methanol leading to 6a exhibited poor reproducibility
(16–68%); therefore, the indolyl NH group of 5a was protected
to afford 5b in 94% yield. In contrast to 5a, hydrogenolysis of
5b proceeded cleanly to afford lactone 6b in 81% yield after
protection of the primary amino group. Oxidation of the primary
hydroxyl group of 6b with PDC gave carboxylic acid 7 in 69%
yield. Finally, deprotection of two Boc groups followed by
alkaline hydrolysis gave (2)-monatin (1) in 92% yield.‡
In conclusion, synthesis of (2)-monatin (1) featuring a
highly stereoselective cycloaddition of cyclic nitrone 2 with
allyl alcohol 3a in the presence of MgBr2·OEt2 has been
accomplished in 47% overall yield from nitrone 2. Thus,
(2)-monatin (1) can now be obtained easily with high
stereoselectivity and in sufficient amounts for further studies of
1 as a sweetener. In addition, the strategy used here would be
effective for synthesis of other g-substituted g-oxyglutamic
acids such as lycoperdic acid7 and dysiherbaine.8
Fig. 1
This work was supported by a Grant-in-Aid for Scientific
Research from the Ministry of Education, Culture, Sports,
Science and Technology.
Notes and references
‡ Synthetic (2)-monatin (1): [a]25D 210.95 (c 1.0, 1 N HCl), [lit.1 [a]20
D
27.6 (c 1.0, 1 N HCl)]; dH (500 MHz, D2O): 2.04 (1H, dd, J = 15.1, 11.7
Hz), 2.66 (1H, br d, J = 15.1 Hz), 3.12 (1H, d, J = 14.6 Hz), 3.32 (1H, d,
J = 14.6 Hz), 3.62 (1H, br d, J = 11.2 Hz), 7.19 (1H, br t, J = 7.3 Hz), 7.26
(1H, br d, J = 7.3 Hz), 7.28 (1H, s), 7.53 (1H, d, J = 8.3 Hz), 7.77 (1H, d,
J = 8.3 Hz); dC (125 MHz, D2O): 38.1, 41.7, 56.6, 83.1, 111.9, 114.4,
121.8, 122.0, 124.3, 127.6, 130.7, 138.6, 179.0, 181.8. These spectral data
of synthetic 1 are identical with those of natural 1 reported in ref. 1.
1 R. Vleggaar, L. G. J. Ackerman and P. S. Steyn, J. Chem. Soc., Perkin
Trans. 1, 1992, 3095.
2 For synthesis of optically active 1, see: (a) K. Nakamura, T. J. Baker and
M. Goodman, Org. Lett., 2000, 2, 2967; (b) T. Kitahara and H. Watanabe,
Japan Kokai Tokkyo Koho, 2002, 60382A.
3 For synthetic study of optically active 1, see: D. J. Oliveira and F. Coelho,
Tetrahedron Lett., 2001, 42, 6793.
4 For syntheses of racemic 1, see: (a) C. W. Holzapfel, K. Bischofberger
and J. Olivier, Synth. Commun., 1994, 24, 3197; (b) E. Abushanab and S.
Arumugan, US Patent, 1999, 5,994,559.
5 (a) O. Tamura, K. Gotanda, R. Terashima, M. Kikuchi, T. Miyawaki and
M. Sakamoto, Chem. Commun., 1996, 1861; (b) O. Tamura, K. Gotanda,
J. Yoshino, Y. Morita, R. Terashima, M. Kikuchi, T. Miyawaki, N. Mita,
M. Yamashita, H. Ishibashi and M. Sakamoto, J. Org. Chem., 2000, 65,
8544.
Scheme 2 Reagents and conditions: i, TBSCl, imidazole, DMF, 97%; ii,
Boc2O, DMAP, CH3CN, 97%; iii, HF·pyridine, THF, 100%; iv, H2,
Pd(OH)2/C, MeOH; v, Boc2O, CH3CN, 16–68% for 6a from 5a, 81% for 6b
from 5b; vi, PDC, DMF, 69% from 6b; vii, HCl, HCO2H; viii, NaOH,
MeOH then Amberlite® IR-120-H+ form, aq. NH3, 92%.
6 For MgBr2-promoted cycloaddition of nitrones with allyl alcohols, see:
(a) S. Kanemasa, T. Tsuruoka and E. Wada, Tetrahedron Lett., 1993, 34,
87; (b) S. Kanemasa and T. Tsuruoka, Chem. Lett., 1995, 49; (c) O.
Tamura, T. Kuroki, Y. Sakai, J. Takizawa, J. Yoshino, Y. Morita, N.
Mita, K. Gotanda and M. Sakamoto, Tetrahedron Lett., 1999, 40, 895.
For a review on chiral zinc complex mediated cycloaddition of nitrones
with allyl alcohols, see (d) Y. Ukaji and K. Inomata, Synlett, 2003, 1075.
For chelation-promoted cycloaddition of nitrile oxides with allyl
alcohols, see (e) S. Kanemasa, M. Nishiuchi, A. Kamimura and K. Hori,
J. Am. Chem. Soc., 1994, 116, 2324; (f) J. W. Bode, N. Fraefel, D. Muri
and E. M. Carreira, Angew. Chem., Int. Ed., 2001, 40, 2082; (g) J. W.
Bode and E. M. Carreira, J. Am. Chem. Soc., 2001, 123, 3611.
7 (a) J. Lamotte, B. Oleksyn, L. Dupont, O. Dideberg, H. Campsteyn, M.
Vermeire and N. R-Banga, Acta Crystallogr., 1978, B34, 3635; (b) N. R-
Banga, A. Welter, J. Jadot and J. Casimir, Phytochemistry, 1979, 18,
482.
alcohol 3 from the si-face (see formula III).5 Taking into
account the fact that MgBr2·OEt2 accelerates the cycloaddition,
chelated transition state models IV and V may be involved in
the reaction of 2 and 3a.6 Because model IV has a severe steric
interaction between MgBr2 and the phenyl group, it is
reasonable to assume that the cycloaddition proceeds via model
V to give cycloadduct 5a with high stereoselectivity. In the
absence of MgBr2·OEt2, the cycloaddition may proceed via III
(R = CH2OH, RA = 3-indolylmethyl group), in which the bulky
3-indolylmethyl group occupies an exo-position, to afford 5aA as
a major product.
Cycloadduct 5a was then elaborated to (2)-monatin (1)
8 R. Sakai, H. Kamiya, M. Murata and K. Shimamoto, J. Am. Chem. Soc.,
1997, 119, 4112.
(Scheme 2). Hydrogenolysis of 5a with Pearlman’s catalyst in
CHEM. COMMUN., 2003, 2678–2679
2679