Journal of Electroanalytical Chemistry 626 (2009) 6–13

Contents lists available at ScienceDirect

Journal of Electroanalytical Chemistry

journal homepage: www.elsevier.com/locate/jelechem

Electrochemical amperometric immunoassay for carcinoembryonic antigen

based on bi-layer nano-Au and nickel hexacyanoferrates nanoparticles
modified glassy carbon electrode
Yan-Ru Yuan, Ruo Yuan *, Ya-Qin Chai, Ying Zhuo, Xiang-Min Miao
Chongqing Key Laboratory of Analytical Chemistry, College of Chemistry and Chemical Engineering, Southwest University, Chongqing 400715, PR China

a r t i c l e i n f o a b s t r a c t

Article history: This study demonstrates a new approach towards development of novel immunosensor based on gold
Received 15 May 2008 nanoparticles (nano-Au) and nickel hexacyanoferrates nanoparticles (NiHCFNPs) for determination of
Received in revised form 13 October 2008 carcinoembryonic antigen (CEA) in clinical immunoassay. The fabrication steps of the immunosensor
Accepted 16 October 2008
as follows: firstly, nano-Au was immobilized on the surface of bare glassy carbon electrode (GCE) by
Available online 20 November 2008
using a simple method – electrochemical reduction of HAuCl4 solution; secondly, NiHCFNPs as an elec-
troactive substance was immobilized on the layer of gold nanoparticles. Microstructure and surface mor-
Keywords:
phology of NiHCFNPs have been characterized by transmission electron microscopy (TEM) and scanning
Amperometric immunosensor
Carcinoembryonic antigen
electron microscopy (SEM); thirdly, nano-Au was again immobilized on the surface of NiHCFNPs, which
Nano-Au can offer a favorable microenvironment and biocompatibility to immobilize anti-CEA. Electrochemical
Hexacyanoferrates nanoparticles impedance spectroscopy (EIS) and cyclic voltammetry (CV) were applied to characterize the electrochem-
ical properties of modified process. Effect of deposition time of nano-Au, pH of working buffer, incubation
temperature and time were studied in detail for optimization of analytical performance. Under optimal
conditions, the peak current of CV of the immunosensor decreased linearly with increasing CEA concen-
tration in two ranges from 0.5 to 10.0 ng mL1 and from 10.0 to 160.0 ng mL1, with a detection limit
0.1 ng mL1 at three times background noise. The proposed immunosensor show good repeatability
and reproducibility, acceptable accuracy, high sensitivity and would be valuable for diagnosis and mon-
itoring of carcinoma.
Ó 2008 Elsevier B.V. All rights reserved.

1. Introduction such as time consuming, requiring highly qualified personnel,

In the tumor process, increased level of tumor markers in hu-
man serum, are associated with certain tumor. Therefore, determi-
nation of tumor markers, potential prognostic factors for tumors,
plays an important role in clinical research and diagnosis [1,2]. Car-
cinoembryonic antigen (CEA), is typically associated with certain
tumors and the developing fetus [3,4] and is widely used as clinical
tumor marker for some familiar cancers [5–8]. The normal range
for CEA in an adult non-smoker is <2.5 ng mL1 and for a smoker
<5.0 ng mL1. A rising CEA level indicates progression or recur-
rence of the cancer [9,10]. Thus, the detection of CEA levels in hu-
man serum is necessary in clinical assay.
The technique based on high specific molecular recognition of
antigen by antibody and used for quantitative determination of tu-
mor markers is usually immunoassay. Many kinds of immunoassay
methods [11–17], have been applied for the detection of CEA. How-
ever, methods described above have always the disadvantages,
* Corresponding author. Tel.: +86 23 68252277; fax: +86 23 68254000.
E-mail address: yuanruo@swu.edu.cn (Y.-R. Yuan).
sophisticated instrumentation, poor precision, difficult to realize
automation [18,19]. Electrochemical immunoassay combining the
features of fast analysis, sensitive and precise measurement, sim-
ple pretreatment, inexpensive and miniaturizable instrumentation,
has drown more attention in a wide range of uses [20]. Electro-
chemical amperometric immunoassay with amperometric trans-
ducer is receiving intense attention because it can achieve a
relatively low detection limit and high sensitivity. The amperomet-
ric immunoassay also plays an increasing role in immunosensors.
Thus, looking for a novel immobilization method for amperometric
immunosensor with great improvement in sensitivity, selectivity,
and response time is of considerable interest.
Advanced material based on inorganic and organic nanoparti-
cles is nowadays one of the key research fields of today material
science. Among these nanomaterials, noble metal nanoparticles
have been used to fabricate biosensor owing to their excellent
properties to immobilize biomolecules (such as enzyme, anti-
body, antigen, DNA, RNA) [21–24]. The nano-Au has drawn much
attention in constructing electrical and optical sensors due to
their small size and correspondingly unique optical, electronic,

0022-0728/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.jelechem.2008.10.031
 Y.-R. Yuan et al. / Journal of Electroanalytical Chemistry 626 (2009) 6–13
and catalysis properties. The nano-Au not only can offer a micro-
environment similar to nature and retain the bioactivity of the
immobilized biomolecules, but also can provide a high surface
to volume ratio and enhance the electron transfer kinetics by giv-
ing more freedom energy to the immobilized biomolecular in ori-
entation which makes active sites closer to conducting electrode
and permit the biomolecules to orient in conformation more
favorable for direct electron transfer [25,26]. In our present work,
a nano-Au layer on the electrode was prepared by using a simple
method – electrochemical reduction of HAuCl4 solution that
agrees with literatures [27–29]. The nano-Au film formed by this
method can be achieved in a relatively short time and can pro-
vide a stable and rough surface that is more favorable to immo-
bilize biomolecules.
Metal hexacyanoferrate complexes (MHCFs) modified elec-
trodes have attracted considerable interest due to their potential
application in many fields [30,31]. Among PB analogues, nickel
hexacyanoferrate exhibits attractive redox properties in term of
charge compensation [32]. In addition, it has been to the electro-
catalysis and determination of organic and inorganic species [33].
So far, little attention has paid to use nickel hexacyanoferrate
nanoparticles (NiHCFNPs) to construct biosensors. NiHCFNPs is
of lager surface area, a high surface to volume ratio, and high
affinity to nano-Au due to existing strong interaction between –
NH2, –SH, –CN, and nano-Au [34–37]. It will be an important
improvement to modify NiHCFNPs onto nano-Au layer for biosen-
sing application. In the present work, NiHCFNPs was synthesized
by simply mixing nickel ions and hexacyanoferrate (K3Fe(CN)6) at
room temperature. After that, NiHCFNPs as a good electroactive
substance at the same time was immobilized on the electrode,
which greatly simplifies the immunoassay system. However,
many amperometric immunoassay techniques need an electroac-
tive substance to analytical system, which lead to more complex
immunoassay system and increased analytical time and expense
[38,39].
Recently, we also developed some electrochemical immunosen-
sors based on nanomaterials with favorable biocompatibility for
the determination of CEA [40–42]. Sensitivities of the immunosen-
sors need to be improved. In the present investigation, we simul-
taneously took advantages of two kinds of nanomaterial (nano-
Au and NiHCFNPs) with good properties, tried to develop a simple
and sensitive immobilization strategy for constructing a compati-
ble amperometric immunoassay. At first, nano-Au layer with
favorable biocompatibility and large surface area was formed by
simple electrochemical reduction method. Secondly, NiHCFNPs as
a good electroactive substance was self-assembled on nano-Au
layer by strong interaction between CN (NiHCFNPs) and nano-
Au [34–37]. Thus, the second nano-Au layer was again self-assem-
bled onto NiHCFNPs by simply electrochemical reduction, which
offers a biocompatible interface to adsorb anti-CEA by chemical
adsorption between nano-Au and NH2 of anti-CEA. In addition,
nano-Au layer formed by this method can offer a stable, rough
and compact surface, which can immobilize more amount anti-
body. There are two novel strategies in designing the immunosen-
sor as follows: First, promising polymeric inorganic compound
NiHCFNPs was not only an excellent electroactive substance, but
also could provide a favorable interface for nano-Au immobiliza-
tion, which greatly improve performance of immunosensor. The
nano-Au and NiHCFNPs both with excellent properties were
simultaneously used to design an accuracy, sensitivity and stabil-
ity immunosensor. Second, the preparation of immunosensor is
simple, easy controlled and less time consuming. The performance
and factors influencing the immunosensor’s performance were
also investigated in detail. The proposed immunosensor was ap-
plied to determine CEA in human serum samples with satisfactory
results.
7
2. Experimental
2.1. Reagents and materials
CEA and anti-CEA (Biocell Company, Zhengzhou, China), Bovine
serum albumin (BSA, 96–99%), gold chloride (HAuCl4) and sodium
citrate (Sigma, St. Louis, MO, USA), K3Fe(CN)6 and NiCl26H2O
(Chemical Reagent Co, Sichuan, China). All other chemicals and sol-
vents (analytical grade, regular sources), Bi-distilled water was
employed throughout this study. Phosphate buffered solutions
(PBS) at various pH were prepared using 0.02 M Na2HPO4 and
0.02 M KH2PO4 stock solution. The supporting electrolyte was
0.1 M KCl. The CEA was stored in the frozen state, and its standard
solutions were prepared freshly with bi-distilled water when in
use.
2.2. Apparatus
CHI 610A electrochemistry workstation (Shanghai CH Instru-
ments, China), Model IM6e (ZAHNER Elektrick, Germany), pH me-
ter and digital ion analyzer (Model PHS-3C, DaPu Instruments,
Shanghai, China), transmission electron microscopy (TEM) (TECNAI
10, PHILIPS, Holand), S-3400 scanning electron microscope (SEM)
(Hitashi High Technologies Corporation, Japan).
2.3. Preparation of nickel hexacyanoferrates nanoparticles (NiHCFNPs)
The synthesis of NiHCFNPs was according to the literature [43]
with a little modification: 70 mL of 0.01 M NiCl2 aqueous solution
was drop by drop added to 70 mL of 0.05 M K3Fe(CN)6 aqueous
solution containing 0.05 M KCl under stirring. After finished addi-
tion, the mixture solution was vigorously agitated for 5 min. Fol-
lowing that, the mixture solution was immediately centrifuged,
and washed with bi-distilled water for several times. Then dried
NiHCFNPs overnight in a vacuum at room temperature and finally
gave a powered substance.
2.4. Preparation of the immunosensor
The glassy carbon electrode (GCE) (U = 4 mm) was first polished
to a mirror finish respectively with 1.0, 0.3 lm alumina slurry, fol-
lowed by rinsing thoroughly with bi-distilled water after each pol-
ishing step. After that, the electrodes were successively sonicated
in 1:1 nitric acid, ethanol, bi-distilled water. Following dried in air.
The electrode was modified immediately after the cleaning
steps. The preparation of nano-Au layer was constructed by im-
mersed the cleaning electrode in HAuCl4 aqueous solution
(2 mg mL1) and applied constant potential 0.2 V for 60 s. It
was rinsed with a copious amount of water and a yellow nano-
Au layer can be seen.
Before the step of modification, 0.05 M NiHCFNPs aqueous solu-
tion need be prepared firstly. When NiHCFNPs was dispersed in bi-
distilled water, its aqueous solution was very stable and no precip-
itate was observed after 2 months when stored at 4 °C. A 10 lL
NiHCFNPs aqueous solution (0.05 M) was dropped onto the surface
of nano-Au layer formed right now. Immediately transfer the elec-
trode to refrigerator and keep at 4 °C for about 2 h for NiHCFNPs
film dry. When NiHCFNPs film dried, the electrode was washed
thoroughly with bi-distilled water.
The NiHCFNPs/nano-Au modified electrode was soaked in 3 mL
HAuCl4 aqueous solution (2 mg mL1) and applied constant poten-
tial 0.2 V for 30 s. Following that, the modified electrode washed
with bi-distilled water. Subsequently, the nano-Au/NiHCFNPs/
nano-Au modified electrode was immersed in anti-CEA solution at
4 °C for about 12 h. Finally, the proposed electrode was incubated
8 Y.-R. Yuan et al. / Journal of Electroanalytical Chemistry 626 (2009) 6–13

Scheme 1. Schematic diagram of the stepwise immunosensor construction process: (a) electrochemical reduction formation of nano-Au; (b) assembled the NiHCFNPs; (c)
formation of nano-Au film; (d) anti-CEA loading; and (e) blocked nonspecific sites with BSA.

in BSA for 1 h at 37 °C. The accomplished immunosensor was

stored at 4 °C when not in use. The schematic diagram of the step-
wise self-assemble procedure of the immunosensor was shown in
Scheme 1.

2.5. Experiment measurement

The electrochemical impedance technique is employed to de-

tect impedance change and thickness of surface of the electrode
during self-assemble (SA) process. The stepwise immunosensor
fabrication processes were characterized by using electrochemical
impedance measurements (EIS). It was carried out in the presence
of a 5 mM FeðCNÞ4 6 =
3
as a redox probe.
Electrochemical experiments were performed in a conventional
three-electrode electrochemical cell. A three-electrode electro-
chemical cell contained a platinum wire auxiliary electrode, a sat-
urated calomel reference electrode (SCE) and the modified glassy
carbon electrode (GCE) (U = 4 mm) as working electrode. The CV
measurement was taken from 0.0 to 0.8 V (vs. SCE) at 50 mV s1
in 0.02 M PBS (pH = 7.0) at room temperature.
Fig. 1. The TEM images of nickel hexacyanoferrates nanoparticles (NiHCFNPs).
The amperometric detection is based on the change in the
amperometric response (4ip) before and after antigen-antibody
reaction. When the antibody immobilized on the electrode has was about 10 nm. A relatively uniform structure appeared in TEM
bound with antigen, the complex of antigen–antibody coating on of NiHCFNPs. Moreover, the surface morphology of nano-Au,
the surface of the electrode acts as inert substance and badly NiHCFNPs and anti-CEA on the electrode was characterized with
blocks the electron transfer tunnel to the electrode surface. More- SEM. As shown in Fig. 2a, TEM of nano-Au photograph displayed
over, with increasing the concentration of antigen, the amount of a chemically relative congregate surface microstructure, and the
antigen–antibody complex also increases, which result in less tun- size of nano-Au is about 200 nm. The aggregates of NiHCFNPs on
nels of electron transfer and the decrease of current signal of elec- the electrode surface showed a very narrow particle size distribu-
troactive substance. Meantime, the peaks current signal of tion (Fig. 2b); when gold nanoparticles were again immobilized on
electroactive substance decrease directly proportional to the con- the surface of NiHCFNPs, a relatively uniform distribution of nano-
centration of antigen that has reacted with antibody immobilized Au was obtained (Fig. 2c). Fig. 2d showed images of anti-CEA/nano-
on the electrode surface. Au/NiHCFNPs/nano-Au. Big biomolecule anti-CEA displayed hill
like microstructure.

3. Results and discussion 3.2. Electrochemical characterization of the immunosensor

3.1. Morphology of NiHCFNPs
The morphology of NiHCFNPs is an important factor affecting
the performance of an immunosensor. TEM is an effective method
to provide the information on particle size and shape. Fig. 1 exhib-
ited the microstructure of NiHCFNPs. The average size of NiHCFNPs
CV is an effective and convenient method for probing the fea-
ture of the modified electrode surface. Thus, CV was selected as a
marker to investigate electrochemical behavior after each assem-
bly step. CVs of the different modified stepwise of immunosensor
are shown in Fig. 3. No peak was observed at the bare GCE as the
lack of electroactive substance (Fig. 3a). It can be observed a little
 Y.-R. Yuan et al. / Journal of Electroanalytical Chemistry 626 (2009) 6–13 9

Fig. 2. The typical SEM images of (a) nano-Au modified surface obtained by electrochemical reduction of HAuCl4 solution; (b) NiHCFNPs/nano-Au surface; and (c) nano-Au/
NiHCFNPs/nano-Au surface; (d) anti-CEA/nano-Au/ NiHCFNPs/nano-Au surface.

300 creased greatly and appeared a well defined CV, which can attri-
d bute to NiHCFNPs be an excellent electroactive substance and
favorably enhance transfer of the electron (Fig. 3c). When the
200 c NiHCFNPs/nano-Au modified electrode was further modified by
e the nano-Au, the peak current increased a lot owing to noble metal
100 f Au nanoparticles layer could form more electron transfer tunnels
that facilitated electron transmission (Fig. 3d). Peak current de-
creased clearly after the anti-CEA was adsorbed on the nano-Au
I/µA

0
layer (Fig. 3e), which suggested the big molecular protein anti-
CEA severely blocks the electron transfer tunnel and reduced effec-
-100 g tive area of electron transfer. Subsequently, when BSA was used to
a
b block nonspecific sites, the peak current further decreased (Fig. 3f).
a The reason may be that the big biomolecular BSA hinders the tun-
-200
nel of electron transfer to the electrode surface. After the immuno-
sensor incubated in 20.0 ng mL1 of CEA for 20 min, a dramatically
-300 decrease in current is observed (Fig. 3g), that contribute to the for-
0.0 0.2 0.4 0.6 0.8 mation of immunocomplex retardation of the electron transfer
E/V tunnel.
The CVs of the resulting immunosensor incubated in
Fig. 3. Cyclic voltammograms of the different electrodes measured in 0.02 M PBS
(pH7.0) at 25 °C: (a) bare GCE; (b) nano-Au modified electrode; (c) NiHCFNPs/nano- 20.0 ng mL1 CEA in 0.02 M PBS at different scan rates ranging
Au modified electrode; (d) nano-Au/NiHCFNPs/nano-Au modified electrode; (e) from 20 mV s1 to 250 mV s1 was investigated. As Fig. 4 shown,
anti-CEA/nano-Au/NiHCFNPs/nano-Au modified electrode; (f) BSA/ anti-CEA/ nano- the potential and peak currents are dependent on the scan rate.
Au/NiHCFNPs/nano-Au modified electrode (the resulting immunosensor); and (g) It can be also observed that the redox peak currents were propor-
the resulting immunosensor incubated in 20.0 ng mL1 CEA. The potential scan rate
was 50 mV s1.
tional to the scan rate, shown in the inset, suggesting that a surface
limited process.
Moreover, the impedance changes of immunosensor surface in
increase in background current due to increase of the double layer the fabrication process and the formation of antigen–antibody
capacitance from nano-Au modification (Fig. 3b). However, after complex can be observed by EIS. Thus, the stepwise construction
NiHCFNPs was modified onto the nano-Au layer, peak current in- process of the immunosensor was characterized by EIS. Fig. 5a,
10 Y.-R. Yuan et al. / Journal of Electroanalytical Chemistry 626 (2009) 6–13

(Fig. 5e). The result is consistent with the fact that the hydrophobic
layer of protein insults the conductive support and hinders the
200 interfacial electron transfer. Ret increased in the same way after
I/µA

BSA was used to block nonspecific sites (Fig. 5f), which may attri-
bute to the same reason with loading the antibody. After the
100
resulting immunosensor was incubated in 20.0 ng mL1 of CEA,
v Ret further increases (Fig. 5g), which indicates the formation of
I/µA

0 hydrophobic immunocomplex layer embarrassing the electron

transfer.

-100 3.3. Optimization of analytical conditions for the immunoassay

3.3.1. Reduction time of nano-Au

-200 Time of electrochemical reduction HAuCl4 has great effect on
performance. Less reduction time, nano-Au layer formed is very
thin and amount of nano-Au is few, which directly affect the
0.0 0.2 0.4 0.6 0.8 immobilization of NiHCFNPs and further influence performance
E/V of the immunosensor. The more reduction time, the more amount
of nano-Au, but if reduction time exceeding some value might re-
Fig. 4. CVs of the immunosensor incubated in 20 ng mL1 CEA solution at various
scan rates in 0.02 M PBS (pH = 7.0) at 25 °C. Scan rates from inner to outer: 20, 50,
sult in excessive compactness film of Au particles, which seriously
80, 100, 120, 150, 180, 200, and 250 mV s1. The inset shows the dependence of the obstructs electron transfer tunnels and further affected the re-
redox peak currents on scan rates. sponse of the immunosensor. The effect of first nano-Au layer
was researched in FeðCNÞ4 6 =
3
solution when reduction time ran-
ged from 20 s to 100 s. As Fig. 6a shown, at 60 s, the current re-
-1200 sponse reached a maximum value. A 60 s was chosen as the
reduction time of first nano-Au layer. Moreover, the reduction time
of second nano-Au layer also need to be optimized. The influence of
-1000
second nano-Au layer was studied in 0.02 M PBS with the reduc-
tion time ranged from 10 s to 60 s. Why a 0.02 M PBS was selected
-800 as characterization solution? The reason might be that before mod-
ified the second nano-Au layer, NiHCFNPs as an excellent electro-
zim/ohm

active substance had been assembled on the nano-Au/GCE, which

-600
d offered enough sensitivity to optimize the reduction time of second
-400 a f
b e
c 30 s was used as the reduction time of second nano-Au layer.
-200
0 gated between 5.0 and 9.0 in 0.02 M PBS. As shown in Fig. 7, the
0 200 400 600 800 1000 1200 1400
Zre/ohm
Fig. 5. Electrochemical impedance spectroscopy (EIS) of the different electrodes
measured in 5 mM FeðCNÞ4 6 =
3
at 25 °C: (a) bare GCE; (b) nano-Au modified
electrode; (c) NiHCFNPs/nano-Au modified electrode; (d) nano-Au/NiHCFNPs/nano-
Au modified electrode; (e) anti-CEA/ nano-Au/NiHCFNPs/nano-Au modified elec-
trode; (f) BSA/anti-CEA/ nano-Au/NiHCFNPs/nano-Au modified electrode (the
resulting immunosensor); and (g) the resulting immunosensor incubated in
20.0 ng mL1 CEA.
showed EIS of the bare GCE. A very small semicircle at high fre-
quencies and linear part at low frequencies was observed, which
agrees with literature [44], suggesting very low Ret to redox probe
FeðCNÞ4 3
6 = . After the bare electrode was modified by nano-Au
layer (Fig. 5b), the EIS exhibits a very low interfacial Ret, indicating
the resulting nano-Au layer greatly enhanced electron transfer of
the redox probe. When NiHCFNPs again modified the nano-Au/
GCE, it can be found that interfacial Ret increases but less than that
of bare GCE (Fig. 5c). The reason may be that the microstructure of
NiHCFNPs gives birth to some barrier obstructing the electron
transfer. Then the Ret decreases obviously when nano-Au was again
assembled on the NiHCFNPs (Fig. 5d), which maybe ascribed to
nano-Au be a kind of conductive material and its conductivity be
better than that of NiHCFNPs. Subsequently, when the anti-CEA
was loaded on the surface of nano-Au, Ret increases dramatically
nano-Au layer. The result shows that the reduction time of second
nano-Au layer arrive at maximum response at 30 s (Fig. 6b). Thus
g
3.3.2. pH of working buffer
The effect of pH on the immunosensor behavior was investi-
1600 peak current increases with increasing pH value from 5.0 to 7.0
and decreases when further increases pH value. The test results
show that the maximum current response occurs at pH 7.0. So a
pH 7.0 PBS is chosen throughout this study.
3.3.3. Incubation temperature and time
To our knowledge, the optimal temperature of immunoreaction
is 37 °C. It is also the temperature of human being. But taking con-
sideration of the activity of biomolecules and life-time of immuno-
sensor, 25 °C was selected as incubation temperature for whole
assays. At this temperature, the immunosensor was incubated in
a standard solution of CEA with known constant concentration of
20.0 ng mL1 for different time. As Fig. 8 shown, the response cur-
rent of immunosensor was rapidly up with the duration of incuba-
tion time from 2 min to 20 min, and then leveled off slowly, which
indicated an equilibrium state reached. Thus, 20 min was used for
incubation time of all the subsequent assays.
3.4. Performance of immunosensor
3.4.1. Amperometry response of the proposed immunosensor
After the immunosensor was incubated in a standard CEA solu-
tion with constant concentration 20.0 ng mL1 for 20 min at 25 °C,
the amperometry response can be obtained in 0.02 M PBS
(pH = 7.0). As shown in Fig. 3g, peak current decreased as expected;
correspondingly, in Fig. 5g, the Ret increased remarkably. When the
 Y.-R. Yuan et al. / Journal of Electroanalytical Chemistry 626 (2009) 6–13 11

174
a b
160

171

156
168

I/µA
I/µA

165
152

162

148
159
20 40 60 80 100 10 20 30 40 50 60
t/sec t/sec
Fig. 6. The effect of electrochemical reduction time of nano-Au on response currents of modified electrode characterized by CV in different solution at 25 °C: (a) the electrode
of nano-Au/GCE in 5 mM FeðCNÞ4 6 =
3
; (b) the electrode nano-Au/NiHCFNPs/nano-Au in 0.02 M PBS (pH = 7.0). The potential scan rate was 50 mV s1.

-68
112

-72
105

-76
I/µA

I/µA

98

-80
91

-84
84

5 6 7 8 9
0 5 10 15 20 25
pH
Time/min
Fig. 7. The influence of pH of the PBS on the response of the immunosensor in
0.02 M PBS at 25 °C. The scan rate was 50 mV s1. Fig. 8. The effects of incubation incubation time; CV determination was performed
in 0.02 M PBS (pH = 7.0) at 25 °C. The scan rate was 50 mV s1.

proposed immunosensor was incubated in standard solution of

CEA with different concentration with other conditions the same
as described above, the calibration plot of the determination of
CEA can be gained along with obtaining CVs of different concentra-
tion of CEA. Fig. 9 illustrates that the calibration plot for the detec-
tion of CEA. The amperometry response linearly increased with the
increase of CEA concentration in two ranges from 0.5 to
10.0 ng mL1 and from 10.0 to 160.0 ng mL1, with a detection lim-
it 0.1 ng mL1 at a signal to noise ratio of 3. The linear slopes were
2.6696 and 0.1704 lA (ng mL1)1, and the correlation coefficients
were 0.9874 and 0.9955, respectively. Why calibration curve pre-
sented two slopes? The reason may be: the amount of antibody
immobilized on the immunosensor is some certain. When the
immunosensor is used to detect antigen, antigen can bind with ac-
tive sites of antibody which was immobilized on the immunosen-
sor. With increase of antigen concentration, active sites of antibody
on the immunosensor become fewer and fewer. As a result, when
the immunosensor is used to determine higher concentration of
antigen, the sensitivity can decline. There are literatures [45–50]
whose calibration curves refer two slopes. The proposed method
is thus absolutely means to quantify CEA in the sample, which indi-
cated that this immunosensor has great potential to be used in
clinical diagnosis.
3.4.2. Selectivity against interferences
Selective determination of target analytes plays an important
role in analyzing biological samples in situ without separation.
The effect of substances that might interfere with the response of
immunosensor was investigated. The evaluation selectivity of the
immunosensor was carried out by incubated the immunosensor
in 20.0 ng mL1 CEA containing some potential co-existed species
with CEA, such as alpha-fetoprotein (AFP), hepatitis B surface anti-
gen (HBsAg), human chorionic gonadotrophin (HCG), BSA, ascorbic
acid, dopamine, L-glucose, L-cystein, L-lysine, serine, arginine, his-
tidine, leucine. The degree of interference from substances de-
scribed above can be judged from the value of the current ratio.
In this study, the current ratios can be obtained from the current
reading of the immunosensor in 0.02 M PBS (pH = 7.0) with the
immunosensor incubated in 20.0 ng mL1 CEA and 20.0 ng mL1
interfering substances versus the current reading with the immu-
nosensor incubated in 20.0 ng mL1 CEA. The test results were
12 Y.-R. Yuan et al. / Journal of Electroanalytical Chemistry 626 (2009) 6–13

Table 2
-30 Comparison of serum levels by using two methods.

Serum samples 1 2 3 4 5 6
-40 This method (ng mL ) 1
1.8 4.6 8.3 15.7 29.2 56.4
CLIA (ng mL1) 1.7 4.3 8.5 16.4 28.6 58.3
Relative deviation (%) 5.9 7.0 2.4 4.3 2.1 3.3
-50
I/µA

-60 Table 3
Comparision of the performance of electrochemical immunosensors for the determi-
nation of CFA.
-70
Type of Linear ranges Incubation DL Reference
immunosensor (ng mL1) time (min) (ng mL1)
-80
Electrochemical 0.5–3.0 and 3.0– 30 0.4 [45]
120
-90 Electrochemical 25–150 60 1.2 [46]
0 30 60 90 120 150 180 Electrochemical 0.5–25 35 0.22 [38]
-1 Electrochemical 0.5–80.0 10 0.14 [47]
C/(ng mL ) Quartz crystal 3.0–50 35 1.5 [48]
microbalance
Fig. 9. Calibration plot of the immunosensor for detection of CEA. The amperometry Piezoelectric 66.7–466.7 20 66.7 [49]
measurement was carried out by CVs in 5 mL 0.02 M PBS (pH = 7.0) at 25 °C. The Chemiluminescent 1.0–100.0 100 0.53 [50]
scan rate was 50 mV s1.
Chemiluminescent 1.0–120 20 0.6 [51]
Electrochemical 0.5–10 and 20 0.1 This work
10.0–160
listed in Table 1. The results suggested that the substance selected
did not interfere remarkably. However, these non-specific species
might result in current shift or current change in certain error
range. cretely, the proposed immunosensor was first incubated in 1 mL
human serum samples for 20 min at 25 °C. Then, simply washed
3.4.3. Stability, repeatability, and reproducibility with bi-distilled water. Subsequently, the CV measurement was
The successive stability of the immunosesor was researched. carried out in 0.02 M pH7.0 PBS. The test results were listed in Ta-
After the imunosensor was successively scanned for 80 circles in ble 2. The results may imply that the detectable concentration of
working buffer, a RSD of 4.3% was gained. The immunosensor CEA in this assay satisfies the requirement of clinical analysis
had acceptable storage stability with 89% and 75% of initial re- and that may provide a feasible alternative tool for the direct
sponse remaining after the storage periods of 14 days and 30 days determination of CEA in real serum.
in 0.02 M PBS (pH = 7.0) at 4 °C. When the immunosensor was
measured for 11 times in working buffer, respectively, it yielded 3.4.5. Characteristics comparison with other electrochemical CEA
a 1.9% RSD. immunosensors
The reproducibility of immunosensors plays a key factor for The analytical performance of the developed CEA immunosen-
developing a practical immunosensor. A regeneration study of sors has been compared with those of other CEA immunosensors
the prepared immunosensor was implemented by using regenera- reported in the literatures [51–57]. Characteristics such as linear
tion solutions to break the antibody–antigen linkage. In this study, ranges, detection limit (DL) and incubation time of every immuno-
the two different immunosensors that had been incubated in sensor were summarized in Table 3. As Table 3 shown, the linear
20.0 ng mL1 CEA for 20 min, were immersed in a stirred 0.2 M gly- ranges, DL, and incubation time of the proposed immunosensor
cine-hydrochloric acid (Gly-HCl) solution (pH = 2.8) and a stirred are relative good.
4 M urea solution for 5 min, respectively. RSD of 4.7% and 6.2%
was acquired, respectively (4 regenerations and measurement). 4. Conclusions

3.4.4. Preliminary analysis of real human serum samples In the present paper, a novel approach for construction of an
To demonstrate the application of the proposed immunosensor immunosensor was developed based on BSA/anti-CEA/nano-Au/
for the detection of the CEA in human serum, six serum samples NiHCFNPs/nano-Au complex matrix modified electrode. The
with different infection degrees from the Ninth people’s Hospital immunosensor can be used for directly monitoring the concentra-
of Chongqing, were examined by the proposed electrochemical tion of CEA in serum samples. The amperometric immunoassay
immunoassay and chemiluminescene immunoassay (CLIA). Con- strategy offered several advantages including simple fabrication,
high sensitivity, low detection limit, satisfactory regenerations,
quantitative immunobinding reactions, no need for labeled re-
Table 1
Possible interferences tested with the proposed electrochemical immunosensor. agents, cost-effectiveness. Thus, this method could be further
developed for practical clinical diagnosis of serum CEA level. More-
Possible interferences Current Possible Current
over, anti-CEA/nano-Au/NiHCFNPs/nano-Au efficiently immobilize
ratio interferences ratio
antibody and could be used in the preparation of other ampero-
Alpha-fetoprotein 0.89 L-Cysteine 1.04
metric immunosensors for detection of clinically or environmental
Hepatitis B surface antigen 0.92 L-Lysine 0.96
Human chorionic 0.93 Histidine 1.03 interested biospecies.
gonadotrophin
Bovin serum antigen 1.05 Leucine 1.01 Acknowledgments
L-Glutamate 1.03 Ascorbic acid 0.94
Serine 1.02 Dopamine 0.95
This work was supported by the National Natural Science Foun-
Arginine 0.98
dation of China (No. 20675064), the Chinese Education Ministry
 Y.-R. Yuan et al. / Journal of Electroanalytical Chemistry 626 (2009) 6–13
Foundation for Excellent Young Teachers (No. 2002-40), the Na-
tional Science Foundation of Chongqing City (CSTC-2004BB4149,
2005BB4100), and High Technology Project Foundation of South-
west University (XSGX 02), China.
References
[1] A. Szilvas, A. Blazovices, G. Szekely, E. Dinya, J. Feher, G. Mozsik, J. Physiol. Paris
95 (2001) 247.
[2] D. Faraggi, A. Kramar, Urol. Oncol. 5 (2000) 211.
[3] W. Limbut, P. Kanatharana, B. Mattaiasson, P. Asawatreratanakul, P.
Thavarungkul, Anal. Chim. Acta 561 (2006) 55.
[4] J.E. Shively, J.D. Beatty, Crit. Rev. Oncol. Hematol. 2 (1985) 355.
[5] F. Naghibalhossaini, P. Ebadi, Cancer Lett. 234 (2006) 158.
[6] S.P. Prete, L. Rossi, P.P. Correale, M. Turriziani, S. Baier, G. Tamburrelli, L.D.
Vecchis, E. Bonmassar, A. Aquino, Pharmacol. Res. 52 (2005) 167.
[7] Y. Sakao, S. Tomimitsu, Y. Takeda, M. Natsuaki, T. Itoh, Eur. J. Cardio-thorac. 25
(2004) 520.
[8] A. Aquino, V. Formica, S.P. Prete, P.P. Correale, M.C. Massara, M. Turriziani, L.D.
Vecchis, E. Bonmassar, Pharmacol. Res. 49 (2004) 383.
[9] K. Kashiwabaraa, H. Nakamurab, T. Yokoic, Clin. Chim. Acta 303 (2001) 25.
[10] K. Matsuoka, S. Sumitomo, N. Nakashima, D. Nakajima, N. Misaki, Eur. J.
Cardio-thorac. 32 (2007) 435.
[11] M. Juweid, L.C. Swayne, R.M. Sharkey, R. Dunn, A.D. Rubin, T. Herskovic, D.M.
Goldenberg, Gynecol. Oncol. 67 (1997) 259.
[12] K. Sarkar, A. Bose, S. Laskar, S.K. Choudhuri, S. Dey, P.K. Roychowdhury, R.
Baral, Int. Immunopharmacol. 7 (2007) 306.
[13] J.H. Lina, F. Yan, H.X. Ju, Clin. Chim. Acta 341 (2004) 109.
[14] N.R. Gonzales, P. Schuck, J. Schlom, S.V.S. Kashmiri, J. Immunol. Methods 268
(2002) 197.
[15] P.J. Yazaki, L. Shively, C. Clark, C.W. Cheung, W. Le, B. Szpikowska, J.E. Shively,
A.A. Raubitschek, A.M. Wu, J. Immunol. Methods 253 (2001) 95.
[16] Y.P. Luo, D.O. Hagan, H. Zhou, M. Singh, J. Ulmer, R.A. Reisfeld, F.J. Primus, R.
Xiang, Vaccine 21 (2003) 1938.
[17] B. Zhang, X. Zhang, H.H. Yan, S.J. Xu, D.H. Tang, W.L. Fu, Biosensors Bioelectron.
23 (2007) 19.
[18] G. Davis, Biosensors 2 (1986) 101.
[19] F. Darain, S.U. Park, Y.B. Shim, Biosensors Bioelectron. 20 (2003) 773.
[20] Z.P. Aguilar, W.R. Vandaveer, I. Fritsch, Anal. Chem. 74 (2002) 3321.
[21] A.P. Baioni, M. Vidotti, P.A. Fiorito, E.A. Ponzio, S.I.C. Torresi, Langmuir 23
(2007) 6796.
[22] G. Zhao, J.J. Feng, Q.L. Zhang, S.P. Li, H.Y. Chen, Chem. Mater. 17 (2005) 3154.
[23] B.Y. Wu, S.H. Hou, L. Huang, F. Yin, Z.X. Zhao, J.I. Anzai, Q. Chen, Mater. Sci. Eng.
C (2007), doi:10.1016/j.msec.2007.04.035.
13
[24] Y. Du, X.L. Luo, J.J. Xu, H.Y. Chen, Bioelectrochem. 70 (2007) 342.
[25] A.L. Crumbliss, S.C. Perine, J. Stonehuerner, K.R. Tubergen, J.G. Zhao, R.W.
Henkens, J.P. O’Daly, Biotechnol. Bioeng. 40 (1992) 483.
[26] S.Q. Liu, D. Leech, H.X. Ju, Anal. Lett. 36 (2003).
[27] V.C. Sanz, M.L. Mena, A.G. Cort́es, P.Y. Sede~no, J.M. Pingarŕon, Anal. Chim. Acta
528 (2005) 1.
[28] M.L. Mena, P.Y. Seden, J.M. Pingarroń, Anal. Biochem. 336 (2005) 20.
[29] V. Carralero, A.G. Cortés, P.Y. Sedeño, J.M. Pingarrón, Electroanal. 17 (2005)
289.
[30] P. Wang, Y. Yuan, X.Y. Jing, G.Y. Zhu, Talanta 53 (2001) 863.
[31] M.A. Malik, K. Miecznikowski, P.J. Kulesza, Electrochim. Acta 45 (2000) 3777.
[32] D.M. Zhou, H.X. Ju, H.Y. Chen, J. Electroanal. Chem. 408 (1996) 219.
[33] S. Zamponi, M. Berrettoni, P.J. Kulesza, K. Miecznikowski, M.A. Malik, O.
Makowski, R. Marassi, Electrochim. Acta 48 (2003) 426.
[34] K.C. Grabar, P.C. Smith, M.D. Musick, J.A. Davis, D.G. Walter, M.A. Jackson, A.P.
Guthrie, M.J. Natan, J. Am. Chem. Soc. 120 (1996) 1148.
[35] K.C. Grabar, R.G. Freeman, M.B. Hommer, M.J. Natan, Anal. Chem. 67 (1995)
735.
[36] H.Y. Gu, A.M. Yu, H.Y. Chen, J. Electroanal. Chem. 516 (2001) 119.
[37] C.C. Wang, H. Wang, Z. yang, Z.Y. Wu, G.L. Shen, R.Q. Yu, Acta Chim. Sin. 61
(2003) 608.
[38] J. Wu, J.H. Tang, Z. Dai, F. Yan, H.X. Ju, N.E. Murrc, Biosensors Bioelectron. 22
(2006) 102.
[39] Y.T. Shi, R. Yuan, Y.Q. Chai, X.L. He, Electrochim. Acta 52 (2007) 3518.
[40] X.L. Li, R. Yuan, Y.Q. Chai, L.Y. Zhang, Y. Zhuo, Y. Zhang, J. Biotechnol. 123
(2006) 356.
[41] Y.T. Shi, R. Yuan, Y.Q. Chai, M.Y. Tang, X.L. He, J. Electroanal. Chem. 604 (2007)
9.
[42] D.P. Tang, R. Yuan, Y.Q. Chai, J. Phys. Chem. B 110 (2006) 11640.
[43] M.H. Yang, Y.H. Yang, F.L. Qu, Y.S. Lu, G.L. Shen, R.Q. Yu, Anal. Chim. Acta 571
(2006) 211.
[44] S.A. Yuan, S.S. Hu, Electrochim. Acta 49 (2004) 4287.
[45] Z. Dai, S. Serban, H. Ju, N. Murr, Biosens. Bioelectron. 22 (2007) 1700.
[46] J. Chen, J. Tang, F. Yan, H. Ju, Biomaterials 27 (2006) 2313.
[47] X. Li, R. Yuan, Y. Chai, L. Zhang, Y. Zhuo, Y. Zhang, J. Biotechnol. 123 (2006) 356.
[48] Z. Dai, J. Chen, F. Yan, H. Ju, Cancer Detect. Prevent. 29 (2005) 233.
[49] H. Yu, F. Yan, Z. Dai, H. Ju, Anal. Biochem 331 (2004) 98.
[50] Z. Dai, F. Yan, H. Yu, X. Hu, H. Ju, J. Immunol. Methods 287 (2004) 13.
[51] F. Tan, F. Yan, H. Ju, Electrochem. Commun. 8 (2006) 1835.
[52] M.S. Wilson, Anal. Chem. 77 (2005) 496.
[53] Y. Zhuo, R. Yuan, Y. Chai, A. Sun, Y. Zhang, J. Yang, Biomaterials 27 (2006) 5420.
[54] D.Q. Tang, D.J. Zhang, D.Y. Tang, H. Ai, J. Immunol. Methods 316 (2006) 144.
[55] G.Y. Shen, H. Wang, T. Deng, G.L. Shen, R.Q. Yu, Talanta 67 (2005) 217.
[56] H.S. Zhuang, J.L. Huang, G.N. Chen, Anal. Chim. Acta 512 (2004) 347.
[57] Z. Fu, H. Liu, H. Ju, Anal. Chem. 78 (2006) 6999.