Biosensors and Bioelectronics 58 (2014) 68–74
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
DNA-based hybridization chain reaction for an ultrasensitive cancer marker EBNA-1 electrochemical immunosensor Chao Song a, Guoming Xie a, Li Wang a, Lingzhi Liu a, Guang Tian b, Hua Xiang a,n a Key Laboratory of Laboratory Medical Diagnostics of Education, Department of Laboratory Medicine, Chongqing Medical University, No. 1 Yi Xue Yuan Road, Chongqing 400016, PR China b Guiyang Medical College, Guiyang 55004, PR China
art ic l e i nf o
a b s t r a c t
Article history: Received 4 December 2013 Received in revised form 10 February 2014 Accepted 11 February 2014 Available online 28 February 2014
An ultrasensitive and selective electrochemical immunosensor was developed for the detection of Epstein Barr virus nuclear antigen 1 (EBNA-1). Firstly, a suspension of graphene sheets (GS) and multiwalled carbon nanotubes (MWCNTs) was prepared with the aid of chitosan (CS) solution and then modified on a glassy carbon electrode (GCE). Gold nanoparticles (AuNPs) were then electrodeposited onto the surface of the GS–MWCNTs film by cyclic voltammetry (CV) to immobilize the captured antibodies. After that, specific sandwich immunoreactions were formed among the captured antibody, EBNA-1, and secondary antibody, DNA-coated carboxyl multi-wall carbon nanotubes (DNA–MWCNTs– Ab2). DNA initiator strands (S0) and secondary antibodies linked to the MWCNTs and double-helix DNA polymers were obtained by hybridization chain reaction (HCR), and here S0 on the MWCNTs propagates a chain reaction of hybridization events between two alternating hairpins to form a nicked double-helix. Finally, electroactive indicator doxorubicin hydrochloride was intercalated into the CG–GC steps between the HCR products and could produce an electrochemical signal, which was monitored by differential pulse voltammetry (DPV). Under optimum conditions, the amperometric signal increased linearly with the target concentrations (0.05–6.4 ng mL 1), and the immunosensor exhibited a detection limit as low as 0.7 pg mL 1 (S/N ¼3). The proposed method showed acceptable stability and reproducibility, as well as favorable recovery for EBNA-1 in human serum. The proposed immunosensor provides a novel avenue for signal amplification and potential applications in bioanalysis and clinical diagnostics. & 2014 Elsevier B.V. All rights reserved.
Keywords: EBNA-1 Hybridization chain reaction GS–MWCNTs Doxorubicin hydrochloride Immunosensor
1. Introduction The Epstein–Barr virus (EBV) is a ubiquitous human herpesvirus that infects a majority of the world's population, that is 490% (Marrazza et al., 1999). EBV infection usually takes places asymptomatically during childhood, but it is possible to be symptomatic when infection is put off until adolescence (Sixbey et al., 1986; Yao et al., 1985). EBV has been associated with several lymphoid and epithelial cell malignancies, including Burkitt lymphoma, Hodgkin disease (HD), immunodeficiency-related B cell lymphoma, gastric carcinoma, and nasopharyngeal carcinoma (NPC) (Thompson and Kurzrock, 2004). In these diseases, Epstein–Barr nuclear antigen 1 (EBNA-1) is the only protein consistently expressed and one of the earliest viral proteins expressed after infection (Garai-Ibabe et al., 2012; Humme,
n
Corresponding author. Tel./fax.: þ 86 23 68485009. E-mail address: [email protected] (H. Xiang).
http://dx.doi.org/10.1016/j.bios.2014.02.031 0956-5663 & 2014 Elsevier B.V. All rights reserved.
2003; Middeldorp et al., 2003). EBNA-1 plays a fundamental role in replication and transcription of the viral genome (Young and Murray, 2003). It is important to highlight that EBNA-1 is not only a EBV related cancer marker (Garai-Ibabe et al., 2011), but is also discriminated from infection by other pathogens. The previously reported methods to detect EBNA-1 were gel electrophoresis (Ambinder et al., 1990) and immunoprecipitation (Canaan et al., 2009), which are time-consuming, expensive, labor-intensive and not suitable for clinical applications. Compared with conventional methods, electrochemical biosensors exhibit several advantages, such as high sensitivity, small analyte volume, and simple pretreatment (Garai-Ibabe et al., 2011, 2012). Therefore, we developed a new method to make biosensors for detection of EBNA-1. Graphene sheets (GS) are the flat monolayers of carbon atoms tightly packed into a two-dimensional honeycomb lattice (Brownson and Banks, 2010). GS have been extensively used as immobilized substrates for the development of biosensors (Jiang et al., 2013; Wang et al., 2014) due to rapid electron transportation, high electrical conductivity and good biocompatibility (Myung
C. Song et al. / Biosensors and Bioelectronics 58 (2014) 68–74
et al., 2011; Sun et al., 2011). However, GS are generally insoluble in water and may leak out upon immobilization onto the electrode surface (Saha et al., 2010). Chitosan (CS) is a biological cationic macromolecule with abundant –NH2 and –OH functional groups (Banks et al., 2005; Lu et al., 2002). A water-soluble polymer, CS was used as a disperser to prepare homogeneous GS solution by the covalent functionalization method (Kang et al., 2009). However, the introduction of CS could not promote electron transfer well. To avoid this problem, GS–MWCNTs hybrid was used. The MWCNTs have extensively been used due to excellent electron conductivity and high surface area-to-volume ratio, which proved to promote electron transfer (Kauffman et al., 2010; Wang et al., 2006). Ma et al. have prepared a GS and MWCNTs hybrid (Yen et al., 2011). Herein, we tried to combine GS with MWCNTs via π–π stacking. Au nanoparticles (AuNPs) have been widely used for their chemical stability, strong adsorption ability, and excellent electrical conductivity (Parker et al., 2010); meanwhile, they also have large capacity to immobilize protein (antibody) via Au-NH2 (Lu et al., 2012a; Ma et al., 2013). Furthermore, several studies have been reported on functional MWCNTs loaded with HRP and Ab2 based on a covalent bond. Xie et al. have constructed an electrochemical immunosensor based on multi-sHRP-DNA-coated MWNCTs as signal labels with high sensitivity (Ma et al., 2013). Recently, DNA-based hybridization chain reaction (HCR) was extensively used for the signal amplification (Liu et al., 2013; Wang et al., 2013; Zhou et al., 2012; Zhuang et al., 2013) in the electrochemical biosensors. Tang et al. and Chen et al. have constructed an electrochemical biosensor based on
69
nanogold-based biobar codes for label-free immunosensing of proteins coupled with an in situ HCR, which has gained satisfactory results for CEA detection (Zhang et al., 2012). The concept of HCR was firstly described by Dirks (2004). HCR is an enzyme-free process; the DNA initiator strand (S0) propagated a cascade of hybridization events between alternating H1 and H2 hairpin probes to yield a nicked double-helix. Doxorubicin hydrochloride was used as an electroactive indicator, which bound to double stranded DNA to produce an electrochemical signal (Wang et al., 1998; Zhang et al., 2009). Doxorubicin hydrochloride can remarkably improve the sensitivity of the immunosensor. Xie et al. used doxorubicin hydrochloride to fabricate an immunosensor that showed a higher sensitivity than the one without it (Lu et al., 2012b). However, to the best of our knowledge, the immunosensor based on the HCR using doxorubicin hydrochloride intercalated the long DNA concatemers, as an electroactive indicator has not yet been reported. In this study, we attempted to combine the specificity of electrochemistry immunoassay with the signal amplification capability of the HCR to construct an electrochemical immunosensor (Scheme 1). The protocol mainly involves the following: (i) The construction of the sandwiched immunocomplex between the immobilized Ab1 on the AuNPs/GS–MWCNTs/GCE and Ab2 on the MWCNTs conjugated with DNA. The double strand DNA was formed from HCR between initiator strands (S0) on the S0–MWCNTs–Ab2 and a stable mixture of H1 and H2 hairpins. (ii) Doxorubicin hydrochloride was used as an electroactive indicator in the system. (iii) DPV was used to measure the electrochemical signal, which is correlated with EBNA-1
Scheme 1. (a) The preparation of DNA–MWCNTs–Ab2; (b) the fabricating procedure for EBNA-1 immunosensor .
70
C. Song et al. / Biosensors and Bioelectronics 58 (2014) 68–74
concentration. Therefore, the system can indirectly determine the concentration of EBNA-1. The signal amplification function of HCR improved the sensitivity of the immunosensor.
2. Experimental 2.1. Reagents EBV Nuclear Antigen 1, monoclonal EBV Nuclear antibody (Ab1) and polyclonal EBV Nuclear antibody (Ab2) were obtained from Abcam Ltd. (Hong Kong, China). Graphene sheets (GS) were purchased from Pioneer Nanotechnology Co. (Nanjing, China). Multi-walled carbon nanotubes with carboxylic acid groups (MWCNTs) were acquired from Chengdu Organic Chemicals Co. Ltd. (Chengdu, China). Bovine serum albumin (BSA, 96–99%), chloroauric acid (HAuCl4), N-hydroxysuccinimide (NHS), 1-ethyl-3-3(3-dimethylaminopropyl) carbodiimide (EDC), and mecatohexanol (MCH) were produced from Sigma-Aldrich (St. Louis, MO, USA). 2-(N-morpholino) ethanesulfonic acid buffer (MES buffer), doxorubicin hydrochloride and chitosan (CS) were obtained from Shanghai Sangon Biological Engineering Technology & Services Co. Ltd. (Shanghai, China). Deionized distilled water purified through a Millipore system (Z18 MΩ) was employed in all runs. All other reagents were of analytical grade and were used without further purification. All oligonucleotide sequences including S0 and two hairpins were ordered from Shanghai Sangon Biological Engineering Technology & Services Co. Ltd. (Shanghai, China). The sequences are listed in Table S1. 2.2. Apparatus All electrochemical measurements were carried out at room temperature (RT) (25 70.5 1C) using a CHI660D electrochemical workstation (Shanghai Chenhua Instruments, China) with a conventional three-electrode cell. A bare or modified glassy carbon electrode (GCE) was used as the working electrode, while Ag/AgCl (in saturated KCl) and a platinum (Pt) wire were served as the reference and the counter electrodes, respectively. The morphology of GS–MWCNTs nanocomposite was characterized by a scanning electron microscopy (SEM) system (FEI Nova-400, USA). An atomic force microscope (AFM) NanoWizard II (JPK Instruments Inc., Germany) was used to observe images of the MWCNTs and DNA–MWCNTs–Ab2.
400 mM EDC and 100 mM NHS solution for 2 h. The activated MWCNTs were separated from the free EDC/NHS by centrifugation. Next, S0 (21 mL, 1 OD) and Ab2 (50 mL, 25 mg mL 1) were added to the activated MWCNTs and stirred for 12 h at RT. By this step, S0 and Ab2 were chemically bonded to MWCNTs via the formation of an amide bond between the –NH2 groups of Ab2, S0, and the –COOH groups of MWCNTs. Following that, the mixture was centrifuged (8000 rpm) for 10 min at 4 1C. The bioconjugates were collected and washed with phosphate buffer solution (PBS, pH 7.0) three times to eliminate the free S0 and Ab2. Afterward, 1 mM MCH was added into S0–MWCNTs–Ab2 for 1 h to block the remaining sites. After centrifugation, the mixture was resuspended in Tris–HCl buffer (pH 7.0, 600 mL) and hybridization chain reaction was carried out by employing two hairpin species (H1 and H2). About 10 mM H1 and H2 was injected into S0–MWCNTs–Ab2 and incubated at 37 1C for 80 min. The resulting DNA–MWCNTs–Ab2 bioconjugates were washed thrice with Tris–HCl buffer (pH 7.0, 10 mM). Ultimately, DNA–MWCNTs–Ab2 bioconjugates were store at 4 1C for future use. 2.5. Fabrication of immunosensor The stepwise procedure for the electrochemical immunosensor is illustrated in Scheme 1b. The fabrication of the electrochemical immunosensor consists of the following steps: (i) Prior to use, the glassy carbon electrodes (GCE, 3 mm diameter) were polished to a mirror-like surface with 1.0, 0.3, and 0.05 mm alumina slurry sequentially, followed by sonication in deionized water/acetone/ deionized water for 5 min and dried at RT. (ii) About 8 mL of the asprepared GS–MWCNTs homogeneous suspension was initially deposited on the electrode surface, and then dried naturally at RT. (iii) The GS–MWCNTs/GCE electrode was immersed into 3 mM HAuCl4 solution and electrodeposition was performed by cyclic voltammetry at a potential from 0.2 to 0.6 V for 100 seconds to obtain the AuNPs/GS–MWCNTs modified electrode. (iv) Approximately 8 mL of Ab1 solution (50 mg mL 1) in PBS was spread on the AuNPs/GS–MWCNTs modified electrode, followed by overnight incubation at 4 1C. (v) The modified electrode was blocked by dipping in 0.25% BSA solution for 30 min at RT. After another round of washing with PBS (pH 7.0, 0.01 M), the immunosensor was obtained. 2.6. Electrochemical measurement
2.3. Synthesis of GS–MWCNTs nanocomposite GS–MWCNTs nanocomposite was synthesized according to the literature (Diaconu et al., 2010; Jiang et al., 2005; Wu et al., 2009) with slight modification. Firstly, CS solution (0.5 wt%) was prepared by dissolving 250 mg of CS powder in 50 mL of 1% (v/v) acetic acid solution with 2 h ultrasonication to form a homogeneous solution. The pH of the CS solution was adjusted to 4.0–4.5 using 1.0 M NaOH. Then, 1 mg of GS and MWCNTs was dispersed in 1 mL of CS solution, followed by 2 h ultrasonication to make black homogeneous GS–MWCNTs. 2.4. Preparation of DNA–MWCNTs–Ab2 DNA–MWCNTs–Ab2 were synthesized and prepared according to a previous publication (Ma et al., 2013) with light modification, as illustrated in Scheme 1a. Firstly, 1 mg of MWCNTs was dissolved thoroughly in 2-(N-morpholino) ethanesulfonic acid buffer (MES buffer, 1 mL, 50 mM, pH 7.0) and the pH of MES was adjusted to 7.0 with 1.0 M NaOH. The mixture was treated ultrasonically for 3 h until a uniform gloss suspension was obtained. Then, the –COOH groups of MWCNTs were activated by treatment with 1 mL of
In this work, EBNA-1 samples were assayed with a sandwichtype immunoassay format using DNA–MWCNTs–Ab2 as the secondary antibody and doxorubicin hydrochloride as the electroactive indicator. Various concentrations of EBNA-1 (Ag) were tested. The assay was carried out as follows: (i) The blocked immunosensors were incubated with different antigen concentrations for 30 min, followed by washing three times with PBS to remove nonspecifically bound conjugates. (ii) The as-prepared DNA–MWCNTs–Ab2 (8 mL) was deposited onto the immunosensor and incubated for another 30 min to form a sandwiched immunocomplex. (iii) After rinsing with PBS, the modified electrode was immersed into PBS (pH 7.0, 0.01 M) solution containing 5 mM doxorubicin hydrochloride for 40 min, followed by rinsing three times with PBS to remove uninteracted doxorubicin hydrochloride. The electrochemical measurement with differential pulse voltammetry (DPV) values was carried out from 0.3 to 0.9 V at 50 mV s 1 in pH 7.5 PBS. All measurements were conducted at RT (2572 1C). The standards were prepared using EBNA-1 solution at different concentrations. Blood specimens were gifted by the First People's Hospital, Jiulongpo District of Chongqing, and the First Affiliated Hospital, Chongqing Medical University.
C. Song et al. / Biosensors and Bioelectronics 58 (2014) 68–74
71
Fig. 1. SEM images of (a) GS–MWCNTs and (b) AuNPs/GS–MWCNTs. AFM images of (c) MWCNTs and (d) DNA–MWCNTs–Ab2.
3. Results and discussion
3.2. Cyclic voltammetric characterization
3.1. Characterization of the GS–MWCNTs, AuNPs/GS–MWCNTs, MWCNTs and DNA–MWCNTs–Ab2
The different modified electrodes were investigated by cyclic voltammetry in PBS (pH 7.5, 0.01 M) containing 5 mM K3Fe(CN)6/ K4Fe(CN)6 and 0.1 M KCl at a scan rate of 50 mV s 1. As expected, there was an obvious increase in current responses after GS–MWCNTs modified on the electrode (Fig. 2b) when compared with bare GCE (Fig. 2a). The peak current obtained could be ascribed to the redox behavior of [Fe(CN)6]3 /4 . Then, after the electrodeposition of AuNPs onto the GS–MWCNTs/GCE surface, the highest redox peaks were obtained, with well-defined redox peaks at 310 and 181 mV (peak potential difference of 129 mV) (Fig. 2c). AuNPs on the modified electrode could effectively enhance the electron transfer between [Fe(CN)6]3 /4 and the electrode. With the immobilization of the captured antibodies and BSA blocking solution on the modified electrode surface, redox peak currents gradually decreased (Fig. 2d and e). This finding suggested that the Ab1 was successfully immobilized on the electrode surface and the transmission of ferricyanide to the electrode surface was hindered. When the EBNA-1 was captured by its antibody, the peak current further decreased (Fig. 2f), which was consistent with the fact that the hydrophobic layer of the protein insulated the conductive support (Yang et al. 2013). The peak current increased with the DNA–MWCNTs–Ab2 immobilization on the electrode (Fig. 2g),
The morphologies of GS–MWCNTs and AuNPs/GS–MWCNTs were characterized using field-emission scanning electronic microscopy (SEM). Fig. 1a shows an SEM image obtained for the GS–MWCNTs modified surface, GS–MWCNTs dispersed by the CS solution. This material could obviously increase the electrode effective surface and enhance the electron transfer ability. As seen from Fig. 1b, a large number of gold nanoparticles were coated on the modified materials (relative to the GS–MWCNTs, Fig. 1a), indicating that AuNPs could be successfully electrodeposited on the modified electrode. To demonstrate the successful conjugation of DNA and Ab2 on the MWCNTs, the nanomaterial before and after modification was monitored by AFM. Two three-dimensional height maps display AFM images of MWCNTs and DNA–MWCNTs– Ab2 (Fig. 1c and d, respectively). As shown in Fig. 1c, a dense and spiky appearance was observed, which is consistent with a previous report (Ma et al., 2013). Then, due to the bioconjugation of DNA and Ab2, the result of AFM showed a blunt image in Fig. 1d, indicating that DNA and Ab2 were successfully immobilized onto MWCNTs (Ma et al., 2013).
72
C. Song et al. / Biosensors and Bioelectronics 58 (2014) 68–74
to 2116 Ω after EBNA-1 was successively absorbed onto the electrode surface (Fig. 3f). Subsequently, after DNA–MWCNTs– Ab2 was immobilized on the electrode surface, the Ret decreased to 1196 Ω in turn. So, the EIS results were in accordance with the CV curves shown in Fig. 2. 3.4. Comparison of the proposed protocol with a control experiment
Fig. 2. CVs of different electrodes in PBS (pH 7.5) containing 5 mM K3Fe(CN)6/K4Fe (CN)6 and 0.1 M KCl at a scan rate of 50 mV s 1: (a) bare GCE electrode; (b) GS– MWCNTs/GCE; (c) AuNPs/GS–MWCNTs/GCE; (d) Ab1/AuNPs/GS–MWCNTs/GCE; (e) BSA/Ab1/AuNPs/GS–MWCNTs/GCE; (f) EBNA-1/BSA/Ab1/AuNPs/GS–MWCNTs/ GCE; and (g) DNA–MWCNTs–Ab2 /EBNA-1/BSA/Ab1/AuNPs/GS–MWCNTs/GCE.
To demonstrate that HCR can truly improve the signal amplification, a HCR-free control experiment was performed (Fig. S1). As expected, when detecting the same concentration EBNA-1 (400 pg mL 1), the HCR-free immunosensor displayed lower peak current (Fig. S1a) compared with the HCR immunosensor (Fig. S1b). This result is attributed to the fact that HCR provides a lot of sites to the intercalated amount of doxorubicin hydrochloride, which greatly enhances detection signals. In order to prove that GS–MWCNTs and AuNPs could really improve the analytical performance of the immunoassay, the characteristics of the biosensor were investigated by CV (Fig. S2). The peak currents of AuNPs/MWCNTs/GCE (cure a) and AuNPs/GS/ GCE (cure b) increased gradually, but the signal of AuNPs/GS– MWCNTs/GCE (cure c) was obviously the highest under the same conditions, which should be attributed to the reasonable combination of GS–MWCNTs and AuNPs. 3.5. Optimization of experimental conditions
Fig. 3. EIS spectra for each immobilization step: (a) bare GCE electrode; (b) GS– MWCNTs/GCE; (c) AuNPs/GS–MWCNTs/GCE; (d) Ab1/AuNPs/GS–MWCNTs/GCE; (e) BSA/Ab1/AuNPs/GS–MWCNTs/GCE; (f) EBNA-1/BSA/Ab1/AuNPs/GS–MWCNTs/ GCE; and (g) DNA–MWCNTs–Ab2 /EBNA-1/BSA/Ab1/AuNPs/GS–MWCNTs/GCE.
which was attributed to the good electronic conductivity of the MWCNTs. 3.3. Electrochemical impedance spectroscopy characterization Electric impedance spectroscopy (EIS) is an effective method to monitor the impedance changes of surface-transfer electrodes. In the Nyquist diagram, the semicircle diameter equals the charge transfer resistance (Ret). Fig. 3 shows the characteristic EIS corresponding to the stepwise modification processes. Compared with the bare GCE electrode (Fig. 3a), the diameter of the high frequency semicircle decreased when GS–MWCNTs film (Fig. 3b) was modified on the GCE surface. This result can be attributed to the coupling of GS–MWCNTs producing a synergic effect, such as excellent electroconductivity. After AuNPs were electrodeposited on the modified electrode, the Ret was close to zero (Fig. 3c). This result indicates that AuNPs accelerate the electron transfer process of [Fe(CN)6]3 /4 . However, after Ab1 and BSA were immobilized on the electrode surface, the Ret obviously increased to 625 Ω and 1279 Ω (Fig. 3d and e), respectively, because these biomacromolecules obstructed the electron transfer. The Ret further increased
To achieve maximum assay performance, various experimental parameters such as the ratio of GS to MWCNTs, the HCR hybridization reaction time, Ab–Ag interaction time, and the pH of the PBS solution were optimized by DPV. These experiments were conducted using EBNA-1 at 400 pg mL 1. The effect of ratio of GS to MWCNTs was examined to improve the analytical performance with different GS:MWCNT ratios (3:1; 2:1; 1:1; 1:2; 1:3). The highest peak current was achieved when the ratio was 1:1. Thus, the optimal GS:MWCNTs ratio was chosen as 1:1 (Fig. S3a). The effect of HCR time was also investigated in the range of 10 to 90 min (Fig. S3b). The current response gradually increased with HCR time from 10 to 70 min, and reached plateaus at 70 min. This result indicated that the HCR was primitively completed at 70 min. Thus, the optimum HCR time was considered as 70 min. The effect of Ab and Ag interaction time was studied in the range of 10–35 min (Fig. S3c). The current response gradually increased from 10 to 30 min and reached plateaus at 30 min. Hence, the optimum interaction time of 30 min was used in this study. The effect of the pH of the PBS solution was examined over the pH range of 4.5–9.5. Fig. S3d shows that the current response gradually increased from pH 4.5 to pH 7.5 and then decreased. Thus, the pH was considered to be 7.5. 3.6. Amperometric response and calibration curve Analytical calibrations for the EBNA-1 were recorded by DPV under optimal conditions. Fig. 4 shows that the peak currents of doxorubicin hydrochloride increased with increased concentrations of EBNA-1. As can be seen in the inset of Fig. 4, there is a linear relationship between the peak currents and the logarithm of the concentrations of EBNA-1 in the range from 0.05 to 6.4 ng mL 1. The peak current values were obtained from the mean value with three independent experiments. The linear regression equation was Y (mA) ¼24.81X 19.11, with a linear correlation coefficient (R2) of 0.99249, and the detection limit (LOD) was estimated to be 0.7 pg mL 1 (S/N¼3). Therefore, the proposed electrochemical immunosensor had a good analytical performance for the detection of EBV nuclear antigen (EBNA-1).
C. Song et al. / Biosensors and Bioelectronics 58 (2014) 68–74
73
4.0%, and 5.2%, respectively. The results confirmed that the immunosensor exhibited acceptable precision and fabrication reproducibility. Regeneration is also a key parameter in immunosensor development. The surface of the prepared immunosensor was immersed in 8 M urea solution after the completion of each assay, followed by rinsing with double distilled water to dissociate the antigen–antibody complex. After five cycles of regeneration, the immunosensor retained 88.2% of its origin current, and the relative standard deviation (RSD) was 3.5%. 3.8. Recovery
Fig. 4. The DPV responses of the different concentrations of EBNA-1 detected by the proposed immunoassay (from (a) to (i): 0, 50, 100, 200, 400, 800, 1600, 3200, and 6400 pg mL 1); inset: the calibration curve of the peak current response versus the logarithm of EBNA-1 concentration with the proposed immunoassay.
4. Conclusion
Table 1 Recovery studies of EBNA-1 from human serum samples. EBNA-1 added (pg mL 1)
Sample no.
EBNA-1 in serum (pg mL 1)
1
0
50.00
2
0
100.00
3
0
200.00
Recovery tests were performed to evaluate the feasibility of the immunosensor for clinical applications. Serum samples spiked with EBNA-1 at concentrations of 50, 100 and 200 pg mL 1 were tested using the proposed immunosensor, and the corresponding results are given in Table 1. The recovery values ranged from 101.7% to 102.1%. This result shows that the developed immunosensor has potential clinical applicability for the detection of EBNA-1.
EBNA-1 detected (pg mL 1)
RSD% (n¼ 5)
R (%)
50.18, 49.04, 52.64, 50.94, 51.71 101.24, 98.21, 101.83, 102.63, 103.47 207.83, 204.79, 195.57, 205.62, 206.85
2.61
101.7
1.98
101.4
2.41
102.1
R: recovery percentage and RSD: relative standard deviation percentage.
The analytical performance of our EBV immunosensor has been compared to those reported previously in Table S2. It is clear that the proposed biosensor displayed a better performance than other methods in the analytical range and detection limit. 3.7. Stability, selectivity, reproducibility, repeatability and regeneration of the immunosensor In this study, the stability of the immunosensor was evaluated. When the biosensor was stored in the refrigerator at 4 1C for 9 days, the response only decreased by 8.1%, indicating good stability. In order to evaluate the binding specificity of the immunosensor to EBNA-1, some possible interfering agents such as BSA, HBV surface antigen (HBsAg), human IgG and CEA were tested under the same experimental conditions. As shown in Fig. S4, when the biosensor was incubated 400 pg mL 1 pure EBNA-1 and 1 ng mL 1 interfering agents, a significant current response with EBNA-1 was observed compared to BSA, HBsAg, human IgG and CEA, suggesting the high selectivity of the immunosensor. The repeatability and reproducibility of the immunosensor were also investigated by the variation coefficients (CVs), by making 5 successive measurements of three concentrations of EBNA-1 using an identical immunosensor. The CVs of the intraassay were determined as 2.5%, 4.0%, and 2.8% at 50, 100, and 200 pg mL 1 of EBNA-1, respectively. Similarly, a set of 5 immunosensors was prepared to detect 50, 100, and 200 pg mL 1 of EBNA-1, and the inter-assay CVs of the measurement were 3.9%,
In summary, an ultrasensitive electrochemical sandwich immunosensor was developed using AuNPs/GS–MWCNTs as a sensor platform and coupling MWCNTs with the hybridization chain reaction (DNA–MWCNTs–Ab2) as a signal amplifying probe. GS– MWCNTs and AuNPs were, respectively, assembled onto the GCE electrode surface to greatly enhance the conductivity and improve effective surface area, which increased the amount of antibodies to be immobilized on the sensor. More significantly, a novel amplification strategy using DNA–MWCNTs–Ab2 was introduced, which exhibited a broad dynamic range and low detection limit. Basis on this study, more electrochemical immunosensors could be designed for the detection of other target analytes.
Acknowledgments This work was financially supported by the National Natural Science Foundation of China (81171415) and the Foundation of National Key Discipline in Laboratory Medicine, Chongqing Medical University, China (2010103).
Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios.2014.02.031. References Ambinder, R.F., Shah, W., Rawlins, D., Hayward, G.S., Hayward, S.D., 1990. J. Virol. 64 (5), 2369–2379. Banks, C.E., Davies, T.J., Wildgoose, G.G., Compton, R.G., 2005. Chem. Commun. 7, 829–841. Brownson, D.A., Banks, C.E., 2010. Analyst 135 (11), 2768–2778. Canaan, A., Haviv, I., Urban, A.E., Schulz, V.P., Hartman, S., Zhang, Z., Palejev, D., Deisseroth, A.B., Lacy, J., Snyder, M., 2009. Proc. Natl. Acad. Sci. 106 (52), 22421–22426. Diaconu, M., Litescu, S.C., Radu, G.L., 2010. Sens. Actuators B: Chem. 145 (2), 800–806. Dirks, R.M., 2004. Proc. Natl. Acad. Sci. 101 (43), 15275–15278. Garai-Ibabe, G., Grinyte, R., Canaan, A., Pavlov, V., 2012. Anal. Chem. 84 (14), 5834–5837. Garai-Ibabe, G., Grinyte, R., Golub, E.I., Canaan, A., de la Chapelle, M.L., Marks, R.S., Pavlov, V., 2011. Biosens. Bioelectron. 30 (1), 272–275. Humme, S., 2003. Proc. Natl. Acad. Sci. 100 (19), 10989–10994.
74
C. Song et al. / Biosensors and Bioelectronics 58 (2014) 68–74
Jiang, L., Wang, R., Li, X., Jiang, L., Lu, G., 2005. Electrochem. Commun. 7 (6), 597–601. Jiang, S., Hua, E., Liang, M., Liu, B., Xie, G., 2013. Colloids Surf. B: Biointerfaces 101, 481–486. Kang, X., Wang, J., Wu, H., Aksay, I.A., Liu, J., Lin, Y., 2009. Biosens. Bioelectron. 25 (4), 901–905. Kauffman, D.R., Sorescu, D.C., Schofield, D.P., Allen, B.L., Jordan, K.D., Star, A., 2010. Nano Lett. 10 (3), 958–963. Liu, S., Wang, Y., Ming, J., Lin, Y., Cheng, C., Li, F., 2013. Biosens. Bioelectron. 49, 472–477. Lu, G., Yao, X., Zhou, X., Zhan, T., Long, D., 2002. Chem. Res. Chin. Univ. 18 (3), 316–320. Lu, J., Liu, S., Ge, S., Yan, M., Yu, J., Hu, X., 2012a. Biosens. Bioelectron. 33 (1), 29–35. Lu, L., Liu, B., Zhao, Z., Ma, C., Luo, P., Liu, C., Xie, G., 2012b. Biosens. Bioelectron. 33 (1), 216–221. Ma, C., Liang, M., Wang, L., Xiang, H., Jiang, Y., Li, Y., Xie, G., 2013. Biosens. Bioelectron. 47 (0), 467–474. Marrazza, G., Chianella, I., Mascini, M., 1999. Biosens. Bioelectron. 14 (1), 43–51. Middeldorp, J.M., Brink, A.A., van den Brule, A.J., Meijer, C.J., 2003. Crit. Rev. Oncol./ Hematol. 45 (1), 1–36. Myung, S., Solanki, A., Kim, C., Park, J., Kim, K.S., Lee, K.B., 2011. Adv. Mater. 23 (19), 2221–2225. Parker, J.F., Fields-Zinna, C.A., Murray, R.W., 2010. Acc. Chem. Res. 43 (9), 1289–1296. Saha, A., Basiruddin, S., Ray, S., Roy, S., Jana, N.R., 2010. Nanoscale 2 (12), 2777–2782. Sixbey, J., Lemon, S., Pagano, J., 1986. Lancet 328 (8516), 1122–1124.
Sun, J.-Y., Huang, K.-J., Wei, S.-Y., Wu, Z.-W., Ren, F.-P., 2011. Colloids Surf. B: Biointerfaces 84 (2), 421–426. Thompson, M.P., Kurzrock, R., 2004. Clin. Cancer Res. 10 (3), 803–821. Wang, C., Zhou, H., Zhu, W., Li, H., Jiang, J., Shen, G., Yu, R., 2013. Biosens. Bioelectron. 47, 324–328. Wang, H.-S., Li, T.-H., Jia, W.-L., Xu, H.-Y., 2006. Biosens. Bioelectron. 22 (5), 664–669. Wang, J., Ozsoz, M., Cai, X., Rivas, G., Shiraishi, H., Grant, D.H., Chicharro, M., Fernandes, J., Paleček, E., 1998. Bioelectrochem. Bioenerg. 45 (1), 33–40. Wang, L., Hua, E., Liang, M., Ma, C., Liu, Z., Sheng, S., Liu, M., Xie, G., Feng, W., 2014. Biosens. Bioelectron. 51, 201–207. Wu, H., Wang, J., Kang, X., Wang, C., Wang, D., Liu, J., Aksay, I.A., Lin, Y., 2009. Talanta 80 (1), 403–406. Yang, F., Chai, Y., Yuan, R., Han, J., Yuan, Y., Liao, N., Yang, Z., 2013. Anal. Methods 5, 5279–5285. Yao, Q.Y., Rickinson, A.B., Epstein, M.A., 1985. Int. J. Cancer 35 (1), 35–42. Yen, M.-Y., Hsiao, M.-C., Liao, S.-H., Liu, P.-I., Tsai, H.-M., Ma, C.-C.M., Pu, N.-W., Ger, M.-D., 2011. Carbon 49 (11), 3597–3606. Young, L.S., Murray, P.G., 2003. Oncogene 22 (33), 5108–5121. Zhang, B., Liu, B., Tang, D., Niessner, R., Chen, G., Knopp, D., 2012. Anal. Chem. 84 (12), 5392–5399. Zhang, Y., Zhang, K., Ma, H., 2009. Anal. Biochem. 387 (1), 13–19. Zhou, J., Xu, M., Tang, D., Gao, Z., Tang, J., Chen, G., 2012. Chem. Commun. 48 (100), 12207. Zhuang, J., Fu, L., Xu, M., Zhou, Q., Chen, G., Tang, D., 2013. Biosens. Bioelectron. 45, 52–57.
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
DNA-based hybridization chain reaction for an ultrasensitive cancer marker EBNA-1 electrochemical immunosensor Chao Song a, Guoming Xie a, Li Wang a, Lingzhi Liu a, Guang Tian b, Hua Xiang a,n a Key Laboratory of Laboratory Medical Diagnostics of Education, Department of Laboratory Medicine, Chongqing Medical University, No. 1 Yi Xue Yuan Road, Chongqing 400016, PR China b Guiyang Medical College, Guiyang 55004, PR China
art ic l e i nf o
a b s t r a c t
Article history: Received 4 December 2013 Received in revised form 10 February 2014 Accepted 11 February 2014 Available online 28 February 2014
An ultrasensitive and selective electrochemical immunosensor was developed for the detection of Epstein Barr virus nuclear antigen 1 (EBNA-1). Firstly, a suspension of graphene sheets (GS) and multiwalled carbon nanotubes (MWCNTs) was prepared with the aid of chitosan (CS) solution and then modified on a glassy carbon electrode (GCE). Gold nanoparticles (AuNPs) were then electrodeposited onto the surface of the GS–MWCNTs film by cyclic voltammetry (CV) to immobilize the captured antibodies. After that, specific sandwich immunoreactions were formed among the captured antibody, EBNA-1, and secondary antibody, DNA-coated carboxyl multi-wall carbon nanotubes (DNA–MWCNTs– Ab2). DNA initiator strands (S0) and secondary antibodies linked to the MWCNTs and double-helix DNA polymers were obtained by hybridization chain reaction (HCR), and here S0 on the MWCNTs propagates a chain reaction of hybridization events between two alternating hairpins to form a nicked double-helix. Finally, electroactive indicator doxorubicin hydrochloride was intercalated into the CG–GC steps between the HCR products and could produce an electrochemical signal, which was monitored by differential pulse voltammetry (DPV). Under optimum conditions, the amperometric signal increased linearly with the target concentrations (0.05–6.4 ng mL 1), and the immunosensor exhibited a detection limit as low as 0.7 pg mL 1 (S/N ¼3). The proposed method showed acceptable stability and reproducibility, as well as favorable recovery for EBNA-1 in human serum. The proposed immunosensor provides a novel avenue for signal amplification and potential applications in bioanalysis and clinical diagnostics. & 2014 Elsevier B.V. All rights reserved.
Keywords: EBNA-1 Hybridization chain reaction GS–MWCNTs Doxorubicin hydrochloride Immunosensor
1. Introduction The Epstein–Barr virus (EBV) is a ubiquitous human herpesvirus that infects a majority of the world's population, that is 490% (Marrazza et al., 1999). EBV infection usually takes places asymptomatically during childhood, but it is possible to be symptomatic when infection is put off until adolescence (Sixbey et al., 1986; Yao et al., 1985). EBV has been associated with several lymphoid and epithelial cell malignancies, including Burkitt lymphoma, Hodgkin disease (HD), immunodeficiency-related B cell lymphoma, gastric carcinoma, and nasopharyngeal carcinoma (NPC) (Thompson and Kurzrock, 2004). In these diseases, Epstein–Barr nuclear antigen 1 (EBNA-1) is the only protein consistently expressed and one of the earliest viral proteins expressed after infection (Garai-Ibabe et al., 2012; Humme,
n
Corresponding author. Tel./fax.: þ 86 23 68485009. E-mail address: [email protected] (H. Xiang).
http://dx.doi.org/10.1016/j.bios.2014.02.031 0956-5663 & 2014 Elsevier B.V. All rights reserved.
2003; Middeldorp et al., 2003). EBNA-1 plays a fundamental role in replication and transcription of the viral genome (Young and Murray, 2003). It is important to highlight that EBNA-1 is not only a EBV related cancer marker (Garai-Ibabe et al., 2011), but is also discriminated from infection by other pathogens. The previously reported methods to detect EBNA-1 were gel electrophoresis (Ambinder et al., 1990) and immunoprecipitation (Canaan et al., 2009), which are time-consuming, expensive, labor-intensive and not suitable for clinical applications. Compared with conventional methods, electrochemical biosensors exhibit several advantages, such as high sensitivity, small analyte volume, and simple pretreatment (Garai-Ibabe et al., 2011, 2012). Therefore, we developed a new method to make biosensors for detection of EBNA-1. Graphene sheets (GS) are the flat monolayers of carbon atoms tightly packed into a two-dimensional honeycomb lattice (Brownson and Banks, 2010). GS have been extensively used as immobilized substrates for the development of biosensors (Jiang et al., 2013; Wang et al., 2014) due to rapid electron transportation, high electrical conductivity and good biocompatibility (Myung
C. Song et al. / Biosensors and Bioelectronics 58 (2014) 68–74
et al., 2011; Sun et al., 2011). However, GS are generally insoluble in water and may leak out upon immobilization onto the electrode surface (Saha et al., 2010). Chitosan (CS) is a biological cationic macromolecule with abundant –NH2 and –OH functional groups (Banks et al., 2005; Lu et al., 2002). A water-soluble polymer, CS was used as a disperser to prepare homogeneous GS solution by the covalent functionalization method (Kang et al., 2009). However, the introduction of CS could not promote electron transfer well. To avoid this problem, GS–MWCNTs hybrid was used. The MWCNTs have extensively been used due to excellent electron conductivity and high surface area-to-volume ratio, which proved to promote electron transfer (Kauffman et al., 2010; Wang et al., 2006). Ma et al. have prepared a GS and MWCNTs hybrid (Yen et al., 2011). Herein, we tried to combine GS with MWCNTs via π–π stacking. Au nanoparticles (AuNPs) have been widely used for their chemical stability, strong adsorption ability, and excellent electrical conductivity (Parker et al., 2010); meanwhile, they also have large capacity to immobilize protein (antibody) via Au-NH2 (Lu et al., 2012a; Ma et al., 2013). Furthermore, several studies have been reported on functional MWCNTs loaded with HRP and Ab2 based on a covalent bond. Xie et al. have constructed an electrochemical immunosensor based on multi-sHRP-DNA-coated MWNCTs as signal labels with high sensitivity (Ma et al., 2013). Recently, DNA-based hybridization chain reaction (HCR) was extensively used for the signal amplification (Liu et al., 2013; Wang et al., 2013; Zhou et al., 2012; Zhuang et al., 2013) in the electrochemical biosensors. Tang et al. and Chen et al. have constructed an electrochemical biosensor based on
69
nanogold-based biobar codes for label-free immunosensing of proteins coupled with an in situ HCR, which has gained satisfactory results for CEA detection (Zhang et al., 2012). The concept of HCR was firstly described by Dirks (2004). HCR is an enzyme-free process; the DNA initiator strand (S0) propagated a cascade of hybridization events between alternating H1 and H2 hairpin probes to yield a nicked double-helix. Doxorubicin hydrochloride was used as an electroactive indicator, which bound to double stranded DNA to produce an electrochemical signal (Wang et al., 1998; Zhang et al., 2009). Doxorubicin hydrochloride can remarkably improve the sensitivity of the immunosensor. Xie et al. used doxorubicin hydrochloride to fabricate an immunosensor that showed a higher sensitivity than the one without it (Lu et al., 2012b). However, to the best of our knowledge, the immunosensor based on the HCR using doxorubicin hydrochloride intercalated the long DNA concatemers, as an electroactive indicator has not yet been reported. In this study, we attempted to combine the specificity of electrochemistry immunoassay with the signal amplification capability of the HCR to construct an electrochemical immunosensor (Scheme 1). The protocol mainly involves the following: (i) The construction of the sandwiched immunocomplex between the immobilized Ab1 on the AuNPs/GS–MWCNTs/GCE and Ab2 on the MWCNTs conjugated with DNA. The double strand DNA was formed from HCR between initiator strands (S0) on the S0–MWCNTs–Ab2 and a stable mixture of H1 and H2 hairpins. (ii) Doxorubicin hydrochloride was used as an electroactive indicator in the system. (iii) DPV was used to measure the electrochemical signal, which is correlated with EBNA-1
Scheme 1. (a) The preparation of DNA–MWCNTs–Ab2; (b) the fabricating procedure for EBNA-1 immunosensor .
70
C. Song et al. / Biosensors and Bioelectronics 58 (2014) 68–74
concentration. Therefore, the system can indirectly determine the concentration of EBNA-1. The signal amplification function of HCR improved the sensitivity of the immunosensor.
2. Experimental 2.1. Reagents EBV Nuclear Antigen 1, monoclonal EBV Nuclear antibody (Ab1) and polyclonal EBV Nuclear antibody (Ab2) were obtained from Abcam Ltd. (Hong Kong, China). Graphene sheets (GS) were purchased from Pioneer Nanotechnology Co. (Nanjing, China). Multi-walled carbon nanotubes with carboxylic acid groups (MWCNTs) were acquired from Chengdu Organic Chemicals Co. Ltd. (Chengdu, China). Bovine serum albumin (BSA, 96–99%), chloroauric acid (HAuCl4), N-hydroxysuccinimide (NHS), 1-ethyl-3-3(3-dimethylaminopropyl) carbodiimide (EDC), and mecatohexanol (MCH) were produced from Sigma-Aldrich (St. Louis, MO, USA). 2-(N-morpholino) ethanesulfonic acid buffer (MES buffer), doxorubicin hydrochloride and chitosan (CS) were obtained from Shanghai Sangon Biological Engineering Technology & Services Co. Ltd. (Shanghai, China). Deionized distilled water purified through a Millipore system (Z18 MΩ) was employed in all runs. All other reagents were of analytical grade and were used without further purification. All oligonucleotide sequences including S0 and two hairpins were ordered from Shanghai Sangon Biological Engineering Technology & Services Co. Ltd. (Shanghai, China). The sequences are listed in Table S1. 2.2. Apparatus All electrochemical measurements were carried out at room temperature (RT) (25 70.5 1C) using a CHI660D electrochemical workstation (Shanghai Chenhua Instruments, China) with a conventional three-electrode cell. A bare or modified glassy carbon electrode (GCE) was used as the working electrode, while Ag/AgCl (in saturated KCl) and a platinum (Pt) wire were served as the reference and the counter electrodes, respectively. The morphology of GS–MWCNTs nanocomposite was characterized by a scanning electron microscopy (SEM) system (FEI Nova-400, USA). An atomic force microscope (AFM) NanoWizard II (JPK Instruments Inc., Germany) was used to observe images of the MWCNTs and DNA–MWCNTs–Ab2.
400 mM EDC and 100 mM NHS solution for 2 h. The activated MWCNTs were separated from the free EDC/NHS by centrifugation. Next, S0 (21 mL, 1 OD) and Ab2 (50 mL, 25 mg mL 1) were added to the activated MWCNTs and stirred for 12 h at RT. By this step, S0 and Ab2 were chemically bonded to MWCNTs via the formation of an amide bond between the –NH2 groups of Ab2, S0, and the –COOH groups of MWCNTs. Following that, the mixture was centrifuged (8000 rpm) for 10 min at 4 1C. The bioconjugates were collected and washed with phosphate buffer solution (PBS, pH 7.0) three times to eliminate the free S0 and Ab2. Afterward, 1 mM MCH was added into S0–MWCNTs–Ab2 for 1 h to block the remaining sites. After centrifugation, the mixture was resuspended in Tris–HCl buffer (pH 7.0, 600 mL) and hybridization chain reaction was carried out by employing two hairpin species (H1 and H2). About 10 mM H1 and H2 was injected into S0–MWCNTs–Ab2 and incubated at 37 1C for 80 min. The resulting DNA–MWCNTs–Ab2 bioconjugates were washed thrice with Tris–HCl buffer (pH 7.0, 10 mM). Ultimately, DNA–MWCNTs–Ab2 bioconjugates were store at 4 1C for future use. 2.5. Fabrication of immunosensor The stepwise procedure for the electrochemical immunosensor is illustrated in Scheme 1b. The fabrication of the electrochemical immunosensor consists of the following steps: (i) Prior to use, the glassy carbon electrodes (GCE, 3 mm diameter) were polished to a mirror-like surface with 1.0, 0.3, and 0.05 mm alumina slurry sequentially, followed by sonication in deionized water/acetone/ deionized water for 5 min and dried at RT. (ii) About 8 mL of the asprepared GS–MWCNTs homogeneous suspension was initially deposited on the electrode surface, and then dried naturally at RT. (iii) The GS–MWCNTs/GCE electrode was immersed into 3 mM HAuCl4 solution and electrodeposition was performed by cyclic voltammetry at a potential from 0.2 to 0.6 V for 100 seconds to obtain the AuNPs/GS–MWCNTs modified electrode. (iv) Approximately 8 mL of Ab1 solution (50 mg mL 1) in PBS was spread on the AuNPs/GS–MWCNTs modified electrode, followed by overnight incubation at 4 1C. (v) The modified electrode was blocked by dipping in 0.25% BSA solution for 30 min at RT. After another round of washing with PBS (pH 7.0, 0.01 M), the immunosensor was obtained. 2.6. Electrochemical measurement
2.3. Synthesis of GS–MWCNTs nanocomposite GS–MWCNTs nanocomposite was synthesized according to the literature (Diaconu et al., 2010; Jiang et al., 2005; Wu et al., 2009) with slight modification. Firstly, CS solution (0.5 wt%) was prepared by dissolving 250 mg of CS powder in 50 mL of 1% (v/v) acetic acid solution with 2 h ultrasonication to form a homogeneous solution. The pH of the CS solution was adjusted to 4.0–4.5 using 1.0 M NaOH. Then, 1 mg of GS and MWCNTs was dispersed in 1 mL of CS solution, followed by 2 h ultrasonication to make black homogeneous GS–MWCNTs. 2.4. Preparation of DNA–MWCNTs–Ab2 DNA–MWCNTs–Ab2 were synthesized and prepared according to a previous publication (Ma et al., 2013) with light modification, as illustrated in Scheme 1a. Firstly, 1 mg of MWCNTs was dissolved thoroughly in 2-(N-morpholino) ethanesulfonic acid buffer (MES buffer, 1 mL, 50 mM, pH 7.0) and the pH of MES was adjusted to 7.0 with 1.0 M NaOH. The mixture was treated ultrasonically for 3 h until a uniform gloss suspension was obtained. Then, the –COOH groups of MWCNTs were activated by treatment with 1 mL of
In this work, EBNA-1 samples were assayed with a sandwichtype immunoassay format using DNA–MWCNTs–Ab2 as the secondary antibody and doxorubicin hydrochloride as the electroactive indicator. Various concentrations of EBNA-1 (Ag) were tested. The assay was carried out as follows: (i) The blocked immunosensors were incubated with different antigen concentrations for 30 min, followed by washing three times with PBS to remove nonspecifically bound conjugates. (ii) The as-prepared DNA–MWCNTs–Ab2 (8 mL) was deposited onto the immunosensor and incubated for another 30 min to form a sandwiched immunocomplex. (iii) After rinsing with PBS, the modified electrode was immersed into PBS (pH 7.0, 0.01 M) solution containing 5 mM doxorubicin hydrochloride for 40 min, followed by rinsing three times with PBS to remove uninteracted doxorubicin hydrochloride. The electrochemical measurement with differential pulse voltammetry (DPV) values was carried out from 0.3 to 0.9 V at 50 mV s 1 in pH 7.5 PBS. All measurements were conducted at RT (2572 1C). The standards were prepared using EBNA-1 solution at different concentrations. Blood specimens were gifted by the First People's Hospital, Jiulongpo District of Chongqing, and the First Affiliated Hospital, Chongqing Medical University.
C. Song et al. / Biosensors and Bioelectronics 58 (2014) 68–74
71
Fig. 1. SEM images of (a) GS–MWCNTs and (b) AuNPs/GS–MWCNTs. AFM images of (c) MWCNTs and (d) DNA–MWCNTs–Ab2.
3. Results and discussion
3.2. Cyclic voltammetric characterization
3.1. Characterization of the GS–MWCNTs, AuNPs/GS–MWCNTs, MWCNTs and DNA–MWCNTs–Ab2
The different modified electrodes were investigated by cyclic voltammetry in PBS (pH 7.5, 0.01 M) containing 5 mM K3Fe(CN)6/ K4Fe(CN)6 and 0.1 M KCl at a scan rate of 50 mV s 1. As expected, there was an obvious increase in current responses after GS–MWCNTs modified on the electrode (Fig. 2b) when compared with bare GCE (Fig. 2a). The peak current obtained could be ascribed to the redox behavior of [Fe(CN)6]3 /4 . Then, after the electrodeposition of AuNPs onto the GS–MWCNTs/GCE surface, the highest redox peaks were obtained, with well-defined redox peaks at 310 and 181 mV (peak potential difference of 129 mV) (Fig. 2c). AuNPs on the modified electrode could effectively enhance the electron transfer between [Fe(CN)6]3 /4 and the electrode. With the immobilization of the captured antibodies and BSA blocking solution on the modified electrode surface, redox peak currents gradually decreased (Fig. 2d and e). This finding suggested that the Ab1 was successfully immobilized on the electrode surface and the transmission of ferricyanide to the electrode surface was hindered. When the EBNA-1 was captured by its antibody, the peak current further decreased (Fig. 2f), which was consistent with the fact that the hydrophobic layer of the protein insulated the conductive support (Yang et al. 2013). The peak current increased with the DNA–MWCNTs–Ab2 immobilization on the electrode (Fig. 2g),
The morphologies of GS–MWCNTs and AuNPs/GS–MWCNTs were characterized using field-emission scanning electronic microscopy (SEM). Fig. 1a shows an SEM image obtained for the GS–MWCNTs modified surface, GS–MWCNTs dispersed by the CS solution. This material could obviously increase the electrode effective surface and enhance the electron transfer ability. As seen from Fig. 1b, a large number of gold nanoparticles were coated on the modified materials (relative to the GS–MWCNTs, Fig. 1a), indicating that AuNPs could be successfully electrodeposited on the modified electrode. To demonstrate the successful conjugation of DNA and Ab2 on the MWCNTs, the nanomaterial before and after modification was monitored by AFM. Two three-dimensional height maps display AFM images of MWCNTs and DNA–MWCNTs– Ab2 (Fig. 1c and d, respectively). As shown in Fig. 1c, a dense and spiky appearance was observed, which is consistent with a previous report (Ma et al., 2013). Then, due to the bioconjugation of DNA and Ab2, the result of AFM showed a blunt image in Fig. 1d, indicating that DNA and Ab2 were successfully immobilized onto MWCNTs (Ma et al., 2013).
72
C. Song et al. / Biosensors and Bioelectronics 58 (2014) 68–74
to 2116 Ω after EBNA-1 was successively absorbed onto the electrode surface (Fig. 3f). Subsequently, after DNA–MWCNTs– Ab2 was immobilized on the electrode surface, the Ret decreased to 1196 Ω in turn. So, the EIS results were in accordance with the CV curves shown in Fig. 2. 3.4. Comparison of the proposed protocol with a control experiment
Fig. 2. CVs of different electrodes in PBS (pH 7.5) containing 5 mM K3Fe(CN)6/K4Fe (CN)6 and 0.1 M KCl at a scan rate of 50 mV s 1: (a) bare GCE electrode; (b) GS– MWCNTs/GCE; (c) AuNPs/GS–MWCNTs/GCE; (d) Ab1/AuNPs/GS–MWCNTs/GCE; (e) BSA/Ab1/AuNPs/GS–MWCNTs/GCE; (f) EBNA-1/BSA/Ab1/AuNPs/GS–MWCNTs/ GCE; and (g) DNA–MWCNTs–Ab2 /EBNA-1/BSA/Ab1/AuNPs/GS–MWCNTs/GCE.
To demonstrate that HCR can truly improve the signal amplification, a HCR-free control experiment was performed (Fig. S1). As expected, when detecting the same concentration EBNA-1 (400 pg mL 1), the HCR-free immunosensor displayed lower peak current (Fig. S1a) compared with the HCR immunosensor (Fig. S1b). This result is attributed to the fact that HCR provides a lot of sites to the intercalated amount of doxorubicin hydrochloride, which greatly enhances detection signals. In order to prove that GS–MWCNTs and AuNPs could really improve the analytical performance of the immunoassay, the characteristics of the biosensor were investigated by CV (Fig. S2). The peak currents of AuNPs/MWCNTs/GCE (cure a) and AuNPs/GS/ GCE (cure b) increased gradually, but the signal of AuNPs/GS– MWCNTs/GCE (cure c) was obviously the highest under the same conditions, which should be attributed to the reasonable combination of GS–MWCNTs and AuNPs. 3.5. Optimization of experimental conditions
Fig. 3. EIS spectra for each immobilization step: (a) bare GCE electrode; (b) GS– MWCNTs/GCE; (c) AuNPs/GS–MWCNTs/GCE; (d) Ab1/AuNPs/GS–MWCNTs/GCE; (e) BSA/Ab1/AuNPs/GS–MWCNTs/GCE; (f) EBNA-1/BSA/Ab1/AuNPs/GS–MWCNTs/ GCE; and (g) DNA–MWCNTs–Ab2 /EBNA-1/BSA/Ab1/AuNPs/GS–MWCNTs/GCE.
which was attributed to the good electronic conductivity of the MWCNTs. 3.3. Electrochemical impedance spectroscopy characterization Electric impedance spectroscopy (EIS) is an effective method to monitor the impedance changes of surface-transfer electrodes. In the Nyquist diagram, the semicircle diameter equals the charge transfer resistance (Ret). Fig. 3 shows the characteristic EIS corresponding to the stepwise modification processes. Compared with the bare GCE electrode (Fig. 3a), the diameter of the high frequency semicircle decreased when GS–MWCNTs film (Fig. 3b) was modified on the GCE surface. This result can be attributed to the coupling of GS–MWCNTs producing a synergic effect, such as excellent electroconductivity. After AuNPs were electrodeposited on the modified electrode, the Ret was close to zero (Fig. 3c). This result indicates that AuNPs accelerate the electron transfer process of [Fe(CN)6]3 /4 . However, after Ab1 and BSA were immobilized on the electrode surface, the Ret obviously increased to 625 Ω and 1279 Ω (Fig. 3d and e), respectively, because these biomacromolecules obstructed the electron transfer. The Ret further increased
To achieve maximum assay performance, various experimental parameters such as the ratio of GS to MWCNTs, the HCR hybridization reaction time, Ab–Ag interaction time, and the pH of the PBS solution were optimized by DPV. These experiments were conducted using EBNA-1 at 400 pg mL 1. The effect of ratio of GS to MWCNTs was examined to improve the analytical performance with different GS:MWCNT ratios (3:1; 2:1; 1:1; 1:2; 1:3). The highest peak current was achieved when the ratio was 1:1. Thus, the optimal GS:MWCNTs ratio was chosen as 1:1 (Fig. S3a). The effect of HCR time was also investigated in the range of 10 to 90 min (Fig. S3b). The current response gradually increased with HCR time from 10 to 70 min, and reached plateaus at 70 min. This result indicated that the HCR was primitively completed at 70 min. Thus, the optimum HCR time was considered as 70 min. The effect of Ab and Ag interaction time was studied in the range of 10–35 min (Fig. S3c). The current response gradually increased from 10 to 30 min and reached plateaus at 30 min. Hence, the optimum interaction time of 30 min was used in this study. The effect of the pH of the PBS solution was examined over the pH range of 4.5–9.5. Fig. S3d shows that the current response gradually increased from pH 4.5 to pH 7.5 and then decreased. Thus, the pH was considered to be 7.5. 3.6. Amperometric response and calibration curve Analytical calibrations for the EBNA-1 were recorded by DPV under optimal conditions. Fig. 4 shows that the peak currents of doxorubicin hydrochloride increased with increased concentrations of EBNA-1. As can be seen in the inset of Fig. 4, there is a linear relationship between the peak currents and the logarithm of the concentrations of EBNA-1 in the range from 0.05 to 6.4 ng mL 1. The peak current values were obtained from the mean value with three independent experiments. The linear regression equation was Y (mA) ¼24.81X 19.11, with a linear correlation coefficient (R2) of 0.99249, and the detection limit (LOD) was estimated to be 0.7 pg mL 1 (S/N¼3). Therefore, the proposed electrochemical immunosensor had a good analytical performance for the detection of EBV nuclear antigen (EBNA-1).
C. Song et al. / Biosensors and Bioelectronics 58 (2014) 68–74
73
4.0%, and 5.2%, respectively. The results confirmed that the immunosensor exhibited acceptable precision and fabrication reproducibility. Regeneration is also a key parameter in immunosensor development. The surface of the prepared immunosensor was immersed in 8 M urea solution after the completion of each assay, followed by rinsing with double distilled water to dissociate the antigen–antibody complex. After five cycles of regeneration, the immunosensor retained 88.2% of its origin current, and the relative standard deviation (RSD) was 3.5%. 3.8. Recovery
Fig. 4. The DPV responses of the different concentrations of EBNA-1 detected by the proposed immunoassay (from (a) to (i): 0, 50, 100, 200, 400, 800, 1600, 3200, and 6400 pg mL 1); inset: the calibration curve of the peak current response versus the logarithm of EBNA-1 concentration with the proposed immunoassay.
4. Conclusion
Table 1 Recovery studies of EBNA-1 from human serum samples. EBNA-1 added (pg mL 1)
Sample no.
EBNA-1 in serum (pg mL 1)
1
0
50.00
2
0
100.00
3
0
200.00
Recovery tests were performed to evaluate the feasibility of the immunosensor for clinical applications. Serum samples spiked with EBNA-1 at concentrations of 50, 100 and 200 pg mL 1 were tested using the proposed immunosensor, and the corresponding results are given in Table 1. The recovery values ranged from 101.7% to 102.1%. This result shows that the developed immunosensor has potential clinical applicability for the detection of EBNA-1.
EBNA-1 detected (pg mL 1)
RSD% (n¼ 5)
R (%)
50.18, 49.04, 52.64, 50.94, 51.71 101.24, 98.21, 101.83, 102.63, 103.47 207.83, 204.79, 195.57, 205.62, 206.85
2.61
101.7
1.98
101.4
2.41
102.1
R: recovery percentage and RSD: relative standard deviation percentage.
The analytical performance of our EBV immunosensor has been compared to those reported previously in Table S2. It is clear that the proposed biosensor displayed a better performance than other methods in the analytical range and detection limit. 3.7. Stability, selectivity, reproducibility, repeatability and regeneration of the immunosensor In this study, the stability of the immunosensor was evaluated. When the biosensor was stored in the refrigerator at 4 1C for 9 days, the response only decreased by 8.1%, indicating good stability. In order to evaluate the binding specificity of the immunosensor to EBNA-1, some possible interfering agents such as BSA, HBV surface antigen (HBsAg), human IgG and CEA were tested under the same experimental conditions. As shown in Fig. S4, when the biosensor was incubated 400 pg mL 1 pure EBNA-1 and 1 ng mL 1 interfering agents, a significant current response with EBNA-1 was observed compared to BSA, HBsAg, human IgG and CEA, suggesting the high selectivity of the immunosensor. The repeatability and reproducibility of the immunosensor were also investigated by the variation coefficients (CVs), by making 5 successive measurements of three concentrations of EBNA-1 using an identical immunosensor. The CVs of the intraassay were determined as 2.5%, 4.0%, and 2.8% at 50, 100, and 200 pg mL 1 of EBNA-1, respectively. Similarly, a set of 5 immunosensors was prepared to detect 50, 100, and 200 pg mL 1 of EBNA-1, and the inter-assay CVs of the measurement were 3.9%,
In summary, an ultrasensitive electrochemical sandwich immunosensor was developed using AuNPs/GS–MWCNTs as a sensor platform and coupling MWCNTs with the hybridization chain reaction (DNA–MWCNTs–Ab2) as a signal amplifying probe. GS– MWCNTs and AuNPs were, respectively, assembled onto the GCE electrode surface to greatly enhance the conductivity and improve effective surface area, which increased the amount of antibodies to be immobilized on the sensor. More significantly, a novel amplification strategy using DNA–MWCNTs–Ab2 was introduced, which exhibited a broad dynamic range and low detection limit. Basis on this study, more electrochemical immunosensors could be designed for the detection of other target analytes.
Acknowledgments This work was financially supported by the National Natural Science Foundation of China (81171415) and the Foundation of National Key Discipline in Laboratory Medicine, Chongqing Medical University, China (2010103).
Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios.2014.02.031. References Ambinder, R.F., Shah, W., Rawlins, D., Hayward, G.S., Hayward, S.D., 1990. J. Virol. 64 (5), 2369–2379. Banks, C.E., Davies, T.J., Wildgoose, G.G., Compton, R.G., 2005. Chem. Commun. 7, 829–841. Brownson, D.A., Banks, C.E., 2010. Analyst 135 (11), 2768–2778. Canaan, A., Haviv, I., Urban, A.E., Schulz, V.P., Hartman, S., Zhang, Z., Palejev, D., Deisseroth, A.B., Lacy, J., Snyder, M., 2009. Proc. Natl. Acad. Sci. 106 (52), 22421–22426. Diaconu, M., Litescu, S.C., Radu, G.L., 2010. Sens. Actuators B: Chem. 145 (2), 800–806. Dirks, R.M., 2004. Proc. Natl. Acad. Sci. 101 (43), 15275–15278. Garai-Ibabe, G., Grinyte, R., Canaan, A., Pavlov, V., 2012. Anal. Chem. 84 (14), 5834–5837. Garai-Ibabe, G., Grinyte, R., Golub, E.I., Canaan, A., de la Chapelle, M.L., Marks, R.S., Pavlov, V., 2011. Biosens. Bioelectron. 30 (1), 272–275. Humme, S., 2003. Proc. Natl. Acad. Sci. 100 (19), 10989–10994.
74
C. Song et al. / Biosensors and Bioelectronics 58 (2014) 68–74
Jiang, L., Wang, R., Li, X., Jiang, L., Lu, G., 2005. Electrochem. Commun. 7 (6), 597–601. Jiang, S., Hua, E., Liang, M., Liu, B., Xie, G., 2013. Colloids Surf. B: Biointerfaces 101, 481–486. Kang, X., Wang, J., Wu, H., Aksay, I.A., Liu, J., Lin, Y., 2009. Biosens. Bioelectron. 25 (4), 901–905. Kauffman, D.R., Sorescu, D.C., Schofield, D.P., Allen, B.L., Jordan, K.D., Star, A., 2010. Nano Lett. 10 (3), 958–963. Liu, S., Wang, Y., Ming, J., Lin, Y., Cheng, C., Li, F., 2013. Biosens. Bioelectron. 49, 472–477. Lu, G., Yao, X., Zhou, X., Zhan, T., Long, D., 2002. Chem. Res. Chin. Univ. 18 (3), 316–320. Lu, J., Liu, S., Ge, S., Yan, M., Yu, J., Hu, X., 2012a. Biosens. Bioelectron. 33 (1), 29–35. Lu, L., Liu, B., Zhao, Z., Ma, C., Luo, P., Liu, C., Xie, G., 2012b. Biosens. Bioelectron. 33 (1), 216–221. Ma, C., Liang, M., Wang, L., Xiang, H., Jiang, Y., Li, Y., Xie, G., 2013. Biosens. Bioelectron. 47 (0), 467–474. Marrazza, G., Chianella, I., Mascini, M., 1999. Biosens. Bioelectron. 14 (1), 43–51. Middeldorp, J.M., Brink, A.A., van den Brule, A.J., Meijer, C.J., 2003. Crit. Rev. Oncol./ Hematol. 45 (1), 1–36. Myung, S., Solanki, A., Kim, C., Park, J., Kim, K.S., Lee, K.B., 2011. Adv. Mater. 23 (19), 2221–2225. Parker, J.F., Fields-Zinna, C.A., Murray, R.W., 2010. Acc. Chem. Res. 43 (9), 1289–1296. Saha, A., Basiruddin, S., Ray, S., Roy, S., Jana, N.R., 2010. Nanoscale 2 (12), 2777–2782. Sixbey, J., Lemon, S., Pagano, J., 1986. Lancet 328 (8516), 1122–1124.
Sun, J.-Y., Huang, K.-J., Wei, S.-Y., Wu, Z.-W., Ren, F.-P., 2011. Colloids Surf. B: Biointerfaces 84 (2), 421–426. Thompson, M.P., Kurzrock, R., 2004. Clin. Cancer Res. 10 (3), 803–821. Wang, C., Zhou, H., Zhu, W., Li, H., Jiang, J., Shen, G., Yu, R., 2013. Biosens. Bioelectron. 47, 324–328. Wang, H.-S., Li, T.-H., Jia, W.-L., Xu, H.-Y., 2006. Biosens. Bioelectron. 22 (5), 664–669. Wang, J., Ozsoz, M., Cai, X., Rivas, G., Shiraishi, H., Grant, D.H., Chicharro, M., Fernandes, J., Paleček, E., 1998. Bioelectrochem. Bioenerg. 45 (1), 33–40. Wang, L., Hua, E., Liang, M., Ma, C., Liu, Z., Sheng, S., Liu, M., Xie, G., Feng, W., 2014. Biosens. Bioelectron. 51, 201–207. Wu, H., Wang, J., Kang, X., Wang, C., Wang, D., Liu, J., Aksay, I.A., Lin, Y., 2009. Talanta 80 (1), 403–406. Yang, F., Chai, Y., Yuan, R., Han, J., Yuan, Y., Liao, N., Yang, Z., 2013. Anal. Methods 5, 5279–5285. Yao, Q.Y., Rickinson, A.B., Epstein, M.A., 1985. Int. J. Cancer 35 (1), 35–42. Yen, M.-Y., Hsiao, M.-C., Liao, S.-H., Liu, P.-I., Tsai, H.-M., Ma, C.-C.M., Pu, N.-W., Ger, M.-D., 2011. Carbon 49 (11), 3597–3606. Young, L.S., Murray, P.G., 2003. Oncogene 22 (33), 5108–5121. Zhang, B., Liu, B., Tang, D., Niessner, R., Chen, G., Knopp, D., 2012. Anal. Chem. 84 (12), 5392–5399. Zhang, Y., Zhang, K., Ma, H., 2009. Anal. Biochem. 387 (1), 13–19. Zhou, J., Xu, M., Tang, D., Gao, Z., Tang, J., Chen, G., 2012. Chem. Commun. 48 (100), 12207. Zhuang, J., Fu, L., Xu, M., Zhou, Q., Chen, G., Tang, D., 2013. Biosens. Bioelectron. 45, 52–57.
