The growing global demand for healthier and more sustainable diets has fueled the popularity of plant-based diets rich in legumes. Legumes are excellent sources of protein, fiber, and starch, and their components can be separated through various technological processes to create supplements and ingredients. However, legumes are vulnerable to fungal infections, which can lead to mycotoxin contamination. Mycotoxins are toxic secondary metabolites produced by fungi, posing a significant health risk. This research aimed to investigate the fate of mycotoxins throughout the production process of legume-derived products, from raw materials to final products, to understand the potential contribution of these raw materials to mycotoxin contamination in the final commercial products. The rising global awareness of health and nutrition, combined with environmental concerns regarding meat production, is driving the increased consumption of plant-based foods, making it crucial to understand and mitigate potential hazards, such as mycotoxin contamination, within this sector. Understanding mycotoxin behavior during processing is critical to ensure the safety and quality of legume-derived products for consumers.

The literature extensively documents the occurrence of mycotoxins in various food commodities. Many studies have investigated the impact of different food processing methods on mycotoxin levels, revealing that while some processes reduce mycotoxin concentrations, complete elimination is rarely achieved. The stability of mycotoxins during processing varies significantly depending on the mycotoxin type and the specific processing method. Some mycotoxins are heat-stable, while others are more sensitive to changes in pH or other processing conditions. Moreover, processing can alter the chemical structure of mycotoxins, potentially leading to the formation of degradation products with different toxicity profiles. While studies have examined mycotoxin presence in various crops, detailed information on the fate of mycotoxins specifically during the production of legume proteins and derived products is extremely limited, underscoring the necessity for the present research. The impact of climate change on mycotoxin prevalence has also been highlighted in several studies, adding to the urgency of understanding and managing mycotoxin contamination in food production.

A total of 41 legume samples (1 kg each) from two production batches were analyzed. Samples were collected at various stages of the production process, including raw materials, intermediate products, final products, and side streams. A validated method, detailed in Chilaka et al. (2019), was used for mycotoxin extraction and analysis. This involved homogenization, milling, spiking with internal standards (zearalanone and deepoxy-deoxynivalenol), extraction with acetonitrile/water/acetic acid, clean-up using C18 solid-phase extraction and MultiSep® 226 AflaZon+ multifunctional columns, evaporation, and redisolution in mobile phase before LC-MS/MS analysis. An LC-MS/MS system (Waters Acquity HPLC coupled with a Micromass Quattro Premier XE triple quadrupole mass spectrometer) was employed with a Symmetry C18 column. A gradient elution program was used, with selected reaction monitoring (SRM) in positive electrospray ionization (ESI) mode. Method validation parameters (LOD, LOQ, accuracy, precision) are provided in the supplementary material. The identification criteria for mycotoxins in food and feed as stated in SANTE/12089/2016 were used. Matrix-matched calibration curves were constructed for quantification. The specific mycotoxins analyzed included enniatin B (ENN B), alternariol monomethyl ether (AME), deoxynivalenol (DON), T2-toxin, nivalenol (NIV), fumonisin B1 (FB1), and sterigmatocystin (STC).

The study detected several mycotoxins at different stages of the legume processing. Alternariol monomethyl ether (AME) was detected in one sample of the hulls and one sample of the final product B. Deoxynivalenol (DON), T2-toxin, and enniatin B (ENN B) were detected in one sample of the hydrated milled legume. The liquid fraction following decantation contained T2 and ENN B. Nivalenol (NIV), fumonisin B1 (FB1), and ENN B were found in one concentrate sample (intermediate product). Product C and D (intermediate products) were contaminated with T2 and ENN B. Dust samples contained high concentrations of DON and AME, with STC also present in one dust sample. Washing water and soluble C and D samples also showed contamination with ENN B in certain cases. The alkaline solubilization step was effective in reducing ENN B levels. The absence of mycotoxins in other samples might be due to low initial concentrations or the effectiveness of the processing steps. Ochratoxin A (OTA), regulated in legumes, was not detected in the analyzed samples.

The findings indicate that certain processing steps can significantly impact mycotoxin levels. Milling potentially releases mycotoxins from their matrix, while thermal treatment and solubilization reduce concentrations in the final products. The high AME concentration in dust highlights the importance of managing this byproduct. The absence of detectable levels of some mycotoxins in the final products could be attributed to either the sampled fraction not containing any of the mycotoxins, or the effectiveness of specific processing steps. The effectiveness of the alkaline solubilization in reducing ENN B levels may be due to ENN B’s ability to form complexes with alkaline ions. The study shows that while the final products had reduced mycotoxin levels, effective management of byproducts and rigorous monitoring throughout the production process are necessary to minimize consumer exposure. The low initial mycotoxin contamination level in the raw materials might have influenced the outcome. Further studies using artificially contaminated or fungus-infected raw materials are needed to establish a complete mass balance and better understand the fate of other mycotoxins.

This study provides novel insights into the fate of multiple mycotoxins during legume processing. It demonstrates the potential for processing steps to either release or reduce mycotoxin levels. The results highlight the importance of considering mycotoxin contamination throughout the entire production process of legume-derived products. Future research should focus on a larger sample size, analyses from different seasons and regions, and artificial contamination studies to refine the understanding of mycotoxin behavior and develop more effective mitigation strategies.

The study was conducted on a limited number of samples from a single processing company. The low initial mycotoxin contamination levels in the raw materials may have influenced the results. The lack of a mass balance for all mycotoxins is a limitation. Further studies using artificially contaminated raw materials are needed to gain a more complete understanding of the fate of all mycotoxins examined. More analysis of OTA is also warranted given its regulatory status in legumes.