Determination of Multimycotoxin in Cereal-Based Products Sold in Open-Air Markets

In this study, a total of 140 cereal-based foods sold in temporary open-air markets were analyzed by LC-MS/MS for aflatoxin B1, B2, G1, G2, ochratoxin (OTA), zearalenone (ZEN), deoxynivalenol (DON), fumonisin B1, fumonisin B2, citrinin (CIT), HT-2, and T-2 toxins. Breakfast cereals (n:27), cornmeal (n:41), extruded maize (n:32), and oatmeal (n:40) purchased from these alternative shopping areas created to meet the food needs of low-income people in the suburbs formed the sample set of the study. These foods, which are sold in areas that are out of legal control and greatly affected by external environmental conditions, are more open to health risks. Mycotoxins, chemicals of a biological origin, are some of the most important of these risks. In terms of public health, it is important to investigate the presence of mycotoxins in foods, which can cause acute and chronic diseases such as immunosuppression, genotoxic, estrogenic, teratogenic effect, cancer, and liver and kidney dysfunctions. Grain-based foods are often contaminated with a large number of mycotoxins, but legal regulations have not been prepared that consider the health risks associated with the co-existence of mycotoxins. Many of the studies have focused on the presence of a single mycotoxin and the risks it poses. As a result, aflatoxin B1 levels in 28.57% of the samples and total aflatoxin (B1 + B2 + G1 + G2) levels in 26.43% of the samples were determined to exceed the limits defined in the “Turkish Food Codex Contaminants Regulation”. Citrinin could not be detected in any of the samples. The rate of mycotoxin occurrences above the limit of detection (LOD) in grain-based food samples ranged from 22.86% to 99.29%. Total aflatoxin (TAF) + Total Fumonisin (FUM) were found in 83.57% of the samples; TAF + FUM + OTA in 82.14%; TAF + FUM + OTA + T-2 in 44.29%; TAF + FUM + OTA + DON + HT-2, TAF + FUM + OTA + DON + T-2, and TAF + FUM + OTA + DON + ZEN in 22.86% of the samples.


Introduction
Today, the demographic structure of cities is changing with intense internal and external migration. Demands vary depending on culture, habits, and income level of people in cities with increasing populations. Alternative shopping areas are created in cities, especially in suburbs, to meet the food needs of people of different socio-economic levels [1]. Open-air markets are one of these alternatives. These are temporary outdoor marketplaces on private property where individual vendors offer produce, food, or other products for sale, directly to the consumer. In these areas, which are out of legal control and greatly affected by external environmental conditions, the possible public health risks of the foods offered for sale are also increasing [2]. Cereals and cereal-based products such as cornmeal, extruded maize snacks, maize-based breakfast cereals, and oatmeal are among the foods sold in such places and mostly preferred by low-income consumers. Due to poorly managed production processes, poor harvesting conditions, long drying times in the open air, and poor storage practices, grains are subject to a wide variety of fungal contamination from farm to fork [3].
Production of mycotoxins in cereals is largely dependent on factors such as preand/or post-harvest temperature, moisture content, and insect damage. Climate is the main determinant of fungal formation and mycotoxin production [4][5][6]. Some of the mycotoxins are subdivided into field mycotoxins, which are mainly produced by Fusarium before or just after harvest in the crop, and the other part as storage mycotoxins secreted by Aspergillus and Penicillium species during the drying of grains, conversion into commercial products, storage, and sale [4,6]. Mycotoxins are heat-resistant natural toxic compounds with a high bioaccumulation ability [7]. The fact that mycotoxins are resistant to cooking methods paves the way for serious health problems [8]. For this reason, because they may contain mycotoxins, controlling these foods is of great importance in terms of food safety and protection of consumer health.
Many countries have made important regulations on "acceptable health risks" to control mycotoxin contamination in foods and prohibit the trade of contaminated products. These regulations generally depend on a country's level of economic development, the rate of consumption of high-risk products, and the susceptibility of crops to contamination [9,10]. There is a Turkish Food Codex Contaminants Regulation in force for these products in Türkiye (Table 1). Cereal-based foods are often contaminated with a large number of mycotoxins, but legal regulations have not been prepared considering the health risks associated with the co-existence of mycotoxins. The interactions of mycotoxins in foods can be classified into three main categories: antagonistic, additive, and synergistic. In particular, the additive and synergistic effect may cause the picture to become more intractable in terms of public health [4].

Sample Collection
In this study, 140 cereal-based (cornmeal (n:41), extruded maize snacks (n:32), maizebased breakfast cereals (n:27), and oatmeal (n:40)) samples were purchased from temporary open-air markets in suburbs ofİstanbul. Samples were collected in nylon bags (100 g) from May 2022 to August 2022. The samples were delivered to the laboratory in their original packages and were stored at 7 • C until analysis.
As part of our research, we took a closer look at the techniques utilized by Kim et al. [17], De Santis et al. [18], and Solfrizzo et al. [19], and made some modifications to them for further examination.

Preparation of Mobile Phases
The extraction solution was prepared by mixing 800 mL of methanol (Merck, Germany) with 200 mL of LC-MS-grade water (Merck, Darmstadt, Germany). The mobile phases were constituted of 0.252 g of ammonium formate (Merck, Darmstadt, Germany) in 1 L of LC-MS-grade water containing 1 mL formic acid (mobile phase A) and 0.252 g of ammonium formate in 1 L of LC-MS-grade methanol containing 1 mL of formic acid (Merck, Darmstadt, Germany) (mobile phase B).

Sample Preparation and Extraction
The samples were homogenised in the grinder (Retsch GM 200, Haan, Germany). Five grams of the samples were weighed. The weighed samples were mixed with 12 mL of the extraction solution for two hours while shaken. At the end of the period, the samples were centrifuged at 3000× g for 5 min, passed through a 0.45 µm filter (Millipore, Merck KgaA, Darmstadt, Germany) and collected in a 15 mL centrifuge tube. The filtrate was transferred into a 1.5 mL vial, with a volume of 800 µL. After adding 200 µL of mobile phase A, the solution was analysed. Two replicates were performed in one day.
where : LOQ = SD × 10 (3) LOD: limit of detection; LOQ: limit of quantification; SD: standard deviation; xmean: the arithmetic mean of the data; xi: each value of the dataset; n: the total number of data.
For the determination of measurement uncertainties, ISO [21] and Ellison and Williams [22] references were used. Repeatability, reproducibility, and uncertainty budgets from recovery were combined as indicated in the references. Uncertainty budgets for each active ingredient were calculated separately. Uncertainties about the equipment used were included in the repeatability, reproducibility, and recovery uncertainty budgets. LOD, LOQ, recovery, and standard uncertainty ratios are given in Table S2, R2 and regression equations are provided in Table S3.

Results
The results for our cereal-based products are presented in Table 2. We conducted 12 mycotoxin analyses on the collected samples using the LC-MS/MS technique. The findings revealed that 28.57% of the samples exceeded the maximum limits specified in the Turkish Food Codex Contaminants Regulation in terms of AFB 1 and 26.43% in terms of total aflatoxin. It was determined that 74.29% of the samples were contaminated with AFB 1 , 45% with AFB 2 , 36.43% with AFG 1 , and 40.71% with AFG 2 at levels above the detectable limits. While this rate was 22.86% for FB 1 , contamination with FB 2 was detected in 90% of samples. On the other hand, 87.86% of the samples were contaminated with OTA, 44.29% with T-2, 40.71% with HT-2, and 23.57% with ZEN and DON. CIT was not detected in any of the samples.

Discussion
In our study, multiple mycotoxin analyses in cereal-based foods (n:140) were performed using the liquid chromatography-tandem mass chromatography method, which provides a fast, reliable, and economic analysis (n:140). LC-MS/MS is mostly preferred for mycotoxin analysis due to its high selectivity and sensitivity. According to a report by Amirahmadi et al. [23], the matrix effect could potentially have a detrimental effect on quantitative performance, particularly as the electrospray ionization (ESI) source is approached. The methods used in our study meet the requirements of EC No: 401/2006 [24] and EC No: 1881/2006 [25] commission regulations. The data we reported on mycotoxin contamination in cereal-based products in Türkiye provide crucial data in terms of the diversity of analytes investigated, including some less frequently investigated secondary fungal metabolites, which were first reported in products offered for sale in open-market places.
The EU stated that no direct human consumption product should be present at levels higher than 2 µg·kg −1 for AFB 1 and more than 4 µg·kg −1 for total aflatoxin [25,26]. The regulation in TFC is also in this direction for grain-based products that make up our sample set [16].
In our study, it was observed that none of the cornmeal samples exceeded the limit value reported in TFC for AFB 1 and TAF. On the other hand, 95% and 90% of oatmeal samples were found to be above the limit values for AFB 1 and TAF, respectively. The sample rate exceeding the AFB 1 and TAF limit values in extruded maize snack products was determined as 3.13%. However, 3.7% of maize-based breakfast cereal products exceeded the AFB1 limit while none of the samples exceeded the TAF limit value.
There are many studies in the literature about the presence of aflatoxins in corn and oat-based samples. Alborch et al. [27] found that the samples were contaminated with 0.19-3.1 µg·kg −1 of aflatoxin in their study on 60 cornmeal and corn grain samples. Similarly, Zhou et al. [28] also detected AFB 2 in the range of 0.67-1.3 µg·kg −1 in 16 corn samples they examined. In another study, the mean AFB 2 level was measured in 79 different types of corn samples, with very little difference, but still close (1.5 µg·kg −1 ) [29]. When the data in the mentioned studies were compared with those in our study, it was noticed that they were similar. However, in some studies, it was found that the average aflatoxin values in the samples examined were higher than our data [17,[30][31][32], while in some others, they were below the detection limits, in other words, at very low levels contrary to ours [33][34][35].
In terms of oat-based products, there is a slight difference in the presence of aflatoxin compared to corn-based products. For example, in some studies, it is reported that the types of aflatoxins detected in oat-based products are more limited, but also their levels are lower. According to Ortiz et al. [36]'s study, only 2% of the 42 oat flakes samples they examined were contaminated with AFB 1 , while other aflatoxin types could not be detected. Juan et al. [37] also stated that AFG 1 was found in only 0.1% of the samples they analysed, while other types were not present. This has been associated with the correct functioning of the food safety system in the processing of products. In addition, in some studies, it has been reported that the total amount of aflatoxins in the oat-based samples examined was below the detection limits, in other words, no type of aflatoxins was found in these products [38][39][40][41]. It is predicted by the researchers that this situation may be caused by various factors such as product type differences, processing techniques, and temperature changes encountered in the process. However, there are also studies reporting a higher rate of aflatoxin content in oat products. For example, Al-Taher et al. [42] reported that they found AFB 2 at the level of 1.1-2.6 µg·kg −1 and AFG 2 at the level of 0.4-1.7 µg·kg −1 in 40% of the 20 breakfast samples they examined. Additionally, Kuzdralinski et al. [43] stated that 41% of 58 breakfast oat samples they analysed were contaminated with aflatoxin at an average level of 1.4 µg·kg −1 .
Fumonisin, one of the most toxic types of mycotoxins known, unlike aflatoxins, multiplies by contaminating the developing crop prior to harvest, so that, generally, it is difficult to control or prevent its development. There are three types of fumonisin, FB 1 , FB 2 , and FB 3 , and FB 1 is the most common type in foods, especially in corn and corn-based products. The fact that fumonisin B 1 (average 146.17 µg·kg −1 ) was detected only in corn-based samples among the samples examined within the framework of our study is a finding that confirms this information. In the literature review, it is seen that the studies mostly focus on FB 1 . However, although it is not as strong as FB 1 regarding toxicity potential, it is also known that if FB 2 is present in the flora, it increases toxicity by showing a synergistic effect with FB 1 . Among the samples in this study, an average of 58-60 µg·kg −1 was detected in corn-based samples and an average of 166.55 µg·kg −1 in oat-based samples. Many studies in the literature also support our findings from corn products [17,33,44]. On the other hand, there are also studies reporting that fumonisin species are less common in oat-based products compared to corn-based products [29,42,45].
In addition to the toxins we mentioned in corn and oat-based agricultural products, other toxins can be found. For example, heterogeneous trichothecenes and mycooestrogens (type A: T-2 and HT-2; type B: DON, ZEN, and CIT) produced by some Fusarium species fungi and different Aspergillus species (A. ochraceus, A. steyni, A. westerdijkiae) and OTA, classified in Group 2, are the most known toxins for humans in this regard. The presence of these toxins, as with the toxins we mentioned before, is mostly affected by errors in agricultural practices, storage defects, insect attacks, genetic factors, and negative climatic conditions.
According to the data of our study, it was observed that those toxins instead of OTA were found at higher rates and values in oat-based products compared to corn-based ones, while OTA was found to be higher only in corn-based products (Table 2). Moreover, in the results section we mentioned, T-2 remained below the detection limits in the products we examined, while CIT could not be detected in any of the samples. In this context, it is known that there are studies on the presence of multi-mycotoxin equivalent to our study. Alborch et al. [27] detected OTA at a level similar to our study (0.79-1.71 µg·kg −1 ) in 33.3% of 60 corn-based products they collected from open markets, while Kim et al. [17] reported that they detected 180.4 µg·kg −1 DON in 13% of 507 corn samples in their study. In another study, researchers reported that corn samples (n:37) were contaminated with DON (100%; 171 µg·kg −1 ) and OTA (19%; 1 µg·kg −1 ) equivalent to our data [46].
Another important issue that needs to be emphasized is that these toxins can be found in different and multiple combinations in foods (natural co-occurrence). This can cause toxins to interact with each other in undesirable ways. TAF and FUM, DON and ZEN, TAF and OTA, and FUM and ZEN are mostly observed in the literature in cereals and grain product samples. However, due to the synergistic negative effects of co-occurring mycotoxins on human and animal health, it is important to evaluate their co-occurrence [50]. Unfortunately, it seems that most studies have focused on the formation of a single mycotoxin. Only a few studies reported the number of co-occurring mycotoxins and the proportion of co-contaminated samples together, as well as the main combinations found [4]. Moreover, legal regulations do not consider the combined effects of mycotoxins. In this context, we evaluated the natural co-occurrence of these toxins in our samples. In our study, the co-occurrence rates of any two mycotoxins in the samples are given in Figure 1. The coexistence of any three mycotoxins in our samples is indicated in Figure 2. ZEN + DON + OTA was found in 22.14% of the samples, DON + TAF + FUM in 23.57%, and TAF + ZEN + FUM in 23.57%. In addition, in Figures 3 and 4, the co-existence rates of four or more mycotoxins in our samples are shown.  [46]. In a study on oat-based products, it was reported that approximately 50% of the 20 oat breakfast samples were contaminated with DON, OTA, and ZEN at levels of 2.5-146.5 μg•kg −1 , 0.6-11.8 μg•kg −1 , and 1.2-13.6 μg•kg −1 , respectively [42]. There are also studies in which the mycotoxin levels detected are much lower [34,37,39,44] and much higher [40,43,47,48,49] when compared to the data of our study.
Another important issue that needs to be emphasized is that these toxins can be found in different and multiple combinations in foods (natural co-occurrence). This can cause toxins to interact with each other in undesirable ways. TAF and FUM, DON and ZEN, TAF and OTA, and FUM and ZEN are mostly observed in the literature in cereals and grain product samples. However, due to the synergistic negative effects of cooccurring mycotoxins on human and animal health, it is important to evaluate their cooccurrence [50]. Unfortunately, it seems that most studies have focused on the formation of a single mycotoxin. Only a few studies reported the number of co-occurring mycotoxins and the proportion of co-contaminated samples together, as well as the main combinations found [4]. Moreover, legal regulations do not consider the combined effects of mycotoxins. In this context, we evaluated the natural co-occurrence of these toxins in our samples. In our study, the co-occurrence rates of any two mycotoxins in the samples are given in Figure 1. The co-existence of any three mycotoxins in our samples is indicated in Figure 2. ZEN + DON + OTA was found in 22.14% of the samples, DON + TAF + FUM in 23.57%, and TAF + ZEN + FUM in 23.57%. In addition, in Figure 3 and Figure 4, the coexistence rates of four or more mycotoxins in our samples are shown.

Conclusions
This research contributes to eliminating the deficiency observed in the literature regarding the presence and concentration of mycotoxins in cereal-based products marketed in open-air markets in Türkiye. Mycotoxin contaminations, mycotoxin

Conclusions
This research contributes to eliminating the deficiency observed in the literature regarding the presence and concentration of mycotoxins in cereal-based products marketed in open-air markets in Türkiye. Mycotoxin contaminations, mycotoxin combinations, contamination status, and frequency of occurrence in these foodstuffs were determined. It has been revealed that effective control of mycotoxins, which may pose a risk to human health even at low doses, should be carried out with monitoring systems in this and similar types of foodstuffs. On the other hand, considering the additive and synergistic interactions of different mycotoxins, new legal regulations will be needed regarding the presence and limit values of mycotoxin combinations in foods. It is highly recommended to continuously monitor the mycotoxins that were studied in vulnerable food items and to conduct research aimed at identifying masked mycotoxins and other fungal metabolites. Additionally, efforts should be focused on reducing and managing mycotoxins within the food supply chain.

Data Availability Statement:
The data used to support the findings of this study are available by the corresponding author upon request.

Conflicts of Interest:
The authors declare no conflict of interest.