Synthesis and electrochemical studies of an anthraquinone-conjugated nucleoside and derived oligonucleotides

Article abstract of DOI:10.1039/b816820b

The synthesis of a 2′-deoxyuridine nucleoside linked to an anthraquinone moiety, and its incorporation into oligonucleotides are described, including a facile oxidative demethylation with phenyliodine(iii) bis(trifluoroacetate) to reveal the anthraquinone motif. Furthermore, some useful physical and electrochemical properties of the obtained oligonucleotide are also reported, which allow its principal use in electrochemical DNA assays.

Full text of DOI:10.1039/b816820b

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Synthesis and electrochemical studies of an anthraquinone-conjugated
nucleoside and derived oligonucleotides†
Mikkel F. Jacobsen, Elena E. Ferapontova and Kurt V. Gothelf*
Received 25th September 2008, Accepted 18th November 2008
First published as an Advance Article on the web 23rd January 2009
DOI: 10.1039/b816820b
The synthesis of a 2¢-deoxyuridine nucleoside linked to an anthraquinone moiety, and its incorporation
into oligonucleotides are described, including a facile oxidative demethylation with phenyliodine(III)
bis(trifluoroacetate) to reveal the anthraquinone motif. Furthermore, some useful physical and
electrochemical properties of the obtained oligonucleotide are also reported, which allow its principal
use in electrochemical DNA assays.
Anthraquinone-modified oligonucleotides have proven to be ver-
satile tools in stabilization of duplex DNA by intercalation,1
electrochemical detection of single-base mismatches (SNPs),2 and
as photoexcitable probes for the study of DNA hole transport.3
Charge-transfer phenomena in DNA either through oxidative
or reductive pathways have received considerable attention in
recent years due to their importance in biological environments
such protein–DNA complexes, DNA damage, mutations and
cancer.4 As previously employed for pyrene-5 and phenothiazine-
linked nucleosides,6 we aimed for the synthesis of an AQ-dU
nucleoside 1 with conjugation between the anthraquinone and
the uridine moiety that could be employed in studies of charge-
transfer processes in DNA (Fig. 1).7 Here we disclose the synthesis
of an AQ-dU nucleoside, its incorporation into oligonucleotides,
and electrochemical studies of their use for detection of comple-
mentary oligonucleotides.
Scheme 1 Synthesis of AQ-dU nucleoside 1.
furnished the anthraquinone boronic ester 3 in reasonable yield.
The synthesis of the analogous pinacol ester of 3 by the Miyaura
Fig. 1 AQ-dU nucleoside 1.
procedure8 was unsuccessful.
Unfortunately, the subsequent Suzuki coupling with 2¢-deoxy-5-
iodouridine derivatives was unsuccessful, and led to extensive de-
composition. Instead, the synthesis of dimethylated hydroquinone
boronic ester 5 was carried out, that subsequent to any Suzuki
reaction requires both demethylation and oxidation in order to
reveal the anthraquinone motif. Hence, one-pot reduction and
dimethylation of 2 yielded 4. The organolithium species of 4
was subjected to borate trapping and further converted to the
pinacolate 5 in good yield. Nucleoside 6 is rather insoluble in
organic solvents, but it reacted well with 5 using a THF–MeOH–
H2O solvent mixture to afford 7 in 52% yield.
For the construction of the C–C bond between the nucle-
obase and the anthraquinone moiety we aimed for the use
of a Suzuki coupling employing an appropriate boronic ester
derivative of anthraquinone and 2¢-deoxy-5-iodouridine. As out-
lined in Scheme 1 two different boronic ester derivatives of an-
thraquinone were investigated. Starting from commercially avail-
able 2-bromoanthraquinone 2 the modified Miyaura procedure8
Danish National Research Foundation: Centre for DNA Nanotechnology
(CDNA) at the Interdisciplinary Nanoscience Center (iNANO) and the
Department of Chemistry, University of Aarhus, Langelandsgade 140, 8000
Aarhus C, Denmark. E-mail: kvg@chem.au.dk; Fax: +45 86196199
† Electronic supplementary information (ESI) available: Experimental
procedures and spectroscopic data for compounds 1, 4, 5, 7, 8 and 9,
cyclic voltammogram of AQ-dU (1) and differential pulse voltammogram
of O1 and O1+O3 duplex. See DOI: 10.1039/b816820b
An efficient oxidative demethylation of 7 employing hypervalent
iodine reagent PIFA9 smoothly afforded pure 1 in 98% yield, which
could be isolated by simple filtration (Scheme 2). The structure
of 1 was confirmed by spectroscopic measurements, including
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 905–908 | 905
©

View Article Online
Scheme 2 Synthesis of phosphoramidite 10 for oligonucleotide synthesis.
ESI-mass spectrometry and 2D-NMR experiments, while NOE
experiments confirmed the presence of a preferred anti glycosidic
torsion angle in solution. The UV-Vis spectrum of a yellow EtOH-
solution of 1 indicates a strong electronic interaction between
the uridine moiety and the covalently attached anthraquinone as
evident from the bathochromic shift of absorbance between 1 and
pure anthraquinone (e.g. lmax (anthraquinone) = 324 nm → lmax
(1) = 365 nm) (Fig. 2). Subsequently, protection of the 5¢-hydroxyl
group with DMTrCl gave 8 which could easily be converted into
the phosphoramidites 9 required for oligonucleotide synthesis.
Thermal denaturation experiments were performed in PBS-
solution (Fig. 3 and Table 1). These duplexes exhibit cooperative,
monophasic melting transitions. No hysteresis was observed
for sequential heating–cooling cycles which demonstrate rapid
hybridization kinetics.
Fig. 3 Oligonucleotides for thermal denaturation and electrochemical
studies.
Table 1 Thermal denaturation of duplexesa
Entry
Duplex
Tm/◦Cb,c
Fig. 2 UV-Vis spectrum of 1 and anthraquinone (both 50 mM in EtOH).
1
2
3
4
5
O1/O3
O2/O3
O3/O4
O2/O5
O4/O5
52.7
52.3
60.7
53.0
47.7
The incorporation of 9 into the oligonucleotide sequences O110,
O2 and O610 was carried out by standard automated solid-
phase synthesis for phosphoramidite nucleosides on a DNA-
synthesizer, which gave high coupling yields for 9. Following
standard deblocking and deprotection with aq. NH3-solution, the
oligonucleotide was purified by semi-preparative RP-HPLC and
quantified by UV-Vis absorption spectroscopy. The identity of O1,
O2 and O6 was confirmed by MALDI-TOF mass spectrometry
(see ESI†).
a Thermal denaturation was performed in PBS solution (1 M NaCl, 10 mM
phosphate buffer, pH 7.0) on a Cary 100 UV-Vis spectrophotometer
(Varian Inc.), and monitored at 260 nm with 1.0 ◦C min-1 in the range 15–
90 ◦C using three cycles. 2 mM strand conc. b All Tm values are reproducible
within 1◦C. c Melting temperatures were determined as the average of
three heating–cooling cycles (six measurements).
906 | Org. Biomol. Chem., 2009, 7, 905–908
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
The presence of 5¢-disulfide modification in O1 has virtually
no influence on duplex stability compared to oligonucleotide O2
(entries 1 and 2, Table 1). Evidently, the inclusion of AQ-dU into
the oligonucleotide decreases the duplex stability compared to
the unmodified duplex (entries 2 and 3). Furthermore, there is
an absence of discrimination between matched and mismatched
nucleotides for AQ-dU (entries 2 and 4), although AQ-dU
stabilizes the duplex compared to a CT mismatch (entries 2
and 5). The melting transition of duplex O1/O3 observed in
the region 330–380 nm, where only AQ-dU absorbs, also shows
significant hypochromicity which demonstrates the hydrophobic
interactions of AQ-dU with adjacent bases (Fig. 4). Interestingly,
these findings suggest that AQ-dU is mainly stabilizing the duplex
by an intercalation mode of binding despite its hydrogen-bonding
capacity.11
redox chemistry,12 and the amount of the immobilized AQ-DNA,
estimated from the redox peak area, was 8.4 1.6 pmol cm-2 (data
are averaged for 5 individual electrodes). This surface coverage
corresponds to a loosely packed surface monolayer,13 which
is advantageous for surface hybridization assays. The electron
transfer (ET) constant ks, determined from the cathodic and
anodic peak separation,14 was 1.1 0.2 s-1. Considering negligible
conductivity of a single-stranded (ss) DNA,15 we assume that the
observed ET phenomenon is due to the direct, i.e. non-mediated
by DNA, ET reaction between the electrode and AQ moiety.
Upon hybridization of O1 with a complementary strand O3,
the AQ redox peak separation increased slightly, signifying that
the kinetics of ET between AQ-dU and the electrode have not
essentially changed. The ks decreased to 0.8 0.2 s-1, and the
redox potential for AQ-dU shifted only slightly to the less negative
-411 2 mV. These very minor variations in the ET kinetics
may be ascribed to structural rearrangement of DNA due to
hybridization, enabling, (a) directional ET through the dsDNA
strands and, (b) less favorable “back-side” redox chemistry of
AQ-dU. The persistence length of double-stranded (ds) DNA is
around 50 nm, while that of ssDNA is around 1 nm.16 Thus, with
more flexible ssDNA, the redox probe can approach the electrode
surface closer than with a more rigid dsDNA structure, where AQ-
dU, fixed at a sequence predetermined position, should actually be
removed from the electrode surface when O1 hybridizes with O3.
The presence of a one-point mutation (hybridization of O1 with
O5) did not produce any additional effects on the ET efficiency,
which could be expected from thermal denaturation data. The
matched O1–O3 duplex has melting point characteristics close to
those of the one-mismatch O1(O2)–O5 duplex (Table 1).
In DNA assaying, one aims for clear discrimination between
ssDNA and dsDNA states. As seen from the data above, it
is difficult to achieve when AQ-dU is placed too close to the
electrode surface. Therefore we extended this to investigate the
electrochemistry of a 30 nts long AQ-dU DNA O6, which had
AQ-dU modification in its 21 nt position relative the electrode
surface (Fig. 3). For such ET distances, any ET in dsDNA is
expected to be low, while non-directional ET reactions of a more
flexible ssDNA are expected to proceed.
The CV of O6 showed well-resolved AQ redox chemistry,
though efficiency of ET was lower than that for O1: separation be-
tween AQ peaks corresponded to ks < 0.1 s-1. At scan rates higher
than 20 mV s-1, the cathodic peak shifted to the potentials of thiol
desorption; thus for estimations of hybridization–dehybridization
phenomena, we used the anodic peak clearly detectable in CVs
(Fig. 6, inset). Upon hybridization of O6 with complementary
O7, the peak totally disappeared (Fig. 6), which is consistent
with the removal of AQ-dU from the electrode surface upon the
formation of a more rigid dsDNA structure. Hence, the efficiency
of ET through the dsDNA appeared also to be very low at this
distance. After subsequent dehybridization, the AQ oxidation
peak reversibly appeared in the CV, and further disappeared
upon the next hybridization process. This on–off redox process
enabled us to electrochemically distinguish between ss and ds
DNA surface states. It is worth mentioning that addition of a
non-complementary DNA strand did not produce any changes in
electrochemistry of O6.
Fig. 4 Melting transition of AQ-dU in O1/O3 duplex (10 mM phosphate
buffer, 1.0 M NaCl, pH 7.0; 20 mM strand conc.).
We explored principal applicability of AQ-dU for electrochemi-
cal DNA assays. Cyclic voltammetry (CV) of an O1-modified gold
electrode revealed a pair of redox peaks with a mean potential of
-414 3 mV (Fig. 5). These values are in a good agreement with the
potential data obtained from the solution electrochemistry of AQ-
dU, -422 3 mV (see ESI†), which suggests that electrochemical
features of the built-in DNA AQ-dU are close to those of the
free AQ-dU base. The peak currents linearly depended on the
potential scan rate, which is characteristic of surface-confined
In summary, we have devised an efficient synthesis of an AQ-
derivatized nucleoside which is easily incorporated into
Fig. 5 Background-corrected cyclic voltammogram of a gold electrode
modified with O1. Potential scan rate is 50 mV s-1.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 905–908 | 907
©

View Article Online
1996, 118, 8747–8748; (c) S. Kumamoto, M. Watanabe, N. Kawakami,
M. Nakamura and K. Yamana, Bioconjugate Chem., 2008, 19, 65–
69; (d) M. Castano-Alvarez, M. T. Fernandez-Abedul and A. Costa-
Garcia, Electrophoresis, 2007, 28, 4679–4689; (e) T. T. Williams, C.
Dohno, E. D. A. Stemp and J. K. Barton, J. Am. Chem. Soc., 2004, 126,
8148–8158; (f) R. E. McKnight, J. Zhang and D. W. Dixon, Bioorg.
Med. Chem. Lett., 2004, 14, 401–404; (g) S. Kumamoto, H. Nakano,
Y. Matsuo, Y. Sugie and K. Yamana, Electrochemistry, 2002, 70, 789–
794.
2 (a) A. Okamoto, T. Kamei and I. Saito, J. Am. Chem. Soc., 2006, 128,
658–662; (b) A. Okamoto, T. Kamei, K. Tanaka and I. Saito, J. Am.
Chem. Soc., 2004, 126, 14732–14733.
3 Review: H.-A. Wagenknecht, Nat. Prod. Rep., 2006, 23, 973–1006 and
references therein.
4 Charge-transfer in DNA: From mechanism to Application, ed. H.-A.
Wagenknecht, Wiley-VCH, Weinheim, 2005.
5 E. Mayer-Enthart and H.-A. Wagenknecht, Angew. Chem., Int. Ed.,
2006, 45, 3372–3375.
6 C. Wagner and H.-A. Wagenknecht, Chem.–Eur. J., 2005, 11, 1871–
1876.
7 A. A. Gorodetsky, O. Green, E. Yavin and J. K. Barton, Bioconjugate
Chem., 2007, 18, 1434–1441.
Fig. 6 Cyclic voltammetry for a gold electrode modified with O6, before
(solid line) and after hybridization with a complementary O7 (dashed line).
Scan rate is 20 mV s-1. Inset: background-subtracted peak for oxidation
of AQ-dU, derived from the CV of O6 in the main figure.
8 L. Zhu, J. Duquette and M. Zhang, J. Org. Chem., 2003, 68, 3729–
3732.
9 H. Tohma, H. Moriaka, Y. Harayama, M. Hashizume and Y. Kita,
Tetrahedron Lett., 2001, 42, 6899–6902.
10 The 5¢-C6-disulfide modification was incorporated such that the
oligonucleotide may be immobilized on gold surfaces subsequently by
standard thiol-Au chemistry.
11 (a) K. Yamana, Y. Nishijima, T. Ikeda, T. Gokota, H. Ozaki, H.
Nakano, O. Sangen and T. Shimidzu, Bioconjugate Chem., 1990, 1, 319–
324; (b) H. Deshmukh, S. P. Joglekar and A. D. Broom, Bioconjugate
Chem., 1995, 6, 578–586.
oligonucleotides. Electrochemical studies demonstrated principal
applicability of the synthesized AQ-dU base for DNA assays.
Future work will focus on the implementation of such oligonu-
cleotides as photoexcitable probes for studies of DNA hole trans-
port and the further development of the electrochemical DNA
assays exploiting electrochemical and photochemical properties
of the designed artificial nucleotide, i.e. estimations of effects of
sequence positions of AQ-dU on selectivity and sensitivity of the
DNA assay.
12 A. J. Bard, and L. R. Faulkner, Electrochemical Methods - Fundamen-
tals and Applications, Wiley, New York, 1980.
13 (a) M. Chahma, J. S. Lee and H. B. Kraatz, J. Electroanal. Chem., 2004,
567, 283–287; (b) K. A. Peterlinz and R. M. Georgiadis, J. Am. Chem.
Soc., 1997, 119, 3401–3402; (c) D. Y. Petrovykh, H. Kimura-Suda, L. J.
Whitman and M. J. Tarlov, J. Am. Chem. Soc., 2003, 125, 5219–5226;
(d) A.-E. Radi, J. L. A. Sa´nchez, E. Baldrich and C. K. O’Sullivan,
J. Am. Chem. Soc., 2006, 128, 117–124.
14 E. Laviron, J. Electroanal. Chem., 1979, 101, 19–28.
15 G. Hartwich, D. J. Caruana, T. de Lumley-Woodyear, Y. Wu, C. N.
Campbell and A. Heller, J. Am. Chem. Soc., 1999, 121, 10803–
10812.
Acknowledgements
Financial support for this work by the Danish National Research
Foundation, and the Carlsberg Foundation is gratefully acknowl-
edged.
Notes and references
16 S. B. Smith, Y. Cui and C. Bustamante, Science, 1996, 271, 795–
1 (a) H. Ihmels and D. Otto, Top. Curr. Chem., 2005, 258, 161–204;
(b) D. Ly, Y. Kan, B. Armitage and G. B. Schuster, J. Am. Chem. Soc.,
799.
908 | Org. Biomol. Chem., 2009, 7, 905–908
This journal is
The Royal Society of Chemistry 2009
©