Scientists ignore gene flow in wheat

June 2012

Picture of conventional wheat
Photo by User:Bluemoose (Own work) [GFDL (
CC-BY-SA-3.0 ( or CC-BY-SA-2.5-2.0-1.0
(], via Wikimedia Commons
The apparently orchestrated media storm surrounding the GM wheat trial at the Rothamstead research facility near London (see THE MEDIA DEBACLE AROUND THE ROTHAMSTED GM WHEAT PICNIC-PROTEST - June 2012) led to a final summing-up article in the Guardian.

Like the GM-concern groups, the Guardian noted that, in contrast to past years, the media reporting and comment have been “mostly neutral and positive about the experiments, or actively hostile to the protester” and that the scientists involved have “adopted a new tactic”. It, too, asked “What has changed?”

The article went on to answer its own question. Seemingly, the researchers have adopted a new strategy of acting with “reason and openness”: they are coming “across as genuine in their desire simply to find out the answers” with a message to the activists “to discuss the issues rather than resort to criminal damage”.

To illustrate the new “reason and openness”, the author described how keen the scientists had been to 'point out' that “the risk of pollen from their wheat reaching surrounding crops is vanishingly small because the crop is self-fertilising”. Thus, the logic continued, the urgency claimed by the protesters to stop an open-air GM trial because it could have profound effects on the environment “simply did not exist”.

Is this true?

There are two sources of GM pollen flow from the Rothamstead field trials to consider. One is the pollen produced by the GM crops in the test plots. The other will be produced by subsequent GM plants growing from the seed produced during the trial. Are both these dangers to the environment vanishingly small?

A Swiss paper, published in December 2011, mentions some important facts about past GM research, and goes on to describe some very uncomfortable results of an experiment into GM gene flow in wheat.

Conventional wheat tends to be predominantly self-pollinating. This happens because fertilisation of the flower usually happens before it opens. Add to this that the flowers are only open to release or accept pollen for a very restricted period time and their pollen is heavy enough not to be carried far by a normal wind. Cross pollination is usually restricted to 1-2% in very close proximity (less than a meter) to the pollen-donating plant. These facts form the basis for the Rothamstead scientists' dismissal of any potential for gene pollution emanating from their GM trial.

However, there are “large differences among wheat cultivars concerning pollen-mediated gene flow”. The GM wheat team admit that their novel crop can out-cross at a higher-than-normal rate of 4-5%, and other scientific studies have found that direct contact between plants can achieve up to 10% cross-fertilisation. Also, successful pollination by conventional wheat pollen has been recorded at a distance of 2.75 kilometers from its source plant.

The Swiss paper notes “It is usually assumed that GM-wheat would behave similar to conventional varieties, but only scant evidence corroborates this standpoint”. In other words there is not only a known, very large, variation in the tendency for cross-pollination amongst different conventional breeds of wheat, but no one's ever done the science to check if GM wheat behaves anything like its counterpart.

To fill this knowledge gap, the Swiss researchers devised a comprehensive experiment. They created a range of GM lines containing genes for fungus resistance plus their non-GM sister controls, and used these to measure the extent of cross-pollination arising from other GM and non-GM background crops.

What they found was that, on average, 3.36% of the seed from plants exposed to 'foreign' pollen were cross-breeds: the worst-case recorded was actually 8.5%. In a real field situation, pockets of contamination at 3.36% which would continue to expand at that rate year-on-year isn't exactly 'vanishingly' small, and 8.5% definitely isn't. Over 14% of the pollen-recipient plants had some degree of cross-fertilisation, while the cross-pollinated seed produced by these plants amounted to nearly a fifth of their total. The picture emerging here is that pollen flow from wheat generally happens only over very short distances, but when is does happen it's certainly not at a 'vanishingly' small level.

The Swiss researchers noted that there were not only clear physical differences between their GM strains (all contained the same transgene inserted into the same parent strain of wheat, but each was derived from a different transformation event i.e. the novel gene had landed in a different location), but the potential for gene flow turned out to be equally diverse. One GM strain cross-fertilised at over six times the rate of its non-GM equivalent. These variations may be explained by differences in flower and pollen morphology and physiology, susceptibility to infection, plant height etc., but what's more to the point, the genetic transformations produced individuals which were physically and physiologically unpredictable: no generalisations based on conventional wheat characteristics can be made about their GM relatives.

The Swiss team concluded from their experiment that “it may be difficult to develop universal models for pollen-mediated gene flow in wheat. Our results suggest that a case-by-case approach will be required instead”.

To go back to the two sources of pollen flow from the Rothamstead trial. Most of the pollen produced in the trial (source No.1) will stay in the stand. However, the Swiss study found that the gene-flow it recorded happened in the opposite direction to the prevailing wind. Local gusts are clearly a very important determinant of pollen movement within wheat stands. And, given the wild weather Britain has been experiencing this year, the possibility of the odd cloud of grains riding the wind to reach, and pollinate, wheat farther afield can't be ruled out.

Regarding source No.2, no matter how rigorous the 'clean up' procedure in the test plots, small quantities of seeds from the trial field have a high chance of re-emerging somewhere in Britain.

Even the Rothamstead scientists admit that “dispersal of seed prior to harvest (is) possible by wildlife”: GM seed can hitch a ride in or on birds and other small animals. Spillage is also inevitable. Wherever it ends up, some GM seed may lie dormant to emerge in future years.

Once any of these things have happened, the offspring of the resulting rogue GM plant could generate over 8% GM seeds in any wheat nearby, year on year. Be mindful that the global rice contamination with experimental GM genes came from a small-scale field test many years previously (rice is also reckoned to present a 'low risk' of cross-fertilisation because it's predominately self-pollinating with short-lived pollen and no known insect pollinators); and, how it started is anybody's guess. The short-term risk of gene pollution may be vanishingly small, but the long-term risk is very real, very tangible and is being ignored by the Rothamstead scientists.

There was another unexpected, and very ominous, finding by the Swiss team. They demonstrated a tendency for the GM plants to cross-fertilise preferentially with other GM plants (and vice versa), leading to 'natural ... pyramiding' of transgenes. Indeed, “The proportion of GM plants within a population is therefore likely to increase.” The bottom line here is that GM wheat of any kind may be dangerous to commercialise because of the increased risk that novel transgenic varieties will be generated.

The future 'pyramiding' of genes threatened by the current GM trial involves two antibiotic-resistance genes (at least one of which compromises the efficacy of an antibiotic still in clinical use and known to cross-react with other such life-saving drugs), plus a herbicide-tolerance gene besides entirely artificial DNA constructs which produce a wildlife-disrupting insect-smell.


In summary, pollen flow from wheat can happen, especially over short distances. Within a crop such local levels of pollution are not vanishingly small. In the long-term, the gene pollution arising from a few rogue GM plants gone to seed would not only become significant, but would have a habit of creating novel and unpredictable GM plants.

The suggestion that the protesters are raising a needless alarm about an immediate threat to the environment because wheat is self-fertilising is simplistic, wrong and designed to mislead. The Rothamstead scientists who are acting with such 'reason and openness' have in reality been disseminating a scientifically-baseless assumption about the potential for GM gene flow.

  • Silvan Rieben, et al., 2011, Gene Flow in Genetically Modified Wheat, PLoS ONE, 6:12, December 2012
  • James Randerson, The GM debate is growing up, Guardian 30.05.12
  • GM Wheat: Cross-pollination and contamination, GM Freeze, 3.05.12
  • Behind the GM Wheat Trial, Institute of Science in Society Report, 20.06.12
  • GM rice trial, CSIRO Plant Trial Application DIR 052/2004, approved 2005

No comments:

Post a comment

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