Quantitation of mycotoxins in plant-based meats 


Hitha P P1, Sashank Pillai1, Craig M. Butt2

1SCIEX, India; 2SCIEX, USA2

Abstract


In this technical note, a sample preparation and LC-MS/MS method is demonstrated for the quantitation of 10 mycotoxins in 3 plant-based meat products (Figure 1). Using the SCIEX 7500 system, matrix spikes at 1 and 10 ng/g showed good absolute recovery (range: 65-128%) and precision (range: 1.0-15.6%) in the 3 matrices (2 plant-based chicken and 1 plant-based mutton products). Trace levels of mycotoxins were detected in the unspiked samples but significant matrix effects were observed for some mycotoxins, including zearalenone, alternariol and alternariol monomethyl ether. Therefore, standard addition experiments were performed to improve the quantitation accuracy and the automated standard addition calculation feature in the SCIEX OS software was used to streamline data processing. Limits of quantation (LOQs) in the solvent-based calibration standards ranged from 0.01 to 0.025 ng/mL, demonstrating low parts-per-trillion sensitivity for mycotoxin analysis using the SCIEX 7500 system.

Figure 1. Schematic diagram showing the quantitation workflow for mycotoxin analysis in plant-based meat using the SCIEX 7500 system. Extracted ion chromatograms (XICs) at the bottom demonstrate the matrix blank and 1 ng/g spike of Ochratoxin A in a plant-based chicken sample.

Key benefits of the analysis of mycotoxins using the SCIEX 7500 system
 

  • Low ng/g sensitivity in plant-based meat samples. All 10 mycotoxins were detected in the 1 ng/g and 10 ng/g matrix spikes into the 3 plant-based matrices, with good absolute recovery (range: 65-128%) and precision (range: 1.0-15.6%) 

  • Parts-per-trillion mycotoxin sensitivity in solvent-based standards. In-vial LOQs ranged from 0.01 ng/mL to 0.025 ng/mL, mean LOQ accuracy ranged from 81% to 104% and mean LOQ precision ranged from 2.0% to 12.3% 

  • Automated standard addition calculations in SCIEX OS. Standard addition feature within SCIEX OS rapidly performed calculations for improved quantitation accuracy in complex food matrices

Introduction


Mycotoxins are low molecular weight compounds that are naturally produced as secondary metabolites by fungi (moulds)1. Contamination of food and feed by mycotoxins is a significant problem worldwide and can cause disease and potentially death in humans and animals2,3. Many countries have regulatory guidelines, such as maximum tolerated levels, for mycotoxins in food, feed and dairy products. In July 2024, the European Union established new or lowered maximum levels for deoxynivalenol (DON) and T-2/HT-2 in some foods.

Plant-based foods are an important alternative of animal-based foods and are widely consumed by people all over the world. Plant-based foods are frequently made with raw ingredients, such as soybeans, chickpeas and grains, which are commonly the main foodstuff that are contaminated with mycotoxins. Further, due to their processed and complex nature, plant-based foods represent a difficult analytical matrix for residue testing. This technical note demonstrates a sample preparation and LC-MS/MS analysis method for the quantitation of 10 mycotoxins in three commercially available plant-based meat products, with good analytical data quality. 

Methods


Standard preparation: The mycotoxin standards were purchased from Evolution Life Sciences (original source: Trilogy Analytical Laboratory, USA). The intermediate spiking solutions were prepared in 50:50 (v/v) acetonitrile/water. The calibration standards were prepared in 50:50 (v/v) acetonitrile/water with 0.05% (v/v) formic acid at concentrations ranging from 0.01 to 100 ng/mL.

Matrix spike recovery experiments: The three plant-based meat samples were spiked before (“pre-spike”) and after (“post-spike) extraction at 1 ng/g and 10 ng/g (n=6 replicates per spiking level). The post-spike samples were also prepared at 0.1 and 1 ng/mL, equivalent to the 1 and 10 ng/g pre-spike samples. The absolute recovery was calculated as the ratio of the pre- to post-extraction area counts. 

Sample preparation: A 2 g sample of the plant-based product was placed in a 50 mL polypropylene tube, 10 mL of 80:20 acetonitrile/water (v/v) with 0.1% formic acid was added and the mixture vortexed for 1 hr, sonicated for an additional 15 min and centrifuged for 10 min at 4500 rpm. 2 mL of the supernatant was transferred to a new tube containing 300 mg MgSO4, 200 mg NaCl, and 100 mg C18. The mixture was vortexed thoroughly and centrifuged at 4500 rpm for 5 minutes. The supernatant was diluted with water at a 1:1 (v/v) ratio, filtered, collected into an amber vial for instrumental analysis.

Chromatography: The chromatographic separation was achieved using the Phenomenex Kinetex C18 column (100 x 2.1 mm, 2.6 μm, Part No. 00D-4462-AN). Mobile phase A was 0.1% (v/v) formic acid in water and B was acetonitrile. The total run time was 13 min, gradient conditions are presented in Table 1. The flow rate was 0.50 mL/min, the injection volume was 5 μL and the column oven was set to 40°C.

Table1. LC gradient conditions for the analysis of mycotoxins using the SCIEX 7500 system.

Mass spectrometry: Samples were analyzed using the SCIEX QTRAP 7500 system with the OptiflowPro ion source. Data was acquired using multiple reaction monitoring (MRM) mode with polarity switching. Two transitions were monitored per analyte4. The compound and source parameters were optimized and summarized in Tables 2 and 3.

Data processing: Samples were acquired and processed using the SCIEX OS software (version 3.3.1).

Table 2: Optimized source and gas parameters for the analysis of mycotoxins using the SCIEX7500 system.

Table 3. Optimized MRM conditions for the analysis of mycotoxins in plant-based meats using the SCIEX 7500 system. Quantifier transitions are designated by “1” and qualifier transitions are designated by “2” 

Sensitivity and analytical performance of the solvent-based calibration standards


The solvent-based calibration standards were injected in triplicate to evaluate the sensitivity, accuracy, precision and linear dynamic range of the 10 mycotoxins on the SCIEX 7500 system. The in-vial LOQ ranged from 0.01 ng/mL to 0.025 ng/mL (Table 4). Figure 2 shows the extracted ion chromatograms (XICs) for fumonisin B1 and aflatoxin G2 at their LOQ concentrations of 0.01 ng/mL and 0.025 ng/mL, respectively. The mean LOQ accuracy ranged from 81% to 104% and mean LOQ precision ranged from 2.0% to 12.3%. The linear dynamic range was 3-4 orders of magnitude, dependent on the individual mycotoxin, and r2 values were >0.994 for all compounds. Overall, these results demonstrate the capability of the SCIEX 7500 system to analyze mycotoxins at low parts-per-trillion levels with good accuracy and precision. 

Figure 2. XICs of fumonisin B1 and aflatoxin G2 at the LOQ level of 0.01 ng/mL and 0.025 ng/mL, respectively. Both the quantifier (blue) and qualifier (pink) traces are shown with the lines representing the user-defined ion ratio tolerance thresholds.

Table 4. Sensitivity, accuracy and precision at the LOQ level, linear dynamic range and correlation coefficient for the solvent-based calibration standards (n=3)

Method recovery in plant-based meat matrix spikes


Method performance was evaluated through matrix spikes in the 3 plant-based meat products at 1 ng/g and 10 ng/g (n=6 per spiking level). Figure 3 shows XICs for the matrix blank and 1 ng/g spike in the plant-based chicken I sample. Initial quantitation of the unspiked (blank) samples was performed against the solvent-based calibration standards. While trace levels were detected for some mycotoxins (ochratoxin A, tentoxin and alternariol monomethyl ether), quantitation was performed using standard addition due to significant matrix effects (see later section). The absolute recovery was calculated as the ratio of the pre- to post-extraction area counts and ranged from 78% to 128% in the three matrices. The only exception was fumonisin B1 in the plant-based mutton matrix, which showed mean absolute recoveries of 67% and 65% for the 1 ng/g and 10 ng/g spikes, respectively. The mean absolute recovery precision in the matrix spikes ranged from 1.7% to 15.6% at both spiking levels. Additionally, the observed absolute recovery and precision were similar in all the three matrices.

Figure 3: Representative XIC chromatogram for aflatoxin B1, B2,G1, and G2 (left to right, top panel) in the unspiked plant-based chicken I matrix and 1 ng/g pre-spike sample (bottom panel).

Table 5. Mean absolute recovery (%CV) for the 10 mycotoxins spiked into 3 commercially-available plant-based meats at 1 ng/g and 10 ng/g levels (n=6).

Matrix effects for mycotoxins in plant-based meats


Plant-based meat products are complex processed foods and prone to potential matrix effects which may impact quantitative accuracy. The matrix effects were evaluated in the 1 ng/g matrix spikes using the equation,

Matrix effects were minimal for 6 of the 10 mycotoxins with mean matrix effects ranging from -7% to -18%, indicating slight suppression. However, significant matrix suppression was observed for 3 mycotoxins in the 3 plant-based meats, including zearalenone (mean matrix effect = -75%), alternariol (-64%), and alternariol monomethyl ether (-81%). Ochratoxin A showed moderate matrix suppression with a mean matrix effect of -38%.

Automated standard addition calculations in SCIEX OS

Several mycotoxins were detected at low levels in the unspiked plant-based meat products, including ochratoxin A, fumonisin B1, tentoxin, and alternariol monomethyl ether. Overall, this finding highlights the sensitivity of the SCIEX 7500 system to detect mycotoxins in plant-based meats. But, as noted above, some mycotoxins showed significant matrix suppression in the plant-based meat extracts. Matrix suppression is known to impact quantitation accuracy when using solvent-based calibration standards. Stable-isotope standards are commonly used to correct for matrix effects in routine testing. However, these were not used due to commercial availability and high cost. Matrix-matched calibration standards are also used to account for significant matrix effects but are limiting when analyzing diverse food matrices. 

In this project, standard addition was used to accurately quantify the background mycotoxin levels in the plant-based meats due to potential bias from the matrix effects.. The matrix blank samples were spiked post-extraction (n=3) at 0.1, 0.5 and 1.0 ng/mL, which was equivalent to in-sample concentrations of 0.8, 4 and 8 ng/g. A unique feature of SCIEX OS is that it contains an algorithm to help the user perform the standard addition calculations, significantly simplifying and expediting the process. Figure 4 shows the example of Ochratoxin A in plant-based chicken I. These calculations were accessed through selecting the “Quantitate by standard addition” option in the processing method, ultimately generating the “Standard Addition Calculated Concentration” in the Results Table. Since the standard addition calculation output represented the in-vial concentration, the Calculated Columns feature in SCIEX OS was used to back-calculate the in-sample (ng/g) concentration in the plant-based meat samples. Calculated Columns allows the user to generate complex calculations, typical of third party spreadsheet software, with the output directly within the Results Table.

Figure 4. Use of automated standard addition in SCIEX OS to calculate mycotoxin concentrations in plant-based meat. Example standard addition curve (A) and Results Table (B) are shown for Ochratoxin A in the plant-based chicken I sample. The in-sample concentration was back-calculated using a Calculated Column (C).

Conclusion
 

  • A sample preparation and LC-MS/MS method was developed for the analysis of 10 mycotoxins in plant-based meat products using the SCIEX 7500 system 
  • Good chromatographic separation was shown using the Phenomenex Kinetex C18 column and 13 min runtime 
  • The LOQs in the solvent-based calibration standards ranged from 0.01 ng/mL to 0.025 ng/mL; mean LOQ accuracy ranged from 81% to 104% and mean LOQ precision ranged from 2.0% to 12.3% 
  • Matrix spikes at 1 ng/g and 10 ng/g demonstrated good absolute recovery (range: 65-128%) and precision (range: %CV 1.0-15.6%) in the 3 plant-based meat matrices 
  • Significant matrix suppression was observed for 3 mycotoxins (zearalenone, alternariol and alternariol monomethyl ether) 
  • The automated standard addition quantitation feature in SCIEX OS was used for improved quantition accuracy in the complex plant-based food matrices

References
 

  1. Bennett, J.W.; Klich, M. Mycotoxins. Clinical Microbiology Reviews 2003, 16(3), 497-516.

  2. Marin, S.; Ramos, A.J.; Cano-Sancho, G.; Sanchis, V. Mycotoxins: Occurrence, toxicology, and exposure assessment. Food Chem. Toxicol. 2013, 60, 218-237.

  3. Hussein, H.S.; Brasel, J.M. Toxicity, metabolism, and impact of mycotoxins on humans and animals. Toxicology 2001, 167(2), 101-134.

  4. Gonçalves, C.; Tölgyesi, A.; Bouten, K.; Robouch, P:.; Emons, H.; Stroka, J. Determination of Alternaria toxins in tomato, wheat, and sunflower seeds by SPE and LC-MS/MS—A method validation through a collaborative trial. J. AOAC Int. 2022, 105(1), 80-94. 
Request information