Glycosidation of thioglycosides in the presence of bromine: Mechanism, reactivity, and stereoselectivity

Article abstract of DOI:10.1021/jo2019174

Elaborating on previous studies by Lemieux for highly reactive "armed" bromides, we discovered that β-bromide of the superdisarmed (2-O-benzyl-3,4,6-tri-O-benzoyl) series can be directly obtained from the thioglycoside precursor. When this bromide is glycosidated, α-glycosides form exclusively; however, the yields of such transformations may be low due to the competing anomerization into α-bromide that is totally unreactive under the established reaction conditions.

Full text of DOI:10.1021/jo2019174

Article
pubs.acs.org/joc
Glycosidation of Thioglycosides in the Presence of Bromine:
Mechanism, Reactivity, and Stereoselectivity
Sophon Kaeothip, Jagodige P. Yasomanee, and Alexei V. Demchenko*
Department of Chemistry and Biochemistry, University of MissouriSt. Louis, One University Boulevard, St. Louis, Missouri 63121,
United States
S
* Supporting Information
ABSTRACT: Elaborating on previous studies by Lemieux for
highly reactive “armed” bromides, we discovered that β-
bromide of the superdisarmed (2-O-benzyl-3,4,6-tri-O-benzo-
yl) series can be directly obtained from the thioglycoside
precursor. When this bromide is glycosidated, α-glycosides
form exclusively; however, the yields of such transformations
may be low due to the competing anomerization into α-
bromide that is totally unreactive under the established
reaction conditions.
Scheme 1. Lemieux’s in Situ Anomerization Concept
INTRODUCTION
■
Owing to many recent breakthroughs in the field, the formation
of many glycosidic bonds can be readily achieved.1−3 However,
the ability to effectively control the stereoselectivity of
glycosylation, a remarkably complex and multistep reaction
that involves activation, dissociation, nucleophilic attack, and
deprotonation,4−6 is still limited. Building upon our earlier
discoveries of the O-2/O-5 cooperative effect in glycosylation
and the superdisarmed glycosyl donors bearing the 2-O-benzyl-
3,4,6-O-triacyl protecting group pattern,7 herein we present a
study toward the development of a highly stereocontrolled
glycosylation using glycosyl bromides as glycosyl donors
generated in situ. Glycosyl bromides are the most studied
glycosyl donors,8 and their halide-ion-catalyzed glycosidation
introduced by Lemieux et al. in 1975 is arguably the most
significant application in synthesis.9 Overall, Lemieux’s
approach is among the most stereoselective procedures for
direct synthesis of 1,2-cis-glycosides developed to date.10 Thus,
it was found that a rapid equilibrium could be established
between α-halide A and its more reactive β-counterpart I by
adding tetraalkylammonium bromide (Scheme 1). At first,
expulsion of the α-halide A leads to ion-pair B. Since no
inverted 1,2-trans-glycoside E is formed, ion-pair triplet F
leading to the anomerized β-linked bromide I was assumed to
be a more energetically favorable pathway. The existence of
alternate conformations for intermediates G and H en route to/
from I was deemed necessary to form/activate the equatorial
bond.11 The highly reactive β-halide I rapidly dissociates back
into ion-pair G, whereupon it quickly undergoes nucleophilic
attack to form the 1,2-cis product L. As an end result,
nucleophilic substitution of the β-bromide I occurs favorably,
whereas the α-bromide A anomerizes before glycosylation can
occur.
RESULTS AND DISCUSSION
■
Does the Lemieux concept offer a practical application to
synthesis of glycosides and oligosaccharides? Yes, but it lacks
generality, as it requires very reactive perbenzylated bromides
or iodides8,12−14 as glycosyl donors and provides satisfactory
results mainly with the highly reactive series, fucose and
galactose. In our hands, this procedure was found less effective,
and thorough investigation of differently protected glucosyl
bromides 1a−5a in reactions with glycosyl acceptor 615 showed
that it is mainly applicable to glycosidation of perbenzylated
bromide 1a.16 Even then, the reaction was extremely sluggish
and proceeded only halfway in 120 h (entry 1, Table 1). The
stereoselectivity obtained for disaccharide 717 was good (α/β =
9/1) but far from being exclusive. In addition, the application of
such uniformly protected building blocks to oligosaccharide
synthesis is limited to the introduction of the terminal units
only. Our consecutive attempt to diversify the protecting group
Received: September 16, 2011
Published: December 4, 2011
© 2011 American Chemical Society
291
dx.doi.org/10.1021/jo2019174 | J. Org. Chem. 2012, 77, 291−299

The Journal of Organic Chemistry
Article
Table 1. Glycosidation of Differently Protected Glycosyl
Bromides 1a−5a in the Presence of Bu4NBr
Table 2. Glycosidation of Thioglycosides 12b, 13, and 14b
with Glycosyl Acceptor 6 via the Intermediacy of β-
Bromides
a
a
Herein and throughout the article, α- and β-linked glycosyl donors or
reactive intermediates are designated with letters a and b in the
b
compound numbering, respectively. Herein and below, whenever the
reaction was balanced to estimate the amount of the unreacted
glycosyl donor remaining, a combined yield for disaccharide and donor
c
remaining was typically in the range of 87−97%. Herein and below,
the anomeric ratios were determined by comparison of the integral
intensities of the corresponding signals in NMR spectra. Data for
spectra wherein no opposite anomer could be detected or recorded
ratios were above 25/1 are listed as a conservative estimate of >25/1.
studies, however, the reaction of thioglycosides with bromine
was used to generate the corresponding glycosyl bromide,
which upon isolation is usually seen as the α-anomer only.16,27
It is possible that in the absence of the glycosyl acceptor the
kinetic β-bromide rapidly equilibrates into the thermodynami-
cally more stable α-anomer. Alternatively, if the glycosyl
acceptor is present from the beginning, β-bromide can undergo
direct nucleophilic displacement that arguably takes place via
the concerted bimolecular (or close ion-pair) pathway, resulting
in the formation of the inverted α-linked glycoside only. Since
the disaccharide 11 was isolated in only 28% yield along with
63% of α-bromide 5a, we hypothesized that there is a
competition between β-bromide glycosidation and anomerization
and that the latter is favored under these reaction conditions.
With both glycosylation and anomerization reactions being
irreversible under these reaction conditions, the resulting α-
bromide 5a is kinetically stable and cannot be glycosidated.
In a similar fashion, perbenzoylated thioglycoside 1328 was
glycosidated in the presence of bromine to afford β-linked
product disaccharide 1029 in 45% yield (entry 3). Since α-
bromide 4a did not undergo any glycosidation under these
reaction conditions (entry 4), we conclude that the formation
of 10 also takes place via the intermediacy of β-bromide 4b. It
should be noted that, in contrast to the previous example with
5b, β-bromide 4b is equipped with the participating group at C-
2. Therefore, 4b undergoes the sequential two-step displace-
ment via the intermediacy of the bicyclic acyloxonium ion (vide
infra, Scheme 3). Since no α-linked disaccharide formation was
detected, this appears to be the only route leading to
glycosylation, which agrees well with the well-understood
pattern by adding a single benzoyl substituent led to the
notable decrease in the rate of glycosylation and yields (entries
2 and 3). Stereoselectivity and yields observed with 6-O-
benzoyl bromide 2a and 4-O-benzoyl bromide 3a differed
drastically, indicating that the electron withdrawal is very
position dependent.18,19 Not surprisingly, our attempts to
activate bromides 4a20 and 5a bearing the disarming and
superdisarming protecting group pattern, respectively, have
failed completely under Lemieux’s inversion conditions (entries
4 and 5).
Interestingly, β-ethylthioglycoside 12b21 reacted readily in
the presence of bromine and provided disaccharide 11 as the α-
linked diastereomer only (entry 1, Table 2). This result was
rather unexpected, since we first believed that glycosyl bromide
5a was the reactive intermediate in this reaction. We also
noticed that no reaction would take place if acceptor 6 was
added at the later stage when all thioglycoside 12b has been
consumed. We also observed that bromide 5a cannot be
glycosidated in the presence of bromine (entry 2). These
observations led us to a working hypothesis that thioglycoside
12b is first converted into the corresponding β-bromide 5b that
reacts readily with the glycosyl acceptor to give the α-linked
glycoside. Literature precedent for this exists, and the formation
of β-bromides from thioglycosides and selenium glycosides has
been documented.22,23 Along similar lines, β-chlorides have
been also synthesized and used.24−26 In a majority of previous
292
dx.doi.org/10.1021/jo2019174 | J. Org. Chem. 2012, 77, 291−299

The Journal of Organic Chemistry
Article
effect of neighboring participating groups.30−33 When a similar
glycosylation experiment was set up with perbenzylated
thioglycoside 14b,34 disaccharide 7 was readily obtained in a
good yield of 71%, but as a mixture of diastereomers (entry 5).
Also perbenzylated α-bromide 1a reacted readily under these
reaction conditions (entry 6), providing an outcome similar to
that obtained with thioglycoside 14b. Perhaps this reaction
proceeds via the classic SN1 pathway, via the intermediacy of
the flattened oxacarbenium ion (vide infra) that has a tendency
to result in scrambled stereoselectivity,4,35 even under carefully
controlled reaction conditions.
Table 3. NMR Monitoring of Reactions of Thioglycosides
12−14 with Bromine Followed by the Anomerization of
Glycosyl Bromides 1, 4, and 5
Therefore, we decided to investigate the reaction between
superdisarmed thioglycoside 12b and glycosyl acceptor 6 that
in the presence of bromine leads to complete stereoselectivity.
In order to understand why only a modest yield of 28% could
be achieved, we needed to delineate between two possible
reactions pathways by which the reaction may proceed. First,
the rates of formation of the diastereomeric bromides 5a and
5b can be similar, but only the β-anomer 5b then elaborates
into glycoside 11, while the α-bromide 5a remains intact.
Second, it is also possible that the kinetic β-bromide 5b forms
first and then anomerizes into its more stable α-counterpart 5a
concomitantly with glycosidation. Inspired by previous
mechanistic studies,17,36−39 we hoped that monitoring the
reaction by NMR would help to differentiate between the two
possible pathways. We initially investigated the reaction of
perbenzoylated β-thioglycoside 13 with bromine in CDCl3 that
was performed in the standard 5 mm NMR tube directly. As
depicted in Scheme 2, α- and β-bromides (4a and 4b) form
entry
1
reaction
temp
rt
time recorded
ratio α/β
13 → 4a/4b
5 min
15 min
30 min
3 h
1/10.7
1/10.3
1/9.0
1/4.3
1/0.8
16 h
2
3
14b → 1a/1b
14b → 1a/1b
rt
5 min
>25/1
a
low
5 min
10 min
20 min
1 h
7.3/1
8.9/1
10.2/1
12.7/1
4
12b → 5a/5b
rt
rt
5 min
15 min
1 h
2.1/1
2.2/1
2.4/1
Scheme 2. In Situ NMR Monitoring of the Reaction of
Thioglycoside 13 with Bromine
5
6
14a → 1a/1b
14a → 1a/1b
5 min
>25/1
a
low
5 min
10 min
15 min
30 min
1 h
>1/25
1/2.5
1/0.9
1/0.3
>25/1
7
12a → 5a/5b
rt
5 min
15 min
30 min
3 h
1/20
1/17.3
1/16.3
1/11.5
1/4.5
16 h
a
See text and the Experimental Section
The reaction of perbenzylated thioglycoside 14b was
extremely fast at room temperature, and by the time the first
1H NMR was acquired, about 5 min after the addition of
bromine, α-bromide 1a had formed exclusively (entry 2).
Therefore, we attempted to perform the reaction monitoring at
lower temperature as follows. Thioglycoside 14b was dissolved
in CDCl3 and placed in the standard 5 mm NMR tube
equipped with a septum, which was then placed into liquid
nitrogen (−196 °C) for 10 min. Bromine (1.0 equiv) was
added via syringe, the NMR tube was immediately inserted into
the magnet, and the proton spectrum was recorded (approx-
imately 5 min after Br2 addition). This approach allowed us to
detect a 7.3/1 α/β mixture of 1a/1b, which was then observed to
further equilibrate into the α-bromide 1a (entry 3).
rapidly (5 min), with a vast predominance of 4b (α/β = 1/11).
This was then followed by a relatively slow anomerization of
the β-bromide 4b into its α-counterpart 4a, resulting in a nearly
equal mixture of anomers with a slight predominance of 4a
after 16 h (see also entry 1 in Table 3). A very similar result was
obtained from an analogous reaction performed in CD2Cl2.
The predominant formation of the β-bromide 4b in this case
can be attributed to the assistance of the neighboring benzoyl
substituent at C-2. Hence, it was important to establish how the
β-thioglycosides of the armed and superdisarmed series 14b
and 12b, respectively, both bearing a nonparticipating 2-O-
benzyl substituent, react. The results of this study are
summarized in Table 3.
Next, the reaction of thioglycoside 12b equipped with the
superdisarming protecting group pattern (2-O-benzyl-3,4,6-tri-
O-benzoyl) was investigated. Room-temperature experiment
293
dx.doi.org/10.1021/jo2019174 | J. Org. Chem. 2012, 77, 291−299

The Journal of Organic Chemistry
Article
resulted in the formation of a 2.1/1 mixture with predominance
of α-anomer 5a (entry 4). This mixture then began
equilibrating slowly, and further accumulation of α-anomer 5a
was monitored by NMR. The outcome of this experiment
revealed that even though only β-anomer 5b can be
glycosidated under these reaction conditions, the predominant
direct formation of α-anomer 5a is a conceptual barrier for the
possibility to improve the yield on this reaction. Indeed, if one
assumes that the β-anomer in a 2.1/1 α/β-mixture was to react
entirely, the theoretical yield of its glycosidation still could not
exceed 32% based on the starting thioglycoside 12b. This
calculated yield correlates well with the observed yield of 28%
(see Table 2, entry 1) and corresponds to about 88% of
conversion β-bromide 5b into disaccharide 11, leaving about
12% as a result of the concomitant competing anomerization of
5b into 5a.
This result is significant as it demonstrates that the
glycosidation of 12b (at least with glycosyl acceptor 6) is
significantly faster than its competing anomerization into 5a.
Since no Lemieux-like equilibrium could be established with the
unreactive bromides (vide supra), the only possibility to
improve the yield of such glycosylation is to ensure that β-
bromide is generated predominantly. The necessity to improve
the reaction conditions stimulated us to look into possible
mechanistic pathways by which this transformation may
proceed. In the case of thioglycosides equipped with the
neighboring participating group at C-2, such as 13, or
schematically represented as A in Scheme 3a, the β-thioglyco-
In the case of β-thioglycosides 12b and 14b bearing a
nonparticipating 2-O-benzyl substituent, schematically shown
as F in Scheme 3b, there is no direct route that would lead to
the stereoselective formation of the corresponding β-bromides.
Although arguably the β-bromide is the kinetic product from
the oxacarbenium intermediate G, our results showed that the
formation of the more stable α-bromide predominated in both
cases, with the highest β-bromide content obtained in the case
of the superdisarmed series (5a/5b, α/β = 2.1/1, see Table 3,
entry 4). We conceptualized that the use of α-thioglycosides as
an alternative starting material may offer a more direct route to
obtain the corresponding β-bromides. Our working hypothesis
is depicted in Scheme 3c. Thus, promoter-assisted activation of
the leaving group in α-thioglycoside I with bromine will lead to
the formation of α-sulfonium ion J. Two possible pathways by
which J converts into bromides H that would be reasonable to
assume are as follows. First, the formation of β-H takes place
directly via the concerted nucleophilic (or close-ion-pair)
displacement of the BrSEt leaving group, leading to complete
inversion. Second, upon monomolecular departure of the
activated leaving group in J, the reaction would then proceed
via the intermediacy of the oxacarbenium ion G, which would
likely result in the formation of the anomeric mixture of α/β-H.
This mechanism implies that a stoichiometric amount of
bromine is not required because it would be partially
regenerated upon disproportionation of BrSEt intermediate
(vide infra).
To delineate the most viable reaction pathway by which α-
SEt glycosides elaborate into β-bromides, we first obtained
perbenzylated derivative 14a.41 Disappointingly, the NMR
experiment with 14a performed at rt led to the formation of α-
bromide 1a (entry 5, Table 3). When essentially the same
experiment as described for 14b was performed at low
temperature, β-bromide 1b was formed predominantly (α/β
> 1/25, entry 6), with a trace amount of unreacted α-
thioglycoside still remaining. This was an ultimate proof of
concept and a solid indication that we have begun advancing in
the right direction. Subsequently, we observed that 1b begins
equilibrating very rapidly, with complete anomerization
detected within 1 h after the addition of bromine (judged by
the formation of 1a). Assuming that the rate of this
transformation would be difficult to control in a typical
competing glycosylation experiment, we had hoped that
thioglycoside 12a,21 bearing a superdisarming 2-benzyl-3,4,6-
tribenzoyl protecting group pattern, would allow for reaction
rates that are easier to comprehend, and hence allow us to
achieve better control of its glycosidations. Figure 1 shows a
series of NMR spectra that have been acquired at different time
points of such reaction performed at rt (see also the summary
in entry 7, Table 3).
Scheme 3. Mechanistic Rationale for the Transformation of
Thioglycosides A, F, and I into Bromides E and H in the
Presence of Bromine
After a spectrum of the starting material 12a was recorded
(Figure 1a), bromine (1 equiv) was injected into the NMR
tube. A subsequent NMR spectrum recorded 5 min after the
addition of Br2 showed that bromides 5a/5b are formed in a
ratio of 1/20, favoring the β-anomer (Figure 1b). The ratios
were measured by comparison of the integral intensities of the
stand-alone signals corresponding to H-1 of α-bromide 5a
(6.48 ppm) and H-5 of β-bromide 5b (4.20 ppm), which were
clearly separable from the rest of the signals. At this time, about
24% of the starting material 12a was still remaining, as judged
by measuring the integral intensity of its H-3 signal (5.96 ppm).
Subsequent NMR spectra recorded at 15 and 30 min time
points after the addition of bromine showed that 12a was
side undergoes a double displacement via the intermediacy of
the bicyclic acyloxonium ion D. Thus, upon interaction of A
with bromine, a reactive sulfonium intermediate B is
generated.21,40 The promoter-assisted leaving-group departure
results in the formation of the oxacarbenium intermediate C.
The leaving group first departs as an unstable BrSEt species
that then disproportinates into ethyl disulfide (easily detectable
by NMR) and bromine. Oxacarbenium ion C is then stabilized via
anchimeric assistance of the neighboring benzoyl group, resulting
in the formation of cyclic acyloxonium intermediate D. The forma-
tion of the latter can be used to explain the preferential formation
of β-bromide E, as was detected for 4b (entry 1, Table 3).
294
dx.doi.org/10.1021/jo2019174 | J. Org. Chem. 2012, 77, 291−299

The Journal of Organic Chemistry
Article
Table 4. Glycosidation of Thioglycoside 12a with Various
Glycosyl Acceptors
Figure 1. NMR monitoring of the conversion of thioglycoside 12a
into bromides 5a/5b at rt.
acceptor 6 (67% of 11, entry 3). Unfortunately, glycosylations
of secondary acceptors 15 and 16 were below preparative value,
and although also complete stereoselectivity was recorded
herein, 1→3- and 1→2-linked disaccharides 19 and 20 were
isolated in only 35% (entry 4) and 27% (entry 5) yields,
respectively. As mentioned previously, the overall stoichiometry
of this reaction implies that a stoichiometric amount of
bromine is not required because it would be partially
regenerating upon disproportionation of BrSEt intermediate
(vide infra). At this stage, however, reactions in the presence of
a substoichiometric amount of bromine appear to be much
slower, and the yields obtained are below the preparative value.
We do not yet currently have any data on the formation and
roles of other side products, such as HBr, which might also be
forming as a result of the deprotonation of glycoside products
with bromonium ion.
Certainly, if an efficient equilibrium between unreactive
bromide 5a into its very reactive counterpart 5b could be
established, this would help us to regenerate 5b and ultimately
drive the glycosylation to completion, which would result in
improved yields. Since our preliminary study (Table 1) clearly
demonstrated that bromide 5a cannot be anomerized in the
presence of Bu4NBr, we assumed that its anomerization might
be possible with the assistance of heavy-metal-based promoters.
As discovered in 1901 by Koenigs and Knorr44 (and
independently by Fischer and Armstrong),45 glycosyl halides
can be activated in the presence of Ag2CO3 or Ag2O. The silver
salt was used with the primary intention to scavenge the
hydrogen halide byproduct, and it was not until the early 1930s
that it was realized that the silver salt plays an active role by
assisting in the leaving group's departure.46,47 Additionally, as
shown by Helferich et al.,48−50 mercury(II) salts offer even
more potent assistance for the leaving group's departure.
completely consumed (by 30 min, Figure 1d), and β-bromide
5b began a relatively slow anomerization into its more stable α-
counterpart 5a. Therefore, these results confirm the initial
assumption that the anomerization of superdisarmed bromide
5b is much slower than that of its armed, perbenzylated
counterpart 1b. Moreover, anomerization of 5b is even slower
than that of perbenzoylated disarmed bromide 4b, as 5b is still
present as the major product after 16 h (Figure 1f), whereas 4b
was already surpassed by 4a by the same time point (compare
entries 1 and 7, Table 3). It should be noted that, by this time
point, traces of the hydrolysis product (hemiacetal) have begun
to appear, as evident from the new anomeric signal at ∼6.7 ppm
(Figure 1f).
Having established the most favorable reaction conditions for
the formation of β-bromide 5b at rt, we turned our attention to
studying its glycosidation. For this purpose we obtained a
number of glycosyl acceptors ranging from simple aliphatic
alcohols to representative secondary hydroxyls of monosac-
charide acceptors 1542 and 16.43 Since under the established
reaction conditions only 5b is able to react while 5a remains
inert (vide supra), all glycosylations summarized in Table 4
provided the respective products with complete α-stereo-
selectivity. All ratios of the final products are recorded as our
conservative estimate α/β > 25/1, although in most cases no
traces of the β-linked products were even detected. Although
the issue of the stereoselectivity of this glycosylation has been
resolved, it also has become apparent that the yield of these
reactions ultimately depends on the rate with which the
products are forming and the ability of the glycosylation to
outperform the competing isomerization. For instance,
relatively high yields could be achieved in all cases of
glycosylations of highly reactive glycosyl acceptors: MeOH
(78% of 17,21 entry 1), BnOH (75% of 18, entry 2), and 6-OH
295
dx.doi.org/10.1021/jo2019174 | J. Org. Chem. 2012, 77, 291−299

The Journal of Organic Chemistry
Article
Therefore, the investigation of metal salt-based activators
appealed to us as attractive for further study. In particular, we
assumed that mercury(II) bromide would be capable of both
assisting the leaving group's departure and providing bromide
anion to effect the subsequent in situ anomerization of 5a into
5b. Helferich’s study also demonstrated that faster reactions
often result in a decreased stereoselectivity. Therefore, a delicate
balance between the reactivity and stereoselectivity should be kept
in place to ensure success of the metal-assisted activation.
Significantly improved yields have been also obtained with
secondary glycosyl acceptors 15 and 16. Thus, disaccharides 19
and 20 were obtained in 89% and 87% yield, respectively
(entries 5 and 6). We conclude that the metal-assisted protocol
is predominantly suitable for glycosylations of less reactive
glycosyl acceptors, as in the absence of the metal promoter the
anomerization easily outperforms glycosylation. Even glyco-
sylation of the unreactive 4-OH glycosyl acceptor 2151 was
possible under these reaction conditions. Thus, 1→4-linked
disaccharide 22 was readily obtained in 82% yield, whereas Br2-
only activation of 12a provided only a trace amount of 22 (data
not shown).
To investigate the role that HgBr2 may have on the outcome
of glycosylation, we performed a series of experiments,
summarized in Table 5. Glycosidation of 12a with standard
CONCLUSIONS
Table 5. Glycosidation of Thioglycoside 12a in the Presence
of Br2 and HgBr2 with Various Glycosyl Acceptors
■
The conceptual similarity between this approach and Lemieux’s
halide-assisted procedure introduced in 19759 is undeniable.
However, because no equilibrium between superdisarmed
bromides 5a and 5b could be established by the addition of
tetraalkylammonium bromide, we believe that the method
described herein offers a complementary approach. The fact
that α-bromide 5a was found totally unreactive in the presence
of bromine is a particularly significant result, as it assures that its
β-counterpart 5b will be the key (and apparently the only)
species en route to the glycosylation product under these
reaction conditions.
Both simple aliphatic alcohols and the less reactive sugar
acceptor 6 provided the corresponding glycosides with
complete α-stereoselectivity. Glycosylation of secondary accept-
ors is slower, which gives a window for the β-bromide to
anomerize, and this is ultimately reflected in lower yields of
glycosylations in the presence of bromine only. The efficiency
of glycosylation of secondary acceptors can be significantly
improved by the addition of mercury(II) bromide copromoter.
This modification allowed for very good yields and complete
stereoselectivity. Although the yield of glycosylation of primary
acceptor 6 can also be significantly improved by the addition of
mercury(II) bromide copromoter, it also leads to decreased
stereoselectivity.
EXPERIMENTAL SECTION
General. Column chromatography was performed on silica gel 60,
and reactions were monitored by TLC on Kieselgel 60 F254
■
.
Preparative layer chromatography was performed on PLC silica gel
60 glass plates, Kieselgel 60 F254, 1 mm. The compounds were
detected by examination under UV light and by charring with 10%
sulfuric acid in methanol. Solvents were removed under reduced
pressure at <40 °C. Mercury(II) bromide (99%+, reagent grade) and
tetrabutylammonium bromide (99%+, reagent grade) were used as
received. CH2Cl2 and ClCH2CH2Cl were distilled from CaH2 directly
prior to application. Pyridine was dried by refluxing with CaH2 and
then distilled and stored over molecular sieves (3 Å). Molecular sieves
(3 Å) used for reactions were crushed and activated in vacuo at 390 °C
for 8 h in the first instance and then for 2−3 h at 390 °C directly prior
6-OH acceptor 6 was initiated in the presence of Br2 (1.3
equiv), and HgBr2 (1.3 equiv) was subsequently added at
different time points into the reaction mixture at rt. Since no
activation of 12a in the presence of HgBr2 and in the absence of
Br2 took place, glycosyl bromide is indeed the reactive
intermediate in this reaction. Along with the notable increase
of the yield up to 86% (entries 1−4), the stereoselectivity
remained very high, but in all cases the presence of β-linked
disaccharide 11 could be detected. The reduced stereo-
selectivity obtained here can serve as an indication that 5a
reacted with glycosyl acceptor either directly or via the
intermediacy of the oxacarbenium ion, but not via 5b. Not
surprisingly, higher stereoselectivity was detected with a longer
time period (time 1) before the addition of HgBr2 (entry 4).
1
to application. Optical rotations are listed in deg dm−1 cm3 g−1. H
NMR spectra were recorded in CDCl3 at 300 and 500 MHz, and 13C
NMR spectra were recorded in CDCl3 at 75 and 125 MHz. HR-FAB-
MS determinations were made with the use of a matrix of m-
nitrobenzyl alcohol, with NaI as necessary.
Synthesis of Glycosyl Donors. 6-O-Benzoyl-2,3,4-tri-O-ben-
zyl-α-D-glucopyranosyl Bromide (2a). A solution of ethyl 6-O-
benzoyl-2,3,4-tri-O-benzyl-1-thio-β-D-glucopyranoside52 (0.50 g, 0.83
mmol) and activated molecular sieves (3 Å, 0.40 g) in CH2Cl2 (12.5
mL) was stirred under argon for 1 h. A freshly prepared solution of Br2
in CH2Cl2 (8 mL, 1/165, v/v) was then added, and the reaction
mixture was kept for 15 min at rt. After that, the solid was filtered off,
296
dx.doi.org/10.1021/jo2019174 | J. Org. Chem. 2012, 77, 291−299

The Journal of Organic Chemistry
Article
and the filtrate was concentrated in vacuo at rt. The residue was
purified by silica gel flash column chromatography (ethyl acetate−
toluene gradient elution) to obtain the title compound as a colorless
syrup in 59% yield (0.31 g, 0.50 mmol). Analytical data for 2a: Rf =
4.54−4.70 (m, 4H, H-5, 6b, CH2Ph), 5.60 (dd, 1H, J4,5 = 9.6 Hz, H-4),
6.03 (dd, 1H, J3,4 = 9.6 Hz, H-3), 6.48 (d, 1H, J1,2 = 3.9 Hz, H-1),
7.22−7.57 (m, 14H, aromatic), 7.93 (dd, 4H, aromatic), 8.03 (dd, 2H,
aromatic) ppm; 13C NMR δ 62.4, 68.4, 72.3, 72.8 (×2), 76.8, 89.3,
128.2 (×2), 128.4, 128.5 (×2), 128.5 (×2), 128.6 (×2), 128.7 (×2),
128.8, 129.5, 129.7, 129.9 (×4), 130.1 (×2), 133.3, 133.4, 133.7, 136.8,
165.4, 165.6, 166.3 ppm; HR-FAB-MS [M+Na]+ calcd for
C34H29O8BrNa 667.0943, found 667.0970.
24
0.43 (ethyl acetate−hexane, 1/4, v/v); [α]D +122.6 (c = 1.0,
1
CHCl3); H NMR δ 3.56 (dd, 1H, J2,3 = 9.2 Hz, H-2), 3.73 (dd, 1H,
J4,5 = 9.1 Hz, H-4), 4.11 (dd, 1H, J3,4 = 9.1 Hz, H-3), 4.27 (m, 1H, J5,6a
= 3.0 Hz, H-5), 4.50−5.02 (m, 8H, 3×CH2Ph, H-6a, 6b), 6.41 (d, 1H,
J1,2 = 3.8 Hz, H-1), 7.20−7.60 (m, 18H, aromatic), 7.90 (d, 2H,
aromatic) ppm; 13C NMR δ 62.6, 72.9, 73.9, 75.5, 76.1, 76.2, 80.0,
82.2, 91.1, 128.1, 128.2 (×3), 128.3 (×2), 128.4 (×3), 128.6 (×2),
128.7 (×4), 128.8 (×2), 129.8 (×3), 164.1 ppm; HR FAB-MS [M
+Na]+ calcd for C34H33O6BrNa 639.1358, found 639.1332.
3,4,6-Tri-O-benzoyl-2-O-benzyl-β-D-glucopyranosyl Bromide
(5b). A solution of ethyl 2-O-benzyl-3,4,6-tri-O-benzoyl-1-thio-α-D-
glucopyranoside21 (0.03 mg, 0.04 mmol) in CDCl3 (0.5 mL) in an
NMR tube was frozen in liquid nitrogen. After that, 98% bromine
1
solution (2.4 μL, 0.04 mol) was added directly to the NMR tube. H
1
NMR was recorded immediately. Selected analytical data for 5b: H
4-O-Benzoyl-2,3,6-tri-O-benzyl-α-D-glucopyranosyl Bromide (3a).
Benzoyl chloride (0.14 mL, 1.21 mmol) was added dropwise to a
stirred solution of ethyl 2,3,6-tri-O-benzyl-1-thio-β-D-glucopyrano-
side53 (0.50 g, 1.01 mmol) in pyridine (5.0 mL) at 0 °C. The
resulting reaction mixture was kept for 1 h, methanol (2 mL) was
added, and the mixture was concentrated under reduced pressure. The
residue was dissolved in CH2Cl2 (30 mL) and washed with 1 N HCl
(15 mL) and water (3 × 15 mL). The organic phase was separated,
dried with MgSO4, and concentrated in vacuo. The residue was
purified by column chromatography on silica gel (ethyl acetate−
hexane gradient elution) to obtain ethyl 4-O-benzoyl-2,3,6-tri-O-
benzyl-1-thio-β-D-glucopyranoside (23) as a white solid in 89% yield
(0.54 g, 0.90 mmol). Analytical data for 23: Rf = 0.52 (ethyl acetate−
hexane, 3/7, v/v); [α]D22 −49.8 (c = 1.0, CHCl3); 1H NMR δ 1.42 (t,
3H, SCH2CH3), 2.86 (m, 2H, SCH2CH3), 3.60−3.90 (m, 5H, H-2, 3,
5, 6a, 6b), 4.53 (dd, 2H, 2J = 12.3 Hz, CH2Ph), 4.62 (d, 1H, J1,2 = 9.8
Hz, H-1), 4.68−5.00 (m, 4H, 2×CH2Ph), 5.35 (dd, 1H, J3,4 = 9.7 Hz,
H-4), 7.17−8.00 (m, 20H, aromatic) ppm; 13C NMR δ 15.4, 25.3,
70.1, 71.7, 73.7, 75.5, 75.8, 78.0, 81.8, 83.8, 85.3, 127.7, 127.8 (×3),
128.1 (×3), 128.4 (×4), 128.6 (×5), 129.9 (×2), 133.4, 137.9, 138.0
(×2), 165.5 ppm; HR-FAB-MS [M+Na] calcd for C36H38O6SNa
621.2287, found 621.2299.
NMR δ 4.07 (dd, 1H, J2,3 = 7.8 Hz, H-2), 4.20 (m, 1H, H-5), 4.46 (dd,
1H, J5,6a = 5.1 Hz, J6a,6b = 12.4 Hz, H-6a), 4.57 (dd, 1H, J5,6b = 3.1 Hz,
2
1
2
H-6b), 4.72 (d, 1H, J = 10.7 Hz, /2 CH2Ph), 4.91 (d, 1H, J = 10.8
1
Hz, /2 CH2Ph), 5.67−5.81 (m, 3H, H-1, 3, 4), 7.19−8.05 (m, 20H,
aromatic) ppm; 13C NMR δ 63.0, 68.8, 74.8, 75.3, 76.4, 81.7, 82.6,
128.2, 128.4 (×2), 128.5 (×4), 128.6 (×4), 129.2, 129.6, 129.9 (×2),
130.0 (×3), 130.1 (×2), 133.3, 133.6, 133.7, 136.9, 165.3, 165.6, 166.2
ppm; FAB-MS [M+Na]+ for C34H29O8BrNa 667.1.
Synthesis of Glycosides and Disaccharides. Method A:
General Bu4NBr-Promoted Glycosylation Procedure. A mixture of
the glycosyl donor (0.10 mmol), glycosyl acceptor (0.15 mmol), and
freshly activated molecular sieves (3 Å, 200 mg) in (ClCH2)2 (2 mL)
was stirred under argon for 1 h. Bu4NBr (0.20 mmol) was added, and
the reaction mixture was stirred at rt. Upon completion, the reaction
mixture was diluted with CH2Cl2, the solid was filtered off, and the
residue was washed with CH2Cl2. The combined filtrate (30 mL) was
washed with 20% aq NaHCO3 (10 mL) and water (3 × 15 mL), and
the organic phase was separated, dried with MgSO4, and concentrated
in vacuo. The residue was purified by column chromatography on silica
gel (ethyl acetate−hexane gradient elution) to give the corresponding
disaccharide. Anomeric ratios (if applicable) were determined by
1
comparison of the integral intensities of relevant signals in H NMR
A solution of 23 (0.5 g, 1.59 mmol) and activated molecular sieves
(3 Å, 0.40 g) in CH2Cl2 (12.5 mL) was stirred under argon for 1 h. A
freshly prepared solution of Br2 in CH2Cl2 (8 mL, 1/165, v/v) was
then added, and the reaction mixture was kept for 15 min at rt. After
that, the solid was filtered off, and the filtrate was concentrated in
vacuo at rt. The residue was purified by silica gel flash column
chromatography (ethyl acetate−toluene gradient elution) to obtain the
title compound 3a as a colorless syrup in 70% yield (0.36 g, 0.58
mmol). Analytical data for 3a: Rf = 0.42 (ethyl acetate−hexane, 1/4, v/
v); [α]D24 +110.6 (c = 1.0, CHCl3); 1H NMR δ 3.55−3.72 (m, 3H, H-
2, 6a, 6b), 4.16 (dd, 1H, J3,4 = 9.3 Hz, H-3), 4.34 (m, 1H, J5,6a = 3.7 Hz,
H-5), 4.51 (dd, 2H, 2J = 12.0 Hz, CH2Ph), 4.77 (dd, 2H, 2J = 12.7 Hz,
spectra.
Method B: General Br2-Promoted Glycosylation Procedure. A
mixture of the glycosyl donor (0.10 mmol), glycosyl acceptor (0.15
mmol), and freshly activated molecular sieves (3 Å, 200 mg) in
(ClCH2)2 (2 mL) was stirred under argon for 1 h. Bromine (0.10−
0.13 mmol) was added, and the resulting mixture was stirred at rt.
Upon completion, the reaction mixture was diluted with CH2Cl2, the
solid was filtered off, and the residue was washed with CH2Cl2. The
combined filtrate (30 mL) was washed with 20% aq NaHCO3 (10
mL) and water (3 × 15 mL), and the organic phase was separated,
dried with MgSO4, and concentrated in vacuo. The residue was
purified by column chromatography on silica gel (ethyl acetate−
hexane gradient elution) to give the corresponding disaccharide.
Anomeric ratios (if applicable) were determined by comparison of the
2
1
CH2Ph), 4.88 (d, 1H, J = 11.2 Hz, /2 CH2Ph), 5.56 (dd, 1H, J4,5
=
9.7 Hz, H-4), 6.47 (d, 1H, J1,2 = 3.7 Hz, H-1), 7.13−7.50 (m, 17H,
aromatic), 7.62 (m, 1H, aromatic), 8.00 (d, 2H, aromatic) ppm; 13C
NMR δ 67.9, 69.4, 73.1, 73.7, 74.1, 75.5, 79.3, 79.5, 91.4, 127.7, 127.8,
128.0 (×2), 128.2 (×3), 128.3 (×3), 128.3 (×4), 128.4 (×2), 128.5
(×2), 128.7 (×3), 129.6, 129.9 (×2), 133.4, 137.4, 137.5, 137.9, 165.2
ppm; HR-FAB-MS [M+Na]+ calcd for C34H33BrO6Na 639.1358,
found 639.1343.
1
integral intensities of relevant signals in H NMR spectra.
Method C: General Br2/HgBr2-Promoted Glycosylation Proce-
dure. A mixture of the glycosyl donor (0.10 mmol), glycosyl acceptor
(0.15 mmol), and freshly activated molecular sieves (3 Å, 200 mg) in
(ClCH2)2 (2 mL) was stirred under argon for 1 h. Bromine (0.13
mmol) was added, and the resulting mixture was stirred at rt and
monitored with TLC. After time 1 listed in Table 5, mercury(II)
bromide (0.13 mmol) was added, and the reaction mixture was stirred
for time 2 (Table 5). Upon completion, the reaction mixture was
diluted with CH2Cl2, the solid was filtered off, and the residue was
washed with CH2Cl2. The combined filtrate (30 mL) was washed with
20% aq NaHCO3 (10 mL) and water (3 × 15 mL), and the organic
phase was separated, dried with MgSO4, and concentrated in vacuo.
The residue was purified by column chromatography on silica gel
(ethyl acetate−hexane gradient elution) to give the corresponding
disaccharide. Anomeric ratios (if applicable) were determined by
3,4,6-Tri-O-benzoyl-2-O-benzyl-α-D-glucopyranosyl Bromide
(5a). A solution of ethyl 3,4,6-tri-O-benzoyl-2-O-benzyl-1-thio-β-D-
glucopyranoside21 (1.0 g, 1.59 mmol) and activated molecular sieves
(3 Å, 0.79 g) in CH2Cl2 (24 mL) was stirred under argon for 1 h. A
freshly prepared solution of Br2 in CH2Cl2 (15 mL, 1/165, v/v) was
added, and the reaction mixture was kept for 15 min at rt. After that,
the solid was filtered off, and the filtrate was concentrated in vacuo at
rt. The residue was redissolved in CH2Cl2 and then washed with
distilled water. The organic phase was separated, dried with MgSO4,
and concentrated in vacuo at rt, and the residue was purified by column
chromatography on silica gel (ethyl acetate−toluene gradient elution)
to afford the title compound as a white solid in 71% yield (0.76 g, 1.18
mmol). Analytical data for 5a: Rf = 0.52 (ethyl acetate−hexane, 2/3, v/
1
comparison of the integral intensities of relevant signals in H NMR
spectra.
23
1
v); [α]D +65.2 (c = 1.0, CHCl3); H NMR δ 3.78 (dd, 1H, J2,3
=
Methyl 6-O-(6-O-Benzoyl-2,3,4-tri-O-benzyl-α/β-D-glucopyrano-
syl)-2,3,4-tri-O-benzyl-α-D-glucopyranoside (8). The title compound
9.6 Hz, H-2), 4.45 (dd, 1H, J5,6a = 4.7 Hz, J6a,6b = 12.3 Hz, H-6a),
297
dx.doi.org/10.1021/jo2019174 | J. Org. Chem. 2012, 77, 291−299

The Journal of Organic Chemistry
Article
was obtained by method B from 2a and 6 in 45% yield as a colorless
syrup (α/β = 24/1). Analytical data for α-8: Rf = 0.41 (ethyl acetate−
Hz, H-6a′), 4.34 (dd, 1H, J3,4 = 9.3 Hz, H-3), 4.34 (dd, 1H, J5′,6b′ = 2.3
2
2
Hz, H-6b′), 4.38 (dd, 2H, J = 11.9 Hz, CH2Ph), 4.49 (dd, 2H, J =
12.1 Hz, CH2Ph), 4.65 (dd, 2H, 2J = 12.0 Hz, CH2Ph),4.69 (dd, 2H, 2J
= 11.5 Hz, CH2Ph), 4.73 (d, 1H, J1,2 = 3.5 Hz, H-1), 4.94 (m, 1H, H-
5′), 5.43 (dd, 1H, J4′,5′ = 10.0 Hz, H-4′), 5.68 (d, 1H, J1′,2′ = 3.4 Hz, H-
1′), 6.08 (dd, 1H, J3′,4′ = 9.8 Hz, H-3′), 6.80−8.10 (m, 35H, aromatic)
ppm; 13C NMR δ 55.3, 63.3, 67.7, 68.6, 69.9, 72.3, 72.7, 73.5 (×2),
73.7, 76.5, 77.2, 77.3, 78.4, 78.8, 97.1, 97.6, 126.7 (×2), 127.4, 127.8,
127.9, 128.0 (×2), 128.1 (×2), 128.2, 128.3 (×2), 128.4 (×11), 128.6
(×2), 128.7 (×2), 129.3, 129.9 (×2), 129.9 (×4), 130.1 (×2), 130.3,
132.9, 133.1, 133.3, 137.3, 137.9, 138.0, 138.6, 165.6, 166.0, 166.5
ppm; HR-FAB-MS [M+Na]+ calcd for C62H60O14Na 1051.3881,
found 1051.3897.
1
hexane, 3/7, v/v); H NMR δ 3.32 (dd, 1H, J2′,3′ = 9.6 Hz, H-2′),
3.46−3.54 (m, 3H, H-2, 4, 4′), 3.62 (m, 1H, J6a,6b = 10.1 Hz, H-6a),
3.72 (m, 2H, H-5, 6b), 3.90 (m, 3H, H-3, 3′, 5′), 4.31 (dd, 1H, J5′,6a′
=
4.4 Hz, J6a′,6b′ = 11.9 Hz, H-6a′), 4.43 (dd, 1H, J5′,6b′ = 2.0 Hz), 4.47−
4.64 (m, 7H, J1′,2′ = 3.3 Hz, H-1′, 3×CH2Ph), 4.70−4.75 (m, 2H,
CH2Ph), 4.81−4.93 (m, 5H, J1,2 = 3.6 Hz, H-1, 3×CH2Ph) ppm; 13C
NMR δ 55.4, 63.6, 66.2, 69.1, 70.6, 72.6, 73.5, 75.2 (×2), 75.9 (×2),
77.4, 78.0, 80.3, 80.4, 81.9, 82.3, 97.1, 98.1, 127.8 (×3), 127.9 (×7),
128.0 (×2), 128.1 (×2), 128.2 (×4), 128.4 (×3), 128.5 (×6), 128.6
(×4), 129.8 (×3), 130.2, 133.2, 138.2, 138.3, 138.5 (×2), 138.7, 139.0,
166.4 ppm; HR-FAB-MS [M+Na]+ calcd for C62H64O12Na 1023.4295,
found 1023.4277.
Methyl 2-O-(3,4,6-Tri-O-benzoyl-2-O-benzyl-α-D-glucopyrano-
syl)-3,4,6-tri-O-benzyl-α-D-glucopyranoside (20). The title com-
pound was obtained by method C from 12a and 16 in 87% yield as
a colorless syrup (α/β > 25/1). Selected analytical data for α-20: Rf =
Methyl 6-O-(4-O-Benzoyl-2,3,6-tri-O-benzyl-α/β-D-glucopyrano-
syl)-2,3,4-tri-O-benzyl-α-D-glucopyranoside (9). The title compound
was obtained by method B from 3a and 6 in 15% yield as a colorless
syrup (α/β = 6.2/1). Selected analytical data for α-9: Rf = 0.41 (ethyl
23
0.54 (ethyl acetate−toluene, 3/17, v/v); [α]D +107.8 (c = 1.0,
CHCl3); 1H NMR δ 3.49 (s, 3H, OCH3), 3.69 (dd, 1H, J4,5 = 9.3 Hz,
H-4), 3.73 (dd, 1H, J5,6a = 1.8 Hz, J6a,6b = 10.7 Hz, H-6a), 3.81 (dd,
1H, J5,6b = 3.5 Hz, H-6b), 3.81−3.86 (m, 1H, H-5), 3.85 (dd, 1H, J2′,3′
= 10.0 Hz, H-2′), 4.01 (dd, 1H, J2,3 = 9.8 Hz, H-2), 4.13 (dd, 1H, J5′,6a′
= 5.3 Hz, J6a′,6b′ = 12.2 Hz, H-6a′), 4.18 (dd, 1H, J2,3 = 9.6 Hz, H-3),
4.39 (dd, 1H, J5′,6b′ = 2.1 Hz, H-6b′), 4.62 (m, 1H, H-5′), 4.62 (dd, 2H,
1
acetate−hexane, 3/7, v/v); H NMR δ 3.38 (s, OCH3), 3.47 (m, 2H,
H-2′, 4), 3.98−4.11 (m, 3H, H-3, 3′, 5), 4.59 (d, 1H, J1′,2′ = 3.8 Hz, H-
1′), 4.96 (d, 1H, J1,2 = 2.5 Hz, H-1), 5.31 (dd, 1H, J4′,5′ = 9.5 Hz, H-4′)
ppm; 13C NMR δ 55.3, 69.1, 69.3, 70.6, 71.1, 72.7, 73.6, 73.7, 75.1,
75.2, 75.9, 77.4, 78.1, 78.4, 79.9, 80.3, 82.4, 97.5, 98.2, 127.5 (×2),
127.6 (×2), 127.8 (×5), 127.9 (×3), 128.1 (×2), 128.2 (×4), 128.3
(×3), 128.4, 128.5 (×3), 128.6 (×6), 129.9 (×2), 130.1 (×3), 133.2,
138.0, 138.3, 138.5, 138.7, 139.0, 165.5 ppm; HR-FAB-MS [M+Na]+
calcd for C62H64O12Na 1023.4295, found 1023.4365.
2
2J = 12.0 Hz, CH2Ph), 4.67 (s, 2H, CH2Ph), 4.68 (dd, 2H, J = 10.7
Hz, CH2Ph), 4.99 (d, 1H, H-1), 5.10 (dd, 2H, 2J = 11.3 Hz, CH2Ph),
5.14 (d, 1H, J1′,2′ = 3.5 Hz, H-1′), 5.45 (dd, 1H, J4′,5′ = 10.0 Hz, H-4′),
6.06 (dd, 1H, J3′,4′ = 9.8 Hz, H-3′), 7.10−8.11 (m, 35H, aromatic)
ppm; 13C NMR δ 55.2, 63.0, 68.0, 68.7, 69.7, 70.5, 72.3, 72.7, 73.8,
75.3, 75.4, 75.9, 76.5, 78.4, 80.7, 94.2, 96.6, 127.5 (×2), 127.6, 127.9
(×2), 128.0 (×2), 128.1 (×2), 128.2, 128.5 (×9), 128.4 (×11), 128.6
(×7), 129.1, 129.9 (×2), 130.0 (×2), 130.1 (×2), 133.1, 133.2, 133.4,
137.8, 138.2, 138.3, 138.7, 165.6, 165.7, 166.2 ppm; HR-FAB-MS [M
+Na]+ calcd for C62H60O14Na 1051.3881, found 1051.3908.
Methyl 6-O-(3,4,6-Tri-O-benzoyl-2-O-benzyl-α-D-glucopyrano-
syl)-2,3,4-tri-O-benzyl-α-D-glucopyranoside (11). The title com-
pound was obtained by method B from 12a and 5 in 67% yield as a
colorless syrup (α/β > 25/1). Analytical data for α-11: Rf = 0.51 (ethyl
acetate−toluene, 3/17, v/v); [α]D23 +18.9 (c = 1.0, CHCl3); 1H NMR
δ 3.30 (dd, 1H, J2,3 = 9.8 Hz, H-2), 3.38 (s, 3H, OCH3), 3.49 (dd, 1H,
J4,5 = 9.4 Hz, H-4), 3.65 (dd, 1H, J2′,3′ = 10.0 Hz, H-2′), 3.69−3.81 (m,
3H, H-5, 6a, 6b), 3.93 (dd, 1H, J3,4 = 9.3 Hz, H-3), 4.20−4.34 (m, 2H,
J5,6a = 5.5 Hz, H-5, 6a), 4.48 (dd, 2H, 2J = 11.6 Hz, CH2Ph), 4.39−4.55
Methyl 4-O-(3,4,6-Tri-O-benzoyl-2-O-benzyl-α-D-glucopyrano-
syl)-2,3,6-tri-O-benzyl-α-D-glucopyranoside (22). The title com-
pound was obtained by method C from 12a and 21 in 82% yield as
a colorless syrup (α/β > 25/1). Selected analytical data for α-22: Rf =
2
(m, 3H, J1,2 = 3.7 Hz, H-1, 5′, 6b′), 4.58 (dd, 2H, J = 12.1 Hz,
CH2Ph), 4.76 (dd, 2H, 2J = 12.1 Hz, CH2Ph), 4.89 (dd, 2H, 2J = 10.9
Hz, CH2Ph), 5.01 (d, 1H, J1′,2′ = 3.4 Hz, H-1′), 5.36 (dd, 1H, J4′,5′= 9.8
Hz, H-4′), 5.89 (dd, 1H, J3′,4′ = 9.7 Hz, H-3′), 6.90−8.95 (m, 35H,
aromatic) ppm; 13C NMR δ 55.4, 63.4, 66.3, 67.9, 70.0, 70.6, 72.0,
72.3, 73.5, 75.2, 75.9, 77.4, 78.0, 80.0, 82.3, 96.8, 98.1, 127.7, 127.8,
127.9 (×2), 128.0 (×3), 128.1 (×3), 128.2 (×2), 128.4 (×2), 128.5
(×9), 128.6 (×4), 129.2, 129.8 (×2), 129.9 (×3), 130.0 (×2), 133.2
(×2), 133.5, 137.8, 138.3, 138.6, 139.0, 165.7, 165.9, 166.3 ppm; HR-
FAB-MS [M+Na]+ calcd for C62H60O14Na 1051.3881, found
1051.3904.
23
0.51 (ethyl acetate−toluene, 3/17, v/v); [α]D +42.3 (c = 1.0,
CHCl3); 1H NMR δ 3.37 (s, 3H, OCH3), 3.56 (dd, 1H, J2,3 = 9.1 Hz,
H-2), 3.62 (dd, 1H, J2′,3′ = 9.8 Hz, H-2′), 3.66 (dd, 1H, J6a,6b = 11.1 Hz,
H-6a), 3.88 (m, 1H, H-5), 4.01 (dd, 1H, H-6b), 4.04−4.14 (m, 2H, H-
3, 4), 4.18 (dd, 1H, J5′,6a′ = 5.1 Hz, J6a′,6b′ = 12.2 Hz, H-6a′), 4.28−4.39
(m, 2H, J5′,6b′ = 2.6 Hz, H-5′, 6b′), 4.31 (dd, 2H, 2J = 12.4 Hz, CH2Ph),
4.53 (s, 2H, CH2Ph), 4.59 (d, 1H, J1,22 = 3.9 Hz, H-1), 4.64 (dd, 2H, 2J
= 12.1 Hz, CH2Ph), 4.93 (dd, 2H, J = 11.6 Hz, CH2Ph), 5.40 (dd,
1H, J4′,5′ = 9.8 Hz, H-4′), 5.64 (d, 1H, J1′,2′ = 3.5 Hz, H-1′), 5.94 (dd,
1H, J3′,4′ = 9.8 Hz, H-3′), 6.59−8.01 (m, 35H, aromatic) ppm; 13C
NMR δ 55.4, 63.3, 68.5, 69.2, 69.8, 69.9, 71.8, 72.7, 73.6, 73.7, 74.4,
74.7, 77.4, 80.2, 81.7, 96.6, 98.0, 127.1 (×2), 127.2, 127.7 (×2), 127.8,
127.9, 128.0 (×2), 128.1, 128.3 (×3), 128.4 (×2), 128.5 (×7), 128.6
(×2), 128.7 (×2), 129.2 (×2), 129.8, 129.9 (×3), 130.0 (×3), 133.1,
133.2, 133.5, 137.4, 138.0, 138.2, 139.3, 165.6, 165.9, 166.2 ppm; HR-
FAB-MS [M+Na]+ calcd for C62H60O14Na 1051.3881, found
1051.3893.
Benzyl 3,4,6-Tri-O-benzoyl-2-O-benzyl-α-D-glucopyranoside (18).
The title compound was obtained by method B from 12a and benzyl
alcohol as a clear film in 75% yield. Analytical data for 18: Rf = 0.35
(ethyl acetate−toluene, 1/19, v/v); [α]D25 +29.3 (c = 1.0, CHCl3); 1H
NMR (CDCl3) δ 3.78 (dd, 1H, J2,3 = 9.9 Hz, H-2), 4.37−4.51 (m, 5H,
2
1
CH2Ph, H-5, 6a, 6b), 4.62 (d, 1H, J = 12.3 Hz, /2 CH2Ph), 4.98 (d,
2
1
1H, J1,2 = 3.5 Hz, H-1), 4.99 (d, 1H, J = 12.0 Hz, /2 CH2Ph), 5.49
(dd, 1H, J4,5 = 9.93 Hz, H-4), 6.03 (dd, 1H J3,4 = 9.7 Hz, H-3), 7.14−
8.02 (m, 25H, aromatic); 13C NMR (CDCl3) δ 31.4, 63.4, 68.1, 69.6,
70.1, 72.2, 77.4, 77.8, 95.5, 128.0 (×2), 128.1, 128.2, 128.4 (×2), 128.5
(×2), 128.6 (×5), 128.7 (×3), 128.8, 129.1, 129.9 (×3), 130.0 (×3),
130.1 (×2), 133.2, 133.3, 133.5, 136.9, 137.7, 165.7, 165.9, 166.4 ppm;
HR-FAB-MS [M+Na]+ calcd for C41H36O9Na 695.2257, found
695.2244.
Low-Temperature in Situ NMR Monitoring. A thioglycoside
(10 mg) was dissolved in CDCl3 (0.7 mL) and placed in the standard
5 mm NMR tube equipped with a septum, and the tube was then
placed into a Dewar filled with liquid nitrogen (−196 °C) for 10 min.
Bromine (1.0 equiv) was added via syringe, the NMR tube was
removed from the Dewar (upon which the temperature was allowed to
increase to ambient) and immediately inserted into the magnet, and
the proton spectrum was recorded (approximately 5 min after Br2
addition).
Methyl 3-O-(3,4,6-Tri-O-benzoyl-2-O-benzyl-α-D-glucopyrano-
syl)-2,4,6-tri-O-benzyl-α-D-glucopyranoside (19). The title com-
pound was obtained by method C from 12a and 15 in 89% yield as
a colorless syrup (α/β > 25/1). Selected analytical data for α-19: Rf =
23
0.54 (ethyl acetate−toluene, 3/17, v/v); [α]D +88.9 (c = 1.0,
ASSOCIATED CONTENT
■
CHCl3); 1H NMR δ 3.31 (s, 3H, OCH3), 3.60 (dd, 1H, J5,6a = 1.7 Hz,
J6a,6b = 10.7 Hz, H-6a), 3.66 (dd, 1H, J5,6b = 3.4 Hz, H-6b), 3.69 (dd,
1H, J2,3 = 9.8 Hz, H-2), 3.71−3.78 (m, 2H, J2′,3′ = 9.8 Hz, H-2′, 5), 3.83
(dd, 1H, J4,5 = 9.3 Hz, H-4), 4.00 (dd, 1H, J5′,6a′ = 4.7 Hz, J6a′,6b′ = 12.3
S
* Supporting Information
NMR spectra of all new compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
298
dx.doi.org/10.1021/jo2019174 | J. Org. Chem. 2012, 77, 291−299

The Journal of Organic Chemistry
Article
(31) Ber
Chem. Soc. 2001, 123, 5460−5464.
́
ces, A.; Enright, G.; Nukada, T.; Whitfield, D. M. J. Am.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: demchenkoa@umsl.edu
(32) Nukada, T.; Berces, A.; Zgierski, M. Z.; Whitfield, D. M. J. Am.
Chem. Soc. 1998, 120, 13291−13295.
(33) Stalford, S. A.; Kilner, C. A.; Leach, A. G.; Turnbull, W. B. Org.
Biomol. Chem. 2009, 7, 4842−4852.
ACKNOWLEDGMENTS
■
(34) Andersson, F.; Fugedi, P.; Garegg, P. J.; Nashed, M. Tetrahedron
Lett. 1986, 27, 3919−3922.
This work was supported by awards from the NSF (CHE-
0547566 and CHE-1058112).
(35) Nukada, T.; Berces, A.; Wang, L.; Zgierski, M. Z.; Whitfield, D.
M. Carbohydr. Res. 2005, 340, 841−852.
(36) Capon, B. Chem. Rev. 1969, 69, 407−496.
(37) Callam, C. S.; Gadikota, R. R.; Krein, D. M.; Lowary, T. L. J. Am.
Chem. Soc. 2003, 125, 13112−13119.
DEDICATION
■
Dedicated to the memory of David Gin.
(38) Crich, D.; Chandrasekera, N. S. Angew. Chem., Int. Ed. 2004, 43,
5386−5389 and references therein.
REFERENCES
■
(39) Boebel, T. A.; Gin, D. Y. J. Org. Chem. 2005, 70, 5818−5826.
(40) Nokami, T.; Nozaki, Y.; Saigusa, Y.; Shibuya, A.; Manabe, S.; Ito,
Y.; Yoshida, J. Org. Lett. 2011, 13, 1544−1547.
(41) Garcia, B. A.; Poole, J. L.; Gin, D. Y. J. Am. Chem. Soc. 1997,
119, 7597−7598.
(1) Galonic, D. P.; Gin, D. Y. Nature 2007, 446, 1000−1007.
(2) Handbook of Chemical Glycosylation: Advances in Stereoselectivity
and Therapeutic Relevance; Demchenko, A. V., Ed.; Wiley-VCH:
Weinheim, Germany, 2008.
(3) Zhu, X.; Schmidt, R. R. Angew. Chem., Int. Ed. 2009, 48, 1900−
1934.
(4) Whitfield, D. M. Adv. Carbohydr. Chem. Biochem. 2009, 62, 83−
159.
(5) Mydock, L. K.; Demchenko, A. V. Org. Biomol. Chem. 2010, 8,
(42) Zhang, Z.; Wong, C. H. Tetrahedron 2002, 58, 6513−6519.
(43) Sollogoub, M.; Das, S. K.; Mallet, J.-M.; Sinay, P. C. R. Acad. Sci.
Ser. 2 1999, 2, 441−448.
(44) Koenigs, W.; Knorr, E. Ber. Deutsch. Chem. Ges. 1901, 34, 957−
981.
(45) Fischer, E.; Armstrong, E. F. Ber. Dtsch. Chem. Ges. 1901, 34,
2885−2900.
(46) Isbell, H. S. Annu. Rev. Biochem. 1940, 9, 65−92.
(47) Igarashi, K. Adv. Carbohydr. Chem. Biochem. 1977, 34, 243−283.
(48) Helferich, B.; Bohn, E.; Winkler, S. Ber. Dtsch. Chem. Ges. 1930,
63B, 989−998.
(49) Helferich, B.; Gootz, R. Ber. Dtsch. Chem. Ges. 1931, 64B, 109−
114.
497−510.
(6) Crich, D. Acc. Chem. Res. 2010, 43, 1144−1153.
(7) Kamat, M. N.; Demchenko, A. V. Org. Lett. 2005, 7, 3215−3218.
(8) Kulkarni, S. S.; Gervay-Hague, J. In Handbook of Chemical
Glycosylation; Demchenko, A. V., Ed.; Wiley-VCH: Weinheim,
Germany, 2008, p 59−93.
(9) Lemieux, R. U.; Hendriks, K. B.; Stick, R. V.; James, K. J. Am.
Chem. Soc. 1975, 97, 4056−4062 and references therein..
(10) Demchenko, A. V. Synlett 2003, 1225−1240.
(11) Andrews, C. W.; Fraser-Reid, B.; Bowen, J. P. J. Am. Chem. Soc.
1991, 113, 8293−8298.
(12) Gervay, J.; Nguyen, T. N.; Hadd, M. J. Carbohydr. Res. 1997,
300, 119−125.
(13) Meloncelli, P. J.; Lowary, T. L. Carbohydr. Res. 2009, 344,
1110−1122.
(14) Lam, S. N.; Gervay-Hague, J. Org. Lett. 2002, 4, 2039−2042.
(15) Liptak, A.; Jodal, I.; Nanasi, P. Carbohydr. Res. 1975, 44, 1−11.
(16) Weygand, F.; Ziemann, H. Justus Liebigs Ann.Chem. 1962, 657,
179−198.
(17) Garcia, B. A.; Gin, D. Y. J. Am. Chem. Soc. 2000, 122, 4269−
4279.
(50) Helferich, B.; Wedemeyer, K. F. Justus Liebigs Ann. Chem. 1949,
563, 139−145.
(51) Garegg, P. J.; Hultberg, H. Carbohydr. Res. 1981, 93, C10−C11.
(52) van Straten, N. C. R.; Kriek, N. M. A. J.; Timmers, C. M.;
Wigchert, S. C. M.; van der Marel, G. A.; van Boom, J. H. J. Carbohydr.
Chem. 1997, 16, 947−966.
(53) Koto, S.; Uchida, T.; Zen, S. Bull. Chem. Soc. Jpn. 1973, 46,
2520−2523.
(18) Jensen, H. H.; Lyngbye, L.; Jensen, A.; Bols, M. Chem.Eur. J.
2002, 8, 1218−1226.
(19) Jensen, H. H.; Bols, M. Acc. Chem. Res. 2006, 39, 259−265.
(20) Lemieux, R. U. In Methods in Carbohydrate Chemistry; Whistler,
R. L., Wolform, M. L., Eds.; Academic Press Inc.: New York/London,
1963; Vol. 2, pp 226−228.
(21) Mydock, L. K.; Kamat, M. N.; Demchenko, A. V. Org. Lett.
2011, 13, 2928−2931.
(22) Crich, D.; Dai, Z.; Gastaldi, S. J. Org. Chem. 1999, 64, 5224−
5229.
(23) Yamago, S.; Yamada, T.; Hara, O.; Ito, H.; Mino, Y.; Yoshida, J.
Org. Lett. 2001, 3, 3867−3870.
(24) Chittenden, G. J. F. Carbohydr. Res. 1992, 242, 297−301.
(25) Ibatulin, F. M.; Selivanov, S. I. Tetrahedron Lett. 2002, 43,
9577−9580.
(26) Pozsgay, V.; Robbins, J. B. Carbohydr. Res. 1995, 277, 51−66.
(27) Kihlberg, J. O.; Leigh, D. A.; Bundle, D. R. J. Org. Chem. 1990,
55, 2860−2863.
(28) Lonn, H. J. Carbohydr. Chem. 1987, 6, 301−306.
(29) Codee, J. D. C.; Van den Bos, L. J.; Litjens, R. E. J. N.;
Overkleeft, H. S.; Van Boeckel, C. A. A.; Van Boom, J. H.; Van der
Marel, G. A. Tetrahedron 2004, 60, 1057−1064.
(30) Goodman, L. Adv. Carbohydr. Chem. Biochem. 1967, 22, 109−
175.
299
dx.doi.org/10.1021/jo2019174 | J. Org. Chem. 2012, 77, 291−299