Trichloro-oxazolines as Activated Donors for Aminosugar Coupling

Article abstract of DOI:10.1021/ol0359620

(Equation presented) Starting from tri-O-acetyl-o-glucal, a combination of the Overman rearrangement and subsequent dihydroxylation produces a range of aminosugars. These can be activated by formation of the corresponding trichloro-oxazolines, which are e

Full text of DOI:10.1021/ol0359620

ORGANIC
LETTERS
2003
Vol. 5, No. 26
4995-4998
Trichloro-oxazolines as Activated
Donors for Aminosugar Coupling
,†
Timothy J. Donohoe,* James G. Logan,‡ and David D. P. Laffan§
Dyson Perrins Laboratory, UniVersity of Oxford, South Parks Road,
Oxford, OX1 3QY, UK, Department of Chemistry, UniVersity of Manchester,
Oxford Road, Manchester, M13 9PL, UK, and AstraZeneca Pharmaceuticals,
Silk Road Business Park, Macclesfield, SK10 2NA, UK
timothy.donohoe@chem.ox.ac.uk
Received October 8, 2003
ABSTRACT
Starting from tri-O-acetyl-D-glucal, a combination of the Overman rearrangement and subsequent dihydroxylation produces a range of
aminosugars. These can be activated by formation of the corresponding trichloro-oxazolines, which are excellent glycosyl donors as they
form disaccharides with good (trans) stereoselectivity under mild conditions. Propagation of these trichloro-oxazolines gave trisaccharides
that can then be dehalogenated under a variety of conditions.
2-Amino-2-deoxy sugars are widely distributed throughout
Nature occurring as key constituents of many glycoproteins
Scheme 1
and glycolipids.1 They are also essential components of
various other natural products such as the antibiotics strep-
tothricin, streptomycin, and adriamycin.2 As a consequence,
there is considerable interest in new strategies for the
synthesis of such molecules, especially 2-amino-2-deoxy
oligosaccharides.
Recently, we disclosed a strategy for accessing a range of
2-amino-2-deoxy sugars that was based upon Ferrier gly-
cosidation (tri-O-acetyl-D-glucal f 1),3 Overman rearrange-
ment (1 f 2/3), and subsequent dihydroxylation, Scheme
1.4 While the stereochemistry at C-4 in structure 1 is initially
† University of Oxford.
‡ University of Manchester.
§ AstraZeneca.
(1) Banoub, J.; Boullanger, P.; Lafont, D. Chem. ReV. 1992, 92, 1167.
(2) See: Carbohydrates Structure and Biology; Lehmann, L.; Thieme:
Stuttgart, 1998.
(3) Ferrier, R. J. J. Chem. Soc. C 1964, 5443. For a recent reference,
see: Das, S. K.; Reddy, K. A.; Abbineni, C.; Roy, J.; Rao, K. V. L. N.;
Sachwani, R. H.; Iqbal, J. Tetrahedron Lett. 2003, 44, 4507.
(4) Donohoe, T. J.; Blades, K.; Helliwell, M. Chem. Commun. 1999,
1733. See Supporting Information.
set as shown, it could also be readily inverted by a Mitsunobu
reaction.4 As the Overman rearrangement is stereospecific,
this meant that we could fully control the stereochemistry
at C-2 in trichloroacetamides 2 and 3 and gain access to many
allosamine and talosamine derivatives not readily accessible
by other methods.
10.1021/ol0359620 CCC: $25.00 © 2003 American Chemical Society
Published on Web 11/22/2003

To ascertain whether compounds of general structure 4
might prove to be useful for oligosaccharide synthesis, we
recently investigated ways of converting them into a range
of trichloro-oxazoline glycosyl donors so that we could study
subsequent coupling with glycosyl acceptors.
Initially, we chose the allosamine-derived sugar 5 as an
example on which to test our synthetic plans, Scheme 3. This
Scheme 3a
It should be noted that the use of trichloroacetamide-
protected aminosugars as glycosyl donors was originally
established by Jacquinet,5 but that there are only two sugar-
derived trichloro-oxazolines reported in the literature (derived
from glucosamine5 and galactosamine6). Furthermore, the
formation of C-2 mannose-configured trichloro-oxazolines
was unknown. Coupling reactions of isolated trichloro-
oxazoline donors are rare, and only the glucosamine deriva-
tive, prepared from the anomeric O-acetate by Jacquinet, had
been subjected to a handful of glycosylations.5,7
Herein, we now report success in our endeavor, having
successfully accomplished the synthesis of several rare
disaccharide derivatives containing allosamine and talosamine
substructures.
Our general strategy for preparing the appropriate trichloro-
oxazoline donors from 1 exploited a â-naphthylmethyl8 as
the O(1) protecting group and used the Overman/dihydroxy-
lation tactic to fashion the appropriately O-hydroxylated
amino sugar; for example, see talosamine, Scheme 2. Next,
Scheme 2
a Reagents and conditions: (a) DDQ, H2O, CH2Cl2; (b) Ms2O,
MeCN, Et3N; (c) SugOH (1-1.2 equiv), TMSOTf (20 mol %),
-30 to -10 °C, CH2Cl2.
was prepared via the previously published route (see Scheme
1), which involves a directed dihydroxylation reaction using
catalytic osmium tetroxide.4 Removal of the NAP group was
straightforward using DDQ,9 and we then chose methane-
sulfonic anhydride (Ms2O) as a mild and effective reagent
for oxazoline formation 6. The trichloro-oxazoline 6 was
stable at room temperature for 1-2 days and for longer
periods in the freezer.
Next we investigated the glycosylation of 6 using three
sugar acceptors chosen to provide useful challenges. The
acceptors were 7 (protected allosamine),10 8 (protected
glucose),11 and 9 (galactose diisopropylidene acetal). Pleas-
ingly, donor 6 glycosylated easily with the three acceptors
(1-1.2 equiv) at temperatures between -30 and -10 °C
using catalytic TMSOTf as an activator.12,5 In each case, a
single disaccharide product 10-12 was isolated from the
reaction mixture in 83-87% yield.
Clearly, trichloro-oxazolines are promising derivatives for
glycosyl coupling because the halo groups increase the
reactivity of the donor relative to a methyl-substituted
oxazoline (which can require vigorous conditions for cou-
pling).13 The enhanced reactivity of a trichloro-oxazoline
we detached the naphthylmethyl group from the protected
product with DDQ (stage 1, Scheme 2). Oxazoline formation
was then effected by activation of the anomeric hydroxyl
group (stage 2). Thereafter, the trichloro-oxazoline was
coupled with a glycosyl acceptor after activation with
TMSOTf (stage 3). Finally, we converted the trichloro-
acetamide unit into a N-acetyl group (since this is the
naturally occurring derivative of most C-2 aminosugars).
(5) Blatter, G.; Beau, J.-M.; Jacquinet, J.-C. Carbohydr. Res. 1994, 260,
189.
(6) Bartek, J.; Mu¨ller, K. Carbohydr. Res. 1998, 308, 259. Be´lot, F.;
Jacquinet, J.-C. Carbohydr. Res. 2000, 325, 93.
(7) Sherman, A. A.; Olga, N. Y.; Mironov, Y. V.; Sukhova, E. V.;
Shashkov, A. S.; Menshov, V. M.; Nifantiev, N. E. Carbohydr. Res. 2001,
336, 13.
(8) Sarkar, A. K.; Rostand, K. S.; Jain, R. K.; Matta, K. L.; Esko, J. D.
J. Biol. Chem. 1997, 272, 25608. Gaunt, M. J.; Yu, J.; Spencer, J. B. J.
Org. Chem. 1998, 63, 4172.
(9) Xia, J.; Abbas, S. A.; Locke, R. D.; Piskork, C. F.; Alderfer, J. L.;
Matta, K. L. Tetrahedron Lett. 2000, 41, 169.
(10) See Supporting Information.
(11) Bundle, D. R.; Purse, B. W.; Nitz, M.; Org. Lett. 2000, 2, 2939.
4996
Org. Lett., Vol. 5, No. 26, 2003

activating group can be explained by analogy to the anomeric
trichloroacetimidates introduced by Schmidt.14
trichloro-oxazolines would yield the R-linked disaccha-
rides.
The configuration at the newly formed anomeric center
As predicted, the glycosylation of 14 and 19 took place
readily at low temperature and again they each gave a single
disaccharide product in 85-91% yield. For each disaccharide
1
was proven to be â by JCH coupling constants that are
particularly diagnostic of the stereochemistry at C-1.15 In this
case, the large 3J coupling between C1-H and C2-H (10-
that was prepared, the JCH coupling constant showed that
1
3
12, J 7-7.6 Hz, trans diaxial) is also indicative of the
the configuration at the new anomeric center was R.16
Extensive NOE experiments also helped to confirm the
structures as shown, and in most cases NOESY cross-peaks
between the two sugar rings established the connectivity of
the disaccharides. Although compound 21 could be isolated
from the glycosylation reaction, it was contaminated (e10%)
with a byproduct of unknown composition.
stereochemistry indicated. The outcome from coupling is
consistent with an SN2-like reaction between glycosyl donor
and (activated) glycosyl acceptor
Next, we examined the effect of changing the configuration
at C-2, by preparing and then glycosylating two donors
derived from talosamine and altrosamine, Scheme 4 (13 and
To further extend this methodology, we investigated the
propagation of trichloro-oxazolines to make trisaccharide-
derived aminosugars, Scheme 5. The disaccharide 10 was
Scheme 4a
Scheme 5a
a Reagents and conditions: (a) DDQ, H2O, CH2Cl2; (b) Ac2O,
py, DMAP; (c) NH2NH2‚CH3CO2H, DMF; (d) Ms2O, MeCN, Et3N;
(e) 7 (1.1 equiv), TMSOTf (20 mol %), -10 °C, CH2Cl2.
chosen for deprotection, formation of a trichloro-oxazoline,
and subsequent glycosylation. It is important to note that
the â-configured dimer of allosamine 10 is the disaccharide
portion of the natural product allosamidin (a potent inhibitor
of chitin synthase)17 and so this sequence will also define a
possible synthetic route to allosamidin whereby the aglycone
could be attached (via 26) to this sugar portion in subsequent
synthetic studies.
a Reagents and conditions: (a) DDQ, H2O, CH2Cl2; (b) Ms2O,
MeCN, Et3N; (c) SugOH (1-1.2 equiv), TMSOTf (20 mol %),
-30 to -10 °C, CH2Cl2.
18 were both prepared by using the general chemistry in
Scheme 14). In these cases, an SN2 displacement on the
This methodology worked as planned, and selective
deprotection of 10 with DDQ gave the diol 23, which was
(12) BF3‚OEt2 can also be used as an activator, although the couplings
are slower.
(13) See: Shrader, W. D.; Imperiali, B. Tetrahedron Lett. 1996, 37, 599.
Wittmann, V.; Lennartz, D. Eur. J. Org. Chem. 2002, 8, 1363.
(14) Schmidt, R. R. Angew. Chem., Int. Ed. Engl. 1986, 25, 212.
(15) 1JCH (Hz) 10 (C1, 176; C1′, 167); 11 (C1, 157; C1′, 162); 12 (C1,
178; C1′, 167) numbering from the reducing end first. Block, K.; Pedersen,
C. J. Chem. Soc., Perkin Trans. 2 1974, 293.
(16) 1JCH (Hz) 15 (C1, 159; C1′, 179); 16 (C1, 177; C1′, 176); 17 (C1,
181; C1′, 176); 20 (C1, 157; C1′, 172); 21 (C1, 177; C1′, 173); 22 (C1,
182; C1′, 170) numbering from the reducing end first.
(17) Sakuda, S.; Isogai, A.; Matsumoto, S.; Susuki, A.; Koeski, K.;
Tetrahedron Lett. 1986, 27, 2475.
Org. Lett., Vol. 5, No. 26, 2003
4997

subsequently peracetylated (24) before the anomeric OAc
was cleaved by reaction with hydrazine acetate to give 25.18
Subsequent formation of trichloro-oxazoline 26 occurred
smoothly, as did stereoselective formation of trisaccharide
27 with acceptor 7, Scheme 5.19
Scheme 6a
Finally, for this methodology to be useful, access to a
variety of conditions for conversion of the trichloroacetyl
group into the naturally occurring N-acetyl derivative is
required.
Originally, Jacquinet reported radical-based chemistry
(Bu3SnH) to remove the three chlorines from a trichloro-
acetamide, and we also found this method to be effective
with the saccharides that we had prepared, Scheme 6.5 If
one is to use this methodology in a synthetic sequence, then
ideally there should be plenty of alternative methods for
accomplishing the interconversion of a TCA group into its
parent acetamide. To this end, we investigated a range of
different methods for reducing 12 and found that, in addition
to tin hydride, hydrogenolysis20 (55 psi) and cleavage with
NaOH (followed by reacetylation)3 were also viable condi-
tions, Scheme 6.
a Reagents and conditions: (a) Bu3SnH, AIBN, ∆; (b) H2, Pd-
C, MeOH, 55 psi; (c) NaOH, EtOH, then Ac2O, py, DMAP.
compatible with this sequence and give high stereoselectivity
for the trans arrangement at C1-C2. Moreover, we have
shown that the substituted oxazolines can be elaborated to
form trisaccharides and finally that there are a variety of
different methods for the reduction of the TCA group to an
acetamide.
To conclude, we have defined a new protecting group
(ONAP) arrangement for oligosaccharide synthesis that is
introduced via the Ferrier rearrangement and compatible with
subsequent Overman rearrangement and a dihydroxylation
sequence. By using DDQ to deprotect the ONAP group, we
were able to prepare trichloro-oxazolines from rare amino
sugars and then investigated the glycosylation of these
activated derivatives. Both possible configurations at C-2 are
Acknowledgment. We thank the EPSRC and AstraZen-
eca for funding this project. We also thank K. Blades for an
experimental contribution and G. Sharman for assistance with
NMR.
(18) Excoffier, G.; Gagnaire, D.; Utille, J.-P. Carbohydr. Res. 1975, 39,
368.
Supporting Information Available: Full experimental
procedures and spectroscopic data. This material is available
free of charge via the Internet at http://pubs.acs.org.
(19) 1JCH (Hz) 27 (C1, 177; C1′, 163; C1′′, 162) numbering from the
reducing end first.
(20) Armstrong, P. L.; Coull, I. C.; Hewson, A. T.; Slater, M. J.
Tetrahedron Lett. 1995, 36, 4311.
OL0359620
4998
Org. Lett., Vol. 5, No. 26, 2003