Large flanking sequence effects in single nucleotide mismatch detection using fluorescent nucleoside ?f

Article abstract of DOI:10.1016/j.bmc.2010.06.060

The first systematic study of flanking sequence effects on mismatch detection by a fluorescent nucleotide is described, using fluoroside ^f. Although a high degree of variance was observed in fluorescence intensity of mismatched duplexes between different flanking sequences, ^f was able to distinguish a mismatch from the fully base-paired duplex in 13 out of 16 sequences, and even identify each mismatch in 10 of those flanking sequences. For the flanking sequences where fluoroside ^f did not unambiguously determine its base-pairing partner, the experimental conditions were varied in an attempt to facilitate mismatch identification. No beneficial effect on the relative fluorescence intensities was achieved by changing the temperature, adding organic co-solvents or potassium iodide. In contrast, mercuric ions selectively quenched the fluorescence intensity of the ^f·T mismatch, effectively resolving the overlap of all emission spectra and thereby facilitating identification of all base-pairing partners in any flanking sequence by ^f. This is the first time mercuric ions have been used to selectively quench the fluorescence of a single mismatch. A noticeable characteristic of ^f is that, unlike most fluorosides, the fluorescence intensity of ^f was not quenched to a discernable degree by a flanking G-C pair.

Full text of DOI:10.1016/j.bmc.2010.06.060

Bioorganic & Medicinal Chemistry 18 (2010) 6121–6126
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Large flanking sequence effects in single nucleotide mismatch detection
using fluorescent nucleoside Çf
*
Haraldur Gardarsson, Snorri Th. Sigurdsson
University of Iceland, Science Institute, Dunhagi 3, 107 Reykjavik, Iceland
a r t i c l e i n f o
a b s t r a c t
Article history:
The first systematic study of flanking sequence effects on mismatch detection by a fluorescent nucleotide
is described, using fluoroside Çf. Although a high degree of variance was observed in fluorescence inten-
sity of mismatched duplexes between different flanking sequences, Çf was able to distinguish a mismatch
from the fully base-paired duplex in 13 out of 16 sequences, and even identify each mismatch in 10 of
those flanking sequences. For the flanking sequences where fluoroside Çf did not unambiguously deter-
mine its base-pairing partner, the experimental conditions were varied in an attempt to facilitate mis-
match identification. No beneficial effect on the relative fluorescence intensities was achieved by
changing the temperature, adding organic co-solvents or potassium iodide. In contrast, mercuric ions
selectively quenched the fluorescence intensity of the ÇfꢀT mismatch, effectively resolving the overlap
of all emission spectra and thereby facilitating identification of all base-pairing partners in any flanking
sequence by Çf. This is the first time mercuric ions have been used to selectively quench the fluorescence
of a single mismatch. A noticeable characteristic of Çf is that, unlike most fluorosides, the fluorescence
intensity of Çf was not quenched to a discernable degree by a flanking G–C pair.
Received 18 March 2010
Revised 11 June 2010
Accepted 16 June 2010
Available online 22 June 2010
Keywords:
Single nucleotide polymorphisms (SNP)
Phenoxazine
Fluorescence quenching
Mercuric ions
Base-discriminating fluorosides (BDFs)
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
Second, the excitation wavelength of the fluorescent probe must
be different from where DNA absorbs. Third, it is preferable that
Single nucleotide polymorphisms (SNPs) are single nucleotide
mutations and the most abundant type of genetic variation in the
human genome.1–3 SNPs have been linked to various diseases
and disorders.4,5 For example, SNPs positioned in the 50-flanking
region or the regulatory intronic region have been reported to
cause allele-specific effects on the expression of genes.6,7 The iden-
tification of SNPs, as disease-associated variants, is therefore an ac-
tive area of research.
Various methods exist to detect SNPs, most of which rely on
monitoring the hybridization of an oligonucleotide probe to a tar-
get sequence of interest.8 The probe binds selectively to the wild-
type sequence and not to sequences containing a SNP site.9 Most
commonly the hybridization is monitored by a fluorescent probe
which is attached to the oligonucleotide probe.8 Hybridization as-
says require very specific conditions, that is, careful selection of the
probe sequence and more importantly, the temperature for anneal-
ing must be carefully controlled. A fluorescent probe that could di-
rectly identify its base-pairing partner without the cumbersome
hybridization process would therefore be advantageous.
the probe has a high Stokes shift (>100 nm) and that the fluores-
cence maximum is in the visible or near-infrared region, to facili-
tate detection and interpretation.
Saito and co-workers have reported several fluorescent nucleo-
tides that, after incorporation into duplex DNA, are able to discrim-
inate between fully base-paired and mismatch-containing
duplexes. Such nucleosides have been called ‘base-discriminating
fluorescent nucleosides’ (BDFs).11 However, BDFs have limitations,
in that they can usually only distinguish one mismatch from the
fully base-paired duplex.12,13 Furthermore many fluorescent
probes experience quenching when incorporated into DNA,14
which can be severe when the fluorescent nucleoside is flanked
by a G–C pair.15–18 Other methods of detecting a SNP site are for
example, forced intercalation probes (FIT-probes) reported by Seitz
et al.19–21 and internal charge transfer probes (ICT-probes) re-
ported by Wagenknecht,22 both of which can detect mismatches,
but are unable to identify the mismatch in question.
Our group has previously described the fluorescent nucleoside
Çf (Fig. 1), which, after incorporation into duplex DNA, is able to
uniquely identify each of the four DNA bases in duplex DNA.23
Upon further examination it became evident that flanking se-
quence had an effect on the mismatch detection.23 Here we report
a systematic study of the effects of all the flanking sequences on
mismatch detection using fluorescent nucleoside Çf. We found a
large variation in the fluorescence intensities of the mismatches
In order to be able to use such a fluorescent probe, several as-
pects must be taken into consideration. First, the fluorescent probe
must have appreciable fluorescence brightness for general use.10
* Corresponding author. Tel.: +354 525 4801; fax: +354 552 8911.
E-mail address: snorrisi@hi.is (S.Th. Sigurdsson).
0968-0896/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2010.06.060

6122
H. Gardarsson, S. Th. Sigurdsson / Bioorg. Med. Chem. 18 (2010) 6121–6126
Figure 1. (A) The fluorescent nucleoside Çf. (B) The fluoroside Çf base-paired to
guanosine, where dR denotes 20-deoxyribose.
Figure 2. Normalized absorption (–) and emission (- -) spectra of 10 l
M Çf in abs.
EtOH.
between different flanking sequences. However, it was still possi-
ble to identify the base-pairing partner of Çf in the majority of
the sequences. In the sequences in which the fluoroside was unable
to discriminate between the mismatches, we were able to alter the
experimental conditions so that Çf could identify any base-pairing
partner in any sequence. Mercuric ions were especially useful as
they enabled selective quenching of the fluorescence intensity of
T-mismatched duplexes by utilizing the T-Hg+ complex formation.
This is the first example where mercuric ions have been utilized in
this fashion, that is, selective quenching of a fluorophore in close
proximity to T, and should be applicable with other fluorescent re-
porter groups.
the spin labelled DNA with Na2S, highly fluorescent DNA was ob-
tained, which was attributed to the amine derivative (Çf) of the
spin label.23 As the amine is a precursor for the nitroxide, we
sought to incorporate the phosphoramidite of Çf directly, avoiding
the oxidation and subsequent reduction steps (Scheme 1). A dime-
thoxytrityl group was introduced at the 50-OH group of the fluoro-
side, followed by phosphitylation of the 30-OH. The Çf-
phosphoramidite was purified by repetitive precipitation from a
CH2Cl2/n-hexane mixture and obtained in 66% yield over two steps.
2. Results and discussion
2.2. Synthesis of modified oligonucleotides (ODNs)
2.1. Synthesis of Çf phosphoramidite
Çf phosphoramidite (2) was used in automated ODN synthesis
to prepare the 16 ODNs for all possible immediately flanking se-
quences adjacent to the fluorescent nucleotide. The sequences of
the 14-mer ODNs are the same except for the Çf-flanking bases.
Syntheses of the modified ODNs was performed by following stan-
dard protocols, except for the incorporation of the modified phos-
phoramidite, for which the coupling time was extended (from
60 s + 30 s to 2 ꢁ 300 s). All ODNs were purified by denaturing
polyacrylamide gel electrophoresis (DPAGE) and their identity con-
firmed by MALDI-TOF mass spectrometry.
Our group has previously described the synthesis of Ç, a nitrox-
ide spin label, and its incorporation into DNA.24 Upon reduction of
Table 1
Photophysical properties of modified ODNs having the general sequence 50-d(GAC-
CTC-NÇfN-TCGTG), where N is G, C, A or T
FBd
a
b
c
kab
kem
UF
50-d(GÇfG)
50-d(GÇfC)
50-d(GÇfA)
50-d(GÇfT)
50-d(GÇfG)
50-d(GÇfC)
50-d(GÇfA)
50-d(GÇfT)
50-d(GÇfG)
50-d(GÇfC)
50-d(GÇfA)
50-d(GÇfT)
50-d(GÇfG)
50-d(GÇfC)
50-d(GÇfA)
50-d(GÇfT)
360
361
364
361
364
365
365
365
361
362
359
362
362
364
364
364
445
442
443
442
448
443
447
447
449
449
450
449
448
449
451
450
0.201 0.007
0.212 0.005
0.252 0.005
0.215 0.006
0.127 0.003
0.132 0.004
0.137 0.003
0.102 0.002
0.207 0.004
0.119 0.003
0.176 0.006
0.156 0.004
0.179 0.003
0.133 0.009
0.190 0.004
0.140 0.004
1351
1319
2599
1328
921
960
999
761
1347
726
948
866
1219
888
1150
795
a
b
c
Absorbance maxima.
Emission maxima.
Quantum yield of ODNs in H2O.
Fluorescence brightness, which is the multiplication product of the extinction
Scheme 1. Synthesis of phosphoramidite 2 from fluorescent nucleoside Çf. DMTCl:
4,40-dimethoxytrityl chloride, DMAP: 4-dimethylaminopyridine, DIPAT: diisopro-
pylammonium tetrazolide.
d
coefficient at 365.5 nm and the quantum yield.

H. Gardarsson, S. Th. Sigurdsson / Bioorg. Med. Chem. 18 (2010) 6121–6126
6123
Figure 3. (A) The DNA sequence used for the flanking study, where N is G, C, A or T and X is Çf’s base-pairing partner. (B) The color codes of the fluorescence curves for the
different Çf base pairs. (C) The relative fluorescence intensity of all 64 duplexes. The nucleotides flanking the 50- and 30-side of Çf are red and blue, respectively and change
from G, C, A and T horizontally (50-flanking) and in the same order vertically for 30-flanking side. The sequences that can readily distinguish between all base-pairing partners
have a green background, while those that are only able to identify a mismatch from the fully base-paired duplex have a yellow background. The sequences in which Çf is
unable to distinguish between the fully base-paired G and one of the mismatches have a red background. Each panel has been normalized separately by defining the emission
intensity of the most fluorescent duplex as 1.00. Panels are highlighted green if the difference in area under the corrected emission curves exceeds 15% for all duplexes,
excluding the A-mismatched duplex, which can easily be identified due to its unique 3-peak pattern in all duplexes. The panels are highlighted red, if any mismatched curve,
excluding the A-mismatch, does not exceed 15% deviation from that of the fully base-paired duplex (black curve) and yellow if the deviation between mismatched curves,
excluding the A-mismatch, does not exceed 15%.
2.3. Photophysical properties of the fluorescent nucleoside Çf
tum yields of Çf in EtOH, water, 1,4-dioxane and CH3CN were found
to be 0.32, 0.31, 0.40 and 0.37, respectively. The normalized
absorption and emission profile of the fluoroside in EtOH can be
seen in Figure 2. The optical densities of the solutions used for
quantum yield measurements were 0.02–0.07 and the quantum
yield values were corrected with the refractive indexes of the sol-
The UV–vis and steady-state fluorescence emission spectra of
the fluoroside were obtained in several solvents. Fluorescence
emission quantum yields of Çf were determined relative to a dilute
solution of anthracene (UF = 0.27) in absolute ethanol.25 The quan-

6124
H. Gardarsson, S. Th. Sigurdsson / Bioorg. Med. Chem. 18 (2010) 6121–6126
vents.26 Degassing the solutions by sonication, or passing argon
through the solution, before quantum yield determination had no
significant effect on the emission intensity.
these three flanking sequences is that the emission profiles of
the C- and T-mismatch overlap. A third and final class of modi-
fied sequences are the red panels in Figure 3C (3 out of 16),
which contain those sequences in which the fluoroside is incapa-
ble of distinguishing all mismatches from the fully base-paired
duplex. In the red panels the T-mismatch and G-match emission
profiles are nearly superimposable.
As mentioned above, the duplexes containing an A-mismatch
exhibit a unique spectral profile in all sequences. Rather than one
emission peak, three distinct peaks are observed at approximately
420, 450 and 480 nm, with an additional shoulder at 525 nm. The
reason for the appearance of these vibrational peaks is unclear, but
is possibly associated with a tautomer of the fluoroside which, if
formed, would stabilize the A-mismatched duplex with an addi-
tional hydrogen bond. Another intriguing result is that the fluores-
cence intensity of the C-mismatch is consistently higher than that
of the fully base-paired duplex. Therefore, Çf can distinguish and
discriminate the A- and C-mismatch from the fully base-paired du-
plex in all sequences.
All UV–vis spectra contain one absorption band near the visible
region, with the absorption maximum ranging between 359 and
364 nm. The fluorescence spectra lie within k = 380–600 nm, with
the emission maximum between 446 and 450 nm, yielding blue
fluorescence. The molar extinction coefficient of the fluoroside in
EtOH was calculated from Beer’s law and found to be
e
365.5 = 11,330 190 Mꢂ1 cmꢂ1. Therefore, Çf is a relatively bright
fluorescent nucleoside.10 Experiments in which the ion concentra-
tion was altered from 0 mM to 1000 mM NaCl, exhibited no signif-
icant change on the fluorescence intensity (data not shown).
2.4. Photophysical properties of single stranded Çf-modified
ODNs
The UV–vis and steady-state fluorescence emission spectra
were obtained in H2O with concentration of single stranded DNA
ranging from 3.0 to 12.5
l
M. The UV–vis and fluorescence spectra
All the flanking sequences that do not show either mismatch
identification (green) or mismatch detection (yellow) have one
thing in common: The emission of the T-mismatched duplex over-
laps with the emission of another base-pairing partner. Figure 3
clearly shows that in 10 out of 16 sequences, Çf can distinguish
and identify the T-mismatch from the other possibilities. However,
that leaves the 6 sequences 50-d(TÇfT), 50-d(AÇfT), 50-d(CÇfA), 50-
d(AÇfC), 50-d(GÇfC) and 50-d(CÇfG), where the fluorescent probe is
unable to unambiguously identify the T-mismatch. Therefore, we
varied the conditions under which fluorescence was recorded, in
an attempt to facilitate discrimination between all base-pairing
partners.
were similar to those of the fluorescent nucleoside in solution, that
is, a single absorption peak near the visible region and the emission
maximum around 450 nm. Table 1 lists the photophysical proper-
ties of the 16 Çf-modified ODNs. The fluorescence intensity of Çf is
generally higher when the fluoroside has a 50-flanking G. The fluo-
rescence brightness is especially high in the sequence 50-d(GÇfA),
attributed to an especially high molar extinction coefficient, in
addition to the highest quantum yield of all modified ODNs. The re-
lated sequence 50-d(AÇfG) also exhibits the highest emission inten-
sity of its series, that is, when the fluoroside has a 50-flanking A. The
quantum yield, and therefore the brightness, is lower in sequences
where there is a 50-flanking C, but the quenching effects are not
nearly as severe as described for other BDFs.15 Some of the ob-
served variations in emission intensity could be due to secondary
structure formation, however, no such stable structures were pre-
dicted by mFold (not shown).
2.6. Organic co-solvents, potassium iodide and temperature as
external effectors to induce mismatch discrimination
As the emission intensity of the fluorescent nucleoside showed
some variation in fluorescence with changes in solvent, we postu-
lated that the addition of an organic co-solvent to the DNA sample
would induce changes in the emission. Of the four organic co-sol-
vents examined (EtOH, DMF, THF and 1,4-dioxane), only THF in-
duced better discrimination between mismatches, but to a small
extent (data not shown). That may be due to the fact that the
amount of organic co-solvent could only be as high as ca. 25%, at
which point the DNA samples started to precipitate. A KI titration
was also performed to assess the solvent accessibility of the
probe.27,28 The relative fluorescence intensity of matched and mis-
matched duplexes did not change upon addition of KI (data not
shown), indicating a similar exposure of the fluorescent nucleoside
to the organic solvent.
Temperature-dependence of fluorescence was also investi-
gated. Fluorescence intensities were measured as a function of
temperature using two different sequences, 50-d(TÇfT) and 50-
d(TÇfA). These sequences were chosen because the former is an
example of sequence where Çf is unable to identify its base-pair-
ing partner, but in the latter Çf can distinguish and identify all
base-pairing partners. While shifting the temperature from 10 °C
to 40 °C, a gradual increase in mismatch discrimination was affor-
ded in 50-d(TÇfT) (Fig. 4A). However, the mismatch discrimination
was lowered in 50-d(TÇfA) (Fig. 4B). The emission intensity of the
T-mismatch does not change significantly for these two flanking
sequences upon increasing the temperature. The emission intensi-
ties of the A-mismatch and the fully base-paired duplex were
lowered, while the emission intensity of the C-mismatch in-
creased. Therefore, while temperature changes can enhance the
discrimination in some sequences, it will reduce the discrimina-
tion in others.
2.5. Fluorescence of Çf-modified complementary and
mismatched duplexes
Each of the 16 fluorescent ODNs was annealed to its comple-
mentary strand (G opposing Çf) as well as the three possible mis-
matches (A, C and T opposing Çf). All duplexes are expected to be
stable at 20 °C, based on TM values of 46–63 °C for the oligonucleo-
tide with flanking sequence 50-d(GÇfA).23 The relative fluorescence
intensities of the 64 duplexes at 20 °C is shown in Figure 3.
It is clear from the data in Figure 3 that there is a high degree
of variance in the order of fluorescence intensity for the individ-
ual base-pairings between the different flanking sequences. Some
modified sequences are clearly superior to others when it comes
to discrimination between mismatches, or simply to distinguish a
mismatch altogether. The flanking sequences can be grouped into
three distinct classes, based on the difference in fluorescence;
spectra are considered to overlap one another if the difference
in the area under the corrected emission curves is less than
15%, except for A-mismatched duplexes which can clearly be
identified due to the unique 3 peak pattern in the emission spec-
tra. In the first class (Fig. 3C green background), Çf is able to dis-
tinguish and identify the mismatches. Çf can identify its base-
pairing partner in 10 out of 16 sequences. In the second class
(Fig. 3C yellow background), the fluorescent probe can distin-
guish the fully base-paired duplex from a mismatch, but is un-
able to identify the mismatch in question (3 out of 16). In
these sequences the fully base-paired duplex has comparatively
lower fluorescence intensity than the mismatches. The reason
why Çf is incapable of identifying its base-pairing partner in

H. Gardarsson, S. Th. Sigurdsson / Bioorg. Med. Chem. 18 (2010) 6121–6126
6125
2.7. The effects of mercuric ions (Hg2+) on mismatch detection
As previously mentioned, the reason that the base-pairing part-
ner of Çf cannot be identified in all flanking sequences, is due to
overlap of the emission profile of the T-mismatch with other
base-pairing partners. To enhance mismatch discrimination, we
therefore sought a way to selectively change the emission proper-
ties of ÇfꢀT. Mercuric ions bind to nucleotides, in particular T,29–31
and are also known to quench the fluorescence of aromatic hydro-
carbons.32,33 The quenching effect is based on the spatial distance
between the fluorophore and the mercuric ions.34 Therefore, we
hypothesized that the selectivity of the T-Hg+ complex formation
could be utilized to bring the mercuric ion into close proximity
to Çf and thereby selectively quench the emission of the T-
mismatch.
Upon titrating a 1 mM HgCl2 solution into the DNA duplexes of
50-d(AÇfT), only a gradual decrease was initially observed in the
fluorescence intensity, presumably due to the presence of EDTA
Figure 4. Normalized fluorescence spectra of DNA oligomers 50-d(TÇfT) (A) and 50-
d(TÇfA) (B) at 10 °C (left) and at 40 °C (right), ÇfꢀG (black), ÇfꢀC (red), ÇfꢀA (blue) and
ÇfꢀT (green). All spectra are normalized in reference to the ÇfꢀC mismatched duplex
at 40 °C, within each panel (A and B), as this shows the highest emission intensity.
Figure 6. The relative fluorescence spectra of the sequences titrated with a 1 mM
HgCl2 solution. In the absence (left) and presence (right) of mercuric ions. (A) 50-
d(TÇfT), (B) 50-d(AÇfT), (C)50-d(CÇfA), (D) 50-d(AÇfC), (E)50-d(GÇfC), (F)50-d(CÇfG). The
Figure 5. Titration curves of the fluorescence intensity of duplexes as a function of
mercuric ion concentration, using ODN 50-d(AÇfT). ÇfꢀG (black), ÇfꢀC (red), ÇfꢀA (blue)
and ÇfꢀT (green). 10
l
L aliquots of 1 mM Hg2+ solution were added to a solution of
final concentrations of added Hg2+ ions were 91, 70, 91, 48, 91 and 111
lM,
respectively. The color codes are the same for all spectra, ÇfꢀG (black), ÇfꢀC (red), ÇfꢀA
the oligonucleotides in a phosphate buffer (400
lL, 10 mM Na2HPO4, 100 mM NaCl,
(blue) and ÇfꢀT (green).
0.1 mM Na2EDTA, pH 7.00).

6126
H. Gardarsson, S. Th. Sigurdsson / Bioorg. Med. Chem. 18 (2010) 6121–6126
in the buffer (Fig. 5). After that, a dramatic decrease in the emission
intensity of the ÇfꢀT duplex was observed. The emission of the C-
mismatched duplex was also quenched, but to a smaller degree,
as C has a lower affinity for mercuric ions than T.29 The A-mis-
match and the fully base-paired duplex were affected much less.
Acknowledgments
We thank the Icelandic Research Fund (060028021) for finan-
cial support, Dr. Ajay Kumar Kale for technical assistance, Dr. Pavol
Cekan for valuable discussions, and the reviewers of this paper for
valuable suggestions.
At the highest concentration of added mercuric ions (200 lM), all
duplexes experienced considerable quenching, due to the general
quenching effects of mercury.34
Supplementary data
All flanking sequences where mismatch identification had
been problematic, 50-d(TÇfT), 50-d(AÇfT), 50-d(CÇfA), 50-d(AÇfC),
50-d(GÇfC) and 50-d(CÇfG), were treated with mercuric ions
(Fig. 6). All flanking sequences show the same trend as described
for the 50-d(AÇfT) mercuric titration, that is, a dramatic decrease
in emission of the ÇfꢀT duplex, while the C-mismatch is quenched
to a lesser extent. Furthermore, the two sequences 50-d(GÇfG)
and 50-d(CÇfT), which fall very close to the 15% cutoff, were sub-
jected to mercuric titrations. They followed the same trend as
other sequences (Supplementary data). Thus, by the addition of
mercuric ions, the mismatch identification problems in all flanking
sequences were resolved.
Supplementary data (experimental procedures, characteriza-
tion data and photophysical data) associated with this article can
be found, in the online version, at doi:10.1016/j.bmc.2010.06.060.
References and notes
1. Nakatani, K. Chembiochem 2004, 5, 1623.
2. Wang, D. G.; Fan, J. B.; Siao, C. J.; Berno, A.; Young, P.; Sapolsky, R.; Ghandour,
G.; Perkins, N.; Winchester, E.; Spencer, J.; Kruglyak, L.; Stein, L.; Hsie, L.;
Topaloglou, T.; Hubbell, E.; Robinson, E.; Mittmann, M.; Morris, M. S.; Shen, N.;
Kilburn, D.; Rioux, J.; Nusbaum, C.; Rozen, S.; Hudson, T. J.; Lipshutz, R.; Chee,
M.; Lander, E. S. Science 1998, 280, 1077.
3. Yamada, R. Nat. Clin. Pract. Rheumatol. 2008, 4, 210.
4. Suh, Y.; Vijg, J. Mutat. Res. 2005, 573, 41.
5. LaFramboise, T. Nucleic Acids Res. 2009, 37, 4181.
3. Conclusion
6. Kochi, Y.; Yamada, R.; Suzuki, A.; Harley, J. B.; Shirasawa, S.; Sawada, T.; Bae, S.
C.; Tokuhiro, S.; Chang, X.; Sekine, A.; Takahashi, A.; Tsunoda, T.; Ohnishi, Y.;
Kaufman, K. M.; Kang, C. P.; Kang, C.; Otsubo, S.; Yumura, W.; Mimori, A.; Koike,
T.; Nakamura, Y.; Sasazuki, T.; Yamamoto, K. Nat. Genet. 2005, 37, 478.
7. Okamoto, K.; Makino, S.; Yoshikawa, Y.; Takaki, A.; Nagatsuka, Y.; Ota, M.;
Tamiya, G.; Kimura, A.; Bahram, S.; Inoko, H. Am. J. Hum. Genet. 2003, 72, 303.
8. Beaudet, A. L.; Belmont, J. W. Annu. Rev. Med. 2008, 59, 113.
9. Twyman, R. M. Curr. Top. Med. Chem. 2004, 4, 1423.
The development of a fluorescent probe that can identify all
base-pairing partners is of considerable interest due to their possi-
ble use in SNP assays. Herein, we have described the emission
properties of fluorescent nucleoside Çf, its incorporation into
DNA and the emissive properties of the Çf-labelled ODNs and du-
plexes. In particular we have examined the effects of flanking se-
quence on mismatch detection. These studies have shown that Çf
is a promising probe for SNP assays using synthetic oligonucleo-
tides. However the applicability of identifying all mismatches in
small amounts of unknown samples remains to be seen.
10. Sandin, P.; Börjesson, K.; Hong, L.; Martenson, J.; Brown, T.; Wilhelmsson, L. M.;
Albinsson, B. Nucleic Acids Res. 2008, 36, 157.
11. Okamoto, A.; Saito, Y.; Saito, I. J. Photochem. Photobiol., C 2005, 6, 108.
12. Okamoto, A.; Tanaka, K.; Saito, I. J. Am. Chem. Soc. 2003, 125, 9296.
13. Krueger, A. T.; Kool, E. T. J. Am. Chem. Soc. 2008, 130, 3989.
14. Hudson, R. H.; Ghorbani-Choghamarani, A. Synlett 2007, 870.
15. Dohno, C.; Saito, I. Chembiochem 2005, 6, 1075.
When reviewing the results of the mismatch detection, it is
apparent that there is a high degree of variance in the order of fluo-
rescence intensity for the individual base-pairings between the dif-
ferent flanking sequences. The sequences can be categorized into
three groups. First, in 10 out of 16 sequences, Çf is able to distin-
guish and identify a mismatch. Second, in 3 out of 16 sequences,
Çf is able to distinguish a mismatch from the fully base-paired du-
plex. In the remaining 3 sequences, Çf is unable to distinguish all
mismatches from the fully base-paired duplex. In all sequences
the A-mismatched duplex has a unique peak pattern, which effec-
tively makes Çf an ideal SNP probe for the A-allele of any SNP site.
Furthermore, the fluorescence intensity of Çf is not severely
quenched by a flanking G–C pair, unlike many other BDFs.
16. Driscoll, S. L.; Hawkins, M. E.; Balis, F. M.; Pfleiderer, W.; Laws, W. R. Biophys. J.
1997, 73, 3277.
17. Gaied, N. B.; Glasser, N.; Ramalanjaona, N.; Beltz, H.; Wolff, P.; Marquet, R.;
Burger, A.; Mély, Y. Nucleic Acids Res. 2005, 33, 1031.
18. Saito, Y.; Hanawa, K.; Motegi, K.; Omoto, K.; Okamoto, A.; Saito, I. Tetrahedron
Lett. 2005, 46, 7605.
19. Kohler, O.; Jarikote, D. V.; Seitz, O. Chembiochem 2005, 6, 69.
20. Dose, C.; Seitz, O. Bioorg. Med. Chem. 2008, 16, 65.
21. Socher, E.; Jarikote, D. V.; Knoll, A.; Roglin, L.; Burmeister, J.; Seitz, O. Anal.
Biochem. 2008, 375, 318.
22. Wagenknecht, H. A. Ann. N.Y. Acad. Sci. 2008, 1130, 122.
23. Cekan, P.; Sigurdsson, S. T. Chem. Commun. 2008, 29, 3393.
24. Barhate, N.; Cekan, P.; Massey, A. P.; Sigurdsson, S. T. Angew. Chem., Int. Ed.
2007, 46, 2655.
25. Dawson, W. R.; Windsor, M. W. J. Phys. Chem. 1968, 72, 3251.
26. Demas, J. N.; Crosby, G. A. J. Phys. Chem. US 1971, 75, 991.
27. Flowers, S.; Biswas, E. E.; Biswas, S. B. Biochemistry 2003, 42, 1910.
28. Huang, S.; Tu, S.-C. Photochem. Photobiol. 2005, 81, 425.
29. Yamane, T.; Davidson, N. J. Am. Chem. Soc. 1961, 83, 2599.
30. Kosturko, L. D.; Folzer, C.; Stewart, R. F. Biochemistry 1974, 13, 3949.
31. Tanaka, Y.; Oda, S.; Yamaguchi, H.; Kondo, Y.; Kojima, C.; Ono, A. J. Am. Chem.
Soc. 2007, 129, 244.
For the sequences in which Çf was incapable of uniquely identi-
fying its base-pairing partner, mercuric ions proved extremely use-
ful to facilitate discrimination. The mercuric ions selectively
quenched the emission of the T-mismatched duplexes, eliminating
any spectral overlap. This enabled mismatch detection in all the se-
quences and shows that Çf is a probe that identifies individual mis-
matches, independent of the flanking sequence. To our knowledge,
this is the first time that mercuric ions have been used to selec-
tively quench the fluorescence of a probe that is proximal to T,
and should find use in other fluorescence-based assays.
32. Rurack, K. Spectrochim. Acta A Mol. Biomol. Spectrosc. 2001, 57, 2161.
33. Rurack, K.; Bricks, J. L.; Schulz, B.; Maus, M.; Reck, G.; Resch-Genger, U. J. Phys.
Chem. A 2000, 104, 6171.
34. Masuhara, H.; Shioyama, H.; Saito, T.; Hamada, K.; Yasoshima, S.; Mataga, N. J.
Phys. Chem. 1984, 88, 5868.