Chemical synthesis of UDP-Glc-2,3-diNAcA, a key intermediate in cell surface polysaccharide biosynthesis in the human respiratory pathogens B. pertussis and P. aeruginosa

Article abstract of DOI:10.1039/b819607a

In connection with studies on lipopolysaccharide biosynthesis in respiratory pathogens we had a need to access potential biosynthetic intermediate sugar nucleotides. Herein we report the chemical synthesis of uridine 5′-diphospho 2,3-diacetamido-2,3-dideoxy-α-D-glucuronic acid (UDP-Glc-2,3-diNAcA) (1) from N-acetyl-D-glucosamine in 17 steps and ～9% overall yield. This compound has proved invaluable in the elucidation of biosynthetic pathways leading to the formation of 2,3-diacetamido-2,3-dideoxy-D- mannuronic acid-containing polysaccharides. The Royal Society of Chemistry 2009.

Full text of DOI:10.1039/b819607a

View Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Chemical synthesis of UDP-Glc-2,3-diNAcA, a key intermediate in cell
surface polysaccharide biosynthesis in the human respiratory pathogens
B. pertussis and P. aeruginosa†
Martin Rejzek,a,b Velupillai Sri Kannathasan,a Corin Wing,a Andrew Preston,c Erin L. Westman,d
Joseph S. Lam,d James H. Naismith,e Duncan J. Maskell f and Robert A. Field*a,b
Received 4th November 2008, Accepted 8th December 2008
First published as an Advance Article on the web 2nd February 2009
DOI: 10.1039/b819607a
In connection with studies on lipopolysaccharide biosynthesis in respiratory pathogens we had a need
to access potential biosynthetic intermediate sugar nucleotides. Herein we report the chemical synthesis
of uridine 5¢-diphospho 2,3-diacetamido-2,3-dideoxy-a-D-glucuronic acid (UDP-Glc-2,3-diNAcA) (1)
from N-acetyl-D-glucosamine in 17 steps and ~9% overall yield. This compound has proved invaluable
in the elucidation of biosynthetic pathways leading to the formation of 2,3-diacetamido-
2,3-dideoxy-D-mannuronic acid-containing polysaccharides.
However, the reducing terminal lipid A component of LPS is
Introduction
an endotoxin that is capable of causing septic shock.7 Hence, a
Bordetella pertussis is an important human respiratory pathogen
source of LPS lacking this core is key to therapeutic utility. In
that causes whooping cough (pertussis).1 Despite high levels of
the case of B. pertussis, the glycan in question is the so-called
vaccine coverage in recent years, there has been a worldwide
Band A trisaccharide, which contains two highly functionalised
monosaccharides that are not found in humans (Fig. 1).8
resurgence of B. pertussis infection—up ca. 15–20 fold compared
to 20 years ago. With pertussis continuing to pose a serious threat
to infants, and also affecting adolescents and adults, there remains
a need for improved diagnosis and treatment.2,3 Vaccination
represents the most practical way to deal with whooping cough,
so alternatives to the established lysed cell and acellular vaccines
may offer opportunities to address this matter. Analysis of
genome sequences for three Bordetella spp. gives clues to the
differences in host range and the severity of the diseases that
they cause.4 Such top-down analyses may eventually underpin the
identification of new vaccine candidates. However, an alternative,
bottom-up approach based on knowledge of bacterial cell surface
composition is also feasible. Bacterial lipopolysaccharides (LPS)5
are important bacterial surface components. Immunologically, the
carbohydrate component of LPS can give rise to a highly specific
immune response, rendering it suitable for use in glycoconjugate
vaccines, as is the case for bacterial capsular polysaccharides.6
Fig.
1 Band A trisaccharide consisting of N-acetylglucosamine
(GlcNAc), 2,3-diacetamido-2,3-dideoxy-D-mannuronic acid (Man-2,3-
diNAcA) and 2,4-dideoxy-2-N-acetyl-4-N-methylfucos-2,4-diamine
(Fuc2NAc4NMe). ROS designates a connection to an oligosaccharide.
The central 2,3-diacetamido-2,3-dideoxy-D-mannuronic acid
(Man-2,3-diNAcA) unit of this repeating trisaccharide is also
found in the LPS-associated O antigen of Pseudomonas aerug-
inosa,9,10 an opportunistic pathogen that frequently infects the
airways of cystic fibrosis patients.11 It is thought that B. pertussis
Band A trisaccharide may make a good vaccine component
because it is highly immunogenic, highly conserved, and passive
transfer of anti-band A antibodies was protective in a mouse
challenge model.12 Whilst the research scale chemical synthesis
of a GlcNAc-Man-2,3-diNAcA disaccharide fragment of Band A
trisaccharide has been reported,13 scale-up would be tortuous. We
were drawn to consider biotransformation routes instead, which
led us to investigate the biosynthesis of Band A trisaccharide. It
has been shown that the biosynthesis of this glycan is dependent on
the Wlb gene locus,14 which potentially provides the full repertoire
of genes/enzymes necessary to synthesise the key trisaccharide
by in vitro biotransformation, or with a genetically engineered
heterologous expression host.15 However, review of the initially
aSchool of Chemical Sciences and Pharmacy, University of East Anglia,
Norwich, UK NR4 7TJ
bBiological Chemistry Department, John Innes Centre, Norwich Research
Park, Colney Lane, Norwich, UK NR4 7UH. E-mail: rob.field@bbsrc.ac.uk;
Fax: +44 (0) 1603 450018; Tel: +44 (0) 1603 450720
cDepartment of Clinical Veterinary Science, University of Bristol, Langford,
North Somerset, UK BS40 5DU
dDepartment of Molecular and Cellular Biology, University of Guelph,
Guelph, ON N1G 2W1, Canada
eSchool of Chemistry and Centre for Biomolecular Sciences, University of
St Andrews, North Haugh, St Andrews, UK KY16 9ST
f
Department of Veterinary Medicine, University of Cambridge, Madingley
Road, Cambridge, UK CB3 0ES
† Electronic Supplementary Information (ESI) available: Experimental
details for the synthesis of intermediates, NMR spectra of compounds
13, 14 and 15, and NMR data tables for several sugar nucleotides. See
http://dx.doi.org/10.1039/b819607a/
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1203–1210 | 1203
©

View Online
Fig. 2 Biosynthetic pathway for UDP-Man-2,3-diNAcA in B. pertussis and P. aeruginosa (previously proposed version in upper line and revised version
in bottom line). Protein names are indicated in the following order: B. pertussis (P. aeruginosa).
proposed biosynthetic pathway for uridine 5¢-diphospho 2,3-
diacetamido-2,3-dideoxy-a-D-mannosuronic acid (UDP-Man-
2,3-diNAcA) formation14 (Fig. 2) alerted us to potential issues
arising from the assignment of Wlb gene function based on weak
sequence homology to proteins in the Protein Data Bank.
the biosynthesis of UDP-Man-2,3-diNAcA from UDP-GlcNAc
has been proposed (Fig. 2).23
An important intermediate in the new pathway is repre-
sented by uridine 5¢-diphospho 2,3-diacetamido-2,3-dideoxy-a-
D-glucuronic acid (UDP-Glc-2,3-diNAcA) (1) (Fig. 3). This sugar
nucleotide is not readily available.
In Gram-negative enterobacteria, such as Escherichia coli,
ManNAcA is found in the cell-surface enterobacterial common
antigen (ECA),16 where it is derived from uridine 5¢-diphospho
2-acetamido-2-deoxy-a-D-mannuronic acid (UDP-ManNAcA).
This uronic acid is generated from uridine 5¢-diphospho 2-
acetamido-2-deoxy-a-D-glucose (UDP-GlcNAc) by the sequen-
tial action of UDP-GlcNAc 2-epimerase and UDP-ManNAc 6-
dehydrogenase.17 Hence it seemed reasonable to anticipate that B.
pertussis would employ a similar order of events to produce manno-
configured uronic acids (i.e. epimerisation preceding oxidation). B.
pertussis WlbD, with 32% amino acid sequence identity to the E.
coli UDP-GlcNAc 2-epimerase RffE,18 seemed a logical choice
for the requisite 2-epimerase involved in UDP-Man-2,3-diNAcA
biosynthesis. However, despite having established UDP-GlcNAc
2-epimerase assays and validated them with authentic E. coli
RffE,18 we were unable to detect epimerase activity with recombi-
nant WlbD,19 which caused us to review the proposed biosynthetic
pathway leading to UDP-Man-2,3-diNAcA. On reflection, we
were struck by the proposed replacement of a hydroxyl group
with an amino group at the C3 position of UDP-GlcNAc—a pro-
posed single step (bio)transformation that is without precedent,
although the 2-step conversion (oxidation plus transamination) of
UDP-GlcNAc to uridine 5¢-diphospho 2-acetamido-3-amino-2,3-
dideoxy-a-D-glucose (UDP-3-NH2-GlcNAc) has been demon-
strated in Acidithiobacillus ferrooxidans lipid A biosynthesis.20
Parallel studies in P. aeruginosa led to the biochemical characteri-
zation of WbpA, demonstrating it to be an NAD-dependent UDP-
N-acetyl-D-glucosamine 6-dehydrogenase that generates uridine
5¢-diphospho 2-acetamido-2-deoxy-a-D-glucuronic acid (UDP-
GlcNAcA).21 This suggested that, in contrast to ECA biosynthesis,
the first committed step in UDP-Man-2,3-diNAcA biosynthesis is
the oxidation of UDP-GlcNAc to the corresponding uronic acid,
UDP-GlcNAcA. This notion was further supported by studies on
the B. pertussis WlbA oxidase,‡ which was shown to require a 6-
carboxyl group on GlcNAc in order to effect oxidation of the sugar
ring.22 Drawing all of these points together, a revised pathway for
Fig.
3
Uridine 5¢-diphospho 2,3-diacetamido-2,3-dideoxy-a-D-
glucuronic acid (UDP-Glc-2,3-diNAcA) (1) and the precursor sugar nu-
cleotide uridine 5¢-diphospho 2,3-diacetamido-2,3-dideoxy-a-D-glucose
(UDP-Glc-2,3-diNAc) (2). Numbering scheme for NMR purposes.
We were therefore minded to chemically synthesise authentic
UDP-Glc-2,3-diNAcA in order to assess its ability to act as a
substrate for B. pertussis WlbD, and the orthologous P. aeruginosa
WbpI. A positive result in such experiments would add weight
to, and evidence for, the proposed revised biosynthetic pathway
leading from UDP-GlcNAc to UDP-Man-2,3-diNAcA. Herein
we report the chemical synthesis and spectroscopic characterisa-
tion of the key target molecule UDP-Glc-2,3-diNAcA (1) and the
precursor sugar nucleotide uridine 5¢-diphospho 2,3-diacetamido-
2,3-dideoxy-a-D-glucose (UDP-Glc-2,3-diNAc) (2).
Results and discussion
N-Acylated versions of the naturally occurring 2,3-diamino-2,3-
dideoxy-D-glucose are found in the lipid A of some Rhodospiril-
laceae.24 This sugar has previously been synthesised by approaches
that involve a double inversion at C-3 of a gluco-configured sugar
building block, such as methyl 2-benzamido-4,6-O-benzylidene-2-
deoxy-b-D-glucopyranoside.25 Another approach to this sugar has
used methyl 2-O-acetyl-4,6-O-benzylidene-3-deoxy-3-nitro-b-D-
glucopyranoside, which was converted by the action of ammonia
in aqueous tetrahydrofuran into a mixture of two stereoiso-
meric nitroamines and separated by fractional crystallisation.26
‡ WlbA was originally proposed (Fig. 1)14 to be required for the generation
of the 6-carboxyl group, but this assignment was suspicious, based on the
lack of sequence homology to established UDP-sugar 6-dehydrogenases.
1204 | Org. Biomol. Chem., 2009, 7, 1203–1210
This journal is
The Royal Society of Chemistry 2009
©

View Online
A further approach to derivatives of 2,3-diamino-2,3-dideoxy-
D-glucose was based on ring opening of the aziridine in
methyl 2,3-benzoylepimino-4,6-O-benzylidene-2,3-dideoxy-a-D-
allopyranoside by sodium azide27 to form methyl 3-amino-2-
benzamido-2,3-dideoxy-a-D-glucopyranoside. An improved ver-
sion of this approach was employed later28 in a glycophospholipid
synthesis. The 2,3-diacetamido-mannuronate unit has also been
synthesised from D-glucose via azide opening of a 2,3-epoxide,
followed by azide introduction at C-3 by means of double inversion
chemistry.13 In our work, we chose to use an approach based on
double inversion at C-3 of benzyl 2-acetamido-4,6-O-benzylidene-
2-deoxy-a-D-glucopyranoside, involving neighbouring group par-
ticipation of the C-2 acetamido function.31 This approach offers
full control of the stereochemical outcome of the C-3 double
inversion, purification of intermediates is straightforward, as is
scale-up to multigram quantities. Overall, our strategy was to
prepare the gluco-configured 2,3-diacetamido-2,3-dideoxy-hexose
as its anomeric phosphate, couple this material to give the
corresponding UDP-adduct, and finally effect a selective oxidation
of the hexose primary alcohol to the desired uronic acid.
derivative 5 (Scheme 1). The C-3 hydroxyl group of 5 was
converted into an acetamido derivative 12 with the required
gluco-configuration by a double inversion at C-3. This involved
initial mesylation, followed by acetate displacement with in
situ de-O-acetylation to form the allo-configured intermediate
7, installation of the C-3 azide 9 via a mesylate, benzylidene
deprotection, azide reduction followed by acetylation, and finally
hydrogenation to cleave the benzyl glycoside to yield 12. These
transformations followed a literature procedure31 with some
modifications: tetrabutylammonium hydrogen sulfate (TBAHS)-
assistance32 was required for the efficient formation of azide 9,
and hydrogenation of benzyl glycoside 11 to give the free sugar
12 was more effective when using Pearlman’s catalyst than with
palladium on charcoal. An attempt to selectively deprotonate the
anomeric hydroxyl group of the free sugar 12 using LDA in a
mixture of THF and DMF and to react this species in situ with
tetrabenzyl pyrophosphate33,34 resulted in a low yield of the desired
dibenzyl sugar-1-phosphate, along with a number of unidentified
side products.
Therefore the deprotected sugar 12 was first peracetylated,
to give 13, followed by selective deacetylation of the anomeric
position using hydrazine acetate in DMF35 to give hemiacetal
14 in 88% yield. This time, deprotonation of the anomeric
hydroxyl group in 14 with LDA in THF followed by treatment
Commercially available N-acetyl-D-glucosamine (3) was chosen
as a starting material. This material was converted to the corre-
sponding benzyl glycoside 4 following a literature procedure29 and
the C-4/6 hydroxyl groups were protected,30 giving benzylidene
Scheme 1 Reagents and conditions: (a) BnOH, HCl(g); (b) PhCHO, ZnCl2, 89% over 2 steps; (c) MsCl, pyridine; (d) NaOAc, methyl cellosolve,
H2O, 89% over 2 steps; (e) MsCl, pyridine; (f) NaN3, DMF, TBAHS, 82% over 2 steps; (g) 80% AcOH, reflux; (h) H2, Pd-C then Ac2O, 84%
over 2 steps; (i) H2, Pd(OH)2, MeOH, 92%; (j) Ac2O, pyridine, DMAP, 71%; (k) NH2-NH2–AcOH, DMF, rt to 60 ◦C, 88%; (l) LDA, THF then
[(BnO)2PO]2O, 92%; (m) Pd-C, H2, MeOH then Et3N, 88%; (n) MeOH–H2O–Et3N 7 : 3 : 1, 4 days, quantitative; (o) uridine 5¢-monophosphomorpholidate
4-morpholine-N,N¢-dicyclohexylcarboxamidine salt (UMPM), pyridine, 4 ◦C, 14 days then SAX HPLC (gradient NH4HCO3), 49%; (p) Pt, O2, 100 ◦C,
NaHCO3 then SAX HPLC, 69%; (q) Dowex 50Wx8-200 (Na+ form), quantitative.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1203–1210 | 1205
©

View Online
of the resulting lithium salt with tetrabenzyl pyrophosphate, gave
smoothly the desired sugar-1-phosphate 15 in 92% yield after
chromatography on silica gel. This material proved to be solely
the a-anomer, as judged by 1H NMR spectroscopy (H1, dd, 3J1,2
= 3.2 Hz, 3J1,P = 5.6 Hz). Removal of the benzyl protecting groups,
followed by addition of Et3N yielded a bistriethylammonium salt
of 16 in 88% yield. Selective O-deacetylation gave the deprotected
sugar-1-phosphate 17 in quantitative yield.
solubility and consequently some spectral analyses were difficult
to perform (especially 13C NMR). In addition ES-MS of the
triammonium salt showed a very weak molecular ion and HR-
MS failed altogether. For these reasons, the triammonium salt of
1 was converted into the much more water soluble trisodium salt by
treatment with a cation exchange resin (Dowex 50Wx8-200, Na+
form). The trisodium salt of 1 gave a stronger 13C NMR spectrum
and reliable MS data.
With the sugar-1-phosphate 17 in hand, the next stage of
the preparation required installation of the sugar nucleotide
pyrophosphate bridge. Initially the enzymatic formation of the
pyrophosphate bond in 2 was attempted using the galactose-
1-phosphate uridyltransferase (Gal-1-PUT) protocol reported
by Errey et al.36 Although Gal-1-PUT has been shown to
possess a remarkably relaxed substrate specificity, on this oc-
casion 2,3-diacetamido sugar phosphate 17 proved not to
be a substrate for this enzyme. The formation of the py-
rophosphate bond in 2 was therefore achieved by chemi-
cal means. The bistriethylammonium salt§ of 17 was treated
with uridine 5¢-O-monophosphomorpholidate 4-morpholine-
N,N¢-dicyclohexylcarboxamidine salt in dry pyridine at 4 ◦C,37,38
the reaction being monitored by strong anion exchange HPLC. Af-
ter 2 weeks the conversion reached ~82%. The sugar nucleotide
was then isolated in 49% yield¶ by strong anion exchange
chromatography, as an ammonium salt of 2 and characterised
by NMR (see also Table 1 and 2 in ESI†). 31P NMR spectrum
confirmed successful formation of the pyrophosphate moiety (two
Synthetic UDP-Glc-2,3-diNAcA (1) has been subjected to a
number of biochemical analyses, including its assessment in (B.
pertussis) WlbD- and (P. aeruginosa) WbpI-mediated epimeri-
sation to give UDP-Man-2,3-diNAcA, as shown by capillary
electrophoresis and NMR spectroscopy.23 Neither enzyme utilised
UDP-GlcNAc, UDP-GlcNAcA or UDP-Glc-2,3-diNAc (2) as
substrates. With these authentic sugar nucleotides in hand, we
have also been able to demonstrate that the biosynthesis of the di-
N-acetylated sugar in the lipopolysaccharides of both B. pertussis
and P. aeruginosa occurs via an identical route, despite differences
in the arrangement of the gene clusters concerned.45 Further, we
have been able to reconstitute the UDP-MandiNAcA biosynthetic
pathway in vitro.42 Beyond assessment of their direct physiological
function, synthetic materials generated in this study have also
been useful in studies aimed at predicting protein function from
structure in the short chain dehydrogenase/reductase family43 and
for exploring the potential of Pasteurella heparosan synthases for
the chemoenzymatic synthesis of distinct monodisperse polymers
with unnatural structures.44
doublets at d -7.60 (Pb, JPa,Pb = 20.5 Hz) and d -9.38 (Pa, JPa,Pb
=
20.5 Hz)).
Conclusions
The final step in the synthesis required conversion of the hex-
ose unit to the corresponding uronic acid. Using a procedure that
we reported recently39 for the synthesis of UDP-GlcNAcA from
UDP-GlcNAc, the primary hydroxyl group in 2 was selectively ox-
idised using platinum-catalysed oxidation with molecular oxygen
to give the desired uronic acid. Whilst the practical oxidation of
2-acetamido-2-deoxy-sugar glycosides to the corresponding
uronic acids can be effected with TEMPO-mediated oxidation,40
in our hands the oxidation of the corresponding sugar nucleotides
proved problematic. However, this issue was resolved by resorting
to the more forcing conditions of Pt/O2 in boiling water.39 Again,
the course of the reaction was monitored by anion exchange
HPLC. After 24 h the conversion typically reached ~35%. A
new batch of the catalyst was then added and the reaction was
continued for a further 24 h, when the conversion reached ~70%.
The uronic acid was then isolated, by strong anion exchange
chromatography, as the triammonium salt of 1 in 69% yield. In
13C NMR spectra of 2 (diammonium salt) the C-6¢¢ signal changes
markedly (d 60.9 ppm, t, becomes 176.7 ppm, s) on oxidation
to 1. Moreover, on oxidation to 1 (triammonium salt) the C-4¢¢
signal in 2 (diammonium salt) sustained a downfield shift (from
67.9 to 70.6 ppm), a behaviour typical for the transformation
of hexoses into the corresponding uronic acids.41 Interestingly,
the triammonium salt of uronic acid 1 exhibited limited water
In conclusion, starting from N-acetyl-D-glucosamine the title
compound UDP-Glc-2,3-diNAcA (1) was synthesised in 17 steps
and in an overall yield of 9%. Using this compound we have
been able to provide evidence that B. pertussis WlbD and P.
aeruginosa WbpI are uridine diphosphate 2,3-diacetamido-2,3-
dideoxy-a-D-glucuronic acid 2-epimerases.23,45 This observation,
together with evidence for several related sugar nucleotides not
being substrates for these enzymes, has allowed us to confirm a
revised biosynthetic pathway for the synthesis of UDP-Man-2,3-
diNAcA.23 The availability of authentic, synthetic materials has
been key to these developments.
Experimental
General procedures
All chemicals were purchased as reagent grade and used without
further purification. TLC were performed on precoated silica
plates containing a fluorescence indicator. Silica gel (63 mm) was
used for analytical TLC. Compounds were visualised under UV
(254 nm) and by heating after dipping in a solution of 5% H2SO4
in EtOH.
HPLC was performed on a Biocad Sprint Perfusion Chro-
matography instrument. For analytical purposes or small scale
separations the mixtures were applied on a Poros HQ 50 strong
anion exchange column (10 ¥ 50 mm, CV ~4 mL). The column was
first equilibrated with 5 CV of 5 mM NH4HCO3 buffer, then the
mixture was applied and eluted with linear gradient of NH4HCO3
from 5 mM to 250 mM in 15 CV, then with NH4HCO3 250 mM in
§ The triethylammonium counterion in 17 facilitates the solubility of
the salt in pyridine and thus substantially improves the yield of the
pyrophosphate formation. Attempts to perform this reaction with the
sodium salt resulted in much lower yields.
¶ Some material was lost in the separation due to co-eluting impurities.
1206 | Org. Biomol. Chem., 2009, 7, 1203–1210
This journal is
The Royal Society of Chemistry 2009
©

View Online
5 CV at a flow rate of 12 mL min-1 and detection with an on-line
detector to monitor A265. Finally, the column was washed with
3 CV of 1 M NH4HCO3 followed by 1 CV of 5 mM NH4HCO3
at the same flow rate after each run. In the case of large scale
separations a Poros HQ 20 strong anion exchange column (16 ¥
100 mm, CV ~20 mL) was used under similar conditions (with a
different flow rate of 30 mL min-1 and detection with an on-line
detector to monitor A280).
(m, H5a, H6a), 2.13, 2.01, 2.00, 1.88, 1.87 (s, 2 ¥ CH3CONH a
anomer, 3 ¥ CH3(CO)O a anomer); dC (100 MHz; CDCl3; TMS)
170.9, 169.9, 169.7, 169.6, 168.0 (s, 5C, 2 ¥ CH3CONH a anomer,
3 ¥ CH3(CO)O a anomer), 92.0 (d, 1C, C1b), 89.3 (d, 1C, C1a),
69.1 (d, 1C, C5a), 66.9 (d, 1C, C4a), 60.8 (t, 1C, C6a), 50.9 (d,
1C, C2a), 49.6 (d, 1C, C3a), 22.2, 21.9, 20.0, 19.7, 19.6 (q, 5C, 2 ¥
CH3CONH a anomer, 3 ¥ CH3(CO)O a anomer); m/z (CI+) 406
([M + NH4]+, 23.1%), 389 ([M + H]+, 26.3), 329 (46.2), 77 (100);
HR-MS calcd for C16H25N2O9 [M + H]+ 389.1555, found 389.1559.
NMR spectra were recorded on a Varian Unity Plus spectrom-
eter at 400 MHz (1H) or 100.6 MHz (13C), or on a Bruker Avance
2,3-Diacetamido-4,6-di-O-acetyl-2,3-dideoxy-a,b-D-glucose (14).
Acetyl 2,3-diacetamido-4,6-di-O-acetyl-2,3-dideoxy-a,b-D-glu-
copyranoside (13) (525 mg, 1.35 mmol) was dissolved in N,N-
dimethylformamide (15 mL) and hydrazine hydrate (73 mL,
1.5 mmol) followed by glacial acetic acid (83 mL, 1.5 mmol) were
added. The mixture was stirred for 20 h at room temperature
but TLC showed incomplete conversion. For this reason, another
aliquot of hydrazine hydrate (73 mL, 1.5 mmol) followed by glacial
acetic acid (83 mL, 1.5 mmol) were added and the mixture was
heated to 60 ◦C for 3 h when the conversion was complete (by
TLC). The volatiles were evaporated under reduced pressure and
the residue was purified by column chromatography on silica gel
(dichloromethane–MeOH gradient from 20 : 1 to 5 : 1) to give pure
14 (413 mg, 88%) as an amorphous solid in a 7 : 1 a–b equilibrium
(by 13C NMR). Rf = 0.76 for a and 0.74 for b (dichloromethane–
methanol 5 : 1); [a]2D3 +51.7 (c 1.0, MeOH); dH (400 MHz; MeOH-
1
DPX-300 MHz NMR spectrometer at 121 MHz (31P). H NMR
spectra were referenced to dH 7.26 for CDCl3, dH 3.34 for CD3OD,
d
H 4.63 ppm for D2O, and 13C NMR spectra were referenced to dC
77.36 for CDCl3 and dC 49.86 for CD3OD. 31P NMR spectra were
referenced to dP 0.0 for external 85% phosphoric acid. Chemical
shifts of NMR signals recorded in D2O are reported with respect
to the methyl resonance of internal acetone at dH 2.22 ppm and
dC 30.89 ppm. Assignments were made with the aid of gCOSY
and gHSQC experiments. Coupling constants are reported in Hz.
In the NMR spectra of anomeric mixtures, only those signals
attributable to the minor anomer that are clearly discernable
in 1D spectra are reported. Optical rotations were measured at
ambient temperature on a Perkin-Elmer model 141 polarimeter
using a sodium lamp. Accurate mass electrospray ionization mass
spectra (ESI-MS) were obtained from the EPSRC National Mass
Spectrometry Service Centre, Swansea using positive ionization
mode on a Finnigan MAT 900 XLT mass spectrometer or from
the John Innes Centre metabolite analysis service on a Thermo
Finningan DecaXPplus mass spectrometer. For full experimental
data related to compounds 5, 7, 11 and 12 see ESI†.
d4; referenced to solvent residual peak at 3.34 ppm) 5.12 (d, 3J1,2
=
3.2 Hz, H1a), 4.94–4.89 (m, H4a overlapping with HDO signal),
3
4.74 (d, J1,2 = 8.8 Hz, H1b), 4.42–4.37 (m, H3a), 4.29–4.24 (m,
H5a, H6aa), 4.11–4.06 (m, H2a, H6ba), 2.07, 2.04 (s, 2 ¥ CH3CONH
a anomer), 1.97, 1.90 (s, 2 ¥ CH3(CO)O a anomer); dC (100 MHz;
MeOH-d4; referenced to solvent peak at 49.86 ppm) 174.8, 174.4,
173.3, 172.5 (s, 4C, 2 ¥ CH3CONH a anomer, 2 ¥ CH3(CO)O a
anomer), 98.2 (d, 1C, C1b), 92.9 (d, 1C, C1a), 71.6 (d, 1C, C4a),
69.5 (d, 1C, C5a), 64.7 (t, 1C, C6a), 54.4 (d, 1C, C2a), 51.9 (d, 1C,
C3a), 23.6, 23.3 (q, 2C, 2 ¥ CH3(CO)O a anomer), 21.5 (q, 2C, 2 ¥
CH3CONH a anomer); m/z (CI+) 364 ([M + NH4]+, 3.4%), 347
([M + H]+, 10.6), 77 (100); HR-MS calcd for C14H23N2O8 [M +
H]+ 347.1449, found 347.1450.
Attention. Extreme care should be observed when catalytic
hydrogenation is performed. Before any manipulation (addition
of new catalyst or work-up), hydrogen was removed by a water
aspirator and replaced by nitrogen (repeated 3 times). To prevent
self-ignition, the catalyst was collected on a filter paper, wetted
with a small amount of water and sealed in an air-tight container.
Acetyl 2,3-diacetamido-4,6-di-O-acetyl-2,3-dideoxy-a,b-D-glu-
copyranoside (13). To a suspension of 2,3-diacetamido-2,3-
dideoxy-a,b-D-glucose (12) (565 mg, 2.2 mmol) in pyridine
(20 mL), were added 4-dimethylaminopyridine (1 mg, 0.4 mol%)
and acetic anhydride (2.0 mL, 21.5 mmol). The mixture was stirred
at room temperature overnight. MeOH (1 mL) was added and
the volatiles were evaporated under reduced pressure. Traces of
pyridine were co-evaporated with toluene (2 ¥ 5 mL). The residue
was dissolved in ethyl acetate (10 mL), washed with 0.5 M HCl
(5 mL), water (5 mL) and dried over MgSO4. The aqueous layer
was saturated with NaCl and extracted with dichloromethane (4 ¥
20 mL) and dried over MgSO4. The combined organic extracts
were evaporated under reduced pressure and the residue was puri-
fied using column chromatography on silica gel (dichloromethane–
MeOH 10 : 1) to give pure 13 (593 mg, 71%) as an oil in a 8 : 1
2,3-Diacetamido-4,6-di-O-acetyl-2,3-dideoxy-a-D-glucopyra-
nose
1-dibenzylphosphate
(15). 2,3-Diacetamido-4,6-di-O-
acetyl-2,3-dideoxy-a,b-D-glucose (14) (698 mg, 2.02 mmol)
was azeotropically dried with THF and cooled to -78 ◦C
in absolute, freshly distilled THF (120 mL), and lithium
diisopropylamide (1.455 mL, 2.93 mmol of a 2 M solution
in heptane–THF–ethylbenzene) was added. After 10 min,
tetrabenzyl pyrophosphate (1.52 g, 2.8 mmol) was added. The
reaction was allowed to stir for 40 min and then warmed to 4 ◦C
and allowed to stir for a further 12 h. The precipitated solid
(lithium salt of dibenzyl phosphate) was filtered off, washed with
ethyl acetate and the filtrate was concentrated under reduced
pressure. The residue was dissolved in ethyl acetate (20 mL)
and washed with sat. aq. solution of NaHCO3 (10 mL) and
brine (10 mL), and then dried over Na2SO4 and concentrated
under reduced pressure. The residue was purified using column
chromatography on silica gel (dichloromethane–MeOH 20 : 1)
to give pure 15 (1.13 g, 92%) as a pale yellow oil. Rf = 0.63
(dichloromethane–methanol 10 : 1); [a]2D3 +52.6 (c 1.0, CHCl3);
dH (400 MHz; CDCl3; referenced to solvent residual peak at 7.26
1
a–b equilibrium (by H NMR). Rf = 0.86 for b and 0.82 for a
(dichloromethane–methanol 5 : 1); [a]2D3 +56.0 (c 1.0, CHCl3); dH
(400 MHz; CDCl3; TMS) 6.75 (d, 3J2,NH = 9.2 Hz, 2b–NH), 6.46
(d, 3J3,NH = 9.2 Hz, 3b–NH), 6.37 (d, 3J3,NH = 8.4 Hz, 3a–NH), 6.28
(d, 3J2,NH = 8.4 Hz, 2a–NH), 6.10 (d, 3J1,2 = 3.2 Hz, H1a), 5.61 (d,
3
3J1,2 = 8.8 Hz, H1b), 4.89 (t, 3J4,5 = J3,4 = 10.0 Hz, H4a), 4.43–4.35
(m, H3a), 4.31–4.21 (m, H2a, H6a), 4.19–4.13 (m, H2b), 4.06–3.97
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1203–1210 | 1207
©

View Online
ppm) 7.41–7.38 (m, 10H, aromatic), 6.11 (d, 1H, 3J2,NH = 8.8 Hz,
3J2,3 = 11.2 Hz, H3), 4.04 (td, 1H, 3J2,3 = 11.2 Hz, 3J1,2 = 2.0 Hz,
H2), 3.97 (dd, 1H, 3J4,5 = 9.4 Hz, 3J5,6 = 4.0 Hz, H5), 3.86 (bd, 1H,
2J6a,6b = 11.8 Hz, 3J5,6b ª 0 Hz, H6b), 3.73 (dd, 1H, 2J6a,6b = 11.8 Hz,
3J5,6a = 4.0 Hz, H6a), 3.50 (dd, 1H, 3J4,5 = 9.4 Hz, 3J3,4 = 10.4 Hz,
H4), 3.19 (q, 12H, 3JCH2,CH3 = 7.2 Hz, 2 ¥ (CH3CH2)3NH+), 1.97
2–NH), 5.86 (d, 1H, 3J3,NH = 8.8 Hz, 3–NH), 5.69 (dd, 1H, 3J1,2
=
3
3.2 Hz, J1,P = 5.6 Hz, H1), 5.16–5.04 (m, 4H, 2 ¥ CH2Ph), 4.88
(dd, 1H, 3J4,5 = 10.4 Hz, 3J3,4 = 9.6 Hz, H4), 4.39 (ddd, 1H, 3J3,4
= 9.6 Hz, 3J2,3 = 10.0 Hz, 3J3,NH = 8.8 Hz, H3), 4.24–4.15 (m, 2H,
H2, H6a), 4.08–4.02 (m, 1H, H5), 4.91–3.87 (m, 1H, H6b), 2.07,
2.02 (s, 6H, 2 ¥ CH3CONH), 1.92, 1.75 (s, 6H, 2 ¥ CH3(CO)O);
dC (100 MHz; CDCl3; referenced to solvent peak at 77.36 ppm)
171.2, 171.0, 170.7 (s, 4C, 2 ¥ CH3CONH, 2 ¥ CH3(CO)O), 135.5
3
(s, 6H, 2 ¥ CH3CONH), 1.33 (t, 18H, JCH3,CH2 = 7.2 Hz, 2 ¥
(CH3CH2)3NH+); dC (100 MHz; CH3OH-d4; referenced to solvent
peak at 49.86 ppm) 175.2, 174.5 (s, 2C, 2 ¥ CH3CONH), 95.7 (dd,
1C, 3JC1,P = 6.1 Hz, C1), 75.8 (d, 1C, C5), 70.4 (d, 1C, C4), 63.4 (t,
1C, C6), 54.9 (d, 1C, C3), 54.8 (dd, 1C, 3JC2,P = 10.0 Hz, C2), 48.3
(t, 6C, 2 ¥ (CH3CH2)3NH+), 23.7, 23.5 (q, 2C, 2 ¥ CH3CONH),
10.0 (q, 6C, 2 ¥ (CH3CH2)3NH+); m/z (ESI+) 545 ([M + H]+,
12.2%), 444 ([M-Et3NH + 2H]+, 63.5), 267 (100); HR-MS calcd
for C22H50N4O9P [M + H]+ 545.3310, found 545.3313.
3
(ds, 2C, JC arom.,P = 6.0 Hz, aromatic), 129.2, 129.1, 129.0, 128.3,
128.3 (d, 10C, aromatic), 96.0 (dd, 1C, 2JC1,P = 6.0 Hz, C1), 70.2
2
(dt, 2C, JCH2,P = 5.4 Hz, 2 ¥ CH2Ph), 70.2 (d, 1C, C5), 67.8 (d,
1C, C4), 61.5 (t, 1C, C6), 52.9 (dd, 1C, 3JC2,P = 7.6 Hz, C2), 50.7
(d, 1C, C3), 23.4, 23.0 (q, 2C, 2 ¥ CH3(CO)O), 20.9, 20.8 (q, 2C,
2 ¥ CH3CONH); m/z (ESI+) 629 ([M + Na]+, 43.2%), 607 ([M +
H]+, 6.8), 329 (91.9), 150 (100); HR-MS calcd for C28H36N2O11P
[M + H]+ 607.2051, found 607.2053.
UDP-2,3-Diacetamido-2,3-dideoxy-a-D-glucose, diammonium
salt (2). 2,3-Diacetamido-2,3-dideoxy-a-D-glucopyranose-1-
phosphate, bistriethylammonium salt (17) (243 mg, 0.45 mmol)
and uridine 5¢-O-monophosphomorpholidate 4-morpholine-
N,N¢-dicyclohexylcarboxamidine salt (287 mg, 0.417 mmol) were
separately dried by co-evaporation with absolute pyridine (2 ¥
5 mL). Both components were dissolved in absolute pyridine
(7 mL each) and the solutions were combined. The resulting
mixture was kept without stirring at 4 ◦C for 14 days. The
conversion reached 82% (by SAX HPLC on Poros HQ50, for
details see General procedures in the Experimental section). The
mixture was concentrated to dryness under reduced pressure.
The resulting syrup was dissolved in water (1 mL) and purified
using SAX HPLC on Poros HQ20 as indicated in the General
procedures in the Experimental section to give pure diammonium
salt of 2 (138 mg, 49%). dH (400 MHz; D2O; referenced to the
methyl resonance of internal acetone at 2.22 ppm) 7.93 (d, 1H,
2,3-Diacetamido-4,6-di-O-acetyl-2,3-dideoxy-a-D-glucopyra-
nose-1-phosphate, bistriethylammonium salt (16). 2,3-Diacet-
amido-4,6-di-O-acetyl-2,3-dideoxy-a-D-glucopyranose 1-diben-
zylphosphate (15) (308 mg, 0.51 mmol) was dissolved in MeOH
(10 mL) and palladium, 10% on carbon (20 mg) was added. The
mixture was hydrogenated for 24 h at 1 atm pressure. The catalyst
was removed by filtration through 2 filter papers and washed
with MeOH. To the filtrate triethylamine (1.2 mL) was added
and the volatiles were evaporated under reduced pressure to give
pure 16 (281 mg, 88%) as a white amorphous solid. Rf = 0.59
(dichloromethane–methanol–H2O 6 : 3 : 1); [a]2D3 +53.2 (c 1.0,
MeOH); dH (400 MHz; CH3OH-d4; referenced to solvent residual
3
3
peak at 3.34 ppm) 5.52 (dd, 1H, J1,2 = 2.8 Hz, J1,P = 7.2 Hz,
H1), 5.03 (dd, 1H, J4,5 = 10.0 Hz, J3,4 = 10.0 Hz, H4), 4.43
(dd, 1H, J3,4 = 10.0 Hz, J2,3 = 10.0 Hz, H3), 4.36 (dt, 1H, J4,5
3
3
3
3
3
3
3J5,6 = 8.2 Hz, H6), 6.00 (d, 1H, J1¢,2¢ = 4.0 Hz, H1¢), 5.94 (d,
3
2
3
3
3
= 10.0 Hz, J5,6 = 2.6 Hz, H5), 4.29 (dd, 1H, J6a,6b = 9.2 Hz,
3J5,6 = 2.6 Hz, H6a), 4.17–4.13 (m, 2H, H2, H6b), 3.13 (q, 12H,
3JCH2,CH3 = 7.2 Hz, 2 ¥ (CH3CH2)3NH+), 2.06, 2.03 (s, 6H, 2 ¥
CH3CONH), 1.97, 1.88 (s, 6H, 2 ¥ CH3(CO)O), 1.32 (t, 18H,
3JCH3,CH2 = 7.2 Hz, 2 ¥ (CH3CH2)3NH+); dC (100 MHz; CH3OH-
d4; referenced to solvent peak at 49.86 ppm) 174.6, 174.4, 173.2,
172.5 (s, 4C, 2 ¥ CH3CONH, 2 ¥ CH3(CO)O), 95.2 (dd, 1C, 3JC1,P
= 5.3 Hz, C1), 71.1 (d, 1C, C4), 70.5 (d, 1C, C5), 64.2 (t, 1C,
1H, J5,6 = 8.2 Hz, H5), 5.53 (dd, 1H, J1¢¢,Pb = 7.2 Hz, J1¢¢,2¢¢ =
3.2 Hz, H1¢¢), 4.38–4.34 (m, 2H, H2¢, H3¢), 4.29–4.27 (m, 1H,
H4¢), 4.25–4.23 (m, 1H, H5¢a), 4.22–4.19 (m, 1H, H5¢b), 4.12 (dd,
1H, 3J3¢¢,4¢¢ = 9.6 Hz, 3J2¢¢,3¢¢ = 11.2 Hz, H3¢¢), 4.08 (dt, 1H, 3J2¢¢,3¢¢
=
3
4
11.2 Hz, J1¢¢,2¢¢ = 3.2 Hz, J2¢¢,Pb = 2.8 Hz, H2¢¢), 3.98 (ddd, 1H,
3J4¢¢,5¢¢ = 10.0 Hz, J5¢¢,6¢¢a = 4.0 Hz, J5¢¢,6¢¢b = 2.4 Hz, H5¢¢), 3.87
3
3
(dd, 1H, 2J6¢¢a, 6¢¢b = 12.6 Hz, 3J5¢¢,6¢¢b = 2.4 Hz, H6¢¢b), 3.81 (dd, 1H,
2J6¢¢a, 6¢¢b = 12.6 Hz, 3J5¢¢,6¢¢a = 4.0 Hz, H6¢¢a), 3.62 (dd, 1H, 3J3¢¢,4¢¢
=
3
C6), 54.6 (dd, 1C, JC2,P = 6.8 Hz, C2), 53.2 (d, 1C, C3), 47.8 (t,
9.6 Hz, 3J4¢¢,5¢¢ = 10.0 Hz, H4¢¢), 2.01, 1.98 (s, 6H, 2 ¥ CH3CONH);
dC (100 MHz; D2O; referenced to the methyl resonance of internal
acetone at 30.89 ppm) 175.4, 175.1 (s, 2C, 2 ¥ CH3CONH), 168.1
(s, 1C, C4), 153.4 (s, 1C, C2), 142.1 (d, 1C, C6), 103.3 (d, 1C,
6C, 2 ¥ (CH3CH2)3NH+), 23.6, 23.5 (q, 2C, 2 ¥ CH3(CO)O), 21.6
(q, 2C, 2 ¥ CH3CONH), 10.1 (q, 6C, 2 ¥ (CH3CH2)3NH+); m/z
(ESI+) 629 ([M + H]+, 10.1%), 528 ([M-Et3NH + 2H]+, 19.8), 102
(100); HR-MS calcd for C26H54N4O11P [M + H]+ 629.3521, found
629.3519.
2
C5), 94.6 (dd, 1C, JC1¢¢,Pb = 6.0 Hz, C1¢¢), 89.1 (d, 1C, C1¢), 83.8
3
(dd, 1C, JC4¢,Pa = 9.9 Hz, C4¢), 74.4 (d, 1C, C2¢), 73.8 (d, 1C,
2
C5¢¢), 70.3 (d, 1C, C3¢), 67.9 (d, 1C, C4¢¢), 65.7 (dt, 1C, JC5¢,Pa
=
2,3-Diacetamido-2,3-dideoxy-a-D-glucopyranose-1-phosphate,
bistriethylammonium salt (17). 2,3-Diacetamido-4,6-di-O-ace-
tyl-2,3-dideoxy-a-D-glucopyranose-1-phosphate, bistriethylam-
monium salt (16) (281 mg, 0.45 mmol) was dissolved in a 7 :
3 : 1 mixture of MeOH–H2O–Et3N (11 mL) and stirred at room
temperature. After 4 days the conversion was complete (by TLC).
The volatiles were evaporated under reduced pressure to give 17
(243 mg, quantitative) as a pale yellow solid that was sufficiently
pure for analysis. Rf = 0.41 (dichloromethane–methanol–H2O 6 :
3 : 1); [a]2D3 +27.9 (c 1.0, MeOH); dH (400 MHz; CH3OH-d4;
referenced to solvent residual peak at 3.34 ppm) 5.50 (dd, 1H,
3J1,2 = 2.0 Hz, 3J1,P = 6.0 Hz, H1), 4.23 (dd, 1H, 3J3,4 = 10.4 Hz,
5.3 Hz, C5¢), 60.9 (t, 1C, C6¢¢), 53.1 (d, 1C, C3¢¢), 52.6 (dd, 1C,
3JC2¢¢,Pb = 9.2 Hz, C2¢¢), 22.7, 22.5 (q, 2C, 2 ¥ CH3CONH); dP
(D2O, 121 MHz, referenced to external 85% phosphoric acid at 0
ppm) -7.60 (d, Pb, JPa,Pb = 20.5 Hz), -9.38 (d, Pa, JPa,Pb = 20.5
Hz); m/z (ESI+) for acid 649 ([M + H]+, 1.0%), 294 (95.9), 245
(100); m/z (ESI-) for acid 647 ([M - H]-, 20.3%), 158 (40.5), 79
(100); HR-MS calcd C19H31N4O17P2 [M + H]+ 649.1154, found
649.1151.
UDP-2,3-Diacetamido-2,3-dideoxy-a-D-glucuronic acid, tri-
ammonium and trisodium salts (1). Adam’s catalyst (PtO2; 10 mg)
was hydrogenated in water (5 mL) under atmospheric pressure
1208 | Org. Biomol. Chem., 2009, 7, 1203–1210
This journal is
The Royal Society of Chemistry 2009
©

View Online
for 2 h. The apparatus was evacuated and flushed with nitrogen
3 times. The freshly prepared high-surface platinum in water was
added to a solution of UDP-2,3-diacetamido-2,3-dideoxy-a-D-
glucose, diammonium salt (2) (63 mg, 92 mmol) and NaHCO3
(40 mg, 480 mmol) in water (10 mL). The vigorously stirred
mixture was heated to 100 ◦C and oxygen was passed into the
mixture through a porosity 1 sinter. After 24 h the conversion
reached 35% (by SAX HPLC on Poros HQ50, for details see the
General procedures in the Experimental section). A new batch
of the catalyst was added prepared by hydrogenation of Adam¢s
catalyst (10 mg) in water (5 mL) and the oxidation was continued
overnight. After 48 h the conversion reached 71% (by SAX HPLC).
The catalyst was removed by filtration through 2 filter papers
and the filtrate was freeze-dried. The resulting solid was dissolved
in water (1 mL) and the solution was filtered using a 0.45 mm
Milipore Whatman filter. The filtrate was purified using SAX
HPLC on Poros HQ20 as indicated in the General procedures in
the Experimental section. Fractions containing the product were
pooled and freeze-dried to give incompletely neutralised material
(by 1H NMR). The resulting solid was taken into water (1 mL) and
aqueous ammonia (0.1 mL) was added. The solution was freeze-
dried again to give the triammonium salt of 1 (46 mg, 69%) as a
white solid in a purity of >98% (as judged by SAX HPLC). dH
(400 MHz; D2O; referenced to the methyl resonance of internal
acetone at 2.22 ppm) 7.85 (d, 1H, 3J5,6 = 8.0 Hz, H6), 6.01 (d, 1H,
3J1¢,2¢ = 4.4 Hz, H1¢), 5.91 (d, 1H, 3J5,6 = 8.0 Hz, H5), 5.56 (dd, 1H,
3J1¢¢,Pb = 7.2 Hz, 3J1¢¢,2¢¢ = 2.4 Hz, H1¢¢), 4.37–4.33 (m, 2H, H2¢, H3¢),
4.28–4.20 (m, 3H, H4¢, H5¢b, H5¢¢), 4.19–4.16 (m, 2H, H5¢a, H3¢¢),
Trust International Research Collaboration grant to R.A.F, and by
a grant from the Canadian Institutes of Health Research (CIHR)
to J.S.L. (MOP 14687). E.L.W. holds a CIHR Canada Graduate
Scholarship Doctoral Research Award, and J.S.L. holds a Canada
Research Chair in Cystic Fibrosis and Microbial Glycobiology.
We thank Dr Martin Tanner, UBC, for helpful discussions and
for provision of E. coli over-expression constructs used in assay
development and validation.
References
1 R. Parton, Biologicals, 1999, 27, 71–76.
2 F. G. A. Versteegh, J. F. P. Schellekens, A. Fleer and J. J. Roord, Rev.
Med., Microbiol., 2005, 16, 79–89.
3 E. S. Bamberger and I. Srugo, Eur. J. Pediatrics, 2008, 167, 133–139.
4 J. Parkhill, M. Sebaihia, A. Preston, L. D. Murphy, N. Thomson,
D. E. Harris, M. T. G. Holden, C. M. Churcher, S. D. Bentley, K. L.
Mungall, A. M. Cerdeno-Tarraga, L. Temple, K. James, B. Harris,
M. A. Quail, M. Achtman, R. Atkin, S. Baker, D. Basham, N. Bason,
I. Cherevach, T. Chillingworth, M. Collins, A. Cronin, P. Davis, J.
Doggett, T. Feltwell, A. Goble, N. Hamlin, H. Hauser, S. Holroyd, K.
Jagels, S. Leather, S. Moule, H. Norberczak, S. O¢Neil, D. Ormond, C.
Price, E. Rabbinowitsch, S. Rutter, M. Sanders, D. Saunders, K. Seeger,
S. Sharp, M. Simmonds, J. Skelton, R. Squares, S. Squares, K. Stevens,
L. Unwin, S. Whitehead, B. G. Barrell and D. J. Maskell, Nat. Genet.,
2003, 35, 32–40.
5 C. H. Raetz and C. Whitfield, Annu. Rev. Biochem., 2002, 71, 635–700.
6 Conjugate Vaccines, ed. J. M. Cruse and R. E. Lewis, Jr., S Karger AG,
Basel 1989.
7 C. Alexander and E. T. Rietschel, J. Endotoxin Res., 2001, 7, 167–202.
8 M. Caroff, J.-R. Brisson, A. Martin and D. Karibian, FEBS Lett., 2000,
477, 8–14.
9 Y. A. Knirel and N. K. Kochetkov, Biochemistry (Moscow), 1994, 59,
1325–1383.
3
3
4.16–4.12 (m, 1H, H2¢¢), 3.65 (dd, 1H, J3¢¢,4¢¢ = J4¢¢,5¢¢ = 9.6 Hz,
H4¢¢), 2.00, 1.99 (s, 6H, 2 ¥ CH3CONH); dC (100 MHz; D2O;
referenced to the methyl resonance of internal acetone at 30.89
ppm) 176.7 (s, 1C, C6¢¢OO-), 175.5, 175.1 (s, 2C, 2 ¥ CH3CONH),
167.0 (s, 1C, C4), 159.8 (s, C4 or C2 enol form), 156.2 (s, 1C, C2),
141.6 (d, 1C, C6), 103.5 (d, 1C, C5), 94.3 (dd, 1C, 2JC1¢¢,Pb = 5.6 Hz,
C1¢¢), 89.1 (d, 1C, C1¢), 83.5 (dd, 1C, 3JC4¢,Pa = 9.3 Hz, C4¢), 74.3
(d, 1C, C2¢), 74.0 (d, 1C, C5¢¢), 70.6 (d, 1C, C4¢¢), 70.3 (d, 1C, C3¢),
65.8 (dt, 1C, 2JC5¢,Pa = 5.5 Hz, C5¢), 52.6 (d, 1C, C3¢¢), 52.3 (dd, 1C,
3JC2¢¢,Pb = 8.8 Hz, C2¢¢), 22.7, 22.5 (q, 2C, 2 ¥ CH3CONH); dP (D2O,
121 MHz, referenced to external 85% phosphoric acid at 0 ppm)
-7.46 (d, Pb, JPa,Pb = 19.7 Hz), -9.32 (d, Pa, JPa,Pb = 19.7 Hz); m/z
(ESI-) for acid 683 ([M + Na - 2H]-, very weak), 705 ([M + 2Na -
3H]-, very weak); HR-MS failed to give a molecular ion. To a
dispersion of the triammonium salt of 1 (46 mg, 64 mmol) in water
(1 mL) was added a small amount of Dowex 50Wx8-200 (Na+
form) to achieve complete dissolution of the sugar nucleotide. This
mixture was applied on a column charged with Dowex 50Wx8-200
(Na+ form; 5 g) and the column was eluted with water. Fractions
containing the product were pooled and freeze-dried to give the
10 Y. A. Knirel, N. A. Hocharova, A. S. Shashkov, B. A. Dmitriev and
N. K. Kochetkov, Carbohydr. Res., 1981, 93, C12–C13.
11 D. Sibley, H. Rabin and M. G. Surette, Future Microbiology, 2006, 1,
53–61.
12 E. T. Harvill, A. Preston, P. A. Cotter, A. G. Allen, D. J. Maskell and
J. F. Miller, Infect. Immun., 2000, 68, 6720–6728.
13 M. Nilsson and T. Norberg, Carbohydr. Res., 2000, 327, 261–267.
14 A. Allen and D. Maskell, Mol. Microbiol., 1996, 19, 37–52.
15 T. Endo and S. Koizumi, Curr. Opin. Struct. Biol., 2000, 10, 536–541.
For representative examples of oligosaccharide production with en-
gineered bacteria see: (review); J. B. Zhang, P. Kowal, X. Chen and
P. G. Wang, Org. Biomol. Chem, 2003, 1, 3048–3053; C. Schaffer and
P. Messner, Glycobiology, 2004, 14, 31R–42R; L. Jolly, V. Tornare
and S. Kochhar, Synthesis of Carbohydrates Through Biotechnology,
ACS Symp. Ser., 2004, 873, 139–152; M. F. Feldman, M. Wacker, M.
Hernandez, P. G. Hitchen, C. L. Marolda, M. Kowarik, H. R. Morris,
A. Dell, M. A. Valvano and M. Aebi, Proc. Natl. Acad. Sci. U. S. A.,
2005, 102, 3016–3021; S. Drouillard, H. Driguez and E. Samain, Angew.
Chem., Int. Ed., 2006, 45, 1778–1780; N. Fierfort and E. Samain,
J. Biotechnol., 2008, 134, 261–265.
16 H.-M. Kuhn, U. Meier-Dieter and H. Mayer, FEMS Microbiol. Rev.,
1988, 54, 195–222.
17 U. Meier-Dieter, R. Starman, K. Barr, H. Mayer and P. D. Rick, J. Biol.
Chem., 1990, 265, 13490–13497; U. Meier-Dieter, K. Barr, R. Starman,
L. Hatch and P. D. Rick, J. Biol. Chem., 1992, 267, 746–753.
18 R. E. Campbell, S. C. Mosimann, M. E. Tanner and N. C. J. Strynadka,
Biochemistry, 2000, 39, 14993–15001.
19 V. Sri Kannathasan, A. G. Staines, C. J. Dong, R. A. Field, A. G.
Preston, D. J. Maskell and J. H. Naismith, Acta Crystallogr., Sect. D,
2001, D57, 1310–1312.
20 C. R. Sweet, A. A. Ribeiro and C. R. H. Raetz, J. Biol. Chem., 2004,
279, 25400–25410.
1
trisodium salt of 1 (46 mg, quantitative). H, 13C and 31P NMR
spectra were identical to the triammonium salt of 1. m/z (ESI+)
for acid 751 ([M + 4Na - 3H]+, weak), 685 [M + Na]+; m/z
(ESI-) for acid 749 ([M + 4Na - 5H]-, weak), 705 ([M + 2Na -
3H]-, weak); HR-MS calcd C19H25N4O18P2Na4 [M + 4Na - 3H]+
751.0224, found 751.0229.
21 W. L. Miller, C. Q. Wenzel, C. Daniels, S. Larocque, J.-R. Brisson and
J. S. Lam, J. Biol. Chem., 2004, 279, 37551–37558.
22 C. Wing, M. Rejzek, and R. A. Field, unpublished observations.
23 E. L. Westman, D. J. McNally, M. Rejzek, W. L. Miller, V. S.
Kannathasan, A. Preston, D. J. Maskell, R. A. Field, J.-R. Brisson
and J. S. Lam, Biochem. J., 2007, 405, 123–130.
Acknowledgements
These studies were supported by the UK Biotechnology and
Biological Sciences Research Council and through a Wellcome
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1203–1210 | 1209
©

View Online
24 J. Roppel, H. Mayer and J. Weckesser, Carbohydr. Res., 1975, 40, 31–40.
25 W. Meyer zu Reckendorf, Chem. Ber., 1964, 97, 1275–1285.
26 H. H. Baer and T. Neilson, J. Org. Chem., 1967, 32, 1068–1072.
27 R. D. Guthrie and D. Murphy, J. Chem. Soc., 1965, 3828–3834.
28 D. Charon, M. Mondange, J.-F. Pons, K. L. Blay and R. Chaby, Bioorg.
Med. Chem., 1998, 6, 755–765.
29 S. J. Hecker and M. L. J. Minich, Org. Chem., 1990, 55, 6051–6054.
30 P. H. Gross and R. W. J. Jeanloz, Org. Chem., 1967, 32, 2759–2763.
31 W. Meyer zu Reckendorf, Methods Carbohydr. Chem., 1972, 6, 266–269.
32 A. Calvo-Mateo and F. G. de las Heras, Carbohydr. Res., 1986, 155,
266–271.
33 H. G. Khorana and A. R. Todd, J. Chem. Soc., 1953, 2257–2260.
34 J. Elhalabi and K. G. Rice, Carbohydr. Res., 2002, 337, 1935–1940.
35 Y. Shoyama, M. Matsumoto and I. Nishioka, Phytochemistry, 1987,
26, 983–986.
36 J. C. Errey, B. Mukhopadhyay, K. P. R. Kartha and R. A. Field, Chem.
Commun., 2004, 2706–2707.
37 J. G. Moffatt and H. G. Khorana, J. Am. Chem. Soc., 1959, 81, 1265–
1265.
38 J. G. Moffatt and H. G. Khorana, J. Am. Chem. Soc., 1961, 83, 649–
658.
39 M. Rejzek, B. Mukhopadhyay, C. Q. Wenzel, J. S. Lam and R. A. Field,
Carbohydr. Res., 2007, 342, 460–466.
40 Z. Gyorgudeak and J. Thiem, Carbohydr. Res., 1995, 268, 85–92.
41 P. A. J. Gorin and M. Mazurek, Can. J. Chem., 1975, 53, 1212–1223.
42 E. L. Westman, D. J. McNally, A. Charchoglyan, D. Brewer, R. A.
Field, and J. S. Lam, manuscript in preparation.
43 J. D. King, N. J. Harmer, A. Preston, C. Palmer, M. Rejzek, R. A.
Field, T. L. Blundell and D. J. Maskell, J. Mol. Biol., 2007, 374, 749–
763.
44 A. E. Sismey-Ragatz, D. E. Green, M. Rejzek, R. A. Field and P. L.
DeAngelis, J. Biol. Chem., 2007, 282, 28321–28327.
45 E. L. Westman, A. Preston, R. A. Field and J. S. Lam, J. Bacteriol.,
2008, 190, 6060–6069.
1210 | Org. Biomol. Chem., 2009, 7, 1203–1210
This journal is
The Royal Society of Chemistry 2009
©