An improved method for synthesizing antennary β-D-mannopyranosyl disaccharide units

Article abstract of DOI:10.1016/j.tetlet.2004.09.034

The merits of an indirect protecting method for hydroxyl groups using allyl groups via allyloxycarbonyl groups in the synthesis of antennary β-D-mannopyranosyl disaccharides from β-D-galactopyranosyl disaccharides were studied. Regioselective allyloxycarbonylation and conversion reactions involving simultaneous double S_N2 nucleophilic substitution at C-2′ and C-4′ of benzyl O-[β-D-galactopyranosyl]-(1-4)-3,6- di-O-benzyl-2-deoxy-2-N-phthalimido-β-D-glucopyranoside were examined for comparison with the direct allylation method. The required β-D- mannopyranosyl disaccharide having proper protecting groups was obtained using this indirect method in 52% yield. In contrast, the reported direct allylation method using methyl O-(β-D-galactopyranosyl) disaccharide gave the corresponding β-D-mannopyranosyl disaccharide in only 7.5% yield.

Full text of DOI:10.1016/j.tetlet.2004.09.034

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 8199–8201
An improved method for synthesizing antennary
b-D-mannopyranosyl disaccharide units
Ken-ichi Sato,* Shoji Akai, Akira Yoshitomo and Yoshimitsu Takai
Laboratory of Organic Chemistry, Faculty of Engineering, Kanagawa University, 3-27-1, Rokkakubashi,
Kanagawa-ku, Yokohama 221-8686, Japan
Received 20 July 2004; revised 30 August 2004; accepted 3 September 2004
Available online 22 September 2004
Abstract—The merits of an indirect protecting method for hydroxyl groups using allyl groups via allyloxycarbonyl groups in the
synthesis of antennary b-D-mannopyranosyl disaccharides from b-D-galactopyranosyl disaccharides were studied. Regioselective
allyloxycarbonylation and conversion reactions involving simultaneous double SN2 nucleophilic substitution at C-20 and C-40 of
benzyl O-[b-D-galactopyranosyl]-(1-4)-3,6-di-O-benzyl-2-deoxy-2-N-phthalimido-b-D-glucopyranoside were examined for compari-
son with the direct allylation method. The required b-D-mannopyranosyl disaccharide having proper protecting groups was
obtained using this indirect method in 52% yield. In contrast, the reported direct allylation method using methyl O-(b-D-galactopy-
ranosyl) disaccharide gave the corresponding b-D-mannopyranosyl disaccharide in only 7.5% yield.
ꢀ 2004 Elsevier Ltd. All rights reserved.
The practical synthesis of antennary b-D-mannopyranosyl
oligosaccharides is an important subject in bioactive
oligosaccharide synthesis. The difficulties of the neces-
sary 1,2-cis coupling are well known. To date, several
papers1 on the direct construction of the b-D-manno
structure have been reported, but obtaining the proper
protecting groups at the required positions of the
intermediate compounds has proven to be a difficult
problem. Several indirect methods involving SN2 nucleo-
philic substitution aimed at synthesizing a suitably pro-
tected antennary b-D-mannopyranosyl unit have shown
promise. David et al. has reported2 a mode of building
the very important b-mannopyranosyl residue by inver-
sion of C-2, thereby bypassing the difficult step. Further-
more, Alais and David have also developed3 a method
for preparing disaccharides containing b-D-mannopy-
ranosyl groups starting from N-phthaloyllactosamine
derivatives by two simultaneous SN2 substitutions. The
authors were inspired by the work of David on con-
structing b-D-mannopyranosyl residues and have devel-
oped a practical method based on that work.4
lectively. In this case, the regioselectivities and the
yields are usually affected by the reaction conditions
and the kind of protecting groups used. In the syntheses
of antennary b-D-mannopyranosyl oligosaccharide
units, the b-D-galactopyranosyl residue appears to be a
better substrate than b-D-glucopyranosyl due to the abil-
ity to regioselectively protect the branching positions at
3 and 6 followed by SN2 substitutions at the 4 and then
the 2-positions (from galacto to gluco and then manno
derivatives).4 In previous work, Sato et al. have exam-
ined the effect of protecting groups at the 3- and 6-posi-
tions of b-D-galactosides on these double SN2
substitution reactions using benzyl 3,6-di-O-acyl-
or 3,6-di-O-allyl- and -O-benzyl-2,4-bis(O-trifluoro-
methansulfonyl)-b-D-galactopyranoside with CsOAc
and found that 3,6-di-O-acyl derivatives give better
yields than 3,6-di-O-allyl- or -benzyl derivatives.5 In
contrast, it has been reported that the regioselective pro-
tection of the 3- and 6-positions of the b-D-galactopy-
ranosyl disaccharide, methyl 3,6-di-O-benzyl-2-deoxy-
4-O-b-D-galactopyranosyl-2-phthalimido-b-D-glucopyr-
anoside (5) with allyl groups using bis(tributyltin) oxide
give the expected 30,60-di-O-allyl derivative (6) in only
16% yield despite otherwise typically good results for
regioselective protection using the tin oxide method.6
This decreased nucleophilicity has been reported3 to be
a property of disaccharides, especially toward organotin
reagents. Considering the above results, we wanted to
develop a new methodology that would overcome the
In oligosaccharide syntheses, it is an important chal-
lenge to protect the individual hydroxyl group regiose-
Keywords: Carbohydrates; b-Mannopyranoside; Oligosaccharide syn-
thesis; Protecting groups.
*
Corresponding author. Tel.: +81 45 481 5661x3853; fax: +81 45 413
9770; e-mail: satouk01@kanagawa-u.ac.jp
0040-4039/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.09.034

8200
K. Sato et al. / Tetrahedron Letters 45 (2004) 8199–8201
1
difficulties of synthesizing b-D-mannopyranosyl disac-
charide units. We wanted to develop an indirect method
involving as the key step, the transformation of allyloxy-
carbonyl protecting groups to allyl groups (Scheme 1).7
In this paper, we describe the merits of this in-
direct method for synthesizing b-mannohexopyranosyl
disaccharides, using the synthesis of benzyl O-[3,6-di-
O-allyl-b-D-mannopyranosyl]-(1-4)-3,6-di-O-benzyl-2-
deoxy-2-N-phthalimido-b-D-glucopyranoside (10) via
benzyl O-[3,6-di-O-allyloxycarbonyl-b-D-galactopyrano-
syl]-(1-4)-3,6-di-O-benzyl-2-deoxy-2-N-phthalimido-
D-glucopyranoside (7) as an example.
tate = 2:1). The structure of 7 was determined by H
NMR10 (down field shift of H-30, H-60a, and H-60b).
Compound 7 was then treated with 3.0equiv of Tf2O
in pyridine–CH2Cl2 (6:1) at ꢁ19ꢁC then room tempera-
ture under argon to give syrupy 20,40-bis(triflate) 8 quan-
1
titatively, the structure of which was determined by H
NMR10 (down field shift of H-20 and H-40). The labile
compound 8 was directly used for the following reaction
without purification (Scheme 2).
A simultaneous double SN2 substitution of 20,40-bis(tri-
flate)
8 with CsOAc (3.0equiv) and 18-crown-6
(3.0equiv) in toluene with ultrasonication4 in a water
bath was carried out until the disappearance of 8
The starting material, benzyl O-[2,3,4,6-tetra-O-acetyl-
b-D-galactopyranosyl]-(1-4)-3,6-di-O-benzyl-2-deoxy-2-
N-phthalimido-b-D-glucopyranoside (3) was prepared
by the method described by Alais and David.3 Coupling
of benzyl 3,6-di-O-benzyl-2-deoxy-2-N-phthalimido-b-
D-glucopyranoside (2)8 with 1,2,3,4,6-penta-O-acetyl-b-
D-galactpyranose (1) in the presence of trimethylsilyl
(for 12h) gave the corresponding syrupy 20,40-di-O-ace-
27
tyl-b-D-mannopyranosyl derivative 9 [½aꢀ ꢁ11 (c 0.4,
D
CHCl3)] in 86% yield, which was purified on a column
of silica gel (hexane/ethyl acetate = 2/1). The structure
1
of 9 was determined by H NMR10 (8: J1 ,2
0
0
=
7.9Hz, J2 ,3 = 10.4Hz, J3 ,4 = 3.7Hz, J4 ,5 = 0Hz; 9:
0
0
0
0
0
0
triflate as the promoter9 gave the protected lactosamine
J1 ,2 =0Hz, J2 ,3 = 3.7Hz, J3 ,4 = 9.7Hz, J4 ,5 = 9.7Hz).
In a similar reaction, 20,40-bis(trifrate) of methyl
3,6-di-O-benzyl-2-deoxy-4-O-(3,6-di-O-allyl-b-D-galacto-
0
0
0
0
0
0
0
0
25
D
derivative 3 [mp 194–202ꢁC; ½aꢀ +18 (c 1.0, CHCl3)] in
72% yield. This was de-O-acetylated by alkaline met-
hanolysis to give disaccharide 4 in quantitative yield,
which has an unprotected b-D-galactopyranosyl group.
The reaction of 4 and bis(tributyltin)oxide (1.5equiv)
in toluene under reflux conditions by using a Dean Stark
apparatus with molecular sieves (MS 4A) for 3h gave
the corresponding (tributyltin)oxide derivative. This
reaction mixture was treated with allyloxycarbonyl chlo-
ride (3.0equiv) at room temperature until the disappear-
ance of the starting material (for 24h) gave the
pyranosyl)-2-phthalimido-b-D-glucopyranoside
with
tetrabutylammonium benzoate has been reported3 to
give the corresponding 20,40-di-O-benzoyl mannopyran-
osyl disaccharide 11 in 47% yield. From the results, the
double SN2 substitution of these 20,40-bis(triflate) com-
pounds with cesium acylate appears to provide better
yields than that with the use of tetrabutylammonium
acylate. It is also possible to obtain a mono-substituted
b-D-glucopyranosyl product if we want, because this
reactions proceeded firstly at C-40. These results
may promise the wide applications in the syntheses of
b-D-mannopyranosyl as well as b-D-glucopyranosyl
corresponding 30,60-di-O-allyloxycarbonyl derivative 7
25
D
purified on a column of silica gel (hexane/ethyl ace-
[syrup; ½aꢀ +37 (c 0.9, CHCl3)] in 84% yield. It was
O
O
O
O
OAll
2% Pd(OAc)2, PPh3 / Benzene
y. 82%
C
O
C
O
OCOAll
O
Scheme 1. A reported conversion method of allyloxycarbonyl groups into allyl ether groups.7
R1O
R2O
AcO
AcO
OAc
O
OR2
NPhth
O
NPhth
O
O
a
BnO
HO
BnO
O
OBn
OR
OAc
OR1
OAc
OBn
OBn
1
2
3 : R1 = R2 = Ac, R = Bn
4 : R1 = R2 = H, R = Bn
5 : R1 = R2 = H, R = Me
6 : R1 = H, R2 = All, R = Me
7 : R1 = H, R2 = AllOCO, R = Bn
8 : R1 = Tf, R2 = AllOCO, R = Bn
b
c
d
OR1
O
R2O
NPhth
O
BnO
R1O
OR
O
R2O
OBn
e
f
8
9 R1 = Ac, R2 = AllOCO, R = Bn
10 R1 = Ac, R2= All, R = Bn
11 R1 = Bz, R2 = All, R = Me
Scheme 2. Reagents and conditions: (a) TMSOTf/CH2Cl2, ꢁ25ꢁC to rt (72%); (b) NaOMe/MeOH, 0ꢁC (quant.); (c) (i) (Bu3Sn)2O/toluene, reflux; (ii)
AllOCOCl/toluene, rt (84%); (d) Tf2O/Py.–CH2Cl2, ꢁ19ꢁC to rt (quant.); (e) CsOAc, 18-crown-6/toluene, ultrasonication (86%); (f) (Ph3P)4Pd/
benzene, ꢁ19ꢁC to rt (72%).

K. Sato et al. / Tetrahedron Letters 45 (2004) 8199–8201
8201
12.8Hz, –CH2Ph), 4.78, 4.60 (1H · 2, each d, JA,B
=
disaccharides having a variety of functional groups at
C-20.4 Allyloxycarbonyl derivative 9 was then trans-
formed with (Ph3P)4Pd7 into the corresponding allyl
ether derivative 10 in 72% yield. The structure of 10
12.8Hz, –CH2Ph), 4.78, 4.45 (1H · 2, each d,
JA,B = 12.8Hz, –CH2Ph), 4.64 (1H, d, H-10, J1 ,2
0
0
=
7.3Hz), 4.66, 4.45 (2H · 2, each m, –CH2–CH@CH2 · 2),
4.60 (1H, dd, H-30, J2 ,3 = 11.0Hz, J3 ,4 = 2.3Hz), 4.36
(1H, dd, H-3, J2,3 = 10.4Hz, J3,4 = 8.6Hz), 4.25 (1H, dd,
0
0
0
0
1
was supported by H NMR10 (high field shift of H-30,
H-60a, and H-60b). Compound 10 may be useful for syn-
thesizing antennary mannopyranosyl oligosaccharides,
because it is possible to remove each protecting group
individually.
H-60a, J5 ,6 a = 5.2Hz, J6 a,6 b = 11.3Hz), 4.22 (1H, dd,
0
0
0
0
H-60b, J5 ,6 b = 7.9Hz), 4.20 (1H, dd, H-2), 4.14 (1H, dd,
0
0
H-4, J4,5 = 9.8Hz), 4.06 (1H, dd, H-40, J4 ,5 = 0Hz,
0
0
0
J4 ,OH
=
4.9Hz), 4.03 (1H, dd, H-6a, J5,6a = 3.1Hz,
J6a,6b = 11.6Hz), 3.93 (1H, ddd, H-20, J2 ,OH = 3.7Hz),
3.80 (1H, dd, H-6b, J5,6b = 1.8Hz), 3.72 (1H, d, H-20OH),
3.67 (1H, ddd, H-5), 3.55 (1H, ddd, H-50), 2.49 (1H, d,
40-OH).
0
In summary, the combination of indirect allylation and
a double SN2 substitution of the 20,40-bis(triflate) deriva-
tive of b-D-galactopyranosyl disaccharide using CsOAc
was shown to work well for the synthesis of b-D-manno-
pyranosyl disaccharides, which form the core of anten-
nary b-D-mannopyranosyl oligosaccharides. The ease
of reaction and the dramatically improved yields, should
make this a very useful method for performing this and
similar kinds of oligosaccharide syntheses.
Compound 8:d 7.78–6.83 (5H · 3 and 4H, m, Ph · 3 and
Phth), 5.98, 5.93 (1H · 2, each m, –CH2–CH@CH2 · 2),
5.42, 5.33 (1H · 2, each m, –CH2–CH@CH2), 5.41, 5.33
(1H · 2, each m, –CH2–CH@CH2), 5.19 (1H, d, H-
40,J4 ,3 = 3.1Hz), 5.13 (1H, d, H-1, J1,2 = 8.2Hz), 4.90,
4.34 (1H · 2, each d, JA,B = 11.9Hz, –CH2Ph), 4.81, 4.48
(1H · 2, each d, JA,B = 12.5Hz, –CH2Ph), 4.73–4.67
(2H · 2, each m, –CH2–CH@CH2 · 2), 4.65 (1H, dd, H-
20), 4.64, 4.32 (1H · 2, each d, JA,B = 11.9Hz, –CH2Ph),
0
0
4.58 (1H, dd, H-30, J3 ,2 = 10.4Hz, J3 ,4 = 3.1Hz), 4.49
0
0
0
0
Acknowledgements
(1H, d, H-10, J1 ,2 = 7.6Hz), 4.23 (1H, dd, H-3,
J2,3 = 6.7Hz), 4.22 (1H, dd, H-2), 4.20 (1H, dd, H-60a),
4.17 (1H, dd, H-4, J4,3 = 9.8Hz), 3.92 (1H, dd, H-6a,
J6a,6b = 11.3Hz, J6a,5 = 2.8Hz), 3.82 (1H, dd, H-60b,
0
0
This work was partially supported by a ÔHigh-Tech Re-
search Center ProjectÕ and a Grant-in-Aid (15750148)
for Scientific Research from the Ministry of Education,
Science, Sports and Culture, Japan. The authors thank
Professor T. Nakagawa for helpful discussions.
0
0
0
0
J6 b,6 a = 11.3Hz, J6 b,5 = 8.2Hz), 3.79 (1H, dd, H-6b,
J6b,5 = 1.2Hz), 3.59 (1H, dd, H-5, J5,4 = 9.8Hz), 3.46 (1H,
dd, H-50, J5 ,6 a = 6.4Hz).
0
0
Compound 9: d 7.65 (4H, m, Phth), 7.43–6.76 (5H · 3, m,
Ph · 3), 5.92, 5.87 (1H · 2, each m, –CH2–CH@CH2 · 2),
5.48 (1H, dd, H-20, J1 ,2 = 0.9Hz, J2 ,3 = 3.4Hz), 5.35,
5.28 (1H · 2, each m, –CH2–CH@CH2), 5.31, 5.22
(1H · 2, each m, –CH2–CH@CH2), 5.17 (1H, dd, H-40,
References and notes
0
0
0
0
1. For example: (a) Nagai, H.; Matsumura, S.; Toshima, K.
Carbohydr. Res. 2003, 338, 1531–1534; (b) Tsuda, T.; Sato,
S.; Nakamura, S.; Hashimoto, S. Heterocycles 2003, 59,
509–515; (c) Aloui, M.; Chambers, D. J.; Cumpstey, I.;
Fairbanks, A. J.; Seward, C. M. P. Chem. Eur. J. 2002, 8,
2608–2621; (d) Kim, K. S.; Jin, H.; Lee, Y. J.; Lee, Y. J.;
Park, J. J. Am. Chem. Soc. 2001, 123, 8477–8481; (e)
Plante, O. J.; Palacci, E. R.; Seeberger, P. H. Org. Lett.
2000, 2, 3841–3843; (f) Gridly, J. J.; Osborn, H. M. I. J.
Chem. Soc., Perkin Trans. 1 2000, 1471–1491.
2. David, S.; Malleron, A.; Dini, C. Carbohydr. Res. 1989,
188, 193–200.
3. Alais, J.; David, S. Carbohydr. Res. 1990, 201, 69–77.
4. Sato, K.; Yoshitomo, A.; Takai, Y. Bull. Chem. Soc. Jpn.
1997, 70, 885–890.
5. Sato, K.; Seki, H.; Yoshitomo, A.; Nanaumi, H.; Takai,
Y.; Ishido, Y. J. Carbohydr. Chem. 1998, 17, 703–727.
6. (a) Ogawa, T.; Matsui, M. Carbohydr. Res. 1977, 56, C1–
C6; (b) Ogawa, T.; Nukada, T.; Matsui, M. Carbohydr.
Res. 1982, 101, 263–270.
0
0
0
0
J
3 ,4 J4 ,5 = 10.1Hz), 5.05 (1H, d, H-1, J1,2@8.2Hz), 4.81,
4.37 (1H · 2, each d, JA,B = 12.5Hz, –CH2Ph), 4.79, 4.54
(1H · 2, each d, JA,B = 12.3Hz, –CH2Ph), 4.77, 4.46
(1H · 2, each d, JA,B = 12.2Hz, –CH2Ph), 4.77 (1H, d,
H-10, J1 ,2 = 0.6Hz), 4.69 (1H, dd, H-30), 4.63, 4.56
(2H · 2, each m, –CH2–CH@CH2 · 2), 4.22 (1H, dd,
0
0
H-60a, J5 ,6 a = 5.2Hz, J6 a,6 b = 11.9Hz), 4.21 (1H, dd, H-
3, J2,3 = 10.7Hz, J3,4 = 8.2Hz), 4.17 (1H, dd, H-2,
J2,1 = 8.2Hz), 4.12 (1H, dd, H-4, J4,5 = 9.8Hz), 4.06 (1H,
0
0
0
0
dd, H-60b, J5 ,6 b = 3.3Hz), 3.80 (1H, dd, H-6a,
J5,6a = 3.0Hz, J6a,6b = 11.6Hz), 3.76 (1H, dd, H-6b,
J5,6b = 1.8Hz), 3.51 (1H, ddd, H-5), 3.39 (1H, ddd, H-
50), 2.14, 2.04 (3H · 2, each s, OAc · 2).
0
0
Compound 10: d 7.68 (4H, m, Phth), 7.65–6.86 (5H · 3, m,
Ph · 3), 5.79, 5.76 (1H · 2, each m, –CH2–CH@CH2 · 2),
5.34 (1H, dd, H-20, J1 ,2 = 0Hz, J2 ,3 = 3.4Hz), 5.23, 5.17
(1H · 2, each m, –CH2–CH@CH2), 5.20, 5.15 (1H · 2,
each m, –CH2–CH@CH2), 5.10 (1H, d, H-1, J1,2 = 8.3Hz),
0
0
0
0
4.92 (1H, dd, H-40, J3 ,4 J4 ,5 = 9.8Hz), 4.84, 4.40 (1H · 2,
each d, JA,B = 11.9Hz, –CH2Ph), 4.80, 4.52 (1H · 2, each
d, JA,B = 12.2Hz, –CH2Ph), 4.79, 4.48 (1H · 2, each d,
JA,B = 12.5Hz, –CH2Ph), 4.65 (1H, s, H-10), 4.24 (1H, dd,
H-3, J2,3 = 10.7Hz, J3,4 = 8.0Hz), 4.20 (1H, dd, H-2), 4.09
(1H, dd, H-4, J4,5 = 10.1Hz), 4.06, 4.02 (1H · 2, each m,
–CH2–CH@CH2), 4.03, 4.01 (1H · 2, each m, –CH2–
CH@CH2), 3.82 (1H, dd, H-6a, J5,6a = 3.4Hz,
J6a,6b = 11.0Hz), 3.77 (1H, dd, H-6b, J5,6b = 2.1Hz), 3.57
0
0
0
0
7. Guibe, F.; Saint MÕLeux, Y. Tetrahedron Lett. 1981, 22,
3591–3594.
8. Ogawa, T.; Nakabayashi, S. Carbohydr. Res. 1981, 97, 81–
86.
9. Paulsen, H.; Paal, M. Carbohydr. Res. 1984, 135, 53–
69.
10. 1H NMR (500MHz, in CDCl3) data of compound 7–10.
Compound 7:d 7.63 (4H, m, Phth), 7.52–6.70 (5H · 3,
m, Ph · 3), 5.93, 5.86 (1H · 2, each m, –CH2–CH@
CH2 · 2), 5.37, 5.27 (1H · 2, each m, –CH2–CH@CH2),
5.29, 5.23 (1H · 2, each m, –CH2–CH@CH2), 5.13 (1H,
(1H, ddd, H-5), 3.50 (1H, dd, H-60a, J5 ,6 a = 1.9Hz),
0
0
3.35 (1H, dd, H-60b,J6 b,6 a = 12.6Hz, J6 b,5 = 6.8Hz),
3.28 (1H, dd, H-30), 3.23 (1H, ddd, H-50), 2.12, 2.06
(3H · 2, each s, OAc · 2).
0
0
0
0
d, H-1, J1,2 = 8.5Hz), 4.79, 4.40 (1H · 2, each d, JA,B
=