Defra - Department for Environment, Food and Rural Affairs.

Science Search

Science and Research Projects

Return to Science Search homepage   Return to Project List

Methods for the detection of adventitious GM material in non-GM seed - VS0123


With the increasing availability of genetically engineered plant products, it is necessary to have robust and accurate techniques for the identification of such products as a means of monitoring adherence to labelling requirements and other regulations. Methodology will be based on a robust and statistically valid sampling system that will be based upon ISTA/PHSI established systems with due consideration of homogenous vs stratified samples. (Improved sampling systems that will encompass consideration of possible problems associated with seed imported into the UK and potential inaccuracies inherent in the current sequence of sample handling to the laboratory will be addressed in a separate desk study financed by DEFRA to be carried out at NIAB).

Following collection of details of construct sequences of the relevant GM events, which are already available from ISPRA and collected for use by competent authorities. A three-step approach is envisaged as most appropriate; based on a widely applicable PCR-based screening method which will target sequences present in genetic elements most commonly found in transgenic crops, such as the aminoglycoside-3`-phosphotransferase (nptII) marker gene, the cauliflower mosaic virus 35S promoter (P-35S) or the nopaline-synthase - terminator (nos 3`). The design and orientation of the primers will be chosen so that a unique profile will be generated for each different t-DNA that has been used to produce an approved transgenic plant variety (event). Such an approach, in conjunction with information from around the insertion site, will differentiate approved events from non-approved events and natural sources of sequences often found in GM material (due to bacterial or viral infection) in one step. The levels of contamination with respect to approved and non-approved events will be quantified.

The work proposed will lay a foundation for future work that will be necessary to enable determination of the number of t-DNA insertion events present. This will lead to an accurate interpretation of the number of GM sample units (e.g. seeds) to be made. This will be achieved through the combination of a PCR based technique that gives an unambiguous figure for the number of t-DNAs present and a sequential sampling strategy designed to eliminate sampling and analytical error. Sampling error and analytical error both contribute to the final error of the estimate and if these two factors are not taken into account it is quite possible to have unacceptable levels of ‘false accepts’ (false negatives) or ‘false rejects’ (false positives).

The prime objective of this research is to develop methods that will allow laboratories to distinguish between true and false positives – and will allow detection of contaminants at the levels set for both approved and non-approved events.
The set of three methods will allow characterisation and quantification of transgenic components in a sample of plant material. Unambiguous identification of copy number will be possible in part three, however it is envisaged that this will be taken only as far as proof of concept for this study. Further work would be needed to in the future to fully test this element.

1) Development of a general statistical sampling model. This will comprise development and testing of a general model that is applicable to the range of seeds requiring sampling currently and in the near future and testing for contamination by GM seeds at various levels. The model will take account of the nature of the seed (size and type), the acceptable trading or legal standard and the analytical error of the method to be used to test the sample. Initial work to test the modal has already been completed on a laboratory scale: simulated by using colour-marked grains added to a bulk, which was then sub-sampled.

2) A method will be designed to unambiguously identify a transgene construct in an unknown genetic background and allows the simultaneous detection of the presence of known T-DNA components; it will confirm their relative positions and orientations. This would allow the construction of characteristic profiles for constructs that are components of events approved for release, or non-approved. This information in conjunction with information from the insertion site/s (see 3 below), will enable constructs that are part of approved events to be distinguished, both from the same constructs in non-approved events, and from wild type bacterial sequences.

3) A method will be designed to quantify the number of insertions present in the genome of the tested plant species. This will be achieved through the PCR amplification of the junctions between the t-DNA and the genomic region into which it is inserted. A method has been developed that does not require prior sequence knowledge of the insertion site or interpretation of amounts of PCR product. The combination of information about the construct from (2) above the the additional information derived from the insertion site will unambiguously identify the event concerned and whether it is approved or not.

The robustness and repeatability of the approaches developed will be assessed through the performance of a ring test involving three independent laboratories. The two main partners engaged on this study (RHM and NIAB) will work closely and will include the laboratory in Belgium, run by Advanta Seeds in the ring test stage. It is also envisaged that other laboratories may be invited to participate in the ring test or after completion of the project, following consultation with DEFRA.

By combining the data from the three methods above the number and identity of GM seeds etc in a sample can be accurately calculated even where different t-DNAs are present in the sample and where seeds with different copy numbers are present. The latter is possible because material is generally selected from a single isolated clone following selection.
1.1 To develop a method that will allow the unambiguous identification of any transgene cassette or t-DNA region in an unknown genetic background. This method will allow the simultaneous detection of the presence of all known t-DNA components and confirm their relative positions and orientations.

1.2 To develop a PCR based method that will unambiguously quantify the number of transgene insertions present in the genome of the tested plant species. This method will not require prior knowledge of the plant genome and will work with admixtures of seeds derived from different transgenic events. The number and sizes of the products in conjunction with the profiles from 1.1 will unambiguously identify approved from non-approved events

To develop a general statistical sampling model. This will comprise developing and testing a general model that is applicable to a range of seeds requiring sampling and testing for contamination by GM seeds at various levels.

1.4 To achieve technically and commercially acceptable test methods, verifying repeatability and robustness of the techniques by performing a ring-test among three independent laboratories.
Time-Scale and Cost
From: 2001

To: 2002

Cost: £40,000
Contractor / Funded Organisations
RHM Group Ltd, National Institute of Agricultural Botany
GM Risk Assessment              
Plant Varieties and Seeds              
Plants and Animals              
Fields of Study
Plant Varieties and Seeds