Dec 7, 2010

Debate

Send

Debate: The future of biosafety research

“We need to develop a clear, efficient system for analysing new plants.”

GMO Safety: Over the next few years plants will be bred with new traits – plants that produce renewable raw materials for industry, or plants that produce biomass for energy generation, plants that are better adapted to abiotic stress factors, such as drought, and plants with new disease resistance. What biosafety problems could these new plants pose?

Inge Broer: Any biosafety problems will always depend on the crop plant and the specific trait in question. In principle, one can conjecture that in the case of types of stress tolerance, e.g. disease resistance, the crop plant will become less dependent on cultivation by humans and will therefore be more likely to become established in surrounding areas. Some crop types that are already highly competitive could perhaps even end up growing on non-arable land. If a new trait increases a plant’s competitive strength, very careful checks should be carried out to see whether the modified crop plant has any crossing partners among the wild plants in the region where it is to be grown. We should not run the risk of such hybrids being more invasive. But, again, this depends on the trait and the crop plant. For instance, oilseed rape has wild crossing partners here in Germany, whereas maize and potatoes do not.

In the case of plants with higher biomass production, I think they will be more dependent on humans, in other words the risk of them spreading in the environment will be smaller.

GMO Safety: How does one identify the biosafety issues that must be resolved before a plant is cultivated, and how should they be dealt with?

Inge Broer: We need to develop a clear, efficient system for analysing new plants, a kind of decision tree. First you look at the possible risks for a specific combination of transgene and plant. For instance, if a plant produces a protein that kills insect pests, there is the question of whether it harms other insects. If a plant produces a protein that kills bacteria, there is the question of whether it harms soil bacteria.

Then you look at whether it is actually possible for the harm to occur. For instance, does an insect that might be harmed by Bt maize actually come into contact with Bt maize? Does the protein that kills bacteria actually enter the soil? Can a potentially allergenic protein actually pass into the blood?

The next question is how great do the differences between the genetically modified plant and the unmodified plant have to be for us to assume that harm is possible, and which differences are biologically relevant? To investigate this we need indicators. These can be organisms that are particularly sensitive, or test reactions that are very sensitive and could indicate a possible risk. We need to define science-based threshold values for these indicators. Further tests may be necessary if the value measured using a particular measuring system exceeds the threshold value.

Then one has to think carefully about which questions need to be investigated in the field and which can be worked on in the laboratory. Generally, one should always start in the laboratory using standardised conditions. If you find an effect in the laboratory, it must then be checked in the field. However, it is much harder and more time-consuming to make robust statements in field research because there are conditions that are constantly changing and over which one has no control.

A decision-making system like this would be much more efficient than what we do today. At the moment we do not have defined indicators or threshold values.

GMO Safety: Shouldn’t field trials form part of safety research in every case? Or are there plants or traits for which field trials are not necessary? If so, which are they and what criteria can one use to define them?

Inge Broer: Usually, measurement systems in the laboratory are much more sensitive. For some questions, like for instance the impacts on soil micro-organisms, the fluctuations in the field are so high that it has never been possible to demonstrate effects in the field that were clearly visible in the laboratory. So, if I cannot detect anything in the laboratory, there is no point conducting field trials. For other issues, where changing environmental conditions have a big effect on the potential risk, field trials are vital. This applies, for instance, to the amount of the new protein in the plant, which, like all other plant substances, is influenced by the environment. Here we need field data as a basis for the risk assessment.

GMO Safety: But surely you can’t know in advance what effects you will find in the field. A study published just recently by the Swiss Federal Institute of Technology (ETH) in Zurich, reports that transgenic plants presented completely unexpected changes in a field trial, that had not been found in the greenhouse, particularly in relation to yield and pest infestation.

Inge Broer: The fact that you cannot necessarily transfer the results from the greenhouse to the field is not a new finding. Plants grow differently in the greenhouse, where the conditions are different, and people are always finding that yields can vary, and that plants are more or less vigorous or more or less sensitive to pests. For instance, leaves are usually much softer in the greenhouse because they are not exposed to such extreme environmental fluctuations. This can alter pathogen infestation levels. And yields depend for instance on how well the root system can develop and on whether there is interaction with micro-organisms. What happens in a pot or raised bed is very different from what occurs in the field. These differences are also found when conventional varieties are tested in the field for the first time, and field research is important for these as well.

But because the environment changes every year, biosafety research cannot cover all unforeseen effects. That is why we have post-market monitoring, which looks for unexpected effects throughout the entire approval period. The moment a real problem is spotted, the approval is withdrawn.

GMO Safety: Which questions should be researched as part of state-funded biosafety research?

Inge Broer: This type of research can only deal with general questions that need to be asked when government authorities or employees of public institutions are required to assess the risks of a genetically modified plant. For instance, if one wants to investigate whether Cry proteins are stable in the soil, state-funded biosafety research should research the issue for one Cry protein and, using this one protein, obtain enough information to be able to assess whether the measurements for other Cry proteins given by firms applying for approval are realistic. Or, if one is analysing whether certain methods prevent transgenic plants from spreading via pollen, one should develop methods that enable one to carry out the analysis effectively, but not analyse every system.

It is not the government’s job to conduct research for firms that want to submit a transgenic plant for approval. It is about creating a scientific, independent basis for deciding whether a certain group of plants can be approved or not.

GMO Safety: How can the public credibility of biosafety research be improved?

Inge Broer: The first point, which always annoys me, is that our results are rarely used by politicians and are not mentioned as the basis for their decisions. If some politicians do not even have confidence in the research they themselves commission, then the public cannot be expected to trust it either.

Secondly, I think that we need to inform people more about what we do and what results we have produced, so that they realise how absurd the repeated claims are that say we only plant fields to get money and do not actually do any research, and so on. If you see the research results you know that these claims are nonsense.

Thirdly, I think we scientists need to stay professional. In other words, if we cannot see a hypothesis for a certain research question, if we can’t see a way of following a science-based procedure like the one I’ve just described, then we shouldn’t do it. We should not let ourselves be pushed into conducting experiments for which we see no scientific basis just because someone says I am worried because I have a gut feeling about something. This kind of research undermines our credibility. I think we need to assume our responsibility here and use the public resources made available to us to research only those questions that we believe to be relevant.

And fourthly, we scientists should not let ourselves be pushed into making final statements. This is impossible for a serious scientist because there is no 100% certainty, and science is in a constant state of flux. We can only say that the plants that we have studied are, according to everything we know, safer than any other plant you eat. But we cannot say that they are 100% safe or that nothing will ever happen. The politicization of science means that we are being induced to make final statements, and that is a mistake. Science is not definite.

GMO Safety: Thank you for talking to us.