Glycobiology Advance Access originally published online on June 12, 2006
Glycobiology 2006 16(9):874-890; doi:10.1093/glycob/cwl011
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Isomer and glycomer complexities of core GlcNAcs in Caenorhabditis elegans
4 Department of Chemistry, Center for Structural Biology, University of New Hampshire, Durham, NH 03824
1 To whom correspondence should be addressed; e-mail: vnr{at}unh.edu
2 Present address: Wyeth BioPharma, Andover, MA 01833
3 Present address: Faculty of Medicine of Ribeirão Preto, University of São Paulo, Av. Bandeirantes 3900, 14049-900, Ribeirão Preto, São Paulo, Brazil
Received on April 16, 2006; revised on May 31, 2006; accepted on June 2, 2006
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
Analysis of protein glycosylation within the nematode Caenorhabditis elegans has revealed an abundant and unreported set of core chitobiose modifications (CCMs) to N-linked glycans. With hydrazine release, an array of glycomers and isobars were detected with hexose extensions on the 3- and 3,6-positions of the penultimate and reducing terminus, respectively. A full complement of structures includes a range of glycomers possessing a Galß(1-4)Fuc disaccharide at the 3- and 6-positions of the protein-linked GlcNAc. Importantly, enzymatic (PNGase F/A) release failed to liberate many of these extended structures from reduced and alkylated peptides and, as a consequence, such profiles were markedly deficient in a representation of the worm glycome. Moreover, the 3-linked Galß(1-4)Fuc moiety was notably resistant to a range of commercial galactosidases. For identification, the fragments were spectrum-matched with synthetic products and library standards using sequential mass spectrometry (MSn). A disaccharide observed at the 3-position of penultimate GlcNAc, indicating a Hex-Fuc branch on some structures, was not further characterized because of low ion abundance in MSn. Additionally, a Hex-Hex-Fuc trisaccharide on the 6-position of proximal GlcNAc was also distinguished on select glycomers. Similar branch extensions on 6-linked core fucosyl residues have recently been reported among other invertebrates. Natural methylation and numerous isobars complement the glycome, which totals well over 100 individual structures. Complex glycans were detected at lower abundance, indicating glucosaminyltransferase-I (GnT-I) and GnT-II activity. A range of phosphorylcholine (PC)-substituted complex glycans were also confirmed following a signature two-stage loss of PC during MSn analysis, although the precursor ion was not observed in the mass profiles. In a similar manner, numerous other minor glycans may be present but unobserved in hydrazine-release profiles dominated by fucosylated structures. All CCM structures, including multiple isomers, were determined without chromatography by gas-phase disassembly (MSn) in Paul and linear ion trap (IT) instruments.
Key words: C. elegans / core chitobiose modifications / Galß(1-4)Fuc / MSn / paucimannose
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
Caenorhabditis elegans is a free-living nematode that, like humans, undergoes a complex developmental process from embryogenesis through adulthood, has a nervous system including a "brain," and is even capable of rudimentary learning. In these remarkable features, it shares many indispensable biological characteristics of human biology, including nerve function, morphogenesis, development, and senescence. With but 959 somatic cells and a transparent 1-mm long body, this eukaryote provides researchers the perfect compromise between complexity and tractability (Wood, 1988
). In the company of such features, it is easy to see why this organism was chosen as the first genome to be fully sequenced. Post-genomic research is currently targeting dynamic aspects of the worm’s biology, including proteomics and post-translational modifications. Glycobiologists have cloned and expressed some of the worm genes involved in N-glycosylation (DeBose-Boyd et al., 1998; Chen et al., 2002; Warren et al., 2002; Zheng et al., 2002). These studies, as well as others addressing protein glycosylation using the worm model, have been well summarized (Schachter, 2004). We have investigated GnT-I triple null (Gly-12, Gly-13, and Gly-14) worms, noting unusual and deficient fucosylation patterns among the phenotypically distinct worms (Zhu et al., 2004). It is clear that these three genes (which all encode GnT-I) have different protein targets and different functions (Schachter et al., 2005). To better understand the observed changes noted in mutant worms, we have undertaken a detailed structural analysis of the N-glycans synthesized by wild-type C. elegans. These hydrazine-released fucosylated glycans comprise a major fraction of the worm’s glycome and have been missed or markedly reduced in earlier studies relying on enzymatic release.
Previous investigations of N-glycans in C. elegans have frequently employed release using endoglycosidases A and F (Altmann et al., 2001; Cipollo et al., 2002; Haslam et al., 2002; Schachter et al., 2002; Paschinger et al., 2004). Hydrazinolysis was used in one previous study to release N-glycans from a crude membrane fraction (Natsuka et al., 2002), and in two other investigations, lectins were employed to capture soluble glycoproteins before hydrazinolysis (Hirabayashi et al., 2002; Natuska et al., 2005). Results from these prior studies have been summarized (Haslam and Dell, 2003) and together indicate at least five structural classes comprise the N-glycome of C. elegans: these include high-mannose, truncated complex, phosphorylcholine (PC)-containing, paucimannose, and the high-fucose glycans. Because many fucosylated structures include modifications of the core chitobiose, and are not always highly fucosylated, we further identify new and intriguing structures as core chitobiose modifications (CCMs). Recently, a wider range of hybrid and complex glycans has also been reported (Cipollo et al., 2005; Natsuka et al., 2005), and results from the latter study also indicate a variety of PC-substituted complex structures. Correlation of glycan mass profiles with physiological change, or gene modulation, provides a powerful approach to glycan function, but in using such strategies a representative glycome remains fundamental, as does a comprehensive understanding of all components comprising each ion. This latter feature is an often-overlooked aspect in structural characterization.
In this study, we detail a dominant set of CCM structures that have not been previously reported in the N-glycans of C. elegans. In addition to the enzymatically released glycans, hydrazinolysis complements such structures with a set of fucosylated and galactosylated glycans not readily obtained with endoglycosidase treatment. As a consequence, a large proportion of the N-glycome has apparently been missed. Methylation and sequential mass spectrometry (MSn) analysis of these hydrazine-released glycans exposed a diverse mixture of structures displaying branching from the reducing terminal and penultimate GlcNAc residues. These CCMs released from mixed-stage worms would be difficult to resolve chromatographically because of subtle structural variations and a multiplicity of isomers. Although hydrazine release was used in some of the earlier studies, either detailed analysis was not the goal (Hirabayashi et al., 2002; Natsuka et al., 2005) or the glycans were missed for some other reason such as a lack of comparable standards (Natsuka et al., 2002). In this report, the protein lysates and the released glycans were handled as a single pool throughout the analysis. Thus, the resulting mass profiles represent a comparative measure of the worm proteoglycome. To assist detailed structural characterization, the glycans were either methylated or reduced first and then methylated before MSn analysis. Methylation amplifies MSn disassembly by rendering well-defined glycosidic and cross-ring cleavage ions that can be differentiated from the open hydroxyl groups produced during collision induced dissociation (CID). As such, fragments can be unambiguously tracked to their specific location in the original structure through repeated rounds of MSn even in the presence of structural isomers (Ashline et al., 2005).
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
N-glycan profiles
Both enzymatic and chemical procedures were used to release glycans. The products were mass profiled and compared as native, reduced-native, methylated, and reduced-methylated derivatives. This correlation of multiple profiles provided an analytical starting point with important structural insight well before detailed analysis by MSn. This immediately established the limitations of an enzymatic release, and an early preliminary report discussed these observations (Hanneman and Reinhold, 2003
). Glycan profiles following PNGase-F and PNGase-A treatment were comparable to published reports that were dominated by a series of high-mannose structures (Altmann et al., 2001; Cipollo et al., 2002, 2005; Haslam et al., 2002; Schachter et al., 2002; Paschinger et al., 2004; Griffitts et al., 2005). Following hydrazine treatment, the high-mannose dominance was lost to a new set of highly fucosylated ions that also extended to higher mass (Figure 1). Typical results comparing release profiles of PNGase-A and hydrazine are provided (Table I). Within these profiles, only four glycans were noted to possess more than two HexNAcs, reflecting a relatively low abundance of complex glycans. Methylation and MSn analysis of the two ions containing three HexNAcs confirmed them to be GnT-I products, GlcNAc3Man3 and its homolog fucosylated at the 6-position of the reducing terminal GlcNAc. Two minor ions in the MS profiles included four HexNAcs, indicating GnT-II activity. Structural details regarding truncated complex glycans were not further pursued, and information about complex glycans in C. elegans has been reported elsewhere (Haslam et al., 2002; Cipollo et al., 2005; Natsuka et al., 2005). Ion compositions indicating paucimannose-type structures bearing only two hexose residues were also observed at low levels (Gn2Hex2 Fuc1Me and Gn2Hex2Fuc2Me), as well as glycans bearing two endogenous methyl groups (Gn2Hex4Fuc1HexMe2 and Gn2Hex5Fuc1Me2).
|
|
O-Linked glycans
Profiles and compositions following ß-elimination corresponded to the mucin-type O-glycans reported previously (Guerardel et al., 2001; Cipollo et al., 2004). All O-linked structures were confirmed by MSn following methylation to outline general topologies and to identify isomers; however, linkage details were not pursued and are assumed to be as published. Under the strong basic conditions of O-glycan elimination, high-mannose N-glycans were also observed in the profiles, corresponding to a previous observation (Guerardel et al., 2001).
Phosphatidylcholine
In native profiles, phosphatidylcholine conjugation was detected on six glycans occurring at or below baseline levels (Table II), confirming recent reports (Paschinger et al., 2004; Cipollo et al., 2005). For positive identification, each composition was verified to be a PC conjugate in MS2 and MS3 of the reduced unmethylated sample. This was accomplished by CID that resulted in a stepwise loss of PC-associated products, trimethylamine [-N(CH3)3, –59 amu] in MS2, followed by a cyclic ethylphosphoryl derivative (–124 amu) in MS3 (Friedl et al., 2003). MS4 disassembly of Hex3Hex NAc3PC showed the PC moiety was attached to a nonreducing terminal GlcNAc. MSn disassembly of Hex3Hex NAc3FucPC also indicated PC attached to a nonreducing terminal GlcNAc, whereas the single fucosyl residue was on the reducing terminal. Unfortunately, unmethylated samples do not provide adequate structural details to assign PC-GlcNAc branching and linkage information. Clearly, a prefractionation of these low-abundance materials in combination with methylation needs to be pursued for detailed MSn structural assignment. Characterization of the PC-GlcNAc residues associated with the glycolipids of the parasitic nematode Ascaris suum may be noteworthy, where the NMR spectra indicated a 6-linked PC-group on GlcNAc (Friedl et al., 2003).
|
Complexity at N-linked cores
Contrasting an enzymatic and hydrazine release, the relative abundances of Mans were similar and comparable. However, for a series of other ions of identical mass their abundance was markedly enhanced, and the spectra were different when compared by MS2. This was found to be the case with different release methods and a clear sign of multiple isomers (isobars). Facile glycosidic bond rupture between GlcNAcs of the chitobiose N-linked core dominated the MS2 spectra, producing B- and Y-type ions. Because the major spectral fragments were B-type ions, the neutral loss residues were reducing termini, (Figure 2A, m/z 1447 (-)GlcNAc, and m/z 1273 (-)GlcNAcFuc). Quite unexpected were neutral loss compositions that matched two fucose residues and even a hexose with GlcNAc (Figure 2A, m/z 1099 (-)GlcNAcFuc2, m/z 1069 (-)GlcNAcHexFuc, and (-)GlcNAcHexFuc2, m/z 895). Equally interesting was the observation that these neutral losses (reducing termini) were more abundant when using hydrazine (Figure 2B and D). This reinforced the early suspicion that hydrazine was able to release N-linked glycans with a higher degree of core substitution. To establish a direct structural relationship, the reducing terminal GlcNAc residue was specifically labeled by borohydride reduction converting it to an alditol before methylation. Profile analysis of these products showed that all ions exhibited the expected 16-Da increment of the sodium-adducted parent ion, and MS2 analysis provided the complementary B- and Y-ions as a consequence of the chitobiositol moiety rupture. A difference, however, was that all reducing termini could be identified by a respective 16 amu differential mass shift (cf. Figure 3A and B). Also obvious were a series of sodiated fragments (Y-ions) fitting the compositions GlcNAcFuc, GlcNAcFuc2, GlcNAcHexFuc, GlcNAcHexFuc2, and GlcNAcHex2Fuc2 (Figure 3A) and their alditol counterparts 16 Da higher in the NaBH4-reduced spectra, confirming their terminal position (Figure 3B). As expected, B ions would remain constant in both spectra, m/z 1650.8, 1476.7, 1302.6, 1272.6, 1098.5, and 894.4. These structural differences were subsequently unraveled with the combined efforts of a fragment library (Zhang et al., 2005), synthesis, and MSn. The resultant compositions at the reducing terminus could now be assigned (Table III), which provided an appreciation of the remarkable heterogeneity originating from only one portion of the structure (reducing terminus). Considering this core heterogeneity and when compounded by antennal glycomers, it is easy to see how misleading structural analysis could become when based on ion composition and anticipated biosynthetic data only. The core compositions do indicate core fucosylation precedes hexose addition (never a hexose without a fucose), and monomer addition (to the reducing GlcNAc) approaches a maximum of Hex2Fuc2 (observed on 11 of the 13 profiled ions). When contrasting glycan release strategies with hydrazine and PNGase A (Kubelka et al., 1993), both single and double fucosylated core structures were released. The present data, however, suggest that hydrazine may more efficiently release doubly fucosylated core structures and that a single hexose residue can significantly hinder enzymatic release (Figure 2). Additionally, the (Hex2Fuc2)-substituted core structures noted herein were only released with hydrazine, possibly indicating a total blockage of PNGase-A.
|
|
|
Interestingly, the abundance of core fucose-/hexose-substituted glycans in C. elegans is relatively high. Semiquantitative comparisons among the multiply fucosylated glycans released using the two methods indicate that hydrazinolysis consistently releases significantly higher levels than PNGase-A (Table I). The difference between the two release methods corresponds to differing levels of core hexose substitution, as in Figure 2. As an example, glycans corresponding to Gn2Hex5Fuc2 were typically observed at 
40% intensity (vs. Man3) in the hydrazine-released fractions and 10% in the PNGase-A-released fractions. Hydrazine-released glycans corresponding to Gn2Hex5Fuc2 included isomers ranging from unsubstituted to fully substituted (Hex2Fuc2) core GlcNAc (Table III). As indicated in Figure 3, the most abundant isomers observed in the MS2 spectrum of Gn2Hex5Fuc2 are fucose-/hexose-core substituted. In fact, proximal GlcNAc fucose/hexose substitution dominates the multiply fucosylated glycans (Table III), especially among the highest molecular weight structures that are exclusively hexose elaborated in C. elegans.
A determination of monomer composition by GC-MS of hydrazine-released glycans exhibited significant ratios of galactose in addition to mannose, GlcNAc, and fucose. This information, coupled with hexose residues being associated with the reducing terminus, further suggested an unusual N-linked core. In that regard, both octopus (Zhang et al., 1997) and squid rhodopsin (Takahashi et al., 2003) have been reported to possess N-glycans bearing a Galß(1-4)Fuc
(1-6) on the proximal GlcNAc. Hemocyanin from the keyhole limpet has also been reported to have comparable structures, and even with a further Gal extension, Galß(1-4)Galß(1-4)Fuc(1-6). These hemocyanin glycans could only be released in reasonable amounts by PNGase-F under denaturing conditions; however, linkages at the 3-GlcNAc position were not observed (Wuhrer et al., 2004).
Disassembly of N-linked cores
The earlier reports of N-linked core complexity provided insight to probe for comparable structures in C. elegans. Analysis involved recognition of facile glycosidic rupture intervals and oligomer termini, which were selected for further fragmentation to render diagnostic cross-ring cleavages by MSn (Reinhold et al., 1995, 1996; Sheeley and Reinhold, 1998; Ashline et al., 2005). As a starting point, the monofucosylated ion Gn2Hex3Fuc (m/z 1362, reduced and permethylated) in MSn was noted to be the Man3 homolog Gn2Man3Fuc, fucosylated only at the 6-position, as indicated by FragLib comparison of the MS3 spectrum against monofucosylated core fragments generated from ovalbumin N-glycans (Zhang et al., 2005). This observation confirmed recent findings regarding the order of fucosylation in C. elegans (Paschinger et al., 2005). With the consideration that core-GlcNAc fucosylation at position-6 precedes addition at position-3, MSn was then used to gather linkage and branching details for a novel glycan observed among the isomers noted in Figure 3A (m/z 1928.0, Hex5Gn2Fuc2). The structure for this glycan (Figure 4A) was proposed from data gathered following the steps of disassembly outlined in Figure 4B. In summary, the MS2 spectrum exhibited the described facile chitobiose rupture yielding B-type (m/z 1272.6) and Y-type ions (m/z 678.4). These fragments were known to be associated with the nonreducing and reducing termini from borohydride reduction and mass composition data (Hex4FucGn and HexFucGnol). Again, methylation was fundamental to this understanding, which positions terminal and internal residues. When the B-ion (m/z 1272.6) was analyzed by MS3, an unsubstituted internal GlcNAc was lost (–245 amu), leaving the trimannosyl portion of the core and its associated antennal extensions (Figure 4C). Isolation of this fragment (m/z 1027.5) for MS4 analysis provided a spectrum (Figure 4D) that confirmed two antennae by the respective loss of a terminal fucose, m/z 839, and a terminal hexose, m/z 809. This possibility was further supported by the loss of two terminal disaccharides, m/z 635 and m/z 605, (-)Fuc-Hex and (-)Hex-Hex, respectively, and further confirmed with B- and C-type ions corresponding to their compositions, m/z 415, 433 and m/z 445, 463. Further disassembly (MS5) was obtained on selected ions (Figure 4B) to corroborate each branch to the central mannose. In the latter case, the cross-ring cleavage ions m/z 503 and 475 were indicative of a 6-linked Fuc-Hex branch. Diminished signal intensity beyond MS5 prevented linkage position assignments for the branch-terminating hexose and fucose residues. Structural characterization of the reducing terminal Y-ion m/z 678.4 (Figure 4B) provided a spectrum (Figure 5A) from which the reducing-terminal fragment was deduced. The glycosidic cleavage fragments m/z 241/259 and m/z 415/433 define a Hex-Fuc sequence linked through the 6-position to GlcNAc as characterized by the ions, m/z 475 and 489 (Figure 5B). Double elimination and multiple cross-ring cleavages indicating a 6-linked Hex-Fuc disaccharide can be considered to fit the fragment ions m/z 573, 533, 505, 489, and 475 to fully define the glycan terminus. Thus, all major and minor fragment ions in the spectrum point to a 6-linked Hex-Fuc branch, corresponding to the order of core fucosylation in C. elegans.
|
|
Studies in other organisms (Zhang et al., 1997; Takahashi et al., 2003; Wuhrer et al., 2004) have shown the Hex-Fuc disaccharide core modification to possess a ß(1-4) inter-residue linkage, and a minor ion, m/z 299, supports this possibility (Figure 5A and B). To confirm this observation, the disaccharide B-ion m/z 415 was isolated for MS4 analysis (Figure 5C and D). The glycosidic fragments m/z 241/259 confirmed a terminal hexose and the base ion m/z 299 would be expected for a cross-ring cleavage fragment (3,5A-type) to a 4-linked fucose. To further characterize this reducing-terminal disaccharide (Hex-Fuc), the MS4 spectrum (Figure 5D) was compared to a series of five synthetic standards. The standards were a set of Gal-Fuc-O-benzyl disaccharides comprising different linkages and anomericities and were a generous gift of Dr Khushi Matta (Xue et al., 2004). To utilize these effectively, the samples were methylated and electrosprayed with 1-mmol lithium. MS2 spectra were taken to provide the corresponding Gal-Fuc(ene) B-type precursor ions (m/z 399, lithium adducted). This precursor was analyzed by MS3 for a comparative set of library spectra (Figure 6). The best spectral match with the C. elegans sample was observed to be Galß1-4Fuc (Figure 6B). It is important to note that these disaccharide spectra may not provide conclusive stereoisomer information on the monomers.
|
Monomer identification
Molecular disassembly (MSn) of small oligomers provides extensive structural understanding of linkage, sequence, anomericity and, when extended to monomers, even stereospecificity (Ashline et al., 2005). In this latter case, metal ions find differential binding stability in the paired electrons of equatorially and axially positioned methoxyl groups. From this binding specificity, terminal galactose residues can be distinguished from mannose or glucose based on a high-intensity marker ion (m/z 109) formed by CID of the precursor lithium-adducted B-ion (m/z 225). Using this approach, we confirmed the terminal hexose of the Hex-Fuc branch (Figures 4–
6) to be galactopyranose, indicating the reducing end to consist of Galß(1-4)Fuc(1-6)GlcNAc.
Galactosidase treatment
The above data, together with reports of similar structures in the keyhole limpet, octopus, and squid support the possibility of Galß(1-4)Fuc on C. elegans N-linked cores. To confirm this expectation, hydrazine-released glycans were treated with - and ß-galactosidases. Coffee bean -galactosidase produced no changes to MS profiles or the MSn spectra; however, treatment with different commercially available ß-galactosidases produced significant intensity changes among the highly fucosylated glycans. For example, results using the lactase Biolacta FN5 (Daiwa Kasei, Osaka, Japan) clearly showed a loss of galactose from core-substituted glycans. The control- and enzyme-treated samples were reduced, methylated, and submitted to MS2 analyses (Figure 7). The intensity ratios corresponding to the peaks (-)Gn, (-)GnFuc, and (-)GnFuc2 remain unchanged, whereas the ion peak, (-)GnHexFuc, showed a marked abundance decrease. Additionally, the GnHex2Fuc2 Y-ion (m/z 1072, Figure 7A) was no longer observed in the MS2 spectra (Figure 7B). Surprisingly, the enzymatic release of galactose was incomplete, with a relative abundance increase of a probable intermediate product, GnHexFuc2 (m/z 868, Y-ion, Figure 7B). This could be a consequence of slow kinetics, or the original ions were isomeric mixtures on which ß-galactosidase had selective hydrolytic influence. A comparative study of the ion that appeared to increase (m/z 868) indicated identical MS3 spectra both from Gn2Hex5Fuc2 (m/z 1944) and from hexose-incremented higher glycan Gn2Hex6Fuc2 (m/z 2148) before galactosidase treatment (Figure 7C and D). Sample treatment with ß-galactosidase and a comparable workup, however, provided the spectrum in Figure 7F, suggesting that structures galactosylated at the 6-position (Figure 7E) were largely replaced by structures hexose substituted at the 3-position (Figure 7G). Clearly, enzyme hydrolysis provided a greatly altered spectrum with significant intensity changes and considering the fragments proposed for each structure (Figure 7E and G), the spectral difference can be understood to be a loss of structure 7E resulting in a preponderance of structure 7G. Thus, the data suggest that ß-galactosidase hydrolyzes the Gal-Fuc disaccharide linked at the 6-position of the reducing terminal GlcNAc but not at the 3-linked position. These same changes were observed among all the glycans displaying the GnolHex2Fuc2 reducing terminus (m/z 1072 ion). The experiment was repeated using a nonspecific ß-galactosidase (G5160, Sigma-Aldrich, St. Louis, MO) and a positionally-specific ß(1-4)galactosidase (G0413, Sigma-Aldrich) yielding the same result.
|
Complete glycomer structure
A structural understanding of reducing termini (Y-type ions) and their glycomer distributions (Table III) opens the possibility of resolving their complement fragments (B-type) for a total assignment of structure. As an example, MS2 analysis on the glycan Gn2Hex4Fuc2 (m/z 1740) (Figure 8) provided the previously noted reducing-end Y-ion m/z 868, as well as the complementary nonreducing terminal B-ion (m/z 894, GnHex3). In MS3, m/z 868 gave a spectrum matching those in Figure 7C and D. The nonreducing-end fragment (m/z 894) was then observed to match the m/z 894 fragment obtained from the standard N-glycan GlcNAc2Man3 (Calbiochem, San Diego, CA), indicating a shared Man3 nonreducing-end moiety (Figure 8).
|
Enzymatically blocked termini
Eleven multiply fucosylated glycomers (Table III) were observed to include reducing ends, with the composition GnolHex2Fuc2 (m/z 1072 ion in MS2) structurally linking Gnol with Hex-Fuc disaccharides linked at both 3- and 6-positions. The 6-linked Galß(1-4)Fuc moiety was supported by MSn (Figures 5 and 7) and confirmed by a library match with synthetic standards (Figure 6). Even though MSn data suggested the 3-linked Hex-Fuc structure might be identical, different commercial preparations of ß-galactosidase consistently failed to fully release both hexose residues from reducing-end moieties, although the ion m/z 1072 was always completely removed from the MS2 spectra following ß-galactosidase treatment. Further examination of m/z 1072 fragments in MS3 also indicated that some reducing termini were composed of isobars as a consequence of variable capping of a second hexose residue (Figure 9). The ions, m/z 519, 589, 602, and 662, indicate a disaccharide (Hex-Fuc) linked to the 6-position of the terminal Gnol and the ions m/z 680 and 433 specify a comparable disaccharide at the 3-position (Figure 9B). An open hydroxyl (C-4) would be expected from MS2 rupture of the chitobiose moiety and this consideration was supported by the observed fragments. Complicating that interpretation, however, was the fragment ion m/z 866, accounted for by the neutral loss of a terminal fucosyl residue (–206 amu) (Figure 9A). Also suggestive of an alternative isobar was the fragment m/z 723, an increment of 204 amu over the Gnol-cleaving fragment m/z 519. These data and other ions support an isomer with an additional hexose pendant to the 6-linked disaccharide on the terminal Gnol (Figure 9C), suggesting a structure similar to the Galß(1-4)Galß(1-4)Fuc(1-6) motif previously reported in keyhole limpet hemocyanin (KLH; Wuhrer et al., 2004). However, although the (Hex2Fuc2)-substituted core GlcNAc Y-ion (m/z 1072) was consistently removed from the MS2 spectra following treatments with ß-galactosidases, including those that are ß1-4 specific (Sigma-Aldrich G0413), low abundance precluded confident linkage assignments for this Hex-Hex-Fuc moiety using MSn.
|
Nevertheless, these findings do not explain the inability to completely remove hexose residues from Gnol (Hex-Fuc)-disubstituted cores (m/z 1072), so an MSn experiment was designed to specifically isolate and characterize the 3-linked Gnol component of this structure. This was initiated by hydrazine release and lactase treatment, followed by reduction and methylation. In the example shown, the MS2 spectrum confirmed removal of ß(1-4)galactose from the 6-linked fucose of the reducing terminal Gnol of GnGnolHex5Fuc2 (m/z 1944, cf. Figure 7G). Earlier studies had demonstrated that stereochemistry-revealing spectral matches were more exacting with Li+ adduction (Ashline et al., 2005). Thus, the parent ion now showing the expected lithium-adducted 16-amu downshift from sodium adduction (m/z 1928, GnGnolHex5Fuc4+Li+) was fragmented to provide the corresponding lactase-treated and lithium-adducted reducing-end Y-ion (m/z 852), Gnol HexFuc2 (Figure 10A). From this spectrum, the C-type disaccharide ion (m/z 417) was isolated and fragmented in MS4 (Figure 10B). The resulting spectrum was observed to be identical to the (m/z 417) C-ion MS4 spectrum obtained from the reduced lithium-adducted moiety: Galß(1-4) Fuc(1-6)Gnol from the glycan in Figures 4–6 (Figure 10C). Next, the monomer B-ion (m/z 225) from the 3-linked Hex-Fuc was fragmented in MS5 using the pathway m/z 852
417225 to give a spectrum (Figure 10D) showing the prominent lithium-adducting marker ion (m/z 109) that also had been observed for 6-linked Galß(1-4)Fuc. These results indicate a 3-linked fucose terminated by galactopyranose, and together the spectral comparisons suggest Galß(1-4)Fuc occupying both the 3- and 6-position of the reducing-end Gn residue. Many fucosylated N-glycans that could only be obtained by hydrazinolysis shared this reducing-end motif. The lowest molecular weight example is a glycan with the Man3 reducing-end motif (cf. Figure 8B). Interestingly, glycans bearing this motif released ß-galactose only from 6-position core fucose, suggesting, ß-galactosidase resistance at the 3-position.
|
Penultimate GlcNAc substitution
The initial findings of galactose pendant on core-fucosyl moieties of C. elegans N-glycans suggested some identity with core-fucosylation patterns in the parasitic nematode Haemonchus contortus. In that study, PNGase-A- and PNGase-F-released glycans exhibited both core 3- and 6-GlcNAc fucosylation, as well as 3-linked substitution of the penultimate GlcNAc residue. No galactose capping of these fucosyl residues was reported (Haslam et al., 1996). This may be the result of complex CCM blocking enzymatic release. With penultimate fucosylation a possibility, a specific focus was initiated to ascertain this possibility in C. elegans. The precursor ion corresponding to Gn2Hex6Fuc4 (m/z 2496.3, reduced and methylated, Table III) was disassembled, providing a prominent MS2 chitobiose cleavage and complementary B-/Y-ion pairs: the respective m/z 1650 and m/z 868 fragments (Figure 11). The Y-type ion, m/z 868, upon further disassembly provided a spectrum identical to Figure 7C and D, and the nonreducing-end B-ion (m/z 1650.5) identified a new fragment ion (m/z 646), corresponding in composition to a putative internal GlcNAc substituted with a Hex-Fuc branch (Figure 11C and D). A series of confirming ions in the MS3 spectrum, m/z 1650.5 (Figure 11B), indicate a branching pattern similar to the nonreducing end of the glycan in Figure 4. Further evidence for internal-GlcNAc substitution was a linkage indicating 3,5A-type ion (m/z 1115) locating a Hex-Fuc disaccharide to the 3-position of the penultimate GlcNAc. MS4 of m/z 646 (Figure 11D) showed the neutral loss of GlcNAc, providing a single C-type ion m/z 433, a terminal Hex-Fuc moiety. The product ion m/z 433 following CID of m/z 646 could not be detailed further by MSn because of low signal intensity.
|
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
A glycome mass profile seeks to present all glycans in a manner suitable for qualitative and quantitative analyses. Differential profiles impart functional relationships and disassembly of selected peaks supplies detailed structural understanding. Modulated profiles, as with mutants and glycosyltransferase gene knockouts, can provide a chemical connection even in the absence of visual phenotypes (Cipollo et al., 2004
; Paschinger et al., 2004; Zhu et al., 2004). Functional conclusions must assume all the relevant glycoconjugates were present in the extract and quantitatively reflected in MS profiles. Quantitative glycan release and capture techniques are not always easy to assess and are rarely supported with internal standards or orthogonal strategies. Chemical techniques can be tested with known samples, but biological approaches are particularly defenseless because the full range of function is usually unknown. This is particularly the case for commonly used lectins and endoglycosidases, and much of the "complexity and controversy" that has been summarized (Haslam et al., 2003) may be attributed to these factors.
The aim of this investigation was to present an improved understanding of the most abundant class of N-glycans in C. elegans—a range of CCMs that dominated the hydrazine-release profiles. These glycans have been largely missed in a number of previous reports, which appears related to hindered endoglycosidase activity. All other glycans appeared to be as described. Using a chemical release, the complex and PC-substituted glycans are not prominent by relative abundance measures. The latter could not be observed above baseline levels; nevertheless, by selecting the appropriate mass in an MS profile, these could be trapped and disassembled to demonstrate their presence. Although truncated complex glycans do appear in these MS profiles, evidence for hybrid or chain-extended complex glycans is not observed. These results are also comparable to previous findings that PNGase-A freely releases N-glycans that are not substituted or substituted with (1-3) and/or (1-6) core fucose residues at the proximal GlcNAc (Kubelka et al., 1993). In contrast, PNGase-F is completely hindered by (1-3) fucose (Tretter et al., 1991). Additionally, whereas PNGase-F freely releases 6-linked core fucose structures, 6-linked Gal(ß1-4) extensions have been noted to require more stringent denaturing conditions for release (Wuhrer et al., 2004). The present data further suggest that galactosyl extensions on core difucosylated structures may impose limitations on PNGase-A activity, and 3-/6-linked Gal(ß1-4)Fuc disubstitutions apparently block the enzyme completely. Interestingly, these data also suggest that hydrazine may improve the release of core difucosylated N-glycans. MSn disassembly additionally indicated Hex-Fuc substitutions on penultimate GlcNAc residues in selected cases; however, this motif could not be confidently assigned because of low ion abundance in MSn, and effects of penultimate-GlcNAc substitution on endoglycosidase release will require further investigation.
These findings may be significant in light of previous studies where biological insights have been pursued within the context of glycosylation. Apparent difficulties with endoglycosidase release of core-substituted N-glycans have been overlooked, possibly limiting relationships to function. The MSn spectra presented here indicate galactose pendant to proximal fucose residues among C. elegans N-glycans. These results are also comparable to recent reports of the same substitutions among other invertebrates. The N-glycans released from squid and octopus rhodopsin and KLH include such moieties. While the motif Galß(1-4) Galß(1-4)Fuc(1-6) observed in KLH is hinted at in our MSn spectra, the Hex-Hex-Fuc(1-6)Gnol motif noted in C. elegans will require further investigation because of a low ion abundance in MSn. MSn was used to confidently define the high-abundance proximal-GlcNAc substitution patterns, and ß-galactosidase treatments confirmed the presence of Gal(ß1-4)Fuc(1-6)GlcNAc. On the contrary, although employing a variety of commercially available galactosidases ranging from general ( and ß) to specific (ß1-4), in all cases null results were obtained for the 3-linked Hex-Fuc motif. Coffee bean -galactosidase produced no changes in the spectra, whereas ß-galactosidases released galactose only from proximal (1-6) fucose. By employing lithium adduction and a spectrum-matching approach (Zhang et al., 2005) following ß-galactosidase treatment, however, it was noted that the 3-linked motif gave an identical MS4 spectrum to MS4 from the enzyme labile 6-linked Gal(ß1-4)Fuc moiety before enzyme treatment. MS5 clearly confirmed that both the 3- and 6-linked Hex-Fuc moieties are capped by galactopyranose. Recognizing that enzyme activities toward core-elaborating galactosyl linkages may be unclear, further studies employing different exoglycosidase-sequencing strategies may be required, whereas the MSn data appear conclusive.
Ion trap (IT) mass spectrometry was fundamental for the identification of this CCM family, and when used in conjunction with methylation, these strategies provided a clear identification of multiple isobaric structures. An additional outcome was a direct understanding of glycomer distributions without the time-consuming efforts of chromatography, which at its best, would be exceedingly challenged with this glycome effort. All samples were handled by direct infusion for electrospray ionization-MSn (ESI-MSn) or laser ablation for matrix-assisted laser desorption ionization (MALDI)-MSn applications. O-Glycans could not initially be ruled out; however, these fucosylated structures did not appear in the MS profiles of base-released glycans. Interestingly, Man were observed at low abundance under the basic conditions applied to release O-linked structures. Galactosylation of proximal fucose residues also explained the abundance of galactose and the paucity of GalNAc observed by monosaccharide composition analysis of hydrazine-released glycan pools. Additionally, the protein pellets were washed with 10:10:1 chloroform : methanol : water to avoid possible cross-contamination with glycosphingolipids; nevertheless, the MSn fragmentation spectra were not consistent with the glycosphingolipid structures described for C. elegans (Griffitts et al., 2005). Most convincing, the MSn fragment patterns outlined herein are highly consistent with those we have noted for other N-glycans; in particular, the products obtained in MS2–3 reflect well-defined fragments involving GlcNAcs of the chitobiose core.
Borohydride reduction and MS2 core chitobiose rupture provided the gas-phase separation of three glycomer arrays; composites contributed by multiple antenna, reducing and penultimate GlcNAc residues. A selection of those was compiled in Table III. These data expose an additional point: that molecular glycosylation from composite cells and tissues may be best represented by an array of structures, including isobars as well as glycomers. The symbol of a single structure defining such ion peaks may miss the more accurate representation that depicts the overall plasticity of the process.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
Protein preparation
Mixed-stage C. elegans cultures were grown and harvested as described (Sulston and Hodgkin, 1988
). Worms sufficient to yield 10–15 mg of protein by Bradford assay were suspended in 1.5 mL of CHAPS lysis buffer (4% CHAPS, 8 M urea, 65 mM DTT, 35 mM Tris, pH 8.0), and homogenized by bead beating at room temperature using four 20-second cycles, with cooling on ice between each cycle. The lysates were then centrifuged at 12,000 rcf for 15 min and transferred to Falcon tubes. A second aliquot of lysis buffer was then added, the procedure was repeated, and the supernatants were combined.
Glycan sample preparation
For hydrazinolysis, the lysate was transferred to a glass culture tube (13 x 100 mm) and the protein precipitated using 15% trichloroacetic acid (TCA) on ice for 1 h, then centrifuged at low speed for 15 minutes. After discarding the supernatant, the protein pellet was washed and vortexed with 2 x 1 mL cold acetone, and 2 x 1 mL chloroform : methanol : water (10:10:1), with centrifugation between each addition. The protein pellet was briefly dried under nitrogen, lyophilized overnight, and stored at –20°C if not used immediately. Before hydrazine release, the protein pellet was vacuum-dried over P2O5 for 48 h. Hydrazinolysis was carried out at 100°C for 8 h using 500 µL of anhydrous hydrazine (Sigma-Aldrich). After cooling, the hydrazine was removed under nitrogen, followed by lyophilization overnight. The glycans were re-N acetylated using acetic anhydride (150 µL) in saturated sodium bicarbonate (300 µL) for 30 min on ice, followed by 1 hour at room temperature. The sample was desalted on AG50, and the glycans were purified using a hand-packed column containing 1 mL of medium fibrous cellulose. The column was washed with 4:1:1 n-butanol : MeOH : water to remove protein contaminants, and the glycans were eluted with 1:1 ethanol : water. In selected cases, glycans were reduced with sodium borohydride (300 µL of 15 mg/mL NaBH4 in 100 mM NaOH) initially on ice, followed by room temperature overnight. To remove the borate, acetic acid was added dropwise until pH 5, followed by coevaporation with 3 mL of ethanol, 2 x 2 mL of 1% acetic acid in methanol, and 2 x 2 mL of toluene. The reduced glycans were dissolved in water and desalted on a column packed with 200 µL of graphitized carbon slurry (Carbograph 120/400, Alltech Associates, Deerfield, IL). The column was washed with water, and glycans were eluted using 3 mL of 25% acetonitrile/water. Including the reduction step clarified MSn fragmentation analysis by clearly distinguishing the glycan reducing ends and was also noted to improve the signal-to-noise ratio for MALDI-time-of-flight (TOF) analysis. For N-glycan release using PNGase-A (Calbiochem), a volume of CHAPS lysate containing 1.0-mg protein (by Bradford assay) was placed in an Eppendorf tube, TCA precipitated, washed and lyophilized as above. The protein pellet was resuspended in 100 µL of 400 mM ammonium bicarbonate buffer containing 8 M Urea (pH 8.5). The protein was reduced by adding 25 µL of 45-mM DTT for 15 min at 50°C and alkylated by adding 25 µL of 100 mM IAA at room temperature for 15 min. The sample was diluted to 400 µL, and 50 µg of trypsin (Sigma, St. Louis, MO, type IX-S) was added. The mixture was heated to 50°C for 60 min, whereupon clearing of the solution was observed. A 20 µg aliquot of sequencing grade trypsin (Sigma) was then added for overnight digestion at 37°C. Trypsin was deactivated by heat treatment (100°C for 15 min). After cooling on ice, the solution volume was adjusted to 800 µL with water, and the pH was adjusted to 5.5 with acetic acid. PNGase-A (2.5 mU) was added to the peptide mixture and incubated at 37°C for 48 h with occasional mixing. The sample was dried in a Speedvac and taken up in a solvent consisting of 95% water, 5% ACN, and 0.1% TFA. Peptides were removed by passing through a C18 Sep-Pak, and the flow-through fraction was desalted on graphitized carbon. N-Glycans were also released with PNGase-F using the manufacturer’s protocol (New England Biolabs, Ipswich, MA); detergents were removed by C18 Sep-Pak, and the glycans were desalted on graphitized carbon as described above. Reductive beta elimination (Carlson, 1966) and cleanup of O-glycans was performed as described elsewhere (Karlsson and Packer, 2002). Reduced and nonreduced glycans were methylated as described (Ciucanu and Kerek, 1984; Ciucanu and Costello, 2003). Galactosidase treatments were performed using aliquots of 10 µg of glycans. Treatment with -galactosidase (Sigma-Aldrich G8507) was carried out using 80 mU of enzyme and raffinose as a positive control. Treatment with a nonspecific ß-galactosidase (G5160, Sigma-Aldrich) used 90 mU of enzyme and O-nitrophenyl-ß-D-galactopyranoside (ONPG) as a positive control. Lactase treatment (Biolacta FN5, Daiwa Kasei) used 80 mU of enzyme and ONPG as the positive control. Additionally, a positionally specific ß1-4 galactosidase (G0413, Sigma-Aldrich) was used with 1 µg of an enriched fraction of CCM glycans purified by column chromatography. In the latter case, 9 mU of enzyme was added and ONPG used as a positive control. For all enzyme treatments, both the treated sample and an identical negative control were worked up in tandem using the manufacturer’s recommended conditions. To obtain an enriched fraction of CCM glycans, graphitized carbon column chromatography was carried out using an 8 x 100 mm glass column hand-packed with 4 mL of Carbograph slurry.
Glycan analysis
MS profiles were obtained by reflectron-mode MALDI-TOF (Axima-CFR, Kratos-Shimadzu Biotech, Manchester, UK) using dihydroxybenzoic acid (DHB, 10 mg/mL) as the matrix and external calibration using peptide standards (Proteomass MALDI-MS calibration kit, Sigma). Internal calibration was performed as needed using the Man present in glycan pools. Additionally, MS profiles were obtained by ESI, carried out by direct infusion using a linear IT mass spectrometer equipped with a nanoflow source (ThermoFinnigan LTQ, San Jose, CA). MSn was also performed by MALDI-QIT-TOF (Axima-QIT, Kratos-Shimadzu Biotech) with DHB as the matrix. The ethanol recrystallization of MALDI spots was noted to improve the signal-to-noise ratio for MALDI-QIT-TOF analysis. For MSn, ESI-IT was preferred for deep-pathway fragmentation spectra (i.e. MS4–6); MALDI-QIT-TOF was preferred for use with small sample amounts and where fragmentation pathways beyond MS3 were not required. All recorded peaks were sodium-adducted or lithium-adducted monoisotopic mass ions. Generation of sodium adducts did not require any additional sample handling techniques; however, for electrospray of lithium-adducted glycans the samples were dissolved in a solution of 90% MeOH/10% water containing 1-mM lithium acetate. Lithium adduction improved signal-to-noise ratio and provided better precision where both m/z and intensity are used for spectral–library matching. The spectra were submitted for analysis using the in-house-developed bioinformatics tools: FragLib (Zhang et al., 2005) and OSCAR (Lapadula et al., 2005). Monosaccharide composition analysis was carried out by methanolysis (Reinhold, 1972), followed by electron impact GC-MS (Thermo Finnigan, GCQ, San Jose, CA) using a DB-5, 30 m x 0.25 mm x 25 micron column (J & W Scientific, Folsom, CA).
|
Conflict of interest statement |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
None declared.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
We are grateful to Hailong Zhang for development of a glycan fragment database and to Tony Lapadula and Dr. Suddham Singh for their assistance with MSn fragmentation analysis. We thank Prof. Khushi Matta, Roswell Park Cancer Institute, Buffalo, NY 14263, for providing the Gal-Fuc-O-benzyl standards. We also acknowledge Cristina Silvescu and Heidi Geiser for assistance with some experiments. This study was initiated in collaboration with the late Prof. Charles Warren (Biochemistry and Molecular Biology, UNH) and with Harry Schachter (Hospital for Sick Children, University of Toronto), and we thank them for their guidance and encouragement. The work was supported by NIH grants GM 54045 (V.N.R.) and NCRR-BRIN grant RP016459 (V.N.R.).
|
Footnotes |
|---|
Dedicated to Professor Charles Edward Warren, 1963–2005.
|
Abbreviations |
|---|
CCM, core chitobiose modification; CID, collision induced dissociation; ESI, electrospray ionization; Gn, 2-acetamidoglucopyranose; Gnol, 2-acetamidoglucositol; GnT, glucosaminyltransferase; IT, ion trap; KLH, keyhole limpet hemocyanin; MALDI, matrix-assisted laser desorption ionization; Man, high-mannose glycan; MSn, sequential mass spectrometry; ONPG, O-nitrophenyl-ß-D-galactopyranoside; PC, phosphorylcholine.
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
Altmann, F., Fabini, G., Ahorn, H., and Wilson, I.B.H. (2001) Genetic model organisms in the study of N-glycans. Biochimie, 83, 703–712.[Medline]
Ashline, D., Singh, S., Hanneman, A., and Reinhold, V. (2005) Congruent strategies for carbohydrate sequencing I. Mining structural details by MSn. Anal. Chem., 77, 6250–6262.[Medline]
Carlson, D. (1966) Oligosaccharides isolated from pig submaxillary mucin. J. Biol. Chem., 241, 2984–2986.
Chen, S., Tan, J., Reinhold, V., Spence, A., and Schachter, H. (2002) UDP-N-acetylglucosamine:alpha-3-D-mannoside beta-1,2-N-acetylglucosaminyltransferase I and UDP-N-acetylglucosamine:alpha-6-D-mannoside beta-1,2-N-acetylglucosaminyltransferase II in Caenorhabditis elegans. Biochim. Biophys. Acta, 1573, 271–279.[Medline]
Cipollo, J.F., Awad, A.M., Costello, C.E., and Hirshberg, C.B. (2004) srf-3, a mutant of Caenorhabditis elegans, resistant to bacterial infection and to biofilm binding, is deficient in glycoconjugates. J. Biol. Chem., 279, 52893–52903.
Cipollo, J.F., Awad, A.M., Costello, C.E., and Hirschberg, C.B. (2005) N-glycans of Caenorhabditis elegans are specific to developmental stages. J. Biol. Chem., 280, 26063–26072.
Cipollo, J.F., Costello, C.E., and Hirschberg, C.B. (2002) The fine structure of Caenorhabditis elegans N-glycans. J. Biol. Chem., 277, 49143–49157.
Ciucanu, I. and Costello, C.E. (2003) Elimination of oxidative degradation during the per-O-methylation of carbohydrates. J. Am. Chem. Soc., 125, 16213–16219.[CrossRef][Web of Science][Medline]
Ciucanu, I. and Kerek, F. (1984) A simple and rapid method for the permethylation of carbohydrates. Carbohydr. Res., 131, 209–216.[CrossRef]
DeBose-Boyd, R.A., Nyame, A.K., and Cummings, R.D. (1998) Molecular cloning and characterization of an 1,3 fucosylransferase, CEFT-1, from Caenorhabditis elegans. Glycobiology, 8, 905–917.
Friedl, C.H., Lochnit, G., Zahringer, U., Bahr, U., and Geyer, R. (2003) Structural elucidation of zwitterionic carbohydrates derived from glycosphingolipids of the porcine parasitic nematode Ascaris suum. Biochem. J., 369, 89–102.[CrossRef][Web of Science][Medline]
Griffitts, J.S., Haslam, S.M., Yang, T., Garczynski, S.F., Mulloy, B., Morris, H., Cremer, P.S., Dell, A., Adang, M.J., and Aroian, R.V. (2005) Glycolipids as receptors for Bacillus thuringiensis crystal toxin. Science, 307, 922–925.
Guerardel, Y., Balanzo, L., Maes, E., Leroy, Y., Coddeville, B., Oriol, R., and Strecker, G. (2001) The nematode Caenorhabditis elegans synthesizes unusual O-linked glycans: identification of glucose-substituted mucin-type O-glycans and short chondroitin-like oligosaccharides. Biochem. J., 357, 167–182.[CrossRef][Web of Science][Medline]
Hanneman, A. and Reinhold, V.N. (2003) Abundant and unusual N-glycans from the eukaryote C. elegans. Glycobiology, 13, 889.
Haslam, S.M., Coles, G.C., Munn, E.A., Smith, T.S., Smith, H.F., Morris, H.R., and Dell, A. (1996) Haemonchus contortus glycoproteins contain n-linked oligosaccharides with novel highly fucosylated core structures. J. Biol. Chem., 271, 30561–30570.
Haslam, S.M. and Dell, A. (2003) Hallmarks of Caenorhabditis elegans N-glycosylation: complexity and controversy. Biochimie, 85, 25–32.[Medline]
Haslam, S.M., Gems, D., Morris, H.R., and Dell, A. (2002) The glycomes of Caenorhabditis elegans and other model organisms. Biochem. Soc. Symp., 69, 117–134.
Hirabayashi, J., Hayama, K., Kaji, H., Isobe, T., and Kasai, K. (2002) Affinity capturing and gene assignment of soluble glycoproteins produced by the nematode Caenorhabditis elegans. J. Biochem. (Tokyo), 132, 103–114.
Karlsson, N.G. and Packer, N.H. (2002) Analysis of O-linked reducing oligosaccharides released by an in-line flow system. Anal. Biochem., 305, 173–185.[CrossRef][Web of Science][Medline]
Kubelka, V., Altmann, F., Staudacher, E., Tretter, V., März, L., Hård, K., Kamerling, J.P., and Vliegenthart, J.F.G. (1993) Primary structures of the N-linked carbohydrate chains from honeybee venom phospholipase A2. Eur. J. Biochem., 213, 1193–1204.[Web of Science][Medline]
Lapadula, A.J., Hatcher, P.J., Hanneman, A.J., Ashline, D.J., Zhang, H., and Reinhold, V. (2005) Congruent strategies for carbohydrate sequencing III. OSCAR: an algorithm for assigning oligosaccharide topology from MSn data. Anal. Chem., 77, 6271–6279.
Natsuka, S., Adachi, J., Kawaguchi, M., Nakakita, S., Hase, S., and Ikura, K. (2002) Structural analysis of N-linked glycans in Caenorhabditis elegans. J. Biochem. (Tokyo), 131, 807–813.
Natsuka, S., Kawaguchi, M., Wada, Y., Ichikawa, A., Ikura, K., and Hase, S. (2005) Characterization of wheat germ agglutinin ligand on soluble glycoproteins in Caenorhabditis elegans. J. Biochem. (Tokyo), 138, 209–213.
Paschinger, K., Dubravko, R., Lochnit, G., Jantsch, V., and Wilson, I.B.H. (2004) Molecular basis of anti-horseradish peroxidase staining in Caenorhabditis elegans. J. Biol. Chem., 279, 49588–49598.
Paschinger, K., Staudacher, E., Stemmer, U., Fabini, G., and Wilson, I.B.H. (2005) Fucosyltransferase substrate specificity and the order of fucosylation in invertebrates. Glycobiology, 15, 463–474.
Reinhold, V. (1972) Gas liquid chromatographic analysis of constituent carbohydrates in glycoproteins. Methods Enzymol., 25, 244–249.
Reinhold, V., Reinhold, B., and Chan, S. (1996) Carbohydrate sequence analysis by electrospray ionization-mass spectrometry. Methods Enzymol., 271, 377–402.[Web of Science][Medline]
Reinhold, V., Reinhold, B., and Costello, C. (1995) Carbohydrate molecular weight profiling, sequence, linkage, and branching data: ES-MS and CID. Anal. Chem., 67, 1772–1784.[Medline]
Schachter, H. (2004) Protein glycosylation lessons from Caenorhabditis elegans. Curr. Opin. Struct. Biol., 14, 607–616.[CrossRef][Web of Science][Medline]
Schachter, H., Chen, S., Zhang, W., Spence, A.M., Zhu, S., Callahan, J.W., Mahuran, I., Fan, X., Bagshaw, R.D., She, Y., and others. (2002) Functional post-translational proteomics approach to study the role of N-glycans in the development of Caenorhabditis elegans. Biochem. Soc. Symp., 69, 1–21.
Schachter, H., Shi, H., and Spence, A.M. (2005) N-Acetylglucosaminyltransferase I-dependent N-glycans are involved in the response of Caenorhabditis elegans to bacterial pathogens. Glycobiology, 15, 1191.
Sheeley, D. and Reinhold, V. (1998) Structural characterization of carbohydrate sequence, linkage, and branching in a quadrupole ion trap mass spectrometer: neutral oligosaccharides and N-linked glycans. Anal. Chem., 70, 3053–3059.[Medline]
Sulston, J. and Hodgkin, J. (1988) Methods. In Wood, W.B. (ed.), The Nematode Caenorhabditis elegans. Cold Springs Harbor Laboratory Press, Cold Springs Harbor, New York, pp. 587–606.
Takahashi, N., Masuda, K., Hiraki, K., Yoshihara, K., Huang, H., Khoo, H., and Kato, K. (2003) N-Glycan structures of squid rhodopsin. Existence of the 1–3 and 1–6 difucosylated innermost GlcNAc residue in a molluscan glycoprotein. Eur. J. Biochem., 270, 2627–2632.[Web of Science][Medline]
Tretter, V., Altmann, F., and März, L. (1991) Peptide-N4 (N-acetyl-ß-glucosaminyl) asparagines amidase F cannot release glycans with fucose attached 13 to the asparagines-linked N-acetylglucosamine residue. Eur. J. Biochem., 199, 647–652.[Web of Science][Medline]
Warren, C.E., Krizus, A., Roy, P.J., Culotti, J.G., and Dennis, J.W. (2002) The Caenorhabditis elegans gene, gly-2, can rescue the N-acetylglucosaminyltransferase V mutation of Lec4 cells. J. Biol. Chem., 277, 22829–22838.
Wood, W.B. (1988) Introduction to C. elegans biology. In Wood, W.B. (ed.), The Nematode Caenorhabditis elegans. Cold Springs Harbor Laboratory Press, Cold Springs Harbor, New York, pp. 587–606.
Wuhrer, M., Robijn, M.L.M., Koelman, C.A.M., Balog, C.I.A., Geyer, R., Deedler, A.M., and Hokke, C.H. (2004) A novel Gal (ß1–4)Gal(ß1–4)Fuc(1–6)-core modification attached to the proximal N-acetylglucosamine of keyhole limpet haemocyanin (KLH) N-glycans. Biochem. J., 378, 625–632.[CrossRef][Web of Science][Medline]
Xue, J., Song, L., Khaja, S.D., Locke, R.D., West, C.W., Laine, R.L., and Matta, K.L. (2004) Determination of linkage position and anomeric configuration in Hex-Fuc disaccharides using electrospray ionization tandem mass spectrometry. Rapid Commun. Mass Spectrom., 18, 1947–1955.[CrossRef][Web of Science][Medline]
Zhang, Y., Iwasa, T., Tsuda, M., Kobata, A., and Takasaki, S. (1997) A novel monoantennary complex-type sugar chain found in octopus rhodopsin: occurrence of the Galß4Fuc group linked to the proximal N-acetylglucosamine residue of the trimannosyl core. Glycobiology, 7, 1153–1158.
Zhang, H., Singh, S., and Reinhold, V. (2005) Congruent strategies for carbohydrate sequencing II. FragLib: an MSn spectral library. Anal. Chem., 77, 6263–6270.
Zheng, Q., Van Die, I., and Cummings, R.D. (2002) Molecular cloning and characterization of a novel 1,2-fucosyltransferase (CE2FT-1) from Caenorhabditis elegans. J. Biol. Chem., 277, 39823–39832.
Zhu, S., Hanneman, A., Reinhold, V., Spence, A., and Schachter, H. (2004) Caenorhabditis elegans triple null mutant lacking UDP-N-acetyl-D-glucosamine: -3-D-mannoside ß1,2-N-acetylglucosaminyltransferase I. Biochem. J., 382, 995–1001.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. Takeuchi, K. Nishiyama, K.-i. Sugiura, M. Takahashi, A. Yamada, S. Kobayashi, H. Takahashi, H. Natsugari, and K.-i. Kasai Caenorhabditis elegans galectins LEC-6 and LEC-1 recognize a chemically synthesized Gal{beta}1-4Fuc disaccharide unit which is present in Protostomia glycoconjugates Glycobiology, December 1, 2009; 19(12): 1503 - 1510. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. G. T. Hagen, L. Zhang, E Tian, and Y. Zhang Glycobiology on the fly: Developmental and mechanistic insights from Drosophila Glycobiology, February 1, 2009; 19(2): 102 - 111. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Takeuchi, K. Hayama, J. Hirabayashi, and K.-i. Kasai Caenorhabditis elegans N-glycans containing a Gal-Fuc disaccharide unit linked to the innermost GlcNAc residue are recognized by C. elegans galectin LEC-6 Glycobiology, November 1, 2008; 18(11): 882 - 890. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Prien, L. C. Huysentruyt, D. J. Ashline, A. J. Lapadula, T. N. Seyfried, and V. N. Reinhold Differentiating N-linked glycan structural isomers in metastatic and nonmetastatic tumor cells using sequential mass spectrometry Glycobiology, May 1, 2008; 18(5): 353 - 366. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Gutternigg, D. Kretschmer-Lubich, K. Paschinger, D. Rendic, J. Hader, P. Geier, R. Ranftl, V. Jantsch, G. Lochnit, and I. B. H. Wilson Biosynthesis of Truncated N-Linked Oligosaccharides Results from Non-orthologous Hexosaminidase-mediated Mechanisms in Nematodes, Plants, and Insects J. Biol. Chem., September 21, 2007; 282(38): 27825 - 27840. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Aoki, M. Perlman, J.-M. Lim, R. Cantu, L. Wells, and M. Tiemeyer Dynamic Developmental Elaboration of N-Linked Glycan Complexity in the Drosophila melanogaster Embryo J. Biol. Chem., March 23, 2007; 282(12): 9127 - 9142. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Paschinger, M. Hackl, M. Gutternigg, D. Kretschmer-Lubich, U. Stemmer, V. Jantsch, G. Lochnit, and I. B. H. Wilson A Deletion in the Golgi {alpha}-Mannosidase II Gene of Caenorhabditis elegans Results in Unexpected Non-wild-type N-Glycan Structures J. Biol. Chem., September 22, 2006; 281(38): 28265 - 28277. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||