A glycomics platform for the analysis of permethylated oligosaccharide alditols.
Journal: 2007/November - Journal of the American Society for Mass Spectrometry
ISSN: 1044-0305
Abstract:
This communication reports the development of an LC/MS platform for the analysis of permethylated oligosaccharide alditols that, for the first time, demonstrates routine online oligosaccharide isomer separation of these compounds before introduction into the mass spectrometer. The method leverages a high-resolution liquid chromatography system with the superior fragmentation pattern characteristics of permethylated oligosaccharide alditols that are dissociated under low-energy collision conditions using quadrupole orthogonal time-of-flight (QoTOF) instrumentation and up to pseudo MS(3) mass spectrometry. Glycoforms, including isomers, are readily identified and their structures assigned. The isomer-specific spectra include highly informative cross-ring and elimination fragments, branch position specific signatures, and glycosidic bond fragments, thus facilitating linkage, branch, and sequence assignment. The method is sensitive and can be applied using as little as 40 fmol of derivatized oligosaccharide. Because permethylation renders oligosaccharides nearly chemically equivalent in the mass spectrometer, the method is semiquantitative and, in this regard, is comparable to methods reported using high field NMR and capillary electrophoresis. In this postgenomic age, the importance of glycosylation in biological processes has become clear. The nature of many of the important questions in glycomics is such that sample material is often extremely limited, thus necessitating the development of highly sensitive methods for rigorous structural assignment of the oligosaccharides in complex mixtures. The glycomics platform presented here fulfills these criteria and should lead to more facile glycomics analyses.
Relations:
Content
Citations
(53)
References
(51)
Grants
(1K+)
Chemicals
(2)
Processes
(2)
Affiliates
(2)
Similar articles
Articles by the same authors
Discussion board
J Am Soc Mass Spectrom 18(10): 1799-1812

A Glycomics Platform for the Analysis of Permethylated Oligosaccharide Alditols

Introduction

It is estimated that up to 50% of all proteins are glycosylated 12. Protein glycosylation is involved in many, if not most, biological processes and there is extensive literature that documents its involvement in diverse areas such as cell - cell recognition, cell - extracellular matrix interactions, tissue development, immune response, host - pathogen recognition, protein folding editing control and protein stability 36. Glycans are not directly encoded in the genome. Therefore, their structures cannot be predicted strictly by a reading or manipulation of the genetic code as is the case with nucleic acids and proteins. Although it seems unlikely that the full range of potential structures is realized in nature, an oligosaccharide of only six monosaccharide components could theoretically give rise to greater than 1 × 10 chemically distinct structures 78. This is many orders of magnitude higher than that of a hexapeptide because, as cyclic polyols, the components of oligosaccharides have more linkage points, the resulting structures can be branched, and linkages can be formed in either an α or β anomeric configuration. Therefore, structural identification of oligosaccharides presents a considerable technical challenge.

The development of analytical methods for routine analysis of oligosaccharides has lagged behind those of proteins and nucleic acids. Often the relevant biological questions dictate that the amount of glycan material available for analysis is extremely limited. Furthermore, the glycans are almost always present as a mixture of closely related structures, only a few of which may have biological activity. Aberrant glycosylation, e.g., in a disease state, can be dramatic, such as loss of nearly an entire class or subclass of glycans, or more subtle, such as loss of only a few isomers within a subclass 3911. Naturally, the ability to detect the dramatic as well as the subtle changes in extremely limited sample amounts will be required as questions of glycomic consequence are considered. Ultimately, this means that complete structural assignment of all the glycans in complex mixtures must be accomplished over a wide dynamic range, often with limited sample amounts. With the growing interest in glycomics, it is clear that more sensitive and informative platforms must be developed for oligosaccharide and glycoconjugate structural investigation.

Liquid chromatography mass spectrometry (LC/MS) has been used in various configurations for oligosaccharide analysis. The LC/MS method can be considered to have three components. While the more obvious components are the LC and MS instruments, the third variable is the analyte itself, in that the choice of the particular derivative to be used can have a dramatic effect on spectral properties.

Permethylation of oligosaccharides increases their ionization efficiency up to 20 fold 12. The tandem mass spectra of permethylated oligosaccharides contain a higher abundance of cross-ring fragments than their native counterparts, thus allowing a higher probability of linkage assignment 1316. Elimination of the C3 substituent from HexNAc residues in permethylated glycans is useful for linkage assignment 1719. Also, a chemical signature for branch substitution is inherent in permethylated oligosaccharide spectra, since fragments that expose formerly internal residues in newly formed terminal positions are 14 Da lower in mass at each former branch site due to the presence of an hydroxyl group at the glycosidic cleavage site, whereas all the hydroxyls of native terminal units are methylated. Permethylation also imparts additional hydrophobic character to oligosaccharides, thus making them more amenable to the LC methods most suitable for LC/MS applications, which require organic solvents and low buffer concentrations. Permethylated derivatives of oligosaccharides have been extensively studied by matrix-assisted laser desorption/ionization (MALDI) and electrospray ionization (ESI) tandem MS and their fragmentation properties are well understood. Surprisingly little effort has been made to develop LC/MS methods for these compounds. This, in part, may be due to their lack of a chromophore and the consequent barrier to off-line methods development. LC/MS methods have been reported for analysis of permethylated derivatives using reversed phase chromatography 2021, but improvements on these methods are desired, in order to provide isomer separation.

Many liquid chromatography formats have been used for separation of oligosaccharides. With some examples cited these including normal phase (NPC) 22, reversed phase (RPC) 2324, porous graphitized carbon (PGC) 25 hydrophilic interaction (HILIC) 26, ion exchange (IEC) 27 and high pH anion exchange (HPAEC) 27 chromatographies. PGC, HILIC and HPAEC provide superior resolution of oligosaccharides and allow separation of some isomers. Each has been employed in LC/MS applications for separation of native glycans and/or glycans derivatized at the reducing end 2731. PGC is an attractive candidate for application to characterization of permethylated glycans as it has been reported to provide excellent separation of analytes ranging from highly polar to hydrophobic.

Naturally, it would be advantageous to leverage the superior information content of permethylated oligosaccharide fragmentation spectra with an LC/MS platform capable of isomer differentiation. Although LC/MS analysis of permethylated glycans has been reported 202132, until this report, LC/MS isomer separation of these compounds has not been demonstrated. The capacity to separate permethylated oligosaccharide isomers by LC prior to infusion into the mass spectrometer can make their analysis more facile. Here we report the first separation and identification of permethylated oligosaccharide isomers using an LC/MS platform. This has been accomplished using a PGC LC system, in combination with the further separation capacity of the mass spectrometer. This platform should significantly enhance the capacity for the bioanalyst to identify oligosaccharides in complex mixtures.

Experimental

Materials

Ribonuclease B, fetal calf serum fetuin, bovine articular cartilage decorin and corn syrup malto-oligosaccharides were from Sigma Aldrich. All reagents were ACS grade or higher. Recombinant PNGase F was expressed and purified in-house.

Oligosaccharide Release and Reduction

N-glycans were released from protein by treatment with PNGase F at 4000 U/ml at 37 °C for 16 h. The digest solution contained 0.1% sodium dodecyl sulfate and 0.75% NP-40 in 50 mM ammonium bicarbonate (NH4HCO3) at pH 8.0. After release, the oligosaccharides were purified by solvent extraction as previously described 33. Briefly, detergent and salts were removed by solvent precipitation in 80% acetone at pH 5.5. The released oligosaccharides were purified by solvent extraction in 50% methanol acidified to pH 5.5 with drop-wise addition of acetic acid. Dried glycan samples (1-100 μg) were placed in 200 μl of 0.1 M NaOH/ 1M NaBH4 for one hour at ambient temperature. Borate was removed by rotary evaporation from acidified methanol (1% acetic acid).

Bovine articular cartilage was processed as follows. To 50 ug of decorin proteoglycan was mixed 10 uL water, 2 uL 1M tris-HCl buffer (pH 7.4), 1 uL 1M NH4OAc and 20 mU of chondroitinase ABC and digestion was allowed to proceed at 37°C. Completion of reaction was monitored by size exclusion chromatography until a single peak of the core protein was observed. This required a total of 60 mU of chondroitinase ABC, where 20 mU of the chondroitinase was added every 8 hours. The mixture was allowed to digest for a total of 24 hours after which the reaction was stopped by boiling the mixture for 2 minutes. Decorin core protein was purified using a Superdex Peptide 3.2/30 column (Amersham Biosciences), which was equilibrated in 10% ACN, 0.1M ammonium acetate at 100 uL/min and with detection at 232 nm. N-glycans were release from 25 μg of dried decorin core protein as described above.

C. elegans O-glycans were released from whole nematode extracts as previously described 34. Briefly, glycans were released by reductive β-elimination in 0.1M NaOH/ 1M NaBH4 at 60 °C for 16 h followed by neutralization with drop-wise addition of glacial acetic acid. The mixture was passed over a 10 ml column of Dowex 50W cation exchange resin (hydrogen form) and oligosaccharide material was eluted with water. Borate was removed as tetramethylborate by rotary evaporation from acidified methanol (1% glacial acetic acid). After reconstitution, the samples were applied to a 10 ml column of AG1x-2 acetate anion exchange resin. Neutral glycans were eluted with water and acidic ones eluted with 100 mM NH4OAc. The pools were dried by rotary evaporation.

Released N- and O-glycans were further purified using porous graphitized carbon (PGC) cartridges (Thermo Electron). The cartridges were pre-wet with 100% acetonitrile and rinsed sequentially with two ml each of 60%, 30% and 0% acetonitrile, all of which contained 0.1% trifluoroacetic acid. The glycans were applied to PGC in water, washed extensively with 0.1% trifluoroacetic acid and eluted by application of 1.5 ml of 30% acetonitrile in 0.1% trifluoroacetic acid. The samples were then dried in a Savant™ speed evacuation device (Thermo).

Permethylation of oligosaccharides

Oligosaccharides were permethylated using the method of Ciucanu and Kerek 35 with minor modifications as described previously 36. Sample recovery was 70-95% based on phenol sulfuric acid assay for neutral hexose and/or peak intensity on MALDI-TOF MS analysis using permethylated oligosaccharide standards. Sample purity and suitability for LC/MS analysis was verified by MALDI-TOF MS on a Reflex IV instrument (Bruker, Billerica, MA), using 2,5-dihydroxybenzoic acid as matrix and a nitrogen laser (337 nm, 3 nsec pulse width).

Liquid Chromatography / Mass Spectrometry

Porous graphite liquid chromatographic separation was achieved with either a 2 × 50 mm or 0.320 × 100 mm Thermo Hypersil-Keystone porous graphite chromatography (PGC) column (Thermo) and a Beckman Coulter System Gold 125 solvent module. Solvent A was 0.05% formic acid. Solvent B was 1:1 acetonitrile: isopropanol (v: v) in 0.05% formic acid. The gradient was 30% Solvent B for 5 min, 30 to 65% Solvent B over 5 min, 65 to 85% solvent B over 50 min, 85 to 100% solvent B over 5 min, 100% solvent B for 10 min, return to 30% solvent B in 0.1 min followed by equilibration to original conditions for 5 min. The entire gradient program was 80.1 min. This gradient was used for all experiments presented. The column eluate was delivered through a post-column split to a QStar Pulsar i quadrupole orthogonal time-of-flight mass spectrometer (Applied Biosystems/Sciex) equipped with a Turbospray ion source. The post-column split flow rate to the spectrometer source was ten and five μl/min for the 2-mm and 320-μm diameter columns respectively. Post-column infusion of formic acid was used in a final infused concentration of 1%. 10% formic acid was prepared in methanol and infused at 1:10 (V:V). Data dependent acquisition experiments were performed using Analyst software (Applied Biosystems). Include and exclude lists were used to target select ions for CID experiments. Collision voltages were applied automatically according to a linear calibration curve determined for 1+, 2+ and 3+ charge states using permethylated oligosaccharide standards. The mass spectrometer parameters were as follows for MS experiments: DP1 75.0 V, FP 245.0 V, DP2 30.0 V, CG 3.0 psi, IRD 6.0 V, IRW 5.0 V, GS1 5.0 psi, GS2 10.0 psi, CUR 12.0 psi and the ion spray voltage was between 4000 and 4500 V. Collision gas was Ar. For pseudo MS experiments the following parameters were used: DP1 130.0 V, FP 300.0 V, DP2 40.0 V, CG 3.0 psi, IRD 6.0 V, IRW 5.0 V, GS1 5.0 psi, GS2 10.0 psi, CUR 12.0 V.

Materials

Ribonuclease B, fetal calf serum fetuin, bovine articular cartilage decorin and corn syrup malto-oligosaccharides were from Sigma Aldrich. All reagents were ACS grade or higher. Recombinant PNGase F was expressed and purified in-house.

Oligosaccharide Release and Reduction

N-glycans were released from protein by treatment with PNGase F at 4000 U/ml at 37 °C for 16 h. The digest solution contained 0.1% sodium dodecyl sulfate and 0.75% NP-40 in 50 mM ammonium bicarbonate (NH4HCO3) at pH 8.0. After release, the oligosaccharides were purified by solvent extraction as previously described 33. Briefly, detergent and salts were removed by solvent precipitation in 80% acetone at pH 5.5. The released oligosaccharides were purified by solvent extraction in 50% methanol acidified to pH 5.5 with drop-wise addition of acetic acid. Dried glycan samples (1-100 μg) were placed in 200 μl of 0.1 M NaOH/ 1M NaBH4 for one hour at ambient temperature. Borate was removed by rotary evaporation from acidified methanol (1% acetic acid).

Bovine articular cartilage was processed as follows. To 50 ug of decorin proteoglycan was mixed 10 uL water, 2 uL 1M tris-HCl buffer (pH 7.4), 1 uL 1M NH4OAc and 20 mU of chondroitinase ABC and digestion was allowed to proceed at 37°C. Completion of reaction was monitored by size exclusion chromatography until a single peak of the core protein was observed. This required a total of 60 mU of chondroitinase ABC, where 20 mU of the chondroitinase was added every 8 hours. The mixture was allowed to digest for a total of 24 hours after which the reaction was stopped by boiling the mixture for 2 minutes. Decorin core protein was purified using a Superdex Peptide 3.2/30 column (Amersham Biosciences), which was equilibrated in 10% ACN, 0.1M ammonium acetate at 100 uL/min and with detection at 232 nm. N-glycans were release from 25 μg of dried decorin core protein as described above.

C. elegans O-glycans were released from whole nematode extracts as previously described 34. Briefly, glycans were released by reductive β-elimination in 0.1M NaOH/ 1M NaBH4 at 60 °C for 16 h followed by neutralization with drop-wise addition of glacial acetic acid. The mixture was passed over a 10 ml column of Dowex 50W cation exchange resin (hydrogen form) and oligosaccharide material was eluted with water. Borate was removed as tetramethylborate by rotary evaporation from acidified methanol (1% glacial acetic acid). After reconstitution, the samples were applied to a 10 ml column of AG1x-2 acetate anion exchange resin. Neutral glycans were eluted with water and acidic ones eluted with 100 mM NH4OAc. The pools were dried by rotary evaporation.

Released N- and O-glycans were further purified using porous graphitized carbon (PGC) cartridges (Thermo Electron). The cartridges were pre-wet with 100% acetonitrile and rinsed sequentially with two ml each of 60%, 30% and 0% acetonitrile, all of which contained 0.1% trifluoroacetic acid. The glycans were applied to PGC in water, washed extensively with 0.1% trifluoroacetic acid and eluted by application of 1.5 ml of 30% acetonitrile in 0.1% trifluoroacetic acid. The samples were then dried in a Savant™ speed evacuation device (Thermo).

Permethylation of oligosaccharides

Oligosaccharides were permethylated using the method of Ciucanu and Kerek 35 with minor modifications as described previously 36. Sample recovery was 70-95% based on phenol sulfuric acid assay for neutral hexose and/or peak intensity on MALDI-TOF MS analysis using permethylated oligosaccharide standards. Sample purity and suitability for LC/MS analysis was verified by MALDI-TOF MS on a Reflex IV instrument (Bruker, Billerica, MA), using 2,5-dihydroxybenzoic acid as matrix and a nitrogen laser (337 nm, 3 nsec pulse width).

Liquid Chromatography / Mass Spectrometry

Porous graphite liquid chromatographic separation was achieved with either a 2 × 50 mm or 0.320 × 100 mm Thermo Hypersil-Keystone porous graphite chromatography (PGC) column (Thermo) and a Beckman Coulter System Gold 125 solvent module. Solvent A was 0.05% formic acid. Solvent B was 1:1 acetonitrile: isopropanol (v: v) in 0.05% formic acid. The gradient was 30% Solvent B for 5 min, 30 to 65% Solvent B over 5 min, 65 to 85% solvent B over 50 min, 85 to 100% solvent B over 5 min, 100% solvent B for 10 min, return to 30% solvent B in 0.1 min followed by equilibration to original conditions for 5 min. The entire gradient program was 80.1 min. This gradient was used for all experiments presented. The column eluate was delivered through a post-column split to a QStar Pulsar i quadrupole orthogonal time-of-flight mass spectrometer (Applied Biosystems/Sciex) equipped with a Turbospray ion source. The post-column split flow rate to the spectrometer source was ten and five μl/min for the 2-mm and 320-μm diameter columns respectively. Post-column infusion of formic acid was used in a final infused concentration of 1%. 10% formic acid was prepared in methanol and infused at 1:10 (V:V). Data dependent acquisition experiments were performed using Analyst software (Applied Biosystems). Include and exclude lists were used to target select ions for CID experiments. Collision voltages were applied automatically according to a linear calibration curve determined for 1+, 2+ and 3+ charge states using permethylated oligosaccharide standards. The mass spectrometer parameters were as follows for MS experiments: DP1 75.0 V, FP 245.0 V, DP2 30.0 V, CG 3.0 psi, IRD 6.0 V, IRW 5.0 V, GS1 5.0 psi, GS2 10.0 psi, CUR 12.0 psi and the ion spray voltage was between 4000 and 4500 V. Collision gas was Ar. For pseudo MS experiments the following parameters were used: DP1 130.0 V, FP 300.0 V, DP2 40.0 V, CG 3.0 psi, IRD 6.0 V, IRW 5.0 V, GS1 5.0 psi, GS2 10.0 psi, CUR 12.0 V.

Results and Discussion

Rationale and Key Experimental Observations

Reduced and permethylated malto-oligosaccharides, RNase B released N-glycans, articular cartilage released N-glycans and C. elegans whole nematode extract released O-glycans were analyzed. This selection provided linear, high mannose, complex, hybrid, and short, highly branched oligosaccharides for study. Reduction was performed to avoid separation of the α and β anomers of the permethylated oligosaccharides on the PGC columns, as well as to enable distinction of potentially isobaric fragment ions. The LC system used here was optimized for the oligosaccharides presented. However, an increased amount of isopropanol can be used in solvent B to provide better chromatographic properties for more highly retained permethylated oligosaccharide alditols. The branch isomers containing Fuc had clearly unique retention times, whereas positional isomers that varied in the location of Man had retention times that were more similar to one another. The peak shapes of sialylated glycans tended to be more broad than were those of other compounds having similar masses (data not shown). For glycans such as those released from α-one acid glycoprotein where a high proportion of higher molecular weight glycomers occurs solvent B can be replaced with 2:1 isopropanol: acetonitrile. Essentially the same gradient can be used. Typically, these conditions allow resolution of more glycomers than that obtained with 1:1 isopropanol: acetonitrile where these tend to elute as an unresolved group during the 100% solvent B ramp (data not shown).

In this study some key fragment ions were observed and these reiterated trends that have been observed previously in fragmentation patterns of permethylated glycans. The branching pattern of the oligosaccharides could be deduced from CID spectra containing fragments bearing a mass signature of -14 Da for each lost branch. In samples containing Hex and HexNAc residues, abundant A, A, A, and A cross-ring fragments were often observed, thus allowing the assignment of 1-2, 1-4, and 1-6-linkages 3738. Elimination at the C3 of HexNAc was also observed, and provided a chemical signature that could be used to assign C3 substitutions as has been previously reported 1718. In high mannose and hybrid glycans, secondary fragmentation gave rise to product ions that clearly defined the high mannose portion. In highly branched O-glycans, fragmentation across the C2-C3, C3-C4 and C4-C5 bonds was key to structural assignments. Results were highly reproducible in terms of extracted ion chromatogram (XIC) intensity, CID fragment ion intensity and sensitivity. The nomenclature used is that of Costello and Domon 39. Monosaccharide symbols are those suggested by Varki et al4041 with some modification allowing for Hex.

Porous Graphite LC/MS of Malto-oligosaccharides

As shown in Figure 1A, Glc4-8 were well separated when malto-oligosaccharide aliquots of approximately 300 ng of reduced and permethylated malto-oligosaccharides were analyzed. The CID spectrum of Glc5 [M+Na], m/z 1105.64, is shown in Figure 1C. For the malto-oligosaccharides, permethylation without prior reduction resulted in isobaric Y and C as well as Z and B ions. This situation was avoided by carrying out reduction prior to permethylation. As seen in the Figure 1C inset, Y and C and Z and B ions are all unique. An (n = 2, 3, 4) cross-ring fragments were observed at m/z 329.12, 533.27 and 737.42, respectively, thus confirming each β1,4 glycosidic bond in the series. Also observed were secondary (internal) fragments such as the Y4/C3 and Y3/C4 fragment at m/z 449.22. Fragments that correspond to methanol loss from the Cn ion, e.g. m/z 431.22 could also be assigned as Y4/B3 or Y3/B4. The presence of the abundant peak at m/z 839.51 indicated that the methanol loss is important for C4 but the facile Bn and Yn pathways suggest that, at lower m/z values, the Yx/By isomer may be dominant. Therefore, for simplicity, in the following sections these type fragments will be referred to as the Yx/By type. These data demonstrate the capacity of PGC to separate permethylated malto-oligosaccharide alditols and also show the capacity of the method to provide informative fragmentation patterns where linkage, branch number and sequence can be completely assigned.

An external file that holds a picture, illustration, etc.
Object name is nihms-31741-f0001.jpg

LC/MS analysis of permethylated maltooligosaccharide alditols. A, Extracted ion chromatograms of permethylated oligosaccharide alditol maltooligosaccharide series Glc4-8. The Glc4, Glc5 Glc6, Glc7 [M+Na] ions, Glc8 [M+2Na] were detected at m/z 901.45, 1105.56, 1309.66, 1513.76 and 870.43 respectively. Integrated peak areas are proportional to abundances of the glycoforms. B and C, fragment origin and CID QoTOF MS/MS spectra of Glc5, [M+Na]m/z 1105.56. Fragmentation nomenclature is that of Domon and Costello 39. All fragments contain sodium. Monosaccharide component symbols are those suggested by Varki et al40. Key ions are indicated. The inset in C demonstrates that distinct B, Z, C and Z ions are produced from the reduced and permethylated Hexose oligosaccharide.

Porous Graphite LC/MS of High Mannose N-glycans from RNase B

Permethylated high mannose N-glycans have been studied by tandem mass spectrometry and their fragmentation patterns are well established 4243. RNase B contains exclusively high mannose oligosaccharides. While some variation can be expected from differences in the protein source, method of release and purification, the approximate abundance of each glycan size is known, as is the relative distribution of the isomers within each size pool 4445. Man5GlcNAc2 and Man6GlcNAc2 are reported to be ∼57% and ∼31% respectively of the total glycans of RNase B and proton NMR has established that each is present as a single glycoform. There are three Man7GlcNAc2 isomers, which combined are ∼ 4% of total glycans. The isomers are referred to as D1, D2 and D3 and differ only in the position of one α1,2Man residue (see Figure 3). D1 refers to substitution of the α1,2Man lower arm, D2 the central α1,3Man arm, and D3 the upper α1,6Man arm. The D1, D2 and D3 isomers are present in approximately equimolar proportions. Man8GlcNAc2 is reported to be ∼7% of total glycans and is also present as three isomers. The D1D3 isomer is most abundant and is approximately 80 - 85% of Man8GlcNAc2. The D1D2 isomer is ∼10 - 15% and the D2D3 isomer is ∼5% of the total Man8GlcNAc2. Man9GlcNAc2 is ∼ 1% of glycans and is present as a single isoform. High mannose glycan was released from approximately 2 milligrams of lyophilized ribonuclease B powder. Approximately 120 micrograms of mannan was obtained giving a yield of approximately 80% based on 100% glycosylation site occupancy and an average oligosaccharide molecular weight of 1320 using a weighted average for glycans present.

An external file that holds a picture, illustration, etc.
Object name is nihms-31741-f0003.jpg

CID QoTOF MS/MS spectra of Man7GlcNAc2 D1, D2, and D3 isomers. Major product ions, derived from [M+2Na]m/z 1013.49 are indicated in each spectrum and their origins depicted to the structural representations to the right of each spectrum. Spectra and structural diagrams are shown in the order of their elution: A, Man7GlcNAc2 D2; B, Man7GlcNAc2 D3, and C, Man7GlcNAc2 D1. All fragments contain sodium.

To the LC/MS system equipped with a 2 × 50 mm PGC column, approximately 350-ng aliquots of permethylated RNase N-glycan alditols were applied, providing an estimated dynamic range in chemically distinct glycoforms of ∼ 200 to 5.25 ng. The extracted ion chromatogram (XIC) of the RNase B Man5-9GlcNAc2 is shown in Figure 2A. Permethylation renders oligosaccharides within the same compound class nearly chemically equivalent. As a result, mass spectral responses are similar over a fairly broad mass range 46. Integration of all ions from three independent experiments provided a distrubution of Man5-9GlcNAc2 that was comparable to previous estimates determined by NMR and CE analyses 4445 as shown in Table I. The chromatographic positions of each isomer in Man7GlcNAc2 and Man8GlcNAc2 are shown in Figure 2B.

An external file that holds a picture, illustration, etc.
Object name is nihms-31741-f0002.jpg

Extracted ion chromatogram of permethylated oligosaccharide alditols Man5-9GlcNAc2: A, a stacked plot of molecular ions for Man5-9GlcNAc2is shown; B, Three partially overlapping peaks are observed for Man7GlcNAc2 detected at [M+2Na]m/z 1013.49 (bottom) and are labeled as D1, D2, and D3 and three are observed for Man8GlcNAc2 detected at [M+2Na]m/z 1115.55 and are labeled as D1D2, D2D3 and D1D3 (top); C, Selected ion chromatogram of Man9GlcNAc2 produced from approximately 160 fmol. The molecular ion region of the mass spectrum is shown in the inset. See text for details concerning nomenclature.

Table I

Comparison of semiquantitative methods for the analysis of RNase B oligosaccharides

Glycan% abundance LC/MS% abundance NMRa% Isomer abundance CEb
overallisomeroverallisomer
Man5GlcNAc255.1 ± 6.2-57--
Man6GlcNAc230.8 ± 5.2-31--
Man7GlcNAc2 D13.5 ± 1.3441.53138
Man7GlcNAc2 D22.7 ± 1.3341.53337
Man7GlcNAc2 D31.8 ± 0.7221.03525
Man8GlcNAc2 D1D20.2 ± 0.130.41017
Man8GlcNAc2 D2D30.3 ± 0.150.764
Man8GlcNAc2 D1D34.8 ± 0.6925.98478
Man9GlcNAc21.6 ± 0.4-1.0--

Error is expressed as standard error of the mean calculated from three independent analyses.

Values are from reference 49
Values are from reference 50

The CID spectra of the three Man7GlcNAc2 isomers produced during a representative experiment are shown in Figure 3. The CID spectrum of Man7GlcNAc2, [M+2Na]m/z 1013.49, in the time domain 17.05 to 17.26 min is shown in Figure 3A. The derived structure and all key fragments are shown to the right of the spectrum. The ions seen at m/z 1288.60 and 1492.72 are consistent with fragments Y/B5 for the former and Y4x/B5 and Y5α’/B5 for the latter and define an upper arm that contained four Hex residues. Cross ring fragments A4 and A4 at m/z 913.47 and 941.49 define the 1,6-linked arm. Both D2 and D3 isomers can contain these fragments. Prediction of the fragmentation of the D2 structure would forecast the presence of A and A fragments at m/z 301.1 and 329.1, whereas fragmentation of the D3 structure should yield A and A fragments at m/z 505.2 and 533.2. Indeed, fragments are observed at m/z 301.15 and 329.17 and not at m/z 505.2 or 533.2, demonstrating that, in this time domain, the D2 isomer is the major isomer present. Thus the partially resolved peak centered at 17.1 minutes is isomer D2.

The CID spectrum of the time domain 17.93 to 18.13 min is shown in Figure 3B. As in the D2 spectra, ions were observed at m/z 1288.60 for Y/B5, and m/z 1492.72 for Y4x/B5 and Y/B5. As described above, these fragments are consistent with the presence of a branch that contained four Hex units. Cross-ring fragments A4 and A4 at m/z 913.47 and 941.49, were also observed. These define the Man4 branch as linked to the 1,6 arm. However, A and A fragments were detected at m/z 505.24 and 533.27 and not at m/z 301.1 and 329.1. Therefore, the predominant glycan in this time domain was the D3 isomer. The derived structure and the origins of all key fragment ions are shown to the right of the spectrum.

The CID spectrum of the time domain 19.02 to 19.23 min is shown in Figure 3C. An ion is observed at m/z 1084.54, which is consistent with a Y3x/B5 secondary fragment. This ion can not be generated by either isomer D2 or D3. The D1 α1,6 arm can be defined by cross-ring fragments A3a, A3a, A4 and A4, all of which are present in the spectrum at m/z 301.18, 329.17, 709.34 and 737.35 respectively. Thus, the predominant isomer within this time domain is the D1 isomer.

An XIC of the Man9GlcNAc2 [M+2Na] at m/z 1217.63 is shown in Figure 2C from an analysis of a 20 ng aliquot of total mannan using a 0.32 × 100 mm PGC column. Nearly identical tandem mass spectra were obtained across the entire peak region supporting the presence of only one isomer as expected (not shown). An estimated 160 fmol of Man9GlcNAc2 was present in the mixture and its mass spectrum is shown in the inset in Figure 2C. The signal-to-noise ratio was over 8000:1. When the amount of mannan was reduced four-fold, a signal-to-noise ratio of over 500:1 was obtained from approximately 40 fmol of Man9GlcNAc2 (data not shown).

Rationale and Key Experimental Observations

Reduced and permethylated malto-oligosaccharides, RNase B released N-glycans, articular cartilage released N-glycans and C. elegans whole nematode extract released O-glycans were analyzed. This selection provided linear, high mannose, complex, hybrid, and short, highly branched oligosaccharides for study. Reduction was performed to avoid separation of the α and β anomers of the permethylated oligosaccharides on the PGC columns, as well as to enable distinction of potentially isobaric fragment ions. The LC system used here was optimized for the oligosaccharides presented. However, an increased amount of isopropanol can be used in solvent B to provide better chromatographic properties for more highly retained permethylated oligosaccharide alditols. The branch isomers containing Fuc had clearly unique retention times, whereas positional isomers that varied in the location of Man had retention times that were more similar to one another. The peak shapes of sialylated glycans tended to be more broad than were those of other compounds having similar masses (data not shown). For glycans such as those released from α-one acid glycoprotein where a high proportion of higher molecular weight glycomers occurs solvent B can be replaced with 2:1 isopropanol: acetonitrile. Essentially the same gradient can be used. Typically, these conditions allow resolution of more glycomers than that obtained with 1:1 isopropanol: acetonitrile where these tend to elute as an unresolved group during the 100% solvent B ramp (data not shown).

In this study some key fragment ions were observed and these reiterated trends that have been observed previously in fragmentation patterns of permethylated glycans. The branching pattern of the oligosaccharides could be deduced from CID spectra containing fragments bearing a mass signature of -14 Da for each lost branch. In samples containing Hex and HexNAc residues, abundant A, A, A, and A cross-ring fragments were often observed, thus allowing the assignment of 1-2, 1-4, and 1-6-linkages 3738. Elimination at the C3 of HexNAc was also observed, and provided a chemical signature that could be used to assign C3 substitutions as has been previously reported 1718. In high mannose and hybrid glycans, secondary fragmentation gave rise to product ions that clearly defined the high mannose portion. In highly branched O-glycans, fragmentation across the C2-C3, C3-C4 and C4-C5 bonds was key to structural assignments. Results were highly reproducible in terms of extracted ion chromatogram (XIC) intensity, CID fragment ion intensity and sensitivity. The nomenclature used is that of Costello and Domon 39. Monosaccharide symbols are those suggested by Varki et al4041 with some modification allowing for Hex.

Porous Graphite LC/MS of Malto-oligosaccharides

As shown in Figure 1A, Glc4-8 were well separated when malto-oligosaccharide aliquots of approximately 300 ng of reduced and permethylated malto-oligosaccharides were analyzed. The CID spectrum of Glc5 [M+Na], m/z 1105.64, is shown in Figure 1C. For the malto-oligosaccharides, permethylation without prior reduction resulted in isobaric Y and C as well as Z and B ions. This situation was avoided by carrying out reduction prior to permethylation. As seen in the Figure 1C inset, Y and C and Z and B ions are all unique. An (n = 2, 3, 4) cross-ring fragments were observed at m/z 329.12, 533.27 and 737.42, respectively, thus confirming each β1,4 glycosidic bond in the series. Also observed were secondary (internal) fragments such as the Y4/C3 and Y3/C4 fragment at m/z 449.22. Fragments that correspond to methanol loss from the Cn ion, e.g. m/z 431.22 could also be assigned as Y4/B3 or Y3/B4. The presence of the abundant peak at m/z 839.51 indicated that the methanol loss is important for C4 but the facile Bn and Yn pathways suggest that, at lower m/z values, the Yx/By isomer may be dominant. Therefore, for simplicity, in the following sections these type fragments will be referred to as the Yx/By type. These data demonstrate the capacity of PGC to separate permethylated malto-oligosaccharide alditols and also show the capacity of the method to provide informative fragmentation patterns where linkage, branch number and sequence can be completely assigned.

An external file that holds a picture, illustration, etc.
Object name is nihms-31741-f0001.jpg

LC/MS analysis of permethylated maltooligosaccharide alditols. A, Extracted ion chromatograms of permethylated oligosaccharide alditol maltooligosaccharide series Glc4-8. The Glc4, Glc5 Glc6, Glc7 [M+Na] ions, Glc8 [M+2Na] were detected at m/z 901.45, 1105.56, 1309.66, 1513.76 and 870.43 respectively. Integrated peak areas are proportional to abundances of the glycoforms. B and C, fragment origin and CID QoTOF MS/MS spectra of Glc5, [M+Na]m/z 1105.56. Fragmentation nomenclature is that of Domon and Costello 39. All fragments contain sodium. Monosaccharide component symbols are those suggested by Varki et al40. Key ions are indicated. The inset in C demonstrates that distinct B, Z, C and Z ions are produced from the reduced and permethylated Hexose oligosaccharide.

Porous Graphite LC/MS of High Mannose N-glycans from RNase B

Permethylated high mannose N-glycans have been studied by tandem mass spectrometry and their fragmentation patterns are well established 4243. RNase B contains exclusively high mannose oligosaccharides. While some variation can be expected from differences in the protein source, method of release and purification, the approximate abundance of each glycan size is known, as is the relative distribution of the isomers within each size pool 4445. Man5GlcNAc2 and Man6GlcNAc2 are reported to be ∼57% and ∼31% respectively of the total glycans of RNase B and proton NMR has established that each is present as a single glycoform. There are three Man7GlcNAc2 isomers, which combined are ∼ 4% of total glycans. The isomers are referred to as D1, D2 and D3 and differ only in the position of one α1,2Man residue (see Figure 3). D1 refers to substitution of the α1,2Man lower arm, D2 the central α1,3Man arm, and D3 the upper α1,6Man arm. The D1, D2 and D3 isomers are present in approximately equimolar proportions. Man8GlcNAc2 is reported to be ∼7% of total glycans and is also present as three isomers. The D1D3 isomer is most abundant and is approximately 80 - 85% of Man8GlcNAc2. The D1D2 isomer is ∼10 - 15% and the D2D3 isomer is ∼5% of the total Man8GlcNAc2. Man9GlcNAc2 is ∼ 1% of glycans and is present as a single isoform. High mannose glycan was released from approximately 2 milligrams of lyophilized ribonuclease B powder. Approximately 120 micrograms of mannan was obtained giving a yield of approximately 80% based on 100% glycosylation site occupancy and an average oligosaccharide molecular weight of 1320 using a weighted average for glycans present.

An external file that holds a picture, illustration, etc.
Object name is nihms-31741-f0003.jpg

CID QoTOF MS/MS spectra of Man7GlcNAc2 D1, D2, and D3 isomers. Major product ions, derived from [M+2Na]m/z 1013.49 are indicated in each spectrum and their origins depicted to the structural representations to the right of each spectrum. Spectra and structural diagrams are shown in the order of their elution: A, Man7GlcNAc2 D2; B, Man7GlcNAc2 D3, and C, Man7GlcNAc2 D1. All fragments contain sodium.

To the LC/MS system equipped with a 2 × 50 mm PGC column, approximately 350-ng aliquots of permethylated RNase N-glycan alditols were applied, providing an estimated dynamic range in chemically distinct glycoforms of ∼ 200 to 5.25 ng. The extracted ion chromatogram (XIC) of the RNase B Man5-9GlcNAc2 is shown in Figure 2A. Permethylation renders oligosaccharides within the same compound class nearly chemically equivalent. As a result, mass spectral responses are similar over a fairly broad mass range 46. Integration of all ions from three independent experiments provided a distrubution of Man5-9GlcNAc2 that was comparable to previous estimates determined by NMR and CE analyses 4445 as shown in Table I. The chromatographic positions of each isomer in Man7GlcNAc2 and Man8GlcNAc2 are shown in Figure 2B.

An external file that holds a picture, illustration, etc.
Object name is nihms-31741-f0002.jpg

Extracted ion chromatogram of permethylated oligosaccharide alditols Man5-9GlcNAc2: A, a stacked plot of molecular ions for Man5-9GlcNAc2is shown; B, Three partially overlapping peaks are observed for Man7GlcNAc2 detected at [M+2Na]m/z 1013.49 (bottom) and are labeled as D1, D2, and D3 and three are observed for Man8GlcNAc2 detected at [M+2Na]m/z 1115.55 and are labeled as D1D2, D2D3 and D1D3 (top); C, Selected ion chromatogram of Man9GlcNAc2 produced from approximately 160 fmol. The molecular ion region of the mass spectrum is shown in the inset. See text for details concerning nomenclature.

Table I

Comparison of semiquantitative methods for the analysis of RNase B oligosaccharides

Glycan% abundance LC/MS% abundance NMRa% Isomer abundance CEb
overallisomeroverallisomer
Man5GlcNAc255.1 ± 6.2-57--
Man6GlcNAc230.8 ± 5.2-31--
Man7GlcNAc2 D13.5 ± 1.3441.53138
Man7GlcNAc2 D22.7 ± 1.3341.53337
Man7GlcNAc2 D31.8 ± 0.7221.03525
Man8GlcNAc2 D1D20.2 ± 0.130.41017
Man8GlcNAc2 D2D30.3 ± 0.150.764
Man8GlcNAc2 D1D34.8 ± 0.6925.98478
Man9GlcNAc21.6 ± 0.4-1.0--

Error is expressed as standard error of the mean calculated from three independent analyses.

Values are from reference 49
Values are from reference 50

The CID spectra of the three Man7GlcNAc2 isomers produced during a representative experiment are shown in Figure 3. The CID spectrum of Man7GlcNAc2, [M+2Na]m/z 1013.49, in the time domain 17.05 to 17.26 min is shown in Figure 3A. The derived structure and all key fragments are shown to the right of the spectrum. The ions seen at m/z 1288.60 and 1492.72 are consistent with fragments Y/B5 for the former and Y4x/B5 and Y5α’/B5 for the latter and define an upper arm that contained four Hex residues. Cross ring fragments A4 and A4 at m/z 913.47 and 941.49 define the 1,6-linked arm. Both D2 and D3 isomers can contain these fragments. Prediction of the fragmentation of the D2 structure would forecast the presence of A and A fragments at m/z 301.1 and 329.1, whereas fragmentation of the D3 structure should yield A and A fragments at m/z 505.2 and 533.2. Indeed, fragments are observed at m/z 301.15 and 329.17 and not at m/z 505.2 or 533.2, demonstrating that, in this time domain, the D2 isomer is the major isomer present. Thus the partially resolved peak centered at 17.1 minutes is isomer D2.

The CID spectrum of the time domain 17.93 to 18.13 min is shown in Figure 3B. As in the D2 spectra, ions were observed at m/z 1288.60 for Y/B5, and m/z 1492.72 for Y4x/B5 and Y/B5. As described above, these fragments are consistent with the presence of a branch that contained four Hex units. Cross-ring fragments A4 and A4 at m/z 913.47 and 941.49, were also observed. These define the Man4 branch as linked to the 1,6 arm. However, A and A fragments were detected at m/z 505.24 and 533.27 and not at m/z 301.1 and 329.1. Therefore, the predominant glycan in this time domain was the D3 isomer. The derived structure and the origins of all key fragment ions are shown to the right of the spectrum.

The CID spectrum of the time domain 19.02 to 19.23 min is shown in Figure 3C. An ion is observed at m/z 1084.54, which is consistent with a Y3x/B5 secondary fragment. This ion can not be generated by either isomer D2 or D3. The D1 α1,6 arm can be defined by cross-ring fragments A3a, A3a, A4 and A4, all of which are present in the spectrum at m/z 301.18, 329.17, 709.34 and 737.35 respectively. Thus, the predominant isomer within this time domain is the D1 isomer.

An XIC of the Man9GlcNAc2 [M+2Na] at m/z 1217.63 is shown in Figure 2C from an analysis of a 20 ng aliquot of total mannan using a 0.32 × 100 mm PGC column. Nearly identical tandem mass spectra were obtained across the entire peak region supporting the presence of only one isomer as expected (not shown). An estimated 160 fmol of Man9GlcNAc2 was present in the mixture and its mass spectrum is shown in the inset in Figure 2C. The signal-to-noise ratio was over 8000:1. When the amount of mannan was reduced four-fold, a signal-to-noise ratio of over 500:1 was obtained from approximately 40 fmol of Man9GlcNAc2 (data not shown).

Porous Graphite LC/MS of Articular Cartilage Decorin N-glycans

Decorin is a small leucine-rich proteoglycan of the extracellular matrix involved in cellular proliferation, migration and phenotype. N-glycosylation is thought to be required for its secretion 4748. Bovine articular cartilage decorin (ACD) N-glycans are diverse and some contain antennae with LacdiNAc and fucosyl LacdiNAc and the core can be fucosylated or non-fucosylated. A thorough characterization of ACD N-glycans will be presented in detail in a manuscript currently in preparation by J. Contado-Miller et al.

ACD is a particularly challenging glycoprotein to analyze. As it contains chondroitin and/or dermatan sulfate as well as N-glycan chains. The intact molecular weight averages approximately 100 kDa of which approximately 60% is glycosaminoglycan (GAG). There are three potential glycosylation sites. The N-glycosylation site prediction algorithm NetNGlc 1.0 predicts only two are occupied. The actual site occupancy is not known. In processing first the GAG must be removed followed by the N-glycan. Thus considerable handling of the glycoprotein is required. For this study approximately 25 micrograms of AC decorin was used. Based on total ion intensity compared to permethylated maltooligosaccharide standard, MALDI-TOF spectra of the permethylated AC decorin N-glycans indicated that 1 to 1.5 micrograms of glycan material was released. This is approximately a 70% yield base on an average of 75% site occupancy and a composite average molecular weight glycan of 2000 Da.

In the analysis of decorin glycans carried out as part of the investigation reported here, two dHex1Hex4HexNAc4 isomers were detected. The isomers eluted at 18.6 and 29.1 min, as shown in Figure 4, and are labeled as Peak A and B respectively. CID data were collected for the apparent Fuc1Man4GalNAc1GlcNAc4 [M+2Na]ion, m/z 1039.56, detected from 18.36 to 18.56 minutes. The monosaccharide composition proposed was based on GC/MS monosaccharide analysis of the trimethylsilylated derivatives (manuscript in preparation). The CID spectrum is shown in Figure 4B. Fragments were consistent with a hybrid-type glycan. The core is clearly defined as the unsubstituted chitobiose by the Y1 and Y2 fragments observed at m/z 316.16 and 561.27 respectively. The glycan was hybrid type with a high mannose type upper arm, as evidenced by the B fragment at m/z 445.22 and secondary fragments Y/B5 and Y/B5 at m/z 880.44 and 1084.52. Other fragments that supported a high mannose hybrid-type upper arm were A4 and A4 fragments seen at m/z 505.24 and 533.26. Potential A or A type fragments were not observed at the calculated m/z value of 301.1 and 329.1. Therefore, the terminal Man is not likely to be 1,6-linked and is likely 1,3-linked resulting from incomplete trimming by mannosidases in the Golgi. The lower arm contained a dHex1HexNAc2 extension; this assignment is supported by the presence of a B fragment at m/z 701.35. A B1α’ fragment was observed at m/z 282.13 and is consistent with a terminal HexNAc. The presence of internal dHex-HexNAc is supported by a Y5α’/B fragment observed at m/z 442.22. The 3-O fucosyl substitution is supported by a Y5α’/B5/E1α” fragment at m/z 1297.58, which has undergone a C-3 elimination of O-Fuc. Such fragmentations have been previously reported 1719. The derived structure is shown to the right of the spectrum and the origins of key fragments are indicated.

An external file that holds a picture, illustration, etc.
Object name is nihms-31741-f0004.jpg

LC/MS analysis of articular cartilage decorin N-glycans. A, The extracted ion chromatogram of articular cartilage decorin dHex1Hex4HexNAc4 permethylated oligosaccharide alditol is shown. Two well resolved peaks are observed for [M+2Na]m/z 1039.56 and are labeled as A and B. B and C, CID QoTOF MS/MS spectra of articular cartilage decorin dHex1Hex4HexNAc4 permethylated oligosaccharide alditol isomers. The [M+2Na] ions were isolated at m/z 1039.56. Key ions are indicated in each spectrum (A and C) and their origins are depicted in the structural representations to the right of each spectrum (B and C). All fragments contain sodium.

CID data were collected for the second Fuc1Man4GalNAc1GlcNAc3 [M+2Na] ion, m/z 1039.56, eluting from 29.918 to 30.119 minutes. The spectrum is shown in Figure 4C. Assignment of core fucosylation was supported by a Y1 fragment at m/z 490.28. The presence of a LacdiNAc antenna was supported by fragments B, B, Z and Y at m/z 282.13, 527.28, 1329.56 and 1551.98, respectively. Evidence for a high mannose-type upper arm included cross ring fragments A4 and A4, at m/z 505.24 and 533.29, which define the 1,6-linked upper arm. Secondary fragmentation ions Y/B and Y/B at m/z 880.44 and 1084.51 add further support for the hybrid structure. The derived structure is shown to the right of the spectrum.

Porous Graphite LC/MS of Caenorhabditis elegans O-glycans

The C. elegans O-glycan used here was from a previous study 34 generated from approximately 10 grams wet weight of nematodes. LC/MS analyses were performed on approximately 20 ng aliquots of permethylated alditols using the 300 micron PGC column.

If the analytes are sufficiently separated by LC and/or there are no co-incident isobaric product ions from different but co-eluting parent ions, it is often useful to perform a pseudo MS experiment using QoTOF mass spectrometry. Here the voltage offset relationships between the orifice, ring, skimmer and Q0 are set to favor prompt fragmentation. Results from ESI-TOF MS, MS and pseudo MS experiments are presented below for the analysis of the C. elegans permethylated alditols, Hex3HexNAc1-ol O-glycan isomers.

C. elegans produces neutral O-glycans that contain a mammalian-like Galβ1,3GalNAc type I core and are highly branched and short 3449. Both core residues can be highly substituted. The core Gal of the neutral O-glycans has been reported to be 4,6-disubstituted, while only GlcA has been reported to substitute the C-3 position 3449. The core GalNAc can be up to 3,4,6-trisubstituted, whereas the C-3 position has been reported to contain Gal exclusively 49. The GalNAc C-6 position has been reported to be Gal- or Glc-substituted 49. Mutant strains that are affected in O-glycan biosynthesis have altered susceptibilities to bacterial and fungal pathogens and are currently used to study host - pathogen interactions and innate immunity 345052. In the experiments reported here, the O-glycans were chromatographically separated into neutral and acidic fractions prior to derivatization and analysis by LC/MS. The results for the neutral O-glycans are reported here. The XIC of Hex3HexNAc1-ol, m/z[M+Na] 942.55, is shown in Figure 5. Three major peaks, A, B and C, were observed and were centered at 38.0, 48.5 and 56.0 min respectively.

An external file that holds a picture, illustration, etc.
Object name is nihms-31741-f0005.jpg

Trace ion chromatogram of reduced and permethylated C. elegans Hex3HexNAc1. Three major peaks are observed for [M+Na]m/z 942.55 and are labeled as A, B and C.

The summed CID spectrum of Peak A generated in the time domain 37.54 - 37.74 minutes is shown in Figure 6A. The derived oligosaccharide structure is shown to the right of the spectrum. The presence of a 3,4,6-trisubstituted HexNAc is supported by secondary fragment Y1x/Y1y/Z1x at m/z 270.14. Individual Hex losses were observed at m/z 706.39 and 724.41 for Z1x and Y1x fragments, respectively. The dominant Z1x intensity likely results, in part, from a C-3 elimination of 1,3-linked Hex from the core HexNAc-ol 1719. The abundant fragment at m/z 415.21 results from formation of the secondary fragment C1x/B1y with loss of CH3NHCOCH3. This was supported by the pseudo MS fragmentation of the m/z 415.21 fragment shown in Figure 7A. Ions consistent with the C-3 cleavage are observed at m/z 329.18 for the A1 fragment. The fragment observed at m/z 383.21 may have resulted from a methanol loss which could take the form of a A1 fragment. Further support is given by the presence of C1 and Z1 - CH3OH ions at m/z 259.16 and 165.09. The derived structure is a highly branched 3,4,6-trisubstituted core1-type glycan as shown in Figure 6A.

An external file that holds a picture, illustration, etc.
Object name is nihms-31741-f0006.jpg

CID QoTOF MS/MS spectra of C. elegans Hex3GalNAc1 permethylated oligosaccharide alditol isomers isolated at [M+Na]m/z 942.55. The spectra of Peaks A, B and C are shown. Key ions are indicated in each spectrum and their origins and resultant linkage assignments are depicted in the structural representations to the right of each spectrum. Peaks B and C each contained two isomers, the lower arm of which was either Hex1,4Hex1,3- or Hex1,6Hex1,3-linked to the core GalNAc-ol. Specific structural assignments are based on pseudo MS spectrum shown in Figure 9. All fragments contain sodium. Monosaccharide identities are based on those of Varkii et al., [49] as in other figures except that hexoses that can be either Gal or Glc are represented as circles with an X.

An external file that holds a picture, illustration, etc.
Object name is nihms-31741-f0007.jpg

Pseudo MS spectra of key product ions of Hex3GalNAc1 permethylated alditol isomers. A:X2m/z 415.19 ion of Peak A, B; Zm/z 502.26 ion for Peak B, C; Zm/z 502.26 ion for Peak C, and D; Am/z 463.22 ion of Peak C. Proposed structures are shown above their spectra. In B and C, the Zm/z 502.26 ion can be either a 1,4 or 1,6 glycosidic bond. See text for details. All fragments contain sodium.

The CID spectra of Peaks B and C generated in the time domiains 47.80 - 48.00 and 55.22 - 55.42 min were similar, as shown in Figure 6B and C. The isomers are both biantennary, as indicated by the observation of secondary fragments Y1x/Z1y and Z1y/Y1x at m/z 284.15. Evidence for the presence of a Hex2 arm in both spectra was supported by the appearance of the product ion at m/z 463.25, which was consistent with a C fragment. No ion intensity is observed at m/z 270 in either spectrum, consistent with the absence of a trisubstituted configuration. The abundant Z fragment ion at m/z 502.29 likely reflects a C-3 elimination of the 1,3-linked disaccharide substituent. The presence of cross-ring fragments A2, A2 and A2 at m/z 301.13, 329.16 and 359.18 suggests that both 1,4 and 1,6 linkages could be present in the Hex2 moiety. Taken together, these isomers fit the general linkage configuration Hex1-4,6Hex1,3(Hex1-4,6)HexNAc-ol. Pseudo MS experiments were performed to resolve the remaining ambiguity in linkage assignment.

MS CID spectra from both Peaks B and C contained the C-type fragment at m/z 463.22. The pseudo MS analysis of both produced nearly identical spectra. A representative example is shown in Figure 7B. In addition to glycosidic fragments C1, Y1 and C1-CH3OH observed at m/z 259.12, 245.09 and 227.09, respectively, crossring fragments were observed at m/z 301.12, 329.12, 359.17 and 389.17. A 1,4-linked pair can generate m/z 329.12, 359.17 and 389.17 for A2, A2 and A2 fragments. A 1,6-linked pair can generate m/z 301.12, 329.12 and 389.17 for A2, A2, and A2. The m/z 301.12 and 359.17 fragments are unique to the 1,6- and 1,4-linked pairs, respectively. As the abundance of the m/z 359.17 fragment is very high, it is likely that the majority of the disaccharide pair is present as the 1,4-linked pair with a minor component containing a 1,6-linked pair. Therefore, two isomeric 1,3-linked Hex2 arms exist in each of Peaks B and C.

The C-type fragments isolated at m/z 502.26 in the respective time domains for Peaks B and C were also subjected to pseudo MS. As shown in Figure 7, C and D, each produced a unique spectrum. As each has undergone a C-3 substituent loss, only assignments to 1,4- or 1,6-linkages are possible. The pseudo MS spectrum produced from Peak C is shown in Figure 7C. Loss of CH2CO (ketene) is observed at m/z 460.25 and is followed by two CH3OH losses shown by fragments observed at m/z 428.22 and 396.20. The authenticity of this series is supported by a Y1-type fragment that has lost CH2CO and two equivalents of CH3OH, as seen at m/z 178.08. These fragments likely originate from a 1,4-linked configuration, as rationalized in Scheme I. The pseudo MS spectrum produced from Peak C is shown in Figure 7D. Fragments consistent with A2 and A2 - CH2O were observed at m/z 373.18 and 343.17. Fragments consistent with C1 and C1 with loss of H2O and CH2O were observed at m/z 259.12 and 211.08. A fragment was also observed at m/z 165.09, which is consistent with the presence of a Y1 fragment that has lost CH3CONHCH3 and two CH2O groups. It is known that in C. elegans the C-6 position of the GalNAc core can be C-6 substituted with Hex as either Gal or Glc. It is likely that the m/z 415.19 fragment produced from Peak A contained a 1,6-linkage, as described previously. The pseudo MS spectrum of the m/z 502.26 fragment produced from Peak B was similar to that produced by the m/z 415.19 fragment of Peak A, in that both contained prominent fragments at m/z 165.09 and 259.12. The m/z 502.26 fragment from Peak C produced a spectrum that was different from both and arises from a 1,4-linked fragment ion. Therefore, it is likely that both isomers in Peak B contain a 1,6-linked Hex, while isomers in Peak C contains a 1,4-linked Hex. Taken together, these data show that two major isomers exist in each peak. Both Peaks B and C contain isomers with Hex1,6Hex1,3- and Hex1,4Hex1,3-linked arms. Peak B contains a Hex1,6-arm and Peak C contains a Hex1,4-arm. These tetrasaccharide assignments are consistent with the general structural paradigm seen in O-glycans that have been documented in C. elegans3449. The structural assignments are shown to the right of the spectra in Figure 6.

An external file that holds a picture, illustration, etc.
Object name is nihms-31741-f0008.jpg

Decomposition of the 1-4-linked disaccharide isolated at m/z 502.26. The proposed pathway for the formation of key fragment ions is shown.

Conclusions

An LC/MS method for analysis of reduced and permethylated oligosaccharides has been developed that utilizes a porous graphite based LC system coupled with a QoTOF mass spectrometer. LC/MS separation and identification of permethylated oligosaccharide alditol isomers have been demonstrated for the first time. The method is rapid, reproducible and sensitive. When a 320-μm column was used, 40 fmol of oligosaccharide could be easily detected and fully characterized. Further reduction of chromatographic scale should significantly decrease the sample requirement.

The CID spectra of reduced and permethylated oligosaccharides provide informative fragmentation. The utility of permethylation in oligosaccharide structural determination is well documented. The linkage, branch and sequence information provided in their fragmentation spectra is considered superior to most other derivatives. In the examples presented here, it is made clear that the utility of the highly informative CID spectra of permethylated oligosaccahrides is enhanced by high resolution liquid chromatography prior to introduction of the sample into the mass spectrometer since isomers are temporally resolved before ionization, thus reducing ambiguity in mass spectrometry-based structural assignments. Examples are given to demonstrate assignment of linkage and branch patterns in a variety of oligosaccharides ranging from linear to highly branched glycans and for both positional and branch isomers. Although the capacity of the system presented to differentiate isomers could be saturated by highly complex whole organism glycomes, the present system was employed successfully for structural assignments using up to the pseudo MS dimension. Further improvements to the LC system and additional levels of MSn should enhance the capacity for glycoform resolution and assignment.

It is important to note that the method presented here has some inherent limitations. For some compounds, permethylation is not suitable. During permethylation, some labile groups such as O-acetyl groups are lost and polar groups, such as phosphate and sulfate, can cause poor recovery during standard extractions used in the derivatization protocol. Therefore, if these substituents are suspected, an alternative approach should be considered. As with all chromatographic methods, resolution of the presented method is limited. Therefore, when highly complex mixtures are expected, prior sub-fractionation may be required. In its present form the method is most useful for analysis of neutral saccharides up to those with tetraantennary configurations. Those oligosaccharides that contain Sialic acids tend to give broader peaks as the number of sialic acids increases.

Use of the permethylated derivatives for structural analysis of oligosaccharides is an active area of research. An online permethylation method was recently developed 53. These authors demonstrate that oligosaccharides could be permethylated in minutes, reagents and byproducts could be efficiently removed and analytes directly introduced into an LC/MS system. Online derivatization coupled with LC/MS of permethylated oligosaccharide alditols has obvious advantages and we are in the process of adapting online derivatization to the LC/MS platform. Reinhold et al. have recently reported the complete assignment of oligosaccharide structures of permethylated glycans using a database library, FragLib, in conjunction with their Oligosaccharide Subtree Constraint Algorithm (OSCAR) 385455. This algorithm incorporates analyst-selected MSn ion fragmentation pathways for elucidation of oligosaccharide topology using a top-down sequencing strategy. Complete assignment was reported for a series of oligosaccharides, where mass, branch, linkage, diastereomer specific composition and anomeric configuration were determined. Undoubtedly, coupling of online derivatization, LC/MS and bioinformatics will lead to even more facile structural assignment of oligosaccharides. This method will aid in efforts toward the overall goal of the bioanalyst to provide complete structural assignments for glycosylation details, even when sample material is limited.

Acknowledgements

We thank Dr. Patrick Van Roey of the Wadsworth Center, Albany NY, for the kind gift of the PNGase F plasmid used in this study. This work was supported by National Institutes of Health grants P41 RR10888 and S10 RR015942.

Mass Spectrometry Resource, Department of Biochemistry, Boston University School of Medicine, 670 Albany St, Boston, Massachusetts 02118-2646, USA
Address reprint requests to: John F. Cipollo, Ph. D, Food and Drug Administration, Center for Biologics Evaluation and Research, Division of Bacterial, Parasitic and Allergenic Products, 8800 Rockville Pike, Building 29A, Bethesda MD 20892-0001, Phone 301-496-2902, Fax 301-402-2776 email: vog.shh.adf@ollopic.nhoJ
Publisher's Disclaimer

Abstract

This communication reports the development of an LC/MS platform for the analysis of permethylated oligosaccharide alditols that, for the first time, demonstrates routine online oligosaccharide isomer separation of these compounds prior to introduction into the mass spectrometer. The method leverages a high resolution liquid chromatography system with the superior fragmentation pattern characteristics of permethylated oligosaccharide alditols that are dissociated under low-energy collision conditions using quadrupole orthogonal time-of-flight (QoTOF) instrumentation and up to pseudo MS mass spectrometry. Glycoforms, including isomers, are readily identified and their structures assigned. The isomer-specific spectra include highly informative cross-ring and elimination fragments, branch position specific signatures and glycosidic bond fragments, thus facilitating linkage, branch and sequence assignment. The method is sensitive and can be applied using as little as 40 fmol of derivatized oligosaccharide. Because permethylation renders oligosaccharides nearly chemically equivalent in the mass spectrometer, the method is semi-quantitative and, in this regard, is comparable to methods reported using high field NMR and capillary electrophoresis. In this post - genomic age, the importance of glycosylation in biological processes has become clear. The nature of many of the important questions in glycomics is such that sample material is often extremely limited, thus necessitating the development of highly sensitive methods for rigorous structural assignment of the oligosaccharides in complex mixtures. The glycomics platform presented here fulfills these criteria and should lead to more facile glycomics analyses.

Keywords: LC/MS, permethylation, isomer, QoTOF MS, C. elegans, tandem mass spectrometry
Abstract

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Footnotes

References

  • 1. Apweiler R, Hermjakob H, Sharon NOn the frequency of protein glycosylation, as deduced from analysis of the SWISS-PROT database. Biochim Biophys Acta. 1999;1473(1):4–8.[PubMed][Google Scholar]
  • 2. Ben-Dor S, Esterman N, Rubin E, Sharon NBiases and complex patterns in the residues flanking protein N-glycosylation sites. Glycobiology. 2004;14(2):95–101.[PubMed][Google Scholar]
  • 3. Kornfeld SDiseases of abnormal protein glycosylation: an emerging area. J Clin Invest. 1998;101(7):1293–5.[Google Scholar]
  • 4. Gorelik E, Galili U, Raz AOn the role of cell surface carbohydrates and their binding proteins (lectins) in tumor metastasis. Cancer Metastasis Rev. 2001;20(3-4):245–77.[PubMed][Google Scholar]
  • 5. Lowe JBGlycan-dependent leukocyte adhesion and recruitment in inflammation. Curr Opin Cell Biol. 2003;15(5):531–8.[PubMed][Google Scholar]
  • 6. Haltiwanger RS, Lowe JBRole of glycosylation in development. Annu Rev Biochem. 2004;73:491–537.[PubMed][Google Scholar]
  • 7. Lis H, Sharon N. Protein glycosylation. Structural and functional aspects. Eur J Biochem. 1993;218(1):1–27.[PubMed]
  • 8. Laine RAA calculation of all possible oligosaccharide isomers both branched and linear yields 1.05 × 10(12) structures for a reducing hexasaccharide: the Isomer Barrier to development of single-method saccharide sequencing or synthesis systems. Glycobiology. 1994;4(6):759–67.[PubMed][Google Scholar]
  • 9. Freeze HHUpdate and perspectives on congenital disorders of glycosylation. Glycobiology. 2001;11(12):129R–43R.[PubMed][Google Scholar]
  • 10. Jaeken J, Matthijs GCongenital disorders of glycosylation. Annu Rev Genomics Hum Genet. 2001;2:129–51.[PubMed][Google Scholar]
  • 11. Butler M, Quelhas D, Critchley AJ, Carchon H, Hebestreit HF, Hibbert RG, Vilarinho L, Teles E, Matthijs G, Schollen E, Argibay P, Harvey DJ, Dwek RA, Jaeken J, Rudd PMDetailed glycan analysis of serum glycoproteins of patients with congenital disorders of glycosylation indicates the specific defective glycan processing step and provides an insight into pathogenesis. Glycobiology. 2003;13(9):601–22.[PubMed][Google Scholar]
  • 12. Harvey DJMatrix-assisted laser desorption/ionization mass spectrometry of carbohydrates. Mass Spectrom Rev. 1999;18(6):349–450.[PubMed][Google Scholar]
  • 13. Mechref Y, Novotny MV, Krishnan CStructural characterization of oligosaccharides using MALDI-TOF/TOF tandem mass spectrometry. Analytical Chemistry. 2003;75(18):4895–4903.[PubMed][Google Scholar]
  • 14. Muzikar J, Mechref Y, Huang Y, Novotny MEnhanced post-source decay and cross-ring fragmentation of oligosaccharides facilitated by conversion to amino derivatives. Rapid Communi Mass Spectrom. 2004;18(13):1513–1518.[PubMed][Google Scholar]
  • 15. Morelle W, Slomianny MC, Diemer H, Schaeffer C, van Dorsselaer A, Michalski JCFragmentation characteristics of permethylated oligosaccharides using a matrix-assisted laser desorption/ionization two-stage time-of-flight (TOF/TOF) tandem mass spectrometer. Rapid Commun Mass Spectrom. 2004;18(22):2637–49.[PubMed][Google Scholar]
  • 16. Perreault H, Costello CEStereochemical effects on the mass spectrometric behavior of native and derivatized trisaccharide isomers: comparisons with results from molecular modeling. J Mass Spectrom. 1999;34(3):184–97.[PubMed][Google Scholar]
  • 17. Viseux N, de Hoffmann E, Domon BStructural assignment of permethylated oligosaccharide subunits using sequential tandem mass spectrometry. Anal Chem. 1998;70(23):4951–4959.[PubMed][Google Scholar]
  • 18. Viseux N, de Hoffmann E, Domon BStructural analysis of permethylated oligosaccharides by electrospray tandem mass spectrometry. Anal Chem. 1997;69(16):3193–3198.[PubMed][Google Scholar]
  • 19. Morelle W, Faid V, Michalski JCStructural analysis of permethylated oligosaccharides using electrospray ionization quadrupole time-of-flight tandem mass spectrometry and deutero-reduction. Rapid Commun Mass Spectrom. 2004;18(20):2451–64.[PubMed][Google Scholar]
  • 20. Delaney J, V PLiquid chromatography ion trap mass spectrometric analysis of oligosaccharides using permethylated derivatives. Rapid Communi Mass Spectrom. 2001;15:325–334.[PubMed][Google Scholar]
  • 21. Novotny MV, Mechref YNew hyphenated methodologies in high-sensitivity glycoprotein analysis. J Sep Sci. 2005;28(15):1956–68.[Google Scholar]
  • 22. Rudd PM, Colominas C, Royle L, Murphy N, Hart E, Merry AH, Hebestreit HF, Dwek RAA high-performance liquid chromatography based strategy for rapid, sensitive sequencing of N-linked oligosaccharide modifications to proteins in sodium dodecyl sulphate polyacrylamide electrophoresis gel bands. Proteomics. 2001;1(2):285–94.[PubMed][Google Scholar]
  • 23. Saba JA, Shen XD, Jamieson JC, Perreault HEffect of 1-phenyl-3-methyl-5-pyrazolone labeling on the fragmentation behavior of asialo and sialylated N-linked glycans under electrospray ionization conditions. Rapid Communi Mass Spectrom. 1999;13(8):704–711.[PubMed][Google Scholar]
  • 24. Cipollo JF, Awad AM, Costello CE, Hirschberg CBN-Glycans of Caenorhabditis elegans are specific to developmental stages. J Biol Chem. 2005;280(28):26063–72.[PubMed][Google Scholar]
  • 25. Kawasaki N, Haishima Y, Ohta M, Itoh S, Hyuga M, Hyuga S, Hayakawa TStructural analysis of sulfated N-linked oligosaccharides in erythropoietin. Glycobiology. 2001;11(12):1043–1049.[PubMed][Google Scholar]
  • 26. Wada Y, Tajiri M, Yoshida SHydrophilic affinity isolation and MALDI multiple-stage tandem mass spectrometry of glycopeptides for glycoproteomics. Anal Chem. 2004;76(22):6560–5.[PubMed][Google Scholar]
  • 27. Bruggink C, Maurer R, Herrmann H, Cavalli S, Hoefler FAnalysis of carbohydrates by anion exchange chromatography and mass spectrometry. J Chromatogr A. 2005;1085(1):104–9.[PubMed][Google Scholar]
  • 28. Tolstikov VV, Fiehn OAnalysis of highly polar compounds of plant origin: combination of hydrophilic interaction chromatography and electrospray ion trap mass spectrometry. Anal Biochem. 2002;301(2):298–307.[PubMed][Google Scholar]
  • 29. Dage JL, Ackermann BL, Halsall HBSite localization of sialyl Lewis(x) antigen on alpha1-acid glycoprotein by high performance liquid chromatography-electrospray mass spectrometry. Glycobiology. 1998;8(8):755–760.[PubMed][Google Scholar]
  • 30. Kawasaki N, Ohta M, Hyuga S, Hashimoto O, Hayakawa TAnalysis of carbohydrate heterogeneity in a glycoprotein using liquid chromatography/mass spectrometry and liquid chromatography with tandem mass spectrometry. Anal Biochem. 1999;269(2):297–303.[PubMed][Google Scholar]
  • 31. Ninonuevo M, An H, Yin H, Killeen K, Grimm R, Ward R, German B, Lebrilla CNanoliquid chromatography-mass spectrometry of oligosaccharides employing graphitized carbon chromatography on microchip with a high-accuracy mass analyzer. Electrophoresis. 2005;26(19):3641–9.[PubMed][Google Scholar]
  • 32. Boulenguer P, Leroy Y, Alonso JM, Montreuil J, Ricart G, Colbert C, Duquet D, Dewaele C, Fournet BContinuous-flow fast atom bombardment-mass spectrometry of permethylated oligosaccharides: a comparative study of direct mixture analysis with packed capillary column liquid chromatography-fast atom bombardment-mass spectrometry. Anal Biochem. 1988;168(1):164–70.[PubMed][Google Scholar]
  • 33. Verostek MF, Lubowski C, Trimble RBSelective organic precipitation/extraction of released N-glycans following large-scale enzymatic deglycosylation of glycoproteins. Anal Biochem. 2000;278(2):111–22.[PubMed][Google Scholar]
  • 34. Cipollo JF, Awad AM, Costello CE, Hirschberg CBsrf-3, a mutant of Caenorhabditis elegans, resistant to bacterial infection and to biofilm binding, is deficient in glycoconjugates. J Biol Chem. 2004;279(51):52893–903.[PubMed][Google Scholar]
  • 35. Ciucanu I, Kerek FA simple and rapid method for the permethylation of carbohydrates. Carbohydr Res. 1984;131(2):209–217.[PubMed][Google Scholar]
  • 36. Cipollo JF, Costello CE, Hirschberg CBThe fine structure of Caenorhabditis elegans N-glycans. J Biol Chem. 2002;277(51):49143–57.[PubMed][Google Scholar]
  • 37. Sheeley DM, Reinhold VNStructural characterization of carbohydrate sequence, linkage, and branching in a quadrupole Ion trap mass spectrometer: neutral oligosaccharides and N-linked glycans. Anal Chem. 1998;70(14):3053–3059.[PubMed][Google Scholar]
  • 38. Ashline D, Singh S, Hanneman A, Reinhold V. Congruent Strategies for Carbohydrate Sequencing. 1. Mining Structural Details by MSn. Anal Chem. 2005
  • 39. Domon B, Costello CEA systematic nomenclature for fragmentatios in FAB-MS/MS spectra of glycoconjugates. Glycoconj J. 1988;5:397–409.[PubMed][Google Scholar]
  • 40. Varki A, Cummings R, Esko J, Freeze H, Hart G, Marth G Essentials of Glycobiology. Cold Spring Harbor Laboratory Press; Cold Spring Harbor, NY: 1999. [PubMed][Google Scholar]
  • 41. , In[PubMed]
  • 42. Stephens E, Maslen SL, Green LG, Williams DHFragmentation characteristics of neutral N-linked glycans using a MALDI-TOF/TOF tandem mass spectrometer. Anal Chem. 2004;76(8):2343–54.[PubMed][Google Scholar]
  • 43. Wuhrer M, Deelder AMMatrix-assisted laser desorption/ionization in-source decay combined with tandem time-of-flight mass spectrometry of permethylated oligosaccharides: targeted characterization of specific parts of the glycan structure. Rapid Commun Mass Spectrom. 2006;20(6):943–51.[PubMed][Google Scholar]
  • 44. Fu D, Chen L, O’Neill RAA detailed structural characterization of ribonuclease B oligosaccharides by 1H NMR spectroscopy and mass spectrometry. Carbohydr Res. 1994;261(2):173–86.[PubMed][Google Scholar]
  • 45. Guttman A, Pritchett TCapillary gel electrophoresis separation of high-mannose type oligosaccharides derivatized by 1-aminopyrene-3,6,8-trisulfonic acid. Electrophoresis. 1995;16(10):1906–11.[PubMed][Google Scholar]
  • 46. Reinhold VN, Reinhold BB, Costello CECarbohydrate molecular weight profiling, sequence, linkage, and branching data: ES-MS and CID. Anal Chem. 1995;67(11):1772–1784.[PubMed][Google Scholar]
  • 47. Seidler DG, Faiyaz-Ul-Haque M, Hansen U, Yip GW, Zaidi SH, Teebi AS, Kiesel L, Gotte MDefective glycosylation of decorin and biglycan, altered collagen structure, and abnormal phenotype of the skin fibroblasts of an Ehlers-Danlos syndrome patient carrying the novel Arg270Cys substitution in galactosyltransferase I (beta4GalT-7) J Mol Med. 2006[PubMed][Google Scholar]
  • 48. Seo NS, McQuillan DJ, Hook MThe role of glycosylation in the secretion of proteoglycans. Scientific World Journal. 2006;6:491–3.[Google Scholar]
  • 49. Guerardel Y, Balanzino L, Maes E, Leroy Y, Coddeville B, Oriol R, Strecker GThe nematode Caenorhabditis elegans synthesizes unusual O-linked glycans: identification of glucose-substituted mucin-type O-glycans and short chondroitin-like oligosaccharides. Biochem J. 2001;357(Pt 1):167–82.[Google Scholar]
  • 50. Nicholas HR, Hodgkin JResponses to infection and possible recognition strategies in the innate immune system of Caenorhabditis elegans. Mol Immunol. 2004;41(5):479–93.[PubMed][Google Scholar]
  • 51. Gravato-Nobre MJ, Hodgkin JCaenorhabditis elegans as a model for innate immunity to pathogens. Cell Microbiol. 2005;7(6):741–51.[PubMed][Google Scholar]
  • 52. Silverman MA, Blaxter ML, Link CDBiochemical Analysis of Caenorhabditis elegans Surface Mutants. J Nematology. 1997;29(3):296–305.[Google Scholar]
  • 53. Kang P, Mechref Y, Klouckova I, Novotny MVSolid-phase permethylation of glycans for mass spectrometric analysis. Rapid Commun Mass Spectrom. 2005;19(23):3421–8.[Google Scholar]
  • 54. Lapadula AJ, Hatcher PJ, Hanneman AJ, Ashline DJ, Zhang H, Reinhold VN. Congruent strategies for carbohydrate sequencing. 3. OSCAR: an algorithm for assigning oligosaccharide topology from MS(n) data. Anal Chem. 2005;77(19):6271–9.
  • 55. Zhang H, Singh S, Reinhold VN. Congruent Strategies for Carbohydrate Sequencing. 2. FragLib: An MSn Spectral Library. Anal Chem. 2005
Collaboration tool especially designed for Life Science professionals.Drag-and-drop any entity to your messages.