Talanta 132 (2015) 126–131

Contents lists available at ScienceDirect

Talanta journal homepage: www.elsevier.com/locate/talanta

Quantum-DoT submicrobead-based immunochromatographic assay for quantitative and sensitive detection of zearalenone Hong Duan a, Xuelan Chen b, Wei Xu a, Jinhua Fu c, Yonghua Xiong a,n, Andrew Wang d a

State Key Laboratory of Food Science and Technology, Nanchang University, Nanchang 330047, China Key Laboratory of Functional Small Organic Molecule (Ministry of Education), Jiangxi Normal University, Nanchang 330022, China c Jiangxi Institute of Veterinary Drug and Feedstuff Control, Nanchang 330047, China d Ocean NanoTech, LLC., San Diego, CA 92126, USA b

art ic l e i nf o

a b s t r a c t

Article history: Received 24 July 2014 Received in revised form 28 August 2014 Accepted 31 August 2014 Available online 6 September 2014

Mycotoxin pollutants are commonly related to cereal products and cause fatal threats in food safety, and therefore require simple and sensitive detection. In this work, quantum-dot (QD) submicrobeads (QBs) were synthesized by encapsulating CdSe/ZnS QDs using the microemulsion technique. The resultant QBs, with approximately 2800 times brighter luminescence than the corresponding QDs, were explored as novel fluorescent probes in the immunochromatographic assay (ICA) for sensitive and quantitative detection of zearalenone (ZEN) in corns. Various parameters that influenced the sensitivity and stability of QB-based ICA (QB–ICA) were investigated and optimized. The optimal QB–ICA exhibits good dynamic linear detection for ZEN over the range of 0.125 ng/mL to 10 ng/mL with a median inhibitory concentration of 1.0170.09 ng/mL (n¼3). The detection limits for ZEN in a standard solution and real corn sample (dilution ratio of 1:30) are 0.0625 ng/mL and 3.6 mg/kg, respectively, which is much better than that of a previously reported gold nanoparticle-based ICA method. Forty-six natural corn samples are assayed using both QB–ICA and enzymelinked immunosorbent assay. The two methods show a highly significant correlation (R2 ¼0.92). Nine ZENcontaminated samples were further confirmed with liquid chromatography–tandem mass spectrometry (LC–MS/MS), and the QB–ICA results also exhibited good agreement with LC–MS/MS method. In brief, this work demonstrates that QB–ICA is capable of rapid, sensitive screening of toxins in food analysis, and shows great promise for point-of-care testing of other analytes. & 2014 Elsevier B.V. All rights reserved.

Keywords: Quantum dot submicrobeads Immunochromatographic assay Zearalenone Quantitative detection

1. Introduction Zearalenone (ZEN) is an estrogenic and carcinogenic mycotoxin produced by some Fusarium species, which can be found in corn, wheat, and cereal products. ZEN commonly remains in the food chain in the form of original molecules, and produces severe damage on the reproductive system of humans and animals [1]. Many countries have established regulations governing levels of ZEN in agricultural products. To minimize the risk to humans and animals, several analytical methods have been developed for the detection of ZEN in food samples, including gas chromatography (GC) [2], high-performance liquid chromatography (HPLC) [3,4], liquid chromatography-tandem mass spectrometry (LC-MS/MS) [5], as well as some immunochemical methods, such as enzyme-linked immunosorbent assay (ELISA) [6,7], fluorescence polarization immunoassay (FPIA) [8], and use of other n Correspondence to: State Key Laboratory of Food Science and Technology, Nanchang University, 235 Nanjing East Road, Nanchang 330047, PR China. Tel.: þ 86 791 8818 2405; fax: þ 86 791 8833 3708. E-mail address: [email protected] (Y. Xiong).

http://dx.doi.org/10.1016/j.talanta.2014.08.076 0039-9140/& 2014 Elsevier B.V. All rights reserved.

immunobiosensors [9,10]. However, these methods are unsuitable for on-site routine screening because they require professional laboratory conditions, skilled operators, expensive instruments, and timeconsuming sample pretreatments. Recently, use of gold nanoparticle-based immunochromatographic assay (AuNP–ICA) has been recommended for ZEN qualitative and quantitative detection because of its rapidity, ease of use, and suitability for on-site analysis [11]. For example, Shim et al. and Li et al. reported AuNPs–ICA for qualitative determination of ZEN with visual detection limits (vLOD) at 2.5 ng/mL and 1.0 ng/mL, respectively [12,13]. Wang et al. also used AuNP-ICA for rapid simultaneous quantification of ZEN and Fumonisin B1 (FB1), in which the LODs for ZEN in a standard solution and real corn sample (dilution ratio of 1:60) were 0.35 ng/mL and 21 mg/kg, respectively [14]. Compared with fluorescent nanoparticle probes, traditional AuNP–ICA is always limited by its relatively low sensitivity. Quantum dots (QDs) are ideal fluorescent labels and have been widely used for improving the detection sensitivity of ICA due to their incomparable optical properties, such as broad UV excitation with narrow fluorescent emission spectra, large molar extinction coefficient, and high quantum yield

H. Duan et al. / Talanta 132 (2015) 126–131

[15]. A few research groups have reported QD-based ICA methods (QD–ICA) for rapid detection of protein biomarkers and other small molecule compounds, and achieved high sensitivity [16,17]. Quantumdot submicrobeads (QBs), in which numerous QDs are embedded in a polymer matrix, exhibit advantages in improving the sensitivity of immunoassay due to their stronger fluorescence intensity than the corresponding single QD. Zhang et al. used QBs as luminescent amplification probes and developed a dot-blot sandwich immunoassay for ultrasensitive detection of hepatitis B-virus surface antigen with a LOD at 0.078 ng/mL [18]. Li et al. also reported a QB-based sandwich ICA for quantitative and sensitive detection of prostatespecific antigen, in which the LOD achieved about 12-fold enhancement compared with QDs as probes [19]. However, the use of QBs as fluorescent probes for competitive ICA sensors has not been reported. In the present study, highly luminescent QBs with carboxyl groups were introduced as ICA signal-amplification probes for sensitive and quantitative determination of ZEN in corns. A portable strip reader was used to record the fluorescent intensity of the test (FIT) and control (FIC) lines. The FIT/FIC ratio was adjusted to offset the interference from the inherent heterogeneity of the test strips according to our previous work [20]. Various parameters that influence the sensitivity and reproducibility of QB–ICA were systematically optimized. The performance of QB-ICA, including its LOD, median inhibitory concentration (IC50), specificity, accuracy and precision, was evaluated. Moreover, the practicality and reliability of the proposed quantitative strip were compared with a commercial ELISA kit and were further confirmed by LC-MS/MS.

127

Chemical (St. Louis, MO, USA). Donkey anti-mouse IgG antibodies were purchased from Beijing Zhongshan Biotechnology Inc. (Beijing, China). ZEN–BSA conjugates (mole ratio of 8:1) and anti-ZEN monoclonal antibodies (anti-ZEN mAbs) were prepared in our laboratory. The sample pad, nitrocellulose (NC) membrane, and absorbent pad were purchased from Schleicher and Schuell GmbH (Dassel, Germany). The ELISA kit was provided by Shanxi Defcred Biotech Co., Ltd (Shanxi, China). All other reagents were of analytical grade and purchased from Sinopharm Chemical Corp. (Shanghai, China). Forty-six natural corn samples were collected from the grain procurement agencies in Shandong Province, China. All samples were ground and well mixed prior to use. The sample extraction was as follows: 5.0 g of the pulverized sample was extracted with 25 mL methanol–water mixture (6:4, v/v) for 20 min with vigorous shaking. After centrifuging at 5000 g for 10 min, the supernatant solutions were diluted six-fold with 0.01 M phosphate buffer saline (pH 5.0, PBS) for further use. The BioDot XYZ platform, combined with a motion controller, BioJet Quanti3000 k dispenser, and AirJet Quanti3000 k dispenser for solution dispensing, was supplied by BioDot (Irvine, CA). The automatic programmable cutter was purchased from Shanghai Jinbiao Biotechnology Co., Ltd. (Shanghai, China). The portable fluorescent strip reader was obtained from Huguo Science Instrument Co., Ltd. (Shanghai, China). The water used for all the experiments was purified by Elix-3 and Milli-QA apparatus (Molsheim, France).

2.2. Preparation and characterization of QBs 2. Materials and methods 2.1. Reagents and apparatus ZEN, N-(3-dimethylaminopropyl)-N0 -ethylcarbodiimide hydrochloride (EDC  HCl), bovine serum albumin (BSA), and N-hydroxysuccinimide (NHS) were purchased from Sigma-Aldrich

As shown in Schematic 1A, the quantum-dot submicrobeads (QBs) were synthesized as previously reported with some modifications [21]. Briefly, CdSe/ZnS QDs with a maximum emission wavelength at 623 nm in CHCl3 (1 mL; 20 mg/mL) were mixed with 1 mL CHCl3 containing polymethyl methacrylate (60 mg/mL) and poly (maleic anhydride-alt-1-octadecene) (40 mg/mL). An

Schematic 1. (A) Schematic illustration of the quantum-dot submicrobead formation. (B) Schematic illustration of the QB-ICA method for quantitative detection of ZEN.

128

H. Duan et al. / Talanta 132 (2015) 126–131

emulsion was obtained with sodium dodecyl sulfate (3 mg/mL; 5 mL) in water, using an ultrasonic homogenizer for 2 min. The nonpolar solvent (CHCl3) was then evaporated using a rotary evaporator. The resulting water-soluble QBs were purified by centrifugation (6500 g, 10 min) and washed three times with pure water. Fluorescence intensity of the resultant QBs was monitored using a Hitachi F-4500 fluorescence spectrophotometer (Tokyo, Japan). The average size and morphology of the spMNBs were determined using a high-resolution transmission electron microscope (JEOL JEM 2100, Tokyo, Japan). 2.3. Preparation and characterization of anti-ZEN mAb-labeled QBs The unpurified ascitic fluid containing anti-ZEN mAbs was used to directly conjugate with QBs as previously reported with some modifications [22]. Briefly, 5 mg EDC, 0.6 mg QBs, and 300 mL of unpurified anti-ZEN ascitic fluid (0.3 mg/mL) were added to 1.8 mL of 0.01 M PB buffer (pH 6.0). After reacting at room temperature for 1 h under magnetic stirring, the mixtures were centrifuged at 13 500 g for 10 min. The QBs and anti-ZEN mAb conjugates (QBmAbs) were resuspended with 2.0 mL of PBS (0.01 M, pH 7.4) containing 2% fructose, 1% polyethylene glycol (PEG 20000), 5% sucrose, 1% BSA, and 0.4% Tween-20. The resuspension solution was stored at 4 1C for further use. Dynamic light scattering (DLS) analysis of the free QBs and QB-mAbs probe was performed using a particle size analyzer (Malvern Instruments Ltd., Worcestershire, UK) to confirm the immobilization of anti-AFB1 mAbs on the QB surface. 2.4. Preparation of QB-ICA sensor The QB-ICA sensor was made of three parts: sample pad, nitrocellulose (NC) membrane, and absorbent pad. The ZEN–BSA conjugates (0.8 mg/mL) and goat anti-mouse IgG (1.0 mg/mL) were spotted onto NC membranes as the test (T) and control (C) lines, respectively, and then dried at 37 1C for 12 h. The sample pads were treated with 20 mmol/L sodium borate buffer (pH 8.0) containing 1.0% (w/v) BSA, 0.1% (w/v) NaN3, and 0.25% Tween-20. The distance between the T and C lines was about 5 mm. After drying at 60 1C for 2 h, the sample pad, NC membrane, and absorption pad were attached to a plastic backing plate, and then cut into 4-mm-wide strips and packaged in a plastic casing for subsequent storage in a drying cylinder at room temperature. 2.5. Immunoassay procedure As presented in Schematic 1B, the QB-mAbs probe and sample solution were pre-mixed at room temperature for 1.0 min, and then added onto the sample pad. With the aid of the capillarity of the absorbent pad, the QB-mAbs and ZEN immunocomplex formed migrate across the NC membrane and react with the goat anti-mouse IgG while excess QB-mAbs is captured by the ZEN– BSA, which results in a fluorescent band on the control and test lines, respectively. The more ZEN in the sample extracts, the lower the fluorescence intensity appears on the test line. 2.6. Assay validation The accuracy and precision of QB–ICA were evaluated by analyzing the recovery and variation coefficients of the intra- and interassays [23]. The intra-assays were estimated by one batch of test strips for five replicates, while the inter-assays were analyzed once every day for three sequential days. The QB–ICA specificity was evaluated by analyzing the cross-reaction rate (Cr%) with five other common mycotoxins, including citrinin (CIT), ochratoxin A (OTA),

deoxynivalenol (DON), FB1, and aflatoxin B1 (AFB1). Cr% was calculated according to the following equation: Cr%¼[(IC50 ZEN)/(IC50 analog)]  100% [24]. The practicality of the QB–ICA sensor was compared with a commercial ELISA kit by analyzing 46 natural corn samples. To evaluate the method reliability, nine ZEN-contaminated samples were further confirmed with LC-MS/MS system (Agilent Corporation, MA, USA), which composed of a triple-quadrupole instrument (Agilent 6410) and an LC system (Agilent 1200 series). The sample cleaning up and LC–MS/MS operation were performed according to local standard GB/T 22286-2008 (Sichuan, China) with some modifications. Briefly, 5 g of ZEN-contaminated sample was extracted with acetonitrile: water (90:10, v/v) for 20 min on a horizontal shaker. After centrifugation at 8000 g for 10 min, 1 mL of supernatant and 9 mL of ultrapure water were mixed, and then passed through an immune affinity column which purchased from Beijing Rapid Bioscience Co., Ltd. (Beijing, China). The purified solution was filtered through a 0.22 μm cellulose membrane for the LC–MS/MS analysis. The LC system conditions for ZEN detection were as follows: the column was Agilent Zorbax Eclipse XDB-C18 (100 mm  2.1 mm, 2.4 μm), which was maintained at 35 1C for chromatographic separation. The flow rate of the mobile phase was 0.30 mL/min and the injection volume of sample solution was 5 μL. The mobile phase consisted of solvent A (ammonium acetate, 10 mM) and solvent B (methanol). The chromatographic separation was performed as a gradient elution as follows: 0 to 6.0 min (80% A, 20% B), 6.0 to 6.1 min (15% A, 85% B), 6.1 to 8.1 min (5% A, 95% B), 8.1 to 12.0 min (80% A, 20% B). Ionization was achieved using electrospray ionization in anion mode with a spray voltage of  3500 V. The MS/MS acquisition was performed in multiplereaction monitoring mode. The temperature and flow velocity for the desolvation gas (high-purity nitrogen, 99.99%) were 350 1C and 13 L/min, respectively. The monitoring ion pairs were chosen as ZEN m/z 317.1/175.2 (quantitation ion) and 317.1/131.2 (qualitative ions).

3. Results and discussion 3.1. Characterization of QBs and QBs-mAbs conjugates The highly luminescent QBs were obtained by encapsulation of organic-soluble CdSe/ZnS QDs according a microemulsion method. To evaluate the fluorescent enhancement of the resultant QBs, fluorescence intensities of water-soluble CdSe/ZnS QDs with a maximum emission wavelength at 623 nm and the resultant QBs were monitored. As shown in Fig. 1A, the fluorescence intensity of QBs at concentration of 3.5  10  3 nM (dissolved in ultrapure water) was the same as that of water-soluble CdSe/ZnS QDs at 10 nM (dissolved in ultrapure water). Hence for the same numbers of QB and QD particles, the luminescence intensity of QBs was 2857 times higher than that of QDs according to the ratio of their concentrations under the same fluorescence intensity [Con.QD/ Con.QB ¼ (10 nmol/L)/(3.5  10  3 nmol/L) ¼2857]. High-resolution transmission electron microscope (HRTEM) images (Fig. 1B) show that the QBs have relatively uniform size distribution with an average diameter of 247 nm, while Fig. 1B (inset) further indicates that numerous dark QDs are tightly embedded in the polymer matrix. Fig. 1C confirms a macroscopic metal lattice structure of CdSe/ZnS QDs. DLS analysis shows that the hydrodynamic diameter of the free QBs is about 255 nm, whereas that of the QBsmAbs increases to about 295 nm. The results confirm that the antiZEN mAbs was successfully coupled on the surface of the QBs. 3.2. Optimization of the QB-based ICA and detection conditions The QBs-mAbs probe was obtained by coupling the amino group of anti-ZEN mAbs with the carboxyl group of the QBs using

H. Duan et al. / Talanta 132 (2015) 126–131

Fig. 1. Characterization of QBs. (A) The fluorescence intensities of water-soluble CdSe/ZnS QDs and the resultant QBs. The concentrations of QDs and QBs were 10 nM and 3.5  10  3 nM, respectively. (B) High-resolution transmission electron microscope image of the quantum-dot submicrobeads. Inset: image of individual quantum-dot submicrobeads at high magnification. (C) Part of the individual quantum-dot submicrobead image at high magnification.

an active ester method. The unpurified anti-ZEN ascites were used to label QBs directly to reduce the QB nonspecific bond with the NC membrane and the ZEN–BSA on the test line, because the mass of miscellaneous proteins in the ascites blocked the excess carboxyl groups of the QBs when the anti-ZEN mAbs bound on the surface of the QBs. To achieve the best sensitivity and high FI signals on both lines, a similar checkerboard titration was performed with a series of ZEN–BSA concentrations on the test line for various different contents of the QB-mAbs probe. The FI signals on both lines were recorded at 30 min. The means of FI on both lines were based on three duplicate measurements with the blank corn extract solution. The competitive inhibition rates are obtained by (1  Bx/B0)  100%, where B0 and Bx represent FIT/FIC of the negative sample and a ZEN-spiked corn extract solution (1.0 ng/mL), respectively. As shown in Table 1, the optimal combinations were as follows: 0.8 mg/mL ZEN–BSA on the test line and 1 mL of QB-mAbs (60 mg/mL) pre-mixed with 69 mL sample solution. Under the optimal combination, the means of the FIT and FIC signals on both lines were 748 718 and 539 720, respectively. The competitive inhibition rate for the 1 ng/mL spiked sample was 59.9 70.7% (n ¼ 3). In a conventional strip, the labeled probes are irregularly sprayed on the conjugated pad which can hardly be standardized and easily lead to certain variability of FIs on both lines between the different strips [25,26]. In the present study, the conjugated pad was integrated with sample pad, and the QB-mAbs probe was introduced to pre-incubate with the ZEN standard (or sample) solution for 1 min to improve the quantitative reproducibility of QB-ICA sensor (Schematic 1B). The immunological dynamic analysis was used to optimize the detection conditions that influence the stability and sensitivity of ICA, including the interpretation time, the pH value, and methanol content in the sample solution. The immunological dynamic curve for QB-ICA was obtained as described in our previous work [20]. Briefly, 1 min after the solution was added onto the sample pad, the strip was inserted into a portable fluorescent reader. The FIT, FIC, and FIT/FIC ratio were recorded every 1 min for 30 min. The kinetic curves representing the interaction between the QB–mAbs probe and antigens (on NC membrane) are described by plotting the values of FIT, FIC, and FIT/FIC against running time. As shown in Fig. 2A, FI signals on both lines continue to increase sharply in the initial 25 min, and then kept gently increasing for the succeeding 30 min observation

129

time. However, the FIT/FIC ratio quickly stabilized after 10 min. Thus, the interpretation time for ZEN quantitative analysis was set at 10 min for all succeeding studies. The effect of pH value on the sensitivity of the strip was evaluated by running a series of solutions with pH values in the range of 5.0 to 9.0. The results in Fig. 2B indicate that the FIT/FIC value of the negative sample significantly decreased from 1.54 7 0.10 to 0.60 70.08 as the pH increased from 5.0 to 9.0. The competitive inhibition rate for the 1.0 ng/mL ZEN-spiked sample reached a maximum of 63% 76% in the pH 5.0 solution. These results indicate that weak acidic conditions (pH 5.0) improve the sensitivity of QB-ICA by promoting anti-ZEN mAbs and ZEN-BSA recognition. The effect of methanol concentration on the sensitivity of QB-ICA was also investigated. The results, shown in Fig. 2C, indicate that the inhibition of QB-ICA continuously declined from 62.5%7 4% to 32.8% 76% as the methanol concentration increased from 0 to 15%, and then remained relatively stable from 32.8 7 6% to 34.4 71% as the methanol content increased further to 25%. Due to the hydrophobicity of the ZEN molecule, the methanol content in the buffer was set at 10%. Considering the above results, the optimized experimental conditions of QB-ICA for ZEN quantitative analysis are recommended as follows: the corn extract containing 60% methanol was diluted six-fold with 0.01 M PBS (pH 5.0); 69 mL of the diluted solution and 1 mL of QB-mAbs probe (60 mg/mL) were pre-mixed at room temperature for 1.0 min and then added onto the sample pad; after 10 min, the strip was scanned by a portable fluorescent reader for ZEN quantitative analysis.

3.3. Assay validation Under optimal experimental conditions, a standard curve for the QB-ICA method was constructed by plotting the ratio of BX/B0 against the logarithm of ZEN concentrations. The standard solutions were prepared by spiking a ZEN stock solution (1.0 mg/mL) with a blank corn sample extract to final concentrations of 0, 0.0625, 0.125, 0.35, 0.75, 2.0, 5.0, 7.5, 10, 15, and 20 ng/mL, respectively. The results, shown in Fig. 2D, indicate that the method exhibits a wide dynamic linear range of 0.125 ng/mL to 10 ng/mL with a median inhibitory concentration (IC50) of 1.0 ng/mL (n¼3). The regression equation of the standard curve was y¼  0.160ln(x)þ0.504 with a reliable correlation coefficient (R2 ¼ 0.994). The LOD of the QB-ICA method was calculated at 10% inhibitory concentration to be 0.0625 ng/mL, which is 5.6 times higher than that of the previously reported AuNPs–ICA quantitative method. The LOD for a real corn sample was 3.6 mg/kg, according to the mean determined concentration of 20 randomly negative corn samples plus three-fold standard deviation [27,28], and then multiplied by the dilution factor of 30. The intra-assay recovery rates for ZEN-spiked concentrations of 0.5, 1.0, and 2.0 ng/mL were 108.0 73.0%, 97.071.5%, and 89.5 71.7%, and those for the interassay were 106.0 71.9%, 95.0 73.5%, and 93.5 74.4%, respectively (Table 2). The above results indicate that the precision and accuracy of the QB–ICA method are acceptable for ZEN quantitative detection. The QB-ICA specificity exhibits negligible crossreactivity with other five common mycotoxins: CIT, OTA, DON, FB1, and AFB1. To verify the practicability of the QB–ICA sensor, 46 real corn samples were analyzed using both QB–ICA and a commercial ELISA kit. The results indicate that the two methods have a good agreement (R2 ¼0.92, Fig. 3A). Nine random corn samples, found to be ZEN-contaminated using QB-ICA, were further analyzed by LC–MS/MS to confirm the reliability of QB-ICA. The results from

130

H. Duan et al. / Talanta 132 (2015) 126–131

Table 1 Optimization of the concentrations of QB-mAbs probe and ZEN-BSA using a checkerboard titration. No.

The concentration of ZEN-BSA (mg/mL)

The volume of QB-mAbs (μL)

The FI of test linesa

The FI of control linesa

FIT/FIC

The inhibitionrate (%)b

1 2 3c 4 5 6 7 8 9

0.8 0.8 0.8 0.6 0.6 0.6 0.4 0.4 0.4

2.0 1.5 1.0 2.0 1.5 1.0 2.0 1.5 1.0

10497 33 8447 16 748 7 18 906 7 14 7267 6 480 7 21 669 7 28 532 7 44 328 7 15

839 7 4 6277 5 539 7 20 9337 24 7087 3 4077 20 1029 7 31 7577 45 5337 27

1.25 70.03 1.3570.02 1.39 70.02 0.97 70.01 1.02 70.01 1.2 70.01 0.65 70.04 0.7070.07 0.6 70.04

35.417 1.66 44.54 7 1.60 59.90 7 0.70 28.45 7 2.80 37.86 7 2.17 52.85 7 2.96 33.38 7 4.57 42.03 7 0.22 30.417 5.96

a b c

The means of the FIT and FIC values are based on three duplicate measurements with the negative corn extract solution. The inhibition rates are obtained from the 1 ng/mL ZEN-spiked sample. The optimal parameters of QB–ICA.

Fig. 2. Optimization detection conditions of the QB-ICA. (A) Immunoreaction dynamics of FIT, FIC, and FIT/FIC ratios. (B) The effects of pH on the FIT/FIC and inhibition ratios. (C) The effects of methanol concentration on the FIT/FIC and inhibition ratios. (D) Calibration curves for ZEN using mouse anti-ZEN mAbs-labeled QBs.

QB-ICA sensor also shows a highly significant correlation with LCMS/MS method (R2 ¼0.96, Fig. 3B).

4. Conclusions In summary, highly luminescent QBs have been employed to provide a competitive ICA system for rapid (10 min) and quantitative

detection of ZEN in corn samples, in which the QBs have more than 2800 times brighter luminescence than the corresponding CdSe/ZnS QDs. The optimal QB-ICA exhibits a detection limit of 0.0625 ng/mL for ZEN, which is 5.6 times higher than that of a previously reported AuNPs-ICA quantitative method. Moreover, QB-ICA shows acceptable precision and accuracy, and has a highly significant correlation to the conventional ELISA and LC–MS/MS methods. On account of its significant advantages in luminescent signal amplification, the

H. Duan et al. / Talanta 132 (2015) 126–131

Table 2 Precision and accuracy of QB-based ICA in ZEN-spiked samples. Spiked ZEN (ng/mL)

0.5 1.0 2.0 a b

Inter-assaya

Intra-assay Recovery (%)b

CV (%)

Recovery (%)b

CV (%)

108.0 97.0 89.5

3.0 1.5 1.7

106.0 95.0 93.5

1.9 3.5 4.4

Assay was completed every 1 days for 3 days continuously. Recovery was obtained from five replicates at each spiked concentration.

Fig. 3. Methodology comparison. (A) Methodology comparison between the QB-ICA and ELISA methods (n ¼46). (B) Methodology comparison between the QB-ICA and LC-MS/MS methods (n¼ 9).

proposed QB-ICA method offers great potential for rapid and sensitive analysis of various contaminants in food safety monitoring. Acknowledgments This work was supported by a grant from the National Basic Research Program of China (2013CB127804), “Twelfth Five-Year

131

Plan“ for National Science and Technology Support Program (2013BAD19B02), the National Natural Science Foundation of China (Grant No. 31160323), the Training Plan for the Main Subject of Academic Leaders of Jiangxi Province (20142BCB22004), and National Institute of Health of United State (1R43AI092962).

References [1] A. Zinedine, J.M. Soriano, J.C. Molto, J. Mane, Food Chem. Toxicol. 45 (2007) 1–18. [2] V.G. Amelin, N.M. Karaseva, A.V. Tretyakov, J. Anal. Chem 68 (2013) 61–67. [3] Y.L. Wang, T.J. Chai, G.J. Lu, Environ. Res. 107 (2008) 139–144. [4] E.O. Bennekom, L. Brouwer, E.H.M. Laurant, H. Hooijerink, M.W.F. Nielen, Anal. Chim. Acta. 473 (2002) 151–160. [5] E. Perez-Torrado, J. Blesa, J.C. Molto, Food Control 21 (2010) 399–402. [6] Y.H. Huang, Y. Xu, Q.H. He, Food Agric. Immunol. 25 (2014) 186–199. [7] A. Kuzdralinski, E. Solarska, M. Muszynska, Food Control 29 (2013) 22–24. [8] A.P. Bondarenko, S.A. Eremin, J. Anal. Chem 67 (2012) 790–794. [9] E. Schneider, E. Usleber, E. Martbaur, R. Dietrich, G. Terplan, Food Addit. Contam. 12 (1995) 387–393. [10] J.L. Urraca, J.L. Benito-Pen, C. Perez-Conde, M.C. Moreno-Bondi, J. Agric. Food Chem. 53 (2005) 3338–3344. [11] A.Y. Kolosova, S. De Saeger, L. Sibanda, R. Verheijen, C.V. Peteghem, Anal. Bioanal. Chem. 2389 (2007) 2103–2107. [12] X. Li, P.W. Li, Q. Zhang, Biosens. Bioelectron. 49 (2013) 426–432. [13] W.B. Shim, K.Y. Kim, K.Y. Chung, J. Agric. Food Chem. 57 (2009) 4035–4041. [14] Y.K. Wang, Y.X. Yan, W.H. Ji, H.A. Wang, J. Agric. Food Chem. 61 (2013) 5031–5036. [15] X. Michalet, F.F. Pinaud, L.A. Bentolila, Science 307 (2005) 538–544. [16] Q. Yang, X. Gong, T. Song, J. Yang, S. Zhu, Y. Li, Biosens. Bioelectron. 30 (2011) 145–150. [17] Z. Zou, D. Du, J. Wang, J. Smith, C. Timchalk, Y. Li, Y. Lin, Anal. Chem 82 (2010) 5125–5133. [18] P.F. Zhang, Q. Li, J. Chen, H. Yan, W. Ma, Theranostics 4 (2014) 307–315. [19] X. Li, W.B. Li, Q.H. Yang, X.Q. Gong, W.S. Guo, C.H. Dong, J.Q. Liu, L.X. Xuan, J. Chang, Mater. Interfaces (2014), http://dx.doi.org/10.1021/am5012782. [20] X.L. Huang, X.L. Aguilar, H.M. Li, W.H. Lai, H. Wei, H.Y. Xu, Y.H. Xiong, Anal. Chem. 85 (2013) 5120–5128. [21] R. Xie, U. Kolb, J. Li, T. Basché, A. Mews, J. Am. Chem. Soc. 127 (2005) 7480–7488. [22] R.K. Mendes, R.F. Carvalhal, D.R. Stach-Machado, L.T. Kubota, Biosens. Bioelectron. 24 (2009) 2483–2487. [23] W. Xu, X.L. Chen, X.L. Huang, Y.H. Xiong, Talanta 114 (2013) 160–166. [24] H.M. Li, W.H. Lai, H.Y. Xu, B. Xu, W. Luo, H. Wei, S.X. Wang, Y.H. Xiong, Chinese J. Anal. Chem. 39 (2011) 1647–1652. [25] V. Gubala, L.F. Harris, A.J. Ricco, M.X. Tan, D.E. Williams, Anal. Chem. 84 (2011) 487–515. [26] L. Anfossi, M. Calderara, C. Baggiani, C. Giovannoli, E. Arletti, G. Giraudi, Anal. Chim. Acta 682 (2010) 104–109. [27] S.W. Oh, Y.M. Kim, H.J. Kim, H.J. Kim, J.S. Cho, E.Y. Choi, Clin. Chim. Atca 406 (2009) 18–22. [28] J. Pleadin, J. Vulić, J. Perši, N. Vahčić, Meat Sci 86 (2010) (2010) 733–737.