Design and Development of a Common Synthetic Strategy for a Variety of 1-N-Iminosugars

Article abstract of DOI:10.1021/ol026711e

(Matrix Presented) A new synthetic strategy has been developed for a general approach toward the synthesis of a variety of 1-N-iminosugar-type glycosidase inhibitors utilizing precursors developed by the PET-mediated cyclization of α-trimethylsilylmethyla

Full text of DOI:10.1021/ol026711e

ORGANIC
LETTERS
2002
Vol. 4, No. 22
3883-3886
Design and Development of a Common
Synthetic Strategy for a Variety of
1-N-Iminosugars
Ganesh Pandey* and Manmohan Kapur
DiVision of Organic Chemistry (Synthesis), National Chemical Laboratory,
Pune 411008, India
pandey@ems.ncl.res.in
Received August 9, 2002
ABSTRACT
A new synthetic strategy has been developed for a general approach toward the synthesis of a variety of 1-N-iminosugar-type glycosidase
inhibitors utilizing precursors developed by the PET-mediated cyclization of r-trimethylsilylmethylamine radical cation to a tethered π-functionality.
Glycosidase inhibitors, also called the “sugar-shaped alka-
loids”, are carbohydrate analogues in which one or more of
the oxygen atoms have been substituted by a nitrogen and
are found to be widespread in the plants and micro-
organisms.1
Due to the tremendous potential of these unique molecules
in studying the biological functions of oligosaccharides2 and
by virtue of these being postulated as therapeutics for a
variety of carbohydrate-mediated diseases such as HIV,3
diabetes,4 hepatitis,5 cancer,6 and Gaucher’s disease7 and viral
infections such as influenza,8 an interest in the chemistry,
biochemistry, and pharmacology of these compounds has
been sparked.9 Although a plethora of natural as well as
synthetic R-glycosidase inhibitors are known, the develop-
ment of anomer-selective â-glycosidase inhibitors took place
only in the past decade, pioneered by the studies from the
groups of Bols9f and Ichikawa.9g
These carbohydrate mimics, in which the anomeric carbon
is substituted by a nitrogen, have been termed 1-azasugars
or 1-N-iminosugars. A variety of transition state analogues
such as 1-4 have been designed and studied (Figure 1).
(1) Asano, N.; Nash, R. J.; Molyneux, R. J.; Fleet, G. W. J. Tetra-
hedron: Asymmetry 2000, 11, 1645.
(2) Dwek, R. A. Chem. ReV. 1996, 96, 683.
(3) Ratner, L.; Heyden, N. V.; Dedera, D. Virology 1991, 181, 180 and
references therein.
(4) (a) Anzeveno, P. B.; Creemer, L. J.; Daniel, J. K.; King, C. -H. R.;
Liu, P. S. J. Org. Chem. 1989, 54, 2539. (b) Balfour, J. A.; McTavish, D.
Drugs 1993, 46, 1025.
Figure 1. Some of the â-glycosidase inhibitors developed.
(5) Zitzman, N.; Mehta, A. S.; Carrouee´, S.; Butters, T. D.; Platt, F. M.;
McCauley, J.; Blumberg, B. S.; Dwek, R. A.; Block, T. M. PNAS 1999,
96, 11878.
(6) Gross, P. E.; Baker, M. A.; Carver, J. P.; Dennis, J. W. Clin. Cancer
Res. 1995, 1, 935.
Isofagomine (1a) and noeuromycin (1b) (D-glucose-type 1-N-
iminosugars) have been found to be extremely potent
â-glucosidase inhibitors.9f Also notable is compound 4, the
5-hydroxy analogue of isofagomine, which has been found
to be an inhibitor of glycolipid biosynthesis.10
(7) Alper, J. Science 2001, 291, 2338.
(8) Kim, C. U.; Lew, W.; Williams, M. A.; Liu, H.; Zhang, L;
Swaminathan, S.; Bischofberger, N.; Chen, M. S.; Mendel, D. B.; Tai, C.
Y.; Laver, W. G.; Stevens, R. C. J. Am. Chem. Soc. 1997, 119, 681.
10.1021/ol026711e CCC: $22.00 © 2002 American Chemical Society
Published on Web 10/02/2002

Although these molecules look relatively simple, their
syntheses have not been straightforward. The greatest hurdle
in the synthesis of these piperidines has been the introduction
of the aminomethyl group next to a stereocenter. Although
many intelligent approaches for the syntheses of these have
been reported,9f,g,11 a general synthetic strategy toward these
is unknown. Approaches utilizing the chiral pool have had
to change the starting material for each of these compounds.
A close look at these 1-N-iminosugars reveals a general
substitution pattern, as in 5, varying only in the stereochem-
istry of the hydroxy groups at C-3 and C-4 and the nature
of the R1 and R2 groups (Figure 2). Therefore, we wondered
area toward the development of new azasugars and also to
broaden the scope of our strategy, we report herein the further
development of this strategy for the synthesis of a variety
of 1-N-iminosugars.
The successful synthesis of both isomers of isofagomine
was followed by the synthesis of the 5-hydroxy analogue of
5-epi-isofagomine (Scheme 1).
Scheme 1a
a Reagents and conditions: (a) OsO4, NMO, pyridine, acetone-
water (9:1), from 0 °C to room temperature, 24 h, 95%; (b)
Pd(OH)2, H2, EtOH, 65 psi, 6 h, 90%; (c) HCl, MeOH, rt, 4 h,
quant.
Figure 2. General precursors of D- and L-threo classes.
whether a precursor of the type 6 could be used for the design
of a general route for the synthesis of these molecules. By
having the correct stereochemistry at C-3 and C4 and by
functionalizing the exocyclic double bond, one can have an
access to a variety of these azasugars. In this context, we
designed and successfully synthesized 7 and 8, precursors
of the D- and L-threo classes, respectively.12 These precursors
were obtained by utilizing our methodology13 for synthesiz-
ing cyclic amines via PET (photoinduced electron transfer)-
promoted cyclizations of R-trimethylsilylmethylamine radical
cation to a tethered π-functionality.
Osmium tetroxide dihydroxylation of 7 afforded 9 as a
single diastereomer. The stereochemical outcome was ad-
judged by COSY, HETCOR, and NOESY experiments. The
benzyl and acetonide protecting groups were removed to
afford 11.
We also synthesized 5′-deoxy-5-epi-isofagomine (13) from
7. One-pot olefin reduction, N-debenzylation, and acetonide
deprotection of 7 led to a 80:20 mixture of nonseparable
diastereomers. These diastereomers, however, could be
separated by careful column chromatography as their cor-
responding N-Boc derivatives. The stereochemistry at this
stage was established by COSY. Removal of the N-Boc
moiety afforded 13 (Scheme 2).
We also synthesized (+)- and (-)-isofagomine (1) utilizing
these precursors.12 As a part of our continuing study in this
(9) For some excellent accounts on glycosidases and their inhibitors,
see: (a) Zechel, D. L.; Withers, S. G. Acc. Chem. Res. 2000, 33, 11. (b)
Heightman, T. D.; Vasella, A. T. Angew. Chem., Int. Ed. 1999, 38, 750.
(c) Ganem, B. Acc. Chem. Res. 1996, 29, 340. (d) Sears, P.; Wong, C. -H.
Angew. Chem., Int. Ed. 1999, 38, 2300 (e) Stu¨tz, A. E. Iminosugars as
Glycosidase Inhibitors: Nojirimycin and Beyond; Wiley-VCH: Weinheim,
Germany, 1999. (f) Lillelund, V. H.; Jensen, H. H.; Liang, X.; Bols, M.
Chem. ReV. 2002, 102, 515. (g) Ichikawa, Y.; Igarashi, Y.; Ichikawa, M.;
Suhara, Y. J. Am. Chem. Soc. 1998, 120, 3007.
Scheme 2a
(10) Ichikawa, M.; Igarashi, Y.; Ichikawa, Y. Tetrahedron Lett. 1995,
36, 1767.
(11) For some recent syntheses, see: (a) Andersch, J.; Bols, M. Chem.
Eur. J. 2001, 7, 3744 and references therein. (b) Zhao, G.; Deo, U. C.;
Ganem, B. Org. Lett. 2001, 3, 201. (c) Kim, Y. J.; Ichikawa, M.; Ichikawa,
Y. J. Org. Chem. 2000, 65, 2599. (d) Mehta, G.; Mohal, N. Tetrahedron
Lett. 2000, 41, 5747. (e) Liu, H.; Liang, X.; Søhoel, H.; Bu¨low, A.; Bols,
M. J. Am. Chem. Soc. 2001, 123, 5116. (f) Liang, X.; Lohse, A.; Bols, M.
J. Org. Chem. 2000, 65, 7432. (g) Søhoel, H.; Liang, X.; Bols, M. J. Chem.
Soc., Perkin Trans. 1 2001, 1584.
(12) (a) Pandey, G.; Kapur, M. Tetrahedron Lett. 2000, 41, 8821. (b)
Pandey, G.; Kapur, M. Synthesis, 2001, 1263.
(13) (a) Pandey, G.; Kumaraswamy, G.; Bhalerao, U. T. Tetrahedron
Lett. 1989, 30, 6059. (b) Pandey, G.; Reddy, G. D.; Kumaraswamy, G.
Tetrahedron 1994, 50, 8185.
a Reagents and conditions: (a) (i) Pd/ C, MeOH, HCl, H2, 1atm,
rt, 12 h, 89%; (ii) (Boc)2O, TEA, DCM, rt, 48 h, 75%. (b) HCl,
MeOH, from 0 °C to room temperature, 4 h, quant.
3884
Org. Lett., Vol. 4, No. 22, 2002

Although a fair degree of diastereoselection was obtained
in the above case, we were not completely satisfied by the
outcome. We decided to modify the approach for this
molecule, utilizing our PET cyclization strategy13 for the
construction of this substituted piperidine. The modified
synthesis began with alcohol 14,14 also derived from
D-(-)-tartaric acid (Scheme 3).
medicine as a diuretic, antipyretic, emmenagogue, and
antidiabetic agent. Triols 17-20 have also been synthesized
by Ganem16 and others17 and have been shown to be good
selective inhibitors of glycosidases.
We realized that our precursors 7 and 8 could be ideal for
the synthesis of these molecules. The synthesis of des(hydroxy-
methyl)deoxymannojirimycin (19) commenced with diol 9,
previously synthesized from 7. In anticipation that the amine
would be affected by periodate oxidation, we replaced the
benzyl group in 9 by an N-Boc moiety (Scheme 4). However,
Scheme 3a
Scheme 4a
a Reagents and conditions: (a)(i) TsCl, pyridine, DCM, rt, 24 h,
95%; (ii) PhCH2NHCH2TMS, Cs2CO3, TBAI, CH3CN, reflux, 72
h, 58%. (b) hν, DCN, 2-PrOH, 2 h, 55%. (c) Pd(OH)2 on C, HCl,
MeOH, H2, 1 atm, 28 h, quant.
a Reagents and conditions: (a) (Boc)2O, TEA, DCM, from 0 °C
to room temperature, overnight, 80%; (b) NaIO4, EtOH-H2O (4:
1), rt, 30 min, 70%; (c) NaBH4, MeOH, rt, 36 h, then saturated
NaCl, rt, 24 h, 25%.
Tosylation followed by nucleophilic substitution utilizing
PhCH2NHCH2TMS afforded 15. To our pleasant surprise,
amine 15, when subjected to PET cyclization conditions,
afforded 16 as a single diastereomer (>97%). Removal of
the benzyl and acetonide groups afforded 13 in good yield.
The above successes led us to set our sight on the synthesis
of some of the most important 3,4,5-piperidine triols (Figure
3). These trihydroxy piperidines are regarded as the deriva-
to our surprise and for reasons unknown, the sodium
borohydride reduction of the ketone 22 failed to give good
yield, and therefore this route was abondoned.
The above failure forced us to revert back to diol 9, which
upon periodate oxidation, as shown in Scheme 5, gave 24
Scheme 5a
Figure 3. Some of the most important 3,4,5-piperidine triols.
tives of their parent deoxynojirimycin and have been shown
to possess moderate to good glycosidase inhibitory activity.
Careful consideration of these structural moieties reveals that
these molecules can also be classified as 1-N-iminosugar-
type glycosidase inhibitors, although they do not resemble
any of the existing pyranose sugars.
Triols 17, 18, and 20 were isolated from Eupatorium
fortunei TURZ by Kusano and co-workers15 in 1995 and have
been shown to be the active components of the extracts of
this plant, traditionally used in Chinese and Japanese folk
a Reagents and conditions: (a) NaIO4, EtOH-H2O (4:1), rt, 1
h, 80%; (b) NaBH4, MeOH, rt, 40 h, then saturated NaCl, rt, 24 h,
85%; (c) Pd(OH)2 on C, HCl, MeOH, H2, 1 atm, rt, 36 h, quant.
in 80% yield. However, since 24 was found to be extremely
unstable, it was quickly subjected to sodium borohydride
reduction, which afforded 25 in a 90:10 diastereomeric ratio.
(14) Kang, S. K.; Lee, S. B. Chem. Commun. 1998, 761.
(15) Sekioka, T.; Shibano, M.; Kusano, G. Nat. Med. 1995, 49, 332.
Org. Lett., Vol. 4, No. 22, 2002
3885

One-pot N-debenzylation and acetonide removal afforded
19, which could be purified completely by column chroma-
tography as a free base and was converted back to its
hydrochloride salt for spectral characterization. The stereo-
chemistry of the newly formed hydroxy group was proved
by COSY experiments. The spectral data and optical rotation
for 19 were found to be in close agreement with those
reported.16
We moved on further to synthesize des(hydroxymethyl)-
deoxynojirimycin (18). This was synthesized by inverting
the stereochemistry of the hydroxy group of 26 (ent-25) using
Mitsunobu conditions (Scheme 6). Removal of the benzyl
In a similar manner, ent-11, ent-13, and des(hydroxy-
methyl)deoxygalactonojirimycin (20) were synthesized from
the L-threo precursor 8.
In summary, we have demonstrated the general nature of
our approach by synthesizing various 1-N-iminosugars from
a common precursor. It is obvious that the D- and L-erythro
precursors (in which the C-3 and C-4 hydroxy groups are
syn) would lead to other 1-N-iminosugars of the required
stereochemistry. Currently, the new molecules 11 and 13 and
their enantiomers are being studied for their inhibitory
activity, and the results will be published in due course.
Acknowledgment. We thank Dr. P. R. Rajmohanan, Mrs.
U. D. Phalgune and Ms. Rupali Bhagwat for the special
NMR experiments. M.K. thanks CSIR, New Delhi for the
award of Research Fellowship.
Scheme 6a
Supporting Information Available: Spectral data and
1
copies of H and 13C NMR spectra of compounds 11, 13,
15, 16, 18, and 19. This material is available free of charge
via the Internet at http://pubs.acs.org.
a Reagents and conditions: (a) (i) diisopropyl azodicarboxylate,
PPh3, p-nitrobenzoic acid, THF, rt, overnight; (ii) LiOH, MeOH,
60% over two steps. (b) Pd(OH)2 on C, HCl, MeOH, H2, 1 atm, rt,
20 h, quant.
OL026711E
(16) Bernotas, R. C.; Papendreou, G.; Urbach, J.; Ganem, B. Tetrahedron
Lett. 1990, 31, 3393.
(17) (a) Tschamber, T.; Backenstrass, F.; Neuburger, M.; Zehnder, M.;
Streith, J. Tetrahedron, 1994, 50, 1135. (b) Legler, G.; Stu¨tz, A. E.; Immich,
H. Carbohydr. Res. 1995, 272, 17. (c) Godskesen, M.; Lundt, I.; Madsen,
R.; Winchester, B. Bioorg. Med. Chem. 1996, 4, 1857. (d) Sames, D.; Polt,
R. Synlett 1995, 552. (e) Amat, M.; Llor, N.; Huguet, M.; Molins, E.;
Espinosa, E.; Bosch, J. Org. Lett. 2001, 3, 3257. (f) Patil, N. T.; John, S.;
Sabharwal, S. G.; Dhawale, D. D. Bioorg. Med. Chem. 2002, 10, 2155.
and acetonide groups, followed by chromatographic purifica-
tion yielded 18, which was characterized as its hydrochloride
salt. Spectral data for 18 were in excellent agreement with
those reported.16
3886
Org. Lett., Vol. 4, No. 22, 2002