N-linked Glycosylation Profiling of Pancreatic Cancer Serum Using Capillary Liquid Phase Separation Coupled with Mass Spectrometric Analysis Jia Zhao,† Weilian Qiu,† Diane M. Simeone,‡ and David M. Lubman*,†,§,| Department of Chemistry, The University of Michigan, Ann Arbor, Michigan 48109-1055, Department of Surgery and Molecular Integrative Physiology, The University of Michigan Medical Center, Ann Arbor, Michigan 48109-0656, Comprehensive Cancer Center, University of Michigan Medical Center, Ann Arbor, Michigan, and Department of Surgery, The University of Michigan Medical Center, Ann Arbor, Michigan 48109-0656 Received August 30, 2006
Glycoproteins play important roles in various biological processes including intracellular transport, cell recognition, and cell-cell interactions. The change of the cellular glycosylation profile may have profound effects on cellular homeostasis and malignancy. Therefore, we have developed a sensitive screening approach for the comprehensive analysis of N-glycans and glycosylation sites on human serum proteins. Using this approach, N-linked glycopeptides were extracted by double lectin affinity chromatography. The glycans were enzymatically cleaved from the peptides and then profiled using capillary hydrophilic interaction liquid chromatography coupled online with ESI-TOF MS. The structures of the separated glycans were determined by MALDI quadrupole ion-trap TOF mass spectrometry in both positive and negative modes. The glycosylation sites were elucidated by sequencing of PNGase F modified glycopeptides using nanoRP-LC-ESI-MS/MS. Alterations of glycosylation were analyzed by comparing oligosaccharide expression of serum glycoproteins at different disease stages. The efficiency of this method was demonstrated by the analysis of pancreatic cancer serum compared to normal serum. Ninety-two individual glycosylation sites and 202 glycan peaks with 105 unique carbohydrate structures were identified from ∼25 µg glycopeptides. Forty-four oligosaccharides were found to be distinct in the pancreatic cancer serum. Increased branching of N-linked oligosaccharides and increased fucosylation and sialylation were observed in samples from patients with pancreatic cancer. The methodology described in this study may elucidate novel, cancer-specific oligosaccharides and glycosylation sites, some of which may have utility as useful biomarkers of cancer. Keywords: hydrophilic interaction chromatography • serum • glycan • pancreatic cancer • lectin affinity
1. Introduction Pancreatic carcinoma has the worst prognosis of any cancer, with a 5-year survival rate less than 3%.1 It is the fourth most frequent cause of cancer death in Europe and the U.S.A.2 Despite years of research in this area, there is no reliable method for detection of pancreatic cancer in asymptomatic patients. Thus, there is currently great interest in developing protein-based serum biomarkers for early cancer detection. The glycoproteome is one of the major subproteomes of human serum. Glycosylation plays a crucial role in various biological functions, including intracellular transport, cell * To whom correspondence should be addressed. Dr. David M. Lubman, The University of Michigan Medical Center, Department of Surgery, MSRB1, Rm A510B, 1150 West Medical Center Drive, Ann Arbor, MI 48109-0656. Fax, +1-734-615-2088; E-mail:
[email protected]. † Department of Chemistry. ‡ Department of Surgery and Molecular Integrative Physiology. § Comprehensive Cancer Center. | Department of Surgery.
1126
Journal of Proteome Research 2007, 6, 1126-1138
Published on Web 01/24/2007
recognition, and cell-cell interactions.3-5 Alterations of glycan structures are frequently observed in cancer cells of various tumors types. These alterations include increased glycan branching6-8 and increased Lewis antigen expression.9 The change of the cellular glycosylation profile has been shown to be causally related to aggressive cancer cell behavior such as tumor cell invasion and metastasis.10-12 Therefore, comparative studies of the carbohydrate chains of glycoproteins produced by malignant cells, as compared to their corresponding normal cells, may provide useful information for the diagnosis, prognosis, and immunotherapy of tumors. Separation of oligosaccharides is challenging because of the complexity of structures. A method commonly used for liquid separation of complex carbohydrate mixtures is HPLC followed by online ESI-MS detection.13-15 Fragile groups, such as sialic acid, which are unstable in MALDI due to in-source decay, are retained using this soft ionization technique.16 For liquid-based separation of glycans, analytical scale normal phase chromatography following fluorescent labeling17,18 and graphitized 10.1021/pr0604458 CCC: $37.00
2007 American Chemical Society
Glycosylation Profiling of Pancreatic Cancer Serum
carbon chromatography are generally used.19,20 Normal phase separations are based upon hydrophilic interactions of the oligosaccharide hydroxyl groups with the stationary phase. Isocratic structures can be separated using the high resolution of this technique.21 In recent work, hydrophilic interaction online-LC-ESI-MS analysis of underivatized oligosaccharides to nanoscale and low-femtomole sensitivity was achieved with oligosaccharides cleaved from standard glycoproteins using mass detection.22 The site of carbohydrate attachment to the peptide backbone is also essential for the understanding of the glycan-protein interactions. Glycosylation site occupancy may be altered in tumor cells.23 It is known that Asn-Xaa-Ser/Thr (where Xaa is not Pro) is the consensus N-glycosylation site on human proteins. PNGase F is an amidase that cleaves between the innermost GlcNAc and asparagine residues from N-linked glycoproteins. PNGase F converts asparagines to aspartic acid during deglycosylation and results in a 1 Da shift; thus, the glycosylation site can be elucidated using this modification. For enrichment of glycopeptides, methods have been developed to nonspecifically trap glycopeptides with hydrazide chemistry.24 Lectin affinity chromatography has been applied to extract glycoproteins with specific structures.25-27 Other work has used a 2-fold lectin affinity chromatography to enrich N-glycopeptides for elucidation of N-glycosylation sites on human platelet proteins.28,29 Herein, we have developed a streamlined strategy for comprehensive analysis of N-glycans and glycosylation sites in human serum. We demonstrate the presence of specific glycans in pancreatic cancer serum as a proof-of-principle for the workflow introduced. N-linked glycopeptides from normal and pancreatic cancer serum were enriched using two Concanavalin A (ConA) lectin affinity columns. Glycopeptides (0.5 µg) modified by PNGase F were separated by nanoscale reverse phase HPLC followed by online MS/MS sequencing. N-glycosylation sites were identified when the consensus sequence N-X-S/T was matched with a 1 Da shift observed on asparagine. Purified oligosaccharides from ∼25 µg glycopeptides were separated by capillary hydrophilic interaction chromatography and detected online by ESI-TOF MS. The retention time generally increases with elongation of the oligosaccharide chain. A comparison was made between normal and cancer serum to study the alteration of oligosaccharide expression in pancreatic cancer serum. The structures of neutral carbohydrates were studied using a MALDI-quadrupole-ion trap TOF MS in positive mode whereas the sialylated oligosaccharides were analyzed in negative mode. Oligosaccharides characteristic of pancreatic cancer cells but absent in normal cells were observed. As an example, we demonstrate the potential utility of this approach to identify oligosaccharide markers for malignancy.
2. Experimental Section 2.1. Samples. Human normal serum and pancreatic cancer serum were obtained from patients seen at the University of Michigan Health Center. All patient identifiers were removed from the serum samples. Consent was received from all patients, and the protocol was approved by the University of Michigan Institutional Review Board. Blood (40 cm3) was provided by each patient. The samples were permitted to sit at room temperature for a minimum of 30 min (and a maximum of 60 min) to allow the clot to form in the red-top tubes, and then they were centrifuged at 1300 × g at 4 °C for
research articles 20 min. The serum was removed, transferred to a polypropylene-capped tube, and frozen at -70 °C until use. 2.2. Concanavalin A Affinity Extraction and Protein Assay. Protein assays were carried out in a 250 µL transparent 96well plate (Fisher, Barrington, IL) according to the Bradford assay method, because the plate-based approach enables the simultaneous reading of all the samples and standards. Five milliliters of agarose-bound Concanavalin A (ConA) (Vector Laboratories, Burlingame, CA) was packed into an empty PD10 column. The column was first equilibrated and washed with 10 mL of binding buffer (20 mM Tris, 1 mM MnCl2, 1 mM CaCl2, 0.15 M NaCl, pH7.4). Twenty-five milligrams of serum proteins (approximately 0.5 mL of serum) were diluted with 10 mL of binding buffer. Two normal and two cancer samples were studied in this work. The protease inhibitor stock solution was prepared by dissolving one complete EDTA-free Protease inhibitor cocktail tablet (Roche, Indianapolis, IN) in 1 mL of H2O. The stock solution was added to a diluted sample at a ratio of 1:50. Serum samples were loaded onto the column and incubated for 15 min. The column was washed with 10 mL of binding buffer twice to wash off the nonspecific binding. The captured glycoproteins were released with 10 mL of elution buffer (0.4 M R-methyl-mannoside in 20 mM Tris and 0.5 M NaCl, pH 7.0). The extracted glycoproteins represented ∼16% of the total proteins in the serum. N-glycopeptides were extracted in the same manner. Tryptic digests of about 1.5-2.5 mg glycoproteins were diluted 10 times with binding buffer and then applied to a ConA-agarose column (1.5 cm ID, 2 cm long). The glycopeptides were collected by elution with 0.3 M R-methyl-mannoside in 20 mM Tris and 0.5 M NaCl, pH 7.0. 2.3. N-Glycoprotein Denaturation and Digestion. Repeated ultrafiltration (Amicon, 5k MWCO, Millipore, Bedford, MA) was performed to exchange buffer against 0.1% SDS, 50 mM Tris (pH 8.5). DTT (7 mM) (Sigma-Aldrich, St. Louis, MO) was added, and then the sample was incubated at 60 °C for 45 min. After cooling to room temperature, the proteins were alkylated by 15 mM iodoacetamide in the dark for 45 min. Denaturing buffer was further exchanged with 25 mM ammonium bicarbonate pH 7.8 through repeated ultrafiltration (YM-10, 10k MWCO, Millipore) at 5000 × g. Trypsin was added to the solution at an enzyme:protein ratio of 1:50 w/w. Digestion was performed at 37 °C for approximately 16 h. The digestion was stopped by addition of acetic acid (final pH 4). 2.4. Deglycosylation with PNGase F and Sample Purification. Sample glycopeptides were adsorbed to a nonporous ODSII C18 (33 × 8.0 mm) reverse phase column (Eprogen, Inc., Darien, IL), recovered by a fast gradient of 5% (v/v) acetonitrile (ACN) to 100% ACN in 2 min at a flow rate of 0.5 mL/min. Samples were dried to 50 µL with a centrifugal vacuum concentrator (Thermo, Milford, MA). Sodium phosphate (50 mM), pH 7.5, was added to the solution along with 3 µL (0.015 U) of enzyme PNGase F (QA-Bio, Palm Desert, CA). The glycopeptides were deglycosylated overnight at 37 °C. Approximately 3 µg of deglycosylated peptides remained and were stored for peptide sequencing experiments. Peptides and undigested proteins were removed from the remaining sample by a reverse-phase cartridge (glycoclean R cartridges) (Prozyme, San Leandro, CA). Glycan samples were loaded and washed with 2.5% acetic acid. Salts were further removed from glycans by a black graphitized carbon cartridge (150 mg, 0.5 mL column volume) (Alltech, Deerfield, IL). The samples were loaded in water and eluted with 10, 25, and 40% ACN in 0.05% TFA. Journal of Proteome Research • Vol. 6, No. 3, 2007 1127
research articles 2.5. Reverse Phase Separation of Peptide Mixtures. Before separation, ∼0.5 µg of peptide samples were dried down and redissolved in 10 µL of HPLC grade water. The complex deglycosylated peptide mixtures were separated using a reverse phase column attached to a Paradigm HPLC pump (Michrom Bio Resources Inc, Auburn, CA). For nanoLC-ESI-MS/MS experiments, a nanotrap platform (Michrom) was set up prior to the electrospray source. It includes a peptide nanotrap (0.15 × 50 mm, Michrom) and a separation column (0.075 mm × 150 mm, C18, michrom). The peptide sample was injected and first desalted on the trap column with 5% solvent B (0.3% formic acid in 98% ACN) at 50 µL/min for 5 min. Subsequently the peptides were eluted using a 2 h gradient from 5 to 95% B at a flow rate of 0.25 µL/min, where solvent A was 0.3% formic acid in 2% ACN. 2.6. Hydrophilic Interaction Separation of Glycan Mixtures. The capillary normal phase column (200 µm × 120 mm) was packed in-house using TSKgel Amide-80 resin (5 µm, 80 Å, Tosoh Biosciences, Montgomeryville, PA). A capillary pump (Ultra-Plus II MD, Micro-Tech Scientific, Vista, CA) was used for separation. The capillary column was directly mounted to a micro-injector with a 500 nL internal sample loop (Valco Instruments, Houston, TX). The flow from the solvent delivery pump was split precolumn to generate a flow rate of ∼3 µL/ min. Before injection, complex N-glycan mixtures were dried down and redissolved in 80% ACN. A 50 pmol glucose ladder standard (Beckman-Coulter, Fullerton, CA) or glycan mixtures obtained from approximately 25 µg serum glycopeptides were injected to the normal phase column. The glycans were separated in a 45 min linear gradient that is 100% B to 40% B in 45 min. Solvent A is 50 mM formic acid with pH 4.5 adjusted by ammonium hydroxide. Solvent B is 20% solvent A in ACN. 2.7. Mass Spectrometric Analysis. Peptide Sequencing and Data Interpretation. A Finnigan LTQ mass spectrometer (Thermo electron, Waltham, MA) equipped with a nanoESI source was used to acquire spectra. A 75 µm metal spray tip (Michrom) was used and the spray voltage was set at 3 kV. The instrument was operated in data-dependent mode with dynamic exclusion enabled. The MS/MS spectra on the five most abundant peptide ions in full MS scan were obtained. All MS/ MS spectra were searched against the human protein database from SwissProt using SEQUEST algorithm incorporated in Bioworks software (Thermo). The sequence database search tool was set to expect the following variable modifications: carboxymethylated cysteines, oxidized methiones, and an enzyme-catalyzed conversion of asparagines to aspartic acid (0.984 Da shift) at the N-glycosylation site. Trypsin was used as a specific protease with two missed cleavages allowed. Searches were repeated without specification of a protease to search for nonspecific cleavage products. Positive protein identification was accepted for a peptide with Xcorr of greater than or equal to 3.0 for triply, 2.5 for doubly, and 1.9 for singly charged ions. Glycan Mapping. The micro hydrophilic interaction LC system was directly coupled with an ESI-TOF (LCT premier, Micromass/Waters, Milford, MA). The capillary voltage for electrospray was set at 3200 V, sample cone at 75 V. Desolvation was accelerated by maintaining the desolvation temperature at 150 °C and source temperature at 100 °C. The desolvation gas flow was 200 L/h. The data was acquired in “V” mode and the TOF was externally calibrated by Sodium Iodide and Cesium Iodide mixtures. The instrument was controlled by MassLynx 4.0 software. 1128
Journal of Proteome Research • Vol. 6, No. 3, 2007
Zhao et al.
Glycan Structure Analysis. MS and MS2 spectra of glycan samples were acquired on an Axima QIT MALDI quadrupole ion trap-TOF (MALDI-QIT) (Shimadzu Biotech, Manchester, UK). For positive mode analysis, 25 mg/mL 2,5-dihydroxybenzoic acid (DHB) (LaserBio Labs, France) was prepared in 50% ACN. The dried oligosaccharides were resuspended in water and were applied to the MALDI probe (∼0.3 mg/mL, 0.5 µL) followed by 0.01 M NaCl (0.5 µL). After the sample was dried in air, 0.5 µL matrix was added. For negative mode, matrix solution was prepared by dissolving 5 mg/mL 2′,4′,6′-trihydroxyacetophenone monohydrate (THAP) and 10 mM ammonium citrate in 50% ACN as described by Papec et al.30 Acquisition and data processing were controlled by Launchpad software (Karatos, Manchester, UK). A pulsed N2 laser light (337 nm) with a pulse rate of 5 Hz was used for ionization. Each profile results from 2 laser shots. Argon was used as the collision gas for CID and helium was used for cooling the trapped ions. TOF was externally calibrated using 500 fmol/µL of bradykinin fragment 1-7 (757.40 m/z), angiotensin II (1046.54 m/z), P14R (1533.86 m/z), and ACTH (2465.20 m/z) (Sigma-Aldrich) for both positive and negative mode.
3. Results and Discussion 3.1. Separation of Glucose Ladder Using Normal-Phase Micro-LC-ESI-MS. A capillary hydrophilic interaction column (200 µm × 120 mm) was connected online to ESI-TOF MS with a flow rate of 3 µL/min. The glucose ladder (consisting of at least 20 individual linear glucose oligomers) was used to test the separation efficiency and sensitivity. Figure 1 shows the selected ion chromatogram and corresponding combined mass spectra of 50 pmol of the glucose ladder. Each detected mass represents the successive increase of one glucose residue (162 Da). The separation was based on polar interactions of the oligosaccharide hydroxyl groups with the stationary phase. These linear oligomers consisting of 3-14 glucose residues were eluted in the order of the oligosaccharide chain length where the glucose trimer (G3) elutes earliest at 9 min and the 14glucose polymers (G14) elutes latest at 23.2 min as shown in Figure 1. The broad peaks observed for G3-G5 are due to the saturation of these glucose oligomer ions in the mass spectrometer. The elution of the glucose ladder could be used for standardizing retention of oligosaccharides in terms of glucose units as in electrophoresis31 and as an internal calibration for the correction of experimental variations. 3.2. Double-Lectin Affinity Extraction of Glycopeptides and Deglycosylation. In this work, we developed a streamlined approach for comprehensive analysis of N-glycans and glycosylation sites in human serum. The general strategy used in this work is outlined in Figure 1. N-glycopeptides were enriched by double ConA lectin affinity chromatography. ConA recognizes a commonly occurring sugar structure, R-linked mannose, which is the core structure of a wide variety of serum and membrane glycoproteins. Thus, most glycoproteins can be purified with ConA although the binding efficiency to the high mannose core structure proteins is higher than other glycan structure proteins. Twenty-five milligrams of either normal or pancreatic cancer serum proteins (∼0.5 mL serum) were loaded on the first ConA column. A protein assay was performed to make sure that the same amount of protein from normal and cancer serum were analyzed. Approximately 16% of the total proteins were extracted. Because albumin (HSA) composes ∼65% of total serum proteins, the recovery would be ∼50% when albumin is not considered. The lectin affinity column is
research articles
Glycosylation Profiling of Pancreatic Cancer Serum
Figure 1. Capillary hydrophilic interaction (0.2 mm × 120 mm) separation of 50 pmol glucose ladder detected online by ESI-TOF MS. (a) Selected ion chromatogram of different glucose oligomers eluting at around 9, 11, 14.5, 16.2, 17.4, 18.5, 19.5, 20.4, 21.2, 21.9, 22.5, and 23.2 min, respectively, in I-XII. The flow rate is 3 µL/min. (b) Combined spectra of the corresponding selected ion chromatogram in Figure 2(a). The detected masses of glucose oligomers were circled. (I-XII): Singly charged oligomers of 3-14 glucoses were detected at m/z 522.1689, 684.2122, 846.2546, 1008.2871, 1170.3445, 1332.4380, 1494.5304, 1656.5780, 1818.6814, 1980.7017, 2142.7532, and 2304.8547, respectively. Each oligosaccharide band was designated as G3-G14.
Figure 2. Outline of the analysis of N-glycans and N-glycosylation sites of human serum sample.
overloaded by the HSA in this work where more low abundant proteins could otherwise be extracted. When the loading capacity is not reached, the recovery could be ∼90% for the ConA affinity column for standard glycoproteins. In addition,
the recovery for N-linked glycan structures with a mannose core such as complex type glycans is lower than the high mannose glycan structure proteins where nearly 100% recovery could be obtained. Journal of Proteome Research • Vol. 6, No. 3, 2007 1129
research articles
Zhao et al.
Figure 3. N-linked carbohydrates were separated by microhydrophilic interaction LC and online detected by ESI-TOF MS. The combined mass spectra of retention time 27-28 min and retention time 28-29 min are shown in (a) and (b). All of the peaks presented here are doubly charged. The spectrum for cancer sample is shown above the normal sample. The proposed glycan structures are shown on the corresponding peaks. 2, fucose; f, N-acetyl neuraminic acid (sialic acid); 9, HexNAc; O, mannose; ), galactose.
The first lectin column removes nonglycosylated proteins, including most of the albumin, thereby reducing the complexity of the sample. After denaturation and reduction, followed by alkylation, the enriched glycoproteins were digested. Since nonglycosylated peptides are much more abundant than the glycosylated peptides, a second ConA column was utilized to remove the nonglycosylated peptides. The typical yield of the 1130
Journal of Proteome Research • Vol. 6, No. 3, 2007
glycopeptides is around 5% of the weight of glycoprotein digests used. We tested the reproducibility of glycopeptide trapping by injecting the extracted glycopeptides from the same amount of glycoproteins to the mass spectrometer. The total ion intensities are very similar which shows good reproducibility. Following purification with a reverse phase column, the enriched glycopeptides for each sample were deglycosylated
research articles
Glycosylation Profiling of Pancreatic Cancer Serum Table 1. N-glycosylation Sites Identified in Human Pancreatic Cancer Serum protein namea
acc#
peptide sequenceb (glycosylaiton site @)
MH+
charge
XC
site
knownc
Alpha-1-antitrypsin Alpha-1-antitrypsin Alpha-1-antitrypsin Alpha-1B-glycoprotein* Alpha-1B-glycoprotein Alpha-2-macroglobulin Alpha-2-macroglobulin Alpha-2-macroglobulin Alpha-2-macroglobulin Alpha-2-macroglobulin* Alpha-1-antichymotrypsin* Alpha-1-antichymotrypsin Alpha-1-antichymotrypsin Alpha-albumin AMBP protein* Apolipoprotein B-100 Apolipoprotein B-100 Beta-2-glycoprotein 1 Beta-2-glycoprotein 1 ADAMTS-9* Complement C1r subcomponent C4b-binding protein alpha chain* Ceruloplasmin Ceruloplasmin Complement factor B Complement factor B Complement factor H Complement factor H Complement factor I Complement factor I Clusterin Clusterin Complement C3 Complement C3 Complement C4-A Complement C4-A Complement C5 Complement component C7 Carboxypeptidase N subunit 2 Carboxypeptidase N subunit 2 Alpha-2-HS-glycoprotein Alpha-2-HS-glycoprotein Fibronectin Glia-derived nexin* Hemopexin Hemopexin Hemopexin Heparin cofactor 2 Haptoglobin Haptoglobin Haptoglobin Haptoglobin-related protein Haptoglobin-related protein Haptoglobin-related protein Histidine-rich glycoprotein Histidine-rich glycoprotein Plasma protease C1 inhibitor Plasma protease C1 inhibitor* Plasma protease C1 inhibitor Plasma protease C1 inhibitor Ig alpha-1 chain C region Ig alpha-1 chain C region Ig gamma-1 chain C region Immunoglobulin J chain Inter-alpha-trypsin inhibitor heavy chain H1 Inter-alpha-trypsin inhibitor heavy chain H1 I Inter-alpha-trypsin inhibitor heavy chain H2 Inter-alpha-trypsin inhibitor heavy chain H3
P01009 P01009 P01009 P04217 P04217 P01023 P01023 P01023 P01023 P01023 P01011 P01011 P01011 P43652 P02760 P04114 P04114 P02749 P02749 Q9P2N4 P00736 P04003 P00450 P00450 P00751 P00751 P08603 P08603 P05156 P05156 P10909 P10909 P01024 P01024 P0C0L4 P0C0L4 P01031 P10643 P22792 P22792 P02765 P02765 P02751 P07093 P02790 P02790 P02790 P05546 P00738 P00738 P00738 P00739 P00739 P00739 P04196 P04196 Q96FE0 Q96FE0 Q96FE0 Q96FE0 P01876 P01876 P01857 P01591 P19827
R.QLAHQSN@STNIFFSPVSIAT AFAMLSLGTK.A
[email protected] [email protected] [email protected] [email protected] K.SLGNVN@FTVSAEALESQELC#GTEVPSVPEHGR.K
[email protected] K.GC#VLLSYLN@
[email protected] K.GC#VLLSYLN@
[email protected] I.N@TTNVM*GTSLTVRVN@YKDRSPC#Y.G
[email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] T.CGQGRATRQVM*
[email protected] [email protected] E.N@ETIGVWRPSPPTC#EK.I
[email protected] [email protected] [email protected] [email protected] R.WQSIPLC#VEKIPC#
[email protected] [email protected] R.SIPAC#VPWSPYLFQPN@DTC#IVSGWGR.E
[email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] K.LYLGSNN@
[email protected] K.LYLGSNN@
[email protected] [email protected] [email protected] [email protected] F.EDPASACDSIN@AWVKN@ETRDM*.I K.ALPQPQN@VTSLLGC#TH
[email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] R.VIDFN@C#TTSSVSSALANTK.D
[email protected] [email protected] [email protected] [email protected] [email protected] R.LAGKPTHVN@VSVVMAEVDGTC#Y R.PALEDLLLGSEAN@LTC#TLTGLR.D
[email protected] [email protected] [email protected] 3181.64 1757.86 3691.82 3479.78 3779.65 3416.61 2165.15 2728.37 2728.37 2697.23 2135.96 2213.06 2400.19 1195.56 2222.12 1685.82 1798.9 1468.75 1193.54 2370.03 2538.17 1872.91 1892.84 1582.77 2529.1 506.22 3263.64 1504.63 3009.45 1254.58 1685.80 2413.12 2260.08 2841.41 1774.81 1104.6 3318.57 2307.96 2314.26 2314.26 1657.8 2365.15 2299.02 2371.95 1737.88 1347.65 1851.79 1482.78 1796.98 1458.73 1458.73 2682.34 2734.38 2734.38 2017.96 1623.74 2719.37 1932.85 2101.09 2316.32 2348.16 2359.24 1189.51 2151.13 2831.40
3 2 3 3 3 3 2 3 3 3 2 3 3 2 2 2 2 2 2 2 2 2 2 2 3 1 3 2 2 2 2 2 2 3 2 2 3 2 2 2 2 3 2 2 2 2 2 2 2 2 2 3 3 3 2 2 3 2 3 2 2 2 2 3 3
4.05 3.87 4.25 4.1 3.9 4.23 4.85 3.87 3.87 3.55 3.47 4.31 4.78 4.23 3.97 2.89 2.96 4.79 4.42 3.28 2.89 3.10 3.40 2.98 3.35 1.99 4.14 3.07 4.35 3.63 3.88 6.00 4.24 3.57 3.70 4.12 4.12 3.91 4.15 4.15 3.74 4.55 2.81 2.94 3.11 4.67 3.95 4.13 3.29 2.79 2.79 5.57 4.70 4.67 6.48 4.11 4.23 3.26 4.01 3.27 3.35 5.44 3.77 3.52 3.95
70 271 107 44 179 869 1424 55 70 410 33 127 106 33 36 1523 2982 162 253 1267 125 221 138 358 122 142 882 911 464 103 374 354 85 1617 226 1328 1115 202 348 359 156 176 528 159 453 187 246 49 241 207 211 126 149 153 125 344 238 25 253 352 340 144 180 49 588
yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes potential yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No No No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
P19827
[email protected] 2872.4
3
3.64
285 Yes
P19823 P19823
[email protected] [email protected] 2125.08 1473.7
3 2
4.22 3.24
96 yes 118 yes
Q06033
[email protected] 2272.19
2
3.47
87 yes
Journal of Proteome Research • Vol. 6, No. 3, 2007 1131
research articles
Zhao et al.
Table 1 (Continued) protein namea
acc#
Inter-alpha-trypsin inhibitor heavy chain H4 Inter-alpha-trypsin inhibitor heavy chain H4 Plasma kallikrein Plasma kallikrein Plasma kallikrein Kininogen-1 Ig kappa chain V-I region CAR Lumican Ig mu chain C region Ig mu chain C region Ig mu heavy chain disease protein Ig mu heavy chain disease protein N-acetylmuramoyl-L-alanine amidase N-acetylmuramoyl-L-alanine amidase Proteolipid protein 2 Serum amyloid A-4 protein Serum amyloid P-component Delta-sarcoglycan Prothrombin Serotransferrin Serotransferrin 5,6-dihydroxyindole-2carboxylic acid oxidase Vitronectin
Q14624
peptide sequenceb (glycosylaiton site @)
knownc
MH+
charge
XC
site
[email protected] 2811.32
2
4.21
517
yes
Q14624
K.KAFITN@
[email protected] 2622.27
3
3.89
81
yes
P03952 P03952 P03952 P01042 P01596 P51884 P01871 P01871 P04220
[email protected] [email protected] [email protected] [email protected] [email protected] [email protected] R.GLTFQQN@ASSMC#VPDQDTAIR.V K.STGKPTLYN@VSLVMSDTAGTCY K.STGKPTLYN@VSLVM*SDTAGTC#Y.-
2360.16 1436.8 1984.01 1874.88 2093.05 2195.17 2341.07 2308.08 2383.10
2 2 2 2 2 3 2 2 2
4.00 3.98 4.09 3.53 3.87 3.98 6.34 3.41 6.34
396 453 308 48 28 160 210 441 378
yes yes yes yes yes yes yes yes no
P04220
[email protected] 2341.04
2
3.34
147
no
Q96PD5
R.LEPVHLQLQCMSQEQLAQVA
[email protected] 2750.4
3
3.31
367
yes
Q96PD5
[email protected] 2093.09
2
4.37
485
yes
Q04941 P35542 P02743 Q92629 P00734 P02787 P02787 P17643
[email protected] [email protected] R.ESVTDHVNLITPLEKPLQN@FTLC#FR.A
[email protected] [email protected] R.QQQHLFGSN@VTDC#SGNFC#LFR.S
[email protected] R.TCHC#NGN@
[email protected] 2624.54 2626.28 2974.52 2414.29 1220.63 2518.14 1419.73 2368.89
3 2 3 3 2 2 2 2
3.52 3.46 6.18 4.37 3.67 2.97 3.01 2.81
108 94.00 51 109 416 630 432 104
yes yes yes potential yes yes yes potential
P04004
[email protected] 2381.17
3
3.54
86
yes
a Unspecific cleavages are marked by *. b @: glycosylation site on the sequence; #: carbamidomethylation on cysteine; *: methione oxidation unknown, and potential sites were marked according to Swiss-Prot annotation.
using PNGase F. PNGase F cleaves the oligosaccharides from the N-glycopeptides, resulting in aspartic acid in place of asparagines. 3.3. Analysis of Serum N-Glycans Resolved by Capillary Hydrophilic Interaction Chromatography. A reverse phase and graphitized carbon solid-phase extraction were performed to remove peptides and most of the salts and, at the same time to preconcentrate the released oligosaccharides. Underivatized glycan mixtures obtained from 25 µg serum glycopeptides were separated by the capillary hydrophilic interaction column and mass measured online by ESI-TOF MS. To reduce the experimental variance, a glucose ladder is injected between each run for calibrating the LC-MS condition. GlycoMod tool (http:// www.expasy.org/tools/glycomod/) was used to predict the possible oligosaccharide structures and composition from these experimentally determined masses. Only glycan structures that have been included in the GlycoSuite database (https:// tmat.proteomesystems.com/glycosuite/) were selected. MALDIMS/MS was performed on selected glycan peaks to further validate their structures (data shown in Section 3.5). During the 50 min gradient, 202 glycan peaks were detected. The fractionation of oligosaccharides before mass detection simplified the mass spectrum where glycans with different structures but similar masses could be resolved. The use of separation also reduces ion suppression in these complex samples. The monosaccharide composition, mass value, mass shift, and abundance in two samples are listed in the supplementary Table 1. Compared to the theoretical mass value, most of the glycan peaks have a mass shift less than 0.2 Da. The mass difference does not exceed 1 Da for masses under 4000 Da and 1.5 Da for masses over 4000 Da. Many glycans were detected in more than one charge state. In addition, some 1132
Journal of Proteome Research • Vol. 6, No. 3, 2007
c
Known,
glycans were present in proton adduct form and also in potassium and/or sodium adduct form. The observation of multiple charge states or adduct forms further confirmed the presence of this glycan structure. We searched the MS/MS spectra from the glycan analysis against a human protein database. The ions corresponding to peptides were eliminated. Combining all the factors discussed above, 105 carbohydrate structures were unique from the 202 detected glycan peaks in these two samples as shown in supplementary Table 1. Most of the carbohydrate structures, especially in the low mass range (
