Synthesis and biological activity of 4'-methoxy derivatives of abscisic acid

Article abstract of DOI:10.1016/S0960-894X(00)00302-4

Replacing the 4'-carbonyl group of abscisic acid with a methoxy group does not affect the abscisic acid (ABA)-like activities of the product in barley aleurone protoplasts, although the reduction of ABA to 4'-hydroxyl derivatives significantly reduces the

Full text of DOI:10.1016/S0960-894X(00)00302-4

Bioorganic & Medicinal Chemistry Letters 10 (2000) 1571±1574
Synthesis and Biological Activity of 40-Methoxy Derivatives of
Abscisic Acid
Tadao Asami,a,* Yong Ki Min,a,b Takeshi Nakano,a Tomoki Matsuyama,a
Noboru Murofushi,b Isomaro Yamaguchib and Shigeo Yoshidaa
aRIKEN (The Institute of Physical and Chemical Research), 2-1 Hirosawa, Wako, Saitama 351-0178, Japan
bDepartment of Applied Biological Chemistry, The University of Tokyo, Bunkyo-Ku, Tokyo 113-8657, Japan
Received 24 March 2000; accepted 9 May 2000
AbstractÐReplacing the 40-carbonyl group of abscisic acid with a methoxy group does not a€ect the abscisic acid (ABA)-like
activities of the product in barley aleurone protoplasts, although the reduction of ABA to 40-hydroxyl derivatives signi®cantly
reduces the ABA-like activity of the products. This suggests that methoxy derivatives of abscisic acid might be used to produce
probes for ABA binding proteins. # 2000 Elsevier Science Ltd. All rights reserved.
Abscisic acid (ABA) is involved in the control of many
processes in plants, including the acceleration of abscis-
sion, induction of dormancy, inhibition of rooting, and
stimulation of stomatal closure.1 In addition, ABA has
attracted considerable attention because it is thought to
play an important role in the response to environmental
stresses. In spite of understanding the roles of ABA in
plant physiology, there is no information about the
actions of ABA on its receptors. To develop probes for
ABA-binding proteins, as well as to develop plant growth
regulators targeted towards speci®c biological activities, it
is important to identify regions of the ABA molecule that
can be modi®ed without a€ecting its biological activity.
like activity in a gibberellin-induced a-amylase synthesis
test in barley aleurone protoplasts,5 which suggests that
ABA should retain its activity with the substitution of
other functional groups at its 40-position.
Rose et al. synthesized ABA analogues with a methyl
ether at C-10 for probing the role of the hydroxyl group
of ABA.6 (+)-C-10-O-Methyl-ABA (2) is rapidly meta-
bolized by suspension-cultured maize cells to (+)-C-10-O-
methyl-phaseic acid (3) in a process seen as ana0logous to
ABA.7 The presence of a methyl ether on the 1 -position
of ABA does not interfere with enzymatic oxidation at
the 80-carbon. Although both enantiomers were less active
than ABA in a wheat embryo germination inhibition
assay, in maize cell growth inhibition, (+)-C-10-O-methyl
ABA exhibited stronger activity than (+)-ABA, suggest-
ing that a free C-10-hydroxyl group is not essential for
the biological activity of ABA in maize (Zea mays).
Todoroki et al. reported the potent activity of 80- and 90-
methoxyabscisic acids (4 and 5, respectively) as anti-
metabolic analogues of abscisic acid.8 These results sug-
gest that if a methoxy group is properly introduced into the
ABA molecule, derivatives likely keep the intrinsic activity
of ABA. In addition, if methoxy derivatives of abscisic acid
possess ABA-like activity, we might synthesize a molecular
probe for ABA binding protein by replacing the methoxy
group with a functionalized alkoxy group.
To determine the structural features of the ABA molecule
that are needed for initiating ABA responses in many
speci®c physiologic processes, the modi®cation and
derivatization of ABA molecules and biological assays
have been carried out since the discovery of ABA.2 In a
review, Milborrow3 concluded that almost any change of
the ABA molecule destroys or attenuates its activity in
stomatal bioassays, including replacement of the 40 car-
bonyl with a group that is not rapidly metabolized.
However, Hornberg and Weiler4 reported later that 0.1
mM ABA-40-tyrosylhydrazone had ABA-like activity in a
stomatal bioassay. A ¯uorescence-labeled abscisic acid
(1), in which a ¯uorescent functional group was tagged
to ABA-40-aminophenylcarbohydrazone, showed ABA-
In this context, we substituted a methoxy group for the 40
carbonyl of ABA to develop new compounds to use as
molecular probes to investigate ABA-binding proteins in
barley aleurone cells.
*Corresponding author. Tel.: +81-48-462-111: fax: +81-48-462-4674;
e-mail: tasami@postman.riken.gp.jp
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00302-4

1572
T. Asami et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1571±1574
Figure 1. Inhibition of cress germination by ABA and ABA derivatives.
Synthesis
(S)-(+)-ABA, a kind gift from Toray Co. Ltd. Tokyo,
was modi®ed to methyl abscisate by diazomethane treat-
ment and reduced successively with NaBH4 to give 40-
hydroxy derivatives of ABA, 6 and 7, respectively. The
epimers of the diols were easily separated by silica gel
column chromatography. Both diols were then methy-
lated with methyl iodide after being deprotonated with
1.1 equivalents of NaH in DMF. Finally, hydrolysis of the
methyl esters with KOH±EtOH gave the corresponding
target compounds, 8 and 9, respectively (Scheme 1). The
structures of 6 and 7 were elucidated according to the
report by Hirai et al.9,10
Activities
One of the typical activities of ABA is the inhibition of seed
germination.11 The inhibition of GA3-induced a-amylase
induction and induction of the accumulation of dehydrin
in unstressed barley seeds are also well-demonstrated
e€ects of ABA that have been used to determine whether
chemicals have ABA-like activity.12,13 Therefore, the
activities of chemicals were evaluated using cress seeds
and barley aleurone protoplasts, as reported by Asami
et al.14 The results of the bioassays are shown in Figures
1±3. In the cress germination test (Fig. 1), 40-methoxy
derivative 8 shows inhibitory activity at the same level
as ABA, whereas 40-methoxy derivative 9, in which the
methoxy group has a di€erent con®guration from 8, is
Figure 2. Inhibition of GA-inducible a-amylase induction by ABA
and ABA derivatives in barley aleurone protoplasts.
less active than 8. In the a-amylase induction test using
barley aleurone protoplasts (Fig. 2), 8 and ABA possess
similar potencies, but 9 is not as active as 8 and ABA.
The e€ects of 8, 9, and ABA on the induction of dehy-
drin gene expression were also tested in aleurone proto-
plasts by measuring dehydrin promoter activity in
transient assays (Fig. 3).20 (Æ)-ABA (10 mM) increased
Scheme 1.

T. Asami et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1571±1574
1573
catabolic inactivation to phaseic acid-type compounds
(11), since 8 and 9 have a double bond in their ring systems
that is not conjugated with the carbonyl group and makes
it dicult to cyclize the hydroxylated intermediates (10).
This may make them lead compounds as plant growth
regulators because inactivation of ABA to phaseic acid
is suggested as one of the reasons why ABA is not used
in the ®eld.19
Figure 3. Dehydrin induction activity of ABA and ABA derivatives in
barley aleurone protoplasts.
References and Notes
GUS activity about 3-fold over control levels. 8 and 9
also increased the GUS activity signi®cantly, but were
less active than (Æ)-ABA. The order of e€ectiveness of
these compounds is the same as that in the a-amylase
induction inhibition bioassay: (Æ)-ABA>8>9.
1. Zeevaart, J. A. D.; Creelman, R. A. Ann. Rev. Plant. Phy-
siol. Plant. Mol. Biol. 1988, 39, 439.
2. Ohkuma, K. Agric. Biol. Chem. 1966, 30, 434.
3. Milborrow, B. V. In Abscisic Acid; Addicott, F. T., Ed.;
Praeger Scienti®c: New York, 1983; pp 79±111.
4. Hornberg, C.; Weiler, E. W. Nature 1984, 310, 321.
5. Asami, T.; Tao, L.; Yamamoto, S.; Robertson, M.; Min, Y.
K.; Yoshida, S. Biosci. Biotech. Biochem. 1997, 71, 1198.
6. Rose, P. A.; Lei, B.; Shaw, A. C.; Barton, D. L.; Walker-
Simmons, M. K.; Abrams, S. R. Phytochem. 1996, 41, 1251.
7. Rose, P. A.; Cutler, A. J.; Loewen, M. K.; Hogge, L. R.;
Abrams, S. R. Phytochem. 1996, 42, 575.
Discussion
40-Methoxy ABA derivatives (8 and 9) showed ABA-like
activity in inhibiting cress germination and GA-inducible
a-amylase. However, these results alone are not sucient
to demonstrate that 8 and 9 have the same mechanism
of action as ABA and interact with an ABA receptor. 8
and 9 may be general inhibitors of metabolic processes.
Some phenyl compounds structurally mimicking ABA
are reported to inhibit plant growth, transpiration, or
GA-induced a-amylase synthesis, but they do not mimic
ABA in all the assays for ABA.15 17 Alternatively, 8 and
9 may inhibit steps in signal transduction between GA
perception and the expression of a-amylase gene by a
mechanism that di€ers from that of ABA. For example,
okadaic acid, a protein phosphatase inhibitor, blocked
GA-inducible a-amylase production and greatly reduced
the accumulation of a-amylase mRNA, but did not lead
to the accumulation of ABA-inducible products.18
Therefore, to determine whether 8 and 9 mimic ABA, we
examined the activity of 8 and 9 on ABA up-regulated
expression of the dehydrin gene. The promoter activity
indicated that 8 and 9 have dehydrin-induction activity,
and the combined results show that 8 and 9 act as ABA
agonists, not as so-called general toxins. The ABA-like
activity of 8 and 9 suggests that the analogue may have
the same mechanism of action as ABA and may be
recognized as active ABA in in vitro assay systems.
8. Todoroki, Y.; Hirai, N.; Koshimizu, K. Biosci. Biotech.
Biochem. 1994, 58, 707.
9. Hirai, N.; Okamoto, M.; Koshimizu, K. Phytochem. 1986,
25, 1865.
10. 1H NMR (300 MHz, CDCl3, d): 0.91 (3H, s), 1.05 (3H, s),
1.68 (1H, dd, J=13.4 and 9.8 Hz), 1.68 (3H, s), 1.80 (1H, dd, m),
2.02 (3H, d, J=1.1 Hz), 3.39 (3H, s), 3.92 (1H, m), 5.66 (1H, s),
5.73 (1H, s), 6.21(1H, d, J=16.2 Hz), 7.75 (1H, d, J=16.2 Hz). 9:
1H NMR (300 MHz, CDCl3, d): 0.95 (3H, s), 1.04 (3H, s), 1.71
(3H, brs), 1.72 (1H, dd, J=13.7 and 7.0 Hz),1.74 (1H, brs), 2.02
(3H, d, J=1.1 Hz), 3.39 (3H, s), 3.80 (1H, m), 5.73 (1H, s), 5.74
(1H, m), 6.05 (1H, d, J=16.2 Hz), 7.78 (1H, d, J=16.2 Hz).
11. Reeve, D. R.; Crozier, A. In Encyclopedia of Plant Phy-
siology; MacMillan, J. Ed.; Springer-Verlag: Berlin, 1980; New
Series Vol. 9, pp 203±280.
12. Chrispeels, M. J.; Varner, J. E. Nature 1966, 212, 1066.
13. Chandler, P. M.; Robertson, M. Ann. Rev. Plant Physiol.
Plant Mol. Biol. 1994, 45, 113.
14. Asami, T.; Robertson, M.; Yamamoto, S.; Yoneyama, K.;
Takeuchi, Y.; Yoshida, S. Plant Cell Physiol. 1998, 39, 342.
15. Bittner, S.; Gorodetsky, M.; Har-Paz, I.; Mizrahi, Y.;
Richmond, A. E. Phytochem. 1977, 16, 1143.
16. Ladyman, J. A.; Sanborn, J. R.; Eelkema, E. E. Phy-
tochem. 1988, 27, 3751.
17. Yoshikawa, H.; Fujimoto, E.; Doi, K. Biosci. Biotech.
Biochem. 1992, 56, 256.
18. Kuo, A.; Cappelluti, S.; Cervantes-Cervantes, M.; Rodri-
guez, M.; Bush, D. S. Plant Cell 1996, 8, 259.
19. Dunstan, D. I.; Bock, C. A.; Abrams, G. D.; Abrams, S.
R. Phytochem. 1992, 31, 1451.
20. Dehydrin promoter assay was conducted according to the
procedure below. Protoplasts were resuspended in 1 mL of
incubation medium (IM) consisting of 0.5 M mannitol, 0.11 M
glucose, 0.055 M sucrose, 14 mM l-arginine, 10 mM MES,
and 0.32% B5-Gamborg salt, pH 5.5. They were mixed gently
We have demonstrated that ABA analogues of 8 and 9,
which have a methoxy group instead of a carbonyl group
at the 40-position of ABA, have ABA-like activity in
various biological responses. This result suggests that
modi®cation of the 40-position of ABA with an alkoxy
group might be a way to design probes for ABA binding
proteins. Furthermore, it is possible that they may tolerate

1574
T. Asami et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1571±1574
with 100 mg sheared salmon sperm DNA and 100 mg of Hv41
resuspended in an appropriate volume (generally 15 mL for
protoplasts isolated from 150 grains) of IM containing 50
unit/mL nystatin, 150 mg/mL cefotaxime, 20 mM CaCl2, 1.5
mg/mL aprotinin and 1.5 mg/mL leupeptin. One milliliter of the
protoplasts was aliquoted into ¯asks, and samples were incu-
bated in the dark at 25 ꢀC for 24 h in the presence of 0, 10 mM
(Æ) ABA, 10 mM 8, or 9 in triplicate for each treatment.
Fluorometric assays of GUS activity were performed in tripli-
cate for each sample.
21
(-935)-IGN plasmid DNA,
and the suspension was left
undisturbed for 1 min Protoplast transfection medium (PTM)
consisting of 17.3% (w/v) polyethylene glycol 6000 (PEG), 10
mM Tris±HCl pH 9.0, 0.67 M mannitol, and 0.133 M
Ca(NO3)2 was ®lter sterilized. Three volumes of PTM were
added to the protoplasts, mixed, and left at room temperature
for 20 min with occasional swirling. Then 40 mL of IM was
added in 10 mL aliquots with 2 min between additions. The
protoplasts were collected by centrifugation for 2 min at 50 g
and washed twice in 30 mL of IM. Pelleted protoplasts were
21. Robertson, M.; Cuming, A. C.; Chandler, P. M. Physiol.
Plant. 1995, 94, 470.