Metallic Element Chelated Tag Labeling (MeCTL) for Quantitation of N

Jun 21, 2017 - Metallic Element Chelated Tag Labeling (MeCTL) for Quantitation of N-Glycans in MALDI-MS ... E-mail:[email protected]., *Phone: +86 21 ...
1 downloads 8 Views 2MB Size
Download PDF
Subscriber access provided by EAST TENNESSEE STATE UNIV

Article

Metallic Element Chelated Tag Labeling (MeCTL) for Quantitation of N-Glycans in MALDI-MS Lijun Yang, Ye Peng, Jing Jiao, Tao Tao, Jun Yao, Ying Zhang, and Haojie Lu Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.7b01051 • Publication Date (Web): 21 Jun 2017 Downloaded from http://pubs.acs.org on June 23, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 7

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Metallic Element Chelated Tag Labeling (MeCTL) for Quantitation of N-Glycans in MALDI-MS Lijun Yangab, Ye Pengb, Jing Jiaoa, Tao Taob, Jun Yaob, Ying Zhang*ab and Haojie Lu*ab a

Shanghai Cancer Center and Department of Chemistry, Fudan University, Shanghai, 200032, P. R. China Institutes of Biomedical Sciences and Key Laboratory of Glycoconjugates Research Ministry of Public Health, Fudan University, Shanghai 200032, P. R. China b

ABSTRACT: N-glycosylation plays an important role in chief biological and pathological processes. Quantifying the N-glycan is important since glycan alterations are related to many diseases. In this study, we developed a novel N-glycan quantitation approach using metallic element chelated tag labeling (MeCTL) through reductive amination. The MeCTL strategy is of high labeling efficiency and accurate in quantitation with high reproducibility (CV0.99) within two orders of magnitude of dynamic range. Additionally, it provides significant cross-ring fragmentation to distinguish N-glycan isomers. Furthermore, multiplex quantitation by chelation with several different rare earth elements can be achieved. At last, this strategy has been successfully used for evaluation of N-glycan changes in human serum associated with CRC, indicating its potential in clinical applications including disease N-glycome profiling and relative quantitation. Glycosylation is one of the most widespread and complicated post-translational modifications (PTMs) of proteins. It plays an important role in biological activities, such as protein folding, cell signaling and communication, as well as cell matrix interactions.1-3 Alterations in glycosylation are associated with many diseases including immunological disorders, dementia and cancer.4,5 Several studies have shown that N-glycans could be used as cancer serum biomarkers.6 Therefore, the elucidation of their structures and the further quantitative analysis of N-glycan can lead us to a deeper understanding of glycan’s functions in biological processes.7 Mass spectrometry (MS) has become a powerful tool for qualitative and quantitative profiling of the N-glycome.8 Various methods based on MS have been developed to obtain quantitative information of N-glycans, which could be loosely divided into metabolic labeling,9 enzymatic 18O-labeling10 and chemical labeling.11-27 A high degree of accurate quantitation can be achieved using metabolic labeling. Since the introduction of isotopes can be achieved during the culturing of cells, variations during the preparation of samples are minimized. However, this method is limited to the investigation of cultured cell systems. 18 O-labeling is an alternative approach for N-glycan quantitation, in which 18O is incorporated into the reducing end of glycans when N-glycans are enzymatically released from glycoproteins in the presence of H218O. While simple, it only creates a 2 Da mass difference between 16O- and 18O-labeled N-glycans, which requires an extra deconvolution step to achieve accurate quantitation. Among these three types of approaches, chemical labeling is the most popular, during which isotope-coded tags are introduced through a permethylation11-13 reaction at the hydroxyl group on the glycans or through other chemical reactions, including reductive amination14-23 and by formation of hydrazone24,25 or oxime26,27 at the reducing end of the N-glycans. It is an simple and straightforward way and several isotope-coded compounds have been developed, including aniline,14 2-aminobenzoic acid (2-AA),15-17 2-aminopyridine (2-AP),18 arginine (Arg),19 p-toluidine,20 Girard’s reagent25 and other isobaric tags, such as glyco-TMTs,21 iARTs,22 QUANTITY,23 aminoxyTMT.26,27 All these methods were based on a mass shift generated by stable isotopes like 2D, 13C, 15N and

18 O. Recently, a pseudo-isotope labeling based quantitation approach via metallic element chelated tags have been applied in quantitative proteomics. Specifically, rare earth metals’ chelated macrocyclic 1, 4, 7, 10-tetraazacyclododecane-N, N’, N’’, N’’’-tetraacetic acid (DOTA)28-30 or bicyclic anhydride diethylenetriamine-N, N, N’, N’’, N’’-pentaacetic acid (DTPA) 31 were used to introduce mass shift between different samples. Many rare earth elements are quasi-monoisotopic or naturally monoisotopic. The stability constants of complexes formed between DOTA32 and most of the rare earth elements are similar and high. In addition, these complexes have nearly the same ionization efficiency in MS. Despite these metallic element chelated tags has been well understood in quantitative proteomics, there is no knowledge in quantitative glycomics. Herein, a new method based on metallic element chelated tag for quantitation of N-linked glycans was developed. S-2-(4-Aminobenzyl)-1, 4, 7, 10-tetraazacyclododecane tetraacetic acid (p-NH2-Bn-DOTA) was covalently coupled to the reducing end of N-glycan through reductive amination, followed by a chelating reaction with the rare earth metal Lu/Ho salts, producing a 10 Da mass difference. The MeCTL strategy offers the following advantages for quantitative analysis of N-glycans by MS. First, compared with some self-development reagents, p-NH2-Bn-DOTA and lanthanide salts are commercially available reagents. Second, the labeling efficiency of the reactions was high (almost 100%), minimizing the potential sample loss due to inefficient derivatization. Third, many rare earth elements are naturally monoisotopic or quasi monoisotopic, so multiple labeling, with a mass shift from 2 to 86 Da, could be obtained. Fourth, metal chelated glycans can be detected in the negative ion mode, providing a clean background and diverse MS/MS fragments. Other than a series of Y ions, A ions were created via cross-ring cleavages in the negative ion reflection mode, which could help researchers to distinguish glycan isomers. Moreover, the metal chelated tag can improve the ionization efficiency of most glycans, leading to more sensitive detection and accurate quantitation.

MATERIALS AND EXPERIMENTAL PROCEDURE

ACS Paragon Plus Environment

1

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Materials and Chemicals. S-2-(4-Aminobenzyl)-1, 4, 7, 10-tetraazacyclododecane tetraacetic acid (p-NH2-Bn-DOTA) was purchased from Macrocyclics (Dallas, TX), Maltoheptaose (DP7, 95%) was purchased from Hayashibara Biochemical Laboratories (Okayama, Japan). 2,5-Dihydroxybenzoic acid (DHB), 2', 4', 6'-Trihydroxyacetophenone monohydrate (THAP), trifluoroacetic acid (TFA), Ammonium citrate, ammonium bicarbonate (ABC), sodium cyanoborohydride (NaBH3CN), asialofetuin from fetal calf serum (ASF), albumin from chicken egg white (OVA), fetuin from fetal bovine serum and ribonuclease B from bovine pancreas (RNase B), Lutetium (III) chloride (LuCl3), Holmium (III) chloride (HoCl3), Thulium (III) chloride (TmCl3), Terbium (III) chloride (TbCl3) were purchased from Sigma Aldrich (St. Louis, MO, USA). Peptide N-glycosidase (PNGase F, 500 U/µL) was purchased from New England Biolabs (Ipswich, MA, USA). Centrifugal filters with MWCO of 3 kDa and 10 kDa were purchased from Millipore (Bedford, MA, USA). HyperSep Hypercarb SPE was purchased from Thermo Fisher Scientific (CA, USA). HPLC-grade acetonitrile (ACN) and methanol were purchased from Merck (Darmstadt, Germany). Analytical grade acetic acid (HAC) was purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Distilled water was purified by a Milli-Q system (Milford, MA, USA). Sample preparation. To obtain the N-glycans from model glycoproteins, RNase B, ASF, OVA and fetuin were dissolved in 25 mM ABC buffer (pH 8.0) with a concentration of 2 mg mL-1. After denaturation, PNGase F solution was added and then the solution was incubated at 37 °C overnight and then stored at 4 °C for further use. Human serum samples with three colorectal cancer (CRC) patients with one female (stage 2, age 58) and two male (stage 1, age 46, 64 respectively) and three healthy control were obtained from Fudan University Shanghai Cancer Center. The research was handled in accordance with ethical and legal standards. Releasing N-glycans from human serum was processed according to the procedure reported previously with minor modifications.19 Briefly, the pooled serum was firstly quantified by Bradford method and then equal amounts of serum (containing 600 µg protein) was reduced in 10 mM DTT followed by alkylation in 20 mM IAA. After that, low molecular compounds were removed by ultrafiltration (MWCO, 3 kDa). The collected proteins were redissolved in 25 mM ABC and 1 µL PNGase F was added. The sample was incubated at 37 °C overnight. The released N-glycans were collected by ultrafiltration (MWCO, 10 kDa) and further purified using HyperSep Hypercarb SPE. The neutral N-glycans were eluted by 2 mL H2O/ACN (70/30, v/v), and the sialylated N-glycans were eluted by 2 mL H2O/ACN (70/30, v/v, containing 0.05% TFA). The collected N-glycans were then lyophilized and redissolved in ultrapure water for further analysis. The p-NH2-Bn-DOTA derivatization and metal chelated. DP7 was dissolved at a concentration of 10 mg mL-1 in distilled water as a storage solution. The p-NH2-Bn-DOTA were dissolved at a concentration of 5.8 mg mL-1 in methanol. Before labeling, DP7 and p-NH2-Bn-DOTA were lyophilized (molar ratio of DP7 to p-NH2-Bn-DOTA was 1:5) and redissolved in 25 µL of methanol acetic acid (85/15, v/v) and incubated at 37 °C for 1 h. After that, NaBH3CN was added at a final concentration of 5 mg mL-1. Then the sample solution was incubated at 37 °C for 2 h. For derivatizing sialylated N-glycans, methylamidation of sialic acids were performed before reductive amination to avoid loss of sialic acid, and the reaction conditions are consistent with previously reported. 15 After labeled with DOTA, the solution was lyophilized and redissolved with 10

Page 2 of 7

mM NH4 Ac buffer (pH 5.6) containing with 10-fold molar excess amount of either HoCl3 or LuCl3, and then incubated at 37 °C for 1 h. At last, the labeled glycans were diluted 10 times with ultrapure water and then subjected to MALDI-TOF MS analysis. MALDI Sample Preparation and Analysis. DHB matrix was dissolved in H2O/ACN (50/50, v/v, containing 0.1% TFA) at a concentration of 10 mg mL-1. One µL of sample solution and one µL of matrix were spotted on the target plate for MALDI-TOF MS analysis. MALDI-TOF MS and MS/MS experiments were performed in positive or negative ion reflection mode on a 5800 Proteomics Analyzer (Applied Biosystems, Framingham, MA, USA) with a Nd: YAG laser (355 nm), an acceleration voltage of 20 kV and a repetition rate of 400 Hz. Besides, MS/MS spectra were interpreted manually with the assistance of the GlycoWorkbench software.

Results and Discussion Optimization of the derivatization conditions. The MeCTL strategy, which includes two derivatization processes including reductive amination and metal chelation, is illustrated in Scheme 1. Therefore, we first investigated the preferred reaction conditions. DP7 was used as a model glycan to optimize the reaction conditions. From several preliminary experiments, we found that the concentration of NaBH3CN was important for reductive amination (Figure 1a). When the concentration of NaBH3CN was increased to 5 mg mL-1, the reaction efficiency increased to almost 100%. For the metal chelating reaction, when the concentration of NH4Ac reached 10 mM, high labeling efficiency (>95%) for both metal (Lu and Ho) salts could be achieved (Figure 1b and 1c). Therefore, the optimized condition for reductive amination was with 5 mg mL-1 NaBH3CN at 37 °C for 2 h, and for the metal chelating reaction, it was in 10 mM NH4Ac at 37 °C for 1 h with the molar ratio of DOTA to Lu or Ho salts at 1:10. As shown in Figure 2a, native DP7 was detected as [M+Na]+ and [M+K]+ at m/z 1175.28 and 1191.25, respectively. After labeled with DOTA, [M+DOTA+H] + , [M+DOTA+Na]+ and [M+DOTA+K] + were detected with the predicted mass shift. No signals of native DP7 were observed, indicating the nearly 100% labeling efficiency of the reductive amination reaction between DP7 and DOTA. After chelating with metal, we found that metal labeled DP7 showed better ionization in the negative ion reflection mode than in positive ion reflection mode with a higher S/N ratio and a cleaner background (Figure S1). The MALDI-TOF mass spectra of [M+DOTA+Lu]- and [M+DOTA+Ho]- were shown in Figure 2c and 2d, with a 10 Da mass difference between Ho and Lu adducts. Thanks to a permanent negative charge at the reducing end of glycan, resulting in only [M]- ion, the ambiguities caused by metal adduct peaks such as [M+K] - and [M + Na]- were avoided. And none of DOTA labeled DP7 signals were detected, revealing the high efficiency of metal chelation (Figure S1). Scheme 1. The MeCTL strategy for N-glycan relative quantitation.

ACS Paragon Plus Environment

2

Page 3 of 7

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Figure1. Optimization of the reaction conditions for (a) reductive amination and (b, c) metal chelating reaction. Derivatization yield was calculated by dividing the peak intensity of derivatized products by the total peak intensities of substrates and derivatized products. Equal amounts of DP7 and [DOTA+Lu] labeled DP7 were mix to determine the influence of the metal chelated tag on glycan ionization efficiency. As shown in Figure S1, native DP7 was detected in the positive ion reflection mode and [DOTA+Lu] labeled DP7 was detected in the negative ion reflection mode. [DOTA+Lu] labeled DP7 showed better S/N and signal intensity increased over five-fold over that of DP7. Therefore, metal labeled glycans were analyzed in the negative ion mode in the following experiments.

Establishment and validation of the MeCTL strategy.

Figure 2. MALDI-TOF mass spectra of (a) native, (b) DOTA labeled, (c) [DOTA+Ho] labeled and (d) [DOTA+Lu] labeled DP7. (a, b) were obtained in positive ion reflection mode and (c, d) were obtained in negative ion reflection mode.

Figure 3. MALDI-TOF mass spectra of (a) [DOTA+Ho] and [DOTA+Lu] labeled DP7 with a molar ratio of 1:1 in negative ion reflection mode, (b) Dynamic range and accuracy of the MeCTL strategy (n=3).

DP7 was also used to assess the stability and reproducibility of MeCTL strategy. First, we investigated the stability of metal chelated DOTA. Ten-fold (molar ratio) of LuCl3 solution was added to [DOTA+Ho] derivatized DP7 solution and incubated at 37 °C for 24 h. No metal exchange was observed in MALDI-MS, suggesting the metal chelated complex was stable in the presence of another rare earth metal ion (Figure S2). The accuracy, reproducibility and dynamic range of the MeCTL strategy for quantitation were then evaluated. We calculated relative abundance (measured ratios) by comparison of signal intensity of the monoisotopic peak of Ho or Lu labeled DP7. Figure 3a shows the mass spectra of the mixture of [DOTA+Ho] and [DOTA+Lu] labeled DP7 with a molar ratio of 1:1. A mass shift of 10 Da between [DOTA+Ho] and [DOTA+Lu] labeled DP7 were observed. The measured ratios were consistent with the theoretical ratios. We also investigated other different mixture ratios of Ho/Lu labeled DP7 (1:10, 1:5, 1:2, 2:1, 5:1, and 10:1). The dual-logarithm plots of the theoretical ratios vs. the measured ratios exhibited a good coefficient of variation (CV) of less than 5 % and a high correlation coefficient (R2 = 0.9995) within two orders of magnitude in three repeated experiments (Figure 3b). The ratio of [DP7+DOTA+Ho]/[DP7+DOTA+Lu] didn’t change, even after being stored at -20 °C for six months, indicating good stability of the metal labeled complex, as shown in Figure S3. We also investigated the sensitivity of MeCTL strategy. We found that when the loading amount of [DOTA+Ho] labeled DP7 was 8.68 fmol, it still could be well detected and also provide an accurate quantitative result (Figure S4). We then investigated the tandem mass (MS/MS) behavior of metal labeled DP7. Tandem mass spectrometry is demonstrated to be a useful tool in structural characterization. The metallic element labeled glycan is better ionized in negative ion reflection mode, which can provide more fragment information.33,34 The MALDI-TOF MS/MS spectra of native DP7 and metal labeled DP7 are shown in Figure 4. Fragmentation of native DP7 (precursor ion [DP7+Na]+) only

ACS Paragon Plus Environment

3

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 7

Figure 4. MALDI-TOF/TOF tandem mass spectra of precursor ion (a) [DP7+Na]+ (m/z 1175.21) in positive ion mode, (b) [DP7+DOTA+Ho]- (m/z 1806.52), and (c) [DP7+DOTA+Lu](m/z 1816.54) in negative ion mode. Figure 6. MALDI-TOF mass spectra of an equimolar mixture of [DOTA+Ho] and [DOTA+Lu] labeled N-glycans from (a) RNase B, (b) ASF, (c) OVA and (d) fetuin. “#” denotes the dehydrated metal chelated complex ([M-2H2O]-).

Figure 5. MALDI-TOF/TOF tandem mass spectra of [DOTA+Lu] labeled isomeric N-glycans (a) hybrid-type N-glycan from OVA and (b) complex-type N-glycan from ASF after derivatization with the m/z at 2304.75. provided a series of B/Y ions, formed by cleavages in glycosidic bonds, which gave limited information about the composition and sequence of the glycan moiety. After derivatization with the metallic element chelated tags, a series of Y ions were predominately observed which were conveniently used for deducing the glycan sequence. The prominence of Y ions is due to the negative charge localized at the reducing terminus of the labeled glycan. Additionally, more fragment ions, mainly produced during cross-ring cleavages, were obtained from which the linkage and branching information of the glycan could be deduced. This information is valuable for in-depth structural characterization of the glycan. To further demonstrate the ability of metallic derivatization in the structural determination of the isomeric N-glycan, a pair of isomeric N-glycans with the m/z at 1663.58 were both derivatized with [DOTA+Lu] and analyzed. The N-glycan from OVA was hybrid-type N-glycan and the other from ASF was

complex-type N-glycan. The MS/MS spectrum of the corresponding precursor ions at m/z 2304.75 were shown in Figure 5. The Y3γ/Y4β ion at 2011.56 m/z in Figure 5a indicated the presence of hybridized N-glycan from OVA. The 1,3A3 ion at m/z 424.15 in Figure 5b could be used to confirm the Hex-GlcNAc composition of the antennae. Additionally, the Y3 ion at m/z 1777.42 only presented in Figure 5b indicated the presence of complex N-glycan from ASF. These results demonstrated metal derivatization could facilitate rapid structural assignments and discrimination of isomeric N-glycan. In addition, we also investigated the chromatographic behavior of different rare metal labeled glycan. The chromatographic behavior of [DOTA+Ho] and [DOTA+Lu] labeled DP7 were compared. As shown in Figure S5, [DOTA+Ho] and [DOTA+Lu] labeled glycans were eluted at the same retention time in the HILIC-MS analysis, indicating no chromatographic isotope effects. The chromatographic conditions are shown in Table S1. Finally, we tried to label the glycan with other different rare earth metals to achieve multiplex quantitation. Since many other rare earth elements have been used for quantitation, we extended the MeCTL strategy to Tb, Ho, Tm, Lu, and Figure S6 shows different molar ratios (1:1:1:1, 1:2:2:4, 4:2:2:1, 1:2:5:10 and 10:5:2:1) within two orders of magnitude of these four tags. All of the measured ratios were close to the theoretical ratios.

Quantitation of N-glycans released from glycoproteins. In order to further validate the MeCTL strategy, a high-mannose type N-glycan from RNB, two complex type N-glycans from ASF, a hybrid type N-glycan and a bisected hybrid N-glycan from OVA and a sialylated N-glycan from fetuin were derivatized with Ho/Lu chelated DOTA. Figure 6 shows the mass spectra of the equimolar mixture of N-glycans

ACS Paragon Plus Environment

4

Page 5 of 7

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Figure 7. Dynamic range and accuracy of quantitation of the N-glycans from model proteins (n = 3). (a) from RNB, (b, c) from ASF, (d, e) from OVA and (f) from fetuin.

Figure 8. Relative quantitation of N-glycans from CRC and normal human serum. mentioned above labeled with [DOTA+Ho] and [DOTA+Lu]. Dehydration products of labelled N-glycan from RNB could be observed in the spectra. But the extent of dehydration is estimated to be less than 20%, and this situation happened in both Lu and Ho labeled N-glycans. We found there was no significant influence on the accuracy of quantitative results. All the quantitation results we used was obtain with the non-dehydrated ones. Since we found that the side products also provided a good relationship which is consistent with the proportion of non-dehydrated ones. The dual logarithm plots of the theoretical ratios vs the measured ratios of each N-glycan were displayed in Figure 7. The results exhibit good linearity (R2>0.99) within two orders of magnitude (1:10 to 10:1) (Table S2). The CVs (n=3) varied from 1.24% to 17.03% (Table S2), indicating high quantitation accuracy and reproducibility of this strategy. It was worth mentioning that, sialic acids containing N-glycans were methylamidated before reductive amination to prevent the loss of sialic acids. The methylamidation was nearly complete and the relative abundance of sialylated N-glycan from fetuin remained similar after methylamidation (Figure S7). Thanks to methylamidation protection, the quantitation reproducibility of sialylated N-glycan was good (CV=14.69%, n=3) without significant loss of sialic acid. In addition, we try to use the unique fragment ions of isomeric N-glycans in MS/MS spectra to relative quantitation of isomeric glycans from different glycoproteins. In order to confirm the feasibility of this method, relative quantitation between MS and MS/MS were compared. The quantitation based on MS were obtained by comparing the relative intensity of [DOTA+Ho] labeled N-glycans from ASF and [DOTA+Lu] labeled isomeric

N-glycans from OVA with the m/z at 1663.58 before derivatization. And the quantitation based on MS/MS were obtained by comparing the relative intensity of 1,3A3 and 1,3A2 fragment from [DOTA+Lu] labeled isomeric N-glycans from ASF and OVA, respectively. 1,3A3 and 1,3A2 fragment were chosen due to the same kind of debris. As shown in Figure S8, the quantitation based on MS were consistent with MS/MS. Therefore, it is feasible to achieve accurate quantitation by taking advantage of high quality MS/MS spectra of metal ion labeled glycans. Human serum samples in N-glycan analysis. Colorectal cancer (CRC) is one of the most commonly diagnosed cancers in both women and men, with over one million new cases occurring every year.35 The change in N-glycans is considered relative to CRC progression and may be used as a new biomarker. To verify the quantitation ability of MeCTL strategy in complex biological samples, N-glycans from healthy human serum and CRC patient serum were labeled with different metallic element chelated tags. Before labeling, equimolar native DP7 was spiked into both samples as an inner standard to correct the ratio. Besides, Equimolar Ho and Lu labeled N-glycans from the same healthy human serum were mixed and analyzed to set the criterion for unambiguous changes of the N-glycan. The normal-Ho/normal-Lu ratios of N-glycans were between 0.8 and 1.2, so CRC/normal ratio 1.2 were regarded as downregulation or upregulation. As shown in Figure S9, a total of 30 N-glycans were detected in both normal and CRC serum in all three replicates. The glycoforms, CRC/normal ratios and CVs (n=3) are shown in Table S3. The quantitative results show good reproducibility with the CVs varying from 0.25% to 19.31%. As shown in Figure 8, it could be inferred that 12 glycans were upregulated in the CRC patient serum. The abundance of glycans with bisecting structures, core fucosylation and sialic acid changed significantly in CRC serum. The results are in accordance with previous reports.19, 36 Therefore, the MeCTL strategy is appropriate for quantitative analysis of complex biological samples.

Conclusion A novel N-glycan relative quantitation method (MeCTL) was developed utilizing the metallic element chelated tags. This method provides remarkable advantages in qualitative and quantitative analysis of N-glycans. For qualitative analysis of N-glycans, the metallic element chelated glycans show high ionization efficiency in the negative ion mode of MALDI-MS. Derivatized glycans could provide various types of fragments (Y ion and A ions) during MS/MS analysis, which allows in-depth structural elucidation of N-glycans. For relative quantitation of N-glycans, the MeCTL strategy shows good linearity (R2 > 0.99) and high reproducibility (CV