Simultaneous analysis of sulfated and phosphorylated glycans by

Subscriber access provided by Kaohsiung Medical University

Article

Simultaneous analysis of sulfated and phosphorylated glycans by serotoninimmobilized-column enrichment and hydrophilic-interaction chromatography Keita Yamada, Haruna Kayahara, Mitsuhiro Kinoshita, and Shigeo Suzuki Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.8b00714 • Publication Date (Web): 10 Jun 2018 Downloaded from http://pubs.acs.org on June 10, 2018

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 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 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.

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 31 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

Simultaneous analysis of sulfated and phosphorylated glycans by serotonin-immobilized-column enrichment and hydrophilic-interaction chromatography Keita Yamada1*, Haruna Kayahara2, Mitsuhiro Kinoshita2, Shigeo Suzuki2. 1. The Laboratory of Toxicology, Faculty of Pharmacy, Osaka Ohtani University, 3-11-1 Nishikiori-kita, Tondabayashi, Osaka 584-8540, Japan, 2. Faculty of Pharmacy, Kindai University, 3-4-1 Kowakae, Higashi-osaka, Higashi-Osaka 577-8502, Japan

E-mail: [email protected] Tel: +81-721-24-9987 Fax: +81-721-24-9890

*To whom correspondence should be addressed.

1 ACS Paragon Plus Environment

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

Abstract Changes in the structures and quantities of sulfated glycans play important roles in inflammatory and neurological diseases, and canceration. Therefore, sulfated glycans are expected to become diagnostic markers for a variety of diseases, such as Alzheimer's disease and cancer. On the other hand, structural abnormalities in the phosphorylated glycans on lysosomal enzymes cause a number of lysosomal diseases, while novel phosphorylated glycans have been found in other proteins. As with sulfated glycans, structural and quantitative changes in these phosphorylated glycans, and their associations with disease, are also of interest. In this article, we introduce a new method for the simultaneous analysis of sulfated and phosphorylated glycans. We first employ a serotonin-immobilized column to enrich these glycans. Glycans obtained in this manner were sequentially subjected to other analytical techniques without desalting. We employed hydrophilic-interaction chromatography to distinguish the sulfate and phosphate groups of the glycans and were able to analyze sulfated and phosphorylated N-glycans expressed on thyroglobulin, ovalbumin, and myozyme. We showed that our method not only analyzes sulfated and phosphorylated glycans, but also glycans containing the GlcNAc-HPO3 residue. We comparatively analyzed sulfated glycans, phosphorylated glycans, and GlcNAc-HPO3-residue-containing glycans expressed on MKN7 cells (well-differentiated gastric-cancer cells) and MKN45 cells (poorly differentiated gastric-cancer cells). To the best of our knowledge, this is the first report of a method for the simultaneous and quantitative analysis of sulfated and phosphorylated glycans.


Page 2 of 31

Page 3 of 31 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


Introduction Glycosylation is one of the most frequently observed post-translational modifications of proteins. In addition, glycans are often modified with sulfate or phosphate groups, and various research groups have reported structural and functional analyses of sulfated glycans. For instance, Fiete et al. reported that sulfated N-glycans attached to luteinizing hormone and thyroid-stimulating hormone control the clearance of

these

hormones.1

Kannagi

et

al.

revealed

that

6-sulfated

Lewis

X

(Galβ1-4(Fucα1-3)(SO3-6)GlcNAcβ1-R) residues are selectin ligands used during the routine site-specific homing of various subsets of T cells.2 In addition, sulfated glycans are associated with disease states such as cancer3,4 and osteoarthritis.5,6 Thus, sulfated glycans are of growing importance in terms of biomarker discovery. Although there are many fewer functional analyses of phosphorylated glycans compared to sulfated glycans, phosphorylated N-glycans have been observed to be expressed on soluble lysosomal proteins and assist in their transportation through Man-6-P receptors (MPRs)7. Recently phosphorylated O-glycans were found in HEK-293 cells8, and novel phosphorylated glycans including ribitol-HPO3 and glycerol-HPO3 residues were identified in α-dystroglycan.9 In addition, phosphorylated glycans containing

GlcNAc-HPO3

residues have also been reported.10 However, little is known about the functions of these phosphorylated glycans and further studies are required. The analyses of sulfated and phosphorylated glycans remain challenging. Since these glycans are minor components of glycome, they are difficult to detect by conventional methods; therefore an efficient enrichment method is required for their analysis. Although lectin-affinity chromatography is a powerful method for the enrichment of characteristic glycans,11 no lectins capable of comprehensively collecting 3 ACS Paragon Plus Environment


phosphorylated and sulfated glycans have been discovered. Some enrichment methods based on anion-exchange chromatography and solvent-distribution have been reported for sulfated glycans. In these methods, sulfated glycans are purified after chemically neutralizing the negative charges of the carboxylic acids on the glycans and peptides. These methods facilitate the analyses of sialic-acid-bearing sulfated glycans in their native states.12-15 However, these methods are labor-intensive because desalting is required after enrichment. Furthermore, these methods are not used for phosphorylated glycans and only target sulfated glycans. On the other hand, effective enrichment methods for phosphorylated glycans have, to the best of our knowledge, not been reported. The chelating properties of some metals, such as Mn, Zn and Ti, toward the phosphate group have often been applied to the enrichment of phosphorylated peptides.16 Even using this method, desalting is required. In addition, these metals do not form stable chelates with phosphate groups modified with GlcNAc, ribitol, or glycerol. Therefore, the applications of these methods to the comprehensive analyses of phosphorylated glycans is difficult. Distinguishing sulfated and phosphorylated glycans during their analyses presents another difficulty, as sulfated and phosphorylated glycans have very similar physicochemical properties. In particular, sulfated and phosphorylated groups on glycans have almost identical molecular weights (HSO3: Mw = 79.9568, H2PO3: Mw = 79.9663); consequently, they are not easily distinguished by mass spectrometry. While high-resolution mass spectrometry (e.g., Orbitrap) techniques are able to distinguish known sulfated and phosphorylated glycan structures,17 they struggle to distinguish unknown glycans derived from biological samples. Although methanolysis is sometimes used to determine sulfate groups, it is difficult to apply this technique to trace amounts 4 ACS Paragon Plus Environment

Page 4 of 31

Page 5 of 31 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


of samples.18 Alkaline phosphatase digestion has also been used to determine phosphorylated groups.19 Although alkaline phosphatase has been confirmed to remove the phosphate groups from phosphorylated high-mannose-type glycans, its specificity for other phosphorylated glycans has not been evaluated; consequently, it cannot be used as an analysis tool for unknown phosphorylated glycans. Glycan biosynthesis is very complicated and involves several factors, such as glycosyltransferase, glycan hydrolase, carrier proteins, and sugar nucleotides. In addition, glycan biosynthesis is generally understood to be sensitive and can change in response to environmental changes. Therefore, it can be assumed that changes in the structures and amounts of sulfated and phosphorylated glycans occur at the same time under certain circumstances. Unfortunately, the profiles of sulfated and phosphorylated glycans have been reported independently, and the relationship between sulfated- and phosphorylated-glycan biosynthesis has not been clarified. Therefore, comprehensive analyses of these glycans are required. We reported highly sensitive methods for the analysis of glycans through the use of 2-aminobenzoic acid (2AA) derivatization20,21 and developed methods for the enrichment of 2AA-labeled acidic glycans, such as sialylated glycans and glycosaminoglycans, using a serotonin-immobilized column.21-23 Separation on the serotonin-immobilized column is achieved on the basis of weak anionic exchanges. The glycans enriched in this manner can be analyzed by HPLC and MS without the requirement for further purification because the glycans retained on this column are eluted with volatile salts at low concentrations. This is one of the advantages of the serotonin-immobilized column. In this study, we developed a novel method for the simultaneous analysis of 5 ACS Paragon Plus Environment


Page 6 of 31

2-AA-labeled sulfated and phosphorylated glycans by combining the serotonin immobilized column and HILIC. These methods readily enable the enrichment of phosphorylated and sulfated glycans from biological samples, and allow them to be completely distinguished based on retention-time differences. We also applied this method to the analysis of phosphorylated and sulfated glycans expressed in gastric-cancer cell lines. The developed method has the great potential to contribute to breakthroughs in our understanding of the novel functions of phosphorylated and sulfated glycans.

EXPERIMENTAL SECTION Materials Information on the reagents, materials, and cell lines used in this study is provided in the supporting information.

Preparation of 2AA-labeled N-glycan pools from some glycoproteins and whole cell proteins These preparations were carried out following our previous report24, with some minor modifications. Detailed protocols are provided in the supporting information.

Enrichment

of

the

sulfated

and

phosphorylated

glycans

using

a

serotonin-immobilized column To enrich the sulfated and phosphorylated glycans, the pooled 2AA-labeled glycans were separated on a serotonin-immobilized column following removal of the sialic acids. Separation was performed on a Jasco HPLC instrument equipped with two


Page 7 of 31 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


PU980 pumps and a Jasco FP-920 fluorescence detector (Hachio-ji, Tokyo, Japan). The pooled 2AA-labeled glycans were dissolved in water (180 µL). Neuraminidase (20 mU/20 µL) was added to the solution, and the mixture was incubated at 37 °C for 24 h. A 10 µL aliquot (corresponding to 50 µg of the glycoproteins or 1.0 x 106 gastric-cancer cells) of the asialo glycans was separated on a serotonin-immobilized column (150 mm × 4.6 mm i.d., J-Oilmils, Chu-o ku, Tokyo, Japan) using linear stepwise gradients, from water (eluent A) to 50 mM ammonium acetate of pH 6.8 (eluent B), at a flow rate of 0.5 mL/min. Initially, 5% eluent B was used for 2 min, and then solvent B was increased to 20% over 10 min, after which solvent B was immediately increased to 100% and the eluent was maintained at this composition for a further 25 min. The column was then equilibrated with the starting eluent. The sulfated and phosphorylated glycans eluted between 12 and 24 min. After enrichment, the sulfated and phosphorylated glycan fractions were analyzed by MALDI–TOF MS and HILIC.

Distinguishing the phosphorylated and sulfated glycans by HILIC The phosphorylated and sulfated glycan fractions obtained using the serotonin-immobilized column were separated by HILIC, using the instrument described for serotonin-immobilized column chromatography. A 10 µL sample (corresponded to 10 µg of the glycoproteins or 5.0 x 105 gastric-cancer cells) was analyzed using an amino-bound polymer column (NH2 column; Asahi Shodex NH2P-50 4E column (Showa Denko, Hachio-ji, Tokyo: 250 mm, 4.6 mm i.d.)) using a linear eluent gradient of 2% acetic acid in acetonitrile (solvent A) and 5% acetic acid in water containing 3% trimethylamine (solvent B) at a flow rate of 0.5 mL/min. The column



was initially equilibrated and eluted with 70% solvent A for 2 min, after which the proportion of solvent B was increased to 95% over 80 min. The amounts of sulfated and phosphorylated N-glycans were calculated from the corresponding peak areas on the basis of the standard curve prepared using 2AA-labeled maltopentaose.

MALDI-TOF MS and MS/MS analyses Detailed methods are provided in the supporting information.

Results and discussion Our strategy for the simultaneous analysis of sulfated and phosphorylated glycans is shown in Figure 1. Fig. 1 Glycans released from the biological samples are labeled with 2AA and digested with neuraminidase in order to remove sialic acid that interferes with the enrichment of the phosphorylated and sulfated glycans. Neuraminidase digestion also eliminates structural variations within the sulfated and phosphorylated glycans due to differences in the numbers of sialic acids, and converts them to single structures. This method not only facilitates analysis of the data but also improves detection sensitivity toward sulfated and phosphorylated glycans by MS and HPLC, despite not being able to directly observe the sialylated glycans. Following neuraminidase digestion, the sulfated and phosphorylated glycan fraction is enriched using a serotonin-immobilized column. Sulfated and phosphorylated glycans enriched with the serotonin column do not require desalting, which often results in sample loss. This feature of the serotonin-immobilized column is a significant advantage when analyzing trace amounts of phosphorylated and 8 ACS Paragon Plus Environment

Page 8 of 31

Page 9 of 31 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


sulfated glycans in a biological sample. The enriched fraction is analyzed by HILIC. Since the elution times of sulfated glycans, phosphorylated glycans, and the glycans containing R-HPO3 residues, such as GlcNAc-HPO3, are significantly different during HILIC, these glycans are completely discriminated. In addition, the glycans can be quantitatively analyzed through their corresponding peak areas.

Sulfated- and phosphorylated-glycan enrichment using a serotonin-immobilized column We found that the sulfated and phosphorylated glycans are retained on the serotonin-immobilized column. 6-Sulfated mannose, 6-phosphorylated mannose, and mannose were labeled with 2AA and injected onto the serotonin-immobilized column (Figure S-1). Sulfated mannose and phosphorylated mannose were more strongly retained on this column than mannose. In our past research, we showed that the serotonin-immobilized column has a weak anion exchange capacity.22 Serotonin has the amine nitrogen serving as a proton acceptor at neutral pH. Therefore, it is considered that the amino nitrogen of serotonin is important for retaining sulfated and phosphorylated glycans. The results indicate that the serotonin-immobilized column usefully enriches the sulfated and phosphorylated glycans from the glycan mixture. Based on these results, we applied the serotonin-immobilized column to the enrichment of sulfated and phosphorylated N-glycans from thyroglobulin, ovalbumin, and myozyme. Sulfated complex-type N-glycans, sulfated hybrid-type glycans, and phosphorylated high-mannose and hybrid-type glycans have been reported to be expressed on thyroglobulin,25 ovalbumin,26 and myozyme,27 respectively. The total N-glycans released from these three glycoproteins were labeled with 2AA and digested 9 ACS Paragon Plus Environment


with neuraminidase. We enriched the sulfated and phosphorylated glycan fractions from these asialo glycan mixtures with the serotonin-immobilized column (Figure S-2) and analyzed them by MALDI-QIT-TOF MS (Figure 2). Fig. 2 Phosphorylated and sulfated N-glycans were not observed when asialo glycan mixtures derived from the three glycoproteins were analyzed by MALDI-QIT-TOF MS prior to enrichment by the serotonin-immobilized column (Figure 2, left panels). On the other hand, after enrichment, peaks corresponding to sulfated and phosphorylated N-glycans were clearly visible (Figure 2, right panels). Based on past reports, molecular-ion peaks observed for three glycoproteins were assigned. Three sulfated glycans were observed in the spectrum of thyroglobulin (Figure 2A, right panel); the major ion peak observed at m/z 1986.7 was confirmed to a monosulfated biantennary N-glycan; this structure is the major component of sulfated glycans expressed on thyroglobulin.25 The molecular ion peak at m/z 2351.8 corresponds to a monosulfated triantennary N-glycan. We also detected a monosulfated biantennary N-glycan, including the α-Gal epitope (m/z 2148.7). The molecular ion peaks at m/z 1906.5 and 2271.8 correspond to desulfated biantennary and triantennary N-glycans, which were generated during the MALDI-MS ionization process. The enriched fraction from ovalbumin exhibited a complex mass spectrum (Figure 2B, right panel). The three molecular ion peaks at m/z 1637.3, m/z 1799.3, and m/z 1961.4 correspond to monosulfated hybrid-type glycans reported previously.26 In addition, several molecular ion peaks corresponding to high-mannose-type glycans bearing a phosphate or a sulfate group were observed (m/z 1596.3, m/z 1758.4, m/z 1920.4, and m/z 2082.4). However, whether these glycans are phosphorylated or sulfated could not be determined on the 10 ACS Paragon Plus Environment

Page 10 of 31

Page 11 of 31 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


basis of this analysis. Sulfated complex-type glycans have not been reported for ovalbumin, but some minor ion peaks that seem to correspond to them were observed. Six molecular ion peaks were observed in the enriched myozyme fraction. Although five molecular ion peaks at m/z 1434.3, m/z 1596.7, m/z 1758.8, m/z 1920.6, and

m/z

2104.6

were

confirmed

to

correspond

to

monophosphorylated

high-mannose-type glycans, we also detected a phosphorylated hybrid-type glycan in myozyme (m/z 1961.9). Although these six molecular ion peaks were the same as those corresponding to monosulfated high-mannose-type and hybrid-type glycans, we assigned them to phosphorylated glycans on the basis of past reports.27 From the above results, we conclude that the serotonin-immobilized column is useful for enriching sulfated and phosphorylated N-glycans.

Distinguishing sulfated and phosphorylated glycans by HILIC We succeed in enriching sulfated and phosphorylated glycans using the serotonin-immobilized column. However, it was still difficult to distinguish the sulfated from the phosphorylated glycans by MS alone. Therefore, we subjected these mixtures to chromatography using the NH2 column in order to distinguish the sulfated and phosphorylated glycans. Since the stationary phase of this column tends to be positively charged, electrostatic interactions play important roles during separation.23,28 The separation results for 2AA-labeled sulfated and phosphorylated mannose are displayed in Figure 3A. Fig. 3 Sulfated mannose and phosphorylated mannose were observed at 25 min and 58 min, respectively. Phosphorylated mannose was retained on the column much more strongly 11 ACS Paragon Plus Environment


than the sulfated mannose because the phosphate groups of the glycans each bear one more negative charge than the analogous sulfated groups at near-neutral pH (Figure 3B). These results indicate that the NH2 column is useful for the separation of these groups and for distinguishing sulfated and phosphorylated glycans. We also analyzed the sulfated and phosphorylated N-glycan fractions derived from thyroglobulin, ovalbumin, and myozyme by HILIC with the NH2 column (Figure 4). We collected the compounds corresponding to the observed peaks and analyzed them by MALDI-QIT-TOF MS (Figure S-3). The proposed structures, based on previously reported data25-27, are summarized in the supporting information (Table S-1, S-2, and S-3) Fig. 4 All sulfated N-glycans eluted between 35 min and 41 min. Two main peaks were observed in the 35–41 min range in the chromatogram corresponding to thyroglobulin (peaks 1 and 2). Peak 2, at 38 min, exhibited molecular ion peaks at m/z 1986.7, 2351.7, and 2716.9, which correspond to monosulfated biantennary, triantennary, and tetraantennary N-glycans respectively. A monosulfated biantennary N-glycan, including the α-Gal epitope (m/z 2148.7), was also observed in the spectrum of peak 2. Monosulfated triantennary, and tetraantennary N-glycans, including α-Gal epitope (m/z 2513.8 and m/z 2879.0) were observed in peak 1. No glycans were observed after 40 min. Five peaks (peaks 3–7) were observed in the 35–41 min range in the chromatogram of ovalbumin. Peak 6 exhibited a molecular ion peaks at m/z 1799.9 and m/z 1961.9. Peak 7 also gave a molecular ion peak at m/z 1637.9; these molecular ion peaks correspond to monosulfated hybrid-type glycans. In addition, Peak 4 exhibited a 12 ACS Paragon Plus Environment

Page 12 of 31

Page 13 of 31 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


molecular ion peak at m/z 1596.4 in its MS, which corresponds to a monophosphorylated or monosulfated Man6 high-mannose-type glycan. Peak 5 also showed molecular ion peaks at m/z 1758.5, 1920.1, and 2082.1 that correspond to monosulfated or monophosphorylated Man7, Man8, and Man9 high-mannose-type glycans, respectively. The elution time of this peak is quite different to that of the phosphorylated glycans; however it is the same as that corresponding to the sulfated glycans. The existence of hybrid-type glycans having 4-sulfated mannose residues has been confirmed in ovalbumin,26 as evidenced by molecular ion peaks that correspond to high-mannose-type glycans bearing 4-sulfated mannose residues. Although the details of the biosyntheses leading to these compounds is still unknown, we believe that any molecule in the 4-sulfotransferase family may be involved in their biosyntheses.29 Peak 4

gave

a

molecular

ion

peak

at

m/z

1678.9,

which

corresponds

to

HSO3Hex4HexNAc4-2AA; this monosaccharide composition corresponds to sulfated complex-type glycans. Phosphorylated N-glycans eluted after 49 min. Seven peaks (peaks 16–22) were observed after 49 min in the chromatogram of myozyme. Peaks 17, 18, 21, and 22 exhibited some molecular ion peaks that correspond to monophosphorylated high-mannose-type glycans (m/z 1596.7, m/z 1758.8, m/z 1942.4, and m/z 2104.6). Some molecular ion peaks confirmed as belonging to monophosphorylated hybrid glycans were also observed in peaks 16, 19, and 20. Although these molecular ion peaks appear to correspond to the monosulfated hybrid-type and high-mannose-type observed in ovalbumin, their elution times are significantly different from those of sulfated glycans and are more similar to that of phosphorylated mannose. In addition, when these samples were digested with alkaline phosphatase, peaks 17–22 were no longer observed in the chromatogram (Figure S-4), 13 ACS Paragon Plus Environment


Page 14 of 31

confirming that these peaks correspond to phosphorylated glycans. More interestingly, two small peaks were observed upon elution of the sulfated glycans in myozyme (peak 11 and peak 12). Myozyme was not previously known to contain sulfated glycans; consequently, the presence of sulfated glycans was confirmed for the first time in this study. Furthermore, we detected some glycans containing the GlcNAc-HPO3 residue at 41–45 min. Peaks 8–10 observed for ovalbumin gave molecular ion peaks at m/z 1799.9, 1637.5, and 1961.2 respectively. These molecular ion peaks could correspond to sulfated or phosphorylated hybrid-type N-glycans, however their observed elution times were different to those of either type of glycan. A fragment ion was clearly observed at m/z 1596.0 in the MS/MS spectrum of the m/z 1799.9 in peak 8; this ion corresponds to a sulfated or phosphorylated high-mannose-type glycan devoid of GlcNAc residues (Figure 5A). Fig. 5 In addition, we subjected peak 8 to partial acidic hydrolysis and analyzed the products using the NH2 column (Figure 5B). Peak 8 was no longer present in the chromatogram following this procedure, and a new peak was clearly observed at 55 min when the phosphorylated high-mannose-type glycan was eluted. These results clearly show that peak 8 contains a high-mannose-type glycan that contains the GlcNAc-HPO3 residue. Peaks 9 and 10 also appear to contain the same type of glycan as peak 8; these glycans were also observed in myozyme (peaks 13–15). These

results

reveal

that

sulfated

glycans,

phosphorylated

glycans,

and

GlcNAc-HPO3-residue-containing glycans are able to be separated and distinguished using the NH2 column. 14 ACS Paragon Plus Environment

Page 15 of 31 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


Analysis of sulfated and phosphorylated N-glycans in gastric-cancer cell lines We applied the developed method to the analysis of sulfated and phosphorylated

N-glycans

expressed

on

MKN7

(well-differentiated

human

gastric-cancer cell line) and MKN45 (poorly differentiated human gastric-cancer cell line) cells. After purification of the sulfated and phosphorylated N-glycan fractions from the MKN7 and MKN45 cells with the serotonin-immobilized column, these fractions were analyzed by HILIC with the NH2 column. (Figure 6A). The fractions corresponding to the peaks observed in each chromatogram were collected and analyzed by MALDI-QIT-TOF MS (Figure S-5) and their structures are assigned as shown in the supporting information (Table S-4) Fig. 6 Peaks at elution times corresponding to sulfated glycans were observed in the chromatogram from each gastric-cancer cell line (peaks 1 and 2). Monosulfated biantennary, triantennary, and tetraantennary N-glycans were observed in both cancer cells. We also detected monosulfated N-glycans, including two fucose residues, on the MKN7 cells. Naka et al. reported that multi-fucosylated N-glycans were expressed on MKN7 and MKN45 cells.30 These glycans corresponded to the monosulfated forms of multi-fucosylated N-glycans. There were no significant differences in the phosphorylated N-glycan profiles of the MKN7 and MKN45 cells (peaks 10–14). Peaks 12–14 exhibited molecular ion peaks at m/z 1920, 1596, and 1758, respectively, in their mass spectra. These molecular ions correspond to monophosphorylated high-mannose-type glycans observed in myozyme. We also observed monophosphorylated hybrid-type glycans in peaks 10 and 15 ACS Paragon Plus Environment


11. Some glycans containing the GlcNAc-HPO3 residue were observed at peaks 3– 9. Peaks 3 and 4 contain hybrid-type GlcNAc-HPO3-residue-containing glycans, while and peaks 5–9 contain high-mannose-type glycans bearing the GlcNAc-HPO3 residue. Interestingly, the amounts of peaks 5 and 8 from the MKN45 cells were markedly lower than those from the MKN7 cells. Peak 5 exhibited a molecular ion peak at m/z 1799.4, which was due to a Man6 high-mannose-type glycan that contains the GlcNAc-HPO3 residue. Peak 8 exhibited a molecular ion peak at m/z 2123, which corresponds to a Man8 high-mannose type-glycan that contains the GlcNAc-HPO3 residue. We also detected glycans bearing two GlcNAc-HPO3 residues in the MKN7 and MKN45 cells; Peak 16 exhibited a molecular ion peak at m/z 2244, which corresponded to a Man7 high-mannose-type glycan that contains two GlcNAc-HPO3 residues (Figure S-5). In addition, fragment ion peaks were observed at m/z 2041 and m/z 1838, which correspond to losses of one GlcNAc residue and two GlcNAc residues from the molecular ion peak at m/z 2244. In this study, we found that the ester bond between GlcNAc and the phosphate group is chemically unstable and fragments easily during MALDI-MS analysis (Figure 5). Actually, ion peaks corresponding to the loss of GlcNAc were often observed in the MS spectra of glycans containing GlcNAc-PO3H residues in this study (Figures S-3 and S-5). Therefore, these peaks were judged to be fragment ions generated during the MALDI-MS ionization process. When pseudo MS3 analysis was performed on the m/z 1838 precursor ion, a fragment ion peak at m/z 1678 (Y4α,/5α) corresponding to a loss of two phosphate groups from m/z 1838 were observed(Figure 6B). In addition, two fragment ion peaks at m/z 848 (Y3β,/B5) and m/z 564 (Y4α/4β/B4 or Y5α/4β/B4) were observed in the MS2 spectrum. The fragment ion peak 16 ACS Paragon Plus Environment

Page 16 of 31

Page 17 of 31 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


at m/z 848.0 corresponds to (H2PO3)2Man3GlcNAc1, while the fragment ion peak at m/z 644.8 is due to (H2PO3)2Man3. Based on these results, the peak at m/z 1838 was confirmed to correspond to the structure depicted in Figure 6B. Clearly, our method is capable of detecting diphosphorylated glycans. We also calculated the amounts of sulfated and phosphorylated glycans from the peak areas observed in the HILIC traces (Figure 7). While the sulfated glycans and the glycans bearing GlcNAc-PO3H residues in the MKN cells were poorly separated, we confirmed by MS analysis that there was essentially no cross-contamination between the

fraction

containing

the

sulfated

glycans

and

that

containing

the

GlcNAc-PO3H-residue-bearing glycans (Figure S-5). Therefore, both fractions were distinguished and quantified. Although it did not present a problem this time, when contaminated, further separation of the fraction by capillary electrophoresis or a similar technique is necessary. Fig. 7 The amounts of sulfated glycans in the MKN45 cells were slightly higher than those of the MKN7 cells (n=3, P=0.07). During the process of malignant transformation and differentiation, variations in O-glycan, glycolipid, and glycosaminoglycan sulfation have often been reported31-33; however, to the best of our knowledge, there are no reports relating to N-glycans. Although further studies are required, we speculate that this observation is related to the differentiation of cancer cells and the amounts of sulfated N-glycans. On the other hand, no differences in the amounts of phosphorylated glycans were observed between the two cells, but the expressed amounts of glycans containing the GlcNAc-HPO3 residue were remarkably different; their amounts were three times higher in the MKN7 cells than in MKN45 (n=3, P=0.004). In addition, we 17 ACS Paragon Plus Environment


calculated the ratio of glycans containing the GlcNAc-HPO3 residue to total phosphorylated glycans (Figure 7B). The ratio of GlcNAc-HPO3-residue-containing glycans from MKN7 cells was twice that of the MKN45 cells (n=3, P=0.02). Our results show that the amounts of these glycans are higher in the differentiated cells; consequently, glycans that contain the GlcNAc-HPO3 residue may be biomarkers of differentiation and malignant transformation. We also considered that many of these phosphorylated glycans are present on lysosomal enzymes. Lysosomal function has been revealed to be impaired when abnormalities occur during the biosyntheses of these glycans.34,35 On the other hand, MKN cells have been shown to have abnormal lysosome structures.30 We are also interested in the relationship between changes in these glycan profiles and the lysosomal function of MKN cells.

Conclusion In the present study, we developed a novel method for the simultaneous analysis of sulfated and phosphorylated glycans. We revealed that a serotonin-immobilized column can be used to enrich phosphorylated and sulfated glycans. We were also able to discriminate between enriched sulfated and phosphorylated glycans with the NH2 column. The combination of these two columns facilitates the trace analysis of phosphorylated and sulfated glycans in biological samples. This method also facilitates the detection of glycans that contain the GlcNAc-HPO3 residue. In addition, we applied the present method to the analysis of sulfated and phosphorylated N-glycans expressed in two cancer-cell lines. A variety of sulfated glycans, phosphorylated glycans, and GlcNAc-HPO3-residue-containing glycans were observed in the two gastric-cancer cell lines. We also compared the 18 ACS Paragon Plus Environment

Page 18 of 31

Page 19 of 31 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


amounts of these glycans in these cell lines; the amounts of GlcNAc-HPO3 residues in well-differentiated gastric-cancer cells (MKN7 cells) are much higher than in poorly differentiated gastric-cancer cells (MKN45 cells); these glycans appear to be indicators of differentiation and malignancy in cancer cells. Comprehensive analyses of N-glycans expressed on these cancer cell lines have been previously reported23,30; however, sulfated and phosphorylated glycans were not detected in these previous studies. Clearly, our method is able to detect sulfated and phosphorylated glycans that are not detected by conventional methods. Since no glycans bearing two or three sulfate or phosphate groups were detected in this study, this method needs to be applied to further biological samples in the future, and the number of analysis examples needs to be increased. In order to analyze biological samples, it is also important to acquire sulfate- and phosphate-group positional information for these glycans; accordingly we intend to analyze phosphorylated and sulfated glycans in biological samples by combining our method with detailed MS analyses. We believe that the present method makes a significant contribution to the discovery of new disease markers and novel functions of sulfated and phosphorylated glycans.

References (1) Fiete, D.; Srivastava, V.; Hindsgaul, O.; Baenziger, J. U. Cell 1991, 67, 1103-1110. (2) Kannagi, R. Curr. Opin. Struct. Biol. 2002, 12, 599-608. (3) Fukushima, K.; Ohkura, T.; Kanai, M.; Kuroki, M.; Matsuoka, Y.; Kobata, A.; Yamashita, K. Glycobiology 1995, 5, 105-115. (4) Magro, G.; Perissinotto, D.; Schiappacassi, M.; Goletz, S.; Otto, A.; Muller, E. C.; Bisceglia, M.; Brown, G.; Ellis, T.; Grasso, S.; Colombatti, A.; Perris, R. Am. J. Pathol. 2003, 163, 183-196. (5) Plaas, A. H.; West, L. A.; Wong-Palms, S.; Nelson, F. R. J. Biol. Chem. 1998, 273, 19 ACS Paragon Plus Environment


12642-12649. (6) Bayliss, M. T.; Osborne, D.; Woodhouse, S.; Davidson, C. J. Biol. Chem. 1999, 274, 15892-15900. (7) Dahms, N. M.; Lobel, P.; Kornfeld, S. J. Biol. Chem. 1989, 264, 12115-12118. (8) Breloy, I.; Pacharra, S.; Ottis, P.; Bonar, D.; Grahn, A.; Hanisch, F. G. J. Biol. Chem. 2012, 287, 18275-18286. (9) Yagi, H.; Kuo, C. W.; Obayashi, T.; Ninagawa, S.; Khoo, K. H.; Kato, K. Mol. Cell. Proteomics 2016, 15, 3424-3434. (10) Braulke, T.; Pohl, S.; Storch, S. J. Inherit. Metab. Dis. 2008, 31, 253-257. (11) Mitsui, Y.; Yamada, K.; Hara, S.; Kinoshita, M.; Hayakawa, T.; Kakehi, K. J. Pharm. Biomed. Anal. 2012, 70, 718-726. (12) Lei, M.; Mechref, Y.; Novotny, M. V. J. Am. Soc. Mass. Spectrom. 2009, 20, 1660-1671. (13) Cheng, C. W.; Chou, C. C.; Hsieh, H. W.; Tu, Z.; Lin, C. H.; Nycholat, C.; Fukuda, M.; Khoo, K. H. Anal. Chem. 2015, 87, 6380-6388. (14) Kumagai, T.; Katoh, T.; Nix, D. B.; Tiemeyer, M.; Aoki, K. Anal. Chem. 2013, 85, 8692-8699. (15) Toyoda, M.; Narimatsu, H.; Kameyama, A. Anal. Chem. 2009, 81, 6140-6147. (16) Kinoshita-Kikuta, E.; Kinoshita, E.; Yamada, A.; Endo, M.; Koike, T. Proteomics 2006, 6, 5088-5095. (17) Szabo, Z.; Thayer, J. R.; Reusch, D.; Agroskin, Y.; Viner, R.; Rohrer, J.; Patil, S. P.; Krawitzky, M.; Huhmer, A.; Avdalovic, N.; Khan, S. H.; Liu, Y.; Pohl, C. J. Proteome Res. 2018, 17, 1559-1574. (18) Taguchi, T.; Iwasaki, M.; Muto, Y.; Kitajima, K.; Inoue, S.; Khoo, K. H.; Morris, H. R.; Dell, A.; Inoue, Y. Eur. J. Biochem. 1996, 238, 357-367. (19) Sandra, K.; Van Beeumen, J.; Stals, I.; Sandra, P.; Claeyssens, M.; Devreese, B. Anal. Chem. 2004, 76, 5878-5886. (20) Yamada, K.; Hyodo, S.; Matsuno, Y. K.; Kinoshita, M.; Maruyama, S. Z.; Osaka, Y. S.; Casal, E.; Lee, Y. C.; Kakehi, K. Anal. Biochem. 2007, 371, 52-61. (21) Yamada, K.; Kinoshita, M.; Hayakawa, T.; Nakaya, S.; Kakehi, K. J. Proteome Res. 2009, 8, 521-537. (22) Yamada, K.; Mitsui, Y.; Kakoi, N.; Kinoshita, M.; Hayakawa, T.; Kakehi, K. Anal. Biochem. 2012, 421, 595-606. (23) Naka, R.; Kamoda, S.; Ishizuka, A.; Kinoshita, M.; Kakehi, K. J. Proteome Res. 2006, 5, 88-97. (24) Kinoshita, M.; Mitsui, Y.; Kakoi, N.; Yamada, K.; Hayakawa, T.; Kakehi, K. J. Proteome Res. 2014, 13, 1021-1033. (25) de Waard, P.; Koorevaar, A.; Kamerling, J. P.; Vliegenthart, J. F. J. Biol. Chem. 1991, 266, 4237-4243. (26) Yamashita, K.; Ueda, I.; Kobata, A. J. Biol. Chem. 1983, 258, 14144-14147. (27) Jongen, S. P.; Gerwig, G. J.; Leeflang, B. R.; Koles, K.; Mannesse, M. L.; van Berkel, P. H.; Pieper, F. R.; Kroos, M. A.; Reuser, A. J.; Zhou, Q.; Jin, X.; Zhang, K.; Edmunds, T.; Kamerling, J. P. Glycobiology 2007, 17, 600-619. (28) Anumula, K. R.; Dhume, S. T. Glycobiology 1998, 8, 685-694. (29) Baenziger, J. U. Biochem. Soc. Trans. 2003, 31, 326-330. (30) Ishizuka, A.; Hashimto, Y.; Naka, R.; Kinoshita, M.; Kakehi, K.; Seino, J.; Funakoshi, Y.; Suzuki, T.; Kameyama, A.; Narimatsu, H. Biochem. J. 2008, 413, 20 ACS Paragon Plus Environment

Page 20 of 31

Page 21 of 31 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


227-237. (31) Knelson, E. H.; Nee, J. C.; Blobe, G. C. Trends Biochem. Sci. 2014, 39, 277-288. (32) Sugiyama, T.; Miyazawa, M.; Mikami, M.; Goto, Y.; Nishijima, Y.; Ikeda, M.; Hirasawa, T.; Muramatsu, T.; Takekoshi, S.; Iwamori, M. Int. J. Gynecol. Cancer 2012, 22, 1192-1197. (33) Tanaka-Okamoto, M.; Mukai, M.; Takahashi, H.; Fujiwara, Y.; Ohue, M.; Miyamoto, Y. Glycobiology 2017, 27, 400-415. (34) Tiede, S.; Storch, S.; Lubke, T.; Henrissat, B.; Bargal, R.; Raas-Rothschild, A.; Braulke, T. Nat. Med. 2005, 11, 1109-1112. (35) Kollmann, K.; Pohl, S.; Marschner, K.; Encarnacao, M.; Sakwa, I.; Tiede, S.; Poorthuis, B. J.; Lubke, T.; Muller-Loennies, S.; Storch, S.; Braulke, T. Eur. J. Cell Biol. 2010, 89, 117-123.



Page 22 of 31

Figure legends Figure 1. Strategy for the simultaneous analysis of sulfated and phosphorylated glycans

Figure

2.

Serotonin-immobilized-column

enrichment

of

sulfated

and

phosphorylated N-glycans. 2AA-labeled asialo N-glycans derived from (A) thyroglobulin, (B) ovalbumin (B), and (C) myozyme were analyzed by MALDI-QIT-TOF MS. MS spectra of the (left) un-enriched

asialo

mixtures

serotonin-immobilized-column-enriched

of sulfated

N-glycans, and

and

(right)

phosphorylated

N-glycans.

Symbols: yellow filled circles, galactose; green filled circles, mannose; blue filled squares, N-acetylglucosamine; red filled triangles, fucose.

Figure 3. Distinguishing the sulfated and phosphorylated glycans by HILIC (A) The 2AA-labeled 6-sulfated mannose and 6-phosphorylated mannose were analyzed by HILIC. Analytical conditions are as described in the Experimental Section. (B) Interactions that help to distinguish the 6-sulfated and 6-phosphorylated mannoses by HILIC with the NH2 column.

Figure 4. HILIC analyses of sulfated and phosphorylated N-glycans derived from three glycoproteins. Sulfated and phosphorylated N-glycan fractions from thyroglobulin, ovalbumin, and myozyme were enriched. These fractions were then analyzed by HILIC. The analytical HILIC conditions are described in the Experimental Section. 22 ACS Paragon Plus Environment

Page 23 of 31 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


Figure 5. Structural analysis of glycans containing the GlcNAc-HPO3 residue. Glycans containing the GlcNAc-HPO3 residue in ovalbumin were analyzed by (A) MALDI-QIT-TOF MS/MS, and (B) HILIC following acid hydrolysis. The analytical HILIC conditions are described in the Experimental Section.

Symbols: green filled

circles, mannose; blue filled squares, N-acetylglucosamine.

Figure 6. Analysis of sulfated and phosphorylated N-glycans derived from gastric-cancer cells. (A) Sulfated and phosphorylated N-glycan fractions from MKN7 and MKN45 cells were enriched and analyzed by HILIC. (B) The molecular ion peaks at m/z 1838.9 observed in Fraction 16 were analyzed by pseudo MS3. The analytical HILIC conditions are described in the Experimental Section. Symbols: green filled circles, mannose; blue filled squares, N-acetylglucosamine.

Figure 7. Quantitative analyses of sulfated and phosphorylated N-glycans derived from gastric-cancer cells (A)

Total

amounts

of

sulfated

glycans,

phosphorylated

glycans,

and

GlcNAc-HPO3-residue-containing glycans on MKN7 and MKN45 cells. (B) Ratios of GlcNAc-HPO3-residue-containing glycans to total phosphorylated glycans. The MKN7 and MKN45 glycan contents are displayed as blue and orange bars, respectively. The total amounts and ratios were calculated from the peak areas determined by HILIC. Values are shown as mean ± SD (n = 3, *P < 0.05 and **P < 0.01).



Fig. 1.


Page 24 of 31

Page 25 of 31 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


Fig. 2.



Fig. 3.


Page 26 of 31

Page 27 of 31 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


Fig. 4.



Fig. 5.


Page 28 of 31

Page 29 of 31 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


Fig. 6.



Fig. 7.


Page 30 of 31

Page 31 of 31 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


for TOC only


Simultaneous analysis of sulfated and phosphorylated glycans by

Recommend Documents