3′-Azidoabscisic acid as a photoaffinity reagent for abscisic acid binding proteins

Article abstract of DOI:10.1016/S0960-894X(01)00431-0

3′-Azidoabscisic acid was synthesized as a potential photoaffinity reagent for abscisic acid binding proteins. This compound was stable in organic and aqueous solutions in the dark, but was decomposed by UV irradiation. Its biological activity was equivalent to that of abscisic acid, suggesting that it may be an effective photoaffinity reagent.

Full text of DOI:10.1016/S0960-894X(01)00431-0

Bioorganic & Medicinal Chemistry Letters 11 (2001) 2381–2384
30-Azidoabscisic Acid as a Photoaffinity Reagent
for Abscisic Acid Binding Proteins
Yasushi Todoroki,a,* Takahiro Tanaka,a
Masayuki Kisamoria and Nobuhiro Hiraib
aDepartment of Applied Biological Chemistry, Faculty of Agriculture, Shizuoka University, Shizuoka 422-8529, Japan
bDivision of Applied Life Sciences, Graduate School of Agriculture, Kyoto University, Kyoto 606-8502, Japan
Received 21 May 2001; accepted 21 June 2001
Abstract—30-Azidoabscisic acid was synthesized as a potential photoaffinity reagent for abscisic acid binding proteins. This com-
pound was stable in organic and aqueous solutions in the dark, but was decomposed by UV irradiation. Its biological activity was
equivalent to that of abscisic acid, suggesting that it may be an effective photoaffinity reagent. # 2001 Elsevier Science Ltd. All
rights reserved.
Abscisic acid [(10S)-(+)-ABA (1)] is the primary hor-
mone that induces adaptive reactions to protect plants
from environmental stresses such as desiccation and
freezing.1ꢀ4 A thorough understanding of the mechan-
ism of ABA action is important not only as a point of
academic interest, but also to promote the agricultural
application of ABA. The initial perception and meta-
bolic inactivation of ABA by target cells remain obscure
at the molecular level. ABA receptors are believed to be
membrane proteins,5,6 and the catabolic enzyme that
catalyzes the first step of ABA catabolism is likely to be
the microsomal cytochrome P-450 monooxygenase.7
However, the corresponding proteins and genes have
not yet been isolated.
Milborrow synthesized 1-azido-ABA (2), which was
10% as effective as ABA in stomatal closure and Lemna
gibba growth assays.10 Photolysis of tritiated 2 in a
solution of bovine serum albumin (BSA) gave labeled
BSA. However, compound 2 was so unstable and
rapidly broken down that it may not be useful as a
photoaffinity label for the ABA receptor. The 40-car-
bonyl group is a good site for tethering functional
groups for photoaffinity, since most ABA derivatives
modified at C-40 exhibit moderate biological activities.
Kohler et al. synthesized [125I]-labeled ABA that teth-
ered an aromatic hydrazide at C-40 (3) as the first radio-
iodinated ABA photoaffinity probe.11 This compound
was about one-tenth as active as ABA in the inhibition
of GA-induced a-amylase. An anthracenone ABA ana-
logue (4) was reported by Irvine et al.12 This analogue
was designed based on the fact that benzophenone-con-
taining substrate analogues are advantageous with
regard to chemical stability, a long activation wave-
length (350–360 nm), and many excitation–relaxation
cycles. The activity of 4 was one-third that of ABA in
the inhibition of corn cell growth. There is some doubt
whether these analogues function as a specific photo-
affinity reagents for the ABA receptor because these
modifications of the parent compound involved the loss
of a significant group for bioactivity, alteration of the
ABA skeleton, or the addition of a large substituent
which may interfere with ABA–receptor interactions,
although the investigation of ABA-binding proteins
using these probes has not yet been reported.
A powerful technique for identifying hormone-binding
proteins is photoaffinity labeling. ABA can naturally
form covalent bonds under UV irradiation, since ABA
contains
a UV-sensitive a,b-unsaturated carbonyl
group, an enone structure, in its six-membered ring.
Indeed, Hornberg and Weiler used this characteristic of
ABA in their study.8 More recently, Cornelussen et al.
tried to optimize the conditions for UV-induced cross-
linking of ABA in a model experiment.9 ABA analogues
with a more photosensitive functional group, azide,
have also been synthesized.10 In 1993, Willows and
*Corresponding author. Tel.: +81-54-238-4871; fax: +81-54-238-
4871; e-mail: aytodor@agr.shizuoka.ac.jp
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00431-0

2382
Y. Todoroki et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2381–2384
was treated with 2 equivalents of NaN3 in DMF in the
presence of NH4Cl at 80 ꢁC to give the methyl ester of
30-azido-ABA (10a) in 33% yield. The racemate 10a was
optically resolved by HPLC with a chiral column,
Chiralcel OD (250ꢂ4.6 mm, Daicel; solvent, 5% i-PrOH
in hexane; flow rate, 1 mL minꢀ1; detection, 254 nm) to
give (+)- and (ꢀ)-10a. Each enantiomer was hydrolyzed
with porcine liver esterase (EC 3.1.1.1, Sigma E3019) in
MeOH (3 mL) and potassium phosphate buffer (0.1 M,
pH 8.0, 16 mL). The obtained free acids (+)- and (ꢀ)-10
were purified by chromatography on silica gel with tol-
uene–EtOAc (7:3) containing 5% AcOH. These enan-
tiomers were identified by spectral data.18 Since the
Cotton effects in the CD spectra of (+)- and (ꢀ)-10
were similar to those of (10S)-(+)- and (10R)-(ꢀ)-1,19
respectively, the absolute configuration at C-10 was
determined to be S for (+)-10 and R for (ꢀ)-10.
The 30-position of ABA can also be modified while
maintaining bioactivity. In our previous paper,13 we
reported that the bioactivities of 30-halogenated ABA
(5–8) and 30-methyl-ABA (9) were slightly stronger than
that of ABA. Introducing a hydrophobic substituent at
C-30 may cause a hydrophobic interaction to reinforce
binding with the ABA receptor. This means that acti-
vation of a photolabile group at C-30 may result in
effectively labeling the ABA-binding proteins. We
selected an azide as the photolabile group because of its
size and easy introduction at C-30. By introducing it at
C-30, we can build an aryl azide, which is a popular and
useful group for photoaffinity labeling studies.14ꢀ16 Its
relatively small size would result in high affinity and
specificity for ABA-binding proteins. In this paper, we
describe the synthesis and biological activity of 30-azido-
ABA (10). We also describe the stability and photolysis
of 10 in various solutions to estimate its potential as a
photoaffinity reagent.
Biological Activity
The biological activities of optically active 1 and 10
were evaluated in the two bioassays (Fig. 2).20 In the
inhibition of lettuce seed germination, the activity of
(+)-10 was equivalent to that of (+)-1, whereas (ꢀ)-10
was not effective at the concentrations tested. In the
inhibition of rice seedling elongation, (+)-10 showed
1/3 the activity of (+)-1, whereas (ꢀ)-10 showed no
activity. The high bioactivities of (+)-10 mean that
(+)-10 may be an effective photoaffinity reagent. The
ABA receptors should interact with the azide (+)-10 in
a manner similar to (+)-1. On the other hand, the lack
of an effect of (ꢀ)-10 in the bioassays can be explained
by destruction of the pseudo-symmetrical structure of
ABA. The high bioactivity of unnatural (ꢀ)-1 has been
observed in most bioassays.1ꢀ4 ABA receptors can
recognize unnatural (ꢀ)-1 to induce ABA-activity,
probably due to its pseudo-symmetrical structure,21
which is derived from the pseudo-symmetrical cyclo-
hexenone ring that has a chiral center at C-10 and
methyl groups at both C-20 and C-60. Therefore, the
introduction of a substituent on the ring of (ꢀ)-1 tends
to eliminate the pseudo-symmetry and thus the bioac-
tivity. Indeed, most of the unnatural enantiomers of
ring-modified analogues are biologically inactive.13,22ꢀ27
Synthesis
Racemic ABA, purchased from Tokyo Kasei Kogyo
Co. Ltd., Tokyo, Japan, was converted to the methyl
ester of 20a,30a-deoxy-20a,30a-epoxy-ABA (11) in two
steps as reported elsewhere (Fig. 1).17 The epoxide 11
Figure 1. Synthesis and optical resolution of 10.

Y. Todoroki et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2381–2384
2383
Figure 4. The time-course of UV absorption for the photolysis
(365 nm) of (+)-10 in Tris–Mes buffer at (a) 0 s, (b) 20 s, (c) 40 s, (d)
80 s, (e) 160 s, (f) 320 s and (g) 640 s. The initial concentration of (+)-
10 was 328 mM (100 mg mLꢀ1).
handled. UV irradiation at 254 and 365 nm was per-
formed using compact UV lamps.31 Sample solutions
(0.15 mg per 1.5 mL solution) were prepared in quartz
cells, and 10-mL aliquots were analyzed by HPLC using
an ODS column (solvent: 50% MeOH in 0.1% aqueous
AcOH at 1.0 mL minꢀ1; detection: 254 nm) at appro-
priate intervals after UV irradiation. Since the photo-
lysis profiles of (+)-1 and (+)-10 were scarcely
influenced by changing the solvent, those in Tris–Mes
buffer are shown in Figure 3. A 10-min irradiation at
254 nm converted (+)-1 to an equilibrium mixture of
2Z- and 2E-isomers (ca. 1:1). Decomposition of both
isomers was observed upon further irradiation. Irradia-
tion at 365 nm caused only Z/E isomerization, which
proceeded more slowly than with irradiation at 254 nm.
These results were similar to those reported by Corne-
lussen et al.9 Compound (+)-10 was completely decom-
posed to many compounds by 10 min of irradiation at
both 254 and 365 nm. Most of the degradation products
were more polar than the initial compound. The time-
course of UV absorption for the photolysis (365 nm) of
(+)-10 is shown in Figure 4. Absorption maxima of
(+)-10 before irradiation were observed at 250 and
280 nm; the former is due to the dienoic acid side-chain,
and the latter is due to the enone moiety with an azide
group in the ring. The decrease in absorption after irra-
diation at 280 nm was greater than that at 250 nm, sug-
gesting decomposition of the 30-azide. This was
supported by the fact that the absorption of an azide in
the IR spectrum disappeared after irradiation. The pre-
sent results suggest that (+)-10 is a potential photo-
affinity reagent for ABA-binding proteins.
Figure 2. Inhibitory activities of optically active 1 and 10 in (a) lettuce
seed germination and (b) rice seedling elongation. &, (+)-1; *, (+)-
10; &, (ꢀ)-1; *, (ꢀ)-10.
Figure 3. Photolysis of (+)-1 and (+)-10 by UV irradiation. The sam-
ples were analyzed by HPLC. a: & (+)-1, * (+)-2E-1, 254nm; b: *
(+)-10, 254 nm; c: & (+)-1, * (+)-2E-1, 365nm; d: * (+)-10, 365 nm.
References and Notes
Stability and Photolysis
1. Hirai, N. In Comprehensive Natural Products Chemistry;
Mori, K., Ed.; Elsevier Science: Amsterdam, 1999; Vol. 8, pp
72–91.
The stability and photolysis of (+)-10 were tested in
MeOH and Tris–Mes (10 mM, pH 6.5)28 and citrate–
KOH (10 mM, pH 4) buffers, which are often used in
photoaffinity labeling for a plasma membrane frac-
tion.29,30 Compound (+)-10 was as stable as (+)-1 in
these three solvents in the dark at ambient temperature,
and was relatively stable even under the fluorescent
lamps (ꢃ500 lux) in the laboratory, and so was easily
2. Abscisic Acid; Davies, W. J.; Jones, H. G., Ed.; BIOS Sci-
entific: Oxford, 1991.
3. Zeevaart, J. A. D.; Creelman, R. A. Annu. Rev. Plant Phy-
siol. Plant Mol. Biol. 1988, 39, 439.
4. Abscisic Acid; Addicott, F. T., Ed.; Praeger: New York,
1983.
5. Leyman, B.; Geelen, D.; Quintero, F. J.; Blatt, M. R.
Science 1999, 283, 537.

2384
Y. Todoroki et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2381–2384
6. Pedron, J.; Brault, M.; Nake, C.; Miginiac, E. Eur. J. Bio-
chem. 1998, 252, 385.
were placed on two sheets of Advantec No. 2 filter paper
soaked in 2 mL of a test solution in a polystyrene dish (60 mm
diam) and allowed to germinate under illumination (6400 lux)
at 25 ꢁC. After 48 h, the inhibition ratio was defined as
[(AꢀB)/A]ꢂ100, where A=the number (25) of seeds that ger-
minated when water was used, and B=the number of seeds
that germinated when a test compound was used. The rice
elongation assay was performed as follows. Seeds of rice
(Oryza sativa L. cv. Nihonbare) were soaked in EtOH for
5 min, sterilized with 1% antiformin for 1 h, and washed with
running tap water for 3 h. The sterilized seeds were allowed to
germinate in water for 3 days at 25 ꢁC. The resulting seedlings
were placed in a glass tube containing 2 mL of a test solution,
and grown with the tube sealed by a sheet of polyethylene film
under continuous illumination (6400 lux) at 25 ꢁC. The length
of the second leaf sheath was measured after 7 days, and the
inhibition ratio was calculated. The inhibition ratio was
defined as [(AꢀB)/A]ꢂ100, where A=the mean length
(25 mm) 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.
21. Milborrow, B. V. In Plant Growth Substances 1985; Bopp,
M., Ed.; Springer: Berlin, 1986; pp 108–119.
7. Krochko, J. E.; Abrams, G. D.; Loewen, M. K.; Abrams,
S. R.; Cutler, A. J. Plant Physiol. 1998, 118, 849.
8. Hornberg, C.; Weiler, E. W. Nature 1984, 310, 321.
9. Cornelussen, M. H. M.; Karssen, C. M.; van Loon, L. C.
Phytochemistry 1995, 39, 959.
10. Willows, R. D.; Milborrow, B. V. Phytochemistry 1993,
32, 869.
11. Kohler, A. D.; Beale, M. H.; Rollason, R.; Barratt,
D. H. P.; Lewis, M. J.; Van der Muelen, R. M.; Wang, M. J.
Chem. Soc., Perkin Trans. 1 1997, 1543.
12. Irvine, N. M.; Rose, P. A.; Cutler, A. J.; Squires, T. M.;
Abrams, S. R. Phytochemistry 2000, 53, 349.
13. Arai, S.; Todoroki, Y.; Ibaraki, S.; Naoe, Y.; Hirai, N.;
Ohigashi, H. Phytochemistry 1999, 52, 1185.
14. Feldwisch, J.; Zettl, R.; Hesse, F.; Schell, J.; Palme, K.
Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 475.
15. Macdonald, H.; Jones, A. M.; King, P. J. J. Biol. Chem.
1991, 266, 7393.
16. Hicks, G. R.; Rayle, D. L.; Jones, A. M.; Lomax, T. L.
Proc. Natl. Acad. Sci. U.S.A. 1989, 86, 4948.
17. Todoroki, Y.; Hirai, N.; Ohigashi, H. Tetrahedron 1995,
51, 6911.
22. Todoroki, Y.; Hirai, N.; Koshimizu, K. Biosci. Biotechnol.
Biochem. 1994, 58, 705.
23. Todoroki, Y.; Hirai, N.; Koshimizu, K. Phytochemistry
1995, 38, 561.
24. Todoroki, Y.; Hirai, N.; Ohigashi, H. Phytochemistry
1995, 40, 633.
25. Nakano, S.; Todoroki, Y.; Hirai, N.; Ohigashi, H. Biosci.
Biotechnol. Biochem. 1995, 59, 1699.
1
18. (+)-10. H NMR (270 MHz, CD3OD): d 1.01 (3H, s, H3-
90), 1.05 (3H, s, H3-80), 1.84 (3H, s, H3-70), 1.99 (3H, d,
J=1.0 Hz, H3-6), 2.30 (1H, d, J=16.5 Hz, H-50), 2.61 (1H, d,
J=16.5 Hz, H-50), 5.77 (1H, s, H-2), 6.12 (1H, d, J=16.0 Hz,
H-5), 7.67 (1H, d, J=16.0 Hz, H-4); IR nmax (CHCl3) cmꢀ1
:
2122 (N3); FAB-MS m/z (rel. int.): 306 [M+1]+ (5); UV lmax
(MeOH) nm (e): 251.6 (15,000), 281.4 (12,700); [a]2D5 +447.2ꢁ
(MeOH; c 0.197); CD lext (MeOH) nm (y): 239 (ꢀ27,786),
26. Todoroki, Y.; Nakano, S.; Hirai, N.; Ohigashi, H. Tetra-
hedron 1996, 52, 8081.
1
284 (+93,953). (ꢀ)-10. H NMR (270 MHz, CD3OD): d 1.01
(3H, s, H3-90), 1.05 (3H, s, H3-80), 1.84 (3H, s, H3-70), 1.96
(3H, s, H3-6), 2.29 (1H, d, J=16.5 Hz, H-50), 2.62 (1H, d,
J=16.5 Hz, H-50), 5.79 (1H, s, H-2), 6.02 (1H, d, J=16.2 Hz,
H-5), 7.58 (1H, d, J=16.2 Hz, H-4); UV lmax (MeOH) nm
(e): 251.0 (12,800), 281.4 (11,000); [a]2D5 ꢀ452.4ꢁ (MeOH; c
0.116); CD lext (MeOH) nm (y): 237 (+31,700), 283
(ꢀ94,900).
27. Todoroki, Y.; Nakano, S.; Arai, S.; Hirai, N.; Ohigashi,
H. Biosci. Biotechnol. Biochem. 1997, 61, 2043.
28. Tris: tris(hydroxymethyl)aminomethane (TRIZMA1
BASE, Sigma T1503). Mes: 4-morpholineethanesulfonic acid
(Sigma M2933).
29. Matsubayashi, Y.; Sakagami, Y. J. Biol. Chem. 2000, 275,
15520.
30. Feldwisch, J.; Zettl, R.; Hesse, F.; Schell, J.; Palme, K.
Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 475.
19. Koreeda, M.; Weiss, G.; Nakanishi, K. J. Am. Chem. Soc.
1973, 95, 239.
20. Lettuce germination assay was performed as follows.
Twenty seeds of lettuce (Lactuca sativa L. cv. Grand Rapids)
31. 254 nm: UVG-11 (4 W), UVP, Inc. 365 nm: UVG-21 (4
W), UVP, Inc.