A new fermentation process allows large-scale production of human milk oligosaccharides by metabolically engineered bacteria

Synthesis of α-galactosyl epitopes by metabolically engineered Escherichia coli

The α-Gal epitope is a carbohydrate structure, Galα-3Galβ-4GlcNAc-R, expressed on glycoconjuguates in many mammals, but not in humans. Species that do not express this epitope have present in their serum large amounts of natural anti-Gal antibodies, which contribute to organ hyperacute rejection during xenotransplantation. We first describe the efficient conversion of lactose into isoglobotriaose (Galα-3Galβ-4Glc) using high cell density cultures of a genetically engineered Escherichia coli strain expressing the bovine gene for α-1,3-galactosyltransferase. Attempts to produce the Galili pentasaccharide (Galα-3Galβ-4GlcNAcβ-3Galβ-4Glc) by additionally expressing the Neisseria meningitis lgtA gene for β-1,3-N-acetylglucosaminyltransferase and the Helicobacter pylori gene for β-1,4-galactosyltransferase were unsuccessful and led to the formation of a series of long chain oligosaccharides formed by the repeated addition of the trisaccharide motif [Galβ-4GlcNAcβ-3Galα-3] onto a lacto-N-neotetraose primer. The replacement of LgtA by a more specific β-1,3-N-acetylglucosaminyltransferase from H. pylori, which was unable to glycosylate α-galactosides, prevented the formation of these unwanted compounds and allowed the successful formation of the Galili pentasaccharide and longer α-Gal epitopes.

Profiles of human milk oligosaccharides and production of some human milk oligosaccharides in transgenic animals

During the decade of the 1990s and the first years of the current century, our group embarked on a project to study and synthesize human milk oligosaccharides. This report describes 2 unexpected collateral observations from that endeavor. The first observation was the detection and confirmation of 2 rare neutral human milk oligosaccharides profiles that were uncovered while assessing oligosaccharide content in hundreds of samples of human milk. One of these lacked fucosylated structures altogether, and the other lacked the oligosaccharide 3-fucosyllactose [Galβ1-4(Fucα1-3)Glc]. We used glycoconjugate probes to determine whether the unusual profiles were mirrored by fucosylation of milk glycoproteins. The results show that the lack of fucosylated oligosaccharides in these samples corresponds to the absence of equivalent fucosylated motifs in milk glycoproteins. The second finding was a shortened and distinct lactation process in transgenic rabbits expressing the human fucosyltransferase 1. During the first day of lactation, these animals expressed milk that contained both lactose and 2'-fucosylactose, but on the second day, the production of milk was severely diminished, and by the fourth day, no lactose was detected in their milk. Meanwhile, the concentration of fucosylated glycoproteins increased from the onset of lactation through its premature termination. These 2 findings may shed light on the glycobiology of milk and perhaps on mammary gland differentiation.

Sialic acid metabolism and sialyltransferases: natural functions and applications

Sialic acids are a family of negatively charged monosaccharides which are commonly presented as the terminal residues in glycans of the glycoconjugates on eukaryotic cell surface or as components of capsular polysaccharides or lipooligosaccharides of some pathogenic bacteria. Due to their important biological and pathological functions, the biosynthesis, activation, transfer, breaking down, and recycle of sialic acids are attracting increasing attention. The understanding of the sialic acid metabolism in eukaryotes and bacteria leads to the development of metabolic engineering approaches for elucidating the important functions of sialic acid in mammalian systems and for large-scale production of sialosides using engineered bacterial cells. As the key enzymes in biosynthesis of sialylated structures, sialyltransferases have been continuously identified from various sources and characterized. Protein crystal structures of seven sialyltransferases have been reported. Wild-type sialyltransferases and their mutants have been applied with or without other sialoside biosynthetic enzymes for producing complex sialic acid-containing oligosaccharides and glycoconjugates. This mini-review focuses on current understanding and applications of sialic acid metabolism and sialyltransferases.

Toward automated oligosaccharide synthesis

Carbohydrates have been shown to play important roles in biological processes. The pace of development in carbohydrate research is, however, relatively slow due to the problems associated with the complexity of carbohydrate structures and the lack of general synthetic methods and tools available for the study of this class of biomolecules. Recent advances in synthesis have demonstrated that many of these problems can be circumvented. In this Review, we describe the methods developed to tackle the problems of carbohydrate-mediated biological processes, with particular focus on the issue related to the development of the automated synthesis of oligosaccharides. Further applications of carbohydrate microarrays and vaccines to human diseases are also highlighted.

Glycomimicry: display of fucosylation on the lipo-oligosaccharide of recombinant Escherichia coli K12

We recently described the design of Escherichia coli K12 and Salmonella enterica sv Typhimurium to display the gangliomannoside 3 (GM3) antigen on the cell surface. We report here the fucosylation of modified lipooligosaccharide in a recombinant E.coli strain with a truncated lipid A core due to deletion of the core glycosyltransferases genes waaO and waaB. This truncated structure was used as a scaffold to assemble the Lewis Y motif by consequent action of the heterologously expressed β-1,4 galactosyltransferase LgtE (Neisseria gonorrheae), the β-1,3 N-acetylglucosaminyltransferase LgtA and the β-1,3 galactosyltransferase LgtB from Neisseria meningitidis, as well as the α-1,2 and α-1,3 fucosyltransferases FutC and FutA from Helicobacter pylori. We show the display of the Lewis Y structure by immunological and chemical analysis.

Marine bacterial sialyltransferases

Sialyltransferases transfer N-acetylneuraminic acid (Neu5Ac) from the common donor substrate of these enzymes, cytidine 5′-monophospho-N-acetylneuraminic acid (CMP-Neu5Ac), to acceptor substrates. The enzymatic reaction products including sialyl-glycoproteins, sialyl-glycolipids and sialyl-oligosaccharides are important molecules in various biological and physiological processes, such as cell-cell recognition, cancer metastasis, and virus infection. Thus, sialyltransferases are thought to be important enzymes in the field of glycobiology. To date, many sialyltransferases and the genes encoding them have been obtained from various sources including mammalian, bacterial and viral sources. During the course of our research, we have detected over 20 bacteria that produce sialyltransferases. Many of the bacteria we isolated from marine environments are classified in the genus Photobacterium or the closely related genus Vibrio. The paper reviews the sialyltransferases obtained mainly from marine bacteria.

Efficient synthesis of 6'-sialyllactose, 6,6'-disialyllactose, and 6'-KDO-lactose by metabolically engineered E. coli expressing a multifunctional sialyltransferase from the Photobacterium sp. JT-ISH-224

We have previously reported the efficient conversion of lactose into 3'-sialyllactose by high cell density cultures of a genetically engineered Escherichia coli strain expressing the Neisseria meningitidis gene for alpha-(2-->3)-sialyltransferase [Fierfort, N.; Samain, E. J. Biotechnol. 2008, 134, 261-265.]. First attempts to use a similar strategy to produce 6'-sialyllactose with a strain expressing alpha-(2-->6)-sialyltransferase from the Photobacterium sp. JT-ISH-224 led to the production of a trisaccharide that was identified as KDO-lactose (2-keto-3-deoxy-manno-octonyllactose). This result showed that alpha-(2-->6)-sialyltransferase was able to use CMP-KDO as sugar donor and preferentially used CMP-KDO over CMP-Neu5Ac. By reducing the expression level of the sialyltransferase gene and increasing that of the neuABC genes, we have been able to favour the formation of 6'-sialyllactose and to prevent the formation of KDO-lactose. However, in this case, a third lactose derivative, which was identified as 6,6'-disialyllactose, was also produced. Formation of 6,6'-disialyllactose was mainly observed under conditions of lactose shortage. On the other hand, when the culture was continuously fed with an excess of lactose, 6'-sialyllactose was almost the only product detected and its final concentration was higher than 30g/L of culture medium.

Advances in the biology and chemistry of sialic acids

Sialic acids are a subset of nonulosonic acids, which are nine-carbon alpha-keto aldonic acids. Natural existing sialic acid-containing structures are presented in different sialic acid forms, various sialyl linkages, and on diverse underlying glycans. They play important roles in biological, pathological, and immunological processes. Sialobiology has been a challenging and yet attractive research area. Recent advances in chemical and chemoenzymatic synthesis, as well as large-scale E. coli cell-based production, have provided a large library of sialoside standards and derivatives in amounts sufficient for structure-activity relationship studies. Sialoglycan microarrays provide an efficient platform for quick identification of preferred ligands for sialic acid-binding proteins. Future research on sialic acid will continue to be at the interface of chemistry and biology. Research efforts not only will lead to a better understanding of the biological and pathological importance of sialic acids and their diversity but also could lead to the development of therapeutics.

Assessment of the Two Helicobacter pylori α-1,3-Fucosyltransferase Ortholog Genes for the Large-Scale Synthesis of LewisX Human Milk Oligosaccharides by Metabolically Engineered Escherichia coli

We previously described a bacterial fermentation process for the in vivo conversion of lactose into fucosylated derivatives of lacto-N-neotetraose Gal(beta1-4)GlcNAc(beta1-3)Gal(beta1-4)Glc (LNnT). The major product obtained was lacto-N-neofucopentaose-V Gal(beta1-4)GlcNAc(beta1-3)Gal(beta1-4)[Fuc(alpha1-3)]Glc, carrying fucose on the glucosyl residue of LNnT. Only a small amount of oligosaccharides fucosylated on N-acetylglucosaminyl residues and thus carrying the LewisX group (Le(X)) was also produced. We report here a fermentation process for the large-scale production of Le(X) oligosaccharides. The two fucosyltransferase genes futA and futB of Helicobacter pylori (strain 26695) were compared in order to optimize fucosylation in vivo. futA was found to provide the best activity on the LNnT acceptor, whereas futB expressed a better Le(X) activity in vitro. Both genes were expressed to produce oligosaccharides in engineered Escherichia coli (E. coli) cells. The fucosylation pattern of the recombinant oligosaccharides was closely correlated with the specificity observed in vitro, FutB favoring the formation of Le(X) carrying oligosaccharides. Lacto-N-neodifucohexaose-II Gal(beta1-4)[Fuc(alpha1-3)]GlcNAc(beta1-3)Gal(beta1-4)[Fuc(alpha1-3)]Glc represented 70% of the total oligosaccharide amount of futA-on-driven fermentation and was produced at a concentration of 1.7 g/L. Fermentation driven by futB led to equal amounts of both lacto-N-neofucopentaose-V and lacto-N-neofucopentaose-II Gal(beta1-4)[Fuc(alpha1-3)]GlcNAc(beta1-3)Gal(beta1-4)Glc, produced at 280 and 260 mg/L, respectively. Unexpectedly, a noticeable proportion (0.5 g/L) of the human milk oligosaccharide 3-fucosyllactose Gal(beta1-4)[Fuc(alpha1-3)]Glc was produced in futA-on-driven fermentation, underlining the activity of fucosyltransferase FutA in E. coli and leading to a reassessment of its activity on lactose. All oligosaccharides produced by the products of both fut genes were natural compounds of human milk.

Negative ion mode matrix-assisted laser desorption/ionisation time-of-flight mass spectrometric analysis of oligosaccharides using halide adducts and 9-aminoacridine matrix

9-Aminoacridine was established as a matrix for the detection of neutral oligosaccharides in negative ion mode matrix-assisted laser desorption/ionisation time-of-flight (MALDI-TOF) mass spectrometry. Sodium iodide proved to be a useful additive inducing formation of stable iodide adducts of the analytes, in particular for oligosaccharides with a degree of polymerisation (DP) of three and higher. Lower oligomers (DP <3) and monosaccharides show more stable adducts with chloride ions. After optimisation of the sample preparation procedure, limits of quantitation were determined for alpha-cyclodextrin and cellopentaose at 7 and 13 pmol, respectively, with a linear detector response over two concentration orders. The iodide additive could be successfully employed on MALDI-TOF mass spectrometers with vacuum and atmospheric pressure ion sources. The value of the new method to solve biological problems has been demonstrated by the analysis of a mixture of beta-glucane elicitors isolated from the cell walls of Phytophthora sojae.

Exploring specificity of glycosyltransferases: synthesis of new sugar nucleotide related molecules as putative donor substrates

We investigated the specificity of glycosyltransferases toward donor substrates in two complementary directions. First we prepared simple N-acetyl-alpha-D-glucosamine 1-diphosphates: methyl-(2-acetamido-2-deoxy-alpha-D-glucopyranosyl)-diphosphate, benzyl-(2-acetamido-2-deoxy-alpha-D-glucopyranosyl)-diphosphate, 4-phenylbutyl-(2-acetamido-2-deoxy-alpha-D-glucopyranosyl)-diphosphate, by the coupling of the corresponding activated alkyl phosphates with N-acetyl-alpha-D-glucosamine 1-phosphate. These diphosphates as well as 2-acetamido-2-deoxy-alpha-D-glucopyranose 1-diphosphate, tested as donors of N-acetylglucosamine in a reaction catalyzed by Neisseria meningitidis N-acetylglucosaminyltransferase (LgtA), proved to be devoid of activity. Evaluated as inhibitors, only 2-acetamido-2-deoxy-alpha-D-glucopyranose 1-diphosphate showed some inhibitory activity with an IC50 value of 7 mM. In the second approach, we prepared sugar nucleotide mimics having the diphosphate bridge replaced by the oxycarbonylaminosulfonyl linker. The surrogate of GDP-Fuc was synthesized as a 9:1 alpha/beta anomeric mixture, in 40% yield, starting from chlorosulfonyl isocyanate, perbenzylated l-fucopyranose, and a guanosine derivative, protected on the exocyclic amine and secondary hydroxyl functions of ribose. Then two deprotection steps, hydrogenolysis and enzymatic hydrolysis catalyzed by penicillin G amidase afforded the target molecule to be tested as fucose donor with recombinant human alpha-(1-->3/4)-fucosyltransferase (FucT-III). Tested as a 4:1 alpha/beta anomeric mixture, both in the absence and in the presence of cationic cofactors, this new guanosine fucose conjugate proved to be ineffective. Its inhibitory activity toward FucT-III evaluated through a competition fluorescence assay was very poor (IC50 value of 20 mM). The surrogate of UDP-GlcNAc that was already known as its protected acetylated derivative, tested as N-acetylglucosamine donor with LgtA in the presence of Mn(2+) turned out not to be active either.

Synthesis of globopentaose using a novel β1,3-galactosyltransferase activity of theHaemophilus influenzaeβ1,3-N-acetylgalactosaminyltransferase LgtD

We have previously described a bacterial system for the conversion of globotriaose (Gb3) into globotetraose (Gb4) by a metabolically engineered Escherichia coli strain expressing the Haemophilus influenzae lgtD gene encoding beta1,3-N-acetylgalactosaminyltransferase [Antoine, T., Bosso, C., Heyraud, A. Samain, E. (2005) Large scale in vivo synthesis of globotriose and globotetraose by high cell density culture of metabolically engineered Escherichia coli. Biochimie 87, 197-203]. Here, we found that LgtD has an additional beta1,3-galactosyltransferase activity which allows our bacterial system to be extended to the synthesis of the carbohydrate portion of globopentaosylceramide (Galbeta-3GalNAcbeta-3Galalpha-4Galbeta-4Glc) which reacts with the monoclonal antibody defining the stage-specific embryonic antigen-3. In vitro assays confirmed that LgtD had both beta1,3-GalT and beta1,3-GalNAcT activities and showed that differences in the affinity for Gb3 and Gb4 explain the specific and exclusive formation of globopentaose.

Metabolic engineering of microbes for oligosaccharide and polysaccharide synthesis

Metabolic engineering has recently been embraced as an effective tool for developing whole-cell biocatalysts for oligosaccharide and polysaccharide synthesis. Microbial catalysts now provide a practical means to derive many valuable oligosaccharides, previously inaccessible through other methods, in sufficient quantities to support research and clinical applications. The synthesis process based upon these microbes is scalable as it avoids expensive starting materials. Most impressive is the high product concentrations (up to 188 g/L) achieved through microbe-catalyzed synthesis. The overall cost for selected molecules has been brought to a reasonable range (estimated $ 30–50/g).

Microbial synthesis of oligosaccharides and polysaccharides is a carbon-intensive and energy-intensive process, presenting some unique challenges in metabolic engineering. Unlike nicotinamide cofactors, the required sugar nucleotides are products of multiple interacting pathways, adding significant complexity to the metabolic engineering effort. Besides the challenge of providing the necessary mammalian-originated glycosyltransferases in active form, an adequate uptake of sugar acceptors can be an issue when another sugar is necessary as a carbon and energy source. These challenges are analyzed, and various strategies used to overcome these difficulties are reviewed in this article. Despite the impressive success of the microbial coupling strategy, there is a need to develop a single strain that can achieve at least the same efficiency. Host selection and the manner with which the synthesis interacts with the central metabolism are two important factors in the design of microbial catalysts. Additionally, unlike in vitro enzymatic synthesis, product degradation and byproduct formation are challenges of whole-cell systems that require additional engineering. A systematic approach that accounts for various and often conflicting requirements of the synthesis holds the key to deriving an efficient catalyst.

Metabolic engineering strategies applied to selected polysaccharides (hyaluronan, alginate, and exopolysaccharides for food use) are reviewed in this article to highlight the recent progress in this area and similarity to challenges in oligosaccharide synthesis. Many naturally occurring microbes possess highly efficient mechanisms for polysaccharide synthesis. These mechanisms could potentially be engineered into a microbe for oligosaccharide and polysaccharide synthesis with enhanced efficiency.

Non-isosteric C-glycosyl analogues of natural nucleotide diphosphate sugars as glycosyltransferase inhibitors

A series of C-glycosyl ethylphosphonophosphate analogues of UDP-Glc, UDP-Gal, UDP-GlcNAc and GDP-Fuc were synthesized from the corresponding C-glycosyl ethylphosphonic acids. Analogues were obtained as alpha-anomers through either diastereoselective photo-induced radical addition of glycosyl bromides (D-Glc, D-Gal and L-Fuc) to diethyl vinylphosphonate, or a multi-step sequence (D-GlcNAc), with subsequent coupling with morpholidate-activated nucleotide monophosphates. The in vitro inhibitory activity of UDP-Gal, GDP-Fuc and UDP-GlcNAc analogues towards glycosyltransferases (beta-1,4-GalT, FUT3 and LgtA) was evaluated through a competition fluorescence assay and IC(50) values of 40 microM, 2 mM and 3.5 mM were obtained, respectively.

Production of Lewis x Tetrasaccharides by Metabolically Engineered Escherichia coli

Two tetrasaccharides carrying the trisaccharidic Lewis x motif on a GlcNAc or a Gal residue were produced on the gram-scale by high-cell-density cultures of metabolically engineered Escherichia coli strains that overexpressed the Helicobacter pylori futA gene for alpha-3 fucosyltransferase and the Neisseria meningitidis lgtB gene for beta-4 galactosyltransferase. The first compound Galbeta-4(Fucalpha-3)GlcNAcbeta-4GlcNAc was produced by glycosylation of chitinbiose, which was endogenously generated in the bacterial cytoplasm by the successive action of the rhizobial chitin-synthase NodC and the Bacillus circulans chitinase A1, whose genes were additionally expressed in the E. coli strain. The second compound, Galbeta-4(Fucalpha-3)GlcNAcbeta-3Gal, was produced from exogenously added Gal by a strain that was deficient in galactokinase activity and overexpressed the additional N. meningitidis lgtA gene for beta-3 N-acetylglucosaminyltransferase.

Genetic engineering of Escherichia coli for the production of NI,NII-diacetylchitobiose (chitinbiose) and its utilization as a primer for the synthesis of complex carbohydrates

Chitinbiose was produced at more than 4 g L-1 by a high cell density culture of an Escherichia coli strain that co-expressed the rhizobial chitinoligosaccharide synthase gene nodC and a truncated form of the chitinase gene chiA which has been designed to be functionally produced in the E. coli cytoplasm. Chitinpentaose, which has previously been shown to be produced by the nodC protein in growing E. coli, was formed as an intermediate that was subsequently hydrolyzed into chitinbiose by the chitinase encoded by chiA. Chitinbiose was mainly recovered in the extracellular medium and to prevent its catabolism, the genes for the chitinbiose PTS permease had to be disrupted. When the additional gene lgtB for beta1,4-galactosyltransferase was expressed, intracellular chitinbiose was converted into the trisaccharide Galbeta-4GlcNAcbeta-4GlcNAc which could serve as acceptor for glycosyltransferase that recognize the terminal N-acetyllactosamine structure.

Synthesis of unnatural sugar nucleotides and their evaluation as donor substrates in glycosyltransferase-catalyzed reactions

New unnatural sugar nucleotides, UDP-Fuc and CDP-Fuc were synthesized from fucose-beta-1-phosphate and nucleotide monophosphates activated as morpholidates. Furthermore, a nucleotide analogue was prepared by phosphorylation of 1-(beta-D-ribofuranosyl)cyanuric acid, itself obtained as a protected derivative by condensation of the persilylated derivative of cyanuric acid with 1-O-acetyl-2,3,5-tri-O-benzoyl-beta-D-ribofuranose in 74% yield. This phosphate activated according to the same procedure was condensed with fucose-beta-1-phosphate, affording a new sugar nucleotide conjugate (NDP-Fuc) which was evaluated together with UDP-Fuc, CDP-Fuc and ADP-Fuc, as fucose donors in alpha-(1-->4/3)-fucosyltransferase (FucT-III) catalyzed reaction. Fucose transfer could be observed with each of the donors and kinetic parameters were determined using a fluorescent acceptor substrate. Efficiency of the four analogues towards FucT-III was in the following order: UDP-Fuc=ADP-Fuc>NDP-Fuc>CDP-Fuc. According to the same strategy ADP-GlcNAc was prepared from AMP-morpholidate and N-acetylglucosamine-alpha-1-phosphate; tested as a glucosaminyl donor towards Neisseria meningitidis N-acetylglucosaminyl transferase (LgtA), ADP-GlcNAc was recognized with 0.1% efficiency as compared with UDP-GlcNAc, the natural donor substrate.

Large scale in vivo synthesis of globotriose and globotetraose by high cell density culture of metabolically engineered Escherichia coli

Large amounts of globotriose (Galalpha-4Galbeta-4Glc) are shown to be produced by the high cell density culture of an Escherichia coli strain over-expressing the Neisseria meningitidis lgtC gene for alpha-1,4-Gal transferase. The strain which was devoid of both alpha and beta galactosidase activity was fed with glycerol as the energy and carbon source and with lactose as precursor for globotriose synthesis. After complete exhaustion of lactose, globotriose could serve as an alternative acceptor for LgtC and the formation of a series of polygalactosylated compounds was observed. The system was extended to the synthesis of globotetraose (GalNAcbeta-3Galalpha-4Galbeta-4Glc) by overexpressing two additional genes: lgtD from Haemophilus influenzae Rd which encodes a beta-1,3-GalNAc transferase and wbpP from Pseudomonas aeruginosa which encodes a UDP-GalNAc C4 epimerase. Globotetraose could also be produced from exogenous globotriose which was shown to be actively taken up by the cells.

Enzymatic synthesis of poly-N-acetyllactosamines as potential substrates for endo-β-galactosidase-catalyzed hydrolytic and transglycosylation reactions

Enzymatic synthesis of GlcNAc-terminated poly-N-acetyllactosamine beta-glycosides GlcNAcbeta1,3(Galbeta1,4GlcNAcbeta1,3)(n)Galbeta1,4GlcNAcbeta-pNP (n=1-4) was demonstrated using a transglycosylation reaction of Escherichia freundii endo-beta-galactosidase. The enzyme catalyzed a transglycosylation reaction on GlcNAcbeta1,3Galbeta1,4GlcNAcbeta-pNP (1), which served both as a donor and an acceptor, and converted 1 into p-nitrophenyl beta-glycosides GlcNAcbeta1,3(Galbeta1,4GlcNAcbeta1,3)(1)Galbeta1,4GlcNAcbeta-pNP (2), GlcNAcbeta1,3(Galbeta1,4GlcNAcbeta1,3)(2)Galbeta1,4GlcNAcbeta-pNP (3), GlcNAcbeta1,3(Galbeta1,4GlcNAcbeta1,3)(3)Galbeta1,4GlcNAcbeta-pNP (4) and GlcNAcbeta1,3(Galbeta1,4GlcNAcbeta1,3)(4)Galbeta1,4GlcNAcbeta-pNP (5). When 2 was used as an initial substrate, it led to the preferential synthesis of nonasaccharide beta-glycoside 4 to heptasaccharide beta-glycoside 3. This suggests that 4 is directly synthesized by transferring the tetrasaccharide unit GlcNAcbeta1,3Galbeta1,4GlcNAcbeta1,3Gal to nonreducing end GlcNAc residue of 2 itself. The efficiency of production of poly-N-acetyllactosamines by E. freundii endo-beta-galactosidase was significantly enhanced by the addition of BSA and by a low-temperature condition. Resulting 2 and 3 were shown to be useful for studying endo-beta-galactosidase-catalyzed hydrolytic and transglycosylation reactions.

Biosynthesis of conjugatable saccharidic moieties of GM2 and GM3 gangliosides by engineered E. coli

Oligosaccharidic moieties of GM(2) and GM(3) gangliosides bearing an allyl or a propargyl aglycon, are efficiently biosynthesized on the gram scale by growing metabolically engineered Escherichia coli cells in the presence of the corresponding lactoside acceptors and sialic acid.

Combining fold recognition and exploratory data analysis for searching for glycosyltransferases in the genome of Mycobacterium tuberculosis

Fold recognition was applied to the systematic analysis of the all sequences encoded by the genome of Mycoplasma tuberculosis H37Rv in order to identify new putative glycosyltransferases. The search was conducted against a library composed of all known crystal structures of glycosyltransferases and some related proteins. A clear relationship appeared between some sequences and some folds. It appears necessary to complete the fold recognition approach with a statistical approach in order to identify the relevant data above the background noise. Exploratory data analysis was carried out using several methods. Analytical methods confirmed the validity of the approach, while predictive methods, although very preliminary in the present case, allowed for identifying a number of sequences of interest that should be further investigated. This new approach of combining bioinformatics and chemometrics appears to be a powerful tool for analysis of newly sequenced genomes. Its application to glycobiology is of great interest.

Large-scale in vivo synthesis of the carbohydrate moieties of gangliosides GM1 and GM2 by metabolically engineered Escherichia coli

Two metabolically engineered Escherichia coli strains have been constructed to produce the carbohydrate moieties of gangliosides GM2 (GalNAcbeta-4(NeuAcalpha-3)Galbeta-4Glc; Gal = galactose, Glc = glucose, Ac = acetyl) and GM1 (Galbeta-3GalNAcbeta-4(NeuAcalpha-3)Galbeta-4Glc. The GM2 oligosaccharide-producing strain TA02 was devoid of both beta-galactosidase and sialic acid aldolase activities and overexpressed the genes for CMP-NeuAc synthase (CMP = cytidine monophosphate), alpha-2,3-sialyltransferase, UDP-GlcNAc (UDP = uridine diphosphate) C4 epimerase, and beta-1,4-GalNAc transferase. When this strain was cultivated on glycerol, exogenously added lactose and sialic acid were shown to be actively internalized into the cytoplasm and converted into GM2 oligosaccharide. The in vivo synthesis of GM1 oligosaccharide was achieved by taking a similar approach but using strain TA05, which additionally overexpressed the gene for beta-1,3-galactosyltransferase. In high-cell-density cultures, the production yields for the GM2 and GM1 oligosaccharides were 1.25 g L(-1) and 0.89 g L(-1), respectively.

Prebiotics and synbiotics: towards the next generation

Recent research in the area of prebiotic oligosaccharides and synbiotic combinations with probiotics is leading towards a more targeted development of functional food ingredients. Improved molecular techniques for analysis of the gut microflora, new manufacturing biotechnologies, and increased understanding of the metabolism of oligosaccharides by probiotics are facilitating development. Such developments are leading us to the time when we will be able to rationally develop prebiotics and synbiotics for specific functional properties and health outcomes.

In vivo fucosylation of lacto-N-neotetraose and lacto-N-neohexaose by heterologous expression of Helicobacter pylori alpha-1,3 fucosyltransferase in engineered Escherichia coli

We report here the in vivo production of type 2 fucosylated-N-acetyllactosamine oligosaccharides in Escherichia coli. Lacto-N-neofucopentaose Galbeta1-4GlcNAcbeta1-3Galbeta1-4(Fucalpha1-3)Glc, lacto-N-neodifucohexaose Galbeta1-4(Fucalpha1-3)Glc-NAcbeta1-3Galbeta1-4(Fucalpha1-3)Glc, and lacto-N-neodifucooctaose Galbeta1-4GlcNAcbeta1-3Galbeta1-4(Fucalpha1-3)GlcNAcbeta1-3Galbeta1-4(Fucalpha1-3)Glc were produced from lactose added in the culture medium. Two of them carry the Lewis X human antigen. High cell density cultivation allowed obtaining several grams of fucosylated oligosaccharides per liter of culture. The fucosylation reaction was catalyzed by an alpha-1,3 fucosyltransferase of Helicobacter pylori overexpressed in E. coli with the genes lgtAB of N. meningitidis. The strain was genetically engineered in order to provide GDP-fucose to the system, by genomic inactivation of gene wcaJ involved in colanic acid synthesis and overexpression of RcsA, positive regulator of the colanic acid operon. To prevent fucosylation at the glucosyl residue, lactulose Galbeta1-4Fru was assayed in replacement of lactose. Lactulose-derived oligosaccharides carrying fucose were synthesized and characterized. Fucosylation of the fructosyl residue was observed, indicating a poor acceptor specificity of the fucosyltransferase of H. pylori.