You are here: Agriculture > Genfood and Food > Food from GMO > Methods for identifying > Genfood analyses

3.2 Various nucleotide-based amplification methods and their applicability

Most of the methods mentioned in this section have generally not yet been used widely for the identification of genetically engineered food or food stuffs. This survey, therefore, very much restricts itself to survey review articles that may simplify access to additional readings. Some of the techniques may, under certain circumstances, be appropriate for food analyses. Ongoing research projects (see later sections) include the evaluation of the applicability of some of these methods for the detection of genetically engineered food.

3.2.1 Ligase Chain Reaction (LCR)

The ligase chain reaction is a DNA amplification method based on repeated cycles of oligonucleotide hybridisation and ligation (Backman and Young, 1989; Carrino and Lee, 1995). The method employs sets of oligonucleotides specific to stretches of the target sequence that are in close proximity to each other, as well as another set of oligonucleotides that is complementary to the first set. The protocol is very similar to PCR, except that LCR uses a heat-stable ligase. Polymerase activity is not needed since the primers basically constitute virtually the entire length of the target sequence. Therefore, the length of the amplicon will generally be limited by the availability of longer oligonucleotides. Although known for years now, LCR or variations of this technique (e.g. Gap-LCR) is by far not as significant in routine diagnostics as is PCR (Carrino and Lee, 1995; Pfeffer et al., 1995).

3.2.2 Nucleic Acid Sequence-Based Amplification (NASBA)

This technique mimics the process of retroviral replication (Compton, 1991) and has been used until now primarily for the amplification of RNA molecules (Carrino and Lee, 1995). The method might be applicable for the detection of expressed transgenes and/or viable microorganisms (Blais et al., 1997). Because RNA molecules are present in much higher copy numbers than the respective gene (provided the gene is expressed), NASBA may demonstrate a greater degree of sensitivity compared to PCR for certain applications (Lunel et al., 1995). However, RNA is much more sensitive to degradation than DNA; therefore, the probe material must necessarily be very fresh and appropriately handled. For heat-treated and other processed foods the applicability of NASBA seems very limited. As PCR assays of fresh foods are normally sufficiently sensitive, it seems unlikely that NASBA will find broad application in food analysis.

3.2.3 'Self-sustained sequence replication' (3SR) and 'Q replicase amplification'

Methods for the identification of pathogenic microorganisms have already been developed based on the isothermal 3SR and Q replicase amplification techniques (Carrino and Lee, 1995; Pfeffer et al., 1995). Despite a high amplification rate, these techniques are of less significance in diagnostics as compared to PCR (Pfeffer et al., 1995). Moreover, the alleged technical advantage of an isothermal reaction (Pfeffer et al., 1995), with fast amplification that is not limited by defined temperature and time-cycles and requiring less special equipment, can actually be a disadvantage when compared to methods such as PCR and LCR, which employ pre-set cycles: discrete obligatory temperature cycles have been considered to be a main cause for the relatively minor tendency of PCR for certain experimental artefacts ('in vitro evolution', i.e. amplifying artificially small DNA fragments), whereas isothermal techniques favour fast replicators (Bull and Pease, 1995) and thus short amplicons.

3.2.4 Fingerprinting techniques (RFLP, AFLP, RAPD, etc.)

Fingerprinting techniques such as RFLP (Restriction Fragment Length Polymorphism), AFLP (Amplified Fragment Length Polymorphism) or RAPD (Random Amplified Polymorphic DNA) are used in forensic analysis and for the classification of organisms. They have been successfully used in combination with PCR amplification to classify microorganisms (Tichy and Simon, 1994) and for other applications (Welsh et al., 1995). Fingerprint techniques are applicable for the analysis of complex mixtures of microorganisms used as starter cultures. In this context, fingerprinting may allow to confirm, if a given genetic modification is indeed present in the expected genetic background of a given microorganism. These techniques are based on the comparison of the genomes of related organisms but they may not be sensitive enough to resolve the difference between the DNA of transgenic organisms and their conventional counterparts. The genetic differences among varieties of the same crop are by far greater than differences between a genetically engineered crop and its conventional counterpart. Therefore, with fingerprinting methods it is essential that the DNA compared be derived from exactly the same crop variety before and after transformation. If more than one transgenic product of a certain species (e.g. corn) exists, the DNA of all the respective hosts will be required. Such conditions are difficult to satisfy. Furthermore, fingerprint techniques apparently cannot be used for analysing complex food mixtures or processed foods.

3.2.5 Probe hybridisation

Hybridisations using DNA probes have been frequently used for the detection of pathogens in food (Jones, 1991). One model system for the detection of genetically modified bacteria in milk has been published (Casey et al., 1993). The degree of sensitivity and specificity of probe hybridisation is significantly lower than that achieved through the previously described amplification techniques. Since plants have particularly large genomes but transgenes are present only one or a few copies (thus the relative concentration of target sequence to total DNA is low), the application of probe hybridisation for detecting GMO crops does not seem very promising. However, provided that the target sequence is present in sufficient concentrations (multiple copies of the transgenes, small genome size [e.g. bacteria]), probe quantity is not significantly limited, and highly specific oligonucleotide probes are available, probe hybridisation may provide a simple technique worth considering for screening purposes.