Introduction

Tenuazonic acid (3-acetyl-5-s-butyl-4-hydroxy-3-pyrrolin-2-one, TeA) is an Alternaria toxin produced by Pyricularia, Phoma, and Alternaria alternate. Among Alternaria toxins, TeA was listed as the most toxic one by US Food and Drug Administration (Mikula et al. 2013; Ostry 2008). Improper storage or transportation was prone to induce food spoilage by TeA-producing fungi, therefore leading to contamination of TeA (Bhat et al. 2010; Fernández-Cruz et al. 2010; Yang et al. 2013). It has been reported that TeA was the predominant Alternaria toxin in food samples. For example, up to 77% cereal products were contaminated by TeA (Bruce et al. 1984). A survey about wheat flour showed 99.4% (180/181) samples contained TeA at levels ranging from 1.76 to 520 μg/kg (mean = 79.80 μg/kg) and 31.5% (67 samples) were contaminated with TeA at a concentration higher than 100 μg/kg with a maximum of 520 μg/kg (Zhao et al. 2015b). In tomato products (Stack et al. 1985), TeA level could vary from 0.4 to 70 mg/kg. Besides, TeA could be frequently detected in beers, potatoes, pepper, wines, and even animal products (Fontana et al. 2016; Li and Yoshizawa 2000; Lohrey et al. 2013; López et al. 2016; Siegel et al. 2010; Walravens et al. 2016; Zhao et al. 2015a).

However, until now, most countries have not issued a tolerance limit for TeA as well as Alternaria toxin (Janardhanan and Husain 1984; López et al. 2016; Rychlik et al. 2016; Zhou et al. 2019). Although the exposure risk of TeA existed in dietary is still unclear, its potential hazard should be further evaluated based on continuous and wider monitoring. Thus, a rapid, effective, and costly detection method for TeA would be necessary and deserves to be developed.

Compared with the analytical methods such as high-performance liquid chromatography (HPLC) (Fan et al. 2016; Myresiotis et al. 2015), liquid chromatography coupled with tandem mass spectrometry (LC-MS) (Fraeyman et al. 2015; Rasmussen et al. 2010; Xu et al. 2016), enzyme-linked immunosorbent assay (ELISA) has the advantages of ease, rapid, and accuracy as well as low cost. Previously, the competitive ELISAs based on polyclonal antibody (pAb) for the TeA derivatives, namely tenuazonic acid hemisuccinate and 5-(sec-butyl)-3-(1-hydrazonoethyl)-4-hydrooxy-2(5H)-one (TeAH), was established to detect TeA through derivatization pretreatment of the samples (Gross et al. 2011; Yang et al. 2012). However, in these assays, derivatization is complex and derivative rate for different samples might be different therefore influencing the assay accuracy. Moreover, pAbs from antiserum generally have unstable quality caused by the batch difference of immunized animals. In this study, we prepared a specific monoclonal antibody with TeA as direct target via the hybridoma technique for the first time. Furthermore, an mAb-based ic-ELISA subsequently was developed for the detection of TeA, and its accuracy was validated with spiked food samples and confirmed by HPLC analysis.

Experimental Methods

Materials

The antigen tenuazonic acid (TeA), derivatized antigen tenuazonic acid coupled with carboxymethoxylamine hemihydrochloride (TeA-CMO), and the SP2/0 myeloma cell line were stored in our laboratory. Complete Freund’s adjuvants, incomplete Freund’s adjuvants, N, N-dimethylformamide (DMF), dicyclohexylcarbodiimide (DCC), N-hydroxysuccinimide (NHS), polyethylene glycol (PEG) 4000, and horseradish peroxidase-labeled goat anti-mouse IgG (IgG-HRP) were purchased from Sigma company (USA). Hypoxanthine-aminopterin-thymidine (HAT) and hypoxanthine-thymidine (HT) were obtained from Gibco company (USA). Culture media RPMI-1640 was purchased from HyClone company (USA). All other organic solvents and chemicals used were of analytical grade. Female Balb/c mice were purchased from Guangdong Medical Laboratory Animal Center.

Immunization and Hybridoma selection

Hapten-protein conjugates TeA-CMO-BSA and TeA-CMO-KLH were prepared by active ester methods according to the references (Lin et al. 2019; Ni et al. 2019). Seven-week-old female Balb/c mice were firstly immunized by subcutaneous injection of 50 μg TeA-CMO-KLH emulsified with the same volume of complete Freund’s adjuvant. Booster injections were given at 2-week intervals with the immunogen and an equal volume of incomplete Freund’s adjuvant. The mice tail blood was collected at the interval of 7 days after each immunization, and detected by ic-ELISA using TeA-CMO-BSA as the coating antigen. The mouse with the highest titer and inhibition rate received a final soluble intraperitoneal injection of TeA-CMO-KLH in PBS, 3 days prior to cell fusion. Cell fusion procedures were carried out essentially as described by the methods (Xiao et al. 2018).

Preparation and Identification of mAb

The positive hybridomas were cultivated and injected into abdominal cavity of female Balb/c mice after paraffin injections 1 week in advance. The collected ascites was purified by the caprylic acid-ammonium sulfate precipitation method (Abad et al. 1999; Zhao et al. 2002). The purity of mAb was identified by 12% SDS-PAGE and the concentration was measured using a NanoDrop 2000C system (thermo scientific, USA). The isotype of mAb was identified by a mouse mAb isotyping kit (sigma, USA).

The affinity of the mAb were determined by surface plasmon resonance (SPR, BIAcore T200, USA) with a continuous flow of 10 μL/min of PBS buffer at 25 °C. The coating antigen TeA-CMO-BSA as a ligand in pH 4.0 sodium acetate was immobilized to a CM5 sensor chip by amine coupling method. Then, different concentrations of anti-TeA mAb were diluted and injected. Glycine-HCl (10 mM, pH 2.0) was used as a dissociation reagent. All data was analyzed by Biacore T200 evaluation software.

Establishment of an Indirect Competitive ELISA

The 96-well microplates were coated with 100 μL/well of TeA-CMO-BSA overnight at 37 °C, then washed two times by PBST (0.1% Tween 20 in PBS) and blocked with 120 μL/well of blocking solution (2% skimmed milk in PBS) for 3 h at 37 °C. After washing, 50 μL of different concentrations of standard TeA solution along with 50 μL of antibodies were added each well and incubated at 37 °C for 40 min. The plate was washed five times by PBST and added with 5000-fold diluted IgG-HRP for 40 min at 37 °C. The color was developed with the chromogenic reagent (100 μg/mL 3,3′,5,5′-tetramethylbenzidine and 0.02 μL/mL H2O2 in 100 mM NaOAc, pH 6.0), and the reaction was stopped with 50 μL per well of 10% H2SO4. The absorbance was read at 450 nm using a microplate reader (thermo scientific, USA).

The concentration of antibody and coating antigen were optimized via checkerboard titration (Liang et al. 2007; Xiao et al. 2018). To further improve the sensitivity and repeatability of ic-ELISA, other assay conditions were also optimized, including incubation times of antigen-antibody and secondary antibodies, as well as concentration of PBS buffer.

Under the optimized conditions, the standard curve was fitted to a four-parameter logistic equation with Origin 9.0 software. The percentage of cross-reactivity (CR) (Wang et al. 2018; Zhang et al. 2018) was calculated as follows: CR (%) = IC50 (TeA, ng/mL)/IC50 (TeA analogues, ng/mL) × 100.

Analysis of Spiked Samples

Wheat beer, apple juice, and grape juice purchased from a local supermarket were verified as TeA-free by HPLC analysis. The negative samples were spiked with different concentrations of TeA and filtered by a 0.22-μm membrane respectively for analysis.

The accuracy of the established ic-ELISA was validated with HPLC analysis. The ODS-C18 was used as reverse phase column under 40 °C temperature. Methanol-water (v/v, 85%: 15%) containing zinc sulfate (300 mg/L) was as mobile phase at a continuous flow rate of 0.4 mL/min and the injection volume was 10 μL. Analytes were determined by UV at 280 nm wavelength.

Results and Discussion

Preparation and Identification of mAb

The mouse with high serum titer was chosen for cell fusing. After four rounds of limiting dilution, a single-cell clone 3F10-producing specific antibody was selected. Furthermore, the purified antibodies with the concentration of 6.3 mg/mL were prepared from the ascites fluid and identified by SDS-PAGE (Fig. 1a). The isotype of purified mAb was identified to be immunoglobulin G1 type (Fig. 1b), consistent with the monoclonal antibody isotype for the most of small molecule antigen reported (Liu et al. 2016). In addition, to assess the antibody quality, the affinity of antibody-analyte was determined by SPR. As shown in Fig. 1c, with the antibody concentration increasing, response unit (RU) values became higher and appeared an obvious dose-dependent relationship. The affinity constant of the mAb was 3.045E−3 M (Fig. 1d), which was among the range about binding of proteins to small molecules.

Fig. 1
figure 1

Characterization of purified anti-TeA mAb. a SDS-PAGE analysis, nonreducing gel: 1, mAb after purification, 2, mAb before purification; reducing gel: 3, mAb before purification, 4, mAb after purification; b determination of monoclonal antibody isotypes; c kinetics tests for analysis of anti-TeA mAb and analyte; and d affinity test

Full size image

Establishment of ic-ELISA Based on the mAb

To establish an ic-ELISA based on mAb for the detection of TeA, different parameters were optimized firstly. The ratio of Amax/IC50 was employed to evaluate the influence of each condition, and a higher ratio indicated a higher sensitivity of the assay (Abad and Montoya 1997; Chen et al. 2014). As shown in Fig. 2, the optimized reaction conditions were 250 ng/mL of coating antigen, a 1:32000 dilutions fold antibody (125 ng/mL) in PBS buffer (pH = 7.4, 0.01 M), a 40-min competition time of antibody-antigen and 40 min IgG-HRP incubation time, respectively. Under these conditions, an ic-ELISA standard curve was finally established with the IC50 value of 18.50 ng/mL, the limit of detection (LOD) of 1.00 ng/mL, and the liner concentration range of 3.56~96.24 ng/mL (Fig. 3). The IC50 value was nearly 20 times lower than the reported pAb-based ELISA (Gross et al. 2011). Meanwhile, unlike derivative-ELISA established by previous study, in which at least 2 h is required for derivatization pretreatment of the samples (Yang et al. 2012), the whole assay in this project can be finished within 1.5 h.

Fig. 2
figure 2

Optimization of assay conditions: a coating antigen concentration and antibody dilution, b competitive time of antibody and antigen, c incubation time of IgG-HRP, and d concentration of PBS buffer

Full size image
Fig. 3
figure 3

Standard curve for the detection of TeA by ic-ELISA (n = 4). The vertical bars indicate the mean results of the standard deviation

Full size image

The CR of the assay was evaluated by IC50 values of seven kinds of analogues. The results indicated that mAb was specific and sensitive to TeA (Table 1) and showed less than 2% CR to all analogues including the iso-tenuazonic acid (ITeA), an isomer of TeA, which only appears methyl isomerization in spatial structure. Likewise, in our previous work (Xiao et al. 2018), an anti-ITeA mAb was prepared and showed low CR for TeA. This further confirmed the potential advantages of immunoassay based on the specific antibody to different target analysis.

Table 1 Cross-reactivity (CR) of mAb with TeA and other compounds using ic-ELISA
Full size table

Detection of TeA in Spiked Samples by ic-ELISA and HPLC

Wheat beer, apple juice, and grape juice spiked with TeA at three concentrations (0, 0.8, 2.0, and 3.2 mg/L) were detected by the developed ic-ELISA. Generally, dilution is a convenient and simple way to eliminate the matrix effect (Sheng et al. 2012), especially for liquid samples. In this study, when wheat beer, apple juice, and grape juice samples were diluted by PBS buffer to 40 times respectively, the matrix interference was effectively eliminated (Fig. 4). In addition, the ic-ELISA gives an acceptable average recovery ranged from 85.0 to 120.0% (Table 2), and the coefficients of variation (CV) were all below 15%. Furthermore, the samples spiked with TeA were also detected by HPLC to evaluate the accuracy of ic-ELISA. As shown in Fig. 5, the squared correlation coefficient (R2) of ic-ELISA and HPLC was 0.9732. This means that the established ic-ELISA could be used to monitor TeA residues in food samples with acceptable accuracy and reproducibility.

Fig. 4
figure 4

Sample preparation and elimination of matrix effect (n = 3). a Apple juice, b grape juice, and c beer

Full size image
Table 2 Recoveries of TeA in apple juice, grape juice, and beer by mAb-based ic-ELISA (n = 3)
Full size table
Fig. 5
figure 5

Correlations of analysis for spiked samples between ic-ELISA and HPLC (n = 3)

Full size image

Conclusions

In this study, a sensitive and specific ic-ELISA based on monoclonal antibody was developed for the determination of TeA in food samples. A novel monoclonal antibody against TeA was prepared for the first time in this study, and the SPR results showed its high affinity. After optimization of four parameters, a mAb-based ic-ELISA was established with high sensitivity and specificity. Furthermore, ideal recoveries in spiked food samples were obtained and its accuracy was confirmed via HPLC method. Overall, this work should broaden the application of immunological methods involved in hazardous mycotoxins. Future studies will be focused on the optimization of screening hybridoma process for anti-TeA mAb with higher affinity and the development of a rapid immunoassay kit.