Having offered a DNA test for horse for more than ten years, RSSL is advising the food industry to understand the limitations of testing before rushing to test. With so many companies in the food chain under pressure to publish results, there is a danger that companies may make the wrong choices in testing.
There are two routes to detect horse meat in beef. One relies on a test for protein (ELISA) and the other on a test for DNA. The ELISA methods have a detection limit of 1 - 2%, but there are separate test kits for raw and cooked meat. In the current testing situation, this information may not be readily available as the source of contamination is largely unknown, so this choice of testing may not be suitable. DNA analysis has a verifiable detection limit of 1%, although it is possible for DNA methods to detect DNA at much lower levels. RSSL offers two types of DNA analysis.
The first, known as real-time polymerase chain reaction (real-time PCR), involves using primer sequences (short lengths of DNA) specific to horse, which can pair with any complementary horse DNA found in the sample. The test then amplifies the specific DNA, like a biological photocopier (PCR), and these copies can be counted in real time as they are produced. The second DNA method is PCR, RFLP (restriction fragment length polymorphism). Here, specific animal genes from within the meat sample are replicated many times, and the copies are 'cut' using restriction enzymes. The enzymes can be chosen to cut the DNA at points that give fragments specific to the species of interest. When these fragments are separated on a gel, the pattern of fragments can be compared against a reference sample to determine whether horse DNA (or DNA from any chosen species) is present.
Both of these methods work by replicating mitochondrial DNA, which is more highly conserved and more abundant in animals, than nuclear DNA. However, this also means that the methods cannot be used to quantify the amount of horse DNA found in beef. In any given sample of comminuted meat, there will be muscle meat, and potentially cells from other parts of the animal. Muscle cells contain many more mitochondria than cells in other tissues, but in any event, there is no way of telling how many mitochondria might be in any given cell and they can vary from 1 per cell to 10,000. So, there is no way of knowing how much DNA was in the sample. As a simple illustration, the same amount of starting DNA could arise from a small amount of muscle meat containing many thousands of mitochondria, or from a large amount of other tissue containing many thousand fewer mitochondria. Therefore this type of testing cannot be used to quantify the amount of DNA in the starting material.
Therefore, unfortunately, there are no quick answers to testing, and no definitive way of saying how much horse DNA is in any given sample. The DNA methods also require careful preparation of samples as well as expert application to over come the risks of false positive or false negative results. With some laboratories rushing to add DNA testing to their portfolio of services, there is a danger that testing will create confusion when the industry is looking for clarity says RSSL.