Selective detection of 5-formyl-2′-deoxyuridine, an oxidative lesion of thymidine, in DNA by a fluorogenic reagent

Article abstract of DOI:10.1002/anie.201004087

Chemoselectivity in DNA: The thymidine lesion 5-formyl-2′- deoxyuridine (1) induces the mutation of DNA. Its derivatization with the specific fluorogenic reagent 2-amino-4,5-dimethoxythiophenol (2) gives 3, which shows strong fluorescence. This fast method for the detection of 1 requires no enzymatic digestion, HPLC separation, or MS analysis. ROS=reactive oxygen species.

Full text of DOI:10.1002/anie.201004087

Communications
DOI: 10.1002/anie.201004087
DNA-Damage Detection
Selective Detection of 5-Formyl-2’-deoxyuridine, an Oxidative Lesion
of Thymidine, in DNA by a Fluorogenic Reagent**
Wataru Hirose, Kousuke Sato, and Akira Matsuda*
DNA in living cells is damaged by reactive oxygen species
derived from UV light, ionizing radiation, and cellular
respiration. 5-Formyl-2’-deoxyuridine (fodUrd), an oxidized
thymidine lesion generated in yields comparable to that of 2’-
deoxy-8-oxoguanosine,[1] induces mutation in DNA (AT-to-
GC transition) through the mispairing of an ionized form of
fodUrd with 2’-deoxyguanosine during DNA replication.[2] It
appears that the formation of fodUrd may cause carcinoge-
nicity and/or aging of cells.[3] A selective and more convenient
method for detecting fodUrd would be highly desirable, since
the existing methods are complicated, involve time-consum-
ing analysis by HPLC and/or mass spectrometry following
complete enzymatic hydrolysis of the target DNA, and
require isotope-labeled fodUrd (13C and/or 15N) as an internal
standard.[4]
Several 5-heteroaryl 2’-deoxyuridines with good fluorescence
properties have been reported.[9] Thus, fodUrd in DNA could
be detected directly by fluorescence measurement without
enzymatic hydrolysis of the target DNA to its corresponding
nucleosides, HPLC separation, and analysis.
First, we synthesized the reagent bis(4,5-dimethoxyanilin-
2-yl)disulfide (3) from 3,4-dimethoxyaniline (1) in two steps
(Scheme 1).[10] Since oxidation by air leads to the spontaneous
dimerization of reagent 3’, the disulfide 3 was reduced to 3’ by
the addition of dithiothreitol (DTT) just before treatment
with fodUrd. Oxidation with H2O2 in the presence of
Some other approaches have been developed for the
detection of DNA damage.[5] Molecular recognition through
the formation of a base pair between an oligonucleotide
probe containing a synthetic complimentary nucleobase and
O6-benzyl-dG was developed by Gong and Sturla.[6]
A
common method is to detect DNA damage with enhanced
reactivity by selective tagging with a molecule that contains a
reporter group. Ide et al. first demonstrated the detection of
abasic sites by the use of an aldehyde reactive probe.[7] Other
selective detection methods with chemical reagents have
recently been reported.[8] However, the signal-to-noise (S/N)
ratio is a problem in these methods, as the fluorescent and/or
colorimetric molecules used have higher background signals.
Herein, we report a new concept for the simple detection
of fodUrd in damaged DNA with a fluorogenic reagent. The
reagent 2-amino-4,5-dimethoxythiophenol (3’) shows no
fluorescence before reaction with the target fodUrd in
DNA. However, upon the reaction of 3’ with fodUrd, the
formyl group at the 5-position of fodUrd is converted into a
benzothiazol-2-yl group, which is directly conjugated with the
uracil group. This ring system is similar to luciferin, which
undergoes the luciferase reaction to produce luminescence.
[*] W. Hirose, Dr. K. Sato, Prof. Dr. A. Matsuda
Faculty of Pharmaceutical Sciences, Hokkaido University
Kita-12, Nishi-6, Kita-ku, Sapporo 060-0812 (Japan)
Fax: (+81)11-706-4980
E-mail: matuda@pharm.hokudai.ac.jp
[**] This research was supported in part by a Grant-in-Aid for Scientific
Research from the Ministry of Education, Science, Sports and
Culture of Japan. We thank S. Oka and A. Tokumitsu (Center for
Instrumental Analysis, Hokkaido University) for elemental analysis
and measurement of mass spectra.
Scheme 1. Synthesis of btdUrd (4): a) NH4SCN, Br2, AcOH, 108C,
78%; b) KOH, ethylene glycol, H2O, reflux, 82%; c) DTT, DMF; d) 3’,
DMF; e) H2O2, Sc(OTf)3, H2O, 75% (2 steps). DMF=N,N-dimethyl-
formamide, Tf=trifluoromethanesulfonyl.
Supporting information for this article, including experimental
details, is available on the WWW under http://dx.doi.org/10.1002/
anie.201004087.
8392
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 8392 –8394

Angewandte
Chemie
Figure 1. UV absorption, fluorescence spectra, and photophysical data
of 4. Thin lines show UV absorbance; thick lines show fluorescence.
Fluorescence spectra (5ꢀ10À7 m) were measured by excitation at the
wavelength of each absorbance maximum at 258C in aqueous HCl
(100 mm), phosphate buffer (10 mm, pH 7.0), and aqueous NaOH
(100 mm). The quantum yield (F) was obtained from the half width of
the spectrum by comparison with a known compound (quinine in
0.1m H2SO4).
Figure 2. Conversion of ODN-foU into ODN-btU with 3’. a) Reaction
scheme with the sequences of ODN-foU and ODN-btU; b) reversed-
phase HPLC chromatogram of ODN-foU (before reaction); c) reversed-
phase HPLC chromatogram of ODN-btU (after reaction).
correlation between the concentration of ODN-btU and the
fluorescence intensity over the range 1–100 pmol (Figure 3).
In control experiments, we also measured the fluorescence
intensity (lex: 345 nm, lem: 458 nm) of the 15-mers ODN-T,
which does not contain fodUrd, and ODN-AP, which contains
one abasic site, after their fluorogenic reaction with 3 under
the same conditions. Since abasic sites are generated as a
major aldehyde source in DNA at a rate of approximately 104
abasic sites per cell per day,[14] if an aldehyde reacts with 3’ and
the resulting benzothiazole derivative generates fluorescence,
the selective fluorogenic detection of fodUrd may be
disturbed. However, there was no fluorescence of ODN-T
and ODN-AP at 458 nm (although the abasic site reacted with
3’ in 58% yield, as detected by HPLC: see Figures S2 and S3
in the Supporting Information). According to the data in
Figure 3, the detection limit of our method is around
[11]
Sc(OTf)3 then gave 5-(5,6-dimethoxybenzothiazol-2-yl)-2’-
deoxyuridine (btdUrd, 4) in 75% yield.
The UV absorbance and fluorescence spectra of 4 were
dependent on the pH value of the solution containing 4
(Figure 1). BtdUrd showed much weaker fluorescence under
acidic (aqueous HCl, 100 mm) and neutral conditions (phos-
phate buffer, 10 mm, pH 7.0), with a quantum yield of 0.160
and 0.074, respectively, than under basic conditions. In
aqueous NaOH (100 mm), btdUrd showed the strongest
fluorescence at 458 nm (excitation at 345 nm), with a
quantum yield of 0.701 and a Stokes shift of 113 nm
(Figure 1). These are excellent values for a fluorescent
compound in aqueous solution,[12] and the results suggest
that deprotonation at N3 of the uracil moiety contributes to
the strong fluorescence and high quantum yield.
Next, we synthesized a 15-mer oligodeoxyribonucleotide
(ODN) containing fodUrd (ODN-foU)[13] and treated it with 3’
under our optimized conditions (Figure 2). Thus, ODN-foU
(1.0 mm) and 3 (50 mm), pretreated with DTT (250 mm) at
room temperature for 30 minutes, underwent the desired
reaction in acetate buffer (10 mm, pH 5.0) at room temper-
ature for 2 hours to give the ODN containing 4, ODN-btU, in
greater than 95% yield. No further oxidation reaction was
necessary to generate ODN-btU when a lower concentration
(1.0 mm) of ODN-foU was used in an aqueous buffer.
Dissolved oxygen could be responsible for oxidation of the
intermediate.
Figure 3. Fluorescence intensity (at 458 nm) of the reaction mixture
containing various amounts of the ODN derived from ODN-foU
(square), ODN-AP (diamond), and ODN-T (triangle) in aqueous
NaOH (100 mm). Fluorescence measurements were preformed at
258C with excitation at 345 nm.
The fluorescence intensity (lex: 345 nm, lem: 458 nm) of
the crude ODN-btU was measured in aqueous NaOH
(100 mm) after removal of the excess reagents by extraction
from the reaction mixture with AcOEt. We observed a linear
Angew. Chem. Int. Ed. 2010, 49, 8392 –8394
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8393

Communications
150 fmol, as the minimum S/N ratio is 3:1 (ODN-foU signal to
g irradiation) reported by Frelon et al.,[4a] although the g-
irradiation conditions differed.
ODN-T noise). Although this sensitivity is not yet sufficient,
detection by this method is more accurate than by previous
methods.[4] Hydrolysis by enzymatic digestion often failed
around nonnatural nucleotide moieties, and this failure made
detection by HPLC and/or mass spectrometry difficult.
However, our method can be used to detect even the
presence of nonhydrolyzed fodUrd in an oligonucleotide by
direct fluorescence measurement.
In conclusion, we have developed a new method for the
selective and convenient detection of the oxidized lesion of
thymidine, 5-formyl-2’-deoxyuridine (fodUrd), in DNA by
treatment with the fluorogenic reagent 3’. Our method does
not need long reaction times, any enzymatic digestion, HPLC
separation, or mass spectrometric analysis. Abasic sites and
fodCyd in DNA do not disturb fluorescence detection by our
fluorogenic method. Therefore, the aminothiophenol reagent
3’ would be useful for the selective detection and measure-
ment of fodUrd formed in DNA exposed to various forms of
oxidative stress.
We also attempted the reaction of ODN-foC, which
contains a 5-formyl-2’-deoxycytidine (fodCyd) moiety,[15]
with 3’, because fodCyd may be a product of an oxidized
lesion of 5-methyl-2’-deoxycytidine at 5-methyl CpG sequen-
ces in DNAunder oxidation conditions.[1b,14] Since no reaction
was observed by HPLC (see Figure S4 in the Supporting
Information), the existence of fodCyd in DNA may not
disturb the selective detection of fodUrd. We also attempted
the detection of ODN-foU in the presence of ODN-AP (0–
50 equiv) and found that ODN-AP did not interfere with the
formation of ODN-btU nor with the fluorescence detection of
ODN-btU (see Figure S5 in the Supporting Information).
Finally, g-irradiated calf-thymus DNA was prepared for
the detection of fodUrd formation. First, we confirmed the
chemical reactivity of 3’ with fodUrd in g-irradiated DNA
after enzymatic digestion followed by HPLC analysis (see
Figure S6 in the Supporting Information). FodUrd was found
at a retention time of about 10 minutes. After fluorogenic
derivatization by 3, the fodUrd peak disappeared, and a new
peak appeared around 26 minutes. DNA (50 mg) subjected to
g irradiation in various doses (0–300 Gy) was treated with 3’
(500 mm) in acetate buffer (10 mm, pH 5.0) at room temper-
ature for 2 hours, and the reaction mixture was then extracted
three times with EtOAc. After redissolution of the DNA in
NaOH (100 mm), the fluorescence emission at 458 nm (lex:
345 nm) was measured to estimate the yield of btdUrd and
calibrated by using an identical ODN containing btdUrd
(ODN-btU; Figure 4). The results showed that 8.9 fodUrd/106
bases/Gy could be detected by our method (equivalent to
0.94 Gy/min 60Co g irradiation) from the calibration curve.
This result is in reasonable agreement with the formation rate
of 15.3 fodUrd/106 bases/Gy (equivalent to 20 Gy/min 60Co
Received: July 5, 2010
Published online: September 22, 2010
Keywords: chemoselectivity · DNA damage · fluorescent probes ·
.
nucleosides · oligodeoxyribonucleotides
[1] a) H. Kasai, A. Iida, Z. Yamaizumi, S. Nishimura, H. Tanooka,
Mutat. Res. 1990, 243, 249 – 253; b) N. Murata-Kamiya, H.
Kamiya, N. Karino, Y. Ueno, H. Kaji, A. Matsuda, H. Kasai,
Nucleic Acids Res. 1999, 27, 4385 – 4390.
[2] A. Masaoka, H. Terato, M. Kobayashi, Y. Ohyama, H. Ide, J.
Biol. Chem. 2001, 276, 16501 – 16510.
[3] H. ꢀnensen, F. Provan, A. T. Lian, S.-H. H. S. Reinertsen, Y.
Ueno, A. Matsuda, E. Seeberg, S. Bjelland, Mutat. Res. 2001,
476, 99 – 107.
[4] a) S. Frelon, T. Douki, J.-L. Ravanat, J.-P. Pouget, C. Tornabe-
nen, J. Cadet, Chem. Res. Toxicol. 2000, 13, 1002 – 1010; b) H.
Hong, Y. Wang, Anal. Chem. 2007, 79, 322 – 326.
[5] H. A. Dahlmann, V. G. Vaidyanathan, S. J. Sturla, Biochemistry
2009, 48, 9347 – 9359.
[6] J. Gong, S. J. Sturla, J. Am. Chem. Soc. 2007, 129, 4882 – 4883.
[7] a) H. Ide, K. Akamatsu, Y. Kimura, K. Michiue, K. Makino, A.
Asaeda, Y. Takamori, K. Kubo, Biochemistry 1993, 32, 8276 –
8283; b) J. Lhomme, J.-F. Constant, M. Demeunynck, Biopoly-
mers 1999, 52, 65 – 83.
[8] a) K. Sato, M. M. Greenberg, J. Am. Chem. Soc. 2005, 127,
2806 – 2807; b) S. Dhar, T. Kodama, M. M. Greenberg, J. Am.
Chem. Soc. 2007, 129, 8702 – 8703; c) N. Kojima, T. Takebayashi,
A. Mikami, E. Ohtsuka, Y. Komatsu, J. Am. Chem. Soc. 2009,
131, 13208 – 13209.
[9] a) N. J. Greco, Y. Tor, J. Am. Chem. Soc. 2005, 127, 10784 –
10785; b) N. J. Greco, Y. Tor, Tetrahedron 2007, 63, 3515 –
3527; c) W. Hirose, K. Sato, A. Matsuda, Nucleic Acids Symp.
Ser. 2009, 53, 135 – 136.
[10] H. Matsuoka, N. Ohi, M. Mihara, H. Suzuki, K. Miyamoto, N.
Maruyama, K. Tsuji, N. Kato, T. Akimoto, Y. Takeda, K. Yano,
T. Kuroki, J. Med. Chem. 1997, 40, 105 – 111.
[11] a) T. Itoh, K. Nagata, H. Ishikawa, A. Ohsawa, Heterocycles
2004, 63, 2769 – 2783; b) B. Bi, K. Maurer, K. D. Moeller, Angew.
Chem. 2009, 121, 5986 – 5988; Angew. Chem. Int. Ed. 2009, 48,
5872 – 5874.
[12] L. D. Lavis, R. T. Raines, ACS Chem. Biol. 2008, 3, 142 – 155.
[13] H. Sugiyama, S. Matsuda, K. Kino, Q.-M. Zhang, S. Yonei, I.
Saito, Tetrahedron Lett. 1996, 37, 9067 – 9070.
[14] a) T. Lindahl, B. Nyberg, Biochemistry 1972, 11, 3610 – 3618;
b) A. Sancar, G. B. Sancar, Annu. Rev. Biochem. 1988, 57, 29 –
67; c) T. Lindahl, Nature 1993, 362, 709 – 715.
[15] N. Karino, Y. Ueno, A. Matsuda, Nucleic Acids Res. 2001, 29,
2456 – 2463.
Figure 4. Quantification of fodUrd in g-irradiated calf-thymus DNA
(50 mg) at various irradiation doses by calculation from the calibration
curve of ODN-foU (Figure 3). Each value was calculated from three
independent experiments; the error bars indicate standard deviation.
Fluorescence measurements (at 458 nm) were performed at 258C with
excitation at 345 nm.
8394 www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 8392 –8394