Nanoparticle-Based Sandwich Electrochemical Immunoassay for

Jan 22, 2010 - Key Laboratory of Analysis and Detection for Food Safety (Ministry of ... Signal-On Photoelectrochemical Immunoassay for Aflatoxin B1 B...
5 downloads 0 Views 2MB Size
Download PDF
Anal. Chem. 2010, 82, 1527–1534

Nanoparticle-Based Sandwich Electrochemical Immunoassay for Carbohydrate Antigen 125 with Signal Enhancement Using Enzyme-Coated Nanometer-Sized Enzyme-Doped Silica Beads Dianping Tang,* Biling Su, Juan Tang, Jingjing Ren, and Guonan Chen* Key Laboratory of Analysis and Detection for Food Safety (Ministry of Education), and Fujian Provincial Key Laboratory of Analysis and Detection for Food Safety, Department of Chemistry, Fuzhou University, Fuzhou 350108, China A novel nanoparticle-based electrochemical immunoassay of carbohydrate antigen 125 (CA125) as a model was designed to couple with a microfluidic strategy using antiCA125-functionalized magnetic beads as immunosensing probes.Toconstructtheimmunoassay,thionine-horseradish peroxidase conjugation (TH-HRP) was initially doped into nanosilica particles using the reverse micelle method, and then HRP-labeled anti-CA125 antibodies (HRP-antiCA125) were bound onto the surface of the synthesized nanoparticles, which were used as recognition elements. Different from conventional nanoparticle-based electrochemical immunoassays, the recognition elements of the immunoassay simultaneously contained electron mediator and enzyme labels and simplified the electrochemical measurement process. The sandwich-type immunoassay format was used for the online formation of the immunocomplex in an incubation cell and captured in the detection cell with an external magnet. The electrochemical signals derived from the carried HRP toward the reduction of H2O2 using the doped thionine as electron mediator. Under optimal conditions, the electrochemical immunoassay exhibited a wide working range from 0.1 to 450 U/mL with a detection limit of 0.1 U/mL CA125. The precision, reproducibility, and stability of the immunoassay were acceptable. The assay was evaluated for clinical serum samples, receiving in excellent accordance with results obtained from the standard enzyme-linked immunosorbent assay (ELISA) method. Concluding, the nanoparticle-based assay format provides a promising approach in clinical application and thus represents a versatile detection method. Immunoassay, as a promising approach for selective and sensitive analysis, has recently gained increasing attention.1-4 Various immunoassays have been developed in different fields including environmental monitoring, food safety, and clinical diagnosis.5-11 Nanoparticle-based assays hold great promise in realizing highly sensitive and selective detection at attomolar * To whom correspondence should be addressed. E-mail: dianping.tang@ fzu.edu.cn or [email protected] (D.T.); [email protected] (G.C.). Phone/Fax: +86 591 2286 6135. 10.1021/ac902768f  2010 American Chemical Society Published on Web 01/22/2010

protein concentrations without requiring complex amplification methods including polymerase chain reaction (PCR).12-15 Mirkin and his colleagues reported a nanoparticle-based approach for the detection of free prostate-specific antigen (PSA) at low attomolar concentration by adding an antibody-labeled magnetic microparticle, DNA barcodes, and conjugating a second antibody to the DNA-conjugated gold nanoparticle.16 Liu’s group reported a nanoparticle-based assay for the highly sensitive PSA detection at concentrations as low as 500 attomolar on a single disposable chip using a light scattering method.12 The nanoparticle-based assay allows for the detection of low concentration levels with signal amplification and reduces the sample pretreatment requirement owing to the presence of magnetic particles.17-19 Usually, the nanoparticle-based assay consisted of two kinds of particles: (i) magnetic microparticles coated with monoclonal antibodies as immunosensing probe and (ii) multifunctional nanoparticles (including nanogold particles) decorated with polyclonal antibodies or unique “barcode” DNA sequences as recognition elements. Iron oxide (Fe3O4) nanoparticles, as a well-known (1) Thangawng, A.; Howell, P., Jr.; Richards, J.; Erickson, J.; Ligler, F. Lab Chip 2009, 9, 3126. (2) Golden, J.; Kim, J.; Erickson, J.; Hilliard, L.; Howell, P.; Anderson, G.; Nasir, M.; Ligler, F. Lab Chip 2009, 9, 1942. (3) Rieger, M.; Cervino, C.; Sauceda, J.; Niessner, R.; Knopp, D. Anal. Chem. 2009, 81, 2373. (4) Wolter, A.; Niessner, R.; Seidel, M. Anal. Chem. 2008, 80, 5854. (5) Lai, G.; Yan, F.; Ju, H. Anal. Chem. 2009, 81, 9730. (6) Liu, H.; Fu, Z.; Yang, Z.; Yan, F.; Ju, H. Anal. Chem. 2008, 80, 5654. (7) Kim, J.; Anderson, G.; Erickson, J.; Golden, J.; Nasir, M.; Ligler, F. Anal. Chem. 2009, 81, 5426. (8) Wojciechowski, J.; Shriver-Lake, L.; Yamaguchi, M.; Fureder, E.; Pieler, R.; Schamesberger, M.; Winder, C.; Ligler, F. Anal. Chem. 2009, 81, 519. (9) Ligler, F. S. Anal. Chem. 2009, 81, 519. (10) Erickson, J.; Ligler, F. Nature 2008, 456, 178. (11) Johnson, B.; Delehanty, J.; Lin, B.; Ligler, F. Anal. Chem. 2008, 80, 2113. (12) Goluch, E.; Nam, J.; Georganopoulou, D.; Chiesl, T.; Shaikh, K.; Ryu, K.; Barron, A.; Mirkin, C.; Liu, C. Lab Chip 2006, 6, 1293. (13) Kim, D.; Daniel, W.; Mirkin, C. Anal. Chem. 2009, 81, 9183. (14) Dhar, S.; Daniel, W.; Giljohann, D.; Mirkin, C.; Lippard, S. J. Am. Chem. Soc. 2009, 131, 14652. (15) Hurst, S.; Hill, H.; Macfarlane, R.; Wu, J.; Dravid, V.; Mirkin, C. Small 2009, 5, 2156. (16) Nam, J.; Thaxton, C.; Mirkin, C. Science 2003, 301, 1884. (17) Banholzer, M.; Qin, L.; Millstone, J.; Osberg, K.; Mirkin, C. Nat. Protoc. 2009, 4, 838. (18) Daniel, W.; Han, M.; Lee, J.; Mirkin, C. J. Am. Chem. Soc. 2009, 131, 6362. (19) Millstone, J.; Stoeva, S.; Lee, J.; Shaikh, K.; Mirkin, C.; Liu, C. Biosens. Bioelectron. 2009, 24, 2397.

Analytical Chemistry, Vol. 82, No. 4, February 15, 2010

1527

hard magnetic material, were extensively applied in the nanoparticle-based assays since they have good biocompatibility and can be separated very readily from reaction mixtures with the help of an external magnetic field.20-23 The functionalized probes could pull antibodies bound to magnetic nanoparticles from one laminar flow path to another by applying a local magnetic field gradient and selectively removing them from flowing biological fluids without any washing step. Moreover, it has been successfully applied as matrixes for the electrochemical detection of biomarkers in our past works.24-27 So, Fe3O4 nanoparticles were used for the fabrication of the nanoparticle-based immunosensing probes in this study. For the successful development of nanoparticle-based electrochemical immunoassays, there are two basal issues to realize the application of electrochemical technique.28 First, signaling amplification and noise reduction are very crucial for obtaining low detection limits. In general, they are achieved by using an indicator system that results in the amplification of the measured product including enzyme label and nanolabels.29-32 Nanogold particles labeled with HRP-bound antibodies are commonly exploited as signal transduction tools in the electrochemical immunoassays.33,34 The traditional method often exhibited low signaling; however, there is because of sterical reasons usually a ratio of 1:1 for enzyme and detection antibody used. To improve the labeled method, we utilize HRP-doped silica nanoparticles for the label of HRP-bound antibodies. With the use of this method, there are many HRP molecules inside and outside of the synthesized bionanolabels. The carried HRP molecules will be entered and participated in the catalytic reaction, suspecting, when one antibody among them reacts with the corresponding antigen. This high efficiency makes them especially suitable for ultrasensitive bioanalysis with signal amplification. Another key factor is to provide a good pathway of electron transfer between the redox center of enzymes and the electrode surface.35-38 Recently, we developed a reusable electrochemical immunoassay for aflatoxin B1 in food using a multifunctional magnetic bead as an immunosensing probe with a magnetic CoFe2O4 nanoparticle as core, Prussian blue nanoparticle-doped (20) Yang, Z.; Liu, H.; Zong, C.; Yan, F.; Ju, H. Anal. Chem. 2009, 81, 5484. (21) Corchero, J.; Villaverde, A. Trends Biotechnol. 2009, 27, 468. (22) Schubayev, V.; Pisanic, T., II; Jin, S. Adv. Drug Delivery Rev. 2009, 61, 467. (23) Xie, J.; Huang, J.; Li, X.; Sun, X.; Chen, X. Curr. Med. Chem. 2009, 16, 1278. (24) Tang, D.; Yuan, R.; Chai, Y. Clin. Chem. 2007, 53, 1323. (25) Tang, D.; Yuan, R.; Chai, Y.; An, H. Adv. Funct. Mater. 2007, 17, 976. (26) Tang, D.; Yuan, R.; Chai, Y. J. Phys. Chem. 2006, 110, 11640. (27) Tang, D.; Zhong, Z.; Niessner, R.; Knopp, D. Analyst 2009, 134, 1554. (28) Tushar, R.; Amit, A.; Shuming, N. Anal. Chem. 2006, 78, 5627. (29) Yu, X.; Munge, B.; Patel, V.; Jensen, G.; Bhirde, A.; Gong, J.; Kim, S.; Gillespie, J.; Gutkind, J.; Papadimitrakopoulos, F.; Rusling, J. J. Am. Chem. Soc. 2006, 128, 11199. (30) Sun, N.; McMullan, M.; Papakonstantinou, P.; Gao, H.; Zhang, X.; Mihailovic, D.; Li, M. Anal. Chem. 2008, 80, 3593. (31) Das, J.; Aziz, M.; Yang, H. J. Am. Chem. Soc. 2006, 128, 16022. (32) Wang, J.; Liu, G.; Engelhard, M.; Lin, Y. Anal. Chem. 2006, 78, 6974. (33) Shim, S.; Lim, D.; Nam, J. Nanomedicine 2008, 3, 215. (34) Kim, E.; Stanton, J.; Korber, B.; Krebs, K.; Bogdan, D.; Kunstman, K.; Wu, S.; Wolinsky, S. Nanomedicine 2008, 3, 293. (35) Lanci, M.; Remy, M.; Kaminsky, W.; Mayer, J.; Sanford, M. J. Am. Chem. Soc. 2009, 131, 15618. (36) Palacin, T.; Le Khanh, H.; Jousselme, B.; Jeqou, P.; Filoramo, A.; Ehli, C.; Guldi, D.; Campidelli, S. J. Am. Chem. Soc. 2009, 131, 15394. (37) Dai, Z.; Yan, F.; Yu, H.; Hu, X.; Ju, H. J. Immunol. Methods 2004, 287, 13. (38) Wu, J.; Yan, F.; Tang, J.; Zhai, C.; Ju, H. Clin. Chem. 2007, 53, 1495.

1528

Analytical Chemistry, Vol. 82, No. 4, February 15, 2010

silica as shell, and gold nanoparticles labeled with HRP-bound anti-AFB1 antibodies as recognition elements.27 Although the doped Prussian blue could enhance the electron communication between the immobilized HRP and the base electrode, the reactive time of the redox was relatively long, and the sensitivity was not so high (i.e., nanoamps of current response). The reason might be the fact that the electron mediator (Prussian blue) and bioactive HRP enzyme were immobilized on two sets of nanoparticles, respectively, which lengthened the pathway of electron transfer and lessened the redox of the HRP-H2O2 system. To solve this problem, we prepared a novel multifunctional silica nanoparticle, which simultaneously contained bioactive HRP enzyme and electroactive thionine mediator, and the synthesized nanosilica particles were labeled onto HRP-bound antibodies as recognition elements in this paper. The aim of this work is to exploit an advanced nanoparticlebased electrochemical immunoassay for the detection of carbohydrate antigen 125 (CA125) as a model analyte in clinical diagnosis. The magnetic immunosensing probe not only favors the rapid separation and purification of sandwich-type bionanocomposites but also facilitates the fabrication of the microfluidic immunosensing interface via using an external magnet. The electrochemical signaling of the captured multifunctional silica nanoparticles could be rapidly and adequately achieved because of the catalytic reaction of the carried HRP relative to the H2O2-thionine system. The performance and factors influencing the nanoparticle-based assay are investigated and discussed in the following sections. EXPERIMENTAL SECTION Materials. Mouse monoclonal anti-CA125 (primary antibodies), HRP-labeled anti-CA125 (HRP-anti-CA125, secondary antibodies), CA125, and HRP (EC 1.11.1.7, RZ > 3.0, A > 250 U/mg) were purchased from Sigma (St. Louis, MO). Thionine (TH), gultaraldehyde, bovine serum albumin (BSA, 96%-99%), (3glycidyloxypropyl) trimethoxysilane (C9H20O5Si, GOPS), tetramethoxysilane (TEOS), and bis-(2-ethylhexyl) sodium sulfosuccinate (AOT) were obtained from Sigma-Aldrich. Cyclohexane, n-hexanol, and ammonium hydroxide (25 wt %) were purchased from Merck (Darmstadt, Germany). Magnetic Fe3O4 nanoparticles (particle size: 100 nm) in an aqueous suspension with a concentration of 25 mg/mL were obtained from Chemicell GmbH (Berlin, Germany). All other reagents were of analytical grade and were used without further purification. Deionized and distilled water was used throughout the study. The 0.1 M phosphate-buffered saline (PBS, pH 7.4) was prepared by adding 12.2 g of K2HPO4, 1.36 g of KH2PO4, and 8.5 g of NaCl in 1000 mL of deionized water. Fabrication of Immunosensing Probes. Prior to the experiment, Fe3O4 nanoparticles were separated using an external magnet, and dried in the vacuum at 80 °C for 1 h. Following that, 250 mg of Fe3O4 nanoparticles was added into 5.0 mL of 5% GOPS (v/v) in dry toluene and stirred with 500 rpm for 12 h at room temperature (RT).39 With the aid of an external magnet, the GOPSfunctionalized Fe3O4 nanoparticles were separated and purified. Afterward, the synthesized nanoparticles were washed three times with toluene and ethanol solution, respectively. The purified nanoparticles were dried and activated in an oven at 80 °C for 1 h (39) Wu, Y.; Chen, C.; Liu, S. Anal. Chem. 2009, 81, 1600.

Scheme 1. Construction of the Immunosensing Probe and Recognition Element (RE) and Measurement Protocol of the Nanoparticle-Based Electrochemical Immunoassay with a Sandwich-Type Format

under a nitrogen atmosphere. Afterward, 100 µL of anti-CA125 (500 ng/mL) was injected into the functionalized Fe3O4 aqueous solution with the concentration of 25 mg/mL. The mixture was slightly stirred for 12 h at 4 °C to make the -NH2 groups of antiCA125 antibodies conjugate to epoxy groups. The obtained antiCA125-Fe3O4 bionanoparticles (i.e., immunosensing probes, Scheme 1) were incubated with 10 mg/mL BSA-PBS at 37 °C for 1 h to block the unreacted and nonspecific sites. Finally, the as-prepared bionanoparticles were stored in pH 7.4 PBS with a final concentration of 25 mg/mL at 4 °C when not in use. Preparation of Thionine-HRP Conjugation (TH-HRP). Amounts of 500 µL of HRP (1.5 × 10-3 mol/L) and 500 µL of thionine (3.0 × 10-5 mol/L) were initially dissolved into 1 mL of 0.01 M PBS, and the pH was adjusted to 9.5 by using 10 wt % K2CO3, and then 100 µL of gultaraldehyde solution was added into the mixture. After being stirred for 2 h, the mixture was adjusted to pH 7.0 by using 1.0 M NaH2PO4. The unbound thionine was removed using ultrafiltration for 12-15 times until the peak corresponding to thionine in the elution disappeared. Finally, the obtained TH-HRP was prepared into 2 mL of aqueous solution. Synthesis of Recognition Elements. Scheme 1 displays the construction of the nanosilica-based recognition elements (RE). The TH-HRP-doped silica nanoparticles were prepared at RT using a reverse micelle method according to these literatures with some modification,40,41 which contained the following three steps: (i) 5.3 g of Triton X-100, 5.0 mL of n-hexanol, and 1.0 mL of TH-HRP were added into 20 mL of cyclohexane solution in turn and continuously stirred for 20 min, (ii) 2.5 mL of TEOS and 1.8 mL of NH3 · H2O (25 wt %) were slowly dropped into the stirring mixture and vigorously stirred for 24 h, and (iii) 5.0 mL of acetone was added, and the mixture was centrifuged and washed for five times with ethanol and water. The obtained TH-HRP-doped silica nanoparticles (designed as SiO2(TH-HRP)) were used for the fabrication of recognition elements. Similar with the preparation process of the immunosensing probes, the functionalization of the SiO2(TH-HRP) was carried out using the amine-epoxy interaction. To maintain the bioactivity of the doped HRP, the drying and activation of the SiO2(TH-HRP) (40) Kumar, R.; Maitra, A.; Patanjali, P.; Sharma, P. Biomaterials 2005, 26, 6743. (41) Zhong, Z.; Li, M.; Xiang, D.; Dai, N.; Qing, Y.; Wang, D.; Tang, D. Biosens. Bioelectron. 2009, 24, 2246.

were performed at 40 °C. The process included the synthesis of epoxy-functionalized SiO2(TH-HRP) and the modification of HRP-anti-CA125 on the epoxy-functionalized SiO2(TH-HRP). Finally, the synthesized HRP-anti-CA125-SiO2(TH-HRP) bionanoparticles were fixed with a concentration of 25 mg/mL and still stored at 4 °C when not in use. Electrochemical Measurement. Electrochemical measurements were carried out with a microAutoLab Type III system (Eco Chemie, The Netherlands) in combination with a flow-through detection cell. The detection system comprised an indium-tin oxide (ITO, 5 wt % In2O3 + SnO2) working electrode, a platinum wire as auxiliary electrode, and a Ag/AgCl reference electrode (the schematic illustration is shown in our previous report27). The flow injection system consists of a six-way connected valve equipped with a 1 mL syringe pump and connected through a Teflon tubing to the flow cell. The analytical flow stream entered from the other side into the center of the flow cell at 0.5 mL/ min. The ITO electrode was installed at the bottom of the cell, and the bar electromagnet was set under the ITO electrode. All the experiments were carried out in pH 7.4 PBS containing 1.0 mM H2O2 with a potential sweep rate of 50 mV/s at RT. With the sandwich-type immunoassay format, the carried HRP on the recognition elements exhibited a stable current at +120 mV (vs Ag/AgCl). The changes in cathodic currents were collected and recorded versus the CA125 concentrations. Scheme 1 represents the sandwich-type immunoassay protocol and measurement method. During the process, the collection, incubation, and detection of the nanoparticle-based immunoassay were performed in the flow cell. The measurement process consisted of five steps: (i) 50 µL of immunosensing probes (25 mg/mL) was injected into the cell and collected on the ITO surface with an external magnet; (ii) 50 µL of CA125 standards/ specimens with various concentrations was added to the cell and incubated for 18 min without the magnetic field to form the immunocomplexes on the Fe3O4 surface; (iii) the immunocomplexes were accumulated and washed with the external magnetic field; (iv) 100 µL of recognition elements (25 mg/mL) was flowed into the cell and incubated for another 18 min without the magnetic field to construct a sandwich-type immunocomplex; (v) the H2O2-PBS (pH 7.4) was flowed through the cell with the Analytical Chemistry, Vol. 82, No. 4, February 15, 2010

1529

Figure 1. TEM images of (a) Fe3O4 and (b) HRP-TH-doped SiO2 nanoparticles. Insets: TEM images of anti-CA125-Fe3O4 and HRP-antiCA125-SiO2(TH-HRP), respectively.

Figure 2. UV-vis absorption spectra of (a) anti-CA125, (b) HRP, (c) thionine, (d) nano-Fe3O4 colloids, (e) anti-CA125-Fe3O4 colloids, (f) nano-SiO2 colloids, and (g) HRP-TH-doped nano-SiO2 colloids.

magnetic field, and cyclic voltammograms were registered at the flow stop. After each step, the cell should be washed with pH 7.4 PBS. RESULTS AND DISCUSSION Characteristics of the Nanoparticle-Based Immunoassay. In this study, the nanoparticle-based immunoassay contains two kinds of nanoparticles including anti-CA125-Fe3O4 and HRP-antiCA125-SiO2(TH-HRP). In the presence of the analyte (CA125 antigen), these two sets of nanoparticles form a sandwich-type immunocomplex, and the immunocomplex increases with the increment of the CA125 concentration in the sample, thus fabricating a three-dimensional network with the interleaving of Fe3O4 and SiO2 nanoparticles. With an external magnet, the immunocomplex is captured on the surface of the ITO electrode. The electrochemical detection is based on the immobilized HRP on the silica nanoparticles toward the reduction of H2O2 with the help of the doped (thionine) electron mediator. The more the amount of CA125 antigens in the sample was, the more the amount of the cross-linked HRP-anti-CA125-SiO2(TH-HRP) in the immunocomplex was; thus, the catalytic current was increased. To successfully carry out the nanoparticle-based immunoassay, the preparation, modification, and functionalization of the nanoparticles were very crucial. In this paper, anti-CA125 and HRP-antiCA125 were immobilized on the surface of nanoparticles through the epoxy-amino reaction.42,43 Figure 1 shows the transmission (42) Hu, P.; Huang, C.; Li, Y.; Ling, J.; Liu, Y.; Fei, Y.; Xie, J. Anal. Chem. 2008, 80, 1819. (43) Wu, Y.; Chen, C.; Liu, S. Anal. Chem. 2009, 81, 1600.

1530

Analytical Chemistry, Vol. 82, No. 4, February 15, 2010

electron microscopy (TEM, H600, Hitachi Instrument, Japan) images of Fe3O4 and SiO2 nanoparticles before and after modified with the biomolecules, respectively. Seen from Figure 1, the mean sizes of the formed bionanoparticles were 100 and 60 nm for Fe3O4 and SiO2(TH-HRP), respectively. Moreover, the surface of these nanoparticles became rougher after modification, indicating antiCA125 and HRP-anti-CA125 could be linked to the nanoparticles (inset of Figure 1). To further verify the formation of the functional nanoparticles, we used UV-vis absorption spectrometry (Hitachi Instrument, Japan) to investigate the fabricated process of the bionanoparticles (Figure 2). Figure 2a-c shows the absorption spectra of antiCA125 antibodies, pure HRP, and thionine, respectively. After magnetic Fe3O4 nanoparticles were modified with anti-CA125 antibodies, a 278 nm absorption peak was observed in comparison with that of pure Fe3O4 nanoparticles (Figure 2, curves d and e), suggesting antibodies could be covalently bound to Fe3O4 nanoparticles. Seen from Figure 2, curves f and g, the absorption peaks of HRP and thionine could be simultaneously appeared at the SiO2(TH-HRP) nanoparticles. This result indicated that the HRP-thionine conjugation could be doped into the silica particles using the reverse micelle method. Furthermore, the 268 nm absorption peak was achieved when HRP-anti-CA125 biomolecules were conjugated to the SiO2(TH-HRP) nanoparticles. The absorption wavelength was obviously less than that of pure antiCA125 (Figure 2a). The reason might be attributed to the interoverlapping of 278 nm (for HRP) and 250 nm (for thionine). These results suggested that anti-CA125-Fe3O4 and HRP-anti-

Figure 3. (A) Cyclic voltammograms of (a) anti-CA125-Fe3O4, (b) CA125-anti-CA125-Fe3O4, (c) CA125-anti-CA125-Fe3O4 after being reacted with HRP-anti-CA125-SiO2(TH-HRP) in pH 7.4 PBS, and (d) complex c in pH 7.4 PBS containing 1.0 mM H2O2. (B) Comparison of current responses of the developed immunoassay using various detection antibodies: (a) HRP-anti-CA125-SiO2(TH-HRP) and (b) HRP-antiCA125.

CA125-SiO2(TH-HRP) could be constructed via using the reverse micelle method and the epoxy-amino reaction. Electrochemical Characteristics of Nanoparticle-Based Immunoassay. Figure 3A displays the cyclic voltammograms of the nanoparticle-based immunoassay for the various steps at 50 mV/s in pH 7.4 PBS. The formed immunocomplex was captured on the surface of the ITO electrode with an external magnet after being reacted with 100 U/mL CA125. No peak was observed for the anti-CA125-Fe3O4 and CA125-anti-CA125-Fe3O4 (Figure 3, curves a and b). The high background current after the antigen-antibody interaction was mainly attributed to the formation of the immunocomplex on the nanoparticle surface, which hindered the electron transfer. After HRP-antiCA125-SiO2(TH-HRP) bionanoparticles were reacted with the CA125-anti-CA125-Fe3O4 bionanoparticles, however, the formed immunocomplex exhibited a couple of redox peak at +120 and 220 mV (Figure 3c). The result indicated that the doped thionine, as a good electron mediator, still remained their redox properties, which could provide a fast pathway of electron transfer for the reduction of H2O2. Seen from Figure 3d, an obvious redox reaction was appeared with a distinct increase of the reduction current and a decrease of the oxidation current upon the addition of H2O2 in pH 7.4 PBS. Moreover, the reduction current (ipc) increased with the increment of CA125 concentration in the sample. The catalytic current mainly derived from the immobilized HRP toward the reduction of H2O2 with the aid of the doped thionine as mediator.44,45 Thus, we might quantitatively evaluate the concentration of CA125 according to the reduction current. To clarify the advantage of the developed immunoassay using multifunctional silica nanoparticles in place of the conventional nanogold particles, we fabricated two kinds of recognition elements including HRP-anti-CA125-SiO2(TH-HRP) and HRP-antiCA125-nanogold, which were used for the detection of CA125 by using the sandwich-type immunoassay and the same antiCA125-Fe3O4 probes. The preparation and labeling of nanogold particles were according to our previous report46 and measured in pH 7.4 thionine-PBS. As shown in Figure 3B, HRP-antiCA125-SiO2(TH-HRP) exhibited higher response and sensitivity (44) Tang, D.; Yuan, R.; Chai, Y. Anal. Chem. 2008, 80, 1582. (45) Tang, D.; Ren, J. Anal. Chem. 2008, 80, 8064. (46) Tang, D.; Sauceda, J.; Lin, Z.; Ott, S.; Basova, E.; Goryacheva, I.; Biselli, S.; Niessner, R.; Knopp, D. Biosens. Bioelectron. 2009, 25, 514.

(i.e., slope) than that of using HRP-anti-CA125-nanogold as recognition elements. The reason might be attributed to the proximity of the TH and the fact that the HRP molecules were immobilized inside and outside of silica nanoparticles, whereas they were only on the surface of gold nanoparticles. We might speculate that the immobilized amount of the HRP molecules for HRP-anti-CA125-SiO2(TH-HRP) was more than that of HRP-antiCA125-nanogold and thus exhibited high catalytic efficiency toward the reduction of H2O2. In addition, the synthesized HRP-anti-CA125-SiO2(TH-HRP) could avoid the addition of electron mediator. Optimization of Experimental Conditions. To achieve an optimal electrochemical signaling, the ratio of HRP and thionine doped into silica nanoparticles was very crucial owing to the almost same amount of the HRP-anti-CA125 on the nanosilica surface. Various molar ratios of HRP and thionine including 4:1, 3:1, 2:1, 1:1, 1:2, 1:3, and 1:4 (w/w) were used for the preparation of recognition elements, which was applied for the detection of 100 U/mL CA125. Seen from Figure 4a, the optimal current response was obtained at the molar ratio of 2:1. Higher or lower concentration of thionine as an electron mediator would affect the catalytic performance of the immobilized HRP toward the reduction of H2O2. Thus, 2:1 molar ratio of HRP and thionine was selected for the preparation of recognition elements. In the sandwich-type immunoassays, temperature and time for the antigen-antibody interaction greatly influenced the sensitivity of the developed immunoassay. Considering the practical application of the proposed system in clinical immunoassays, all experiments were carried out at RT (25 ± 1.0 °C). Following that, we investigated the effect of various incubation times (from 5 to 30 min) on the current of the immunoassay toward 100 U/mL CA125. As shown in Figure 4b, the reduction current increased with the increasing incubation time and trended to level off after 18 min, indicating the optimal formation of the sandwich-type immunocomplexes. Longer incubation time could not improve the response. So, 18 min of incubation time was used for the detection of CA125 in this study. Calibration Curve of the Nanoparticle-Based Immunoassay. Under the optimal conditions, a sandwich-type immunoassay format was employed for the detection of CA125 standards using the biomolecules-functionalized silica nanoparticles (HRP-antiAnalytical Chemistry, Vol. 82, No. 4, February 15, 2010

1531

Figure 4. Effects of (a) molar ratios of HRP and thionine for the preparation of HRP-TH-doped silica nanoparticles and (b) incubation time on the current responses in the presence of 100 U/mL CA125.

Figure 5. Calibration curves of the electrochemical immunoassays using various detection antibodies: (a) HRP-antiCA125-SiO2(TH-HRP), (b) HRP-anti-CA125-SiO2(TH), and (c) anti-CA125-SiO2(TH-HRP) in pH 7.4 PBS containing 1.0 mM H2O2 after incubation with CA125 standards.

CA125-SiO2(TH-HRP)) as tracer and H2O2 as enzyme substrate. The catalytic current increased with the increment of CA125 concentration in the sample (Figure 5a), which exhibits a wide linear dynamic range of 0.1-450 U/mL and a detection limit (LOD) of 0.1 U/mL CA125. The linear regression equation is ipc (µA) ) 12.3 + 7.7 × Ln C[CA125] (U/mL, R2 ) 0.943). When the CA125 concentration was higher than 300 U/mL, an appropriate dilution of sample was necessary in the preincubation step. Since the cutoff value of CA125 in diagnostic is 35 U/mL, the sensitivity of the developed immunoassay was enough to practical application. For comparison, we prepared another two sets of recognition elements, including HRP-anti-CA125-labled thionine-doped silica nanoparticles (HRP-anti-CA125-SiO2(TH)) and anti-CA125-labled thionine and HRP-doped silica nanoparticles (antiCA125-SiO2(TH-HRP)), which were used for the determination of CA125 standards using the same assay protocol. The linear ranges and LODs toward CA125 were 2.5-230 U/mL with an LOD of 2.5 U/mL and 4.5-200 U/mL with an LOD of 4.5 U/mL for HRP-anti-CA125-SiO2(TH) and anti-CA125-SiO2(TH-HRP), respectively (Figure 5, curves b and c). These results adequately suggested that the synthesized multifunctional silica nanoparticles could greatly improve the sensitivity and working range of the 1532

Analytical Chemistry, Vol. 82, No. 4, February 15, 2010

electrochemical immunoassay. In addition, we also compared the properties of the immunoassay with other CA125 immunoassays reported previously including standard enzyme-linked immunosorbent assay (ELISA). Seen from Table 1, the immunoassay exhibits wide linear range and low detection limit. Significantly, the developed method was capable of continuously carrying out all steps in less than 40 min for one sample including 36 min of incubation, separation, and detection, which is shorter than that of the commercial ELISA. Precision, Reproducibility, Selectivity, and Stability. The precision and reproducibility of the immunoassay were evaluated by using the variation coefficient (CV) of the intra- and interassays. Three CA125 standards (including 1.0, 150, and 350 U/mL CA125) were used as examples, and each CV represents the average value of three assays. Analyzed from experimental results, the CVs of the intra-assay were 6.9%, 5.7%, and 6.2% at 1.0, 150, and 350 U/mL CA125, respectively, whereas the CVs of the interassay using different nanoparticles with various batches were 7.2%, 5.8%, and 6.5% at 1.0, 150, and 350 U/mL CA125, respectively. Thus, the precision and reproducibility of the immunoassay were acceptable. To investigate the differences in response of the immunoassay to interference degree or crossing recognition level, carcinoembryonic antigen (CEA), R-fetoprotein (AFP), human chorionic gonadotropin (HCG), and prostate-specific antigen (PSA) with various concentrations were injected into the detection system, respectively. The current responses to each type of antigen were recorded, and the results are described in Table 2. Seen from Table 2, the interference degree of variability between lineagedifferent tumor markers is acceptable. When normal (negative) serum samples were analyzed using the developed immunoassay as the control tests, the current shift before and after the (47) http://www.biocompare.com/ProductDetails/1044399/Human-CA-125-EIAKit.html. (48) Sok, D.; Clarizia, L.; Farris, L.; McDonald, M. Anal. Bioanal. Chem. 2009, 393, 1521. (49) Yang, Z.; Xie, Z.; Liu, H.; Yan, F.; Ju, H. Adv. Funct. Mater. 2008, 18, 3991. (50) Suwansa-ard, S.; Kanatharana, P.; Asawatreratanakui, P.; Wongkittisuksa, B.; Limsakul, C.; Thavarungkul, P. Biosens. Bioelectron. 2009, 24, 3436. (51) Bangar, M.; Shirale, D.; Chen, W.; Myung, N.; Mulchandani, A. Anal. Chem. 2009, 81, 2168. (52) Tang, D.; Yuan, R.; Chai, Y. Analyst 2008, 133, 933. (53) Wu, J.; Yan, Y.; Yan, F.; Ju, H. Anal. Chem. 2008, 80, 6072. (54) Fu, Z.; Yan, F.; Liu, H.; Lin, J.; Ju, H. Biosens. Bioelectron. 2008, 23, 1422. (55) Fu, Z.; Yang, Z.; Tang, J.; Liu, H.; Yan, F.; Ju, H. Anal. Chem. 2007, 79, 7376. (56) Fu, Z.; Liu, H.; Ju, H. Anal. Chem. 2006, 78, 6999.

Table 1. Analytical Performances of Various CA125 Immunosensors and Immunoassays measurement protocol

linear range (U/mL)

CA125 ELISA kit fluoroimmunoassay chemiluminescent immunoassay surface plasmon resonance immunosensor electrochemical immunosensor quartz crystal microbalance immunoassay electrochemical immunosensor array chemiluminescent immunoassay chemiluminescent immunoassay chemiluminescent immunoassay electrochemical immunoassay

LOD (U/mL)

15-400

assay time

5.0 1.5 0.17 0.1 1.0 0.5 0.03 0.7 0.15 1.0 0.1

0.5-400 0.1-40 e1000 1.5-180 0.11-13 1.0-50 0.5-80 5.0-100 0.1-450

200 min 20 min 12 min 90 min e5 min 18 min 37 min 35 min 40 min

refs 47 48 49 50 51 52 53 54 55 56 this work

Table 2. Interference/Crossing Recognition Level of the Developed Immunoassays assay time; current change (µA)b CA125 + crossing reagent

a

CA125 CA125 + CEA CA125 + AFP CA125 + HCG CA125 + PSA negative serum positive serumc

1

2

3

4

mean ± SD (µA)

48.1 49.7 53.2 56.4 52.4 3.4 58.9

49.4 50.2 51.3 53.1 53.1 2.1 61.3

47.2 48.9 49.4 53.3 51.4 1.9 57.9

48.9 54.3 52.7 50.6 52.9 2.7 58.2

48.4 ± 0.8 50.8 ± 2.1 51.7 ± 1.5 53.4 ± 2.1 52.5 ± 0.7 2.5 ± 0.6 59.1 ± 1.3

a All the concentrations of CA125 and crossing reagents were 100 U/mL (or ng/mL) in the detection solution. b Each sample was assayed in triplicate. c The concentration of positive serum sample was 248 U/mL CA125, which was obtained by the standard ELISA.

incubation was very low (∆i < 4.0 µA) in contrast to the results obtained when positive serum samples were assayed. These results revealed the significant response difference between the lineage-specific recognition and the nonspecific adsorption. When the immunosensing probes and recognition elements were not in use, they could be stored in pH 7.4 PBS containing 0.1% NaN3 at 4 °C for at least 1 month without obvious signal change. We speculate that the long-term stability mainly attributed to the following two issues: (i) anti-CA125 and HRP-anti-CA125 molecules were covalently immobilized on the surface of the nanoparticles, and (ii) the synthesized method could efficiently prevent the leakage of the doped HRP and thionine from recognition elements. Evaluation of Clinical Specimens and Intralaboratory Validation. To further investigate the analytical reliability and possible application of the electrochemical immunoassay for testing real samples, we collected 30 clinical serum specimens from epithelial ovarian carbohydrate patients according to the rules of the local ethical committee, which were gifted from Chongqing Institute of Cancer Prevention and Cure, China. Prior to measurement, these samples were gently swirled at room temperature (note: all handling and processing were performed carefully, and all tools in contact with patient specimens and immunoreagents were disinfected after use) and then evaluated by using the electrochemical immunoassay and referenced ELISA method, respectively. The experimental results are summarized in Figure 6. The regression equation (linear) for these data is as follows: y ) 2.06 + 0.936x (R2 ) 0.998) (y-axis, by the developed immunoassay; x-axis, by referenced ELISA). These data shows an acceptable agreement with these results given by the two methods. Thus, we might utilize the nanoparticle-based electrochemical immunoassay for the preliminary determination of CA125 in clinical diagnostics.

Figure 6. Evaluation of clinical serum samples using the electrochemical immunoassay and the referenced ELISA method (error bar: 5%).

CONCLUSIONS In summary, this manuscript describes a sandwich-type electrochemical immunoassay for the detection of CA125 using anti-CA125-coated magnetic beads for target capture and localization to the electrode and HRP-anti-CA125-coated silica beads containing entrapped HRP and thionine for signal enhancement. Highlights of the developed immunoassay are as follows: (i) the recognition elements simultaneously contained the bioactive enzyme and electron mediator, avoiding the addition of electroactive materials in the detection solution; (ii) a large number of HRP enzymes were simultaneously immobilized outside and inside of silica nanoparticles, amplifying the sensitivity of the immunoassay. Meanwhile, the proposed method was simple, rapid, and without the need of sample preconcentration. Including, the developed method could greatly favor the nanoparticle-based assay with high sensitivity and selectivity to successfully apply for electrochemical immunoassays in clinical diagnosis. Analytical Chemistry, Vol. 82, No. 4, February 15, 2010

1533

ACKNOWLEDGMENT This work was jointly supported by the National Basic Research Program of China (2010CB732403), the National Natural Science Foundation of China (20877019 and 20735002), the Key Natural Sciences Foundation of Fujian Province, China (D0520001), the Key Program of Science and Technology Department of Fujian Province, China (2007Y0026), NTU-MOE Academic Research

1534

Analytical Chemistry, Vol. 82, No. 4, February 15, 2010

Funds (RG65/08), and the High-Qualified Talent Funding of Fuzhou University (0460-022275).

Received for review December 5, 2009. Accepted January 12, 2010. AC902768F