A novel clean gene transformation technology that eliminates antibiotic resistance genes from transgenic chloroplasts has been developed in the UK.

Despite the potential of genetically modified (GM) plants for improving the quantity and quality of crops, they have not received the worldwide consumer acceptance that might have been expected. GM-antagonists argue that the widespread use of transgenic plants poses an unacceptable risk to the environment and human health. Whilst the strength of this anxiety over GM technology is difficult to justify, it has meant that transgenic plants themselves and the technologies used to generate them are facing ever-increasing scrutiny. The presence of antibiotic resistance genes in GM crops and their dissemination via pollen are particularly controversial issues. In the UK, a subgroup of the Advisory Committee on Releases to the Environment (ACRE) has been given the specific task of examining emerging technologies that could contribute to the development of transgenic plants designed to minimize environmental exposure to transgenes and their products (http://www.environment.detr.gov.uk/acre/bestprac/index.htm).

Antibiotic resistance genes are common components of gene transfer technologies. Transgenic plants are usually made by transferring one or two genes, commonly from another plant or bacterium, into chromosomes located in the nucleus of plant cells. Once transferred, these foreign genes are inherited along with the 25,000 to 50,000 native genes present on plant chromosomes. Gene transfer methods are inefficient and only a tiny proportion of cells usually take up foreign genes. To identify cells that take up foreign DNA, a foreign marker gene that confers a selectable property is required. These selectable marker genes are added alongside genes of interest. Antibiotic resistance genes are one of the most commonly used marker genes. Only plant cells that take up foreign genes proliferate in the presence of an antibiotic that kills unmodified cells. Once these transgenic plants have been selected, antibiotic resistance genes are no longer required, but they are usually retained. Whilst the antibiotics widely used to select GM plants are of limited oral use and the rate of transfer of antibiotic resistance genes from plant DNA to gut bacteria appears to be low (preventing its measurement), the presence of superfluous genes in GM plants is increasingly viewed as undesirable. A report (December 1998) by a House of Lords select committee on the risks to human health of eating GM crops suggests that antibiotic resistance genes should be phased out of GM crops as swiftly as possible since alternatives are available (http://www.parliament.the-stationery-office.co.uk/pa/ld199899/ldselect/ldeucom/11/8121501.htm).

Chloroplasts are a suitable site for locating foreign genes Plant cells contain DNA in three subcellular compartments. Approximately 80% of the DNA is located in the nucleus as chromosomes, 10-20% in chloroplasts, and around 1% in mitochondria. There are approximately 10-100 chloroplasts per cell. The DNA present in chloroplasts is circular and contains about 100 genes. Around 500 to 10,000 copies of these chloroplast DNA circles are present per cell. Methods to introduce foreign genes into chloroplasts lagged behind those used to insert foreign genes into plant nuclei. The stable integration of foreign DNA into chloroplasts was first demonstrated in 1988 using Chlamydomonas, a unicellular green alga, and particle bombardment technology developed at Cornell University(1). Microscopic tungsten particles coated with foreign DNA were accelerated into target cells. The particles penetrated the tough cell walls and delivered foreign DNA into chloroplasts. This success in Chlamydomonas was repeated in tobacco(1). Procedures based on bathing plant cells in media containing foreign DNA and polyethylene glycol have also allowed foreign genes to be introduced into chloroplasts(1).

The chloroplast is a suitable location for a wide range of foreign genes including those involved in photosynthesis, starch synthesis, fatty acid synthesis, oxidative stress tolerance, and those conferring tolerance to herbicides. Moreover, chloroplasts are useful compartments for storing polymers and pharmaceuticals. Although all foreign genes that we might wish to introduce into plants cannot be localized to chloroplasts, many of the genes that are perceived as "high risk" with respect to environmental impact, such as herbicide(2) and insect resistance genes(3), are functional in chloroplasts.

The insertion of genes into chloroplast DNA exhibits a number of desirable features. Since chloroplast DNA is small, it has been characterized in many plant species. The complete sequences of 18 different chloroplast genomes are known. Moreover, the gene content and organization of chloroplast DNA does not vary greatly between related species. This means we have detailed knowledge of the genomic environment into which we are inserting foreign genes, and the insertion of foreign genes into chloroplast DNA can be controlled with complete precision. Foreign genes can be targeted to specific predetermined sites in chloroplast DNA using the native homologous DNA recombination machinery present in chloroplasts. Foreign genes are usually propagated in bacteria by linking them to vector sequences, which allow their replication in bacteria before they are integrated into plant DNA. Vector sequences usually remain attached to foreign genes when they are transferred into plant cells. Although the overall risk of gene transfer from GM plant to bacteria is small, removal of vector sequences will reduce this risk further by deleting sequences that are known to promote replication in bacteria. When foreign genes integrate into chloroplast DNA by homologous recombination, bacterial vector sequences are excluded and are not present in genetically modified chloroplast genomes.

The release of pollen from many GM crops is a major route for transgene dispersal into the environment. Unlike nuclear chromosomal genes, which are transmitted in equal proportions by egg and sperm cells, chloroplast genes are transmitted solely via egg cells in many crops. This maternal pattern of inheritance prevents the pollen-mediated spread of foreign genes located in chloroplasts and is of considerable benefit for the environmental containment of transgenes. When two chloroplast types are present in the same plant, they tend to sort out into two pure populations of each chloroplast type. This cytoplasmic sorting process is unique to genes located in chloroplasts and mitochondria. Maternal inheritance, cytoplasmic sorting, and the rarity of DNA exchange between two chloroplast types, reduce, if not eliminate, the possibility of stacking chloroplast types resistant to different herbicides.

Efficient transformation of chloroplasts has relied on the aadA marker gene(4) that confers resistance to the antibiotics streptomycin and spectinomycin. Because there are hundreds and in many cases thousands of copies of chloroplast DNA per cell, the introduction of foreign genes into chloroplasts is a two-step process. In the first step, the aadA gene integrates into a fraction of the chloroplast DNA molecules present in a cell. In the second step, modified chloroplast genomes containing aadA are selected with spectinomycin and streptomycin until they replace all wildtype chloroplast genomes after repeated cell and chloroplast divisions. Once a plant is homoplasmic, that is contains only modified chloroplast genomes, the aadA gene is no longer required.

Removing antibiotic genes from chloroplasts(5)

Removal of aadA from chloroplasts is desirable. Although spectinomycin and streptomycin are rarely used clinically and not dispensed in the community, streptomycin is used occasionally for treatment of resistant Mycobacterium tuberculosis in hospitals. The task of removing aadA from chloroplast DNA presents a challenge since it requires aadA excision from many copies of chloroplast DNA per cell. We decided to utilize the homologous DNA recombination machinery in chloroplasts to excise aadA genes rather than adding foreign recombinases and their target sites, such as the Cre/lox system, that have been used to excise antibiotic resistance genes from nuclear chromosomes. Our approach obviates the subsequent need to remove foreign recombinase genes from GM plants.

We used short direct DNA repeats to confer instability to the aadA gene. The degree of instability was important. If loss of aadA was too high, we would have been unable to use aadA to select plants containing modified chloroplast genomes. Correspondingly, a low frequency of aadA loss would have prevented us from isolating aadA-free plants. We found the correct level of aadA instability by increasing the number of short direct repeats in a construct from two to three. Once we were able to accumulate a high proportion of aadA-free chloroplast genomes within a plant, we relied on cytoplasmic sorting to isolate aadA-free tobacco plants. The aadA-free plants we isolated either contained a uidA reporter gene or herbicide resistance gene located in chloroplasts.

This method will allow a large range of genes, for example those encoding pharmaceutical proteins and insect resistance, to be introduced into chloroplasts without antibiotic resistance genes. Although the procedure was developed in tobacco, it is likely to work in a range of plant species. The small number of species amenable to stable chloroplast transformation is a current limitation. The combination of maternal inheritance, precise gene targeting, and removal of vector and antibiotic resistance genes makes the chloroplast a favorable site for locating transgenes in order to reduce their environmental impact. These positive features should accelerate the pace of research on chloroplast transformation of major crops.

Sources:

1. Heifetz P. 2000. Genetic engineering of the chloroplast. Biochimie 82: 655-666.

2. Daniell H, et al. 1998. Containment of herbicide resistance through genetic engineering of the chloroplast genome. Nature Biotechnology 16: 345-348.

3. Kota M, et al. 1999. Overexpression of the Bacillus thuringiensis (Bt) Cry2Aa2 protein in chloroplasts confers resistance to plants against susceptible and Bt-resistant insects. Proceedings of the National Academy of Science U.S.A. 96: 1840-1845.

4. Svab Z and Maliga P. 1993. High frequency plastid transformation in tobacco by selection for a chimeric aadA gene. Proceedings of the National Academy of Science U.S.A. 90: 913-917.

5. Iamtham S and Day A. 2000. Removal of antibiotic resistance genes from transgenic tobacco plastids. Nature Biotechnology 18: 1172-1176.

** NOTICE: In accordance with Title 17 U.S.C. Section 107, this material is distributed for research and educational purposes only. **