Urea glycoside synthesis in water

Article abstract of DOI:10.1055/s-2004-822902

A novel approach to the synthesis of urea glycosides in aqueous media has been developed. Reaction of Steyermark's glucosyl carbamate 1 with amines was carried out in water to afford urea glucosides in good yields. This method was successfully applied to

Full text of DOI:10.1055/s-2004-822902

LETTER
1019
Urea Glycoside Synthesis in Water
Urea
G
lycoside Sy
o
nthesis in
W
s
ater hiyasu Ichikawa,* Yohei Matsukawa, Minoru Isobe
Laboratory of Organic Chemistry, School of Bioagricultural Sciences, Nagoya University, Chikusa, Nagoya 464-8601, Japan
Fax +81(52)7894100; E-mail: ichikawa@agr.nagoya-u.ac.jp
Received 25 February 2004
Since usual carbamates do not undergo acetylation under
such mild reaction conditions,5 the highly reactive nature
of 1 can be attributed to its twisted structure, in which
overlap of the lone pair on the nitrogen atom with the sys-
Abstract: A novel approach to the synthesis of urea glycosides in
aqueous media has been developed. Reaction of Steyermark’s
glucosyl carbamate 1 with amines was carried out in water to afford
urea glucosides in good yields. This method was successfully
applied to develop a new route to the synthesis of urea-tethered tem of the carbonyl group is prevented by the rigid geo-
neoglycoconjugates and pseudooligosaccharides.
metry of 1. Consequently, it is expected that ring-opening
reaction of 1 with nucleophiles is prone to occur with a de-
crease in ring strain.6 In fact, Pinter reported, albeit only
one example, the reaction of 1 with N-methylpiperazine in
water to afford urea glucoside.7 Herein we report the syn-
thetic potential of this Steyermark’s carbamate 1 for the
preparation of urea-tethered neoglycoconjugates and
pseudooligosaccharides in aqueous media.
Key words: carbohydrates, glycosylations, glycosides
Urea glycoside is found in nature as a unique structural
motif of glycocinnamoylspermidine antibiotics,1 and sev-
eral years ago, we initiated work on the synthesis of this
class of amino sugar antibiotics. Parallel to this synthetic
effort for the natural products, we also paid attention to
the urea-tethered neoglycoconjugates and pseudo-
oligosaccharides. While several synthetic routes to the
urea glycoside had been reported, they appeared to be not
suitable for our purpose with respect to the functional
group compatibility and stereoselectivity.2 Accordingly,
we developed new synthetic protocols which involve the
reaction of glycopyranosyl isocyanate with amines under
anhydrous conditions in organic solvents.3 During this in-
vestigation, we felt it necessary to explore urea glycoside
synthesis which utilizes unprotected sugars in water. We
envisioned that such an approach would be useful for
bioconjugate synthesis under physiological conditions,
because most parts of biomolecules are soluble in aqueous
solution.
Steyermark’s carbamate 1 was prepared from glucosyl
azide 3 according to the procedure of Pinter (Scheme 2).8
The Staudinger reaction of b-D-glucopyranosyl azide 3
with triphenylphosphine in acetone gave a solution of
phosphinimine 4, which was successively treated with
carbon dioxide in one-pot process. The precipitate was
collected to furnish 1 in 78% yield.
OH
O
OH
O
CO2
78%
HO
HO
HO
HO
PPh3
1
acetone
N
N3
PPh3
OH
OH
3
4
Scheme 2
After surveying the chemical literature, we were most in-
terested in the anomalous reactivity of a glucose deriva-
tive 1 bearing 1b,2a-oxazolidone ring (Scheme 1). The
first synthesis of 1 was reported in 1962 by Steyermark,
who observed that acetylation of 1 (Ac2O, pyridine, r.t.,
20 h) occurred on both the hydroxyl groups and nitrogen
in carbamate ring to afford the tetraacetate 2.4
Initially, we examined the reaction of 1 with 2-phenyleth-
ylamine, because the product could be readily purified by
reverse phase chromatography (Scheme 3). The ring-
opening reaction of 1 with 2-phenylethylamine in water
occurred smoothly at room temperature for one hour, and
the urea glucoside 5 was isolated in 87% yield after ODS
(Cosmosil® 75C18-OPN) column chromatography.
OAc
O
OH
O
Ac2O, pyridine,
room temperature, 20 h
AcO
AcO
HO
HO
HO
1
2
H2N
1)
ref. 4
NAc
NH
O
HO
HO
H2O, r.t., 1 h
O
O
O
O
1
O
2) ODS column
87%
2
1
N
H
N
H
OH
5
Scheme 1
Scheme 3
SYNLETT 2004, No. 6, pp 1019–1022
0
6.
0
5.
2
0
0
4
Advanced online publication: 01.04.2004
DOI: 10.1055/s-2004-822902; Art ID: U05904ST
© Georg Thieme Verlag Stuttgart · New York

1020
Y. Ichikawa et al.
LETTER
In order to evaluate the reactivity of 1 and to identify the amine with 1 (1.2 equiv) afforded the products 6c and 6d
steric factors of the alkyl group in amine, the reaction of 1 in moderate yields (67% and 68%). Slightly more carba-
with a variety of amines was examined using reaction mate 1 (1.5 equiv), longer reaction time (1.5 h) and keep-
conditions similar to those employed in Scheme 3. Re- ing the reaction mixture at 40 °C (entry 4) improved the
sults are summarized in Table 1.
yields (80% and 84%). With secondary amines (entries 5
and 6), use of 2.0 equivalents of 1 gave the products 6e
and 6f in 80% and 71% yields. The difference between en-
tries 5 and 6 can be rationalized by the fact that because of
its cyclic structure, pyrrolidine has lone-pair electrons that
are more accessible for nucleophilic ring-opening reac-
tion than those of diethylamine. The steric effect at the b-
position of secondary amine considerably reduces the
Primary amines, n-butylamine and benzylamine, having
no branches at the a-carbon reacted with 1 to afford the
urea glucosides 6a and 6b in good yields (entries 1 and 2).
Alkyl substituents at the a-position of primary amine had
a considerable steric effect (entries 3 and 4). Thus, reac-
tion of (S)-(–)-a-methylbenzylamine and cyclohexyl-
Table 1 Results of the Reaction of Carbamate 1 with a Variety of Amines
H
H
O
N
R
1)
R
H2O
H
O
N
HO
HO
HO
HO
O
O
2) ODS column
OH
O
OH
OH
6a–h
1
Entry
1
Product (R = )
Time (h)
1.0
Temp (°C)
r.t.
Equiv of 1
Yield (%)a
93b
1.2
N
H
6a
2
3
1.0
r.t.
1.2
88
N
H
6b
1.0
2.0
r.t.
r.t.
1.2
1.5
67
80
N
H
6c
4
1.0
2.0
r.t.
40
1.2
1.5
68
84
N
H
6d
5
6
3.0
6.0
60
50
2.0
2.0
80c
71b
N
6e
N
6f
7
8
12
12
50
70
2.0
1.5
8
N
6g
75
N
H
6h
a Unless otherwise noted, yields were calculated after ODS column purification.
b Since the product is hygroscopic, the yield was determined after acetylation.
c The product was isolated after acetylation, because purification by ODS column was difficult.
Synlett 2004, No. 6, 1019–1022 © Thieme Stuttgart · New York

LETTER
Urea Glycoside Synthesis in Water
1021
OAc
O
OAc
yield. For example, when diisobutylamine was employed
(entry 7), only 8% of 6g was isolated. Aniline (entry 8) un-
derwent urea glycosylation in water at 70 °C for 12 hours
to furnish 6h in 75% yield. In this case, raising the reac-
tion temperature did not accelerate the hydrolysis of 1 due
to the weaker basicity of aniline.
phenol,
SnCl4
AcO
AcO
AcO
AcO
1) Zn, HOAc, THF
O
CH2Cl2
89%
2) CbzCl, CH2Cl2,
aq. NaHCO3
76%
OAc
NHTroc
OPh
NHTroc
9
10
Given the encouraging results stated above, we next
turned our attention to develop a bioconjugation method
using 1, and a preliminary example is illustrated in
Scheme 4. Spermidine (7) underwent urea-glycosylation
with 1 in water, and the corresponding urea glycoside 8
was isolated after acetylation in 71% yield.9
OAc
O
OH
Et3N,
MeOH
AcO
AcO
HO
CBr4, PPh3, pyridine
then Ac2O
O
OPh
NHCbz
11
HO
OPh
NHCbz
12
54%
N3
O
Br
1) NaN3, DMF, 80 °C
2) Et3N, MeOH
HO
HO
AcO
AcO
O
90%
OPh
NHCbz
14
OPh
NHCbz
13
Scheme 5
OH
HO
O
H2N
NH
HO
PPh3, H2O,
t-BuOH
O
O
16
HO
O
Scheme 4
14
HO
H2O/ MeOH
(1:4)
O
CbzHN
15
Finally, we undertook the synthesis of urea-linked oli-
gosaccharide mimics11 in aqueous solvents. For this
purpose, we set up to prepare phenyl 2,6-diamino-glu-
copyranoside (14), which bears two reactive sites for the
reaction with carbamates (Scheme 5). Phenyl group in 14
was planned to facilitate the final purification step with
ODS column chromatography. Glycosylation of glu-
cosamine derivative 9 with phenol, catalyzed with tin(IV)
chloride in dichloromethane, proceeded smoothly to af-
ford a-phenyl glycoside predominantly in 89% yield.10
Treatment of 10 with zinc in a mixture of HOAc and THF
removed Troc carbamate to afford the corresponding
amine, which was immediately reprotected to its Cbz-de-
rivative 11 in 76% yield for two steps. Cleavage of acetyl
groups in 11 (Et3N, MeOH) gave the triol 12. Selective
bromination (CBr4, PPh3, pyridine) of the primary hy-
droxyl group in 12 and acetylation (Ac2O added in one
pot) furnished bromide 13 (54% in two steps). Displace-
ment of the bromide 13 with sodium azide in DMF and
methanolysis of the acetate groups (Et3N, MeOH) afford-
ed 14, which was employed as precursor in the synthesis
of an urea-linked pseudotrisaccharide 18 (Scheme 6).
OH
HO
O
HO
O
OH
N
HO
N
H
HO
O
H
1) H2, Pd-C, H2O, MeOH
CbzHN
17
O
2)
1
H2O
85%
OH
O
HO
HO
HO
O
HO
HO
N
H
HO
O
HO
O
N
H
OH
N
H
HO
N
O
O
H
18
Scheme 6
was subsequently treated with 1 in water. The resultant
product was purified by ODS chromatography to furnish
18 in 85% overall yield for four steps.
In conclusion, we have demonstrated that Steyermark’s
carbamate is a useful synthon for the synthesis of urea-
tethered neoglycoconjugates and pseudooligosaccharides
in aqueous media. This approach represents a useful pro-
tocol for anchoring a carbohydrate moiety onto an amine
in water, which may mimic the enzyme-catalyzed glyco-
sylation in water.
The synthesis of 18 began with the reduction of azide 14
with triphenylphosphine. The resultant amine 15 under-
went urea-glycosylation with galactopyranosyl carbamate
16 in aqueous media (H2O/MeOH, 1:4) to afford pseudo-
disaccharide 17. Deprotection of the Cbz group in 17 gave
the corresponding 2-amino-pseudodisaccharide, which
Synlett 2004, No. 6, 1019–1022 © Thieme Stuttgart · New York

1022
Y. Ichikawa et al.
LETTER
General Procedure of Urea-Glycosylation in Water. To a solu-
tion of phenethylamine (54 mg, 0.45 mmol) in H2O (10.0 mL) was
added carbamate 1 (110 mg, 0.54 mmol) in a single portion. After
stirring at r.t. for 1.0 h, the reaction mixture was directly passed
through a column of ODS (Cosmosil® C18-OPN, H2O followed by
H2O/MeOH, 10:1 as eluent). The urea glucoside 5 was obtained as
a white solid (126 mg, 87%).12
OAc
O
OH
O
AcO
AcO
HO
HO
NR
NH
O
O
O
O
i
ii R = H
iii R = Ac
References
Scheme 7
(1) (a) Ellestad, G. A.; Cosulich, D. B.; Broschard, R. W.;
Martin, J. H.; Kunstmann, M. P.; Morton, G. O.; Lancaster,
J. E.; Fulmor, W.; Lovell, F. M. J. Am. Chem. Soc. 1978,
100, 2515. (b) Dobashi, K.; Nagaoka, K.; Watanabe, Y.;
Nishida, M.; Hamada, M.; Takeuchi, T.; Umezawa, H. J.
Antibiot. 1985, 1166.
(2) (a) Schoorl, M. N. Recl. Trav. Chim. Pays-Bas 1903, 22, 31.
(b) Benn, M. H.; Jones, A. S. J. Chem. Soc. 1960, 3837.
(c) Fischer, E. Chem. Ber. 1914, 47, 1377. (d) Johnson, T.
B.; Bergmann, W. J. Am. Chem. Soc. 1932, 54, 3360.
(e) Bannister, B. J. Antibiot. 1972, 25, 377. (f) Pinter, I.;
Kovacs, J.; Toth, G. Carbohydr. Res. 1995, 273, 99.
(3) (a) Ichikawa, Y.; Nishiyama, T.; Isobe, M. Synlett 2000,
1253. (b) Ichikawa, Y.; Nishiyama, T.; Isobe, M. J. Org.
Chem. 2001, 66, 4200. (c) Nishiyama, T.; Ichikawa, Y.;
Isobe, M. Synlett 2003, 47. (d) Ichikawa, Y.; Matsukawa,
Y.; Nishiyama, T.; Isobe, M. Eur. J. Org. Chem. 2004, 586.
(e) Ichikawa, Y.; Nishiyama, T.; Isobe, M. Tetrahedron
2004, 60, 2621.
(6) For the reactivity of strained carbamates, see the reference:
Hall, H. K. Jr.; El-Shekeil, A. J. Org. Chem. 1980, 45, 5325.
(7) (a) Pinter, I.; Kovacs, J.; Toth, G. Carbohydr. Res. 1995,
273, 99. (b) Analogous O-unprotected glucosyl
thiocarbamate and its reaction with amines in water has been
reproted. See the reference: Maya, I.; López, ; Fernández-
Bolaños, J. G.; Robina, I.; Fuentes, J. Tetrahedron Lett.
2001, 42, 5413. (c) López, Ó.; Maya, I.; Fuentes, J.;
Fernández-Bolaños, J. G. Tetrahedron 2004, 60, 61.
(8) Kovacs, J.; Pinter, I.; Messmer, A. Carbohydr. Res. 1985,
141, 57.
(9) The separation of glucose, hydrolyzed product of 1, by ODS
column was difficult in this case.
(10) Nishiyama, T.; Ichikawa, Y.; Isobe, M. Synlett 2004, 89; and
references therein.
(11) For recent examples of the pseudooligosaccharides with
thiourea linkage, see: Jiméne Blanco, J. L.; Bootello, P.;
Ortiz Mellet, C.; Gutiérrez Gallego, R.; García Fernández, J.
M. Chem. Commun. 2004, 92.
(4) Steyermark, P. R. J. Org. Chem. 1962, 27, 1058.
(5) For mannose type, carbamate i underwent acetylation
(Ac2O, pyridine, r.t.) to afford triacetate ii. Acetylation of i
under more forcing conditions (Ac2O, NaOAc, refluxed for
1 h) gave N-acetate iii (Scheme 7). See the reference:
Kovacs, J.; Pinter, I. Carbohydr. Res. 1991, 210, 155.
(12) Spectroscopic data of 5: 1H NMR (400 MHz, CD3OD): d =
2.77 (2 H, t, J = 7.0 Hz), 3.13 (1 H, t, J = 9.0 Hz), 3.25 (1 H,
t, J = 9.0 Hz), 3.32 (1 H, ddd, J = 9.0, 5.5 and 2.0 Hz), 3.37
(2 H, t, J = 7.0 Hz), 3.38 (1 H, t, J = 9.0 Hz), 3.63 (1 H, dd,
J = 12.0 and 5.5 Hz), 3.81 (1 H, dd, J = 12.0 and 2.0 Hz),
4.73 (1 H, dd, J = 9.0 Hz), 7.15–7.29 (5 H). 13C NMR (100
MHz, CD3OD): d = 37.3, 42.6, 62.8, 71.6, 74.3, 79.1, 79.2,
82.8, 127.3, 129.5, 129.8, 140.7, 160.5.
Synlett 2004, No. 6, 1019–1022 © Thieme Stuttgart · New York