Rapid Quantitative Profiling of N-Glycan by the Glycan-Labeling

Article pubs.acs.org/ac

Rapid Quantitative Profiling of N‑Glycan by the Glycan-Labeling Method Using 3‑Aminoquinoline/α-Cyano-4-hydroxycinnamic Acid Kaoru Kaneshiro,*,† Makoto Watanabe,† Kazuya Terasawa,‡ Hiromasa Uchimura,‡ Yuko Fukuyama,† Shinichi Iwamoto,† Taka-Aki Sato,† Kazuharu Shimizu,‡ Gozoh Tsujimoto,‡ and Koichi Tanaka† †

Koichi Tanaka Laboratory of Advanced Science and Technology, Shimadzu Corporation, 1, Nishinokyo-Kuwabaracho Nakagyo-ku, Kyoto 604-8511, Japan ‡ World-Leading Drug Discovery Research Center, Graduate School of Pharmaceutical Sciences, Kyoto University, 46-29 Yoshida-Shimo-Adachi-cho, Sakyo-ku, Kyoto 606-8501, Japan ABSTRACT: Protein glycosylation is a crucial phenomenon for understanding protein functions, since its patterns and degree are associated with many biological processes, such as intercellular signaling and immune response. We previously reported a novel glycan-labeling method using a 3-ainoquinoline/α-cyano-4-hydroxycinnamic acid (3-AQ/CHCA) liquid matrix for highly sensitive detection by matrix-assisted laser desorption/ionization (MALDI)-mass spectrometry (MS). In the present study, we examined the practicality of this method for qualitative and quantitative glycan profile analysis. We first investigated the reproducibility of the data for 16 N-glycans prepared from human epidermal growth factor receptor type 2 (HER2). All of the data obtained in intra-assays and interassays were highly correlated with statistical significance (R2 > 0.9, p < 0.05). In addition, the HER2 glycosylation pattern differed significantly between different breast cancer cell lines SK-BR-3 and BT474 in a comparative analysis of profile data. Finally, the quantitative capability of this method was examined by using PAlabeled monosialylated N-glycan as an internal standard (IS). Using IS for AQ-labeled neutral and sialylated standard glycans, the ion peak intensity was highly linear (R2 > 0.9) from 0.5 to 5000 fmol. Furthermore, using IS for HER2 N-glycans, all of the Nglycans were highly linear with their dilution factors (R2 > 0.9). These results suggest that our developed AQ labeling method enabled rapid qualitative and quantitative analyses of glycans. This glycan analysis method should contribute to the field of biomarker discovery and biomedicine in applications such as quality control of biotechnology-based drugs.

P

depends on the kind of lectin. It should be noted that this technique cannot obtain detailed information on glycan structures.11 Recently, mass spectrometry (MS) is becoming an indispensable tool for glycan analysis because MSn analysis provides detailed information on glycan structures. However, glycan analysis based on MS is time-consuming because of complicated pretreatment, such as the removing impurities including excess reagents. In general, glycans enzymatically or chemically released from glycoproteins need to be chemically derivatized to increase their sensitivity in matrix-assisted laser desorption/ionization (MALDI)-MS analysis.12,13 The most widely used derivatization technique is reductive amination, in which the reducing-terminus aldehyde group of glycan is labeled with alkylamines such as 2-aminopyridine (PA).14,15 This derivatization method increases the hydrophobicity and the charge property of glycans, facilitating enhanced ionization efficiency of glycans in MS.16 However, this technique requires additional processes to reduce the Schiff

rotein glycosylation is a widely recognized complex posttranslational modification.1 More than half of all proteins are estimated to be glycosylated.2 Protein glycosylation is defined by many factors (e.g., amino acid sequences, local peptide conformations at the glycosylation sites, and activities of enzymes and cofactors), thus its patterns can be very diverse and dynamic. A variety of biological processes, such as extra/ intercellular signal transduction, cell division, cell adhesion, and immune response, are regulated by protein glycosylation.3−5 In addition, changes in glycosylation patterns have been associated with a number of diseases, including rheumatoid arthritis and other malignancies. Therefore, it is important to determine glycan structures and glycosylation patterns to understand the roles of glycan at the biological level and to identify useful biomarkers.6,7 Generally, a lectin microarray based on glycan binding to the arrayed lectins is used to conduct high-throughput glycosylation pattern analyses of proteins.8 This method enables glycosylation pattern analysis of glycoproteins, such as cell surface proteins, without liberating glycans from proteins.9,10 Therefore, this technique is widely used for various research fields as a high-throughput screening tool. However, the information on glycan expression obtained with this technique © 2012 American Chemical Society

Received: May 31, 2012 Accepted: July 25, 2012 Published: July 25, 2012 7146

dx.doi.org/10.1021/ac301484f | Anal. Chem. 2012, 84, 7146−7151

Analytical Chemistry

Article

Figure 1. Negative ion mass spectra of N-linked glycans released from HER2 in SK-BR-3 and BT-474. Approximately 30 ion peaks of N-glycans were observed in either of the SK-BR-3 or BT-474 lines, with the N-glycan profiling pattern in SK-BR-3 differing from that in BT-474. Among these, 17 peaks (G1−G17) were observed in both cell lines. While AQ-labeled neutral glycans were detected as phosphate adduct ions [M + 126 + H2PO4]− derived from ammonium dihydrogen phosphate (ADP) in 3-AQ/CHCA, sialylated glycans were detected as deprotonated ions [M + 126 − H]−.

■

base and to remove excess reagents, which makes sample preparation time-consuming and can cause sample loss.17 To address this problem, we previously developed a novel glycan analysis method (the 3-ainoquinoline (AQ) labeling method) using 3-ainoquinoline/α-cyano-4-hydroxycinnamic acid (3-AQ/CHCA) liquid matrix (LM) for highly sensitive detection of glycans by MALDI-MS.18 This method does not require the additional pretreatment processes described above. In addition, using 3-AQ/CHCA LM significantly enhances the sample homogeneity19,20 (compared with using solid matrixes), providing a shorter measurement time and quantitative analysis by MALDI-MS. These observations suggest that this method is suitable for rapid quantitative analysis of glycans by MALDIMS. Human epidermal growth factor receptor type 2 (HER2) (also known as NEU or ERBB2) is a member of the human epidermal growth factor receptor (EGFR) family. Approximately 20% of breast cancers exhibit HER2 gene amplification or overexpression, resulting in an aggressive tumor phenotype and reduced survival.21−23 Although the extracellular region of HER2 contains eight potential N-glycosylation sites, its glycosylation pattern and glycan structure are not yet fully clarified. In the present study, N-glycans released from HER2 protein were subjected to glycan analysis using the AQ labeling method to perform glycosylation pattern analysis and evaluate its applicability to biological samples. The data reproducibility was examined by correlation analysis of the obtained data. Furthermore, the linear dynamic range of quantification with this method was investigated using an internal standard for standard glycans and N-glycans of HER2.

MATERIALS AND METHODS

Sample Preparations. Materials. Breast cancer cell lines (SK-BR-3 and BT-474) were obtained from American Type Culture Collection (Manassas, VA). Commercially available Nlinked glycan standards PA-labeled A1 glycan [Sia(GalGlcNAc)2Man3(GlcNAc)2] were purchased from Takara Bio Inc. (Shiga, Japan). NA2 glycan [(GalGlcNAc)2Man3(GlcNAc)2] was purchased from Sigma-Aldrich Japan K.K. (Tokyo, Japan). A1 glycan [Sia(GalGlcNAc) 2 Man 3 (GlcNAc) 2 ] and A2 glycan [Sia 2 (GalGlcNAc)2Man3(GlcNAc)2] were purchased from Ludger Ltd. (Oxford, U.K.). Preparation for Total Cell Lysate and Isolation of HER2 Protein. SK-BR-3 and BT-474 were cultured in Roswell Park Memorial Institute medium containing 10% fetal bovine serum (FBS) or Dulbecco’s Modified Eagle Medium containing 100 units/mL of penicillin or streptomycin. The cells were maintained at 37 °C in 5% CO2. Semiconfluent cells in a 10cm dish were lysed with 1 mL of lysis buffer containing 20 mM Hepes, pH 7.5, 1 mM MgCl2, 1 mM EGTA, 150 mM NaCl, 1% (v/v) Nonidet P-40, and 1% (v/v) Proteinase Inhibitor Cocktail (Nacalai Tesque, Kyoto, Japan). To immunoprecipitate HER2 proteins, HER2 antibody (Clone N24, Thermo Fisher Scientific, Fremont, CA) was labeled with biotin using Biotin Labeling Kit-NH2 (Dojindo Laboratories, Kumamoto, Japan) according to the manufacturer’s instructions. Cell lysates were incubated with 5 μg of biotin-labeled HER2 antibody with gentle rocking overnight at 4 °C and further incubated in the presence of streptavidin magnetic beads (Dynabeads MyOne Streptavidin T1, Invitrogen Oslo, Norway) (50 μL) for 2 h at 4 °C. The beads were collected using a magnetic stand and washed three times with lysis buffer and one time with TBS (20 7147

dx.doi.org/10.1021/ac301484f | Anal. Chem. 2012, 84, 7146−7151

Analytical Chemistry

Article

Figure 2. Reproducibility of HER2 N-glycans profile data with the AQ-labeling method. The ion peak intensities for G1−G16 were individually normalized to the total ion intensity of G1−G16. The MS peak intensity of each N-glycan (G1−G16) in experiment 1 is plotted on the horizontal line in parts a and b. Those for experiments 2 and 3 are plotted on the vertical line in parts a and b. (c) The averaged intensities of N-glycan obtained on different days (lot 1 (n = 4) and lot 2 (n = 4)) are plotted on the horizontal and vertical lines.

the intraexperiment variations. The ion peak intensity for G1− G16 was normalized to the total ion intensity of G1−G16. AQ labeling followed by MS analysis was carried out five times on a single sample prepared from SK-BR-3. The obtained data (n = 4) had high correlation coefficients (R2 > 0.9, p < 0.05) (Figure 2a,b). In turn, when the coefficient between data from samples prepared on different days was examined to investigate the interexperimental variations, they also exhibited high correlation (R2 > 0.9, p < 0.05) (Figure 2c). Similar results were observed in BT-474 (data not shown). These observations demonstrate the reliability of the glycan profile data obtained with this method. Next, the HER2 glycan (G1−G16) profile data in SK-BR-3 and BT-474 were compared to examine whether the glycopatterns differed between different breast cancer cell lines. The profile data from SK-BR-3 had a low correlation with that from BT-474 (R2 = 0.079, p = 0.272, Figure 3). This result

mM Tris, pH 7.5, 150 mM NaCl). To elute HER2 protein, the beads were resuspended in 30 μL of TBS buffer containing 0.2% SDS followed by incubation at 60 °C for 10 min. After magnetic separation, the supernatants were obtained as an HER2 protein fraction. N-Glycan Labeling with 3-AQ. Samples were dissolved in LDS sample buffer containing 50 mM DTT and heated at 70 °C for 10 min. Protein separation was performed with 10% BisTris Gel according to the manufacturer’s protocol. After gel electrophoresis, the gel was stained with CBB followed by gel excision of HER2 protein bands. The excised gel pieces were destained and then incubated with peptide-N-glycosidase F (Sigma-Aldrich, St. Louis, MO) (0.25 units) at 37 °C overnight. After the incubation, N-glycans released from HER2 were extracted with 10 μL of distilled water. The extracted N-glycans were purified with NuTip Carbon (Glygen Corporation, Columbia, MD) and labeled with 3-AQ reagent according to our previous report.18 MS Measurement. All MS and MS/MS spectra were acquired by MALDI-QIT-TOF MS (AXIMA-Resonance; Shimadzu/Kratos, U.K.) in the negative-ion reflectron mode. Data acquisition and processing were controlled by Shimadzu Biotech Launchpad 2.9.1 (Shimadzu/Kratos; Shimadzu/Kratos, U.K.). In this study, the ion peak intensity for each glycan was calculated from the area of the monoisotopic ion (Mascot Distiller Software, version 2.4.2.0, Matrix Science; London, U.K.).

■

RESULTS AND DISCUSSION We prepared HER2 protein by immunoprecipitation from breast-cancer cell lines SK-BR-3 and BT474. These cell lines are known to highly express HER2 protein. N-Glycans released from immunoprecipitated HER2 were subjected to glycosylation pattern analysis using the AQ-labeling method in negativeion mode (Figure 1). Approximately 30 ion peaks of N-glycans were observed in at least one of the cell lines. Among these, 17 peaks (G1−G17) were observed in both cell lines (Figure 1) and confirmed to be AQ-labeled N-glycan with MS/MS analysis (data not shown). While AQ-labeled neutral glycans were detected as phosphate adduct ions [M + 126 + H2PO4]− derived from ammonium dihydrogen phosphate (ADP) in 3AQ/CHCA, sialylated glycans were detected as deprotonated ions [M + 126 − H]−. Using these 16 peaks (G1−G16) and omitting the lowintensity G17, we examined the correlation coefficient of Nglycan profile data in each different measurement to evaluate

Figure 3. Comparison of N-glycosylation pattern of HER2 in SK-BR-3 and BT-474. The ion peak intensities for G1−G16 were individually normalized to the total ion intensity of G1−G16. The averaged peak intensities (n = 4) of the N-glycans in SK-BR-3 (BT-474) are plotted on the horizontal (vertical) line.

suggests that unique N-glycosylation patterns were observed in 16 glycans, reflective of differences in the glycosylation state of HER2 in SK-BR-3 and BT-474. When we investigated glycosylation patterns in other cell lines, the N-glycosylation pattern in each cell line was different (data not shown). These results demonstrate the possibility that this method could be applied for distinguishing between cell lines. 7148

dx.doi.org/10.1021/ac301484f | Anal. Chem. 2012, 84, 7146−7151

Analytical Chemistry

Article

Figure 4. Negative ion mass spectra of N-linked glycans of HER2 in SK-BR-3 and BT-474 using an internal standard glycan. In total, 5 fmol of PAlabeled monosialylated N-glycan was applied as an internal standard (IS). Its molecular weight and ion intensity were comparable to those of G1− G16, and the PA-labeled glycan invariably had a different mass than the AQ-labeled glycan.

Figure 5. Linear dynamic ranges of quantification with the AQ-labeling method. The ion peak intensity for standard glycan was normalized to that for 5 fmol of PA-labeled monosialylated N-glycan (IS). The peak intensities of (a) NA2 ion [M + 126 + H2PO4]−, (b) A1 ion [M − H]−, and (c) A2 ion [M + 126 − H]− are individually plotted against the concentration of standard glycan (0.5, 5, 50, 500, 1000, 2000, 3000, 4000, and 5000 fmol).

corresponding dynamic range for signal intensity in MS measurements.24 Next, N-glycans prepared from a biological sample were analyzed to confirm the quantitative performance. An HER2 Nglycan sample from SK-BR-3 was serially diluted to 1:32 and subjected to glycan analysis. Peak intensities for G1−G16 were normalized by the IS peak intensity. The normalized intensities of G5−G16 were highly linear with their dilution factors (R2 > 0.9, p < 0.05) over the concentration range (Figure 6). In contrast, the intensities of G1−G4 exhibited high linearity (R2 > 0.9, p < 0.05) with the exception of the highest-concentration sample (Figure 6). The concentration of the ADP additive used may affect the quantification linearity range for G1−G4 because neutral glycans are detected as phosphate adduct ions. Therefore, the quantification linearity range for neutral glycan may be extended by optimizing the concentration of the ADP additive. Results comparable to the above were observed in BT-474 (data not shown), suggesting that the relative quantification of glycans using the AQ-labeling method is applicable to complex mixture samples.

Although N-glycan profile patterns can be compared between samples by the present method, each N-glycan expression between samples cannot be directly compared. Therefore, we evaluated the relative quantitative performance of N-glycans with this method by employing an internal standard (IS). PA-labeled monosialylated N-glycan (5 fmol) was used as the IS because it would not have any negative effect on the MS data (e.g., ion suppression) (Figure 4). Neutral glycan (NA2), monosialylated glycan (A1), and disialylated glycan (A2), applied as a standard glycan, are all commercially available glycans. We investigated the linear dynamic range of quantification using the IS for these AQ-labeled standard glycans. The ion peak intensity for the standard glycan was normalized to that for IS. NA2 ion [M + 126 + H2PO4]− and A1 ion [M + 126 − H]− exhibited high linearity (R2 > 0.9) from 0.5 to 5000 fmol (Figure 5a,b). The A2 ion [M + 126 − H]− also exhibited high linearity (R2 > 0.9) over the same range in spite of the dissociation of the sialic acid moiety (Figure 5c). The quantitative capability of this method in combination with IS was demonstrated to be excellent, given that the dynamic range for quantification (1−1 × 104) was comparable to the 7149

dx.doi.org/10.1021/ac301484f | Anal. Chem. 2012, 84, 7146−7151

Analytical Chemistry

Article

Figure 6. Linearity of quantification over several orders of magnitude. The ion peak intensities for G1−G16 were individually normalized by the IS. A dilution series extending to 1:32 of N-glycans released from HER2 in SK-BR-3 was subjected to glycan profiling using the AQ-labeling method. The normalized peak intensities for the N-glycans are plotted against their dilution.

additive as discussed above. These observations together suggest that the AQ-labeling method in combination with ADP enabled simultaneous quantitative profiling of neutral and sialylated glycans without complicated pretreatments such as the removal of excess reagent in AA labeling. This last feature would be particularly advantageous for rapid quantitative analysis with MS. However, it should be noted that sialic acid dissociated to some degree in the negative mode. To suppress this dissociation, we will modify the carboxyl group of sialic acid, such as methyl esterification,26,34,35 methylamidation,36 permethylation,37−39 perbenzoylation,40 and amidation.41

Glycans derived from biological samples are usually comprised of a mixture of neutral and sialylated glycans. In general, neutral glycans are mainly observed in the positive-ion mode, while sialylated glycans are observed in the negative-ion mode. In addition, sialic acid moiety in glycans is dissociated in the ionization process of MALDI.25,26 These facts make it difficult to simultaneously profile neutral and sialylated glycans by MALDI-MS. The dissociation of sialic acid moiety is minimized (reduced) by MALDI-MS analysis in the negative-ion mode.27 Furthermore, negative-ion fragmentation spectra of glycans are more informative than positive-ion spectra.28−32 Considering these facts, we analyzed glycans in negative-ion mode in our study. However, neutral glycans need to be negatively charged for highly sensitive detection in negative-ion mode. Anumula K. R. et al. reported that glycan labeling with 2amonobenzoic acid (AA) carries one negative charge, enabling glycan profiling of neutral glycans in addition to sialylated glycans.33 In our study, we quantitatively detected both neutral and sialylated glycans in negative-ion mode by using ADP as an

■

CONCLUSIONS We demonstrated that the AQ-labeling method enables highly reproducible N-glycan profiles and is applicable to biological samples. In addition, using an internal standard enabled the relative quantification of N-glycans, including sialylated glycans without derivatization, in different samples with a wide dynamic range. This method would apply to various N- and O-glycans 7150

dx.doi.org/10.1021/ac301484f | Anal. Chem. 2012, 84, 7146−7151

Analytical Chemistry

Article

(22) Slamon, D. J.; Godolphin, W.; Jones, L. A.; Holt, J. A.; Wong, S. G.; Keith, D. E.; Levin, W. J.; Stuart, S. G.; Udove, J.; Ullrich, A.; et al. Science 1989, 244, 707. (23) Slamon, D. J.; Clark, G. M.; Wong, S. G.; Levin, W. J.; Ullrich, A.; McGuire, W. L. Science 1987, 235, 177. (24) Makarov, A.; Denisov, E.; Lange, O.; Horning, S. J. Am. Soc. Mass Spectrom. 2006, 17, 977. (25) Talbo, G.; Mann, M. Rapid Commun. Mass Spectrom. 1996, 10, 100. (26) Powell, A. K.; Harvey, D. J. Rapid Commun. Mass Spectrom. 1996, 10, 1027. (27) Thaysen-Andersen, M.; Thogersen, I. B.; Nielsen, H. J.; Lademann, U.; Brunner, N.; Enghild, J. J.; Hojrup, P. Mol. Cell. Proteomics 2007, 6, 638. (28) Harvey, D. J. J. Am. Soc. Mass Spectrom. 2005, 16, 622. (29) Harvey, D. J. J. Am. Soc. Mass Spectrom. 2005, 16, 631. (30) Harvey, D. J. J. Am. Soc. Mass Spectrom. 2005, 16, 647. (31) Harvey, D. J.; Jaeken, J.; Butler, M.; Armitage, A. J.; Rudd, P. M.; Dwek, R. A. J. Mass Spectrom. 2010, 45, 528. (32) Domann, P.; Spencer, D. I.; Harvey, D. J. Rapid Commun. Mass Spectrom. 2012, 26, 469. (33) Ruhaak, L. R.; Huhn, C.; Waterreus, W. J.; de Boer, A. R.; Neususs, C.; Hokke, C. H.; Deelder, A. M.; Wuhrer, M. Anal. Chem. 2008, 80, 6119. (34) Liu, X.; Li, X.; Chan, K.; Zou, W.; Pribil, P.; Li, X. F.; Sawyer, M. B.; Li, J. Anal. Chem. 2007, 79, 3894. (35) Miura, Y.; Shinohara, Y.; Furukawa, J.; Nagahori, N.; Nishimura, S. Chemistry 2007, 13, 4797. (36) Liu, X.; Qiu, H.; Lee, R. K.; Chen, W.; Li, J. Anal. Chem. 2010, 82, 8300. (37) Ciucanu, I.; Costello, C. E. J. Am. Chem. Soc. 2003, 125, 16213. (38) Kang, P.; Mechref, Y.; Klouckova, I.; Novotny, M. V. Rapid Commun. Mass Spectrom. 2005, 19, 3421. (39) Liu, X.; McNally, D. J.; Nothaft, H.; Szymanski, C. M.; Brisson, J. R.; Li, J. Anal. Chem. 2006, 78, 6081. (40) Chen, P.; Werner-Zwanziger, U.; Wiesler, D.; Pagel, M.; Novotny, M. V. Anal. Chem. 1999, 71, 4969. (41) Sekiya, S.; Wada, Y.; Tanaka, K. Anal. Chem. 2005, 77, 4962.

and rapidly determines alterations in glycosylation pattern or glycan structures on glycoconjugates with high sensitivity, facilitating novel biomarker discovery for disease. In the future, the rapid and highly reproducible glycosylation pattern analysis made possible by this method will contribute to the field of biomedicine in areas such as quality control in cell cultures and biotechnology-based drugs.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: +81 (0) 75 8232897. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Dr. Sekiya and Dr. Nishikaze for helpful comments and discussion. 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) Opdenakker, G.; Rudd, P. M.; Ponting, C. P.; Dwek, R. A. FASEB J. 1993, 7, 1330. (2) Apweiler, R.; Hermjakob, H.; Sharon, N. Biochim. Biophys. Acta 1999, 1473, 4. (3) Hollingsworth, M. A.; Swanson, B. J. Nat. Rev. Cancer 2004, 4, 45. (4) Schroeder, H. W., Jr.; Cavacini, L. J. Allergy Clin. Immunol. 2010, 125, S41. (5) Rudd, P. M.; Elliott, T.; Cresswell, P.; Wilson, I. A.; Dwek, R. A. Science 2001, 291, 2370. (6) Abd Hamid, U. M.; Royle, L.; Saldova, R.; Radcliffe, C. M.; Harvey, D. J.; Storr, S. J.; Pardo, M.; Antrobus, R.; Chapman, C. J.; Zitzmann, N.; Robertson, J. F.; Dwek, R. A.; Rudd, P. M. Glycobiology 2008, 18, 1105. (7) Saldova, R.; Royle, L.; Radcliffe, C. M.; Abd Hamid, U. M.; Evans, R.; Arnold, J. N.; Banks, R. E.; Hutson, R.; Harvey, D. J.; Antrobus, R.; Petrescu, S. M.; Dwek, R. A.; Rudd, P. M. Glycobiology 2007, 17, 1344. (8) Kuno, A.; Uchiyama, N.; Koseki-Kuno, S.; Ebe, Y.; Takashima, S.; Yamada, M.; Hirabayashi, J. Nat. Methods 2005, 2, 851. (9) Kuno, A.; Kato, Y.; Matsuda, A.; Kaneko, M. K.; Ito, H.; Amano, K.; Chiba, Y.; Narimatsu, H.; Hirabayashi, J. Mol. Cell. Proteomics 2009, 8, 99. (10) Li, Y.; Tao, S. C.; Zhu, H.; Schneck, J. P. Current Protocols in Protein Science; John Wiley and Sons, Inc.: New York, 2011; Chapter 12, Unit 12 9. (11) Patwa, T.; Li, C.; Simeone, D. M.; Lubman, D. M. Mass Spectrom. Rev. 2010, 29, 830. (12) Zaia, J. Chem. Biol. 2008, 15, 881. (13) Zaia, J. OMICS 2010, 14, 401. (14) Kuraya, N.; Hase, S. Anal. Biochem. 1996, 233, 205. (15) Harvey, D. J. J. Am. Soc. Mass Spectrom. 2000, 11, 900. (16) Anumula, K. R. Anal. Biochem. 2006, 350, 1. (17) Naven, T. J.; Harvey, D. J. Rapid Commun. Mass Spectrom. 1996, 10, 1361. (18) Kaneshiro, K.; Fukuyama, Y.; Iwamoto, S.; Sekiya, S.; Tanaka, K. Anal. Chem. 2011, 83, 3663. (19) Li, Y. L.; Gross, M. L. J. Am. Soc. Mass Spectrom. 2004, 15, 1833. (20) Tholey, A.; Zabet-Moghaddam, M.; Heinzle, E. Anal. Chem. 2006, 78, 291. (21) Gusterson, B. A.; Gelber, R. D.; Goldhirsch, A.; Price, K. N.; Save-Soderborgh, J.; Anbazhagan, R.; Styles, J.; Rudenstam, C. M.; Golouh, R.; Reed, R.; et al. J. Clin. Oncol. 1992, 10, 1049. 7151

dx.doi.org/10.1021/ac301484f | Anal. Chem. 2012, 84, 7146−7151