Machine Learning Applied to the Detection of Mycotoxin in Food: A Systematic Review

Mycotoxins, toxic secondary metabolites produced by certain fungi, pose significant threats to global food safety and public health. These compounds can contaminate a variety of crops, leading to economic losses and health risks to both humans and animals. Traditional lab analysis methods for mycotoxin detection can be time-consuming and may not always be suitable for large-scale screenings. However, in recent years, machine learning (ML) methods have gained popularity for use in the detection of mycotoxins and in the food safety industry in general due to their accurate and timely predictions. We provide a systematic review on some of the recent ML applications for detecting/predicting the presence of mycotoxin on a variety of food ingredients, highlighting their advantages, challenges, and potential for future advancements. We address the need for reproducibility and transparency in ML research through open access to data and code. An observation from our findings is the frequent lack of detailed reporting on hyperparameters in many studies and a lack of open source code, which raises concerns about the reproducibility and optimisation of the ML models used. The findings reveal that while the majority of studies predominantly utilised neural networks for mycotoxin detection, there was a notable diversity in the types of neural network architectures employed, with convolutional neural networks being the most popular.

Changing climate, shifting mycotoxins: A comprehensive review of climate change impact on mycotoxin contamination

Climate change (CC) is a complex phenomenon that has the potential to significantly alter marine, terrestrial, and freshwater ecosystems worldwide. Global warming of 2°C is expected to be exceeded during the 21st century, and the frequency of extreme weather events, including floods, storms, droughts, extreme temperatures, and wildfires, has intensified globally over recent decades, differently affecting areas of the world. How CC may impact multiple food safety hazards is increasingly evident, with mycotoxin contamination in particular gaining in prominence. Research focusing on CC effects on mycotoxin contamination in edible crops has developed considerably throughout the years. Therefore, we conducted a comprehensive literature search to collect available studies in the scientific literature published between 2000 and 2023. The selected papers highlighted how warmer temperatures are enabling the migration, introduction, and mounting abundance of thermophilic and thermotolerant fungal species, including those producing mycotoxins. Certain mycotoxigenic fungal species, such as Aspergillus flavus and Fusarium graminearum, are expected to readily acclimatize to new conditions and could become more aggressive pathogens. Furthermore, abiotic stress factors resulting from CC are expected to weaken the resistance of host crops, rendering them more vulnerable to fungal disease outbreaks. Changed interactions of mycotoxigenic fungi are likewise expected, with the effect of influencing the prevalence and co-occurrence of mycotoxins in the future. Looking ahead, future research should focus on improving predictive modeling, expanding research into different pathosystems, and facilitating the application of effective strategies to mitigate the impact of CC.

Investigating Metabolic Plant Response toward Deoxynivalenol Accumulation in Four Winter Cereals

Deoxynivalenol (DON) is a phytotoxic agent supporting the spread of fungal diseases in cereals worldwide, i.e., fusarium head blight. It is known that DON accumulation may elicit changes in plant secondary metabolites in response to pathogen attack. This study maps the changes in selected secondary metabolite classes upon DON contamination occurring in fifteen Triticum spp. genotypes, among them emmer, spelt, and soft wheat, and 2 tritordeum varieties, cultivated in two different sites and over two harvest years. The main phenolic classes (i.e., alkylresorcinols, soluble, and cell-wall bound phenolic acids) were targeted analyzed, while changes in the lipidome signature were collected through untargeted HRMS experiments. The results, obtained across multiple Triticum species and in open fields, confirmed the modulation of first-line biological pathways already described in previous studies involving single cereal species or a limited germplasm, thus reinforcing the involvement of nonspecific chemical defenses in the plant response to pathogen attack.

Mycotoxin occurrence in kernels and straws of wheat, barley, and tritordeum

Thirty-two varieties of common and durum wheat, hordeum, barley, and tritordeum collected over two harvesting years (2020 and 2021) were investigated for the presence of multiple Fusarium-related mycotoxins in asymptomatic plants. DON, 3-AcDON, 15-AcDON, T-2, HT-2, and ZEN together with the emerging mycotoxin ENN B and the major modified form of DON, namely DON3Glc, were quantified by means of UHPLC-MS/MS. Overall, DON and ENN B were the most frequently detected mycotoxins, albeit large inter-year variability was observed and related to different climate and weather conditions. Straws had higher mycotoxin contents than kernels and regarding DON occurrence tritordeum was found to be the most contaminated group on average for both harvesting years, while barley was the less contaminated one. Emerging mycotoxin ENN B showed comparable contents in kernels compared to straw, with a ratio close to 1 for tritordeum and barley. Regarding the occurrence of the other evaluated mycotoxins, T-2 and HT-2 toxins have been spotted in a few tritordeum samples, while ZEN has been frequently found only in straw from the harvesting year 2020. The data collected confirms the occurrence of multiple Fusarium mycotoxins in straws also from asymptomatic plants, highlighting concerns related to feed safety and animal health. The susceptibility of Tritordeum, hereby reported for the first time, suggests that careful measures in terms of monitoring, breeding, and cultural choices should be applied when dealing with his emerging crop.

Aflatoxin Decontamination in Maize Steep Liquor Obtained from Bioethanol Production Using Laccases from Species within the Basidiomycota Phylum

Maize (Zea mays L.) is an important crop in Argentina. Aspergillus section Flavi can infect this crop at the pre-harvest stage, and the harvested grains can be contaminated with aflatoxins (AFs). During the production of bioethanol from maize, AF levels can increase up to three times in the final co-products, known as, dry and wet distiller's grain with solubles (DDGS and WDGS), intended for animal feed. Fungal enzymes like laccases can be a useful tool for reducing AF contamination in the co-products obtained from this process. The aim of the present study was to evaluate the ability of laccase enzymes included in enzymatic extracts (EE) produced by different species in the Basidiomycota phylum to reduce AF (AFB1 and AFB2) accumulation under the conditions of in vitro assays. Four laccase activities (5, 10, 15, and 20 U/mL) exerted by nine isolates were evaluated in the absence and presence of vanillic acid (VA), serving as a laccase redox mediator for the degradation of total AFs. The enzymatic stability in maize steep liquor (MSL) was confirmed after a 60 h incubation period. The most effective EE in terms of reducing AF content in the buffer was selected for an additional assay carried out under the same conditions using maize steep liquor obtained after the saccharification stage during the bioethanol production process. The highest degradation percentages were observed at 20 U/mL of laccase enzymatic activity and 1 mM of VA, corresponding to 26% for AFB1 and 26.6% for AFB2. The present study provides valuable data for the development of an efficient tool based on fungal laccases for preventing AF accumulation in the co-products of bioethanol produced from maize used for animal feed.

An integrated model for pre- and post-harvest aflatoxin contamination in maize

Aflatoxin contamination caused by colonization of maize by Aspergillus flavus continues to pose a major human and livestock health hazard in the food chain. Increasing attention has been focused on the development of models to predict risk and to identify effective intervention strategies. Most risk prediction models have focused on elucidating weather and site variables on the pre-harvest dynamics of A. flavus growth and aflatoxin production. However fungal growth and toxin accumulation continue to occur after harvest, especially in countries where storage conditions are limited by logistical and cost constraints. In this paper, building on previous work, we introduce and test an integrated meteorology-driven epidemiological model that covers the entire supply chain from planting to delivery. We parameterise the model using approximate Bayesian computation with monthly time-series data over six years for contamination levels of aflatoxin in daily shipments received from up to three sourcing regions at a high-volume maize processing plant in South Central India. The time series for aflatoxin levels from the parameterised model successfully replicated the overall profile, scale and variance of the historical aflatoxin datasets used for fitting and validation. We use the model to illustrate the dynamics of A. flavus growth and aflatoxin production during the pre- and post-harvest phases in different sourcing regions, in short-term predictions to inform decision making about sourcing supplies and to compare intervention strategies to reduce the risks of aflatoxin contamination.

The Effect of Ustilago maydis and Delayed Harvesting on A- and B-Type Trichothecene Concentrations in Maize Grain

The aim of this study was to investigate whether, in the context of a higher incidence of Ustilago maydis and Fusarium spp. at optimal and delayed harvest times, a higher incidence of mycotoxin contamination in maize grains could be expected. The field experiment was carried out at the Lithuanian Research Centre for Agriculture and Forestry experimental fields over three consecutive years (2020-2022). Two maize hybrids (Duxxbury and Lapriora) with different FAO numbers were used. The experimental design in the field was a randomized complete block design. Harvesting took place at three different times: first at physiological maturity, and then 10 (±2) and 20 (±2) days after the first harvest. Each hybrid had four repetitions at different harvest times. The U. maydis infection was only detected in 2021 and after the first harvest cobs were further divided into four different groups with four repetitions: healthy cobs, cobs visually infected with Fusarium spp., cobs visually infected with common smut, and cobs visually infected with both pathogens. No U. maydis-damaged maize cobs were found in 2020 and 2022. The levels of Fusarium microscopic fungi in maize grains were also from 4 to 16 times higher in 2021 than in 2020 and 2022. Harvest delays in 2020 led to a significant deoxynivalenol concentration increase in the Duxxbury hybrid and an HT-2 concentration increase in the Lapriora hybrid. In 2021, deoxynivalenol, 3-acetyl-deoxynivalenol, 15-acetyl-deoxynivalenol, and HT-2 concentrations significantly rose in both hybrids, but the T-2 concentration significantly increased only in the Lapriora hybrid. Deoxynivalenol concentrations were, respectively, 110 and 14.6 times higher than in cobs only infected with Fusarium spp. or U. maydis. Concentrations of 15-acetyl-deoxynivalenol were, respectively, 60, 67, and 43 times higher than in asymptomatic cobs and cobs only infected with Fusarium spp. or U. maydis. Cobs contaminated with both pathogens also had higher concentrations of 3-acetyl-deoxynivalenol. T-2 and HT-2 were detected in maize grains harvested from cobs infected only with Fusarium spp. The presence of U. maydis and Fusarium fungi in maize cobs, along with harvest delays, led to significant increases in mycotoxin concentrations, highlighting the importance of timely harvesting and pathogen management to mitigate mycotoxin contamination in maize grains.

Storage Fungi and Mycotoxins Associated with Rice Samples Commercialized in Thailand

The study focused on the examination of the different fungal species isolated from commercial rice samples, applying conventional culture techniques, as well as different molecular and phylogenic analyses to confirm phenotypic identification. Additionally, the mycotoxin production and contamination were analyzed using validated liquid chromatography-tandem mass spectrometry (LC-MS/MS). In total, 40 rice samples were obtained covering rice berry, red jasmine rice, brown rice, germinated brown rice, and white rice. The blotting paper technique applied on the 5 different types of rice samples detected 4285 seed-borne fungal infections (26.8%) for 16,000 rice grains. Gross morphological data revealed that 19 fungal isolates belonged to the genera Penicillium/Talaromyces (18 of 90 isolates; 20%) and Aspergillus (72 of 90 isolates; 80%). To check their morphologies, molecular data (fungal sequence-based BLAST results and a phylogenetic tree of the combined ITS, BenA, CaM, and RPB2 datasets) confirmed the initial classification. The phylogenic analysis revealed that eight isolates belonged to P. citrinum and, additionally, one isolate each belonged to P. chermesinum, A. niger, A. fumigatus, and A. tubingensis. Furthermore, four isolates of T. pinophilus and one isolate of each taxon were identified as Talaromyces (T. radicus, T. purpureogenum, and T. islandicus). The results showed that A. niger and T. pinophilus were two commonly occurring fungal species in rice samples. After subculturing, ochratoxin A (OTA), generated by T. pinophilus code W3-04, was discovered using LC-MS/MS. In addition, the Fusarium toxin beauvericin was detected in one of the samples. Aflatoxin B1 or other mycotoxins, such as citrinin, trichothecenes, and fumonisins, were detected. These preliminary findings should provide valuable guidance for hazard analysis critical control point concepts used by commercial food suppliers, including the analysis of multiple mycotoxins. Based on the current findings, mycotoxin analyses should focus on A. niger toxins, including OTA and metabolites of T. pinophilus (recently considered a producer of emerging mycotoxins) to exclude health hazards related to the traditionally high consumption of rice by Thai people.

Advanced mycotoxin control and decontamination techniques in view of an increased aflatoxin risk in Europe due to climate change

Aflatoxins are toxic secondary metabolites produced by Aspergillus spp. found in staple food and feed commodities worldwide. Aflatoxins are carcinogenic, teratogenic, and mutagenic, and pose a serious threat to the health of both humans and animals. The global economy and trade are significantly affected as well. Various models and datasets related to aflatoxins in maize have been developed and used but have not yet been linked. The prevention of crop loss due to aflatoxin contamination is complex and challenging. Hence, the set-up of advanced decontamination is crucial to cope with the challenge of climate change, growing population, unstable political scenarios, and food security problems also in European countries. After harvest, decontamination methods can be applied during transport, storage, or processing, but their application for aflatoxin reduction is still limited. Therefore, this review aims to investigate the effects of environmental factors on aflatoxin production because of climate change and to critically discuss the present-day and novel decontamination techniques to unravel gaps and limitations to propose them as a tool to tackle an increased aflatoxin risk in Europe.

DEFHAZ: A Mechanistic Weather-Driven Predictive Model for Diaporthe eres Infection and Defective Hazelnut Outbreaks

The browning of the internal tissues of hazelnut kernels, which are visible when the nuts are cut in half, as well as the discolouration and brown spots on the kernel surface, are important defects that are mainly attributed to Diaporthe eres. The knowledge regarding the Diaporthe eres infection cycle and its interaction with hazelnut crops is incomplete. Nevertheless, we developed a mechanistic model called DEFHAZ. We considered georeferenced data on the occurrence of hazelnut defects from 2013 to 2020 from orchards in the Caucasus region and Turkey, supported by meteorological data, to run and validate the model. The predictive model inputs are the hourly meteorological data (air temperature, relative humidity, and rainfall), and the model output is the cumulative index (Dh-I), which we computed daily during the growing season till ripening/harvest time. We established the probability function, with a threshold of 1% of defective hazelnuts, to define the defect occurrence risk. We compared the predictions at early and full ripening with the observed data at the corresponding crop growth stages. In addition, we compared the predictions at early ripening with the defects observed at full ripening. Overall, the correct predictions were >80%, with <16% false negatives, which confirmed the model accuracy in predicting hazelnut defects, even in advance of the harvest. The DEFHAZ model could become a valuable support for hazelnut stakeholders.

Functional Study of Lipoxygenase-Mediated Resistance against Fusarium verticillioides and Aspergillus flavus Infection in Maize

Mycotoxin contamination of maize kernels by fungal pathogens like Fusarium verticillioides and Aspergillus flavus is a chronic global challenge impacting food and feed security, health, and trade. Maize lipoxygenase genes (ZmLOXs) synthetize oxylipins that play defense roles and govern host-fungal interactions. The current study investigated the involvement of ZmLOXs in maize resistance against these two fungi. A considerable intraspecific genetic and transcript variability of the ZmLOX family was highlighted by in silico analysis comparing publicly available maize pan-genomes and pan-transcriptomes, respectively. Then, phenotyping and expression analysis of ZmLOX genes along with key genes involved in oxylipin biosynthesis were carried out in a maize mutant carrying a Mu transposon insertion in the ZmLOX4 gene (named UFMulox4) together with Tzi18, Mo17, and W22 inbred lines at 3- and 7-days post-inoculation with F. verticillioides and A. flavus. Tzi18 showed the highest resistance to the pathogens coupled with the lowest mycotoxin accumulation, while UFMulox4 was highly susceptible to both pathogens with the most elevated mycotoxin content. F. verticillioides inoculation determined a stronger induction of ZmLOXs and maize allene oxide synthase genes as compared to A. flavus. Additionally, oxylipin analysis revealed prevalent linoleic (18:2) peroxidation by 9-LOXs, the accumulation of 10-oxo-11-phytoenoic acid (10-OPEA), and triglyceride peroxidation only in F. verticillioides inoculated kernels of resistant genotypes.

Changes in Fusarium and Aspergillus Mycotoxin Content and Fatty Acid Composition after the Application of Ozone in Different Maize Hybrids

Mycotoxins in maize represent a great threat to human health. For this reason, novel technics such as ozone treatment are used to reduce the content of maize mycotoxins. However, there is little knowledge about the effect of ozone treatment on maize quality parameters. This study investigated the changes in Fusarium and Aspergillus mycotoxins and the changes in fatty acids during the ozone treatment of maize samples. Sixteen maize hybrids were visually tested for the naturally occurring ear rot severity and treated with three different concentrations of ozone (40, 70, and 85 mg/L). Mycotoxin content in maize samples was determined using a high-performance liquid chromatography (HPLC) system, whereas dominant fatty acids were determined using gas chromatography coupled with a flame ionization detector (GC-FID). Ozone treatments could be successfully applied to reduce the content of mycotoxins in maize below the detection limit. Ozone treatments increased the content of monounsaturated fatty acids (MUFAs) and decreased the content of polyunsaturated fatty acids (PUFAs), i.e., linoleic acid (36.7% in relation to the lowest applied ozone concentration), which negatively affected the nutritional value of maize.

Designing a monitoring program for aflatoxin B1 in feed products using machine learning

Agricultural commodities used for feed and food production are frequently contaminated with mycotoxins, such as Aflatoxin B1 (AFB1). In Europe, both the government and companies have monitoring programs in place for the presence of AFB1. With limited resources and following risk-based monitoring as prescribed in EU Regulation 2017/625, these monitoring programs focus on batches with the highest probability of being contaminated. This study explored the use of machine learning algorithms (ML) to design risk-based monitoring programs for AFB1 in feed products, considering both monitoring cost and model performance. Historical monitoring data for the presence of AFB1 in feed products (2005–2018; 5605 records in total) were used. Four different ML algorithms, including Decision tree, Logistic regression, Support vector machine and Extreme gradient boosting (XGB), were applied and compared to predict the high-risk feed batches to be considered for further AFB1 sampling and analysis. The monitoring cost included the cost of: sampling and analysis, disease burden, storage, and of recalling and destroying contaminated feed batches. The ML algorithms were able to predict the high-risk batches, with an AUC, recall, and accuracy higher than 0.8, 0.6, and 0.9, respectively. The XGB algorithm outperformed the other three investigated ML. Its incorporation would result into up to 96% reduction in monitoring cost in 2016–2018, as compared to the official monitoring program. The proposed approach for designing risk based monitoring programs can support authorities and industries to reduce the monitoring cost for other food safety hazards as well.

Assessment of Fusarium-Damaged Kernels in Common Wheat in Romania in the Years 2015 and 2016 with Extreme Weather Events

This article assesses the occurrence of Fusarium-damaged kernels (FDKs) in common wheat (Triticum aestivum) under the influence of environmental factors and extreme weather events in Romania (exceptionally high air temperatures and extreme pedological drought produced by a dipole block in summer 2015, and extreme precipitation and floods produced by an omega block in spring 2016). Wheat samples (N = 272) were analyzed for FDKs via visual estimation and manual weighing according to ISO 7970 and are statistically evaluated using SPSS. The dipole block in 2015 reduced the effects of environmental factors to non-significant correlations with FDKs, while the omega block in 2016 was non-significantly to very significantly correlated with FDKs in the northwestern and western regions. The occurrence of FDKs was favored for wheat cultivation in acidic soils and inhibited in alkaline soils. Wheat samples with FDKs ≥ 1% were sampled from crops grown in river meadows with high and very high risks of flooding. Knowing the contaminants’ geographical and spatial distributions under the influence of regular and extreme weather events is important for establishing measures to mitigate the effects of climate change and to ensure human and animal health.

Mass Spectrometry Imaging Disclosed Spatial Distribution of Defense-Related Metabolites in Triticum spp

Fusarium Head Blight is the most common fungal disease that strongly affects Triticum spp., reducing crop yield and leading to the accumulation of toxic metabolites. Several studies have investigated the plant metabolic response to counteract mycotoxins accumulation. However, information on the precise location where the defense mechanism is taking place is scarce. Therefore, this study aimed to investigate the specific tissue distribution of defense metabolites in two Triticum species and use this information to postulate on the metabolites' functional role, unlocking the "location-to-function" paradigm. To address this challenge, transversal cross-sections were obtained from the middle of the grains. They were analyzed using an atmospheric-pressure (AP) SMALDI MSI source (AP-SMALDI5 AF, TransMIT GmbH, Giessen, Germany) coupled to a Q Exactive HF (Thermo Fisher Scientific GmbH, Bremen, Germany) orbital trapping mass spectrometer. Our result revealed the capability of (AP)-SMALDI MSI instrumentation to finely investigate the spatial distribution of wheat defense metabolites, such as hydroxycinnamic acid amides, oxylipins, linoleic and α-linoleic acids, galactolipids, and glycerolipids.

Lipid Signaling Modulates the Response to Fumonisin Contamination and Its Source, Fusarium verticillioides, in Maize

Fumonisin-contaminated maize (Zea mays L.) products are a major health concern because of their toxic effects in humans and animals. Breeding maize for increased mycotoxin resistance is one of the key sustainable strategies for mitigating the effects of fumonisin contamination. Recent studies suggest a link between fumonisin accumulation and plant lipid and oxylipin profiles. However, the data collected so far do not reveal a cause-and-effect relationship. In this study, to decipher the multifactorial nature of mycotoxin resistance and plant-pathogen interaction mechanisms, we examined the oxylipin and complex lipid profiles of two maize hybrids (H21 and H22, the latter showing significantly lower FBs content) grown in the open field in two locations over 3years. Untargeted ultra-high performance liquid chromatography coupled with quadrupole-time-of-flight (UHPLC-Q-TOF), together with chemometrics analysis, successfully distinguished between the two hybrids as having low- and high-level fumonisin contamination. Considering that H21 and H22 were exposed to the same environmental factors, the higher activation of lipid signaling systems in H22 suggests that other routes are enabled in the less susceptible hybrids to limit fumonisin B (FB) accumulation. Our results highlighted the crucial role played by oxylipin and sphingolipid signaling in modulating the complex maize response to F. verticillioides infection. Overall, our results returned a global view on the changes in lipid metabolites related to fumonisin accumulation under open field conditions, and revealed a strong activation of the lipid signaling cascade in maize in the presence of FB₁.

Overall Exposure of European Adult Population to Mycotoxins by Statistically Modelled Biomonitoring Data

This study presents the exposure scenario to mycotoxins of adult population throughout Europe. The urinary biomarkers values were obtained by modelling data from two European projects. Exposure to AFB1, OTA, CIT, FBs, DON, NIV and T2/HT2 are presented. The main output obtained refers to a concern for public health about AFM1, FBs, T2/HT2 and NIV, and low concern for OTA, DON and CIT. The margin of exposure for AFM1 did not respect the reference value of 10,000 considered of low priority for risk; for Fusarium toxins, FBs and T2/HT2, probable daily intake (PDI) values resulted about ten times higher than their tolerable daily intake and NIV presented the most critical situation with a calculated PDI 30 times higher than the reference TDI value. North and South Europe scenarios were also depicted by clustering biomonitoring data. OTA and DON showed to be prevalent in Northern countries and the opposite was noticed for ZEN, higher in Southern countries. The critical issues of the availability of records feeding the dataset and of the accuracy of excretion rate for some mycotoxins are source of uncertainty for the reliability of the outputs, nevertheless the time is ripe for asking for more concrete HBM values and/or HBM-HBGV which would help in interpreting the burden of mycotoxins in Europe.

Fusarium verticillioides and Aspergillus flavus Co-Occurrence Influences Plant and Fungal Transcriptional Profiles in Maize Kernels and In Vitro

Climate change will increase the co-occurrence of Fusarium verticillioides and Aspergillus flavus, along with their mycotoxins, in European maize. In this study, the expression profiles of two pathogenesis-related (PR) genes and four mycotoxin biosynthetic genes, FUM1 and FUM13, fumonisin pathway, and aflR and aflD, aflatoxin pathway, as well as mycotoxin production, were examined in kernels and in artificial medium after a single inoculation with F. verticillioides or A. flavus or with the two fungi in combination. Different temperature regimes (20, 25 and 30 °C) over a time-course of 21 days were also considered. In maize kernels, PR genes showed the strongest induction at 25 °C in the earlier days post inoculation (dpi)with both fungi inoculated singularly. A similar behaviour was maintained with fungi co-occurrence, but with enhanced defence response at 9 dpi under 20 °C. Regarding FUM genes, in the kernels inoculated with F. verticillioides the maximal transcript levels occurred at 6 dpi at 25 °C. At this temperature regime, expression values decreased with the co-occurrence of A. flavus, where the highest gene induction was detected at 20 °C. Similar results were observed in fungi grown in vitro, whilst A. flavus presence determined lower levels of expression along the entire time-course. As concerns afl genes, considering both A. flavus alone and in combination, the most elevated transcript accumulation occurred at 30 °C during all time-course both in infected kernels and in fungi grown in vitro. Regarding mycotoxin production, no significant differences were found among temperatures for kernel contamination, whereas in vitro the highest production was registered at 25 °C for aflatoxin B1 and at 20 °C for fumonisins in the case of single inoculation. In fungal co-occurrence, both mycotoxins resulted reduced at all the temperatures considered compared to the amount produced with single inoculation.

Deoxynivalenol Occurrence in Triticale Crops in Romania during the 2012-2014 Period with Extreme Weather Events

This article aims to evaluate deoxynivalenol occurrence in triticale crops in Romania in years with extreme weather events (2012: Siberian anticyclone with cold waves and heavy snowfall; 2013 and 2014: “Vb” cyclones with heavy precipitation and floods in spring). The deoxynivalenol level in triticale samples (N = 236) was quantified by ELISA. In Romania, the extreme weather events favoured deoxynivalenol occurrence in triticale in Transylvania and the Southern Hilly Area (44–47° N, 22–25° E) with a humid/balanced-humid temperate continental climate, luvisols and high/very high risk of floods. Maximum deoxynivalenol contamination was lower in the other regions, although heavy precipitation in May–July 2014 was higher, with chernozems having higher aridity. Multivariate analysis of the factors influencing deoxynivalenol occurrence in triticale showed at least a significant correlation for all components of variation source (agricultural year, agricultural region, average of deoxynivalenol, average air temperature, cumulative precipitation, soil moisture reserve, aridity indices) (p-value < 0.05). The spatial and geographic distribution of deoxynivalenol in cereals in the countries affected by the 2012–2014 extreme weather events revealed a higher contamination in Central Europe compared to southeastern and eastern Europe. Deoxynivalenol occurrence in cereals was favoured by local and regional agroclimatic factors and was amplified by extreme weather events.

Predicted Aflatoxin B1 Increase in Europe Due to Climate Change: Actions and Reactions at Global Level

Climate change (CC) is predicted to increase the risk of aflatoxin (AF) contamination in maize, as highlighted by a project supported by EFSA in 2009. We performed a comprehensive literature search using the Scopus search engine to extract peer-reviewed studies citing this study. A total of 224 papers were identified after step I filtering (187 + 37), while step II filtering identified 25 of these papers for quantitative analysis. The unselected papers (199) were categorized as “actions” because they provided a sounding board for the expected impact of CC on AFB₁ contamination, without adding new data on the topic. The remaining papers were considered as “reactions” of the scientific community because they went a step further in their data and ideas. Interesting statements taken from the “reactions” could be summarized with the following keywords: Chain and multi-actor approach, intersectoral and multidisciplinary, resilience, human and animal health, and global vision. In addition, fields meriting increased research efforts were summarized as the improvement of predictive modeling; extension to different crops and geographic areas; and the impact of CC on fungi and mycotoxin co-occurrence, both in crops and their value chains, up to consumers.

Machine Learning for Predicting Mycotoxin Occurrence in Maize

Meteorological conditions are the main driving variables for mycotoxin-producing fungi and the resulting contamination in maize grain, but the cropping system used can mitigate this weather impact considerably. Several researchers have investigated cropping operations' role in mycotoxin contamination, but these findings were inconclusive, precluding their use in predictive modeling. In this study a machine learning (ML) approach was considered, which included weather-based mechanistic model predictions for AFLA-maize and FER-maize [predicting aflatoxin B1 (AFB1) and fumonisins (FBs), respectively], and cropping system factors as the input variables. The occurrence of AFB1 and FBs in maize fields was recorded, and their corresponding cropping system data collected, over the years 2005-2018 in northern Italy. Two deep neural network (DNN) models were trained to predict, at harvest, which maize fields were contaminated beyond the legal limit with AFB1 and FBs. Both models reached an accuracy >75% demonstrating the ML approach added value with respect to classical statistical approaches (i.e., simple or multiple linear regression models). The improved predictive performance compared with that obtained for AFLA-maize and FER-maize was clearly demonstrated. This coupled to the large data set used, comprising a 13-year time series, and the good results for the statistical scores applied, together confirmed the robustness of the models developed here.

Recent Advances in Mycotoxin Analysis and Detection of Mycotoxigenic Fungi in Grapes and Derived Products

Mycotoxins are secondary metabolites of filamentous fungi that can cause toxic effects in human and animal health. Most of the filamentous fungi that produce these mycotoxins belong to four genera, namely, Aspergillus, Penicillium, Fusarium, and Alternaria. Mycotoxigenic fungi, along with mycotoxins, create a constant and serious economic threat for agriculture in many terms, counting product losses due to crop contamination and food spoilage, as well malnutrition when considering nutritional quality degradation. Given the importance of robust and precise diagnostics of mycotoxins and the related producing fungi in the grape food chain, one of the most important agricultural sectors worldwide, the present review initially delivers a comprehensive presentation of mycotoxin reports on grape and derived products, including a wide range of commodities such as fresh grapes, raisins, wine, juices, and other processed products. Next, based on worldwide regulations’ requirements for mycotoxins, and referring to the relative literature, this work presents methodological approaches for mycotoxin determination, and stresses major methods for the detection of fungal species responsible for mycotoxin production. The principle of function and basic technical background on the available analytical and molecular biology techniques developed—including chromatography, mass spectrometry, immunochemical-based assays, biosensors, and molecular assays—is briefly given, and references for their application to grape and derived product testing are highlighted.

Loss of ZmLIPOXYGENASE4 Decreases Fusarium verticillioides Resistance in Maize Seedlings

Fusarium verticillioides is one of the most relevant fungal species in maize responsible for ear, stalk and seedling rot, as well as the fumonisin contamination of kernels. Plant lipoxygenases (LOX) synthesize oxylipins that play a crucial role in the regulation of defense mechanisms against pathogens and influence the outcome of pathogenesis. To better uncover the role of these signaling molecules in maize resistance against F. verticillioides, the functional characterization of the 9-LOX gene, ZmLOX4, was carried out in this study by employing mutants carrying Mu insertions in this gene (named as UFMulox4). In this regard, the genotyping of five UFMulox4 identified the mutant UFMu10924 as the only one having an insertion in the coding region of the gene. The impact of ZmLOX4 mutagenesis on kernel defense against F. verticillioides and fumonisin accumulation were investigated, resulting in an increased fungal susceptibility compared to the inbred lines W22 and Tzi18. Moreover, the expression of most of the genes involved in the LOX, jasmonic acid (JA) and green leaf volatiles (GLV) pathways, as well as LOX enzymatic activity, decreased or were unaffected by fungal inoculation in the mutant UFMu10924. These results confirm the strategic role of ZmLOX4 in controlling defense against F. verticillioides and its influence on the expression of several LOX, JA and GLV genes.

Pre-Harvest Modelling and Mitigation of Aflatoxins in Maize in a Changing Climatic Environment-A Review

Aflatoxins (AFs) are harmful secondary metabolites produced by various moulds, among which Aspergillus flavus is the major AF-producer fungus. These mycotoxins have carcinogenic or acute toxigenic effects on both humans and food producing animals and, therefore, the health risks and also the potential economic damages mounted by them have led to legal restrictions, and several countries have set maximum allowable limits for AF contaminations in food and feed. While colonization of food and feed and AF production by A. flavus are highly supported by the climatic conditions in tropical and subtropical geographic regions, countries in the temperate climate zones are also increasingly exposed to AF-derived health risks due to climate change. In the present study, we have reviewed the available mathematical models as risk assessment tools to predict the possibility of A. flavus infection and levels of AF contaminations in maize in a changing climatic environment. After highlighting the benefits and possible future improvements of these models, we summarize the current agricultural practices used to prevent or, at least, mitigate the deleterious consequences of AF contaminations.

The fate of mycotoxins during secondary food processing of maize for human consumption

Mycotoxins are naturally occurring fungal metabolites that are associated with health hazards and are widespread in cereals including maize. The most common mycotoxins in maize that occur at relatively high levels are fumonisins (FBs), zearalenone, and aflatoxins; furthermore, other mycotoxins such as deoxynivalenol and ochratoxin A are frequently present in maize. For these toxins, maximum levels are laid down in the European Union (EU) for maize raw materials and maize-based foods. The current review article gives a comprehensive overview on the different mycotoxins (including mycotoxins not regulated by EU law) and their fate during secondary processing of maize, based on the data published in the scientific literature. Furthermore, potential compliance with the EU maximum levels is discussed where appropriate. In general, secondary processing can impact mycotoxins in various ways. Besides changes in mycotoxin levels due to fractionation, dilution, and/or concentration, mycotoxins can be affected in their chemical structure (causing degradation or modification) or be released from or bound to matrix components. In the current review, a special focus is set on the effect on mycotoxins caused by different heat treatments, namely, baking, roasting, frying, (pressure) cooking, and extrusion cooking. Production processes involving multiple heat treatments are exemplified with the cornflakes production. For that, potential compliance with FB maximum levels was assessed. Moreover, effects of fermentation of maize matrices and production of maize germ oil are covered by this review.

Awareness of mycotoxins and occurrence of aflatoxins in poultry feeds and feed ingredients in selected regions of Uganda

Background

Aflatoxins are a category of mycotoxins produced by certain molds naturally occurring as food and feed contaminants with toxic effects to both animals and humans. In Uganda, previous studies on aflatoxins mainly considered human foods, but scarce information exists for animal feeds. The study aimed at establishing the current status of aflatoxins contamination of poultry feeds, level of awareness and the existing technological challenges and innovations to mycotoxin control in Uganda.

Method

Mycotoxin awareness, predisposing factors and existing strategies for managing mycotoxin contamination were investigated through focus group discussions and questionnaires with selected processors and farmers. Poultry feed and feed ingredient samples were collected and analyzed for total aflatoxins using VICAM Fluorimeter procedure for foods and animal feeds.

Results

Majority of the farmers and processors (> 50%) had limited knowledge about aflatoxins; contamination predisposing factors; dangers to animals and humans; and mitigation strategies. The study further revealed poor feed and feed ingredients handling and storage practices that predispose to mold/aflatoxin contamination. Forty feed samples from feed processing plants had aflatoxins in the range 7.5 ± 0.71 to 393.5 ± 19.09 parts per billion (ppb) with only twelve samples being within the generally acceptable limits of 20 ppb as recommended by the Food and Agricultural Organisation and the United States Federal Department of Agriculture. Additionally, all 27 feed samples obtained from the farmers had aflatoxins in the range of 19.0 ± 1.41 to 188.5 ± 2.12 ppb and were above the acceptable limit. Generally, broiler feeds were the most contaminated with aflatoxins. Of the feed ingredients tested, silver fish (Rastrineobola argentea, locally known as “mukene”) had the least concentration (8.7 ± 3.18 ppb) of aflatoxins while maize bran had the highest level of contamination, 103.3 ± 22.98 ppb.

Conclusion

Given the lack of awareness and the high prevalence of aflatoxin contaminated poultry feeds and feed ingredients in Uganda, a higher risk for carryover of the toxins into the animal products for human consumption exist. Hence, the need for comprehensive establishment of the prevention, control and surveillance strategies for reducing mycotoxins in foods.

Use of Competitive Filamentous Fungi as an Alternative Approach for Mycotoxin Risk Reduction in Staple Cereals: State of Art and Future Perspectives

Among plant fungal diseases, those affecting cereals represent a huge problem in terms of food security and safety. Cereals, such as maize and wheat, are very often targets of mycotoxigenic fungi. The limited availability of chemical plant protection products and physical methods to control mycotoxigenic fungi and to reduce food and feed mycotoxin contamination fosters alternative approaches, such as the use of beneficial fungi as an active ingredient of biological control products. Competitive interactions, including both exploitation and interference competition, between pathogenic and beneficial fungi, are generally recognized as mechanisms to control plant pathogens populations and to manage plant diseases. In the present review, two examples concerning the use of competitive beneficial filamentous fungi for the management of cereal diseases are discussed. The authors retrace the history of the well-established use of non-aflatoxigenic isolates of Aspergillus flavus to prevent aflatoxin contamination in maize and give an overview of the potential use of competitive beneficial filamentous fungi to manage Fusarium Head Blight on wheat and mitigate fusaria toxin contamination. Although important steps have been made towards the development of microorganisms as active ingredients of plant protection products, a reasoned revision of the registration rules is needed to significantly reduce the chemical based plant protection products in agriculture.