Analytica Chimica Acta 818 (2014) 46–53
Contents lists available at ScienceDirect
Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca
Application of ZnO quantum dots dotted carbon nanotube for sensitive electrochemiluminescence immunoassay based on simply electrochemical reduced Pt/Au alloy and a disposable device Fang Liu a , Wenping Deng a , Yan Zhang a , Shenguang Ge b , Jinghua Yu a,∗ , Xianrang Song c,∗ a Key Laboratory of Chemical Sensing & Analysis in Universities of Shandong, School of Chemistry and Chemical Engineering, University of Jinan, Jinan, 250022, PR China b Shandong Provincial Key Laboratory of Preparation and Measurement of Building Materials, University of Jinan, Jinan, 250022, PR China c Cancer Research Center, Shandong Tumor Hospital, Jinan, 250012, PR China
h i g h l i g h t s
g r a p h i c a l
a b s t r a c t
• A sandwich-type electrochemluminence immunosensor was fabricated. electrochemical reduced Pt/Au alloy was selected as immunosensing probes. • ZnO@CNT composite was first employed as signal amplification label.
• Simply
a r t i c l e
i n f o
Article history: Received 6 September 2013 Received in revised form 17 December 2013 Accepted 22 January 2014 Available online 4 February 2014 Keywords: ZnO quantum dots dotted carbon nanotube Pt/Au alloy Screen-printed electrode Electrochemiluminescence Immunosensor
a b s t r a c t We report on a disposable microdevice suitable for sandwich-type electrochemiluminescence (ECL) detection of prostate specific antigen (PSA). The method is making use of ZnO quantum dots dotted carbon nanotube (ZnO@CNT) and simply electrochemical reduced Pt/Au alloy. The latter was selected as immunosensing probe to modify screen-printed carbon electrode, due to its excellent electrical property. For further ultrasensitive, low-potential and stable ECL detection, ZnO@CNT composite was first synthesized using a facile solvothermal method, and employed as signal amplification label. In this work, two working electrodes in one device were used for one determination to obtain more exact results based on screen-print technique. Taking advantage of dual-amplification effects of the Pt/Au and ZnO@CNT, this immunosensor could detect the PSA quantitatively, in the range of 0.001–500 ng mL−1 , with a low detection limit of 0.61 pg mL−1 . The resulting versatile immunosensor possesses high sensitivity, satisfactory reproducibility and regeneration. This simple and specific strategy has vast potential to be used in other biological assays. © 2014 Elsevier B.V. All rights reserved.
1. Introduction The development of highly sensitive protein detection based on antigen–antibody binding plays a key role in the prognostic treatment of human diseases. Generally, protein biomarkers have
∗ Corresponding authors. Tel.: +86 531 82767161. E-mail addresses: [email protected] (J. Yu), [email protected] (X. Song). 0003-2670/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.aca.2014.01.047
been detected using conventional immunoassays [1]. However, conventional immunoassays are complicated, time-consuming, tedious, expensive, and labor-intensive [2–5]. As an alternative to the conventional immunoassay procedures, the development of electrochemiluminescence (ECL) immunosensors has shown great promise because of high sensitivity, simple instrumentation, low cost, and good compatibility [6,7]. Thus, it has received particular attention for the detection of proteins in clinical applications [8–11]. However, improvement of sensitivity is crucial for
F. Liu et al. / Analytica Chimica Acta 818 (2014) 46–53
the successful clinical applications. Sensitivity can be improved by improving the platform for immobilizing primary antibodies and increasing the amount of signal labels. To fulfill these improvements, most of recent developments in ultrasensitive immunosensing are based on nanomaterials and nanotechnologies [12–14]. As for the fabrication process of ECL immunosensors, emphasis was placed on the step of immobilizing biomolecules on the sensing surface. Metal nanoparticles have drawn considerable interests in various fields of science and engineering because of their unique physical and chemical properties, leading to potential applications in electronics, in optical and for electroanalytical purposes [15,16]. Bimetallic inorganic nanoparticles often present more versatile properties in comparison with their monometallic counterparts. A case in point is Pt/Au alloy, which not only exhibit a high surface area ratio, but also accelerate the electron transfer rate effectively. Several studies involving Pt/Au alloy have been reported in the literatures [17–20]. However, these approaches to synthesize Pt/Au alloy reported previously are often somewhat complex and time-consuming, thereby limiting their applicability to the mass production of biosensors. Herein, according to the literature [21], a simple electrodeposition method was applied to synthesize Pt/Au alloy as carrier for the capture antibody. In previous reports, ECL from nanocrystal luminophors of semiconductors such as Si [22], CdSe [23] and CdS [24] was studied. However, the ECL properties of ZnO nano-structures have rarely been studied due mainly to the wide band gap of ZnO and the instability of the electrogenerated reactants. Recently, Zhu’s group [25] has reported ECL of ZnO, ZnO/ZnS and ZnO/ZnSe core/shell nanostructures in aqueous systems. Although with a reasonably good stability and enhanced intensity, the ECL needs more indepth studies. In this paper, a facile solvothermal method is first reported to synthesize nanocomposite of ZnO quantum dots (ZnO QDs) dotted carbon nanotube (CNT) (ZnO@CNT). The special surface structure of the ZnO@CNT nanocomposites exhibits remarkable ECL characteristics. Compared with pure ZnO QDs, about a significant enhancement of ECL from the ZnO@CNT nanocomposites has been observed, which can be attributed to the superior electron transfer ability of CNT. In addition, the ECL peak voltage shifts positively from −1.6 to −1.5 V, manifesting the effectiveness of the CNT in delivering and injecting the needed electrons. And, ZnO@CNT was employed as ECL signal reporter to label secondary antibody (PSAAb2 ). Until now, screen-print is a well-established technique for the reliable fabrication of robust biosensors, and chemical sensors, due to its low cost and applicability to mass production [26,27]. The adaptability of screen printed electrodes (SPEs) is also of great benefit in areas of research. And the electrodes can be modified through differing inks commercially available for the reference, counter, and working electrodes [28]. Polyvinyl chloride (PVC) was chosen as the substrate material in this work for fabricating low-cost SPE, and two working electrodes were designed for one determination to obtain more exact results. Based on the above considerations, we attempted to construct a highly sensitive and stable ECL immunosensor based on a disposable device. Initially, Pt/Au alloy was immobilized on the screen-printed carbon electrode to immobilize primary antibody (PSA-Ab1 ) by a potentiometric stripping analysis technique. On the other hand, it would be an effective way to integrate ZnO QDs with functionalized CNT as ECL signal magnified elements to label Ab2 for constructing sensitive ECL biosensor. The ZnO@CNT labels were brought to the Pt/Au modified electrode surface through a subsequent “sandwich” immunoreaction, which allowed sensitive detection of PSA, with a detection limit of 0.61 pg mL−1 .
47
2. Experimental 2.1. Reagents All reagents were of analytical-reagent grade or the highest purity available and directly used for the following experiments without further purification and the aqueous solutions unless indicated were prepared with Milli-Q water (Millipore). Chloroauric acid (HAuCl4 ), chloroplatinic acid (H2 PtCl6 ), hydrated zinc acetate (Zn(Ac)2 ·2H2 O) were obtained from Shanghai Chemical Reagent Company (Shanghai, China). Ethylene glycol (EG) N-(3-dimethylaminopropyl)-N -ethylcarbodiimidehydrochloride (EDC) and N-hydro-xysuccinimide (NHS) were purchased from Alfa Aesar China Ltd. PSA, PSA-Ab1 , PSA-Ab2 and bovine serum albumin (BSA) were purchased from Shanghai Linc-Bio Science Co. Ltd (Shanghai, China). The clinical serum samples were provided by Shandong Tumor Hospital. 2.2. Apparatus The ECL experiments were performed on MPI-B ECL analyzer (Remax, Xi’an) equipped with a PMT. UV–vis spectra were obtained on a Shimadzu UV-2550 UV–vis spectrophotometer (Shimadzu Corporation, Japan). Scanning electron microscope (SEM) analyses were performed using a QUANTA FEG 250 thermal field emission scanning electron microscopy (FEI Co., USA), and the microscope was equipped with an Oxford X-MAX50 energy dispersive spectrometer (EDS) (Oxford, Britain). Transmission electron microscope (TEM) images were obtained from a JEOL JEM-1400 microscope (JEOL, Japan). X-ray photoelectron spectroscopy (XPS) measurements were taken by using a spectrometer (Thermo Fisher Scientific, ESCALAB 250) with a monochromic Al Ka source at 1486.7 eV, at a voltage of 15 kV and an emission current of 10 mA. Electrochemical impedance spectra (EIS) were performed on a CHI 600D Electrochemical Workstation (Shanghai CH Instruments Inc., China). 2.3. Fabrication of SPEs As shown in Scheme 1A, the SPEs containing two carbon working electrodes (diameter: 2.5 mm), an Ag/AgCl reference, and a carbon counter electrode were prepared with screen-printed technology
Scheme 1. Schematic representation of the immunoassay procedure for the ECL device. SPE: (a) PVC film, (b) carbon ink, (c) carbon ink counter electrode, (d) Ag/AgCl reference electrode, (e) two carbon ink working electrodes, (f) insulating dielectric (A); Pt/Au modified SPE (B); after immobilization of Ab1 (C); blocking with BSA (D); capture with PSA (E); immobilization with the ZnO@CNT labeled Ab2 (F).
48
F. Liu et al. / Analytica Chimica Acta 818 (2014) 46–53
on PVC film according to the reported work [29]. The insulating layer printed around the working area constituted a reservoir of the ECL microcell. The surface of Ag/AgCl chloride ink and carbon ink printed on the film were polished to and fro along the direction of printing the conductive ink using an agate lapping hammer until the surface turned into smooth and shiny. The surface of the SPEs was rinsed several times by ultrapure water and dried. Then the polished SPEs were activated in PBS (pH 7.4) containing 5.0 mM K3 [Fe(CN)6 ] and 0.2 mol L−1 KCl for 3 min by cyclic voltammogram with potential from −0.2 to +0.6 V. Then, the electrodes were washed with deionized water and dried before use. 2.4. Preparation of Pt/Au alloy Pt/Au alloy was prepared according to the literature procedure [21]. Simply, a mixed solution containing 1 mM HAuCl4 , 1 mM H2 PtCl6 and a small amount of sodium sulfate was adopted as the electrodeposition base solution. After constant potential (−0.2 V) deposition for 30 s, Pt/Au alloy was fabricated on the electrode surface. 2.5. Preparation of ZnO@CNT composite labeled Ab2 The ZnO@CNT was first synthesized through the one-pot ethanol-thermal reaction of Zn–EG–Ac and the CNTs. Zn–EG–Ac complex was obtained by reacting Zn(Ac)2 with EG as described in previous report [30]. CNTs were first carboxyl functionalized and shortened by sonicating in a mixture of concentrated HNO3 and H2 SO4 (v/v, 1:3) for 4 h followed by extensive washing in water until the filtrate was at neutral pH. After this procedure, the CNTs were better separated and more active centers were formed in the surface. 0.04 g of the acid-treated CNTs were added into 20 mL ethanol under ultrasonication to get the dispersion solution A. Solution B was prepared by the addition of 0.5 g of Zn–EG–Ac precursor into 20 mL ethanol, and then stirring for 30 min. After mixing solution A and B, the final solution was further stirred for 30 min, and then transferred to a 50 mL Teflon-lined stainless steel autoclave. After heating at 150 ◦ C for 8 h and naturally cooling to room temperature, the resulting suspension was filtered and washed with water three times, and dried at 60 ◦ C for 24 h to obtain the ZnO@CNT composite. 1 mL of the ZnO@CNT suspension (5 mg mL−1 ) was mixed with 1 mL of Ab2 solution (20 g mL−1 , in 0.01 M pH 7.4 PBS) to obtain ZnO@CNT composite labeled Ab2 conjugate (Ab2 -ZnO@CNT). Subsequently, 100 L of freshly prepared EDC (20 mg mL−1 , in 0.01 M pH 7.4 PBS) and 100 L of NHS (10 mg mL−1 , in 0.01 M pH 7.4 PBS) were added. After incubation at room temperature for 1 h, Ab2 ZnO@CNT was redispersed in 1% BSA solution for 30 min, again under stirring, to block the excess amino group and nonspecific binding sites. 2.6. Fabrication of the immunosensor A sandwich-type immunosensing platform was constructed on the two working electrodes of SPEs, as shown in Scheme 1. The Pt/Au alloy modified working electrodes were incubated with Ab1 (5 L, 20 g mL−1 ) for 30 min, followed by washing with PBS buffer to remove unspecific physically adsorption. Then the modified electrodes was soaked in 1% BSA solution at 4 ◦ C for 30 min to block the nonspecific binding sites on the electrode surface. The as-prepared immunosensors (designed as SPE/Pt/Au-Ab1 ) were used for detection of PSA analyte. 5 L of PSA solutions with different concentrations were mixed with the above immunosensor and the incubation was conducted at 37 ◦ C for 40 min to react with the limited binding sites of Ab1 . Then
the PSA immobilized SPE/Pt/Au-Ab1 (designed as SPE/Pt/Au-Ab1 PSA) was gotten through the antigen–antibody specific interaction on the surface of Pt/Au alloy. After that, the conjugate Ab2 ZnO@CNT was incubated with SPE/Pt/Au-Ab1 -PSA for 40 min at 37 ◦ C to construct the sandwich-type immunocomplex SPE/Pt/AuAb1 -PSA-Ab2 -ZnO@CNT. The excess Ab2 -ZnO@CNT was washed with PBS buffer. This fabrication process of the ECL immunosensor was illustrated in Scheme 1. 2.7. ECL detection of PSA with the immunosensor ECL measurements were done at room temperature and the potential swept from 0 to −1.5 V with scan rate of 100 mV s−1 in 2 mL of PBS buffer solution (pH 7.4) containing 0.05 M K2 S2 O8 as the coreactant and 0.1 M KCl with a photomultiplier tube voltage of 800 V. The ECL signals related to the different PSA concentrations could be measured. 3. Results and discussion 3.1. Characterization of Pt/Au alloy Typical morphologies of Pt/Au alloy nanoparticles were analyzed by SEM. Fig. 1A exhibited a relatively uniform distributed layer of spheres (Pt/Au alloy) with average particle sizes of 200 ± 50 nm. EDS analysis was used to further confirm the successful modification of Pt/Au alloy, and the result was shown in Fig. 1B. The electrochemical behaviors for Au and Pt/Au were investigated by cyclic voltammogram in deaerated 0.5 M H2 SO4 , and the data were presented in Fig. 1C. As shown in Fig. 1C, a peak at 0.24 V was observed and a much higher current value was obtained for Pt/Au, which were attributed to the presence of Pt [31]. To further confirm the better electrical conductivity of Pt/Au alloy, we investigated the electrochemical behavior of the different nanoparticles modified SPEs in pH 7.4 PBS. From Fig. 1D, we can see that Pt/Au alloy modified SPE bring the obvious increase in current compaired with Pt nanoparticles or Au nanoparticles modified SPEs. This demonstrated that Pt/Au alloy effectively accelerated the electron transfer rate so as to increase the ECL signal as well as provided a biocompatible microenvironment for the immobilization of more Ab1 . 3.2. Characterization of ZnO@CNT composite SEM and TEM images were selected to characterize the morphology of ZnO@CNT composite. As shown in Fig. 2A, the CNTs were decorated by ZnO QDs (with an average diameter of 18 nm), forming chain-like structures along the CNT, while the morphology of the CNT remained unaltered. To further explore the structure of ZnO@CNT composite, TEM (Fig. 2B) was carried out. The UV–visible absorption spectrum of the as-prepared ZnO@CNT composite was measured along with that of the pure ZnO QDs for comparison, and the results were shown in Fig. 2C. The absorption peak of the ZnO@CNT composites (357 nm) had an 18 nm blue-shift of band edge with respect to that of pure ZnO QDs (375 nm). The blue shift and the change of peak can be explained by a strong small-size confinement effect if the ZnO QDs were composed of ZnO nanodomains with a size of a few nanometers dispersed on the CNTs. The chemical composition and further details of the formation of the ZnO@CNT composites were confirmed by XPS. As can be observed in Fig. 2D, we can see that the XPS spectrum of C1s can be deconvoluted into two peaks centred at 284.7 and 288.8 eV. The peak at 284.7 eV was attributed to the sp2 carbon atom, while the peak positioned at 288.8 eV was assigned to the C from O C OH species. Besides the C1s peak, the O1s peak was obtained at 530.1 eV in Fig. 2E. As shown in Fig. 2F, we also found the Zn2p peaks located
F. Liu et al. / Analytica Chimica Acta 818 (2014) 46–53
49
Fig. 1. SEM (A) and EDS (B) analysis of Pt/Au alloy; cyclic voltammogram of Au and Pt/Au modified SPEs in deaerated 0.5 M H2 SO4 (C); Cyclic voltammograms of different modified SPEs (D).
at 1021.2 eV (Zn2p3/2 ) and 1044.3 eV (Zn2p5/2 ), respectively. Therefore the XPS data, together with the above microscopy results, unambiguously indicated the formation of ZnO at the surface of CNTs. 3.3. Possible mechanism for the ECL behaviors of the ZnO@CNT composite Nanocrystals can be reduced and oxidized by charge injection during the potential cycling at the electrodes [32]. Here, ECL was based on the electron-transfer reaction between reduced species formed in ZnO and oxidized species of the co-reactant (SO4 •− ) [33]. This can be correlated to the aggregation morphology of the ZnO nanocrystals and the addition of the co-reactant (S2 O8 2− ), which helped overcome either a limited potential window of a solvent or the poor stability of electrogenerated reduced species. S2 O8 2− was reduced and generated a strong oxidant SO4 •− , then SO4 •− reacted with the electrogenerated species (ZnO•− ) to generate higher intensity light emission. The corresponding ECL processes are shown as follows [34]: ZnO + ne− −→ nZnO• − S2 O8
2−
−
+ e −→ SO4
2−
(1) + SO4
•−
ZnO• − + SO4 • − −→ ZnO∗ + SO4 2− ∗
ZnO −→ ZnO + h
(2) (3) (4)
In order to identify the superior ECL performance of ZnO@CNT composite, the cyclic voltammogram and ECL-potential curves for
the ZnO@CNT composite constructed sandwich-type immunosensor were measured, as well as those of pure ZnO QDs. From Fig. 3, it can be seen that the cathodic peak current and ECL intensity of ZnO@CNT composite were all increased compared with those of pure ZnO QDs. Compared with pure ZnO QDs, it can be seen that the ECL intensity of ZnO@CNT composite was 3-fold greater than that of pure ZnO QDs, and the ECL peak voltage (−1.5 V) was much more positive than that of pure ZnO QDs (−1.6 V) as well as the ECL onset voltage. Cyclic voltammograms demonstrated that the onset reduction potential of ZnO also shifted positively and the cathodic peak current was enhanced in the ZnO@CNT composite. These phenomena indicated that the presence of CNTs obviously decreased the potential barriers [35]. On the other hand, CNTs with high conductivity will induce more SO4 •− improving the quantity of the transmission state (ZnO* ), which can also contribute to the enhanced ECL intensity. 3.4. EIS characterization of the immunosensor EIS is an effective tool for probing the features of surfacemodified and biomaterial-functionalized electrodes. Fig. 4A showed the nyquist plots of EIS corresponding to the stepwise modification processes. For a bare SPE surface, the impedance spectrum (curve a) exhibited a small electron-transfer resistence (Ret ). When Pt/Au alloy (curve b) was modified on the SPE surface the resistance decreased. The big drop of impedance in curve b can be owing to the high electron transfer rate of Pt/Au alloy. Then, Ab1 (curve c), BSA (curve d), and PSA (curve e) were constructed step by step, and the Ret of each related electrode increased correspondingly. The reason for the resistance increase was that nonconductive property
50
F. Liu et al. / Analytica Chimica Acta 818 (2014) 46–53
Fig. 2. SEM (A) and TEM (B) image of ZnO@CNT; Absorption spectrum of ZnO and ZnO@CNT (C); XPS spectra of ZnO@CNT: C1s (D); O1s (E); Zn2p (F).
of biomacromolecule would obstruct electron transfer and mass transport of the electrochemical probe. After Ab2 -ZnO@CNT was added, the Ret showed a small decrease, which can be attributed to the excellent conductivity of CNT. The EIS results indicated that the biomacromolecules were successfully assembled on the electrode surface. 3.5. Characterization of the improved ECL performance for our immunoassay The SPEs were scanned from 0 V to −1.5 V with scan rate of 100 mV s−1 to collect ECL signals. To investigate the amplification technique of the Pt/Au alloy and ZnO@CNT composite for ECL
analysis, control experiments without the involvement of Pt nanoparticles or CNTs were carried out and the results were shown in Fig. 4B. We compared the ECL intensity of different immunosensors constructed with pure ZnO QDs labeled Ab2 (curve a) and ZnO@CNT composite labeled Ab2 (curve c). The quantity of the ZnO QDs and ZnO@CNT composite was equal in both labels and Pt/Au alloy was used to capture Ab1 with the same quantity. As can be observed that the ZnO@CNT composite labeled Ab2 revealed excellent ECL performance compared with pure ZnO QDs. The inset showed plots of ECL intensity of the immunosensor vs. the target PSA concentration of pure ZnO QDs (a) and ZnO@CNT composites labeled Ab2 (c), respectively. From the inset in Fig. 4B, we can see that the ZnO@CNT composite can be used as excellent ECL labels.
Fig. 3. ECL-potential curves (A) and cyclic voltammograms (B) of ZnO QDs (curve a), and ZnO@CNT composite (curve b) in pH 7.4 PBS containing 0.05 M K2 S2 O8 and 0.1 M KCl.
F. Liu et al. / Analytica Chimica Acta 818 (2014) 46–53
51
Fig. 4. EIS of the modified SPEs in 5.0 mmol L−1 Fe(CN)6 4−/3− containing 0.1 mol L−1 KCl (A); Comparison of different immunoassays with different biocomplexes: (a): SPE/Pt/Au-Ab1 -PSA-Ab2 -ZnO; (b): SPE/Au-Ab1 -PSA-Ab2 -ZnO@CNT; (c): SPE/Pt/Au-Ab1 -PSA-Ab2 -ZnO@CNT (B); Relationship between ECL intensity and PSA concentration, each point is the average of five measurements (Insert: logarithmic calibration curve for PSA) (C).
Similarly, comparative experiments were carried out based on pure Au nanoparticles (curve b) and Pt/Au alloy (curve c), respectively. As shown in Fig. 4B, it was observed that a greatly enhanced ECL performance was obtained by applying Pt/Au alloy to immobilize Ab1 instead of pure Au nanoparticles. In a conclusion, the proposed ECL immunosensor showed a greatly amplified ECL signal based on ZnO@CNT composite and Pt/Au alloy and hence enhanced the detection sensitivity. 3.6. Analytical performance The analytical performance of this method was verified by applying human PSA standard solutions at various concentrations in PBS under the optimum conditions (seeing in supplemental materials). With the increasing concentrations of PSA, good correlation between the logarithm concentration of PSA and the ECL peak intensity with a wide dynamic range (0.001–500 ng mL−1 ) was also observed (Fig. 4C). The linear regression equation was ECL = 6957.07 + 2058.16 lgcPSA (ng mL−1 ). The limit of detection at a signal-to-noise ratio of 3 for PSA was 0.61 pg mL−1 . For comparison, the analytical properties of the fabricated biosensor were compared with the previously reported PSA biosensors based on utilization of different materials and methods (Table 1). These results adequately suggest that our immunoassay owns greatly improved sensitivity and working range. The greatly improved performance for our immunoassay can be attributed to the excellent electrical properties of Pt/Au alloy and good electrical conductivity of ZnO@CNT composite based on our ECL assay. 3.7. Specificity, throughput, reproducibility, and stability of the immunosensor To confirm that the observed ECL response was generated from the antibody–antigen specific interaction not nonspecific protein
interaction, selectivity was investigated. And the immunosensor was incubated with the samples containing the following two kinds of potential interfering substances: carcinoma embryonic antigen (CEA, 1.0 ng mL−1 ) and human chorionic gonadotrophin (HCG, 1.0 ng mL−1 ). Then the ECL response of the mixture was detected through the as-fabricated immunosensor. As shown in Fig. 5A, the ECL responses to 1.0 ng mL−1 PSA standard solutions with and without interferences showed a difference of less than 5.1%, which suggested that the selectivity of the immunosensor based on the specific antigen–antibody immunoreaction is acceptable. High sample throughput is a long-cherished goal in the developments of both immunosensor and analyte testing. It is of great significance in early disease diagnosis. To achieve this goal, appropriate analytical time per assay is necessary. The total testing process can be completed for one SPEs device within 143 min, including a total incubation time of 141 min for the fabrication of the two immunosensors, 2 min for the ECL signals collection, and washing. Furthermore, parallel incubation could be conveniently performed on a series of SPEs devices, and the ECL detection needed only 1 min. A higher throughput was obtained when more SPEs devices were used for parallel incubation and immunoassay. The stability played vital role in the performance of the prepared immunosensor. As shown in Fig. 5B, after the immunosensors were stored in pH 7.4 PBS at 4 ◦ C for two weeks, the ECL signal declined 1.5% and 2.4% of the immediate value for the prepared biosensor at PSA concentrations of 1.0 ng mL−1 (A) and 0.05 ng mL−1 (B), respectively. In addition, the inset in Fig. 5C was the ECL signal-time curve under continuous potential scanning for 10 cycles. As can be observed, after running for 10 cycles, only a 2.1% decline of the original ECL was presented. These results indicated the superior stability of our immunosensor. Reproducibility was a vital parameter, too. So we tested it for 1.0 ng mL−1 PSA with six fabricated immunosensors
Table 1 Comparison of analytical properties of different immunoassys toward PSA. Immunoassay format
Modified platform
Signal antibody
Linear range (ng mL−1 )
Detection limit (pg mL−1 )
Ref.
Electrochemical immunoassay
Single wall carbon nanotube forest Diaminoheptane modified nitrocellulose membrane Graphene
Horseradish peroxidase
1–40
3
[36]
Quantum dots
0.05–4
20
[37]
Gold nanorods labeled with glucose oxidase Carbon nanotubes- horseradish peroxidase Multiwalled carbon nanotube ZnO quantum dots functionalized carbon nanotube
0.01–8
3
[38]
0.4–40
4
[39]
0.005–4 0.001–500
5 0.61
[40] Thiswork
Electrochemical immunoassay Electrochemilumi-nescence immunoassay Electrochemical immunoassay Electrochemical immunoassay Electrochemilumi-nescence immunoassay
Single-wall carbon nanotube Screen-printed carbon electrode array Pt/Au alloy
52
F. Liu et al. / Analytica Chimica Acta 818 (2014) 46–53
Fig. 5. Specificity (A) and stability (B, C) of the immunosensor.
Table 2 Comparison of serum PSA levels determined using two methods. Serum samples Immunosensor (ng mL ELISA (ng mL−1 )a Relative deviation (%) a
−1 a
)
1
2
3
50.52 49.15 2.71
29.78 30.41 −2.12
12.96 12.55 3.16
The average value of three successive determinations.
Table 3 The recoveries assayed by using the electrochemical immunosensor. Sample no.
Spiking value (ng mL−1 )
Assayed value (ng mL−1 )a
Recovery (%)
1 2 3 4
1.0 5.0 10.0 15.0
1.023 4.834 10.12 14.73
102.3 96.68 101.2 98.20
a
The average of three successive assays.
independently. The consequence showed a relative standard deviation (RSD) of 3.75%, giving an acceptable fabrication reproducibility of the immunosensors. 3.8. Application of the immunosensor in human serum The feasibility of applying the immunosensor in clinical systems was investigated via analyzing several real clinical serum samples. Samples 1–3 were obtained from three prostate cancer patients and were diluted before the assay, if the concentration was found to exceed the dynamic range of the calibration graphs. The accuracy of PSA determination was examined by comparing the results with those from the enzyme-linked immunosorbent assay (ELISA) analysis. Table 2 showed the correlation results obtained using the proposed immunosensor and the ELISA method. The relative deviations of the proposed immunosensor ranged from −2.12% to 3.16%. It obviously suggested that there is no significant difference between the results given by two methods. Therefore, the proposed sensor could be reasonably applied in the clinical determination of PSA in human plasm. We also researched the recovery, by spiking the serum sample obtained from prostate cancer patient into pH 7.4 PBS, and the content of PSA was assayed by using the proposed ECL immunosensor. The results listed in Table 3 showed satisfying recoveries of 96.68–102.3% for PSA. 4. Conclusions SPEs device was applied to a sensitive ECL sandwich-type strategy for the specific detection of PSA based on Pt/Au alloy and ZnO@CNT composite. The main advantages of the present
biosensor contribute to several aspects: (1) The ECL sensing platform that constructed with Pt/Au alloy makes great contributions to the high ECL intensity. (2) The proposed strategy which based on ZnO@CNT composite greatly amplifies the ECL signal and allows the sensitive detection at a lower potential. (3) SPE-based microdevice with two working electrodes was used for one time determination to obtain more exact results. (4) The immunosensor can detect as low as 0.61 pg mL−1 , which is superior to other methods. These features, as well as its other advantages, such as high specificity, low cost, excellent stability, make it a promising candidate for biomedical and bioanalytical applications. Acknowledgements This work was financially supported by Natural Science Research Foundation of China (21277058, 21175058, 21207048) and Natural Science Foundation of Shandong Province, China (ZR2012BZ002, ZR2011BQ019). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.aca.2014.01.047http://dx.doi.org/10.1016/j.aca.2014.01.047. References [1] A.M. Yates, S.J. Elvin, D.E. Williamson, Journal of Immunoassay 20 (1999) 31–44. [2] B. Zhang, X. Zhang, H.H. Yan, S.J. Xu, D.H. Tang, W.L. Fu, Biosensors and Bioelectronics 23 (2007) 19–25. [3] D.P. Tang, J.J. Ren, Analytical Chemistry 80 (2008) 8064–8070. [4] D.P. Tang, R. Yuan, Y.Q. Chal, Analytical Chemistry 80 (2008) 1582–1588. [5] P.C. Mathias, N. Ganesh, B.T. Cunningham, Analytical Chemistry 80 (2008) 9013–9020. [6] J. Qian, C.Y. Zhang, X.D. Cao, S.Q. Liu, Analytical Chemistry 82 (2010) 6422–6429. [7] J. Wang, H.Y. Han, X.C. Jiang, L. Huang, L.N. Chen, N. Li, Analytical Chemistry 84 (2012) 4893–4899. [8] N. Liao, Y. Zhuo, Y.Q. Chai, Y. Xiang, Y.L. Cao, R. Yuan, J. Han, Chemical Communications 48 (2) (2012) 7610–7612. [9] S.Y. Deng, J.P. Lei, Y. Huang, Y. Cheng, H.X. Ju, Analytical Chemistry 85 (2013) 5390–5396. [10] L. Wang, W. Wei, J. Han, Z.F. Fu, Analyst 137 (2012) 735–740. [11] J. Zhang, J.P. Lei, C.L. Xu, L. Ding, H.X. Ju, Analytical Chemistry 82 (2010) 1112–1117. [12] G.F. Jie, P. Liu, S.S. Zhang, Chemical Communications 46 (2010) 1323–1325. [13] H. Dai, C.P. Yang, Y.J. Tong, G.F. Xu, X.L. Ma, Y.Y. Lin, G.N. Chen, Chemical Communications 48 (2012) 3055–3057. [14] S.Y. Deng, J.P. Lei, X.N. Yao, Y. Huang, D.J. Lin, H.X. Ju, Journal of Materials Chemistry C 1 (2013) 299–306. [15] K. Qian, L.F. Luo, H.Z. Bao, Q. Hua, Z.Q. Jiang, W.X. Huang, Catalysis Science & Technology 3 (2013) 679–687. [16] G.R. Zhang, B.Q. Xu, Nanoscale 2 (2010) 2798–2804. [17] L.J. Bai, R. Yuan, Y.Q. Chai, Y.L. Yuan, L. Mao, Y. Wang, Analytica Chimica Acta 698 (2011) 14–19. [18] X.H. Kang, Z.B. Mai, X.Y. Zou, P.X. Cai, J.Y. Mo, Analytical Biochemistry 369 (2007) 71–79. [19] Y. Li, R. Yuan, Y.Q. Chai, Z.J. Song, Electrochimica Acta 56 (2011) 6715–6721.
F. Liu et al. / Analytica Chimica Acta 818 (2014) 46–53 [20] S. Shahrokhian, S. Rastgar, Analyst 137 (2012) 2706–2715. [21] X.H. Kang, Z.B. Mai, X.Y. Zou, P.X. Cai, J.Y. Mo, Analytical Biochemistry 369 (2007) 71–79. [22] H.C. Choi, J.M. Buriak, Chemistry of Materials 12 (2000) 2151–2156. [23] B. Liu, T. Ren, J.R. Zhang, H.Y. Chen, J.J. Zhu, C. Burda, Electrochemistry Communications 9 (2007) 551–557. [24] C.Z. Wang, L.Z. Fan, Z.H. Wang, H.B. Liu, Y.L. Li, S.H. Yang, Y.L. Li, Advanced Materials 19 (2007) 3677–3681. [25] J. Geng, B. Liu, L. Xu, F.N. Hu, J.J. Zhu, Langmuir 23 (2007) 10286–10293. [26] L. Ge, J.X. Yan, X.R. Song, M. Yan, S.G. Ge, J.H. Yu, Biomaterials 33 (2012) 1024–1031. [27] D. Hernández-Santos, M. Díaz-González, M.B. González-García, A. Costa-García, Analytical Chemistry 76 (2004) 6887–6893. [28] J.M. Zen, Y.S. Song, H.H. Chung, C.T. Hsu, A. Senthil Kumar, Analytical Chemistry 74 (2002) 6126–6130. [29] M. Yan, D.J. Zang, S.G. Ge, L. Ge, J.H. Yu, Biosensors and Bioelectronics 38 (2012) 355–361. [30] C.G. Tian, W. Li, Q. Zhang, K. Pan, H.G. Fu, Materials Research Bulletin 46 (2011) 1283–1289.
53
[31] Q. Zhu, R. Yuan, Y.Q. Chai, J. Han, Y. Li, N. Liao, Analyst 138 (2013) 620– 626. [32] T. Ren, J.Z. Xu, Y.F. Tu, S. Xu, J.J. Zhu, Electrochemistry Communications 7 (2005) 5–9. [33] M.M. Richter, Chemical Reviews 104 (2004) 3003. [34] G.F. Jie, B. Liu, J.J. Miao, J.J. Zhu, Talanta 71 (2007) 1476–1480. [35] R.X. Zhang, L.Z. Fan, Y.P. Fang, S.H. Yang, Journal of Materials Chemistry 18 (2008) 4964–4970. [36] B.V. Chikkaveeraiah, A. Bhirde, R. Malhotra, V. Patel, J.S. Gutkind, J.F. Rusling, Analytical Chemistry 81 (2009) 9129–9134. [37] G.D. Liu, Y.Y. Lin, J. Wang, H. Wu, C.M. Wai, Y.H. Lin, Analytical Chemistry 79 (2007) 7644–7653. [38] S.J. Xu, Y. Liu, T.H. Wang, J.H. Li, Analytical Chemistry 83 (2011) 3817– 3823. [39] X. Yu, B. Munge, V. Patel, G. Jensen, A. Bhirde, J.D. Gong, S.N. Kim, J. Gillespie, J.S. Gutkind, F. Papadimitrakopoulos, J.F. Rusling, Journal of the American Chemical Society 128 (2006) 11199–11205. [40] Y. Wan, W.P. Deng, Y. Su, X.H. Zhu, C. Peng, H.Y. Hu, H.Z. Peng, S.P. Song, C.H. Fan, Biosensors and Bioelectronics 30 (2011) 93–99.
Contents lists available at ScienceDirect
Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca
Application of ZnO quantum dots dotted carbon nanotube for sensitive electrochemiluminescence immunoassay based on simply electrochemical reduced Pt/Au alloy and a disposable device Fang Liu a , Wenping Deng a , Yan Zhang a , Shenguang Ge b , Jinghua Yu a,∗ , Xianrang Song c,∗ a Key Laboratory of Chemical Sensing & Analysis in Universities of Shandong, School of Chemistry and Chemical Engineering, University of Jinan, Jinan, 250022, PR China b Shandong Provincial Key Laboratory of Preparation and Measurement of Building Materials, University of Jinan, Jinan, 250022, PR China c Cancer Research Center, Shandong Tumor Hospital, Jinan, 250012, PR China
h i g h l i g h t s
g r a p h i c a l
a b s t r a c t
• A sandwich-type electrochemluminence immunosensor was fabricated. electrochemical reduced Pt/Au alloy was selected as immunosensing probes. • ZnO@CNT composite was first employed as signal amplification label.
• Simply
a r t i c l e
i n f o
Article history: Received 6 September 2013 Received in revised form 17 December 2013 Accepted 22 January 2014 Available online 4 February 2014 Keywords: ZnO quantum dots dotted carbon nanotube Pt/Au alloy Screen-printed electrode Electrochemiluminescence Immunosensor
a b s t r a c t We report on a disposable microdevice suitable for sandwich-type electrochemiluminescence (ECL) detection of prostate specific antigen (PSA). The method is making use of ZnO quantum dots dotted carbon nanotube (ZnO@CNT) and simply electrochemical reduced Pt/Au alloy. The latter was selected as immunosensing probe to modify screen-printed carbon electrode, due to its excellent electrical property. For further ultrasensitive, low-potential and stable ECL detection, ZnO@CNT composite was first synthesized using a facile solvothermal method, and employed as signal amplification label. In this work, two working electrodes in one device were used for one determination to obtain more exact results based on screen-print technique. Taking advantage of dual-amplification effects of the Pt/Au and ZnO@CNT, this immunosensor could detect the PSA quantitatively, in the range of 0.001–500 ng mL−1 , with a low detection limit of 0.61 pg mL−1 . The resulting versatile immunosensor possesses high sensitivity, satisfactory reproducibility and regeneration. This simple and specific strategy has vast potential to be used in other biological assays. © 2014 Elsevier B.V. All rights reserved.
1. Introduction The development of highly sensitive protein detection based on antigen–antibody binding plays a key role in the prognostic treatment of human diseases. Generally, protein biomarkers have
∗ Corresponding authors. Tel.: +86 531 82767161. E-mail addresses: [email protected] (J. Yu), [email protected] (X. Song). 0003-2670/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.aca.2014.01.047
been detected using conventional immunoassays [1]. However, conventional immunoassays are complicated, time-consuming, tedious, expensive, and labor-intensive [2–5]. As an alternative to the conventional immunoassay procedures, the development of electrochemiluminescence (ECL) immunosensors has shown great promise because of high sensitivity, simple instrumentation, low cost, and good compatibility [6,7]. Thus, it has received particular attention for the detection of proteins in clinical applications [8–11]. However, improvement of sensitivity is crucial for
F. Liu et al. / Analytica Chimica Acta 818 (2014) 46–53
the successful clinical applications. Sensitivity can be improved by improving the platform for immobilizing primary antibodies and increasing the amount of signal labels. To fulfill these improvements, most of recent developments in ultrasensitive immunosensing are based on nanomaterials and nanotechnologies [12–14]. As for the fabrication process of ECL immunosensors, emphasis was placed on the step of immobilizing biomolecules on the sensing surface. Metal nanoparticles have drawn considerable interests in various fields of science and engineering because of their unique physical and chemical properties, leading to potential applications in electronics, in optical and for electroanalytical purposes [15,16]. Bimetallic inorganic nanoparticles often present more versatile properties in comparison with their monometallic counterparts. A case in point is Pt/Au alloy, which not only exhibit a high surface area ratio, but also accelerate the electron transfer rate effectively. Several studies involving Pt/Au alloy have been reported in the literatures [17–20]. However, these approaches to synthesize Pt/Au alloy reported previously are often somewhat complex and time-consuming, thereby limiting their applicability to the mass production of biosensors. Herein, according to the literature [21], a simple electrodeposition method was applied to synthesize Pt/Au alloy as carrier for the capture antibody. In previous reports, ECL from nanocrystal luminophors of semiconductors such as Si [22], CdSe [23] and CdS [24] was studied. However, the ECL properties of ZnO nano-structures have rarely been studied due mainly to the wide band gap of ZnO and the instability of the electrogenerated reactants. Recently, Zhu’s group [25] has reported ECL of ZnO, ZnO/ZnS and ZnO/ZnSe core/shell nanostructures in aqueous systems. Although with a reasonably good stability and enhanced intensity, the ECL needs more indepth studies. In this paper, a facile solvothermal method is first reported to synthesize nanocomposite of ZnO quantum dots (ZnO QDs) dotted carbon nanotube (CNT) (ZnO@CNT). The special surface structure of the ZnO@CNT nanocomposites exhibits remarkable ECL characteristics. Compared with pure ZnO QDs, about a significant enhancement of ECL from the ZnO@CNT nanocomposites has been observed, which can be attributed to the superior electron transfer ability of CNT. In addition, the ECL peak voltage shifts positively from −1.6 to −1.5 V, manifesting the effectiveness of the CNT in delivering and injecting the needed electrons. And, ZnO@CNT was employed as ECL signal reporter to label secondary antibody (PSAAb2 ). Until now, screen-print is a well-established technique for the reliable fabrication of robust biosensors, and chemical sensors, due to its low cost and applicability to mass production [26,27]. The adaptability of screen printed electrodes (SPEs) is also of great benefit in areas of research. And the electrodes can be modified through differing inks commercially available for the reference, counter, and working electrodes [28]. Polyvinyl chloride (PVC) was chosen as the substrate material in this work for fabricating low-cost SPE, and two working electrodes were designed for one determination to obtain more exact results. Based on the above considerations, we attempted to construct a highly sensitive and stable ECL immunosensor based on a disposable device. Initially, Pt/Au alloy was immobilized on the screen-printed carbon electrode to immobilize primary antibody (PSA-Ab1 ) by a potentiometric stripping analysis technique. On the other hand, it would be an effective way to integrate ZnO QDs with functionalized CNT as ECL signal magnified elements to label Ab2 for constructing sensitive ECL biosensor. The ZnO@CNT labels were brought to the Pt/Au modified electrode surface through a subsequent “sandwich” immunoreaction, which allowed sensitive detection of PSA, with a detection limit of 0.61 pg mL−1 .
47
2. Experimental 2.1. Reagents All reagents were of analytical-reagent grade or the highest purity available and directly used for the following experiments without further purification and the aqueous solutions unless indicated were prepared with Milli-Q water (Millipore). Chloroauric acid (HAuCl4 ), chloroplatinic acid (H2 PtCl6 ), hydrated zinc acetate (Zn(Ac)2 ·2H2 O) were obtained from Shanghai Chemical Reagent Company (Shanghai, China). Ethylene glycol (EG) N-(3-dimethylaminopropyl)-N -ethylcarbodiimidehydrochloride (EDC) and N-hydro-xysuccinimide (NHS) were purchased from Alfa Aesar China Ltd. PSA, PSA-Ab1 , PSA-Ab2 and bovine serum albumin (BSA) were purchased from Shanghai Linc-Bio Science Co. Ltd (Shanghai, China). The clinical serum samples were provided by Shandong Tumor Hospital. 2.2. Apparatus The ECL experiments were performed on MPI-B ECL analyzer (Remax, Xi’an) equipped with a PMT. UV–vis spectra were obtained on a Shimadzu UV-2550 UV–vis spectrophotometer (Shimadzu Corporation, Japan). Scanning electron microscope (SEM) analyses were performed using a QUANTA FEG 250 thermal field emission scanning electron microscopy (FEI Co., USA), and the microscope was equipped with an Oxford X-MAX50 energy dispersive spectrometer (EDS) (Oxford, Britain). Transmission electron microscope (TEM) images were obtained from a JEOL JEM-1400 microscope (JEOL, Japan). X-ray photoelectron spectroscopy (XPS) measurements were taken by using a spectrometer (Thermo Fisher Scientific, ESCALAB 250) with a monochromic Al Ka source at 1486.7 eV, at a voltage of 15 kV and an emission current of 10 mA. Electrochemical impedance spectra (EIS) were performed on a CHI 600D Electrochemical Workstation (Shanghai CH Instruments Inc., China). 2.3. Fabrication of SPEs As shown in Scheme 1A, the SPEs containing two carbon working electrodes (diameter: 2.5 mm), an Ag/AgCl reference, and a carbon counter electrode were prepared with screen-printed technology
Scheme 1. Schematic representation of the immunoassay procedure for the ECL device. SPE: (a) PVC film, (b) carbon ink, (c) carbon ink counter electrode, (d) Ag/AgCl reference electrode, (e) two carbon ink working electrodes, (f) insulating dielectric (A); Pt/Au modified SPE (B); after immobilization of Ab1 (C); blocking with BSA (D); capture with PSA (E); immobilization with the ZnO@CNT labeled Ab2 (F).
48
F. Liu et al. / Analytica Chimica Acta 818 (2014) 46–53
on PVC film according to the reported work [29]. The insulating layer printed around the working area constituted a reservoir of the ECL microcell. The surface of Ag/AgCl chloride ink and carbon ink printed on the film were polished to and fro along the direction of printing the conductive ink using an agate lapping hammer until the surface turned into smooth and shiny. The surface of the SPEs was rinsed several times by ultrapure water and dried. Then the polished SPEs were activated in PBS (pH 7.4) containing 5.0 mM K3 [Fe(CN)6 ] and 0.2 mol L−1 KCl for 3 min by cyclic voltammogram with potential from −0.2 to +0.6 V. Then, the electrodes were washed with deionized water and dried before use. 2.4. Preparation of Pt/Au alloy Pt/Au alloy was prepared according to the literature procedure [21]. Simply, a mixed solution containing 1 mM HAuCl4 , 1 mM H2 PtCl6 and a small amount of sodium sulfate was adopted as the electrodeposition base solution. After constant potential (−0.2 V) deposition for 30 s, Pt/Au alloy was fabricated on the electrode surface. 2.5. Preparation of ZnO@CNT composite labeled Ab2 The ZnO@CNT was first synthesized through the one-pot ethanol-thermal reaction of Zn–EG–Ac and the CNTs. Zn–EG–Ac complex was obtained by reacting Zn(Ac)2 with EG as described in previous report [30]. CNTs were first carboxyl functionalized and shortened by sonicating in a mixture of concentrated HNO3 and H2 SO4 (v/v, 1:3) for 4 h followed by extensive washing in water until the filtrate was at neutral pH. After this procedure, the CNTs were better separated and more active centers were formed in the surface. 0.04 g of the acid-treated CNTs were added into 20 mL ethanol under ultrasonication to get the dispersion solution A. Solution B was prepared by the addition of 0.5 g of Zn–EG–Ac precursor into 20 mL ethanol, and then stirring for 30 min. After mixing solution A and B, the final solution was further stirred for 30 min, and then transferred to a 50 mL Teflon-lined stainless steel autoclave. After heating at 150 ◦ C for 8 h and naturally cooling to room temperature, the resulting suspension was filtered and washed with water three times, and dried at 60 ◦ C for 24 h to obtain the ZnO@CNT composite. 1 mL of the ZnO@CNT suspension (5 mg mL−1 ) was mixed with 1 mL of Ab2 solution (20 g mL−1 , in 0.01 M pH 7.4 PBS) to obtain ZnO@CNT composite labeled Ab2 conjugate (Ab2 -ZnO@CNT). Subsequently, 100 L of freshly prepared EDC (20 mg mL−1 , in 0.01 M pH 7.4 PBS) and 100 L of NHS (10 mg mL−1 , in 0.01 M pH 7.4 PBS) were added. After incubation at room temperature for 1 h, Ab2 ZnO@CNT was redispersed in 1% BSA solution for 30 min, again under stirring, to block the excess amino group and nonspecific binding sites. 2.6. Fabrication of the immunosensor A sandwich-type immunosensing platform was constructed on the two working electrodes of SPEs, as shown in Scheme 1. The Pt/Au alloy modified working electrodes were incubated with Ab1 (5 L, 20 g mL−1 ) for 30 min, followed by washing with PBS buffer to remove unspecific physically adsorption. Then the modified electrodes was soaked in 1% BSA solution at 4 ◦ C for 30 min to block the nonspecific binding sites on the electrode surface. The as-prepared immunosensors (designed as SPE/Pt/Au-Ab1 ) were used for detection of PSA analyte. 5 L of PSA solutions with different concentrations were mixed with the above immunosensor and the incubation was conducted at 37 ◦ C for 40 min to react with the limited binding sites of Ab1 . Then
the PSA immobilized SPE/Pt/Au-Ab1 (designed as SPE/Pt/Au-Ab1 PSA) was gotten through the antigen–antibody specific interaction on the surface of Pt/Au alloy. After that, the conjugate Ab2 ZnO@CNT was incubated with SPE/Pt/Au-Ab1 -PSA for 40 min at 37 ◦ C to construct the sandwich-type immunocomplex SPE/Pt/AuAb1 -PSA-Ab2 -ZnO@CNT. The excess Ab2 -ZnO@CNT was washed with PBS buffer. This fabrication process of the ECL immunosensor was illustrated in Scheme 1. 2.7. ECL detection of PSA with the immunosensor ECL measurements were done at room temperature and the potential swept from 0 to −1.5 V with scan rate of 100 mV s−1 in 2 mL of PBS buffer solution (pH 7.4) containing 0.05 M K2 S2 O8 as the coreactant and 0.1 M KCl with a photomultiplier tube voltage of 800 V. The ECL signals related to the different PSA concentrations could be measured. 3. Results and discussion 3.1. Characterization of Pt/Au alloy Typical morphologies of Pt/Au alloy nanoparticles were analyzed by SEM. Fig. 1A exhibited a relatively uniform distributed layer of spheres (Pt/Au alloy) with average particle sizes of 200 ± 50 nm. EDS analysis was used to further confirm the successful modification of Pt/Au alloy, and the result was shown in Fig. 1B. The electrochemical behaviors for Au and Pt/Au were investigated by cyclic voltammogram in deaerated 0.5 M H2 SO4 , and the data were presented in Fig. 1C. As shown in Fig. 1C, a peak at 0.24 V was observed and a much higher current value was obtained for Pt/Au, which were attributed to the presence of Pt [31]. To further confirm the better electrical conductivity of Pt/Au alloy, we investigated the electrochemical behavior of the different nanoparticles modified SPEs in pH 7.4 PBS. From Fig. 1D, we can see that Pt/Au alloy modified SPE bring the obvious increase in current compaired with Pt nanoparticles or Au nanoparticles modified SPEs. This demonstrated that Pt/Au alloy effectively accelerated the electron transfer rate so as to increase the ECL signal as well as provided a biocompatible microenvironment for the immobilization of more Ab1 . 3.2. Characterization of ZnO@CNT composite SEM and TEM images were selected to characterize the morphology of ZnO@CNT composite. As shown in Fig. 2A, the CNTs were decorated by ZnO QDs (with an average diameter of 18 nm), forming chain-like structures along the CNT, while the morphology of the CNT remained unaltered. To further explore the structure of ZnO@CNT composite, TEM (Fig. 2B) was carried out. The UV–visible absorption spectrum of the as-prepared ZnO@CNT composite was measured along with that of the pure ZnO QDs for comparison, and the results were shown in Fig. 2C. The absorption peak of the ZnO@CNT composites (357 nm) had an 18 nm blue-shift of band edge with respect to that of pure ZnO QDs (375 nm). The blue shift and the change of peak can be explained by a strong small-size confinement effect if the ZnO QDs were composed of ZnO nanodomains with a size of a few nanometers dispersed on the CNTs. The chemical composition and further details of the formation of the ZnO@CNT composites were confirmed by XPS. As can be observed in Fig. 2D, we can see that the XPS spectrum of C1s can be deconvoluted into two peaks centred at 284.7 and 288.8 eV. The peak at 284.7 eV was attributed to the sp2 carbon atom, while the peak positioned at 288.8 eV was assigned to the C from O C OH species. Besides the C1s peak, the O1s peak was obtained at 530.1 eV in Fig. 2E. As shown in Fig. 2F, we also found the Zn2p peaks located
F. Liu et al. / Analytica Chimica Acta 818 (2014) 46–53
49
Fig. 1. SEM (A) and EDS (B) analysis of Pt/Au alloy; cyclic voltammogram of Au and Pt/Au modified SPEs in deaerated 0.5 M H2 SO4 (C); Cyclic voltammograms of different modified SPEs (D).
at 1021.2 eV (Zn2p3/2 ) and 1044.3 eV (Zn2p5/2 ), respectively. Therefore the XPS data, together with the above microscopy results, unambiguously indicated the formation of ZnO at the surface of CNTs. 3.3. Possible mechanism for the ECL behaviors of the ZnO@CNT composite Nanocrystals can be reduced and oxidized by charge injection during the potential cycling at the electrodes [32]. Here, ECL was based on the electron-transfer reaction between reduced species formed in ZnO and oxidized species of the co-reactant (SO4 •− ) [33]. This can be correlated to the aggregation morphology of the ZnO nanocrystals and the addition of the co-reactant (S2 O8 2− ), which helped overcome either a limited potential window of a solvent or the poor stability of electrogenerated reduced species. S2 O8 2− was reduced and generated a strong oxidant SO4 •− , then SO4 •− reacted with the electrogenerated species (ZnO•− ) to generate higher intensity light emission. The corresponding ECL processes are shown as follows [34]: ZnO + ne− −→ nZnO• − S2 O8
2−
−
+ e −→ SO4
2−
(1) + SO4
•−
ZnO• − + SO4 • − −→ ZnO∗ + SO4 2− ∗
ZnO −→ ZnO + h
(2) (3) (4)
In order to identify the superior ECL performance of ZnO@CNT composite, the cyclic voltammogram and ECL-potential curves for
the ZnO@CNT composite constructed sandwich-type immunosensor were measured, as well as those of pure ZnO QDs. From Fig. 3, it can be seen that the cathodic peak current and ECL intensity of ZnO@CNT composite were all increased compared with those of pure ZnO QDs. Compared with pure ZnO QDs, it can be seen that the ECL intensity of ZnO@CNT composite was 3-fold greater than that of pure ZnO QDs, and the ECL peak voltage (−1.5 V) was much more positive than that of pure ZnO QDs (−1.6 V) as well as the ECL onset voltage. Cyclic voltammograms demonstrated that the onset reduction potential of ZnO also shifted positively and the cathodic peak current was enhanced in the ZnO@CNT composite. These phenomena indicated that the presence of CNTs obviously decreased the potential barriers [35]. On the other hand, CNTs with high conductivity will induce more SO4 •− improving the quantity of the transmission state (ZnO* ), which can also contribute to the enhanced ECL intensity. 3.4. EIS characterization of the immunosensor EIS is an effective tool for probing the features of surfacemodified and biomaterial-functionalized electrodes. Fig. 4A showed the nyquist plots of EIS corresponding to the stepwise modification processes. For a bare SPE surface, the impedance spectrum (curve a) exhibited a small electron-transfer resistence (Ret ). When Pt/Au alloy (curve b) was modified on the SPE surface the resistance decreased. The big drop of impedance in curve b can be owing to the high electron transfer rate of Pt/Au alloy. Then, Ab1 (curve c), BSA (curve d), and PSA (curve e) were constructed step by step, and the Ret of each related electrode increased correspondingly. The reason for the resistance increase was that nonconductive property
50
F. Liu et al. / Analytica Chimica Acta 818 (2014) 46–53
Fig. 2. SEM (A) and TEM (B) image of ZnO@CNT; Absorption spectrum of ZnO and ZnO@CNT (C); XPS spectra of ZnO@CNT: C1s (D); O1s (E); Zn2p (F).
of biomacromolecule would obstruct electron transfer and mass transport of the electrochemical probe. After Ab2 -ZnO@CNT was added, the Ret showed a small decrease, which can be attributed to the excellent conductivity of CNT. The EIS results indicated that the biomacromolecules were successfully assembled on the electrode surface. 3.5. Characterization of the improved ECL performance for our immunoassay The SPEs were scanned from 0 V to −1.5 V with scan rate of 100 mV s−1 to collect ECL signals. To investigate the amplification technique of the Pt/Au alloy and ZnO@CNT composite for ECL
analysis, control experiments without the involvement of Pt nanoparticles or CNTs were carried out and the results were shown in Fig. 4B. We compared the ECL intensity of different immunosensors constructed with pure ZnO QDs labeled Ab2 (curve a) and ZnO@CNT composite labeled Ab2 (curve c). The quantity of the ZnO QDs and ZnO@CNT composite was equal in both labels and Pt/Au alloy was used to capture Ab1 with the same quantity. As can be observed that the ZnO@CNT composite labeled Ab2 revealed excellent ECL performance compared with pure ZnO QDs. The inset showed plots of ECL intensity of the immunosensor vs. the target PSA concentration of pure ZnO QDs (a) and ZnO@CNT composites labeled Ab2 (c), respectively. From the inset in Fig. 4B, we can see that the ZnO@CNT composite can be used as excellent ECL labels.
Fig. 3. ECL-potential curves (A) and cyclic voltammograms (B) of ZnO QDs (curve a), and ZnO@CNT composite (curve b) in pH 7.4 PBS containing 0.05 M K2 S2 O8 and 0.1 M KCl.
F. Liu et al. / Analytica Chimica Acta 818 (2014) 46–53
51
Fig. 4. EIS of the modified SPEs in 5.0 mmol L−1 Fe(CN)6 4−/3− containing 0.1 mol L−1 KCl (A); Comparison of different immunoassays with different biocomplexes: (a): SPE/Pt/Au-Ab1 -PSA-Ab2 -ZnO; (b): SPE/Au-Ab1 -PSA-Ab2 -ZnO@CNT; (c): SPE/Pt/Au-Ab1 -PSA-Ab2 -ZnO@CNT (B); Relationship between ECL intensity and PSA concentration, each point is the average of five measurements (Insert: logarithmic calibration curve for PSA) (C).
Similarly, comparative experiments were carried out based on pure Au nanoparticles (curve b) and Pt/Au alloy (curve c), respectively. As shown in Fig. 4B, it was observed that a greatly enhanced ECL performance was obtained by applying Pt/Au alloy to immobilize Ab1 instead of pure Au nanoparticles. In a conclusion, the proposed ECL immunosensor showed a greatly amplified ECL signal based on ZnO@CNT composite and Pt/Au alloy and hence enhanced the detection sensitivity. 3.6. Analytical performance The analytical performance of this method was verified by applying human PSA standard solutions at various concentrations in PBS under the optimum conditions (seeing in supplemental materials). With the increasing concentrations of PSA, good correlation between the logarithm concentration of PSA and the ECL peak intensity with a wide dynamic range (0.001–500 ng mL−1 ) was also observed (Fig. 4C). The linear regression equation was ECL = 6957.07 + 2058.16 lgcPSA (ng mL−1 ). The limit of detection at a signal-to-noise ratio of 3 for PSA was 0.61 pg mL−1 . For comparison, the analytical properties of the fabricated biosensor were compared with the previously reported PSA biosensors based on utilization of different materials and methods (Table 1). These results adequately suggest that our immunoassay owns greatly improved sensitivity and working range. The greatly improved performance for our immunoassay can be attributed to the excellent electrical properties of Pt/Au alloy and good electrical conductivity of ZnO@CNT composite based on our ECL assay. 3.7. Specificity, throughput, reproducibility, and stability of the immunosensor To confirm that the observed ECL response was generated from the antibody–antigen specific interaction not nonspecific protein
interaction, selectivity was investigated. And the immunosensor was incubated with the samples containing the following two kinds of potential interfering substances: carcinoma embryonic antigen (CEA, 1.0 ng mL−1 ) and human chorionic gonadotrophin (HCG, 1.0 ng mL−1 ). Then the ECL response of the mixture was detected through the as-fabricated immunosensor. As shown in Fig. 5A, the ECL responses to 1.0 ng mL−1 PSA standard solutions with and without interferences showed a difference of less than 5.1%, which suggested that the selectivity of the immunosensor based on the specific antigen–antibody immunoreaction is acceptable. High sample throughput is a long-cherished goal in the developments of both immunosensor and analyte testing. It is of great significance in early disease diagnosis. To achieve this goal, appropriate analytical time per assay is necessary. The total testing process can be completed for one SPEs device within 143 min, including a total incubation time of 141 min for the fabrication of the two immunosensors, 2 min for the ECL signals collection, and washing. Furthermore, parallel incubation could be conveniently performed on a series of SPEs devices, and the ECL detection needed only 1 min. A higher throughput was obtained when more SPEs devices were used for parallel incubation and immunoassay. The stability played vital role in the performance of the prepared immunosensor. As shown in Fig. 5B, after the immunosensors were stored in pH 7.4 PBS at 4 ◦ C for two weeks, the ECL signal declined 1.5% and 2.4% of the immediate value for the prepared biosensor at PSA concentrations of 1.0 ng mL−1 (A) and 0.05 ng mL−1 (B), respectively. In addition, the inset in Fig. 5C was the ECL signal-time curve under continuous potential scanning for 10 cycles. As can be observed, after running for 10 cycles, only a 2.1% decline of the original ECL was presented. These results indicated the superior stability of our immunosensor. Reproducibility was a vital parameter, too. So we tested it for 1.0 ng mL−1 PSA with six fabricated immunosensors
Table 1 Comparison of analytical properties of different immunoassys toward PSA. Immunoassay format
Modified platform
Signal antibody
Linear range (ng mL−1 )
Detection limit (pg mL−1 )
Ref.
Electrochemical immunoassay
Single wall carbon nanotube forest Diaminoheptane modified nitrocellulose membrane Graphene
Horseradish peroxidase
1–40
3
[36]
Quantum dots
0.05–4
20
[37]
Gold nanorods labeled with glucose oxidase Carbon nanotubes- horseradish peroxidase Multiwalled carbon nanotube ZnO quantum dots functionalized carbon nanotube
0.01–8
3
[38]
0.4–40
4
[39]
0.005–4 0.001–500
5 0.61
[40] Thiswork
Electrochemical immunoassay Electrochemilumi-nescence immunoassay Electrochemical immunoassay Electrochemical immunoassay Electrochemilumi-nescence immunoassay
Single-wall carbon nanotube Screen-printed carbon electrode array Pt/Au alloy
52
F. Liu et al. / Analytica Chimica Acta 818 (2014) 46–53
Fig. 5. Specificity (A) and stability (B, C) of the immunosensor.
Table 2 Comparison of serum PSA levels determined using two methods. Serum samples Immunosensor (ng mL ELISA (ng mL−1 )a Relative deviation (%) a
−1 a
)
1
2
3
50.52 49.15 2.71
29.78 30.41 −2.12
12.96 12.55 3.16
The average value of three successive determinations.
Table 3 The recoveries assayed by using the electrochemical immunosensor. Sample no.
Spiking value (ng mL−1 )
Assayed value (ng mL−1 )a
Recovery (%)
1 2 3 4
1.0 5.0 10.0 15.0
1.023 4.834 10.12 14.73
102.3 96.68 101.2 98.20
a
The average of three successive assays.
independently. The consequence showed a relative standard deviation (RSD) of 3.75%, giving an acceptable fabrication reproducibility of the immunosensors. 3.8. Application of the immunosensor in human serum The feasibility of applying the immunosensor in clinical systems was investigated via analyzing several real clinical serum samples. Samples 1–3 were obtained from three prostate cancer patients and were diluted before the assay, if the concentration was found to exceed the dynamic range of the calibration graphs. The accuracy of PSA determination was examined by comparing the results with those from the enzyme-linked immunosorbent assay (ELISA) analysis. Table 2 showed the correlation results obtained using the proposed immunosensor and the ELISA method. The relative deviations of the proposed immunosensor ranged from −2.12% to 3.16%. It obviously suggested that there is no significant difference between the results given by two methods. Therefore, the proposed sensor could be reasonably applied in the clinical determination of PSA in human plasm. We also researched the recovery, by spiking the serum sample obtained from prostate cancer patient into pH 7.4 PBS, and the content of PSA was assayed by using the proposed ECL immunosensor. The results listed in Table 3 showed satisfying recoveries of 96.68–102.3% for PSA. 4. Conclusions SPEs device was applied to a sensitive ECL sandwich-type strategy for the specific detection of PSA based on Pt/Au alloy and ZnO@CNT composite. The main advantages of the present
biosensor contribute to several aspects: (1) The ECL sensing platform that constructed with Pt/Au alloy makes great contributions to the high ECL intensity. (2) The proposed strategy which based on ZnO@CNT composite greatly amplifies the ECL signal and allows the sensitive detection at a lower potential. (3) SPE-based microdevice with two working electrodes was used for one time determination to obtain more exact results. (4) The immunosensor can detect as low as 0.61 pg mL−1 , which is superior to other methods. These features, as well as its other advantages, such as high specificity, low cost, excellent stability, make it a promising candidate for biomedical and bioanalytical applications. Acknowledgements This work was financially supported by Natural Science Research Foundation of China (21277058, 21175058, 21207048) and Natural Science Foundation of Shandong Province, China (ZR2012BZ002, ZR2011BQ019). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.aca.2014.01.047http://dx.doi.org/10.1016/j.aca.2014.01.047. References [1] A.M. Yates, S.J. Elvin, D.E. Williamson, Journal of Immunoassay 20 (1999) 31–44. [2] B. Zhang, X. Zhang, H.H. Yan, S.J. Xu, D.H. Tang, W.L. Fu, Biosensors and Bioelectronics 23 (2007) 19–25. [3] D.P. Tang, J.J. Ren, Analytical Chemistry 80 (2008) 8064–8070. [4] D.P. Tang, R. Yuan, Y.Q. Chal, Analytical Chemistry 80 (2008) 1582–1588. [5] P.C. Mathias, N. Ganesh, B.T. Cunningham, Analytical Chemistry 80 (2008) 9013–9020. [6] J. Qian, C.Y. Zhang, X.D. Cao, S.Q. Liu, Analytical Chemistry 82 (2010) 6422–6429. [7] J. Wang, H.Y. Han, X.C. Jiang, L. Huang, L.N. Chen, N. Li, Analytical Chemistry 84 (2012) 4893–4899. [8] N. Liao, Y. Zhuo, Y.Q. Chai, Y. Xiang, Y.L. Cao, R. Yuan, J. Han, Chemical Communications 48 (2) (2012) 7610–7612. [9] S.Y. Deng, J.P. Lei, Y. Huang, Y. Cheng, H.X. Ju, Analytical Chemistry 85 (2013) 5390–5396. [10] L. Wang, W. Wei, J. Han, Z.F. Fu, Analyst 137 (2012) 735–740. [11] J. Zhang, J.P. Lei, C.L. Xu, L. Ding, H.X. Ju, Analytical Chemistry 82 (2010) 1112–1117. [12] G.F. Jie, P. Liu, S.S. Zhang, Chemical Communications 46 (2010) 1323–1325. [13] H. Dai, C.P. Yang, Y.J. Tong, G.F. Xu, X.L. Ma, Y.Y. Lin, G.N. Chen, Chemical Communications 48 (2012) 3055–3057. [14] S.Y. Deng, J.P. Lei, X.N. Yao, Y. Huang, D.J. Lin, H.X. Ju, Journal of Materials Chemistry C 1 (2013) 299–306. [15] K. Qian, L.F. Luo, H.Z. Bao, Q. Hua, Z.Q. Jiang, W.X. Huang, Catalysis Science & Technology 3 (2013) 679–687. [16] G.R. Zhang, B.Q. Xu, Nanoscale 2 (2010) 2798–2804. [17] L.J. Bai, R. Yuan, Y.Q. Chai, Y.L. Yuan, L. Mao, Y. Wang, Analytica Chimica Acta 698 (2011) 14–19. [18] X.H. Kang, Z.B. Mai, X.Y. Zou, P.X. Cai, J.Y. Mo, Analytical Biochemistry 369 (2007) 71–79. [19] Y. Li, R. Yuan, Y.Q. Chai, Z.J. Song, Electrochimica Acta 56 (2011) 6715–6721.
F. Liu et al. / Analytica Chimica Acta 818 (2014) 46–53 [20] S. Shahrokhian, S. Rastgar, Analyst 137 (2012) 2706–2715. [21] X.H. Kang, Z.B. Mai, X.Y. Zou, P.X. Cai, J.Y. Mo, Analytical Biochemistry 369 (2007) 71–79. [22] H.C. Choi, J.M. Buriak, Chemistry of Materials 12 (2000) 2151–2156. [23] B. Liu, T. Ren, J.R. Zhang, H.Y. Chen, J.J. Zhu, C. Burda, Electrochemistry Communications 9 (2007) 551–557. [24] C.Z. Wang, L.Z. Fan, Z.H. Wang, H.B. Liu, Y.L. Li, S.H. Yang, Y.L. Li, Advanced Materials 19 (2007) 3677–3681. [25] J. Geng, B. Liu, L. Xu, F.N. Hu, J.J. Zhu, Langmuir 23 (2007) 10286–10293. [26] L. Ge, J.X. Yan, X.R. Song, M. Yan, S.G. Ge, J.H. Yu, Biomaterials 33 (2012) 1024–1031. [27] D. Hernández-Santos, M. Díaz-González, M.B. González-García, A. Costa-García, Analytical Chemistry 76 (2004) 6887–6893. [28] J.M. Zen, Y.S. Song, H.H. Chung, C.T. Hsu, A. Senthil Kumar, Analytical Chemistry 74 (2002) 6126–6130. [29] M. Yan, D.J. Zang, S.G. Ge, L. Ge, J.H. Yu, Biosensors and Bioelectronics 38 (2012) 355–361. [30] C.G. Tian, W. Li, Q. Zhang, K. Pan, H.G. Fu, Materials Research Bulletin 46 (2011) 1283–1289.
53
[31] Q. Zhu, R. Yuan, Y.Q. Chai, J. Han, Y. Li, N. Liao, Analyst 138 (2013) 620– 626. [32] T. Ren, J.Z. Xu, Y.F. Tu, S. Xu, J.J. Zhu, Electrochemistry Communications 7 (2005) 5–9. [33] M.M. Richter, Chemical Reviews 104 (2004) 3003. [34] G.F. Jie, B. Liu, J.J. Miao, J.J. Zhu, Talanta 71 (2007) 1476–1480. [35] R.X. Zhang, L.Z. Fan, Y.P. Fang, S.H. Yang, Journal of Materials Chemistry 18 (2008) 4964–4970. [36] B.V. Chikkaveeraiah, A. Bhirde, R. Malhotra, V. Patel, J.S. Gutkind, J.F. Rusling, Analytical Chemistry 81 (2009) 9129–9134. [37] G.D. Liu, Y.Y. Lin, J. Wang, H. Wu, C.M. Wai, Y.H. Lin, Analytical Chemistry 79 (2007) 7644–7653. [38] S.J. Xu, Y. Liu, T.H. Wang, J.H. Li, Analytical Chemistry 83 (2011) 3817– 3823. [39] X. Yu, B. Munge, V. Patel, G. Jensen, A. Bhirde, J.D. Gong, S.N. Kim, J. Gillespie, J.S. Gutkind, F. Papadimitrakopoulos, J.F. Rusling, Journal of the American Chemical Society 128 (2006) 11199–11205. [40] Y. Wan, W.P. Deng, Y. Su, X.H. Zhu, C. Peng, H.Y. Hu, H.Z. Peng, S.P. Song, C.H. Fan, Biosensors and Bioelectronics 30 (2011) 93–99.
