Rapid detection of zearalenone and its metabolite in corn flour with the immunochromatographic test strip

ABSTRACT Zearalenone (ZEN) is a mycotoxin produced by fungal species which are widespread in nature and commonly contaminate many cereal grains, such as wheat, maize, corn, barley, and other cereal grains. An immunochromatographic (ICA) test strip was developed for the rapid simultaneous detection of ZEN and its metabolite in corn flour samples. For the ICA test, antigen (ZEN-BSA) and goat anti-mouse IgG were, respectively, drawn on nitrocellulose membrane as control line and test line. Under the optimized condition, the cut-off limits of test strips for ZEN, α-ZAL, and β-ZAL were found to be 0.5 ng/mL in 0.01 M PBS and 50 ng/mL in corn flour samples, meanwhile α-ZOL, β-ZOL, and ZEA were found to be 1 ng/mL in 0.01 M PBS and 75 ng/mL in corn flour samples. Our data indicate that the ICA is sensitive, rapid, cost-effective, and specific on-site screening for ZEN and its metabolite.


Introduction
Mycotoxins are toxic metabolites produced under particular environmental conditions by fungi and contaminate a large variety of grains and feedstuffs worldwide (Anater et al., 2016;Serrano, Font, Maes, & Ferrer, 2016). Infection occurs through ingestion, inhalation, or skin contact with fungi. Cereal grains can be infected by mycotoxigenic fungi at preharvest or post-harvest. These grains can be consumed by humans leading to contamination of plant origin food (cereal grain) (Glenn, 2007). It is estimated that approximately 25% of the world's crop production is contaminated with mycotoxins, which can result in the reduction of livestock production, pose a great potential health risk to consumers, and will lead to tremendous economic losses (Chen et al., 2016;Sirot, Fremy, & Leblanc, 2013) Zearalenone (Figure 1, ZEN), a non-steroidal β-resorcyclic acid lactone with the systematic name of 3, 4, 5, 6, 9, 10-hexalydro-14, 16-dihydroxy-3-methyl-[S-(E)]-1H-2-benzoxacyl-otetradecin-1, 7(8H)-dione, mainly produced by several Fusarium species, such as Fusarium graminearum (Gibberella zeae), Fusarium culmorum, Fusarium cerealis, Fusarium equiseti, Fusarium crookwellense, and Fusarium semitectum, which are widespread in nature and commonly contaminate many cereal grains, such as wheat, maize, corn, barley, and other cereal grains (Cetin & Bullerman, 2006). ZEN shows only a low acute toxicity and there is limited evidence for its carcinogenicity and was proposed to be better fit the classification of a non-steroidal oestrogen or a myco-oestrogen. Because of the potential health risk, the provisional maximum tolerable daily intake (PMTDI) was 0.5 µg/kg body weight per day for ZEN. To reduce the intake ZEN maximum limits (MLs) for ZEN, it has been regulated in many countries and global commissions ranging from 20 to 3000 µg/kg in various food and feed stuffs (Briones-Reyes, Gomez-Martinez, & Cueva-Rolon, 2007;Chang et al., 2017;Liu et al., 2012;Meneely et al., 2010) Due to the potential impact on human health and economical aspects, the development of a simple, rapid, and low-cost analytical methods of ZEN and its metabolite was urgently in need. The most common analytical techniques include thin-layer chromatography (TLC) (Swanson, Corley, White, & Buck, 1984), high-performance liquid chromatography (Chen, Luan, Wang, Wang, & Shao, 2017;Ok, Choi, Kim, & Chun, 2014;Sadeghi et al., 2016), and liquid chromatography-tandem mass spectrometry (LC-MS/MS) (De Santis et al., 2017;Kim et al., 2017). TLC is now rarely used because of its low sensitivity and specificity (Khaemba et al., 2016;Suryoprabowo, Liu, Peng, Kuang, & Xu, 2015). HPLC and LC-MS/MS are the methods that are most frequently used due to high sensitivity and accuracy, but these methods have disadvantages as they are complex, time-consuming, and labour intensive Ding, Liu, Song, Kuang, & Xu, 2017;Kong et al., 2017;Liu, Suryoprabowo, Zheng, Song, & Kuang, 2017b).
A variety of immunoassay methods used for ZEN detection have recently been described, including enzyme-linked immunosorbent assay (ELISA) (Bolechova et al., 2015;Du et al., 2014;Huang et al., 2014;Pei et al., 2013). The ELISA has high specificity, is applicable for high-throughput applications, requires simple sample preparation steps, and is cost-effective Peng, Liu, Kuang, Cui, & Xu, 2017;Suryoprabowo et al., 2015). However, the lateral-flow immunochromatographic assay (ICA) is increasingly popular because it is simple, rapid, specific, and sensitive. The ICA method has advantage that all the reagents are included in the strip and the results can be obtained within 5-10 min Liu, Suryoprabowo, Zheng, Song, & Kuang, 2017a;.
Nitrocellulose (NC) high-flow plus membranes (Pura-bind RP) were obtained from Whatman-Xinhua Filter Paper Co., Ltd. (Hangzhou, China). Glass fibre membrane (CB-SB08) used for the sample pad, polyvinylchloride (PVC) backing material and absorbance pad (SX18) were supplied by Goldbio Tech Co., Ltd. (Shanghai, China). A BioDot TSR 3000 Membrane Strip Reader (Gene Company Limited, Shanghai Branch, Shanghai (China)) was used to test the colour intensities of colloidal gold in the test zone. The complete Freund's adjuvant (FCA), incomplete Freund's adjuvant (FIA), and enzyme immunoassay-grade HRP-labeled goat anti-mouse immunoglobulin were obtained from Sigma (St. Louis, MO, USA). BSA and specific monoclonal antibodies (anti-ZEN monoclonal antibody (mAb), No. #3) were generated in our laboratory.
Preparation and characterization of monoclonal anti-ZEA antibody ZEA was initially derived using carboxymethoxylamine (CMO) (Figure 2) with protein and then the ZEA-CMO-BSA or ZEA-CMO-KLH was synthesized by the active ester method (Kong, Liu, Song, Suryoprabowo, et al., 2016;Suryoprabowo et al., 2015). Briefly, a mixture of ZEN-CMO, carboxyl-reactive carbodiimide cross linker (1-ethyl-3-[3dimethylaminopropyl]carbodiimide hydrochloride) (EDC), and N-hydroxysuccinimide (NHS) was added to N,N-dimethyl formamide (DMF) and incubated for 24 h (solution 1). KLH or BSA was mixed with 0.05 mol/L NaHCO 3 (solution 2). Solution 1 was slowly added to solution 2 with stirring and the mixture was then stirred continuously for 8 h at room temperature. The conjugates were dialysed against distilled water for 3 days and subsequently with 0.01 M PBS for a further 3 days. Ultraviolet absorption was used to check the conjugation of protein and ZEN (Figure 3).
We used FCA in the first immunization and FIA was used in the subsequent booster injections. The mice were immunized with ZEN-KLH every three weeks with 100 µg (first immunization) and 50 µg (subsequent injections; two to five times). We analysed the antibodies in blood samples of mice using ELISA. The mouse with the highest titre was sacrificed and its splenocytes were fused with Sp 2/0 murine myeloma cells. The resulting hybridomas were screened using indirect ELISA. The selected hybridoma cells were expanded and injected into BALB/c mice to produce mAbs. The ascites were harvested and purified by the caprylic acid-ammonium sulphate precipitation method , and the purified mAbs were stored at −20°C.

Development of the immunochromatographic strip
Preparation of colloidal gold nanoparticles All the solvents were prepared with ultrapure water and passed through a 0.22 µm filter membrane. Chlorauric acid (0.1 g/L, 25 mL) was heated to boiling under constant stirring. Following the addition of 1.0 mL of 1% w/v sodium citrate tribasic dehydrate, the mixture was stirred for 30 min. When the colour of the solution turned wine-red, it was allowed to cool to room temperature, and stored at 4°C. Observations by transmission electron microscopy (TEM) revealed that the gold nanoparticles (GNPs) had a nearly uniform particle size of 15 nm. Preparation of colloidal gold-labelled mAb We used anti-ZEN mAb (no. #3) with a half-maximal inhibitory concentration (IC 50 ) of 0.1 ng/mL. Colloidal GNP solution (10 mL) was adjusted to pH 7.0 with 0.1 M K 2 CO 3 (40 µL). Briefly, mAbs (0.50 mL) were added to colloidal GNP solution dropwise. After 35 min, 1 mL of 10% (w/v) BSA was added, and the mixture was stirred for 2 h. The product was centrifuged at 8000 rpm for 45 min to remove any gold aggregates. The solution was separated into two layers; the lower layer (gold-labelled mAbs) was rinsed with 0.02 M phosphate buffer containing 5% sucrose, 1% BSA, and 0.5% polyethylene glycol (PEG) 6000 (pH 7.4). The conjugation products were reconstituted to 1 mL using goldlabeling resuspension buffer (0.02 M PBS, 5% sucrose, 2% sorbitol, 1% mannitol, 0.1% PEG, 0.1% tween, and 0.04% NaN 3 ) and stored at 4°C.

Test procedure and principle
Prior to the test, 50 µL of gold-labelled mAbs is mixed with 150 µL of the sample solution, allowed to react for about 5 min, and added onto the sample pad. The solution migrates to the absorbent pad, and the test results are visualized within 5 min. In ZENpositive samples, ZEN competes with ZEN-BSA conjugates embedded in the test (T) line for the finite amount of anti-ZEN mAb. When a sufficient amount of ZEN is present, free ZEN binds to gold-labelled mAb, preventing mAb from binding to ZEN-BSA in the T line. Therefore, the higher the ZEN contents in the sample, the weaker the colour of the T line. In ZEN-negative samples, the limited amount of colloidal gold-labelled mAb is trapped by the immobilized ZEN-BSA conjugate, and a clearly visible red T line appears.
The sample must reach the control (C) line, which contains goat anti-mouse IgG antibody. Therefore, the C line must always appear in a successful test, whereas the T line will only appear when the sample is negative (Figure 4(A)). The appearance of the C line alone is indicative that ZEN is present (Figure 4(B)), whereas if neither the C line nor the T line appears (Figure 4(C)), the test procedure is incorrect or the strip is invalid, indicating that the test should be repeated with a new strip.

Sensitivity of the test strip
The sensitivity of the test strip was determined by testing ZEN, α-ZAL, β-ZAL, α-ZOL, β-ZOL, and ZEA-containing samples. ZEN, α-ZAL, and β-ZAL standard was diluted to 0, 0.01, 0.025, 0.05, 0.1, 0.25, and 0.5 ng/mL, meanwhile α-ZOL, β-ZOL, and ZEA standard was diluted to 0, 0.05, 0.1, 0.25, 0.5, and 1 ng/mL in 0.01 M PBS (pH 7.4) and the detection limit was determined. The sample solution (150 µL) was mixed with 50 µL gold-labelled mAb, allowed to react for 5 min, and added onto the sample pad. After 5 min, the test strip reader was used to measure the colour intensities. The lowest detection limit (LDL) that could be detected by the naked eye was defined as the amount of ZEN, α-ZAL, β-ZAL, α-ZOL, β-ZOL, and ZEA that produced a colour reaction on the strip that was visibly different in intensity from that obtained from negative (0 ng/mL). Six replicates for each concentration were analysed on the same day.

Preparation of antibody-AuNPs (mAb-GNPs) conjugate
In the production of AuNPs, the particle size is inversely proportional to the sodium citrate volume. In this study, we choose 1 mL of 10% sodium citrate. The colour of HAuCl 4 changed from yellow to wine-red. Colloidal GNPs with a diameter 15 nm were selected because of their stability, simplicity, and bright colour. The GNPs with size 5 nm do not have the bright red colour, hence they give only a very weak to no signal at the detection line. Amount of 0.1 M KC 2 CO 3 was added into the AuNPs solution to ensure the AuNPs solutions are stable.
The concentration of mAbs in GNPs of the coating antigen affects the sensitivity of the assay. In this study, two different concentrations of mAbs (8 and 10 µg/mL) were allowed to react with ZEN-negative (0 ng/mL) and ZEN-positive spiked in 0.01 M PBS (1 ng/mL). Based on Figure 5, there were significant differences in colour intensity between 8 and 10 µg/mL. We observed that the optimum concentration of mAb in GNP was 10 µg/ mL. Additionally, there was a significant difference in the colour intensity at 0.05 and 0.1 mg/mL. It was found that the coating antigen with a concentration of 0.05 mg/mL had a weaker colour intensity than at 0.1 mg/mL, suggesting that more sensitive detection could be achieved at 0.1 mg/mL. Therefore, the optimum conditions for our developed assay consisted of 0.1 mg/mL coating antigen and 8 µg/mL.

Optimization of the strip test
The coating antigen and surfactants were optimized. The strip materials must be hydrophilic and have consistent flow characteristics in the immunochromatographic assay. NC must be hydrophobic. ICA with different coating concentrations (0.05 and 0.1 mg/mL) on the T line were evaluated with 0.01 M PBS (pH 7.4) containing negative (0 ng/mL) or positive (0.5 ng/ mL). Figure 6 shows that T red lines appeared with different colour intensities at 0.05 and 0.1 mg/mL. At 0.1 mg/mL the T red lines had a deeper colour intensity than 0.05 mg/mL, suggesting that a more sensitive detection could be achieved at 0.1 mg/mL coating antigen. It is important to choose the appropriate surfactant when developing these assays because not every protein is compatible with every surfactant. In this study, suspension buffer (0.02 M PBS, 5% sucrose, 2% sorbitol, 1% mannitol, 0.1% PEG, 0.1% tween, and 0.04% NaN 3 ) was mixed with two reagents including polyvinylpyrrolidone (PVP) and PEG. Based on Figure 5, the suspension buffer mix with 5% PEG gave best signal colour on the T line. Therefore, the optimum conditions of our developed assay consisted of 0.1 mg/mL coating antigen and suspension buffer mixed with 5% PEG was carried out in the ICA assay.
Detection of ZEN, α-ZAL, β-ZAL, α-ZOL, β-ZOL, and ZEA in corn flour One of the main effects to be considered in developing these assays is the choice of surfactant reagents because the correct reagent is required for the success of the test. In this study, suspension buffer (0.02 M PBS, 5% sucrose, 2% sorbitol, 1% mannitol, 0.1% PEG, 0.1% tween, and 0.04% NaN 3 ) was added to four kinds of reagents, tween-20, brij-35, triton X-100, and Rhodasurf® On-870 (an ethoxylated oleyl alcohol). As shown in Figure 8, triton X-100 which produced good and stable colour reactions was used for subsequent experiments.

Conclusions
A simple and rapid analytical method for simultaneous determination of ZEN, α-ZAL, β-ZAL, α-ZOL, β-ZOL, and ZEA was developed by the lateral-flow immunochromatography assay. Our results demonstrate that quantitative and semi-quantitative utility of the immunochromatography strip assay, with cut-off limits for the semi-quantitative test strip as low as 0.5 ng/mL in 0.01 M PBS (pH 7.4) and 50 ng/mL in corn flour samples for ZEN, α-ZAL, β-ZAL, meanwhile 1 ng/mL in 0.01 M PBS (pH 7.4) and 75 ng/mL in corn flour samples for α-ZOL, β-ZOL, and ZEA and 50 ng/mL. The results were obtained in 10 min. The lateral-flow immunochromatographic assay represents a sensitive and rapid detection method of ZEN, α-ZAL, β-ZAL, α-ZOL, β-ZOL, and ZEA in corn flour.

Disclosure statement
No potential conflict of interest was reported by the authors.