Chiral Bronsted acid mediated glycosylation with recognition of alcohol chirality

Article abstract of DOI:10.1002/anie.201304830

Sugar sugar: In the glycosylation of racemic alcohols with 1 using the chiral phosphoric acid 2 as an activator, one enantiomer of the racemic alcohol selectively reacts with 1 to give the corresponding glycoside with good to excellent α/β-stereo- and diastereoselectivity in high yield. The reaction was successfully applied to the synthesis of a chiral natural flavan glycoside using a racemic aglycon. Copyright

Full text of DOI:10.1002/anie.201304830

Angewandte
Chemie
DOI: 10.1002/anie.201304830
Carbohydrates
Chiral Brønsted Acid Mediated Glycosylation with Recognition of
Alcohol Chirality**
Tomoya Kimura, Maiko Sekine, Daisuke Takahashi, and Kazunobu Toshima*
Many carbohydrate-containing natural products, which pos-
sess mono- and oligosaccharides, such as proteoglycans,
glycoproteins, glycolipids, and antibiotics, are found in
nature as important biological substances. A large number
of recent biological studies on these glycomolecules at the
molecular level have shed light on the biological significance
of their carbohydrate units (glycons) in molecular recognition
for the transmission of biological information.[1] It is now
recognized that carbohydrates are at the heart of a multitude
of biological events.[1] Additionally, some glycomolecules
have been developed as new functional materials.[2] For
example, certain alkyl glycosides are expected to be biode-
gradable surfactants. Therefore, glycomolecules continue to
be the central focus of much research in chemistry, biology,
and material science. With this stimulating background, the
efficient synthesis of not only the carbohydrate itself, but also
carbohydrate-containing products, is of particular interest
both in academia and in industry. In this context, glycosyla-
tion, which is a crucial organic synthetic method to attach
a sugar to other sugar moieties or other molecules (aglycons),
is becoming more and more important in synthetic organic
chemistry and carbohydrate chemistry, and considerable
attention has been directed towards the efficiency of the
glycosylation method.[3] From a synthetic standpoint, the
efficiency of the glycosylation reaction generally is evaluated
by a high chemical yield, regioselectivity, and a/b-stereose-
lectivity. Unfortunately, little attention has been focused on
the important issue of diastereoselectivity of the reaction
between the aglycon and glycon units in glycosylation. Herein
we report a novel chemical glycosylation method that alters
the chiral recognition ability of the aglycon. To the best of our
knowledge, this is the first demonstrated example of a chem-
ical glycosylation with recognition of alcohol chirality.
resolution of racemic tetrahydroberberrubine was reported
by Yu et al.[5] While our continuing efforts regarding this issue
were unsuccessful, Fairbanks et al. recently reported a/b-
stereoselective glycosylation of a galactosyl trichloroacetimi-
date and chiral alcohols using a chiral Brønsted acid.[6,7] In
that study, high b stereoselectivity, induced by a chiral
Brønsted acid, was demonstrated. In this context, we
expected that use of a chiral Brønsted acid in glycosylation
would realize a/b stereoselective and diastereoselective
chemical glycosylation by tuning the reaction conditions.
To investigate our hypothesis, we selected the glucosyl
trichloroacetimidate 1a[8] and the binol-derived Brønsted
acid 2 (Akiyama–Terada catalyst)[9] as glycosyl donor and
chiral Brønsted acids, respectively (Figure 1). In this case,
Figure 1. Structures of the glycosyl donor 1a, chiral Brønsted acids 2
and achiral acids 3 and 4.
Previously, we demonstrated an enzymatic glycosylation
of o-nitrophenyl b-d-galactoside and racemic secondary
alcohols using b galactosidase from E. coli.[4] In this study,
complete b stereoselectivity and moderate diastereoselectiv-
ity were observed. Furthermore, biocatalytic glycosylation
a benzyl group was chosen as the protecting group for 1a
because it does not exhibit the participation effect, which
would influence both a/b-stereo- and diastereoselectivity in
the glycosylation.
First, we investigated the glycosylations of 1a and the
racemic secondary alcohol (Æ)-5 using either the chiral
phosphoric acids (R)-2 and (S)-2, an achiral phosphoric acid
(3), or a typical Lewis acid, TMSOTf (4), under several
different conditions. These results are summarized in Table 1.
It was found the glycosylations of 1a and (Æ)-5 using 2–4 as
catalysts in PhMe proceeded at either 0 or À208C to give the
four corresponding glycosides 6 in moderate to high yields
(entries 1–8). When 4 was employed as the activator, a high
total yield of the four glycosides 6 was obtained with low a/b-
stereo- and diastereoselectivities (entries 1 and 5), whereas
a moderate total yield with low a/b-stereo- and diastereose-
lectivities was observed in the glycosylation using 3 (entries 2
[*] T. Kimura, M. Sekine, Dr. D. Takahashi, Prof. Dr. K. Toshima
Department of Applied Chemistry
Faculty of Science and Technology, Keio University
3-14-1 Hiyoshi, Kohoku-ku, Yokohama 223-8522 (Japan)
E-mail: toshima@applc.keio.ac.jp
[**] This research was supported in part by the Program for the Strategic
Research Foundation at Private Universities, 2012–2016, from the
Ministry of Education, Culture, Sports, Science and Technology of
Japan (MEXT).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201304830.
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
Table 1: Glycosylation reaction of 1a and (Æ)-5 several reaction conditions.
(b,R)-6 with excellent a/b-stereo-
and diastereoselectivity in high
yield.
With these favorable results in
hand, we next examined the gener-
ality of the present glycosylation
method. First we compared the
result using (S)-2 with those using
(R)-2, 3, and 4 under the optimized
reaction conditions (see Table 1).
As shown in Table 2, only the use of
(S)-2 in the glycosylation reaction
of 1a and (Æ)-5 showed excellent a/
b-stereo- and diastereoselectivity,
and gave only (b,R)-6 in high yield
(entries 1–4). In addition, when the
alcohols (Æ)-7 and (Æ)-12 were
used, similar excellent a/b-stereo-
and diastereoselectivities were
observed, and only the glycosides
(b,R)-13 and (b,R)-18, respectively,
were obtained in high yields
(entries 5 and 10). Notable the
Entry
Activator
(equiv)
Solvent
(50 mm)
T [8C]
t [h]
Yield [%][b]
(a,S)-6 (b,R)-6
(a,R)-6
(b,S)-6
1
2
3
4
5
6
7
8
4 (0.3)
3 (0.3)
(R)-2 (0.3)
(S)-2 (0.3)
4 (0.3)
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
THF
0
0
0
24
24
24
24
24
24
24
24
13
13
13
13
13
13
34
12
10
3
15
16
12
2
0
2
6
4
47
9
7
10
3
2
0
15
4
3
0
0
0
16
19
33
52
28
28
50
52
42
68
23
87
85
9
4
0
À20
À20
À20
À20
À40
À40
À40
À40
À40
À40
17
8
12
3
3 (0.3)
(R)-2 (0.3)
(S)-2 (0.3)
(S)-2 (0.3)
(S)-2 (0.3)
(S)-2 (0.3)
(S)-2 (0.3)
(S)-2 (0.3)
(S)-2 (0.6)
9[a]
10[a]
11[a]
12[a]
13[a]
14[a]
0
2
CH2Cl2
MeCN
Et2O
10
3
3
3
23
0
products possess
a b-glycosidic
3
0
bond, and the R enantiomer of the
corresponding alcohol 7 and 12.
When the alcohols (Æ)-8–11 were
employed as glycosyl acceptors, the
glycosylations of 1a using (S)-2
Et2O
0
0
[a] 5 ꢀ molecular sieves added. [b] Yield of isolated product.
and 6). Similarly, the use of (R)-2 gave a moderate total yield
of the four glycosides 6 with low a/b-stereo- and diastereo-
selectivity (entries 3 and 7). In sharp contrast, when (S)-2 was
used as the activator, although the total yield of was
moderate, the a/b-stereo- and diastereoselectivity was good
to high, and the total yield as well as the relative amount of
(b,R)-6 increased at lower temperature (entries 4 and 8).
Furthermore, the glycosylation of 1a and (Æ)-5 using (S)-2 in
the presence of 5 ꢀ molecular sieves at lower temperature
(À408C) for 13 hours afforded only (b,R)-6 in moderate yield
(52%) with complete a/b-stereo- and diastereoselectivity
(entry 9). Based on these results, we next examined the
solvent effect on the glycosylation using (S)-2 in Et2O, MeCN,
THF, and CH2Cl2. It was found that when THF was used, the
result was similar to that obtained using PhMe (entry 10 in).
In the cases of Et2O and CH2Cl2, a high total yield of the four
glycosides 6 was obtained while retaining high a/b-stereo- and
diastereoselectivities (entries 11 and 13). Interestingly, it was
found that although a/b stereoselectivity was high, diastereo-
selectivity was low for the glycosylation in MeCN (entry 12).
This phenomenon may come from the strong coordination of
MeCN[10] with an oxonium cation intermediate of the
glycosylation reaction, and thus prevents coordination of
(S)-2 to the glycosylation intermediate. Finally, we found the
use of 0.6 equivalents of (S)-2 at À408C for 13 hours in Et2O
gave the best result with high reproducibility, thus producing
only (b,R)-6 in high (85%) yield (entry 14). These results
clearly indicated that in the glycosylations of (Æ)-5 and 1a
using the chiral phosphoric acid (S)-2 as an activator, the
R enantiomer of 5 selectively reacted with 1a to give only
proceeded with excellent a/b stereoselectivity and high
diastereoselectivity to predominately give the corresponding
glycosides (b,R)-14,[11] (b,R)-15, (b,R)-16, and (b,R)-17 in
good to high yields (entries 6–8). These results clearly
indicated the high generality of the present glycosylation
method.
Next, we performed mechanistic studies of the present
glycosylation reaction (Figure 2a). It was found that when
Figure 2. Mechanistic study (a) and proposed mechanism (b) of the
glycosylation. M.S.=molecular sieves.
2
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Table 2: Glycosylations of 1a and several alcohols using (S)-2.
Et2O also gave similar excellent a/
b-stereo- and diastereoselectivity,
the yield of 22 was lower than that
obtained when using CH2Cl2. It
was also confirmed that the use of
(R)-2, 3, and 4 did not give such
high a/b-stereo- and diastereose-
lectivities as expected (see the
Supporting Information). These
results again clearly indicated the
high efficiency and generality of
the present glycosylation method.
Removal of the benzyl groups from
22 using Pd(OH)2 in MeOH/
EtOAc under an H2 atmosphere
completed the synthesis of the
flavan glycoside 19 in optically
pure form. 1H NMR, 13C NMR,
HRMS (ESI-TOF), and optical
rotation data for an analytical
sample of the synthetic 19 matched
those obtained for an authentic
sample.[14]
Entry[a]
Alcohol[b]
Activator[c]
T [8C]
t [h]
Yield [%][d]
(a,S)-6 (b,R)-6
(a,R)-6
(b,S)-6
7
2
13
0
1
2
3
4
(Æ)-5
(Æ)-5
(Æ)-5
(Æ)-5
4
3
À40
À40
À40
À40
13
13
13
13
13
11
8
17
7
5
37
10
31
85
(R)-2
(S)-2
0
0
(a,R)-13
0
(a,S)-13
(b,R)-13
(b,S)-13
0
5
(Æ)-7
(S)-2
(S)-2
(S)-2
(S)-2
(S)-2
(S)-2
À40
À40
À40
À40
À40
À40
13
13
13
13
13
13
0
87
(a,R)-14
0
(a,S)-14
(b,R)-14
(b,S)-14
13
6
(Æ)-8
0
67
(a,R)-15
0
(a,S)-15
(b,R)-15
(b,S)-15
17
7
(Æ)-9
0
75
(a,R)-16
0
(a,S)-16
(b,R)-16
(b,S)-16
8
8
(Æ)-10
(Æ)-11
(Æ)-12
0
77
(a,R)-17
0
(a,S)-17
(b,R)-17
(b,S)-17
7
In conclusion, we have devel-
oped the first a/b-stereo- and dia-
9
0
81
(a,R)-18
0
(a,S)-18
(b,R)-18
(b,S)-18
0
stereoselective
glycosylation
10
0
84
method. The use of the glucosyl
a-trichloroacetimidate 1a and the
chiral phosphoric acid (S)-2 in
Et2O or CH2Cl2 was proven to be
very effective for selective reaction
[a] 5 ꢀ molecular sieves added. [b] 2.0 equiv. [c] 0.6 equiv. [d] Yield of isolated product. THF=tetrahy-
drofuran.
(b,R)-6 was treated with (S)-2 in the absence of (Æ)-5 under
the glycosylation conditions, no isomerization occurred, and
(b,R)-6 was quantitatively recovered. This indicates that the
a/b-stereo- and diastereoselectivities were determined by
kinetic control. In addition, when the b-anomer 1b was used
as the glycosyl donor, high a/b-stereo- and diastereoselectiv-
ity was not observed, and the glycosides, (a,R)-6, (a,S)-6,
(b,R)-6, and (b,S)-6 were produced in 32, 22, 21, and 0%
yields, respectively. These results strongly suggest that the
present glycosylation of 1a using (S)-2 proceeds by a (S)-2-
mediated Sn2 reaction mechanism, which explains the high
b stereoselectivity. Although the mechanism for the high
diastereoselectivity is still not totally clear, higher stability of
the oxonium cation/(S)-2/R alcohol intermediate compared
to that of the oxonium cation/(S)-2/S alcohol complex may
induce the high diastereoselectivity (Figure 2b).
with the R enantiomer of a racemic mixture of secondary
alcohols. Furthermore, we successfully applied the present
glycosylation method to the synthesis of a natural product,
a flavan glycoside, from a racemic aglycon. These results
introduce a new and innovative strategy for the synthesis of
natural products possessing sugar(s). Detailed mechanistic
studies of the present glycosylation, applications to other
types of glycosyl donors,[15] and additional studies with respect
to the synthesis of natural products using the present
glycosylation method are currently underway in our labora-
tory.
Finally, we applied the present glycosylation method to
the synthesis of a natural product, the flavan glycoside 19. The
flavan glycoside was isolated from the bark of Cinnamomum
cassia by Morimoto and Nishioka in 1986.[12] The synthesis of
19 is summarized in Scheme 1. We first synthesized the
benzyl-protected aglycon (Æ)-21 from the benzene-1,3,5-triol
derivative 20 using OsO4 according to the modified version of
the procedure reported by Schroeter et al.[13] We then
conducted the glycosylation of 1a and (Æ)-21 using (S)-2 in
Et2O or CH2Cl2 at À408C. It was found that only the desired
glycoside 22 was obtained in 65% yield with excellent a/b-
stereo- and diastereoselectivity when using CH2Cl2. Although
Scheme 1. Total synthesis of the flavan glycoside 19 by glycosylation
using 1a, (Æ)-21, and (S)-2.
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
Winschel, A. Borovika, P. Nagorny, J. Am. Chem. Soc. 2012, 134,
8074 – 8077.
[8] a) R. R. Schmidt, J. Michael, Angew. Chem. 1980, 92, 763 – 764;
Angew. Chem. Int. Ed. Engl. 1980, 19, 731 – 732; b) R. R.
Schmidt, Angew. Chem. 1986, 98, 213 – 236; Angew. Chem. Int.
Ed. Engl. 1986, 25, 212 – 235.
[9] a) T. Akiyama, H. Morita, J. Itoh, K. Fuchibe, Org. Lett. 2005, 7,
2583 – 2585; b) M. Terada, Chem. Commun. 2008, 4097 – 4112;
c) T. Akiyama, Chem. Rev. 2007, 107, 5744 – 5758.
[10] S. Hashimoto, M. Hayashi, R. Noyori, Tetrahedron Lett. 1984, 25,
1379 – 1382.
[11] a) H. Imagawa, A. Kinoshita, T. Fukuyama, H. Yamamoto, M.
Nishizawa, Tetrahedron Lett. 2006, 47, 4729 – 4731; b) S. Hotha,
S. Kashynap, J. Am. Chem. Soc. 2006, 128, 9620 – 9621.
[12] S. Morimoto, G. Nonaka, I. Nishioka, Chem. Pharm. Bull. 1986,
34, 633 – 642.
Received: June 5, 2013
Revised: August 26, 2013
Published online: && &&, &&&&
Keywords: diastereoselectivity · glycosides · glycosylation ·
.
organocatalysis · synthetic methods
[1] a) Carbohydrates in Chemistry and Biology, Vols. 1 – 4 (Eds.: B.
Ernst, G. W. Hart, P. Sinaꢁ), Wiley-VCH, Weinheim, 2000;
b) Glycoscience, Chemistry and Chemical Biology, Vols. 1 – 3
(Eds.: B. O. Fraser-Reid, K. Tatsuta, J. Thiem), Springer, Berlin,
2001.
[2] Functional Carbohydrates (Eds.: C. G. Biriaderis, M. S. Izydor-
czyck), CPC, Boca Raton, FL, 2006.
[13] a) S.-B. Wan, D. Chen, P. Dou, T.-H. Chan, Bioorg. Med. Chem.
2004, 12, 3521 – 3527; b) P. K. Sharma, M. He, L. J. Romanc-
zyk, Jr., H. J. Schroeter, J. Labelled Compd. Radiopharm. 2010,
53, 605 – 612.
[14] Although the optical rotation value of the synthetic 19 unfortu-
nately did not match that reported in Ref. [12], the value
matched that obtained an authentic sample, which was prepared
using 1a and chiral 21 [S. Stadlbauer, K. Ohmori, F. Hattori, K.
Suzuki, Chem. Commun. 2012, 48, 8425 – 8427].
[3] a) A. V. Demchenko in Handbook of Chemical Glycosylation:
Advances in Stereoselectivity and Therapeutic Relevance (Ed.:
A. V. Demchenko), Wiley-VCH, Weinheim, 2008, pp. 1 – 27;
b) K. Toshima, K. Sasaki in Comprehensive Glycoscience (Eds.:
J. P. Karmerling, G.-J. Boons, Y. C. Lee, A. Suzuki, N. Taniguchi,
A. G. J. Voragen), Elsevier, Oxford, UK, 2007, pp. 261 – 311.
[4] S. Matsumura, H. Yamazaki, K. Toshima, Biotechnol. Lett. 1997,
19, 583 – 586.
[5] H.-X. Ge, J. Zhang, Y. Dong, K. Cui, B.-Y. Yu, Chem. Commun.
2012, 48, 6127 – 6129.
[6] D. J. Cox, M. D. Smith, A. J. Fairbanks, Org. Lett. 2010, 12, 1452 –
1455.
[15] It was found that an acyl protecting group such as the acetyl
group was not suitable for the glycosyl donor in the present
glycosylation.
[7] For chiral phosphoric acid mediated spiroacetalization, a) I.
ˇ
´
Coric, B. List, Nature 2012, 483, 315 – 319; b) Z. Sun, G. A.
4
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Communications
Carbohydrates
T. Kimura, M. Sekine, D. Takahashi,
K. Toshima*
&&&& — &&&&
Chiral Brønsted Acid Mediated
Glycosylation with Recognition of Alcohol
Chirality
Sugar sugar: In the glycosylation of
racemic alcohols with 1 using the chiral
phosphoric acid 2 as an activator, one
enantiomer of the racemic alcohol selec-
tively reacts with 1 to give the corre-
sponding glycoside with good to excellent
a/b-stereo- and diastereoselectivity in
high yield. The reaction was successfully
applied to the synthesis of a chiral natural
flavan glycoside using a racemic aglycon.
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!