Trichoderma Enzymes for Degradation of Aflatoxin B1 and Ochratoxin A

The contamination of agricultural products with mycotoxins causes risks to animal and human health and severe economic losses. Mycotoxicoses can be reduced by preventing fungal infection using chemical and biological approaches. The chemical strategies can release toxic molecules; therefore, strategies for biological control are being evaluated, such as using nontoxic fungi and their metabolites. This work evaluated the effect of exoenzymes produced by the beneficial fungus Trichoderma afroharzianum strain T22 in degrading Aflatoxin B1 (AFB1) and Ochratoxin A (OTA). The ability of Trichoderma to produce hydrolases was stimulated by using different inducing substrates. The highest AFB1 and OTA degradation activity was obtained using a medium containing lyophilized mushrooms and crude fiber. The T. afroharzianum T22’s ability to reduce mycotoxins may be attributed to peroxidase enzymes. This study showed that T. afroharzianum strain T22 or its peroxidase supplementation could represent a sustainable strategy for the degradation of AFB1 and OTA in feed and food products.


Introduction
Mycotoxins are secondary metabolites produced by plant-parasitic fungi, which can produce acute or chronic diseases in animals and humans [1]. Some of these compounds are lipophilic molecules mainly absorbed in the intestine and distributed in fat and soft tissues [2], with subsequent detoxification and detriment of the kidney and liver [3]. The International Agency for Research on Cancer (IARC) classified some mycotoxins (i.e., Aflatoxin B1-AFB1 and Ochratoxin A-OTA) as the most potent natural carcinogens [4]. Mycotoxins contamination can occur in the field or during processing and storage periods. Cereals; fruits; coffee; cocoa; herbal infusions; spices; fodder; products of animal origin; and derivatives (e.g., juices, fermented beverages, milk, and dairy products) are at high risk of contamination, considering their resistance to several processes used in food preparation [5]. The aflatoxins and ochratoxins (especially AFB1 and OTA) are among the most toxic and studied fungal compounds and have a significant economic and biological impact. The consumption of contaminated food and feed induces neurotoxic, immunosuppressive, and carcinogenic effects in humans and animals [6].
In this work, we evaluated the ability of Trichoderma afroharzianum strain T22 to synthesize peroxidases using different substrates. Then, we tested the effect of the exoenzymes on the degradation of mycotoxins AFB1 and OTA in different doses and at different incubation times.

Peroxidases Production
Mycotoxins are natural compounds produced by filamentous fungi, which, under suitable temperature and humidity conditions, may develop on various foods and feeds, causing severe risks to human and animal health [38]. Some works have been performed to safeguard foodstuffs from the contamination of mycotoxins and detoxify contaminated products [39].
Biological detoxification, or the enzymatic degradation of mycotoxins by microorganisms, represents an attractive approach to conventional mycotoxin inactivation. For example, Trichoderma fungi can synthesize several enzymes, such as peroxidase, chitinase, β-1, 3-glucanase, and phenylalanine ammonia-lyase, to overcome environmental stresses [40][41][42]. Previous works have shown the involvement of peroxidases in mycotoxin (especially OTA and AFB1) degradation and decontamination [43][44][45] as they are capable of oxidizing aflatoxins into polar and less toxic molecules [46]. The fungal strain and growing condition (i.e., media) affect the Trichoderma's ability to produce peroxidases [47]. In this work, the ability of T. afroharzianum T22 to release peroxidases was evaluated on different substrates at two different times (7 and 14 days). The best performance was obtained using a medium containing lyophilized mushrooms and crude fiber (enzymatic units = 0.075) after 14 days ( Figure 1). Therefore, this substrate was used to test Trichoderma's ability to induce mycotoxin degradation.

Aflatoxin B1 and OTA In Vitro Degradation
The degradation of AFB1 and OTA with the T22 enzymatic mixture was determined at different incubation times, considering the importance of this parameter and pH for the enzyme activity [48]. Times and mycotoxin concentrations were chosen based on the results already reported in the literature [49][50][51]. Figure 2 reports the kinetics degradation of OTA (at 0.01, 0.1, and 1 mg/L) following treatment with Trichoderma enzyme mixtures. After 8 days, the concentration of OTA decreased in a different and nonproportional way from the initial mycotoxin concentration. The best degradation of OTA was achieved at 0.01 mg/L with a degradation rate of 46% ( Figure 2B), followed by 37% when OTA was 0.001 mg/L ( Figure 2A) and 31% when OTA was 0.1 mg/L ( Figure 2C). Figure 3A-C reports the kinetics degradation of AFB1. After 8 days of incubation, the degradation rate was 100% at 0.001 mg/L, 63% at 0.01 mg/L, and 28% at 0.1 mg/L. Thus, the degradation of AFB1 after treatment with the T22 culture filtrate was inversely proportional to the mycotoxin concentration.
The reduction of both mycotoxins confirmed the involvement of fungal enzymes in the mycotoxin degradation (Figures 2 and 3). In both cases, the optimum reduction time was 8 days. The different decontamination yields depending on the dose of mycotoxin treated suggested a relationship between the levels of decontaminating agent and toxin. Different mechanisms could explain the slightly different behaviors of the two enzymatic kinetics. The T22 culture filtrate decontamination inversely proportional to the mycotoxin concentration could depend on the direct action of peroxidases on AFB1, which generates lower toxicity products (i.e., Aflatoxins P1, M1, B2a, Q1, and aflatoxicol) by opening the difurocoumarin moiety of Aflatoxin B1 and hydrolyzing the vinylene group [52]. The changes in the lactone and furan rings negatively affect the AFB1 toxicity, since the unsaturated 8,9-bond in the furofuran ring and the lactone ring are interested in the AFB1-DNA and AFB1-protein interactions [50,51]. The severity's order of toxicity (acute to chronic: AFB1 > AFB2 > AFM1) is ascribable to the chirality, steric hindrances, and resonance energy [53]. It has not fully elucidated the peroxidase action on ochratoxin A. It is known that the OTA contamination determines ROSs production, lipid peroxidation, DNA damage, and the loss of cell function [54]. ROSs production is due to the inhibition of antioxidative enzymes. Lipid peroxidation determines an increase in the malondialdehyde (MDA) level [54].
On the contrary, Trichoderma spp. releases antioxidant enzymes and decreases the MDA content [54]. Therefore, at the same concentration of Trichoderma culture filtrates and different OTA levels, the peroxidase level decreases as the OTA concentration increases. Moreover, the maximum degradation levels of OTA occurred at 0.01 mg/L and could depend on the release of H 2 O 2 in the reaction mixture. H 2 O 2 is the peroxidase cosubstrate, which can create an intermediate molecule that transforms a phenolic substrate into a free radical [54]. It is known that the peroxidases reveal a kinetic performance of inactivation, both in the excess or absence of H 2 O 2 [55]. Noteworthy, the prolongation of detoxification for one week at 37 • C suggests that these enzymes are stable at this time and temperature.  Error bars represent the standard deviation, and the significant difference was p < 0.05.

Aflatoxin B1 and OTA Degradation on Maize Flour and Enzyme Isolation
The formulation based on T22 exoenzymes, tested on maize flour artificially contaminated with Aflatoxin B1 (0.1 mg/L), decreased (−30%) the AFB1 levels ( Figure 4). Numerous evidence supports the application of peroxidases as food additives to degrade mycotoxins in food and feed [56]. They are easier to dose and store than producing fungi while retaining the ability to make nontoxic or less toxic products than chemical approaches.
HPLC gel filtration chromatography and SDS-page electrophoresis were used to isolate the enzyme responsible for the mycotoxin degradation. The HPLC chromatogram of T. afroharzianum strain T22 culture filtrates ( Figure 5) showed ten main peaks. Standard solutions of AFB1 were added to each isolated fraction to test the degrading activity. Fraction number 3 (peak at Rt 12.8 min- Figure 5) showed the best activity, with a degradation rate of 100% at 0.001 mg/L, 52% at 0.01 mg/L, and 51.63% at 0.1 mg/L (data not shown). SDS page electrophoresis showed three prominent bands that were isolated and used to determine which was responsible for the AFB1 reduction. Fraction 2 gave the maximum degradation percentage of AFB1. Thus, it was possible to assume that one or more Trichoderma enzymes with a molecular weight range of 60-90 kDa were involved in the degradation of this mycotoxin.

Fungal Strains and Growing Conditions
A pure culture of the Trichoderma afroharzianum strain T22 was obtained from the collection available at the Department of Agricultural Sciences of the University of Naples Federico II, Italy. Fungal cultures were incubated at 25 • C on Petri dishes containing Potato Dextrose Agar (PDA; HiMedia Laboratories, Mumbai, India). After 7-10 days, the spores were collected in a 20% (v/v) glycerol solution, and the concentration of conidial suspensions was determined using a counting chamber.

Peroxidase Assay
The assay is based on the reaction between hydrogen peroxide, phenol, and aminoantipyrine catalyzed by peroxidase, which produces a pink color whose intensity is proportional to the concentration of the enzyme.
The enzymatic activity was calculated as the difference between the final and initial absorbances. The values obtained were transformed into enzymatic units by comparison with the calibration line ( Figure S1) made using increasing concentrations of commercial "horseradish peroxidase" (Merck).

Degradation of Aflatoxin B1 and Ochratoxin A Using Enzymatic Mixtures Produced by Trichoderma
The enzymatic mixture produced by T. afroharzianum strain T22 was concentrated 20 times and dialyzed against distilled water (100 µL). Subsequently, 800 µL of physiological solution (containing 0.9% sodium chloride in water, Merck) and 100 µL of AFB1 (Sigma-Aldrich, Merck, Darmstadt, Germany) or OTA (Sigma-Aldrich) at concentrations of 0.01, 0.1, and 1 mg/L were added. The tubes were incubated under agitation for 8 days at 37 • C. Every 2 days, a sample (20 µL) was analyzed by High-Performance Liquid Chromatography (HPLC), as described below.

Aflatoxin B1 Quantification
The quantification of AFB1 in liquid samples and planta matrices was obtained by an HPLC method [47]. The chromatographic separations were carried out in a Shimadzu HPLC system (Shimadzu, Kyoto, Japan) equipped with LC-10 ADBP pumps, a SCL 10 ABP controller fluorimetric detector (λ 360 nm absorption, λ 440 nm emission), a C18 column with a 5-mm particle diameter (Phenomenex Gemini, Torrance, CA, USA), and a security Guard C-18 pre-column. The mobile phase was water:acetonitrile:methanol (50:25:25, v/v) (Sigma-Aldrich). The flow rate was 1 mL/min. The retention time of AFB1 was 8.5 min.
The analyses were performed in triplicate, and the results were analyzed with Class Vp software (Shimadzu). The areas underlying the peaks were transformed into mg/L by comparison with a calibration line obtained with different concentrations of the AFB1 standard solutions. Figure S1 reported the AFB1 calibration line.

Ochratoxin A Quantification
The quantification of OTA was obtained by the HPLC method [46]. The chromatographic separations were carried out in the isocratic mode using 65% phase A (water + 1% acetic acid) and 35% phase B (acetonitrile + 1% acetic acid) (Sigma-Aldrich). The flow rate was 1 mL/min. The injection volume was 20 µL. The experiments were performed using a Shimadzu HPLC system (Shimadzu), equipped as described above, and a Bio-Sil C18 HL 90-5 column with a diameter of 5-mm particles (4.6 × 250 mm, Bio-Rad, Richmond, CA, USA). The retention time of OTA was 5.5 min. The data were processed using Class Vp software (Shimadzu). The areas underlying the peaks were transformed into mg/L by comparison with a calibration line obtained with different concentrations of OTA standard solutions. Figure S2 reported the OTA calibration line.

Degradation of Aflatoxin B1 in Cornflour
The Trichoderma T22 enzymatic mixture's ability to degrade AFB1 was evaluated when inoculated on cornflour (solid matrix) with a known concentration of AFB1 (Sigma). Contamination was carried out on maize flour samples obtained by grinding 300 g of kernels. The flour was autoclaved (120 • C for 5 min) and cooled with a laminar flow hood for 6 h. Aliquots of 50 g were placed in Petri dishes (diameter = 20 cm; Thermo Fisher Scientific) containing 15 mL of AFB1 (1 mg/L). The solvent was removed by evaporation under a fume hood at room temperature. Triplicate samples were treated with the enzymatic mixture (5 mL), dialyzed, and then concentrated (20 times). Three samples treated with 25 mL of sterile distilled water were used as controls. The extraction of AFB1 was carried out after 8 days of incubation (37 • C), according to the Vicam Afla test method (VICAM, Milford, MA, USA), and the quantification was performed by HPLC analysis, as described previously.

Extraction of Aflatoxin B1 from Cornflour
The extraction of AFB1 from cornflour was carried out according to Vicam's Afatest instruction manual (VICAM).
The eluate (20 µL) was used to evaluate the degradation levels of AFB1 by HPLC analysis, as previously described.

Gel Filtration
The culture filtrate of Trichoderma T22, used for the AFB1 degradation assays, was fractionated by HPLC (Shimadzu), equipped as described above, using a gel filtration Superdex 75 HR 10/30 column (Pharmacia LKB Biotechnology, Uppsala, Sweden). The chromatographic course was conducted under isocratic conditions (1 mL/min flow) using water as a mobile phase.

Electrophoresis
The fractions were concentrated to obtain a final volume of 1 mL. The samples (4 µL) were diluted in the running buffer (5 µL of β-mercaptoethanol and a 95-µL Laemmli sample buffer) (Pharmacia LKB Biotechnology) and separated by vertical electrophoresis under tricine denaturing conditions -SDS-PAGE (PHAST GEL ® electrophoresis system, Pharmacia LKB Biotechnology).
The running buffers were made with 0.1 M Tris, 0.1 M tricine, and 0.1% SDS pH 8.5 at the cathode and 0.2 M Tris-HCl pH 8.9 at the anode.
The time course was 30 min, the power 80 volts. The gel at the end of the electrophoretic course was colored with silver nitrate according to the procedure described by Blum et al. [36]. The gel was fixed (10 min) in a solution containing 50% ethanol and 10% acetic acid (v/v) to remove the electrophoresis buffer and the urea. Then, it was washed (three minutes) with deionized water, incubated (1 min) in a solution consisting of 0.02 g/100 mL of sodium thiosulfate pentahydrate, and washed with deionized water (3 times for 30 s). After fixing, the gel was placed (20 min) in a staining solution (containing 0.1 g/50 mL of silver nitrate + 38 µL of 37% formaldehyde). Subsequently, it was rinsed in deionized water (twice for 20 s) to remove the silver in excess and developed in an alkaline solution containing sodium carbonate (0.3 g/50 mL) + 37% formaldehyde (25 µL) + 1 mL Na 2 S 2 O 3 ·5H 2 O (0.02 g solution/100 mL). When visible bands appeared on the gel, two rinses of 2 min were carried out to eliminate the excess reagents, and the reaction was stopped with a solution containing ethanol (50%) + acetic acid (10%). The gel was analyzed using Imaging Densitometer GS-800 (Bio-Rad), which allowed the acquisition of a digital image and subsequent analysis with Quantity One software (Bio-Rad).
All chemicals were purchased by Honeywell Fluka (Charlotte, NC, USA).

Statistical Analysis
Statistical analyses were made using Statistica software (StatSoft, Tulsa, OK, USA). The differences among the means were tested at p < 0.05.

Conclusions
Contamination of agricultural products with mycotoxins is dangerous for animal and human health. The risk of mycotoxicosis can be reduced by using beneficial microorganisms and their enzymes. Numerous studies have demonstrated the ability of Trichoderma spp. as a biocontrol agent, but only a few studies have been concerned about its ability to degrade mycotoxins. For the first time, our results demonstrated the ability of Trichoderma afroharzianum T22 to degrade AFB1 and OTA using an inducing media to release exohydrolases with peroxidase activity. High levels of degradation have been demonstrated both in vitro (up to 100% of degradation at a low concentration of AFB1) and in contaminated maize flour (30% of degradation of AFB1).
The results obtained in this study showed that T22 or Trichoderma peroxidase supplementation could represent sustainable strategies for the degradation of AFB1 and OTA. The characterization of Trichoderma peroxidases, the adverse effects of enzyme preparations on food/feed, and the toxicity of the degradation products of the selected mycotoxins are underway in our laboratory.

Conflicts of Interest:
The authors declare no conflict of interest.