Biosensors and Bioelectronics 63 (2015) 546–551

Contents lists available at ScienceDirect

Biosensors and Bioelectronics

http://d0956-56

n Corrnn CorE-m

jwchoi@1 Th

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

3D label-free prostate specific antigen (PSA) immunosensor based ongraphene–gold composites

Hee Dong Jang a,n,1, Sun Kyung Kim a,b,1, Hankwon Chang a, Jeong-Woo Choi b,nn

a Rare Metals Research Center, Korea Institute of Geoscience and Mineral Resources, Daejeon 305-350, Koreab Interdisciplinary Program of Integrated Biotechnology, Sogang University, Seoul 121-742, Korea

a r t i c l e i n f o

Article history:Received 23 May 2014Received in revised form23 July 2014Accepted 7 August 2014Available online 13 August 2014

Keywords:Label-free immunosensorCrumpled graphene–gold compositeAerosol spray pyrolysisProstate specific antigen (PSA)

x.doi.org/10.1016/j.bios.2014.08.00863/& 2014 Elsevier B.V. All rights reserved.

esponding author. Tel.: þ82 42 868 3612; faxresponding author. Tel.: þ82 2 705 8480; faxail addresses: hdjang@kigam.re.kr (H.D. Jang),sogang.ac.kr (J.-W. Choi).ese authors contributed equally to this work

a b s t r a c t

Highly sensitive and label-free detection of the prostate specific antigen (PSA) remains a challenge in thediagnosis of prostate cancer. Here, a novel three-dimensional (3D) electrochemical immunosensorcapable of sensitive and label-free detection of PSA is reported. This unique immunosensor is equippedwith a highly conductive graphene (GR)-based gold (Au) composite modified electrode. The GR-based Aucomposite is prepared using aerosol spray pyrolysis and the morphology of the composite is the shape ofa crumpled GR ball decorated with Au nanoparticles. Unlike the previous research, this novel 3Dimmunosensor functions very well over a broad linear range of 0–10 ng/mL with a low detection limit of0.59 ng/mL; furthermore, it exhibits a significantly increased electron transfer and high sensitivitytoward PSA. The highest rate of current change with respect to the PSA concentration is 5 μA/(ng/mL).Satisfactory selectivity, reproducibility, and stability of the 3D immunosensor are also exhibited.

& 2014 Elsevier B.V. All rights reserved.

1. Introduction

Immunosensors based on antibody–antigen binding are one ofthe most widely used to detect disease related substances, whichare known as biomarkers, in clinical diagnostics (Li et al., 2013).Due to the specific binding of an antibody to its correspondingantigen, antibodies are immobilized on the immunosensor surfacein order to capture specific biomarkers (Mao et al., 2012). Amongthe numerous immunosensor species, the prostate specific antigen(PSA) for the specificity of prostate cancer markers has beenwidely used in prostate cancer screening, diagnosis, and treatmentafter monitoring (Huang et al., 2005; Qu et al., 2011). It is wellknown that the PSA concentration for a normal person rangesfrom 0 to 4 ng/mL (Qu et al., 2008; Yang et al., 2010).

There are two types of PSA immunosensors: sandwich-typeimmunosensors and label-free immunosensors. Sandwich-typeimmunosensors are primary composed of an antibody, secondaryantibody, and antigen. This immunosensor can be preparedthrough the label protocol with the primary antibody immobilizedon the solid surface and the specific antigen bound to the antibodysite. The labeled secondary antibody can bind to the PSA antigen(Yang et al., 2010). In sandwich-type immunosensors, the labeled

: þ82 42 868 3415.: þ82 2 3273 0331.

as the first author.

antibodies are used for signal amplification, and much attentionhas been paid to the development of materials for immobilizingmore enzymes in order to increase the efficiency and sensitivity(Sun et al., 2013; Yang et al., 2011). In contrast, the label-freeimmunosensor can be prepared via direct detection of antibody–antigen interaction. This immunosensor has become a remarkableanalytical tool for detection of biomolecules as a result of itssimple preparation, fast detection, high sensitivity, and low cost (T.Li et al., 2011, R. Li et al., 2011).

In order to develop enhanced label-free immunosensors, gra-phene (GR)-based composites are of interest as electrode materialsbecause GR has a high biocompatibility and fast electron trans-portation (Peng et al., 2012; Zhao et al., 2013). Akhavan et al.(2012) showed GR can effectively contribute to the development ofultra-high-sensitive electrochemical biosensors. T. Li et al. (2011)reported label-free electrochemical detection of PSA markersbased on GR–cobalt hexacyanoferrate nanocomposites. They sta-ted that the immunosensor using GR materials had advantages ofhigh sensitivity, good selectivity, and stability. Mao et al. (2012)developed label-free electrochemical immunosensors for PSAbased on GR–methylene blue nanocomposites. They reported thatthe amperometric signal decreased linearly with the PSA concen-tration (0.05–5.00 ng/mL) and that the immunosensor had a lowlimit of detection (0.013 ng/mL). The prepared PSA immunosensorwas used in the analysis of PSA in serum samples with satisfactoryresults. As with the previous studies, the GR-based compositematerials and nanomaterials garnered more attention as electrodematerials due to the synergistic effect of two or more functional

H.D. Jang et al. / Biosensors and Bioelectronics 63 (2015) 546–551 547

components in the potential sensor applications. Therefore, it isexpected that a composite of GR and gold (Au) nanoparticles couldexhibit enhanced sensitivities and simplify the assay system as alabel-free PSA immunosensor. This is because GR can provide abetter environment for electrochemical reactions for potentialimmunosensors, and Au nanoparticles are expected to enhancethe conductivity, biocompatibility, and strong adsorption abilities(Huang et al., 2013; Liu et al., 2013; De et al., 2008, Kavosi et al.,2014). However, label-free immunosensors for PSA detection usingGR–Au composites have not yet been reported.

In order to prepare label-free immunosensors fabricated withGR-based composites, previous studies have used a two-step orliquid phase reaction. However, these methods require numeroustime consuming stages such as filtering, washing, and drying; theyalso require supporting materials that assist with the immobiliza-tion in order to prepare the composite materials. Because themorphology of the composite prepared via a liquid phase reactionis flat and two-dimensional (2D), 2D layered GR–Au sheets have atendency to form aggregates and restack easily through π–πstacking and van der Waals attraction (Luo et al., 2011). Thus, it isexpected that the 2D GR–Au sheets not only reduce their effectivesensing ability but also compromise their properties, such as theaccessible sensing surface area on the working electrode.

In our previous studies, we developed a crumpled GR-basedcomposite exhibiting a three-dimensional (3D) morphology viaaerosol spray pyrolysis (ASP) (Jang et al., 2012, 2013a, 2013b). Wedemonstrated that the crumpled GR-based composites had both ahigh free volume and a high compressive strength, and that theycould be tightly packed without significantly reducing the acces-sible surface area, unlike the 2D GR sheets. The fabricatedcrumpled GR-based composites could also deliver much higherspecific capacitances in the sensing performance. Therefore,crumpled GR–Au composites are expected to have a highlysensitive electrocatalytic activity of a PSA immunosensor com-pared with a 2D GR–Au composite. Furthermore, the Au nanopar-ticles in crumpled GR–Au composites can interact with many

Fig. 1. Schematic illustration of (a) 2D GR–Au and 3D GR–Au electrodes, (b) the formatioof 3D label-free PSA immunosensor using crumpled GR–Au composites.

biomolecules (Fig. 1(a)). When crumpled GR–Au composites areused to fabricate label-free PSA immunosensors, 3D label-free PSAimmunosensors can be prepared with high accessible surfaceareas that lead to enhanced biomolecule absorption and promotedirect electron transfer between the materials and electrodesurface.

In this study, 3D label-free PSA immunosensors based oncrumpled GR–Au composites are introduced. The effects of theAu/GR ratio on the composite particle properties such as morphol-ogy and crystallinity are examined. Then, label-free immunosen-sors for the sensitive detection of PSA were prepared using thecrumpled GR–Au composites modified with glassy carbon electro-des. The linear response range, low detection limit, selectivity,stability, and reproducibility were investigated for the detection ofPSA as an immunosensor using cyclic voltammetry measurements.

2. Experimental

2.1. Preparation of crumpled GR–Au composites

Graphene oxide (GO) colloid was prepared via the oxidation ofgraphite powder (Alfa Aesar, 99.9% purity) using a modifiedHummers' method (Hummers and Offeman, 1958; Cote et al.,2009a, 2009b). Our previous study proved the characteristics ofGO by XRD, UV, Raman and XPS analysis. These results confirmedthat GO was successfully synthesized from graphite by themodified Hummers' method and matched well with previousresults (Jang et al., 2012, 2013b, Kim et al., 2014). For the synthesisof the various crumpled GR–Au composites, the precursor wasprepared with different weight ratios of Au/GR from 1 to 3, whilethe concentration of the GO was fixed at 0.5 wt% in the colloidalmixture. The crumpled GR–Au composites were prepared fromHAuCl4 �3H2O and GO using the ASP method (Jang et al., 2013b).The experimental apparatus for the ASP process consisted of anultrasonic atomizer, an electrical tubular furnace, and a filter

n of crumpled GR–Au composites via aerosol spray pyrolysis and (c) fabricating step

H.D. Jang et al. / Biosensors and Bioelectronics 63 (2015) 546–551548

sampler. The ultrasonic atomizer was used to generate micron-sized droplets of the GO–Au precursor. Then, the droplets werecarried into the furnace via 1.0 l/min of inert gas at an operatingtemperature of 800 °C. The evaporation of water, the thermalreduction of the GO and HAuCl4 �3H2O precursor, and the self-assembly between GR and Au nanoparticles were conducted inseries in a tubular furnace. Finally, the fabricated crumpled GR–Aucomposites were collected using a filter (Fig. 1(b)).

2.2. Immunosensor fabrication

Fig. 1(c) presents a schematic of the fabrication procedure forthe 3D label-free PSA immunosensors. The glassy carbon electrode(GCE, 3 mm diameter) was cleaned with ethanol and deionized(DI) water before use. Then, 5 ml of crumpled GR–Au colloid wasdropped onto the surface of the glassy carbon electrode (GCE).After the crumpled GR–Au modified electrode dried, 5 ml of anti-body solution (10 mg/mL) was dropped on the surface of electrode.Subsequently, after drying, 10 ml of BSA (0.1 wt%) was added to theelectrode surface. Then, the various concentrations of PSA antigenwere added to the electrode surface and the electrode wasincubated at room temperature. Phosphate buffer solution (PBS)and − −Fe(CN)6

3 /4 were used as the electrolytes for the PSAimmunosensor.

For comparison, we also fabricated 2D label-free PSA immuno-sensors using 2D GR–Au composite. The content of Au nanopar-ticles of 2D GR–Au composite was the same as those of 3D GR–Aucomposites. The 2D GR–Au composite was synthesized using thefollowing electro-deposition method: 5 ml of GO solution wasdropped onto the surface of a GCE. After the GO-coated electrodedried, the electrode was transferred to an electrochemical cellcontaining PBS solution. The GR/GCE was acquired by scanningelectro-reduction for 50 cycles from �1 to 1 V at a 50 mV/s scanrate. Then, the Au modified GR/GCE was fabricated via scanningelectro-reduction in HAuCl4 �3H2O for 50 cycles. The remainingexperimental procedures were the same as those for the 3D label-free PSA immunosensors.

Fig. 2. FE-SEM and TEM images of the crumpled GR–Au composites prepared at differentcolloid was 0.5 wt%, the operating temperature was 800 °C, and the carrier gas flow rat

2.3. Analysis

The particle morphologies and sizes of the crumpled GR–Aucomposites were characterized using a field emission scanningelectron microscope (FE-SEM; FEI, Sirion) and a transmissionelectron microscope (TEM; JEOL, JEM-ARM200F). Elemental com-position of the composite was measured by energy dispersiveX-ray spectroscopy (EDS; JEOL, JSM-6380LA) spectrum analysis.The crystallinity and specific surface area of the composites wereanalyzed using X-ray diffractometry (XRD; Rigaku, RTP 300 RC)and the Brunauer–Emmett–Teller method (BET; Micromeritics,Tristar 3000). The electrochemical properties of the PSA immuno-sensor were measured using a cyclic voltammetry (CV) methodwith an electrochemical interface instrument (Bio-Logics, VSP). Aconventional three-electrode cell was used with a glassy carbonelectrode (3 mm) as the working electrode, an Ag/AgCl electrodeas the reference electrode, and a platinum foil as the counterelectrode. The peak current had a potential ranging from �1.0 to1.0 V at a scan rate of 50 mV/s.

3. Result and discussion

3.1. Characterization of the crumpled GR–Au composites

Fig. 2 presents the FE-SEM and TEM images of the crumpledGR–Au composites prepared with different Au/GR weight ratios of(a1, a2) 1, (b1, b2) 2, and (c1, c2) 3, while the GO concentration ofthe colloid was 0.5 wt%. The FE-SEM images demonstrate that thecomposites appear to be crumpled paper balls and are approxi-mately 1 mm in diameter regardless of the Au/GR ratio. The TEMimages revealed that the Au nanoparticles were uniformly dis-tributed on the surface of the crumpled GR. As the Au/GR ratioincreased, large numbers of Au nanoparticles were generated andformed agglomerates due to interaction force among particles.Since some of the agglomerates were sintered while those werepassed through the heating zone, the TEM images as shown inFig. 2 indicated larger sizes and higher numbers of Au

Au/GR ratios of (a1, a2) 1, (b1, b2) 2 and (c1, c2) 3 while the GO concentration of thee was 1 l/min.

Fig. 3. Comparison of electrochemical properties of modified crumpled GR–Au and2D GR–Au electrodes in (a) cyclic voltammograms and (b) electrochemical im-pedance spectroscopy, respectively. Au/GR ratio was varied from 1.0 to 3.0 in(b) while the GO concentration was 0.5 wt%.

Fig. 4. Variation of cyclic voltammograms with the fabricating process of 3D label-free PSA immunosensors crumpled GR–Au, crumpled GR–Auþantibody, andcrumpled GR–Auþantibodyþantigen modified GCE in PBS containing 5 mM

− −Fe(CN)63 /4 while the Au/GR ratio was 3.0, GO was 0.5 wt%.

H.D. Jang et al. / Biosensors and Bioelectronics 63 (2015) 546–551 549

nanoparticles having a wide particle size distribution were ob-served on the surface of GR at the higher Au content. The particlesize and distribution of the Au particles in the GR–Au compositeswere determined by counting the more than 300 particles fromTEM images. The mean diameter of Au particles was 16, 21, 28 nm,and the geometric standard deviation, which represents the sizedistribution, was 1.59, 1.75, 1.65 when the Au/GR ratio was 1, 2, 3,respectively (Fig. S1). In Fig. S2, the TEM dark field images andEDS-mapping images reveal that the Au nanoparticles were welldeposited on the crumpled GR sheets. At higher Au/GR ratios,larger sizes and higher numbers of Au particles were observed inboth the TEM dark field images and EDS-mapping images. The GRand Au contents in crumpled GR–Au composites were determinedfrom SEM-EDS spectra, as shown in Fig. S3. Fig. S3 shows that themeasured weight ratio of Au/GR was 1.2, 2.0, 3.2 when the pre-determined Au/GR ratio in the aerosol precursor was 1, 2, 3,respectively. This result thus exhibits that the Au/GR ratio of as-prepared samples are consistent with the pre-determined Au/GRratio of the aerosol precursor and the aerosol process is veryefficient tool to control the composition of nanoparticles decoratedGR composite materials.

Fig. S4 presents FE-SEM images of the 2D GR–Au composite.The morphology of the 2D GR–Au composites demonstrated thatthe 50 nm Au nanoparticles were well deposited on the surface ofthe flat GR sheets. The Au/GR weight ratio of the 2D compositeobtained from SEM-EDS spectra was 2.8 when the pre-determinedAu/GR ratio was 3 in the precursor.

Fig. S5 demonstrates that the crystallinity of the crumpled GR–Au composites indicated the same peaks as those of the Aureference (JCPDS 04-0784). The peak weakly detected near 25°exhibited GR because the intensity of the GR phase was muchlower than that of the Au phase. The crystallite sizes of Aucalculated using the Scherrer equation were 17, 23, and 24 nmwith respect to Au/GR ratio. This indicated that the crumpled GR–Au composites successfully crystallized at 800 °C with variable Aucontent using the ASP. The Raman spectrum of the as-preparedcrumpled GR–Au composite showed a D band at 1346 cm�1, Gband at 1600 cm�1 and 2D band at 2760 cm�1, which indicatedthe presence of GR (Fig. S6). The specific surface areas of thecrumpled GR–Au composites decreased from 170 to 130 m2/g withthe increase in the Au content. The decrease of the specific surfacearea was attributed to the increased size and weight of the Aunanoparticles on the GR surface.

3.2. Evaluation of the PSA immunosensor

The performance of the 3D label-free PSA immunosensors wasmeasured using cyclic voltammetry measurements. First, in orderto discern the role of crumpled GR–Au and the possible synergisticeffects, the cyclic voltammetry and electrochemical impedancespectra of the crumpled GR–Au and 2D GR–Au modified electrodewere compared in − −Fe(CN)6

3 /4 electrolyte (Fig. 3). The crumpledGR–Au composite modified electrode exhibited stronger redox,higher current flow, and smaller electron transfer resistancecompared with the 2D GR–Au modified electrode in the detectionof − −Fe(CN)6

3 /4 . This result demonstrated that the crumpled GR–Aucomposites have superior abilities in their catalyst in terms ofelectrochemical reactions (Zhu et al., 2013). Sun et al. (2013)showed the high performance was attributed to the higheraccessible surface area. The performance of the crumpled GR–Auhigher than that of the 2D GR–Au, and thus the electric conduc-tivity improved. In addition, the electron transfer resistance of thecrumpled GR–Au composite decreased with increases in the Au/GRratio. This indicated that the electric conductivity of the modifiedelectrode improved as a result of increases in the amount of Aunanoparticles (Kong et al., 2011).

In order to characterize the fabrication process of the immu-nosensor, the cyclic voltammetry at each immobilization step wasrecorded and is presented in Fig. 4. When the antibody was

H.D. Jang et al. / Biosensors and Bioelectronics 63 (2015) 546–551550

conjugated onto the electrode, there was a clear decrease in theamperometric current. This decrease in the amperometric currentcould be explained by the antibody functioning as an electrontransfer blocking layer, thus hindering the electron transfer towardthe electrode surface (Yu et al., 2012). Furthermore, when theantigen was conjugated, the amperometric current decreasedmore, which indicated that the immunocomplex formed afterthe specific immunoreaction between the antibody and antigensignificantly obstructed the reaction on the electrode surface (T. Liet al., 2011). Therefore, the crumpled GR–Au composite modifiedelectrode could be used for immunosensor to detect PSAs. FT-IRspectra of the as-prepared crumpled GR–Au composites weremeasured before and after antibody and antigen immobilization(Fig. S7). The FT-IR spectra of antibody captured composite showedthe presence of CH2 deformation (2948 cm�1), NH bonding (1450cm�1), and CN stretching vibration (1117, 1046 cm�1), indicatingthat antibodies were successfully immobilized. When the antigenwas immobilized into the antibody captured composite, the FT-IRspectrum showed no difference in the chemical bondings as thespectrum of antigen captured composite.

The cyclic voltammetry of the immunosensor with differentPSA concentrations was also investigated. The cyclic voltammetryof the 3D label-free PSA immunosensor based on crumpled GR–Aucomposites was measured with respect to the Au/GR ratio (Fig. S8).All immunosensors based on various crumpled GR–Au compositesamples exhibited high current flows as well as clear redox peaks.They indicated good characteristics of sensor activity but exhibiteddifferences in the rate of current change. The redox peak of thecurrent flows decreased with an increasing PSA concentrationbecause the antibody–antigen immunocomplex prevented theelectron transfer on the electrode surface.

The calibration plots were obtained at 0.35 V by recording thecurrent response versus the PSA concentration (Fig. 5). The plotsexhibited good linear relationships between the current changesand different PSA concentrations from 0 to 10 ng/mL and the lowdetection limit was 0.59 ng/mL at 3 of Au/GR ratio. The limit ofdetection was calculated statistically, intercept of the regressionline plus 3 times of standard deviation (Miller and Miller, 1988).The rates of current change with respect to the PSA concentrationfrom 0 to 10 ng/mL indicated 0.8, 3.5, 4.8, and 4.9 μA/(ng/mL) withAu/GR content amounts of 1, 1.5, 2, and 3, respectively. Fig. 5 showsthat the immunosensor prepared at the Au/GR ratio higher than1.5 exhibited a higher current change rate, but it was found that

Fig. 5. Comparison of calibration curves of 3D label-free PSA immunosensors atdifferent Au contents toward different PSA concentration while the Au/GR ratio was1.0–3.0, GO was 0.5 wt%, and scan rate was 50 mV/s (error bar¼7standarddeviation (n¼5)).

there was no further increase in the current change rate when theAu/GR ratio was higher than 2. This result may exhibit that there isan optimum Au content to plays an important role to obtain highercurrent change rate.

When current change rate of 3D GR–Au immunosensors wascompared with that of 2D GR–Au immunosensors, all the 3Dimmunosensors exhibited much higher current change rates thanthe 2D immunosensor (Fig. S9). This result thus confirms that thecrumpled GR-supported Au nanoparticles have superior ability intheir catalyst regarding their electrochemical reactions, whichindicates that assembled biomolecules with high bioactivity wereimmobilized (Yu et al., 2012; Yang et al., 2013). When wecompared the performance of the PSA immunosensor based onGR–Au composite to those of previously reported sensors, oursensor showed the highest current change rate in the PSAconcentration ranged from 0 to 10 ng/mL, even though Maoet al. (2012) showed the highest current change rate from 0 to5 ng/mL (Table S1). The crumpled GR–Au composites having largesurface area and high loading of Au nanoparticles could conjugatea larger amount of antibody biomolecules, and then capture moreantigen biomolecules than the 2D GR–Au composite. In addition,they could increase the loaded amount of biomolecules and thenamplify the response at the higher Au content of crumpled GR–Aucomposites. Compared with the previous results of label-free PSAimmunosensors, the as-prepared 3D PSA immunosensors exhib-ited the highest rate of current change (5 μA/(ng/mL)) with respectto the PSA concentrations from 0 to 10 ng/mL. Therefore, the as-prepared 3D PSA immunosensors were superior to other 2D PSAimmunosensors in the sensitivity of the current change rate.

Characterizations of the selectivity, reproducibility, and stabi-lity using the as-prepared 3D immunosensors were also con-ducted. The selectivity of the immunosensors was examined asshown in Fig. 6. The amperometric response of the alpha-fetopro-tein (AFP), glucose (Glu), uric acid (UA), bovine serum albumin(BSA), and vitamin C (VitC) were examined for their interferenceeffect: 1 ng/mL PSA and 100 ng/mL interference materials weremixed and the current was measured. This result demonstratedthat there was no change in the current after adding the inter-ference material, which indicates that the selectivity of theimmunosensor was acceptable. In order to evaluate the reprodu-cibility of the immunosensor, the five electrodes were measured ina 1 ng/mL PSA solution. The relative standard deviation (RSD) of

Fig. 6. Selectivities of 3D label-free PSA immunosesnors prepared with thecomposite (Au/GR¼3) on amperometric response to 1 ng/mL PSA, 1 ng/mLPSAþ100 ng/mL AFP, 1 ng/mL PSAþ100 ng/mL UA, 1 ng/mL PSAþ100 ng/mL BSA,1 ng/mL PSAþ100 ng/mL Glu, and 1 ng/mL PSAþ100 ng/mL VitC (error bar¼7standard deviation (n¼5)).

H.D. Jang et al. / Biosensors and Bioelectronics 63 (2015) 546–551 551

the current response was 3.1% for the five measurements, whichconfirms the reproducibility of the PSA immunosensor. Theamperometric response of crumpled GR–Au immunosensors wasshown in Fig. S10. The as-prepared electrode showed a rapid andsensitive electrocatalytic response as soon as the PSA added. Thefast response time may be due to fast detection of PSA electrolyteby the PSA sensor consisted of the 3D crumpled structure having ahigh surface area on the electrode. The stability of the preparedPSA immunosensors was measured with respect to the number ofscanned cycles in an electrolyte solution using the crumpled GR–Au composite modified electrode. The cyclic voltammetry of theas-prepared immunosensor demonstrated that the response de-crease in the current was approximately 1%, 5%, and 9% after theelectrode was scanned continuously for 10, 50, and 100 cycles,respectively. Then, it was confirmed that the immunosensor of thecrumpled GR–Au composite was stable and repeatable because theCV curves exhibited stable changes for 100 cycles. Therefore, theas-prepared 3D label-free PSA immunosensors can provide asimple, sensitive, and specific method for prostate cancer markerdetection that can have potential applications in clinical analysis.

4. Conclusions

A novel 3D label-free PSA immunosensor based on GR–Aucomposites was successfully developed for the detection of pros-tate cancer. The GR–Au composite synthesized via ASP was acrumpled GR ball decorated with Au nanoparticles. The ampero-metric response of the 3D label-free PSA immunosensor based onthe GR–Au composite exhibited stronger redox, higher currentflow, and smaller electron transfer resistance compared with the2D GR–Au modified electrode. The highest rate of current changefor the 3D label-free PSA immunosensor was 5 μA/(ng/mL). The 3Dlabel-free PSA immunosensor also exhibited a good linear relation-ship between the current change and different concentrations ofPSA from 0 to 10 ng/mL and the detection limit was 0.59 ng/mL.Furthermore, the as-prepared immunosensor exhibited satisfac-tory selectivity, reproducibility, and stability. The as-preparedimmunosensor provides a promising approach for clinical researchand diagnostic applications.

Acknowledgment

This research was supported by the General Research Project ofthe Korea Institute of Geoscience and Mineral Resources (KIGAM)funded by the Ministry of Science ICT and Future Planning.

Appendix A. Supplementary information

Supplementary data associated with this article can be found inthe online version at http://dx.doi.org/10.1016/j.bios.2014.08.008.

References

Akhavan, O., Ghaderi, E., Rahighi, R., 2012. ACS Nano 6, 2904–2916.Cote, L.J., Silva, R.C., Huang, J., 2009a. J. Am. Chem. Soc. 131, 11027–11032.Cote, L.J., Kim, F., Huang, J., 2009b. J. Am. Chem. Soc. 131, 1043–1049.De, M., Ghosh, P.S., Rotello, V.M., 2008. Adv. Mater. 20, 4225–4241.Huang, L., Reekmans, G., Saerens, D., Friedt, J.M., Frederix, F., Francis, L., Muylder-

mans, S., Campitelli, A., Hoof, C.V., 2005. Biosens. Bioelectron. 21, 483–490.Huang, K.J., Li, J., Wu, Y.Y., Liu, Y.M., 2013. Bioelectrochemistry 90, 18–23.Hummers, W.S., Offeman, R.E., 1958. J. Am. Chem. Soc. 80, 1339.Jang, H.D., Kim, S.K., Chang, H., Roh, K.M., Choi, J.W., Huang, J., 2012. Biosens.

Bioelectron. 38, 184–188.Jang, H.D., Kim, S.K., Chang, H., Choi, J.W., Luo, J., Huang, J., 2013a. Aerosol Sci.

Technol. 47, 93–98.Jang, H.D., Kim, S.K., Chang, H., Choi, J.W., Huang, J., 2013b. Mater. Lett. 106, 277–

280.Kavosi, B., Salimi, A., Hallaj, R., Amani, K., 2014. Biosens. Bioelectron. 52, 20–28.Kim, S.K., Chang, H., Choi, J.W., Huang, J., Jang, H.D., 2014. Kona Powder Part J. 31,

111–125.Kong, F.Y., Xu, M.T., Xu, J.J., Chen, H.Y., 2011. Talanta 85, 2620–2625.Li, H., He, J., Li, S., Turner, A.P.F., 2013. Biosens. Bioelectron. 43, 25–29.Li, T., Yang, M., Li, H., 2011. J. Electroanal. Chem. 655, 50–55.Li, R., Wu, D., Li, H., Xu, C., Wang, H., Zhao, Y., Cai, Y., Wei, Q., Du, B., 2011. Anal.

Biochem. 414, 196–201.Liu, N., Chen, X., Ma, Z., 2013. Biosens. Bioelectron. 48, 33–38.Luo, J., Jang, H.D., Sun, T., Xiao, L., He, Z., Katsoulidis, A.P., Kanatzidis, M.G., Gibson, J.

M., Huang, J., 2011. ACS Nano 5, 8943–8949.Mao, K., Wu, D., Li, Y., Ma, H., Ni, Z., Yu, H., Luo, C., Wei, Q., Du, B., 2012. Anal.

Biochem. 422, 22–27.Miller, J.C., Miller, J.N., 1988. Statistics for Analytical Chemistry, 2nd edition Ellis

Horwood Limited, England.Peng, S., Zou, G., Zhang, X., 2012. J. Electroanal. Chem. 686, 25–31.Qu, B., Chu, X., Shen, G., Yu, R., 2008. Talanta 76, 785–790.Qu, F., Li, T., Yang, M., 2011. Biosens. Bioelectron. 26, 3927–3931.Sun, G., Lu, J., Ge, S., Song, X., Yu, J., Yan, M., Huang, J., 2013. Anal. Chim. Acta 775,

85–92.Yang, M., Javadi, A., Li, H., Gong, S., 2010. Biosens. Bioelectron. 26, 560–565.Yang, M., Javadi, A., Gong, S., 2011. Sens. Actuators B 155, 357–360.Yang, G., Li, L., Rana, R.K., Zhu, J.J., 2013. Carbon 61, 357–366.Yu, S., Cao, X., Yu, M., 2012. Microchem. J. 103, 125–130.Zhao, L., Li, S., He, J., Tian, G., Wei, Q., Li, H., 2013. Biosens. Bioelectron. 49, 222–225.Zhu, L., Xu, L., Jia, N., Huang, B., Tan, L., Yang, S., Yao, S., 2013. Talanta 116, 809–815.