Neighbouring group participation vs. addition to oxacarbenium ions: Studies on the synthesis of mycobacterial oligosaccharides

Article abstract of DOI:10.1039/b914417j

Neighbouring group participation is frequently used to control the stereoselectivity of chemical reactions. Herein, we investigate the use of neighbouring group participation for the synthesis of disaccharides incorporating the mycobacterial sugar methylt

Full text of DOI:10.1039/b914417j

View Article Online / Journal Homepage / Table of Contents for this issue
www.rsc.org/obc
Volume 7 | Number 23 | 7 December 2009 | Pages 4801–5036
ISSN 1477-0520
FULL PAPER
W. Bruce Turnbull et al.
Highlightsin
Neighbouring group participation vs.
addition to oxacarbenium ions: studies
on the synthesis of mycobacterial
oligosaccharides
ChemicalBiology
In this issue...
1477-0520(2009)7:23;1-B

PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Neighbouring group participation vs. addition to oxacarbenium ions: studies
on the synthesis of mycobacterial oligosaccharides†
Susanne A. Stalford,a,b Colin A. Kilner,a Andrew G. Leachc and W. Bruce Turnbull*a,b
Received 17th July 2009, Accepted 11th August 2009
First published as an Advance Article on the web 21st September 2009
DOI: 10.1039/b914417j
Neighbouring group participation is frequently used to control the stereoselectivity of chemical
reactions. Herein, we investigate the use of neighbouring group participation for the synthesis of
disaccharides incorporating the mycobacterial sugar methylthioxylose. A bicyclic thioglycoside was
activated by methylation to generate a methylsulfonium group that would act both as the anomeric
leaving group, and also provide the methylsulfide group in the product. Model reactions indicated that
the bicyclic intermediate would also act as a participating group to direct the acceptor alcohol to the
lower a-face of the sugar. While the key sulfonium intermediate could be detected in the reaction
mixture, the glycosylation reaction proceeded with moderate stereoselectivity, apparently via an
SN1-type mechanism. Density functional theory calculations were used to compare our
methylthioxylose sulfonium ion with a trans-decalin-like sulfonium ion described by Boons and
co-workers to be an a-directing participating group (J. Am. Chem. Soc. 2005, 127, 12090). Our studies
show that even where a bicyclic sulfonium ion can be detected in the reaction mixture, caution should
be applied before invoking it as an intermediate on the reaction pathway.
Introduction
Neighbouring group participation has long been used to control
the stereochemistry of chemical reactions.1 Nowhere is this
concept more important than in the stereoselective synthesis of
carbohydrates.2-7 For example, glycosyl donors bearing an ester
group adjacent to the anomeric carbon can provide 1,2-trans-
b-glycosides with very high stereoselectivity.3 Amino,4 imido,5
iodo,6 and thio7 groups at this position can also control stere-
ochemistry effectively. More recently, the neighbouring group
participation strategy has been extended to the synthesis of
1,2-cis-a-glycosides.8,9 Boons and co-workers described an elegant
strategy to make 1,2-cis-a-glycosides using a chiral auxiliary that
contains a sulfide group (Scheme 1a).9 Upon activation of the
glycosyl donor, the oxacarbenium ion is trapped by the auxiliary
group to form a trans-decalin-like sulfonium ion. The incoming
nucleophile is thus directed to approach from the lower face
of the sugar to yield a 1,2-cis glycoside. Functional groups at
positions more distant from the anomeric centre can also influence
stereoselectivity,10,11 but the role of neighbouring group participa-
Scheme 1 (a) Boons’ 1,2-cis a-directing participating group; (b) NGP
with a 5-methylthio substituent; (c) a bicyclic thioglycoside which is
tion in these processes remains under debate.12,13 Herein we report
studies toward applying neighbouring group participation to the
pre-organised for NGP.
synthesis of mycobacterial oligosaccharides, and we use density
functional theory (DFT) calculations to rationalise the scope and
limitations of neighbouring group participation for stereoselective
a-glycosylation.
aSchool of Chemistry, University of Leeds, Leeds, LS2 9JT, UK. E-mail: w.b.
Methylthioxylofuranose (MTX) is a natural thiosugar that
turnbull@leeds.ac.uk; Fax: (+44) 1133436565; Tel: (+44) 1133437438
is attached to the lipoarabinomannan polysaccharide from
Mycobacterium tuberculosis.14-16 The MTX-a-1,4-mannose disac-
charide has been shown to inhibit production of the cytokine
TNF-a,16 and may also play a role in protecting M. tuberculosis
from oxidative stress.17 In the course of preparing some MTX-
mannosyl oligosaccharides for biological studies, we wondered
if a glycosyl donor containing the methylthio group could be
bAstbury Centre for Structural Molecular Biology, University of Leeds,
Leeds, LS2 9JT, UK
cAstraZeneca, Alderley Park, Macclesfield, Cheshire, UK
† Electronic supplementary information (ESI) available: Crystal structure
details for compound 8; coordinates for DFT structures; 1H NMR
spectra for novel compounds. CCDC reference number 735329. For ESI
and crystallographic data in CIF or other electronic format see DOI:
10.1039/b914417j
4842 | Org. Biomol. Chem., 2009, 7, 4842–4852
This journal is
The Royal Society of Chemistry 2009
©

used to control the stereochemistry of the glycosylation reaction
via neighbouring group participation (Scheme 1b). An oxacarbe-
nium ion intermediate could be trapped intramolecularly by the
methylthio group to give a 2.2.1 bicyclic sulfonium ion. If this
intermediate was to undergo an SN2 reaction with the acceptor
alcohol, it would yield the desired a-glycoside. We reasoned that
pre-organising the MTX glycosyl donor in the desired bicyclic
configuration could maximise the probability of it reacting by
neighbouring group participation. For example, activation of a
bicyclic thioglycoside by methylation (Scheme 1c),18 would lead
directly to the desired sulfonium ion intermediate. As both the
activating methyl group and the sulfide leaving group would form
part of the target molecule, such a glycosylation reaction could
also be considered to be atom efficient.19
Results and discussion
Glycosyl donor 8 was prepared from isopropylidene xylofuranose
1 in seven steps and 28% overall yield (Scheme 2). Benzylidene
acetal 2 was prepared and subjected to oxidative bromination
according to the method of Hollenberg et al.,20 to provide bromide
3 in 83% yield. Hydrolysis of the acetonide gave hemi-acetal 4
which was acetylated to yield triester 5 in 68% yield over two
steps. The bridging sulfur atom was introduced by activating the
anomeric acetate group with TMSOTf in the presence of thiourea,
followed by treatment with Et3N to give the anhydrosugar 6 in 85%
yield from 5. Much lower yields of the product were obtained when
BF3·OEt2 was used to activate the anomeric acetate group. The
ester protecting groups were removed under Zemple´n conditions,
and the resulting diol 7 was reprotected with benzyl ethers to yield
glycosyl donor 8. Single crystals of compound 8 were grown from
aqueous MeOH and subjected to X-ray analysis to prove that the
bicyclic ring system had been formed (Scheme 2).
Scheme 2 Reagents: (a) PhCH(OMe)2/CSA/DMF (84%); (b) NBS/
BaCO3/CCl4 (83%); (c) 60% aq. AcOH/H2SO4 (68%); (d) Ac2O/pyr
(100%); (e) (i) CS(NH2)2/TMSOTf/CH3CN (ii) Et3N (85%); (f)
NaOMe/MeOH (77%); (g) NaH/DMF/BnBr (91%).
Methylation of sulfide 8 with MeOTf in CD2Cl2, or with either
Me3OBF4 or MeOTf in nitromethane-d3, led to a single product
(Fig. 1). The H NMR spectrum displayed a new methyl signal
at 2.87 ppm and substantial downfield shifts for most of the
xylose protons, in particular H-19,13 and the pro-S H-5.21 Greater
deshielding of the pro-S H-5 (compared to the pro-R H-5) would
be consistent with the formation of the exo-(S)-sulfonium ion
9. Density functional theory (DFT) calculations also indicated
that the exo-(S)-isomer would be lower in energy than endo-
(R)-isomer (2.8 kcal mol-1 in the gas phase; 3.0 kcal mol-1 with
CH2Cl2 solvation); therefore, we would assign the S-configuration
to sulfonium ion 9. The spectrum in nitromethane-d3 remained
unchanged even after six days at room temperature indicating that
the sulfonium ion was stable in a polar aprotic solvent.
1
Fig. 1 1H NMR spectra (500 MHz, CD3NO2, 300 K) for (a) sulfide 8 and (b) methylsulfonium ion 9·BF4.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 4842–4852 | 4843
©

Methylation of sulfide 8 (87 mM) by Me3OBF4 in CD3CN
initially led to the same product as observed previously (Fig. 2a).
However, over the course of a few hours at room temperature,
the NMR spectrum changed dramatically, indicating that the
sulfonium ion had been converted to a new species which was
stable for several days (Fig. 2b-c). The signal for H-2 moved
1.7 ppm downfield, while that for H-4 moved 1.3 ppm upfield,
suggesting that the compound had undergone a substantial change
in conformation. Upfield shifts for H-5, H-5¢ and the methyl signal
were consistent with opening of the sulfonium ring to give a sulfide
group.15 The multiplicity of the anomeric proton signal changed
from singlet to doublet (J1,2 = 5.9 Hz) which is typical for an
a-configured xylofuranosyl product.15 The high chemical shift for
H-1 indicated that there was still a strongly electron-withdrawing
group at the anomeric centre which we presume to be the nitrilium
ion 10 (Fig. 2d; R = Bn).22
DFT calculations predict that the a-configured nitrilium ion 10
(R = Me) would be favoured in the gas phase by 0.3 kcal mol-1, as
◦
≡
O-2 can stabilise the adjacent cation. The C-N C angle is 173 in
the lowest energy conformation suggesting that the stabilisation
is not only electrostatic in nature, but there is also a hybridisation
change which would be consistent with orbital overlap between
O-2 and the nitrilium p* orbital. The lowest energy conformation
of the b-configured nitrilium is similarly stabilized by interaction
between the electron poor p system of the nitrilium group and
the sulfur atom at C-5. This transannular interaction reduces the
surface available for solvation and hence, when solvation is added
to the model, the energy difference between the a- and b-isomers
is increased to 1.8 kcal mol-1. Repeating the reaction in a 1:1
mixture of CD3CN and CD3NO2 led to the same intermediate,
but the reaction rate was reduced by approximately a factor of
two. As nitromethane and acetonitrile have essentially the same
dielectric constant,23 the change in rate may be attributed solely
to the change in nucleophile concentration. Therefore, our data
are consistent with an SN2 mechanism. Alternatively, it is possible
that the reaction could be stepwise if trapping the oxacarbenium
ion by MeCN were slower than cyclisation onto the sulfur atom.
The exclusive formation of the a-configured nitrilium ion may be
under kinetic control, as has been reported for pyranosyl nitrilium
ions,24 although this configuration would also be expected were
the reaction to run to equilibrium. As nitrilium intermediates
are usually b-directing in glycosylation reactions,25 all subsequent
experiments were performed in dichloromethane.
Donor 8 was treated with methyl triflate in the presence of the
hindered base 2,6-di-t-butyl-4-methylpyridine (DTBMP), and the
reaction was monitored by TLC until all of the starting material
had been consumed. Addition of secondary alcohol acceptors
12-14 led to the formation of disaccharides 15-17, respectively, in
modest to good yields (Scheme 3). In contrast to the reaction with
acetonitrile, mixtures of a- and b-glycosides were obtained in all
cases. Lowering the reaction temperature gave little improvement
to the anomeric ratio. If both the SN1 and SN2 pathways were to
be in operation, one would expect an increase in flux down the
SN2 pathway upon increasing the concentration of the reactants.
However, neither the product ratio, nor the reaction rate changed
significantly on increasing the concentration of the reactants;
therefore, the glycosylation reaction appears to proceed exclusively
through an SN1 mechanism. The acetyl groups were removed from
disaccharide 17 under Zemple´n conditions before Birch reduction
of the benzyl ethers. The resulting disaccharides 18a and 18b
were readily separated by flash chromatography, to provide an
a-linked disaccharide that is suitable for further elaboration into
glycoconjugates.
The synthetic strategy outlined above provides a convenient
route to MTX-oligosaccharides, in which the methylthio sub-
stituent is introduced during the glycosylation step. However,
the stereoselectivity of the glycosylation reaction was lower
than anticipated, in particular when compared to the highly
stereoselective ring opening of sulfonium ion 9 by acetonitrile.
So, why is the Boons system (Scheme 1a) more stereoselec-
tive than other bicyclic sulfonium ions, e.g., our MTX system
(Scheme 1c)? Woerpel and co-workers recently observed that the
presence of a glycosyl sulfonium ion in a reaction mixture does
Fig. 2 1H NMR spectra (300 MHz, CD3CN, 300 K) showing the
conversion of methylsulfonium ion 9 (87 mM; R = Bn) into putative
nitrilium adduct 10 (R = Bn) at different times after mixing sulfide 8 and
Me3OBF4: (a) 30 minutes, (b) 5 hours and (c) 6 days. (d) structures for the
two possible acetonitrile adducts (10, 11; R = Me) with relative energies
from DFT calculations (B3LYP/6-31+G*).
4844 | Org. Biomol. Chem., 2009, 7, 4842–4852
This journal is
The Royal Society of Chemistry 2009
©

To address the question of stereoselectivity for our MTX donor
and the Boons’ system, we undertook DFT calculations on two
model sulfonium ions 26 and 30 (Schemes 5 and 6). In order
to reduce computational time, simplified structures were used in
which benzyl protecting groups, the equatorial substituent on
the oxathiane ring and primary carbon on the glucopyranosyl
ring were all abbreviated to methyl groups. A correction for the
energy of solvation was applied to each structure using parameters
for MeCN or CH2Cl2, as appropriate. Generally, trapping of
cations by nucleophiles is barrierless in the gas phase but in this
intramolecular case, there is a conformational barrier between
the sulfonium ion and the oxacarbenium ion. The SN1 transition
states are therefore those for an internal rotation. In general, the
gap between SN1 and SN2 activation energies was reduced after
inclusion of the correction for solvation; presumably, the SN2
transition state benefited less from the solvation term than either
the sulfonium ions or SN1 transition states.
Calculations for the reaction of xylofuranosyl sulfonium ion
26 with MeCN indicated only a small preference for the SN2
pathway (Scheme 5b; DDG‡ = -0.3 kcal mol-1 for SN2 vs. SN1;
MeCN solvation parameters), suggesting that both mechanisms
would be in operation under standard state conditions (i.e., 1 M
concentration). Therefore, we conclude that the SN2 process ob-
served when using MeCN as the solvent arises from the very high
concentration of the nucleophile (ca. 19 M) that would increase
the SN2 reaction rate relative to that for the SN1 mechanism.
Ammonia was chosen as a model nucleophile for comparison
of the Boons-type sulfonium ion 30 and our MTX model 26.
Ammonia has higher symmetry and basicity than an alcohol
which remove the issues of (1) the directionality of the O-H bond
vector as it approaches the glycosyl donor, and (2) the precise
point at which the proton is removed during the reaction.26 As
the structures do not directly reproduce the reactions studied
in vitro, it is important that one should apply caution when directly
comparing the SN1 and SN2 activation barriers for a given glycosyl
donor. Nevertheless, as the same nucleophile is used for reaction
with ions 26 and 30, general trends arising from the calculations
should still be valid.
The SN2 reaction of MTX sulfonium ion 26 (Scheme 5c) with
ammonia is predicted to have a lower activation barrier than for
reaction with MeCN (Scheme 5b). This observation is consistent
with the greater nucleophilicity of ammonia. Both gas phase and
solution calculations also suggest that sulfonium ion 26 should
undergo a more facile SN2 reaction than the Boons-type ion
30. In this case, the lower activation energy for MTX ion 26
probably arises from release of ring strain in the SN2 transition
state,13 and also a more optimal angle between the nucleophile,
C-1 and sulfide leaving group (168◦ vs. 136◦ for transition states
25 and 29, respectively). The difference between these activation
energies is amplified further when solvation effects are added to the
model.
Scheme 3 Reagents: (a) donor 9/DTBMP/CH2Cl2 (64% 15; 48% 16; 47%
17); (b) NaOMe/MeOH (69%); (c) Na/NH3 (l)/THF (46% 18a; 17% 18b).
not necessarily mean that it is an intermediate on the reaction
pathway.13 For example, sulfonium ion 21 (Scheme 4) could
be detected by NMR spectroscopy; however, C-glycosylation
products from this intermediate were consistent with an SN1
mechanism that proceeded through oxacarbenium ion 20. They
also reported that donor 22 failed to react via neighbouring group
participation. In this case, the authors were unable to detect the
formation of the bicyclic intermediate 24 in the reaction mixture,
even though modelling had predicted that the sulfonium ion
should be 5-6 kcal mol-1 more stable than the oxacarbenium ion
intermediate 23. It is unclear if the lack of stereocontrol in this
example results from a failure to form the sulfonium ion under
the reaction conditions, or a Curtin-Hammett kinetic scenario
proposed for the analogous 2.2.2. bicyclic system 21.13
The calculations also predict that the MTX sulfonium ion 26
would have a lower SN1 activation barrier than trans-decalin
sulfonium ion 30 in CH2Cl2 solution (Schemes 5c and 6b).
Although one might attribute this difference to releasing ring
strain in bicycle 26, the gas phase energies for SN1 transition
states 27 and 31 are essentially the same. It is likely that the
effect of ring strain is offset by the greater intrinsic stability of an
alkyl sulfonium ion relative to an aryl sulfonium ion.13 The more
Scheme 4 Model systems studied by Woerpel and co-workers (ref. 13).
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 4842–4852 | 4845
©

Scheme 5 (a) Possible reaction pathways for sulfonium ion 26. Energy profiles for reaction of sulfonium ion 26 with (b) MeCN or (c) NH3 were
determined using density functional theory calculations performed at the B3LYP/6-31+G* level of theory. DG‡ and DG◦ values are quoted in kcal mol-1
relative to the lowest energy conformation of sulfonium ion 26. Values for gas phase calculations are given in normal text, while values corrected for
solvation effects are given in bold. Solvation correction employed the integral equation formalism polarisable continuum model (IEFPCM) with united
atom Kohn–Sham (UAKS) radii.
Scheme 6 (a) Possible reaction pathways for sulfonium ion 30. (b) Energy profiles for reaction of sulfonium ion 30 with NH3 were determined as
described in Scheme 5. Values for gas phase calculations are given in normal text, while values corrected for solvation effects (CH2Cl2) are given in bold.
facile SN1 pathway calculated for ion 26 arises only from different
effects of solvation when compared to the same reaction of
ion 30.
When taken together, our calculations would predict that MTX
sulfonium ion 26 should be more likely to react by an SN 2
mechanism than the Boons-type ion 30 (DDG‡(SN2 - SN1) =
-4.0 vs. -0.3 kcal mol-1 for ions 26 and 30, respectively, in CH2Cl2
solution). However, neither the stereochemistry of the products
nor the reaction kinetics observed experimentally are consistent
with an SN2 pathway (vide supra).
4846 | Org. Biomol. Chem., 2009, 7, 4842–4852
This journal is
The Royal Society of Chemistry 2009
©

One additional factor that has not been considered thus far
is the relative energy of the sulfonium ion reactants and their
oxacarbenium ions 28 and 32. The smaller this gap, the larger
the concentration of oxacarbenium ion, and the more significant
the contribution of an SN1 reaction that would presumably be
less stereoselective. The MTX oxacarbenium ions 28 (Fig. 3) are
closer in free energy to the cyclised sulfonium ion 26 by about
5 kcal mol-1 in CH2Cl2 solution when compared to sulfonium ion
30 and oxacarbenium ion 32. It may be that these systems operate
at the borderline between SN1 and SN2 mechanisms and that small
changes in the equilibrium concentration of the oxacarbenium
relative to the sulfonium, or very high concentrations of the
trapping nucleophile (e.g., 19 M MeCN), can tip the balance one
way or the other.
conformation should be dictated by placing O-3 in a pseudoaxial
position.27
While it is difficult to model the intermolecular addition of
nucleophiles to oxacarbenium ions, covalent adducts of the nucle-
ophile and cation may give some insight into the stereoselectivity
of the addition step.26 The lowest energy a- and b-ammonium
adducts 33 and 34, respectively, also adopted ring conformations
in which the O-2 and O-3 substituents were in pseudoaxial
conformations (Fig. 3c and d). The b-anomer was favoured by
1.2 kcal mol-1 in the gas phase and 0.6 kcal mol-1 in CH2Cl2.
If we can assume that the ammonium ion adducts reflect the
structures and relative energies of the transition states, then
the model would suggest that the b-configured products would
be favoured. This observation is in qualitative agreement with
Woerpel’s experimental observations that a nucleophile should
add cis to a pseudoaxial O-3 group and trans to a pseudoaxial O-2
group on a furanosyl oxacarbenium ion.27 However, we observe
a-selectivity in all of our glycosylation reactions (Scheme 3).
One possible explanation is that the MTX donor reacts through
neither the sulfonium ion 9, nor the oxacarbenium ion 35
(Scheme 7). Instead, ion 35 is initially trapped as a glycosyl triflate
36a/b,11,28,29 and the stereoselectivity is governed by the relative
stabilities (or reactivities) of the two triflates.28 Alternatively, the
a-glycoside 37a could be formed principally from the b-triflate
36b, while the b-glycoside 37b could be formed from either the
a-triflate 36a or the oxacarbenium ion 35. The major species
present in the reaction mixture was always the sulfonium ion,
and it was not possible to detect the presence of a glycosyl triflate.
Therefore, if triflates 36a/b are intermediates in the reaction, they
always remain minor components of the reaction mixture.
Scheme 7 Putative glycosyl triflate intermediates.
Conclusions
We have demonstrated that a bicyclic thioglycoside provides a
concise route to MTX oligosaccharides. NMR experiments in
acetonitrile and DFT calculations would have predicted that
bicyclic sulfonium ion 9 should react via an SN2 mechanism
to yield a-glycosides with high stereoselectivity. However, the
glycosylation results demonstrate that such seductive predictions
can not always be trusted. The observed stereoselectivity is
more consistent with an SN1 mechanism involving rate-limiting
formation of an oxacarbenium ion which forms a contact ion pair
with its triflate counterion. Our results support the observations
of Woerpel13 that the presence of a glycosyl sulfonium ion in the
reaction mixture is not sufficient information to conclude that it
will react via an SN2 mechanism to yield the products.39 The Boons
Fig. 3 Overlay of the five lowest energy conformations of oxacarbenium
ion 28 having O-2 and O-3 substituents in (a) pseudoaxial and (b) pseu-
doequatorial positions. (c) Lowest energy conformation of a-ammonium
adduct 33. (d) Lowest energy conformation of b-ammonium adduct 34.
Our experimental data would imply that MTX donor 9 reacts by
an SN1 mechanism via the oxacarbenium ion. MTX oxacarbenium
ion 28 can adopt two distinct families of conformations in
which O-2 and O-3 both adopt pseudoaxial or pseudoequatorial
positions (Fig. 3a and b). The conformations with pseudoaxial
methoxy substituents are on average favoured by 3 kcal mol-1 in
the gas phase and 1.5 kcal mol-1 in CH2Cl2. This result is in line
with Woerpel’s observation that a furanosyl oxacarbenium ion
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 4842–4852 | 4847
©

participating group (Scheme 1a)9 is much more stereoselective
than the bridged bicyclic system reported here; however, we note
that our DFT calculations predict that the Boons-type sulfonium
ion 30 is less likely to react via an SN2 mechanism than MTX
sulfonium ion 26. Indeed, we have observed that trans-decalin
sulfonium ions that are even more stable than Boons’ compound
do not necessarily lead to a-glycosides; these studies will be
reported in due course.
(1H, dd, J1,OH 5.8, J1,2 0.8, H-1), 5.39 (1H, dd, J3,4 5.1, J2,3 1.7, H-3),
4.78 (1H, ddd, J3,4 5.1, J4,5 2.0, J4.5¢ 1.8, H-4), 4.37 (1H, ddd, J2,OH
2.75, J2,3 1.7, J1,2 0.8, H-2), 3.69 (1H, d, J4,5 2.0, H-5), 3.67 (1H,
d, J4,5¢ 1.8, H-5¢), 3.01 (1H, d, J1,OH 5.8, OH-1), 2.75 (1H, d, J2,OH
3.6, OH-2); dC (75 MHz; CDCl3); a-anomer 133.8-128.7 (PhCO),
96.4 (C-1), 79.4 (C-3), 77.3 (C-4), 76.1 (C-2), 28.3 (C-5); b-anomer
133.8-128.7 (PhCO), 103.3 (C-1), 80.7 (C-2), 80.3 (C-4), 78.7 (C-3),
29.3 (C-5); m/z (ES); 339.0, 341.0 ([M+Na]+), 655.0, 657.0, 659.0
([2M+Na]+); Found: [M+Na]+ 338.9842, C12H13NaO5Br requires
338.9839.
Experimental section
All solvents were dried prior to use, according to standard
methods.30 Otherwise, commercial reagents were used without
further purification. All concentrations and evaporations were
performed in vacuo unless stated. Analytical TLC was performed
on silica gel 60-F254 (Merck) with detection by fluorescence and/or
by charring following immersion in 5% H2SO4/MeOH. Flash
chromatography was performed with silica gel 60 (Merck). Melting
points were determined on a Reichert hot stage apparatus. Optical
rotations were measured at the sodium D-line with an Optical
Activity AA-1000 polarimeter. [a]D values are given in units of
10-1 deg cm2 g-1. 1H and 13C NMR spectra were recorded at
300 K on either a Bruker Avance 500 spectrometer or a Bruker
1,2-Di-O-acetyl-3-O-benzoyl-5-bromo-5-deoxy-D-xylofuranose (5)
Acetic anhydride (10 mL) was added to a stirred solution of 3-O-
benzoyl-5-bromo-5-deoxy-D-xylofuranose (4) (0.55 g, 1.73 mmol)
in pyridine (10 mL). The reaction was stirred under a nitrogen
atmosphere for 1 hour. The reaction mixture was diluted with
EtOAc (100 mL) and washed with aq. 1M HCl solution (100 mL),
sat. aq. NaHCO3 solution (100 mL) and sat. aq. NaCl solution
(100 mL). The organic extracts were dried (Na2SO4), filtered and
evaporated to give an orange oil, which was purified by flash
chromatography (silica gel; Hex-EtOAc, 2:1) giving the acetylated
bromide 5 (696 mg, 100%) as a colourless syrup; a-b 59:41 (Found:
C, 48.0; H, 4.4. C16H17O7Br requires C, 47.9; H 4.3%); nmax/cm-1
1
DPX 300 spectrometer. H NMR and 13C NMR spectra were
=
1755 and 1729 (C O); dH (500 MHz; CDCl3); a-anomer 8.08-
referenced using tetramethylsilane or residual solvent signals as
internal standards.31 Signals were assigned using a combination of
COSY and HMQC experiments, and where appropriate NOESY
experiments. The following abbreviations were used to explain the
signal multiplicities or characteristics: s, singlet; bs, broad singlet;
d, doublet; dd, double doublet; ddd, double double doublet; dt,
doublet of triplets; m, multiplet; t, triplet. For disaccharides 15-18
the MTX is designated residue “a” and the glucosyl/mannosyl
residue is designated ring “b”. Mass spectra were acquired
on a Micromass LCT-KA111 electrospray mass spectrometer,
or Bruker MicroTOF electrospray mass spectrometer. Infra-red
spectra were recorded on a Perkin Elmer Spectrum One FT-IR
Spectrometer. Elemental analyses were performed by the School
of Chemistry microanalysis service using a Carlo Erba 1108
Elemental Analyzer.
7.47 (5H, m, PhCO), 6.53 (1H, d, J1,2 4.6, H-1), 5.79 (1H, dd,
J3,4 5.9, J2,3 5.0, H-3), 5.51 (1H, dd, J2,3 5.0, J1,2 4.6, H-2), 4.80
(1H, ddd, J4,5 6.3, J4,5¢ 5.5, J3,4 5.9, H-4), 3.54 (1H, dd, J5,5¢ 10.8,
J4,5 6.3, H-5), 3.45 (1H, dd, J5,5¢ 10.8, J4,5¢ 5.5, H-5¢), 2.23 (3H, s,
MeCO), 2.16 (3H, s, MeCO); b-anomer 8.08-7.47 (5H, m, PhCO),
6.21 (1H, s, H-1), 5.65 (1H, dd, J3,4 4.8, J2,3 0.9, H-3), 5.31 (1H,
d, J2,3 0.9, H-2), 4.81 (1H, m, H-4), 3.59 (1H, s, H-5), 3.58 (1H,
d, J4,5¢ 0.8, H-5¢), 2.16 (3H, s, MeCO), 2.11 (3H, s, MeCO); dC
(75 MHz; CDCl3); a-anomer 133.9-128.7 (PhCO), 93.3 (C-1),
77.7 (C-4), 76.8 (C-2), 75.9 (C-3), 28.0 (C-5), 20.9 (MeCO), 20.4
(MeCO); b-anomer 133.9-128.7 (PhCO), 99.1 (C-1), 82.3 (C-4),
79.5 (C-2), 74.9 (C-3), 27.6 (C-5), 22.2 (MeCO), 20.7 (MeCO); m/z
(ES); 423.0, 425.0 ([M+Na]+), 823.0, 825.0, 827.0 ([2M+Na]+);
Found: [M+Na]+ 423.0048, C16H17NaO7Br requires 423.0050.
3-O-Benzoyl-5-bromo-5-deoxy-D-xylofuranose (4)
2-O-Acetyl-1,4-anhydro-3-O-benzoyl-5-deoxy-5-thio-a-D-
xylopyranose (6)
Sulfuric acid (100 mL) was added to a solution of bromide 320 (1.0 g,
2.80 mmol) and 60% v/v aq. acetic acid (10 mL). The solution
was heated at 70 ◦C for 2 hours, before cooling to room temp.
The mixture was neutralised with sat. aq. NaHCO3 solution and
extracted with EtOAc (100 mL). The organic extracts were washed
with sat. aq. NaCl solution (2 ¥ 100 mL), dried (Na2SO4), filtered
and evaporated to give an orange oil, which was purified by flash
chromatography (silica gel; Hex-EtOAc, 3:2) to give 3-O-benzoyl-
5-bromo-5-deoxy-D-xylofuranose (4) (605 mg, 68%) as a colourless
amorphous solid, a-b 81:19 (Found: C, 45.7; H, 4.3. C12H13O5Br
Trimethylsilyl trifluoromethanesulfonate (4.16 mL, 23.0 mmol)
was added to a solution of diacetate 5 (6.15 g, 15.3 mmol) and
thiourea (1.28 g, 16.8 mmol) in acetonitrile (100 mL) and stirred
under a nitrogen atmosphere at 80 ◦C for 1 hour. The reaction
mixture was cooled to RT while bubbling nitrogen gas through
the solution. Triethylamine (6.41 mL, 46.0 mmol) was added and
the solution was stirred for a further hour. The reaction mixture
was diluted with EtOAc (250 mL) and washed with sat. aq. NaCl
solution (100 mL). The organic extracts were dried (Na2SO4),
filtered and evaporated to give a brown oil, which was purified
by flash chromatography (silica gel; Hex-EtOAc, 3:1) to give the
epithiosugar 6 (3.82 g, 85%) as a yellow syrup (Found: C, 57.2; H,
requires C, 45.5; H 4.1%); nmax/cm-1 3425 (OH), 1718 (C O); dH
=
(500 MHz; CDCl3); a-anomer 8.05-7.47 (5H, m, PhCO), 5.61 (1H,
dd, J1,OH 6.1, J1,2 4.3, H-1), 5.44 (1H, dd, J3,4 4.8, J2,3 2.8, H-3), 4.80
(1H, ddd, J4,5 6.9, J4,5¢ 6.5, J3,4 4.8, H-4), 4.38 (1H, ddd, J2,OH 4.6,
J1,2 4.3, J2,3 2.8, H-2), 3.85 (1H, d, J1,OH 6.1, OH-1), 3.57 (1H, dd,
26
4.5. C14H14SO5 requires C, 57.1; H 4.8%); [a]D +142.4 (c 1.0 in
CHCl3); nmax/cm-1 1731 and 1725 (C O); dH (500 MHz; CDCl3);
=
J
5,5¢ 10.4, J4,5 6.9, H-5), 3.53 (1H, dd, J5,5¢ 10.4, J4,5¢ 6.5, H-5¢), 3.37
8.06-7.46 (5H, m, PhCO), 5.57 (1H, s, H-1), 5.40 (1H, dd, J3,4 5.5,
(1H, d, J2,OH 4.6, OH-2); b-anomer 8.05-7.47 (5H, m, PhCO), 5.42
J
4,5¢ 4.3, H-4), 5.16 (1H, dd, J3,4 5.5, J2,3 1.2, H-3), 5.11 (1H, d, J2,3
4848 | Org. Biomol. Chem., 2009, 7, 4842–4852
This journal is The Royal Society of Chemistry 2009
©

1.2, H-2), 3.11 (1H, d, J5,5¢ 9.9, H-5), 2.91 (1H, dd, J5,5¢ 9.9, J4,5¢
4.3, H-5¢), 2.12 (3H, s, CH3CO); dC (75 MHz; CDCl3); 134.0-128.9
(PhCO), 85.8 (C-1), 82.4 (C-2), 79.4 (C-4), 78.9 (C-3), 30.3 (C-5),
21.1 (CH3CO); m/z (ES); 317.0 ([M+Na]+), 611.1 ([2M+Na]+);
Found: [M+Na]+ 317.0450, C14H14NaSO5 requires 317.0454.
standing. Recrystallisation from EtOAc gave pent-4-enyl 2,3-di-
O-acetyl-4,6-O-benzylidene-a-D-mannopyranoside (1.534 g, 36%)
as colourless needles; m.p. 94-95 ◦C (from EtOAc); [a]D25 +26.0 (c
1.1 in CHCl3) (Found: C, 63.1; H, 6.7. C22H28O8 requires C, 62.9; H
6.7%); nmax/cm-1 1750 (C O), 1641 (C C); dH (500 MHz; CDCl3);
7.47-7.35 (5H, m, PhC), 5.81 (1H, m, CH2=CH), 5.58 (1H, s,
PhCH), 5.42 (1H, dd, J2,3 3.4, J3,4 10.0, H-3), 5.35 (1H, dd, J1,2 1.3,
=
=
1,4-Anhydro-5-deoxy-5-thio-a-D-xylopyranose (7)
J
2,3 3.4, H-2), 5.07-4.99 (2H, m, CH2=CH), 4.75 (1H, d, J1,2 1.3, H-
Sodium methoxide in MeOH (9.5 mL, 4.76 mmol) was added to
a solution of ester-protected epithiosugar 6 (3.5 g, 11.9 mmol)
in anhydrous MeOH (35 mL) and stirred under a nitrogen
atmosphere for 19 hours. The reaction mixture was neutralized
with Amberlite IRC-50 H+ ion exchange resin and diluted with
MeOH (50 mL). The solution was filtered and evaporated to give
an orange oil, which was purified by flash chromatography (silica
gel; DCM-MeOH, 95:5) to give 1,4-anhydro-5-deoxy-5-thio-a-D-
xylopyranose (7) (1.359 g, 77%), as a colourless amorphous glass
1), 4.28 (1H, dd, J5,6 4.6, J6,6¢ 10.2, H-6), 4.03 (1H, dd, J3,4 10.0, J4,5
9.8, H-4), 3.97 (1H, ddd, J5,6 4.6, J5,6¢ 10.1, J4,5 9.8, H-5), 3.85 (1H,
dd, J5,6¢ 10.1, J6,6¢ 10.2, H-6¢), 3.72 (1H, m, OCH2CH2), 3.44 (1H, m,
OCH2CH2), 2.17 (3H, s, CH3CO), 2.12 (2H, m, CH2CH2CH), 2.03
(3H, s, CH3CO), 1.72 (2H, m, OCH2CH2); dC (125 MHz; CDCl3);
170.1 and 170.0 (CH3CO), 138.0 (CH2=CH), 129.3, 128.4 and
126.4 (PhC), 115.4 (CH2=CH), 102.1 (PhCH), 98.8 (C-1), 76.4
(C-4), 70.4 (C-2), 68.9 (C-6), 68.5 (C-3), 67.8 (OCH2CH2), 64.0 (C-
5), 30.3 (CH2CH2CH), 28.6 (OCH2CH2) 21.1 and 21.0 (CH3CO);
m/z (ES); 443.2 ([M+Na]+), 438.2 ([M+NH4]+), 421.2 ([M+H]+);
Found: [M+Na]+ 443.1675, C22H28NaO8 requires 443.1676.
Sodium cyanoborohydride (1.01 g, 16.1 mmol) was added to
a solution of pent-4-enyl 2,3-di-O-acetyl-4,6-O-benzylidene-a-D-
mannopyranoside (0.5 g, 1.2 mmol) and methyl orange (a few
crystals) in tetrahydrofuran (10 mL). 1M HCl in ether was added
dropwise to the mixture until the colour changed from yellow to
pink and the mixture was stirred under a nitrogen atmosphere
for 24 hours. The mixture was diluted with EtOAc (100 mL) and
washed with sat. aq. NaHCO3 solution (100 mL) and sat. aq.
NaCl solution (100 mL). The organic extracts were evaporated to
give red oil, which was purified by flash chromatography (silica gel;
Hex-EtOAc, 3:2→1:1) giving the 4-hydroxy mannoside 14 (355 mg,
71%) as a colourless syrup; [a]D28 +30.8 (c 1.0 in CHCl3); nmax/cm-1
3435 (OH), 1746 (C O), 1641 (C C); dH (500 MHz; CDCl3);
7.36-7.28 (5H, m, PhCH2), 5.80 (1H, m, CH=CH2), 5.22 (1H,
m, H-3), 5.19 (1H, dd, J2,3 3.4, J 9.8 H-2), 5.05-4.96 (2H, m,
CH=CH2), 4.77 (1H, s, H-1), 4.62 (2H, m, PhCH2), 4.02 (1H, m,
H-4), 3.85-3.80 (2H, m, H-5 and H-6), 3.75 (1H, m, H-6¢), 3.70
(1H, m, OCH2CH2), 3.43 (1H, m, OCH2CH2), 2.61 (1H, d, J4,OH
3.8, OH-4), 2.12 (3H, s, CH3CO), 2.12 (2H, m, CH2CH2CH),
2.08 (3H, s, CH3CO), 1.70 (2H, m, OCH2CH2); dC (75 MHz;
CDCl3); 171.0 and 170.3 (MeCO), 138.0 (CH2=CH), 128.7-127.8
(PhCH2), 115.3 (CH2=CH), 97.9 (C-1), 73.9 (PhCH2), 72.1 (C-2),
71.0 (C-5), 70.4 (C-6), 70.1 (C-3), 67.7 (C-4), 67.6 (OCH2CH2),
30.4 (CH2CH2CH), 28.7 (OCH2CH2) 21.2 and 21.1 (CH3CO);
m/z (ES) 445.2 ([M+Na]+), 867.4 ([2M+Na]+), 440.2 ([M+NH4]+),
423.2 ([M+H]+); Found: [M+Na]+ 445.1820, C22H30NaO8 requires
445.1833.
26
(Found: C, 40.6; H, 5.6. C5H8SO3 requires C, 40.5; H 5.4%); [a]D
-12.3 (c 0.4 in MeOH); nmax/cm-1 3367 (OH); dH (500 MHz; D2O);
5.44 (1H, s, H-1), 5.12 (1H, dd, J3,4 5.6, J4,5¢ 4.7, H-4), 4.04 (1H,
d, J3,4 5.6, H-3), 3.94 (1H, s, H-2), 3.22 (1H, d, J5,5¢ 10.0, H-5),
2.81 (1H, dd, J5,5¢ 10.0, J4,5¢ 4.7, H-5¢); dC (75 MHz; D2O); 88.1 (C-
1), 83.1 (C-2), 81.3 (C-4), 79.9 (C-3), 29.1 (C-5); m/z (EI); 148.0
([M]+); Found: [M]+ 148.0191, C5H8SO3 requires 148.0194.
1,4-Anhydro-2,3-di-O-benzyl-5-deoxy-5-thio-a-D-xylopyranose (8)
Sodium hydride (60% dispersion in oil, 297 mg, 3.4 mmol)
was added to a solution of 1,4-anhydro-5-deoxy-5-thio-a-D-
xylopyranose (7) (0.5 g, 7.4 mmol) in DMF (10 mL) at 0 ◦C,
and stirred under a nitrogen atmosphere for 30 minutes. Benzyl
bromide (883 mL, 7.4 mmol) was added to the mixture and the
reaction was stirred as before, warming to RT for 3 hours. MeOH
(10 mL) was added to quench the reaction. The mixture was
diluted with EtOAc (100 mL), and washed with sat. aq. NaCl
solution (2 ¥ 100 mL). The organic extracts were dried (Na2SO4),
filtered and evaporated to give a yellow oil, which was purified
by flash chromatography (silica gel; Hex-EtOAc, 6:1) to give the
dibenzylated epithiosugar 8 (1.01 g, 91%) as colourless needles
(Found: C, 69.5; H, 6.2. C5H8SO3 requires C, 69.5; H 6.1%); m.p.
=
=
49-50 ◦C (from 1:1 H2O-MeOH); [a]D -8.6 (c 0.5 in CHCl3); dH
26
(500 MHz; CDCl3); 7.37-7.28 (10H, m, PhCH2), 5.47 (1H, s, H-1),
4.97 (1H, dd, J3,4 5.5, J4,5¢ 4.5, H-4), 4.57-4.47 (4H, m, PhCH2), 3.92
(1H, d, J3,4 5.5, H-3), 3.85 (1H, s, H-2), 3.30 (1H, d, J5,5¢ 9.3, H-5),
2.78 (1H, dd, J5,5¢ 9.3, J4,5¢ 4.5, H-5¢); dC (75 MHz; CDCl3); 128.7-
128.0 (PhCH2), 88.6 (C-2), 86.0 (C-3), 85.6 (C-1), 79.3 (C-4), 73.2
and 71.8 (PhCH2), 30.1 (C-5); m/z (ES); 351.1 ([M+Na]+), 679.2
([2M+Na]+); Found: [M+Na]+ 351.1035, C19H20NaSO3 requires
351.1025.
3-O-(2,3-Di-O-benzyl-5-deoxy-5-methylthio-D-xylofuranosyl)-
1,2:5,6-di-O-isopropylidene-a-D-glucofuranose (15)
Pent-4-enyl 2,3-di-O-acetyl-6-O-benzyl-a-D-mannopyranoside (14)
Methyl trifluoromethanesulfonate (25.8 mL, 0.23 mmol) was added
to a solution of glycosyl donor 8 (50 mg, 0.15 mmol), DTBMP
(94 mg, 0.46 mmol) and molecular sieves (3A) in DCM (0.25 mL)
Acetic anhydride (25 mL) was added to a solution of pent-4-
enyl 4,6-O-benzylidene-a-D-mannopyranoside32 (4.45 g, 13.2 mmol)
in pyridine (25 mL) and stirred under a nitrogen atmosphere
for 24 hours. The reaction mixture was diluted with EtOAc
(250 mL) and washed with aq. 1M HCl solution (2 ¥ 250 mL),
sat. aq. NaHCO3 solution (250 mL) and sat. aq. NaCl solution
(250 mL). The organic extracts were dried (Na2SO4), filtered
and evaporated to give an orange oil, which crystallized on
˚
and stirred under a nitrogen atmosphere for three hours. 1,2:5,6-
Di-O-isopropylidene-a-D-glucofuranose (79 mg, 0.30 mmol) and
˚
molecular sieves (3A) in DCM (0.25 mL) was added to the
reaction mixture and stirred for a further 20 hours. The reaction
mixture was filtered through celite, diluted with DCM (25 mL)
and washed with sat. aq. NaCl (25 mL). The organic extracts were
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 4842–4852 | 4849
©

dried (Na2SO4), filtered and evaporated to give a colourless oil.
Purification by flash chromatography (silica gel; Hex-EtOAc 3:1)
gave MTX-disaccharide 15 (59 mg, 64%) as a colourless syrup;
a:b 77:23 (Found: C, 64.0; H, 7.1. C32H42SO9 requires C, 63.8;
H 7.0%); dH (500 MHz; CDCl3); a-anomer 7.36-7.27 (10H, m,
PhCH2), 5.89 (1H, d, J1b,2b 3.5, H-1b), 5.28 (1H, d, J1a,2a 4.4, H-
1a), 4.77-4.46 (4H, m, PhCH2), 4.76 (1H, m, H-2b), 4.38 (1H,
m, H-4a), 4.32 (1H, m, H-5b), 4.21 (1H, m, H-3b), 4.19 (1H, dd,
J2a,3a 6.3, J3a,4a 5.1, H-3a), 4.11 (1H, dd, J4b,5b 8.6, J3b,4b 2.9, H-4b),
4.02 (3H, m, H-2a, H-6b, H-6b’), 2.81 (1H, dd, J5a,5a’ 13.9, J4a,5a
5.0, H-5a), 2.70 (1H, dd, J5a,5a’ 13.9, J4a,5a’ 7.7, H-5a’), 2.16 (3H, s,
CH3S), 1.50, 1.42, 1.32, 1.23 (12H, s, CH3C); b-anomer 7.36-7.27
(10H, m, PhCH2), 5.81 (1H, d, J1b,2b 3.7, H-1b) 5.13 (1H, s, H-
1a), 4.77-4.46 (4H, m, PhCH2), 4.45 (1H, m, H-4a), 4.39 (1H,
m, H-2b), 4.27 (1H, m, H-5b), 4.21 (1H, m, H-3b), 4.11 (1H, m,
H-4b), 4.02 (2H, m, H-2a, H-3a), 3.92 (2H, d, J5b,6b 6.1, H-6b,
H-6b’), 2.91 (1H, dd, J5a,5a’ 13.9, J4a,5a 6.5, H-5a), 2.84 (1H, dd,
J5a,5a’ 13.9, J4a,5a’ 7.3, H-5a’), 2.16 (3H, s, CH3S), 1.48, 1.37, 1.30,
1.22 (12H, s, CH3C); dC (75 MHz; CDCl3); a-anomer 128.7-127.8
(PhCH2), 112.0, 109.0 (2 ¥ (Me)2CO), 105.3 (C-1b), 104.5 (C-1a),
86.0 (C-2a), 84.0 (C-2b), 82.5 (C-3a), 82.2 (C-3b), 81.6 (C-4b), 77.3
(C-4a), 72.5 (C-5b), 72.5, 72.3 (2 ¥ PhCH2), 66.9 (C-6b), 34.3 (C-
5a), 27.1-25.5 (2 ¥ (Me)2CO); 16.8 (MeS); b-anomer 128.7-127.8
(PhCH2), 112.0, 109.2 (2 ¥ (Me)2CO), 105.5 (C-1b), 101.5 (C-
1a), 84.0 (C-2a, C-2b), 81.9 (C-3a), 81.8 (C-3b), 81.4 (C-4b), 78.0
(C-4a), 72.6 (C-5b), 72.4 (2 ¥ PhCH2), 67.6 (C-6b), 33.7 (C-5a),
27.0-25.4 (2 ¥ (Me)2CO); 16.3 (MeS); m/z (ES); 625.2 ([M+Na]+);
Found: [M+Na]+ 625.2414, C32H42NaSO9 requires 625.2442.
(CH3CO), 35.0 (C-5a), 16.8 (CH3S); m/z (ES); 829.3 ([M+Na]+);
Found: [M+Na]+ 829.3413, C48H54NaSO9 requires 829.3381.
Pent-4-enyl 2,3-di-O-acetyl-6-O-benzyl-4-O-(2,3-di-O-benzyl-5-
deoxy-5-methylthio-D-xylofuranosyl)-a-D-mannopyranoside (17)
Methyl trifluoromethanesulfonate (129.2 mL, 1.14 mmol) was
added to a solution of glycosyl donor 8 (250 mg, 0.76 mmol),
˚
DTBMP (469 mg, 2.28 mmol) and molecular sieves (3A) in DCM
(1 mL) and stirred under a nitrogen atmosphere for three hours.
Pentenyl mannoside 14 (804 mg, 1.90 mmol) and molecular sieves
˚
(3A) in DCM (1 mL) were added to the reaction mixture and
stirred for a further 24 hours. The reaction mixture was filtered
through celite, diluted with DCM (50 mL) and washed with sat. aq.
NaCl (50 mL). The organic extracts were dried (Na2SO4), filtered
and evaporated to give a colourless solid, which was purified
by flash chromatography (silica gel; DCM-EtOAc 39:1) to give
benzylated MTX-disaccharide 17 (274 mg, 47%) as a colourless
syrup; a:b 77:23; nmax/cm-1 1750 (C O), 1640 (C C), 1599,
1454 (C C aromatic); dH (500 MHz; CDCl3); a-anomer 7.37-7.24
=
=
=
(15H, m, PhCH2), 5.81 (1H, m, CH2=CH), 5.38 (1H, dd, J3b,4b 9.6,
J2b,3b 3.3, H-3b), 5.32 (1H, d, J1a,2a 4.2, H-1a), 5.27 (1H, dd, J2b,3b
3.3, J1b,2b 1.5, H-2b), 4.99 (2H, m, CH2=CH), 4.79 (1H, d, J1b,2b
1.5, H-1b), 4.71-4.39 (6H, m, PhCH2), 4.13-4.09 (3H, m, H-3a,
H-4a, H-4b), 3.98 (1H, m, H-2a), 3.80-3.69 (3H, m, H-5b, H-6b,
H-6b’), 3.68 (1H, m, OCH2CH2), 3.45 (1H, m, OCH2CH2), 2.72
(1H, dd, J5a,5a’ 13.6, J4a,5a 4.8, H-5a), 2.56 (1H, dd, J5a,5a’ 13.6, J4a,5a’
6.8, H-5a’), 2.12 (2H, m, CH2CH2CH), 2.10 (3H, s, CH3S), 2.07
(3H, s, CH3CO), 1.98 (3H, s, CH3CO), 1.71 (2H, m, OCH2CH2);
b-anomer 7.37-7.24 (15H, m, PhCH2), 5.81 (1H, m, CH2=CH),
5.37 (1H, m, H-3b), 5.20 (1H, dd, J2b,3b 3.4, J1b,2b 1.5, H-2b), 5.07
(1H, d, J1a,2a 1.3, H-1a), 4.99 (2H, m, CH2=CH), 4.76 (1H, d,
J1b,2b 1.5, H-1b), 4.71-4.39 (6H, m, PhCH2), 4.26 (1H, m, H-4a),
4.13-4.09 (1H, m, H-4b), 3.92 (1H, dd, J3a,4a 4.7, J2a,3a 1.9, H-
3a), 3.89 (1H, m, H-2a), 3.80-3.69 (3H, m, H-5b, H-6b, H-6b’),
3.68 (1H, m, OCH2CH2), 3.45 (1H, m, OCH2CH2), 2.88 (1H,
dd, J5a,5a’ 13.5, J4a,5a 7.4, H-5a), 2.69 (1H, dd, J5a,5a’ 13.5, J4a,5a’
6.3, H-5a’), 2.12 (2H, m, CH2CH2CH), 2.12 (3H, s, CH3S), 2.10
(3H, s, CH3CO), 2.01 (3H, s, CH3CO), 1.71 (2H, m, OCH2CH2);
dC (75 MHz; CDCl3); a-anomer 138.2 (CH2=CH), 128.7-127.6
(PhCH2), 115.1 (CH2=CH), 100.8 (C-1a), 97.5 (C-1b), 82.8
(C-2a), 81.9 (C-4a), 77.2 (C-3a), 73.4, 72.9, 72.6 (PhCH2), 72.7 (C-
3b), 70.8 (C-4b), 70.6 (C-2b), 70.0 (C-5b), 69.5 (OCH2CH2), 67.7
(C-6b), 34.9 (C-5a), 30.4 (CH2CH2CH), 28.6 (OCH2CH2), 21.2,
21.1 (CH3CO), 16.7 (CH3S); b-anomer 138.1 (CH2=CH), 128.7-
127.6 (PhCH2), 115.2 (CH2=CH), 108.3 (C-1a), 97.7 (C-1b), 86.5
(C-2a), 81.3 (C-4a), 80.9 (C-3a), 73.7, 73.0, 72.1 (PhCH2), 72.1 (C-
4b), 71.3 (C-2b), 70.5 (C-3b), 70.1 (C-5b), 69.0 (OCH2CH2), 67.6
(C-6b), 33.6 (C-5a), 30.1 (CH2CH2CH), 29.2 (OCH2CH2), 21.3,
21.1 (CH3CO), 16.6 (CH3S); m/z (ES); 787.3 ([M+Na]+); Found:
[M+Na]+ 787.3104, C42H52NaSO11 requires 787.3123.
Methyl 2,3,6-tri-O-benzyl-4-O-(2,3-di-O-benzyl-5-deoxy-5-
methylthio-D-xylofuranosyl)-a-D-mannopyranoside (16)16
Methyl trifluoromethanesulfonate (77.5 mL, 0.69 mmol) was added
to a solution of glycosyl donor 8 (150 mg, 0.46 mmol), DTBMP
˚
(281.4 mg, 1.37 mmol) and molecular sieves (3A) in DCM (0.5 mL)
and stirred under a nitrogen atmosphere for three hours. Methyl
2,3,6-tri-O-benzyl-a-D-mannopyranoside33 (424.3 mg, 0.92 mmol)
˚
and molecular sieves (3A) in DCM (0.5 mL) was added to the
reaction mixture and stirred for a further 25 hours. The reaction
mixture was filtered through celite, diluted with DCM (50 mL)
and washed with sat. aq. NaCl (50 mL). The organic extracts
were dried (Na2SO4), filtered and evaporated to give a colourless
syrup, which was purified by flash chromatography (silica gel;
DCM-EtOAc 39:1→19:1) to give benzylated MTX-disaccharide
16 (178 mg, 48%) as colourless syrup; a:b 83:17; dH (500 MHz;
CDCl3); a-anomer 7.38-7.04 (25H, m, PhCH2), 5.55 (1H, d, J1a,2a
4.4, H-1a), 4.84 (1H, s, H-1b), 4.75-4.37 (10H, m, PhCH2), 4.05
(2H, m, H-4a, H-4b), 3.96-3.69 (6H, m, H-2a, H-2b, H-3a, H-
3b, H-5b, H-6b), 3.39 (3H, s, CH3O), 2.66 (1H, dd, J5a,5a’ 13.8,
J4a,5a 4.7, H-5a), 2.50 (1H, dd, J5a,5a’ 13.8, J4a,5a’ 7.2, H-5a’), 2.05
(3H, s, CH3S); b-anomer 7.38-7.04 (25H, m, PhCH2), 5.39 (1H, s,
H-1a), 4.84 (1H, s, H-1b), 4.75-4.37 (10H, m, PhCH2), 4.24 (1H,
m, H-4a), 4.05 (1H, m, H-4b), 3.96-3.69 (6H, m, H-2a, H-2b, H-
3a, H-3b, H-5b, H-6b), 3.32 (3H, s, CH3O), 2.87 (1H, dd, J5a,5a’
13.5, J4a,5a 7.6, H-5a), 2.70 (1H, dd, J5a,5a’ 13.5, J4a,5a’ 6.3, H-5a’),
2.07 (3H, s, CH3S); dC (75 MHz; CDCl3); a-anomer only 128.6-
126.9 (PhCH2), 100.8 (C-1a), 98.6 (C-1b), 82.7-76.9 (C-2a, C-2b,
C-3a, C-3b, C-4a, C-4b), 73.3-70.2 (PhCH2, C-5b, C-6b), 54.9
Pent-4-enyl 4-O-(5-deoxy-5-methylthio-a-D-xylofuranosyl)-a-D-
mannopyranoside (18a) and pent-4-enyl 4-O-(5-deoxy-5-
methylthio-b-D-xylofuranosyl)-a-D-mannopyranoside (18b)
Sodium methoxide in MeOH (0.193 mL, 0.10 mmol) was added
to a solution of acetylated MTX-disaccharide 17 (185 mg,
0.24 mmol) in anhydrous MeOH (2 mL) and stirred under a
4850 | Org. Biomol. Chem., 2009, 7, 4842–4852
This journal is
The Royal Society of Chemistry 2009
©

nitrogen atmosphere for 19 hours. The reaction mixture was
neutralized with Amberlite IRC-50 H+ ion exchange resin and
diluted with MeOH (50 mL). The solution was filtered and
evaporated to give a colourless oil, which was purified by
flash chromatography (silica gel; Hex-EtOAc, 1:1) to give pent-
4-enyl 6-O-benzyl-4-O-(2,3-di-O-benzyl-5-deoxy-5-methylthio-D-
xylofuranosyl)-a-D-mannopyranoside (114 mg, 69%), as a colour-
dd, J5a,5a’ 13.8, J4a,5a’ 7.4, H-5a’), 2.17 (3H, s, CH3S), 2.15 (2H, m,
CH2CH2CH), 1.69 (2H, m, OCH2CH2); dC (75 MHz; CD3OD);
139.4 (CH2=CH), 115.3 (CH2=CH), 104.9 (C-1a), 101.6 (C-1b),
80.5 (C-4a), 78.9 (C-2a), 77.2 (C-3a), 76.1, 73.4, 72.5, 72.4 (C-2b,
C-3b, C-4b, C-5b), 68.0 (OCH2CH2), 62.8 (C-6b), 34.5 (C-5a), 31.5
(CH2CH2CH), 29.9 (OCH2CH2), 16.3 (CH3S); m/z (ES); Found:
[M+Na]+ 433.1514, C17H30NaSO9 requires 433.1503.
b-Anomer 18b [a]D +1.6 (c 1.5 in CHCl3); nmax/cm-1 3350
26
less syrup; a:b 77:23 (Found: C, 66.8; H, 7.0. C38H48SO9 requires C,
-1
=
=
67.0; H 7.1%); nmax/cm 3435 (OH), 1639 (C C), 1593, 1453 (C C
aromatic); dH (500 MHz; CDCl3); a-anomer 7.37-7.22 (15H, m,
PhCH2), 5.80 (1H, m, CH2=CH), 5.22 (1H, d, J1,2 4.5, H-1a),
4.99 (2H, m, CH2=CH), 4.83 (1H, s, H-1b), 4.72-4.56 (6H, m,
PhCH2), 4.35 (1H, m, H-4a), 4.26 (1H, dd, J3a,4a 6.5, J2a,3a 6.1,
H-3a), 4.07 (1H, dd, J2a,3a 6.1, J1a,2a 4.5, H-2a), 3.91 (2H, m, H-
2b, H-4b), 3.79-3.71 (4H, m, H-3b, H-5b, H-6b, H-6b’), 3.68 (1H,
m, OCH2CH2), 3.42 (1H, m, OCH2CH2), 2.73 (1H, dd, J5a,5a’ 13.9,
(OH); dH (500 MHz; CD3OD); 5.84 (1H, m, CH2=CH), 5.01
(1H, s, H-1a), 5.00 (2H, m, CH2=CH), 4.74 (1H, d, J1b,2b 1.4,
H-1b), 4.36 (1H, ddd, J4a,5a’ 7.5, J4a,5a 6.6, J3a,4a 3.9, H-4a),
4.07 (1H, d, J2a,3a 1.9, H-2a), 4.02 (1H, dd, J3a,4a 3.9, J2a,3a 1.9,
H-3a), 3.86 (1H, m, H-2b), 3.84-3.75 (4H, m, H-3b, H-4b, H-6b,
H-6b’), 3.73 (1H, m, OCH2CH2), 3.60 (1H, m, H-5b), 3.44 (1H,
m, OCH2CH2), 2.91 (1H, dd, J5a,5a’ 13.6, J4a,5a 6.6, H-5a), 2.80 (1H,
dd, J5a,5a’ 13.6, J4a,5a’ 7.5, H-5a’), 2.18 (3H, s, CH3S), 2.15 (2H, m,
CH2CH2CH), 1.69 (2H, m, OCH2CH2); dC (75 MHz; CD3OD);
139.4 (CH2=CH), 115.3 (CH2=CH), 110.4 (C-1a), 101.4 (C-1b),
83.6 (C-4a), 82.1 (C-2a), 76.7 (C-3a), 76.3, 73.2, 71.7, 71.4 (C-2b,
C-3b, C-4b, C-5b), 68.0 (OCH2CH2), 62.3 (C-6b), 34.1 (C-5a), 31.5
(CH2CH2CH), 29.9 (OCH2CH2), 16.0 (CH3S); m/z (ES); Found:
[M+Na]+ 433.1505, C17H30NaSO9 requires 433.1503.
J
4a,5a 5.0, H-5a), 2.58 (1H, dd, J5a,5a’ 13.9, J4a,5a’ 7.4, H-5a’), 2.12 (2H,
m, CH2CH2CH), 2.10 (3H, s, CH3S), 1.67 (2H, m, OCH2CH2);
b-anomer 7.37-7.22 (15H, m, PhCH2), 5.80 (1H, m, CH2=CH),
5.07 (1H, d, J1,2 1.1, H-1a), 4.99 (2H, m, CH2=CH), 4.87
(1H, s, H-1b), 4.72-4.40 (6H, m, PhCH2), 4.26 (1H, m, H-
4a), 3.98 (1H, m, H-3a), 3.96 (1H, m, H-2a), 3.85-3.72 (5H,
m, H-3b, H-4b, H-5b, H-6b, H-6b’), 3.68 (1H, m, OCH2CH2),
3.42 (1H, m, OCH2CH2), 2.88 (1H, dd, J5a,5a’ 13.5, J4a,5a 6.4,
H-5a), 2.80 (1H, dd, J5a,5a’ 13.5, J4a,5a’ 7.8, H-5a’), 2.13 (3H, s,
CH3S), 2.12 (2H, m, CH2CH2CH), 1.67 (2H, m, OCH2CH2);
dC (75 MHz; CDCl3); a-anomer 138.2 (CH2=CH), 128.8-127.6
(PhCH2), 115.1 (CH2=CH), 102.0 (C-1a), 99.6 (C-1b), 83.8
(C-2a), 81.9 (C-3a), 77.8 (C-4a), 77.2 (C-3b), 73.6, 73.3, 73.0
(PhCH2), 71.5 (C-2b), 71.0 (C-4b), 70.7 (C-5b), 69.2 (OCH2CH2),
67.3 (C-6b), 35.0 (C-5a), 30.5 (CH2CH2CH), 28.9 (OCH2CH2),
16.9 (CH3S); b-anomer 138.2 (CH2=CH), 128.8-127.6 (PhCH2),
115.1 (CH2=CH), 108.0 (C-1a), 97.3 (C-1b), 85.7 (C-2a), 81.2
(C-4a), 80.9 (C-3a), 77.4 (C-3b), 73.7, 73.3, 72.3 (PhCH2), 70.5
(C-2b), 70.4 (C-4b), 70.2 (C-5b), 69.3 (OCH2CH2), 67.2 (C-
6b), 33.7 (C-5a), 30.5 (CH2CH2CH), 28.8 (OCH2CH2), 16.3
(CH3S); m/z (ES); 703.3 ([M+Na]+); Found: [M+Na]+ 703.2926,
C38H48NaSO9 requires 703.2911.
Density functional theory calculations
Conformations were generated in MacroModel using the OPLS
forcefield.34 For the MTX system, six low energy conformations
of the reactant were obtained (sampling the six available con-
formations of the two OMe sidechains) each of which was used
to generate the corresponding transition states for SN1 and SN2
reactions. For other species, a selection of 20-25 of the lowest
energy conformations according to the forcefield were chosen.
The geometries generated were subject to density functional
theory calculations performed in Gaussian0335 and using Becke’s
3-parameter hybrid exchange functional36 and the Lee-Yang-Parr
exchange functional (B3LYP/6-31G*).37 All stationary points
were verified by frequency calculations. All energies are free
energies quoted as differences between the species being referred
to in its lowest energy conformation and the corresponding
sulfonium ion in its lowest energy conformation. Solvation free
energies were computed with the IEFPCM method in combination
with B3LYP/6-31+G* and used UAKS radii.38
Ammonia (25 mL) was condensed into a flask containing pent-
4-enyl 6-O-benzyl-4-O-(2,3-di-O-benzyl-5-deoxy-5-methylthio-D-
xylofuranosyl)-a-D-mannopyranoside (90 mg, 0.13 mmol) in
tetrahydrofuran (5 mL) at -78 ◦C. Shavings of sodium were added
until a dark blue colour persisted and the solution was stirred
under a nitrogen atmosphere for one and a half hours. MeOH
(2 mL) was added and the mixture was left open to air at RT
overnight. The solution was diluted with MeOH (20 mL) and
neutralised with acetic acid, before evaporation to give a yellow
residue, which was purified by flash chromatography (silica gel;
DCM-MeOH, 9:1) to yield a-glycoside 18a (25 mg) as colourless
syrup and b-glycoside 18b (9 mg) as colourless syrup (overall yield
63%).
Acknowledgements
We thank the Royal Society, AstraZeneca and the University of
Leeds for financial support. WBT is the recipient of a Royal Society
University Research Fellowship.
Notes and references
1 S. Winstein and R. E. Buckles, J. Am. Chem. Soc., 1942, 64, 2780.
2 L. Goodman, Adv. Carbohydr. Chem. Biochem., 1967, 22, 109.
3 R. U. Lemieux, Adv. Carbohydr. Chem., 1954, 9, 1; D. B. Werz and
P. H. Seeberger, Angew. Chem., Int. Ed., 2005, 44, 6315; G. Ragupathi,
F. Koide, P. O. Livingston, Y. S. Cho, A. Endo, Q. Wan, M. K. Spassova,
S. J. Keding, J. Allen, O. Ouerfelli, R. M. Wilson and S. J. Danishefsky,
J. Am. Chem. Soc., 2006, 128, 2715; X. Zhu and R. R. Schmidt, Angew.
Chem., Int. Ed., 2009, 48, 1900.
a-Anomer 18a [a]D +54.2 (c 3.1 in CHCl3); nmax/cm-1 3350
26
(OH); dH (500 MHz; CD3OD); 5.85 (1H, m, CH2=CH), 5.38 (1H,
d, J1,2 4.2, H-1a), 5.00 (2H, m, CH2=CH), 4.74 (1H, d, J1b,2b 1.2,
H-1b), 4.34 (1H, ddd, J4a,5a’ 7.4, J4a,5a 5.7, J3a,4a 4.6, H-4a), 4.14
(1H, dd, J3a,4a 4.6, J2a,3a 4.1, H-3a), 4.08 (1H, dd, J1a,2a 4.2, J2a,3a 4.1,
H-2a), 3.89-3.85 (2H, m, H-3b, H-6b’), 3.82-3.79 (3H, m, H-2b, H-
4b, H-6b), 3.75 (1H, m, OCH2CH2), 3.62 (1H, m, H-5b), 3.44 (1H,
m, OCH2CH2), 2.75 (1H, dd, J5a,5a’ 13.8, J4a,5a 5.7, H-5a), 2.63 (1H,
4 H. Jiao and O. Hindsgaul, Angew. Chem., Int. Ed., 1999, 38, 346.
5 R. U. Lemieux, T. Takeda and B. Y. Chung, ACS Symp. Ser., 1976,
39, 90; J. S. Debenham, R. Madsen, C. Roberts and B. Fraser-Reid,
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 4842–4852 | 4851
©

J. Am. Chem. Soc., 1995, 117, 3302; H. Shimizu, Y. Ito, Y. Matsuzaki,
H. Iijima and T. Ogawa, Biosci., Biotechnol., Biochem., 1996, 60, 73.
6 W. R. Roush and C. E. Bennett, J. Am. Chem. Soc., 1999, 121, 3541.
7 K. C. Nicolaou, R. M. Rodr´ıgue, H. J. Mitchell and F. L. v. Delft,
Angew. Chem., Int. Ed., 1998, 37, 1874.
8 J.-H. Kim, H. Yang and G.-J. Boons, Angew. Chem., Int. Ed., 2005,
44, 947; D. J. Cox and A. J. Fairbanks, Tetrahedron: Asymmetry,
2009, 20, 773; M. A. Fascione, S. J. Adshead, S. A. Stalford, C. A.
Kilner, A. G. Leach and W. B. Turnbull, Chem. Commun., 2009,
DOI: 10.1039/b913308a.
24 I. Braccini, C. Derouet, J. Esnault, C. H. e. de Penhoat, J. M. Mallet,
V. Michon and P. Sinay¨, Carbohydr. Res., 1993, 246, 23.
25 P. Sinay, Pure Appl. Chem., 1991, 63, 519.
26 T. Nukada, A. Berces, L. Wang, M. Z. Zgierski and D. M. Whitfield,
Carbohydr. Res., 2005, 340, 841.
27 C. H. Larsen, B. H. Ridgway, J. T. Shaw, D. M. Smith and K. A.
Woerpel, J. Am. Chem. Soc., 2005, 127, 10879.
28 J. R. Krumper, W. A. Salamant and K. A. Woerpel, Org. Lett., 2008,
10, 4907.
29 Lowary and co-workers have demonstrated that 2,3-anhydro-furanosyl
sulfoxide donors form 1,2-trans-triflate intermediates: C. S. Callam,
R. R. Gadikota, D. M. Krein and T. L. Lowary, J. Am. Chem. Soc.,
2003, 125, 13112.
30 D. D. Perrin and W. L. F. Armarego, Purification of Laboratory
Chemicals, 3rd Ed, Pergamon, Oxford, 1988.
31 H. E. Gottlieb, V. Kotlyar and A. Nudelman, J. Org. Chem., 1997, 62,
7512.
32 I. Cumpstey, T. D. Butters, R. J. Tennant-Eyles, A. J. Fairbanks, R. R.
France and M. R. Wormald, Carbohydr. Res., 2003, 338, 1937.
33 R. Madiyalakan, M. S. Chowdhary, S. S. Rana and K. L. Matta,
Carbohydr. Res., 1986, 152, 183.
9 J. H. Kim, H. Yang, J. Park and G. J. Boons, J. Am. Chem. Soc., 2005,
127, 12090.
10 C. A. A. van Boeckel, T. Beetz and S. F. van Aelst, Tetrahedron, 1984,
40, 4097; N. Ustyuzhanina, B. Komarova, N. Zlotina, V. Krylov, A.
Gerbst, Y. Tsvetkov and N. Nifantiev, Synlett, 2006, 921; C. De Meo,
M. N. Kamat and A. V. Demchenko, Eur. J. Org. Chem., 2005, 706.
11 D. Crich, W. L. Cai and Z. M. Dai, J. Org. Chem., 2000, 65, 1291.
12 D. Crich, T. S. Hu and F. Cai, J. Org. Chem., 2008, 73, 8942.
13 M. G. Beaver, S. B. Billings and K. A. Woerpel, J. Am. Chem. Soc.,
2008, 130, 2082.
14 A. Treumann, X. Feng, L. McDonnell, P. J. Derrick, A. E. Ashcroft,
D. Chatterjee and S. W. Homans, J. Mol. Biol., 2002, 316, 89.
15 W. B. Turnbull, K. H. Shimizu, D. Chatterjee, S. W. Homans and
A. Treumann, Angew. Chem., Int. Ed., 2004, 43, 3918.
34 W. L. Jorgensen and J. Tirado-Rives, J. Am. Chem. Soc., 1988, 110,
1657.
35 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, J. J. A. Montgomery, T. Vreven, K. N. Kudin,
J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone,
B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson,
H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene,
X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo,
J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,
R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma,
G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski,
S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick,
A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui,
A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu,
A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox,
T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M.
Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong,
C. Gonzalez and J. A. Pople, eds., Gaussian03: Revision D.02, Gaussian,
Inc., Wallingford CT, 2004.
36 A. D. Becke, J. Chem. Phys., 1993, 98, 1372.
37 C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter Mater.
Phys., 1988, 37, 785.
38 E. Cances, B. Mennucci and J. Tomasi, J. Chem. Phys., 1997, 107, 3032.
39 Note added in proof: Lowary and co-workers have recently reported an
analogous study of 2-deoxy-2-thioarylxylofuranosyl cations; they also
conclude that their glycosylation reactions proceed via oxacarbenium
ion intermediates rather than through episulfonium ions. D. Hou, H. A.
Taha and T. L. Lowary, J. Am. Chem. Soc., 2009, 131, 12937.
16 M. Joe, D. Sun, H. Taha, G. C. Completo, J. E. Croudace, D. A.
Lammas, G. S. Besra and T. L. Lowary, J. Am. Chem. Soc., 2006, 128,
5059.
17 S. A. Stalford, M. A. Fascione, S. J. Sasindran, D. Chatterjee,
S. Dhandayuthapani and W. B. Turnbull, Chem. Commun., 2009, 110.
18 I. Lundt and B. Skelbaek-Pedersen, Acta Chem. Scand., Ser. B, 1981,
B35, 637; A. Fleetwood and N. A. Hughes, Carbohydr. Res., 1999,
317, 204; R. V. Stick, D. Matthew, G. Tilbrook and S. J. Williams,
Aust. J. Chem., 1999, 52, 685; P. R. Sridhar, V. Saravanan and
S. Chandrasekaran, Pure Appl. Chem., 2005, 77, 145; W. B. Turnbull,
M. A. Fascione and S. A. Stalford, Sci. Synth., 2007, 29, 923.
19 B. M. Trost, Science, 1991, 254, 1471.
20 D. H. Hollenberg, K. A. Watanabe and J. J. Fox, Carbohydr. Res., 1975,
42, 241.
21 The crystal structure for compound 6 has H-4/H-5 dihedral angles of
42◦ and 79◦ for the pro-R and pro-S hydrogens, respectively (41◦ and
85◦ for the DFT structure of sulfonium ion 7). The NMR signals for
the pro-R and pro-S hydrogens were assigned on the basis of the J4,5
coupling constants.
22 The NMR chemical shifts for H-1, H-2 and H-3 were similar to
those reported previously for a glucopyranosyl nitrilium ion (see ref.
23). Attempts to trap the nitrilium ion using Sinay¨’s method were
unfortunately inconclusive: J.-R. Pougny and P. Sinay¨, Tetrahedron
Lett., 1976, 17, 4073.
23 A. D’Aprano, A. Capalbi, M. Iammarino, V. Mauro, A. Princi and
B. Sesta, J. Solution Chem., 1995, 24, 227.
4852 | Org. Biomol. Chem., 2009, 7, 4842–4852
This journal is
The Royal Society of Chemistry 2009
©