A multiplex nested PCR assay for the simultaneous detection of genetically modified soybean, maize and rice in highly processed products

Food Control 22 (2011) 1617e1623

Contents lists available at ScienceDirect

Food Control journal homepage: www.elsevier.com/locate/foodcont

A multiplex nested PCR assay for the simultaneous detection of genetically modified soybean, maize and rice in highly processed products Ao Jinxia, Li Qingzhang, Gao Xuejun*, Yu Yanbo, Li Lu, Zhang Minghui Detecting Center of Agricultural Genetically Modified Organism, Research Center of Life Science and Biotechnique, Northeast Agricultural University, 59 Mucai Road, Harbin 150030, PR China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 17 July 2010 Received in revised form 9 March 2011 Accepted 11 March 2011

The use of genetically modified organisms (GMOs) as food products becomes more and more widespread. The European Union has implemented a set of very strict procedures for the approval to grow, import and/or utilize GMOs as food or food ingredients. Thus, analytical methods for detection of the GMOs are necessary in order to verify compliance with labeling requirements. There are few effective screening methods for highly processed GM (genetically modified) products. Four genes (CP4-EPSPS, Cry1A(b), BAR, and, PAT) are common exogenous genes used in commercialized transgenic soybean, maize, and rice. In the present study, a multiplex nested polymerase chain reaction (PCR) method was developed to simultaneously detect the four exogenous genes and one endogenous gene in two runs. We tested eleven representative highly processed products samples (soya lecithin, soya protein powder, chocolate beverage, infant rice cereal, soybean refine oil, soybean salad oil, maize oil, maize protein powder, maize starch, maize jam) using the developed method, and amplicons of endogenous gene and transgenic fragments were obtained from all the processed products except for soybean refined oil, soybean salad oil and maize oil, and the sensitivity was 0.005%. These results indicate that multiplex nested PCR is appropriate for qualitative detection of transgenic soybean, maize and rice in highly processed products except for refined oil. Ó 2011 Elsevier Ltd. All rights reserved.

Keywords: Transgenic soybean Transgenic maize Transgenic rice Highly processed products Multiplex nested PCR

1. Introduction The use of genetically modified organisms (GMOs) as food products becomes more and more widespread (Ying et al., 2005; Nikoli et al., 2009) The European Union has implemented a set of very strict procedures for the approval to grow, import and/or utilize GMOs as food or food ingredients. Thus, analytical methods for the detection of GMOs are necessary in order to verify compliance with labeling requirements (Lu, Lin, & Pan, 2010; Litao et al., 2007). The most accepted analytical methods for GMO detection are based on DNA techniques such as polymerase chain reaction (PCR), since the protein-based methods are not reliable for highly processed food analysis. PCR gel-based assay is rapid, sensitive and simple, and used as a routine GMO detection method in many countries because of the relatively inexpensive and ordinary equipment required. Recently, several multiplex PCR assays based on the simultaneously amplification multiple sequences have been developed.

* Corresponding author. Tel.: þ86 0451 55190244. E-mail address: [email protected] (G. Xuejun). 0956-7135/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.foodcont.2011.03.018

Randhawa, Chhabra, and Singh (2009) developed a multiplex PCRBased simultaneous amplification of selectable marker and reporter genes for the screening of genetically modified crops. Bahrdt, Krech, Wurz, and Wulff (2010) developed a hexaplex realtime PCR assay for screening for presence of GMOs in food, feed and seed. These studies demonstrated that multiplex PCR was a costeffective and efficient assay for GM detection. Multiplex nested PCR can be generated by using the product of the first as a template when high specificity is required. Multiplex nested PCR method fully combines the high sensitivity and specificity of nested PCR and the rapidness of the multiplex PCR (Anna et al., 2010; Rondini et al., 2008). Application of the multiplex nested PCR not only can greatly improve sensitivity, but also can save considerable time and effort by decreasing the number of reactions required to assess the possible presence of GMOs in a food sample. Several methods using multiplex nested PCR for the detection of GM maize and/or soy have already been described. One multiplex nested PCR assay has been commercialized which simultaneously detects the presence of the 35S promoter as well as zein (maize) and lectin (soya) (Biosmart Allin 1.0 GMO Screening system, Promega, WI, USA). We also reported a triplex nested PCR

1618

A. Jinxia et al. / Food Control 22 (2011) 1617e1623

assay for the simultaneous detection of lectin and transgenic construct 35S-CTP and EPSPS-NOS of soybean (Minghui et al., 2007), but these analytical methods detect only one endogenous gene and several transgenic markers of one genetically modified organism or endogenous gene of two genetically modified organism and 35S promoter. However, a positive test for this CaMV35S promoter is not always conclusive. This sequence also occurs naturally in plants and plants may be naturally infected with the Cauliflower mosaic virus, the source of the CaMV35S promoter. Therefore a positive result of CaMV35S will not be sufficient to confirm the presence of GMO, but will suggest that it is probable. In such cases further PCR tests should be run with primers designed to amplify the specific transgenic DNA. In addition, more and more GMOs lacking the 35S promoter will not be detected by such screening and an alternative screening test will be required. Here, we describe a new multiplex nested PCR assay to simultaneously detect four mainly specific transgenic sequences and a common endogenous reference gene in highly processed products of transgenic soybean, maize and rice. It would be advantageous to detect more than one sequence per genetically modified organism (one endogenous gene and several transgenic sequences) or to screen several GMOs in one analysis. In this study, a multiplex nested PCR procedure that provides a simple and reliable identification of four main exogenous genes, 5-enolpyruvylshikimate-3-phosphate synthase (CP4-EPSPS) gene, Bacillus thuringiensis subsp (Cry1A(b)) gene, ribonuclease gene from Bacillus amyloliquefaciens (BAR), phosphinothricin acetyltransferase (PAT) gene and a common endogenous reference ribulose bisphosphat4e carboxylase/oxygenase large subunit (RBCL) gene in highly processed products of genetically modified soybean (GTS40-3-2, MON89788, A2704-12, A5547-127), maize (MON810, Bt176, Bt11, T25, TC1507, CBH351, NK603, MON88017, DAS-59122) and transgenic rice has been developed. The methods are reproducible since each sample has been repeated at least three times. Host specific internal target (the gene of RBCL) has been tested in all assays as a control to evaluate DNA quality and PCR efficacy, reducing the risk of false negatives, thereby increasing reliability. The method we described here is simple, reliable, efficient and sensitive, which offers a cost-effective alternative for routine GMO identification in food product analysis. 2. Materials and methods 2.1. Materials Transgenic soybean (GTS40-3-2, MON89788, A2704-12, A5547127), transgenic maize (Bt11, Bt176, MON810, TC1507, T25, CBH351, NK603, MON88017, DAS-59122) and transgenic rice were obtained from the Academy of Agriculture Science of China, as reference materials. The non-transgenic soybean, maize and rice were gifts from the Institute of Soybean of Northeast Agricultural University (NEAU, Harbin, China). The GM mixture samples included equal weights of each of four GMO (GM soybean containing CP4-EPSPS gene, GM maize T25 containing PAT genes, CBH351 containing BAR gene and GM rice containing Cry1A(b) gene) in non-GMO (soybean, maize and rice) were used as positive controls and the non-GM mixture samples included equal weights of non-transgenic soybean, maize and rice as negative controls. Eleven highly processed products containing soybean, maize and rice ingredients (labeled in their trademark) including soya lecithin, soya protein powder, chocolate beverage, infant rice cereal, soybean refined oil, soybean salad oil, maize oil, maize starch, maize protein powder, oatmeal, maize paste were chosen as blind samples bought from local markets.

2.2. DNA extraction and purification DNA extraction was performed with the WizardÒ Magnetic DNA Purification System for Food Kit (Promega, Madison, WI, USA) according to the procedure outlined in the technical manual; the DNA concentrations were measured by absorbance at 260 nm and DNA purities were measured by calculating the ratio of absorbance at 260e280 nm with the spectrophotometer DU-600 (Beckman, Fullerton, CA). Three replicate extracts for per sample were measured. Also the extracted DNA was loaded onto the agarose gel (0.8%) to check its purity. 2.3. Selection of primers for the multiplex nested PCR In order to detect the highly processed products of transgenic soybean, maize and rice, we first needed to obtain sequence information on exogenous genes to design primers. The sequences of the same exogenous gene in different breeds are much different after the gene has been transfered with different modifications. Furthermore, sometimes the sequences of the same exogenous gene such as Cry1A(b) gene in different lines: Bt11, Bt176 and MON810, are also different. So different primers need to be designed for one exogenous gene in different breeds or lines in routine PCR detection. In this study, the different sequences of the same exogenous gene in different breeds or lines were compared, and the consensus sequence regions were determined, then the general primers for different breeds or lines were designed to simplify the primer design and increase the detection efficiency. Using this method, we selected and designed Cry1A(b), PAT, BAR and CP4-EPSPS gene primers by the information in the available database (GMDD, http://gmdd.shgmo.org/). The Cry1A(b) gene primers were from Takeshi, Hideo, and Ken (2002), who successfully used them to amplify Cry1A(b) gene in transgenic maize Bt11, Bt176, MON810, and Delano, AnnaeMary, Erika, Margaret, and Saad (2003) got the same result. Primer pairs CP4 WF and CP4 WR (498 bp product), Primer pairs Rbcl WF and Rbcl WR (433 bp product), Primer pairs Bar WF and Bar WR (175 bp product) were from references (Jiang et al., 2003; Quan et al., 2002; Tan et al., 2003). And the other primer pairs were from this study (listed in Table 1). In addition, the product length was also an important factor which interfered with the PCR reaction. The DNA in processed products was destroyed and cut into little fragments, so these primers were designed to yield amplicon sizes within 100e500 bp. Each primer pair was also designed to distinguish the length of the amplified product from other amplified products. PCR primers were designed with Primer Premier V5.0 software and listed in Table 1. The larger fragments were amplified in the first round of the multiplex nested PCR, and afterward the smaller ones were amplified in the second round. To equalize as much as possible the intensities of all the PCR products in gel, according to different amplifying efficiencies, the suitable concentration of each primer pair was accurately calibrated. The primer concentrations for multiplex PCR were optimized by determination the minimum primer concentration. The primers were premixed to minimize the differences among the primer concentrations due to pipetting variability; the primer mixture prepared at a 4 concentration was ready to be diluted during the PCR assembly. The oligonucleotides were synthesized by Shanghai Sangon Bioengineering Technological Service Ltd. (Shanghai, China). 2.4. Multiplex nested PCR conditions The procedures of the first multiplex PCR round were performed in a final volume of 50 ml with the following reagent concentrations: genomic DNA 50 ng, PCR buffer 2, 0.4 mmol/L of each dNTPs

A. Jinxia et al. / Food Control 22 (2011) 1617e1623

1619

Table 1 List of primers in multiplex nested PCR.

The primer in the first multiplex PCR

The primer in the Second multiplex PCR

Primer pair

Sequence(50 /30 )

Amplified gene

Amplified fragment/bp

CP4 WF CP4 WR Rbcl WF Rbcl WR Cry WF Cry WR Bar WF Bar WR Pat WF Pat WR Rbcl NF Rbcl NR CP4 NF CP4 NR Bar NF Bar NR Cry NF Cry NR Pat NF Pat NR

CCTTCATGTTCGGCGGTCTCG GCGTCATGATCGGCTCGATG AATCTTCTACTGGTACATGGAC TCATCATCTTTGGTAAAATCAAG GGACAACAACCCAAACATCAAC TTGGTACAGGTTGCTCAGGCCCTC GTCTGCACCATCGTCAACC ACTCGGCCGTCCAGTCGTA AGATTAGGCCAGCTACAGCAGC GCAACCAACCAAGGGTATC AATCTTCTACTGGTACATGGAC GACCGTACTTGTTCAACTTATCC GGCGACGCCTCGCTCACAA GCGTCATGATCGGCTCGATG ACAAGCACGGTCAACTTCC ACTCGGCCGTCCAGTCGTA GGACAACAACCCAAACATCAAC GCACGAACTCGCTGAGCAG AGATTAGGCCAGCTACAGCAGC ACTCTTGTGGTGTTTGTGGC

CP4-EPSPS

498

RBCL

433

Cry1A(b)

322

BAR

201

PAT

160

RBCL

321

CP4-EPSPS

240

BAR

175

Cry1A(b)

152

PAT

100

(Tiangen, Beijing, China), primer mix I1  CP4 WF/CP4 WR, 0.8 mmol/L; Rbcl WF/Rbcl WR, 0.3 mmol/L; Cry WF/Cry WR, 0.25 mmol/L; Bar WF/Bar WR, 0.4 mmol/L; Pat WF/Pat WR, 0.4 mmol/ L, Taq DNA Polymerase (Tiangen, Beijing, China) 4U. Thermal cycler conditions were as follows: preincubation at 95  C for 10 min; 10 cycles consisting of dsDNA denaturation at 95  C for 30 s, primer annealing at 65  C for 60 s, primer extension at 72  C for 60 s. 10 cycles consisting of dsDNA denaturation at 95  C for 30 s, primer annealing at 60  C for 60 s, primer extension at 72  C for 60 s; 10 cycles consisting of dsDNA denaturation at 95  C for 30 s, primer annealing at 58  C for 60 s, primer extension at 72  C for 60 s; and final elongation at 72  C for 10 min. All the reactions were carried out on a Tgradient (Biometra, Germany) thermal cycler. The second multiplex nested PCR reaction was carried out in a final volume of 50 ml and contained 1 ml products (1:50 dilution) of the first amplification reaction, PCR buffer2, primer mix II1  (CP4 NF/CP4 NR, 0.6 mmol/L; Rbcl NF/Rbcl NR, 0.25 mmol/L; CryNF/Cry NR, 0.5 mmol/L; Bar NF/Bar NR, 0.3 mmol/L; Pat NF/Pat NR, 0.6 mmol/L), Taq DNA Polymerase 4U, the other reagents and the amplification condition were performed as described above. The PCR products were separated on 3% (w/v) agarose gel in TBE buffer and then visualized by ethidium bromide staining on UV transilluminator. 2.5. Real-time PCR conditions The amplification by real-time PCR were carried out in 20 ml containing 2 ml of DNA extract, 1 of TaqManÒ Premix Ex TaqTM (TaKaRa, DaLian, China), 0.2 mmol/L of each primer and 0.1 mmol/L probe (Sequences of primers and probes were named in Table 2), 1 of ROX Reference Dye. The real-time PCR assays were performed on ABI 7300 system (Applied Biosystems, Darmstadt, Germany), using the following conditions: 95  C for 30 s, 40 cycles at 95  C for 5 s and 60  C for 31 s, with collection of fluorescence signal at the end of each cycle. Data were collected and processed using 7300 Real-time PCR System Sequence Detection Software version 1.3. The GM content in a sample was determined by using comparative DCt values method, and the relative amount of the GM target sequence was compared to that of reference gene sequence. The standard curves were obtained by loading a series of known concentration standards (5, 1, 0.5, 0.1, 0.01%) of reference materials

for both the reference genes and exogenous genes. The GM content values were obtained by calculation using DCt and standard values. Real-Time PCR analysis was performed in triplicates for each sample from market as well as for reference samples. 3. Results and discussion 3.1. DNA extraction The genomic DNAs were extracted from samples with or without transgenic components by the WizardÒ Magnetic DNA Purification System for Food Kit, quantified by absorption at 260 nm and adjusted with TE to 10 ng/ml. The OD260/OD280 ratio of the final DNA ranged from 1.40 to 1.9. No DNA band could be detected with electropheresis in all blind samples, except the positive and negative controls (Fig. 1). This was probably because of the intensity of the highly processing procedures for production of these foods, and the breakdown of DNA in the physical processing under long term heating and/or low pH.

Table 2 List of primers and probes in real-time PCR. Primer/probe name Sequence(50 /30 )

Gene

Lec-1 Lec-2 Probe for Zein-1 Zein-2 Probe for SPS-1 SPS-2 Probe for CP4-1 CP4-2 Probe for Bar-1 Bar-2 Probe for Pat-1 Pat-2 Probe for Cry-1 Cry-2 Probe for

Lectin

CCTCCTCGGGAAAGTTACAA GGGCATAGAAGGTGAAGTT Lectin FAM-CCCTCGTCTCTTGGTCGCGCCCTCT-TAMRA TGAACCCATGCATGCAGT GGCAAGACCATTGGTGA Zein FAM-TGGCGTGTCCGTCCCTGATGC-TAMRA TTGGCCTGAACGGATAT CGGTTGATCTTTTCGGGATG Sps FAM-TCCGAGCCGTCCGTGCGTC-TAMRA CCGACGCCGATCACCTA GATGCCGGGCGTGTTGAG Cp4 FAM-CCGCGTGCCGATGGCCTCCGCA-TAMRA ACAAGCACGGTCAACTTCC ACTCGGCCGTCCAGTCGTA BAR FAM-CCGAGCCGCAGGAACCGCAGGAG-TAMRA GTCGACATGTCTCCGGAGAG GCAACCAACCAAGGGTATC PAT FAM-TGGCCGCGGTTTGTGATATCGTTAA-TAMRA GGGAAATGCGTATTCAATTCAAC TTCTGGACTGCGAACAATGG Cry1A(b) FAM-ACATGAACAGCGCCTTGACCACAGC-TAMRA

Zein

SPS

CP4-EPSPS

BAR

PAT

Cry1A(B)

1620

A. Jinxia et al. / Food Control 22 (2011) 1617e1623

Fig. 1. Electrophoresis of DNA extracts from negative, positive controls and eleven highly processed product. Lane M, Marker; lane 1, negative soybean; lane 2, negative rice; lane 3, negative maize; lane 4, GTS40-3-2; lane 5, MON89788; lane 6, A2704-12; lane 7, A5547-127; lane 8, transgenic rice; lane 9, Bt11; lane 10, Bt176; lane 121, Mon810; lane 12, T25; lane 13, TC1507; lane 14, CBH351; lane 15, NK603; lane 16, MON88017; lane 17, DAS-59122; lane 18, soya lecithin; lane 19, soya protein powder; lane 20, chocolate beverage; lane 21, infant rice cereal; lane 22, soybean refine oil; lane 23, soybean salad oil; lane 24, maize oil; lane 25, maize starch; lane 26, maize protein powder; lane 27, oatmeal; lane 28, maize paste.

However, the following PCR detection results suggested that the qualities of DNAs purified with the WizardÒ Magnetic DNA Purification System for Food Kit were enough for the subsequent PCR analysis of soya lecithin, soya protein powder, chocolate beverage, infant rice cereal, maize starch, maize protein powder, oatmeal and maize paste. However, the soybean refined oil, soybean salad oil and maize oil were not amplificated. The lack of amplification was due to the presence of PCR inhibitors in the DNA preparation or DNA content of oil was very low, or that the DNA were broken down by processing. Several difficulties to obtain amplifiable DNA from oil matrices have been reported. The refining process of oil, either physical or chemical, which may lead to an inferior quality of the final product, especially the heat treatments, the use of activated clays and pH variations, may affect the quantity and quality of the DNA that remains in the fully refined oil (Gryson et al., 2002). Previous reports evidenced positive results in the extraction and amplification of DNA from crude soybean oils (Gryson, Messens, & Dewettinck, 2004; Gryson et al., 2002; Pauli, Liniger, & Zimmermann, 1998). But for the positive detection of DNA in fully refined vegetable oils, very few studies are available. Pauli et al. showed that no genetic material can be recovered after the refined processing steps of soybean oil, and does not need to be labeled as a GMO product in Switzerland (Pauli et al., 1998). UK chief scientific advisor claimed that there’s no detectable GM DNA in refined food oils so they don’t need labeling (May, 1999). Bogani et al. (2009) achieved the qualitative detection of RR soybean at an industrial soybean processing chain until degummed oil and lecithin, but no data for the

Fig. 2. Electrophoresis of the amplification products of eleven highly processed products using external primers. Lane M, DL-2000; lane 1, blank control; lane 2, negative control; lane 3, positive control; lane 4, soya lecithin; lane 5, soya protein powder; lane 6, chocolate beverage; lane 7, infant rice cereal; lane 8, soybean refine oil; lane 9, soybean salad oil; lane 10, maize oil; lane 11, maize starch; lane 12, maize protein powder; lane 13, oatmeal; lane 14, maize paste.

subsequent steps and fully refined soybean oil. In another, the amplification of RR soybean by PCR assays was also achieved for all the extraction and refining steps, except for the intermediate steps of refining (neutralization, washing and bleaching) (Joana et al., 2010). These results show that the standardization of extraction methods to obtain pure enough DNA from refined oil is one of the main challenges to fulfill regulation requirements and much more effort and experiments will need to be performed to improve levels of refined oil DNA extraction in further study. 3.2. Application of multiplex nested PCR detection method The detection system developed consisted of a two-part approach: detection of RBCL gene (endogenous control gene of soybean, maize and rice) and detection of the transgenic elements CP4-EPSPS, Cry1A(b), BAR and PAT genes (exogenous genes of transgenic soybean, maize and rice). It’s very important to set an endogenous control for the transgenic detection. We chose the housekeeping gene (RBCL gene) which exists in soybean, maize and rice as the endogenous control. If the RBCL gene is amplified well, we can make the conclusion that the quality and the quantity of the template is suitable to the demand of the detection, otherwise we will make the opposite conclusion, and the condition of the DNA extraction should be optimized. Obviously, it is rather important to ensure the quality and quantity of the DNA so as to eliminate the false-negative amplification. Quan et al. (2002) found the general endogenous control gene (RBCL gene) from 23 kinds of plants including soybean, maize and rice. RBCL gene is now widely used as

Fig. 3. Electrophoresis of the amplification products of eleven highly processed products using internal primers. Lane M, DL-2000; lane 1, blank control; lane 2, negative control; lane 3, positive control; lane 4, soya lecithin; lane 5, soya protein powder; lane 6, chocolate beverage; lane 7, infant rice cereal; lane 8, soybean refine oil; lane 9, soybean salad oil; lane 10, maize oil; lane 11, maize starch; lane 12, maize protein powder; lane 13, oatmeal; lane 14, maize paste.

A. Jinxia et al. / Food Control 22 (2011) 1617e1623 Table 3 Parameters of regression equation of standard curves. Gene name

R2

Slope

Y-intercept

CP4-EPSPS Cry1A(b) (in maize) Cry1A(b) (in rice) PAT BAR

0.9972 0.9986 0.9992 0.9979 0.9988

0.2320 0.3021 0.3379 0.2883 0.3489

0.5020 0.5003 0.4572 0.3935 0.3736

endogenous control gene in the detection of the transgenic plants such as transgenic papaya, tomato, rice and strawberry (Gonsalves, 1998; Liu et al., 2006). Ten pairs of primers were selected in the detection assay, including primer group I: CP4 WF/CP4 WR, Rbcl WF/Rbcl WR, CryWF/Cry WR, Bar WF/Bar WR, Pat WF/Pat WR, and group II: CP4 NF/CP4 NR, Rbcl NF/Rbcl NR, CryNF/Cry NR, Bar NF/Bar NR, Pat NF/ Pat NR. The two groups were used in the first and second round of the multiplex nested PCR, respectively, amplifying fragments of 433, 498, 322, 201 and 160 bp in the first round and 321, 240, 175, 152 and 100 bp in the second. In the first round of multiplex PCR reaction, the RBCL, CP4-EPSPS, Cry1A(b), BAR and PAT amplicons in positive control were successfully amplified. But in all the blind samples, the specific amplicons of the expected size could not be detected, as shown in Fig. 2. The results showed that in all the eleven products, transgenic elements could not be detected by the primer group I (outer primers). Primer group II (inner primers) was selected for identification of the transgenic soybean, maize and rice in highly processed products in the second round multiplex nested PCR, as shown in Fig. 3. From soya lecithin, soya protein powder, chocolate beverage, infant rice cereal, the distinct amplicons were obtained, whose sizes were 321 bp (for RBCL) and 240 bp (CP4-EPSPS), showing that these four samples had the transgenic CP4-EPSPS gene. From maize starch and maize paste, the distinct amplicons were obtained, which sizes were 321 bp (for RBCL), 152 bp (Cry1A(b)) and 100 bp (PAT), showing that, these two sample had the transgenic Cry1A(b) gene and PAT gene. The distinct amplicons were obtained from oatmeal, which sizes were 321 bp (for RBCL) and 152 bp (Cry1A(b)), showing that, this sample had the transgene Cry1A(b). The distinct amplicons were obtained from maize protein powder, which sizes were 321 bp (for RBCL), 175 bp (BAR) and 152 bp (Cry1A(b)), showing that, this sample had the transgenic Cry1A(b) gene and BAR gene. All studies failed to extract the DNA from soybean refined oil, soybean salad oil and maize oil. No other transgenes were identified, and no unspecific bands were present. In another experiment, the general use of the designed primers was also individually assessed by PCR tests. The Cry1A(b) gene primers were used for the detection of transgenic maize Bt11, Bt176, MON810 and transgenic rice; the PAT gene primers were used to detect PAT fragment in transgenic maize (Bt11,T25, TC1507, DAS-59122) and soybean (A2704-12, A5547-127); the BAR gene primers were used to detect BAR fragment in transgenic maize Bt176 and CBH351; CP4-EPSPS gene primers were used to detect transgenic soybean (GTS40-3-2, MON89788) and maize (NK603, MON88017). The expected amplification products could be obtained from transgenic maize (Bt11, Bt176, MON810) and transgenic rice using Cry1A(b) gene primers in the PCR reaction, this is consistent with previous studies that have shown that under the right conditions amplification is possible with primers that are 63e83% complementary, with 5e10 perfectly matched bases at the 30 end of the primer (Delano et al., 2003; Takeshi et al., 2002). The Cry1A(b) gene primers used in this study were described and used by Takeshi et al. (2002), but they did not report that the primers also were used in transgenic rice.

1621

The expected amplification products could be obtained from transgenic maize (Bt11, T25, TC1507, DAS-59122) and soybean (A2704-12, A5547-127) using PAT gene primers in the PCR reaction, Correct PCR products could be obtained from transgenic maize (Bt176, CBH351) using BAR gene primers; from transgenic soybean (GTS40-3-2, MON89788) and maize (NK603, MON88017) using CP4-EPSPS gene primers; and from soybean, maize and rice using RBCL gene primers. These results show that these primers correctly amplified the transgenes and endogenous gene in different transgenic lines. To confirm the multiplex nested PCR results and to have an estimate of DNA amount, all the extracts were amplified by realtime PCR using specific fluorescent probes proposed by SN/T 1204 (Zhu et al., 2003). The lectin gene, zein gene and SPS (sucrose phosphate synthase) gene were used as reference genes of soybean, maize and rice, respectively. For quantitative analysis, the standard curves were obtained by loading a series of known concentration standards (5, 1, 0.5, 0.1, 0.01%) of reference materials. The linear correlation coefficient of the standard curves (R2) and real-time PCR amplification results for all samples were presented at Table 3 and 4, respectively. The detection of all samples was also in good agreement with the multiplex nested PCR results and also no amplifications in soybean refined oil, soybean salad oil and maize oil, which indicated the DNA content was very low in these samples. In a conclusion, the presented data showed that a highly specific and sensitive multiplex nested PCR system for the transgenic soybean, maize and rice in highly processed products has been developed. One reason we failed to detect the transgenic elements from the soybean refined oil, soybean salad oil and maize oil was because of the broken down DNA by processing as a result of long term heating and/or low pH; in addition, PCR inhibitors might be present. These results, repeated three times, did not show variability and highlighted a high reproducibility. The multiplex nested PCR system could be used to achieve maximum specificity and sensitivity. 3.3. Specificity and sensitivity analysis of the multiplex nested PCR The routine analysis of multiplex PCR products may lead to false positive results if an artifact of similar size as a given target sequence is amplified. This limitation can be overcome by simultaneous detection of two PCR. The methods of confirmation include: gel electrophoresis, Southern blot assay, nested PCR and sequencing. The results of nested PCR had high specificity and could eliminate the identification assay (Khairnar & Parija, 2007). As expected, five bands for transgenic soybean, maize and rice were Table 4 Real-time PCR results of in food samples. Food sample

CP4-EPSPS gene

Soya lecithin Soya protein powder Chocolate beverage Infant rice cereal Soybean refined oil Soybean salad oil Maize oil Maize starch Maize protein powder Oatmeal Maize paste

4.48 4.69 7.44 6.74 e e e e e e e

Cry1A(b) gene

PAT gene

BAR gene

DCt GMO(%) DCt GMO(%) DCt GMO(%) DCt GMO(%) 2.87 2.57 0.59 0.86 e e e e e e e

e e e e e e e 6.67 6.35 6.07 6.59

e e e e e e e 0.31 0.39 0.32 0.33

Note: the “e” mean that no product was detected.

e e e e e e e 7.30 e e 7.18

e e e e e e e 0.34 e e 0.37

e e e e e e e

e e e e e e e

6.95 0.35 e e e e

1622

A. Jinxia et al. / Food Control 22 (2011) 1617e1623

Fig. 4. Sensitivity analysis of the first round multiplex nested PCR by eight level of simulated GM mixture samples. Lane M, DL-2000 Marker; lane 1, blank control; lane 2, negative control; lane 3, positive control; lane 4e11, the contents of transgenic soybean, maize and rice are 5%, 1%, 5  101%, 5  102%, 5  103%, 1  103%, 5  104%, 1  104%.

simultaneously amplified, one corresponding to the endogenous control (RBCL gene), others corresponding to the transgenic soybean, maize and rice specific amplicons of CP4-EPSPS, Cry1A(b), BAR and PAT amplicons, confirming the specificity of the primer pairs chosen for each transgenic element. The assay was able to amplify certain genes of the genome of most blind samples and reference materials. The amplicons being of different sizes, allowed a simple and specific detection of each element in transgenic soybean, maize and rice. To evaluate the sensitivity of this multiplex nested PCR detection system, we used the equivalent DNA mix of four GMO (GM soybean with CP4-EPSPS gene, GM maize T25 with PAT gene, CBH351 with BAR gene and GM rice with Cry1A(b)) as positive samples (per exogenous genes concentration was 25%), and used the equivalent DNA mix of three non-GMO (soybean, maize and rice) as negative samples. We prepared eight level of simulated GM mixture samples by mixing different weights of positive samples and negative samples, each mixture contained 5%, 1%, 0.5%, 0.05%, 5  103%, 1  103%, 5  104% and 1  104% (w/w) of each of four GMO in non-GMO. After DNA extraction, 100 ng of DNA from each mixture was used in the multiplex nested PCR detection system. In the first round of multiple PCR, the bands of BAR and CP4EPSPS genes in the 5  102% samples were not detected, while five target genes were simultaneously amplified in the 5  101% samples. So the sensitivity of the first round of multiple PCR was 5  101% (Fig. 4). The bands of BAR and PAT genes in the 1  103% samples were not detected in the second multiplex PCR, while the bands of five target genes in the 5  103% samples were clearly visible. So the sensitivity of the second round of multiple PCR was 5  103%. The sensitivity of second round of multiplex PCR was chosen as the sensitivity of multiplex nested PCR. So target gene

amplicons (RBCL, CP4-EPSPS, Cry1A(b), BAR and PAT gene) were detected in mixtures with as little as 5  103% transgenic DNA (Fig. 5). Amplicons having the expected sizes were visible, and the multiplex PCR mixture did not reduce the sensitivity of the test (Dario et al., 2009), and the nested PCR was 1000-fold more sensitive than the common PCR. Multiplex nested PCR overcome the limit of “platform stage effect” in a single amplification, increased amplification multiples, and greatly improved the sensitivity of PCR. Alternatively, template and primer changes in the second multiplex reaction, reduced nonspecific reaction on the possibility of continuous amplification, ensured the reaction of specificity. In addition, inside primer amplification template was the amplification product of the outside primer. The second multiplex reactions could be carried on, which could identify the accuracy of the first multiplex reaction, so the accuracy and feasibility of the entire reaction could be realized (Edwards & Gibbs, 1994). These studies indicated that at conditions used in this experiment, the presence of non-transgenic soybean, maize and rice DNA did not decrease the sensitivity of the detection system. In short, this detection assay of multiplex nested PCR for identification of positive signals is a fast, sensitive and specific detection system, so it can be used as a reliable choice for routine detection method for transgenic soybean, maize and rice in highly processed products. The sensitivity of the multiplex nested PCR is higher than that of the multiplex PCR for the detection of GM soy (Forte et al., 2005) (sensitivity ¼ 0.5%), maize (Germini et al., 2004; Mari et al., 2005) (sensitivity ¼ 0.25%) and canola (James, Schmidt, Wall, Green, & Masri, 2003) (sensitivity ¼ 0.1%). Furthermore, the multiplex nested PCR assay discriminated the transgenic soybean, maize and rice very quickly, reproducibly and in a cost-saving and less time-consuming way. It is also a flexible assay because it is carried out in the same tube for two runs. The advantage of multiplex nested PCR is that fewer reactions are needed to test a sample for potential presence of transgenic soybean, maize and rice-derived DNA. In conclusion, a rapid method is proposed for the simultaneous detection of five target sequences in genetically modified soybean (GTS40-3-2, MON89788, A2704-12, A5547-127), maize (MON810, Bt176, Bt11, T25, TC1507, CBH351, NK603, MON88017, DAS-59122) and transgenic insect-resistant rice in two runs. Hence, the proposed method comprises a rapid, reliable, simple and sensitive (down to 5  103%) multiplex nested PCR, suitable for the detection of CP4-EPSPS, Cry1A(b), BAR, PAT exogenous genes of genetically modified soybean, maize and rice in highly processed products except for refined oil. But this multiplex nested PCR assay could not discriminate the transgenic soybean, maize and rice, further tests are needed if we want to discriminate them. The main limitation for the qualitative analysis of highly processed products was the quality of DNA obtained. The standardization of extraction methods to obtain pure enough DNA from refined oil is one of the main challenges to fulfill regulation requirements and we will make detailed study in the future.

Acknowledgments

Fig. 5. Sensitivity analysis of the second round multiplex nested PCR by eight level of simulated GM mixture samples. Lane M, DL-2000 Marker; lane 1, blank control; lane 2, negative control; lane 3, positive control; lane 4e11, the contents of transgenic soybean, maize and rice are 5%, 1%, 5  101%, 5  102%, 5  103%, 1  103%, 5  104%, 1  104%.

This study was supported by the Funds of China Postdoctoral Science Foundation (No. 20100471242), the Fund of Young Scientist Foundation of Heilongjiang China (No. QC2009C49), the Fund Supported by Heilongjiang Postdoctoral Science Foundation (No. LBH-Z10250) and the Fund of National Genetically Modified Organisms Major Projects China (No. 2008ZX08012-001). This is gratefully acknowledged.

A. Jinxia et al. / Food Control 22 (2011) 1617e1623

References Anna, K. W., Raymond, C. C., Nidhi, A., Manoj, K. S., Stephen, N. W., & Shikha, B. (2010). Human papillomavirus genotypes in anal intraepithelial neoplasia and anal carcinoma as detected in tissue biopsies. Modern Pathology, 23, 144e150. Bahrdt, C., Krech, A. B., Wurz, A., & Wulff, D. (2010). Validation of a newly developed hexaplex real-time PCR assay for screening for presence of GMOs in food, feed and seed. Analytical and Bioanalytical Chemistry, 396(6), 2103e2112. Bogani, P., Minunni, M., Spiriti, M. M., Zavagliam, M., Tombellim, S., Buiatti, M., et al. (2009). Transgenes monitoring in an industrial soybean processing chain by DNA-based conventional approaches and biosensors. Food Chemistry, 113, 658e664. Dario, D. M., Fabrizio, A., Gary, M. W., Miia, L., Ute, M., Clare, F., et al. (2009). Multiplex PCR for detection of botulinum neurotoxin-producing clostridia in clinical, food, and environmental samples. Applied and Environment Microbiology, 75(20), 6457e6461. Delano, J., AnnaeMary, S., Erika, W., Margaret, G., & Saad, M. (2003). Reliable detection and identification of genetically modified maize, soybean, and canola by multiplex PCR analysis. Journal of Agriculture and Food Chemistry., 51, 5829e5834. Edwards, M. C., & Gibbs, R. A. (1994). Multiplex PCR: advantages, development, and applications. Genome Research, 3, 65e75. Forte, V. T., Di Pinto, A., Martino, C., Tantillo, G. M., Grasso, G., & Schena, F. P. (2005). A general multiplex-PCR assay for the general detection of genetically modified soya and maize. Food Control, 16, 535e539. Germini, A., Zanetti, A., Salati, C., Rossi, S., Forré, C., Schmid, S., et al. (2004). Development of a seven-target multiplex PCR for the simultaneous detection of transgenic soybean and maize in feeds and foods. Journal of Agriculture and Food Chemistry, 52(11), 3275e3280. Gonsalves, D. (1998). Control of papaya ringspot virusin papaya: a case study. Annual Review of Phytopathology, 36(1), 415e437. Gryson, N., Messens, K., & Dewettinck, K. (2004). Influence of different oil-refining parameters and sampling size on the detection of genetically modified DNA in soybean oil. Journal of the American Oil Chemists Society, 81, 231e234. Gryson, N., Ronsse, F., Messens, K., De Loose, M., Verleyen, T., & Dewettinck, K. (2002). Detection of DNA during the refining of soybean oil. Journal of the American Oil Chemists Society, 79, 171e174. James, D., Schmidt, A. M., Wall, E., Green, M., & Masri, S. (2003). Reliable detection and identification of genetically modified maize, soybean, and canola by multiplex PCR analysis. Journal of Agriculture and Food Chemistry, 51(20), 5829e5834. Jiang, Y., Zhu, C. Q., & Lin, H. (2003). Protocol of the qualitative polymerase chain reaction for detecting genetically modified component in soybeans, industry standard of entryeexit inspection and quarantine of the People’s Republic of China. SN/T 1195. China: Industry Standard. Joana, C., Isabel, M., Joana, S. A., & Oliveira, M. B. P. P. (2010). Monitoring genetically modified soybean along the industrial soybean oil extraction and refining processes by polymerase chain reaction techniques. Food Research International, 43, 301e306. Khairnar, K., & Parija, S. C. (2007). A novel nested multiplex polymerase chain reaction (PCR) assay for differential detection of Entamoeba histolytica, E.

1623

moshkovskii and E. dispar DNA in stool samples. Biomedical Chromatography Microbiology, 7, 47. Liu, J. Y., Deng, X., Kang, L., & Chen, D. M. (2006). Detection of transgenic papaya by SYBR-green real time PCR. Journal of Hunan Agriculture University, 32(4), 371e374. Litao, Y., Jinchao, G., Aihu, P., Haibo, Z., Kewei, Z., Zhengming, W., et al. (2007). Event-specific quantitative detection of nine genetically modified maizes using one novel standard reference molecule. Journal of Agriculture Food Chemistry, 5, 15e24. Lu, I. J., Lin, C. H., & Pan, T. M. (2010). Establishment of a system based on universal multiplex-PCR for screening genetically modified crops. Analytical and Bioanalytical Chemistry, 396, 2055e2064. Mari, O., Takeshi, M., Takeshi, K., Koichi, K., Satoshi, F., Hiroshi, A., et al. (2005). Development of a multiplex polymerase chain reaction method for simultaneous detection of eight events of genetically modified maizes. Journal of Agriculture Food Chemistry, 53, 9713e9721. May, R. M. (1999). Genetically modified foods: Facts, worries, policies, and public confidence. London, UK: Office of Science and Technology. [online]. http://www. dti.gov.uk/ost/genetic/geni.htm. Minghui, Z., Xuejun, G., Yanbo, Y., Jinxia, A., Jun, Q., Yonghao, Y., et al. (2007). Detection of roundup ready soy in highly processed products by triplex nested PCR. Journal of Food Control, 18, 1277e1281. Nikoli, Z., Talki-Ajdukovi, K., Tati, M., & Balelevi-Tubi, S. (2009). Monitoring of the roundup ready soybean in the Vojvodina province in Serbia. Industrial Crops and Products, 29, 2e3. Pauli, U., Liniger, M., & Zimmermann, A. (1998). Detection of DNA in soybean oil. Z Lebensmittel Untersuch Forsch A, 207, 264e267. Quan, J. X., Zhang, Y. B., Cheng, C. F., Liang, C. Z., & Wei, X. T. (2002). The construction of reference system for plant genetically modified PCR detection. Research of Plant in Yunnan, 24(3), 333e340. Randhawa, G. J., Chhabra, R., & Singh, M. (2009). Multiplex PCR-based simultaneous amplification of selectable marker and reporter genes for the screening of genetically modified crops. Journal of Agriculture Food Chemistry, 57(12), 5167e5172. Rondini, S., Pingle, M. R., Das, S., Tesh, R., Rundell, M. S., Hom, J., et al. (2008). Development of multiplex PCR-ligase detection reaction assay for detection of West Nile virus. Journal of Clinical Microbiology, 46(7), 2269e2279. Takeshi, M., Hideo, K., & Ken, T. (2002). Detection of Recombinant DNA segments introduced to genetically modified maize (Zea mays). Journal of Agricultural and Food Chemistry, 50, 2100e2109. Tan, W., Dong, J., Deng, H. L., Wu, H. Z., & Xu, B. L. (2003). Protocol of the qualitative polymerase chain reaction for detection genetically modified plant components in food, industry standard of entryeexit inspection and quarantine of the People’s Republic of China. SN/T 1202. China: Industry Standard. Ying, C., Yuan, W., Yiqiang, G., & Baoliang, X. (2005). Degradation of endogenous and exogenous genes of roundup ready soybean during food processing. Journal of Agricultural and Food Chemistry, 53, 10239e10243. Zhu, S. F., Tan, W., Cao, J. J., Zhang, G. M., Pan, L. W., Huang, W. S., et al. (2003). Protocol of the real-time PCR for detecting genetically modified plants and their derived products, industry standard of entry-exit inspection and quarantine of the People’s Republic of China. SN/T 1204. China: Industry Standard.