S

Fee

FSa

b

c

d

a

ARRAA

KMGCJE

1

nt(2setnsteo

s

0d

Biosensors and Bioelectronics 26 (2011) 4253–4256

Contents lists available at ScienceDirect

Biosensors and Bioelectronics

journa l homepage: www.e lsev ier .com/ locate /b ios

hort communication

acile fabrication of magnetic gold electrode for magnetic beads-basedlectrochemical immunoassay: Application to the diagnosis of Japanesencephalitis virus

ang Lia,b, Li Meia, Yaoming Lia, Kaihong Zhaoa,c, Huanchun Chena, Peng Wud, Yonggang Hua,c,∗,hengbo Caoa,∗

State Key Laboratory of Agricultural Microbiology, Huazhong Agricultural University, Wuhan 430070, ChinaSchool of Environmental Science and Engineering, Huazhong University of Science and Technology, Wuhan 430074, ChinaCollege of Life Science and Technology, Huazhong Agricultural University, Wuhan 430070, ChinaDepartment of Clinical Laboratory, Taihe Hospital, Hubei Medical University, Shiyan 442000, China

r t i c l e i n f o

rticle history:eceived 29 January 2011eceived in revised form 14 April 2011ccepted 15 April 2011vailable online 22 April 2011

eywords:agnetic gold electrode

a b s t r a c t

A novel magnetic beads-based electrochemical immunoassay strategy has been developed for the detec-tion of Japanese encephalitis virus (JEV). The magnetic gold electrode was fabricated to manipulatemagnetic beads for the direct sensing applications. Gold-coated magnetic beads were employed as theplatforms for the immobilization and immunoreaction process, and horseradish peroxidase was chosenas an enzymatic tracer. The proteins (e.g., antibodies or immunocomplexes) attached on the surface ofmagnetic beads were found to induce a significant decline in their electric conductivity. Multiwalledcarbon nanotubes were introduced to improve sensitivity of the assay. The envelope (E) protein, a major

old-coated magnetic beadarbon nanotube

apanese encephalitis viruslectrochemical immunoassay

immunogenic protein of JEV, was utilized to optimize the assay parameters. Under the optimal conditions,the linear response range of E protein was 0.84 to 11,200 ng/mL with a detection limit of 0.56 ng/mL. Whenapplied for detection of JEV, the proposed method generated a linear response range between 2 × 103

and 5 × 105 PFU/mL. The detection limit for JEV was 2.0 × 103 PFU/mL, which was 2 orders of magnitudelower than that of immunochromatographic strip and similar to that obtained from RT-PCR. This methodwas also successfully applied to detect JEV in clinical specimens.

. Introduction

Japanese encephalitis virus (JEV), which contributes to huge eco-omic losses in animal industry in southern and eastern Asia andhreatens public health, is a single-strand positive-sense RNA virusWang et al., 2009; Misra and Kalita, 2010; Van den Hurk et al.,009). The diagnosis of JEV is generally carried out by reverse tran-criptase polymerase chain reaction (RT-PCR). However, RT-PCR isxpensive and may not be appropriate for the developing coun-ries. In addition, RT-PCR method has the risk of releasing amplifieducleic acids into the environments, thereby contaminating theubsequent analysis. Development of a rapid, cost-effective, sensi-ive and specific antigen detection method is, therefore, urgent for

arly and reliable clinical diagnosis as well as effective surveillancef new JEV infection.

∗ Corresponding authors. Tel.: +86 27 87282608; fax: +86 27 87282608.E-mail addresses: yongganghu@mail.hzau.edu.cn (Y. Hu),

bcao@mail.hzau.edu.cn (S. Cao).

956-5663/$ – see front matter © 2011 Elsevier B.V. All rights reserved.oi:10.1016/j.bios.2011.04.028

© 2011 Elsevier B.V. All rights reserved.

Immunoassay, quantitative analysis of antibody or antigenconcentrations based on the highly biospecific recognition interac-tions, is of paramount importance for clinical diagnosis (Aytur et al.,2006; Du et al., 2010; Fu et al., 2008; Kerman et al., 2007). Specialattention has been paid to electrochemical detection method due toits high sensitivity, simple instrumentation, excellent compatibil-ity, low cost and power requirements (De la Escosura-Muniz et al.,2010; Du et al., 2010; Nie et al., 2009; Zhong et al., 2009). Typ-ically, the electrochemical immunoassay involves two processes:the recognition elements (antibodies or antigens) are fixed onthe electrode surface to react with target analytes, and then theimmunological responses are converted into a measurable electri-cal signal. These two different events are performed at the sameelectrode surface. Therefore, it is not surprising that the proper-ties of the optimum surface for immunoreaction can hardly be thesame as those for electrochemical detection (Palecek and Fojta,2007), thereby causing high nonspecific binding, poor reproducibil-

ity, et al.

One major emerging strategy is the incorporation of magneticbeads (MBs) into electrochemical immunoassay (Campuzano et al.,2010; De la Escosura-Muniz et al., 2010; Liebana et al., 2009; Hervas

4254 F. Li et al. / Biosensors and Bioelectronics 26 (2011) 4253–4256

F ) immp chem

ecaaMmpttttwrcec

ctw(ksIgci

ig. 1. Schematic representation of the magnetic electrochemical immunoassay: (1rotein or JEV; (3) incubation with 2A2 IgG-HRP; (4) mixed with MCNTs; (5) electro

t al., 2009; Palecek and Fojta, 2007; Zacco et al., 2006). This canircumvent the problems in generic technique discussed above, byllowing the immobilization and immunoreaction to be performedway from the electrode surface. As an ideal platform candidate,Bs can automatically purify and concentrate target analytes by aagnet and effectively minimize the matrix effect from the sam-

les. Additionally, MBs provide a high surface area to immobilizehe biomolecules as many as possible, leading to a lower detec-ion limit. Ideally, after the process of immunoreaction, MBs carriedhe relevant information would be captured by the magnetic elec-rode and focused on the detection surface for the direct sensingithout the need for further processing. This concept effectively

educes the complexity and time required for the sensing appli-ation (Goon et al., 2010). Thus, the development of a magneticlectrode to manipulate MBs is a key step to achieve the directapturing and sensing.

A gold electrode, which has a wide potential window in theathode region and efficient electric conductivity, is consideredo be a major metallic electrode. In the last decade, it has beenidely explored as a transducer in electrochemical immunoassay

Chaki and Vijayamohanan, 2002; Lucarelli et al., 2004). To our bestnowledge, the fabrication of a magnetic gold electrode and itsubsequent applications has not been reported in the literature.

n the present work, we first and foremost fabricated magneticold electrodes, and then developed a novel MBs-based electro-hemical immunoassay strategy for the detection of JEV. As shownn Fig. 1, gold-coated magnetic beads (GMBs) were chosen as the

obilization of 4D1 IgG on the surface of GMBs; (2) immunological reactions with Eical detection. (not in scale).

platform for the immobilization and the selective capture (step1–3). After the immunoreactions, multiwalled carbon nanotubes(MCNTs) were mixed with immunocomplex-coated GMBs (IGMBs)to improve the performance of the assay (step 4). Then, the mix-ture was captured and focused on the detection surface with helpof magnetic gold electrode, and the electrochemical detection wascarried out with the use of H2O2-hydroquinone-HRP system (step5). Finally, this assay was applied in clinical diagnosis of JEV.

2. Experiments

The magnetic electrode (Supplementary Fig. S1) was developedin our lab for the assay. It consists of a neodymium hollow cylindermagnet (6 mm diameter × 52 mm length) that was press-fitted intoa PTFE cylindrical tube (10 mm diameter × 65 mm length). One endof the PTFE cylindrical tube was packed with gold to about 5 mmdepth and 3 mm diameter. A silver wire was then fitted throughthe neodymium magnet hole and was connected with the goldfor electric conductivity. The apparatus, reagents and methods aredescribed in detail in the Supplementary Content section.

3. Results and discussion

3.1. Characterization of the magnetic gold electrode

The internal resistances of the magnetic gold electrode and anormal gold electrode were examined, respectively. The difference

oelectronics 26 (2011) 4253–4256 4255

wnfiestwFaaimsa

3

iaenatimw

rsciteawttomoptdMrtdHwsiti(s

aetH

3

c

Fig. 2. (a) Cyclic voltammograms obtained from IGMBs with MCNTs (—)and without MCNTs (. . .) in 10 mmol/L PBS solutions (pH = 7.4) containing2.5 mmol/L K3[Fe(CN)6]/K4[Fe(CN)6], scan rate 100 mV/s; (b) DPV response with dif-ferent amounts of MCNTs in PBS solutions (pH = 7.4) containing 1 mmol/L H2O2 and1 mmol/L hydroquinone; (c) Influence of the amounts of MCNTs on signal-to-noiseratio (the ratio of DPV response in the presence of E protein to DPV response with-

F. Li et al. / Biosensors and Bi

as found not to be significant, which indicates that this mag-etic gold electrode had good electric conductivity. The magneticeld strengths measured at 0, 1.5 and 5 mm from the centre of thelectrode surface were 35.9, 34.7, and 22.3 mT, respectively. Thetrength of the central magnetic field was moderate, and higherhan that of the surrounding field. This fact allowed IGMBs to beell captured and focused on the detection surface (Supplementary

ig. S2), making the process of collection and detection readilyvailable and simultaneously performed on the same platform. Inddition, electrode regeneration can be easily achieved, i.e., remov-ng IGMBs away from the detection surface without the need for the

agnet to be taken out from the PTFE tube. These factors demon-trate that the magnetic gold electrode was robust, user-friendly,nd show a commercial promise for electrochemical immunoassay.

.2. Optimization of the experimental conditions

Taking account of biosafety, the envelope (E) protein, a majormmunogenic protein of JEV, was utilized to produce monoclonalntibodies (2A2 and 4D1) (Li et al., 2010), optimize the assay param-ters and evaluate the performance of this assay. Due to the inertature of most MBs, 10–75 �g of GMBs were selected and examineds the platform. As expected, the sensitivity rose with an increase ofhe GMBs and up to 40 �g. Excess GMBs was not necessary, becauset was not in close contact with the electrode surface, even mini-

ized the electric conductivity of the assay. The amount of 40 �gas, thus, chosen as a compromise for the subsequent experiment.

Cyclic voltammograms (CVs) analysis using [Fe(CN)6]3−/4−

edox probe was used to characterize this immunoassay. Ashown in the Supplementary Fig. S3, strong declines in electriconductivity were observed when biomolecules (4D1 IgG andmmunocomplex) attached on the surface of GMBs. To addresshis problem, MCNTs that possess high roughness and area, uniquelectronic were selected to improve the assay. After the immunore-ction, MCNTs were mixed with IGMBs, and then the poorlyrapped MCNTs were removed by washing. Transmission Elec-

ron Microscope (TEM) was used to describe the morphology ofhe IGMBs with and without MCNTs, respectively. The structuref IGMBs in the presence of MCNTs (Supplementary Fig. S4(b)) isore incompact than that of IGMBs (Supplementary Fig. S4(a)),

wing to the fact that MCNTs can nonspecifically absorb onto therotein surface (Baldrich et al., 2011). The CVs in Fig. 2a showshat �Ep became smaller and the redox peak currents increasedramatically when MCNTs were added. This demonstrates thatCNTs can obviously enhance the electron transfer between the

edox probe and the electrode (Yang et al., 2010). The influence ofhe amounts of MCNTs (0–15 �g) on sensitivity was examined byifferential pulse voltammograms (DPVs) in H2O2-hydroquinone-RP system. As shown in Fig. 2b, the cathodic currents enhancedith an increase of the amount of MCNTs from 0–12 �g. The best

ignal-to-noise ratio could be attained when the amount of MCNTss 12 �g (Fig. 2c). Further increasing the amount of MCNTs leadso IGMBs partially confined on the detection surfaces, and resultsn the declines of cathodic currents and the signal-to-noise ratioFig. 2 b and c). MCNTs at 12 �g were therefore, selected for theubsequent assay.

The concentration of mediator, H2O2 and PBS solution, as wells the pH of PBS solution were further investigated. Optimization oflectrochemical detection was achieved in the 10 mmol/L PBS solu-ion (pH = 7.0) containing 3 mmol/L hydroquinone and 0.8 mmol/L2O2.

.3. Performances of the immunoassay

Under optimal conditions, the increase of the cathodic peakurrent in the DPV response was proportional to the logarithm

out E protein) in PBS solutions (pH = 7.4) containing 1 mmol/L H2O2 and 1 mmol/Lhydroquinone.

of the concentration of E protein in the range from 0.84 to11200 ng/mL (Supplementary Fig. S5(a)). The linear regressionequation is i = 5.4387 lg CE protein + 6.2031 (R2 = 0.9934), where i isthe increase of cathodic current and CE protein is the concentrationof E protein in ng/mL. The detection limit is 0.56 ng/mL (S/N = 3).The reproducibility and repeatability of the developed assay were

also examined. A series of repetitive measurements of 56 ng/mLE protein gave sensor-to-sensor reproducible results with a rela-tive standard deviation (RSD) of 7.64% (n = 11) for this biosensor.

4256 F. Li et al. / Biosensors and Bioelect

Table 1Comparative results of the proposed method and RT-PCR of 60 clinical samples.

RT-PCR

Positive Negative Total

The magneticelectrochemical assay

Positive 17 0 17Negative 3 40 43

Total 20 40 60

TEt

Tiffiah

3

acTpteiTwcnstridgstTcpt

ptAbpva

sflA9pd

Yang, W.R., Ratinac, K.R., Ringer, S.P., Thordarson, P., Gooding, J.J., Braet, F., 2010.Angew. Chem. Int. Ed. 49, 2114–2138.

he efficiency value was calculated using following equations (Shukla et al., 2009):fficiency = (TP + TN) × 100/Total = 95.0 (TP = 17, TN = 40, Total = 60). TP: true posi-ive; TN: true negative.

his suggested a good reliability of the MBs-based electrochemicalmmunoassay. The repeatability of the electrochemical detectionsor 56 ng/mL E protein was checked by taking the reduction peaks ofve successive voltammograms from one immunosensor. The rel-tive standard deviation of 3.8% indicated that the proposed assayad a good stability.

.4. Application of the immunoassay in clinical diagnosis

The magnetic electrochemical immunoassay strategy waspplied in JEV diagnosis of clinical samples. Under the optimizationonditions, JEV at 1.0 × 106 PFU/mL was serially diluted and tested.he results show that the increase of the cathodic current was pro-ortional to the logarithm of the titer of JEV in the range of 2 × 103

o 5 × 105 PFU/mL (Supplementary Fig. S5(b)). The linear regressionquation is i = 7.1232 lg CJEV–20.423 (R2 = 0.9816), where i is thencrease of cathodic current and CJEV is the titer of JEV in PFU/mL.he relative standard deviation of this approach was 9.38% (n = 11)ith the detection of 5 × 104 PFU/mL JEV. The minimal amount that

ould be detected was 2.0 × 103 PFU/mL, which is 2 orders of mag-itude lower than that of immunochromatographic strip (ICS) andimilar to that obtained from RT-PCR (Li et al., 2010). As a poten-ial preliminary screening tool for JEV, ICS has the characteristics ofapid, inexpensive and ease of handling. The major limitation of ICSs its low sensitivity which prevents the use of ICS for confirmatoryiagnosis of JEV. At present, RT-PCR with remarkable sensitivity isenerally recognized as gold standard in JEV confirmatory diagno-is. Nevertheless, its inherent drawbacks such as the high cost andhe requirement of sophisticate instrument limit the prevalence.he method developed in this study is rapid, sensitive, convenient,ompact and cost-effective, indicating that it can be employed as aromising fast screening and confirmatory diagnostic method forhe detection of JEV.

Several viruses associated with epizootic viral diseases in swineroduction were used as nonspecific virus samples to investigatehe specificity for the diagnosis of JEV (Supplementary Fig. S6).s expected, the negative results were obtained from pseudora-ies virus (PRV), porcine circovirus (PCV), porcine parvovirus (PPV),orcine respiratory syndrome virus (PRRSV), classical swine feverirus (CSFV) and swine influenza virus (SIV), suggesting that thisssay has high specificity for JEV.

The clinical specimens (n = 60), collected from mosquitoes,wine (brain tissue), and the human patients (the cerebrospinaluid), were detected by this method and RT-PCR, respectively.

s shown in Table 1, the proposed method had an efficiency of5% compared to the results of RT-PCR. This is telling that theroposed method can be satisfactorily used to detect JEV in clinicaliagnostic test.

ronics 26 (2011) 4253–4256

4. Conclusions

In summary, we have reported a novel MBs-based electro-chemical immunoassay strategy for the diagnosis of JEV basedon the homemade magnetic gold electrode for the capture anddirect sensing. It effectively reduces the complexity and timerequired for sensing applications. MCNTs were successfully intro-duced to improve the electric conductivity and enhance thesensitivity. The presented strategy was applied in the clinicaldiagnosis of JEV, and had a good diagnostic agreement with theresults from RT-PCR. This impressive, sensitivity, high specificityand low-cost electrochemical immunoassay provides a new con-cept for capture and direct sensing, and has a great promisefor clinical diagnosis, environmental monitoring and food analy-sis.

Acknowledgements

This work was supported by the National Natural Science Foun-dation of China (Grant No. 20005005), the State Key Laboratoryof Agricultural Microbiology, Huazhong Agricultural University)(Grant No. AML-200905) and Wuhan Chenguang Plan (Grant No.201050231070).

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at doi:10.1016/j.bios.2011.04.028.

References

Aytur, T., Foley, J., Anwar, M., Boser, B., Harris, E., Beatty, P.R., 2006. J. Immunol.Methods 314, 21–29.

Baldrich, E., Gomez, R., Gabriel, G., Munoz, F.X., 2011. Biosens. Bioelectron. 26,1876–1882.

Campuzano, S., De vila, B.E., Yuste, J., Pedrero, M., Garcia, J.L., Garcia, P., Garcia, E.,Pingarron, J.M., 2010. Biosens. Bioelectron. 26, 1225–1230.

Chaki, N.K., Vijayamohanan, K., 2002. Biosens. Bioelectron. 17, 1–12.De la Escosura-Muniz, A., Maltez-da, C.M., Sanchez-Espinel, C., Diaz-Freitas, B.,

Fernandez-Suarez, J., Gonzalez-Fernandez, F., Merkoci, A., 2010. Biosens. Bio-electron. 26, 1710–1714.

Du, D., Zou, Z.X., Shin, Y.S., Wang, J., Wu, H., Engelhard, M.H., Liu, J., Aksay, I.A., Lin,Y.H., 2010. Anal. Chem. 82, 2989–2995.

Fu, Z.F., Yan, F., Liu, H., Lin, J.H., Ju, H.X., 2008. Biosens. Bioelectron. 23, 1422–1428.Goon, I.Y., Lai, L.M.H., Lim, M., Amal, R., Gooding, J.J., 2010. Chem. Commun. 46,

8821–8823.Hervas, M., Lopez, M.A., Escarpa, A., 2009. Anal. Chim. Acta 653, 167–172.Kerman, K., Endo, T., Tsukamoto, M., Chikae, M., Takamura, Y., Tamiya, E., 2007.

Talanta 71, 1494–1499.Liebana, S., Lermo, A., Campoy, S., Cortes, M.P., Alegret, S., Pividori, M.I., 2009. Biosens.

Bioelectron. 25, 510–513.Li, Y.M., Hou, L.D., Ye, J., Liu, X.Q., Dan, H.B., Jin, M.L., Chen, H.C., Cao, S.B., 2010. J.

Virol. Methods 168, 51–56.Lucarelli, F., Marrazza, G., Turner, A.P.F., Mascini, M., 2004. Biosens. Bioelectron. 19,

515–530.Misra, U.K., Kalita, J., 2010. Prog. Neurobiol. 91, 108–120.Wang, H.Y., Li, Y.X., Liang, X.F., Liang, G.D., 2009. Jpn. J. Infect. Dis. 62, 331–336.Nie, H.G., Liu, S.J., Yu, R.Q., Jiang, J.H., 2009. Angew. Chem. Int. Ed. 48, 9862–9866.Palecek, E., Fojta, M., 2007. Talanta 74, 276–290.Shukla, J., Khan, M., Tiwari, M., Sannarangaiah, S., Sharma, S., Rao, P.V.L., Parida, M.,

2009. Diagn. Micro. Infect. Dis. 65, 142–149.Van den Hurk, A.F., Ritchie, S.A., Mackenzie, J.S., 2009. Ann. Rev. Entomol. 54, 17–35.

Zhong, Z.Y., Li, M.X., Xiang, D.I., Dai, N., Qing, Y., Wang, D., Tang, D.P., 2009. Biosens.Bioelectron. 24, 2246–2249.

Zacco, E., Pividori, M.I., Alegret, S., 2006. Anal. Chem. 78, 1780–1788.