Magneto-Controlled Graphene Immunosensing Platform for

Jun 3, 2011 - The assay was based on the catalytic reduction of H2O2 at the .... CEA standards (0, 5, 10, 20, 40, and 80 ng mL–1) and AFP standards ...
0 downloads 0 Views 4MB Size
ARTICLE pubs.acs.org/ac

Magneto-Controlled Graphene Immunosensing Platform for Simultaneous Multiplexed Electrochemical Immunoassay Using Distinguishable Signal Tags Juan Tang,† Dianping Tang,*,† Reinhard Niessner,‡ Guonan Chen,† and Dietmar Knopp*,‡ †

Key Laboratory of Analysis and Detection for Food Safety (Fujian Province & Ministry of Education of China), Department of Chemistry, Fuzhou University, Fuzhou 350108, People’s Republic of China ‡ Chair for Analytical Chemistry, Institute of Hydrochemistry, Technische Universit€at M€unchen, Marchioninistrasse 17, D-81377 M€unchen, Germany

bS Supporting Information ABSTRACT: A novel flow-through multiplexed immunoassay protocol for simultaneous electrochemical determination of carcinoembryonic (CEA) and alpha-fetoprotein (AFP) in biological fluids was designed using biofunctionalized magnetic graphene nanosheets (MGO) as immunosensing probes and multifunctional nanogold hollow microspheres (GHS) as distinguishable signal tags. The probes were fabricated by means of co-immobilization of primary anti-CEA (Ab1) and anti-AFP (Ab2) antibodies on the Fe3O4 nanoparticle-coated graphene nanosheets (MGO-Ab1,2). The reverse-micelle method was used for the synthesis of distinguishable signal tags by encapsulation of horseradish peroxide (HRP)-thionine and HRP-ferrocene into nanogold hollow microspheres, respectively, which were utilized as labels of the corresponding GHS-Ab1 and GHS-Ab2. A sandwich-type immunoassay format was employed for the online detection of CEA and AFP by coupling a flow-through detection cell with an external magnet. The assay was based on the catalytic reduction of H2O2 at the various peak potentials in the presence of the corresponding mediators. Experimental results revealed that the multiplexed electrochemical immunoassay enabled the simultaneous monitoring of AFP and CEA in a single run with wide working ranges of 0.01200 ng mL1 for AFP and 0.0180 ng mL1 for CEA. The detection limits (LODs) for both analytes at 1.0 pg mL1 (at 3sB) were very low. No obvious nonspecific adsorption and cross-talk were observed during a series of analyses to detect target analytes. Intraassay and interassay coefficients of variation were 250 U mg1), thionine (TH, g85%), ferrocenecarboxylic acid (Fc-COOH, Fc), bovine serum albumin (BSA, 9699%), and 3-glycidyloxypropyl trimethoxysilane (C9H20O5Si, GOPS) were obtained from SigmaAldrich (USA). Diethylene glycol (DEG, 99%) was obtained from Alfa Aesar China (Beijing). Graphene oxide nanosheets were prepared and characterized as described.14 All other reagents were of analytical grade and were used without further purification. Ultrapure water obtained from a Millipore water purification system (g18 MΩ, Milli-Q, Millipore) was used in all runs. Phosphate-buffered saline (PBS, 0.1 M) solution of various pH values was prepared and 0.1 M KCl was used as the supporting electrolyte. Clinical serum samples were made available by Fujian Provincial Hospital, China. Synthesis of Magnetic Graphene Nanosheets (MGO). Graphene nanosheets coated with Fe3O4 nanoparticles (MGO) were synthesized according to the literature with a slight modification.15 Prior to experiment, a 10 mg mL1 NaOH/DEG stock solution was initially prepared via the addition of 200 mg of NaOH into 20 mL of DEG; the mixture then was heated at 120 °C in an oil bath for 1 h under the protection of N2 and cooled to 70 °C. Following that, 15 mg of graphene oxide nanosheets and 60 mg of FeCl3 were added into 10 mL of the asprepared DEG solution and stirred for 1 h at room temperature (RT); the mixture then was transferred to an oil bath and heated to 220 °C for 30 min with stirring under N2. Afterward, 5 mL of DEG stock solution at 70 °C was added rapidly into the 220 °C mixture and further heated for another 1 h. Finally, magnetic graphene nanosheets (MGO, ∼40 mg) were obtained by centrifugation (8.000 g) and washing with ethanol several times. Preparation of Immunosensing Probe (MGO-Ab1,2). The immunosensing probe was prepared by simultaneous conjugation of anti-AFP and anti-CEA antibodies onto the surface of the as-synthesized MGO, according to our recent report.10 Briefly, 20 mg of MGO were initially dried at 100 °C for 1 h, and then incubated with 5 mL of GOPS (5%, v/v) in dry toluene for 12 h at RT under gentle stirring. During this process, GOPS molecules were conjugated onto the MGO through the reaction between OH groups on the MGO and OCH3 groups on the GOPS. The GOPS-functionalized MGO was separated by an external magnet and washed thoroughly with toluene and 5408

dx.doi.org/10.1021/ac200969w |Anal. Chem. 2011, 83, 5407–5414

Analytical Chemistry

ARTICLE

Scheme 1. Schematic Illustration of the Multiplexed Electrochemical Immunoassay Protocol and the Measurement Principle of the Sandwich Immunoassay

ethanol to remove the physically adsorbed GOPS. Following that, the obtained GOPS/MGO was heated for 1 h at 100 °C under N2 to achieve active epoxy groups on the surface. The obtained GOPS/MGO was dispersed into 2 mL of PBS, pH 7.4, (C[MGO] ≈ 10 mg mL1). Afterward, 50 μL of anti-AFP antibodies (1.0 mg mL1) and 50 μL of anti-CEA antibodies (1.0 mg mL1) were simultaneously added to the 2 mL of GOPS/MGO suspension and incubated for 12 h at 4 °C under gentle stirring. The excess biomolecules were removed by magnetic separation. Finally, the as-prepared immunosensing probe (designated as MGO-Ab1,2) was stored in 2 mL of PBS, at pH 7.4 and 4 °C, when not in use (C[MGO] ≈ 10 mg mL1). Preparation of Distinguishable Signal Tags (GHS-Ab1 and GHS-Ab2). In this work, the reverse-micelle method was used for the preparation of two signal tags, according to our earlier report.13b Briefly, solution A, which contained 200 μL of CaCl2 (50 mM), 50 μL of HRP (1 mg mL1), 50 μL of thionine (5 mM), and 15 mL of bis-(2-ethylhexyl) sodium sulfosuccinate (AOT, 0.1 M) iso-octane suspension was added dropwise into solution B, which contained 200 μL of Na2CO3 (50 mM), 50 μL of HRP (1 mg mL1), 50 μL of thionine (5 mM), and 15 mL of AOT (0.1 M) iso-octane suspension with vigorous stirring until the formation of HRP-thionine-doped CaCO3 light-blue colloids (∼30 min). Following that, 100 μL of HAuCl4 solution (5 wt %) was added dropwise to the reverse-micellar solution and stirred for 2 h at RT. Then, 100 μL of hydrazine hydrate (50 mM) was added to reduce the Au3þ on the CaCO3 surface to Au0. Afterward, 3 mL of absolute ethanol was added to form two immiscible layers of aqueous ethanol and iso-octane. The nanogold-coated HRP-thionine-doped CaCO3 microspheres were separated with a separating funnel, washed with iso-octane, and centrifuged for 10 min at 8000 g to remove any residual surfactant. Finally, the HRP-thionine-encapsulated nanogold hollow microspheres (designated as HTGHS) were obtained by adding 0.1 M HCl to the microsphere suspension and used for the labeling of anti-CEA antibodies (anti-CEAHTGHS, designated as GHS-Ab1) through the interaction between nanogold and thiols or alkylamines of the protein.13b Similarly, HRP-Fc-COOH-encapsulated nanogold hollow microspheres (designated as HFGHS) and anti-AFP-HFGHS (designated as GHS-Ab2) were synthesized using the same

method. The signal tags were stored in PBS, at pH 7.4 and 4 °C, until use. Flow-Through Multiplexed Electrochemical Immunoassay. The multiplexed electrochemical immunoassay was carried out by coupling the CHI 630D Electrochemical Analyzer (CH Instruments Inc., Shanghai, PRC) with a flow-through detection cell using an indium tin oxide (ITO, 5 wt % In2O3 þ SnO2) working electrode, a platinum wire auxiliary electrode, and a saturated Ag/AgCl reference electrode. [Note: The detection cell is schematically illustrated in our previous report.13a] The flowthrough system consisted of a six-way valve equipped with a 1-mL syringe pump and connected through a Teflon tubing with the flow cell. The ITO electrode was installed at the bottom of the cell, and an external permanent BaFe12O19 magnet with pot shape (10 mm in diameter and 5 mm in depth, 410430 mT) was set under the ITO electrode. The carrier buffer (0.1 M PBS, pH 6.5), MGO-Ab1,2 (C[MGO] ≈ 10 mg/mL), GHS-Ab1, GHSAb2, and the detection solution (0.1 M PBS, pH 6.5, containing 3.5 mM H2O2) were introduced at a rate of 100 μL min1 via a control valve-based injection loop, respectively. The analyte was directly injected into the cell using a microsyringe, followed by the MGO-Ab1,2. Scheme 1 shows the fabrication process and measurement principle of the integrated multiple electrochemical immunoassay. The assay mainly consists of the following steps: (i) 100 μL of MGO-Ab1,2 (C[MGO] ≈ 10 mg mL1) was flowed into the detection cell and collected on the ITO surface with an external magnet; (ii) 200 μL of the sample containing AFP and/or CEA with various concentrations was injected into the cell, and incubated for 13 min without the external magnet to form the antibodyantigen immunocomplex on the MGO surface; (iii) 200 μL of signal tags (excess) was injected into the cell and incubated for another 13 min without the magnet to construct the sandwich immunocomplex; and (iv) the H2O2PBS (pH 6.5) was flowed through the cell in the presence of the external magnet, and a differential pulse voltammetric (DPV) measurement from 600 mV to 600 mV (vs Ag/AgCl) with a pulse amplitude of 50 mV and a pulse width of 50 ms was registered as the sensor signals. All incubations and measurements were performed using stopped-flow technique. Analyses were always made in triplicate. 5409

dx.doi.org/10.1021/ac200969w |Anal. Chem. 2011, 83, 5407–5414

Analytical Chemistry

ARTICLE

Figure 1. (a) TEM image of magnetic graphene nanosheets (inset shows graphene nanosheets); (b) photograph showing the as-prepared graphene sample and magnetic graphene sample with an external magnet; and (c) TEM image of Fc-COOH and HRP-encapsulated nanogold hollow microspheres.

Figure 2. XPS analysis of (a) magnetic graphene nanosheets and (b) Fc-COOH and HRP-encapsulated nanogold hollow microspheres, and (c) N2 adsorptiondesorption isotherms of Fc-COOH and HRP-encapsulated nanogold hollow microspheres. (Inset shows the pore size distribution.)

’ RESULTS AND DISCUSSION Characteristics of Immunosensing Probe and Signal Tags. In this work, the multiplexed immunoassay for simultaneous detection of CEA and AFP was constructed using the same type of primary antibodies both for conjugation to magnetic graphene nanosheets (immunosensing probe) and to multifunctional nanogold hollow microspheres (signal tags). Since the advantages of the nanoparticle-based immunoassay protocol have been reported in our previous works,10,13a the aim of the present paper was to design a feasible and sensitive nanoparticle-based multianalyte immunoassay as a proof-of-concept by further tuning the nanoparticles’ configuration. Figure 1a shows typical transmission electron microscopy (TEM) images (Model H-7650, Hitachi Instruments, Japan) of the synthesized MGO. Compared with pure graphene oxide (inset of Figure 1a), many nanoparticles could be observed on the MGO (Figure 1a). The graphene nanosheets provided a large surface area for the assembly of the magnetic nanoparticles at the top and bottom of nanosheets. Moreover, the synthesized graphene nanosheets and MGO could be homogeneously dispersed in the water. With the external magnet, however, the synthesized MGO were aggregated together (Figure 1b), which might be attributed to the presence of magnetic nanoparticles on the graphene nanosheets. Figure 1c displays the TEM image of the multifunctional nanogold hollow microspheres (HTGHS as an example), and the average size was 50 nm. The scraggy nanostructures could provide a large surface area for the immobilization of biomolecules.

To further demonstrate the formation of MGO and multifunctional nanogold hollow microspheres, XPS spectrometry was used to characterize the synthesized nanostructures. Figure 2a shows spectra obtained by the X-ray photoelectron spectroscopy (XPS) (Thermo Fisher Scientific, Model Escalab 250 spectrometer, Al KR X-ray, 1486.6 eV) of the MGO. The characteristic peaks for Fe 2p, O 1s, and C 1s core-level regions could be obviously observed at the synthesized MGO. According to the literature,16 the Fe 2p and C 1s core levels were derived from the magnetic nanoparticles and graphene oxide, respectively. Because of the simultaneous existence of elemental oxygen in the magnetic nanoparticles and graphene oxide, a higher characteristic peak at O 1s was achieved. Figure 2b displays the Au 4f7 and 4f5, C 1s, N 1s, S 2p, O 1s, and Fe 2p core-level regions of the multifunctional nanogold hollow microspheres. The Fe 2p core-level region might originate from the encapsulated HRP and/or ferrocene. The existence of the Au 4f doublet (83 and 86 eV for the 4f7/2 and 4f5/2) indicated the presence of gold nanoparticles.17 Furthermore, Type IV adsorptiondesorption isotherms and H1 hysteresis loops of functional nanogold hollow microspheres were investigated with an ASAP 2000 instrument (Micromeritics, Norcross, GA, USA) in the range of 0.71.0 Pa. As shown in Figure 2c, the BrunauerEmmettTeller (BET) surface area and pore size with BarrettJoynerHalenda (BJH) diameter of the most probable distribution were 367.1 m2 g1 and 1.7 nm for the multifunctional hollow microspheres, respectively. The results indicated that nanogold hollow microspheres could be synthesized via the reverse-micelle method. 5410

dx.doi.org/10.1021/ac200969w |Anal. Chem. 2011, 83, 5407–5414

Analytical Chemistry

ARTICLE

Figure 3. Cyclic voltammograms of MGO-Ab1,2/ITO (trace a), the probe described by trace “a” after incubation with 10 ng mL1 AFP and 10 ng mL1 CEA (trace b), the probe described by trace “b” after incubation with excess GHS-Ab1 (trace c), the probe described by trace “b” after incubation with excess GHS-Ab2 (trace d), and the probe described by trace “b” after incubation with excess GHS-Ab1 and GHS-Ab2 in PBS, pH 6.5, without H2O2 at 50 mV s1 (trace e). (Note: The cyclic voltammogram of MGO-Ab1,2/ITO probe described by trace “e” in PBS, pH 6.5, containing 3.5 mM H2O2 is shown in panel “f”.)

Characteristics of the Multiplexed Immunoassay. Figure 3 displays the electrochemical behavior of the nanoparticles-based immunoassay after each step in PBS, pH 6.5. As shown in Figure 3a, an obvious reduction peak around 0.3 V to 0.4 V was observed at the MGO-Ab1,2. The peak might be attributed to the reduction of oxygen in aerated buffer solution, pH 6.5.18 When the MGO-Ab1,2 reacted with the mixture containing 10 ng mL1 AFP and 10 ng mL1 CEA (as an example), the background currents were decreased (Figure 3b). The reason might be attributed to the fact that the antigenantibody complexes formed on the MGO hindered the electron transfer. The resulting immunosensing probe then was further incubated with the signal tags. Figure 3c shows the cyclic voltammogram of the immunosensing probe after reaction with excess GHS-Ab1, where a couple of well-defined redox peaks appeared at 180 mV and 130 mV at the working potential range in PBS, pH 6.5. The redox wave mainly originated from the encapsulated thionine.13c The result indicated that the nanoparticles-based sandwich immunocomplex could be formed between the MGOAb1,2 and the GHS-Ab1. Furthermore, after the immunosensing probe was incubated again with excess GHS-Ab2, another new pair of redox peaks at þ280 mV and þ340 mV was obtained (Figure 3d), which was mainly ascribed to the encapsulated ferrocene. Importantly, when the immunosensing probe was simultaneously incubated with GHS-Ab1 and GHS-Ab2, two pairs of redox waves appeared (Figure 3e). The peak separation between two pairs of the reduction peaks was ∼450 mV (ΔEpc). Thus, a differentiation of CEA and AFP was possible, according to the position of the corresponding reduction peak. Significantly, upon the addition of 3.5 mM H2O2 into PBS, pH 6.5, an obvious catalytic process with the decrease of anodic peak and the increase of cathodic peak was observed for the two couples of redox waves with the sandwiched immunosensing probe (Figure 3f). The results indicated that the bioactivity of the encapsulated HRP into the nanogold hollow microspheres still remained. The reaction process of the multiplexed electrochemical immunoassay could be simply summarized as follows: Immunoreaction process:

Electrochemical measurement and the bioelectrocatalytic principle: H2 O2 þ HRP f H2 O þ Compound I

ð2Þ

Compound II þ THðredÞ =FcðredÞ f Compound II þ THðoxÞ =FcðoxÞ

ð3Þ

Compound II þ THðoxÞ =FcðoxÞ þ 2Hþ f HRP þ THðoxÞ =FcðoxÞ þ H2 O THðoxÞ =FcðoxÞ þ 2e þ 2Hþ f THðredÞ =FcðredÞ

ð4Þ ð5Þ

Comparison of Multiplexed Immunoassays Using Different Types of Immunosensing Probes. To monitor the effect of

MGO on the signal amplification of the multiplexed electrochemical immunoassay, we prepared two types of immunosensing probes with and without graphene nanosheets (i.e., MGOAb1,2 and Fe3O4-Ab1,2) for the simultaneous detection of AFP and CEA using GHS-Ab1 and GHS-Ab2 as signal tags. [Note: The Fe3O4-Ab1,2 was synthesized according to our previous report.10] The comparison was performed based on the shift in the reduction currents, relative to zero analyte condition. As seen from Figure 4, the MGO-based immunosensing probe exhibited higher shifts in the currents for both AFP (Figure 4A) and CEA (Figure 4B), compared to Fe3O4-Ab1,2 as signal tags, even at low analyte concentrations. The reason might be the fact that the synthesized MGO could act as a filterlike network, which could capture more AFP and CEA biomolecules than that of the individually dispersed functionalized nanoparticles. In contrast to MGO, the probability for Fe3O4-Ab1,2-based immunosensing probe to bind with randomly distributed analytes in the solution might be reduced. Therefore, the MGO-Ab1,2 was utilized in this work. Comparison of Multiplexed Immunoassays Using Different Signal Tags. For the successful development of a multianalyte immunoassay, signal amplification is crucial. To verify the advantages of the as-synthesized GHS-Ab1 and GHS-Ab2, we also prepared another two types of signal tags, i.e., HRP-labeled primary antibodies (HRP-Ab1 and HRP-Ab2) and nanogoldlabeled HRP-Ab conjugates (16 nm in diameter as an example, Au-HRP-Ab1 and Au-HRP-Ab2), which were employed for the 5411

dx.doi.org/10.1021/ac200969w |Anal. Chem. 2011, 83, 5407–5414

Analytical Chemistry

ARTICLE

Figure 4. Electrochemical responses of the flow-through multiplexed immunoassay with various immunosensing probes: (a) MGO-Ab1,2 and (b) Fe3O4-Ab1,2 toward various concentrations of (A) AFP and (B) CEA standards, using GHS-Ab1 and GHS-Ab2 as signal tags.

Figure 5. Electrochemical responses of the flow-through multiplexed immunoassay with various signal tags ((a) GHS-Ab, (b) Au-HRP-Ab, and (c) HRP-Ab) toward various concentrations of (A) AFP and (B) CEA standards, using MGO-Ab1,2 as an immunosensing probe.

detection of AFP and CEA, using MGO-Ab1,2 as an immunosensing probe. The comparative study was carried out using the single-analyte immunoassay format. [Note: The corresponding electron mediators, thionine or ferrocene, were added to the detection solution when HRP-Ab or Au-HRP-Ab were used as signal tags.] The results are listed in Figure 5. As indicated from this figure, the shifts of the reduction currents using GHS-Ab1 and GHS-Ab2 were significantly higher than those of the other two signal tags. At the same time, we also observed that the immunoassays using HRP-Ab and Au-HRP-Ab as signal tags exhibited low and smooth current changes at the low levels. The results suggested that the use of multifunctional nanogold hollow microspheres not only displayed high sensitivity, but also favor the determination of the analytes at very low concentrations. Evaluation of Cross-Talk and Cross-Reactivity. Prior to experiments, the assay conditions, including the incubation time (13 min) and incubation temperature (RT) for both antigenantibody reactions, and the pH of PBS (pH 6.5), were optimized (see Figure S1 in the Supporting Information). To evaluate the cross-reactivity and cross-talk of the multiplexed electrochemical immunoassay, three control tests were carried out as follows: (i) single analyte, i.e. AFP or CEA, was assayed using the MGO-Ab1,2 as an immunosensing probe, and GHSAb1 and GHS-Ab2 as signal tags; (ii) AFP and CEA were simultaneously monitored using the MGO-Ab1,2 as an immunosensing probe, and GHS-Ab1 and GHS-Ab2 as signal tags; and (iii) normal serum (i.e., negative human serum without AFP and CEA) and positive serum samples containing AFP and CEA were simultaneously measured using the MGO-Ab1,2 as an immunosensing probe, and GHS-Ab1 and GHS-Ab2 as signal tags. With

tests (i) and (ii), 1.0 and 50 ng mL1 of AFP and CEA were assayed as examples. The electrochemical responses of the three control tests are displayed in Table S1 in the Supporting Information. Significantly higher current response was observed with the corresponding target analyte. Meanwhile, we also observed that the current of the immunoassay displayed a substantially low increase (the shift in peak currents were near zero) when using normal (negative) serum samples as the control tests in contrast with those of the positive serum. The results indicated that the multiplexed immunoassay exhibited low cross-talk and cross-reactivity between both analytes. Analytical Performance of the Multiplexed Immunoassay. Under optimal conditions, the sensitivity and dynamic range of the multiplexed immunoassay was evaluated with AFP and CEA standards. A differential pulse voltammetry (DPV) measurement was carried out in PBS, pH 6.5, containing 3.5 mM H2O2 after incubation with various analyte levels and excess signal tags for 13 min at RT. As indicated in Figure 6a, the DPV reduction currents of the multiplexed immunoassay increased as the AFP and CEA concentrations increased. Both calibration plots displayed a good linear relationship between the reduction peak currents and the logarithm of the analyte concentration in the ranges of 0.01200 ng mL1 for AFP (Figure 6b) and 0.01 80 ng mL1 for CEA (Figure 6c). The correlation coefficients were 0.987 and 0.967 for AFP and CEA (n = 27), respectively. The limit of detection (LOD) values for AFP and CEA were determined at 1.0 pg mL1, estimated at the 3sB criterion, which were partially lower than those of enzyme-linked immunosorbent assay (3.7 ng mL1), carbon nanotube/Prussian blue/nanogold-modified immunosensor (3.0 pg mL1), 5412

dx.doi.org/10.1021/ac200969w |Anal. Chem. 2011, 83, 5407–5414

Analytical Chemistry

ARTICLE

Figure 6. (a) Differential pulse voltammetry (DPV) responses and (b,c) calibration curves of the flow-through multiplexed electrochemical immunoassay toward AFP and CEA standards, using MGO-Ab1,2 as an immunosensing probe and GHS-Ab as signal tags in PBS, pH 6.5, containing 3.5 mM H2O2.

graphene sheets/multienzyme-functionalized carbon nanospheres (20 pg mL1), label-free photoelectrochemical immunoassay (40 pg mL1), and electrochemical protein chip (2.0 ng mL1) for AFP,19 and time-resolved immunofluorometric assay (240 pg mL1), quantum dot barcode-based electrochemical immunoassay (3.3 pg mL1), surface-enhanced Raman scattering of hollow gold nanospheres (110 pg mL1), and carbon nanoparticles-enhanced immunoelectrochemical assay (32 pg mL1) for CEA.20 The results indicated that the multiplexed electrochemical immunoassay enabled wide linear ranges and low LODs. Furthermore, the assay format shortened the assay time and decreased the sample volume and, therefore, analysis costs. Since the threshold values in normal human serum is ∼10 ng mL1 for AFP and 3 ng mL1 for CEA,21 the developed multiplexed immunoassay could completely meet the requirements of clinical diagnosis. If the concentrations in the sample are >200 ng mL1 for AFP and >80 ng mL1 for CEA, an appropriate dilution should be preferable. To investigate the reproducibility of determinations, we repeatedly assayed three different concentrations of both tumor markers, using identical batches of immunosensing probes and signal tags throughout. Experimental results indicated that the coefficients of variation (CVs) of the intra-assay between six runs were 9.1%, 6.4%, and 5.7% for 1.0, 50, and 150 ng mL1 AFP, and 5.7%, 8.9%, and 7.2% for 1.0, 30, and 60 ng mL1 CEA, respectively, whereas the CVs of the interassay with various batches were 8.3%, 9.8%, and 7.6% for AFP, and 8.9%, 7.2%, and 9.4% for CEA toward the above-mentioned analyte. The low CVs indicated that the multiplexed immunoassay could be regenerated and used repeatedly, and further verified the possibility of batch preparation. When the immunosensing probe and signal tags were not in use, they were stored in PBS, at pH 6.5 and 4 °C. No obvious change was observed after storage for 17 days, but a 10% decrease of its initial currents was noticed after 43 days. Analysis of Clinical Serum Specimens and Method Validation. Furthermore, the multiplexed immunoassay was validated by assaying nine clinical serum specimens containing AFP and CEA. The obtained results were compared with those obtained by a commercially available electrochemiluminescence technique (ECL, Roche 2010, Switzerland) as a reference method (see Table S2 in the Supporting Information). The relative standard deviations (RSDs) were 1.4%7.2% and 2.3%9.1% for AFP and CEA, respectively. No significant differences were encountered, thereby revealing a good correlation between the two methods.

In addition, the multiplexed immunoassay was also used for the simultaneous determination of AFP and CEA with ultralow concentrations obtained by spiking various AFP and CEA standards into the blank calf serum (Dingguo Biotechnol. Co. Ltd., Beijing). [Note: The aim of using normal calf serum is to avoid the possible interference by AFP and CEA in the normal human serum.] Three levels of AFP and CEA were used: 15, 200, and 500 pg mL1. Experimental results revealed 16.7, 189.6, and 567.3 pg mL1 for AFP, and 17.3, 215.4, and 612.3 pg mL1 for CEA. In conclusion, the recoveries for AFP and CEA were 111.3%, 94.8%, and 113.5% for AFP, and 115.3%, 107.7%, and 122.5% for CEA, respectively, indicating satisfactory accuracy of the multiplexed immunoassay.

’ CONCLUSIONS In this work, a novel highly sensitive multiplexed electrochemical immunoassay for simultaneous detection of alphafetoprotein (AFP) and carcinoembryonic (CEA) was developed using biofunctionalized magnetic graphene nanosheets (MGO) as immunosensing probes and multifunctional nanogold hollow microspheres as distinguishable signal tags. Magnetic graphene nanosheets were not only used as a substrate for the immobilization of biomolecules, but also facilitated the rapid separation and purification after synthesis. With the flow-through system, the honeycomb-like magnetic graphene immunosensing probe could provide a large surface area for target capture. Highlights of this work can be summarized as follows: (i) the synthesized MGO can be used for the simultaneous immobilization of two types of antibodies and avoid the fabrication of a position-sensitive array; (ii) the multiplexed immunoassay method can pull antibodies bound to magnetic nanoparticles from one laminar flow path to another by applying a local magnetic field gradient, and selectively remove them from flowing biological fluids without any washing steps; (iii) the system can be regenerated; and (iv) the simultaneous determination of two relevant clinical targets will improve human phenotype decision-making. Significantly, the multiplexed immunoassay approach does not require sophisticated fabrication and is well-suited for high-throughput biomedical sensing and application to other areas. ’ ASSOCIATED CONTENT

bS

Supporting Information. This material is available free of charge via the Internet at http://pubs.acs.org.

5413

dx.doi.org/10.1021/ac200969w |Anal. Chem. 2011, 83, 5407–5414

Analytical Chemistry

’ AUTHOR INFORMATION Corresponding Author

*Phone: þ86-591-22866125 (D.T.), þ49-89-218078252 (D.K.). Fax: þ86-591-22866135 (D.T.), þ49-89-218078255 (D.K.). E-mail addresses: [email protected] (D.T.), [email protected] (D.K.).

’ ACKNOWLEDGMENT Support by the National Natural Science Foundation of China (Nos. 21075019 and 20735002), the Research Fund for the Doctoral Program of Higher Education of China (No. 20103514120003), the Award Program for Minjiang Scholar Professorship (No. XRC-0929), the Alexander von Humboldt-Foundation of Germany, and the “973” National Basic Research Program of China (No. 2010CB732403) is gratefully acknowledged. ’ REFERENCES (1) (a) Freedland, S. Cancer 2011, 117, 1123–1135. (b) Wickstrom, M.; Larsson, R.; Nygren, P.; Gullbo, J. Cancer Sci. 2011, 102, 501–508. (2) (a) Li, H.; Cao, Z.; Zhang, Y.; Lau, C.; Lu, J. Analyst 2011, 136, 1399–1405. (b) Lai, G. S.; Wu, J.; Leng, C.; Ju, H. X.; Yan, F. Biosens. Bioelectron. 2011, 26, 3782–3787. (3) (a) Wang, J.; Mi, X.; Guan, H.; Wang, X.; Wu, Y. Chem. Commun. 2011, 47, 2940–2942. (b) Bonilla, D.; Mallen, M.; De La Rica, R.; Fernandez-Sanchez, C.; Baldi, A. Anal. Chem. 2011, 83, 1726–1731. (c) Li, H.; Leulmi, R.; Juncker, D. Lab Chip 2011, 11, 528–534. (d) Song, T.; Zhang, Q.; Lu, C.; Gong, X.; Yang, Q.; Li, Y.; Liu, J.; Chang, J. J. Mater. Chem. 2011, 21, 2169–2177. (e) Tamarit-Lopez, J.; Morais, S.; Puchades, R.; Maquieira, T. Biosens. Bioelectron. 2011, 26, 2694–2698. (f) Yang, Z. J.; Liu, H.; Zong, C.; Yan, F.; Ju, H. X. Anal. Chem. 2009, 81, 5484–5489. (4) (a) Song, T.; Zhang, Q.; Lu, C.; Gong, X.; Yang, Q.; Li, Y.; Liu, J.; Chang, J. J. Mater. Chem. 2011, 21, 2169–2177. (b) Agger, S.; Marney, L.; Hoofnagle, A. Clin. Chem. 2010, 56, 1804–1813. (c) Barbee, K.; Hsiao, A.; Roller, E.; Huang, X. Lab Chip 2010, 10, 3084–3093. (d) Hazarika, P.; Jickells, S.; Wolff, X.; Russell, D. Anal. Chem. 2010, 82, 9150–9154. (e) Fu, Z. F.; Liu, H.; Ju, H. X. Anal. Chem. 2006, 78, 6999–7005. (f) Wu, J.; Yan, F.; Zhang, X. Q.; Yan, Y. T.; Tang, J. H.; Ju, H. X. Clin. Chem. 2008, 54, 1481–1488. (g) Wu, J.; Yan, Y. T.; Yan, F.; Ju, H. X. Anal. Chem. 2008, 80, 6072–6077. (5) Morais, S.; Tortajada-Genaro, L.; Arnandis-Chover, T.; Puchades, R.; Maquieira, A. Anal. Chem. 2009, 81, 5646–554. (6) (a) Bonilla, D.; Mallen, M.; De La Rica, R.; Fernandez-Sanchez, C.; Baldi, A. Anal. Chem. 2011, 83, 1726–1731. (b) Ma, L.; Wang, C.; Hong, Y.; Zhang, M.; Su, M. Anal. Chem. 2010, 82, 1186–1190. (c) Lai, G.; Yang, F.; Ju, H. Anal. Chem. 2009, 81, 9730–9736. (7) (a) Willner, I.; Katz, E. Angew. Chem., Int. Ed. 2003, 42, 4576–4588. (b) Katz, E.; Sheeney-Haj-Ichia, L.; Buchmann, A.; Willner, I. Angew. Chem., Int. Ed. 2002, 41, 1343–1346. (c) Nam, J.; Thaxton, C.; Mirkin, C. Science 2003, 301, 1884–1886. (8) (a) Kricka, L.; Park, J. Clin. Chem. 2009, 55, 1058–1060. (b) Azzazy, H.; Mansour, M.; Kazmierczak, S. Clin. Chem. 2006, 52, 1238–1246. (9) (a) Giliohann, D.; Mirkin, C. Nature 2009, 462, 461–464. (b) Fishbein, I.; Levy, R. Nature 2009, 461, 890–891. (c) Hill, H.; Mirkin, C. Nat. Protoc. 2006, 1, 324–336. (10) Tang, D.; Su, B.; Tang, J.; Ren, J.; Chen, G. Anal. Chem. 2010, 82, 1527–1534. (11) (a) Craciun, M.; Russo, S.; Yamamoto, M.; Tarucha, S. Nano Today 2011, 6, 42–60. (b) Jia, X.; Campos-Delgado, J.; Terrones, M.; Meunier, V.; Dresselhaus, M. Nanoscale 2011, 3, 86–95. (c) Wei, Q; Mao, K.; Wu, D.; Dai, Y.; Yang, J.; Du, B.; Yang, M.; Li, H. Sens. Actuators B 2010, 149, 314–318.

ARTICLE

(12) (a) Ambrosi, A.; Castaneda, M.; Killard, A.; Smyth, M.; Alegret, S.; Merkoci, A. Anal. Chem. 2007, 79, 5232–5234. (b) Zhong, Z.; Wu, W.; Wang, D.; Wang, D.; Shan, J.; Qing, Y.; Zhang, Z. Biosens. Bioelectron. 2010, 25, 2379–2383. (c) Liu, S.; Zhang, X.; Luo, W.; Wang, Z.; Guo, X.; Steigerwald, M.; Fang, X. Angew. Chem., Int. Ed. 2011, 50, 2496–2502. (d) Holland, J.; Malin, J.; Jordan, D.; Geiger, F. J. Am. Chem. Soc. 2011, 133, 2567–2570. (e) Khan, H.; Jang, J.; Kim, J.; Knoll, W. J. Am. Chem. Soc. 2011, 133, 2170–2176. (f) Lee, S.; Lee, K.; Ahn, J.; Lee, J.; Kim, M.; Shin, Y. ACS Nano 2011, 5, 897–904. (13) (a) Tang, D.; Yu, Y.; Niessner, R.; Miro, M.; Knopp, D. Analyst 2010, 135, 2661–2667. (b) Tang, D.; Ren, J. Anal. Chem. 2008, 80, 8064–8070. (c) Tang, D.; Yuan, R.; Chai, Y. Anal. Chem. 2008, 80, 1582–1588. (d) Tang, D.; Tang, J.; Su, B.; Chen, G. J. Agric. Food Chem. 2010, 58, 10824–10830. (e) Li, Q.; Tang, D.; Tang, J.; Su, B.; Huang, J.; Chen, G. Talanta 2011, 84, 538–546. (14) Lu, C.; Yang, H.; Zhu, C.; Chen, X.; Chen, G. Angew. Chem., Int. Ed. 2009, 28, 4785–4787. (15) He, H.; Gao, C. ACS Appl. Mater. Interfaces 2010, 2, 3201–3210. (16) (a) Pan, L.; Zhang, G.; Fang, C.; Qiu, H.; Wu, P.; Wang, F.; Zhang, Y. Thin Solid Films 2005, 473, 63–67. (b) Yang, D.; Velamakanni, A.; Bozoklu, G.; Park, S.; Stoller, M.; Piner, R.; Stankovich, S.; Jung, I.; Field, D.; Ventrice, C., Jr.; Ruoff, R. Carbon 2009, 47, 145–152. (17) (a) Chandra, S.; Deepak, F.; Gruber, J.; Sardar, D. J. Phys. Chem. C 2010, 114, 874–880. (b) Tripathy, P.; Mishra, A.; Ram, S.; Fecht, H.; Bansmann, J.; Behm, R. Nanotechnology 2009, 20, Article No. No. 075701. (18) (a) Lai, M.; Bergel, A. J. Electroanal. Chem. 2000, 494, 30–40. (b) Raj, I.; Vasu, K. J. Appl. Electrochem. 1993, 23, 728–734. (19) (a) Gianetto, M.; Elviri, L.; Careri, M.; Mangia, A.; Mori, G. Biosens. Bioelectron. 2011, 26, 2232–2236. (b) Jiang, W.; Yuan, R.; Chai, Y.; Yin, B. Anal. Biochem. 2010, 407, 65–71. (c) Du, D.; Zou, Z.; Shin, Y.; Wang, J.; Wu, H.; Engelhard, M.; Liu, J.; Aksay, I.; Liu, Y. Anal. Chem. 2010, 82, 2989–2995. (d) Wang, G.; Xu, J.; Chen, H.; Fu, S. Biosens. Bioelectron. 2009, 25, 791–796. (e) Wilson, M.; Nie., W. Anal. Chem. 2006, 78, 6476–6483. (20) (a) Wang, Z.; Huang, Y.; Li, S.; Xu, H.; Linder, M.; Qiao, M. Biosens. Bioelectron. 2010, 26, 1074–1079. (b) Xiang, Y.; Zhang, Y.; Chang, Y.; Chai, Y.; Wang, J.; Yuan, R. Anal. Chem. 2010, 82, 1138–1141. (c) Chon, H.; Lee, S.; Son, S.; Oh, C.; Choo, J. Anal. Chem. 2009, 81, 3029–3034. (e) Ho, J.; Lin, Y.; Wang, L.; Hwang, K.; Chou, P. Anal. Chem. 2009, 81, 1340–1346. (21) Wu, J.; Fu, Z.; Yan, F.; Ju, H. TrAC, Trends Anal. Chem. 2007, 26, 679–688.

5414

dx.doi.org/10.1021/ac200969w |Anal. Chem. 2011, 83, 5407–5414