Introduction

Mycotoxin contamination of animal and human grain and peanut storage facilities poses a serious health hazard and can result in severe economic losses. Mycotoxins are produced by molds or microfungi, are ubiquitous in soil and water, and are easily distributed via air currents. This renders storage facilities with improper humidity controls fertile ground for mold growth. Mycotoxins such as aflatoxin B1 (AFB1) are some of the most highly toxic compounds known for humans and ingestion can cause acute hepatic necrosis, cirrhosis, and liver cancer [1]. AFB1 was classified as a Group 1 carcinogen by the World Health Organization (WHO) in 1993.

In addition to AFB1 there are hundreds of mycotoxins and compounds like zearalenone (ZEN) produced by species of Fusarium and Gibberella. These compounds have strong estrogenic effects resulting in infertility, abortions, and other breeding abnormalities in farm animals. Indeed, a link between ZEN exposure and estrogen receptor-positive breast cancer in humans has recently been shown [2]. As a result of the potential hazards associated with these toxins, many governments require regular monitoring for multiple mycotoxins and have set maximum residue limits for these compounds. The most prominent and most dangerous that are associated with food safety are the aflatoxins, ochratoxin A (OTA), ZEN, the T-2 toxins, and the penultimate precursor to AFB1, sterigmatocystin (ST) [3–8].

These mycotoxins can be absorbed through the skin so that ingestion is not the only mode of toxicity. A potential risk of contamination extends throughout the processing cycle of feeds, i.e., from raw material harvest and storage to processing and transportation. Moreover, feeds can be contaminated with multiple mycotoxins that originate from different raw materials. For example, aflatoxins and OTA have been found together in improperly stored peanut meal, and ZEN and T-2 have been identified in feed stocks simultaneously [9]. Thus, it is necessary to develop a simultaneous determination method for multiple mycotoxins in feeds.

There have been a number of analytical methods developed for monitoring mycotoxin levels in feeds and range from enzyme-linked immunosorbent assays (ELISA) [10] and antibody-based test strips [11] to various chromatographic methods including gas chromatography (GC) and liquid chromatography (HPLC) [12] coupled with mass spectrometry (MS) [13]. ELISA and mycotoxin test strips are the most rapid procedures but are of limited value if quantitative determinations are necessary. Typical GC or GC–MS methods are time-consuming and require sample derivatization. HPLC methods coupled with fluorescence or ultraviolet detection are limited by the separating capacity of the columns; however, the technique is sensitive and easy to use. Therefore, the most accurate and precise measurements of multiple mycotoxins in feeds must rely on chromatography-based methods coupled with MS for unambiguous mycotoxin identification.

There have been several advances on this front for the simultaneous determination of multiple mycotoxins. Ren et al. established an ultra-performance liquid chromatography (UPLC)–MS/MS method that could detect 14 mycotoxins [14]. Jackson et al. improved an LC/MS protocol that provided a practical method for large-scale detection and quantification of multiple mycotoxins in animal feed matrices [15]. However, these methods had disadvantages including low sensitivity and specificity due to matrix effects caused by other sample components. This results in increased background noise from commercial solid-phase extraction columns.

In order to more effectively monitor and assess the co-occurrence of multiple mycotoxins in feeds, we developed alternative cleanup methods that encompassed a wide range of mycotoxins. Sample preparation plays a key role in this process, especially when the toxins are present in complex matrices such as feeds. Ideally, traces of multiple mycotoxins should be extracted in a simple and rapid manner with high recoveries. A 20-min ultrasound-assisted extraction was reported by Kong et al. to have these qualities [16]; however, this method was time-consuming and required a specific device not usually available in most laboratories.

In this work, a rapid sample preparation method was developed using a single multiple immunoaffinity column (mIAC) aided by a group of monoclonal antibodies we produced in the laboratory. The mIAC allowed the specific capture of multiple mycotoxins which could be eluted for downstream processing using UPLC. This procedure was more rapid and performed better than HPLC methods and resulted in a decreased consumption of organic solvents.

The present study focuses on the simultaneous determination method for eight mycotoxins in feeds by using UPLC–MS/MS via a mIAC. The UPLC method allowed rapid separation, and MS/MS provided a high selectivity for the analytes of interest. The overall performance of our method meets the regulatory criteria set by the Food and Agriculture Organization of the United Nations (FAO), European Union (EU), and China and is ready to be field tested for monitoring of multiple mycotoxins in foods and feed.

Materials and methods

All experiments were performed in compliance with relevant laws and institutional guidelines. The experimental animal management committee in Hubei Province approved the experiments.

Regents and materials

Standard solutions of aflatoxins B1, B2, G1, G2 as well as OTA, ZEN, ST, and T-2 mycotoxins were purchased from Sigma-Aldrich (USA). Methanol and acetonitrile were both of HPLC grade. All other chemicals were of reagent grade and were obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). CNBr-activated Crystarose 4B microspheres were purchased from Jingcheng Biological (Wuhan, China). Water (18 MΩ cm) was obtained from a Milli-Q purification system (Millipore, USA).

Preparation of monoclonal antibodies against mycotoxins

Monoclonal antibodies (mAbs) against AFT, OTA, ZEN, ST, and T-2 were produced in our laboratory using standard hybridoma technology as reported previously [17–19]. Briefly, antigens were coupled to bovine serum albumin (BSA) and inoculated subcutaneously injected into Balb/c mice at multiple sites. Hybridomas were screened using ELISA, and mAbs were further purified via the caprylic acid–ammonium sulfate precipitation and a protein G immunoaffinity column (IAC), and then stored at −20 °C before use.

mIAC preparation

The gel-coupling method of mAb against Crystarose 4B was performed by treating the mAbs against AFT, OTA, ZEN, ST, and T-2 with CNBr-activated Crystarose 4B. Crystarose 4B (4 g) was swollen in 50 mL of coupling buffer (0.1 M NaHCO3 0.5 M NaCl, pH 8.3) containing 20 mg of each mAb generated against AFT, OTA, ZEN, ST, and T-2. The coupling reaction was carried out at 4 °C for 12 h while shaking at 200 rpm. This reaction mixture was then filtered through a 2–5-μm membrane, and unbound mAbs were removed using three washes of coupling buffer. The remaining active sites were blocked by gentle shaking with 50 mL of blocking buffer (0.1 M Tris–HCl, pH 8.0) for 2 h at room temperature. The adsorbent was washed three times with 0.1 M acetate buffer (pH 4.0) containing 0.5 M NaCl and with 0.1 M Tris–HCl, 0.5 M NaCl pH 8.0 three times, each time for 5 min. The gel was equilibrated with phosphate buffer saline (PBS, pH 7.4) and stored in PBS solution containing 0.01 % NaN3. The mAb-coupled Crystarose 4B (0.3 mL) was then gravity packed into a solid-phase extraction column and stored at 4 °C until use.

Sample preparation

Eighty feed samples were obtained from local markets in China and homogenized and extracted in a single step using a commercial homogenizer (Jiuyang soybean milk machine, China). Samples of 20 g were added to 100 mL acetonitrile (ACN)/water/acetic acid (80:18:2, v/v/v) and homogenized for 2 min. The resulting supernatant was then filtered and diluted with 1:3 with PBS. Ten milliliters of this solution was then used for the mIAC cleanup step.

mIAC cleanup optimization

Loading conditions were evaluated using ACN/PBS ratios of 5:95, 10:90, 20:80, 30:70, 40:60, 50:50, 60:40, and 70:30 to determine the most effective ACN concentration to load onto mIACs. Columns were loaded with 10 mL mycotoxin standard solution (2 ng mL−1 final concentration of each in solutions using varied acetonitrile concentrations). The columns were washed with 20 mL of PBS, and the mycotoxins were eluted with 1 mL of MeOH. This eluent was then filtered with a 0.22-μm membrane before injection into the UPLC–MS/MS system. A 10-mL aliquot of standard solution containing all eight mycotoxins in 20 % ACN/PBS was loaded onto the mIACs at a final concentration of 1 ng mL−1. The flow rates were varied between 0.5 and 5 mL min−1, the columns were then washed using 20 mL PBS, and the samples were eluted with 1 mL MeOH. Matrix effects on mIAC column washing were studied in the presence of multiple mycotoxins bound to the columns. A series PBS and water wash volumes were performed using 5, 10, 15, 20, 30, 40, and 50 mL. Elution conditions were also varied using 1 mL of MeOH/water in ratios of 100:0, 80:20, 60:40, and 40:60, v/v.

UPLC–MS/MS parameters

Multiple mycotoxins from these column eluates were identified via an ACQUITY H-Class UPLC system coupled to a Xevo TQ-S (Waters, Milford, MA) in selected reaction monitoring (SRM) mode. The UPLC separation was conducted using an ACQUITY UPLC BEH C18 column (2.1 × 50 mm, 1.7 μm particle size) equipped with an ACQUITY UPLC BEH C18 guard column (Waters, Milford, MA) at 40 °C. Mobile phase A consisted of 0.05 % formic acid in water, and mobile phase B contained 0.05 % formic acid in ACN. A linear gradient elution program was developed and used mobile phase A with the following conditions for mobile phase B: 0–3 min, 15–50 % B; 4–5 min, 50–70 % B; 6.5–8 min, 70–100 % B; 8–10 min, 100–50 % B, 10–11 min, 50–15 % B, 11–15 min, 15 % B. The elution flow rate was set at 200 μL min−1, and the injection volume was 10 μL. Spectra were obtained using a triple quadrupole coupled with an electrospray interface (ESI). The MS/MS was operated using ESI sources in the positive or negative mode. The conditions were set as follows: capillary voltage, 3.0 kV for ESI+ and −3.0 kV for ESI−; extractor voltage, 3.0 V; source temperature, 150 °C; desolvation temperature, 350 °C. Argon was used as collision gas (collision cell, 0.8 V), and nitrogen was used for both the nebulizing (50 L h−1) and desolvation gases (650 L h−1). Data was acquired using Mass Lynx 4.0 software.

mIAC method validation

Limits of detection/quantitation and linear range determinations

The limits of detection (LOD) for each mycotoxin were determined as the lowest concentration of the analyte that produced a chromatographic peak with a signal to noise ratio (S/N) of 3 using a matrix of spiked pig, cattle, chicken, and rabbit feed samples. The limit of quantitation (LOQ) was recorded at an S/N of 10 in a similar manner, and each experiment was repeated five times.

The matrix-matched calibration curves of each mycotoxin showed a concentration range of three orders of magnitude. The starting point was the blank sample, while the ending point was set to exceed the maximum residue limits for each mycotoxin. In brief, blank feed samples were spiked at seven different concentrations of AFB1 (0.1, 0.3, 2.5, 10, 25, 50, 75 μg kg−1), AFB2 (0.04, 0.12, 1, 4, 10, 20, 30 μg kg−1), AFG1 (0.1, 0.3, 2.5, 10, 25, 50, 75 μg kg−1), AFG2 (0.04, 0.12, 1, 4, 10, 20, 30 μg kg−1), OTA (0.2, 0.6, 1, 5, 10, 30, 50 μg kg−1), ZEN (0.1, 0.3, 2.5, 10, 25, 50, 75 μg kg−1), ST (0.1, 0.3, 2.5, 10, 25, 50, 75 μg kg−1), and T-2 (0.4, 1.2, 10, 20, 40, 60, 100 μg kg−1), respectively, along with the blank sample. Each calibration curve point was analyzed three times. The linear range was determined by triplicate experiments of spiked samples at these eight different concentrations by fitting a linear model for each mycotoxin with correlation coefficients of R 2 > 0.99.

Recovery

The mIAC recoveries were determined using blank pig, cattle, chicken, and rabbit feed samples spiked with the mycotoxins. Using optimized cleanup conditions, we determined the recoveries from three independent experiments, and the mean recoveries for each mycotoxin were calculated using this data.

Capacity

The mIAC capacity was determined using an excess amount of a mycotoxin that would saturate all accessible sites as described previously [20]. A 10-mL mixed mycotoxin standard solution containing 500 ng mycotoxin extracted from spiked pig feed samples was loaded on the mIAC under optimized conditions. Each experiment was repeated five times.

General assay evaluations

Intra- and inter-assay tests were employed to the evaluate repeatability and reproducibility of the procedure. Each blank pig, cattle, chicken, and rabbit feed sample was spiked with multiple mycotoxins resulting in a final concentration of 10 μg kg−1 of each. These samples were extracted and stored until use. Assay repeatability was determined using a single mIAC and performing six runs. Reproducibility was established by running a single test extraction on six different mIACs in triplicate. Column stability was tested using one mIAC batch that was run multiple times over a 1-year period.

Real sample assays

A total of 20 pig feed, 20 cattle feed, 20 chicken feed, and 20 rabbit feed samples were collected from local markets (Wuhan, China). These agricultural product samples were assayed using optimized conditions for sample preparation, purification, and UPLC–MS/MS analysis.

Results and discussion

UPLC–MS/MS method

As a result of the different polarities of this collection of mycotoxins, a series of preliminary experiments were performed to develop an acceptable UPLC separation method. We tested several elution mixtures including ACN, MeOH, or mixtures of these with water containing ammonium acetate, acetic acid, or formic acid at different concentrations. UPLC parameters including the flow rates and elution gradients were also investigated (data not shown).

For our eight test mycotoxins, optimal separation conditions were achieved using water containing 0.05 % formic acid in water (eluent A) and 0.05 % formic acid in ACN (eluent B). A representative chromatogram of a mixture of eight mycotoxins extracted from a peanut sample is shown in Fig. 1. The MS conditions were also studied to improve sensitivity. The MS parameters were optimized using flow injection for each compound separately. The ESI+ and ESI− modes were alternatively applied, and the optimized MS/MS parameters are listed in Table 1.

Fig. 1
figure 1

Chromatograms of 8 mycotoxins in peanut samples. AFB1, 10 ng mL−1; AFB2, 5 ng mL−1; AFG1, 10 ng mL−1; AFG2, 5 ng mL−1; OTA, 5 ng mL−1; ZEN, 5 ng mL−1; T-2, 5 ng mL−1; ST, 5 ng mL−1

Full size image
Table 1 Parent ion/product ion pairs of mycotoxins in UPLC–ESI–MS/MS mode
Full size table

Sample preparation

Acetonitrile (ACN) and mixtures of ACN/water/acetic acid (60:38:2, 70:28:2, 80:18:2, 90:8:2, and 98:2:0, v/v/v) were evaluated for extraction of multiple mycotoxins. Our results indicated that the 80:18:2 mixture gave the best recovery for each of the mycotoxins that had been spiked into pig feed samples (Table 2). For cattle, chicken, and rabbit feed samples, similar results were found. Sample grinding coupled with extraction reduced the processing time to 2 min per sample, which suggested its feasibility as part of a rapid determination method. The samples were then diluted threefold to decrease the acetonitrile concentration before proceeding to the mIAC cleanup step.

Table 2 Recoveries using a graduated series of extraction solutions
Full size table

mIAC cleanup optimization

The mycotoxins used in our samples have both hydrophilic and hydrophobic character which may be problematic when using a single cleanup procedure. Previous studies have successfully used dispersive solid-phase extraction using C18 silica/MgSO4 as sorbent [21]. However, matrix interferences lowered the sensitivity of MS determination and did not fulfill the requirements for multiple mycotoxin monitoring. Thus, it is unlikely that this method would be employed in a developing country where high levels of contamination with multiple mycotoxins are a risk due to improper storage conditions. This mIAC method could reduce matrix effects and without sacrificing economy. Herein, the loading, washing, and elution conditions were optimized using optimized parameters.

Loading conditions

Mycotoxins can dissolve in moderately polar solvents such as acetonitrile and MeOH. However, the high solvent concentrations are deleterious to mAb stability, which would interfere with antibody–antigen interactions. Therefore, different acetonitrile concentrations were tested to minimize these effects.

Recoveries of mIAC were studied by using spiked feed samples in triplicate. The recoveries from the mIACs are shown in Fig. 2a. Thus, 20 % ACN in PBS showed the greatest recovery (101.3 %); as the ACN content was increased to 70 %, the recoveries decreased to 43.4 % (Fig. 2a). Therefore, the loading buffer was optimized as a 20 % ACN solution in PBS.

Fig. 2
figure 2

Effects of different a loading solvents and b flow rates on IAC efficiency

Full size image

The flow rate of sample loading influenced the mIAC recoveries as well. Sample recoveries were reduced with flow rates greater than 3 mL min−1 while slower rates increased the elution time (Fig. 2b). A flow rate of 2 mL min−1 was chosen as the optimal flow rate for sample loading.

Column washing conditions

Sample matrixes can increase background noise in a chromatogram and thus decrease instrument sensitivity. We wanted to elute interfering substances from our extractions so we examined the column washing conditions. We used a series of PBS and water washes using different volumes. We determined that the PBS washes gave lower RSD values than water, which suggested that PBS probably acts as a stabilizer of the mycotoxin–mAb interaction. Washing volumes below 20 mL resulted in an RSD increase of up to 14.4 % while those washes greater than 20 mL showed RSD values of 3.2–7.1 % (Fig. 2). Therefore, 20 mL PBS was used as washing volume.

Sample elution conditions

We initially used MeOH to disrupt the antigen–antibody interactions and elute the mycotoxins from the mIACs. Bound mycotoxins were tested for elution with 1 mL of MeOH/water in ratios of 100:0, 80:20, 60:40, and 40:60, v/v, as shown in Fig. 3a. When we used a 40:60 mixture, the analytes were not quantitatively eluted. This could be due to a decreased solvent access to nonpolar interactions that are necessary for the release of the mycotoxins from the mAb. On the other hand, a recovery around 100 % was found using no less than 1 mL MeOH. Interestingly, it should be noted that the mIAC can be reused three times and still produce recoveries between 70 % and 110 %. This suggested that MeOH did not disrupt the mAb specificity. Therefore, 1 mL of MeOH was employed as the elution solvent.

Fig. 3
figure 3

a mIAC average recoveries vs elution conditions (MeOH/water; v/v): turquoise, 100:0; blue, 80:20; green, 60:40; red, 40:60). b mIAC stability

Full size image

mIAC method validation

LOD, LOQ, and linear range

A twofold dilution series was employed to investigate LODs and LOQs. On the basis of the chromatographic peaks with an S/N of 3 and 10 for LOD and LOQ, respectively, LODs were 0.02, 0.02, 0.04, 0.03, 0.12, 0.25, 0.006, and 0.12 ng mL−1 and LOQs were 0.06, 0.06, 0.12, 0.09, 0.36, 0.75, 0.018, and 0.36 ng mL−1 for AFB1, AFB2, AFG1, AFG2, OTA, ZEN, ST, and T-2, respectively. Compared with previous reports [22–24], our method provided a lower LOD/LOQ ratio by a factor of 5–20 [25]. The mIAC method therefore meets the requirements set by the EU (EC1881/2006 and 1126/2007) and China (GB/T 5009.23-2006, SN/T 3136-2012).

Matrix-matched calibration curves and linear range

The linear range was then estimated for all the mycotoxins from feed matrices. There were no significant differences using different feed matrices with the IAC cleanup. The linear range was obtained by triplicate analysis of spiked pig, cattle, chicken, and rabbit feed samples using an eight-point calibration curve with correlation coefficients of R 2 > 0.99. For AFB1, AFB2, AFG1, AFG2, OTA, ZEN, ST, and T-2, the linear ranges were found to be 0.30–25, 0.12–20, 0.30–20, 0.12–20, 0.60–30, 0.30–25, 0.30–25, and 1.2–40 μg kg−1, respectively. These linear ranges were larger than those of previous studies on multiple mycotoxins in body fluids (urine and plasma) [22].

Recovery

Under optimal conditions, mIAC recoveries were measured by spiking pig, cattle, chicken, and rabbit feed samples. The mean recoveries ranged from 91.2 % to 104.1 % for pig feed samples, from 93.7 % to 98.7 % for cattle feed samples, from 92.2 % to 102.1 % for chicken feed samples, and from 96.4 % to 101.7 % for rabbit feed samples. This fulfilled the requirements for the simultaneous determination of multiple mycotoxins from a complex mixture.

mIAC capacity

The maximum column binding capacity was determined as outlined by Qiao et al. [20]. A 10-mL volume of a multiple mycotoxin standard solution (containing AFB1, AFB2, AFG1, AFG2, OTA, ZEN, ST, and T-2 at 50 ng mL−1 each) was cleaned up using the mIACs. Using the optimized conditions as described above and a value of 8 mg of mAb, we determined that the binding capacity was approximately 1117.2 ng for a multiple mycotoxin mixture (Table 3).

Table 3 mIAC capacity for each mycotoxin
Full size table

Repeatability, reproducibility, and stability

An intra-assay test was used to evaluate the repeatability of our recoveries. A spiked pig feed sample was detected six times by using the mIACs in the same batch. The recoveries of AFB1, AFB2, AFG1, AFG2, OTA, ZEN, ST, and T-2 were found to be 93.9–102.0 %, 94.3–102.0 %, 94.7–100.5 %, 95.2–101.5 %, 95.6–99.7 %, 90.9–96.5 %, 92.9–97.7 %, and 93.1–98.7 %, respectively, with RSDs of 3.4 %, 2.9 %, 2.3 %, 3.1 %, 2.3 %, 2.1 %, 1.8 %, and 2.2 %, respectively. This indicated that the mIAC method had good repeatability.

In order to estimate reproducibility, six batches of mIACs were employed in an inter-assay recovery procedure. The inter-assay experiment was conducted in a similar manner as the intra-assay one. The recoveries of AFB1, AFB2, AFG1, AFG2, OTA, ZEN, ST, and T-2 were found to be 92.2–102.2 %, 93.5–101.4 %, 89.7–102.5 %, 91.8–100.1 %, 91.5–97.5 %, 92.7–98.7 %, 92.3–100.3 %, and 90.6–101.1 %, respectively, with RSDs of 4.2 %, 5.1 %, 5.6 %, 3.8 %, 7.7 %, 6.1 %, 4.1 %, and 5.2 %, respectively, indicating excellent reproducibly of these mIACs. The RSDs below 10 % are regarded as qualified values.

In order to evaluate its stability, the mIACs were stored at both 4 °C and room temperature for 360 days and at set intervals were used to test recoveries. Typically, 10 mL of a multiple mycotoxin standard solution (containing AFB1, AFB2, AFG1, AFG2, OTA, ZEN, and T-2 at a concentration of 1 ng mL−1 each) was cleaned up using the mIACs. Recovery of mIACs stored at room temperature decreased dramatically, which suggested that these IACs had expired. On the other hand, the recovery remained over 80 % up to 360 days when stored at 4 °C (Fig. 3b). These results indicated that the columns had a 30-day shelf life when stored at room temperature and a 360-day shelf life when stored at 4 °C.

Comparison with commercial a multiple IAC

Two commercially available kits (commercial 1 and commercial 2) and two previously published IACs were tested and compared with our mIAC. Our column could screen eight mycotoxins simultaneously with the exception of ST mycotoxin, which is regarded as an early warning sign of aflatoxin contamination. A comparison of recoveries with two commercially IACs indicated that there was excellent agreement among the three IACs with RSD values below 10 %. (Table 4).

Table 4 Comparison of 5 multiple IAC for mycotoxin determination
Full size table

Application of IACs to real samples

To further apply the mIAC to the determination of multiple mycotoxins in feed samples, 80 feed samples were tested for the presence of our eight test mycotoxins. Interestingly, all eight mycotoxins were found in the feed samples (Table 5). About 38.8 % (31/80) of the feed samples were contaminated with multiple mycotoxins, while only 8.8 % (7/80) of those were contaminated only with one mycotoxin. Overall, 52.5 % (42/80) of the feed samples were not contaminated with any of these mycotoxins.

Table 5 Determination for 8 mycotoxins via mIAC for pig feed, cattle feed, chicken feed, and rabbit feed
Full size table

These results suggested that our method could be employed in the risk monitoring of feed. In these contaminated feed samples, AFB1 had the highest frequency of occurrence (32.5 %) with concentrations ranging from not detected (ND) to 32.12 μg kg−1. OTA was found in 23.7 % of the feed samples with concentration from ND to 15.64 μg kg −1 and was usually present when the feed sample contained wheat and corn. Cotton and corn matrix-based feed samples were found to be contaminated with ZEN, which accounted for 28.8 % of the total feed with concentrations from ND to 61.59 μg kg −1 (Table 5).

The co-occurrence of three mycotoxins in a single sample was less frequent than two. There were six feed samples that were contaminated with combinations such as AFB1, OTA, and ZEN or AFB1, OTA, ST. A few feed samples exceeded the maximum residue limit set by China including for AFB1 and ZEN. Considering the diverse origins of these feed samples, multiple mycotoxin contamination was not predictable and the monitoring of multiple mycotoxins using the mIAC would fit into this type of testing program.

Conclusion

As are result of multiple feed sources, it is likely that more than one mycotoxin will be found in a particular feed batch. Monitoring of this situation requires a highly sensitive determination method to assess the relative contamination risk. We developed an mIAC preloaded with mAbs directed against eight distinct mycotoxins and can simultaneously monitor AFB1, AFB2, AFG1, AFG2, OTA, ZEN, ST, and T-2 from a single sample. Sample preparation was simplified and utilized an ACN/water/acetic acid solution in a one-step grinding–extraction step. The extract is then purified using the mIAC followed by UPLC–MS/MS for quantification. The mIAC cleanup process was also optimized, and the sensitivity and recoveries of our columns were lower than those of commercialized IACs. The mIACs demonstrated considerable repeatability, reproducibility, and stability. Compared with the commercialized IACs for multiple mycotoxins, this method covered more mycotoxins than usually tested in China, and the sensitivities were dramatically enhanced. This mIAC is ready to be used in the field to monitor the presence of multiple mycotoxins in complex mixtures.