Cassia obtusifolia MetE as a cytosolic target for potassium isolespedezate, a leaf-opening factor of cassia plants: Target exploration by a compact molecular-probe strategy

Article abstract of DOI:10.1002/asia.201100392

Affinity chromatography by using ligand-immobilized bead technology is generally the first choice for target exploration of a bioactive ligand. However, when a ligand has comparatively low affinity against its target, serious difficulties will be raised in affinity-based target detection. We report here that the use of compact molecular probes (CMP) will be advantageous in such cases; it enables the retention of moderate affinity between the ligand and its target in contrast to immobilizing the ligand on affinity beads that will cause a serious drop in affinity to preclude target detection. In the CMP strategy, a CMP containing an azide handle is used for an initial affinity-based labeling of target, and subsequent tagging by CuAAC with a large FLAG tag will give a tagged target protein. By using the CMP strategy, we succeeded in the identification of Cassia obtusifolia MetE as a cytosolic target protein of potassium isolespedezate (1), a moderately bioactive ligand. Affinity chromatography by using ligand-immobilized bead technology is the first choice for target exploration of a bioactive ligand. However, when a ligand has comparatively low affinity against its target, serious difficulties will arise. Herein, we report that the use of a compact molecular probe (CMP) will be advantageous in such cases (see scheme).

Full text of DOI:10.1002/asia.201100392

FULL PAPERS
DOI: 10.1002/asia.201100392
Cassia obtusifolia MetE as a Cytosolic Target for Potassium Isolespedezate,
a Leaf-Opening Factor of Cassia plants: Target Exploration by a Compact
Molecular-Probe Strategy
Minoru Ueda,*[a] Yoshiyuki Manabe,[a] Yuki Otsuka,[b] and Nobuyuki Kanzawa[b]
Abstract: Affinity chromatography by
using ligand-immobilized bead technol-
ogy is generally the first choice for
target exploration of a bioactive ligand.
However, when a ligand has compara-
tively low affinity against its target, se-
rious difficulties will be raised in affini-
ty-based target detection. We report
here that the use of compact molecular
probes (CMP) will be advantageous in
such cases; it enables the retention of
moderate affinity between the ligand
and its target in contrast to immobiliz-
ing the ligand on affinity beads that
will cause a serious drop in affinity to
preclude target detection. In the CMP
strategy, a CMP containing an azide
handle is used for an initial affinity-
based labeling of target, and subse-
quent tagging by CuAAC with a large
FLAG tag will give a tagged target
protein. By using the CMP strategy, we
succeeded in the identification of
Cassia obtusifolia MetE as a cytosolic
target protein of potassium isolespede-
zate (1), a moderately bioactive ligand.
Keywords: affinity
chromatog-
raphy · Cassia obtusifolia · molec-
ular probes · natural products ·
proteins
Introduction
circadian rhythmic leaf opening.[3] Previous studies suggest-
ed that 1 affects motor cells in pulvini of Cassia mimo-
soides[4] through binding membrane proteins of molecular
weights 180 and 210 kDa.[5] In this paper, we report on the
identification of Cassia obtusifolia MetE as a cytosolic
target protein of 1 by using our compact molecular probe
approach (CMP). Additionally, a hypothetical mechanism
including the involvement of identified membrane proteins
will be discussed.
Target identification of bioactive metabolites is an emerging
field of natural product chemistry and bioorganic chemis-
try.[1] In particular, target identifi-
cation of biologically active metab-
olites that induce a biologically in-
triguing phenomena have the pos-
sibility of opening a new aspect of
chemical biology based on natural
product chemistry.
Potassium isolespedezate (1) was
Results and Discussion
isolated as a leaf-opening factor of
Target Identification of Potassium Isolespedezate: Issues to
be Solved
Cassia plants.[2] Observed diurnal change in the endogenous
content of 1 suggested that it might be involved in triggering
Affinity chromatography (AC) by using ligand-immobilized
bead technology is generally the first choice for target ex-
ploration.[6] This potent technology has been applied to
many target explorations.[7] However, it is dependent on the
extremely high bioactivity of the ligand, as strong as
10À9 m,[8a] because the bioactivity of the ligand is tightly cor-
related with its affinity to the target. AC is based on the
noncovalent interaction between ligand and target, and thus
is dependent on a chemical equilibrium (Scheme 1a). The
affinity would dramatically decrease when the ligand is im-
mobilized on beads. Thus, successful target exploration re-
quires both a high affinity ligand and an abundant target
[a] Prof. M. Ueda, Y. Manabe
Department of Chemistry, Tohoku University
6-3 Aramaki-aza Aoba, Aoba-ku
Sendai 980-8578 (Japan)
Fax : (+81)22-795-6553
E-mail: ueda@m.tohoku.ac.jp
[b] Y. Otsuka, Prof. N. Kanzawa
Department of Material & Life Science
Faculty of Science & Technology, Sophia University
7-1, Kioi-cho, Chiyoda-ku, Tokyo 102-8554 (Japan)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.201100392.
3286
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 3286 – 3297

functional group, molecular
tag, and a long linker.[14] Addi-
tionally, changes in physical
properties, such as solubility
and polarity, caused by the in-
troduction of a relatively large
functional unit, should be care-
fully considered. As almost all
of these functionalities are hy-
drophobic, they are prone to
nonspecific interactions, and
when a biotinyl group is used
as a reporter tag, the abundant
endogenous biotinylated pro-
teins[15] will interfere with the
identification of the desired
true signal for the chemically
labeled target protein.
Scheme 1. Concepts of a) the AC separation of target protein and b) the CMP approach to target protein.
Target Identification of Potassium Isolespedezate: Target
Identification by a Combination of CMP and the Stepwise
Tagging Approach
protein.[8] Ligands of much lower affinity cannot give a clear
result because of nonspecific binding.
On the other hand, our ligand 1 showed moderate bioac-
tivity (10À5–10À6 m) on Cassia plants and was presumed to
have moderate affinity with the corresponding target. Such
a comparatively low affinity is expected to cause difficulties
in affinity-based target detection.[9] In contrast to ligand-im-
mobilized AC, a molecular-probe approach would be benefi-
cial because a well-designed molecular probe can retain bio-
activity and affinity to the target (Scheme 1b). Additionally,
formation of a covalent bond between ligand and target will
effectively remove the complex from chemical equilibrium,
which is advantageous for the formation of an energetically
unfavorable complex. We realized that by using a molecular
probe for which high bioactivity can be maintained would
be advantageous for target exploration of moderately bioac-
tive ligands. Successful target identification by a molecular-
probe approach, such as phytosulfokaine,[10] Pep1,[11] and
RALF,[12] usually involves the use of a probe with bioactivity
maintained as high as the naturally occurring ligand. These
results suggested that retention of the original bioactivity
should be the top priority when designing such a molecular
probe. A well-designed molecular probe is indispensable for
successful target exploration of 1.
Two major issues must be addressed to successfully ach-
ieve a well-designed molecular probe for target detection of
1. The first is a compact structure of the probe. Introduction
of a relatively large functional unit, such as linker-connected
biotin, in the compact structure of 1 caused a 100-fold de-
crease in its bioactivity.[5]
Another serious issue regarding design of molecular
probes lies in the occurrence of countless nonspecific inter-
actions between the probe and false targets along with the
genuine ligand–receptor interactions.[13] A major cause of an
increase in such nonspecific binding can be addressed in the
hydrophobic properties of a molecular probe that is brought
by the functional units in the probe itself, such as a reactive
In comparatively large bioactive molecules, such as oligo- or
polypeptidehormones: phytosulfokaine,[10] Pep1,[11] and
RALF,[12] it is not so difficult to introduce troublesome func-
tional units without decreasing bioactivity because the phar-
macophore and these functional units can be separated spa-
tially because of their large molecular size. However, the
same approach cannot be adopted for a small ligand, such
as potassium isolespedezate (1). Then, the molecular design
of the probe is the most important issue to be scrutinized:
the more compact the probe is, the better the bioactivity
that may be expected. On the other hand, the larger the mo-
lecular tag becomes, the less nonspecific binding can occur.
This is because the larger tag of many functional groups can
be expected to have more sites for molecular recognition,
thus leading to highly selective recognition with less nonspe-
cific binding. These conflicting requirements can be made
compatible by the exclusion of functional units, such as tags,
from the probe. The use of a copper-catalyzed azide alkyne
cycloaddition (CuAAC)[13]-based stepwise tagging approach
for the target protein[14,15] would be the best way to put this
into practice. In this approach, a CMP containing an azide
handle can be used for an initial affinity-based labeling of
target, and subsequent tagging by CuAAC with a large mo-
lecular tag will give a tagged target protein. The use of
CMP will enable the retention of moderate affinity between
the ligand CMP and its target, in contrast to immobilizing
the ligand on affinity beads, to secure selective and efficient
binding even for only moderately bioactive ligands.
We designed and synthesized azide-containing lespede-
zates 2–4[15] as CMPs (Scheme 2a). The arrangement of af-
finity-labeling functionalities and an azide handle was deter-
mined according to previous SAR studies[16] on 1: both func-
tional units were introduced within the glycone moiety be-
cause structural modification of the glycon moiety of 1 little
Chem. Asian J. 2011, 6, 3286 – 3297
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
3287

M. Ueda et al.
FULL PAPERS
target labeling. This result sug-
gested that avoiding a large
molecular tag enables develop-
ing a CMP that is expected to
retain affinity with its target
without a severe decrease in
bioactivity.
A
FLAG peptide[20] was
chosen as a highly potent mo-
lecular tag for affinity-based
exploration and purification of
the target protein of 1. The
FLAG peptide is an artificial
octapeptide,
DYKDDDDK,
and highly selective anti-
a
FLAG antibody with little
nonspecific binding is commer-
cially available. We also devel-
oped the alkyne-connected
FLAG tag 9[21] (Scheme 2b),
which exerts good reactivity
against azide
2
(Scheme 3,
Figure 1) under in vitro
CuAAC conditions. This large
tag is introduced after the
azide labeling of target protein
in the living cell.
Our previous results suggest-
ed that fluorescein-labeled po-
tassium isolespedezate (13, see
Figure 7) is localized in motor
cells of C. mimosoides pulvi-
ni;[4] thus, we concluded that
the target cell for 1 is motor
cells. We chose isolated motor
cell protoplasts as a material
for target exploration because
use of isolated protoplasts
greatly decreased nonspecific
binding.[6] Additionally, we
chose C. obtusifolia instead of
C. mimosoides as the plant ma-
terial. We expected that the
larger pulvini of C. obtusifolia
Scheme 2. Syntheses of a) azide-CMPs (2–4) and b) the FLAG-alkyne unit (9). DIPEA=N,N-diisopropylethyl-
amine; HBTU=[2-(1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate]; HFIP=hexa-
fluoroisopropanol; HOBt=1-hydroxybenzotriazole; TFA=trifluoroacetic acid.
affects its bioactivity. Probe 2 was effective for leaf opening
of C. obtusifolia leaves, with 80% of the leaves opening at
1ꢁ10À5 m, which was as strong as the response to naturally
occurring 1 (effective for leaf opening of 87% of leaves at
1ꢁ10À5 m). Probes 3 and 4 were effective for leaf opening of
40% of leaves at 1ꢁ10À5 m, which is half as strong as that of
1 (40% at 5ꢁ10À6 m). A carbon electrophile (iodoacetyl[17] in
2) and photoreactive groups (benzophenone[18] in 3 and tri-
fluoromethydiazirine[19] in 4) were selected as reactive func-
tionalities for protein labeling. The remarkably stronger ac-
tivity of CMPs 2–4 than that of a biotin-connected probe
(1ꢁ10À4 m: for structure),[5] used for the detection of 180 and
210 kDa membrane targets, promises successful specific
(1ꢁ3 mm) than that of C. mimosoides (0.5ꢁ0.5 mm) would
facilitate the collection of source materials for motor cell
protoplasts. Potassium isolespedezate (1) was effective for
leaf opening of both plants (5ꢁ10À6 for C. obtusifolia and at
1ꢁ10À6 m for C. mimosoides). Motor cell protoplasts were
prepared according to the method of Satter and co-work-
ers[21] and were used for the following examination. Next,
we moved to target exploration of 1 by using living motor
cell protoplasts.
We developed a highly effective stepwise tagging strategy
for target exploration of a bioactive metabolite. CMPs 2–4
were used for in vivo introduction of an azide handle into
the target protein (Figure 2a). Subsequently, the FLAG-tag
3288
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 3286 – 3297

A Cytosolic Target for Potassium Isolespedezate
Scheme 3. CuAAC between 2 and 9 to give 11 with fragment ions found in MS/MS analysis.
Figure 1. HPLC profile of CuAAC between 2 and 9: Upper: FLAG unit 9, middle: CMP 2, bottom: CuAAC after 1 h (column: Deverosil ODS HG-3,
detection: 280 nm, mobile phase: linear gradient in 48 min from 2% CH3CN aq. containing 0.1% HCO2H to 98% CH3CN aq. containing 0.1%
HCO2H).
was coupled with the azide-labeled target protein by a
CuAAC reaction. The microsomal fraction yielded no
FLAG-tagged protein, whereas the cytosolic fraction gave a
FLAG-tagged protein at 83 kDa when the iodoacetyl-type
probe 2 was used (Figure 2b, lane 2). This band disappeared
in a competitive inhibition experiment in the presence of
excess isolespedezate 1 (Figure 2b, lane 3), and thus was
confirmed as a putative cytosolic target protein of 1, named
cytosolic target protein for lespedezate (CTPL). Considering
the difference found in bioactivity among the three probes
(80% at 1ꢁ10À5 m for 2, and 40% at 1ꢁ10À5 m for 3 and 4),
no discriminating difference can be predicted for their affin-
ity with CTPL. Thus, the difference in results in Figure 2 can
be attributed to the superiority of the iodoacetyl functionali-
ty over a photoaffinity tag. The advantage of iodoacetyl-
based protein labeling would lie in its high yield in protein
labeling[22] relative to photoaffinity-based protein labeling,
Chem. Asian J. 2011, 6, 3286 – 3297
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
3289

M. Ueda et al.
FULL PAPERS
Figure 3. a) Immunoprecipitation (IP) of CTPL by using
a panel of
probes with linkers of various lengths and structure. b) SDS-PAGE analy-
ses of immunoprecipitated CTPL by using probe 20; left: chemilumines-
cence (CL) detection, right: silver-staining (Ag-stain) detection.
Figure 2. a) Procedures of the CMP approach combined with step-wise
FLAG-tagging of target protein. b) SDS-PAGE analyses of separated mi-
crosomal and cytosolic fractions.
which usually results in around a few percent of labeling
yield percent.[23]
tagging of the target by FLAG was successfully applied to
the target identification of 1 and led to the discovery of the
novel cytosolic target CTPL.
The isolated CTPL was analyzed by de novo sequencing
by using LCMS/MS after in-gel digestion by trypsin. The se-
These results suggested that our CMP approach combined
with stepwise FLAG-tagging, first forming a covalent bond
between ligand and target by using CMP with bioactivity
maintained, followed by CuAAC-based introduction of a
FLAG tag, is a potent approach to target exploration of li-
gands with moderate bioactivity. The resulting tagged CTPL
was then immunoprecipitated by using a corresponding anti-
FLAG antibody immobilized on beads. Among probes (14–
20, see Schemes S1–S5 in the Supporting Information) of
various lengths and structures of linker-connected tags, a
3XFLAG peptide antigen as a molecular tag (20)[24] gave
the most efficient IP result (Figure 3). The linker structure
and length strongly affect IP results.[24] As a result, stepwise
quences
GNASVPAMEMTK,
ALGVETVPVLVGPV-
SYLLLSKPAK, KLNLPILPTTTIGSFPQTLDLR, IQEEL-
DIDVLVHGEPER, GMLTGPVTILNWSFVR, and AGIT-
VIQIDEAALR, YGAGIGPGVYDIHSPR were identified
by using a MASCOT search of the NCBInr database to con-
clude that CTPL is highly similar to MetE (a 5-methyltetra-
hydropteroyl- triglutamate-homocysteinemethyltransferase)
from Ricinus communis (Figure 4a).[25] MetE is a cobala-
mine-independent methyltransferase participating in the
final step of methionine biosynthesis, and Arabidopsis thali-
3290
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 3286 – 3297

A Cytosolic Target for Potassium Isolespedezate
Figure 4. a) Comparison between the partial sequence of CTPL and Ricinus communis MetE. b) Chemiluminescence detection of CTPL by using anti-
MetE antibody raised against Lotus japonicus MetE.
Membrane Targets of 180 and 210 kDa
ana has three isoforms with distinct expression profiles.[26]
The sequence corresponded well with a consensus sequence
for MetE (Figure 4a, colored in yellow).[27] Additionally, an
anti-MetE antibody raised against L. japonicus MetE cross-
reacted with isolated CTPL (Figure 4b). Thus, CTPL was
identified as a MetE homologue from C. obtusifolia and
named CoMetE (Cassia obtusifolia MetE).
In the current study, previously observed membrane targets
(180 and 210 kDa)[5] were not found in the microsomal frac-
tion. This may be attributed to the different quantities of
microsomal fractions used in these experiments: in the pre-
vious study, the microsomal fraction, which was directly pre-
pared from chopped pulvini, contained as much as 134 mg
FLAG-tagged CTPL (=CoMetE) was detected only in
the precipitate obtained by CMP-IP (Figure 5a). This indi-
protein
(membrane
ATPase
activity:[28]
0.67 mmolmgÀ1 protein/min).[5] On the other hand, a micro-
somal fraction prepared from isolated motor cell protoplasts
contained as little as 2.84 mg protein (membrane ATPase ac-
tivity: 2.81 mmolmgÀ1 protein/min) because the azide label-
ing using motor cells was carried out with few protoplasts
due to the low yield during protoplast preparation. A large
amount of motor cells was lost and disrupted during enzyme
digestion during protoplast preparation. The use of proto-
plasts was effective for decreasing nonspecific binding by
avoiding incorporation of other plant materials, such as vas-
cular bundles or epidermis, but the overwhelming difference
in the quantity of the microsomal fraction would make it
difficult to find trace amounts of a membrane target of 180
and 210 kDa.
Figure 5. Chemiluminescence detection of IP results by using a) anti-
FLAG antibody or b) anti-MetE antibody: from the left, supernatant
without CuAAC (cell lyzate as a control), precipitate obtained by CMP-
IP, and supernatant after CMP-IP.
cates that all of FLAG-tagged CTPL was recovered by IP.
However, non-FLAG-tagged CTPL, which can be detected
by using anti-MetE antibody, was also found in supernatant
after CMP-IP (Figure 5b, right). Overall isolation yield of
CTPL (=CoMetE) by the CMP-IP strategy was estimated
to be 20% by comparing the content of CoMetE in each su-
pernatant without CuAAC (motor cell lyzate as a control),
precipitate obtained by CMP-IP, and supernatant after
CMP-IP (Figure 5a,b).
CoMetE
Target identification of a bioactive metabolite requires ap-
propriate validation. In general, the most reliable approach
would be to use a knock-out or -down strain impaired in the
response to the bioactive metabolite. However, no genetic
mutant has been obtained in Cassia plants, and it is impossi-
ble to use model legumes, such as Lotus japonicus, instead
Chem. Asian J. 2011, 6, 3286 – 3297
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
3291

M. Ueda et al.
FULL PAPERS
of Cassia because potassium isolespedezate (1) is not effec-
tive for leaf movement of L. japonicus, even at 3ꢁ10À4 m.
Additionally, to the best of our knowledge, no inhibitor is
known for MetE. We thus carried out validation by using
unique features in the bioactivity of 1.
Fortunately, potassium isolespedezate (1) has two unique
features in its bioactivity. One is a unique SAR pattern:[16]
modification of the aglycone moiety severely lowers its leaf-
opening activity, whereas modification of the glycone
moiety does not affect it at all. The other is its unique
genus-specific leaf-opening activity:[2,29] potassium isolespe-
dezate (1) is effective only for plants belonging to the genus
Cassia. The correspondence between these unique features
of 1 and its affinity with CoMetE was examined to validate
MetE as a target.
Binding affinity between CoMetE and biologically active
(2)/inactive (21, 22) CMPs showed a good correspondence
Scheme 4. Syntheses of biologically inactive CMPs (21 and 22).
with this unique SAR pattern of 1 in cell-based binding
studies (Figure 6a). We synthesized biologically inactive
CMPs 21 and 22 (Scheme 4), in which the aglycone moiety
was modified to exert no leaf-opening activity even at 1ꢁ
10À3 m. CoMetE in Cassia motor cells was not labeled by
either of these probes, and a clear relationship between leaf-
opening activity and affinity with CoMetE was demonstrat-
ed. Additionally, CMP 2 gave a FLAG-labeled 83 kDa pro-
tein in the motor cells of both C. mimosoides and C. obtusi-
folia, whereas no labeled protein was found in L. japonicus
(Figure 6b). This result also reflects the other unique feature
of 1, its genus-specific leaf-opening activity. Thus, these two
validation experiments gave results in accordance with the
unique features of bioactivity in 1.
CoMetE and Cytosolic Localization
Two of the three isoforms of A. thaliana MetE (AtMetE)
were revealed to be localized in the cytosol.[26] CoMetE was
also isolated from the cytosol fraction. This means that
highly polar potassium isolespedezate (1) can be transported
into the cytosol through the plasma membrane of the motor
cell to bind CoMetE.
Waring[30] reported that logD and molecular weight are
the most important factors in determining the membrane
permeability of small molecules based on a study by using a
panel of over 9500 compounds. According to the study,
logD>1.1 for MW: 300–350, logD>1.7 for MW: 350–400,
logD>3.1 for MW: 400–450, logD>3.4 for MW: 450–500,
logD>4.5 for MW: >500, are expected criteria for success-
ful membrane permeability. The calculated logD[36] of 1
(MW: 342, logD=À4.5 at pH 7, À2.5 at pH 5) and of CMP
2 (MW: 534, logD=À3.7 at pH 7, À1.7 at pH 5) are lower
than the criteria necessary for passive (diffusion-driven)
transport. Additionally, the localization of fluorescein-la-
Figure 6. Experiments for the varidation of CTPL as a target protein of
1: a) A clear correlation was observed between the leaf-opening activity
of CMPs (2, 21, and 22) and affinity with CTPL. b) Genus-specific label-
ing of CTPL was observed by using CMP 2.
3292
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 3286 – 3297

A Cytosolic Target for Potassium Isolespedezate
beled 1 (13)[4] in motor cell protoplasts showed that 13 can
be transported into the cytosol through the plasma mem-
brane (Figure 7). This may be due to transport by a carrier-
mediated facilitated diffusion process, perhaps involving a
transporter protein, because 13 at a pH below 7 lost its
membrane permeability due to dissociation of the carboxyl-
ate to form a salt.
Experimental Section
Compound 6
Compound 5[5] (56.5 mg, 0.103 mmol) was dissolved in ethanol (2 mL) and
hydrazine monohydrate (40 mL, 0.8 mmol) was added to this solution.
After overnight stirring at RT under an Ar atmosphere, the reaction mix-
ture was evaporated to dryness. The residue was dissolved in EtOAc,
washed with H2O, and dried over anhydrous Na2SO4. After filtration, the
filtrate was evaporated to give the crude amine (49.5 mg). Iodoacetic
acid N-hydroxysuccinimide ester[35] (5.1 mg, 17.9 mmol) was added to a
N,N-dimethylformamide (DMF; 200 mL) solution of the crude amine
(6.3 mg) and the mixture was stirred for 6 h at RT under an Ar atmos-
phere. The reaction was quenched by the addition of AcOH (3 drops).
After drying, the residue was purified by pTLC (CHCl3/MeOH 10:1) to
give
6
(4.0 mg, 6.5 mmol 50%) in two steps. 1H NMR (500 MHz,
CD3OD): d=7.68 (d, J=8.5 Hz, 2H), 6.82 (s, 1H), 6.76 (d, J=8.5 Hz,
2H), 5.32 (d, J=8.5 Hz, 1H), 4.21 (dd, J=11.0, 8.5 Hz, 1H), 3.77 (s, 2H),
3.74 (dd, J=3.5, 1.0 Hz, 1H), 3.67 (dd, J=11.0, 3.5 Hz, 1H), 3.59 (ddd,
J=8.5, 4.0, 1.0 Hz, 1H), 3.50 (dd, J=12.5, 8.5 Hz, 1H), 3.23 (dd, J=12.5,
4.0 Hz, 1H), 1.55 ppm (s, 9H); 13C NMR (125 MHz, CD3OD): d=172.0,
165.0, 159.5, 140.4, 133.7, 126.2, 125.8, 116.2, 100.5, 82.8, 76.2, 73.0, 70.0,
54.9, 52.4, 28.5, À1.58 ppm; IR (film): n˜ =3295, 2978, 2927, 2102, 1698,
1606, 1584, 1556, 1512, 1369, 1316, 1276, 1254, 1158, 1115, 1072, 837,
755 cmÀ1; [a]2D7 =À52.7 (c=0.10 in MeOH); HRMS (ESI, positive): m/z:
calcd for C21H28IN4NaO8: 613.0771 [M+Na]+; found: 613.0800.
Figure 7. Cytosolic localization of fluorescein-labeled 1 (13) in motor cell
protoplasts.
The previously identified 180 and 210 kDa membrane
proteins were reported to be able to recognize the structure
related to the leaf-opening activity of 1,[5] and thus, can be
considered potential transporters with the appropriate selec-
tivity. Additionally, the detection of cytosolic CoMetE by
using 2 implies that CMP 2 can be transported into the cyto-
sol with putative carrier protein, as naturally occurring 1.
This means that our CMP strategy was successful in main-
taining transport ability and bioactivity of naturally occur-
ring 1.
Compound 2
TFA (1 mL) was added to compound 6 (12.5 mg, 21.2 mmol). After stir-
ring for 5 min at RT under an Ar atmosphere, the reaction mixture was
dried in vacuo. The residue was purified by HPLC (25% CH3CN aq. con-
taining 0.1% TFA) to give
2
(6.3 mg, 11.3 mmol, 56%). 1H NMR
(500 MHz, CD3OD): d=7.71 (d, J=8.5 Hz, 2H), 7.00 (s, 1H), 6.76 (d,
J=8.5 Hz, 2H), 5.26 (d, J=8.5 Hz, 1H), 4.21 (dd, J=10.5, 8.5 Hz, 1H),
3.77 (s, 2H), 3.74 (dd, J=3.0, 0.5 Hz, 1H), 3.66 (dd, J=10.5, 3.0 Hz, 1H),
3.60 (ddd, J=8.0, 4.5, 0.5 Hz, 1H), 3.49 (dd, J=13.0, 8.0 Hz, 1H),
3.23 ppm (dd, J=13.0, 4.5 Hz, 1H); 13C NMR (125 MHz, CD3OD): d=
172.2, 167.6, 159.7, 139.8, 133.9, 127.2, 126.2, 116.1, 101.2, 76.1, 73.3, 70.0,
55.0, 52.3, À1.71 ppm; [a]2D3 =À24.6 (c=0.10 in MeOH); HRMS (ESI,
positive): m/z: calcd for C17H20IN4NaO8: 557.0145 [M+Na]+; found:
557.0157.
Conclusions
Our CMP strategy was proven to be effective for the detec-
tion of a moderate affinity target for which AC cannot be
applied successfully. Then, our CMP approach can be con-
sidered as a powerful alternative of AC when the ligand has
moderate bioactivity. Additionally, it is also emphasized that
our CMP approach enables the tracing of the fate of bioac-
tive metabolites to detect cytosolic targets because CMP can
maintain transport ability and bioactivity equivalent to natu-
rally occurring 1.
MetE is a ubiquitous enzyme participating in the biosyn-
thesis of methionine. However, in Cassia motor cells, the pu-
tative membrane transporter recognizes potassium isolespe-
dezate (1) and transports it into the cytosol by a carrier-
mediated diffusion process. Recently, it was revealed that a
leaf-closing substance of Samanea saman worked as a trigger
for motor cell shrinkage by activation of ion channels.[31] On
the other hand, the leaf-opening factor 1 can be involved in
a different mechanism and induce leaf opening in Cassia
plants.
Compound 7
Compound 5[5] (32.8 mg, 59.4 mmol) was dissolved in ethanol (2 mL) and
hydrazine monohydrate (40 mL, 0.8 mmol) was added to this solution.
After overnight stirring at RT under an Ar atmosphere, the reaction mix-
ture was evaporated to dryness. The residue was purified by RP-pTLC
(water/acetonitrile 1:1, containing 1% TFA) to give the crude amine
(33.0 mg, crude). 4-Bromomethylbenzophenone (32.5 mg, 0.119 mmol)
and TEA (41.3 mL, 0.297 mmol) were added to the DMF solution
(1 mL) of the crude amine (33.0 mg, crude). After stirring for 2.5 h at RT
under an Ar atmosphere, the reaction was quenched with AcOH. After
being dried, the residue was purified by pTLC (CHCl3/MeOH 10:1) to
give 7 (16.3 mg, 26.5 mmol 45% in two steps). 1H NMR (500 MHz,
CD3OD): d=7.75–7.67 (m, 6H), 7.61 (t, J=6.5 Hz, 1H), 7.55 (d, J=
8.5 Hz, 2H), 7.49 (t, J=7.8 Hz, 2H), 6.92 (s, 1H), 6.70 (d, J=8.5 Hz,
2H), 5.13 (d, J=8.0 Hz, 1H), 4.22 (d, J=13.5 Hz, 1H), 4.19 (d, J=
13.5 Hz, 1H), 3.70 (d, J=3.0 Hz, 1H), 3.61 (dd, J=10.0, 3.5 Hz, 1H),
3.58 (m, 1H), 3.44 (dd, J=13.0, 8.3, 1H), 3.17 (dd, J=13.0, 4.3, 1H), 3.07
(dd, J=10.0, 8.3, 1H), 1.55 ppm (s, 9H); 13C NMR (125 MHz, CD3OD):
d=198.5, 165.2, 159.7, 147.0, 140.6, 139.0, 137.4, 133.7, 131.3, 130.9, 129.8,
129.5, 126.9, 126.1, 116.1, 103.8, 82.9, 76.1, 74.3, 70.0, 60.5, 53.7, 52.4,
28.5 ppm; IR (film): n˜ =3381, 2978, 2931, 2102, 1698, 1651, 1606, 1512,
1369, 1317, 1279, 1159 cmÀ1; [a]D22 =À43.8 (c=0.10 in CH3OH); HRMS
(ESI, positive): m/z: calcd for C33H37N4O8: 617.2606 [M+H]+; found:
617.2602.
Chem. Asian J. 2011, 6, 3286 – 3297
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
3293

M. Ueda et al.
FULL PAPERS
Compound 3
supernatant was removed. The collected precipitate was re-suspended in
MeOH (2 mL) and the supernatant was decanted. This sequence was re-
peated twice and the resin was dried by flushing with N2. The resultant
resin was suspended in CH2Cl2/HFIP/TFA 5:5:90 (1 mL) and stirred for
1 h at RT under an Ar atmosphere. After the reaction mixture was con-
centrated in vacuo, purification by HPLC (linear gradient over 40 min
from 20% CH3CN aq. containing 0.1% TFA to 40% CH3CN aq. con-
taining 0.1% TFA) gave 9 (3.3 mg, 2.5 mmol, 50%); HRMS (ESI, posi-
TFA (1 mL) was added to compound 7 (8.2 mg, 13.2 mmol). After stirring
for 1 h at RT under an Ar atmosphere, the reaction mixture was dried in
vacuo. The residue was purified by RP-pTLC (water/acetonitrile 1:1, con-
1
taining 1% TFA) to give 3 (5.0 mg, 9.9 mmol, 75%). H NMR (500 MHz,
CD3OD): d=7.74–7.30 (m, 11H), 6.89 (s, 1H), 6.62 (d, J=8.5 Hz, 2H),
4.96 (d, J=8.5 Hz, 1H), 4.36 (d, J=13.0 Hz, 1H), 4.28 (d, J=13.0 Hz,
1H), 3.96 (dd, J=11.0, 3.0 Hz, 1H), 3.75 (d, J=3.0 Hz, 1H), 3.63 (m,
1H), 3.42 (dd, J=12.5, 7.8, 1H), 3.35 (dd, J=11.0, 8.5, 1H), 3.26 ppm
(dd, J=12.5, 4.0, 1H); 13C NMR (125 MHz, CD3OD): d=197.7, 169.3,
160.7, 139.8, 138.4, 137.1, 134.4, 134.0, 131.5, 131.4, 130.9, 130.1, 129.6,
126.5, 125.1, 116.2, 101.0, 76.2, 70.7, 69.6, 61.3, 52.3, 52.0 ppm; IR (film):
tive): m/z: calcd for C54H66F6N10O21: 652.2149 [M+2H]2+
; found:
652.2149.
Compound 11 (in vitro CuAAC)
Azide 2 (2ꢁ10À8 mol) was dissolved in ligation buffer (90 mL, 25 mm
HEPES (pH 8.0)) and then FLAG unit 9 (2ꢁ10À8 mol in 90 mL ligation
n˜ =3304, 2924, 2102, 1658, 1607, 1512, 1366, 1279, 1064, 703 cmÀ1; [a]D22
=
À38.3 (c=0.10 in CH3OH); HRMS (ESI, positive): m/z: calcd for
buffer) with [CuACTHNUTRGNEUNG
(CH3CN)4]PF6 (2ꢁ10À8 mol) and 12 (2ꢁ10À8 mol) in liga-
C29H28N4NaO8: 583.1799 [M+Na]+; found: 583.1801.
tion buffer (10 mL) containing 10% DMSO were added. This solution
was incubated with gentle agitation of 50 rpm for 1 h at 308C. The
CuAAC reaction was monitored by LCMS/MS analysis (linear gradient
over 48 min from 2% CH3CN aq. containing 0.1% HCOOH to 98%
CH3CN aq. containing 0.1% HCOOH, flow rate: 0.2 mLminÀ1, detection
280 nm coupled with positive mode-ESI by using an esquire 4000, Bruker
Compound 8
Compound 5[5] (28.0 mg, 50.7 mmol) was dissolved in ethanol (1 mL) and
hydrazine monohydrate (20 mL, 0.8 mmol) was added to this solution.
After overnight stirring at RT under an Ar atmosphere, the reaction mix-
ture was evaporated to dryness. The residue was dissolved in EtOAc,
washed with H2O, and dried over anhydrous Na2SO4. After filtration, the
filtrate was evaporated to give crude amine (26.1 mg). 3-[4-(Bromome-
thyl)phenyl]-3-(trifluoromethyl)diazirine[34] (28.2 mg, 0.101 mmol) and
TEA (21.2 mL, 0.152 mmol) were added to a solution of the crude amine
(26.1 mg, crude) in DMF (0.5 mL) and the mixture was stirred for 1 h at
08C under an Ar atmosphere. Then, the mixture was slowly allowed to
stand to RT. After overnight stirring, the reaction was quenched with
AcOH. After being dried, the residue was purified by pTLC (CHCl3/
MeOH 10:1) to give 8 (12.5 mg, 16.7 mmol 33%) in two steps. 1H NMR
(500 MHz, CD3OD): d=7.67 (d, J=8.5 Hz, 2H), 7.48 (d, J=8.5 Hz, 2H),
7.16 (d, J=8.5 Hz, 2H), 6.91 (s, 1H), 6.71 (d, J=8.5 Hz, 2H), 5.11 (d, J=
8.5 Hz, 1H), 4.15 (d, J=13.5 Hz, 1H), 4.09 (d, J=13.5 Hz, 1H), 3.68 (d,
J=3.5 Hz, 1H), 3.56 (m, 2H), 3.44 (dd, J=13.0, 8.5 Hz, 1H), 3.16 (dd,
J=13.0, 4.0 Hz, 1H), 3.01 (d, J=8.5 Hz, 1H), 1.54 ppm (s, 9H);
13C NMR (125 MHz, CD3OD): d=165.2, 159.7, 143.9, 140.5, 133.7, 130.4,
128.5, 127.5, 127.0, 126.1, 123.5 (q, J=261.2 Hz), 116.1, 103.8, 82.9, 76.1,
74.2, 69.9, 60.5, 53.4, 52.4, 29.5 (q, J=40.2 Hz), 28.5 ppm; IR (film): n˜ =
3312, 2980, 2932, 2103, 1697, 1606, 1585, 1513, 1455, 1394, 1370, 1346,
1316, 1277, 1254, 1233, 1158, 1114, 1067, 759 cmÀ1; [a]D27 =À11.1 (c=0.10
in MeOH); HRMS (ESI, positive): m/z: calcd for C28H32F3N6O7: 621.2285
[M+H]+; found: 621.2259.
Daltonics). HRMS (ESI, positive): m/z: calcd for C71H84F6I1N14O29
:
1837.4477 [M+H]+; found: 1837.4490.
Compound 24
Compound 23[5] (22.1 mg, 39.8 mmol) was dissolved in ethanol (2 mL) and
hydrazine monohydrate (40 mL, 0.8 mmol) was added to this solution.
After overnight stirring at RT under an Ar atmosphere, the reaction mix-
ture was evaporated to dryness. The residue was dissolved in EtOAc,
washed with H2O, and dried over anhydrous Na2SO4. After filtration, the
filtrate was evaporated to give crude amine (22.3 mg). Iodoacetic acid N-
hydroxysuccinimide ester[34] (22.6 mg, 79.8 mmol) was added to a DMF
(1 mL) solution of the crude amine (22.3 mg) and the mixture was stirred
for 45 min at RT under an Ar atmosphere. The reaction was quenched
with AcOH (3 drops). After being dried, the residue was purified by
pTLC (CHCl3/MeOH 10:1) to give 24 (11.1 mg, 18.0 mmol 47% in two
steps). 1H NMR (500 MHz, CD3OD): d=7.04 (d, J=8.5 Hz, 2H), 6.65
(d, J=8.5 Hz, 2H), 4.53 (d, J=8.5 Hz, 1H), 4.51 (t, J=6.0 Hz, 1H), 3.92
(dd, J=10.5, 8.5 Hz, 1H), 3.75 (s, 2H), 3.72 (d, J=3.0 Hz, 1H), 3.67–3.63
(m, 3H), 3.22 (q, J=8.5 Hz, 1H), 2.97 (dd, J=14.0, 6.0 Hz, 1H), 2.90
(dd, J=14.0, 6.0 Hz, 1H), 1.33 ppm (s, 9H); 13C NMR (125 MHz,
CD3OD): d=172.3, 171.8, 157.1, 132.0, 128.4, 115.8, 101.0, 82.9, 77.7,
76.2, 72.7, 70.2, 54.8, 52.5, 39.3, 28.3, À1.21 ppm; IR (film): n˜ =3319,
2979, 2931, 2100, 1720, 1656, 1517, 1369, 1250, 1154, 1114, 1073, 757 cmÀ1
;
Compound 4
[a]2D7 =À24.4 (c=0.10 in MeOH); HRMS (ESI, positive): m/z: calcd for
Compound 8 (8.2 mg, 13.2 mmol) was dissolved in TFA (1 mL). After stir-
ring for 1 h at RT under an Ar atmosphere, this solution was dried in
vacuo. The residue was purified by HPLC (40% CH3CN aq. containing
0.1% TFA) to give 4 (6.8 mg, 12.0 mmol, 91%). 1H NMR (500 MHz,
CD3OD): d=7.75 (d, J=8.5 Hz, 2H), 7.65 (d, J=8.5 Hz, 2H), 7.31 (d,
J=8.5 Hz, 2H), 7.24 (s, 1H), 6.74 (d, J=8.5 Hz, 2H), 5.26 (d, J=8.5 Hz,
1H), 4.68 (d, J=13.0 Hz, 1H), 4.48 (d, J=13.0 Hz, 1H), 4.03 (dd, J=
11.0, 3.0 Hz, 1H), 3.83 (dd, J=3.0, 0.5 Hz, 1H), 3.69 (ddd, J=8.0, 5.0,
0.5 Hz, 1H), 3.52 (dd, J=11.0, 8.5 Hz, 1H), 3.46 (dd, J=13.0, 8.0 Hz,
1H), 3.26 ppm (dd, J=13.0, 5.0 Hz, 1H); 13C NMR (125 MHz, CD3OD):
d=169.2, 160.7, 134.8, 134.3, 132.8, 132.1, 131.3, 129.8, 128.2, 125.1, 123.4
(q, J=273.1 Hz), 116.1, 101.0, 76.5, 76.2, 70.6, 69.6, 61.4, 52.0, 29.3 ppm
(q, J=40.3 Hz); IR (film): n˜ =3311, 2926, 2855, 2358, 2106, 1672, 1606,
1513, 1439, 1374, 1261, 1232, 1175, 1156, 1079, 940, 838, 801, 763, 722,
505, 458, 444, 418 cmÀ1; [a]D27 =+20.1 (c=0.10 in MeOH); HRMS (ESI,
positive): m/z: calcd for C24H24F3N6O7: 565.1659 [M+H]+; found:
565.1651.
C21H29IN4NaO8: 615.0928 [M+Na]+; found: 615.0935.
Compound 21
TFA (1 mL) was added to compound 24 (5.3 mg, 8.95 mmol). After stir-
ring for 5 min at RT under an Ar atmosphere, the reaction mixture was
dried in vacuo. The residue was purified by HPLC (25% CH3CN aq. con-
taining 0.1% TFA) to give 21 (4.6 mg, 8.2 mmol 96%). 1H NMR
(500 MHz, CD3OD): d=7.05 (d, J=8.5 Hz, 2H), 6.64 (d, J=8.5 Hz, 2H),
4.58 (t, J=6.0 Hz, 1H), 4.54 (d, J=8.0 Hz, 1H), 3.91 (dd, J=11.0, 8.0 Hz,
1H), 3.71 (d, J=3.0 Hz, 1H), 3.70 (s, 2H), 3.64 (m, 3H), 3.23 (q, J=
8.5 Hz, 1H), 2.99 (dd, J=14.5, 6.0 Hz, 1H), 2.96 ppm (dd, J=14.5,
6.0 Hz, 1H); 13C NMR (125 MHz, CD3OD): d=181.2, 174.9, 157.1, 131.9,
128.6, 115.9, 101.2, 77.8, 76.1, 72.9, 70.2, 54.9, 52.5, 39.1, À1.41 ppm; IR
(film): n˜ =3290, 2924, 2358, 2342, 2105, 1733, 1653, 1558, 1541, 1517,
1457, 1122, 1069, 470, 456, 426 cmÀ1; [a]D27 =À12.7 (c=0.10 in MeOH);
HRMS (ESI, positive): m/z: calcd for C17H21IN4NaO8: 559.0302
[M+Na]+; found: 559.0317.
Compound 9
Compound 26
Protected-FLAG (12.5 mg on beads, 5 mmol) was suspended in DMF
(200 mL). Compound 10[36] (15.4 mg, 50 mmol), HOBt (6.8 mg, 50 mmol),
iPr2NEt (8.7 mL, 50 mmol), and HBTU (19.0 mg, 50 mmol) were added to
this suspension. After shaking for 8 h with a shaker at RT under an Ar at-
mosphere, the reaction mixture was diluted with MeOH (2 mL) and the
K2CO3 (0.8 mg, 5.93 mmol) and MeI (3.7 mL, 59.3 mmol) were added to a
DMF (0.6 mL) solution of 25[15] (3.5 mg, 5.93 mmol). After the mixture
was stirred for 4 h at RT under an Ar atmosphere, the reaction was
quenched by the addition of water (10 mL). After adding EtOAc
3294
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 3286 – 3297

A Cytosolic Target for Potassium Isolespedezate
(10 mL), the organic layer was extracted and dried over anhydrous
Na2SO4. After filtration, the filtrate was evaporated. The residue was pu-
rified by pTLC (CHCl3/MeOH 10:1) to give 26 (2.0 mg, 3.2 mmol, 56%).
1H NMR (500 MHz, CD3OD): d=7.68 (d, J=9.0 Hz, 2H), 6.81 (d, J=
9.0 Hz, 2H), 6.74 (s, 1H), 5.26 (d, J=8.5 Hz, 1H), 4.13 (dd, J=11.0,
8.5 Hz, 1H), 3.72 (s, 3H), 3.69 (s, 2H), 3.65 (dd, J=3.5, 1.0 Hz, 1H), 3.58
(dd, J=11.0, 3.5 Hz, 1H), 3.51 (ddd, J=8.5, 4.0, 1.0 Hz, 1H), 3.42 (dd,
J=13.0, 8.5 Hz, 1H), 3.14 (dd, J=13.0, 4.0 Hz, 1H), 1.47 ppm (s, 9H);
13C NMR (125 MHz, CD3OD): d=172.0, 164.8, 161.6, 141.0, 133.5, 127.4,
125.2, 114.8, 100.5, 82.9, 76.3, 73.0, 70.1, 55.7, 54.8, 52.5, 28.5, À1.67 ppm;
IR (film): n˜ =3309, 3084, 2931, 2102, 1704, 1604, 1551, 1510, 1369, 1320,
1303, 1254, 1159, 1119, 1074, 1034, 757 cmÀ1; [a]D27 =À60.7 (c=0.10 in
MeOH); HRMS (ESI, positive): m/z: calcd for C22H29IN4NaO8: 627.0928
[M+Na]+; found: 627.0951.
Preparation of Protoplasts
Protoplasts were isolated from primary pulvini of eight- to 12-week-old
C. obtusifolia seedlings according to the method of Gorton and Satter15)
with modifications. About 30 pulvini were chopped finely with a sharp
razor blade and soaked in a predigestion solution (1 mL, 50 mm MES
(pH 5.5), 0.3m sorbitol, 0.2% bovine serum albumin (BSA), 8 mm CaCl2,
Gamborgꢂs B-5) for 10 min. The osmotic pressure of the predigestion so-
lution was then raised to the desired level (0.6m sorbitol) for plasmolysis
in three steps over 30 min with osmotic adjustment solution (50 mm MES
(pH 5.5), 4.0m sorbitol, 0.2% BSA, 8 mm CaCl2, Gamborgꢂs B-5). The so-
lution was replaced with an enzyme solution (2 mL, 50 mm MES (pH 5.5)
buffer containing 0.6m sorbitol, 0.3% (w/v) Macerozyme R-10, and 1%
(w/v) cellulase Onozuka RS). This suspension containing the protoplasts
was placed under reduced atmospheric pressure for 2 min. Protoplasts
were released by incubation of the tissue in the solution with gentle incu-
bation at 50 rpm for 3 h at 308C. Debris was removed by filtration of the
protoplast suspension through a 50 mm nylon mesh. Protoplasts in the fil-
trate were sedimented at 110ꢁg for 7 min and the supernatant was dis-
carded. The protoplasts were re-suspended in HEPES buffer (4 mL,
25 mm HEPES-KOH (pH 7.0), 0.6m sorbitol, 1 tablet/500 mL complete
protease inhibitor cocktail (Roche, 1 tablet/500 mL) was added and the
mixture was sedimented again at 100ꢁg for 7 min. This washing proce-
dure was repeated three times. The yield of protoplasts was 4ꢁ105.
Compound 22
TFA (1 mL) was added to compound 26 (2.5 mg, 4.1 mmol). After stirring
for 5 min at RT under an Ar atmosphere, the reaction mixture was dried
in vacuo. The residue was purified by HPLC (35% CH3CN aq. contain-
ing 0.1% TFA) to give 25 (1.8 mg, 3.1 mmol, 80%). 1H NMR (500 MHz,
CD3OD): d=7.81 (d, J=9.0 Hz, 2H), 7.00 (s, 1H), 6.90 (J=9.0 Hz, 2H),
5.30 (d, J=8.5 Hz, 1H), 4.22 (dd, J=10.5, 8.5 Hz, 1H), 3.81 (s, 3H), 3.77
(s, 2H), 3.74 (dd, J=3.5, 1.0 Hz, 1H), 3.66 (dd, J=10.5, 3.5 Hz, 1H), 3.61
(ddd, J=8.0, 4.5, 1.0 Hz, 1H), 3.49 (dd, J=13.0, 8.0 Hz, 1H), 3.23 ppm
(dd, J=13.0, 4.5 Hz, 1H); 13C NMR (125 MHz, CD3OD): d=172.1,
167.4, 161.8, 140.5, 133.7, 127.3, 126.6, 114.7, 101.1, 76.2, 73.3, 70.0, 55.7,
55.0, 52.3, À1.70 ppm; IR (film): n˜ =3309, 2925, 2854, 2104, 1700, 1604,
FLAG-Tagging of CTPL by using Protoplasts of C. occidentalis Motor
Cells
Each sample (6ꢁ10À10 mol of 2, 6ꢁ10À10 mol of 2 with 6ꢁ10À8 mol of 1,
6ꢁ10À10 mol of 3, 6ꢁ10À10 mol of 4) was added to a suspension of proto-
plasts (about 5ꢁ104 protoplasts in 20 mL of wash solution (25 mm
HEPES-KOH (pH 7), complete protease inhibitor cocktail (1 tablet/
500 mL), 0.6m sorbitol)). Biologically inactive analogues (21, 22: 6ꢁ
10À10 mol) were also used instead of 2, 3, and 4 in control experiments
with a suspension of protoplasts (about 3ꢁ104). Cross-linking with CTPL
was conducted at 48C as follows: cross-linking by using iodoacetoamide-
type probes (2) was achieved by incubation for 5 min, whereas photo-
cross-linking by using photoaffinity-type probes (3 and 4) was achieved
by UV irradiation (365 nm, 1820 mWcmÀ2) at a distance of 5 cm from UV
lamp for 30 min with ice cooling. After cross-linking, the protoplasts
were sedimented three times by centrifugation (100ꢁg, 7 min, 48C) with
wash solution (100 mL) to remove excess unreacted probe. The sediment
was resuspended in 25 mm HEPES buffer (pH 7) (6 mL); FLAG unit 9
(1ꢁ10À9 mol) in 25 mm HEPES buffer (2 mL, pH 8) was added to this sus-
pension. The CuAAC reaction was started by the addition of 25 mm
HEPES buffer (2 mL, pH 8), 5% DMSO solution containing ligand 12
1558, 1511, 1458, 1424, 1303, 1254, 1175, 1121, 1069, 1030, 883, 831 cmÀ1
;
[a]2D7 =À41.4 (c=0.10 in MeOH); HRMS (ESI, positive): m/z: calcd for
C18H21IN4NaO8: 571.0302 [M+Na]+; found: 571.0327.
Plant Materials and Bioassay
C. obtusifolia and C. mimosoides were grown in a Biotron LPH-1000S
chamber (Nippon Medical & Chemical Instruments) under a condition of
16 h light from 5:00 to 21:00 at 32Æ28C with 80% humidity and
150 mmolmÀ2 s PAR/8 h dark from 21:00 to 5:00 at 22Æ28C with 80%
humidity.
Eight- to 12-week-old C. obtusifolia leaves were used for the bioassay.
All of the procedures were carried out in a Biotron chamber because the
results were strongly affected by changes in temperature, climate, and
other environmental factors. The leaves were detached from the stem
with a sharp razor blade in water around 16:00 h. One leaf was placed in
a 5 mL glass tube with H2O and allowed to stand overnight. The leaves,
which opened again in the morning (around 9:30 h), were then used for
the bioassay. Each test solution was transferred into test tubes with a mi-
cropipette around 10:00 h. The LOF activity was determined as the con-
centration at which over 40% of sample leaves had opened at 21:00 h.
ACTHNUTRGNEUNG
(1ꢁ10À8 mol), and [Cu(CH3CN)4]PF6 (1ꢁ10À8 mol, Aldrich). After incu-
bation for 1 h at 308C, extraction buffer (40 mL, 25 mm Tris-MES buffer
(pH 7.2) containing 0.25m sucrose, 3 mm EDTA-2K, 2.5 mm DTT, and
complete protease inhibitor cocktail (1 tablet/mL)) was added to this solu-
tion and the protoplasts were crushed by ultrasonification (DU-200 with
UR-20P, TOMY Industry). Centrifugation of the lysate twice (1st: 3,000ꢁ
g, 15 min, 48C; 2nd: 100000ꢁg, 1 h, 48C) gave a crude cytosolic homoge-
nate as the supernatant (this fraction was designated crude CTPL) with a
crude membrane fraction as the pellet. After lyophilization, electropho-
resis buffer (15 mL, 0.3m Tris-HCl buffer (pH 6.8) containing 10%
sodium dodecyl sulfate (SDS), 30% glycerol, and 9.3% di-
On the other hand, leaves of L. japonicus were bioassayed according to
the procedures in references [33]. The bioactivity of a fraction was as-
sessed by the status of leaves at 21:00 h. Potassium isolespedezate (1;
Table 1) was not effective in this assay even at 1ꢁ10À4 (0/4) or 3ꢁ
10À4 molLÀ1 (0/4), whereas potassium (R)-eucomate,[33] a LOF of L. japo-
nicus, was effective at 1ꢁ10À5 molLÀ1 (5/10).
thiothreitol (DTT) was added to each fraction and each solu-
Table 1. Leaf-opening activities of naturally occurring 1 and CMPs (2–4, 21, and 22).
tion was heated at 958C for 5 min. The reaction mixtures
were analyzed by SDS-PAGE on Ready Gel J 5–10% poly-
acrylamide gradient gels (Bio-Rad Laboratories) with a mo-
lecular weight marker. After western blotting by using
Hybond-P PVDF membrane (GE Healthcare), the membrane
was exposed to a rabbit anti-FLAG antibody (Delta Biolabs)
or an anti-AtMetE polyclonal antibody raised against bacteri-
ally expressed L. japonicus MetE. Subsequently, the mem-
brane was exposed to goat anti-rabbit IgG-HRP (Santa Cruz
Biotechnology). These procedures were carried out by using a
BenchPro TM4100 (Invitrogen). The protein bands were de-
tected by chemiluminescence by using an ECL Advance west-
Compounds
Number of opened leaves
at 21:00/sample leaves
Percentage of LOF activity
blank
0/24
13/15
4/10
8/10
4/10
5/10
0/5
0
87
40
80
40
50
0
1 (1ꢁ10À5 molLÀ1
1 (5ꢁ10À6 molLÀ1
2 (1ꢁ10À5 molLÀ1
3 (1ꢁ10À5 molLÀ1
4 (1ꢁ10À5 molLÀ1
)
)
)
)
)
21 (1ꢁ10À3 molLÀ1
22 (1ꢁ10À3 molLÀ1
)
)
0/5
0
Chem. Asian J. 2011, 6, 3286 – 3297
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
3295

M. Ueda et al.
FULL PAPERS
ern blotting detection kit (GE Healthcare UK) with an LAS-4000 IR
multicolor Image Analyzer (Fujifilm Corp.).
Preparation of Protoplasts from C. mimosoides and L. japonicus
Protoplasts were isolated from primary pulvini of eight- to 12-week-old
C. mimosoides and L. japonicus seedlings according to the method of
Gorton and Satter[21] with modifications.
Immunoprecipitation of CTPL by using Protoplasts or Cytosolic
Fractions
About 120 of C. mimosoides pulvini or 200 of L. japonicus pulvini were
chopped finely with a sharp razor blade and soaked in a predigestion so-
lution (1 mL, 50 mm MES (pH 5.5), 0.3m sorbitol, 0.2% BSA, 8 mm
CaCl2, Gamborgꢂs B-5) for 10 min. The osmotic pressure of the prediges-
tion solution was then raised to the desired level (0.6m sorbitol) for plas-
molysis in three steps over 30 min with osmotic adjustment solution
(50 mm MES (pH 5.5), 4.0m sorbitol, 0.2% BSA, 8 mm CaCl2, Gamborgꢂs
B-5). The solution was replaced by an enzyme solution (2 mL, 50 mm
MES (pH 5.5) buffer containing 0.6m sorbitol, 0.3% (w/v) Macerozyme
R-10, and 1% (w/v) cellulase Onozuka RS). This solution was placed
under reduced atmospheric pressure for 2 min. Protoplasts were released
by incubation of the tissue in the solution with gentle incubation at
50 rpm for 3 h at 308C. Debris was removed by filtration of the proto-
plast suspension through a 50 mm nylon mesh. Protoplasts in the filtrate
were sedimented at 100ꢁg for 7 min and the supernatant was discarded.
The protoplasts were resuspended in HEPES buffer (4 mL, 25 mm
HEPES-KOH (pH 7.0), 0.6m sorbitol, 1 complete tablet/500 mL), and
sedimented again at 100ꢁg for 7 min. This washing procedure was re-
peated three times. The yield of protoplasts was 4ꢁ104 from 120 C. mim-
osoides pulvini or 200 L. japonicus pulvini.
FLAG tagging by using protoplasts and subsequent immunoprecipitation
were carried out by probe 2. After FLAG-tagging by 2 and subsequent
ultracentrifugation (10000ꢁg, 15 min, 48C), the cytosolic fraction (50 mL)
obtained as a supernatant was mixed with four volumes of acetone and
allowed to stand for 1 h at À808C. After centrifugation (10000ꢁg,
15 min, 48C), the pellet was dissolved in lysis buffer (100 mL, 50 mm Tris-
HCl (pH 7.4) containing 150 mm NaCl, 1 mm EDTA, and 1% Triton X-
100). The solution was incubated with anti-FLAG antibody-connected
beads (FLAG M Purification Kit, Sigma–Aldrich) for 12 h at 48C. After
centrifugation (5,000ꢁg, 48C, 30 sec), the supernatant was lyophilized,
whereas the precipitated affinity beads were resuspended in wash buffer
(100 mL, 50 mm Tris-HCl buffer (pH 7.4) containing 150 mm NaCl) and
centrifuged (5,000ꢁg, 48C, 30 s). This washing process was repeated five
times. The precipitated affinity beads were suspended in wash buffer, and
FLAG-tagged CTPL was eluted from the affinity beads by incubation
with 10 mg of 3ꢁFLAG (Sigma) at RT for 10 min. After centrifugation
(5,000ꢁg, 48C, 30 sec), the supernatant was lyophilized to give purified
CTPL. The purified CTPL was analyzed by using an ECL Advance west-
ern blotting detection kit (GE Healthcare UK) or a silver staining kit
(Wako Pure Chemical Industries). The amount of purified CTPL and the
ratio of supernatant to precipitate in Figure 4 were obtained by using an
LAS-4000 Bioimager (Fujifilm). The amount of purified CTPL was esti-
mated from a comparison of the CTPL band with a standard protein on
silver staining.
FLAG tagging of CTPL by using Motor Cells of C. Mimosoides or L.
Japonicus
Iodoacetamide probe 2 (6ꢁ10À10 mol) was added to a suspension of C.
obtusifolia, C. mimosoides, or L. japonicus protoplasts (about 1ꢁ104) in
wash solution (20 mL, 25 mm HEPES-KOH (pH 7), complete protease in-
hibitor cocktail (1 tablet/500 mL), 0.6m sorbitol). The remainder of the
procedure was carried out as for C. obtusifolia cells.
Immunoprecipitation by probes 14–20 was carried out by using the cyto-
solic fraction instead of living protoplasts. Protoplasts (1ꢁ104) in extrac-
tion buffer (50 mL, 25 mm Tris-MES (pH 7.2), 0.25m sucrose, 3 mm
EDTA-2K, 2.5 mm DTT, 1 complete tablet/50 mL) were crushed by ultra-
sonification (DU-200 with UR-20P, TOMY Industry). The solution was
centrifuged twice (1st: 3,000ꢁg, 15 min, 48C, 2nd: 100000ꢁg, 1 h, 48C),
and each probe (14–20, 5ꢁ10À11 mol each) was added to the cytosolic
fraction obtained as a supernatant. After cross-linking by incubation for
1 h at 308C, the FLAG-tagged cytosolic fraction was mixed with four vol-
umes of acetone and allowed to stand for 1 h at À808C, then treated ac-
cording to the above procedure. The yield of immunoprecipitated CTPL
was calculated by comparing the density of bands detected at 83 kDa:
the cytosolic fraction containing CMP-labeled CTPL was applied to IP
procedure without CuAAC reaction and the density of 83 kDa band de-
tected in supernatant was defined as control (100%). In this case, no
83 kDa band was detected in the IP fraction. On the other hand, the cy-
tosolic fraction containing CMP-labeled CTPL was applied to the IP
after the CuAAC reaction by using 9, and the density of 83 kDa bands in
both of the IP fractions and supernatant were compared with the control.
Acknowledgements
This work was supported by a Grant-in-Aid for Scientific Research from
MEXT (no. 20310128). We also thank the Experimental Station for Me-
dicinal Plant Studies, Tohoku University, for cultivation of plants, and
Mr. S. Park for carrying out bioassays by using L. japonicus.
[1] a) N. Jessani, B. F. Cravatt, Curr. Opin. Chem. Biol. 2004, 8, 54;
b) M. J. Evans, B. F. Cravatt, Chem. Rev. 2006, 106, 3279–3301; c) U.
Hillaert, M. Verdoes, B. I. Florea, A. Saragliadis, K. L. L. Habets, J.
Kuiper, S. V. Calenbergh, F. Ossendrop, G. A. van der Marel, C.
Driessen, H. S. Overkleeft, Angew. Chem. 2009, 121, 1657–1660;
Angew. Chem. Int. Ed. 2009, 48, 1629–1632.
Microsequencing of CTPL
[2]
Immunoprecipitation by using probe 20 was run at a 10-fold scale by
using 1ꢁ105 protoplasts. The resulting supernatant and eluent from the
affinity beads were lyophilized and dissolved in 20 mL of sample buffer
each. An aliquot of 2 mL each was separated by SDS-PAGE followed by
chemiluminescence detection and the remaining 18 mL was separated by
SDS-PAGE with silver-staining detection (1 ngꢁ5 lanes). Protein bands
detected by silver staining with a Silver Stain MS Kit (Wako Pure Chemi-
cal Industries) were excised from the gel, destained, and in-gel digested
with trypsin at 358C for 20 h. The peptide fragments were analyzed on a
Q-Tof2 positive mode nanoflow-LC ESIMS (Waters Micromass, UK, ca-
pillary voltage: 1.8 kV, collision energy: 20–56 eV) equipped with an l-
column ODS (f 0.1ꢁ50 mm). The linear gradient conditions were: 95%
A:5% B to 45% A:55% B over 0–35 min (solvent A: 2% CH3CN aq.
containing 0.1% HCOOH, solvent B: 90% CH3CN aq. containing 0.1%
HCOOH).
H. Shigemori, E. Miyoshi, Y. Shizuri, S. Yamamura, Tetrahedron
Lett. 1989, 30, 6389–6392.
[3] T. Ohnuki, M. Ueda, S. Yamamura, Tetrahedron 1998, 54, 12173–
12184.
[4] T. Sugimoto, Y. Wada, S. Yamamura, M. Ueda, Tetrahedron 2001,
57, 9817–9825.
[5] T. Fujii, Y. Manabe, T. Sugimoto, M. Ueda, Tetrahedron 2005, 61,
7874–7893.
[6] B. J. Leslie, P. J. Hergenrother, Chem. Soc. Rev. 2008, 37, 1347–1360.
[7] a) M. W. Harding, A. Galat, D. E. Uehling, S. L. Schreiber, Nature
1989, 341, 758–760; b) J. Taunton, C. A. Hassig, S. L. Schreiber, Sci-
ence 1996, 272, 408–411.
[8] a) L. Burdine, T. Kodadek, Chem. Biol. 2004, 11, 593–597; b) S.
Sato, A. Murata, T. Shirakawa, M. Uesugi, Chem. Biol. 2010, 17,
616–623.
3296
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 3286 – 3297

A Cytosolic Target for Potassium Isolespedezate
[9] a) J. L. Erion, J. E. Fox, Plant Physiol. 1981, 67, 156–162; b) S.
Sakai, T. Hanagata, Plant Cell Physiol. 1983, 24, 685–693; c) S.
Sakai, Plant Cell Physiol. 1985, 26, 185–192.
[23] M. Hashimoto, Y. Hatanaka, Eur. J. Org. Chem. 2008, 2513–2523.
[24] M. Ueda, Y. Manabe, M. Mukai, Bioorg. Med. Chem. Lett. 2011, 21,
1359–1362.
[10] Y. Matsubayashi, Y. Sakagami, J. Biol. Chem. 2000, 275, 15520–
15525.
[11] Y. Yamaguchi, G. Pearce, C. A. Ryan, Proc. Natl. Acad. Sci. USA
2006, 103, 10104–10109.
[12] J. M. Scheer, G. Pearce, C. A. Ryan, Planta 2005, 221, 667–674.
[13] a) C. W. Tornøe, C. Christensen, M. Meldal, J. Org. Chem. 2002, 67,
3057–3064; b) V. V. Rostovtsev, L. G. Green, V. V. Fokin, K. B.
Sharpless, Angew. Chem. 2002, 114, 2708–2711; Angew. Chem. Int.
Ed. 2002, 41, 2596–2599.
[14] L. Ballell, K. J. Alink, M. Slijper, C. Versluis, R. M. Liskamp, R. J.
Pieters, ChemBioChem 2005, 6, 291–295.
[15] Y. Manabe, M. Mukai, S. Ito, N. Kato, M. Ueda, Chem. Commun.
2010, 46, 469–471; all of the physical data on 3–5 and 6 were pre-
sented in the Supporting Information of this paper.
[16] a) H. Shigemori, N. Sakai, E. Miyoshi, Y. Shizuri, S. Yamamura, Tet-
rahedron 1990, 46, 383–394; b) M. Ueda, Y. Sawai, S. Yamamura,
Tetrahedron 1999, 55, 10925–10936.
[17] C. Huggins, E. V. Jensen, J. Biol. Chem. 1949, 179, 645–654.
[18] a) R. E. Galardy, L. C. Craig, J. D. Jamieson, M. P. Printz, J. Biol.
Chem. 1974, 249, 3510–3518; b) G. Dorman, G. D. Prestwich, Bio-
chemistry 1994, 33, 5661–5673.
[25] a) S. Ravanel, B. Gakiere, D. Job, R. Douce, Proc. Natl. Acad. Sci.
USA 1998, 95, 7805–7812; b) http://ca.expasy.org/cgi-bin/nice-
prot.pl?B9RQ33.
[26] S. Ravanel, M. A. Block, P. Rippert, S. Jabrin, G. Curien, F. Rebeille,
R. Dounce, J. Biol. Chem. 2004, 279, 22548–22557.
[27] R. Pejchal, M. L. Ludwig, PLoS Biol. 2005, 2, 254–265.
[28] R. P. Sanderson, A. H. de Boer, T. L. Lomax, R. E. Cleland, Plant
Physiol. 1987, 85, 693.
[29] M. Ueda, H. Shigemori, N. Sata, S. Yamamura, Phytochemistry
2000, 53, 39–44.
[30] M. J. Waring, Bioorg. Med. Chem. Lett. 2009, 19, 2844–2851.
[31] a) I. V. Tetko, G. I. Poda, J. Med. Chem. 2004, 47, 5601–5604;
b) I. V. Tetko, P. Bruneau, J. Pharm. Sci. 2004, 93, 3103–3110.
[32] a) Y. Nakamura, A. Mithçfer, E. Kombrink, W. Boland, S. Hama-
moto, N. Uozumi, K. Tohma, M. Ueda, Plant Physiol. 2011, 155, in
press; b) Y. Nakamura, R. Miyatake, M. Ueda, Angew. Chem. 2008,
120, 7399–7402; Angew. Chem. Int. Ed. 2008, 47, 7289–7292.
[33] M. Okada, S. Park, T. Koshizawa, M. Ueda, Tetrahedron 2009, 65,
2136–2141.
[34] H. Nakashima, M. Hashimoto, Y. Sadakane, T. Tomohiro, Y. Hata-
naka, J. Am. Chem. Soc. 2006, 128, 15092–15093.
[19] Y. Hatanaka, M. Hashimoto, H. Kurihara, H. Nakayama, Y. Kanao-
ka, J. Org. Chem. 1994, 59, 383–395.
[35] I. Schechter, E. Clerrici, E. Zazepitzki, Eur. J. Biochem. 1971, 18,
561–572.
[20] A. Einhauer, A. Jungbauer, J. Biochem. Biophys. Methods 2001, 49,
455–465.
[36] Y. Nakamura, S. Inomata, M. Ebine, Y. Manabe, I. Iwakura, M.
Ueda, Org. Biomol. Chem. 2011, 9, 83–85.
[21] a) H. Gorton, R. L. Satter, Plant Physiol. 1984, 76, 680–684; b) N.
Moran, D. Fox, R. L. Satter, Plant Physiol. 1990, 94, 424–431.
[22] A. Saghatelian, B. F. Cravatt, Nat. Chem. Biol. 2005, 1, 130–142.
Received: April 18, 2011
Published online: August 3, 2011
Chem. Asian J. 2011, 6, 3286 – 3297
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
3297