Rational design for cooperative recognition of specific nucleobases using β-cyclodextrin-modified DNAs and fluorescent ligands on DNA and RNA scaffolds

Article abstract of DOI:10.1002/chem.201300985

We propose a binary fluorimetric method for DNA and RNA analysis by the combined use of two probes rationally designed to work cooperatively. One probe is an oligonucleotide (ODN) conjugate bearing a β-cyclodextrin (β-CyD). The other probe is a small repo

Full text of DOI:10.1002/chem.201300985

FULL PAPER
Biosensors
A. Futamura, A. Uemura, T. Imoto,
Y. Kitamura, H. Matsuura, C.-X. Wang,
T. Ichihashi, Y. Sato, N. Teramae,
S. Nishizawa,* T. Ihara* . . . . &&&& — &&&&
Rational Design for Cooperative
Recognition of Specific Nucleobases
Using b-Cyclodextrin-Modified DNAs
and Fluorescent Ligands on DNA and
RNA Scaffolds
All lit up: A nucleobase-specific recog-
nition system has been constructed by
combining DNA/RNA-binding fluores-
cent reporter ligands and oligonucleo-
tide conjugates bearing b-cyclodextrin
(see scheme). The two molecules work
cooperatively to recognize and signal
the presence of specific nucleobases.
A binary fluorimetric method……for DNA and
RNA analysis is proposed based on the combination
of two probes designed to work cooperatively. One
is an oligonucleotide conjugate bearing a b-cyclo-
dextrin. The other is a small reporter ligand, which
comprises linked molecules of a nucleobase-specific
heterocycle and an environment-sensitive fluoro-
phore. The reporter ligand recognizes a single nucle-
obase displayed in a gap on the target labeled with
the conjugate, and the fluorophore moiety forms a
luminous inclusion complex with nearby cyclodex-
trin. For more details see the Full Paper by S. Nishi-
zawa, T. Ihara et al. on page && ff.
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

DOI: 10.1002/chem.201300985
Rational Design for Cooperative Recognition of Specific Nucleobases Using
b-Cyclodextrin-Modified DNAs and Fluorescent Ligands on DNA and RNA
Scaffolds
Akika Futamura,[a] Asuka Uemura,[a] Takeshi Imoto,[a] Yusuke Kitamura,[a, b]
Hirotaka Matsuura,[a] Chun-Xia Wang,[c] Toshiki Ichihashi,[c] Yusuke Sato,[c]
Norio Teramae,[c] Seiichi Nishizawa,*[c] and Toshihiro Ihara*[a, b]
Abstract: We propose a binary fluori-
metric method for DNA and RNA
analysis by the combined use of two
probes rationally designed to work co-
operatively. One probe is an oligonu-
cleotide (ODN) conjugate bearing a b-
cyclodextrin (b-CyD). The other probe
is a small reporter ligand, which com-
prises linked molecules of a nucleo-
base-specific heterocycle and an envi-
ronment-sensitive fluorophore. The
heterocycle of the reporter ligand rec-
ognizes a single nucleobase displayed
in a gap on the target labeled with the
conjugate and, at the same time, the
fluorophore moiety forms a luminous
inclusion complex with nearby b-CyD.
Three reporter ligands, MNDS (naph-
gates. For the DNA target, the b-CyD
tethered to the 3’-end of the ODN
facing into the gap interacted with the
fluorophore sticking out into the major
groove of the gap site (MNDS and
DPDB). Meanwhile the b-CyD on the
5’-end of the ODN interacted with the
fluorophore in the minor groove
(MNDB and DPDB). The results ob-
tained by this study could be a guide-
line for the design of binary DNA/
RNA probe systems based on control-
ling the proximity of functional mole-
cules.
thyridine–dansyl
linked
ligand),
MNDB (naphthyridine–DBD), and
DPDB (pyridine–DBD), were used for
DNA and RNA probing with 3’-end or
5’-end modified b-CyD–ODN conju-
Keywords: biosensors
· cyclodex-
trin · DNA recognition · fluores-
cence · host–guest systems
Introduction
cannot be used to separate bound and free targets. Probes
for use in solution are expected to emit a signal only when
they bind to their targets.[6–12] This is an essential require-
ment for bioprobes providing a spatiotemporal response.
In the design of molecular probes, the structural require-
ments for binding to the target and for signal generation are
often different. Most of the molecular probes that are
widely used in current molecular biological research meet
one of the two requirements. However, it is very difficult to
design probes that meet both requirements in one mole-
cule.[13–22]
Various fluorescent molecular probes have been developed
for targeting certain biomolecules, cell organelles, or intra-
cellular processes.[1–5] In general, free or nonspecifically
bound probes must be thoroughly washed from the sub-
strates on which the targets are immobilized to obtain a
signal with a good contrast, because of signal interference
from unbound or nonspecific probes. However, for targets
dissolved or dispersed in homogeneous solutions, washing
We propose a general solution for designing molecular
probes that recognize their targets and switch their signal on
at the same instant. This could be achieved using coopera-
tive action between probes. We have shown the validity of
the design through the analysis of DNA and RNA. To date,
several DNA-probing systems based on specific photochemi-
cal reactions,[23–25] luminous metal complex formation,[26,27]
and electrochemically modulated inclusion-complex forma-
tion[28] on DNA have been proposed.
In this work, a convenient technique for SNP (single nu-
cleotide polymorphism) genotyping in a homogeneous solu-
tion is presented. We prepared DNA conjugates, b-cyclodex-
trin (b-CyD)-modified oligodeoxyribonucleotides (CyD–
ODN), and nucleobase-specific fluorescent reporter ligands
and used them simultaneously for SNP analysis in aqueous
solution. The reporters are linked hybrid molecules consist-
[a] A. Futamura, A. Uemura, T. Imoto, Dr. Y. Kitamura,
Dr. H. Matsuura, Prof. Dr. T. Ihara
Department of Applied Chemistry and Biochemistry
Graduate School of Science and Technology, Kumamoto University
Kurokami, Chuo-ku, Kumamoto 860-8555 (Japan)
Fax : (+81)96-342-3679 (office) or 3873
E-mail: toshi@chem.kumamoto-u.ac.jp
[b] Dr. Y. Kitamura, Prof. Dr. T. Ihara
CREST Japan Science and Technology Agency
Gobancho, Chiyoda-ku, Tokyo 102-0076 (Japan)
[c] Dr. C.-X. Wang, T. Ichihashi, Dr. Y. Sato, Prof. Dr. N. Teramae,
Prof. Dr. S. Nishizawa
Department of Chemistry, Graduate School of Science
Tohoku University, Aoba-ku, Sendai 980-8578 (Japan)
E-mail: nishi@m.tohoku.ac.jp
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201300985.
&
2
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Cooperative Recognition of Specific Nucleobases
FULL PAPER
ing of two parts: a nucleobase-specific heterocyclic ligand
and an environment-sensitive fluorophore. In this assay,
SNP bases (N: A, T, G, or C) on the targets are displayed at
a gap in ternary duplexes (N-gap duplexes) consisting of the
targets (DNA or RNA), the CyD–ODN, and the mask, as
shown in Figure 1. That is, the sequences of the CyD–ODN
and the mask are designed to be complementary to both se-
quences adjacent to the SNP base on the target (Figure 2a).
Therefore, both ODNs (oligodeoxyribonucleic acids) form a
Figure 1. Schematic of nucleobase detection by binary probing. N indi-
cates the target nucleobase: A, T, G, or C.
stable tandem duplex, regardless of the type of N. The re-
porter ligand is then added to these ternary duplexes with a
displayed SNP base. The fluorophore moiety of the reporter
is expected to form a luminous inclusion complex with
nearby a b-CyD[29–35] only when the heterocyclic moiety
binds to the N displayed in the gap. Base discrimination is
not based on the hybridization specificity of a long DNA
probe, as in conventional methods, but on the complemen-
tarity of a small ligand with only one displayed base. This
would be expected to increase the contrast of the signal.
In our previous report, we showed the preliminary result
of the system, in which G-specific detection was performed
by the combined use of MNDS (Figure 2b) and 3’-modifed
CyD–ODN (3CyD–ODN). The system, however, did not
work with 5’-modified CyD–ODN (5CyD–ODN).[36] We
thought that the disparity between 3CyD–ODN and 5CyD–
ODN would be derived from the asymmetric microenviron-
ment around the gap of the DNA duplex. The orientations
of the bound reporter ligand and the modified ends of b-
CyD might explain the results. Here, we conducted a sys-
tematic study using three reporter ligands, MNDS and two
newly prepared ligands MNDB and DPDB, having different
binding selectivities and fluorophores. The method provides
a general principle for the rational design of systems for nu-
cleobase targeting of DNA or RNA. In this report, the re-
sults for all of the three ligands are shown in a comprehen-
sive style to ensure the completeness of the whole picture of
the bimolecular probing scheme.
Figure 2. Probes and targets. a) Sequences of the ODN probes (green
line, CyD–DNA; red line, mask) and the target DNA and RNA. N indi-
cates the target nucleobase: A, T, G, or C. b) Structures of fluorescent re-
porter ligands: MNDS, MNDB, and DPDB. c) Putative pairing of the flu-
orescent ligands with nucleobases. The red and green areas indicate
major and minor grooves, respectively.
Results and Discussion
Design of the system: Figure 2a, b show the sequences and
structures of the ODNs and the three fluorescent reporter li-
gands used in this study. Figure 2c shows putative pairings
of the fluorescent reporters and nucleobases. Part of the se-
quence of the TPMT gene was used as a target. SNP bases,
N, on the targets are displayed in a gap in the duplexes (N-
gap) consisting of the targets CyD–ODN and mask. That is,
the sequences of CyD–ODN and mask are designed to be
complementary to both sequences adjacent to the N on the
target. Both ODNs form a fully matched tandem duplex re-
gardless of the type of N. The mask is expected to play a
role to define the target base to the one adjacent to CyD–
ODN and to form a comfortable space to accommodate the
reporter ligands.
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

S. Nishizawa, T. Ihara et al.
Fluorescent reporter molecules were systematically de-
signed by covalently linking two functionally different ele-
ments, a heterocycle enabling complementary hydrogen
bonding with a particular nucleobase and an environment-
sensitive fluorophore. As heterocycles for nucleobase recog-
nition, 1,8-naphthyridine (MND) and 3,5-diamino-6-chloro-
pyradine (DAP) were used. MND binds with G or C de-
pending on the structure of the linker chain tethered to the
fluorophore moiety.[37] DAP is complementary to T/U.[17]
The binding behavior of these heterocycles has been studied
in detail separately for the nucleobases in an AP site (apur-
inic/apyrimidinic site ). The selectivity and the binding affin-
ity were investigated in regard to the effect of the introduc-
tion of functional groups[16,38–40] and flanking nucleobas-
es.[39–43] At the same time, 2,6-dansyl and 4-(N,N-dimethyla-
minosulfonyl)benzofurazan (DBD) were employed as envi-
ronment-sensitive fluorophores. MNDS is a hybrid molecule
of MND (monomethyl) and 2,6-dansyl, which are the base-
recognition and fluorescent parts, respectively. The dansyl
moiety protrudes into the major groove of the duplex, when
MND forms a supposed complex with G. MNDB consists of
MND (trimethyl) and DBD. It is expected to bind with C
with its DBD moiety protruding into the minor groove.
DPDB consists of DAP and DBD. The DBD moiety could
stick out into both grooves, because the hydrogen bonding
surface of T/U is symmetric. The fluorescent reporter li-
gands could be designed to meet the demands of different
target nucleobases, fluorescent colors, and directions that
the fluorophore protrudes by simply combining the elemen-
tary functional modules.
Figure 3. Fluorometric DNA analysis using MNDS. a) Fluorescence spec-
tra (top) (blue, G-target; black, A-target; red, C-target; green, T-target
DNA) and normalized fluorescence intensities at 443 nm. Left and right
indicate the results for the detection of N-target DNAs using 3CyD–
ODN and 5CyD–ODN conjugates, respectively. 1.0 mm N-gap duplexes,
5.0 mm MNDS, 1m NaCl, 10 mm phosphate buffer (pH 7.0), 0.83%
DMSO, lex =328 nm, 08C. b) Fluorescence images of the solutions of A-
target (left) and G-target DNAs (right) with 3CyD-ODN measured at
58C. Excitation source: low-pressure mercury lamp (6 W, 365 nm).
Interaction of MNDS with N-gap DNA duplexes: Figure 3a
(top) shows the fluorescence spectra of MNDS in the pres-
ence of N-gap duplexes with the 3CyD–ODN (left) and
5CyDODN (right), measured at 08C. Normalized fluores-
cence intensities at 443 nm are shown as the relative values
in the bar graph (bottom) after subtraction of the fluores-
cence intensity of the solution containing only MNDS. The
signal was enhanced significantly only for G-target DNA
when using the 3CyD–ODN. As expected, the MND moiety
in MNDS seemed to recognize G by complementary hydro-
gen bonding in the gap. The relative fluorescence signals
(with respect to G) for A, C, and T were 10.7, 11.6, and 22.7,
respectively. The signal from dansyl moiety was observed as
blue emission, and the contrast is large enough to be recog-
nized by the naked eye, as shown in Figure 3b. On the other
hand, the specific signal was not observed when using the
5CyD–ODN.
The interactions of MNDS with the G-gap DNA duplexes
were further studied by fluorescence titration at 08C (see
Figure S1 in the Supporting Information). The observed
spectral changes resulted from the formation of an inclusion
complex between the dansyl and b-CyD moieties, because
the MNDS spectra scarcely changed by the addition of the
control G-gap duplex lacking b-CyD. The changes in the flu-
orescence intensities were fitted to the theoretical curve,
which was derived assuming a 1:1 interaction. The binding
constants of MNDS with b-CyD and MNDS with a G-gap
duplex containing 3CyD–ODN were calculated to be 5.2ꢁ
102 mÀ1 and 2.4ꢁ105 mÀ1, respectively. These results show the
validity of the molecular design of MNDS. Although the
binding of each elemental unit of MNDS, MND and dansyl,
with its supposed counterpart, G-gap and b-CyD, is very
weak, the synergistic effect of linking the two elements
made the binding constant of their integrated hybrid mole-
cule, MNDS, significantly higher in the microenvironment
of G-gap duplex with the 3CyD-ODN. That is, MND inserts
into the G-gap to form complementary hydrogen bonding
with G and, simultaneously, the dansyl group is accommo-
dated in the nearby b-CyD modified at the 3’-end of the
ODN. The binding constants with A-, C-, and T-gap duplex-
es carrying the 3CyD–ODN were too small to be calculated.
Meanwhile, the binding constant of MNDS with the G-gap
duplex containing the 5CyD–ODN was much less than that
with the 3CyD–ODN. It was roughly estimated to be
<104 mÀ1. Therefore, the difference in fluorescence signal in-
tensities observed for 3CyD–ODNs and 5CyD–ODNs
&
4
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Cooperative Recognition of Specific Nucleobases
FULL PAPER
shown in Figure 3a could be attributed to the difference in
the binding constant for both duplexes. Thus, each of the
two elementary units of MNDS does not contribute suffi-
ciently to the MNDS binding in the G-gap duplex with the
5CyD–ODN. The poor signal contrast observed for the
5CyD–ODN system could be the result of inadequate inter-
action of the MND moiety with N-gaps and/or dansyl with
b-CyD.
To explain the disparity observed in the interaction of
MNDS with 3CyD–ODN and 5CyD–ODN, the steric effect
derived from the asymmetric nature of the DNA duplex
structure may have to be taken into account. MNDS pro-
trudes its dansyl moiety into the major groove of the duplex
when MND forms complementary hydrogen bonds with G
in the gap, as shown in Figure 4. The distances from the
ends of both ODNs that face into the gap to the dansyl
MNDS. The latter case needs to be demonstrated by experi-
ments using other reporter ligands.
Interaction of MNDB with N-gap DNA duplexes: To com-
plete the working principle of the system derived from the
unique binding behavior of MNDS, we prepared new fluo-
rescent reporter molecules, MNDB and DPDB (see
Figure 2 and the next section). MNDB consists of MND and
DBD as recognition and signaling moieties, respectively.
The resonance effect from the amino linker of MNDB
raises the basicity of the endocyclic nitrogens of MND and
changes the target nucleobase to C. MNDB protrudes the
DBD moiety into the minor groove when it forms a comple-
mentary pairing with C displayed in the gap, as shown in
Figure 2c. According to the hypothetical principle men-
tioned above, the DBD moiety in MNDB would be accom-
modated into the b-CyD of a 5CyD–ODN.
Figure 5a shows the fluorescence spectra and normalized
fluorescence intensities of MNDB in the presence of N-gap
DNA duplexes with the 3CyD–ODN (left) and the 5CyD-
Figure 4. One of the possible 3D structures of the MNDS/G-gap DNA
duplex. The model was geometry-optimized by AMBER* force field
with a GB/SA (generalized Born/surface area) solvent model using Mac-
roModel version 9.1. The distances from the dansyl moiety of bound
MNDS to both (3’- and 5’-) ends of ODNs are quite different. b-CyD
moiety of the CyD–ODN was omitted from the structure for clarity.
group of the bound MNDS would be quite different. Molec-
ular modeling studies showed that the distances from the 3’-
and 5’-ends of both ODNs to the center of dansyl were
about 10.4 and 17.8 ꢂ, respectively, in the optimized com-
plex structure, in which the short linker chain of MNDS
took a trans conformation. The conformational freedom
would be limited for the linker chain that connects the 5’-
end of the ODN with b-CyD, when the dansyl group of
MNDS is included in the 5’-modified b-CyD. This would
make the binding of MNDS with the G-gap duplex of the
5CyD-ODN weaker, because this entropic disadvantage
would deteriorate the synergy between the two elementary
units of MNDS on binding to the G-gap duplexes. The re-
sults suggest that while the reporter ligand that protrudes its
fluorophore into the major groove of the N-gap duplex
works cooperatively with the 3CyD–ODN, the ligand that
protrudes its fluorophore into the minor groove works with
the 5CyD–ODN. The former case has been verified with
Figure 5. Fluorometric DNA analysis using MNDB. a) Fluorescence spec-
tra (blue, G-target; black, A-target; red, C-target; green, T-target DNA)
and normalized fluorescence intensities of MNDB at 535 nm. Left and
right indicate the results for the N-target DNAs detection using 3CyD–
ODN and 5CyD–ODN conjugates, respectively. 1.0 mm N-gap duplexes,
5.0 mm MNDB, 1m NaCl, 10 mm phosphate buffer (pH 7.0), 0.83%
DMSO, lex =447 nm, 08C. b) Fluorescence images of the solutions of
tandem duplexes of A-target (left) and C-target DNA (right) containing
the 5CyD–ODN in the presence of MNDB at 58C. Excitation source:
low-pressure mercury lamp (6 W, 365 nm).
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

S. Nishizawa, T. Ihara et al.
ODN (right). The signal was enhanced significantly not only
upon addition of the C-target but also the T-target in the
presence of the 5CyD–ODN. Signal changes observed in the
interaction with A- and G-targets were very small. A dual
response to C and T could be caused by protonation on two
different endocyclic nitrogens of MND. While protonation
of the N1 position makes MND complementary to C, proto-
nation of the N8 position makes it complementary to T. Ac-
tually, an MND derivative is known to bind to both C and T
in the AP site of a DNA duplex with binding constants of
1.9ꢁ107 and 0.91ꢁ107 mÀ1, respectively.[16] The emission from
the DBD moiety was observed as a green image, as shown
in Figure 5b. By contrast, selective recognition of the nucle-
obases failed in the presence of the 3CyD–ODN. The inter-
actions of MNDB with the C-gap duplexes were studied fur-
ther by fluorescence titration at 08C (see Figure S2 in the
Supporting Information). The binding constants of MNDB
with the C-gap and T-gap duplexes carrying the 5CyD–
ODN were calculated to be 1.8ꢁ105 and 1.3ꢁ105 mÀ1, re-
spectively. Meanwhile, the binding constant of MNDB with
the A-gap duplex was too small to be calculated. Incidental-
ly, the binding constant of MNDB with b-CyD was estimat-
ed to be 4.3ꢁ103 mÀ1 under the same conditions (data not
shown). These results indicate that pyrimidines (C and T)
could be detected by the combined use of MNDB and the
5CyD–ODN.
Figure 6. Fluorescence spectra (blue, G-target; black, A-target; red, C-
target; green, T-target DNA) and normalized fluorescence intensities of
DPDB at 550 nm. Left and right show the results for the N-target DNA
detection using 3CyD–ODN and 5CyD–ODN conjugates, respectively.
1.0 mm N-gap duplexes, 5.0 mm DPDB, 1m NaCl, 10 mm phosphate buffer
(pH 7.0), 0.83% DMSO, lex =427 nm, 08C.
Thus, we have completed the entire picture of recogni-
tion/signaling for the target nucleobases using the combina-
tion of a fluorescent reporter ligand and a CyD–ODN. That
is, 3CyD–ODNs and 5CyD–ODNs form luminous inclusion
complexes with the reporter ligands at the major and minor
grooves, respectively, of N-gap DNA duplexes. According to
this simple principle, the system can be logically designed to
detect specific nucleobases.
Their binding constants were too small to be calculated.
Separately, the binding constant of DPDB with b-CyD was
estimated to be 8.8ꢁ102 mÀ1 under the same conditions.
DPDB binding to both T-gap DNA duplexes seems to be
enhanced by a synergistic effect of both constituents, DAP
and DBD.
The working principles of the fluorescent reporter ligands
in the binary probing system, which were derived from the
studies of MNDS and MNDB, were supported by the
DPDB studies. We confirmed that the effective interactions
involving the 3’- and 5’-ends of ODNs facing the N-gap
occur at the major and the minor grooves, respectively
(Figure 7).
Interaction of DPDB with N-gap DNA duplexes: DPDB
was designed to target T, based on the principle mentioned
above. DPDB consists of DAP and DBD as recognition and
signaling moieties, respectively. As shown in Figure 2c, bi-
modal binding is expected for DPDB on T recognition, be-
cause the hydrogen bonding surface of T is symmetric (as is
that of DAP). Therefore, DPDB might work cooperatively
with both 3CyD–ODNs and 5CyD–ODNs, because DPDB
is allowed to direct the DBD moiety to either groove.
Figure 6 shows the fluorescence spectra of DPDB and the
normalized fluorescence intensities at 550 nm with the addi-
tion of N-target DNAs in the presence of the 3CyD–ODN
(left) and the 5CyD–ODN (right). As expected, the signal
was enhanced significantly only for the T-target with both
the 3CyD–ODN and 5CyD–ODN. Signal changes observed
for the interactions with A-, C-, and G-targets were margin-
al. The binding constants of DPDB with the T-gap DNA du-
plexes carrying the 3CyD–ODN and the 5CyD–ODN were
calculated as 2.1ꢁ107 and 9.2ꢁ105 mÀ1, respectively (see Fig-
ure S3 in the Supporting Information). The interaction of
DPDB and other N-gap DNA duplexes carrying the 3CyD–
ODN or the 5CyD–ODN were examined in the same way.
Binary probing for RNA targets: This system can also be
used to detect RNA targets with the same sequences.
Hence, DNA/RNA heteroduplexes were used as scaffolds
for nucleobase recognition. Figure 8 shows the normalized
fluorescence intensities of the three reporter ligands for N-
target RNAs using the 3CyD–ODN (left) and the 5CyD–
ODN (right). Measurements were performed under the
same conditions used for the DNA targets mentioned
above. Results obtained for RNA targets were totally differ-
ent from those for the DNA targets. The most pronounced
difference from the DNA targets was shown in the response
of MNDS (Figure 8a). MNDS did not work with either of
the CyD–ODNs in the heteroduplex systems. None of the
RNA targets increased the fluorescence signal of MNDS.
Meanwhile, although the pyrimidine-specific response of
&
6
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Cooperative Recognition of Specific Nucleobases
FULL PAPER
Table 1. Nucleobases that can be targeted by the binary probing system
for DNA and RNA.[a]
Target
Conjugate
Reporter ligand
MNDS
MNDB
DPDB
5CyD-ODN
3CyD-ODN
5CyD-ODN
3CyD-ODN
–
C, T (m)
–
C, T (m)
C, T (m)
T (m)
T (M)
U (m)
U (m)
DNA
RNA
G (M)
–
–
[a] M and m in parentheses show the putative grooves, where the two
probes (the conjugate and the reporter ligand) meet with each other.
They stand for major (M) and minor (m) grooves.
with DNA (Figure 8c). DPDB showed a specific response
to U using either the 3CyD–ODN or the 5CyD–ODN.
The profiles for DNA and RNA recognition are summar-
ized in Table 1. The profiles of DNA and RNA with the
same sequence were very different from each other. UV
melting studies showed that all the duplexes (N-gap homo-
heteroduplexes) are stable under the experimental condi-
tions (data not shown). Therefore, the difference in the re-
sponse might come from the differences in the structures of
the DNA duplexes and DNA/RNA heteroduplexes. Figure 9
shows CD spectra of the G-gap duplexes lacking b-CyD. In
the case of the DNA homoduplexes, the spectrum showed a
positive peak around 280 nm and an intense negative peak
around 250 nm (Figure 9a). The CD spectrum of the RNA/
DNA heteroduplex indicated an intense positive peak
Figure 7. Simplified 3D image of the gap in the tandem DNA duplex.
The target nucleobase, N, is displayed in the gap. The nucleobase-specific
fluorescent ligands (blue) inserted into the gap protrude their fluoro-
phores (shown as “R”) into the major or minor groove. The fluorophores
in the major and the minor groove locate in close proximity with the 3’-
and 5’-end of the probe ODNs (green), respectively.
Figure 8. Normalized fluorescence intensities of a) MNDS, b) MNDB,
and c) DPDB for N-target RNAs detection. Left and right indicate the
results obtained using the 3CyD–ODN and the 5CyD–ODN, respectively.
Fluorescence of each reporter molecule (5.0 mm) with N-gap heterodu-
plexes (1.0 mm) was measured in a solution containing NaCl (1m), phos-
phate buffer (10 mm, pH 7.0), and DMSO (0.83%) at 08C. lex =328, 447,
and 427 nm for MNDS, MNDB, and DPDB, respectively.
Figure 9. CD spectra of G-gap duplexes lacking b-CyD: a) DNA homo-
duplex and b) RNA/DNA heteroduplex. The duplexes were dissolved
(1.25 mm) in a solution containing NaCl (1.0m), phosphate buffer (10 mm,
pH 7.0), and DMSO (0.83%). Measurements were carried out at ambient
temperature.
MNDB was preserved for the RNA targets, the fluorescence
was enhanced not only with the 5CyD–ODN but also with
the 3CyD–ODN (Figure 8b). Response characteristics of
DPDB for RNA targeting were almost the same as those
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!

S. Nishizawa, T. Ihara et al.
around 275 nm and a small negative peak around 230 nm
(Figure 9b). These are the typical features of B-form and A-
form duplexes.[44,45] That is, N-gap DNA homoduplexes and
RNA/DNA heteroduplexes used in this study should take
the structures of B-form and A-form duplexes, respectively.
One of the notable features of an A-form duplex is that the
major groove is extremely narrow and very deep, while the
minor groove is very broad and shallow.[46,47] This structure
might explain the signal profile for RNA. MNDS protrudes
its dansyl moiety to the major groove. A modeling study
showed that the narrow and deep major groove of the heter-
oduplex does not leave space for b-CyD after the dansyl
moiety sticks out (Figure 10a). Therefore, b-CyD could not
the SNP assay should be quite wide compared with tradi-
tional probe hybridization. Fluorometric SNP analyses using
MNDS and 3CyD–ODN, for example, were performed to
investigate an effect of the temperature on the signal con-
trast of this system. Normalized signals measured at differ-
ent temperatures are shown in Figure 11 with the UV melt-
ing curve of the duplex. The signal quality (signal contrast)
did not deteriorate much, even at 358C, which is below the
Figure 10. One of the possible 3D structures of a) MNDS/G-gap and
b) MNDB/C-gap RNA/DNA heteroduplexes. The model was geometry-
optimized by AMBER* force field with GB/SA (generalized Born/sur-
face area) solvent model using MacroModel version 9.1. The fluorophore
of MNDS is buried deeply in the major groove of the heteroduplex,
while that of MNDB is exposed to the bulk solution from the shallow
minor groove of the heteroduplex. The b-CyD moiety of the CyD–ODN
was omitted from the structure for clarity.
Figure 11. UV melting curve (top) and normalized fluorescence intensi-
ties (bottom) of MNDS in the presence of N-gap duplexes containing the
3CyD–ODN. Fluorescence measurements were performed at 08C (left),
358C (center), and 758C (right). All other experimental conditions were
the same as Figure 3.
melting temperature (Tm ꢀ508C) of this duplex. The meas-
urement at 758C was, of course, disabled. Thus, the assay
can be carried out at any temperature that is lower than the
Tm, because the duplex framework that provides the envi-
ronment for nucleobase recognition/signaling is formed at
temperatures below the Tm. The temperature dependence of
the interaction of MNDS with G-gap DNA duplex also has
to be taken into account. The decrease in the signal intensi-
ties at 358C compared with those at 08C observed in
Figure 11 can be attributed to the temperature dependence
of MNDS binding. As commonly observed in most bimolec-
ular interactions, the bindings of MND and dansyl with the
G-gap and b-CyD, respectively, are thought to weaken with
a rise in temperature. However, the result showed that the
temperature dependence of the MNDS binding was not so
significant, at least in the temperature range of the present
measurements. Even so, the measurement at 08C is the
most desirable condition, which endorses a highest stabiliza-
tion for all the duplexes in aqueous solutions. This unique
feature (low temperature susceptibility) could be very im-
portant in massively parallel processing systems such as
DNA arrays or chips.
access the dansyl moiety in the major groove of the hetero-
duplex, even from the 3’-end of CyD–ODN. The DBD
moiety of MNDB protrudes into the minor groove. The
DBD in the shallow minor groove seems to be accessible
from both 3’- and 5’-ends of the CyD–ODNs, as shown in
Figure 10b. Only the reporter ligands that protrude their sig-
naling group into the minor groove could be used for RNA
analysis.
Effect of temperature on the signal contrast: One of the
merits of the present method is that the measurements for
all of the targets using different DNA probes could be per-
formed under the same conditions, because the recognition
does not rely on subtle differences in thermal stabilities of
the duplexes formed with the probes. Generally, DNA prob-
ing relies on the specificity of probe hybridization. Accord-
ingly, to recognize one-base displacement, the temperature
should be controlled in a narrow range between the melting
temperatures of the full match and mismatch duplexes. In
the present system, however, all the duplexes that interact
with the reporter ligands are fully matched duplexes; only
the unpaired base displayed in the gap is different. There-
fore, the response is not completely independent of temper-
ature, but the temperature range in which we can conduct
Remaining problems and perspective: The general concerns
for molecular probes are mainly selectivity and sensitivity.
&
8
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Cooperative Recognition of Specific Nucleobases
FULL PAPER
To improve probes according to these two requirements, we
have been conducting preliminary studies, separately, using
various nucleobase-recognition ligands. For selectivity, we
now have ligands that are complementary to each of the
four nucleobases. For example, the C/T selectivity of naph-
thyridine can be controlled by introducing electron-with-
drawing groups (such as a trifluoromethyl group) to the
ring.[39] The protonation to the endocyclic nitrogens of naph-
thyridine is delocalized between the N1 and N8 positions.
The electron-withdrawing effect seems to localize the proto-
nation to the N1 position and makes the naphthyridine ring
complementary to C. We have just reported that the C/T se-
lectivity could also be controlled by competitive binding of
additional ligands.[48] We have also published reports that al-
loxadine and lumazine have adenine selectivity.[38,49] All of
these findings could be fed into the present system.
The sensitivity of the present system is adequate for the
products of polymerase chain reactions (PCR), but still
should be improved more for practical applications. Consid-
ering the properties of the fluorophores employed in this
study, the signal intensities were supposed to be enhanced
more than observed here. It was apparent that only a por-
tion of the reporter ligands bound with the N-gaps under
the experimental conditions. This is due to the low binding
constant of the reporter ligands to the target N-gaps. If we
use ligands with higher binding constants in this system, we
could reduce the concentration of the ligand and, conse-
quently, improve the signal contrast. We already succeeded
in the synthesis of such ligands for some cases.[40] Alterna-
tively or additionally, modification of b-CyD would be effec-
tive. For example, by making b-CyD hydrophobic (e.g., by
methylation), both the binding constant and the fluores-
cence intensity of the inclusion complexes could be en-
hanced.[50]
dependently to design the desired reporting ligands for spe-
cific nucleobases and fluorescence colors.
The design of the proposed system is general; therefore,
the signal is not be limited to fluorescence. Various signals
(e.g., electrochemical, colorimetric, or catalytic) could be
modulated by each of the specific counterparts through the
controlled proximity in the major and minor grooves of the
N-gap DNA duplexes.
Experimental Section
General: b-CyD and N-succinimidyl 3-(2-pyridyldithio) propionate
(SPDP) were purchased from Sigma–Aldrich (Saint Louis, MO, USA)
and Dojindo Laboratories (Kumamoto, Japan), respectively. RNA targets
were purchased from Japan Bio Services (Saitama, Japan). All ODNs
were synthesized by an automated DNA synthesizer (Expedite 8900)
using conventional phosphoramidite methods. The phosphoramidite mon-
omers were purchased from Proligo (Hamburg, Germany) and Glen Re-
search (Sterling, VA, USA). After purification by HPLC, all ODNs and
synthesized ODN conjugates were identified using MALDI-TOF mass
spectrometry on a Bruker Daltonics Autoflex-III (Billerica, MA, USA).
All other reagents were obtained as the highest grade and used without
further purifications.
CyD–ODNs (5CyD–ODNs and 3CyD–ODNs) were synthesized accord-
ing to Scheme S1 (Supporting Information). 3’- or 5’-end aminopropyl-
linked DNA was modified with a bifunctional linker molecule and then
coupled with monothiolated b-CyD.
Synthesis of monotosylated b-CyD:[50] b-CyD (0.50 g, 0.43 mmol) was dis-
solved in dried pyridine (4.3 mL) under an atmosphere of argon. To the
solution, p-toluene sulfonylchloride (0.16 g, 0.85 mmol) was added in an
ice bath and then stirred at room temperature. The progress of the reac-
tion was occasionally monitored by TLC (1-buthanol/ethanol/water=
5:4:3, indicator: p-anisaldehyde). The reaction was quenched by addition
of water (0.35 mL) after 3 h. Analysis by TLC indicated the presence of
the three spots corresponding to b-CyD (Rf =0.30), monotosylated b-
CyD (0.48), and ditosylated b-CyD (0.57) at almost the same density.
The solution was concentrated to a half in vacuo and poured into acetone
(8.5 mL) with vigorous stirring. The resulting white solid was collected
and repeatedly recrystallized from water.
1
White solid 72 mg (12.7%); H NMR (399.65 MHz, [D6]DMSO): d=2.43
Conclusions
(s, 3H), 3.10–3.45 (m, 14H), 3.45–3.66 (m, 28H), 4.10–4.60 (m, 6H), 4.76
(s, br, 2H), 4.83 (s, br, 5H), 5.60–5.85 (m, 14H), 7.42 (d, 2H, J=8.3 Hz),
7.74 ppm (d, 2H, J=8.3 Hz).
A nucleobase-specific recognition system was constructed
by the rational design of a combination between DNA/
RNA-binding fluorescent reporter ligands and CyD–ODN
conjugates. The two molecules work cooperatively to recog-
nize/report specific nucleobases displayed in the gap of the
duplexes. The groove in which the expected cooperation
proceeds can be pre-assigned according simple rules. For
DNA targeting, the signaling moiety located in the major
groove interacts with the counterpart modified on the 3’-end
of ODN, while the reporter moiety in the minor groove in-
teracts with that on the 5’-end of ODN. The system permits
the design of various reporting molecules in a logical
manner. The reporting ligands could be prepared by cova-
lently linking the selected recognition and fluorescent mole-
cules through an alkyl chain, because the two elementary
processes (recognition and reporting) are separated and al-
lotted to two different sites on the duplex structure. There-
fore, the two elementary functional groups can be chosen in-
Synthesis of monothiolated b-CyD:[51,52] Monotosylated b-CyD (0.50 g,
0.39 mmol) and thiourea (0.50 g, 6.6 mmol) were dissolved in aqueous
methanol (25 mL, 80%) and refluxed for 72 h. The solution was evapo-
rated in vacuo. The solid was suspended in methanol (7.6 mL) and stirred
for 1 h at room temperature. The solid was filtered and dissolved in aque-
ous solution of NaOH (17 mL, 10%) and stirred for 5 h at 508C. After
the solution was acidified with HCl (1m) to pH 2, trichloroethylene
(1.2 mL) was added. After stirring overnight, the precipitate was filtered
and washed with water. Evaporation of trichloroethylene in vacuo fol-
lowed by repeated recrystallization from water gave a white solid.
White solid 0.27 g (59.4%); TLC (silica), one spot, Rf =0.23 (CH3CO2Et/
n-PrOH/H2O, 7:7:5); MS (MALDI-TOF): m/z calcd for [M+H]+:
1150.54; found: 1150.06.
Synthesis of SPDP–DNA conjugate: The purified 3’- or 5’-end amino-
propyl-linked ODN (100 nmol) was dissolved in carbonate Na buffer
(100 mL, 0.5m, pH 9.3). To this solution, SPDP (1.5 mg, 4.6 mmol) dis-
solved in DMSO (50 mL) was added. The resulting suspension was stirred
at ambient temperature overnight. The solution was diluted to 400 mL
with water. The mixture was purified by RP-HPLC under the following
conditions. Column: Wakosil-II 5C18 RS, room temperature, flow rate:
1.0 mLminÀ1, eluent A: TEAA (triethylamine–acetic acid, 0.1m, pH 7.0),
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
9
&
ÞÞ
These are not the final page numbers!

S. Nishizawa, T. Ihara et al.
eluent B: acetonitrile, linear gradient: 5–30% B in 30 min, detection
wavelength: 260 nm.
Acknowledgements
MS (MALDI-TOF): m/z calcd for [MÀH]À (3’-modified): 2746.46;
found: 2746.81; m/z calcd for [MÀH]À (5’-modified): 2724.50; found:
2724.37.
We thank Ms. M. Miyabe for assistance in the mass spectrometry meas-
urements for CyD–ODN conjugate identification. We also thank Prof. H.
Ihara and Dr. H. Jintoku for their support in measuring CD and fluores-
cence spectra. This work was supported by the Grant-in-Aid for Scientif-
ic Research on Innovative Areas (Coordination Programming, Area
2107) [No. 24108734 to T.I.], Scientific Research (B) [No. 24350040 to
T.I.] from the Ministry of Education, Culture, Sports, Science and Tech-
nology, Japan; and the Health Labour Sciences Research Grant (Re-
search on Publicly Essential Drugs and Medical Devices) [to T.I.] from
the Ministry of Health and Labour, Japan.
Synthesis of b-CyD–DNA conjugate (CyD–ODN): SPDP–DNA conju-
gate (50 nmol) was dissolved in phosphate–Na buffer (100 mL, 10 mm,
pH 7.2). To this solution was added monothiolated b-CyD (5.8 mg, 5.0
mmol) dissolved in DMSO (60 mL). The resulting suspension was stirred
at ambient temperature overnight. The solution was diluted to 400 mL
with water. The mixture was purified by RP-HPLC under the following
conditions. Column: Wakosil-II 5C18 RS, room temperature, flow rate:
1.0 mLminÀ1, eluent A: 0.1m TEAA (pH 7.0), eluent B: acetonitrile,
linear gradient: 5–30% B in 30 min, detection wavelength: 260 nm.
[1] D. M. Kolpashchikov, Chem. Rev. 2010, 110, 4709–4723.
[2] A. B. Iliuk, L. Hu, W. A. Tao, Anal. Chem. 2011, 83, 4440–4452.
[3] J. Liu, Z. Cao, Y. Lu, Chem. Rev. 2009, 109, 1948–1998.
[4] A. C. Syvꢃnen, Nat. Rev. Genet. 2001, 2, 930–942.
[5] V. Lyamichev, A. L. Mast, J. G. Hall, J. R. Prudent, M. W. Kaiser, T.
Takova, R. W. Kwiatkowski, T. J. Sander, M. de Arruda, D. A. Arco,
B. P. Neri, M. A. D. Brow, Nat. Biotechnol. 1999, 17, 292–296.
[6] S. Tyagi, F. R. Kramer, Nat. Biotechnol. 1996, 14, 303–308.
[7] B. Dubertret, M. Calame, A. J. Libchaber, Nat. Biotechnol. 2001, 19,
365–370.
[8] H. Kashida, T. Takatsu, T. Fujii, K. Sekiguchi, X. Liang, K. Niwa, T.
Takase, Y. Yoshida, H. Asanuma, Angew. Chem. 2009, 121, 7178–
7181; Angew. Chem. Int. Ed. 2009, 48, 7044–7047.
[9] T. Ihara, M. Mukae, Anal. Sci. 2007, 23, 625–629.
[10] Y. N. Teo, E. T. Kool, Chem. Rev. 2012, 112, 4221–4245.
[11] J. D. Wulfkuhle, L. A. Liotta, E. F. Petricoin, Nat. Rev. Cancer 2003,
3, 267–275.
[12] Y. Urano, Anal. Sci. 2008, 24, 51–53.
[13] J. Li, W. Zhou, X. Ouyang, H. Yu, R. Yang, W. Tan, J. Yuan, Anal.
Chem. 2011, 83, 1356–1362.
[14] D. C. Harris, B. R. Saks, J. Jayawickramarajah, J. Am. Chem. Soc.
2011, 133, 7676–7679.
[15] K. Gorska, A. Manicardi, S. Barluenga, N. Winssinger, Chem.
Commun. 2011, 47, 4364–4366.
[16] Y. Sato, S. Nishizawa, K. Yoshimoto, T. Seino, T. Ichihashi, K.
Morita, N. Teramae, Nucleic Acids Res. 2009, 37, 1411–1422.
[17] Y. Sato, T. Ichihashi, S. Nishizawa, N. Teramae, Angew. Chem. 2012,
124, 6475–6478; Angew. Chem. Int. Ed. 2012, 51, 6369–6372.
[18] J. Wu, Y. Zou, C. Li, W. Sicking, I. Piantanida, T. Yi, C. Schmuck, J.
Am. Chem. Soc. 2012, 134, 1958–1961.
[19] K. Onizuka, Y. Taniguchi, S. Sasaki, Bioconjugate Chem. 2009, 20,
799–803.
[20] Y. Taniguchi, R. Kawaguchi, S. Sasaki, J. Am. Chem. Soc. 2011, 133,
7272–7275.
[21] N. Tabatadze, C. Tomas, R. McGonigal, B. Lin, A. Schook, A. Rout-
tenberg, Hippocampus 2012, 22, 1228–1241.
[22] S.-H. Fujishima, R. Yasui, T. Miki, A. Ojida, I. Hamachi, J. Am.
Chem. Soc. 2012, 134, 3961–3964.
[23] P. Arslan, A. Jyo, T. Ihara, Org. Biomol. Chem. 2010, 8, 4843–4848.
[24] M. Mukae, T. Ihara, M. Tabara, A. Jyo, Org. Biomol. Chem. 2009, 7,
1349–1354.
[25] T. Ihara, T. Fujii, M. Mukae, Y. Kitamura, A. Jyo, J. Am. Chem.
Soc. 2004, 126, 8880–8881.
MS (MALDI-TOF): m/z calcd for [MÀH]À (3’-modified): 3771.77; found:
3770.94; m/z calcd for [MÀH]À (5’-modified): 3749.81; found: 3748.90.
Synthesis of MNDS: MNDS was synthesized as described previously.[36]
The outline of the procedure is as follows. Aminonaphthyridine was cou-
pled with activated ester of N-protected aminopropionate to form a
naphthyridine connected with a protected aminolinker chain through an
amide bond. After deprotection, it was coupled with dansyl chloride to
obtain MNDS.
Synthesis of MNDB: MNDB was synthesized as described previously.[53]
The outline of the procedure is as follows. Aminonaphthyridine was
chlorinated and then coupled with 1,2-diaminoethane to form a naphthyr-
idine derivative with an aminolinker chain, which is tethered through the
secondary amine. Finally, it was coupled with DBD-F (7-fluoro-4-(N,N-
dimethylaminosulfonyl)benzofurazan) to obtain MNDB.
Synthesis of 3,5-diamino-N-(2-((2-aminoethyl)amino)ethyl)-6-chloropyra-
zine-2-carboxamide:[54] Methyl-3,5-diamino-6-chloropyrazine-2-carboxy-
late (1.0 g, 4.9 mmol) was added to diethylenetriamine (1.5 g, 15 mmol)
and stirred for 24 h at 1008C. The solution was suspended in CHCl3 and
subsequently extracted with HCl (0.1m). The aqueous layer was neutral-
ized with NaOH and extracted with CHCl3. The organic layer was con-
centrated in vacuo and the crude residue was purified by column chroma-
tography on amino-group-modified silica gel (CHCl3/MeOH). The ob-
tained residue was identified by MALDI-TOF MS and was used for the
next step without further purification.
Synthesis of DPDB: To a mixture of 3,5-diamino-N-(2-((2-aminoethyl)a-
mino)ethyl)-6-chloropyrazine-2-carboxamide (50 mg, 0.21 mmol) and
DBD-F (70 mg, 0.31 mmol) in DMF (8 mL), triethylamine (4 mL) was
added and the reaction mixture was stirred and refluxed under N2 atmos-
phere for 12 h. The solvent was concentrated in vacuo and the crude
product was purified by column chromatography on an amino-group-
modified silica gel (CHCl3/MeOH).
1
Yellow solid 29 mg (28%); H NMR (400 MHz, DMSO): d=7.87 (d, 1H,
J=7.5 Hz), 7.51 (br, 1H), 6.85 (br, 1H), 6.09 (d, 1H, J=7.5 Hz), 5.11 (s,
2H), 3.51 (m, 2H), 3.42 (m, 2H), 3.08 (m, 2H), 2.87 (m, 2H), 2.85 ppm
(s, 6H); MS (ESI): m/z calcd for C17H24ClN10O4S: 499.1391 [M+H]+;
found: 499.1386.
Circular dichroism (CD) measurements: CD spectra were obtained using
a JASCO J-725 spectropolarimeter equipped with a Peltier thermal con-
troller. CD spectra were measured from 360 to 200 nm in a 0.1 cm path
length cell at 08C during N2 purging to prevent moisture condensation
on the cell. The concentration of the samples was 1.25 mm in 10 mm phos-
phate buffer (pH 7.0) containing NaCl (1m).
[26] T. Ihara, Y. Kitamura, Y. Tsujimura, A. Jyo, Anal. Sci. 2011, 27,
585–590.
[27] Y. Kitamura, T. Ihara, Y. Tsujimura, Y. Osawa, D. Sasahara, M. Ya-
mamoto, K. Okada, M. Tazaki, A. Jyo, J. Inorg. Biochem. 2008, 102,
1921–1931.
Fluorescence measurements: Fluorescence measurements were per-
formed at 08C using a JASCO FP-6500 spectrofluorometer and a Perki-
nElmer LS55 equipped with a Peltier thermal controller during N2 purg-
ing to prevent moisture condensation on the quartz cell. Each of the re-
porter molecules (5.0 mm) was added into the solution of N-gap duplexes
(1.0 mm) dissolved in phosphate buffer (10 mm, pH 7.0), NaCl (1m), and
DMSO (0.83%), and subjected to a measurement. Normalized fluores-
cence intensities were estimated after subtraction of the signal of the re-
porter molecule alone from each measurement.
[28] T. Ihara, T. Wasano, R. Nakatake, P. Arslan, A. Futamura, A. Jyo,
Chem. Commun. 2011, 47, 12388–12390.
[29] T. Ogoshi, A. Harada, Sensors 2008, 8, 4961–4982.
[30] A. Ueno, S. Minato, I. Suzuki, M. Fukushima, M. Ohkubo, T. Osa,
F. Hamada, K. Murai, Chem. Lett. 1990, 19, 605–608.
[31] R. Corradini, A. Dossena, R. Marchelli, A. Panagia, G. Sartor, M.
Saviano, A. Lombardi, V. Pavone, Chem. Eur. J. 1996, 2, 373–381.
&
10
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Cooperative Recognition of Specific Nucleobases
FULL PAPER
[32] Molecular encapsulation: Organic reactions in constrained systems
(Eds.: U. H. Brinker, J.-L. Mieusset), Wiley, Chichester, 2011.
[33] A. Kuzuya, T. Ohnishi, T. Wasano, S. Nagaoka, J. Sumaoka, T.
Ihara, A. Jyo, M. Komiyama, Bioconjugate Chem. 2009, 20, 1643–
1649.
[34] H. Takakusa, K. Kikuchi, Y. Urano, T. Higuchi, T. Nagano, Anal.
Chem. 2001, 73, 939–942.
[35] H.-J. Schneider, A. K. Yatsimirsky, Principles and Methods in Supra-
molecular Chemistry, Wiley, Chichester, 2000.
[36] T. Ihara, A. Uemura, A. Futamura, M. Shimizu, N. Baba, S. Nishiza-
wa, N. Teramae, A. Jyo, J. Am. Chem. Soc. 2009, 131, 1386–1387.
[37] K. Yoshimoto, S. Nishizawa, M. Minagawa, N. Teramae, J. Am.
Chem. Soc. 2003, 125, 8982–8983.
[38] B. Rajendar, A. Rajendran, Z. Ye, E. Kanai, Y. Sato, S. Nishizawa,
M. Sikorski, N. Teramae, Org. Biomol. Chem. 2010, 8, 4949–4959.
[39] Y. Sato, Y. Zhang, T. Seino, T. Sugimoto, S. Nishizawa, N. Teramae,
Org. Biomol. Chem. 2012, 10, 4003–4006.
[40] B. Rajendar, A. Rajendran, Y. Sato, S. Nishizawa, N. Teramae,
Bioorg. Med. Chem. 2009, 17, 351–359.
[41] S. Miura, S. Nishizawa, A. Suzuki, Y. Fujimoto, K. Ono, Q. Gao, N.
Teramae, Chem. Eur. J. 2011, 17, 14104–14110.
[44] W. A. Baase, W. C. Johnson, Jr., Nucleic Acids Res. 1979, 6, 797–
814.
[45] C. R. Cantor, P. R. Schimmel, Biophysical Chemistry: the Behavior
of Biological Macromolecules, W. H. Freeman, New York, 1980.
[46] R. E. Dickerson, H. R. Drew, B. N. Conner, R. M. Wing, A. V. Frati-
ni, M. L. Kopka, Science 1982, 216, 475.
[47] S. Neidle, Oxford Handbook of Nucleic Acid Structure, Oxford Uni-
versity Press, Oxford, 1999.
[48] Y. Sato, T. Kageyama, S. Nishizawa, N. Teramae, Anal. Sci. 2013, 29,
15–19.
[49] Z. Ye, B. Rajendar, D. Qing, S. Nishizawa, N. Teramae, Chem.
Commun. 2008, 6588–6590.
[50] H. Dodziuk, Cyclodextrins and Their Complexes. Wiley-VCH, Wein-
heim. 2006.
[51] F. S. Damos, R. Luz, A. A. Sabino, M. N. Eberlin, R. A. Pilli, L. T.
Kubota, J. Electroanal. Chem. 2007, 601, 181–193.
[52] K. Fujita, T. Ueda, T. Imoto, I. Tabushi, N. Toh, T. Koga, Bioorg.
Chem. 1982, 11, 108–114.
[53] C.-X. Wang, Y. Sato, M. Kudo, S. Nishizawa, N. Teramae, Chem.
Eur. J. 2012, 18, 9481–9484.
[54] T. Russ, W. Ried, F. Ullrich, E. Mutschler, Arch. Pharm. 1992, 325,
761–767.
[42] A. Rajendran, C. Zhao, B. Rajendar, V. Thiagarajan, Y. Sato, S.
Nishizawa, N. Teramae, Biochim. Biophys. Acta Gen. Subj. 2010,
1800, 599–610.
[43] C. Zhao, A. Rajendran, Q. Dai, S. Nishizawa, N. Teramae, Anal. Sci.
2008, 24, 693–695.
Received: March 14, 2013
Published online: && &&, 0000
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
11
&
ÞÞ
These are not the final page numbers!