Mycotoxin testing – From multi-toxin analysis to metabolomics

Introduction

The occurrence of fungal and subsequently, mycotoxin contamination in various crops is of major concern since it has significant implications for food and feed safety, food security and international trade. The European Union (EU) annually produces about 133 million tons (MT) wheat (~M€ 29038), 68 MT maize (~M€ 13571) and 8 MT oats (~M€ 1543) (EUROSTAT online, 2016a, EUROSTAT online, 2016b). The Food and Agricultural Organization (FAO) has estimated that about 25% of the world’s crops are affected by mycotoxins each year. However, results using latest state-of-the-art multi-analyte methods show that almost 100% of the crops are contaminated with one or more mycotoxins (Berthiller et al., 2014). With annual losses due to mycotoxins estimated at 5-10%, this equates to €1.2-2.4 billion in lost income for wheat alone, a reduction in these losses of only 1% would save €12-24 Mio. Moreover, with “one in eight persons worldwide suffering from chronic undernourishment” (Schmale and Munkvold, 2009); there is clearly a moral obligation to curb these significant food and feed losses.

Grain and foods based on these grain (e.g. pasta, bread, bakery products) account for the largest contribution to mycotoxin exposure in all age classes of the EU population, in particular due to the mycotoxins produced by Fusarium spp. (deoxynivalenol (DON), T-2/HT-2 toxins, zearalenone, ochratoxin A and fumonisins). In 2013, the EU’s RASFF (EU, 2014) (Rapid Alert System for Food and Feed) showed that of the total rejections at the EU borders, 18.7% were due to mycotoxin contamination exceeding the EU legislative limits (All about feed, 2014). Exposure of livestock to mycotoxins may lead to substantial economic losses (Wu and Munkvold, 2008). While mycotoxicoses in farm animals are regularly reported (Jovanović et al., 2015), it is difficult to get an accurate estimate of the burden of market loss associated with animal health and productivity, and loss of animal feed. Up to 139 different fungal metabolites were identified in a recent feed survey which has again demonstrated the importance of multi-analyte methods in mycotoxin analysis (Streit et al., 2013). There is a pressing need to mobilize the wealth of knowledge that exists from the mycotoxin research conducted in Europe and internationally to reduce mycotoxins along the entire food and feed chain. (Multi-)Mycotoxin-testing plays a key role in determining the natural occurrence of mycotoxins in food and feed and the exposure of humans and animals to these toxic secondary fungal metabolites. In addition, accurate quantification of mycotoxins is crucial to evaluate the success of measures to reduce the mycotoxin levels along the food and feed chains. To provide a means for the basic understanding of biological systems, metabolomics has been introduced to obtain a holistic picture of all the secondary metabolites of plants and fungi which are involved in the interaction of different organisms.

Mycotoxin occurrence and detection

Common mycotoxins include trichothecenes (such as deoxynivalenol, fumonisins, zearalenone, ochratoxin A and aflatoxins). The potential health risks to animals and humans posed by food- and feed-borne mycotoxin intoxication have been recognized by national and international institutions and organizations such as the European Commission (EC), the US Food and Drug Administration (FDA), the World Health Organization (WHO) and the FAO which have addressed this problem by adopting regulatory limits for major mycotoxin classes and selected individual mycotoxins (MYCOTOXINSANNUALLY, 2015). The necessity to obey these regulatory limits has prompted the development of appropriate sampling plans and validated analytical methods for the determination of mycotoxins in various food and feed commodities which will lead to improved exposure estimates and risk assessment strategies with respect to these toxic secondary metabolites. However, sampling is still the major issue in mycotoxin analysis due to the sometimes very heterogeneous distribution of the toxic metabolites in agricultural commodities and products intended for human and animal consumption. The proper selection of a sample from the lot and the subsequent steps undertaken to produce a portion for the determination of the mycotoxin of interest is essential for the production of sound analytical data.

The chemical diversity of mycotoxins and their occurrence in a wide range of agricultural commodities and foods poses a great challenge for sample separation and methods of analysis. In order to deal with the increasing demand for mycotoxin analyses, rapid screening methods for single mycotoxins or whole mycotoxin classes have been developed, which are mainly based on immunochemical techniques. Highly sophisticated multi-analyte LC-MS based methods have become available which enable the simultaneous quantification of more than 300 metabolites of fungi, plants and bacteria (Malachova et al., 2015) which allow for a comprehensive assessment of the range of mycotoxins humans and animals are exposed to. Despite the enormous progress in mycotoxin analysis, major challenges remain. Among these are the determination of conjugated (masked) mycotoxins, the matrix effects observed when performing LC-MS(/MS) measurements, the lack of certified reference materials and the need for reliable rapid methods particularly for the simultaneous quantification of mycotoxins in foods and feeds. Another challenge is the increasing number of unexpected findings of mycotoxins in food and feed commodities, e.g. due to extreme weather conditions as a result of climate change. Such unexpected mycotoxins are referred to as so-called emerging toxins, which include enniatins, beauvericin, moniliformin, fusaproliferin and Alternaria metabolites. Moreover, the impact of these cocktails of secondary metabolites in food and feed and the potential additive or synergistic effects of these substances are still mostly unknown. But even well known mycotoxins, such as aflatoxins, can occur unexpectedly due to global warming. Despite huge research investments, prevention and control of these toxic secondary metabolites remain difficult and the agriculture and food industries continue to be vulnerable to problems of contamination: In February-March 2013 Romania, Serbia and Croatia, reported aflatoxin M1 contamination of milk. Severe droughts in Serbia in 2012 resulted in 70% of the maize crop being contaminated with aflatoxins (All about feed, 2013). Use of this maize to feed dairy cattle led to high levels of aflatoxin M1 in milk, two times higher than the EU legal limit. The milk scandal was fueled when the permitted level of aflatoxin M1 in milk in Serbia was temporarily raised to 0.5 µg/L milk, 10 times the EU legal limit. In recent years, a dramatic increase in DON contamination in Nordic countries has been observed.

Obviously, farmers around the world and also in Asia have to prepare themselves for such unexpected occurrences of mycotoxins as climate change is still ongoing. Over the last decades, in mycotoxin analysis, a shift from thin layer, ELISA and LC based methods (Berthiller et al., 2015) can be observed. Until recently, most of the available analytical methods (e.g. HPLC-UV/FLD) for the determination of these toxic metabolites only covered single mycotoxin classes (e.g. aflatoxins, type-B trichothecenes or fumonisins). In the meanwhile, mass spectrometry based analytical methods have been the key for the determination of a variety of mycotoxins and their metabolites in plants and foods and for the investigation of the metabolism of these toxic compounds in body fluids such as serum and urine. One example is a multi-analyte LC-MS/MS (liquid chromatography - tandem mass spectrometry) method which has recently been developed by us and which is capable of determining 380 fungal, bacterial and plant metabolites, respectively, in cultures, cereals, food and feed products. LC-MS-based screening has also been playing a vital role in the discovery of novel mycotoxin conjugates so called “masked” - forms of mycotoxins in the past, and it is believed that this will also continue in the future. This unique mycotoxin LC-MS/MS method has also been employed in BIOMIN’s Spectrum 380® screening program. For the first time, surveys can be conducted routinely which include concentration levels of more than 380 mycotoxins and fungal metabolites. Spectrum 380® employs latest state-of-the art LC-MS/MS and revealed that a typical agricultural commodity contains on average 30 different mycotoxin metabolites.

Metabolomics of plant-fungi interactions

Metabolomics has emerged as the latest of the so-called “omics” disciplines and shows great potential to determine thousands of metabolites at once over a wide range of concentrations. Metabolomics or metabolome analysis has been introduced to designate the set of all low-molecular-mass compounds, i.e. metabolites, synthesized by an organism (Villas-Bôas et al., 2005). Metabolites are the products of cellular regulatory processes, and their levels can be regarded as the ultimate response of biological systems to genetic or environmental changes (Fien, 2002). At the Center for Analytical Chemistry of the IFA-Tulln, both metabolomics as well as metabolite profiling is carried out by means of modern high resolution mass spectrometric techniques (LC-HR-MS/MS). In contrast to mere metabolite profiling, metabolomics always shows fitness for a functional genomics context. All studies in functional genomics aim to narrow the gap between gene sequence and gene function to yield new insights into the behavior of biological systems. Metabolite profiling represents an extremely useful tool that finds applications in many aspects of drug discovery, food safety issues and disorders of cells and organisms. One example is the multi-analyte method which has recently been developed by researchers of the IFA-Tulln capable of determining 39 and 91 fungal metabolites, respectively, in cultures and grains (Sulyok et al., 2006).

Metabolomic analyses are increasingly being performed, and both profiling as well as targeted analyses in which a defined list of metabolites is analyzed, are currently applied. The IFA-Tulln has both, the expertise as well as the required latest state-of-the-art analytical instrumentation available to successfully perform high-level research in this important field of life-science. The ultimate goal of our metabolomics research is to understand and to predict the behavior of complex biological systems such as plants and microbes. An example for successful research of the IFA-Tulln in this area was the finding that the ability of wheat to detoxify the relevant mycotoxin DON into its non-toxic glucosidic form can directly be linked to recently identified resistance genes (Lemmens et al., 2005).

By non-directional metabolomics-method, the untargeted detection of yet unknown secondary metabolites of mildew and metabolic products of mold-resistant plants becomes feasible. With such a method, the plants are treated with native and 13C-marked mycotoxins, whose metabolic products are then traced by high-resolution mass spectrometry (Kluger et al., 2013). After measurement of biological/food samples treated with a 1+1 mixture of labelled and non-labeled precursors, labeling-specific isotopic patterns can be reliably and automatically detected by means of the novel software tool (“MetExtract”), which was developed by C. Bueschl (Bueschl et al., 2013). In a preliminary study, the great potential of the presented approach is further underlined by the successful and automated detection of eight novel plant-derived biotransformation products of the most prevalent Fusarium mycotoxin DON. By applying this method, Bernhard Kluger and colleagues were able to identify DON-3-glucoside (see Figure 1) and DON-S-, which was a new form of conjugated mycotoxins (Kluger et al., 2015). The relevance of the novel glutathione metabolites for food and feed safety is still being investigated.

In the future, integrated approaches will increasingly be developed to reduce mycotoxins along the whole food and feed chain, which combines effective pre- and post-harvest measures with proper monitoring techniques for the determination of mycotoxins (www.mytoolbox.eu). Such integrated approaches will also include metabolomics studies to reveal the fundamental biological processes behind mycotoxin production and its reduction.