Natural Occurrence of Nivalenol, Deoxynivalenol, and Deoxynivalenol-3-Glucoside in Polish Winter Wheat

The presence of mycotoxins in cereal grain is a very important food safety factor. The occurrence of “masked” mycotoxins has been intensively investigated in recent years. In this study, the occurrence of nivalenol, deoxynivalenol-3-glucoside, and deoxynivalenol in 92 samples of winter wheat from Polish cultivars was determined. The frequency of the occurrence of deoxynivalenol and nivalenol in the samples was 83% and 70%, respectively. The average content of the analytes was: for deoxynivalenol 140.2 µg/kg (10.5–1265.4 µg/kg), for nivalenol 35.0 µg/kg (5.1–372.5 µg/kg). Deoxynivalenol-3-glucoside, the formation of which is connected with the biotransformation pathway in plants, was present in 27% of tested wheat samples; its average content was 41.9 µg/kg (15.8–137.5 µg/kg). The relative content of deoxynivalenol-3-glucoside (DON-3G) compared to deoxynivalenol (DON) in positive samples was 4–37%. Despite the high frequency of occurrence of these mycotoxins, the quality of wheat from the 2016 season was good. The maximum content of DON, as defined in EU regulations (1250 µg/kg), was exceeded in only one sample. Nevertheless, the presence of a glycosidic derivative of deoxynivalenol can increase the risk to food safety, as it can be hydrolyzed by intestinal microflora.


Introduction
Wheat is a basic ingredient used for food preparation around the world. Poland, with approximately 11 million metric tons cultivated each year, is the fourth largest producer of this crop in the European Union (after France, Germany, and the United Kingdom) and 15th in the world [1]. Cereal plants are vulnerable to infections of pathogenic fungi of the Fusarium genus. Fusarium head blight (FHB) is an infection widespread in wheat-producing countries; it is induced mainly by F. graminearum, however F. culmorum and F. avenaceum can also be dominating species. The optimal conditions for the fungal infection and propagation of these fungi are moderate temperature and high air humidity [2]. The geographic occurrence of F. graminearum and F. culmorum may vary depending on the climatic conditions (temperature and relative air humidity): F. graminearum on the climatic conditions (temperature and relative air humidity): F. graminearum occurs mostly in warmer and more humid regions (e.g., North America, eastern Europe, Australia, southern China), while F. culmorum occurs mainly in colder climatic regions (e.g., Western Europe) [3]. This is not a rule, thus in some regions, sometimes the main FHB-inducing factors were F. culmorum, F. avenaceum, and F. poae [4]. FHB leads to economic losses caused by the decrease in crop yields, but also can have an impact on food safety. This is connected with accumulation of mycotoxins in grains. Nivalenol (NIV) and deoxynivalenol (DON), classified as type B trichotecenes, are fungal metabolites present in agricultural products [5,6].
NIV is not found in food as commonly as DON; however, it demonstrates higher toxicity in animal studies. The LD50 values for DON and NIV in tests in mice were 78 and 39 mg/kg, respectively [7]. The toxicity of NIV is often compared to the toxicity of DON; however, the amount of toxicological data on NIV impact is much lower. On the molecular level DON and NIV, similarly to other trichotecenes, show many limiting effects on the primary eukaryotic cell metabolism including inhibition of protein, DNA, and RNA synthesis [8,9].
DON acts as a virulence factor in the development of FHB and facilitates the spread of fungus from the infection site [10]. After synthesis of this mycotoxin in the infected tissue, DON inhibits protein synthesis, disrupts signal transmission, and eventually causes cell death [11]. Plants possess developed detoxication systems. In the case of cereal plants in the scope of these reactions, there is a group of mycotoxins, which undergo biological modification. The modification is made by two major reactions. In the first stage xenobiotics are oxidized or hydrolyzed; the second stage is a conjugation in which the xenobiotics functional group binds with glucose, malonic acid, or glutathione. In the reaction of the second phase of detoxication deoxynivalenol-3-glucoside (DON-3G), which is the main DON metabolite, may be formed [12,13]. The structures of these compounds are presented in Figure 1. Mycotoxin metabolites that were synthesized in this manner are described as "masked mycotoxins" [14,15]. DON-3G was detected in naturally contaminated wheat and maize for the first time in 2005 [16]. Afterwards, the presence of the DON-3G biosynthesis pathway was confirmed also in durum wheat and barley [17,18]. DON-3G, compared to the basic analogue, exhibits lower synthesis of ribosomal protein of wheat in in vitro conditions [19]. In the biosynthesis of this metabolite, uridine-5'-diphospho-glucuronosyltransferase (UDP-glucosyltransferase) participates, which catalyses the transfer of glucose from UDP-glucose to the hydroxyl group of DON on the third carbon atom. The DON-resistant wheat cultivars are more efficient in conversion of DON to DON-3G than more susceptible cultivars. Thus, it is believed that implementing a detoxication mechanism based on glicosylation in cultivars susceptible to Fusarium will increase wheat resistance to FHB [20]. One of the most recent attempts to increase the resistance was using transgenic wheat containing barley UDP-glucosyltransferase gene (HVUGT13248), which demonstrated a high level of resistance [21]. The majority of the studies on the toxicological features of DON-3G were conducted in vitro. It is still not clear how it impacts living organisms. In general, it is stated that DON-3G is less toxic than its parent toxin. It is resistant to digestion processes in digestive systems and is not absorbed by the intestinal epithelium, although it can by hydrolyzed to DON or deepoxy-deoxynivalenol (DOM-1) by intestinal microflora [22][23][24].   Because of the reasons listed above, knowledge about the natural occurrence of masked mycotoxins in wheat samples is very important from the viewpoint of food safety. In Poland, there are no routine analyses of DON-3G (in mills or elevators). Similarly, due to the lack of legal regulations, there is no obligatory NIV screening. The aim of this study was to evaluate natural occurrence of NIV, DON and DON-3G (a product of DON metabolism in plants) in wheat grain harvested in Poland.

Sample Processing
Wheat samples fortified with analytical standards added to determine method recovery and precision were processed identically to real samples. 2 g of ground wheat grain and 8 mL of de-ionised water put into a 50 mL falcon tube were first homogenized (Unidrive X 1000 homogenizer manufactured by CAT Scientific, Inc., Pase Robles, CA, USA) for 2 min, then centrifuged (MPV, Med. Instruments, Warsaw, Poland) at 10,730 × g for 10 min. 3 mL of the extract dissolved with 3 mL of phosphate buffered saline (PBS) was again centrifuged at 10,730 × g for 10 min. 5 mL of the extract was passed through a DON-NIV (wide-bore, WB) immunoaffinity column (Vicam, Watertown, MA, USA) at a speed of 1-2 drops/s. Next, the column was rinsed with 10 mL of PBS and 10 mL of de-ionised water at a speed of 2-3 drops/s. Analytes washed out of the column with first 0.5 mL of methanol then with 1.5 mL of acetonitrile were collected into a reaction vial. The solvent was evaporated in a stream of nitrogen, and the residues were re-dissolved in 300 µL of 10% acetonitrile solution.
The samples ready for injection were filtered through a nylon syringe filter with 0.45 µm pore diameter. Each sample was analysed twice; two independent repetitions were sufficient since the used immunoassay columns (IAC) columns (Vicam, Watertown, MA, USA) feature very high sample purification degree and repeatability. Chromatograms taken from naturally contaminated wheat samples are shown in Figure 2, respectively, on top of chromatograms of mixtures of NIV, DON and DON-3G standards. Because of the reasons listed above, knowledge about the natural occurrence of masked mycotoxins in wheat samples is very important from the viewpoint of food safety. In Poland, there are no routine analyses of DON-3G (in mills or elevators). Similarly, due to the lack of legal regulations, there is no obligatory NIV screening. The aim of this study was to evaluate natural occurrence of NIV, DON and DON-3G (a product of DON metabolism in plants) in wheat grain harvested in Poland.

Sample Processing
Wheat samples fortified with analytical standards added to determine method recovery and precision were processed identically to real samples. 2 g of ground wheat grain and 8 mL of deionised water put into a 50 mL falcon tube were first homogenized (Unidrive X 1000 homogenizer manufactured by CAT Scientific, Inc., Pase Robles, CA, USA) for 2 min, then centrifuged (MPV, Med. Instruments, Warsaw, Poland) at 10,730 × g for 10 min. 3 mL of the extract dissolved with 3 mL of phosphate buffered saline (PBS) was again centrifuged at 10,730 × g for 10 min. 5 mL of the extract was passed through a DON-NIV (wide-bore, WB) immunoaffinity column (Vicam, Watertown, MA, USA) at a speed of 1-2 drops/s. Next, the column was rinsed with 10 mL of PBS and 10 mL of deionised water at a speed of 2-3 drops/s. Analytes washed out of the column with first 0.5 mL of methanol then with 1.5 mL of acetonitrile were collected into a reaction vial. The solvent was evaporated in a stream of nitrogen, and the residues were re-dissolved in 300 µL of 10% acetonitrile solution.
The samples ready for injection were filtered through a nylon syringe filter with 0.45 µm pore diameter. Each sample was analysed twice; two independent repetitions were sufficient since the used immunoassay columns (IAC) columns (Vicam, Watertown, MA, USA) feature very high sample purification degree and repeatability. Chromatograms taken from naturally contaminated wheat samples are shown in Figure 2, respectively, on top of chromatograms of mixtures of NIV, DON and DON-3G standards.
Recovery (R) and repeatability (precision) (relative standard deviation, RSD) of the method was checked for three fortification levels: 225/450/900 µg/kg. Ground blank wheat grain was used in the validation experiment. Depending on the level, the analyte, the obtained R values ranged from 76.9 to 109.4% (see Table 1), while the RSD values ranged from 2.2 to 7.9%. The ranges are quite satisfactory.

Occurrence of Mycotoxins in Wheat Grain
The presence of NIV/DON in wheat grain proves the plants must have been infected with the Fusarium fungi. DON-3G is a product of plant metabolism produced from DON by plant enzymes. NIV (above the method LOD threshold) was found in 70% of the studied wheat samples, DON-in 83% of the samples ( Table 2)

Discussion
Capability to improve method selectivity by removing interfering compounds is one of big advantages of immunoassay columns (IAC). Since samples get concentrated, LOD/LOQ thresholds for analytes may be improved. Highly purified samples produced by IAC columns used to isolate the to-be-determined analytes from extracts may be readily analysed with the help of the liquid chromatography (LC) technique using various detectors. In our analyses it was an Ultraviolet-Visible (UV-VIS) detector (Knauer, Wissenschaftliche Geräte GmbH, Berlin, Germany) operated at 218 nm wavelength. The DON-NIV WB columns (Vicam, Watertown, MA, USA) applied by us contain some antibodies allowing them to feature a high cross-reactivity for modified mycotoxins. Therefore they enabled simultaneous analyses of NIV, DON, and DON-3G. Advantages of the columns have recently been successfully utilized also by other authors (Trombete et al. [25], Geng et al. [26], Yoshinari et al. [27]).
Mycotoxins produced by Fusarium fungi are commonly found in grain cultivated on every continent. A high fraction of our samples were NIV-and DON-positive (70% and 83% of the samples, respectively), while DON-3G was found in only 27% of the samples. The results obtained in this work are quite important for food safety, especially since literature data on the pollution of food with DON-3G are quite scarce. Although the frequency of occurrence of the discussed mycotoxins was high, the quality of the wheat was generally high. In must be emphasized that DON-3G can increase the risk to food safety, because, as was reported in the literature, it can be hydrolysed in the intestine by microflora.

Samples
The 92 samples of winter wheat were collected in 2016 from crops harvested in various regions of Poland. The samples were gathered at collection points of The National Center for Agricultural Support located in 15 voivodeships; the map of them is presented in Figure 5.

HPLC-UV
A Knauer K 1001 HPLC instrument (Knauer, Wissenschaftliche Geräte GmbH, Berlin, Germany) equipped with an autosampler, a thermostat, and a 45 µL injection loop was used. Analytes were separated using a Cosmosil 5C18-AR-II, 4.6 × 250 mm chromatographic column (NacalaiTesque, Kyoto, Japan) kept at 45 °C constant temperature. Separation was done in isocratic mode. The mobile phase was 90:10 v/v water: acetonitrile mixture flowing at 1 mL/min. An UV detector manufactured by Knauer was operated at 218 nm wavelength. Knauer Eurochrom HPLC Software ver. 1.65 was used to integrate chromatographic peaks.

HPLC-UV
A Knauer K 1001 HPLC instrument (Knauer, Wissenschaftliche Geräte GmbH, Berlin, Germany) equipped with an autosampler, a thermostat, and a 45 µL injection loop was used. Analytes were separated using a Cosmosil 5C18-AR-II, 4.6 × 250 mm chromatographic column (NacalaiTesque, Kyoto, Japan) kept at 45 • C constant temperature. Separation was done in isocratic mode. The mobile phase was 90:10 v/v water: acetonitrile mixture flowing at 1 mL/min. An UV detector manufactured by Knauer was operated at 218 nm wavelength. Knauer Eurochrom HPLC Software ver. 1.65 was used to integrate chromatographic peaks.