Glycobiology Advance Access originally published online on December 12, 2007
Glycobiology 2008 18(2):152-157; doi:10.1093/glycob/cwm134
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Communication |
Glucuronylation in Escherichia coli for the bacterial synthesis of the carbohydrate moiety of nonsulfated HNK-1
2 Centre de Recherches sur les Macromolécules Végétales, 601 rue de la Chimie, BP 53X, 38041 Grenoble, cedex 09, France
3 Faculty of Life Sciences, University of Manchester, Manchester, UK
1 To whom correspondence should be addressed: Fax: +33-476-547-203; e-mail: priem{at}cermav.cnrs.fr
Received on October 31, 2007; revised on December 3, 2007; accepted on December 4, 2007
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsSupplementary DataConflict of interest statementReferences |
|---|
We have previously reported the large-scale synthesis of neolactotetraose (Galβ-4GlcNAcβ-3Galβ-4Glc) from lactose in engineered Escherichia coli cells (Priem B, Gilbert M, Wakarchuk WW, Heyraud A and Samain E. 2002. A new fermentation process allows large-scale production of human milk oligosaccharides by metabolically engineered bacteria. Glycobiology. 12:235–240). In the present study we analyzed the adaptation of this system to glucuronylated oligosaccharides. The catalytic domain of mouse glucuronyl transferase GlcAT-P was cloned and expressed in an engineered strain which performed the in vivo synthesis of neolactotetraose. Under these conditions, efficient glucuronylation of neolactotetraose was achieved, but some residual neolactotetraose was still present. Although E. coli K-12 has an indigenous UDP-glucose dehydrogenase, the yield of glucuronylated oligosaccharides was greatly improved by the additional expression of the orthologous gene kfiD from E. coli K5. Glucuronylation of neolactohexaose and lactose was also observed. The final glucuronylated oligosaccharides are precursors of the brain carbohydrate motif HNK-1, involved in neural cell adhesion.
Key words: Escherichia coli / glucuronyltransferase / HNK-1 / metabolic engineering / UDP-glucose dehydrogenase
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary DataConflict of interest statementReferences |
|---|
Carbohydrate molecules on the cell surface are involved in many physiological processes. The Human Natural Killer cell-1 (HNK-1) carbohydrate motif sulfo-3GlcAβ-3Galβ-4GlcNAc-R is shared by glycolipids and glycoproteins. Its expression is predominant in the nervous system, where it plays a role in cell adhesion. It is highly expressed in embryonic brain, where it participates in neural crest development (Jungalwala 1994
). Isolated sulfoglucuronyl glycolipids bearing the HNK-1 epitope (SGGLs) promote neurite outgrowth of motor neurons in vitro (Martini et al. 1992). On the other hand, 52% of the IgM found in antimyelin-associated glycoprotein (anti-MAG) neuropathy cross-reacts with SGGLs (Steck et al. 2006).
The design of artificial sulfoglucuronyl polymers would be of great interest when studying nerve regeneration and stem cell differentiation. However, neither HNK-1 carbohydrates nor SGGLs are commercially available and their synthesis is proving very difficult. The chemical synthesis of SGGLs has been described (Chevalier et al. 2006), but the methodology is complicated due to the many protection and de-protection steps required. Their enzymatic synthesis would represent an interesting alternative, as the genes encoding sulfotransferase and glucuronyltranferase involved in HNK-1 synthesis have been identified. However the development of such systems is hampered by the high cost of the activated substrates used for the sulfotransferase (PAPS) and glucuronyltranferase (UDP-GlcA), even if these costs can be reduced by the use of multienzymatic systems for the regeneration of PAPS and UDP-GlcA.
A far more economical approach is the production of oligosaccharides by metabolically engineered Escherichia coli. This system has been successfully used for the synthesis of a large variety of complex sugars, including sulfated oligosaccharides (Samain et al. 1999). However up to now this approach has never been used for the synthesis of structures containing glucuronic acid, and its use for the production of the HNK-1 epitope is conditional on the production of UDP-GlcA in E. coli cells. Like most enterobacterial species, E. coli K-12 secretes an exopolysaccharide, which contains glucuronic acid and can therefore synthesize UDP-GlcA from UDP-glucose in a reaction catalyzed by the UDP-glucose dehydrogenase (Ugd). The E. coli ugd gene has been identified but the regulation of its expression has not been studied. In the closely related Salmonella species, it was shown that ugd was regulated by different inducers, including the RcsA protein (Mouslim and Groisman 2003), which is known to act as a positive regulator of the colanic acid operon (Stevenson et al. 1996). In addition to colanic acid, there are other bacterial polysaccharides that contain glucuronic acid, and it is worth noting that E. coli K5, which produces a capsular polysaccharide with a basal repeat structure similar to that of heparin, has an additional Ugd encoded by the kfiD gene (Petit et al. 1995).
We have previously described a system for the bacterial production of oligosaccharides that contains the terminal type 2 lactosaminyl motif used as an acceptor by mammalian glucuronyltransferases (GlcAT-P and GlcAT-S). In this paper, we report on the extension of this process to the production of glucuronylated oligosaccharides that could serve as precursors for the synthesis of the HNK-1 epitope, and we show that the expression of the kfiD gene is required for an efficient glucuronylation.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary DataConflict of interest statementReferences |
|---|
Production of GlcAβ-3Galβ-4GlcNAcβ-3Galβ-4Glc (GlcAnLc4)
The catalytic domain of glucuronyltransferase of mouse (GlcAT-P) involved in the biosynthesis of HNK-1 carbohydrate was cloned to assess its ability to act in the cytoplasm of E. coli. The gene encoding GlcAT was expressed together with the necessary genes to synthesize neolactotetraose (nLc4). The positive regulator of colanic acid synthesis (rcsA) was also expressed, glucuronic acid being one of the constituents of colanic acid. The biosynthetic pathway of the expected product is shown in Figure 1. The synthesis was carried out by high-density culture as previously reported for the synthesis of nLc4 (Priem et al. 2002
). As judged by the TLC profile (Figure 2, Table I), a compound having a lower migration than nLc4 accumulated in the cells of strain HN1. This compound was purified by ion-exchange chromatography and gel permeation and was identified as GlcAβ-3Galβ-4GlcNAc-β-3Galβ-4Glc. The presence of residual nLc4 in the final products led us to try several combinations of plasmid constructions in order to improve the glucuronylation of nLc4 (Table I). The overexpression of RcsA (strain HN1) was found to have little effect on glucuronylation, since the yield of GlcAnLc4 without RcsA (strain HN2) did not significantly change. Neither did the overexpression of Ugd of E. coli (Ugd, strain HN4) improve the production of GlcAnLc4. In fact, only one combination involving the overexpression of Ugd involved in the synthesis of glucuronic acid in E. coli K5 (KfiD, strain HN6), resulted in a high consumption of nLc4 (Figure 2(A), lane b). Under these conditions, the yield of purified GlcAnLc4 from strain HN6 was about 5.1 g/L of the culture medium.
|
|
|
Production of GlcAβ-3Galβ-4GlcNAcβ-3Galβ-4GlcNAcβ- 3Galβ-4Glc (GlcAnLc6)
From our previous work dealing with nLc4 synthesis, it was known that nLc4 itself was a good substrate for further action by the endogenous glycosyltransferases, thus giving neolactohexaose among the final products (Priem et al. 2002
). In the present experimental approach, nLc4 is a possible acceptor for both (subsequent action of) LgtA (GlcNAc transferase) and GlcAT-P. The tendency of the system to drive either one or the other reaction depends on the relative activities of LgtA versus GlcAT-P and the duration of culture. We constructed a plasmid (pBSkfiDglcAT) in which kfiD was placed before the GlcAT-P encoding sequence (strain HN5), thus favoring the expression of kfiD rather than glcAT-P (the expression level of genes inside the plasmid operon decreases with distance from the Lac promoter). The production period was also increased to 48 h. These conditions gave time for the production of GlcAnLc6, obtained together with GlcAnLc4 (Figure 2(A), lane d). Although TLC analysis of the intracellular extract did not allow for GlcAnLc6 to be clearly distinguished, the gel permeation chromatography of the acidic fraction resolved the purification of that compound (Figure 2(B), profile d, peak 2). The yield of purified GlcAnLc6 from strain HN5 was 0.5 g/L, leading also to a yield of 2.4 g/L of GlcAnLc4.
Production of GlcAβ-3Galβ-4Glc (GlcALac)
During the production of GlcAnLc4 and GlcAnLc6, we noticed the accumulation of an unknown compound in the extracellular fraction, having a TLC migration distance between lactose and nLc4 (Figure 2(A), lanes e, f, h). A smaller amount of this compound was also present in the intracellular fraction (Figure 2(A), lanes a, b, d). Measurement of the uronic acid content confirmed the accumulation of an acidic compound in the medium (data not shown). This was successfully purified from the extracellular fraction (culture medium) and was identified as GlcAβ-3Galβ-4Glc, thus attesting to a glucuronylation of lactose. The ability of glucuronyltransferase to accept lactose was confirmed by running a culture with a strain lacking the necessary genes for the synthesis of nLc4 (strain HN7, Figure 2(A), lane g). Under those conditions, GlcALac alone was produced, with a yield of 4.0 g/L of medium.
Sensitivity of the glucuronylated oligosaccharides to endogenous β-glucuronidase
E. coli possess a β-glucuronidase (GUS) which is highly expressed, and fluorescent β-glucuronides are used for the detection of coliforms. We investigated the possibility that a futile cycle was taking place, thus lowering the yield of glucuronylation and driving out the central energy-generating pathway of the cells. We thus performed incubations at different time intervals of the purified oligosaccharides with the E. coli enzyme (30 units/µmole of substrate) and did not notice any degradation (data not shown). This explains the accumulation of glucuronylated oligosaccharides in the bacteria, despite GUS activity; this can be explained by the preference of the enzyme for glycolipids, which are normal detoxification products of the human body and constitute a carbon source for intestinal bacteria.
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary DataConflict of interest statementReferences |
|---|
Our data clearly show that the (catalytic domain of) glucuronyltransferase (GlcAT-P) of mouse brain that we expressed in E. coli cells along with another glycosyltransferases efficiently attached glucuronic acid onto neolactotetraose and neolactohexaose. Moreover, a noticeable activity was also observed with lactose, the first exogenous acceptor in oligosaccharide synthesis.
The homologous gene GlcAT-P of rat brain was previously described as accepting the acceptor N-acetyllactosamine (Galβ1-4GlcNAc) in vitro (Kakuda et al. 2005); the acceptance of neolactotetraose containing this motif was thus quite expected. In contrast, lactose was reported to be a poor acceptor for the enzyme (4% transfer compared to LacNAc, Kakuda et al. 2005); we show here that lactose is efficiently glucuronylated under our experimental conditions.
The challenging step we had to overcome was the capacity of the bacteria to furnish UDP-glucuronic acid. Since glucuronic acid is a precursor of the exopolysaccharide, colanic acid, we first overexpressed RcsA, a positive regulator of colanic acid synthesis. In our previous work, we used a comparable strategy involving RcsA to perform fucosylation in E. coli (Dumon et al. 2001), since fucose, as well as glucuronic acid, is a constituent of colanic acid. In the present work, neither RcsA was required to perform glucuronylation, nor did its overexpression increase the yield of glucuronylated products. We thus decided to clone the gene encoding the Ugd of K-12 responsible for the synthesis of UDP-glucuronic acid into an IPTG-inducible vector, but this strategy also proved to be unsuccessful. Surprisingly, the overexpression of another Ugd of E. coli K5 (KfiD) was able to boost the synthesis until total depletion of the acceptor neolactotetraose. A possible explanation for this phenomenon is that there is a posttranslational negative control of Ugd in E. coli K-12, which does not apply to the strain K5 enzyme. It is known that the K-12 enzyme is positively regulated by phosphotyrosine-protein kinase (Grangeasse et al. 2003). The kinase itself can be dephosphorylated by phosphotyrosine-protein phosphatase. Both kinase (Wzc) and phosphorylase (Wzb) belong to the colanic acid gene cluster. It is possible that in K-12, this system specifically counter-acts the overexpression of Ugd, but not of KfiD. Indeed, kfiD belongs to the capsule gene cluster in charge of the synthesis of heparin-like polysaccharides. Group K enterobacteria also possess the colanic acid cluster, but the two metabolisms are regulated independently. As a matter of fact, the expression of the K-antigen capsule is not positively regulated by the rcs system, unlike that of colanic acid (Keenleyside et al. 1993).
The three oligosaccharides we have obtained in this study are particularly valuable molecules. The nonsulfated HNK-1 epitope is present in some tissues such as kidney (Tagawa et al. 2005), but the biological role of nonsulfated HNK-1 is unknown. Moreover, a recent report mentions that glucuronyl neo-carbohydrates are ligands for the pathogenic strain Helicobacter pylori (Weikkolainen et al. 2007).
The next step will be the endogenous sulfation of glucuronic acid by the subsequent action of the HNK-1 sulfotransferase. We are currently investigating such a goal, thereby leading to the total synthesis of HNK-1 carbohydrate in recombinant E. coli.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary DataConflict of interest statementReferences |
|---|
Bacterial strains, plasmids, and cloning procedures
A summary of plasmids and strains used is presented in Supplementary data.
E. coli mutant strain DJ.
An E. coli mutant derived from DH1 previously used in our lab, was further mutated in order to inactivate wcaJ. Primers 5'-CAAGGATCCAGATGAC- AAATCTAAAAAAGCGCGA (a) and 5'-TGCGTGCGGA- CGTGGACCGACAATCCGTTTCGCGTCCTCGACCAG (b) were used for PCR amplification of 0.64 kb of DNA flanking the 5' end of wcaJ, whereas primers 5'-ATTGTCGGTCC- ACGTCCGCACC (c) and 5'-TTAATTTGGATCCAATCGGG- TTACCTACGGAGC (d) were used for PCR amplification of 1.27 kb of DNA flanking the 3' end of wcaJ. The reverse primer (c) sequence was contained in primer (b), allowing a fusion of both DNA fragments by PCR amplification with primers (a + d). The fused DNA of 1.91 kb containing a truncated wcaJ was cloned into the BamHI site of the suicide plasmid pKO3 (Link et al. 1997). The resulting recombinant suicide plasmid was transformed into DJ competent cells; mutants were obtained according to the author's instructions. Positive clones were screened by PCR with primers (a) and (c).
Construction of pBS-glcAT.
The catalytic domain of UDP-glucuronyltransferase of mouse (GlcAT-P) was cloned by PCR amplification from brain cDNA as a template. A first amplification was done with a set of primers 5'-CTTGG- AGATGCCGAAGAGAC and 5'-GGGTCTGAGAAGGAGG- TTCC designed with primer3 software (Rozen and Skaletsky 2000). The PCR mix containing an amplicon of 1.04 kb was used as template for a second amplification with primers 5'-GATATCTTGAAGGAGATATACATATGCTGCCCACCAT- CCATGTGGTG and 5'-CTGCAGTCAGATCTCCACTGAG- GGGTCG, thus allowing the amplification of a 0.78 bp DNA product encoding the coding region Leu85-Ile336, flanked by a Shine–Dalgarno sequence and an initiation codon. It was digested with EcoRV and PstI and ligated into the EcoRV-PstI sites of plasmid pBlue Script (KS) to give pBS-glcAT.
Construction of pBS-glcAT-ugd.
The gene ugd UDP-glucose-6-dehydrogenase was cloned by PCR amplification from E.coli K-12 genomic DNA as a template. A first amplification was done with a set of primers 5'-TTGTTTTTCGTTGCGTTGA and 5'-AGGGCGTAAATAGCCCTGAT designed with primer3 software. The PCR mix containing the amplicon 1.3 kb was used as template for a second amplification with primers 5'-TCTAGACTGCAGGTTAATTCTGAGAGCATGAAATGAA- AATCAC and 5'-GAGCTCACTAGTTTAGTCGCTGCCAAA- GAGATCGC. The amplified 1.2 kb fragment was cloned in the pCR4 blunt vector to be subsequently subcloned into the PstI SpeI site of the pBS-glcAT plasmid.
Construction of pBS-glcAT-kfiD.
The gene kfiD encoding the UDP-GlcA dehydrogenase of E. coli strain K5 was obtained from plasmid pGPR105 (Petit et al. 1995). A single digestion with PstI was done in order to release a DNA fragment containing kfiD, and this was cloned into the unique PstI site of pBS-glcAT located downstream of the gene encoding GlcAT-P. A restriction map with SalI was done in order to select the constructions in which kfiD was cloned in the right direction.
Construction of pBS-KfiD-GlcAT.
The kfiD sequence was obtained from a NotI digestion of pGPR105, but the NotI ends were then filled with DNA polymerase I (a large fragment). It was thus cloned into the unique EcoRV site located upstream of the gene encoding GlcAT-P. A restriction map with SalI was done in order to select the constructions in which kfiD was cloned in the right direction.
Construction of pBBR-galT-rcsa.
The HP0826 gene encoding β-4GalT (Endo et al. 2000) was cloned by PCR using genomic DNA from H. pylori ATCC 26695 as a template and the primers : 5'-TAGGGCCCAGGAGGTTAAGCTATGCGTG- TTTTTGCCATTTCTTTAAATC and 5'-TCTCGAGTTATA- CAAACTGCCAATATTTCAAATATTTAAAATGG. The amplified 0.83 kb fragment was cloned into the pCR4 blunt vector to be subsequently subcloned into the ApaI PstI site of the plasmid pLNTR1T, which was a pBBR1-MCS3 derivative containing the rcsA gene (Dumon et al. 2001). Both lgtA and lgtB, which were located between the ApaI and PstI sites of pLNTR1T, were excised during cloning and the resulting plasmid pBBRGalT-rcsa contained the HP0826 gene upstream of rcsA.
Construction of pBBR-galT.
The construction was done by subcloning the β1,4-GalT enzyme gene from pBBRgalT-rcsa into the pBBR1-MCS3 vector again (Kovach et al. 1995) using the KpnI and PstI sites.
Construction of pWKSlgtA.
The 2.2 kb fragment containing the lgtB and lgtA genes in pLNTR1T was subcloned into the ApaI and PstI sites of the pWKS130 vector (Wang and Kushner 1991). The lgtB gene, which was flanked by two BamH1 sites, was then excised by a BamH1 digestion, yielding pWKSlgtA.
Production of oligosaccharides in high-cell-density culture
Cultures were carried out in 2-L reactors containing a mineral culture medium (1 L), as previously described (Priem et al. 2002). The high-cell-density culture consisted of three phases: an exponential-growth phase, which started with inoculation of the fermenter and lasted until exhaustion of the initially added glucose (17.5 g/L), a 5-h fed-batch phase with a high substrate-feeding rate (glycerol, 4.5 g/L/h) and a 19- to 48-h fed-batch phase with a lower feeding rate (glycerol, 2.4 g/L/h). The temperature was maintained at 34°C for the first phase and 28°C for the last two phases unless otherwise indicated in the text. The pH of the medium was regulated at 6.8 with 14% NH4OH. The acceptor (lactose 5 g/L and inducer (IPTG 50 mg) were added at the end of the exponential phase. For the culture of HN6 (Table I), a temperature of 28°C was maintained, and glucose was kept as the carbon and energy source for all three phases of the cell culture.
Purification of oligosaccharides
At the end of the fermentation period, the bacterial cells were recovered by centrifugation (5000 x g, 25 min). The supernatant was kept as the extracellular fraction. The cell pellets were resuspended in distilled water, and the cells were permeabilized by autoclaving at 100°C for 50 min. The mixture was then centrifuged (5000 x g, 25 min), and the supernatant was recovered as the intracellular fraction. The negatively charged oligosaccharides were separated from the neutral oligosaccharides by using Dowex 1 x 4 – 400 (HCO–3 form) resin. The samples were first fixed in the column and then recovered with a linear NaHCO3 gradient (0–1 M). The sodium bicarbonate was then eliminated by Dowex 50 x 4– 400 (H+ form) resin. The oligosaccharides were then separated by size-exclusion chromatography carried out on a HW40 Toyopearl (5 x 100 cm) column at 50°C with 100 mM NaHCO3 as the mobile phase and a flow rate of 35 mL/min.
Analysis of oligosaccharides
Thin Layer Chromatography.
Culture samples, collected at different culture intervals (1 mL), were centrifuged in microfuge tubes (3 min at 13,000 x g). The supernatants were kept as the extracellular fraction. The pellets were then re-suspended in distilled water (1 mL), boiled for 25 min, and centrifuged (3 min at 13,000 x g). The supernatant was kept as the intracellular oligosaccharides fraction. TLC-plate analyses were carried out on silica gels, and the oligosaccharides were eluted with n-butanol/acetic acid/water (2:1:1). Sugars were detected by dipping the plate in orcinol sulfuric reagent and heating.
Glucuronic Acid Content Determination.
The glucuronic acid content was determined by the m-hydroxydiphenyl method (Blumenkrantz and Asboe-Hansen 1973). The samples from the extracellular and intracellular fractions were diluted to 200 µL. After the addition of 1.2 mL sodium tetraborate (0.0125 M in H2SO4), the solutions were kept at 100°C for 5 min in the dark. Then m-hydroxydiphenyl was added (20 µL, 0.15% in 0.5% NaOH). The samples were vortexed until coloration appeared and analyzed at 
= 520 nm.
Glucuronidase Assay.
The ability of β-glucuronidase Type IX-A from E.coli (Sigma G7396–25 KU) to degrade the oligosacharides obtained in this study was assayed. The three purified oligosaccharides (2 mg/mL) were incubated with 31 units of the enzymes for 15 min, 30 min, 3 h, and 26 h at pH 7 in buffer Z with a final volume of 50 µL. The reaction was ended by the addition of 1M Na2CO3. One microliter of each sample was then subjected to TLC analysis as described above.
Carbohydrate Structural Analysis.
1H and 13C NMR spectra were obtained with a 300-MHz Bruker AVANCE spectrometer.
|
Supplementary Data |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary DataConflict of interest statementReferences |
|---|
Supplementary data for this article is available online at http://glycob.oxfordjournals.org.
|
Conflict of interest statement |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary DataConflict of interest statementReferences |
|---|
None declared.
|
Acknowledgements |
|---|
This research project has been supported by a Marie Curie Early Stage Research Training Fellowship of the European Community's Sixth Framework Programme, under contract number MEST-CT-2004-5033.
|
Abbreviations |
|---|
anti-MAG, antimyelin-associated glycoprotein; GlcALac, glucuronyllactose; GlcAnLc4, glucuronylneolactotetraose; GlcAnLc6, glucuronylneolactohexaose; GlcAT-P, mouse β-glucuronyltransferase; HNK-1, Human Natural Killer cell-1; Lac, lactose; nLc4, neolactotetraose; nLc6, neolactohexaose; SGGLs, sulfoglucuronyl glycolipids bearing the HNK-1 epitope; TLC, thin layer chromatography; Ugd, UDP-glucose dehydrogenase
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary DataConflict of interest statementReferences |
|---|
Blumenkrantz N, Asboe-Hansen G. New method for quantitative determination of uronic acids. Anal Biochem (1973) 54:484–489.[CrossRef][Web of Science][Medline]
Chevalier R, Colsh B, Afonso C, Baumann N, Tabet J-C, Mallet J-M. Synthetic sulfated glucuronosyl paragloboside (SGPG) and its use for the detection of autoimmune peripheral neuropathies. Tetrahedron (2006) 62:563–577.[CrossRef][Web of Science]
Dumon C, Priem B, Martin SL, Heyraud A, Bosso C, Samain E. In vivo fucosylation of neolactotetraose and neolactohexaose by heterologous expression of Helicobacter pylori alpha-1,3 fucosyltransferase in engineered Escherichia coli. Glycoconj J (2001) 18:465–474.[CrossRef][Web of Science][Medline]
Endo T, Koizumi S, Tabata K, Ozaki A. Cloning and expression of beta1,4-galactosyltransferase gene from Helicobacter pylori. Glycobiology (2000) 10:809–813.
Grangeasse C, Obadia B, Mijakovic I, Deutscher J, Cozzone AJ, Doublet P. Autophosphorylation of the Escherichia coli protein kinase Wzc regulates tyrosine phosphorylation of Ugd, a UDP-glucose dehydrogenase. J Biol Chem (2003) 278:39323–39329.
Jungalwala FB. Expression and biological functions of sulfoglucuronyl glycolipids (SGGLs) in the nervous system–a review. Neurochem Res (1994) 19:945–957.[CrossRef][Web of Science][Medline]
Kakuda S, Sato Y, Tonoyama Y, Oka S, Kawasaki T. Different acceptor specificities of two glucuronyltransferases involved in the biosynthesis of HNK-1 carbohydrate. Glycobiology (2005) 15:203–210.
Keenleyside WJ, Bronner D, Jann K, Jann B, Whitfield C. Coexpression of colanic acid and serotype-specific capsular polysaccharides in Escherichia coli strains with group II K antigens. J Bacteriol (1993) 175:6725–6730.
Kovach ME, Elzer PH, Hill DS, Robertson GT, Farris MA, Roop RM 2nd, Peterson KM. Four new derivatives of the broad-host-range cloning vector pBBR1MCS, carrying different antibiotic-resistance cassettes. Gene (1995) 166:75–176.
Link AJ, Phillips D, Church GM. Methods for generating precise deletions and insertions in the genome of wild-type Escherichia coli: Application to open reading frame characterization. J Bacteriol (1997) 179:6228–6237.
Martini R., Xin Y, Schmitz B, Schachner M. The L2/HNK-1 carbohydrate epitope is involved in the preferential outgrowth of motor neurons on ventral roots and motor nerves. Eur J Neurosci (1992) 4:628–639.[CrossRef][Web of Science][Medline]
Mouslim C, Groisman EA. Control of the Salmonella ugd gene by three two-component regulatory systems. Mol Microbiol (2003) 47:335–344.[CrossRef][Web of Science][Medline]
Petit C, Rigg GP, Pazzani C, Smith A, Sieberth V, Stevens M, Boulnois G, Jann K, Roberts IS. Region 2 of the Escherichia coli K5 capsule gene cluster encoding proteins for the biosynthesis of the K5 polysaccharide. Mol Microbiol (1995) 17:611–620.[CrossRef][Web of Science][Medline]
Priem B, Gilbert M, Wakarchuk WW, Heyraud A, Samain E. A new fermentation process allows large-scale production of human milk oligosaccharides by metabolically engineered bacteria. Glycobiology (2002) 12:235–240.
Rozen S, Skaletsky H. Primer3 on the WWW for general users and for biologist programmers. Methods Mol Biol (2000) 132:365–386.[Medline]
Samain E, Chazalet V, Geremia RA. Production of O-acetylated and sulfated chitooligosaccharides by recombinant Escherichia coli strains harboring different combinations of nod genes. J Biotechnol (1999) 72:33–47.[CrossRef][Web of Science][Medline]
Steck AJ, Stalder AK, Renaud S. Anti-myelin-associated glycoprotein neuropathy. Curr Opin Neurol (2006) 19:458–463.[Web of Science][Medline]
Stevenson G, Andrianopoulos K, Hobbs M, Reeves PR. Organization of the Escherichia coli K-12 gene cluster responsible for production of the extracellular polysaccharide colanic acid. J Bacteriol (1996) 178:4885–4893.
Tagawa H, Kizuka Y, Ikeda T, Itoh S, Kawasaki N, Kurihara H, Onozato ML, Tojo A, Sakai T, Kawasaki T, et al. A non-sulfated form of the HNK-1 carbohydrate is expressed in mouse kidney. J Biol Chem (2005) 280:23876–23883.
Wang RF, Kushner SR. Construction of versatile low-copy-number vectors for cloning, sequencing and gene expression in Escherichia coli. Gene (1991) 100:195–199.[CrossRef][Web of Science][Medline]
Weikkolainen K, Helin J, Niemelä R, Miller-Podraza H, Natunen J. Use of natural and synthetic oligosaccharide, neoglycolipid and glycolipid libraries in defining lectins from pathogens. In: Lectins: Analytical Technologies—Nilsson Carol, ed. (2007) 1st ed. Amsterdam (The Netherlands): Elsevier. p. 129–166.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||