DNA testing is an important part of our life now — as we have seen with proving that a skeleton in Leicester comes from the body of the last Plantagenet King Richard III, and not a mediaeval Monk, or with the implications of finding horse meat in a lot of beef products. I’ve just done an interview with Ben Jackson on BBC Radio Leicester about tracking the provenance of our food. Here are my notes and some comments about the things discussed at the live interview by Ben Jackson. It will be available for the next week at http://www.bbc.co.uk/programmes/p013nxz2 starting after the news and traffic roundup at about 17.15, 2hr15min from the start.
There is a very welcome move towards traceability of our food from farm to fork — that’s what why there are ever more complex codes stamped on every food item you buy. But the traceability relies on a paper trail, and where there is money to be made by passing off something cheaper as something more expensive, somebody is going to try to falsify the documents — fraudulent labelling of inferior products. So that is where testing comes in: a retailer, or in the horsemeat case it seems even the manufacturers, can check that some parts of the paper trail are correct. For those reading this blog outside the UK and Ireland, at the end of last year and published in January, a high proportion of horsemeat was found in beefburgers (that’s English for hamburgers internationally) on sale in Ireland, by the Food Standards Agency of Ireland, and the same manufacturing plants were supplying similar product to many UK shops. Not surprisingly, nobody had really tested beef products for horse, horsemeat almost never being eaten in the UK. But now the test are being done, even after the originally suspect products were pulled from the shelf, both contamination of beef with pork, and outright fraud of selling horse as beef in processed foods, has come to light in many products still on sale. Now the UK Food Standards Agency has started testing on a larger scale, the latest news this week is a ‘beef’ lasagne where over 60% of the meat is horse. (I’m afraid the true cynics amongst us are surprised that there was indeed meat in such products; personally I would prefer a more botanical dish!)
The Radio Leicester interview was played in with the Prime Minister talking at a press conference this morning, “It’s important to say there’s no reason to believe any frozen food currently on sale is unsafe or a danger to health. It’s not about food safety – it’s about proper food labelling and about confidence in retailers”. “Economical with the truth”, in the UK at least, has become a polite expression, as used in Parliament, for accusing someone of telling an outright lie. So I’m not quite sure how I say that our Prime Minister was being economical with the truth, but in the most literal sense. There are exceptionally rigorous rules about every aspect of treatment of our food animals from birth, through death and onwards to processing for food. Most people are entirely unaware of the huge recording, licensing and monitoring overheads that are there for keeping farm animals: for some species, the paperwork needed for moving them in a van from one field to another is the longest part of the task, while every drug is recorded, their diet strictly defined, and both welfare and health monitored and recorded. (As a friend said to me when his wife had to wait 10 weeks for a hospital appointment, he’d be in jail if his lame cow had to wait two weeks to see the vet.) While these rules are onerous, the Prime Minister should know that they are in place for two reasons: food safety and animal welfare. If the paper trail in the supply chain is so compromised that there is no proper record even of the species in our food, it is almost certain that the safety of the food is in jeopardy, and will not meet anything like the standards of farming or of processing that we have come to expect.
Fortunately, it seems people aren’t poisoned or allergic to horsemeat, but most horses are treated at some point in their life and have a residue of an anti-inflammatory painkiller, ‘bute’, or phenylbutazone which is not safe enough for human use. Perhaps the worst recent food contamination in the last decade was addition of melamine powder, from a plastic which includes nitrogen in its molecules, to milk powder. Tests at that time could not separate melamine from protein, and about 300,000 children in China were given milk with melamine to make it look like high-protein milk.
DNA is a remarkably stable molecule, and is present in all the foods we eat that is made up of cells. Each cell has hundreds of millions of four DNA letters (A, T, C and G) in a particular sequence, and in that sequence of four letters (bases) in the DNA, there are characteristic signatures of whichever animal or plant made the cells. There are several ways to see which organism made the DNA depending what is required. Many tests will only tell you what is there when you test for that particular animal. That seems to be the case with the testing of processed food in Ireland and the UK until now: nobody was testing for horse. How far should you go in making tests? Rabbit, donkey … or less palatable products: mink or dog; or rare but perhaps cheap meat: river buffalo, camel, elephant?
Now, most DNA tests will use a molecular method called PCR to amplify very characteristic pieces of DNA from a test sample. If they amplify, then that animal was used in the product — and you can see the product easily in the laboratory. But this test will only say whether what you are testing for is there: you need to do a different test for each species. The test will take typically 8 hours.
More specific, but much more expensive is reading the DNA letters to see which species if comes from. In this test, again specific pieces of DNA are amplified from the test sample (exactly the same as was done in Leicester with Richard III), and then they are sequenced, and the DNA code made up of the four letters is compared with reference samples. Anybody can see these reference DNA sequences over the internet — search at www.ebi.ac.uk for example. This will tell you all the animals that are present in a sample. I saw one supermarket company reported these tests will cost £400 or £500 per sample ($/€ 750): this is higher than my lab but what a well-documented result would cost (my raw costs without labour or paying for facilities would be £5 for DNA extraction, £5 for amplification, £10 for separation and purification, £50 for the cloning of the DNA and isolation, then say 20 sequence runs totalling £200 so a total of c. £300 per sample). Altogether, this will take something like a week: this is acceptable for frozen food although expensive in terms of storage and stock capital. For fresh food, a week waiting for a test would not be possible. Sequencing DNA will identify all the species present in the sample, but not the proportion present with any accuracy. For that, another quantitative PCR methods with specific primers is needed.
Other tests are also possible: each animal has characteristic proteins, and antibodies (made in other animals) can be used to test to origin of the proteins in meat. Formerly, hybridization with radioactive DNA probes was used: now my lab. does this to see how DNA is organized in a species, but not to examine the origin of the DNA.
DNA testing today needs the most minute sample: less than the weight of a pin is ample. This sensitivity might be valuable for identifying a crook from a drop of blood, or a King from a few fragments of bone marrow, but it brings another problem for food. A factory will typically be handling hundreds if not thousands of tons of food each week, so how do you obtain a ‘typical’ sample? A fragment of meat will originate from one animal — but will certainly not be an ‘average’ of the whole production of the factory. For sampling grain arriving from a truck off the combine, there are many elaborate but accurate bulk sampling approaches (pictured) but how can it be done with meat in a factory? Another problem is contamination: if samples are not carefully taken and stored, they can become mixed with other samples. Moreover, modern DNA tests are so sensitive that a stray hair in a sample, perhaps pulled from a pet, would easily be detected.
Over the last decades, food testing has been a growing industry: food safety has been driving the increase, but followed closely by identification of food composition, whether that is with respect to allergens, or misrepresentation of ingredients as with horse in beef lasagne or burgers. The testing of food — and feed — is mostly done by official feed and food control laboratories which in the UK are mostly designated by the Food Standards Agency. As far as I know, there are none in Leicestershire.
International trade has always been important for food — think of the spice routes of the middle ages. But now it is a global market — some products can’t be grown in the UK, in other cases, they are surplus in some countries. Kidney and liver not typically eaten in US, or ducks feet in the UK, but high value elsewhere (I once sat next to the UK duck-foot export expert on a flight to China!). But the scale of food transport is increasing — we want cheap food, but would rather have a less productive natural landscape around us, and want the same food year-round, so hence we are importing and transporting on an unprecedented scale. In the horsemeat case, the retailers pass the blame to the manufacturers, who pass it to their suppliers. What is notable to me is just how diverse and distant these suppliers, who seem to have had the tampered documentation, are from the producers. My laboratory in Leicester works closely with the researchers at Teagasc - The Irish Agriculture and Food Development Authority, on the genetics of grass crops, a multibillion euro industry used to feed animals. It is clear from my in-box and Twitter traffic the potential devastation of the Irish cattle industry, up to now a byword for quality and purity, by the malpractices revealed by the DNA tests.
My lab and the University of Leicester does not do ‘food testing’ on a commercial scale, but we develop the science that is underpinning the tests that are applied to food. This work is also important for understanding the genetic and evolutionary relationships between food species — both animals and plants. We want to characterize their nearest wild relatives, trace the genetic differences with modern species, and see how new breeds of varieties are developed which are disease resistant and productive. My research group works with both plants and animals. For example, for a number of years we worked with the relationships of all animals in the bovid group. This means we developed what could be used as DNA tests for different animals such as river buffalo (pw and user both ‘visitor’; a close relative of cows), or the many deer species.
I think these paragraphs covered most of the areas I wanted to discuss: the main point was to say DNA testing was now important to verifying the paper-trail showing the origin of our food. We should be worried about the level of deceit and fraud that has been revealed in the last two months; it looks as though it has not had a food safety impact but certainly testing must be increased so we can be sure we buy what we want to eat.