Structure-activity studies on the fluorescent indicator in a displacement assay for the screening of small molecules binding to RNA

Article abstract of DOI:10.1002/chem.201103932

A series of xanthone and thioxanthone derivatives with aminoalkoxy substituents were synthesized as fluorescent indicators for a displacement assay in the study of small-molecule-RNA interactions. The RNA-binding properties of these molecules were investigated in terms of the improved binding selectivity to the loop region in the RNA secondary structure relative to 2,7-bis(2-aminoethoxy)xanthone (X2S) by fluorimetric titration and displacement assay. An 11-mer double-stranded RNA and a hairpin RNA mimicking the stem loop IIB of Rev response element (RRE) RNA of HIV-1 mRNA were used. The X2S derivatives with longer aminoalkyl substituents showed a higher affinity to the double-stranded RNA than the parent molecule. Introduction of a methyl group on the aminoethoxy moiety of X2S effectively modulated the selectivity to the RNA secondary structure. Methyl group substitution at the C1' position suppressed the binding to the loop regions. Substitution with two methyl groups on the amino nitrogen atom resulted in reducing the affinity to the double-stranded region by a factor of 40 %. The effect of methyl substitution on the amino nitrogen atom was also observed for a thioxanthone derivative. Titration experiments, however, suggested that thioxanthone derivatives showed a more prominent tendency of multiple binding to RNA than xanthone derivatives. The selectivity index calculated from the affinity to the double-stranded and loop regions suggested that the N,N-dimethyl derivative of X2S would be suitable for the screening of small molecules binding to RRE. Indicating RNA binding: A fluorescent indicator displacement assay using 2,7-bis(aminoethoxy)xanthone (X2S) as the indicator is reported for the study of small-molecule-RNA interactions (see figure). The substituents at the two alkoxy moieties modulate the binding selectivity. The X2S derivative with N,N-dimethylaminoethoxy substituents is the best fluorescent indicator in the screening of molecules binding to Rev response element (RRE). Copyright

Full text of DOI:10.1002/chem.201103932

FULL PAPER
DOI: 10.1002/chem.201103932
Structure–Activity Studies on the Fluorescent Indicator in a Displacement
Assay for the Screening of Small Molecules Binding to RNA
Shiori Umemoto,[a] Seongwang Im,[a] Jinhua Zhang,[a] Masaki Hagihara,[a]
Asako Murata,[a] Yasue Harada,[a] Takeo Fukuzumi,[a] Takahiro Wazaki,[b]
Shin-ichi Sasaoka,[b] and Kazuhiko Nakatani*[a]
Abstract: A series of xanthone and
thioxanthone derivatives with amino-
alkoxy substituents were synthesized as
fluorescent indicators for a displace-
ment assay in the study of small-mole-
cule–RNA interactions. The RNA-
binding properties of these molecules
were investigated in terms of the im-
proved binding selectivity to the loop
region in the RNA secondary structure
relative to 2,7-bis(2-aminoethoxy)xan-
thone (X2S) by fluorimetric titration
and displacement assay. An 11-mer
double-stranded RNA and a hairpin
RNA mimicking the stem loop IIB of
Rev response element (RRE) RNA of
HIV-1 mRNA were used. The X2S de-
rivatives with longer aminoalkyl sub-
stituents showed a higher affinity to
the double-stranded RNA than the
parent molecule. Introduction of
a methyl group on the aminoethoxy
moiety of X2S effectively modulated
the selectivity to the RNA secondary
structure. Methyl group substitution at
the C1’ position suppressed the binding
to the loop regions. Substitution with
two methyl groups on the amino nitro-
gen atom resulted in reducing the affin-
ity to the double-stranded region by
a factor of 40%. The effect of methyl
substitution on the amino nitrogen
atom was also observed for a thioxan-
thone derivative. Titration experiments,
however, suggested that thioxanthone
derivatives showed a more prominent
tendency of multiple binding to RNA
than xanthone derivatives. The selectiv-
ity index calculated from the affinity to
the double-stranded and loop regions
suggested that the N,N-dimethyl deriv-
ative of X2S would be suitable for the
screening of small molecules binding to
RRE.
Keywords: displacement assay · flu-
orescent probes · RNA · small mol-
ecules · xanthones
Introduction
a change of indicator fluorescence is called a fluorescent in-
dicator displacement (FID) assay (Figure 1). In the depicted
FID assay, the fluorescence of the indicator is quenched
upon binding to the receptor. Displacement of the bound in-
dicator with test molecules results in the increase of the
fluorescence intensity emitted from the free unbound indica-
tor.
A displacement assay is a method to investigate the interac-
tions between two molecules, for example, ligand and recep-
tor, by monitoring a signal produced from the indicator, an
auxiliary molecule added as a reporter.[1–5] When the indica-
tor and test molecules compete for the same binding site on
the receptor, the test molecule can replace the indicator
bound to the receptor. The necessary character of the indi-
cator in a displacement assay is to change signals unique to
the indicator, such as absorption and fluorescence, depend-
ing on its bound and free states. In particular, an assay using
[a] Dr. S. Umemoto, S. Im, Dr. J. Zhang, Dr. M. Hagihara,
Dr. A. Murata, Y. Harada, Dr. T. Fukuzumi, Prof. Dr. K. Nakatani
Department of Regulatory Bioorganic Chemistry
The Institute of Scientific and Industrial Research, Osaka University
8-1 Mihogaoka, Ibaraki 567-0047 (Japan)
Figure 1. Illustration of the FID assay.
Among methods to investigate molecular interactions, the
displacement assay is characterized by a simple procedure.
Neither labeling nor immobilization of receptor on the solid
support is necessary for the assay. Covalent modification
with fluorescent labels and surface immobilization intrinsi-
cally hold possibilities leading to an unnatural folding and
structure of receptors. The subsequent assay may investigate
the binding of molecules to the unnatural structures. In con-
Fax : (+81)6-6879-8459
E-mail: nakatani@sanken.osaka-u.ac.jp
[b] T. Wazaki, S. Sasaoka
Research & Development Division
Nitto Kasei Co., Ltd.
17-14 Nishiawaji 3-Chome, Higashiyodogawa-Ku, Osaka 533-0031
(Japan)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201103932.
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

trast, the displacement assay can be performed in a homoge-
neous solution in the presence of the indicator without any
covalent modification of the target. Another important char-
acteristic of the displacement assay is scalability of the assay
format from small to large compound libraries without
much difficulty. The preparation of a large quantity of fluo-
rescently labeled proteins and nucleic acids is still problem-
atic and high in cost, but a displacement assay uses the natu-
ral unlabeled proteins and nucleic acids, which could be sup-
plied by cell-free translation and chemical synthesis or tran-
scription.
The drawback to the displacement assay is that the ob-
served signal is not directly produced from the binding of
the test molecule to the receptor, but indirectly produced
from the dissociation of the indicator bound to the receptor.
The quality of the displacement assay regarding the accura-
cy in the molecular interaction, therefore, would be high
when the indicator and the test ligand competitively bind to
the receptor. In a real situation, however, the binding
among three molecules (e.g., indicator, receptor, and test
molecule) might involve unnecessary binary (e.g., indicator–
ligand) and ternary (receptor–indicator–ligand) molecular
interactions besides the expected ligand–receptor interac-
tion. Nonspecific binding of the indicator to the receptor
leading to multiple binding of the indicator also decreases
the quality of the assay data. In other words, the quality of
the displacement assay depends on the binding properties of
the indicator to the receptor.
We have recently reported an FID assay using 2,7-bis(2-
aminoethoxy)xanthone (X2S, 3a) as a fluorescent indicator
for the study of the ligand–RNA interaction,[5] because
small molecules interacting with biologically important
RNA structures have been envisioned as potential drug can-
didates.[6] We found that the fluorescence of X2S is intense
when the X2S is free in solution but is quenched when it is
bound to RNA. Fluorimetric titration of X2S with double-
stranded RNA (dsRNA), RNA with a single nucleotide
bulge, and a model RNA of stem loop IIb in Rev response
element (RRE) of HIV-1 mRNA[7] showed that X2S favora-
bly bound to stem loops with some selectivity. Displacement
experiments with a model peptide of Rev protein showed
that X2S competitively bound to RRE with Rev peptide.
RRE is an important secondary structure of HIV-1 mRNA,
and is the site for the binding to Rev protein. Rev binding
to RRE is essential for the nuclear export of the full-length
HIV-1 mRNA to cytoplasm without splicing.[7,8] Small mole-
cules binding to RRE and suppressing the Rev binding were
anticipated as potential therapeutics for HIV-1 prolifera-
tion.[9] X2S was applied to the screening of small molecular
ligands binding to RRE from the LOPAC1280 chemical li-
brary. The displacement assay against RRE with X2S as
a fluorescent indicator showed that mitoxantrones[10] and
sanguinarine[11] were suggested to bind to RRE, and the
binding was confirmed separately by isothermal titration cal-
orimetry analysis.
stem loop region than X2S would improve the assay quality
by eliminating the nonspecific binders, for example, mole-
cules binding to dsRNA. Because the loop structures are
more prominent sites of protein binding than the double-
stranded region,[12] fluorescent indicators having weak affini-
ty to the double-stranded region and a binding preference
to the stem loop structures would be suitable for the screen-
ing of small molecules that potentially modulate the RNA–
protein interaction. We realized that another issue in FID is
the overlap of the fluorescence spectra of X2S with those of
library compounds. A library compound with emitting fluo-
rescence that is overlapped with the fluorescence of the in-
dicator could not be correctly assessed by the FID assay.
Under normal screening conditions, the concentration of the
library compounds is usually much higher than that of the
indicator. In our screening of the LOPAC1280 chemical li-
brary with X2S, we first eliminated 59 compounds that emit
fluorescence about three times stronger than X2S. On the
basis of these discussions on our previous screening studies
by FID assay, herein we report studies on a series of xan-
thone derivatives to look for improved binding selectivity to
the stem loop region relative to X2S and thioxanthone de-
rivatives emitting fluorescence at a longer wavelength.[13]
We found that introduction of two methyl groups on a pri-
mary amino group in X2S to give a dimethylamino group
significantly improved the selective binding to the loop
region over the binding to the double-stranded region, and
that the thioxanthone derivative showing fluorescence with
an approximately 45 nm shift to a longer wavelength was
also useful for the FID assay.
Results and Discussion
Synthesis of X2S derivatives: We synthesized a series of
X2S derivatives with respect to the length of the linker,
from three methylene groups 3b (n=3) to six methylene
groups 3e (n=6; Figure 2). Syntheses of these compounds
were carried out by Mitsunobu reaction of 2,7-dihydroxy-
xanthone (1)[14] with the appropriate tert-butoxycarbonyl
(Boc)-protected amino alcohols followed by deprotection
with an acidic treatment (Scheme 1). By using the same pro-
cedure, we also synthesized X2S derivatives 3g and 3 f, con-
taining a methyl substituent at the aminoethoxy moiety as
a mixture of diastereoisomers, and N,N-disubstituted 4, con-
taining two methyl groups at the amino nitrogen atom.
Thioxanthone is known to emit fluorescence at a longer
wavelength than xanthone. Therefore, 2,7-Bis(2-aminoeth-
oxy)thioxanthone (5) and its N,N-dimethyl derivative 6 were
synthesized from 2,7-dihydroxythioxanthone[15] by Mitsuno-
bu reaction. NMR spectra of the synthesized compounds are
shown in the Supporting Information.
Photochemical properties of X2S derivatives: The absorp-
tion and emission spectra of X2S derivatives were measured
in a buffered solution at pH 7. The X2S derivatives having
different linker lengths (3a–e) showed a slight deviation in
We learned from the screening studies with X2S that fluo-
rescent indicators with higher structural selectivity to the
&
2
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Indicator in the Displacement Assay of Small Molecules Binding to RNA
FULL PAPER
Table 1. Absorption and fluorescence properties of xanthone and thio-
xanthone derivatives.
3a
3b
3c
3d
3e
3 f
3g
4
5
6
lex
370
372
457
372
463
376
464
377
464
370
455
370
452
370
451
424
496
424
493
lem 450
ff
0.73 0.70 0.68 0.67 0.65 0.75 0.76 0.66 0.55 0.49
Figure 2. The xanthone and thioxanthone derivatives studied as indicators
for the FID assay.
Figure 3. Normalized fluorescence spectra of a) xanthones 3a–4 and
b) thioxanthones 5 and 6 in sodium cacodylate buffer (pH 7, 10 mm) and
NaCl (100 mm). Data for 3a are shown in b) for comparison.
highest among these five compounds. Introduction of the
methyl group into the linker did not result in a significant
shift of the absorption and emission wavelengths. Com-
pounds 3 f and 3g with methyl substitution at the C1’ and
C2’ position in the linker, respectively, showed absorption
and emission maxima at 370 and 455 nm for 3 f and 370 and
452 nm for 3g. N,N-Dimethyl derivative 4 also showed simi-
lar spectral properties. The fluorescence quantum yield of
3 f and 3g was 0.75 and 0.76, respectively, showing a slightly
higher value than 3a, although substitution with two methyl
groups at nitrogen in compound 4 led to a decrease in the
fluorescence quantum yield to 0.66. Thioxanthone derivative
5 and its N,N-dimethyl derivative 6 showed almost identical
spectra with the absorption maximum at 424 nm and emis-
Scheme 1. Synthesis of xanthones 3a–g and 4. i) Ph3P, diethyl azodicar-
boxylate (DEAD), Boc-protected amino alcohol, THF; ii) 4n HCl/
EtOAc.
the absorption maximum (Table 1). The absorption maxi-
mum of 3a was 370 nm, whereas that of 3e was 377 nm. The
emission maximum of X2S derivatives also showed a shift
depending on the linker length (Figure 3). The fluorescence
spectra of 3a showed an emission peak at 450 nm, whereas
3d and e showed the peak at the longest wavelength of
464 nm. The fluorescence quantum yield of 3a was 0.73, the
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

K. Nakatani et al.
sion maximum at around 495 nm. The fluorescence quantum
yields of 5 and 6 were 0.55 and 0.49, respectively.
Fluorescence quenching by RNA: Fluorimetric titration of
X2S derivatives was investigated with an 11-mer fully com-
plementary dsRNA and a hairpin RNA mimicking RRE
stem loop IIb (hpRNA; Figure 4).[4a] The relative fluores-
cence intensities of X2S derivatives of different linker
length (3a–e) titrated with dsRNA and hpRNA are shown
in Figure 5. The efficiency of fluorescence quenching of X2S
complexes of X2S derivatives with hpRNA at the double-
stranded and loop regions, respectively. Subscript t indicates
the total amount. For the fitting
of KdACTHNUGRTNEUNG(loop) in Equation (2), we
used the Kd(ds) value obtained
from the titration with dsRNA
by Equation (1).
KdACTHNGUTERN(UNG loop) [m]
was obtained with an R2 value
higher than 0.88 and was 5.2ꢁ
10ꢀ8 for 3a, 1.3ꢁ10ꢀ8 for 3b,
1.7ꢁ10ꢀ8 for 3c, 3.7ꢁ10ꢀ7 for
3d, and 1.1ꢁ10ꢀ6 for 3e. The
curve-fitted plots and R2 values
Figure 4. The sequence and secondary structure of dsRNA, hpRNA, and Rev peptide used in these studies.
derivatives by dsRNA was highly dependent on the linker
length (Figure 5a). In the presence of an equimolar amount
of dsRNA, the fluorescence intensity of 3a was still 73%
that of intact 3a, whereas that of 3e in the presence of
dsRNA was reduced to 21%. The fluorescence quenching
by dsRNA became more efficient as the linker length in-
creased. In contrast, the fluorescence intensities of these
compounds titrated with hpRNA were not markedly differ-
ent from each other.
The titration plots obtained with dsRNA were analyzed
by assuming static quenching with a 1:1 binding stoichiome-
try, represented by Equation (1), in which the relative fluo-
rescence intensity F/F0 is described by the free dsRNA con-
centration [RNA] and the apparent dissociation constant
Kd(ds) of the dsRNA–indicator complex. The titration data
with dsRNA were fitted to Equation (1) by the least-squares
curve-fitting method with the coefficient of determination
(R2) being higher than 0.96 (Figure 5). Kd(ds) values for the
complexes of dsRNA with 3a–e obtained by least-squares
curve fitting were 1.7ꢁ10ꢀ6, 5.7ꢁ10ꢀ7, 4.0ꢁ10ꢀ7, 1.8ꢁ10ꢀ7,
and 3.7ꢁ10ꢀ8 m, respectively (Table 2). The titration data ob-
tained with hpRNA, however, could not be fitted by Equa-
tion (1) with a good coefficient of determination, most likely
due to the binding of X2S derivatives simultaneously to
both the double-stranded and loop regions.
are shown in Figure 5. The analysis suggested that the affini-
ty to the double-stranded region increased upon increasing
the linker length, whereas the affinity to the loop region de-
creased by a factor of 20. The affinity of 3d and 3e to
dsRNA was similar to that to hpRNA, whereas the affinity
of 3e to dsRNA was about 30 times stronger than that to
hpRNA. The electrostatic association of two positively
charged ammonium groups to the negatively charged phos-
phate anions may rationalize the higher affinity of the X2S
derivative with a long alkyl chain. It is reasonable to assume
that as the distance between the two ammonium groups be-
comes longer, the probability of simultaneous electrostatic
interactions of the two charged ammonium groups with the
negatively charged phosphate would be higher.
!
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
F
F0
ꢀa þ 2 a2 ꢀ 3b
ꢀ2a3 þ 9ab ꢀ 27c
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
¼
cos arccos
3
3½X2Sꢁt
6
ða2 ꢀ 3bÞ
a ¼ KdðdsÞ þ KdðloopÞ þ 2½RNAꢁt ꢀ ½X2Sꢁt
b ¼ f½RNAꢁt ꢀ ½X2SꢁtgfKdðdsÞ þ KdðloopÞg þ KdðdsÞKdðloopÞ
c ¼ ꢀKdðdsÞKdðloopÞ½X2Sꢁt
ð2Þ
The effect of the methyl substituent at the side chain on
the fluorescence quenching was not apparent for dsRNA,
but was marked for hpRNA (Figure 6). Three methyl-substi-
tuted X2S derivatives 3 f, 3g, and 4 showed similar plots
when titrated with dsRNA. The lowest efficiency in fluores-
cence quenching by dsRNA was observed for the N,N-di-
methyl derivative 4. Both 3 f and 3g with a methyl substitu-
ent on the linker at the C1’ and C2’ position, respectively,
showed a similar plot to that obtained for 3a. In marked
contrast, the titration plot of 3 f obtained with hpRNA was
not much different from that obtained with dsRNA, whereas
derivatives 3g and 4 showed fluorescence quenching in
F
F0
KdðdsÞ
ð1Þ
¼
KdðdsÞ þ ½RNAꢁ
Alternatively, we used the binding model for the evalua-
tion of the binding to hpRNA,[16] in which the indicator
binds independently to the double-stranded and loop re-
gions with a 1:1 binding stoichiometry to each region. The
relative fluorescence intensity obtained by titrating with
hpRNA is described by Equation (2), in which Kd(ds) and
KdACHTUNGTRENNUNG(loop) represent the apparent dissociation constant of
&
4
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Indicator in the Displacement Assay of Small Molecules Binding to RNA
FULL PAPER
&
Figure 5. a–e) Relative fluorescence intensity of xanthone derivatives 3a–e (1 mm), respectively, titrated with various concentrations of dsRNA ( ) and
*
hpRNA ( ). Dotted and solid lines represent nonlinear least-squares-fitted curves derived by fitting the data with Equations (1) and (2), respectively.
The coefficient of determination R2 is shown in the graphs and summarized in Table 2.
a concentration-dependent manner with a reduced efficiency
compared to 3a. This suggested that the methyl group at the
C1’ position effectively hindered the binding of 3 f to the
loop region of hpRNA.
The curve fitting analysis of the titration data for 3 f, 3g,
and 4 with Equation (1) was successfully carried out, with
the coefficient of determination being higher than 0.93
(Figure 6), and showed that the affinity of 4 to the dsRNA
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

K. Nakatani et al.
Table 2. Apparent dissociation constants Kd [m] of the complex of each indicator and double-stranded region
and loop region of RNA.
was 2.5 times weaker than the
affinity of the parent molecule
3a, which showed the lowest af-
3a
3b
3c
3d
3e
3 f
3g
4
Kd(ds)
1.7ꢁ10ꢀ6 5.7ꢁ10ꢀ7 4.0ꢁ10ꢀ7 1.8ꢁ10ꢀ7 3.7ꢁ10ꢀ8 1.6ꢁ10ꢀ6 1.2ꢁ10ꢀ6 4.3ꢁ10ꢀ6 finity to dsRNA among the
R2[a]
G
R
G
E
G
G
R
ACHTUNGTRENNUNG
molecules discussed in these
studies. The apparent dissocia-
tion constant of the complex of
methyl-substituted X2S deriva-
tive at the loop region of
KdACHTUNGTRENNUNG(loop)
R2[a]
N
G
G
N
G
N
N
ACHTUNGTRENNUNG
relKd(ds)[b]
0.4
0.31
25
relK
(loop)
SI[c]
[a] R2 represents the coefficient of determination obtained for nonlinear least-squares fitting. [b] relKd(ds)=
[Kd(ds) of 3a]/Kd(ds). [c] SI (selectivity index)=Kd(ds)/KdA(loop).
hpRNA was obtained by curve
CTHUNGTRENNUNG
fitting analysis using Equa-
tion (2), with the R2 value being
higher than 0.95. The affinity of
3 f to the loop region of hpRNA (2.8ꢁ10ꢀ6 m) was almost
comparable to that to dsRNA (1.6ꢁ10ꢀ6 m), whereas the af-
finity of 3g and 4 to the loop region of hpRNA was 2.0ꢁ
10ꢀ7 and 1.7ꢁ10ꢀ7 m, respectively, and was 6 and 26 times
stronger than that to dsRNA (Table 2). The remarkable
effect of methyl substitution at the C1’ position in 3 f is most
likely ascribable to the steric hindrance of the methyl group.
Conformational analyses of 3 f suggested that the methyl
group of 3 f in low-energy conformers favors the position
perpendicular to the xanthone plane, thus making it less sus-
ceptible to intercalative binding into the base pairs in the
stem loop regions.
The binding characteristics of fluorescent indicators were
assessed by Kd(ds), K
which was calculated by the ratio of two dissociation con-
stants (Kd(ds)/KdA(loop)). The affinities of each compound to
dACHTUNGTRNE(NUNG loop), and the selectivity index (SI),
CTHUNGTRENNUNG
the double-stranded and loop regions were also compared
with the affinities of the parent compound 3a. Thus, relative
Kd(ds) [relKd(ds)] and KdACHTNUGTRNENUG(loop) [relKdHCATUNGTRENN(UGN loop)] values were
calculated by dividing the Kd of 3a by the Kd of each com-
pound. The relKd(ds) values clearly showed that the N,N-di-
methyl derivative 4 is the only compound showing a weaker
affinity to dsRNA than the parent compound 3a. The relKd-
AHCTUNGTREG(NNNU loop) was significantly affected by the linker structure.
Compounds 3b and 3c showed higher affinity to the loop
region than 3a, whereas compound 4 still retained about
one third of the affinity of 3a to the loop region. The selec-
tivity index of the affinity to the loop region over the
double-stranded region decreased in the order of 3b, 3a, 4,
and 3c.
The thioxanthone derivative 5 and its N,N-dimethyl deriv-
ative 6 showed a similar quenching profile to 3a when titrat-
ed with dsRNA (Figure 7). The affinity to dsRNA was calcu-
lated to be 8.3ꢁ10ꢀ7 (R2 =0.97) for 5 and 1.5ꢁ10ꢀ6 (R2 =
0.89) for 6. Fluorescence quenching by hpRNA was more ef-
ficient for both 5 and 6 than for 3a. The titration data for 5
and 6 with hpRNA, however, did not fit well to Equa-
Figure 6. Relative fluorescence intensity of xanthone derivatives a) 3 f,
&
b) 3g, and c) 4 (1 mm) with various concentrations of dsRNA ( ) and
*
hpRNA ( ). Dotted and solid lines represent nonlinear least-squares-
fitted curves derived by fitting the data with Equations (1) and (2), re-
spectively. The coefficient of determination R2 is shown in the graphs and
summarized in Table 2.
&
6
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Indicator in the Displacement Assay of Small Molecules Binding to RNA
FULL PAPER
Figure 7. Relative fluorescence intensity of xanthone derivatives a) 5 and
&
*
b) 6 (1 mm) with various concentrations of dsRNA ( ) and hpRNA ( ).
Dotted and solid lines represent nonlinear least-squares-fitted curves de-
rived by fitting the data with Equations (1) and (2), respectively. The co-
efficient of determination R2 is shown in the graphs and summarized in
Table 2.
Figure 8. Recovered fluorescence intensity of xanthone and thioxanthone
derivatives upon titrating the indicator–hpRNA complex (each 2 mm)
with Rev peptide. Solid lines are provided as a guide to the eye.
tion (2). The coefficient of determination obtained for the
ure 8a, the profile of fluorescence recovery was indicator-
dependent. A sigmoidal profile of fluorescence recovery was
observed for 3a as we reported earlier, the fluorescence in-
tensity at saturation gradually decreasing with increasing
linker length. With 3e, which has the longest linker length
and the highest affinity to dsRNA, we did not observe any
fluorescence recovery even at a Rev peptide concentration
of 4 mm. This is most likely ascribable to the increased affini-
ty of xanthone derivatives to the dsRNA with concomitant
decreased affinity to the hpRNA. As anticipated from the
weak affinity of 3 f to both dsRNA and hpRNA, the fluores-
cence intensity of 3 f in the presence of hpRNA did not
change upon titration with Rev peptide (Figure 8b). Other
xanthone and thioxanthone derivatives showed similar fluo-
rescence recovery to that of 3a with different fluorescence
intensity at saturation with a Rev peptide concentration of
4 mm.
curve fitting was 0.76 for 5 and 0.77 for 6 with an unreason-
able KdACHTUNGTRENNUNG
(loop) of 6.0ꢁ10ꢀ10 and 4.4ꢁ10ꢀ9, respectively. The
slope of the profile of fluorescence quenching for 5 and 6
was much steeper than that for 3a, thus suggesting that thio-
xanthone derivatives are more prominent than xanthone de-
rivatives for multiple binding to the loop region.
Displacement with Rev peptide: Fluorimetric titration ex-
periments with dsRNA and hpRNA revealed that synthetic
xanthone and thioxanthone derivatives showed diverse af-
finity to the dsRNA and hpRNA depending on the length
and structures of the linker as well as the chromophore. We
then examined the displacement of the bound indicators to
hpRNA by the titration of Rev peptide (Figure 8). We ex-
pected an increase of the fluorescence intensity of the indi-
cator upon displacement of the bound indicator with Rev
peptide. First, we examined the displacement of xanthone
derivatives with a different linker length. As shown in Fig-
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!

K. Nakatani et al.
tate. The solvent was evaporated, the residue was dissolved in water, and
the solution was filtered and lyophilized, to give 3a as a white solid.
1H NMR (600 MHz, D2O): d=3.48 (t, J=4.8 Hz, 4H), 4.29 (t, J=4.8 Hz,
4H), 7.36 (d, J=3.0 Hz, 2H), 7.39–7.43 ppm (4H); HRMS (ESI): m/z
calcd for C17H19N2O4: 315.1345 [M+H]+; found: 315.1334.
2,7-Bis(N-Boc-3-aminopropoxy)xanthone (2b): 1H NMR (600 MHz,
CDCl3): d=1.45 (s, 18H), 2.04 (quintet, J=5.5 Hz, 4H), 3.37 (m, 4H),
4.15 (t, J=5.9 Hz, 4H), 4.74 (broad, 2H), 7.33 (dd, J=3.3, 9.1 Hz, 2H),
7.44 (d, J=9.1 Hz, 2H), 7.69 ppm (d, J=2.9 Hz, 2H); 13C NMR
(100 MHz, CDCl3): d=28.38, 29.44, 37.89, 66.35, 79.26, 106.43, 119.37,
121.40, 125.00, 150.96, 155.01, 155.97, 176.79 ppm; HRMS (ESI): m/z
calcd for C29H38N2NaO8: 565.2526 [M+Na]+; found: 565.2526.
2,7-Bis(3-aminopropoxy)xanthone (3b): 1H NMR (600 MHz, D2O): d=
2.20 (quintet, J=6.6 Hz, 4H), 3.24 (t, J=6.6 Hz, 4H), 4.11 (t, J=5.4 Hz,
4H), 7.21 (broad, 2H), 7.28–7.33 ppm (4H); HRMS (ESI): m/z calcd for
C19H23N2O4: 343.1658 [M+H]+; found: 343.1645.
2,7-Bis(N-Boc-4-aminobutoxy)xanthone (2c): 1H NMR (400 MHz,
CDCl3): d=1.45 (s, 18H), 1.71 (quintet, J=7.3 Hz, 4H), 1.87 (quintet,
J=6.2 Hz, 4H), 3.22 (m, 4H), 4.10 (t, J=6.2 Hz, 4H), 4.63 (broad, 2H),
7.32 (dd, J=2.9, 9.2 Hz, 2H), 7.43 (d, J=9.1 Hz, 2H), 7.67 ppm (d, J=
2.9 Hz, 2H); 13C NMR (100 MHz, CDCl3): d=26.48, 26.84, 28.39, 40.28,
68.21, 79.13, 106.37, 119.32, 121.40, 125.00, 150.89, 155.11, 155.97,
176.86 ppm; HRMS (ESI): m/z calcd for C31H42N2NaO8: 593.2839
[M+Na]+; found: 593.2839.
2,7-Bis(4-aminobutoxy)xanthone (3c): 1H NMR (600 MHz, D2O): d=
1.82–1.90 (8H), 3.09 (t, J=7.2 Hz, 4H), 3.98 (t, J=5.1 Hz, 4H), 7.09
(broad, 2H), 7.18–7.26 ppm (4H); HRMS (ESI): m/z calcd for
C21H27N2O4: 371.1971 [M+H]+; found: 371.1959; m/z calcd for
C21H26N2NaO4: 393.1790 [M+Na]+; found: 393.1778.
2,7-Bis(N-Boc-5-aminopentanoxy)xanthone (2d): 1H NMR (600 MHz,
CDCl3): d=1.44 (s, 18H), 1.48–1.60 (8H), 1.84 (quintet, J=7.6 Hz, 4H),
3.16 (m, 4H), 4.06 (t, J=6.2 Hz, 4H), 4.58 (broad, 2H), 7.30 (dd, J=2.8,
8.9 Hz, 2H), 7.41 (d, J=8.9 Hz, 2H), 7.66 ppm (d, J=2.8 Hz, 2H);
13C NMR (100 MHz, CDCl3): d=23.34, 28.40, 28.80, 29.84, 40.44, 68.40,
79.06, 106.31, 119.30, 121.42, 125.05, 150.86, 155.23, 155.97, 176.90 ppm;
HRMS (ESI): m/z calcd for C33H46N2NaO8: 621.3152 [M+Na]+; found:
621.3151.
2,7-Bis(5-aminopentanoxy)xanthone (3d): 1H NMR (600 MHz, CDCl3):
d=1.51 (quintet, J=7.8 Hz, 4H), 1.75 (quintet, J=7.8 Hz, 4H), 1.79
(quintet, J=7.8 Hz, 4H), 3.03 (t, J=7.8 Hz, 4H), 3.86 (t, J=6.3 Hz, 4H),
6.92 (d, J=2.4 Hz, 2H), 7.07 (dd, J=3.3, 9.3 Hz, 2H), 7.13 ppm (d, J=
9.0 Hz, 2H); HRMS (ESI): m/z calcd for C23H31N2O4: 399.2284 [M+H]+;
found: 399.2272; m/z calcd for C23H30N2NaO4: 421.2103 [M+Na]+; found:
421.2090.
2,7-Bis(N-Boc-6-aminohexanoxy)xanthone (2e): 1H NMR (600 MHz,
CDCl3): d=1.44 (s, 18H), 1.38–1.55 (12H), 1.83 (quintet, J=7.6 Hz, 4H),
3.14 (m, 4H), 4.07 (t, J=6.5 Hz, 4H), 4.54 (broad, 2H), 7.31 (dd, J=3.4,
8.9 Hz, 2H), 7.42 (d, J=8.9 Hz, 2H), 7.67 ppm (d, J=3.4 Hz, 2H);
13C NMR (100 MHz, CDCl3): d=25.76, 26.55, 28.41, 29.06, 30.02, 40.51,
68.53, 79.02, 106.35, 119.29, 121.44, 125.06, 150.86, 155.29, 155.98,
176.93 ppm; HRMS (ESI): m/z calcd for C35H50N2NaO8: 649.3465
[M+Na]+; found: 649.3467.
Conclusion
Structurally modified derivatives of 3a were synthesized and
their spectroscopic and RNA-binding properties were inves-
tigated. The N,N-dimethyl derivative 4 showed a lower affin-
ity to the double-stranded region than the parent 3a by
compensating for the decrease of the affinity to the loop
region by one third. From the viewpoint of fluorescent indi-
cators for FID, we concluded that the N,N-dimethyl deriva-
tive 4 would be the choice for screening RRE-binding mole-
cules. Thioxanthone derivatives have intrinsic advantages
over xanthones in terms of their longer excitation wave-
length. The N,N-dimethyl derivative 6 is also useful for the
FID assay, but it is important to note that the thioxanthone
derivatives tend to bind to RNA with multiple stoichiome-
try.
The marked effect of the linker length and structures teth-
ering amino groups on the affinity to RNA suggested the
potential of the auxiliary functional groups and structures
for modulating the affinity of the indicator to specific RNA
structures. In particular, the substituents on the amino
groups having diverse structures and providing hydrogen-
bonding, electrostatic, and hydrophobic interactions to the
nucleotide bases and sugar–phosphate backbone may pro-
vide an important class of compounds, not only for the indi-
cator in FID but also for potential lead compounds of
RNA-binding molecules.
Experimental Section
General: Reagents and solvents were purchased from standard suppliers
and used without further purification. Reactions were monitored with
TLC plates precoated with Merck silica gel 60 F254. Spots were visualized
1
with UV light or ninhydrin. H and 13C NMR spectra were recorded with
JEOL JNM-LA400 and LA600 instruments. The chemical shifts are ex-
pressed in ppm relative to residual solvent as an internal standard. ESI
mass spectra were recorded on a JEOL AccuTOF JMS-T100N mass
spectrometer. Absorption spectra were recorded with a Beckman Coulter
DU800 UV/Vis spectrophotometer, and fluorescence with a Shimazdu
RF-5300PC spectrometer. Quantum yields were recorded with a PL
Quantum Yield Measurement System 9920-02 (Hamamatsu Photonics).
2,7-Bis(N-Boc-2-aminoethoxy)xanthone (2a): 2,7-Dihydroxyxanthone
(1)[14] (92.0 mg, 0.403 mmol) and triphenylphosphine (271 mg, 1.03 mmol)
were dried by their azeotrope with toluene, and dissolved in dry THF
(2.5 mL) under an argon atmosphere. Diethyl azodicarboxylate (460 mL,
40% in toluene, 1.01 mol) was added to the solution at room tempera-
ture, and then N-Boc-aminoethanol (158 mg, 0.949 mmol) was added
dropwise. The reaction mixture was stirred at room temperature over-
night, and then at 558C for 5 h. The solvent was evaporated and the resi-
due was purified by column chromatography on silica gel with elution by
chloroform and ethyl acetate (100:1 to 100:5). The product was further
purified by gel permeation chromatography to give 2a (110 mg, 53%) as
a white solid. 1H NMR (400 MHz, CDCl3): d=1.46 (s, 18H), 3.59 (m,
4H), 4.14 (t, J=5.1 Hz, 4H), 5.00 (broad, 2H), 7.33 (dd, J=3.0, 9.3 Hz,
2H), 7.45 (d, J=9.0 Hz, 2H), 7.68 ppm (d, J=2.9 Hz, 2H); 13C NMR
(150 MHz, CDCl3): d=28.89, 29.69, 40.00, 67.88, 106.84, 119.49, 121.47,
124.82, 151.10, 154.80, 155.82, 176.72 ppm; HRMS (ESI): m/z calcd for
C27H34N2NaO8: 537.2213 [M+Na]+; found: 537.2189.
2,7-Bis(6-aminohexanoxy)xanthone (3e): 1H NMR (400 MHz, D2O): d=
1.49–1.53 (8H), 1.72–1.82 (8H), 3.06 (t, J=7.6 Hz, 4H), 3.85 (t, J=
6.4 Hz, 4H), 6.89 (d, J=2.8 Hz, 2H), 4.58 (broad, 2H), 7.30 (dd, J=2.8,
8.9 Hz, 2H), 7.05 (dd, J=2.8, 9.2 Hz, 2H), 7.12 ppm (d, J=9.2 Hz, 2H);
HRMS (ESI): m/z calcd for C25H35N2O4: 427.2597 [M+H]+; found:
427.2580; m/z calcd for C25H34N2NaO4: 449.2416 [M+Na]+; found:
449.2399.
2,7-Bis(1-N-Boc-amino-2-propoxy)xanthone (2 f): The two diastereomers
could not be separated, therefore, we report the NMR data for the mix-
ture. 1H NMR (600 MHz, CDCl3): d=1.33 (d, J=6.6 Hz, 6H), 1.44 (s,
18H), 3.30 (2H), 3.53 (2H), 4.54 (2H), 4.94 (broad, 2H), 7.31 (dd, J=
2.7, 8.7 Hz, 2H), 7.43 (d, J=9.0 Hz, 2H), 7.72 ppm (d, J=2.4 Hz, 2H);
13C NMR (100 MHz, CDCl3): d=16.81, 28.36, 45.45, 74.05, 79.51, 108.90,
119.48, 121.53, 125.74, 150.98, 153.73, 155.96, 176.64 ppm; HRMS (ESI):
m/z calcd for C29H38N2NaO8: 565.2526 [M+Na]+; found: 565.2519.
2,7-Bis(2-aminoethoxy)xanthone (3a): 2,7-Bis(N-Boc-2-aminoethoxy)-
xanthone (2a; 43.7 mg, 85 mmol) was added to 4n HCl in EtOAc
(5.0 mL) and stirred at room temperature for 4 h, giving a white precipi-
&
8
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Indicator in the Displacement Assay of Small Molecules Binding to RNA
FULL PAPER
2,7-Bis(1-amino-2-propoxy)xanthone (3 f): 1H NMR (400 MHz, D2O):
d=1.44–1.46 (6H), 3.26–3.33 (2H), 3.38–3.44 (2H), 4.80–4.85 (2H), 7.38–
7.48 (4H), 7.49–7.51 ppm (2H); HRMS (ESI): m/z calcd for C19H23N2O4:
343.1658 [M+H]+; found: 343.1649; m/z calcd for C19H22N2NaO4:
365.1477 [M+Na]+; found: 365.1468.
during titration of indicators with hpRNA were fitted to Equation (2), as-
suming two independent binding site models. To calculate KdA(loop), we
employed the Kd(ds) value derived from fitting of the titration data with
CHTUNGTRENNUNG
dsRNA.
Fluorimetric titration with RNA: All titrations were performed with
a working solution of each indicator (1.0 mm) in sodium cacodylate buffer
(10 mm, pH 7.0) containing NaCl (100 mm). During titration, the indica-
tor was excited at its excitation maximum, and the changes in emission
were monitored at the emission maximum of the indicators. Relative
fluorescence intensity is the percentage of fluorescence intensity ob-
served in each measurement relative to that of free indicator. The values
relKd were calculated by dividing the Kd value of 3a by the Kd value of
each indicator as an index to compare the affinity of indicators. The se-
lectivity to the loop region was estimated by the selectivity index (SI) ob-
2,7-Bis(N-Boc-2-aminopropoxy)xanthone (2g): The two diastereomers
could not be separated, therefore, we report the NMR data for the mix-
ture. 1H NMR (600 MHz, CDCl3): d=1.32 (d, J=6.9 Hz, 6H), 1.46 (s,
18H), 4.03 (d, J=3.4 Hz, 4H), 4.11 (m, 2H), 4.79 (broad, 2H), 7.33 (dd,
J=2.8, 8.9 Hz, 2H), 7.42 (d, J=8.9 Hz, 2H), 7.66 ppm (d, J=2.8 Hz,
2H); 13C NMR (100 MHz, CDCl3): d=17.93, 28.38, 45.82, 71.73, 79.50,
106.82, 119.38, 121.42, 124.82, 151.03, 154.96, 155.22, 176.69 ppm; HRMS:
m/z calcd for C29H38N2NaO8: 565.2526 [M+Na]+; found: 565.2520.
2,7-Bis(2-aminopropoxy)xanthone (3g): 1H NMR (400 MHz, D2O): d=
1.47–1.49 (6H), 3.80–3.90 (2H), 4.05–4.15 (2H), 4.20–4.30 (2H), 7.30–
7.50 ppm (6H); HRMS (ESI): m/z calcd for C19H23N2O4: 343.1658
[M+H]+; found: 343.1648; m/z calcd for C19H22N2NaO4: 365.1477
[M+Na]+; found: 365.1467.
2,7-Bis[2-(N,N-dimethylamino)ethoxy]xanthone (4): 1H NMR (600 MHz,
CD3OD): d=2.39 (s, 12H), 2.85 (t, J=5.3 Hz, 4H), 4.22 (t, J=5.3 Hz,
4H), 7.45 (dd, J=2.9, 9.2 Hz, 2H), 7.53 (d, J=9.2 Hz, 2H), 7.66 ppm (d,
J=2.9 Hz, 2H); 13C NMR (100 MHz, CD3OD): d=45.80, 58.95, 67.17,
107.35, 120.81, 122.31, 126.45, 152.49, 156.54, 178.38 ppm; HRMS: m/z
calcd for C21H27N2NaO4: 371.1971 [M+Na]+; found: 371.1965.
tained by dividing the Kd(ds) value by the KdACTHNUTRGENUG(N loop) value of the indica-
tor.
Displacement assay: A solution of each indicator (2.0 mm) mixed with
hpRNA (2.0 mm) in sodium cacodylate buffer (10 mm, pH 7.0) containing
NaCl (100 mm) was titrated with Rev peptide. Changes in the fluores-
cence intensity during the titration were monitored at the emission maxi-
mum.
2,7-Dihydroxythioxanthone: 2,7-Dimethoxythioxanthone[15] (500 mg,
1.84 mmol) was dissolved in dichloromethane (8 mL) under an argon at-
mosphere. The reaction mixture was cooled at 08C and BBr3 (18 mL,
1.0 molLꢀ1 in dichloromethane, 18 mmol) was added carefully. The tem-
perature was allowed to rise to room temperature and the mixture was
stirred overnight. The reaction mixture was cooled at 08C and ice was
added carefully to quench the reaction. The reaction mixture was extract-
ed with ethyl acetate and dried over magnesium sulfate. The solvent was
evaporated and the residue was purified by column chromatography em-
ploying hexane and ethyl acetate (2:1 to 1:1) to give 2,7-dihydroxythio-
Acknowledgements
This work was supported by the Uehara Memorial Foundation, a Grant-
in-Aid for Scientific Research (A; 11000432) from JSPS, and the Pro-
gram for Promotion of Fundamental Studies in Health Sciences of the
National Institute of Biomedical Innovation (NIBIO, 10-22). We thank
Prof. Takashi Morii and Dr. Masatora Fukuda of Kyoto University for
the generous gift of Rev peptide, and Prof. Kazuo Harada of Tokyo Ga-
gugei University for valuable discussions. S.U. is grateful for the support
by JSPS Fellows (21, 1350).
xanthone as
a
yellow powder (363 mg, 81%). 1H NMR (600 MHz,
[D6]DMSO): d=7.23 (dd, J=2.8, 8.8 Hz, 2H), 7.64 (d, J=8.8 Hz, 2H),
7.82 ppm (d, J=2.8 Hz, 2H); 13C NMR (150 MHz, [D6]DMSO): d=
113.09, 122.60, 126.75, 127.96, 129.03, 156.18, 178.40 ppm; HRMS (ESI):
m/z calcd for C13H8NaO3S: 267.0092 [M+Na]+; found: 267.0079.
[1] a) S. L. Wiskur, H. Ait-Haddou, J. J. Lavigne, E. V. Anslyn, Acc.
Chem. Res. 2001, 34, 963–972; b) B. T. Nguyen, E. V. Anslyn,
Coord. Chem. Rev. 2006, 250, 3118–3127; c) L. Zhu, E. V. Anslyn, J.
Am. Chem. Soc. 2004, 126, 3676–3677; d) B. T. Nguyen, S. L.
Wiskur, E. V. Anslyn, Org. Lett. 2004, 6, 2499–2501.
[2] a) W. C. Tse, D. L. Boger, Acc. Chem. Res. 2004, 37, 61–69; b) D. L.
Boger, B. E. Fink, S. R. Brunette, W. C. Tse, M. P. Hedrick, J. Am.
Chem. Soc. 2001, 123, 5878–5891; c) D. L. Boger, B. E. Fink, M. P.
Hedrick, J. Am. Chem. Soc. 2000, 122, 6382–6394.
2,7-Bis(N-Boc-2-aminoethoxy)thioxanthone:
1H NMR
(600 MHz,
CDCl3): d=1.46 (s, 18H), 3.54–3.62 (4H), 4.15 (t, J=5.4 Hz, 4H), 5.00
(broad, 2H), 7.24 (dd, J=3.3, 8.7 Hz, 2H), 7.50 (d, J=9.0 Hz, 2H),
8.03 ppm (d, J=1.8 Hz, 2H); 13C NMR (150 MHz, CDCl3): d=28.55,
40.13, 67.78, 79.79, 111.52, 121.47, 122.71, 127.57, 129.73 129.89, 155.99,
157.37, 179.32 ppm; HRMS (ESI): m/z calcd for C27H34N2NaO7S:
553.1984 [M+Na]+; found: 553.1960.
2,7-Bis(2-aminoethoxy)thioxanthone (5): 1H NMR (400 MHz, D2O): d=
3.48 (t, J=4.8 Hz, 4H), 4.29 (t, J=4.8 Hz, 4H), 7.25 (dd, J=2.8, 8.8 Hz,
2H), 7.50 (d, J=9.2 Hz, 2H), 7.64 ppm (d, J=2.8 Hz, 2H); HRMS (ESI):
m/z calcd for C17H19N2O3S: 331.1116 [M+H]+; found: 331.1102.
2,7-Bis[2-(N,N-dimethylamino)ethoxy]xanthone (6): 1H NMR (600 MHz,
CD3OD): d=2.98 (s, 12H), 3.62 (t, J=5.1 Hz, 4H), 4.51 (t, J=4.8 Hz,
4H), 7.51 (dd, J=2.7, 8.7 Hz, 2H), 7.76 (d, J=9.0 Hz, 2H), 8.16 ppm (d,
J=3.6 Hz, 2H); 13C NMR (150 MHz, CDCl3): 45.94, 58.30, 66.24, 110.97,
127.46, 129.65, 129.74, 157.64, 179.51 ppm; HRMS (ESI): m/z calcd for
C21H27N2O3S: 387.1742 [M+H]+; found: 387.1731.
[3] a) Y. Wang, R. R. Rando, Chem. Biol. 1995, 2, 281–290; b) K. Ha-
masaki, R. R. Rando, Anal. Biochem. 1998, 261, 183–190; c) K. D.
Goodwin, M. A. Lewis, F. A. Tanious, R. R. Tidwell, W. D. Wilson,
M. M. Georgiadis, E. C. Long, J. Am. Chem. Soc. 2006, 128, 7846–
7854; d) D. Monchaud, C. Allain, H. Bertrand, N. Smargiasso, F.
Rosu, V. Gabelica, A. De Cian, J.-L. Mergny, M.-P. Teulade-Fichou,
Biochimie 2008, 90, 1207–1223; e) M. Krishnamurthy, N. T. Schirle,
P. A. Beal, Bioorg. Med. Chem. 2008, 16, 8914–8921; f) P. N. Asare-
Okai, C. S. Chow, Anal. Biochem. 2011, 408, 269–276.
[4] For FID studies with covalent modification, see: a) N. W. Luedtke,
Y. Tor, Biopolymers 2003, 70, 103–199; b) K. J. Blount, Y. Tor, Nu-
cleic Acids Res. 2003, 31, 5490–5500; c) T. D. Bradrick, J. P. Marino,
RNA 2004, 10, 1459–1468; d) Z. Yan, A. M. Baranger, Bioorg. Med.
Chem. Lett. 2004, 14, 5889–5893; e) R. Kawai, M. Kimoto, S. Ikeda,
T. Mitsui, M. Endo, S. Yokoyama, I. Hirao, J. Am. Chem. Soc. 2005,
127, 17286–17295; f) M. Hagihara, M. Fukuda, T. Hasegawa, T.
Morii, J. Am. Chem. Soc. 2006, 128, 12932–12940; g) K. Lang, R.
Rieder, R. Micura, Nucl. Acids Res. 2007, 35, 5370–5378; h) Y. Xie,
A. V. Dix, Y. Tor, J. Am. Chem. Soc. 2009, 131, 17605–17614;
i) H. S. Jeong, S. Kang, J. Y. Lee, B. H. Kim, Org. Biomol. Chem.
2009, 7, 921–925; j) Y. Xie, T. Maxson, Y. Tor, J. Am. Chem. Soc.
2010, 132, 11896–11897.
Spectroscopic measurements: The UV spectra of xanthone derivatives
(100 mm) were recorded in sodium cacodylate buffer (10 mm, pH 7.0) con-
taining NaCl (100 mm). Quantum yields were recorded with xanthone
and thioxanthone derivatives (20–50 mm), the concentration showing ade-
quate absorption of photons. The fluorescence spectra of xanthone deriv-
atives (2 mm) were recorded in sodium cacodylate buffer (10 mm, pH7)
containing NaCl (100 mm).
Nonlinear least-squares fitting of fluorescence titration data: All curve
fitting was carried out by using SigmaPlot 2001. The data obtained
during titration of indicators with dsRNA were fitted to Equation (1), as-
suming a 1:1 binding model, to give the Kd(ds) value. The data obtained
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
9
&
ÞÞ
These are not the final page numbers!

K. Nakatani et al.
[5] J. Zhang, S. Umemoto, K. Nakatani, J. Am. Chem. Soc. 2010, 132,
3660–3661.
[11] M. Maiti, G. S. Kumar, Med. Res. Rev. 2007, 27, 649–695.
[12] a) J. D. Puglisi, R. Tan, B. J. Calnan, A. D. Frankel, J. R. Williamson,
Science 1992, 257, 76–78; b) C. Oubridge, N. Ito, P. R. Evans, C.-H.
Teo, K. Nagai, Nature 1994, 372, 432–438; c) R. N. De Guzman,
Z. R. Wu, C. C. Stalling, L. Pappalardo, P. N. Borer, M. F. Summers,
Science 1998, 279, 384–388; d) P. Legault, J. Li, J. Mogridge, L. E.
Kay, J. Greenblatt, Cell 1998, 93, 289–299; e) D. E. Draper, J. Mol.
Biol. 1999, 293, 255–270.
[13] a) K. A. Abdullah, T. J. Kemp, J. Photochem. 1986, 32, 49–57;
b) J. C. Dalton, F. C. Montgomery, J. Am. Chem. Soc. 1974, 96,
6230–6232.
[6] a) C. S. Chow, F. M. Bogdan, Chem. Rev. 1997, 97, 1489–1514;
b) N. D. Pearson, C. D. Prescott, Chem. Biol. 1997, 4, 409–414; c) T.
Hermann, E. Westhof, Curr. Opin. Biotechnol. 1998, 9, 66–73; d) J.
Gallego, G. Varani, Acc. Chem. Res. 2001, 34, 836–843; e) E. E.
Swayze, E. A. Jefferson, K. A. Sannes-Lowery, L. B. Blyn, L. M.
Risen, S. Arakawa, S. A. Osgood, S. A. Hofstadler, R. H. Griffey, J.
Med. Chem. 2002, 45, 3816–3819; f) Q. Vicens, E. Westhof, Chem-
BioChem 2003, 4, 1018–1023; g) J. R. Thomas, P. J. Hergenrother,
Chem. Rev. 2008, 108, 1171–1224.
[7] V. W. Pollard, M. H. Malim, Annu. Rev. Microbiol. 1998, 52, 491–
532.
[14] a) T. C. S. Pace, S. L. Monahan, A. I. MacRae, M. Kaila, C. Bohne,
Photochem. Photobiol. 2006, 82, 78–87; b) S. H. Szajnman, W. Yan,
B. N. Bailey, R. Docampo, E. Elhalem, J. B. Rodriguez, J. Med.
Chem. 2000, 43, 1826–1840.
[15] a) C. P. Falshaw, N. A. Hashi, G. A. Taylor, J. Chem. Soc. Perkin
Trans. 1 1985, 1837–1843; b) N. Koumura, E. M. Geertsema, M. B.
van Gelder, A. Meetsma, B. L. Feringa, J. Am. Chem. Soc. 2002,
124, 5037–5051.
[8] a) M. H. Malim, J. Hauber, S. Le, J. V. Maizel, B. R. Cullen, Nature
1989, 338, 254–257; b) U. Fischer, J. Huber, W. C. Boelens, I. W.
Mattaj, R. Lꢂhrmann, Cell 1995, 82, 475–483; c) J. L. Battiste, H.
Mao, N. S. Rao, R. Tan, D. R. Muhandiram, L. E. Kay, A. D. Frank-
el, J. R. Williamson, Science 1996, 273, 1547–1550.
[9] a) K. A. Lacourciere, J. T. Stivers, J. P. Marino, Biochemistry 2000,
39, 5630–5641; b) K. Moehle, Z. Athanassiou, K. Patora, A. David-
son, G. Varani, J. A. Robioson, Angew. Chem. 2007, 119, 9260–
9264; Angew. Chem. Int. Ed. 2007, 46, 9101–9104; c) C. Pannecou-
que, D. Daelemans, E. D. Clercq, Nat. Protoc. 2008, 3, 427–434.
[10] a) S. Zheng, Y. Chen, C. P. Donahue, M. S. Wolfe, G. Varani, Chem.
Biol. 2009, 16, 557–566; b) Y. Liu, E. Peacey, J. Dickson, C. P. Do-
nahue, S. Zheng, G. Varani, M. S. Wolfe, J. Med. Chem. 2009, 52,
6523–6526.
[16] a) Z.-X. Wang, FEBS Lett. 1995, 360, 111–114; b) Z.-X. Wang, R.-F.
Jiang, FEBS Lett. 1996, 392, 245–249.
Received: December 15, 2011
Revised: April 4, 2012
Published online: && &&, 0000
&
10
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Indicator in the Displacement Assay of Small Molecules Binding to RNA
FULL PAPER
Fluorescent Probes
S. Umemoto, S. Im, J. Zhang,
M. Hagihara, A. Murata, Y. Harada,
T. Fukuzumi, T. Wazaki, S. Sasaoka,
K. Nakatani* .................. &&&& — &&&&
Structure–Activity Studies on the
Fluorescent Indicator in a Displace-
ment Assay for the Screening of Small
Molecules Binding to RNA
Indicating RNA binding: A fluores-
cent indicator displacement assay using
2,7-bis(aminoethoxy)xanthone (X2S)
as the indicator is reported, for the
study of small-molecule–RNA interac-
tions (see figure). The substituents at
the two alkoxy moieties modulate the
binding selectivity. The X2S derivative
having N,N-dimethylaminoethoxy sub-
stituents is the best fluorescent indica-
tor in the screening of molecules bind-
ing to Rev response element (RRE).
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
11
&
ÞÞ
These are not the final page numbers!