A 90 Day Safety Assessment of Genetically Modified Rice ExpressingCry1Ab/1Ac Protein using an Aquatic Animal ModelHao-Jun Zhu,†,‡ Yi Chen,‡ Yun-He Li,‡ Jia-Mei Wang,†,‡ Jia-Tong Ding,† Xiu-Ping Chen,*,‡

and Yu-Fa Peng*,‡

†College of Animal Science and Technology, Yangzhou University, Yangzhou 225009, People’s Republic of China‡State Key Laboratory for Biology of Plant Diseases and Insect Pests, Institute of Plant Protection, Chinese Academy of AgriculturalSciences, No. 2 West Yuanmingyuan Road, Haidian District, Beijing 100193, People’s Republic of China

ABSTRACT: In fields of transgenic Bt rice, frogs are exposed to Bt proteins through consumption of both target and nontargetinsects. In the present study, we assessed the risk posed by transgenic rice expressing a Cry1Ab/1Ac fusion protein (Huahui 1,HH1) on the development of Xenopus laevis. For 90 days, froglets were fed a diet with 30% HH1 rice, 30% parental rice(Minghui 63, MH63), or no rice as a control. Body weight and length were measured every 15 days. After sacrificing the froglets,we performed a range of biological, clinical, and pathological assessments. No significant differences were found in body weight(on day 90: 27.7 ± 2.17, 27.4 ± 2.40, and 27.9 ± 1.67 g for HH1, MH63, and control, respectively), body length (on day 90: 60.2± 1.55, 59.3 ± 2.33, and 59.7 ± 1.64 mm for HH1, MH63, and control, respectively), animal behavior, organ weight, liver andkidney function, or the microstructure of some tissues between the froglets fed on the HH1-containing diet and those fed on theMH63-containing or control diets. This indicates that frog development was not adversely affected by dietary intake of Cry1Ab/1Ac protein.

KEYWORDS: transgenic rice, Bt protein, frog, safety assessment, nontarget effect

■ INTRODUCTION

Rice is an important crop that feeds half of the world’spopulation. However, it is subject to severe pest damage, with∼2−10% of Asia’s annual rice yield being lost to insect pestssuch as Lepidoptera.1 The rapid development of transgenictechnologies provides new strategies for pest control in rice.Bacillus thuringiensis (Bt) is a Gram-positive soil bacteria that,during sporulation, produces the crystal insecticidal protein δ-endotoxin, which has a strict specificity to target insects.2 Tocontrol insect damage to rice, many transgenic lines expressingBt genes have been successfully developed worldwide.3

In 2000, the Cry1Ab/1Ac fusion gene from Bt wastransformed into the elite rice indica restorer line Minghui 63(MH63), generating the Bt rice line Huahui 1 (HH1), whichhas a high resistance to stem borers.4 In August 2009, China’sMinistry of Agriculture issued a safety certificate for this Bt riceline, but it has not yet been approved for commercialcultivation because detractors argue that Bt rice may posepotential food safety and environmental issues.5 To assess foodsafety, a number of animal feeding studies using Bt rice havebeen conducted, primarily in rats,6−9 broilers,10 pigs,11 andcarp.12 All of the above studies showed that the safety of Bt ricelines was comparable to that of their nontransgenic counter-parts. To assess environmental safety, many field and laboratoryexperiments have also been performed, mainly focusing onterrestrial organisms. Bt rice, protected from the damage byLepidoptera, had no obvious negative impacts on the individualfitness, population abundance, or diversity of nontargetorganisms.13−17 However, the entry of Bt proteins into streamecosystems was initially so insignificant that it was notconsidered in the assessment of risks associated with thecultivation of Bt crops.18 Moreover, because rice, unlike dry-

land crops, requires water during most stages of development,the risk for aquatic organisms in and around Bt rice fieldscannot remain ignored. To date, only a few studies haveassessed the effects of Bt rice on aquatic organisms such asDaphnia magna19 and Chlorella pyrenoidosa.20

Frogs are commonly found in rice fields and play animportant role in maintaining the biodiversity and stability ofthe paddy field ecosystem. In recent decades, frog populationshave declined sharply worldwide.21 Frogs might be affected byBt rice in two ways. First, frogs could ingest Bt proteins directlyby consuming insects that have fed on Bt rice.22 Second,because of the relatively high permeability of frog skin, frogscould take up Bt proteins that are released into the waterthrough exudation from roots, pollen dispersal, and disposal ofpostharvest detritus.18 Although some studies demonstratedthat Bt rice releases detectable amounts of Bt protein intoirrigation water,23,24 it remains unclear whether and how thisaffects frog development. Therefore, it is important to assessthe potential nontarget effects of Bt rice on the development offrog species.Xenopus laevis is a model animal widely used in environ-

mental toxicology because it is easy to feed, readily induced tolay eggs, and very sensitive to external contamination.25

Additionally, unlike wild frogs, which usually prefer movingprey, it will consume static food. Thus, in the present study, X.laevis froglets were fed a nutritionally balanced diet containingHH1 or its nontransformed parental rice line to assess the food

Received: November 19, 2014Revised: March 30, 2015Accepted: March 30, 2015

Article

pubs.acs.org/JAFC

© XXXX American Chemical Society A DOI: 10.1021/jf5055547J. Agric. Food Chem. XXXX, XXX, XXX−XXX

and environmental safety of Bt rice. Results from this study willprovide important information concerning the environmentalsafety of genetically modified (GM) strains of rice.

■ MATERIALS AND METHODSChemicals. Human chorionic gonadotropin (hCG) and meth-

anesulfonate (MS-222) were purchased from Sigma ChemicalCompany (St. Louis, MO, USA). Commercial frog feed was obtainedfrom Cargill Feed Co., LTD (Nanjing, People’s Republic of China).Liver and kidney function detection kits and Bt-Cry1Ab/1Ac proteindetection kits were purchased from JianCheng BioengineeringInstitute (Nanjing, People’s Republic of China) and EnviroLogixInc., (Portland, OR, USA), respectively.Test Materials. The HH1 transgenic rice line expressing the fusion

gene Cry1Ab/1Ac exhibits resistance to stem borers, such as Chilosuppressalis, Scirpophaga incertulas, and Cnaphalocrocis medinalis.4 Theinsecticidal activity of HH1 seeds were previously confirmed at theseedling stage.26 The parental nontransformed control line (MH63) isan elite Indica restorer line for cytoplasmic male sterility commonlygrown in China. These lines were simultaneously planted in twoadjacent plots at a scientific research base at Jiangxi Academy ofAgricultural Sciences (Nanchang, Jiangxi Province, People’s Republicof China). Transgenic rice had not been previously planted in theseplots. Crops were managed in accordance with regulations of theAgricultural Genetically Modified Organism (GMO) Safety Manage-ment of China. Rice was harvested at the end of October 2013, andgrains were collected and stored at −20 °C until use. The conventionalnutrient components of each rice line were analyzed by the BeijingResearch Institute for Nutritional Resources, People’s Republic ofChina.Animals. Mature female and male X. laevis were maintained

separately in glass tanks containing dechlorinated water at 21 ± 2 °Con a 12 h light/12 h dark cycle and were fed chopped pork liver onceper week. One pair of parent frogs was chosen and injected with 100IU hCG to induce breeding. After eggs were laid, the mating pair wasremoved from the breeding tank. Fertilized eggs were incubated at 22± 2 °C on a 12 h light/12 h dark cycle. Tadpoles were started on adaily diet of green algae and D. magna on day 5 after fertilization andswitched to commercial frog feed when they completed metamor-phosis. Approximately 2 months after metamorphosis, froglets withuniform body weight (∼4.30 g) and body length (∼32.0 mm) wereused for experimentation. Frogs were treated in accordance with theNIH Guide for the Care and Use of Laboratory Animals.Diet Formulation. Synthetic frog diets were produced based on

the nutrient composition of a commercial frog feed. Two test dietscontained 30% rice flour, whereas the control diet was formulatedusing cornstarch and soybean meal without rice. The detailed dietcompositions are shown in Table 1.Experimental Design. A total of 96 froglets were equally divided

into three experimental groups of 32 according to body weight, andtheir sexes were not considered. Each group was randomly divided intofour glass jars (20 × 34 × 24 cm) to give four replicates. Froglets werefed daily (∼3% weight/body weight). Fresh dechlorinated water wasreplenished every 2 days and monitored daily to stay at 20−22 °C. A12 h light/12 h dark cycle was maintained throughout the course ofthe experiment. Animals were observed twice daily, and body weightand length (from the tip of the snout to the tip of cloaca) weremeasured every 15 days for 90 days. Animals were fasted overnightbefore being sacrificed.Gross Necropsy and Histopathology. At the end of the

experiment, the frogs were anesthetized by immersion in 1% MS-222.Before dissection, body weight and length were measured, and thegender was determined on the basis of gross gonadal morphology. Acomplete necropsy was then performed, and the following organs wereexcised and weighed: heart, liver, spleen, lung, kidneys, body fat,ovaries or testes, and intestines. Paired organs (lungs, kidneys, ovariesand testes) were weighed as a total of left and right. Additionally, thetissues (stomach, intestines (divided into duodenum, ileum, andrectum), liver, spleen, and gonad) of one male and one female from

each glass jar were fixed for a minimum of 24 h in 4% phosphate-buffered formaldehyde before histological processing. Tissue sampleswere embedded in paraffin, and ∼4 to 6 μm thick sections were cutand stained with hematoxylin and eosin for light microscopy.

Determination of Liver and Kidney Function. The liver andspleen from the remaining six froglets in each glass jar (a total of 24samples per treatment) were used to determine the followingparameters: alkaline phosphatase (AKP) activity, albumin (ALB),alanine aminotransferase (ALT) activity, aspartate aminotransferase(AST) activity, urea (BUN), cholinesterase (CHE) activity, totalprotein (TP), creatinine (CR), glutamic acid (GLU), total cholesterol(TC), and triglycerides (TG). All parameters were detected using kitsprovided by the JianCheng Bioengineering Institute. All of thedetection kits used in this study use a corresponding standardsubstance to show assay validity prior to analysis. Moreover, these kitsare used strictly according to the recommendations and protocols setby the manufacture.

Determination of Cry1Ab/1Ac Content. The Cry1Ab/1Acprotein content in rice grains, frog diets, frog tissues (stomach andintestine), and frog feces (∼0.20 g each) were determined using a Bt-Cry1Ab/1Ac protein kit with a detection limit of 0.1 ng/g totalprotein. Solid feces were collected immediately after the renewal ofrearing water at the later stage of the present study. Before analysis,animal tissues were washed in phosphate-buffered saline/Tween-20 toremove Bt toxins from the outer surface, lyophilized, and thenhomogenized in 1 mL phosphate-buffered saline/Tween-20 using amicropestle and mortar on ice. After centrifugation and appropriatedilution of the supernatants, an ELISA was performed following themanufacturer’s protocol. Optical density values were read using amicroplate spectrophotometer (PowerWave XS2, BioTek, Winooski,VT, USA). A standard curve derived from purified Cry1Ab/1Acsamples was used to calculate Cry1Ab/1Ac levels.

Statistical Analyses. All data are presented as mean ± standarddeviation (SD) unless indicated otherwise. Statistically significantdifferences in conventional nutrient compositions between HH1 andMH63 rice grains were analyzed by Student’s t test. Body weights andlengths were analyzed using repeated measures analysis of variance(ANOVA), whereas a one-way ANOVA, followed by least significantdifference (LSD) multiple comparison tests, was used to analyzedifferences in organ weight, as well as liver and kidney function amongthe three treatment groups. Differences were considered significant atp < 0.05.

Table 1. Composition of Nutritionally Balanced Test andControl Diets for X. laevis Froglets

amt of ingredient (%)

ingredient HH1 MH63 control

transgenic rice 30.0 0.00 0.00nontransgenic rice 0.00 30.0 0.00pregelatinized starch 8.00 8.00 8.00corn 0.00 0.00 17.0soybean meal 6.00 6.00 21.0soybean oil 3.00 3.00 3.50fish meal 52.0 52.0 38.0wheat bran 0.00 0.00 5.00meat and bone meal 0.00 0.00 6.00rock flour 0.00 0.00 0.50additivea 1.00 1.00 1.00total 100 100 100

aContents in mg/kg of diet: iron, 70; copper, 11; manganese, 70; zinc,65; iodine, 0.49; selenium, 0.3; vitamin A, 8000 (IU); vitamin D, 2400(IU); vitamin E, 20 (IU); vitamin K, 0.5 (IU); vitamin B1, 2; vitaminB2, 8; vitamin B6, 3.5; vitamin B12, 0.01; calcium pantothenate, 20;niacin, 35; folic acid, 0.75; biotin, 0.26.

Journal of Agricultural and Food Chemistry Article

DOI: 10.1021/jf5055547J. Agric. Food Chem. XXXX, XXX, XXX−XXX

B

■ RESULTSNutrient Composition of Rice Grains and Froglet

Diets. Levels of conventional nutrients (crude protein, crudefat, crude fiber, crude ash, calcium, and phosphorus) were verysimilar in HH1 and MH63 rice grains (Table 2), and nostatistical differences were observed between the two (p >0.05). Moreover, the measured values of these same nutrientsin the HH1 diet were comparable to those in the MH63 andcontrol diets (Table 2).General Health. Animal wellbeing was observed twice

daily, and body weight and length were measured every 15 daysfor 90 days. No adverse effects on animal behavior wereobserved during the experiment. The mean body weights andbody lengths of animals from each treatment are shown inTables 3 and 4, respectively. Statistically significant differences

were observed among the seven repeated measures within eachtreatment group from day 0 to day 90 (p < 0.01 for bodyweight and body length), but not among the three treatments

(p = 0.81 for body weight, p = 0.98 for body length).Additionally, the only significant difference in organ weightamong the three dietary groups was the relative weight of fatbody, which was significantly higher in the HH1 group incomparison to control (p < 0.05, Table 5).

Liver Function, Kidney Function, and Fat Metabolism.Parameters for liver and kidney function, as well as fatmetabolism, are given in Table 6. There were no statisticallysignificant differences among the three groups (p > 0.05).

Cry1Ab/1Ac Protein Content in Diets and Frogs’Digestive Tracts. Cry1Ab/1Ac was present in HH1 rice grainat an average concentration of 752 ± 78.0 ng/g and was notdetectable (ND) in the nontransformed control strain MH63(Table 7). The measured Cry1Ab/1Ac contents in the HH1,MH63, and control diets (221 ± 21.2 ng/g, ND, and ND,respectively; Table 7) were close to the predicted values (226,0, and 0 ng/g, respectively). To investigate the degradation ofCry1Ab/1Ac protein in the frog digestive tract, intestinalcontents and feces were collected, and the Cry1Ab/1Ac proteincontent was determined by ELISA. Cry1Ab/1Ac protein wasdetected in the intestinal contents (249 ± 16.9 ng/g) and infeces (160 ± 34.7 ng/g) of froglets fed on the HH1 diet but notin those fed on the MH63 or control diet (Table 7).

Gross Necropsy and Histopathology. Histologicalexaminations of stomach, intestine (ileum), liver, kidneys,spleen, testes, and ovaries are shown in Figure 1. None of thethree sections of intestines showed apparent pathologicalabnormalities; only representative images of ileum are shownin Figure 1. Overall, there were no gross pathological findingsduring the necropsies, and no group-related histopathologicalabnormalities were observed.

■ DISCUSSION

An integral part of the safety evaluation of GM plants is testingfor substantial equivalence to the unmodified parent strain,which provides a starting point for the overall assessment.27 Inthe present study, we analyzed the differences in nutrientcomposition between HH1 and MH63 before evaluating theeffects of the GM rice strain on development of froglets. Nosignificant differences in composition, including crude protein,crude fat, and crude ash levels, were found between the two ricelines or between diets containing them (Table 2). Schrøder etal. found significant differences in some nutritional compositionindices between brown rice material from transgenic Cry1Aband its corresponding parental line, but all differences werewithin the normal reference intervals.7 Indeed, most studieshave shown that transgenic Bt rice lines have nutrientcompositions similar to those of their parental lines,6,8,28

which is consistent with our results.

Table 2. Nutrient Composition of Rice Grains and Diets for X. laevis Froglets

rice grain (n = 3) diet (n = 1)

componentamt of componenta in

HH1bamt of componenta in

MH63bp

valuecamt of componenta in

HH1amt of componenta in

MH63amt of componenta in

control

crude protein(%)

6.63 ± 0.24 6.85 ± 0.12 0.22 38.1 38.2 42.2

crude fat (%) 2.83 ± 0.58 2.90 ± 0.00 0.18 8.40 8.80 8.10crude fiber (%) 6.97 ± 0.06 6.65 ± 0.50 0.17 3.30 3.20 3.00crude ash (%) 11.8 ± 0.26 11.9 ± 0.06 0.89 11.2 11.4 10.7calcium (g/kg) 0.66 ± 0.16 0.61 ± 0.13 0.71 3.50 3.50 2.40phosphorus (%) 0.36 ± 0.01 0.37 ± 0.01 0.25 1.81 1.84 1.73aUnits for the amount of component are given in column 1. bData are presented as mean ± SD. cStudent’s t test, p < 0.05.

Table 3. Mean Body Weight over Time in X. laevis FrogletsFed Test versus Control Diets (n = 32)

body weight (g)a

time (days) HH1 MH63 control

0 4.35 ± 0.01 4.36 ± 0.06 4.32 ± 0.0415 6.02 ± 0.23 6.13 ± 0.30 5.85 ± 0.2130 8.85 ± 0.25 9.33 ± 0.57 8.97 ± 0.2745 12.0 ± 0.40 12.3 ± 1.01 12.2 ± 0.3860 16.7 ± 0.39 17.2 ± 1.46 16.8 ± 0.5675 21.2 ± 0.34 22.2 ± 2.11 22.5 ± 1.1290 27.7 ± 2.17 27.4 ± 2.40 27.9 ± 1.67

aData are presented as group mean values ± SD. Repeated measuresANOVA, p < 0.05.

Table 4. Mean Body Length over Time in X. laevis FrogletsFed Test versus Control Diets (n = 32)

body length (mm)a

time (days) HH1 MH63 control

0 31.6 ± 0.44 31.7 ± 0.42 31.8 ± 0.1915 34.8 ± 0.60 34.7 ± 0.48 33.9 ± 0.3930 39.2 ± 0.27 39.5 ± 0.68 39.5 ± 0.5145 43.2 ± 0.39 43.3 ± 0.98 43.7 ± 0.5460 47.8 ± 0.42 48.3 ± 0.50 48.5 ± 0.5375 52.5 ± 0.78 52.0 ± 1.74 52.2 ± 1.3890 60.2 ± 1.55 59.3 ± 2.33 59.7 ± 1.64

aData are presented as group mean values ± SD. Repeated measuresANOVA, p < 0.05.

Journal of Agricultural and Food Chemistry Article

DOI: 10.1021/jf5055547J. Agric. Food Chem. XXXX, XXX, XXX−XXX

C

The consequences and potential risks associated withexpressed insecticidal proteins are of particular concern inany safety assessment.29 Numerous experimental studies haveconsistently indicated that the health and performance ofanimals fed GM crop lines and isogenic non-GM crop lines are

comparable.30 The food and feed safety of Bt rice has beentested in many animals, including rats,6−9 broilers,10 pigs,11 andcarp,12 and all of the above studies showed that the safety of theGM line and that of the parental line were comparable. Forexample, TT51 rice, the same Bt rice line as in the presentstudy, showed no significant differences on reproductionperformance of rats in comparison with MH63 and control.9

Moreover, Xu et al. found that Cry1Ab/Ac fusion proteinproduced in Escherichia coli has no adverse effects in mice bygavage at a high dose level of 5 g/kg body weight.31 Our resultsusing X. laevis froglets were consistent, as there were nosignificant differences among the three dietary groups withrespect to general health, including animal behavior, bodyweight (Table 3), and body length (Table 4). In addition, weassessed organ weight and relative organ weight, which areimportant indicators reflecting organ development. A change inorgan weight can signal adverse effects from the externalenvironment, including food.32 Our findings show that the dietcontaining HH1 rice resulted in no significant differences inabsolute or relative organ weights in comparison with theMH63 and control diets, except for an increase in the relativeweight of fat bodies in the HH1 group in comparison to thecontrol group (Table 5). Furthermore, pathology examsrevealed no obvious lesions in any group (Figure 1), and

Table 5. Absolute and Relative Organ Weights and Intestinal Lengths in X. laevis Froglets Fed Test versus Control Dietsa

HH1 MH63 control statisticsc

Absolute Measurementsbody (g) 27.7 ± 2.17 27.4 ± 2.40 27.9 ± 1.67 t = 0.07; p = 0.93heart (g) 0.14 ± 0.01 0.14 ± 0.02 0.15 ± 0.02 t = 0.41; p = 0.66liver (g) 1.70 ± 0.16 1.60 ± 0.16 1.64 ± 0.07 t = 0.17; p = 0.84spleen (g) 0.03 ± 0.00 0.03 ± 0.00 0.03 ± 0.00 t = 0.02; p = 0.98lung (g) 0.16 ± 0.01 0.16 ± 0.01 0.16 ± 0.02 t = 0.07; p = 0.94kidney (g) 0.21 ± 0.03 0.21 ± 0.03 0.22 ± 0.01 t = 0.16; p = 0.85fat body (g) 1.84 ± 0.14 1.82 ± 0.21 1.73 ± 0.06 t = 0.20; p = 0.82ovary (g) 0.12 ± 0.02 (16) 0.11 ± 0.01 (13) 0.12 ± 0.02 (19) t = 0.11; p = 0.90testis (g) 0.05 ± 0.01 (16) 0.04 ± 0.01 (19) 0.05 ± 0.01 (13) t = 0.95; p = 0.39intestinal length (cm) 15.1 ± 1.11 15.0 ± 1.30 16.2 ± 1.76 t = 1.94; p = 0.15

Relative Valuesb

heart 0.51 ± 0.05 0.49 ± 0.04 0.52 ± 0.06 t = 0.60; p = 0.55liver 6.06 ± 0.14 5.75 ± 0.20 5.70 ± 0.15 t = 2.00; p = 0.14spleen 0.10 ± 0.01 0.11 ± 0.01 0.10 ± 0.01 t = 0.75; p = 0.48lung 0.56 ± 0.04 0.58 ± 0.05 0.56 ± 0.04 t = 0.27; p = 0.76kidney 0.74 ± 0.05 0.76 ± 0.07 0.77 ± 0.04 t = 0.63; p = 0.54fat body 6.56 ± 0.22b 6.49 ± 0.39ab 6.00 ± 0.26a t = 3.99; p = 0.02ovary 0.44 ± 0.06 (16) 0.42 ± 0.06 (13) 0.48 ± 0.09 (19) t = 0.73; p = 0.49testis 0.18 ± 0.03 (16) 0.16 ± 0.02 (19) 0.17 ± 0.01 (13) t = 0.60; p = 0.55intestinal length 0.56 ± 0.02 0.58 ± 0.08 0.63 ± 0.02 t = 2.47; p = 0.09

aData presented as group mean values ± SD (n = 32, except for ovaries and testes, where n is given in parentheses). Different Roman lowercaseletters in the same row indicate a statistical difference of p < 0.05. bRelative values expressed as g or cm per 100 g of body weight. cOne-wayANOVA, p < 0.05.

Table 6. Liver Function, Kidney Function, and FatMetabolism in X. laevis Froglets Fed Test versus ControlDiets (n = 24)

indices HH1 MH63 control statisticsa

Liver FunctionAKP (U/gprotein)

195 ± 12.6 193 ± 31.1 167 ± 12.1 t = 0.81;p = 0.45

ALB (g/L) 7.31 ± 0.55 7.22 ± 0.58 7.04 ± 0.72 t = 0.73;p = 0.49

ALT (U/gprotein)

860 ± 181 902 ± 224 813 ± 36.7 t = 0.55;p = 0.58

AST (U/gprotein)

443 ± 66.5 432 ± 65.1 405 ± 22.0 t = 0.75;p = 0.48

BUN(mg/L)

1.82 ± 0.08 1.79 ± 0.08 1.72 ± 0.08 t = 2.71;p = 0.08

CHE (U/mgprotein)

1.41 ± 0.15 1.40 ± 0.10 1.34 ± 0.12 t = 0.06;p = 0.94

TP (g/L) 9.27 ± 0.48 8.90 ± 0.21 9.10 ± 0.25 t = 1.28;p = 0.28

Kidney FunctionBUN(mg/L)

3.49 ± 0.11 3.32 ± 0.18 3.36 ± 0.48 t = 0.47;p = 0.63

CR(μmol/L)

23.9 ± 4.07 25.1 ± 3.30 20.3 ± 5.56 t = 2.82;p = 0.07

GLU (μmol/g protein)

30.6 ± 4.34 28.25 ± 3.74 29.75 ± 0.82 t = 0.74;p = 0.48

Fat MetabolismTC(mmol/L)

0.66 ± 0.05 0.69 ± 0.10 0.67 ± 0.16 t = 0.19;p = 0.83

TG(mmol/L)

2.32 ± 0.34 2.11 ± 0.10 2.06 ± 0.68 t = 0.79;p = 0.46

aOne-way ANOVA, p < 0.05.

Table 7. Cry1Ab/1Ac Protein Content in Rice Grain, Diet,and Digestive Tract of X. laevis Froglets Fed Test versusControl Diets (ng/g)a

HH1 MH63 control

rice grain (n = 6) 752 ± 78.0 ND NAdiet (n = 6) 221 ± 21.2 ND NDintestinal tract (n = 24) 249 ± 16.9 ND NDfeces (n = 6) 160 ± 34.7 ND ND

aND, not detectable; NA, not applicable.

Journal of Agricultural and Food Chemistry Article

DOI: 10.1021/jf5055547J. Agric. Food Chem. XXXX, XXX, XXX−XXX

D

liver function, kidney function, and fat metabolism (Table 6)were not significantly different among the groups. Takentogether, these results indicate that Cry1Ab/Ac protein had nosignificant adverse effect on organ development.The planting of GM crops brings huge economic and

environmental benefits to humans; however, it also raisesquestions and concerns regarding their safety. Some people fearthat foreign proteins will enter the food chain and eventuallyenter the human body, where they may cause potential harm tohumans. They also fear that foreign proteins may be present inthe feces of animals fed GM crops, where they may affect theenvironment. Whereas some researchers found that transgenic

DNAs and proteins could not be detected in the gastro-intestinal contents or feces of animals fed diets containing GMcrops,11,33,34 others were able to detect foreign transgenicproteins in the gastrointestinal contents and/or feces of pigs,35

bovines,36,37 deer,38 and insects39 that were fed transgenic cropsor diets containing transgenic crops. The amount of transgenicprotein ingested by livestock depends on the concentration ofthe protein in the feed, the amount of feed intake, and thespecies.29 From previous studies we can conclude that Btproteins can survive passage through an animal’s gastro-intestinal tract but are not transferred to the visceral organsor their products.33,36 In the current study, Cry1Ab/1Ac

Figure 1. Histopathological staining of tissues from X. laevis froglets after consuming the HH1 test diet (a), MH63 test diet (b), or control diet (c)for 90 days. For intestine, only representative images of the ileum are shown.

Journal of Agricultural and Food Chemistry Article

DOI: 10.1021/jf5055547J. Agric. Food Chem. XXXX, XXX, XXX−XXX

E

protein was detected in the intestinal contents and feces offroglets in the HH1 group (Table 7), indicating that they wereexposed to Bt proteins by consuming a diet containing theCry1Ab/1Ac protein. However, the microstructure of thedigestive tract (Figure 1) and the froglets’ overall developmentwere not adversely affected. Our results, combined with thoseof others, demonstrate that expressed foreign Bt proteins donot produce harmful effects in the digestive tract or elsewhereafter ingestion by animals.29

Consumption of Bt rice fed arthropods is the main routethrough which frogs are exposed to the rice-produced Cryproteins. Zhang et al. found that the Cry2Aa protein content inmost of the arthropods collected from the Cry2Aa rice fieldwere much lower than 200 ng/g dry weight.22 Studies of Btsoybean and Bt maize showed that the plant-produced Cryproteins are diluted when moving through the foodweb.39,40

Additionally, Wang et al. did not detect Cry1Ab/1Ac protein inthe stomach and intestine of frogs collected from HH1 ricefield.26 In the present study, the Cry1Ab/1Ac proteinconcentration in the HH1 diet was 221 ± 21.2 ng/g (Table7), which was much higher than the likely exposure under fieldconditions.In this 90 day study, X. laevis froglets were exposed to Bt

proteins by consuming a diet containing HH1 rice that carriesthe gene encoding Cry1Ab/1Ac. Froglet development was notadversely affected in comparison with froglets fed a dietcontaining the parental rice (MH63) or no rice. Combining thislaboratory study with our previous field experiment,26 weconclude that the planting of transgenic Cry1Ab/1Ac rice willnot adversely affect the development of frogs.

■ AUTHOR INFORMATIONCorresponding Authors*X.-P.C.: tel, +86-10-62815947; fax, +86-10-62896114; e-mail,cxpyuan@gmail.com.*Y.-F.P.: tel, +86-10-62815947; fax, +86-10-62896114; e-mail,yfpeng@vip.qq.com.

FundingThis work was supported by the National GMO New VarietyBreeding Program of the PRC (2012ZX08011-002 and2014ZX08011-001) and a Priority Academic Program Develop-ment of Jiangsu Higher Education Institutions (PAPD).

NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSWe thank Professor Yongjun Lin (Huazhong AgriculturalUniversity, Wuhan, People’s Republic of China) for kindlyproviding transgenic rice seeds.

■ ABBREVIATIONS USEDAKP, alkaline phosphatase activity; ALB, albumin; ALT, alanineaminotransferase activity; AST, aspartate aminotransferaseactivity; BUN, urea; CHE, cholinesterase activity; TP, totalprotein; CR, creatinine; GLU, glutamic acid; TC, totalcholesterol; TG, triglycerides

■ REFERENCES(1) High, S. M.; Cohen, M. B.; Shu, Q. Y.; Altosaar, I. Achievingsuccessful deployment of Bt rice. Trends Plant Sci. 2004, 9 (6), 1360−1385.

(2) Hofmann, C.; Vanderbruggen, H.; Hofte, H.; Van Rie, J.; Jansens,S.; Van Mellaert, H. Specificity of Bacillus thuringiensis delta-endotoxinsis correlated with the presence of high-affinity binding sites in thebrush border membrane of target insect midgets. Proc. Natl. Acad. Sci.U. S. A. 1988, 85 (21), 7844−7848.(3) Chen, M.; Shelton, A.; Ye, G. Y. Insect-resistant geneticallymodified rice in China: from research to commercialization. Annu. Rev.Entomol. 2011, 56, 81−101.(4) Wang, Y.; Zhang, G.; Du, J.; Liu, B.; Wang, M. Influence oftransgenic hybrid rice expressing a fused gene derived from cry1Ab andcry1Ac on primary insect pests and rice yield. Crop Prot. 2010, 29 (2),28−33.(5) Lu, C. The first approved transgenic rice in China. GM crops2010, 1 (3), 113−115.(6) Wang, Z. H.; Wang, Y.; Cui, H. R.; Xia, Y. W.; Altosaar, I.; Shu,Q. Y. Toxicological evaluation of transgenic rice flour with a syntheticcry1Ab gene from Bacillus thuringiensis. J. Sci. Food Agric. 2002, 82 (7),738−744.(7) Schrøder, M.; Poulsen, M.; Wilcks, A.; Kroghsbo, S.; Miller, A.;Frenzel, T.; Danier, J.; Rychlik, M.; Emami, K.; Gatehouse, A.; Shu, Q.;Engel, K. Z.; Altosaar, I.; Knudsen, I. A 90-day safety study ofgenetically modified rice expressing Cry1Ab protein (Bacillusthuringiensis toxin) in Wistar rats. Food Chem. Toxicol. 2007, 45 (3),339−349.(8) Cao, S.; He, X.; Xu, W.; Luo, Y.; Yuan, Y.; Liu, P.; Cao, B.; Shi,H.; Huang, K. Safety assessment of transgenic Bacillus thuringiensis riceT1c-19 in Sprague-Dawley rats from metabonomics and bacterialprofile perspectives. IUBMB Life 2012, 64 (3), 242−250.(9) Wang, E. H.; Yu, Z.; Hu, J.; Jia, X. D.; Xu, H. B. A two-generationreproduction study with transgenic Bt rice TT51 in Wistar rats. FoodChem. Toxicol. 2014, 65, 312−320.(10) Qin, H. F. Safety assessment of rice genetically modified withCry1Ac and sck by feeding studies on broilers. Dissertation, ChineseAcademy of Agriculture Science, 2012 (in Chinese with Englishabstract).(11) Cao, Z. H.; Wang, Z. B.; Gu, X. H. Residues and organ damageof exogenous gene and protein from transgenic Bt brown rice ingrowing pigs. Chin. J. Anim. Nutr. 2014, 26 (7), 1908−1915 (inChinese with English abstract) .(12) Xu, Z. Y. Safety assessment of rice genetically modified withCry1Ac and sck by feeding studies on carp (Cyprinus carpio).Dissertation, Chinese Academy of Agriculture Science, 2011 (inChinese with English abstract).(13) Bernal, C. C.; Aguda, R. M.; Cohen, M. B. Effect of rice linestransformed with Bacillus thuringiensis toxin genes on the brownplanthopper and its predator Cyrtorhinus lividipennis. Entomol. Exp.Appl. 2002, 102 (1), 21−28.(14) Liu, W.; Lu, H. H.; Wu, W. X.; Wei, Q. K.; Chen, Y. X.; Thies, J.C. Transgenic Bt rice does not affect enzyme activities and microbialcomposition in the rhizosphere during crop development. Soil Biol.Biochem. 2008, 40 (2), 475−486.(15) Akhtar, Z. R.; Tian, J. C.; Chen, Y.; Fang, Q.; Hu, C.; Chen, M.;Peng, Y. F.; Ye, G. Y. Impacts of six bt rice lines on nontarget ricefeeding thrips under laboratory and field conditions. Environ. Entomol.2010, 39 (2), 715−726.(16) Li, Y.; Zhang, X.; Chen, X.; Romeis, J.; Yin, X.; Peng, Y.Consumption of Bt rice pollen containing Cry1C or Cry2A does notpose a risk to Propylea japonica (Thunberg) (Coleoptera:Coccinellidae). Sci. Rep. 2015, 5, 7679.(17) Li, Y.; Chen, X.; Hu, L.; Romeis, J.; Peng, Y. Bt rice producingCry1C protein does not have direct detrimental effects on the greenlacewing Chrysoperla sinica (Tjeder). Environ. Toxicol. Chem. 2014, 33(6), 1391−1397.(18) Viktorov, A. G. Transfer of Bt corn byproducts from terrestrialto stream ecosystems. Russ. J. Plant Physiol. 2011, 58 (4), 543−548.(19) Li, G.; Wang, Y.; Liu, B.; Zhang, G. Transgenic Bacillusthuringiensis (Bt) rice is safer to aquatic ecosystems than its non-transgenic counterpart. PLoS One 2014, 9 (8), e104270.

Journal of Agricultural and Food Chemistry Article

DOI: 10.1021/jf5055547J. Agric. Food Chem. XXXX, XXX, XXX−XXX

F

(20) Wang, J.; Chen, X.; Li, Y.; Zhu, H.; Ding, J.; Peng, Y. Effect ofstraw leachates from Cry1Ca-expressing transgenic rice on the growthof Chorella Pyrenoidosa. Environ. Toxicol. Chem. 2014, 33 (5), 1156−1162.(21) Blaustein, A. R.; Han, B. A.; Relyea, R. A.; Johnson, P. T. J.;Buck, J. C.; Gervasi, S. S.; Kats, L. B. The complexity of amphibianpopulation declines: understanding the role of cofactors in drivingamphibian losses. Ann. N. Y. Acad. Sci. 2011, 1223, 108−119.(22) Zhang, Q. L.; Li, Y. H.; Hua, H. X.; Yang, C. J.; Wu, H. J.; Peng,Y. F. Exposure degree of important non-target arthropods to Cry2Aain Bt rice fields. Chin. J. Appl. Ecol. 2013, 24 (6), 1647−1651 (InChinese with English abstract) .(23) Wang, Y.; Hu, H.; Huang, J.; Li, J.; Liu, B.; Zhang, G.Determination of the movement and persistence of Cry1Ab/1Acprotein released from Bt transgenic rice under field and hydroponicconditions. Soil Biol. Biochem. 2013, 58, 107−114.(24) Wang, H.; Ye, Q.; Gan, J.; Wu, L. Biodegradation of Cry1Abprotein from Bt transgenic rice in aerobic and flooded paddy soils. J.Agric. Food Chem. 2007, 55 (5), 1900−1904.(25) Cong, L.; Qin, Z. F.; Jing, X. N.; Yang, L.; Zhou, J. M.; Xu, X. B.Xenopus laevis is a potential alternative model animal species to studyreproductive toxicity of phytoestrogens. Aquat. Toxicol. 2006, 77 (3),250−256.(26) Wang, J. M.; Chen, X. P.; Liang, Y. Y.; Zhu, H. J.; Ding, J. T.;Peng, Y. F. Influence of transgenic rice expressing a fused Cry1Ab/1Acprotein on frogs in paddy fields. Ecotoxicology 2014, 23 (9), 1619−1628.(27) Alexander, T. W.; Reuter, T.; Aulrich, K.; Sharma, R.; Okine, E.K.; Dixon, W. T.; McAllister, T. A. A review of the detection and fateof novel plant molecules derived from biotechnology in livestockproduction. Anim. Feed Sci. Technol. 2007, 133, 31−62.(28) Li, X.; Huang, K.; He, X.; Zhu, B.; Liang, Z.; Li, H.; Luo, Y.Comparison of nutritional quality between Chinese indica rice with sckand cry1Ac genes and its nontransgenic counterpart. J. Food Sci. 2007,72 (6), S420−424.(29) Aumaitre, A.; Aulrich, K.; Chesson, A.; Flachowsky, G.; Piva, G.New feeds from genetically modified plants: substantial equivalence,nutritional equivalence, digestibility, and safety for animals and thefood chain. Livest. Prod. Sci. 2002, 74 (3), 223−238.(30) Van Eenennaam, A. L.; Young, A. E. Prevalence and impacts ofgenetically engineered feedstuffs on livestock populations. J. Anim. Sci.2014, 2 (10), 255−278.(31) Xu, W.; Cao, S.; He, X.; Luo, Y.; Guo, X.; Yuan, Y.; Huang, K.Safety assessment of Cry1Ab/Ac fusion protein. Food Chem. Toxicol.2009, 47 (7), 1459−1465.(32) Konarzewski, M.; Diamond, J. M. Evolution of basal metabolicrates and organ masses in laboratory mice. Evolution 1995, 49 (6),1239−1248.(33) Ash, J.; Novak, C.; Scheideler, S. E. The fate of geneticallymodified protein from Roundup Ready soybeans in laying hens. J.Appl. Poult. Res. 2003, 12 (2), 242−245.(34) Alexander, T. W.; Reuter, T.; Okine, E.; Sharma, R.; McAllister,T. A. Conventional and real-time polymerase chain reactionassessment of the fate of transgenic DNA in sheep fed RoundupReady rapeseed meal. Br. J. Nutr. 2006, 96 (6), 997−1005.(35) Chowdhury, E. H.; Kuribara, H.; Hino, A.; Sultana, P.; Mikami,O.; Shimada, N.; Guruge, K. S.; Saito, M.; Nakajima, Y. Detection ofcorn intrinsic and recombinant DNA fragments and Cry1Ab protein inthe gastrointestinal contents of pigs fed genetically modified cornBt11. J. Anim. Sci. 2003, 81 (10), 2546−2551.(36) Chowdhury, E. H.; Shimada, N.; Murata, H.; Mikami, O.;Sultana, P.; Miyazaki, S.; Yoshioka, M.; Yamanaka, N.; Hirai, N.;Nakajima, Y. Detection of Cry1Ab protein in gastrointestinal contentsbut not visceral organs of genetically modified Bt11-fed calves. Vet.Human Toxicol. 2003, 45 (2), 72−75.(37) Lutz, B.; Wiedemann, S.; Einspanier, R.; Mayer, J.; Albrecht, C.Degradation of Cry1Ab protein from genetically modified maize in thebovine gastrointestinal tract. J. Agric. Food Chem. 2005, 53 (5), 1453−1456.

(38) Guertler, P.; Lutz, B.; Kuehn, R.; Meyer, H. H. D.; Einspanier,R.; Killermann, B.; Albrecht, C. Fate of recombinant DNA andCry1Ab protein after ingestion and dispersal of genetically modifiedmaize in comparison to rapeseed by fallow deer (Dama dama). Eur. J.Wildlife Res. 2008, 54 (1), 36−43.(39) Li, Y.; Romeis, J. Bt maize expressing Cry3Bb1 does not harmthe spider mite, Tetranychus urticae, or its ladybird beetle predator,Stethorus punctillum. Biol. Control 2010, 53 (3), 337−344.(40) Yu, H.; Romeis, J.; Li, Y.; Li, X.; Wu, K. Acquisition of Cry1Acprotein by non-target arthropods in Bt soybean fields. PLoS One 2014,9 (8), e103973.

Journal of Agricultural and Food Chemistry Article

DOI: 10.1021/jf5055547J. Agric. Food Chem. XXXX, XXX, XXX−XXX

G