Abscinazole-F1, a conformationally restricted analogue of the plant growth retardant uniconazole and an inhibitor of ABA 8′-hydroxylase CYP707A with no growth-retardant effect

Article abstract of DOI:10.1016/j.bmc.2009.07.070

To develop a specific inhibitor of abscisic acid (ABA) 8′-hydroxylase, a key enzyme in the catabolism of ABA, a plant hormone involved in stress tolerance, seed dormancy, and other various physiological events, we designed and synthesized conformationally

Full text of DOI:10.1016/j.bmc.2009.07.070

Bioorganic & Medicinal Chemistry 17 (2009) 6620–6630
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Abscinazole-F1, a conformationally restricted analogue of the plant growth
retardant uniconazole and an inhibitor of ABA 80-hydroxylase CYP707A
with no growth-retardant effect
a
a
a
a
Yasushi Todoroki a, , Kyotaro Kobayashi , Minaho Shirakura , Hikaru Aoyama , Kokichi Takatori ,
Hataitip Nimitkeatkai b, Mei-Hong Jin a, Saori Hiramatsu a, Kotomi Ueno c, , Satoru Kondo b,
Masaharu Mizutani d,à, Nobuhiro Hirai e
*
a Department of Applied Biological Chemistry, Faculty of Agriculture, Shizuoka University, Shizuoka 422-8529, Japan
b Division of Bioresource Science, Graduate School of Horticulture, Chiba University, Matsudo 271-8510, Japan
c The United Graduate School of Agricultural Science, Gifu University, Gifu 501-1193, Japan
d Institute for Chemical Research, Kyoto University, Uji, Kyoto 611-0011, Japan
e Division of Environmental Science and Technology, Graduate School of Agriculture, Kyoto University, Kyoto 606-8501, Japan
a r t i c l e i n f o
a b s t r a c t
To develop a specific inhibitor of abscisic acid (ABA) 80-hydroxylase, a key enzyme in the catabolism of
ABA, a plant hormone involved in stress tolerance, seed dormancy, and other various physiological
events, we designed and synthesized conformationally restricted analogues of uniconazole (UNI), a
well-known plant growth retardant, which inhibits a biosynthetic enzyme (ent-kaurene oxidase) of gib-
berellin as well as ABA 80-hydroxylase. Although most of these analogues were less effective than UNI in
inhibition of ABA 80-hydroxylase and rice seedling growth, we found that a lactol-bridged analogue with
an imidazole is a potent inhibitor of ABA 80-hydroxylase but not of plant growth. This compound, absci-
Article history:
Received 30 June 2009
Revised 27 July 2009
Accepted 28 July 2009
Available online 3 August 2009
Keywords:
P450 inhibitors
Abscisic acid
Uniconazole
nazole-F1, induced drought tolerance in apple seedlings upon spray treatment with a 10 lM solution.
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
Azole-type inhibitors bind to P450 active sites by both coordi-
nating to the heme-iron atom and interacting with surrounding
protein residues. Because heme coordination is a common prop-
erty of azole-containing inhibitors, their affinity and specificity
for individual P450 enzymes depend on structural properties other
than the azole group. In the case of UNI, these are 4-chlorophenyl
and t-butyl-hydroxymethyl groups. UNI inhibits various plant
P450 enzymes whose native substrates are structurally quite dif-
ferent from each other: for example, ABA (sesquiterpene) for
CYP707A; ent-kaurene (diterpene) for CYP701A; and steroids for
brassinosteroid biosynthetic enzymes. Mammalian P450 enzymes
metabolizing xenobiotics have a large conformational flexibility
to adapt to substrates of different sizes.11 Because some plant
P450 species possess xenobiotic detoxification activity,12 they
may be highly flexible in conformation. On the other hand, ABA
80-hydroxylase has relatively high substrate specificity.13 If other
UNI-inhibiting P450 enzymes involved in biosynthesis and catabo-
lism of plant hormones also have high substrate specificity, even
though this has not yet been investigated in detail, explanations
other than enzyme flexibility may be required to explain the adap-
tation of UNI. From this standpoint, we have been probing why UNI
inhibits various plant P450 enzymes. The first reason must be that
UNI is small enough to embed itself into various substrate-binding
pockets. If UNI just invades the pocket, even if its geometry does
Uniconazole-P, S-(+)-uniconazole [S-(+)-E-1-(4-chlorophenyl)-
4,4-dimethyl-2-(1,2,4-triazo-1-yl)-1-penten-3-ol, UNI], is an
azole-containing cytochrome P450 inhibitor developed as a plant
growth retardant in the 1980s.1,2 UNI has since been used as a
plant growth regulator in agriculture and horticulture. The main
site of action of UNI is suggested to be ent-kaurene oxidase
(CYP701A), which catalyzes the three-step oxidation of ent-kau-
rene to ent-kaurenoic acid,3 biosynthetic precursors of the plant
hormone gibberellin (GA). This has prompted researchers to use
UNI as a chemical tool inhibiting GA biosynthesis. However, UNI
also inhibits brassinosteroid biosynthesis4,5 and alters the level of
other plant hormones, such as auxins, cytokinins, ethylene, and ab-
scisic acid (ABA).6 Recently, Kitahata et al.7 and Saito et al.8 re-
vealed that UNI strongly inhibits ABA 80-hydroxylase
(CYP707A9,10), a key enzyme in ABA catabolism (Fig. 1).
* Corresponding author. Tel./fax: +81 54 238 4871.
E-mail address: aytodor@agr.shizuoka.ac.jp (Y. Todoroki).
Present address: Graduate School of Agricultural and Life Sciences, The University
of Tokyo, Tokyo 113-8657, Japan.
à
Present address: Graduate School of Agricultural Science, Kobe University, Kobe
657-8501, Japan.
0968-0896/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2009.07.070

Y. Todoroki et al. / Bioorg. Med. Chem. 17 (2009) 6620–6630
6621
Figure 1. Uniconazole (UNI) is a potent inhibitor of ABA 80-hydroxylase and ent-kaurene oxidase.
not completely coincide with that of the pocket, the structural
plasticity of a P450 active site can allow it to fit into the pocket well
enough to make the aza nitrogen interact with the heme iron. Thus,
enlargement may be an effective approach to increase enzyme
specificity. However, this approach will be reported elsewhere.
In a previous study, we found that the conformer preference of
UNI analogues partially contributes to affinity for the active site of
ABA 80-hydroxylase.14 This is the second possible reason that the
low specific inhibition of UNI for plant P450 enzymes may depend
on its conformational flexibility. We hypothesized that conforma-
tional restriction of UNI may have a drastic influence on its affinity
to ABA 80-hydroxylase or other plant P450 enzymes that have nar-
row ligand specificity. Restriction of conformation is a strategy that
has been widely used in drug design to develop potent and selec-
tive enzyme inhibitors.15 Thus, we tested whether this strategy is
applicable for a plant P450 enzyme in developing a specific inhib-
itor of ABA 80-hydroxylase.
corresponds to a pocket of various P450 enzymes, a conformation-
ally constrained UNI derivative should be more specific than UNI.
We searched and found a known UNI derivative, 1,16 that is con-
formationally restricted (Fig. 3). Compound 1 has a structure
bridged with a carbonyl group between C500 in the triazole ring
and the oxygen of the 3-hydroxy in UNI; therefore, it is expected
to be restricted in a similar conformation to the B-type form of
UNI, which is the crystalline structure in MeOH–H2O, and is theo-
retically estimated to be the most stable conformer in the aqueous
phase (Fig. 4).14 Compound 1, whose C3 stereochemistry was not
shown in the literature, was described as a fungicide and plant
growth retardant. However, its conformational structure–activity
relationship and enzyme selectivity have never been investigated.
Thus, we first synthesized compound 1 and examined its inhibitory
effect on ABA 80-hydroxylase and its activity as a plant growth
retardant. The results showed that compound 1 functioned as a
plant retardant as effectively as UNI, whereas it did not inhibit
ABA 80-hydroxylase at all, unlike UNI. This suggested that a novel,
highly specific P450 inhibitor can be produced by conformational
restriction. Restriction in a different manner from 1 might produce
a selective inhibitor of ABA 80-hydroxylase that does not inhibit
other UNI-inhibited P450s, and thus never cause plant growth inhi-
bition, in contrast to compound 1. ABA 80-hydroxylase is a key en-
Because UNI has t-butyl and hydroxy groups at the 3-position,
the rotation of the C2–C3 bond has a great effect on its molecular
shape and electrostatic distribution. C2–N100 bond rotation affects
the orientation of the aza-nitrogen, which plays an important role
in binding to the P450 heme iron. In a previous study,14 we uncov-
ered four major conformers A–D of UNI (Fig. 2). If each conformer
Figure 2. Four conformers A–D generated by bond rotations of C2–N100 and C2–C3 in UNI.

6622
Y. Todoroki et al. / Bioorg. Med. Chem. 17 (2009) 6620–6630
Figure 3. Conformationally restricted analogues of UNI.
Figure 4. 3D structures of S-UNI (crystalline and theoretical) and S-1 (theoretical).
zyme in the catabolism of ABA, a plant hormone involved in stress
tolerance, seed dormancy, and other physiological events.17 Thus, a
specific inhibitor of ABA 80-hydroxylase is promising not only as a
chemical probe for the mechanism of ABA action, but also because
of its potential use in agriculture and horticulture. We synthesized
and tested various conformationally restricted analogues (Fig. 3) to
identify novel inhibitors of ABA 80-hydroxylase with no growth-
retardant effect.
reported method16 (Scheme 1). UNI (racemic, S, or R) was treated
with trichloromethyl chloroformate in pyridine to give 1 (racemic,
S, or R). Compound 1 was quite sensitive to light and was partially
converted into the Z-isomer by photoisomerization. Although we
used a sample stored in the dark as soon as purification was com-
plete, it contained ca. 10% Z-isomer in enzymatic and biological as-
says. Because the Z-isomer of UNI is much less effective than the
corresponding E-isomer in plant growth and ABA 80-hydroxylase
inhibition,3,7 the activity of 1 will never be overestimated using
the Z-isomer-contaminated sample.
2. Results and discussion
The inhibitory activity of 1 against ABA 80-hydroxylase was
examined using recombinant CYP707A3 expressed by Escherichia
coli. The activity was evaluated based on the decrease in the en-
zyme product, phaseic acid, caused by addition of a test compound
at a concentration two times higher than the substrate S-ABA
2.1. Preparation of compound 1 and its properties
Both racemic and optically pure compound 1 were synthesized,
respectively, from racemic and optically pure UNI according to a

Y. Todoroki et al. / Bioorg. Med. Chem. 17 (2009) 6620–6630
6623
enzyme selectivity. The triazole types 2 and 3 were prepared from
1 according to Scheme 2. The reduction of 1 with LiAlH4 gave 2 and
the ring-opened diol 10. The reduction of 2 with triethylsilane and
trifluoroborane etherate gave the ether-bridged analogue 3. The
imidazole types 4–6 were synthesized from the imidazole ana-
logue of UNI (11)14 according to Scheme 3. Treatment of 11 with
trichloromethyl chloroformate gave the lactone-bridged analogue
4. This compound was reduced with LiAlH4 to give the lactol-
bridged analogue 5 and the ring-opened diol 12. Each stereoisomer
of 5 at C3 was prepared from optically pure 11. However, the ster-
eoisomers derived from the lactol moiety were unable to be re-
solved because of rapid interconversion. 1H NMR analysis
revealed that the diastereomeric ratio was 3:1 in methanol and
5:1 in chloroform, although the absolute configuration of the major
diastereomer could not be determined. The intramolecular dehy-
drated condensation of 12 with ZnCl2 on heating gave mainly 6Z
with E to Z isomerization. Photoisomerization of 6Z gave the
ether-bridged analogue 6.
Scheme 1. Synthesis of S-1. Reagents and conditions: ClCOOCCl3, pyridine, dry
CH2Cl2, 0 °C, 98%. R-1 and racemic 1 were synthesized by the same scheme from R-
UNI and racemic UNI, respectively.
(5 lM). Inhibitory activity of 1 was much weaker than that of S-
UNI (Table 1). The plant growth-retardant effect was examined
using rice seedlings. Compound S-1 exhibited strong activity,
although it was weaker than S-UNI (Table 1); the IC50 value was
2.2.2. A-type restricted analogues
1.3 lM. On the other hand, R-1 was less effective than the S-iso-
A similar bridging strategy to that in compound 1 cannot be
used for A-type restriction of UNI because the aza nitrogen (N200)
of the triazole ring cannot be attached to a bridging moiety without
breaking the triazole ring. Thus, we used only the imidazole ring
for constructing the A-type restricted compound. Our previous re-
port suggests that the replacement of triazole by imidazole has no
significant effect on inhibitory activity on ABA 80-hydroxylase.14
The A-type restricted analogues 7 and 9 were synthesized from
commercially available 5-hydroxymethylimidazole according to
Scheme 4. A coupling reaction of the TBS-protected 5-hydroxyme-
thylimidazole 13 and 1-bromo-3,3-dimethylbutan-2-one gave
compound 14. Aldol condensation of 14 and 4-chlorobenzaldehyde
gave the ketone 15 as an E/Z mixture (1:1), which was UV-irradi-
ated (365 nm) to increase the population of the E-isomer by photo-
isomerization. The E-ketone 15E was reduced to give 16, which
was deprotected to afford the diol 17. MnO2 oxidation of 17 gave
the lactone-bridged analogue 7 and the aldehyde 18, whereas
ZnCl2 treatment of 17 gave the ether-bridged analogue 9. This ana-
logue was optically resolved by HPLC using chiral stationary
phases. The absolute configuration of optically pure 9 was deter-
mined based on Cotton effects in the circular dichroism spectrum
mer; this is the same trend observed for UNI.
The 3D structure of 1, which was theoretically estimated, is very
similar to that of the B-type form of UNI (Fig. 4). This suggested
that conformational freezing of UNI into the B-type form results
in a decrease in affinity for ABA 80-hydroxylase with little signifi-
cant effect on P450 enzymes involved mainly in growth regulation,
although the possibility cannot be excluded that the lactone moi-
ety involved in bridging caused the elevated selectivity, indepen-
dent of conformational freezing. Thus, we designed various UNI
analogues (2–9) with conformations (A and B) restricted by differ-
ent bridge moieties to identify a selective inhibitor of ABA 80-
hydroxylase (Fig. 3).
2.2. Design and synthesis of conformationally restricted UNI
analogues
2.2.1. B-type restricted analogues
We designed B-type restricted analogues with triazole (2 and 3,
in addition to 1) and imidazole rings (4, 5, and 6). A lactone, lactol,
and cyclic ether were selected for bridging between the azole ring
and the oxygen at C3, to compare the effect of these moieties on
Table 1
Inhibitory activity of conformationally restricted analogues of UNI against recombinant CYP707A3 and rice seedling elongation
Compound
Structural propertiesa
Bridge
CYP707A3
Rice
Azole
Conf.
Inhibitionb (%)
100
KI (nM)
IC50 (lM)
c
S-UNI
R-UNI
Trz
Trz
A/B
A/B
10
1450
0.18
7.8
79
1
1
Trz
Trz
Trz
Trz
Lon
Lon
Lon
Lol
Ether
Lon
Lol
Lol
Lol
Ether
Lon
Ether
Ether
Ether
B
B
B
B
B
B
B
B
B
B
A
A
A
A
14
10
27 24
69
31 10
35
91
96
95
13
52
5
7
NMd
NM
NM
NM
NM
NM
NM
970
420
NM
NM
NM
160
520
2.3
1.3
46
3.2
78
S-1
R-1
2
3
4
7
Trz
Imz
Imz
Imz
Imz
Imz
Imz
Imz
Imz
Imz
8
5
4
7
6
2
NIe
NI
NI
NI
28
58
9.0
8.2
12
5 (Abz-F1)
3S-5
3R-5
6
7
9
100
100
87
S-9
R-9
3
a
Azole, type of azole (Trz, triazole; Imz, imidazole); bridge, bridging moiety (Lon, lactone; Lol, lactol; Ether, cyclic ether); conf., type of conformation (see Figs. 2 and 3).
b
c
Inhibition ratio of compounds (10 l lM) (average of at least three sets of experiments).
M) in the 80-hydroxylation of ABA (5
The concentration for 50% inhibition (average of at least two sets of experiments).
Not measured.
No significant inhibition even at 100 lM (max concentration tested).
d
e

6624
Y. Todoroki et al. / Bioorg. Med. Chem. 17 (2009) 6620–6630
unconverted to the closed form 8 in both protic and aprotic sol-
vents. The opened form 18 seems to be thermodynamically more
stable than the closed form 8. On the other hand, the lactol-bridged
analogues 2 and 5 were stable enough to remain unconverted to
the ring-opened form. The thermodynamic instability of the lac-
tol-bridged A-type analogues contrasts with that of B-type ana-
logues, although the reason is not clear.
2.3. Inhibitory potency against ABA 80-hydroxylase
The inhibitory activity of conformationally restricted analogues
1–7 and 9 against ABA 80-hydroxylase was examined using Arabid-
opsis CYP707A3 expressed by E. coli (Table 1). First of all, the activ-
ity of all the analogues was screened at twice the concentration of
the substrate (S-ABA). For the A-type, the cyclic ether-bridged ana-
logue 9 showed significant inhibition, whereas for the B-type, only
the lactol-bridged analogues 2 and 5 exhibited more than 50%
inhibition; 5 was an especially strong inhibitor, as was 9. Other
Scheme 2. Synthesis of 2 and 3. Reagents and conditions: (i) LiAlH4, dry Et2O, 0 °C,
47% (2); (ii) Et3SiH, BF3ꢁEt2O, rt, 13%.
(see Supplementary data): (+)-9 is S-9, and (ꢀ)-9 is R-9. The alde-
hyde 18 is the tautomeric isomer (ring-opened form) of the lactol-
bridged analogue 8 (ring-closed form), which was not identified in
the reaction products. The isolated 18 was stable enough to remain
Scheme 3. Synthesis of 4, 5, and 6. Reagents and conditions: (i) ClCOOCCl3, pyridine, dry CH2Cl2, 0 °C, 94%; (ii) LiAlH4, dry Et2O, 0 °C, 38% (5), 24% (18); (iii) ZnCl2, dry 1,2-
dichloroethane, 100 °C, 51%; (iv) UV 365 nm, EtOAc, rt, 32%.
Scheme 4. Synthesis of 7 and 9. Reagents and conditions: (i) see Ref. 16; (ii) NaH, 1-bromo-3,3-dimethyl-2-butanone, dry Et2O, 0 °C, 9%; (iii) p-chlorobenzaldehyde, K2CO3,
Ac2O, 100 °C, 62%; (iv) UV 365 nm, EtOAc, rt, 63%; (v) NaBH4, MeOH, 0 °C to rt, 49%; (vi) 1 M HCl, EtOH–H2O, rt, 89%; (vii) MnO2, acetone, rt, 35% (7); (viii) ZnCl2, dry CH2Cl2, rt,
27%.

Y. Todoroki et al. / Bioorg. Med. Chem. 17 (2009) 6620–6630
6625
analogues of both types were poor inhibitors. The cyclic ether-
bridged compounds (3, 6, and 9), which are the most simple con-
formationally restricted UNI analogues, have no additional func-
tional group compared to the 3-methoxy-UNI. 3-Methoxy-UNI is
as effective as S-UNI.14 This means that the effect of conformational
restriction is dominant in the interaction with the active site.
Therefore, the difference in CYP707A3 inhibition between the A-
type (9) and B-type (3 and 6) suggests that the CYP707A3 active
site prefers the A-type conformation of S-UNI. On the other hand,
for the lactol- and lactone-bridged analogues, this conformational
preference was not true, probably because the additional hydroxy
and carbonyl groups were advantageous or disadvantageous
according to the orientation in the active site. The potent activity
of B-type analogues 2 and 5 indicates that the lactol moiety of
the B-type analogue has a significant role in binding to compensate
for the conformational disadvantage.
The most potent analogues 5 and 9 were examined in greater
detail for each stereoisomer at C3. Kinetic analysis revealed that
these analogues function as competitive inhibitors (Fig. 5); how-
ever, the affinity to the enzyme active site was estimated to be less
than one-tenth that of S-UNI on the basis of the KI values. Interest-
ingly, 3R-5 was slightly more effective than the 3S-enantiomer
(KI = 420 and 970 nM, respectively), unlike the case of UNI. Ana-
logue 5 may fit into the active site of ABA 80-hydroxylase in quite
a different manner from UNI, because of its B-type conformational
rigidity and lactol moiety. We cannot discuss this in greater detail
due to the lack of structural characterization of ABA 80-
hydroxylase.
Figure 6. Inhibitory effect of 3S- and 3R-5 and S-UNI on rice seedling growth.
imidazole, 4 and 5, showed no significant inhibition even at
100 M (Fig. 6). Because the corresponding B-type analogues with
l
a triazole, 1 and 2, acted as potent retardants, the 200-aza nitrogen
in the azole ring of the lactone- and lactol-bridged B-type ana-
logues may be involved in binding to the active site. These results
suggest that the A/B conformation does not affect primarily the
binding to the active site of P450 enzymes involved in rice seedling
growth that is represented by CYP701A. Because plant phenotypes
can be affected by many factors other than the direct action of an
enzyme inhibitor to the target, including uptake, metabolism, and
unexpected side effects, we cannot directly discuss the structure–
activity relationships for a specific enzyme, for example, CYP701A,
on the basis of findings in this bioassay. Nevertheless, the absence
of an effect of 4 and 5 on rice seedling growth indicates that these
compounds do not inhibit enzymes involved in growth under our
experimental conditions. In future studies, we plan to examine
the inhibitory activity of these analogues using recombinant
CYP701A and other plant P450 enzymes.
2.4. Potency as a plant retardant
UNI, especially the S-enantiomer, functions as a highly potent
plant retardant, mainly because of its inhibition of ent-kaurene oxi-
dase (CYP701A), a biosynthetic enzyme of GA.3 In fact, S-UNI
strongly inhibited elongation of rice (cultivar Nipponbare) seed-
lings even at 300 nM in our assay conditions (Fig. 6). In the same
bioassay, the inhibitory activity of conformationally restricted ana-
logues of UNI declined by at least a factor of 10 compared to S-UNI
on the basis of the IC50 values (Table 1). In an early stage of this
study, the potent retardant activity of 1 implied that the B-type
analogues function as growth inhibitors, whereas the A-type ana-
logues never do. However, the A-type analogues 7 and 9 were ac-
tive; in particular, the simplest, ether-bridged A-type analogue 9
was more potent than the corresponding B-types 3 and 6. More-
over, the lactone- and lactol-bridged B-type analogues with an
2.5. Drought tolerance induced by abscinazole-F1 (Abz-F1) (5)
In enzymatic and biological assays, all the analogues were less
effective than UNI to a greater or lesser extent. This suggests that
the rigid conformation confers a basic disadvantage in binding to
ABA 80-hydroxylase and P450 enzymes involved in plant growth,
probably because of a decrease in induced fit. However, because
the strength of this effect was different for inhibition of ABA 80-
hydroxylase and rice seedling growth, we identified analogue 5,
which we named abscinazole-F1 (Abz-F1), as a new azole-contain-
ing inhibitor of ABA 80-hydroxylase with no inhibition of plant
growth. Abz-F1 was too stable to be converted to the Z-isomer
and decomposed in the usual conditions in the laboratory.
To determine whether in vivo inhibition of Abz-F1 against ABA
80-hydroxylase is enough to intensify ABA-induced physiological
responses, we tested the effect of Abz-F1 on stomatal aperture
and drought tolerance. We tested on 90-day-old apple seedlings
by spraying them with an aqueous solution containing the 3R iso-
mer of Abz-F1 at concentrations of 10, 50, and 100 lM. Application
of 3R-Abz-F1 before dehydration induced stomatal closure during
dehydration (Fig. 7) and drought tolerance (Fig. 8) as effectively
Figure 5. Competitive inhibition of ABA 80-hydroxylase by 3R-5. Assays contained
S-ABA (d) or S-ABA and 0.5 lM 3R-5 (s). The inset is a double-reciprocal plot for
the same data. The same mode of inhibition was observed for 3S-5 and S- and R-9
(data not shown).
as that of S-UNI at all the tested concentrations. In the 100 lM S-

6626
Y. Todoroki et al. / Bioorg. Med. Chem. 17 (2009) 6620–6630
UNI treatment, leaf browning was observed. This side effect, which
may have been caused by inhibition of other enzymes including
P450s, was also observed to a minor extent following 100
3R-Abz-F1 treatment. Figure 9 shows that 10 M 3R-Abz-F1 treat-
l
M
l
ment induced slightly stronger drought tolerance in apple seed-
lings than S-UNI treatment at the same concentration. This is
contrary to the finding that the in vitro inhibitory activity of S-
UNI for ABA 80-hydroxylase was much higher than that of 3R-
Abz-F1 on the basis of KI values for the recombinant Arabidopsis
enzyme (Table 1). In addition to a difference between the recombi-
nant Arabidopsis and endogenous apple enzymes, this inconsis-
tency may be explained by the phenotypes resulting after
treatment with an enzyme inhibitor being affected not only by
inhibitory activity against the target enzyme, but also by other fac-
tors including uptake, stability, metabolism, and unexpected side
effects. The mechanism of action of these azole compounds should
be investigated in detail at the enzymatic, cellular, and organ lev-
els. Nevertheless, the present findings at least show that Abz-F1
is an inhibitor of ABA 80-hydroxylase with no growth-retardant ef-
fect, unlike UNI.
Figure 7. Effect of 3R-Abz-F1 (5) and S-UNI on stomatal aperture of apple seedlings:
control (d), 10
lM 3R-Abz-F1 (s), and 10
lM S-UNI (.). Similar changes were
observed at 50 and 100
lM (data not shown). Dashed line shows changes in water
potential of vermiculite in which apple seedlings were cultivated.
3. Conclusions
We focused on conformational flexibility of UNI to identify a
specific inhibitor of ABA 80-hydroxylase, and synthesized conform-
ationally restricted UNI analogues that are bridged between the
azole ring and hydroxy group at the C3 position. Most of these ana-
logues showed a decrease in inhibitory activity on ABA 80-hydrox-
ylase and growth of rice seedlings. This may be because the
decrease in conformational flexibility results in poor binding to
the P450 active site. Because the magnitude of this negative effect
must differ in different enzymes, the inhibition spectrum of con-
formationally restricted UNI analogues in plant P450 enzymes
should differ from that of UNI. This may be why we found Abz-
F1, a new azole-containing inhibitor of ABA 80-hydroxylase with
no inhibition of plant growth. Abz-F1 is the most specific inhibitor
against ABA 80-hydroxylase, although it is not the strongest on the
basis of in vitro enzyme assays.
4. Experimental
4.1. General
S-(+)- and R-(ꢀ)-uniconazole were prepared by the chiral HPLC
purification of uniconazole-P, which was purchased from Wako
Pure Chemical Industries, Ltd, Osaka, Japan.18 (+)-ABA was a gift
from Toray Industries Inc., Tokyo, Japan. 1H NMR spectra were re-
corded with tetramethylsilane as the internal standard using a
Figure 8. Drought tolerance of apple seedlings treated with (a) 3R-Abz-F1 (5) and
(b) S-UNI.
Figure 9. Drought tolerance of apple seedlings treated with distilled water (left),
10 lM of 3R-Abz-F1 (5) (center), and 10 lM of S-UNI (right).

Y. Todoroki et al. / Bioorg. Med. Chem. 17 (2009) 6620–6630
6627
JEOL JNM-EX270 (270 MHz) and JNM-LA500 (500 MHz) NMR spec-
trometer. 13C NMR and 2D-correlation NMR experiments were re-
corded using a JNM-LA500 (500 MHz) NMR spectrometer. High
resolution mass spectra were obtained with a JEOL JMS-T100LC
‘AccuTOF’. Optical rotations were recorded with a Jasco DIP-1000
digital polarimeter. Circular dichroism spectra were recorded with
a Jasco J-805 spectrophotometer. Column chromatography was
performed on silica gel (Wakogel C-200).
phenyl protons), 8.05 (1H, s, H-300); UV kmax (MeOH) nm (
e
): 270
(17,800); HRMS (ESI-TOF, positive mode): calcd for C16H18ClN3O2-
Na [M+Na]+ 342.0985, found 342.0982. Compound 10: 1H NMR
(270 MHz, CDCl3): d 0.66 (9H, s, t-butyl), 4.50 (1H, d, J = 8.9 Hz,
H-3), 4.78 (1H, d, J = 8.9 Hz, HO-3), 4.83 (1H, d, J = 13.8 Hz, –H2C-
500), 4.98 (1H, d, J = 13.8 Hz, –H2C-500), 6.98 (1H, s, H-1), 7.39 (4H,
m, 4-Cl-phenyl), 7.97 (1H, s, H-300); UV kmax (MeOH) nm (
e): 250
(13,400); HRMS (ESI-TOF, positive mode): calcd for C16H20ClN3O2-
Na [M+Na]+ 344.1142, found 344.1143.
4.2. Synthesis of optically pure (E)-1-(4-chlorophenyl)-4,4-
dimethyl-2-(1H-1,2,4-triazol-1-yl)pent-1-en-3-ol (1)
4.3.2. (E)-6-tert-Butyl-5-(4-chlorobenzylidene)-6,8-dihydro-
5H-[1,2,4]triazolo[5,1-c][1,4]oxazine (3)
S- and R-1 were synthesized according to the method of Saito
To a stirred solution of 2 (30 mg, 93
was added triethylsilane (100 L) and boron trifluoride diethyl
etherate (100 L) at room temperature. The mixture was stirred
lmol) in dry CH2Cl2 (2 mL)
et al.16 To a stirred solution of S-(+)-uniconazole (7 mg, 24.0
l
mol)
L) was added trichlo-
mol) at 0 °C. The mixture
l
in dry CH2Cl2 (0.5 mL) and dry pyridine (50
romethyl chloroformate (80 mg, 404
l
l
l
for 12 d at room temperature. After quenched with satd NaHCO3,
the resulting mixture was extracted with EtOAc (20 mL ꢂ 3). The or-
ganic layer was washed with H2O, dried over Na2SO4, and concen-
trated in vacuo. The residual oil was purified by column
chromatography on silica gel with 10% EtOAc in hexane and then
further purified by reverse-phase HPLC (YMC-Pack ODS-AQ,
150 ꢂ 20 mm, 80% MeOH, 4 mL minꢀ1, 254 nm) to give 3 (3.7 mg,
was stirred for 2 h at 0 °C. After being quenched with H2O at the
same temperature, the resulting mixture was extracted with EtOAc
(20 mL ꢂ 3). The organic layer was washed with H2O, dried over
Na2SO4, and concentrated in vacuo. The residual oil was purified
by silica gel column chromatography with 20% EtOAc in hexane
to obtain S-1 (7.5 mg, 23.6
(5.7 mg, 17.9
mol) was prepared from R-(ꢀ)-uniconazole (6 mg
20.5 mol) in 87% yield in the similar manner to the preparation
of S-1. Optical purity of each enantiomer was more than 99% based
lmol, 98%) as colorless oil. R-1
l
12 lmol, 13%) as colorless crystalline solid and its Z-isomer
l
(1.3 mg). Compound 3: 1H NMR (270 MHz, CDCl3): d 0.75 (9H, s, t-
butyl), 4.94 (1H, s, H-3), 5.06 (1H, d, J = 16.8 Hz, –H2C-500), 5.18
(1H, d, J = 16.8 Hz, –H2C-500), 7.33 (4H, m, 4-Cl-phenyl), 7.42 (1H, s,
on HPLC in
a
chiral column (Daicel Chiralpak AD-H,
250 mm ꢂ 4.6 mm, 10% 2-propanol in hexane, 1 mL minꢀ1
,
H-1), 8.00 (1H, s, H-300); UV kmax (MeOH) nm (
e): 269 (11,800), 213
254 nm). Each enantiomer contains the corresponding Z-isomer
(ca. 15%, determined by 1H NMR) which could not be removed
by purification. S-1: 1H NMR (270 MHz, CD3OD): d 0.75 (9H, s, t-bu-
tyl), 5.84 (1H, d, J = 0.7 Hz, H-3), 7.51 (4H, m, 4-Cl-phenyl), 7.67
(6400); HRMS (ESI-TOF, positive mode): calcd for C16H19ClN3O
[M+H]+ 304.1217, found 304.1213. Z-Isomer of 3: 1H NMR
(270 MHz, CDCl3): d 0.92 (9H, s, t-butyl), 4.22 (1H, s, H-3), 4.98
(1H, d, J = 16.2 Hz, –H2C-500), 5.15 (1H, d, J = 16.2 Hz, –H2C-500), 6.19
(1H, s, H-1), 7.33 (4H, m, 4-Cl-phenyl), 7.91 (1H, s, H-300); UV kmax
(1H, s, H-1), 8.36 (1H, s, H-300); UV kmax (MeOH) nm (
e): 297
(10,500) and shoulders at around 224 and 246 nm; HRMS (ESI-
(MeOH) nm (e): 274 (8600), 219 (7200); HRMS (ESI-TOF, positive
TOF, positive mode): calcd for C16H16ClN3O2Na [M+Na]+
mode): calcd for C16H19ClN3O [M+H]+ 304.1217, found 304.1214.
340.0829, found 340.0827; ½a D26
ꢃ
ꢀ18.1 (MeOH, c 0.555). R-1: 1H
NMR (270 MHz, CD3OD): d 0.74 (9H, s, t-butyl), 5.84 (1H, d,
J = 0.7 Hz, H-3), 7.51 (4H, m, 4-Cl-phenyl), 7.67 (1H, s, H-1), 8.37
4.4. Synthesis of 4, 5, and 6
(1H, s, H-300); UV kmax (MeOH) nm (
e): 297 (10,300) and shoulders
4.4.1. (E)-6-tert-Butyl-5-(4-chlorobenzylidene)-5H-
at around 223 and 247 nm; HRMS (ESI-TOF, positive mode): calcd
imidazo[2,1-c][1,4]oxazin-8(6H)-one (4)
for C16H16ClN3O2Na [M+Na]+ 340.0829, found 340.0827; ½a 2D6
ꢃ
+16.3
To a stirred solution of compound 1114 (0.556 g, 1.9 mmol) in
dry MeCN (250 mL) under Ar was added pyridine (1.47 g, 19 mmol)
and trichloromethyl chloroformate (1.28 g, 7.15 mmol) at 0 °C. The
mixture was stirred for 4 h at the same temperature. After
quenched with H2O, MeCN was removed from the mixture in va-
cuo before extracted with EtOAc (70 mL ꢂ 3). The organic layer
was washed with brine, dried over Na2SO4, and concentrated in va-
cuo. The residual oil was purified by column chromatography on
silica gel with 30% EtOAc in hexane to give 4 (0.567 g, 1.79 mmol,
94%) as pale yellow oil. Compound 4: 1H NMR (500 MHz, CD3OD): d
0.73 (9H, s, t-butyl), 5.58 (1H, s, H-3), 7.33 (1H, s, H-1), 7.43 (1H, s,
triazole), 7.47 (4H, m, 4-Cl-phenyl), 7.98 (1H, s, triazole); UV kmax
(MeOH, c 0.555). Racemic 1 was prepared in the similar manner to
the preparation of optically pure 1. The spectral data of racemic 1
agreed with that of optically pure 1.
4.3. Synthesis of 2 and 3
4.3.1. (E)-6-tert-Butyl-5-(4-chlorobenzylidene)-6,8-dihydro-5H-
[1,2,4]triazolo[5,1-c][1,4]oxazin-8-ol (2)
To a stirred solution of 1 (103 mg, 324
lmol) in dry Et2O
(20 mL) was added LiAlH4 (29 mg, 763 mol) at 0 °C. The mixture
l
was stirred for 1 h at the same temperature, then for 1.5 h at room
temperature before added LiAlH4 (4 mg). The mixture was stirred
for 2 h at room temperature. After quenched with 1 M HCl, the
resulting mixture was extracted with EtOAc (80 mL ꢂ 3). The or-
ganic layer was washed with H2O, dried over Na2SO4, and concen-
trated in vacuo. The residual oil was purified by column
chromatography on silica gel with 30% EtOAc in hexane to give 2
(MeOH) nm (e): 259 (13,300); HRMS (ESI-TOF, positive mode):
calcd for C17H17ClN2O2Na [M+Na]+ 339.0876, found 339.0874.
4.4.2. (E)-6-tert-Butyl-5-(4-chlorobenzylidene)-6,8-dihydro-
5H-imidazo[2,1-c][1,4]oxazin-8-ol (5)
To a stirred solution of 4 (31.6 mg, 0.100 mmol) in dry Et2O
(0.5 mL) was added LiAlH4 (16 mg, 0.34 mmol) at 0 °C. The mixture
was stirred for 1 h at the same temperature. After quenched with
1 M HCl, the resulting mixture was extracted with EtOAc
(20 mL ꢂ 4). The organic layer was washed with brine, dried over
Na2SO4, and concentrated in vacuo. The residual oil was purified
by column chromatography on silica gel with 30% acetone in hexane
to give 5 (12.3 mg, 0.038 mmol, 38%) which was a diastereomeric
mixture (3:1 in CDCl3, and 5:1 in CD3OD, determined by 1H NMR)
as pale yellow powder and the lactone ring-opened diol 12
(48 mg, 151
mined by integrating H-3 singlets in the 1H NMR spectrum) and 10
(21 mg, 65
mol, 20%) as colorless oils. Compound 2: 1H NMR
lmol, 47%) as a diastereomeric mixture (59:41, deter-
l
(270 MHz, CDCl3): major diastereomer: d 0.70 (9H, s, t-butyl),
5.17 (1H, s, H-3), 6.51 (1H, s, –(HO)HC-500), 7.36 (m, 4-Cl-phenyl,
overlapped with H-1), 7.41 (s, H-1, overlapped with 4-Cl-phenyl
protons), 8.03 (1H, s, H-300); minor diastereomer: d 0.74 (9H, s, t-
butyl), 5.12 (1H, s, H-3), 6.43 (1H, s, –(HO)HC-500), 7.36 (m, 4-Cl-
phenyl, overlapped with H-1), 7.37 (s, H-1, overlapped with 4-Cl-

6628
Y. Todoroki et al. / Bioorg. Med. Chem. 17 (2009) 6620–6630
(7.6 mg, 0.023 mmol, 24%) as pale yellow powder. Compound 5: 1H
NMR (500 MHz, CD3OD): major diastereomer: d 0.63 (9H, s, t-butyl),
5.06 (1H, s, H-3), 6.16 (1H, s, –(HO)HC-200), 6.93 (1H, s, H-1), 7.08 (1H,
s, triazole), 7.38 (4H, s, 4-Cl-phenyl), 7.41 (1H, s, triazole); minor dia-
stereomer: d 0.71 (9H, s, t-butyl), 4.94 (1H, s, H-3), 6.07 (1H, s, –
(HO)HC-200), 6.98 (1H, s, H-1), 7.12 (1H, s, triazole), 7.38 (4H, s, 4-
dazole hydrochloride (5.0 g, 37.2 mmol) in 75% yield according to
the method of Amino et al.19 1H NMR spectrum of 13 was coincident
with that reported by Amino et al.19
4.5.2. 1-(5-((tert-Butyldimethylsilyloxy)methyl)-1H-imidazol-
1-yl)-3,3-dimethylbutan-2-one (14)
Cl-phenyl), 7.60 (1H, s, triazole); UV kmax (MeOH) nm (
e
): 264
To a stirred solution of 13 (5.88 g, 27.9 mmol) in dry Et2O
(110 mL) was added sodium hydride (60% in oil, 0.75 g, 31.2 mmol)
at 0 °C. The mixture was stirred for 0.5 h at the same temperature
before being added dropwise 1-bromo-3,3-dimethyl-2-butanone
(5.45 g, 30.5 mmol). The mixture was stirred for 1 h at 0 °C. After
quenched with H2O, the resulting mixture was extracted with
EtOAc (100 mL ꢂ 3). The organic layer was washed with H2O, dried
over Na2SO4, and concentrated in vacuo. The residual oil was puri-
fied by silica gel column chromatography with 20–40% acetone in
hexane to obtain 14 (0.76 g, 2.45 mmol, 9%) and its 4-substituted
isomer (2.17 g, 6.99 mmol, 25%) as pale yellow oil. Compound
14: 1H NMR (500 MHz, CDCl3): d 0.01 (6H, s, two methyls of
TBS), 0.85 (9H, s, t-butyl of TBS), 1.26 (9H, s, t-butyl), 4.51 (2H, s,
–H2C-50), 5.04 (2H, s, H2-1), 6.89 (1H, s, H-40), 7.34 (1H, s, H-20);
13C NMR (125 MHz, CDCl3): ꢀ5.36 (C200), 18.1 (C300), 25.8 and
26.3 (C4 and C400), 43.3 (C3), 49.3 (C1), 55.5 (C100), 127.7 (C40),
130.7 (C50), 139.3 (C20), 207.8 (C2); HRMS (ESI-TOF, positive
mode): calcd for C16H30N2O2SiNa [M+Na]+ 333.1974, found
333.1971. The substituted position in the imidazole ring of 14
was determined based on an HMBC correlation: –C(O)–CH2–N–
C(CH2OTBS)@CH–.
(17,300); HRMS (ESI-TOF, positive mode): calcd for C17H20ClN2O2
[M+H]+ 319.1213, found 319.1210. Compound 12: 1H NMR
(500 MHz, CD3OD): d 0.69 (9H, s, t-butyl), 4.52 (1H, s, H-3), 4.70
(1H, d, J = 12.8 Hz, (HO)H2C-200), 4.76 (1H, d, J = 12.8 Hz, (HO)H2C-
200), 6.79 (2H, s, H-1 and triazole), 7.43 (4H, m, 4-Cl-phenyl), 7.57
(1H, s, triazole); UV kmax (MeOH) nm (e): 250 (13,800); HRMS (ESI-
TOF, positive mode): calcd for C17H22ClN2O2 [M+H]+ 321.1370,
found 321.1367. The 3S and 3R epimers of 5 were prepared from
optically pure 1712 by the similar manner for the preparation of
the epimeric mixture 5. 3S-5: 1H NMR (270 MHz, acetone-d6): d
0.63 (9H, s, t-butyl), 5.08 (1H, s, H-3), 6.17 (1H, s, –(HO)H C-200),
6.92 (1H, s, H-1), 7.01 (1H, d, J = 1.3 Hz, triazole), 7.46 (4H, s, 4-Cl-
phenyl), 7.48 (1H, d, J = 1.3 Hz, triazole); HRMS (ESI-TOF, positive
mode): calcd for C17H19ClN2O2Na [M+Na]+ 341.1033, found
341.1030. 3R-5: 1H NMR (270 MHz, acetone-d6): d 0.63 (9H, s, t-bu-
tyl), 5.08 (1H, s, H-3), 6.17 (1H, s, –(HO)HC-200), 6.92 (1H, s, H-1), 7.01
(1H, d, J = 1.3 Hz, triazole), 7.46 (4H, s, 4-Cl-phenyl), 7.49 (1H, d,
J = 1.3 Hz, triazole); HRMS (ESI-TOF, positive mode): calcd for
C17H19ClN2O2Na [M+Na]+ 341.1033, found 341.1029.
4.4.3. (Z)-6-tert-Butyl-5-(4-chlorobenzylidene)-6,8-dihydro-
5H-imidazo[2,1-c][1,4]oxazine (6Z)
4.5.3. (Z and E)-2-(5-((tert-Butyldimethylsilyloxy)methyl)-1H-
imidazol-1-yl)-1-(4-chlorophenyl)-4,4-dimethylpent-1-en-3-
one (15Z and 15E)
To a stirred solution of 12 (18.5 mg, 58
lmol) in dry 1,2-dichlo-
roethane (1 mL) was added ZnCl2 (70 mg, 510
lmol) at room tem-
perature. The mixture was stirred for 6 h at 100 °C. After quenched
with H2O, the resulting mixture was extracted with CH2Cl2
(25 mL ꢂ 3). The organic layer was washed with brine, dried over
Na2SO4, and concentrated in vacuo. The residual oil was purified
by column chromatography on silica gel with 30% acetone in hex-
To a stirred solution of 14 (2.2 g, 7.1 mmol) in Ac2O (25 mL) was
added K2CO3 (1.3 g, 9.7 mmol) and p-chlorobenzaldehyde (1.2 g,
8.5 mmol) at the room temperature. The mixture was stirred for
3 h at 100 °C. After quenched with satd NaHCO3, the resulting mix-
ture was extracted with EtOAc (150 mL ꢂ 3). The organic layer was
washed with H2O, dried over Na2SO4, and concentrated in vacuo.
The residual oil was purified by silica gel column chromatography
with 20% EtOAc in hexane to obtain a mixture (1:1, based on TLC)
of 15Z and 15E (1.9 g, 4.4 mmol, 62%) as pale yellow oil.
ane to give 6Z (9.0 mg, 29 lmol, 51%) as pale red oil. Compound
6Z: 1H NMR (270 MHz, CDCl3): d 0.89 (9H, s, t-butyl), 4.17 (1H, s,
H-3), 4.80 (1H, d, J = 14.5 Hz, –O–H2C-200), 5.00 (1H, d, J = 14.5 Hz,
–O–H2C-200), 6.11 (1H, s, H-1), 6.61 (1H, s, triazole), 6.93 (1H, s, tri-
azole), 7.12 and 7.28 (each 2H, m, 4-Cl-phenyl).
4.5.4. (E)-2-(5-((tert-Butyldimethylsilyloxy)methyl)-1H-
imidazol-1-yl)-1-(4-chlorophenyl)-4,4-dimethylpent-1-en-3-
one (15E)
4.4.4. (E)-6-tert-Butyl-5-(4-chlorobenzylidene)-6,8-dihydro-
5H-imidazo[2,1-c][1,4]oxazine (6)
A solution of 6Z (9.0 mg, 29
l
mol) in EtOAc (2 mL) was irradi-
The solution of a mixture of 15Z and 15E (1.9 g, 4.4 mmol) in
EtOAc (100 mL) was irradiated with UV light (365 nm, UVP B-
100A) for 14 h. After concentrated in vacuo, the residual oil was
purified by column chromatography on silica gel with 20% EtOAc
in hexane to give 15E (1.2 g, 2.8 mmol, 63%) as pale yellow oil.
Compound 15E: 1H NMR (270 MHz, acetone-d6): d 0.13 (6H, s,
two methyls of TBS), 0.90 (9H, s, t-butyl of TBS), 0.97 (9H, s, t-bu-
tyl), 4.80 (2H, s, –H2C-50), 7.08 (1H, s, H-400), 7.22 (1H s, H-1), 7.43
(2H, m, 4-Cl-phenyl), 7.47 (2H, m, 4-Cl-phenyl), 7.65 (1H, d,
J = 1.3 Hz, H-200); HRMS (ESI-TOF, positive mode): calcd for
C23H33ClN2O2SiNa [M+Na]+ 455.1898, found 455.1896.
ated with UV light (365 nm, UVP B-100A) for 12 h. After concen-
trated in vacuo, the residual oil was purified by column
chromatography on silica gel with 30% EtOAc in hexane to give 6
(2.9 mg, 29
(500 MHz, acetone-d6):
l
mol, 32%) as pale yellow oil. Compound 6: 1H NMR
d
0.72 (9H, s, t-butyl), 4.84 (1H, d,
J = 15.6 Hz, –O–H2C-200), 4.91 (1H, s, H-3), 5.02 (1H, d, J = 15.6 Hz,
–O–H2C-200), 7.01 (1H, s, H-1), 7.04 (1H, d, J = 1.2 Hz, triazole),
7.44 (4H, m, 4-Cl-phenyl), 7.63 (1H, d, J = 1.2 Hz, triazole); 13C
NMR (125 MHz, acetone-d6): d 27.2 (methyls in t-butyl), 38.8 (ter-
tiary carbon of t-butyl), 61.9 (C60), 77.0 (C3), 114.5 (C400 or C500),
116.2 (C1), 129.6 (4-Cl-phenyl), 130.3 (C400 or C500), 131.6 (4-Cl-
phenyl), 133.4 and 133.5 (C10 and C2), 143.1 (C200); UV kmax (MeOH)
4.5.5. (E)-2-(5-((tert-Butyldimethylsilyloxy)methyl)-1H-
imidazol-1-yl)-1-(4-chlorophenyl)-4,4-dimethylpent-1-en-3-ol
(16)
nm (e): 264 (14,300); HRMS (ESI-TOF, positive mode): calcd for
C17H20ClN2O [M+H]+ 303.1264, found 303.1267.
To a stirred solution of 15E (1.2 g, 2.8 mmol) in MeOH (45 mL)
was added NaBH4 (220 mg, 5.8 mmol) at 0 °C. The mixture was
stirred for 1 h at room temperature. After quenched with satd
NH4Cl, the resulting mixture was extracted with EtOAc
(50 mL ꢂ 3). The organic layer was washed with H2O, dried over
Na2SO4, and concentrated in vacuo. The residual oil was purified
by column chromatography on silica gel with 20% acetone in hex-
4.5. Synthesis of 7 and 9
4.5.1. 5-((tert-Butyldimethylsilyloxy)methyl)-1H-imidazole
(13)
TBS-protected 5-hydroxymethylimidazole (13) (5.9 g, 27.8 mmol)
was prepared from commercially available 4(5)-hydroxymethylimi-

Y. Todoroki et al. / Bioorg. Med. Chem. 17 (2009) 6620–6630
6629
ane to give 16 (0.6 g, 1.4 mmol, 49%) as white solid. Compound 16:
1H NMR (270 MHz, CDCl3): d 0.14 (3H, s, methyl of TBS), 0.17 (3H, s,
methyl of TBS), 0.73 (9H, s, t-butyl of TBS), 0.93 (9H, s, t-butyl), 4.39
(1H, d, J = 8.5 Hz, H-3), 4.53 (1H, d, J = 8.5 Hz, HO-3), 4.68 (1H, d,
J = 13.2 Hz, –H2C-500), 4.81 (1H, d, J = 13.2 Hz, –H2C-500), 6.74 (1H,
s, H-1), 7.04 (1H, d, J = 1.0 Hz, H-400), 7.38 (4H, s, 4-Cl-phenyl),
7.70 (1H, s, H-200); HRMS (ESI-TOF, positive mode): calcd for
C23H36ClN2O2Si [M+H]+ 435.2235, found 435.2234.
4.5.9. Optical resolution of 9
Compound 9 (33 mg) was subjected to chiral HPLC under the
following conditions: column, Chiralpak OD-H (250 ꢂ 10 mm, Dai-
cel); solvent, 13% 2-propanol in hexane; flow rate, 5 mL minꢀ1
;
detection, 254 nm. The materials at tR 14.7 and 30.7 min were col-
lected to give (ꢀ)-9 (14.4 mg) and (+)-9 (14.0 mg), respectively,
with an optical purity of each 99.9%. (+)-9: ½a D24
ꢃ
+15.2 (MeOH, c
0.787); CD kmax (MeOH) nm (De): 275 (+7.2), 241 (0), 216
(ꢀ19.4); 1H NMR (500 MHz, CDCl3): d 0.67 (9H, s, t-butyl), 4.81
(1H, d, J = 14.0 Hz, –H2C-500), 4.82 (1H, s, H-3), 4.99 (1H, d,
J = 14.0 Hz, –H2C-500), 6.81 (1H, s, H-1), 6.91 (1H, s, H-400), 7.32
(4H, m, 4-Cl-phenyl), 7.88 (1H, s, H-200); HRMS (ESI-TOF, positive
mode): calcd for C17H20ClN2O [M+H]+ 303.1264, found 303.1264.
4.5.6. (E)-1-(4-Chlorophenyl)-2-(5-(hydroxymethyl)-1H-
imidazol-1-yl)-4,4-dimethylpent-1-en-3-ol (17)
To a stirred solution of 16 (0.6 g, 1.4 mmol) in EtOH (60 mL) and
H2O (20 mL) was added 1 M HCl (10 mL) at room temperature. The
mixture was stirred for 1.5 h at room temperature. After quenched
with satd NaHCO3, the resulting mixture was extracted with EtOAc
(150 mL ꢂ 3). The organic layer was washed with H2O, dried over
Na2SO4, and concentrated in vacuo. The residual oil was purified
by column chromatography on silica gel with 20% acetone in hex-
ane to give 17 (0.39 g, 1.2 mmol, 89%) as white solid. Compound
17: 1H NMR (270 MHz, CDCl3): d 0.75 (9H, s, t-butyl), 4.48 (1H, s,
H-3), 4.58 (1H, d, J = 13.2 Hz, –H2C-500), 4.74 (1H, d, J = 13.2 Hz, –
H2C-500), 6.68 (1H, s, H-1), 7.00 (1H, s, H-400), 7.38 (4H, m, 4-Cl-phe-
nyl), 7.72 (1H, s, H-200); HRMS (ESI-TOF, positive mode): calcd for
C17H22ClN2O2 [M+H]+ 321.1370, found 321.1342.
(ꢀ)-9: ½a 2D4
ꢃ
ꢀ16.2 (MeOH, c 0.600); CD kmax (MeOH) nm (
De):
275 (ꢀ7.1), 244 (0), 214 (+21.0); 1H NMR (500 MHz, CDCl3): d
0.67 (9H, s, t-butyl), 4.81 (1H, d, J = 14.0 Hz, –H2C-500), 4.82 (1H, s,
H-3), 4.99 (1H, d, J = 14.0 Hz, –H2C-500), 6.81 (1H, s, H-1), 6.91
(1H, s, H-400), 7.32 (4H, m, 4-Cl-phenyl), 7.88 (1H, s, H-200); HRMS
(ESI-TOF, positive mode): calcd for C17H20ClN2O [M+H]+
303.1264, found 303.1264. See Supplementary data for determina-
tion of the absolute configuration of optically pure 9.
4.6. Coexpression of recombinant CYP707A3 and Arabidopsis
P450 reductase (AR2) in E. coli
A truncated Arabidopsis CYP707A3 (707A3d28), which lacked
the putative membrane-spanning segment of the N-terminus, resi-
dues 3–28, was constructed. Cells of E. coli strain BL21 were trans-
formed with the pCW-CYP707A3d28 and pACYC-AR2 constructs.
Cultures (3 mL) were grown overnight in Luria-Bertani medium
4.5.7. (E)-6-tert-Butyl-5-(4-chlorobenzylidene)-5H-
imidazo[5,1-c][1,4]oxazin-8(6H)-one (7)
To a stirred solution of 17 (53 mg, 0.17 mmol) in acetone
(30 mL) was added MnO2 (activated, 186 mg, 2.14 mmol) at room
temperature. The mixture was stirred for 5 h at the same temper-
ature. After filtered MnO2 through a filter paper, the filtrate was
concentrated in vacuo. The residual oil was purified by column
chromatography on silica gel with 15% acetone in hexane to give
supplemented with ampicillin (50
(100
g mLꢀ1). Then, 50 mL of Terrific Broth medium supplemented
with ampicillin (50
g mLꢀ1), chloramphenicol (100 g mLꢀ1), and
l
g mLꢀ1) and chloramphenicol
l
l
l
aminolevulic acid (0.5 mM) was inoculated with an aliquot of the
overnight culture (0.5 mL). The culture was incubated with gentle
shaking (225 rpm) at 37 °C until A600 reached 0.6 (for 2.5–3 h), and
the expression of the P450 enzyme was induced by the addition of
7 (19 mg, 59 lmol, 35%) and 18 (31 mg, 97 lmol, 59%) as white
solids. Compound 7: 1H NMR (270 MHz, CDCl3): d 0.76 (9H, s, t-bu-
tyl), 5.32 (1H, s, H-3), 7.05 (1H, s, H-1), 7.36 (4H, m, 4-Cl-phenyl),
7.91 (1H, s, H-400), 8.08 (1H, s, H-200); UV kmax (MeOH) nm (
e): 257.0
isopropyl-b-D-thiogalactopyranoside (0.1 mM). The culture was
(14,100); HRMS (ESI-TOF, positive mode): calcd for C17H18ClN2O2
[M+H]+ 317.1057, found 317.1064. Compound 18: 1H NMR
(270 MHz,CDCl3): d 0.75 (9H, s, t-butyl), 3.95 (1H, s, HO-3), 4.76
(1H, d, J = 3.0 Hz, H-3), 6.72 (1H, s, H-1), 7.40 (4H, m, 4-Cl-phenyl),
7.92 (1H, s, H-400), 7.98 (1H, s, H-200), 9.74 (1H, s, OHC-500); UV kmax
shaken continuously (150 rpm) at 25 °C and cells were harvested
48 h later by centrifugation at 2330g for 20 min at 4 °C. Pelleted cells
were suspended in 2.5 mL of Buffer A (50 mM potassium phosphate
buffer, pH 7.25, 20% glycerol, 1 mM EDTA, 0.1 mM dithiothreitol).
The suspension was sonicated for 30 s, and the enzyme solution in
the supernatant was collected by centrifugation at 23,470g for
30 min at 4 °C. The P450 content was determined by spectrophoto-
metric analysis, using the extinction coefficient of reduced CO differ-
ence spectrum (91.1 mMꢀ1 cmꢀ1).20
(MeOH) nm (e): 257 (15,300); HRMS (ESI-TOF, positive mode):
calcd for C17H20ClN2O2 [M+H]+ 319.1213, found 319.1204.
4.5.8. (E)-6-tert-Butyl-5-(4-chlorobenzylidene)-6,8-dihydro-
5H-imidazo[5,1-c][1,4]oxazine (9)
To a stirred solution of 17 (51 mg, 0.16 mmol) in dry dichloro-
ethane (20 mL) was added ZnCl2 (60 mg, 0.44 mmol) at room tem-
perature. The mixture was stirred for 31 h at the same temperature
before added H2O. The mixture was extracted with CH2Cl2
(50 mL ꢂ 3). The organic layer was washed with brine, dried over
Na2SO4, and concentrated in vacuo. The residual oil was purified
by column chromatography on silica gel with 20% acetone in hex-
4.7. Enzyme assay
Kinetic analysis was performed according to the detailed pro-
tocols described previously.13
A
reaction mixture containing
g mLꢀ1 CYP707A3 microsomes coexpressed AR2 in E. coli,
(+)-ABA (final concn: 0.5–128 M), inhibitors (0 for control, 2–
1000 nM, in 5 L DMF) and 50 M NADPH in 50 mM potassium
25
l
l
l
l
ane to give 9 (13 mg, 43 lmol, 27%) as white solid. Compound 9:
phosphate buffer (pH 7.25) were incubated for 10 min at 30 °C.
Reactions were initiated by adding NADPH, and stopped by addi-
1H NMR (500 MHz, CDCl3): d 0.67 (9H, s, t-butyl), 4.81 (1H, d,
J = 14.0 Hz, –H2C-500), 4.82 (1H, s, H-3), 4.99 (1H, d, J = 14.0 Hz, –
H2C-500), 6.81 (1H, s, H-1), 6.91 (1H, s, H-400), 7.32 (4H, m, 4-Cl-phe-
nyl), 7.88 (1H, s, H-200); 13C NMR (125 MHz, CDCl3): d 26.4 (methyls
of t-butyl), 38.2 (tertiary carbon of t-butyl), 59.2 (–H2C-500), 77.5
(C3), 119.2 (C3), 123.1 (C400), 127.4 (C500), 129.0 (C30 and C50),
130.2 (C20 and C60), 131.4 (C200), 132.3 (C2), 133.0 (C40), 133.6
tion of 50
with 100
l
L of 1 M NaOH. The reaction mixtures were acidified
lL of 1 M HCl. To extract the reaction products, the mix-
tures were loaded onto Oasis HLB cartridges (1 mL, 30 mg;
Waters) and washed with 1 mL of 10% MeOH in H2O containing
0.5% AcOH. The enzyme products were then eluted with 1 mL
of MeOH containing 0.5% AcOH, and the eluate was concentrated
(C10); UV kmax (MeOH) nm (
e): 265 (15,600); HRMS (ESI-TOF, posi-
in vacuo. The dried sample was dissolved in 50
10 L were subjected to HPLC. HPLC conditions were: ODS col-
umn, Hydrosphere C18 (150 ꢂ 6.0 mm, YMC); solvent, 35% MeOH
lL of MeOH, and
tive mode): calcd for C17H20ClN2O [M+H]+ 303.1264, found
303.1264.
l

6630
Y. Todoroki et al. / Bioorg. Med. Chem. 17 (2009) 6620–6630
or 20% MeCN in H2O containing 0.1% AcOH; flow rate,
1.0 mL minꢀ1; detection, 254 nm. Enzyme activity was confirmed
by determining the amounts of phaseic acid (PA) in control exper-
iments before each set of measurements. Inhibition constants
were determined using the Enzyme Kinetics module of SIGMAPLOT
10 software21 after determining the mode of inhibition by plot-
ting the reaction velocities in the presence and absence of inhib-
itor on a double-reciprocal plot. For the non-inhibited enzymatic
4.9. Computational method
The minimum-energy conformer of compound 1 was generated
and minimized using MM3 combined with the molecular dynamics
simulation built into CACHE 3.11.22 This MM3-minimized structure
was fully optimized with density functional theory, using the
Becke three parameter hybrid functional (B3LYP) method and the
6-31G(d) basis set in GAUSSIAN 03.23
reaction, the KM for (+)-ABA was calculated to be 3.4 0.6 lM,
based on five separate experiments. All tests were conducted at
Acknowledgments
least three times.
We thank Toray Industries Inc., Tokyo, Japan, for the gift of (+)-
ABA. This research was supported by a Grant-in-Aid for Scientific
Research (No. 18380192) from the Ministry of Education, Culture,
Sports, Science and Technology of Japan.
4.8. Bioassays
4.8.1. Rice seedling elongation assay
Seeds of rice (Oryza sativa L. cv. Nipponbare) were sterilized
with EtOH for 5 min and washed with running tap water. The
sterilized seeds were soaked in water to germinate for 3 days
at 25 °C. The seeds were then placed in a glass tube (40 mm
i.d.) containing 2 mL of a test solution and grown with the tube
sealed with a plastic cap under continuous light (10,000 lux) at
25 °C. When the seedlings were 7 days old, the length of the sec-
ond leaf sheath was measured, and the inhibition ratio was cal-
culated. The inhibition ratio is defined as [(A ꢀ B)/A] ꢂ 100,
where A = the mean length of the second leaf sheath when water
was used, and B = the mean length of the second leaf sheath
when a test compound was used. All tests were conducted at
least twice.
Supplementary data
Supplementary data (determination of the absolute configura-
tion of optically pure 9, determination of the E- and Z-isomers of
UNI analogues, and 1H NMR spectra of new compounds which
were used in the enzymatical and biological assays) associated
with this article can be found, in the online version, at
doi:10.1016/j.bmc.2009.07.070.
References and notes
1. Funaki, Y.; Ishiguri, T.; Kato, T.; Tanaka, S. J. Pesticide Sci. 1984, 9, 229.
2. Funaki, Y.; Oshita, H.; Yamamoto, S.; Tanaka, S.; Kato, T. (Sumitomo Chemical
Co.). Ger. Offen. DE3010560, 1980.
3. Izumi, K.; Kamiya, Y.; Sakurai, A.; Oshio, H.; Takahashi, N. Plant Cell Physiol.
1985, 26, 821.
4. Iwasaki, T.; Shibaoka, H. Plant Cell Physiol. 1991, 32, 1007.
5. Asami, T.; Mizutani, M.; Fujioka, S.; Goda, H.; Min, Y. K.; Shimada, Y.; Nakano,
T.; Takatsuto, S.; Matsuyama, T.; Nagata, N.; Sakata, K.; Yoshida, S. J. Biol. Chem.
2001, 276, 25687.
6. Izumi, K.; Nakagawa, S.; Kobayashi, M.; Oshio, H.; Sakurai, A.; Takahashi, N.
Plant Cell Physiol. 1988, 29, 97.
7. Kitahata, N.; Saito, S.; Miyazawa, Y.; Umezawa, T.; Shimada, Y.; Min, Y. K.;
Mizutani, M.; Hirai, N.; Shinozaki, K.; Yoshida, S.; Asami, T. Bioorg. Med. Chem.
2005, 13, 4491.
4.8.2. Drought tolerance assay
Apple [Malus sylvestris (L.) Mill. var. domestica (Borkh.) Mansf.]
cultivar Akitabeniakari seeds were soaked overnight in water,
then sown in moist vermiculite and grown in a greenhouse. Seed-
lings (90 days old) were divided into two groups: water-stressed
and well-watered. The seedlings of the water-stressed group were
sprayed uniformly with 2 mL of a test solution (10, 50, or
100 lM) containing 3R-Abz-F1, UNI (positive control), or distilled
water (untreated control) before water-stressing them for 150 h,
then watering them once, and then water-stressing them again
for an additional 330 h. The well-watered group was watered
once daily so that the vermiculite was saturated. Measurement
of stomatal aperture was performed as follows: the state of 10
stomatal apertures per leaf was observed using a TM-1000 micro-
scope (Hitachi High-Technologies Co., Tokyo, Japan); three leaves
from three randomly selected seedlings (one leaf per seedling)
were used; that is, 30 stomatal apertures per treatment were
measured. The water potential of the vermiculite was measured
with a WP4-T water potential meter (Decagon Devices, Inc.,
USA). Apple [M. sylvestris (L.) Mill. var. domestica (Borkh.) Mansf.]
cultivar Fuji seeds were soaked overnight in water, then sown in
moist vermiculite and grown in a greenhouse. The seedlings were
60 days old at the time of the treatment. Each plant was sprayed
uniformly with 2 mL of the test solution. The untreated control
group was sprayed with distilled water, and the treated group
8. Saito, S.; Okamoto, M.; Shinoda, S.; Kushiro, T.; Koshiba, T.; Kamiya, Y.; Hirai,
N.; Todoroki, Y.; Sakata, K.; Nambara, E.; Mizutani, M. Biosci., Biotechnol.,
Biochem. 2006, 70, 1731.
9. Saito, S.; Hirai, N.; Matsumoto, C.; Ohigashi, H.; Ohta, D.; Sakata, K.; Mizutani,
M. Plant Physiol. 2004, 134, 1439.
10. Kushiro, T.; Okamoto, M.; Nakabayashi, K.; Yamagishi, K.; Kitamura, S.; Asami,
T.; Hirai, N.; Koshiba, T.; Kamiya, Y.; Nambara, E. ENBO J. 2004, 23, 1647.
11. Poulos, T. L.; Johnson, E. F. In Cytochrome P450: Structure, Mechanism, and
Biochemistry; Ortiz de Montellano, P. R., Ed., 3rd ed.; Kluwer Academic/Plenum
Publishers: New York, 2005; pp 87–114.
12. Schuler, M. A. Crit. Rev. Plant Sci. 1996, 15, 235.
13. Ueno, K.; Araki, Y.; Hirai, N.; Saito, S.; Mizutani, M.; Sakata, K.; Todoroki, Y.
Bioorg. Med. Chem. 2005, 13, 3359.
14. Todoroki, Y.; Kobayashi, K.; Yoneyama, H.; Hiramatsu, S.; Jin, M.-H.; Watanabe,
B.; Mizutani, M.; Hirai, N. Bioorg. Med. Chem. 2008, 16, 3141.
15. Wang, Z.; Silverman, R. B. Bioorg. Med. Chem. 2006, 14, 2242. and references
cited therein.
16. Saito, J.; Kurahashi, Y.; Goto, T.; Yamaguchi, N. (Nihon Tokushu Noyaku Seizo K.
K.). Eur. Pat. Appl. EP185987, 1986.
17. Nambara, E.; Marion-Poll, A. Annu. Rev. Plant Biol. 2005, 56, 165.
18. See Supplementary data in Ref. 14.
19. Amino, Y.; Eto, H.; Eguchi, C. Chem. Pharm. Bull. 1989, 37, 1481.
20. Omura, T.; Sato, R. J. Biol. Chem. 1964, 239, 2370.
21. SIGMAPLOT 10; Systat Software: Richmond CA, 2006.
was sprayed with a solution of 3R-Abz-F1 (10
(10 M). All test groups were watered before treatment. The
seedlings were water-stressed for 480 h.
lM) or S-UNI
l
22. CACHE, 3.11; Oxford Molecular Ltd: London, 1998.
23. GAUSSIAN 03 Revision D.01; Gaussian: Wallingford, CT, 2004.