Dr. Pusztai initially claimed that GM potatoes damaged the rats' immune systems and caused intestinal damage that could lead to cancer. The real concern arose because he claimed that it was the genetic modification process itself causing the toxicity. A number of scientists voiced criticisms of this conclusion, calling for better 'control' experiments and speculating on other causes for the toxic effects. They identified a number of problems with the design and execution of Pusztai's experiments.

Media Background

After BSE, we've been primed to distrust official assurances about food safety. Our fears have been played on by opposition groups and the media.

In August 1998 Arpad Pusztai appeared in a TV documentary and claimed – based on unpublished experiments - that GM foods pose a risk to human health. There followed a report from his employers criticising his experimental methods and counter-accusations by GM's opponents of a cover-up.

Pusztai was supported by twenty international scientists, including former associates and active opponents of GM; at least one of them later withdrew his support. However, the Royal Society took the unprecedented step of reviewing Pusztai's data and found no evidence for his claims. Despite their rejection of Pusztai conclusions, the editor of the Lancet published the results in November 1999. This was not easy. Even after three revisions the paper was still rejected for publication by two of the six independent referees. The final paper did not carry any of the original claims concerning cancer and suppression of the immune system.

So What Was All the Fuss About?

Pusztai's claim that the GM process itself made the potatoes toxic to the rats.

Scientific Background

Pusztai's experiments were part of a larger three-way collaboration trying to discover whether the gene for lectin (a natural plant insecticide) could be copied from snowdrops and pasted into crops, enabling them to protect themselves from sap-sucking pests.

A group at the University of Durham ‘pasted’ the lectin gene into the potatoes. This group was led by John Gatehouse, who would later criticise Pusztai's feeding experiments in a letter to the Lancet that the editor chose not to publish. Another group at the Scottish Crop Research Institute grew the potatoes and, at the Rowett Research Institute, Pusztai fed the potatoes to rats.

Four groups of young rats (6 rats in each group) were fed either:

• ordinary cooked potatoes
• ordinary raw potatoes
• potatoes laced with lectin (raw and some cooked)
• transgenic potatoes that made lectin themselves (raw and some cooked)

The Results

Before we outline the results, we need to bear in mind that we – as humans - don’t eat raw potatoes because our early ancestors evolved enough sense to realise that, especially when green, potatoes contain poisonous chemicals called alkaloids. At the time, of course, they didn’t have a name for these chemicals but they could tell from the taste that raw potatoes were best avoided. Our mammalian cousins, the rats, also evolved enough sense to avoid raw potatoes.

The rats didn’t particularly want to eat the potatoes (even the cooked ones) and were malnourished in all but the ordinary ‘cooked potato’ group. Yet even this group needed the nutritional supplements it was necessary for Pusztai to give all the rats in every group. Suffice to say, none of the rats ate particularly well during their stay at the Rowett.

What Was Found?

Pusztai found differences in the size of several organs in the young rats. But one problem with malnutrition is that it causes unpredictable fluctuations in the size of the internal organs in such stressed animals. And because the differences between each group were modest, many scientists weren’t convinced that they were anything more than part of this fluctuation. The differences just weren’t big enough to be sure about.

The paper also focused on abnormalities in the rats' intestines. They reported a thickening of the lining of the colon and the jejunum (part of the small intestines) and an increase in a certain type of white blood cell – a sign of something going on with the immune system. But Pusztai and his colleague, Stanley Ewan, used a less than ideal variation on the well-established method for counting these cells, making it difficult for their peers to understand the numbers they generated.

Some of the Criticisms

In the same issue of the Lancet, scientists from the National Institute for Quality Control of Agricultural Products in Wageningen, the Netherlands, said that the toxic effects could have come from the nutritional differences between the potatoes, not because of the GM process. The monotonous diet made the rats protein-starved and was not a good basis to assess toxicity, they argued. The speculation that the GM potatoes caused the jejunal thickening cannot, they said, be supported by these results. The thickening implies an increase in the cycle of cell growth and division, which was not measured.

Scientists use statistics as a way of calculating the ‘significance’ of what they’re seeing; it's a way of trying to decide whether or not they should believe their eyes. Because he had used so few rats in each group, the numbers generated by Pusztai’s experiments weren’t significant enough to many scientists, and they were left in some doubt as to whether or not to believe him. It’s not that they thought he was lying but that – contrary to what some groups would have us believe – the conclusion he drew from his results was not necessarily the most logical explanation. His claim that the genetic modification process itself caused the toxicity is easy to make but very difficult to prove with these experiments. The problem is that no consistent changes were seen in any particular group of rats.


Pusztai would have been helped by better use of control experiments. If he'd included a group of rats that weren't fed any potatoes – but ate the normal fare of lab rats – they could have been used as a 'normal' control against which the other rats could be compared. Such a normal control group would have set a benchmark for the poor diet of the other rats and the difficulty rats have in digesting raw potato.

As it was, he provided no results for normal rats. Instead he compared malnourished rat with malnourished rat and tried to decide which group was the worse off. By omitting a normal control he was assuming that anything found in the group fed GM potatoes would be abnormal.

'Empty Vector'

Pusztai claimed that the toxicity of the potatoes might not be due to the poisonous lectin itself but to the 'promoter' used to control the lectin gene's activity. Promoters are short pieces of DNA that are found at the beginning of every gene (including the those in humans) and act like switches, turning the genes on or off as required. In genetic modification, the promoters used come from viruses because scientists already know how to make them work. Since such viral promoters are commonly used, concern was raised about the whole technology.

The most useful test of whether the GM process itself harmed the rats would have been a 'negative' control: this is performed in exactly the same way as the intended experiment but lacks one vital element. A simple negative control would have been GM potatoes containing every other bit of the foreign DNA except the lectin gene itself. If a group of rats eating these 'empty vector' potatoes became as sickly as the ones eating the lectin-containing potatoes, then this might indicate that the viral promoter made them poisonous. It was suggested that the genetic 'switch' caused the problem, but without such a negative control experiment we cannot tell.

Final Comments

This explanation might still leave us in some confusion. But perhaps this confusion isn't due to unfamiliarity with the words themselves but because they seem to suggest that nothing is black and white. It appears that science is about interpreting shades of grey. This is often the case. To think that science – any science – is about absolute truth is to misunderstand it. Science is more about providing evidence to support as close an approximation of 'black' or 'white' as possible. GM's opponents see this as a weakness of science and use it as a stick with which to beat its practitioners.

They know that balance is as boring as men in grey suits. Their dramatic gestures – gas-masks, trashing sorties etc – make more interesting news. However, these media devices are not arguments and we need to guard against the easy understanding they tempt us into. Many of GM's opponents feel they have nothing to prove and will not be providing any scientific evidence as to why we should believe their interpretation of black and white. They have nothing to lose in their approach to the truth. But if we succumb to their manipulation of the GM issues we risk scape-goating a technology that could – if used responsibly and managed carefully – bring real benefits to the world's environment and the people in it.


Stanley W B Ewen, Arpad Pusztai (1999). Effect of diets containing genetically modified potatoes expressing Galanthus nivalis lectin on rat small intestine. The Lancet, 354, 1353. (http://www.thelancet.com/search/search.isa)

Richard Horton (1999). Genetically modified foods: "absurd" concern or welcome dialogue? The Lancet, 354, 1314. (http://www.thelancet.com/search/search.isa)

Peter Lachmann (1999). Health risks of genetically modified foods. The Lancet, 354, 69.


  questions & answers
25. Don't Pusztai’s experiments with GM potatoes show harm to rats who ate them?