Potassium 2,3,4-trihydroxy-2-methylbutanoate, a leaf-closing substance of Leucaena leucocephalam

Article abstract of DOI:10.1016/S0040-4039(01)00380-X

Potassium 2,3,4-trihydroxy-2-methylbutanoate (1) was identified as a leaf-closing substance of a nyctinastic plant, Leucaena leucocephalam. Compound 1 was quite effective for the leaf-closing of L. leucocephalam at 1×10^-6 M.

Full text of DOI:10.1016/S0040-4039(01)00380-X

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 3109–3111
Potassium 2,3,4-trihydroxy-2-methylbutanoate, a leaf-closing
substance of Leucaena leucocephalam
a,
a
b
a,
Minoru Ueda, * Yoshihiro Sohtome, Katsuhiro Ueda and Shosuke Yamamura *
a
Laboratory of Natural Products Chemistry, Department of Chemistry, Faculty of Science and Technology, Keio University,
Hiyoshi, Yokohama 223-8522, Japan
b
Department of Chemistry, Biology and Marine Science, The University of Ryukyus, Nishihara-cho, Okinawa 903-0213, Japan
Received 29 January 2001; revised 26 February 2001; accepted 2 March 2001
Abstract—Potassium 2,3,4-trihydroxy-2-methylbutanoate (1) was idetified as a leaf-closing substance of a nyctinastic plant,
−
6
Leucaena leucocephalam. Compound 1 was quite effective for the leaf-closing of L. leucocephalam at 1×10 M. © 2001 Elsevier
Science Ltd. All rights reserved.
Most legumes close their leaves in the evening, as if to
sleep, and open them in the morning. This rhythmic
disruption of an ecosystem by L. leucocephalam is a
serious problem. We have tried to develop a herbicide,
which is specifically effective for L. leucocephalam, by
using the leaf-movement factor of this plant. This new
type of highly ecological herbicide will exterminate only
L. leucocephalam and, on the other hand, have no effect
on other plants. Thus, we tried to identify the leaf-
movement factors of L. leucocephalam. In this paper,
we describe the isolation, chemical structure, and bio-
logical activity of the leaf-closing substance of L.
leucocephalam.
1
movement of the leaves is called nyctinasty, which has
been known to be controlled by an internal biological
2
clock. Recently, we have identified several bioactive
substances that regulate this leaf-movement, and
revealed that nyctinastic movement of the plants is
controlled by the interaction between leaf-closing and
-
opening substances, which are different among the
3
plants. Moreover, we have demonstrated the impor-
tance of leaf-movement for the survival of legumes. In
our previous work with artificial leaf-opening substance
the leaf of Cassia mimosoides L. which was kept open
Isolation of the leaf-closing substance was carried out
based on a bioassay using a leaf of L. leucocephalam.
The bioactive fraction closed a leaf within a few hours
after the addition of a solution containing the bioactive
substance.
4
for a week was observed to wither and die. This result
showed that the nyctinastic leaf-movement is essential
for the survival of legumes. From these findings, we
envisioned that plant-specific leaf-movement factors
could be useful as a herbicide. When we use such
leaf-movement factor as a herbicide, it would be effec-
tive only for the legume from which it was isolated, and
have no effect on other vicinal plants.
The fresh whole plant of L. leucocephalam (13.0 kg)
which was collected in Okinawa, Japan, was extracted
with MeOH (72 L) for 1 week and concentrated in
vacuo. The concentrated extract was partitioned with
n-hexane, ethyl acetate, then with n-butanol. The
bioactive aqueous layer was carefully separated by
Amberlite XAD-7 column chromatography eluted with
Nyctinasty is also observed in Leucaena leucocephalam
(
gin-nemu in Japanese), a leguminous tropical plant. L.
leucocephalam is known for its rapid growth and secre-
tion of allelochemicals to inhibit the growth of other
plants around them, which leads to a grove of that
MeOH–H O (0:100, 10:90, 30:70, 50: 50, and 100:0),
2
and the H O fraction showed weak leaf-opening activ-
2
5
plant and elimination of vicinal plants. Thus, the
ity. This H O fraction was further purified with column
2
chromatography using TSK G3000S gel with 30%
EtOH aq. In this step, we were able to separate the
fraction with leaf-opening activity and another one with
leaf-closing activity. The fraction with leaf-closing
activity was purified by HPLC using a preparative
Keywords: plants; nyctinasty; natural products; biologically active
compounds.
*
Corresponding authors. Fax: 81 45 566 1697; e-mail: ueda@
chem.keio.ac.jp; yamamura@chem.keio.ac.jp
0
040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)00380-X

3
110
M. Ueda et al. / Tetrahedron Letters 42 (2001) 3109–3111
The structural determination of 1 was carried out by
7
means of NMR and FAB MS experiments. HMQC
and HMBC experiments gave the planar structure of 1.
A strong molecular ion corresponding to 1 was
observed at m/z 189 in the positive mode FAB mass
spectrum of bioactive fraction, on the other hand, m/z
5
H C
3
H
OH
H
2
4
1
1
49 was observed as a molecular ion in negative mode
KOOC
OH
3
FAB MS experiment, which suggested that 1 exists as a
potassium salt. Daughter ions observed in a linked-scan
experiment are also shown in Fig. 1.
m/z 88
HO
H
m/z 117
The relative stereochemistry of 1 was determined by
8
using g-lactone (2), which was synthesized from a
Figure 1. HMBC correlations and daughter ions observed in
a linked-scan experiment of potassium 2,3,4-trihydroxy-2-
methylbutanoate (1).
mixture of 1 and 2 by treatment with 0.01 M HCl (Fig.
2). 2,3-Dihydroxy-2-methyl-4-butanolides of syn and
anti relative stereochemistry were synthesized by
9
1
Kobayashi et al. Thus, comparison of the H NMR
9
data of 2 with those reported by Kobayashi et al.
Develosil ODS-HG-5 column with 10% MeOH aq., and
then Develosil ODS-HG-5 with 5% MeOH aq. to give
a mixture (1.7 mg) of potassium lactate and potassium
suggested the relative stereochemistry of 2 as shown in
Fig. 2. And an NOE correlation observed in 2 confir-
med this result (Fig. 2). Thus, the relative stereochem-
istry of 1 was determined as shown in Fig. 1.
2,3,4-trihydroxy-2-methylbutanoate (1) in the ratio of
:2. We carried out further purification of 1 with HPLC
1
with plural ODS columns, which also gave a mixture of
the same compounds. Then, it was found that com-
pound 1 was unstable around neutral or basic pH, and
gradually decomposed to give lactate, which was iden-
We are now examining the effect of leaf-closure with 1
on the survival of the L. leucocephalam. On the other
hand, we have observed the leaf-opening activity in a
fraction separated by the column chromatography
using a TSK G3000S gel, and the isolation of the
leaf-opening substance is now in progress. Either (or
both) of these leaf-movement factors would be a poten-
tial herbicide of highly selective action against L.
leucocephalam.
1
tified by negative-mode FAB MS and H NMR spectra.
Then, we tried to isolate 1 by HPLC using a combina-
tion of two Develosil ODS-SR-5 columns under acidic
conditions (5% MeOH aq. containing 0.5% TFA).
Under this condition, we obtained 1 as a free acid form
and the corresponding g-lactone, 2,3-dihydroxy-2-
methyl-4-butanolide (2), which was produced from 1
under acidic condition, as a mixture of 1:1. Despite6
many trials, further purification of 1 was unsuccessful.
However, isolated potassium lactate showed no leaf-
closing activity against a leaf of L. leucocephalam; thus,
co-existing 1 was assumed to be the genuine leaf-closing
substance.
Acknowledgements
We are indebted to the Ministry of Education, Science,
Sports and Culture (Japan) for Grant-in-Aid for Scien-
tific Research (No. 12045259 and No. 12680598), Pio-
neering Research Project in Biotechnology given by the
Ministry of Agriculture, Forestry and Fisheries, Kihara
Foundation, Saneyoshi Foundation, Sumitomo Foun-
dation, and Kanagawa Academy of Science and Tech-
nology Research Grant for financial support. And we
thank Dr. Kazuhiko Nakamura (Biomolecules Depart-
ment, National Institute of Bioscience and Human-
Technology) for the collection of the plant material.
We carried out a bioassay using a mixture of 1 and
lactate in a ratio of 2:1 which was dissolved in a 0.1
mM phosphate buffer solution at pH 5.4. From the
content of 1 in this 2:1 mixture, 1 was estimated to be
effective for the leaf-closing of L. leucocephalam at
−
6
1
(
×10 M. On the other hand, corresponding g-lactone
−4
2) showed no leaf-closing activity even at 1×10 M.
NOE
(
3.7%)
H C
3
HO
OH
CH
OH
H
3
2
3
0.01 M HCl aq.
5
KOOC
OH
4
O
OH
O
1
2
Figure 2. The preparation of g-lactone (2) and the determination of stereochemistry in potassium 2,3,4-trihydroxy-2-methylbu-
tanoate (1).

M. Ueda et al. / Tetrahedron Letters 42 (2001) 3109–3111
3111
References
Paulsen, B. S.; Nordal, A. Phytochemistry, 1980, 19, 543–
45.
5
1
2
3
4
5
6
. Darwin, C. The Power of Movement in Plants. Third
Thousand; John Murray: London, 1882.
. B u¨ nning, E. The Physiological Clock, 3rd ed; English
Univ. Press: London, 1973.
. Ueda, M.; Yamamura, S. Angew. Chem., Int. Ed. 2000, 39,
7. The NMR and MS experiments were carried out on a
mixture of 1 and potassium lactate. Potassium 2,3,4-trihy-
1
droxy-2-methylbutanoate (1): H NMR (400 MHz, D O,
2
rt): 3.80 (1H, t, J=5.5 Hz, H ), 3.57 (2H, d, J=5.5 Hz,
3
13
H ), 1.34 (3H, s, Me) ppm.; C NMR (100 Mhz, D O, rt):
4
2
1
400–1414.
. Ueda, M.; Sawai, Y.; Yamamura, S. Tetrahedron 1999, 55,
0925–10936.
183.0 (C ), 80.0 (C ), 78.6 (C ), 65.0 (C ), 25.0 (Me) ppm.;
1 2 3 4
−
FAB-MS (negative): [M−H] m/z 149; FAB-MS (positive):
+
1
[M+H] m/z 189.
8. 2,3-Dihydroxy-2-methyl-4-butanolide(2): H NMR (400
1
. Rice, E. L. Allelopathy, 2nd ed; Academic Press: Orlando,
FL, 1984.
. No example of the isolation of 1 has been reported.
Only the detection of 1 in plant extract using GC–MS
was reported in the following reference: Kringstad,
R.; Singsaas, A. O.; Rusten, G.; Baekkemoen, G.;
MHz, D O, rt): 4.58 (1H, dd, J=3.7, 11.0 Hz, H ), 4.29
2
4a
(1H, d, J=11.0 Hz, H ), 4.23 (1H, d, J=3.7 Hz, H ), 1.45
4b 3
(3H, s, H ) ppm.
5
9. Kobayashi, S.; Horibe, M.; Saito, Y. Tetrahedron 1994, 50,
9629–9642.
.
.