Perennial ryegrass (Lolium perenne L.) is the most important pasture grass for meat, dairy and wool production in New Zealand, covering more than seven million hectares (Siegal et al., 1985). It is an out-crossing, wind pollinated and highly self-incompatible species and, for these reasons, the pace of genetic improvement has been slow through conventional breeding methods. Biotechnology can be a tool to accelerate the improvement of perennial ryegrass traits such as drought tolerance that are recalcitrant to conventional breeding techniques. High frequency genetic transformation of perennial ryegrass and other Lolium spp. has been achieved using biolistic bombardment (Altpeter et al., 2000; Takahashi et al., 2005) but Agrobacterium-mediated transformation of Lolium spp. has proven difficult, and only a few transformed lines have been produced (Wu et al., 2005). We have developed a high-frequency Agrobacterium-mediated transformation of perennial ryegrass (Bajaj et al., 2006) for candidate gene analysis. We have produced more than 1,000 independent transformed lines from several constructs selected using our SAGE™ analysis of gene expression in ryegrass taken from on-farm pastures.

Ryegrass transformation
Embryogenic calli derived from meristematic regions of the tillers of selected ryegrass lines were used for transformation (Fig. 1).

Agrobacterium tumefaciens strain EHA101 carrying binary vector pCAMBIA 1305.1 (CAMBIA, Australia) or a modified pMH vector (MetaHelix, India) were used for transformation experiments. Both plasmids carry the hygromycin phosphotransferase (hptII) gene driven by a double-enhancer version of the CaMV35S promoter and the pCAMBIA vector also harbors a new reporter gene, GUSPlus^™ (Broothaerts et al., 2005; US patent No. 6,391,547; WO 99/13085) driven by a typical CaMV35S promoter. Genes of interest from ryegrass were cloned into pMH for expression via either a double-enhancer version of the CaMV35S promoter or promoters sourced from perennial ryegrass (for an example see Fig. 2). Embryogenic calli were immersed with overnight-grown Agrobacterium cultures for 30 minutes with continuous shaking. Calli resistant to hygromycin were selected after subculturing them on co-cultivation medium for four weeks and verified by GUS staining (Fig. 3). After selection, the resistant calli were subcultured on regeneration medium every two weeks until the plants regenerated. The regenerants that continued to grow after two or three rounds of selection proved to be stable transformants (Fig. 3) as shown by GUS staining. Each regenerated plant was then multiplied on maintenance medium to produce clonal plantlets and subsequently rooted on MS medium without hormones. A rooted plant was transferred from each clone into contained glasshouse conditions while retaining a clonal counterpart in tissue culture as backup. Molecular analyses by Southern hybridization or reverse transcription PCR on some lines confirmed integration and/or expression of the transgene. In these experiments, the number of insertions in transformed plants ranged from 1-3 with the exception of one plant that showed four insertions (see Bajaj et al., 2006).

Functional Genomics in Ryegrass
After optimizing the transformation protocol, several genes cloned from ryegrass were transformed back into ryegrass for functional analysis. More than 1,000 plants were generated from 15 different constructs. Of particular interest is the ryegrass version of the Arabidopsis vacuolar H⁺ pyrophosphatase gene AVP1, the overexpression of which in transgenic Arabidopsis conferred drought tolerance (Gaxiola et al., 2001). We cloned the cDNA and the genomic DNA homologue of AVP1 from ryegrass under the double-enhancer version of the CaMV35S promoter or promoters sourced from perennial ryegrass, and produced several independent transformed lines. Our preliminary experiments on T₀ plants show high drought tolerance in some of the transformed lines.

Future Directions
A high throughput transformation system allows rapid analysis of a gene function and phenotype that can in turn help in screening gene candidates for marker-assisted breeding and other non-transgenic approaches to genetic improvement. For example, there is a gain-of-function mutant of the AVP1 gene (AVP1-D) involving a single residue change (E229D) reported to increase drought tolerance by a coordinated increase of both inorganic pyrophosphate (PPi) hydrolytic activity and PPi-dependent H⁺-translocation (Park et al. 2005). Screening of germplasm and/or ecoTilling lines with markers associated with the analogous ryegrass mutation will identify the best plant material for incorporation in ryegrass breeding programs.

Conclusions
In conclusion, "we have developed an efficient Agrobacterium-mediated genetic transformation protocol for perennial ryegrass and produced a large number of plants expressing various genes of interest." We have demonstrated that our protocol, coupled with publicly available genomic information (Jaiswal et al., 2006), allows perennial ryegrass to be used as a model plant for functional genomic studies of pasture grasses. Our transformation protocol will also help in applying Cisgenic^® technologies, in which only ryegrass genes are used in ryegrass, and marker-assisted breeding for drought tolerance, increased biomass, and other targets important for pasture-based agriculture.

References
Altpeter F, Xu J & Ahmed S (2000). Generation of large numbers of independently transformed fertile perennial ryegrass (Lolium perenne L.) plants of forage and turf-type cultivars. Mol Breed 6, 519-528

Bajaj S et al. (2006). A high throughput Agrobacterium tumefaciens-mediated transformation method for functional genomics of perennial ryegrass (Lolium perenne L.). Plant Cell Rep 25, 651-659

Broothaerts W et al. (2005). Gene transfer to plants by diverse species of bacteria. Nature 433, 629-633

Gaxiola RA, et al. (2001). Drought-and salt tolerant plants result from overexpression of the AVP1 H⁺ -pump. Proc Natl Acad Sci USA 98, 11444-11449

Jaiswal P (2006) Gramene: a bird's eye view of cereal genomes. Nucleic Acids Res. 34, D717-D723

Park S et al. (2005).Up-regulation of a H⁺-pyrophosphatase (H⁺-PPase) as a strategy to engineer drought-resistant crop plants. Proc Natl Acad Sci USA 102, 18830-18835

Siegal MR, Latch GC & Johnson MC (1985). Acremonium fungal endophytes of tall fescue and perennial ryegrass: significance and control. Plant Dis 69, 179-183

Takahashi W, et al. (2005). Increased resistance to crown rust disease in transgenic Italian ryegrass (Lolium multiflorum Lam.) expressing the rice chitinase gene. Plant Cell Rep 23, 811-818

Wu Y-Y et al. (2005). Salt-tolerant transgenic perennial ryegrass (Lolium perenne L.) obtained by Agrobacterium tumefaciens-mediated transformation of the vacuolar Na⁺/H⁺ antiporter gene. Plant Sci 169, 65-73

GENE FLOW FROM GE TO CONVENTIONAL MAIZE USING REAL-TIME PCR
Maria Pla, Joaquima Messeguer & Enric Melé

Worldwide commercialization and increasing acreage of genetically engineered organisms (GEO) have propitiated the approval of labeling regulations in several countries to protect the consumers' right to information. As an example, EEC regulations (Commission Regulation (EC) No 258/97, 1997; Commission Regulation (EC) No 49/2000, 2000; Commission Regulation (EC) No.50/ 2000, 2000; Commission Regulation (EC) No 1829/2003, 2003 Commission Regulation (EC) 1830/2003, 2003) established the compulsory labeling of foods containing more than 0.9% GE ingredients. In order to harmonize the necessary coexistence between GE and non-GE crops grown in parallel, standards, which strongly benefit from experimental gene flow data, are being established or prepared in many countries. It is important to take into consideration the extent of pollen dissemination from transgenic to conventional crops under field conditions to properly establish containment strategies to minimize the adventitious presence of transgenes in conventional or organic crops. Of particular concern is maize, of which there are an increasing number of GE varieties cultivated worldwide.

TRANSGENIC PLANTS THAT MAKE NON-TRANSGENIC POLLEN
Ludmila Mlynarova & Jan-Peter Nap

A major challenge in the agronomy of genetically engineered (GE) crops is to prevent gene flow to non-GE crops to wild relatives.¹ GE material should not enter the food production chain or the environment when and where this is not desired and/or not sufficiently controlled. This challenge requires the design of GE crop management protocols that generate added value to agriculture, while coexistence of organic, non-GE and GE crops satisfies the desire of consumers and/or markets. A major route for unwanted mixing of GE crops and non-GE plants is gene flow by pollen. Together with Dr. A. J. (Tony) Conner, Crop & Food Research Institute, New Zealand, we have demonstrated that gene flow by pollen can be effectively eliminated in a new approach that incorporates transgene removal into the biology of pollen formation.²

Pollen without transgenes from plants with transgenes
The new approach to remove transgenes from pollen makes novel use of existing methods using recombination for gene removal. Previously, such recombination-mediated approaches were largely aimed at removing undesired selectable marker genes from incoming DNA.³ The gene for the necessary recombinase enzyme, often, but not limited to, the enzyme Cre, as well as its characteristic recombination sites (loxP), are added to the incoming DNA in such a way that the selectable marker gene along with the Cre gene is removed in a process known as auto-excision. This way, all incoming DNA can be removed, except for a single loxP site. Although it may sound silly to transform plants with transgenes with the aim to remove all incoming DNA, the crucial issue is where in the life cycle of the crop the transgenes are removed.

We have employed a tightly controlled microspore-specific promoter that is activated at the start of microsporogenesis that results in mature pollen.² The tobacco NTM19 promoter originates from a tobacco gene encoding a microspore-specific protein with unknown function. This NTM19 promoter was coupled to a plant intron-containing Cre recombinase gene. Activation of the promoter during pollen formation resulted in recombinase-mediated auto-excision of all genetic material between recombination sites in pollen only. As a result, all pollen of the GE plant is wild type, except for the presence of a single recombination site. In more popular terms, the GE plant 'cleans' itself while preparing for sex, although we hesitate to suggest that transgenes are by definition 'dirty'.

Surprisingly efficient 'clean-up' of transgenes from pollen
This fairly simple approach proved surprisingly efficient in both GE Nicotiana tabacum (tobacco) and GE Arabidopsis thaliana (thale cress). Actual gene flow, that is, forced outcrossing to a non-GE (wild type) plant, generated only non-GE seeds (with less than 0.03% escapes). PCR analyses and crossings confirmed the generation of WT pollen from GE plant material. Growing plants or germinating seeds at elevated temperatures (30°C) in the greenhouse showed no evidence for either premature activation (transgene loss from somatic cells; a production risk) or absence of activation (transgene presence in pollen; a safety issue) in either tobacco or Arabidopsis. More environmental stresses should be evaluated to show how robust and tightly controlled the activation of the NTM19 promoter actually is. We expect that in the actual field the fraction of escapes will be considerably lower. More research is also required to see whether particular places of integration will give different (lower? zero?) escape rates. For application, it will be necessary to test other plant species: it is not known how the NTM19 promoter behaves in, for example, a monocot species such as rice.

The highly efficient excision of plant transgenes linked to pollen development does not require external stimuli such as heat shock or spraying with chemicals. Such treatments are not easily accomplished in natural ecosystems, so incorporating transgene removal in the biological process of pollen development is a clear advantage after any undesired escape. Pollen promoter-mediated removal of transgenes is an intrinsically iterative process: each plant with transgenes maintained via the female lineage will produce transgene-free pollen, effectively preventing fixation in wild populations and reducing the potential area of unwanted spread. This way, transgene 'clean-up' has become an integral part of pollen biology.

Additional safety issues to consider
Obviously this new approach does not solve all biosafety issues of unwanted transgene spread. There is still a 34 bp piece (a loxP site) of 'foreign' DNA in the plant genome carried by pollen. This DNA does not allow any protein production. Biosafety discussions regarding the presence of 'foreign' DNA in the plant genome could and should be limited to the presence of this 34 bp segment. In addition, it should be pointed out that transgene spread through seeds and plants is not prevented. Spillage of GE plant-produced seed, either deliberate or unintentional, could result in fertile GE plants outside the intended agricultural field. However, as outlined above, the removal of transgenes is made an integral part of the biology of the plant. As a result, the removal will happen again each generation, so the further spread of that plant and its transgenes will be considerably slower compared to plants without the removal system.

The transgenes-of-interest are maintained in the transgenic plant via the female lineage. This creates additional challenges for breeding and seed production. GE plants generated this way cannot become homozygous and should be maintained for breeding as so-called hemizygous lines. Commercial applications of this new approach are being discussed. For vegetatively propagated crops, such as potato, fruit, or trees, the issue of hemizygosity would seem to pose no additional challenges. In crops in which sexual transmission of the transgenes is not necessary, generating transgene-free pollen can have immediate application. This may be especially important to reduce gene flow to wild relatives of clonal cultivars, especially near the centers of genetic diversity in such crops.

The economy of seed production
In seed propagated crops, however, extra steps will have to be taken for breeding and/or seed production. Upon selfing, or outcrossing to another plant with the same insert, half of the seeds will be wild type. For seed production, standard herbicide tolerance (e.g., BASTA) could replace the kanamycin resistance used in the proof-of-concept research phase. Kanamycin resistance could be combined with seed selection methods either before or after planting in order to be an economically viable enterprise. This could be achieved by either a seed priming step with herbicide solutions prior to sowing or at an early stage in crop development. The sowing of mixtures with herbicide-resistant and herbicide-sensitive seed at a higher density followed by chemical thinning of the herbicide-sensitive seeds may also facilitate crop establishment. It could be investigated whether induction of distorted segregation towards 'true breeding' herbicide-tolerant plants by the application of herbicide treatments to the hemizygous plants is feasible. For herbicides with high translocation in plants, such treatments have been reported to effectively prevent the full development of sensitive pollen, ovule, and seeds on the otherwise tolerant plants.

The feasibility of such applications and additional steps will depend on the economics of production and the added value of the new GE trait. The additional work on breeding and seed production should be cost effective, and this will largely be decided upon by the added value of the transgene-encoded trait. The new approach seems particularly useful for specialty crops, such as the production of chemicals in tobacco, also, or particularly, when grown in contained environments. As always in risk – benefit balancing,¹ the potential disadvantages and additional costs of the use of hemizygous plants will have to be weighed against the perceived and desired gain of safety in the field.

A democratic form of 'terminator technology'?
The auto-excision of transgenes achieved in this novel approach could be considered a new variant of GURT (genetic use restriction technology) for which notably the 'terminator technology' got very bad press.⁴ Terminator technology aimed at modifying plants so they produce sterile, non-germinating seeds. This approach is now promoted because of its potential value in the biological containment of GE crops. Terminator technology was, however, criticized much more because it provides a biological protection of property that is much stronger than any patent or breeder's right. It could lead to an undesired (or unacceptable) dependence for poor smallholder farmers. The biosafety value of the new approach using pollen promoter-mediated excision is the elimination of transgene dispersal to neighboring non-GE crops or crossable relatives (either weeds or in pristine ecosystems). Compared to terminator technology, the transgenes are not present in pollen and outcrossed seeds, which could be considered an additional food safety advantage.

For biosafety, this new method can be an interesting alternative, as it does not imply the biological protection of property of the terminator technology and does not result in more dependence of farmers. Any farmer can reproduce the seed and any breeder can use it in further breeding, subject to the national seed laws and intellectual property systems. Yet, farmers, breeders, and seed producers are all faced with additional work to maintain, enrich or screen for the transgene-of-interest in the seed. In this sense the non-GE pollen approach could be considered a different and "democratic" application of terminator technology: both producers and growers will have to take additional steps to maintain the trait-of-interest, whereas undesired spread is 'terminated'.

Improved biological containment for biosafety
The new approach can prevent the adventitious presence of transgenes in non-GE crops or related wild species by gene flow. It does not require external stimuli and is an intrinsically iterative process: transgene 'clean-up' has become an integral part of pollen biology. As such, it may coexist with chloroplast containment.⁵ Chloroplasts are generally (but not in all plants) maternally inherited and considerable progress in chloroplast transformation technology has been achieved. As the new approach presented here is based on established nuclear transformation technology, it may be more straightforward to achieve. Biological containment of transgenes, by whatever means, will help the deployment and management of coexistence practices to support consumer choice and can promote clean molecular farming for the production of high-value compounds in plants. When wisely adopted, it is likely to open up exciting new possibilities for the future use and safety of GE crops.

References
1. Conner AJ, Glare TR, and Nap JP (2003). The release of genetically modified crops into the environment: Part II. Overview of ecological risk assessment. Plant Journal 33, 19-46

2. Mlynarova L, Conner AJ, and J.P. Nap (2006). Directed microspore-specific recombination of transgenic alleles to prevent pollen-mediated transmission of transgenes. Plant Biotechnology Journal 4, 445-452

3. Mascia PN & Flavell RB (2004). Safe and acceptable strategies for producing foreign molecules in plants. Current Opinion in Plant Biology 7, 189-195

4. Louwaars NP et al. (2002). Policy response to technological developments: The case of GURTs. Journal of New Seeds 4, 89-102

5. Grevich JJ & Daniell H. (2005). Chloroplast genetic engineering: Recent advances and future perspectives. Critical Reviews in Plant Sciences 24, 83-107

U.S. farmers have adopted genetically engineered (GE) crops widely since their introduction in 1996, notwithstanding uncertainty about consumer acceptance and economic and environmental impacts. Soybeans and cotton genetically engineered with herbicide-tolerant traits have been the most widely and rapidly adopted GE crops in the U.S., followed by insect-resistant cotton and corn. This product summarizes the extent of adoption of herbicide-tolerant and insect-resistant crops since their introduction in 1996. Three tables, devoted to corn, cotton, and soybeans covering the 2000-2006 period by State, can be found at the following websites: