A TIERED APPROACH TO ASSESSING RISK OF GE PLANTS TO NON-TARGET ORGANISMS
Thomas E. Nickson, Monica Garcia-Alonso & Francesca Tencalla
Recently, several companies, members of the Plant Biotechnology Unit of the "Bioindustries" association EuropaBio (http://www.europabio.org), published an approach to assessing the risks of biotechnology-derived (genetically engineered, GE) crops to non-target organisms1. The paper intended to clarify the position of industry with regard to non-target organism risk assessment in the context of the European Union (EU). This is an important publication because it reflects a consensus among a group of scientists who have much experience in evaluating the risks of GE crops, and has broader applicability beyond the EU. It is also important because it reflects a desire on the part of the GE crop industry to promote better understanding and harmonization of risk assessment globally. The key points made in the publication are summarized below.
Scientifically sound risk assessment is critical to regulatory decision-making in general. In the case of GE crops, environmental risk assessment is one of the key elements that must be completed before authorization for commercialization can be granted. Since the first GE crops were introduced into the environment, broad agreement has been reached that certain principles underpin an appropriate environmental risk assessment: science-based consideration of all available information, including expert judgment, case-by-case assessment, comparison of risks to an appropriate conventional (non-GE) counterpart crop, and a recursive process (risk conclusions are re-evaluated based on new information). These principles have been noted in publications from the European Union (EU), the Organization for Economic Cooperation and Development (OECD), and most recently the Cartagena Protocol on Biosafety (CPB) under the Convention on Biological Diversity. Another important principle developed by the OECD and referenced in environmental risk assessment guidance from the EU is the concept of "familiarity," which states that experience with the trait, the crop, the intended receiving environment, and the interactions among these, is valuable in organizing and initiating the environmental risk assessment.
Since risk is a function of both the potential harm (hazard) and the likelihood of occurrence (exposure), environmental risk assessment examines the properties of the GE crop to understand what harm is possible, how likely it is to be realized, and consequences of the harm should it occur. Based on the requirements of EU legislation and on various approaches published in peer-reviewed literature, the EuropaBio paper proposes a tiered approach to environmental risk assessment, 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). A Tier 4 has been introduced to account for the EU regulatory requirement for post-commercial monitoring (case-specific or general surveillance) of GE crops. Tier 4 was developed in this publication to encompass post-marketing activities, rather than the refinements of risk that occur in the earlier Tiers. Tiering enables the environmental risk assessment to progress in a structured manner to generate reasonable risk hypotheses that are relevant for decision-making. Tiering also allows risk assessors to clearly link measurement endpoints with specific questions that must be answered and manages the complexity inherent in ecological sciences.
Tier 0, also referred to as "problem formulation," is the step in which all available information is gathered and assessment planning occurs. This structuring of information is supplemented by a detailed plant characterization–an exhaustive experimental approach to look at the properties of the GE crop to determine what, if any, differences exist that would be relevant to an environmental risk assessment. Tier 0 experiments include detailed molecular, expression, compositional, and phenotypic analyses. These studies form part of the basic set of data required by EU legislation and must be provided by all applicants. Molecular and expression analyses confirm that the gene(s) of interest are stably inserted and expressed in a manner consistent with the intent of the product concept. Compositional and agronomic/phenotypic analyses enable risk assessors to determine what, if any, unintended changes have occurred that could be relevant to the nutritional quality, toxicity, and potential weediness/invasiveness of the GE crop, including the foods and feeds derived from it. The experimental basis of the Tier 0 compositional and phenotypic analyses is a comparison with the traditional crop. Importantly, Tier 0 forms a weight of evidence that allows risk assessors to determine what changes or properties of the GE crop need more detailed and focused assessment. Use of an appropriate experimental design and a familiar basis for comparison–the traditional counterpart–provides the basis for determining the significance of the experimental result, i.e., whether results are within or outside the range of expected values, and if meaningful changes are consistent with a change toward increased weediness, reduced nutritional quality, or increased toxicity. All GE crops undergo Tier 0 examination, which may take several years to complete.
Often, a risk assessor has sufficient information at Tier 0 to complete the comparative assessment and reach a conclusion of "no increased risk to the environment compared to the conventional counterpart." Nevertheless, additional valuable information is often collected as part of basic research programs whose results are reviewed in light of the original risk conclusions.
Tier 1 experiments have several key characteristics: 1) they are conducted in laboratories with several single species, typically using validated/well-recognized protocols; and 2) they use single or multiple doses (dose response) of pure protein or plant tissues that are determined based on conservative assumptions about the expected environmental exposure. Typically a Tier 1 non-target organism environmental risk assessment comprises hazard tests on a series of selected species that broadly represent ecological functions and communities of organisms expected in an agricultural field. The functions include pollinators, predators, parasitoids, and detritivores, and it is common to use specific taxa logically associated with the crop and the gene of interest, e.g., tests with non-target coleopterans have been used to assess coleopteran active (Cry3) Bt products. Dosing in Tier 1 experiments is targeted to exceed the expected environmental concentration to account for uncertainty or lack of realism. Importantly, when much is known about the trait/protein at the beginning of the non-target organism environmental risk assessment, a risk assessor has more certainty in deciding the appropriateness of the test species and test materials used in the Tier 1 experiment. An important element of Tier 1 experiments is that they are designed to be conservative to avoid false negatives (Type II errors) and address the uncertainty associated with their lack of realism. An appropriately robust Tier 1 protocol enables a risk assessor and registrant to deal with the very real problem that one simply cannot test every non-target organism, and avoids the uncertainty created by field tests that do not test a well-defined risk hypothesis.
Tiers 2 and 3 take specific risk hypotheses developed from results in Tiers 0 and 1 and enable the risk assessor to refine estimates of hazard and exposure in either laboratory and/or field experiments. Tier 2 is distinguished from Tier 1 experiments by the element of realism in the experimental design and the fact that the hypothesis was generated as a result of a lower tier study. For example, more refined exposure estimates might be used with test materials like pollen to more precisely assess hazard to an organism likely to be exposed. Tier 3 experiments are characterized by more environmental realism and also complexity. A field and semi-field experiment to assess impacts to the abundance of a specific species or family would be an example of a Tier 3 study.
In common across all Tiers is that, when a conclusion of negligible risk is made with acceptable certainty, then higher tiered experiments are unnecessary. Conversely, when a conclusion of acceptable risk cannot be made based on information from lower tiered experiments, higher tiered experiments must be conducted to refine the risk conclusions.
In conclusion, the biotech industry has published a consensus approach to assessing the risks to non-target organisms from GE crops. The foundation of the approach is based on consensus principles previously developed for environmental risk assessment and utilizes a tiered approach. Tiering in the non-target organism environmental risk assessment allows a risk assessor to progress in a scientific and structured manner. In addition, tiering also enables the collection of only the information that is needed to utilize well-constructed, reasonable risk hypotheses and ultimately characterize risk. Progression through the Tiers is straightforward when starting with a detailed "Problem Formulation" and Tier 0 "Plant Characterization." This approach balances the scientific curiosity that often leads to the proposal for field studies intended to assess risks to non-target organisms with the regulatory requirement to answer only risk-related questions, which can only be done through specific hypothesis generation and testing. The plant biotechnology industry believes that this approach is robust, sufficiently protective of the environment without creating unnecessary costs, and will serve as a solid basis for future work in harmonizing environmental risk assessment practices around the EU and world.
Reference
Garcia-Alonso M, Jacobs E, Raybould A, Nickson TE, Sowig P, Willekens H, Van der Kouwe P, Layton R, Amijee F, Fuentes AM and Tencalla F. 2006. A tiered system for assessing the risk of genetically modified plants to non-target organisms. Environ. Biosafety Res. 5. 57-65
Thomas E. Nickson (Monsanto, USA)
thomas.nickson@monsanto.com
Monica Garcia-Alonso (Syngenta, UK)
Francesca Tencalla (Monsanto, Europe)
With coauthors: Alan Raybould1, Erik Jacobs2, Peter Sowig3, Hilde Willekens4, Pier Van der Kouwe5, Raymond Layton6, Firoz Amijee7, and Angel M. Fuentes7
1 Syngenta, United Kingdom; 2 Monsanto Europe, Belgium; 3 Bayer CropScience, Germany; 4 Syngenta International AG, Belgium; 5 Bayer CropScience, France; 6 Pioneer Hi-Bred International, USA; 7 Pioneer Overseas Corporation, Belgium.
MINICHROMOSOMES: THE NEXT GENERATION TECHNOLOGY FOR PLANT GENETIC ENGINEERING
Weichang Yu & James A. Birchler
Genetic transformation occurs frequently in nature in prokaryotes. The transfer of genes from one organism to another is termed horizontal gene transfer1. For example, bacteria can acquire virulence factors, as well as antibiotic resistance genes, which may lead to the breakdown of the efficacy of antibiotics. Horizontal gene transfers are rare in higher eukaryotes, but years ago scientists found that a pathogenic bacterium, Agrobacterium, could transfer genes from its genome to its plant hosts, where expression of the transferred genes caused crown gall disease2.
The development of Agrobacterium-mediated genetic transformation, and direct transformation by biolistics, i.e., the high velocity delivery of DNA attached to metal particles, led to the first generation of transgenic plants and the rapid application of this technology to crop improvement.
Genetic engineering as a driving force for modern agriculture
Genetic engineering is a powerful tool for improving crop quality and productivity, and reducing labor and resource utilization of farming. For example, farmers saved up to an estimated 60% of costs for pest control by growing Bt (insect resistance) cotton in certain regions of the US in 1997, according to a USDA report3. The reduction of pesticide spray has other safety benefits for both the environment and humans. Because of these and other benefits, the adoption of three primary GE crops (corn, cotton and soybean) for two traits, HT (herbicide tolerance) and Bt, has dramatically increased since the first commercial introduction in 1996. According to recent USDA survey data (http://www.ers.usda.gov/Data/biotechcrops/), the acreage of GE soybeans reached 89% in the US in 2006, followed by 83% for cotton, and 61% for corn. Although most of the current GE crops are either herbicide tolerant or insect resistant, there is a trend toward a small but steadily increasing percentage of GE crops with stacked genes (containing both HT plus Bt traits). For example, 39% of GE cotton planted last year were varieties with stacked genes, compared with only 20% in 2000; 15% of GE corn had stacked genes, compared with 1% in 2000.
The next generation – engineered minichromosomes
Despite many successes, genetic engineering does have limitations. For example, the technology used to create stacked gene varieties is not efficient. When stacking genes in inbred lines, the two genes located on different chromosomes will segregate at a ratio of 1:2:1 following first a genetic cross, and then self-pollinate to make each homozygous. This means that only one-sixteenth of the progeny will maintain the two homozygous stacked transgenes in the next generation. Gene stacking or pyramiding is also widely used in conventional crop breeding programs to combine multiple characteristic traits.
Another difficulty is linkage drag, which refers to the reduction in fitness of a cultivar due to deleterious genes introduced along with the beneficial gene during backcrossing. Linkage drag may occur because genes closely linked to the desired genes or transgenes are usually difficult to recombine during the transfer from one line to another. Many generations of backcrosses are usually required to introduce a good character gene to an elite cultivar without linkage drag. This problem has created a bottleneck that limits the advancement of plant genetic engineering technology from the first generation of GE crops, characterized as 'input traits', to the next generation, characterized as 'output traits'4.
In a recent paper5, Yu et al. describe an engineered minichromosome system that can be used as a genetic engineering platform for the expression of an almost unlimited number of genes that can be inherited together as an independent unit. This engineered minichromosome technology should remove the bottleneck described above, and may lead to the second generation of genetic engineered crops.
Chromosome engineering and minichromosomes
Minichromosomes have been produced in yeast and mammalian cells through either de novo construction using the minimum constituent parts of chromosomes or telomere-mediated chromosomal truncation of existing chromosomes6,7 (Fig. 1).
