Electrochemical investigations of M-DNA self-assembled monolayers on gold electrodes

Article abstract of DOI:10.1021/jp026792w

Monolayers of double-stranded DNA on polycrystalline gold electrodes were prepared by chemisorbing SS-5a?2-GTCACGATGGCCCAGTAGTT-3a?2 (1) hybridized with its complementary strand 5a?2-AACTACTGGGCCATCGTGAC-3a?2 (2) onto gold through a disulfide linkage, giving a monolayer with hydroxylalkyl chains adjacent to the double-stranded DNA (ds-DNA). The system forms a stable and well-behaved ds-DNA monolayer on gold, which reduces the signal transduction between a solution electrophore and the gold surface. Upon addition of Zn²⁺ to the solution at pH 8.6, Zn²⁺ associates with the ds-DNA and form M-DNA. In contrast to ds-DNA, the signal transduction is significantly enhanced. The heterogeneous electron-transfer rate (k_ET) between the ferricyanide electrophore and the surface was determined for bare gold, ds-DNA, and M-DNA monolayers. k_ET for electron transfer from the solution electrophore to the gold surface through the M-DNA monolayer is 1.2 ?? 10^-4 ?± 0.2 ?? 10^-4 cm s^-1, which is smaller than for bare gold under the same conditions (k_ET = 9.1 ?? 10^-4 ?± 0.2 ?? 10^-4 cm s^-1). For simple ds-DNA, k_ET was too slow to be measured.

Full text of DOI:10.1021/jp026792w

J. Phys. Chem. B 2003, 107, 2291-2296
2291
Electrochemical Investigations of M-DNA Self-Assembled Monolayers on Gold Electrodes
Chen-Zhong Li,†,‡ Yi-Tao Long,†,‡ Heinz-Bernhard Kraatz,*,† and Jeremy S. Lee*,‡
Department of Chemistry, UniVersity of Saskatchewan, 110 Science Place,
Saskatoon, SK, Canada S7N 5C9, and Department of Biochemistry, UniVersity of Saskatchewan,
107 Wiggins Road, Saskatoon, SK, Canada S7N 5E5
ReceiVed: August 19, 2002; In Final Form: December 5, 2002
Monolayers of double-stranded DNA on polycrystalline gold electrodes were prepared by chemisorbing
SS-5′-GTCACGATGGCCCAGTAGTT-3′ (1) hybridized with its complementary strand 5′-AACTACTGGGC-
CATCGTGAC-3′ (2) onto gold through a disulfide linkage, giving a monolayer with hydroxylalkyl chains
adjacent to the double-stranded DNA (ds-DNA). The system forms a stable and well-behaved ds-DNA
monolayer on gold, which reduces the signal transduction between a solution electrophore and the gold surface.
Upon addition of Zn2+ to the solution at pH 8.6, Zn2+ associates with the ds-DNA and form M-DNA. In
contrast to ds-DNA, the signal transduction is significantly enhanced. The heterogeneous electron-transfer
rate (kET) between the ferricyanide electrophore and the surface was determined for bare gold, ds-DNA, and
M-DNA monolayers. kET for electron transfer from the solution electrophore to the gold surface through the
M-DNA monolayer is 1.2 × 10-4 ( 0.2 × 10-4 cm s-1, which is smaller than for bare gold under the same
conditions (kET ) 9.1 × 10-4 ( 0.2 × 10-4 cm s-1 ). For simple ds-DNA, kET was too slow to be measured.
Introduction
backbone phosphates as do other cations such as Mg2+ 10
.
M-DNA can be converted back to ordinary ds-DNA by the
addition of ethylenediaminetetraacetic acid (EDTA) to remove
the metal ions.
The electron-transfer properties of various biomolecules such
as proteins, peptides, and DNA have attracted a significant
amount of attention over the past few years, largely driven by
the desire to develop devices able to detect mutations or to
prepare nanoscale biological devices such as molecular wires.1-4
In particular, the conductivity of double-stranded DNA (ds-
DNA) has been studied in detail. The results were often
contradictoryssome reports have indicated that ds-DNA is a
conductor and mediates electron transfer via the π-stacked base
pairs whereas other reports suggested that DNA is an insulator
under certain conditions.5,6
Electrochemical techniques have proven invaluable for prob-
ing the electron-transfer properties of immobilized biomolecules
and have yielded information regarding the electron-transfer
kinetics and potential pathways.11,12 The immobilization of DNA
on an electronically conductive substrate has been described in
the literature.13-16 In particular, DNA linked to a gold surface
through a Au-S linkage will often form very stable monolayer
assemblies with a well-defined 2D structure.17-21
Stable monolayers on gold have been formed from alkylthiols
and disulfides, giving what is believed to be a stable alkyl-
thiolate-gold linkage.22,23 However, the possibility of a disulfide
bond coordinating to the surface cannot be excluded.24a
Here we provide a detailed account of our study of the
electrochemical properties of M-DNA supported on gold
surfaces, including the examination of the electron-transfer
kinetics. Our kinetic studies demonstrate that M-DNA allows
electron transfer more facile than ds-DNA.
Recent reports of the direct measurements of electrical
transport through a 10.4-nm ds-DNA oligomer7 demonstrate that
ds-DNA is a semiconductor with a wide band gap. Although
native DNA is an inadequate conductor, it can be improved by
coating individual DNA fibers with metallic silver.8 A novel
method for improving the conductivity of DNA was reported
recently and involves close association or even incorporating
divalent metal ions such as Zn2+, Ni2+, and Co2+ into the helix
at pH > 8.5 to form metalated DNA (M-DNA).9,10 Other
divalent metals such as Mg2+ and Ca2+ do not form M-DNA.
In photophysical studies using ds-DNA where one strand is
labeled with a fluorescein fluorophore and the complementary
strand carries a rhodamine quencher, the fluorescein fluorescence
is quenched only under conditions where M-DNA forms.
Preliminary evidence demonstrates that M-DNA is an efficient
conductor of electrons over distances as long as 500 base pairs
and possibly as long as several micrometers.9 It is proposed
that in M-DNA the Zn2+ cations replace the imino protons of
every thymine and guanine in addition to saturating the
Experimental Section
General. [Ru(NH3)6]Cl3, [Ru(NH3)6]Cl4, K3[Fe(CN)6], and
K4[Fe(CN)6] were purchased from Aldrich and used without
further purification. Zn(ClO4)2, Mg(ClO4)2, and tris-ClO4 were
purchased from Fluka Co. Deionized water (18 MΩ cm
resistivity) from a Millipore Milli-Q system was used throughout
this work.
Oligonucleotide Synthesis. HO-(CH2)6SS(CH2)6OPO3H
was used to modify the oligonucleotides at the 5′ terminus. The
5′-modified oligonucleotides and underivatized complements
used in the study were synthesized by standard phosphoramidite
chemistry25 using a fully automated DNA synthesizer at the
Plant Biotechnology Institute (PBI-NRC, Saskatoon). The
oligonucleotides were purified by two-step reversed-phase
* Corresponding authors. E-mail: kraatz@skyway.usask.ca. Fax: +306-
966-4730.
† Department of Chemistry, University of Saskatchewan.
‡ Department of Biochemistry, University of Saskatchewan.
10.1021/jp026792w CCC: $25.00 © 2003 American Chemical Society
Published on Web 02/19/2003

2292 J. Phys. Chem. B, Vol. 107, No. 10, 2003
Li et al.
HPLC and then characterized by MALDI-TOF. The sequences
and nomenclature used are as follows:
SS-5′- GTCACGATGGCCCAGTAGTT -3′
5′- AACTACTGGGCCATCGTGAC -3′
1
2
Note: SS 5′ refers to HO3PO-(CH2)6-SS-(CH2)6-OH
Electrode Characterization and Pretreatment. Gold disk
electrodes (Bioanalytical Systems Inc., 1.6-mm diameter, ca.
0.02 cm2 geometrical area, and roughness coefficients of 1.2-
1.4) were used for the electrochemical measurement. Before
modification, the electrode surface was cleaned by electro-
chemical sweeping in 0.1 M H2SO4 from 0 to 1.4 V, rinsed
with water, and then ultrasonicated for 5 min in fresh piranha
solution (30% H2O2, 70% H2SO4). Warning: Piranha solution
reacts Violently with organic solVents. The electrode was then
sonicated by freshly distilled and degassed ethanol and then
rinsed with Milli-Q water. A cyclic voltammagram recorded in
0.1 M H2SO4 (scan rate 100 mV s-1) was used to determine
the active area of the electrode surface. The real electrode
surface area and roughness factors were obtained by integrating
the gold oxide reduction peak.26,27 The gold oxide was reduced
in situ to a clean gold surface by immersion in degassed freshly
distilled ethanol.
Figure 1. Schematic representation of (A) the model of di-6-alkyl
disulfide-modified double-stranded DNA giving rise to (B) the proposed
model of the mixed SAMs on a gold surface.
Figure 2. Comparison of the capacitive background currents on bare
(s) and ds-DNA-modified gold electrode (- - -) in 20 mM tris-ClO4
buffer solution (pH 8.6) at a sweep rate of 100 mV s-1
.
Preparation of DNA SAMs. ds-DNA was hybridized in 20
mM tris-ClO4 (pH 8.6) from disulfide-modified DNA 1 and its
complementary strand 2 at room temperature for 10 h. Hybrid-
ization was confirmed by the ethidium fluorescence assay.28 The
cleaned gold electrodes (vide supra) were immersed in an
aqueous solution of tris-ClO4 (100 mM), NaClO4 (100 mM),
and the hybridized ds-DNA conjugate at pH 8.6. The concentra-
tion of the DNA was 0.2 mM. After 3 days, the electrodes were
repeatedly rinsed with tris-ClO4 (50 mM, pH 8.6) for 5 min.
The resulting electrode was then soaked again in the same ds-
DNA solution overnight and rinsed with buffer three times. The
derivatized electrode surface coverage was quantified to be over
90% by the underpotential deposition (UPD) of Cu.29-31 The
ds-DNA monolayer was converted into the M-DNA monolayer
by exposing the mononlayer to a solution of 0.3 mM Zn(ClO4)2
in 20 mM tris-ClO4 buffer (pH 8.6) for at least 2 h.
Electrochemical Measurements. All electrochemical experi-
ments were performed in a conventional three-electrode cell
consisting of the DNA-SAM-modified Au electrode as the
working electrode, a Ag/AgCl/3 M NaCl reference electrode
(BAS), and a platinum wire auxiliary electrode. All potentials
are reported with the respect to the Ag/AgCl/3 M NaCl reference
electrode. A BAS 50-W potentiostat was used for the electro-
chemical measurements. Typical CV experiments employed a
5 mM solution of [Ru(NH3)6]Cl4/[Ru(NH3)6]Cl3 or a 2.5 mM
solution of K3[Fe(CN)6]/K4[Fe(CN)6] in 20 mM tris-ClO4
solution at pH 8.6 with or without divalent M2+ at room
temperature (20 ( 2 °C). The open-circuit or rest potential of
the system was measured prior to all electrochemical experi-
ments to prevent sudden potential-related changes in the SAM.
All electrochemical experiments were started from the rest
potential. UPD experiments were carried out in 1 mM Cu(ClO4)2
in 0.1 M HClO4 at a scan rate of 10 mV s-1 starting at 500 mV
(vs Ag/AgCl) followed by cathodic scanning to 50 mV and an
anodic sweep to 600 mV. Simulation of the CV curves was
done using the DigiSim 21 simulation program (BAS).
CATCGTGAC-3′ (2) were synthesized by automated solid-
phase synthesis. Chemisorption of the ss-DNA-hydroxyalkyl
disulfide 1 or the ds-DNA prepared by the hybridization of 1
and 2 resulted in a mixed monolayer having DNA-thiolate and
hydroxyalkylthiolate-gold linkages with the DNA thiolate
adjacent to the hydroalkylthiolate residue, as indicated in Figure
1. Experimental evidence by others strongly suggests that S-S
bond cleavage occurs upon chemisorption.24b,c Nonspecifically
bound DNA was readily removed by washing with buffer.19
It is predicted that this arrangement should prevent the
interaction of DNA with the gold surface through the nitrogen
of the nucleotides.32 Furthermore, the disulfide adsorption should
result in a more uniform monolayer structure.24 Usually “dilute”
DNA monolayers are prepared by soaking a monolayer prepared
from DNA-thiol in a solution of alkylthiol solution.18,33,34 This
result is a two-component film, which often shows phase
separation and incomplete mixing of the two components on
the surface. Furthermore, this procedure leads to a displacement
of the bound DNA-thiolate by the alkylthiol.19,34 In the present
study, unlike previous studies, systems derived from the
asymmetric DNA conjugate 1 were used to modify the electrode
surface. In particular, the choice of the terminal hydroxyl
function of the alkyl appears to play an important role in
determining the monolayer stability and packing.35,36 The
terminal OH groups appear to be an essential element in the
production of stable monolayers. H-bonding interactions in-
volving HO-PO3 in monolayer systems containing cystamine
were reported before and lead to very stable monolayer
systems.37,38
At the DNA-modified electrodes, the capacitive background
currents (Figure 2) are smaller compared to those of a bare gold
electrode, as expected for the presence of a monolayer with a
low dielectric constant.
Figure 3 shows the results of a UPD study of Cu on bare
gold and a ds-DNA-modified gold electrode taken in 1.0 mM
Cu(ClO4)2/50 mM HClO4 aqueous solution at a scan rate of 10
mV s-1. As expected, the Cu UPD on the bare gold electrode
gives a broad deposition peak during the cathodic sweep at about
190 mV and a stripping wave on the anodic scan at 300 mV.30,39
Results and Discussion
The oligonucleotide SS-5′-GTCACGATGGCCCAGTAGTT-
3′ (1) and its complementary strand 5′-AACTACTGGGC-

M-DNA Self-Assembled Monolayers on Au Electrodes
J. Phys. Chem. B, Vol. 107, No. 10, 2003 2293
Figure 3. Under potential deposition of copper on bare (s) and ds-
DNA-modified gold electrodes (- - -); 1 mM Cu(ClO4)2 + 0.05 M
HClO4, scan rate was 10 mV s-1. The gold electrode area was 0.02
cm2.
In contrast to bare gold, the UPD of Cu on the ds-DNA-modified
gold electrode is strongly suppressed. The amount of charge
involved in the stripping wave is decreased by 95% compared
to that of the UPD on bare gold (15 µC cm-2 as compared to
275 µC cm-2 for bare gold). These results suggest that the
monolayer of ds-DNA prepared from 1 and 2 is well packed
and has a limited amount of exposed surface area that is directly
accessible to the electrolyte solution.
DNA self-assembled monolayers immobilized on gold elec-
trode surfaces provide an excellent means for studying electron-
transfer kinetics through DNA. The electron transfer through
the DNA monolayer from solution-based redox couples to the
electrode surface is good a measure of the conductivity of the
molecular assembly. Anionic [Fe(CN)6]3-/4- and cationic
[Ru(NH3)6]3+/4+ were chosen to study the electron transfer
between a gold surface and a solution-based redox probe. They
have been used extensively to study electrode kinetics and to
examine solid electrode surfaces.36,40 Both exhibit reversible or
quasi-reversible but well-behaved redox chemistry in aqueous
solution.
Figure 4. Cyclic voltammograms of bare and ds-DNA-modified
electrodes. (a) Bare gold electrode; (b) ds-DNA-coated gold electrode;
(c) ds-DNA-modified gold electrode soaked in 0.3 mM MgClO4 for
2.5 h; (d) ds-DNA-modified gold electrode soaked in a solution
containing 0.3 mM Zn(ClO4)2 for 2.5 h. The concentration of ferri-
cyanide/ferrocyanide was 2.5 mM in 20 mM tris-ClO4 buffer (pH 8.6).
All CVs were recorded at a sweep rate of 100 mV s-1
.
Figure 4 shows a series of four cyclic voltammagrams of 2.5
mM [Fe(CN)6]3-/4- solutions at pH 8.6 using a bare gold
working electrode (4a), a ds-DNA-modified gold electrode (4b),
a ds-DNA-modified electrode in the presence of 0.3 mM Mg2+
(4c), and a ds-DNA-modified electrode in the presence of 0.3
mM Zn2+ (4d). As shown in Figure 4a, the CV curve for bare
gold shows the reduction and oxidation peaks that are typical
for a diffusion-limited one-electron redox process. The CV of
ds-DNA on gold shown in Figure 4b exhibits a much reduced
signal because of the [Fe(CN)6]3-/4- redox couple (only 5%
compared to bare gold) in what can be described as an
irreversible system with a large peak-to-peak separation (∆Ep).
Thus, the electronic communication between the solution
electrophore and the gold surface is effectively suppressed,
indicating blocking of the gold surface by the ds-DNA. Electron
transfer between the solution electrophore and the gold surface
now has to proceed though the ds-DNA/hydroxyl-C6 monolayer.
This increase in distance between the gold surface and the
solution electrophore will result in a slower electron-transfer
rate, which in CV will result in an increase in ∆Ep. In addition,
the electrostatic repulsion between the ds-DNA molecules on
the surface and the anionic [Fe(CN)6]3-/4- has to be considered,
which does not allow for an effective penetration of the
monolayer by the negatively charged electrophore. Better
blocking was observed by Kelly and co-workers for a monolayer
of 5′HS-(CH2)6-p-15 base oligonucleotides, which appears to
be free of any measurable pinholes and provides complete
blocking.41 Similarly, a monolayer using 418-bp DNA com-
pletely blocks the [Fe(CN)6]3-/4- electrophore, and no signal
transduction was observed.19 In contrast, our monolayer consists
of a 20-bp ds-DNA-hydroxylalkyl disulfide.
Both residues are deposited on the surface, and the resulting
monolayer does not completely block the redox activity of the
[Fe(CN)6]3-/4- electrophore. We propose that the electron passes
through the hydroxyl “defect” sites rather than through the ds-
DNA, which may allow for closer proximity between the redox
probe and the surface. Because of the charge repulsion between
the phosphates and [Fe(CN)6]3-/4-, monolayer penetration by
the solution anion will have a low probability. As demonstrated
by the UPD experiments with Cu2+, some of the gold is
accessible to the electrolyte, and thus we cannot rule out some
contributions from these sites. However, the contribution should
be the same for M-DNA and native ds-DNA.
A comparison of a monolayer prepared by the method
outlined above to that of a method described by Steel33 shows
a notable difference in the blocking ability. Steel’s method,
briefly outlined, uses a mixed monolayer system (ss-DNA-SH
and C6-SH) employing only thiols, and the DNA-hybridization
step occurs after the monolayer formation is established. The
electrode surface blocking results clearly show that Steel’s
method produces a surface that can permit electron transfer to
the redox probe, whereas a monolayer produced using the
method described here is effectively blocked to electron transfer.
According to the Steel study, the surface density of ds-DNA
was about (1-10) × 10-12 mol cm-2 of a 25-bp DNA,
indicating low surface coverage.33 On the basis of the small
Faradaic current produced from the electrophore, the ds-DNA-
SS monolayer appears to be tightly packed, resulting in better
insulation of the Au surface electrode.
In contrast, the positively charged [Ru(NH3)6]3+/4+ is not
effectively blocked by the ds-DNA. Figure 5 shows a CV curve
of a ds-DNA-modified gold electrode and a bare gold electrode

2294 J. Phys. Chem. B, Vol. 107, No. 10, 2003
Li et al.
Figure 7. Anodic peak current as a function of incubation time for
the ds-DNA-modified electrode in (b) 2.5 mM ferricyanide/ferrocyanide
in tris-ClO4 buffer solution (20 mM, pH 8.6) with the addition of (9)
0.3 mM Zn(ClO4)2 or (O) 0.3 mM MgCl2. The error bars represent the
standard deviation on three independent determinations.
Figure 5. Cyclic voltammograms at bare (s) and ds-DNA-modified
electrodes (- - -) in 5.0 mM hexaammineruthenium(III/IV) chloride in
20 mM tris-ClO4 buffer (pH 8.6) at a sweep rate of 100 mV s-1
.
the signal intensity on bare gold, with a very large ∆Ep of 0.68
V indicating slow electron transfer between the solution and
the surface.
However, upon adding 0.3 mM Zn(ClO4)2 to the buffer
solution at pH 8.6 to form M-DNA, a large signal due to the
[Fe(CN)6]3-/4- electrophore, which has 80% of the integrated
signal intensity compared to that of bare gold (Figure 4d), is
observed. The CV curve exhibits a significantly increased signal
as compared to that of the blocked ds-DNA monolayer electrode.
The peak separation in the presence of Zn2+ is larger (∆Ep )
0.40 V) than that for bare gold (∆Ep ) 0.22 V). Thus, in the
presence of Zn2+, under favorable M-DNA formation conditions,
effective signal transduction takes place from the solution
electrophore to the gold electrode. In comparison, the addition
of Mg2+ does not allow for an effective signal transduction
(Figure 4c) under M-DNA conditions. As a final control, adding
Zn2+ at pH 7.0 does not give rise to increased electron transfer
because M-DNA is proposed not to form at this pH.
Figure 6. Cyclic voltammagrams of ds-DNA-modified gold electrodes
in blank buffer (20 mM tris-ClO4, pH 8.6) after incubation in buffered
solutions of ferricyanide/ferrocyanide (- - -) or hexaammineruthe-
nium(III/IV) chloride (s) and thorough rinsing with buffer at a sweep
rate of 100 mV s-1. A Faradaic current was observed only for
3+/4+
Ru(NH3)6
.
with the [Ru(NH3)6]3+/4+ electrophore in solution. Signal trans-
duction through the monolayer proceeds virtually unimpeded.
The peak separation ∆Ep is slightly increased from 0.17 to 0.20
V as compared to that of the bare gold electrode. These results
suggest that the [Ru(NH3)6]3+/4+ ion can interact to a significant
degree with the electrode surface. Presumably, there is a
favorable interaction between the negatively charged phosphate
backbone and the positively charged [Ru(NH3)6]3+/4+ ion,
allowing for effective penetration of the monolayer.
Figure 7 shows a time study on the effects of the addition of
Zn2+ and Mg2+ on the CV of the ds-DNA/hydroxyalky mono-
layer. Initially, the monolayer is blocking the electron-transfer
process effectively. After the addition of Zn2+ to the buffer
solution, the signal intensity of the CV of the [Fe(CN)6]3-/4-
electrophore in solution steadily increases up to a limit, and
the peak separation decreases. Figure 7 shows the time
dependence of the anodic peak current with increasing incuba-
tion time in the presence of Zn2+ or Mg2+. For Zn2+, the current
increases significantly, reaches a maximum after about 2 h, and
then remains at this maximum, corresponding to about 80% of
the signal intensity of bare gold. For Mg2+, the increase over
the same period is comparatively small. Thus, the electrochemi-
cal properties of the ds-DNA/hydroxyalkyl monolayer in the
presence of Mg2+ are not affected. The time-dependent increase
of the electrical response probably represents the dynamics of
the incorporation of Zn2+ into the ds-DNA monolayer. To a
first approximation, the M-DNA formation on the surface
follows pseudo-first-order kinetics with a pseudo-first-order rate
In the case of the anionic [Fe(CN)6]3-/4-, the monolayer is
more resistant to ion and electrolyte penetration. Even after
washing the [Ru(NH3)6]3+/4+-exposed ds-DNA/hydroxy-
alkyl monolayer with copious amounts of buffer solution, a
significant signal due to electroactive [Ru(NH3)6]3+/4+ is still
present in the monolayer. Figure 6 compares the CV curve of
[Ru(NH3)6]3+/4+-exposed and [Fe(CN)6]3-/4--exposed mono-
layers after extensive rinsing. Note that the [Ru(NH3)6]3+/4+
signal can still be observed after washing with buffer whereas
the signal due to [Fe(CN)6]3-/4- is completely absent after
washing. This is evidence of redox probe penetration or
constant of k ) 10 min-1
.
3-
electrostatic charge coupling along the PO4 backbone.
The anodic peak current exhibits a linear dependence that
increases with the square root of the scan rate in the range of
50 to 500 mV (Figure 8), suggesting that the electrochemical
reaction is primarily controlled by linear diffusion and not by
pinholes or defects.40
Thus, the Faradic current is due to electron transfer from the
[Fe(CN)6]3-/4- electrophore through the M-DNA helix to the
electrode surface. Figure 9 shows the CV curve of M-DNA in
the absence and presence of EDTA. It was shown in previous
The ds-DNA/hydroxyalkyl monolayer shown in Figure 4
effectively blocks the signal transduction of [Fe(CN)6]3-/4-. The
addition of 0.3 mM Mg2+ to the buffer at pH 8.6 changes the
signal transduction only slightly (Figure 4c). It is expected that
Mg2+ will coordinate to the phosphate backbone and thereby
reduce the electrostatic repulsion, allowing for better ion
penetration of the monolayer by [Fe(CN)6]3-/4-. However,
Figure 4d shows that the signal represents only about 10% of

M-DNA Self-Assembled Monolayers on Au Electrodes
J. Phys. Chem. B, Vol. 107, No. 10, 2003 2295
Figure 10. Cyclic voltammograms (s) of 2.5 mM ferricyanide/
ferrocyanide in 20 mM tris-ClO4 buffer solution (pH 8.6) at (A) a bare
gold electrode and (B) a M-DNA-modified gold electrode at 25 °C
Figure 8. Plot of anodic peak current (Ipa) vs the square root of the
sweep rate. The cyclic voltammograms were obtained using a gold
electrode modified by a M-DNA SAM in 20 mM tris-HClO4 buffer
(pH 8.6) in the presence of 5 mM ferricyanide/ferrocyanide.
together with simulated data (O). Scan rate used was 100 mV s-1
.
Digital simulations were made using the kET, E0, and R (transfer
coefficient) values of (A) 9.1 × 10-4, 0.21 vs Ag/AgCl/3 M NaCl,
and 0.5 and (B) 1.23 × 10-4 cm s-1, 0.19 V vs Ag/AgCl/3 M NaCl,
and 0.4. The diffusion coefficients of Dox and Dred used for A were 8
× 10-6 cm2 s-1 for ferricyanide and ferrocyanide. Both CVs have the
background current subtracted.
Figure 9. Cyclic voltammograms at M-DNA-modified gold electrodes
(s) or after treatment with 1 mM EDTA (- - -) for 90 min measured
in 2.5 mM ferricyanide/ferrocyanide in 20 mM tris-ClO4 buffer (pH
8.6) at a sweep rate of 100 mV s-1
.
studies that EDTA will complex Zn2+ and convert back to native
ds-DNA.10 Upon adding EDTA to the buffer containing the
M-DNA monolayer, the signal intensity is significantly de-
creased.
In addition, the peak separation increases, indicating a loss
of signal transduction from the [Fe(CN)6]3-/4- electrophore
through the monolayer after the removal of the Zn2+ ions. The
resulting CV curve exhibits the same features as that of the
ds-DNA monolayer. The signal intensity is reduced compared
to that of the M-DNA monolayer showing a large ∆Ep (0.60
V). Thus, the M-DNA can be converted back to native ds-DNA
on the gold surface.
Figure 11. Chronocoulometric transients at -350 mV of 5 mM
ferricyanide in 20 mM tris-ClO4 buffer (pH 8.6) at (a) bare gold, (b) a
ds-DNA-modified gold electrode in the presence of 0.3 mM Zn(ClO4)2
in the solution, (c) a ds-DNA-modified gold electrode in the presence
of 0.3 mM MgClO4 in the solution, and (d) a ds-DNA-modified
electrode.
systems under identical experiment conditions. Recently, several
authors have used chronocoulometry to characterize mismatched
DNA43 or to determine the surface density of immobilized
DNA33,44 quantitatively on the basis of the charge transport to
the surface. A comparison of the charge passed at the various
modified electrodes is shown in Figure 11. The initial potential
started at 200 mV versus Ag/AgCl, where no electrolysis of
ferricyanide occurs. During a single step of 12 s to the potential
of -350 mV at which essentially all of the ferricyanide is
reduced to ferrocyanide, the amount of charge passed on the
M-DNA-modified electrode is significantly larger than that on
the ds-DNA-modified electrode. These experiments confirm the
results of CV in that blocking electron transfer of from
[Fe(CN)6]3-/4- through M-DNA is much more effective than
through conventional ds-DNA.
We evaluated the heterogeneous electron-transfer kinetics for
bare gold and the M-DNA modified electrode by comparing
the experimental curve with a simulated CV curve. Figure 10
shows the CV curves for a bare Au electrode (Figure 10a) and
the M-DNA SAMs’ modified Au electrode after ds-DNA SAM
background subtraction (Figure 10b) together with the simulated
CV plot. The diffusion coefficients for the [Fe(CN)6]3-/4- redox
couple are Dox ) Dred ) 8 × 10-6 cm2 s-1 42
.
In 20 mM tris-ClO4 buffer at a scan rate of 100 mV s-1, a
∆Ep of 220 mV was measured for the [Fe(CN)6]3-/4- redox
couple at the bare electrode, and a ∆Ep of 400 mV, for the
M-DNA monolayer-modified Au electrode. Under these condi-
In conclusion, we have confirmed by CV and chrono-
coulometry that blocking electron transfer on a ds-DNA-
modified electrode can be overcome by converting to M-DNA,
through which electron transfer is much faster. Therefore,
M-DNA may find widespread applications in nanoelectronics
or biosensing since a direct electrical readout of hybridization
or DNA binding is now possible.
tions, the electron-transfer rate constant kET was 9.1 × 10-4
(
(
0.2 × 10-4 cm s-1 for the bare Au electrode and 1.2 × 10-4
0.2 × 10-4 cm s-1 for the M-DNA/hydroxylalkyl-modified Au
electrode. In the case of the ds-DNA/hydroxylalkyl-modified
Au electrode, the electron-transfer rate constant was too small
to be measured, thus kET must be orders of magnitude smaller.
To compare and verify the results obtained from the analysis
of the sampled voltammetry, a charge-integration technique,
chronocoulometry, was employed on the same set of redox
Acknowledgment. We thank CIHR, NSERC, and UMDI
for financial support. H-B.K. holds the Canada Research Chair
in Biomaterials, and J.S.L. is supported by a Senior Investigators

2296 J. Phys. Chem. B, Vol. 107, No. 10, 2003
Li et al.
(23) Bain, C. D.; Biebuyck, H. A.; Whitesides, G. M. Langmuir 1989,
5, 723-727.
Award from the Regional Partnership Program of CIHR. In
addition, we thank Dr. T. Sutherland for helpful discussions.
(24) (a) Schonherr, H.; Ringsdorf, H.; Jaschke, M.; But, H.-J.; Bamberg,
E.; Allinson, H.; Evans, S. D. Langmuir 1996, 12, 3898. (b) Biebuyck, H.
A.; Whitesides, G. M. Langmuir 1993, 9, 1766-1770. (c) Ishida, T.;
Yamamoto, S.; Mizutani, W.; Motomatsu, M.; Tokumoto, H.; Hokari, H.;
Azehara, H.; Fujihara, M. Langmuir 1997, 13, 3261-3265.
(25) Wincoff, F.; Direnzo, A.; Shaffer, C.; Sweedler, D.; Gonzalez, C.;
Scarinje, S.; Usman, N. Nucleic Acids Res. 1995, 25, 2677-2684.
(26) Hallmark, V. M.; Chiang, S.; Rabolt, J. F.; Swalen, J. D. Phys.
ReV. Lett. 1987, 59, 2879.
(27) Woods, R.; Bard, A. J.; Dekker, M. 1 ed.; New York, 1980; Chapter
9.
(28) Pecq, J. B. L.; Paoletti, C. Anal. Biochem. 1966, 17, 100-107.
(29) Eliadis, E. D.; Nuzzo, R. G.; Gewirth, A. A.; Alkire, R. C. J.
Electrochem. Soc. 1997, 144, 96-105.
(30) Sun, L.; Crooks, R. M. J. Electrochem. Soc. 1991, 138, L23-L25.
(31) Nishizawa, M.; Sunagawa, T.; Yoneyama, H. Langmuir 1997, 13,
5215-5217.
(32) Satjapipat, M.; Sanedrin, R.; Zhou, F. Langmuir 2001, 17, 7637-
7644.
(33) Steel, A. B.; Herne, T. M.; Tarlov, M. J. Anal. Chem. 1998, 70,
4670-4677.
(34) Steel, A. B.; Levicky, R. L.; Herne, T. M.; Tarlov, M. J. Biophys.
J. 2000, 79, 975-981.
(35) Zamborini, F. P.; Crooks, R. M. Langmuir 1998, 14, 3279-3286.
(36) Takehara, K.; Takemura, H.; Ide, Y. Electrochim. Acta. 1994, 39,
817-822.
(37) Cha, X.; Ariga, K.; Kunitake, T. Bull. Chem. Soc. Jpn. 1996, 69,
163-168.
(38) Clegg, R. S.; Hutchinson, J. E. Langmuir 1996, 12, 5239-5243.
(39) Sondag-Huethorst, J. A. M.; Fokkink, L. G. J. Langmuir 1995, 11,
4823-4831.
(40) Chailapakul, O.; Crooks, R. M. Langmuir 1993, 9, 884-888.
(41) Kelley, S. O.; Barton, J. K.; Jackson, N. M.; Hill, M. G.
Bioconjugate Chem. 1997, 8, 31-37.
(42) Winkler, K. J. Electroanal. Chem. 1995, 388, 151-159.
(43) Kelley, S. O.; Boon, E. M.; Barton, J. K.; Jackson, N. M.; Hill, M.
G. Nucleic Acids Res. 1999, 27, 4830-4837.
(44) Patolsky, F.; Lichtenstein, A.; Willner, I. J. Am. Chem. Soc. 2001,
123, 5194-5205.
References and Notes
(1) Elghanian, R.; Storhoff, J. J.; Mucic, R. C.; Letsinger, R. L.; Mirkin,
C. A. Science (Washington, D.C.) 1997, 277, 1078-1081.
(2) Bobinson, B. H.; Seeman, N. C. Protein Eng. 1987, 1, 295-300.
(3) Gooding, J. J.; Hibbert, D. B.; Yang, W. Sensors 2001, 1, 75-90.
(4) Niemeyer, C. M. Appl. Phys. 1999, 68, 119-124.
(5) Wilson, E. K. Chem. Eng. News 1997, 24, 33-39.
(6) Murphy, C. J.; Arkin, M. R.; Jenkins, Y.; Ghatlia, N. D.; Bossmann,
S. H.; Turro, N. J.; Barton, J. K. Science (Washington, D.C.) 1993, 262,
1025-1029.
(7) Porath, D.; Bezryadin, A.; de Vries, S.; Dekker, C. Nature (London)
2000, 635-638.
(8) Braun, E.; Eichen, Y.; Slvan, U.; Ben-Yoseph, G. Nature (London)
1998, 391, 775-778.
(9) Aich, P.; Labiuk, S. L.; Tari, L. W.; Delbaera, L. J. T.; Roesler,
W. J.; Falk, K. J.; Steer, R. P.; Lee, J. S. J. Mol. Biol. 1999, 294, 477-
485.
(10) Lee, J. S.; Latimer, L. J. P.; Reid, R. S. Biochem. Cell Biol. 1993,
162-168.
(11) Mikkelsen S. R. In Encyclopedia of Electrochemistry; Bard, A. J.,
Stratmann, M., Eds.; Wiley-VCH: Weinheim, Germany, 2002; Chapter 11.
(12) Galka, M. M.; Kraatz, H. B. ChemPhysChem 2002, 3, 356-359.
(13) Huang, E.; Zhou, F.; Deng, L. Langmuir 2000, 16, 3272-3280.
(14) Thorp, H. H. Trends Biotechnol. 1998, 16, 117-121.
(15) Watterson, J. H.; Piunno, P. A. E.; Wust, C. C.; Krull, U. J. Anal.
Chim. Acta 1999, 395, 273-284.
(16) Palecek, E. Bioelectrochem. Bioenerg. 1986, 15, 275-295.
(17) Kelley, S. O.; Barton, J. K.; Jackson, N. M.; Mcpherson, L. D.;
Potter, A. B.; Spain, E. M.; Allen, M. J.; Hill, M. G. Langmuir 1998, 14,
6781-6784.
(18) Herne, T. M.; Tarlov, M. J. J. Am. Chem. Soc. 1997, 119, 8916-
8920.
(19) Yang, M.; Yau, H. C. M.; Chan, H. L. Langmuir 1998, 14, 6121-
6129.
(20) Okahata, Y.; Matsunobu, Y.; Kunihara, I.; Masayuki, M.; Mura-
kami, A.; Makino, K. J. Am. Chem. Soc. 1992, 114, 8299-8300.
(21) Tarlov, M. J.; Newman, J. G. Langmuir 1992, 8, 1398-1405.
(22) Nuzzo, R. G.; Allara, D. L. J. Am. Chem. Soc. 1983, 105, 4481-
4483.