Biosensors and Bioelectronics 54 (2014) 351–357
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Sensitivity enhancement of an electrochemical immunosensor through the electrocatalysis of magnetic bead-supported non-enzymatic labels Rashida Akter a, Choong Kyun Rhee a,b,n, Md. Aminur Rahman b,nn a b
Department of Chemistry, Chungnam National University, Daejeon 305-764, South Korea Graduate School of Analytical Science and Technology (GRAST), Chungnam National University, Daejeon 305-764, South Korea
art ic l e i nf o
a b s t r a c t
Article history: Received 16 August 2013 Received in revised form 25 October 2013 Accepted 28 October 2013 Available online 4 November 2013
An ultrasensitive non-enzymatic electrochemical carcinoembryonic antigen (CEA) immunosensor was fabricated by the immobilization of a monoclonal CEA antibody (anti-CEA) on a protein A (PA) attachedgold nanoparticles (AuNPs)-deposited electrochemically prepared polydopamine film (e-PD/AuNPs). Magnetic beads (MB)-supported and CEA-conjugated multiple 3,3′,5,5′-tetramethylbenzidine (TMB) was used as electrochemical labels. The detection was based on the measurements of the electrocatalyzed oxidation of ascorbic acid (AA) by the multiple TMB labels after competitive binding between MB/TMBconjugated-CEA and free-CEA. The electrocatalyzed oxidation current of AA by TMB decreased with increasing concentration of the free-CEA as the amount of CEA/MB/TMB labels decreased at the immunosensor probe. The immunosensor surface was characterized using electrochemical impedance spectroscopy, Fourier transform infrared spectroscopy, quartz crystal microbalance, and scanning electron microscopy techniques. Cyclic voltammetry (CV) and differential pulse voltammetry (DPV) techniques were used to monitor the electrocatalyzed response. The proposed immunosensor exhibited a wide linear dynamic range (1.0 pg/mL to 10.0 ng/mL), low detection limit (1.07 0.04 pg/mL), good selectivity, and long-time stability. It was successfully applied to various CEA spiked human serum samples for the detection of CEA. & 2013 Elsevier B.V. All rights reserved.
Keywords: Non-enzymatic electrocatalysis 3,3′,5,5′-Tetramethylbenzidine Electrochemical immunosensor Nanoparticles Polydopamine film Protein biomarker detection
1. Introduction The measurement of trace amounts of protein biomarkers plays essential roles in biochemical and biomedical researches as well as for the development of practical medical diagnostics for early disease detection, prognostic treatment, and patient curative effect monitoring. Therefore, measuring their presences in body fluids remains a great challenge in disease diagnostics. Conventional immunoassay methods such as enzyme-linked immunosorbent assay (ELISA) (Yates et al., 1999), fluorescence immunoassay (Matsuya et al., 2003), chemiluminescence assay (Fu et al., 2006), radioimmunoassay (Teppo and Maury, 1987), mass spectrometric immunoassay (Hu et al., 2007), metalloimmunoassay (Noh et al., 2011), and electrophoretic immunoassay (Schmalzing and Nashabeh, 1997) are commonly used for trace amounts protein detection. However, each of these immunoassay methods has drawbacks: complicated, time-consuming, tedious, expensive, labor-intensive, and not suitable for point-of-care applications. As n
Corresponding author. Tel.: þ 82 42 821 5483; fax: þ 82 42 821 8896. Corresponding author. Tel.: þ 82 42 821 8546; fax: þ 82 42 821 8541. E-mail addresses: [email protected] (C. Kyun Rhee), [email protected] (Md. Aminur Rahman). nn
0956-5663/$ - see front matter & 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bios.2013.10.058
alternative approaches, various immunosensor methods based on surface plasma resonance (SPR) (Chang et al., 2010), quartz crystal microbalance (QCM) (Knudsen et al., 2006), chemiluminescence (Jie at al., 2010), and electrochemistry (Blonder et al., 1996; Wang et al., 2006) have been developed for the protein detection. Among them, electrochemical immunosensors have shown a great promise because of simple instrumentation, high sensitivity, fast response time, miniaturization, low cost, and point-of-care applications. Coupling with nanomaterials and various amplification techniques, several electrochemical immunosensors have been successfully developed for detecting various protein biomarkers (Akter et al., 2012; Jeong et al., 2013; Tang et al., 2008; Du et al., 2011). In most cases, nanomaterial-supported enzymes are used as electrochemical labels for increasing the detection sensitivity. However, enzymes are not advantageous because of costly preparation and purification process (Chen et al., 2009), which limits their uses as labels in electrochemical immunosensors. Thus, the search for substitutes of enzymes in electrochemical protein immunosensors would be very attractive. Various non-enzymatic labels including quantum dots (QD) (Liu et al., 2004; Wang et al., 2008), metallic nanoparticles (Ting et al., 2009; Ho et al., 2010; Das et al., 2006), hydrazine (Shiddiky et al., 2007), and metal ion-cysteamine complex (Noh et al., 2011), have been used in
352
R. Akter et al. / Biosensors and Bioelectronics 54 (2014) 351–357
electrochemical protein immunosensors. Although the above nonenzymatic labels are able to enhance the detection sensitivity, however, they suffer from some inherent disadvantages. For example, QD-based detection requires harsh condition for nanocrystal dissolution for stripping analysis, hydrazine-based detection requires deoxygenation of the detection medium, and metal– ion–cysteamine complex-based detection requires multistep processes for synthesizing the complex prior to electrochemical detection. Metallic nanoparticles label such as gold (Au) and silver (Ag) can be directly detected by stripping voltammetry. However, the oxidation peaks of Au and Ag occur at relatively positive potentials where other biomolecules can be oxidized. Thus, it is a major challenge to incorporate new non-enzymatic labels having enhanced electrocatalytic activities in electrochemical immunosensors. 3,3′,5,5′-Tetramethylbenzidine (TMB) is a widely used chromogen in enzyme-linked immunosorbent assay (ELISA). It is a nonmutagenic and non-carcinogenic compound that has been widely used for the detection of blood and urine sugar, hypothermia, horseradish peroxidase enzyme, and some antibodies in clinical samples (Serrat, 1994). TMB is an electroactive compound, showing a reversible redox reaction under appropriate conditions (Liu et al., 2008), so that it has been used for the fabrication of a reagentless amperometric immunosensor (Wu et al., 2009). The amine groups of TMB are easy to conjugate with antibody or other proteins by covalent cross-linking. Thus, TMB is a promising electroactive label for the fabrication of highly sensitive electrochemical immunosensor. In order to increase the amount of electroactive labels, various nanomaterials are used as nanocarriers. Gold nanoparticles (AuNPs) (Lai et al., 2009), magnetic bead (MB) (Mani et al., 2009), multi wall carbon nanotubes (MWCNT) (Akter et al., 2012; Yu et al., 2006), silica nanoparticles (Lin et al., 2011), carbon spheres (Du et al., 2010), and dendrimer/AuNPs (Den/AuNPs) (Jeong et al., 2013) have been used as nanocarriers in electrochemical immunosensor. For example, Rusling group has enhanced the detection limit of prostate specific antigen by using MB-supported enzymatic bioconjugates, which shows 8-fold better performances than carbon nanotube (CNT)-supported enzymatic bioconjugates (Mani et al., 2009). The MB-supported non-enzymatic labels (MB-TMB) would be a prospective amplification strategy for the high sensitive cancer biomarker detection because enhanced amount of non-enzymatic TMB labels can be loaded onto the MB. In the present study, the electrocatalytic activity of the magnetic bead-supported TMB labels has been demonstrated for the first time for developing an electrochemical carcinoembryonic antigen (CEA) immunosensor. CEA has been chosen as a model protein as it is associated with colorectal cancer and also identified as a biomarker for breast tumors and ovarian carcinoma (Letilovic et al., 2006). For the fabrication of the immunosensor probe, monoclonal anti-CEA antibody has been immobilized onto the protein A (PA)-attached electrochemically prepared gold nanoparticles/polydopamine (e-PD/e-AuNPs/PA) film. The purpose of using the e-PD film is to make a biocompatible and stable immunosensor platform that allows nanoparticles deposition and biomolecules immobilization. The major advantage of this material is that it has high biocompatibility (Shi et al., 2013) and remarkable adhesive ability (Zhang et al., 2012) and could deposit very stably on all kinds of organic and inorganic surfaces, including nonsticking surfaces (Lee et al., 2007). However, the conductivity of the e-PD film is low. To increase the surface conductivity of the e-PD film, electrochemically prepared AuNPs were deposited on the surface of e-PD film. The CEA detection has been performed by monitoring the electrocatalytic activity of the MB-supported TMB labels towards AA oxidation in phosphate buffer saline solution (PBS) after the competitive binding reaction between free-CEA and MB-supported TMB-conjugated CEA (CEA/MB/TMB).
2. Experimental 2.1. Preparation of CEA/MB/TMB conjugates For the preparation of CEA/MB/TMB conjugates, the COOH groups of MB (2 mg/mL) were activated by treating with 10 mM NHS/EDC in 50 mM MES buffer at pH 5.8 for 6 h. The activated COOH-MB was separated from the free NHS/EDC by magnetic separation. 5.0 ng/mL CEA and 1.0 mg/mL TMB were then added to the activated COOH-MB and stirred for 24 h at 4 1C. By this step, CEA and TMB were attached to the MB through the covalent bond formation between the carboxylic acid groups of MB and amine groups of CEA and TMB. The CEA/MB/TMB conjugate was separated from the free CEA and TMB by magnetic separation and was washed several times with a PBS solution. Finally, the CEA/MB/ TMB conjugate was diluted with 1.0 mL of 0.1 M PBS (pH 7.4) and kept at 4 1C. 2.2. Fabrication of the immunosensor probe and the sensing principle Fig. 1 shows the schematic illustration of the fabrication of immunosensor and the sensing principle of the proposed CEA detection. At first, an e-PD film was grown on gold (Au) electrodes through the electropolymerization of dopamine using a potential cycling method. Prior to electropolymerization, Au electrodes were polished with a 0.05 μm alumina/water slurry on a polishing cloth to a mirror finish, followed by sonicating and rinsing with distilled water. Then, the polished electrodes were rinsed with fresh piranha solution (70% H2SO4, 30% H2O2; CAUTION: piranha solution reacts violently with most organic materials and must be handled with extreme care) followed by cycling the potential between þ1.4 and 0.2 V at 100 mV/s for 50 times in a 1.0 M H2SO4 solution. The electropolymerization of dopamine (1.0 mg/mL) was carried out in a 0.1 M PBS solution of pH 6.0 by cycling the potential between 0 and 0.7 V for nine times. The pH of the PBS solution was chosen 6.0 as the chemical polymerization of dopamine occurred under basic solution (Lee et al., 2007). After electropolymerization, the e-PD film coated Au electrode was washed with a PBS solution (pH ¼6.0) to remove any remaining dopamine from the electrode surface. AuNPs were then electrochemically deposited on the e-PD film by sweeping the potential between 0 and þ1.4 V in a 0.1 M H2SO4 solution containing 1 mM HAuCl4 at the scan rate of 100 mV/s (Singh et al., 2008). PA was then immobilized onto the e-AuNPs deposited e-PD film (Au/e-PD/ e-AuNPs) by incubating the Au/e-PD/e-AuNPs electrode in a PBS solution containing 3.0 mg/mL of PA for 8 h at 4 1C. PA was attached onto the e-PD/e-AuNPs through the formation of Au-NH bond. Immobilization of a monoclonal anti-CEA antibody was carried out by incubating the Au/e-PD/e-AuNPs/PA electrodes in a 0.1 M phosphate PBS solution (pH 7.0) containing 0.1 mg/mL anti-CEA for 24 h at 4 1C. After washing three times with 0.1 M PBS, the Au/e-PD/e-AuNPs/PA/anti-CEA electrodes (immunosensors) were blocked by dipping them in 0.1% BSA solution for 2 h at 4 1C. After washing three or four times with PBS and drying with N2, the blocked immunosensors were incubated in the mixtures of a fixed amount of CEA/MB/TMB and various concentrations of free-CEA at 4 1C for the competitive binding of free- and labeledCEA to the active sites of anti-CEA. After the competitive binding, the immunosensors were washed three or four times with PBS to remove nonspecifically bound free- and labeled- CEA. Then, the immunosensors were dipped in a 0.1 M PBS (pH 7.0) solution for monitoring the CV and DPV responses of the oxidation process of MB/TMB labels. The electrocatalytic activity of the MB/TMB label towards the oxidation of ascorbic acid (AA) was studied for the signal amplification.
R. Akter et al. / Biosensors and Bioelectronics 54 (2014) 351–357
353
Fig. 1. (a) Schematic representation of the immunosensor and (b) the mechanism of CEA detection through the TMB electrocatalyzed AA oxidation.
3. Results and discussion 3.1. Characterization of the e-PD and e-PD/e-AuNPs modified electrode The e-PD film grown onto an Au electrode was characterized by using the CV technique. Fig. 2a shows a series of CVs recorded during the electropolymerization of PD in a 0.1 M PBS solution of pH 6.0 at the scan rate of 0.1 V/s. The CV exhibited an oxidation peak at þ 0.35 V during the first anodic scan, which was due to the oxidation of PD to form e-PD. The e-PD film progressively covered the electrode surface, which increased the thickness of the film and decreased the surface area of the electrode. Thus, the peak currents were found to be decreased as the cycle numbers increased. These results clearly demonstrate that the e-PD film immediately formed after the oxidation of the PD at þ 0.35 V (Kang et al., 2012). A small reduction peak of the e-PD was observed at þ0.048 V during the cathodic scan. In order to determine the amount of e-PD formed, electrochemical quartz crystal microbalance (EQCM) experiment was carried out. Fig. 2b shows the frequency changes obtained from an EQCM experiment. During the electropolymerization of PD, the frequency decreased due to the formation of e-PD film. Deposition of e-PD, took place rapidly as the cycle number increased. The Δf value was 161.5 Hz after nine cycles. The amount of e-PD grown was determined to be 404 722 ng by using the sensitivity factor of 2.5 (Akter et al., 2012), corresponding to a surface coverage of 2.1 10 9 mol/cm2. The redox activity of the e-PD film was recorded for an e-PD modified Au electrode in a PBS solution (Fig. 2c). A pair of oxidation and reduction peaks was observed at þ0.15/ þ 0.25 V, indicating that the e-PD film is electrochemically active. The peak currents of these oxidation and reduction peaks were proportional to the scan rate due to the redox behavior of the adsorbed e-PD film. The electrochemically deposited AuNPs on e-PD (e-PD/e-AuNPs) was characterized using the CV technique (Fig. S1). Au oxidation and reduction peaks were observed at þ0.97 and þ0.45 V, respectively,
which was not observed in the case of e-PD electrode, indicating clearly that AuNPs successfully deposited on the e-PD film. Fig. 2d shows the SEM image obtained for the e-PD/e-AuNPs surface, showing the AuNPs were uniformly distributed on the e-PD film with the particle sizes ranges between 5 and 20 nm. 3.2. Characterizations of CEA/MB/TMB conjugates and the immunosensor probe Fig. 3 compares the SEM images of MB before (a) and after (b) conjugating with CEA and TMB. The SEM image obtained after conjugation showed that there are overlayers on MB, whose radius is 0.1 μm. The existence of such overlayers on MB confirmed that CEA and TMB conjugated to the MB. The Fourier transform infrared (FT-IR) spectra also confirmed that CEA and TMB conjugated to the MB (Fig. S2). The immunosensor probe was characterized using electrochemical impedance spectroscopy (EIS) and the quartz crystal microbalance (QCM) techniques. Fig. 3c shows the Nyquist plot observed for a e-pD (i), e-pD/e-AuNPs (ii), e-pD/e-AuNPs/PA (iii), e-pD/e-AuNPs/PA/anti-CEA (iv) modified electrodes recorded in a 5.0 mM of Fe(CN)63 /4 solution. The inset of Fig. 3c shows a general equivalent circuit that can be used to model the impedance data. The general equivalent circuit contains the solution resistance (Rs), the charge transfer resistance (Rct), the Warburg element (W), and the charge of the constant phase element (CPE). The Rct can be estimated from the diameter of the semicircle part at higher frequencies in the Nyquist plot. For the e-PD modified electrode, the Rct value was estimated as 60 kΩ, which significantly decreased to about 15 kΩ after e-AuNPs deposition. This result indicates that e-AuNPs increased the conductivity of the e-PD film. However, the Rct value increased to about 20 kΩ after PA immobilization due to the increase of the thickness of the electrode surface. PA was chemically bonded onto the e-PD/e-AuNPs through the interaction of Au-NH bonds. After the immobilization of anti-CEA on the e-PD/e-AuNPs/PA, the Rct value further increased to 25 kΩ. The above results clearly show that the PA and anti-CEA successfully immobilized on the modified e-PD/e-AuNPs electrode. The immunosensor surface was also
354
R. Akter et al. / Biosensors and Bioelectronics 54 (2014) 351–357
Fig. 2. (a) Consecutive CVs (9 cycles) recorded for the electropolymerization of 1.0 mg/mL dopamine in a 0.1 M PBS solution of pH 6.0. The san rate was 0.05 V/s. (b) EQCM responses concurrently recorded with the CV for the electropolymerization of dopamine. (c) CVs recorded for the e-PD modified electrode in a 0.1 M PBS solution at various scan rates: (i–vi) 0.05-0.5 V/s. Inset shows the current vs. scan rate plot. (d) SEM image of the e-PD/e-AuNPs modified electrode. Inset shows the image of an e-PD modified electrode.
characterized using the QCM technique (Fig. S3) and is discussed in the Supplementary materials.
3.3. Electrochemical characteristics of CEA immunosensor Fig. 4a shows the CV recorded for an Au/e-PD/e-AuNPs/PA/antiCEA modified electrode (immunosensor probe) after binding with CEA/MB/TMB conjugates (200 μL) in a PBS solution (pH 7.4) in the absence (i) or presence (ii) of 1.0 mM AA. In the absence of AA, well-defined oxidation and reduction peaks were observed at 0.39 and 0.25 V, respectively, which was not observed when the CVs were recorded after binding with CEA and CEA/MB. The Au/e-PD/ e-AuNPs/PA/anti-CEA immunosensor probe itself also did not show any peak. These results indicate that the redox peak came solely from the conjugated TMB, which was bonded to the immunosensor probe as CEA/MB/TMB. For a comparison, the redox behavior of the CEA/TMB conjugate was studied and compared with that of CEA/MB/TMB (Fig. S3). The peak separation of only CEA/TMB was about 0.06 V, whereas it was about 0.16 V for CEA/MB/TMB. These results indicate that the electron transfer (ET) process of the CEA/MB/TMB shows more irreversibility than that of CEA/TMB. This more irreversible ET process of CEA/MB/TMB was due to the large electron transfer distance between the electrode and the MB-supported TMB molecules. Another interesting result is that the electrochemical behavior of CEA/MB/TMB only shows a single redox peak, which is in contrast to the two one-electron reversible waves previously observed for TMB in solution at neural or acidic pH (Volpe et al., 1998). The redox peak currents were directly proportional to the scan rate up to 0.5 V/s, confirming that the redox reaction was involved in a surface-confined process (Murray, 1984) of TMB þ þ þ2e⇌TMB.
When 1.0 mM AA was added to the PBS solution, the catalytic characteristics observed with a dramatic increase of the oxidation peak and sharp decrease of the reduction current (Fig. 4a (ii)), which was due to the electrocatalytic oxidation of AA by the conjugated TMB. The electrocatalytic mechanism of AA oxidation can be expressed by the following equation: TMB⇌TMB þ þ 2e ðat electrodeÞ
ð1Þ
TMB þ þ AA-TMB þ DHAA
ð2Þ
The electrocatalytic AA oxidation current was much higher than the TMB oxidation current. To confirm the electrocatalytic process of AA results from the TMB labels and not from the electrodeposited PD film, the AA electrooxidation was measured with the immunosensor probe without PD film. The electrooxidation of AA was observed even without electrodeposition of PD film (Au/PA/anti-CEA/CEA/MB/TMB) (Fig. 4a (iii)). The electrocatalytic peak slightly shifted to the positive potential and the peak current was found to be decreased. The decrease in peak current might be related to the immobilization of lesser amounts of PA and anti-CEA due to the absence of PD/AuNPs, resulted in a lesser amount of TMB labels. The electrocatalytic peak slightly shifted to the positive potential might be due to low conductivity of the immunosensor surface in the absence of AuNPs. The existence of the AA electrocatalytic peak in the absence of PD clearly showed the electrocatalytic response results from the TMB label on the MB not from the PD film. Fig. 4b shows the DPVs recorded for an immunosensor after binding with CEA/MB/TMB conjugates in a PBS solution in the absence (i) or presence (ii) of 1.0 mM AA. In the absence of AA, the DPV response showed an anodic peak at about þ0.32 V for the oxidation of TMB. In the presence of AA, the dramatic increase of oxidation current due to the electrocatalytic
R. Akter et al. / Biosensors and Bioelectronics 54 (2014) 351–357
355
Fig. 3. SEM images of (a) COOH-MB and (b) CEA/MB/TMB conjugate. Insets of both figures show the images of a single particle. (c) EIS analyses of (i) Au/e-PD, (ii) Au/e-PD/eAuNPs, (iii) Au/e-PD/e-AuNPs/PA, (iv) Au/e-PD/e-AuNPs/PA/anti-CEA, and (v) bare Au electrodes in 5.0 mM Fe(CN)63 /4 solution. Inset shows the general equivalent circuit.
Fig. 4. (a) CV and (b) DPV responses recorded for Au/e-PD/e-AuNPs/PA/anti-CEA/CEA/MB/TMB in 0.1 M PBS solutions containing no AA (i) and 1.0 mM AA (ii), and for Au/PA/ anti-CEA/CEA/MB/TMB with 1.0 mM AA (iii). (c) DPV responses on the electrocatalytic AA oxidation recorded for an Au/e-PD/e-AuNPs/PA/anti-CEA-based immunosensor in a 0.1 M PBS solution before (i) and after (ii) competitive binding between free-CEA and CEA/MB/TMB conjugates. The concentration of free-CEA was 2.0 ng/mL and the amount of CEA/MB/TMB was 200 μL.
356
R. Akter et al. / Biosensors and Bioelectronics 54 (2014) 351–357
3.4. Detection of CEA Under the optimized experimental condition, the DPV current responses were measured before and after competitive binding of conjugated-CEA and free-CEA. Fig. 5a shows the DPV responses, where the electrocatalytic oxidation current of AA linearly decreased with the increasing concentration of free-CEA. Fig. 5b shows the calibration plot constructed by plotting the differences in current responses (ΔI) between before and after competitive binding (ICEA/MB/TMB ICEA/MB/TMB þ free-CEA) of CEA/MB/TMB and free-CEA. The calibration plot at the lower concentration range is shown in the inset of Fig. 5b. The immunosensor exhibits a wide dynamic range between 1.0 pg/mL and 10.0 ng/mL. The reproducibility expressed in terms of the relative standard deviation (RSD) was about 4.9% (n ¼5) at a CEA concentration of 2.0 ng/mL. The experimental detection limit of CEA was determined to be 1.0 70.04 pg/mL, which was 10 times lower than that obtained without magnetic nanoparticles assisted electrocatalytic activities of multiple TMB labels (Akter et al., 2012). The detection limit of the proposed immunosensor was also lower than HRP-anti-CEA/nanogold/chitosan (0.22 ng/mL) (Wu et al., 2006), anti-CEA/glutathione// nanogold/poly-o-aminophenol /Au (0.1 ng/mL) (Tang et al., 2007), thionine-doped magnetic gold nanpshheres (10 pg/mL) (Tang et al., 2008), anti-CEA/GA/GPMS/Fe3O4/SiO2/CPE (0.5 ng/mL) (Pan and Yang, 2007), and dendrimer-encapsulated gold nanoparticles and carbon nanotube-assisted multiple bienzymatic labels-based electrochemical immunosensors (4.4 pg/mL) (Jeong et al., 2013). This means that the electrocatalytic activity of MB-supported TMB labels improved the sensitivity of the CEA detection. 3.5. Stability and the real sample analysis
Fig. 5. (a) DPV responses on the electrocatalytic AA oxidation recorded at various free-CEA concentrations: (i) 0, (ii) 1.0, (iii) 10, (iv) 100 pg/mL, (v) 1.0, (vi) 2.0, (vii) 5.0, (viii) 10, and (ix) 20 ng/mL. (b) Calibration plot for the CEA detection. Insets of both figures show the magnified portion of the peaks and calibration plot at the low concentration range (1.0 100 pg/mL).
oxidation of AA by TMB was observed. The both CV and DPV studies clearly revealed that the TMB oxidation current could be enhanced by using electrocatalytic oxidation of AA by TMB. The electrocatalytic oxidation current of AA was considered as a base immunosensor response and monitored this response during the competitive binding between the fixed amount of TMB-labeled CEA and various concentrations of free-CEA. Fig. 4c shows the DPVs recorded for the immunosensor probe after incubating it in only (i) CEA/MB/TMB (before competitive binding) and a (ii) mixture of CEA/MB/TMB þ2 ng/mL of free-CEA (after competitive binding) solutions. The electrocatalytic oxidation current response after competitive binding was found to be decreased. In the absence of free-CEA, there was no competition, thus all the active sites of anti-CEA were bonded by the CEA/MB/TMB. However, when 2 ng/mL of free-CEA was present with CEA/MB/TMB, the competition between the free-CEA and conjugate-CEA for the antiCEA binding sites occurred. As a result, the amount of CEA/MB/ TMB at the immunosensor probe reduced and a decrease in the current response was observed. The difference in electrocatalytic current response (ΔI) before and after competitive binding was directly proportional to the concentration of free-CEA. To maximize the immunosensor response, various experimental parameters such as amount of PA, PA/anti-CEA interaction, anti-CEA dilution factor, competitive interaction time and temperature between CEA and CEA/MB/TMB with anti-CEA, and the pH of the binding medium were optimized (Fig. S5) and are discussed in the Supplementary materials.
The stability of the proposed CEA immunosensor was checked by measuring the competitive response of 1.0 ng/mL CEA for two months. After each measurement, the immunosensor surface was regenerated by dipping it into a 0.2 M glycine-hydrochloric acid (Gly-HCl) solution (pH ¼2.8) for 5 min followed by washing with a PBS solution. When stored in dry condition at 4 1C, the initial ΔI response did not change significantly for a period of two months (Fig. 6a), indicating that the long time stability of the immunosensor was good. The proposed non-enzymatic CEA immunosensor was applied in CEA spiked human serum samples for the competitive detection of CEA. After measurements, the standard addition plot was constructed to determine the CEA concentration. Fig. 6b shows the standard addition plots for the detection of CEA in human serum samples. The recovery results are shown in Table S1 in the Supplementary materials. The CEA recoveries were between 97 and 102% with RSD values ranging between 4.4 and 6.2%, which clearly indicated the good applicability of the proposed CEA immunosensor in real human serum samples. In addition, the CEA spiked human serum samples were also analyzed using ELISA experiments and a good correlation was found between immunosensor and ELISA results.
4. Conclusions An ultrasensitive electrochemical competitive immunosensor based on MB-supported multiple non-enzymatic labels was developed for the detection of CEA. The use of MB not only supports the multiple TMB labels in CEA/MB/TMB conjugates but also provides very quick and efficient separation and purification of the labels, which enhanced the performances of TMB-electro-catalyzed AA oxidation for highly sensitive CEA detection. The multiple non-enzymatic labels not only enhanced the sensitivity of the
R. Akter et al. / Biosensors and Bioelectronics 54 (2014) 351–357
357
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.2013.10.058. References
Fig. 6. (a) The long-term stability responses and (b) the standard addition plots for the quantification of free-CEA in human serum samples: (i) 1.0, (ii) 2.0, (iii) 5.0, and (iv) 10 ng/mL.
immunosensor but also increased the stability for two months. The detection limit of the proposed non-enzymatic CEA immunosensor was about one order lower than those obtained for the enzymatic labels-based immunosensors. The easy fabrication of the proposed non-enzymatic immunosensor is expected to provide a promising platform for other protein biomarkers detection.
Acknowledgments This research was supported by the Basic Research Program for regional university (2013-0730) and a NRF grant (2013-R1A1A2007139) funded by the National Research Foundation, Korean government (MEST).
Akter, R., Rahman, M.A., Rhee, C.K., 2012. Anal. Chem. 84, 6407–6415. Blonder, R., Katz, E., Cohen, Y., Itzhak, N., Riklin, A., Willner, I., 1996. Anal. Chem. 68, 3151–3157. Chang, C.-C., Chiu, N.-F., Lin, D.S., Chu-Su, Y., Liang, Y.-H., Lin, C.-W., 2010. Anal. Chem. 82, 1207–1212. Chen, W.W., Li, B.X., Xu, C.L., Wang, L., 2009. Biosens. Bioelectr. 24, 2534–2540. Das, J., Aziz, M.A., Yang, H., 2006. J. Am. Chem. Soc. 128, 16022–16023. Du, D., Wang, L., Shao, Y., Wang, J., Engelhard, M.H., Lin, Y., 2011. Anal. Chem. 83, 746–752. Du, D., Zou, Z., Shin, Y., Wang, J., Wu, H., Engelhard, M.H., Liu, J., Aksay, I.A., Lin, Y., 2010. Anal. Chem. 82, 2989–2995. Fu, Z.F., Hao, C., Fei, X., Ju, H.X., 2006. J. Immunol. Methods 312, 61–67. Ho, J.A., Chang, H.C., Shih, N.Y., Wu, L.C., Chang, Y.F., Chen, C.C., Chou, C., 2010. Anal. Chem. 82, 5944–5950. Hu, S.H., Zhang, S.C., Hu, Z.C., Xing, Z., Zhang, X.R., 2007. Anal. Chem. 79, 923–929. Jeong, B., Akter, R., Han, O.H., Rhee, C.K., Rahman, M.A., 2013. Anal. Chem. 85, 1784–1791. Jie, G., Li, L., Chen, C., Xuan, J., Zhu, J., 2010. Biosens. Bioelectr. 25, 1781–1788. Kang, K., Lee, S., Kim, R., Choi, I.S., Nam, Y., 2012. Angew. Chem. Int. Ed. 51, 13101–13104. Knudsen, S.M., Lee, J., Ellington, A.D., Savran, C.A., 2006. J. Am. Chem. Soc. 128, 15936–15937. Lai, G.S., Yan, F., Ju, H.X., 2009. Anal. Chem. 81, 9730–9736. Lee, H., Dellatore, S.M., Miller, W.M., Messersmith, P.B., 2007. Science 318, 426–430. Letilovic, T., Vrhovac, R., Verstovsek, S., Jaksik, B., Ferrajoli, A., 2006. Cancer 107, 925–934. Lin, M., Liu, Y., Liu, C., Yang, Z., Huang, Y., 2011. Biosens. Bioelectr. 26, 3761–3767. Liu, G.D., Wang, J., Kim, J., Jan, M.R., 2004. Anal. Chem. 76, 7126–7130. Liu, M., Zhang, Y., Chen, Y., Xie, Q., Yao, S., 2008. J. Electroanal. Chem. 622, 184–192. Mani, V., Chikkaveeraiah, B.V., Patel, V., Gutkind, J.S., Rusling, J.F., 2009. ACS Nano 3, 585–594. Matsuya, T., Tashiro, S., Hoshino, N., Shibata, N., Nagasaki, Y., Kataoka, K., 2003. Anal. Chem. 75, 6124–6132. Murray, R.W., 1984. Chemically modified electrodes. In: Bard, A.J. (Ed.), In Electroanalytical Chemistry, vol. 13. Marcel Dekker, New York, pp. 191–368. Noh, H.-B., Rahman, M.A., Yang, J.E., Shim, Y.-B., 2011. Biosens. Bioelectr. 26, 4429–4435. Pan, J., Yang, Q., 2007. Anal. Bioanal. Chem. 388, 279–286. Schmalzing, D., Nashabeh, W., 1997. Electrophoresis 18, 2184–2193. Serrat, F.B., 1994. Talanta 41, 2091–2094. Singh, K., Rahman, M.A., Son, J.I., Kim, K.C., Shim, Y.-B., 2008. Biosens. Bioelectr. 23, 1595–1601. Shi, J., Yang, C., Zhang, S., Wang, X., Jiang, Z., Zhang, W., Song, X., Ai, Q., Tian, C., 2013. ACS Appl. Mater. Interfaces 5, 9991–9997. Shiddiky, M.J.A., Rahman, M.A., Shim, Y.-B., 2007. Anal. Chem. 79, 6886–6890. Tang, D., Yuan, R., Chai, Y., 2008. Anal. Chem. 80, 1582–1588. Tang, H., Chen, J., Nie, L., Kuang, Y., Yao, S., 2007. Biosens. Bioelectr. 22, 1061–1076. Teppo, A.M., Maury, C.P., 1987. Adv. Clin. Chem. 33, 2024–2027. Ting, B.P., Zhang, J., Khan, M., Yang, Y.Y., Ying, J.Y., 2009. Chem. Commun., 6231–6233. Volpe, G., Compagnone, D., Draisci, R., Palleschi, G., 1998. Analyst 123, 1303–1307. Wang, J., Liu, G., Engelhard, M.H., Lin, Y., 2006. Anal. Chem. 78, 6974–6979. Wang, J., Liu, G.D., Wu, H., Lin, Y.H., 2008. Small 4, 82–86. Wu, J., Tang, J., Dai, Z., Yan, F., Ju, H., Murr, N.E., 2006. Biosens. Bioelectr. 22, 102–108. Wu, Y., Zheng, J., Li, Z., Zhao, Y., Zhang, Y., 2009. Biosens. Bioelectr. 24, 1389–1393. Yates, A.M., Elvin, S.J., Williamson, D.E., 1999. J. Immunoassay Immunochem. 20, 31–44. Yu, X., Munge, B., Patel, V., Jensen, J., Bhirde, A., Gong, J.D., Kim, S.N., Gillespie, J., Gutkind, J.S., Papadimitrakopoulos, F., Rusling, J.F., 2006. J. Am. Chem. Soc. 128, 11199–11205. Zhang, L., Wu, J., Wang, Y., Long, Y., Zhao, N., Xu, J., 2012. J. Am. Chem. Soc. 134, 9879–9881.
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Sensitivity enhancement of an electrochemical immunosensor through the electrocatalysis of magnetic bead-supported non-enzymatic labels Rashida Akter a, Choong Kyun Rhee a,b,n, Md. Aminur Rahman b,nn a b
Department of Chemistry, Chungnam National University, Daejeon 305-764, South Korea Graduate School of Analytical Science and Technology (GRAST), Chungnam National University, Daejeon 305-764, South Korea
art ic l e i nf o
a b s t r a c t
Article history: Received 16 August 2013 Received in revised form 25 October 2013 Accepted 28 October 2013 Available online 4 November 2013
An ultrasensitive non-enzymatic electrochemical carcinoembryonic antigen (CEA) immunosensor was fabricated by the immobilization of a monoclonal CEA antibody (anti-CEA) on a protein A (PA) attachedgold nanoparticles (AuNPs)-deposited electrochemically prepared polydopamine film (e-PD/AuNPs). Magnetic beads (MB)-supported and CEA-conjugated multiple 3,3′,5,5′-tetramethylbenzidine (TMB) was used as electrochemical labels. The detection was based on the measurements of the electrocatalyzed oxidation of ascorbic acid (AA) by the multiple TMB labels after competitive binding between MB/TMBconjugated-CEA and free-CEA. The electrocatalyzed oxidation current of AA by TMB decreased with increasing concentration of the free-CEA as the amount of CEA/MB/TMB labels decreased at the immunosensor probe. The immunosensor surface was characterized using electrochemical impedance spectroscopy, Fourier transform infrared spectroscopy, quartz crystal microbalance, and scanning electron microscopy techniques. Cyclic voltammetry (CV) and differential pulse voltammetry (DPV) techniques were used to monitor the electrocatalyzed response. The proposed immunosensor exhibited a wide linear dynamic range (1.0 pg/mL to 10.0 ng/mL), low detection limit (1.07 0.04 pg/mL), good selectivity, and long-time stability. It was successfully applied to various CEA spiked human serum samples for the detection of CEA. & 2013 Elsevier B.V. All rights reserved.
Keywords: Non-enzymatic electrocatalysis 3,3′,5,5′-Tetramethylbenzidine Electrochemical immunosensor Nanoparticles Polydopamine film Protein biomarker detection
1. Introduction The measurement of trace amounts of protein biomarkers plays essential roles in biochemical and biomedical researches as well as for the development of practical medical diagnostics for early disease detection, prognostic treatment, and patient curative effect monitoring. Therefore, measuring their presences in body fluids remains a great challenge in disease diagnostics. Conventional immunoassay methods such as enzyme-linked immunosorbent assay (ELISA) (Yates et al., 1999), fluorescence immunoassay (Matsuya et al., 2003), chemiluminescence assay (Fu et al., 2006), radioimmunoassay (Teppo and Maury, 1987), mass spectrometric immunoassay (Hu et al., 2007), metalloimmunoassay (Noh et al., 2011), and electrophoretic immunoassay (Schmalzing and Nashabeh, 1997) are commonly used for trace amounts protein detection. However, each of these immunoassay methods has drawbacks: complicated, time-consuming, tedious, expensive, labor-intensive, and not suitable for point-of-care applications. As n
Corresponding author. Tel.: þ 82 42 821 5483; fax: þ 82 42 821 8896. Corresponding author. Tel.: þ 82 42 821 8546; fax: þ 82 42 821 8541. E-mail addresses: [email protected] (C. Kyun Rhee), [email protected] (Md. Aminur Rahman). nn
0956-5663/$ - see front matter & 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bios.2013.10.058
alternative approaches, various immunosensor methods based on surface plasma resonance (SPR) (Chang et al., 2010), quartz crystal microbalance (QCM) (Knudsen et al., 2006), chemiluminescence (Jie at al., 2010), and electrochemistry (Blonder et al., 1996; Wang et al., 2006) have been developed for the protein detection. Among them, electrochemical immunosensors have shown a great promise because of simple instrumentation, high sensitivity, fast response time, miniaturization, low cost, and point-of-care applications. Coupling with nanomaterials and various amplification techniques, several electrochemical immunosensors have been successfully developed for detecting various protein biomarkers (Akter et al., 2012; Jeong et al., 2013; Tang et al., 2008; Du et al., 2011). In most cases, nanomaterial-supported enzymes are used as electrochemical labels for increasing the detection sensitivity. However, enzymes are not advantageous because of costly preparation and purification process (Chen et al., 2009), which limits their uses as labels in electrochemical immunosensors. Thus, the search for substitutes of enzymes in electrochemical protein immunosensors would be very attractive. Various non-enzymatic labels including quantum dots (QD) (Liu et al., 2004; Wang et al., 2008), metallic nanoparticles (Ting et al., 2009; Ho et al., 2010; Das et al., 2006), hydrazine (Shiddiky et al., 2007), and metal ion-cysteamine complex (Noh et al., 2011), have been used in
352
R. Akter et al. / Biosensors and Bioelectronics 54 (2014) 351–357
electrochemical protein immunosensors. Although the above nonenzymatic labels are able to enhance the detection sensitivity, however, they suffer from some inherent disadvantages. For example, QD-based detection requires harsh condition for nanocrystal dissolution for stripping analysis, hydrazine-based detection requires deoxygenation of the detection medium, and metal– ion–cysteamine complex-based detection requires multistep processes for synthesizing the complex prior to electrochemical detection. Metallic nanoparticles label such as gold (Au) and silver (Ag) can be directly detected by stripping voltammetry. However, the oxidation peaks of Au and Ag occur at relatively positive potentials where other biomolecules can be oxidized. Thus, it is a major challenge to incorporate new non-enzymatic labels having enhanced electrocatalytic activities in electrochemical immunosensors. 3,3′,5,5′-Tetramethylbenzidine (TMB) is a widely used chromogen in enzyme-linked immunosorbent assay (ELISA). It is a nonmutagenic and non-carcinogenic compound that has been widely used for the detection of blood and urine sugar, hypothermia, horseradish peroxidase enzyme, and some antibodies in clinical samples (Serrat, 1994). TMB is an electroactive compound, showing a reversible redox reaction under appropriate conditions (Liu et al., 2008), so that it has been used for the fabrication of a reagentless amperometric immunosensor (Wu et al., 2009). The amine groups of TMB are easy to conjugate with antibody or other proteins by covalent cross-linking. Thus, TMB is a promising electroactive label for the fabrication of highly sensitive electrochemical immunosensor. In order to increase the amount of electroactive labels, various nanomaterials are used as nanocarriers. Gold nanoparticles (AuNPs) (Lai et al., 2009), magnetic bead (MB) (Mani et al., 2009), multi wall carbon nanotubes (MWCNT) (Akter et al., 2012; Yu et al., 2006), silica nanoparticles (Lin et al., 2011), carbon spheres (Du et al., 2010), and dendrimer/AuNPs (Den/AuNPs) (Jeong et al., 2013) have been used as nanocarriers in electrochemical immunosensor. For example, Rusling group has enhanced the detection limit of prostate specific antigen by using MB-supported enzymatic bioconjugates, which shows 8-fold better performances than carbon nanotube (CNT)-supported enzymatic bioconjugates (Mani et al., 2009). The MB-supported non-enzymatic labels (MB-TMB) would be a prospective amplification strategy for the high sensitive cancer biomarker detection because enhanced amount of non-enzymatic TMB labels can be loaded onto the MB. In the present study, the electrocatalytic activity of the magnetic bead-supported TMB labels has been demonstrated for the first time for developing an electrochemical carcinoembryonic antigen (CEA) immunosensor. CEA has been chosen as a model protein as it is associated with colorectal cancer and also identified as a biomarker for breast tumors and ovarian carcinoma (Letilovic et al., 2006). For the fabrication of the immunosensor probe, monoclonal anti-CEA antibody has been immobilized onto the protein A (PA)-attached electrochemically prepared gold nanoparticles/polydopamine (e-PD/e-AuNPs/PA) film. The purpose of using the e-PD film is to make a biocompatible and stable immunosensor platform that allows nanoparticles deposition and biomolecules immobilization. The major advantage of this material is that it has high biocompatibility (Shi et al., 2013) and remarkable adhesive ability (Zhang et al., 2012) and could deposit very stably on all kinds of organic and inorganic surfaces, including nonsticking surfaces (Lee et al., 2007). However, the conductivity of the e-PD film is low. To increase the surface conductivity of the e-PD film, electrochemically prepared AuNPs were deposited on the surface of e-PD film. The CEA detection has been performed by monitoring the electrocatalytic activity of the MB-supported TMB labels towards AA oxidation in phosphate buffer saline solution (PBS) after the competitive binding reaction between free-CEA and MB-supported TMB-conjugated CEA (CEA/MB/TMB).
2. Experimental 2.1. Preparation of CEA/MB/TMB conjugates For the preparation of CEA/MB/TMB conjugates, the COOH groups of MB (2 mg/mL) were activated by treating with 10 mM NHS/EDC in 50 mM MES buffer at pH 5.8 for 6 h. The activated COOH-MB was separated from the free NHS/EDC by magnetic separation. 5.0 ng/mL CEA and 1.0 mg/mL TMB were then added to the activated COOH-MB and stirred for 24 h at 4 1C. By this step, CEA and TMB were attached to the MB through the covalent bond formation between the carboxylic acid groups of MB and amine groups of CEA and TMB. The CEA/MB/TMB conjugate was separated from the free CEA and TMB by magnetic separation and was washed several times with a PBS solution. Finally, the CEA/MB/ TMB conjugate was diluted with 1.0 mL of 0.1 M PBS (pH 7.4) and kept at 4 1C. 2.2. Fabrication of the immunosensor probe and the sensing principle Fig. 1 shows the schematic illustration of the fabrication of immunosensor and the sensing principle of the proposed CEA detection. At first, an e-PD film was grown on gold (Au) electrodes through the electropolymerization of dopamine using a potential cycling method. Prior to electropolymerization, Au electrodes were polished with a 0.05 μm alumina/water slurry on a polishing cloth to a mirror finish, followed by sonicating and rinsing with distilled water. Then, the polished electrodes were rinsed with fresh piranha solution (70% H2SO4, 30% H2O2; CAUTION: piranha solution reacts violently with most organic materials and must be handled with extreme care) followed by cycling the potential between þ1.4 and 0.2 V at 100 mV/s for 50 times in a 1.0 M H2SO4 solution. The electropolymerization of dopamine (1.0 mg/mL) was carried out in a 0.1 M PBS solution of pH 6.0 by cycling the potential between 0 and 0.7 V for nine times. The pH of the PBS solution was chosen 6.0 as the chemical polymerization of dopamine occurred under basic solution (Lee et al., 2007). After electropolymerization, the e-PD film coated Au electrode was washed with a PBS solution (pH ¼6.0) to remove any remaining dopamine from the electrode surface. AuNPs were then electrochemically deposited on the e-PD film by sweeping the potential between 0 and þ1.4 V in a 0.1 M H2SO4 solution containing 1 mM HAuCl4 at the scan rate of 100 mV/s (Singh et al., 2008). PA was then immobilized onto the e-AuNPs deposited e-PD film (Au/e-PD/ e-AuNPs) by incubating the Au/e-PD/e-AuNPs electrode in a PBS solution containing 3.0 mg/mL of PA for 8 h at 4 1C. PA was attached onto the e-PD/e-AuNPs through the formation of Au-NH bond. Immobilization of a monoclonal anti-CEA antibody was carried out by incubating the Au/e-PD/e-AuNPs/PA electrodes in a 0.1 M phosphate PBS solution (pH 7.0) containing 0.1 mg/mL anti-CEA for 24 h at 4 1C. After washing three times with 0.1 M PBS, the Au/e-PD/e-AuNPs/PA/anti-CEA electrodes (immunosensors) were blocked by dipping them in 0.1% BSA solution for 2 h at 4 1C. After washing three or four times with PBS and drying with N2, the blocked immunosensors were incubated in the mixtures of a fixed amount of CEA/MB/TMB and various concentrations of free-CEA at 4 1C for the competitive binding of free- and labeledCEA to the active sites of anti-CEA. After the competitive binding, the immunosensors were washed three or four times with PBS to remove nonspecifically bound free- and labeled- CEA. Then, the immunosensors were dipped in a 0.1 M PBS (pH 7.0) solution for monitoring the CV and DPV responses of the oxidation process of MB/TMB labels. The electrocatalytic activity of the MB/TMB label towards the oxidation of ascorbic acid (AA) was studied for the signal amplification.
R. Akter et al. / Biosensors and Bioelectronics 54 (2014) 351–357
353
Fig. 1. (a) Schematic representation of the immunosensor and (b) the mechanism of CEA detection through the TMB electrocatalyzed AA oxidation.
3. Results and discussion 3.1. Characterization of the e-PD and e-PD/e-AuNPs modified electrode The e-PD film grown onto an Au electrode was characterized by using the CV technique. Fig. 2a shows a series of CVs recorded during the electropolymerization of PD in a 0.1 M PBS solution of pH 6.0 at the scan rate of 0.1 V/s. The CV exhibited an oxidation peak at þ 0.35 V during the first anodic scan, which was due to the oxidation of PD to form e-PD. The e-PD film progressively covered the electrode surface, which increased the thickness of the film and decreased the surface area of the electrode. Thus, the peak currents were found to be decreased as the cycle numbers increased. These results clearly demonstrate that the e-PD film immediately formed after the oxidation of the PD at þ 0.35 V (Kang et al., 2012). A small reduction peak of the e-PD was observed at þ0.048 V during the cathodic scan. In order to determine the amount of e-PD formed, electrochemical quartz crystal microbalance (EQCM) experiment was carried out. Fig. 2b shows the frequency changes obtained from an EQCM experiment. During the electropolymerization of PD, the frequency decreased due to the formation of e-PD film. Deposition of e-PD, took place rapidly as the cycle number increased. The Δf value was 161.5 Hz after nine cycles. The amount of e-PD grown was determined to be 404 722 ng by using the sensitivity factor of 2.5 (Akter et al., 2012), corresponding to a surface coverage of 2.1 10 9 mol/cm2. The redox activity of the e-PD film was recorded for an e-PD modified Au electrode in a PBS solution (Fig. 2c). A pair of oxidation and reduction peaks was observed at þ0.15/ þ 0.25 V, indicating that the e-PD film is electrochemically active. The peak currents of these oxidation and reduction peaks were proportional to the scan rate due to the redox behavior of the adsorbed e-PD film. The electrochemically deposited AuNPs on e-PD (e-PD/e-AuNPs) was characterized using the CV technique (Fig. S1). Au oxidation and reduction peaks were observed at þ0.97 and þ0.45 V, respectively,
which was not observed in the case of e-PD electrode, indicating clearly that AuNPs successfully deposited on the e-PD film. Fig. 2d shows the SEM image obtained for the e-PD/e-AuNPs surface, showing the AuNPs were uniformly distributed on the e-PD film with the particle sizes ranges between 5 and 20 nm. 3.2. Characterizations of CEA/MB/TMB conjugates and the immunosensor probe Fig. 3 compares the SEM images of MB before (a) and after (b) conjugating with CEA and TMB. The SEM image obtained after conjugation showed that there are overlayers on MB, whose radius is 0.1 μm. The existence of such overlayers on MB confirmed that CEA and TMB conjugated to the MB. The Fourier transform infrared (FT-IR) spectra also confirmed that CEA and TMB conjugated to the MB (Fig. S2). The immunosensor probe was characterized using electrochemical impedance spectroscopy (EIS) and the quartz crystal microbalance (QCM) techniques. Fig. 3c shows the Nyquist plot observed for a e-pD (i), e-pD/e-AuNPs (ii), e-pD/e-AuNPs/PA (iii), e-pD/e-AuNPs/PA/anti-CEA (iv) modified electrodes recorded in a 5.0 mM of Fe(CN)63 /4 solution. The inset of Fig. 3c shows a general equivalent circuit that can be used to model the impedance data. The general equivalent circuit contains the solution resistance (Rs), the charge transfer resistance (Rct), the Warburg element (W), and the charge of the constant phase element (CPE). The Rct can be estimated from the diameter of the semicircle part at higher frequencies in the Nyquist plot. For the e-PD modified electrode, the Rct value was estimated as 60 kΩ, which significantly decreased to about 15 kΩ after e-AuNPs deposition. This result indicates that e-AuNPs increased the conductivity of the e-PD film. However, the Rct value increased to about 20 kΩ after PA immobilization due to the increase of the thickness of the electrode surface. PA was chemically bonded onto the e-PD/e-AuNPs through the interaction of Au-NH bonds. After the immobilization of anti-CEA on the e-PD/e-AuNPs/PA, the Rct value further increased to 25 kΩ. The above results clearly show that the PA and anti-CEA successfully immobilized on the modified e-PD/e-AuNPs electrode. The immunosensor surface was also
354
R. Akter et al. / Biosensors and Bioelectronics 54 (2014) 351–357
Fig. 2. (a) Consecutive CVs (9 cycles) recorded for the electropolymerization of 1.0 mg/mL dopamine in a 0.1 M PBS solution of pH 6.0. The san rate was 0.05 V/s. (b) EQCM responses concurrently recorded with the CV for the electropolymerization of dopamine. (c) CVs recorded for the e-PD modified electrode in a 0.1 M PBS solution at various scan rates: (i–vi) 0.05-0.5 V/s. Inset shows the current vs. scan rate plot. (d) SEM image of the e-PD/e-AuNPs modified electrode. Inset shows the image of an e-PD modified electrode.
characterized using the QCM technique (Fig. S3) and is discussed in the Supplementary materials.
3.3. Electrochemical characteristics of CEA immunosensor Fig. 4a shows the CV recorded for an Au/e-PD/e-AuNPs/PA/antiCEA modified electrode (immunosensor probe) after binding with CEA/MB/TMB conjugates (200 μL) in a PBS solution (pH 7.4) in the absence (i) or presence (ii) of 1.0 mM AA. In the absence of AA, well-defined oxidation and reduction peaks were observed at 0.39 and 0.25 V, respectively, which was not observed when the CVs were recorded after binding with CEA and CEA/MB. The Au/e-PD/ e-AuNPs/PA/anti-CEA immunosensor probe itself also did not show any peak. These results indicate that the redox peak came solely from the conjugated TMB, which was bonded to the immunosensor probe as CEA/MB/TMB. For a comparison, the redox behavior of the CEA/TMB conjugate was studied and compared with that of CEA/MB/TMB (Fig. S3). The peak separation of only CEA/TMB was about 0.06 V, whereas it was about 0.16 V for CEA/MB/TMB. These results indicate that the electron transfer (ET) process of the CEA/MB/TMB shows more irreversibility than that of CEA/TMB. This more irreversible ET process of CEA/MB/TMB was due to the large electron transfer distance between the electrode and the MB-supported TMB molecules. Another interesting result is that the electrochemical behavior of CEA/MB/TMB only shows a single redox peak, which is in contrast to the two one-electron reversible waves previously observed for TMB in solution at neural or acidic pH (Volpe et al., 1998). The redox peak currents were directly proportional to the scan rate up to 0.5 V/s, confirming that the redox reaction was involved in a surface-confined process (Murray, 1984) of TMB þ þ þ2e⇌TMB.
When 1.0 mM AA was added to the PBS solution, the catalytic characteristics observed with a dramatic increase of the oxidation peak and sharp decrease of the reduction current (Fig. 4a (ii)), which was due to the electrocatalytic oxidation of AA by the conjugated TMB. The electrocatalytic mechanism of AA oxidation can be expressed by the following equation: TMB⇌TMB þ þ 2e ðat electrodeÞ
ð1Þ
TMB þ þ AA-TMB þ DHAA
ð2Þ
The electrocatalytic AA oxidation current was much higher than the TMB oxidation current. To confirm the electrocatalytic process of AA results from the TMB labels and not from the electrodeposited PD film, the AA electrooxidation was measured with the immunosensor probe without PD film. The electrooxidation of AA was observed even without electrodeposition of PD film (Au/PA/anti-CEA/CEA/MB/TMB) (Fig. 4a (iii)). The electrocatalytic peak slightly shifted to the positive potential and the peak current was found to be decreased. The decrease in peak current might be related to the immobilization of lesser amounts of PA and anti-CEA due to the absence of PD/AuNPs, resulted in a lesser amount of TMB labels. The electrocatalytic peak slightly shifted to the positive potential might be due to low conductivity of the immunosensor surface in the absence of AuNPs. The existence of the AA electrocatalytic peak in the absence of PD clearly showed the electrocatalytic response results from the TMB label on the MB not from the PD film. Fig. 4b shows the DPVs recorded for an immunosensor after binding with CEA/MB/TMB conjugates in a PBS solution in the absence (i) or presence (ii) of 1.0 mM AA. In the absence of AA, the DPV response showed an anodic peak at about þ0.32 V for the oxidation of TMB. In the presence of AA, the dramatic increase of oxidation current due to the electrocatalytic
R. Akter et al. / Biosensors and Bioelectronics 54 (2014) 351–357
355
Fig. 3. SEM images of (a) COOH-MB and (b) CEA/MB/TMB conjugate. Insets of both figures show the images of a single particle. (c) EIS analyses of (i) Au/e-PD, (ii) Au/e-PD/eAuNPs, (iii) Au/e-PD/e-AuNPs/PA, (iv) Au/e-PD/e-AuNPs/PA/anti-CEA, and (v) bare Au electrodes in 5.0 mM Fe(CN)63 /4 solution. Inset shows the general equivalent circuit.
Fig. 4. (a) CV and (b) DPV responses recorded for Au/e-PD/e-AuNPs/PA/anti-CEA/CEA/MB/TMB in 0.1 M PBS solutions containing no AA (i) and 1.0 mM AA (ii), and for Au/PA/ anti-CEA/CEA/MB/TMB with 1.0 mM AA (iii). (c) DPV responses on the electrocatalytic AA oxidation recorded for an Au/e-PD/e-AuNPs/PA/anti-CEA-based immunosensor in a 0.1 M PBS solution before (i) and after (ii) competitive binding between free-CEA and CEA/MB/TMB conjugates. The concentration of free-CEA was 2.0 ng/mL and the amount of CEA/MB/TMB was 200 μL.
356
R. Akter et al. / Biosensors and Bioelectronics 54 (2014) 351–357
3.4. Detection of CEA Under the optimized experimental condition, the DPV current responses were measured before and after competitive binding of conjugated-CEA and free-CEA. Fig. 5a shows the DPV responses, where the electrocatalytic oxidation current of AA linearly decreased with the increasing concentration of free-CEA. Fig. 5b shows the calibration plot constructed by plotting the differences in current responses (ΔI) between before and after competitive binding (ICEA/MB/TMB ICEA/MB/TMB þ free-CEA) of CEA/MB/TMB and free-CEA. The calibration plot at the lower concentration range is shown in the inset of Fig. 5b. The immunosensor exhibits a wide dynamic range between 1.0 pg/mL and 10.0 ng/mL. The reproducibility expressed in terms of the relative standard deviation (RSD) was about 4.9% (n ¼5) at a CEA concentration of 2.0 ng/mL. The experimental detection limit of CEA was determined to be 1.0 70.04 pg/mL, which was 10 times lower than that obtained without magnetic nanoparticles assisted electrocatalytic activities of multiple TMB labels (Akter et al., 2012). The detection limit of the proposed immunosensor was also lower than HRP-anti-CEA/nanogold/chitosan (0.22 ng/mL) (Wu et al., 2006), anti-CEA/glutathione// nanogold/poly-o-aminophenol /Au (0.1 ng/mL) (Tang et al., 2007), thionine-doped magnetic gold nanpshheres (10 pg/mL) (Tang et al., 2008), anti-CEA/GA/GPMS/Fe3O4/SiO2/CPE (0.5 ng/mL) (Pan and Yang, 2007), and dendrimer-encapsulated gold nanoparticles and carbon nanotube-assisted multiple bienzymatic labels-based electrochemical immunosensors (4.4 pg/mL) (Jeong et al., 2013). This means that the electrocatalytic activity of MB-supported TMB labels improved the sensitivity of the CEA detection. 3.5. Stability and the real sample analysis
Fig. 5. (a) DPV responses on the electrocatalytic AA oxidation recorded at various free-CEA concentrations: (i) 0, (ii) 1.0, (iii) 10, (iv) 100 pg/mL, (v) 1.0, (vi) 2.0, (vii) 5.0, (viii) 10, and (ix) 20 ng/mL. (b) Calibration plot for the CEA detection. Insets of both figures show the magnified portion of the peaks and calibration plot at the low concentration range (1.0 100 pg/mL).
oxidation of AA by TMB was observed. The both CV and DPV studies clearly revealed that the TMB oxidation current could be enhanced by using electrocatalytic oxidation of AA by TMB. The electrocatalytic oxidation current of AA was considered as a base immunosensor response and monitored this response during the competitive binding between the fixed amount of TMB-labeled CEA and various concentrations of free-CEA. Fig. 4c shows the DPVs recorded for the immunosensor probe after incubating it in only (i) CEA/MB/TMB (before competitive binding) and a (ii) mixture of CEA/MB/TMB þ2 ng/mL of free-CEA (after competitive binding) solutions. The electrocatalytic oxidation current response after competitive binding was found to be decreased. In the absence of free-CEA, there was no competition, thus all the active sites of anti-CEA were bonded by the CEA/MB/TMB. However, when 2 ng/mL of free-CEA was present with CEA/MB/TMB, the competition between the free-CEA and conjugate-CEA for the antiCEA binding sites occurred. As a result, the amount of CEA/MB/ TMB at the immunosensor probe reduced and a decrease in the current response was observed. The difference in electrocatalytic current response (ΔI) before and after competitive binding was directly proportional to the concentration of free-CEA. To maximize the immunosensor response, various experimental parameters such as amount of PA, PA/anti-CEA interaction, anti-CEA dilution factor, competitive interaction time and temperature between CEA and CEA/MB/TMB with anti-CEA, and the pH of the binding medium were optimized (Fig. S5) and are discussed in the Supplementary materials.
The stability of the proposed CEA immunosensor was checked by measuring the competitive response of 1.0 ng/mL CEA for two months. After each measurement, the immunosensor surface was regenerated by dipping it into a 0.2 M glycine-hydrochloric acid (Gly-HCl) solution (pH ¼2.8) for 5 min followed by washing with a PBS solution. When stored in dry condition at 4 1C, the initial ΔI response did not change significantly for a period of two months (Fig. 6a), indicating that the long time stability of the immunosensor was good. The proposed non-enzymatic CEA immunosensor was applied in CEA spiked human serum samples for the competitive detection of CEA. After measurements, the standard addition plot was constructed to determine the CEA concentration. Fig. 6b shows the standard addition plots for the detection of CEA in human serum samples. The recovery results are shown in Table S1 in the Supplementary materials. The CEA recoveries were between 97 and 102% with RSD values ranging between 4.4 and 6.2%, which clearly indicated the good applicability of the proposed CEA immunosensor in real human serum samples. In addition, the CEA spiked human serum samples were also analyzed using ELISA experiments and a good correlation was found between immunosensor and ELISA results.
4. Conclusions An ultrasensitive electrochemical competitive immunosensor based on MB-supported multiple non-enzymatic labels was developed for the detection of CEA. The use of MB not only supports the multiple TMB labels in CEA/MB/TMB conjugates but also provides very quick and efficient separation and purification of the labels, which enhanced the performances of TMB-electro-catalyzed AA oxidation for highly sensitive CEA detection. The multiple non-enzymatic labels not only enhanced the sensitivity of the
R. Akter et al. / Biosensors and Bioelectronics 54 (2014) 351–357
357
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.2013.10.058. References
Fig. 6. (a) The long-term stability responses and (b) the standard addition plots for the quantification of free-CEA in human serum samples: (i) 1.0, (ii) 2.0, (iii) 5.0, and (iv) 10 ng/mL.
immunosensor but also increased the stability for two months. The detection limit of the proposed non-enzymatic CEA immunosensor was about one order lower than those obtained for the enzymatic labels-based immunosensors. The easy fabrication of the proposed non-enzymatic immunosensor is expected to provide a promising platform for other protein biomarkers detection.
Acknowledgments This research was supported by the Basic Research Program for regional university (2013-0730) and a NRF grant (2013-R1A1A2007139) funded by the National Research Foundation, Korean government (MEST).
Akter, R., Rahman, M.A., Rhee, C.K., 2012. Anal. Chem. 84, 6407–6415. Blonder, R., Katz, E., Cohen, Y., Itzhak, N., Riklin, A., Willner, I., 1996. Anal. Chem. 68, 3151–3157. Chang, C.-C., Chiu, N.-F., Lin, D.S., Chu-Su, Y., Liang, Y.-H., Lin, C.-W., 2010. Anal. Chem. 82, 1207–1212. Chen, W.W., Li, B.X., Xu, C.L., Wang, L., 2009. Biosens. Bioelectr. 24, 2534–2540. Das, J., Aziz, M.A., Yang, H., 2006. J. Am. Chem. Soc. 128, 16022–16023. Du, D., Wang, L., Shao, Y., Wang, J., Engelhard, M.H., Lin, Y., 2011. Anal. Chem. 83, 746–752. Du, D., Zou, Z., Shin, Y., Wang, J., Wu, H., Engelhard, M.H., Liu, J., Aksay, I.A., Lin, Y., 2010. Anal. Chem. 82, 2989–2995. Fu, Z.F., Hao, C., Fei, X., Ju, H.X., 2006. J. Immunol. Methods 312, 61–67. Ho, J.A., Chang, H.C., Shih, N.Y., Wu, L.C., Chang, Y.F., Chen, C.C., Chou, C., 2010. Anal. Chem. 82, 5944–5950. Hu, S.H., Zhang, S.C., Hu, Z.C., Xing, Z., Zhang, X.R., 2007. Anal. Chem. 79, 923–929. Jeong, B., Akter, R., Han, O.H., Rhee, C.K., Rahman, M.A., 2013. Anal. Chem. 85, 1784–1791. Jie, G., Li, L., Chen, C., Xuan, J., Zhu, J., 2010. Biosens. Bioelectr. 25, 1781–1788. Kang, K., Lee, S., Kim, R., Choi, I.S., Nam, Y., 2012. Angew. Chem. Int. Ed. 51, 13101–13104. Knudsen, S.M., Lee, J., Ellington, A.D., Savran, C.A., 2006. J. Am. Chem. Soc. 128, 15936–15937. Lai, G.S., Yan, F., Ju, H.X., 2009. Anal. Chem. 81, 9730–9736. Lee, H., Dellatore, S.M., Miller, W.M., Messersmith, P.B., 2007. Science 318, 426–430. Letilovic, T., Vrhovac, R., Verstovsek, S., Jaksik, B., Ferrajoli, A., 2006. Cancer 107, 925–934. Lin, M., Liu, Y., Liu, C., Yang, Z., Huang, Y., 2011. Biosens. Bioelectr. 26, 3761–3767. Liu, G.D., Wang, J., Kim, J., Jan, M.R., 2004. Anal. Chem. 76, 7126–7130. Liu, M., Zhang, Y., Chen, Y., Xie, Q., Yao, S., 2008. J. Electroanal. Chem. 622, 184–192. Mani, V., Chikkaveeraiah, B.V., Patel, V., Gutkind, J.S., Rusling, J.F., 2009. ACS Nano 3, 585–594. Matsuya, T., Tashiro, S., Hoshino, N., Shibata, N., Nagasaki, Y., Kataoka, K., 2003. Anal. Chem. 75, 6124–6132. Murray, R.W., 1984. Chemically modified electrodes. In: Bard, A.J. (Ed.), In Electroanalytical Chemistry, vol. 13. Marcel Dekker, New York, pp. 191–368. Noh, H.-B., Rahman, M.A., Yang, J.E., Shim, Y.-B., 2011. Biosens. Bioelectr. 26, 4429–4435. Pan, J., Yang, Q., 2007. Anal. Bioanal. Chem. 388, 279–286. Schmalzing, D., Nashabeh, W., 1997. Electrophoresis 18, 2184–2193. Serrat, F.B., 1994. Talanta 41, 2091–2094. Singh, K., Rahman, M.A., Son, J.I., Kim, K.C., Shim, Y.-B., 2008. Biosens. Bioelectr. 23, 1595–1601. Shi, J., Yang, C., Zhang, S., Wang, X., Jiang, Z., Zhang, W., Song, X., Ai, Q., Tian, C., 2013. ACS Appl. Mater. Interfaces 5, 9991–9997. Shiddiky, M.J.A., Rahman, M.A., Shim, Y.-B., 2007. Anal. Chem. 79, 6886–6890. Tang, D., Yuan, R., Chai, Y., 2008. Anal. Chem. 80, 1582–1588. Tang, H., Chen, J., Nie, L., Kuang, Y., Yao, S., 2007. Biosens. Bioelectr. 22, 1061–1076. Teppo, A.M., Maury, C.P., 1987. Adv. Clin. Chem. 33, 2024–2027. Ting, B.P., Zhang, J., Khan, M., Yang, Y.Y., Ying, J.Y., 2009. Chem. Commun., 6231–6233. Volpe, G., Compagnone, D., Draisci, R., Palleschi, G., 1998. Analyst 123, 1303–1307. Wang, J., Liu, G., Engelhard, M.H., Lin, Y., 2006. Anal. Chem. 78, 6974–6979. Wang, J., Liu, G.D., Wu, H., Lin, Y.H., 2008. Small 4, 82–86. Wu, J., Tang, J., Dai, Z., Yan, F., Ju, H., Murr, N.E., 2006. Biosens. Bioelectr. 22, 102–108. Wu, Y., Zheng, J., Li, Z., Zhao, Y., Zhang, Y., 2009. Biosens. Bioelectr. 24, 1389–1393. Yates, A.M., Elvin, S.J., Williamson, D.E., 1999. J. Immunoassay Immunochem. 20, 31–44. Yu, X., Munge, B., Patel, V., Jensen, J., Bhirde, A., Gong, J.D., Kim, S.N., Gillespie, J., Gutkind, J.S., Papadimitrakopoulos, F., Rusling, J.F., 2006. J. Am. Chem. Soc. 128, 11199–11205. Zhang, L., Wu, J., Wang, Y., Long, Y., Zhao, N., Xu, J., 2012. J. Am. Chem. Soc. 134, 9879–9881.
