To be or not to be biosafe : an evaluation of transgenic phosphinothricin-tolerant oilseed rape (Brassica napus L.)

Genetic modification is an additional tool for conventional plant breeding to improve the application and quality of crop plants. No longer hampered by natural crossing barriers, application of genetic modification results in a nearly infinite pool from which genes, after isolation, can be introduced in crop plants. At this moment more and more genetically modified crops are coming on the market and these crops will probably significantly contribute to near-future agriculture. However, since the introduction of transgenic plants in the environment the biosafety regulation of these crops has been discussed and developed. Preceding the release of transgenic plants within existing legal frameworks several stages of containment were passed through and a new regulation concerning biotechnological products was set up. In the process leading to commercialization of transgenic plants, herbicide- tolerant crops, such as phosphinothricin (PPT)-tolerant oilseed rape are at the forefront and the first varieties are at present on the market. This fact played five years ago a decisive role to use this particular trait-crop combination for biosafety studies.The aim of the study described in this thesis was to gain knowledge about and familiarity with transgenic PPT-tolerant oilseed rape in relation to its biosafety. This was made in two ways: 1) scientific data concerning oilseed rape were reviewed and the ecological and toxicological impact of the PPT tolerance transgene and the herbicide PPT was evaluated (Chapters 1 and 2); 2) experiments were performed to investigate whether or not PPT tolerance could be transmitted to intra-specific, inter-specific and inter-generic hybrids and if so, what the fate of the transgene and its expression was in these different genetic backgrounds and in successive selfings and backcrosses (Chapters 3, 4 and 5).Reviewing the taxonomy and cytogenetics of the family of Cruciferae revealed that there were ample possibilities for inter-specific and inter-generic hybridization, either with or without embryo- rescue techniques (Chapter 1). Pollen dispersal by both insects and wind is the main factor through which transgenes in oilseed rape may spread. Gene dispersal from transgenic oilseed rape to its (wild) relatives can not be ruled out. Hybridizations, such as oilseed rape (genome constitution AACC) with B. rapa (AA) and B. juncea (AABB) have been described to occur spontaneously under field conditions.Assuming outcrossing occur, attention should be given to the ecological and toxicological impact of the introduced PPT tolerance transgenes in combination with the use of PPT (Chapter 2). To illustrate the so-called 'transgene-centered approach' in which all characteristics of a particular transgene and its product are assayed the pat and bar transgenes, whose gene products, confer PPT tolerance were reviewed. The use of PPT-tolerant crops in combination with PPT could imply a considerable environmental gain compared to currently used herbicide cocktails. Assuming responsible use of PPT-tolerance in agronomy, the consequences with respect to weediness or spread of this trait are minor. Consumption of unspread transgenic PPT-tolerant plant material containing the bar or pat transgenes and/or the gene product PAT will have no adverse effects. Bar and pat DNA will not differ from any other DNA that passes the digestive tract daily and all data found, indicate that no toxicity or allergenicity of PAT are to be expected. Upon spraying PPT-tolerant plants, PPT or derivatives might be present in food and feed. To date, it is insufficiently clear to what extent consumers are exposed to PPT (metabolites) and what the toxicological impact of such exposure might be. As long as there is not much familiarity with the trait, pre-market evaluation of the levels of PPT metabolites in PPT-tolerant plant food will indicate, which further toxicological data are necessary for safe consumption.Because biological containment cannot be obtained for oilseed rape, studying the transgene transmission and its fate in different genetic backgrounds and over generations can indicate the transgene impact in time in sexual offspring. Only when the transgenes are transmitted and expressed in a predictable, consistent and stable manner in subsequent generations during seed multiplication or subsequent steps in a breeding program, transgenic crops have commercial value. The bar gene was successfully transmitted to intraspecific, inter-specific and inter-generic hybrids. In crosses among independent transgenics of one variety and between transgenics of different varieties, no transgene inactivation was observed (Chapter 3). This is what has been expected, because the phenotypic expression of the transgene in homozygous and hemizygous nature in these transgenics was stable. However, independent from its homozygous or hemizygous nature, infrequent loss of expression of the PPT tolerance transgene was found after selfings and backcrosses of some individual transgenic plants with non-transgenic oilseed rape. Molecular analyses of susceptible plants showed that the transgene was still present. Gene inactivation might be caused by methylation or co-suppression while also somaclonal variation might be one of the mechanisms responsible for a reduced of even a loss of phenotypic expression in later generations. These observations indicate that breeders should test whether selected lines stably express PPT tolerance during subsequent generations as is also required in conventional breeding programs.By inter-specific hybridization between B. rapa (AA) and two transgenic oilseed rape lines, the PPT tolerance transgene was relatively easily transmitted into the F 1 , hybrids and retained active (Chapter 4). During backcrossing, between offspring of the two investigated transgenic lines large differences in transmission frequency of the transgene were noted. The line showing low transmission contained the transgene most probably integrated into a C-genome chromosome and in the line showing high transmission it was probably integrated into a chromosome of the A-genome. Therefore, gene transfer from oilseed rape (AACC) to B. rapa (AA) and B. juncea (AABB) can be limited considerably by integration of the transgene on chromosomes of the C-genome. An alternative approach to prevent gene dispersal through pollen transfer is integration of the transgene into the DNA of plastids, since these organelles are maternally inherited in most plants.The inter-generic crosses between transgenic PPT-tolerant oilseed rape and radish (Raphanus sativus) have no biosafety impact (Chapter 5). Potential spread of transgenes from oilseed rape to radish is negligible, because hybridization can only be accomplished using a modified flower culture method. Hybrids produced small amounts of stainable pollen, but they could not be selfed and backcrosses on radish did not yield any viable seed.In the biosafety assessment of genetically modified plants a distinction between 'biosafety in the narrow sense' and 'biosafety in the broad sense' was proposed (Chapter 6). With respect to 'biosafety in narrow sense', ecological concerns focus on weediness and vertical and horizontal transgene spread and toxicological concerns focus on food safety and consumption. With respect to 'biosafety in the broad sense', concerns also reflect social, ethical and/or economic views related to current agriculture. Regulatory authorities in the UK and the Netherlands tend to consider mainly the 'narrow sense' biosafety questions of transgenic plants. Austria and the Scandinavian EU members take the position that 'broad sense' effects should also include linkage of safety aspects of transgenic plants with criteria such as sustainability, socio-economics and ethics.At the end of 1996 transgenic glyphosate-tolerant soybeans were shipped to Europe, which led to protests. Arguments put forward by environmentalists against these soybeans concern primarily the herbicide, which was already allowed to be applied pre-harvest for wild type soybean (USA) and other crops (The Netherlands). Both application of the herbicide and tolerant plants are approved following Dutch and EU regulations. Permission was given to import, store and process in food these glyphosate-tolerant soybeans. However, this does not mean that all herbicide-tolerant crops are biosafe. When a herbicide tolerance transgene is evaluated to be 'biosafe in the narrow sense', it might still possess undesirable characteristics with respect to its 'biosafety in the broad sense'. For example, a particular transgene confers tolerance to a herbicide with an adverse environmental or toxicological impact, such as bromoxynil or the persistent herbicide chlorsulfuron. Introduction of crops tolerant for such herbicides might stimulate the use of these herbicides. In the cases of bromoxynil and chlorsulfuron it is questionable whether or not this is a benign development due to respectively their toxicity and persistence in the soil for years.The major issues described in this thesis are summarized as follows:- the spread of the PPT tolerance transgene from oilseed rape to (wild) relatives occurs, especially when the transgene is integrated into a chromosome of the A-genome- gene flow from PPT-tolerant oilseed rape to B. rapa (AA) and B. juncea (AABB) can be limited considerably through selection for the presence of the PPT tolerance transgene on one of the chromosomes of the C-genome of oilseed rape or through integration of the transgene into the chloroplast genome- the transgene-centered approach shows that without spraying with PPT, PPT-tolerant crops are ecologically and toxicologically biosafe. Upon spraying, however, it is currently insufficiently clear whether consumers are exposed to what levels of PPT and/or its metabolites. No toxicological data are available of PPT-derived metabolites or how they behave upon food processing- in intra-specific crosses involving PPT-tolerant oilseed rape occasional loss of phenotypic expression of the PPT tolerance was observed. This implies that breeders have to follow time- consuming selection procedures to ensure stable expression of transgenic traits, which are similar to those followed in conventional breeding- because inter-generic hybrids between transgenic oilseed rape and radish are difficult to make and almost sterile, transgenes cannot spread in the environment through radish and these hybrids, therefore, have no impact from a biosafety point of view- in the assessment of the biosafety of genetically modified crops a distinction can be made between 'biosafety in the narrow sense' and 'biosafety in the broad sense'. 'Biosafety in the narrow sense' involves the ecology and toxicology of both release and use of transgenic plants. 'Biosafety in the broad sense' also implies social, ethical and/ or economic aspects of transgenic crops with respect to current agriculture- permission for commercialization of a particular herbicide tolerance-herbicide-crop combination does not create a precedent for other herbicide tolerance-herbicide-crop combinations, but their 'biosafety in the broad sense' should be evaluated as a new case.