How can you tell a genetically modified soybean from a conventionally produced soybean? This is the question many countries are grappling with as they look at the issue of labelling of foods produced through biotechnology. You can't tell just by looking or tasting the soybean. And unless it has been modified in some way that changes its nutritional characteristics, such as a lower fat content, you also can't tell by doing a routine nutritional analysis.

The introduction of mandatory labelling for products resulting from the application of biotechnology in some countries means that attention has now turned to the need for quantitative testing methods that are able to detect the recombinant DNA (rDNA) ingredients resulting from genetic modification. Yet herein lies the difficulty because while significant progress has been made over the past year, such tests are still being refined. Difficulties arise in sampling methods, costs, lack of standards and false results. In this article, Food Facts Asia looks at the most common methods of testing genetically modified foods and the pros and cons of each.

The two most common methods for detecting whether a food contains products of biotechnology are the polymerase chain reaction (PCR) method and immunoassays.

The PCR method detects DNA sequences resulting from genetic modification. The analysis must be undertaken in a laboratory by trained staff using highly specialised equipment. This method is extremely sensitive and, in theory, it should be able to detect the presence of genetically modified ingredients even at very low levels.

DNA is firstly extracted from the sample and purified. A PCR amplification of the DNA is then made followed by electrophoretic analysis. The way in which the DNA is extracted needs to be varied by the type of food because various components in food can interact with the reagents used in the assay.

One of the main problems with PCR methods is that DNA is generally not detectable in highly heat-processed foods, hydrolysed plant proteins, purified lecithin, starch derivatives and refined oils that have come from genetically modified crops. The method cannot be used therefore for products such as refined oils and sugars in which the rDNA cannot be detected. This problem is starting to be overcome by the use of Real Time PCR that uses fluorescence to monitor the PCR amplification process. Real time PCR however requires very expensive laboratory equipment, which has limited its widespread adoption and availability.

Another problem is that the sensitivity of PCR methods results in a significant number of false positive readings. The test also takes more time than immunoassays with results normally taking one day for the assay and 3 to 5 days to be processed. PCR and related analyses typically cost between US$100 and US$300 per sample.

Rather than measuring the rDNA in a sample, immunoassays measure the levels of proteins expressed by DNA sequences inserted by genetic modification. The test works by using anti-bodies specific for proteins encoded by rDNA sequences. One example of a commonly used immunoassay is the Enzyme-Linked Immunosorbent Assay (or the ELISA).

The ELISA relies on a reaction between antibodies (soluble proteins that are produced by the immune system in response to a foreign substance) and a "foreign substance" (typically the inserted protein), called the "antigen". The reaction is detected by a colour change or a flourometric reaction that can be measured quantitatively.

Immunoassays are less sensitive than PCR methods, which means they are less susceptible to false positives caused by minor levels of contamination. However, careful validation of each food type is needed before the test can be performed because of the large diversity of foods. This means that new assays need to be continually developed as new products of biotechnology are developed. Many hundreds of different assays will eventually be required to enable accurate detection methods.

One of the key advantages of this method of analysis is that the results are generally available within minutes. The costs of immunoassays are also much lower than that for PCR techniques (about US$2-10 per sample) however the costs for the assay development and the generation of antibodies and protein standards mean that the up-front costs can be significant.

One problem with immunoassays is that the technique does not distinguish between different sources of biotechnology-derived rDNA that may express similar protein characteristics. For example, the technique won't detect if a specific protein, such as the Bt protein, is derived from corn or soy. In addition, proteins are denatured by many food processing methods so this technique is more useful for raw foods or food ingredients that have undergone minimal processing.

The choice of technique largely depends on the type of product being analysed and the availability of equipment and funds for the analysis. Both techniques also share some common problems such as a lack of internationally recognised sampling methods or agreement on the number or size of samples required.

In countries where the labelling of foods produced using biotechnology is to be introduced on either a voluntary or a mandatory basis, the ability to accurately identify the presence or otherwise of genetically modified materials is imperative. In addition to accurate analytical techniques, the way in which the ingredient or product is brought to market (the distribution channel) is also critical. It has been estimated that up to 70 per cent of all processed foods contain corn or soy in one form or another because these grains are the source of thousands of ingredients in processed foods.

The United States provides a significant proportion of the world supply of corn and soy. In the United States, labelling for foods produced through biotechnology is not required unless that product has undergone substantial changes such as an alteration of nutrient composition. This means that grains are not generally segregated after harvesting or during distribution and processing and genetically modified grains and conventionally produced grains are often mixed together.

Demands from some markets for grains not produced using biotechnology has caused the US to examine the supply chain. Complete segregation of crops however is difficult and costly to implement. In some cases, buyers accept the assurances of the supplier (the farmer in the first instance) that the product has been produced without the aid of biotechnology. The grain then needs to be totally segregated from genetically modified grains or grains produced without the same assurances, all the way along the shipping channel. This would necessitate separate storage and handling facilities and separate trucks or shipping containers. In some cases, testing, certification and verification are required on the final product.

All of these assurances result in increased costs. One estimate of the costs of such segregation is an additional 6 to 10 per cent premium for non-genetically modified grains. These costs would need to be absorbed by the customer in much the same way as organic products are priced at a premium.

There are many issues that need to be resolved on the analysis and labelling of foods produced using biotechnology.

Neither the PCR nor the ELISA technique has been accepted as quantitatively reliable because of a lack of standardised sampling techniques and a lack of reliable control standards. The good news is that agencies in Europe and the US are working on reference standards and validation programmes for rDNA testing methods. This will enable labelling regulations to be introduced where required and provide the necessary support for resolving disputes arising from trade in genetically modified products. More work needs to be done to ensure that crop segregation methods are both manageable and affordable.