Skip to content

Biodetoxification to mitigate mycotoxin DON in grains

Principal Investigator: Ting Zhou

Research Institution: Agriculture and Agri-Food Canada (AAFC)

Timeline: November 2019 – March 2023   


  • To develop a green technology to mitigate DON (a mycotoxin also known as vomitoxin) contamination in grains based on the microbial detoxification system recently discovered by AAFC which is able to transform DON into less / non-toxic compounds (i.e., keto-DON and epi-DON) by microbial and enzymatic conversions.


  • Assure grain farmers sustainable profits with possibilities of valorization of contaminated grains.
  • Provide multiple grain industries with innovative, highly-specific​, environmentally​ green, yet safe and effective mycotoxin mitigation tools to proactively mitigate the risks of DON.
  • Ensure consumer confidence in Canadian grains by reducing food-safety risks associated with DON.   

Scientific Summary:

Contamination of grains, such as corn and wheat, with the mycotoxin deoxynivalenol (DON) has long been a serious problem in Canada and many other countries around the world. While there are yearly variations, the prevalence and severity of DON contamination in Ontario corn has been higher in the past five years with 2018 as the worst year. In the survey by the Ontario Ministry of Agriculture and Food, 40% of the corn samples showed 2 ppm DON or higher, and 25% of the total samples had 5 ppm or more DON; much higher than previous years. Calculated using acreage and yield, Ontario produced more that 360 million bushels of corn in 2018, resulting in an estimated 90 million bushels of corn highly (5 ppm or higher) contaminated with DON. Grain Farmers of Ontario estimated a $200 million revenue loss to farmer-members in 2018 alone due to significant reduction or complete loss of market value. The severe contamination of Ontario corn crops in 2018 undoubtedly signified that new strategies are needed for mitigating this mycotoxin.

The contamination of animal feed with DON causes millions of dollars of loss to Canadian producers annually due to reduced weight gain and/or higher prices for feed with lower DON levels. These losses can be especially high during years of heavy Fusarium infection (such as feed produced from the 2016 and 2018 harvest). Currently there is no satisfactory way of mitigating the effects of DON in feed; no chemical means of destroying mycotoxins has been approved in any jurisdiction and mycotoxin binders have not been approved by either the CFIA or the FDA. There have been no binders shown to be effective for DON, and in one study binders were shown to increase DON concentration in plasma. It is believed that, even in feed that is below the threshold for DON contamination, interactions between multiple toxins may lead to deleterious effects on animals. Bioconversions / biodetoxifications with microorganisms and their enzymes offer an effective and environment-friendly green technology for mitigating the serious DON contamination problem; DON can be detoxified, not just bound to be later released after feeding as is potentially the case with binders. Enzymes are currently used in food and feed production processes, demonstrating that adding enzymes can be safe and does not disrupt the process.

Recently Dr. Ting Zhou’s research group at AAFC-Guelph Research and Development Center developed an effective biological detoxification system, known as the DON epimerization (Dep) system. A bacterial strain, Devosia mutans 17-2-E-8, originating from an Ontario agricultural soil sample, was discovered to have the capability to convert the DON molecule at its C3 position, hence detoxifying the mycotoxin under more practical feed processing conditions before the toxin enters the animal alimentary canal. This system transforms DON to 3-keto-DON, a molecule with approximately 5-fold less toxicity than DON and converts 3-keto-DON to 3-epi-DON, a compound shown to be at least 50-fold less toxic than DON. The two enzymes, designated as DepA and DepB, responsible for the DON detoxification have been identified. It was shown that DepA can convert 100% of DON to 3-keto-DON and 3-keto-DON completely disappears from solution when treated with DepB. These enzymes have been optimized, expressed, and purified in the lab and an initial characterization has shown these enzymes function over a wide pH range and are stable at 60°C. Under lab conditions these enzymes can easily be produced, purified, and used to rapidly detoxify DON.

This project further advanced the system with the main goal of applying it to detoxify DON in grains and animal feeds that have been contaminated. The project determined the range of mycotoxins that the Dep system can detoxify and assessed its effectiveness in comparison to DON. It also explored the feasibility of incorporating the Dep system into a liquid feeding system, as well as devised strategies for implementing the Dep enzymes in various feed production processes.

Additionally, the project established techniques for integrating the detoxification microorganisms at different stages of feed production. The project’s outcomes were meticulously analyzed, including a thorough comparison of different methods (using enzymes or microorganisms) developed for detoxifying DON at various points in the feed production process. These findings pave the way for future industrial applications.


Objective 1: Determine the capability of bacteria / enzymes in detoxifying multiple trichothecene mycotoxins

Expanded detoxification abilities: A Breakthrough for Enzymatic Biocatalysis.

The combination of DepA and DepB enzymes, along with the PQQ/PMS and NADPH cofactors, successfully transformed another trichothecene mycotoxin, 15-Acetyl-DON, into two different substances: 3-keto-15-Acetyl-DON and 3-epi-15-Acetyl-DON. When it came to nivalenol (NIV) and 3-Acetyl-DON, these enzymes didn’t show any activities.

How effective DepA is at dealing with 15-Acetyl-DON and DON has been determined. Under specific conditions, the enzyme managed to process 698 units of 15-Acetyl-DON and 755 units of DON per minute per milligram of enzyme. Interestingly, the enzyme showed similar efficiency for both mycotoxins. This also revealed that DepA struggles with NIV due to a specific part of NIV’s structure that doesn’t interact well with DepA.

This research unveiled that the DepA and DepB enzymes can also neutralize 15-Acetyl-DON, a mycotoxin even more toxic than DON. The fact that these enzymes can tackle multiple mycotoxins adds a significant advantage to their potential applications in treating grains contaminated with these harmful substances. This discovery broadens the horizons for using this enzyme system effectively.

Deacetylation and/or Epimerization Activities of bacterial strain Devosia mutans 17-2-E-8: The bacterial strain efficiently transformed DON within one day. For 15ADON, 3-keto-15ADON was converted to an unknown product resembling 3-keto-15ADON. 3-epi-15ADON appeared after Day 1. 3ADON epimerization to 3-epi-DON occurred in 2 days, indicating a deacetylation mechanism of the bacterial strain. Although NIV transformation was not observed, the bacteria can work on three trichothecenes, i.e., DON, 3ADON and 15ADON.

Understanding DepA’s Catalytic Mechanism: Docking analysis revealed DepA’s specificity. Trichothecenes with C4 functional groups (e.g., nivalenol) hindered DepA activity, while DON and 15ADON, without C4 functional groups, bound effectively, explaining their detoxification. Insights emphasized system limitations for Type A and certain Type B trichothecenes and the necessity of unblocked C3 hydroxyl groups.

Objective 2: Determine the factors affecting the applications of the enzymes.

Cofactor Effects: The impact of cofactors (PQQ, PMS, Ca2+) in the DepA system was probed by their exclusion from reactions. The absence of Ca2+, PMS, or both resulted in substantial reduction of DON transformation. Ca2+ functions as a pivotal divalent metal ion, vital for PQQ and amino acid interactions. PMS serves as an oxidant to regenerate PQQ from PQQH2.

Optimization via RSM: Employing Response Surface Methodology (RSM), DepA and DepB enzyme activities were evaluated under various enzyme and co-factor concentrations. A robust quadratic model was established. DepA (X1), PQQ (X2), and PMS (X3) emerged as significant contributors to the reaction rate. In contrast, DepB’s reaction rate was influenced primarily by DepB (X1) and secondarily by NADPH (X2).

Ascorbic Acid’s Role: The introduction of ascorbic acid during DepA/DepB reactions exhibited a marked enhancement in 3-keto-DON biotransformation, particularly across different DepB and NADPH concentrations. Ascorbic acid, functioning as a potent reducing agent, safeguarded NADPH and thereby amplified the efficiency of DepB. Ascorbic acid’s role in preserving NADPH enhances DepB’s cost-effectiveness in commercial applications, particularly benefiting agriculture through its approved status as a feed additive.

Pathway and Products Insights:  In the stepwise reaction, where allowed, DepA reaction ran for 2 hours followed by DepB reaction for additional 3 hours. Within the DepA phase, complete conversion (100%) of DON to 3-keto-DON was observed. In the subsequent DepB stage, 3-keto-DON underwent rapid reduction to yield 3-epi-DON (major) and DON (minor) within 30 mins. In the simultaneous reaction where both DepA and DepB were added at the same time DepA/DepB reaction yielded a predominant 3-keto-DON product alongside a minor 3-epi-DON product after 5 hours.

DON Balancing Act: In the stepwise reaction, the balance between DepA and DepB actions was disrupted, as DepA’s oxidation activity counteracted DepB’s reduction. The rapid depletion of NADPH due to strong reducibility within the system led to compromised DepB efficacy. The reduced NADPH availability caused DepB to predominantly produce minor amounts of 3-epi-DON.

Kinetics of Dep Enzymes: Kinetics of DepA and DepB were assessed using DON, 15ADON, and substrates produced from complete DON and 15ADON transformation by DepA.

pH and Temperature Effects on Dep Enzyme Activity:  DepA and DepB activities were evaluated at various pH (3.5-9.0) and temperatures (4-60°C). Optimal activity for both enzymes was observed at pH 6.0-8.0 and 22°C. Acidic conditions (pH 4.5) did not severely affect activity, indicating acid tolerance. Alkaline conditions (pH 9.0) led to reduced activity. Both enzymes showed optimal activity at 22°C.

Cloning, Expression, and Structural Insights: Successful cloning and expression of various DepB variants, along with unraveling the crystal structure of DepB_Rl.

In summary, this objective comprehensively elucidates factors influencing DepA and DepB enzyme applications for mycotoxin detoxification. Insights into enzyme kinetics, pH and temperature preferences, structural nuances, and strain-specific activities offer practical directives for optimizing enzyme utilization across diverse grain products and scenarios.

Objective 3: Determine the ability of the detoxification system to work in an animal liquid feeding system.

Effects of DepA/DepB cofactors in liquid animal feed (LF): Enzyme activities were evaluated using a response surface methodology and central composite design. Optimal enzyme and cofactor concentrations were determined for DepA and DepB reactions. PMS was identified as an essential cofactor for DepA activity, while NADPH was crucial for DepB activity.

Evaluation of DepA activity to 15ADON in LF: DepA effectively converted both DON and 15ADON to their detoxified forms. The reaction rate for 15ADON was slightly lower than that for DON.

Estimation of the cost to detoxify DON in LF: The cost of enzyme production and cofactors was estimated to find the most economical conditions for detoxification.

Objective 4: Determine the ability of the detoxification system in grain processing systems.

Effects of DepA/DepB cofactors in corn soaking /steeping water (CSW): DepA and DepB were effective in detoxifying DON in CSW, with optimal conditions for each enzyme determined.

Effect of steeping temperature on the DepA activity in CSW: Steeping temperature influenced the rate of DON detoxification in CSW, with higher temperatures resulting in faster detoxification.

Evaluation of IM-DepA and IM-DepB activities in CSW: Immobilized enzymes showed reduced activity compared to free enzymes, but IM-DepA retained its activity through multiple batches.

Determination of epimerization activities of Devosia mutans 17-2-E-8 to DON, 3ADON, and 15ADON in LF: Devosia mutans 17-2-E-8 exhibited the ability to transform DON to its detoxified form, and partial transformation of 15ADON was also observed.

In summary, objectives 3 and 4 demonstrated the potential of the DepA/DepB enzyme system for detoxifying DON and related toxins in various agricultural products and matrices. The optimal conditions for enzyme activity were determined, and the cost-effectiveness of the detoxification process was estimated. Immobilized enzyme systems showed promise for continuous detoxification processes. The study provides valuable insights into developing strategies for safer agricultural products and animal feed.

Objective 5: Produce microorganism(s) capable of detoxifying DON.

Production of Microorganisms for DON Detoxification: DON detoxifying enzymes, DyDepA and DmDepB, were expressed in E. coli BL21 (DE3) LOBSTR and G. oxydans ATCC 621. G. oxydans, despite its GRAS status and natural cofactor biosynthesis, did not show activity for DON biotransformation. DyDepA and DmDepB expressed individually in E. coli successfully oxidized DON to 3-keto-DON and transformed 47% of 3-keto-DON to 3-epi-DON respectively. Co-expressing DyDepA and DmDepB in E. coli led to complete DON biotransformation within 2 hours, including 15ADON conversion.

Objective 6: Use microorganism(s) to detoxify DON under conditions mimicking the feed production process.

Determination of DON detoxification activities of Devosia mutans 17-2-E-8 and IM- D. mutans in various feed matrices: Both free and immobilized Devosia mutans 17-2-E-8 effectively transformed DON in different matrices (e.g., DGGS, CSW and H(hydrolyzed)SW, etc.) with some variations in transformation rates and intermediate product formation.

Whole-cell biocatalysts of DyDepA and DmDepB expressed in E. coli:  Following verification in buffer, the effectiveness of whole-cell biocatalysts in DON epimerization was assessed in animal feed matrices: corn soaking water (CSW), Phase I liquid animal feed, and distiller’s dried grains with solubles (DDGS).

In CSW, the whole-cell biocatalysts achieved DON detoxification to 3-epi-DON, albeit requiring 48 hours. In animal feed, a 24% conversion of DON to 3-epi-DON occurred within 24 hours. However, no significant DON biotransformation was observed in DDGS, attributed partly to low pH-induced acid stress that may hinder the detoxification process. CSW, with its neutral pH, proved more amenable to biotransformation by bacterial biocatalysts. The pH environment emerged as a crucial factor. CSW’s neutral pH favored detoxification, while liquid animal feed’s mildly acidic pH (5.5-6) and DDGS’s extremely low pH (approx. 4) challenged biotransformation. The unique complexities of the feed matrices might also influence the efficacy of the biocatalysts.

These findings hold significance as they demonstrate that the expression of Dep enzymes in bacterial strains mitigates the need for costly exogenous NADPH supplementation. The use of whole-cell biocatalysts presents a potentially viable and cost-effective strategy for leveraging the Dep system in industrial applications.

Conclusion and Discussion

In conclusion, this study has brought forth groundbreaking advancements in the field of enzymatic biocatalysis for the detoxification of trichothecene mycotoxins. The combined efforts of DepA and DepB enzymes, accompanied by essential cofactors, showcased remarkable capabilities in transforming 15-Acetyl-DON, in addition to DON, into detoxified products. It was also discovered that the soil bacterial strain can transform 3ADON into 3-epi-DON. These achievements significantly expand the spectrum of mycotoxin detoxification and hold great promise for agricultural industries and consumer safety. The versatility of the DepA/DepB system in addressing multiple mycotoxins, including the highly toxic 15-Acetyl-DON, opens avenues for sustainable grain farming, innovative processing, and heightened confidence in agricultural products.

The potential applications of this enzymatic biocatalysis system are noteworthy. By effectively neutralizing a variety of mycotoxins, the DepA/DepB system presents itself as a transformative tool in grain industries. The implications for sustainable agriculture, food safety, and consumer trust are substantial, positioning the DepA/DepB system as a pivotal solution for addressing mycotoxin contamination challenges.

However, this study also casts light on the challenges and areas for future research that lie ahead in realizing the full potential of the DepA/DepB system for industrial applications. The sensitivity to pH variations and reduced activity of immobilized enzymes present hurdles that necessitate further refinement for practical implementation. Moreover, the transition from laboratory-scale successes to large-scale industrial processes requires careful consideration of compatibility with diverse grain matrices and optimization of enzymatic efficacy.

External Funding Partners:

Funding for this project has been provided by Agriculture and Agri-Food Canada.

Project Related Publications:

Abraham, N., Chan, E.T.S., Zhou, T. and Seah, S.Y.K. 2022. Microbial detoxification of mycotoxins in food. Frontiers in Microbiology. 13 (957148).

Abraham, N., Schroeter, K.L., Zhu, Y. and Zhou, T. et., al. 2022. Structure–function characterization of an aldo–keto reductase involved in detoxification of the mycotoxin, deoxynivalenol. Scientific Reports. 12 (14737).

Wang, W., Zhu, Y., Abraham, N., Li, X.Z., Kimber, M., Zhou, T. 2020. The ribosome-binding mode of trichothecene mycotoxins rationalizes their structure—activity relationships. International Journal of Molecular Science. 22(4): 1-17.

Li, X.Z., Hassan Y. I., Lepp, D., Zhu, Y., and Zhou, T. 2022. 3-keto-DON, but not 3-epi- DON, retains the in planta toxicological potential after the enzymatic biotransformation of deoxynivalenol.  International Journal of Molecular Biology. 23(13): 7230.