Bulge-like asymmetric heterodye clustering in DNA duplex results in efficient quenching of background emission based on the maximized excitonic interaction

Article abstract of DOI:10.1002/chem.201201365

Asymmetric dye clusters with a single fluorophore (Cy3) and multiple quenchers (4′-methylthioazobenzene-4-carboxylate, methyl red, and 4′-dimethylamino-2-nitroazobenzene-4-carboxylate) were prepared. The dye and one-to-five quenchers were tethered through

Full text of DOI:10.1002/chem.201201365

FULL PAPER
DOI: 10.1002/chem.201201365
Bulge-like Asymmetric Heterodye Clustering in DNA Duplex Results in
Efficient Quenching of Background Emission Based on the Maximized
Excitonic Interaction
Taiga Fujii,[a] Yuichi Hara,[a] Takuya Osawa,[a] Hiromu Kashida,*[a, b] Xingguo Liang,[a, b]
Yasuko Yoshida,[b] and Hiroyuki Asanuma*[a, b]
Abstract: Asymmetric dye clusters
with a single fluorophore (Cy3) and
multiple quenchers (4’-methylthioazo-
benzene-4-carboxylate, methyl red, and
4’-dimethylamino-2-nitroazobenzene-4-
carboxylate) were prepared. The dye
and one-to-five quenchers were teth-
ered through d-threoninol to opposite
strands of a DNA duplex. NMR analy-
sis revealed that the clusters with
a single fluorophore and two quenchers
formed a sandwich-like structure (anti-
parallel H-aggregates). The melting
temperatures of all the heteroclusters
were almost the same, although struc-
tural distortion should become larger,
as the number of quenchers increased.
An asymmetric heterocluster of
a single fluorophore and two quenchers
showed larger excitonic interaction
(i.e., hypochromicity of Cy3), than did
a single Cy3 and a single quencher.
Due to the larger exciton coupling be-
tween the dyes, the 1:2 heterocluster
suppressed the background emission
more efficiently than the 1:1 cluster.
However, more quenchers did not en-
hance quenching efficiency due to the
saturation of exciton coupling with two
quenchers. Finally, this asymmetric 1:2
heterocluster was introduced into the
stem region of a molecular beacon
(MB; also known as an in-stem MB)
targeting the fusion site in the L6
BCR-ABL fusion gene. With this MB
design, the signal/background ratio was
as high as 68 due to efficient suppres-
sion of background emission resulting
from the maximized excitonic interac-
tion.
Keywords: cluster compounds
·
coherent quenching · DNA · dyes/
pigments · NMR spectroscopy
Introduction
room temperature.[4] The relationship between cluster size
and excitonic interaction in heteroclusters of different dyes
has scarcely been investigated due to the difficulty of their
preparation.[5] Especially, only a limited number has been
reported on the exciton coupling between a dye and the
other dyes (1:n type heteroclusters).
The excitonic interaction is a static interaction among close-
ly stacked dye molecules[1] that induces narrowing of the ab-
sorption band and a bathochromic or a hypsochromic shift
depending on stacking geometry.[2] H- and J-bands that
show large hypsochromicity and bathochromicity due to ver-
tical and stair-like stacking, respectively, are typical exam-
ples of effects of excitonic interactions.[2–3] UV/Vis spectra
depend on the size of the dyes as well as their mutual orien-
tations. Notably, in homodye clusters, exciton coupling
among the cluster (spectroscopic aggregation number or co-
herence length) is saturated with 10–20 dye molecules at
Recently, DNA and RNA have been used as scaffolds to
prepare organized assemblies of dyes: nucleic acid scaffolds
allow control of the number of the dyes in an array and of
their mutual orientation.[6] For example, Matray and Kool
reported oligofluorosides composed of fully artificial base
surrogates tethered to fluorophores: these single-stranded
scaffolds based on the d-ribose were used to prepare various
functional heteroclusters.[7] Hꢀner and co-workers developed
heteroclusters of dyes linked to dialkynylpylene–perylene-
diimide pairs that were introduced into DNA duplexes.[8]
Wagenknecht and co-workers incorporated 3-amino-1,2-pro-
panediols (C2 linkers) in a DNA duplex by placing abasic
sites at the counterpart of the base surrogates to ensure in-
terstrand helical assembly.[9] We have also developed a new
methodology to prepare organized assemblies of homo- and
heterodyes by using d-threoninol as a scaffold.[10] Introduc-
tion of base surrogates linked to dyes into the centers of
two complementary DNAs resulted in a stacked H-type dye
cluster in antiparallel manner in the duplex. Because it is
difficult to prepare well-defined heterodye clusters, this
method was applied first to verify molecular exciton theory
[a] Dr. T. Fujii, Y. Hara, T. Osawa, Dr. H. Kashida, Prof. Dr. X. Liang,
Prof. Dr. H. Asanuma
Department of Molecular Design and Engineering
Graduate School of Engineering, Nagoya University
Furo-cho, Chikusa-ku, Nagoya 464-8603 (Japan)
Fax : (+81)52-789-2528
E-mail: asanuma@mol.nagoya-u.ac.jp
kashida@mol.nagoya-u.ac.jp
[b] Dr. H. Kashida, Prof. Dr. X. Liang, Y. Yoshida,
Prof. Dr. H. Asanuma
Innovative Research Center for PME
Nagoya University
Furo-cho, Chikusa-ku, Nagoya 464-8603 (Japan)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201201365.
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

of heteroclustered dyes from scientific point of view.[10f]
With the exception of spectral shifts, experimental results
qualitatively agreed with theoretical predictions. Thus, het-
erodimerization induced hyperchromism of the shorter
wavelength band (in-phase transition), but hypochromism of
the longer wavelength band. These absorbance changes
were enhanced, when the Dlmax of the two dyes was de-
creased.
Next, our method was applied to design a type of highly
sensitive molecular beacon (MB), the in-stem MB (ISMB),
by introducing a heterodimer of fluorophore and quencher
into the stem region of MB to facilitate the quenching of
emission in the closed state.[11] Interestingly, as Johansson
et al. proposed,[12] our systematic studies on the spectroscop-
ic behavior of fluorophore–quencher dimer revealed that
a stronger excitonic interaction between the dyes suppressed
the background emission in the closed MB more efficiently
and resulted in a higher signal/background (S/B) ratio. Fur-
ther enhancement of excitonic interaction between the dyes
should raise quenching efficiency, leading to highly sensitive
detection of a target DNA. Herein, we incorporated a single
fluorophore (Cy3) and several quenchers (azo dyes) on d-
threoninols into a duplex (Scheme 1). We found that an
Scheme 2. Chemical structures and sequences of the ODNs synthesized
in this study. In the MB sequences, the stem regions are underlined; the
italicized region denotes the loop that recognizes the target sequence.
Scheme 1. Schematic illustration of the design of the asymmetric dye
clusters by using d-threoninol as dye tethers. One strand with a fluoro-
phore and the other strand with 1 to 5 quenchers (n denotes the number
of the quenchers) were hybridized resulting in dye clustering.
heterocluster duplex was conducted to determine the align-
ment of the dyes, and then the stabilities of 1:n clustered du-
plexes were evaluated.
asymmetric heterocluster of a single fluorophore and two
quenchers suppressed the background emission more effi-
ciently than single fluorophore and quencher pair due to the
larger exciton coupling of the duplex with two quenchers.
Based on this result, a highly sensitive ISMB was designed
with asymmetric 1:2 heterocluster in the stem region.
NMR analysis of 1:2 heterocluster: To discriminate proton
signals of all the dye molecules in the duplex, three different
dyes were introduced: 4’-dimethylaminoazobenzene-4-car-
boxylate (methyl red, M), 4’-methylthioazobenzene-4-car-
boxylate (S), and 4’-methoxyazobenzene-4-carboxylate (O).
Two complementary strands, NMR-MSa and NMR-Ob,
were synthesized (Scheme 2). NMR-MSa has M and S resi-
dues in the center, and NMR-Ob has an O. There are sever-
al possible alignments of the dyes as shown in Scheme 3.[13]
The imino region (d=11.5–14.0 ppm) of the 1D NMR spec-
trum of the duplex of these two strands recorded in H2O is
depicted in Figure 1. Signals corresponding to the six natural
base pairs (C1–G12, G2–C11, A3–T10, G4–C9, T5–A8, and C6–
G7) were assigned on the basis of NOESY and chemical
shifts, indicating that the non-natural M, S, and O did not in-
terrupt the base pairing.[14] Figure 2 depicts the NOEs be-
tween the imino-proton signal (d=11.5–14.0 ppm) and the
aromatic-proton signal (d=6.0–6.7 ppm) regions. A distinct
Results and Discussion
Structures and stabilities of duplexes with asymmetric dye
clusters: Scheme 1 shows an illustration of a 1:n asymmetric
dye cluster formed in the middle of the duplex. One strand
has single dye and the other has consecutive dye residues at
the center, and their hybridization affords a 1:n asymmetric
dye cluster when the dyes are stacked with each other. Se-
quences of the DNAs incorporating dyes synthesized in this
study are listed in Scheme 2. First, NMR analysis of a 1:2
&
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!

Asymmetric Heterodye Clustering in DNA Duplex
FULL PAPER
have NOE to M, but was in close proximity to H8 (H12) of
the S residue. Thus, we could unambiguously exclude the
possibilities that the single dye (shown in Scheme 3 in
white) stacks with a natural base (shown in Scheme 3, pan-
els b and c).
Figure 3 shows the NOESY between the regions of d=
5.0–8.0 and 1.6–3.0 ppm. Proton signals from the O residue
were correlated only with those of M and S residues, and no
Scheme 3. Schematic illustration of four possible alignments of the asym-
metric dye clusters. The dye (white) and quenchers (black and gray) are
located on opposite strands. a) The dye is located between quenchers
(corresponding to MOS orientation in NMR-MSa/NMR-Ob); b) natural
nucleobases are to the 5’-side of the dye (corresponding to MSO orienta-
tion); c) natural nucleobase are to the 3’-side of the dye (corresponding
to OMS orientation); or d) the dye or the quencher is flipped out from
the duplexes.
Figure 1. Imino region of the 1D NMR spectrum of the NMR-MSa/
NMR-Ob duplex in H2O/D2O 9:1 at 278 K in phosphate buffer (20 mm,
pH 7.0) and NaCl (200 mm). The duplex concentration was 1.0 mm. As-
signments of the imino protons and the residue number are denoted
above the peaks.
Figure 3. 2D NOESY spectrum (mixing time=150 ms) between the re-
gions of d=5.0–8.0 and 1.6–3.0 ppm for the NMR-MSa/NMR-Ob duplex
in H2O/D2O 9:1 at 278 K in phosphate buffer (20 mm, pH 7.0) and NaCl
(200 mm).
cross-peaks were observed between M and S residues as
schematically shown in Figure 4a. Cross-peaks were ob-
À
À
served between aromatic proton signals of O H2 (O H6)
À
À
À
À
and O H3 (O H5) with those of both S CH3 of S and N
CH3 of M. Moreover, relatively strong signals were also ob-
À
À
served between O CH3 of O and aromatic protons of M
À
À
À
À
À
À
À
H2 (M H6), M H3 (M H5), S H2 (S H6), and S H3 (S
H5). These NOE signals indicate that M, O, and S were
stacked as depicted in Figure 4a and in Scheme 3a: The O
residue is sandwiched between M and S residues in the
DNA duplex. The computer model of NMR-MSa/NMR-Ob
was generated by using the InsightII/Discover3 software,
and it is entirely consistent with the NMR analyses (Fig-
ure 4b).
Figure 2. 2D NOESY spectrum (mixing time=150 ms) between the
imino-proton signals (d=11.5–14.5 ppm) and the aromatic-proton signals
(d=6.0–6.7 ppm) for the NMR-MSa/NMR-Ob duplex in H2O/D2O 9:1 at
278 K in phosphate buffer (20 mm, pH 7.0) and NaCl (200 mm). Assign-
ments of the methyl red (M) and 4’-methylthioazobenzene (S) protons
are denoted on the 1D spectra (F1 axis) by using the numbers designated
in Scheme 2. The NOE signals surrounded by broken circles demonstrate
intercalation of the methyl red and 4’-methylthioazobenzene.
Duplex stabilities of the asymmetric dye clusters: Introduc-
tion of several quencher dyes should induce a severe distor-
tion of the duplex structure due to the increased asymmetry
of the strands, which should affect the stability of the
duplex. Thus, melting temperatures (Tm) of duplexes with
1 to 5 quenchers across from a single dye was measured
(Table 1). In these experiments, the dye is not 4’-methoxy-
azobenzene (O), but rather Cy3 (Y); however, we assumed
that a sandwich-type cluster was formed. Insertion of single
Y–R pair (Y1a/R1b: 44.58C) slightly decreased the stability
NOE signal was observed between the imino proton of T10
and H8 (H12) of the M protons, indicating that the M resi-
due was located near T10. The imino proton of G4 did not
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!

H. Kashida, H. Asanuma et al.
Table 1. Effect of the number and quencher dyes on the melting temper-
ature (Tm).[a]
Tm [8C]
Y1a/Rnb
Mna/Y1b
Sna/Y1b
N/C
47.7
44.5
44.8
46.1
47.1
47.1
–
–
n=1
n=2
n=3
n=4
n=5
45.3
46.0
47.5
48.2
49.3
46.2
48.2
49.2
50.0
49.1
[a] Solution conditions were as follows: [DNA]=5 mm, [NaCl]=100 mm,
phosphate buffer (10 mm, pH 7.0).
Table 1), although distortion presumably increased dramati-
cally. Similar results were observed when M or S was used
as the quencher instead of R. These results are in contrast
with our previous observation that asymmetric combinations
of natural DNA with a strand containing a methyl red conju-
gate markedly destabilized the duplex.[10e,15] Presumably,
stacking of Cy3 with quenchers linked through d-threoninols
offset the destabilizing effect of the increased distortion.
Effect of the number of quenchers on excitonic interaction
among the dyes: Next, the effect of the number of quencher
dyes on excitonic interaction was investigated. To estimate
the strength of exciton coupling, hypochromicity of Cy3 ab-
sorption by the quencher(s) was quantified by the compari-
son of UV/Vis spectrum of Cy3 before and after hybridiza-
tion of quencher strand. Figure 5a shows typical excitonic
interaction between Cy3 in Y1a and R in R1b. Each single-
stranded Y1a (pink line) and R1b (purple line) gave ab-
sorption maxima at l=549 and 512 nm assigned to Cy3 and
R, respectively, whereas hybridization of these two strands
resulted in a blueshift (hypsochromicity) of the R band at
l=513 nm and decrease in absorbance (hypochromicity) of
Cy3 band at l=550 nm compared with the summed spec-
trum (blue line). Herein, the decrease in the absorbance
(Figure 5a, black arrow) corresponds to the strength of the
excitonic interaction, which was estimated as 0.32 (Table 2).
An increase in the excitonic interaction was observed with
Y1a/R2b (strength of excitonic interaction 0.47) as shown
in Figure 5b and Table 2. However, introduction of addition-
al quencher dyes did not significantly increase the interac-
tion (Table 2 and Figure S1 in the Supporting Information),
indicating that the excitonic interaction between Y and R
was saturated with two quenchers. The other quenchers
(methyl red and 4’-methylthioazobenzene) showed the same
saturation behavior (Table 2, Figures S2 and S3 in the Sup-
porting Information), although the maximum strength of the
excitonic interaction depended on the dye.[16] Based on
these results, we conclude that introduction of two quench-
ers results in the maximum excitonic interaction among a flu-
orophore and quenchers.
Figure 4. a) Stacking and orientations of M, O, and S residues in the
duplex determined from NOESY. Broken lines indicate the observed
NOE signals. b) Energy-minimized structure of the NMR-MSa/NMR-Ob
duplex calculated with InsightII/Discover3 from the initial structure de-
termined by NMR analyses. 4’-Methoxyazobenzene, 4’-methylthioazo-
benzene, and methyl red are shown in cyan, yellow, and red, respectively.
relative to the native duplex (N/C: 47.78C). The Tm of the
cluster with a 1:2 bulge (Y1a/R2b: 44.88C) was almost
same as that with a Y–R pair. Interestingly, addition of
more R residues did not affect Tm significantly (46.1–47.18C,
Quenching efficiency of a fluorophore: We previously re-
ported that a larger excitonic interaction between a fluoro-
phore and a quencher induced more efficient quenching 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!

Asymmetric Heterodye Clustering in DNA Duplex
FULL PAPER
the fluorophore.[11b] Emission
spectra of Sna/Y1b and Y1a/
Rnb are shown in Figure 5c
and d. As shown in the insets,
all clusters (Y1a/Rnb and Sna/
Y1b) exhibited low fluores-
cence compared with that of
single-stranded Y1a and Y1b.
We found that two quenchers
(S2a/Y1b) reduced emission
more (96.8%) than
a single
quencher (S1a/Y1b; 93.9%).
However, addition of more S
residues did not significantly
improve quenching efficiency.
4’-Dimethylamino-2-nitroazo-
benzene quenched Cy3 more
efficiently than S due to the
stronger excitonic interaction
(compare blue lines in Fig-
ure 5d and c) as reported pre-
viously.[11b] However, more than
two R residues did not further
improve the quenching efficien-
cy (Table 2). The same effect
was also observed with methyl
red (Figure S5 in the Support-
ing Information). We conclude
that two consecutive quenchers
maximized the quenching of
a fluorophore due to the satu-
rated excitonic interaction.
Figure 5. a) UV/Vis spectra of the Y1a/R1b duplex (blue line), single-stranded Y1a (pink dotted line), R1b
(purple dashed line), and a simple sum of their spectra (green line); b) UV/Vis spectra of the Y1a/R2b duplex
(blue line), single-stranded Y1a (pink dotted line), R2b (purple dashed line), and a simple sum of their spec-
tra (green line). Hypochromicity (differences of absorbance of Cy3) before and after clustering were defined
as strength of the exciton couplings (shown by dotted arrows). Solution conditions of UV/Vis spectra were
DNA (5 mm), phosphate buffer (10 mm, pH 7.0), NaCl (100 mm); all spectra were recorded at 208C. Fluores-
cence emission spectra of Cy3 (Y) excited at l=546 nm and the duplexes containing c) 4’-methylthioazobe-
nene and d) 4’-dimethylamino-2-nitroazobenzene as quenchers. DNA concentrations of fluorescence spectra
were [Sna]=[Rnb]=0.4 mm, [Y1a]=[Y1b]=0.2 mm. Solutions contained phosphate buffer (10 mm, pH 7.0),
NaCl (100 mm); all spectra were recorded at 208C.
Design of highly sensitive in-
stem MB: Based on the above
quenching
experiments,
ISMBs[17] with two consecutive
R residues as quenchers in the stem region, referred to as
MB2 and MB4, targeting a fusion site of L6 BCR-ABL
fusion gene was designed.[18] Control ISMBs with single
quencher, MB1 and MB3, were also synthesized. Back-
ground emission of MB2 was 56% lower than that of MB1
in the absence of target (Figure 6a). Consequently, the S/B
ratio, which was defined as the ratio of emission intensity of
MB2 in the presence versus absence of the target, is 68,
whereas that of MB1 is 27 (compare Figure 6b and c). This
remarkable improvement of S/B ratio was also observed for
MB3 and MB4 with the fluorophore–quencher pair at a dif-
ferent position in the ISMB (Figure S6 in the Supporting In-
formation). Thus, highly sensitive ISMBs were successfully
prepared by using the double-quencher design.
Table 2. Effect of the number and type of quencher dye on the exciton
couplings of the quenchers to a fluorophore and the quenching efficien-
cies.
Y1a/Rnb
Mna/Y1b
Sna/Y1b
H[a,b]
QE[c] [%]
H[a,b]
QE[c] [%]
H[a,b]
QE[c] [%]
n=1
n=2
n=3
n=4
n=5
0.32
0.47
0.48
0.47
0.52
98.5
99.0
99.0
99.3
99.0
0.25
0.36
0.38
0.36
0.39
96.5
97.3
96.8
96.4
97.7
0.10
0.15
0.18
0.17
0.19
93.9
96.8
96.8
97.3
97.1
[a] Solutions contained DNA (5 mm), phosphate buffer (10 mm, pH 7.0),
NaCl (100 mm); all spectra were recorded at 208C. [b] Hypochromicity
(H) was evaluated from difference of absorbance between the heteroclus-
ters and the sum spectrum of each single strand at Cy3 absorption region
(Y1a/Rnb; l=549 nm, Mna/Y1b; l=552 nm, Sna/Y1b; l=552 nm).
[c] Quenching efficiencies (QE) are the differences in fluorescent intensi-
ties between the clusters and the single strand with a Cy3 residue. The
emission spectra of Y1a/Rnb, Mna/Y1b, and Sna/Y1b at l=564, 563,
and 562 nm, respectively, excited at l=546 nm were used for these calcu-
lations. DNA concentrations were [Mna]=[Sna]=[Rnb]=0.4 mm,
[Y1a]=[Y1b]=0.2 mm. Solutions contained phosphate buffer (10 mm,
pH 7.0), NaCl (100 mm); all spectra were recorded at 208C.
Conclusion
We have prepared asymmetric dye clusters in DNA duplex-
es by using d-threoninol as a scaffold for the dyes. In-stem
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!

H. Kashida, H. Asanuma et al.
Experimental Section
Materials: All conventional phosphoramidite monomers, CPG columns,
reagents for DNA synthesis, and Poly-Pak II cartridges were purchased
from Glen Research. Other reagents for the synthesis of phosphorami-
dite monomers were purchased from Tokyo Kasei and Sigma–Aldrich.
Synthesis of DNA modified with M, O, R, S, or Y: All the modified
DNAs were synthesized by using an automated DNA synthesizer (ABI-
3400 DNA synthesizer, Applied Biosystems) by using conventional and
dye-carrying phosphoramidite monomers. 4’-Methylthioazobenzene,
methyl red, and 4’-dimethylamino-2-nitroazobenzene phosphoramidite
monomers were synthesized as reported previously.[10e,f] The compound
4’-methoxyazobenzene was synthesized according to the literature proce-
dure.[19] Dyes were converted to phosphoramidite monomers as described
in the Supporting Information, Scheme S1. The coupling efficiencies of
the monomers with modified residues were as high as those of the con-
ventional monomers as was judged from the intensity of the color of the
released trityl cation. After work-up, the monomers were purified by re-
verse-phase HPLC and characterized by MS (MALDI-TOF; Autoflex II,
Bruker Daltonics).
MS (MALDI-TOF) data: found versus calculated (calcd) for the mono-
mers: M1a, found: 4062 (calcd for [M1a+H+]: 4063); M2a: found 4481
(calcd for [M2a+H+]: 4481); M3a: found 4899 (calcd for [M3a+H+]:
4899); M4a: found 5318 (calcd for [M4a+H+]: 5317); M5a: found 5735
(calcd for [M5a+H+]: 5735); S1a: found 4067 (calcd for [S1a+H+]:
4066); S2a: found 4488 (calcd for [S2a+H+]: 4487); S3a: found 4909
(calcd for [S3a+H+]: 4908). S4a: found 5332 (calcd for [S4a+H+]:
5329); S5a: found 5752 (calcd for [S5a+H+]: 5750); R1b: found 4108
(calcd for [R1b+H+]: 4108); R2b: found 4572 (calcd for [R2b+H+]:
4571); R3b: found 5034 (calcd for [R3b+H+]: 5034); R4b: found 5501
(calcd for [R4b+H+]: 5497); R5b: found 5966 (calcd for [R5b+H+]:
5960); Y1a: found 4234 (calcd for [Y1a+H+]: 4237); Y1b: found 4235
(calcd for [Y1b+H+]: 4237); NMR-MSa: found 2630 (calcd for [NMR-
MSa+H+]: 2632); NMR-Ob: found 2198 (calcd for [NMR-Ob+H+]:
2197); MB1: found 10300 (calcd for [MB1+H+]: 10305). MB2: found
10772 (calcd for [MB2+H+]: 10768); MB3: found 10302 (calcd for
[MB3+H+]: 10305); MB4: found 10769 (calcd for [MB4+H+]: 10768).
Determination of molar extinction coefficients: Molar extinction coeffi-
cients for dye monomers at l=260 nm were determined by comparison
[20]
of ratios between absorbances at l=260 nm and their lmax
.
These
ratios were calculated from UV/Vis spectra of the dye monomers (such
as compound 4 in Scheme S1 in the Supporting Information). The coeffi-
cients for dye monomers are listed in Table S1 in the Supporting Infor-
mation.
Spectroscopic measurements: The UV/Vis spectra were measured on
a JASCO model V-550 spectrophotometer and on a Shimadzu UV-1800
instrument in 10 mm quartz cells. Fluorescent spectra were measured
with a JASCO model FP-6500 with a microcell. The excitation wave-
length was 546 nm for Cy3. All models were equipped with programma-
ble temperature controllers. All samples of DNA-dye conjugates were
heated at 808C for 5 min in the dark to thermally isomerize the cis- to
the trans-form before spectroscopic measurement.[21] Before fluorescent
measurements of the MBs, all samples were cooled to 18CminÀ1 from
808C.
Figure 6. Fluorescence emission spectra of a) MB1 (broken line) and
MB2 (solid line) excited at l=546 nm without T-MB; b) MB1 with (solid
line) and without (broken line) T-MB; and c) MB2 with (solid line) and
without (broken line) T-MB. DNA concentrations were [T-MB]=0.4 mm,
[MB1]=[MB2]=0.2 mm. Solutions contained phosphate buffer (10 mm,
pH 7.0), NaCl (100 mm); all spectra were recorded at 208C.
Measurement of melting temperatures: The melting curves of duplex
DNAs were obtained by measurement of the change in absorbance at
l=260 nm versus temperature. The melting temperature (Tm) was deter-
mined from the maximum of the first derivative of the melting curve.
Both the heating and cooling curves were measured and the obtained Tm
values were within 2.08C.
MB prepared by using a single fluorescent dye and two
quenchers had quite low background emission. We are now
applying this probe design to high-throughput detection of
wild-type and chimeric gene expression associated with
chronic myeloid leukemia in a microarray format and
cells.[18]
NMR measurements: DNA duplexes were lyophilized three times from
buffer containing sodium phosphate (20 mm, pH 7.0) and then dissolved
in the a H2O/D2O 9:1 solution or D2O to give a duplex concentration of
1.0 mm. NaCl was added to give a final sodium concentration of 200 mm.
NMR spectra were recorded with
a Varian INOVA spectrometer
(700 MHz) equipped for triple resonance at a probe temperature of
278 K. Resonances were assigned by standard methods by using a combi-
&
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!

Asymmetric Heterodye Clustering in DNA Duplex
FULL PAPER
nation of 1D, total correlation spectroscopy (TOCSY, 60 ms mixing
time), double-quantum filtered (DQF)-COSY, and NOESY (150 ms
mixing time) experiments. All spectra in the H2O/D2O 9:1 solution were
recorded by using the 3:9:19 WATERGATE pulse sequence for water
suppression.[22]
Studer, V. L. Malinovskii, R. Hꢀner, Angew. Chem. 2011, 123, 5604–
5608; Angew. Chem. Int. Ed. 2011, 50, 5490–5494; d) S. M. Biner, D.
Kummer, V. L. Malinovskii, R. Haner, Org. Biomol. Chem. 2011, 9,
2628–2633.
[9] a) C. Wagner, H.-A. Wagenknecht, Org. Lett. 2006, 8, 4191–4194;
b) D. Baumstark, H.-A. Wagenknecht, Chem. Eur. J. 2008, 14, 6640–
6645.
Computer modeling: Molecular modeling by conformational energy min-
imization was performed with the InsightII/Discover3 software (Molecu-
lar Simulation, Inc.) on a Silicon Graphics O2+ workstation with the op-
erating system IRIX64, release 6.5. An AMBER force field was used for
the calculations. The results of the NMR analyses served as a starting
point for the modeling. For the analysis a trimer of methyl red, 4’-meth-
oxyazobenzene, and 4’-methylthioazobenzene, the trimer was prepared
by positioning the three dye molecules in a cofacial orientation replacing
three native base pairs.
[10] a) H. Asanuma, K. Shirasuka, M. Komiyama, Chem. Lett. 2002, 31,
490–491; b) H. Asanuma, K. Shirasuka, T. Takarada, H. Kashida,
M. Komiyama, J. Am. Chem. Soc. 2003, 125, 2217–2223; c) H. Ka-
shida, H. Asanuma, M. Komiyama, Angew. Chem. 2004, 116, 6684–
6687; Angew. Chem. Int. Ed. 2004, 43, 6522–6525; d) H. Kashida,
M. Tanaka, S. Baba, T. Sakamoto, G. Kawai, H. Asanuma, M. Ko-
miyama, Chem. Eur. J. 2006, 12, 777–784; e) H. Kashida, T. Fujii,
H. Asanuma, Org. Biomol. Chem. 2008, 6, 2892–2899; f) T. Fujii, H.
Kashida, H. Asanuma, Chem. Eur. J. 2009, 15, 10092–10102.
[11] a) 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; b) Y. Hara, T.
Fujii, H. Kashida, K. Sekiguchi, X. Liang, K. Niwa, T. Takase, Y.
Yoshida, H. Asanuma, Angew. Chem. 2010, 122, 5634–5638; Angew.
Chem. Int. Ed. 2010, 49, 5502–5506.
Acknowledgements
This work was partially supported by a JSPS research fellowship for
young scientists (to T.F.), a Grant-in-Aid for Scientific Research (A,
21241031), and a Grant-in-Aid for Young Scientists (B, 22750149) from
the Ministry of Education, Culture, Sports, Science and Technology,
Japan. Partial support by the SENTAN program, Japan Science and
Technology Agency (JST) is also acknowledged.
[12] a) M. K. Johansson, H. Fidder, D. Dick, R. M. Cook, J. Am. Chem.
Soc. 2002, 124, 6950–6956; b) M. K. Johansson, R. M. Cook, Chem.
Eur. J. 2003, 9, 3466–3471.
[13] Because the melting temperature of the NMR-MSa/NMR-Ob
duplex was determined to be 25.68C, NMR measurements were per-
formed at 58C (278 K), at which temperature a stable duplex is
formed. To monitor imino protons that are exchangeable with water
molecules, NMR spectra were recorded in H2O/D2O 9:1, with
a 3:9:19 WATERGATE pulse sequence for H2O suppression. When
these data were combined with the NOESY, DQF-COSY, and
TOCSY spectra, most of the signals of the duplex could be assigned.
[14] The imino-proton signals of the terminal G7 and G12 were slightly
broad because of the rapid exchange with water, whereas signals
from the other residues remained sharp.
[1] A. S. Davydov, Theory of Molecular Excitons, Plenum, New York,
1971, p.18.
[2] M. Kasha, Radiat. Res. 1963, 20, 55–70.
[3] J-Aggregates (Ed.: T. Kobayashi), World Scientific, Singapore, 1996,
pp. 1–40.
[4] a) M. van Burgel, D. A. Wiersma, K. Duppen, J. Chem. Phys. 1995,
102, 20–33; b) J-aggregates (Ed.: T. Kobayashi), World Scientific,
Singapore, 1996, p. 111–160; c) A. S. R. Koti, J. Taneja, N. Periasa-
my, Chem. Phys. Lett. 2003, 375, 171–176.
[5] a) K. Murata, S.-i. Kuroda, K. Saito, Thin Solid Films 1997, 295, 73–
76; b) F. D. Lewis, Y. Wu, L. Zhang, X. Zuo, R. T. Hayes, M. R. Wa-
sielewski, J. Am. Chem. Soc. 2004, 126, 8206–8215; c) F. D. Lewis,
L. Zhang, X. Liu, X. Zuo, D. M. Tiede, H. Long, G. C. Schatz, J.
Am. Chem. Soc. 2005, 127, 14445–14453; d) A. Yamaguchi, N. Ko-
metani, Y. Yonezawa, Thin Solid Films 2006, 513, 125–135; e) A. L.
Burin, M. E. Armbruster, M. Hariharan, F. D. Lewis, Proc. Natl.
Acad. Sci. USA 2009, 106, 989–994; f) V. Czikkely, G. Dreizler,
H. D. Fçrsterling, H. Kuhn, J. Sondermann, P. Tillmann, J. Z. Wie-
gand, Z. Naturforsch. 1969, 24a, 1821–1830.
[6] a) J. L. Seifert, R. E. Connor, S. A. Kushon, M. Wang, B. A. Armit-
age, J. Am. Chem. Soc. 1999, 121, 2987–2995; b) M. Wang, G. L.
Silva, B. A. Armitage, J. Am. Chem. Soc. 2000, 122, 9977–9986; c) S.
Bernacchi, Y. Mꢂly, Nucleic Acids Res. 2001, 29, 62e; d) R. A.
Garoff, E. A. Litzinger, R. E. Connor, I. Fishman, B. A. Armitage,
Langmuir 2002, 18, 6330–6337; e) S. Bernacchi, E. Piꢂmont, N.
Potier, A. v. Dorsselaer, Y. Mꢂly, Biophys. J. 2003, 84, 643–654;
f) M. Hariharan, Y. Zheng, H. Long, T. A. Zeidan, G. C. Schatz, J.
Vura-Weis, M. R. Wasielewski, X. Zuo, D. M. Tiede, F. D. Lewis, J.
Am. Chem. Soc. 2009, 131, 5920–5929; g) T. M. Wilson, T. A.
Zeidan, M. Hariharan, F. D. Lewis, M. R. Wasielewski, Angew.
Chem. 2010, 122, 2435–2438; Angew. Chem. Int. Ed. 2010, 49, 2385–
2388; h) A. L. Stadler, B. R. Renikuntla, D. Yaron, A. S. Fang, B. A.
Armitage, Langmuir 2011, 26, 1472–1479.
[15] The Tm of M5a/C, which was reported in the previous paper (see
Ref. [10b]), was 36.38C under the buffer conditions used. This large
destabilization was considered to be induced by the structural distor-
tion. Thus, introduction of the dye residue (Y) on d-threoninol into
the complementary strand significantly stabilized the duplexes.
[16] An increase in the number of quenchers induced hypsochromicity of
quencher band (see Figure S4 in the Supporting Information).
[17] For highly sensitive molecular beacons, see: a) O. Seitz, Angew.
Chem. 2000, 112, 3389–3392; Angew. Chem. Int. Ed. 2000, 39, 3249–
3252; b) S. Tyagi, S. A. E. Marras, F. R. Kramer, Nat. Biotechnol.
2000, 18, 1191–1196; c) B. Dubertret, M. Calame, A. J. Libchaber,
Nat. Biotechnol. 2001, 19, 365–370; d) H. Du, M. D. Disney, B. L.
Miller, T. D. Krauss, J. Am. Chem. Soc. 2003, 125, 4012–4013;
e) C. J. Yang, H. Lin, W. Tan, J. Am. Chem. Soc. 2005, 127, 12772–
12773; f) C. J. Yang, M. Pinto, K. Schanze, W. Tan, Angew. Chem.
2005, 117, 2628–2632; Angew. Chem. Int. Ed. 2005, 44, 2572–2576;
g) L. Wang, C. J. Yang, C. D. Medley, S. A. Benner, W. Tan, J. Am.
Chem. Soc. 2005, 127, 15664–15665; h) C. Crey-Desbiolles, D.-R.
Ahn, C. J. Leumann, Nucleic Acids Res. 2005, 33, e77; i) D. M. Kol-
pashchikov, J. Am. Chem. Soc. 2006, 128, 10625–10628; j) Y. Kim,
C. J. Yang, W. Tan, Nucleic Acids Res. 2007, 35, 7279–7287; k) T. N.
Grossmann, L. Rçglin, O. Seitz, Angew. Chem. 2007, 119, 5315–
5318; Angew. Chem. Int. Ed. 2007, 46, 5223–5225; l) E. Socher, L.
Bethge, A. Knoll, N. Jungnick, A. Herrmann, O. Seitz, Angew.
Chem. 2008, 120, 9697–9701; Angew. Chem. Int. Ed. 2008, 47, 9555–
9559; m) S. Nakayama, L. Yan, H. O. Sintim, J. Am. Chem. Soc.
2008, 130, 12560–12561; n) E. Socher, D. V. Jarikote, A. Knoll, L.
Rçglin, J. Burmeister, O. Seitz, Anal. Biochem. 2008, 375, 318–330;
o) S. Berndl, H.-A. Wagenknecht, Angew. Chem. 2009, 121, 2454–
2457; Angew. Chem. Int. Ed. 2009, 48, 2418–2421; p) S. Kummer, A.
Knoll, E. Socher, L. Bethge, A. Herrmann, O. Seitz, Angew. Chem.
2011, 123, 1972–1975; Angew. Chem. Int. Ed. 2011, 50, 1931–1934;
[7] a) T. J. Matray, E. T. Kool, J. Am. Chem. Soc. 1998, 120, 6191–6192;
b) J. Gao, C. Strꢀssler, D. Tahmassebi, E. T. Kool, J. Am. Chem. Soc.
2002, 124, 11590–11591; c) J. N. Wilson, Y. N. Teo, E. T. Kool, J.
Am. Chem. Soc. 2007, 129, 15426–15427; d) S. S. Tan, S. J. Kim,
E. T. Kool, J. Am. Chem. Soc. 2011, 133, 2664–2671.
[8] a) H. Bittermann, D. Siegemund, V. L. Malinovskii, R. Hꢀner, J.
Am. Chem. Soc. 2008, 130, 15285–15287; b) V. L. Malinovskii, F.
Samain, R. Hꢀner, Angew. Chem. 2007, 119, 4548–4551; Angew.
Chem. Int. Ed. 2007, 46, 4464–4467; c) A. L. Nussbaumer, D.
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!

H. Kashida, H. Asanuma et al.
q) C. Holzhauser, H.-A. Wagenknecht, Angew. Chem. 2011, 123,
7406–7410; Angew. Chem. Int. Ed. 2011, 50, 7268–7272.
[19] a) W. H. Nutting, R. A. Jewell, H. Rapoport, J. Org. Chem. 1970, 35,
505–508; b) M. A. Rahim, P. N. P. Rao, E. E. Knaus, J. Heterocycl.
Chem. 2002, 39, 1309–1314.
[20] X. Liang, T. Mochizuki, H. Asanuma, Small 2009, 5, 1761–1768.
[21] H. Nishioka, X. Liang, H. Kashida, H. Asanuma, Chem. Commun.
2007, 4354–4356.
[18] a) H. Dçhner, S. Stilgenbauer, K. Fischer, M. Schrçder, M. Bentz, P.
Lichter, Stem Cells 1995, 13, 76–82; b) C. Preudhomme, F. Rꢂvil-
lion, A. Merlat, L. Hornez, C. Roumier, N. Duflos-Grardel, J. P.
Jouet, A. Cosson, J. P. Peyrat, P. Fenaux, Leukemia 1999, 13, 957;
c) C. L. Sawyers, N. Engl. J. Med. 1999, 340, 1330–1340; d) P. Bolu-
fer, G. F. Sanz, E. Barragan, M. A. Sanz, J. Cervera, E. Lerma, L.
Senent, I. Moreno, M. D. Planelles, Haematologica 2000, 85, 1248–
1254.
[22] M. Liu, X.-a. Mao, C. Ye, H. Huang, J. K. Nicholson, J. C. Lindon, J.
Magn. Reson. 1998, 132, 125–129.
Received: April 20, 2012
Published online: && &&, 0000
&
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!

Asymmetric Heterodye Clustering in DNA Duplex
FULL PAPER
Asymmetric Dye Clusters
T. Fujii, Y. Hara, T. Osawa,
H. Kashida,* X. Liang, Y. Yoshida,
H. Asanuma* ................. &&&& — &&&&
Bulge-like Asymmetric Heterodye
Clustering in DNA Duplex Results in
Efficient Quenching of Background
Emission Based on the Maximized
Excitonic Interaction
For reduction of the background emis-
sion with the DNA probes, asymmetric
dye clusters were prepared by using d-
threoninol as a dye–cluster scaffold in
DNA. The cluster of single fluoro-
phore and two quenchers suppressed
the emission more efficiently than the
heterodimer due to the strong exciton
coupling between the dyes. This design
was used to prepare a highly-sensitive
in-stem molecular beacon (MB; see
figure).
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!