Benzoquinone and naphthoquinone based redox-active labels for electrochemical detection of modified oligonucleotides on Au electrodes

Article abstract of DOI:10.1016/j.tetlet.2015.12.035

The reaction of 1,4-benzoquinone and 1,4-naphthoquinone with β-thiopropionic acid yielded 3-[(3,6-dioxocyclohexa-1,4-diene-1-yl)thio]propionic and 3-[(1,4-dioxo-1,4-dihydronaphthalene-2-yl)thio]propionic acids, respectively. These compounds were used to modify oligonucleotides to allow their electrochemical detection on the surface of Au electrodes using cyclic voltammetry. Electrochemical reduction of the corresponding amides, modeling the bonding of the redox-active labels with molecules that are not electrochemically active, was also studied.

Full text of DOI:10.1016/j.tetlet.2015.12.035

Accepted Manuscript
Benzoquinone and naphthoquinone based redox-active labels for electrochem-
ical detection of modified oligonucleotides on Au electrodes
Leonid A. Shundrin, Irina G. Irtegova, Nadezhda V. Vasilieva, Irina A. Khalfina
PII:
S0040-4039(15)30450-0
DOI:
http://dx.doi.org/10.1016/j.tetlet.2015.12.035
TETL 47082
Reference:
Tetrahedron Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
10 September 2015
1 December 2015
8 December 2015
Please cite this article as: Shundrin, L.A., Irtegova, I.G., Vasilieva, N.V., Khalfina, I.A., Benzoquinone and
naphthoquinone based redox-active labels for electrochemical detection of modified oligonucleotides on Au
electrodes, Tetrahedron Letters (2015), doi: http://dx.doi.org/10.1016/j.tetlet.2015.12.035
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Leave this area blank for abstract
Benzoquinone and naphthoquinone based redox-
active labels for electrochemical detection of
modified oligonucleotides on Au electrodes
Leonid A. Shundrin, Irina G. Irtegova, Nadezhda V. Vasilieva, Irina A. Khalfina
1

Benzoquinone and naphthoquinone based redox-active labels for
electrochemical detection of modified oligonucleotides on Au electrodes
Leonid A. Shundrina,b,c*, Irina G. Irtegovaa,b, Nadezhda V. Vasilievaa, Irina A. Khalfinaa,b
a N. N. Vorozhtsov Institute of Organic Chemistry, Siberian Branch of the Russian Academy of
Sciences, Academician Lavrent’ev Avenue, 9 , 630090 Novosibirsk, Russian Federation
b Department of Natural Sciences, National Research University – Novosibirsk State University,
Pirogova Str. 2, 630090 Novosibirsk, Russian Federation
c The Scientific Institute of Clinical and Experimental Lymphology, Siberian Branch of the
Russian Academy of Medical Sciences, Timakova Str. 4, 630060 Novosibirsk, Russian
Federation
Keywords: quinones; redox-active labels; electrochemical reduction; cyclic voltammetry;
modified oligonucleotides
Abstract: The reaction of 1,4-benzoquinone and 1,4-naphthoquinone with -thiopropionic acid
yielded
3-[(3,6-dioxocyclohexa-1,4-diene-1-yl)thio]propionic
and
3-[(1,4-dioxo-1,4-
dihydronaphthalene-2-yl)thio]propionic acids, respectively. These compounds were used to
modify oligonucleotides to allow their electrochemical detection on the surface of Au electrodes
using cyclic voltammetry. Electrochemical reduction of the corresponding amides, modeling the
bonding of the redox-active labels with molecules that are not electrochemically active, was also
studied.
Elaboration of electrochemical DNA biosensors, also known as genosensors, has achieved
noticeable progress in the last decade.1 Methods used for DNA detection on the surface of
electrodes can be classified into two main types depending on the nature of the electrochemical
signal: label-free and label-based. The former use the signal resulting from hybridization itself2
while the latter use small electrochemically active molecules, which are inserted into
oligonucleotides and ds-DNA fragments (or amplicons) as redox-active labels. These labels must
meet strict demands: low redox potentials, chemical stability and reversibility towards electron
transfer in water buffers to avoid label decay under potential sweep. The main types of
compounds studied as redox-active labels in electrochemical genosensing technologies are based
* Corresponding author. Tel.: +7-383-330-94-32; fax: +7-383-330-97-52; e-mail: shundrin@nioch.nsc.ru
2

on ferrocene,3 9,10-anthraquinone derivatives,4 transition metals complexes with organic
electrochemically active moieties5 and some types of redox-active intercalators.6 Several types of
organic electrochemically active labels reported to date have close redox potentials, for example,
derivatives of 9,10-anthraquinone.4 Thus, the synthesis of redox-active labels with different and
well-discriminated potentials is a vital problem because a set of labels with various redox
potentials can be used for simultaneous electrochemical detection of different site-specific DNA
fragments in electrochemical microarray technologies. From this point of view the task of
synthesising a large number of redox-active labels is similar to the analogous task in
fluorescence based microarray analysis, for which a set of fluorescent labels with different
spectral properties has been synthesized.7 The goal of this study was the synthesis and testing of
redox-active modifiers of oligonucleotides to allow their electrochemical detection at the surface
of Au electrodes using cyclic voltammetry (CV). Well distinguished potentials of the redox
active modifiers in their cyclic voltammograms was the main requirement of this study, and the
creation of specific genosensing technologies was not considered at this stage.
Naphthoquinone (NQ), benzoquinone (BQ) and their derivatives have low electrochemical
reduction potentials, and their electrochemistry has been well studied in various media.8-11 The
stable electrochemical activity of NQ and BQ was exploited for their application in biosensors in
recent years. Charge transfer interaction between quinones and melamine was used to determine
the latter by voltammetric methods.12 NQ-based polymers have been applied to design
electrochemical immunosensors.13 Several intracellular nanosensors based on BQ- and NQ-
modified nanoparticles were recently proposed for the determination of cell oxidative stress. 14,15
NQ and BQ moieties possess well distinguished potentials in water buffers making them suitable
to use as redox-active labels for oligonucleotides.
NQ(BQ)-based modifiers (1, 2) for oligonucleotides were synthesized using S-nucleophilic
addition of -thiopropionic acid to the double bond of NQ and BQ followed by aerobic oxidation
or using PbO2 (for details see ESI). To test 1, 2 as redox active labels, the corresponding amides
(3, 4) were synthesised. These compounds model the bonding of 1, 2 with molecules that are not
electrochemically active. Then, the electrochemical reduction (ECR) of 3, 4 in MeCN and
MeCN:H2O mixtures with variable water content were studied using cyclic voltammetry.16
3

O
O
O
O
S
R
S
R
O
O
2 (4)
1 (3)
Figure 1. Structures of the compounds. R= OH (1, 2); NHiPr (3, 4).
The mechanisms of NQ and BQ ECR are known to be different in aprotic solvents and
H2O.7 ECR of both compounds is an electron-electron (EE) - process in aprotic solvents with the
formation of a long-lived radical anion (RA) and dianion (DA), whereas protonation of RA and
DA should be taken into account in the aqueous media.8 Cyclic voltammograms (CV) of 3, 4 in
MeCN and its mixtures with H2O were characterized by two or three (for 3 in MeCN only)
diffusion controlled ( IpiС –1/2 = const, where IpiC is the peak current17, and  is the potential
sweep rate) reduction peaks (Fig. 3a-d for 3). The CV of 4 are shown in ESI Fig. S1. The
corresponding peak potentials of 3, 4 (Table 1) become less negative with the increase of the
H2O content in the mixture. The second peak becomes irreversible when the molar fraction of
H2O (m) exceeds 0.1 for 3 and 0.2 for 4, and in both cases is not observable when m>0.4. The
only diffusion controlled and one-electron reversible peak (Ep= Ep1A  Ep1C =0.058 V) was
observed in the CV of both compounds in H2O (Fig. 3d,f, ESI Fig. S1d). To understand the
changes in the CV of 3, 4 in binary MeCN:H2O mixtures, we considered a common ECR
mechanism, which includes three one-electron electrochemical reactions and three non
electrochemical protolytic equilibria where Q is 3 or 4 (Figure 2).
4
QH
+
+e
+H O
3
-e
-H O
2
+
+
+H O
+H O
3
3
+e
-e
+e
-e
Q2-
3
QH-
5
Q
1
Q
2
QH2
6
-H O
-H2O
2
Figure 2. Common ECR mechanism of quinone-type compounds.
4

Table 1. Experimental reductive peak potentials
of compounds 3, 4 in MeCN:H2O
mixtures.
3
4
Ep1C , V Ep2C , V
Ep1C , V Ep2C , V
m
0a
m
−0.70 −1.19
0
0.1
−0.52 −1.11
−0.40 −0.63
0.1 −0.59 −0.82
0.244 −0.54 −0.69 0.218 −0.34 −0.53
0.301 −0.53 −0.67 0.334 −0.31 −0.50
0.447 −0.51
0.519 −0.51
0.921 −0.37
0.977 −0.33
1.0 −0.32
—
—
—
—
—
0.456 −0.29
0.528 −0.27
0.737 −0.23
0.808 −0.21
—
—
—
—
—
1.0
−0.13
a
E
3C = 1.38 V was observed for 3 only when m=0.
p
Figure 3. Cyclic voltammograms (solid line) and their simulations (circles) of 3 in MeCN
(a), MeCN:H2O mixtures (m= 0.1 (b), 0.244 (c)), H2O (d), model CV curves with
various H3O+ concentrations when the mechanism III (Fig. 2) is assumed (e) and
CVs of 4 in H2O at =100, 300, 700, 900, 1100, 1300, 1500, 1700, 1900 mVs-1 (f).
Arrows indicate the direction of potential sweep.
It is convenient to present possible schemes of mechanisms as graphs. The points of the graphs
are molecules involved in the reaction mechanism (1-6), and the edges are one-electron transfer
reactions or protolytic equilibria (Fig. 4).
5

Figure 4. Topological representation of the
ECR mechanisms of 3, in
4
MeCN:H2O mixtures (yellow edges
are one-electron transfer reactions,
black edges are protolytic equilibria).
The ECR of 4 in MeCN (ESI Fig. S1) is an example of a classical EE-process (I in Fig. 4)
with two well-separated one-electron reversible peaks in the CV curve. Meanwhile, the ECR of 3
in MeCN can be adequately simulated18 only when mechanism II is used (Fig. 3a). In
accordance with the CV simulation data, the 3C peak corresponds to the ECR of the protonated
form of 3 RA.
The EPR spectra of 3, 4 RAs together with their protonated forms were observed under
Ep1С
electrochemical reduction of 3, 4 at potential
. Both spectra were characterized by broadened
lines in MeCN due to proton exchange between the oxygen atoms of the quinone moieties. This
exchange became fast in H2O leading to line narrowing (ESI Fig. S2).
The increase of the water content in the MeCN:H2O mixture results in a decrease of the
heterogeneous electron transfer constants at the second electrochemical stage, whereas the first
stage remained reversible over the whole range of m (ESI Table S1). In accordance with the CV
simulation data, the ECR mechanism changes as follows: II (m=0)III(m>0) for 3 (Fig. 3a-d)
and I (m=0)II (m~0.1)III(m>0.1) for 4 (ESI Fig. S1). To understand why only one
reversible one-electron peak was observed in the CVs of 3, 4 in H2O (Fig. 3d, ESI Fig. S1), we
considered a model simulation of the CV of 3 at different H3O+ concentrations (Fig. 3e) using
mechanism III (Fig. 4). The decrease in the H3O+ concentration from 0.25 mM to 0.001 mM (pH
of the aqueous solution was 6.2 under experimental conditions) resulted in a decrease of the
protonated 3 RA concentration (particle 4 in graph III). In turn, this was accompanied by the
Ip2C
decrease of
, corresponding to one-electron transfer to particle 4. As a result, the second
electrochemical stage became not observable. The corresponding CV in this case has a one-
electron reversible diffusion-controlled peak, which was dependent on the potential sweep rate as
0.5 (Fig. 3f , ESI Fig. S1a, example for 4). The situation is reversed if the ECR mechanism,
Ip2C
including DA formation is accepted. In this case, the
decrease was not observed when the
6

H3O+ concentration is decreased. This proves that only mechanism III is correct for an adequate
description of the ECR of 3, 4 in H2O. Note, that the ECR of 3, 4 in 50 mM phosphate buffer at
pH 7.2 using an Au disk electrode gave the same results. When the Au electrode was used, the
peak potentials of the ECR of 3, 4 were equal to 0.34 and +0.10 V, respectively.
In accordance with literature data,8,14 quinoid substances in general undergo two-electron
ECR in aqueous buffers. Contrary to this, we showed that the ECR of 3, 4 in H2O at pH 6.2-7.2
was in fact a one-electron reversible process. The transfer of the second electron was not
observed due to the kinetic reasons. This meant that the corresponding acids 1, 2 could be used
as redox-active modifiers of oligonucleotides without limitation in potential sweep, because no
irreversible processes accompanying the loss of the label electrochemical activity are observed.
To test acids 1, 2 as redox-active labels in the model oligonucleotide structures, the
modified oligonucleotides were prepared by modification of the corresponding amino-terminated
27-base oligonucleotide via N-succinimide derivatives of 1, 2:
NQ(BQ)S(CH2)2C(O)NH(CH2)65’-TTGGCAGAAGCTATGAAACGATATGGG-3’
(for oligonucleotide synthesis and modification see ESI). The base sequence of the model
oligonucleotide was identical to the sequence of the quality control oligo probe (QC) which was
19
used in the microarray analysis.
Labeled QC was purified by reprecipitation from ethanol
followed by gel-filtration using Sephadex G-25 gel column chromatography. The final
concentration of the labeled QC was adjusted to 100 M.
A specially designed thin-layer electrochemical cell with three working electrodes
equipped with a manual switch between the working electrodes was used in the experiments. An
Ag disk electrode was used as a reference electrode. A standard microscope glass with 200 Å Au
sputtering served as an auxiliary electrode (ESI Fig. S3a,b). The working electrode surface was
20,21
prepared in accordance with a described previously procedure
(ESI). After purification of
the working electrode surface, the electrochemical cell was dried in an Ar flow. Then, 5 l of a
100 M solution of BQ- and NQ-labeled QC were dropped onto the surface of the working
electrodes (electrode 1 and 2, respectively, ESI Fig. S3b) followed by drying overnight in air. A
solution of unmodified QC was dropped onto electrode 3 for comparison. To remove any
unadsorbed oligonucleotide, the cell was washed twice for 3 min in 50 mM phosphate buffer
(pH=7.2). Then, 1.2 ml of the buffer was added into the cell electrode cavity and covered by the
auxiliary electrode (ESI Fig. S3c). CV were sequentially measured at each electrode using the
manual switch between the electrodes. CV are shown in the overlay mode (Fig. 5) and contain
redox peaks corresponding to the electrochemical responses from the QC labeled with BQ
7

(electrode 1) and NQ (electrode 2), whose potentials were sufficiently different to identify the
type of the label. No electrochemical signal from the unlabeled QC was detected.
Figure 5. CV scans of the working electrodes of the thin-
layer electrochemical cell with adsorbed QC
oligonucleotides labeled with NQ (1 – electrode
1), BQ (2- electrode 2) and without a label (3 –
electrode 3). (50 mM phosphate buffer, the
initial delay in potential sweep was 5 s, =0.1
Vs1, arrows indicate the beginning and
direction of the potential sweep).
The sensor electrodes modified with the labeled oligonucleotides were quite stable and
displayed electrochemical activity for at least two weeks when stored under an inert atmosphere.
The redox signal from both labeled oligonucleotides could be observed over 80-90 potential
sweep cycles.
Thus, we report the synthesis and electrochemical properties of benzoquinone and
naphthoquinone based redox-active modifiers of oligonucleotides, which make electrochemical
detection of oligonucleotides on the surface of Au electrodes possible. The mechanism study of
the label electrochemical reduction demonstrated the one-electron and reversible nature of the
first step of their ECR in H2O and phosphate buffer with pH=7.2. Other electrochemical
processes resulting in the loss of the labels electrochemical activity were not observed under the
studied experimental conditions. Both labels were stable in the oligonucleotide structure and
gave quite steady electrochemical signals after adsorption of the labeled oligonucleotides onto
Au electrodes. The elaborated labels possess well-distinguished ECR potentials, the values of
which fall into “free” potential area between the potentials of anthraquinone and ferrocene based
redox-active labels4 and can be used in electrochemical genosensing technologies, for which the
simultaneous use of a number of labels with different redox potentials is required.
8

Acknowledgments
We are grateful to the Russian Federation Ministry of Education and Science for support
within the Project of Joint Laboratories of the Siberian Branch of the Russian Academy of
Sciences and National Research Universities.
9

References
1. Tosar, J. P.; Branas, G.; Laiz, J. Biosens. Bioelectron. 2010, 26, 1205.
2. Caruso, F.; Rodda, E.; Furlong D. N. Anal. Chem. 1997, 69, 2043.
3 Anne, A.; Bouchardon, A.; Moiroux, J. J. Am. Chem. Soc. 2003, 125, 1112.
4 Di Giusto, D. A.; Wlassoff, W. A.; Giesebrecht, S.; Gooding, J. J.; King G. C. J. Am. Chem.
Soc. 2004, 126, 4120.
5 Gao, Z.; Yu, Y. H. Biosens. Bioelectron. 2007, 22 (6), 933.
6 Tansil, N. C.; Xie, F.; Xie, H.; Gao, Z. Chem. Commun. 2005, 8, 1064.
7 Shena M. Microarray Analysis. Ed.; John Whiley  Sons, Inc.: Hoboken, New Jersey, 2003,
pp 1-664.
8 Guin, P. S.; Das, S.; Mandal P.C. Intern. J. Electrochem. 2011, 2011, 1-22.
9 Soriaga, M. P.; Wilson, P. H.; Hubbard, A. T.; Benton, C. S. J. Electroanal. Chem. 1982, 142,
317.
10 He, P.; Crooks, R. M.; Faulkner, L. R. J. Phys. Chem. 1990, 94, 1135.
11 Ji, X.; Banks, C. E.; Silvester, D. S.; Wain, A. J.; Compton R. G. J. Phys. Chem. C 2007, 111,
1496.
12 Zhang, L.; Li, T.; Yu, P.; Ohsaka, T.; Mao, L.; Electrochem. Comm. 2013, 26, 89.
13 Piro, B.; Reisberg, S.; Anquetin, G.; Duc, H.-T.; Pham, M.-C. Biosensors. 2013, 3, 58.
14 Thomson, P. I. T.; Camus, V. L.; Hu, Y.; Campbell, C. J. Anal. Chem. 2015, 87(9), 4719.
15 Jiang, J.; Auchinvole, C.; Fisher, K.; Campbell, C. J. Nanoscale. 2014, 6, 12104.
16 Details of CV experiments, the electrochemical and kinetic parameters for CV simulations are
described in Electronic Supplementary Information (ESI).
17 i is a number of peaks, symbols “A” or “C” indicate the anode or cathode branch of the CV
curve. The designation of potentials is the same.
18 CV simulations were performed using electrochemical simulation package (ESP, v.2.4)
designed by Carlo Nervi, Università di Torino, Italy. A simplex algorithm was used for
multiparametric optimization implemented in the package.
19 Ryabinin, V. A.; Shundrin, L. A.; Kostina, E. V.; Laassri, M.; Chizhikov V.; Shchelkunov, S.
N.; Chumakov, K.; Sinyakov, A. N. Journal of Medical Virology, 2006, 78, 1325.
20 Pang D.-W.; Abruña H. D. Anal. Chem., 1998, 70, 3162.
21 Pang D.-W.; Abruña H. D. Anal. Chem., 2000, 72, 4700.
10