It is difficult to control GMOs and their release into the wild. To protect the environment, experiments and cultures are well framed by European directives. Speech to Marc Fellous (INRA, Agronomy UMR, Chairman of the Commission for Biomolecular Engineering – Text from 2002 but whose general principles remain true).
Experiments with genetically modified plants in France
The construction or creation of a genetically modified organism (GMO) then its development is a long process that takes place in several stages. This process starts first in a laboratory and includes confined experiments in vitro , then in vivoin an air-conditioned chamber, then in a greenhouse to eventually lead to experiments in an open environment. Some GMOs, designed as scientific tools and intended to improve the knowledge and understanding of plant biology, are used exclusively in a confined environment. On the other hand, field experimentation is considered when it is necessary to acquire certain scientific data, or a development of the GMO for commercial purposes is possible and technically possible. The succession of steps and experiments leads to the construction of knowledge on both the intrinsic characteristics of the GMO and its potential interests and the risks that it is likely to present for health and the environment.
It is notably on the basis of these data that the development of the GMO with a view to valorization in the agricultural, industrial or pharmaceutical field may be considered interesting. On the other hand, the development of the GMO may be abandoned, in particular by the petitioner, if the data acquired shows that it does not meet the desired objectives or that it does not satisfactorily meet the safety requirements.
Moreover, in the field of plant genomics and in particular GMOs, field trials are essential for assessing the risks and potential benefits of growing and marketing plants on large plots. Otherwise, knowledge of the behavior of the GMO in natural conditions and the potential risks it may represent for public health and the environment would be fragmentary. Their appreciation might be too exclusively focused on theoretical approaches.
When a commercial use of the GMO is envisaged, the experimentation of transgenic plants in the field is imposed by the European regulation: Directive 2001/18 / CE on the deliberate release of GMOs, replacing the directive 901220 / CEE indicates, “That the placing on the market of any product consisting of or containing GMOs and intended to be deliberately released without first having been submitted at the stage of research should not be considered and development, with satisfactory field trials, in ecosystems that are likely to be affected by its use “.
Unintended effects of genetically modified plants
The isolation of trials from neighboring crops and other measures taken to limit the dispersal of pollen , seeds or organs for vegetative reproduction .
1.) Factors affecting the extent of these unintended effects
In the context of experimental releases of GMOs, the size of the plots, the number of plants tested, the location of these plots, and the conditions under which the experimental culture is conducted, are factors taken into consideration, as factors which affect the extent of the unintended effects considered.
Unintended effects are a function of several factors depending on the introduced trait, the speciesconcerned and the general context of the use of the GMO. This justifies an assessment approach on a case-by-case basis, a common principle of risk assessment of GMOs within the European Union.
Paragraph 2 proposes an inventory of the unintended effects that can be attributed to genetically modified plants. However, it should be remembered that most of these unintended effects are not usually specific to transgenic plants, but may be relevant to any innovation in plant breeding or breeding. In the assessment, it is therefore necessary to distinguish between the eigenvalues of transgenic plants and those of broader problems.
2.) Inventory of possible unintended effects
The unintended effects that GMOs are likely to produce when used on a commercial scale are of several types: toxicological, environmental, agronomic and economic. Some justify a prior evaluation prior to any deliberate release, especially in the preliminary stages of research and development.
- Unintended effects on the environment
Unintended effects depend on the species considered and its ability to disperse genes into the environment via pollen, seeds or organs that allow vegetative propagation. Depending on the nature of the introduced traits, the GMO may acquire new properties giving it a selective advantage over other varieties or plant species. This may result in unintended effects on the equilibrium of the species in ecosystems, by proliferation in the ecosystem of the transgenic plant itself or the interfertile species that it would have fertilized.
The possible consequences, in particular in the long term, of these flows on the environment can be difficult to evaluate. These possible consequences generate greater concern when they concern regions that correspond to centers of genetic diversity of the genetically modified species tested. The possibility of acquiring several genetically modified characters from the same plant must also be taken into consideration.
GMOs resistant to pests are likely to exert a selection pressure favoring the selection of populations resistant to the toxin expressed by the plant. The unintended effect is then linked to the possible proliferation of these pests due to the difficulties of controlling them.
The release into the soil of native or modified toxins secreted by the plant is also a source of unintended effects on soil ecosystems and, consequently, biological diversity in the immediate environment of the plant.
Plants grown as part of the experiment may also have an unintended effect on “non-target” wildlifespecies that are likely to visit experimental plots.
- Unintentional effects of agricultural practice on these crops
The use of herbicide tolerant GMOs may create indirect unintended effects on the environment due to the use of plant protection molecules on these crops.
In addition, there is also the question of possible economic effects, and the area of agricultural land management linked to new gene flows to commercial production plots. The result is questions about the coexistence of experimentation and cash crops in the same or near areas, and more broadly about the coexistence of different models of agriculture . The choice of organic farming to accept no use of GMOs in its productions, as well as the demands of consumer associations to be able to have a free choice of food without any trace of GMOs, are preoccupations. of this order.
Social or economic criteria do not fall within the regulatory scope of the evaluation advisory bodies that specifically address the issue of genetically modified organisms. Agronomic issues related to the overall management of agricultural areas are not taken into account in the context of the evaluation, on a case-by-case basis, but can be addressed in the framework of specific discussions and made the subject of recommendations. in general opinions of the Biomolecular Engineering Commission.
Prevention means to limit the risks
1.) Exclusion clauses
Prior risk assessment is the first element that leads to risk prevention .
Any experimentation with a genetically modified plant is subject to a prior risk assessment. This risk assessment is based on a scientific and technical file. This dossier provides information on the safety for humans and the environment of the “new genetic construction”. The purpose of the evaluation is to exclude any file for which the information would be considered insufficient or reveal proven risks.
Elements on the origin, history and length of the gene sequence and genetic construct transferred to the host organism must be known. The information specifies in which cellular compartment of the host, this genetic construction is likely to be inserted (nucleus, organelles , cytoplasm ). Evaluations are also based on available information about the organism receiving the genetic construct.
2.) Limitation of unintentional releases
The measures taken around the trials aim to limit to a very low level the possibility of unintended release into the environment of transgenes from the genetically modified organism tested. The measures recommended and imposed by the individual decision to authorize field trials are defined according to the conclusions of the risk assessment carried out beforehand. They are therefore determined case by case according to the biological characteristics of the species considered, the level of knowledge on the transformation event considered and the objectives of the experiment.
Isolation, reproductive, geographical or physical measures make it possible to limit the dispersion of transgenes by means of pollen, seeds or organs that allow vegetative propagation.
- Reproductive isolation
In the case of strict reproductive isolation, the experienced plants produce neither pollen nor seeds: the flowers or inflorescences are eliminated from the plant as soon as they appear. Reproductive isolation is ensured for example by castration of the male organs of the genetically modified plant, sterility-male systems or the disjunction of the flowering periods between the experienced plants and the plants grown nearby. Regular monitoring ensures complete elimination of plant reproductive organs.
- Geographical or physical isolation
Geographical isolations or physical isolation measures have a less dramatic effect on the reduction of this dispersion. In this case, experienced genetically modified plants will breed during the test period. The isolation distance is determined according to the pollen dispersal characteristics of the plant. The objective is to create a buffer zone between the experimental plot of the GMO and neighboring crops in order to reduce the probability of fertilizationcrossed between the GMO and the cultivated plant of the same species, or compatible with it. Isolations using physical measurements such as plant nets or bags on inflorescences further limit this probability, but they require more monitoring and are not technically feasible when the number of plants is important. Finally, complementary devices, made up of peripheral plantations of non-transgenic plants, are often recommended in addition.
3.) Monitoring before and after harvests
Isolation measures are complemented by other provisions that provide control of plants and their dispersion. These provisions concern in particular the monitoring and the destruction of the plants likely to constitute relays of dissemination of the transgenes. These may be plants of the same species or related species present on roadsides or paths. The fertilization of these plants by the pollen of the genetically modified plant would allow the production of hybrids likely, the following year, to disperse in turn the transgene in the environment.
Parcel management measures prevent the deep burial of seeds that have fallen to the ground, and ensure the control of the regrowth that may ensue in the years following the experimentation. They shall also include, where appropriate, the control of propagating organs such as potato tubers or beet root chips. These management and monitoring operations work in the same way as isolation measures to control the unintentional release of transgenes.
Data on gene dispersion
1.) Consequences of pollen dispersal
The conditions by framing the seed production make it possible to guarantee a varietal purity conforming to the requirements of a quality production. A technical regulation of production defines in particular the modalities allowing to ensure a control of pollen pollutions coming from outside the plot of production (incoming flow). The result of these measurements is expressed through the varietal purity rates measured in the batches of basic seed (SB) or certified seed (CS) marketed. These varietal purity rates are good indicators of the effects of isolation measures on the probability of cross-fertilization beyond a certain distance.
These values are not directly transposable to determine the rate of fertilization to be expected from a plot to the outer plots (outflow). Experiments and analysis of samples on plots placed near the source of pollen from genetically modified plants provide additional information on this aspect. The measurement of varietal impurity rates in the samples makes it possible to evaluate the effectiveness of the isolation distances on the diffusion of pollen from the parcel considered towards the neighboring parcels. These are relevant indicators of gene flow from the experimental plot of transgenic plants (outflow).
Analyzes of this type have been conducted, particularly in France, in the case of different species such as maize , rapeseed and beetroot.
Discussions and recommendations
- For the environment
With respect to the environmental impact of herbicides, the limited areas of experimental cultures suggest that such impact is limited in time and space. In addition, if GM plants are new, the herbicides considered are not, and are already authorized for other uses, and have therefore been the subject of a toxicological and ecotoxicological evaluation.
Du fait de leur nombre et de leurs surfaces, les cultures expérimentales d’OGM résistants à des ravageurs ne sont pas susceptibles d’exercer une pression de sélection notable, et de suffisamment longue durée, pour favoriser la sélection de populations de ravageurs résistants. Des effets non intentionnels liés à la possible prolifération de ravageurs résistants, du fait de ces cultures expérimentales, ne sont donc pas à craindre. On rappellera que l’une des principales mesures de gestion du risque d’émergence de ravageurs résistants, est de préserver des zones refuge (i.e. sans culture d’OGM) de suffisamment grande taille. Dans ce cas, on peut donc considérer que la zone refuge est constituée par les cultures conventionnelles couvrant le territoire national.
The isolation and monitoring constraints surrounding the trials seem to provide some assurance as to the lack of consequences of these genetically modified plant experiments on the environment.
It emerges that under the current conditions in which GM experiments are conducted in the field, the risks appear framed and limited. In addition, if the European Union has common legislation onGMOs, it gives each Member State the opportunity to adopt risk management measures of its own. It should be noted that the isolation and supervision measures of the tests adopted by France are among the most stringent in Europe in terms of isolation and monitoring of trials.
However, the lack of certainty about the potential residual unintended effects should be cautious and warrant a rigorous prior risk assessment. Experiments are an important means of acquiring precise information on the unintended effects of GMOs, and should thus make it possible to evolve, if necessary, the devices in place.
- What about the coexistence of trials with various types of agriculture?
The coexistence of different agricultural models is an issue that currently exists between conventional and organic farming. The very principle of a possible coexistence supposes first of all to recognize a right equal to exist for each of the models freely chosen by one or the other.
Organic farming has opted for non-use (with the exception of veterinary medicines), genetically modified products, or products derived from them, as well as seeds or plants obtained from GMOs (Production rules – Article 6 of Regulation 2092/91 of 24 June 1991 as amended). However, as in the case of plant protection products, which are also considered in the production rules of the same European regulation, this is an obligation of means and not an obligation of results in terms of total absence molecules considered, in the finished products.
It shows that, in the same way that organic farming is able to coexist with a conventional agriculture, using plant protection products, it could be coexistent with agriculture using genetically modified plants, whether genetically modified crops for commercial or experimental purposes. This coexistence, however, supposes that the partners, from the producer to the consumers, clearly recognize that the productions resulting from organic farming are acceptable when they present traces of GMOs whereas the farmer has, in good faith, used seeds sold as conventional. This coexistence also presupposes the definition of precise specifications and tolerance thresholds.
In addition, since genetically modified plant varieties have been authorized at Community level for placing on the market, this question of coexistence is no longer specific to the question of field experimentation of such plants.
- What type of trials admit?
Field experiments must be part of a rigorous process that contributes to the acquisition of essential knowledge, the development of independent and accessible expertise, the control and transparentmonitoring of transgenic plants. In the logic of these ends, we can distinguish a gradation of experiments in two orders.
The first-order tests provide indispensable and new answers to questions that have already been addressed and not solved by confined experiments. It is therefore necessary to subject genetically modified plants to open environmental conditions. These tests bring new elements of understanding. This may involve the design or development of methodologies, the assessment of the agronomic effectiveness of processing events under various agro-climatic conditions, the characterization and assessment of risks, or the production of transgenic material for subsequent tests or for food or toxicology studies. In all cases, the rationale for conducting trials may be supported by results obtained in confined greenhouses or other locations. Experimental approaches and modeling may be coupled whenever complementarity of inputs provides new results or questions. These tests are conducted by both public and private laboratories. The first-order experiments concern level 1, 2 or 3 tests as well. Characterization and evaluation of impacts on the biotic or abiotic environmentmay require experimentation on several hectares for several years. Insofar as these experiments base part of their social legitimacy on the fact that they bring unpublished information, it is essential that the results relating to the risk assessment are widely disseminated, whether they come from the public sector or private. According to some experts, the inability to conduct first-rate tests on the national territory would not be without consequences on the dynamism of French research and its ability to acquire, on its own, data and knowledge.
The second-level experiments are intended to compare the data already acquired with the requirements of the regulations or those of development. They are set up only at level 3 in accordance with the requirements of the European regulations described above.
- How to organize these tests and manage them?
The unintended effects of experimental plots depend on the characteristics of the genetically modified plant as well as factors related to experimental conditions, such as plot size. The precise data on these different effects are not always well known, especially as they depend on different factors. These potential effects depend to a large extent on pollen dispersal in the environment.
Geographical localization is therefore an important factor to consider in the management of field trials, both in terms of the distance of the trial compared to sexually compatible plant species, and the location of the trial. The geographical and reproductive isolation distances currently applied make it possible to limit the risks of accidental presence in conventional seeds to a low level.
With regard to pollen dispersal, it seems logical to consider that the greater the distance, the lower the probability of cross fertilization. However, the data obtained show that beyond a certain distance, the probability of cross-fertilization remains at a very low level. According to the knowledge acquired in most allogamous species, pollen dispersaldecreases, in fact, exponentially at short distance from the source (specific to the species considered) and then stays at a very low level (a few per thousand) over longer distances (specific to the species considered) . It therefore seems difficult to reliably predict the dispersion rate in the case of these longer distances because the decay law can not be determined from the currently available data. In addition, pollen dispersal varies according to the species considered. It depends on its biological characteristics and in particular the size, weight and shape of the pollen. The dispersal of pollen over long distances is subject to variable factors among which the heterogeneity of the landscape ( hedgerows, rivers …) and transport by animals. The lifespan of the pollen and its resulting germination capacity are important parameters to consider as well, since it is not dispersion as such that matters, but the consequences of this dispersion.
All these elements make it possible to consider that the increase in isolation distances currently defined would have little impact.on the reduction of the probability of cross-fertilization between the genetically modified plant and other plants. In addition, available data on dispersal by means other than pollen (seeds, organs allowing vegetative reproduction, etc.) are more fragmentary than those available on pollen. They should be taken into account to validly base new measures limiting long-range dispersal. An effort in this area of research should therefore be encouraged. Similarly, additional attention should be paid to the question of unintended effects that concern the natural heritage (flora and fauna). These effects are currently less well known, less well evaluated and warrant a special effort.
While small plots are often sufficient to obtain some information, large-scale experiments may also be necessary. As the size of the experimental plots also influences the probability of occurrence of unintended effects, a more global management of plots implantation in space is desirable especially when the tests concern several sites.