5.00 (m, 2H), 4.92 (d, 1H, J 11.1), 4.84 (d, 1H, J 11.1), 4.72 (d, 1H, J 11.1),
4.56 (d, 1H, J 12.8), 4.53 (d, 1H, J 12.4), 4.29 (d, 1H, J 11.9), 4.18 (bs, 1H),
3.90 (d, 1H, J 13.2), 3.83 (dd, 1H, J 7.8, 3.2), 3.67 (d, 1H, J 13.2), 3.56 (bs,
1H), 2.66 (bs, 1H), 2.45 (m, 1H), 2.29 (m, 1H); dC (CDCl3) 140.5, 139.1,
139.0, 138.7, 135.8, 135.1, 128.2–126.6 (20C), 118.9, 116.7, 82.6, 81.5,
80.1, 74.7, 74.5, 69.7, 56.7, 50.7, 34.5.
§ Spectral data for 15: dH (CDCl3) 7.43–7.25 (m, 20H), 6.00 (ddd, 1H, J
17.7, 10.4, 8.1), 5.79 (m, 1H), 5.39 (bd, 1H, J 17.5), 5.34 (bd, 1H, J 10.4),
5.10 (bd, 1H, J 17.9), 5.09 (bd, 1H, J 9.8), 4.83 (d, 1H, J 11.5), 4.75 (d, 2H,
J 11.5), 4.71 (d, 1H, J 11.9), 4.58 (d, 1H, J 11.1), 4.44 (d, 1H, J 11.9), 4.29
(m, 1H), 3.89 (m, 1H), 3.83 (d, 1H, J 12.8), 3.82 (m, 1H), 3.77 (d, 1H, J
12.8), 3.08 (dt, 1H, J 8.0, 3.9), 2.51 (m, 1H), 2.36 (m, 1H); dC (CDCl3)
140.6, 138.8, 138.6, 138.5, 136.4, 136.2, 128.8–125.9 (20C), 118.1, 117.0,
82.3, 81.1, 79.4, 74.4, 73.1, 70.4, 57.1, 51.7, 34.6.
a one pot procedure to give a 3+1 mixture of ketones. Major
ketone isomer 8 was isolated in 61% yield and hydrogenated to
give calystegine B2 ([a]20 +23.8 (c 0.31, H2O)). NMR data and
optical rotation were in Daccordance with those reported for the
natural product.13
The same sequence was then applied to the synthesis of
calystegine B3 and B4 (Scheme 2). Subjecting galactose-derived
iodoglycoside 9 to the domino reaction gave a 2+1 mixture of
amino dienes. Major diastereomer 10‡ was isolated in 59%
yield and then converted into cycloheptene 12. Hydroboration
and oxidation gave a 3+1 mixture of ketones and the major
isomer 13 was isolated in 64% yield. Hydrogenation then
furnished calystegine B3 ([a]20 +75.6 (c 0.55, H2O)) with NMR
data and optical rotation in Daccordance with the data for the
natural compound.13 Subjecting mannose-derived iodoglyco-
side 14 to the domino reaction gave a 8+1 mixture of amino
dienes and the major diastereomer 15§ was isolated in 71%
yield. The stereochemical outcome in these allylations is
noteworthy. In all three cases the major product is the (R)-
benzylamine which is the correct stereochemistry for the
calystegines. The major product 15 from mannose is consistent
with predictions from the Felkin-Anh model while the major
isomers from glucose and galactose are not. This, however, does
correspond with our previous experience on zinc-mediated
alkylations of glucose and mannose substrates.4a Finally, amine
15 was converted into cycloheptene 17 which was hydroborated
and oxidised to give a 3+1 ratio of ketones. The major isomer 18
was isolated in 63% yield and deprotected to give calystegine
B4 ([a]2D0 246.4 (c 0.18, H2O)). NMR data and optical rotation
were similar to those reported for the natural product.7
In conclusion, a general strategy for preparation of the
calystegines has been devised. Calystegine B3 and B4 have been
prepared for the first time and their absolute configuration
confirmed. These syntheses should hold great promise for
making the calystegines and their analogues more readily
available for biological investigations.
1 Iminosugars as Glycosidase Inhibitors, ed. A. E. Stütz, Wiley-VCH,
Weinheim, 1999.
2 N. Asano, R. J. Nash, R. J. Molyneux and G. W. J. Fleet, Tetrahedron:
Asymmetry, 2000, 11, 1645; A. A. Watson, D. R. Davies, N. Asano, B.
Winchester, A. Kato, R. J. Molyneux, B. L. Stegelmeier and R. J. Nash,
in ACS Symposium Series, vol. 745, ed. A. T. Tu and W. Gaffield, ACS,
Washington, DC, 2000, p. 129.
3 (a) F.-D. Boyer and I. Hanna, Tetrahedron Lett., 2001, 42, 1275; (b) T.
Faitg, J. Soulié, J.-Y. Lallemand and L. Ricard, Tetrahedron: Asym-
metry, 1999, 10, 2165; (c) C. R. Johnson and S. J. Bis, J. Org. Chem.,
1995, 60, 615; (d) F.-D. Boyer and J.-Y. Lallemand, Tetrahedron, 1994,
50, 10443; (e) O. Duclos, M. Mondange, A. Duréault and J. C. Depezay,
Tetrahedron Lett., 1992, 33, 8061.
4 (a) L. Hyldtoft and R. Madsen, J. Am. Chem. Soc., 2000, 122, 8444; (b)
L. Hyldtoft, C. S. Poulsen and R. Madsen, Chem. Commun., 1999,
2101.
5 For application to seven- and eight-membered rings, see: I. Hanna and
L. Ricard, Org. Lett., 2000, 2, 2651.
6 Recently, the method in ref. 4 was used for a very similar synthesis of
1 (see ref. 3a) which has prompted us to publish our results now.
7 N. Asano, A. Kato, H. Kiza, K. Matsui, A. A. Watson and R. J. Nash,
Carbohydr. Res., 1996, 293, 195.
8 B. Bernet and A. Vasella, Helv. Chim. Acta, 1979, 62, 1990.
9 M. Scholl, S. Ding, C. W. Lee and R. H. Grubbs, Org. Lett., 1999, 1,
953.
We thank the Danish Natural Science Research Council for
financial support.
10 For a recent review on olefin metathesis in carbohydrate chemistry, see:
M. Jørgensen, P. Hadwiger, R. Madsen, A. E. Stütz and T. M.
Wrodnigg, Curr. Org. Chem., 2000, 4, 565.
11 W. Wang, Y. Zhang, M. Sollogoub and P. Sinaÿ, Angew. Chem., Int.
Ed., 2000, 39, 2466.
Notes and references
† All Cbz-protected compounds showed mixtures of rotamers by NMR.
‡ Spectral data for 10: dH (CDCl3) 7.35–7.22 (m, 20H), 5.98 (ddd, 1H, J
17.9, 10.0, 8.5), 5.65 (m, 1H), 5.31 (bd, 1H, J 10.0), 5.15 (bd, 1H, J 17.5),
12 D. B. Dess and J. C. Martin, J. Am. Chem. Soc., 1991, 113, 7277.
13 N. Asano, A. Kato, K. Oseki, H. Kizu and K. Matsui, Eur. J. Biochem.,
1995, 229, 369.
Chem. Commun., 2001, 1106–1107
1107