Serial Affinity Chromatography as a Selection Tool in Glycoproteomics

Article

Serial Affinity Chromatography as a Selection Tool in Glycoproteomics Kwanyoung Jung, and Wonryeon Cho Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/ac400653z • Publication Date (Web): 11 Jun 2013 Downloaded from http://pubs.acs.org on June 23, 2013

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

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

Page 1 of 28

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

Analytical Chemistry

Serial Affinity Chromatography as a Selection Tool in Glycoproteomics

Kwanyoung Jung1 and Wonryeon Cho2, ∗ 1

Department of Chemistry, Seoul Science High School, 63 Hyehwa-ro, Jongno-gu, Seoul 110-

530, Republic of Korea 2

Department of Chemistry, Wonkwang University, 460 Iksandae-ro, Iksan, Jeonbuk 570-749,

Republic of Korea

Glycan-targeting affinity chromatography systems are becoming increasingly important as tools in the purification, enrichment, and identification of glycoproteins. The great advantage of this strategy is that immobilized lectin and antibody selectors allow specific glycan structures to be matched with a particular protein. It is also possible to show that a glycan seen at one site in a glycoprotein may not be present at another glycosylation site in the same glycoprotein. A problem with single column affinity chromatography is how to obtain information on glycan diversity within the oligosaccharide portions of captured glycoproteins. Although all the glycoprotein species bearing a particular glycan feature will be captured by an affinity column, there is no way of knowing whether the ligand being targeted appears alone or co-resides with a series of other glycan features in the same oligosaccharide conjugate. The work being described here examines the utility of serial affinity columns in determining whether individual glycan structures appear alone or together with other glycans in specific proteins. Four different types of affinity columns were examined in two series; the LEL→HPA→anti∗

To whom correspondence should be addressed. E-mail: [email protected] 1 ACS Paragon Plus Environment


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

LexAb→anti-sLexAb

series

and

the

anti-sLexAb→anti-LexAb→HPA→LEL

Page 2 of 28

series.

Patterns in protein capture from these two series were very different. Thus serial affinity chromatography (SAC) can be a valuable tool in recognizing diversity in protein glycosylation, especially when the order of columns in the SAC series is varied. Two clear types of diversity were recognized. One is the independent occurrence of different affinity targetable glycan features in the same glycoprotein. The other is that multiple targetable glycan features were co-resident in the same glycoprotein. The great advantage of this method is that it couples easily with current methods used in glycoproteomics.

2 ACS Paragon Plus Environment

Page 3 of 28

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


Although blood has been used for decades as a source of protein biomarkers for clinical medicine, discovery and quantification of disease associated proteins through classical chromatographic, electrophoretic, and sequencing methods are slow and tedious. Complicating the search for new disease biomarkers is the fact that diagnostic proteins may be an isoform bearing a post-translational modification (PTM). Fifty percent or more of the thousands of proteins in blood are thought to be PTM bearing. Gene sequence libraries and mass spectrometry based sequencing methods are of minimal value in searching for such biomarkers without structure targeting selection tools that allow enrichment and fractionation. Although a wide variety of affinity selectors are now being used to select PTM bearing proteins and peptides in what is coming to be known as targeted proteomics, how they work and what they do are sometimes not well understood. The focus of the work presented here is on targeting glycans in glycoproteins. Glycoprotein and glycopeptide capture has been achieved most widely by lectin affinity chromatography (LAC)1-5. The fact that glyco-conjugates are easily retrieved from lectin columns6-9 makes this approach very attractive, especially in glycoproteomics. Thus affinity selected glycoproteins are trypsin digested and the product peptides are identified by tandem mass spectrometry (MS/MS). N-Glycosylation sites are subsequently identified by treatment with PNGase F to remove the glycan and to identify the deglycosylated peptide(s). A complication associated with this approach is that critical information is lost as a consequence of deglycosylation. An interesting feature of a glycoprotein is that it may bear multiple lectin targetable features within an oligosaccharide at a single site or at several sites, which has been shown with concanavalin A, wheat germ agglutinin and jacalin columns4, 10-12. When lectin affinity columns



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

are loaded and eluted in series, the method is referred to as serial affinity chromatography (SAC). If a hypothetical glycoprotein bearing a β-D-galactosyl residue (A); a D-N-acetylgalactosaminyl residue (B); D-mannosyl groups (C); and an α-L-fucosyl residue (D) were loaded onto a SAC column series, it is clear that the selection pattern would vary depending on the order in which columns binding these monosaccharides were coupled. Alternatively the protein might exist as isoforms that have either ABC or BCD residues in the oligosaccharide at the site but no single species would have ABCD. Using four affinity columns that individually target A, B, C, or D respectively, it should be possible to differentiate between these cases. Among the many lectin targetable features of glycans a large number of SAC combinations will be possible across the glycoproteome. This may be of value in structure specific glycan selection from biological mixtures. Although having different glycan combinations will obviously impact separation characteristics, what is the biological significance? It has been recognized for many years that aberrations in protein glycosylation often accompany disease progression and that these aberrations can be targeted with lectins13,

14

. Moreover, glycan patterns within the same

glycoprotein synthesized in different tissues will vary, perhaps in association with disease15, 16. Glycosylation plays such a large part in disease because it plays a major role in cell growth17 and differentiation18, adhesion19, proliferation20, multiple diseases21, aging22, 23, and environmental stress24. Exploitation of this fact through LAC has now become a popular route in proteomics to capture and identify protein biomarkers with which disease-related glycans are associated. The problem with this approach is that lectin binding affinity varies widely along with selectivity toward glycan isoforms.


Page 4 of 28

Page 5 of 28

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


The focus of the work described here was to assess whether serial lectin and glycantargeting antibody affinity chromatography is of any value in assessing glycan occurrence and heterogeneity among glycoproteins in blood. Glycoproteins were chosen instead of glycopeptides for several reasons. Identification of a parent protein based on glycopeptides requires deglycosylation. Although that is relatively easy with N-glycosylated proteins, it is difficult to deglycosylate and identify O-glycosylated peptides. In contrast, affinity selected Nand O-glycosylated proteins are easily identified by their non-glycosylated peptides. There is also the issue that the requisite trypsin digestion necessary to select glycopeptides for identification destroys critically important structural information. That is not necessary in the glycoprotein capture method. Finally, affinity selected glycoproteins are still trypsin digested for identification. Glycopeptides from a protein can still be deglycosylated and identified with glycosylation sites at that time. The major advantage of this approach is that it can be demonstrated whether a targeted glycan appearing at one site in an affinity selected protein may not be present at other sites in the protein. That is not possible with the glycopeptides approach.

MATERIALS AND METHODS Materials and Chemicals. Agarose bound Lycopersicon esculentum lectin (LEL) sorbent was purchased from Vector Laboratories (Burlingame, CA, USA). Agarose bound Helix pomatia agglutinin (HPA) lectin sorbent was purchased from EY laboratory (San Mateo, CA, USA). Agarose-conjugated anti-Lewis x and anti-sialyl Lewis x IgM affinity sorbents were purchased from Santa Cruz Biotech (Santa Cruz, CA, USA). Normal pooled human plasma from 100 subjects was generously supplied by the National Institute of Standards and Technology (NIST, Gaithersburg, MD, USA). Acetic acid, sodium hydroxide, formic acid, calcium chloride,



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

magnesium chloride, and HPLC grade acetonitrile (ACN) were purchased from Mallinckrodt Chemicals (Phillipsburg, NJ, USA). Ammonium bicarbnonate, glycine, manganese chloride, proteomics grade N-p-tosyl-phenylalanine chloromethyl ketone (TPCK)-treated trypsin, N-αTosyl-L-lysine chloromethyl ketone hydrochloride (TLCK), 4-(2-hydroxyethyl)-1-piperazine ethanesulfonic acid (HEPES), iodoacetic acid (IAA), and L-cysteine were obtained from SigmaAldrich (St. Louis, MO, USA).

Dithiothreitol (DTT) and urea were provided by Bio-Rad

Laboratories (Hercules, CA, USA). PNGase F(glycerol free) was purchased from New England BioLabs (Ipswich, MA, USA). C18 microspin column was obtained from The Nest Group, Inc. (Southborough, MA, USA). HLB Oasis SPE cartridges were provided by Waters (Milford, MA, USA). The DI water system was purchased from Millipore (Boston, MA, USA). Centrivap Concentrator was purchased from Labconco, Corp. (Kansas city, MO, USA). Serial Affinity Chromatography (SAC). Agarose bound LEL and HPA sorbents were individually self-packed in 4.6 x 50 mm columns. Anti-Lewis x antibody (anti-LexAb) and antisialyl Lewis x antibody (anti-sLexAb) sorbents were also separately self-packed in 4.6 x 50 mm columns. Protein concentration in healthy human plasma was estimated using the Bradford assay to prepare samples with equal amounts of total proteins. For the first SAC experiment, the four columns, LEL, HPA, anti-LexAb, and anti-sLexAb columns, were connected in series (Table 1). Healthy human plasma was loaded directly onto serially connected soft-gel affinity columns with mobile phase A (0.10 M HEPES buffer, pH 7.5 containing 1 mM CaCl2 and 1 mM MgCl2) at a flow rate of 0.3 mL/min. Following extensive washing with mobile phase A to remove nonspecifically and weakly bound proteins, the serial affinity columns were disassembled and affinity selected proteins were eluted from each column individually with solution B (0.5M acetic acid-HCl solution, pH 2.5). Then four affinity columns were connected in the reversed


Page 6 of 28

Page 7 of 28

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


column order; anti-sLexAb→anti-LexAb→HPA→LEL columns. The same amount of human plasma was loaded directly onto serially connected soft-gel affinity columns in this order and bound proteins were eluted with solution B from each column separately after extensive washing. All the elution curves were obtained with an absorbance detector operating at 280 nm using a 20 AD LC systems from Shimadzu Scientific Instruments, Inc. (Kyoto, Japan). Saturation Test of Columns. A saturation test was done with an LEL column, because the LEL column had captured the highest amount of proteins from human plasma among four selectors based on the previous papers3, 25-28. The amount of injected sample was increased from 25 µg to 250 µg by 25 µg increments. The saturation amount of healthy human plasma samples was 200 µg of human plsma in the LEL column. (Data not shown) Therefore, the injection amount was decided as 100 µg of plasma below the saturation amount in order to prevent overloading. Proteolysis.

Captured proteins were digested as previously described25. The resulting

peptide mixture was desalted with HLB Oasis SPE cartridges, dried into 50–100 µL with the centrivap concentrator, after being adjusted to pH 7.5 with the 0.5 M NH4HCO3 buffer. PNGase F Digestion.

N-linked glycopeptides in a digested peptide mixture were

deglycosylated by treatment with PNGase F. A 50 mM ammonium bicarbonate buffer was added to the desalted tryptic peptides to adjust to pH 7.0 - 8.0. Five microliters (2500 U) of PNGase F were added to the pH-adjusted tryptic peptides and then incubated overnight at 37 oC. Following deglycosylation, samples were desalted using HLB Oasis SPE cartridges and concentrated using C18 microspin column. The PNGase F-treated peptide mixtures were reconstituted in 0.1 % formic acid solution, stored at -80 oC until analysis with LTQ-Orbitrap instrument.



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

LC-MS/MS based Protein Identification. Proteins were identified with an LTQ-Orbitrap hybrid mass spectrometer (Thermo Fisher, Bremen, Germany). The peptide mixtures resulting from PNGase F treatment were separated on an Agilent 1100 HPLC system using a 75 µm × 120 mm C18 RPC column packed with 5 µm C18 Magic beads. RPC separations were achieved using a 60 minutes linear mobile phase gradient from 98% solvent A with 2% solvent B to 60% solvent A with 40% solvent B at a flow rate of 300 nL/min. Solvent A was composed of DI water to which formic acid had been added to a concentration of 0.1%. Solvent B was prepared with ACN to which formic acid had been added to a concentration of 0.1%. The electrospray ionization emitter tip was generated on the pre-packed column with a laser puller (Model P-2000, Sutter Instrument Co.). The HPLC system was coupled directly to the LTQ Orbitrap hybrid mass spectrometer equipped with a nano-electrospray ion source (Proxeon Biosystems, Odense, Denmark). The MS was operated in the data-dependent mode, in which a survey full scan MS spectrum (from m/z 300 to 1600) was acquired in the Orbitrap with a resolution of 60,000 at m/z 400. This was then followed by MS/MS scans of the 3 most abundant ions with +2 to +3 charge states. Target ions already selected for MS/MS were dynamically excluded for 180 seconds. The resulting fragment ions were recorded in the linear ion trap. Automated MS/MS data analysis was performed using Protein Pilot Software 4.0 as described in the previous paper25. The identified proteins are listed according to their Swiss-Prot entry names and accession numbers in Table S-1 and Table S-2 of the Supporting Information.

RESULTS The focus of the work described here was to evaluate immobilized lectin and glycantargeting antibody columns coupled in series for the capture and differentiation of glycans in


Page 8 of 28

Page 9 of 28

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


glycoproteins from complex biological mixtures. Glycoprotein identifications were achieved subsequent to affinity chromatography in a series of steps involving trypsin digestion of affinity selected protein fractions, further fractionation of tryptic digests by RPC, electrospray ionization and mass spectral analysis of peptides by tandem mass spectrometry (ESI-MS/MS). The minimum criterion for identification of captured proteins was the presence of two or more nonglycosylated or deglycosylated peptides in a sample at a confidence level of at least 99% in two independent analyses. This protein-affinity selection has several advantages; one being the availability of large numbers of non-glycosylated peptides for protein identification and a second being that the deglycosylation step is precluded. Controlled deglycosylation is difficult with O-glycosylated proteins and peptides. Moreover, glycosylation sites in most glycoproteins are known. This study elucidates some N-glycosylation sites with the PNGase F treatment. Consequently, more N- and O- glycoproteins were identified by using non-glycosylated peptides as well as deglycosylated peptides than when only one approach was used (Table 2). A more detailed elucidation of structure at specific sites in a glycoprotein can be carried out on affinity selected glycoproteins through trypsin digestion of affinity selected glycoprotein fractions. Although not done in these studies, glycopeptides from all the glycopeptides in a protein bearing a specific glycan can be individually isolated and their glycan profiles established. This allows one to investigate whether single or other sites in a protein bear the same glycan profile. Four, relatively specific affinity selectors were examined in immobilized affinity chromatography columns; LEL, HPA, anti-LexAb, and anti-sLexAb. LEL binds to Nacetylglucosamine (GlcNAc) containing oligomers such as (GlcNAcβ1-4)1-4 whereas HPA has



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

Page 10 of 28

an affinity for terminal α-N-acetylgalactosamine (α-GalNAc) residues alone or in conjugation with galactose (GalNAcβ1-4Gal). Figure S-1 of the Supporting Information shows Lex and sLex antigen structures. Anti-LexAb binds to Lex antigen while anti-sLexAb (derived from clone CHO-131) used in this experiment binds sLex antigen coupled to a core 2(β1-6)-O-glycan through GlcNacβ1-6GalNAc branching. Interest in these affinity selectors stems from their use in recognizing various types of cancer at both tissue and blood levels29, 30. The binding capacity of affinity selectors was assessed qualitatively and quantitatively with 4.6 x 50 mm columns packed with immobilized LEL, HPA, anti-LexAb, and anti-sLexAb sorbents in four soft-gel columns individually. Increasing loads of healthy human plasma were applied to columns in 25 µg increments from 25 µg to 250 µg. Based on absorbance measurements with the four types of columns, the LEL column showed saturation at approximately 200 µg of human plasma. Healthy pooled plasma samples were examined through two combinations of SAC as shown in Table 1; one being the SAC series of LEL→HPA→anti-LexAb→anti-sLexAb and another being the anti-sLexAb→anti-LexAb→HPA→LEL combination. Four individual 4.6 x 50 mm SAC columns were coupled in series and loaded with plasma samples at a mobile phase velocity of 0.3 mL/min to preclude resin compression during loading and elution. After extensive washing to remove nonspecifically and weakly bound proteins, the serial assembly was disassembled and affinity selected proteins were individually eluted from each column with an acidic mobile phase at the same flow rate as the loading step. Typical loading and elution profiles are seen in Figure 1. Elution was monitored at 280 nm to minimize baseline perturbation from the acidic eluant. The relative amount of glycoprotein selected by any of these selectors was determined by relative absorbance measurements, assuming the extinction coefficients of all


Page 11 of 28

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


the proteins were the same. Each of the four affinity selectors captured less than 0.1% of the plasma proteome. The selected proteins were identified as described in the METHODS section. Overall Results. A total of 102 proteins were identified using the LEL→HPA→antiLexAb→anti-sLexAb series as seen in Figure 2. A more detailed protein lists can be found in Table S-1 of the Supporting Information. Among these proteins, 69 (68%) were captured by the LEL column, 45 (44%) with the HPA column, 48 (47 %) by the anti-LexAb column, and 40 (39 %) with the anti-sLexAb column. Of the total, 20 (20%) proteins were selected by each of the four selectors, 33 (32 %) proteins were not selected by the first (LEL) affinity column but were selected by the second, third, or fourth affinity columns, while 29 (28%) of the proteins were identified only from the LEL affinity column. Thirteen (13%) of the proteins were identified only by the HPA column, four (4%) proteins were identified only by the anti-LexAb column, and eight (8%) proteins were identified only by the anti-sLexAb column (Figure 2). Altering the order in which these four affinity selector columns were coupled produced very different results. With the anti-sLexAb→anti-LexAb→HPA→LEL serial combination, 86 proteins were identified, which is 16% less than the LEL→HPA→anti-LexAb→anti-sLexAb series in Figure 3. A more detailed protein lists can be found in Table S-2 of the Supporting Information. Of the 86 identified proteins, 52 (60%) were captured by the anti-sLexAb column, 24 (28%) by the anti-LexAb column, seven (8%) from the third (HPA) column, and 75 (87%) were captured by the LEL column. Of the total, six (7%) proteins were captured by each of the four selectors, and 34 (40%) of the proteins were not selected by the first anti-sLexAb affinity column but were selected by the second, third, or fourth affinity columns. With the antibody columns, four (5%) proteins were identified exclusively in the anti-sLexAb column, and six (7%)



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

Page 12 of 28

with the anti-LexAb column. One (1%) of the proteins was selected solely by the HPA column while 27 (31%) of the proteins were identified only from the LEL column.

DISCUSSION One of the greatest challenges in glycoprotein characterization is to understand the heterogeneity of oligosaccharide structures conjugated to glycoproteins. Enzymatic degradation and mass spectral methods allow structure analysis of glycans released from the glycoproteome, but information on the glycoproteins to which they were attached is lost during the process. Obviously, the glycan-protein conjugate is the biologically important entity, not the individual components of the conjugate. It is in this light that glycan-targeting antibodies and lectins have become important. They have allowed targeting and recognizing species in biological problems as diverse as cancer29, 31-34 and diabetes35, 36 to aging37. Clearly there is a need for fractionation methods that allow a global approach to glycoprotein heterogeneity. It was in this context that this research was undertaken. It is intriguing that doing nothing more than connecting well known lectin and antibody affinity chromatography columns in series and switching their order between studies is useful in glycoproteomics. The 102 proteins captured by the LEL→HPA→anti-LexAb→anti-sLexAb series are glycoproteins or proteins associated with the captured glycoproteins by protein-protein interactions (Figure 2 and Table S-1 of the Supporting Information). The 69 proteins captured by the LEL column either bore GlcNAc containing oligomers such as (GlcNAcβ1-4)1-4 or were associated with the glycoprotein(s) selected by the LEL column. It is also possible that proteins captured by the LEL column bear α-GalNAc residues or one of the Lewis antigens. This possibility was examined by switching the column order and placing the HPA column ahead of


Page 13 of 28

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


the LEL column as in the anti-sLexAb→anti-LexAb→HPA→LEL series of columns. A weakness of structure analysis by SAC is that there will be some false positives arising from proteins captured by the mechanisms such as protein-protein interactions other than the direct affinity selection of a targeted ligand. Human serum albumin (HSA) identified in this experiment may be a case of a protein associated with affinity captured glycoprotein(s). This is supported by the fact that HSA was not found in the second (anti-LexAb) and third (HPA) columns but was found in the fourth (LEL) column as seen in Table 3-b. If HSA was captured, it should clearly be captured in the second and third columns because HSA is the most abundant plasma protein. The affinity selectors used in this study have narrow selectivity. Thus the numbers of identified proteins from the SAC series are about 100, as expected based on the literatures and from the previous studies.26, 27, 28 Recognizing that GlcNAc containing oligomers have been stripped from the effluent exiting the LEL column, α-GalNAc bearing proteins were captured by the HPA column in the LEL→HPA→anti-LexAb→anti-sLexAb series. Effluent leaving the HPA columns will be stripped of GlcNAc and α-GalNAc bearing glycoproteins, and the proteins associated with these glycoproteins. This would apply to Lex and sLex antigen bearing proteins as well if they are co-resident in any of the proteins selected by the LEL and HPA columns. Again this possibility was examined by switching the order of the SAC columns. The Lex and sLex antigens are identical in their structures except for the presence of sialic acid in sLex (Figure S-1 of the Supporting Information). Fortunately the antibodies used in this work were of such very high specificity that they easily differentiate between the presence and absence of the sialic acid residue. This gives a high level of confidence to the Lewis antigen data in Table S-1 and Table S-2 of the Supporting Information. Among the 102 proteins found in this 13 ACS Paragon Plus Environment


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

Page 14 of 28

first SAC combination, 59 carried either the Lex, or sLex antigens, or both, or were associated with these glycoproteins after stripping GlcNAc and α-GalNAc bearing glycoproteins with the LEL and HPA columns. Nineteen of the selected proteins bore the Lex antigen alone while eleven exclusively carried the specific sLex antigen. Lex and sLex antigens occurred together in 29 (49%) proteins of the observed cases. It is interesting that these two similar antigens have been suggested to be synthesized in independent pathways38, 39. The fact that 20 (20%) of the proteins selected by the LEL→HPA→anti-LexAb→antisLexAb tandem column series were identified from each of the four columns while 33 (32 %) proteins were not selected by the first (LEL) affinity column but were selected by the second, third, or fourth affinity columns suggests a huge amount of heterogeneity within oligosaccharide structures bound to these proteins. The origin and significance of this heterogeneity are yet to be explained. Given that much of glycan structure remodeling occurs in Golgi, it is highly unlikely that such large diversity arose from a single set of Golgi. There is a high probability that this diversity is arising from Golgi in different organs and being mixed in plasma40. Altering the order in which these four affinity selector columns were coupled produced strikingly different results. Eighty-six proteins were identified with the anti-sLexAb→antiLexAb→HPA→LEL series, as shown in Figure 3 and Table S-2 of the Supporting Information. It is not clear why reversing the column order caused 16 fewer proteins to be identified. However, it is interesting that putting the anti-sLexAb column first in the SAC set led to the capture of 52 proteins as opposed to 40 when it was the last in the series. This is because at least twelve of the sLex bearing proteins also carried GlcNAc, α-GalNAc, or glycoproteins with Lex groups, so they were stripped from the effluent before entering the anti-sLexAb column in the LEL→HPA→antiLexAb→anti-sLexAb series. Placing the anti-LexAb column after the anti-sLexAb column in the 14 ACS Paragon Plus Environment

Page 15 of 28

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


anti-sLexAb→anti-LexAb→HPA→LEL series led to the capture of 24 proteins as opposed to 48 captured when it was placed before the anti-sLexAb column. This suggests that at least 24 of the sLex bearing proteins also carry the Lex antigen, or are associated with Lex bearing proteins. With the anti-sLexAb→anti-LexAb→HPA→LEL column series six proteins were selected exclusively by the anti-LexAb column whereas four were captured exclusively with the LEL→HPA→anti-LexAb→anti-sLexAb column series. The fact that seven proteins were selected by the HPA column in the anti-sLexAb→antiLexAb→HPA→LEL column set as opposed to 45 in the case of the LEL→HPA→antiLexAb→anti-sLexAb series suggests that at least 38 of the Lex and sLex bearing proteins also carry α-GalNAc, or were associated with α-GalNAc bearing proteins. In contrast, 75 proteins were captured by the LEL column with the anti-sLexAb→anti-LexAb→HPA→LEL column series whereas 69 were selected with the LEL→HPA→anti-LexAb→anti-sLexAb series. Although this makes it difficult to explain why six fewer proteins were captured when the LEL was the first column as opposed to the last column in the series, it is still relatively clear that GlcNAc is far less likely to occur with Lex and sLex than with α-GalNAc. The last (LEL) column of the second SAC combination captured more low abundance proteins than the first (LEL) column of the first SAC series, because, in the case of the first LEL column, low abundant proteins were suppressed by the high abundance proteins in the plasma. Twenty-seven proteins were identified as being captured exclusively with the anti-sLexAb→anti-LexAb→HPA→LEL column series, whereas 29 were captured exclusively with the LEL→HPA→anti-LexAb→antisLexAb series on the LEL column. Based on this study, it is clear that the SAC method can be of utility in biomarker-discovery for disease diagnosis through the comparison of the types and amount of specific glycoproteins



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

Page 16 of 28

between normal and disease plasma samples, especially when aberrations in protein glycosylation accompany disease progression as in cancer. There will be specific SAC combinations that are best for the enrichment, identification, and differentiation of specific disease-related proteins.

CONCLUSIONS It is concluded that SAC can be a valuable tool in recognizing diversity in protein glycosylation, especially when the order of columns in the SAC series is varied. It is further concluded that two clear types of diversity were recognized in this study. One is the independent occurrence of different affinity targetable glycan features in the same glycoprotein. The second is the case in which multiple targetable glycan features were co-resident in the same glycoprotein. The great advantage of this method is that it couples easily with current methods used in glycoproteomics.

ACKNOWLEDGEMENT This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (grant number 2013R1A1A3012431).

SUPPORTING INFORMATION AVAILABLE Protein lists identified by SAC as noted in the context are in Table S-1 and Table S-2 of the Supporting Information. Structures of Lewis x (Lex) and sialy-Lewis x (sLex) are in Figure S-1 of


Page 17 of 28

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


the Supporting Information. This material is available free of charge via the Internet at http://pubs.acs.org.



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

Page 18 of 28

Table 1. Two combinations of serial affinity chromatography

Combination 1

Selector 1 LEL

Selector 2 HPA

Selcetor 3 anti-LexAb

Slector 4 anti-sLexAb

2

anti-sLexAb

anti-LexAb

HPA

LEL

In the first combination, LEL, HPA, anti-LexAb, and anti-sLexAb columns were connected in series. Four affinity columns were also connected in the reversed order; anti-sLexAb→antiLexAb→HPA→LEL.

Table 2. Numbers of identified proteins and numbers of peptides used for the identification of those proteins in two combinations of SAC a) The first combination of SAC; LEL→HPA→anti-LexAb→anti-sLexAb Selector

Numbers of Identified Proteins

LEL HPA anti-LexAb anti-sLexAb

69 45 48 40

Numbers of Peptides used for Protein Identification Non-glycoslated 366 141 213 207

Deglycosylated 33 9 19 17

Total 399 150 232 224

b) The second combination of SAC; anti-sLexAb→anti-LexAb→HPA→LEL Selector x

anti-sLe Ab anti-LexAb HPA LEL

Numbers of Identified Proteins 52 24 7 75

Numbers of Peptides used for Protein Identification Non-glycoslated 278 89 26 429 18

ACS Paragon Plus Environment

Deglycosylated 28 6 0 49

Total 306 95 26 478

Page 19 of 28

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


Table 3. Numbers of identified proteins and numbers of peptides used for the identification of HSA in two combinations of SAC a) The first (LEL→HPA→anti-LexAb→anti-sLexAb) combination of SAC

a

Selector


Rankinga

LEL HPA anti-LexAb anti-sLexAb

69 45 48 40

4 -b 46 -b

HSA Numbers of Peptides used for Protein Identification 17 0 1 0

Ranking is the order of identified proteins in each selector with ranking number 1 being the

identified protein that used the most number of peptides and the number 69(LEL), 46(HPA), 47(anti-LexAb), and 39(anti-sLexAb) being the protein with the least number of peptides used. b

- indicates that HSA was not found.

b) The second (anti-sLexAb→anti-LexAb→HPA→LEL) combination of SAC

b c

Selector


Rankingc

anti-sLexAb

52

5

HSA Numbers of Peptides used for Protein Identification 14

anti-LexAb

24

-b

0

HPA

7

b

-

0

LEL

75

11

14

- indicates that HSA was not found.

Ranking is the order of identified proteins in each selector with ranking number 1 being the

identified protein that used the most number of peptides and the number 52(anti-sLexAb), 24(anti-LexAb), 7(HPA), and 75(LEL) being the protein with the least number of peptides used.



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

Page 20 of 28

FIGURE LEGENDS. Figure 1. (a) Loading Chromatogram on the LEL→HPA→anti-LexAb→anti-sLexAb SAC assembly. A 100 µg of healthy human plasma was loaded directly onto serially connected four soft-gel affinity columns with mobile phase A (0.10 M HEPES buffer, pH 7.5 containing 1 mM CaCl2 and 1 mM MgCl2) at a flow rate of 0.3 mL/min. (b) Elution Chromatogram of the LEL column after the SAC columns were disassembled. Affinity selected proteins by the LEL column were eluted with solution B (0.5M acetic acid-HCl solution, pH 2.5) after disassembled. Fractions were collected from 13 minutes to 30 minutes of retention time.

Figure 2. Numbers of proteins identified from LEL→HPA→anti-LexAb→anti-sLexAb SAC. Numbers in the black, red, green, and blue ovals show the proteins identified from LEL, HPA, anti-LexAb, and anti-sLexAb columns respectively. Of the 102 identified proteins in this series, 69 were captured by the LEL column shown as a black oval, 45 with the HPA column with a red oval, 48 by the anti-LexAb column as a green oval, and 40 with the anti-sLexAb column with a blue oval. Of the total, 20 proteins were selected by each of the four selectors, 33 proteins were not selected by the LEL affinity column but were selected by the HPA, the anti-LexAb, or the anti-sLexAb affinity columns, while 29 proteins were identified only by the LEL affinity column. Thirteen proteins were identified only by the HPA column, four proteins were identified only by the anti-LexAb column, and eight proteins were identified only by the anti-sLexAb column. A more detailed presentation of the protein lists can be found in the Table S-1 of the Supporting Information.


Page 21 of 28

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 3. Numbers of proteins identified from anti-sLexAb→anti-LexAb→HPA→LEL SAC. Numbers in the black, red, green, and blue rectangles show the proteins identified from LEL, HPA, anti-LexAb, and anti-sLexAb columns respectively. Of the 86 identified proteins in this series, 52 were captured by the anti-sLexAb column shown as a blue oval, 24 by the anti-LexAb column with a green oval, seven from the HPA column as a red oval, and 75 were captured by the LEL column with a black oval. Of the total, 6 proteins were captured by each of the four selectors, and 34 proteins were not selected by the first anti-sLexAb affinity column but were selected by the anti-LexAb, the HPA, or the LEL affinity columns. With the antibody columns, four proteins were identified exclusively in the anti-sLexAb column, and six with the anti-LexAb column. One proteins was selected solely by the HPA column while 27 proteins were identified only from the LEL column. A more detailed presentation of the protein lists can be found in in the Table S-2 of the Supporting Information.



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

Page 22 of 28

Figure 1.

(a) uV 3000000 2750000 2500000 2250000 2000000 1750000 1500000 1250000 1000000 750000 500000 250000 0 0.0

2.5

5.0

7.5

10.0

12.5

15.0

17.5

(b)


20.0

22.5

25.0

27.5

min

Page 23 of 28

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 2.



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 3.


Page 24 of 28

Page 25 of 28

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


REFERENCES. (1)

Qiu., R.; Regnier., F. E. Analytical Chemistry 2005, 77 7225–7231.

(2)

Madera., M.; Mechref., Y.; Klouckova., I.; Novotny., M. V. Journal of Proteome Research 2006, 5 2348–2363.

(3)

Drake., P. M.; Schilling., B.; Niles., R. K.; Prakobphol., A.; Li., B.; Jung., K.; Cho., W.; Braten., M.; Inerowicz., H. D.; Williams., K.; Albertolle., M.; Held., J. M.; Iacovides., D.; Sorensen., D. J.; Griffith., O. L.; Johansen., E.; Zawadzka., A. M.; Michael P. Cusack, S. A.; Gormley., M.; Hal., S. C.; Witkowska., H. E.; Gray., J. W.; Regnier., F.; Gibson., B. W.; Fisher., S. J. Journal of Proteome Research 2012, 11, 2508-2520.

(4)

Zeng., Z.; Hincapie., M.; Pitter., S. J.; Hanash., S.; Schalkwijk., J.; Hogan., J. M.; Wang., H.; Hancock., W. S. Analytical Chemistry 2011, 83, 4845-4854.

(5)

Donadio., S.; Pascual., A.; Thijssen., J. H. H.; Ronin., C. Clinical Chemistry 2006, 52, 286-297.

(6)

Wang., Y.; Ao., X.; Vuong., H.; Konanur., M.; Miller., F. R.; Goodison., S.; Lubman., D. M. Journal of Proteome Research 2008, 7 4313–4325.

(7)

Pan., S.; Wang., Y.; Quinn., J. F.; Peskind., E. R.; Waichunas., D.; Wimberger., J. T.; Jin., J.; Li., J. G.; Zhu., D.; Pan., C.; Zhang., J. Journal of Proteome Research 2006, 5 2769– 2779.

(8)

Kreunin., P.; Zhao., J.; Rosser., C.; Urquidi., V.; Lubman., D. M.; Goodison., S. Journal of Proteome Research 2007, 6 2631–2639.

(9)

Mao., X.; Luo., Y.; Dai., Z.; Wang., K.; Du., Y.; Lin., B. Analytical Chemistry 2004, 76 6941–6947.

(10)

Kullolli., M.; Hancock., W. S.; Hincapie., M. Journal of Separation Science 2008, 31, 2733-2739.

(11)

Dayarathna., M. K. D. R.; Hancock., W. S.; Hincapie., M. Journal of Separation Science 2008, 31, 1156-1166.

(12)

Plavina., T.; Wakshull., E.; Hancock., W. S.; Hincapie., M. Journal of Proteome Research 2007, 6, 662-671.



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)

Page 26 of 28

Ueda, K.; Fukase, Y.; Katagiri, T.; Ishikawa, N.; Irie, S.; Sato, T.-A.; Ito, H.; Nakayama, H.; Miyagi, Y.; Tsuchiya, E.; Kohno, N.; Shiwa, M.; Nakamura, Y.; Daigo, Y. PROTEOMICS 2009, 9, 2182-2192.

(14)

Julka, S.; Regnier, F. Journal of Proteome Research 2004, 3, 350-363.

(15)

Dube, D. H.; Bertozzi, C. R. Nature Reviews. Drug Discovery 2005, 4, 477-488.

(16)

Caterson, B.; Mahmoodian, F.; Sorrell, J. M.; Hardingham, T. E.; Bayliss, M. T.; Carney, S. L.; Ratcliffe, A.; Muir, H. Journal Of Cell Science 1990, 97 ( Pt 3), 411-417.

(17)

Julien., S.; Adriaenssens., E.; Ottenberg., K.; Furlan., A.; Courtand., G.; VercoutterEdouart., A.-S.; Hanisch., F.-G.; Delannoy., P.; Bourhis., X. L. Glycobiology 2006, 16, 54-64.

(18)

Cao., R.; Chen., K.; Song., Q.; Yi Zang, J. L.; Wang., X.; Chen., P.; Liang., S. Journal of Proteome Research 2012, 11 829–838.

(19)

Gu., J.; Sato., Y.; Kariya., Y.; Isaji., T.; Taniguchi., N.; Fukuda., T. Journal of Proteome Research 2009, 8 431–435.

(20)

Singh., R.; Subramanian., S.; Rhodes., J. M.; Campbell., B. J. Glycobiology 2006, 16, 594-601

(21)

Telford., J. E.; Bones., J.; McManus., C.; Saldova., R.; Manning., G.; Doherty., M.; Leweke., F. M.; Rothermundt., M.; Guest., P. C.; Rahmoune., H.; Bahn., S.; Rudd., P. M. Journal of Proteome Research 2012, 11, 3743-3752.

(22)

Stewart., E. M.; Aquilina., J. A.; Easterbrook-Smith., S. B.; Murphy-Durland., D.; Jacobsen., C.; Moestrup., S.; Wilson., M. R. Biochemistry 2007, 46, 1412-1422.

(23)

Robinson., L. J.; Karlsson., N. G.; Weiss., A. S.; Packer., N. H. Journal of Proteome Research 2003, 2, 556-557.

(24)

Chandramouli., K. H.; Zhang., Y.; Wong., Y. H.; Qian., P.-Y. Journal of Proteome Research 2012, 11, 1330-1340.

(25)

Cho., W.; Jung., K.; Regnier., F. E. Analytical Chemistry 2008, 80, 5286-5292.

(26)

Jung., K.; Cho., W.; Regnier., F. E. Journal of Proteome Research 2009, 8, 643-650.

(27)

Cho., W.; Jung., K.; Regnier., F. E. Journal of Proteome Research 2010, 9, 5960-5968.

(28)

Cho., W.; Jung., K.; Regnier., F. E. Journal of Separation Science 2010, 33, 1438-1447.


Page 27 of 28

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


(29)

Saeland., E.; Vliet., S. J. v.; Bäckström., M.; M., V. C.; Berg., V. d.; Geijtenbeek., T. B. H.; Meijer., G. A.; Kooyk., Y. v. Cancer Immunology, Immunotherapy 2007, 56, 12251236.

(30)

Avent., N.; Judson., P. A.; Parsons., S. F.; Mallinson., G.; Anstee., D. J.; Tanner., M. J.; Evans., P. R.; Hodges., E.; Maciver., A. G.; Holmes., C. Biochemical Journal 1988 251, 499–505.

(31)

Zhang., X.; Teng., Y.; Fu., Y.; Xu., L.; Zhang., S.; He., B.; Wang., C.; Zhang., W. Analytical Chemistry 2010, 82

(32)

Li., Y.; Tao., S.-C.; Bova., G. S.; Liu., A. Y.; Chan., D. W.; Zhu., H.; Zhang., H. Analytical Chemistry 2011, 83 8509–8516.

(33)

Belle., J. T. L.; Gerlach., J. Q.; Svarovsky., S.; Joshi., L. Analytical Chemistry 2007, 79 6959–6964.

(34)

Wu., J.; Xie., X.; Liu., Y.; He., J.; Benitez., R.; Buckanovich., R. J.; Lubman., D. M. Journal of Proteome Research 2012, 11 4541–4552.

(35)

Chen., H.-H.; Wu., C.-H.; Tsai., M.-L.; Huang., Y.-J.; Chen., S.-H. Analytical Chemistry 2012, 84 8635–8641.

(36)

Bouwman., L. H.; Eerligh., P.; Terpstra., O. T.; Daha., M. R.; Knijff., P. d.; Ballieux., B. E. P. B.; Bruining., G. J.; Slik., A. R. v. d.; Roos., A.; Roep., B. O. Diabetes 2005 54 3002-3006.

(37)

Jaisson., S.; Gillery., P. Clinical Chemistry 2010, 56, 1401-1412.

(38)

Saldova R; Wormald MR; Dwek RA; PM., R. Disease Markers 2008, 25(4,5), 219-232.

(39)

St Hill Catherine A; Farooqui M; Mitcheltree G; Gulbahce HE; Jessurun J; Cao Q; B., W. BioMedCentral cancer 2009, 9, 79.

(40)

Freeze, H. H. Nat Rev Genet 2006, 7, 537-551.



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

For TOC only


Page 28 of 28

Serial Affinity Chromatography as a Selection Tool in Glycoproteomics

Recommend Documents