The Nutritional Value and Safety of Genetically Unmodified Soybeans and Soybean Feed Products in the Nutrition of Farm Animals

: Post ‐ extraction soybean ( Glycine max (L.) Merr.) meal is widely used as a basic protein feed for farm animals, especially poultry and pigs. Products made from unmodified soybean seeds are an alternative to imported GMO soybean meal. The aim of the study was to develop feed products from popular European varieties of genetically unmodified soybeans, which can be produced on small and medium ‐ sized farms, and to assess their nutritional value and safety to livestock. The research was conducted on the seeds of three soybean varieties and two types of feed products resulting from thermobaric treatment (extrudate) and oil pressing (soybean press cake). The mould and yeast contamination of domestic seeds was negligible. The thermobaric and pressing treatments lowered the content of fungi by 97%. The products were considered free from mycotoxins. In comparison with full ‐ fat soybean seeds, the protein content in the products was up to 19% higher, and 92% of the total lysine remained available. The products had lower content of antinutritional ingredients (trypsin inhibitors) and the urease activity was reduced by 52–59% and 99%, respectively. The experiment showed that the European genetically unmodified soybean feed products were characterised by good quality, mycotoxicological for farm animals.


Introduction
Post-extraction soybean meal is widely used to feed various species of farm animals and currently it is used as a basic protein feed for poultry (Galus galus domesticus), pigs (Sus scrofa domestica) and cattle (Bos taurus Taurus) [1]. It is relatively expensive when imported to Europe from Brazil and Argentina, where soybeans are cultivated in vast areas, which used to be occupied by tropical forests. Post-extraction soybean meal largely comes from genetically modified plants, which is a matter of concern to some consumers. According to public opinion polls, 56.8% of consumers prefer food products from animals receiving feed without genetically modified feeds, 65% do not think that food containing or produced from genetically modified plants is safe, 42% of consumers definitely and 23% moderately support a ban on genetically modified crops [2].
Products made from unmodified soybean varieties grown in Europe are an alternative to genetically modified soybean meal [3]. In order to maintain animal production at a high-level animals need to be fed properly to satisfy their demand for nutrients, including protein, which is an important and expensive ingredient. In 2016 import of genetically modified soybean meal covered about 65% of the feed protein consumption by farm animals, whereas the yield of legume seeds, such as yellow lupin (Lupinus luteus L.), faba beans (Vicia faba L.), peas (Pisum sativum L.), covered only about 11% of the feed protein consumption [4]. The area of genetically unmodified soybean plantations is increasing rapidly [5]. Moreover, soybean oil pressing products have become more available. There are more and more varieties with improved chemical composition, i.e., with high content of protein (37-41% DM) and fat (22)(23), and with reduced content of antinutrients. Modern soybean varieties are better adapted to the climate and soil conditions in northern countries, which is manifested by higher yields. The intensity of selection works on genetically unmodified soybean varieties is evidenced by their increasing number in registers and on recommendation lists [6]. The final composition of the seeds is known to vary by genotype, region and the environmental conditions during seed development [7]. In comparison with cereals, the contamination of American or Asian soybean seeds with mycotoxins is not considered a serious problem due to the local climates [8]. Mycotoxins (deoxynivalenol, zearalenone, trichothecenes) can be a major challenge in northern countries because soybean seeds are harvested late, when temperature is low, whereas humidity is elevated.
It is necessary to provide a wide range of new sources of protein feeds and test their properties because genetically modified feeds may be banned in the near future due to increasing consumer pressure [9][10][11][12]. Home-grown protein feeds, especially those that can be prepared on small or medium-sized farms, are also essential for organic livestock production, because there is an increasing number of informed consumers interested in the quality and origin of food. However, breeders and feed producers face the problem of low availability of feed products made from local genetically unmodified soybeans, because it is mainly large industrial plants that process large amounts of seeds for feed companies. Such plants require large batches of homogeneous material, which is difficult to obtain at the current stage of development of local soybean cultivation. On the other hand, small farms significantly support rural communities by offering employment and they play various key functions in rural economy. There is a demand for farms processing small amounts of seeds obtained from their own crops to feed their own livestock. Individual farms are increasingly interested in purchasing the equipment which will help them to use soybean seeds as a high-protein feed (oral information from manufacturers of equipment for processing small amounts of seeds). Therefore, farmers should be encouraged to use small-scale seed processing equipment, e.g., toasters to make products of an increased protein nutritional value to feed cattle, extruders to make full-fat products of an increased protein and energy nutritional value by means of extrusion or skimmed extrusion products of an increased protein nutritional value, as well as screw presses to make high-quality press cakes [13].
The main goal of our study was to assess the quality of different genetically unmodified varieties of soybean seeds and the products made from these seeds in terms of their safety, nutritional value and usefulness for feeding farm animals. The assumption was to develop soybean seed products that can be made on small and medium-sized farms. The results of the study will broaden the market of protein feeds as an alternative to feeds derived from genetically modified plants. They will help local markets to become independent of imported genetically modified feed products and to meet the expectations of consumers looking for products made from animals not receiving genetically modified feeds.

Soybeans Seeds
The research was conducted on the seeds of three soybean (Glycine max (L.) Merr.) varieties (Erica, Petrina and Viola-breeder: company DANKO Hodowla Roślin sp. z o.o.) recommended for cultivation in the climate and soil conditions of northeastern Europe as well as two types of feed products made as a result of thermobaric treatment (extrudate) and oil pressing (press cake).

Heat Treatment of Soybean Seeds
Soybean press cake was made from full-fat soybean seeds by means of a production line at the Department of Biosystems Engineering, Poznań University of Life Sciences, Poland ( Figure 1). The production line was composed of the following machines and devices: • extruder, • screw conveyor with a water vapour extraction hood, • screw press, • roller mill.

Extrudate Drying
Before oil pressing the extrudate was dried with a 2.2 kW conveyor with a water vapour extractor equipped with a 1.1 kW fan and two PT-100 temperature sensors (installed at the beginning and at the end of the conveyor).

Oil Pressing
Oil was pressed from the soybean extrudate by means of a PS-60 screw press (AgroFeedingTech, Poznań, Poland). The press had an inverter controlling the rotational speed of the working shaft and two PT-100 temperature sensors (one on each working chamber). The technical specifications of the press were as follows: year of manufacture: 2020; drive motor power (kW): 7.5; screw rotational speed (rpm): 42; exit slot (mm): 7; and yield (kg.h −1 ): 40-50.

Analyses
First the soybean seeds from domestic crops were checked for the absence of genetic modifications. The DNA of a sample was isolated according to the I-02/3 edition of 4 January 2016, using the method based on CTAB (cetyltrimethylammonium bromide). The genetic analyses are shown in Table 1. The analytical procedure was prepared according to ISO/TS 21,098 and the Joint Research Centre validation products [16,17]. The soybean seeds and feed products were tested for the presence of mould and yeast cells in October 2020, i.e., 36 months after the harvest of Erica seeds and 24 months after the harvest of Petrina and Viola seeds. The mycological analysis was conducted according to the standard method PN-ISO7954:1999 with the authors' modification [19]. The surface culture test was performed in triplicate. Soybean samples were plated on YGC agar medium composed of yeast extract and glucose with 100 ppm chloramphenicol, pH 6.6. The dishes were incubated at 25 °C for 5 days.
The presence and content of mould metabolites in the soybean seeds and feed products was measured in a mycotoxicological test. High-performance liquid chromatography (HPLC) with fluorescence detection was used for the analysis of aflatoxins and ochratoxin A-the samples were purified on AflaTest immunoaffinity columns (Vicam) for aflatoxins and OchraPrep (R-Biopharm Rhóne Ltd.) for ochratoxin A according to the procedure provided by the producers. The HPLC-MS/MS method was used for the analysis of fumonisins-the samples were purified on MultiSep ® 2II Fum columns (Romer Labs ® ), according to the procedure provided by the producer. The HPLC-MS/MS method was used for the analysis of deoxynivalenol, nivalenol, diacetoxyscirpenol, zearalenone, T2 and HTż toxins-the samples were purified on Bond Elut ® Mycotoxin columns (Agilent). Mycotoxins were detected by means of a Nexera high-performance liquid chromatograph (HPLC) (Shimadzu, Tokyo, Japan) with an API 4000 mass detector (Sciex, Foster City, CA, USA). The mycotoxins were separated on a Gemini NX-C18 chromatographic column (150 mm × 4.6 mm, 3 μm) (Phenomenex, Torrance, CA, USA). The flow rate was 0.75 mL/min, and the injection volume was 7 μL. Mobile phases were: A = 1% AcOH in water and B = 1% AcOH in methanol (both phases contained 5 mmol/L ammonium acetate) with the following gradient: 10% B up to 2.0 min, 97% B 2.0-14.0 min, 97% B up to 16.0 min, then 10% B.
The chemical compositions of the feeds were analysed according to the AOAC methods (2005) [20]. Amino acids were analysed with an AAA 400 automatic analyser (INGOS Ltd., Prague, Czech Republic). The essential amino acid index (EAAI) and limiting amino acid score (AS) were calculated according to the Oser (1951) model. The content of available lysine as a percentage of total lysine was measured according to the Polish standard PN-ISO 5510:2000 [21].
The content of minerals in the feed was measured by flame atomic absorption spectroscopy, according to the Polish standard PN-EN ISO 6869:2002 [22]. Samples were mineralised in a mixture of oxidising acids in a microwave dryer. After evaporation of the acid mixture the samples were dissolved in hydrochloric acid and filtered. After filtration and dilution the solution was aspirated into the flame of an air-acetylene atomic absorption spectrometer. The absorbance of the sample was measured and compared with the absorbance of the calibration solutions. An AVANTA Σ spectrometer (GBC) was used for the analyses.
Fatty acids in all the feeds were analysed as methyl esters by means of gas chromatography (Shimadzu GC 2010 Plus, Japan). Samples were extracted with a chloroform/methanol mixture (2:1 v/v), as described by Folch et al. [23]. Fatty acids were saponified (0.5 N NaOH in methanol, 80 °C) and next they were esterified with boron trifluoride/methanol (ISO 12966-2:2011). After extraction with hexane the compounds were separated on an RTX-2330 column (105 m, 0.32 mm, 0.2 μ, flow rate: 4 mL . min −1 ) with a flame ionisation detector and AOC-20i autosampler (Shimadzu, Japan). The average extraction efficiency was 97.1%. The temperature programme in the column oven was as follows: 60-120 °C (20°.min −1 ) and then 240 °C (3°.min −1 ). The injector and detector were set at 250°C. Fatty acids were identified by comparing their retention times with those of methylated FA standards (Supelco Inc., PA, USA). Fatty acids were calculated with chromatogram peak areas and expressed as g/100 g fatty acid methyl esters. The iodine value was calculated by means of the following equation [24]: The saturation index (S/P) of soybean seeds and fat products was calculated according to the formula presented by [25]: S/P = (C14:0 + C16:0 + C18:0)/(MUFA + PUFA) (2) where, MUFA is the sum of monounsaturated fatty acids and PUFA is the sum of polyunsaturated fatty acids.

Statistical Analyses
The results were analysed statistically by means of one-way analysis of variance. The means were compared with Duncan's multiple range test at a significance level p ≤ 0.05. All analyses of variance were conducted with the Statistica 12 package (Copyright © StatSoft, Inc. 1984-2014, Texas, USA).

Results
There were no: target sequence of the 35S promoter, NOS terminator, PAT gene, BAR gene or CTP2-CP4-EPSPS construct in the seeds of the Petrina and Viola soybean varieties. The target sequence of the 35S promoter was found only in the sample of the Erica seeds. This may indicate the presence of genetically modified organisms or may be derived from a natural donor, i.e., the cauliflower mosaic virus (CaMV), whose DNA was identified in the seeds of this variety. The DNA of modified soybean lines DP 356043, DP 305423, CV 127, MON 87708, MON 87701, MON 87769 was not found in any of the seeds. This means that the selected soybean seeds were not genetically modified.
The results of the mycological examination of soybean seeds are shown in Table 2. The fungi of the Alternaria or Penicillum genera were found in the soybean seeds of the Erica and Viola varieties. The seeds of the Petrina variety had fungi of the Eurotium genus only.  Table 3 lists the content of individual mycotoxins in the soybean seeds. None of the varieties contained ochratoxin, fumonisins, aflatoxins, nivalenol, diacetoxyscirpenol, T2 or HT2 toxins. None of the varieties had higher deoxynivalenol level in the seeds than 0.3 ppb. There were trace amounts of zearalenone only in the Erica and Petrina soybean seeds.
The total count of mould and yeast colonies was measured only in the raw seeds of the Petrina variety, and in the feed products formed after the thermobaric treatment of Petrina soybean seeds (extrudate) and the extraction of oil from them (press cake) ( Figure  2). This was linked to organoleptic evaluation of soybean seeds, after which fungal and mould infestation was found only in soybean seeds of the Petrina variety. When raw seeds were exposed to temperatures of 86.8 °C, 115.2 °C, 143.7 °C and 144.2 °C in individual chambers of the extruder, the content of living fungi decreased by 94.3%. When oil was pressed from the extrudate at working chamber temperatures of 120.8 °C and 109.5 °C, the content of fungi decreased by 43%. As a result, the content of fungi in the soybean press cake was 97% lower than in the raw seeds. This amount of fungi and yeast spore colonies, both in the seeds and feed products, can be considered negligible and the products can be considered free from mycotoxins. To sum up, at the time of the analysis the products did not pose a threat in terms of mycotoxin contamination. The total count of mould and yeast colonies in raw soybean seeds of the Petrina variety and in the feed products made from these seeds. The percentage difference was calculated with reference to the raw seeds in which the count of mould colonies was assumed to be 100%.
All the three soybean varieties had a similar content of trypsin inhibitors (from 16.2 to 18.4 mg/g), but the highest content of trypsin inhibitors was noted in the soybean seeds of the Petrina variety ( Table 4). The thermal treatment (extrusion) significantly reduced the content of this antinutrient to 9-11 mg/g in the extrudates, as compared with the raw seeds. The differences (decrease) in the content of trypsin inhibitors between the raw and extruded soybean seeds amounted to 30.2% (Erica), 43.5% (Petrina) and 45.2% (Viola) (p < 0.001). It is noteworthy that the soybean seeds of the Viola variety were the most susceptible to temperature and they deactivated trypsin inhibitors most effectively. The pressing of oil from the extrudate and the preparation of a soybean press cake resulted in a further decrease in the content of trypsin inhibitors-from 6.6 mg/g in the press cake made from the Erica seeds to 8.6 mg/g in the press cake made from the Petrina seeds. In comparison with the raw seeds, the feed material in the form of soybean press cake was characterised by reduced activity of trypsin inhibitors, which amounted to 59.3%, 53.3% and 52.4% for the Erica, Petrina and Viola varieties, respectively (p < 0.001). Table 4. The content of trypsin inhibitors (mg . g −1 feed) in the raw soybean seeds and in the feed products subjected to technological processes. The urease activity in the raw soybean seeds (over 4 mgN•g −1 •min −1 ) and in the feed products made from these seeds is shown in Table 5. The urease activity was almost completely (by about 98-99%) reduced by temperature during the seed extrusion, regardless of the soybean cultivar (p < 0.001). Total count of fungal colony (yeast + moulds) Table 5. The urease activity in the soybean seeds and processed feed products. The percentage difference shows a decrease in the urease activity in the processed feed materials, as compared with the raw seeds, in which the urease activity was assumed as 100%. In contrast to the raw seeds, the processed feed products had a higher content of total protein (3.4% higher in the extrudate and 19% higher in the press cake), ash (7.7% higher in the extrudate and 28% higher in the press cake) and nitrogen-free extracts, whereas the content of crude fibre was lower (44% lower in the extrudate and 33% lower in the press cake) (p < 0.001), as shown in Table 6. Due to the oil extraction the content of crude fat in the press cake was lower than in the raw and extruded seeds (p < 0.001). The technological processing of soybean seeds did not significantly affect the content of amino acids in the protein of these products (Figure 3). The protein of all these feed products was mainly composed of glutamic acid (16-17%) and aspartic acid (10%), as well as arginine and leucine (8-9%). The protein contained slightly more than 6% of lysine. Among the essential amino acids, sulphur-containing amino acids Met + Cys have the limiting effect. Their AS index was 42% in the seeds and 41-45% in the feed products. The index for lysine was the highest (about 106-107%).

Type of Material
On average, the availability of lysine for all the three soybean varieties was reduced by about 8-9% both by the effect of temperature during the production of the extrudate and by the subsequent oil extraction during the production of the press cake ( Table 7). However, regardless of the soybean seed treatment technology, 91-92% of the total lysine remained available for pigs. The biological value of the protein, in relation to the reference egg white protein (EAAI), was higher in the processed feed products than in the raw seeds (78% vs. 76%).
The total content of all products in the extrudate measured in the experiment did not differ significantly from the amount of these elements in the raw seeds ( Table 8). The only exception was iodine, whose content in the feed materials was slightly lower than in the raw seeds (p < 0.001). After pressing the oil, the total content of minerals in the soybean press cake increased by 10% in the dry matter and by 16% as the feed basis, as compared with the raw seeds (p < 0.001). The highest increase was observed for sodium-100% (p > 0.05), iron-61% (p = 0.0001) and cooper-31% (p > 0.05). The amount of other minerals increased by 15-19%, as compared with their content in the raw seeds. Figure 3. The content of amino acids (g) in 100 g of protein in raw soybean seeds and processed feed products. Table 7. The content of total and available lysine and the index of protein value of the raw soybeans and processed feed products (the mean values from 3 soybean varieties). EAAI-essential amino acid index. Table 8. The content of minerals in the raw soybean seeds and feed products and the reference to crude ash and dry matter (the mean values from 3 soybean varieties ± SD). The thermobaric treatment of the soybean seeds influenced their fatty acid profile (Table 9). In comparison with the raw seeds, the content of saturated fatty acids in 100 g of all acids under analysis increased by 13%-14%, whereas the content of unsaturated fatty acids decreased by about 2% (p < 0.001). The MUFA remained unchanged, whereas the PUFA level dropped by 4.8-6.4% (p = 0.0174). The n-6/n-3 PUFA ratio in the extrudate and press cake decreased by 13-14%, as compared with the raw seeds (p = 0.0009). These changes slightly increased the degree of fat saturation in the extrudate and press cake, as compared with the raw seeds-the S/p ratio was 16% higher, whereas the iodine value was about 3% lower (p < 0.05) ( Table 9). A higher degree of fat saturation indicates its greater resistance to oxidation, which translates into better durability during storage, and thus a higher quality of fat in the extrudate and press cake than in the raw seeds. Table 9. The fatty acid profile and fat quality indicators for the raw soybean seeds and feed products (the mean values from 3 soybean varieties ± SD).

Discussion
The main problem concerning the replacement of soybean meal made from genetically modified soybeans in the nutrition of fast-growing, high-production animals is their high requirement for protein. This problem particularly concerns animals with the highest demand for nutrients, i.e., feeding sows, high-yield cows, as well as piglets, weaners and calves. The latter groups include young, fast-growing animals, for which the weaning period is a critical stage of rearing. Such animals need a feed which not only meets their requirement for nutrients and energy but is also free from antinutritional components and of good microbiological and toxicological quality.
Like other raw products, soybean seeds can be contaminated with fungi while growing in the field, during storage and/or processing. This is a serious global problem because fungal contamination may damage seeds and result in low germination, discoloration, off-flavours, softening and rotting. In addition, even small amounts of fungal secondary metabolites, known as mycotoxins, are toxic to animals when consumed or inhaled. They may cause serious production loss and they also exhibit mutagenic, carcinogenic and estrogenic properties [26]. Piotrowska et al. [8] summarised the results of monitoring in the most important regions of soybean production. In North America the most common species of fungi found in soybean seeds were: Aspergillus, Fusarium, Chaetomium, Penicillium, Alternaria and Colletotrichum. Aspergillus flavus and A. ochraceus were the most prevalent fungal species in soybean seeds in Ecuador. The most common Fusarium species were F.verticillioides and F.semitecium. In eastern Argentina, the most prevalent species were Aspergillus flavus and Fusarium verticillioides. The authors concluded that the most fungi were found in mature seeds prior to storage and only about 10% of them was commonly referred to as storage fungi. Among all the isolated fungal species 93.2% was toxigenic [8]. The contamination frequency varied by year and depended more on the environmental and cultural practices than on the variety. Our study revealed the presence of fungi in the soybean seeds. There were fungi of the Alternaria and Penicillum genera in the seeds of the Erica and Viola cultivars. The seeds of the Petrina cultivar contained fungi of the Eurotium genus only. The mycological state of feed can be considered good when it contains less than 10 3-10 5 CFU per gram [27]. In our study the total counts of fungal colonies in the soybean seeds were very low, i.e., <50 or 100 CFU/g. It was 2.2 × 10 3 CFU/g only in one variety. Fungi of the Alternaria and Penicillum genera are potentially toxic, whereas Eurotium is considered to be mild. According to Frisvad and Samson [28], this genus does not produce mycotoxins, although Weidenbörner [29] reported that Eurotium moulds can produce ochratoxin A. Contrary to these observations, where soybeans were infected mainly with field fungi, in our experiment the growth of fungi occurred during the transport and storage of the seeds. Even though seeds may be in good condition after the harvest in autumn and the count of fungi may be very low, spores may start the focal formation of ochratoxin A during longer storage (e.g., until spring).
The contamination of soybeans with mycotoxins is not considered a significant problem around the world in comparison with the fungal contamination of maize (Zea mays), cottonseeds (Gossypium L.), peanuts (Arachis hypogaea L.), barley (Hordeum L.) and other grains. The study conducted by the US Department of Agriculture on soybean samples collected from different regions of the United States revealed low levels of aflatoxin (7-14 ppb) in only 0.2% of the samples [8]. Valenta et al. [30] detected aflatoxin (AFB1) in 63% of non-suspicious samples, but the maximum concentration was only 0.41 ppb. ZEA was detected in 45% of the samples (the maximum concentration was 18 ppb). DON was detected only in one suspicious sample at a low concentration of 104 ppb. OTA was detected in 10% of the samples (the maximum concentration was 1 ppb). Binder et al. [31] conducted a study on soybeans in the Asian and Pacific region. The researchers found aflatoxin in only 2% of the samples (the maximum concentration was 13 ppb), zearalenone-in 17% (the maximum concentration-1078 ppb), ochratoxin in 13% (the maximum concentration-11 ppb), DON in 7% (the maximum concentration-1347 ppb) and fumonisins in 7% (the maximum concentration-331 ppb). Our study did not reveal the presence of ochratoxin, fumonisins, aflatoxins, nivalenol, diacetoxyscirpenol, T2 or HT2 toxins in any of the cultivars. The deoxynivalenol level in the seeds of no variety exceeded 0.3 ppb. There were trace amounts of zearalenone only in the seeds of two varieties. To sum up, at the time of analysis none of the products was hazardous in terms of mycotoxin contamination. The amount of fungi and yeast spore colonies found in both the seeds and feed products were negligible and these products can be considered free from mycotoxins. Our research results confirmed the purity and nutritional safety of the seeds of the most popular soybean varieties in Poland. Therefore, even high proportions of domestic soybean products can be used in mixtures for pigs, poultry or calves without exceeding the maximum permissible mycotoxin content (Commission Recommendation EC 576/2006), assuming that the other components are equally clean.
It is cheaper to use full-fat soybean seeds in mixtures for farm animals because it is not necessary to use additional fat and the fat content in the feed mixture can be significantly reduced. However, raw seeds contain anti-nutritional substances, which are harmful especially to pigs and poultry. Soybean seeds contain trypsin and chymotrypsin inhibitors, lectins, antigenic proteins, oestrogens, saponins, phytic acid and non-starch polysaccharides, which are classified as substances exhibiting anti-nutritional effects on monogastric animals [32]. The use of raw soybean seeds is also limited by some endogenous enzymes which are necessary for seed germination. These are lipases and lipoxygenases as well as urease-the enzyme hydrolysing urea [33]. These compounds, depending on their concentration, may be toxic to microorganisms in the rumen. Therefore, raw soybean seeds cannot be used to feed high-yielding cows and calves under the age of 4 months. Raw soybean seeds contain protease inhibitors (trypsins, chymotrypsins), which inhibit the action of protein-digesting enzymes and thus limit the use of protein by monogastric animals [32]. These factors do not have such a significant influence on the use of soybean feed for ruminants due to the specificity of metabolism occurring in the rumen, where it is possible to deactivate anti-nutritional substances. All the three soybean varieties analysed in our study had a similar content of trypsin inhibitors (16-18 mg•g −1 ). The heat treatment of the seeds and the preparation of the extrudate reduced the content of trypsin inhibitors to 9-11 mg/g. The extraction of oil from the extrudate and the preparation of the soybean press cake further reduced the level of trypsin inhibitors to 8-6 mg•g −1 . It was 65-53% lower than in the standard GMO soybean meal analysed in this study (2.8 mg•g −1 ).
Chen et al. [34] also observed a lower amount of trypsin inhibitor and reduced urease activity in differently processed soybean seeds, as compared with raw soybean seeds. The trypsin inhibitor ranked as follows: raw soybean seeds > expeller soybean meal (SBM) > full-fat-extruded SBM ≥ heat-inactivated full-fat SBM > solvent-extracted SBM. The urease activity ranked as follows: raw soybean seeds > heat-inactivated full-fat SBM > full-fatextruded SBM ≥ solvent-extracted SBM ≥ expeller SBM [34].
Raw soybean seeds also contain urease-the enzyme which hydrolyses ammonia from urea. The grinding of soybean seeds, which needs to be done before they are included in the concentrate mixture, increases the urease activity. During the seed extrusion process in our experiment temperature effectively reduced the urease activity by about 98-99%. The urease activity of the extruded soybean seeds was as low as in the standard soybean meal analysed in this study (≤0.05 mgN•g −1 •min −1 ). A sudden appearance of a large amount of ammonia in the gastrointestinal tract deteriorates the free intake of dry matter, it may damage the liver function, and in extreme cases it may cause poisoning [33]. A similar decrease in the urease activity (by 96-97%) was observed by Jaśkiewicz et al. [35], who treated various varieties of domestic soybean seeds with a temperature of 121 °C for 20 min. After 30 min the urease activity was completely reduced. The maximum urease activity level recommended for feed products is <0.4 mgN‧g −1 ‧min −1 (Official Journal 2005, No.16, item 137). In our experiment the urease activity in the raw seeds exceeded 4 mgN‧g −1 ‧min −1 , but in extruded seeds it ranged from 0.08 to less than 0.05 mgN‧g −1 ‧min −1 , depending on the variety. Veum et al. [36] confirmed the fact that after proper thermobaric treatment of soybeans the content of anti-nutritional ingredients, especially trypsin inhibitors, was significantly reduced. The treatment improved the digestibility of this soy product and resulted in high protein digestibility and energy.
The results of experiments conducted on soybeans in Poland showed that depending on the variety, the total protein content ranged from 360 to 433 g‧kg −1 of dry matter, whereas the fat content ranged from 200 to 240 g‧kg −1 [6]. In our experiment the content of these components was slightly lower; the average content of protein in the three varieties of seeds was 350 g‧kg −1 , whereas the fat content was 160-190 g‧kg −1 . The fat content and fatty acid profile of soybeans are believed to depend not only on the variety but also on the weather conditions during the growing season. Pisulewska et al. [37] observed that at low temperatures and high air humidity the crude fat content in the seeds increased and the fat was characterised by a lower content of saturated fatty acids and a higher content of polyunsaturated fatty acids. The crude fat and protein content in the seeds analysed in our experiment was lower than average, and this may have been caused by the dry and hot weather conditions during the growing season. Our observation is in line with the findings of the research conducted by Rotundo and Westgate [7], who found that water shortage during the soybean seed filling period decreased the crude fat content by up to about 35% and that elevated temperature reduced the crude fat content by over 15%. Water shortage and high temperature had less negative influence on the protein content in soybean seeds (3-7%).
In our experiment the total protein content in the soybean feed products was higher than in the raw seeds (3.4% higher in the extrudate and 19% higher in the press cake). The oil pressing reduced the crude fat content in the press cake (88 g‧kg −1 ), which was lower than in the raw (175 g‧kg −1 ) and extruded seeds. Similarly, Milczarek et al. [38] used domestic soybean seeds and observed that the content of protein (305 g‧kg −1 ) and fat (221 g‧kg −1 ) in the raw seeds was lower than in the extruded ones (338 and 217 g‧kg −1 , respectively). The lower crude fat content reduces the energy value of feeds for pigs and poultry. However, in ruminants the fat in soybeans undergoes changes in the rumen, where fatty acids are released, and unsaturated acids undergo biohydrogenation. If there is too much fat in the diet, these transformations disturb the course of fermentation in the rumen. If the fat content is higher than 4% of the dry matter of the feed ration, it may be toxic to microorganisms in the rumen of young cattle; for adult cattle the fat content limit is 6% [39]. When composing a diet containing whole, full-fat raw soybeans, it is necessary to remember that such a form of seeds protects unsaturated fatty acids from biohydrogenation in the rumen [40] and that the amount of crude fat in seeds limits their content in the diet [39]. As mentioned above, the processed seeds of locally grown genetically unmodified soybean varieties can be an alternative feed material for farm animals replacing post-extracted soybean meal. The research results show that the soybean press cake is the most similar product to soybean meal. The soybean meal analysed in our study contained 452 g‧kg −1 of crude protein and 19 g‧kg −1 of crude fat. This content was close to the mean values in tables listing the chemical composition and nutritional value of feed: 466 g‧kg −1 and 14 g‧kg −1 , respectively, for crude protein and crude fat. Soybean meal contains on average about 27 g‧kg −1 of lysine, 14 g‧kg −1 of methionine with cysteine, 18 g‧kg −1 of threonine and 6 g‧kg −1 of tryptophan [41,42], which are the most important essential amino acids for farm animals, especially for monogastric ones.
The technological processing of soybean seeds in our experiment did not significantly affect the composition of amino acids in the protein of the feed products. 100 g of protein contained the most glutamic and aspartic acids, which were followed by arginine and leucine. Lysine constituted slightly more than 6% of the protein. Among the essential amino acids, sulphur-containing amino acids were the limiting ones (CS index above 40%). It is known that during heat processing the amino group of lysine can react with reducing sugars and form biologically unavailable lysine derivatives (unreactive Lys). During the production of the extrudate in our experiment temperature did not affect the amount of total lysine but limited its availability in all the varieties by about 8-9%. Jaśkiewicz et al. [35] treated the seeds of three domestic soybean varieties with a temperature of 121°C and pressure of 0.25 MPa for 10, 20, 30 or 2 × 30 min and observed that after 10 min the amount of available lysine decreased by 2-4%. After 30 min the decrease ranged from 6% to 25%, depending on the soybean variety. Fontaine et al. [43] noted that toasting damaged reactive lysine by about 15%, as compared with untoasted soybean seeds. Kim et al. [44] observed that the total lysine content in soybean meal heated at 135°C decreased linearly along with the heating time. After 28 min the loss of total lysine amounted to about 30%, while the amount of reactive lysine decreased by 50%.
The biological value of fat mostly depends on its content and fatty acids profile. An excess of n-6 acids in the diet and a very high ratio of n-6 to n-3 acids are considered a factor causing various diseases, including cardiovascular, inflammatory and autoimmune diseases as well as cancers [45] A high n-3 PUFA level and a low n-6/n-3 ratio has the opposite effect. In the n-3 PUFA group α-linolenic acid (ALA), which is converted into EPA and DHA, is the dominant acid in the diet, but the scale of its conversion in the human body is relatively low. It is estimated that about 5-10% of ALA is converted into EPA, and less than 1% into DHA [46]. The thermobaric treatment of the soybean seeds and oil pressing influenced their fatty acid profile. In comparison with the raw seeds, the SFA level increased by 13-14%, whereas the UFA level decreased by about 2%. The MUFA level increased slightly, whereas the PUFA level decreased by 4.8-6.4%. It was a good tendency for the dietary value of the feed products, as the n-6/n-3 PUFA ratio in the extrudate and press cake was 13-14% lower than in the raw seeds. The fat saturation of the extrudate and press cake increased, as compared with the raw seeds. A higher fat saturation indicates greater resistance to oxidation, and in consequence fat is characterised by better quality and stability during storage [47]. In our experiment the extrudate and press cake exhibited greater resistance to oxidation than the raw soybeans, where the fat was slightly more unsaturated.

Conclusions
The research confirmed the good quality and mycotoxicological purity of genetically unmodified soybeans. The feed products obtained from full-fat soybean seeds and subjected to thermobaric treatment (extrudate) and oil pressing (press cake) had a lower content of anti-nutritional ingredients, including trypsin inhibitors, whereas the urease activity was reduced by 52-59% and 99%, respectively. The mould and yeast contamination of the seeds was negligible. The thermobaric and pressing treatments lowered the fungal content by 97%. The products were considered free from mycotoxins. In comparison with the full-fat soybean seeds, the protein content in the products was up to 19% higher, and 92% of the total lysine remained available. Even though the level of anti-nutritional substances was reduced in both feed products, the soybean press cake, which contained less fat, may be more beneficial for ruminants and the technological groups of pigs which do not require high concentration of energy in the feed mixture or a high ratio of protein to metabolic energy. The analysis confirmed the good quality and nutritional value of the feed products made from local genetically unmodified soybean seeds. They can be a safe and valuable alternative to genetically modified soybean meal in the nutrition of farm animals. They may also help to popularise the cultivation of genetically unmodified soybeans and the processing of full-fat soybeans for feed products in northern countries.