The biotechnology industry has a clear need and a natural responsibility to address the concerns of its customers and those of the public at large if it is to contribute meaningfully to the current debate over its future. The strongest voices raised against the use of new genetic technologies in agriculture are those of activists whose ethos is based on objections to contamination, usually manifested as environmental pollution. These organizations also tend to oppose perceived political, cultural, and commercial colonialism, and often fail to disentangle these opinions from those that are based on an ecological stance. Emerging technologies can deal with arguments against genetic modification (GM) based on legitimate scientific standpoints, and thereby impact upon the political debate.

Runaway Genes

The major stated concern of many opponents of GM is the movement of genes from modified organisms into other organisms in the natural environment, a process termed `gene escape' in scientific discourse and `genetic pollution' in activism pamphlets. One brand of anti-GM zealotry anticipates apocalyptic consequences in the event of any gene derived from a GM event moving out into the biosphere, but research does not support this.(1) For most people who express an opinion, the creation of antibiotic-resistant pathogens (`superbugs') or herbicide-resistant uncultivated plant species (`superweeds') are seen as the greatest potential dangers of gene escape, and regulatory approval around the world is increasingly being withheld for GM plants with such genes. Except in cases in which engineered resistance is the desired outcome (e.g., Roundup Ready maize), such genes are unnecessary beyond the laboratory-based phase of a biotechnology program, and the failure to remove them or avoid their use entirely implies that the industry is woefully or willfully ignorant of one of the public's major concerns. Fishermen cannot leave their hooks in the catch and expect their customers to return. Techniques exist for the removal of such genes before field release(2), but these delay the development of products and introduce additional complexity and unpredictability.

All methods for inserting DNA into plants are inefficient and result in large numbers of plants of which only a tiny proportion are transformed. Resistance to an antibiotic or herbicide `marks' a transformed organism as carrying the modified DNA. Marker genes are therefore essential in laboratory research, being the equivalent of using an X-ray machine in the proverbial search for a needle in a haystack. Unpredictability of transformation outcome further contributes to the need for markers; many transgenic plants must be manufactured via transformation, identified by the presence of markers, and assessed for useful traits before one with the optimum balance of characteristics can be found. To extend the metaphor, a large number of needles must be found, as some needles are better than others.

Markers are therefore a prerequisite for all current methods of plant transformation. Unfortunately, the most widely used marker genes are precisely those that raise the most public concern by conferring abilities that are advantageous beyond the laboratory, for example, the capacity to degrade a potent herbicide before the plant succumbs to its effects. This is an example of `negative selection'— the entire population is subjected to a negative (toxic) selection pressure that only transformants can bear. The main advantage of negative selection is that it negates some of the inefficiencies of transformation; untransformed plants are culled by the antibiotic or herbicide, leaving a population enriched with transformants. One disadvantage of negative selection systems, in addition to concerns over gene escape, is that they cause destruction of untransformed tissues, which then release toxic, inhibitory, or suicide-signalling compounds to the detriment of the transformants.

A different kind of marker gene exists that can signal its presence but does not provide a fitness advantage, obviating the improbable apocalyptic scenarios peddled by some anti-GM activists and the practical problem of toxin release during the experiment. The first such gene was isolated from the laboratory workhorse bacterium E.coli and encodes the ß-glucuronidase (GUS) protein; this confers upon transformed tissues the ability to break down a synthetic chemical added to their growth medium into a fluorescent product.(3) `Neutral' markers of this kind are screenable rather than selectable, meaning that the identification of transformants involves more labor and time. There is a clear need for a marker technology that poses no realistic or hypothetical danger of superbug or superweed formation, yet is as efficient as the negative selection systems currently favored.

Building A Bigger Biotechnology Toolkit

While the development of entirely novel selectable or screenable marker technologies is desirable, it is also important that the range of options within the current paradigms is expanded. The tools available for plant modification at the genetic level are limited to a handful of well-understood regulatory or targeting elements and a few options for selection and transformation systems. There is only limited knowledge about the characteristics and potential uses of the many other genome elements such as introns, terminators, enhancers, and repressors.

The torrent of information from the large-scale automation of biological research embodied in genomics and proteomics must be trawled for new tools for the genetic engineer. Exigent additions to the toolkit are non-viral promoters (another public concern) and new marker systems to permit greater control and precision in genetic modification. A recent paper by Gough and co-workers(4) describes the development of a new negative selection system for the toolkit in which, rather than detoxifying a herbicide, the transformed plant possesses a backup metabolic step to replace the essential one attacked by the herbicide. By the introduction of an altered cyanobacterial enzyme called glutamate-1-semialdehyde aminotransferase, which, unlike the plant's own copy, is not susceptible to the effects of the toxin gabaculine, negative selection is achieved. This is analogous to the use of the glyphosate resistance gene, a mutant and immune form of the glyphosate target enzyme.

A positive selection system is one that enhances the performance of transformants over that of normal plants, and could be a viable replacement for negative selection systems, yet offers advantages over neutral screenable systems. In this type of marking, a fitness advantage is conferred in an artificial situation that is harmless to other plants. This is the most active and interesting area of current research in marker development and (to extend the haystacks metaphor yet further) allows the seeker to use magnetism in the hunt for needles. The GUS neutral marker gene can be adapted for use in a positive selectable marker system.(5) In this system, one of the required growth hormones of plant culture medium is supplied in a form activated only by the GUS protein, so only transformants can grow. The widespread use of GUS in plant GM could permit existing research programs to switch painlessly from screening to applying selection.

Similarly, the use of the phosphomannose isomerase gene isolated from E.coli in a positive selection marker system has recently been demonstrated in some commercially important crop plants.(6) Transformed plant tissues expressing this gene can grow on culture medium containing the sugar mannose as the only source of carbon, while untransformed tissues can maintain their size but, lacking utilizable carbon, do not grow further. The gene confers no advantage in the natural environment where plants are self-sufficient in carbon derived from the atmosphere using photosynthesis, and so cannot contribute to the generation of a superweed. Other carbohydrates that do not support plant growth can be used similarly when there exists a non-plant enzyme that can convert them into a form usable by the plant, e.g., xylose and the xylose isomerase gene isolated from Thermoanaerobacterium thermosulfurogenes.(7)

Broader Use Of The Better Toolkit

The most controlled method of gene delivery into plants is via an intermediate modification of the plant pathogen Agrobacterium. This method introduces into the plant only the genetic material contained within defined boundaries on a large loop of DNA. Other methods of transformation, the most popular of which is microprojectile bombardment (`biolistics'), often introduce the rest of the DNA loop. This extraneous genetic material is only necessary during the construction of the transforming DNA and commonly contains antibiotic resistance genes as components of negative selection systems used in E.coli. Unfortunately, most commercially important agricultural plants are not amenable to Agrobacterium-mediated transformation, and so biolistic methods are employed without removing these undesired elements; this is one route by which antibiotic-resistance genes arrive in the genomes of modified plants.

LaFayette and Parrott(8) have developed a positive selection marker system for E.coli that avoids the use of formerly ubiquitous antibiotic resistance markers. In their system, the presence of the rtl gene permits the growth of transformed bacteria on culture medium containing the sugar alcohol ribitol as the sole carbon source; untransformed bacteria lack the gene and therefore cannot multiply. DNA that has been built in bacteria using this system can be transferred to a plant since the bacterial marker gene is incapable of contributing to superweed creation. Such plants must still harbor a second selection system to identify transformed plants, such as those described above, and these plants would then address scientifically founded concerns over gene escape.

The expansion of the range of available markers and the applications of positive selection systems promise to increase the abilities and precision of plant genetic engineering, and consequently raise the confidence of the public in the biotechnology industry. Several prominent scientific concerns over the current applications of gene technologies are addressed by the emerging technologies in this field. Until an unlikely and unpredictable quantum leap in the efficiency of transformation technology takes place (allowing us to prevent the formation of metaphorical haystacks entirely), the development of new marker technologies is a keystone of advancing biotechnology research and development.

Sources

1. See for example Crawley MJ, et al. 2001. Transgenic crops in natural habitats. Nature 409: 682-683.

2. Either via time-consuming multiple breeding steps or more advanced methods such as reported by Zuo JR, Nui Q-W, Møller SG, and Chua N-H. 2001. Chemical-regulated, site-specific DNA excision in transgenic plants. Nature Biotechnology 19: 157-161.

3. Originally reported by Jefferson RA, Kavanagh TA, and Bevan MW. 1987. GUS fusions: ß-glucuronidase as a sensitive and versatile gene fusion marker in higher plants. EMBO Journal 6: 3901-3907.

4. Gough KC, Hawes WS, Kilpatrick J, and Whitelam GC. 2001. Cyanobacterial GR6 glutamate-1-semialdehyde aminotransferase: a novel enzyme-based selectable marker for plant transformation. Plant Cell Reports 20: 296-300.

5. Joersbo M and Okkels FT. 1996. A novel principle for the selection of transgenic plant cells: positive selection. Plant Cell Reports 16: 219-221.

6. Wright M, et al. 2001. Efficient biolistic transformation of maize (Zea mays L.) and wheat (Triticum aestivum L.) using the phosphomannose isomerase gene, pmi, as the selectable marker. Plant Cell Reports 20: 429-436.

7. Haldrup A, Petersen GS, and Okkels FT. 1998. The xylose isomerase gene from Thermoanaerobacterium thermosulfurogenes allows effective selection of transgenic plant cells using D-xylose as the selection agent. Plant Molecular Biology 37: 287-296.

8. LaFayette PR and Parrott WA. 2001. A non-antibiotic marker for amplification of plant transformation vectors in E.coli. Plant Cell Reports 20: 338-342.