GMO and non-GMO maize and soybean samples were sent to independent laboratory Kogenebiotech Co., LTD (Seoul, Korea) for GMO analysis. GMO qualitative analysis of maize was performed by polymerase chain reaction (PCR) with the specific primer pairs for SSIIb (reference gene), 35S Promoter, NOS Terminator, DP-098140-6, and DAS-40278-9 genes respectively. GMO qualitative analysis of soybean was performed by PCR with the specific primer pairs for Lectin (reference gene), 35S Promoter, NOS Terminator, MON89788, DP305423-1, DO356043-5, MON87701, CV127, MON87708, MON87769, and DAS-68416-4 genes respectively. The results are shown in Table 1 and 2. The results confirmed that the maize and soybean meal were non-GMO.

Samples of each of the four lots of ingredients were used to carry out proximate analysis [12]. All raw materials formulated (non-GMO and GMO maize and soybean) in diet were provided by Daehan feed mill company which were imported from USA. Amino acid contents were determined, following acid hydrolysis with 6 N HCl at 110°C for 24 h, using an amino acid analyzer (Biochrom 20, Pharmacia Biotech, Cambridge, England) (Table 3).

A total of 840 male Ross 308 two-day-old (body weight [BW] of 43.03±5 g) broiler chicks were obtained from a commercial hatchery (Yang Ji Company, Cheonan, Korea). All birds were randomly assigned into 2 dietary treatment groups by BW in a randomized complete block design. Each treatment had 28 replicate pens of 15 broilers in each pen. i) Trt 1, GMO maize-soybean meal based diet; ii) Trt 2, non-GMO maize soybean meal based diet. Both diets were maize-soybean meal diets. All birds were kept in stainless steel pens of identical size (1.75× 1.55 m) in a house with concrete floors covered with clean rice bran in an area that was provided with continuous light and were given free access to water and mash feed. During the experiment, house’s temperature was regulated around 32°C. All diets were formulated to contain approximately equal amounts of the 3 first limiting dietary essential amino acids (methionine, cystine, and lysine), Ca, absorbable P, and Na, based on the analytical data from the feedstuffs. All diets were formulated to meet or exceed the NRC [13] requirements for broilers. Chromium contents were 330 and 320 μg/kg in starter and finisher diets, respectively, as measured by atomic absorption spectrophotometer (Analyst 100, Perkin-Elmer, Norwalk, CT, USA). The initial and final diet compositions are shown in Table 4. All birds used in this trial were handled in accordance with the guidelines set forth by the Animal Care and Use Committee of Dankook University.

The broilers were weighed by pen and feed intake (FI) was recorded on d 0, 7, 17, and 32. This information was then used to calculate body weight gain, and FCR. For deaths during the middle of a weighing period, the dead animal’s weight was recorded, and the gain of the dead bird was counted towards pen gain in figuring feed conversion. Number of dead birds was examined as well. At the end of the experiment, 56 broilers were randomly selected from each treatment (2 birds per pen) and blood samples were collected in 5 mL vacuum tubes (Becton Dickinson Vacutainer System, Franklin Lakes, NJ, USA) and then centrifuged (3,000×g, 15 min, 4°C) within one hour after the collection of the sample to separate the serum. The blood urea nitrogen (BUN), creatinine, and glucose in the serum samples were analyzed with an automatic biochemical analyzer (HITACHI 747, Tokyo, Japan) using colorimetric methods.

After blood collection, total 56 broilers were weighed individually and slaughtered by cervical dislocation. The stomach, breast meat, bursa of Fabricius (Bursa cloacalis), liver, spleen, and abdominal fat were then removed by trained personnel and weighed. The breast muscles were stored at −20°C for the following analysis. Organ weight was expressed as a percentage of BW. The breast muscle Hunter L* (lightness), a* (redness), and b* (yellowness) values were determined using a Minolta CR410 chromameter (Konica Minolta Sensing Inc., Osaka, Japan). Cooking loss was determined using 5 g of breast meat, which was heat-treated in plastic bags separately in a water bath (100°C) for 5 min. Samples were cooled at room temperature. Cooking loss was calculated as (sample weight) before cooking – sample weight after cooking)/sample weight before cooking×100. A piece of breast meat was chilled at 2°C for 26 h, duplicate pH values for each sample were measured using a pH meter (Fisher Scientific, Pittsburgh, PA, USA).

All data were subjected to the statistical analysis as a randomized complete block design using the general linear model procedures of SAS [14], and the cage was used as the experimental unit. Differences among treatment means were determined using the Duncan’s multiple range test. Statements of statistical significance were based on p<0.05.

Only maize and soybean meal were included in the formula as the main raw materials to ensure that the non-GMO diets were applied in this experiment. The GMO qualitative analysis results are presented in Table 1 and 2. Four genes including 35S Promoter, NOS Terminator, DP-098140-6, DAS40278-9 were not detected non-GMO maize. 35S Promoter, NOS Terminator, MON89788, DP305423-1, DO356043-5, MON87701, CV127, MON87708, MON87769, DAS-68416-4 were not detected in non-GMO soybean meal. Those results confirmed that the maize and soybean meal which applied in non-GMO meal were non-GMOs.

The results of proximate analysis and amino acid analysis presented in Table 3 were reported as the percentage by weight on an as-is basis. In maize, the crude protein of non-GMO maize was 0.17% higher than that of GMO maize. Nevertheless, the higher content of Lys, Met, Thr was found in GMO maize. In soybean meal, the crude fiber of non-GMO soybean was 3.02% higher than that of GMO soymeal meal. A higher content of crude protein was observed in GMO soybean meal as well. Besides, the contents of Lys and Met in GMO soybean were 0.6% and 0.8% higher than that in non-GMO soybean, respectively. The development of GMO crops, especially the first generation, enhance insect or herbicide resistance, abiotic stress tolerance. The nutrient compositions of GMO grain are better than that of non-GMO materials. Probably due to fewer challenging factors (insect, herbicide or abiotic stress) affecting the accumulation of nutrients in the growth process. Besides, Rayan et al [15] reported that there were some statistical differences between the GMO corn samples and non-GMO control in some biochemical components. But he believed that those results were unlikely to be biologically significant, since they were well within the range of literature values.

The results of growth performance and nutrient digestibility are presented in Table 5 and 6. The FI and FCR were greater (p<0.05) in broilers provided with non-GMO diet feed than that in the GMO group from day 17 to 32. A decrease FCR was observed when birds were fed with the GMO diet through the whole experiment (p<0.05). In 1997, A program was started to assess GMO including Bt-maize, Pat-maize, Pat-sugar beets and Gt-soybeans, which tried to determine an effective system to assess GMO in feed stuffs from the view of nutrition. In 2001, the series of experiments reported by Flachowsky et al [16] had been published by Aulrich et al [17,18], Bohme et al [19], Daenicke et al [20,21], and Halle et al [22]. Results of all the experiments did not show any significant difference in growth performance and nutrient digestibility between GMO diet and non-GMO diet. The latest article on non-GMO Feed stuff in broilers was published in 2010 by Świątkiewicz et al [23], which revealed that no statistical difference was observed in any of the performance parameters across dietary treatments. Only two articles showed some improvements in growth performances when birds were fed the GMO-diet, one of the two papers published in 1998 showed the birds receiving GMO-corn diets exhibited improved adjusted feed conversion ratios at 28 and 38 days of age. The author considered that the improved feed conversion ratios cannot necessarily be attributed to the corn source, but these data did show an absence of any deleterious effects associated with the diets made from GMO when compared to diets made from non-GMO corn [24]. However, our study found that the GMO-diet exhibited a better growth performance of FCR than non-GMO diet. Chicks fed non-GMO diet have higher FCR than those fed GMO diet which was attributed to the higher food intakes. The first reason for these differences may be a higher apparent metabolic energy value assumed for diet formulation. The metabolic energy (ME) value of the experimental diets are calculated based on CVB Table Booklet Feeding of poultry recommendation [25]. It is not clear whether standard non-GMO maize or soybean meal supplied the similar ME as much as GMO maize for diet. During age of d 17 to d 32, broilers are in faster growth than that at age of 2 weeks. Goliomytis [26] documented the growth rate from 3 weeks to 6 weeks of BW, breast weight and leg weight were greater than that in first 2 weeks in broilers. Therefore, the protein and amino acids could be crucial one of all elements. The second reason may be different amino acids compositions or protein characteristics between GMO and non-GMO diet. The lower content of first 3 limiting essential amino acids occur in non-GMO feed stuff, the more crystalline amino acids have to be used in a non-GMO diet. But, the nutritional value of protein in animal feed may be related not only to amino acid content and composition, but also to the different resource of protein characteristics. Differences in physical and chemical properties of protein resource also affect the release dynamics of amino acids in the digestive tract of animals. The GMO varieties of plants which were called second-generation crops have traits to enhance nutritional properties [27]. Furthermore, the branched-chain amino acids which plays several critical roles in metabolism homeostasis and cell functions including immunity, survival and growth, energy homeostasis, and protein and lipid metabolism regulation [28] were not calculated in diet. Our amino acid analysis showed the quantity of branched-chain amino acids in non-GMO grain were around 1.15 times as much as that in non-GMO grain. Therefore, the relatively lower content of branched-chain amino acids in non-GMO diets might be another reason for the influence.

No significant impact on blood profile, meat quality, organ weight was found in response to the two treatments throughout the experimental period (Tables 7, 8) (p>0.05). Serum creatinine (a blood measurement) is an important indicator of renal health because it is an easily-measured by-product of muscle metabolism that is excreted unchanged by the kidneys. Plasma or BUN concentration may be useful as an indicator of protein status within a group of animals as well as nitrogen utilization and could help to fine-tune diets or identify problems with a feeding program [29]. Those results might demonstrate that the metabolism of protein and amino acids are not affected in the birds fed non-GMO and GMO diets. Consistent with our research, almost all previous published papers demonstrated that no statistic difference in carcass and organ weight was found between GMO and non-GMO feed. For example, the study of Tatlor et al [30] showed that no difference in carcass characteristics was present between birds were fed either GMO-maize or non-GMO maize.