As the horse meat issue continues to unfold I’ve been asked a number of times why we can’t simply use DNA techniques to test for all species in all products – and test for them quickly.
I hope that the following overview of DNA testing will help to explain why it’s not as easy as it appears on CSI.
First, it’s important to say that DNA-based testing is highly specialised, although it is becoming more routine in many food analysis laboratories. Second, I should say that while it’s possible to test for more than one species at a time, the more species you test for the more complex it becomes and the more time and effort is required.
Laboratory analysts often use a two-stage process. Stage one of this processinvolves qualitative analysis to identify whether a sample contains undeclared meat/DNA. Two commonly used DNA-based methods are described. The first method involves the Polymerase Chain Reaction (PCR), often known as ‘basic PCR’. In simple terms, this method can be used to amplify, or copy, trace amounts of a DNA target sequence that is unique to an animal species, so it can be detected. Theoretically, if a food sample contains meat from two different species, then it should be possible to ‘amplify’ target DNA sequences unique to those different species, to show that they are present. The copies of the different DNA sequences present can be visualised according to their size as bands on a gel.
The second method used in the first stage of analysis is Restriction Fragment Length Polymorphisms (RFLP) analysis, which is often used in conjunction with PCR techniques (RFLP-PCR). In the most basic sense, RLFP involves cutting known DNA sequences (with restriction enzyme/s) to produce DNA fragments of various lengths that are then separated by length (which would produce a pattern of bands on a gel). The resulting pattern of bands can be then compared. Theoretically, if DNA from two different species were compared there should be a difference in certain bands (markers/patterns unique to each species). RFLP can be performed quickly in commercial labs.
While these tests can’t determine how much undeclared meat/DNA the sample contains, they are relatively quick and cost effective to run.
The second stage is where the samples containing DNA from undeclared meat species then undergo two confirmatory analyses for quantification and species identification. These methods are more expensive and time consuming than the methods in stage one. However, some clients can afford to skip Stage 1 and directly undertake these more expensive analyses. Used together, both confirmatory methods are the gold standard for DNA analysis for meat species identification.
In basic PCR, DNA of interest can only be detected when enough copies are made (amplified). This makes quantification difficult because by that time the amplification process is not linear (not an exact doubling of copies). A variation of PCR, what is often called ‘real time PCR’, looks at the linear phase where copying theoretically doubles during the PCR process. This activity can be detected as signals (usually by fluorescence). The DNA quantity can then be measured either as the number of copies of DNA or as a relative amount of DNA when normalised to the amount of DNA input.
DNA sequencing is an accurate and reliable technology for determining nucleotide sequences. A band on a gel may appear to be the right size to indicate the presence of undeclared meat species but it is important to check that the correct band (sequence) has been amplified. Once the nucleotide sequence has been determined it can then be compared to others sequences in a reference database, using alignment software tools such as the Basic Local Alignment Search Tool (BLAST). BLAST finds regions of local similarity between sequences. The output can be used to identify the species present (for example, is the DNA detected from horse, cow, pig, etc...). DNA sequencing is very good at identifying the species present but is not a quantitative method.
Sensitivity of tests
Most DNA methods used by analysts can confidently determine levels of different species present to at least 1% weight for weight (w/w) on a meat basis. There are also some commercial primer kits that determine the limit of detection to levels below 1% w/w on a meat basis. The differences in limits of detection depend on the PCR methodology, the way the DNA sample was prepared and whether the DNA extracted was from the nucleus or the mitochondria. Assays based on mitochondrial DNA tend to be the preferred choice for a number of reasons, including a tendency to have a higher number of copies of the target sequence per gram of tissue (more sensitive) and because they may be more heat tolerant. However, single copy target sequences from the nucleus are more suitable for quantification purposes.
Viewers of the CSI TV series may be familiar with many of these techniques but it is fair to comment that in reality DNA-based analyses take much longer than that portrayed in a single episode!