Patent Application
20110144317
The present invention relates to mutants of glycoside hydrolases and uses thereof in chemo-enzymatic synthesis of complex oligosaccharides, in particular fragments of S. flexneri 2b, 3a, 5a, 5b, and X O-antigens.
Carbohydrates displayed at the surface of cells and pathogens are involved in a wide range of biological processes, among which several intercellular recognition events, as well as host-pathogen interactions possibly resulting in microbial or viral infections. The understanding of the molecular events involved in carbohydrate-mediated interactions has long been impaired by the difficult access to relevant oligosaccharides and glycoconjugates in pure form and sufficient amounts. The exquisite diversity of possible structures, varying in monosaccharide composition, linkage and branching pattern (ref. 1), is indeed a major roadblock to easy availability. In recent years however, important developments in the preparation of carbohydrate derivatives, based on (i) multistep chemical synthesis, (ii) enzymatic strategies using recombinant glycosyltransferases (i.e., enzymes catalyzing the transfer of a monosaccharide residue from an activated sugar phosphate to an acceptor molecule; EC 2.4) with in situ regeneration of sugar nucleotides, (iii) combinations thereof, or (iv) biosynthesis using metabolically engineered cell-factories (ref. 2), have opened the way to significant progress in the fields of glycobiology and glycotherapeutics (ref. 3-5). A number of efficient and elegant synthetic methods such as one-pot oligosaccharide synthesis (ref. 6) or automated synthetic tools (ref 7) have been developed to provide more straightforward access to structurally-defined carbohydrates. The use of lightly protected precursors and the regioselective one-pot protection of monosaccharides were recently emphasized (ref 8).
Nevertheless, despite accomplished advancements, chemical approaches towards specific usable microbial oligosaccharides still need considerable effort. They remain, for the most part, highly dependent on the design of appropriate combinations of multiple protection, deprotection, and efficient glycosylation steps, which often involve numerous tedious chromatographic separations (ref 9). Avoiding the need for protecting groups, organisms engineered to express several glycosyltransferase genes have been used to produce a nice range of biologically active complex carbohydrates, but they remain to date limited to the synthesis of short oligosaccharides which can passively cross the cell membrane (ref. 2, 10).
Following the early success of polysaccharide vaccines in the second half of the 20th century, polysaccharide-protein conjugate vaccines were seen as a major progress in antibacterial vaccination (ref. 11, 12). Indeed, made from bacterial polysaccharides purified from pathogen cell cultures, eventually shortened following partial-chemical hydrolysis or enzymatic depolymerisation of the native antigen, and subsequently covalently coupled to a protein carrier, these second generation carbohydrate vaccines are suitable for use in human (ref. 12). Potential extrapolations are numerous since for a large number of pathogens, surface carbohydrates behave as key “protective antigens”. Interestingly, this long known property of a range of bacterial polysaccharides was extended in recent years to other microbial carbohydrates of fungal (ref. 13) and parasitic origin (ref. 14). Besides, the disclosure that cancer cells could among other features be differentiated from healthy ones by the presence of surface glycoconjugates, often termed tumor-associated carbohydrate antigens, contributed to additional interest in carbohydrate antigens (ref 15). Overall, interest in synthetic carbohydrate-based vaccines has emerged as one amongst the many exploding fields of carbohydrate medical applications (ref 15, 16). The development of synthetic microbial carbohydrate-protein conjugates, thus termed third generation carbohydrate vaccines, was proposed as an alternative to conventional polysaccharide vaccines, compatible with the increasingly demanding requirements in terms of safety, efficiency, product definition, and needs for multivalent vaccines (ref. 17). Most interestingly, use of the natural polysaccharide, and consequently risks associated to antigens of biological origin, are avoided. However, chemical synthesis of carbohydrate-protein conjugates, more precisely of appropriate carbohydrate haptens is seen as a drawback. By way of example,
The use of enzymes as catalysts has emerged as a practical alternative to a number of limitations encountered in chemical synthesis (ref. 18, 19). Leloir-type glycosyltransferases and transglycosidases constitute the two major classes of enzymes that can be used for the synthesis of glycosidic linkages. Both are enzymes transferring a glycosyl group from a donor to an acceptor. Glycosyltransferases require nucleotide sugar as donor substrate whereas glycosidases (also called glycoside hydrolases) usually employ mono- and/or oligosaccharides as donor substrates.
The term “donor” refers to a molecule that provides a glycosyl moiety which will be transferred to an acceptor molecule.
The term “acceptor” refers to a molecule that will receive the glycosyl moiety through the formation of a chemical bond, preferentially C—O-linkage.
However, despite the increasing number of available transglycosidases and glycosyltransferases (Leloir type), the lack of appropriate enzymatic tools with requisite substrate specificity has prevented extensive exploration of chemo-enzymatic strategies when dealing with complex bacterial carbohydrate antigens. The use of multiple overexpressed native glycosyltransferases was shown to be highly rewarding for the synthesis of the upstream pentasaccharide terminus of the Neisseria meningitidis lipo-oligosaccharide (ref. 20), but examples of enzymatic synthesis of complex carbohydrates remain scarce (ref. 21). Indeed, certain membrane Leloir-type glycosyltransferases are not easily available. Their nucleotide activated sugar substrates are expensive. They may be generated in situ, but the process necessitates additional enzymes (ref. 21).
Replacement of glycosyltransferases by transglycosidases has been proposed to proceed from different types of glycosyl donors, and to be compatible with a larger variety of acceptors (ref. 22). Interestingly, modified donors were occasionally used successfully (ref 23). Nonetheless, the availability of these enzymes is often critical, especially when considering the appropriate regio- and stereospecificity required for a given target (ref 22, 24 and 25).
Protein engineering based on rational, semi-rational or fully combinatorial approaches (directed evolution) has also proven to be extremely useful to generate catalysts with improved natural properties but also to create new substrate specificities (ref 26, 27). In the field of carbohydrate-enzymes, glycosyltransferase substrate specificity has been successfully modified by site-directed mutagenesis assisted by computational modeling or directed evolution for the synthesis of biologically relevant carbohydrate structure (ref 27). Promiscuous glycosidases showing altered and new specificities towards the donor or the acceptor sugar have been generated (ref 28-31). Engineering of new glycosynthases (i.e., enzymes catalyzing the condensation of sugar residues for synthesizing a glycoside) from β-glycosidases (classified into EC 3.2.1) also emerged as a powerful way to generate modified transferases, even if they use only fluoride donors (ref 26, 32, 33). However, this methodology is difficult to apply to α-retaining enzymes. Interestingly, only one active glycosynthase derived from an α-retaining enzyme has been described up to now, and there is yet no description of substrate acceptance enlargement for α-retaining transglycosidases (or glycosynthases) (ref 79).
In addition, enzymatic glycosylation towards complex carbohydrates, relying on compatibility between enzyme and acceptor, mostly involves fully deprotected acceptors and thus comes at the latest stages in a chemo-enzymatic synthetic process (ref 21, 34 and 35).
Interestingly, the use of intermediates issued from enzymatic glycosylation in the subsequent generation of glycoconjugates of higher complexity has also been reported (ref 36-38). In all cases, the building blocks, conceived by action of native glycosyltransferases, were converted to donors and used as such, following peracylation.
The term “building block” refers to a suitably protected carbohydrate intermediate occurring in the chemical pathway of synthesis of complex oligosaccharides, e.g., said carbohydrate can be a disaccharide.
The term “intermediate” refers to a compound, protected or not, issued from an enzymatic and/or synthetic step, and involved in the multi-step synthesis of a specific target, e.g., said compound can be a disaccharide.
The design of an appropriate enzymatic glycosylation tool that would allow an optimal combination of the chemical and enzymatic steps involved in the synthesis of complex oligosaccharides has never been attempted although it would be of major interest to develop new synthetic pathways.
The Inventors have thus investigated the applicability of enzymatic glycosylation for the synthesis of building blocks compatible with chemical chain extension both at the reducing and non-reducing ends that is compatible with subsequent conversion into donors as well as acceptors. In the course of their investigation, the Inventors have surprisingly demonstrated the applicability of engineered amylosucrases (AS)—which are glycoside hydrolases—to the synthesis of oligosaccharide fragments of Shigella flexneri lipopolysaccharide by in-vitro chemo-enzymatic synthetic methodologies, implicating an enzymatic step at an early stage in the synthesis.
Amylosucrases (EC 2.4.1.4) as well as sucrose hydrolases (EC 3.2.1.-) belong to the family 13 of the glycoside hydrolases (GH13), and more particularly the subfamily 4 (GH13.4) as defined per the CAZY nomenclature (ref. 66-69). Amylosucrases and sucrose hydrolases operate on the same substrate (sucrose) with the same molecular mechanism (ref. 80). The difference between the amylosucrases and the sucrose hydrolases is only different transglycosylation abilities (ref. 69).
The structure of amylosucrase from Neisseria polysaccharea is the only known structure of enzymes from family GH13.4 (ref. 81). The single polypeptide chain (628 amino acid residues) of amylosucrase from Neisseria polysaccharea is folded into a tertiary structure consisting of five domains named N (residues 1-90), A (residues 98-184; 261-395; 461-550), B (residues 185-260), B′ (residues 395-460) and C (residues 555-628). Domains A, B and C are common domains found in family GH13. Domains N and B′ are specific to family GH13.4. Domain N is the N-terminal domain composed of 6 α-helices. Domain A is made up of eight alternating β-sheets (β1-β8) and α-helices (α1-α8) building up the catalytic core: the (β/α)8 barrel common to family GH13. It contains also eight loops connecting helices to strands (labelled loop1 to loop8). Domain B, or loop 3, is an extension of domain A, containing two short antiparallel β-sheets flanked by two α-helices. Domain B′, or loop 7, is another extension of domain A, composed of two α-helices followed by a β-sheet and another short α-helice. Domain C is an eight-stranded β-sandwich found C-terminal to the (β/α)8 barrel.
Unexpectedly, the Inventors have now found eleven consensus amino acid sequences to characterize glycoside hydrolases: eight consensus motifs defined hereafter (SEQ ID NO: 1; SEQ ID NO: 2; SEQ ID NO: 3; SEQ ID NO: 6; SEQ ID NO: 7; SEQ ID NO: 8; SEQ ID NO: 10; SEQ ID NO: 11) are localised in said β-sheets (6) or said loops (2) constituting domain A, two consensus motifs (SEQ ID NO: 4; SEQ ID NO: 5) are found in said domain B and one consensus motifs (SEQ ID NO: 9) is found is said domain B′.
Shigella is the causal agent of shigellosis, or bacillary dysentery. In developing countries, it induces about 1 million deaths per year, most of which involve children under five years of age (ref. 39). In countries where disease is endemic, a number of S. flexneri serotypes and to a lesser extent S. sonnei are isolated, emphasizing the need for a multivalent vaccine. Noteworthy, despite numerous clinical trials (ref. 40), no vaccine is available so far. Epidemiological as well as experimental data point to the polysaccharide part, or O-antigen (O—Ag), of the bacterial lipopolysaccharide as an important virulence factor (ref. 41) and the major target of protective humoral response against reinfection (ref. 42). S. flexneri is divided into 14 serotypes based on known O—Ag structures. Interestingly, protein-conjugates of short synthetic oligosaccharides mimicking S. flexneri 2a O—Ag induced in mice a potent anti-O—Ag humoral immune response, which was shown to be protective in a murine model of infection (ref. 43). The diversity, associated to a close resemblance in composition, of the known S. flexneri O—Ag repeating units was found of utmost interest to challenge the investigation. Indeed, except for serotype 6, all S. flexneri O—Ag repeating units share a linear tetrasaccharide backbone (ref. 41). Diversity resides in the branching pattern, which involves O-acetyl and/or α-D-glucopyranosyl decorations (ref. 41, 44). Interestingly, at least 4 different patterns of α-D-glucosylation, have been characterized for this family of bacterial polysaccharides. L-Rhamnopyranose residue is implicated as branching acceptor. By way of example, S. flexneri 2b, 3a, 5a, 5b, and X O—Ags have the α-D-glucopyranosyl-(1→3)-α-L-rhamnopyranosyl (EA) branching pattern in common:
2b O-antigen: 2)-[α-D-Glcp(1→3)]-α-L-Rhap-(1→2)-α-L-Rhap-(1→3)-[α-D-Glcp(1→4)]-α-L-Rhap-(1→3)-β-D-GlcpNAc-(1→
3a O-antigen: 2)-[α-D-Glcp(1→3)]-α-L-Rhap-(1→2)-α-L-Rhap-(1→3)-[2Ac]-α-L-Rhap-(1→3)-β-D-GlcpNAc-(1→
5a O-antigen: 2)-α-L-Rhap-(1→2)-[α-D-Glcp(1→3)]-α-L-Rhap-(1→3)-α-L-Rhap-(1→3)-β-D-GlcpNAc-(1→
5b O-antigen: 2)-[α-D-Glcp(1→3)]-α-L-Rhap-(1→2)-[α-D-Glcp(1→3)]-α-L-Rhap-(1→3)-α-L-Rhap-(1→3)-β-D-GlcpNAc-(1→
X O-antigen: 2)-[α-D-Glcp(1→3)]-α-L-Rhap-(1→2)-α-L-Rhap-(1→3)-α-L-Rhap-(1→3)-β-D-GlcpNAc-(1→
Abbreviations: Glcp=Glucopyranosyl, Rhap=Rhamnopyranosyl, GlcNAcp=N-acetyl-Glucosaminopyranosyl, Ac=acetyl
Here-under are examples of repeating units and/or cores of bacterial surface polysaccharides containing the disaccharide motives synthesized by glucansucrases and that can be obtained by the method of the invention (ref. 38, 43, 44, 45):
S. flexneri
S. flexneri
S. flexneri
S. flexneri
S. flexneri
S. pneumoniae
S. pneumoniae
S. pneumoniae
Within the framework of research that has lead to the present invention, the Inventors have demonstrated the chemo-enzymatic synthesis of S. flexneri serotype-specific oligosaccharides (
Accordingly, the present invention provides a method for preparing the synthetic intermediate corresponding to the disaccharide [α-D-Glcp(1→3)]-α-L-Rhap-OMe of formula (I) (ref. 74 and 62), and more generally to the disaccharide [α-D-Glcp(1→3)]-α-L-Rhap-YR of formula (Ia), wherein Y is selected from —O— and —S— and R is selected from the group consisting of: C1-C6 alkyl, C1-C6 alkenyl, aryl, allyl, —CO-alkyl (C1-C6), —CO-alkenyl (C1-C6), —CO-aryl; aryl designating an aromatic group like phenyl, benzyl, possibly substituted by one or several of the following groups: C1-C4 alkyl, —NO2, a halogen atom, —O-alkyl (C1-C6).
said method being characterized in that it comprises the step of using a mutant of a wild type glycoside hydrolase, wherein said wild type glycoside hydrolase has 450 to 850 amino acids, preferably 580 to 735 amino acids, and comprises from the N- to C-terminus eleven motifs defined by the following consensus motifs:
(1) the amino acid sequence LGVNYLHLMPL (SEQ ID NO: 1), which is located in the β-strand 2 of said wild type glycoside hydrolase;
(2) the amino acid sequence DGGYAV (SEQ ID NO: 2), which is located in the loop 2 of the (β/α)8-barrel of said wild type glycoside hydrolase;
(3) the amino acid sequence DFVFNH (SEQ ID NO: 3) which is located in the β-strand 3 of said wild type glycoside hydrolase;
(4) the amino acid sequence LREIFPDTAPGNF (SEQ ID NO: 4), which is located in the domain B of said wild type glycoside hydrolase;
(5) the amino acid sequence FNSYQWDLN (SEQ ID NO: 5), which is located in the C-terminal part of the domain B of said wild type glycoside hydrolase;
(6) the amino acid sequence ILRLDAVAFLWK (SEQ ID NO: 6), which is located in the β-strand 4 of said wild type glycoside hydrolase;
(7) the amino acid sequence EAIV (SEQ ID NO: 7), which is located in the β-strand 5 of said wild type glycoside hydrolase;
(8) the amino acid sequence YVRCHDDI (SEQ ID NO: 8), which is located in the β-strand 7 of said wild type glycoside hydrolase;
(9) the amino acid sequence RISGTLASLAG (SEQ ID NO: 9), which is located in the domain B′ of said wild type glycoside hydrolase;
(10) the amino acid sequence GIPLIYLGDE (SEQ ID NO: 10), which is located in the β-strand 8 of said wild type glycoside hydrolase;
(11) the amino acid sequence RWVHRP (SEQ ID NO: 11), which is located in the loop 8 of the (β/α)8-barrel,
and the sequence formed by said eleven motifs joined end-to-end from motif (1) to motif (11) of said wild type glycoside hydrolase has at least 65%, preferably at least 70%, and by order of increasing preference, at least 75%, 80%, 85%, 90%, 95%, 95%, 97%, 98%, and 99%, or 100% sequence identity or at least 80%, preferably at least 85%, and by order of increasing preference, at least 90%, 95%, 95%, 97%, 98%, and 99%, or 100% sequence similarity with the amino acid sequence SEQ ID NO: 12, which is formed by the concatenation of the eleven consensus motifs ordered from (1) to (11);
wherein said mutant has a mutation consisting of:
the substitution of the amino acid residue at position 4 in said motif (4) with any amino acid selected from the group consisting of alanine (A), cysteine (C), glycine (G), histidine (H), lysine (K), asparagine (N), arginine (R), serine (S), threonine (T), tryptophan (W) and tyrosine (Y), or
the substitution of the amino acid residue at position 9 in said motif (6) with any amino acid selected from the group consisting of cysteine (C), glutamic acid (E), isoleucine (I) and valine (V).
A “wild type glycoside hydrolase” refers to an amylosucrase (EC 2.4.1.4) or sucrose hydrolase (EC 3.2.1.-), preferably an amylosucrase. A wild type glycoside hydrolase belongs to the family 13, subfamily 4, of the glycoside hydrolases (GH13.4) as defined per the CAZY nomenclature (ref. 66-69, http://www.cazy.org).
The eleven consensus motifs of the wild type glycoside hydrolases have been found by the inventors by aligning 34 wild type glycoside hydrolases as shown in
By way of example, the glycoside hydrolase 1G5A (gi|16974797, SEQ ID NO: 13) comprises, from the N- to C-terminus, the eleven following motifs: (1) 125-134, (2) 144-149, (3) 182-187, (4) 225-237, (5) 250-258, (6) 282-293, (7) 328-331, (8) 388-395, (9) 446-456, (10) 480-489 and (11) 509-514 of SEQ ID NO: 13. These eleven motifs joined end-to-end form motif (1) to motif (11) form an amino acid sequence which has 83% sequence identity and 92% sequence similarity with the sequence SEQ ID NO: 12.
In order to identify the eleven motifs from a wild type glycoside hydrolase, one of skilled in the art can align the amino acid sequence of this wild type glycoside hydrolase against the amino acid sequence of the glycoside hydrolase 1G5A (SEQ ID NO: 13) for example, and therefore identify the eleven motifs thereof matching the eleven motifs of 1G5A as described above.
Unless otherwise specified, sequence alignments are performed using the well-known MUSCLE program under default parameters (http://phylogenomics.berkeley.edu/cgi-bin/muscle/input_muscle.py). Jalview software can be used for visualizing the alignment and generating the eleven motifs joined end-to-end. The sequence identity and similarity values provided herein are calculated using the Vector NTI AlignX program (V9.1.0, Invitrogen, USA) on a comparison window including the whole set of eleven consensus motifs ordered from 1 to 11 as defined above.
In a particular embodiment of said wild type glycoside hydrolase, it is an amylosucrase selected from the group consisting of the proteins available in the GENBANK database under the following accession number: gi|16974797 (named 1G5A and also reproduced herein as SEQ ID NO: 13), gi|99031739 (named 1ZS2), gi|27574003 (1MVY), gi|27574004 (named 1MW0), gi|47169012 (named 1S46), gi|16974938 (named 1JGI), gi|27574006 (named 1MW2), gi|27574007 (named 1MW3), gi|27574005 (named 1MW1), gi|16974937 (named 1JG9), gi|27728142 (named Q84HD6), gi|116670577, gi|32473567, gi|149179129, gi|76260974, gi|77163753, gi|158336602, gi|87310603, gi|149187214, gi|119944090, gi|109900119, gi|88795755, gi|152994364, gi|88800970, gi|114778050, gi|117926788, gi|78486138, gi|87300744, gi|91776960, gi|88804711, gi|158438431, gi|153811783, gi|153810451, gi|15805957, gi|94984679, gi|119715503, gi|113941581, gi|16125387, gi|119477809 and gi|89092061.
In a more preferred embodiment of said wild type glycoside hydrolase, it is an amylosucrase from Neisseria polysaccharea, and is preferably selected from the group consisting of 1G5A, 1ZS2, 1MVY, 1MW0, 1S46, 1JGI, 1MW2, 1MW3, 1MW1 and 1JG9 proteins.
In another preferred embodiment of said wild type glycoside hydrolase, it is a sucrose hydrolase from Xanthomonas, and is preferably selected from the group consisting of the proteins available in the GENBANK database under the following accession number: gi|78049174, gi|21244215, gi|58580721, gi|84622653, gi|21232788.
The Table I below shows the sequence identity and similarity percent of the eleven motifs joined end-to-end for each of 34 glycoside hydrolases as described above with the sequence SEQ ID NO: 12.
Neisseria polysaccharea
Neisseria meningitidis
Arthrobacter sp. FB24
Chloroflexus aurantiacus J-10-fl
Rhodopirellula baltica SH 1
Nitrosococcus oceani ATCC 19707
Planctomyces maris DSM 8797
Acaryochloris marina MBIC11017
Pseudoalteromonas atlantica T6c
Psychromonas ingrahamii 37
Alteromonas macleodii ‘Deep ecotype’
Vibrio shilonii AK1
Marinomonas sp. MWYL1
Blastopirellula marina DSM 3645
Reinekea sp. MED297
Herpetosiphon aurantiacus ATCC
Deinococcus radiodurans R1
Deinococcus geothermalis DSM
Thiomicrospira crunogena XCL-2
Robiginitalea biformata HTCC2501
Nocardioides sp. JS614
Mariprofundus ferrooxydans PV-1
Magnetococcus sp. MC-1
Xanthomonas campestris pv.
campestris str.
Xanthomonas axonopodis pv.
citri str. 306
Xanthomonas campestris pv.
vesicatoria str. 85-10
Xanthomonas oryzae pv. oryzae
Synechococcus sp. WH 5701
Methylobacillus flagellatus KT
Xanthomonas oryzae pv. oryzae
Ruminococcus obeum ATCC 29174
Clostridium bolteae ATCC BAA-613
Ruminococcus obeum ATCC 29174
Caulobacter crescentus CB15
According to a preferred embodiment of said wild type glycoside hydrolase, it contains an isoleucine (I) or valine (V) residue at position 4 in said motif (4), preferably an isoleucine.
According to another preferred embodiment of said wild type glycoside hydrolase, it contains a phenylalanine (F) or tyrosine (Y) residue at position 9 in said motif (6), preferably a phenylalanine.
By way of example, the amino acid residues at position 4 in said motif (4) and at position 9 in said motif (6) of the amylosucrase 1G5A, corresponds respectively to an isoleucine (I) at position 228 and a phenylalanine (F) at position 290 in the amino acid sequence of 1G5A (SEQ ID NO: 13).
According to a preferred embodiment of said mutant of a wild type glycoside hydrolase, the amino acid residue at position 4 in said motif (4) is substituted with any amino acid selected from the group consisting of histidine, tryptophan and tyrosine.
According to another preferred embodiment of said mutant of a wild type glycoside hydrolase, the amino acid residue at position 9 in said motif (6) is substituted with any amino acid selected from the group consisting of glutamic acid and isoleucine.
Unexpectedly, the mutants of a wild type glycoside hydrolase, in particular mutants of an amylosucrase, as defined in the present invention show a specific activity towards L-Rhap, α-L-Rhap-OMe, α-L-Rhap-OAllyl. The use of an appropriate combination of a mutant of wild type glycoside hydrolase, in particular an amylosucrase, with a donor and an acceptor as defined in the present invention at an earlier stage of a multi-step synthesis leads to the synthesis of any complex oligosaccharides, such as fragments of S. flexneri 3a, X, 5a, 5b, and 2b O-antigens.
Especially, the invention is directed to a method for preparing the synthetic intermediate corresponding to the disaccharide [α-D-Glcp(1→3)]-α-L-Rhap-OMe of formula (I):
and more generally to the disaccharide [α-D-Glcp(1→3)]-α-L-Rhap-YR of formula (Ia),
wherein Y is selected from —O— and —S— and R is selected from the group consisting of: C1-C6 alkyl, C1-C6 alkenyl, aryl, allyl, —CO-alkyl (C1-C6), —CO-alkenyl (C1-C6), —CO-aryl:
wherein aryl designates an aromatic group like phenyl, benzyl, possibly substituted by one or several of the following groups: C1-C4 alkyl, —NO2, a halogen atom, —O-alkyl (C1-C6),
said method being characterized in that it comprises the step of reacting a mutant of the wild type glycoside hydrolase as defined above, with the acceptor of formula (II) (methyl α-L-rhamnopyranoside (ref. 78)) or with the acceptor of formula (IIa), respectively:
and with a donor of formula (IIIa):
wherein R1 represents a group selected from:
and preferably with the sucrose donor of formula (III):
Advantageously, the method further comprises at least one step of acetylation by treatment with Ac2O, so that the disaccharide of formula (XX1) or (XX1a):
is obtained as intermediate molecule.
Another object of the invention is a method comprising an enzymatic glucosylation step for the preparation of the building block corresponding to the disaccharide of formula (XX5):
wherein it follows the steps according to scheme 1 below. Said method is illustrated in the examples and in
Another object of the invention is a method comprising an enzymatic glucosylation step as illustrated on
comprising at least one step of reacting a mutant of the wild type glycoside hydrolase as defined above in the presence of the acceptor of formula (IIa), advantageously with the acceptor of formula (II):
and with a donor of formula (IIIa):
wherein R1 can represent a group selected from:
preferably the donor of formula (III):
Some of the molecules obtained by the method of the invention are new and are another object of the invention. Among these are the following molecules:
Another object of the invention is a mutant of a wild type glycoside hydrolase, said wild type glycoside hydrolase being defined as above, and said mutant having a mutation consisting of:
the substitution of the amino acid residue at position 4 in said motif (4) with any amino acid selected from the group consisting of lysine (K) and arginine (R), or
the substitution of the amino acid residue at position 9 in said motif (6) with any amino acid selected from the group consisting of cysteine (C), glutamic acid (E), isoleucine (I) and valine (V).
In a preferred embodiment of said mutant of a wild type glycoside hydrolase, it is a mutant of an amylosucrase from Neisseria polysaccharea having the amino acid sequence SEQ ID NO: 13, wherein said mutant, has in reference to SEQ ID NO: 13, a mutation consisting of:
the substitution of the isoleucine (I) residue at position 228 (I228) with any amino acid selected from the group consisting of lysine (K) and arginine (R), or
the substitution of the phenylalanine (F) residue at position 290 (F290) with any amino acid selected from the group consisting of cysteine (C), glutamic acid (E), isoleucine (I) and valine (V).
The present invention also provides polynucleotides encoding a mutant of a glycoside hydrolase according to the present invention.
Polynucleotides of the invention may be obtained by the well-known methods of recombinant DNA technology and/or of chemical DNA synthesis. These methods also allow introducing the desired mutations in a naturally occurring DNA sequence.
The invention also provides recombinant DNA constructs comprising a polynucleotide of the invention, such as expression cassettes wherein said polynucleotide is linked to appropriate control sequences allowing the regulation of its transcription and translation in a host cell and optionally to a sequence encoding a GST tag allowing a rapid purification of the mutant enzymes and recombinant vectors comprising a polynucleotide or an expression cassette of the invention.
In addition to the preceding features, the invention further comprises other features which will emerge from the following description, which refers to examples illustrating the present invention, as well as to the appended figures.
Plasmid pGST-AS, derived from the pGEX-6P-3 (GE Healthcare Biosciences) and containing the N. polysaccharea amylosucrase encoding gene (ref. 45) was used for the construction of the AS single mutant library.
E. coli JM109 was used as host for the plasmid library transformation, gene expression and large-scale production of the selected mutants.
Sucrose, N-acetyl-D-glucosamine and glycogen were purchased from Sigma-Aldrich (Saint-Louis, Mo., USA).
Methyl α-L-rhamnopyranoside (ref. 78), and the disaccharides of reference α-D-Glcp-(1→3)-α-L-Rhap-OMe (ref. 74) and α-D-Glcp-(1→4)-α-L-Rhap-OMe (ref. 63) were chemically synthesised at the Institut Pasteur (Paris, France).
Ampicillin, lysozyme and isopropyl β-D-thiogalactopyranoside (IPTG) were purchased from Euromedex (Souffelweyersheim, France), and DpnI restriction enzyme from New England Biolabs (Beverly, Mass., USA).
Oligonucleotides were synthesised by Eurogenetec (Liege, Belgium).
DNA extraction (QIASpin) and purification (QIAQuick) columns were purchased from Qiagen (Chatsworth, Calif.).
Wild type glycoside hydrolase: amylosucrase (ASNPwt) 1G5A of sequence SEQ ID NO: 13.
Starting models for the disaccharide and for AS: The disaccharide α-D-Glcp-(1→3)-α-L-RhapOMe (formula I) was constructed with the monosaccharide obtained from a database of carbohydrate three-dimensional structures. All molecular modeling calculations were performed using the SYBYL 7.3 software. The coordinates of amylosucrase were taken from the 2.0 Å resolution crystal structures of amylosucrase from N. polysaccharea in complex with sucrose (PDB: 1JGI) and maltoheptaose, a reaction product (PDB: 1MW0). All hydrogen atoms were added to the enzyme and their position optimized with the Tripos force field.
Systematic conformational search for the disaccharides: α-D-Glcp-(1→3)-α-L-RhapOMe (formula I) was subjected to a systematic grid search study of the glycosidic linkage conformation. Starting from minimized disaccharide, a two-dimensional systematic conformational search was performed by rotating the two torsion angles defining the glycosidic linkages, Φ and Ψ by 20° steps: Φ=O5′-C1′-O3-C3′ and Ψ=C1′-O3-C3′-C2′ for α-D-Glcp-(1→3)-α-L-RhapOMe. The MM3 force field implemented in SYBYL 7.3 software was used for this purpose together with the energy parameters appropriate for carbohydrates. Different maps were constructed with the dielectric constant set to 4.0 and 78.0 (to mimic gas phase and water environment, respectively). The geometries were optimized at each point of the grid with the driver option that keeps fixed the atoms defining the torsion angles. The solvent specific relaxed conformational maps obtained for the disaccharide were then used to locate the different energy minima that were subsequently fully relaxed.
Docking of disaccharide in the binding site of AS: The lowest energy conformations identified on the disaccharide potential energy maps were used as starting structures to be docked in the binding site of amylosucrase. This was performed by superimposing the glucosyl unit of the disaccharide at (−1) subsite onto the glucosyl unit of the crystallographic maltoheptaose. Each of these AS-disaccharide complexes was optimized by means of the appropriate energy parameters. The annealing method implemented in SYBYL 7.3 software was used to optimize the complexes. Two shells of amino acids were considered: a 12 Å shell centered on the binding site was taken into account for the energy calculations. A 6 Å shell region closest to the carbohydrate was defined as the hot region to be optimized. The position of all atoms included in this region was optimized using Powell's method.
Single mutagenesis, focused on +1 subsite amino acids retained from ligand docking, was carried out with the QuickChange Site-Directed Mutagenesis Kit (Stratagene, La Jolla, Calif.) according to the manufacturer's instructions, and using pGST-AS G537D as vector template. It was checked that this mutation had no impact on the native enzyme catalytic properties. The complementary primers listed below were used to obtain the single mutant library (Table II below). XXX codon indicates the bases which were used to obtain the replacement by the desired amino acids and are listed in Table III below.
PCR amplification was carried out with Pfu DNA polymerase (2.5 U) for 16 cycles (95° C., 30 s; 55° C., 30 s; 72° C., 12 min). The DNA was digested with DpnI to eliminate methylated parental template and purified using Qiaquick spin column, following manufacturer's recommendations. E. coli JM109 was transformed with the plasmid and plated on LB agar supplemented with 100 μg/mL ampicillin. For each construction, two clones were isolated and their corresponding plasmids stored at −20° C. 17 mutants (I228A1, I228V1, I228Y1, A289D1, F290D1, F290K1, F290Q1, I330A1, I330D1, I330E1, I330F1, I330T1, I330W1, V331A1, V331S1, D394V1 and R446K1) were sequenced on the entire gene and showed no other mutations by Millegen (Labège, France) or Cogenics (Meylan, France).
The protocol was established to enable the rapid identification of clones able to glucosylate α-L-Rhap-OMe (formula II). To obtain higher amounts of enzymes and facilitate detection of glucosylated compounds upon HPLC screening, mutants were produced in 96-DeepWell Format plates. Storage microplates containing monomutants were thawed and replicated to inoculate a starter culture in 96-well microplates containing, in each well, 150 μL LB medium supplemented with ampicillin (100 μg/mL). After growth for 24 h at 30° C. under agitation (200 rpm), plates were duplicated into 96-Deep Well plates containing, in each well, 1.1 mL LB medium supplemented with ampicillin (100 μg/mL) and IPTG (1 mM) to induce GST-AS expression. Cultures were then grown for 24 h at 30° C. under agitation (200 rpm). Plates were centrifuged (20 min, 3000 g, 4° C.) and the supernatant was removed. The cell pellet was resuspended in 200 μL of lysozyme (0.5 mg/mL), followed by freezing at −80° C. for 8 to 12 h. After thawing at room temperature, 100 μL of sucrose and 100 μL of acceptor (each at a final concentration of 73 mM) were added to each well. Enzymatic reaction was incubated at 30° C. during 24 h under agitation. The DeepWell plates were then centrifuged (20 min, 3000 g, 4° C.) and 300 μL of the supernatant was transferred to a filter micro-plate (PVDF 0.2 μm) to be clarified. Supernatant filtration was carried out by centrifugation of the filter micro-plate (5 mM, 2000 g, 4° C.) into a novel microplate for screening.
Efficiency of the glucosylation reaction was evaluated by HPLC analysis of the acceptor reaction product synthesized when using α-L-Rhap-OMe as acceptor using a Dionex P 680 series pump, a Shodex RI 101 series refractometer, a Dionex UVD 340 UV/Vis detector and an autosampler HTC PAL. HPLC analyses were performed using two columns: 1) a Biorad HPLC Carbohydrate Analysis column (HPX-87K column (300×7.8 mm)) maintained at 65° C., using ultra-pure water as eluent with a flow rate of 0.6 mL/min; 2) a reversed phase analytical column (Synergi C18RPFusion, 4 μm, 30×4.6 mm) kept at room temperature, with 1 ml/min of ultra-pure water as eluent. HPX-87K column was used to determine sucrose consumption by RI detection. C18RPfusion column served to detect the production of α-D-Glcp-(1→3)-α-L-RhapOMe.
Production and purification of AS variant were performed as previously described (ref. 45). Since pure GST/AS fusion protein possesses the same function and the same efficiency as pure AS, enzymes were purified to the GST/AS fusion protein stage (96 kDa). The enzyme was desalted by size exclusion chromatography using P6DG columns (GE Healthcare Biosciences) at +4° C. and stored in elution buffer (50 mM Tris-HCl, pH 7.0, 150 mM NaCl) at −80° C. The protein content was determined by micro-Bradford method, using bovine serum albumin as standard (ref. 55).
All assays were performed at 30° C. in 50 mM Tris buffer, pH=7.0. For the acceptor reactions, I228Y and wild type AS were tested on α-L-Rhap-OMe (formula II) used as acceptor.
Standard activity determination. Specific activity of the purified enzymes was determined by measuring the initial rate of released fructose under standard conditions (146 mM sucrose). Fructose concentration was determined using the dinitrosalycilic acid (DNS) method (ref. 56). One unit of AS variant corresponds to the amount of enzyme that catalyses the production of 1 μmole fructose per minute in the assay conditions.
Comparison of products synthesized by wild type and AS variant. Reactions were performed in the presence of 146 mM sucrose alone or supplemented with 146 mM acceptor. The purified wild-type or mutated GST/AS were employed at 0.5 U/mL. The reactions were stopped by heating at 95° C. for 5 mM. The final mixture was centrifuged at 18 000 g for 5 min. Different carbohydrate analyses were performed to compare the product profiles synthesized by ASNPwt and AS variant (I228Y):
Soluble and insoluble oligosaccharides produced during the reaction were identified by HPAEC using a Dionex Carbo-Pack PA100 column at 30° C. Before analysis, the insoluble fraction was solubilised in KOH at a final total sugar concentration of 10 g/kg. Mobile phase (150 mM NaOH) was set at 1 mL/min flow rate with a sodium acetate gradient (going from 6 to 500 mM within 120 min). Detection was performed using a Dionex ED40 module with a gold working electrode and an Ag/AgCl pH reference. Note that α-L-Rhap-OMe (acceptor) and its derivatives are not oxidable products and thus are not detectable by HPAEC. Sucrose, glucose, fructose, α-L-Rhap-OMe (acceptor) and its derivatives (glucosylation products) were quantified by HPLC, as previously described.
Determination of Kinetic Parameters.
Enzyme assays were carried out in a total volume of 2 mL containing pure enzyme (0.115 mg and 2.6 mg when using ASNPwt and I228Y, respectively). Catalytic efficiency (Eff=kcat/KM) of ASNPwt and I228Y variant were determined with both sucrose (D=Donor) and methyl α-L-rhamnopyranoside (A=Acceptor) as variable substrates. For the determination of the catalytic efficiency for D, A was held constant at 250 mM and D was varied between 0 and 600 mM. For the determination of the catalytic efficiency for A, D was held constant at 250 mM and A was varied between 0 and 500 mM. Experiments were performed until A and D solubility limits were reached. For each experiment, two distinct reaction velocities were determined: (1) the donor consumption determined by the release of fructose (called Vi (F)), corresponding to the overall kinetics of the different reactions described in
Initial velocities were fitted to the Michaelis-Menten equation using Sigma-Plot. As saturation was not achieved with the mutant, efficiency was calculated by linear regression analysis of the velocity versus substrate concentration plot. Aliquots (200 μL) were removed between 0 and 60 min (at which time product formation was still linear with respect to time), heated at 95° C. for 5 min and centrifuged at 18 000 g for 5 min. The final mixtures were filtered on a 0.22 μm membrane and analyzed using HPLC material previously described. HPLC analyses were performed using two columns: 1) a Biorad HPLC Carbohydrate Analysis column (HPX-87H column (300×7.8 mm)) maintained at 30° C., using ultra-pure water as eluent with a flow rate of 0.6 mL/min; 2) a reversed phase analytical column (Synergi C18RPFusion, 4 μm, 30×4.6 mm) kept at room temperature, with 1 ml/min of ultra-pure water as eluent. HPX-87H column was used to determine the released fructose by RI detection. C18RPfusion column was used to detect the formation of α-D-Glcp-(1→3)-α-L-RhapOMe.
Preparative Synthesis of the Acceptor Reaction Product (cf.
Methyl α-
A 500 ml mixture containing 146 mM sucrose, 100 mM methyl α-L-rhamnopyranoside (formula II) (ref. 78), and 0.25 U/ml of non-purified I228Y mutant extract was incubated at 30° C. for 24 h. Then, the reaction mixture was centrifuged (4800 rpm, 20 min, 4° C.) to remove proteins and filtered on a 0.22 μm membrane
Purification and Characterisation of the Target Glucosylation Product (
Methyl(2,3,4,6-tetra-O-acetyl-α-
Methyl α-
The choice of a suitable rhamnosyl residue to be used as acceptor in the enzymatic glucosylation step was crucial. Three major features were taken into account: (i) light protecting pattern compatible with the limited ability of selected glucansucrases to modulate their acceptor binding site, (ii) easy synthetic access, and (iii) the possible conversion of the glucosylation product into a protected disaccharide building block known to be compatible with additional chemical elongation. In view of previous work on S. flexneri 5a synthetic oligosaccharides (ref. 64), methyl α-
Glucansucrases are α-retaining transglucosidases found in families 13 and 70 of glycoside-hydrolases (ref. 51). They catalyze the synthesis of α-glucan polymers by successive transfers of α-D-glucopyranosyl units from sucrose without any mediation of sugar nucleotides. Using the high energy of the sucrose bond to catalyze condensation reaction, they stand among the most efficient transglucosidases in the glycoside-hydrolase family. Depending on regiospecificity of the enzyme, distinct types of glucosidic linkage are found in the polymer formed. Notably, polymerization reaction can be redirected toward the glucosylation of exogenous acceptors, when the latter are well recognized (
Consequently, to overcome the limited substrate recognition by the enzymes, it was opted for the engineering of novel transglucosidases with altered regio and stereospecificities. The combination of combinatorial and rational enzyme engineering in the form of focused small size libraries was used. Further, the three-dimensional structure of AS (ref. 52 and 81) was available in complex with either the substrate or the natural reaction product.
To modify enzyme specificity, an approach based on site-directed evolution targeted at the binding pocket was followed. The catalytic site pocket is defined by the subsites (−1) and (+1) according to the nomenclature earlier described for glycoside hydrolases (
It was combined both the rational selection of mutation targets at the acceptor binding site (noted +1 subsite in
Structural Analysis of α-D-Glcp-(1→3)-α-L-Rhap-OMe [EA-OMe]: AS Complex
The desired disaccharide α-D-Glcp-(1→3)-α-L-Rhap-OMe was docked into the AS active site using the crystallographic maltose glucosyl units (ie α-D-Glcp-(1→4)-D-Glcp: native product) bound at (−1) and (+1) subsites (PDB: 1MW0) as a template for the starting location. The conformation of the disaccharide α-D-Glcp-(1→3)-α-L-Rhap-OMe in the enzyme pocket with best binding energy was considered out of all docking solutions. Comparison of the docking modes adopted by α-D-Glcp-(1→4)-D-Glcp and α-D-Glcp-(1→3)-α-L-Rhap-OMe, in particular at subsite (+1), revealed a ‘flip’ of the pyranose ring, due to the change in the chair ring conformation adopted by the α-L-Rhap-OMe unit (4C1 vs 1C4, respectively) as well as a ‘rotation’ of the pyranose ring due to the altered osidic linkage (α-1→4 vs α-1→3, respectively). As a result of these main structural differences, the network of interactions with the wild type enzyme was drastically impaired. A detailed comparison of both disaccharides bound in the enzyme pocket allowed to highlight the key amino acid side-chains in close contact with α-D-Glcp-(1→3)-α-L-Rhap-OMe that could be selected to facilitate the binding of the acceptor in a catalytically productive position and to provide a new network of interactions specific to α-L-Rhap-OMe (
Out of the 18 residues identified as surrounding the (+1) subsite, 7 positions that were presumed to be not critical for sucrose binding but beneficial for target acceptor glucosylation have been selected for mutagenesis: Ile228, Ala289, Phe290, Ile330, Val331, Asp394 and Arg446. It was systematically mutated the selected amino acids by the 19 other possible residues to create a small size library focused on +1 subsite.
For each of the 7 selected positions, 19 single mutants were generated (corresponding to each possible amino acid change). Site-directed mutagenesis has been performed. A first library of 133 monomutants (7×19) was thus obtained and stored in cryotubes and 96-wells microplates. The mutants were tested for the glucosylation of the target acceptor α-L-Rhap-OMe in microtiter format experiments. HPLC screening was performed to identify those able to form the desired disaccharide. Both sucrose consumption and disaccharide formation were determined in order to calculate the glucosylation rate defined as the molar ratio of monoglucosylated acceptor versus/sucrose consumed.
The wild type AS does not recognize the α-L-RhapO-Me as an acceptor. As shown in
In overall, the most improved mutant of interest for the chemo-enzymatic pathway was retained and further characterized: I228Y which is specific for the production of α-D-Glcp-(1→3)-α-L-RhapO-Me.
I228Y was produced in a larger amount and purified to homogeneity for further characterization. Glucosylation reactions with I228Y and wild type AS were performed using sucrose and in the absence or presence of α-L-RhapO-Me. Distributions of the acceptor reaction products obtained with I228Y and wild type AS are shown in
Comparison of kinetic parameters determined using sucrose as variable donor and α-L-RhapO-Me as constant acceptor showed that catalytic efficiency towards sucrose is 13 to 150 times less active for I228Y variant than for wild type AS depending on sucrose concentration (FIGS. 7D(a) and 7E). However, the catalytic efficiency towards the desired reaction (acceptor glucosylation) clearly indicates that an enzyme with novel specificity has been obtained, adapted to sucrose usage for methyl α-L-rhamnopyranoside glucosylation. Noteworthy, FIG. 7D(b) shows a clear activation of I228Y in the presence of α-L-RhapO-Me. Activation of I228Y is induced by increasing concentration of acceptor and leading to an increase of Vi(Dp2) and therefore of Vi(F) in the same order of magnitude (FIG. 7D(b)). Conversely, no activation of wild type AS is detected in the presence of the acceptor, as shown in
The seven 1G5A mutants I228A, I228G, I228H, I228N, I228S, I228W and I228Y were tested on other rhamnose derivatives (
Regarding glucosylation with
Efficiency of the glucosylation reaction was evaluated by HPLC analysis of the acceptor reaction product synthesized using a reversed phase analytical column (Synergi C18RPFusion, 4 μm, 30×4.6 mm) (when testing
General methods. TLC were performed on precoated slides of Silica Gel 60 F254 (Merck). Detection was effected when applicable, with UV light, and/or by charring in orcinol (35 mM) in 4N aqueous sulfuric acid and ethanol (95/5). Preparative chromatography was performed by elution from columns of Silica Gel 60 (particle size 0.040-0.063 mm). NMR spectra were recorded at 25° C. for solutions in CDCl3 or MeOD (400 MHz for 1H, 100 MHz for 13C). Residual CHCl3 (7.28 ppm for 1H and 77.0 ppm for 13C), MeOH (3.31 ppm for 1H and 49.0 ppm for 13C), and HOD (4.79 ppm) were used as internal references for solutions in CDCl3, MeOD, and D2O, respectively. Proton-signal assignments were made by first-order analysis of the spectra, as well as analysis of 2D 1H—1H correlation maps (COSY). Of the two magnetically non-equivalent geminal protons at C-6, the one resonating at lower field is denoted H-6a, and the one at higher field is denoted H-6b. The 13C NMR assignments were supported by 2D 13C—1H correlations maps (HMBC and HSQC). Interchangeable assignments are marked with an asterisk in the listing of signal assignments. Sugar residues in disaccharides are serially lettered according to the lettering of the repeating unit of the S. flexneri 3a O-antigen (glucopyranosyl: E, and rhamnopyranosyl: A) and identified by a subscript in the listing of signal assignments. Electrospray Ionisation-Time of flight (ESI-TOF) mass spectra were recorded in the positive-ion mode using a 1/1 acetonitrile (CH3CN)/water containing 0.1% formic acid ESI-TOF spectrometer-solution. Anhydrous dichloromethane (DCM) and dichloroethane (DCE) sold on molecular sieves were used as such. 4 Å powder molecular sieves was activated before use by heating at 250° C. under vacuum.
Methyl(2,3,4,6-tetra-O-acetyl-α-
Acetic anhydride (100 mL) was added dropwise to a solution of the whole mixture (17.5 g) in anhydrous pyridine (100 mL) stirred at 0° C. The solution was stirred overnight at room temperature. TLC (Toluene-EtOAc, 6:4) showed the complete disappearance of the starting materials and the presence of three less polar products. The mixture was concentrated under reduced pressure, and volatiles were eliminated by repeated coevaporation with toluene. The residue was purified by column chromatography (Toluene-EtOAc, 7:3) to give XX1 (4.9 g, 17% from methyl α-
(2,3,4,6-Tetra-O-acetyl-α-
(2,3,4,6-Tetra-O-acetyl-α-
(2,3,4,6-Tetra-O-acetyl-α-
α-
(2,3,4,6-Tetra-O-benzyl-α-
Allyl(2,3,4,6-tetra-O-benzyl-α-
The β anomer XX5A, (R3=All) had Rf=0.45 (Cyclohexane-EtOAc, 7.3:2.7); 1H NMR (CDCl3) δ 7.43-7.05 (m, 25H, Ph), 5.88 (m, 1H, CH═), 5.66 (d, 1H, H-2A), 5.33 (d, 1H, J1,2=3.5 Hz, H-1E), 5.28 (m, 1H, Jtrans=17.2 Hz, ═CH2), 5.25 (m, 1H, Jcis=10.2 Hz, ═CH2), 5.03 (d, 1H, J=11.0 Hz, OCH2Ph), 4.96 (d, 1H, J=10.2 Hz, OCH2Ph), 4.94 (d, 1H, OCH2Ph), 4.90-4.85 (m, 2H, OCH2Ph), 4.78 (d, 1H, J=11.4 Hz, OCH2Ph), 4.72-4.56 (m, 3H, OCH2Ph), 4.82 (bs, 1H, H-1A), 4.77 (d, 1H, J=12.1 Hz, OCH2Ph), 4.31 (m, 1H, OCH2), 4.14-4.06 (m, 3H, H-3E, OCH2, H-5E), 3.95 (dd, 1H, J2,3=3.2 Hz, J3,4=9.6 Hz, H-3A), 3.69 (pt, 1H, J4,5=9.6 Hz, H-4E), 3.65-3.55 (m, 4H, H-2E, H-6aE, H-6bE, H-4A), 3.35 (dq, 1H, J4,5=9.2 Hz, H-5A), 1.98 (s, 3H, Ac), 1.45 (d, 3H, J5,6=6.2 Hz, H-6A); 13C NMR (CDCl3) δ 170.7 (C═O), 138.8-137.5 (CPh), 133.6 (CH═), 128.9-127.4 (CHPh), 117.5 (CH2═), 97.5 (C-1A, 1JCH=156.3 Hz), 92.0 (C-1E, 1JCH=169.0 Hz), 82.0 (C-3E), 79.4 (C-4A), 79.3 (C-2E), 77.9 (C-4E), 76.3, 75.6, 75.0, (3C, 2OCH2Ph), 73.8 (C-3A), 73.1, 72.9 (2C, 2OCH2Ph), 71.9 (C-5A), 70.0 (C-5E), 69.8 (OCH2), 68.4 (C-6E), 66.7 (C-2A), 20.8 (Ac), 17.9 (C-6A). HRMS (ESI+) of C52H58O11Na (M+Na, 881.3877) m/z 881.3854 ([M+Na]+); HRMS (ESI+) of C52H58O11NH4 (M+NH4, 876.4323) m/z 876.4322 ([M+NH4]+).
Allyl(2,3,4,6-tetra-O-benzyl-α-
(2,3,4,6-Tetra-O-benzyl-α-
Compound XX5 may be obtained according to the same protocole when using Compound XX4 (R=All) as starting material. Alternatively, Compound XX5 can also result from conventional acidic hydrolysis of Compounf XX4 (X=All, Me, Pent).
(2,3,4,6-Tetra-O-benzyl-α-
(2,3,4,6-tetra-O-acetyl-α-
α-
(2,3,4,6-Tetra-O-benzyl-α-
Pentenyl(2,3,4,6-tetra-O-benzyl-α-
Pentenyl(2,3,4,6-tetra-O-benzyl-α-
| Number | Date | Country | Kind |
|---|---|---|---|
| 08290238.8 | Mar 2008 | EP | regional |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/IB09/05343 | 3/12/2009 | WO | 00 | 12/27/2010 |
