Influence of O6 in mannosylations using benzylidene protected donors: Stereoelectronic or conformational effects?

Article abstract of DOI:10.1021/jo302455d

The stereoselective synthesis of β-mannosides and the underlying reaction mechanism have been thoroughly studied, and especially the benzylidene-protected mannosides have gained a lot of attention since the corresponding mannosyl triflates often give exce

Full text of DOI:10.1021/jo302455d

Featured Article
pubs.acs.org/joc
Influence of O6 in Mannosylations Using Benzylidene Protected
Donors: Stereoelectronic or Conformational Effects?
Tobias Gylling Frihed,† Marthe T. C. Walvoort,‡ Jeroen D. C. Codee,‡ Gijs A. van der Marel,‡
́
Mikael Bols,† and Christian Marcus Pedersen*,†
†Department of Chemistry, University of Copenhagen, Universitetsparken 5, 2100 Copenhagen, Denmark
‡Leiden Institute of Chemistry, Leiden University, Einsteinweg 55, 2333 CC Leiden, The Netherlands
S
* Supporting Information
ABSTRACT: The stereoselective synthesis of β-mannosides
and the underlying reaction mechanism have been thoroughly
studied, and especially the benzylidene-protected mannosides
have gained a lot of attention since the corresponding
mannosyl triflates often give excellent selectivity. The
hypothesis for the enhanced stereoselectivity has been that
the benzylidene locks the molecule in a less reactive
conformation with the O6 trans to the ring oxygen (O5),
which would stabilize the formed α-triflate and subsequent
give β-selectivity. In this work, the hypothesis is challenged by using the carbon analogue (C7) of the benzylidene-protected
mannosyl donor, which is investigated in terms of diastereoselectivity and reactivity and by low-temperature NMR. In terms of
diastereoselectivity, the C-7-analogue behaves similarly to the benzylidene-protected donor, but its low-temperature NMR reveals
the formation of several reactive intermediate. One of the intermediates was found to be the β-oxosulfonium ion. The reactivity
of the donor was found to be in between that of the “torsional” disarmed and an armed donor.
glycoside bond formation, anomeric O-alkylation8 has been
found to be useful, but more complex glycoside bonds than 1,6-
mannosidic linkages are seldom possible with this method. The
INTRODUCTION
■
Mannosides play a central role in mammalian biology, where
both the α- and the β-anomers are present, e.g., in the core
pentasaccharide of N-glycoconjugates. It is therefore highly
influence of the donors reactivity on the β-selectivity was
realized by Schuerch9 and later explored by others;10 its general
attractive to synthesize glycoconjugates containing mannosides,
use has recently been developed by Kim and co-workers, who
and amazing progress in synthetic glycoconjugates has been
systematically studied the influence of electron-withdrawing
achieved during the past decade in order to decipher the
glycocode. The α-mannosides are readily accessible via
neighboring group participation and additionally favored due
substituents on the 3, 4, and 6 position, respectively. This
demonstrated that not only is the 2-position important, but all
the substituents influence the reactivity and hence the outcome
to the anomeric effect. The β-anomer, however, is challenging
of the reaction.11
since both the anomeric effect, the Δ2-effect1 as well as steric
effects disfavor its formation. This synthetic problem has
attracted a lot of interest since the early days of oligosaccharide
A tremendous improvement of direct β-mannosylation
appeared in 1996 from the Crich group,12 who discovered
that when a 4,6-benzylidene-protected donor was preactivated
synthesis. The approaches can roughly be divided into direct
using Kahne’s conditions13 a good to excellent β-selectivity was
and indirect methods, where the direct method provides the β-
obtained. Over the years, the reaction has proved to be a
mannoside from a donor and an acceptor without pre- or
general approach for this difficult linkage, and it has gained a lot
postmodifications. The indirect methods demand additional
of attention not only for its use as an applied method but also
steps but are often more efficient in terms of diastereose-
lectivity.2 Some examples of indirect approaches are intra-
for the underlying mechanism responsible for the selectivity.
molecular aglycon delivery (IAD),3 2-O-inversion of the
Early studies of the reaction using low-temperature NMR
corresponding gluco derivative,4 stereoselective reduction of
revealed that the donor was transformed into the α-triflate
the 2-oxo derivative (uloside derivative),5 or reductive opening
under the reaction conditions, and a SN2-type mechanism was
proposed to be responsible for the β-selectivity.35 The decisive
role of the triflate ions in terms of selectivity has been
demonstrated by the substantial amount of work in the field.14
The 4,6-tethering of the donor, most often by a benzylidene, is
of the tricyclic orthoester.6 Direct methods are, however,
preferred, especially in the synthesis of complex compounds,
where the number of steps is required to be kept at a minimum.
One of the first successful β-mannosylations was based on the
classic Konigs−Knorr method applying solid silver salts as
̈
promoters.7 The yields are, however, often moderate, and their
use in complex oligosaccharide synthesis is limited. For simple
Received: November 28, 2012
Published: January 21, 2013
© 2013 American Chemical Society
2191
dx.doi.org/10.1021/jo302455d | J. Org. Chem. 2013, 78, 2191−2205

The Journal of Organic Chemistry
Featured Article
another necessity in the method. The influence of the tethering
has been explained by either conformational restriction favoring
the β-attack10,15 or stereoelectronic effects by locking the O6
antiperiplanar (trans−gauche) to the ring oxygen (O5), making
it more electron poor (disarming the donor). This results in a
less stable oxocarbenium ion and hence a more stable covalent
triflate intermediate, which results in a SN2 type reaction.16
Expanding the ring with one extra atom reduces the β-
selectivity dramatically, which supports both hypotheses, i.e.,
deactivation by either stereoelectronics or strain.17 Exchange of
the O6 with the less electronegative sulfur atom has also been
shown to reduce the β-selectivity either by stereoelectronic or
torsional effects. The selectivity could, however, be restored by
introducing an electron-withdrawing cyano substituent on the
benzylidene.18 The enhanced electron-withdrawing capacity of
an antiperiplanar oxygen (trans−gauche) has been used to
explain the increased stability of the α-triflates when a
benzylidene protective group is present. The effect was
originally observed in a gluco-model system, where the rate
of hydrolysis was hampered by locking the oxygen in different
positions (Figure 1).19 The influence on rate of hydrolysis and
thereby the reactivity is clear, but is it responsible for the β-
selectivity in mannosides?
synthesized from mannose in six steps using standard
carbohydrate chemistry.
Protective group manipulations gave the diol 3, which was
tritylated followed by acylation in one-pot to give the
compounds 4a−c. Upon acid-mediated removal of the trityl
group, the 4-O-acetyl group readily migrated, even in the
presence of trifluoroacetic anhydride, to trap the product as the
6-O-trifluoroacetyl (entry 3 and 4 in Table 1, Supporting
Information).23 The more bulky esters, the pivaloyl 4b and the
benzoyl 4c, solved that problem at this stage of the synthesis
(entries 5 and 6, Table 1, Supporting Information).
With the access to the alcohols 5a−c, oxidation to the
aldehydes 7a,b by Dess−Martin periodinane (DMP) could
readily be carried out. Wittig olefination of the crude aldehydes
gave 9a and 9b in overall good yields (Scheme 1). The reaction
Scheme 1. Synthesis of the Precursor for the C7-Analogue
Starting from the Common Mannosyl Donor 3 via
Temporary Tritylation of the 6-OH, 4-OH Acylation, 6-O
Deprotection, and Oxidation to the Aldehydes, Followed by
Wittig Olefination To Give the Alkene 9
Figure 1. Relative rate of hydrolysis of dinitrophenyl (DNP)
glycosides.
The reaction mechanism of the preactivated β-mannosides
has been extensively investigated during the last two decades,
both experimentally and in silico. Kinetic isotope effect studies
have supported the SN2-like mechanism when having mannosyl
triflates.20 The importance of the nucleophile, i.e., the acceptor
used in triflate substitution on tetrahydropyrans, has also been
investigated, and a SN2-like mechanism was suggested for
strong nucleophiles.21 Conformational preferences of oxocar-
benium ions and the influence on stereochemical outcome have
been thoroughly studied on model compounds, and a
preference, in the manno system, for a pseudoequatorial 2-O
was found. All other hydroxyl groups preferred pseudoaxial
arrangement due to electrostatic stabilization (the 3H4
conformation).22 Is the O6 (when being trans−gauche)
responsible for the increased β-selectivity obtained with
benzylidene-protected mannosyl donors? In this paper, the
reaction is studied using a carbon (C7) analogue 1 of the 4,6-
benzylidene mannosyl donor 2 in order to answer this question.
with the benzoylated substrate was, however, unpredictable
when scaling up the reaction. On a gram scale, the major
product turned out to be the benzoyl-eliminated compound
10.17,24
Selective reduction of the olefins was mediated by freshly
prepared Stryker’s reagent [CuH(PPh3)]6,25 which gave good
to high yield of the corresponding saturated compounds,
Scheme 2. Saponification of the pivaloyl or the benzoyl 4-O-
Scheme 2. Selective Alkene Reduction Using Stryker’s
Reagent Followed by Deacylation
́
protective group using Zemplen conditions in MeOH/THF
gave the mono-ol 12 ready for ring formation with the ketone.
The deprotection of the bulky pivaloyl 11a was very slow at
room temperature. Even at elevated temperatures, 6 days were
required to obtain a reasonable conversion. Despite the
synthetic challenges, enough of compound 12 could be
obtained to investigate the key reaction: ring closure to the
C7 donor 1.
RESULTS
■
Synthesis. In contrast to previous related studies on the
influence of the O6 on the anomeric reactivity in a gluco-model
system,19 it was decided to include the phenyl substituent as
part of the bicyclic system resembling the benzylidene; this
demanded a new approach for the synthesis. The diol 3 was
2192
dx.doi.org/10.1021/jo302455d | J. Org. Chem. 2013, 78, 2191−2205

The Journal of Organic Chemistry
Featured Article
Reductive etherification inspired by Hung and co-workers26
was attempted, but no conversion of the 4-O-TMS 13 was
observed (Scheme 3). Treating 12 with acid in MeOH−
finally 23 as a more challenging secondary 4-OH acceptor,
Scheme 5.
Scheme 5. Glycosylation with the C7-Analogue 1 vs the
Benzylidene 2
Scheme 3. Ring Closure Followed by Reduction To Give the
C7-Analogue 1
The mannosylations of donor 1 and 2 were first compared
using NIS/TfOH as the promoter system, which mainly gave
the α-product for both donors; 1:5.7 for the 2 and 1:8.0 for 1,
entries 1 and 2, Table 1. The yield was somewhat better for the
benzylidene donor 2. Changing to the preactivation conditions
Ph2SO/TTBP and Tf2O with preactivation in 5 min resulted in
complete α-selectivity (entry 3) and not the expected β-
selectivity (9:1 β:α).28 Prolonging the preactivation to 30 min
significantly improved the selectivity to 6.7:1, clearly under-
lining the importance of preactivation time, entry 4. When the
exact same conditions were applied on the C7-analogue a
practically identical selectivity was obtained (5.4:1) though in a
slightly lower yield, entry 5. Changing the molecular sieves to 4
Å instead of 3 Å did not significantly alter the selectivity in this
example. Performing the mannosylations using the BSP
promoter system developed by Crich29 resulted in complete
α-selectivity after 5 min preactivation,28 but a 6.1:1 β-selectivity
after 30 min, entry 7 vs 8. The same result was obtained with
the C7-analogue, both in terms of excellent yield and selectivity,
entry 9. Mannosylation of the more demanding carbohydrate-
based acceptors 21−23 were performed as described for the
benzylidene donor using the BSP promoter system, but with a
preactivation for 30 min since it was found to be crucial for
selectivity. Glycosylation of diacetone glucose 21 gave the β-
product 26β as the major (4.4:1), a bit lower than the reported
>9:1, entry 10. The selectivity with the 6OH acceptor 22 was
also slightly lower (6.1:1 vs >9:1), but the yield was slightly
higher, entry 11. The reaction with the 4-OH acceptor 23
turned out to give complex mixtures, where one of the major
side products was the triflated acceptor. The ratio of anomers
was estimated by 1H NMR to be 2:1 (β/α). All glycosides were
carefully analyzed by NMR and their anomeric configuration
determined from the JCH coupling constant, which followed the
trend seen in mannosides, i.e., JCHα ∼ 165−175 Hz and JCHβ ∼
155 Hz (or ca.10 ppm lower than the α).30
toluene, however, gave a mixture of the bicyclic compounds 14
and 15, which upon reduction gave one major product 1 in
97% yield, Scheme 3. The new donor 1 was carefully analyzed
by NMR spectroscopy, and the equatorial position of the
phenyl could be confirmed and coupling constants within the
sugar ring corresponded to the ones found in the benzylidene
donor 2.
With the chemistry developed for the final crucial ring
closure, the steps leading to the 4-OH 12 had to be optimized
in order to obtain large amounts (grams) of the donor to use in
mechanistic studies and glycosylations. The main problem in
the synthesis was clearly associated with the (in)stability of the
acyl protective groups and their ability as leaving groups.
To solve these problems, p-methoxybenzyl (PMB) was
introduced, as the 4-O protective group, Scheme 4. The diol 3
Scheme 4. Optimized Synthesis of the C7-Analogue 1
was first selectively 6-O-silylated using TBDMSCl, followed by
alkylation of the 4-OH under basic conditions using PMB-Cl.
TBAF-mediated desilylation, 6-OH oxidation by DMP to the
aldehydes 7c and Wittig olefination gave the alkene, which was
selectively reduced by Stryker’s reagent to give 11c. Removal of
the PMB group using acid surprisingly resulted in a
rearrangement to the bicyclic compound 19; this side reaction
could however be avoided using DDQ mediated oxidative
deprotection giving 12, which could be transformed into 1
following the route in Scheme 3.27
Glycosylation. With a solid and scalable synthesis in hand,
the glycosylation properties of the donor 1 could be
investigated and compared to the standard benzylidene donor
2. As acceptors, some common alcohols were chosen in order
to compare with literature results; 1-adamantanol 20 was
chosen as an achiral bulky acceptor, diacetone glucose 21 as a
hindered secondary 3-OH, 22 as a primary 6-OH acceptor, and
Competition Experiments. To get further insight into the
reactivity of the C7-analogue donor 1 vs the benzylidene 2,
competition experiments were performed. One equivalent of
each donor was activated using 1 equiv of promoter (NIS/
TfOH) and was allowed to react with 3 equiv of acceptor,
Figure 2.32,33 To validate the method, a competition between
the benzylidene 2 and the perbenzylated mannosyl donor 31
was carried out. From integrals in crude NMR it was estimated
that the ratio of reaction was 19:1 in favor of the perbenzylated,
determined from the ratio of the remaining donor. The value
fits well with the reaction difference obtained by Wong and co-
workers by measuring the donor reactivities, where the
perbenzylated has a relative reactivity value (RRV) of 5000
2193
dx.doi.org/10.1021/jo302455d | J. Org. Chem. 2013, 78, 2191−2205

The Journal of Organic Chemistry
Featured Article
Table 1. Glycosylation between Benzylidene Donor 2 or C7-Analogue 1 and Different Acceptors with Different Activation
Methods
a
b
c
The acceptor was added after the preactivation time. Determined from crude NMR for donor 2 and purified products for donor 1. Not isolated.
d
e
The yield and ratio are estimated from NMR. The yields and selectivities are obtained under conditions with 5 min preactivation. BSP = 1-
benzenesulfinyl piperidine, MS = molecular sieves, NIS = N-iodosuccinimide, Ph2SO = diphenyl sulfoxide, Tf2O = triflic anhydride
(trifluoromethanesulfonic anhydride), TfOH = triflic acid (trifluoromethanesulfonic acid), TTBP = 2,4,6-tri-tert-butylpyrimidine.
2194
dx.doi.org/10.1021/jo302455d | J. Org. Chem. 2013, 78, 2191−2205

The Journal of Organic Chemistry
Featured Article
Figure 3. Low-temperature 1H NMR (one experiment) for the
activation of 1 with Ph2SO/Tf2O (1.3 equiv each) give rise to three
new anomeric peaks (δ 5.78, 6.06, and 6.47 ppm) which were stable
over time. Addition of extra Ph2SO (1.7 equiv) resulted in the
disappearance of the peak at δ 6.06 ppm while the other peak
increased. Warming the sample gradually resulted in the peak at δ 5.78
ppm decreasing while the peak at δ 6.47 increased (−80 °C → −40
°C) (see the Supporting Information for more details).
Figure 2. Competitive glycosylation experiments between thiomanno-
syl donors. Ratios are determined from the H NMR of the crude
reactions by integration of the H-1 signals of the remaining donors.
The reaction was performed in CH2Cl2 at −40 °C and allowed to
reach 0 °C in 3 h.
1
oxosulfonium ion into the more stable α-oxosulfonium ion
33α (Figure 3 and Scheme 6). When CD3OD was added to the
mixture of the α- and β-oxosulfonium ions at −80 °C a fast
consumption of the oxosulfonium ion at 5.78 ppm was
observed; the α-anomer reacted somewhat slower, and at
−30 °C it was completely consumed (Figure 3, Supporting
Information). From crude 1H NMR the β-anomer was found to
be the major product. The slow equilibrium between the two
oxosulfonium ions might be hampering the diastereoselectivity
in the reaction, and one might suspect that it will be even more
predominant with less nucleophilic acceptors. The formation of
the oxosulfonium ions could be avoided by using stoichiometric
amounts of the diphenyl sulfoxide, which gave a clean, but not
complete, formation of the α-triflate (Figure 4, Supporting
Information). When the C7-analogue was activated using 1.3
equiv of BSP as the promoter only the formation of the α-
triflate was observed (Figure 6, Supporting Information).
From the low-temperature NMR it is clear that the C7-
analogue 1 has a surprising preference for forming
oxosulfonium ions 33 over triflates 32, which is opposite that
for the benzylidene donor 2, where the α-triflate was formed
exclusively (Figure 8, Supporting Information). α-Oxosulfo-
nium intermediates have been observed previously with other
glycosyl donors using the same or a similar promoter system36
and used as glycosyl donors.37 β-Oxosulfonium ions have not
previously been observed as an intermediate in glycosylation
reactions, but other related reactive intermediates have been
reported.38
Surprisingly, it was observed that whereas the benzylidene
donor cleanly furnished the expected α-triflate, the C7-analogue
gave a mixture of reactive intermediates (the α-, β-
oxosulfonium ions and the α-triflate) when the exact conditions
were used. The stability of the two α-triflates (benzylidene and
C7-analogue) could also be studied, since their decomposition
temperatures are directly related to the reactivity of the donor.
To study the stability of the intermediate triflates, both triflates
were generated at low temperature after which the temperature
of the NMR sample was gradually increased until decom-
position set in. The decomposition for the C7-analogue proved
to be around −30 °C (Figures 4 and 9, Supporting
and the benzylidene 315 (a ratio of ca. 16).34 When the
competition between the two tethered donors was performed a
difference between the residue donor integrals of 1:2.6 (1/2)
was observed, indicating that the C7-analogue 1 is slightly more
reactive. Competition experiment between the C7-analogue 1
and the armed perbenzylated mannosyl donor 31 resulted in a
residue donor ratio of 1:2.6 (31/1), i.e., the benzyl donor being
more reactive. From these experiments it can be concluded that
the C7-analogue 1 has a reactivity in between that of the armed
donor 31 and the “conformational” disarmed 2.
Low-Temperature NMR Studies. To further investigate
the reactivity of the C7-analogue 1 and get insight into the
reactive intermediates at play, low-temperature NMR experi-
ments35 were carried out. The stability of activated donor over
1
time was also studied by low-temperature H NMR, since it
reflects the reactivity of the intermediate(s) and hence the
donor.
The donor, Ph2SO (1.3 equiv), and TTBP (2.4 equiv) were
1
dissolved in CD2Cl2 and cooled to −80 °C, where a H NMR
was recorded (t = 0 min) followed by activation with Tf2O,
Figure 3. Spectra were then recorded keeping the temperature
at −80 °C at various time (t = 5, 10, and 60 min and 16 h).
Surprisingly, three new anomeric peaks, δ 5.78, 6.06, and 6.47
ppm, were instantly formed upon activation and were stable
over time at −80 °C. To elucidate the identity of the three new
species, extra Ph2SO (1.7 equiv) was added and a spectrum
recorded after 5 min and 3 h. The anomeric peak at δ 6.06 ppm
disappeared and the peaks at δ 5.78 and 6.47 ppm increased,
indicating that the former peak belongs to an α-anomeric
triflate species and the latter two peaks come from an
intermediate oxosulfonium species. This indicated that a β-
oriented oxosulfonium intermediate had been formed (Scheme
6). To our knowledge, this is the first example of a β-
oxosulfonium ion as a reactive intermediate in a glycosylation
reaction.
Increasing the temperature, in intervals of 10 °C starting
from −80 °C and ending at −40 °C, led to the slow decrease of
the anomer peak at δ 5.78 ppm and an increased signal of the
anomer at δ 6.47 ppm, i.e., anomerization of the β-
2195
dx.doi.org/10.1021/jo302455d | J. Org. Chem. 2013, 78, 2191−2205

The Journal of Organic Chemistry
Featured Article
Scheme 6. Relative Ratios of Reactive Intermediates Observed in Low-Temperature NMR Experiments, when Activating the
C7-Analogue at −80 °C
Information), where the benzylidene triflate decomposed
between −20 and −10 °C, which is in line with the
decomposition temperature reported for its 2,3-di-O-methyl
counterpart.39 This places the C7-analogue 1’s reactivity in
between the perbenzylated sugar 31 (decompose between −40
and −30 °C)11 and the benzylidene 2. This result is in line with
the competition experiments described above. The difference,
being only around 10−20 °C, is surprising taking the more
significant difference in rate of hydrolysis into account in the
related gluco derivatives (Figure 1) and the inherently higher
reactivity of 6-deoxy sugars.19
observed. The effect of having an oxygen antiperiplanar (trans−
gauche) to the ring oxygen is minor when it comes to triflate
stability and diastereoselectivity in the glycosylations. Despite
being smaller than anticipated, there is, however, an effect that
becomes more clear when the system is challenged by more
difficult glycosylations involving carbohydrate acceptors. This is
exemplified when the 4-OH of a glucoside is glycosylated,
which is less nucleophilic and more hindered. The reaction
resulted in a complex reaction mixture with a low yield, and the
selectivity dropped from around 10:1 (mannosylation using the
benzylidene donor 2) to 2:1. The marginal difference in triflate
stability, which is important in the borderline cases such as the
glycosylation on the 4-OH, is also confirmed by triflate
decomposition observed at low-temperature NMR.
DISCUSSION
■
The small differences between the diastereoselectivities of the
C7-analogue 1 and the benzylidene donor 2 are surprising.
From previous studies it is evident that a fixed trans−gauche
oxygen is more electron withdrawing than an unrestricted or all
other staggered positions.19 It is also well established that a 6-
deoxy is more reactive (armed) compared with a 6-hydroxy
ether, such as the higher reactivity of rhamnosyl or fucosyl
donors (compared to mannosyl or galactosyl donors,
respectively).40 The results in this work suggest that not only
the stereoelectronics of C6−O but also the conformational
restriction is important for the stereochemical outcome, and the
combination. The clearly disarming effect of having a “4,6
tethering” is only partly due to the 6-O being antiperiplanar to
the ring oxygen. Torsional effects, as suggested by Fraser-Reid
et al.,41 does probably also play a role. The locked 4C1 ground-
state conformation dictates a certain reaction path and limits
the number of possible oxocarbenium ion conformations to
If the antiperiplanar 6-O is not decisive for the increased β-
selectivity when having a 4,6-tethering, the strain induced by
the tethering together with the limited conformational freedom
should be taken into account. With a fixed conformation, the α-
triflate is very favored over the β-triflate in mannosides due to
the anomeric as well as the Δ2-effect.1 These effects are
amplified by a locked conformation, which thereby stabilizes
the α-triflate further and give raise to an increased β-selectivity.
The correlation between α-triflate stability and β-selectivity in
mannosylation using unrestricted donors has earlier been
observed.46 Crich and Vinogradova47 prepared a series of 6-
deoxy-6-fluoro donors with one, two, and three fluorides. The
decomposition of the respective α-triflates increased with
roughly 10 degrees for each fluoride attached, and this resulted
in an increased β-selectivity. The β-selectivity in the
glycosylations decreased when weaker nucleophiles were
used. A similar trend was observed by Kim and co-workers11
when using sulfonyl protective groups on the 3, 4, and 6
positions in a mannosyl donor. It was found that an
unrestricted mannosyl donor’s β-selectivity could be improved
by lowering its reactivity with strongly electron-withdrawing
groups; the weakly associated triflate was again crucial for the
selectivity. The strongest deactivating effect was found when
having a 6-OSO2Bn protective group followed by the 3 position
and the 4 position. The triflate stabilities were again correlated
to the β-selectivity.46 These studies demonstrate that the locked
conformation is not decisive for β-mannosylation and the
selectivity is related to the overall reactivity of the donor.
A less stabilized oxocarbenium ion results in a tighter triflate
and hence a better shielding of the α-site. The oxocarbenium
ion can be destabilized by different means, conformational or
stereoelectronical, but a fixed conformation maximizes the
stereoelectronic effects, i.e., the anomeric and the Δ2 effect in
the covalent intermediate, and hence push the equilibrium
toward it (illustrated by the α-triflate 30 in Scheme 7). Having
a fixed conformation, here restricted by the benzylidene, limits
the conformational freedom and results in further stabilization
3
essentially two; the B2,5 and H4, which both are in a fast
equilibrium with the α-triflate or, depending on the exact
glycosylation conditions, the oxosulfonium ions.
From other studies it has been demonstrated that a restricted
chair conformation is not enough; a 3,4-tethering42 does not
provide the high β-selectivities and neither does 2,3-tethering,
despite the enhanced stability of the respective α-triflates.43
Conformational flip to the axial-rich conformation, however,
does favor α-selectivity in an L-rhamnoside model44 suggesting
the importance of the “ground-state chair” as the starting point.
Whitfield45 has recently suggested, on the basis of computa-
tional chemistry, that the C6-OR tend to shield the β-face of
the oxocarbenium ion, due to electrostatic interaction, and
hence favor the α-product. This is avoided when the O6 and
O4 are tethered together. The C1−O5 flexibility is also more
restricted by this tethering than, e.g., a 3,4-tethering (such as
butane-2,3-diacetal (BDA) protection), which only has a minor
influence on the conformation at C1−O5.
From the glycosylation properties of the C7-analogue 1, in
terms of selectivity, reactivity, and the low-temperature NMR
studies, a surprisingly small difference from the benzylidene 2 is
2196
dx.doi.org/10.1021/jo302455d | J. Org. Chem. 2013, 78, 2191−2205

The Journal of Organic Chemistry
Featured Article
Scheme 7. Suggested Reaction Pathways To Give Either the
α- or β-Mannosides
EXPERIMENTAL SECTION
■
Phenylthio 2,3-Di-O-benzyl-α-D-mannopyranoside (3).49 To
a solution of 2,3-di-O-benzyl-4,6-O-benzylidene-α-D-mannopyranoside
(5.00 g, 9.25 mmol) in MeOH/CH2Cl2 (4:1, 200 mL) was added p-
TsOH monohydrate (440 mg, 2.31 mmol, 0.25 equiv), and the
reaction mixture was stirred under nitrogen at rt for 18 h. Afterward,
the reaction was neutralized with Et3N and concentrated. The crude
was purified by flash column chromatography (EtOAc/petroleum
ether 1:10 → 1:1) which gave the desired product phenylthio 2,3-di-O-
benzyl-α-D-mannopyranoside in 87% (3.64 g) as crystals. The product
can be recrystallized from iPrOH: HRMS (ESI) m/z calcd for
C26H28O5SNa+ 475.1555, found 475.1557. Spectra were in accordance
with literature.49 1H NMR (300 MHz, CDCl3) δ 7.38−7.18 (m, 15H;
2
Ph), 5.49 (s, 1H; H-1), 4.60 (d, J = 12.2 Hz, 1H; CH(Ha)-Ph), 4.51
2
2
(dd, J = 11.9, J = 4.6 Hz, 2H; CH2Ph), 4.40 (d, J = 11.7 Hz, 1H;
CH(Hb)-Ph), 4.07−4.01 (m, 2H; H-4, H-5), 3.95 (s, 1H; H-2), 3.86−
3.70 (m, 2H; H-6), 3.62 (dd, J = 8.9, J = 2.7 Hz, 1H; H-3), 2.32 (s,
1H; OH-4), 1.83 (t, J = 6.4 Hz, 1H; OH-6). 13C NMR (75 MHz,
CDCl3) δ 138.3 (CipsoBn), 138.0 (CipsoBn), 134.4 (CipsoPhS), 132.1 (2
CPh), 129.5 (2 CPh), 129.3 (2 CPh), 128.8 (2 CPh), 128.8 (2 CPh), 128.4
(2 CPh), 128.3 (CparaPh), 128.2 (CparaPh), 127.9 (CparaPh), 86.4 (C-1),
79.9 (C-3), 76.3 (C-5), 74.1 (C-2), 72.6 (CH2Ph), 72.4 (CH2Ph), 67.3
(C-4), 62.5 (C-6).
of the intermediate. The enhanced stability of the covalent
intermediate is unique for the manno stereochemistry since the
optimal stabilization, most often the α-triflate, can be reached
due to the combination of the stereoelectronic effects
mentioned above. This is not the case with, e.g., gluco
stereochemistry, which is only stabilized from the anomeric
effect. When having less stable covalent intermediates or less
nucleophilic acceptors the competing reaction, i.e., the Curtin−
Hammett pathway,48 involving oxocarbenium intermediates
(limited to mainly two species the H3 and B2,5, where the H3
is less stable and therefore more reactive) dominate and the β-
selectivity drops (Scheme 7). With more conformational
freedom and no restriction the oxocarbenium ion is stabilized
(more reactive donor) and the mechanism shifts away from the
now less stable covalent intermediate (such as 30 in Scheme 7)
resulting in less shielding of the α-side and thereby diminished
β-selectivity.
Phenylthio 4-O-Acyl-2,3-di-O-benzyl-6-O-trityl-α-D-manno-
pyranoside (4a−c). Representative procedure: A solution of
phenylthio 2,3-di-O-benzyl-α-D-mannopyranoside 3 (3.95 g, 7.31
mmol), trityl chloride (4.48 g, 16.07 mmol, 2.2 equiv), and DMAP
(179 mg, 1.46 mmol, 0.2 equiv) in pyridine (15 mL) was stirred under
nitrogen at rt for 22 h by which TLC indicated conversion (petroleum
ether/EtOAc 2:1). Additional TrCl (2.24 g, 8.10 mmol, 1.1 equiv) was
added and the reaction mixture stirred for 1 h. Afterward, acetic
anhydride (6.9 mL, 73.06 mmol, 10 equiv) was added, and the mixture
was stirred under nitrogen at rt for 6.5 h. Subsequently, the reaction
mixture was concentrated, coevaporated with toluene, and diluted with
CH2Cl2. To remove excess DMAP and pyridine the solution was
acidified with 1.0 M HCl, and the phases were separated. Afterward,
the organic phase was neutralized with satd aq NaHCO3, washed with
brine, dried (MgSO4), and concentrated. The crude was purified by
flash column chromatography (petroleum ether/EtOAc 25:1→ 10:1)
to give the desired product phenylthio 4-O-acetyl-2,3-di-O-benzyl-6-O-
trityl-α-D-mannopyranoside (4a) in 81% (4.35 g) as a white foam:
[α]RDT +40 (c 1.0, CHCl3); HRMS (ESI) m/z calcd for C47H44O6SNa+
759.2756, found 759.2784; 1H NMR (500 MHz, CDCl3) δ 7.72−7.67
(m, 2H; PhS), 7.53 (d, J = 8.0 Hz, 6H; Ph), 7.45−7.30 (m, 22H; Ph),
5.73 (s, 1H; H-1), 5.44 (t, J3,4 = 9.7 Hz, 1H; H-4), 4.81−4.72 (m, 2H;
4
4
CONCLUSION
■
In conclusion, we have shown that the C7 carbon analogue 1
indeed is β-selective when using the preactivation procedures
developed by Crich. The selectivity is, however, generally lower
than the benzylidene donor 2. The selectivity is therefore not
totally dependent on the electron-withdrawing capacity of the
6-O but also on other effects; conformational, stereoelectronical
and torsional.
2
2
CH2Ph), 4.65 (d, J = 12.2 Hz, 1H; CH(Ha)-Ph), 4.54 (d, J = 12.2
Hz, 1H; CH(Hb)-Ph), 4.50−4.42 (m, 1H; H-5), 4.10 (s, 1H; H-2),
3.83 (dd, J3,4 = 9.4, J2,3 = 1.6 Hz, 1H; H-3), 3.41 (dd, 2J6a,6b =10.1, J5,6a
= 6.7 Hz, 1H; H-6a), 3.23 (d, 2J6a,6b = 10.4 Hz, 1H; H-6b), 1.86 (s, 3H;
CH3CO); 13C NMR (126 MHz, CDCl3) δ 169.5 (CO), 144.0 (3
CipsoPh‑trityl), 138.0 (2 CipsoBn), 134.4 (CipsoPhS), 131.6 (CPh), 129.1
(CPh), 128.9 (CPh), 128.5 (CPh), 127.9 (CPh), 127.8 (CPh), 127.5
(CPh), 127.0 (CPh), 86.8 (Ctert‑trityl), 85.6 (C-1), 77.3 (C-3), 76.1 (C-2),
72.3 (2 Cbenzyl), 71.9 (C-5), 71.8 (Cbenzyl), 68.6 (C-4), 63.4 (C-6), 20.8
(CH3acetyl).
Low-temperature NMR revealed, in contrast to the
benzylidene-protected mannosyl donor, that other reactive
intermediates than the triflate were formed upon activation.
The β-oxosulfonium was observed for the first time as a reactive
intermediate in a glycosylation reaction.
The reactivity of the C7-analogue 1 was found to be in
between that of the perbenzylated (armed) and the benzylidene
2 (“torsional” disarmed) as determined from α-triflate
decomposition. This was further confirmed by competition
experiments between the respective donors. Since the O6 is not
required the scope of the preactivation procedure is broader
and with the optimal conformational restriction other
challenging 1,2-cis β-glycosides, such as the β-rhamnosides,
should be within reach. Following parameters seems to be of
major importance for the β-selectivity: α-triflate stability (by
the anomeric and the Δ2 effect), acceptor reactivity and to
some extent conformational preference of the activated donor.
The O-6 does contribute to the β-selectivity but is not the only
decisive factor.
Phenylthio 2,3-Di-O-benzyl-4-O-pivaloyl-6-O-trityl-α-D-manno-
pyranoside (4b). Pivaloyl chloride (8.3 mg/8.5 mL, 68.94 mmol, 5
equiv). The desired product phenylthio 2,3-di-O-benzyl-4-O-pivaloyl-
6-O-trityl-α-D-mannopyranoside was contaminated with Ph3COH:
HRMS (ESI) m/z calcd for C50H50O6SNa+ 801.3226, found 801.3222;
1H NMR (500 MHz, CDCl3) δ 7.67 (ddd, J = 4.7, J = 2.3, J = 1.3 Hz,
2H; PhS), 7.47−7.44 (m, 6H; Ph), 7.37−7.23 (m, 48H; Ph), 5.71 (d, J
= 1.8 Hz, 1H, H-1), 5.36 (t, J = 9.7 Hz, 1H; H-4), 4.69 (s, 2H;
CH2Ph), 4.54 (s, 2H; CH2Ph), 4.52−4.47 (m, 1H; H-5), 4.06−3.98
2
(m, 1H; H-2), 3.81 (dd, J = 9.4, 3.0 Hz, 1H; H-3), 3.34 (dd, J6a,6b
=
10.2, J = 7.4 Hz, 1H; H-6a), 3.07 (dd, 2J6a,6b = 10.2, J = 1.7 Hz, 1H; H-
6b), 2.82 (brs, 1H, Ph3OH), 0.96 (s, 9H; 3 × CH3);13C NMR (126
MHz, CDCl3) δ 176.7 (CO), 146.9, 143.9, 138.0 (CipsoBn), 137.8
(CipsoBn), 134.6 (CipsoPhS), 131.2 (CPh), 129.1 (CPh), 128.9 (CPh), 128.4
2197
dx.doi.org/10.1021/jo302455d | J. Org. Chem. 2013, 78, 2191−2205

The Journal of Organic Chemistry
Featured Article
(CPh), 128.0 (CPh), 127.8 (CPh), 127.7 (CPh), 127.6 (CPh), 127.3
(CPh), 126.8 (CPh), 86.6(Ctert‑trityl), 85.5 (C-1), 77.8 (C-3), 76.2 (C-2),
72.2 (2 × CH2Ph), 72.0 (C-5), 68.0 (C-4), 63.2 (C-6), 38.6
(Ctert‑pivaloyl), 26.9 (3 × CH3pivaloyl).
137.7(CipsoBn), 133.7 (CipsoPhS), 131.6 (CPh), 129.2 (CPh), 128.5
(CPh), 128.4 (CPh), 128.0 (CPh), 127.9 (CPh), 127.8 (CPh), 127.7
(CPh), 85.9 (C-1), 76.9 (C-3), 75.7 (C-2), 72.4 (Cbenzyl), 72.3 (C-5),
72.0 (Cbenzyl), 68.6 (C-4), 61.7 (C-6), 20.9 (CH3acetyl).
Phenylthio 4-O-Benzoyl-2,3-di-O-benzyl-6-O-trityl-α-D-manno-
pyranoside (4c). Benzoyl chloride (4.40 g/3.63 mL, 31.28 mmol, 2
equiv): [α]RDT + 44 (c 1.0, CHCl3); HRMS (ESI) m/z calcd for
C52H46O6SNa+ 821.2913, found 821.2930; 1H NMR (500 MHz,
CDCl3) δ 7.87 (d, J = 7.5 Hz, 2H; PhS), 7.65−7.58 (m, 3H; Ph), 7.41
(m, 10H; Ph), 7.30 (m, 6H; Ph), 7.24−7.17 (m, 5H; Ph), 7.16−7.09
(m, 9H; Ph), 5.69 (s, 1H; H-1), 5.62 (t, J3,4 = 9.6 Hz, 1H; H-4), 4.74
(s, 2H; CH2Ph), 4.57 (d, J = 12.3 Hz, 1H; CH(Ha)Ph), 4.54−4.48 (m,
1H; H-5), 4.45 (d, J = 12.3 Hz, 1H; CH(Hb)Ph), 4.08 (s, 1H; H-2),
3.89 (dd, J3,4 = 9.3, J = 2.9 Hz, 1H; H-3), 3.40 (dd, J = 10.5, J = 6.5 Hz,
1H; H-6a), 3.27−3.19 (m, 1H; H-6b); 13C NMR (126 MHz, CDCl3) δ
165.2 (CO), 143.8 (CPh‑ipso), 137.9 (CBn‑ipso), 137.7 (CBn‑ipso), 134.3
(CPh‑ipso), 132.9 (CPh), 131.5 (CPh), 129.9 (CPh‑ipso), 129.8 (CPh), 129.1
(CPh), 128.7 (CPh), 128.4 (CPh), 128.3 (CPh), 128.2 (CPh), 127.9
(CPh), 127.7 (CPh), 127.4 (CPh), 126.8 (CPh), 86.75(Ctrityl), 85.67 (C-
1), 75.94 (C-3), 72.34 (CBn), 72.10 (C-5), 71.68 (CBn), 69.18 (C-4),
63.40 (C-6).
Phenylthio 4-O-Acetyl-2,3-di-O-benzyl-α-D-mannopyrano-
side (5a) and Phenylthio 6-O-Acetyl-2,3-di-O-benzyl-α-D-
mannopyranoside (6). Procedure 1. Phenylthio 4-O-acetyl-2,3-di-
O-benzyl-6-O-trityl-α-D-mannopyranoside (4a) (0.49 g, 0.66 mmol)
was dissolved in CH2Cl2/MeOH (9:2, 22 mL) and cooled on ice.
Subsequently, Amberlite IR-120 was added until the solution was
acidic, and the reaction mixture was stirred under nitrogen at rt. The
reaction mixture was diluted with CH2Cl2, and the resin was filtered
and washed twice with CH2Cl2. Afterward, the organic phase was
neutralized with Et3N, concentrated, and purified by flash column
chromatography (petroleum ether/EtOAc 15:1 → 4:1), which gave
the product phenylthio 4-O-acetyl-2,3-di-O-benzyl-α-D-mannopyrano-
side (5a) as a syrup in 55% yield (180 mg) and phenylthio 6-O-acetyl-
2,3-di-O-benzyl-α-D-mannopyranoside (6) in 34% yield (112 mg).
Procedure 2. Phenylthio 4-O-acetyl-2,3-di-O-benzyl-6-O-trityl-α-D-
mannopyranoside (4a) (0.49 g, 0.66 mmol) was dissolved in benzene
(135 mL) and heated under reflux and nitrogen overnight in the
presence of anhydrous CuSO4.(10 g). After cooling and concentration,
the reaction mixture was poured on a dry silica gel column and purified
by flash column chromatography (pure benzene → petroleum ether/
EtOAc 15:1 → 4:1), which gave phenylthio 6-O-acetyl-2,3-di-O-
benzyl-α-D-mannopyranoside (6) in 55% (186 mg) along with
phenylthio 4-O-acetyl-2,3-di-O-benzyl-α-D-mannopyranoside (5a) in
12% yield (40 mg).
Phenylthio-6-O-acetyl-2,3-di-O-benzyl-α-D-mannopyranoside (6):
[α]DRT +27 (c 0.1, CHCl3); HRMS (ESI) m/z calcd for
C28H30O6SNa+: 517.1661, found 517.1684. 1H NMR (500 MHz,
CDCl3) δ 7.36 (dd, J = 8.0, J = 1.5 Hz, 2H; PhS-Hortho), 7.26−7.15 (m,
13H; Ph), 5.51 (d, J1,2 = 1.1 Hz, 1H; H-1), 4.60−4.38 (m, 4H; 2
2
CH2Ph), 4.32 (dd, J6a,6b =12.0, J5,6a = 5.8 Hz, 1H; H-6a), 4.25 (dd,
2J6a,6b =12.0, J5,6b = 2.1 Hz, 1H; H-6b), 4.18 (ddd, J4,5 = 9.4, J5,6a = 5.8,
J5,6b = 1.9 Hz, 1H; H-5), 3.96−3.90 dd, J3,4 = 9.6, J4,OH = 2.0, 2H; H-4),
3.90 (dd, J2,3 = 2.9, J1,2 = 1.5 Hz, 1H; H-2), 3.59 (dd, J3,4 = 9.5, J2,3
=
3.1 Hz, 1H; H-3), 2.75 (d, J4,OH = 2.2 Hz, 1H; 4-OH), 1.93 (s, 3H;
CH3−CO). 13C NMR (126 MHz, CDCl3) δ 171.3 (CO), 137.8
(CipsoBn), 137.7 (CipsoBn), 134.0 (CipsoPhS), 131.7 (CorthoPhS), 129.1
(CorthoPhS), 128.6 (CPh), 128.5 (CPh), 128.1 (CPh), 128.0 (CPh), 127.9
(CPh), 127.7 (CPh), 85.7 (C-1), 79.4 (C-3), 75.6 (C-2), 72.0 (Cbenzyl),
71.9 (Cbenzyl), 71.6 (C-4), 66.8 (C-5), 63.7 (C-6), 20.9 (CH3 acetyl).
Phenylthio 2,3-Di-O-benzyl-4-O-pivaloyl-α-D-mannopyrano-
side (5b). Same procedure as procedure 3 (see above): [α]RDT +48 (c
0.1, CHCl3). HRMS (ESI) m/z calcd for C31H36O6SNa+ 559.2130,
1
found 559.2113; H NMR (500 MHz, CDCl3) δ 7.51−7.47 (m, 2H;
PhS), 7.42−7.30 (m, 13H; Ph), 5.65 (d, J = 1.6 Hz, 1H; H-1), 5.47 (t,
J = 9.8 Hz, 1H; H-4), 4.71 (dd, J = 27.9, 12.3 Hz, 2H; CH2Ph), 4.63−
4.54 (dd, 2H; CH2Ph), 4.19 (ddd, J = 9.9, 4.4, 2.7 Hz, 1H; H-5), 4.05
(dd, J = 3.0, 1.8 Hz, 1H; H-2), 3.96 (dd, J = 9.7, 3.0 Hz, 1H; H-3),
3.73−3.62 (m, 2H; H-6), 2.57 (s, 1H; 6-OH), 1.28 (s, 9H; 3 ×
CH3pivaloyl);. 13C NMR (126 MHz, CDCl3) δ 178.5 (CO), 137.7 (2
× Cipso‑benzyl), 133.8 (CipsoPhS), 131.6 (CPh), 129.2 (CPh), 128.5 (2 CPh),
128.1 (CPh), 127.9 (2 CPh), 127.8 (CPh), 127.7 (CPh), 86.0 (C-1), 77.2
(C-3), 75.90 (C-2), 72.5 (C-5), 72.3 (Cbenzyl), 72.2 (Cbenzyl), 68.2 (C-
4), 61.7 (C-6), 39.0 (Ctert‑pivaloyl), 27.2 (3 × CH3pivaloyl).
Phenylthio 2,3-Di-O-benzyl-4-O-benzoyl-α-D-mannopyrano-
side (5c). Same as procedure 3 (see above): [α]RDT +114 (c 0.3,
CHCl3); HRMS (ESI) m/z calcd for C33H32O6SNa+ 579.1817, found
1
579.1838; H NMR (500 MHz, CDCl3) δ 8.12−8.06 (m, 2H; PhS),
7.69−7.61 (m, 1H; Ph), 7.51 (m, 4H; Ph), 7.43 (m, 2H; Ph), 7.38−
7.24 (m, 11H; Ph), 5.73 (t, J4,5 = 9.8 Hz, 1H; H-4), 5.71 (d, J1,2 = 1.6
Hz, 1H; H-1), 4.77 (dd, J = 31.2, J = 12.4 Hz, 2H; CH2Ph), 4.57 (dd, J
= 48.2, J = 12.2 Hz, 2H; CH2Ph), 4.31 (dt, J4,5 = 9.8, J5,6 = 3.4 Hz, 1H;
H-5), 4.14 (dd, J2,3 = 2.9, J1,2 = 1.9 Hz, 1H; H-2), 4.06 (dd, J3,4 = 9.6,
J2,3 = 3.0 Hz, 1H; H-3), 3.76 (d, J5,6 = 3.5 Hz, 2H; 2 × H-6), 2.53 (br s,
1H; 4-OH); 13C NMR (126 MHz, CDCl3) δ 166.4 (CO), 137.7
(Cipso‑benzyl), 137.6 (Cipso‑benzyl), 133.8 (CipsoPh), 133.5−130.0 (in total 5
CPh), 129.5 (CipsoBz), 129.3−127.8 (in total 15 CPh), 86.1 (C-1), 76.7
(C-3), 75.8 (C-2), 72.5 (C-5), 72.4 (Cbenzyl), 71.9 (Cbenzyl), 69.1 (C-4),
61.7 (C-6).
Procedure 3. Phenylthio 4-O-acetyl-2,3-di-O-benzyl-6-O-trityl-α-D-
mannopyranoside (4a) (500 mg, 0.68 mmol) was dissolved in dry
CH2Cl2 (20 mL) and cooled on ice. Detritylation was affected by the
addition of trifluoroacetic acid (78 μL, 1.02 mmol, 1.5 equiv) and
trifluoroacetic anhydride (TFFA, 141 μL, 1.02 mmol, 1.5 equiv), and
the mixture was stirred overnight under nitrogen at 4 °C. The reaction
mixture was neutralized with Et3N (0.5 mL) by which excess TFAA
was removed and the alcohol detrifluoroacetylated by pouring the
mixture into MeOH (10 mL). After concentration, the crude was
dissolved in CH2Cl2, washed with brine, dried (MgSO4), filtered, and
concentrated. The crude was purified by dry column chromatog-
raphy50 (heptane/EtOAc 15:1 → 4:1) which gave the desired product
phenylthio 4-O-acetyl-2,3-di-O-benzyl-α-D-mannopyranoside (5a) in
66% yield (223 mg) along with phenylthio 6-O-acetyl-2,3-di-O-benzyl-
α-D-mannopyranoside 6 in 10% yield (35 mg).
(E)-Phenylthio 2,3-Di-O-benzyl-4-O-pivaloyl-6,7-dideoxy-8-
C-phenyl-α-D-manno-oct-6-eno-8-ulo-pyranoside (9a). To a
solution of phenylthio 2,3-di-O-benzyl-4-O-pivaloyl-α-D-mannopyra-
noside 5b (969 mg, 1.81 mmol) in dry CH2Cl2 (25 mL) was added
Dess−Martin periodinane (DMP, 1.15 g, 2.71 mmol, 1.5 equiv), and
the reaction mixture was stirred under nitrogen at rt for 22 h. The
resulting suspension was diluted with CH2Cl2, and satd aq NaHCO3
was added followed by Na2S2O3 (560 mg).The mixture was stirred for
2 h to secure complete reduction of excess DMP. Subsequently, the
water phase was extracted with CH2Cl2 (3×), and the combined
organic phases was washed with satd aq NaHCO3 and brine, dried
(MgSO4), filtered, and concentrated to give 973 mg of the crude
phenylthio 2,3-di-O-benzyl-4-O-pivaloyl-α-D-manno-hexodialdo-1,5-
pyranoside (7b). The crude aldehyde 7b (888 mg, 1.66 mmol) was
dissolved in dry CH2Cl2 (30 mL), and (benzoylmethylene)-
triphenylphosphorane (950 mg, 2.49 mmol, 1.5 equiv) was added.
The reaction mixture was stirred under nitrogen at rt for 22 h.
Afterward, the reaction mixture was concentrated, and cold Et2O was
added by which some Ph3PO precipitated and was removed by
filtration through a silica pad. Subsequently, the mixture was
concentrated and the crude was purified by dry column chromatog-
raphy (heptane with 1.5% gradient of EtOAc) which gave the desired
Phenylthio 4-O-acetyl-2,3-di-O-benzyl-α-D-mannopyranoside (5a):
[α]RDT +12 (c 0.1, CHCl3); HRMS (ESI) m/z calcd for C28H30O6SNa+
517.1661, found 517.1667; 1H NMR (500 MHz, CDCl3) δ 7.34 (dd, J
= 8.0, 1.5 Hz, 2H; PhS-Hortho), 7.29−7.18 (m, 13H; Ph), 5.51 (d, J1,2
=
1.6 Hz, 1H; H-1), 5.30 (t, J3,4 = 9.8, J42,5 = 9.8, Hz, 1H; H-4), 4.60 (dd,
2J = 12.4 Hz, 2H; CH2Ph), 4.46 (dd, J = 12.1 Hz, 2H; CH2Ph), 4.03
(dt, J4,5 = 9.9, J5,6 = 3.5 Hz, 1H; H-5), 3.92 (dd, J2,3 = 3.0, J1,2 = 1.8 Hz,
1H; H-2), 3.76 (dd, J3,4 = 9.7, J2,3 = 3.1 Hz, 1H; H-3), 3.57 (d, J5,6a
=
3.6 Hz, 2H; H-6), 2.27 (s, 1H, 4-OH), 1.99 (s, 3H; CH3CO). 13C
NMR (126 MHz, CDCl3) δ 170.9 (CO), 137.9 (CipsoBn),
2198
dx.doi.org/10.1021/jo302455d | J. Org. Chem. 2013, 78, 2191−2205

The Journal of Organic Chemistry
Featured Article
product (E)-phenylthio 2,3-di-O-benzyl-4-O-pivaloyl-6,7-dideoxy-8-C-
phenyl-α-D-manno-oct-6-eno-8-ulo-pyranoside (9a) in 73% yield (838
mg): Rf 0.3 (heptane/EtOAc 5.5:1); [α]RDT +67 (c 0.1, CHCl3), Rf 0.3
(heptane/EtOAc 5:1); HRMS (ESI) m/z calcd for C39H40O6SNa+
659.2443, found 659.2469; 1H NMR (500 MHz, CDCl3) δ 7.78 (dd, J
= 8.3, J = 1.2 Hz, 2H; PhS), 7.46−7.42 (m, 1H; Ph), 7.37−7.31 (m,
4H; Ph), 7.27−7.17 (m, 13H; Ph), 6.92 (dd, J6,7trans = 15.6, J = 1.6 Hz,
1H; H-7), 6.75 (dd, J6,7trans = 15.6, J5,6 = 4.9 Hz, 1H; H-6), 5.50 (d, J =
2.0 Hz, 1H; H-1), 5.38 (t, J = 9.5 Hz, 1H; H-4), 4.78 (ddd, J4,5 = 9.7,
J5,6 = 4.9, J = 1.3 Hz, 1H; H-5), 4.58 (s, 2H; CH2Ph), 4.47 (s, 2H;
CH2Ph), 3.95−3.89 (m, 1H; H-2), 3.80 (dd, J3,4 = 9.4, J = 2.9 Hz, 1H;
H-3), 1.10 (s, 9H; 3 × CH3pivaloyl). 13C NMR (126 MHz, CDCl3) δ
191.0 (PhCO), 177.0 (CO), 142.1 (C-6), 137.7 (2 × Cipso‑benzyl),
137.4 (CipsoPh), 133.6 (CipsoPhS), 132.8 (CPh), 131.8 (CPh), 129.2 (CPh),
128.9 (CPh), 128.6 (CPh), 128.5 (CPh), 128.0 (CPh), 127.9 (CPh), 127.7
(CPh), 127.2 (C-7), 86.2 (C-1), 77.5 (C-3), 76.0 (C-2), 72.7 (Cbenzyl),
72.3 (Cbenzyl), 71.4 (C-5), 70.2 (C-4), 38.9 (Ctert‑pivaloyl), 27.2 (3 ×
CH3pivaloyl).
After standing overnight under oxygen-free argon, the red crystals were
collected by filtration, washed with degassed acetonitrile (3 × 50 mL),
and dried by applying oil pump vacuum. Stryker’s reagent was
obtained as dark red crystals in 82% and used directly in the next step.
Phenylthio 2,3-Di-O-benzyl-4-O-pivaloyl-6,7-dideoxy-8-C-
phenyl-α-D-manno-oct-8-ulopyranoside (11a). Schlenk techni-
ques were used for the 1,4-reduction. Freshly made Stryker’s reagent
(1.90g, 0.97 mmol, 0.37 equiv) was dissolved in dry and degassed
benzene (5 mL, dried by passing through basic Al2O3 and degassed by
three freeze−thaw cycles, refilling with argon). Then (E)-phenylthio
2,3-di-O-benzyl-4-O-p-pivaloyl-6,7-dideoxy-8-C-phenyl-α-D-manno-
oct-6-eno-8-ulo-pyranoside (9a) (1.68 g, 2.64 mmol) dissolved in dry
and degassed benzene (25 mL, see above) was added by cannulation
along with degassed water (118 μL). The resulting dark red solution
was stirred under argon until completion as indicated by NMR of the
crude reaction mixture (30 min). Afterward, the system was opened,
and air was bubbled through the solution by which copper-containing
degradation products precipitated. The solution was filtered through a
bed of Celite and washed with Et2O followed by concentration in
vacuo. The crude was purified by flash column chromatography
(heptane/EtOAc 15:1 → 5:1) to give the product as crystals in 86%
yield (1.45 g): Rf 0.3 (heptane/EtOAc 5:1); [α]RDT +62 (c 0.8, CHCl3);
mp 108.5−108.9 °C; HRMS (ESI) m/z calcd for C39H42O6SNa+
661.2600, found 661.2626. 1H NMR (500 MHz, CDCl3) δ 8.13 (dd, J
= 8.4, J = 1.3 Hz, 2H; Ph), 7.87 (dd, J = 8.4, J = 1.3 Hz, 2H; Ph),
7.67−7.61 (m, 1H; Ph), 7.58−7.54 (m, 1H; Ph), 7.53−7.49 (m, 2H;
Ph), 7.46−7.26 (m, 12H; Ph), 7.24−7.17 (m, 5H; Ph), 5.71 (t, J = 9.7
Hz, 1H; H-4), 5.69 (d, J = 1.7 Hz, 1H; H-1), 4.84−4.77 (m, 2H;
CH2Ph), 4.62 (d, J = 12.2 Hz, 1H; C(Ha)HPh)), 4.51 (d, J = 12.2 Hz,
1H; C(Hb)HPh)), 4.41 (td, J = 9.3, J = 2.8 Hz, 1H; H-5), 4.11 (dd, J =
2.9, J = 1.9 Hz, 1H; H-2), 4.00 (dd, J = 9.5, J = 3.0 Hz, 1H; H-3), 3.11
(ddd, J = 17.9, J = 8.2, J = 5.1 Hz, 1H; H-7a), 2.97 (dt, J = 17.9, J = 7.4
Hz, 1H; H-7b), 2.24 (tdd, J = 9.9, J = 7.6, J = 2.9 Hz, 1H; H-6a), 2.03−
1.93 (m, 1H; H-6b); 13C NMR (126 MHz, CDCl3) δ 199.4 (PhC
Oketone), 165.8 (PhCOester), 137.9 (CBn‑ipso), 137.7 (CBn‑ipso), 136.9
(CPh‑ipso), 134.0−133.8 (2 CPh), 133.6 (CPh‑ipso), 133.2−127.5 (23
CPh), 85.7 (C-1), 77.1 (C-3), 75.9 (C-2), 72.6 (CBn), 72.0 (C-4), 71.8
(CBn), 70.6 (C-5), 33.6 (C-7), 25.3 (C-6).
(E)-Phenylthio 4-O-Benzoyl-2,3-di-O-benzyl-6,7-dideoxy-8-
C-phenyl-α-D-manno-oct-6-eno-8-ulo-pyranoside (9b) and (E)-
Phenylthio 2,3-Di-O-benzyl-4,6,7-trideoxy-8-C-phenyl-α-L-er-
ythro-oct-6-eno-8-ulopyranoside (10). Same procedure as above
which on small scale gave the desired compound in 73% (838 mg) but
on larger scale gave the eliminated product 10 in 97% yield (2.79 g).
(E)-Phenylthio 4-O-benzoyl-2,3-di-O-benzyl-6,7-dideoxy-8-C-phenyl-
α-D-manno-oct-6-eno-8-ulo-pyranoside: Rf 0.3 (heptane/EtOAc
5.5:1); [α]RDT +114 (c 0.1, CHCl3); HRMS (ESI) m/z calcd for
C41H36O6SNa+ 679.2130, found 679.214; 1H NMR (500 MHz,
CDCl3) δ 8.10−8.04 (m, 2H; Ph), 7.84−7.78 (m, 2H; Ph), 7.64 (t, J =
7.4 Hz, 1H; Ph), 7.55−7.20 (m, 20H; Ph), 7.08 (dd, J6,7 = 15.6, J = 1.4
Hz, 1H; H-7), 6.97 (dd, J6,7 = 15.6, J = 4.7 Hz, 1H; H-6), 5.73 (t, J =
9.5 Hz, 1H; H-4), 5.67 (d, J = 1.9 Hz, 1H; H-1), 5.06−4.96 (m, 1H;
H-5), 4.77 (s, 2H; CH2Ph), 4.62 (d, J = 12.2 Hz, 1H; C(Ha)HPh),
4.51 (d, J = 12.2 Hz, 1H; C(Hb)HPh), 4.15−4.10 (m, 1H; H-2), 4.01
(dd, J = 9.3, J = 2.9 Hz, 1H; H-3); 13C NMR (126 MHz, CDCl3) δ
190.6 (PhCOketone), 165.3 (PhCOester), 141.9 (C-6), 137.7
(CPh‑ipso), 137.5 (CPh‑ipso), 137.4 (CPh‑ipso), 133.6 (CPh‑ipso), 133.3−
129.9 (4 CPh), 129.7 (CPh‑ipso), 129.2−127.8 (21 CPh), 127.0 (C-7),
86.1 (C-1), 76.8 (C-3), 75.8 (C-2), 72.6 (CBn), 71.9 (CBn), 71.3 (C-5),
71.2 (C-4). (E)-Phenylthio 2,3-di-O-benzyl-4,6,7-trideoxy-8-C-phenyl-
α-L-erythro-oct-6-eno-8-ulo-pyranoside (10): [α]DRT +210 (c 1.0,
CHCl3); mp 122−123 °C; HRMS (ESI) m/z calcd for
C34H30O4SNa+ 557.1757, found 557.1749; 1H NMR (500 MHz,
CDCl3) δ 7.82 (dd, J = 8.4 Hz, J = 1.4 Hz, 2H; Ph), 7.45 (m, 3H; Ph),
7.38 (m, 2H; Ph), 7.31−7.18 (m, 13H; Ph), 7.12 (d, J6,7 = 15.2 Hz,
1H; H-7), 6.99 (d, J6,7 = 15.2 Hz, 1H; H-6), 5.61 (d, J = 4.9 Hz, 1H;
H-1), 5.38 (d, J = 3.4 Hz, 1H; H-4), 4.63 (s, 2H; CH2Ph), 4.54 (s, 2H;
CH2Ph), 4.26 (t, J = 3.8 Hz, 1H; H-3), 3.84 (t, J = 4.5 Hz, 1H; H2);
13C NMR (126 MHz, CDCl3) δ 190.2 (PhCOketone), 148.7 (C-5),
Phenylthio 4-O-Benzoyl-2,3-di-O-benzyl-6,7-dideoxy-8-C-
phenyl-α-D-manno-oct-8-ulopyranoside (11b). Same procedure
as above using (E)-phenylthio 4-O-benzoyl-2,3-di-O-benzyl-6,7-
dideoxy-8-C-phenyl-α-D-manno-oct-6-eno-8-ulopyranoside (9b) (5.00
g, 7.61 mmol) giving the desired product in 66% yield (3.29 g): Rf 0.3
(heptane/EtOAc 4:1); [α]RDT +8 (c 0.5, CHCl3); HRMS (ESI) m/z
1
calcd for C41H38O6SNa+ 681.2287, found 681.2322; H NMR (500
MHz, CDCl3) δ 8.13 (dd, J = 8.4, J = 1.3 Hz, 2H; Ph), 7.87 (dd, J =
8.4, J = 1.3 Hz, 2H; Ph), 7.67−7.61 (m, 1H; Ph), 7.58−7.54 (m, 1H;
Ph), 7.53−7.49 (m, 2H; Ph), 7.46−7.26 (m, 12H; Ph), 7.24−7.17 (m,
5H; Ph), 5.71 (t, J = 9.7 Hz, 1H; H-4), 5.69 (d, J = 1.7 Hz, 1H; H-1),
4.84−4.77 (m, 2H; CH2Ph), 4.62 (d, J = 12.2 Hz, 1H; C(Ha)HPh)),
4.51 (d, J = 12.2 Hz, 1H; C(Hb)HPh)), 4.41 (td, J = 9.3, J = 2.8 Hz,
1H; H-5), 4.11 (dd, J = 2.9, J = 1.9 Hz, 1H; H-2), 4.00 (dd, J = 9.5, J =
3.0 Hz, 1H; H-3), 3.11 (ddd, J = 17.9, J = 8.2, J = 5.1 Hz, 1H; H-7a),
2.97 (dt, J = 17.9, J = 7.4 Hz, 1H; H-7b), 2.24 (tdd, J = 9.9, J = 7.6, J =
2.9 Hz, 1H; H-6a), 2.03−1.93 (m, 1H; H-6b); 13C NMR (126 MHz,
CDCl3) δ 199.4 (PhCOketone), 165.8 (PhCOester), 137.9 (CBn‑ipso),
137.7 (CBn‑ipso), 136.9 (CPh‑ipso), 134.0−133.8 (2 CPh), 133.6 (CPh‑ipso),
133.2−127.5 (23 CPh), 85.7 (C-1), 77.1 (C-3), 75.9 (C-2), 72.6 (CBn),
72.0 (C-4), 71.8 (CBn), 70.6 (C-5), 33.6 (C-7), 25.3 (C-6).
Phenylthio 2,3-Di-O-benzyl-6,7-dideoxy-8-C-phenyl-α-D-
manno-oct-8-ulopyranoside (12). Procedure 1 (acyl deprotec-
tion): Phenylthio 2,3-di-O-benzyl-4-O-pivaloyl-6,7-dideoxy-8-C-phe-
nyl-α-D-manno-oct-8-ulopyranoside (300 mg, 470 mmol) was
dissolved in MeOH/THF (5:2, 14 mL) followed by the addition of
NaOMe (0.3 mL) until the reaction mixture was slightly basic. The
reaction mixture was stirred under nitrogen at rt and followed by TLC
(heptane/EtOAc 5:1, Rf 0.21). After 6 days, TLC only showed half
conversion but the reaction was acidified by the addition of Amberlite
IR-120. Subsequently, concentration and dry column chromatography
(heptane with a 1.7% gradient of EtOAc) gave the starting material in
138.0 (CPh‑ipso), 137.95 (CPh‑ipso), 137.9 (C-6), 137.7 (CPh‑ipso), 133.3
(CPh), 133.1 (CPh), 132.6 (CPh‑ipso), 129.2 (CPh), 128.7 (CPh), 128.6
(CPh), 128.4 (CPh), 128.3 (CPh), 128.1 (CPh), 128.0 (CPh), 127.9
(CPh), 123.0 (C-7), 110.3 (C-4), 84.1 (C-1), 72.6 (C-2), 72.3 (CBn),
71.4 (CBn), 69.3 (C-3).
Stryker’s Reagent: [CuH(PPh)3]6.51 Schlenk techniques were
used for the synthesis of the air-sensitive “copper hydride”. Anhydrous
copper(II) acetate (3.75 g, 20.64 mmol, prepared from the
monohydrate by heating at 105 °C in vacuo) and PPh3 (10.83 g
41.29 mmol, 2 equiv) were weighed into an dried, argon-filled round-
bottomed two-necked flask equipped with a sintered glass. Benzene
(45 mL), dried by passing through basic Al2O3 and degassed by three
freeze−thaw cycles, refilling with argon, was added, followed by
diphenylsilane (4.57 mL, 24.77 mmol, 1.2 equiv). The reaction
mixture turned from a blue suspension to a green solution within 5
min. The color of the reaction further changed from green to dark red
in 1 h. The resulting homogeneous solution was stirred for another 1 h
and concentrated under vacuum to approximately one-third of its
volume. Anhydrous acetonitrile (45 mL, from a solvent purification
system), degassed by three freeze−thaw cycles, was slowly layered
onto the benzene solution to promote crystallization of the product.
2199
dx.doi.org/10.1021/jo302455d | J. Org. Chem. 2013, 78, 2191−2205

The Journal of Organic Chemistry
Featured Article
36% yield (107 mg) and the desired product in 57% yield (148 mg).
Procedure 1 was also used for the deprotection of phenylthio 4-O-
benzoyl-2,3-di-O-benzyl-6,7-dideoxy-8-C-phenyl-α-D-manno-oct-8-ulo-
pyranoside (11b) giving the desired product 12 in quantitative yield:
[α]RDT +3.0 (c 0.4, CHCl3); HRMS (ESI) m/z calcd for C34H34O5SNa+
577.2025, found 577.2034; 1H NMR (500 MHz, CDCl3) δ 7.86−7.74
(m, 2H; Ph), 7.50−7.44 (m, 1H; Ph), 7.36−7.16 (m, 17H; Ph), 5.57
(d, J1,2 = 1.0 Hz, 1H; H-1), 4.52 (m, 4H; 2 × CH2Ph), 4.00 (td, J =
9.0, J = 2.7 Hz, 1H; H-5), 3.94 (dd, J2,3 = 3.0, J1,2 = 1.4 Hz, 1H; H-2),
(C-1), 79.6 (C-3), 75.9 (C-2), 75.9, 73.3, 72.07 (CBn), 72.0 (CBn),
68.7, 64.2 (C-6), 26.1 (Cmethyl), 18.5 (Cquart.), −5.2 (Cmethyl).
Phenylthio 2,3-Di-O-benzyl-4-O-(p-methoxybenzyl)-6-(tert-
butyldimethylsiloxy)-α-D-mannopyranoside (17). A solution of
phenylthio 2,3-di-O-benzyl-6-(tert-butyldimethylsiloxy)-α-D-manno-
pyranoside (16) (1.43 g, 2.53 mmol) and p-methoxybenzyl chloride
(0.44 mL, 27.79 mmol, 1,1 equiv) in DMF was cooled to −8 °C.
Then, NaH (60% in mineral oil, 0.25 g) was added, and the mixture
was allowed to stir under N2 for 1.5 h. The reaction was quenched by
addition of MeOH, and the mixture was concentrated and purified by
dry column chromatography (heptane with a 0.8% gradient of EtOAc)
to give the product 17 as a colorless syrup in 66% yield (1.15 g): Rf 0.3
(heptane/EtOAc 12:1); HRMS (ESI) m/z calcd for C40H50O6SSiNa+
3.83 (t, J3,4 = 9.4, J4,OH = 1.9 Hz, 1H; H-4), 3.64 (dd, J3,4 = 9.4, J2,3
=
3.1 Hz, 1H; H-3), 3.00 (ddd, J7a,7b = 17.9, J = 8.8, J = 5.4 Hz, 1H; H-
7a), 2.80 (ddd, J7a,7b = 17.9, J = 8.5, J = 6.3 Hz, 1H; H-7a), 2.50 (d,
J4,OH = 2.3 Hz, 1H; 4-OH), 2.34−2.23 (m, 1H; H-6a), 1.92 (dtd, J =
14.2, J = 8.6, J = 5.5 Hz, 1H; H-6b); 13C NMR (126 MHz, CDCl3) δ
199.9 (Ph−CO), 137.8 (2 × CBn‑ipso), 136.9 (CPh‑ipso), 134.0
(CPh‑ipso), 132.9 (CPh), 131.5 (CPh), 129.1 (CPh), 128.6 (CPh), 128.5
(CPh), 128.1 (CPh), 128.0 (CPh), 127.9 (CPh), 127.4 (CPh), 85.2 (C-1),
79.7 (C-3), 75.5 (C-2), 72.4 (C-5), 72.2 (Cbenzyl), 71.8 (Cbenzyl), 70.2
(C-4), 34.1 (C-7), 25.4 (C-6).
709.2996, found 709.3000; [α]RDT +76 (c 0.4, CHCl3); H NMR (500
1
MHz, CDCl3) δ 7.46−7.42 (dd, J = 8.1, 1.6 Hz, 2H; Ph), 7.39−7.24
2
(m, 16H; Ph), 6.88−6.85 (m, 2H; Ph), 5.57−5.54 (d, J1,2 = 1.7 Hz,
1H; H-1), 4.88−4.84 (d, J = 10.5 Hz, 1H; CH(Ha)Ph), 4.70−4.61 (m,
4H; 2 × CH2Ph), 4.60−4.57 (d, 2J = 10.6 Hz, 1H; CH(Hb)Ph), 4.10−
4.05 (m, 1H), 4.01−3.95 (m, 2H; H-2), 3.91−3.87 (m, 1H), 3.86−
3.82 (m, 2H; ; H-6a, H-6b), 3.82−3.80 (s, 3H; CH3O), 0.90−0.88 (s,
Phenylthio 2,3-Di-O-benzyl-6,7-dideoxy-8-C-phenyl-4-O-
(trimethylsilyl)-α-D-manno-oct-8-ulopyranoside (13). To a sol-
ution of phenylthio 2,3-di-O-benzyl-6,7-dideoxy-8-C-phenyl-α-D-
manno-oct-8-ulopyranoside (12) (137 mg, 0.25 mmol) in DMF (3
mL) were added imidazole (50 mg, 0.74 mmol, 3 equiv) and
trimethylsilyl chloride (94 μL, 0.74 mmol, 3 equiv). The reaction
mixture was stirred under nitrogen at rt for 3 h. Subsequently, the
mixture was diluted with EtOAc and H2O and separated. The water
phase was extracted with EtOAc (3×), and the combined organic
phases were washed with brine, dried (MgSO4), filtered, and
concentrated. The crude was purified by dry column chromatography
(heptane with a 0.9% gradient of EtOAc) giving the desired product in
92% yield (142 mg): Rf 0.3 (heptane/EtOAc 10:1); [α]RDT +12 (c 0.7,
CHCl3); HRMS (ESI) m/z calcd for C37H42O5SSiNa+ 649.2425,
found 649.2418; 1H NMR (500 MHz, CDCl3) δ 7.81 (dt, J = 8.5, J =
1.5 Hz, 2H; Ph), 7.55−7.49 (m, 1H; Ph), 7.42−7.20 (m, 17H;), 5.57
(d, J1,2 = 1.5 Hz, 1H; H-1), 4.65 (s, 2H; CH2Ph), 4.59 (dd, J = 32.1, J
= 11.9 Hz, 2H; CH2Ph), 3.97 (td, J5,6 = 9.5, J4,5 = 2.3 Hz, 1H; H-5),
3.94−3.88 (m, 2H; H-2, H-4), 3.66 (dd, J3,4 = 8.8, J2,3 = 3.0 Hz, 1H; H-
3), 2.98 (ddd, J7a,7b = 17.8, J = 9.9, J6b,7a = 4.8 Hz, 1H; H-7a), 2.74
(ddd, J7a,7b = 17.8, J = 9.5, J6a,7b = 5.8 Hz, 1H; H-7b), 2.38−2.26 (m,
1H; H-6a), 1.79 (dtd, J6a,6b = 14.4, J5,6 = 9.6, J6b,7a = 4.8 Hz, 1H; H-6b),
0.15 (s, 9H; 3 × CH3); 13C NMR (126 MHz, CDCl3) δ 199.7 (C
O), 138.4 (CBn‑ipso), 138.3 (CBn‑ipso), 137.1 (CPh‑ipso), 134.2(CPh‑ipso),
132.9 (CPh), 131.5 (CPh), 129.1 (CPh), 128.6 (CPh), 128.5 (CPh), 128.1
(CPh), 127.9 (2 × CPh), 127.8 (CPh), 127.7 (CPh), 127.3 (CPh), 85.5
(C-1), 80.4 (C-3), 76.6 (C-2), 73.6 (C-5), 72.7 (CBn), 72.4 (CBn), 72.3
(C-4), 34.8 (C-7), 25.9 (C-6), 0.9 (3 × CTMS).
9H; 3 × CH3), 0.07−0.06 (s, 3H; CH3), 0.06−0.05 (s, 3H; CH3); 13
C
NMR (126 MHz, CDCl3) δ 159.4 (CPh‑ipso), 138.5 (CPh‑ipso), 138.2
(CPh‑ipso), 135.0 (CPh‑ipso), 131.6 (CPh), 131.0 (CPh), 129.8 (CPh), 129.0
(CPh), 128.5 (CPh), 128.4 (CPh), 128.0 (CPh), 127.9 (CPh), 127.8
(CPh), 127.7 (CPh), 127.4 (CPh), 113.9 (CPh), 85.7 (C-1), 80.3, 76.7,
75.0 (CBn), 74.8, 74.37, 72.3 (CBn), 71.9 (CBn), 62.8 (C-6), 55.4
(CMeO), 26.1 (Cmethyl), 18.5 (Cquart), −5.0 (Cmethyl), −5.1 (Cmethyl).
Phenylthio 2,3-Di-O-benzyl-4-O-(p-methoxybenzyl)-α-D-
mannopyranoside (18). To a solution of phenylthio 2,3-di-O-
benzyl-4-O-(p-methoxybenzyl)-6-(tert-butyldimethylsiloxy)-α-D-man-
nopyranoside (17) (11.50 g, 16.74 mmol) in 200 mL of dry CH2Cl2
was added tetra-n-butylammonium fluoride (1.0 M in THF, 20.1 mL,
20.1 mmol). The reaction was kept stirred under N2 for 18 h. Then,
the reaction mixture was concentrated, and the residue was purified by
dry column chromatography (heptane with a 2.2% gradient of EtOAc)
to give the pure compound 18 as a colorless syrup in 99% yield (9.51
g): Rf 0.36 (heptane/EtOAc 2:1); HRMS (ESI) m/z calcd for
C34H36O6SNa+ 595.2130, found 595.2124; [α]DRT +90 (c 0.3, CHCl3);
1H NMR (500 MHz, CDCl3) δ 7.46−7.31 (m, 17H; Ph), 6.93 (d, J =
8.6 Hz, 2H; Ph), 5.58 (d, J1,2 = 1.8 Hz, 1H; H-1), 4.94 (d, 2J = 10.6 Hz,
1H; CH(Ha)Ph), 4.75 (m, 2H; CH2Ph), 4.73−4.65 (m, 3H; CH2Ph,
CH(Hb)Ph), 4.18 (m, 1H; H-5), 4.10 (t, J3,4 = 9.4 Hz, 1H; H-4), 4.06
(t, J = 2.4 Hz, 1H; H-2), 3.95 (dd, J3,4 = 9.2, J = 3.0 Hz, 1H; H-3),
3.90−3.83 (m, 6H; MeO, 2 × H-6, 6-OH); 13C NMR (126 MHz,
CDCl3) δ 159.4 (CPh‑ipso), 138.2 (CBn‑ipso), 137.9 (CBn‑ipso), 134.0
(CSPh‑ipso), 131.9 (CPh), 130.5 (CPh), 129.8 (CPh), 129.2 (CPh), 128.5
(CPh), 128.5 (CPh), 128.0 (CPh), 127.9 (CPh), 127.8 (CPh), 127.8
(CPh), 127.7 (CPh), 113.9 (CPh), 86.1 (C-1), 80.2 (C-3), 76.5 (C-2),
75.0 (CBn), 74.6 (C-4), 73.4 (C-5), 72.4 (CBn), 72.3 (CBn), 62.3 (C-6),
55.4 (CMeO).
Phenylthio 2,3-Di-O-benzyl-6-(tert-butyldimethylsiloxy)-α-D-
mannopyranoside (16).52 A solution of phenylthio 2,3-di-O-benzyl-
α-D-mannopyranoside (3) (22.20 g, 49.05 mmol) in DMF was cooled
to 0 °C. Imidazole (3.51 g, 51.51 mmol, 1.05 equiv) and tert-
butyldimethylsilyl chloride (7.76 g, 51.51 mmol, 1.05 equiv) were
added, and the reaction mixture was warmed to rt under N2. After
being stirred for 3 h, the reaction mixture was quenched with MeOH
and concentrated. The residue was taken up in Et2O and washed two
times with water, one time with brine, dried (MgSO4), and
concentrated. The crude was purified by dry column chromatography
(heptane with a 1.3% gradient of EtOAc) to give the product 15 as a
colorless syrup in 78% yield (21.63 g): Rf 0.3 (heptane/EtOAc 7:1);
HRMS (ESI) m/z calcd for C32H42O5SSiNa+ 589.2420, found
589.2413; [α]RDT +24 (c 0.4, CHCl3); spectra were in accordance
with literature;9 1H NMR (500 MHz, CDCl3) δ 7.38−7.35 (dd, J =
(E)-Phenylthio 2,3-Di-O-benzyl-4-O-(p-methoxybenzyl)-6,7-
dideoxy-8-C-phenyl-α-D-manno-oct-6-eno-8-ulopyranoside
(9c). To a solution of phenylthio 2,3-di-O-benzyl-4-O-(p-methox-
ybenzyl)-α-D-mannopyranoside (18) (4.00 g, 6.98 mmol) in 100 mL
of dry CH2Cl2 was added Dess−Martin periodinane (4.44 g, 10.48
mmol, 1.5 equiv). The reaction was stirred under N2 for 5.5 h before
the reaction mixture was quenched with satd aqueous NaHCO3 and
Na2S2O3 (1.65 g). The reaction mixture was stirred for an additional 1
h followed by extraction of the water phase with CH2Cl2. The
combined organic phases were washed with satd NaHCO3 and brine,
dried (MgSO4), and concentrated. To a solution of crude phenylthio
2,3-di-O-benzyl-4-O-(p-methoxybenzyl)-α-D-manno-hexodialdo-1,5-
pyranoside 7c (3.99 g, 6.99 mmol) in dry CH2Cl2 (100 mL) was
added (benzoylmethylene)triphenylphosphorane (3.19 g, 8.39 mmol,
1.2 equiv). The reaction mixture was stirred for 13 h under N2
followed by concentration in vacuo and purification by dry column
chromatography (heptane with a 1.4% gradient of EtOAc. This gave
the product 9c as a colorless syrup in 74% (3.49 g) over two steps: Rf
0.3 (heptane/EtOAc 5:1); HRMS (ESI) m/z calcd for C42H40O6SNa+
8.1, J = 1.5 Hz, 2H; Ph), 7.27−7.15 (m, 13H; Ph), 5.51−5.48 (d, J1,2
=
1.6 Hz, 1H; H-1), 4.60−4.56 (m, 1H; CH(Ha)Ph), 4.52−4.44 (m, 3H;
CH2Ph; CH(Hb)Ph), 4.04−3.98 (m, 2H; H-4, H-5), 3.91−3.88 (dd,
J2,3 = 3.1, J1,2 = 1.6 Hz, 1H; H-2), 3.85−3.78 (m, 2H; H-6a, H-6b),
3.63−3.57 (m, 1H; H-3), 0.83−0.80 (s, 9H; 3 × CH3), 0.00 (s, 6H; 2
× CH3). 13C NMR (126 MHz, CDCl3) δ 138.1 (CBn‑ipso), 138.0
(CBn‑ipso), 134.7 (CPhS‑ipso), 131.6 (CPh), 129.1 (CPh), 128.7 (CPh),
128.5 (CPh), 128.1 (CPh), 128.0 (CPh), 127.9 (CPh), 127.5 (CPh), 86.0
695.2438, found 695.2430; [α]RDT +67 (c 0.2, CHCl3); H NMR (500
1
2200
dx.doi.org/10.1021/jo302455d | J. Org. Chem. 2013, 78, 2191−2205

The Journal of Organic Chemistry
Featured Article
MHz, CDCl3) δ 7.82 (dd, J = 8.3, 1.4 Hz, 2H; Phortho), 7.47 (m, 1H;
Ph), 7.36 (m, 2H; Phortho), 7.31−7.16 (m, 15H; Ph), 7.13 (dd, J = 6.4,
2.1 Hz, 3H; 2 × Ph, H-6), 7.07 (m, 1H; H-7), 6.71 (m, 2H; Phortho),
5.50 (d, J1,2 = 1.8 Hz, 1H; H-1), 4.76 (d, 2J = 10.4 Hz, 1H;
CH(Ha)Ph), 4.72 (m, 1H; H-5), 4.65−4.54 (m, 4H; 2 × CH2Bn),
2H; H-6); 13C NMR (126 MHz, CDCl3) δ 158.0 (CPh‑ipso), 148.7 (C-
7), 138.6 (CBn‑ipso), 137.8 (CBn‑ipso), 135.5 (CPh‑ipso), 134.5 (CSPh‑ipso),
132.1 (CPh‑ipso), 131.0 (CPh), 129.3 (CPh), 129.1 (CPh), 129.0 (CPh),
128.5 (CPh), 128.4 (CPh), 128.3 (CPh), 128.2 (CPh), 128.1 (CPh), 128.0
(CPh), 127.8 (CPh), 127.6 (CPh), 127.5 (CPh), 127.3 (CPh), 113.9
(CPh), 107.4 (C-8), 86.5 (C-1), 77.6 (C-2), 76.9 (C-3), 76.8 (C-4),
73.0 (CBn), 72.7 (CBn), 67.3 (C-5), 55.3 (CMeO), 37.5 (CBn), 31.2 (C-
6).
2
4.45 (d, J = 10.4 Hz, 1H; CH(Hb)Ph), 3.94 (dd, J2,3 = 3.0, J1,2 = 1.8
Hz, 1H; H-2), 3.83 (dd, J2,3 = 3.0, J3,4 = 1.9 Hz, 1H; H-3), 3.76 (t, J4,5
=
9.4 Hz 1H; H-4), 3.65 (s, 3H; MeO); 13C NMR (126 MHz, CDCl3) δ
190.4 (CO), 159.4 (CPh‑ipso), 143.9 (C-6), 138.1 (CBn‑ipso), 137.9
(CBn‑ipso), 137.7 (CBn‑ipso), 134.0 (CSPh‑ipso), 132.9 (CPh), 131.8 (CPh),
130.0 (CPh‑ipso), 129.9 (CPh), 129.1 (CPh), 128.7 (CPh), 128.6 (CPh),
128.5 (CPh), 128.5 (CPh), 128.0 (CPh), 127.89 (CPh), 127.87 (CPh),
127.85 (CPh), 127.7 (CPh), 125.8 (C-7), 113.9 (CPh), 86.0 (C-1), 80.1
(C-3), 78.1 (C-4), 76.6 (C-2), 75.3 (CBn), 72.5 (CBn), 72.4 (CBn), 72.2
(C-5), 55.3 (CMeO).
Phenylthio 2,3-Di-O-benzyl-6,7-dideoxy-8-C-phenyl-α-D-
manno-oct-8-ulopyranoside (12). Procedure 2 (DDQ deprotec-
tion): To a solution of phenylthio 2,3-di-O-benzyl-4-O-(p-methox-
ybenzyl)-6,7-dideoxy-8-C-phenyl-α-D-manno-oct-8-ulopyranoside
(2.00 g, 2.96 mmol) dissolved in CH2Cl2/H2O (100 mL, 20:1) was
added DDQ (0.81 g, 3.55 mmol, 1.2 equiv), and the reaction was
stirred for 2 h. Then, the reaction was quenched by addition of satd
NaHCO3, and the mixture was extracted with CH2Cl2, washed with
satd NaHCO3 and brine, dried (MgSO4), and concentrated. The crude
was purified by dry column chromatography (heptane with a 2.5%
EtOAc gradient). This gave the product as a colorless syrup in 89%
(1.46 g): Rf 0.26 (heptane/EtOAc 3:1); HRMS (ESI) m/z calcd for
C34H34O5SNa+ 577.2025, found 577.2034; [α]DRT +8 (c 0.4, CHCl3);
1H NMR (500 MHz, CDCl3) δ 7.86−7.74 (m, 2H; Ph), 7.50−7.44
Phenylthio 2,3-Di-O-benzyl-4-O-(p-methoxybenzyl)-6,7-di-
deoxy-8-C-phenyl-α-D-manno-oct-8-ulopyranoside (11c).
Schlenk techniques were used for the 1,4-reduction. Freshly made
Stryker’s reagent (7.38 g, 3.76 mmol, 0.82 equiv) was dissolved in dry
and degassed benzene (5 mL, dried by passing through basic Al2O3
and degassed by three freeze thaw cycles, refilling with argon). Then
(E)-phenylthio 2,3-di-O-benzyl-4-O-(p-methoxybenzyl)-6,7-dideoxy-8-
C-phenyl-α-D-manno-oct-6-eno-8-ulopyranoside (9c) (3.10 g, 4.61
mmol) dissolved in dry and degassed benzene (35 mL, see above)
was added by cannulation along with degassed water (207 mL, 11.52
mol, 2.5 equiv). The resulting dark red solution was stirred under
argon until completion as indicated by NMR of the crude reaction
mixture (30 min). Afterward, the system was opened, and air was
bubbled through the solution by which copper-containing degradation
products precipitated. The solution was filtered through a bed of
Celite and washed with Et2O followed by concentration in vacuo. The
crude was purified by dry column chromatography (heptane with a
0.91% gradient of EtOAc) to give the product 11c as a colorless syrup
in 99% (3.10 g): Rf 0.3 (heptane/EtOAc 5:1); HRMS (ESI) m/z calcd
for C42H42O6SNa+ 697.2594, found 697.2579; [α]RDT +93 (c 0.2,
CHCl3); 1H NMR (500 MHz, CDCl3) δ 7.87 (dd, J = 8.4, 1.4 Hz, 2H;
Phortho), 7.52 (m, 1H; Ph), 7.47−7.28 (m, 19H; Ph), 6.89 (m, 2H;
(m, 1H; Ph), 7.36−7.16 (m, 17H; Ph), 5.57 (d, J1,2 = 1.0 Hz, 1H; H-
1), 4.52 (m, 4H; 2 × CH2Ph), 4.00 (td, J = 9.0, J = 2.7 Hz, 1H; H-5),
3.94 (dd, J2,3 = 3.0, J1,2 = 1.4 Hz, 1H; H-2), 3.83 (t, J3,4 = 9.4, J4,OH
1.9 Hz, 1H; H-4), 3.64 (dd, J3,4 = 9.4, J2,3 = 3.1 Hz, 1H; H-3), 3.00
(ddd, J7a,7b = 17.9, J = 8.8, J = 5.4 Hz, 1H; H-7a), 2.80 (ddd, J7a,7b
=
=
17.9, J = 8.5, J = 6.3 Hz, 1H; H-7a), 2.50 (d, J4,OH = 2.3 Hz, 1H; 4-
OH), 2.34−2.23 (m, 1H; H-6a), 1.92 (dtd, J = 14.2, J = 8.6, J = 5.5 Hz,
1H; H-6b); 13C NMR (126 MHz, CDCl3) δ 199.9 (Ph−CO), 137.8
(2 × CBn‑ipso), 136.9 (CPh‑ipso), 134.0 (CPh‑ipso), 132.9 (CPh), 131.5
(CPh), 129.1 (CPh), 128.6 (CPh), 128.5 (CPh), 128.1 (CPh), 128.0
(CPh), 127.9 (CPh), 127.4 (CPh), 85.2 (C-1), 79.7 (C-3), 75.5 (C-2),
72.4 (C-5), 72.2 (Cbenzyl), 71.8 (Cbenzyl), 70.2 (C-4), 34.1 (C-7), 25.4
(C-6).
Phenylthio 4,8-Anhydro-2,3-di-O-benzyl-6,7-dideoxy-8-me-
thoxy-8R-C-phenyl-α-D-manno-octopyranoside (14) and (Z)-
Phenylthio 4,8-Anhydro-2,3-di-O-benzyl-6,7-dideoxy-8-C-phe-
nyl-α-D-manno-oct-7-enopyranoside (15). To a solution of
phenylthio 2,3-di-O-benzyl-6,7-dideoxy-8-C-phenyl-α-D-manno-oct-8-
ulopyranoside (12) (460 mg, 0.83 mmol) dissolved in dry MeOH/
THF (5:1, 15 mL) was added freshly activated Amberlite IR-120H.
The reaction mixture was stirred under N2 at rt for 48 h. When TLC
(heptane/EtOAc 5:1) indicated 90% conversion the resin was
removed and washed several times with MeOH. The solution was
afterward neutralized with a few drops of Et3N and concentrated. The
crude was purified by dry column chromatography (heptane with a
1.0% gradient of EtOAc) to give two products in an overall yield of
71% (333 mg). The starting material 12 could be recovered in 28%
yield (129 mg). (Z)-Phenylthio 4,8-anhydro-2,3-di-O-benzyl-6,7-
dideoxy-8-C-phenyl-α-D-manno-oct-7-enopyranoside (15): Rf 0.31
(heptane/EtOAc 5:1); HRMS (ESI) m/z calcd for C34H34O4S+
537.2100, found 537.2109; [α]RDT +13 (c 0.3, CHCl3); 1H NMR
(500 MHz, CDCl3) δ 7.49 (m, 2H; Ph), 7.37−7.15 (m, 18H; Ph), 5.49
(d, J1,2 = 1.6 Hz, 1H; H-1), 5.25 (dd, J = 5.0, J = 3.1 Hz, 1H; H-7),
4.84 (d, J = 12.0 Hz, 1H; CH(Ha)-PH), 4.66 (m, 3H; Bn), 4.31 (td, J
= 9.3, J = 6.7 Hz, 1H; H-5), 4.22 (t, J3,4 = 9.6 Hz, 1H; H-4), 3.99 (dd,
J2,3 = 3.1, J1,2 = 1.6 Hz, 1H; H-2), 3.93 (dd, J3,4 = 9.6, J2,3 = 3.2 Hz, 1H;
H-3), 2.37 (m, 2H; H-6). 13C NMR (126 MHz, CDCl3) δ 151.6 (C-
8), 138.5 (CPh‑ipso), 137.9 (CPh‑ipso), 135.2 (CPh‑ipso), 134.3 (CPh‑ipso),
131.4−124.8 (CPh‑H), 95.3 (C-7), 86.6 (C-1), 77.6 (C-2),77.1 (C-3),
76.7 (C-4), 73.1 (CBn), 72.8 (CBn), 67.0 (C-5), 27.6 (C-6). Phenylthio
4,8-anhydro-2,3-di-O-benzyl-6,7-dideoxy-8-methoxy-8R-C-phenyl-α-D-
manno-octopyranoside (14): Rf 0.24 (heptane/EtOAc 5:1); HRMS
(ESI) m/z calcd for C35H36O5NaS+ 591.2181, found 591.2174; [α]RDT
+45 (c 0.3, CHCl3); 1H NMR (500 MHz, CDCl3) δ 7.44 (dd, J = 8.3, J
= 1.4 Hz, 2H), 7.32−7.18 (m, 18H), 5.46 (d, J1,2 = 1.5 Hz, 1H; H-1),
4.83 (d, J = 12.2 Hz, 1H; CH(Ha)-Ph), 4.72 (m, 1H; CH(Ha)-Ph),
4.68 (m, 2H; 2 x CH(Hb)-Ph), 4.18 (t, J = 9.8 Hz, 1H; H-4), 4.00 (dd,
J = 3.3, J1,2 = 1.5 Hz, 1H; H-2), 3.95 (m, 1H; H-5), 3.84 (dd, J = 9.9,
2
Phortho), 5.66 (d, J1,2 = 1.7 Hz, 1H; H-1), 4.94 (d, J = 10.4 Hz, 1H;
CH(Ha)Ph), 4.81−4.73 (m, 2H; CH2Ph), 4.73−4.66 (m, 3H; CH2Ph,
CH(Hb)Ph), 4.12 (td, J4,5 = 9.3, J5,6 = 2.8 Hz, 1H; H-5), 4.04 (dd, J2,3
= 3.0, J1,2 = 1.8 Hz, 1H; H-2), 3.96 (dd, J = 9.1, J2,3 = 3.1 Hz, 1H; H-3),
3.84 (t, J4,5 = 9.3 Hz, 1H; H-4), 3.79 (s, 3H; MeO), 2.98 (m, 1H; H-
7a), 2.83 (ddd, J = 17.8, 9.8, 5.5 Hz, 1H; H-7b), 2.42 (m, 1H; H-6b),
1.99 (dtd, J = 14.2, 9.6, 4.8 Hz, 1H; H-6a); 13C NMR (126 MHz,
CDCl3) δ 199.6 (CO), 159.3 (CPh‑ipso), 138.3 (CBn‑ipso), 138.0
(CBn‑ipso), 136.9 (CBn‑ipso), 134.40 (CPh), 134.38 (CPh), 134.1
(CSPh‑ipso), 133.9 (CPh), 133.7 (CPh), 132.9 (CPh), 131.5 (CPh), 130.5
(CPh‑ipso), 130.2 (CPh), 129.9 (CPh), 129.1 (CPh), 128.8 (CPh), 128.6
(CPh), 128.51 (CPh), 128.49 (CPh), 128.46 (CPh), 128.45 (CPh), 128.04
(CPh), 127.99 (CPh), 127.87 (CPh), 127.84 (CPh), 127.80 (CPh), 127.78
(CPh), 127.3 (CPh), 113.9 (CPh), 85.1 (C-1), 80.3 (C-3), 78.2 (C-4),
76.4 (C-2), 75.0 (CBn), 72.32 (CBn), 72.28 (CBn), 72.1 (C-5), 55.3
(CMeO), 34.6 (C-7), 25.7 (C-6).
Phenylthio 4,7-Anhydro-2,3-di-O-benzyl-6-deoxy-7,8-ene-8-
(p-methoxybenzyl)-8-phenyl-α-D-manno-octopyranoside (19).
Phenylthio 2,3-di-O-benzyl-4-O-(p-methoxybenzyl)-6,7-dideoxy-8-C-
phenyl-α-D-manno-oct-8-ulopyranoside (11c) (103 mg, 0.15 mmol)
was dissolved in a 10% TFA solution in CH2Cl2 (5 mL). The reaction
was completed after 5 min (estimated from TLC). The mixture was
concentrated in vacuo and purified by dry column chromatography
giving the entitled compound 19 as a colorless syrup in 50% (49 mg):
HRMS (ESI) m/z calcd for C42H40O5SNa+ 679.2494, found 679.5109;
[α]RDT +35 (c 0.2, CHCl3); H NMR (500 MHz, CDCl3) δ 7.36 (m,
1
2H; Ph), 7.44−7.27 (m, 18H; Ph), 7.10 (dd, J = 6.5, 2.1 Hz, 2H; Ph),
2
6.86 (m, 2H; Ph), 5.56 (d, J1,2 = 1.6 Hz, 1H; H-1), 4.87 (d, J = 12.0
2
Hz, 1H; CH(Ha)Ph), 4.81−4.71 (q, J = 12.4 Hz, 2H; CH2Ph), 4.69
(d, 2J = 12.0 Hz, 1H; CH(Hb)Ph), 4.44 (td, J4,5 = 9.5, J = 6.7 Hz, 1H;
H-5), 4.29 (t, J = 9.7 Hz, 1H; H-4), 4.08 (m, 1H; H-2), 3.97 (dd, J3,4
=
2
9.6, J = 3.3 Hz, 1H; H-3), 3.82 (s, 3H; MeO), 3.54 (d, J = 15.4 Hz,
2
1H; CH(Hb)Ph), 3.37 (d, J = 15.4 Hz, 1H; CH(Ha)Ph), 2.28 (m,
2201
dx.doi.org/10.1021/jo302455d | J. Org. Chem. 2013, 78, 2191−2205

The Journal of Organic Chemistry
Featured Article
3.3 Hz, 1H; H-3), 3.03 (s, 3H; MeO), 2.18−2.05 (m, 2H; H-6a, H-7a),
1.76 (m, 1H; H-6b), 1.63 (m, 1H; H-7b). 13C NMR (126 MHz,
CDCl3) δ 142.0 (CPh‑ipso),, 138.9 (CBn‑ipso), 138.1 (CBn‑ipso), 134.6
(CPh‑ipso), 131.1−124.8 (20 × Ph-H), 99.7 (C-8 benzylidene), 87.1 (C-1),
78.0 (C-2), 76.9 (C-3), 73.1 (Cbenzyl), 72.9 (Cbenzyl), 72.0 (C-4), 69.4
(C-5), 49.2 (CMeO), 37.6 (C-7), 25.4 (C-6).
Tf2O (2 equiv) was added and the mixture stirred for 30 min (−60
°C) after which the acceptor in 0.5 mL of dichloromethane (1.5 equiv,
0.1 M with respect to donor) was added. After the reaction was
complete as judged by TLC (10 min −3 h, heptane/EtOAc 5:1), the
temperature was raised to −40 °C and several drops of Et3N were
added. Then, the mixture was filtered through Celite, concentrated,
and purified by dry column chromatography on silica gel giving the α/
β products.
1-Adamantanyl 2,3-Di-O-benzyl-4,6-O-benzylidene-D-man-
nopyranoside (24).53 Method A: β/α 1:5.7. 91% yield (98 mg,
starting with 100 mg). Method B: β/α 6.7:1. 75% yield (80 mg,
starting with 100 mg). Method C: β/α 7.0:1. 83% yield (89 mg,
starting with 100 mg). Method D: β/α 6.1:1. Quantitative yield: 107
mg, starting with 100 mg. HRMS (ESI): m/z calcd for C37H42O6Na+
605.287, found 605.2876
1-Adamantanyl 4,8-anhydro-2,3-di-O-benzyl-6,7-dideoxy-8-
C-phenyl-D-glycero-α-D-manno-octopyranoside (25α): HRMS
(ESI) m/z calcd for C38H44O5Na+ 603.3086, found 603.3098; 1H
NMR (500 MHz, CDCl3) δ 7.56 (m, 1H; Ph), 7.40−7.27 (m, 14H;
Ph), 5.13 (d, J1,2 = 1.8 Hz, 1H; H-1), 4.84 (t, J = 12.4 Hz, 2H;
CH2Ph), 4.68 (m, 2H; CH2Ph), 4.53 (dd, J = 11.2, J = 2.3 Hz, 1H; H-
8), 3.92 (m, 1H; H-3), 3.88 (m, 1H; H-4), 3.70 (ddd, J = 11.2, J = 9.0,
J = 4.2 Hz, 1H; H-5), 3.62 (dd, J = 3.0, J1,2 = 1.8 Hz, 1H; H-2), 2.10
(dd, J = 6.4, J = 3.2 Hz, 3H; Hadamantanyl), 2.03 (m, 2H; H-6a, H-7a),
1.78 (m, 2H; H-6b, H-7b), 1.69 (d, J = 2.8 Hz, 6H; Hadamantanyl), 1.59
(m, 6H; Hadamantanyl).
Phenylthio 4,8-Anhydro-2,3-di-O-benzyl-6,7-dideoxy-8-phe-
nyl-D-glycero-α-D-manno-octopyranoside (1). A solution of acetal
14 and alkene 15 (1.21 g, 2.13 mmol, from acetal) was dissolved in dry
CH2Cl2 (20 mL), kept under nitrogen, and cooled to 0 °C. Then,
trifluoroacetic anhydride (0.60 mL, 4.26 mmol, 2 equiv) and
triethylsilane (1.70 mL, 10.60 mmol, 5 equiv) were added, and the
resulting solution was stirred for 5 min. Afterward, trifluoroacetic acid
(0.81 mL, 10.60 mmol, 5 equiv) was added dropwise over a period of
2 min. Then the ice bath was removed, and the reaction mixture was
left until completion (4 h, judged by TLC). The mixture was diluted
with CH2Cl2, washed with saturated aqueous NaHCO3 and water,
dried (MgSO4), and concentrated. The crude was purified by dry
column chromatography (heptane with a 1.0% gradient of EtOAc) to
give the product 1 as a colorless syrup in 97% (1.15 g): Rf 0.4
(heptane/EtOAc 5:1); HRMS (ESI) m/z calcd for C34H34O4SNa+
561.2076, found 561.2090; [α]RDT +101 (c 0.1, CHCl3); 1H NMR (500
MHz, CDCl3) δ 7.47−7.29 (m, 20H; Ph), 5.60 (d, J1,2 = 1.4 Hz, 1H:
H-1), 4.87 (d, J = 12.4 Hz, 1H; C(Ha)HPh), 4.79 (q, J = 12.3 Hz, 2H;
CH2Ph), 4.73 (d, J = 12.4 Hz, 1H; C(Hb)Ph), 4.61 (dd, J7,8 = 2.0 Hz,
J7,8 = 11.7 Hz, 1H; H-8axial), 4.10 (dd, J1,2 = 1.4 Hz, J2,3 = 3.1 Hz, 1H;
H-2), 4.07 (m, 1H; H-5), 4.05 (m, 1H; H-4), 3.94 (dd, J2,3 = 3.1 Hz,
1H; H-3), 2.15 (dt, J = 9.0, J = 2.6 Hz, 2H; H-6equatorial, H-7equatorial),
1.95 (m, 1H; H-6axial), 1.79 (m, 1H; H-7axial). 13C NMR (126 MHz,
CDCl3) δ 142.4 (CPh‑ipso), 138.8 (C3‑Bn ipso), 138.0 (C2‑Bn ipso), 134.7
(CSPh‑ipso), 131.2−125.8 (20 CPh), 86.90 (C-1), 79.44 (C-4), 79.19 (C-
8), 78.13 (C-2), 77.00 (C-3), 73.05 (C3‑Bn), 72.84 (C2‑Bn), 69.83 (C-
5), 33.52 (C-7), 29.31 (C-6).
1-Adamantanyl 4,8-anhydro-2,3-di-O-benzyl-6,7-dideoxy-8-
C-phenyl-D-glycero-β-D-manno-octopyranoside (25β): [α]DRT
−11 (c 4.2, CHCl3); HRMS (ESI) m/z calcd for C38H44O5Na+
603.3081, found 603.3087. Spectra were in accordance with the
literature:53 1H NMR (500 MHz, CDCl3) δ 7.43−7.40 (m, 2H; Ph),
7.28−7.15 (m, 13H; Ph), 4.93 (d, 2J = 12.7 Hz, 1H; CH(Ha)-Ph), 4.85
2
(d, J = 12.7 Hz, 1H; CH(Hb)-Ph), 4.60 (d, J1,2 = 1.1 Hz, 1H; H-1),
2
2
4.53 (d, J = 12.8 Hz, 1H; CH(Ha)-Ph), 4.49 (d, J = 12.8 Hz, 1H;
CH(Hb)-Ph), 4.43 (dd, J = 11.5, J = 2.3 Hz, 1H; H-8), 3.77 (t, J = 9.4
Hz, 1H; H-4), 3.64 (dd, J2,3 = 3.2, J1,2 = 1.0 Hz, 1H; H-2), 3.39 (dd, J3,4
= 9.8, J2,3 = 3.2 Hz, 1H; H-3), 2.98 (ddd, J = 11.2, J = 9.0, J = 4.2 Hz,
1H; H-5), 2.06 (m, 3H; Cadamantanyl), 1.98 (m, 2H; H-6, H-7), 1.83
(tdd, J = 12.7, J = 11.3, J = 4.0 Hz, 1H; H-7), 1.72 (m, 6H; Cadamantanyl),
1.53 (m, 7H; 6 × Cadamantanyl, C-7); 13C NMR (126 MHz, CDCl3) δ
142.6 (CPh‑ipso), 138.9 (CBn‑ipso), 138.8 (CBn‑ipso), 128.8−125.6 (15 ×
CPh), 94.5 (C-1), 79.1−78.8 (three peaks; C-3, C-4, C-8), 76.6 (C-2),
74.8 (Cquart‑adamantanyl next to O), 74.3 (CBn), 72.0 (CBn), 71.6 (C-5), 42.5
(Csec‑adamantanyl), 36.3 (Csec‑ adamantanyl), 33.3 (C-7), 30.6 (Ctert‑adamantanyl),
29.5 (C-6).
Methyl (4,8-Anhydro-2,3-di-O-benzyl-6,7-dideoxy-8-C-phe-
nyl-D-glycero-D-manno-octopyranosyl)-(1→3)-1,2:5,6-di-O-iso-
propylidene-α-D-glucofuranoside (26α). The product was purified
twice by dry column chromatography, but it was not possible to
remove unreacted BSP (ratio 1:0.8) which can be seen in the spectra.
The peaks from BSP are in accordance with previous published data.54
Because of the multiple purifications, heptane is present in the spectra:
HRMS (ESI) m/z calcd for C40H48O10Na+ 711.3145, found 711.3153;
1H NMR (500 MHz, CDCl3) δ 7.69−7.63 (dd, J = 8.0, 1.6 Hz, 2H;
Ph), 7.37−7.28 (m, 13H; Ph), 5.85−5.82 (d, J = 3.5 Hz, 1H; H-1′),
5.20−5.17 (d, J1,2 = 1.3 Hz, 1H; H-1), 4.79−4.71 (m, 3H; 3 ×
CH(H)Ph), 4.61−4.58 (m, 1H; CH(H)Ph), 4.57−4.53 (m, 2H; H-2′,
H-8), 4.29−4.27 (d, J = 2.0 Hz, 1H; H′), 4.10−4.07 (m, 1H; H-6a′),
4.05−4.03 (dd, J = 3.6, 2.3 Hz, 2H; H′, H-6b′), 4.03−4.00 (m, 1H;
H′), 3.95−3.91 (t, J = 9.3 Hz, 1H; H-4), 3.82−3.78 (m, 2H; H-2, H-
3), 3.58−3.53 (m, 1H; H-5), 2.16−2.10 (ddd, J = 16.2, 10.8, 3.3 Hz,
2H; H-6a, H-7a), 1.90−1.85 (m, 1H; H-6a), 1.76−1.72 (m, 1H, H-7a),
1.50−1.49 (s, 3H; CH3C), 1.41−1.40 (s, 3H; CH3C), 1.34−1.33 (s,
3H; CH3C), 1.32−1.31 (s, 3H; CH3C). BSP: 7.52−7.47 (m, 3H; Ph),
7.37−7.28 (m, 2H; Ph), 3.14−3.10 (dt, J = 7.6, 3.2 Hz, 1H ;), 2.98−
2.94 (m, 1H), 1.63−1.58 (m, 6H; 3 x CH2).13C NMR (126 MHz,
CDCl3) δ 143.5 (CPh ipso BSP), 142.3 (CPh‑ipso), 139.0 (CPh‑ipso), 138.3
(CPh‑ipso), 130.8 (CPh BSP), 128.9 (CPh BSP), 128.5, 128.4, 128.3, 128.3,
128.1, 127.8, 127.6, 127.5, 127.4, 126.4 (CPh BSP), 125.8, 112.2 (Cacetal),
109.5 (Cacetal), 105.4 (C-1′), 99.8 (C-1), 84.2 (C-2′), 81.6 (C′), 80.1
Glycosylation Procedures. Method A: NIS/TfOH Method.
Mannosyl donor (100 mg, 185 μmol, 1 equiv) and acceptor (1.1
equiv) were coevaporated twice with toluene. Afterward, the donor
and acceptor were dried for 30 min at a Schlenk line along with 3 Å
molecular sieves. To the mixture of donor, acceptor, and 3 Å molecular
sieves was added 2 mL of dry dichloromethane (0.1 M with respect to
donor) at 0 °C under argon, and the reaction mixture was stirred for 1
h. Then, NIS (2.5 equiv) was added followed by a catalytic amount of
TfOH (1 drop). The reaction mixture was stirred until completion, as
judged by TLC (ca. 5−10 min, heptane/EtOAc 3:1), after which the
solution was diluted with dichloromethane and filtered through Celite.
Afterward, the filtrate was washed with 10% aqueous Na2S2O3,
saturated aqueous NaHCO3, and brine, dried (MgSO4), filtered, and
concentrated. The crude reaction mixture was purified by dry column
chromatography on silica gel.
Methods B and C: Ph2SO/TTBP/Tf2O Method. Mannosyl donor
(100 mg, 185 μmol, 1 equiv), TTBP (1.1 equiv), and Ph2SO (1 equiv)
were coevaporated twice with toluene. Afterward, the donor, TTBP,
and Ph2SO were dried for 30 min on a Schlenk line along with either 3
Å or crushed 4 Å molecular sieves. To the mixture of donor, TTBP,
Ph2SO, and molecular sieves was added 1 mL of dry dichloromethane
at −78 °C under argon, and the reaction mixture was stirred for 1 h.
Afterward, Tf2O (2 equiv) was added and the mixture stirred for 30
min (−78 °C) after which the acceptor in 0.5 mL of dichloromethane
(1.5 equiv, 0.1 M with respect to donor) was added. After the reaction
was complete as judged by TLC (10 min −2 h, heptane/EtOAc 3:1),
the temperature was raised to −40 °C, and several drops of Et3N were
added. Then, the mixture was filtered through Celite, concentrated,
and purified by dry column chromatography on silica gel to give the α/
β products.
Method D: BSP/TTBP/Tf2O Method. Mannosyl donor (100 mg, 185
μmol, 1 equiv), TTBP (1.1 equiv), and 1-benzenesulfinyl piperidine
(BSP, 1 equiv) were coevaporated twice with toluene. Afterward, the
donor, TTBP, and BSP were dried for 30 min on a Schlenk line along
with 3 Å molecular sieves. To the mixture of donor, TTBP, BSP, and
molecular sieves was added 1 mL of dry dichloromethane at −60 °C
under argon, and the reaction mixture was stirred for 1 h. Afterward,
2202
dx.doi.org/10.1021/jo302455d | J. Org. Chem. 2013, 78, 2191−2205

The Journal of Organic Chemistry
Featured Article
(C′), 79.4 (C-4), 79.2 (C-8), 76.7 (C-2). 75.8 (C-3), 73.2 (C′), 73.1
(CBn), 72.6 (CBn), 69.4 (C-5), 67.9 (C-6′), 47.1 (Csec. BSP), 33.3 (C-7),
29.5 (C-6), 27.1(Ctert), 26.5 (Ctert), 26.1 (Csec‑BSP), 25.7 (Ctert), 24.1
(Csec‑BSP).
itself. It was possible to estimate the α/β mixture from 1H NMR after
partial purification (β/α ≈2:1). Furthermore, spectra of the α-anomer
were also obtained but were contaminated with BSP in 1:3 (α-
anomer/BSP). The β-anomer was contaminated with the α-anomer
and acceptor.
Methyl (4,8-anhydro-2,3-di-O-benzyl-6,7-dideoxy-8-C-phe-
nyl-D-glycero-β-D-manno-octopyranosyl)-(1→3)-1,2:5,6-di-O-
isopropyliden-β-D-glucofuranoside (26β): [α]DRT −50 (c 2.2,
CH2Cl2); HRMS (ESI) m/z calcd for C40H48O10Na+ 711.3145,
found 711.3163; 1H NMR (500 MHz, CDCl3) δ 7.36 (dd, J = 8.1, J =
1.4 Hz, 2H; Ph), 7.29−7.21 (m, 13H; Ph), 5.83 (d, J1′,2′ = 3.7 Hz, 1H:
H-1′), 4.82−4.73 (m, 2H; CH2Ph), 4.65−4.54 (m, 2H; CH2Ph), 4.46
(dd, J = 11.5, J = 2.3 Hz, 1H; H-8), 4.43 (d, J1,2 = 1.0 Hz, 1H; H-1),
4.40 (d, J4′,5′ = 5.0 Hz, 1H; H-5′), 4.33 (d, J1′,2′ = 3.8 Hz, 1H; H-2′),
4.26 (dd, J4′,5′ = 4.9, J3′,4′ = 3.1 Hz, 1H; H-4′), 4.23 (d, J3′,4′ = 3.1 Hz,
1H; H-3′), 4.07 (dd, J = 8.5, J = 6.6 Hz, 1H; H-6′a), 4.01−3.97 (m,
28α: Rf 0.50 (heptane/EtOAc 5:2); HRMS (ESI) m/z calcd for
C56H60O10Na+ 915.4079, found 915.4060; 1H NMR (500 MHz,
CDCl3) δ 7.66 (m, 2H), 7.40−7.20 (m, 26H), 7.15 (m, 2H), 5.23 (d,
J1,2 = 1.8 Hz, 1H; H-1), 5.07 (d, J = 11.5 Hz, 1H; CH(H)Ph), 4.79 (m,
1H; CH(H)Ph), 4.68 (m, 1H; CH(H)Ph), 4.63−4.56 (m, 5H; H-1′, 2
× CH2Bn), 4.51−4.47 (m, 2H; H-8, CH(H)Ph), 4.44 (d, J = 12.0 Hz,
1H; CH(H)Ph), 4.23 (d, J = 11.9 Hz, 1H; CH(H)Ph), 3.88 (m, 1H;
H-4), 3.86−3.82 (m, 2H; H-3′, H-4′), 3.76 (dd, J2,3 = 2.9 Hz, J1,2 = 1.8
Hz, 1H; H-2), 3.75−3.69 (m, 4H; H-3, H-5′, 2 × H-6′), 3.58 (m, 1H;
H-5), 3.54 (dd, J2′,3′ = 9.6 Hz, J1′,2′ = 3.5 Hz, 1H; H-2′), 3.40 (s, 3H;
CH3O), 1.99 (ddd, J = 13.3 Hz, J = 5.6 Hz, J = 3.0 Hz, 2H; H-7eq),
1.90 (dq, J = 11.1 Hz, J = 3.6 Hz, 2H; H-6eq), 1.77 (ddd, J = 24.3 Hz, J
= 12.5 Hz, J = 4.1 Hz, 2H; H-6ax), 1.67−1.46 (m, 1H; H-7ax). BSP:
7.55−7.45 (m, 3H), 7.40−7.20 (m, 12H), 3.15−3.08 (m, 3H), 3.02−
2.92 (m, 3H), 1.67−1.46 (m, 9H). 13C NMR (126 MHz, CDCl3) δ
143.5 (CPh ipso BSP), 142.6 (CPh‑ipso), 139.2 (CBn‑ipso), 138.9 (CBn‑ipso),
138.7 (CBn‑ipso), 138.3 (CBn‑ipso), 138.00 (CBn‑ipso), 130.79 (CPh BSP),
128.9 (CPh BSP), 128.6 (CPh), 128.60 (CPh), 128.56 (CPh), 128.50
(CPh), 128.47 (CPh), 128.41 (CPh), 128.35 (CPh), 128.31 (CPh), 128.25
(CPh), 128.2 (CPh), 128.1 (CPh), 127.68 (CPh), 127.67 (CPh), 127.64
(CPh), 127.6 (CPh), 127.5 (CPh), 127.4 (CPh), 127.3 (CPh), 127.0
(CPh), 126.4 (CPh), 125.8 (CPh), 101.6 (C-1, JC,H = 169 Hz), 97.9 (C-
1′, JC,H = 170 Hz), 81.6, 80.1 (C-2′), 79.4 (C-8), 79.1, 78.0, 77.7, 75.2
(CBn), 73.7 (CBn), 73.4 (CBn), 73.3 (CBn), 73.0 (CBn), 69.9, 69.4 (C-
6′), 55.4 (CMeO), 47.1 (Csec. BSP), 33.6 (C-7), 29.9 (C-6), 26.3
(Csec. BSP), 24.1 (Csec. BSP).
Methyl 2,3,6-Tri-O-benzyl-4-O-trifluoromethanesulfonyl-α-
D-glucopyranoside: Rf 0.55 (heptane/EtOAc 5:2); HRMS (ESI)
m/z calcd for C29H31O8SF3Na+ 619.1584, found 619.1586; [α]RDT 21 (c
2.0, CHCl3); 1H NMR (500 MHz, CDCl3) δ 7.32−7.18 (m, 15H; Ph),
4.95 (t, J = 9.6 Hz, 1H; H-4), 4.87 (d, J = 10.2 Hz, 1H; CH(Ha)Ph),
4.76 (d, J = 10.2 Hz, 1H CH(Hb)Ph), 4.67 (d, J = 12.0 Hz, 1H
CH(Ha)Ph), 4.50 (m, 2H; CH2Ph), 4.47 (s, 1H; H-1), 4.40 (d, J =
12.0 Hz, 1H CH(Hb)Ph), 3.98 (t, J = 9.4 Hz, 1H; H-3), 3.89 (ddd, J =
10.0, 3.6, 2.1 Hz, 1H; H-5), 3.60 (m, 2H; 2 x H-6), 3.52 (dd, J = 9.6,
3.5 Hz, 1H; H-2), 3.31(s, 3H; CH3O); 13C NMR (126 MHz, CDCl3)
δ 137.8 (CBn), 137.6 (CBn), 137.5 (CBn), 128.7 (CPh), 128.5 (CPh),
128.4 (CPh), 128.3 (CPh), 128.1 (CPh), 128.0 (CPh), 127.9 (CPh), 127.8
(CPh), 122.3, 119.8, 117.2, 114.7 (q, F3CSO2R), 98.0 (C-1), 81.5 (C-
4), 80.2 (C-2), 78.1 (C-3), 75.6 (CBn), 73.8 (CBn), 73.8 (CBn), 67.9
(C-5), 67.7 (C-6), 55.9 (CMeO).
1H; H-6′b), 3.83−3.78 (m, 1H; H-4), 3.76−3.75 (dd, J2,3 = 3.1, J1,2
=
1.0 Hz, 1H; H-2), 3.45−3.40 (dd, J = 9.7, J2,3 = 3.1 Hz, 1H; H-3),
3.04−2.98 (ddd, J5,6 = 11.1, J = 9.1, J = 4.3 Hz, 1H; H-5), 2.08−1.98
(m, 2H; H-6a,H-7a), 1.88−1.81 (ddd, J = 12.8, J5,6 = 11.4, J = 8.7 Hz,
1H; H-6b), 1.62−1.52 (m, 1H; H-7b), 1.42 (s, 3H, CH3C), 1.37 (s,
3H; CH3C), 1.26 (s, 3H; CH3C), 1.24 (s, 3H; CH3C); 13C NMR (126
MHz, CDCl3) δ 142.4 (CPh‑ipso), 138.8 (CPh‑ipso), 138.5 (CPh‑ipso),
128.6−125.8 (15 × CPh), 112.0 (Cacetal), 108.7 (Cacetal), 105.1 (C-1′),
100.2 (C-1), 83.0 (C-2′), 80.9 (C-4′), 80.8 (C-3′), 79.2 (C-8), 78.9
(C-4), 78.9 (C-3), 76.2 (C-2), 74.8 (CBn), 74.8 (C-5), 73.4 (C-5′),
72.5 (CBn), 66.2 (C-6′), 33.2 (C-7), 29.9 (heptane), 29.2 (C-6), 26.9
(CMe), 26.7 (CMe), 26.5 (CMe), 25.5 (CMe).
Methyl (4,8-Anhydro-2,3-di-O-benzyl-6,7-dideoxy-8-C-phe-
nyl-D-glycero-β-D-manno-octopyranosyl)-(1→3)-2,3,4-tri-O-
acetyl-α/β-D-glucopyranoside (27). The crude was purified by dry
column chromatography (heptane with a 1.67% gradient of EtOAc) to
give a mixture of α/β in a 1:6.1 ratio with a overall yield of 78% (108
mg). Only a small amount of the β-anomer could be separated for
characterization.
27α. Rf 0.39 (heptane/EtOAc 2:1). The α-anomer could not be
separated from the β-anomer. In the mixture of both anomers the α-
anomer was found to resonate 1H NMR (500 MHz, CDCl3) at δ 4.80
(d, J1,2 = 1.60 Hz, 1H; H-1) and in 13C NMR (126 MHz, CDCl3) at δ
99.23 (C-1, JC1,H1 = 169.3 Hz).
27β: Rf 0.38 (heptane/EtOAc 2:1); HRMS (ESI) m/z calcd for
C41H48O13Na+ 771.2987, found 771.3002; [α]DRT +3 (c 0.1, CHCl3);
1H NMR (500 MHz, CDCl3) δ 7.46 (m, 2H; Ph), 7.36−7.26 (m, 13H;
Ph), 5.51 (dd, J = 10.3, J = 9.2 Hz, 1H; H-3′), 4.99−4.93 (m, 3H;
CH(Ha)Ph, H-1′, H-4′), 4.89−4.84 (m, 2H; CH(Hb)Ph, H-2′), 4.60
2
(q, J = 12.6, 2H; CH2Ph), 4.52 (dd, J7ax,8 = 11.4, J7eq,8 = 2.3 Hz, 1H;
H-8), 4.43 (d, J1,2 = 1.1 Hz, 1H; H-1), 4.07−4.00 (m, 2H; H-5′, H-
6a′), 3.96 (d, J2,3 = 3.2 Hz, 1H; H-2), 3.86 (t, J = 9.5 Hz, 1H; H-4),
3.51−3.45 (m, 2H; H-3, H-6b′), 3.38 (s, 3H; MeO), 3.09 (ddd, J =
11.0, J = 9.1, J = 4.2 Hz, 1H; H-5), 2.15 (dd, J = 12.1, 3.6 Hz, 1H; H-
6eq), 2.09 (s, 3H; CH3CO), 2.05 (dt, J = 4.2, J = 2.6 Hz, 1H; H-7eq),
2.03 (s, 3H; CH3CO), 2.02 (s, 3H; CH3CO), 1.95−1.86 (m, 1H; H-
6ax), 1.64 (m, 1H; H-7ax). 13C NMR (126 MHz, CDCl3) δ 170.4 (C
O), 170.2 (CO), 170.1 (CO), 142.5 (Cipso Ph), 138.8 (Cipso Ph),
128.6 (CPh), 128.4 (CPh), 128.28 (CPh), 128.27 (CPh), 127.7 (CPh),
127.6 (CPh), 127.5 (CPh), 127.4 (CPh), 125.8 (CPh), 102.3 (C-1, JC1,H1
= 156.6 Hz), 96.6 (C-1′, JC1′,H1′ = 175.1 Hz), 79.2 (C-8), 78.9 (C-4),
78.5 (C-3), 76.0 (C-2), 74.9 (CBn), 72.3 (CBn), 72.2 (C-5), 71.1 (C-
2′), 70.3 (C-3′), 69.4 (C-4′), 68.6 (C-6′), 68.6 (C-5′), 55.4 (CMeO),
33.3 (C-7), 29.3 (C-6), 20.9 (Cacetyl), 20.88 (Cacetyl), 20.85 (Cacetyl).
Methyl (4,8-Anhydro-2,3-di-O-benzyl-6,7-dideoxy-8-C-phe-
nyl-D-glycero-β-D-manno-octopyranosyl)-(1→4)-2,3,6-tri-O-
benzyl-α/β-D-glucopyranoside (28). With the 4-OH as acceptor
slightly changed conditions were used. Donor (95 mg, 0.18 mmol),
TTBP (110 mg, 0.44 mmol, 2.5 equiv), BSP (37 mg, 0.18 mmol, 1
equiv), Tf2O (59 μL, 0.35 mmol, 2 equiv), and acceptor (123 mg, 0.26
mmol, 1.5 equiv). The glycosylation gave a mixture of ≈2:1 α/β
anomers in 71% yield, which was estimated from NMR. The major
byproduct was methyl 2,3,6-tri-O-benzyl-4-O-trifluoromethanesulfon-
yl-α-D-glucopyranoside (36 mg). The anomers could not be clearly
separated from each other, the triflated acceptor and the acceptor
Phenylthio 2,3,4,6-Tetra-O-benzyl-α-D-mannopyranoside
(31).55 To a 0 °C solution of phenylthio-α-D-mannopyranoside
(0.50 g, 1.84 mmol) in DMF (15 mL) was added NaH (60% in
mineral oil, 0.59 g 14.7 mmol, 8 equiv), and the mixture was stirred for
30 min. Then, benzyl bromide (0.98 mL, 8.26 mmol, 4.5 equiv) was
added and the ice bath removed. The reaction was left until
completion (3 h, estimated by TLC). Methanol was added to quench
the reaction, and the mixture was afterward concentrated, poured into
water, and extracted with EtOAc (3×). The combined organic layers
were washed with brine, dried (MgSO4), concentrated, and purified by
dry column chromatography (heptane with a 1.3% gradient of EtOAc)
to give the product as a yellow syrup in 87% yield (1.01 g): Rf 0.45
(heptane/EtOAc 6:1); ESI m/z calcd for C40H40O5SNa 655.2494,
1
found 655.2514; H NMR (500 MHz, CDCl3) δ 7.44 (m, 2H; PhS),
7.36−7.20 (m, 23H; Ph), 5.61 (d, J1,2 = 1.8 Hz, 1H; H-1), 4.90 (d, J =
10.8 Hz, 1H; CH(Ha)Ph), 4.31 (d, J = 12.3 Hz, 1H; CH(Ha)Ph), 4.64
(m, 2H; CH2Ph), 4.61 (m, 2H; CH2Ph), 4.54 (d, J = 10.8 Hz, 1H;
CH(Hb)Ph), 4.49 (d, J = 12.0 Hz, 1H; CH(Hb)Ph), 4.28 (ddd, J = 9.9,
5.1, 1.9 Hz, 1H; H-5), 4.07 (t, J = 9.6 Hz, 1H; H-4), 4.00 (dd, J2,3 = 3.1
Hz, J1,2 = 1.8 Hz, 1H; H-2), 3.86 (m, 2H; H-3, H-6b), 3.75 (dd, J =
10.9 Hz, J = 1.9 Hz 1H; H-6b); 13C NMR (126 MHz, CDCl3) δ 138.6
(CBn‑ipso), 138.5 (CBn‑ipso), 138.3 (CBn‑ipso), 138.1 (CBn‑ipso), 134.6
(CSPh‑ipso), 131.8 (CPh), 129.1 (CPh), 128.6 (CPh), 128.5 (CPh), 128.5
(CPh), 128.4 (CPh), 128.11 (CPh), 128.06 (CPh), 128.0 (CPh), 127.9
2203
dx.doi.org/10.1021/jo302455d | J. Org. Chem. 2013, 78, 2191−2205

The Journal of Organic Chemistry
Featured Article
(CPh), 127.84 (CPh), 127.76 (CPh), 127.6 (CPh), 127.5 (CPh), 85.9 (C-
1), 80.3 (C-3), 76.4 (C-2), 75.3 (CBn), 75.1 (C-4), 73.4 (CBn), 72.9
(C-5), 72.3 (CBn), 72.0 (CBn), 69.3 (C-6).
Notes
The authors declare no competing financial interest.
Competitions Experiments. General procedure for competitive
NIS/TfOH-promoted glycosylation: The donors (ca. 01 mmol, 1
equiv each) and 1-adamantanol (acceptor, 3 equiv) were coevaporated
with toluene (2×) and placed under vacuum for 30 min together with
activated molecular sieves (3 Å). Then, freshly distilled CH2Cl2 (donor
conc 0.05 M) was added, and the mixture was stirred under argon at rt
for 30 min. Subsequently, NIS (1 equiv) was added, and the mixture
was cooled to −40 °C followed by the addition of TfOH (0.1 equiv,
0.1 m stock solution in distilled CH2Cl2). Then, the reaction mixture
was slowly warmed to 0 °C in about 3 h after which Et3N (0.1 mL)
was added. The reaction mixture was diluted with EtOAc, washed with
satd aq Na2S2O3 and brine, dried (MgSO4), and concentrated. NMR
of the crude mixture was used to determine the consumption of
donors.
Phenylthio 2,3,4,6-Tetra-O-benzyl-α-D-mannopyranoside vs
Phenylthio 2,3-Di-O-benzyl-4,6-O-benzylidene-α-D-mannopyr-
anoside (31 vs 2). Perbenzylated donor 31 (63 mg, 1.0 mmol) and
benzylidene donor 2 (54 mg, 0.1 mmol) were used in a competitive
glycosylation using 1-adamantanol (46 mg, 3.0 mmol) following the
general procedure described above. The experiment yielded a 1:19
(perbenzyl vs benzylidene donor) mixture of starting material
determined from crude NMR. Diagnostic peaks used for determi-
nation of the donor ratio: 1H NMR (500 MHz, CDCl3) δ 5.63 (d, J =
1.8 Hz, 1H; α-SPh H-1 of perbenzyl), 5.53 (d, J = 1.4 Hz, 19.2H; α-
SPh H-1 of benzylidene). 13C NMR (126 MHz, CDCl3) δ 87.2 (2JC1,H1
= 167.3 Hz, C-1 of α-SPh benzylidene), 85.8 (2JC1,H1 = 167.0 Hz, C-1
of α-SPh perbenzyl).
ACKNOWLEDGMENTS
■
Christian Tortzen is acknowledged for assistance with low-
temperature NMR experiments. The Lundbeck foundation is
acknowledged for financial support.
REFERENCES
■
(1) Reeves, R. E. J. Am. Chem. Soc. 1950, 72, 1499−1506.
(2) For examples, see: Demchenko, A. V. Curr. Org. Chem. 2003, 7,
35−79. El Ashry, E. S. H.; Rashed, N.; Ibrahim, E. S. I. Curr. Org.
Synth. 2005, 2, 175−213. Zhu, X.; S; Schmidt, R. R. Angew. Chem., Int.
Ed. 2009, 48, 1900−1934. Ishiwata, A.; Lee, Y. J.; Ito, Y. Org. Biomol.
Chem. 2010, 8, 3596−3608.
(3) Recently reviewed in: Ishiwata, A.; Lee, Y. J.; Ito, Y. Org. Biomol.
Chem. 2010, 8, 3596−3608.
(4) (a) Miljkovic, M.; Gligorijevic, M.; Glisin, D. J. Org. Chem. 1974,
39, 3223−3226. (b) Kunz, H.; Gunther, W. Angew. Chem., Int. Ed.
̈
1988, 27, 1086−1087. (c) David, S.; Malleron, A.; Dini, C. Carbohydr.
Res. 1989, 188, 193−200.
(5) (a) Theander, O. Acta Chem. Scand. 1958, 12, 1883−1885.
(b) Ekborg, G.; Lindberg, B.; Lonngren, J. Acta Chem. Scand. 1972, 26,
̈
3287−3292. (c) Lichtenthaler, F. W.; Klaeres, U.; Lergenmueller, M.;
Schwidetzky, S. Synthesis 1992, 179−184. (d) Lichtenthaler, F. W.
Chem. Rev. 2011, 111, 5569−5609.
(6) (a) Ohtake, H.; Iimori, T.; Ikegami, S. Tetrahedron Lett. 1997, 38,
3413−3414. (b) Iimori, T.; Ohtake, H.; Ikegami, S. Tetrahedron Lett.
1997, 38, 3415−3418. (c) Ikegami, S.; Ohtake, H.; Ichiba, N. J. Org.
Chem. 2000, 65, 8171−8179.
(7) (a) Gorin, P. A. J.; Perlin, A. S. Can. J. Chem. 1961, 39, 2474−
2485. (b) Paulsen, H.; Lockhoff, O. Tetrahedron Lett. 1974, 4027−
4030. (c) Garegg, P. J.; Iversen, T. Carbohydr. Res. 1979, 70, C13−
C14.
Phenylthio 2,3,4,6-Tetra-O-benzyl-α-D-mannopyranoside vs
Phenylthio 4,8-Anhydro-2,3-di-O-benzyl-6,7-dideoxy-8-phe-
nyl-D-glycero-α-D-manno-octopyranoside (31 vs 1). Perbenzy-
lated donor 31 (63 mg, 1.0 mmol) and C7-analogue donor 1 (54 mg,
0.1 mmol) were used in a competitive glycosylation using 1-
adamantanol (46 mg, 3.0 mmol) following the general procedure
described above. The experiment yielded a 1:2.6 (perbenzyl vs C-7
benzylidene analogue donor) mixture of starting material determined
from crude NMR. Diagnostic peaks used for determination of the
(8) Srivastava, V. K.; Schuerch, C. Tetrahedron Lett. 1979, 3269−
3272.
(9) (a) Srivastava, V. K.; Schuerch, C. Carbohydr. Res. 1980, 79,
C13−C16. (b) Srivastava, V. K.; Schuerch, C. J. Org. Chem. 1981, 46,
1121−1126. (c) El Ashry, E. S. H.; Schuerch, C. Carbohydr. Res. 1982,
105, 33−43.
(10) (a) Abdel-Rahman, A.A.-H.; Jonke, S.; El Ashry, E. S. H.;
Schmidt, R. R. Angew. Chem., Int. Ed. 2002, 41, 2972−2974.
Corrections: Abdel-Rahman, A.A.-H.; Jonke, S.; El Ashry, E. S. H.;
Schmidt, R. R. Angew. Chem., Int. Ed. 2004, 43, 4389. Abdel-Rahman,
A.A.-H.; Jonke, S.; El Ashry, E. S. H.; Schmidt, R. R. Angew. Chem., Int.
Ed. 2008, 47, 5277. Crich, D.; Li, M.; Jayalath, P. Carbohydr. Res. 2009,
344, 140−144. (b) Crich, D.; Huton, T. K.; Banerjee, A.; Jayalath, P.;
Picione, J. Tetrahedron: Asymmetry 2005, 16, 105−119.
(11) Baek, J. Y.; Lee, B.-Y.; Jo, M. G.; Kim, K. S. J. Am. Chem. Soc.
2009, 131, 17705−17713.
(12) (a) Crich, D.; Sun, S. J. Org. Chem. 1996, 61, 4506−4507.
(b) Crich, D.; Sun, S. J. Org. Chem. 1997, 62, 1198−1199.
(13) Kahne, D.; Walker, S.; Cheng, Y.; Engen, D. V. J. Am. Chem. Soc.
1989, 111, 6881−6882.
(14) Crich, D. J. Carb. Chem. 2002, 21, 667−690.
(15) For computational studies of the influence of oxo-carbenium ion
conformation on the selectivity see: (a) Walvoort, M. T. C.; Dinkelaar,
1
donor ratio: H NMR (500 MHz, CDCl3) δ 5.63 (d, J = 1.8 Hz, 1H;
α-SPh H-1 of perbenzyl), 5.55 (d, J = 1.5 Hz, 1H; α-SPh H-1 of C7-
analogue). 13C NMR (126 MHz, CDCl3) δ 86.9 (2JC1,H1 = 166.4 Hz,
C-1 of α-SPh C7-analogue), 85.8 (2JC1,H1 = 167.0 Hz, C-1 of α-SPh
perbenzyl).
Phenylthio 4,8-Anhydro-2,3-di-O-benzyl-6,7-dideoxy-8-phe-
nyl-D-glycero-α-D-manno-octopyranoside vs Phenylthio 2,3-Di-
O-benzyl-4,6-O-benzylidene-α-D-mannopyranoside (1 vs 2).
C7-analogue donor 1 (54 mg, 1.0 mmol) and benzylidene donor 2
(54 mg, 0.1 mmol) were used in a competitive glycosylation using 1-
adamantanol (46 mg, 3.0 mmol) following the general procedure
described above. The experiment yielded a 1:2.6 (C7-analogue vs
benzylidene donor) mixture of starting material determined from
crude NMR. Diagnostic peaks used for determination of the donor
1
ratio: H NMR (500 MHz, CDCl3) δ 5.55 (d, J = 1.5 Hz, 1H; α-SPh
H-1 of C7-analogue), 5.53 (d, J = 1.4 Hz, 2.6H; α-SPh H-1 of
benzylidene). 13C NMR (126 MHz, CDCl3) δ 87.2 (2JC1,H1 = 167.3
Hz, C-1 of α-SPh benzylidene), 86.9 (2JC1,H1 = 166.4 Hz, C-1 of α-SPh
C7-analogue).
ASSOCIATED CONTENT
́
J.; van den Bos, L. J.; Lodder, G.; Overkleeft, H. S.; Codee, J. D. C.;
■
van der Marel, G. A. Carbohydr. Res. 2010, 345, 1252−1263.
(b) Whitfield, D. M. Carbohydr. Res. 2007, 342, 1726−1740.
(c) Whitfield, D. M. Carbohydr. Res. 2002, 356, 180−190.
(d) Satoh, S.; Hansen, H. S.; Manabe, S.; van Gunnsteren, W. F.;
S
* Supporting Information
Characterization data, low-temperature NMR studies, competi-
tion experiments, and NMR spectra. This material is available
free of charge via the Internet at http://pubs.acs.org
Hunenberger, P. H. J. Chem. Theor. Comput. 2010, 6, 1783−1797.
(16) Experimental and computational studies in favor of an SN2-type
̈
reaction: (a) Crich, D. Acc. Chem. Res. 2010, 43, 1144−1153.
AUTHOR INFORMATION
■
́
(b) Huang, M.; Retailleau, P.; Bohe, L.; Crich, D. J. Am. Chem. Soc.
Corresponding Author
*E-mail: cmp@chem.ku.dk.
2012, 134, 14746−14749.
(17) Crich, D.; Banerjee, A. J. Am. Chem. Soc. 2006, 128, 8078−8086.
2204
dx.doi.org/10.1021/jo302455d | J. Org. Chem. 2013, 78, 2191−2205

The Journal of Organic Chemistry
Featured Article
(18) Crich, D.; Li, L. J. Org. Chem. 2009, 74, 773−781.
(19) Jensen, H. H.; Nordstrøm, L. U.; Bols, M. J. Am. Chem. Soc.
2004, 126, 9205−9213.
(42) Crich, D.; Cai, W.; Dai, Z. J. Org. Chem. 2000, 65, 1291−1297.
(43) (a) Crich, D.; Cai, W.; Dai, Z. J. Org. Chem. 2000, 65, 1291−
1297. (b) Crich, D.; Vinod, A. U.; Picione, J. J. Org. Chem. 2003, 68,
8453−8458.
́
(20) (a) Huang, M.; Garrett, G. E.; Birlirakis, N.; Bohe, L.; Pratt, D.
(44) Heuckendorff, M.; Pedersen, C. M.; Bols, M. J. Org. Chem. 2012,
77, 5559−5568.
A.; Crich, D. Nature Chem. 2012, 4, 663−667. (b) Crich, D.;
Chandrasekera, N. S. Angew. Chem., Int. Ed. 2004, 43, 5386−5389.
(21) Krumper, J. R.; Salamant, W. A.; Woerpel, K. A. J. Org. Chem.
2009, 74, 8039−8050.
(22) (a) Yang, M. T.; Woerpel, K. A. J. Org. Chem. 2009, 74, 545−
553. (b) Lucero, C. G.; Woerpel, K. A. J. Org. Chem. 2006, 71, 2641−
2647.
(45) Whitfield, D. M. Adv. Carb. Chem. Biochem. 2009, 62, 83−159.
See also ref 22a.
(46) Increased triflate stability does not necessary lead to a higher
degree of selectivity in glycosylation reactions, e.g., mannosylations.
(47) Crich, D.; Vinogradova, O. J. Am. Chem. Soc. 2007, 129, 11756−
11765.
(23) Krainer, E.; Naider, F.; Becker, J. Tetrahedron Lett. 1993, 34,
(48) For a computational study, see: Li, Z. Carbohydr. Res. 2010, 345,
1952−1957.
1713−1716.
(24) Similar eliminations during oxidation reactions have previously
been described when there is an acyl-protective group on O4:
(a) Mackie, D. M.; Perlin, A. S. Carbohydr. Res. 1972, 24, 67−86.
(b) Jordan, A. M.; Osborn, H. M. I.; Stafford, P. M.; Tzortzis, G.;
Rastall, R. A. J. Carbohydr. Chem. 2003, 22, 705−717.
(25) (a) Brestensky, D. M.; Huseland, D. E.; McGettigan, C.; Stryker,
J. M. Tetrahedron Lett. 1988, 29, 3749−3752. (b) Mahoney, W. S.;
Brestensky, D. M.; Stryker, J. M. J. Am. Chem. Soc. 1988, 110, 291−
293. (c) Brestensky, D. M.; Stryker, J. M. Tetrahedron Lett. 1989, 30,
5577−5680. (d) Koenig, T. M.; Daeuble, J. F.; Brestensky, D. M.;
Stryker, J. M. Tetrahedron Lett. 1990, 31, 3237−3240.
(26) Wang, C. C.; Lee, J. C.; Luo, S. Y.; Fan, H. F.; Pai, C. L.; Yang,
W. C.; Lu, L. D.; Hung, S. C. Angew. Chem., Int. Ed. 2002, 41, 2360−
2362.
(49) Crich, D.; Smith, M. J. Am. Chem. Soc. 2002, 124, 8867−8869.
(50) (a) Harwood, L. M. Aldrichimica Acta 1985, 18, 25.
(b) Pedersen, D. S.; Rosenbohm, C. Synthesis 2001, 2431−2434.
(51) (a) Brestensky, D. M.; Huseland, D. E.; McGettigan, C.; Stryker,
J. M. Tetrahedron Lett. 1988, 29, 3749−3752. (b) Brestensky, D. M.;
Stryker, J. M. Tetrahedron Lett. 1989, 30, 5677−5680. (c) Koenig, T.
M.; Daeuble, J. F.; Brestensky, D. M.; Stryker, J. M. Tetrahedron Lett.
1990, 31, 3237−3240. (d) Mahoney, W. S.; Brestensky, D. M.;
Stryker, J. M. J. Am. Chem. Soc. 1988, 110, 291−293.
(52) Crich, D.; Banerjee, A. Org. Lett. 2005, 7, 1395−1398.
(53) Crich, D.; Sun, S. Tetrahedron 1998, 54, 8321−8348.
(54) Crich, D.; Banerjee, A.; Li, W.; Yao, Q. J. Carbohydr. Chem.
2005, 24, 415−424.
(55) Charbonnier, F.; Penades
3650−3656.
́
, S. A. Eur. J. Org. Chem. 2004, 17,
(27) Oikawa, Y.; Yoshioka, T.; Yonemitsu, O. Tetrahedron Lett. 1982,
23, 885−888.
(28) Crich, D.; de la Mora, M.; Vinod, A. U. J. Org. Chem. 2003, 68,
8142−8148.
(29) Crich, D.; Smith. J. Am. Chem. Soc. 2001, 123, 9015−9020.
(30) Bock, K.; Pedersen, C. J. Chem. Soc., Perkin Trans. 2 1974, 293−
297. Duss, J. O.; Gotfredsen, C. H.; Bock, K. Chem. Rev. 2002, 100,
4589−4614.
(31) Crich, D.; Jayalath, P.; Hutton, T. K. J. Org. Chem. 2006, 71,
3064−3070.
(32) Procedure inspired by: Christina, A. E.; Muns, J. A.; Olivier, J.
Q. A.; Visser, L.; Hagen, B.; van den Bos, L. J.; Overkleeft, H. S.;
́
Codee, J. D. C.; van der Marel, G. A. Eur. J. Org. Chem. 2012, 29,
5729−5737.
(33) Adamantanol was chosen as acceptor, since it is bulky and
achiral, which eliminates problems with match−mismatch between
donor and acceptor.
(34) The relative reactivity values (RRVs) were obtained from:
Koeller, K. M.; Wong, C.-H. Chem. Rev. 2000, 100, 4465−4493.
(35) Crich, D.; Sun, S. J. Am. Chem. Soc. 1997, 119, 11217−11223.
(36) (a) Crich, D.; Vinogradova, O. J. Org. Chem. 2006, 71, 8473−
8480. (b) Honda, E.; Gin, D. Y. J. Am. Chem. Soc. 2002, 124, 7343−
7352. (c) Boebel, T. A.; Gin, D. Y. J. Org. Chem. 2005, 70, 5818−5826.
(d) Crich, D.; Li, W. Org. Lett. 2006, 8, 959−962. (e) Walvort, M. T.
C.; Lodder, G.; Overkleeft, H. S.; Codee
Org. Chem. 2010, 75, 7990−8002.
(37) Garcia, B. A.; Gin, D. Y. J. Am. Chem. Soc. 2000, 122, 4269−
4279.
́
, J. D. C.; van der Marel, G. J.
(38) β- or equatorial triflates have been observed previously: (a) See
ref 33. (b) Crich, D.; Bowers, A. A. J. Org. Chem. 2006, 71, 3452−
3463. (c) Wei, P.; Kerns, R. J. J. Org. Chem. 2005, 70, 4195−4198.
(d) Walvoort, M. T. C.; Lodder, G.; Mazurek, J.; Overkleeft, H. S.;
Codee, J. D. C.; van der Marel, G. A. J. Am. Chem. Soc. 2009, 131,
́
12080−1208. (e) See refs 31 and 35.
(39) A decomposition temperature of −10 °C for 2 appears in the
recent account by Crich, but without a reference to original work:
Crich, D. Acc. Chem. Res. 2010, 43, 1144−1153. The decomposition
temperature for the 2,3-di-O-methyl analogue can be found in ref 11.
(40) Wong, C. H.; Zhang, Z.; Ollmann, I. R.; Ye, X. S.; Wischnat, R.;
Baasov, T. J. Am. Chem. Soc. 1999, 121, 734−753.
(41) Fraser-Reid, B.; Wu, Z.; Andrews, C. W.; Skowronski, E.;
Bowen, J. P. J. Am. Chem. Soc. 1991, 113, 1434−1435.
2205
dx.doi.org/10.1021/jo302455d | J. Org. Chem. 2013, 78, 2191−2205