Representatives of the developers of modern agricultural biotechnology are proposing a tiered approach for conducting non-target organism risk assessment for genetically modified (GM) plants in Europe. The approach was developed by the Technical Advisory Group of the EuropaBio Plant Biotechnology Unit (http://www.europabio.org/TAG.htm) and complements other international activities to harmonize risk assessment. In the European Union (EU), the principles and methods to be followed in an environmental risk assessment for the placing on the market of GM plants are laid out in Annex II of Directive 2001/18/EC on the deliberate release into the environment of GMOs, Commission Decision 2002/623/EC and Regulation (EC) No. 1829/2003. Additional information is provided in the European Food Safety Authority guidance document of 2004. However, risk assessment for effects to non-target organisms could benefit from further clarification and remains the subject of much discussion in Europe. The industry-wide approach developed by EuropaBio is based on the fundamental steps of risk evaluation, namely hazard and exposure assessment. It follows a structured scheme including assessment planning, product characterization and assessment of hazard/exposure (Tier 0), single high dose and dose response testing (Tier 1), refined hazard characterization and exposure assessment (Tier 2) and further refined risk assessment experiments (Tier 3). An additional tier (Tier 4) was included to reflect the fact that post-market activities such as monitoring are required under Directive 2001/18/EC. The approach is compatible with conditions of commercial release in the EU and around the world.

Environmental risk assessments can provide high confidence of minimal risk by testing theories, “risk hypotheses”, that predict the likelihood of unacceptable harmful events. The creation of risk hypotheses and a plan to test them is called problem formulation. Effective problem formulation seeks to maximize the possibility of detecting effects that indicate potential risk; if such effects are not detected, minimal risk is indicated with high confidence. Two important implications are that artificial test conditions can increase confidence, whereas prescriptive data requirements can reduce confidence (increase uncertainty) if they constrain problem formulation. Poor problem formulation can increase environmental risk because it leads to the collection of superfluous data that may delay or prevent the introduction of environmentally beneficial products.

The scientific organizers of the symposium put much emphasis on the identification and definition of hazard and the potential consequences thereof and three full sessions with a total of 13 presentations encompassing a wide range of related themes were planned for this topic. Unfortunately, one talk had to be cancelled because of illness of the speaker (BM Khadi, India). Some presentations covered conceptual approaches for environmental risk assessment (ERA) of GM plants (problem formulation in the risk assessment framework, familiarity approach, tiered and methodological frameworks, non-target risk assessment) and the use of models in assessing invasiveness and weediness of GM plants. Other presentations highlighted the lessons learned for future ERA from case studies and commercialized GM crops, and from monitoring of unintended releases to the environment. When the moderators of the three sessions came together after the presentations to align their summaries, there was an obvious need to restructure the 12 presentations in a way that allowed for a consistent summarizing discussion. The following new organization of the 12 talks was chosen:(1) Concepts for problem formulation and non-target risk assessment(2) Modeling as a tool for predicting invasiveness of GM plants(3) Case-studies of ERA of large-scale release(4) Lessons learned for ERA from a commercialized GM plant(5) Monitoring of unintended release of Bt maize in Mexico.The new thematic structure facilitates a more in-depth discussion of the presentations related to a specific topic, and the conclusions to be drawn are thus more consistent. Each moderator agreed to take responsibility for summarizing one or more themes and to prepare the respective report.

Recapitulating how genetic modification technology and its agro-food products aroused strong societal opposition in the European Union, this paper demonstrates how this opposition contributed to shape the European regulatory frame on GM crops. More specifically, it describes how this opposition contributed to a de facto moratorium on the commercialization of new GM crop events in the end of the nineties. From this period onwards, the regulatory frame has been continuously revised in order to slow down further erosion of public and market confidence. Various scientific and technical reforms were made to meet societal concerns relating to the safety of GM crops. In this context, the precautionary principle, environmental post-market monitoring and traceability were adopted as ways to cope with scientific uncertainties. Labeling, traceability, co-existence and public information were installed in an attempt to meet the general public request for more information about GM agro-food products, and the specific demand to respect the consumers' and farmers' freedom of choice. Despite these efforts, today, the explicit role of public participation and/or ethical consultation during authorization procedures is at best minimal. Moreover, no legal room was created to progress to an integral sustainability evaluation during market procedures. It remains to be seen whether the recent policy shift towards greater transparency about value judgments, plural viewpoints and scientific uncertainties will be one step forward in integrating ethical concerns more explicitly in risk analysis. As such, the regulatory frame stands open for further interpretation, reflecting in various degrees a continued interplay with societal concerns relating to GM agro-food products. In this regard, both societal concerns and diversely interpreted regulatory criteria can be inferred as signaling a request – and even a quest – to render more explicit the broader-than-scientific dimension of the actual risk analysis.

When cultivating genetically modified varieties, the spontaneous gene flow between crop and wild relatives could be of concern. We analyzed spontaneous hybridization between a transgenic male-sterile line of oilseed rape (Brassica napus, 2n=38, AACC) and, as pollen donors, three European populations of wild radish (Raphanus raphanistrum, 2n=18, Rr,Rr) and a variety of cultivated radish (Raphanus sativus, 2n=18, RR). Seeds showed size and shape dimorphism that correlated to the frequency of hybrids. The offspring were scored morphologically and analyzed using DNA markers (inter-simple sequence repeats) to quantify hybrid frequencies. Seed set ranged from 0.4–1.2 seeds per pod, and 0.02–0.6 seeds per pod were confirmed as hybrids. The frequency of confirmed hybrids differed significantly among populations of R. raphanistrum. In the cross with a French population, all offspring were hybrids; in the cross with a Swiss population, 53% of the offspring were hybrids; and in the cross with a Danish population, only 2% of the offspring were found to be hybrids. The remaining offspring apparently belonged to two groups: the majority was B. napus-like plants, possibly of matromorphic origin, and a minority from the Danish cross seemed to carry fragments of the Raphanus genome. In the cross with a cultivated R. sativus, all offspring were found to be hybrids. This is the first report on spontaneous hybridization between B. napus and R. sativus. Hybrids from all cross-combinations had low pollen fertility (0–15%). If R. raphanistrum occurs where male-sterile B. napus is cultivated, large regional differences in hybridization frequencies between the species could complicate environmental risk assessment of transgenic oilseed rape.

The occurrence of transgenic herbicide-resistant oilseed rape (Brassica napus) in ruderal (non-crop disturbed) areas has not been investigated previously in Canada. The primary objective of this study was to document their occurrence in two main ruderal areas (along railways and roads) in the province of Saskatchewan, where half of all oilseed rape is grown, and at the port of Vancouver, British Columbia on the west coast of Canada, where most oilseed rape destined for export is transported by rail. During the 2005 growing season, leaf samples of oilseed rape plants were collected at randomly-selected sites along railways and roads across Saskatchewan ecoregions and at Vancouver; infestation area, density, and plant height of oilseed rape were measured at each site. The presence of the glyphosate and glufosinate resistance traits was determined using test strips. The infestation area of oilseed rape, averaged across 155 sampled sites in the Saskatchewan survey, was markedly smaller in populations along railways than roads; in contrast, infestation area averaged across 54 sites in the Vancouver survey was greater for populations along railways than roads. In both surveys, mean plant density was greater for populations found along railways than roads. Two-thirds of oilseed rape plants sampled across Saskatchewan ecoregions and at Vancouver were transgenic, although the relative proportion of plants with the glyphosate or glufosinate resistance trait varied between surveys. Frequency of occurrence of transgenic plants in ruderal areas was similar to the proportion of the oilseed rape area planted with transgenic cultivars in the recent preceding years. A single transgenic B. rapa x B. napus hybrid was found along a road in Vancouver, confirming the relatively high probability of hybridization between these two Brassica species. With current control measures, transgenic oilseed rape populations may persist and spread in these ruderal areas.

Modeling pollen dispersal to predict cross-pollination is of great importance for the ongoing discussion of adventitious presence of genetically modified material in food and feed. Two different modeling approaches for pollen dispersal were used to simulate two years of data for the rate of cross-pollination of non-GM maize (Zea mays (L.)) fields by pollen from a central 1 ha transgenic field. The models combine the processes of wind pollen dispersal (transport) and pollen competition. Both models used for the simulation of pollen dispersal were Lagrangian approaches: a stochastic particle Lagrange model and a Lagrangian transfer function model. Both modeling approaches proved to be appropriate for the simulation of the cross-pollination rates. However, model performance differed significantly between years. We considered different complexity in meteorological input data. Predictions compare well with experimental results for all simplification steps, except that systematic deviations occurred when only main wind direction was used. Concluding, it can be pointed out that both models might be adapted to other pollen dispersal experiments of different crops and plot sizes, when wind direction statistics are available. However, calibration of certain model parameters is necessary.

Brassica rapa grows as a wild and weedy species throughout the world and is the most likely recipient of transgenes from GM oilseed rape. For transgene introgression to occur, the critical step which must be realized, is the formation of an F1 hybrid. Concerns exist that hybrid populations could be more vigorous and competitive compared to the parental species. This study examines the effect of simulated herbivory and interspecific competition on the vegetative and reproductive performance of non-transgenic F1 hybrids and their parental lines. Several vegetative and reproductive performance measures were used to determine the effect of simulated herbivory and competition on the Brassica lines, including leaf length and biomass for herbivory and seedling height and biomass for competition. For defoliation experiments, B. rapa showed little response in terms of leaf length but B. napus and the F1 hybrid responded negatively. Brassica rapa showed elevated biomass responses, but B. napus and the hybrid demonstrated negative responses to defoliation. Defoliation at the cotyledon stage had a slight effect upon final biomass with the F1 hybrid performing significantly worse than B. napus, although seed counts were not significantly different. For the series of competition experiments, hybrids seemed to be more similar to B. rapa in terms of early seedling growth and reproductive measures. The underperformance of hybrid plants when challenged by herbivory and competition, could potentially decrease survivorship and explain the rarity of hybrids in field surveys. However, should transgene introgression occur, the dynamics of hybrids could change radically thus increasing the risk of gene flow from a transgenic oilseed rape crop to the wild recipient.

Repeated monitoring for escaped transgenic crop plants is sometimes necessary, especially in cases when the crop has not been approved for release into the environment. Transgenic oilseed rape (Brassica napus) was detected along roadsides in central Japan in a previous study. The goal of the current study was to monitor the distribution of transgenic oilseed rape and occurrence of hybridization of transgenic B. napus with feral populations of its closely related species (B. rapa and B. juncea) in the west of Japan in 2005. The progenies of 50 B. napus, 82 B. rapa and 283 B. juncea maternal plants from 95 sampling sites in seven port areas were screened for herbicide-resistance. Transgenic herbicide-resistant seeds were detected from 12 B. napus maternal plants growing at seven sampling sites in two port areas. A portion of the progeny from two transgenic B. napus plants had both glyphosate-resistance and glufosinate-resistance transgenes. Therefore, two types of transgenic B. napus plants are likely to have outcrossed with each other, since the double-herbicide-resistant transgenic strain of oilseed rape has not been developed intentionally for commercial purposes. As found in the previous study, no transgenic seeds were detected from B. rapa or B. juncea, and more extensive sampling is needed to determine whether introgression into these wild species has occurred.

Natural out-crossing rates were evaluated for conventional soybeans (Glycine max (L.) Merr.) cultivated adjacent to genetically modified (GM) glyphosate-tolerant soybeans under field conditions during a four-year period in Japan. A total of 107 846 progeny of 2772 plants harvested from conventional varieties were screened for glyphosate herbicide tolerance. The highest out-crossing rates, 0.19% in 2001 and 0.16% in 2002, were observed in adjacent rows 0.7 m from the pollen source. The highest rate in 2004 was 0.052%, which was observed at 2.1 m. No out-crossing was observed in the rows 10.5 m from the pollen source over the four-year period. The farthest distances between receptor and pollen source at which out-crossing was observed were 7 m in 2001, 2.8 m in 2002, and 3.5 m in 2004. The greatest airborne pollen density during the flowering period, determined by Durham pollen samplers located between the rows of each variety, was 0.368 grains.cm-2.day-1, with the average value at 0.18 grains.cm-2.day-1, indicating that the possibility of out-crossing by wind is minimal. Thrips species and predatory Hemiptera visited the soybean flowers more frequently during the four-year period than any other common pollinators, such as bees.

The release of genetically engineered apple trees raises the question of their potential environmental impact, and the transfer via pollen of transgenes to cross-compatible cultivars of Malus domestica and Malus species is deemed to be the greatest source for environmental exposure. The hybrid TNR 31-35, a descendant of Malus sieversii var. sieversii f. niedzwetzkyana, carrying a homozygous, dominant gene responsible for red pigmentation in all plant parts, was used to assess gene flow in an apple scion repository of genetic resources. The red pigmentation provides a morphological marker that enables large-scale evaluation of cross-fertilization under natural conditions. In two consecutive years, 60 and then 56 apple trees of 38 different Malus domestica cultivars were selected to serve as pollen-receptor trees. In these two years, 6876 and then 5513 seeds, respectively, were gathered from pollen-receptor trees located at different distances, 2–100 m from 15 pollen-dispenser trees. In total, 11 797 seedlings were examined. An average of 1.8% and 1.4%, respectively, of all seedlings obtained showed red-colored leaves. Considering both years of sampling, 69% of the seeds fertilized by TNR 31-35 were found at less than 10 m from the nearest pollen-dispenser tree. Almost 91% of all seeds fertilized by TNR 31-35 were found at less than 60 m from the nearest pollen-dispenser tree, which is equal to 30 adjacent trees along the row. In this study, pollen was dispersed at least 104 m. After phenotypical evaluation, seedlings selected as red-colored were investigated by simple sequence repeat (SSR) analysis. Each seedling was tested with at least one heteromorphic SSR-marker, which allows the verification of TNR 31-35 as the male parent. All but four seedlings showed one allele specific for the appropriate fruiting tree and the second allele specific for the pollen-dispenser TNR 31-35.

Over the past decades, genes of Bacillus thuringiensis var. kurstaki (Berliner) (Bt) coding for protein toxins have been engineered into maize for protection against the European Corn Borer (Ostrinia nubilalis (Hbn.)). However, these transgenic plants may have an impact on non-target organisms. In particular, a potential hazard was identified for non-target lepidopteran larvae, if they consume Bt maize pollen on their host plants. Risk can be defined as a function of the effect of an event (hazard) and the likelihood of this event occurring. Although data on toxicity (hazard) are available from many lab and field studies, knowledge about the environmental exposure of European lepidopteran larvae is incomplete at the population level. Therefore we studied the distribution of small tortoiseshell caterpillars (Aglais urticae (L.)) and its host plant in an agricultural landscape in Germany, to estimate the potential population exposure to maize pollen. The results showed that larvae of the small tortoiseshell developed primarily on freshly sprouted nettle stands (Urtica dioica (L.)) in field margins, rather than adjacent to hedges and groves. However, the main distribution was at margins of cereal (non-maize) fields, where 70% of all larvae were found. This may be due the fact that cereals covered 54% of the survey area, while maize only covered 6.1%. On the other hand, maize fields seem so show higher food plant densities than cereal crops. The results must be interpreted carefully, as the data basis of the present study is very small, and the situation can vary between years due to crop rotation or other changes in agricultural practices. Therefore it is still questionable whether the small tortoiseshell is significantly exposed to maize pollen. For a conclusive risk assessment, more replications and surveys of larger areas in different intensively managed agricultural landscapes over several years are needed.

The objectives of the present study were (1) to investigate the qualitative composition of rhizodeposits leached from soils cropped with non-transgenic and genetically modified (GM) potatoes, and disclose if there were GM-specific modifications in potato rhizodeposition, and (2) to compare these results with conventional bulk parameters of microbial activity in soil. We have raised potatoes from a non-transgenic line (Solanum tuberosum L. cv. Désirée) and three GM lines, which expressed a gene for the resistance to kanamycin (DLH 9000) and a gene for T4 lysozyme (DL10 and DL12). A sandy soil placed in 340 cm3-“CombiSart” containers was used, from which the rhizodeposit was leached after a six-week growth period. The freeze-dried leachates were analyzed by pyrolysis-field ionization mass spectrometry (Py-FIMS). The Py-FI mass spectra gave detailed molecular-chemical information about the composition of leachates, indicating that the potato growth generally altered the composition of the soil solution. Moreover, a principal component analysis of the mass spectra showed differences between the leachates from the non-transgenic parent line and the GM potatoes as well as among the latter group. However, these differences in molecular composition could not be assigned to the release of T4-lysozyme into soil. Dehydrogenase activity and substrate-induced soil respiration as more common bulk parameters of soil microbial activity failed to disclose any significant effects of the various potatoes grown. The limitations of the described rhizodeposit leaching and analysis for risk assessment of GM potato cropping under field conditions are discussed critically. However, it could be concluded that the Py-FI mass spectrometric “fingerprint” can be developed as a fast, comprehensive, highly sensitive and reproducible analytical approach to discern any effects GM-crops may exert on soil ecological parameters.

This study deals with hybrids between Aegilops geniculata and bread wheat, Triticum aestivum L., detected in two Ae. geniculata populations in the “Meseta Central”, Spain’s central plateau where wheat is a major crop. Morphological traits and pentaploid chromosome numbers were used to identify hybrids in 2004 and 2005. The frequency of hybridization under natural conditions was calculated for one of these populations. Six hybrids were detected from a sample of 3158 seeds from this population. The hybrids were mainly sterile but seed set can be found. This study extends the range of documented hybridization to include central Spain. The potential risks associated with natural hybridization in the context of transgenic wheat cultivation are discussed.

We have assessed the utility of morphological and microsatellite markers for tracing field hybridization between Lolium multiflorum and Lolium perenne in cereal-enclosed gene flow plots. The presence of awns on the inflorescence of F1 hybrids was found to be a reliable, but underscoring, indicator of L. multiflorum paternity in L. perenne derived seed as determined by inheritance of species-specific alleles at the microsatellite locus ‘H01 H06’ in these progeny. A positive correlation was evident in the experimental treatment between the number of pollen donor plants in a given plot and the frequency of hybrid F1 seed harvested from pollen receptor plants in that plot. These experiments have established the utility of naturally occurring heritable markers for the measurement of gene flow rates in field Ryegrass populations, with particular significance for risk assessment modeling of potential gene flow from transgenic grass cultivars.