Amplified Electrochemical Detection of a Cancer Biomarker by

Jun 30, 2012 - An electrochemical nanoimmunosensor based on multiwall carbon nanotubes (MWCNTs)/gold nanoparticles (AuNPs) was developed for the ampli...
2 downloads 4 Views 2MB Size
Download PDF
Article pubs.acs.org/ac

Amplified Electrochemical Detection of a Cancer Biomarker by Enhanced Precipitation Using Horseradish Peroxidase Attached on Carbon Nanotubes Rashida Akter,† Md. Aminur Rahman,*,‡ and Choong Kyun Rhee*,†,‡ †

Department of Chemistry, Chungnam National University, Daejeon 305-764, South Korea Graduate School of Analytical Science and Technology, Chungnam National University, Daejeon 305-764, South Korea

‡

ABSTRACT: An electrochemical nanoimmunosensor based on multiwall carbon nanotubes (MWCNTs)/gold nanoparticles (AuNPs) was developed for the amplified detection of prostate specific antigen (PSA). The amplified detection was achieved by the enhanced precipitation of 4-chloro-1naphthol (CN) using a higher number of horseradish peroxidase (HRP) molecules attached on MWCNTs. The PSA nanoimmunosensor was fabricated by immobilizing a monoclonal anti-PSA antibody (anti-PSA) on the AuNPattached thiolated MWCNT on a gold electrode. The sensor surface was characterized using scanning electron microscope, transmission electron microscope, quartz crystal microbalance, and electrochemical techniques. Cyclic and square wave voltammetric techniques were used to monitor the enhanced precipitation of CN that accumulated on the electrode surface and subsequent decrement in the electrode surface area by monitoring the reduction process of the Fe(CN)63−/Fe(CN)64− redox couple. Under the optimized experimental condition, the linear range and the detection limit of PSA immunosensor were determined to be 1.0 pg/mL to 10.0 ng/mL and 0.40 ± 0.03 pg/mL, respectively. The validity of the proposed method was compared with an enzyme-linked immunosorbent assay method in various PSA spiked human serum samples.

T

conducting polymer layer has been recently reported for the fabrication of a sensitive horseradish peroxidase (HRP)-based electrochemical biosensor.21 Amplification through the enzymatic precipitation of a nonconductive insoluble product has also been used for the development of biosensors.22,23 In this particular strategy, an insoluble product precipitates by an enzyme horseradish peroxidase (HRP) (biocatalyzed precipitation) and makes an insulating layer on an electrode surface, which completely blocks the electron transfer process of a redox probe (such as ferricyanide ion). The extent of electrode insulation was monitored by Faradaic impedance spectroscopy and cyclic voltammetry. This biocatalyzed precipitation has also been successfully applied to the fabrication of voltammetric antibiotin IgG antibody immunosensor24 with a biotin-HRP conjugate, where a biotin molecule is conjugated with an HRP molecule (single-HRP strategy). However, the dynamic detection range (0.1−100 μg/mL) is not satisfactory for most of the proteins detection whose concentrations are normally less than 1 ng/mL. Thus, a more sensitive amplification strategy is needed for the biocatalyzed precipitation method to be practical. In this work, a novel amplification strategy using a higher number of HRP molecules induced biocatalyzed precipitation

he sensitive detection of cancer biomarkers play an important role in early detection, monitoring disease recurrence, and therapeutic treatment efficacy to improve longterm survival of cancer patients.1 Conventional detection methods for cancer biomarkers are enzyme-linked immunosorbent assay (ELISA),2 radioimmunoassay (RIA),3 fluorescence,4 chemiluminescence,5 electrophoretic,6 and mass spectrometric immunoassays.7 However, these conventional immunoassays method are complicated, time-consuming, tedious, expensive, and labor-intensive. Moreover, these methods are not suitable for point-of-care applications. Thus, it is urgently needed to develop a highly sensitive, less expensive, and fast detection method for the point-of-care applications. As an alternative to the conventional immunoassay procedures, the electrochemical immunosensors8,9 received a particular attention because of its simple instrumentation, fast response time, and portability. One of the critical issues for the sensitive electrochemical immunosensors is signal amplification. Signal amplification strategies reported so far include metal and semiconductor nanomaterials label,10,11 catalytically deposited nanoparticles,12,13 metal nanocarriers,14,15 muiltiple enzyme particles,8,16 dual amplification,17,18 catalytic chemical process,19 and metal-cysteamine complex.20 On the other hand, enzymatic reactions are able to form polymeric layers on the electrode surface and this strategy can be used for the signal amplification. For example, an enzymatically synthesized © XXXX American Chemical Society

Received: January 12, 2012 Accepted: June 30, 2012

A

dx.doi.org/10.1021/ac300110n | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

Figure 1. Schematic illustrations of the single-HRP and multiple-HRP strategy-based PSA immunosensor.

(multiple-HRP strategy) that decreases the electroactive area of the electrode; thus, an increase in the current decrement of the redox probe is used. This amplification idea is applied for the detection of a model protein, prostate specific antigen (PSA, a 28-kDa serine protease), which is a biomarker for the screening and early detection of prostate cancer.25 PSA lacks specificity as it can be elevated in men with cancer as well as in men with benign prostate conditions. The normal range of PSA level is regarded to be 0 to 4 ng/mL.26 However, there is a nonnegligible risk of prostate cancer at any PSA level. For example, A PSA level of 0.6 to 1.0 ng/mL gives a 10.1% risk of developing prostate cancer. Higher PSA levels are directly associated with the increased risk of prostate cancer: a level of 4.1 to 10 ng/mL yields a 58.2% risk.27 Thus, a highly sensitive method (less than ng/mL) is essential for PSA detection because real samples are generally 10 to 100 times diluted before analysis. Figure 1 describes the new amplification strategy of biocatalyzed precipitation, the multiple enzyme label method using CNTs, and gold nanoparticles (AuNPs). In Steps 1 and 2, a Au electrode surface is coated with multiwall carbon nanotubes (MWCNTs; Au/MWCNT), and then, AuNPs are deposited on the MWCNT surfaces (Au/MWCNT/AuNP). The sequential coating of MWCNTs and AuNPs increases the area of Au surface on which antibody will be immobilized. Monoclonal anti-PSA (Ab1) antibody is then immobilized onto the AuNPs of Au/MWCNTs/AuNPs (Au/MWCNTs/AuNPs/ anti-PSA) (Step 3). Thus, an immunosensor probe for PSA is fabricated. In Step 4, the target PSA proteins interact with the antibodies on the surface of the Au/MWCNTs/AuNPs/antiPSA probe. Two methods are able to be applied in the detection of PSA: single and multiple HRP routes. In the

single-HRP route (Steps 5 and 6), secondary antibodies (polyclonal anti-PSA, Ab2) conjugated with a single HRP molecule interact with the attached PSA proteins and the HRP induces conversion of 4-chloro-1-naphthol (CN) to benzo-4chlorocyclohexadienone in the presence of hydrogen peroxide generated insoluble precipitates, which formed a blue-colored insulating film24 on the immunosensor surface. In the multipleHRP route (Steps 7 and 8), on the other hand, more HRP molecules conjugated with Ab2 through MWCNTs produces more precipitates. Therefore, the multiple-HRP route results in a more decrement of the electrode surface area than the singleHRP route does, thus enhancing the current reduction of the redox probe. The immunosensors fabricated along the multipleHRP route were characterized using scanning electron microscope (SEM), transmission electron microscope (TEM), quartz crystal microbalance (QCM), and cyclic voltammetric (CV) techniques. CV and square wave voltammetric (SWV) techniques were used for measuring the current response changes of a redox probe before and after the precipitation. The linear dynamic range and the detection limit were determined after optimizing various experimental parameters. Besides, the selectivity, stability, and the regeneration of the immunosensor surface were studied. It was applied to the PSA spiked human serum sample for the detection of PSA, and the results were compared with a conventional ELISA method.

■

EXPERIMENTAL SECTION Reagents. N-Hydroxysuccinimide (NHS), 1-ethyl-3 (3(dimethylamino)-propyl) carbodiimide (EDC), hydrogen peroxide, 4-chloro-1-naphthol, peroxidase from horseradish (E.C. 232-668-6, Type VI, 250-330 units/mg), glutaraldehyde, B

dx.doi.org/10.1021/ac300110n | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

of ice-cold 0.0035 wt % NaBH4 solution was slowly added into the solution mixture being stirred to produce AuNPs. Fabrication of the Immunosensor Probe and the Sensing Procedure. At first, Au/MWCNTs electrodes were prepared through the interaction between thiol groups and the Au electrode by dipping a Au electrode for 18 h in an ethanol solution of SH-MWCNT (Step 1, Figure 1). Prior to the modification, polycrystalline gold electrodes (3 mm in diameter) 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 +1.4 V and −0.2 V at 100 mV/s for 50 times in a 1.0 M H2SO4 solution. After covering the remaining empty sites of the Au electrode with mercaptoethanol, AuNPs were then chemically adsorbed on the Au/MWCNTs electrode by immersing the electrode in AuNP solution for 18 h (Step 2, Figure 1). By this step, AuNPs were attached through the interaction of thiol groups of the other side of MWCNTs and AuNPs. The AuNP modified electrodes (Au/MWCNTs/AuNPs) were rinsed with water and dried by nitrogen gas. Then, monoclonal anti-PSA antibody (Ab1) was immobilized onto the Au/MWCNTs/AuNPs electrode by incubating the electrode in a 0.1 M phosphate PBS solution (pH 7.0) containing 0.1 mg/mL Ab1 for 24 h at 4 °C. Ab1 was immobilized on the Au/MWCNTs/AuNPs through the interaction between amine groups of Ab1 and AuNPs. After washing three times with 0.1 M PBS, the Au/MWCNTs/ AuNPs/Ab1 electrodes (immunosensors) were blocked by dipping them in 0.1% BSA solution for 2 h at 4 °C. In our previous work,33 we found that blocking the immunosensor surface with 0.1% bovine serum albumin (BSA) solution was effective to minimize the NSB events. Thus, in the present study, we also used BSA blocking for minimizing nonspecific binding events. After washing three or four times with PBS and drying with N2, the blocked immunosensors were incubated with various concentrations of PSA antigen (Ag) at 37 °C. After the immunointeraction between Ab1 and Ag, the immunosensors were immersed into Ab2-HRP and Ab2-MWCNTHRPs bioconjugates for 1 h followed by washing three or four times with PBS to remove nonspecifically bound bioconjugates. Then, the biocatalyzed precipitation of benzo-4-chlorohexadienone at the immunosensor surface was followed by dipping the electrode for a specified time in a PBS solution containing 1.0 mM CN and 1.0 mM H2O2. After thoroughly washing with 0.1 M PBS, the immunosensors were placed in a detection solution of 0.1 M PBS (pH 7.0) containing 0.1 mM of potassium ferricyanide. SWV measurements of the reduction process of 0.1 mM ferricyanide ions were then monitored for the quantification of the PSA. Preparation of Secondary Antibody (Ab2)-Single-HRP Conjugates. The secondary antibody (Ab2) conjugated singleHRP (Ab2-single-HRP) conjugate was prepared according to a previous report.34 Briefly, 0.5 mg/mL of HRP in 0.1 M PBS and 0.5 mL of glutaraldehyde (25%) were mixed at room temperature for 18 h. Excess glutaraldehyde was removed by passing the solution through a sephadex G-25 column equilibrated with 0.9% NaCl. Five μg/mL of Ab2 in 0.5 M sodium carbonate buffer (pH 9.5) was added to the HRP solution, and the mixture was incubated for 24 h at 4 °C with stirring. The remaining active sites of Ab2 were blocked with a

ferricyanide, sodium chloride, sodium citrate, sodium borohydride (NaBH4), sephadex G-25, sodium carbonate, sulfuric acid, nitric acid, cysteamine, glycine, hydrogen tetrachloroaurate (HAuCl4·H2O), immunoglobulin G (IgG), human serum albumin, bovine serum albumin (produced in mouse), carcinoembryonic antigen, prostate specific antigen, monoclonal anti-PSA antibody (produced in mouse, clone 1B1), and polyclonal anti-PSA (produced in rabbit) antibodies were obtained from Sigma Co. Human α-thrombin (specific activity: 3725 units/mg; concentration: 9.7 mg/mL) was obtained from Haematologic Technologies Inc. The multiwall carbon nanotubes (MWCNTs) were obtained from JEIO Co. Korea. The phosphate buffer saline solution (PBS) was prepared by mixing 10 mM NaH2PO4 and 10 mM Na2HPO4 with 0.9% sodium chloride (NaCl). All other chemicals were of extra pure analytical grade and used without further purification. All aqueous solutions were prepared with deionized distilled water obtained from a Milli-Q water purifying system (18 MΩcm). Instruments. Cyclic voltammograms (CV) and square wave voltammograms (SWV) were recorded using a model 430B potentiostat/galvanostat (CH Instruments Inc. USA). In CV and SWV experiments, Ag/AgCl (in saturated KCl) and a platinum (Pt) wire were used as reference and counter electrodes, respectively. In SWV, the potential was scanned from +0.4 to 0 V with 5 mV pulse height, 25 mV amplitude, and 60 Hz frequency. Scanning electron microscopic (SEM) and transmission electron microscopic (TEM) images were obtained using a field emission scanning electron microscope (FE-SEM, Model JSM-7000F JEOL) and a field-emission transmission electron microscope (JEM-2100F, JEOL), respectively. Quartz crystal microbalance (QCM) experiments were performed at an 8 MHz AT-cut quartz crystal (area, 0.205 cm2). The mass adsorbed was calculated from the following relationship28 assuming that Δf is affected by only mass change of a film: Δm = −C Δf

(1)

where C is the sensitivity or conversion factor. The sensitivity factor was obtained by plotting the Δf from the EQCM experiments and Δm, which was calculated from the voltammetric deposition of copper from a copper nitrate solution by integrating the cathodic peak and using the Faraday’s Law. From the slope of the Δm vs Δf plot, the sensitivity factor was calculated as 2.5 ng/Hz. Thiolation of MWCNTs and Preparation of AuNPs. MWCNTs were thiolated according to previous reports.29,30 Prior to thiolation, MWCNTs were shortened through chemical oxidation by treating MWCNTs in a 3:1 mixture (v/v) of sulfuric and nitric acids at 60 °C for 12 h. The shortened MWCNTs were filtered and washed repeatedly with water until the pH reached 7.0. The shortening process removed metallic and carbonaceous impurities and generated carboxylic acid groups at the surfaces of MWCNTs (hereafter designated as COOH-MWCNTs). The COOH-MWCNTs were further modified with thiol (SH) groups by the formation of covalent bonds between COOH groups of MWCNTs and amine (NH2) groups of cysteamine (NH2-(CH2)2-SH). The SH groups modified MWCNTs are hereafter designated as SHMWCNTs. Gold nanoparticles (AuNP) were prepared according to the previous reports.31,32 Briefly, 800 mL of 1 wt % sodium citrate solution was mixed with 800 mL of a solution containing 0.0666 g of HAuCl4·H2O. A 10 mL aliquot C

dx.doi.org/10.1021/ac300110n | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

Figure 2. TEM images of (a) COOH-MWCNTs, (b) SH-MWCNTs, and (c) AuNPs and SEM images of (d) MWCNTs and (e) MWCNTs/ AuNPs.

edly washed for 3−4 times with PBS buffer and were blocked with a 0.1% (w/v) BSA solution for 1 h. After centrifugation at 15 000 rpm for 5 min, the supernatant was discarded. Finally, 1.0 mL of 0.1 M PBS was added to the bioconjugate residues and stored in a refrigerator at 4 °C.

0.1 mL of 0.1% (w/v) BSA solution and dialyzed against 0.1 M PBS (pH 7.4) at 4 °C for 24 h with 4 changes of PBS. Finally, the conjugate was filtered through a sterile Millipore membrane (0.20 μm), and the filtrate was stored at −20 °C until use. Preparation of Secondary Antibody (Ab2)-MultipleHRP Conjugates. The secondary antibody (Ab2) conjugated multiple-HRPs (Ab2-multiple-HRPs) were prepared using COOH-MWCNTs as a bioconjugation platform according to a previous report16 with a slight modification. The COOHMWCNTs were first mixed with a 0.1 M PBS solution (pH 7.0) containing 10 mM EDC and 10 mM NHS for 6 h at room temperature.35 The resulting mixture was centrifuged at 15 000 rpm for 5 min. After discarding the supernatant, the activated COOH-MWCNTs residue was washed repeatedly (3 times) with PBS buffer in order to remove excess EDC and NHS. Ab2 (5.0 μg/mL) and horseradish peroxidase (HRP, 0.5 mg/mL) in 0.1 M PBS were added to the activated COOH-MWCNTs and stirred for 24 h at 4 °C. The reaction mixture was then centrifuged at 15 000 rpm for 5 min, and the supernatant was discarded. The Ab2/MWCNTs/HRPs conjugates were repeat-

■

RESULTS AND DISCUSSION Characterization of the Functionalized MWCNTs, AuNPs, and the Immunosensor Probe. The functionalized MWCNTs (COOH-MWCNTs, SH-MWCNTs) and AuNPs were characterized using field-emission TEM and are shown in Figure 2. Figure 2a,b shows the TEM images obtained for COOH-MWCNTs and SH-MWCNTs, respectively. The lengths of the COOH-MWCNTs obtained after the shortening process of MWCNTs were similar to that obtained in our previous study.29 Upon thiolation, some of the MWCNTs were found to be aggregated (Figure 2b). This might be due to the formation of disulfide bonds between SH-MWCNTs. Figure 2c shows a TEM image of the prepared AuNPs, where most of the D

dx.doi.org/10.1021/ac300110n | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

indicating the completion of the Ab1-Ag interaction. The Δf was 17 ± 3 Hz, which corresponds to a Δm of 42.5 ± 7.5 ng. Using the molecular weight of PSA (28 000 Da), the surface coverage of PSA was calculated to be 1.5 ± 0.3 × 10−12 mol/ cm2. Because the number of Ab1 is close enough to that of Ag, it becomes clear that Ag combines with Ab1 quantitatively. The interactions of Ab2-single-HRP and Ab2-multiple-HRP conjugates at the Au/MWCNTS/AuNPs/Ab1/Ag were also studied using QCM experiments. During the interactions of Ab2-single-HRP (line (3) in Figure 3a), the Δf was found to be 33.3 ± 6 Hz, which corresponds to a Δm value of 83.3 ± 15 ng. On the other hand, the Δf and Δm values were determined to be 90 ± 6.5 Hz and 225 ± 16.25 ng, respectively, for Ab2multiple-HRPs (line (4) in Figure 3a). The increased Δf and Δm values observed for the Ab2-multiple-HRP conjugate was due to the presence of MWCNTs and higher number of HRP molecules. The above QCM results confirmed the successful immobilization of Ab1 on the Au/MWCNTS/AuNPs and stable bindings of Ag with Ab2-single-HRP and Ab2-multipleHRP bioconjugates on the Au/MWCNTS/AuNPs/Ab1 probe. The nonspecific binding of other proteins on the immunosensor surface was also examined using QCM experiments. Figure 3b shows the time-dependent frequency responses of the immunosensor probe in PSA spiked human serum samples (1) without or (2) with BSA blocking. It is very clear that other proteins/matrixes can be nonspecifically adsorbed on the immunosensor surface. However, after blocking the immunosensor surface by BSA, the nonspecific binding of the sensor probe was completely minimized. Voltammetric Responses of the PSA Immunosensor. Figure 4a shows the cyclic voltammetric responses recorded for PSA immunosensors in a 0.1 M PBS solution (1) without or

AuNPs are well separated from each other. The average particle sizes of AuNPs were determined to be about 4.0 ± 1.1 nm. The surface of the Au/MWCNTs/AuNPs electrode was characterized using SEM experiments. Figure 2d shows a SEM image obtained after dipping the precleaned Au electrode in an ethanol solution of SH-MWCNTs for 16 h. SH-MWCNTs were chemically adsorbed on the Au electrodes through the interaction of SH groups of MWCNTs and Au. Figure 2e shows a SEM image obtained after dipping the Au/MWCNTs electrode in an aqueous solution of AuNPs for 2 h. The SEM image revealed that AuNPs were chemically adsorbed on the surface of Au/MWCNTs. The QCM technique was performed for the confirmation of the immobilization of monoclonal anti-PSA antibody (Ab1) on the Au/MWCNTs/AuNPs electrode. For this purpose, the Au quartz crystal electrode was coated with SH-MWCNTs by dipping a QCM electrode into an ethanol solution of SHMWCNTs for 18 h at the room temperature. The MWCNTs/ AuNPs coated QCM electrode was then placed in the QCM cell containing a 1000 times diluted (2 ng/mL) Ab1 solution in PBS (pH 7.0). The frequency changes during the immobilization of Ab1 onto the Au/MWCNTs/AuNPs electrode are shown in Figure 3a (line (1)). The frequency decrease reached

Figure 3. (a) Time-dependent frequency changes during (1) monoclonal anti-PSA antibody (Ab1) immobilization, (2) Ab1-Ag interaction, (3) Ag-Ab2/HRP, and (4) Ag-Ab2/MWCNTs/HRPs. (b) Time-dependent frequency responses of the immunosensor probe in PSA spiked human serum samples (1) without or (2) with BSA blocking.

to a steady state within 2 h, which indicates that immobilization of Ab1 was completed within 2 h. After a 2 h immobilization, the frequency change (Δf) was found to be about 81 ± 5 Hz corresponding to a mass change (Δm) of 202.5 ± 12.5 ng. Assuming the anti-PSA antibody has similar molecular weight of anti-IgG antibodies (150 000 Da), the surface coverage of Ab1 was calculated as 1.35 ± 0.08 × 10−12 mol/cm2. The interaction of PSA (Ag) at the Au/MWCNTS/AuNPs/ Ab1 electrode was characterized using the QCM technique (line (2) in Figure 3a). During the interaction between Ab1 and Ag, the Δf decreased and reached to a steady state within 2 h,

Figure 4. (a) CVs recorded for a PSA immunosensor in PBS (1) without or (2) with 1.0 mM Fe(CN)63− and after biocatalytic precipitation reaction with (3) single-HRP and (4) multiple-HRP bioconjugates. (B) SWVs recorded for a PSA immunosensor in PBS (1) without or (2) with 0.1 mM Fe(CN)63− and after biocatalytic precipitation reaction with (3) single-HRP and (4) multiple-HRP bioconjugates. The experimental conditions were as follows: [CN] = 1.0 mM; [H2O2] = 1.0 mM; precipitation time = 10 min, pH = 7.5. E

dx.doi.org/10.1021/ac300110n | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

(2) with 1.0 mM ferricyanide ion (Fe(CN)63−). A well-defined redox peak was observed, indicating that the modified electrode had efficient electron transfer ability. The peak potential separation (ΔEp) between the anodic and cathodic peaks was determined to be 0.09 V. The peak currents were proportional to the square root of scan rates, indicating that the electron transfer was controlled by the diffusion process of Fe(CN)63−. The Au/MWCNTs/AuNPs/Anti-PSA electrode was subjected to antigen (1 ng/mL) binding, BSA blocking, and Ab2-singleHRP and Ab2-multiple-HRP bioconjugate association. The modified electrode was then used for CV measurements after (3) single-HRP and (4) multiple-HRP biocatalyzed precipitation reactions for 10 min. The ΔEps were determined to be 0.14 and 0.17 V for single-HRP and multiple-HRPs, respectively. The peak currents decreased for both singleHRP and multiple-HRPs. This might be due to the fact that the precipitate generated by the labeled HRP accumulated on the immunosensor surface decreased the electrode surface area. However, the magnitude of the current decrement is much higher for multiple-HRPs due to the generation of more precipitates by the higher number of HRP molecules attached to Ab2 through MWCNTs. Thus, the multiple-HRP strategy was used for the analytical detection of PSA. The difference in current responses (ΔI) before and after the precipitation reaction was taken as analytical signal for PSA detection. Figure 4b shows the square wave voltammograms (SWVs) recorded for PSA immunosensors in a 0.1 M PBS solution (1) without or (2) with 0.1 mM Fe(CN)63−. The SWVs showed a reduction peak at about 0.2 V. After the biocatalyzed precipitation with (3) single-HRP and (4) multiple-HRP strategies, the SWV peak current at 0.2 V was decreased and the magnitude of the current decrement was much higher in the case of multiple-HRP. These results further confirmed that the amount of precipitation that accumulated on the sensor surface was much higher in the multiple-HRP biocatalyzed precipitation. Both CV and SWV studies clearly revealed that the immunosensor response could be more amplified through the enhanced precipitation by the higher number of HRP molecules attached to Ab2 through MWCNT. The various experimental parameters such as Ab1 antibody dilution factor, Ab1-Ag interaction time, pH of the Ab1-Ag binding medium, etc. for the detection of PSA with multipleHRP strategy were optimized. The effect of Ab1 dilution factor on the Fe(CN)63− reduction was examined between 1:10 000 and 1:100 (Figure 5a). The ΔI response gradually increased from 1:10 000 to 1:1000. Over the dilution factor of 1:1000, the ΔI response did not change. Thus, the optimum anti-PSA antibody dilution factor was optimized as 1:1000. The effect of anti-PSA antibody and PSA interaction time was investigated between 2 and 40 min (Figure 5b). The ΔI response gradually increased from 2 to 20 min. Over 20 min of interaction time, the ΔI response did not significantly change due to the saturation effect. The maximum response was found at the interaction time of 20 min. Thus, the optimum interaction time was considered as 20 min. The effect of pH of the Ab1-Ag binding medium was studied over the pH range of 5−9 (Figure 5c). The ΔI response gradually increased from pH 5 to 7.5 and then rapidly decreased at pH values higher than 8. The ΔI responses between pH 7 and 8 did not change significantly. The decrease in the ΔI responses below 6 and over pH 8 might be due to the weak immunointeraction at lower and higher pHs. The

Figure 5. Effects of (a) anti-PSA dilution, (b) Ab1-Ag interaction time, and (c) the pH of the medium of Ab1-Ag binding on the SWV responses of the multiple-HRP-based immunosensor after the biocatalytic precipitation reaction of 1.0 mM CN and H2O2 for 10 min. Other experimental conditions were the same as Figure 4.

Figure 6. (a) Square wave voltammograms recorded for multipleHRP-based immunosensors at various PSA concentrations before (1) and after (2−l2) precipitation: (2) 0.001, (3) 0.005, (4) 0.01, (5) 0.05, (6) 0.1, (7) 0.5, (8) 1.0, (9) 5.0, (10) 10, (11) 20, and (12) 30 ng/mL. Experimental conditions were the same as in Figure 4. (b) A calibration plot for the PSA detection. Insets showed the calibration plots at a low concentration range.

maximum reduction current was observed at the pH of 7.5. Thus, the optimum pH was chosen as 7.5. Analytical Performance of the PSA Immunosensor. The SW current responses were measured by varying the PSA concentration with the multiple-HRP strategy as shown in Figure 6a. The reduction current responses of Fe(CN)63− were inversely proportional to the PSA concentration. The calibration plot was constructed by plotting the ΔI responses against logarithm of PSA concentrations and is shown in Figure F

dx.doi.org/10.1021/ac300110n | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

on three measurements of the standard deviation of the blank noise (95% confidence level, k = 3, n = 5). The observed dynamic range is much wider than a carbon nanotube (0.4 to 40 ng/mL)8 or AuNPs (1.0 to 40 ng/mL)16 amplificationbased PSA immunosensors. The detection limit is also lower than previously reported values obtained from a AuNP filmbased immunosensor (0.5 pg/mL),8 CNTs amplified immunodetection (4 pg/mL),16 enzyme-linked immunosorbent assay (3 pg/L),36 and Si-nanowire transistors-based sensor (0.9 pg/ mL).37 Selectivity and Stability of the PSA Immunosensor. In order to assess the possibility of interference, the SWV responses were measured for other common proteins such as human serum albumin (HSA), human immunoglobin G (IgG), and human thrombin (TB) which coexist with PSA. Figure 7a shows the SWV responses of the multiple-HRP-based PSA immunosensor (1) before and (2−5) after biocatalyzed precipitation using (2) HSA, (3) IgG, (4) TB, and (5) PSA at the concentration of 1.0 ng/mL. As shown in the Figure 7a, no significant current decrements were observed for HSA, IgG, and TB, indicating the above proteins did not interfere in the PSA detection. The stability of the multiple-HRP-based PSA immunosensor was determined by measuring the response of 1.0 ng/mL PSA twice a day for one month. After each measurement, the immunosensor was washed with a PBS solution and dipped into a 0.2 M glycine-hydrochloric acid (Gly-HCl) buffer solution (pH = 2.8) for 5 min. After three times washing with a PBS solution, the immunosensor was stored in a dry condition at 4 °C until it was used for further measurement. The ΔI response was not found to change significantly for one month (Figure 7b). The immunosensor retained almost 94% of its initial response for the one month time period. After one month, the ΔI response gradually decreased due to the gradual decrease in the activity of anti-PSA, PSA, and HRP. These

Figure 7. (a) Square wave voltammograms recorded for multipleHRP-based immunosensors (1) before and (2−5) after the precipitation reaction using 1.0 ng/mL of each (2) human HSA (red), (3) human IgG (blue), (4) human TB (green), and (5) 0.5 ng/ mL of PSA (orange). (b) Long-term stability of the multiple-HRPbased PSA immunosensor. Experimental conditions were the same as in Figure 4.

6b. Under the optimized condition, the immunosensor exhibits a wide dynamic range between 1.0 pg/mL and 10.0 ng/mL. The calibration plot at lower concentration ranges is presented in the inset of Figure 6b. The reproducibility expressed in terms of the relative standard deviation (RSD) was about 4.2% (n = 5) at a PSA concentration of 1.0 ng/mL. The detection limit of PSA was determined to be 0.4 ± 0.03 pg/mL, which was based

Figure 8. Real sample analysis of PSA spiked human serum samples (a) 1 ng/mL, (b) 2 ng/mL, and (c) 5 ng/mL using the standard addition method; (1) 0, (2) 0.5, (3) 1.0, (4) 1.5, and (5) 2.0 ng/mL additions of standard PSA. Insets showed the background subtracted SWV responses (1−5) after and (6) before biocatalytic precipitation. (d) Corresponding standard addition plots. Experimental conditions were the same as in Figure 4. G

dx.doi.org/10.1021/ac300110n | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

Table 1. Spike and Recovery Results (n = 3) Obtained from the PSA Immunosensor and ELISA Methods PSA found (ng/mL)

RSD (%)

immunosensor

ELISA

immunosensor

ELISA

immunosensor

ELISA

1 2 5

0.96 ± 0.04 1.97 ± 0.08 5.10 ± 0.30

0.99 ± 0.04 1.95 ± 0.06 4.9 ± 0.20

4.2 4.1 5.9

4.0 3.1 4.1

96.0 99.0 102.0

99.0 98.0 98.0

results indicated that the PSA immunosensor exhibited not only high selectivity but also long-term stability. Real Sample Analysis. To examine the applicability of the proposed PSA immunosensor in human serum sample, we performed spike and recovery experiments. The serum sample was spiked with 1.0, 2.0, and 5.0 ng/mL PSA. Figure 8a−c shows the SWV responses obtained in various PSA spiked serum samples. The ΔI response obtained in serum samples was lower than that obtained in PBS. This might be due to the matrix effect of the serum samples. To minimize the matrix effect, the standard addition method was followed for PSA quantification by adding four PSA standard solutions. Figure 8d shows the standard addition plots, from where the PSA concentrations were determined by extrapolating the linear lines to the negative x-axis. The recovery results were listed in Table 1, which showed acceptable results with RSD values ranging between 3.4% and 4.5%. The PSA recovery was between 96% and 103%, which clearly indicated the potentiality of the present multiple-HRP-based PSA immunosensor for PSA detection in real human serum samples. In addition to immunosensor experiments, the PSA spiked samples were also analyzed using a conventional ELISA method. The results are also presented in Table 1. A good agreement between the immunosensor and ELISA results was found, indicating the validity of the proposed PSA immunosensor in real biological samples.

■

ACKNOWLEDGMENTS

■

REFERENCES

This research was financially supported by a Research Fund of Chungnam National University in 2010 (Grant No. 20100721) and the Basic Research Program (2011-0004814) through a National Research Foundation of Korea grant funded by the MEST.

(1) Kulasingam, V.; Diamindis, E. P. Nat. Clin. Pract. Oncol. 2008, 5, 588−599. (2) Yates, A. M.; Elvin, S. J.; Williamson, D. E. J. Immunoassay 1999, 20, 31−44. (3) Teppo, A. M.; Maury, C. P. J. Clin. Chem. 1987, 33, 2024−2027. (4) Matsuya, T.; Tashiro, S.; Hoshino, N.; Shibata, N.; Nagasaki, Y.; Kataoka, K. Anal. Chem. 2003, 75, 6124−6132. (5) Fu, Z. F.; Yan, F.; Liu, H.; Yang, Z. J.; Ju, H. X. Biosens. Bioelectron. 2008, 23, 1063−1069. (6) Schmalzing, D.; Nashabeh, W. Electrophoresis 1997, 18, 2184− 2193. (7) Hu, S. H.; Zhang, S. C.; Hu, Z. C.; Xing, Z.; Zhang, X. R. Anal. Chem. 2007, 79, 923−929. (8) Mani, V.; Chikkaveeraiah, B. V.; Patel, V.; Gutkind, S. G.; Rusling, J. F. ACS Nano 2009, 3, 585−594. (9) Shiddiky, M. J. A.; Rahman, M. A.; Shim, Y.-B. Anal. Chem. 2007, 79, 6886−6890. (10) Liu, G. D.; Lin, Y. H. J. Am. Chem. Soc. 2007, 129, 10394− 10401. (11) Wang, J.; Xu, D.; Kwade, A.-N.; Polsky, R. Anal. Chem. 2001, 73, 5576−5581. (12) Hwang, G.; Kim, E.; Kwak, J. Anal. Chem. 2005, 77, 579−584. (13) Lai, G.; Yan, F.; Wu, J.; Leng, C.; Ju, H. Anal. Chem. 2011, 83, 2726−2732. (14) Wang, J.; Liu., G. D.; Wu, H.; Lin, Y. H. Anal. Chim. Acta 2008, 610, 112−118. (15) Jie, G.; Huang, H.; Sun, X.; Zhu, J. J. Biosens. Bioelectron. 2008, 23, 1896−1899. (16) 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. J. Am. Chem. Soc. 2006, 128, 11199−11205. (17) Wang, G.; Huang, H.; Zhang, G.; Zhang, X.; Fang., B.; Wang, L. Langmuir 2011, 27, 1224−1231. (18) Du, D.; Zou, Z.; Shin, Y.; Wang, J.; Wu, H.; Engelhard, M. H.; Liu, J.; Aksay, I. A.; Lin, Y. Anal. Chem. 2010, 82, 2989−2995. (19) Das, J.; Aziz, M. A.; Yang, H. J. Am. Chem. Soc. 2006, 128, 16022−16023. (20) Noh, H.-B.; Rahman, M. A.; Yang, J. E.; Shim, Y.-B. Biosens. Bioelectron. 2011, 26, 4429−4435. (21) Kausaite-Minkstimiene, A.; Mazeiko, V.; Ramanaviciene, A.; Ramanaviciene, A. Biosens. Bioelectron. 2010, 26, 790−797. (22) Patolsky, F.; Zayats, M.; Katz, E.; Willner, I. Anal. Chem. 1999, 71, 3171−3180. (23) Alfonta, L.; Bardea, A.; Khersonsky, O.; Katz, E.; Willner, I. Biosens. Bioelectron. 2001, 16, 675−687. (24) Yoon, H. C.; Yang, H.; Kim, Y. T. Analyst 2002, 127, 1082− 1087. (25) Gretzer, M. B.; Partin, A. W. Urol. Clin. North. Am. 2003, 30, 677−86.

■

CONCLUSIONS An amplified voltammetric detection of PSA was developed with a MWCNTs/AuNPs-based PSA immunosensor employing a multiple-HRP strategy. The signal amplification was achieved through the generation of more precipitate by the higher number of HRP molecules that accumulated on the electrode surface and decreased the electrode surface area. Thus, more current decrement was observed before and after precipitation even for a low concentration of PSA change. The linear dynamic range and the detection limit were determined to be 1.0 pg/mL to 10.0 ng/mL and 0.4 ± 0.03 pg/mL, respectively. The immunosensor showed a good selectivity and long-term stability. The proposed immunosensor can be applied for detecting other proteins by simply changing the respective antibodies and has the potential future for reliable point-of-care cancer diagnostics.

■

recovery (%)

PSA added (ng/mL)

AUTHOR INFORMATION

Corresponding Author

*Phone: 82-42-821-8546 (M.A.R.); 82-42-821-5483 (C.K.R.). Fax: 82-42-821-8541 (M.A.R.); 82-42-821-8896 (C.K.R.). Email: [email protected] (M.A.R.); [email protected] (C.K.R.). Notes

The authors declare no competing financial interest. H

dx.doi.org/10.1021/ac300110n | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

(26) Catalona, W. J.; Smith, D. S.; Ratliff, T. L.; Dodds, K. M.; Coplen, D. E.; Yuan, J. J.; Petros, J. A.; Andriole, G. L. N. Engl. J. Med. 1991, 324, 1156−1161. (27) Thompson, I. M.; Pauler, D. K.; Goodman, P. J.; Tangen, C. M.; Lucia, M. S.; Parnes, H. L.; Minasian, L. M.; Ford, l. G.; Lippman, S. M.; Crawford, E. D.; Crowley, J. J.; Coltman, C. A., Jr. N. Engl. J. Med. 2004, 350, 2239−2246. (28) Buttry, D. A. In Electroanalytical Chemistry; Bard., A. J., Ed.; Marcel Dekker: New York, 1991; Vol. 17, pp 1−86. (29) Park, S.; Kim, H. R.; Kim, J.; Jung, C.; Rhee, C. K.; Kwon, K.; Kim, Y. Carbon 2011, 49, 487−494. (30) Konya, Z.; Zhu, J.; Niessz, K.; Mehn, D.; Kiricsi, I. Carbon 2004, 42, 2001−2008. (31) Kim, S.; Jung, C; Kim, J.; Rhee, C. K.; Choi, S.-M.; Lim, T.-H. Langmuir 2010, 26, 4497−4505. (32) Hayat, M. A. Colloidal Gold: Principles, Methods, and Applications; Academic Press: San Diego, CA, 1989. (33) Noh, H.-B.; Rahman, M. A.; Yang, J. E.; Shim, Y.-B. Biosens. Bioelectron. 2011, 26, 4429−4435. (34) Tijssen, P. Practice and Theory of enzyme Immunoassay. In Laboratory Techniques in Biochemistry and Molecular Biology; Elsevier, Amsterdam, 1985; Vol. 15. (35) Rahman, M. M.; Shiddiky, M. J. A.; Rahman, M. A.; Shim, Y.-B. Anal. Biochem. 2009, 384, 159−165. (36) Ward, M. A.; Catto, J. W. F.; Hamdy, F. C. Ann. Clin. Biochem. 2001, 38, 633−651. (37) Zheng, G.; Patolsky, F.; Cui, Y.; Wang, W. U.; Leiber, C. M. Nat. Biotechnol. 2005, 23, 1294−1301.

I

dx.doi.org/10.1021/ac300110n | Anal. Chem. XXXX, XXX, XXX−XXX