greatly facilitates the profiling of isomeric glycans and the analysis of trace-level samples.

P

spectrometry (LC-MS) has become the technique of choice to profile glycan isomers.11,12 Common chromatographic methods for separating glycans include reversed-phase chromatography (RPC), hydrophilic interaction liquid chromatography (HILIC), and porous graphitized carbon (PGC) chromatography.13 RPC and HILIC enable the separation of the derivatized glycans, such as permethylated and fluorescently labeled glycans.14−17 In comparison to RPC and HILIC, PGC has higher separation

rotein glycosylation is one of the most complex and widespread post-translational modifications, which regulates the biological activities and alters the biochemical environment.1,2 The biosynthesis of the glycans is via a nontemplate-driven system, which is regulated by the availability of nucleotide donors and expression of enzymes,3,4 resulting in the glycans with various isomers. It is a great challenge to characterize the detailed structures of the glycans. Current analytical strategies for probing the intricate structural information on glycans include capillary electrophoresis,5 liquid chromatography (LC),6,7 tandem mass spectrometry (MS/ MS),8,9 and ion mobility−mass spectrometry.10 Although each strategy has respective merits, liquid chromatography−mass © 2014 American Chemical Society

Received: May 17, 2014 Accepted: July 14, 2014 Published: July 14, 2014 7913

dx.doi.org/10.1021/ac501844b | Anal. Chem. 2014, 86, 7913−7919

Analytical Chemistry

Article

Figure 1. XICs of the glycans with different concentrations. (a) XICs of the underivatized H5N4S2 (m/z 1113.45) from 100, 50, 10, and 1 μg of Fetuin. (b) XICs of the derivatized H5N4S2 (m/z 1126.44) from 100, 50, 10, and 1 μg of Fetuin. (c) XICs of the underivatized H6N5S3 (m/z 961.37) from 100, 50, 10, and 1 μg of Fetuin. (d) XICs of the derivatized H6N5S3 (m/z 974.75) from 100, 50, 10, and 1 μg of Fetuin. Structural schemes are given as follows: blue square, N-acetylglucosamine; green circle, mannose; yellow circle, galactose; purple diamond, N-acetylneuraminic acid; red triangle, fucose.

■

EXPERIMENTAL PROCEDURES Chemicals and Materials. N-Glycosidase F (PNGase F) and endoglycosidase buffer pack were from New England Biolabs (Ipswich, MA, USA). Dimethyl sulfoxide (DMSO), methylamine hydrochloride, N-methylmorpholine, (7-azabenzotriazol-1-yloxy) trispyrrolidinophosphonium hexafluorophosphate (PyAOP), ammonia−borane complex, trifluoroacetic acid (TFA), acetic acid, 1-butanol, ethanol, PGC, microcrystalline cellulose (MCC), Fetuin, and human serum were obtained from Sigma-Aldrich (St. Louis, MO, U.S.A.). Acetonitrile (ACN) and methanol were purchased from Merck KGaA (Darmstadt, Germany). Formic acid (FA) and pure water were obtained from Thermo Fisher Scientific (Waltham, MA, USA). The empty cartridges and frits were purchased from Sinopharm Chemical Reagent Co. Ltd. (Shanghai, China). N-Glycan Preparation. The aliquots of Fetuin or human serum sample (5 μL) were diluted to 10 μL and mixed with 90 μL of buffer solution containing 20 mM sodium phosphate (pH = 7.5), 0.13% dodecyl sulfate sodium, and 10 mM dithiothreitol. The sample was denatured at 100 °C for 10 min. After cooling, 12 μL of 10% octylphenoxypolyethoxyethanol (NP40) was added and the solution was allowed to equilibrate for 10 min. The reaction mixture was incubated with PNGase F (5 units) at 37 °C for 18 h. The N-glycans were initially released from the glycoproteins with PNGase F as N-glycosylamine derivatives, which would be hydrolyzed to the native glycans over time.29,30 Since there was a mass difference of 1 Da between the native glycans and their corresponding Nglycosylamines, 1 μL of acetic acid was added to the reaction mixture following incubation at room temperature for 30 min to ensure complete hydrolysis of the released N-glycosylamines.31 Next, the solution was dried in a vacuum concentrator (Eppendorf, Germany). The sample was subsequently purified using PGC cartridge (Supporting Information).32

efficiency due to its retention mechanism, including hydrophobic and ionic interactions.18,19 In addition, PGC is more orthogonal to the subsequent mass separation.20,21 Thus, PGC has been widely used for glycan profiling, especially for isomerspecific glycans.12,22−24 Previous literature showed that PGC provided unstable chromatographic resolution for native sialylated glycans with different concentrations25 because the interactions between PGC and the sialylated glycans partly depend on the polar and the ionic nature of the sialic acid residues, resulting in prolonged retention of the sialylated glycans.18 Neutralization of the sialic acid residues prior to LC-MS analysis might reduce their influence during the PGC separation, improving the chromatographic resolution of sialylated glycans. As an important neutralization method for sialic acid residues, methylamidation has recently been used for the identification of sialylated glycans by matrix-assisted laser desorption ionization−mass spectrometry (MALDI-MS) in the positive-ion mode.26−28 In this work, a new strategy based on methylamidation was developed for isomer-specific glycan profiling by coupling nanoflow liquid chromatography (nanoLC) to electrospray ionization−mass spectrometry (ESI-MS). Native glycans were neutralized and reduced prior to LC-MS analysis. The neutralization stabilized the retention time and peak width and improved the detection sensitivity of sialylated glycans. Furthermore, it significantly increased the separation efficiency of PGC, facilitating the identification of isomeric glycans. As a result, a total of 293 glycan species, corresponding to over 70 N-glycan compositions, were identified from the human serum sample. The strategy opened up a new avenue for comprehensive isomer-specific glycan profiling with high sensitivity and reproducibility. 7914

dx.doi.org/10.1021/ac501844b | Anal. Chem. 2014, 86, 7913−7919

Analytical Chemistry

Article

N-Glycan Derivatization. The dried glycans were derivatized according to the method described previously.26 Briefly, the sample was dissolved in 25 μL of DMSO containing 1 M methylamine hydrochloride and 0.5 M N-methylmorpholine prior to the addition of 25 μL of DMSO containing 50 mM PyAOP. The reaction mixture was incubated at room temperature for 30 min and then diluted with 0.5 mL of binding solution consisting of 1-butanol/ethanol/H2O (4:1:1, v/v/v). The purification of the derivatized glycans was performed using the self-packed MCC cartridge. The cartridge was washed with 3 mL of water, followed by 3 mL of binding solution. The diluted sample was applied to the cartridge and washed with 3 mL of binding solution. Next, the glycans were eluted with 1.2 mL of ethanol/H2O (1:1, v/v) and dried under vacuum. Finally, the glycans were reduced by the ammonia−borane complex (Supporting Information).33 NanoLC-MS Analysis. The trap and elute mode was applied to separate samples using a NanoLC Ultra System (Eksigent, USA) equipped with a trap column (150 μm i.d. × 1 cm long; PGC, 5 μm; Proteomics Front, China) and a separation column (75 μm i.d. × 10 cm long; PGC, 5 μm; Proteomics Front, China.). The solvent A consisted of 5% ACN solution containing 0.1% FA (v/v), and the solvent B consisted of 95% ACN solution containing 0.1% FA (v/v). The reduced glycans were dissolved in 40 μL of solvent A, and a 2 μL aliquot of the solution was loaded into the trap column at a flow rate of 2.0 μL/min for 10 min. The analytical separation was conducted at a flow rate of 400 nL/min with gradient elution. To evaluate the detection sensitivity, a simple 20 min gradient was used: 0 min 15% solvent B; 10 min 80% solvent B; 11 min 95% solvent B; 15 min 95% solvent B; 16 min 15% solvent B; 20 min 15% solvent B. For the separation of isomeric glycans, a 50 min gradient was used: 0 min 5% solvent B; 5 min 5% solvent B; 30 min 30% solvent B; 40 min 80% solvent B; 41 min 95% solvent B; 45 min 95% solvent B; 46 min 5% solvent B; 50 min 5% solvent B. Data were acquired using a TripleTOF 5600 System (AB SCIEX, USA) equipped with a nanospray source. Data acquisition was conducted using an ion source gas of 3 psi, a curtain gas of 35 psi, an ion spray voltage of 2.3 kV, an interface heater temperature of 150 °C, and a collision energy of 20 eV for collision-induced dissociation (CID). MS was operated in the positive-ion mode with a mass range of 500−3000 m/z, and MS/MS was acquired in the information dependent acquisition (IDA) mode with a mass range of 100−2000 m/z. The 20 most abundant precursor ions with charge numbers from 2 to 5 were scanned in the IDA mode. Each cycle consisted of a MS acquisition for 0.25 s and a total of 20 MS/MS scannings for 2 s. Data were processed with PeakView 1.2 software (AB SCIEX, USA), and the illustrations of N-glycans were edited by the GlycoWorkbench 2.1 software.34 Glycan compositions were abbreviated as follows: hexose (H), N-acetylhexosamine (N), N-acetylneuraminic acid (S), and fucose (F).

Figure 2. Isomeric sialylated glycans separated by the PGC approach. (a) XIC of the underivatized H5N4S2 at m/z 1113.41 from Fetuin. (b) XIC of the derivatized H5N4S2 at m/z 1126.44 from Fetuin. (c) XIC of the underivatized H5N4S2 at m/z 1113.43 from the human serum sample. (d) XIC of the derivatized H5N4S2 at m/z 1126.44 from the human serum sample. The compound peaks are indicated by the letters and numbers.

the use of methylamidation for the analysis of the sialylated glycans by nanoLC-MS. Previous literature showed that no byproducts were observed when methylamidation was used for MALDI-MS analysis.26,27 In our case, we observed several byproducts in the ESI-MS spectra, which might result from less ion suppression of the electrospray during LC separation (Figure S1, Supporting Information). However, the reaction efficiency of methylamidation was relatively stable, which was suitable for the following studies. Enhanced Separation Stability. It has been demonstrated that the chromatographic resolution was associated with the sample concentration with PGC as the stationary phase, especially for native sialylated glycans.25 Figure 1a,c illustrated the extracted ion chromatograms (XICs) of the underivatized H5N4S2 and H6N5S3 from 100, 50, 10, and 1 μg of Fetuin, which showed different peak widths and retention time shifts. However, the peak widths and retention times remained unchanged for different concentrations of the derivatized glycans (Figure 1b,d). These results demonstrated that methylamidation enhanced separation stability of the sialylated glycans. The change was likely associated with the retention mechanisms of PGC, involving polar and ionic interactions. The interactions between PGC and the sialic acid residues promote the retention of the sialylated glycans.18,19 Moreover, more sialic acid residues and more antennas have a greater retention than less sialylated and branched glycans. Thus, native sialylated glycans could be well retained on PGC.12 However, the molecule properties were changed with the carboxyl group converting to the neutral group, which reduced the polar and

■

RESULTS AND DISCUSSION PGC has been widely employed for the separation of isomeric glycans due to its high separation efficiency. However, it is difficult for PGC to separate the isomers of native sialylated glycans. Here, methylamidation was used to neutralize the sialic acid residues to overcome this issue. Fetuin is a model protein that exists in a variety of glycoforms including mono-, di-, tri-, and tetrasialylated glycans, which was first studied to evaluate 7915

dx.doi.org/10.1021/ac501844b | Anal. Chem. 2014, 86, 7913−7919

Analytical Chemistry

Article

Figure 3. MS/MS spectra of the derivatized H5N4S2 released from Fetuin. (a) MS/MS spectra of peak b1 from Figure 2b. (b) MS/MS spectra of peak b7 from Figure 2b. The fragment ions are assigned according to Domon and Costello nomenclature.37

residues (Figure S4, Supporting Information). In fact, the separation efficiency of PGC was not improved significantly for the derivatized monosialylated glycans. However, methylamidation of the sialylated glycans with two or more sialic acid residues significantly increased their chromatographic resolution. It is worth mentioning that it was difficult to detect the compositions of native tetrasialylated glycans due to their low abundance (Figure S4g). With the increased detection sensitivity and separation efficiency after methylamidation, the tetrasialylated glycans including isomers could be successfully observed in this study (Figure S4h). In addition, methylamidation had little influence for the separation of the neutral glycans (Figure S5, Supporting Information). As a typical disialylated glycan, H5N4S2, released from both Fetuin and the human serum sample was studied in detail. For the underivatized H5N4S2, the XICs presented severe peak tailing and retention time shifts (Figure 2a,c). The results mainly resulted from the unsuitable concentrations of samples and strong interactions between PGC and the sialylated glycans. Especially, the polar and the ionic nature of the sialic acid residues could significantly promote their affinity toward the PGC,18 resulting in poor chromatographic separation. In contrast, the XICs of the derivatized glycans presented narrow peak width and stable retention time (Figure 2b,d). The high separation efficiency stemmed from the neutralization of the sialic acid residues, which reduced the interactions between PGC and the sialylated glycans. Therefore, some other structures perhaps appeared to play an increasing role in the separation of isomeric glycans. As a result, up to seven isomers were observed in H5N4S2 released from Fetuin (Figure 2b), and three isomers were identified for H5N4S2 from the human

the ionic interactions between PGC and the sialylated glycans and led to narrower peak width and less retention. Consequently, methylamidation stabilized the peak width and retention time of the sialylated glycans, enhancing separation stability. Increased Sensitivity and Reproducibility. Methylamidation could reduce the loss of a carboxyl moiety of native sialic acid residues and increase the detection sensitivity of sialylated glycans in the positive-ion mode (Figures S2 and S3, Supporting Information). In general, the enhancement exhibited a dynamic range based on the numbers of the sialic acid residues and the concentrations of samples. For the former case, more sialic acid residues resulted in larger enhancement. For the latter, the sensitivity enhancement varied with different concentrations of samples. As a result, it was a 2−80-fold enhancement depending on different concentrations and types of sialylated glycans. Particularly, the detection limit was nearly 100-fold lower than that of the underivatized tetrasialylated glycans in the positive-ion mode. Since native sialylated glycans are less sensitive in LC-MS analysis partly due to the presence of the carboxylic acid moiety,35 the enhancement may primarily benefit from the neutralization of the carboxyl groups. In addition, it was found that methylamidation also improved the reproducibility of MS detection in comparison to those of underivatized sialylated glycans (Figure S2, Supporting Information). Separation of Isomeric N-Glycans. To obtain the isomerspecific information, a 50 min gradient was used for the separation of the glycans from both Fetuin and the human serum sample. After methylamidation, the separation efficiency of PGC was improved based on different number of sialic acid 7916

dx.doi.org/10.1021/ac501844b | Anal. Chem. 2014, 86, 7913−7919

Analytical Chemistry

Article

Figure 5. All compound peaks of the glycans from the human serum sample. (a) Extracted compound chromatograms of all glycans found in the derivatized serum sample. Colors denote different glycan groups. Six groups are given: high-mannose, high-mannose type of glycans; neutral, neutral glycans (not including high-mannose type of glycans); monosia, monosialylated glycans; disia, disialylated glycans; trisia, trisialylated glycans; tetrasia, tetrasialylated glycans. (b) Peak area of all ion species (1 ≤ z ≤ 4) associated with a single group. The numbers on the horizontal axis show the sum of compound peaks from a single group.

Figure 4. Typical MS/MS spectra of the derivatized glycans with the fucose residue. (a) XIC of the derivatized H5N4S2F1 (m/z 799.98) from the human serum sample. The peaks are indicated by the numbers. (b) MS/MS spectra of peak 1. (c) MS/MS spectra of peak 2.

serum sample (Figure 2d). The MS/MS spectra showed the different fragment patterns of isomeric H5N4S2 released from Fetuin (Figure S6, Supporting Information). Based on the fragment species, these seven isomers could be divided into two groups. One group included four isomers (peaks b1, b2, b3, and b4), which have been identified in previous studies.12,36 They were also observed in this study by nanoLC-MS. As shown in Figure 3a, the MS/MS spectra of peak b1 revealed the structure of the disialylated glycan. Sixteen ion species were observed by CID mode, including three B ions, three B/Y ions, and 10 Y ions. The ion species of peaks b2, b3, and b4 were almost the same as that of peak 1 (Figure S6, Supporting Information). Although the CID-based MS/MS method often failed to produce sufficient numbers of cross-ring fragments,38 it provided abundant Y ions which produced a ladder of ions. The ladder consisted of 913.35 (Y4α/Y4β), 1116.44 (Y4α/Y5β or Y5α/Y4β or Y3α/Y6β or Y6α/Y3β), 1278.48 (Y6α/Y4β or Y4α/Y6β), 1582.62 (Y4), 1785.68 (Y5), and 1947.76 (Y6), which could facilitate the structural assignment by database searching. The other group included three new isomeric structures, which were first identified in this study (peaks b5, b6, and b7). The MS/MS spectra were largely consistent with that shown in Figure 3a, but two particular fragment ions at m/z 508.21 (B3β/ Y5β) and 1258.44 (Y3α/Y5β) were observed in their spectra (Figure 3b), indicating the presence of a sialic acid residue linked to the N-acetylhexosamine residue of H5N4S2. Generally, this structure was observed in the tetrasialylated glycan. 36 However, our results demonstrated that the disialylated glycan also contained this structure. The identi-

fication of the novel structure suggested that the strategy could greatly facilitate the isomer-specific glycan profiling. It should be mentioned that this structure was not observed in the human serum sample, which indicated that the strategy had the capacity of identifying glycan isomers from different sources in comparison to previous methods. Identification of Fucosylated Glycans. It was difficult to identify the fucose position in the fucosylated glycans by MS/ MS because the fucose residues could be transferred between the antennae or from the antennae to the trimannosyl core.39 The complete separation of the fucosylated glycans would be helpful to elucidate their structures. Only two peaks from the underivatized H5N4S2F1 were observed (Figure S4, Supporting Information), whose MS/MS spectra were identical (data not shown). However, four peaks representing the isomers of the H5N4S2F1 were detected after derivatization (Figure 4a). The MS/MS spectra of peak 1 was different than that of the other three peaks, in which the ion species at m/z 370.17 was absent (Figure 4b). These results indicated that the glycan compound corresponding to peak 1 contained a terminal fucose residue. However, the presence of the ions at m/z 1059.41 showed the fucose migration from the antennae to the trimannosyl core, which was consistent with previous findings reported for the fucose migrations.39 In contrast, the glycan compound corresponding to peak 2 contained a core fucose residue due to the presence of Y1 ion at m/z 370.17 (Figure 7917

dx.doi.org/10.1021/ac501844b | Anal. Chem. 2014, 86, 7913−7919

Analytical Chemistry

Article

4c). A minor ion at m/z 512.20 might result from the fucose migration from the core to the antennae. In addition, although the relative abundances were different, the MS/MS fragmentation patterns from peaks 3 and 4 were almost the same as that from peak 2 (data not shown). Therefore, the terminal elaboration structures of these isomers could not be effectively identified in this study. To characterize precise structures of the glycans, a combination of exoglycosidase digestion and other MS-based approaches, such as MSn, electronic excitation dissociation, and electron capture dissociation, may need to be employed for further research.9,40−42 Profiling of N-Glycans Released from Human Serum Sample. All N-glycan species observed in the derivatized serum sample were divided into six groups (Figure 5), highmannose (high-mannose type of glycans), neutral (neutral glycans, not including high-mannose type of glycans), monosia (monosialylated glycans), disia (disialylated glycans), trisia (trisialylated glycans), and tetrasia (tetrasialylated glycans). The abundance of the extracted compounds spanned 4 orders of magnitude, showing high sensitivity and resolution of the developed method (Figure 5a). In order to evaluate the performances of methylamidation in the analysis of the human serum sample, the peak area of all ion species (1 ≤ z ≤ 4) associated with a single group was calculated. The peak area of each group is shown in Figure 5b, and the numbers of the horizontal axis show the sum of compound peaks from the underivatized and the derivatized samples. It was found that the peak areas of the neutral glycans were reduced after derivatization. The decrease might result from the loss of the neutral glycans after derivatization, but the loss was less than 50%, which did not impact the detection of the neutral glycans. However, the areas and the numbers of the peaks were both increased for the sialylated glycans after methylamidation (Figure 5b). Up to 192 compound peaks with 44 distinct compositions were identified in the derivatized sample, representing an increase of 102 peaks and 16 compositions in comparison to the underivatized sample. Particularly, nearly 20 tetrasialylated glycan species were identified in the derivatized sample, which were difficult to detect in the sample without derivatization. Due to high detection sensitivity and chromatographic resolution, a total of 293 compound peaks and over 70 distinct glycan compositions from the human serum sample were identified using the analytical strategy (Table S1, Supporting Information). In contrast, less than 200 glycan species were identified in a previous study.43 It was shown that the developed strategy could greatly facilitate the isomerspecific glycan profiling with high sensitivity.

Cancer patients frequently display the glycans with different levels or structures than those observed in healthy individuals.44 The aberrant glycosylation is partly associated with the glycosyltransferases,45 which change the relative abundances or the structures of isomeric glycans, leading to potential isomer-specific biomarkers. In comparison to compositional glycan profiling, structure-specific profiling may be more useful in revealing glycan biomarkers due to higher specificity. In this work, the developed strategy greatly promoted the identification of isomer-specific sialylated glycans with high sensitivity and reproducibility. Therefore, it holds a high potential for discovering glycan-based cancer biomarkers in the future.

■

ASSOCIATED CONTENT

S Supporting Information *

Additional figures and data as indicated in text. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. Tel: +86-27-87792203. *E-mail: [email protected]. Tel: +86-27-87756662. Author Contributions §

Q.Z. and X.F. contributed equally to this work.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors gratefully acknowledge the financial support from the National Natural Science Foundation of China (20905027).

■

REFERENCES

(1) Dwek, R. A. Chem. Rev. 1996, 96, 683−720. (2) Turnbull, J. E.; Field, R. A. Nat. Chem. Biol. 2007, 3, 74−77. (3) Muthana, S. M.; Campbell, C. T.; Gildersleeve, J. C. ACS Chem. Biol. 2012, 7, 31−43. (4) Han, L.; Costello, C. E. Biochemistry (Moscow) 2013, 78, 710− 720. (5) Mechref, Y.; Muzikar, J.; Novotny, M. V. Electrophoresis 2005, 26, 2034−2046. (6) Lipniunas, P. H.; Neville, D. C.; Trimble, R. B.; Townsend, R. R. Anal. Biochem. 1996, 243, 203−209. (7) Takegawa, Y.; Deguchi, K.; Ito, H.; Keira, T.; Nakagawa, H.; Nishimura, S. J. Sep. Sci. 2006, 29, 2533−2540. (8) Prien, J. M.; Prater, B. D.; Cockrill, S. L. Glycobiology 2010, 20, 629−647. (9) Yu, X.; Jiang, Y.; Chen, Y.; Huang, Y.; Costello, C. E.; Lin, C. Anal. Chem. 2013, 85, 10017−10021. (10) Plasencia, M. D.; Isailovic, D.; Merenbloom, S. I.; Mechref, Y.; Novotny, M. V.; Clemmer, D. E. J. Am. Soc. Mass Spectrom. 2008, 19, 1706−1715. (11) Hua, S.; Williams, C. C.; Dimapasoc, L. M.; Ro, G. S.; Ozcan, S.; Miyamoto, S.; Lebrilla, C. B.; An, H. J.; Leiserowitz, G. S. J. Chromatogr. A 2013, 1279, 58−67. (12) Palmisano, G.; Larsen, M. R.; Packer, N. H.; Thaysen-Andersen, M. RSC Adv. 2013, 3, 22706−22726. (13) Melmer, M.; Stangler, T.; Premstaller, A.; Lindner, W. J. Chromatogr. A 2011, 1218, 118−123. (14) Hu, Y.; Mechref, Y. Electrophoresis 2012, 33, 1768−1777. (15) Anumula, K. R. Anal. Biochem. 2006, 350, 1−23. (16) Deguchi, K.; Keira, T.; Yamada, K.; Ito, H.; Takegawa, Y.; Nakagawa, H.; Nishimura, S. J. Chromatogr. A 2008, 1189, 169−174. (17) Saldova, R.; Huffman, J. E.; Adamczyk, B.; Muzinic, A.; Kattla, J. J.; Pucic, M.; Novokmet, M.; Abrahams, J. L.; Hayward, C.; Rudan, I.;

■

CONCLUSIONS Here, we demonstrated methylamidation to be a useful strategy for the analysis of the N-glycans using nanoLC-MS with PGC as the stationary phase. Methylamidation stabilized the retention time and ionization efficiency, improving the detection sensitivity in the positive-ion mode and providing higher chromatographic resolution for the isomeric sialylated glycans. Using the developed strategy, over 290 N-glycan species and over 70 distinct N-glycan compositions were identified from the human serum sample. The strategy has shown potential applications in efficient and information-rich screening of glycans, involving the identification of structures and the analysis of trace-level samples. 7918

dx.doi.org/10.1021/ac501844b | Anal. Chem. 2014, 86, 7913−7919

Analytical Chemistry

Article

Wild, S. H.; Wright, A. F.; Polasek, O.; Lauc, G.; Campbell, H.; Wilson, J. F.; Rudd, P. M. J. Proteome Res. 2012, 11, 1821−1831. (18) Pereira, L. J. Liq. Chromatogr. Relat. Technol. 2008, 31, 1687− 1731. (19) Ross, P.; Knox, J. H. Adv. Chromatogr. 1997, 37, 121−162. (20) Pabst, M.; Altmann, F. Proteomics 2011, 11, 631−643. (21) Pabst, M.; Wu, S. Q.; Grass, J.; Kolb, A.; Chiari, C.; Viernstein, H.; Unger, F. M.; Altmann, F.; Toegel, S. Carbohydr. Res. 2010, 345, 1389−1393. (22) Hua, S.; Jeong, H. N.; Dimapasoc, L. M.; Kang, I.; Han, C.; Choi, J. S.; Lebrilla, C. B.; An, H. J. Anal. Chem. 2013, 85, 4636−4643. (23) Jensen, P. H.; Karlsson, N. G.; Kolarich, D.; Packer, N. H. Nat. Protoc. 2012, 7, 1299−1310. (24) Karlsson, N. G.; Wilson, N. L.; Wirth, H. J.; Dawes, P.; Joshi, H.; Packer, N. H. Rapid Commun. Mass Spectrom. 2004, 18, 2282−2292. (25) Hua, S.; An, H. J.; Ozcan, S.; Ro, G. S.; Soares, S.; DeVereWhite, R.; Lebrilla, C. B. Analyst 2011, 136, 3663−3671. (26) Liu, X.; Qiu, H.; Lee, R. K.; Chen, W.; Li, J. Anal. Chem. 2010, 82, 8300−8306. (27) Nishikaze, T.; Kawabata, S.; Tanaka, K. Anal. Chem. 2014, 86, 5360−5369. (28) Zhou, H.; Warren, P. G.; Froehlich, J. W.; Lee, R. S. Anal. Chem. 2014, 86, 6277−6284. (29) Palm, A. K.; Novotny, M. V. Rapid Commun. Mass Spectrom. 2005, 19, 1730−1738. (30) Nakano, M.; Higo, D.; Arai, E.; Nakagawa, T.; Kakehi, K.; Taniguchi, N.; Kondo, A. Glycobiology 2009, 19, 135−143. (31) Kamoda, S.; Nakano, M.; Ishikawa, R.; Suzuki, S.; Kakehi, K. J. Proteome Res. 2005, 4, 146−152. (32) Desantos-Garcia, J. L.; Khalil, S. I.; Hussein, A.; Hu, Y.; Mechref, Y. Electrophoresis 2011, 32, 3516−3525. (33) Alley, W. R., Jr.; Vasseur, J. A.; Goetz, J. A.; Svoboda, M.; Mann, B. F.; Matei, D. E.; Menning, N.; Hussein, A.; Mechref, Y.; Novotny, M. V. J. Proteome Res. 2012, 11, 2282−2300. (34) Ceroni, A.; Maass, K.; Geyer, H.; Geyer, R.; Dell, A.; Haslam, S. M. J. Proteome Res. 2008, 7, 1650−1659. (35) Kirsch, S.; Bindila, L. Bioanalysis 2009, 1, 1307−1327. (36) Green, E. D.; Adelt, G.; Baenziger, J. U.; Wilson, S.; Van Halbeek, H. J. Biol. Chem. 1988, 263, 18253−18268. (37) Domon, B.; Costello, C. E. Glycoconjugate J. 1988, 5, 397−409. (38) Zaia, J. Mass Spectrom. Rev. 2004, 23, 161−227. (39) Wuhrer, M.; Koeleman, C. A.; Hokke, C. H.; Deelder, A. M. Rapid Commun. Mass Spectrom. 2006, 20, 1747−1754. (40) Takegawa, Y.; Deguchi, K.; Ito, S.; Yoshioka, S.; Sano, A.; Yoshinari, K.; Kobayashi, K.; Nakagawa, H.; Monde, K.; Nishimura, S. Anal. Chem. 2004, 76, 7294−7303. (41) Kurimoto, A.; Daikoku, S.; Mutsuga, S.; Kanie, O. Anal. Chem. 2006, 78, 3461−3466. (42) Zhao, C.; Xie, B.; Chan, S. Y.; Costello, C. E.; O’Connor, P. B. J. Am. Soc. Mass Spectrom. 2008, 19, 138−150. (43) Chu, C. S.; Ninonuevo, M. R.; Clowers, B. H.; Perkins, P. D.; An, H. J.; Yin, H.; Killeen, K.; Miyamoto, S.; Grimm, R.; Lebrilla, C. B. Proteomics 2009, 9, 1939−1951. (44) Dube, D. H.; Bertozzi, C. R. Nat. Rev. Drug Discovery 2005, 4, 477−488. (45) Noda, K.; Miyoshi, E.; Gu, J.; Gao, C. X.; Nakahara, S.; Kitada, T.; Honke, K.; Suzuki, K.; Yoshihara, H.; Yoshikawa, K.; Kawano, K.; Tonetti, M.; Kasahara, A.; Hori, M.; Hayashi, N.; Taniguchi, N. Cancer. Res. 2003, 63, 6282−6289.

7919

dx.doi.org/10.1021/ac501844b | Anal. Chem. 2014, 86, 7913−7919