The first 'science' of Roundup Ready soya

March 2011

- The basis for regulatory and commercial acceptance of RR soya

In the beginning, there was Roundup Ready soya, a crop with a unique ability to survive a drenching with Roundup herbicide while every other living, green thing around it curled up and died. Its creator saw that it was good, and that there was a lot of profit to be made from it

Its creator also saw that, while regulators would easily be persuaded of Roundup Ready wonders, the public would not.

Therefore, the creator of Roundup Ready soya took steps to oil the wheels of acceptance.

In 1996 and 1997, just as the first Roundup Ready (RR) soya crops were entering the food chain, letters of concern from the British public to Monsanto were enthusiastically answered with a folio of papers extolling the benefits and safety of its new GM beans:
“Roundup Ready soyabeans are just like any other soyabeans in safety, nutrition, composition and the way they process into high-protein animal fed and ingredients in the food we eat, such as margarine, salad dressings and bakery products.”
At the same time, the Company website quoted “1800 evaluations” concluding RR soyabeans are the same as other commercially available varieties. In the UK, The Soya Bean Information Centre was set up to assure the sceptical that “extensive testing” procedures had found “No health risk of any sort”.

All these glowing assurances were based on three studies which Monsanto had published in 1996.

The papers were accepted by the Journal of Nutrition after “the standard peer-review process”, and were usefully published all together as an information package. Because Monsanto had paid part of the costs of publication, the articles were marked “advertisement”, but readers would have to wade through the small print to discover this.

The soya used in all these studies seems to have been derived from various combinations of beans from the 5th, 6th, and 7th generations after genetic transformation, grown in 13 different sites over 2 different years. The varieties grown were the GM soya commercialised in 1996, and the untransformed parent control. Another GM soya, transformed using the same method (particle gun bombardment, a method notorious for causing random genomic disruption) and using a similar DNA construct (with an added marker gene) was included to double the experimental materials, although this variety was not considered for commercialisation. The beans were processed in various ways: toasted (as in animal feed) from which meal, oil, lecithin, protein isolate and protein concentrate were derived, and untoasted (as in human food).

COMMENT That's a lot of different test materials and a lot of variables to cram into a single study. And data from a different transformation event, not intended for the market, are irrelevant.

The first paper sought to demonstrate that “The Composition of Glyphosate-Tolerant Soybean Seeds Is Equivalent to That of Conventional Soybeans”. This it did by chemical analysis of the various consumed portions of the beans, including:
  1. Macronutrients, such as total protein, energy, fibre etc. (7 parameters in all)
  2. Amino acids resulting from digested proteins (18 parameters in all)
  3. Fatty acids resulting from digested fats (12 parameters, including 2.3% 'unknowns')
  4. Known anti-nutrients and other biologically active substances found in soya (16 parameters in all)
  5. Lecithin fractions (4 parameters)
  6. Oil quality (8 parameters)
This is a total of some 65 tests, which, multiplied by the number of different test materials explains where Monsanto's impressive “1800 evaluations” came from.

These chemical compositional tests give a general nutritional overview, and are long-established measurements of the theoretical products of digestion which are used to check the commercial quality of any new feed crop. They are relevant to the desirability of the crop in terms of food processing and animal feed. What Monsanto presented is a huge array of data on known substances highly unlikely to reveal the presence of novel material. Safety, and unknowns, simply don't figure in such tests.

The authors' conclusions were that “Several minor differences” between glyphosate-tolerant soya and control soya were found to be statistically significant, but were “considered biologically unimportant.”

Presumably, the composition of the fibre fraction was also considered by Monsanto's scientists to be unimportant since only the total fibre content was measured (and like all the other macronutrients, was found not to be significantly different from the control). However, subsequent analyses by other scientists showed that the lignin portion of the fibre was nearly double in the GM soya compared to the control. This unnoticed difference was definitely of biological importance: the excess lignin in GM soya crops has since been identified as a major cause of crop failure in times of environmental stress.

Buried in the 'discussion' of this paper was a note that the parent plants “from which the tested and analysed seed was derived were not treated with glyphosate in order to focus the analysis on any effects of the introduced gene and protein”.

After it was pointed out that, in the absence of the herbicide, the data on these GM crops were not, representative of the GM soya actually being eaten by animals and humans, Monsanto carried out a brief repeat of some of the measurements on Roundup-treated crops.

These subsequent “Quality studies performed on glyphosate-treated, glyphosate

This follow-up study focused particularly on the parameters connected to the 'aromatic biosynthetic' pathway. This pathway is what glyphosate interferes with. It is also the pathway in which the novel protein of the inserted gene is active to overcome the otherwise toxic effects of the herbicide. The study, therefore, specifically included amino-acids and phyto-oestrogens. Lignin is also a product of the aromatic biosynthetic pathway, but the Monsanto scientists' reasoning on what might be disturbed by the engineered gene in action doesn't seem to have extended that far.

The only way to assess complete dietary unknowns is to feed them to animals and monitor the effects, and this indeed was the subject of Monsanto's next advertisement/study.

The second paper was entitled “The Feeding Value of Soybeans Fed to Rats, Chickens, Catfish and Dairy Cattle Is Not Altered by Genetic Incorporation of Glyphosate Tolerance”.

The feed used in these trials was the same as that used for the compositional analyses of the first paper. In other words, half the test animals were fed an irrelevant GM soya, and the source and glyphosate treatment of the feed was indeterminable.

The lactating dairy cow study lasted four weeks. The chicken and catfish study lasted as long as these livestock normally live before slaughter, 6 and 10 weeks respectively. These are short-term studies aimed at commercial concerns, they are not safety studies.

In the Abstract, the tests were referred to as 'animal feeding studies' conducted “as part of a safety assessment program”. In the Introduction, they were “animal feeding trials ... undertaken to provide further support for commercial acceptance.” In the discussion, they are “animal feeding studies (to) provide reassurance”. Either Monsanto's scientists are confused, or are trying to sow confusion.

'Feeding value', as described in the title, determines the 'gain-to-feed performance', or, how much product (milk, meat etc.) you get in return for the feed given. These tests are commercially important, but (as Monsanto itself points out) there are well-established feed nutrient values (as measured in the first publication) which can realistically predict 'feeding value' without actually feeding anything to find out. Despite what Monsanto tells us about providing a “safety assessment” and “assurance”, feeding value studies are not designed to detect toxic effects beyond the very gross. (This limitation was recognised by the UK government when it commissioned Arpad Pusztai to develop the GM safety tests which went inconveniently wrong in 1998). 'Feeding value' trials should not be confused with the animal-feeding experiments GM concern groups have been asking for since the mid-1990s.

The closest Monsanto gets to any kind of toxicity assessment is the 4-week rat feeding study in which the animals' organs were weighed and eye-balled. Since all soya contains a range of anti-nutrients, some of which interfere with pancreatic function, this organ received some extra attention to check for gross tissue changes. The variables measured in all these trials were declared ”similar”, and the differences found during the rat study were easily explained away or declared not “meaningful”.

COMMENT Ignoring data in this way is easy if your measurements are gross enough. Try measuring your dining table rounded up to the nearest centimeter and meter, and use your two sets of data to work out what minimum size of doorway will be needed to move it through. The grossest measurements will give you some very unrealistic answers, and you will have to dismiss them for practical purposes.

In 2003, scientists reviewing the feeding studies on health consequences of GM food and feed, made a number of important criticisms of Monsanto's experiment on rats. They pointed out that the test diet was so artificially high in protein (close to 25% of the chow) that it would almost certainly mask, or reduce, any effects. Monsanto's authors were described as giving a bland assurance that no differences had been recorded without presenting any data. Despite the obvious short-comings of the 1996 studies, these reviewers found only one other feeding study on RR soya worth mentioning in the seven years after Monsanto's publications.

The third paper was designed to test whether the transgenic protein generated by the commercialised GM soya strain was safe to eat. Scientifically, it showed that eating a similar protein produced by a GM bacterium is safe, providing it is pure (in other words not complexed with other food materials), provided it by-passes the mouth to enter the stomach, and provided you have the super-efficient digestive system of a laboratory beaker full of digestive juices. There is minimal possibility of such tests identifying any adverse reaction to the actual protein as actually consumed.


Monsanto's studies of its RR soya are poorly controlled, over-generalised, gross measurements of inappropriate or indefinable materials, and are designed not to reveal unexpected qualities.

Either the Journal of Nutrition's 'standard peer-review' process is poor, or, the 'standard peer-review' applied to an advertisement is very different from the 'Standard peer-review' applied to a scientific contribution.

In summary, the basis for regulatory and commercial acceptance of Monsanto's RR soya seems to be three advertisements dressed up as science.

(This article adapted from an archived article entitled Adversciencing, which first appeared on GM-free Scotland in February 2009)

Padgette et al., The Composition of Glyphosate-Tolerant Soybean seeds Is Equivalent to That of
Conventional Soybeans, Journal of Nutrition, 1996, 126
Padgette et al., Quality studies performed on glyphosate-treated glyphosate-tolerant soybeans (GTS), Monsanto
Hammond et al., The Feeding Value of Soybeans Fed to rats, Chickens, Catfish and Dairy Cattle Is Not Altered by Genetic Incorporation of Glyphosate Tolerance, Journal of Nutrition, 1996 126
Harrison et al. The Expressed Protein in Glyphosate-Tolerant Soybean, 5-Enolypyruvylshikimate-3-Phosphate Synthase from Agrobacterium sp. Strain CP4, Is Rapidly Digested In Vitro and Is Not Toxic to Acutely Gavaged Mice, Journal of Nutrition, 1996, 126
Tudisco et al., Genetically modified soya bean in rabbit feeding: detection of DNA fragments and evaluation of metabolic effects by enzymatic analysis, Animal Science, 2006, 82
Pryme and Lembcke, In vivo studies on possible health consequences of genetically modified food and feed – with particular regard to ingredients consisting of genetically modified plant materials, Nutrition and Health, 2003, 17

No comments:

Post a comment

Thanks for your comment. All comments are moderated before they are published.