*For offprint requests: fax: (910) 334-5839; email: nstewart@goodall.uncg.edu

SUMMARY

The significant increase in the development of transgenic plants has caused concern among researchers about the potential escape of transgenes into the environment. The availability of a cost efficient, in vivo, real-time system is needed, therefore, to track genetically engineered crop plants containing transgenes that can confer an increment of fitness to could-be weedy crops and their weedy relatives. In order to obtain such a system, we are transforming plants with green fluorescent protein (GFP,) a protein which enables the whole plant to visibly fluoresce green under UV light. The GFP protein masks the natural pink autofluorescence chlorophyll produces under UV light. We envisage a system where GFP may be fused or linked to other economically important genes. These green plants are obviously distinguishable from pink non-transgenic plants in a field, so in the event that a gene is found to cause an ecological risk, the plants can be closely monitored for escapes. These escapees could then be easily identified by scanning the area with UV or blue light and looking for green plants among the pink plants.

Keywords: GFP, gene flow, marker genes, transgene persistence, transgene monitoring

INTRODUCTION

The rapid development of transgenes and transformation technology in plants has provoked increased concern from regulatory officials and scientists regarding the release of transgenes into the environment. Even though previous field trials involved small-scale releases under strict regulatory control, current trends towards deregulation and commercialization (almost 1800 field releases from 1987 to 9/30/95, APHIS Permits Sept, 1996) may have impacts on the environment that cannot be ignored.

Since the commercialization of transgenic crops will be an uncontrolled experiment, the ramifications of escaped transgenes conveying fitness traits such as herbicide, disease or insect resistance to surrounding plant populations could be severe. Research, therefore, has consisted mostly of gene flow studies since this is the first step in assessing environmental risk for each plant species. To do this, a selectable or scorable marker gene must be used to track transgenic plants. In the past, selectable marker genes like drought or insect resistance linked to the gene of interest have been used, but these are a problem because they are not neutral genes. Two other commonly used methods, such as linked scorable marker genes like those coding for B-glucuronidase (GUS) which necessitates the use of the expensive substrate X-Gluc, or directly analyzing the transgene DNA using PCR, are expensive or time-consuming (reviewed in Stewart, 1996a). Therefore, an ideal assay would be a real-time, in vivo, marker gene that could be inserted into any plant species.

A gene recently isolated and cloned from the jellyfish Aequoria victoria, called green fluorescent protein (GFP) meets these requirements. GFP is a 27 kDa monomer that fluoresces green under UV (365 nm) or blue (490) light (Chalfie et al., 1994). The significance of GFP is that it does not require any substrate, enzyme, or co-factor to fluoresce, making it an in vivo marker. In order for a plant containing the transgene to fluoresce green under UV or blue light though, the GFP protein must be highly expressed to mask the pink autofluorescence of chlorophyll. The wild type GFP gene has been modified by Jim Haseloff (mGFP4) by altering codons at the site of prior mis-splicing and now provides stable and high expression in transgenic plants (Haseloff et al., 1996). mGFP4 provides a several fold increase in protein expression over native GFP because it has altered codons at prior mis-spliced sites (Haseloff et al., 1996). Therefore, plants containing high amounts of fluorophore-active GFP can be distinguished from non-transgenics by a visual screening using a portable hand-held UV light (Stewart, 1996a; b).

MATERIALS AND METHODS

Using an Agrobacterium tumefaciens expression vector (courtesy of Jim Haseloff) containing the mGFP4 construct under the control of the CaMV 35S promoter, Nicotiana tabacum (tobacco) cv 'Xanthi' was transformed using Agrobacterium-mediated transformation (Shardl et al., 1987).

RESULTS AND DISCUSSION

Of the 25 transgenic lines that were recovered, only 2 fluoresced macroscopically. This is a low frequency of high expressors. Recently it has been shown that high GFP transgene expression does assure high fluorescence within plant lines (Leffel et al., unpublished data). This suggests that there may be environmental factors affecting proper fluorophore formation sGFP. However, this proof-of-concept study demonstrates that some form of GFP and other in vivo markers can be used as valuable tools for transgene monitoring.

Our future ecological work will involve deploying transgenic seedlings in a field with non-transgenics planted at various distances in concentric circles around them. The plants will be allowed to flower and set seed. The goal is to assess gene flow and persistence. We will estimate gene flow from transgenic, pollen-donor populations to non-transgenic, pollen-recipient populations using a standard "bullseye" design. Progeny plants will be observed at night using a simple hand-held UV or blue light and an electric generator, then will be analyzed for transgene frequency by determining if they are positive or negative for fluorescence, and therefore, poistive or negative for the transgene. the power of this approach is that results can be obtained in less time because many plant species may be tested together in the same field instead of testing them individually. Since the plants will be allowed to re-seed in situ, results can be assessed in real time, discarding the need for complex molecular or biochemical analysis. More importantly and the ultimate goal of the project is to provide biotech companies with a system to tag the genetically engineered plants they produce. No transgenic populations have been in existence for more than a few years. This fact is not trivial since ecological effects can range from years to centuries. If commercialized transgenic plants contain an in vivo marker such as GFP, then the scientific community would have a powerful tool to address ecological effects well into the twentieth century.

ACKNOWLEDGMENTS

We appreciate the gift of plasmids from Jim Haseloff. Funds from the USDA BRARGP were used to partially fund this research.

REFERENCES

Chalfie M, Tu Y, Euskirchen G, Ward WW, Prasher DC. 1994. Green fluorescent protein as a marker for gene expression. Science 263:725-888.

Haseloff J, Siemering K, Hodge S, Golbik R, Prasher D. 1996. The green fluorescent protein gene must be modified for use as a vital marker in Arabidopsis thaliana. Plant Physiol 111:17. (Abstract).

Schardl CL, Byrd AD, Benzion G, Altschuler MA, Hildebrand DF,Hunt AG. 1987. Design and construction of a versatile system for the expression of foreign genes in plants. Gene 61:1-11.

Stewart, CN, Jr. 1996a Transgene flow and persistence may be monitored using in vivo markers such as GFP. Biosafety Volume 2, (BY96003), September 2nd 1996. Refereed Online Journal, URL http://www.bdt.org.br/bioline/by. Stewart, C. N., Jr. Monitoring of transgenic plants with in vivo markers.

Stewart, C N, Jr. 1996b. Monitoring transgenic plants using in vivo markers. Nature Biotechnology 14:682.