Skip Navigation


Glycobiology Advance Access originally published online on August 30, 2007
Glycobiology 2007 17(12):1333-1343; doi:10.1093/glycob/cwm090
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
17/12/1333    most recent
cwm090v2
cwm090v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (4)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Bernatchez, S.
Right arrow Articles by Wakarchuk, W. W
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bernatchez, S.
Right arrow Articles by Wakarchuk, W. W
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Published by Oxford University Press 2007.

Variants of the ß1,3-Galactosyltransferase CgtB from the Bacterium Campylobacter Jejuni have Distinct Acceptor Specificities

Stéphane Bernatchez2, Michel Gilbert2, Marie-Claude Blanchard2, Marie-France Karwaski2, Jianjun Li2, Shawn DeFrees3 and Warren W Wakarchuk1,2

2 Institute for Biological Sciences, National Research Council Canada, 100 Sussex Drive, Ottawa, Ontario K1A 0R6, Canada
3 NEOSE Technologies, Inc. 102 Witmer Road, Horsham, PA 19044, USA

1 To whom correspondence should be addressed: Tel. +613 952 4299; Fax +613 952 9092; e-mail: warren.wakarchuk{at}nrc-cnrc.gc.ca

Received on May 17, 2007; revised on July 23, 2007; accepted on August 23, 2007


   Abstract
Top
 Abstract
Introduction
 Results
 Discussion
 Materials and methods
 Supplementary data
 Conflict of interest statement
 References
 
The gene clusters encoding the lipooligosaccharide biosynthesis glycosyltransferases from Campylobacter jejuni have previously been divided in eight classes based on their genetic organization. Here, three variants of the ß1,3-galactosyltransferase CgtB from two classes were purified as fusions with the maltose-binding protein (MalE) from Escherichia coli and their acceptor preference was determined. The acceptor preference of each CgtB variant was directly related to the presence or absence of sialic acid in the acceptor, which correlated with the core oligosaccharide structure in vivo. The three variants were evaluated for their ability to use a derivitized monosaccharide, a GM2 ganglioside mimic, a GA2 ganglioside mimic as well as a peptide containing {alpha}-linked GalNAc. This characterization shows the flexibility of these galactosyltransferases for diverse acceptors. The CgtB variants were engineered via carboxy-terminal deletions and inversion of the gene fusion order. The combination of a 20 to 30 aa deletion in CgtB followed by MalE at its carboxy terminus significantly improved the glycosyltransferase activity (up to a 51.8-fold increase of activity compared to the full length enzyme) in all cases regardless of the acceptor tested. The improved enzyme CgtBOH4384{Delta}C-MalE was used to galactosylate a glyco-peptide acceptor based on the interferon {alpha}2b protein O-linked glycosylation site as confirmed by the CE-MS analysis of the reaction products. This improved enzyme was also used successfully to galactosylate the human therapeutic protein IFN{alpha}2b[GalNAc{alpha}]. This constitutes the first report of the in vitro synthesis of the O-linked T-antigen glycan on a human protein by a bacterial glycosyltransferase and illustrates the potential of bacterial glycosyltransferases as tools for in vitro glycosylation of human proteins of therapeutic value.

Key words: ß1,3-galactosyltransferase / CgtB / enzyme improvement / glycosyltransferase


   Introduction
Top
 Abstract
 Introduction
Results
 Discussion
 Materials and methods
 Supplementary data
 Conflict of interest statement
 References
 
The cell surface glycolipids (lipooligosaccharides, LOS) of Campylobacter jejuni show considerable structural diversity, with many ganglioside mimics being found in pathogenic strains and this has been correlated to the genetic diversity of the locus (Gilbert et al. 2002Go). These ganglioside mimics have been extensively examined in the context of their role in the development of Guillain-Barré syndrome as a result of infection with C. jejuni (Yuki and Odaka 2005Go; Godschalk et al. 2007Go). The biological role of these diverse ganglioside mimics prompted us to explore in detail the structure–function relationships of the various glycosyltransferases involved in their synthesis. The presence of a diverse set of enzymes in this bacterial species which are sequence related, and which make the same carbohydrate linkage on different acceptors represent a resource to probe the molecular basis of this acceptor specificity.

The LOS biosynthesis gene clusters of a large number of C. jejuni strains have been examined and have been divided in eight classes ("A" to "H") based on their gene content and genetic organization (Parker et al. 2005Go). The first three classes ("A", "B", and "C") contain the neuBCA and cst-II genes needed for the biosynthesis of sialic acid and its incorporation into the growing LOS to constitute ganglioside mimics (Gilbert et al. 2002Go; Parker et al. 2005Go). As we are interested in the synthesis of sialylated glycans, we have focused our research efforts on the enzymes encoded by genes from clusters belonging to classes "A", "B", and "C" (described in Gilbert et al. 2002Go).

There exists considerable amino acid sequence variation between C. jejuni strains for the same LOS biosynthesis glycosyltransferase (Gilbert et al. 2002Go). The investigation of the LOS biosynthesis gene clusters from 11 C. jejuni strains expressing eight different serotypes led to the identification of five distinct mechanisms by which this bacterium can vary the structure of its LOS (Gilbert et al. 2002Go). One of these mechanisms consists in the occurrence of single or multiple mutations leading to "allelic" glycosyltransferases with different acceptor specificities (Gilbert et al. 2002Go). This was clearly illustrated by the investigation of variants of CgtA (ß1,4-N-acetylgalactosaminyltransferase) and of Cst-II ({alpha}2,3-/ 2,8-sialyltransferase). Six CgtA variants (sharing 34% overall amino acid sequence identity) from classes "A", "B", and "C" displayed different specific activities toward nonsialylated (lactose), mono (GM31)- or di-sialylated (GD31) 6-(fluorescein-5-carbaxamido)hexanoic acid, succinimidyl ester (FCHASE)-labeled acceptors (Gilbert et al. 2002Go). Similarly, six variants of Cst-II (sharing 92% overall amino acid sequence identity) from classes "A" and "B" displayed different specific activities toward nonsialylated (lactose) or sialylated (GM31) FCHASE-labeled acceptors (Gilbert et al. 2002Go). A Cst-II variant belonging to class "C" had previously been shown to be active with lactose-FCHASE and inactive with GM3-FCHASE (Gilbert et al. 2000Go). These differences in acceptor preference and in specific activity levels are a consequence of amino acid sequence divergence between enzyme variants and this has been recently examined structurally with the crystal structures of two members of this protein family (Chiu et al. 2004, 2007Go). It has also been reported that the presence of the cst-II gene has been associated with immune-mediated neuropathy of the of C. jejuni strains (Van Belkum et al. 2001Go).

As is the case for CgtA, there exists significant sequence divergence among CgtB (ß1,3-galactosaminyltransferase) enzyme variants: Only 47% amino acid sequence identity has been reported between the sequences from 11 strains belonging to classes "A", "B", and "C" (Gilbert et al. 2002Go). The activity and acceptor preference of CgtB variants from classes "A", "B", and "C" need to be determined to evaluate their potential use for the synthesis of glycans of therapeutic interest as has been done for CgtA and Cst-II (Gilbert et al. 2000Go, 2002Go).

The CgtB sequences from strains ATCC 43432, OH4384, and ATCC 43460 (serotypes HS:4, HS:19, and HS:41, respectively; all members of class "A") share 99% sequence identity (Gilbert et al. 2002Go). CgtB from strain OH4384 (CgtBOH4384, also of serotype HS:19; Figure 1A) was chosen for further characterization since it had already been partially characterized (Gilbert et al. 2000Go). CgtB from strain NCTC 11168 (CgtB11168; serotype HS:2; Figure 1B), a member of the class "C" group was also chosen as it has been used for the in vitro synthesis of the glycone moiety of ganglioside GM1a and for the synthesis of ganglioside mimics (Linton et al. 2000Go; Blixt et al. 2005Go). A second member of class "A", that from strain ATCC 43438 (CgtBHS:10; serotype HS:10), was included for the comparison because its sequence differs from those of the other members of class "A". Compared to the other members of its class, the carboxy-terminal of CgtBHS:10 contains a large number of amino acid substitutions (Gilbert et al. 2002Go). These changes may have occurred to accommodate the presence of the nonsialylated acceptor in the inner core LOS of C. jejuni ATC 43438 (Figure 1C; Gilbert et al. 2002Go). The investigation of CgtBHS:10 will allow the verification of this hypothesis. No CgtB variant from class "B" was chosen, as their amino acid sequences are not distinct from those of class "A" (Gilbert et al. 2002Go).


Figure 1
View larger version (24K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Ganglioside mimics in the core oligosaccharides from three C. jejuni strains. (A) Strain OH4384 (Aspinall et al. 1994Go). (B) Strain NCTC 11168 (Szymanski et al. 2003Go) with the phase variable terminal Gal residue (St-Michael et al. 2002Go) in bold. (C) Strain HS:10 ATCC 43438 (Nam Shin et al. 1997Go). In all the cases, the galactosyl residue added by CgtB is underlined. The nonganglioside mimic portion of the LOS is shown in grey. The structures have been simplified for this figure to emphasize the ganglioside mimic and the full LOS core structures can be found in the references.

 
The in vitro synthesis of glycoconjugates to investigate biological activity has been aided by the incorporation of enzymatic glycosylation steps, which eliminates the need for chemical protection groups, and the formation of isomers which are then difficult to purify. The availability of enzymes with a wide spectrum of acceptor specificity is invaluable for the synthesis of natural and nonnatural oligosaccharides to evaluate the structure–activity relationships of these glycoconjugates. Many glycosyltransferases have now been characterized, and it appears that mammalian enzymes are more rigid in their acceptor specificity, and are still more difficult to express in quantity than many bacterial glycosyltransferases (Blixt and Razi 2006Go). The diverse glycosyltransferases from C. jejuni have been successfully incorporated in various chemo-enzymatic schemes. The glycosyltransferases CgtA, CgtB, and Cst-II from C. jejuni have been used in vitro for the synthesis of a series of mono-, di-, and tri-sialylated 2-azidoethyl-ganglio-oligosaccharides (Blixt et al. 2005Go). CgtA and CgtB have also been used for the biosynthesis of the carbohydrate moieties of gangliosides GM1 and GM2 (Linton et al. 2000Go; Antoine et al. 2003Go). We are presenting the detailed characterization of three distinct variants of CgtB. This work provides insight into the diversity of function of these enzymes and we report on the improvement of their enzymatic activity for in vitro synthesis using protein engineering. This broadens the in vitro synthetic capability of this diverse set of enzymes from C. jejuni.


   Results
Top
 Abstract
 Introduction
 Results
Discussion
 Materials and methods
 Supplementary data
 Conflict of interest statement
 References
 
CgtB variants sequence comparison and analysis
A multiple sequence alignment of the CgtB variants is shown in Figure 2A. Pairwise comparisons using any two of the three CgtB variants show that they share 54–59% sequence identity and 67–75% sequence similarity over their whole sequence (Figure 2B). The highly conserved amino-terminal domain of CgtB is thought to be the donor-binding domain (see discussion). In CgtB, this domain has been delineated as the first 108 positions based on sequence identity between the three sequences. Pairwise comparisons using this domain done between any two variants show that it is highly conserved with any two variants sharing at least 87% sequence identity (Figure 2B).


Figure 2
View larger version (62K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 (A) Multiple sequence alignment of representative CgtB enzyme variants. OH4384 is CgtBOH4384, 11168 is CgtB11168, HS:10 is CgtBHS:10. Identical amino acid residues in all the three sequences are white on a black background; those identical between any two sequences are white on a gray background. This alignment was made using ClustalX (Nicholas and Nicholas 1997Go) and was reformatted in GeneDoc (Thompson et al. 1997Go). (B) Percentages of identity between any two CgtB variants as given in Genedoc.

 
The carboxy-terminal domain is the acceptor-binding domain and comprises positions 109 to the end of CgtB. Pairwise comparisons using this domain reveal more sequence divergence with any two variants sharing at least 37% sequence identity and at least 54% sequence similarity (Figure 2B). The sequence divergence observed in the acceptor-binding domains supports the observation that they have different acceptor LOS molecules with or without branched sialic acid residues in their outer core.

Acceptor preferences of CgtB variants
Each variant of CgtB was expressed as a fusion with the MalE protein from E. coli (Table I) and was purified as described in the materials and methods section. The three purified enzymes were then assayed with the synthetic acceptors GalNAc{alpha}-, GalNAcß-, GM21-, lyso-GM21-, and GA21-FCHASE to investigate their acceptor preference (Table II). This data represents a screening of the activity of each CgtB variant on various acceptors. The assay conditions had not been optimized, nor had the purification of the enzymes been optimized. The data of Table II is presented in relative activity, in which the activity measured on GM2-FCHASE was used as the reference for MalE-CgtBOH4384 and MalE-CgtB11168 while the activity on GA2-FCHASE was the reference for MalE-CgtBHS:10. The reference acceptor for each CgtB variant was chosen on the basis of its structural similarity with the structure of the LOS of the corresponding strain (Figure 1).


View this table:
[in this window]
[in a new window]

  		Table I Bacterial strains, plasmids and oligonucleotides used in this work

 

View this table:
[in this window]
[in a new window]

  		Table II Initial substrate survey of the acceptor preference of MalE-CgtB fusions

 
The activity trends indicated that MalE-CgtBOH4384 and MalE-CgtB11168 were most active with the lyso-GM2-FCHASE acceptor, compared to the nonlipid containing acceptor GM2-FCHASE. The activity with either monosaccharide acceptor was also lower than that of the tetra-saccharide GM2-FCHASE (Table II). In the case of MalE-CgtBHS:10, its activity with the nonsialylated acceptor GA2-FCHASE was significantly higher than with any sialylated acceptor or monosaccharide acceptor (Table II). Taken together, these data indicate that the three variants have very different acceptor preferences. In particular, the differences observed between sialylated and nonsialylated acceptors (GM2-, lyso-GM2- vs. GA2-FCHASE) show that the sialylation of the acceptor is an important determinant in acceptor preference. This perfectly correlates with the sialylation state of the inner core LOS of each strain in vivo. The higher activity of MalE-CgtBOH4384 and MalE-CgtB11168 with lyso-GM2-FCHASE is attributable to the sphingosine lipid aglycone of the acceptor, which mimics the lipid A portion of the natural LOS acceptor.

CgtB enzyme improvement: construction of carboxy-terminal deletions
The expression of cgtB on its own in E. coli does not lead to visible over-production of CgtB and the protein is completely associated with the membrane (Gilbert et al. 2000Go; Linton et al. 2000Go). We have shown with other bacterial glycosyltransferases that removal of short stretches of C-terminal residues can lead to an increase in solubility for the recombinant enzymes in E. coli (Wakarchuk et al. 1998Go; Chiu et al. 2004Go). Deletions of 10, 20, and 30 aa residues were made on the CgtB from OH4384 and NCTC11168, but these modified proteins were not any more soluble nor were they expressed at a much higher level than the parent construct (data not shown). These deletions were next incorporated into the MalE-CgtB fusion proteins as these proteins were somewhat soluble to begin with. The deletions did not improve much the solubility of these fusion proteins, and in fact the longer deletions were deleterious to the activity (Table III).


View this table:
[in this window]
[in a new window]

  		Table III Specific activity of purified CgtB variants using FCHASE-labeled acceptors

 
With CgtBHS:10, only the 20 aa deletion was studied as the other two deletions were not very active. In this case, the decrease in activity is seen only on the {alpha}-anomer of GalNAc, but the activity on the ß-anomer actually increases slightly. The activity on the "best" acceptor GA2-FCHASE also decreased as was observed with the other enzyme/acceptor combinations. Clearly, the change in the activity of CgtBHS:10 suggests more subtle changes to the active site than observed in the other two enzyme variants.

Inversion of the fusion order and impact on CgtB specific activity
The data from the previous section showed that the carboxy-terminal end of CgtB was very important for acceptor interactions and sensitive to deletions. It was then reasoned that the inversion of the gene order of the fusion might improve the solubility and might compensate the loss of specific activity caused by the 30 amino acids deletion made in CgtB. These fusions were constructed, purified, and assayed as previously described. Once these truncated CgtB{Delta}C-MalE fusions had been generated, they were used in side-by-side comparisons with the other available CgtB constructs to evaluate which were the most active, and on which acceptor (Table III). The data show a surprising result in that the reversed gene order produces enzymes which are far more active than any CgtB construct previously made. A schematic of these constructs is shown in Figure 3. The range of improvements was indeed impressive; where the construct CgtBOH4384{Delta}C-MalE shows as much as a 11-fold increase over the full length construct on the tetrasaccharide GM2-FCHASE acceptor. This improvement was also significant in that a glycopeptide acceptor, which is a model system for producing the core 1 O-glycosylation on a human protein (see the following section), could now be effectively modified.


Figure 3
View larger version (17K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Schematic of the CgtB enzyme constructs used in this study. Full-length CgtB enzymes did not express well, so they were expressed as fusions with the MalE protein of E. coli on the N-terminal end of the enzyme. C-terminal deletions (C {Delta}) were introduced, and these proteins were examined as N-terminal or C-terminal MalE fusions.

 
The surprising trend of activity from the CgtB{Delta}C-MalE fusions was also seen with the other two variants of CgtB, although these enzymes were never as active as those derived from CgtBOH4384 (Table III). There were some differences seen, and one notable exception is that of CgtB11168{Delta}C-MalE on the {alpha}-anomer of GalNAc where the activity is unchanged. This may simply reflect the intrinsically low activity on this anomer. The CgtBHS:10 enzyme always displayed very weak activity on the monosaccharide acceptors, and so it was unexpected to see CgtBHS:10{Delta}C-MalE fusion have as much as a 25-fold increase on the ß-anomer of GalNAc.

The fusion CgtBOH4384{Delta}C-MalE is the engineered variant with the highest specific activities among all tested (Table III). Therefore, its kinetic parameters were investigated using the monosaccharide acceptors GalNAc{alpha}-FCHASE and GalNAcß-FCHASE and for its donor sugar UDP-Gal (Table IV). Previous attempts to obtain kinetic parameters with MalE-CgtB had been hampered by the low solubility of the monosaccharide acceptors, and that conditions of acceptor saturation could never be approached (data not shown). Because of the solubility limitations of the two monosaccharide acceptors, the assays were not performed under saturating conditions; the values reported in Table IV are therefore, estimates of the actual kinetic parameter. The Km(app) value for GalNAc{alpha}-FCHASE is 1.68 mM and that for GalNAcß-FCHASE is 3.18 mM. The Km(app), Vmax, and kcat values for GalNAcß-FCHASE are 2.1 times higher than those for GalNAc{alpha}-FCHASE. However, the specificity constant, kcat/Km, is very similar for the two acceptors which indicates that variant CgtBOH4384{Delta}C-MalE does not discriminate strongly between these two anomers, which makes this a very good catalyst for producing the core 1 type disaccharide.


View this table:
[in this window]
[in a new window]

  		Table IV Kinetic parameters of enzyme from CJL-136

 
Activity on the IFN{alpha}2b[Tn]-FCHASE peptide
The highest activity measured for GalNAc{alpha}-FCHASE among all the CgtB variants and CgtB constructs made so far has been that of CgtBOH4384{Delta}C-MalE (531.5 mU/mg, Table III). This warranted the evaluation of it as a tool to elaborate O-linked glycans on peptides and proteins. The other CgtBOH4384 constructs were also tested for comparison. The analysis of the galactosylation of IFN{alpha}2b[Tn]-FCHASE after 30 min by capillary electrophoresis showed that 91.9% of the material had been converted into IFN{alpha}2b[T-Ag]-FCHASE and there was also 4.4% of unwanted side-products (4.3% and 0.1% of IFN{alpha}2b[Gal-T-Ag]-FCHASE and IFN{alpha}2b[Gal-Gal-T-Ag]-FCHASE, respectively). These unwanted polygalactosylation products were seen in much great proportion on the simpler monosaccharide acceptors, and to a lesser extent with the GM2-FCHASE acceptor (data not shown). The reaction conditions were optimized to minimize the polygalactosylation of the peptide acceptor by keeping the concentration of donor in the reaction lower than 2 mM. The level of polygalactosylation was variable: whereas never more than 10% of the side products were observed on the peptide acceptor, more than 30% of the polygalactosylated product could be easily obtained with the monosaccharide acceptor (data not shown).

The reaction products were analyzed by capillary electrophoresis mass spectrometry (CE-MS; Figure 4A and B). The MS spectrum of two peaks from this separation were analyzed and are shown in Figure 3. The peak corresponding to starting material, IFN{alpha}2b[Tn]-FCHASE, (Figure 4A) contains a singly-protonated molecular ion at m/z 1377.2 and a doubly-protonated ion at m/z 689.4. Doubly-protonated ions with sodium and potassium adducts were detected at m/z 700.3 and 708.3, respectively. In addition, ions corresponding to IFN{alpha}2b[Tn]-FCHASE having lost the {alpha}-GalNAc and IFN{alpha}2b-FCHASE having lost its carboxy-terminal prolyl residue were also observed at m/z 1174.1 and 1059.0, respectively. The putative product peak gave a spectrum shown in Figure 3B, where the predominant species on the CE-MS spectrum of IFN{alpha}2b[T-Ag]-FCHASE were a singly-charged ion at m/z 1538.9, a doubly-protonated ion at m/z 770.0 with the corresponding ammonium adduct at m/z 778.7. As expected, the molecular weight of IFN{alpha}2b-[T-Ag]-FCHASE was 162 Dalton higher than that of IFN{alpha}2b[Tn]-FCHASE, thus confirming the galactosylation of the peptide by CgtBOH4384{Delta}C-MalE.


Figure 4
View larger version (22K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 CE-MS analysis of (A) IFN{alpha}2b[Tn]-FCHASE and (B) IFN{alpha}2b[T-Ag]-FCHASE. To confirm the conversion of this peptide substrate to the T-Ag containing form, we analyzed samples by CE-MS. Two peaks were analyzed which corresponded to the starting material and putative reaction product. The acceptor peak spectrum contains a singly-protonated molecular ion at m/z 1377.2 and a doubly-protonated ion at m/z 689.4. In this analysis, some of the GalNAc is released giving rise to the m/z 1174.1 peak. The predominant species on the CE-MS spectrum of the putative product peak, IFN{alpha}2b[T-Ag]-FCHASE, (Figure 3B) were a singly-charged ion at m/z 1538.9, a doubly-protonated ion at m/z 770.0 with the corresponding ammonium adduct at m/z 778.7. As expected, the molecular weight of IFN{alpha}2b[T-Ag]-FCHASE was 162 Dalton higher than that of IFN{alpha}2b[Tn]-FCHASE.

 
Activity on the protein IFN{alpha}2b[Tn]
DeFrees et al. (2006)Go have shown that it is possible to glycosylate the IFN{alpha}2b protein in vitro. The improved CgtBOH4384{Delta}C-MalE was therefore, evaluated for making the Core 1 disaccharide on this protein. The MALDI spectra of IFN{alpha}2b (Figure 5A) and IFN{alpha}2b[Tn] (Figure 5B) present peaks at m/z 19293 and 19512, which are consistent with the expected molecular weights of IFN{alpha}2b and the corresponding glycosylated product. The MALDI spectrum of IFN{alpha}2b[T-Ag] (Figure 5C), generated from the galactosylation of IFN{alpha}2b[Tn] with the improved variant shows a main species with a m/z of 19665 which is consistent with the presence of an additional hexose residue when compared to the main peak in Figure 4B. The presence of the species at m/z 19665 is the proof that the IFN{alpha}2b[Tn] protein can be specifically galactosylated with this enzyme. Within the limits of detection, we do not see any evidence for the presence of polygalactosylation products.


Figure 5
View larger version (11K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 MALDI analysis of (A) IFN{alpha}2b, (B) IFN{alpha}2b[Tn] and (C) IFN{alpha}2b[T-Ag]. In order to follow the conversion of the protein to the T-Ag containing form, we analyzed samples by MALDI-MS. Protein samples from the reactions with ppGalNAcT-2 and then with CgtBOH4384{Delta}C-MalE show a main species with a m/z of 19665 which is consistent with the presence of the disaccharide modification of the protein. The presence of the species at m/z 19665 is the proof that the IFN{alpha}2b[Tn] protein can be efficiently galactosylated by CgtBOH4384{Delta}C-MalE.

 

   Discussion
Top
 Abstract
 Introduction
 Results
 Discussion
Materials and methods
 Supplementary data
 Conflict of interest statement
 References
 
The multiple sequence alignments of the three variants of CgtB (Figure 2) shows that there are only 19 differences over the amino-terminal domain, resulting in any two of these three variants sharing at least 87% identity over positions 1–108 (Figure 2B). The high similarity between the three sequences in this region is easily explained by the fact that it forms the UDP-galactose binding domain. A BLAST similarity search performed using the amino-terminal domain of CgtBOH4384 detected the conserved domain of the GT-2 glycosyltransferase family (data not shown) and found significant sequence similarities with many confirmed or putative glycosyltransferases, including several from C. jejuni.

The sequences of three CgtB variants are much more divergent over the carboxy-terminal domain, with any two sequences sharing at least 37% identity (Figure 2B). A BLAST similarity search performed using the carboxy-terminal domain of CgtBOH4384 as the query sequence detected significant similarities with many confirmed or putative C. jejuni glycosyltransferases. The sequences sharing the highest similarities with the query were all annotated as putative ß1,3-galactosyltransferases, however, none of them have been biochemically characterized (data not shown). The highest similarity reported with a sequence other than from a Campylobacter species is that to a putative glycosyltransferase from Bacteroides thetaiotaomicron strain VPI-5482 and the similarity is considerably lower (BLAST Score of 44 with only 27% identity and 49% similarity over 110 positions (data not shown)) than those to C. jejuni. The absence of significant similarity to sequences other than those of CgtB variants suggests that the acceptor specificity of CgtB is determined by its carboxy-terminal domain. In the case of CgtB, various evolutionary mechanisms produced single or multiple mutations which led to "allelic" CgtBs with different acceptor specificities. These variants give rise to some of the structural diversity observed in the core oligosaccharide that contributes to the ability of some C. jejuni strains to present multiple ganglioside mimics to evade the human immune response during infection.

CgtB belongs to the second family of glycosyltransferases (GT2) in the CAZy database (Carbohydrate-Active enzymes; http://afmb.cnrs-mrs.fr/CAZY/index.html and Coutinho et al. 2003Go). With more than 6200 members, family GT2 is the largest by far of the 90 families of glycosyltransferases currently indexed in the CAZy database. The only three-dimensional structure currently available for a member of the GT-2 family is that of SpsA, a glycosyltransferase involved in the synthesis of the spore coat in the bacterium Bacillus subtilis (Charnock and Davies 1999Go; Tarbouriech et al. 2001Go). The donor and acceptor of SpsA remain unknown although its three-dimensional structure has been resolved with bound metal and various nucleotides (Charnock and Davies 1999Go; Tarbouriech et al. 2001Go). Based on this structure, enzymes of family GT2 are expected to contain the glycosyltransferase fold GT-A, which consists of two tightly associated and abutting ß/{alpha}/ß domains that tend to form a continuous central sheet of at least eight ß-strands. The amino-terminal domain of glycosyltransferases containing the GT-A fold appears to be the site of nucleotide–sugar binding (UDP-galactose in the case of CgtB), while the acceptor is bound by the carboxy-terminal domain. It was therefore assumed that any modifications to the amino-terminal domain would have a deleterious effect on the enzyme's activity, hence the focus on the carboxy-terminal end of the enzyme. CgtBOH4384 and SpsA share 24% and 54% of sequence identity and similarity over the first 117 positions (almost exclusively the donor-binding domain) of CgtBOH4384, but share no significant sequence similarity over the acceptor-binding domain. Important amino acid residues for acceptor binding remain unknown, although some of the amino acid residues conserved in all the three CgtB sequences (Figure 2) could be involved in this process.

The determination of the acceptor preference for three CgtB variants has illustrated the role played by sialylation which has a significant impact on the ganglioside mimic displayed by these different strains. The LOS of strains OH4384 and NCTC 11168 both contain a sialic acid residue {alpha}-2,3-linked to an inner-core galactosyl residue (Figure 1A and B) and this galactosyl residue is not sialylated in the inner core LOS oCf strain HS:10 ATCC 43438 (Figure 1C). The correlation between the presence of sialic acid in the LOS and the preference for sialylated or nonsialylated acceptors has previously been observed with variants of the ß1,4-N-acetylgalactosaminyltransferase CgtA: CgtAOH4384, CgtA11168, and CgtAO:36 (strain ATCC 43456, serotype HS:36) are active with GM3-FCHASE as acceptor; CgtAOH4384 has a much lower specific activity (140 times) with lactose-FCHASE, while CgtA11168 and CgtAO:36 are inactive with this acceptor (Gilbert et al. 2002Go). On the other hand, CgtAHS:10 is active with lactose-FCHASE and inactive with GM3-FCHASE (Gilbert et al. 2002Go). The results we presented here for the variants of CgtB correlate perfectly with those reported earlier for variants of CgtA.

In the absence of a three-dimensional structure for CgtB to explain the acceptor specificity, we have to speculate based on the activity differences seen between the variants described in this report. We hypothesize that the deletion in the terminal domain of this protein has weakened or eliminated some of the productive interactions between the glycosyltransferase and its acceptor. It is likely that some changes in the local folding around the active site has altered these interactions (or introduced new ones), and that perhaps the active site is now more accessible to the nonnatural acceptors we used. With regard to the higher specific activity of the CgtB{Delta}C-MalE constructs, the presence of MalE at the carboxy-terminal end of CgtB may compensate for the deletion by providing new interactions with the end of the transferase, again by a slight perturbation of the active site through steric and/or charge interactions. Structural studies are planned on this protein in order to better understand the increase in activity with these unique galactosyltransferases.

Our observations regarding polygalactosylation suggest that it is acceptor-dependent and that it is more important with CgtBOH4384 (data not shown). Polygalactosylated products have not been seen with the other two variants of CgtB. We do not yet know if this activity is simply an artifact based on the use of small synthetic acceptors, or is a consequence of the higher activity of this CgtB variant with synthetic acceptors. Clearly, the active site structure of the CgtBOH4384 enzyme must be somewhat different from the other two as it more readily accepts the {alpha}-anomer of GalNAc, and appears to have a higher turnover of acceptor, although we have not measured real kinetic constants with this enzyme yet. The planned structural studies on CgtB will hopefully explain some of these phenomena.

The T-Ag disaccharide has been synthesized on peptides using various chemical strategies (reviewed in Brocke and Kunz 2002Go). Enzymatic strategies have relied on the use of glycosyltransferases from various eukaryotic organisms. The use of the recombinant mammalian Core 1 ß-galactosyltransferase was originally hampered by the fact that it has a chaperone required for enzyme activity (Ju and Cummings 2002Go). More recently, the Drosophilia Core 1 synthetase has been described (Müller et al. 2005Go) and has been used to produce T-Ag on granulocyte-macrophage colony stimulating factor (GM-CSF[GalNAc{alpha}]) (DeFrees et al. 2006Go). Bacterial galactosyltransferase enzymes, which make the T-Ag structure have not to our knowledge been previously described. A Streptococcus agalactiae type 1b ß1,3-galactosyltransferase enzyme has been characterized which makes a Gal-ß-1,3-GlcNAc-beta-R (Watanabe et al. 2002Go), but it is not known if this enzyme would use other acceptors such as GalNAc{alpha}. Similarly, there is an enzyme from E. coli VW187 O:7 which has ß1,3-galactosyltransferase activity on bactoprenylpyrophosphate linked GlcNAc{alpha}, but has not been reported to use GalNAc based acceptors (Riley et al. 2005Go). There are some reports in which eukaryotic glycosyltransferases have been used to synthesize the Tn, S-Tn (NeuAc{alpha}2,6-GalNAc{alpha}-), the T-Ag, and sialyl T-Ag (ST-Ag, NeuAc{alpha}2,3-Galß1,3-GalNAc{alpha}-) (George et al. 2001Go; Blixt et al. 2002Go; DeFrees et al. 2006Go; Sørensen et al. 2006Go). The T-Ag may then be modified by other glycosyltransferases to generate more complex O-linked structures. Some of these modifications consist in the sialylation of the galactosyl residue via an {alpha}-2,3 linkage to generate ST-Ag. The ST-Ag may in turn be further sialylated by the addition of a second sialic acid residue to the first one via an {alpha}-2,8 linkage. The ease of production of bacterial glycosyltransferase enzymes has made them more prominent in glyconjugate synthesis than many of the mammalian counterparts (Blixt and Razi 2006Go). The usefulness of CgtBOH4384{Delta}C-MalE for the in vitro synthesis of IFN{alpha}2b[T-Ag] certainly justifies an investigation of the potential usefulness of other C. jejuni glycosyltransferases for the synthesis of O-linked glycans on proteins. C. jejuni contains two sialyltransferases, Cst-I and Cst-II, which sialylate the LOS (Gilbert et al. 2000Go, 2002Go). One or both of these enzymes may be capable of sialylating IFN{alpha}2b[T-Ag] to generate IFN{alpha}2b[ST-Ag] and IFN{alpha}2b[di-ST-Ag].

The galactosylation of IFN{alpha}2b[Tn] with an engineered variant of CgtB clearly demonstrates that a bacterial glycosyltransferase is able to glycosylate a human protein in vitro. To the best of our knowledge, this is the first time that the direct glycosylation of a human protein by a bacterial glycosyltransferase is reported. Considering the ease of production of typical bacterial proteins compared to that of mammalian or human proteins in E. coli, the potential of bacterial glycosyltranferases for the in vitro or in vivo glycosylation of other human therapeutic proteins certainly deserves to be further explored.


   Materials and methods
Top
 Abstract
 Introduction
 Results
 Discussion
 Materials and methods
Supplementary data
 Conflict of interest statement
 References
 
Bacterial strains and plasmids
The bacterial strains and plasmids used in this work are listed in Table I. E. coli strains were maintained on 2YT (BIO 101, Carlsbad, CA) plates. For growth of E. coli in liquid medium (2YT), cultures were inoculated from fresh overnight cultures, grown at 37°C for 2 h, supplemented with isopropyl-1-thio-ß-D-galactopyranoside (IPTG) to a final concentration of 1 mM and were grown at 20°C for an additional 24 h before harvest. When required, ampicillin was added to 150 µg/mL.

Molecular biology
All the oligonucleotide primers used for this work are listed in Supplementary data (Table I). A schematic of the various constructs used to evaluate the enzyme activity is shown in Figure 3. Amplification reactions were done using purified C. jejuni DNA as described previously (Gilbert et al. 2002Go). Amplicons were purified using the QIAquick polymerase chain reaction (PCR) Purification Kit (QIAGEN Inc., Mississauga, Ontario, Canada) and cloned as fragments with either NdeI and SalI ends (in pCWori+ or pCWmalE-N) or with EcoRI and SalI termini (in pCWmalE-C). The construction of plasmids pCWmalE-N and pCWmalE-C is described in the Supplementary data section. DNA sequencing reactions were performed as described previously (Gilbert et al. 2002Go). The nucleotide sequences of cgtB from various serotypes of C. jejuni are available in Genbank (entries AF130984 [GenBank] , AF400048 [GenBank] , and AL139077 for cgtBOH4384, cgtBHS:10, and cgtB11168, respectively). Similarity and identity percentages between CgtB sequences were determined using Pairwise BLAST (BLAST 2 Sequences; http://www.ncbi.nlm.nih.gov/blast/bl2seq/bl2.html).

Protein purification
Cells grown in liquid cultures as described above were resuspended at 10% (w/v) in Na N-(2-hydroxyethyl)piperazine-N'-2-ethanesulfonic acid (HEPES) 20 mM pH 7.0, 200 mM NaCl, 5 mM ß-mercaptoethanol, 1 mM EDTA, 10% glycerol (Buffer A) and were lysed by two passages through an Emulsiflex (Avestin, Ottawa, Ontario, Canada). The cell lysate was centrifuged at 20,000 x g for 30 min at 4°C (SS-34 rotor). The supernatant was diluted as needed in buffer A and applied to a 20 mL column of amylose resin (New England Biolabs Ltd., Pickering, Ontario, Canada). After sample application, the column was washed with two column volumes to elute unbound proteins. The bound protein was eluted by washing the column with buffer B (Buffer A with 20 mM maltose) and the eluate was collected in 2 mL fractions. The fractions containing the eluted protein were pooled and dialyzed overnight against 100 volumes of sodium acetate 50 mM pH 6.0, glycerol 20% at 4°C. Protein quantitation was done using the BCA reagent (Pierce, Rockford, IL).

Enzyme assays
All of the CgtB variants have been previously shown either by NMR or glycosidase digestion to produce only ß-1,3-linked transfer products when used as recombinant proteins for in vitro synthesis (Linton et al. 2000Go; Antoine et al. 2003Go; Blixt et al. 2005Go). For that reason, no verification of the linkages formed with the new constructs was performed as we felt a complete change in specificity was very unlikely to have occurred with our constructs. The oligosaccharide acceptors and the interferon alpha 2b (IFN{alpha}2b) peptide (VGVTETP, corresponding to positions 103–109 of the mature IFN{alpha}2b protein) were labeled with FCHASE as previously described for aminophenylglycosides (Wakarchuk et al. 1996Go). IFN{alpha}2b-FCHASE was purified by reverse-phase chromatography using a PRP-1 column (10 x 250 mm, Hamilton Company, Reno, NV) in 10 mM ammonium acetate pH 5.5 with a gradient of acetonitrile (0–100% over four column volumes). CgtB activity assays were done in cell lysates or using purified enzyme in 50 mM 2-(N-Morpholino)ethanesulfonic acid (MES) pH 6.0, 0.5 mM of FCHASE-acceptor (unless otherwise indicated), 10 mM MnCl2, 10 mM DTT, 1 mM UDP-Gal at 37°C from 5 to 30 min. All reactions were stopped by addition of 10 µL of 50% acetonitrile, 10 mM EDTA, and 1% SDS and were diluted with H2O to obtain 10–15 µM final concentration of the FCHASE-labeled compounds. The samples were analyzed by capillary electrophoresis as described previously (Wakarchuk and Cunningham 2003Go) except that a P/ACE MDQ Capillary Electrophoresis System equipped with a Laser module 488 (Beckman Coulter, Fullerton, CA) was used. Quantitation of the reactions was performed by integration of the trace peaks using the MDQ 32 Karat software.

Glycosylation of IFN{alpha}2b-FCHASE and IFN{alpha}2b
The IFN{alpha}2b protein was purchased from Cell Sciences (Canton, MA). IFN{alpha}2b-FCHASE and IFN{alpha}2b were converted into IFN{alpha}2b[Tn]-FCHASE and IFN{alpha}2b[Tn] using ppGalNAc-T2-MBP (a gift from NEOSE Technologies) (DeFrees et al. 2006Go). IFN{alpha}2b[Tn]-FCHASE was purified by reverse-phase chromatography as described above. IFN{alpha}2b[Tn] was purified by cation-exchange chromatography using a Mini S 4.6/5.0 column (GE Healthcare Bio-Sciences, Baie d’Urfé, Québec, Canada) using a 0.01–1 M gradient of NH4OAc pH 4.5.

Purified IFN{alpha}2b[Tn]-FCHASE 500 µg (360 nmole) was converted into IFN{alpha}2b[T-Ag]-FCHASE in a reaction containing 15 mU of purified CgtBOH4384{Delta}C-MalE, 50 mM MES pH 6.0, 10 mM MnCl2, 1 mM DTT, and 2 mM UDP-Gal. The reaction was stopped after 30 min, cleaned on a Sep Pak C18 column (Waters Corporation, Milford, MA) and the IFN{alpha}2b [T-Ag]-FCHASE was purified by reverse-phase chromatography as described above.

IFN{alpha}2b[Tn] was converted into IFN{alpha}2b[T-Ag] in 1 mL reactions containing 2.5 mg of IFN{alpha}2b[Tn], 10 mM MnCl2, 1 mM DTT, 2 mM UDP-Gal, 50 mM NaCl, in 50 mM NaOAc pH 6.0 and 5 mU of purified CgtBOH4384{Delta}C-MalE. The reaction was left overnight at room temperature. The next day, the reaction was supplemented with 1 mM UDP-Gal and 2.5 mU of purified CgtBOH4384{Delta}C-MalE. The newly synthesized IFN{alpha}2b[T-Ag] was purified by cation-exchange chromatography as described above.

Mass spectrometry
A Prince CE system (Prince Technologies, Emmen, The Netherlands) was coupled to a 4000 QTRAP mass spectrometer (Applied Biosystems/MDS Sciex, Streetsville, Canada). A sheath solution (isopropanol–methanol, 2:1) was delivered at a flow rate of 1.0 L/min. Separations were obtained on about 90 cm length bare fused-silica capillary using 15 mM ammonium acetate in deionized water, pH 9.0. The 5 kV of electrospray ionization voltage were used for positive ion mode. MALDI-TOF mass spectra were acquired on a Voyager-DE STR mass spectrometer (Applied Biosystems, Foster City, CA) equipped with a pulsed nitrogen laser (337 nm), with a voltage of 20 kV as the accelerating voltage in the positive mode.


   Supplementary data
Top
 Abstract
 Introduction
 Results
 Discussion
 Materials and methods
 Supplementary data
Conflict of interest statement
 References
 
Supplementary data for this article is available online at http://www.glycob.oxfordjournals.org/


   Conflict of interest statement
Top
 Abstract
 Introduction
 Results
 Discussion
 Materials and methods
 Supplementary data
 Conflict of interest statement
References
 
S.D. is an employee of and holds stock in Neose Technologies, which sponsored this research.


   Acknowledgements
 
We thank Melissa Schur and Denis Brochu for running reaction analysis on the capillary electrophoresis apparatus, Sonia Leclerc and Michael Masotti for DNA sequencing and oligonucleotide synthesis. We also acknowledge Dave Watson and Sonia Leclerc for the construction of pCWmalE-N and pCWmalE-C as well as Anna Cunningham and Elizabeth Willis for technical assistance with generating and characterizing deletion mutants. We also thank Jean-René Barbier and Gordon Willick for the synthesis of the IFN{alpha}2b peptide as well as Lisa Morrison and Jacek Stupak for the spectrometric analysis of samples by MALDI. This work was partially funded by a collaborative research project with NEOSE Technologies (Horsham, PA).


   Abbreviations
 
CE, capillary electrophoresis; FCHASE, 6-(fluorescein-5-carbaxamido)hexanoic acid, succinimidyl ester; HEPES, N-(2-hydroxyethyl)piperazine-N'-2-ethanesulfonic acid; IPTG, isopropyl-1-thio-ß-D-galactopyranoside; LOS, lipooligosaccharide; MES, 2-(N-Morpholino) ethanesulfonic acid; PCR, polymerase chain reaction


   References
Top
 Abstract
 Introduction
 Results
 Discussion
 Materials and methods
 Supplementary data
 Conflict of interest statement
 References
 
Akiyama Y, Ito K. SecY protein, a membrane-embedded secretion factor of E. coli, is cleaved by the OmpT protease in vitro. Biochem Biophys Res Commun (1990) 167:711–715.[CrossRef][Web of Science][Medline]

Antoine T, Priem B, Heyraud A, Greffe L, Gilbert M, Wakarchuk WW, Lam JS, Samain E. Large-scale in vivo synthesis of the carbohydrate moieties of gangliosides GM1 and GM2 by metabolically engineered Escherichia coli. Chembiochem (2003) 4:406–412.[CrossRef][Web of Science][Medline]

Aspinall GO, McDonald AG, Pang H, Kurjanczyk LA, Penner JL. Lipopolysaccharides of Campylobacter jejuni serotype O:19: structures of core oligosaccharide regions from the serostrain and two bacterial isolates from patients with the Guillain-Barré syndrome. Biochemistry (1994) 33:241–249.[CrossRef][Medline]

Blixt O, Allin K, Pereira L, Datta A, Paulson JC. Efficient chemoenzymatic synthesis of O-linked sialyl oligosaccharides. J Am Chem Soc (2002) 124:5739–5746.[CrossRef][Web of Science][Medline]

Blixt O, Razi N. Chemoenzymatic synthesis of glycan libraries. Methods Enzymol (2006) 415:137–153.[Medline]

Blixt O, Vasiliu D, Allin K, Jacobsen N, Warnock D, Razi N, Paulson JC, Bernatchez S, Gilbert M, Wakarchuk W. Synthesis of 2-azidoethyl-ganglio-oligosaccharides GD3, GT3, GM2, GD2, GT2, GM1 and GD1a. Carbohydrate Res (2005) 340:1963–1972.[CrossRef][Web of Science][Medline]

Brocke C, Kunz H. Synthesis of tumor-associated glycopeptide antigens. Bioorg Med Chem (2002) 10:3085–3112.[CrossRef][Medline]

Charnock SJ, Davies GJ. Structure of the nucleotide-diphospho-sugar transferase, SpsA from Bacillus subtilis, in native and nucleotide-complexed forms. Biochemistry (1999) 38:6380–6385.[CrossRef][Medline]

Chiu CP, Lairson LL, Gilbert M, Wakarchuk WW, Withers SG, Strynadka NC. Structural analysis of the alpha-2,3-sialyltransferase Cst-I from Campylobacter jejuni in Apo and substrate-analogue bound forms. Biochemistry (2007) 46:7196–7204.[CrossRef][Medline]

Chiu CP, Watts AG, Lairson LL, Gilbert M, Lim D, Wakarchuk WW, Withers SG, Strynadka NC. Structural analysis of the sialyltransferase CstII from Campylobacter jejuni in complex with a substrate analog. Nat Struct Mol Biol (2004) 11:163–170.[CrossRef][Web of Science][Medline]

Coutinho PM, Deleury E, Davies GJ, Henrissat B. An evolving hierarchical family classification for glycosyltransferases. J Mol Biol (2003) 328:307–317.[CrossRef][Web of Science][Medline]

DeFrees S, Wang ZG, Xing R, Scott AE, Wang J, Zopf D, Gouty DL, Sjoberg ER, Panneerselvam K, Brinkman-Van der Linden EC, Bayer RJ, Tarp MA, Clausen H. TGlycoPEGylation of recombinant therapeutic proteins produced in Escherichia coli. Glycobiology (2006) 16:833–843.[Abstract/Free Full Text]

George SK, Schwientek T, Holm B, Reis CA, Clausen H, Kihlberg J. Chemoenzymatic synthesis of sialylated glycopeptides derived from mucins and T-cell stimulating peptides. J Am Chem Soc (2001) 123:11117–11125.[CrossRef][Web of Science][Medline]

Gilbert M, Brisson JR, Karwaski MF, Michniewicz J, Cunningham AM, Wu Y, Young NM, Wakarchuk WW. Biosynthesis of ganglioside mimics in Campylobacter jejuni OH4384. Identification of the glycosyltransferase genes, enzymatic synthesis of model compounds, and characterization of nanomole amounts by 600-MHz 1H and 13C NMR analysis. J Biol Chem (2000) 275:3896–3906.[Abstract/Free Full Text]

Gilbert M, Karwaski MF, Bernatchez S, Young NM, Taboada E, Michniewicz J, Cunningham AM, Wakarchuk WW. The genetic bases for the variation in the lipo-oligosaccharide of the mucosal pathogen, Campylobacter jejuni: Biosynthesis of sialylated ganglioside mimics in the core oligosaccharide. J Biol Chem (2002) 277:327–337.[Abstract/Free Full Text]

Godschalk PC, Kuijf ML, Li J, St Michael F, Ang CW, Jacobs BC, Karwaski MF, Brochu D, Moterassed A, Endtz HP, van Belkum A, Gilbert M. Structural characterization of Campylobacter jejuni lipooligosaccharide outer cores associated with Guillain-Barre and Miller Fisher syndromes. Infect Immun (2007) 75:1245–1254.[Abstract/Free Full Text]

Ju T, Cummings RD. A unique molecular chaperone Cosmc required for activity of the mammalian core 1 ß3-galactosyltransferase. Proc Natl Acad Sci USA (2002) 99:16613–16618.[Abstract/Free Full Text]

Linton D, Gilbert M, Hitchen PG, Dell A, Morris HR, Wakarchuk WW, Gregson NA, Wren BW. Phase variation of a ß-1,3 galactosyltransferase involved in generation of the ganglioside GM1-like lipo-oligosaccharide of Campylobacter jejuni. Mol Microbiol (2000) 37:501–514.[CrossRef][Web of Science][Medline]

Müller R, Hülsmeier AJ, Altmann F, Ten Hagen K, Tiemeyer M, Hennet T. Characterization of mucin-type core-1 ß1–3 galactosyltransferase homologous enzymes in Drosophila melanogaster. FEBS Journal (2005) 272:4295–4305.[CrossRef][Web of Science][Medline]

Nam Shin JE, Ackloo S, Mainkar AS, Monteiro MA, Pang H, Penner JL, Aspinall GO. Lipo-oligosaccharides of Campylobacter jejuni serotype HS:10. Structures of core oligosaccharide regions from a bacterial isolate from a patient with the Miller–Fisher syndrome and from the serotype reference strain. Carbohydrate Res (1997) 305:223–232.[CrossRef][Web of Science][Medline]

Nicholas KB, Nicholas HB. GeneDoc: Analysis and visualization of genetic variation. (1997) EMBNEW.News. 4:14 (www.psc.edu/biomed/genedoc).

Parker CT, Horn ST, Gilbert M, Miller WG, Woodward DL, Mandrell RE. Comparison of Campylobacter jejuni lipooligosaccharide biosynthesis loci from a variety of sources. J Clin Microbiol (2005) 43:2771–2781.[Abstract/Free Full Text]

Riley JG, Menggad M, Montoya-Peleaz PJ, Szarek WA, Marolda CL, Valvano MA, Schutzbach JS, Brockhausen I. The wbbD gene of E. coli strain VW187 (O7:K1) encodes a UDP-Gal: GlcNAc a-pyrophosphate-R ß1,3-galactosyltransferase involved in the biosynthesis of O7-specific lipopolysaccharide. Glycobiology (2005) 15:605–613.[Abstract/Free Full Text]

Sørensen AL, Reis CA, Tarp MA, Mandel U, Ramachandran K, Sankaranarayanan V, Schwientek T, Graham R, Taylor-Papadimitriou J, Hollingswor MA, Burchell J, Clausen H. Chemoenzymatically synthesized multimeric Tn/STn MUC1 glycopeptides elicit cancer-specific anti-MUC1 antibody responses and override tolerance. Glycobiology (2006) 16:96–107.[Abstract/Free Full Text]

St-Michael F, Szymanski CM, Li J, Chan KH, Khieu NH, Larocque S, Wakarchuk WW, Brisson JR, Monteiro MA. The structures of the lipooligosaccharide and capsule polysaccharide of Campylobacter jejuni genome sequenced strain NCTC 11168. Eur J Biochem (2002) 269:5119–5136.[Web of Science][Medline]

Szymanski CM, St-Michael F, Jarrell HC, Li J, Gilbert M, Larocque S, Vinogradov E, Brisson JR. Detection of conserved N-linked glycans and phase-variable lipooligosaccharides and capsules from campylobacter cells by mass spectrometry and high resolution magic angle spinning NMR spectroscopy. J Biol Chem (2003) 278:24509–24520.[Abstract/Free Full Text]

Tarbouriech N, Charnock SJ, Davies GJ. Three-dimensional structures of the Mn and Mg dTDP complexes of the family GT-2 glycosyltransferase SpsA: a comparison with related NDP-sugar glycosyltransferases. J Mol Biol (2001) 314:655–661.[CrossRef][Web of Science][Medline]

Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG. The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res (1997) 25:4876–4882.[Abstract/Free Full Text]

Van Belkum A, Van Den Braak N, Godschalk P, Ang W, Jacobs B, Gilbert M, Wakarchuk W, Endtz H. A Campylobacter jejuni gene associated with immune-mediated neuropathy. Nat Medicine (2001) 7:752–753.[CrossRef][Web of Science][Medline]

Wakarchuk WW, Campbell RL, Sung WL, Davoodi J, Yaguchi M. Mutational and crystallographic analyses of the active site residues of the Bacillus circulans xylanase. Prot Sci (1994) 3:467–475.[Web of Science][Medline]

Wakarchuk W, Martin A, Jennings MP, Moxon ER, Richards JC. Functional relationships of the genetic locus encoding the glycosyltransferase enzymes involved in expression of the lacto-N-neotetraose terminal lipopolysaccharide structure in Neisseria meningitidis. J Biol Chem (1996) 271:19166–19173.[Abstract/Free Full Text]

Wakarchuk WW, Cunningham A, Watson DC, Young NM. Role of paired basic residues in the expression of active recombinant galactosyltransferases from the bacterial pathogen Neisseria meningitidis. Protein Eng (1998) 11:295–302.[Abstract/Free Full Text]

Wakarchuk WW, Cunningham AM. Capillary electrophoresis as an assay method for monitoring glycosyltransferase activity. Methods Mol Biol (2003) 213:263–274.[Medline]

Watanabe M, Miyake K, Kouji Yanae, Kataoka Y, Koizumi S, Endo T, Ozaki A, Iijima S. Molecular characterization of a novel ß1, 3-galactosyltransferase for capsular polysaccharide synthesis by Streptococcus agalactiae. (2002) type Ib. J Biochem (Tokyo). 131:183– 191.

Yuki N, Okada M. Ganglioside mimicry as a cause of Guillain-Barré syndrome. Curr Opin Neurol (2005) 18:557–61.[Web of Science][Medline]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 GlycobiologyHome page
R S. Houliston, S. Bernatchez, M.-F. Karwaski, R. E Mandrell, H. C Jarrell, W. W Wakarchuk, and M. Gilbert
Complete chemoenzymatic synthesis of the Forssman antigen using novel glycosyltransferases identified in Campylobacter jejuni and Pasteurella multocida
Glycobiology, February 1, 2009; 19(2): 153 - 159.
[Abstract] [Full Text] [PDF]

Home page
 GlycobiologyHome page
L. M Willis, M. Gilbert, M.-F. Karwaski, M.-C. Blanchard, and W. W Wakarchuk
Characterization of the {alpha}-2,8-polysialyltransferase from Neisseria meningitidis with synthetic acceptors, and the development of a self-priming polysialyltransferase fusion enzyme
Glycobiology, February 1, 2008; 18(2): 177 - 186.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
17/12/1333    most recent
cwm090v2
cwm090v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (4)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Bernatchez, S.
Right arrow Articles by Wakarchuk, W. W
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bernatchez, S.
Right arrow Articles by Wakarchuk, W. W
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?