Total synthesis and determination of the stereochemistry of 2-amino-3-cyclopropylbutanoic acid, a novel plant growth regulator isolated from the mushroom Amanita castanopsidis Hongo

Article abstract of DOI:10.1039/b108752e

The unknown stereostructure of 2-amino-3-cyclopropylbutanoic acid 1, a novel plant growth regulator isolated from the mushroom Amanita castanopsidis Hongo, was determined to be (2S,3S)-2 through its racemic and enantioselective syntheses employing the che

Full text of DOI:10.1039/b108752e

Total synthesis and determination of the stereochemistry of
2-amino-3-cyclopropylbutanoic acid, a novel plant growth regulator
isolated from the mushroom Amanita castanopsidis Hongo
Yoshiki Morimoto,*a Mamoru Takaishi,a Takamasa Kinoshita,a Kazuhiko Sakaguchib and
Kozo Shibata†a
a Department of Chemistry, Graduate School of Science, Osaka City University, Sumiyoshi-ku, Osaka
558–8585, Japan. E-mail: morimoto@sci.osaka-cu.ac.jp
b Department of Material Science, Graduate School of Science, Osaka City University, Sumiyoshi-ku, Osaka
558–8585, Japan
Received (in Cambridge, UK) 27th September 2001, Accepted 14th November 2001
First published as an Advance Article on the web 14th December 2001
The unknown stereostructure of 2-amino-3-cyclopropylbu-
tanoic acid 1, a novel plant growth regulator isolated from
the mushroom Amanita castanopsidis Hongo, was deter-
mined to be (2S,3S)-2 through its racemic and enantiose-
lective syntheses employing the chelate–enolate Claisen
rearrangement as a key step.
of (Z)-3b, a steric repulsion between the axial substituent RZ and
the rigid chelate-bridged zinc enolate moiety in the transition
state A (R1 = RE = H, R2 = Z, RZ = Me) leading to the anti
4b is more serious than in that of (E)-3a. Therefore, since the
2-Amino-3-cyclopropylbutanoic acid 1 was isolated as a novel
a-amino acid from the mushroom Amanita castanopsidis
Hongo (Koshiroonitake in Japanese) by Yoshimura et al. in
1999, and the plane structure was elucidated by spectroscopic
analyses.1 The relative and absolute configurations at the two
chiral centres C2 and C3 have not, however, been determined.
It is reported that 1 substantially inhibits root elongation in
lettuce seedlings, and the biological activity is like indole-
3-acetic acid known as an important plant growth regulator.
There have also been many cyclopropane ring-containing a-
amino acids closely related to 1, such as coronatine,2 cyclopro-
pylalanine,3 methylenecyclopropylglycine and hypoglycin A4
and so forth, possessing a broad spectrum of biological
activities. Thus, it appeared desirable to clarify the stereos-
tructure of 1 for exploring structure–activity relationships
between the cyclopropane ring-containing a-amino acids
mentioned above and 1. In this paper, we report that the
stereochemistry of 1 is 2S,3S through the stereoselective
syntheses of two possible syn and anti diastereomers 7a and 7b,
respectively, in racemic form and the optically active anti 2.
Scheme 1 Reagents and conditions: i, LDA, ZnCl2, THF, 278 °C to RT, 30
min, 47%; ii, DCC, DMAP, BnOH, Et2O, RT, 20 h; iii, CH2N2, Pd(OAc)2,
Et2O, 0 °C, 1 h; iv, H2, 10% Pd–C, MeOH, RT, 15 min.
First, we decided to synthesize two possible diastereomers,
syn 7a and anti 7b, to elucidate the unknown relative
configuration. In 1994, it has been reported by Kazmaier that
syn amino acid 4a can diastereoselectively be obtained in a ratio
of syn+anti = 95+5 from (E)-ester 3a via the [3,3]-sigmatropic
rearrangement of a chelate-bridged zinc enolate (Scheme 1).5
The high diastereoselectivity was explained in terms of a
chairlike transition state A (R1 = RZ = H, R2 = Z, RE = Me)
like that generally postulated for [3,3]-sigmatropic rearrange-
ments of acyclic systems (Scheme 2).6 Therefore, another anti
amino acid 4b could be constructed by applying Kazmaier’s
method to (Z)-ester 3b.
In practice, the [3,3]-sigmatropic rearrangement of (Z)-ester
3b‡ under Kazmaier’s conditions predominantly afforded the
anti amino acid 4b7 in a modest yield (anti+syn = ca. 80+20).
The somewhat lowered diastereoselectivity in the rearrange-
ment of (Z)-3b may be due to the following reason. In the case
Scheme 2 Plausible transition states for the [3,3]-sigmatropic rearrange-
ment of chelate-bridged zinc enolates.
† Deceased on June 18, 1998.
42
CHEM. COMMUN., 2002, 42–43
This journal is © The Royal Society of Chemistry 2002

energy difference between A and the boatlike transition state B
leading to the syn 4a becomes smaller than that in (E)-3a, the
lowering of the diastereoselectivity might be observed. Any-
way, both the diastereomers 4a and 4b were converted into the
desired syn and anti 2-amino-3-cyclopropylbutanoic acids 7a
and 7b, respectively, by a sequence of reactions: (i) protection
of the carboxyl group as a benzyl ester; (ii) Pd(OAc)2-catalyzed
cyclopropanation of the terminal double bond with diazo-
methane;8 and (iii) hydrogenolysis of the benzylic protective
groups. Comparing the 1H and 13C NMR of synthetic 7a and 7b
with those of the natural product 1,1 it has been found that 1 is
identical to 7b possessing the anti relative stereochemistry.
Next, we embarked on the enantioselective synthesis of
optically active anti 2 bearing 2S stereochemistry, because
Yoshimura et al. have deduced that the absolute configuration at
the C2 position of 1 is S.1 Recently, Sakaguchi et al., one of the
authors in this paper, have developed silyl-assisted [3,3]-sigma-
tropic rearrangements of (1-acyloxy-2-alkenyl)trialkylsilanes to
provide optically active vinylsilane-containing a-amino acids
in a complete chirality-transferring manner.9 In conjunction
with the key step for the syntheses of racemates 7a and 7b, we
adopted the method to synthesize the optically active 2.
Concurrent removal of the TBDMS and Boc groups in the
11
rearrangement product 14 with 42% HBF4 furnished free
amino acid 15, N-Z protection of which yielded 16 consistent
with ( )-4b. Finally, according to the racemic route, the
carboxylic acid 16 was converted into the amino acid (2S,3S)-2
in almost quantitative yield. The optical rotation of synthetic 2,
[a]3D1 29.08 (c 0.5, H2O), was identical with that reexamined for
the natural amino acid 1 generously gifted by Professor
Wakabayashi (Osaka City University), [a]2D6 29.46 (c 0.035,
H2O). Thus, it has been found that the hitherto unknown
absolute configuration of 1 is assigned to 2S,3S.
In conclusion, we have accomplished the stereoselective
syntheses of two possible syn and anti diastereomers 7a and 7b,
respectively, in racemic form and the optically active anti 2
using the chelate–enolate Claisen rearrangement as a key step,
and determined the relative and absolute configurations of
2-amino-3-cyclopropylbutanoic acid 1, a novel plant growth
regulator. The elucidation of the stereochemistry of 1 will be
useful for structure–activity relationships and conformational
analysis.
We thank Professor K. Wakabayashi (Osaka City University)
for generously supplying natural amino acid 1, and are also
grateful to Ms H. Yoshimura and Mr H. Suzuki (Osaka City
University) for helpful discussions.
The preparation of the rearrangement precursor (Z)-13 (J =
11.0 Hz between the olefinic protons) was carried out by
esterfication of the readily available chiral (S)-alcohol 11,10
[a]2D3 290.2 (c 1.01, CHCl3); 98% ee, with commercially
available Boc-glycine followed by partial reduction of the
resulting alkyne 12 with Lindlar catalyst (Scheme 3). The
[3,3]-sigmatropic rearrangement of 13 diastereo- and enantiose-
lectively proceeded under Kazmaier’s conditions to give only
the (E)-anti product 14 (J = 18.5 Hz between the olefinic
protons) in 97% yield with complete chirality transfer.§ In spite
of the same (Z)-geometry as 3b, the high diastereo- and
enantioselectivity observed in this rearrangement could be
attributed to the bulky tert-butyldimethylsilyl group. In the
Notes and references
‡ The (Z)-ester 3b (J = 10.8 Hz between the olefinic protons) was prepared
by condensation of commercially available Z-glycine with but-2-yn-1-ol
(DCC, DMAP, Et2O, RT, 22 h, 97%) and subsequent Lindlar reduction of
the resulting alkyne (H2, 5% Pd–CaCO3, Py, EtOAc, RT, 40 min, 89%). All
new compounds were satisfactorily characterized using 1H and 13C NMR,
IR, MS and HRMS spectra and also by elemental analyses whenever
possible. Selected data for 7a: mp 262–264 °C; dH (400 MHz, D2O) 3.74
(1H, d, J = 4.1 Hz), 1.34 (1H, ddq, J = 9.8, 4.1, 7.0 Hz), 0.98 (3H, d, J =
7.1 Hz), 0.68–0.44 (3H, m), 0.25–0.14 (2H, m); dC (100 MHz, D2O) 174.6,
60.1, 39.8, 14.12, 14.10, 4.3, 4.0; nmax (KBr)/cm21 3600–2200, 3410, 1672;
m/z (FAB-MS) 144 [(M + H)+, 100%] (FAB-HRMS: calc. for C7H14O2N
[(M + H)+], 144.1025; found, 144.1032) (Calc. for C7H13O2N: C, 58.72; H,
9.15; N, 9.78. Found: C, 58.54; H, 9.30; N, 9.64%). 2: mp 265–268 °C; dH
(400 MHz, D2O) 3.64 (1H, d, J = 4.6 Hz), 1.30 (1H, ddq, J = 9.8, 4.6, 7.1
Hz), 1.06 (3H, d, J = 7.1 Hz), 0.65 (1H, m), 0.47 (2H, m), 0.22 (1H, m),
0.09 (1H, m); dC (100 MHz, D2O) 174.3, 60.2, 39.6, 16.1, 12.9, 4.0, 2.7;
nmax (KBr)/cm21 3600–2200, 3342, 1628; m/z (FAB-MS) 144 [(M + H)+,
100%] (FAB-HRMS: calc. for C7H14O2N [(M + H)+], 144.1025; found,
144.1010) (Calc. for C7H13O2N: C, 58.72; H, 9.15; N, 9.78. Found: C,
58.25; H, 9.13; N, 9.55%).
transition state B (R1 = TBDMS, R2 = Boc, RZ = Me, RE
=
H) leading to syn 8 and C leading to 10 enantiomeric to 9 except
for the alkene geometry, sterically encumbered A1,3-strain and
1,3-diaxial-like interaction, respectively, occur between the
bulky R1 and RZ. Therefore, because the transition state A
without such a repulsive interaction becomes relatively more
stable than B and C, 14 might exclusively be formed.
§ The absolute configuration of (2S,3S)-14, [a]2D4 +9.94 (c 1.00, CHCl3), has
been confirmed by leading the enantiomer of 14, [a]1D9 –8.6 (c 1.05, CHCl3),
to (2R,3R)- -isoleucine (ref. 9). The optical purity of 14 was determined to
D
be > 99% ee by derivatization of 15 (R1 = R2 = R3 = H) to 15A [R1 = Me,
R2 = (R)-a-methoxy-a-(trifluoromethyl)phenylacetyl, R3 = H] and in-
tegration of the signals in the1 H NMR spectrum.
1 H. Yoshimura, K. Takegami, M. Doe, T. Yamashita, K. Shibata, K.
Wakabayashi, K. Soga and S. Kamisaka, Phytochemistry, 1999, 52,
25.
2 A. Ichihara and H. Toshima, in Biologically Active Natural Products:
Agrochemicals, ed. H. Cutler and S. Cutler, CRC Press, Washington,
DC, 1999, p. 93.
3 T. Ohta, S. Nakajima, Z. Sato, T. Aoki, S.-I. Hatanaka and S. Nozoe,
Chem. Lett., 1986, 511.
4 D. Li, G. Agnihotri, S. Dakoji, E. Oh, M. Lantz and H.-W. Liu, J. Am.
Chem. Soc., 1999, 121, 9034 and references cited therein.
5 U. Kazmaier, Angew. Chem., Int. Ed. Engl., 1994, 33, 998.
6 P. Wipf, in Comprehensive Organic Synthesis, ed. B. M. Trost and I.
Fleming, Pergamon, Oxford, 1991, vol. 5, p. 827 and references cited
therein; U. Kazmaier, Liebigs Ann./Recueil, 1997, 285.
7 P. A. Bartlett and J. F. Barstow, J. Org. Chem., 1982, 47, 3933.
8 M. Suda, Synthesis, 1981, 714.
Scheme 3 Reagents and conditions: i, Boc-glycine, 1-(3-dimethylamino-
propyl)-3-ethylcarbodiimide hydrochloride (EDCI), DMAP, CH2Cl2, RT, 2
h, 94%; ii, H2, 5% Pd–CaCO3, Py, EtOAc, RT, 1 h, 81%; iii, LDA, ZnCl2,
THF, 278 °C to RT, 2 h, 97%; iv, 42% aq. HBF4, 1,4-dioxane, 65 °C, 14
h, 100%; v, ZCl, K2CO3, 1,4-dioxane–H2O (1+1), 0 °C to RT, 12 h, 73%;
vi, BnOH, EDCI, DMAP, CH2Cl2, RT, 4 h, 94%; vii, iii in Scheme 1, 100%;
viii, iv in Scheme 1, 100%.
9 K. Sakaguchi, H. Suzuki and Y. Ohfune, Chirality, 2001, 13, 357.
10 K. Sakaguchi, M. Fujita, H. Suzuki, M. Higashino and Y. Ohfune,
Tetrahedron Lett., 2000, 41, 6589.
11 R. E. Ireland and M. D. Varney, J. Am. Chem. Soc., 1984, 106, 3668.
CHEM. COMMUN., 2002, 42–43
43