Bioremediation of aflatoxin B1-contaminated maize by king oyster mushroom (Pleurotus eryngii)

Aflatoxin B1 (AFB1) is the most harmful mycotoxin that occurs as natural contaminant of agricultural commodities, particularly maize. Practical solutions for detoxification of contaminated staples and reduction of agricultural wastes are scarce. We investigated the capability of the white-rot and edible fungus Plerotus eryngii (king oyster mushroom) to degrade AFB1 both in vitro and in a laboratory-scale mushroom cultivation, using a substrate similar to that routinely used in mushroom farms. In malt extract broth, degradation of AFB1 (500 ng/mL) by nine isolates of P. eryngii ranged from 81 to 99% after 10 days growth, and reached 100% for all isolates after 30 days. The growth of P. eryngii on solid medium (malt extract-agar, MEA) was significantly reduced at concentrations of AFB1 500 ng/mL or higher. However, the addition of 5% wheat straw to the culture medium increased the tolerance of P. eryngii to AFB1 and no inhibition was observed at a AFB1 content of 500 ng/mL; degradation of AFB1 in MEA supplemented with 5% wheat straw and 2.5% (w/v) maize flour was 71–94% after 30 days of growth. Further, AFB1 degradation by P. eryngii strain ITEM 13681 was tested in a laboratory-scale mushroom cultivation. The mushroom growth medium contained 25% (w/w) of maize spiked with AFB1 to the final content of 128 μg/kg. Pleurotus eryngii degraded up to 86% of the AFB1 in 28 days, with no significant reduction of either biological efficiency or mushroom yield. Neither the biomass produced on the mushroom substrate nor the mature basidiocarps contained detectable levels of AFB1 or its metabolite aflatoxicol, thus ruling out the translocation of these toxins through the fungal thallus. These findings make a contribution towards the development of a novel technology for remediation of AFB1- contaminated corn through the exploitation of the degradative capability of P. eryngii and its bioconversion into high nutritional value material intended for feed production.


Introduction
Aflatoxin B 1 (AFB 1 ) is a mycotoxin, produced mainly by isolates of the species Aspergillus flavus and A. parasiticus, which has potent hepatotoxic, carcinogenic and mutagenic effects on a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 humans and animals [1]. Beside AFB 1 , other aflatoxins that are structurally correlated to AFB 1 , occur as natural contaminants of foods and feeds or are generated from the metabolic transformation of AFB 1 , but have considerably lower incidence and toxicity than AFB 1 [2]. AFB 1 has been listed as a group I agent (carcinogenic to humans) by the International Agency for Research on Cancer and epidemiological studies have correlated the incidence of hepatocellular carcinoma in humans to consumption of AFB 1 -contaminated food in some world regions [3]. Human exposure to aflatoxin can result directly from ingestion of contaminated food or indirectly from consumption of products from animals that have been fed with contaminated feed. As a result of ingestion of such feeds, aflatoxins are transformed into metabolites that contaminate meat, eggs and dairy products, such as milk and cheese [4]. Aflatoxin occurrence is a major problem in a number of crops, including cereals, groundnuts, legumes and cotton seeds, which can be contaminated at any stage of production, processing, transportation, and storage [5]. Amongst the cereal grains, aflatoxin contamination concerns primarily maize and maize by-products [6].
Several approaches have been attempted for the removal of aflatoxins from contaminated commodities, including degradation of the toxin [7]. Degradation of aflatoxin requires the alteration of one or both of the molecule sites that are important for its toxic properties, namely the double bond of the difuran ring and the lactone ring of the coumarin moiety [8]. Chemical and physical methods have been found to be effective in detoxification of AFB 1 from various materials [9,10], but their use in the practice is limited, due to safety issues, possible loss of nutritional value of the treated commodities and cost implications [11]. Microbial degradation of aflatoxins has been attempted with some success, although in most cases only in axenic cultures, [12][13][14][15][16][17]. The microbial degradation of aflatoxin is achieved by the activity of enzymes able to break down the recalcitrant polyheterocyclic molecule of aflatoxin. Among fungi, the so called "white-rot" fungi are known to possess very efficient enzymatic systems for degradation of polycyclic aromatic hydrocarbons [18]. Indeed, the enzymes produced by the white-rot fungi have the crucial role of breaking down the complex molecules of lignin and other plant raw materials into low molecular weight compounds that can be assimilated by the organism. This process involves multiple ligninolytic enzyme systems consisting in extracellular oxido-reductases [19]. Encouraging results in aflatoxin degradation have been obtained with specific enzymes purified from Pleurotus spp., a genus that includes several edible and cultivable mushroom species. Motomura and colleagues [20] reported degradation of AFB 1 by culture supernatants of P. ostreatus and isolated a novel enzyme with aflatoxin-degradation activity. More recently, Yehia [21] has shown that a Mn-peroxidase (MnP) purified from P. ostreatus was able to detoxify up to 90% of AFB 1 , depending on enzyme concentration and exposure time. Other enzymes produced by Pleurotus spp. that have received attention because of their aflatoxin-degradative capability are laccases, a group of enzymes of low specificity that catalyse the oxidation of phenolic substrates via the reduction of molecular oxygen to water. Alberts and colleagues [22] reported a significant correlation between laccase activity and AFB 1 -degradative capability of P. ostreatus isolates. Loi and colleagues [23] purified a laccase isoform (Lac2) from P. pulmonarius and found that AFB 1 degradation by Lac2 via direct oxidation was 23%, which raised up to 90% in the presence of natural phenols acting as redox mediators.
Pleurotus eryngii is a cultivable edible white-rot fungus, commonly known as king oyster mushroom (KOM). The KOM is cultivated worldwide and is highly appreciated for its firm texture, taste, flavor and nutritional value [24]. The goals of the work herein presented were: a) investigate the capability of P. eryngii strains that are exploited for commercial production of edible mushrooms to degrade AFB 1; b) carry out a study on bioconversion of aflatoxin-contaminated maize into valuable feeds by P. eryngii in a laboratory-scale cultivation that mimicked the mushroom farm cultivation process; and c) assess the aflatoxin content in the residual mushrooms growth medium and investigate translocation of AFB 1 or its toxic metabolite aflatoxicol (AFOL) through the thallus to the basidiocarps (fruit bodies) of P. eryngii. The results obtained make a first contribution towards the development of a novel technology for remediation of AFB 1 -contaminated corn and its bioconversion into safe materials intended for feeds.

Fungal strains
The nine isolates of P. eryngii used in the study are strains commercially exploited for production of mushrooms. The isolates were obtained from the culture collection of the

Determination of aflatoxin B 1 and aflatoxicol
The standard solution of AFB 1 was prepared by dissolving the solid commercial toxin (Sigma-Aldrich, Milan, Italy) in toluene-acetonitrile (9:1, v/v) into amber silanized vials to obtain a 1 mg/mL solution. The exact concentration of aflatoxin solution was determined spectrometrically according to AOAC Official Method 971.22 [26]. Aliquots of the stock solution were transferred to 4 mL amber silanized glass vials and evaporated to dryness under a stream of nitrogen at 50˚C. The residue was dissolved with water-methanol (60:40, v/v) to obtain calibrant standard solutions with 0.2, 0.4, 1.2, 2.0, 4.0, 5.0, 10.0 ng/mL of AFB 1 . The standard solution of AFOL, Sigma-Aldrich, Milan, Italy) in acetonitrile was transferred to amber silanized glass vials and dried under a stream of nitrogen at 50˚C, then was reconstituted with watermethanol (50:50, v/v), to obtain calibrant standard solution with 2.0, 5.0, 10.0, 25.0, 50.0, 100 ng/mL of AFOL. Standard solutions were stored at -20˚C and warmed to room temperature before use. AFB 1 was determined by using UPLC apparatus with Acquity UPLC system (Waters, Milford, MA, USA). Data acquisition and instrument control were performed by Empower 2 software (Waters). The column used was a 100 mm × 2.1 mm i.d., 1.7 μm, Acquity UPLC 1 BEH RP-18, with an Acquity UPLC column in-line filter (0.2 μm), detected by fluorometric detector without postcolumn derivatization. The fluorometric detector was set at wavelengths of 365 nm (excitation) and 435 nm (emission). The mobile phase was a mixture of water-acetonitrilemethanol (64:18:18, v/v/v) at a flow rate of 0.4 mL/min. The temperature of the column was maintained at 40˚C. AFB 1 was quantified by measuring the peak areas at the retention time of aflatoxin standard and comparing these areas with the calibration curve of AFB 1 in the range 0.2 to 10.0 ng/mL. With this mobile phase, the retention time of AFB 1 was about 3.7 min. The limit of quantification (LOQ) of the method was 0.2 ng/mL for AFB 1 , based on a signal to noise ratio of 10:1.
One hundred microliters samples were injected into the HPLC apparatus with a full loop injection system. The fluorometric detector was set at wavelengths of 333 nm (excitation) and 418 nm (emission). The mobile phase consisted of a mixture of H 2 O-ACN (70:30, v/v) and the flow rate was 0.8 mL/min. The temperature of the column was maintained at 40˚C. AFOL was quantified by measuring peak areas at the retention time of standard solutions and comparing these areas with the relevant calibration curve at 2.0-100 ng/mL. In this analytical conditions, the retention time was about 16 min. The limit of quantification (LOQ) values were 20 μg/kg, calculated in according to s/n = 10.

Stability of AFB 1 in malt extract broth
A preparatory study was carried out to assess the decay of AFB 1 in the incubation conditions utilized for the subsequent experiments. AFB 1 was dissolved in malt extract broth (MEB) (Oxoid) to the final concentration of 1000 ng/mL and the solution was placed at 30 ± 1˚C for 30 days. Triplicate 1 mL samples were collected from the batch solution of AFB 1 after 0, 1, 5, 10, 15 and 30 days of storage at 30 ± 1˚C and the AFB 1 content of the samples was determined by UPLC/FLD.

Inhibitory effect of AFB 1 on growth of P. eryngii
The inhibitory effect of AFB 1 on growth of P. eryngii was studied in agar media supplemented with a range of concentrations of AFB 1 , i. e. 60, 120, 250, 500 and 1000 ng/mL. The media used were MEA, MEA supplemented with 5% (w/v) wheat straw (MEAS), and MEA supplemented with 5% wheat straw and 2.5% (w/v) maize flour (MEASM).
Aliquots of a 1 mg/mL solution of AFB 1 in toluene: acetonitrile (9:1, v/v) were added to melted MEA (50˚C) to obtain the desired test dilutions (60 to 1000 ng/mL). The dilutions were thoroughly mixed and poured into 9 cm-diameter Petri dishes (12 ml per dish). For preparation of MEAS and MEASM, wheat straw obtained from a local dealer was fragmented in pieces 5 mm long and maize kernels were finely ground in a laboratory mill (mulino Cyclone, International PBI, Milano, Italy) to particles 0.2 mm; 0.5 g of straw and/or 0.25 g of ground maize were transferred into 2.5 cm-diameter and 15 cm long test tubes and autoclaved at 121˚C for 30 min. After cooling, the glass tubes were filled with 12 ml of melted sterile MEA supplemented with the test dilutions of AFB 1 , thoroughly mixed and poured into 9 cmdiameter Petri dishes.
Then, each plate was inoculated with a 8 mm-diameter mycelial plug from a 20-day-old culture of P. eryngii on MEA and incubated at 30 ± 1˚C in the dark for 30 days. Five replicates per each tested isolate were prepared. Control plates were prepared with P. eryngii on MEA, MEAS and MEASM not supplemented with AFB 1 . The growth of P. eryngii was assessed by the colony diameter of three representative isolates (ITEM 13681, ITEM 13697, ITEM 13688), measured with a ruler under a dissecting microscope every 48 hours.

Degradation of AFB 1 by P. eryngii
The capability of different isolates of P. eryngii to degrade AFB 1 was tested in both liquid and solid cultures.
Liquid cultures of P. eryngii were prepared in MEB supplemented with the toxin. The assays were carried out in 12-well plates. Each well was filled with 2 mL of MEB containing 500 ng/ mL of AFB 1 and inoculated with a 8 mm-diameter mycelial plug from a 20-day-old culture of P.eryngii on MEA. Triplicate wells were prepared for each treatment and each sampling time and the cultures were incubated at 30 ± 1˚C for 30 days in the dark. After 10, 20 and 30 days of incubation, 1 mL of culture medium was withdrawn from each well, filtered through 0.45-μm-pore-size cellulose filters (Labochem Science, Sant'Agata Li Battiati, Italy) and stored at -20˚C until the analysis. Five hundred microliters of each sample were diluted with 500 μl of ultrapure water produced by a Milli-Qsystem (Millipore, Bedford, MA, USA), filtered through 0.2μm-pore-size regenerated cellulose (RC) filters (Grace, Deerfield, IL, USA) and 10 μL of the filtrate were injected directly into the UPLC apparatus through a full loop injection system. The percent degradation (D) was calculated by the formula D (%) = [(C i −C f )/ C i ] x 100, where C i was the concentration of AFB 1 in the non-inoculated control and C f was the concentration of AFB 1 in culture filtrates 10, 20 and 30 days post inoculation of P. eryngii.
Solid cultures of P. eryngii were grown on MEASM supplemented with 500 ng/mL of AFB 1 . The medium was prepared and inoculated as described above. Non-inoculated plates containing the medium supplemented with AFB 1 were used as controls. After 15 and 30 days of growth, six 10 mm-diameter plugs of culture were excised with a cork-borer along one radius of the colony, at regular distances from the initial inoculation point to the edge of the colony, and transferred to a test tube. The samples (apx. 1 g) were precisely weighted and extracted with 5 mL of a methanol-water (80:20, v/v) solution in a KS 4000i orbital shaker (IKA Werke GmbH & Co. KG., Staufen, Germany) at 250 rpm for 60 min. at room temperature. Five hundred microliters of each extract were then diluted, processed and analyzed for AFB 1 by UPLC/ FLD as described for liquid cultures. The percent degradation (D) was calculated like for liquid cultures.

Cultivation of P. eryngii in a AFB 1 -contaminated mushroom medium
Pleurotus eryngii was cultivated on a substrate similar to that used for production of commercial mushrooms, which contained ground maize spiked with aflatoxin. To produce AFB 1 , the aflatoxigenic A. flavus strain ITEM 7764 was cultured on potato-dextrose-agar (PDA) for 7 days at 25˚C in the dark; a conidial suspension in sterile distilled water was prepared from the PDA cultures and used to inoculate 200 g of autoclaved (121˚C for 30 min.) maize kernels in 1000 mL Erlenmeyer flask, to reach a final concentration of 1 × 10 4 conidia/g. Flasks were incubated for 21 days at 25˚C in the dark and analyzed for aflatoxin content. To extract aflatoxins, 300 mL methanol-water (80:20, v/v) were added into each flask, shaken on a rotary shaker for 60 min. at 250 rpm and filtered through Whatman No. 4 filter paper. Five hundred microliters of extract were diluted with 500 μL of water, filtered through RC 0.20 μm filters and 10 μL of the extract was injected into the UPLC apparatus for aflatoxin quantification. The extract was used to spike non contaminated ground maize to have the final concentration of 500 μg/kg of maize.
The substrate used for cultivation of P. eryngii contained 50% wheat straw, 25% spiked maize, 12.5% sugar beet and 12.5% field beans (Vicia faba minor). All the raw materials were obtained from a local supplier (Gruppo I.F.E srl, Matera, Italy). Wheat straw was fragmented in pieces about 2-3 mm long, sugar beet and field beans were ground in a blender Sorvall Omnimixer (Dupont Intruments, Newton, CT, USA) for 2 min. Magenta™ vessels (Sigma, 77 × 77 ×77 mm) were filled with 18 g of growth substrate with 1% (w/w) of calcium carbonate. Dry ingredients were carefully mixed and 32 mL of tap water were added to reach approx. 65% (w/w) of moisture content; the mix was finally autoclaved at 121˚C for 60 min. Once cooled, the substrate was inoculated with 3 g of spawn of P. eryngii ITEM 13681, prepared as described by Estrada and co-workers [27], with some minor modifications. Briefly 100 g of durum wheat kernels were mixed with 70 mL of warm tap water in Magenta vessels and then incubated for 28 days at 30 ± 1˚C in the dark. The AFB 1 content in the growth substrate was assessed by UPLC/FLD 0, 7, 14, 21, and 28 days post inoculation (d.p.i.), in triplicate vessels. Non-inoculated vessels containing aflatoxin-contaminated substrate were used as controls.
In order to assess the possible carry-over of AFB 1 or its metabolite AFOL in the basidiocarps, 28 d.p.i. vessels were opened, covered with a thin layer (1-2 cm) of soil and placed in greenhouse at 26˚C/ 21˚C day/night with a 12 h photoperiod to promote the fruiting of carpophores. The cultures were maintained constantly moist by nebulization of moderate amounts of water. Mushrooms were harvested after 15 additional days of growth, when the basidiomata was ripe and the mushroom cap was flat. Fresh mushrooms from each vessel were individually weighed to determine the yield and the biological efficiency (BE) as the ratio fresh mushroom weight/dry weight of the substrate, expressed as percentage. The spent substrate and the basidiocarps were analyzed for AFB 1 and AFOL. Cultures prepared with non-contaminated maize were used as controls. The experiment was performed in triplicate and repeated once.

Determination of AFB 1 and AFOL in spent mushrooms substrate and basidiocarps of P. eryngii
The aflatoxin-contaminated growth substrate of P. eryngii of each vessel was dried at 50˚C until constant weight. A portion of 20 g of substrate from each replicate was extracted with 100 mL of acetonitrile-water (84:16, v/v) by blending at high speed for 3 min with a Sorvall Omnimixer. The extract was filtered through Whatman No. 4 paper filters and 50 μL of acetic acid were added. Aliquots of 8 mL were purified with Mycosep 1 224 AflaZon column (Romer Labs). The extract was diluted 1:1 with pure water and 10 μL were injected into the UPLC apparatus [28]. The percent degradation (D) of AFB 1 in P. eryngii growth substrate was calculated like for liquid cultures.
To investigate the carry-over of AFB 1 and AFOL in the basidiocarps, 2 g of dry fruit bodies from each replicated vessel were extract in a Sorvall Omnimixer for 3 min with 50 mL of acetonitrile-water (84:16, v/v). The extract was then filtered, processed and analyzed as described for the growth substrate above.

Statistical analysis
Data were analyzed by one-way analysis of variance (ANOVA) and Tukey-Kramer multiple comparison test. The statistical analyses were performed using the GraphPad Instat 3.0 software (GraphPad Software, San Diego, CA).

Chemical stability of AFB 1
Experiments to test stability of AFB 1 in MEB showed no significant degradation in respect to control after 5, 10, 15 and 30 days at 30 ± 1˚C in the dark. The experiment was highly reproducible (coefficient of variation = 97.2 ± 3.5%).

Inhibitory effect of AFB 1 on growth of P. eryngii in agar medium
Although the three isolates showed a different sensitivity to AFB 1 , all exhibited a statistically significant growth inhibition on MEA when exposed to 500 ng/mL or higher concentration of AFB 1 (Fig 1). Concentrations of AFB 1 of 250 ng/mL or lower did not significantly affect the mycelial growth; P. eryngii ITEM 13688 was the most sensitive isolate, showing 60 ± 2% and 30 ± 2% of growth inhibition when exposed for 15 days to 1000 and 500 ng/mL of AFB 1 , respectively; P. eryngii ITEM 13681 was the most tolerant isolate, showing a growth inhibition of 37± 1% and 20 ± 3% at 1000 and 500 μg/mL of AFB 1 respectively (Fig 1).
For all the isolates tested, the inhibition of growth on MEAS and MEASM was lower than on MEA in the presence of 1000 ng/mL of AFB 1 and not statistically significant at the concentrations of 500 ng/mL or lower (Fig 2).

Degradation of AFB 1 by P. eryngii
Degradation of AFB 1 in the liquid medium (MEB) are shown in Table 1    significantly degraded the AFB 1 supplemented to the medium, in percentages that ranged from 43 to 59%. After 30 days the degradation of AFB 1 by the strain ITEM 17015 increased to 84%, and that of the strains ITEM 13696 and ITEM 13662 were 83 and 82% respectively. The strains ITEM 13688 and ITEM 13676, the least effective strains in degradation of AFB 1 , degraded AFB 1 in the medium by 71 and 65% respectively.

Mushroom cultivation on aflatoxin-contaminated substrate
The artificially contaminated cultivation substrate contained 128 ± 6 μg/kg of AFB 1 at the time of inoculation with P. eryngii ITEM 13681 (day 0). The percentages of degradation of AFB 1 at 7, 14, 21, and 28 d.p.i. are shown in Fig 3. Seven d.p.i. degradation was 13 ± 6% and 28 d.p.i. the level of degradation reached the maximum value of 86 ± 0.2% that did not change significantly until the end of the experiment (42 d.p.i). AFOL was not detected in the fungal biomass developed in the contaminated mushroom substrate (Fig 4). The analysis of basidiocarps of P. eryngii obtained on the contaminated mushrooms substrate did not reveal the presence of either AFB 1 or AFOL. In order to evaluate a possible effect of AFB 1 on production of basidiocarps, yield and BE were assessed when the basidiocarps were fully ripe and the mushroom cap was flat (Fig 5). Yield and BE of P. eryngii ITEM 13681 grown on the AFB 1 contaminated substrate were 9 ± 1 grams and 72 ± 11% respectively, not significantly different from values of the control cultures (9 ± 0,8 g; 70 ± 5%).

Discussion
Previous research has proven the capability of some white-rot fungi to detoxify aflatoxins. In particular the genus Pleurotus, which comprises edible cultivable mushrooms such as P. ostreatus and P. eryngii, produce the enzymes laccase and peroxidase that have been proved to degrade aflatoxins [20-23, 29, 30].The aim of this work was to investigate the AFB 1 -degrading capability of the species P. eryngii and explore the potential of this mushroom for development of practical technologies aimed at recovery and valorization of aflatoxin-contaminated cereal wastes.
Aflatoxin B 1 has a dose dependent inhibitory effect on the growth of P. eryngii. The tolerated level of AFB 1 varies among P. eryngii isolates and depends on the substrate composition. It is not clear if the individual variation in the tolerance to AFB 1 is due to differences in mycelium sensitivity or to different levels in AFB 1 -degrading capability of the strains. The latter hypothesis seems more likely, as the most affected isolate (ITEM 13688, Fig 1) was also the  Bioremediation of aflatoxin B1-contaminated maize by Pleurotus eryngii least effective in AFB 1 degradation (Tables 1 and 2). The presence of 5% wheat straw in the growing medium doubled (from 250 to 500 ng/mL) the concentration of AFB 1 that was tolerated without resulting in any significant growth inhibition (Figs 1 and 2). The ligninolytic enzymes involved in AFB 1 detoxification are inducible enzymes [31][32][33], therefore the higher tolerance to AFB 1 shown by P. eryngii when grown of wheat straw may be conceivably due to the increase in the synthesis of such enzymes that is induced by the lignocellulosic materials in the culture medium.
Differences in the AFB 1 -degrading capability among the nine tested P. eryngii strains were found both in liquid culture on a semi-synthetic medium, (MEB) and on an agar medium supplemented with wheat straw and maize (MEASM) (Tables 1 and 2), even if the differences among the isolates were not dramatic. In liquid medium, where the degrading enzymes can diffuse freely, the levels of degradation were quite high (81 ± 1.4 to 99 ± 0.2%) already after 10 days of growth, to reach 100% in all the tested strains after 30 of growth. On the contrary, in the agar medium, where the enzymes presumably diffused more slowly through the agar layer, AFB 1 was degraded by 71-94% after 30 days of growth. Based on these results, the capability to degrade AFB 1 exhibited by P. eryngii appears to be widely distributed in members of the species. Individual differences in the efficiency of degradation, however, exist and might be due to differences in the kinetics of biosynthesis of single enzymes, in the composition of the enzyme pool implicated in AFB 1 degradation (laccase and Mn-peroxidases), or in the degradative efficiency of different enzyme isoforms [29].
One more goal of our work was the exploitation of the AFB 1 -degrading capability of P. eryngii for the development of a technology aimed at the recovery and valorization of contaminated cereals. So far, cultivation of mushrooms is the only biotechnological process that allows for bioconversion of lignocellulosic organic waste into protein-rich and high nutritional value nutriment, thus contributing to reduction of environmental pollution [34]. We have demonstrated that this technology may also be applied to the recycling of aflatoxin-contaminated cereals, which should be otherwise destroyed or directed to alternative uses, such as production of bioethanol. In a laboratory-scale cultivation on a growth medium that is routinely used in mushroom farms, the isolate P. eryngii ITEM 13681 was able to bioconvert up to 86% of the AFB 1 in the medium (128 μg/kg) in 28 days. Although the mushroom growth medium contained 25% (w/w) of maize contaminated with 500 ppb of AFB 1 , P. eryngii did not show any significant reduction of either biological efficiency or mushroom yield.
AFOL is a derivative of AFB 1 that originates from reduction of the cyclopentanone carbonyl of the coumarine moiety [35]. The biological conversion of AFB 1 to AFOL by intracellular enzymes has been demonstrated for several fungi [35,36]. Although AFOL is 18 times less toxic than AFB 1 [35], it maintains a significant toxicity and can be reconverted to AFB 1 [36], thus still representing a safety risk. In this study AFOL was not found, as a by-product of enzymatic or pH-dependent reduction of AFB 1 by P. eryngii, in the fungal biomass. Also, the mature basidiocarps did not contain detectable levels of either AFB 1 or AFOL, thus ruling out translocation or "carry-over" of these toxins through the fungal thallus.
In conclusion, our results show that P. eryngii is able to degrade aflatoxin B 1 in both liquid and solid media. The biodegradation of AFB 1 -contaminated cereals by P. eryngii that we have demonstrated to occur in a laboratory-scale mushroom cultivation needs appropriate validation prior to become a practical technology. Nevertheless our results highlight that it may be regarded as a candidate process for bioconversion of contaminated staple cereals into valuable products intended for animal feeding. These findings make a contribution towards the development of preventative strategies to reduce AFB 1 contamination of feeds and animal-derived foods. Further research will be aimed at the identification of degradation products of aflatoxins and to the assessment of the technical and economic sustainability of remediation of aflatoxincontaminated commodities through mushroom bioconversion into non-toxic feeds.