Deoxynivalenol concentrations in feed ingredients and swine diets measured by enzyme-linked immunosorbent assay and high-performance liquid chromatography

ABSTRACT The objective was to compare deoxynivalenol (DON) concentrations in feed ingredients and commercial swine diets measured by enzyme-linked immunosorbent assay (ELISA) and high-performance liquid chromatography (HPLC). Seventy feed ingredient samples consisted of corn, corn dried distillers grains with solubles, corn gluten feed, corn gluten meal, palm kernel expellers, rice bran, soy hulls, soybean meal, and wheat. Commercial swine diet samples (n = 92) were collected from 23 swine farms of varying regions in Korea and different growth stages of pigs. The DON concentration of all samples was determined in duplicate. Statistical comparisons were performed to compare the analytical methods (ELISA vs. HPLC), diet phases, and regions. The DON concentrations in most ingredients and all diets determined by ELISA method were greater than those determined by HPLC. The DON concentrations determined by the ELISA method were less than 1 mg/kg in all ingredients except corn dried distillers grains with solubles and corn gluten feed, and those determined by the HPLC were less than 0.5 mg/kg in all ingredients. The DON concentrations in complete diets did not vary by region or growth stages of pigs. The DON concentrations in most feed ingredients and commercial swine diets determined by ELISA method are greater than those determined by HPLC, but does not vary by regions in Republic of Korea or pig growth stages.


Introduction
Deoxynivalenol (DON), naturally produced by the Fusarium species, is one of the mycotoxins commonly found in feed ingredients including cereal grains and byproducts (Tiemann et al., 2006;Rasmussen et al., 2012).Mycotoxins can be produced during the pre-harvesting period, as well as in the postharvesting period.The environmental conditions of storage are one of the major factors contributing to the occurrence of mycotoxins (Homdork et al., 2000;Mannaa and Kim, 2017).A hot and humid environment, for example, can cause mycotoxin contamination during storage of feed ingredients and animal diets (Agriopoulou et al., 2020).
Dietary DON causes feed intake reduction, growth retardation, nutrient digestibility reduction, and even vomiting (Diekman and Green, 1992;Jo et al., 2016).In Korea, most feed ingredients used to produce commercial swine diets are imported from other countries.Feed ingredients are shipped for long distances, which can be a reason for the productions of molds and mycotoxins (Tang et al., 2019).In addition, the summer season in Korea is hot and humid, resulting in reduced feed intake for pigs, so the detrimental effects of dietary DON including feed intake reduction and growth retardation can be more critical during that period.Moreover, the recent climate change can increase occurrence risk of DON in the feed ingredients and diets.
Generally, DON concentrations in feed ingredients and diets are analyzed by using enzyme-linked immunosorbent assay (ELISA) and high-performance liquid chromatography (HPLC).However, large variations have been reported in values for DON concentrations among different analytical methods (Ghali et al., 2009;Ekwomadu et al., 2021).One of our hypotheses was that the analytical methods potentially affect DON concentrations in feed ingredients and diets.We also hypothesized that dietary DON concentrations may vary by region or growth stage of pigs.The objective of this study was to compare DON concentrations in feed ingredients and commercial swine diets by regions and growth phases measured by ELISA and HPLC.
All experiments and analyses were conducted in a biosafety level 1 containment facility in Seoul, Republic of Korea (37°32'24.8"N, 127°04'27.5"E).The ELISA test kit (Romer Labs Inc., Singapore, Republic of Singapore) for DON with a quantification range of 250 to 5,000 ng/mL (AgraQuant ® Deoxynivalenol; COKAQ4000) was used to determine DON concentrations in ingredient and diet samples.The analysis was performed according to the manufacturer's instructions.Briefly, 20 g of finely ground (< 0.1 mm) sample was moved into Erlenmeyer flask and shaken for 3 min with 100 mL of distilled water.The extract was filtered with Whatman #1 filter, and 10 mL of extract was diluted with 30 mL of distilled water.The DON concentrations in the prepared samples were measured by using the ELISA test kit and microplate reader at 450 nm.All samples and standards were analyzed in duplicate.
For the HPLC procedure, 25 g of ground sample and 5 g of sodium chloride were transferred into a 500 mL Erlenmeyer flask.Then, 200 mL of distilled water was added and homogenized for 2 min with a shaking incubator.After this, the 40 mL of supernatant extract was transferred into a 50 mL conical tube and centrifuged at 2,469 g for 10 min (Hettich Rotofix 32A, Andreas Hettich GmbH & Co. KG, Tuttlingen, Germany), and adjusted to approximately pH 7.4 using 2 M sodium hydroxide.The whole supernatant extract was filtered through a microfiber filter paper.Two milliliters of filtrate were passed through the immunoaffinity chromatography (IAC) column (ZORBAX Eclipse XDB-C18 column, 4.6 × 250 mm, 5 μM, Agilent, USA) at a flow rate of 2 mL per min.The columns were washed with 10 mL of distilled water at a flow rate of 5 mL per min.The mycotoxins were eluted by passing 1.5 mL of 100% methanol through the column at a flow rate of one drop per second.The extracts were collected in a glass tube and dried with nitrogen at 60 °C.One milliliter of mobile phase (0.01% of Deoxynivalenol concentrations in feed ingredients and swine diets measured by enzyme-linked... Ahn et al. 3 acetic acid in deionized water:methanol (85:15, v/v)) was added to the dried glass tube and vortexed for 1 min, and then 100 µL of solution was transferred to an HPLC vial for the deoxynivalenol analysis.Deoxynivalenol was determined using a mobile phase for 20 min at a flow rate of 0.8 mL/min under UV light at a wavelength of 220 nm at 45 °C.The DON peak has a retention time of 14.9 min.All samples and standards were analyzed in duplicate.
Data were analyzed by MIXED procedure of SAS (Statistical Analysis System, version 9.4) to compare the DON concentration by the analytical methods (ELISA vs. HPLC) or by the diet phases and regions.
The statistical model used for the comparison of the analytical methods was: in which Y ij is the response variable measured in the i-th analytical methods at the j-th replication block, μ is the overall mean, A i is the fixed effect of i-th analytical method, b j is the random effect of the j-th replication block, and ε ij is the random error term.
The statistical model used for testing the influence of region and growth stage of pigs was: in which Y ijk is the response variable measured in the i-th region and j-th growth stage at the k-th replication block, μ is the overall mean, R i is the fixed effect of i-th region, G j is the fixed effect of j-th growth stage, b k is the random effect of the k-th replication block, and ε ijk is the random error term.
The experimental unit was each sample.The statistical significance was declared at an alpha less than 0.05.

Results
The DON concentrations determined by the ELISA method in all ingredients, except corn DDGS (2.614 mg/kg) and corn gluten feed (2.344 mg/kg), were less than 1 mg/kg (Table 1).The DON concentrations determined by the HPLC method in all ingredients were less than 0.5 mg/kg.The DON concentrations in corn DDGS, corn gluten feed, rice bran, soy hulls, soybean meal, and wheat determined by the ELISA method were greater (P<0.05)than those determined by the HPLC method.
The DON concentrations in commercial feeds determined by the ELISA method were greater (P<0.001)than those determined by the HPLC method (Table 2).The DON concentrations did not vary by farm region or pig growth stage regardless of the analytical methods (ELISA or HPLC).

Discussion
Dietary DON leads to harmful effects on growth performance in pigs, derived from reduced feed intake and weight gain (Mok et al., 2013;Kim et al., 2021).Moreover, ingested DON via contaminated diets can be accumulated in various organs of pigs (Goyarts et al., 2007).Most countries regulate mycotoxin concentrations in feed ingredients and complete diets, and regulatory limits or guideline values are provided (van Egmond and Jonker, 2004).In the current study, DON concentrations in all feed ingredients and diets were less than the recommended concentrations in the Republic of Korea (ingredients = 10 mg/kg, swine diets = 0.9 mg/kg regardless of growth stages).
Several analytical methods are available for determining DON concentrations in animal feeds (Turner et al., 2015).The HPLC method has been considered as a reference method for mycotoxin determinations (Zheng et al., 2003;Binder, 2007).The advantages of the HPLC method include high sensitivity and repeatability, but the HPLC method demands time, costs, and practiced skills (Wolf and Schweigert, 2018).In contrast, the ELISA method is one of most rapid techniques for detection of mycotoxins.Due to several advantages including simplicity, less time-consuming task, and multiple analyses, the ELISA method has been widely employed for mycotoxin analysis (Wolf and Schweigert, 2018).However, there has been a concern on the accuracy of measured values resulting from matrix effects in ELISA methods, which can cause underestimation or overestimation compared with the reference values determined by HPLC or gas chromatography methods (Janik et al., 2021).
The greater DON concentrations in the ELISA method compared with the HPLC method observed in the present study are in agreement with a previous study (Matić et al., 2011), indicating that the ELISA method potentially overestimates DON concentrations in feed ingredients given that the HPLC method provides true values likely due to the matrix effect and quantification of multiple DON precursors as DON in the ELISA method.In the present work, the direct ELISA method employing the antigen-antibody reaction was used for the analysis of DON concentrations in feed ingredients (Xu et al., 1988) and the HPLC method using ultraviolet detection was used for quantifying DON concentrations (Visconti and Bottalico, 1983).In DON-contaminated feed ingredients, DON precursors including DON-3-glucoside, 3-acetyl-DON, and 15-acetyl-DON can co-occur with DON (Pestka, 2007), and these precursors are sometimes analyzed along with DON due to their similar molecular structures in the direct ELISA method but not in the HPLC method (Cavaliere et al., 2005).Although analyzed values from the ELISA method were greater than those from the HPLC method, the first is considered as a comparable method for screening the DON in feed ingredients due to a high correlation between analyzed values from these two methods (Zheng et al., 2003;Tangni et al., 2011).Continuous monitoring of DON values analyzed by the ELISA with the HPLC method, as the reference values, should be needed for accuracy and reliability from the ELISA method (Matić et al., 2011).
A previous study reported large variations in mycotoxin concentrations in different regions in China (Guan et al., 2011).In addition, it has been shown that differences in occurrence and concentration of mycotoxins in feed ingredients according to the countries and regions (Rodrigues and Naehrer, 2012; Schatzmayr and Streit, 2013) are likely due to different climates during harvesting crops.With the same token, the concentrations of DON in corn and corn DDGS in the US have been reported to have year-to-year and regional variations (Zhang and Caupert, 2012).However, the regional differences in the DON concentrations were not observed in this study.This discrepancy may be due to the fact that most feed ingredients used in Korea are imported from other countries.
Mycotoxin concentrations in diets may vary by pig growth stage likely due to the different DON concentrations in ingredients and the different inclusion rates of individual ingredients.A previous study reported variations in aflatoxin B1 and ochratoxin A concentrations in complete diets among pig growth stages in the Beijing region, China, but not in DON concentrations (Li et al., 2014).This indicates that DON concentrations may have not deviated much among feed ingredients.In the present work, similarly, DON concentrations were not affected by pig growth phase.However, if a specific ingredient such as corn gluten feed, corn DDGS, or barley is highly contaminated with DON (Kong et al., 2015;Leni et al., 2019), the DON concentration in complete diets can dramatically increase depending on the inclusion rate of high-DON ingredient.

Conclusions
Overall, deoxynivalenol concentrations in most ingredients and commercial pig diets determined by the enzyme-linked immunosorbent assay method are greater than those determined by high-performance liquid chromatography method, but do not vary by region in Republic of Korea or growth stage of pigs.The deoxynivalenol concentrations in feed ingredients and commercial pig diets are lower than regulatory levels in Republic of Korea.

Table 1 -
Deoxynivalenol concentrations in feed ingredients measured by enzyme-linked immunosorbent assay (ELISA) and high-performance liquid chromatography (HPLC) 1 DDGS -dried distillers grains with solubles; ND -not detected; SEM -standard error of the mean. 1 All samples were analyzed in duplicate.

Table 2 -
Deoxynivalenol concentrations in commercial swine diets measured by enzyme-linked immunosorbent assay (ELISA) and high-performance liquid chromatography (HPLC) 1 1All samples were analyzed in duplicate.