Carbohydrate mimics: Analogues of aza-di-(or tri-)saccharides

Article abstract of DOI:10.1016/S0040-4039(01)00242-8

Four enantiopure pseudo-aza-di-(or tri-)saccharides have been synthesized by aminocyclization of a C₂-symmetrical d-manno-bis-epoxide either with a primary amine having a polyhydroxylated tetrahydrofuran skeleton or with a polyhydroxylated pipe

Full text of DOI:10.1016/S0040-4039(01)00242-8

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 2661–2663
Carbohydrate mimics: analogues of aza-di-(or tri-)saccharides
Yves Le Merrer,* Miche`le Sanie`re, Isabelle McCort, Catherine Dupuy and Jean-Claude Depezay
Universite´ Rene´ Descartes, Laboratoire de Chimie et Biochimie Pharmacologiques et Toxicologiques, UMR 8601 CNRS,
45, rue des Saints-Pe`res, 75270 Paris Cedex 06, France
Received 10 January 2001; accepted 7 February 2001
Abstract—Four enantiopure pseudo-aza-di-(or tri-)saccharides have been synthesized by aminocyclization of a C2-symmetrical
D
-manno-bis-epoxide either with a primary amine having a polyhydroxylated tetrahydrofuran skeleton or with a polyhydroxylated
piperidine or azepan. © 2001 Elsevier Science Ltd. All rights reserved.
Carbohydrates play a critical role in biological events
such as intercellular communication and cell-mediated
processes. Thus, inhibitors of glycosidases or glycosyl-
transferases involved in these processes may find thera-
peutic applications to treat various diseases such as
diabetes,1 cancer2 and viral infections.3 Enzymatic
hydrolysis of a glycosidic bond is assumed to proceed
via a general acid–base catalysis. Monosaccharide ana-
logues with nitrogen in place of the oxygen are potent
glycosidase inhibitors because protonation at physio-
logical pH allows them to mimic the oxycarbenium
ion-like transition state.4,5 Furthermore, since the agly-
con part of the glycoside plays an important role in
achieving selectivity towards different glycosidases,
pseudo-disaccharides or pseudo-sugars containing an
aglycon part are worth considering.6,7
With the goal of developing new routes to carbohydrate
mimics as potential drugs or pharmacological tools, our
ongoing program is directed towards the synthesis of
OH
HO
OH
OH
OH
OH OH
HO
HO
OH
OH
OH
O
A :
B :
N
N
N
HO
HO OH
OH
OH
HO
OH
Figure 1.
OP
1a : P = Bn
O
1b : P = C(CH3)2
O
PO
Nu
OH
OBn
Nu
HO
BnO
Nu
HO
O
Nu
BnO
O
Nu
OH
OP
HO
Nu
OH
O
OH
PO
OBn
I
III
IV
II
(Nu = NH2)
1b
1a (Nu = NH)
A
B
Figure 2.
Keywords: glycosidases; pseudo-azasaccharide; bis-epoxide; polycyclic heterocyclic compounds.
* Corresponding author. Tel.: +33(0)142862181; fax: +33(0)142868387; e-mail: yves.le-merrer@biomedicale.univ-paris5.fr
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)00242-8

2662
Y. Le Merrer et al. / Tetrahedron Letters 42 (2001) 2661–2663
enantiopure aza-di-(or tri-)saccharides from C2-sym-
metrical bis-epoxides derived from -mannitol (Fig. 1).
having an azepan or a piperidine skeleton derived from
III and IV, respectively (Nu=NH).
D
In previous papers, we have reported the regiospecific
opening of a C2-symmetric -manno or -ido-bis-epox-
Pseudo-azadisaccharides were obtained according to
Scheme 1.9 Azidomethyl-tetrahydrofuran formation
D
L
ide by various nucleophiles to afford iminosugars, thio-
sugars, guanidinosugars and aminothiazolinosugars.8
Nucleophilic opening of bis-epoxide 1 depends on dif-
ferent factors, such as the nature of the nucleophile or
the 3,4-O-protecting groups of the bis-epoxide, and the
experimental conditions (Fig. 2).
was initiated by opening of the 3,4-di-O-benzyl
D-
manno bis-epoxide 1a with sodium azide in the presence
of silica gel (95% yield).10 Protection of the primary
alcohol function of 2 into the tert-butyldimethylsilyl
ether followed by selective heterogeneous catalytic
hydrogenation led to the aminomethyl-tetrahydrofuran
4 (95% overall yield). Aminocyclization of a second
bis-epoxide 1a, with the primary amine 4 cleanly
occurred in methanol affording a mixture of bicycles
N-methylfurano-piperidine 5a and N-methylfurano-
azepan 6a, which could be separated by flash chro-
matography (30 and 40%, respectively, for 5a and 6a).11
Subsequent desilylation (nBu4NF, THF) led to 5b and
6b in 80 and 85% yields, respectively. After removal of
all benzyl protecting groups by hydrogenolysis in the
presence of palladium black in acetic acid, the desired
azadisaccharides with an N-methylfurano-piperidine 5c
or azepan 6c skeleton were isolated at its acetate salt
after purification by flash chromatography (80 and 85%
yields, respectively).
After nucleophilic opening of the first epoxide moiety
at the least substituted site, three different evolutions
can occur.8 Opening of the other epoxide by a second
equivalent of nucleophile leads to the C2-symmetric
acyclic compound I. O-Cyclization according to a 5-
exo-tet process gives mainly the tetrahydrofuran com-
pound II. Nu-Cyclization, after acid–base reaction
between the alkoxide and the introduced nucleophile,
leads to a seven- and a six-membered ring III and IV,
according to the regioselectivity of the second epoxide
opening.
In order to obtain pseudo-azadi-(or tri-)saccharides A
and B, respectively, our synthetic strategy involves the
aminocyclization of bis-epoxide 1, on the one hand
with a primary amine having a polyhydroxylated tetra-
hydrofuran skeleton derived from II (Nu=NH2) and
on the other hand with a secondary heterocyclic amine
The second strategy to afford pseudo-azatrisaccharides
B involves the bis-epoxide opening with a sterically
hindered heterocyclic amine. We initially investigated
O
A
OTBDMS
N3
OH
O
b
O
a
BnO
OBn
BnO
c
OBn
BnO
OBn
O
3 : A = N3
2
1a
4 : A = NH2
HO
HO
P1O
P1O
OP2
N
d
O
P1O
P1O
N
P1O
OP2
O
+
HO
P1O
OP1
OP1
HO
6a : P1 = Bn, P2 = TBDMS
6b : P1 = Bn, P2 = H
6c : P1 = P2 = H
5a : P1 = Bn, P2 = TBDMS
5b : P1 = Bn, P2 = H
5c : P1 = P2 = H
e
f
e
f
Scheme 1. Reagents and conditions: (a) NaN3, SiO2, CH3CN, D, 48 h; (b) TBDMSCl, imidazole, DMF, 15 h; (c) H2, Pd black,
EtOAc; (d) 1a, CH3OH, 8 days; (e) Bu4NF, THF; (f) H2, Pd black, CH3CO2H.
a
c
8 a
7a : P = Bn
1a
1b
HO OP
N
OH
O
N
a or b
7b : P = C(CH3)2
N
PO OH
BnO
OBn
Scheme 2. Reagents and conditions: (a) neat piperidine, 20°C, 4 days, 80% of 7a or 90% of 7b (b) piperidine (2 equiv.), MeOH,
20°C, 90% of 7b; (c) piperidine, 2 equiv., CH3CN, D, 85% of 8a.

Y. Le Merrer et al. / Tetrahedron Letters 42 (2001) 2661–2663
2663
HO
OH
OH
OH
HO
HO
HO
HO OP
OH
OH
10a : P = C(CH3)2
10b : P = H
9
N
H
a
HO
HO
N
b
b
N
PO OH
OH
O
OH
O
HO
N
OH
OH
OH
HO
OH
O
O
HO
HO
HO OP
12a : P = C(CH3)2
12b : P = H
N
N
H
11
OH
PO OH
1b
a
HO
OH
Scheme 3. Reagents and conditions: (a) 9 or 11, 2 equiv. in MeOH, 12 h, 20°C; 75%, or 95% for 10a or 12a, respectively; (b)
CF3CO2H/H2O 4/1, 20°C, 70%, or 80% for 10b or 12b, respectively.
the reactivity of piperidine towards the 3,4-di-O-benzyl
and 3,4-O-isopropylidene- -manno-bis-epoxide, 1a and
1b, respectively (Scheme 2).9
in due course. Synthesis of pseudo-azatrisaccharides
could also be achieved following the described strategy
for pseudo-azadisaccharides A according to a recursive
process, which involves transformation of the primary
hydroxy group into a primary amine.
D
Nucleophilic opening of the 3,4-O-dibenzyl-D-manno-
bis-epoxide 1a in neat piperidine at room temperature
led cleanly to the acyclic C2-symmetric compounds 7a
(80%). On the other hand, treatment of the same bis-
epoxide 1a with two equivalents of piperidine in reflux-
References
ing
acetonitrile
afforded
the
piperidino-
1. Kordik, A.; Rietz, A. B. J. Med. Chem. 1999, 42, 181–
201.
2. Gross, P. E.; Baker, M. A.; Carver, J. P.; Dennis, J. W.
Clin. Cancer Res. 1995, 1, 935–944 and references cited
therein.
methyl-tetrahydrofuran 8a (85%). In this case, only the
O-cyclization occurred by a 5-exo-tet process (no tetra-
hydropyran has been isolated after flash chromatogra-
phy). These results interestingly proved that
experimental conditions could modify the course of the
reaction. Furthermore, the more rigid 3,4-O-isopropyli-
dene bis-epoxide 1b afforded the acyclic compound 7b
(90%), whatever the experimental conditions were
(piperidine as solvent or 2 equiv. of piperidine in
methanol). In connection with these results leading
to acyclic C2-symmetrical compounds, we looked at
3. Fenouillet, E.; Papandreou, M. J.; Jones, I. M. Virology
1997, 231, 89–95.
4. Lohse, A.; Bols, M. Tetrahedron 1997, 53, 6917–6924.
5. Sinnot, M. L. Chem. Rev. 1990, 90, 1171–1202.
6. Saotome, C.; Kanie, Y.; Kanie, O.; Wong, C.-H. Bioorg.
Med. Chem. 2000, 8, 2249–2261 and references cited
therein.
7. Johns, B. A.; Jonhson, C. R. Tetrahedron Lett. 1997, 39,
749–752.
8. Le Merrer, Y.; Poitout, L.; Depezay, J.-C.; Dosbaa, I.;
Geoffroy, S.; Foglietti, M.-J. Bioorg. Med. Chem. 1997, 5,
519–533.
the opening reaction of the 3,4-O-acetonide-D-manno-
bis-epoxide 1b by the polyhydroxylated azepan 9 and
piperidine 11 (Scheme 3). Each of these hetero-
cyclic secondary amines 9 and 118 cleanly reacted with
the 3,4-O-isopropylidene- -manno-bis-epoxide 1b in
D
methanol at room temperature to afford 10a (75%) or
12a (95%). Subsequent treatment with aqueous trifl-
uoroacetic acid at room temperature and purification
by flash chromatography gave access to the pseudo-aza-
trisaccharide 10b (70%) or 12b (80%).
9. Satisfactory analytical and/or spectroscopic data were
obtained on all new compounds.
10. Poitout, L.; Le Merrer, Y.; Depezay, J.-C. Tetrahedron
Lett. 1995, 36, 6887–6890.
11. In the absence of the silyl protecting group of the primary
alcohol function, the N-amino cyclization also occurs in
similar yield, but we were unable to separate the corre-
sponding bicycles 5a and 6a for which P2=H by flash
chromatography.
In summary, we have synthesized four potential
inhibitors of glycosidases, pseudo-azadisaccharides A
and pseudo-azatrisaccharides B. Glycosidase inhibitory
evaluations are currently underway and will be reported
.
.