Glycoproteomics of Plasma Based on Narrow Selectivity Lectin Affinity Chromatography Kwanyoung Jung, Wonryeon Cho, and Fred E. Regnier* Department of Chemistry, Purdue University, West Lafayette, Indiana 47907 Received September 9, 2008
Lectin affinity chromatography using concanavalin A (Con A), Helix pomatia agglutinin (HPA), Lycopersicon esculentum lectin (LEL), Aleuria aurantia lectin (AAL) and Lens culinaris agglutinin (LCA) was used to investigate the utility of narrow selectivity lectins in the characterization of plasma glycoproteome diversity and to recognize cancer associated aberrations in glycosylation. Following affinity chromatographic selection, proteins were tryptically digested, the peptide fragments separated by reversed phase chromatography (RPC), and fractions from RPC identified by tandem mass spectrometry. The diversity of glycosylation found with narrow selectivity lectins was generally 2/3 that of Con A and not related to protein abundance. Small groups of proteins were found with each of the affinity columns, HPA, LEL, AAL, and LCA, that changed 3-fold or more in concentration between normal and breast cancer patient plasma. Although the number of cancer patients examined was too small to validate cancer marker candidates, they are clearly worth examining in a larger, more diverse patient population. Keywords: Lectins • affinity chromatography • glycoproteomics • glycoproteins • aberrant glycosylation • breast cancer • plasma • diversity • post-translational modifications • quantification • iTRAQ • HPA • LEL • AAL • LCA
Introduction On the basis of the widespread use of blood in clinical medicine, there is broad interest in defining the plasma proteome. Through multiple mechanisms, proteins from most organs in the body make their way into the circulatory system, creating a plasma proteome of great complexity.1 Many of these proteins can exist in one to over 50 post-translationally modified isoforms. The plasma proteome is probably the most complicated proteome in the human body. Among the more than 100 possible post-translational modifications, glycosylation is one of the more common. Recent studies suggest that up to 50% of the proteins in the plasma proteome are glycosylated2 and that glycosylation is altered in association with cell growth and differentiation,3 multiple diseases,4 aging,5 and environmental stress.6 It is important that methods for the isolation and identification of glycoproteins be advanced along with their application to the study of biological systems. Glycoprotein isolation and identification from complex mixtures has been achieved in several ways. One is by targeting the vicinal diol moiety of appended carbohydrates.7 In this approach, vicinal diols are first oxidized to aldehydes or ketones with periodate and then contacted with a hydrazide containing resin where a Schiff base is formed that couples oxidized glycoproteins to the support. After reduction of Schiff bases to more stable amines and washing to remove unbound proteins, glycoproteins are proteolytically digested on the solid phase support, leaving the peptide at the glycosylation site still * To whom correspondence should be addressed. E-mail: fregnier@ purdue.edu. 10.1021/pr8007495 CCC: $40.75
2009 American Chemical Society
attached to the resin. Following stable isotope coding for potential quantification, peptides are enzymatically released from the resin with PNGase F and identified through conventional mass spectral based proteomics methods.8 A limitation of this approach is that it is difficult to differentiate between glycan variants on a glycoprotein after periodate oxidation. An alternative is to affinity select glycoproteins or glycopeptides with an immobilized glycan targeting protein such as an antibody9 or lectin.10 Advantages of this affinity selection approach are that binding is reversible after selection, multiple affinity selectors are available that allow different glycoforms to be targeted, glycan selectors can be applied in serial fashion,11 and glycans can be retrieved for characterization and quantification.12,13 Plasma and serum have already been examined with broad selectivity lectins such as concanavalin A, wheat germ agglutinin and jacalin.14,15 The objectives of the work described here were to (1) expand lectin based characterization of glycoproteins in plasma to narrower selectivity lectins that target more specific features of glycoproteins and (2) determine whether these lectins might be of value in studying disease associated aberrations in glycosylation.
Materials and Methods Materials and Chemicals. Immobilized lectin sorbents were purchased from Vector Laboratories (Burlingame, CA). Helix pomatia lectin was purchased from EY laboratory (San Mateo, CA). iTRAQ reagents and the ABI 4700 Proteomics Analyzer Calibration Mixture (4700 Cal Mix, bradykinin, angiotensin I, glu1-fibrinopeptide B, ACTH fragment 1-17, ACTH fragment 18-39, and ACTH fragment 7-38) were purchased from Journal of Proteome Research 2009, 8, 643–650 643 Published on Web 12/19/2008
research articles Applied Biosystems (ABI, Foster City, CA). Normal pooled human plasma was supplied by the National Institute of Standards and Technology (NIST, Gaithersburg, MD). Human breast cancer plasma from ductal carcinoma patients was purchased from Asterand, Inc. (Detroit, MI). Acetic acid, sodium hydroxide, calcium chloride, magnesium chloride, trifluoroacetic acid (TFA), and HPLC grade acetonitrile were purchased from Mallinckrodt Chemicals (Phillipsburg, NJ). Glycine, manganese chloride, R-cyano-4-hydroxy-cinnamic acid (CHCA), proteomics grade N-p-tosyl-phenylalanine chloromethyl ketone (TPCK)-treated trypsin, 4-(2-hydroxyethyl)-1piperazine ethanesulfonic acid (HEPES), iodoacetic acid (IAA), Bradford reagent, and L-cysteine were obtained from SigmaAldrich (St. Louis, MO). Sodium dodecyl sulfate (SDS) was purchased from Fluka Biochemika (Buchs, Switzerland). Dithiothreitol (DTT) and urea were provided by Bio-Rad Laboratories (Hercules, CA). The DI water system and C18 spin columns were purchased from Millipore (Boston, MA). Affinity Chromatography. Agarose bound LCA sorbent was self-packed in a 4.6 × 100 mm column. Other lectin sorbents were self-packed in a 4.6 × 50 mm column. Protein concentration in human plasma in the control sample from NIST and ductal carcinoma patient samples was estimated using the Bradford assay as needed to prepare samples with equal amounts of total protein. Human plasma was loaded directly onto soft-gel lectins column with mobile phase A (0.10 M HEPES buffer, pH 7.5, containing 1 mM CaCl2, 1 mM MgCl2, and 0.1 mM MnCl2) at a flow rate of 0.3 mL/min. Following extensive washing with mobile phase A, lectin columns were eluted with buffer B (0.1 M glycine/2% acetic acid-HCl buffer, pH 2.5). Elution curves were obtained with an absorbance detector operating at 280 nm using a HPRP Module liquid chromatograph from Beckman Coulter, Inc. (Fullerton, CA). Trypsin Digestion. Captured glycoproteins were dried completely with a Speed-Vac after they were adjusted to pH 7.5 with 1.0 M NaOH and then reconstituted with 8 M urea in 50 mM HEPES buffer containing 10 mM CaCl2. Proteins thus denatured were reduced with 10 mM DTT. After 2 h incubation at 50 °C, iodoacetic acid was added to a final concentration of 20 mM for alkylation and incubated in darkness for 2 h more. L-cysteine was then added to the reaction mixture to a final concentration of 40 mM and the mixture incubated for 30 min at room temperature. After dilution with 50 mM HEPES buffer to a final urea concentration of 1.0 M, proteomics grade trypsin (2%, w/w, enzyme to protein) was added and incubated for 24 h at 37 °C. The digest was then stored at 0 °C until needed for identification and quantification. Mass Spectrometry Based Protein Identification. Glycoproteins were identified with an ABI 4800 Proteomics Analyzer mass spectrometer. The pH of the glycoprotein fraction selected by lectin affinity chromatography and eluted with 0.1 M glycine/2% acetic acid-HCl buffer (pH 2.5) was adjusted to pH 7.5 and the volume reduced by Speed-Vac. Following volume reduction, the sample was tryptically digested and the peptide fragments were subjected to reversed phase chromatography (RPC) using a Pepmap C18 trap column, a nanocolumn (Zorbax 300sB-C18, 3.5 µm, 100 µm i.d., 15 cm length, Agilent Technologies, Santa Clara, CA), and an Agilent 1100 Series HPLC (Agilent Technologies). The RPC separation was achieved using a 40 min linear gradient from 98% solvent A with 2% solvent B to 60% solvent A with 40% solvent B at a flow rate of 800 nL/ min. Solvent A was composed of DI water to which trifluoroacetic acid (TFA) had been added to a concentration of 0.1%. 644
Journal of Proteome Research • Vol. 8, No. 2, 2009
Jung et al. Solvent B was prepared with acetonitrile (ACN) to which TFA had been added to a concentration of 0.1%. Peptides were collected directly from the RPC column onto a 384 target, stainless steel MALDI plate utilizing a microfraction collector and MALDI spotter driven by an Agilent 1100 LC System. Ninety-six of the 384 MALDI targets were used for each sample. Column effluent was combined in a mixing tee with MALDI matrix (R-cyano-4-hydroxycinnamic acid, 8 mg/mL in 60% ACN/0.1% TFA) delivered at 1.2 µL/min. Peptides were analyzed with an ABI 4800 Proteomics Analyzer mass spectrometer in the positive ion mode. The 4800 Proteomics Analyzer was equipped with TOF/TOF ion optics and a 200 Hz Nd:Yag laser. Automated acquisition of MS and MS/MS data was controlled by 4000 Explorer software. Automated MS/MS data analysis was performed utilizing Protein Pilot software 2.0 using the Pro Group algorithm (ABI) for protein identification. The minimum acceptance criterion for peptide identification was the 99% confidence level. Most of proteins were identified based on the presence of two peptides from a protein identified by the Pro Group algorithm at the 99% confidence level. An unused score cutoff of 4 was the minimum value for identifying proteins with the Protein Pilot 2.0 software. Proteins were also identified with MASCOT Version 2.2.2 using the Swiss-Prot R48 database. The following parameters were used in the MASCOT analysis: (1) database, SwissProt; (2) enzyme, trypsin; (3) allowed missed cleavage, two; (4) fixed modification, carboxymethyl (C); (5) variable modification, oxidation (M); (6) peptide tolerance, 1.2 Da; (7) MS/MS tolerance, 0.6 Da; (8) peptide charge, +1; (9) instrument, MALDI-TOF-TOF. In addition, Decoy was used to estimate false positive rates. Results were filtered with a significance threshold of p < 0.05. The false discovery rate was generally under 1%. Proteins are listed in Tables according to their Swiss-Prot entry names and accession numbers. Quantitative Comparison of Protein Abundance with iTRAQ. Tryptic-digested proteins from either lectin affinity chromatography fractions were labeled with iTRAQ reagent to compare control plasma and cancer patient plasma samples. Trypsin digestion and labeling with iTRAQ reagent were performed according to the supplier’s guidelines (ABI). The NIST control sample was labeled with the 114-Da iTRAQ labeling agent, while cancer samples were labeled with the 115Da iTRAQ labeling agent. The 116-Da and 117-Da iTRAQ reagents were also used with cancer patient analyses. According to the literature, the relative standard deviation of the iTRAQ method is near 10%.16,17 Again, the Pepmap C18 trap column and a nano-RPC column were used for desalting and RPC of peptides from affinityselected proteins as described above in the mass spectrometry based protein identification section. Peptides were analyzed on the ABI 4800 Proteomics Analyzer mass spectrometer. Automated acquisition of MS and MS/MS data was controlled by 4000 Series Explorer software. Automated MS/MS data analysis was performed utilizing Protein Pilot software 2.0 with the Pro Group algorithm for protein identification and quantification of iTRAQ reporter ions. Only peptides that were completely labeled with iTRAQ reagent at their N-terminus and lysine residues and had a nonzero relative isotope ratio were considered in comparative proteomics measurements.
Results Glycoprotein characterization through the use of immobilized glycan binding proteins in the isolation phase of glyco-
research articles
Glycoproteomics of Plasma
Figure 1. Lectin affinity chromatography of human plasma. The blue chromatogram is from the LCA affinity column, while the black is from the AAL affinity column and the red from the HPA affinity column at a sensitivity of 2 AUFS. The inset is a chromatogram from the HPA column at 0.1 AUFS. Plasma samples were loaded directly onto columns with 0.10 M HEPES mobile phase (pH 7.5 with 1 mM CaCl2, 1 mM MgCl2, and 0.1 mM MnCl2) at a flow rate of 0.3 mL/min. Elution was achieved with 0.1 M glycine/2% acetic acid-HCl buffer (pH 2.5). The absorbance detector was operated at 280 nm.
proteomics has been achieved in several ways. One has been to use lectin affinity chromatography to capture glycopeptides from tryptic digests of a proteome and following deglycosylation with PNGase F to identify the remaining peptides by conventional proteomic methods involving mass spectral analysis of an RPC fractionated peptide mixture.18 The advantages of this approach are that N-glycosylation sites are readily identified and the peptide mixtures being examined are simpler than a tryptic digest of the proteome. The disadvantage is that there is only one peptide per glycosylation site and ion suppression can limit the availability of that peptide for identification. A second strategy is to lectin-select glycoproteins from a proteome before proteolysis.19 Following lectin affinity chromatography, the glycoprotein fraction can either be (1) further fractionated by RPC or electrophoresis, and protein fractions thus resolved identified by ESI-MS/MS or MALDI-MS/MS or (2) tryptic digested directly and the cleavage fragments separated by RPC and identified by ESI-MS/MS or MALDI-MS/MS. The advantage of this method is that both O- and N-glycosylated proteins can be identified. The second of these two alternatives was chosen for this work based on preliminary studies showing slightly more proteins were identified. Lectin Affinity Selection. Affinity selection was carried with soft-gel sorbents packed in 4.6 × 50 mm high performance columns. Mobile phase velocities of no more than 0.3 mL/min were used to preclude resin compression during loading and elution. Typical elution profiles are seen in Figure 1. Elution was monitored at 280 nm to minimize baseline perturbation from the acetic acid desorbing agent. The relative amount of glycoprotein selected by any particular lectin was determined
Table 1. Relative Amount of Protein Captured by Lectin Affinity Chromatography lectina sample
HPA
LEL
AAL
LCA
Con A
Normal Breast cancer
