Effect of Dietary Inclusion of Genetically Modified Maize (MON89034) on Marine Medaka Oryzias dancena
With recent increases in the international market price of the fish meal, which is an essential source of protein in aquaculture diets, and increasingly limited supplies, alternative sources of protein are being sought. In particular, the use of alternatives including soybean meal, fermented soybean meal, and corn gluten meal (CGM) has been studied [1,2]. Given the –2% grain self-sufficiency rate in Korea , the CGM derived from genetically modified (GM) maize is a major source of protein.
GM crops produced using genetic engineering carry foreign genes for various characteristics including herbicide tolerance, insect resistance, and drought resistance, and hybrid lines have recently been produced . The insect resistance lines include cry genes that code for the insecticidal δ-endotoxins (Cry-protein) of the soil bacterium Bacillus thuringiensis (Bt) . There are many types of Cry-proteins produced by Bt. They have strong host specificity, and it has been reported that they have no impact on non-target organisms, human health and the environment .
The Cry1A.105 gene, which is a compound of the Cry2Ab2 and Cry genes, was inserted into maize to prevent damage by lepidopteran insects. The MON89034 maize line has been found to be nutritionally equivalent to non-genetically modified (non-GM) maize . Studies on the influence of MON89034 × MON88017 maize crosses on non-target organisms includingAglaisurticae and Caenorhabditis elegans have been undertaken [8,9]. However, no studies have been reported on the effect on fish of a diet containing MON89034.
Horizontal gene transfer (HGT) occurs among prokaryotes, and van den Eede et al.  reported that HGT of intact antibioticresistant genes to the microflora of the intestinal tract of animals eating GM crops is possible. Furthermore, a number of studies have been conducted to assess the potential risks of GM diets for decreased biodiversity in aquatic environments and the marine ecosystem, impacts on non-target aquatic organisms, the transfer of recombinant proteins, and increases in tolerance [11-14]. In terms of fish culture, there have been reports on the nutritional equivalence and effects on growth of GM diets, and the feasibility of in-fish detection of recombinant genes [15-19].
The marine medaka Oryzias dancena (Beloniformes) is a euryhaline teleost that is easy to rear in culture, highly fecund, and has a short life span [20,21] It has already been approved of its possibility as a risk assessment model for aquatic environments and marine ecosystems .
In this study, GM maize MON89034 was provided in the diet of O. dancena to investigate its effects on the fish, including on growth, fertility, and histopathological changes. We also investigated whether the insect resistant gene Cry1A.105 in the diet could be detected in various fish organs following feeding.
Materials and Methods
Experimental diets and design
The approximate composition and component analysis for the ingredients used in the experimental diets are shown in Table 1. We used fish meal, casein, and maize as sources of protein, and squid oil as a source of lipid. The total content of maize (non-GM plus GM maize) in each of the experimental diets was kept constant at 21% (Table 1). The control diet (Con) was formulated to contain 21% non-GM maize, and the three experimental diets contained MON89034 at rates of 7% (GM7), 14% (GM14) and 21% (GM21). The nutritional composition of the experimental diets was formulated to be approximately 54% crude protein and 10% crude lipids. Each dried ingredient was ground to a powder, and the powdered ingredients were thoroughly mixed mechanically in distilled water (30%). Each diet was pelletized using a lab pellet extruder, dried for 48 h at room temperature, and stored at –25°C until used.
The GM maize used in this experiment (MON89034) was produced by Monsanto Co., USA, while the non- GM maize was cultivated and produced in Korea, and obtained from Eomgung Agriculture Market, Busan, Korea (http://www.eomgung-market.busan.kr/). Prior to using the maize for use in the experimental diets was analyzed using PCR and a primer set for the Cry1A.105 gene. The procedure is described in Notification No.2015-34 of the Ministry of Food and Drug Safety of Korea (Table 2).The Cry1A.105 gene was detected only in the MON89034 maize and not in the non-GM maize.
Con, control group; GM7, MON89034 maize content 7%; GM14,MON89034 maize content 14%; GM21,MON89034 maizecontent 21%.
1Vitamin premix contained the following, diluted in cellulose (g/kg premix): L-ascorbic acid, 121.2; DL-α-tocopheryl acetate, 18.8; thiamin hydrochloride, 2.7; riboflavin, 9.1; pyridoxine hydrochloride, 1.8; niacin,36.4; Ca-D-pantothenate, 12.7; myo-inositol, 181.8; D-biotin, 0.27; folic acid, 0.68; p-aminobenzoic acid, 18.2; menadione, 1.8; retinyl acetate, 0.73; cholecalciferol, 0.003; cyanocobalamin, 0.003.
2Mineral premix contained the following ingredients (g/kg premix):MgSO47H2O, 80.0; NaH2PO42H2O, 370.0; KCl, 130.0; Ferric citrate, 40.0; ZnSO47H2O, 20.0; Ca-lactate, 356.5; CuCl, 0.2; AlCl36H2O, 0.15; KI, 0.15; Na2Se2O3, 0.01; MnSO4H2O, 2.0; CoCl26H2O, 1.0.
Table 1. Composition (% dry matter) of the experimental diets containing genetically modified maize (MON89034), used for feeding O. dancena.
Table 2. Oligon1u cleo tide primers and thermal cycling conditions used in this study.
Experiment fish and breeding management
The O. dancena used in this study were from a laboratory stock at the Institute of Marine Living Modified Organisms,Pukyong National University, Busan, Korea. The juveniles were acclimated to laboratory conditions by feeding with brineshrimp (Artemia nauplii; INVE, Salt Lake City, Utah, USA) for 9 weeks following hatching. The mature fishes, which had aninitial mean body weight of 0.071±0.001 g, were randomly distributed into 8 square 10 L fiberglass tanks. Each tankcontained 12 male and 12 female fishes and had an independent water recirculation system that was continuously aerated. The water temperature and salinity were maintained at 25±1°C and 5 psu, respectively. Duplicate groups of fish were fed on one of the experimental diets (control, or one of the three GMcontaining diets) at a rate of five times per day for 30 weeks.
Measurement of fish weight
All surviving fish in each diet group at the 12th week and at the end of the 30-week experiment were collectively weighed after 12 h of starvation and anesthetized in ice-cold water.
Collection and observation of fertilized eggs
Fertilized eggs were collected once per week and immediately placed in a net (12 × 5 × 5 cm; mesh size: 330 × 330 μm) in 30 L fiberglass tanks until hatched (approximately 10 days). The average temperature and salinity were maintained at 25±1°C and 0 psu, respectively.
The eggs were examined using a dissecting microscope (AZ100, Nikon Co., Japan), and the rates of fertilization and abnormalities were recorded, based on embryonic development criteria detailed by Song et al. . The hatching rate was calculated from the numbers of hatched fishes and fertilized eggs.
To investigate the level of maturity of the experiment fishes, and for histopathological analysis, 2 male and 2 female fisheswere selected randomly from the control and each GM diet treatment at the end of the feeding experiment. The gonads,liver, spleen and kidneys were surgically removed, visually inspected, then fixed in 10% formaldehyde for >24 h. The organs were dehydrated through a graded series of ethanol, rinsed in xylene, embedded in paraffin, sectioned (5 μm thickness), and stained using hematoxylin-eosin (H & E). The presence of tissue lesions was assessed using a light microscope (Eclipse E400; Nikon Co., Japan) and photographed using a digital camera attached to the microscope (MoticampPro 205A; Motic Co., China).
At the completion of the feeding experiment, at 5-day intervals, 2 male and 2 female fishes from each diet were chosenrandomly following 12-h of starvation, and fixed in 70% ethanol. Ten types of the tissue sample (caudal fin, liver, brain,intestine, kidneys, muscle, eyes, heart, gonads, and gills) were carefully dissected from the fish, and pure genomic DNA was extracted from each tissue type for PCR analysis. A PCR premix (Accupower ProFiTaq PCR premix; Bioneer Co., Korea) was used for the analysis, and the PCR was conducted using three pairs of primers designed to amplify the corn starch synthase gene (SSIIb gene) and the Cry1A.105 gene (Table 2).
Growth and the rates of fertilization, hatching, and abnormality were assessed using ANOVA (analysis of variance), followed by Duncan’s multiple range test . The Student’s t-test was used to assess the amplification frequency for the SSIIb gene, and the Cry1A.105 gene from the GM maize. Differences were considered to be significant at the P <0.05 level.
Results and Discussion
At the end of the feeding trial no difference was observed in the survival of O. dancena fed experimental diets containing non- GM and GM maize, and little mortality of fish occurred during the 30-week trial (≥93% survival); from this it was concluded that inclusion of MON89034 maize had no lethal effect on the experimental fish. Raybould and Viachos  reported that survival of the water flea (Daphnia magna)and the catfish (Ictarus punctatus) was not affected when fed insect resistant recombinant maize MIR162 for 10 and 30 days, respectively
Con, control group; GM7, MON89034 maize content7%; GM14, MON89034 maize content14%; GM21, MON89034 maizecontent21%.
aValues (mean of duplicate groups±SE) in each row having a different superscript are not significantly different (P < 0.05).
Table 3. Growth (total weight in g) of O. dancena fed experimental diets containing genetically modified maize for 12 and 30 weeks.
The effect on O. dancena (initial mean body weight, 0.071 g) of the four experimental diets fed to the fish for 12 and 30 weeks is shown in Table 3. No significant differences (P<0.05) in mean weight were found among fish fed the various diets. This is consistent with previous studies on the growth performance of Atlantic salmon (Salmo salar), zebrafish (Danio rerio), common carp (Cyprinus carpio), olive flounder (Paralichthys olivaceus), and the rockfish (Sebastes schlegeli) [18,19,25-27] fed diets involving GM crops, indicating that the GM-based diets are substantially equivalent nutritionally to the non- GM maize. However, unlike the above studies, the inclusion of insect-resistant MON810 GM maize in diets decreased the growth of D. magna  and Atlantic salmon , whereas the growth of zebrafish increased . Additionally, the growth of red seabream (Pagrus auratus) increased on a diet including GM lupin . The differences in these results may be because of differences in the ability of the test animals to digest and absorb the vegetable proteins in the GM crops provided in the diet.
No studies on the fertility of fish fed diets containing recombinant GM crop materials have been reported. To investigate the influence of diets containing specific GM crops on the fertility of O. dancena, we evaluated the rates of fertilization, abnormality, and hatching after 30 weeks of feeding. No significant differences were found among the experimental groups (P<0.05) (Table 4), suggesting that MON89034 maize has no influence on the fertility of this species.
Con, control group; GM7, MON89034 maize content7%; GM14, MON89034 maize content14%; GM21, MON89034 maize content21%.
aValues (mean of duplicate groups±SE) in each row having a different superscript are not significantly different (P < 0.05).
Table 4. Fertilization, hatching, and abnormalities (%) in spawning eggs of O. dancena fed experimental diets containing genetically modified maize for 30 weeks.
Figure 1. Transverse sections of the gonad of O. dancena fed experimental diets containing genetically modified maize for 30 weeks. Con, control group; GM7, MON89034 maize content 7%; GM14, MON89034 maize content 14%; GM21, MON89034 maize content 21%. Scale bars indicate 25 μm.
Figure 2. Transverse sections of kidney, spleen and liver of O. dancena fed experimental diets containing genetically modified maize for 30 weeks. Con, control group; GM7, MON89034 maize content 7%; GM14, MON89034 maize content 14%; GM21, MON89034 maize content 21%. Scale bars indicate 25 μm.
Figure 3. Average PCR-based detection of the Cry1A.105 gene from MON89034 maize in O. dancena fed experimental diets containing GM and non-GM maize for 30 weeks. Con, control group; GM7, MON89034 maize content 7%; GM14, MON89034 maize content 14%; GM21, MON89034 maize content 21%.
aValues (mean of duplicate groups±SE) in each having a different superscript are significantly different (P < 0.05).
In contrast, Bøhn et al.  reported a decrease in the percentage of D. magna females reaching maturity and in total production of eggs, and also showed  that there were differences in the maturation period and fecundity. However, the applicability of direct comparisons between D. magna and fish is questionable.
There were no changes in the size and appearance of the gonads, liver, spleen or kidneys of fish fed the experimental diets (the gonads were very mature in all experiment groups), and no differences in histopathology were observed among theexperimental groups (Figure 1 and 2). These results indicate that inclusion of MON89034 in the diet caused no histologicalchanges to major organs of O. dancena. However, Hemre et al.  reported an increase in the mass of the liver and distalintestine of Atlantic salmon fed a diet containing MON810, but no histological differences were observed. Bakke-McKellep et al.  also found no histological differences in five organs, including the liver of Atlantic salmon fed a diet containing GM soybean meal (Roundup Ready; RR) and GM maize (MON810). Kitagima et al.  also found no expression of the fluorescent protein in the gut and internal organs of rainbow trout fed on a diet that included plasmid DNA containing the green fluorescence protein (GFP).
Con, control group; GM7, MON89034 maize content 7%; GM14, MON89034 maize content 14%; GM21, MON89034 maize content 21%.
abcValues (mean±SE of duplicate groups) in each row having a different superscript are significantly different (P < 0.05).
Table 5. Average PCR-based detection of the endogenous SSIIb gene from maize in O. dancena fed experimental diets containing GM and non-GM maize for 30 weeks.
There are concerns related to the possibility of HGT of transgenic DNA to animals, as transgenic DNA fragments from GM diets have been found in the tissues of chickens (Gallus fallus), pigs (Sus scrofa), mice (Mus musculus), Atlantic salmon,and rainbow trout [16,34-38]. In particular, transgenic DNA fragments were detected in the internal organs of starved Nile tilapia (Oreochromis niloticus) and Atlantic salmon fed GM diets [17,39]. Figure 3 shows changes in the amplification of the Cry1A.105 gene following fish starvation. The gene was amplified in all experimental groups, and there were no significant differences in the amplification frequency among the groups (P<0.05). Consistent with previous studies, the amplification frequency decreased in all experimental groups as the period of starvation increased, with the exception of GM14. In most of the experimental group’s amplification of Cry1A.105 was found on the 20th day of starvation. It is assumed that this result is a consequence of the longer period of feeding in the present study compared with the preceding studies.
Hohlweg and Doerfler  reported the detection of the endogenous plant gene ribulose-1, 5-bisphosphatecarboxylase/oxygenase (Rbc) in mice. In fish, Sissener et al.  reported amplification of an endogenous non-GM soybean gene from Atlantic salmon, while Chainark et al.  reported amplification of a chloroplast gene in rainbow trout. In this study, analysis of fish tissue types for amplification of the SSIIb maize gene using PCR (Table 5) showed that the gene was present in the control and all treatment groups.
The frequency was 12.5%–60.0% higher than that of the Cry1A.105 gene (0.0%–11.5%). In contrast to the Cry1A.105gene, in all experimental groups the amplification frequency did not decrease with starvation. This may be related to the SSIIb gene having a significantly higher copy number compared with Cry1A.105 from MON89034. Aumaitre et al.  also reported that in the guts of ruminants the amplification frequency of transgenic DNA was lower than that of a plant chloroplast gene.
Tao et al.  confirmed the amplification of transgenic DNA fragments in all internal tissues of Nile tilapia, and in the present study, the Cry1A.105 and SSIIb genes were amplified from the various tissues of O. dancena. Furthermore, the amplification frequency for the shorter (114 bp) DNA fragment was higher than that for the 151 bp DNA fragment (Table 5). Sanden et al.  reported amplification of transgenic DNA from the gut, liver, brain, and muscles of Atlantic salmon, with smaller DNA fragments being amplified at a higher frequency. They also reported that the DNA fragments in the gastrointestinal tract (GIT) may be protected from decomposition by a DNA-protein complex introduced into cells through mast cells. Therefore, it may be that the DNA-protein complex present in the GIT is transferred into cells, explaining the successful amplification from various internal organs.
We conclude that the influence of GM maize in the diet of O. dancena was not significant in terms of growth, eggdevelopment, and tissue histopathology. Moreover, the amplification of fragments of transgenic DNA was not significantly different between the diet containing GM maize or non-GM maize. The results indicate the substantial equivalency of MON89034maize with non-GM maize for feeding in O. dancena.
This work was supported by a Research Grant of Pukyong National University (2016 year).
4. Lee CS. On the sustainability of grouper farming. In Book of the abstracts, International symposium on grouper. Edited by Wu WC, Tseng CT, Tseng HR, Chen SC, Liu DC, Su MS and Su WS. Keelung: Fisheries Research Institute; 2010.
5. Hu SH. Toward sustain culture of groupers. In Book of the abstracts, International symposium on grouper. Edited by Wu WC, Tseng CT, Tseng HR, Chen SC, Liu DC, Su MS and Su WS. Keelung: Fisheries Research Institute; 2010.
6. Harikrishnan R, Balasundaram C, Heo MS. Molecular studies, disease status and prophylactic measures in grouper aquaculture: Economic importance, diseases and immunology. In Aquaculture . 2010, 309(1): 1-14.
10. Asche F, Khatun F. Aquaculture: issues and opportunities for sustainable production and trade. Geneva: International Centre for Trade and Sustainable Development; 2006.
20. Bombeo-Tuburan I, Coniza EB, Rodriguez EM, Agbayani RF. Culture and economics of wild grouper (Epinephelus coioides) using three feed types in ponds. In Aquaculture 2001, 201(3): 229-240.
21. Pomeroy RS, Parks JE, Balboa CM. Farming the reef: is aquaculture a solution for reducing fishing pressure on coral reefs? Mar. Policy 2006, 30(2): 111-130.
22. Afero F, Miao S, Perez AA. Economic analysis of tiger grouper Epinephelus fuscoguttatus and humpback grouper Cromileptes altivelis commercial cage culture in Indonesia. In Aquacult Int. 2010, 18(5): 725-739.