A 3D graphene oxide microchip and a Au-enwrapped silica nanocomposite-based supersandwich cytosensor toward capture and analysis of circulating tumor cells

Abstract

Determination of the presence and number of circulating tumor cells (CTCs) in peripheral blood can provide clinically important data for prognosis and therapeutic response patterns. In this study, a versatile supersandwich cytosensor was successfully developed for the highly sensitive and selective analysis of CTCs using Au-enwrapped silica nanocomposites (Si/AuNPs) and three-dimensional (3D) microchips. First, 3D microchips were fabricated by a photolithography method. Then, the prepared substrate was applied to bind graphene oxide, streptavidin and biotinylated epithelial-cell adhesion-molecule antibody, resulting in high stability, bioactivity, and capability for CTCs capture. Furthermore, horseradish peroxidase and anti-CA153 were co-linked to the Si/AuNPs for signal amplification. The performance of the cytosensor was evaluated with MCF7 breast cancer cells. Under optimal conditions, the proposed supersandwich cytosensor showed high sensitivity with a wide range of 10(1) to 10(7) cells per mL and a detection limit of 10 cells per mL. More importantly, it could effectively distinguish CTCs from normal cells, which indicated the promising applications of our method for the clinical diagnosis and therapeutic monitoring of cancers.

Full text

Nanoscale
PAPER
Cite this: Nanoscale, 2015, 7, 16354
Received 23rd July 2015,
Accepted 25th August 2015
DOI: 10.1039/c5nr04798f
www.rsc.org/nanoscale
A 3D graphene oxide microchip and a
Au-enwrapped silica nanocomposite-based
supersandwich cytosensor toward capture and
analysis of circulating tumor cells
Na Li,
a,b
Tingyu Xiao,
a,b
Zhengtao Zhang,
a
Rongxiang He,
a,b
Dan Wen,
a
Yiping Cao,*
a,b
Weiying Zhang*
a,b
and Yong Chen
a,b,c
Determination of the presence and number of circulating tumor cells (CTCs) in peripheral blood can
provide clinically important data for prognosis and therapeutic response patterns. In this study, a versatile
supersandwich cytosensor was successfully developed for the highly sensitive and selective analysis of
CTCs using Au-enwrapped silica nanocomposites (Si/AuNPs) and three-dimensional (3D) microchips.
First, 3D microchips were fabricated by a photolithography method. Then, the prepared substrate was
applied to bind graphene oxide, streptavidin and biotinylated epithelial-cell adhesion-molecule antibody,
resulting in high stability, bioactivity, and capability for CTCs capture. Furthermore, horseradish peroxidase
and anti-CA153 were co-linked to the Si/AuNPs for signal amplification. The performance of the cyto-
sensor was evaluated with MCF7 breast cancer cells. Under optimal conditions, the proposed super-
sandwich cytosensor showed high sensitivity with a wide range of 10
1
to 10
7
cells per mL and a detection
limit of 10 cells per mL. More importantly, it could effectively distinguish CTCs from normal cells, which
indicated the promising applications of our method for the clinical diagnosis and therapeutic monitoring
of cancers.
Introduction
Cancer is considered a worldwide mortal sickness and has
become a major public concern. Metastasis is the most
common cause of cancer-related deaths in patients with solid
tumors. During the progression of metastasis, tumor cells
detach from the solid primary tumor, enter the blood stream
and travel to different tissues of the body.
1
These “break-away”
tumor cells in the peripheral blood are known as circulating
tumor cells (CTCs). In addition to conventional diagnostic
imaging and serum marker detection, CTCs analysis as a
“liquid biopsy”can provide valuable information on prognosis,
facilitate monitoring of systemic anticancer therapy, and help
in identifying appropriate therapeutic targets.
2
However, as the
concentration of CTCs is generally extremely low (a few to
hundreds per milliliter) in the bloodstream, detection and
characterization of CTCs present a tremendous technical chal-
lenge.
3,4
Over the past decade, various technology platforms
for isolating/counting CTCs have been developed using
different strategies such as flow cytometry, immune magnetic
beads, mechanical separation, or microfluidic devices.
5–8
However, the sensitivity of these emerging technologies relies
on the degree of enrichment of CTCs. Thus, the development
of a novel and sensitive platform that enhances CTCs capture,
allows for imaging of the captured CTCs, and enables quanti-
tative analysis would dramatically increase the use of CTCs in
diagnostics and prognostics.
Graphene and its derivatives have attracted considerable
attention due to their extremely large surface area, biocompati-
bility, excellent electrocatalytic activity, fast electron transfer,
high mechanical strength and high chemical stability.
9
Owing
to these extraordinary properties, graphene and its nano-
composites have been widely applied to modify a glassy carbon
electrode (GCE) for immobilizing various biomolecules.
10–14
Thus, most of the graphene based electrochemical sensors
developed to date show two designs: the assembly of two-
dimensional (2D) graphene on the GCE or multi-stacked
graphene films. Further integration of the widely available
graphene sheets as 2D-nanoscale building blocks into a func-
tional three-dimensional (3D) system is essential to extend
a
Flexible Display Mater. & Tech. Co-Innovation Center of Hubei, Institute for
Interdisciplinary Research, Jianghan University, Wuhan 430056, PR China.
E-mail: zwy2428@163.com, cyp@jhun.edu.cn
b
Key Laboratory of Optoelectronic Chemical Materials and Devices of Ministry of
Education, Jianghan University, Wuhan 430056, PR China
c
Ecole Normale Supérieure, CNRS-ENS-UPMC UMR 8640, 24 Rue Lhomond, Paris
75005, France
16354 |Nanoscale,2015,7, 16354–16360 This journal is © The Royal Society of Chemistry 2015

their biomedical applications.
15
Recently, a 3D hierarchical
nanostructured graphene oxide (GO) platform combined with
a ZnO nanorod array was successfully utilized to recognize/
capture epithelial cell-adhesion molecule (EpCAM)-expressing
cancer cells.
16
The mechanism relies on the enhanced local
topographic interactions between the substrate and nanoscale
components of the cellular surface, which improve the capture
efficiency of the target cells to a certain extent.
17–22
Herein, we present a convenient and cost-efficient 3D
microchip coated with GO and EpCAM antibody (anti-EpCAM,
Ab1) to serve as a cytosensor for the capture and analysis of
CTCs. With the remarkable development of nanotechnology,
nanomaterials have been extensively used to enhance the
sensitivity of cytosensors. Among them, gold nanoparticles
(AuNPs) are of particular interest owing to their merits of high
surface reactivity, good solubility, unique optical and electronic
characteristics as well as their excellent biocompatibility.
23–25
Thus, clever combinations of different types of functional
nanostructured materials will enable the development of multi-
functional nanomedical platforms for diagnosis and therapy.
Among the various integrated nanocomposite-based systems,
mesoporous silica-based nanostructured materials have
attracted great interest, since they exhibit low cytotoxicity and
excellent chemical stability, and their surfaces can be easily
modified.
26–29
Therefore, they are ideal platforms for con-
structing multifunctional materials that incorporate a variety
of functional nanostructured materials.
In this study, we developed a novel CTC-sensitive quantitative
detection system that integrates two functional components: (1) a
3D microchip with GO and an anti-EpCAM coating for recogniz-
ing/capturing EpCAM-expressing cells, and (2) Au-enwrapped
silica nanocomposites (Si/AuNPs) loaded with horseradish peroxi-
dase (HRP) and anti-CA153 (Ab2) to improve the selectivity of the
target cells and amplify the electrochemical detection signal. The
3D microchip arrays were initially prepared as illustrated in
Scheme 1A. The microchips were subsequently modified by GO,
streptavidin (SA), and biotinylated anti-EpCAM, respectively,
which served as 3D bioelectronic interfaces (Scheme 1B). In
addition, the Si/AuNPs were synthesized (Scheme 1C), and then
used as matrices to load the high-density signaling enzymes HRP
and Ab2. As a proof-of-concept, MCF7 breast cancer cells were
selected as model cells because of their high levels of EpCAM and
CA153 expression to demonstrate the feasibility of the supersand-
wich strategy. Finally, when sandwich-type immunoreactions were
carried out (Scheme 1D), the HRP on the surface could catalyze
the redox of H
2
O
2
in the presence of thionine, which served as
an electron transfer mediator. The proposed strategy exhibited
excellent sensitivity and selectivity, indicating its wide applica-
bility for research and clinical evaluations of cancer progression.
Experimental
Materials and reagents
The polydimethylsiloxane (PDMS) prepolymer kit was pur-
chased from Momentive Performance Materials (Waterford,
NY). The positive photoresist AZ 40XT-11D and the developer
AZ-300MIF were obtained from AZ Electronic Materials Corp.
(Philadelphia, PA). DMEM medium for cell culture was
obtained from GIBCO. 1-(3-(Dimethylamino)-propyl)-3-ethyl-
carbodiimide hydrochloride (EDC), N-hydroxysulfosuccinimide
(NHS), HPR, fluorescein diacetate (FDA) and SA were pur-
chased from Sigma (St. Louis, MO). The biotinylated goat IgG
polyclonal anti-EpCAM antibody and mouse monoclonal anti-
CA153 antibody were obtained from R&D Systems (Minneapo-
lis, MN). GO was obtained from Graphene Laboratories Inc.
(Calverton, NY). All other chemicals used in this study were of
analytical-grade. All solutions were freshly prepared using
ultrapure water (≥18 MΩ, Millipore).
Fabrication and surface modification of the 3D microchip
device
All the 3D microchip devices were designed with L-Edit soft-
ware and fabricated using a lithography fabrication technique,
which is illustrated in Scheme 1. The device was approximately
1 × 1 cm. All of the microchip masters were fabricated by spin-
coating with a positive photoresist (AZ 40XT-11D) on a silicon
substrate. Briefly, step 1 (Scheme 1A a to c): a layer of AZ
Scheme 1 (A) Schematic representation of the integrated fabrication of
polydimethylsiloxane (PDMS)-based 3D microchip arrays. (B) The micro-
chips were modified by GO, SA and biotinylated anti-EpCAM, respect-
ively. (C) Si/AuNPs were synthesized and used as nanocarriers to load
HPR and anti-CA153. (D) MCF7 cells bound to biotinylated anti-EpCAM
and HRP–Si/AuNP–Ab2 were recognized by MCF7 cells on the
microchips.
Nanoscale Paper
This journal is © The Royal Society of Chemistry 2015 Nanoscale,2015,7, 16354–16360 | 16355

40XT-11D was thrown at a clean and dry silicon wafer, which
was then exposed to ultraviolet light through a high-resolution
transparency mask (25 000 dpi). Step 2 (Scheme 1A d): a PDMS
prepolymer mixture (oligomer: the curing agent mass ratio =
10 :1) was poured over the negative grooves, which were placed
in suitable Petri dishes. Step 3 (Scheme 1A e): after curing at
80 °C for 2 h and peeling offthe PDMS, positive linear arrays
of 3D PDMS-based microchips were obtained. Scheme 1A(f )
shows the scanning electron microscopy (SEM) images
(Hitachi, S-3400N) of the prepared 3D microchip arrays.
A total of 100 mg of GO was dispersed in 200 mL ethanol
(with 0.05% chitosan) with ultrasonication for 30 min, and
then 1 mL of the GO solution was mixed with 400 mM EDC
and 100 mM NHS in 1 mL of MES buffer ( pH 5.2) for 30 min
at room temperature to activate the carboxyl groups on the sur-
faces of GO. After washing the mixture with phosphate-
buffered saline (PBS), SA (50 μgmL
−1
in PBS) was added to the
solution at 4 °C for 2 h and subsequently washed three times
with PBS to remove unreacted SA. Biotinylated anti-EpCAM
(10 μgmL
−1
in PBS) was finally added. The microchips were
incubated with the resulting solution for 1 h at room tempera-
ture and dried to fabricate a functional film.
Synthesis of Si/AuNP nanocomposites
The Si/AuNPs were prepared according to a published pro-
cedure with slight modification.
30
Briefly, alcohol (EtOH;
2600 μL), H
2
O (500 μL), and ammonia-water (NH
3
·H
2
O;
300 μL) were mixed in a round-flask with stirring for 5 min.
Tetraethyl orthosilicate (170 μL) was then added to the mixture
and stirred for 8 h at room temperature. After the reaction, the
mixture was centrifuged at 15 000gfor 10 min, rinsed succes-
sively with H
2
O and EtOH three times, and then dried to
obtain the silica particles.
The Au-coated silica complex was obtained by chemical
reduction of an Au(III) solution with hydroxylamine in the pres-
ence of suspended silica nanoparticles. Firstly, SiO
2
(8 mg) was
well dissolved in HAuCl
4
·3H
2
O (1 mL, 0.1 mg mL
−1
), and then
the solution was adjusted to pH 7–8 with 2 M NaOH, resulting
in a colour change to pink. NH
2
OH·HCl (30 μL, 0.22 M) was
added for the chemical reduction of Au(III) to Au(0), and the
colour immediately turned dark brown. The mixture was
placed in an orbital shaker at 120 rpm for 10 h followed by
resting for 10 h to discard the supernatant. The resulting
nanocomplex was resuspended in H
2
O. This procedure was
repeated and the precipitated conjugates were dispersed in
double-distilled water and stored at 4 °C.
Preparation of HRP–Si/AuNP–Ab2 conjugates
First, the Si/AuNPs were sonicated with mercaptoacetic acid
for 2 h to generate carboxylated groups, and then the mixture
was washed repeatedly with PBS. Second, the carboxylated Si/
AuNPs were added to 1.0 mL 400 mM EDC and 100 mM NHS
in MES buffer (pH 5.2), and sonicated for 30 min. The mixture
was centrifuged for 5 min at 13 000gand the supernatant was
discarded. Third, 500 μL of 400 ng mL
−1
anti-CA153 and
500 μLof80μgmL
−1
HRP were added to the mixture, which
was stirred for 4 h at room temperature. The mixture was
washed with PBS several times to remove unbound HRP
and Ab2. Finally, the particles were dispersed in 1.0 mL of
PBS (pH 7.4) containing 3% bovine serum albumin and stored
at 4 °C.
Cell culture
MCF7 cells were cultured in DMEM supplemented with 10%
fetal bovine serum, penicillin (100 μgmL
−1
), and streptomycin
(100 μgmL
−1
) in an incubator (5% CO
2
, 37 °C). After the con-
centration of cells reached 1 × 10
5
cells per mL, the cells were
collected using centrifugation at 1000gfor 3 min.
Construction of the supersandwich cytosensor
The MCF7 cell suspension (100 μL) was introduced onto the
modified 3D microchip device and retained in an incubator
(5% CO
2
, 37 °C) for 50 min, followed by washing with 0.05%
Tween-20 and PBS. Then, 100 μL HRP–Si/AuNP–Ab2 was
dropped into the device and incubated at 37 °C for 50 min.
After rinsing and staining, the immobilized cells were imaged
by using SEM and a fluorescence microscope (Zeiss, Observer
Z1), and analyzed with the CHI 660E electrochemical work-
station. As a control, a flat substrate modified with GO/SA–Ab1
was also examined in parallel.
Results and discussion
Characterization of Si/AuNP and HRP–Si/AuNP–Ab2 conjugates
The electrochemical responses were greatly influenced by the
Ab2. So we first synthesized novel Si/AuNPs that were used as
antibodies and HRP carriers to amplify the signals. Fig. 1A
shows the SEM and transmission electron microscopy images
of the Si/AuNPs, indicating that the silica nanospheres were
well covered with Au nanoparticles and the Si/AuNPs were
mono-dispersed with a size around 125 nm in diameter. Such
a uniform dispersion of the Si/AuNPs was vital for their conju-
gation with HRP and Ab2. X-ray photoelectron spectroscopy
(XPS) spectra were explored to confirm formation of the HRP–
Si/AuNP–Ab2 conjugate. As shown in Fig. 1B, the conjugate
exhibited some single sharp N, Si and Au peaks in the XPS
spectra to indicate the successful formation of the HRP–
Si/AuNP–Ab2 conjugate.
Efficient capture of CTCs with the 3D microchip device
The 3D microstructure cell-capture device was prepared as
illustrated in Scheme 1. To test the cell-capture performance of
the 3D microchip device, we prepared a cell suspension
(10
5
cells per mL) in cell culture medium (DMEM). The cell
suspension (100 μL) was introduced onto the microchip device,
which was placed in an incubator (5% CO
2
, 37 °C) for 50 min.
As controls, 3D microchips modified with SA–Ab1 and flat
substrates modified with GO/SA–Ab1 were examined in parallel.
After rinsing and staining, the substrate-immobilized cells were
characterized through microscopy imaging and electrochemical
behaviors. Fig. 2A shows the SEM images of immobilized
Paper Nanoscale
16356 |Nanoscale,2015,7,16354–16360 This journal is © The Royal Society of Chemistry 2015

MCF7 cells at different magnifications after glutaraldehyde fix-
ation. The immobilized MCF7 cells, which were stained with
FDA, were also counted under a fluorescence microscope. As
shown in Fig. 2B, the 3D microchips modified with GO/SA–
Ab1 could capture three times more cells than the microchips
only modified with SA–Ab1, and eight times more cells than
the flat substrate modified with GO/SA–Ab1. This result
suggests that the introduction of GO and the 3D microstruc-
ture was possibly responsible for the enhanced cell-capture
yields. This effect was likely due to the synergistic interplay
between GO and the 3D microchip, which provided more
surface area for SA and antibody binding and enhanced local
topographic interactions between the microarrays and CTCs.
Characterization of the supersandwich cytosensor
As shown in Fig. 3 (curve a), the cyclic voltammogram did not
display any detectable signal on the Ab1–SA/GO/microchip in
PBS (pH 7.4). However, upon addition of thionine and H
2
O
2
,a
couple of stable and well-defined redox peaks were observed at
−0.248 and −0.289 V (curve b), which corresponded to the elec-
trochemical response of thionine. When the cytosensor was
incubated with MCF7 cells at 10
5
cells per mL, no obvious
change in the signal was observed (data not shown). However,
after the sensor was incubated with the HRP–Ab2 solution, the
reduction current obviously increased on the HRP–Ab2/MCF7/
Ab1–SA/GO/microchip, which was due to the catalysis of the
immobilized HRP toward the reduction of H
2
O
2
. Moreover,
when replacing HRP–Ab2 with HRP–Si/AuNP–Ab2 as the detec-
tion antibody, the electrocatalytic current at the HRP–Si/AuNP–
Ab2/MCF7/Ab1–SA/GO/microchip (curve d) increased substan-
tially. As expected, the achieved amplification of the signal
was ascribed to the increased amounts of enzymes introduced
on the microchips compared with the traditionally labeled
HRP–Ab2.
Fig. 1 (A) TEM (a) and SEM (b) images of Si/AuNPs. Inset: SEM image of
a silica nanosphere. (B) XPS spectra of HRP–Si/AuNP–Ab2 conjugates.
Fig. 2 (A) SEM images of microchips on which MCF7 cells were cap-
tured, shown at different magnifications. (B) Fluorescence micrographs
of 3D microchips modified with GO/SA–Ab1 (left), 3D microchips
modified with SA–Ab1 (middle), and the flat substrate modified with GO/
SA–Ab1 (right) on which MCF7 cells were captured.
Fig. 3 Cyclic voltammograms of (a) Ab1–SA/GO/microchip in pH 7.4
PBS and (b) Ab1–SA/GO/microchip, (c) HRP–Ab2/MCF7/Ab1–SA/GO/
microchip, (d) HRP–Si/AuNP–Ab2/MCF7/Ab1–SA/GO/microchip in pH
7.4 PBS containing 50 µM thionine and 2 mM H
2
O
2
. A concentration of
10
5
MCF7 cells per mL was used.
Nanoscale Paper
This journal is © The Royal Society of Chemistry 2015 Nanoscale,2015,7, 16354–16360 | 16357

Optimization of experimental conditions
The electrochemical performance of the cytosensor is influ-
enced by some parameters such as the ratio of HRP to Ab2 and
the incubation time. To improve the sensitivity, the ratio of
HRP to Ab2 was optimized first because of the co-immobili-
zation of enzymes and antibodies on the Si/AuNP nanocarrier.
As shown in Fig. 4A, a marked increase of the peak current
was observed when increasing the HRP/Ab2 ratio from 50/1 to
200/1. This increased ratio caused an increase in the total
amount of HRP loaded per Si/AuNP, which was expected to
enhance the response during the immunoassay. However, the
excessive HRP reduced the available binding sites of Si/AuNPs
with Ab2, which would decrease the immune coupling
efficiency for capturing MCF7 cells and result in a decreased
response. Therefore, the HRP/Ab2 ratio of 200/1 was selected
as the optimal condition for the preparation of HRP–Si/AuNP–
Ab2 conjugates. The BCA protein assay
31
showed that the con-
centration of active HRP in the HRP–Si/AuNP–Ab2 dispersion
was 4.31 μgmL
−1
.
In addition, the effects of incubation time on capturing
MCF7 cells and specifically recognizing HRP–Si/AuNP–Ab2
were investigated. With increasing incubation time in the
MCF7 solution, the peak current increased and tended to be
steady after 50 min (red curve in Fig. 4B), indicating thorough
capturing of MCF7 cells on the 3D microchip device. When
HRP–Si/AuNP–Ab2 was incubated with MCF7 in the second
step, the current increased and reached a plateau at 50 min
(black curve in Fig. 4B), reflecting the saturation of binding
sites between MCF7 and Ab2. A longer incubation time could
result in a large nonspecific signal. Therefore, the optimal
incubation time for the first and second immunoreactions was
determined to be 50 min, respectively.
Electrochemical detection of CTCs
For the detection of CTCs, high sensitivity and selectivity play
important roles in improving the treatment. In view of this,
the proposed supersandwich cytosensor was challenged with
different concentrations of MCF7 cells, as shown in Fig. 5A.
The differential pulse voltammetry currents increased with the
increasing concentration of MCF7 cells. Fig. 5B shows the
linear calibration plots of the peak current (i
p
)versus the con-
centration of MCF7 cells. A linear relationship between i
p
and
the logarithm of the cell concentration is found in the range of
10
1
–10
7
cells per mL, with a correlation coefficient of 0.9921
(n= 3). The detection limit at a signal-to-noise ratio of 3σ
(where σis the standard deviation of the signal in a blank solu-
tion) was estimated to be 10 cells per mL, which is comparable
and much lower than the values reported in the literature
(summarized in Table 1). The excellent sensitivity is attributed
to double signal amplifications from GO and Si/AuNPs,
which greatly accelerate the rate of electron transfer on the
cytosensor.
Performance of the supersandwich cytosensor
To investigate the specificity of the supersandwich cytosensor
toward MCF7 cells, different possible interfering cells such as
HeLa and A549 were used. As shown in Fig. 5A (curves b and
c), the peak current of the cytosensor showed a negligible
change with the addition of the interfering cells, whereas a
substantial electrochemical signal increase was observed with
the addition of MCF7 cells. These results demonstrated the
good selectivity of the cytosensor for MCF7 cells. Thus, the
proposed strategy can be applied to different types of CTCs
using the corresponding Ab2.
To evaluate the intra- and inter-assay coefficients of vari-
ation of the cytosensor array, the intra-assay precision of the
proposed method was measured for six replicate measure-
ments. The coefficient of variation of the intra-assay was 3.5%
at 10
5
cells per mL. Likewise, the inter-assay coefficient of vari-
ation on six cytosensors was 4.3% at 10
5
cells per mL. The pre-
Fig. 4 (A) Effects of the HRP to Ab2 ratio on the current responses in
the presence of 10
5
MCF7 cells per mL. (B) Effect of incubation time on
the current responses by capturing MCF7 (red curve) and recognizing
between HRP–Si/AuNP–Ab2 and MCF7 (black curve) in the presence of
10
5
MCF7 cells per mL.
Fig. 5 (A) Differential pulse voltammetry responses of the supersand-
wich cytosensor incubated with (b) HeLa cells, (c) A549 cells at 10
5
cells
per mL and (a, d–j) different concentrations of MCF7 cells: 0, 10
1
,10
2
,
10
3
,10
4
,10
5
,10
6
,10
7
cells per mL. (B) Calibration curve of MCF7.
Table 1 Comparison of the sensitivity of different CTC cytosensors
Cytosensor type
Linear range
(cells per mL)
Detection limit
(cells per mL) Ref.
Microchip cytosensor 10
1
to 10
7
10 Present study
LSAW aptasensor 10
2
to 10
7
32 32
PEC biosensor 10
2
to 10
6
58 33
Colorimetric aptasensor 10
2
to 10
4
40 34
Aptamer/QD cytosensor 10
2
to 10
6
50 35
Paper Nanoscale
16358 |Nanoscale,2015,7, 16354–16360 This journal is © The Royal Society of Chemistry 2015

cision and reproducibility of the cytosensor are acceptable.
Moreover, when the cytosensor was stored at 4 °C, the analyti-
cal performance did not decline obviously after one week, and
93% of the initial response remained after one month. The
stability is attributed to the stable structure of the Si/AuNPs
and the strong interaction between anti-EpCAM and GO modi-
fied 3D microchips. Thus, the designed strategy shows good
performance for the detection of CTCs with a broad detection
range, low detection limit, excellent selectivity, good reproduci-
bility and stability.
In this study, the supersandwich cytosensor shows attrac-
tive performance for the quantification of MCF7 cells, such as
its wide linear range and low detection limit. First, the GO
modified 3D microchips provide large specific surface areas to
facilitate abundant binding of SA. Combined with the extra-
ordinary biocompatibility of GO, the microchips are very suit-
able for immobilizing anti-EpCAM with excellent stability and
bioactivity. Therefore, an ideal interface for cell capture is pro-
vided, which improves the sensitivity of detection compared to
previous designs. Second, the supersandwich strategy helps to
further enhance the sensitivity via the signal amplification of
Si/AuNPs. Indeed, the introduction of an HRP–Si/AuNP–Ab2
probe led to a remarkable increase of the signal.
Conclusions
In summary, due to the increase of local topographic inter-
actions between the substrate and CTCs, the formation of
sandwich-like structures and double signal amplifications
through the use of GO and Si/AuNPs, the proposed supersand-
wich cytosensor shows excellent performance for analysis of
CTCs with a wide linear range, low detection limit, and accep-
table stability, reproducibility and accuracy. Among the various
technologies developed to specifically recognize and capture
CTCs to date, this is the first strategy to combine 3D micro-
chips and an electrochemical method for quantitative detec-
tion of CTCs. Therefore, we anticipate that this 3D microchip
device can be extended for the determination of other CTCs
and shows great potential in the field of disease diagnostics
and clinical analysis.
Acknowledgements
This work was supported by the National Natural Science
Foundation of China (no. 81402466) and the Natural Science
Foundation of Hubei (no. 2015CFB610). W. Z. would like to
acknowledge the support from a Jianghan University start-up
grant. We would like to thank Editage (http://www.editage.cn/)
for English language editing.
References
1 M. Cristofanilli, G. T. Budd, M. J. Ellis, A. Stopeck,
J. Matera, M. C. Miller, J. M. Reuben, G. V. Doyle,
W. J. Allard, L. W. M. M. Terstappen and D. F. Hayes,
N. Engl. J. Med., 2004, 351, 781.
2 K. Pantel and C. Alix-Panabieres, Trends Mol. Med., 2010,
16, 398.
3 V. Murlidhar, M. Zeinali, S. Grabauskiene, M. Ghannad-
Rezaie, M. S. Wicha, D. M. Simeone, N. Ramnath,
R. M. Reddy and S. Nagrath, Small, 2014, 10, 4895.
4 T. JunáHuang, Lab Chip, 2013, 13, 602.
5 S. Nagrath, L. V. Sequist, S. Maheswaran, D. W. Bell,
D. Irimia, L. Ulkus, M. R. Smith, E. L. Kwak,
S. Digumarthy, A. Muzikansky, P. Ryan, U. J. Balis,
R. G. Tompkins, D. A. Haber and M. Toner, Nature, 2007,
450, 1235.
6 S. K. Singh, C. Hawkins, I. D. Clarke, J. A. Squire, J. Bayani,
T. Hide, R. M. Henkelman, M. D. Cusimano and
P. B. Dirks, Nature, 2004, 432, 396.
7 S. S. Banerjee, A. J. Badhwar, K. R. Zope, K. J. Todkar,
R. R. Mascarenhas, G. P. Chate, G. V. Khutale, A. Bharde,
M. Calderon and J. J. Khandare, Nanoscale, 2015, 7,
8684.
8 X. Hu, C. W. Wei, J. Xia, I. Pelivanov, M. O’Donnell and
X. Gao, Small, 2013, 9, 2046.
9 A. K. Geim, Science, 2009, 324, 1530.
10 G. S. Kulkarni, K. Reddy, Z. Zhong and X. Fan, Nat.
Commun., 2014, 5, 4376.
11 R. Alam, I. V. Lightcap, C. J. Karwacki and P. V. Kamat, ACS
Nano, 2014, 8, 7272.
12 J. Wang and X. Qu, Nanoscale, 2013, 5, 3589.
13 H. K. Choi, H. Y. Jeong, D. S. Lee, C. G. Choi and
S. Y. Choi, Carbon, 2013, 14, 186.
14 X. Zhang, Y. Zhang, Q. Liao, Y. Song and S. Ma, Small,
2013, 23, 4045.
15 H. J. Yoon, T. H. Kim, Z. Zhang, E. Azizi, T. M. Pham,
C. Paoletti, J. Lin, N. Ramnath, M. S. Wicha, D. F. Hayes,
D. M. Simeone and S. Nagrath, Nat. Nanotechnol., 2013, 8,
735.
16 S. Yin, Y. L. Wu, B. Hu, Y. Wang, P. Cai, C. K. Tan, D. Qi,
L. Zheng, W. R. Leow, N. S. Tan, S. Wang and X. Chen, Adv.
Mater. Interfaces, 2014, 1, 1300043.
17 H. J. Yoon, M. Kozminsky and S. Nagrath, ACS Nano, 2014,
8, 1995.
18 N. Zhang, Y. Deng, Q. Tai, B. Cheng, L. Zhao, Q. Shen,
R. He, L. Hong, W. Liu, S. Guo, K. Liu, H. R. Tseng,
B. Xiong and X. Z. Zhao, Adv. Mater., 2012, 24, 2756.
19 S. Hou, L. Zhao, Q. Shen, J. Yu, C. Ng, X. Kong, D. Wu,
M. Song, X. Shi, X. Xu, W. H. OuYang, R. He, X. Z. Zhao,
T. Lee, F. C. Brunicardi, M. A. Garcia, A. Ribas, R. S. Lo and
H. R. Tseng, Angew. Chem., Int. Ed., 2013, 52, 3379.
20 S. Hou, H. Zhao, L. Zhao, Q. Shen, K. S. Wei, D. Y. Suh,
A. Nakao, M. A. Garcia, M. Song, T. Lee, B. Xiong, S. C. Luo,
H. R. Tseng and H. H. Yu, Adv. Mater., 2013, 25, 1547.
21 S. Wang, K. Liu, J. Liu, Z. T. F. Yu, X. Xu, L. Zhao, T. Lee,
E. K. Lee, J. Reiss, Y. K. Lee, L. W. K. Chung, J. Huang,
M. Rettig, D. Seligson, K. N. Duraiswamy, C. K. F. Shen and
H. R. Tseng, Angew. Chem., Int. Ed., 2011, 50, 3084.
22 X. Liu and S. Wang, Chem. Soc. Rev., 2014, 43, 2385.
Nanoscale Paper
This journal is © The Royal Society of Chemistry 2015 Nanoscale,2015,7, 16354–16360 | 16359

23 O. Yehezkeli, R. Tel-Vered, S. Raichlin and I. Willner, ACS
Nano, 2011, 5, 2385.
24 Z. K. Han and Y. Gao, Nanoscale, 2015, 7, 308.
25 A. Li, P. Zhang, X. Chang, W. Cai, T. Wang and J. Gong,
Small, 2015, 11, 1892.
26 W. Chen, P. H. Tsai, Y. Hung, S. Chiou and C. Y. Mou, ACS
Nano, 2013, 7, 8423.
27 D. Tarn, D. P. Ferris, J. C. Barnes, M. W. Ambrogio,
J. F. Stoddart and F. I. Zink, Nanoscale, 2014, 6, 3335.
28 N. Z. Knezevic and V. S. Y. Lin, Nanoscale, 2013, 5,
1544.
29 L. Zhu, H. Wang, X. Shen, L. Chen, Y. Wang and H. Chen,
Small, 2012, 8, 1857.
30 K. Leopold, M. Foulkes and P. J. Worsfold, Anal. Chem.,
2009, 81, 3421.
31 M. M
´evel, G. Breuzard, J. J. Yaouanc, J. C. Cĺement,
P. Lehn, C. Pichon, P. A. Jaffr
`es and P. Midoux, ChemBio-
Chem., 2008, 9, 1462.
32 K. Chang, Y. Pi, W. Lu, F. Wang, F. Pan, F. Li, S. Jia, J. Shi,
S. Deng and M. Chen, Biosens. Bioelectron., 2014, 60, 318.
33 F. Liu, Y. Zhang, J. Yu, S. Wang, S. Ge and X. Song, Biosens.
Bioelectron., 2014, 51, 413.
34 X. Zhang, K. Xiao, L. Cheng, H. Chen, B. Liu, S. Zhang and
J. Kong, Anal. Chem., 2014, 86, 5567.
35 H. Liu, S. Xu, Z. He, A. Deng and J. Zhu, Anal. Chem., 2013,
85, 3385.
Paper Nanoscale
16360 |Nanoscale,2015,7, 16354–16360 This journal is © The Royal Society of Chemistry 2015