Glycobiology Advance Access originally published online on October 24, 2008
Glycobiology 2009 19(2):144-152; doi:10.1093/glycob/cwn116
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
O-Fucosylation of an antibody light chain: Characterization of a modification occurring on an IgG1 molecule
2 Department of Analytical and Formulation Sciences
3 Department of Molecular Sciences, Amgen, 1201 Amgen Court West, Seattle, WA 98119-3105, USA
1 To whom correspondence should be addressed: Tel: +(206)-265-7426; Fax: +(206)-217-0492; E-mail: byan{at}amgen.com
Received on August 3, 2008; revised on September 29, 2008; accepted on October 21, 2008
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterial and methodsFundingConflict of interest statementReferences |
|---|
We describe the characterization of an O-fucosyl modification to a serine residue on the light chain of a recombinant, human IgG1 molecule expressed in Chinese hamster ovary (CHO) cells. Cation exchange chromatography (CEX) and hydrophobic interaction chromatography (HIC) were used to isolate a Fab population which was 146 Da heavier than the expected mass. Isolated Fab fragments were treated with a reducing agent to facilitate mass spectrometric analysis of the reduced light chain (LC) and fragment difficult (Fd). An antibody light chain with a net addition of 146 Da was detected by mass spectrometric analysis of the modified Fab. A light chain tryptic peptide in complementarity determining region-1 (CDR-1) was subsequently identified with a net addition of 146 Da by a peptide map. Results from a nanospray infusion of the modified peptide into a linear ion trap mass spectrometer with electron transfer dissociation (ETD) functionality indicated that the modified residue was a serine at position 30 in the light chain. Acid hydrolysis of the modified tryptic peptide followed by fluorescent labeling with 2-aminoanthranilic acid (2AA) and HPLC comparison with monosaccharide standards confirmed the presence of fucose on the light chain peptide. The presence of O-fucose on an antibody has not been previously reported. Currently, O-fucose has been described as occurring on mammalian proteins with amino acid sequence motifs associated with epidermal growth factor (EGF)-like repeats or thrombospondin type 1 repeats (TSRs). The amino acid sequence around the modified Ser in the IgG1 molecule does not conform to any known O-fucosylation sequence motif and thus is the first description of this type of modification on a nonconsensus sequence in a mammalian protein.
Key words: antibody / ETD mass spectrometry / light chain / monosaccharide analysis / O-fucosylation
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsFundingConflict of interest statementReferences |
|---|
The presence of fucose residues linked to proteins through the hydroxyl groups on serine and threonine side chains is a rare post-translational modification. O-Fucosylation was initially found on urinary plasminogen activator (u-PA) obtained from human samples (Kentzer et al. 1990
; Buko et al. 1991). O-Fucose on recombinantly expressed molecules was described initially by Harris et al. (1991) in their analysis of tissue-type plasminogen activator (t-PA). Subsequent analyses indicated that O-fucosylation could occur on molecules bearing epidermal growth factor (EGF)-like repeats (Harris and Spellman 1993). More recently, O-fucosylation has been discovered on molecules bearing thrombospondin type 1 repeats (TSRs) (Hofsteenge et al. 2001; Gonzalez de et al. 2002). O-Fucosylated glycans on EGF-like repeats may be present as a single fucose linked 
1 to Ser/Thr (Bjoern et al. 1991) or as a more complex structure composed of Sia-2,3/6-Gal-β1,4-GlcNAc-β1,3-Fuc (Harris et al. 1993; Moloney et al. 2000). O-Fucosylation on TSRs occurs as part of a disaccharide composed of Glc-β1,3-Fuc that is linked 1 to Ser/Thr (Moloney et al. 1997; Moloney and Haltiwanger 1999; Wang et al. 2007).
The analytical characterization of N-linked glycosylation is fairly straightforward due to the well-established consensus sequence of NXS/T where X is not proline (Bause and Hettkamp 1979). However, O-linked glycosylation is more difficult to predict as this modification is not associated with a single universal consensus site. The consensus sequence of XTPXP has been described for O-glycosylation with N-acetylgalactosamine on glycoproteins bearing mucin-type motifs (Yoshida et al. 1997; Tetaert et al. 2001), and consensus sequences of PVST and TTA have been proposed for O-glycosylation with N-acetylglucosamine (Vosseller et al. 2006). On EGF-like repeats, the consensus sequence for O-linked fucosylation is CXXGG(T/S)C (Harris and Spellman 1993). The consensus sequence proposed for O-fucosylation of TSRs was initially determined to be CSX(S/T)CG (Hofsteenge et al. 2001) and later refined to WX5CX2/3S/TCX2G based on the analysis of O-fucosylation occurring on recombinant F-spondin (Gonzalez de et al. 2002).
Determination of O-glycan site occupancy is challenging because the sugar moiety is highly labile under typical collision-induced dissociation (CID)-MS conditions. N-Terminal sequence analysis using Edman degradation chemistry has been successfully used to identify the occupied serine or threonine residue following degalactosialylation by treatment with a trifluoromethanesulfonic acid (TFMSA) anisole mixture (Gerken et al. 1997). O-Glycan site occupancy has been determined by chemical release of the oligosaccharide at alkaline pH (β-elimination) followed by CID-MS detection of dehydroalanine and dehydrobutyrate resulting from the occupied serine or threonine residues. The ethylene bond created during β-elimination of glycans from serine and threonine residues has been modified via covalent Michael-type addition with bifunctional thiol containing molecules which can then be used as the basis for affinity purification of formerly glycosylated peptides (Wells et al. 2002). In some reports, hybrid quadrupole-time of flight (Q-TOF) mass spectrometry has been used to determine the occupied serine or threonine residues. Such analysis requires careful optimization of ESI parameters and collision energy to preserve the glycan on the molecule during CID MS (Alving et al. 1999; Chalkley and Burlingame 2001, 2003; Hofsteenge et al. 2001). Recently, electron transfer dissociation (ETD) mass spectrometry has proved to be a superior approach to identify amino acid sites of attachment due to its capacity for maintaining labile post-translational modifications (Syka et al. 2004). In these analyses, fragmentation of the peptide is initiated by the transfer of an electron to the multiply charged peptide from an odd electron anion. Radical-induced fragmentation occurs preferentially along the peptide backbone generating c- and z-type ions while labile covalent modifications are maintained at the modified residue. Subsequently, ETD mass spectrometry has been applied to the analysis of O-glycosylated proteins and peptides and found to be appropriate for maintaining the glycan at the attached Ser/Thr residue during MS/MS experiments (Mikesh et al. 2006). The advent of ETD mass spectrometry has considerably simplified site identification of O-glycosylated proteins and peptides.
The work presented here to identify the O-fucose modification on an IgG1 antibody began initially as an inquiry into the charge-based modifications that were affecting the cation exchange (CEX) profile of the molecule. During the course of orthogonal subfractionation of intact and papain-digested molecules, it was found that a small population of molecules contained a 146 Da modification in the light chain. Mass spectrometric analyses of modified populations revealed that a fucose was attached to a serine in CDR-1 of the light chain.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsFundingConflict of interest statementReferences |
|---|
The IgG1 antibody used in this study was a human recombinant molecule expressed in Chinese hamster ovary (CHO) cells. Examination of the primary sequence indicated that the heavy chain and light chain of the antibody contained N-terminal glutamine (Gln) and glutamic acid (Glu), respectively. N-Terminal sequencing data indicated that the N-terminus of the heavy chain was cyclized and thus resistant to Edman degradation chemistry. The peptide map data indicated that 98% of the heavy chain was cyclized to pyroglutamic acid (data not shown).
HIC isolation and analytical characterization of a modified light chain
The acidic variants isolated by CEX chromatography were compared to the CEX main peak after digesting the molecules with papain and analyzing the resulting Fab and Fc subdomains by hydrophobic interaction chromatography (HIC). The results from the HIC comparison indicated that an early eluting species was enriched in the acidic variants (Figure 1, peak A). In order to fully characterize the early eluting HIC peak, the CEX acidic variants were digested with papain and preparatively purified by HIC. Preparative isolation of peaks A and B (Figure 1) by HIC yielded sufficient material for subsequent mass spectrometric characterization. Mass analysis of peak A indicated that it was comprised of two species: a molecule with a mass that was consistent with Fab containing an uncyclized heavy chain (47,749.25 Da, Figure 2A) and a molecule with a mass that was 146 Da heavier than Fab with a cyclized heavy chain N-terminus (47,877.22 Da, Figure 2A). The mass of HIC peak B was 47,731.28 Da, a value consistent with a Fab having a heavy chain which is cyclized to pyroglutamate on the N-terminus (Figure 2B). Mass analysis of the HIC-purified molecules reduced with Tris[2-carboxyethyl] phosphine (TCEP) indicated that a species with a mass that was 146 Da heavier (23,539.39 Da, Figure 2C) than the observed light chain mass of 23,393.37 Da (Figure 2D) was also enriched in the CEX acidic variants. The observation of a molecule with a mass that was 146 Da heavier than the predominant Fab species and a molecule that was 146 Da heavier than the light chain following reduction of the disulfides suggested that there was a light chain population in the CEX acidic variants with a 146 Da modification.
|
|
Capillary scale LC-MS was used to analyze tryptic peptide maps of HIC peak A which contained the presumptive modified Fab and HIC peak B which contained unmodified Fab. The results indicated that the third tryptic peptide (L3) in the light chain contained an apparent addition of 146 Da to its base mass. In order to retain the earlier eluting unmodified peptide, the column temperature was lowered to 40°C and sample loading occurred in the absence of acetonitrile. The mobile phase TFA concentration was also found to be important for separating the modified L3 peptide from a closely eluting heavy chain peptide. A TFA concentration of 0.03% in the mobile phases was found to be optimal for this purpose. Based on the peptide map analysis of the early eluting HIC Fab peak, the level of modified L3 peptide was found to be 33.2% (Figure 3). Although this modification was easily enriched by HIC separation of the papain-digested molecules, the pre-main peak which contained the modified Fab only constituted 3.6% of the total Fab population in the acidic variants.
|
Quantitation of a modified light chain in an unfractionated antibody
The level of the light chain modification in the bulk drug substance (BDS) was assessed by the analytical scale peptide map analysis of trypsin-digested BDS. As was the case for the capillary scale peptide maps, the light chain L3 peptide carrying the 146 Da modification eluted after the unmodified peptide. Extracted ion current quantitation of modified L3 using the 2+ ion indicated that it was present at the approximate level of 0.76% (assuming equal ionization efficiencies of the modified and unmodified peptides) in the BDS. MS2 fragmentation of the modified L3 peptide 2+ ion at m/z of 432.8 yielded unmodified L3 as the major product ion due to the loss of the apparently labile 146 Da modification (Figure 4). Subsequent MS3 fragmentation of the dominant 2+ product ion at m/z of 359.8 yielded b- and y-type ion fragments which were nearly identical to the MS2 spectrum derived from fragmentation of unmodified L3 peptide 2+ ion at m/z of 359.7. The sequence-specific ions obtained in the MS3 analysis of the modified peptide confirmed that the 146 Da modification could be localized to the 7-residue sequence of L3. However, the results also indicated that the 146 Da modification was highly labile and thus could not be definitively assigned to a particular residue using CID-MS.
|
Preparative isolation and ETD-MS analysis of L3 species
The observed lability of the 146 Da modification during standard CID-MS experiments necessitated the use of ETD mass spectrometry for identification of the modified residue in the L3 peptide. ETD functionality enables peptide dissociation by transferring electrons to positively charged peptides leading to a rich ladder of sequence ions derived from cleavage at the amide groups along the peptide backbone to produce c and z ions. Amino acid side chains and labile post-translational modifications are preserved intact and can be readily assigned as a result of this analysis. Eight milligrams of unfractionated mAb was digested with trypsin to isolate L3 and modified L3 peptides for subsequent characterization. As described previously, the preparative reversed-phase (RP)-HPLC separations were carried out in relatively low levels of TFA in the mobile phases as well as low column temperatures to limit the loss of the modification from acid hydrolysis. One milligram of the digested material was loaded onto an RP-HPLC column and eluted as described and the composition of the collected fractions was determined by matrix-assisted laser desorption ionization time of flight mass spectrometry (MALDI-TOF MS). The purified L3 peptides were infused into the mass spectrometer by static nanospray and ETD-MS2 of the modified L3 peptide at m/z of 433.0 resulted in a clear z-ion series covering the entire sequence (Figure 5). The unmodified L3 peptide at m/z of 359.0 was also analyzed by nanospray ETD-MS2 to facilitate a direct comparison of the product ions resulting from the modified and unmodified peptides. At low m/z, the dominant species observed were z-type ions that carried two positive charges and an electron derived from the anionic ETD reagent, fluoranthene, giving an apparent m/z that was indistinguishable from the singly charged C13 isotope for each fragment. The z5 and z6 ions were singly charged and did not carry an electron; thus the apparent m/z agreed with the theoretical value for the singly charged monoisotopic species. A full z-ion series was observed for both the unmodified and modified L3 peptides. The modified z2 ion and unmodified z1 ion in the ETD-MS2 spectrum obtained from the purified L3 peptide containing the 146 Da modification indicated that this modification was associated with the second Ser in the sequence ASQGISR.
|
Fluorescent-labeled monosaccharide analysis of modified L3
The observation of the 146 Da modification on the Ser residue was highly suggestive of O-fucosylation. However, based on previous reports already described, the lack of a consensus sequence for O-fucosylation around the modified residue required confirmation by monosaccharide analysis. L3 and modified L3 peptides as well as several monosaccharide standards were subjected to acid hydrolysis in the presence of TFA followed by labeling with the fluorescent tag 2-aminoanthranilic acid (2AA). RP-HPLC analysis of the fluorescent products revealed a peak from modified L3 which had the same retention time as the labeled fucose standard indicating that the source of the 146 Da modification was in fact an O-linked fucose (Figure 6). The absence of this species in the unmodified L3 hydrolysis product confirmed that this peak was due to a fucose monosaccharide and not a side reaction resulting in a labeled peptide.
|
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsFundingConflict of interest statementReferences |
|---|
Different mechanisms, either enzymatic or nonenzymatic, could lead to a single monosaccharide addition on a protein. Nonenzymatic sugar addition, glycation, results from the condensation of an amino group with the reducing group of the carbohydrate. Monosaccharide is incorporated on either the N-terminus or epsilon amino group of lysine residues (Neglia et al. 1983
). Enzymatic addition of a monosaccharide on proteins is generally associated with O-linked glycosylation of serine and threonine residues or, more rarely, with C-linked glycosylation (Hofsteenge et al. 1994). On the basis of the information found in the scientific literature, we believe that the fucosylated Ser we have observed on the light chain is likely the result of the mechanistic action of an O-fucosyltransferase enzyme. The GDP-fucose protein O-fucosyltransferase enzymes catalyze the transfer of fucose from GDP-fucose to the side-chain hydroxyl of an acceptor Ser or Thr residue that is part of an O-fucosylation sequence motif. In the case of EGF-like repeats, transfer of the fucose to the protein is catalyzed by POFUT1 which was initially identified in CHO cells (Wang et al. 1996; Wang and Spellman 1998) and subsequently in humans (Wang et al. 2001). While an O-fucosylation consensus sequence motif is required for constitutive O-fucosylation of the target Ser or Thr residue, Wang et al. have shown previously that peptides with the O-fucosylation sequence motif were not sufficient acceptor molecules and that a properly folded EGF-like repeat was necessary (Wang and Spellman 1998). Luo et al. found that fucosylation of TSRs is mediated by POFUT2 and that a correctly folded TSR was necessary for O-fucosylation of the Ser/Thr within the consensus sequence (Luo, Koles, et al. 2006). The divergent pathways for the initial O-fucosylation event occurring on Ser/Thr for EGF-like repeats and TSRs are maintained in the subsequent elongation of the monosaccharide to more complex structures. O-Fucose on EGF-like repeats can be elongated to a tetrasaccharide from the action of fringe, a fucose-specific 1,3-GlcNAc transferase (Moloney et al. 2000). The findings of Rampal et al. (Rampal et al. 2005) indicated that the elongation of O-fucosylated EGF-like repeats may be carried out by mammalian homologs of fringe: lunatic fringe, manic fringe, and radical fringe each recognizing a particular amino acid sequence on O-fucosylated EGF-like repeats. Elongation of O-fucose on TSRs to a disaccharide is a similarly specific event but, unlike elongation of O-fucosylated EGF repeats, appears to require particular secondary structures as opposed to amino acid sequences (Kozma et al. 2006).
Although the sequence around the modified light chain Ser residue in our antibody did not conform to a known O-fucosylation sequence motif, the presence of an O-fucosylated population extended to either a disaccharide or a tetrasaccharide could provide information about the class of transferases that acted upon the modified residue and hence information about whether POFUT1 or POFUT2 was responsible for the initial modification. This would then indicate that the secondary structure of the light chain CDR had some local structural homology to EGF-like repeats or TSRs. We did not detect any low level populations in our mass spectral data that would be consistent with the extension of the fucosylated Ser residue to a disaccharide or tetrasaccharide. While disaccharides are observed on O-fucosylated TSRs in mammalian cells (Hofsteenge et al. 2001; Gonzalez de et al. 2002; Luo, Nita-Lazar, et al. 2006), the modification by a single O-fucose monosaccharide has been observed on EGF-like repeats as illustrated by the example of factor VII, where O-fucose is present as a monosaccharide (Bjoern et al. 1991). We must note that O-fucose on another EGF domain containing molecule of the clotting cascade, factor IX, is observed as the full-length tetrasaccharide mentioned above (Nishimura et al. 1992). In our case, the absence of a tetrasaccharide structure on the antibody light chain is not conclusive evidence against O-fucosylation resulting from the action of POFUT1. The absence of O-fucose monosaccharide on TSRs, however, does suggest that the O-fucosylated light chain did not result from the action of POFUT2.
Apart from the well-characterized occurrence of O-fucosylation on EGF-like repeats and TSRs, there have been instances of this modification occurring on proteins which do not adhere to the binary categorization that has been described. The occurrence of O-fucose monosaccharide on a Thr residue in the insect protein PMP-C isolated from Locusta migratoria has been described by Nakakura et al. (1992). The amino acid sequence associated with the modification does not match the consensus sequence for O-fucosylation of EGF-like repeats or TSRs and would seem to be a unique case from which a corollary to our results could be drawn. However, the sequence surrounding the O-fucosylated residue in PMP-C has virtually no similarity to the IgG1 light-chain sequence on which we have observed the O-fucose modification. The IgG1 modeling structure used in this study was derived from the work of Saphire et al. (2001), and the Ser at position 30 on CDR-1 of the LC is found located in a loop structure which is solvent exposed (details not shown). The structural feature in the IgG1 appears to be different from the strand structure that contains O-fucosylated Thr residue in PMP-C. It should be noted that there is an extraordinary variability in the CDR sequences of antibodies which reflects the great diversity of epitopes available in biological systems. While N-glycosylation of antibody variable regions requires a consensus sequence and has been extensively reported in Spiegelberg et al. (1970) and Tomlinson et al. (1992)), examples of O-glycosylation, particularly with a single monosaccharide, are less common (Martinez et al. 2007). At this time, we are unaware of any CDR sequences which contain a consensus sequence for O-fucosylation.
Despite the significant progress made in identifying O-fucosylation and understanding the biological processes that cause it, many areas relating to O-fucosylation are presently unknown. Our results indicate that the IgG1 antibody we have described is O-fucosylated on a light-chain Ser residue at position 30 at a total level of 0.76%. Our findings are unique in that we describe a modification to a mammalian protein occurring on an amino acid residue that is not part of a known O-fucosylation sequence motif. While the occurrence of O-fucosylation is a rare event, analytical strategies that could be used to screen for this modification are not well developed at this time and this may contribute to underreporting of this phenomenon. Isolation of sugars and nucleic acids through boronate affinity was initially described by Weith et al. (1970) and more recently, isolation of glycated recombinant antibodies by boronate chromatography has been described (Brady et al. 2007). However, we were unable to enrich the O-fucosylated species using these techniques. It is possible that this negative result may have occurred due to inaccessibility of the modified light chain to the boronate resin. Lectin affinity chromatography has been used to isolate several different classes of glycans on the intact molecule and at the peptide level but to the best of our knowledge there are currently no known O-fucose-specific lectins. We believe that the lack of an affinity chromatographic-based technique for O-fucose suggests that chromatographic techniques which are sensitive to changes in molecule hydrophobicity and charge profile, e.g., HIC and CEX chromatography may be the most appropriate techniques currently available for enrichment and identification of O-fucosylation sites in molecules. We believe that the emergence of more resolving technical procedures will shed light on underreported post-translational modifications such as the one we described for this human antibody.
|
Material and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsFundingConflict of interest statementReferences |
|---|
Materials
Human Recombinant Antibody
The antibody used in the study is a human recombinant antibody of the IgG1 subclass. The molecule was expressed in CHO cells and chromatographically purified using conventional techniques.
Preparative Fractionation Cation Exchange Chromatography
Weak CEX separation of the intact IgG1 was performed using a Dionex ProPac WCX-10 column, 22 x 250 mm (Dionex, Sunnyvale, CA) and an AKTA HPLC (GE Amersham Biosciences, Uppsala, Sweden). Antibody populations were isolated by gradient elution with a phosphate buffered sodium chloride solution. All fractions were buffer exchanged into the original mildly acidic buffer, using Amicon Ultra centrifugal filter devices with a molecular weight cutoff (MWCO) of 30 kDa (Millipore, Billerica, MA). With this purification step, we obtained an acidic variants fraction with 95% purity.
HIC Fractionation
The CEX-isolated acidic variants were digested with papain according to previous methods (Yan et al. 2007) and isolated preparatively by HIC. The papain-digested molecule was bound to two Dionex ProPac HIC-10 columns connected in series in the presence of 1.5 M ammonium sulfate, 20 mM sodium acetate, pH 5.2, and eluted with a linear gradient from 0 to 100% buffer B over 60 min (buffer A: 1.5 M ammonium sulfate, 20 mM sodium acetate, pH 5.2, and buffer B: 20 mM sodium acetate, pH 5.2) at a flow rate of 0.4 mL/min. The separation was monitored by UV absorption at 220 nm. HIC fractionation was carried out on an Agilent HP1200 Quaternary HPLC. Purified fractions were pooled and buffer exchanged into 20 mM sodium acetate (pH 5.3) by centrifugation in Amicon centriprep 15 mL filter units with a 30 kDa MWCO (Millipore, Billerica, MA).
Intact and Reduced Mass Analysis
HIC fractions containing papain-digested antibody subdomains were analyzed by SEC-MS according to Brady et al. (2008). Selected fractions were reduced by incubation at 55°C for 30 min in the presence of 4 M guanidine HCl, 50 mM sodium acetate, pH 5.0, 50 mM TCEP (Pierce, Rockford, IL). Briefly, 0.5–5 µg of reduced antibody subdomains or 10 µg of intact Fab or Fc was injected onto a polyhydroxyethyl aspartamide column (PolyLC, The Nest Group, Southborough, MA) in 0.1% formic acid at a flow rate of 0.1 mL/ min and 98% acetonitrile; 2% formic acid was introduced postcolumn with a "T" union and mixed with the HPLC eluent entering the electrospray ionization (ESI) source. The mass of the desalted reduced or intact antibody was determined with an Agilent LC-MSD time-of-flight instrument (Agilent, Santa Clara, CA). The mass spectrometer capillary voltage, declustering potential, and octapole RF voltage were set to 4000, 300, and 300 V, respectively. Raw MS data were deconvoluted with Agilent MassHunter Qualitative Analysis software version B.01.02.
Capillary Peptide Map Analysis
Aliquots of 20 µg of each purified HIC fraction were dried by vacuum centrifugation (GMI, Ramsey, MN) to a final volume of 2–3 µL. Samples were denatured by the addition of 10 µL of 7.0 M guanidine HCl, 0.1 M 2-amino-2-(hydroxymethyl)-1,3-propanediol, hydrochloride (Tris–HCl) at pH 8.3 and heating at 75°C for 15 min. A volume of 20 µL of 0.1 M Tris–HCl at pH 8.0 was added to each sample, thus diluting the final guanidine concentration to 1.4 M. Each sample was incubated with 2 µg of recombinant proteomics grade trypsin (Roche, Basel, Switzerland) for 5 h at 37°C at which time the samples were frozen at –20°C until analysis. Capillary peptide map separations were carried out on an Agilent HP1100 capillary HPLC. The peptides were detected by UV absorption at 215 nm. Four micrograms of the digested peptides was separated on a Varian Polaris Ether C18 (0.32 x 150 mm, 3 micron particle) column (Varian, Palo Alto, CA) at a solvent flow rate of 8 µL/min. Upon sample injection, the column was held at 100% solvent A (0.03% trifluoroacetic acid, TFA, Pierce, Rockford, IL) for 15 min, and then eluted with a linear gradient to 50% solvent B (80% acetonitrile, 0.03% TFA) over 85 min with the column temperature maintained at 40°C. Peptides were identified by data-dependent MS2 and MS3 fragmentation using a Thermo LTQ mass spectrometer (Thermo Fisher, Waltham, MA).
Analytical Peptide Map Analysis and Preparative Isolation of Fucosylated Peptide
Two hundred micrograms of IgG1 was added to 100 µL of 4.0 M guanidine HCl, 57 mM Tris, pH 8.3 (final volume), and the disulfides were reduced by the addition of dithiothreitol (DTT, Roche, Basel, Switzerland) to a final concentration of 10 mM and incubated at 55°C for 30 min. Reduced cysteines were alkylated with the addition of iodoacetamide (Sigma, St. Louis, MO) to a final concentration of 22 mM and incubated at room temperature in the dark for 15 min. Prior to digestion, the samples were buffer exchanged into 50 mM Tris, pH 7.5, using biospin 6 columns (Biorad, Hercules, CA) according to manufactures instructions. Recombinant proteomics grade trypsin was added to samples at an enzyme to protein ratio of 1:10 (w/w) and digestion occurred at 37°C for 4 h at which time samples were frozen at –20°C until analysis. Analytical peptide maps consisted of loading 100 µg of the digest onto a Varian Polaris Ether C18 (2.0 x 250 mm, 3 micron particle) column heated to a temperature of 50°C. The separation was performed by gradient elution on an Agilent HP 1200 HPLC. The column was held under the initial conditions of 0.5% solvent A (0.02% TFA) at a flow rate of 0.2 mL/min for 5 min at which time the peptides were eluted with a linear gradient to 60% solvent B (80% acetonitrile, 0.02% TFA) over 160 min. Peptides were identified by CID-MS as previously described. The sample preparation procedure for preparative isolation of fucosylated peptide consisted of a reducing and alkylating agent as described previously at an antibody concentration of 10 mg/mL. Prior to digestion, the samples were buffer exchanged into 0.2 M guanidine HCL, 50 mM Tris, pH 7.5, and 8 M guanidine was added after buffer exchange to a final concentration of 0.5 M. Trypsin digestion was carried out as described above and upon completion, 8 M guanidine was again added to each sample increasing the final concentration to 1.0 M. One milligram of digest per run was separated by RP-HPLC as described above except that the column temperature was lowered to 40°C. Fractions were analyzed by MALDI-TOF MS to assess purity prior to pooling.
O-Fucose Site Identification by ETD Ion-Trap MS
Fucosylated and unmodified tryptic peptides isolated from 8 mg of starting material were concentrated to dryness by vacuum centrifugation and reconstituted in 30% acetonitrile and 0.1% formic acid to a final approximate concentration of 20 pmol/µL. Approximately 5 µL of purified L3 or modified L3 peptide was analyzed with a LTQ XL mass spectrometer with ETD capability equipped with a TriVersa NanoMate (Advion, Ithaca, NY). The peptides were infused into the mass spectrometer by static nanospray and c- and z-type ions were generated in ETD mode by reacting the analyte with fluoranthene for 100 ms.
Monosaccharide Analysis of Released Fucose
Acid hydrolysis of O-fucose from modified and unmodified peptide occurred during incubation at 100°C for 2 h in the presence of 3 M TFA. Samples were evaporated to dryness by vacuum centrifugation and labeled with a 2AA labeling kit (Prozyme, San Leandro, CA) according to the manufacturers specifications. Unreacted chemicals were removed using GlykoClean S cartridges (Prozyme, San Leandro, CA) according to the manufacturers specifications, and the purified-labeled monosaccharides were evaporated to dryness by vacuum centrifugation and reconstituted in water prior to analysis. HPLC analysis was carried out on Waters 2695 HPLC equipped with a 2996 diode array detector and a 474 fluorescence detector (Waters, Milford, MA). The labeled monosaccharides were separated on a Thermo Aq column (2.1 x 100 mm, 1.9 micron particle, Thermo Fisher, Waltham, MA). Solvents A and B consisted of 0.25% acetic acid (A) and 10% acetonitrile, 0.25% acetic acid (B). Upon injection, a flow rate of 0.2 mL/min of 95% solvent A was maintained for 2 min and the labeled monosaccharides were subsequently eluted with a linear gradient to 30% B over 90 min. The eluted monosaccharides were detected with fluorescence excitation at 360 nm and emission detection at 425 nm.
|
Funding |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsFundingConflict of interest statementReferences |
|---|
|
Conflict of interest statement |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsFundingConflict of interest statementReferences |
|---|
None declared.
|
Abbreviations |
|---|
2AA, 2-aminoanthranilic acid; BDS, bulk drug substance; CDR-1, complementarity determining region-1; CEX, cation exchange; CHO, Chinese hamster ovary; CID, collision-induced dissociation; EGF, epidermal growth factor; ETD, electron transfer dissociation; Fd, fragment difficult; HIC, hydrophobic interaction chromatography; LC, light chain; MALDI-TOF MS, matrix-assisted laser desorption ionization time of flight mass spectrometry; MWCO, molecular weight cutoff; POFUT, protein O-fucosyltransferase 1; TCEP, Tris[2-carboxyethyl] phosphine; TFA, trifluoroacetic acid; TFMSA, trifluoromethanesulfonic acid; TSR, thrombospondin type 1 repeat
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsFundingConflict of interest statementReferences |
|---|
Alving K, Paulsen H, Peter-Katalinic J. Characterization of O-glycosylation sites in MUC2 glycopeptides by nanoelectrospray QTOF mass spectrometry. J Mass Spectrom. (1999) 34:395–407.[CrossRef][Medline]
Bause E, Hettkamp H. Primary structural requirements for N-glycosylation of peptides in rat liver. FEBS Lett. (1979) 108:341–344.[CrossRef][Web of Science][Medline]
Bjoern S, Foster DC, Thim L, Wiberg FC, Christensen M, Komiyama Y, Pedersen AH, Kisiel W. Human plasma and recombinant factor VII. Characterization of O-glycosylations at serine residues 52 and 60 and effects of site-directed mutagenesis of serine 52 to alanine. J Biol Chem. (1991) 266:11051–11057.
Brady LJ, Martinez T, Balland A. Characterization of nonenzymatic glycation on a monoclonal antibody. Anal Chem. (2007) 79:9403–9413.[Medline]
Brady LJ, Valliere-Douglass J, Martinez T, Balland A. Molecular mass analysis of antibodies by on-line SEC-MS. J Am Soc Mass Spectrom (2008) 19:502–509.[Medline]
Buko AM, Kentzer EJ, Petros A, Menon G, Zuiderweg ER, Sarin VK. Characterization of a posttranslational fucosylation in the growth factor domain of urinary plasminogen activator. Proc Natl Acad Sci USA (1991) 88:3992–3996.
Chalkley RJ, Burlingame AL. Identification of GlcNAcylation sites of peptides and alpha-crystalline using Q-TOF mass spectrometry. J Am Soc Mass Spectrom. (2001) 12:1106–1113.[CrossRef][Web of Science][Medline]
Chalkley RJ, Burlingame AL. Identification of novel sites of O-N-acetylglucosamine modification of serum response factor using quadrupole time-of-flight mass spectrometry. Mol Cell Proteomics. (2003) 2:182–190.
Gerken TA, Owens CL, Pasumarthy M. Determination of the site-specific O-glycosylation pattern of the porcine submaxillary mucin tandem repeat glycopeptide. Model proposed for the polypeptide:GalNAc transferase peptide binding site. J Biol Chem. (1997) 272:9709–9719.
Gonzalez de PA, Klein D, Macek B, Hess D, Peter-Katalinic J, Hofsteenge J. C-Mannosylation and O-fucosylation of thrombospondin type 1 repeats. Mol Cell Proteomics. (2002) 1:11–18.
Harris RJ, Leonard CK, Guzzetta AW, Spellman MW. Tissue plasminogen activator has an O-linked fucose attached to threonine-61 in the epidermal growth factor domain. Biochemistry (1991) 30:2311–2314.[CrossRef][Web of Science][Medline]
Harris RJ, Spellman MW. O-Linked fucose and other post-translational modifications unique to EGF modules. Glycobiology (1993) 3:219–224.
Harris RJ, van Halbeek H, Glushka J, Basa LJ, Ling VT, Smith KJ, Spellman MW. Identification and structural analysis of the tetrasaccharide NeuAc alpha(2
6)Gal beta(14)GlcNAc beta(13)Fuc alpha 1O-linked to serine 61 of human factor IX. Biochemistry (1993) 32:6539–6547.[CrossRef][Web of Science][Medline]
Hofsteenge J, Huwiler KG, Macek B, Hess D, Lawler J, Mosher DF, Peter-Katalinic J. C-Mannosylation and O-fucosylation of the thrombospondin type 1 module. J Biol Chem. (2001) 276:6485–6498.
Hofsteenge J, Muller DR, de BT, Loffler A, Richter WJ, Vliegenthart JF. New type of linkage between a carbohydrate and a protein: C-Glycosylation of a specific tryptophan residue in human RNase Us. Biochemistry (1994) 33:13524–13530.[CrossRef][Web of Science][Medline]
Kentzer EJ, Buko A, Menon G, Sarin VK. Carbohydrate composition and presence of a fucose-protein linkage in recombinant human pro-urokinase. Biochem Biophys Res Commun. (1990) 171:401–406.[CrossRef][Web of Science][Medline]
Kozma K, Keusch JJ, Hegemann B, Luther KB, Klein D, Hess D, Haltiwanger RS, Hofsteenge J. Identification and characterization of a beta1,3-glucosyltransferase that synthesizes the Glc-beta1,3-Fuc disaccharide on thrombospondin type 1 repeats. J Biol Chem. (2006) 281:36742–36751.
Luo Y, Koles K, Vorndam W, Haltiwanger RS, Panin VM. Protein O-fucosyltransferase 2 adds O-fucose to thrombospondin type 1 repeats. J Biol Chem. (2006) 281:9393–9399.
Luo Y, Nita-Lazar A, Haltiwanger RS. Two distinct pathways for O-fucosylation of epidermal growth factor-like or thrombospondin type 1 repeats. J Biol Chem. (2006) 281:9385–9392.
Martinez T, Pace D, Brady L, Gerhart M, Balland A. Characterization of a novel modification on IgG2 light chain. Evidence for the presence of O-linked mannosylation. J Chromatogr A (2007) 1156:183–187.
Mikesh LM, Ueberheide B, Chi A, Coon JJ, Syka JE, Shabanowitz J, Hunt DF. The utility of ETD mass spectrometry in proteomic analysis. Biochim Biophys Acta (2006) 1764:1811–1822.[Medline]
Moloney DJ, Haltiwanger RS. The O-linked fucose glycosylation pathway: Identification and characterization of a uridine diphosphoglucose: fucose-beta1,3-glucosyltransferase activity from Chinese hamster ovary cells. Glycobiology (1999) 9:679–687.
Moloney DJ, Lin AI, Haltiwanger RS. The O-linked fucose glycosylation pathway. Evidence for protein-specific elongation of O-linked fucose in Chinese hamster ovary cells. J Biol Chem (1997) 272:19046–19050.
Moloney DJ, Panin VM, Johnston SH, Chen J, Shao L, Wilson R, Wang Y, Stanley P, Irvine KD, Haltiwanger RS, et al. Fringe is a glycosyltransferase that modifies Notch. Nature (2000) 406:369–375.[CrossRef][Medline]
Nakakura N, Hietter H, Van DA, Luu B. Isolation and structural determination of three peptides from the insect Locusta migratoria. Identification of a deoxyhexose-linked peptide. Eur J Biochem. (1992) 204:147–153.[Web of Science][Medline]
Neglia CI, Cohen HJ, Garber AR, Ellis PD, Thorpe SR, Baynes JW. 13C NMR investigation of nonenzymatic glucosylation of protein. Model studies using RNase A. J Biol Chem. (1983) 258:14279–14283.
Nishimura H, Takao T, Hase S, Shimonishi Y, Iwanaga S. Human factor IX has a tetrasaccharide O-glycosidically linked to serine 61 through the fucose residue. J Biol Chem. (1992) 267:17520–17525.
Rampal R, Li AS, Moloney DJ, Georgiou SA, Luther KB, Nita-Lazar A, Haltiwanger RS. Lunatic fringe, manic fringe, and radical fringe recognize similar specificity determinants in O-fucosylated epidermal growth factor-like repeats. J Biol Chem. (2005) 280:42454–42463.
Saphire EO, Parren PW, Pantophlet R, Zwick MB, Morris GM, Rudd PM, Dwek RA, Stanfield RL, Burton DR, Wilson IA. Crystal structure of a neutralizing human IGG against HIV-1: A template for vaccine design. Science (2001) 293:1155–1159.
Spiegelberg HL, Abel CA, Fishkin BG, Grey HM. Localization of the carbohydrate within the variable region of light and heavy chains of human gamma g myeloma proteins. Biochemistry (1970) 9:4217–4223.
Syka JE, Coon JJ, Schroeder MJ, Shabanowitz J, Hunt DF. Peptide and protein sequence analysis by electron transfer dissociation mass spectrometry. Proc Natl Acad Sci USA (2004) 101:9528–9533.
Tetaert D, Ten Hagen KG, Richet C, Boersma A, Gagnon J, Degand P. Glycopeptide N-acetylgalactosaminyltransferase specificities for O-glycosylated sites on MUC5AC mucin motif peptides. Biochem J. (2001) 357:313–320.[CrossRef][Web of Science][Medline]
Tomlinson IM, Walter G, Marks JD, Llewelyn MB, Winter G. The repertoire of human germline VH sequences reveals about fifty groups of VH segments with different hypervariable loops. J Mol Biol. (1992) 227:776–798.[CrossRef][Web of Science][Medline]
Vosseller K, Trinidad JC, Chalkley RJ, Specht CG, Thalhammer A, Lynn AJ, Snedecor JO, Guan S, Medzihradszky KF, Maltby DA, et al. O-Linked N-acetylglucosamine proteomics of postsynaptic density preparations using lectin weak affinity chromatography and mass spectrometry. Mol Cell Proteomics. (2006) 5:923–934.
Wang LW, Dlugosz M, Somerville RP, Raed M, Haltiwanger RS, Apte SS. O-Fucosylation of thrombospondin type 1 repeats in ADAMTS-like-1/punctin-1 regulates secretion: Implications for the ADAMTS superfamily. J Biol Chem. (2007) 282:17024–17031.
Wang Y, Lee GF, Kelley RF, Spellman MW. Identification of a GDP-l-fucose:polypeptide fucosyltransferase and enzymatic addition of O-linked fucose to EGF domains. Glycobiology (1996) 6:837–842.
Wang Y, Shao L, Shi S, Harris RJ, Spellman MW, Stanley P, Haltiwanger RS. Modification of epidermal growth factor-like repeats with O-fucose. Molecular cloning and expression of a novel GDP-fucose protein O-fucosyltransferase. J Biol Chem. (2001) 276:40338–40345.
Wang Y, Spellman MW. Purification and characterization of a GDP-fucose:polypeptide fucosyltransferase from Chinese hamster ovary cells. J Biol Chem. (1998) 273:8112–8118.
Weith HL, Wiebers JL, Gilham PT. Synthesis of cellulose derivatives containing the dihydroxyboryl group and a study of their capacity to form specific complexes with sugars and nucleic acid components. Biochemistry (1970) 9:4396–4401.
Wells L, Vosseller K, Cole RN, Cronshaw JM, Matunis MJ, Hart GW. Mapping sites of O-GlcNAc modification using affinity tags for serine and threonine post-translational modifications. Mol Cell Proteomics. (2002) 1:791–804.
Yan B, Valliere-Douglass J, Brady L, Steen S, Han M, Pace D, Elliott S, Yates Z, Han Y, Balland A, et al. Analysis of post-translational modifications in recombinant monoclonal antibody IgG1 by reversed-phase liquid chromatography/mass spectrometry. J Chromatogr A (2007) 1164:153–161.[CrossRef][Medline]
Yoshida A, Suzuki M, Ikenaga H, Takeuchi M. Discovery of the shortest sequence motif for high level mucin-type O-glycosylation. J Biol Chem. (1997) 272:16884–16888.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. F. Valliere-Douglass, P. Kodama, M. Mujacic, L. J. Brady, W. Wang, A. Wallace, B. Yan, P. Reddy, M. J. Treuheit, and A. Balland Asparagine-linked Oligosaccharides Present on a Non-consensus Amino Acid Sequence in the CH1 Domain of Human Antibodies J. Biol. Chem., November 20, 2009; 284(47): 32493 - 32506. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||