Rapid Analysis of Cell Surface N-Glycosylation from Living Cells

Oct 28, 2014 - Rapid Analysis of Cell Surface N-Glycosylation from Living Cells ... it is of paramount importance to have robust methods to analyze th...
24 downloads 0 Views 2MB Size
Subscriber access provided by DOKUZ EYLUL UNIVERSITESI

Technical Note

Rapid analysis of cell surface N-glycosylation from living cells using mass spectrometry Houda Hamouda, Matthias Kaup, Mujib Ullah, Markus Berger, Volker Sandig, Rudolf Tauber, and Véronique Blanchard J. Proteome Res., Just Accepted Manuscript • DOI: 10.1021/pr5003005 • Publication Date (Web): 28 Oct 2014 Downloaded from http://pubs.acs.org on November 4, 2014

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.

Journal of Proteome Research 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 29

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

Journal of Proteome Research

Rapid analysis of cell surface N-glycosylation from living cells using mass spectrometry Houda Hamouda,1,2 Matthias Kaup,1 Mujib Ullah,3 Markus Berger,1 Volker Sandig,4 Rudolf Tauber1 and Véronique Blanchard1* 1

Institute of Laboratory Medicine, Clinical Chemistry and Pathobiochemistry, Charité-

Universitätsmedizin Berlin, Augustenburger Platz 1, 13353 Berlin, Germany, 2Freie Universität Berlin, Dept. of Biology, Chemistry and Pharmacy, Takustrasse 3, 14195 Berlin, Germany, 3

Tissue Engineering Laboratory & Berlin-Brandenburg Center for Regenerative Therapies, Dept.

of Rheumatology and Clinical Immunology, Charité-Universitätsmedizin Berlin, Charitéplatz 1, 10117 Berlin, Germany, 4ProBioGen AG, Goethestraße 54, 13086 Berlin, Germany

*

To whom correspondence should be addressed. Dr. Véronique Blanchard, Institute of

Laboratory Medicine, Clinical Chemistry, and Pathobiochemistry, Charité-Universitätsmedizin Berlin, Augustenburger Platz 1, 13353 Berlin, Germany Phone: (+49) 30 450669196, Fax: (+49) 30 450569906 E-mail: [email protected]

Running title: MALDI-TOF investigation of cell surface N-glycans

1

ACS Paragon Plus Environment

Journal of Proteome Research

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

Cell surfaces are covered with a dense carbohydrate layer referred to as the glycocalyx. Since different cell types express different glycan signatures, it is of paramount importance to have robust methods to analyze the glycome of living cells. The common procedure involves first cell lysis and extraction of membranes (glyco)proteins and yields a major proportion of highmannose N-glycans most likely stemming from intracellular proteins derived from the ER. Using HEK 293 cells as a model system, we developed a reproducible, sensitive and fast method to profile surface N-glycosylation from living cells. We directly released glycopeptides from cell surfaces through tryptic digestion of freshly harvested and vital cells. Thereby we improved the detection and quantification of complex-type N-glycans by increasing their relative amount from 14 % to 85 %. It was also possible to detect 25 additional structures in HEK 293, 48 in AGE1.HN, 42 in CHO-K1 and 51 in Hep G2 cells. The additional signals gave a deeper insight into cell type specific N-glycan features such as antennarity, fucosylation and sialylation. Thus this protocol, which can potentially be applied to any cells, will be useful in the fields of glycobiotechnology and biomarker discovery.

Keywords: cell surface, glycocalyx, MALDI-TOF-MS, N-glycosylation

2

ACS Paragon Plus Environment

Page 2 of 29

Page 3 of 29

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

Journal of Proteome Research

Introduction

N-glycosylation is a common protein post-translational modification occurring on asparagine residues of the consensus sequence asparagine-X-serine/threonine, where X may be any amino acid except proline 1. Protein N-glycosylation takes place in the endoplasmic reticulum (ER) as well as in the Golgi apparatus, where the extent and type of glycosylation is determined by the cell type and species 1-3. A detailed structural characterization of the cell surface glycome is of great importance to gain a better understanding of glycan functions in cells, to support the process of biomarker discovery as well as to select cell lines for the production of recombinant glycoproteins.

A classical method used to profile the N-glycosylation pattern of cells is membrane (glyco)protein isolation followed by denaturation and/or tryptic digestion of the (glyco)proteins and an enzymatic release of the N-glycans 4-6 7-9. Alternatively, (glyco)proteins can be extracted by consecutive triton phase partitioning then blotted on PVDF membranes prior to N-glycan release10. However, such protocols require many time-consuming steps and the resulting Nglycan pool contains intra- and extracellular proteins. In addition, we and others found that the N-glycans obtained by this method are mostly of the high-mannose type, which may originate from intracellular proteins derived from the ER.4, 6, 9, 11-13 This hinders accurate detection and quantification of the complex-type N-glycans. We recently improved complex-type N-glycan quantification using first Endo H to release high-mannose and hybrid-type N-glycans and subsequently using PNGase F in order to release complex-type N-glycans 14. However, this method still involves many steps and is therefore time-consuming. Lin et al. 15 reported an approach that uses C18 cartridges to retain permethylated complex-type N-glycans and thus

3

ACS Paragon Plus Environment

Journal of Proteome Research

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

separate them from high-mannose structures and de-N-glycosylated peptides by stepwise elution. In another approach, plasma membranes were blocked on a surface prior to cell lysis so that the intracellular part was washed away, which resulted in an enhanced detection of complex-type Nglycans.11 However, these methods cannot be applied to living cells. The cell surface glycome of living cells can be visualized by lectin staining16 or by incorporating unnatural carbohydrate precursors on cell glycoproteins.17-20 Nevertheless, such strategies are specific to certain glycan epitopes and detailed profiling of the cell surface is not possible. As a result, we aimed at developing an appropriate and simple protocol, which allows the fast analysis of the N-glycome from living cells with significantly lower levels of high-mannose type N-glycans. In order to show the efficiency of the new protocol, we compared it with the above-mentioned classical method, the cell membrane preparation protocol.

Materials and Methods

Chemicals and cell lines

HEK-293, CHO-K1 and Hep G2 and cells were purchased from DSMZ (Braunschweig, Germany). AGE1.HN cells were a generous gift from ProBioGen AG (Berlin, Germany). Dulbecco’s Modified Eagle Medium (DMEM), Hams F12, RPMI, L-glutamine, sodium pyruvate and phosphate-buffered saline (PBS) were purchased from PAA (Pasching, Austria). Penicillin, streptomycin, collagen A as well as fetal calf serum (FCS) were from Biochrom AG (Berlin, Germany). Adenovirus Expression Medium (AEM) was purchased from Gibco (California, USA). KCl, CaCl2 and acetic acid were purchased from Roth (Karlsruhe, Germany). Sodium

4

ACS Paragon Plus Environment

Page 4 of 29

Page 5 of 29

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

Journal of Proteome Research

dodecyl sulfate (SDS) was purchased from Serva (Heidelberg, Germany). Acetonitrile was purchased from VWR (Leuven, Belgium). Super-dihydroxybenzoic acid (sDHB) and methyl iodide were purchased from Sigma-Aldrich (St. Louis, MO, USA). All the other chemicals were purchased from Merck (Darmstadt, Germany), unless stated otherwise.

Cell culture

HEK 293 cells were cultivated in DMEM containing glucose (4.5 g/l), 10 % fetal calf serum (FCS), penicillin (100 U/mL), streptomycin (100 µg/mL), L-glutamine (200 mM) and sodium pyruvate (1 mM). AGE1.HN cells were cultured in adenovirus expression medium (AEM) containing penicillin (100 U/mL), streptomycin (100 µg/mL) and L-glutamine (2 mM). CHO-K1 cells were cultured in Ham’s F12 containing 10 % FCS, penicillin (100 U/mL), streptomycin (100 µg/mL) and sodium pyruvate (1 mM). Hep G2 cells were cultured in RPMI containing 10 % FCS, penicillin (100 U/mL), streptomycin (100 µg/mL), L-glutamine (2 mM), sodium pyruvate (1 mM) and 0.32 % glucose. The adherent cell lines HEK 293, CHO-K1 and Hep G2 were cultured in 60.1 cm2 dishes. Cell culture dishes used for Hep G2 cultivation were first incubated with 5 mL of 10 % collagen A in PBS containing CaCl2 (9 mM) and MgCl2 (5 mM) for 30 min at 4 °C. They were subsequently washed twice with PBS and plated with Hep G2 cells. AGE1.HN suspension cells were cultured in 250 mL flasks. The medium was changed twice per week and adherent cell lines (HEK 293, CHO-K1 and Hep G2) were cultured until about 90 % confluence was reached.

Isolation of cell membrane glycoproteins

5

ACS Paragon Plus Environment

Journal of Proteome Research

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

Approximately 4x106 cells per cell type were harvested for the isolation of cell membrane glycoproteins. AGE1.HN suspension cells were harvested, centrifuged for 15 min at 420 g, washed four times with PBS and resuspended in 100 µl PBS. The adherent cell lines HEK 293, CHO-K1 and Hep G2 were washed with ice-cold PBS in the cell culture dish, suspended in PBS, centrifuged for 15 min at 420 g and additionally washed twice with PBS. Cells were subsequently resuspended in 100 µl PBS. Cell membrane extraction was performed according to Hamouda et al. 14 Briefly, cell pellets were thawed and suspended in 2 mL of homogenization buffer consisting of KCl (150 mM), CaCl2 (2 mM), NaHCO3 (1 mM) and protease inhibitors (complete EDTA-free, Roche Applied Science, Mannheim, Germany) in the concentration recommended by the manufacturer. Cell lysis was then performed by 30 strokes through a syringe having a narrow needle, and 20 mL of NaHCO3 (1 mM) was subsequently added. The lysate was centrifuged at 1,400 g for 30 min at 4°C and the pellet was discarded. The supernatant, which contains cellular membranes, was then collected and centrifuged at 48,000 g for 20 min at 4°C. The resulting pellet, which contains glycoproteins and other membrane proteins, was washed three times with 300 µl of water (26,000 g, 15 min, 4°C). The pellet was resuspended in water, methanol and chloroform (3:8:4) and incubated for 30 min on ice in order to separate lipids from membrane glycoproteins. Proteins were finally precipitated by centrifugation at 26,000 g for 20 min at 4°C. The supernatant was discarded and proteins were washed twice with 400 µl of ethanol (26,000 g, 10 min, 4°C) and then resuspended in 300 µl NaH2PO4/Na2HPO4 buffer (20 mM), pH 7.0, containing 2 % SDS and 2 % NP-40 (Calbiochem). Detergents were removed overnight using CalbiosorbTM adsorbent beads (Merck; Darmstadt, Germany).

N-glycan release and isolation from cell membrane glycoproteins

6

ACS Paragon Plus Environment

Page 6 of 29

Page 7 of 29

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

Journal of Proteome Research

Membrane glycoproteins were dissolved in 300 µl of NaH2PO4/ Na2HPO4 buffer (20 mM, pH 7.0). Trypsin (20 µl, 1 µg/µl, from bovine pancreas, Sigma Aldrich, St. Louis, MO) was then added and samples were incubated for 6 h at 37 °C. A freshly prepared aliquot of trypsin (20 µl, 1 µg/µl) was added to the mixture and samples were digested overnight at 37 °C. After trypsin deactivation (99 °C, 5 min), 1.5 U of peptide-N4-(N-acetyl-β-glucosaminyl) asparagine amidase F (PNGase F) from Flavobacterium meningosepticum (Roche Applied Science, Mannheim, Germany) was added to the glycopeptide samples and digestions were performed overnight at 37 °C. Digestion was continued for 8 h at 37 °C by addition of a second aliquot of 1 U PNGase F. Released N-glycans were isolated from the peptide moieties using C18 Extract-CleanTM cartridges (Alltech, Deerfield, IL). Briefly, samples were acidified with 1 % TFA to a final pH value < 4. Columns were equilibrated with 3 x 400 µl of 80 % ACN in 0.1 % TFA followed by 0.1 % TFA in water, respectively. Samples were applied and columns were washed with 3 x 400 µl of 0.1 % TFA in water. The flow-through, which contained free N-glycans, was collected. Nglycans were then desalted using carbograph Extract-CleanTM columns (Alltech, Deerfield, IL). Columns were equilibrated with 3 x 400 µl of 80 % ACN in 0.1 % TFA followed by 0.1 % TFA in water, respectively. Samples were applied and columns were washed with 3 x 400 µl of 0.1 % TFA in water. The flow-through, which contained free glycans, was collected and evaporated to dryness.

Enzymatic release and isolation of cell surface N-glycans

Approximately 4x106 cells were used for the digestion of cell surface (glyco)proteins. Cells were harvested and washed four times with PBS (15 min at 600 rpm). The resulting cell pellet was resuspended in 500 µl cold PBS containing trypsin (2.5 mg/mL) and shaked at 150 rpm for 15

7

ACS Paragon Plus Environment

Journal of Proteome Research

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

min at 37 °C. Samples were then centrifuged at 15000 rpm and 4 °C for 15 min, and the supernatant, containing the cell surface glycopeptides, was separated from the pellet. Trypsin was then deactivated at 99 °C for 5 min. Tryptic glycopeptides were digested overnight at 37° C with 0.5 U of PNGase F. Released N-glycans were isolated from the peptide moieties using C18 Extract-CleanTM cartridges (Alltech, Deerfield, IL). N-Glycans were subsequently desalted using carbograph Extract-CleanTM columns (Alltech, Deerfield, IL) and evaporated to dryness.

Exoglycosidase digestions

Exoglycosidase digestions were performed to verify the monosaccharide type and linkage. Nglycans, dissolved in sodium acetate (100 mM, pH 5.0), were digested for 18 h at 37 °C using different exoglycosidases with the following concentrations: 3 U/mL Arthrobacter ureafaciens neuraminidase (Roche Applied Science, Indianapolis, IN); 0.8 U/mL β(1–4) galactosidase from Streptococcus pneumoniae (GKX-5014, Prozyme); 0.75 U/mL bovine kidney α(1–2,3,4,6) fucosidase (GKX-5006, Prozyme); 5.4 mU/mL almond meal α(1–3,4) fucosidase (GKX-5019, Prozyme); 12 U/mL β-N-acetylhexosaminidase, recombinant from Streptococcus pneumoniae, expressed in E. coli (GKX-80050, Prozyme, CA). After inhibition at 95 °C for 5 min, samples were desalted on self-made graphite micro-columns and lyophilized. The presence of LewisX epitopes and core-fucosylated structures was shown using the following sequence of digestions: β(1–4) galactosidase, α(1–3,4) fucosidase, β(1–4) galactosidase, and bovine kidney α(1-2,3,4,6) fucosidase. Data are presented in Supplementary Data Table S2 to S5.

Permethylation and MALDI-TOF mass spectrometry

8

ACS Paragon Plus Environment

Page 8 of 29

Page 9 of 29

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

Journal of Proteome Research

Permethylation was performed in DMSO using sodium hydroxide and methyl iodide as described elsewhere 21, 22. Permethylated cell surface N-glycans were further desalted with C18 ExtractCleanTM cartridges (Alltech, Deerfield, IL). N-Glycans were analyzed on an Ultraflex III TOF/TOF mass spectrometer (Bruker Daltonics, Bremen, Germany) equipped with a smartbeamIITM laser and a LIFT-MS/MS facility. After a delayed extraction time of 10 ns, the ions were accelerated with a 25 kV voltage. Measurements were carried out in the positive-ionization mode. External calibration was performed using a dextran ladder. Samples (0.5 µl) were mixed on a ground steel target in a 1:1 ratio (v/v) with the matrix consisting of sDHB(10 mg/mL) dissolved in 10 % acetonitrile. Structural assignments were confirmed by MALDI-TOF/TOF fragmentation (Supplementary Figure S1). Polyhexose contamination, when present, was negligible and excluded from the quantification.

Results

Analysis of cell surface N-glycosylation using the cell membrane preparation protocol

About 4x106 HEK 293 cells were used to investigate the cell surface N-glycome. Membrane (glyco)proteins were isolated from cell lysates by membrane extraction after three centrifugation steps and a lipid separation step. Membrane glycoproteins were subsequently digested with trypsin and N-glycans were then released using PNGase F. After purification using reversedphase C18 and carbograph cartridges, N-glycans were permethylated prior to MALDI-TOF-MS (Supplementary Data Figure S4). A representative MALDI-TOF mass spectrum is shown in Fig. 1A and the average relative abundances derived from four biological replicates are presented in

9

ACS Paragon Plus Environment

Journal of Proteome Research

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

Supplementary Table S1. The mass spectrometric data revealed about 80 different N-glycan structures; the most abundant N-glycan fraction was the high-mannose-type (83 %), followed by the complex-type N-glycans (15 %) and trace amounts of hybrid N-glycans (2 %) were detected as well. Complex-type N-glycans predominantly bore two antennae, however, tri- and tetraantennary complex N-glycans were detected as well. Moreover, they were found to contain 0-3 fucose residues and in most of the cases they were only partially sialylated.

Analysis of cell surface N-glycosylation using the new method (cell surface protocol)

With the aim of releasing only cell surface N-glycans, HEK 293 cells were treated with trypsin in order to digest solely cell surface (glyco)proteins. About 4x106 cells were harvested and washed three times with PBS. The cell pellet was resuspended in 500 µl of freshly prepared 2.5 mg/mL trypsin solution in PBS, which is the trypsin concentration normally used to detach adherent cells from a cell culture dish in order to maintain the cells intact and digestion was carried out for 15 min at 37 °C under slight agitation. For other cell lines, trypsin concentration as well as incubation time may be adjusted accordingly. The cell viability was not different before or after tryptic digestion (2.5 mg/mL, 15 min at 37 °C): HEK 293= 86 %, CHO-K1= 77 %, Hep G2= 91 % viability.

The sample was then centrifuged and the supernatant, which contains cell surface (glyco)peptides, was separated from the pellet. N-glycans were released from glycopeptides using PNGase F, purified using reversed-phase C18 then carbograph cartridges and permethylated prior to MALDI-TOF mass spectrometry (Supplementary Data Figure S4). NGlycans were measured by MALDI-TOF-MS in the positive ionization mode. A representative mass spectrum of cell surface N-glycans of HEK 293 is shown in Figure 1B and the average 10

ACS Paragon Plus Environment

Page 10 of 29

Page 11 of 29

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

Journal of Proteome Research

relative abundance of each detected structure derived from four biological replicates is summarized in Supplementary Table S1. In total, 95 different N-glycan signals were detected. The relative amount of high-mannose N-glycans dropped drastically to 14 % (Supplementary Figure S6), the amount of complex N-glycans increased to 85 % and the amount of hybrid-type N-glycans remained almost the same when compared with the membrane extraction method. In the cell surface N-glycosylation profile obtained from the new method, the presence of 25 additional signals were observed and were assigned to complex-type structures, which accounted for about 7 % of the N-glycan pool and were not observed when the membrane preparation method was applied. Structures were verified by MALDI-TOF/TOF and exoglycosidase digestions (Supplementary Data Figure S1 and Table S2). Interestingly, two of these structures contained the N-acetylgalactosamine (GalNAc) epitope (m/z 2499.9 and m/z 3007.1). Most of the structures were biantennary bi- or trifucosylated N-glycans containing the LewisX epitope and were relatively low in abundance but however, they could not be detected in the N-glycome of HEK 293 obtained from the membrane preparation method. In addition, the high overlap of complex-type N-glycans detected in HEK 293 cells with the membrane preparation and the newly established cell surface protocol is depicted in Supplementary Figure S5. Thus, in comparison with the membrane preparation method, the new method led to an improved sensitivity of detection and therefore a better quantification of complex-type N-glycans.

We tested the selectivity of the tryptic digestion towards cell surface proteins by analyzing the Nglycome of the HEK 293 pellet, which was obtained after tryptic digestion of the cells and subsequent centrifugation. (Glyco)proteins were isolated from this pellet using the membrane preparation protocol and N-glycans were released as described in the first paragraph. We found

11

ACS Paragon Plus Environment

Journal of Proteome Research

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

that the profile was predominantly consisting of high-mannose N-glycans (about 94 %) (Supplementary Figure S2).

Reproducibility

In order to assess the reproducibility of the new method, both protocols were carried out four times. Figure 2 presents the relative amounts of the 27 most abundant structures found in the Nglycosylation profile of HEK 293 cells using both methods. Both N-glycan profiles were highly reproducible as observed from the small deviations obtained within the observed glycan profiles. Although the rest of the structures were present in small amounts, low standard deviations were observed among the four repetitions. While the N-glycan distribution varied with the methods employed, little variation was noticed among biological replicates, showing that the new method is as reproducible as the membrane preparation protocol.

Comparison of cell surface N-glycosylation of AGE1.HN, CHO-K1 and Hep G2 cells using the cell surface protocol and the membrane preparation protocol

The established and optimized method was applied to profile the cell surface N-glycosylation pattern of the AGE1.HN, CHO-K1 and Hep G2 cell lines (Supplementary Data Figure S3 and Table S1). The relative amounts of complex-type N-glycans found in HEK 293, AGE1.HN, CHO-K1 and Hep G2 in case of the cell surface and membrane preparation protocol, which were calculated from relative peak areas of N-glycan signals detected by MALDI-TOF-MS, are summarized in Table 1. It was possible to obtain between 55% and 85% of complex-type structures (Table 1) and to detect 25 additional structures in HEK 293, 48 in AGE1.HN, 42 in CHO-K1 and 51 in Hep G2 cells when compared to the membrane extraction protocol. Structures

12

ACS Paragon Plus Environment

Page 12 of 29

Page 13 of 29

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

Journal of Proteome Research

were verified by MALDI-TOF/TOF and exoglycosidase digestions (Supplementary Data Figure S1 and Table S2-S5). For all the cell lines studied here, the amount of hybrid-type N-glycans was in all cases minimal (about 4 %) and neutral, monosialylated and/or monofucosylated structures were detected. N-Glycans stemming from FCS, namely S3H6N5 at m/z 3602.1 and S4H6N5 at m/z 3964.2, were observed as minor amounts in all the cell lines (Supplementary Data Table S1).

About 85 different N-glycan signals, representing 78% of the glycan pool, were detected in the MALDI-TOF-MS spectrum of AGE1.HN cells generated by the new approach whereas only 37 signals, representing 33% of the glycan pool, were detected when the membrane preparation protocol was applied (Table 1). The additional structures detected by the new approach, which were mainly of the complex-type, also included N-glycans bearing 2-4 fucose residues containing the LewisX epitope in AGE1.HN cells.

The N-glycome of CHO-K1 cells contained about 87 % high-mannose N-glycans in case of the membrane preparation protocol and 42 % when the new approach was applied. Complex Nglycans comprised about 11 % of the total N-glycan pool obtained from the membrane preparation protocol and 55 % in case of the new approach (Table 1). In case of the new approach, it was possible to detect 48 supplementary structures that accounted together about 7 % of the N-glycan pool and were mainly complex-type N-glycans. These structures included the N-glycans G1H5N4 (m/z 2460.9), S1G1H5N4 (m/z 2822.1), G2H5N4 (m/z 2852.1), S1G1H5N4F1 (m/z 2996.1) and G1H6N5F1 (m/z 3084.1), which contain terminal Nglycolylneuraminic acid (Neu5Gc) and were not detected in the N-glycome obtained by the membrane preparation protocol. In addition, additional structures included complex-type N-

13

ACS Paragon Plus Environment

Journal of Proteome Research

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

glycans bearing digalactosylated antennas (H7N5F1 at m/z 2897.1, S2H7N5F1 at m/z 3620.1 and S2H8N6F1 at m/z 4069.3). Most of the complex N-glycans were present as asialylated or monosialylated structures, but some fully sialylated triantennary N-glycans were detected as well. N-glycolylneuraminic acid was present and comprised about 3 % of the N-glycan pool of CHO-K1 and was only detected with the new approach.

Using the cell surface protocol, 51 additional structures were detected in the N-glycome of Hep G2 cells and the complex-type N-glycan fraction increased from 15 % (membrane protocol) to 82 % (cell surface protocol) (Table 1). Biantennary N-glycans comprised about 55 % of the total relative intensity and most of them were fucosylated (1-3 Fuc) but non-fucosylated biantennary N-glycans were detected as well. About 6 % of the signals corresponded to triantennary Nglycans (0-3 Fuc, 0-4 Neu5Ac). A minor amount of tetraantennary and larger N-glycans was detected as well. The supplementary structures detected with the new approach were mainly bi-, tri- and tetraantennary complex-type N-glycans. They included important epitopes such as GalNAc (H4N5F1 at m/z 2284.8, H3N6F1 at m/z 2325.8, S2H4N5F1 at m/z 3007.1) and LewisX in bi-, tri- and tetraantennary N-glycans.

Discussion

Cell surface N-glycans are directly involved in cell-cell or cell-protein interactions that trigger various biological responses. It is therefore important to have robust methods to characterize them. In this study, we developed a simple and fast glycomics method that allows the specific analysis of cell surface N-glycosylation from living cells, which was not reported before.

14

ACS Paragon Plus Environment

Page 14 of 29

Page 15 of 29

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

Journal of Proteome Research

The classical cell membrane extraction protocol, used in many laboratories, yields high levels of high-mannose N-glycans 4, 6 that stem from both membrane proteins as well as proteins from the ER. An et al. confirmed with proteomic analysis the presence of not only plasma membrane proteins but also contamination from the ER and other membranes, showing that N-glycans derived from the ER were included as well in the analysis of stem cell N-glycome 4. Nakano et al. 6 isolated cell membrane proteins then dotted them on polyvinylidene difluoride membrane prior to N-glycan release. A high proportion of high-mannose N-glycans were observed, which is not the case with our new approach. Recently, Mun et al. 11 used an adhesion-based method to immobilize plasma membranes prior to N-glycan analysis. This method has the advantage of washing away intracellular high-mannose N-glycans before the release of the N-glycan fraction but cell lysis is performed prior to N-glycan analysis. Hence, in this study, we aimed at developing a simple and reproducible protocol to analyze cell surface N-glycans from living cells and compared it with the classical method, the cell membrane preparation, in order to demonstrate the superior efficiency of the newly developed method.

Using this protocol we achieved a dramatic decrease of the amount of high-mannose N-glycans from 82 % to 10 % for HEK 293 cells. As a result, the detection of complex-type N-glycans was drastically improved when compared to the membrane preparation method. With the new protocol, we were able to detect 94 different N-glycan structures in HEK 293. More than 20 of them were complex-type N-glycans that were absent from the N-glycosylation profile of HEK 293 cells obtained from the membrane preparation method. When compared with Nakano et al. 6 and Reinke et al. 13 who both used membrane extraction to isolate glyco(proteins), we could detect nearly twice the number of structures. In addition, the abundance of high-mannose Nglycans in the study of Reinke et al. was remarkably high for HEK 293T cells that were used for

15

ACS Paragon Plus Environment

Journal of Proteome Research

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

this study 13. Thus, our method is more suitable for cell surface glycosylation profiling since a higher number of complex-type N-glycans are present and their detection is enhanced.

The new approach requires about 4x106 cells, which is lower when compared with amounts used by other groups for cell membrane extraction, namely from 50 to 100x106 cells 4-6. In addition, with the new approach, cell surface glycopeptides are available in about an hour starting from the cell pellet instead of 2 working days with our cell membrane extraction protocol. This time is comparable with the time needed by Nakano et al. and An et al. to isolate and denature membrane proteins 4, 6. In addition, Mi et al. 5 described an even more complex and timeconsuming method to obtain glycopeptides from cell membranes. Not only was the time of preparation increased but also the costs for N-glycan analysis through cell membrane extraction were higher since additional consumables and centrifugation steps were required. The new approach was applied to profile the cell surface N-glycosylation pattern of the HEK 293, AGE1.HN, CHO-K1 and Hep G2 cell lines and the results were compared to the profiles obtained from the preparation of the same cells with the membrane extraction protocol.

In this study, exoglycosidase digestions and MALDI-TOF/TOF fragmentations revealed the presence of core-fucosylation and of the GalNAc-GlcNAc motif in HEK 293 cells. These results are in line with previous studies of Reinke et al. 23, in which our group reported the high expression level of the glycosyltransferase FucT-VIII and a high content of core-fucose in HEK 293T cells. Furthermore, earlier studies revealed that L-selectin and human protein C expressed in HEK 293 cells contained the GalNAc motif as well 24 25, suggesting that the HEK 293 cell line is able to produce GalNAc epitopes on glycoproteins.

16

ACS Paragon Plus Environment

Page 16 of 29

Page 17 of 29

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

Journal of Proteome Research

The core-fucosylation and traces of the antigenic Neu5Gc found in this study in the N-glycome of CHO-K1 cells are in accordance with the literature published on the N-glycome of recombinant glycoproteins expressed in CHO-K1 cells 26, 27, namely anti-HER2 monoclonal antibodies 28, human plasminogen activator 29, human alpha-1-antitrypsin (A1AT) 30 and human erythropoietin 31. The characteristic feature found in AGE1.HN cells was the high degree of core and antennary fucosylation, which corroborates our previous findings on recombinant A1AT 32. A1AT expressed in AGE1.HN contained core-fucosylated di-, tri and tetraantennary N-glycans. In addition, up to four fucose residues were found on triantennary N-glycans 32. Moreover, studies of Rudd et al. and Stimson et al. also suggest that core-fucosylation and Lex epitopes are characteristic of neuronal N-glycosylation, since their work demonstrated the presence of the two types of fucose linkages in N-glycans of the prion protein in mice and hamsters 33, 34. Core-fucosylation was found to be a singular trait of Hep G2 cells. This finding is in line with hepatoma cell glycoprotein glycosylation, in which α(1–6)-linked core-fucose has been reported on α-fetoprotein and transferrin 35, 36. In parallel, Noda et al. successfully demonstrated that the α(1–6)-fucosyltransferase gene was expressed at high levels in hepatomas of rat models and hepatoma cell lines 37. Finally, the presence of GalNAc in glycans derived from hepatoma was described as well by Johnson et al. 38 on alpha-fetoprotein O-glycans in hepatocellular carcinoma patients. Altogether, the cell surface N-glycome of the four cell lines studied here, namely HEK293, CHO-K1, AGE1.HN and Hep G2, is in line with the N-glycome found on recombinant proteins produced with these cell lines. In all, our data are in line with the literature published on the N-glycome of recombinant glycoproteins expressed in the cell lines studied here.

17

ACS Paragon Plus Environment

Journal of Proteome Research

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

To sum up, we established a new glycomics method to profile the N-glycome of living cells that is fast, robust, sensitive and reproducible. In addition, we were able to detect an increased number and amount of complex type N-glycans with the new protocol. Information about cell surface N-glycan structures are of great importance as they can give important structural information about the possible N-glycan profiles of recombinant glycoproteins, which is of relevance to glycobiotechnologists. In addition, cell surface profiling will be of help not only to study the function of glycans in cells but in the fields of glycobiotechnology and biomarker discovery.

Acknowledgment

This study was supported by grants from the Investitionsbank Berlin and the European Regional Development Fund (grant no: 10147244).

Abbreviations

AEM: Adenovirus expression medium; CHO: chinese hamster ovary; ER: endoplasmic reticulum; DMEM: Dulbecco´s modified eagle medium; DMSO: dimethyl sulfoxide; DSMZ: German Collection of Microorganisms and Cell Cultures; FCS: fetal calf serum; GalNAc: Nacetylgalactosamine; GlcNAc: N-acetylglucosamine; HEK: human embryonic kidney; MALDI: matrix-assisted laser-desorption ionization; MS: mass spectrometry; Neu5Ac: Nacetylneuraminic acid; Neu5Gc: N-glycolylneuraminic acid; PBS: phosphate-buffered saline; 18

ACS Paragon Plus Environment

Page 18 of 29

Page 19 of 29

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

Journal of Proteome Research

PNGaseF: peptide-N4-(N-acetyl-β-glucosaminyl) asparagine amidase F; SDS: sodium dodecyl sulfate; TFA, trifluoroacetic acid; TOF: time-of-flight.

References

1. Spiro, R. G., Protein glycosylation: nature, distribution, enzymatic formation, and disease implications of glycopeptide bonds. Glycobiology 2002, 12, (4), 43R-56R. 2. Rademacher, T. W.; Parekh, R. B.; Dwek, R. A., Glycobiology. Annual Review of Biochemistry 1988, 57, (1), 785-838. 3. Tateno, H.; Uchiyama, N.; Kuno, A.; Togayachi, A.; Sato, T.; Narimatsu, H.; Hirabayashi, J., A novel strategy for mammalian cell surface glycome profiling using lectin microarray. Glycobiology 2007, 17, (10), 1138-1146. 4. An, H. J.; Gip, P.; Kim, J.; Wu, S.; Park, K. W.; McVaugh, C. T.; Schaffer, D. V.; Bertozzi, C. R.; Lebrilla, C. B., Extensive determination of glycan heterogeneity reveals an unusual abundance of high mannose glycans in enriched plasma membranes of human embryonic stem cells. Molecular & Cellular Proteomics 2012, 11, (4). 5. Mi, W.; Jia, W.; Zheng, Z.; Wang, J.; Cai, Y.; Ying, W.; Qian, X., Surface glycoproteomic analysis of hepatocellular carcinoma cells by affinity enrichment and mass spectrometric identification. Glycoconjugate Journal 2012, 29, (5-6), 411-424. 6. Nakano, M.; Saldanha, R.; Göbel, A.; Kavallaris, M.; Packer, N. H., Identification of glycan structure alterations on cell membrane proteins in desoxyepothilone B resistant leukemia cells. Molecular & Cellular Proteomics 2011, 10, (11). 7. Whelan, S. A.; Lu, M.; He, J.; Yan, W.; Saxton, R. E.; Faull, K. F.; Whitelegge, J. P.; Chang, H. R., Mass spectrometry (LC-MS/MS) site-mapping of N-glycosylated membrane proteins for breast cancer biomarkers. Journal of Proteome Research 2009, 8, (8), 4151-4160. 8. Sun, B.; Ma, L.; Yan, X.; Lee, D.; Alexander, V.; Hohmann, L. J.; Lorang, C.; Chandrasena, L.; Tian, Q.; Hood, L., N-Glycoproteome of E14.Tg2a mouse embryonic stem cells. PLoS ONE 2013, 8, (2), e55722. 9. Vercoutter-Edouart, A.-S.; Slomianny, M.-C.; Dekeyzer-Beseme, O.; Haeuw, J.-F.; Michalski, J.-C., Glycoproteomics and glycomics investigation of membrane N-glycosylproteins from human colon carcinoma cells. Proteomics 2008, 8, (16), 3236-3256. 10. Sethi, M. K.; Thaysen-Andersen, M.; Smith, J. T.; Baker, M. S.; Packer, N. H.; Hancock, W. S.; Fanayan, S., Comparative N-glycan profiling of colorectal cancer cell lines reveals unique bisecting GlcNAc and alpha-2,3-linked sialic acid determinants are associated with membrane proteins of the more metastatic/aggressive cell lines. J Proteome Res 2014, 13, (1), 277-88. 11. Mun, J.-Y.; Lee, K. J.; Seo, H.; Sung, M.-S.; Cho, Y. S.; Lee, S.-G.; Kwon, O.; Oh, D.-B., Efficient adhesion-based plasma membrane isolation for cell surface N-glycan analysis. Analytical Chemistry 2013, 85, (15), 7462-7470. 12. North, S. J.; von Gunten, S.; Antonopoulos, A.; Trollope, A.; MacGlashan, D. W., Jr.; Jang-Lee, J.; Dell, A.; Metcalfe, D. D.; Kirshenbaum, A. S.; Bochner, B. S.; Haslam, S. M., Glycomic analysis of human mast cells, eosinophils and basophils. Glycobiology 2012, 22, (1), 12-22.

19

ACS Paragon Plus Environment

Journal of Proteome Research

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

13. Reinke, S. O.; Bayer, M.; Berger, M.; Blanchard, V.; Hinderlich, S., Analysis of Cell Surface N-glycosylation of the human embryonic kidney 293T cell line. Journal of Carbohydrate Chemistry 2011, 30, (4-6), 218-232. 14. Hamouda, H.; Ullah, M.; Berger, M.; Sittinger, M.; Tauber, R.; Ringe, J.; Blanchard, V., N-Glycosylation profile of undifferentiated and adipogenically differentiated human bone marrow mesenchymal stem cells: towards a next generation of stem cell markers. Stem Cells and Development 2013, 22, (23), 3100-13. 15. Lin, C.-H.; Kuo, C.-W.; Jarvis, D. L.; Khoo, K.-H., Facile removal of high mannose structures prior to extracting complex type N-glycans from de-N-glycosylated peptides retained by C18 solid phase to allow more efficient glycomic mapping. Proteomics 2014, 14, (1), 87-92. 16. Cao, J.-T.; Chen, Z.-X.; Hao, X.-Y.; Zhang, P.-H.; Zhu, J.-J., Quantum dots-based immunofluorescent microfluidic chip for the analysis of glycan expression at single-cells. Analytical Chemistry 2013, 84, (22), 10097-10104. 17. Hang, H. C.; Yu, C.; Kato, D. L.; Bertozzi, C. R., A metabolic labeling approach toward proteomic analysis of mucin-type O-linked glycosylation. Proceedings of the National Academy of Sciences 2003, 100, (25), 14846-14851. 18. Hsu, T.-L.; Hanson, S. R.; Kishikawa, K.; Wang, S.-K.; Sawa, M.; Wong, C.-H., Alkynyl sugar analogs for the labeling and visualization of glycoconjugates in cells. Proceedings of the National Academy of Sciences 2007, 104, (8), 2614-2619. 19. Li, B.; Mock, F.; Wu, P.; Conn, P. M., Chapter twenty-one - imaging the glycome in living systems. In Methods in Enzymology, Academic Press: 2012; Vol. Volume 505, pp 401-419. 20. Vocadlo, D. J.; Hang, H. C.; Kim, E.-J.; Hanover, J. A.; Bertozzi, C. R., A chemical approach for identifying O-GlcNAc-modified proteins in cells. Proceedings of the National Academy of Sciences 2003, 100, (16), 9116-9121. 21. Dell, A.; Reason, A. J.; Khoo, K.-H.; Panico, M.; McDowell, R. A.; Morris, H. R., [8] Mass spectrometry of carbohydrate-containing biopolymers. In Methods in Enzymology, William J. Lennarz, G. W. H., Ed. Academic Press: 1994; Vol. Volume 230, pp 108-132. 22. Wada, Y.; Azadi, P.; Costello, C. E.; Dell, A.; Dwek, R. A.; Geyer, H.; Geyer, R.; Kakehi, K.; Karlsson, N. G.; Kato, K.; Kawasaki, N.; Khoo, K.-H.; Kim, S.; Kondo, A.; Lattova, E.; Mechref, Y.; Miyoshi, E.; Nakamura, K.; Narimatsu, H.; Novotny, M. V.; Packer, N. H.; Perreault, H.; Peter-Katalinić, J.; Pohlentz, G.; Reinhold, V. N.; Rudd, P. M.; Suzuki, A.; Taniguchi, N., Comparison of the methods for profiling glycoprotein glycans—HUPO Human Disease Glycomics/Proteome Initiative multi-institutional study. Glycobiology 2007, 17, (4), 411-422. 23. Reinke, S. O.; Bayer, M.; Berger, M.; Blanchard, V. r.; Hinderlich, S., Analysis of cell surface N-glycosylation of the human embryonic kidney 293T cell line. Journal of Carbohydrate Chemistry 30, (4-6), 218-232. 24. Yan, S. B.; Chao, Y. B.; van Halbeek, H., Novel Asn-linked oligosaccharides terminating in GaINAc beta (1-4)[Fuc alpha(1-3)]GlcNAc beta(1-) are present in recombinant human Protein C expressed in human kidney 293 cells. Glycobiology 1993, 3, (6), 597-608. 25. Wedepohl, S.; Kaup, M.; Riese, S. B.; Berger, M.; Dernedde, J.; Tauber, R.; Blanchard, V. r., N-Glycan analysis of recombinant L-selectin reveals sulfated GalNAc and GalNAc−GalNAc motifs. Journal of Proteome Research 2010, 9, (7), 3403-3411. 26. Costa, A. R.; Withers, J.; Rodrigues, M. E.; McLoughlin, N.; Henriques, M.; Oliveira, R.; Rudd, P.; Azeredo, J., The impact of microcarrier culture optimization on the glycosylation profile of a monoclonal antibody. SpringerPlus 2013, 2, (1), 25. 27. Borys, M. C.; Dalal, N. G.; Abu-Absi, N. R.; Khattak, S. F.; Jing, Y.; Xing, Z.; Li, Z. J., Effects of culture conditions on N-glycolylneuraminic acid (Neu5Gc) content of a recombinant fusion protein produced in CHO cells. Biotechnology and Bioengineering 2010, 105, (6), 10481057. 28. Ho, S. C. L.; Koh, E. Y. C.; van Beers, M.; Mueller, M.; Wan, C.; Teo, G.; Song, Z.; Tong, Y. W.; Bardor, M.; Yang, Y., Control of IgG LC:HC ratio in stably transfected CHO cells and

20

ACS Paragon Plus Environment

Page 20 of 29

Page 21 of 29

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

Journal of Proteome Research

study of the impact on expression, aggregation, glycosylation and conformational stability. Journal of Biotechnology 2013, 165, (3–4), 157-166. 29. Bergwerff, A. A.; van Oostrum, J.; Asselbergs, F. A. M.; BÜRgi, R.; Hokke, C. H.; Kamerling, J. P.; Vliegenthart, J. F. G., Primary structure of N-linked carbohydrate chains of a human chimeric plasminogen activator K2tu-PA expressed in Chinese hamster ovary cells. European Journal of Biochemistry 1993, 212, (3), 639-656. 30. Lee, K.; Lee, S.; Gil, J.; Kwon, O.; Kim, J.; Park, S.; Chung, H.-S.; Oh, D.-B., N-glycan analysis of human α1-antitrypsin produced in Chinese hamster ovary cells. Glycoconjugate Journal 2013, 30, (5), 537-547. 31. Hokke, C. H.; Bergwerff, A. A.; Van Dedem, G. W. K.; Kamerling, J. P.; Vliegenthart, J. F. G., Structural analysis of the sialylated N- and O-linked carbohydrate chains of recombinant human erythropoietin expressed in Chinese hamster ovary cells. European Journal of Biochemistry 1995, 228, (3), 981-1008. 32. Blanchard, V.; Liu, X.; Eigel, S.; Kaup, M.; Rieck, S.; Janciauskiene, S.; Sandig, V.; Marx, U.; Walden, P.; Tauber, R.; Berger, M., N-glycosylation and biological activity of recombinant human alpha1-antitrypsin expressed in a novel human neuronal cell line. Biotechnology and Bioengineering 108, (9), 2118-2128. 33. Rudd, P. M.; Endo, T.; Colominas, C.; Groth, D.; Wheeler, S. F.; Harvey, D. J.; Wormald, M. R.; Serban, H.; Prusiner, S. B.; Kobata, A.; Dwek, R. A., Glycosylation differences between the normal and pathogenic prion protein isoforms. Proceedings of the National Academy of Sciences 1999, 96, (23), 13044-13049. 34. Stimson, E.; Hope, J.; Chong, A.; Burlingame, A. L., Site-specific characterization of the N-linked glycans of murine Prion protein by high-performance liquid chromatography/electrospray mass spectrometry and exoglycosidase digestions. Biochemistry 1999, 38, (15), 4885-4895. 35. Hutchinson, W. L.; Du, M.-Q.; Johnson, P. J.; Williams, R., Fucosyltransferases: Differential plasma and tissue alterations in hepatocellular carcinoma and cirrhosis. Hepatology 1991, 13, (4), 683-688. 36. Matsumoto, K.; Maeda, Y.; Kato, S.; Yuki, H., Alteration of asparagine-linked glycosylation in serum transferrin of patients with hepatocellular carcinoma. Clinica Chimica Acta 1994, 224, (1), 1-8. 37. Noda, K.; Miyoshi, E.; Uozumi, N.; Gao, C.-X.; Suzuki, K.; Hayashi, N.; Hori, M.; Taniguchi, N., High expression of α-1-6 fucosyltransferase during rat hepatocarcinogenesis. International Journal of Cancer 1998, 75, (3), 444-450. 38. Johnson, P. J.; Poon, T. C. W.; Hjelm, N. M.; Ho, C. S.; Ho, S. K. W.; Welby, C.; Stevenson, D.; Patel, T.; Parekh, R.; Townsend, R. R., Glycan composition of serum alphafetoprotein in patients with hepatocellular carcinoma and non-seminomatous germ cell tumour. Br J Cancer 1999, 81, (7), 1188-1195.

21

ACS Paragon Plus Environment

Journal of Proteome Research

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 and Table legends

Figure 1. MALDI-TOF-MS of PNGase F-released N-glycans from HEK 293 cells. (Glyco)proteins were isolated from HEK 293 using membrane extraction (A) or cell surface glycoproteins were directly released from the cells using trypsin (B). N-Glycans were enzymatically cleaved from tryptic glycopeptides with PNGase F, permethylated and measured by MALDI-TOF-MS. The 60 most abundant structures are depicted. Polyhexose contaminations are marked with asterisks and are negligible. Black square represents N-acetylglucosamine, grey square N-acetylgalactosamine, white square N-acetylhexosamine, dark grey circle mannose, light grey circle galactose, black triangle fucose, white diamond N-glycolylneuraminic acid and black diamond N-acetylneuraminic acid.

Figure 2. Reproducibility of N-glycan isolation using membrane glycoprotein extraction (black) and direct digestion of cell surface glycoproteins (grey). The average relative intensity of the 27 most abundant structures from MALDI-TOF-MS are shown (n=4). Black square represents Nacetylglucosamine, grey square N-acetylgalactosamine, white square N-acetylhexosamine, dark grey circle mannose, light grey circle galactose, black triangle fucose, white diamond Nglycolylneuraminic acid and black diamond N-acetylneuraminic acid.

22

ACS Paragon Plus Environment

Page 22 of 29

Page 23 of 29

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

Journal of Proteome Research

Table 1. Relative amounts of PNGase F-released complex-type N-glycans found in HEK 293, CHO-K1, AGE1.HN and HEP G2 cells using the membrane preparation (MP) and cell surface (CS) protocols as judged by MALDI-TOF-MS using relative peak areas.

Supporting Information - this material is available free of charge via http://pubs.acs.org.

Supplementary Information

Figure S1. MALDI-TOF/TOF mass spectra of: H4N5F1 (m/z 2284.8) from HEK 293 cells (A), G1H5N4F1 (m/z 2634.9) from CHO-K1 cells (B), H7N6F4 (m/z 3665.1) from AGE1.HN cells (C) and H8N7F1 (m/z 3592.1) from HEP G2 cells (D).

Figure S2. MALDI-TOF-MS of N-glycans obtained using the membrane extraction protocol in the pellet obtained after cell surface tryptic digestion of HEK 293 cells. Cell surface (glyco)proteins were directly released from living HEK 293 cells using trypsin. The supernatant containing cell surface (glyco)peptides was separated from the pellet and the pellet was washed three times with PBS followed by a (glyco)protein isolation using membrane extraction. (Glyco)proteins were digested with trypsin and N-glycans were enzymatically cleaved with PNGase F, permethylated and measured by MALDI-TOF-MS.

Figure S3. MALDI-TOF-MS of PNGase F-released N-glycans from HEK 293 (A), CHO-K1 (B), AGE1.HN (C) and HEP G2 cells (D). Cell surface glycoproteins were directly digested from the cells using the newly established method. N-glycans were enzymatically cleaved from glycopeptides with PNGase F, permethylated and measured by MALDI-TOF-MS. Only the most abundant structures are depicted in the Figure.

23

ACS Paragon Plus Environment

Journal of Proteome Research

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 S4. Analytical workflow used for the cell membrane preparation (left) and cell surface (right) protocol.

Figure S5. Number and overlap of complex-type N-glycans detected in HEK 293, CHO-K1, AGE1.HN and Hep G2 cells using the membrane preparation as well as cell surface protocol.

Table S1. Average relative amounts of all PNGase F-released N-glycans of HEK-293 from cell membrane preparation (MP) and cell surface (CS) as well as from CHO-K1, AGE1.HN and HepG2 cells (4 independent preparations for HEK-293 by MP, 3 independent preparations for HEK-293 , CHO-K1, AGE1.HN and HepG2 by CS) as judged by MALDI-TOF-MS. H: hexose, N: N-acetylhexosamine, F: deoxyhexose, S: N-acetylneuraminic acid, G: N-glycolylneuraminic acid.

Table S2. Exoglycosidase digestions of PNGase F-released N-glycans derived from HEK 293 cells.

Table S3. Exoglycosidase digestions of PNGase F-released N-glycans derived from AGE.HN cells.

Table S4. Exoglycosidase digestions of PNGase F-released N-glycans derived from CHO-K1 cells.

Table S5. Exoglycosidase digestions of PNGase F-released N-glycans derived from Hep G2 cells.

24

ACS Paragon Plus Environment

Page 24 of 29

Page 25 of 29

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

Journal of Proteome Research

Table S6. Average relative amounts of PNGase F-released high-mannose type N-glycans of HEK-293 from cell membrane preparation (MP) and cell surface (CS) (n=4) as judged by MALDI-TOF-MS. H: hexose, N: N-acetylhexosamine.

25

ACS Paragon Plus Environment

Journal of Proteome Research

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

MP CS

% in HEK 293 14 85

% in AGE1.HN 33 78

% in CHO-K1 11 55

% in Hep G2 15 82

Table 1. Relative amounts of PNGase F-released complex-type N-glycans found in HEK 293, CHO-K1, AGE1.HN and Hep G2 cells using the membrane preparation (MP) and cell surface (CS) protocols as judged by MALDI-TOF-MS using relative peak areas.

26

ACS Paragon Plus Environment

Page 26 of 29

2314.8

3590.1

3000 3500 4000

4000

Figure 1

ACS Paragon Plus Environment 4587.4

4226.3

3864.2

3500

4587.4

4226.3

2674.1

3500

4039.3

3964.2

3776.2

2518.9 2599.9 2792.1 2820.1 2850.1

3000

3864.2

2634.9 2645.9

3504.1

3416.1

3228.1

3142.1

2417.9 2458.9 2488.9

4500

4587.4

4226.3

3864.2

3504.1

2395.9

2191.8

Intensity (%)

1783.7

2430.9

3776.2

3054.1

2779.1

2674.1

3241.1

0

3776.2

3661.1

3504.1

3000

3602.1

3416.1

3474.1

2966.1 3054.1

2645.9

2634.9

2500

3474.1

3457.1

3299.1

3241.1

3228.1

2880.1

2779.1

2604.9

2389.9

50

3211.1

2966.1

2820.1

2604.9

2430.9

2314.8

2243.8

2185.8

1987.7

1579.6

4000

4500 4500

2850.1

2820.1

2792.1

2599.9

2518.9

2488.9

2458.9

2417.9

2674.1

3416.1

3142.1

2966.1

2880.1

2693.1

2604.9

50

3112.1

3054.1

2663.1

2850.1

2779.1

2634.9

2458.9

2395.9 2500

3007.1

2880.1

2792.1

2645.9 2674.1

2243.8

2028.8

1835.7

1590.6

1375.6

1345.5

1171.5

100

3024.1

2500

2996.1

2938.1

2734.1

2000

2499.9

2430.9 2488.9

2417.9

2191.8

2069.8

2000

2389.9

2359.9

2325.8

1500 1987.7

1500

2284.8

1783.7

1579.6

0

2213.8

1835.7

1590.6

50

2080.8

2039.8

1416.6

1171.5

Intensity (%)

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 Intensity (%)

Page 27 of 29 Journal of Proteome Research

A

100

Mass (m/z)

Mass (m/z)

100

B

0 Mass (m/z)

Figure 2

ACS Paragon Plus Environment S4H7N6F1

S3H7N6F1

S3H6N5F1

S3H6N5

S2H6N5F1

S1H6N5F1

S2H5N4F1

S1H5N5F1

S1H4N5F2

H3N6F3

S1H4N5F1

S1H6N4

S1H5N4F1

H3N6F2

H5N5F1

H4N5F2

4587.4

4226.3

3776.2

3602.1

3416.1

3054.1

2850.1

2674.1

2645.9

2634.9

2499.9

2488.9

8 2966.1

2820.1

2458.9

2430.9

2243.8

2604.9

2395.9

2191.8

25

S1H5N4

H9N2

2284.8

6

H4N5F1

H5N4F1

1987.7

1783.7

20

H8 N 2

2069.8

4

H 5N 4

H7N2

H3N4F1

1835.7

10

H6N2

12

1579.6

%

H5N2

1171.5

2

H 3 N2

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 Intensity (%)

Journal of Proteome Research Page 28 of 29

HEK 293 membrane glycoproteins HEK 293 cell surface glycoproteins

30

15

14

0

Page 29 of 29

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

Journal of Proteome Research

   

Graphical  abstract    

ACS Paragon Plus Environment