presence of an enzyme that forms these compounds suggests
they are probably Arabidopsis metabolites. Lupeol and
â-amyrin have been shown to be enzymatic oxidosqualene
cyclase products in crude homogenates.10 â-Amyrin syn-
thases11 from Panax ginseng and lupeol synthases12 from
Olea europaea and Taraxacum officinale are known. These
enzymes form single products and are phylogenetically
distinct from LUP1.12 The present study provides the first
demonstration of direct enzymatic conversion of oxido-
squalene to germanicol, taraxasterol, ψ-taraxasterol, and 3,20-
dihydroxylupane.
Scheme 1. LUP1 Converts Oxidosqualene to at Least Five
Different C30H50O Triterpene Alcohols and a C30H52O2 Diol8
All six compounds can be derived from the lupeyl cation
(Scheme 1). Deprotonation from C-29 provides lupeol, and
quenching with water forms 3,20-dihydroxylupane. Expand-
ing the E ring provides the germanicyl cation, which can
deprotonate directly from C-18 to form germanicol or
undergo hydride shifts and deprotonation at C-12 to form
â-amyrin. A methyl shift to the taraxasteryl cation followed
by deprotonation from C-30 or C-21 yields taraxasterol and
ψ-taraxasterol, respectively. Such broad product diversity is
unprecedented in oxidosqualene cyclases. Although traces
of parkeol (∼1%) have been isolated from some incubations
with cycloartenol synthases,13 no byproducts have been
reported from lanosterol synthases. â-Amyrin and lupeol
synthases11,12 are known that generate single products. Some
biological role for LUP1 is implied by Arabidopsis maintain-
ing the LUP1 gene without accumulating mutations that
abolish encoded enzymatic activity. If generating lupeol were
the primary role of LUP1, it should evolve to attain greater
accuracy as other lupeol synthases have done. LUP1 product
heterogeneity could be useful for several biological functions.
Pure compounds form crystals more readily than mixtures
do, and LUP1 may provide triterpenes for a role (such as
epicuticular wax) in which an amorphous solid is physi-
ologically preferable. Alternatively, product heterogeneity
could generate a diverse array of defense compounds
relatively efficiently.
(1H, 13C, DEPT, COSYDEC, HSQC, and HMBC) confirmed
the structures, and some assignments were revised.7 The polar
fraction was shown to be composed of 3,20-dihydroxylupane
(6)9 by similar NMR experiments. Although none of these
compounds have been described from Arabidopsis, the
(7) Lupeol and â-amyrin were characterized in the previous study.5
Germanicol: selected 1H NMR signals (500 MHz, CDCl3, 25 °C) δ 0.733
(3H, s, H-27), 0.844 (3H, s, H-24), 0.849 (3H, s, H-23), 0.904 (3H, s, H-25),
0.937 (3H, s, H-29), 0.944 (3H, s, H-30), 1.017 (3H, s, H-28), 1.078 (3H,
s, H-26), 2.046 (3H, s, acetate), 4.485 (1H, m, H-3R), 4.862 (1H, m, H-19);
13C NMR (125 MHz, CDCl3, 25 °C) δ 14.55 (C-27), 16.08 (C-26), 16.52
(C-24), 16.76 (C-25), 18.14 (C-6), 21.12 (C-11), 21.32 (acetate) 23.69 (C-
2), 25.25 (C-28), 26.18 (C-12), 27.51 (C-15), 27.91 (C-23), 29.18 (C-30),
31.36 (C-29), 32.35 (C-20), 33.33 (C-21), 34.34 (C-17), 34.52 (C-7), 37.14
(C-10), 37.36 (C-22), 37.69 (C-16), 37.82 (C-4), 38.39 (C-13), 38.62 (C-
1), 40.76 (C-8), 43.32 (C-14), 51.13 (C-9), 55.58 (C-5), 80.96 (C-3), 129.77
(C-19), 142.67 (C-18), 171.02 (acetate). ψ-Taraxasterol: selected 1H NMR
signals (500 MHz, CDCl3, 25 °C) δ 0.733 (3H, s, H-28), 0.844 (3H, s,
H-24), 0.854 (3H, s, H-23), 0.879 (3H, s, H-25), 0.948 (3H, s, H-27), 0.989
(3H, d, H-29), 1.045 (3H, s, H-26), 1.635 (3H, s, H-30), 2.046 (3H, s,
acetate) 4.485 (1H, m, H-3R), 5.261 (1H, d, H-21); 13C NMR (125 MHz,
CDCl3, 25 °C) δ 14.69 (C-27), 16.04 (C-26), 16.35 (C-25), 16.52 (C-24),
17.70 (C-28), 18.18 (C-6), 21.32 (acetate), 21.61 (C-11), 21.62 (C-30), 22.53
(C-29), 23.69 (C-2), 27.02 (C-15), 27.59 (C-12), 27.94 (C-23), 34.16 (C-
7), 34.38 (C-17), 36.32 (C-19), 36.69 (C-16), 37.01 (C-10), 37.79 (C-4),
38.44 (C-1), 39.20 (C-13), 41.08 (C-8), 42.17 (C-22), 42.33 (C-14), 48.68
(C-18), 50.33 (C-9), 55.38 (C-5), 80.99 (C-3), 118.88 (C-21), 139.86 (C-
20), 171.02 (acetate). Taraxasterol: selected 1H NMR signals (500 MHz,
CDCl3, 25 °C) δ 0.843 (3H, s, H-24), 0.850 (3H, s, H-23), 0.855 (3H, s,
H-28), 0.877 (3H, s, H-25), 0.927 (3H, s, H-27), 1.021 (3H, s, H-26), 1.021
(3H, d, H-29), 2.046 (3H, s, acetate), 4.485 (1H, m, H-3R), 4.609 (2H, m,
H-30); 13C NMR (125 MHz, CDCl3, 25 °C) δ 14.72 (C-27), 15.88 (C-26),
16.33 (C-25), 16.49 (C-24), 18.18 (C-6), 19.47 (C-28), 21.32 (acetate), 21.46
(C-11), 23.69 (C-2), 25.49 (C-29), 25.62 (C-21), 26.15 (C-12), 26.64 (C-
15), 27.94 (C-23), 33.99 (C-7), 34.53 (C-17), 37.05 (C-10), 37.80 (C-4),
38.29 (C-16), 38.44 (C-1), 38.86 (C-22), 39.16 (C-13), 39.38 (C-19), 40.92
(C-8), 42.04 (C-14), 48.65 (C-18), 50.40 (C-9), 55.45 (C-5), 80.98 (C-3),
107.11 (C-30), 154.67 (C-20), 171.02 (acetate).
Although enzymes that convert a single substrate to
multiple products are unusual, examples are known in terpene
(9) 3â, 20-Dihydroxylupane: selected 1H NMR signals (500 MHz,
CDCl3, 25 °C) δ 0.764 (3H, s, H-24), 0.810 (3H, s, H-28), 0.841 (3H, s,
H-25), 0.957 (3H, s, H-27), 0.973 (3H, s, H-23), 1.059 (3H, s, H-26), 1.122
(3H, s, H-29/30), 1.225 (3H, s, H-29/30), 3.198 (1H, m, H-3R); 13C NMR
(125 MHz, CDCl3, 25 °C) δ 14.83 (C-27), 15.38 (C-24), 16.15 (C-25),
16.15 (C-26), 18.33 (C-6), 19.21 (C-28), 21.38 (C-11), 24.76 (C-29/30),
27.38 (C-2), 27.56 (C-15), 27.98 (C-23), 28.74 (C-21), 29.05 (C-12), 31.54
(C-29/30), 34.54 (C-7), 35.55 (C-16), 37.07 (C-10), 37.44 (C-13), 38.68
(C-1), 38.84 (C-4), 40.20 (C-22), 41.34 (C-8), 43.51 (C-14), 44.64 (C-17),
48.29 (C-18), 49.93 (C-19), 50.27 (C-9), 55.17 (C-5), 73.50 (C-20), 79.00
(C-3).
(10) (a) Corey, E. J.; Ortiz de Montellano, P. R. J. Am. Chem. Soc. 1967,
89, 3362-3363. (b) Barton, D. H. R.; Jarman, T. R.; Watson, K. C.;
Widdowson, D. A.; Boar, R. B.; Damps, K. J. Chem. Soc., Perkin Trans.
1 1975, 1134-1138.
(11) (a) Kushiro, T.; Shibuya, M.; Ebizuka, Y. Eur. J. Biochem. 1998,
256, 238-244. (b) Kushiro, T.; Shibuya, M.; Ebizuka, Y. In Towards
Natural Medicine Research in the 21st Century, Excerpta Medica Inter-
national Congress Series 1157; Ageta, H., Aimi, N., Ebizuka, Y., Fujita,
T., Honda, G., Eds.; Elsevier Science: Amsterdam, 1998; pp 421-428.
(12) Shibuya, M.; Zhang, H.; Endo, A.; Shishikura, K.; Kushiro, T.;
Ebizuka, Y. Eur. J. Biochem. 1999, 266, 302-307.
(8) Relative amounts of C30H50O triterpenes were determined by GC-
FID quantitation after acetylation (<1% error). The diol content was
determined by integrating characteristic signals for lupeol and 3,20-
(13) (a) Hart, E. A.; Hua, L.; Darr, L. B.; Wilson, W. K.; Pang, J.;
Matsuda, S. P. T. J. Am. Chem. Soc. 1999, 121, 9887-9888. (b) Godzina,
S. M.; Lovato, M. A.; Meyer, M. M.; Foster, K. A.; Wilson, W. K.; Gu,
W.; de Hostos, E. L.; Matsuda, S. P. T. Lipids 2000, 36, 249-255.
1
dihydroxylupane in the crude product H NMR spectrum (∼5% error).
2258
Org. Lett., Vol. 2, No. 15, 2000