Synthesis and biological activity of 3'-chloro, -bromo, and - iodoabscisic acids, and biological activity of 3'-fluoro-8'-hydroxyabscisic acid

Article abstract of DOI:10.1016/S0031-9422(99)00444-6

(+)-3'-Chloro, -bromo and -iodoabscisic acids were synthesized, and their biological activities and metabolism were examined, as well as that of (+)-3'-fluoroabscisic acid. This was in order to estimate the effects of altered electron density at the C-2' position of abscisic acid on both biological activity and metabolic stability. The biological activities of (+)-3'-chloro and bromoabscisic acids were similar to, or higher than, those of (+)-abscisic acid in four bioassays, while (+)-3'-iodoabscisic acid was more potent in two bioassays and less potent than (+)-abscisic acid in the other two bioassays. The biological activities of 3'-haloabscisic acids seem to be dependent on the 3'-halogens, but not on the electron density at C-2'. The metabolism of (+)3'-haloabscisic acids to ca. 50% of their original levels in the media of rice cell suspension culture was similar to that of (+)abscisic acid. Metabolites from (+)-3'-chloro, -bromo and -iodoabscisic acids were not found in the culture media and cells, while (+)-3'- fluoroabscisic acid gave (+)-3'-fluoro-8'-hydroxyabscisic acid as a stable metabolite. (+)-3'-fluoro-8'-hydroxyabscisic acid showed biological activities lower than (+)-3'-fluoroabscisic acid, but higher than (-)-3'α- Fluorophaseic acid. These findings suggested that the biological activity of (+)-8'-hydroxyabscisic acid is intermediate between those of (+)abscisic acid and (-)-phaseic acid. Abscisic acid may be inactivated in a stepwise manner by metabolism in plants.

Full text of DOI:10.1016/S0031-9422(99)00444-6

Phytochemistry 52 (1999) 1185±1193
Synthesis and biological activity of 3'-chloro, -bromo, and
-iodoabscisic acids, and biological activity of 3'-¯uoro-
8'-hydroxyabscisic acid
Sho Arai, Yasushi Todoroki, Satomi Ibaraki, Yoshikazu Naoe, Nobuhiro Hirai*,
Hajime Ohigashi
Division of Applied Life Sciences, Graduate School of Agriculture, Kyoto University, Kyoto 606-8502, Japan
Received 20 January 1999; received in revised form 26 May 1999
Abstract
(+)-3'-Chloro, -bromo and -iodoabscisic acids were synthesized, and their biological activities and metabolism were examined,
as well as that of (+)-3'-¯uoroabscisic acid. This was in order to estimate the e€ects of altered electron density at the C-2'
position of abscisic acid on both biological activity and metabolic stability. The biological activities of (+)-3'-chloro and -
bromoabscisic acids were similar to, or higher than, those of (+)-abscisic acid in four bioassays, while (+)-3'-iodoabscisic acid
was more potent in two bioassays and less potent than (+)-abscisic acid in the other two bioassays. The biological activities of
3'-haloabscisic acids seem to be dependent on the 3'-halogens, but not on the electron density at C-2'. The metabolism of (+)-
3'-haloabscisic acids to ca. 50% of their original levels in the media of rice cell suspension culture was similar to that of (+)-
abscisic acid. Metabolites from (+)-3'-chloro, -bromo and -iodoabscisic acids were not found in the culture media and cells,
while (+)-3'-¯uoroabscisic acid gave (+)-3'-¯uoro-8'-hydroxyabscisic acid as
a stable metabolite. (+)-3'-¯uoro-8'-
hydroxyabscisic acid showed biological activities lower than (+)-3'-¯uoroabscisic acid, but higher than ( )-3'a-Fluorophaseic
acid. These ®ndings suggested that the biological activity of (+)-8'-hydroxyabscisic acid is intermediate between those of (+)-
abscisic acid and ( )-phaseic acid. Abscisic acid may be inactivated in a stepwise manner by metabolism in plants. # 1999
Elsevier Science Ltd. All rights reserved.
Keywords: Abscisic acid; 3'-Bromoabscisic acid; 3'-Chloroabscisic acid; 3'-Fluoroabscisic acid; 3'-Fluoro-8'-hydroxyabscisic acid; 8'-Hydroxyabs-
cisic acid; 3'-Iodoabscisic acid; Phaseic acid
1. Introduction
Although any direct measurement of biological activity
of 2 is dicult due to its spontaneous cyclization to 3,
8'-hydoxy-ABA (2) showed a biological activity similar
to that of 1 in stimulating biosynthesis of very long
chain fatty acids in Brassica napus embryos (Zou,
Abrams, Barton, Taylor, Pomeroy & Abrams, 1995).
On the other hand,it displayed only weak biological
activity in inhibition of germination of wheat embryos
The plant hormone (+)-abscisic acid (ABA, 1)
induces adaptive reactions to protect plants from en-
vironmental stresses such as desiccation and low tem-
perature (Davies & Jones, 1991). Plants metabolize 1
to (+)-8'-hydroxy-ABA (2) which cyclizes to ( )-pha-
seic acid (3) which has a low biological activity, by
nucleophilic attack of the 8'-hydroxyl group on C-2',
numbered as in 1, and 3 is metabolized to inactive ( )-
dihydrophaseic acid (4) (Fig. 1) (Hirai, 1999).
(Walker-Simmons, Holappa, Abrams
&
Abrams,
1997). This di€erence in observed responses in the two
bioassays might be due to a di€erential decrease in
concentrations of 2 in the assay media by cyclization
of 2 to 3 during the assays. In other studies, more
stable analogs of 2, namely, 8'-methoxy- and 8'-¯uoro-
* Correponding author.
E-mail address: hirai@kais.kyoto-u.ac.jp (N. Hirai).
0031-9422/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved.
PII: S0031-9422(99)00444-6

1186
S. Arai et al. / Phytochemistry 52 (1999) 1185±1193
Fig. 1. Metabolic pathway of 1 and 5, and structures of 7±9.
ABAs, were synthesized in attempts to estimate the
probable biological efects of 2 (Todoroki, Hirai &
Koshimizu, 1994, 1995b); the cyclization of 8'-meth-
oxy-ABA is blocked by replacement of the 8'-hydroxyl
group with a methoxy group. Additionallly, ¯uorine is
electronically equivalent to the oxygen of the hydroxyl
group; the C±F bond is physicochemically similar to
the C±OH bond, with the distinct di€erence between
the mono¯uoro and hydroxyl groups being the lack of
a dissociable hydrogen, i.e. these properties of ¯uorine
make 8'-¯uoro-ABA a stable mimic of 2 which cannot
cyclize to 3. Interestingly, 8'-methoxy- and 8'-¯uoro-
ABAs were as active as 1 in inhibition of lettuce seed
germination (Todoroki et al., 1994; 1995b), perhaps
implying that 2 might also be as active as 1. In any
case, an ABA analog incapable of cyclization after 8'-
hydroxylation could show high biological activity.
We previously synthesized (+)-3'-¯uoro-ABA (5)
which has an increased electron density at C-2' due to
repulsion between the p-electron at C-2' and the elec-
trons of the outermost shells of the ¯uorine atom, and
reported its bioactivity and metabolism, and the stab-
ility of (+)-3'-¯uoro-8'-hydroxy-ABA (6) (Todoroki,
Hirai & Ohigashi, 1995). In contrast to ¯uorine, other
halogens, i.e., chlorine, bromine and iodine, would be
expected to have very little e€ect on the electron den-
sity at C-2' (Breitmaier & Voetler, 1987). To compare
the e€ects of the di€erent halogens on the electron
densities at C-2', on the bioactivities and on metab-
olism with those of ¯uorine, (+)-3'-chloro-, bromo-
and iodo-ABAs (7, 8 and 9, respectively) were syn-
thesized.
2. Results and discussion
2.1. Synthesis
(+)-2',3'-Dihydro-2'a,3'a-epoxy-ABA (11) was syn-
thesized from 1, and the chloro analog 7 was syn-
thesized from 11 via (+)-chlorohydrin (12) by two
treatments with hydrogen chloride (Fig. 2). The pos-
ition of the chlorine in 12 could not be assigned from
1
its H-NMR spectrum since the signal of the H-3' pro-
ton in the other possible product (i.e. 2'-chloro-3'-
hydroxy derivative) would be expected to have a simi-
lar chemical shift to that of 12; however, conversion of
12 to 7 indicated that the chlorine was bound to C-3',
with both the existence and the position of chlorine in
7 being con®rmed by the molecular formula from
HRMS and by the disappearance of H-3' in the 1H-
NMR spectrum, respectively. This indicated that the
chloride ion attacked the b-face of C-3' of 11 which is
less sterically hindered than C-2' to open the 2'a,3'a-
epoxy ring. The low yield of 7 from 12 would be
caused by the cis relationship between the 2'-hydroxyl
group and H-3'.
The bromo analog 8 was synthesized from 11 via 13
by the same method as used for 7 with hydrogen bro-
mide. However, this method could not be applied to
synthesis of the iodo analog 9. Treatment of (2)-11
with hydrogen iodide gave (2)-1, suggesting that iod-
A previous study showed that the e€ect of 5 in a
rice seedling assay was similar to that of 1, although 6
was more stable than 2 (Todoroki et al., 1995). This
®nding, however, also suggested that the bioactivity of
6 was lower than that expected. We, therefore, tested
the bioactivity of 6 to examine the e€ects of 8'-hydrox-
ylation on the bioactivity of 5, and to further compare
the bioactivity of 2 with that of 6.
In this paper, we describe the synthesis and biologi-
cal activity of (+)-3'-chloro-, bromo- and iodo-ABAs
(7, 8 and 9, respectively), metabolism of 5 and 7±9 in
a rice cell suspension culture, and biological activities
of 6 and ( )-3'a-¯uorophaseic acid (10) (Fig. 1).
Fig. 2. Synthesis of (+)-3'-halo-ABAs, 7±9. R is the side chain of 1.

S. Arai et al. / Phytochemistry 52 (1999) 1185±1193
1187
ide ion attacked the 3'-iodine of iodohydrin (14), a
putative intermediate, to eliminate the 2'-hydroxyl
group. Reaction of (2)-11 with sodium iodide also
gave (2)-1. Direct iodination of methyl ester of (2)-1
with iodine azide gave 2Z-2-iodo-ABA (2)-15
(McIntosh, 1971). 2-Iodo-4-oxoisophorone (16) was
synthesized from 4-oxoisophorone with iodine azide,
but the 4-carbonyl group of 16 could not be protected
before introduction of the side chain. The ethylene
ketal of 4-oxoisophorone did not react with iodine
azide (data not shown). Methyl ester (17) of 9 was
®nally synthesized from the methyl ester (18) of 8 by
copper iodide and potassium iodide (Suzuki, Kondo &
Ogawa, 1985), and hydrolysis of 17 with esterase gave
9 (Fig. 2). The CD spectra of 7±9 showed the same
Cotton e€ects as that of 1, con®rming that the absol-
ute con®guration at C-1' of the analogs was S.
shifts of C-3' in 5 and 7±9 were correlated with their
electronegativities (Breitmaier & Voetler, 1987; Downs
& Adams, 1973). Changes in Mulliken charges at C-2'
and C-3' of 1, 5 and 7±9 paralleled with those of the
chemical shifts.
2.2. Biological activity
The biological activities of 7±9 were evaluated with
four bioassays along with 1, 3, 5, 6 and 10: elongation
of the second leaf sheath of rice seedlings; a-amylase
induction by gibberellin A3 in barley half-seeds with-
out embryos; lettuce seed germination; and stomatal
opening of the epidermal strips of spiderwort. The
results are summarized in Table 2.
In the rice seedling assay, the bioactivities of 7±9
were similar to those of 1 and 5. The bioactivities of
the 3'-halo-ABAs were not a€ected by the electron
density at C-2' or by the 3'-halogens themselves, indi-
cating that this assay is tolerant of steric size and
hydrophobicity of the 3'-halogens. The bioactivity in
this long term assay could be a€ected by metabolism
of analogs in addition to their anity to carriers and
receptors, suggesting that the 3'-halogens did not a€ect
the metabolic rate of the analogs. The results of the a-
amylase assay resembled those of the rice seedling
assay. The bioactivities of 5 and 7±9 were similar to
that of 1. This assay seemed not to be a€ected by
steric size and hydrophobicity of the 3'-halogens. The
bioactivities of 7±9 in the lettuce seed assay were
higher than those of 1 and 5. This result suggested
that the bioactivities of 1, 5 and 7±9 were correlated
not with the electron density at C-2' but with steric
size and hydrophobicity of 3'-substituents (Downs &
Adams, 1973; Moriguchi, Kanada & Komatsu, 1976).
ABA-binding components, such as receptors, of lettuce
seeds might prefer large and hydrophobic substituents
at C-3'.
The electron densities of 7±9 were estimated by 13C-
NMR analyses (Breitmaier & Voetler, 1997) and calcu-
lations, and Table 1 summarizes the chemical shifts
and Mulliken charges of C-2' and C-3' of 1, 5 and 7±
9. C-2' and C-3' signals of 7±9 were assigned by their
HMQC and HMBC spectra. The e€ects of 3'-halogens
on chemical shifts at C-2' and C-3' were as expected
(Breitmaier & Voetler, 1987). The up-®eld shift of C-2'
of 5 was the highest among the compounds examined,
con®rming the e€ect of the 3'-¯uorine on the increase
in electron density at C-2'. On the other hand, the
chemical shift at C-2' of 7 suggested that the 3'-chlor-
ine had a smaller e€ect on p-electron at C-2'. The
e€ects of 3'-bromine on the electron density at C-2'
were very small since the di€erence between the chemi-
cal shifts of C-2' of 1 and 8 was only 1.4 ppm. The
chemical shift of C-2' of 9 showed that the 3'-iodine
slightly decreased the electron density at C-2'. As
expected, the e€ects of the halogens on the chemical
In the stomatal opening assay, the bioactivities of 5,
7 and 8 were similar to that of 1, but 9 showed 1/9
bioactivity of 1. This result showed that bioactivity
was not correlated with electron density at C-2'.
Bioactivity in this assay most re¯ected anity to recep-
Table 1
Chemical Shifts ꢀd ppm) at C-2' and C-3' in 1, 5 and 7±9 in the 13C-NMR spectra (125 MHz, acetone-d6) and Mulliken chargesa
Compound
Chemical shifts
Mulliken charges
C-2'
C-3'
C-2'
C-3'
0.2357
1
5
7
8
9
163.0
127.1
0.0222
138.3 ( 24.7)
158.2 ( 4.8)
161.6 ( 1.4)
167.5 (+4.5)
151.0 (+23.9)
130.6 (+3.5)
124.8 ( 2.3)
119.4 ( 7.7)
0.0662 ( 0.0440)
0.0241 ( 0.0019)
0.0095 (+0.0127)
0.0032 (+0.0190)
0.0611 (+0.1746)
0.2675 ( 0.0318)
0.2591 ( 0.0234)
0.2787 ( 0.0430)
a Shifts relative to 1 are shown in parentheses.

1188
S. Arai et al. / Phytochemistry 52 (1999) 1185±1193
Table 2
IC50 values for 1, 3 and 5±10 in bioassays
Compound
Rice elongation (mM)
Barley a-amylase (mM)
Lettuce germination (mM)
Stomatal opening (nM)
1
3.4
70
4.2
9.2
9.4
16.0
> 100
6.5
2.2
285
4.4
3
5
4.2
17
a
6
±
64
2.9
16.0
0.8
7
3.4
2.4
3.9
100
3.0
1.3
2.7
8
2.5
2.2
1.2
20.4
190
9
10
a
±
> 100
a Not tested.
tors, and the e€ect of metabolism on the bioactivity
was small due to the short assay time, which is in
agreement with the lack of a correlation between the
bioactivity and the electron density at C-2'. The low
bioactivity of 9 seemed to be caused by conformational
changes due to the bulky 3'-iodine, but no such
changes were observed in its NOESY spectrum which
indicated a cross peak between H-5 and H-5'b as well
as that of 1 (Milborrow, 1984). It appears that the
large size of the iodine may be limiting bioactivity
otherwise the other halides are all similarly active.
More bulky groups at C-3' than bromine, the van der
Waals radius of which is 0.198 nm (Downs & Adams,
1973) may decrease anity of receptors in guard cells
for analogs. The bioactivity of a 3'-thioether analog
was also 1/10 that of 1 in the cress seed germination
assay (Balsevich, Bishop & Banowetz, 1997). The
possibility that the electronic characteristics of iodine
rather than steric size a€ects bioactivity in this assay
could not be excluded. The slightly higher bioactivities
of 7 and 8 than that of 1 might have been due to the
hydrophobicity of the 3'-halogens with adequate steric
size.
Changes in the electron density at C-2' in the 3'-
halo-ABAs did not a€ect the bioactivities in the four
assays. The bioactivities of 3'-halo-ABAs seemed to be
dependent on the physicochemical properties of the 3'-
halogens, i.e., electronegativity, steric size and hydro-
phobicity, rather than the e€ect on 2'-electron density.
However, it is unclear at present what characteristics
of chlorine and bromine contribute to increases in the
bioactivity in the lettuce seed and stomatal assays.
Analog (2)-15 was inactive in the rice seedling and let-
tuce seed assays probably due to its having the same
geometry of the side chain as 2E-1.
in their concentrations in the media. Concentrations of
5 and 7 decreased at rates similar to that of 1, while 8
and 9 concentrations decreased faster than that of 1,
with in each case about 50% of each of the four ana-
logs being metabolized in 0.6±0.8 days. 3'-Halo-8'-
hydroxy-ABAs and 3'-halophaseic acids derived from
7±9 were not found in the media and cells while 6
formed from 5 was detected in the media, suggesting
that 7±9 were metabolized by a di€erent pathway from
8'-hydroxylation, or their 8'-hydroxylated derivatives
are unstable and decomposed to unknown metabolites.
This ®nding suggested that the biological activities of
7±9 were dependent not on the metabolic stability but
on the nature of the 3'-substituents themselves. The
concentration of 6 increased to 7 mM after 2 days, and
then decreased gradually thereafter (Fig. 3b).
Formation of 10 and 3'b-¯uorophaseic acid was not
observed. If 6 is as active as 5, this metabolite could
contribute to the increase in bioactivity of 5 in long
term assays.
Inhibitory activities of 6 are summarized in Table 2
along with those of 10 prepared from 6. Bioactivities
of 6 were higher than those of 10, but lower than
those of 5 in rice seedling, lettuce seed, and stomata
assays. The bioactivities of 10 were low similarly to 3.
After incubation of an aqueous solution of 6 under the
same conditions as used for the rice seedling assay for
7 days, the ratio of 6, its 2E isomer, 10 and its 2E iso-
mer in the solution was 64 : 31 : 4 : 1, showing that
cyclization of 6 to 10 was small and the amount of 2-
2.3. Metabolism of 3'-halo-ABAs and biological activity
of (+)-3'-¯uoro-8'-hydroxy-ABA (6)
3'-Halo-ABAs, 5 and 7±9, were added at 20 mM to
suspension cultures of rice cells to examine whether
they were metabolized or not. Fig. 3 showed changes
Fig. 3. Changes in concentrations of 1 and 7±9 (a) and of 5 and 6
(b) in the culture media. (a) w, 1; *, 7; r, 8; R, 9. (b) w, 5; *, 6.

S. Arai et al. / Phytochemistry 52 (1999) 1185±1193
1189
E-6 formed was similar to that of 2E-1 formed from 1
(data not shown). These observations con®rmed that
the bioactivities of 6 were almost entirely due to the
analog itself. The low bioactivity of 6 would explain
why the bioactivity of 5 was similar to that of 1 in the
long term assay although 6 was stable. C-8' and C-9'
of 1 seemed to interact with a hydrophobic site on
ABA-binding proteins since (+)-8' and (+)-9'-alkyl-
ated analogs showed bioactivities similar to that of 1
(Nakano, Todoroki, Hirai & Ohigashi, 1995). If this
was the case, the low bioactivity of 6 may have been
due to a negative e€ect of the 8'-hydroxyl group on
the hydrophobic interaction. A hydrogen bonding in
which the 8'-hydroxyl group acts as a proton donor
may cause the negative e€ect. Methoxy and ¯uorine
groups cannot participate in hydrogen bonding in
which these groups act as proton donor (Todoroki,
Hirai & Koshimizu, 1995a, 1995b), which may explain
the di€erence in the bioactivities between 6, and 8'-
methoxy- and 8'-¯uoro-ABAs. This presumably means
that 6 more re¯ects the e€ect of the 8'-hydroxyl group
on the bioactivities than 8'-methoxy- and 8'-¯uoro-
ABAs.
ARX500 (500 MHz) spectrometers. 13C-NMR spectra
were recorded reference to the acetone-d6 as 30.2 ppm.
For clarity, the conventional ABA numbering system
was used to assign peaks in the 1H- and 13C-NMR
spectra. Mass spectra were obtained using a Jeol JMS-
DX300/DA5000 mass spectrometer, whereas GC±MS
was conducted with a 1% OV-17 column (1 m  3
mm) at a He ¯ow rate of 60 ml/min, and the oven
temperature was increased from 908 to 2008 at a rate
of 328/min. Optical rotations were measured with a
Jasco DIP-1000 digital polarimeter, and CD spectra
were recorded with a Jasco J-720W spectropolarimeter.
3.2. (+)-2',3'-Dihydro-2'a,3'a-epoxy-ABA (11) and
(+)-3'-¯uoro-ABA (5)
These compounds were synthesized as described
1
(Todoroki et al., 1995). H-NMR (500 MHz, CDCl3)
of 11: d 0.94 (3H, s, H-8'), 0.99 (3H, s, H-9'), 1.39
(3H, s, H-7'), 1.84 (1H, dd, J = 14.6 Hz, H-5'), 2.07
(3H, s, H-6'), 2.84 (1H, d, J = 14.6 Hz, H-5'), 3.30
(1H, s, H-3'), 5.79 (1H, s, H-2), 6.11 (1H, d, J = 15.8
1
Hz, H-5), 8.02 (1H, d, J = 15.8 Hz, H-4). H-NMR
Thus, bioactivity of the unstable metabolite 2 can be
estimated from the bioactivity of 6. Bioactivities of 5
and 10 were similar to those of 1 and of 3, respect-
ively, indicating that the e€ect of the 3'-¯uorine on the
bioactivities was small. This suggests that bioactivity
of 2 must be similar to that of 6. These observations
indicated that the biological activity of 2 is probably
lower than that of 1, and higher than that of 3.
Consequently, 1 would be inactivated in a stepwise
manner by metabolism in plants: in the ®rst step, 1
with high bioactivity is converted to 2 with medium
bioactivity by 8'-hydroxylation; then, 2 is cyclized to 3
with low bioactivity; and in the third step, 3 is metab-
olized to inactive 4 by reduction of the 4'-carbonyl
group. Inactivation of 1 is completed probably at 4,
since 3 can isomerize to 2 again unless the 4'-carbonyl
group of 3 is reduced. Low bioactivities of 3 could be
partially caused by a small amount of 2 formed from 3
by isomerization, suggesting that bioactivity of 3 itself
is lower than that observed. The stepwise inactivation
of 1 suggests that an analog resistant to 8'-hydroxyl-
ation such as (+)-8',8',8'-tri¯uoro-ABA would show
long lasting bioactivity as was found by Todoroki et
al. (1995a).
(500 MHz, CDCl3) of 5: d 1.05 (3H, s, H-9'), 1.14
(3H, s, H-8'), 1.87 (3H, d, J = 3.2 Hz, H-7'), 2.05
(3H, s, H-6), 2.44 (1H, br. d, J = 17.1 Hz, H-5'), 2.49
(1H, d, J = 17.1 Hz, H-5'), 5.79 (1H, br. s, H-2), 6.11
(1H, d, J = 16.1 Hz, H-5), 7.81 (1H, d, J = 16.1 Hz,
H-4).
3.3. (+)-2Z,4E-5-[(1S,2S,3R )-3-Chloro-1,2-
dihydroxy-2,6,6-trimethyl-4-oxocyclohexan-1-yl]-3-
methyl-2,4-pentadienoic acid [2',3'-dihydro-3'b-chloro-
2'a-hydroxy-ABA (12)]
To a stirred solution of 11 (89 mg) in MeOH (4.5
ml) was added 10% HCl (4.5 ml), and the mixture was
stirred at 188 for 2.4 h. Then, the reaction mixture was
diluted with H2O to 40 ml and extracted with EtOAc.
The organic layer was washed with H2O, dried
(Na2SO4), ®ltered and evaporated to dryness. The re-
sidual oil was next subjected to silica gel chromatog-
raphy using n-hexane±EtOAc±HOAc (80 : 20 : 1) as
30
eluant to a€ord 12 (33 mg, 33% yield). ‰aŠ +678
max
D
MeOH
1
(MeOH; c 0.049); UV l
nm ꢀlog e): 260 (4.19); H-
NMR (300 MHz, CDCl3): d 0.94 (3H, s, H-8' or H-
9'), 0.99 (3H, s, H-9' or H-8'), 1.39 (3H, s, H-7'), 1.83
(1H, dd, J =14.6 and 0.9 Hz, H-5'), 2.06 (3H, d, J
=1.1 Hz, H-6), 2.83 (1H, d, J =14.6 Hz, H-5'), 3.30
(1H, d, J = 0.6 Hz, H-3'), 5.79 (1H, s, H-2), 6.11 (1H,
d, J = 15.7 Hz, H-5), 8.02 (1H, d, J = 15.7 Hz, H-4).
Compound 12 (1 mg) was methylated with CH2N2 to
give the methyl ester (1 mg). EIMS (probe) 70 eV, m/z
(rel. int.): 332 [M + 2]+ (1.3), 330 [M]+ (4), 312 (1.5),
3. Experimental
3.1. General
1H- and 13C-NMR, NOESY, and HMQC and
HMBC spectra were recorded with TMS as an internal
standard using Bruker AC300 (300 MHz) and
295 [M
(20), 238 (26), 224 (76), 196 (39), 170 (38), 161 (63),
Cl]+ (10), 280 (14), 277 (23), 263 (18), 245

1190
S. Arai et al. / Phytochemistry 52 (1999) 1185±1193
153 (34), 125 (100), 121 (78); HRMS: [M]+ at m/z
or H-8'), 1.42 (3H, s, H-7'), 2.09 (3H, d, J = 1.2 Hz,
H-6), 2.34 (1H, d, J = 12.8 Hz, H-5'), 3.04 (1H, d, J
= 12.8 Hz, H-5'), 5.33 (1H, s, H-3'), 5.80 (1H, s, H-
2), 6.47 (1H, d, J = 15.6 Hz, H-5), 7.88 (1H, d, J =
15.6 Hz, H-4); FABMS (matrix, 3-nitrobenzyl alco-
hol), m=z (rel. int.): 385 [M + Na + 2]+ (12), 383 [M
+ Na]+ (14), 362 [M + 2]+ (13), 360 [M]+ (14), 345
CHCl3
330.1230 (C16H23O5Cl requires 330.1234); IR nmax
cm 1: 3000, 1740, 1730, 1710, 1690, 1600.
3.4. (+)-2Z,4E-5-[(1S )-3-Chloro-1-hydroxy-2,6,6-
trimethyl-4-oxo-2-cyclohexen-1-yl]-3-methyl-2,4-
pentadienoic acid [3'-chloro-ABA (7)]
(25), 343 (30), 327 (34), 303 (11), 281 [M
Br]+ (40),
To a stirred solution of 12 (17 mg) in MeOH (1.0
ml) was added 10% HCl (1.0 ml), and the mixture was
stirred at room temperature for 6.6 h. The reaction
mixture was diluted with H2O to 40 ml and extracted
with EtOAc. The organic layer was washed with H2O,
dried (Na2SO4), ®ltered and evaporated to dryness.
The residual oil was subjected to silica gel chromatog-
263 (51), 245 (26), 235 (15), 223 (58), 207 (100), 182
(32), 163 (75).
3.6. (+)-2Z,4E-5-[(1S )-3-Bromo-1-hydroxy-2,6,6-
trimethyl-4-oxo-2-cyclohexen-1-yl]-3-methyl-2,4-
pentadienoic acid and its methyl ester [3'-bromo-ABA
(8) and its methyl ester(18), respectively]
raphy using n-hexane±EtOAc±HOAc (80 : 20 : 1) as
26
eluant to a€ord 7 (6.5 mg, 42% yield). ‰aŠ +2218
MeOH
To a stirred solution of 13 (396 mg) in MeOH (30
ml) was added 30% HBr (6.0 ml), and the mixture was
stirred at room temperature 3.7 h. The reaction mix-
ture was diluted with H2O to 300 ml and extracted
with EtOAc. The organic layer was washed with H2O,
dried (Na2SO4), ®ltered and evaporated to dryness.
The residual oil was subjected to silica gel chromatog-
D
(MeOH; c 0.175); CD: De240 18.4, De268 +27.2,De329
1.9 (MeOH; c 0.001); UV l
nm ꢀlog e): 260
max
(4.28); 1H-NMR (300 MHz, CD3OD): d 1.03 (3H, s,
H-8' or H-9'), 1.05 (3H, s, H-9' or H-8'), 2.03 (3H, d,
J = 1.1 Hz, H-6), 2.07 (3H, s, H-7'), 2.40 (1H, d, J =
16.7 Hz, H-5'), 2.64 (1H, d, J = 16.7 Hz, H-5'), 5.75
(1H, s, H-2), 6.22 (1H, d, J = 16.1 Hz, H-5), 7.76
(1H, d, J = 16.1 Hz, H-4); 13C-NMR (125 MHz,
acetone-d6): d 17.90 (C-7'), 21.56 (C-6), 23.77 (C-8'),
24.86 (C-9'), 42.26 (C-6'), 50.13 (C-5'), 81.73 (C-1'),
119.45 (C-2), 129.76 (C-4), 130.55 (C-3'), 137.47 (C-5),
151.12 (C-3), 158.15 (C-2'), 167.51 (C-1), 189.83 (C-
4'). Compound 7 (1 mg) was methylated with CH2N2
to give the methyl ester (1 mg). EIMS (probe) 70 eV,
m=z (rel. int.): 314 [M + 2]+ (0.6), 312 [M]+ (1.2),
294 (7), 280 (10), 269 (4), 245 (31), 239 (5), 224 (76),
217 (9), 196 (30), 181 (7), 168 (20), 161 (42), 141 (5),
133 (23), 125 (100), 112 (26); HRMS: [M]+ at m=z
raphy with n-hexane±EtOAc±HOAc (80 : 20 : 1) as
26
eluant to a€ord 8 (86 mg, 23% yield). ‰aŠ +2598
MeOH
D
(MeOH; c 0.062); CD: De244 20.9, De272 +38.1, De325
3.0 (MeOH; c 0.001); UV l
nm ꢀlog e): 262
max
(4.38); 1H-NMR (300 MHz, CD3OD): d 1.03 (3H, s,
H-8' or H-9'), 1.04 (3H, s, H-9' or H-8'), 2.04 (3H, s,
H-6), 2.12 (3H, s, H-7'), 2.44 (1H, d, J = 16.7 Hz, H-
5'), 2.67 (1H, d, J = 16.7 Hz, H-5'), 5.75 (1H, s, H-2),
6.22 (1H, d, J = 16.1 Hz, H-5), 7.77 (1H, d, J = 16.1
Hz, H-4); 13C NMR (75 MHz, acetone-d6): d 21.43 (C-
7'), 21.55 (C-6), 23.78 (C-8'), 24.80 (C-9'), 42.25 (C-
6'), 49.99 (C-5'), 82.38 (C-1'), 119.57 (C-2), 124.78 (C-
3'), 129.83 (C-4), 137.30 (C-5), 151.02 (C-3), 161.60
(C-2'), 167.63 (C-1), 189.84 (C-4'). Compound 8 (56
mg) was methylated with CH2N2 to give methyl ester
(18) (57 mg). 1H-NMR (300 MHz, CDCl3): d 1.03
(3H, s, H-8' or H-9'), 1.11 (3H, s, H-9' or H-8'), 2.01
(3H, d, J = 1.2 Hz, H-6), 2.14 (3H, s, H-7'), 2.58 (1H,
d, J = 17.1 Hz, H-5'), 2.59 (1H, d, J = 17.1 Hz, H-
5'), 3.72 (3H, s, 1-CO2CH3), 5.78 (1H, s, H-2), 6.10
(1H, d, J = 16.0 Hz, H-5), 7.87 (1H, d, J = 16.0 Hz,
H-4); EIMS (probe) 70 eV, m=z (rel. int.): 358 [M +
2]+ (0.1), 356 [M]+ (0.1), 340 (0.2), 338 (0.2), 326
(0.5), 324 (0.5), 315 (0.2), 313 (0.2), 300 (0.1), 277 [M
Br]+ (10), 270 (42), 268 (44), 259 (11), 245 (42), 217
(11), 189 (8), 161 (47), 133 (27), 125 (100), 112 (23);
HRMS: [M]+ at m=z 356.0663 (C16H21O4Br requires
CHCl3
312.1134 (C16H21O4Cl requires 312.1128); IR nmax
cm 1: 2950, 1720, 1700, 1690, 1630, 1600.
3.5. (+)-2Z,4E-5-[(1S,2S,3R )-3-Bromo-1,2-
dihydroxy-2,6,6-trimethyl-4-oxocyclohexan-1-yl]-3-
methyl-2,4-pentadienoic acid [2',3'-dihydro-3'b-bromo-
2'a-hydroxy-ABA (13)]
To a stirred solution of 11 (680 mg) in MeOH (50
ml) was added 10% HBr (50 ml), and the mixture was
stirred at 188 for 1.3 h. The reaction mixture was
diluted with H2O to 400 ml and extracted with EtOAc.
The organic layer was washed with H2O, dried
(Na2SO4), ®ltered and evaporated to dryness. The re-
sidual oil was subjected to silica gel chromatography
356.0623); IR nmax cm 1: 2900, 1785, 1715, 1635,
CHCl3
using n-hexane±EtOAc±HOAc (76 : 24 : 1) as eluant
30
to a€ord 13 (314 mg, 36% yield). ‰aŠ +1008 (MeOH;
1600, 1450, 1435, 1370, 1240, 1160.
D
MeOH
max
CHCl3
c 0.075); UV l
nm ꢀlog e): 261 (4.35); IR nmax
cm 1: 3550, 3300±2500, 2950, 1725, 1685, 1635, 1600,
1450, 1385, 1245, 1180, 1120; 1H-NMR (300 MHz,
CDCl3): d 1.06 (3H, s, H-8' or H-9'), 1.10 (3H, s, H-9'
3.7. Treatment of 11 with HI and NaI
To a stirred solution of (2)-11 (178 mg) in MeOH

S. Arai et al. / Phytochemistry 52 (1999) 1185±1193
1191
(9 ml) was added 30% HI (1.8 ml), and the mixture
3.9. 3-Iodo-2,6,6-trimethyl-2-cyclohexen-1,4-dione [2-
iodo-4-oxoisophorone (16)]
was stirred at room temperature for 3.9 h. The reac-
tion mixture was diluted with H2O to 100 ml and
extracted with EtOAc. The organic layer was washed
with H2O, dried (Na2SO4), ®ltered and evaporated to
dryness. The residual oil was subjected to silica gel
chromatography using n-hexane±EtOAc±HOAc (85 :
15 : 1) as eluant to a€ord (2)-1 (40 mg, 24% yield).
1H-NMR (300 MHz, CD3OD): d 1.03 (3H, s, H-9'),
1.06 (3H, s, H-8'), 1.93 (3H, d, J = 1.3 Hz, H-7'),
2.04 (3H, d, J = 1.1 Hz, H-6'), 2.17 (1H, d, J = 16.9
Hz, H-5'), 2.52 (1H, d, J = 16.9 Hz, H-5'), 5.74 (1H,
br. s, H-2), 5.92 (1H, br. s, H-3'), 6.24 (1H, d, J =
16.2 Hz, H-5), 7.77 (1H, d, J = 16.2 Hz, H-4). To a
stirred solution of (2)-11 (172 mg) in HOAc (0.5 ml)
and propionic acid (1 ml) was added NaOAc (23 mg)
and NaI (187 mg), and the mixture was stirred at 208
for 1.6 h. The reaction mixture was diluted with 5%
NaHCO3 aqueous solution to 45 ml, acidi®ed with 3
M HCl to pH 3, and extracted with EtOAc. The or-
ganic layer was washed with 5% NaHCO3 aqueous
solution and H2O, dried (Na2SO4), ®ltered and concen-
trated to give (2)-1 as the major product.
To a stirred solution of sodium azide (23 g, 346
mmol) in 90 ml of CH3CN was added dropwise iodine
monochloride (8.6 g, 53 mmol) in 17 ml of CH3CN at
208C over a period of 15 min. Then 4-oxoisophorone
(7.6 g) was added to the mixture, followed by stirring
at room temperature for 28.5 h. The mixture was
diluted with H2O to 170 ml, and extracted with
EtOAc. The organic layer was washed with 10%
Na2SO3 aqueous solution and H2O, dried (Na2SO4),
®ltered and concentrated. The residual oil was sub-
jected to silica gel chromatography with toluene±
EtOAc (97 : 3) as eluant to a€ord 16 (5.8 g, 42%
1
yield). H-NMR (300 MHz, CDCl3): d 1.25 (6H, s, 5-
CH3 Â 2), 2.32 (3H, s, 3-CH3), 2.90 (2H, s, H-6); GC-
EIMS 70 eV, m=z (rel. int.): 278 [M]+ (24), 263 (11),
250 (4), 235 (4), 222 (12), 194 (5), 237 (50), 151 [M
I]+ (6), 127 (3), 108 (2), 67 (100).
3.10. (+)-Methyl 2Z,4E-5-[(1S )-1-hydroxy-3-iodo-
2,6,6-trimethyl-4-oxo-2-cyclohexen-1-yl]-3-methyl-2,4-
pentadienoate [methyl 3'-iodoabscisate (17)]
3.8. (2)-2Z,4E-5-(1-Hydroxy-2,6,6-trimethyl-4-oxo-2-
cyclohexen-1-yl]-2-iodo-3-methyl-2,4-pentadienoic acid
[2Z-2-iodo-ABA (15)]
A mixture of 18 (57 mg), potassium iodide (400 mg),
copper(I) iodide (156 mg), and hexamethylphosphora-
mide (0.48 ml) was stirred at 150 for 0.7 h under N2.
After the reaction was quenched by H2O, the mixture
was diluted with 10% Na2SO3 aqueous solution to 45
ml and extracted with EtOAc. The organic layer was
washed with 10% aqueous Na2SO3 and H2O, dried
(Na2SO4), ®ltered and evaporated to dryness. The re-
sidual oil was subjected to silica gel chromatography
using toluene±EtOAc (9 : 1) as eluant to a€ord 17 (29
To a stirred solution of sodium azide (194 mg, 2.98
mmol) in CH3CN (0.76 ml) was added iodine mono-
chloride (73 mg, 0.45 mmol) in 0.14 ml of CH3CN at
208C. The mixture was stirred for 35 min at 208C,
then the methyl ester of (2)-1 (115 mg, 0.41 mmol) in
1.5 ml of CH3CN was added to the mixture. The mix-
ture was stirred at 08C for 4.1 h and at room tempera-
ture for 18.9 h, then diluted with H2O to 45 ml,
acidi®ed with 3 M HCl to pH 3, and extracted with
EtOAc. The organic layer was washed with 10%
Na2SO3 aqueous solution and H2O, dried (Na2SO4),
®ltered and evaporated to dryness. The residual oil
was subjected to silica gel chromatography using n-
1
mg, 44% yield). H-NMR (300 MHz, CDCl3): d 1.03
(3H, s, H-8' or H-9'), 1.09 (3H, s, H-9' or H-8'), 2.01
(3H, d, J = 1.3 Hz, H-6), 2.25 (3H, s, H-7'), 2.63 (2H,
s, H-5'), 3.72 (3H, s, 1-CO2CH3), 5.78 (1H, d, J = 0.4
Hz, H-2), 6.09 (1H, dd, J = 16.1 and 0.3 Hz, H-5),
7.86 (1H, dd, J = 16.1 and 0.5 Hz, H-4); EIMS
(probe) 70 eV, m=z (rel. int.): 404 [M]+ (4), 386 (3),
hexane±EtOAc±HOAc (80 : 20 : 1) as eluant to a€ord
(2)-15 (16 mg, 9% yield). UV l
372 (9), 316 (46), 288 (12), 277 [M
(13), 245 (36), 227 (5), 217 (11), 203 (5), 189 (8), 161
I]+ (12), 259
MeOH
max
nm ꢀlog e): 237
(4.38); H-NMR (300 MHz, CDCl3): d 1.03 (3H, s, H-
8' or H-9'), 1.11 (3H, s, H-9' or H-8'), 1.89 (3H, d, J
= 1.4 Hz, H-7'), 2.31 (3H, s, H-6), 2.34 (1H, dd, J
=17.2 and 0.9 Hz, H-5'), 2.49 (1H, dd, J = 17.2 and
0.5 Hz, H-5'), 5.96 (1H, m, H-3'), 6.47 (1H, d, J =
15.7 Hz, H-5), 6.83 (1H, d, J = 15.7 Hz, H-4);
FABMS (matrix, 3-nitrobenzyl alcohol), m=z (rel. int.):
391 [M + H]+ (9), 358 (5), 341 (13), 316 (9), 265 [M
I]+ (77), 245 (33), 237 (50), 223 (100), 205 (100),
191 (31), 177 (56), 189 (8), 161 (43), 125 (100); IR
(43), 125 (100); HRMS: [M]+ at m=z 404.0477
1
(C16H21O4I requires 404.0483); IR nmax cm 1: 2950,
CHCl3
1700, 1680, 1635, 1600, 1370, 1165, 985.
3.11. (+)-2Z,4E-5-[(1S)-1-Hydroxy-3-iodo-2,6,6-
trimethyl-4-oxo-2-cyclohexen-1-yl]-3-methyl-2,4-
pentadienoic acid [3'-iodo-ABA (9)]
To a solution of 17 (29 mg) in a mixture of MeOH
(0.1 ml) and 0.1 M KH2PO4±K2HPO4 bu€er (pH 8.0,
4 ml) was added porcine liver esterase (EC 3.1.1.1,
2000 units). The mixture was shaken at 308C for 5.8 h,
nmax cm 1: 3603, 3411, 3300±2500, 3017, 2971, 2114,
CHCl3
1667, 1628, 1424, 1366, 1254, 1177, 1123, 984.

1192
S. Arai et al. / Phytochemistry 52 (1999) 1185±1193
then diluted with H2O, acidi®ed with 3 M HCl to pH
3 and extracted with EtOAc. The organic layer was
washed with H2O, dried (Na2SO4), ®ltered and evapor-
ated. The residual oil was subjected to silica gel chro-
3.15. Quanti®cation of 1, 5, and 7±9 in culture media
Cultured cells (4 g fresh weight) were transferred
into 70 ml of fresh medium in a 300 ml ¯ask, and a
solution of the compound (0.37 mg) in 0.5 ml of H2O
was added to yield a ®nal concentration of 20 mM.
The cultures were incubated at 248 for 8 days with
shaking. One-ml samples were collected from ¯asks for
HPLC analysis after 0, 1, 2, 3, 4 and 5 days. The
samples were kept at 208 until HPLC analysis on an
matography using n-hexane±EtOAc±HOAc (82 : 18 :
26
1) as eluant to a€ord 9 (21 mg, 77% yield). ‰aŠ
MeOH
D
+2248 (MeOH; c 0.021); CD: De243 5.2, De283 +23.7,
De329 2.2 (MeOH; c 0.001); UV l
nm ꢀlog e): 258
max
(4.37); 1H-NMR (300 MHz, CD3OD): d 1.03 (6H, s,
H-8' and H-9'), 2.03 (3H, d, J = 1.2 Hz, H-6), 2.22
(3H, s, H-7'), 2.49 (1H, d, J =16.6 Hz, H-5'), 2.70
(1H, d, J = 16.6 Hz, H-5'), 5.75 (1H, s, H-2), 6.20
(1H, d, J = 16.1 Hz, H-5), 7.75 (1H, d, J = 16.1 Hz,
H-4); 13C-NMR (75 MHz, acetone-d6): d 21.56 (C-6),
23.81 (C-8'), 24.65 (C-9'), 27.63 (C-7'), 42.31 (C-6'),
48.96 (C-5'), 82.43 (C-1'), 109.66 (C-3'), 119.40 (C-2),
129.74 (C-4), 137.42 (C-5), 151.16 (C-3), 167.47 (C-2'),
167.62 (C-1), 191.16 (C-4').
ODS column (YMC AQ-311, 6 i.d. Â 100 mm). The
1
column was eluted at 1.0 ml min
with 40±60%
MeOH in 0.1% aqueous HOAc, and the eluent was
monitored at 254 nm. Peaks in the chromatograms
were identi®ed by comparing their retention times with
those of authentic standards, and the amounts of com-
pounds were calculated with reference to calibration
curves of the authentic samples between peak area and
amount. Experiments were performed at least in dupli-
cate.
3.12. Calculation
3.16. Searching metabolites of 7±9 in culture media and
cells
Mulliken charges at C-2' and C-3' of 1, 5 and 7±9
were calculated by the PM3 method using CAChe 3.1.,
Oxford Molecular Ltd., Oxford, 1997.
Four ¯asks containing 7 and cell suspension were
prepared by the same method as that for the quanti®-
cation. The cultures were incubated at 248 on a rotary
shaker at 90 rpm for 3 days, and ®ltrated. The pH of
the ®ltrate was adjusted to pH 3 with 1 N HCl, and
the ®ltrate was extracted four times with 200 ml of
EtOAc. The combined EtOAc layer was washed with
water, dried (Na2SO4), ®ltered and concentrated. The
residue (21 mg) was subjected to silica gel chromatog-
raphy (5 g) using a mixture of toluene and EtOAc as
eluant. 3'-Chloro-8'-hydroxy-ABA and 3'-chloropha-
seic acid of 7 were not found in the eluted fractions.
Recovered cells (22.1 g fresh weight) was soaked in 50
ml of MeOH for three days, and removed by ®ltration.
The ®ltrate was concentrated, and the residue was ana-
lyzed with HPLC in the same condition as that for the
quanti®cation, and silica gel TLC developed with tolu-
ene±EtOAc±HOAc (14 : 6 : 1). 3'-Chloro-8'-hydroxy-
ABA and 3'-chlorophaseic acid of 7 were not detected
with HPLC and TLC analyses. Compounds 8 and 9
were treated and analyzed by the same manner as that
for 7.
3.13. Bioassays
The four bioassays were performed as previously
reported (Todoroki et al., 1994). For the elongation
assays, the length of the second leaf sheath of rice
(Oryza sativa L. cv. Nihonbare) seedlings was
measured after incubation with the test solution in
continuous light at 308C for 7 days. For the germina-
tion assay, the number of germinated lettuce (Lactuca
sativa L. cv. Grand Rapids) seeds was counted after
incubation with the test solution at 258C for 48 h. For
the a-amylase assay, after incubating barley (Hordeum
vulgare L. cv. Himalaya) half-seeds without embryos
in the test solution at 308C for 48 h in the dark, the
absorbance of the test solution at 660 nm was
measured by the Somogyi±Nelson method. For the
stomatal opening assay, the width of the stomatal
aperture on the epidermal strip of spiderwort
(Tradescantia re¯exa Ra®n) was measured after incu-
bation with the test solution in continuous light at
258C for 3 h.
3.17. Preparation of 6 and 10
3.14. Maintenance of rice cells
A solution of 5 (0.4 mg) in 0.5 ml of H2O was
added to 70 ml of a suspension of 10 g of cells in a
¯ask so that the ®nal concentration was 20 mM.
Twelve ¯asks containing 5 and cell suspension were
prepared. The cultures were incubated at 248 on a
rotary shaker at 90 rpm for 24 h, and the cells were
removed by ®ltration. The pH of the ®ltrate was
Suspension cultures of rice (O. sativa L. cv.
Nipponkai) were maintained at 248 in continuous light
on a rotary shaker at 90 rpm in Linsmaier and Skoog
medium. Half of the cells were transferred into new
medium once a week.

S. Arai et al. / Phytochemistry 52 (1999) 1185±1193
1193
adjusted to 3 with 1 N HCl, and the ®ltrate was
extracted four times with 500 ml of EtOAc. The com-
bined EtOAc layer was washed with water, dried
(Na2SO4), ®ltered and concentrated. The residue (52
mg) was chromatographed on silica gel (5.2 g) with a
mixture of toluene±EtOAc as the eluant. Materials
eluted with 50% EtOAc were puri®ed with an ODS
HPLC column (YMC AQ-313, 6 i.d. Â 100 mm) by
eluting with 45% MeOH in 0.1% aqueous HOAc at
2.4 Hz, H-5'), 2.70 (1H, dd, J = 18.0 and 2.5 Hz, H-
3'), 2.80 (1H, d, J = 18.0 Hz, H-3'), 3.66 (1H, d, J =
7.6 Hz, H-8'), 3.94 (1H, dd, J = 7.6 and 2.9 Hz, H-
8'), 5.79 (1H, br. s, H-2), 6.45 (1H, d, J = 15.9 Hz, H-
5), 8.10 (1H, d, J = 15.9 Hz, H-4).
Acknowledgements
1
25
1.0 ml min
(MeOH; c 0.046); UV l
to give 0.61 mg of 6. ‰aŠ +2738
max
D
This study was supported by
a Grant-in-Aid
MeOH
1
nm ꢀlog e): 255 (4.34); H-
(Biocosmos program) from the Ministry of
Agriculture, Forestry and Fisheries of Japan (BCP-98-
I-A-2). We thank Professor Junichi Ueda of the
University of Osaka Prefecture for supplying lettuce
and barley seeds and the spiderwort, Dr. Hiroshi
Okumoto at this university for supplying rice seeds,
Professor Toshiaki Mitsui of the Niigata University
for supplying rice cells, and Kyowa Hakko Kogyo
Co., Ltd., Tokyo, Japan for the gift of (+)-ABA.
NMR (500 MHz, CD3OD): d 1.05 (3H, s, H-9'), 1.85
(3H, d, J = 3.4 Hz, H-7'), 2.02 (3H, s, H-6), 2.43 (1H,
d, J = 17.3 Hz, H-5'), 2.60 (1H, d, J = 17.3 and 4.7
Hz, H-5'), 3.62 (1H, d, J = 11.3 Hz, H-8'), 3.65 (1H,
d, J = 11.3 Hz, H-8'), 5.79 (1H, br. s, H-2), 6.11 (1H,
d, J = 16.1 Hz, H-5), 7.73 (1H, d, J = 16.1 Hz, H-4).
Compound 6 (0.32 mg) was dissolved in 0.7 ml of
0.1% NH3 aqueous soln, and left at 258 for 1 h. The
solution was concentrated, and the residue was puri®ed
with an ODS column under the same conditions as
26
used for 6 to a€ord 0.2 mg of 10. ‰aŠ
max
908 (MeOH; c
nm ꢀlog e): 255 (4.37); H-NMR (500
D
References
MeOH
1
0.021); UV l
MHz, CD3OD): d 1.03 (3H, s, H-9'), 1.32 (3H, d, J =
1.8 Hz, H-7'), 2.01 (3H, s, H-6), 2.38 (1H, dd, J =
16.8 and 6.6 Hz, H-5'), 2.79 (1H, dd, J = 16.8 and 2.8
Hz, H-5'), 3.55 (1H, d, J = 7.6 Hz, H-8'), 3.94 (1H,
dd, J = 7.6 and 2.8 Hz, H-8'), 5.02 (1H, d, J = 47.0
Hz, H-3'), 5.85 (1H, br. s, H-2), 6.29 (1H, d, J = 15.9
Hz, H-5), 8.00 (1H, d, J = 15.9 Hz, H-4).
Balsevich, J. J., Bishop, G. G.,
Phytochemistry, 44, 215.
& Banowetz, G. M. (1997).
Breitmaier, E., & Voetler, W. (1987). Carbon-13 NMR spectroscopy.
Weinheim: VHC Verlagsgesellschaft.
Davies, W. J., & Jones, H. G. (1991). Abscisic acid. Oxford: BIOS
Scienti®c Publishers.
Downs, A. J., & Adams, C. J. (1973). In J. C. Bailer, H. J. Emeleu,
R. Nyholm, & A. F. Trotman-Dickenson, Comprehensive inor-
ganic chemistry, vol. 2 (p. 1107). Oxford: Pergamon Press.
Hirai, N. (1999). In K. Mori, Comprehensive natural products chem-
istry, Vol. 8 (p. 72). Oxford: Elsevier.
3.18. Incubation conditions of 6 in the rice seedling
assay
Hirai, N., Fukui, H., & Koshimizu, K. (1978). Phytochemistry, 17,
1625.
An aqueous solution (2 ml) of 6 at 0.1 mM was
incubated under continuous light at 308C for 7 days.
Aliquots of 5 ml of the solution were analyzed every
day with an ODS column under the same conditions
as used for the preparation of 6. Compound 6 and its
2E-isomer, and 10 and its 2E-isomer were eluted at Rt
8.9, 6.5, 4.8, and 4.3 min, respectively.
McIntosh, J. M. (1971). Canadian Journal of Chemistry, 49, 3045.
Milborrow, B. V. (1984). Biochemical Journal, 220, 325.
Moriguchi, I., Kanada, Y., & Komatsu, K. (1976). Chemical and
Pharmaceutical Bulletin, 24, 1799.
Nakano, S., Todoroki, Y., Hirai, N.,
& Ohigashi, H. (1995).
Bioscience, Biotechnology and Biochemistry, 59, 1699.
Suzuki, H., Kondo, A., & Ogawa, T. (1985). Chemistry Letters,
1985, 411.
Todoroki, Y., Hirai, N.,
Biotechnology and Biochemistry, 58, 707.
& Koshimizu, K. (1994). Bioscience,
3.19. Preparation of 3
Todoroki, Y., Hirai, N., & Koshimizu, K. (1995a). Phytochemistry,
38, 561.
Three mg of 3 was prepared by hydrolysis of the b-
hydroxy-b-methylglutaryl ester of 8'-hydroxy-ABA iso-
lated from immature seeds of Robinia pseudacacia
(Hirai, Fukui & Koshimizu, 1978). 1H-NMR (500
MHz, CD3OD) of 3: d 1.01 (3H, s, H-9'), 1.21 (3H, s,
H-7'), 2.06 (3H, d, J = 0.9 Hz, H-6), 2.39 (1H, dd, J
= 18.0 and 2.5 Hz, H-5'), 2.47 (1H, dd, J = 18.0 and
Todoroki, Y., Hirai, N., & Koshimizu, K. (1995b). Phytochemistry,
40, 633.
Todoroki, Y., Hirai, N., & Ohigashi, H. (1995). Tetrahedron, 51,
6911.
Walker-Simmons, M. K., Holappa, L. D., Abrams, G. D.,
Abrams, S. R. (1997). Physiologia Plantarum, 100, 474.
&
Zou, J., Abrams, G. D., Barton, D. L., Taylor, D. C., Pomeroy, M.
K., & Abrams, S. R. (1995). Plant Physiology, 108, 563.