S-benzoxazolyl (SBox) glycosides in oligosaccharide synthesis: Novel glycosylation approach to the synthesis of β-D-glucosides, β-D-galactosides, and α-D-mannosides

Article abstract of DOI:10.1055/s-2003-40345

Per-acetylated and per-benzoylated S-benzoxazolyl (SBox) glycosides have been synthesized and applied to highly efficient 1,2-trans glycosylation. Complete stereoselectivities and remarkably high yields were achieved for the synthesis of β-D-glucosides, β

Full text of DOI:10.1055/s-2003-40345

LETTER
1287
S-Benzoxazolyl (SBox) Glycosides in Oligosaccharide Synthesis:
Novel Glycosylation Approach to the Synthesis of b-D-Glucosides,
b-D-Galactosides, and a-D-Mannosides
S-Benzoxazolyl
Glycosid
e
O
ligos
x
a
ccharide
S
e
Department of Chemistry and Biochemistry, University of Missouri – St. Louis, 8001 Natural Bridge Rd., St. Louis, MO 63121, USA
Fax +1(314)5165342; E-mail: demchenkoa@umsl.edu
Received 31 March 2003
Dedicated to Ray Lemieux.
We describe here further evaluation of SBox glycosides
Abstract: Per-acetylated and per-benzoylated S-benzoxazolyl
for the stereoselective synthesis of 1,2-trans-linked oli-
gosaccharides. For this purpose a range of the acylated
SBox derivatives 3a–d were synthesized (Scheme 1).
(SBox) glycosides have been synthesized and applied to highly ef-
ficient 1,2-trans glycosylation. Complete stereoselectivities and re-
markably high yields were achieved for the synthesis of b-D-
glucosides, b-D-galactosides, and a-D-mannosides. It was also de- Thus, acetobromoglucose 2a was reacted with 2-mercap-
monstrated that benzoylated ‘disarmed’ SBox glycosides could
tobenzoxazole (HSBox, 1.2 equiv) in the presence of
be selectively activated in the presence of both armed and disarmed
K2CO3 (1.2 equiv) in dry acetone at 50 °C (3.5 h) to afford
ethyl thioglycosides. Convergent synthesis of the tetrasaccharide
3a9–11 in 96% yield. Prolonged reaction times were re-
b-D-Glc-(1-6)-b-D-Gal-(1-6)-b-D-GlcNAc-(1-6)-a-D-GlcOMe was
quired for the transformation of the less reactive benzo-
achieved in three sequential selective glycosylation steps.
bromo derivatives of D-glucose (2b), D-galactose (2c),
Key words: carbohydrates, stereoselective synthesis, glycosyl-
and D-mannose (2d), and, therefore, an alternative method
ation, chemoselectivity
for the synthesis of 3b–d was developed (Scheme 1). Ac-
cording to this procedure, benzobromohexoses12 2b, 2c or
2d were reacted with KSBox (1.2 equiv), synthesized
In the last decade the scientific world has witnessed ex-
from HSBox and K2CO3, in the presence of 18-crown-6 in
plosive growth in the field of glycobiology opening novel,
dry acetone at room temperature (16–48 h) to afford 3b,
3c or 3d, respectively in 90–97% yields. It should be not-
exciting frontiers of modern science.1–4 Along with this
expansion, particular interest has been drawn to complex
carbohydrates that have been found to play important
ed that mannoside 3d was isolated as an a/b-anomeric
mixture (1:1) and used for glycosylation without separa-
roles at the cellular level. These findings have eventually
tion of individual anomers. In addition, an alternative ap-
resulted in appreciation of the difficulties associated with
proach to the synthesis of SBox glycosides has been
the chemical synthesis of carbohydrate-based biomole-
cules, mainly O-linked glycoconjugates and oligosaccha-
rides. Consequently, this has stimulated the development
explored. Thus, glucose pentaacetate was reacted with
HSBox (2–3 equiv) in the presence of BF3–Et2O (2–3
equiv) to afford 3a as an anomeric mixture (a/b, 1:3.5) in
79% yield. Clearly, this transformation is of somewhat
of novel methods for stereoselective oligosaccharide syn-
thesis.5–7
lower efficiency in comparison to that shown in
As part of a program to develop novel, versatile methods
and strategies for oligosaccharide and glycoconjugate
synthesis,8,9 we have recently reported the synthesis of
two partially benzylated S-benzoxazolyl (SBox) glyco-
sides and their evaluation for 1,2-cis glycosylation.9 We
have shown that the glycosyl donors 1a and 1b
(Scheme 1) provide high yields, as well as remarkably
high 1,2-cis stereoselectivity. We have also demonstrated
that these compounds fulfill the requirements for a mod-
ern glycosyl donor, such as accessibility, fairly high sta-
bility toward protecting group manipulations, and mild
activation conditions.
Scheme 1, as prolonged reaction time or large excess of
reagents are typically required.
OR6
O
OR6
O
R4O
R3O
R4O
R3O
R3O
AcO
OAc
O
KSBox
18-cr-6
SBox
SBox
R2O
BnO
OR2
3a–d
Br
1a: R3eq=Ac (Glc)
1b: R3ax=Ac (Gal)
2a: R2eq=R3=R4eq=R6=Ac (Glc)
2b: R2eq=R3=R4eq=R6=Bz (Glc)
2c: R2eq=R3=R4ax=R6=Bz (Gal)
2d: R2ax=R3=R4eq=R6=Bz (Man)
N
Box =
O
Scheme 1 Synthesis of per-acetylated and per-benzoylated SBox
glycosides of b-D-glucose (3a and 3b), b-D-galactose (3c), and a/b-D-
mannose (3d).
Synlett 2003, No. 9, Print: 11 07 2003.
Art Id.1437-2096,E;2003,0,09,1287,1290,ftx,en;S04703ST.pdf.
© Georg Thieme Verlag Stuttgart · New York

1288
A. V. Demchenko et al.
LETTER
Having synthesized SBox glycosides 3a–d, attention was side processes, the most significant of which was found to
focused on the evaluation of their glycosyl donor proper- be hemiacetal formation, these glycosylations had to be
ties. Considering the multifunctional character of the leav- performed at lower temperature (entries 4 and 5).
ing group, we have previously shown that the SBox
moiety can be activated via a number of conceptually dif-
ferent modes.9 For example, we illustrated that AgOTf or
MeOTf serve as efficient promoters for the stereoselective
1,2-cis glycosylation with electronically ‘armed’ glycosyl
donors.
O
OR
O
OBn
O
BzO OR
O
Ph
O
O
RO
BnO
O
OMe
O
RO
BzO
BnO
BnO
O
BzO
O
OMe
OMe
6: R=H
13: R=R1
7: R=H
14: R=R1
8: R=H
15: R=R1
9: R=H
16: R=R1
Ph
To evaluate the applicability of these activation condi-
tions for electronically deactivated (‘disarmed’) benzoy-
lated glycosyl donors, we performed a number of test
glycosylations of the glycosyl acceptor 4.13 Disappoint-
ingly, the glycosidation of 3a was greatly compromised
due to undesired O-acetylation of 4 via a competing acetyl
migration process.14,15 As a result, coupling yields be-
tween 3a and 4 were typically in the range of 40–50%. In
contrast, per-benzoylated glucosyl donor 3b has proven to
be very efficient. For example, AgOTf or MeOTf-promot-
ed glycosylation of 4 provided the disaccharide 5b in 94%
or 95% yield respectively (entries 1 and 2, Table 1). It is
noteworthy that the glycosylations in the presence of Ag-
OTf were typically completed in minutes, whereas com-
pletion of the MeOTf-promoted coupling required at least
1 h. NIS/TfOH-promoted glycosylation in the absence of
the molecular sieves was also proven to be feasible; how-
ever, a somewhat lower yield of the disaccharide 5b was
achieved 86% (entry 3).
OBn
O
R1
BzO
=
BzO
BzO
OR
O
O
OBz
OBn
O
O
RO
BnO
O
SEt
OTE
BzO
BnO
BzO
BzO
OR
SEt
10: R=H
17: R=R1
11: R=H
18: R=R1
12: R=H
19: R=R1
TE=CH2CH2SiMe3
Figure 1 Glycosyl acceptors 6–12 and the corresponding disaccha-
rides 13–19 (see also Table 2).
Being encouraged by these promising results, we decided
to explore this concept for the glycosylation of a range of
glycosyl acceptors. Thus, commercially available 7, and
other common building blocks 6,16 8,17 9,18 10,19 11,20 and
1221 were employed for this purpose (Figure 1).
As evident from the experiments summarized in Table 2,
this novel methodology can be successfully applied to the
glycosylation of both highly reactive primary and sterical-
ly hindered secondary glycosyl acceptors. A number of
Table 1 Glycosylation Reactions between 3b–d and 4
Table 2 Glycosylation Reactions between 3b and 6–12
OBz
O
BzO
Entry Acceptor Conditionsa Time
Product Yield (%)
OBz
O
OH
O
O
BzO
BzO
BzO
BnO
BnO
BzO
BnO
BnO
SBox
+
1
2
6
6
AgOTf
MeOTf
AgOTf
MeOTf
AgOTf
MeOTf
AgOTf
MeOTf
AgOTf
AgOTfc
MeOTf
AgOTf
AgOTf
10 min
1 h
13
13
14
14
15
15
16
16
17
17
17
18
19
85
92
91
88
85
92
66b
86
40b
78
94
99
96
O
BnO
BzO
OMe
BnO
OMe
3b: β-D-Glc
3c: β-D-Gal
3d: α,β-D-Man
4
5b–d
3
7
10 min
30 min
1 h
Entry Donor Conditionsa
Time
Product Yield
(%)
4
7
5
8
1
2
3
4
5
6
7
3b
3b
3b
3c
3c
3d
3d
AgOTf, MS 3 Å, r.t.
MeOTf, MS 3 Å, r.t.
NIS/TfOH, r.t.
5 min
1 h
b-5b 94
b-5b 95
b-5b 86
6
8
16 h
7
9
<5 min
15 min
<5 min
30 min
1 h
30 min
5 min
5 h
8
9
AgOTf, MS 3 Å, 0 °C
MeOTf, MS 3 Å, 0 °C
AgOTf, MS 3 Å, r.t.
MeOTf, MS 3 Å, r.t.
b-5c
b-5c
92
78
9
10
10
10
11
12
10
11
12
13
<5 min
1 h
a-5d 76
a-5d 92
5 min
5 min
a The glycosylations were performed in 1,2-dichloroethane under
argon atmosphere.
a Unless otherwise noted, the glycosylation were performed in the
presence of molecular sieves 3 Å in 1,2-dichloroethane at r.t. under
argon atmosphere.
In this context, glycosyl donors 3c and 3d could be ap-
plied with similar efficiency to galactosylation and man-
nosylation respectively (entries 4–7). It should be noted
that the galactosyl donor 3c appears to be more reactive
than 3b. In order to achieve better control over competing
b Acid-labile protecting groups cleavage was found to have occurred
subsequent to the glycosylation step.
c The reaction was performed at 0 °C.
Synlett 2003, No. 9, 1287–1290 ISSN 1234-567-89 © Thieme Stuttgart · New York

LETTER
S-Benzoxazolyl Glycosides in Oligosaccharide Synthesis
1289
OH
O
experiments afforded lower yields, most likely due to the
marginal stability of the acid-labile moieties, e.g. TMS-
ethyl, in the presence of AgOTf (entry 9). This result was
significantly improved when the coupling reaction was
performed at 0 °C (entry 10). In addition, the distinctive
reaction conditions required for the activation of SBox
glycosides allowed selective glycosidation of 3b with
both disarmed and armed thioglycosides 11 and 12 afford-
ing the corresponding disaccharides 18 and 19 in 99% and
96% yields respectively (entries 13 and 14).
HO
O
BzO
18
PhthN
20
90%
MeOTf
OBz
O
BzO
O
BzO
OBz
O
BzO
BzO
BzO
BzO
O
O
BzO
BzO
4
O
BzO
NIS/TfOH
73%
BzO
HO
O
O
BzO
O
BzO
O
BzO
PhthN
O
HO
BzO
More importantly, disaccharides 18 and 19 could be di-
rectly employed for subsequent glycosylations. As can be
envisaged, the application of MeOTf as the thioglycoside
activator will allow glycosylation of partially protected O-
pentenyl glycosides in accordance with the semi-orthogo-
nal glycosylation strategy recently developed in our labo-
ratory.8 To support this statement we performed the
regioselective glycosylation of the building block 2022
with 18 in the presence of MeOTf to afford the tri-
saccharide derivative 21 in 90% yield (Scheme 2). Subse-
quently, the O-pentenyl function of 21 was activated for
the glycosylation of 4 in the presence of NIS/TfOH to
provide tetrasaccharide 22 in 73% yield (Scheme 2).23
While application of S-alkyl/aryl24,25 and O-pentenyl
glycosides26,27 for oligosaccharide synthesis is well appre-
ciated, the use of novel SBox glycosides is still under con-
sideration. The latter synthesis however, serves as a clear
indication that these potent glycosyl donors may occupy
an important niche in the arsenal of modern glycosylation
methods.
O
O
BnO
BnO
PhthN
BnO
OMe
22
21
Scheme 2 Rapid assembly of the tetrasaccharide 22 from the
building blocks 4, 18, 20, and 21.
vergent oligosaccharide synthesis. This synthesis was
accomplished via a three-step glycosylation approach in-
cluding two crucial steps, selective activation of SBox
functionality over S-ethyl (AgOTf), followed by activa-
tion of the S-ethyl over the O-pentenyl moiety (MeOTf).
Acknowledgement
This work was supported by the University of Missouri Research
Board Award and the UM – St. Louis, and the ‘The Foundation
BLANCEFLOR Boncompagni-Ludovisi, née Bildt.’
References
In a typical experiment a mixture of an SBox glycosyl do-
nor (0.11 mmol), glycosyl acceptor (0.10 mmol), and
freshly activated molecular sieves (3 Å, 200 mg) in 1,2-
dichloroethane (2 mL) was stirred for 2 h under an atmo-
sphere of argon. The reaction mixture was cooled down to
0 °C (if appropriate, see Tables 1 and 2), promoter MeOTf
(0.33 mmol) or AgOTf (0.22 mmol) was added and the re-
action mixture was stirred until the reaction reached com-
pletion (monitored by TLC). The mixture was diluted
with CH2Cl2, the solid was filtered-off and washed with
CH2Cl2. The combined filtrate was washed with 20% aq
NaHCO3, water, and the organic phase was separated,
dried and concentrated in vacuo. The residue was purified
by column chromatography on silica gel. All synthesized
compounds have unambiguous 1H, 13C NMR, and HRMS
data.
(1) Dwek, R. A. Chem. Rev. 1996, 96, 683.
(2) Essentials of Glycobiology; Varki, A.; Cummings, R.; Esko,
J.; Freeze, H.; Hart, G.; Marth, J., Eds.; Cold Spring Harbor
Laboratory Press: Cold Spring Harbor N. Y., 1999.
(3) Bertozzi, C. R.; Kiessling, L. L. Science 2001, 291, 2357.
(4) Ritchie, G. E.; Moffatt, B. E.; Sim, R. B.; Morgan, B. P.;
Dwek, R. A.; Rudd, P. M. Chem. Rev. 2002, 102, 305.
(5) Toshima, K.; Tatsuta, K. Chem. Rev. 1993, 93, 1503.
(6) Nicolaou, K. C.; Mitchell, H. J. Angew. Chem. Int. Ed. 2001,
40, 1576.
(7) Demchenko, A. V. Curr. Org. Chem. 2003, 7, 35.
(8) Demchenko, A. V.; De Meo, C. Tetrahedron Lett. 2002, 43,
8819.
(9) Demchenko, A. V.; Malysheva, N. N.; De Meo, C. Org. Lett.
2003, 5, 455.
(10) Zinner, H.; Pfeifer, M. Berichte 1963, 96, 432.
(11) Zinner, H.; Peseke, K. Chem. Ber. 1965, 98, 3515.
(12) Fletcher, H. G. In Methods in Carbohydrate Chemistry, Vol.
2; Whistler, R. L.; Wolform, M. L., Eds.; Academic Press
Inc.: New York, London, 1963, 226–228.
(13) Kuester, J. M.; Dyong, I. Justus Liebigs Ann. Chem. 1975,
2179.
(14) Betaneli, V. I.; Kryazhevskikh, I. A.; Ott, A. Y.; Kochetkov,
N. K. Bioorg. Khimiya 1989, 15, 217.
In conclusion, we have demonstrated that per-benzoylated
SBox glycosides serve as efficient glycosyl donors for ste-
reoselective 1,2-trans glycosylations. This statement is
substantiated by successful synthesis of a range of disac-
charide derivatives. Compounds 5b–d, 13–19 were ob-
tained by glycosylation of either highly reactive primary
(6, 7, 11) or sterically hindered secondary glycosyl accep-
tors (8–10, and 12). The synthesis of linear tetrasaccharide
22 with the core structure b-D-Glc-(1-6)-b-D-Gal-(1-6)-b-
D-GlcNAc-(1-6)-a-D-GlcOMe was performed to evaluate
the applicability of the developed methodology to con-
(15) Tsvetkov, Y. E.; Kitov, P. I.; Backinowsky, L. V.;
Kochetkov, N. K. Bioorg. Khimiya 1990, 16, 98.
(16) Valashek, I. E.; Shakhova, M. K.; Minaev, V. A.;
Samokhvalov, G. I. Zh. Obshch. Khim. 1974, 44, 1161.
(17) Garegg, P. J.; Hultberg, H. Carbohydr. Res. 1981, 93, C10.
(18) Kondo, Y. Agric. Biol. Chem. 1975, 39, 1879.
Synlett 2003, No. 9, 1287–1290 ISSN 1234-567-89 © Thieme Stuttgart · New York

1290
A. V. Demchenko et al.
LETTER
(19) Jansson, K.; Ahlfors, S.; Frejd, T.; Kihlberg, J.; Magnusson,
G.; Dahmen, J.; Noori, G.; Stenvall, K. J. Org. Chem. 1988,
53, 5629.
3.34 (dd, 1 H, J = 3.1 Hz), 3.28 (dd, 1 H, J = 3.4 Hz, 9.6 Hz),
3.13 (m, 1 H), 3.06 (s, 3 H) ppm; 13C NMR: 166.94, 166.34,
166.30, 165.76, 165.59, 165.41, 165.30, 165.30, 127.70–
138.98 (72 signals), 101.97, 101.38, 98.24, 98.12, 81.99,
79.99, 75.80, 75.25, 74.88, 74.28, 73.57, 73.19, 72.51,
71.94, 71.80, 70.58, 70.06, 69.62, 69.62, 69.42, 69.25,
69.04, 68.26, 68.07, 62.93, 55.30, 54.59, 29.92 ppm; HR-
FAB MS [M + Na]+ calcd for C110H97NNaO30 1934.5993,
found 1934.6022.
(20) Birberg, W.; Lonn, H. Tetrahedron Lett. 1991, 32, 7453.
(21) Garegg, P. J.; Kvarnstrom, I.; Niklasson, A.; Niklasson, G.;
Svensson, S. C. T. J. Carbohydr. Chem. 1993, 12, 933.
(22) Building block 20 was obtained by sequential benzoylation
(BzCl/pyridine) and benzylidene acetal removal (TFA/H2O/
CH2Cl2, 20:1:79, v/v/v) of pent-4-enyl 2-deoxy-4,6-O-
benzylidene-2-phthalimido-b-D-glucopyranoside described
in: Madsen, R.; Udodong, U. E.; Roberts, C.; Mootoo, D. R.;
Konradsson, P.; Fraser-Reid, B. J. Am. Chem. Soc. 1995,
117, 1554.
(24) Garegg, P. J. Adv. Carbohydr. Chem. Biochem. 1997, 52,
179.
(25) Oscarson, S. In Carbohydrates in Chemistry and Biology,
Vol. 1; Ernst, B.; Hart, G. W.; Sinay, P., Eds.; Wiley-VCH:
Weinheim, 2000, 93–116.
(26) Fraser-Reid, B.; Madsen, R. In Preparative Carbohydrate
Chemistry; Hanessian, S., Ed.; Marcel Dekker Inc.: New
York, 1997, 339.
(27) Fraser-Reid, B.; Anilkumar, G.; Gilbert, M. B.; Joshi, S.;
Kraehmer, R. In Carbohydrates in Chemistry and Biology,
Vol. 1; Ernst, B.; Hart, G. W.; Sinay, P., Eds.; Wiley-VCH:
Weinheim, 2000, 135–154.
(23) Selected analytical data for 22: 1H NMR: 6.87–8.01 (m, 59
H), 5.80 (dd, 1 H, J = 9.5 Hz), 5.72–5.76 (m, 2 H), 5.66 (dd,
1 H, J = 7.8 Hz , 10.4 Hz), 5.6 (dd, 1 H, J = 8.4 Hz), 5.46 (dd,
1 H, J = 7.9 Hz, 9.7 Hz), 5.40 (dd, 1 H, J = 3.5 Hz, 10.4 Hz),
5.30 (d, 1 H, J = 8.4 Hz), 5.01 (d, 1 H, J = 7.8 Hz), 4.67 (dd,
2 H, J = 10.9 Hz), 4.66 (d, 1 H, J = 7.9 Hz), 4.56 (dd, 2 H, J
= 12.2 Hz), 4.52 (dd, 1 H, J = 3.2 Hz), 4.26–4.36 (m, 3 H),
3.87–4.24 (m, 7 H), 3.62–3.77 (m, 4 H), 3.42–3.48 (m, 2 H),
Synlett 2003, No. 9, 1287–1290 ISSN 1234-567-89 © Thieme Stuttgart · New York