Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Evaluation of Four Endogenous Reference Genes and Their Real-Time PCR Assays for Common Wheat Quantification in GMOs Detection

  • Huali Huang,

    Affiliation National Center for Molecular Characterization of Genetically Modified Organisms, SJTU-Bor Luh Food Safety Center, School of Life Science and Biotechnology, Shanghai Jiao Tong University, Shanghai, People’s Republic of China

  • Fang Cheng,

    Affiliation National Center for Molecular Characterization of Genetically Modified Organisms, SJTU-Bor Luh Food Safety Center, School of Life Science and Biotechnology, Shanghai Jiao Tong University, Shanghai, People’s Republic of China

  • Ruoan Wang,

    Affiliation National Center for Molecular Characterization of Genetically Modified Organisms, SJTU-Bor Luh Food Safety Center, School of Life Science and Biotechnology, Shanghai Jiao Tong University, Shanghai, People’s Republic of China

  • Dabing Zhang,

    Affiliation National Center for Molecular Characterization of Genetically Modified Organisms, SJTU-Bor Luh Food Safety Center, School of Life Science and Biotechnology, Shanghai Jiao Tong University, Shanghai, People’s Republic of China

  • Litao Yang

    yylltt@sjtu.edu.cn

    Affiliation National Center for Molecular Characterization of Genetically Modified Organisms, SJTU-Bor Luh Food Safety Center, School of Life Science and Biotechnology, Shanghai Jiao Tong University, Shanghai, People’s Republic of China

Evaluation of Four Endogenous Reference Genes and Their Real-Time PCR Assays for Common Wheat Quantification in GMOs Detection

  • Huali Huang, 
  • Fang Cheng, 
  • Ruoan Wang, 
  • Dabing Zhang, 
  • Litao Yang
PLOS
x

Abstract

Proper selection of endogenous reference genes and their real-time PCR assays is quite important in genetically modified organisms (GMOs) detection. To find a suitable endogenous reference gene and its real-time PCR assay for common wheat (Triticum aestivum L.) DNA content or copy number quantification, four previously reported wheat endogenous reference genes and their real-time PCR assays were comprehensively evaluated for the target gene sequence variation and their real-time PCR performance among 37 common wheat lines. Three SNPs were observed in the PKABA1 and ALMT1 genes, and these SNPs significantly decreased the efficiency of real-time PCR amplification. GeNorm analysis of the real-time PCR performance of each gene among common wheat lines showed that the Waxy-D1 assay had the lowest M values with the best stability among all tested lines. All results indicated that the Waxy-D1 gene and its real-time PCR assay were most suitable to be used as an endogenous reference gene for common wheat DNA content quantification. The validated Waxy-D1 gene assay will be useful in establishing accurate and creditable qualitative and quantitative PCR analysis of GM wheat.

Introduction

Although genetically modified organisms (GMOs) have been developed and marketed by many countries during the past two decades, controversies over safety issues have always been hot topics in public discussions. To protect consumers’ rights, many countries and regions have issued regulations and legislations to strengthen commercial GMO release and labeling management [13]. To effectively implement GMO labeling regulations, many efforts have been made to develop sensitive, accurate, and reliable methods for identification and quantification of GMOs, either gene-specific or event-specific [47].

For detection and quantification of GM contents, techniques based on nucleic acid analysis have been widely applied, including conventional PCR and TaqMan real-time PCR analysis [8]. The efficient and accurate quantification of host genome DNA copy numbers using endogenous reference genes is very important during the process [9]. In general, GM contents are expressed as mass/mass ratio or copy number ratio of GM over non-GM of the assayed organism. Therefore, endogenous reference genes and their real-time PCR assays are referred to as “golden standards” in GMO analysis [8]. One desirable endogenous reference gene and its real-time PCR assay should have three characteristics: species specificity, single or low copy number in the genome, and low heterogeneity among different lines [10]. In general, low heterogeneity is determined by minimum number of single nucleotide polymorphisms (SNPs) in the target DNA sequence and high PCR amplification performance among different lines [11,12].

To date, many endogenous reference genes and their real-time PCR assays have been developed for commercial GM crops, such as Invertase I, zein, zSSIIb, adh and hmg-A genes for maize; HMG I/Y, FatA, CruA, and BnACCg8 genes for rapeseed; PLD, SPS, GOS9, and ppi-PPF genes for rice [8]. However, the increased number of reported endogenous reference genes has made it difficult to select the best candidate for a specific GMO analysis, and how to harmonize these endogenous reference genes is becoming not only important but also necessary in some cases. Recently, maize and rice endogenous reference genes and their real-time PCR assays were evaluated and harmonized employing 84 maize varieties and 58 rice varieties from different geographic and phylogenic origins, and the maize zein and/or zSSIIb gene, and the rice SPS gene assays were selected as best candidate in GM maize and rice analysis, respectively [11,12]. These reports demonstrated that proper evaluation and harmonization of different endogenous reference genes would be beneficial to host genome DNA quantification in routine laboratory analysis, proficiency testing, and GMO traceability.

Common wheat (Triticum aestivum L.) is one of the major staple food crops grown on more than 17% of the global cultivated land [13]. Wheat is the last major cereal crop to be genetically modified due to a complex genome structure, its recalcitrance to tissue culture, and challenges using Agrobacterium mediated gene transfer in this species [14,15]. Although no GM wheat has been commercialized so far, several GM wheat events, such as herbicide-tolerant and Fusarium-resistant GM wheat, are in the pipeline [16,17]. Many agricultural biotech companies (Monsanto, Bayer CropScience, and Syngenta, etc) also showed great interest in developing GM wheat events with traits such as drought tolerance, high yield, and more efficient phosphorus absorption [13]. For GM wheat detection and quantification, four different endogenous reference genes, ACC1, PKABA1, ALMT1, and Waxy-D1, as well as their corresponding real-time PCR assays have been previously reported [1821]. The ACC1 gene and its real-time PCR assay was suitable for the quantification of not only common wheat but also durum wheat, with the PCR performance verified in reactions employing 18 common wheat and 10 durum lines as templates [18]. The PKABA1 real-time PCR assay was specific to wheat and barley, and one duplex real-time PCR assay with one set of primers and two TaqMan probes was developed for quantification of wheat and barley. However, it is difficult to develop one suitable PKABA1 assay for GM wheat analysis with high specificity [19]. ALMT1 real-time PCR assay was highly specific to common wheat and gave stable PCR performance in 15 lines [20]. The Waxy-D1 real-time PCR assay was also highly specific to common wheat and generated similar PCR performance in 19 common wheat varieties [21].

To determine the most suitable endogenous reference gene and its real-time PCR assay for common wheat, we evaluated and harmonized the previously reported four endogenous reference genes and their assays according to the PCR target DNA sequence variations and real-time PCR performance among different hexaploid wheat lines. The results suggested that Waxy-D1 gene and its real-time PCR assay was the most suitable because of its minimum sequence variation and best overall PCR performance among 37 tested hexaploid wheat lines.

Materials and Methods

Plant Materials

Seeds of 43 wheat and relatives were kindly provided by the Institute of Crop Germplasm Resources, Chinese Academy of Agricultural Sciences (ICGR-CAAS), including 1 Aegilops tauschii, 1 Aegilops spehoides, 1 Triticum urartu, 1 Triticum durum (Cannizzo), 2 triticale (X triticosecale Wittmack), and 37 hexaploid common wheat (Triticum aestivum) lines from different geographic and phylogenic origins. Brief description of the 43 wheat and relatives were listed in Table S1 in File S1, and detailed information could be obtained from ICGR-CAAS website (icgr.caas.net.cn). We grouped these lines into four categories according to ploidy, which were diploid (genotype BB, DD and AA, including Aegilops tauschii, Aegilops speltoides, and Triticum urartu), tetraploid (Triticum durum, genotype BBAA), Hexaploid (genotype BBAADD, such as zhongmai 175), and octoploid triticale (genotype BBAADDRR, such as Xiaoyan 22 and Zhongguochun). In order to evaluate the PCR performance of the endogenous reference gene assays among different hexaploid common wheat lines, one common wheat was randomly selected (Jimai 19) for constructing of PCR standard curves and related analyses. All plants were grown in our greenhouse from germinated seeds, and fresh leaves were used for genomic DNA extraction.

Genomic DNA Extraction and Purification

Genomic DNAs used for qualitative PCR and quantitative real-time PCR analysis were extracted and purified using Mini Plant Genomic DNA Extraction Kit (Shanghai Ruifeng Agro-tech Co. Ltd., Shanghai, China) according to manufacturer’s manual. The quantity and quality of the extracted DNA were evaluated using a Thermo Nanodrop 1000 spectrometer and 1.0% agarose gel electrophoresis. Each DNA preparation was firstly adjusted to a stock solution of 20 ng/µl, from which further dilutions were made.

PCR Primers and Probes

According to the reported DNA sequence [1821], PCR primers targeting the four endogenous reference genes were designed using Primer Premier 5.0 (PREMIER Biosoft Company, Canada) and used to amplify target gene fragment for sequencing. Two sets of primers and probes were used for real-time PCR analysis of the four endogenous reference genes. The first set was primers and probes described in previous reports [1821]. The second set contained redesigned primers and probes based on the sequencing results of each amplified target DNA fragment to match the observed SNPs, including Q-Acc1-F2 and Q-ALMT1-F2 primers, Q-ALMT1-p2 and Q-PKABA1-p2 probes. All PCR primers and probes were synthesized by either Invitrogen Co., Ltd. or TaKaRa Biotechnology Co., Ltd and listed in Table 1.

PCR typeGene namePrimer/Probe nameSequence(5’–3’)Amplicon size
Sequence PCRAcc1Seq-Acc1-FTGCATCTGCGCTGTTTGT184bp
Seq-Acc1-RAATGACTGAACACGACGACT
ALMT1Seq-ALMT1-FCCAAGGTGCTTAGGGATC171bp
Seq-ALMT1-RCAAGAAGTGTTGCGGTGA
Waxy-D1Seq-Waxy-FCTATTGGTTCTCCGTGTTTG263bp
Seq-Waxy-RGTCATCTGTCATTTCCTGGTT
PKABA1Seq–PKABA1-FAGGGAGATTTAGCGAGGAT211bp
Seq-PKABA1-RTGTGCCGACAGTTGACTT
Real-time PCRAcc1Q-Acc1-FTGCCCATTGTCGGCCTTA93bp
Q-Acc1-RGGGCAGATGGTTGGAATGC
Q-Acc1-pFAM-TGCCTCGACAACACCATCGCTATCC-TAMRA
Q-Acc1-F2TGCCCATCGTCGGCCTTA
ALMT1Q-ALMT1-FAATGACTGTGCCGTCTCCAGT95bp
Q-ALMT1-RACAGAGCCGTGTTCTCTGCA
Q-ALMT1-pVIC-CGTGAAAGCAGCGGAAAGCCTCAGA-TAMRA
Q-ALMT1-F2ACTGACTGTGCCGTCTCCAGT
Q-ALMT1-p2VIC-CGTAAAAGCAGCGGAAAGCCTCAGA-TAMR
Waxy-D1Q-Waxy-FGTCGCGGGAACAGAGGTGT87bp
Q-Waxy-RGGTGTTCCTCCATTGCGAAA
Q-Waxy-pFAM-CAAGGCGGCCGAAATAAGTTGCC-TAMRA
PKABA1Q-PKABA1-FCAAGTATGTCATAGAGATTTGAA87bp
Q-PKABA1-RGTAACCGAAGTCACAAATCT
Q-PKABA1-PFAM-TCGCACCTCGGCT-MGBNFQ
Q-PKABA1-p2FAM-TCGCACCTCGACT-MGBNFQ

Table 1. Nucleotide sequences of the primers and probes used in this study.

CSV
Download CSV

Sequencing of amplified target DNA fragment

Target DNA fragments of Acc1, ALMT1, Waxy-D1 and PKABA1 in all 43 lines were amplified by PCR. The PCR reactions were carried on Verity thermal cycler (Applied Biosystems, Foster City, CA) with a 50 µl reaction volume. The PCR reaction contained 1× PCR reaction buffer for KOD-plus-DNA polymerase, 0.2 mM dNTPs, 1 mM MgCl2, 0.2 µM of each primer, 0.5 unit KOD-plus-DNA polymerase (TOYOBO Co. Ltd.), and 20 ng of extracted genomic DNA. The PCR program was: template denaturation at 94°C for 5 minutes; followed by 35 cycles of template denaturation at 94°C for 30 seconds, primer annealing at 58°C for 30 seconds, product extension at 68°C for 50 seconds; and a final PCR product extension at 68°C for 5 minutes. The PCR products were electrophoresed on 2% agarose gels to verify the correct size of each amplicon and target DNA fragment was excised from the gel and purified using DNA gel extraction kit (Axygen Bio-technology, Hangzhou, China). Each purified DNA was sequenced using ABI 3730 DNA Analyzer (BGI Co. Ltd., Shanghai, China). The sequences were blasted and aligned with target DNA sequence using Vector NTI Advance 10 software (Life Technologies, USA) to reveal the SNPs.

Real-time PCR

Real-time PCR reactions were performed in optical 96-well PCR plates using ABI PRISM 7900HT sequence detection system (Applied Biosystems, Foster City, CA). The PCR reaction volume was 25 µl, containing 1 x universal amplification mix (Applied Biosystems, USA) and 5 µL template DNAs. 160 nM primers and 400 nM probe were used for Acc1 and ALMT1 assays, 80 nM primers and 200 nM probe were used for Waxy-D1 gene assay, and 200 nM primers and 800 nM probe were used for PKABA1 assay. All real-time PCR reactions were performed according to the program: 50°C for 2 minutes; 95°C for 10 minutes; and 45 cycles of 95°C for 15 seconds, 60°C for 60 seconds. Fluorescent signals were monitored at the extension step of 60°C in each cycle. For each sample test, each PCR reaction had 3 replicates and the experiment was repeated three times.

Evaluation of Quantitative PCR Efficiency and Ct values

To evaluate the efficiency of quantitative PCR assay, standard curves were constructed using 5 serially diluted Jimai 19 genomic DNA solutions (10.0, 2.0, 0.4, 0.08, and 0.016 ng/µL). Optimal threshold lines were automatically determined using ABI 7900 software. Efficiency of each quantitative PCR was calculated using equationE=10(1m)1, where E was the PCR efficiency, m was the slope of the log transformed DNA quantities (x) versus Ct values (y) in equation y = m * logx + b. Real-time PCR amplification of each gene was performed at three template DNA concentrations (10.0, 1.0, and 0.2 ng/µL) with same PCR program. Measured Ct values under optimal threshold of their respective standard curves were recorded for all three DNA concentrations.

Statistical Analysis

Ct values of PCR reactions using 50.0, 5.0 and 1.0 ng template DNA were calculated and analyzed for each endogenous reference gene. For each assay, Ct values from nine repeats were statistically analyzed for range, mean value, and standard deviation. The variations of each PCR assay among different lines were evaluated using GeNorm software (version 3.5, http://medgen.ugent.be/~jvdesomp/genorm/). GeNorm analysis is an algorithm widely used in evaluation and selection of suitable reference genes for gene expression studies. It has also been used in evaluating and harmonizing maize and rice endogenous reference genes and assays [11,12]. In this study, Ct values were transformed to relative quantities (Q) by ΔCt method (Q=EΔCt), where E was the PCR efficiency of an assay using random selected DNA calibrator (Jimai 19), and ΔCt was the Ct difference between each individual cultivar and the calibrator (Jimai 19) at the same DNA concentration. The relative quantities were used as input data to automatically calculate the M value, which measured the corresponding average pairwise variation of a single reference gene against all other genes [22]. M value reflects the consistency of a target gene. Higher M value indicates lower consistency of a particular gene. For the three tested DNA concentrations, Q values of each reference gene were shown in a boxplot chart showing median, interquartile range (IQR), and the value ranges; and M values were displayed in a line connected dot plot according to the order of consistency of each gene.

Results

Discovered SNPs in Real-time PCR Primer and Probe Annealing Regions of the Four Endogenous Reference Genes

The target DNA fragments of the four endogenous reference genes from all tested lines were amplified by conventional PCR and sequenced. The expected DNA fragment of Acc1 gene could be amplified from all lines except Aegilops tauschii, Aegilops speltoides, and Triticum urartu; expected DNA fragments of ALMT1 and Waxy-D1 genes were obtained from all lines except Aegilops speltoides, Triticum urartu, and Durum wheat; and expected PKABA1 gene fragment was amplified from all lines except Aegilops tauschii and Aegilops speltoides (Table S1 in File S1). These results were the same as the previously reported results about the species specificity of these four genes [1821], and failed amplifications were caused by genetic variation from all lines.

After blastN and alignment analysis of the obtained DNA sequences from all line, SNPs were discovered in the real-time PCR target regions of Acc1 gene, ALMT1 gene, and PKABA1 gene (Figure 1). For Acc1 gene, there was a SNP of C to T in durum wheat line Cannizzo, which was located at the eighth nucleotide from the 5’ end in the forward PCR primer. For ALMT1 gene, a SNP of A to C was located at the second nucleotide from the 5’ end in the forward primer in common wheat Zhongguochun, and another SNP of G to A was located in the fourth nucleotide from the 5’ end of the probe in common wheat line Zhengmai 9023. For PKABA1 gene, a SNP of G to A was found in the eleventh nucleotide from the 5’ end of the probe region in lines Zhoumai 16 and Zimai 12. No sequence variation was observed in the amplicon of Waxy-D1 gene among all tested lines.

thumbnail
Figure 1. Detected SNPs within the amplified target DNAs of Acc1, ALMT1, and PKABA1 gene.

A SNP of C>T was found at the eighth nucleotide position of Acc1 gene forward primer in durum wheat cultivar Cannizzo; A SNP of A>C was found at the second nucleotide position of ALMT1 gene forward primer in common wheat cultivar Zhongguochun, and one SNP of G>A was at the fourth nucleotide position of ALMT1 gene probe in common wheat cultivar Zhengmai 9023; A SNP of G>A was found at the eleventh nucleotide position of PKABA1 gene probe in common wheat cultivars Zhoumai 16 and Zimai 12. No SNP was found in Waxy-D1 target region.

https://doi.org/10.1371/journal.pone.0075850.g001

Real-time PCR Efficiency and Ct Values

To establish efficient real-time PCR assays of the four wheat endogenous reference genes, five serially diluted genomic DNA solutions of common wheat line Jimai 19 (10.0, 2.0, 0.4, 0.08, and 0.016 ng/µL) were used to construct a PCR standard curve. PCR efficiencies of the four gene assays were between 0.920 and 1.045, indicating that all four real-time PCR assays had acceptable exponential efficiencies that were consistent with previously reported results except for the dramatical improvement of PCR efficiency in PKABA1 assay [1821]. The linear correlation (R2) values of the four constructed PCR standard curves were all above 0.990 (Table 2), indicating these four assays were robust, with a wide dynamic range, and suitable for further quantitative analysis. Mean Ct values of each PCR assay were calculated from three repeats under identical threshold and listed in Table S2 in File S1. These values could be used as references for further evaluation of the performance of each assay.

Real-time PCR assayPrimers and probeCalibratorLinearityPCR EfficiencyStandard curve
Waxy-D1Q-Waxy-F/R, Q-Waxy-pJimai 190.9990.920y = -3.5294*log χ+ 42.152
ALMT1Q-ALMT1-F2/R, Q-ALMT1-pJimai 190.9940.826y = -3.8233*log χ + 45.699
Zhongguochun0.9960.969y = -3.3992*log χ + 43.813
Q-ALMT1-F/R, Q-ALMT1-p2Jimai 190.9960.810y = -3.8792*log χ + 47.025
Zhengmai90230.9940.950y = -3.4484*log χ + 44.96
Q-ALMT1-F/R, Q-ALMT1-pJimai 190.9990.966y = -3.4057*log χ + 42.216
Zhongguochun0.9960.893y = -3.6096*log χ + 44.114
Zhengmai90230.9970.868y = -3.6842*log χ + 44.279
PKABA1Q-PKABA1-F/R, Q-PKABA1-pJimai 190.9981.045y = -3.2178* log χ + 43.31
Zhoumai160.9971.075y = -3.1546*log χ + 44.23
Zimai120.9820.885y = -3.6328*log χ + 45.869
Q-PKABA1-F/R, Q-PKABA1-p2Jimai 190.9950.804y = -3.9012*log χ + 46.58
Zhoumai160.9970.933y = -3.4933*log χ + 44.303
Zimai120.9970.964y = -3.4108*log χ + 43.858
Acc1Q-Acc1-F/R, Q-Acc1-pJimai 190.9950.984y = -3.361*log χ + 40.1
Cannizzo wheat0.9970.808y = -3.8891*log χ + 46.62
Q-Acc1-F2/R, Q-Acc1-pJimai 190.9990.864y = -3.6971*log χ + 45.417
Cannizzo wheat0.9980.928y = -3.5079*log χ + 43.704

Table 2. PCR efficiencies and standard curve linearity of real-time PCR assays in their corresponding calibrator lines.

CSV
Download CSV

Effects of the SNPs on Real-time PCR Assays

To evaluate the effects of the discovered SNPs in Acc1, ALMT1 and PKABA1 genes on their real-time PCR assays, PCR efficiency and Ct values in lines with detectable SNPs were analyzed. The recalculated PCR efficiencies of Acc1, ALMT1, and PKABA1 genes in lines with SNPs were mostly below 0.90 and obviously lower than those in assays using lines containing no SNPs. For instance, the PCR efficiency of ALMT1 gene in Zhongguochun and Zhengmai 9023 were 0.893 and 0.868, respectively; the PCR efficiency of Acc1 gene in durum wheat Cannizzo was 0.808 (Table 2).

Ct values of PCR assays of the four genes in different lines are listed in Table 3. In ALMT1 assay, the Ct values in Zhengmai 9023 (which had one SNP in probe region) using 50 ng, 5 ng, and 1 ng template genomic DNA were 29.34, 32.89 and 35.93, respectively. These Ct values were bigger than those in lines without SNPs, where the range of Ct values for ALMT1 gene were 27.36-29.21, 30.77-32.65, and 32.72-35.74 for 50 ng, 5 ng, and 1 ng template genomic DNA, respectively. For Acc1 assay, the Ct values in Cannizzo (with a SNP of C to T in the forward primer region) were 26.29, 29.48, and 31.78 at 50 ng, 5 ng, and 1 ng template DNA, and these Ct values were also bigger than the values in lines without SNPs, whose corresponding Ct range were 22.77-24.67, 25.97-28.49, and 27.83-30.5. Similar increase of Ct values was also observed for PKABA1 assays in Zhoumai 16 and Zimai 12 which contained SNPs (Table 3).

Real-time PCR assayPrimers and probeWheat cultivarCt Values
50 ng5 ng1 ng
ALMT1Q-ALMT1-F/R and Q-ALMT1-pZhongguochun (A>C)29.4932.9535.25
Zhengmai 9023 (G>A)29.3432.8935.93
Cultivars without SNPs27.36-29.2130.77-32.6532.72-35.74
Q-ALMT1-F2/R and Q-ALMT1-pZhongguochun (A>C)27.9031.1633.22
Q-ALMT1-F/R and Q-ALMT1-p2Zhengmai 9023 (G>A)28.1431.4032.52
PKABA1Q-PKABA1-F/R and Q-PKABA1-pZhoumai 16 (A>G)30.2933.3935.40
Zimai 12 (A>G)30.2733.5435.41
Cultivars without SNPs27.66-29.9330.66-32.6632.49-34.86
Q-PKABA1-F/R and Q-PKABA1-p2Zhoumai 16 (A>G)28.6031.7034.17
Zimai 12 (A>G)28.9132.1734.64
Acc1Q-Acc1-F/R and Q-Acc1-pCannizzo wheat (C>T)26.2929.4831.78
Cultivars without SNPs22.77-24.6725.97-28.4927.83-30.5
Q-Acc1-F2/R and Q-Acc1-pCannizzo wheat (C>T)24.3927.7829.19

Table 3. Ct values of original assays and homogenous assays targeting on the SNPs in five wheat lines with SNPs.

CSV
Download CSV

Confirmation of the SNP Effects on Real-time PCR Efficiencies

In order to confirm that the increased Ct values and lowered PCR efficiencies were indeed caused by the SNPs, PCR primers (Q-ALMT1-F2 for ALMT1 assay and Q-Acc1-F2 for Acc1 assay) and probes (Q-ALMT1-F2 for ALMT1 assay, Q-PKABA1-p2 for PKABA1 assay) that had matched nucleotides at the SNP sites were designed and used to test for Ct values and PCR efficiencies. PCR efficiencies calculated from reconstructed standard curves of Acc1, ALMT1, and PKABA1 genes using new primers or probes are listed in Table 2. The new PCR efficiencies were 0.928 for Acc1 in Cannizzo, 0.969 for ALMT1 in Zhongguochun, 0.950 for ALMT1 in Zhengmai 9023, 0.933 for PKABA1 in Zhoumai16, and 0.964 for PKABA1 in Zimai12. All PCR efficiencies were obviously increased from those in real-time PCR assays using primers and probes containing unmatched SNPs (where PCR efficiencies were mostly below 0.90). On the contrary, when the new primer/probe sets adjusted for SNPs in some lines were used for Jimai 19, where the primer/probe had mismatched SNPs, the PCR efficiencies decreased dramatically (Table 2), indicating that the mismatched SNPs caused the decrease of real-time PCR efficiency.

Ct values of the new homogenous PCR assays were also calculated using the same threshold (Table 3). For ALMT1 assay in Zhongguochun using Q-ALMT1-F2/R and Q-ALMT1-p, the Ct values were 27.90, 31.66, and 33.22 for 50 ng, 5 ng, and 1 ng template DNA. For ALMT1 assay using Q-ALMT1-F/R and Q-ALMT1-p2, the Ct values were 28.14, 31.40, and 32.52 for 50 ng, 5 ng, and 1 ng genomic DNA in Zhengmai 9023. Similarly lowered Ct values were also observed for the other two genes containing SNPs (Table 2). The results demonstrated that real-time PCR Ct values could be lowered into normal ranges when SNPs were eliminated from the primer/probe region, indicating the direct effect of SNPs on PCR amplified efficiency.

Comparison of the Real-time PCR Performance among the Four Endogenous Reference Genes

To evaluate the performance of the four real-time PCR assays, 37 hexaploid wheat lines and 2 octoploid triticale lines were tested in this study. Ct values obtained from all 39 lines with the same threshold were used for statistical analysis and comparison (Table S2 in File S1). The variability of each real-time PCR assay was first evaluated using Ct values across all 39 tested lines. The assay with the lowest variation (judged by standard deviation, SD) of Ct values was the assay for Waxy-D1 gene (0.40, 0.42, and 0.43 using three different amount of template genomic DNAs). The PKABA1 assay had the largest Ct value variation in all 39 lines (0.70, 0.70, and 0.68 using three different amount of template genomic DNAs) (Table 4). To evaluate the consistency of each of the four endogenous reference gene real-time PCR assays, Ct values were converted to relative quantitative values. Boxplot charts were constructed to show median and interquartile range of the relative quantitative values (Figure 2). Based on the chart, the ALMT1 and Waxy-D1 gene assays appeared relatively consistent among all tested lines.

Gene name50 ng5 ng1 ng
Ct rangeCt minCt maxMean Ct±SDCt rangeCt minCt maxMean Ct±SDCt rangeCt minCt maxMean Ct±SD
ALMT12.1327.3629.4928.49±0.522.1830.7732.9531.84±0.543.2132.7235.9334.42±0.67
PKABA12.6327.6630.2928.68±0.702.8830.6633.5431.65±0.702.9232.4935.4134.00±0.68
Acc11.9022.7724.6723.84±0.472.5225.9728.4927.14±0.552.6727.8330.529.26±0.59
Waxy-D11.6829.0130.6929.68±0.401.8632.2734.1333.70±0.421.8234.6436.4635.36±0.43

Table 4. Ct values of the four common wheat reference genes in all wheat lines.

CSV
Download CSV
thumbnail
Figure 2. Quantity of the reference genes in three sets of assays as compared with the Jimai 19 calibrator (Q = EΔCt).

Boxplot shows median, interquartile and range.

https://doi.org/10.1371/journal.pone.0075850.g002

To further evaluate the consistency of each real-time PCR assay, GeNorm analysis was performed using GeNorm software (version 3.5). M values of each reference gene in all lines were calculated. The highest M values were observed for PKABA1 gene (0.884, 0.890, and 0.887 for 50 ng, 5 ng, and 1 ng genomic DNA, respectively), while Waxy-D1 gene had the lowest M values (0.629, 0.723, and 0.711 for 50 ng, 5 ng, and 1 ng genomic DNA, respectively) (Figure 3). In further analyses excluding the four lines with SNPs (Zhongguochun, Zhengmai 9023, Zhoumai 16, and Zimai 12), Waxy-D1 assay still had the lowest M values (0.623, 0.685, and 0.674 for 50 ng, 5 ng, and 1 ng genomic DNA, respectively) (Figure 3). Overall evaluation using calculated M values in GeNorm analysis suggested that the Waxy-D1 real-time PCR assay was more consistent than the other assays and had the best PCR performance among all four assays.

thumbnail
Figure 3. GeNorm analysis of four wheat endogenous reference genes assays among common wheat cultivars.

Solid lines were results of analysis using all 39 cultivars, and dashed lines were results of analysis excluding the four cultivars with SNPs (Zhongguochun, Zhengmai 9023, Zhoumai 16, and Zimai 12).

https://doi.org/10.1371/journal.pone.0075850.g003

Discussion

In order to develop GM wheat detection methods, including selection of proper wheat endogenous reference genes and their assays, researchers have reported several wheat endogenous reference genes and assays for GM wheat analysis in the past few years. However, there are still difficulties in selecting suitable endogenous reference genes for GM wheat analysis because of the complicated genotypes and karyotypes among wheat species (Triticum L genus). Based on qualitative PCR amplification results, amplicons of Acc1 gene and PKABA1 gene were not only observed in all hexaploid common wheat lines, but also in Aegilops speltoides Tausch (BB), Triticum urartu L (AA), and durum wheat (BBAA), indicating that Acc1 and PKABA1 assays had low specificity for common wheat. ALMT1 and Waxy-D1 gene amplicons could not be amplified from Aegilops speltoides Tausch (BB), Triticum urartu L (AA), and durum wheat (BBAA), showing that these two genes were more suitable for identifying common wheat species. The results also showed the ACC1 gene came from B genome, PKABA1 from A genome, and the Waxy-D1 and ALMT1 genes from the D genome of wheat. Therefore, it was impossible to specifically identify the Chinese octoploid triticale from common wheat lines employing any gene from haploid A, B and D genome. These results demonstrated that the genotype and karyotype situations should be considered in selecting new endogenous reference genes for any given species.

The gene allele stability among different lines is another important parameter in validating and harmonizing the endogenous reference gene assays for GMO detection. It can be evaluated by target DNA sequencing/alignment and PCR performance analysis. Sequence alignment of target DNA from different wheat lines in this study revealed SNPs in several lines, such as one SNP of C to T in Acc1 gene, two SNPs of A to C and G to A in ALMT1 gene, and one SNP of G to A in PKABA1 gene. It has been reported previously that SNPs in real-time PCR primer/probe annealing regions could affect PCR efficiency in quantification of GM contents, virus, and microorganisms [2327]. In our study, increased Ct values (0.77<ΔCt<3.41) and decreased PCR efficiencies (0.12<ΔE<0.24) were observed in Acc1, ALMT1, and PKABA1 assays when lines with SNPs in primer or probe region were tested. The SNPs effects on PCR efficiency were also confirmed by new real-time PCR assays employing re-designed homogenous primers or probes eliminating SNPs. These results demonstrated the SNPs could decrease the real-time PCR efficiency, which would result in an over-estimation of GM content in a GMO assay. Although previous reports have suggested that Ct variation for an endogenous reference gene assay among different lines might come from different qualities of extracted DNA, errors associated with DNA dilution, inaccurate measurement of DNA concentrations, or the calculation error between the copy number and DNA quantity because of the complex haploid genome size [19,20], our study indicated that the Ct differences of the four wheat endogenous reference gene assays among different lines could originate from different PCR performances. A collaborative ring trial validation of the selected real-time PCR assays might further improve the reliability and applicability of our results by considering the robustness of the assays in different laboratories, equipment, and brands of mastermix.

Based on the combined results of high species specificity, no SNPs in target sequence, and high PCR performance among different wheat lines, we concluded that Waxy-D1 gene was the most suitable candidate among the four reported wheat endogenous reference genes, and its real-time PCR assay would be very useful in establishing accurate and creditable quantitative PCR assay for GM wheat.

Supporting Information

File S1.

Supporting tables.

Table S1, List of the 43 seed samples of Triticum genus and the specificity test results of the four endogenous reference genes. Table S2, Ct values of the 39 common wheat cultivars and 1 durum wheat cultivar from the four endogenous reference gene assays.

https://doi.org/10.1371/journal.pone.0075850.s001

(DOCX)

Author Contributions

Conceived and designed the experiments: DZ LY. Performed the experiments: HH RW. Analyzed the data: FC HH LY. Contributed reagents/materials/analysis tools: FC. Wrote the manuscript: HH LY. Read and gave suggestions on the manuscript: HH FC RW DZ LY.

References

  1. 1. Regulation (EC) No 1829/2003 of the European Parliament and of the Council of 22 September 2003 on genetically modified food and feed (2003). Official Journal Eur Union 46: 268.
  2. 2. Notification 31 Ministry of Agriculture and Forestry of Korea, Seoul. (2000) Korea.
  3. 3. Order 10, Ministry of Agricultureof the People's Republic of China. (2002) China.
  4. 4. Querci M, Van den Bulcke M, Zel J, Van den Eede G, Broll H (2010) New approaches in GMO detection. Anal Bioanal Chem 396: 1991-2002. doi:10.1007/s00216-009-3237-3. PubMed: 19876618.
  5. 5. Elenis DS, Kalogianni DP, Glynou K, Ioannou PC, Christopoulos TK (2008) Advances in molecular techniques for the detection and quantification of genetically modified organisms. Anal Bioanal Chem 392: 347-354. doi:10.1007/s00216-008-1868-4. PubMed: 18239909.
  6. 6. Zhang M, Liu Y, Chen L, Quan S, Jiang S et al. (2013) One simple dna extraction device and its combination with modified visual loop-mediated isothermal amplification for rapid on-field detection of genetically modified organisms. Anal Chem 85: 75-82. doi:10.1021/ac301640p. PubMed: 23181490.
  7. 7. Guo J, Yang L, Chen L, Morisset D, Li X et al. (2011) MPIC: a high-throughput analytical method for multiple DNA targets. Anal Chem 83: 1579-1586. doi:10.1021/ac103266w. PubMed: 21291179.
  8. 8. Zhang D, Guo J (2011) The development and standardization of testing methods for genetically modified organisms and their derived products. J Integr Plant Biol 53: 539-551. doi:10.1111/j.1744-7909.2011.01060.x. PubMed: 21651724.
  9. 9. Ahmed FE (2002) Detection of genetically modified organisms in foods. Trends Biotechnol 20: 215-223. doi:10.1016/S0167-7799(01)01920-5. PubMed: 11943377.
  10. 10. Hernández M, Río A, Esteve T, Prat S, Pla M (2001) A rapeseed-specific gene, acetyl-CoA carboxylase, can be used as a reference for qualitative and real-time quantitative PCR detection of transgenes from mixed food samples. J Agric Food Chem 49: 3622-3627. doi:10.1021/jf010173n. PubMed: 11513638.
  11. 11. Wang C, Jiang L, Rao J, Liu Y, Yang L (2010) Evaluation of four genes in rice for their suitability as endogenous reference standards in quantitative PCR. J Agric Food Chem 58: 11543-11547. doi:10.1021/jf102092c. PubMed: 20961039.
  12. 12. Papazova N, Zhang D, Gruden K, Vojvoda J, Yang L et al. (2010) Evaluation of the reliability of maize reference assays for GMO quantification. Anal Bioanal Chem 396: 2189-2201. doi:10.1007/s00216-009-3386-4. PubMed: 20069284.
  13. 13. Xia L, Ma Y, He Y, Jones HD (2012) GM wheat development in China: current status and challenges to commercialization. J Exp Bot 63: 1785-1790. doi:10.1093/jxb/err342. PubMed: 22174439.
  14. 14. Bhalla PL (2006) Genetic engineering of wheat–current challenges and opportunities. Trends Biotechnol 24: 305-311. doi:10.1016/j.tibtech.2006.04.008. PubMed: 16682090.
  15. 15. Jones HD (2005) Wheat transformation: current technology and applications to grain development and composition. J Cereal Sci 41: 137-147. doi:10.1016/j.jcs.2004.08.009.
  16. 16. Obert C J-Ridley P, W-Schneider WR, Riordan GS, Nemeth AM, et al. (2004) The composition of grain and forage from glyphosate tolerant wheat MON7180 is equivalent to that of conventional wheat (Triticum aestivum L.). J Agric Food Chem 52: 1375–1384.
  17. 17. Anand A, Zhou T, Trick HN, Gill BS, Bockus WW et al. (2003) Muthukrishnan, S. Greenhouse and field testing of transgenic wheat plants stably expressing genes for thaumatin-like protein, chitinase and glucanase against Fusarium graminearum. J Exp Bot 54: 1101-1111. doi:10.1093/jxb/erg110. PubMed: 12598580.
  18. 18. Hernández M, Esteve T, Pla M (2005) Real-time polymerase chain reaction based assays for quantitative detection of barley, rice, sunflower, and wheat. J Agric Food Chem 53: 7003-7009. doi:10.1021/jf050797j. PubMed: 16131102.
  19. 19. Rønning SB, Berdal KG, Andersen CB, Holst-Jensen A (2006) Novel reference gene, PKABA1, used in a duplex real-time polymerase chain reaction for detection and quantitation of wheat- and barley-derived DNA. J Agric Food Chem 54: 682-687. doi:10.1021/jf052328n. PubMed: 16448168.
  20. 20. Vautrin S, Zhang D (2007) Real-time polymerase chain reaction assay for endogenous reference gene for specific detection and quantification of common wheat-derived DNA (Triticum aestivum L.). J AOAC Int 90: 794-801.
  21. 21. Iida M, Yamashiro S, Yamakawa H, Hayakawa K, Kuribara H et al. (2005) Development of taxon-specific sequences of common wheat for the detection of genetically modified wheat. J Agric Food Chem 53: 6294-6300. doi:10.1021/jf0505731. PubMed: 16076109.
  22. 22. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N et al. (2002) Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol 3: research0034. PubMed: 12184808.
  23. 23. Ghedira R, Papazova N, Vuylsteke M, Ruttink T, Taverniers I et al. (2009) Assessment of primer/template mismatch effects on real-time PCR amplification of target taxa for GMO quantification. J Agric Food Chem 57: 9370-9377. doi:10.1021/jf901976a. PubMed: 19778057.
  24. 24. Stadhouders R, Pas SD, Anber J, Voermans J, Mes THM et al. (2010) The effect of primer-template mismatches on the detection and quantification of nucleic acids using the 5′nuclease assay. J Mol Diagn 12: 109-117. doi:10.2353/jmoldx.2010.090035. PubMed: 19948821.
  25. 25. Bru D, Martin-Laurent F, Philippot L (2008) Quantification of the detrimental effect of a single primer-template mismatch by real-time PCR using the 16S rRNA gene as an example. Appl Environ Microbiol 74: 1660-1663. doi:10.1128/AEM.02403-07. PubMed: 18192413.
  26. 26. Klungthong C, Chinnawirotpisan P, Hussem K, Phonpakobsin T, Manasatienkij W et al. (2010) The impact of primer and probe-template mismatches on the sensitivity of pandemic influenza A/H1N1/2009 virus detection by real-time RT-PCR. J Clin Virol 48: 91-95. doi:10.1016/j.jcv.2010.03.012. PubMed: 20413345.
  27. 27. Broothaerts W, Corbisier P, Schimmel H, Trapmann S, Vincent S et al. (2008) A single nucleotide polymorphism (SNP839) in the adh1 reference gene affects the quantitation of genetically modified maize (Zea mays L.). J Agric Food Chem 56: 8825-8831.