Original ArticleDetection analysis limit of nonlinear characteristics of DNA sensors nanoparticles a Institute of Scientific Research and Applications, Hanoi Pedagogical University 2, 32
Trang 1Original Article
Detection analysis limit of nonlinear characteristics of DNA sensors
nanoparticles
a Institute of Scientific Research and Applications, Hanoi Pedagogical University 2, 32 Nguyen Van Linh, Phuc Yen, Vinh Phuc 280000, Viet Nam
b International Training Institute for Materials Science, Hanoi University of Science and Technology, 1 Dai Co Viet, Hanoi 100000, Viet Nam
c Nano and Energy Center, VNU University of Science, Vietnam National University, Hanoi, 334 Nguyen Trai, Thanh Xuan, Hanoi 100000, Viet Nam
a r t i c l e i n f o
Article history:
Received 19 January 2018
Received in revised form
29 March 2018
Accepted 1 April 2018
Available online 7 April 2018
Keywords:
DNA sensor
Surface modification
Gold nanoparticles
Polypyrrole nanowires
LOD
Differential voltage
EIS
a b s t r a c t
Surface modification of interdigitated DNA sensors by polypyrrole nanowires and gold nanoparticles has been analyzed systematically Polypyrrole nanowires with diameter of 200 nm and length of 5mm were electrochemically synthesized on the gold surface of interdigitated electrodes and subsequently deco-rated with 20 nm gold nanoparticles Electrochemical impedance spectroscopy and differential voltage measurements were conducted to detect DNA concentrations We have observed a logarithmic depen-dence of analytical signals on the DNA concentration A formula for estimating the limit of detection has been derived Instead of using a conventional method in which a blank measurement is performed to record the response of the sensor in a solution containing non-complementary DNA molecules, causing
an errornous estimation of detection limit, we have proposed a novel approach for the calculation of the limit of detection Limits of detection of 60.0 fM for the differential voltage method and of 84.5 fM for the electrochemical impedance spectroscopy method were calculated after taking into account all possible errors These are the lowest values for the DNA sensors reported so far The presence of the gold nanoparticles increases the effective electrode area, leading to an overall improvement of the detection limit From the perspective of the detection limit, the differential voltage method is considered more advantageous as compared to the electrochemical impedance spectroscopy one
© 2018 The Authors Publishing services by Elsevier B.V on behalf of Vietnam National University, Hanoi This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/)
1 Introduction
DNA sensors are promising analytical devices, which have
attracted growing attention from the academic community, spin-off
companies as well as biomedical manufacturers due to their diverse
applications in food processing[1e3], mechanical and biomedical
engineering[4e9]and healthcare[10e14] The working mechanism
of a DNA sensor is based on the hybridization between the probe and
the target DNA, resulting in changes in the physical properties at the
interface of the sensor These changes can be detected by different
techniques such as optical [13e15], mechanical [3] or
electrochemical/electrical[2,16,17]and differential voltage[17,18] The limit of detection (LOD) is often used to evaluate the quality of this type of sensor By definition, LOD is the lowest analyte con-centration that can be distinguished from the absence of the analyte with a confidence limit The electrochemical method is versatile, simple but its LOD is typically higher than analytic requirements This disadvantage can be improved by modifying the electrode surface according to the specific applications[19] Two common surface modifications are adjusting the affinity of the surface to biological entities and increasing the surface area of an electrode Metallic nanoparticles, typically gold nanoparticles (GNPs), are preferably used to adjust the surface affinity of biosensors due to the high capability for surface functionalization with thiol groups con-tained in many organic molecules[20] In addition, at the nanoscale, the dynamic balance between Au0¼ Auþþ eon the particle surface
provides a source of charged entities
* Corresponding author.
E-mail addresses: haopv@hpu2.edu.vn (P Van Hao), xuan@itims.edu.vn
(C.T Xuan), thanh@itims.edu.vn (P.D Thanh), thuatnt@vnu.edu.vn (N.-T Thuat),
nhhai@vnu.edu.vn (N.H Hai), tuan.maianh@hust.edu.vn (M.A Tuan).
Peer review under responsibility of Vietnam National University, Hanoi.
Contents lists available atScienceDirect Journal of Science: Advanced Materials and Devices
j o u r n a l h o m e p a g e : w w w e l s e v i e r c o m / l o c a t e / j s a m d
https://doi.org/10.1016/j.jsamd.2018.04.002
2468-2179/© 2018 The Authors Publishing services by Elsevier B.V on behalf of Vietnam National University, Hanoi This is an open access article under the CC BY license
Journal of Science: Advanced Materials and Devices 3 (2018) 129e138
Trang 2tween the dimers and the monomers[24] Pure PPy, an insulator,
becomes a good conductor when being oxidized The charge
associated with the oxidized state is typically delocalized over
several pyrrole units and it can create a radical cation (polaron) or a
dication (bipolaron)[25] However, the conductivity of PPy is still
significantly lower than that of metals In order to improve the
performance of PPy-based sensors, PPy is usually prepared in the
form of nanomaterials after doping metallic nanoparticles on the
latter[26] The electrodes modification by conducting polymers has
shown its high applicability in combination with many
bio-molecules, such as protein, DNA, RNA etc For instance, DNA sensors
using TAE and PPy could offer a LOD of 1 nM[27]and 2 nM[28],
respectively LOD can be further refined by combining high surface
agents such as carbon nanotubes [29], graphite[30], conducting
polymers[31], and metallic nanoparticles[7,31]
There are several methods for measuring DNA concentrations,
such as electrochemical impedance spectroscopy [15,19,32e34],
optical (fluorescent) [6,35], mechanical (resonance cantilever)
[36,37], colorimetric [38] and magnetic methods [39,40] The
electrochemical impedance spectroscopy (EIS) method requires a
fitting procedure[31,32,41e46] The drawback of the EIS method
lies in its time consuming fitting procedure Another technique
which can be performed on the same electrode geometrics with the
EIS is the differential voltage (DV)[47,48] The DV method relies on
the measurement of the potential difference of the electrodes with
and without DNA hybridization
The LOD is one of the most important factors to evaluate the
quality of a sensor However, the way of calculation varies in
different cases In the simplest way, calculation error in LOD might
be as large as three times of the standard deviation [49e51] In
more complicated ways, calculating LOD must be based on the
standard deviation, the blank data and the calibration data
[30,41,52] The blank measurement plays an important role It is
normally the measurement in a solution without any analytical
agent, which indeed is the target DNA This procedure is not really
easily applicable for DNA sensors because there are as many
LOD in the case of a logarithmic dependence, which includes a blank measurement with the addition of non-complementary target DNA molecules to the analyte We have shown a method to modify the interdigitated DNA sensor with polypyrrole nanowires and gold nanoparticles The electrochemical impedance spectros-copy and differential voltage techniques were used for the DNA detection
2 Experimental 2.1 Fabrication and measurements
A diagram illustrating the surface modification process is pre-sented inFig 1 Gold electrodes on a silica/silicon substrate were used to grow the polypyrrole nanowires, followed by the decoration with gold nanoparticles After immobilizing with probe DNA, the electrodes were used for detecting the DNA concentration through the hybridization process between the probe and the target DNA Pyrrole monomers, gelatin, KCl, K2Cr2O799%, H2SO498%, N299.9%, LiClO4$3H2O, phosphate buffer saline (PBS), 20 nm of diameter gold nanoparticles were commercially obtained from SigmaeAldrich The probe and target DNA, each of which consisting of 25 bases, were supplied by the Invitrogen Life Technologies Company Se-quences of the probe DNA, the complementary and the non-complementary target DNA are shown inTable 1 The schematic diagram of the interdigitated electrodes is presented inFig 2 De-tails of the preparation of the interdigitated electrodes have been reported elsewhere [53] The polymerization process was con-ducted in an electrochemical reaction cell using an Autolab PGSTAT302N Two polymerization experiments were performed according to the connection configurations as shown inFig 2c: (i) arms 1 and 2 were connected as the working electrodes; (ii) all 4 arms were connected as the working electrodes The PPy polymer-ization on the electrode surface was presented in previous articles
[54,55]with the polymerization time of 200 s The bare gold elec-trode (denoted as BG elecelec-trode) surface was modified with
Fig 1 Side view diagram of the surface modification process of an interdigitated sensor From left to right: the BG, PPy, GNP-PPy, GNP-PPy immobilized with the probe DNA, and
Trang 3polypyrrole nanowires (denoted as PPy electrode) Gold
nano-particles were decorated on the PPy electrode by dropping 3ml of a
solution containing 25mM GNPs in PBS After incubating at ambient
conditions for one day, the surface was rinsed with deionized (DI)
waterfive times Finally, the PPy electrode surface decorated with
gold nanoparticles (denoted as GNP-PPy electrode) was dried in air
for further experiments DNA immobilization was conducted by
dropping 3ml of a solution containing 2mM probe DNA in PBS on the
BG, PPy and GNP-PPy electrodes The electrodes were dried under
ambient conditions for one day, then rinsed with DI waterfive times
to remove unbound probe DNA molecules The DNA hybridization
process was studied by conducting the EIS measurement of the
interdigitated sensors with the connection configuration (iii) shown
in Fig 2c The EIS analyses were performed using an Autolab
PGSTAT302N (Metrohm Autolab B.V., Utrecht, The Netherlands)
with a conventional three-electrode setup An alternative potential
with modulus of 250 mV DC and±10 mV AC with frequency of
0.1 Hze10 kHz was applied on the working electrode All
experi-ments were performed in the presence of 10 mM [Fe(CN)6]3/4in
0.1 M KCl as the redox-active indicator The EIS results were then
fitted to an equivalent circuit (in our case, it is the Randle circuit) by
using the ZsimpWin 3.10 software The EIS setup was the same as
that for PPy nanowires as shown inFig 2b The morphology of the
materials and the electrode surface were investigated on a Hitachi
S4800 scanning electron microscope (SEM) The chemical
compo-sition of the gold nanoparticles was studied via energy dispersion
x-ray spectroscopy (EDS) on the same SEM system The change in the
electrical potential between the electrode with the probe DNA and
the electrode without the probe DNA provided the DV signals
De-tails of the DV system were presented in[17] In brief, a potential of
100 mV with the frequency of 10 kHz from an SR830 lock-in
amplifier (Stanford Research system, USA) was applied to the
elec-trodes The hybridization of the target and the probe DNA cause a
change in the conductance of the system The output signals are the
voltage drop across two 1-kUresistors Each measurement was
repeated 5 times
2.2 Limit of detection analysis
The verified method for calculating the LOD value was applied
For the simplest case, LOD is determined from the blank standard
deviation, sy0, using the equation:
where t is the coefficient for a Student's t distribution[56]This equation is primitive so it is not often used in determining LOD recently When the analytical signal is univariating and specific, LOD is evaluated from the average signal value and standard de-viations of repeated measurements of a blank sample and of several samples at concentrations near the detection limit[57] In many cases, the linear regression is applied when the response is linearly proportional to the analyte concentration, y ¼ a þ bx Here, a is the intercept of the calibration curve with the horizontal axis; b is the slope of the calibration curve; x is the analyte concentration; and y
is the response signal The limit of detection is calculated as following:
xD¼tsy0
this definition of LOD does not take into account the standard de-viation of the calibration measurements In many cases, thefirst order dependence of the signal on the analyte concentration is observed This requires a linear regression In our study, a loga-rithmic dependence was observed therefore LOD must be differ-ently calculated than just from Eq.(2) The logarithmic dependence
is y ¼ a þ b ln x Rearranging the equation, we obtain x ¼ eyab In order to calculate the deviation of x, all terms have to be taken into account to calculate the standard deviation of the concentration:
s2
x¼
vx vy
2
s2
yþ
vx va
2
s2
aþ
vx vb
2
s2 b
with sa, sb, sxand syas the deviation of the intercept, the slope, the concentration and the response signal, respectively
Taking the derivatives and the square root, we obtain:
sx¼1beyab
"
s2yþ s2
aþ
y a b
2
s2b
#1=2
This equation can be used to estimate the limit of detection where y is placed by the blank signal y0; syis changed to sy0
xD¼ tsx¼btey0ab
"
s2y0þ s2
aþ
y0 a b
2
s2b
#1=2
The confidence factor t ¼ 3 is chosen corresponding to the probability of making type I and type II errorsa ¼b¼ 0:05, the confidence level of 95% Here, we apply the weighted regression
Fig 2 (a) Front view diagram of the interdigitated sensor consisting of 4 arms (denoted as 1e4) (b) Diagram of the polypyrrole polymerization and the EIS measurement setup In these setups, the interdigitated sensor described in (a) plays as a working electrode (WE) The WE with the counter electrode (CE) and the reference electrode (RE) form a conventional three-electrode system (c) Three connection configurations of the four arms: (i) and (ii) are for the preparation of the polypyrrole polymerization, (iii) is for the EIS measurement.
P Van Hao et al / Journal of Science: Advanced Materials and Devices 3 (2018) 129e138 131
Trang 43 Results and discussion
Fig 3presents the SEM images of PPy grown on gold electrode
surface It can be seen that PPy was formed in wires with an average
size of about 200 nm and a length of about 5mm These nanowires
were fully distributed on the gold electrode surface Some of these
nanowires have branches with a size of about 80 nm In order to
study the polymerization of the pyrrole molecules on the
electri-cally charged electrodes, in one experiment setup (configuration (i)
inFig 2c) the arms 1 and 2 were subjected to a voltage of 0.75 V
whereas the electrodes 3 and 4 were not.Fig 3a shows the PPy
nanowires grown on thefingers of arm 1, but not on those of arm 3
So it was confirmed that the pyrrole molecules have been
poly-merized on the working electrode forming PPy nanowires In
another polymerization setup (configuration (ii) inFig 2c), all four
nanowires and on the silica surface of the gaps
Fig 4shows the Nyquist plots of the BG, PPy and GNP-PPy electrodes before and after DNA immobilization All results show
a semicircle at high frequencies and a straight line at low fre-quencies This feature is typical for a simple Randle's circuit (the inset ofFig 4a) Therefore, the Randle cell was applied to study the system The Randle circuit contains an electrolyte resistance, RS, in series with a parallel combination of a constant phase element,
QCPE, (often called as a double layer capacitance) and an impedance
of the faradaic reaction, which consists of a charge transfer resis-tance, RCT, in series with a Warburg element, W In a typical Randle cell configuration, the Nyquist semicircle possesses two compo-nents: (i) the resistor's impedance on the left of the real part Z0 corresponding to the high frequency range; (ii) the sum of the resistor's and the capacitor's impedance on the right of Z0
Fig 3 SEM images of PPy nanowires covered on BG electrode (a) The SEM result corresponding to the connection configuration (i) in Fig 2 c: PPy nanowires only appear on all fingers of arm 1 (and arm 2) and not on fingers of arm 4 (and arm 3) (b) The SEM result corresponding to the connection configuration (ii) in Fig 2 c: PPy nanowires appear on all fingers of the 4 arms and also bridging the gap between the fingers (c) Higher magnification of (b) (d) SEM images of PPy nanowires decorated with GNPs.
Trang 5corresponding to the low frequency range The Warburg impedance
of the Randle cell is due to the diffusion process of molecules in the
electrolyte under a gradient of concentration This impedance
de-pends on the frequency of the potential At the high frequency
range, the diffusing reactants cannot respond to the fast change of
the applied voltage As a consequence, the impedance is small At
the low frequency range, the reactants can follow the change of the
applied voltage, thus leading to a high value of impedance InFig 4,
the semi-infinite Warburg impedance is important at the low
fre-quency range, which appears as the diagonal line If there is any
change on the surface of the electrode, RCT and QCPE are highly
affected Studying the change of these quantities reveals the
pro-cesses occurring on the electrode surface In the literature, RCThas
been used more often than QCPE [31e34,45,58] In this study, we
used both RCT and QCPE as the EIS analytic signal for determining
the concentration of DNA The results obtained from all approaches
were almost the same Therefore, we only present the data from
RCT
Values of RCT deduced from the Nyquist plots in Fig 4, are
shown onTable 2 RCT was increased from 3982 U for the BG
electrode to 5120Ufor the PPy electrode, then reduced to 3495U
for the GNP-PPy electrode RCTpresents the charge transfer kinetics
in the absence of a mass transfer limitation It is inversely
propor-tional to the exchange current between the electrode surface and
the solution With the presence of the conducting PPy nanowires,
the increase of electrode surface might generate an increase of
exchange current The poor conductance of the PPy nanowires, in
comparison with the gold paticles layer, leads to the increase of RCT
The presence of the gold nanoparticles on the PPy nanowires
sur-face causes the increase of the conductivity of the electrode sursur-face
The gold nanoparticles have a high surface area, on which a
dy-namic balance of Au0¼ Auþþ eis established The appearance of
the charged particles Auþ and the electrons contribute to the
conductance, thus reducing the RCT DNA is a polymer of four types
of nucleotide linking together by a backbone The phosphodiester
bonds in the backbone retain one or two negative charges from the
oxygen atoms, thus making the negatively charged DNA molecules
DNA molecules are not conductive, so consequently, their immo-bilization on the surfaces generates a repulsion of the redox species (in this study, these are [Fe(CN)6]3/4), thus inhibiting the redox reaction and enhancing the charge transfer resistance InTable 2, the parameter n is a dimensionless CPE coefficient, which is directly related to the degree of inhomogeneity and roughness of the electrode surface The value of n can be between 0 (for ideal smooth surface) and 1 (for extremely rough surface) In this study, n varied from 0.73 to 0.94, which indicates the high inhomogeneity of the CPE (originating from the roughness of the PPy nanowires) The value of n is the typical value obtained from other researches[19] The parameterc2 presents the deviation from the Randles model and the experimental data The small value ofc2suggests that the Randles model is well appropriate to the experimental data The role of the PPy nanowires is to improve the surface area of the electrodes However, the role of GNP can be either to increase the conductivity of the electrode or to enhance the affinity of the DNA molecules with the electrode surface To elucidate the role of the PPy nanowires, the cyclic voltammetry study of the BG, PPy and GNP-PPy electrodes were undertaken Using the RandleseSevcik equation[43], which describes the effect of scan rate on the peak current, ip, we can calculate the effective electrode area, A by using the equation:
ip¼ kACn3=2D1=2v1=2: where k¼ 2:69 105 is the experimental coefficient; n is the number of electrons transferred in the redox event (1); D is the diffusion coefficient for K3[Fe(CN)6] (7:6 108 m2/s); C is the concentration of K3[Fe(CN)6] (0.03 M/l);v is the scan rate (25 mV/s) From the cyclic voltammetry shown inFig 5, the effective area of
207 mm2 for BG electrode is reduced to 143 mm2 for the PPy electrode With the presence of the gold nanoparticles, the effective area regained to 226 mm2 The effective area was enhanced significantly with the presence of the gold nanoparticles This can
be explained by the fact that the gold nanoparticles provide charged entities Auþand electrons The presence of the charged entities foster the charge transfer process and thus, induce an in-crease of the effective area
Fig 6presents the Nyquist plots of the EIS measurements of the
BG, PPy and GNP-PPy electrode after the DNA hybridization In all cases, the shape of the plot is very similar to that shown inFig 4 This suggests that the Randles model can be applied to study the system A logarithmic dependence of RCTdeduced from the Randle model on the target DNA concentration was obtained This approach is not as usual as other studies in which a linear depen-dence is often observed[30,33,52] The blank data are normally
defined as the data taken from the measurement in which the
Fig 4 EIS Nyquist plots of the BG (solid square), PPy (solid circle), GNP-PPy electrode (solid triangle) before (a) and after (b) DNA immobilization The inset in (a) presents the equivalent Randles circuit.
Table 2
Impedance parameters obtained from fitting to the Randles model Electrodes after
immobilizing with the probe DNA are denoted as/DNA.
Electrode R s ðUÞ Q CPE (mF) n R CT ðUÞ W ð10 5 Þ c2 ð10 4 Þ
GNP-PPy electrode 110.55 2.82 0.73 3495 40.83 9.47
P Van Hao et al / Journal of Science: Advanced Materials and Devices 3 (2018) 129e138 133
Trang 6analyte does not contain the target DNA The presence of a
non-complementary target DNA may contribute to the usual blank
data Therefore, we used as blank data those taken from the
mea-surement in the presence of the non-complementary target DNA
(the sequences of the DNA are shown inTable 1) Each measure-ment was conducted three times The blank data taken from the analyte in the presence of the non-complementary target DNA are presented in the insets ofFig 6 There is seen a slight difference
Fig 5 Cyclic voltammetry data of the BG (solid square), PPy (solid circle), and GNP-PPy (solid triangle) electrodes before (a) and after (b) DNA immobilization.
Fig 6 Left: EIS Nyquist plots of (a) the BG, (b) PPy and (c) GNP-PPy electrodes with different DNA concentrations Right: DNA concentration dependence of R CT of the (d) BG, (e) PPy and (f) GNP-PPy electrode The insets of (d)e(f) present the blank data The straight lines in (d)e(f) are the fitting lines.
Trang 7between these blank data and those taken from the analyte without
the non-complementary target DNA This fact confirmed that the
presence of the non-complementary DNA has affected the
mea-surements through the physical attachment of the
non-complementary DNA on the electrode surface There was no
sig-nificant change in the value of RCT when increasing the
non-complementary DNA concentration When the non-complementary
target DNA was added to the analyte, RCT increased almost
loga-rithmically with the DNA concentration In the analysis process, we
used the concentration range in which RCT depended
logarithmi-cally, and this logarithmic dependence occurs indeed at high
con-centrations The low DNA concentrations were therefore not
considered in the analysis
There occurs a hybridization process between the probe DNA
and the complementary target DNA Upon hybridization, an
adenine-thymine based pair possessed two intermolecular
hydrogen bonds; a guanine-cytosine based pair possessed three
intermolecular hydrogen bonds In fact, a hydrogen bond has the
electrostatic attraction nature between two polar groups that are
formed when a hydrogen atom bound to a highly electronegative
atom such as nitrogen and oxygen This hybridization process does not create any charge, but simply induces an accumulation of DNA molecules on the electrode surface, thus enhancing the charge transfer resistance
In order to study another possible method for the DNA detection
of the sensor, the differential voltage measurement was used Four arms of the sensor were modified with PPy nanowires and gold nanoparticles Arm 2 and 3 were covered by the probe DNA mol-ecules (playing the role of the working sensors) leaving arm 1 and 4 without probe DNA molecules (playing the role of the reference sensors) The hybridization process between the target and the probe DNA on the working sensor gives rise to the conductance change of the electrode surface, hence leading to the change in the output voltage.Fig 7shows the time dependence of the differential voltageDU¼ jUw Urj of the BG, PPy, and GNP-PPy electrodes Uw
and Ur are the working and the reference electrode voltage, respectively The value DUm is the difference between the mean values ofDU before and after immersing the sensor in the analyte The change in voltage appeared immediately when the sensors immersed in the analyte with the response time of few seconds
Fig 7 Left: Time dependence of the DV signals,DU of the (a) BG, (b) PPy and (c) GNP-PPy electrode with different target DNA concentrations Right: DNA concentration dependence
ofDU of (d) the BG, (e) PPy and (f) GNP-PPy electrode The insets of (d)e(f) present the blank data The straight lines in (d)e(f) are the fitting lines.
P Van Hao et al / Journal of Science: Advanced Materials and Devices 3 (2018) 129e138 135
Trang 8The short response time is an important characteristics of the DNA
sensors Similarly to the EIS data, the change in voltage increases
logarithmically with the concentration of the target DNA in the
same manner of that in the RCT The results of the blank
mea-surements are shown on the insets ofFig 6 Each measurement was
conducted 5 times The hybridization of DNA results in the
accu-mulation of DNA molecules on the electrode surface, thus increases
the resistance and voltage of the signals from the electrode
The interpretation and estimation of the limit of detection are
diversified in terms of calculation formula and the blank
mea-surement Here, we applied the most used formulas to estimate
LOD (Eq.(1)and Eq.(2)) and compared the results of calculation to
that obtained from the calculation using our formula (Eq.(3)) In
the estimation, the value of t equal to 3 was chosen The results are
displayed inTable 3 We have conducted the DNA hybridization
measurements with all three electrode surfaces, namely, BG, PPy
and GNP-PPy surfaces For each surface, the EIS and DV methods
were studied When using Eq.(3), among three types of surface, the
GNP-PPy electrode provided the lowest LOD of the order of 1014M
That value increased to 1011M and 109 M for the PPy and BG
surface, respectively This feature can be explained by the fact that
the presence of the PPy in PPy induces an increase of the electrode
surface area; the gold nanoparticles on the GNP-PPy electrode
provide the charged entities Both of them enforce the charge
transfer process The blank measurements were performed in two
cases: the analyte without any DNA and the analyte with a non-complementary DNA (Data not shown) The limit of detection deduced from the blank data of the former was lower than that of the latter by 101 102M This indicates that the presence of the
non-complementary DNA on the electrode surface causes changes
in the charge transfer resistance and induces the difference in RCTin the blank data by 10e30% The values of LOD obtained from Eq.(1)
and Eq (2) were lower than that obtained from Eq (3) by
101 103 M All equations take into account the standard
de-viation of the blank response signal, sy0 Eq.(3), however, includes the standard deviation of both the intercept sa and the slope sb Therefore, the value of LOD obtained using Eq.(3)is higher, but more meaningful than that from the other approach Our best re-sults obtained from Eq.(3)are for the GNP-PPy electrodes with LOD value of 8:45 1014 M and 6:00 1014 M for the EIS and DV
method, respectively This might originate from the improvement
of the affinity to DNA probe molecules and the conductivity of the electrode surface due to the presence of the gold nanoparticles A similar calculation of LOD was based on the taking into account the constant phase element instead of the charge transfer resistance The results of the two approaches were found almost the same, which suggested that we could choose either RCTor QCPEto analyze the data Since the DV method is simpler than the EIS method and both give the same order of magnitude of the LOD value, the dif-ferential voltage measurement can be considered as more
Table 4
Comparison of LOD values of different DNA sensors.
Electrical
1:70 10 10 Electrochemical Change in R CT signals or meldola's blue
signal due to hybridization
[33] 3:20 10 12 Electrochemical Cyclic voltammetry, differential pulse
voltammetry
[61]
Trang 9advantagous and favorable than the electrochemical impedance
spectroscopy measurement for the DNA sensing detection.Table 4
shows a comparison of the LOD values from our approaches with
those from other DNA sensing methods using Eqs.(1)e(3) Here we
used the blank data taken from the analyte with the presence of the
non-complementary DNA Depending on the equation used, LOD
varied from the order of 1017 1014 In any cases, our values of
LOD are the lowest ones, much lower than those obtained from
other methods The logarithmic dependence on the concentration
of the signals may be the cause of the low limit of detection Other
studies yielding higher LOD values, indeed, based on the linear
dependence of the signal on the concentration
4 Conclusion
A DNA sensor based on the interdigitated electrode has been
successfully prepared The electrochemical impedance
spectros-copy and the differential voltage techniques were used to
deter-mine DNA concentrations in a solution The electrode surface
modification with polypyrrole nanowires and gold nanoparticles
has improved the sensor performance We have derived a formula
to calculate the limit of detection based on a logarithmic
depen-dence of the signals on the concentration The blank measurement
was conducted with the presence of a non-complementary target
DNA in the analyte Our approach for the estimation of the LOD is
more complicated than other procedures, but has brought by quite
meaningful and well explained results It can be, therefore,
effec-tively applied in further researches The polypyrrole nanowires
have been found to induce the increase of the electrode surface area
and the gold nanoparticles to improve the affinity to DNA probe
molecules and the conductivity of the probe surface The
differ-ential voltage method appears to be more versatile, rapid and easier
than the electrochemical impedance spectroscopy one and will be a
potential tool for our future DNA sensors research
Acknowledgements
The authors would like to thank the MOET project
#B2015-01-102 for supporting this work
References
[1] C.L Manzanares-Palenzuela, B Martín-Fernandez, M Sanchez-Paniagua
Lopez, B Lopez-Ruiz, Electrochemical genosensors as innovative tools for
detection of genetically modified organisms, TrAC Trends Anal Chem 66
(2015) 19e31, https://doi.org/10.1016/j.trac.2014.10.006
[2] N Khemthongcharoen, W Wonglumsom, A Suppat, K Jaruwongrungsee,
A Tuantranont, C Promptmas, Piezoresistive microcantilever-based DNA
sensor for sensitive detection of pathogenic Vibrio cholerae O1 in food sample,
Biosens Bioelectron 63 (2015) 347e353, https://doi.org/10.1016/
j.bios.2014.07.068
[3] L.A Hiatt, D.E Cliffel, Real-time recognition of Mycobacterium tuberculosis
and lipoarabinomannan using the quartz crystal microbalance, Sensor
Actu-ator B Chem 174 (2012) 245e252, https://doi.org/10.1016/j.snb.2012.06.095
[4] D.P Kalogianni, T Koraki, T.K Christopoulos, P.C Ioannou, Nanoparticle-based
DNA biosensor for visual detection of genetically modified organisms, Biosens.
Bioelectron 21 (2006) 1069e1076, https://doi.org/10.1016/j.bios.2005.04.016
[5] R Marie, H Jensenius, J Thaysen, C.B Christensen, A Boisen, Adsorption
ki-netics and mechanical properties of thiol-modified DNA-oligos on gold
investigated by microcantilever sensors, Ultramicroscopy 91 (2002) 29e36,
https://doi.org/10.1016/S0304-3991(02)00079-7
[6] R Ravichandran, S Sundarrajan, J.R Venugopal, S Mukherjee, S Ramakrishna,
Applications of conducting polymers and their issues in biomedical
engi-neering, J R Soc Interface 7 (Suppl 5) (2010) S559eS579, https://doi.org/
10.1098/rsif.2010.0120.focus
[7] T Premkumar, K.E Geckeler, Graphene-DNA hybrid materials: assembly,
ap-plications, and prospects, Prog Polym Sci 37 (2012) 515e529, https://
doi.org/10.1016/j.progpolymsci.2011.08.003
[8] S.R Shin, C.K Lee, T.W Eom, S.H Lee, C.H Kwon, I So, S.J Kim, DNA-coated
MWNT microfibers for electrochemical actuator, Sens Actuators B Chem 162
(2012) 173e177, https://doi.org/10.1016/j.snb.2011.12.063
[9] N.V Berezhnoy, N Korolev, L Nordenskiold, Principles of electrostatic in-teractions and self-assembly in lipid/peptide/DNA systems: applications to gene delivery, Adv Colloid Interface Sci 205 (2014) 221e229, https://doi.org/ 10.1016/j.cis.2013.08.008
[10] D Zhang, M.C Huarng, E.C Alocilja, A multiplex nanoparticle-based bio-bar-coded DNA sensor for the simultaneous detection of multiple pathogens, Biosens Bioelectron 26 (2010) 1736e1742, https://doi.org/10.1016/ j.bios.2010.08.012
[11] M Kalofonou, C Toumazou, Semiconductor technology for early detection of DNA methylation for cancer: from concept to practice, Sensor Actuator B Chem 178 (2013) 572e580, https://doi.org/10.1016/j.snb.2012.12.054 [12] M Sirajuddin, S Ali, A Badshah, Drug-DNA interactions and their study by UV-Visible, fluorescence spectroscopies and cyclic voltametry, J Photochem Photobiol B Biol 124 (2013) 1e19, https://doi.org/10.1016/ j.jphotobiol.2013.03.013
[13] G Yunus, S Srivastava, M Kuddus, V.D Gupta, Drug-DNA interaction: a theoretical study on the binding of thionine with DNAs of varying base composition, Curr Appl Phys 13 (2013) 322e326, https://doi.org/10.1016/ j.cap.2012.05.020
[14] Y Shin, A.P Perera, M.K Park, Label-free DNA sensor for detection of bladder cancer biomarkers in urine, Sensor Actuator B Chem 178 (2013) 200e206, https://doi.org/10.1016/j.snb.2012.12.057
[15] T.J Gnanaprakasa, O.A Oyarzabal, E.V Olsen, V.A Pedrosa, A.L Simonian, Tethered DNA scaffolds on optical sensor platforms for detection of hipO gene from Campylobacter jejuni, Sensor Actuator B Chem 156 (2011) 304e311, https://doi.org/10.1016/j.snb.2011.04.037
[16] J.I.A Rashid, N.A Yusof, J Abdullah, U Hashim, R Hajian, The utilization of SiNWs/AuNPs-modified indium tin oxide (ITO) in fabrication of electro-chemical DNA sensor, Mater Sci Eng C 45 (2014) 270e276, https://doi.org/ 10.1016/j.msec.2014.09.010
[17] A.T Mai, T.P Duc, X.C Thi, M.H Nguyen, H.H Nguyen, Highly sensitive DNA sensor based on polypyrrole nanowire, Appl Surf Sci 309 (2014) 285e289, https://doi.org/10.1016/j.apsusc.2014.05.032
[18] T Quang, N Thi, H Hanh, N Thanh, P Van Chung, A novel biosensor based on serum antibody immobilization for rapid detection of viral antigens, Talanta
86 (2011) 271e277, https://doi.org/10.1016/j.talanta.2011.09.012 [19] T.L Tran, T.X Chu, D.C Huynh, D.T Pham, T.H.T Luu, A.T Mai, Effective immobilization of DNA for development of polypyrrole nanowires based biosensor, Appl Surf Sci 314 (2014) 260e265, https://doi.org/10.1016/ j.apsusc.2014.06.068
[20] L Goswami, N Sen Sarma, D Chowdhury, Determining the ionic and elec-tronic contribution in conductivity of polypyrrole/Au nanocomposites, J Phys Chem C 115 (2011) 19668e19675, https://doi.org/10.1021/jp2075012 [21] X Zhang, J Zhang, W Song, Z Lu, Controllable synthesis of conducting pol-ypyrrole nanostructures, J Phys Chem B 110 (2006) 1158e1165, https:// doi.org/10.1021/jp054335k
[22] N Mermilliod, J Tanguy, F Petiot, A study of chemically synthesized poly-pyrrole as electrode material for battery applications, J Electrochem Soc 133 (1986) 1073e1079, https://doi.org/10.1149/1.2108788
[23] S Machida, S Miyata, A Techagumpuch, Chemical synthesis of highly elec-trically conductive polypyrrole, Synth Met 31 (1989) 311e318, https:// doi.org/10.1016/0379-6779(89)90798-4
[24] D Ateh, H Navsaria, P Vadgama, Polypyrrole-based conducting polymers and interactions with biological tissues, J R Soc Interface 3 (2006) 741e752, https://doi.org/10.1098/rsif.2006.0141
[25] G March, T Nguyen, B Piro, Modified electrodes used for electrochemical detection of metal ions in environmental analysis, Biosensors 5 (2015) 241e275, https://doi.org/10.3390/bios5020241
[26] I Chikouche, A Sahari, A Zouaoui, S Tingry, Enhancement of electric prop-erties of polypyrrole by copper electrodeposition, Can J Chem Eng 93 (2015) 1076e1080, https://doi.org/10.1002/cjce.22197
[27] J Cha, J.I Han, Y Choi, D.S Yoon, K.W Oh, G Lim, DNA hybridization elec-trochemical sensor using conducting polymer, Biosens Bioelectron 18 (2003) 1241e1247, https://doi.org/10.1016/S0956-5663(03)00088-5
[28] P.D Tam, M.A Tuan, T.Q Huy, A.-T Le, N Van Hieu, Facile preparation of a DNA sensor for rapid herpes virus detection, Mater Sci Eng C 30 (2010) 1145e1150, https://doi.org/10.1016/j.msec.2010.06.010
[29] R Hajian, Z Mehrayin, M Mohagheghian, M Zafari, P Hosseini, N Shams, Fabrication of an electrochemical sensor based on carbon nanotubes modified with gold nanoparticles for determination of valrubicin as a chemotherapy drug: valrubicin-DNA interaction, Mater Sci Eng C 49 (2015) 769e775, https://doi.org/10.1016/j.msec.2015.01.072
[30] B Rezaei, M.K Boroujeni, A.A Ensafi, Fabrication of DNA, o-phenylenedi-amine, and gold nanoparticle bioimprinted polymer electrochemical sensor for the determination of dopamine, Biosens Bioelectron 66 (2015) 490e496, https://doi.org/10.1016/j.bios.2014.12.009
[31] J Wilson, S Radhakrishnan, C Sumathi, V Dharuman, Polypyrrole-polyani-line-Au (PPy-PANi-Au) nano composite films for label-free electrochemical DNA sensing, Sensor Actuator B Chem 171e172 (2012) 216e222, https:// doi.org/10.1016/j.snb.2012.03.019
[32] T.T.N Lien, Y Takamura, E Tamiya, M.C Vestergaard, Modified screen printed electrode for development of a highly sensitive label-free impedimetric immunosensor to detect amyloid beta peptides, Anal Chim Acta 892 (2015) 69e76, https://doi.org/10.1016/j.aca.2015.08.036
P Van Hao et al / Journal of Science: Advanced Materials and Devices 3 (2018) 129e138 137
Trang 10https://doi.org/10.1016/j.snb.2012.12.111
[38] Y Song, W Wei, X Qu, Colorimetric biosensing using smart materials,
Adv Mater 23 (2011) 4215e4236, https://doi.org/10.1002/adma.
201101853
[39] S Bi, Y Cui, Y Dong, N Zhang, Target-induced self-assembly of DNA
nano-machine on magnetic particle for multi-amplified biosensing of nucleic acid,
protein, and cancer cell, Biosens Bioelectron 53 (2014) 207e213, https://
doi.org/10.1016/j.bios.2013.09.066
[40] F Li, J Kosel, A magnetic method to concentrate and trap biological targets,
IEEE Trans Magn 48 (2012) 2854e2856, https://doi.org/10.1109/
TMAG.2012.2202644
[41] P Van Hao, P.D Thanh, C.T Xuan, N.H Hai, M.A Tuan, Development of a DNA
sensor based on nanoporous Pt-Rich electrodes, J Electron Mater 46 (2017)
3491e3498, https://doi.org/10.1007/s11664-017-5326-y
[42] C Ocana, M del Valle, Three different signal amplification strategies for the
impedimetric sandwich detection of thrombin, Anal Chim Acta 912 (2016)
117e124, https://doi.org/10.1016/j.aca.2016.01.027
[43] A.J Bard, Faulkner, Electrochemical Methods, 2000, https://doi.org/10.1146/
annurev.matsci.30.1.117
[44] A Alsamuraee, H Jaafer, Electrochemical impedance spectroscopic evaluation
of corrosion protection properties of polyurethane/polyvinyl chloride blend
coatings on steel, Am J Sci Ind Res 2 (2011) 761e768, https://doi.org/
10.5251/ajsir.2011.2.5.761.768
[45] A Bonanni, M.I Pividori, S Campoy, J Barbe, M del Valle, Impedimetric
detection of double-tagged PCR products using novel amplification
proced-ures based on gold nanoparticles and protein G, Analyst 134 (2009) 602e608,
https://doi.org/10.1039/b815502j
[46] E Katz, I Willner, Probing biomolecular interactions at conductive and
sem-iconductive surfaces by impedance spectroscopy: routes to impedimetric
immunosensors, DNA-sensors, and enzyme biosensors, Electroanalysis 15
(2003) 913e947, https://doi.org/10.1002/elan.200390114
[47] G Rizzi, F.W Osterberg, A.D Henriksen, M Dufva, M.F Hansen, On-chip
magnetic bead-based DNA melting curve analysis using a magnetoresistive
sensor, J Magn Magn Mater 380 (2015) 215e220, https://doi.org/10.1016/
j.jmmm.2014.09.004
[48] P.D Tam, M.A Tuan, N Van Hieu, N.D Chien, Impact parameters on
hybrid-ization process in detecting influenza virus (type A) using
conductimetric-biosensor, J Mater Chem B 3 (2015) 458e464, https://doi.org/10.1039/ C4TB01320D
[53] C Karuppiah, S Palanisamy, S.M Chen, S.K Ramaraj, P Periakaruppan,
A novel and sensitive amperometric hydrazine sensor based on gold nano-particles decorated graphite nanosheets modified screen printed carbon electrode, Electrochim Acta 139 (2014) 157e164, https://doi.org/10.1016/ j.electacta.2014.06.158
[54] M Labib, P.O Shipman, S Martic, H Kraatz, A bioorganometallic approach for rapid electrochemical analysis of human immunodeficiency virus type-1 reverse transcriptase in serum, Electrochim Acta 56 (2011) 5122e5128, https://doi.org/10.1016/j.electacta.2011.03.063
[55] S.M Radke, E.C Alocilja, A high density microelectrode array biosensor for detection of E coli O157: H7, Biosens Bioelectron 20 (2005) 1662e1667, https://doi.org/10.1016/j.bios.2004.07.021
[56] D Rodbard, Statistical estimation of the minimal detectable concentration (“sensitivity”) for radioligand assays, Anal Biochem 90 (1978) 1e12, https:// doi.org/10.1016/0003-2697(78)90002-7
[57] G.L Long, J.D Winefordner, Limit of detection: a closer look at the IUPAC definition, Anal Chem 55 (1983) 712e724, https://doi.org/10.1021/ ac00258a001
[58] T.T.N Lien, T.D Lam, V.T.H An, T.V Hoang, D.T Quang, D.Q Khieu,
T Tsukahara, Y.H Lee, J.S Kim, Multi-wall carbon nanotubes (MWCNTs)-doped polypyrrole DNA biosensor for label-free detection of genetically modified organisms by QCM and EIS, Talanta 80 (2010) 1164e1169, https:// doi.org/10.1016/j.talanta.2009.09.002
[59] S Niu, Y Jiang, S Zhang, Fluorescence detection for DNA using hybridization chain reaction with enzyme-amplification, Chem Commun (Camb) 46 (2010) 3089e3091, https://doi.org/10.1039/c000166j
[60] M.M.N Nuzaihan, U Hashim, M.K Md Arshad, S.R Kasjoo, S.F.A Rahman, A.R Ruslinda, M.F.M Fathil, R Adzhri, M.M Shahimin, Electrical detection of dengue virus (DENV) DNA oligomer using silicon nanowire biosensor with novel molecular gate control, Biosens Bioelectron 83 (2016) 106e114, https://doi.org/10.1016/j.bios.2016.04.033
[61] M.B Gholivand, L Mohammadi-Behzad, G Paimard, K Gholivand, A Gholami, Solid state electrochemical oxidation of some bisphosphoramidates in aqueous media and their applications in DNA sensing, Electroanalysis 28 (2016) 601e610, https://doi.org/10.1002/elan.201500391