|Current Workshop | Overview||Back | Home | Search ||
Transgenes Going Wild?
Risk Assessment of Transgene Introgression from Crops into Wild Relatives
Transgenes going wild? Setting the scene from basic research to regulatory requirements
Many studies have been performed to examine the ability and the frequency of gene flow from genetically modified crops to wild relatives. Several conferences were organised and summarizing reviews have been published specifically dedicated to gene flow and introgression. For many crop-wild species complexes it is almost certain that introgression of GM traits into wild populations will happen. Then it is not so much a question of where and when introgression will happen, but moreover what the consequences will be in view to environmental impacts. Nevertheless the mechanisms of transgene flow are still important to be well understood. Regulatory risk assessment is concerned with questions on direct effects on persistence and invasiveness of wild populations, or indirect effects on non-target organisms as well as on GM crop cultivation and management techniques. Further, risk management and mitigation measures including environmental monitoring are becoming increasingly important in the future. The aim of this talk is to set the scene for the workshop from the perspectives of basic research, regulatory assessment needs and authorization requirements.
Federal Office for Consumer Protection and Food Safety, Mauerstraße 39-41, D-10117 Berlin, Germany - e-mail: Detlef.Bartsch@BVL.BUND.DE
‘The consequences of repetition and randomness: hazard rates of introgression’
The incorporation of crop genes into wild relatives through hybridization and backcrossing is inherently a random process. That implies that many characteristics, such as whether or not a transgene will be permanently established in a wild population within a limited time cannot be predicted with certainty. When there is (infinitely long) repeated hybridization, however, it is certain that eventually this will happen. Thus, the probability of eventual introgression is one, and cannot be used to quantify such a process. Deterministic models fail to capture the large amount of variation that is inherent in the process. The hazard rate provides a measure of risk that takes stochasticity into account: it is the probability that a permanent lineage that contains the transgene is established in the next generation, given that this has not happened yet. The hazard rate indicates periods of high and low risk, and provides a way to examine the effects of life history, fitness changes, and crop management measures on introgression risk.
Tom de Jong
“Is there ancient and recent gene flow from oilseed rape to wild populations of Brassica rapa in western Europe?”
B. rapa (AA) and B. napus (AACC) can easily be crossed to produce an F1 (AAC). Further backcrosses to B. rapa should be easy to detect in nature because of the variable number of extra C-chromosomes. Yet such plants are not present in hybrid populations in the Netherlands and in the UK. This suggests rather strong selection against hybrids which limits introgression. This point is illustrated with a simple model of the accumulation of unpaired C-chromosomes in wild B. rapa populations. The model compares different selection regimes and biased versus unbiased transmission of unpaired chromosomes.
A comparison of C-specific AFLP markers in mixed B. rapa/ B. napus populations (high introgression risk) versus pure B. rapa populations (low introgression risk) showed no differences, suggesting that recent gene flow was negligible. In contrast, chromosome painting suggested that homeologous substitution (from C- to A-chromosomes) has occurred in nature. This event may have occurred anytime during the cultivation of oilseed rape in the Netherlands which is between c. 1500 and the present day.
I discuss these results against the background of other Brassica studies in the UK, Denmark and Canada.
“Over a Decade of Crop Transgenes Gone Wild”
The first report of spontaneous transgenic plants was published more than a decade ago. Presently, there are about 10 cases in which transgenes have been identified in volunteer, feral, and wild free-living plants. There are an equal number of reports of transgenes identified in crop plants for which they were not intended. I review these cases to address how these novel alleles evolve when not directly managed.
Rosemary S. Hails
“What information is requested for assessing the potential of GM plants for their persistence and invasiveness?”
“Which plant characters influence Brassica population persistence?”
Potential escape of transgenes via pollen and/or seed dispersal to wild populations remains a central issue in the discussion about the ecological risks of GM crops. Oilseed rape (Brassica napus) readily crosses with Brassica rapa, a common self-incompatible herb that occurs throughout Europe, and has the potential to produce persistent feral populations after seed dispersal to non-agricultural habitats. Here, we focus on the establishment of feral populations through seeds. In particular, we investigate how temporal and spatial demographic variation affects the population dynamics of different Brassica accessions, using demographic data collected over three consecutive years. In addition, we quantify whether crops have the ability to adapt to natural environments, by comparing fitness components of feral and cultivated Brassica accessions. Thirty experimental populations were established at two different localities using seeds from feral B. rapa and B. napus populations and from B. napus cultivars of different age (n = 30 in total). Demographic data will be used to parameterize integral projection models, which allow quantification of population dynamics across different ecological scenarios (e.g., seed predation, herbivory and disturbance). Here, we present the results of the first year of the demographic experiment. Authors: Elze Hesse and Tom J. de Jong
“Extrapolating empirical resulst in time and space: modelling tool boxes for predicting introgression of transgenes”
“An international trade regulatory nightmare: Transgenes in agricultural crop commodities”
Transgenic cultivars of soybeans and maize (corn) provided 93% and 86% of US acreage, respectively, in 2010, with much of these crops exported to overseas markets. Like regular genes, transgenes can disperse by outcrossing into related non-transgenic cultivars, outcrossing into relatives and by seed admixtures into any proximal seed stock, at any point from the farm through the grain handling and shipping/delivery systems. As a result, Low Level Presence (LLP) or Adventitious Presence (AP) of transgenes is virtually inevitable. With major markets (such as EU) enforcing 'zero
tolerance' of unapproved transgenes in incoming commodities, one might expect major disruptions due to the presence of detected transgenes. However, with a few notable exceptions, international trade has not suffered greatly so far, mainly because most of the transgenes in commerce have international approvals. However, we can predict that trade disruptions due to unapproved transgenes will accelerate in coming years, for several reasons to be discussed at the workshop. Alan McHughen, Botany and Plant Sciences, University of California, Riverside, Ca. USA
“Potential Ecological Impacts Caused By Transgene Flow To Weedy Rice: A Case Study Of Insect-Resistant Traits”
Weedy rice (Oryza sativa f. spontanea) is a noxious weed occurring in tropic and temperate rice planting regions worldwide. This weed can cause significant yield losses (up to 100%) of crop rice (Oryza sativa) when heavily infest the rice fields, in addition to the reduction of rice grain quality. Since weedy rice is conspecific to cultivated rice, frequent crop-to-weed gene flow has been reported based on field surveys and well-designed field experiments during the past ten years, causing the adaptive evolution of weedy rice populations. Gene flow from transgenic rice allows novel traits to spread to the sexually compatible weedy rice. Traits such as resistance to insects may enhance the fitness of weedy rice, but few studies have tested for these impacts under natural field conditions. We created crop-weed F1 hybrids as well as the F2, F3, and F4 hybrid lineages of genetically engineered (GE) rice, using lines with two transgene constructs, cowpea trypsin inhibitor (CpTI) and a Bt transgene linked to CpTI (Bt/CpTI) to test the potential ecological impacts. Experiments conducted in Fuzhou, China, demonstrated that CpTI alone did not significantly affect fecundity, although it reduced herbivory in the fields. In contrast, under the conditions of heavy insect pressure Bt/CpTI conferred up to 79% less insect damage and 47% greater fecundity in F2, F3, and F4 populations relative to non-transgenic controls, and a 44% increase in fecundity relative to the weedy parent. A small fitness cost was detected in F3 and F4 progeny with Bt/CpTI when grown under low insect pressure and direct competition with transgene-negative controls. We conclude that Bt/CpTI transgenes may introgress into the co-occurring weedy rice populations and contribute to greater seed production when target insects are abundant. However, the net fitness benefits that are associated with Bt/CpTI could be ephemeral if insect pressure is lacking, for example, due to widespread planting of Bt rice cultivars that suppress the target insect populations. Further experiments including weedy rice strains with a wider genetic background, more hybrid generations, determined levels of insect pressure, and different transgenes should be conducted to have a more profound understanding of the ecological impacts caused by transgene flow to weedy rice populations.
Bao-Rong Lu, Xingxing Cai, Xiao Yang - Ministry of Education Key Laboratory for Biodiversity and Ecological Engineering, Institute of Biodiversity Science, Fudan University, Handan Road 220, Shanghai 200433, China
Patrick G. Meirmans
“Modelling the chromosome-wide consequences of transgene introgression”
The risk of transgene escape from a crop into a wild relative depends on many different factors. The most attention has been given to the hybridization rate between the crop and the wild relative, and the fitness effect of the transgene in the wild population, but also factors like the size and geographic structure of the recipient population are often deemed important. A comprehensive modeling study to assess the risks of transgene introgression should therefore take all these elements into account. One factor that is often overlooked is that the transgene is not the only gene that will introgress from the crop into the wild relative. In fact, the transgene is embedded in the whole crop genome and is physically linked to many other crop genes, many of which may be under selection in the wild population. Therefore, it is important to look at the consequences of transgene introgression on a chromosome-wide or even genome-wide level. Here, we use an individuel-based modeling approach to study how transgene introgression and selection on linked crop genes affect the size of the chromosomal region that is swept along. We also test the influence of life-history traits such as the mating system, seed dormancy and engineered male sterility. Patrick G. Meirmans, Peter H. van Tienderen
“Fitness effects of transgenes in wild populations: expectations, methods, and interpretation”
Studying the fitness effects of commercially available transgenes in wild populations is challenging, partly because only a few crop-wild systems are available to researchers. I will review general expectations and evaluate case studies of experiments designed to test for costs and benefits of transgenes that confer resistance to insects or pathogens in wild squash, clover, sunflower, and other species.
“Transgene introgression in crop relatives: molecular evidence and mitigation strategies”
Incorporation of crop genes into wild and weedy relative
populations (i.e. introgression) has long been of interest to ecologists and
weed scientists. Potential negative outcomes that result from crop transgene introgression (e.g. extinction of native wild relative
populations; invasive spread by wild or weedy hosts) have not been documented,
and few examples of transgene introgres-
sion exist. However,
molecular evidence of introgression from non-transgenic crops to their
relatives continues to emerge, even for crops deemed low-risk candidates for transgene introgression. We posit that transgene
introgression monitoring and mitigation strategies are war- ranted in cases in
which transgenes are predicted to confer selective
advantages and disadvantages to recipient hosts. The utility and consequences
of such strategies are examined, and future directions provided.
1 Department of PlantSciences, University of Tennessee, Knoxville, TN 37996, USA
2 Agriculture and Agriood Canada, Eastern Cereal and Oilseeds Research Centre, K.W. Neatby Bldg., C.E.F., Ottawa, Ontario, Canada K1AOC6
"Estimating Short-Term and Long-Term Levels of Gene Flow"
Studies of introgression from crop to wild populations often deal with short-term or geographically localized phenomena. However, long-term term patterns of gene flow are also important to understand, as they can provide a valuable background or baseline for expectations of contemporary gene flow. Measuring gene flow on these different time scales involves complementary and partially overlapping sets of methodologies. I discuss commonly used methods for measuring gene flow among populations or species on both deep and shallow time scales, using some empirical examples from wild sunflowers.
Peter van Tienderen
“Hitchhiking of transgenes from crop to wild relatives due to selection on crop (domestication) genes”
The first generations of hybrid formation between crop and wild relative are thought to be crucial for the likelihood of persistent introgression. Initial hybrids may exhibit low or high relative fitness for several reasons. For instance, crop/wild hybrids may have reduced fitness if they belong to diverged evolutionary lineages that are not fully interfertile. Alternatively, crops may contain domestication genes that are beneficial in an agricultural setting (fast growth, high yield under favourable conditions), but detrimental in the wild. However, they could also contain genes that are beneficial, for instance an increased resistance to pathogens or herbivores.
The goals of this presentation are twofold. Using two-locus population genetic models we show how specific assumptions on the fitness effects of a crop gene affect the likelihood of introgression of the gene and neighbouring genes. We find that bottlenecks for introgression are to be expected, with a considerable lag time before a crop (trans-)gene starts to increase. Second we present data on the actual fitness effects of crop genes in a wild background, for our lettuce model system. We performed field experiments with the Michelmore recombinant inbred line (RIL) population of Iceberg Lettuce (L. sativa, var. salinas) with its wild relative Prickly Lettuce (L. serriola). We find that the crop contains a mixture of QTLs that affect fitness of the wild carrier, some positive some negative. We discuss the implications for the likelihood of introgression of a transgene, given the position where it was inserted in the crop genome.
Peter van Tienderen & Yorike Hartman
Institute for Biodiversity & Ecosystem Dynamics, University of Amsterdam
e-mail: firstname.lastname@example.org / email@example.com