How did top scientists mix up brains from cattle and sheep?

The mystery of how a world-class team of scientists could have spent nearly five years and £217,000 mistakenly testing cow brains instead of sheep brains for signs of BSE deepened yesterday.

Internal links

Labour 'buried report of BSE inquiry'

The mystery of how a world-class team of scientists could have spent nearly five years and £217,000 mistakenly testing cow brains instead of sheep brains for signs of BSE deepened yesterday.

The study was meant to assess whether bovine spongiform encephalopathy had infected sheep at the end of the 1980s when they were fed the same contaminated feed as cattle, but the results have now been declared uninterpretable because of the mix-up.

On a scale of laboratory blunders the error ranks about as high as they come, given that the fate of Britain's 40 million sheep might have rested on its outcome. Finding BSE in sheep could have led to the culling of the entire national flock.

How scientists could have confused brain material from cattle and sheep is now the subject of two inquiries, one by the Institute for Animal Health, the government-funded laboratory where the research was done, and another by the Department for the Environment, Food and Rural Affairs (Defra), which funded the study.

Both inquiries will attempt to explain how two tests early in the experiment appeared to confirm that the material was the mashed-up brains of sheep while a third, more definitive test, which was made public on Wednesday night, found only bovine brain and no evidence of sheep brain at all.

The roots of the story lie in an experiment conceived and run at the height of the BSE epidemic in the late 1980s and early 1990s. At that time, scientists wanted to know what effect different meat-rendering practices would have on the infectious agents behind both BSE and scrapie, a similar brain disease of sheep.

During 1990, scientists from the Government's Central Veterinary Laboratory at Weybridge, Surrey, collected the diseased brains of more than 800 cows with BSE to "pool" the material and test how it would survive different rendering practices. A second phase of the study took place over the next two years when 2,860 brains of sheep affected by scrapie were collected by 18 veterinary laboratories around the country and pooled for a similar rendering experiment.

Material left over from both these experiments was frozen and stored. Not until 1996, when the link was established between BSE and a variant of human Creutzfeldt-Jakob disease (vCJD), did scientists think the stored sheep material might be useful in answering another question: could BSE have got into sheep?

There has always been a theoretical possibility that BSE infected sheep. Laboratory experiments show the agent can be transmitted through feeding and some sheep were known to have eaten the same BSE-contaminated feed as cattle.

The big question is whether infection had actually happened in the late 1980s, before measures were fully enforced to ban the feeding of ruminant- derived feed to ruminants such as sheep and cattle. Realising that they had a store of sheep brains from the early 1990s, scientists suggested "testing" this material for BSE.

However, a possible complication was that the pooled sheep brains may have been cross-contaminated with infected cattle brains. Both sets of brains were removed by veterinary scientists in the same centres using the same instruments and surgical slabs.

With cross-contamination a possibility, the Institute for Animal Health and the Central Veterinary Laboratory did two sets of tests to see whether the material – which looks like porridge – was fundamentally ovine rather than bovine in origin.

Professor Chris Bostock, the institute's director, said one test involved looking for the presence of the amino acid arginine at "position 171" on the molecules of the prion protein present in the sample. The "arginine 171" signature is unique to sheep and the test proved positive, he said. The Central Veterinary Laboratory was also unable to detect bovine material. "This was psychologically good news as far as contamination was concerned," Professor Bostock said.

Convinced that the pooled material was largely if not entirely sheep brains, the complicated experiment took place within the institute's neuropathogenesis unit in Edinburgh, the research centre that did the work proving the link between BSE and vCJD. The research involved injecting the pooled brain material into different strains of laboratory mice, which incubate the disease in a precise pattern depending on whether it is scrapie or BSE. The object was to see if the sheep that had died with "scrapie" were in fact suffering from BSE.

Preliminary results of the experiment were sent to the Food Standards Agency, which issued a statement two months ago to coincide with a separate initiative by the sheep industry to boost the consumption of lamb. The agency warned there was still a "theoretical risk" of BSE infecting sheep and said early results from the experiment involving the pooled brains from the early 1990s "could be compatible with BSE having been in sheep at that time".

Professor Bostock, who was on a holiday when the agency made its statement, was not happy. He said at the time: "I personally think that it is completely unhelpful to start discussing results until they are complete and in the public domain in a way that everyone can see what we are talking about."

The issue of possible cross-contamination had still not been resolved so Defra organised independent DNA tests of the brain material, which began in September. The results were unequivocal: the material was entirely bovine with no trace of sheep tissue. "Extraordinary," declared Professor Peter Smith of the Government's spongiform encephalopathy advisory committee. Professor Bostock said he was "flabbergasted".

One possibility being investigated is that the pool of cattle brains collected in 1990 was confused with the pool of sheep brains collected two years later.

Professor Bostock said: "There has to be an explanation for this discrepancy. We'll have to wait to see what [it] is."

Meanwhile, we are still no nearer to knowing whether BSE has infected sheep.