Rudolf Krska, Eskola Mari, Berthiller Franz, Nielen Michel, Oswald Isabelle,Elliott Chris, McNerney Oonagh, BotanaLuis Miguel, Malachová Alexandra

(1. Institute of Bioanalytics and Agro-Metabolomics, Department of Agrobiotechnology (IFA-Tulln),University of Natural Resources and Life Sciences, Vienna (BOKU), 3430 Tulln, Austria; 2. Institute for Global Food Security, School of Biological Sciences, Queen’s University Belfast, Belfast BT9 5DL,Northern Ireland, UK; 3. Medfiles Ltd, Rajatorpantie 41 C, 01640 Vantaa, Finland; 4. Laboratory of Organic Chemistry, Wageningen University, Wageningen, Netherlands; 5. Toxalim (Research Center in Food Toxicology), University of Toulouse, F-31027 Toulouse cedex 3, France; 6. IRIS Technology Solutions S.L., 08940 Cornellà de Llobregat, Spain; 7. Departamento de Farmacología, Facultad de Veterinaria, Universidade de Santiago de Compostela, Lugo, 27002 Spain; 8. FFoQSI GmbH, FFoQSI Austrian Competence Centre for Feed & Food Quality, Safety and Innovation, 3430 Tulln, Austria)

Abstract: There is a massive and urgent need to ensure safety and security of the food supply of the world´s growing population. However, global agriculture and food industries continue to be vulnerable to problems of contamination with biotoxins produced by plants, algae and particularly by fungi; with global warming and extreme weather events making the occurrence of these toxic metabolites even more unpredictable. In this paper we summarize the multidisciplinary, multi-sectoral complementary competencies needed to innovate in various scientific fields and approaches, so strongly needed to develop improved early warning,monitoring and toxicity assessment of biotoxins in the food and feed chain. These include big data approaches using satellite and drone images, portable monitoring devices and (combined) toxicity testing of(emerging) biotoxins using proteomics and transcriptomics.

Key words: food and feed safety; prediction; early warning; phycotoxins; mycotoxins; plant toxins

1. BIOTOXINS IN THE FOOD AND FEED CHAIN

Contaminations of food, pet food and animal feed with poisonous natural substances - so called biotoxins, which are produced by living organisms -are a major concern for stakeholders along the whole food and feed chain, including the aquaculture industry. These organisms include fungi, plants and algae, which are known to produce mycotoxins,plant toxins and phycotoxins, respectively. These harmful secondary metabolites are typically produced by the defence mechanisms of the organism as response to changes in their growing environment.Biotoxins usually occur in complex mixtures in various aquaculture and agricultural products, including food for human and animal consumption. Biotoxins are toxic to humans and animals and can induce a wide range of acute and chronic adverse health effects such as DNA damage, cancer formation,damages to the central nervous system, the kidney,the liver and the immune system, can lead to detrimental reproductive developmental effects and even death[1]. These toxins not only pose a health risk to humans, but also to farm and companion animals adversely affecting their health, welfare and productivity. Based on the data from the European Rapid Alert System for Food and Feed (RASFF)mycotoxin contamination exceeding the maximum limits are the major reason for rejections of food products at the external EU border[2]. Currently, an average European consumer is exposed to several biotoxins daily at levels which are of potential health concern, especially for children, as concluded by the European Food Safety Authority (EFSA) in recent scientific opinions[3-5].

Despite huge research investments, prevention and control (early warning systems, monitoring and toxicological assessment) of these toxic secondary metabolites remains difficult and the agriculture,food and feed industries continue to be vulnerable to problems of contamination with biotoxins. In addition, climate change is increasingly affecting the global biotoxin map[6-7]. The changing environment impacts the interaction between organisms, can promote the growth of certain toxin-producing organisms and modify the type and quantity of resulting toxins. The impact of biotoxins is expected to increase in agriculture and aquaculture, as well as on humans and animals due to climate change.Among the reasons for that is that temperature,relative humidity and crop damage by pests – all factors which contribute to fungal spread and mycotoxin production - are directly or indirectly affected by climate change. For instance, severe droughts in Serbia resulted in 70% of the maize crops being contaminated with aflatoxins (AFs) in 2013[8]and the use of this maize to feed dairy cattle led to the high levels of aflatoxin M1found in milk,up to twice the EU legal limit. Similar challenges have been reported for plant toxins and phycotoxins.In addition to changes in the global occurrence of phycotoxins, puffer fish carrying the deadly tetrodotoxin (TTX) invading European waters is one such recent example[9]. Approximately 60,000 human intoxications yearly with overall mortality of approximately 1.5% are related to toxins produced by algae[10]. EFSA, at the request of the European Commission has developed a series of scientific opinions concerning marine biotoxins, including groups currently regulated in the EU legislation and emerging toxins[11], which will pose a new global health risk due to the spread of algae and prevalence of these toxin groups in new geographical regions.EFSA also concluded that the exposure to plant toxins, in particular pyrrolizidine alkaloids and tropane alkaloids, are of human health concern and pose a health risk, particularly for children and high consumers of certain plant-based food items[3].

Exceptional weather conditions in Europe during recent years have already reduced food and feed quality and security, due to the high and especially unexpected occurrence of biotoxins[12],which also led to severe economic damage. With annual losses due to mycotoxins estimated by the EU to be at a level of 5%～10%, this equates to 1.2～2.4 billion Euro in lost income for the major food crop wheat alone. At the same time the adverse impacts of biotoxins are an important risk factor to food security for the consumers for a world population estimated to pass 9 billion by 2050[13].With regard to the occurrence of marine biotoxins,toxic harmful algal blooms (HABs) can seriously affect the production of shellfish and fish in sea farms when they occur. Using the shellfish aquaculture data of the Food and Agriculture Organization of the United Nations (FAO), the estimated negative economic impact of HABs in 2000—2009 was 75.4 million dollars in Europe, or 7% of the total shellfish sector value. For the European mussel industry alone, the losses during the period 2000—2009 were 32 million dollars[14].Furthermore, there is increasing evidence linking climate change with emerging phycotoxins in Europe. This is mainly caused by the presence of microalgal species that move North with warming sea waters[15]. The lethal TTX is already a matter of legislative effort in Europe, but the fact that it is associated with bacteria and microalgae makes its monitoring especially complex. There is recent evidence proving chronic levels of low doses of TTX dangerous. Along with TTX, the highly toxic palytoxins and ostreocins are also a matter of concern, especially in Southern Europe. In Mediterranean countries their expansion in the past few years has been extraordinary, and they are now present from Greece to South Portugal. Also, the expansion of ciguatoxins (CTX) in Europe is of serious concern, caused by marine dinoflagellates(Gambierdiscusspp.) moving up North due to warming waters[16]. During the most dramatic periods the fish farmers may lose their entire investment that year due to mass kills of fish caused by ichtyotoxic algal blooms. Consequently, the agricultural,aquacultural and food industries as well as all citizens who need to be fed in Europe and around the world with nutritious, safe and affordable food in a sustainable economic environment are under some form of threat from biotoxins. There is an unprecedented demand for multidisciplinary knowledge and action needed to develop integrated innovative approaches to tackle food safety and security issues -and especially in relation to the presence of biotoxins.These challenges need to be faced with cutting edge research and multidisciplinary training of researchers in innovative and integrated control and reduction of biotoxins in the food and feed chain.

2. ADVANCED TECHNOLOGY FOR EARLY WARNING AND MONITORING OF BIOTOXINS

Forecasting, early-warning and monitoring of biotoxins is of utmost importance to avoid human and animal health consequences as well as economic losses. There is a great need to develop new techniques and deploy these for early warning and monitoring of a range of biotoxins in agriculture and aquaculture.Novel warning system can benefit from cuttingedge technology and expertise in the application of satellite images, on-site testing approaches, and big data handling and management. For on-site measurements and remote sensing of biotoxins, such as mycotoxins and plant toxins, or biotoxin formation aspects, tools are needed for taking appropriate risk mitigation measures.

2.1 Mycotoxins

The accurate forecasting of the (increased)occurrence of mycotoxins, such as the potent liver carcinogen aflatoxin B1(AFB1) in crops, could provide early warning to farmers and to the whole food supply system, to take timely and appropriate measures. A need for such a technology is evident when severe disruptions in the maize production occur due to exceptional weather conditions as happened in Balkan area in 2012 and in France in 2013, which led to a temporary increase of legislative maximum levels and a request to the EC for such a derogation to secure the food supply[17]. The occurrence of aflatoxins and in particular of AFB1in the peanut supply chain poses a clear risk for the Europeans[18]. RASFF notifications of over 400 border rejections have been reported in 2018[19].China is among the top three largest suppliers of peanuts to the EU, accounting for a share of 12% in 2017 according to Eurostat. An integrated approach for detecting and collecting relevant data along the entire peanut chain would allow better risk assessment of aflatoxin occurrence and facilitate better decision making at earlier stages. Modern technology allows that data from multiple sources incl. freely available satellite images, field and weather records, thermal and hyperspectral imagery to develop a big data model. The available high resolution and the large number of satellite images[20]offer a massive influx of spectral indices which properties are a good basis for the prediction of mycotoxin levels in primary production. State-of- the-art algorithms for machine learning techniques need to be further developed,though, to extract hidden dependencies between diverse input data and the presence of e.g. mycotoxins in grains or peanuts. Research on relationships between plant’s spectral response and the presence of mycotoxins at large scale using spectral bands available from satellites, is still lacking. Satellitebased solutions to predict the (increased)occurrence of mycotoxins in crops could allow the mapping of large agricultural sites and the precise control of mycotoxin contamination. Identification of contaminated areas would enable the tailor-made application of post-harvest techniques to reduce mycotoxins.

To predict aflatoxin formation during transport and storage in large peanut silos innovative real-time post-harvest environmental monitoring systems[21]such as the prototype developed within the EU funded project MyToolBox[22]need to be further optimised and applied. Such a prediction is based on modelling of CO2production during the growth ofAspergillus flavus[21].An EC DG Health report[23]recommends that official controls of peanuts are to be carried out at all stages of production,processing and export. However, the confirmatory analytical methods typically used (liquid chromatography coupled with tandem mass spectrometry –HPLC-MS/MS) are tedious and the turnaround time for mycotoxin analysis is typically 3 to 5 days.Therefore, rapid semi-quantitative tests based on lateral flow immunochromatographic assay (LFIA)for fast screening are often used first. The results can be obtained in 5 min[24]. Another option for rapid testing are platforms, which still need to be advanced. These include mid-infrared-attenuated total reflection-fourier transform infrared spectroscopic approaches, affordable handheld near infrared portable instruements, as well as portable instruments operating in the fluorescence-VIS spectral range, as options for a flexible in-field/in-line detection of aflatoxins on peanuts early in the production chain[25].Such solutions could be very beneficial for the internal control, which allow the operators to carry out cost-efficient, frequent and rapid testing which is particular of relevance for heterogeneous aflatoxin contamination as encountered in peanut lots. Due to the relatively low concentrations of aflatoxins to be detected portable mass spectrometric approaches show the greatest potential for providing the sensitivity and fit-for-purpose selectivity for accurate on-site testing of the quantification of aflatoxins in peanuts and peanut products for official control purposes. Such innovative analytical solutions would enhance the monitoring capacity at all stages in the peanut supply chain and would consequently provide the basis for quicker and better decisions.The data generated through the different applications can in the future be transferred and stored in a digital platform and made available for automated risk assessment and the transmission of risk profiles to end-user as depicted in Figure 1.

Fig.1 High level concept diagram for the control and reduction of aflatoxin contamination in peanuts across continents

2.2 Phycotoxins

Under optimal growth conditions, some microalgal species can generate HABs in marine and even freshwater environments. A driving force for more frequently occurring HABs is an overabundance of fertilizer application combined with more intense precipitation, which together lead to increased eutrophication in waterbodies[26].Consequently, high levels of phycotoxins can cause mass mortality of fish. When ingested, marine biotoxins are also highly acutely toxic to humans,adverse effects ranging from mild tingling sensation around the mouth to death. Different countries have established standards about the total amount of phycotoxins that can be present in seafood products and also about the testing methods for detecting these molecules. Legal requirements for paralytic shellfish toxins [usually named paralytic shell-fish poisoning (PSP)], amnesic shellfish toxins [amnesic shellfish poisoning (ASP)], lipophilic toxins, CTXs,brevetoxins or TTXs have been set around the world[27]. The European legislation establishes a maximum level of marine toxins in bivalve molluscs,echinoderms, tunicates and marine gastropods[28].Typical analytical methods for control of marine toxins are based on HPLC/FLD, HPLC/UV and HPLC-MS/MS nowadays. The mice bioassay has been considered for many years as a reference method for monitoring of phycotoxins. However, in accordance with animal health protection, all possible replacement, refinement and reduction on animals must be taken into account when using animal assays.Therefore, alternative methods are being developed,for instance biological assays[29].

Very little information is available on those marine biotoxins which cause massive fish mortalities of wild and farmed fish (ichtyotoxins)[30]. The research of ichtyotoxins is majorly prevented by the poor availability of pure substances and their unknown chemical structures[31]. In order to prevent algaeinduced contamination of fish and shellfish algal bloom forecasting model has been developed within EU-funded project ASIMUTH. The project uses satellite data provided by the marine service of the EU’s Copernicus Earth observation programme. This service offers daily information on a range of marine parameters, including chlorophyll and phytoplankton concentrations. ASIMUTH integrates this satellite data with sea-based monitoring and regional knowledge of current water column and surface current behaviour. This not only facilitates accurate forecasts of the build-up of harmful algal blooms,but also their likely movements over the coming days.This gives to fish farmers a vital early-warning system and allows them to harvest earlier, before the blooms arrive[32].

2.3 Plant toxins

On-site testing of mycotoxins and plant toxins at different stages along the entire food chain, in the field, at border inspection, in industry and retail, is still needed to allow a more efficient and effective control and early warning of quality and safety parameters of their emergence or presence in the food chain. Early discovery provides better opportunities for (re-)use of contaminated crop batches e.g. for animal feed instead of ending up to biofuels. Due to the relatively low concentrations and chemical diversity of mycotoxins and plant toxins, portable mass spectrometric approaches are potentially capable of providing the versatility,sensitivity and fit-for-purpose selectivity for on-site testing of these toxins. However, neither the heavier transportable commercial MS systems, nor the lightweight portable MS prototypes, have been studied and applied to the analysis of mycotoxins or plant toxins. Initial work showed that biotoxins can be detected at relevant levels using lab-based ambient MS approaches without chromatographic separation[33-35]. Within two recent studies with high potential for future portable MS applications, direct MS was optimized, validated and benchmarked for the most prevalentFusariummycotoxin deoxynivalenol[36-37]. Such highly innovative solutions in analytical chemistry would multiply the monitoring capacity at all stages in the food chain and consequently enhance food and feed safety and security.

3. INNOVATIVE APPROACHES TO ASSESS THE TOXICITY OF BIOTOXINS

There is a great need for the creation of innovative approaches to test the toxicity of different classes of biotoxins including mycotoxins and shellfish toxins, in particular when present as mixtures. The lack of toxicity data on mixtures of co-occurring mycotoxins and mixtures of shellfish toxins has prevented EFSA from assessing reliably health risks from exposure to mixtures of these biotoxins. Toxicity data, particularly chronic toxicity data, lack in many areas: e.g. the mode of action(MoA), toxicokinetics, toxic adverse effects in humans and animals, including combined effects of mixtures, and interactions of these toxins at organ and cellular level in mixtures. More research is needed to provide answers to these major data demands[38]. A tiered strategy for risk assessment of mixtures of multiple chemicals derived from multiple sources across different life stages has been studied in EuroMix EU funded project[39].

Chronic exposure toFusariummycotoxins at low to moderate doses, has likely a significant long-term impact on human and animal health, with impaired growth and development or immune dysfunction. However, the associated diseases are often subclinical and/or symptoms are very unspecific. Toxicity of so-called emergingFusariumtoxins in humans and animals is currently poorly known as concluded by EFSA[4-5]. The term“emerging” mycotoxins is used for mycotoxins which are neither routinely determined, nor legislatively regulated; however, the evidence of their incidence is rapidly increasing[40]. As they co-occur in mixtures with regulated mycotoxins, it is highly important to know the nature of their toxic interactions with the regulated ones. Studies to be carried out on intestine barrier function towards emerging mycotoxins using intestinal cell lines and explant cultures employing toxicity assays as well as state-of-the-art transcriptomic and proteomic approaches would give novel data on toxic effects, MoAs and interactions of emergingFusariumtoxins. There are several mathematical and statistical models used for a determination whether the exposure to a mixture of mycotoxins would result in synergistic, additive or antagonistic effects on the cell cultures[41].

The use of biomarkers of exposure does not reflect the effects of biotoxins in the body, but biomarkers of effect, such as alterations at the cellular level and early/low dose-associated pathological changes, would provide powerful tools to evaluate toxic effects. As these biomarkers are typically unspecifically related to the toxin, they have an advantage of having a greater potential to reflect exposures to toxin-mixtures and repeated exposures over time, and could help to identify both active substances of the mixtures/co-exposure and health consequences from the exposure to mixtures[42].Currently, specific and validated biomarkers of effects do not exist for most biotoxins including mycotoxins. Novel approaches could take advantage fromex vivostudies and multi-omics techniques to reveal molecular biomarkers of effects for regulatedFusariumtoxins. Validated biomarkers of effects would also provide unique information on combined toxic effects of toxin mixtures at low doses to enable risk assessors to reliably assess the chronic risks for human and animal health from the exposure to mixtures of biotoxins.

4. CLIMATE CHANGE AND BIOTOXINS

The FAO very recently assessed and summarized the impact of climate change on food safety[26]; in particular considering foodborne pathogens and parasites, pesticides, heavy metals, but also harmful algal blooms and mycotoxins. Already in the executive summary the FAO states that: “Today, we know that increasing temperatures, ocean warming and acidification, severe droughts and wildfires,untimely heavy precipitation and acid rain, meltingglaciers and rising sea levels and amplification of extreme weather events are causing unprecedented damage to our food systems.” Addressing the consequences of anthropogenic changes to the earth´s climate is among the biggest challenges for humankind[43]. Its largest effect is unprecedented warming of the climate system, which in turn elicits more frequent and more extreme weather events including heatwaves and measurable changes to rainfall distribution and intensity. This affects arable land and will lead to adaptation in agricultural practises and the range of crops that can be grown.Both, climate effects and how we change agriculture in response to it, will influence ranges and incidences of fungal pathogens each adapted to distinct climatic conditions. Changing weather conditions are the main driving variable acting on fungi, their metabolism and the cross-talk with the host plant. Therefore,climate variability and related uncertainties, commonly stressed on a large scale, are not only a matter for policymakers, but also for farmers facing every day the impact on fungi and mycotoxin occurrence.The scientific community is not well prepared for this challenge.

The Joint Programming Initiative on Agriculture,Food Security and Climate Change (FACCE-JPI),launched by the Council of the European Union stated: “Addressing food safety with respect to climate change is also an important component of food security. Reducing and controlling these risks is relevant for both public health and economic reasons[44]. The need for mitigating the issue of biotoxins becomes particularly evident in the area of mycotoxins: a very recent review revealed that up to 80% of food items of plant origin worldwide are estimated to be contaminated with these toxic secondary fungal metabolites[45].

5. CONCLUSIONS

A series of measures is needed that can be exploited to establish innovative integrated solutions to the predicted challenges in the control,reduction and management of (co-occurring)biotoxins in foods and feeds to protect human and animal health in a world of changing climate. These include improved early warning, monitoring and toxicity assessment of co-occurring biotoxins in food and feed. A broad spectrum of multidisciplinary scientific skills at top scientific level is required in the various sub-disciplines within agricultural, food science, analytical chemistry, omics-sciences and toxicology which are needed to achieve this goal. In the future more cutting edge research, such as big data approaches using drone images and the use of proteomics, transcriptomics and metabolomics tools,need to be further developed to control and reduce biotoxins from field-to-fork by considering also aquacultural production systems and minimising food and feed losses. A unique workshop held in Beijing, 2019, within the frame of the EU-China project MyToolBox[17]also testified to the fact that there is still a great need for smart and integrated strategies to tackle the mycotoxin issue to achieve safer food and feed products and to minimize losses and export rejections[45].