Article pubs.acs.org/ac
Enzymatic Hydrolysate-Induced Displacement Reaction with Multifunctional Silica Beads Doped with Horseradish Peroxidase− Thionine Conjugate for Ultrasensitive Electrochemical Immunoassay Youxiu Lin,† Qian Zhou,† Yuping Lin,† Dianping Tang,*,† Reinhard Niessner,‡ and Dietmar Knopp*,‡ †
Key Laboratory of Analysis and Detection for Food Safety (Ministry of Education and Fujian Province), Institute of Nanomedicine and Nanobiosensing, Department of Chemistry, Fuzhou University, Fuzhou 350108, People’s Republic of China ‡ Chair for Analytical Chemistry, Institute of Hydrochemistry, Technische Universität München, Marchioninistrasse 17, D-81377 München, Germany ABSTRACT: A novel (invertase) enzymatic hydrolysatetriggered displacement reaction strategy with multifunctional silica beads, doped with horseradish peroxidase−thionine (HRP−Thi) conjugate, was developed for competitive-type electrochemical immunoassay of small molecular aflatoxin B1 (AFB1). The competitive-type displacement reaction was carried out on the basis of the affinity difference between enzymatic hydrolysate (glucose) and its analogue (dextran) for concanavalin A (Con A) binding sites. Initially, thionine−HRP conjugates were doped into nanometer-sized silica beads using the reverse micelle method. Then monoclonal anti-AFB1 antibody and Con A were covalently conjugated to the silica beads. The immunosensor was prepared by means of immobilizing the multifunctional silica beads on a dextran-modified sensing interface via the dextran−Con A binding reaction. Gold nanoparticles functionalized with AFB1−bovine serum albumin conjugate (AFB1− BSA) and invertase were utilized as the trace tag. Upon target AFB1 introduction, a competitive-type immunoreaction was implemented between the analyte and the labeled AFB1−BSA on the nanogold particles for the immobilized anti-AFB1 antibody on the electrode. The invertase followed by gold nanoparticles hydrolyzed sucrose into glucose and fructose. The produced glucose displaced the multifunctional silica beads from the electrode based on the classical dextran−Con A−glucose system, thus decreasing the catalytic efficiency of the immobilized HRP on the electrode relative to that of the H2O2−thionine system. Under optimal conditions, the detectable electrochemical signal increased with the increasing target AFB1 in a dynamic working range from 3.0 pg mL−1 to 20 ng mL−1 with a detection limit of 2.7 pg mL−1. The strong bioconjugation with two nanostructures also resulted in a good repeatability and interassay precision down to 9.3%. Finally, the methodology was further validated for analysis of naturally contaminated or spiked AFB1 peanut samples, giving results matched well with those from a commercialized AFB1 enzyme-linked immunosorbent assay kit. Importantly, the system provides a signal-on competitive-type immunosensing platform for ultrasensitive detection of small molecules.
I
antigen during the measurement.11,12 Generally speaking, the detectable signal decreases with increasing analyte concentration.13,14 In this case, a high background signal is requested toward zero analyte and thus may suffer from a low sensitivity and a narrow linear range. To keep pace with expectations in the future, there is the request for signal-on, yet highly sensitive, quantitative, more flexible, and easy-to-use methods. In the conventional competitive-type enzyme immunoassays, the detectable signal mainly derives from the labeled enzyme toward the catalytic oxidation or reduction of substrate.15,16 The assay is often executed on the basis of a one-step immunoreaction with a signal-off assay mode.17 Over the years, many competitivetype enzyme immunoassays have addressed this challenge, yet most assay formats are very difficult to realize as signal-on assay systems and thus are inflexible to apply to small molecular
mmunoassays, based on the specific antigen−antibody recognition reaction, have recently gained increasing attention in research fields including clinical diagnostics, environmental monitoring, and food quality control. 1−3 Undoubtedly, enzyme immunoassays are widely used for the development of high-efficiency immunoassay methods,4−6 because a single enzyme molecule, e.g., horseradish peroxidase (HRP), may cause the conversion of 107 molecules of substrate per minute.7 Typically, the method usually consists of sandwichtype and competitive-type immunoassay formats.8,9 Despite high sensitivity and specificity of the sandwich-type immunoassays, they have some disadvantages, e.g., a relatively long incubation time and expensive chemicals because of the use of two matched antibodies.10 More unfavorably, the sandwich-type assays are unsuitable for quantitative monitoring of small molecules, e.g., mycotoxins, marine toxins, and food additives. In contrast, competitive-type immunoassays can provide a favorable format for the determination of small molecules, since target analyte can be monitored by its ability to compete with the labeled hapten/ © XXXX American Chemical Society
Received: June 15, 2015 Accepted: July 16, 2015
A
DOI: 10.1021/acs.analchem.5b02253 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry targets.18−21 We speculate that one main disadvantage is to involve one competitive reaction in the whole assay process. To innovate such a signal-off competitive-type immunoassay, an alternative approach that adopts the dialectic viewpoint (double negation equals affirmation) would be advantageous by coupling with two competitive/displacement reactions. Inspiringly, the rapid emergence of the displacement-type assay mode provides a promising platform for the development of novel competitivetype immunoassays.22−24 Concanavalin A (Con A; 104−112 kDa, pI = 4.5−5.5, 3.9 × 4.0 × 4.2 nm for the tetramer, a member of the legume lectin family) is a carbohydrate-binding protein from jack bean with four saccharide binding sites.25,26 Under neutral conditions, each subunit contains one site for hydrophobic recognition, one each for Mn2+ and Ca2+ ions (to activate the specific carbohydrate site of protein), and the fourth one for the specificity of carbohydrates, i.e, for α-D-mannosyl and α-Dglucosyl residues (two hexoses differing only by the hydroxyl group on carbon 2).27,28 Con A is the first lectin to be available on a commercial basis, and is widely utilized in biology and biochemistry to characterize glycoproteins on the surface of various cells.29,30 The molecular basis of its interactions with metals as well as its affinity for monosaccharides mannose and glucose are well-known.31−33 To this end, competitive binding assays with Con A as the receptor have been employed over the years as a potential solution for continuous glucose monitoring application.34,35 Typically, Con A-based sensors track the glucose concentration by following the inhibition the sugar has on the binding between the lectin and a competing ligand. Usually, the systems are utilized by coupling with the glucose−Con A− dextran system since dextran (a polymer of glucose) has a lower affinity for binding to Con A than glucose.22,36 Favorably, glucose can be acquired by the invertase toward sucrose hydrolysis,37 and the invertase can be utilized for the labeling of the molecular tag. In this regard, a novel signal-on competitive-type immunoassay protocol might be constructed by combination with the glucose−Con A−dextran competitive binding system. To the best of our knowledge, there has been no report focusing on the glucose−Con A−dextran binding system for the development of a signal-on competitive-type electrochemical immunoassay until now. Another important concern for high-efficiency competitivetype immunoassay is to obtain a low limit of detection and quantification. Recent research has looked to develop innovative and powerful novel biofunctionalized nanostructures, controlling and tailoring their properties in a very predictable manner to meet the needs of specific applications.38,39 Silica nanoparticles have proved to be an ideal protein host because of their highly chemical/thermal stability, fine suspendability in aqueous solution, and good biocompatibility with the environment.40,41 Some advanced techniques involving doped silica beads with an organic dye, quantum dots, and Raman-active particles have been promoted on the basis of the Stöber technique or reverse micelle method.42,43 Importantly, nanosized silica particles have a high silanol concentration on the surface which facilitates a wide variety of surface reactions and allows conjugation with biomolecules such as proteins and DNA.44,45 With these advantages in mind, our motivation in this work is to synthesize oxidoreductase- and organic dye (e.g., thionine)-doped silica nanoparticles for the bioconjugation of the antibodies to realize the amplification of the detectable signal in competitive-type immunoassays. Aflatoxins, the highly toxic secondary metabolites produced by a number of different fungi, are present in a wide range of food
and feed commodities, and are assumed significant because of their deleterious effects on human beings, poultry, and livestock.46,47 The major aflatoxins of interest are designated as B1, B2, G1, and G2; however, aflatoxin B1 (AFB1) is usually predominant and the most hazardous.48 Thus, exploring validated analytical methods for rapid detection of AFB1 on a large scale is important. Herein, we report the proof-of-concept of a novel and powerful signal-on competitive-type electrochemical immunoassay for the detection of AFB1 by coupling the glucose−Con A−dextran displacement mode with the nanoparticle-labeling strategy. Initially, monoclonal anti-AFB1 antibody and Con A were covalently conjugated onto HRP− thionine-doped silica beads, and then the functionalized silica beads were immobilized onto a dextran-modified sensing interface via the Con A−dextran chemistry for the preparation of the immunosensor. Gold nanoparticles, functionalized with invertase and AFB1−BSA (bovine serum albumin), were employed as the competitors. The assay consists of two competitive reactions: (i) target AFB1 and the labeled AFB1− BSA on the gold nanoparticles for the immobilized anti-AFB1 on the electrode (i.e., competitive reaction) and (ii) glucose generated by the invertase toward sucrose hydrolysis and the immobilized dextran for the Con A binding site owing to the affinity difference (i.e., displacement reaction). The electrochemical signal derives from the doped HRP into the silica beads toward the catalytic reduction of hydrogen peroxide with the assistance of the thionine mediator. The aim of our design is to exploit a signal-on competitive-type electrochemical immunoassay protocol for ultrasensitive screening of small molecules to challenge conventional signal-off competitive-type immunoassays.
■
EXPERIMENTAL SECTION Materials and Reagents. Monoclonal anti-AFB1 antibody (mAb 62, clone 2B7) was synthesized and characterized in our laboratory.49 AFB1−BSA conjugate was obtained from the School of Food Science and Technology, Jiangnan University (Wuxi, China). AFB1 standards in acetonitrile were purchased from Express Technology Co. Ltd. (Beijing, China). Concanavalin A was acquired from Vector Laboratories, Inc. (Burlingame, CA). 1,2-Epoxy-3-phenoxypropane-derived dextran (DexP) and thionine−HRP conjugate (Thi−HRP) were synthesized according to our previously reported methods, respectively.50,51 Graphene oxide nanosheets (2.0 mg mL−1, dispersion in H2O), HRP (lyophilized, powder, beige, ∼150 U mg−1), gold nanoparticles (AuNPs, 15 nm diameter, stabilized suspension in 0.1 mM phosphate-buffered saline (PBS), reactant free), invertase from baker’s yeast (Saccharomyces cerevisiae, grade VII, ≥300 U mL−1 solid), sucrose (Vetec reagent grade, 99%), thionine acetate salt (Thi), bovine serum albumin (BSA), tetraethoxysilane (TEOS), and (3-(glycidyloxy)propyl)trimethoxysilane (C9H20O5Si, GOPS) were obtained from Sigma-Aldrich. Cyclohexane, 1-hexanol, and ammonium hydroxide were obtained from Merck-Millipore (Darmstadt, Germany). All other reagents were of analytical grade and were used without further purification. Ultrapure water obtained from a Millipore water purification system was used in all runs (18.2 MΩ cm−1, Milli-Q). PBS (0.01 M) solution with various pH values containing 0.1 M KCl, 0.1 M Mn2+, and 0.1 M Ca2+ ions was prepared by mixing 0.01 M K2HPO4 and 0.01 M KH2PO4. Preparation of Naturally Contaminated/Spiked AFB1 Peanut Samples. AFB1-spiked peanut standards with various concentrations were prepared via our previously reported B
DOI: 10.1021/acs.analchem.5b02253 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
Scheme 1. Schematic Illustration of the Invertase Hydrolysate-Induced Displacement Reaction with Functionalized Silica Beads Doped with Thionine−Horseradish Peroxidase (Thi−HRP) for Ultrasensitive Electrochemical Immunoassay of Aflatoxin B1 (AFB1) Based on the Typical Dextran−Con A−Glucose Displacement System
method.52 Briefly, 5.0 g of the blank peanut sample milled by using a Midea food mixer (BP252AG, Guangdong, China) was initially extracted with 37.5 mL of MeOH/water (80:20, v/v) after being stirred for 60 min at room temperature (RT) followed by filtration. Then the extract (20 mL) was diluted into 60 mL of ultrapure water to acquire a final concentration of 20% (v/v) MeOH. Finally, standard samples were prepared by spiking AFB1 standards into different-volume-diluted extracts. Naturally contaminated peanut samples were simply made by extracting 7.5 g of peanut slurry with 18 mL of MeOH. Following that, the contaminated peanut samples were obtained by addition of 15 mL of ultrapure water to 5 mL of extract for AFB1 detection. Preparation of Invertase/AFB 1 −BSA-Conjugated AuNPs (En−AuNP−AFB1). The En−AuNP−AFB1 conjugates were prepared on the basis of the electrostatic and hydrophobic interactions between proteins and gold nanoparticles.53 The process mainly consisted of the following steps: (i) gold colloids (5.0 mL, 15 nm in diameter, 5.0 ng mL−1) were adjusted to pH 9.5 by using a 0.1 M Na2CO3 aqueous solution; (ii) 200 μL of invertase (0.5 mg mL−1) and 50 μL of AFB1−BSA (0.5 mg mL−1) were injected into the colloidal gold nanoparticles, and the resulting solution was gently shaken for 60 min at RT on a shaker (MS, IKA GmbH, Staufen, Germany); (iii) 100 μL of polyethylene glycol (1.0 wt %) was added to the suspension, and the mixture was further incubated for 12 h at 4 °C; (iv) En− AuNP−AFB1 conjugates were obtained by centrifugation at 4 °C (10 min, 13000g), and dispersed in 2 mL of PBS, pH 7.4, containing 1.0 wt % BSA for subsequent use. Preparation and Bioconjugation of Thi−HRP-Doped Silica Beads. Silica nanoparticles doped with Thi−HRP conjugates were prepared using a typical reverse micelle method.51 Triton X-100 (5.0 g), 1-hexanol (5.0 mL), and Thi−HRP conjugate (200 μL, 0.5 mg mL−1) were initially added to 10 mL of cyclohexane solution, and the resulting solution was vigorously stirred for 10 min at RT. Following that, 2 mL of TEOS and 1.5 mL of NH3·H2O (25 wt %) were slowly dropped into the mixture for further reaction (6 h at 4 °C). Afterward, 5 mL of acetone was injected into the mixture, which was separated by centrifugation (10 min at 4 °C, 8000g) and washed with ethanol/ultrapure water (five times). The obtained silica beads doped with Thi−HRP conjugates are designated as SiO2(Thi− HRP).
Next, the as-prepared SiO2(Thi−HRP) was utilized for the conjugation of Con A and anti-AFB1 by using GOPS as the crosslinkage reagent as follows: (i) 250 mg of the dried SiO2(Thi− HRP) obtained in a vacuum was added to 5 mL of GOPS (5.0%, v/v) in dry toluene and the resulting solution gently shaken for 6 h at 4 °C on the shaker; (ii) the nanoparticles were centrifuged for 10 min at 4 °C, and washed with toluene and ethanol three times; (iii) the obtained GOPS-modified SiO2(Thi−HRP) was directly added into 5 mL of PBS (pH 7.4) containing 100 μL of anti-AFB1 (0.5 mg mL−1) and 20 μL of Con A (0.5 mg mL−1), and the mixture was reacted for 12 h at 4 °C to make the antiAFB1 and Con A conjugate on the silica nanoparticles via the amine−epoxy reaction.54 Finally, the as-synthesized SiO2(Thi− HRP) conjugated with anti-AFB1 and Con A [designated as mAb−SiO2(Thi−HRP)−Con A] was centrifuged and dispersed into 2 mL of pH 7.4 PBS for further use. Immunosensor Fabrication. First, a graphene-modified glassy carbon electrode (GCE; 2 mm in diameter) was prepared by using the electrodeposition method according to the literature.55 Briefly, graphene oxide nanosheets were dispersed in PBS (67 mM, pH 9.18) by sonication to form 1.0 mg mL−1 graphene colloids. Following that, a cleaned GCE treated with H2SO4 was immersed into the suspension, and a cyclic voltammetric reduction was carried out between −1.5 and 0.5 V at 10 mV s−1 (15 cycles) with N2 bubbling at room temperature. Note: During this process, graphene oxide nanosheets were electrochemically reduced and deposited on the GCE. The aim of using N2 bubbling was to accelerate the deposition of graphene oxide on the electrode. After the GCE was washed with ultrapure water, DexP (10 μL, 3.0 μM) was dropped onto the inverted GCE, followed by incubation for 2 h at RT to cause DexP to adsorb onto the graphene-modified GCE via π-stacking interaction. After that, the resulting GCE was immersed into the aboveprepared mAb−SiO2(Thi−HRP)−Con A. Through dextran− Con A reaction, anti-AFB1 antibodies were immobilized onto the GCE accompanying the silica beads. Finally, the as-prepared immunosensor was stored at 4 °C when not in use. Immunoreaction and Displacement-Type Electrochemical Measurement. Scheme 1 shows the hydrolysateinduced displacement reaction with multifunctional silica beads doped with Thi−HRP conjugates for electrochemical immunoassay of target AFB1. All electrochemical measurements were carried out on a microAutoLab Type III system (Eco Chemie, C
DOI: 10.1021/acs.analchem.5b02253 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
Figure 1. (A) TEM image of En−AuNP−AFB1, (B) DLS data of AuNP and En−AuNP−AFB1, (C) FT-IR spectra of (a) AuNPs and (b) En−AuNP− AFB1, and (D) PGM readout signals toward different components: (a) 10 ng mL−1 En−AuNP−AFB1, (b) 50 mM sucrose + 10 ng mL−1 En−AuNP− AFB1, (c) 50 mM sucrose, (d) 50 mM sucrose + 10 ng mL−1 AuNPs, (e) 50 mM sucrose + 0.5 mg mL−1 AFB1−BSA, and (f) 50 mM sucrose + 10 ng mL−1 AFB1−BSA−AuNP.
utilized as the signal-transduction tags on the basis of the doped HRP toward the catalytic reduction of hydrogen peroxide with the aid of the thionine mediator. The immunosensor was fabricated via the reaction between the labeled Con A on the SiO2(Thi−HRP) and the immobilized dextran on the graphenemodified GCE. In the presence of target AFB1, the analyte initially competed with the labeled AFB1−BSA on the AuNPs for the immobilized anti-AFB1 antibody on the electrode. Upon addition of sucrose, the invertase immobilized on the AuNPs could hydrolyze the sucrose into glucose and fructose. Owing to the higher affinity of concanavalin A for glucose compared to dextran, the produced glucose displaced it from the electrode. Accompanying the displacement reaction, the whole mAb− SiO2(Thi−HRP)−Con A conjugate was separated from the electrode, thus decreasing the amount of the residual Thi−HRP. With increasing concentration of target AFB1 in the sample, the conjugated amount of En−AuNP−AFB1 on the electrode decreased, thereby indirectly increasing the residual amount of Thi−HRP on the electrode. In this case, the catalytic current by the residual Thi−HRP toward H2O2 increased with increasing target AFB1. By monitoring the change in the current, we might quantitatively evaluate the concentration of target AFB1 in the sample. Characterization of the En−AuNP−AFB1 Conjugate. As described above, glucose arises from hydrolysis of sucrose by invertase immobilized on the En−AuNP−AFB1 conjugate. Thus, the successful preparation of En−AuNP−AFB1 was crucial. Figure 1A shows a typical transmission electron microscopy (TEM; H-7650, Hitachi Instruments, Tokyo, Japan) image of the En−AuNP−AFB1, and the nanostructures were homogeneously dispersed in the buffer, which provided a precondition for the assay development. Also, we used dynamic light scattering (DLS;
The Netherlands) comprising a modified GCE working electrode, a platinum wire auxiliary electrode, and a Ag/AgCl reference electrode. The assay was carried out as follows. (i) Competitive-type immunoreaction for the immobilized antiAFB1 antibody on the GCE between target AFB1 and the labeled AFB1−BSA on the AuNP: The immunosensor was initially dipped into a 0.5 mL centrifugal tube containing 100 μL of En− AuNP−AFB1 prepared above and 100 μL of AFB1 standard/ sample, and then incubated for 30 min at 37 °C with gentle shaking on a Heating & Cooling ThermoMixer (MS-100, Aosheng Inc., Hangzhou, China). (ii) Invertase hydrolysateinduced displacement reaction with the glucose−Con A− dextran system: The resulting immunosensor was immersed into 200 μL of pH 6.0 PBS containing 0.5 mg mL−1 sucrose in the presence of Mn2+ and Ca2+ ions and hydrolyzed for 50 min at 37 °C. (iii) Electrochemical measurement: Differential pulse voltammetry (DPV) measurement was performed in pH 6.0 PBS containing 50 μM H2O2 from 0 to −400 mV (amplitude 50 mV, width 50 ms). Note: The resulting immunosensor should be washed with pH 7.4 PBS af ter each step. All measurements were performed in triplicate unless otherwise specified.
■
RESULTS AND DISCUSSION
Construction of a Signal-On Competitive-Type Immunoassay. To achieve a high-sensitivity competitive-type electrochemical immunoassay for quantitative monitoring of small molecules, a signal-on immunosensing protocol would be advantageous, especially for the determination of a lowconcentration target analyte. In this work, gold nanoparticles functionalized with invertase and AFB1−BSA were used for the signal amplification by the hydrolysis of the labeled invertase toward sucrose, while silica beads doped with Thi−HRP were D
DOI: 10.1021/acs.analchem.5b02253 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
Figure 2. (A) TEM image of mAb−SiO2(Thi−HRP)−Con A. (B) Cyclic voltammograms of (a) the bare GCE in pH 6.0 PBS and (b, c) the SiO2(Thi− HRP)-modified GCE in pH 6.0 PBS (b) without and (c) with 50 μM H2O2 at 50 mV s−1. (C) DLS data for SiO2(Thi−HRP), mAb−SiO2(Thi−HRP)− Con A, and En−AuNP−AFB1 + mAb−SiO2(Thi−HRP)−Con A.
Characterization of mAb−SiO2(Thi−HRP)−Con A. In this study, the Thi−HRP doped into silica beads was used as the signal-generation tag by catalytic reduction of HRP toward H2O2 (thionine used as the electron mediator). Therefore, the bioactivity of the doped Thi−HRP would directly affect the sensitivity of the immunoassay. By the same token, the mAb− SiO2(Thi−HRP)−Con A hybrid nanostructures were first characterized by using TEM (Figure 2A). As seen from Figure 2A, the as-synthesized nanostructures were similarly spherical. Importantly, conjugation of anti-AFB1 and Con A did not cause the aggregation of SiO2(Thi−HRP) nanoparticles. Next, the asprepared SiO2(Thi−HRP) colloids were directly dropped onto the cleaned GCE, and the bioactivity of the doped HRP was evaluated by using cyclic voltammetry at 50 mV s−1 in pH 6.0 PBS (Figure 2B). No redox peaks were observed at the bare GCE (curve a). When the GCE was modified with SiO2(Thi−HRP) colloids, however, a pair of well-fined redox peaks at −248 and −165 mV were obtained (curve b), suggesting that the thionine doped into nanometer-sized silica was a good mediator, and facilitated the electron transfer. Upon addition of 50 μM H2O2 into pH 6.0 PBS, significantly, an obvious catalytic characteristic appeared with an increase of the cathodic current and a decrease of the anodic current (curve c). The results revealed that the doped HRP could maintain high catalytic activity toward H2O2 reduction with the help of thionine. The electron-transfer pathway between HRP and thionine was described in detail in our previous paper59 as the following process: (i) H2O2 + HRP(red) → H2O + O2 + HRP(ox); (ii) HRP(ox) + Thi → HRP(red) + Thi(H)+; (iii) Thi(H)+ + H+ + 2e− → Thi. Therefore, Thi− HRP conjugates could be doped into the nanosilica particles by the reverse micelle method. Furthermore, we also employed the DLS technique to investigate the mAb−SiO2(Thi−HRP)−Con A before (top) and after (middle) reaction with En−AuNP−AFB1. As shown in Figure 2C, the average size of the as-prepared mAb−SiO2(Thi− HRP)−Con A was 56.7 nm, which was seemingly larger than that of the TEM result (Figure 2A). The reason might be the fact that the proteins were difficult to observe in the TEM image. Favorably, a mixture of the mAb−SiO2(Thi−HRP)−Con A with the En−AuNP−AFB1 could trigger the increasing size of the nanostructures (bottom, Figure 2C). The results indicated that two nanostructures could be linked together by the specific antigen−antibody reaction, which was conducive for the development of the competitive immunoassay. Characteristics of the Signal-On Electrochemical Immunoassay. To demonstrate that the prepared mAb−
Zetasizer Nano S90, Malvern, London, U.K.) to investigate the size of the gold nanoparticles before and after reaction with invertase and AFB1−BSA. As seen from Figure 1B, the average size of the nanoparticles increased from 15.3 to 26.1 nm after conjugation with invertase and AFB1−BSA. The reason for the increase in size originated from the labeled proteins. 56 Indistinctly, we could also observe from Figure 1A that a layer of materials was coated on the surface of the gold nanoparticles. To further clarify the presence of the proteins on the nanostructures, gold nanoparticles before and after conjugation with invertase and AFB1−BSA were characterized by using Fourier transform infrared (FT-IR; PerkinElmer, United States) spectroscopy (Figure 1C). As is well-known, the shapes of the infrared absorption bands of amide I groups at 1610−1690 cm−1 corresponding to the CO stretching vibration of peptide linkages and amide II groups around 1500−1600 cm−1 from a combination of N−H bending and C−N stretching can provide detailed information on the secondary structure of proteins.57,58 As seen from curve b, two characteristic peaks at 1655 and 1545 cm−1 were obviously observed after gold nanoparticles were reacted with invertase and AFB1−BSA. In contrast, the characteristic peak at 1640 cm−1 for gold colloids might be derived from the nonspecifically adsorbed citrate (curve a). The slight deviation between absorption wave numbers indicated the interaction between nanoparticles and proteins. The results preliminarily revealed that the proteins could be conjugated to the AuNPs, thus resulting in a size increase of the nanostructures. Another important concern for the hydrolysate-induced displacement reaction is determination of whether invertase was successfully, i.e., in the active state, immobilized onto the gold nanoparticles. To demonstrate this point, we used a commercialized Roche personal glucometer (PGM; AccuChem Active, Selangor Darul Ehsan, Malaysia) to monitor the 10 ng mL−1 En−AuNP−AFB1 suspension before (column a) and after (column b) reaction with 50 mM sucrose for 60 min at 37 °C (Figure 1D). As control tests, the PGM signals were recorded toward 50 mM sucrose (column c), 50 mM sucrose + 10 ng mL−1 AuNPs (column d), 50 mM sucrose + 0.5 mg mL−1 BSA−AFB1 (column e), and 50 mM sucrose + 10 ng mL−1 AuNP−BSA− AFB1 (column f) under the same assay modes. As shown in Figure 1D, almost no PGM signals were observed toward these control tests, while a strong PGM signal was obtained when using En−AuNP−AFB1 as the catalyst. Hence, we might make a conclusion that invertase was conjugated to the nanogold particles. E
DOI: 10.1021/acs.analchem.5b02253 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
Figure 3. SEM images of (A) the graphene-modified GCE (inset: bare GCE substrate), (B) dextran-functionalized substrate A, and (C) mAb− SiO2(Thi−HRP)−AFB1-functionalized substrate B. (D) DPV responses of (a) the newly prepared immunosensor, (b) sensor a after incubation with 0 ng mL−1 + En−AuNP−AFB1, (c) sensor b after reaction with sucrose, and (d) sensor a toward 0.01 ng mL−1 AFB1 by using the same assay mode in pH 6.0 PBS containing 50 μM H2O2.
Figure 4. Effects of (A) the volume ratio between 0.5 mg mL−1 invertase and 0.5 mg mL−1 AFB1−BSA for the preparation of En−AuNP−AFB1, (B) the volume ratio between 0.5 mg mL−1 anti-AFB1 and 0.5 mg mL−1 Con A for the preparation of mAb−SiO2(Thi−HRP)−Con A, (C) the incubation time for the antigen−antibody reaction (a) and hydrolysis/displacement time (b), and (D) the pH of PBS on the current of the electrochemical immunoassay (note that 0.01 ng mL−1 AFB1 was used in this case).
SiO2(Thi−HRP)−ConA could be immobilized onto the dextran-modified GCE, we also utilized scanning electron
microscopy (SEM; Hitachi 4800, Japan) to monitor the different types of sensing interfaces. As seen from Figure 3A, a layer of F
DOI: 10.1021/acs.analchem.5b02253 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
Figure 5. (A) Calibration curve of the competitive-type electrochemical immunoassay toward different concentrations of AFB1 standards (inset: corresponding DPV curves in pH 6.0 PBS containing 50 μM H2O2) and (B) specificity of the developed immunoassay toward AFB1 (0.01 ng mL−1), AFB2 (1.0 ng mL−1), AFG1 (1.0 ng mL−1), AFG2 (1.0 ng mL−1), OTA (1.0 ng mL−1), and OA (1.0 ng mL−1). Each data point represents the average value of three measurements, and the error bar represents the 95% confidence interval of the mean for the electrochemical signal; error bars are standard error of the mean (1σ) values.
flakelike structures was observed on the GCE when graphene oxide was electrochemically deposited onto the electrode. Relative to that of the bare GCE (Figure 3A, inset), the surface became more rough. Moreover, the surface roughness increased again when the graphene-modified GCE was incubated with phenoxy-derivatized dextran (Figure 3B), indicating that the dextran derivates were immobilized on the substrate through πstacking interaction between the phenoxy group and graphene nanosheets. Happily, a large number of nanoparticles appeared on the dextran-functionalized interface upon introduction of mAb−SiO2(Thi−HRP)−Con A (Figure 3C). On the basis of the SEM results, we might ensure that the immunosensor could be successfully fabricated by using phenoxy-derivatized dextran as the bridge on the graphene-modified interface. Maybe, one might ask whether the as-prepared immunosensor could be utilized for the detection of the target AFB1 based on the glucose−Con A−dextran competitive-type system. To this end, two AFB1 concentrations, 0 and 0.01 ng mL−1, were monitored by the same-batch immunosensors using DPV in pH 6.0 PBS containing 50 μM H2O2 (Figure 3D). Owing to the presence of SiO2 (Thi−HRP) on the electrode, the newly prepared immunosensor exhibited a high catalytic current toward 50 μM H2O2 (curve a). When the immunosensor was incubated with En−AuNP−AFB1 and 0 ng mL−1 AFB1, the DPV peak current gently decreased (curve b versus curve a). This might be attributed to the conjugated BSA and invertase (as a kind of biomacromolecule) competitive-type immunoreaction which hindered the electron transfer to some extent. Moreover, the DPV peak current largely decreased when sucrose was present in the incubation solution (curve c). In the presence of sucrose, the labeled invertase on the AuNP hydrolyzed sucrose to glucose, and the generated glucose displaced the mAb−SiO2(Thi− HRP)−Con A from the dextran-modified GCE on the basis of the glucose−Con A−dextran system, thus resulting in a decrease of the residual SiO2(Thi−HRP) on the electrode. In this case, the residual Thi−HRP (curve c) displayed a low catalytic current relative to the newly prepared immunosensor (curve a). Moreover, the DPV peak current increased (curve d versus curve c) when the immunosensor was utilized for the detection of 0.01 ng mL−1 AFB1. The results preliminarily revealed that our design could be used for the detection of the target AFB1. Optimization of the Experimental Conditions. To enhance the competitive ability of the target AFB1 with the labeled AFB1−BSA on the AuNPs for anti-AFB1 and increase the
hydrolytic efficiency of invertase, the labeled amount of AFB1− BSA and invertase should be optimized. As shown from Figure 4A, an optimal electrochemical signal could be acquired at a 4:1 volume ratio of invertase and AFB1−BSA by using 0.01 ng mL−1 AFB1 as an example. Although the invertase with the high concentration could increase the hydrolytic ability toward sucrose, it interfered with the reaction of AFB1−BSA with antiAFB1, possibly. Therefore, 200 μL of invertase (0.5 mg mL−1) and 50 μL of AFB1−BSA (0.5 mg mL−1) were used for the preparation of En−AuNP−AFB1 in 5 mL of gold colloids (5.0 ng mL−1) in this work. Certainly, the conjugated amounts of anti-AFB1 and Con A on the SiO2(Thi−HRP) also affect the sensitivity of the electrochemical immunoassay since En−AuNP−AFB1 was captured to the electrode by the conjugated anti-AFB1 antibody on the SiO2(Thi−HRP). At this condition, we not only ensure the highefficiency immobilization of mAb−SiO2(Thi−HRP)−Con A on the GCE by Con A, but also enhance the capture ability of antiAFB1 toward En−AuNP−AFB1. Figure 4B gives the effect of different volume ratios of anti-AFB1 and Con A on the current of the developed immunoassay. A maximum current was obtained at a volume ratio of 5:1 for anti-AFB1 and Con A. Hence, 100 μL of anti-AFB1 (0.5 mg mL−1) and 20 μL of Con A (0.5 mg mL−1) were used for the synthesis of mAb−SiO2(Thi−HRP)−ConA in 5 mL of PBS (pH 7.4) containing 250 mg of SiO2(Thi−HRP). In addition to nanostructure preparation, the sensitivity of the electrochemical immunoassay was influenced by some assay conditions, including the immunoreaction time, displacement time, and pH of PBS. This is because the antigen−antibody reaction, enzymatic hydrolysis, and dextran−Con A−glucose displacement reaction take some time to implement. As displayed in Figure 4C, DPV peak currents initially increased with increasing immunoreaction time and enzymatic hydrolysis time (i.e., dextran−Con A−glucose displacement time), and then tended to level off. An optimal current was achieved after 30 min of immunoreaction and 50 min of hydrolysis/displacement. To save assay time, 30 and 50 min were used for the antigen− antibody reaction and the hydrolysis of invertase toward sucrose, respectively. To adequately exert the catalytic efficiency of the doped HRP, we investigated the effect of the pH of PBS on the current response of the immunosensor. As seen from Figure 4D, the electrochemical signal initially increased with increasing pH values, and then decreased after pH 6.0. More surprisingly, the G
DOI: 10.1021/acs.analchem.5b02253 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
concentration of AFB1 prior to measurement. The results were compared with those obtained from a commercialized AFB1 ELISA kit (Diagnostic Automation Inc., LOD 5.0 pg mL−1). The data are listed in Table 1. Statistical comparison between two
currents tended to zero below pH 5.5. We suspected that this might be ascribed to the following issues: (i) Con A exists as a dimer below pH 5.5 and a tetramer between pH 5.8 and pH 7.0,25 and (ii) the redox reaction of thionine often involves a H+ ion.59 When the pH of PBS was less than 5.5, Con A was dissociated into a dimer, thus resulting in the automatic separation of SiO2(Thi−HRP) from the electrode regardless of the target AFB1. Taking all this into consideration, pH 6.0 PBS was used as the supporting electrolyte for electrochemical measurement. Analytical Performance of the Signal-On Electrochemical Immunoassay toward the Target AFB1. To evaluate the detectability of the electrochemical immunoassay, AFB1 standards with different concentrations were tested on the basis of hydrolysate-induced displacement reaction with multifunctional silica beads on the mAb−SiO2(Thi−HRP)−Con Amodified GCE under the optimal conditions. The assay was carried out by using DPV in pH 6.0 PBS containing 50 μM H2O2. As shown from the inset in Figure 5A, the DPV peak current increased with increasing AFB1 level in the sample. A linear correlation between the peak current and the logarithm of the AFB1 concentration was observed in the dynamic range from 0.003 to 20 ng mL−1 with a detection limit (LOD) of 2.7 pg mL−1 and a limit of quantification (LOQ) of 9.2 pg mL−1 at signal-tonoise ratios of 3 and 10, respectively (Figure 5A). The regression equation could be fitted to y (μA) = 11.371 ln C[AFB1] + 77.998 (ng mL−1, R2 = 0.996, n = 27). Each data point represents the average value obtained from three different measurements. The maximum relative standard deviation (RSD) was 6.4%. Furthermore, we studied the batch-to-batch precision and reproducibility of the immunoassay toward three concentrations (low, middle, high), 0.01, 1.0, and 10 ng mL−1 AFB1, and the RSDs were 9.3%, 5.7%, and 7.4% (n = 3), respectively. These results suggested a good reproducibility and precision of the electrochemical immunoassay. Inspiringly, the LOD of our design was obviously even lower compared to our previously reported value (LOD = 5.0 pg mL−1)52 and the existing commercialized AFB1 enzyme-linked immunosorbent assay (ELISA) kits (e.g., Quicking Biotech (100 ppt), MaxSignal (50 pg mL−1), MyBioSource (250 pg mL−1), and Diagnostic Automation Inc. (5 pg mL−1)). Due to the legal limit of AFB1 (
