As the worldwide commercialization of genetically modified organisms (GMOs) increases and consumers concern the safety of GMOs, many countries and regions are issuing labeling regulations on GMOs and their products. Analytical methods and their standardization for GM ingredients in foods and feed are essential for the implementation of labeling regulations. To date, the GMO testing methods are mainly based on the inserted DNA sequences and newly produced proteins in GMOs. This paper presents an overview of GMO testing methods as well as their standardization.
Since the first commercial genetically modified (GM) plant (the FlavrSavr tomato) was approved for marketing in 1994, recombinant DNA technology has been widely used in modern agriculture. So far, 184 GM events have been authorized for food and feed production in 59 countries, and the planting region of GM crops reached 148 million hectares in 2010 (James 2011). However, owing to the ever-increasing global controversial issues on food safety, environmental risk and ethical concerns (Dlugosch and Whitton 2008), more and more countries and regions have required the labeling of food products and ingredients containing or derived from genetically modified organisms (GMOs) (Ruttink et al. 2010). In addition, with increasing international trade of food and feed, international harmonization of the detection methods of GMO analysis is necessary. Thus the development of reliable testing methods and their standardization for GMO detection, identification, traceability and quantification is a key step in GMO development and commercialization. Current analytical methods are mainly carried out by either detecting the transgenic DNA or the foreign protein(s) produced in GMOs using polymerase chain reaction (PCR), molecular hybridization, microarrays, biosensors, and sequencing methods, etc. (Holst-Jensen 2009). Some novel, high-throughput DNA amplification or PCR-free methods have also been reported (Morisset et al. 2008; Guan et al. 2010; Li et al. 2010; Guo et al. 2011).
Labeling Regulations to GM Foods and Feed
The first labeling regulation (EU Regulation 258/97) of GMOs and GM products was introduced by the European Union (EU) for the consumers’ right to know the information of GM ingredients in foods in 1997. Since then, more than 40 countries and regions have introduced the regulations for tracing and/or labeling GM products. While the majority of these countries and regions are the members of the Organization for Economic Cooperation and Development (OECD), few developing countries have implemented labeling regulations (Gruère and Rao SR 2007).
The established labeling regulations can be classified into two categories: voluntary (e.g., Canada, Hong Kong, and South Africa) and mandatory (e.g., Australia, the EU, Japan, Brazil, and China) (Gruère and Rao 2007). Among the countries with mandatory labeling, there are many different aspects among their regulations (Table 1). In China, the mandatory labeling regulation set zero tolerance as the threshold level of GM ingredients; while in Australia, European Union and Japan, their mandatory labeling regulations have the threshold levels of GM ingredients to certain ranges (from 0.9% to 5%). Furthermore, some countries (e.g., Korea and Japan) require the threshold levels for three or five individual ingredients in foods. In Japan, the threshold is based on the mass fraction as measurement unit when quantifying GMOs, whereas current EU (e.g., EC 2004/787) and some other countries recommend using “the percentage of GM target DNA copy numbers per corresponding target taxon-specific DNA copy numbers calculated in terms of haploid genomes” as the measurement unit (Marmiroli et al. 2008; Trapmann et al. 2010).
Table 1. Status of the rules for labeling of genetically modified (GM) foods
The analysis procedure for GMOs and derived products consist of sampling, sample preparation, extraction of DNA and protein, examination, and result interpretation (Figure 1). The DNA-based method is relatively stable compared with the one formed on the basis of protein analysis, owing to the DNA retainable in the processed samples. At present, the DNA-based methods involve mainly molecular hybridization (e.g., Southern blot), PCR, and microarray. Among the methods, PCR is in fact the most commonly used tool for GMO detection and traceability, due to its rapid and relatively low-cost detection procedures. PCR can be carried out with qualitative and quantitative methods, including singleplex PCR, multiplex PCR, nested PCR, competitive PCR, and real-time quantitative PCR etc. (Holst-Jensen et al. 2003; Michelini et al. 2008; Querci et al. 2010). The targets for PCR-based GMO tests can be grouped into at least four categories corresponding to their levels of specificity: screening-specific sequences (e.g., p35S and tNOS) (Matsuoka et al. 2002; Dörries et al. 2010), gene-specific sequences (e.g., Cry1Ab and CP4-EPSPS) (Grohmann et al. 2009; Randhawa et al. 2009), construct-specific sequences (Waiblinger et al. 2007; Shrestha et al. 2008), event-specific sequences (Yang et al. 2008a; Guo et al. 2009a; Liu et al. 2009; Oguchi et al. 2010). An event-specific PCR detection strategy is based on the unique and specific integration flanking sequences between the host plant genome DNA and the inserted gene (Holst-Jensen 2003). Some event-specific PCRs have been accredited or validated for the nucleic acid detection by official bodies/reference laboratories, and are currently the most widely used DNA detection methods for determining the presence of GMOs content in processed food and feed samples (Dong et al. 2008).
Endogenous reference gene
According to the principle of GMO quantification, one endogenous reference gene should have three typical characters, i.e. species specificity, no allelic variation among various cultivars and low or stable copy number in haploid genome (Ding et al. 2004; Chaouachi et al. 2007). In real-time quantitative PCR analysis, initial amounts of GM and non-GM DNA templates are quantified by standard curves, and GM contents (%) can be calculated by the ratios of specific GMOs target sequence to species-specific endogenous reference gene sequence (Ding et al. 2004; Yang et al. 2005a; Guo et al. 2009b). Consequently, validation of an appropriate endogenous reference gene for each GM plant is necessary for GMO analysis as well as identification of plant ingredients in the mixed samples. So far, great efforts have been made in obtaining reference genes of different crops for detection of GMOs, as shown in Table 2.
Table 2. Endogenous reference genes used in genetically modified organism (GMO) detection
Certified reference material (CRM) includes values of the materials and techniques used, and the choice of CRMs for construction of calibration curves is very important in GMO quantification (ISO 24276: 2006). Generally, the labeling threshold in most countries (usually from 0.9 to 5%) for GMO events in legislation has been widely interpreted as being the mass/mass or DNA copy ratio, but the available methods for GMO quantization mainly depend on the copy number ratios, which should be converted to mass/mass using CRMs (Rodríguez-Lázaro et al. 2007). The availability of suitable CRMs is a fundamental requirement for the validation of GMO detection methods, and can be used as the calibrant and quality control material in PCR amplification (Trapmann et al. 2010). For such purposes, some countries and regions such as EU, USA, China, Japan etc. strengthen the research and applicability of CRMs. The JRC Institute for Reference Materials and Measurements (IRMM, Geel, Belgium) is an institute responsible for the production of appropriate CRMs in Europe.
However, current CRMs have some limitations in practical application, such as the inadequate availability of GM crop materials, limited quantitative ranges, and high costs (Li et al. 2009). The reference molecules, usually the plasmids containing one or more functional fragments suitable for GM crop event detection, have been developed and proved to be a good alternative for CRM in GMO detection (Yang et al. 2007; Zhang et al. 2008). So far, over 20 reference molecules have been reported and used in the detection of several GM crops, for instance MON159855 and MON88913 cotton, GTS 40–3-2 soybean, GT73 canola, MON863, GA21, TC1507, T25, MON810, NK603, Bt176, E-3272, and 59122 maize (Taverniers et al. 2005; Burns et al. 2006; Lee et al. 2007; Yang et al. 2007; Zhang et al. 2008; Li et al. 2009; Shen et al. 2010) etc.
Qualitative PCR-based detection
To detect GMOs, qualitative PCR technologies are mainly used to screen and/or identify single or multiple transgenic DNA fragments, including singleplex PCR, multiplex PCR, and nested PCR etc. In multiplex PCR, several primer pairs are included to permit the simultaneous detection of multiple target sequences. Generally, the qualitative PCR products are distinguished by size in agarose gel electrophoresis. Randhawa et al. (2009) developed a multiplex PCR assay with six marker/reporter genes (aadA, bar, hpt, nptII, pat, and uidA) that have already been introduced into GMOs. This assay could be immensely used to test unintentional mixing of GM seeds with non-GM seed lots. Onishi et al. (2005) and Shrestha et al. (2008) reported respectively applications of multiplex PCR assay for simultaneous detection of up to eight events of GM maize and endogenous reference gene (zSSIIb or zein) within a single reaction. The limits of detection (LODs) were approximately 0.25% in both assays.
However, the sensitivity and resolution are limited in agarose gel electrophoresis: the requirement of amplicons with apparent differences in size and the longer separation time. Recently, the combination of PCR and capillary gel electrophoresis (CGE) was developed for simultaneous detection of multiple DNA targets. The CGE assay accomplishes higher resolutions compared with agarose gel electrophoresis, and the sample separation time in the assay is shorter. Additionally, this assay has sensitivity and the reproducibility similar to real-time PCR (RT-PCR) (Nadal et al. 2006, 2009), and might be the most prospective analytical method for GMO detection. Nadal et al. (2009) presented the development of the CGE technology for the simultaneous detection of up to eight amplicons. The application offers an alternative tool for routine GMO identification. Recently, Guo et al. (2011) developed a robust high-throughput analytical approach named multiplex microdroplet PCR implemented capillary gel electrophoresis (MPIC) (Figure 2). This assay combines the advantages of bipartite primers, microdroplet PCR and CGE for multiple target DNA analysis, and at least 24 different targets can be simultaneously detected and identified. Furthermore, the microarray is a reliable method for amplicon analysis, but it is also time-consuming and costly (Leimanis et al. 2006; Xu et al. 2007). Another alternative is application of combinations of screening methods and matrix table (Table 3), forming the basis of decision tools to conclude which GMP is present in a sample (Holst-Jensen 2009, Querci et al. 2010). The matrix method has been adopted by many laboratories as part of their general GMO screening strategy.
Table 3. Matrix description of the authorized genetically modified (GM) crops in China (2009)
Generally, the purpose of GMO quantification is to calculate the fraction of a certain species that comes from GM materials relying on quantitative PCR (Buh Gasparic et al. 2010). In the quantitative PCR assay, the number of initial template molecules can be calculated based on the amount of the products through the standard curves. There are two specific targets needed in GMO quantification: reference gene sequence and exogenous gene sequence. The total number of taxon-specific haploid genomes and GM-specific haploid genomes that are present in the sample were estimated, respectively. The early quantitative PCR tests were based on double competitive PCR (DC-PCR), but quantitative real-time PCR (qRT-PCR), representing the most powerful current means of quantifying GM ingredients, is the most widely used method for GMO detection (Buh Gasparic et al. 2008).
Real time-PCR allows for the real-time monitoring of the amplification reaction through fluorescence signal corresponding to increased amounts of amplification products at each reaction cycle. Because of its ease of use, high throughput ability, decreased post-PCR manipulation, and lack of cross-contamination of PCR amplicons, the RT-PCR method is becoming the new gold standard method for nucleic acid quantification (Yang et al. 2008b). Currently, a number of RT-PCR fluorogenic signal reagents have been developed and applied for quantitative purposes (Buh Gasparic et al. 2008, 2010), for instance sequence unspecific DNA-binding dyes (e.g., SYBR Green I) (Hernandez et al. 2003b), fluorescence resonance energy transfer (FRET) probes (Wittwer et al. 1997), TaqMan/MGB probes (Terry et al. 2002), LNA (locked nucleic acid) probe (Salvi et al. 2008), Plexor technology (Buh Gasparic et al. 2008), light upon extension (LUX) probe (Nazarenko et al. 2002), molecular beacons (Andersen et al. 2006) and their derivatives (Amplifluor, Sunrise, and scorpion primers) (Whitcombe et al. 1999; Thelwell et al. 2000; Li et al. 2002), and universal template (UT) probe (Zhang et al. 2003) etc. Among them, TaqMan/MGB probes and SYBR green I are the most commonly used RT-PCR chemistries. With the development of detection technology, multiplex TaqMan quantitative PCRs have been recently reported (Bahrdt et al. 2010). Yang et al (2008b) reported a novel set of fluorescent signal devices named Attached Universal Duplex Probes (AUDP), that can not only be used for different target DNA sequences in single PCR assays, but also in duplex PCR assays with higher fluorescent intensity. Moreover, amplified target DNA fragments as long as 1.5 kb can be detected with high efficiency.
Microarrays, also called “DNA chips”, have the advantages of automation, miniaturization and high-throughput. The microarray consists of glass supports containing thousands of specific oligonucleotide capture probes being spotted in array format to their surface. Subsequently, the detection of the target(s) of interest (DNA or RNA) labeled with a fluorescent marker is performed by direct hybridization (Miraglia et al. 2004; Querci et al. 2010). Sometimes, in order to increase the sensitivity of GMO detection and quantification, an amplification step of the DNA targets is also applied prior to hybridization on microarrays (Morisset et al. 2008). Ultimately, the microarray is scanned for individual fluorescence intensity of each spot by computer and the resultant data are analyzed.
Several assays with microarrays combined with multiplex PCR methods have been reported for detection of GM maize, canola, cotton and soybean events (Leimanis et al. 2006; Xu et al. 2007; Kim et al. 2010) by using fluorescent probes. The strategy of padlock probe ligation-microarray detection of multiple (non-authorized) GMOs was reported (Prins et al. 2008). Prins et al. (2008) demonstrated that the method was suitable for large-scale detection of GMOs in real-life samples. Morisset et al. (2008) developed a novel multiplex quantitative DNA-based target amplification method suitable for use in combination with microarray detection (NAIMA). This fast and simple integrated method allows sensitive, specific and fully quantitative on-chip GMO detection in a multiplex format. Although the microarray analytical approach is relatively expensive, it is also one of the most promising discrimination platforms at present for GMO detection owing to its flexibility and automatic and high-throughput capability.
Biosensor technology has been applied in a variety of molecular reactions including protein- and DNA-based testing, showing the advantages of simplicity, speed and cost (Elenis et al. 2008; Holst-Jensen 2009). For DNA biosensor technology, the single-stranded specific oligonucleotide probes (recognition layer) used as capture probes attached to the surface of the sensor is the most widely used. Other DNA biosensor methods based on optical, electrochemical, and piezoelectric transduction have been reported for the detection of amplified or non-amplified GMO-related sequences (Stobiecka et al. 2007; Ahmed et al. 2009; Bai et al. 2010).
Surface plasmon resonance (SPR) is the typical optical biosensors-based method. Gambari and Feriotto (2006) reported the SPR-based assay for rapid (about 40 min), easy, and automatable analysis of GM Roundup Ready soybean in foods. The hybridization of the target DNA with capture probe attached to the surface results in a change in the refractive index of the solution near the surface, and shows a linear relationship to the mass of target DNA hybridized. Enzyme-based electrochemical sensors were developed using disposable oligonucleotide-modified screen-printed gold electrodes. A probe carrying an -SH group at the 5’-end is attached to the gold surface. After the PCR product being denatured is hybridized with a biotinylated probe in solution, the solution is then pipetted onto the electrode and allowed to hybridize with the immobilized probe. The sensor is washed and a streptavidin-alkaline phosphatase conjugate is added. The enzyme catalyzes the hydrolysis of naphthyl phosphate substrate to the electro active naphthol, which is then detected by differential pulse voltammetry (Lucarelli et al. 2005; Elenis et al. 2008). In piezoelectric biosensors assays such as Quartz crystal microbalance (QCM) sensors (Stobiecka et al. 2007), the increase in mass due to hybridization causes a decrease in the resonance frequency, thereby allowing target detection.
Foreign proteins produced in GMOs can be detected by application of immunological and physicochemical techniques (Holst-Jensen 2009). The most common protein based assays are immunoassays: the target proteins are detected by specific monoclonal or polyclonal antibodies followed by immunochemical analysis (Michelini et al. 2008). The Lateral flow devices (LFD) and plate-based enzyme-linked immunosorbent assays (ELISA) are the most widely used methods (Fantozzi et al. 2007; Shim et al. 2007). Multiplex protein detection using immunological methods can be also achieved using microarray formats or flow cytometry by colored beads coated with the antibodies (Fantozzi et al. 2007; Ling et al. 2007). In order to increase the sensitivity of immunoassays, Allen et al. (2006) developed a novel immunoassay method in combination with PCR amplification. A sandwich enzyme-linked immunosorbent assay (S-ELISA) method for the pat and bar genes in GM pepper was developed, showing a detection limit of 0.01 μg/mL in real samples examination (Shim et al. 2007). Other alternative protein-based methods include the use of immunomagnetic electrochemical sensors, 2-dimensional gel electrophoresis, and mass spectrometry (Kim et al. 2006; Volpe et al. 2006; Ocana et al. 2007). However, protein-based methods are not suitable for processed foods because of denaturation occurring during processing. In addition, the higher costs for developing specific antibodies and the fact that antibodies cannot be synthesized simply in comparison to oligonucleotides limits the method's evolution.
Standardization of Testing Methods for GMOs
To date, up to 184 kinds of events have been developed and approved for application in 59 countries worldwide (James 2011), and many countries have issued GMO labeling and traceability policies. With the quick development of economic globalization and international trade, much effort was taken to develop standard methods for GMO detection to reduce national and international trade disputes. Some organizations, such as International Organization for Standardization (ISO), the Community Reference Laboratory for GM Food and Feed (CRL-GMFF), and Standardization Committee for Agricultural GMOs of China have made significant efforts to initiate the validation and standardization of GMO testing methods. ISO provides some international standard for GMO sampling, DNA extraction, and PCR detection (ISO 21569: 2005; ISO 21570: 2005; ISO 21571: 2005; ISO/TS 21098: 2005; ISO 24276: 2006). In the EU, CRL-GMFF and others have also organized some deliberate collaborative ring trials for p35S and tNOS quantitative detection methods (Feinberg et al. 2005; Fernandez et al. 2005; Waiblinger et al. 2007) and event-specific quantitative detection methods for TC1507, MON863, GA21, and MON810 maize events (http://gmo-crl.jrc.ec.europa.eu/).
In China, significant attention has been devoted to standardization of GMO testing methods. The Standardization Committee for Agricultural GMOs is engaged in large numbers of research projects and the development of GM molecular characterization, analytical methods, and CRM etc. Several endogenous reference genes including rice SPS, tamato Lat52, (Ding et al. 2004; Weng et al. 2005; Yang et al. 2005a, 2005b; Guo et al. 2009a) suitable for GMO detection and event-specific qualitative and quantitative PCR detection methods were validated (http://gmdd.shgmo.org/), and about 40 technical standards were developed (http://www.stee.agri.gov.cn/biosafety/). Several international collaborative validations for GM detection methods were also organized for the international harmonization of the detection. For instance Pan et al. (2007) reported the results of a collaborative ring trial for RT73 event-specific detection method, Jiang et al. (2009) and Yang et al. (2008c) respectively organized the international collaborative validations for the SPS and LAT52 gene. These should greatly assist the international harmonization of GMO identification and quantification.
Detection of Unauthorized GMOs
With the increase of more developed GMOs and international trade, some unauthorized GMOs have been monitored in the market during recent years (Dorey et al. 2000; Vermij et al. 2006; Holst-Jensen et al. 2008; Cao et al. 2009). Unauthorized GMOs are divided into two classes: (i) GMOs authorized for commercialization in some countries, but not in other countries (also called asynchronous approval); and (ii) GMOs that are not (yet) authorized in any countries or regions (Ruttink et al. 2010). The unauthorized GMOs have been shown to greatly affect domestic supplies, international trade, reduce the trust in industry and authorities, or pose significant risks to human and animal health and the environment (Holst-Jensen et al. 2008, 2009; Michelini et al. 2008). Owing to the lack of available information about the molecular inserts in unauthorized GMOs, the effective way to inspect unauthorized GMOs is a big challenge for detection laboratories.
Owing to the increasing development of approved GM crop events and more and more GM-derived products being introduced into the market, the detection of GMOs faces increasing challenges. Additionally, hundreds of GMOs detection methods have been developed, and the number is continuously increasing. Rapidly and accurately obtaining the information of GMO background, such as correct inserted gene sequences, validated methods, and reference materials is a fundamental prerequisite for setting up an effective strategy. The development of a GMO database is extremely important. Accordingly, several databases related to GMO safety and risk assessment, application, development, and labeling and regulation etc., have been established. These databases include GMO compass (http://www.gmo-compass.org/eng/home/), Agbios GM Crop Database (http://www.cera-gmc.org/?action=gm_crop_database), Living Modified Organism (LMO) Registry (http://bch.cbd.int/database/lmo-registry/), and Agbioforum (http://www.agbioforum.org/) etc. Recently, Dong et al. (2008) reported the development of a database for GMO detection methods (http://gmdd.shgmo.org/). In this database, almost all the previous developed GMOs detection methods were collected, thus providing a user-friendly search service for GMOs by event name, gene, and protein information, etc. In particular, the database supplies sequence information of exogenous inserts, if available, as well as endogenous reference genes, and standard reference materials for GMOs analysis. Furthermore, registered users can submit new GMO detection methods or sequences to this database, which makes this database open. These databases will certainly be a useful tool for method developers, detection laboratories, and regulatory officers from both industry and governments.
With the development of modern biotechnology, numerous GMOs have been approved for commercial production. Specially, some unauthorized or unknown GMOs are possibly present in the market owing to unintentional release of seed lots with unauthorized GMOs. Demands for testing GMO foods and development of reliable GMO analytical methods have been dramatically increasing. For this reason, many countries especially those in EU have recently issued restricted safety rules for import of GM foods. Various protein- and nucleic acid-based analytical methods have been developed and collected in various databases (http://gmdd.shgmo.org/). In addition, current testing methods also need improvements in their cost, in-field application and specificity and ability to quantify the commercial GMOs. The development of faster, cheaper analytical methods allowing for high-throughput, miniaturization, automation, and quantization will be the future trend. On the another hand, owing to the differences in labeling regulations among different countries, the standardization, exchange of information, and international cooperation on GMO analytical methods will be also extremely important: not only will it facilitate monitoring GMOs, but also reduce possible disputes for global trade.
(Co-Editor: Weicai Yang)
This work was supported by the National Transgenic Plant Special Fund. This work was also supported by the National Special Project of Transgenic Organisms (2008ZX8012-002).