Strategies to prevent and reduce mycotoxins for compound feed manufacturing

Animal Feed Science and Technology 237 (2018) 129–153

Contents lists available at ScienceDirect

Animal Feed Science and Technology journal homepage: www.elsevier.com/locate/anifeedsci

Review article

Strategies to prevent and reduce mycotoxins for compound feed manufacturing W.-X. Penga, J.L.M. Marchalb, A.F.B. van der Poela,

T

⁎

a

Animal Nutrition Group, Wageningen Institute of Animal Sciences, WIAS, Wageningen University, P.O. Box 338, 6700 AH Wageningen, The Netherlands ForFarmers, Kwinkweerd 12, 7241CW Lochem, The Netherlands

b

AR TI CLE I NF O

AB S T R A CT

Keywords: Mycotoxins Prevention Reduction Feed additives Compound feed

Mycotoxins are the secondary metabolites of fungi, especially moulds. They have over 300 types and can be easily produced ubiquitously by moulds. Many mycotoxins have been found to be toxic to most farmed animals through the diets. With the globalization of feed ingredient trade and the rapid climate changes, occurrence of mycotoxins become increasingly difficult to be predicted. Thus, the unnoticeable mycotoxin hazards can directly impact the animal production systems. Preventing or minimizing mycotoxins in feed ingredients has become an important topic from the aspect of feed manufacturing industry. The aim of this literature review is to summarize the effective strategies for feed manufacturers to minimize the mycotoxin hazards. Prevention methods, including pre-harvest field management and post-harvest storage management, are still the most effective strategies, since mycotoxins are hardly to be eliminated once they are present in the ingredients. Moreover, mycotoxin reducing effects of several feed manufacturing technologies are also reviewed. In this review, the mycotoxin reducing methods are mainly categorized into 4 methodologies: physical methods, thermal methods, chemical methods, and mycotoxin controlling feed additives. The first three methodologies mainly focus on how to reduce mycotoxins in feed ingredients during processes, while the last one on how to compensate the adverse impacts of mycotoxin-contaminated diets in animal bodies. The results showed that most of the methods reviewed show evident mycotoxin reducing effects, but of different consistencies. On the other hand, many practical factors that can affect the feasibility of each method in practical manufacturing are also discussed in this review. In conclusion, mycotoxin prevention management and the processing stage of cleaning and sorting are still the most efficient strategies to control mycotoxin hazards in current feed manufacturing.

1. Introduction 1.1. Overview: mycotoxins Mycotoxin contamination is one of the most severe threats to modern feedstuff manufacturing and animal husbandry. Modern academic detection of mycotoxins started in 1961, when mass mortality of turkeys occurred in England and then aflatoxin was discovered and accused as the culprit (Richard, 2007). In following decades, more mycotoxins and their toxicities were gradually discovered, and some mass health problems recorded before 1960s were retrospectively found related to mycotoxins (Richard, 2007).

⁎

Corresponding author. E-mail address: [email protected] (A.F.B. van der Poel).

https://doi.org/10.1016/j.anifeedsci.2018.01.017 Received 6 November 2017; Received in revised form 15 January 2018; Accepted 16 January 2018 0377-8401/ © 2018 Elsevier B.V. All rights reserved.

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

Table 1 Brief information of some concerned mycotoxins (Hussein and Brasel, 2001; Richard, 2007; Streit et al., 2012). Mycotoxin

Main origin

Host crop

Main adverse effects

Aflatoxin

Aspergillus flavus A. parasiticus

Decreased feeding performance, acute damage in organs, carcinogen and sudden death

Deoxynivalenol

Fusarium species

Zearalenone

Fusarium species

Fumonisin

F. proliferatum F. verticillioides

Ochratoxin A

A. ochraceus Penicillium viridicatum P. cyclopium

T-2/HT-2

F. sporotrichioides F. poae

Wheat Barley Maize Peanut Wheat Barley Oats Maize Maize Wheat Barley Sorghum, Rye Maize Sorghum Rice Maize, Raisins Barley Soy Coffee Corn Wheat Barley Oats Rice Rye

Impaired feeding performance, immune-suppressive,

Little acute toxicity, disorder of female estrus and reproduction, smaller litter size and weak piglets

Carcinogen to humans, pulmonary oedema in swine and leukoencephalomalacia in horses

kidney and liver damage, immune suppression, infant deformity

Inhibition of protein synthesis, disruption of DNA and RNA synthesis

Modern animal feed/human food manufacturing suffers from increasing risks of mycotoxin contamination, since mould infection can occur in almost all sectors involved, from crop cultivation in the field to the storage and logistics of finished commodities. Compared to those man-made toxins such as pesticides, preservatives and some feed/food additives, mycotoxins, as a group of natural toxins, are more difficult to be controlled because of the absence of standardized management and sufficient toxicological data (Berthiller et al., 2013). ‘Mycotoxin’ refers to a group of toxic compounds which are produced by fungi and are toxic to vertebrates and other animals in low concentrations, which excludes fungi-originated chemical compounds respectively threatening bacteria (categorised as ‘antibiotics’) and plants (categorised as ‘phytotoxins’). In addition, general discussion of mycotoxicology tends to focus on unnoticeable toxic compounds produced by moulds and exclude mushroom poisons, which threaten animals or human mainly through avoidable accidental ingestions (Bennett and Klich, 2003). Thus, discussion on mycotoxin contamination is highly related to issues of mould infection. One mould species can produce different mycotoxins, and on the other way one mycotoxin can be produced by different mould species. Currently, most concerned mycotoxins are mainly synthesized by five fungal genera: Alternaria, Aspergillus, Cladosporium, Fusarium and Penicillium (Bryden, 2012). Several dangerous mycotoxins that are frequently detected are introduced in this part (Table 1). 1.2. Most detected mycotoxins in feed manufacturing 1.2.1. Aflatoxin Among mycotoxins, aflatoxin (AFL) is the most notorious one, not only because it is the first mycotoxin discovered that lead to animals’ mass mortality, but also due to its current high adverse effects on human and animal health. The major aflatoxins consist of aflatoxin B1, B2, G1 and G2, and they are mainly synthesized by Aspergillus flavus or Aspergillus parasiticus. Aflatoxin B1 (AB1) is listed as a severe carcinogen by World Health Organization (WHO). Aflatoxin M1, a metabolic in ruminants of B1 appearing in animal tissues and fluids (such as milk and urine), has also raised special attention for its residual levels in milk products. Aflatoxins have been observed to be immunosuppressive, teratogenic and mutagenic to animals. In addition, aflatoxins can impair the intestines and decrease the feed intake of animals and animals’ growth performance (Richard, 2007), which have been supported by several longterm studies (Bailey et al., 2006; Dersjant-Li et al., 2003; Han et al., 2008). 1.2.2. Deoxynivalenol Deoxynivalenol (DON) is the most detected mycotoxin in the ingredients for feed/food production (Richard, 2007; Wu et al., 2007). DON is mainly produced by Fusarium, like F. graminearum, F. crookwellense and F. culmorum (Eckard et al., 2011). Grains of some common crops, including wheat, barley, oats and maize, are common hosts of these fungi. Rachis of maize, which can be applied as silage feed material, could also suffer from attack of these fungal. Symptoms of fungal attacks occurring in maize rachis are 130

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

mainly the rots, which, however, are hardly distinguishable from those resulting from insects’ or other animals’ attack (Eckard et al., 2011). This can lead to the ignorance of mycotoxin hazards in silage feed production. DON is one of the least acutely toxic compounds among trichothecenes (Binder, 2007), which makes its chronic toxicity relatively difficult to be diagnosed. The immunosuppressive effect is the most observed chronic disadvantage of DON. Among livestock, the pig is particularly sensitive to DON in the diet. Pigs may vomit soon after they swallow DON, and vomiting can acutely reduce feed intake and growing performance (Richard, 2007). As indicated by Pestka (2007) levels of already 50 μg/kg body weight can cause vomiting where Bryden (2012) associates dietary levels of 2–20 mg/kg in the feed with feed refusal and vomiting, respectively. DON metabolism and accumulation in pig body occurs very soon after the DON contaminated diet is consumed (Danicke et al., 2004). Different from aflatoxin, DON is much less possible to be transferred into animal-originated food (Richard, 2007). 1.2.3. Zearalenone Zearalenone (ZEN) is a frequently detected mycotoxin also synthesized by Fusarium fungi like F. graminearum or F. culmorum, and it usually co-occurs with DON. Most ZEN contamination is observed in maize, but wheat, barley, sorghum and rye are also potential hosts. Generally, ZEN shows little acute toxicity to animals. Because of its structural similarity to estradiol, ZEN can chronically affect the reproductive performance of female animals. For example, ‘fake estrus’ of female animals (especially swine) is a potential symptom of ZEN poisoning (Binder, 2007). Other main symptoms include precocious development of mammae in young gilts and preputial enlargement in young barrows. From the productive aspect, small litter size and weak piglets are the most notorious consequences when ZEN interferes with reproduction of older swine. Comparatively, chickens are more tolerant to this toxin (FinkGremmels and Malekinejad, 2007). 1.2.4. Ochratoxin A Ochratoxin A (OTA) is primarily produced by Aspergillus ochraceus and Penicillium verrucosum (Richard, 2007). This toxin occurs in storage environment more than in the field. It usually occurs in raisins, barley, soy products and coffee at low level (Richard, 2007). Adverse consequences of consuming OTA mainly include organ damage, immune suppression and infant deformity. In body, kidney suffers most from OTA, including the symptoms of vascular lesions, renal haemorrhages, renal lymph nodes enlargement and so on (Stoev et al., 2002). OTA was accused of being responsible to the once-sensational ‘mycotoxic porcine nephropathy (MPN)’ in 1970s (Stoev et al., 2002), and it also had been suspected to have resulted in the human Balkan endemic nephropathy (BEN) occurring in the 1950s (Pfohl-Leszkowicz and Manderville, 2007). OTA is easy to accumulate in organs due to its low metabolizing velocity in animal bodies, and thus human food originating from these organs also have the risks of OTA accumulation. 1.2.5. Fumonisins Fumonisin (FN) covers a group of relatively newly discovered mycotoxins (mainly fumonisin B1, fumonisin B2 and fumonisin B3), produced primarily by F. verticillioides and F. proliferatum. Corn is the major commodity affected by this group of toxins and their occurrence also has been found in sorghum and rice. Consuming fumonisin can lead to tumors growing in lung, liver and kidney of rodents and pigs (Richard, 2007). One distinguishable problems of fumonisin is that it can interfere with the sphingolipid metabolism of horses and foster tumour growth in their brains, and this is quite an unusual pathogenic pattern compared to other mycotoxins. 1.2.6. T2/HT-2 T2 and its derivate HT-2 are also produced by Fusarium, especially F. sporotrichioides. Since mutual transformation between T-2 and HT-2 is frequent under natural conditions, they are often combined as one detecting unit in analytical detections. T-2 and HT-2 have been found to occur in crops including corn, wheat, barley, oats, rice, rye and in temperature ranging from 6 to 24 °C. A 4-year (2002–2005) survey on barleys and oats at harvest in England showed that 16% of the samples were contaminated by high H-2 and HT-2 (average combined concentration > 1000 mg/kg), where DON concentrations were quite low (∼300 mg/kg) in the same survey (Scudamore et al., 2007). T-2 toxin can inhibit protein synthesis, and this can lead to even disruption of DNA and RNA synthesis. Among the farm animals, poultry is quite sensitive to it. 1.3. Growing impacts of mycotoxins on animal production system Different mycotoxins could contaminate many crop species. Among these crops, some are quite important as both human food and animal feed, like maize, wheat, and barley. For example, maize is quite susceptible to various fungi, including A. flavus, A. parasiticus, F. graminearum, F. verticillioides, Penicillium spp., and Diplodia maydisand. Thus, maize could be easily contaminated by aflatoxins, DON, ZEN, Fumonisins etc. (Reddy et al., 2009). In recent years, mycotoxin hazard becomes a worldwide increasing problem in human food or animal feed production. One three-year survey (2009–2011) (Rodrigues and Naehrer, 2012) tested 7049 ingredient samples (corn, soybean, soybean meal, wheat, dried distillers grains with solubles and finished) from America, Europe or Asia, 81% of which were contaminated by at least one of the target mycotoxins (aflatoxin, ZEN, DON, FUM and OTA). Among the samples, corn suffered the highest frequency of mycotoxin contamination, and soybeans or soybean meals met less problems. In the context of modern feed manufacturing, the occurrence of mycotoxins may become more complicated through the frequent international trade of ingredients Legislative or certificating organisations around the world have been trying to establish maximum levels of mycotoxins in feedstuffs (Cheli et al., 2014). For example, European Commission (EC) published maximum levels of several mycotoxins in the EU. The mycotoxins targeted are: AB1, DON, ZEN, OTA, FB1 + B2 and T-2 + HT-2 (Table 2). All these maximum levels are low, and some 131

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

Table 2 Regulated and recommended maximum levels of mycotoxins in feed materials set by the European Commission (moisture content: 12%). Mycotoxin

Products

Regulated maximum level (mg/kg, or ppm)

Aflatoxin B1a

All feed materials Complete feedingstuffs for dairy animals Complete feedingstuffs for calves and lambs Complementary and complete feedingstuffs for cattle, sheep and goats (except complementary feedingstuffs for dairy animals, calves and lambs) Complementary and complete feedingstuffs for adult pigs and poultry Maize by-products Other cereals and cereal products Complementary and complete feedingstuffs except for: -complementary and complete feedingstuffs for pigs -complementary and complete feedingstuffs for calves (< 4 months), lambs and kids Maize by-products Other cereals and cereal products Complementary and complete feedingstuffs for piglets and young sows Complementary and complete feedingstuffs for sows and fattening pigs Complementary and complete feedingstuffs for calves, dairy cattle, sheep (including lamb) and goats (including kids) Cereals and cereal products Complementary and complete feedingstuffs for pigs Complementary and complete feedingstuffs for poultry maize and maize products Complementary and complete feedingstuffs for pigs, horses, rabbits and pet animals Complementary and complete feedingstuffs for fish Complementary and complete feedingstuffs for poultry, calves (< 4 months), lambs and kids Complementary and complete feedingstuffs for adult ruminants (> 4 months) and mink Oat milling products (husks) Other cereal products Compound feed, with the exception of feed for cats

0.02 0.005 0.01 0.02

DONb

ZENb

OTAb

Fumonisin B1 + B2b

c

T-2 + HT-2

a b c

Guidance Value (mg/kg, or ppm)

0.02 12 8 5 0.9 2 2 3 0.1 0.25 0.5 0.25 0.05 0.1 60 5 10 20 50 2 0.5 0.25

European Commission (2002). European Commission (2006a). European Commission (2013).

of them are lower because of special toxicological backgrounds. For example, maximum level of AB1 in feedingstuffs for dairy animals is set lower than that in other feeding strategies because of the concern that AB1 in cattle bodies can be transferred into AM1 in milk. Another example is that the maximum levels of DON and ZEN are set lower because pigs are more sensitive to them. On the other hand, some mycotoxins, having different distinct forms, are detected as one compound in EC legislatives. For example, the reasons can be the easy transformation between forms (T-2 and HT-2) or the similar toxicities and appearing frequency (FB1 and FB2). In addition, EU laws and legislations against mycotoxins are advancing as the accumulation of toxicological studies increase. For example, EC maximum level of T-2 + HT-2 was recently set in 2013, after their toxicological studies had been performed for decades (EC, 2013). Establishment of legal maximum concentrations of mycotoxins does not mean that mycotoxins at low concentrates are safe. Small amounts of mycotoxins can also interrupt the normal cellular functioning, and even result in cell deaths and accumulative pathological changes (Table 3). Moreover, low levels of mycotoxins in diets can lead to subclinical problems which is less evident, Table 3 Cascade of cellular events due to the cytotoxicity of several mycotoxins (Riley, 1998; Bryden, 2012). Mycotoxin

Cascade of cellular events

Aflatoxin

Metabolic activation → DNA modification → cell deregulation → cell death/transformation (metabolic activation → disruption of macromolecular synthesis → cell deregulation → cell death (apoptotic)) Inhibition of protein synthesis → disruption of cytokine regulation → altered cell proliferation → cell death/apoptosis? Sphinganine N-acyltransferase → disrupted lipid metabolism → cell deregulation → cell death/apoptosis (disrupted delta-6-desaturase activity → disrupted fatty acid and arachidonic acid metabolism → cell death) Disruption of phenylalanine metabolism → reduced Phosphoenolpyruvate carboxykinase (PEPCK) → reduced glyconeogenesis → cell death (metabolic activation → inhibition of protein/DNA synthesis → apoptosis?) (altered membrane permeability → disrupt calcium homeostasis → cell deregulation → cell death) Cytosolic oestrogen receptor → estrogenic response → disruption of hormonal control → ? Inhibition of protein synthesis → ? → cell death (apoptotic?) (Transient Ca2+ elevation → endonuclease activation → apoptosis)

Deoxynivalenol Fumonisins Ochratoxin

Zearalenone T-2

132

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

including a slight decrease of feed intake and growth, nutrient utilization and (re)productive performance (Bryden, 2012). Climate change is also fostering mycotoxin hazards. Different fungal strains show preference to specific environments. For example, some Aspergillus flavus strains can prosper and produce aflatoxin under low humidity while other Aspergillus flavus strains not (Magan et al., 2011). Thus, acute climate changes can contribute to changes of mycotoxin distribution. One recent example is the increasing A. flavus infection and aflatoxin contamination during summer in Southern Europe, where these problems were uncommon when summers used to be less hot and dry (Streit et al., 2012). On the other hand, the climate changes indirectly influence mycotoxin accumulation by influencing other local biological factors. For example, draught was associated with increasing insect attack on maize (Miller, 2008), and these broken kernels usually contain higher mycotoxins (Jouany, 2007). The update of toxicological studies also helps to refresh the comprehension towards mycotoxins. For example, it was believed that mycotoxins hazards are much less severe to ruminant animals, but recent studies turned down this concept. This outdated concept was based on the physiological fact that rumen can degrade several mycotoxins, like OTA, DON and ZEN. However, aflatoxins and fumonisins have recently been found much less degradable in the rumen than expected (Fink-Gremmels, 2008). Diverse feed resources, including concentrate feed, silage commodities, and natural pasturing, also increase potential dietary mycotoxin hazards to cattle (Fink-Gremmels, 2008). On the other hand, ensiling was once believed a good process to eliminate fungi and mycotoxins because of its high moisture, low pH and low oxygen concentrate environment (Eckard et al., 2011). However, Penicillium, one recently noticed mycotoxigenic mould, has been found able to prosper under the environment of ensiling. Mycotoxins produced by Penicillium mainly include cyclopiazonic acid (CPA), patulin (PAT), mycophenolic acid (MPA), and roquefortine C (ROC). PAT and ROC have been observed to cause acute mycotoxicoses in cattle, including paralysis, abortion and even death, but MPA results in less acute symptoms like immunosuppression (Mansfield et al., 2008). In 120 maize silage samples randomly collected by Mansfield et al. (2008) from all around US, contamination frequency of these mycotoxins is quite high: 60% (ROC), 42% (MPA), 37% (CPA), and 23% (PAT). Since mycotoxin hazards, accompanying fungal infection, could occur ubiquitously, strategies against mycotoxins should cover a long chain from crop cultivation to feed manufacturing. In the following parts of this review, these strategies are introduced.

2. Prevention strategies against mycotoxin contamination 2.1. Pre-harvest prevention Pre-harvest mycotoxin prevention strategies are mainly performed during the phase of crop cultivation in the field. Based on plant-fungus interaction patterns, toxigenic fungi are classified into four categories (Miller, 2008): 1) fungi as plant pathogens (e.g. Fusarium graminearum); 2) fungi that produce mycotoxins on aging or stressed plants (e.g. Aspergillus flavus on maize); 3) fungi that colonize the crop and potentially contaminate the post-harvested commodity (e.g. Aspergillus flavus in subtropical maize); 4) fungi left over in the soil or decaying plant materials which potentially invade newly planted crops or their commodity (e.g. Aspergillus flavus on many commodities). These interaction patterns lead to the prevention methodologies. First, different strains of one crop species can show different effects on fungal resistance and mycotoxin control, which is mainly determined by genotypes (Brandwagt et al., 2000; Miedaner et al., 2001; Gao et al., 2007). Thus, crop breeding is a logical methodology to relieve crops from mycotoxin contamination. However, practical effects of breeding still have many limitations to overcome. For example, Fusarium head blight (FHB), which usually occurs to corn, is a severe disease caused by Fusarium infection, and infected plants usually contain high Fusarium mycotoxins. Wounds on plant body, which usually result from insects, crucially foster this infection. Thus, the trait resistant to both fungal infection and insect wounds are crucial for breeding works of plants’ mycotoxin resistance (Mesterhazy et al., 2012). However, the mycotoxin resistant trait is strongly limited by environment, and its heritability is also limited (Massman et al., 2010), which greatly bothers the breeding work. Generally, current development of hybrid breeding for mycotoxin resistance has been left far behind the rapid growing of academic mycotoxin knowledge (Mesterhazy et al., 2012). Cultivation management also takes a crucial role in pre-harvest mycotoxin control. Crop rotation with proper species, like wheat and legumes, is effective to control mycotoxin contamination (Richard, 2007). Proper practice of soil cultivation, like ploughing, minimum tillage, no-tillage technology, also contributes to the removal or isolation of the infected crop residues, which decrease the mycotoxin risks invading new cultivated crops (Richard, 2007). Proper sowing date is highly recommended since a delay could increase the risks of DON accumulation (Eckard et al., 2011). On the contrary, improper supplies of water and nutrition can foster the soil stress, which fosters mycotoxin accumulation (Richard, 2007). Preventing the attacks from other organisms, like insects and other animals, is also important, since these attacks can directly interfere with plants’ resistance capacity towards mycotoxins. Even though the utilization of pesticides or fungicides is rarely accepted by some developed farming concepts, these fungi controlling products could contribute to mycotoxin prevention. For example, it was found that crops from organic farms potentially contain six times as much OTA as compared to the crops from conventional farms (Czerwiecki et al., 2002, reviewed by Pfohl-Leszkowicz and Manderville, 2007). Rapid developing of information technology has led to mathematical models for regional mycotoxin prediction on field scale. However, compared to the developed models for predicting crop diseases, models for mycotoxin occurrence are still in their infancy because of the relatively low accuracy (Miller, 2008). Since dynamics between mycotoxin occurrence, climate factors and other biological factors remain unclear, these modelling systems still have high uncertainty to overcome (Magan et al., 2011).

133

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

2.2. Post-harvest prevention Post-harvest management is also crucial to decrease the mycotoxin hazards, since harvested grains lack much defending ability compared to entire plants. Storage is one crucial stage for the post-harvest prevention of mycotoxins. Storage of harvested grains forms a man-made ecosystem, with the dynamics of various biotic (grain, bacteria, yeast, fungi etc.) and abiotic factors (water, air, temperature, etc.). Among these factors, temperature and humidity are crucial for the fungal infection and mycotoxin contamination. When the humidity is high at 17–20%, DON concentrations can increase significantly due to activity of Fusarium spores (Birzele et al., 2000). Keeping a low moisture in storage environments is crucial for controlling the level of fungi and mycotoxins. Generally, a water activity less than 0.7 can efficiently eliminate mould spoilage (Magan et al., 2003), but insufficient drying of grains before storage can significantly decrease the mycotoxin eliminating effects of dry storage environment. Biotic factors, like bacteria and other microbes, also foster the mycotoxin problems in storage. The break-in of these organisms could be due to the impaired tightness of storage, but unnoticed organisms can also sneak into grains before storage. Insect is a common example of biotic factors impairing the bio-dynamics in storage ecosystem. Different from microbes, insects can survive a much broader range of temperature and humidity. Activities of insects can produce metabolite heat and water, which form so-called ‘hot spots’ inside the grain heap. These spots can soon provide a suitable environment for fungal growth (Magan et al., 2003). Thus, keeping insects or similar biotic factors from storage environment is a necessary method for food and feed hygiene and safety. 3. Physical removal of mycotoxins by feed manufacturing technologies It is insufficient to overcome mycotoxin problems only by prevention strategies. To deal with mycotoxin contaminants which are present in feedingstuffs, more strategies have to be taken into account. One classic example of these strategies was ‘dilution’, which means to mix the contaminated materials with clean feed ingredients to achieve a mixture whose mycotoxin concentrations can be legally low. However, ‘hot spot’ hazard, which means a partial high toxicity in the feed materials, was then found to threaten consumers’ health. Another typical strategy was to assign the contaminated materials to the livestock species which were thought to have a higher mycotoxin resistance, like ruminants. In recent years, however, this method has also been found dangerous since these animals have been found not as resistant to mycotoxins as having been expected (Part 1). Currently, both these methods are prohibited in the European Union (EC, 2006b). As required by European Commission (2006b), cereal grains should meet the legal and recommended levels of mycotoxins (Table 2) before entering the ‘first-stage processing’, which means ‘any physical or thermal treatment, other than cleaning, sorting and drying’. Nevertheless, feedingstuffs with legally low mycotoxin levels might still be threatening to consumers (Alassane-Kpembi et al., 2013; Streit et al., 2013). Thus, reducing mycotoxin concentrations as low as possible becomes a crucial task for compound feed manufacturers. This part 3 focuses on several methodologies that can physically remove from feedingstuffs. 3.1. Cleaning and sorting Generally, cleaning means to remove external materials like large foreign trash and dust, and sorting means to pick out the inferior kernels from the healthy ones. These unwanted materials can decrease feed quality and even impair the consumers’ health, as well as attract toxigenic fungi. Moreover, cleaning and sorting practices are expected to reduce concentrates of mycotoxins to a low level in respect to EU requirements, which gives high importance to the mycotoxin reducing effects of cleaning and sorting. One traditional sorting method is hand sorting, which means to pick out the inferior grains by human labour. This method is based on the visible abnormalities of kernels, like smaller size, mould spots or colour changes, which could be associated with higher possibility of mycotoxin contamination (Kabak et al., 2006). Effectiveness of this traditional method is highly dependent on the experiences of workers, and it can be only performed in small-scale production situations. Another traditionally applied sorting method to reduce mycotoxins is wet sorting. Main mechanism of wet sorting is to remove the buoyant of kernels in liquid, based on their different densities. These lighter fractions contain at least within one batch more mycotoxins since their decreased densities could be involved with fungal activities in kernels (Huff, 1980, Huff and Hagler, 1982, 1985). Several articles on mycotoxin reducing effects of these traditional methods were published. Matumba et al. (2015) compared the mycotoxin reducing effects between hand sorting and water flotation. In their study, hand sorting helped to increase the safety of products by eliminating over 90% of all kinds of target mycotoxins (aflatoxins, fumonisins, DON etc.), and in comparison, the reducing effects of liquid flotation were highly varied (32–78%) between mycotoxins (Table 4). Types of solutions could be crucial for the wet sorting’s mycotoxin reducing effects by removing contaminated parts without specificity. For example, Huff and Hagler (1985) observed that liquid flotation with 30% sucrose solution could decrease DON concentration by 76.7% in a maize sample with 6 ppm initial DON, while flotation with water could only reduce that by 11.7% in a backup sample (Table 4), even though similarly large differences between solutions were not repeatedly observed in other samples of the same study. In another study (van der Westhuizen et al., 2011), the mycotoxin reducing effects of simple washing by water were examined, but limited effects were observed, which was much inferior to the effects of hand sorting in the same study (Table 4). The traditional cleaning and sorting methods introduced above, however, are inapplicable in large-scale manufacturing situations, for their low efficiency and high costs. The authors recognize however, that these methods provide valuable information what can be achieved, especially since the fast advances of image processing and computer power will make this economical achievable in the near future. In large-scale feed manufacturing, cleaning and sorting are highly mechanized, which usually combine several mechanical 134

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

Table 4 Mycotoxin reducing effects of cleaning and sorting processes. Samples a,b

Wheat Wheatb,c Oatd,e Maizeg

Maizeh

Conditions

Mycotoxin

Reduction (%)

Mass Loss (%)

Air screen cleaner gravity separator Air screen cleaner gravity separator Air separator Liquid flotation (brine solution)q Liquid flotation (water) Liquid flotation (30% sucrose) Water flotation

DON DON T-2/HT-2 Aflatoxin DON DON NIV DON 3-AcDON + 15-AcDONi Aflatoxin B1 Aflatoxin B2 Aflatoxin G1 Aflatoxin G2 FB1 FB2 FB3 NIV DON 3-AcDON + 15-AcDONi Aflatoxin B1 Aflatoxin B2 Aflatoxin G1 Aflatoxin G2 FB1 FB2 FB3 FB1 + FB2 + FB3i

74.7 88.9 48.3 ± 44.7f 73.4 ± 2.1f 11.7 76.7 57.3 60.4 32.3 62.6 64.2 55.9 68.2 74.3 77.8 67.0 95.4 96.3 95.7 98.7 96.0 99.2 97.7 92.1 95.7 91.6 71 ± 18f 5 ± 8f 13 ± 12f 12 ± 33f 9 49.8, 45.4, 46.7 & 52.7m 100 n 100n,o 22.6 ± 10.5f

28.0 32.9

Hand sorting

Maizej Maizej,k

Wheatl

Hand sorting Washing for 5 min Washing for 10 min Washing for 30 min Washing for 60 min Sieving, scouring and polishing

Durum wheatp

Sieving machine

DON NIV HT-2/T-2 DON

3.4 3.4 7.5 6.4

14.3

2.5 ± 0.5f

a

Wheat cultivar BRS Parrudo. Tibola et al. (2016). c Wheat cultivar BRS 374. d Schwake-Anduschus et al. (2010). e Samples cover 4 cultivars from 10 cultivating sites. f Mean ± Standard deviation based on 24 different samples. g Huff and Hagler (1985). h Matumba et al. (2015). i Different mycotoxins were merged as one index. j van der Westhuizen et al. (2011). k Inferior maize grains had been discarded by hand sorting. l Lancova et al. (2008). m DON reduction respectively for 4 samples with different initial DON concentrations: 909, 108, 92 and 2985 (μg kg−1). n Mycotoxin concentrations in cleaned samples were lower than the limit of detection. o Not all uncleaned samples contained targeted mycotoxins. p Visconti et al. (2004). q ‘Each sample was stirred for 2 min in a saturated sodium chloride solution (brine solution, d-1.2)’. b

equipment, including air separators, sieves, gravity separators and indented cylinders. Alike the wet flotation, the mechanical sorting is mainly based on the lower density of potential infected grains. Several authors have studied the mycotoxin decontaminating effects of these mechanical methods (Lancova et al., 2008; Visconti et al., 2004; Tibola et al., 2016; Schwake-Anduschus et al., 2010). In the study of Tibola et al. (2016), mechanical cleaning was observed with general good DON decontaminating effects in wheat, but difference occurred between wheat cultivars (74.7 and 88.9% respectively). Moreover, some articles also focused on those less commonly occurred mycotoxins, like NIV and T-2/HT-2. Lancova et al. (2008) observed that almost 100% of NIV and T-2/HT-2 were eliminated through cleaning machine (including the process of Sieving, scouring and polishing). However, in the study of SchwakeAnduschus et al. (2010), the cleaning by air separator was observed to result in high variation (2.3–100%) in T-2/HT-2 reduction, and this may be partly due to the high diversity of the cultivars and places of origin of these oat samples (4 cultivars from 10 cultivating sites) (Table 4).

135

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

Table 5 Mycotoxin reducing effects of dehulling processes. Samples Maize

a

Conditions 11% Moisture 20% Moisture

Oatc

4 cultivars from 10 locations

Maizee

Wooden mortar and pestle

a b c d e

Mycotoxins b

Aflatoxin B1 + B2 + G1 + G2 Aflatoxin B1 Aflatoxin B1 + B2 + G1 + G2b Aflatoxin B1 T-2 HT-2 NIV DON 3-AcDON + 15-AcDONb Aflatoxin B1 Aflatoxin B2 Aflatoxin G1 Aflatoxin G2 FB1 FB2 FB3

Reduction (%)

Mass loss (%)

93.4 88.8 91.4 89.8 88.3 ± 14.2d 100 71.8 62.9 51.6 87.8 91.9 94.1 94.4 85.2 90.3 88.9

20

29

Siwela et al. (2005). Different mycotoxins were merged as one index. Schwake-Anduschus et al. (2010). Mean ± Standard Deviation. Matumba et al. (2015).

3.2. Dehulling In feed manufacturing, dehulling is a mechanical process to remove the outer layer of the grains. This process is widely used for cereals, legume seeds and oilseeds with outer fibrous compositions (Rios et al., 2009a). Dehulling also helps to increase the fineness of subsequent milling products, such as flour and semolina (Cheli et al., 2013). The removed hulls can be re-utilized to meet some livestock’s nutritional requirements for fibres. From the aspect of feed hygiene and safety, this removal can incidentally decrease the contaminants attached to the outer layer, including fungi and mycotoxins. Several studies on the mycotoxin detoxification effect of dehulling process have been published. In their results, the mycotoxin contents in the dehulled feed materials have been significantly decreased (Table 5). Dehulling could be performed with different tools. For example, Siwela et al. (2005) studied the aflatoxin reduction in maize through ‘a dehuller’; Schwake-Anduschus et al. (2010) studied the T-2/HT-2 reduction in oat by laboratory huller; Matumba et al. (2015) applied traditional dehulling with wooden mortar and pestle to examine mycotoxin reduction in maize. These studies indicated that many factors, including crop species, crop cultivars and processing duration, could crucially influence the mycotoxin reducing effects of dehulling process. Moreover, Rios et al. (2009a) proposed two curved-line models to explain the relationship between dehulling duration, grain mass loss and the remaining DON proportions in two wheat samples: the first model (Fig. 1) indicated the grain mass loss percentage (Y axis) against duration (X axis), and the second model (Fig. 2) indicated the remaining DON percentage (Y axis) against the grain mass loss percentage (X axis). For example, it could be observed that 20 min’ milling process reduces about 20% grain mass (Fig. 1), which takes away about 60% DON (Fig. 2). Similarly, House et al. (2003) also proposed a similar curving line between grain loss mass and remaining DON percentage in barley (Fig. 3). In this study, high reduction (e.g. 65%) of DON was observed with 15% mass loss within a very short dehulling duration (15 s), and 1.5 min seemed long enough to eliminate 90% of DON, but with a mass loss of 40% (Table 6).

Fig. 1. Relationship between dehulling duration (min) and percentage of grain mass loss in two durum wheat grain samples with different initial DON concentrations (A1 presents the first sample which contains 382 mg kg−1 initial DON; A2 presents the second sample which contains 4203 mg kg−1 initial DON) (Rios et al., 2009a).

136

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

Fig. 2. Relationship between percentage of grain mass loss and proportions of remaining DON in two durum wheat grain samples with different initial DON concentrations (A1 presents the first sample which contains 382 mg kg−1 initial DON; A2 presents the second sample which contains 4203 mg kg−1 initial DON) (Rios et al., 2009a).

Fig. 3. Relationship between percentage of remaining grain mass and proportions of remaining DON in barley samples (House et al., 2003).

Table 6 Effects of dehulling duration on grains’ mass loss and mycotoxin reduction. Sample Durum Wheat

a

Barleyb

a b

Mycotoxin

Dehulling duration

Reduction

Mass loss

DON

5 min 10 min 30 min

45% 60% 70%

10% 20% 35%

DON

15 s 45 s 90 s

66% 82% 90%

15% 26% 40%

Rios et al. (2009b). House et al. (2003).

3.3. Milling Milling is a process to physically break grains into smaller particles. The application of this process is based on that milled fractions of different sizes usually contain different main nutrients. For example, large granules (20 μm) of milled pea seeds contain mainly starch particles, when small particles (1–3 μm) of these milled seeds are protein-rich (Pelgrom et al., 2013). Thus, milled fractions of different sizes are usually separately preserved for manufacturing different products. In practical situations, the air separator equipped within typical milling machine is for separating fractions with different densities, and these fractions can be collected for further manufacturing. Logically, milling processes can only redistribute the existing mycotoxins into different fractions, rather than to eliminate or deactivate these toxins. Since the outer layer of kernels is easier to be contaminated than inner parts (Lancova et al., 2008), milling is expected to result in lower mycotoxin concentrations in fractions from grains’ inner parts (e.g. flour or semolina) compared to that in initial grain samples, meanwhile higher mycotoxin levels in those fractions originating from outer layer (e.g. brans and shorts) (Cheli et al., 2013). While in highly contaminated grains, high mycotoxin levels could also be observed in the finest fractions originating from endosperm. Rios et al. (2009b) studied the distribution of DON in the milled fractions of two French durum wheat samples with different initial DON concentrations (382 and 4204 mg kg−1, Table 7). In this study, samples were successively milled by two types of 137

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

Table 7 Redistribution of mycotoxins in milled durum wheat samples of different initial contaminating levels (Rios et al., 2009b). Sample A1: initial DON concentration before milling: 382.0 ± 9.7 mg kg−1 Mill fraction Total Total Total Total

semolina break flours reduction flours bran & shorts

Yield (%) 76.4 3.2 3.7 16.8

Concentration (mg kg−1)

DON (%) 40.8 4.8 3.5 50.8

199.3 637.3 374.8 908.5

± ± ± ±

a

62.0 166.7a 56.1a 478.0a

Change rate (%)

CV (%)

−47.8 +66.8b −1.9b +137.8b

31.1 26.2 15.0 52.6

Change rate (%)

CV (%)

−78.4b +33.8b +6.3b +116.2b

21.0 33.6 26.1 47.1

b

Sample A2: initial DON concentration before milling: 4203.5 ± 202.7 mg kg−1 Mill fraction

Yield (%)

DON (%)

Concentration (mg kg−1)

Total Total Total Total

75.9 3.3 4.2 16.6

49.6 3.7 4.1 42.6

3015 5623 4467 9087

semolina break flours reduction flours bran & shorts

± ± ± ±

632.6a 1890.4a 1164.0a 4279.5a

a

Mean ± Standard Deviation. Change rate indicates the mycotoxin concentration in milled fractions compared to that in initial unprocessed samples. ‘+’ represents the increase of concentration, and ‘−’ represents the decrease. b

rolls (‘break rolls’ and ‘reducing rolls’), and then different milled fractions were collected for mycotoxin tests. In both samples, the fraction ‘bran and shorts’ were observed to contain high proportions of the total DON (50.8% and 42.6%), but this did not signal that the removal of this fraction was sufficient to overcome the mycotoxin hazard, since the finest fraction ‘total semolina’ still maintain high DON proportions (40.8% and 49.6%). Similar results were observed in the study of Lancova et al. (2008), which was performed on 4 wheat samples of different initial mycotoxin contamination levels (909, 108, 92 and 2985 mg kg−1 respectively). The results indicated that 50–65% of total initial DON were removed with brans and rests (shorts, losses etc.), but certain proportions of total DON remained in the break flour (10–22%) and the reducing flour (24–28%). Visconti et al. (2004) also observed about 47% of total DON in unprocessed durum wheat samples were distributed to milled semolina fraction. However, these proportions of redistributed mycotoxins were only about the ‘amount’ of mycotoxins, which indeed are much less crucial than mycotoxin concentrations. Thus, mycotoxin redistributing effects of the milling process could be interpreted from the aspect of mycotoxin concentrations again. In the study of Rios et al. (2009b), DON concentration in semolina fraction show large reduction (by 47.8% and 78.4 respectively in two samples), regardless that nearly half of initial DON was left in this fraction (Table 7). Similarly, good mycotoxin reduction of mycotoxin concentrations was also showed in the fine milled fractions in other studies (Lancova et al., 2008; Visconti et al., 2004). More rigorously, Tibola et al. (2015) studied samples with different initial DON levels, and found that mycotoxin reduction in fine milled wheat fractions decreased with the grains’ initial contaminating level increasing. In this study, untreated samples were classified, based on different levels of initial DON contamination (A:498.8 mg kg−1, B: 746.5 mg kg−1, C: 1225.0 mg kg−1, D: 2747.5 mg kg−1 and E: 5985.0 mg kg−1). The results of this study showed that DON concentrations in finished flour decreased with these initial concentrations increased (42.1%, 33.6%, 35.8%, 21.5% and 10.4%) (Table 8). In addition, low levels of ZEN co-occurred in sample D and E in this study, and milling seemed efficient enough to eliminate them in finished flours.

3.4. Summary of physical removal methods In this part, mycotoxin reducing effects of several feed manufacturing processes (cleaning, sorting, dehulling and milling) have been introduced. In all these processes, mycotoxins are mainly decreased due to the physical separation rather than chemical and biological deactivation, and thus they are classified as ‘physical removal methods’. Cleaning and sorting is first line methodology to remove contaminants, including mycotoxins in grain samples. In the reviewed studies, hand sorting showed highest mycotoxin reducing effects since 70%–99% of different mycotoxin types were reduced (Matumba et al., 2015; van der Westhuizen et al., 2011). Liquid flotation methods also showed good mycotoxin reducing effects, especially when an optimal solvent is supplied (Matumba et al., 2015). However, these two methods are rarely applied in practical manufacturing situations because of high extra costs. In comparison, reducing effects of cleaning machines (e.g. air separator, sieving machine) are various. Some studies declared to observe high DON reduction (Tibola et al., 2016), while some others observed low DON reducing effects (Visconti et al., 2004). High variation might be easier occur when targeted mycotoxin types usually occur with relatively low concentrations, like T-2/HT-2 and 3AcDON + 15-AcDON (Schwake-Anduschus et al., 2010; Matumba et al., 2015). Mass loss during cleaning and sorting is another highly concerned topic in manufacturers’ eyes, since high mass loss might imply high financial loss. Most of the reviewed studies observed mass loss lower than 8%, but high mass losses were also observed. For example, mass loss of hand sorting method studied by Matumba et al. (2015) was 14%, and the study of Tibola et al. (2016) observed up to 33% mass lost. The high variation in hand sorting methods’ decontaminating effects could be due to the inconsistent subjective judgment of workers. Tibola et al. (2016) also declared it difficult to explain the high mass loss in their results, since it was impossible to record accurate mass balance of each step 138

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

Table 8 Average DON and ZON concentrations in unprocessed wheat and milled fractions, and the change rate of mycotoxin concentrations in milled fractions compared to unprocessed wheat (Tibola et al., 2015). Mill fraction

Yield (%)

DON

ZEN −1

Concentration (mg kg

)

Concentration (mg kg−1)

Change rate (%)

Change rate (%)

−1a

Sample A: initial DON concentration before milling: 498.8 ± 29.3 mg kg Finished flour 43.9 288.7 ± 4.2a Bran 626.3 ± 207.0a

−42.1b +25.6b

Sample B: initial DON concentration before milling: 746.5 ± 70.7 mg kg−1a Finished flour 42.0 495.8 ± 36.1a −33.6b Bran 870.0 ± 175.2a +16.5b Sample C: initial DON concentration before milling: 1225.0 ± 90.0 mg kg−1a Finished flour 43.2 787.0 ± 36.1a −35.8b Bran 1752.5 ± 234.8a +43.1b Sample D: initial DON concentration before milling: 2747.5 ± 140.8 mg kg−1a (with 27.6 ± 1.8 mg kg−1a ZEN co-occurring) Finished flour 42.6 2157.5 ± 79.3 −21.5b NDc Bran 3415.0 ± 212.1 +24.3b 62.6 ± 42.8a −1a

Sample E: initial DON concentration before milling: 5985.0 ± 260.6 mg kg Finished flour 40.1 5360.0 ± 331.4a Bran 7407.5 ± 534.6a

+127.2b

−1a

(with 38.7 ± 6.5 mg kg ZEN co-occurring) −10.4b NDc +23.8b 61.2 ± 158.2a

+58.2b

a

Mean ± Standard Deviation. Change rate indicates the mycotoxin concentration in milled fractions compared to that in initial unprocessed samples. ‘+’ represents the increase of concentration, and ‘−’ represents the decrease. c ND: Below the limit of quantification (< 20 mg kg−1). b

in a continuous processing system. Dehulling and milling processes also showed good mycotoxin reducing effects in reviewed studies. However, these two methodologies have an obvious limitation: their mycotoxin reducing effects must give way to their main functions for feed manufacturing. One evident example is the relationship between dehulling processing duration, remaining mass and remaining mycotoxin (House et al., 2003; Rios et al., 2009b). Longer processing duration could lead to lower remaining mycotoxin, but processing duration is always set stable in practical manufacturing situations in respect to production standards. On the other hand, milling process was effective to reduce mycotoxin concentrations in fine mill fractions (e.g. semolina), but this does not mean that mycotoxin hazards could be ignored in other mill fractions (e.g. break flours and reducing flours), which are commonly utilised for feed manufacturing. In the study of Rios et al. (2009b), these mill fractions even contained higher mycotoxin concentrations compared to the raw samples. Meanwhile, increased initial contaminating levels of grains also limit the reducing rate of these physical removal methods (Tibola et al., 2015), so rejecting highly contaminated samples is a logical choice for feed manufacturers. 4. Thermal processes for mycotoxin reduction Thermal processing has been widely utilized in feed manufacturing. It helps to increase feed quality by deactivating some microbes like Salmonella or transforming kernels’ physicochemical properties. In addition, anti-nutritional factors in cereals and legume seeds can be significantly inactivated through thermal process (Qin et al., 1996). In this part, mycotoxin-reducing abilities of several thermal processes are introduced. 4.1. Dry heating Many mycotoxins are thermally resistant. For example, decomposing temperature of aflatoxins can be as high as 280 °C, and other mycotoxins also have high thermal resistance (Table 9). Nevertheless, mycotoxin reducing effects of dry heating process were observed in some studies. These effects are highly influenced by different factors, including processing duration, temperature and form of processed grain samples. Yumbe-Guevara et al. (2003) studied the reduction of DON, NIV and ZEN in barley through superheating. Table 9 Decomposing temperatures of some mycotoxins (Kabak, 2009). Mycotoxins

Decomposing temperature (∘C)

Aflatoxins Ochratoxin A DON ZON Fumonisins

268–269 169 151–153 164–165 100–120

139

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

Fig. 4. Effects of heating time and temperature on DON, NIV and ZEN reduction in: A) Standard solution; B) Naturally contaminated barley powder; C) Naturally contaminated barley kernels. ○ 140 °C; ◆ 150 °C; ■ 160 °C; ▲ 180 °C; ◇ 200 °C; ● 220 °C (Yumbe-Guevara et al., 2003).

In this study, different processing temperatures and sample forms (standard solution, barley powder and barley kernels) were prepared. As shown in the results, mycotoxins in some treatment groups were quite resistant, but reducing effects of superheating also occurred to other groups and showed positive association with both temperature and processing time (Fig. 4). Moreover, this study supported that sample particle size distribution influences the mycotoxin reducing effects of heating, since the barley powder in this study achieved better mycotoxin reduction compared to the whole kernel samples (Fig. 4). 4.2. Superheated steam Superheated steam is another effective thermal technology in feed/food manufacturing. Compared to conventional thermal processing, superheated steam could bring less oxidation loss and higher energy efficacy to the manufacturing (Anto et al., 2014). An effect of this technology is to bring high friability to kernels, which saves the time and energy in subsequent milling process. What should be pointed out is that superheated steam technology may have drying effects to the samples even though high moisture is involved. For example, prolongation of processing time at high temperature could dry the materials, while at low temperature it might bring higher moisture to the samples (Pronyk et al., 2006). Besides this, depending on processing conditions (temperature and moisture), starch will gelatinise. Pronyk et al. (2006) studied the DON reducing effects in wheat kernels by using a superheated steam process. The results of this study showed that the detoxifying effects of superheated steaming were positively associated with the processing duration and steam temperature (experimental range: 110–185 °C). Speed of steam is a crucial interfering factor specifically for superheated steaming. Fig. 5 was quoted from this study and clearly illustrates that linear results were hardly achieved in the reducing effects in DON contaminated wheat when these factors (processing duration, steam temperature and speed) were all integrated. For example, high steam temperature (185 °C) with sufficient processing time can reduce DON concentration by up to 50%, while no significant reduction effect was brought by most of the lower temperature groups (< 135 °C). So far, the reducing effects of superheated steam on other mycotoxins, such as AFL, ZEN etc., are still in need. 4.3. Extrusion cooking Extrusion cooking is a process that combines high-speed shearing and superheated steaming within a short time. This processing step is usually applied to produce snack foods for humans, or pet food and fish diets. During this process, feed or food ingredients successively experience physical powdering, super heating forming and cooling. Thus, several physical and chemical transformations occur to the processed samples, including starch gelatinization, protein denaturation and microbial count reduction (Castells et al., 2005). Several studies on mycotoxin reducing effects of extrusion cooking observed in samples with different initial contaminating levels were published (Castells et al., 2006a,b) and targeted mycotoxins cover aflatoxins, fumonisins, DON and ZEN etc. (Table 10). Higher processing temperatures during extrusion lead to a higher reduction of mycotoxins (Buser and Abbas, 2002; Katta et al., 1999; Table 11; Figs. 6 and 7) but temperatures of 160° product temperature may have a negative effect on protein quality after extrusion. With many influencing factors involved, the effects of extrusion detoxification indeed are less constant. For example, a study by Castells et al. (2006a) showed that higher aflatoxin concentration unexpectedly occurred in the rice meal sample which was 140

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

Fig. 5. DON concentration in wheat samples treated by superheated steam respectively at: a) 185 °C, b) 160 °C, c) 135 °C, d) 110 °C. Two steam velocities were supplied: 1.3 m/s and 0.65 m/s (Pronyk et al., 2006).

processed at higher temperature, even when other variables (residence time and initial moisture) were controlled (Fig. 8). Extrusion cooking is a promising process for mycotoxin reduction, since it combines different processing methodologies including shearing, making available the inner part of the grains, and heating (Bullerman and Bianchini, 2007). Moreover, the radicals formed during the extrusion cooking (Ullsten et al., 2006; Schaich, 2002; Schaich and Rebello, 1999) may also have an effect on the breakdown of mycotoxins. 4.4. Irradiation Irradiation technologies are applied in feed and food industry usually because of their high efficiency to eliminate the microorganisms and other potential pathogens infecting the grains. Irradiation contains two types: non-ionizing radiation and ionizing radiation. The non-ionizing radiation has a long wavelength and does not contain enough energy to ionize the atoms or molecules. Radio, microwave and infrared ray are heat producing examples of non-ionizing radiation. Ionizing radiation is a by-product when unstable atoms try to reach stability by giving out energy or mass or both, and it has short wavelength and high energy. With the concern of radioactive contamination, only selected types of ionizing radiations are legally applicable in food or feed manufacturing with detailed regulations in Europe, like gamma-radiation, X-rays and electron beams (EC, 1999a; EC, 1999b). Both heating and hydrolysis are the main mechanisms involved with the mycotoxin reduction by irradiation. Heating effects of irradiation methods were directly provided by high doses and long exposing time, and thus mycotoxin reducing effects of irradiation are directly affected by these factors. A study performed by Herzallah et al. (2008) is quite informative. In this study, the non-ionizing treatments (microwave and sunlight), which provides lower energy, showed much less efficient to reduce mycotoxins in chicken feed samples, compared to gamma ray. Also, increase of exposing time of each treatment showed better reducing effects in the same study 141

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

Table 10 Publications on the mycotoxin decontamination effects of extrusion process (Not all authors provided the duration of process). Mycotoxin

Sample

Type of extruder

Additives

Reduction (%)

FB1

FB2 AFB1

Corn flour Corn flourb Alkali-cooked corn flourc Corn flourb Corn flourd

Single-screw Slot hot extruder Twin-screw Slot hot extruder Single-screw

AFM1

Corn flourd

Single-screw

DON

Corn floure

Not reported

66–92 26 0–99 32 45 74 85 63 64 72 20 51 82 61 74 73 95–99.5

ZEN OTA

Wheat kernelsf Corn gritsg Wheat mealh Barley meali

Twin-screw Twin-screw Twin-screw Single-screw

None None None None Control 0.3% Ca(OH)2 0.5% Ca(OH)2 0.75% H2O2 1.5% H2O2 3% H2O2 Control 0.3% Ca(OH)2 0.5% Ca(OH)2 0.75% H2O2 1.5% H2O2 3% H2O2 Control Na2S2O5 NaHSO3 None None None

a

42–92 66–83 8.3–42.2 17.5–86.5

a

Piñeiro et al. (1999). De Girolamo et al. (2001). c Cortez-Rocha et al. (2002). d Elias-Orozco et al. (2002). e Cazzaniga et al. (2001). f Accerbi et al. (1999). g Ryu et al. (1999). h Scudamore et al. (2004). i Castells et al. (2006b). b

Table 11 Effects of different processing temperature (°C) during extrusion on aflatoxin reduction (%)a in cottonseed (Buser and Abbas, 2002). Processing temperature (°C)

Remaining aflatoxin (ppb)

Reduction (%)

104 132 160

182 ± 44 137 ± 34 117 ± 21

72.0 ± 6.7 78.9 ± 5.2 82.0 ± 3.2

a

Initial concentration: 650 ppb.

(Table 12). On the other hand, the hydrolysis effect of irradiation, where free radicals appear to reduce mycotoxins, occurs more easily in aqueous conditions (He et al., 2010). For example, O’Neill et al. (1993) observed that DON and its derivative 3-A-DON in dry maize kernels showed high stability even at 50 kGy (Gy is the SI unit of absorbed radiation) of gamma irradiation, while 5 kGy was sufficient to drastically reduce these toxins in the aqueous solution (Fig. 9). In another study (Stepanik et al., 2007), evident difference of DON reducing effects between wet distillers grains (wet condition, Fig. 10), distillers solubles (wet condition, Fig. 11) and distillers dried grain and solubles (dry condition, Fig. 12). 4.5. Summary of thermal methods for mycotoxin reducing Mycotoxin reducing effects of four thermal methodologies were introduced in this part. Apart from high temperature, some other different functions involved could contribute to mycotoxin reduction, like hydrolysis in irradiation technologies and kernel deformation in extrusion cooking. Like physical removal methods, thermal methods should have little specificity for mycotoxins, but the practical reduction could still be affected by different thermal sensitiveness between mycotoxins. For example, aflatoxin B1 has relatively high sensitiveness to gamma irradiation, and is thus easier to eliminate, compared to other mycotoxins (O’Neill et al., 1993). Some disadvantages of irradiation technologies could limit their promotion. First, extra costs conducted by manufacturers and the public concern on chemical safety of ionizing irradiation are still limiting their application in food/feed manufacturing (Calado et al., 2014). Second, negative nutritional changes could result from the irradiation treatments. For example, Ghanem et al. (2008) observed a significant drop of oil content in the crop samples (including corn, pistachios, and peanuts) after gamma irradiation degraded existing mycotoxins, and the oil loss increased with higher gamma doses. Unsaturated bonds in fat can also be attacked by 142

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

Fig. 6. Fumonisin B1 reduction (%) in extruded corn cereal, with the presence of salt (%), barley malt (%) and sucrose (%). S = salt, M = malt (Castells et al., 2009).

Fig. 7. Effect of temperature during extrusion cooking on fumonisin B1 concentration in corn grit samples with different initial FB1 concentrations (Katta et al., 1999).

irradiation, which leads to changes in taste (D’Ovidio et al., 2007). Some vitamins (especially vitamin C, E, K and thiamin) and proteins are also sensitive to irradiation (D’Ovidio et al., 2007). Third, whether mycotoxin ‘reduction’ is real ‘detoxification’ is still doubtful, since thermal treatments could transfer mycotoxins into some undetectable derivatives with unknown toxicities. In conclusion, thermal technologies are not that ideal for mycotoxin reduction in feed manufacturing. 5. Chemical methods for mycotoxin reduction Some chemicals are effective to reduce the mycotoxins. Generally, chemical methods have some advantages in mycotoxin reduction, including high reducing efficiency and relatively low cost. Different mechanisms are involved with the mycotoxin reducing effects of these chemical methods, including alkalization, oxidation, reduction, hydrolysis, hydration and conjugation (He et al., 2010). However, most of these chemical methods have been concerned due to consumers’ health. Currently, application of chemical treatments for mycotoxin reduction has been banned in feed/food materials by European Commission (2006b). In this review, only two main chemicals (ammonia and ozone application) are discussed. One typical chemical with mycotoxin reducing effects is ammonia. The other chemical is ozone, which is famous for the super oxidizing ability. In practice, both products are in either liquid or gas form, and could be applied because of their excellent effects to eliminate zoonotic bacteria in feedstuffs (Tajkarimi et al., 2008; McKenzie et al., 1998). 143

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

Fig. 8. Effects of extrusion processing with different processing temperature (°C), residence time (s) and initial moisture (A and B) on aflatoxin B1 reduction in rice meal (Castells et al., 2006a). Five bars of different colours represent different residence times (second); ‘A’ represents the samples initially with 24% moisture, and ‘B’ represents the samples initially with 27% moisture.

Table 12 Effects of different irradiation types on AB1 reduction. Sample Wheat bran

a

Irradiation

Dose (kGy)

60

4 6 10 4 6 10 4 6 10 4 6 10 5 5 10 15

Co gamma ray

Corna

Barleya

Peanuta

Coconut agarb Chick feedc

Electron beam 60 Co gamma ray

Microwave

Sunlight

a b c

Duration

Reduction (%)

4 min 8 min 10 min 3h 9h 30 h

52.9 75.5 86 31.2 72.1 84.2 45.0 66.2 89.9 8 28.5 58 75.5 6.3–10.5 12.8–21.9 18.3–37.3 3.6–14.9 7.1–28.6 21.2–33.3 39.8–42.3 49.9–56.4 60.4–75.5

Ghanem et al. (2008). Rogovschi et al. (2009). Herzallah et al. (2008).

These two chemicals were also applied for mycotoxin decontamination. One in vitro study (Young et al., 2006) found that aqueous ozone is highly effective to reduce pure trichothecene mycotoxins (including DON, 3ADON, 15ADON, T2, HT2 etc.) under laboratory conditions, and the reducing effects were positively related with the applied amount of ozone, which is related to both the concentration of ozone and the exposing time. In situ mycotoxin reducing effects of these chemicals were also supported by studies, and their decontaminating effect can be influenced by many factors. First, they can be influenced by the initial mycotoxin contaminating degree. In the study of Savi et al. (2014), ozone successfully reduced DON level in wheat pericarp by 86.6% within 30 min, but was

144

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

Fig. 9. Different degradation effects on DON by gamma irradiation in dry maize or aqueous solution (O’Neill et al., 1993).

Fig. 10. Effect of electron beam with different doses on DON concentrations in wet distillers grains (Stepanik et al., 2007).

Fig. 11. Effect of electron beam with different doses on DON concentrations in distillers solubles (Stepanik et al., 2007).

much less efficient to reduce the DON in wheat endosperm. Since endosperm originates from the inner parts of grains, the rich lipid in that parts could be a reason to grasp this lipo-soluble toxin firmly (Table 13). On the other hand, levels of heat and moisture during the chemical treatment may be positively associated with the effects of 145

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

Fig. 12. Effect of electron beam with different doses on DON concentrations in distillers dried grain and solubles (Stepanik et al., 2007).

Table 13 Effects of ozone and ammonia treatments on mycotoxin reduction. Mycotoxins

Samples

Chemicals

Duration (min)

DONa

Wheat pericarp

Ozoned

Wheat endosperm

Ammoniae

Ground corn

Ammoniaf

30 60 120 30 60 120 60

Aflatoxin B1b

Moisturec (%)

12 16 12 12

60

Temperature (°C)

Reduction (%)

40 40 25 121

86.6 86.6 100 20.8 34.0 100 64.9 93.1 17.5 99.8

a

Savi et al. (2014). Weng et al. (1994). c Initial moisture of samples. d In the gaseous form. e Gaseous NH3 (%2). f Aqueous NH4OH (2%). b

these treatments. For example, In the study of Weng et al. (1994), higher initial moisture in ground corn largely increase the reduction of aflatoxin B1 by aqueous NH4OH. In the same study, aqueous NH4OH could decrease aflatoxin B1 by 99.8%, while only 17.5% aflatoxin B1 was eliminated at 25 °C (Table 13). Third, chemicals show different detoxifying effects when targeting different mycotoxins. For example, ammonia (either gas or liquid) shows good effects to reduce aflatoxins in samples, but is less effective to deal with Fumonisins and DON (Jouany, 2007). Feeding experiment also supported the detoxifying effects of these chemical treatments. An in vivo study on broiler chicks by Allameh et al. (2005) observed that the adverse effects of aflatoxin-contaminated maize could be controlled after the ammonia treatment (vapours with 1% ammonia), and the growth performance, organ conditions and the serum biochemical parameters of chicks fed with the treated diet were comparable to the control group which was fed with uncontaminated grains. The compensating effects of ozone were also observed in another in vivo study, where turkeys were fed with aflatoxin-contaminated corn (McKenzie et al., 1998). These chemical treatments for decontaminating mycotoxins also have their limitations. First, the toxicities of the mycotoxin metabolites resulting from these chemical treatments are still threatening, since, for example, some intermediate compounds originating from aflatoxin B1 treated by ozone or ammonia, including aflatoxin D1 and aflatoxin ozonide, could still have toxicity (Tiwari et al., 2010; McKenzie et al., 1998) (Fig. 13). Second, these chemical treatments may change the quality of materials. These changes include surface oxidation, colour changes, impaired taste, reduced germination energy, and extra energy could be required in following milling processing (Tiwari et al., 2010). In the study of Savi et al. (2014), germination capacity of wheat grains was reduced by 12.5% after 180 min of ozone exposure compared to the untreated grains, but no similar impairment was observed in the samples whose exposure lasted for 120 min. Since chemical treatments for mycotoxin reduction in food or feed materials have been banned, studies on chemicals’ mycotoxin reducing effects were gradually less published in this century (He et al., 2010). 6. Biotransformation of mycotoxins Apart from the inorganic or organic binding agents introduced above, mycotoxin controlling ability of some microbes was also noticed. These methods use biologically transformation to reduce or eliminate mycotoxins as reviewed by He et al. (2010) and 146

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

Fig. 13. General chemical transformation flow chart of aflatoxin B1 treated by ammonia or ozone (McKenzie et al., 1998).

Vanhoutte et al. (2016). He et al. (2010) summarize the following advantages of biotransformation: – – – –

specificity towards less or non-toxic ingredients mild reaction conditions (temperature; pH) application under aerobic and anaerobic conditions potential application of detoxifying enzymes

Some microbes can turn mycotoxins into less toxic or non-toxic derivatives by transforming their physical and chemical characteristic. These materials were usually isolated from natural sources, like crop tissues, soil, and in vivo microbial (Heinl et al., 2010). One example is the reductive de-epoxidation of trichothecenes of microbes in the hindgut of pigs and poultry. In this process, the transformation of epoxide groups, which determines the toxicity of trichothecenes, leads to less toxic products (Wu et al., 2007). These natural detoxifying microbes occurring in animal GI tracts seem to be adaptive. For example, DON or NIV could not be detoxified in animals which have never been exposed to trichothecenes before (Karlovsky, 2011). However, these microbes may function well only when their strict environment requirements are met. For example, the bacteria that can detoxify trichothecenes by de-epoxidating are observed to be active only under anaerobic environment (Karlovsky, 2011). Some enzymes involved in detoxification need NADPH (Hassan et al., 2017) or other co-factors which limit their practical application as an enzyme solution. This situation makes the application of these microbes as animal feed additive quite complicated. Thus, biological feed additives can represent alternative detoxification methods but are still in their infancy and need breakthroughs. 7. Mycotoxin-binding by feed additives In 2009, the European Commission extended their definition of ‘feed additives’ to include materials that ‘suppress or reduce the absorption of mycotoxins, promote the excretion of mycotoxins or modify their mode of action and thereby mitigate possible adverse effects of mycotoxins on animal health’ (EC, 2009). Academic studies on feed additives to reduce mycotoxins in feed were performed long time ago. Unlike the physical removal, thermal or chemical methods which focus on how to reduce mycotoxins in the ingredients during the feed manufacturing stage, these feed additives were designed to reduce the potential adverse effects of mycotoxins after the feed is ingested by animals. Main mycotoxin-restricting mechanisms involved with these additives include: 1) physically binding the mycotoxins and thus decreasing the gastrointestinal absorption of mycotoxins; 2) inactivating mycotoxins; 3) modifying animals’ enzymes to transform mycotoxins into some less toxic metabolites or improve animals’ defence against mycotoxin (Meissonnier et al., 2009). Below, mycotoxin restricting effects of several feed additive products are introduced. 7.1. Inorganic binding agents Binding agents are a group of feed additives which function to block the absorption of mycotoxins in gastrointestinal tracts of animals by physically binding these toxins. Main examples of inorganic binders for mycotoxins include hydrated sodium calcium aluminosilicates (HSCAS), activated coat (AC), montmorillonite clays (MC), etc. (Avantaggiato et al., 2005). Absorbing effects of these inorganic binding agents are based on the physical and chemical structures of both the agents and the mycotoxins, and these binding agents usually have their own specific repertoire of target mycotoxins. For example, AFL, which contain polar functioning 147

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

Table 14 Body weight, feed conversion rate (FCR) and mortality of chicks1,2,3 from 0 to 6 weeks old, provided with 4 different diets (Bailey et al., 2006). Diet

Weight (g)

FCR

Mortality (%)

Phase 1: week 1–3 Aflatoxin (3600 ppb) Aflatoxin (3600 ppb) + MC Control Control + MC

330.1d 338.7c 732.6a 701.3b

2.05a 1.95a 1.33b 1.33b

16.5a 13.0a 3.5b 6.0b

Phase 2: week 4–5 Aflatoxin (3600 ppb) Aflatoxin (3600 ppb) + MC Control Control + MC

566.4c 759.8b 1134.9a 1028.1a

1.91a 1.72b 1.66b 1.68b

15.2a 6.9b 1.6c 0.5c

318.9c 424.5b 550.3a 592.8a

2.32a 2.18ab 2.08ab 1.98b

9.2a 4.2b 0.5c 1.1bc

1215.5c 1565.0b 2417.8a 2376.2a

2.01a 1.89b 1.65c 1.65c

36.0a 22.5b 5.5c 7.5c

Phase 3: week 6 Aflatoxin (3600 ppb) Aflatoxin (3600 ppb) + MC Control Control + MC Cumulative for 6 weeks Aflatoxin (3600 ppb) Aflatoxin (3600 ppb) + MC Control Control + MC a–d

Values with a different superscript within a column differ significantly (P < 0.05). All values represent the average of 10 pens. 2 MC = montmorillonite clays. 3 The added MC is 0.5% of the total feed. 1

groups, are easier to be absorbed by MC than the non-polar ZEN and OTA (Kabak et al., 2006). Among these inorganic binders, bentonite (an absorbent aluminium phyllosilicate clay) receives the support from European Commission (2013, No 1060/2013) as an effective feed additive to bind mycotoixns, especially aflatoxins. An ideal mycotoxin binding agent should: 1) be harmless to the animal health and productive performance, 2) rapidly produce stable detoxified mycotoxin-absorbent complex before mycotoxins enter the blood, 3) be effective to different mycotoxin concentrates under various environments of the entire GI tract, 4) be environment-friendly after excreted out of animal bodies (Jard et al., 2011). Roughly, these agents can be divided into two groups, inorganic and organic, based on their physicochemical characteristics. A long-term study (Bailey et al., 2006) on the protecting effects of dietary calcium montmorillonite clays (MC) was performed on chicks for the entire 6 weeks after hatchery (first 3 weeks as ‘new-born’, then 2 weeks as ‘grower’ and last 1 week as ‘finisher’) (Table 14). Four diet types were set: A) diet contaminated with 3600 ppb of aflatoxin; B) diet contaminated with 3600 ppb of aflatoxin, and containing 0.5% MC; C) control diet without aflatoxin contamination; D) diet without aflatoxin contamination, but containing 0.5% MC. In this experiment, both the health statue and the growth performance were analysed through recording several parameters, including mortality, feed intake, body weight gain, and weights of some organs. In the comparison of groups offered with diet A and B, MC showed limited effects to reduce the mortality of new-born chickens poisoned by aflatoxin, but the weight in group B was still higher than that of group A, which indicated the compensating effects of MC. Compared to the youngest chicks, the compensating effects on the weight gain and mortality reduction became greater in older chicks. On the other hand, the MC diet seemed to reduce the weight gain of new-born chicks, even though this adverse effect became less evident in the older chicks. Moreover, the drastically increased weight of (both A and B) finisher’s liver, kidney and spleen indicated that MC failed to protect these organs from the high level of aflatoxin (much higher than EU requirement, Table 2) in this experiment (Table 15). In similar studies, activated coat failed to prevent weaning pigs from the poisoning of fumonisin B1 (Piva et al., 2005), and the diatomaceous earth (DE) also failed for boilers fed aflatoxin contaminated diets (Modirsanei et al., 2008). Results of these experiments indicate that the effects of inorganic binders are limited.

Table 15 Relative weights of organs in bodies of sacrificed finisher chicks (Bailey et al., 2006). Relative weight (%)

Aflatoxin

Aflatoxin + MC

Control

Control + MC

Liver Kidney Spleen

4.56a 1.50a 0.22a

4.00b 1.11b 0.2 a

1.90c 0.56c 0.12b

1.96c 0.60c 0.12b

a–c

values with a different superscript within a row differ significantly (P < 0.05).

148

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

Table 16 Aflatoxin B1 binding abilities of strains of yeasta and lactic acid bacteriab (Shetty and Jespersen, 2006). Isolates

Quantity of mycotoxin binding strains Percentage of binding < 15

15–39

40–59

≥60

Yeasta

Saccharomyces cerevisiae Candida krusei Saccharomycodes ludwigii

1 4 0

8 5 0

3 1 1

3 1 0

Lactic acid bacteriab

Lactobacillus plantarum Lactobacillus fermentum Selangorensis

0 0 4

0 0 1

4 0 0

1 1 0

a b

For all the yeast strains, 108 cells were incubated in 1 ml of PBS containing 5 mg of aflatoxin B1 for 72 h at 25 °C. For all the lactic acid bacteria strains, 109 cells were incubated in 1 ml of PBS containing 5 mg of aflatoxin B1 for 72 h at 25 °C.

7.2. Organic ‘binding agents’ Some organic materials also work effectively to ‘bind’ mycotoxins. One of the earliest organic binders dealing with mycotoxins might be the extractions from some fibre-rich plants (e.g. oat and alfalfa) (Jouany, 2007). Currently, the most noticed organic binder now is the yeast cell wall. Even though earlier researchers had noticed that live yeast Saccharomyces cerevisiae can effectively suppress adverse effects of some mycotoxins (e.g. aflatoxins) in animals (e.g. chicks) (Stanley et al., 1993), the earliest in vitro study which found the mycotoxin binding function of cell wall of yeast Saccharomyces cerevisiae was performed by Devegowda and colleagues in 1998, as reviewed by Jouany (2007). Shetty and Jespersen (2006) performed a laboratory study on the aflatoxin B1 binding abilities of isolates from different yeast or lactic acid bacteria strains (Table 16). In their study, one group of isolates from western African maize, Saccharomyces cerevisiae were observed with best effect to bind over 60% of aflatoxin, but other strains performed much less effective. Apart from the simple physical absorbing effects, some other mycotoxin controlling mechanisms are also involved with these organic ‘binders’. For example, major fractions of dry yeast cell wall are two polysaccharides: β-D-glucan and α-D-mannan, and the mycotoxin binding function is directly involved with β-D-glucan (Kogan and Kocher, 2007; Jouany, 2007). Apart from binding mycotoxins, these polysaccharides also supply other effects to control mycotoxins’ harm in animal bodies, including modulating immune activities and binding gastrointestinal bacterial pathogens (Kogan and Kocher, 2007). Meissonnier et al. (2009) specifically studied the immuno-modulating effects of glucomannan of dry yeast cell wall. In this study, weaned pigs were exposed to different levels of aflatoxin B1 or T-2 for 28 days. The glucomannan added for some diet groups evidently decreased the severity of liver damages and delayed the impairing of immune reactions. Main mechanisms could be that dietary glucomannan stirred the proliferation of immune enzymes and immunoglobulin G, even though these protective effects might be relatively limited.

7.3. Summary With the banning (EC, 2006a,b) of chemical mycotoxin-decontaminating methods in feed manufacturing, safer methodology for mycotoxin controlling in feed are needed in recent decades. Mycotoxin binding agents are gradually noticed in this context. Different from other methodologies, binding agents mainly function when mycotoxin contaminated diets are ingested by animals. Thus, the in vivo effects of these additives are crucial. These binding agents are roughly classified as ‘inorganic’ and ‘organic’ ones. Inorganic binders were found to have effects to compensate the impacts of mycotoxins in animal diets, and they have been widely noticed and partly applied by local farmers because of their low costs and high operability. Organic binding agents may have other positive functions, like binding other pathogens and improving animal’s immunity, which indirectly help to control mycotoxin hazards. These feed additives also have some disadvantages. Generally, efficacy of the inorganic products is usually low. With both low costs and low efficacy, these inorganic binders are sometimes added into feed at large amount by farmers, which may decrease the total nutritional values of feed (Karlovsky, 2011). In addition, they could bind other nutritional factors like vitamins and trace elements, bringing nutritional imbalance to animals. Some binders, like clay, show low toxicity in animals. They might also contain other negative effects like interrupting nutrient metabolisms, and impairing animal organs. Relatively low natural degradation of inorganic binders also leads to ecological problems after they are excreted out of animals. Even though the organic feed additives could exclude the disadvantages of the inorganic ones, highly varied inner environments (e.g. pH) of animal GI tracts can greatly affect the efficacy of these biological treatments (Karlovsky, 2011). In conclusion, feed additives for mycotoxin detoxifying through binding are useful in current animal feed practice but still less from ideal.

149

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

8. Discussion 8.1. Limitations of current strategies Despite these strategies have been observed to be effective for mycotoxin controlling, several problems and limitations of some of them are still need overcoming. First, an evident limitation is the heterogeneity observed in different studies. One type of heterogeneity is the experimental conditions. For example, several studies noticed that some experimental conditions could affect the mycotoxin reducing effects of extrusion cooking, including type of processing machine (Castells et al., 2005), crop breeds and places of origin (Schwake-Anduschus et al., 2010; Tibola et al., 2016), presence of special ingredients (e.g. sugar, salt) during processing (Castells et al., 2009). Nevertheless, most studies reviewed usually neglected to emphasize the potential impacts of these variables. The other heterogeneity is the initial contaminating levels of feedstuffs. Mycotoxin contamination mainly accumulates on the surface or shallow parts of a single kernel, but contamination may also occur in deeper parts when the level of fungal activity rises. Take Fusarium graminearum, when this mycotoxigenic mould infects cereal grains, the brush hair at the kernel’s tip is the most rapid entrance, which is followed by quite a slow development into deeper parts, and the endosperm is the eventual target (Skadsen and Hohn, 2004). Among the reviewed studies, some on the mycotoxin redistributing effects of milling noticed the impact of different levels of initial contamination and grant it as an experimental variable (Rios et al., 2009b; Tibola et al., 2015), while others did not focus on this variable. The second limitation is the unclear toxicity of mycotoxin derivatives. One typical problem is the ‘masked mycotoxin’. Generally, this term ‘masked mycotoxin’ covers those mycotoxin derivatives which are undetectable by some conventional analytical methods, because of their changed structure compared to the parent toxins (Berthiller et al., 2013). Masked mycotoxins can result from feed/ food processing. For example, amount of deoxynivalenol-3-D-glucopyranoside (D3G) increases after malting of barley grains (Berthiller et al., 2013). Also, thermal treatments could transfer DON into masked mycotoxins, like D3G (Kostelanska et al., 2011). On the other hand, one mycotoxin can be transferred into various ‘masked mycotoxins’, which increases the uncertainty of food or feed safety. For example, through conjugation, DON occurring in some crops can be transferred into less toxic D3G, while through acetylation DON can be turned into 3-acetyl-DON (3ADON) and 15-acetyl-DON (15ADON) (Berthiller et al., 2013). Till now, more and more ‘masked mycotoxins’, originating from DON, ZEN, T-2, HT-2, fumonisins, OTA, etc., have been gradually discovered (reviewed by Berthiller et al., 2013). This issue of the unknown toxicological information brings some warnings to the current knowledge of mycotoxin reduction. First, the mycotoxin reducing effects of some methodologies should be examined again, especially thermal and chemical methods, which could produce the toxic derivatives from common mycotoxins. Second, current legal systems on food/feed safety might not take the new toxicological development into account. For example, α-zearalenol (α-ZEL), a much less detectable derivate of ZEN, can result in more severe estrogenic effect than its precursor, but no current regulation has been proposed to control its presence in food or feed processing (Urraca et al., 2004). The final problem discussed is the limited applicability of some reviewed mycotoxin controlling methods in practical feed manufacturing. Typical examples include the banning of the chemical methods and ionizing irradiation, which shows distinguishable mycotoxin reducing effects, but raises public concern towards the safety of treated food/feed materials. Furthermore, costs of mycotoxin reducing strategies are always an important factor considered by manufacturers. High costs of some methods, like liquid flotation and hand sorting, directly restrict their applicability in large scale manufacturing situations. Advances in automation and digital processing could in the future help to make this a viable option. For those strategies that are applicable in feed manufacturing, their mycotoxin reducing effects may interfere with the desired product quality. For example, the mycotoxin reducing effects of dehulling could be increased when longer duration takes away more outer parts of seeds or hulls, but duration of dehulling is set, based on the time requirements of production, a reason why mycotoxin reducing effects of dehulling could be marginalized. Due to similar logics, mass loss of materials is a crucial factor to judge the applicability of cleaning and sorting methods, since low mass loss could indicate low economic loss. For example, the high mass loss (> 28%) in one reviewed study could bring high extra costs and thus be less acceptable from the aspect of feed manufacturers, regardless of the high mycotoxin reducing effects (> 74%) (Table 4, Tibola et al., 2016). Detoxification methods by using biotransformation agents certainly have future possibilities on the condition they are well identified for their degradative ánd detoxification effects. This means that they have to be examined at the same time for the toxicity of their metabolites. Multiple mycotoxin degradation should have a priority in research, since agents claiming mycotoxin detoxification are not always effective against other, non-related mycotoxins (Vanhoutte et al., 2016). The same authors invite researchers to examine the biodegradation of mycotoxins by microorgansims. 8.2. Recommendations With the comprehensive consideration achieved through this review, several recommendations for feed manufacturers to control mycotoxin hazards could be proposed. First, physical removal strategies are the most recommended, because they contribute to reducing the existing fungi and mycotoxins of little specificity, and less tend to result in the synthesis compared to chemical and thermal methods. Among the physical methods, cleaning and sorting is necessary in most situations because of its function to eliminate not only mycotoxin but also other contaminants. In comparison, dehulling may be more narrowly applicable, since they can only function on limited range of seed species like legumes, and these seeds are relatively less contaminated by mycotoxins. As introduced above, the milling process is only to redistribute mycotoxins into different mill fractions. Even though this process could result in low mycotoxin concentrations in the 150

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

finest fractions, the mycotoxin hazards could be still shrouding other mill fractions like brans, which are often reutilized for animal feed products. Thus, potential mycotoxins in these by-products should be examined. Second, thermal methods generally show limited mycotoxin reducing effects, and is less recommended than physical removal methods. The costs of different thermal methods are highly diverse, which could be an important factor influencing the applicability of each thermal method in practical manufacturing. Cheap methods, like IR, seem less effective towards mycotoxins, while some methods showing evident mycotoxin controlling effects, like gamma irradiation, are costly. In addition, it is still doubtful whether the ‘decreased’ mycotoxins are transferred into undetectable toxic derivatives, like masked mycotoxins. Third, chemical treatments (not meaning enzymatic), without doubts, are banned by legislatives. In comparison, mycotoxin reducing effects of some feed additive products seem more promising, especially after some of these products were officially included as ‘feed additives’ (EC, 2013). Some in vitro studies supported their potential to detoxifying contaminated feedstuffs, but their actual functions in animal bodies still need more proof. Generally, these products’ mycotoxin controlling effects are generally low and inconsistent. It is also challenging for them to function well through the complicated inner environments of GI tracts with little harm to the animals (Le Thanh et al., 2016). The first results to detoxify DON in vivo are promising (Starkl et al., 2015). 9. Conclusions Among the different mycotoxin reducing methods, we should not declare that one individual method is unconditionally effective to eliminate mycotoxin contamination in crops or to prevent the consequent impacts on animal health. The prevention strategies and cleaning and sorting methods are widely applied because they work as first-line barriers to free the materials from various contaminants, including mycotoxins. However, mycotoxin reducing effects of other feed manufacturing technologies, like milling, dehulling and thermal methods, could be inconsistent and limited by various practical conditions. In addition, the evident reducing effects of some physical removal methods requires the costs of high grain mass loss, which could be a dilemma from the aspect of practical manufacturing. Feed additive products specifically against mycotoxins are promising but still in their babyhood, since in vitro functioning of some products are inconsistent, and their in vivo functioning need more proof. As for biotransformation, understanding the reaction for the detoxification of mycotoxins, attention in research on the toxicity of transformation by enzymes and their end-products (He et al., 2010; Vanhoutte et al., 2016) should emphasize the possibilities of the use of micro-organisms or enzymes towards safe properties in animal feed ingredients. Thus, this literature review generally agrees with the discussion of Jouany (2007): ‘Prevention strategies are essential, since it is impossible for other technologies to completely decontaminate the mycotoxin-contaminated feed samples once they are present’. Conflict of interest statement No conflict of interest. References Accerbi, M., Rinaldi, V.E.A., Ng, P.K.W., 1999. Utilization of highly deoxynivalenol contaminated wheat via extrusion processing. J. Food Prot. 62, 1485–1487. Alassane-Kpembi, I., Kolf-Clauw, M., Gauthier, T., Abrami, R., Abiola, F.A., Oswald, I.P., Puel, O., 2013. New insights into mycotoxin mixtures: the toxicity of low doses of type B trichothecenes on intestinal epithelial cells is synergistic. Toxicol. Appl. Pharmacol. 272, 191–198. Allameh, A., Safamehr, A., Mirhadi, S.A., Shivazad, M., Razzaghi-Abyaneh, M., Afshar-Naderi, A., 2005. Evaluation of biochemical and production parameters of broiler chicks fed ammonia treated aflatoxin contaminated maize grains. Anim. Feed Sci. Technol. 122, 289–301. Anto, A., Kiran, B.V., Jeevitha, G.C., Hebbar, U., 2014. Recent developments in superheated steam processing of foods—a review. Crit. Rev. Food Sci. Nutr. 56 (13), 2191–2208. Avantaggiato, G., Solfrizzo, M., Visconti, A., 2005. Recent advances on the use of adsorbent materials for detoxification of Fusarium mycotoxins. Food Addit. Contam. 22 (4), 379–388. Bailey, C.A., Latimer, G.W., Barr, A.C., Wigle, W.L., Haq, A.U., Balthrop, J.E., Kubena, L.F., 2006. Efficacy of montmorillonite clay (NovaSil PLUS) for protecting fullterm broilers from aflatoxicosis. J. Appl. Poult. Res. 15, 198–206. Bennett, J.W., Klich, M., 2003. Myotoxins. Clin. Microbiol. Rev. 16, 497–516. Berthiller, F., Crews, C., Dall’Asta, C., De Saeger, S., Haesaert, G., Karlovsky, P., Oswald, I.P., Seefelder, W., Speijers, G., Stroka, J., 2013. Masked mycotoxins: a review. Mol. Nutr. Food Res. 57, 165–186. Binder, E.M., 2007. Managing the risk of mycotoxins in modern feed production. Anim. Feed Sci. Technol. 133, 149–166. Birzele, B., Prange, A., KrÄmer, J., 2000. Deoxynivalenol and ochratoxin A in German wheat and changes of level in relation to storage parameters. Food Addit. Contam. 17, 1027–1035. Brandwagt, B.F., Mesbah, L.A., Takken, F.L.W., Laurent, P.L., Kneppers, T.J.A., Hille, J., Nijkamp, H.J.J., 2000. A longevity assurance gene homolog of tomato mediates resistance to Alternaria alternate f.sp.lycopersici toxins and fumonisin B1. Proc. Natl. Acad. Sci. U. S. A. 97, 4961–4966. Bryden, W.L., 2012. Mycotoxin contamination of the feed supply chain. Implications for animal productivity and feed security. Anim. Feed Sci. Technol. 173, 134–158. Bullerman, L.B., Bianchini, A., 2007. Stability of mycotoxins during food processing. Int. J. Food Microbiol. 119, 140–146. Buser, M.D., Abbas, H.K., 2002. Effects of extrusion temperature and dwell time on aflatoxin levels in cottonseed. J. Agric. Food Chem. 50, 2556–2559. Calado, T., Venancio, A., Abrunhosa, L., 2014. Irradiation for mold and mycotoxin control: a review. Compr. Rev. Food Sci. Food Saf. 13, 1049–1061. Castells, M., Marín, S., Sanchis, V., Ramos, A.J., 2005. Fate of mycotoxins in cereals during extrusion cooking: a review. Food Addit. Contam. 22, 150–157. Castells, M., Marín, S., Sanchis, V., Ramos, A.J., 2006a. Reduction of aflatoxins by extrusion-cooking of rice meal. J. Food Sci. 71, 369–377. Castells, M., Pardo, E., Ramos, A.J., Sanchis, V., Marin, S., 2006b. Reduction of ochratoxin A in extruded barley meal. J. Food Prot. 69, 1139–1143. Castells, M., Ramos, A.J., Sanchis, V., Marín, S., 2009. Reduction of fumonisin B1 in extruded corn breakfast cereals with salt, malt and sugar in their formulation. Food Addit. Contam. 26, 512–517. Cazzaniga, D., Basilico, J.C., Gonzalez, R.J., Torres, R.L., de Greef, D.M., 2001. Mycotoxins inactivation by extrusion cooking of corn flour. Lett. Appl. Microbiol. 33, 144–147. Cheli, F., Pinotti, L., Rossi, L., Dell’Orto, V., 2013. Effect of milling procedures on mycotoxin distribution in wheat fractions: a review. Food Sci. Technol. Int. 54, 307–314.

151

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

Cheli, F., Battaglia, D., Gallo, R., Dell’Orto, V., 2014. EU legislation on cereal safety: an update with a focus on mycotoxins. Food Control 37, 315–325. Cortez-Rocha, M.O., Trigo-Stockli, D.M., Wetzel, D.L., Reed, C.R., 2002. Effect of extrusion processing on fumonisin B1 and hydrolyzed fumonisin B1 in contaminated alkali-cooked corn. Bull. Environ. Contam. Toxicol. 69, 471–478. Czerwiecki, L., Czajkowska, D., Witkowska-Gwiazdowska, A., 2002. On ochratoxin A and fungal flora in Polish cereals fromconventional and ecological farms. Food Addit. Contam. 19, 470–477. D’Ovidio, K.L., Trucksess, M.W., Devries, J.W., Bean, G., 2007. Effects of irradiation on fungi and fumonisin B1 in corn, and of microwave-popping on fumonisins in popcorn. Food Addit. Contam. 24, 735–743. Danicke, S., Valenta, H., Döll, S., 2004. On the toxicokinetics and the metabolism of deoxynivalenol (DON) in the pig. Arch. Anim. Nutr. 58, 169–180. De Girolamo, A., Solfrizzo, M., Visconti, A., 2001. Effect of processing on fumonisin concentration in corn flakes. J. Food Prot. 5, 701–705. Dersjant-Li, Y., Verstegen, M.W.A., Gerrits, W.J.J., 2003. The impact of low concentrations of aflatoxin, deoxynivalenol or fumonisin in diets on growing pigs and poultry. Nutr. Res. Rev. 16, 223–239. Devegowda, G., Raju, M.V.L.N., Afzali, N., Swamy, H.V.L.N., 1998. Mycotoxin picture worldwide: novel solutions for their counteraction. In: Lyons, T.P., Jacques, K.A. (Eds.), Biotechnology in the Feed Industry: Proceedings of the 14th Annual Symposium. Nottingham University Press, Nottingham, pp. 241–255. European Commission, 1999a. Directive 1999/2/EC of The European Parliament and of the Council: on the approximation of the laws of the Member States concerning foods and food ingredients treated with ionising radiation. Off. J. Eur. Commun. 16–22 Issue 13.3.1999. European Commission, 1999b. Directive 1999/3/EC of The European Parliament and of the Council: on the establishment of a Community list of foods and food ingredients treated with ionising radiation. Off. J. Eur. Commun. 16–22 Issue 13.3.1999. European Commission, 2002 (amended in 2010). Directive 2002/32/EC of the European Parliament and of the Council: on undesirable substances in animal feed. Off. J. Eur. Union, Issue 01.07.2010, 1–27. European Commission, 1999a. Commission Recommendation of 17 August 2006: on the presence of deoxynivalenol, zearalenone, ochratoxin A, T-2 and HT-2 and fumonisins in products intended for animal feeding. Off. J. Eur. Union Issue 08.23.2016. European Commission, 2006b. Commission Regulation No 1881/2006: setting maximum levels for certain contaminants in foodstuffs. Off. J. Eur. Union 7–9 Issue 01.07.2010. European Commission, 2009. Commission Regulation (EC) No 386/2009: amending Regulation (EC) No 1831/2003 of the European Parliament and of the Council as regards the establishment of a new functional group of feed additives. Off. J. Eur. Union 66 Issue 13.5.2009. European Commission, 2013. Commission Implementing Regulation (EU) No 1060/2013: concerning the authorisation of bentonite as a feed additive for all animal species. Off. J. Eur. Union 33–37 Issue 31.10.2013. Eckard, S., Wettstein, F.E., Forrer, H.R., Vogelgsang, S., 2011. Incidence of fusarium species and mycotoxins in silage maize. Toxins 3, 949–967. Elias-Orozco, R., Castellanos-Nava, A., Gaytan-Martinez, M., Figueroa-Cardenas, J.D., Loarca-Pina, G., 2002. Comparison of nixtamalization and extrusion processes for a reduction in aflatoxin content. Food Addit. Contam. 19, 878–885. Fink-Gremmels, J., Malekinejad, H., 2007. Clinical effects and biochemical mechanisms associated with exposure to the mycoestrogen zearalenone. Anim. Feed Sci. Technol. 137 (3), 326–341. Fink-Gremmels, J., 2008. Mycotoxins in cattle feeds and carry-over to dairy milk: a review. Food Addit. Contam. 25, 172–180. Gao, X., Shim, W.B., Göbel, C., Kunze, S., Feussner, I., Meeley, R., Balint-Kurti, P., Kolomiets, M., 2007. Disruption of a maize 9-lipoxygenase results in increased resistance to fungal pathogens and reduced levels of contamination with mycotoxin fumonisin. Mol. Plant Microbe Interact. 20, 922–933. Ghanem, I., Orfi, M., Shamma, M., 2008. Effect of gamma radiation on the inactivation of aflatoxin B1 in food and feed crops. Braz. J. Microbiol. 39, 787–791. Han, X.Y., Huang, Q.C., Li, W.F., Jiang, J.F., Xu, Z.R., 2008. Changes in growth performance, digestive enzyme activities and nutrient digestibility of cherry valley ducks in response to aflatoxin B1 levels. Livest. Sci. 119, 216–220. Hassan, Y.I., He, J.W., Perilla, N., Tang, K., Karlovsky, P., Zhou, T., 2017. The enzymatic epimerization of deoxynivalenol by Devosia mutans proceeds through the formation of 3-keto-DON intermediate. Sci. Rep. 7 Article number: 6929. He, J., Zhou, T., Young, J.C., Boland, G.J., Scott, P.M., 2010. Chemical and biological transformations for detoxification of trichothecene mycotoxins in human and animal food chains: a review. Trends Food Sci. Technol. 21, 67–76. Heinl, S., Hartinger, D., Thamhesl, M., Schatzmayr, G., Moll, W.-D., Grabherr, R., 2010. An aminotransferase from bacterium ATCC 55552 deaminates hydrolyzed fumonisin B1. Biodegradation 22, 25–30. Herzallah, S., Alshawabkeh, K., Al Fataftah, A., 2008. Aflatoxin decontamination of artificially contaminated feeds by sunlight, γ-radiation, and microwave heating. J. Appl. Poult. Res. 17, 515–521. House, J.D., Nyachoti, C.M., Abramson, D., 2003. Deoxynivalenol removal from barley intended as swine feed through the use of an abrasive pearling procedure. J. Agric. Food Chem. 51, 5172–5175. Huff, W.E., Hagler, W.M., 1982. Note: evaluation of density segregation as a means to estimate the degree of aflatoxin contamination of corn. Cereal Chem. 59, 152–154. Huff, W.E., Hagler, W.M., 1985. Density segregation of corn and wheat naturally contaminated with Aflatoxin, Deoxynivalenol and Zearalenone. J. Food Prot. 48, 416–420. Huff, W.E., 1980. A physical method for the segregation of aflatoxin-contaminated corn. Cereal Chem. 57, 236–238. Hussein, H.S., Brasel, J.M., 2001. Toxicity, metabolism, and impact of mycotoxins on humans and animals. Toxicology 167, 101–134. Jard, G., Liboz, T., Mathieu, F., Guyonvarc’h, A. Lebrihi, 2011. Review of mycotoxin reduction in food and feed: from prevention in the field to detoxification by adsorption or transformation. Food Addit. Contam. 28 (11), 1590–1609. Jouany, J.P., 2007. Methods for preventing, decontaminating and minimizing the toxicity of mycotoxins in feeds. Anim. Feed Sci. Technol. 137, 342–362. Kabak, B., Dobson, A.D.W., Var, I., 2006. Strategies to prevent mycotoxin contamination of food and animal feed: a review. Crit. Rev. Food Sci. Nutr. 46, 593–619. Kabak, B., 2009. The fate of mycotoxins during thermal food processing. J. Sci. Food Agric. 89, 549–554. Karlovsky, P., 2011. Biological detoxification of the mycotoxin deoxynivalenol and its use in genetically engineered crops and feed additives. Appl. Microbiol. Biotechnol. 91, 491–504. Katta, S.K., Jackson, L.S., Sumner, S.S., Hanna, M.A., Bullerman, L.B., 1999. Effect of temperature and screw speed on stability of fumonisin b1 in extrusion-cooked corn grits. Cereal Chem. 76, 16–20. Kogan, G., Kocher, A., 2007. Role of yeast cell wall polysaccharides in pig nutrition and health protection. Livest. Sci. 109, 161–165. Kostelanska, M., Dzuman, Z., Malachova, A., Capouchova, I., Prokinova, E., Skerikova, A., Hajslova, J., 2011. Effects of milling and baking technologies on levels of deoxynivalenol and its masked form deoxynivalenol-3-glucoside. J. Agric. Food Chem. 59, 9303–9312. Lancova, K., Hajslova, J., Kostelanska, M., Kohoutkova, J., Nedelnik, J., Moravcova, H., Vanova, M., 2008. Fate of trichothecene mycotoxins during the processing milling and baking. Food Addit. Contam. 25, 650–659. Le Thanh, B.V., Lemay, M., Bastien, A., Lapointe, J., Lessard, M., Chorfi, Y., Guay, F., 2016. The potential effects of antioxidant feed additives in mitigating the adverse effects of corn naturally contaminated with Fusarium mycotoxins on antioxidant systems in the intestinal mucosa, plasma, and liver in weaned pigs. Mycotoxin Res. 32, 99–116. Magan, N., Hope, R., Cairns, V., Aldred, D., 2003. Post-harvest fungal ecology: impact of fungal growth and mycotoxin accumulation in stored grain. Eur. J. Plant Pathol. 109, 723–730. Magan, N., Medina, A., Aldred, D., 2011. Possible climate-change effects on mycotoxin contamination of food crops pre- and postharvest. Plant Pathol. 60, 150–163. Mansfield, M.A., Jones, A.D., Kuldau, G.A., 2008. Contamination of fresh and ensiled maize by multiple penicillium mycotoxins. Phytopathology 98, 330–336. Massman, J., Cooper, B., Horsley, R., Neate, S., Dill-Macky, R., Chao, S., Dong, Y., Schwarz, P., Muehlbauer, G.J., Smith, K.P., 2010. Genome-wide association mapping of Fusarium head blight resistance in contemporary barley breeding germplasm. Mol. Breed. 27, 439–454. Matumba, L., van Poucke, C., Ediage, E.N., Jacobs, B., De Saeger, S., 2015. Effectiveness of hand sorting, flotation/washing, dehulling and combinations thereof on the decontamination of mycotoxin-contaminated white maize. Food Addit. Contam. Part A 32, 960–969. McKenzie, K.S., Kubena, L.F., Denvir, A.J., Rogers, T.D., Hitchens, G.D., Bailey, R.H., Harvey, R.B., Buckley, S.A., Phillips, T.D., 1998. Aflatoxicosis in turkey poults is prevented by treatment of naturally contaminated corn with ozone generated by electrolysis. Poult. Sci. 77, 1094–1102. Meissonnier, G.M., Raymond, I., Laffitte, J., Cossalter, A.M., Pinton1, P., Benoit, E., Bertin, G., Galtier, P., Oswald, I.P., 2009. Dietary glucomannan improves the vaccinal response in pigs exposed to aflatoxin B1 or T-2 toxin. World Mycotoxin J. 2, 161–172.

152

Animal Feed Science and Technology 237 (2018) 129–153

W.-X. Peng et al.

Mesterhazy, K., Lemmens, M., Reid, L.M., 2012. Breeding for resistance to ear rots caused by Fusarium spp. in maize—a review. Plant Breed. 131, 1–19. Miedaner, T., Reinbrecht, C., Lauber, U., Schollenberger, M., Geiger, H.H., 2001. Effects of genotype and genotype-environment interaction on deoxynivalenol accumulation and resistance to Fusarium head blight in rye, triticale, and wheat. Plant Breed. 120, 97–105. Miller, J.D., 2008. Mycotoxins in small grains and maize: old problems new challenges. Food Addit. Contam. 25, 219–230. Modirsanei, M., Mansoori, B., Khosravi, A.R., Kiaei, M.M., Khazraeinia, P., Farkhoy, M., Masoumi, Z., 2008. Effect of diatomaceous earth on the performance and blood variables of broiler chicks during experimental aflatoxicosis. J. Sci. Food Agric. 88, 626–632. O’Neill, K., Damoglou, A.P., Patterson, M.F., 1993. The stability of deoxynivalenol and 3-acetyl deoxynivalenol to gamma irradiation. Food Addit. Contam. 10, 209–215. Pelgrom, P.J.M., Vissers, A.M., Boom, R.M., Schutyser, M.A.I., 2013. Dry fractionation for production of functional pea protein concentrates. Food Res. Int. 53, 232–239. Pestka, J.J., 2007. Deoxynivalenol: toxicity: mechanisms and animal health risks. Anim. Feed Sci. Technol. 137, 283–298. Pfohl-Leszkowicz, A., Manderville, R.A., 2007. Ochratoxin A: an overview on toxicity and carcinogenicity in animals and humans. Mol. Nutr. Food Res. 51, 61–99. Piñeiro, M., Miller, J., Silva, G., Musser, S., 1999. Effect of commercial processing on fumonisin concentrations of maize-based foods. Mycotoxin Res. 15, 2–12. Piva, A., Casadei, G., Pagliuca, G., Cabassi, E., Galvano, F., Solfrizzo, M., Riley, R.T., Diaz, D.E., 2005. Activated carbon does not prevent the toxicity of culture material containing fumonisin B1 when fed to weanling piglets. J. Anim. Sci. 83, 1939–1947. Pronyk, C., Cenkowski, S., Abramson, D., 2006. Superheated steam reduction of deoxynivalenol in naturally contaminated wheat kernels. Food Control 17, 789–796. Qin, G., ter Elst, E.R., Bosch, M.W., van der Poe1, A.F.B., 1996. Thermal processing of whole soya beans: studies on the inactivation of antinutritional factors and effects on ileal digestibility in piglets. Anim. Feed Sci. Technol. 57, 313–324. Reddy, K.R.N., Abbas, H.K., Abel, C.A., Shier, W.T., Oliveira, C.A.F., Raghavender, C.R., 2009. Mycotoxin contamination of commercially important agricultural commodities. Toxin Rev. 28 (2-3), 154–168. Richard, J.L., 2007. Some major mycotoxins and their mycotoxicoses: an overview. Int. J. Food Microbiol. 119, 3–10. Riley, R.T., 1998. Mechanistic interactions of mycotoxins: theoretical considerations. In: Sinha, K.K., Bhatnagar, D. (Eds.), Mycotoxins in Agriculture and Food Safety. Marcel Dekker, Inc., New York, pp. 227–253. Rios, G., Pinson-Gadais, L., Abecassis, J., Zakhia-Rozis, N., Lullien-Pellerin, V., 2009a. Assessment of dehulling efficiency to reduce deoxynivalenol and Fusarium level in durum wheat grains. J. Cereal Sci. 49, 387–392. Rios, G., Zakhia-Rozis, N., Chaurand, M., Richard-Forget, F., Samson, M.F., Abecassis, J., Lullien-Pellerin, V., 2009b. Impact of durum wheat milling on deoxynivalenol distribution in the outcoming fractions. Food Addit. Contam. Part A 26, 487–495. Rodrigues, I., Naehrer, K., 2012. A three-year survey on the worldwide occurrence of mycotoxins in feedstuffs and feed. Toxins 4, 663–675. Ryu, D., Hanna, M.A., Bullerman, L.B., 1999. Stability of zearalenone during extrusion of corn grits. J. Food Prot. 62, 1482–1484. Savi, G.D., Piacentini, K.C., Bittencourt, K.O., Scussel, V.M., 2014. Ozone treatment efficiency on Fusarium graminearum and deoxynivalenol degradation and its effects on whole wheat grains (Triticum aestivum L.) quality and germination. J. Stored Prod. Res. 59, 245–253. Schaich, K.M., Rebello, C.A., 1999. Extrusion chemistry of wheat flour proteins: I. Free radical formation. Cereal Chem. 76 (5), 748–755. Schaich, K.M., 2002. Free radical generation during extrusion: a critical contributor to texturization. ACS Symp. Ser. 807, 35–48. Schwake-Anduschus, C., Langenkämper, G., Unbehend, G., Dietrich, R., Märtlbauer, E., Münzing, K., 2010. Occurrence of Fusarium T-2 and HT-2 toxins in oats from cultivar studies in Germany and degradation of the toxins during grain cleaning treatment and food processing. Food Addit. Contam. 27, 1253–1260. Scudamore, K.A., Banks, J.N., Guy, R.C.E., 2004. Fate of ochratoxin A in the processing of whole wheat grain during extrusion. Food Addit. Contam. 21, 488–497. Scudamore, K.A., Baillie, H., Patel, S., Edwards, S.G., 2007. Occurrence and fate of Fusarium mycotoxins during commercial processing of oats in the UK. Food Addit. Contam. 24, 1374–1385. Shetty, P.H., Jespersen, L., 2006. Saccharomyces cerevisiae and lactic acid bacteria as potential mycotoxin decontaminating agents. Trends Food Sci. Technol. 17, 48–55. Siwela, A.H., Siwela, M., Matindi, G., Dube, S., Nziramasanga, N., 2005. Decontamination of aflatoxin-contaminated maize by dehulling. J. Sci. Food Agric. 85, 2535–2538. Skadsen, R.W., Hohn, T.M., 2004. Use of Fusarium graminearum transformed with gfp to follow infection patterns in barley and Arabidopsis. Physiol. Mol. Plant Pathol. 64, 45–53. Stanley, V.G., Ojo, R., Woldesenbet, S., Hutchinson, D.H., 1993. The use of saccharomyces cerevisiae to suppress the effects of aflatoxicosis in broiler chicks. Poult. Sci. 72, 1867–1872. Starkl, V., Hofstetter, U., Tenier, C., 2015. Efficacité de la souche bactérienne DSM11798 à biotransformer le déoxynivalénol et son métabolite le dé-époxydéoxynivalonenol (DOM-1), dans le sérum des porcs. J. Rech. Porcine 47, 135–136. Stepanik, T., Kost, D., Nowicki, T., Gaba, D., 2007. Effects of electron beam irradiation on deoxynivalenol levels in distillers dried grain and solubles and in production intermediates. Food Addit. Contam. 24, 1001–1006. Stoev, S.D., Paskalev, M., Macdonald, S., Mantle, P.G., 2002. Experimental one year ochratoxin A toxicosis in pigs. Exp. Toxicol. Pathol. 53, 481–487. Streit, E., Schatzmayr, G., Tassis, P., Tzika, E., Marin, D., Taranu, I., Tabuc, C., Nicolau, A., Aprodu, I., Puel, O., Oswald, I.P., 2012. Current situation of mycotoxin contamination and co-occurrence in animal feed—focus on Europe. Toxins 4, 788–809. Streit, E., Schwab, C., Sulyok, M., Naehrer, K., Krska, R., Schatzmayr, G., 2013. Multi-mycotoxin screening reveals the occurrence of 139 different secondary metabolites in feed and feed ingredients. Toxins 5, 504–523. Tajkarimi, M., Riemann, H.P., Hajmeer, M.N., Gomez, E.L., Razavilar, V., Cliver, D.O., 2008. Ammonia disinfection of animal feeds—laboratory study. Int. J. Food Microbiol. 122, 23–28. Tibola, C.S., Fernandes, J.M.C., Guarienti, E.M., Nicolau, M., 2015. Distribution of Fusarium mycotoxins in wheat milling process. Food Control 53, 91–95. Tibola, C.S., Fernandes, J.M.C., Guarienti, E.M., 2016. Effect of cleaning, sorting and milling processes in wheat mycotoxin content. Food Control 60, 174–179. Tiwari, B.K., Brennan, C.S., Curran, T., Gallagher, E., Cullen, P.J., O’Donnell, C.P., 2010. Application of ozone in grain processing. J. Cereal Sci. 51, 248–255. Ullsten, N.H., Gallstedt, M.G., Johansson, E., Eklund, A.G., Hedenqvist, M.S., 2006. Enlarged processing window of plasticized wheat gluten using salicylic acid. Biomacromolecules 7, 771–776. Urraca, J.L., Marazuela, M.D., Moreno-Bondi, M.C., 2004. Analysis for zearalenone and α-zearalenol in cereals and swine feed using accelerated solvent extraction and liquid chromatography with fluorescence detection. Anal. Chim. Acta 524, 175–183. van der Westhuizen, L., Shephard, G.S., Rheeder, J.P., Burger, H.M., Gelderblom, W.C.A., Wild, C.P., Gong, Y.Y., 2011. Optimising sorting and washing of home-grown maize to reduce fumonisin contamination under laboratory-controlled conditions. Food Control 22, 396–400. Vanhoutte, I., Audenaert, K., de Gelder, L., 2016. Biodegradation of mycotoxins: tales from known and unexplored worlds. Front. Microbiol. 7, 1–20. Visconti, A., Haidukowski, E.M., Pascale, M., Silvestri, M., 2004. Reduction of deoxynivalenol during durum wheat processing and spaghetti cooking. Toxicol. Lett. 153, 181–189. Weng, C.Y., Martinez, A.J., Park, D.L., 1994. Efficacy and permanency of ammonia treatment in reducing aflatoxin levels in corn. Food Addit. Contam. 11, 649–658. Wu, X., Murphy, P., Cunnick, J., Hendrich, S., 2007. Synthesis and characterization of deoxynivalenol glucuronide: its comparative immunotoxicity with deoxynivalenol. Food Chem. Toxicol. 45, 1846–1855. Young, J.C., Zhu, H., Zhou, T., 2006. Degradation of trichothecene mycotoxins by aqueous ozone. Food Chem. Toxicol. 44, 417–424. Yumbe-Guevara, B.E., Imoto, T., Yoshizawa, T., 2003. Effects of heating procedures on deoxynivalenol, nivalenol and zearalenone levels in naturally contaminated barley and wheat. Food Addit. Contam. 20, 1132–1140.

153