A Novel Diels-Alder Approach to Heavily Substituted Azasugars

Article abstract of DOI:10.1055/s-2003-43336

Diels-Alder cycloaddition of an appropriately substituted 1,4-oxazin-2-one with vinylene carbonate followed by the chemical manipulation of the bridged bicyclic lactone cycloadduct affords a heavily functionalised azasugar ring.

Full text of DOI:10.1055/s-2003-43336

LETTER
2341
A Novel Diels–Alder Approach To Heavily Substituted Azasugars
N
ovel Diels–Alder
a
A
pproach
m
T
o
H
eavily Substitu
y
ted
A
zasuga
a
rs r Afarinkia,*a Akmal Bahar,a Judi Neussb
a
Department of Chemistry, King’s College London, Strand, London, WC2R 2LS, UK
b
Celltech R&D Ltd., 208 Bath Road, Slough, Berkshire, SL1 3WE, UK
E-mail: kamyar.afarinkia@kcl.ac.uk
Received 29 August 2003
involve a ring forming reductive amination of an in situ
generated w-aminoaldehyde or ketone and have been ap-
plied to aza analogues of furanose as well as pyranose
sugars.9 The starting materials for these syntheses are
readily available carbohydrate molecules of the chiral
pool which results in the formation of the products in an
enantiomerically pure form. However, the requirement
for functional group protection makes the synthesis of cer-
tain analogues laborious and inefficient, particularly those
with a more elaborate pattern of substituents.
Abstract: Diels–Alder cycloaddition of an appropriately substitut-
ed 1,4-oxazin-2-one with vinylene carbonate followed by the chem-
ical manipulation of the bridged bicyclic lactone cycloadduct
affords a heavily functionalised azasugar ring.
Key words: Diels–Alder, cycloaddition, 2H-1,4-oxazin-2-one,
azasugar
‘Azasugars’ (also known as ‘iminosugars’) are structural
analogues of pyranose carbohydrates in which the ring ox-
ygen atom is replaced by a nitrogen atom. It has been pos-
tulated that because of their structural similarity to ‘true’
sugars, azasugars are recognized by carbohydrate han-
dling enzymes and receptors, which are nevertheless inca-
pable of chemically transforming them. The resulting
inhibition of the enzymes or antagonism of the receptors
leads to the numerous pharmacological properties which
are associated with azasugars.1 The opportunity they pro-
vide for ‘rational’ design of inhibitors and their already
proven ability to interfere with the biological processes in-
volving binding or turn-over of native polysaccharides
has made azasugars important biological tools and medic-
inal targets. This is demonstrated with the recent launch
of Zavesca (miglustat) for the treatment of Gaucher’s
disease.2 Crucial to these investigations are the develop-
ment of general as well as specific synthetic methods for
the preparation of azasugars. Indeed, the preparation and
biological properties of a number of naturally occurring
and synthetic azasugars are already documented in the
literature (Figure 1).3–8
O
R9
R6 R5
R7
R8
R4
OH
R3
R2
R5
R7
R6
R8
R9
O
O
R8
R6
R7
N
+
HN
R3
O
HO
Cl
N
R9
R1
R4
R2
R4
R5
1
2
Scheme 1
Therefore, other methods have also been investigated in-
cluding ring contraction of azepanes,10 photocyclisation,11
ring closure metathesis12 and ruthenium catalysed ring re-
arrangement,13 oxidation of a pyridone14 and oxidation of
aminomethylfurans,15 to name a few. Combinatorial syn-
thesis of libraries of azasugars are also reported.16 The
‘aza’ concept has been further expanded to ‘double aza’16b
analogues and amino aza sugars.17
Here, we report a new approach to the synthesis of azasug-
ars. The key advantage of this route is that it allows a
short, very efficient method for the controlled synthesis of
heavily and diversely substituted piperidines e.g. 1, in
These examples typify the many routes to azasugars al-
ready reported in the literature. The vast majority of these
OH
OH
OH
OH
OH
H
N
R1
OH
HO
CO2H HO
HO
OH
OH
CH2OH
AcNH
N
H
N
H
N
R2
7
Swainsonine2
Siastatin B6
calystegin B2
R1 = R2 = H; Fagomine3
R1 = OH; R2 = H; Deoxynojirimycin4
R1 = OH; R2 = n-Bu; Zavesca (Miglustat)
R1 = OH; R2 = CH2CH2OH; Miglitol5
Figure 1
SYNLETT 2003, No. 15, pp 2341–2344
0
1
.1
2
.2
0
0
3
Advanced online publication: 21.11.2003
DOI: 10.1055/s-2003-43336; Art ID: D21703ST
© Georg Thieme Verlag Stuttgart · New York

2342
K. Afarinkia et al.
LETTER
O
O
Later, Hoornaert demonstrated that 2-oxa-5-azabicyc-
lo[2.2.2]oct-5-en-3-one cycloadducts can also be isolated
from the cycloadditions23 and reported on their chemical
manipulation.24
4
O
O
O
Ha
8
i
ii
5
1
7
O
Hb
O
6O
EtO2C
EtO2C
EtO2C
O
O
These observations by Hoornaert encouraged us to con-
sider a route to substituted piperidines from the cycload-
dition of 1,4-oxazin-2-ones. The key advantage of such an
approach is the diversity of the substituents that can be in-
troduced into the piperidine nucleus (Scheme 1). In par-
ticular, this approach is expected to allow control over the
relative configuration of every carbon substituent on the
azasugar ring (Scheme 1).
3
O
4
O
J6,7 = 2.2 Hz; J5,8a = 1.6 Hz
iii, iv
NH2
NH2
O
OH
OH
v
O
HOH2C
EtO2C
O
OH
OH
5
O
O
Scheme 2 i) Vinylene carbonate, 100 °C, 83%; ii) H2, EtOAc, Pd/
C, 100%; iii) NH3 Dioxane, 78%; iv) (CF3CO2)2IPh, MeCN/H2O,
92%; v) LiAlH4, pyridine, 88%
Me
Me
Me
Me
O
O
O
i
O
N
N
N
+
O
O
O
Cl
Cl
Cl
Me
Me
O
O
general, and azasugars, in particular. Our methodology is
based on Diels–Alder cycloadditions of appropriately
substituted 1,4-oxazin-2-ones, e.g. 2, as an azadiene com-
ponent of the [4+2] cycloaddition to regio and stereoselec-
tively bring together the heterocyclic ring which through
further chemo- and stereoselective transformations would
afford target molecules (Scheme 1).
O
8a endo
8b exo
ii
O
O
Me
Me
O
N
O
O
N
+
O
O
Me
Me
Me
Me
O
O
O
This approach is particularly relevant to us due to the
close analogy between the 1,4-oxazin-2-ones and the 2H-
pyran-2-one rings. We have previously shown that the Di-
els–Alder cycloaddition of appropriately substituted 2H-
pyran-2-ones, such as 3, is an excellent starting point for
the synthesis of heavily functionalised carbocyclic six-
membered rings, and in particular carbasugars. This has
been exemplified by the conversion of 3 to epi-validamine
via its chemical manipulation of intermediates 4 and 5
(Scheme 2).18 Carbasugars, naturally occurring or syn-
thetic analogues of pyranose sugars in which the ring ox-
ygen atom is replaced by a methylene group, have a
similar biological relevance as azasugars.19 We expected
that our experience with the cycloadditions of 2H-pyran-
2-one would be very beneficial to addressing the issues
raised by this proposed methodology.
9 endo
10 exo
Scheme 4 i) Toluene, reflux, 24 h, vinylene carbonate (74%); ii)
Toluene, reflux, 48 h, Me4Sn, Pd(PPh3)4 (92%)
3,5-Dichloro-6-methyl-1,4-oxazin-2-one (6) was pre-
pared according to a previously published route by Hoor-
naert.20 Under palladium catalysed conditions, the 3-
chloro substituent was selectively substituted with a me-
thyl group to afford 7 (Scheme 3). There was no evidence
of substitution of the 5-chloro substituent in the NMR data
of the crude product. This azadiene underwent efficient
cycloaddition with vinylene carbonate to afford a 2.5:1
endo/exo ratio of cycloadducts 8a and 8b, which could not
be separated. The chloro substituents in each of the cy-
cloadducts 8a and 8b were replaced with methyl groups
using palladium catalysed Stille coupling. At this stage,
the endo and exo products 9 and 10 could be separated by
chromatography and were isolated in 65% and 27% yields
respectively (Scheme 4).
Cl
Me
O
O
NC
OH
i
ii
N
N
O
O
Me
Cl
Cl
The assignment of the endo and exo isomers proved to be
non-trivial. Based on our extensive expertise in the cy-
cloadditions of 2H-pyran-2-ones, we did expect the major
isomer to have endo configuration. However, the primary
evidence for this is deduced from the coupling constants
of the bridgehead protons in the NMR of the cycload-
ducts, using empirical rules set out previously.25 Since
neither cycloadduct contains a bridgehead proton (i.e.
both positions 1 and 4 are substituted), the assignment
could not be made. We therefore had to resort to a differ-
ent criterion for this assignment. Fortunately, Tomisawa26
and Posner24 had earlier demonstrated that in pyrone cy-
cloadducts, magnetic anisotropy due to the double bond
Me
Me
7
6
Scheme 3 i) Toluene, reflux, 24 h, (COCl)2; ii) Toluene, reflux, 24
h, SnMe4, Pd(PPh3)4
Surprisingly, a general and reliable method for the prepa-
ration of substituted 1,4-oxazin-2-ones was lacking until
Hoornaert published a straightforward route in 1989.20
Starting from cyanohydrins, Hoornaert prepared various
6-substituted 3,5-dichloro-1,4-oxazin-2-ones and subse-
quently demonstrated that they undergo [4+2] cyclo-
addition with concomitant loss of CO2 or cyanogen to
afford pyridines21 and 2H-pyran-2-ones, respectively.22
Synlett 2003, No. 15, 2341–2344 © Thieme Stuttgart · New York

LETTER
Novel Diels–Alder Approach To Heavily Substituted Azasugars
2343
significantly lowers the chemical shifts of the protons at case of reduction of 10, there are no steric encumbrance to
endo positions. Indeed, the chemical shifts for H-7 and H- either face of the imine and therefore, the reducing agent
8 protons in the assigned endo product are at 4.74 ppm and is introduced from either face. Interestingly, close analy-
4.78 ppm whereas the chemical shifts for H-7 and H-8 sis of the NMR of compounds 14a and 14b revealed no
protons in the assigned exo product are at 4.51 ppm. Fur- fine W coupling between H-6 and H-7. The absence of
ther evidence of the endo configuration of the major cy- this W coupling supports the proposed exo configuration
cloadduct was obtained after the next step.
of the starting material.
Hydride reduction of the imine functionality of the endo As before, methanolysis of 14a and 14b followed by basic
bridged ester 9 selectively gave amine 11. The stereo- hydrolysis of the two isomers gave azasugars 15a and 15b
chemistry of the reduction is based on our earlier observa- (Scheme 7).
tions during hydrogenolysis of the structurally related
compound 4.18 The lower face of the imine is sterically en-
cumbered by the carbonate bridge and therefore, the re-
ducing agent is introduced from the lactone bridge face
(Scheme 5). Furthermore, close analysis of the NMR of
compound 11 revealed a fine coupling (1.3 Hz) between
H-6 and H-7. This W coupling, is also a feature of reduc-
tion of cycloadducts of 2H-pyran-2-ones and is observed,
for instance, in compound 5 (Scheme 2). The presence of
this W coupling strongly supports the proposed configu-
ration of the reduction product.
O
OH
Me
Me
OH
O
HO
O
i, ii
HN
Me
Me
O
HO2C
N
Me
Me H
15ab
O
14ab
Scheme 7 i) MeONa (1 equiv), MeOH, reflux, 30 min;
ii) 2 M NaOH, r.t., 30 min (51%)
In conclusion, we have demonstrated that azasugars can
be prepared with control of relative stereochemistry via
the Diels–Alder cycloaddition of an appropriately substi-
tuted 1,4-oxazin-2-one with vinylene carbonate followed
by the chemical manipulation of the bridged bicyclic lac-
tone cycloadduct. We are currently working to expand the
scope of this novel methodology and to synthesise other
azasugars.
O
O
OH
Me
Me
Me
OH
O
O
HO
i
ii
HN
Me
N
7
6
RO2C
iii
N
Me H
Me
O
O
Me
Me
O
O
Me
O
O
R = Me 12
R = H 13
9
11
Scheme 5 i) HOAc, NaBH(OAc)3, r.t., 24 h (67%); ii) MeONa (1
equiv), MeOH, reflux, 30 min (49%); iii) 2 M NaOH, r.t., 1 h (quant.)
Acknowledgement
We thank Oxford GlycoSciences Ltd. (now part of Celltech R&D
Ltd.) for financial support.
Basic aqueous hydrolysis of the amine 11 was not effi-
cient, however, methanolysis of 11 to methyl ester 12 fol-
lowed by basic hydrolysis of compound 12 gave a very
good yield of azasugar 13.
References
(1) (a) Compain, P.; Martin, O. R. Bioorg Med. Chem. 2001, 9,
3077. (b) Watson, A. A.; Fleet, G. W. J.; Asano, N.;
Molyneux, R. J.; Nash, R. J. Phytochemistry 2001, 56, 265.
(c) Asano, N.; Nash, R. J.; Molyneux, R. J.; Fleet, G. W. J.
Tetrahedron: Asymmetry 2000, 11, 1645. (d) Picasso, S.
Chimia 1996, 50, 648.
(2) Cox, T.; Lachmann, R.; Hollak, C.; Aerts, J.; van Weely, S.;
Hrebicek, M.; Platt, F.; Butters, T.; Dwek, R.; Moyses, C.;
Gow, I.; Elstein, D.; Zimran, A. Lancet. 2000, 355, 1481.
(3) Lindsay, K. B.; Pyne, S. G. J. Org. Chem. 2002, 67, 7774.
(4) (a) Desire, J.; Dransfield, P. J.; Gore, P. M.; Shipman, M.
Synlett 2001, 1329. (b) Banba, Y.; Abe, C.; Nemoto, H.;
Kato, A.; Adachi, I.; Takahata, H. Tetrahedron: Asymmetry
2001, 12, 817. (c) Degiorgis, F.; Lombardo, M.; Trombini,
C. Synthesis 1997, 1243.
In contrast to the stereoselective reduction of imine 9, hy-
dride reduction of the imine functionality of the exo
bridged ester 10 gave a mixture of two amines 14a and
14b in a 3:1 ratio which could not be separated
(Scheme 6). The lack of selectivity in the reduction of 10
is again consistent with our previous observations that exo
cycloadducts of 2H-pyran-2-ones undergo non-stereose-
lective hydrogenation of their alkene function.18 In the
O
O
O
Me
Me
Me
O
O
O
O
O
i
O
N
HN
+
HN
Me
(5) (a) Somfai, P.; Marchand, P.; Torsell, S.; Lindstrom, U. M.
Tetrahedron 2003, 59, 1293. (b) O’Brien, J. L.; Tosin, M.;
Murphy, P. V. Org. Lett. 2001, 3, 3353.
Me
O
O
O
6
6 Me
Me
7
7
Me
O
O
O
14b
10
14a
Me
(6) Scott, L. J.; Spencer, C. M. Drugs 2000, 59, 521.
(7) Knapp, S.; Zhao, D. Org. Lett. 2000, 2, 4037.
(8) Garcia-Moreno, M. I.; Benito, J. M.; Mellet, C. O.;
Fernandez, J. M. G. J. Org. Chem. 2001, 66, 7604.
Scheme 6 i) HOAc, NaBH(OAc)3, r.t., 24 h (79%)
Synlett 2003, No. 15, 2341–2344 © Thieme Stuttgart · New York

2344
K. Afarinkia et al.
LETTER
(9) For some recent examples see: (a) Batey, R. A.; Mackay, D.
B. Tetrahedron Lett. 2000, 41, 9935. (b) Schuster, M.; He,
W.-F.; Blechert, S. Tetrahedron Lett. 2001, 42, 2289.
(c) Enders, D.; Kirchhoff, J. H. Synthesis 2000, 2099.
(d) Wrodnigg, T. M.; Gaderbauer, W.; Greimel, P.;
Haeusler, H.; Sprenger, F. K.; Stuetz, A. E.; Virgona, C.;
Withers, S. G. J. Carbohydr. Chem. 2000, 19, 975.
(e) Izquierdo, I.; Plaza, M. T.; Robles, R.; Mota, A. J. Eur. J.
Org. Chem. 2000, 2071.
(10) (a) Poitout, L.; LeMerrer, Y.; Depezay, J. C. Tetrahedron
Lett. 1996, 37, 1609. (b) Poitout, L.; LeMerrer, Y.;
Depezay, J. C. Tetrahedron Lett. 1996, 37, 1613.
(11) (a) Pandey, G.; Kapur, M. Org. Lett. 2002, 4, 3883.
(b) Pandey, G.; Kapur, M. Tetrahedron Lett. 2000, 41,
8821. (c) Pandey, G.; Kapur, M. Synthesis 2001, 41, 1263.
(12) Overkleeft, H. S.; Bruggeman, P.; Pandit, U. K. Tetrahedron
Lett. 1998, 39, 3869.
(17) (a) Kondo, K.; Adachi, H.; Nishimura, Y.; Takeuchi, T. Nat.
Prod. Lett. 2001, 15, 371. (b) Shitara, E.; Nishimura, Y.;
Kojima, F.; Takeuchi, T. Bioorg. Med. Chem. 2000, 8, 343.
(c) Kondo, K.; Adachi, H.; Shitara, E.; Kojima, F.;
Nishimura, Y. Bioorg. Med. Chem. 2001, 9, 1091.
(d) Berges, D. A.; Fan, J. M.; Devinck, S.; Liu, N. N.; Dalley,
N. K. Tetrahedron 1999, 55, 6759. (e) Berges, D. A.;
Zhang, N.; Hong, L. W. Tetrahedron 1999, 55, 14251.
(18) Afarinkia, K.; Mahmood, F. Tetrahedron 1999, 55, 3129.
(19) (a) Vogel, P. Chimia 2001, 55, 359. (b) Yuasa, H.;
Hashimoto, H. Trends Glycosci. Glycotech. 2001, 13, 31.
(20) Meerpool, L.; Hoornaert, G. J. Synthesis 1990, 905.
(21) Vanaken, K. J.; Lux, G. M.; Deroover, G. G.; Meerpoel, L.;
Hoornaert, G. J. Tetrahedron 1994, 50, 5211.
(22) Tutonda, M. G.; Vandenberghe, S. M.; Van Aken, K. J.;
Hoornaert, G. J. J. Org. Chem. 1992, 57, 2935.
(23) Vanaken, K. J.; Lux, G. M.; Deroover, G. G.; Meerpoel, L.;
Hoornaert, G. J. Tetrahedron 1996, 52, 2591.
(24) Wu, X.; Dubois, K.; Rogiers, J.; Toppet, S.; Compernolle,
F.; Hoornaert, G. J. Tetrahedron 2000, 56, 3043.
(13) Voigtmann, U.; Blechert, S. Org. Lett. 2000, 2, 3971.
(14) Amat, M.; Llor, N.; Huguet, M.; Molins, E.; Espinosa, E.;
Bosch, J. Org. Lett. 2001, 3, 3257.
(15) Koulocheri, S. D.; Magiatis, P.; Skaltsounis, A. L.;
Haroutounian, S. A. Tetrahedron 2000, 56, 6135.
(16) (a) Lohse, A.; Jensen, K. B.; Lundgren, K.; Bols, M. Bioorg.
Med. Chem. 1999, 7, 1965. (b) Lohse, A.; Jensen, K. B.;
Bols, M. Tetrahedron Lett. 1999, 40, 3033.
(25) Afarinkia, K.; Vinader, V.; Nelson, T. D.; Posner, G. H.
Tetrahedron 1992, 48, 9111.
(26) Tomisawa, H.; Fujita, R.; Noguchi, K.; Hongo, H. Chem.
Pharm. Bull. 1970, 18, 941.
Synlett 2003, No. 15, 2341–2344 © Thieme Stuttgart · New York