Co-occurrence of fumonisins and T-2 toxins in milling maize fractions under industrial conditions

Abstract

Latest studies indicate frequent co-contamination of food and feed supplies. The aim of this study was to investigate the natural co-occurrence of fumonisins (FB) and T-2 toxins in maize milling fraction obtained under industrial conditions. Both mycotoxins were identified in all tested samples and the FB level was higher than T-2 toxin. This fact was reflected in the content of mycotoxins in milling end products (extra maize meal, superior maize meal, and maize flour) and in hominy feed fractions. Results revealed that mycotoxins are concentrated in the hominy feed fraction, while the lowest level of mycotoxins was identified in the extra maize meal that contains coarse particle, have uniform fineness and low fat content.

Las investigaciones más recientes señalan que frecuentemente se produce una contaminación conjunta de alimentos y de piensos. El objetivo de este estudio se dirigió a investigar la ocurrencia conjunta y natural de dos micotoxinas, las fumonisinas y las toxinas T-2, en las fracciones de maíz molido obtenidas en condiciones industriales. Ambas micotoxinas fueron identificadas en todas las muestras probadas, constatándose que el nivel de fumonisinas fue más elevado que el de la toxina T-2. Este hecho se reflejó en el contenido de micotoxinas presente en los productos finales de la molienda (harina de maíz extra, harina de maíz superior y harina de maíz) y en las fracciones de pienso de maíz blanco. Los resultados mostraron que dichas micotoxinas se concentran en la fracción de pienso de maíz blanco y que el nivel de más bajo de las mismas se encuentra en la harina de maíz extra que contiene partículas gruesas, tiene una finura uniforme y un bajo contenido de grasas.

Keywords:

• fumonisins
• T-2 toxin
• maize
• milling fractions
• dry milling

Palabras claves:

• fumonisinas
• toxina T-2
• maíz
• fracciones de molienda
• molienda en seco

Introduction

Grains contaminated with mycotoxigenic molds can accumulate various mycotoxins, and consequently the safety of food and feed supplies is affected, compromising their use in animal feed and human consumption.

Maize contamination with mycotoxins is an important issue in Romania because this cereal is one of the major local crops. In fact Romania is the largest maize grower in the EU, with over 25% of total EU planted area. Anyway, considering the total production is the third European producer after France and Italy (Institutul National de Statistica, 2012). The average annual consumption of maize and maize products, in grain equivalent, is about 35 kg/capita and about 26 kg/capita, in flour equivalent (Institutul National de Statistica, 2012).

The main mycotoxigenic molds found in maize are Fusarium spp. and trichothecenes (deoxynivalenol (DON), nivalenol, T-2, HT-2, and acetylated DON derivates), zearalenone (ZEA), and fumonisins (FB) are frequently detected in corn grain and derived products (Hazel & Pacale, 2009; Montes, Reyes, Montes, & Cantu, 2009).

Trichothecenes are mainly synthesized by F. sporotrichioides and F. poae, which produce T-2 and HT-2 toxins, while DON, nivalenol and their derivatives are produced by F. culmorum and F. graminearum (Foroud & Eudes, 2009). ZEA is produced by F. graminearum, F. semitectum, F. equiseti, and F. culmorum, while FB are synthesized by F. verticilloides, F. proliferatum, and Aspergillus niger (Garrido, Hernández Pezzani, & Pacin, 2012).

As reviewed by Streit et al. (2012) FB contamination is commonly associated with maize and maize products, while DON and ZEA are detected more often in different other grains. In the last years the occurrence of aflatoxin B1 (AFB1) at high levels in maize was also detected in different zones of Europe. Anyway important change in mycotoxins distribution is expected in direct relation with climate change (Streit et al., 2012). Different studies suggest that maize is the most affected cereal by co-contamination in different combinations: AFB1 and FB (Camargos, Machinski, & Soares, 2001; Moreno et al., 2009; Rocha et al., 2009; Scudamore, Hetmanski, Chan, & Collins, 1997; Vargas, Preis, Castro, & Silva, 2001); FB and DON (Scudamore et al., 1997); FB and ZEA (Garrido et al., 2012; Vargas et al., 2001); 15-Acetyl-Deoxynivalenol, moniliformin, and ZEA (Decastelli et al., 2007); ZEA and Ochratoxin A, DON, and ZEA (Rafai, Bata, Jakab, & Vanyi, 2000).

Three different reasons were proposed by Streit et al. (2012) to explain the co-occurrence of mycotoxins (1) different mycotoxins are simultaneously synthesized by the same species of fungi; (2) commodities can be contaminated by several fungi, and (3) when dealing with co-contaminated non-cereal-based product, the completed feed might by achieved from various commodities.

Due to the high incidence of mycotoxins in food and feed supplies, and knowing the significant health risks associated with their consumption, it is important to understand the mycotoxins distribution in dry milling fractions. The incidence and levels of mycotoxins encountered in Romania have been investigated in maize grain (Curtui, Usleber, Dietrich, Lepschy, & Märtlbauer, 1998; Gagiu et al., 2007; Tabuc, Marin, Guerre, Sesan, & Bailly, 2009; Tabuc, Taranu, & Calin, 2011), but there is a general lack of data about the mycotoxins contents in milling fractions.

According to Marin, Ramos, Cano-Sancho, and Sanchis (2013), in EU in the latest years, the most frequent and highly contaminated products with FB were maize and maize products. The average level of FB content of the maize grains was 346.4 μg·kg^–1, while in the maize flour and polenta the average levels were 408.5 μg·kg⁻¹ and 182.2 μg·kg⁻¹, respectively. Regarding T-2, available data by Marin et al. (2013) indicates average levels of 7.3 μg·kg⁻¹ and maximum levels of 204 μg·kg⁻¹ for the milling products.

The aim of this study was to investigate the natural co-occurrence of FB and T-2 in maize milling fraction under industrial conditions.

Materials and methods

Maize degermination and dry milling

The study was performed on maize (M) grown in the south east of Romania. The experiments were conducted at an industrial roller mill that uses the dry milling process. Dry milling of maize is a physical process conducted in two steps. In the first step, the endosperm is separated from the germ, through an operation named degermination, by passing successively through three machines: degerminator, plansifter, and gravity tables. In the second step, the endosperm is utilized to produce various milling fractions, using roller mills, plansifter, and purifier machines.

Degermination of maize was conducted in an MHXM-M degerminator and the following products were achieved: degermed endosperm with low fat content (MF1), as main product, and hominy feed (MF2a), as by-products (Table 1). The degermed endosperm was sieve using a plansifter and three tailings resulted (Figure 1): the first tailing was conducted to the degerminator (MF3), the second tailing was conducted to the first break roll (MF4), and the third tailing was conducted to the gravity tables (MF5) to separate the germs from the flaking grits. The flaking grits, resulted by combining the MF4 and the product cleaned through separation at gravity table, are processed in order to reduce the particle size to a standard meal size with a minimum quantity of fines particles (Vanara, Reyneri, & Blandino, 2009). The milling fractions resulted from plansifter were combined giving a total of three major milled fractions: superior maize meal (MF6), milling hominy feed (MF2b), and maize flour (MF7). The extra maize meal (MF8) was obtained from the purifier machines. The MF2a and MF2b were obtained from the degermination and milling process, respectively, were combined into MF2.

Table 1. The dry milling fractions obtained through maize processing based on the Romanian specific methodology.

Tabla 1. Fracciones de molienda en seco obtenidas a través del procesado de maíz basado en una metodología específicamente rumana.

Dry milling fraction	Description
Degermed endosperm (MF1)	Coarse to fine fragments consisting of endosperm (flaking grits), small amounts of germ, hull, pericarp, tip cap, cleaned, or attached.
Hominy feed (MF2)	Product consisting of germ, hull, pericarp, tip cap, and fine endosperm particle.
First tailing (MF3)	Coarse fragments of maize kernel containing endosperm with germ and hull attached.
Second tailing (MF4)	Flaking grits – medium fragments of endosperm and some fragments of pericarp, hull, and tip cap.
Third tailing (MF5)	Flaking grits and germ – fine fragments of endosperm and fragments of germ, pericarp, hull, and tip cap.
Superior maize meal (MF6)	Medium granulated product with some germ and pericarp fragments; includes some fine granulated products. Granularity: 0% tailing from mesh sieve of 1000 μm, 12% tailing from mesh sieve of 710 μm, 19% tailing from mesh sieve of 500 μm, and 15% tailing from mesh sieve of 372 μm.
Maize flour (MF7)	Fine granulated product; results from plansifter. Granularity: 21.2% tailing from mesh sieve of 180 μm.
Extra maize meal (MF8)	Coarse granulated product from which bran and germ are removed and the fineness is uniform. Granularity: 0% tailing from mesh sieve of 1000 μm, 11.2% tailing from mesh sieve of 710 μm, 65% tailing from mesh sieve of 500, 4.4% tailing from mesh sieve of 372 μm.

Figure 1. Flow diagram of dry milling process.

Diagrama de flujo del proceso de molienda en seco.

In order to perform further analysis, the sampling was performed according to standard SR EN ISO 13690:2007 (ASRO, 2008) and samples M and MF1 to MF8 were collected from the specific points of the flow diagram of the dry milling process indicated in Figure 1.

Sample preparation

Samples of 500 g consisting of maize fractions were collected on three different days to ensure that differences caused by the milling procedure were taken into account. Immediately after sampling, the sample was characterized in terms of granulometric and chemical composition. Aliquots were stored at −18°C before performing the mycotoxin analysis. For determining the mycotoxins content, all samples, excepted for MF7, were ground in a laboratory mill to diminish the particle size in order to pass a 372-mesh sieve.

Granulometric and chemical composition analysis

The protein content was determined using semimicro-Kjeldahl method and the fat content by extraction with ether through Soxhlet method. Protein (protein factor was N × 5.7) and fat were reported on a dry basis (moisture free). The granularity was determined by sieving the product using 1000-, 710-, 500-, and 372-μm sieves.

Analysis of FB and T-2 toxins

Each sample was extracted by confinable solvent systems; 5 g of ground sample was suspended in 25-ml methanol/distilled water 70/30 (v/v) for extraction. The suspensions were first mixed vigorously for 10 min on a magnetic stirrer (Velp Scientifica), in case of FB, and 2 min in case of T-2, and afterward the extract was filtered. The mycotoxins content was determined by the competitive enzyme immunoassay (ELISA-kit). Ridascreen^® T-2 toxin test and Ridascreen^® Fumonisin test (R-Biopharm Rhone Ltd.) that were specially designed for quantitative analysis of T-2 an FB in cereals and feed were used. The concentrations of T-2 and FB were quantified according to the manufacturer’s description. The optical density of the final extracts was measured at 450 nm using ELISA 96-well plate reader and the special software RIDA^® Soft Win (R-Biopharm AG, Germany) was afterward used for mycotoxin content quantification. All sample solutions were analyzed in duplicate. According to the manufacturer’s description, the detection limit for T-2 by ELISA for cereals was 35 μg kg⁻¹, while in case of FB the detection limits was 25 μg kg⁻¹. Recovery rate corresponding to the standard fumonisin (50 and 500 μg kg⁻¹) from maize fractions was 60%; the standard curve used in the study was in the range of 25 to 2000 μg kg⁻¹. Recovery rate for T-2 from maize fractions was 90%; the standard curve used in the study was in the range of 35 to 560 μg kg⁻¹. The T-2 and FB contents were expressed in μg·kg⁻¹ initial product.

Statistical analysis

The experiments were independently performed at least three times. The statistical significance of the data was analyzed using Student’s t-test.

Results and discussion

The chemical composition of the maize milling fractions is presented in Table 2.

Table 2. Chemical composition of maize milling fractions.

Tabla 2. Composición química de las fracciones de maíz molido.

Maize milling fractions	Proteins*, %	Fat*, %
M	5.45 ± 0.15**	6.10 ± 0.20
MF1	4.80 ± 0.20	2.56 ± 0.16
MF2	6.21 ± 0.25	6.99 ± 0.24
MF3	5.41 ± 0.10	4.21 ± 0.15
MF4	4.88 ± 0.18	1.28 ± 0.10
MF5	5.77 ± 0.17	3.48 ± 0.15
MF6	4.45 ± 0.15	1.58 ± 0.12
MF7	5.55 ± 0.20	2.29 ± 0.15
MF8	4.64 ± 0.14	1.08 ± 0.10

Notes: *The results are given at 14% moisture basis.

**Standard deviation.

As shown in Table 2, the MF2 and MF3 fraction was rich in fat. The MF2 fraction is a complex fraction consisting on germ, hull, pericarp, and fine endosperm particle and it is used as feed supplies. Out of all maize milling fractions, it contains the largest amount of germ in its composition. The MF3 contains a higher fat content, but this content is lower than the whole kernel, and it is redirected to degerminator.

The fractions MF8, MF6, and MF7 represent end-products currently obtained in the dry milling process specific to Romania. MF8 is the most appreciated product by Romanian consumers and has the following characteristics: particles are coarse, fineness is uniform, the fat content is less than 1.1%, and it is obtained only from purifier machines. MF7 is used to obtain composite flours specific for producing certain baked products; fat content is higher than in MF8 and MF6.

The highest content of proteins was observed in the MF2 fraction, which also had the highest fat content.

In Table 3 are presented the levels FB and T-2 in all maize milling fractions investigated.

Table 3. Mycotoxin levels for FB and T-2 in the milling maize fractions.

Tabla 3. Niveles de micotoxina para FB y T-2 en las fracciones de maíz molido.

Maize milling fractions	FB, μg·kg^−1	T-2, μg·kg^−1
M	1334 ± 4.7*	760 ± 3.2*
MF1	600 ± 3.2	355 ± 2.1
MF2	2970 ± 5.2	1900 ± 4.8
MF3	980 ± 2.9	515 ± 3.0
MF4	386 ± 2.5	220 ± 2.0
MF5	462 ± 3.0	270 ± 2.0
MF6	162 ± 1.9	100 ± 1.5
MF7	195 ± 2.8	125 ± 1.5
MF8	56 ± 1.7	nd

Notes: The experiment was carried out in triplicate.

*Coefficient of variation; nd – below the limit of detection.

The degermination process decreased mycotoxin levels, therefore the content of T-2 and FB in MF1 fractions is 2.8 and 2.2 times lower compared to the maize whole kernel.

The distribution of FB and T-2 in the milling fractions (MF1 to MF8) indicated that the mycotoxins are concentrated in the hominy feed fraction (MF2), while the MF8 fraction contains the lowest levels of mycotoxins (Table 3). According to our results significant correlations were established between the contamination level with FB and T-2 and the fat content of the milling fractions, the correlation coefficients being of 0.902 (p < 0.05) and 0.882 (p < 0.05), respectively. Moreover, it is important to mention that the MF2 contains fine endosperm particles, and MF8 contains coarse particle and has a uniform fineness (Table 1). Our results comply with the observations of Vanara et al. (2009) who showed that the contamination level of the horny and mealy endosperm with fumonisin increased as the particle size decreased.

Our results comply with the literature indicating preferential distribution of mycotoxins in different milling fractions. For instance, Castells, Marín, Sanchis, and Ramos (2008), Katta, Cagampang, Jackson, and Bullerman (1997), Scudamore and Patel (2009), and Vanara et al. (2009) observed the high content of FB in animal meal fraction, comparatively with the fractions that contain coarse endosperm. According to these authors, the toxins are concentrated in the bran and germ. They explained that in the animal meal are present the parts from the outer layers, which are colonized by fungi, and consequently the mycotoxins are located in this part of kernel. The high fat contents of both and fumonisin could be due to the presence of the mealy endosperm, located near the germ (Vanara et al. 2009). Burger, Shephard, Louw, Rheeder, and Gelderblom (2013) report the reduction of mycotoxins in various milling end-products obtained under experimental condition; the most important accomplishment is related to the significant removal of mycotoxins from the total hominy feed fraction. These authors agreed that the mycotoxins are generally concentrated in the outer kernel layers, being mainly located in the germ, hull, pericarp, and tip cap fractions. Anyway, they suggest that country-specific milling strategies as well as the consignment of maize cause variations in the mycotoxins content.

It can observed that the content of FB in maize whole kernel was higher than T-2. After dry milling the lowest levels of mycotoxins were identified in the milling end-products (MF8, MF6, and MF7), while the highest levels of mycotoxins were found in the hominy feed (MF2). The mycotoxins contamination levels of the maize whole kernel reflected in the higher levels of FB in the MF8 and MF6 fractions compared to T-2. The maximum limits for fumonisins have been set by European Commission Regulation for milling fractions of maize with particle size > 500 μm at 1400 μg·kg⁻¹ and for particle size ≤ 500 μm at 2000 μg·kg⁻¹ (Commission regulation (EC) No. 1126/2007). According to EU recommendation (Commission Recommendation No. 165/2013), indicative levels for unprocessed maize and for direct human consumption are 200 μg·kg⁻¹ and 100 μg·kg⁻¹, respectively.

The lowest level of T-2 was identified in the MF8 fraction, being below the detection limit (Table 3).

The endosperm texture influences the quality and quantity of extra maize meal. The endosperm hardness allows obtaining the higher quantity of extra maize meal. Castells et al. (2008) and Costa, Môro, Môro, Da Silva, and Panizzi (2003) suggested important correlations between the endosperm texture and mycotoxin distribution within the milling fractions. Therefore, the high-density kernel and the compact structure of the pericarp act like an effective antifungal barrier explaining the resistance to FB contamination. When the particle size of the endosperm decreases the mycotoxin levels increase (Burger et al., 2013). Analyzing the mycotoxins contents from milling end-products we can observe that the fraction containing coarse endosperm particles (MF8) had lower contamination levels compared with the fraction with fine particles (MF7) (Table 3). Practically, the MF8 fraction arises from the endosperm and has lower levels of mycotoxins, compared to the MF7. These results indicate that the transfer of the mycotoxins to the inner structure of the kernel is reduced because of the outer layer which acts like a physical barrier. On the other hand, due to its specific hardness, the endosperm is less susceptible to breaking and cracking and is more like to allow obtaining coarse milling fractions (Burger et al., 2013). The MF6 fraction contains medium size endosperm particles and some germ and pericarp fragments, include some fine granulated fractions, and in consequence have higher mycotoxins content than the MF8 (Table 3).

Conclusions

FB and T-2 were identified in all tested maize samples. The level of contamination in the milling fractions depends on the initial content of mycotoxins in the whole maize. The highest levels of mycotoxins were identified in the hominy feed fractions, whereas in the fractions designated to human consumption (extra maize meal, superior maize meal, and maize flour) the FB and T-2 levels were lower. The results indicated that granularity influences the contents of mycotoxins in maize meal end-products. The lowest level of mycotoxins was identified in extra maize meal that contains coarse particle, has uniform fineness and low fat content.

Disclosure statement

The authors declare that they have no conflict of interest.