Simultaneous Occurrence of Aflatoxins and Fumonisins in Corn Intended for the Pet Feed Industry and for Human Consumption

In this study, 24 samples of corn intended for the pet feed industry (PFI) and 24 samples of corn for human consumption (HC) commercialized in São Paulo, Brazil were analyzed for aflatoxins B1, B2, G1 and G2, and fumonisins B1 and B2 by high performance liquid chromatography. The concentrations of total aflatoxins in positive samples of PFI (33.3%) and HC (70.8%) corn samples ranged from 0.5 to 3.9 μg/kg and 0.5 to 41.8 μg/kg, respectively. For fumonisins, the concentrations in positive samples of PFI (100%) and HC (83.3%) corn samples ranged from 685 to 11,379 μg/kg and 157 to 6,495 μg/kg, respectively. Overall the aflatoxin levels complied with regulations for food products in Brazil (20 μg/kg), except for one sample of HC corn. However, the high incidence of fumonisins indicates the need for better agricultural practices to avoid contamination of corn, especially in the PFI.


Introduction
Mycotoxins are secondary metabolites produced by fungi that grow in food products worldwide, leading to a great variety of toxic effects in vertebrates, including humans [1].Toxigenic fungi may contaminate food products at different phases of production and processing, mainly under favourable humidity and temperature conditions [2].In Brazil, the most common toxigenic fungi found in cereals include species from the genera Aspergillus and Fusarium.Aspergillus species, mainly A. flavus, A. parasiticus and A. nomius, produce aflatoxins, which have high toxicity, teratogenicity, mutagenicity and carcinogenicity [3].
There are four major aflatoxins, namely B1, B2, G1 and G2, which can contaminate feed and foodstuffs.However, aflatoxin B1 (AFB1) is the most commonly occurring and has the higher toxigenic potential, with the liver as the main affected organ [2].The occurrence of AFB1 in food products is a public health concern because the International Agency for Research on Cancer (2015) [4] classified AFB1 in Group 1 -human carcinogen.AFB1 can produce liver tumors in Fisher rats at levels as low as 50 µg/kg [5].Fumonisins are a group of structurally related mycotoxins produced mainly by Fusarium verticillioides.Although 28 fumonisin analogs have been identified since 1988, fumonisin B1 (FB1) is the most prevalent and also the most toxic, FB1 is considered a possible carcinogen in humans (group 2B) by the International Agency for Research on Cancer (2015) [4].Since their identification, fumonisins have been associated with animal diseases such as equine leukoencephalomalacia and porcine pulmonary edema, although at concentrations nearly 1,000 higher than the toxic levels of AFB1 [3].
Outbreaks related to pet food contaminated with aflatoxins have been described in several countries, including USA [6] and Israel [7], with corn as the usual source of aflatoxins in the feed.Although outbreaks caused by mycotoxins in pet animals have not been described in Brazil, a few cases of natural aflatoxicosis in dogs were reported in Rio Grande do Sul [8].Moreover, previous studies reported aflatoxin contamination in the 12% of 4 of samples collected in Santa Catarina State presenting at least one type of mycotoxin [9].Fumonisinrelated outbreaks in pet animals have not been described in the literature.
Brazilian corn production is around 75 million tons per year, of which approximately 65% is destined for the animal feeding industry.Previous reports have indicated that the incidence of aflatoxins in Brazilian corn is highly variable, with mean levels of up to 460 µg/kg [10,11].For fumonisins, higher incidences (up to 100%) were reported in corn, with mean levels ranging from 20-22,600 µg/kg [10][11][12].However, there is very little information on the simultaneous occurrence of aflatoxins and fumonisins in Brazilian corn kernels.The objective of the present trial was to evaluate the simultaneous occurrence of fumonisins and aflatoxins in corn kernels available for the pet feed industry (PFI) and for human consumption (HC) in São Paulo, Brazil.

Sampling procedures
Corn intended for the PFI was collected during the same 6-month period, from a large scale supplier of corn located in Porto Ferreira city, state of São Paulo (4 samples from different batches per month), totaling 24 samples.The supplier received corn from several producers in the state of São Paulo by the time of sampling, stored the product for a period of time and distributed it for at least 8 pet feed factories in the state.Sampling procedures followed the recommendations of Food and Agriculture Organization [13].Incremental portions of corn were collected at different intervals during the transference of kernels from trucks to the storage silos, until reaching 16 kg of corn for each sample.The kernels were collected in sterile polypropylene bags, identified and kept at room temperature until analysis.
Samples of packaged corn kernels for HC were collected in the cities of Araras, Leme, Pirassununga and Porto Ferreira, all located in the Northeast of the state of São Paulo.In each city, samples from the brands that showed the greatest trade volume in supermarkets were collected from January to June 2008 (4 samples per month, one per city), totalling 24 samples of corn.The sampling unit was made up of original closed packages of at least 500 g of kernels, collected from different batches as indicated in the label, avoiding batch repetition.Samples were identified, including data on the manufacturer, batch and/or manufacturing date and expiring date.Products were stored in their respective packages and kept at room temperature (same supermarket conditions) until the moment of analysis.

Extraction of aflatoxins and fumonisins
The extraction and purification of aflatoxins (B 1 , B 2 , G 1 and G 2 ) and fumonisins (B 1 and B 2 ) in PFI and HC samples were performed using immunoaffinity columns, following manufacturer recommendations (Aflatest® or Fumonitest®, Vicam, Watertown, MA, USA).Previously, the total amount of corn samples collected in supermarkets and a 2.5 kg subsample of corn collected in the supplier were grinded in a hammer mill (Marconi, Piracicaba, Brazil) until a particle size of nearly 14 mesh, and mixed thoroughly.An aliquot of each sample (50 g) was weighted in an Erlenmeyer flask containing 5 g of sodium chloride and 100 ml of methanol/water (80:20, v/v) were added.The flask was placed in an orbital shaker (Tecnal, Piracicaba, Brazil) for 30 minutes, the flask content was filtered and 10 ml of the filtrate were placed in a Becker.40 ml of ultra-pure water (Milli Q, Millipore, Bedford, MA, USA) were added and the mixture was filtered in a 1.5 µm microfiber filter.Two 10 ml aliquots were then passed through the immunoaffinity columns (Aflatest® or Fumonitest®) at flow rate of 1-2 drops/sec.After washing with 10 ml of ultrapure water, aflatoxins or fumonisins were eluted from their respective columns with 1 ml of methanol, being each eluate collected in an amber vial.The aflatoxins and fumonisins eluates were evaporated to dryness under nitrogen flow.

Determination of aflatoxins
AFB 1 and AFG 1 in the final extracts were derivatized by adding 200 µL of n-hexane and 200 µL of trifluoroacetic acid (TFA) to the aflatoxin dried extract [14].The mixture was kept at 40 °C for 10 min, evaporated to near-dryness and diluted in 1 ml of methanol: water (50:50, v/v).Final extracts were filtered through a 0.45 µm PTFE membrane and 20 µL were injected into a Shimadzu (Kyoto, Japan) 10VP high performance liquid chromatograph (HPLC) with a 10 AXL fluorescence detector (excitation at 360 nm and emission above 440 nm).A Shim-Pack CLC-ODS Sil column (4.6 X 250 mm, 5 µm) and a Shim-Pack pre-column (4 X 10 mm, 5 µm CLC G-ODS) were used.The isocratic mobile phase consisted of methanol-water (45:55, v/v) with a flow rate of 1.0 ml/min.Calibration curves were prepared using standard solutions of aflatoxins B 1 , B 2 , G 1 and G 2 (Sigma, St Louis, MO, USA) previously evaluated according to Scott (1990) [14].The individual aflatoxins solutions were mixed in convenient volumes to reach working solutions at the concentrations of 2.5, 5.0, 10.0 and 20.0 ng/ml of each aflatoxin.Aflatoxins working solutions were prepared with TFA similarly to the sample extracts.The retention times were approximately 4.2 min for AFG 1 (converted to AFG 2a ), 5.0 min for AFB 1 (converted to AFB 2a ), 7.1 min for AFG 2 and 9.7 min for AFB 2 (Figure 1).

Determination of fumonisins
Fumonisin dried extracts were re-diluted with 200 µL of acetonitrile-water (50:50, v/v) and filtered using a 0.45 µm PTFE membrane.A 100-µL aliquot was placed in a test tube.After that, 200 µL of o-phthaldialdehyde (OPA) reagent (prepared using 40 µg of OPA diluted in 5 mL of sodium tetraborate solution 0.1 M and 50 µL of 2-mercaptoethanol) were added.After 2 minutes, 20 µL were used for quantification of the toxins in the same HPLC system described before, with a

Simultaneous Occurrence of Aflatoxins and Fumonisins in Corn Intended for the Pet Feed Industry and for Human Consumption
Oliveira et al.
reverse phase C 18 column (150 x 4.6 mm, particle size 5 µm -Phenomenex, Torrance, USA) kept at constant temperature of 30 o C. Mobile phase was made up of acetonitrile-wateracetic acid (50:50:1, v/v), and constant flow of 1.0 ml/min was used.Detection of derivatized fumonisins was performed using the same HPLC system as described for aflatoxins, under fluorescence from 335 to 440 nm for excitation and emission, respectively.The calibration curves were performed from successive chromatographic measurements of FB 1 and FB 2 standards in triplicate at concentrations of 0.313, 0.625, 1.25, 2.5 and 5.0 µg/ml of each fumonisin.Retention time was approximately 9.5 minutes for FB 1 and 27.4 minutes for FB 2 (Figure 2).

Validation procedures
The limits of detection (LOD) and quantification (LOQ) were calculated for each method of analysis based on signal:noise ratio of 3:1 and 10:1, respectively.Linearity was evaluated by verifying the coefficient of determination (r 2 ) and visual inspection of residual plots of analytical curves built separately for each mycotoxin.The analytical methods used in the experiment were evaluated using commercially available corn samples previously analyzed for endogenous AFB 1 and FB 1 .Triplicate samples were spiked with AFB 1 and FB 1 at levels of 2 and 20 µg/kg and 50 and 500 µg/kg, respectively.All these fortified samples were analyzed as described for samples collected in supermarkets and in the corn supplier.

Despite the limitations in the sample collection, results
for aflatoxins alone in HC corn in this trial were higher   than previous data reported in Brazil.Bento et al., [18] found aflatoxins in 19% and 23.8% of corn samples from Mato Grosso state harvested in 2009 and 2010 respectively, although at higher levels (1.0 to 108.7 µg/kg).8% of 300 samples of freshly harvested corn collected in the reception and pre-drying steps in Paraná State had aflatoxins at mean concentration of 30.6 µg/kg [11].However, the authors found fumonisins in 100% of the analyzed samples, with similar mean levels (2,082 µg/kg) as obtained in the present study.Moreover, the mean fumonisin level in corn intended for HC as presented in our work were in agreement with those found by Westhuizen et al. [12], in Santa Catarina state (mean: 2,870 µg/kg).In summary, the low levels of aflatoxin and high levels and incidence of fumonisin in corn reported in the present study is in accordance with previous studies indicating that producers of corn for HC have improved control practices to avoid contamination of corn grain with aflatoxin, but not fumonisin.
Samples of corn intended for PFI also presented low levels of aflatoxins, ranging from 0.5 to 3.9 µg/kg with a mean of total aflatoxins of 1.5 µg/kg [Table 2].No sample had levels higher than the recommended level of aflatoxins for feed ingredients marketed in Brazil (50 µg/kg) [19].However, fumonisins were detected in all samples analyzed, at levels ranging from 685 to 11,379 µg/kg (mean: 3,845 + 2,570 µg/kg), which is higher than fumonisin levels found in samples of corn intended for HC.Importantly, there is no regulation for fumonisins in feedstuffs in Brazil, and the recommended level for aflatoxins (50 µg/kg) is not an action level.
The comparison with our results with previous data is difficult because there is no report on the occurrence of aflatoxins or fumonisins in corn intended for the pet industry in Brazil.However, aflatoxin-contaminated pet feeds have been found in surveys conducted in the states of Minas Gerais [20] and Santa Catarina [9].Pleadin et al. [21] confirmed the corn as the main source of aflatoxin contamination in feeds in Croatia, where high incidences of aflatoxins at levels ranging from 1.1 to 2,072 µg/kg were found in corn samples from different farms and feed factories.FB 1 levels in PFI or HC samples obtained in the presented study were higher when compared with those reported by Mngadi et al. [22], who 2 Percentage of positive samples of the total analysed samples of corn for pet feed industry (n = 24) and for human consumption (n = 24)". 3Minimum and maximum levels quantified in analysed samples in duplicate. 4Values are reported as means ± SD, for samples containing quantifiable levels of each mycotoxin.Despite the high incidence and levels of fumonisins in Brazil, there is no legally maximum limit for these mycotoxins in animal feed.The tolerance limit for FBs adopted by the European Union is 60,000 µg/kg for corn and corn products and 5,000 µg/kg for complementary and complete feedingstuff for pet animals [23].In our study, no corn sample had fumonisins levels above this limit.However, one sample had 11,379 µg/kg, which is above the U.S. Food and Drug Administration recommendation for total fumonisins (10,000 µg/kg) in corn [24].
The simultaneous occurrence of aflatoxins and fumonisins was found in 54.2% samples of corn intended for HC, and in 33.3% samples of corn for the PFI, as presented in [Table 3].The levels of fumonisins were higher in corn destined for feed pet industry, hence indicating that better control practices are applied to corn kernels destined for HC, reflecting specially with agronomic practices and climatic conditions.The higher fumonisin contamination in corn for animal consumption can also be due to a greater period that corn is stored before being used in industry, in addition to worst conditions of storage.

Conclusion
The simultaneous occurrence of aflatoxins and fumonisins in corn reinforces the need for regulations for mycotoxins in feedstuffs in Brazil, to avoid potential health risks in feed supply industry.
Samples of corn intended for the PFI, and corn kernels commercially available for HC in the Northeast region of São Paulo State presented high incidences of aflatoxins and fumonisins.Although the mean concentrations of aflatoxins in corn samples were below the maximum limit established adopted in Brazil, high fumonisin levels were found in all samples, especially in corn intended for the PFI.There is a need for regulations for mycotoxins in feedstuffs in Brazil.

Table 1 :
Performance of the analytical methods for aflatoxins and fumonisins in corn intended for human consumption and for the pet feed industry in São Paulo,Brazil.
1Values expressed as mean ± standard deviation of samples analyzed in triplicate.

Table 2 :
Aflatoxins and fumonisins in corn intended for human consumption and for the pet feed industry in São Paulo, Brazil.

Table 3 :
Simultaneous occurrence of aflatoxins and fumonisins in cornintended for human consumption and for the pet feed industry in São Paulo,Brazil.

Simultaneous Occurrence of Aflatoxins and Fumonisins in Corn Intended for the Pet Feed Industry and for Human Consumption Oliveira et al. found
[9]centrations of 15 to 5,900 µg/kg in 23 animal feed from South Africa.The results also showed a contamination by fumonisins in 100% of corn samples evaluated.On the other hand, in Santa Catarina State, Scussel et al.[9], detected fumonisin in 11% of the 123 pet feed samples.