First preparative synthesis of a 3-acetamido-3,6-dideoxy-D-galactopyranose glycosyl donor via intramolecular cyclization of an epoxytrichloroacetimidate

Article abstract of DOI:10.1016/j.tetlet.2004.04.065

The preparative synthesis of a 3-acetamido-3,6-dideoxy-D-galactopyranose N-phenyl-trifluoroacetimidate donor has been accomplished using as key step a silica gel mediated cyclization of an epoxytrichloroacetimidate, while other more conventional routes to

Full text of DOI:10.1016/j.tetlet.2004.04.065

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 4445–4448
First preparative synthesis of a 3-acetamido-3,6-dideoxy-D-
galactopyranose glycosyl donor via intramolecular cyclization
of an epoxytrichloroacetimidate
Emiliano Bedini,* Alfonso Iadonisi, Antonella Carabellese and Michelangelo Parrilli*
Dipartimento di Chimica Organica e Biochimica, Universitꢀa di Napoli ‘Federico II’, Complesso Universitario Monte Santangelo,
Via Cintia 4, 80126 Napoli, Italy
Received 26 March 2004; revised 13 April 2004; accepted 13 April 2004
Abstract—The preparative synthesis of a 3-acetamido-3,6-dideoxy-D-galactopyranose N-phenyl-trifluoroacetimidate donor has
been accomplished using as key step a silica gel mediated cyclization of an epoxytrichloroacetimidate, while other more conventional
routes to aminosugars failed. Test glycosylations with the N-phenyl-trifluoroacetimidate donor are also reported.
Ó 2004 Elsevier Ltd. All rights reserved.
3-Acetamido-3,6-dideoxy-
D
-galactopyranose (
D
-Fucp3-
In an initial attempt, we planned a synthetic strategy
from commercially available -fucose entailing the
conversion of the 3-OH group to an amino functionality
via an SN2 displacement of a triflate with sodium azide.
Therefore the first steps of the synthesis required the
NAc) is a sugar frequently found in the O-antigen
moiety of lipopolysaccharides (LPS)1 extracted almost
exclusively from phytopathogenic Gram-negative bac-
teria.2 The synthesis of several unusual aminodeoxy-
hexoses and some related oligosaccharides has already
been accomplished,3 as some of them are important
constituents of various antibiotics4 and play important
biological roles. In contrast, the biological role of
D
regioselective protection of
D-fucose to obtain a free
alcoholic function exclusively on position 3 (Scheme 1).
We firstly protected the anomeric position with an ethyl
group by a sequence of reactions (Fischer glycosylation,
acetylation, a-anomerization with FeCl3;6 56% over
three steps) that assured the enrichment of the anomeric
mixture with the a-anomer 1. This triacetylated com-
D
-Fucp3NAc has not yet been clearly elucidated, despite
its wide distribution in nature.
ꢀ
Therefore, the synthesis of a suitable
D
-Fucp3NAc
pound was then deacylated under Zemplen conditions
building-block and its incorporation in biologically
important oligosaccharide sequences would be useful for
the purpose. Actually, the only to date reported syn-
(87%) to yield triol 2 that was then subjected to a one-pot
sequence of three reactions (orthoesterification, allyl-
ation and orthoester regioselective opening; 51% over
three steps) to afford the alcohol 3. A direct displace-
ment of the triflated derivative of 3 with sodium azide
would give an azido-sugar with a configuration at C-3
opposite to that desired; therefore, alcohol 3 was epi-
merized in two steps by oxidation with DMSO/Ac2O7
and subsequent reduction of the ketone 4. This sequence
afforded alcohol 5 in 55% yield with excellent stereo-
selectivity, reasonably due to the steric hindrance of the
axial ethyl group at the anomeric position.
thesis of
L
-Fucp3NAc, as a methyl glycoside,5 is not
practical for a preparative purpose, the overall yield
being modest and the product obtained being in an
unsuitable form for incorporation into more complex
oligosaccharides. Therefore, a preparative new proce-
dure for the synthesis of
mandatory.
D-Fucp3NAc donors was
Keywords: Aminosugar; 3-Acetamido-3,6-dideoxy-D-galactopyranose;
Deoxygulose alcohol 5 was then converted into the
corresponding triflated derivative that unfortunately did
not afford the desired azidosugar 6 when treated with
Epoxytrichloroacetimidate; Glycosylation; Lipopolysaccharide.
* Corresponding authors. Tel.: +39-81-674147; fax: +39-81-674393;
e-mail: parrilli@unina.it
0040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.04.065

4446
E. Bedini et al. / Tetrahedron Letters 45 (2004) 4445–4448
OH
OR'
OAc
O
O
O
a
b
HO
HO
AcO
OH
OR
OAc
OH
OEt
OEt
2: R=H, R'=H
D-Fuc
1
c
3: R=All, R'=Ac
OAc
OAc
OAc
O
O
O
d
e
3
N3
OAll
OEt
OAll
O
OEt
OAll
OEt
OH
6
4
5
f
OAc
O
poor yield in
reduction product
N
MeO
OEt
OAll
7
Scheme 1. Reagents and conditions: (a) (i) EtOH, Amberlist-15 (Hþ), reflux, (ii) Ac2O, pyridine, rt, (iii) FeCl3, CH2Cl2, rt, 56% over three steps; (b)
NaOMe, MeOH, rt, 87%; (c) (i) trimethyl-orthoacetate, CSA, DMF, 40 °C, (ii) NaH, AllBr, rt, (iii) 80% AcOH, rt, 51% over three steps; (d) 2:1
DMSO/Ac2O, rt; (e) NaBH4, 9:1 THF/MeOH, 0 °C, 55% over two steps from 3; (f) NH2OMeÆHCl, 64% over two steps from 3.
sodium azide in DMF. Indeed, the reaction produced a
complex mixture with predominant amounts of elimi-
nation products. Every attempt to achieve the predom-
inance of the substitution process over elimination using
crown-ethers and/or different solvents (DMSO, toluene)
failed. Actually, difficulties in nucleophilic displace-
ments of 3-OTf-gulose derivatives have been recently
reported,8;9 and they have been solved by using a 4,6-O-
benzylidene ring8 and/or a 2-O-acyl protecting group.9
Both methods minimize elimination reactions, never-
theless they were unfortunately considered to be not
very useful for our scope, due to the impossible instal-
lation of a 4,6-O-benzylidene on a 6-deoxysugar and the
necessity of avoiding the use of 2-O-acyl protecting
their formation is usually accomplished by electrophile-
induced cyclization of allylic trichloroacetimidate11 or
by intramolecular nucleophilic displacement of bis(tri-
chloroacetimidate),12 whereas here we report this
achievement by an intramolecular cyclization of 2,3-
epoxytrichloroacetimidates.
Starting from allyl a-
readily available from
D
-fucopyranoside 8 (Scheme 2),
-galactose,13 alcohol 9 was ini-
D
tially prepared by a one-pot sequence of three reactions
(orthoesterification, acetylation and orthoester regio-
selective opening; 58% yield over three steps). Conver-
sion of 9 into the corresponding triflate 10 and
ꢀ
subsequent exposure to Zemplen deacylation conditions,
group in the synthesis of a
used for a-glycosidations (see below).
D
-Fucp3NAc donor to be
afforded, with high regio and stereoselectivity, the
epoxyalcohol 11 that was in turn directly converted to the
epoxytrichloroacetimidate 12 with trichloroacetonitrile
and catalytic DBU. The intramolecular cyclization of
2,3-epoxytrichloroacetimidatesto(trichloromethyl)oxazo-
lines has been already applied on open-chain com-
pounds,14 whereas its application to cyclic saccharidic-
like compounds is so far restricted to a single example.15
Its mechanism requires the catalysis of Lewis acids
(Et2AlCl,14a BF3ÆOEt2,14b SnCl4,14b Et3Al,14c;15 CoCl216),
most of which are quite poisonous and moisture sensi-
tive; in contrast, the cyclization of the epoxytrichloro-
acetimidate 12 was simply performed by adsorption on
silica gel at 45 °C and subsequent chromatography. It is
also worthy of note that in the previously reported pro-
cedures the chromatographic purification of the inter-
mediates is required. In this case the oxazoline 13 was
instead obtained in a 64% yield over four steps from the
Thus, we decided to insert the amino functionality on C3
by oxime reduction:10 ketone 4 was transformed in its
O-methyloxime derivative
NH2OMeÆHCl (64% over two steps from 3), but the
subsequent reduction of 7 with either BH3ÆTHF or
LiAlH4 gave a complex mixture with only a small
amount of the desired compound.
7
by treatment with
Since the most conventional strategies for converting a
saccharidic alcohol to an amino group were uneffective
on our compounds, we envisioned a different synthetic
approach, based on the electrophile-induced cyclization
of an epoxytrichloroacetimidate to
methyl)oxazoline. In carbohydrate chemistry (trichlo-
romethyl)oxazoline derivatives have already been used:
a
(trichloro-

E. Bedini et al. / Tetrahedron Letters 45 (2004) 4445–4448
4447
OR
OAc
OH
O
O
O
c
a
HO
RO
OH
OAc
OAll
OAll
O
OAll
8
9: R=H
10: R=Tf
11: R=H
b
d
12: R=C(NH)CCl3
e
OAc
O
Cl3C
O
O
g
AcHN
N
OBn
OR
OR
OAll
15: R=α-All
16: R=H
13: R=H
h
i
f
14: R=Bn
17: R=α-C(NH)CCl3
18: R=C(NPh)CF3
j
Scheme 2. Reagents and conditions: (a) (i) trimethyl-orthoacetate, CSA, DMF, 40 °C, (ii) Ac2O, pyridine, rt, (iii) AcOH 80%, rt, 58% over three
steps; (b) Tf2O, 1:1 CH2Cl2/py, 0 °C; (c) NaOMe, MeOH, rt; (d) Cl3CCN, DBU, CH2Cl2, 0 °C; (e) silica gel (0.063–0.200 mm), CHCl3, 45 °C, in
vacuo, 64% over four steps from 9; (f) BnBr, NaH, DMF, rt, 68%; (g) (i) 1 M HCl, THF, rt, (ii) Ac2O, py, rt, 63% over two steps; (h) PdCl2, 1:1
CH2Cl2/MeOH, rt, 84% (a : b ¼ 1 : 1:5 as determined by 1H NMR analysis); (i) Cl3CCN, DBU, CH2Cl2, 0 °C, 53%; (j) CF3C(NPh)CCl, NaH,
molecular sieves 4 A, CH2Cl2, 0 °C, 67% (a : b ¼ 3 : 1 as determined by 1H NMR analysis).
ꢁ
alcohol 9 without any intermediate chromatography,17
also by means of the excellent regio and stereoselectivity
of the whole synthetic sequence.
glycosyl donor, as it is supposed to exalt the a-stereo-
selectivity in glycosidation reactions through a long
range participation effect,18 even though this effect has
not been observed in a recent study regarding the cou-
pling of various fucosyl donors with linear alcohols.19
Since
D-Fucp3N occurs in natural oligosaccharides
almost exclusively as an a-glycoside, the preparation of
its glycosyl donor required the installation of a nonpar-
ticipating protecting group at position 2. Moreover, a 4-
O-acyl group was chosen in designing the most suitable
Thus, 13 was benzylated to give 14 (68%) that was
subsequently subjected to acid hydrolysis of the oxazo-
line cycle and acetylation to obtain 15 (63% over two
OBn
O
OBn
BzO
AllO
O
O
18, TMSOTf
BzO
MS, CH2Cl2, rt
AllO
O
BnO
OH
19
AcHN
20 (65%, α:β 1.6:1)
OAc
OAc
O
O
O
18, TMSOTf
AcHN
HO
O
O
OBn
O
MS, CH2Cl2, rt
O
O
OMe
OMe
21
22 (61%, α:β 2.1:1)
Scheme 3.

4448
E. Bedini et al. / Tetrahedron Letters 45 (2004) 4445–4448
10. Tsuda, Y.; Okuno, Y.; Iwaki, M.; Kanemitsu, K. Chem.
Pharm. Bull. 1989, 37, 2673–2678.
steps). Anomeric deallylation with PdCl2 afforded 16
(84%) that was finally converted into two different gly-
cosyl donors, namely trichloroacetimidate 17 (53%) by
treatment with Cl3CCN and DBU and N-phenyl-tri-
fluoroacetimidate20 18 (67%) by treatment with
CF3C(NPh)Cl and NaH.21 Since 18 was obtained in a
significantly better yield than 17 after chromatographi-
cal purification, we decided to test the former compound
in two glycosylation reactions.
11. (a) Pauls, H. W.; Fraser-Reid, B. J. Org. Chem. 1983, 48,
1392–1393; (b) Bongini, A.; Cardillo, G.; Orena, M.;
Sandri, S.; Tomasini, C. Tetrahedron 1983, 39, 3801–3806.
12. Link, T.; Raghavan, S.; Danishefsky, S. J. J. Am. Chem.
Soc. 1995, 117, 552–553.
13. Liu, M.; Yu, B.; Wu, X.; Hui, Y.; Fung, K.-P. Carbohydr.
Res. 2000, 329, 745–754.
14. (a) Hatakeyama, S.; Matsumoto, H.; Fukuyama, H.;
Mukugi, Y.; Irie, H. J. Org. Chem. 1997, 62, 2275–2279;
(b) Schmidt, U.; Respondek, M.; Lieberknecht, A.;
Werner, J.; Fischer, P. Synthesis 1989, 256–261; (c)
Bernet, B.; Vasella, A. Tetrahedron Lett. 1983, 49, 5491–
5494.
Coupling 18 with the rhamnosyl acceptors 1922 and 2123
(Scheme 3) under the agency of TMSOTf gave 20 and 22
in 65% and 61% yield, respectively, and a fairly good
a-selectivity. Notably, both 20 and 22 are useful build-
€
15. Schurrle, K.; Beier, B.; Piepersberg, W. J. Chem. Soc.,
Perkin Trans. 1 1991, 2407–2412.
ing-blocks for the synthesis of many
D-Fucp3NAc
containing repeating unit of LPS from phytopathogenic
bacteria. Work is in progress in order to further enhance
a-selectivity of the couplings; the results will be pub-
lished at due time.
€
16. Schmidt, U.; Zah, M.; Lieberknecht, A. J. Chem. Soc.,
Chem. Commun. 1991, 1002–1004.
17. Compound 13: A solution of 9 (1.34 g, 4.61 mmol) in 1:1
CH2Cl2/pyridine (10 mL) was cooled at 0 °C and then
Tf2O (1.6 mL, 9.7 mmol) was slowly added. The solution
was stirred at 0 °C for 400, after that the solution was
diluted with CH2Cl2 (300 mL) and washed with 1 M HCl
(300 mL), 1 M NaHCO3 (300 mL) and water (300 mL).
The organic layer was collected, dried and concentrated to
afford an oily residue that was dissolved in 2:1 MeOH/
CH2Cl2 (21 mL) and treated with a 0.6 M solution of
NaOMe in MeOH (12 mL) at rt. After 2 h, the solution
was diluted with CH2Cl2 (350 mL) and washed with water
(350 mL). The organic layer was collected, dried and
concentrated to afford an oily residue that was then
dissolved in CH2Cl2 (13 mL). The solution was cooled at
0 °C and then treated with Cl3CCN (4.5 mL, 44.8 mmol)
and DBU (360 lL, 0.72 mmol). After 600 under stirring at
0 °C, the solution was concentrated. Silica gel (0.063–
0.200 mm) (5.6 g) was then added to the residue, the
mixture was suspended in CHCl3 (20 mL) and immediately
concentrated in vacuo at 45 °C. After 100 the solvent was
completely evaporated and the solid residue was chroma-
tographed (8:1 petroleum ether/EtOAc) to give 13
Acknowledgements
We thank Centro di Metodologie Chimico-Fisiche of
the University Federico II of Naples for the NMR
spectra, and MIUR, Rome (Progetti di Ricerca di In-
teresse Nazionale 2002, M.P.) for the financial support.
References and notes
1. (a) Knirel, Y.; Kochetkov, N. K. Biochemistry (Moscow)
1994, 59, 1325–1383; (b) Jansson, P.-E. In Endotoxins in
Health and Disease; Brade, H., Opal, S. M., Vogel, S.,
Morrison, D. C., Eds.; Marcel Dekker: New York, 1999;
pp 155–178; (c) Raetz, C. R. H.; Whitfield, C. Ann. Rev.
Biochem. 2002, 71, 535–700.
(965 mg, 64%) as a yellowish oil. ½aꢀ +31.7 (c 0.7,
D
CH2Cl2). 1H NMR (CDCl3, 200 MHz): d 5.90 (m, 1H,
OCH2CH@CH2), 5.28 (dd, 1H, Jvic ¼ 17:2 Hz, Jgem ¼ 1:6
Hz, OCH2CH@CH2 trans), 5.21 (dd, 1H, Jvic ¼ 10:4 Hz,
Jgem ¼ 1:6 Hz, OCH2CH@CH2 cis), 4.85 (dd, 1H,
J4;3 ¼ 9:8 Hz, J4;5 ¼ 1:6 Hz, H4), 4.76 (d, 1H, J1;2 ¼ 4:4 Hz,
H1), 4.72 (dd, 1H, J3;4 ¼ 9:8 Hz, J3;2 ¼ 3:8 Hz, H3), 4.40–
4.28 (m, 3H, H2, H5, OCH2CH@CH2), 4.13 (m, 1H,
OCH2CH@CH2), 3.08 (b s, 1H, OH), 1.29 (d, 3H,
J6;5 ¼ 6:6 Hz, H6); 13C NMR (CDCl3, 50 MHz): d 164.2
(C@N), 133.8 (OCH2CH@CH2), 117.5 (OCH2CH@CH2),
94.6 (C1), 83.5 (C4), 68.3, 66.9, 65.7, 64.8 (C2, C3, C5,
OCH2CH@CH2), 15.9 (C6). ESI-MS for C11H14Cl3NO4
(m=z): Mr (calcd) 329.00, Mr (found) 351.88 (M+Na)þ.
18. Dejter-Juszynski, M.; Flowers, H. M. Carbohydr. Res.
1972, 23, 41–45.
2. Corsaro, M. M.; De Castro, C.; Molinaro, A.; Parrilli, M.
Recent Res. Dev. Phytochem. 2001, 5, 119–138.
3. Some recent examples: (a) Langner, M.; Laschat, S.;
Grunenberg, J. Eur. J. Org. Chem. 2003, 1494–1499; (b)
Nicolaou, K. C.; Baran, P. S.; Zhang, Y.-L.; Barluenga,
S.; Hunt, W.; Kranich, R.; Vega, J. A. J. Am. Chem. Soc.
2002, 124, 2233–2244; (c) Varga, Z.; Bajza, I.; Batta, G.;
ꢂ
Liptak, A. Tetrahedron Lett. 2002, 43, 3145–3148; (d)
Kirkpatrick, P. N.; Scafe, W.; Hallis, T. M.; Liu, H.-W.;
Spencer, J. B.; Williams, D. H. Chem. Commun. 2000,
1565–1566; (e) Tsvetkov, Y. E.; Shashkov, A. S.; Knirel,
€
Y.; Backinowsky, L.; Zahringer, U. Mendeleev Commun.
2000, 90–92; (f) Franck-Neumann, M.; Miesch-Gross, L.;
Gateau, C. Tetrahedron Lett. 1999, 40, 2829–2832.
4. Nicolaou, K. C.; Mitchell, H. J. Angew. Chem., Int. Ed.
2001, 40, 1576–1624.
5. Adinolfi, M.; Barone, G.; Corsaro, M. M.; Lanzetta, R.;
Mangoni, L.; Monaco, P. J. Carbohydr. Chem. 1995, 14,
913–928.
19. Vermeer, H. J.; van Dijk, C. M.; Kamerling, J. P.;
Vliegenthart, J. F. G. Eur. J. Org. Chem. 2001, 193–
203.
20. Yu, B.; Tao, H. Tetrahedron Lett. 2001, 42, 2405–2407.
21. Adinolfi, M.; Barone, G.; Iadonisi, A.; Schiattarella, M.
Tetrahedron Lett. 2002, 43, 5573–5577.
6. Chatterjee, S. K.; Nuhn, P. Chem. Commun. 1998, 1729–
1730.
7. Mancuso, A. J.; Swern, D. Synthesis 1981, 165–185.
8. Liakatos, A.; Kiefel, M. J.; von Itzstein, M. Org. Lett.
2003, 5, 4365–4368.
22. Bedini, E.; Barone, G.; Unverzagt, C.; Parrilli, M.
Carbohydr. Res. 2004, 339, 393–400.
23. Zou, W.; Sen, A. K.; Szarek, W.; MacLean, D. B. Can. J.
Chem. 1993, 71, 2194–2200.
9. Rich, J. R.; Bundle, D. R. Org. Lett. 2004, 6, 897–900.