Jul 18, 2010

An Experiment in Scientific Truth: GM wheat yields 48-56 % less in field experiments

As a student of agriculture, I had worked on a research project on genotype x phenotype interaction in some of the traditional cultivars of wheat found in the hilly State of Himachal Pradesh in northwest India, leading to my master's degree. My research project, supported by a GTZ sponsored fellowship, entailed conducting field trials on 30-odd wheat cultivars that I had painstakingly collected from different regions, for two consecutive years. I wish I had the time and the inclination to continue with the same research project for my doctorate thesis.

It was quite a laborious job. The similarly designed experiments were laid out at three different locations (from three different ecological zones) and altitude in Himachal Pradesh. The frequent travels especially to the higher reaches of Himachal Pradesh (where roads would open for not more than 6-8 months in a year) made it quite a difficult task to monitor the experiments, but at the same time posed a challenge.

Why I am telling you this is to highlight the importance of genotype x environment interactions, which receive less attention nowadays. In my opinion, the absence of multi-locational field trials under diverse environments, is what is leading to the failure of crop varieties, including transgenes, in several parts of the globe. In India, multi-location trials especially in case of genetically modified crops are being used primarily to build on seed supply. There is not enough scientific literature on genotype x environment interactions. "Breeding trials to select lines for further investigation do not need full replication randomization, yet for an assessment of the ecological behaviour of such lines, replicated and randomized ecological experiments would be required. (see the study below)"

It is after long that I have come across an excellent scientific paper that needs proper understanding and more scientific investigation. It has thrown up so many questions, and good science is all about enquiry, that the GM industry may find it difficult to fathom. And knowing its muscle-power, I am sure the GM industry will throttle the scientific community into silence. It knows how to bribe and manipulate the regulatory system (and US FDA as well as India's GEAC are classic examples) into submission, and therefore good science will remain buried.

I draw your attention to a research experiment being reported from Switzerland for transgene x environment interactions in genetically modified wheat. The team of researchers, led by Simon L Zeller from the Institute of Evolutionary Ecology and Environment Studies at the University of Zurich, had used the transgene wheat variety Bobwhite SH 98 26 transformed with a powdery mildew resistance gene Pm3b. They grew four offspring pairs, each consisting of a GM line and its corresponding non-GM line, under different soil nutrient conditions and also treated for fungicide treatments in the glasshouse as well as in the field.

What is interesting is to see the performance of the transgene in the glasshouse conditions and in the fields. It differed quite significantly. This is what the researchers found: Without fungicide treatment, in the glasshouse GM lines had increased vegetative biomass and seed number and a twofold yield compared with control lines. In the field these results were reversed. Fertilization generally increased GM/control differences in the glasshouse but not in the field. Two of four GM lines showed up to 56% yield reduction and a 40-fold increase of infection with ergot disease Claviceps purpurea compared with their control lines in the field experiment; one GM line was very similar to its control.

Interestingly, when you dwell deep, you find that in the glasshouse experiments, the researchers found that while the control lines benefited from the fungicide treatment, the GM lines reacted negatively. The next line is more significant. It says: The yield of GM lines dropped lower than the yield of the sprayed controlled lines. According to researchers, it means that the cost of resistance might be high if the pathogen is absent. I think we need more explanation for this interaction.

Unintended effects of single gene transfers, says the researchers, are always smaller in experiments using naturally occurring genetic variation and wild plants. I agree. "Even when we included crop plants, we could not find any publications where single genes reduced quantitative fitness traits in a plant as strongly as in the present case, yet only in the field and not in the glasshouse."

"Commercial glyphosate-resistant soybean cultivars were found to suffer from a 5 per cent yield depression that might be caused by the transgene or its insertion process; One study tested wheat varieties with introduced resistant genes against leaf and stripe rust and reported a 12 per cent reduction in yield, which was considered to be a very large effect. Compared with these, the yield reductions of 48 to 56 per cent in our two GM lines of wheat expressing the Pm3b gene are much larger."

The researchers also found that differences between GM plants and non-GM plants increased with nutrient levels (read fertiliser application, in the glasshouse). I am so glad that the researchers were honest in admitting that they have no explanation for this result, and have suggested more such tests across a range of environments.

Let me draw out some other salient points from this study:

1. GM plants had significantly fewer seeds and lower seed yield than control plants.

2. In the field, GM plants showed increased infestation by ergot fungus compared with control plants. In the glasshouse, soils were free of ergot fungus.

3. Like in the glasshouse, mildew infection increased with fertiliser application in the field.

4. In the field, yield of GM lines significantly differed when compared with the corresponding control lines.

5. High fungicide dose increased the extent of the stress reaction of GM plants.  In the field too, environment stress reduced the fitness of the GM plants. In other words, GM plants were increasingly prone to biotic and abiotic stress.

6. GM plants differ in morphological, fitness and pathogen-related traits from their control plants.

7. The four GM lines, although with identical transgenes in homozygous condition, significantly differed among themselves. Although the researchers have given several explanations to address this mystery, including the disruption of the native genes by the insertion of the transgene, but have refrained from pointing to any definite reason.

8. There is still a question whether the over expression of the transgene led to an overabundance of its protein product and the subsequent phenotypic effects or of other mechanisms would be involved.

9. Plant morphology changed when the GM plants were exposed to field conditions, with significant differences in the flowering time, and the level of ergot infection.

10. The study concludes by saying that the lines that perform particularly well in a specific environment may perform poorly in other environments.

You can view the scientific paper at:

The scientific community definitely needs to have a fresh look at GM crops in view of the disruptions that ecological and environmental factors can cause to its genetic makeup resulting in serious distortions in performance. This has grave implications for farmers, consumers and the environment. Scientists cannot be pardoned for deliberately ignoring genotype x environment interactions.


Anonymous said...

I want to share with you this letter from Dr Seralini of France in response to this study. It was circulated to a group of people.

Devinder Sharma

Dear friends,

Of course we know now how a transgenic insertion may modify the genome network. In 9 years of experience in the official commissions in France and EU all my contradictors admitted that.

But they said that:
1/ We never eat original GMOs but the 10th or 20th generation, thus an agronomic selection is applied in between (that means for them no matter with that story, there are more chances to get unintended effects with regular hybridization) - I don't believe that a priori but it is important to know their arguments.
2/ They don't define how the substantial equivalence should be met exactly (compare how many fields during how many years ? 3 x 3 to establish SEM with an isogenic non transformed variety, I proposed, this was never performed in the world - from my experience in the WTO conflict between US and EU), and then what parameters to measure in details - it is up to the firms - now it is quite different for different GMOs. Thus nobody cares about that in our contradictors.
3/ They all admit now in EFSA and FDA that if the substantial equivalence is obviously not met (the Amflora potato for instance), it is up to any official committee to require animal tests or not, and no committee did, and thus these tests are not still obligatory. Then they didn't require any mammalian test with blood analyses for most new crossed GM events, or they accepted tests with only 10 rats and 20 controls (for GM potato, during 90 days).

In fact, there is no scientific debate, but laxity against our arguments.

Friendly yours,
GE Seralini

Anonymous said...

Gilles-Eric and all,

You make the essential points on this. And there has also been an unstated assumption that the obvious morphological or agronomic defects that show up in the lab or fields tests, which are screened out, will somehow also remove toxicity problems or less obvious environmental problems--in other words, that substantial alterations in plant metabolism that could cause toxicity or environmental problems will also show up as gross morphological problems in the plant. There is no molecular or biochemical basis for this as a generalization. The corollary is our EPA trying to justify the supposed adequacy of acute toxicity tests for GE proteins by saying (with virtually no support in the science literature) that a single high dose administered to rats will reveal potential chronic toxicity problems by manifesting in some kind of (unspecified) acute toxicity (If you look at earlier EPA approval documents for Bt and so on, they cite a single paper in support of this, which in turn cites only one or a few references of their own previous paper). The whole thing seems like a house of cards, and basically rests on the assumption that these crops will not be more harmful than conventionally bred crops--which most of the supporters of GE also claim.

The field trials are also not adequate to detect many types of environmental harm--especially those that require time to develop, like weed or insect resistance or other adaptations of the broader agroenvironment. But they also won't work for many more-immediate problems. For example, when roundup-ready sugar beets were first being approved by our Department of Agriculture (USDA), field trials at several sites for several years showed much higher, and statistically significant, levels of the most serious fungal disease of sugar beets here--called Cercospora leaf spot. Because higher levels were not detected in other trials, Monsanto and USDA dismissed this as some kind of anomaly. But as any first year plant pathology student knows, diseases are highly dependent on environmental conditions, the presence of adequate amounts of the pathogen, and the genotype of the fungal strains and crop. It is entirely possible that the observed increases in disease were real, and due to increased susceptibility of the GE sugar beets as favorable environmental conditions. Or it could have been a stochastic blip as USDA suggested. The point is--as Gilles-Eric has shown on the toxicity side--the tests are not adequate to reliably detect and sort out these possible problems. It would have been pretty simple, for example, to have done more controlled tests--inoculations of the RR sugar beets with several strains of the pathogen for example--to begin to determine whether the increases in disease were due to the GE trait. But of course USDA did no such thing. Cercospora has been controlled (not sure if this is still true) by the use of organotin--one of the nastiest fungicides and a known carcinogen. So the consequences if USDA was wrong may have been more than economic. All of this (not the specific example I give here) is also why the Ecological Society of America (the professional society of academic and other ecologists in the US) published a position paper several years ago that recommended that continued monitoring and adaptive management be done for some GE crops after they are approved--which has been ignored by USDA.

- Doug

Doug Gurian-Sherman, Ph.D.
Senior Scientist
Food and Environment
Union of Concerned Scientists
1825 K Street, NW
Suite 800
Washington, DC 20006-1232
phone: 202-331-5436
fax: 202-223-6162

(this letter is also from the same group)