'GM-gene-deletor': fused loxP-FRT recognition sequences dramatically improve the efficiency of FLP or CRE recombinase on transgene excision from pollen and seed of tobacco plants

Pollen- and seed-mediated transgene flow is a concern in plant biotechnology. We report here a highly efficient 'genetically modified (GM)-gene-deletor' system to remove all functional transgenes from pollen, seed or both. With the three pollen- and/or seed-specific gene promoters tested, the phage CRE/ loxP or yeast FLP/ FRT system alone was inefficient in excising transgenes from tobacco pollen and/or seed, with no transgenic event having 100% efficiency. When loxP-FRT fusion sequences were used as recognition sites, simultaneous expression of both FLP and CRE reduced the average excision efficiency, but the expression of FLP or CRE alone increased the average excision efficiency, with many transgenic events being 100% efficient based on more than 25 000 T₁ progeny examined per event. The 'GM-gene-deletor' reported here may be used to produce 'non-transgenic' pollen and/or seed from transgenic plants and to provide a bioconfinement tool for transgenic crops and perennials, with special applicability towards vegetatively propagated plants and trees.     

Transgene Containment Using Cytokinin-Reversible Male Sterility in Constitutive, Gibberellic Acid–Insensitive (Δgai) Transgenic Tobacco

Mechanisms are needed to prevent gene flow from transgenic crops, and the later establishment of these transgenes in populations of other varieties, weeds, or wild relatives. Such prevention can be achieved by containing the transgene within a crop, and then mitigating the effects of the inherent leakage and unidirectionality of containment systems. Mitigation lowers the fitness of recipients below that of the wild-type so that transgenes cannot spread. Transplastomic and male-sterility systems suppress transgene outflow, but not the influx of pollen from relatives, requiring mitigation. The Arabidopsis thaliana Δgai (gibberellic acid–insensitive) gene, driven by its own promoter, induced male sterility in transgenic tobacco (Nicotiana tabacum), which is chemically reversible by kinetin applications. Female reproduction was not affected. Kinetin-treated sterile hemizygous and homozygous dwarf tobacco produced viable pollen, becoming self-fertile with copious viable seed, restoring the small amount of seed production needed for such a crop. Thus, Δgai, under its endogenous promoter, can be used as a containment mechanism to prevent transgene outflow. This application is in addition to the previously described highly effective role of Δgai as a dwarfing mitigator gene, which renders the rare transgenic tobacco hybrids unfit and unable to compete with the wild-type in the mixed cultures. Δgai is unique in that it can be used both to prevent transgene outflow and to mitigate the flow should containment fail or should gene influx occur, a dual role for the gene, not previously reported.     

Transgene excision in pollen using a codon optimized serine resolvase CinH-<Emphasis Type="Italic">RS2</Emphasis> site-specific recombination system

Transgene escape, a major environmental and regulatory concern in transgenic crop cultivation, could be alleviated by removing transgenes from pollen, the most frequent vector for transgene flow. A transgene excision vector containing a codon optimized serine resolvase CinH recombinase (CinH) and its recognition sites RS2 were constructed and transformed into tobacco (Nicotiana tabacum cv. Xanthi). CinH recombinase recognized 119 bp of nucleic acid sequences, RS2, in pollen and excised the transgene flanked by the RS2 sites. In this system, the pollen-specific LAT52 promoter from tomato was employed to control the expression of CinH recombinase. Loss of expression of a green fluorescent protein (GFP) gene under the control of the LAT59 promoter from tomato was used as an indicator of transgene excision. Efficiency of transgene excision from pollen was determined by flow cytometry (FCM)-based pollen screening. While a transgenic event in the absence of CinH recombinase contained about 70% of GFP-synthesizing pollen, three single-copy transgene events contained less than 1% of GFP-synthesizing pollen based on 30,000 pollen grains analyzed per event. This suggests that CinH-RS2 recombination system could be effectively utilized for transgene biocontainment.     

Evaluating genetic containment strategies for transgenic plants

One of the primary concerns about genetically engineered crop plants is that they will hybridize with wild relatives, permitting the transgene to escape into the environment. The likelihood that a transgene will spread in the environment depends on its potential fitness impact. The fitness conferred by various transgenes to crop and/or wild-type hybrids has been evaluated in several species. Different strategies have been developed for reducing the probability and impact of gene flow, including physical separation from wild relatives and genetic engineering. Mathematical models and empirical experimental evidence suggest that genetic approaches have the potential to effectively prevent transgenes from incorporating into wild relatives and becoming established in wild populations that are not reproductively isolated from genetically engineered crops.     

Recent advances in development of marker-free transgenic plants: Regulation and biosafety concern

During the efficient genetic transformation of plants with the gene of interest, some selectable marker genes are also used in order to identify the transgenic plant cells or tissues. Usually, antibiotic- or herbicide-selective agents and their corresponding resistance genes are used to introduce economically valuable genes into crop plants. From the biosafety authority and consumer viewpoints, the presence of selectable marker genes in released transgenic crops may be transferred to weeds or pathogenic microorganisms in the gastrointestinal tract or soil, making them resistant to treatment with herbicides or antibiotics, respectively. Sexual crossing also raises the problem of transgene expression because redundancy of transgenes in the genome may trigger homology-dependent gene silencing. The future potential of transgenic technologies for crop improvement depends greatly on our abilities to engineer stable expression of multiple transgenic traits in a predictable fashion and to prevent the transfer of undesirable transgenic material to non-transgenic crops and related species. Therefore, it is now essential to develop an efficient marker-free transgenic system. These considerations underline the development of various approaches designed to facilitate timely elimination of transgenes when their function is no longer needed. Due to the limiting number of available selectable marker genes, in future the stacking of transgenes will be increasingly desirable. The production of marker-free transgenic plants is now a critical requisite for their commercial deployment and also for engineering multiple and complex trait. Here we describe the current technologies to eliminate the selectable marker genes (SMG) in order to develop marker-free transgenic plants and also discuss the regulation and biosafety concern of genetically modified (GM) crops.     

Measuring gene flow from two birdsfoot trefoil (Lotus corniculatus) field trials using transgenes as tracer markers

Genetic engineering is becoming a useful tool in the improvement of plants but concern has been expressed about the potential environmental risks of releasing genetically modified (GM) organisms into the environment. Attention has focused on pollen dispersal as a major issue in the risk assessment of transgenic crop plants. In this study, pollen-mediated dispersal of transgenes via cross-fertilization was examined. Plants of Lotus corniculatus L. transformed with either the Escherichia coli asparagine synthetase gene asnA or the beta-glucuronidase gene uidA, were used as the pollen donor. Nontransgenic plants belonging to the species L. corniculatus L., L. tenuis Waldst. and Kit. ex Willd, and L. pedunculatus Cav., were utilized as recipients. Two experimental fields were established in two areas of central Italy. Plants carrying the uidA gene were partially sterile, therefore only the asnA gene was used as a tracer marker. No transgene flow between L. corniculatus transformants and the nontransgenic L. tenuis and L. pedunculatus plants was detected. As regards nontransgenic L. corniculatus plants, in one location flow of asnA transgene was detected up to 18 m from the 1.8 m2 donor plot. In the other location, pollen dispersal occurred up to 120 m from the 14 m2 pollinating plot.     

Gene flow in genetically modified wheat

Understanding gene flow in genetically modified (GM) crops is critical to answering questions regarding risk-assessment and the coexistence of GM and non-GM crops. In two field experiments, we tested whether rates of cross-pollination differed between GM and non-GM lines of the predominantly self-pollinating wheat Triticum aestivum. In the first experiment, outcrossing was studied within the field by planting "phytometers" of one line into stands of another line. In the second experiment, outcrossing was studied over distances of 0.5-2.5 m from a central patch of pollen donors to adjacent patches of pollen recipients. Cross-pollination and outcrossing was detected when offspring of a pollen recipient without a particular transgene contained this transgene in heterozygous condition. The GM lines had been produced from the varieties Bobwhite or Frisal and contained Pm3b or chitinase/glucanase transgenes, respectively, in homozygous condition. These transgenes increase plant resistance against pathogenic fungi. Although the overall outcrossing rate in the first experiment was only 3.4%, Bobwhite GM lines containing the Pm3b transgene were six times more likely than non-GM control lines to produce outcrossed offspring. There was additional variation in outcrossing rate among the four GM-lines, presumably due to the different transgene insertion events. Among the pollen donors, the Frisal GM line expressing a chitinase transgene caused more outcrossing than the GM line expressing both a chitinase and a glucanase transgene. In the second experiment, outcrossing after cross-pollination declined from 0.7-0.03% over the test distances of 0.5-2.5 m. Our results suggest that pollen-mediated gene flow between GM and non-GM wheat might only be a concern if it occurs within fields, e.g. due to seed contamination. Methodologically our study demonstrates that outcrossing rates between transgenic and other lines within crops can be assessed using a phytometer approach and that gene-flow distances can be efficiently estimated with population-level PCR analyses.     

Site-specific excisional recombination strategies for elimination of undesirable transgenes from crop plants

A major limitation of crop biotechnology and breeding is the lack of efficient molecular technologies for precise engineering of target genomic loci. While transformation procedures have become routine for a growing number of plant species, the random introduction of complex transgenenic DNA into the plant genome by current methods generates unpredictable effects on both transgene and homologous native gene expression. The risk of transgene transfer into related plant species and consumers is another concern associated with the conventional transformation technologies. Various approaches to avoid or eliminate undesirable transgenes, most notably selectable marker genes used in plant transformation, have recently been developed. These approaches include cotransformation with two independent T-DNAs or plasmid DNAs followed by their subsequent segregation, transposon-mediated DNA elimination, and most recently, attempts to replace bacterial T-DNA borders and selectable marker genes with functional equivalents of plant origin. The use of site-specific recombination to remove undesired DNA from the plant genome and concomitantly, via excision-mediated DNA rearrangement, switch-activate by choice transgenes of agronomical, food or feed quality traits provides a versatile “transgene maintenance and control” strategy that can significantly contribute to the transfer of transgenic laboratory developments into farming practice. This review focuses on recent reports demonstrating the elimination of undesirable transgenes (essentially selectable marker and recombinase genes) from the plant genome and concomitant activation of a silent transgene (e.g., a reporter gene) mediated by different site-specific recombinases driven by constitutive or chemically, environmentally or developmentally regulated promoters. These reports indicate major progress in excision strategies which extends application of the technology from annual, sexually propagated plants towards perennial, woody and vegetatively propagated plants. Current trends and future prospects for optimization of excision-activation machinery and its practical implementation for the generation of transgenic plants and plant products free of undesired genes are discussed.     

Letting the gene out of the bottle: the population genetics of genetically modified crops

Genetically modified (GM) plants are rapidly becoming a common feature of modern agriculture. This transition to engineered crops has been driven by a variety of potential benefits, both economic and ecological. The increase in the use of GM crops has, however, been accompanied by growing concerns regarding their potential impact on the environment. Here, we focus on the escape of transgenes from cultivation via crop x wild hybridization. We begin by reviewing the literature on natural hybridization, with particular reference to gene flow between crop plants and their wild relatives. We further show that natural selection, and not the overall rate of gene flow, is the most important factor governing the spread of favorable alleles. Hence, much of this review focuses on the likely effects of transgenes once they escape. Finally, we consider strategies for transgene containment.     

Establishment of transgenic herbicide-resistant creeping bentgrass (Agrostis stolonifera L.) in nonagronomic habitats

Concerns about genetically modified (GM) crops include transgene flow to compatible wild species and unintended ecological consequences of potential transgene introgression. However, there has been little empirical documentation of establishment and distribution of transgenic plants in wild populations. We present herein the first evidence for escape of transgenes into wild plant populations within the USA; glyphosate-resistant creeping bentgrass (Agrostis stolonifera L.) plants expressing CP4 EPSPS transgenes were found outside of cultivation area in central Oregon. Resident populations of three compatible Agrostis species were sampled in nonagronomic habitats outside the Oregon Department of Agriculture control area designated for test production of glyphosate-resistant creeping bentgrass. CP4 EPSPS protein and the corresponding transgene were found in nine A. stolonifera plants screened from 20,400 samples (0.04 +/- 0.01% SE). CP4 EPSPS-positive plants were located predominantly in mesic habitats downwind and up to 3.8 km beyond the control area perimeter; two plants were found within the USDA Crooked River National Grassland. Spatial distribution and parentage of transgenic plants (as confirmed by analyses of nuclear ITS and chloroplast matK gene trees) suggest that establishment resulted from both pollen-mediated intraspecific hybridizations and from crop seed dispersal. These results demonstrate that transgene flow from short-term production can result in establishment of transgenic plants at multi-kilometre distances from GM source fields or plants. Selective pressure from direct application or drift of glyphosate herbicide could enhance introgression of CP4 EPSPS transgenes and additional establishment. Obligatory outcrossing and vegetative spread could further contribute to persistence of CP4 EPSPS transgenes in wild Agrostis populations, both in the presence or absence of herbicide selection.     

Tissue-specific expression in transgenic maize of four endosperm promoters from maize and rice

The tissue-specific, developmental, and genetic control of four endosperm-active genes was studied via expression of GUS reporter genes in transgenic maize plants. The transgenes included promoters from the maize granule-bound starch synthase (Waxy) gene (zmGBS), a maize 27 kDa zein gene (zmZ27), a rice small subunit ADP-glucose pyrophosphorylase gene (osAGP) and the rice glutelin 1 gene (osGT1). Most plants had a transgene expression profile similar to that of the endogenous gene: expression in the pollen and endosperm for the zmGBS transgene, and endosperm only for the others. Histological analysis indicated expression initiated at the periphery of the endosperm for zmGBS, zmZ27 and osGT1, while osAGP transgene activity tended to start in the lower portion of the seed. Transgene expression at the RNA level was proportional to GUS activity, and did not influence endogenous gene expression. Genetic analysis showed that there was a positive dosage response with most lines. Activity of the zmGBS transgene was threefold higher in a low starch (shrunken2) genetic background. This effect was not seen with zmZ27 or osGT1 transgenes. The expression of the transgenes is discussed relative to the known behaviour of the endogenous genes, and the developmental programme of the maize endosperm     

Transgene-induced lesion mimic

Lesion mimic, i.e., the spontaneous formation of lesions resembling hypersensitive response (HR) lesions in the absence of a pathogen, is a dramatic phenotype occasionally found to accompany the expression of different, mostly unrelated, transgenes in plants. Recent studies indicated that transgene-induced lesion formation is not a simple case of necrosis, i.e., direct killing of cells by the transgene product, but results from the activation of a programmed cell death (PCD) pathway. Moreover, activation of HR-like cell death by transgene expression is viewed as an important evidence for the existence of a PCD pathway in plants. The study of lesion mimic transgenes is important to our understanding of PCD and the signals that control it in plants. PCD-inducing transgenes may provide clues regarding the different entry points into the cell death pathway, the relationships between the different branches of the pathway (e.g., developmental or environmental), or the different mechanisms involved in its induction or execution. Cell death-inducing transgenes may also be useful in biotechnology. Some lesion mimic transgenes were found to be induced in plants a state of systemic acquired resistance (SAR). These genes can be used in the development of pathogen-resistant crops. Other cell death-inducing transgenes may be used as specific cell ablation tools. Although mainly revealed unintentionally, and at times considered `an adverse phenotype', lesion mimic transgenes should not be ignored because they may prove valuable for studying PCD as well as developing useful traits in different plants and crops.     

Molecular Strategies for Addressing Gene Flow Problems and Their Potential Applications in Abiotic Stress Tolerant Transgenic Plants

Transgenic plant technology provides a powerful tool to improve abiotic stress tolerance of crop plants. However, introgression of stress tolerance genes into weedy relatives may increase the potential for persistence and invasiveness, resulting in undesirable ecological consequences. A variety of gene confinement strategies have been developed to reduce unwanted transgene movement. In this review, we discuss some of these strategies, such as male and female sterility, GeneSafe?, parthenocarpy, chloroplast transformation and gene deletor technologies. In the case of the gene deletor technology, all transgenes from pollen, seeds, fruits or other organs may be eliminated once the transgene functions are no longer needed at the stage when the presence of the transgene becomes a cause for ecological or public concern. The gene deletor and other technologies can be useful to reduce unintended dispersal of stress tolerance genes and thus may facilitate commercialization of transgenic crops with enhanced tolerance to abiotic stresses.     

Molecular strategies for gene containment in transgenic crops

The potential of genetically modified (GM) crops to transfer foreign genes through pollen to related plant species has been cited as an environmental concern. Until more is known concerning the environmental impact of novel genes on indigenous crops and weeds, practical and regulatory considerations will likely require the adoption of gene-containment approaches for future generations of GM crops. Most molecular approaches with potential for controlling gene flow among crops and weeds have thus far focused on maternal inheritance, male sterility, and seed sterility. Several other containment strategies may also prove useful in restricting gene flow, including apomixis (vegetative propagation and asexual seed formation), cleistogamy (self-fertilization without opening of the flower), genome incompatibility, chemical induction/deletion of transgenes, fruit-specific excision of transgenes, and transgenic mitigation (transgenes that compromise fitness in the hybrid). As yet, however, no strategy has proved broadly applicable to all crop species, and a combination of approaches may prove most effective for engineering the next generation of GM crops.     

Engineered selective plant male sterility through pollen‐specific expression of the EcoRI restriction endonuclease

Unintended gene flow from transgenic plants via pollen, seed and vegetative propagation is a regulatory concern because of potential admixture in food and crop systems, as well as hybridization and introgression to wild and weedy relatives. Bioconfinement of transgenic pollen would help address some of these concerns and enable transgenic plant production for several crops where gene flow is an issue. Here, we demonstrate the expression of the restriction endonuclease EcoRI under the control of the tomato pollen‐specific LAT52 promoter is an effective method for generating selective male sterility in Nicotiana tabacum (tobacco). Of nine transgenic events recovered, four events had very high bioconfinement with tightly controlled EcoRI expression in pollen and negligible‐to‐no expression other plant tissues. Transgenic plants had normal morphology wherein vegetative growth and reproductivity were similar to nontransgenic controls. In glasshouse experiments, transgenic lines were hand‐crossed to both male‐sterile and emasculated nontransgenic tobacco varieties. Progeny analysis of 16 000–40 000 seeds per transgenic line demonstrated five lines approached (>99.7%) or attained 100% bioconfinement for one or more generations. Bioconfinement was again demonstrated at or near 100% under field conditions where four transgenic lines were grown in close proximity to male‐sterile tobacco, and 900–2100 seeds per male‐sterile line were analysed for transgenes. Based upon these results, we conclude EcoRI‐driven selective male sterility holds practical potential as a safe and reliable transgene bioconfinement strategy. Given the mechanism of male sterility, this method could be applicable to any plant species.     

A large-scale field study of transgene flow from cultivated rice (Oryza sativa) to common wild rice (O. rufipogon) and barnyard grass (Echinochloa crusgalli)

The introgression of transgenes into wild relatives or weeds through pollen-mediated gene flow is a major concern in environmental risk assessment of transgenic crops. A large-scale (1.3–1.8 ha) rice gene flow study was conducted using transgenic rice containing the bar gene as a pollen donor and Oryza rufipogon as a recipient. There was a high frequency of transgene flow (11%−18%) at 0–1 m, with a steep decline with increasing distance to a detection limit of 0.01% by 250 m. To our knowledge, this is the highest frequency and longest distance of gene flow from transgenic rice to O. rufipogon reported so far. On the basis of these data, an adequate isolation distance from both conventional and transgenic rice should be taken for in situ conservation of common wild rice. Meanwhile, there is no evidence of transgene introgression into barnyard grass, even when it has coexisted with transgenic rice containing the bar gene for five successive years. Thus, the environmental risk of gene flow to this weedy species is of little concern.     

Pollen dispersal in sugar beet production fields

Pollen-mediated gene flow has important implications for biodiversity conservation and for breeders and farmers’ activities. In sugar beet production fields, a few sugar beet bolters can produce pollen as well as be fertilized by wild and weed beet. Since the crop, the wild beets, and the weed beets are the same species and intercross freely, the question of pollen flow is an important issue to determine the potential dispersal of transgenes from field to field and to wild habitats. We report here an experiment to describe pollen dispersal from a small herbicide-resistant sugar beet source towards male sterile target plants located along radiating lines up to 1,200 m away. Individual dispersal functions were inferred from statistical analyses and compared. Pollen limitation, as expected in root-production fields, was confirmed at all the distances from the pollen source. The number of resistant seeds produced by bait plants best fitted a fat-tailed probability distribution curve of pollen grains (power–law) dependent on the distance from the pollen source. A literature survey confirmed that power–law function could fit in most cases. The b coefficient was lower than 2. The number of fertilized flowers by background (herbicide-susceptible) pollen grains was uniform across the whole field. Airborne pollen had a fertilization impact equivalent to that of one adjacent bolter. The individual dispersal function from different pollen sources can be integrated to provide the pollen cloud composition for a given target plant, thus allowing modeling of gene flow in a field, inter-fields in a small region, and also in seed-production area. Long-distance pollen flow is not negligible and could play an important role in rapid transgene dispersal from crop to wild and weed beets in the landscape. The removing of any bolting, herbicide-resistant sugar beet should be compulsory to prevent the occurrence of herbicide-resistant weed beet, thus preventing gene flow to wild populations and preserving the sustainable utility of the resistant varieties. Whether such a goal is attainable remains an open question and certainly would be worth a large scale experimental study.     

Integration and inheritance of transgenes in crop plants and trees

Transgene integration and inheritance have been investigated in a number of crop plants and few tree species. Transgene integration is predominantly a random process, whether mediated by Agrobacterium or particle bombardment. Depending on the genomic position of the integrated transgene and structure of the integration site as well as copy number of the transgene in the genome, its expression may be stable or variable. Therefore, integration patterns would affect the mode of transgene inheritance in plants, regardless of the method of gene transfer. So far, both Mendelian and non-Mendelian inheritance of transgenes has been reported across several generations (T1–T3) of crop plants. In few tree species (apple, poplar, plum, and American chestnut), mostly Mendelian inheritance of the transgenes has been observed in the T1 or BC1 generations. However, detailed studies in the transgenic papaya trees showed Mendelian segregation of the transgene in the T1 generation but non-Mendelian inheritance in the T2 generation. Variation in transgene inheritance was also detected in transgenic apple and plum trees. Long generation cycles in many economically important tree species preclude investigation of inheritance of transgenes in the tree progeny. Production of early flowering trees, either by genetic modification or by environmental modulation, would facilitate the study of transgene inheritance across generations of transgenic trees. In order to overcome problems of randomness of transgene integration, targeted transgene insertions by homologous or site-specific recombination or by designer recombinases or nucleases offer prospects for stable integration of transgenes in predetermined locations in the plant genome. And perhaps, that might provide a platform for stable expression and Mendelian inheritance of transgenes in plants.