NBD-based green fluorescent ligands for typing of thymine-related SNPs by using an abasic site-containing probe DNA.

Abstract

The binding behavior of green fluorescent ligands, derivatives of 7-nitrobenzo-2-oxa-1,3-diazole (NBD), with DNA duplexes containing an abasic (AP) site is studied by thermal denaturation and fluorescence experiments. Among NBD derivatives, N(1)-(7-nitrobenzo[c][1,2,5]oxadiazol-4-yl)propane-1,3-diamine (NBD-NH(2)) is found to bind selectively to the thymine base opposite an AP site in a DNA duplex with a binding affinity of 1.52 x 10(6) M(-1). From molecular modeling studies, it is suggested that the NBD moiety binds to thymine at the AP site and a protonated amino group tethered to the NBD moiety interacts with the guanine base flanking the AP site. Green fluorescent NBD-NH(2) is successfully applied for simultaneous G>T genotyping of PCR amplification products in a single cuvette in combination with a blue fluorescent ligand, 2-amino-6,7-dimethyl-4-hydroxypteridine (diMe-pteridine).

Full text

Page 1

DOI: 10.1002/cbic.200900530
NBD-Based Green Fluorescent Ligands for Typing of
Thymine-Related SNPs by Using an Abasic Site-Containing
Probe DNA
Viruthachalam Thiagarajan ,[a, b, c]Arivazhagan Rajendran,[a, d]Hiroyuki Satake,[a, b]
Seiichi Nishizawa,[a, b]and Norio Teramae*[a, b]
Introduction
DNA-binding small molecules have attracted considerable at-
tention for the development of drugs to control gene expres-
sion[1]and they have also been used as important probes to
detect specific sequences and lesion sites in DNA strands.[2]
Most of the DNA-binding ligands are either groove binders or
intercalators,[3]and such ligands have been used as a reporter
molecules for the detection of single-nucleotide polymorph-
isms (SNPs) in combination with labeled or unlabeled probe
DNAs.[4]Because SNPs are responsible for the genetic variations
associated with susceptibility to various common diseases and
response to various drugs, a variety of methods have been de-
veloped to detect them, including high-density arrays, primer
extension methods, and the Invador assay.[4]Considering the
future usage of the phenotype of each individual in personal-
ized medicine and medical care,[5]there is still a great demand
for developing new typing methods that are high-throughput,
high in accuracy, low in cost, and easy to operate.
Among the variety of methods for SNP typing, we have re-
cently proposed a new noncovalent ligand-based fluorescence
assay for SNP typing in combination with DNA duplexes con-
taining an abasic or apurinic/apyrimidinic (AP) site.[6]Related to
the DNA-binding ligands and DNA lesions, Nakatani and co-
workers reported specific recognition of a single guanine
bulge by using 2-acrylamino-1,8-naphthyridines,[7]and they
have successfully demonstrated the selective detection of mis-
matched base pairs,[8]SNPs,[8a]and trinucleotide repeats,[9]
based on the binding of naphthyridine derivatives to the
bulge sites in the DNA duplex by using surface plasmon reso-
nance. In contrast, our method is based on the construction of
a chemically stable AP site in a DNA duplex; this allows small
synthetic ligands to bind to a target nucleotide, accompanied
by fluorescence signaling. As shown in Scheme 1, an AP site-
containing probe DNA is hybridized with a target DNA so as to
place the AP site toward a target nucleobase. In this way a hy-
drophobic microenvironment is provided for ligands to recog-
nize the target nucleotide through hydrogen bonding; this is
followed by p-stacking interactions between ligands and the
bases flanking the AP site. Formation of hydrogen bonds be-
tween nucleobases and ligands at the AP site was confirmed
by the appearance of imino protons in
and
charged guanidinium moiety of amiloride with the phosphate
backbone around the AP site.[6d]Thermodynamic analysis
based on isothermal titration calorimetry indicated that the
1H NMR spectra,[6h,10]
31P NMR measurements revealed the interaction of a
The binding behavior of green fluorescent ligands, derivatives
of 7-nitrobenzo-2-oxa-1,3-diazole (NBD), with DNA duplexes
containing an abasic (AP) site is studied by thermal denatura-
tion and fluorescence experiments. Among NBD derivatives,
N1-(7-nitrobenzo[c][1,2,5]oxadiazol-4-yl)propane-1,3-diamine
(NBD-NH2) is found to bind selectively to the thymine base op-
posite an AP site in a DNA duplex with a binding affinity of
1.52?106m?1. From molecular modeling studies, it is suggest-
ed that the NBD moiety binds to thymine at the AP site and a
protonated amino group tethered to the NBD moiety interacts
with the guanine base flanking the AP site. Green fluorescent
NBD-NH2is successfully applied for simultaneous G>T geno-
typing of PCR amplification products in a single cuvette in
combination with a blue fluorescent ligand, 2-amino-6,7-di-
methyl-4-hydroxypteridine (diMe-pteridine).
[a] V. Thiagarajan , A. Rajendran, H. Satake, S. Nishizawa, N. Teramae
Department of Chemistry, Graduate School of Science, Tohoku University
Aoba-ku, Sendai 980-8578 (Japan)
Fax: (+81)22-795-6552
E-mail: teramae@mail.tains.tohoku.ac.jp
[b] V. Thiagarajan , H. Satake, S. Nishizawa, N. Teramae
CREST (Japan) Science and Technology Agency (JST)
Aoba-ku, Sendai 980-8578 (Japan)
[c] V. Thiagarajan
Present address: CEA
Institut de Biologie et Technologies de Saclay (IBITECS) and CNRS
Gif-sur-Yvette, 91191 (France)
[d] A. Rajendran
Present address:
Frontier Institute for Biomolecular Engineering Research (FIBER)
Konan University
7-1-20 Minatojima-minamimachi, Chuo-Ku, Kobe 650-0047 (Japan)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cbic.200900530: Fluorescence and UV–visible
absorption responses of NBD-NH2for various target bases opposite the AP
site, energy-minimized structures of the complexes between NBD-NH2and
thymine, and the possible binding mode of NBD derivatives with T opposite
the AP site.
94
? 2010 Wiley-VCH Verlag GmbH&Co. KGaA, WeinheimChemBioChem 2010, 11, 94–100

Page 2

hydrophobicity of a ligand was responsible for the increase in
the binding affinity at the AP site.[11]
We have successfully discovered a series of fluorescent li-
gands that can recognize a target nucleotide with high affinity
and selectivity at an AP site in DNA duplexes; this series in-
cludes cytosine (C)-selective naphthyridine derivatives,[6a,11]thy-
mine (T)-selective amiloride[6d]and riboflavin,[10]guanine (G)-
selective pteridine derivatives,[6b,c]and adenine (A)-selective al-
loxazine.[6f]By using G- and A-selective ligands, G>A genotyp-
ing was demonstrated in the analysis of PCR amplification pro-
ducts,[6f]in which complexation-induced fluorescence quench-
ing of alloxazine (lem=453 nm) and 2-amino-6,7-dimethyl-4-
hydroxypteridine (diMe-pteridine; lem=436.5 nm) was utilized
to detect the single base mutations. However, both of the
above ligands show blue fluorescence, and the fluorescence
detection for genotyping must be carried out by using a differ-
ent cuvette for each ligand. To achieve the simultaneous de-
tection of the multiple target bases in one cuvette, fluorescent
ligands with different colors are necessary. Accordingly, green
fluorescent 7-nitrobenzo-2-oxa-1,3-diazole (NBD) derivatives
were chosen as new ligands for the detection of SNPs and
their binding behavior was examined for nucleobases opposite
an AP site in DNA duplexes. The green fluorescent probe NBD-
NH2was found to bind selectively to the thymine target base
over other nucleobases at the AP site with a binding affinity
more than 106m?1, and the G>T genotyping of PCR amplifica-
tion products was successfully demonstrated in a single cuv-
ette by using NBD-NH2and diMe-pteridine as T- and G-selec-
tive ligands, respectively.
Results and Discussion
Thermal denaturation studies
First, we performed thermal denaturation experi-
ments monitored by UV absorption at 260 nm as a
function of temperature to examine the binding be-
havior of NBD derivatives with the AP site-contain-
ing DNA duplexes that contained various target nu-
cleobases. Melting temperature (Tm) values were ob-
tained by differentiating the melting temperature
profiles by using eleven-mer DNA duplexes contain-
ing an AP site (5’-TCC AGX GCA AC-3’/3’-AGG TCN
CGT TG-5’, X=AP site (a propylene residue, Spacer
C3), N=A, G, C, or T), and the results are summarized in
Table 1. Tm(?)and Tm(+ +)denote the melting temperatures of
DNA duplexes in the absence and presence of NBD derivatives,
respectively. As can be seen in Table 1, in the absence of li-
gands, Tmvalues are low for pyrimidine bases (T and C) oppo-
site the AP site in contrast to purine bases (A and G). This dif-
ference in Tmvalues can be ascribed to the intra- and extraheli-
cal conformations of nucleobases opposite the AP site.[12]
Upon addition of ligands, a remarkable increase in the Tm
value (DTm) is observed for all ligands in cases in which T is the
base opposite the AP site; this indicates the stronger binding
of NBD derivatives with the DNA duplex containing a T target
base. A moderate increase in Tmis observed for C, and the DTm
values for A and G are quite small. The DTmvalues are in the
order of T>C>A, G for all of NBD ligands. Among three NBD
ligands, NBD-NH2gives the largest increase in Tmfor T, and
NBD-NMe2and NBD-C give moderate increases in Tm; this sug-
gests that the interaction between the tethered NH2group of
NBD-NH2and a duplex containing an AP site is crucial for the
strongest binding of NBD-NH2. If a ligand binds to a fully
matched DNA duplex through nonspecific intercalation or elec-
trostatic interactions, then the Tmvalue of the DNA duplex (no
AP site) will increase in the presence of the ligand.[1b]We could
not observe any significant change in Tmof the fully matched
Scheme 1. Schematic illustration of the ligand-based fluorescence detection of SNPs in
combination with an AP site-containing DNA duplex.
Table 1. Melting temperatures of AP-site-containing DNA duplexes in the
absence and presence of NBD derivatives.
Tm(?)[8C]
without ligand
Tm(+ +)(DTm) [ 8C]
NBD-NMe2
NNBD-NH2
NBD-C
T
C
A
G
27.4?0.8
29.6?0.9
33.5?0.8
33.4?0.7
DNA duplex is 5’-TCCAGXGCAAC-3’/5’-GTTGCNCTGGA-3’, where X is an
AP site (Spacer C3) and N is T, C, A, or G. Tm(?)and Tm(+ +)denote the melt-
ing temperature in the absence and presence of NBD derivatives, respec-
tively. Errors are the standard deviation obtained by four independent re-
peated measurements. [DNA duplex]=20 mm, [ligand]=100 mm in 2%
DMSO, [NaCl]=100 mm, [EDTA]=1 mm, [sodium cacodylate]=10 mm,
pH 7.0.
32.0?1.0 (+ +4.6)
32.0?1.0 (+ +2.4)
34.3?0.1 (+ +0.8)
34.2?0.5 (+ +0.8)
31.0?1.0 (+ +3.6)
32.0?0.8 (+ +2.4)
34.3?0.3 (+ +0.8)
32.8?0.4 (?0.6)
30.5?0.8 (+ +3.1)
31.6?0.8 (+ +2.0)
34.3?0.6 (+ +0.8)
33.9?0.3 (+ +0.5)
ChemBioChem 2010, 11, 94–100? 2010 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim
www.chembiochem.org
95
NBD-Based Fluorescent Ligands for SNP Typing

Page 3

DNA duplex in the presence of NBD-NH2 (fully matched
duplex:5’-TCCAGGGCAAC-3’/5’-GTTGCCCTGGA-3’;
59.2?0.38C; Tm (+ +)=59.6?0.58C; [DNA duplex]=20 mm, [NBD-
NH2]=100 mm). Accordingly, it can be regarded that the inter-
calation and electrostatic interactions do not take place in the
present case.
Tm(?)=
Fluorescence experiments
The binding behavior of NBD derivatives with AP site-contain-
ing DNA duplexes was further examined by UV–visible absorp-
tion and fluorescence measurements by using 23-mer DNA du-
plexes (5’-GT GTGCGTTG CNC TGGACGCAGA-3’/5’-TCTGCG
TCCA GXG CAACGCACAC-3’; X=spacer C3; N=G, C, A or T).
As shown in Figure 1, the fluorescence intensity of NBD li-
gands decreases upon addition of DNA duplexes containing an
AP site without an appreciable shift in the emission maximum,
whereas no significant changes in the fluorescence intensity
are recognized upon addition of a DNA duplex which has no
AP site (fully matched DNA duplex); this indicates that the
NBD moiety interacts solely with a target base opposite an AP
site. The degree of fluorescence quenching of NBD derivatives
is most remarkable in cases in which thymine is the target
base and the strongest quenching is observed for NBD-NH2
among three NBD ligands. To evaluate the binding affinity of
NBD ligands with DNA duplexes containing thymine opposite
the AP site, each ligand was titrated with the DNA duplex con-
taining an AP site, and the resulting fluorescence titration
curves (Figure 1D) were analyzed by a non-linear regression
based on a 1:1 binding isotherm model.[6a]The strongest affini-
ty was obtained for NBD-NH2((1.52?0.07)?106m?1), and mod-
erate and lower affinities were obtained for NBD-NMe2((0.61?
0.02)?106m?1) and NBD-C ((0.14?0.01)?106m?1); this order in
the affinity is in accordance with the results of thermal denatu-
ration experiments (Table 1) and the degree of fluorescence
quenching (Figure 1A–C). By using NBD-NH2as a ligand, the
differences in binding affinities for nucleobases opposite the
AP site were examined by fluorescence titration experiments
(Figure S2 in the Supporting Information). It was found that
the binding affinity of NBD-NH2for T is about fivefold higher
Figure 1. Fluorescence spectra of NBD ligands A) NBD-NH2, B) NBD-NMe2, C) NBD-C in the presence and absence of 23-mer DNA duplexes. i) and ii) in each
figure denote, respectively, the fluorescence spectra of ligands without and with fully matched DNA. iii)–vi) denote the fluorescence spectra of ligands with
AP site-containing DNA duplexes with G, A, C, and T target bases, respectively. D) Fluorescence titration curves for NBD ligands by addition of DNA duplexes
containing Topposite the AP site. F and F0denote the fluorescence intensity with and without DNA duplexes. [ligand]=1.0 mm , [NaCl]=100 mm,
[EDTA]=1 mm, [sodium cacodilate]=10 mm, pH 7.0, temperature 58C. DNA duplex=5’-GTGTGCGTTG CNC TGGACGCAGA-3’/5’-TCTGCGTCCA GXG CAACGCA-
CAC-3’ ; X=Spacer C3, N=G, A, C, or T. In fully matched DNA duplex, N=Tand X=A. For A) to C) [DNA duplex]=1.0 mm. For D) [DNA duplex]=1.0 to
6.0 mm, lex=500 nm (isosbestic point) for NBD-NH2and NBD-NMe2, lex=504 nm (isosbestic point) for NBD-C. Quartz cell (2?10 mm).
96
www.chembiochem.org
? 2010 Wiley-VCH Verlag GmbH&Co. KGaA, WeinheimChemBioChem 2010, 11, 94–100
N. Teramae et al.

Page 4

than C (0.32(?0.03)?106m?1) and more than tenfold higher
than A (0.15(?0.02)?106m?1) and G (0.08(?0.03)?106m?1).
Considering the difference in the binding affinities of NBD-NH2
with nucleobases, NBD-NH2could be applicable to T>A and
T>G genotyping. The order in selectivity (T>C>A>G) is also
well reflected in the spectral changes of UV–visible absorption
of NBD-NH2with and without DNA duplexes that have differ-
ent nucleobases opposite an AP site (Figure S3). Upon addition
of AP site-containing DNA duplexes, absorption intensity de-
creases and the absorption maximum shows a red-shift, which
is due to the p-stacking interaction between NBD-NH2and
bases flanking the AP site, with a clear isosbestic point at
500 nm; this wavelength was selected as the excitation wave-
length for fluorescence measurements. Based on the high af-
finity and selectivity of NBD-NH2for the binding with T oppo-
site an AP site in a DNA duplex, NBD-NH2can be expected to
give sufficient fluorescence response for reliable discrimination
of the T/G and T/A mutations, and the ligand can be applica-
ble for the genotyping of PCR amplification products.
Ionic strength dependency of binding con-
stants
Further, the binding behavior of NBD derivatives with DNA du-
plexes was studied by the ionic strength dependency of
ligand–DNA binding affinities, by which thermodynamic pa-
rameters were obtained for the polyelectrolyte and nonelectro-
static components of overall binding free energy of ligand-
DNA interactions.[13]The binding affinity (K11) was obtained
from fluorescence titration experiments and its value was plot-
ted against the salt concentrations (Figure 2). As shown in
Figure 2, the binding constants for NBD-NH2and NBD-NMe2
decrease remarkably as the salt concentration increases, while
the decrease for NBD-C is quite less. According to the polyelec-
trolyte theory,[13]an apparent charge on the ligand upon bind-
ing to the DNA duplex can be calculated from Equation (1):
SK ¼dlogK11
dlog½NaI?¼ ?ZY
ð1Þ
in which SK is the slope of the linear least-squares analysis
of logK11vs. log[NaI] plot, Z is the number of counterions re-
leased per ligand-binding event, which is equivalent to the ap-
parent charge of the ligand, and Y is a constant equal to the
fraction of counterions associated per phosphate (0.88 for B-
type DNA[14]). The apparent charges on NBD derivatives upon
binding to DNA with a T target base were calculated as + +1.67,
+ +1.28, and + +0.04 for NBD-NH2, NBD-NMe2, and NBD-C, respec-
tively. The contribution of electrostatic interactions towards
the total free energy was then calculated from Equation (2):[13]
DGpe¼ ?ðSKÞRT ln½NaI?
ð2Þ
in which R is the gas constant and T is the absolute temper-
ature. The polyelectrolyte (DGpe) and nonelectrostatic (DGt)
contributions were examined from the overall binding free
energy change (DGobs=-RT ln K11) by using Equation (3):[13]
DGobs¼ DGpeþ DGt
ð3Þ
From the plots shown in Figure 2, polyelectrolyte and non-
polyelectrolyte contributions to DGobswere calculated and the
dissected contributions are summarized in Table 2. It is re-
vealed from the thermodynamic parameters listed in Table 2
that the nonpolyelectrolyte contribution is fundamental for the
complexation of NBD derivatives with DNA duplex. It is also
recognized that the polyelectrolyte contribution is large for
protonated ligands, NBD-NH2and NMe2(NBD-NH2: ?1.79 kcal
mol?1, NBD-NMe2: ?1.37 kcalmol?1), compared to a neutral
ligand NBD-C (?0.04 kcalmol?1).
Molecular modeling: Molecular modeling was carried out
with MacroModel Ver.9.0, considering the 1:1 complexation of
NBD-NH2with a DNA duplex containing an AP site from a qual-
itative point of view. Although both the thymine base (T14)
Figure 2. Salt dependence of binding constants for the interaction between
NBD derivatives and DNA duplex containing thymine opposite an AP site.
DNA duplex: 5’-GTGTGCGTTG CTC TGGACGCAGA-3’/5’-TCTGCGTCCA GXG
CAACG CACAC-3’; X=spacer C3. [NaCl]=60–310 mm, 58C.
Table 2. Thermodynamic parameters for the binding of NBD derivatives
to thymine in 23-mer DNA duplex containing an AP site.
NBD
probes [m?1][a]
K11?106
DGobs
[kcalmol?1]
Z
DGpe
[kcalmol?1]
DGt
[kcalmol?1]
NBD-
NH2
NBD-
NMe2
NBD-C 0.14?0.01 ?6.54?0.05 + +0.04?0.06 ?0.04?0.06 ?6.50?0.08
[a] K11[m?1] determined from fluorescence titration experiments by using
a 1:1 binding isotherm model at 58C. ([sodium cacodylate buffer]=
10 mm; [NaCl]=100 mm; [EDTA]=1 mm, pH 7.0). DGobsis the observed
free energy change; Z is the apparent charge on the ligand upon com-
plexation; DGpeis the polyelectrolyte contribution to the observed free
energy change; DGtis the nonpolyelectrolyte contribution to the ob-
served free energy change; DNA duplex: 5’-GTGTGCGTTG CTC TGGACG-
CAGA-3’/5’-TCTG CGTCCA GXG CAACGCACAC-3’; X=spacer C3.). Errors
denote the standard error.
1.52?0.07 ?7.86?0.02 + +1.67?0.03 ?1.79?0.03 ?6.07?0.04
0.61?0.02 ?7.36?0.02 + +1.28?0.04 ?1.37?0.04 ?5.59?0.04
ChemBioChem 2010, 11, 94–100? 2010 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim
www.chembiochem.org
97
NBD-Based Fluorescent Ligands for SNP Typing

Page 5

opposite the AP site and the AP site adopted an extrahelical
arrangement in the duplex containing an AP site (5’-C1G2T3G4X
G6T7G8C9-3’/3’-G18C11A12C13T14G15T16G17C18-5’, X=AP site)[12]due
to the stacking of G4with G6, intercalation of an aromatic
group conjugated to the DNA strand at the AP site was report-
ed to form a single intrahelical form of a B-type DNA duplex[15]
by using the same sequence reported previously.[12]By using a
DNA duplex containing an AP site opposite thymine, an acri-
dine derivative was reported to bind to the AP site by thread-
ing intercalation, and the acridine derivative stacked with the
bases flanking the AP site keeping the intrahelical structure of
the thymine base opposite the AP site.[16]In our case, CD spec-
tra of AP site-containing DNA duplexes with and without NBD-
NH2show a couplet signal with a positive peak at around
280 nm and a negative peak at around 247 nm (Figure S4),
similar to the CD spectrum of B-type DNA. Accordingly, a B-
type conformation was adopted in the molecular modeling.
Figure 3 shows the energy minimized model of the 1:1 com-
plex of NBD-NH2with the T target base opposite the AP site in
a DNA duplex. The molecular modeling indicates that the for-
mation of two hydrogen bonds is evident between NBD
moiety and T, and that the tethered cation is oriented toward
the major groove of the DNA duplex forming hydrogen bonds
between the protonated amino group and the O6 and N7
atoms of guanine flanking the AP site. This structural feature is
similar to the earlier studies[17]in which the tethered ammoni-
um ion is reported to direct toward the major groove and in-
teract with electronegative atoms (N7 and O6) of guanine. This
structural feature is also responsible for the strongest binding
of NBD-NH2to the T target base opposite the AP site in a DNA
duplex among three NBD derivatives. The nitrogen atom of an
N(CH3)2group of NBD-NMe2can be protonated and charge–
charge interaction can take place, whereas formation of one of
hydrogen bonds between the protonated amino group and an
O6 atom of guanine flanking the AP site is prevented as com-
pared to NBD-NH2. As for NBD-C, charge–charge interaction
cannot be expected and only the NBD moiety contributes to
the binding to the target T base. Thus the substituents govern
the binding affinity of NBD derivatives with DNA duplexes con-
taining an AP site (Figures S5 and S6).
Two-colored SNPs typing: Finally, NBD-NH2was applied to
the detection of the G>T transversion present in codon 12 of
the K-ras gene.[18]Multicolor fluorescence is widely used in bio-
sciences for simultaneous or sequential detection of multiple
targets. We have recently demonstrated the G>A genotyping
of PCR products of 107-mer DNAs (K-ras gene, codon 12) by
using diMe-pteridine and alloxazine as G- and A-selective li-
gands, respectively.[6f]Because the emission wavelengths of
diMe-pteridine and alloxazine are similar (436.5 and 453 nm,
respectively), a different cuvette containing a PCR amplification
product must be used for each ligand to detect the G/A muta-
tion by fluorescence. A mixture of fluorescent ligands that
have different wavelengths of excitation (lex) and emission
(lem) would enable the multicolored typing of gene SNP sites
simultaneously (Figure S7). First, 107-mer sense strands of the
K-ras gene were amplified by asymmetric PCR as described in
Experimental section. Then, an AP site-containing probe DNA
is added into a single sample cuvette containing a PCR-ampli-
fied product of K-ras gene to form DNA duplexes. By the addi-
tion of a mixture of green fluorescent T-selective NBD-NH2(lex/
lem=474/550 nm) and blue fluorescent G-selective diMe-pteri-
dine (lex/lem=372/440 nm) into the cuvette, the type of alleles
at a specific SNP site was determined by the fluorescence in-
tensities of the T- and G-selective ligands at 550 nm and
440 nm, respectively.
Next, 24 samples of PCR amplification products (six G/G-ho-
mozygous, six T/T-homozygous, six G/T heterozygous, and six
control solutions without PCR amplification product) were ana-
lyzed with a 20-mer probe DNA strand containing an AP site
by using green fluorescent NBD-NH2 and blue fluorescent
diMe-pteridine ligands. Figure 4 shows the fluorescence inten-
sity of the three genotypes in the K-ras gene, codon 12, frag-
ment 107. Although the change in fluorescence intensity of
NBD-NH2is not large compared to that of diMe-pteridine, SNP
genotyping of the PCR amplification product samples, G/G, G/
T, and T/T, can be clearly discriminated as three different clus-
ters based on fluorescence responses of NBD-NH2 (lem=
550 nm) and diMe-pteridine (lem=440 nm). The control experi-
ments measured without PCR product are located in the upper
right corner and the plots are clearly separated from the other
three clusters. These results indicate that the use of two differ-
ent colored ligands in our assay allows the detection of both
SNP alleles in a single cuvette, and three genotypes can be
clearly distinguished from one another by measuring the in-
tensity of blue and green fluorescence emission. The analysis
requires no time-consuming steps such as purification of PCR
amplification products and careful control of temperature, and
the result could be readily obtained after PCR.
Figure 3. A) Possible binding mode of NBD-NH2with Topposite an AP site
in a DNA duplex. B) Energy minimized structure obtained by MacroModel
Ver. 9.0., for the complex between NBD-NH2and Topposite an AP site in a
DNA duplex (5’-GTGTGCGTTG CTC TGGAC GCAGA-3’/5’-TCTGCGTCCA GXG
CAACGCACAC-3’ ; X=spacer C3). The NBD-NH2and the thymine base oppo-
site the AP site are colored green and pink, respectively.
98
www.chembiochem.org
? 2010 Wiley-VCH Verlag GmbH&Co. KGaA, WeinheimChemBioChem 2010, 11, 94–100
N. Teramae et al.

Page 6

Conclusions
In summary, we found that a green fluorescent ligand, NBD-
NH2, showed a highly selective and strong binding affinity for
thymine opposite the AP site in DNA duplexes. The NBD-NH2
probe was successfully demonstrated to be applicable for the
simultaneous G>T genotyping of PCR amplification products
in combination with a blue fluorescent ligand, diMe-pteridine.
In the present genotyping, PCR amplification products could
be readily analyzed by measuring fluorescence of added li-
gands without any pretreatment procedure.
Experimental Section
Materials: NBD derivatives were synthesized according to the liter-
ature.[19]All of the oligodeoxynucleotides used in the present study
were custom synthesized and HPLC purified (>97%) by Nihon
Gene Research Laboratories Inc. (Sendai, Japan). For the synthesis
of AP site containing DNAs, a propylene residue (Spacer phosphor-
amidite C3, Spacer C3) was utilized. The concentration of DNA was
determined from the molar extinction coefficient at 260 nm.[20]
Water was deionized (?18.0 MW cm specific resistance) by an Elix
5 UV Water Purification System and a Milli-Q Synthesis A10 system
(Millipore Corp., Bedford, MA). The other reagents were commer-
cially available analytical grade and were used without further pu-
rification.
Melting temperature (Tm) measurements: All measurements were
performed in sodium cacodylate buffer solutions (10 mm, pH 7.0)
containing NaCl (100 mm) and EDTA (1.0 mm). Concentrations of
NBD derivatives and DNA duplexes were 100 and 20 mm, respec-
tively. Before the Tmmeasurements, the sample solutions were an-
nealed as follows: heated at 758C for 10 min, and then gradually
cooled down to 58C (38Cmin?1). Absorbance of DNAs was then
measured at 260 nm as a function of temperature by using an UV-
visible spectrophotometer model 2450 (Shimadzu Corp., Kyoto,
Japan) equipped with a thermoelectrically temperature-controlled
micro multi-cell holder (eight cells; optical path length=1 mm).
The temperature ranged from 2–928C with a heating rate of
1.58Cmin?1. The resulting absorbance versus temperature curves
were differentiated to determine Tmvalues. Due to the poor solu-
bility of NBD-C and NBD-NMe2in aqueous solution, all the melting
temperature measurements (including NBD-NH2) were carried out
by using DMSO solutions (2%).
Fluorescence measurements: Fluorescence spectra of NBD deriva-
tives were measured at 58C with a JASCO FP-6500 spectrofluoro-
photometer (Japan Spectroscopic Co. Ltd., Tokyo, Japan) equipped
with a thermoelectrically temperature-controlled cell holder. All
emission spectra were corrected for the instrumental spectral re-
sponse. Sample solutions were buffered to pH 7.0 with sodium ca-
codylate (10 mm) containing NaCl (100 mm) and EDTA (1.0 mm). As
described in the Tmmeasurements, the sample solutions were an-
nealed before the fluorescence measurements. Due to the poor
solubility of NBD-C and NBD-NMe2in aqueous solution, all the fluo-
rescence studies for these two ligands were carried out by first dis-
solved in DMSO and further diluted by using an aqueous solution.
The percentage of DMSO in the final assay volume is 0.005%. Bind-
ing constants (Kobs) of NBD-derivatives for the complexation with
nucleobases in DNA duplexes containing an AP site were calculat-
ed by analyzing fluorescence titration curves with nonlinear least-
squares regression based on a 1:1 binding isotherm.[6a]
PCR procedures and Multicolor SNPs typing: A 107-mer K-ras
gene[16](sense strand containing a codon 12 site) was amplified by
asymmetric PCR (107-mer DNA: 5’-GACTG AATAT AAACT TGTGG
TAGTT GGAGCTG (G/T) TGGCG TAGGC AAGAG TGCCT TGACG
ATACA GCTAA TTCAG AATCA TTTTG TGGAC GAATA TGATC CAACA
ATAG-3’; 0.5 ng). Eicosomer forward primer (5’-GACTGAATATAAAC
TTGTGG-3’ (300 pmol) and 20-mer reverse primer (5’-CTATT GTTGG
ATCAT ATTCG-3’; 20 pmol) were added to 100 mL of the reaction
mixture containing dNTPs (2.5 mm each), PCR buffer (10?, 10 mL;
TaKaRa), and Ex-Taq (2.5 U; TaKaRa). The thermal cycling program
consisted of initial incubation at 948C for 5 min followed by 40
cycles of denaturation at 948C for 30 s, annealing at 528C for 30 s,
and extension at 728C for 30 s, and then the sample was kept at
728C for 7 min. After thermal cycling and subsequent cooling to
48C, aliquots (40 mL) of the amplified DNA mixtures were buffered
to pH 7.0 with sodium cacodylate (100 mm) containing EDTA
(1.6 mm) and subsequently hybridized with AP site-containing 20-
mer probe DNA (5.0 mm; 5’-CCTAC GCCAX CAGCT CCAAC-3’; X=
AP site). After hybridization, diMe-pteridine (0.1 mm) and NBD-NH2
(0.1 mm) were added to the PCR mixture and fluorescence intensi-
ties were measured at both green- and blue-emission wave-
lengths.
Molecular modeling: Models of the 23-mer oligodeoxynucleotides
and the NBD ligands were constructed by using Maestro 7.0. In
order to construct the AP site in a 23-mer DNA duplex containing
a thymine target base, the complementary adenosine unit was re-
moved and a propylene residue was inserted between two phos-
phate moieties adjacent to the AP site. Molecular modeling was
carried out with MacroModel Ver.9.0 (Schroinger, LLC, Portland,
OR). After inserting ligands manually into the AP site, energy mini-
mization was performed with Amber* force field and GB/SA contin-
uum solvation model for water (with constant dielectric treatment
for the electrostatic part) together with the default cut-off criteri-
ons; the gradient convergence threshold was set to be 0.005.
Figure 4. Scatter plots for G>T genotyping of 107-mer PCR products (K-ras
gene, codon 12, sense strand) based on fluorescence responses of G-selec-
tive diMe-pteridine and T-selective NBD-NH2. PCR product: 5’-GACTG AATAT
AAACT TGTGG TAGTT GGAGC TGNTG GCGTA GGCAA GAGTG CCTTG ACGAT
ACAGC TAATT CAGAA TCATT TTGTG GACGA ATAT GATCCA ACAATAG-3’
(wild-type, N=G; mutant type, N=T); probe DNA: 5’-CCTACGCCAXCAGCTC-
CAAC-3’ (X=AP site, Spacer-C3). Temperature, 58C. Excitation and emission
slit widths are 10 nm each. NBD-NH2(lex=474 nm; lem=550 nm), diMe-pter-
idine (lex=372 nm; lem=440 nm). Standard deviations are presented as el-
lipse.
ChemBioChem 2010, 11, 94–100? 2010 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim
www.chembiochem.org
99
NBD-Based Fluorescent Ligands for SNP Typing

Page 7

Acknowledgements
This work was partially supported by Grants-in-Aid for Scientific
Research (A), No. 17205009, and Scientific Research (B), No.
18350039, from the Ministry of Education, Culture, Sports, Science
and Technology, Japan. V.T. would like to thank the JST for a fel-
lowship.
Keywords: abasic site · DNA recognition · DNA · fluorescence ·
single nucleotide polymorphism
[1] See reviews and references herein: a) J. B. Chaires, Curr. Opin. Struct.
Biol. 1998, 8, 314–320; b) R. Martinez, L. Chac?n-Garc?a, Curr. Med.
Chem. 2005, 12, 127–151; c) P. Palchaudhuri, P. J. Hergenrother, Curr.
Opin. Biotechnol. 2007, 18, 497–503.
[2] See reviews and references herein: a) P. B. Dervan, Bioorg. Med. Chem.
2001, 9, 2215–2235; b) S. Neidle, Nat. Prod. Rep. 2001, 18, 291–309;
c) P. B. Dervan, B. S. Edelson, Curr. Opin. Struct. Biol. 2003, 13, 284–299.
[3] See reviews and references herein: a) D. E. Graves, L. M. Velea, Curr. Org.
Chem. 2000, 4, 915–929; b) S. M. Nelson, L. R. Ferguson, W. A. Denny,
Mutat. Res. 2007, 623, 24–40.
[4] See reviews and references herein: a) A.-C. Syv?nen, Nat. Rev. Genet.
2001, 2, 930–942; b) K. Nakatani, ChemBioChem 2004, 5, 1623–1633;
c) R. T. Ranasinghe, T. Brown, Chem. Commun. 2005, 5487–5502.
[5] a) F. S. Collins, E. D. Green, A. E. Guttmacher, M. S. Guyer, Nature 2003,
422, 835–847; b) W. Sadee, Z. Dai, Hum. Mol. Genet. 2005, 14, R207-
R214; c) M. Hiratsuka, T. Sasaki, M. Mizugaki, Clin. Chim. Acta 2006, 363,
177–186; d) G. W. Duff, Am. J. Clin. Nutr. 2006, 83, 431S-435S.
[6] a) K. Yoshimoto, S. Nishizawa, M. Minagawa, N. Teramae, J. Am. Chem.
Soc. 2003, 125, 8982–8993; b) K. Yoshimoto, C.-Y. Xu, S. Nishizawa, T.
Haga, H. Satake, N. Teramae, Chem. Commun. 2003, 2960–2961; c) Q.
Dai, C.-Y. Xu, Y. Sato, K. Yoshimoto, S. Nishizawa, N. Teramae, Anal. Sci.
2006, 22, 201–203; d) C. Zhao, Q. Dai, T. Seino, Y.-Y. Cui, S. Nishizawa, N.
Teramae, Chem. Commun. 2006, 1185–1187; e) B. Rajendar, Y. Sato, S.
Nishizawa, N. Teramae, Bioorg. Med. Chem. Lett. 2007, 17, 3682–3685;
f) B. Rajendar, S. Nishizawa, N. Teramae, Org. Biomol. Chem. 2008, 6,
670–673; g) Z. Ye, B. Rajendar, D. Qing, S. Nishizawa, N. Teramae, Chem.
Commun. 2008, 6588–6690; h) N. B. Sankaran, Y. Sato, F. Sato, B. Rajen-
dar, K. Morita, T. Seino, S. Nishizawa, N. Teramae, J. Phys. Chem. B 2009,
113, 1522–1529; i) B. Rajendar, A. Rajendran, Y. Sato, S. Nishizawa, N.
Teramae, Bioorg. Med. Chem. 2009, 17, 351–359.
[7] K. Nakatani, S. Sando, I. Saito, J. Am. Chem. Soc. 2000, 122, 2172–2177.
[8] a) K. Nakatani, S. Sando, I. Saito, Nat. Biotechnol. 2001, 19, 51–55; b) K.
Nakatani, A. Kobori, H. Kumasawa, Y. Goto, I. Saito, Bioorg. Med. Chem.
2004, 12, 3117–3123; c) S. Hagihara, H. Kumasawa, Y. Goto, G. Hayashi,
A. Kobori, I. Saito, K. Nakatani, Nucleic Acids Res. 2004, 32, 278–286;
d) A. Kobori, S. Horie, H. Suda, I. Saito, K. Nakatani, J. Am. Chem. Soc.
2004, 126, 557–562.
[9] K. Nakatani, S. Hagihara, Y. Goto, A. Kobori, M. Hagihara, G. Hayashi, M.
Kyo, M. Nomura, M. Mishima, C. Kojima, Nat. Chem. Biol. 2005, 1, 39–
43.
[10] N. B. Sankaran, S. Nishizawa, T. Seino, K. Yoshimoto, N. Teramae, Angew.
Chem. 2006, 118, 1593–1598; Angew. Chem. Int. Ed. 2006, 45, 1563–
1568.
[11] Y. Sato, S. Nishizawa, K. Yoshimoto, T. Seino, T. Ichihashi, K. Morita, N.
Teramae, Nucleic Acids Res. 2009, 37, 1411–1422.
[12] a) P. Cuniasse, G. V. Fazakerly, W. Guschlbauer, B. Kaplan, L. C. Sowers, J.
Mol. Biol. 1990, 213, 303–314; b) J. T. Stivers, Nucleic Acids Res. 1998,
26, 3837–3844.
[13] M. T. Record, C. F. Anderson, T. M. Lohman, Q. Rev. Biophys. 1978, 11,
103–178.
[14] M. T. Record, R. Spolar in The Biology of Nonspecific DNA–Protein Interac-
tions (Ed.: A. Revzin), CRC, Boca Raton, 1990, pp. 33–69.
[15] M. P. Singh, G. G. Hill, D. Peoc’h, B. Rayner, J.-L. Imbach, J. W. Lown, Bio-
chemistry 1994, 33, 10271–10285.
[16] A. Martelli, M. Jourdan, J.-F. Constant, M. Demeunynck, P. Dumy, Bioorg.
Med. Chem. Lett. 2006, 16, 154–157.
[17] a) A. K. Todd, A. Adams, J. H. Thorpe, W. A. Denny, L. P. G. Wakelin, C. J.
Cardin, J. Med. Chem. 1999, 42, 536–540; b) B. Gold, Biopolymers 2002,
65, 173–179.
[18] J. L. Bos, Mutat. Res. 1988, 195, 255–271.
[19] a) S. Fery-Forgues, J.-P. Fayet, A. Lopez, J. Photochem. Photobiol. A 1993,
70, 229–243; b) A. Cotte, B. Bader, J. Kuhlmann, H. Waldmann, Chem.
Eur. J. 1999, 5, 922–936; c) M. Onoda, S. Uchiyama, T. Santa, K. Imai, Lu-
minescence 2002, 17, 11–14.
[20] J. D. Puglisi, I. Tinoco, Methods Enzymol. 1989, 180, 304–325.
Received: August 24, 2009
Published online on November 30, 2009
100
www.chembiochem.org
? 2010 Wiley-VCH Verlag GmbH&Co. KGaA, WeinheimChemBioChem 2010, 11, 94–100
N. Teramae et al.