Phytoremediation is the using of plants to clean up contamination. We are indebted to The-Scientist.com for the following from their ''daily news'' of Oct.7th, 02

PLANTS CAN CLEAR ARSENIC POLLUTION ( ED.note - e.g. after gold mining )
Genetically engineered Arabidopsis plants can sequester arsenic from the soil. [ Tudor P Toma ]

Arsenic is an extremely toxic metalloid pollutant yet the decontamination of polluted sites can be environmentally destructive. An alternative is suggested in October 7 Nature Biotechnology; Om Parkash Dhankher and colleagues at the University of Georgia, Athens, US, show that genetically engineered plants can transport arsenic above ground, reduce it to arsenite, and sequester it in thiol-peptide complexes (Nature Biotechnology, DOI:10.1038/nbt747, October 7, 2002).

Dhankher et al. examined the effects of coexpressing two bacterial genes, arsenate reductase (arsC) and y-glutamylcysteine synthetase (y-ECS), in Arabidopsis plants. They observed that plants expressing SRS1p/ArsC and ACT2p/ y-ECS together showed substantially greater arsenic tolerance than wild-type plants or plants expressing y-ECS alone. In addition, when grown on arsenic, these plants accumulated 4-17-fold greater fresh shoot weight and accumulated 2- 3-fold more arsenic per gram of tissue than wild-type plants or plants expressing y-ECS or ArsC alone.

"These transgenic technologies and the understanding of the physiology of arsenic behavior in plants should lead to the development of high-biomass, fast-growing arsenic hyperaccumulator plants for field use," conclude the authors. © the-scientist.com Oct.7, 2002 "Daily news"

We at MolecularFarming.com are also very pleased to have linked with D. Glass AssociatesWe have two articles below from this world authority.The bottom one is from the 1997 Battelle Bioremediation Symposium. The First [ top ] is an excerpt from his 1999 market report on phytoremediation, that talks about the prospects for use and the regulatory outlook for transgenics, but from a more business and science oriented view than the 1997 paper.

"US and International Markets for Phytoremediation, 1999-2000" © David Glass. D.Glass Phyto Market Reports

{This is excerpted from a larger document, it begins somewhat abruptly and also includes references for which the citations are not specified in this document. But it updates the previous paper, below }

Improving Performance; Research Needs and the Role of Genetic Modification of Plants

Optimal performance is of course the key to phytoremediation's being able to improve its market penetration. With the possible exception of some systems that are already widely studied and understood (e.g., the use of deep rooted poplars for groundwater control), all of phytoremediation's major applications require further basic and applied research in order to optimize in-field performance. This need was summarized in three of the four recommended areas for further R&D identified by the DOE workshop discussed above (DOE 1994):

o Mechanisms of uptake, transport and accumulation: Better understand and utilize physiological, biochemical, and genetic processes in plants that underlie the passive adsorption, active uptake, translocation, accumulation, tolerance and inactivation of pollutants.

o Genetic evaluation of hyperaccumulators: Collect and screen plants growing in soils containing elevated levels of metals or other pollutants for traits useful in phytoremediation.

o Rhizosphere interactions: Better understand the interactive roles among plant roots, microbes, and other biota that make up the rhizosphere, and utilize their integrative capacity in contaminant accumulation, containment, degradation and mineralization.

All of these recommendations primarily are directed towards basic research, aimed at understanding the mechanisms that underlie the biological processes central to phytoremediation. This is primarily the province of academic researchers, and there is quite a bit of this research taking place. Although it will take time, this information will eventually be obtained.

Another reason for gaining this knowledge is to achieve the means to manipulate or control these processes to improve commercial performance, whether simply through selection and use of optimal plants for given waste scenarios, or through more advanced techniques. There is currently a great deal of such research taking place in academic and non-profit labs around the world, some of which is already finding its way into field testing or commercial programs, and we believe that the impact of this work will soon be felt in the phytoremediation market. General strategies for improving phytoremediation's efficacy are summarized in Table 16.

A number of agronomic enhancements are possible, ranging from traditional crop management techniques (use of pesticides, soil amendments, fertilizers, etc.) to approaches more specific to phytoremediation, such as soil chelators. A number of academic and industrial groups have studied the use of metal chelators such as EDTA

Table 16. Strategies to Improve Phytoremediation.

Agronomic Enhancements

o Improving metal solubility in soils through the use of chelators.

o Combining phytoremediation with other in situ technologies (e.g., electro-osmosis)

o Enhancing phytoremediation processes by using exogenous modulators or inducers, or soil amendments that enhance plant growth.

o Enhancing plant growth and biomass accumulation by improved crop management practices.

Genetic Enhancements

o Creating improved plants through classical plant breeding

o Creating improved plants through genetic engineering.

Adapted from the framework of Cunningham & Ow (1996).

and hydroxyethylethylene diaminetriacetic acid (HEDTA): according to Cunningham & Ow (1996), these chelators can cause a thousand-fold enhancement in soil solubility of metals such as Pb and can result in significant increases in plant uptake of metals, as measured by concentrations in plant tissue. Chelation research has also been conducted by Phytotech, and by the Kochian group at Cornell, with these groups studying chelators such as HEDTA and citric acid. Many metal phytoremediation researchers believe that use of chelators or related soil amendments will be necessary to promote sufficient metal solubility to cause phytoremediation to proceed at rates fast enough for economical commercial-scale remediation.

It is likely that greater research efforts will be directed at improving the plants themselves, either using classical genetics or the new techniques of genetic engineering. Traditional plant breeding, where different mutants or variants of a given species (or of different related species) are mated and desired traits are selected, is a well-understood, long-standing process for improving plant germplasm. However, it is best practiced with those commodity crops (particularly food or oilseed crops) that have long been cultivated on a large scale, and for which a great deal is known about the genetics of important "background" traits. Many plant species used in phytoremediation do not have this long history of use, nor is there an accumulated base of knowledge of genetics that would allow breeding to proceed smoothly. Traditional crop breeding can also be time-consuming, with several generations needed to introduce stably inherited traits into an existing genetic background. While there may be limited usefulness for traditional crop breeding (e.g., mating a fast-growing variety within an species like B. juncea with a known hyperaccumulating variety of the same species, to try to combine these traits in the offspring), we expect crop breeding to be of limited value in improving phytoremediation in the near future.

More advanced genetic techniques include plant tissue culture, somaclonal variation, and the molecular techniques of recombinant DNA (see Gasser and Fraley 1989 for an early summary). The tissue culture techniques are potentially powerful, in that they allow individual mutant cells to be selected on the basis of a desired phenotype, in much the way bacterial genetics is commonly done. Since techniques for regeneration of whole (fertile) plants from single plant cells are routine for most species, this allows creation or selection of new plant varieties based on biochemical traits first seen in (or introduced into) individual cells. One can therefore use tissue culture to select for cells having enhanced biodegradative properties (for organics) or enhanced ability to assimilate metals, and regenerate new plant varieties based on these selected cells.

Recombinant DNA genetic engineering makes use of the same techniques, combined with the potentially more powerful ability to more selectively and proactively choose the traits to be introduced into the plant cell, via the introduction of DNA encoding enzymes or other proteins from other living organisms, or even completely synthetic genes designed to encode enhanced enzymes. To do so, the DNA or gene of interest in spliced into a small, circular carrier DNA molecule known as a vector, where it is placed under the control of genetic regulatory sequences that are recognized by the plant cell. The vector is introduced into plant cells either by physical means (electroporation or via high-velocity microprojectiles shot inside the cell), or biological means (utilizing natural biological systems where bacteria such as Agrobacterium can insert DNA into plant cells, and cause the DNA to be incorporated into plant chromosomes). Upon entry into the cell and integration into the plant chromosome, the desired gene is "expressed" in a subset of the cells (that is, its genetic code is read by the plant cell to cause the synthesis of a protein encoded by the gene); these cells are selected in tissue culture and used to regenerate whole plants for subsequent breeding.

After some years of uncertainty caused by technical difficulties and delays imposed by lack of regulatory clarity, industrial and food-producing crop plants created by recombinant DNA methods are now being used on a large scale in commercial agriculture in the U.S., Europe and elsewhere in the world (Glass 1997a), and the technique is at last beginning to transform plant agriculture, as was predicted in the 1980s by the technology's promoters. We expect genetic engineering to have a similar impact on phytoremediation, and we discuss the prospects for the use of this technology in the next section

Prospects for Genetic Engineering

In spite of the recent commercial success of recombinant DNA in crop agriculture, the technology comes to the remediation business carrying some negative "baggage": much-hyped potential uses of genetic engineering to improve microbial bioremediation have never come to pass, except in basic research in microbial biodegradative pathways. There are many reasons why this is so, including the time and cost needed to engineer metabolic changes in bacteria, the lack of a compelling need to turn to genetic engineering to improve microbial performance that, for most purposes, was adequate for bioremediation, and a regulatory and public acceptance climate that was viewed by the industry as hostile to the use of recombinant microbes in the open environment. Although we disagree with the latter view, the combination of these factors presented powerful disincentives to the use of recombinant microbes in bioremediation.

We are quite hopeful, however that unlike the case with microbial bioremediation, genetic engineering will have an important role to play in commercial phytoremediation, perhaps as early as the next 2 to 4 years, if not sooner. Among the reasons for this optimism are the following: (1) genetic engineering of plants is quicker, easier, and more routine than genetic engineering of soil microorganisms; (2) there is a clear need to improve the performance of naturally-occurring plant species, in order to obtain commercially-significant phytoremediation performance; (3) phytoremediation processes are likely to be simpler and easier to understand and manipulate than microbial biodegradative pathways where consortia of organisms are sometimes needed; (4) regulatory and public acceptance barriers are substantially less severe for the use of transgenic plants than they are for engineered microbes, with the precedent that transgenic crop plants have been approved for commercial sale in agriculture in the U.S., Canada, Europe and elsewhere in the world.

Many of the naturally occurring plant species used in phytoremediation activities to date belong to species that can be genetically engineered, including Brassica juncea, Helianthus annuus, and poplars. In general, any dicotyledonous plant species can be genetically engineered using the Agrobacterium vector system, while most monocotyledonous plants can be transformed using particle gun or electroporation techniques.

There are several conceivable strategies for the use of genetic engineering to improve phytoremediation, as shown in Table 17. Strategies to enhance metal phytoremediation can be grouped in several categories. One approach is to enhance the ability of metal ions to enter plant cells, by introducing genes encoding transporter molecules. These are generally proteins that are found in the cell membrane, which have an affinity for metal ions, or which create favorable energetic conditions to allow metals to enter the cell. Examples might include transport proteins such as the Arabidopsis IRT1 gene that encodes a protein that regulates the uptake of iron and other metals (Eide et al. 1996), or the MRP1 gene encoding an Mg-ATPase transporter, also from Arabidopsis (Lu et al. 1997).

Another strategy would be to express proteins, peptides or other molecules within plant cells that have high affinity for metals. Examples might include genes controlling the synthesis of peptides that sequester metals, like phytochelatins (e.g., the Arabidopsis cad1 gene of Howden et al. 1995) metallothioneins or other metalloproteins. In fact, transgenic plants expressing metallothioneins have been created, and although these plants exhibited enhanced tolerance to high metal concentrations, the uptake of metals was not enhanced.

As observed by some researchers in the field, notably David Salt of Northern Arizona University, another need is for better movement of metals into and within the plant biomass. This might be accomplished by creating plants which express metal chelators, which can be secreted into the soil, to enhance root uptake of metals, or expressed within plant tissue to facilitate movement of metals out from the root into above-ground biomass.

Finally, one might introduce genes encoding enzymes that change the oxidation state of heavy metals, like the bacterial merA gene encoding mercuric oxide reductase (Rugh et al. 1996), or that convert metals into less toxic forms, such as enzymes that can methylate Se into dimethylselenate (Hansen et al. 1998). In both these cases, the resulting form of the metal is volatile, so that one can create plant capable of metal remediation by phytovolatilization. Transgenic plants of both types have been created and are being tested in the lab and greenhouse, but as of this writing have not been field tested.

Table 17. Strategies to Improve Phytoremediation using Genetic Engineering.

Metals

Introduce genes encoding transport proteins.
o IRT1 iron transporter
o MRP1 Mg-ATPase transporter
o High affinity Zn transporter from Thalassiosira weissflogii
Introduce genes encoding metal-sequestering proteins or peptides.
o phytochelatins (e.g., cad1)
o metallothioneins
Introduce genes to enhance metal transport into roots, and from roots to other plant biomass.
o genes encoding metal chelators
Introduce genes to change the oxidation state of metals.
o mercuric reductases
o selenium methylation enzymes

Organics

Introduce genes encoding key biodegradative enzymes (plant and microbial origin).
o Laccases
o Dehalogenases
o Nitroreductases
Introduce genes for the stimulation of rhizosphere microflora.

General

Introduce genes to enhance:
o growth rates/biomass production rates
o enhancement of root depth, penetration
Introduce genes encoding insect resistance, disease resistance, etc. to reduce costs of agricultural chemical input, enhance biomass yield.

Sources: Raskin 1996, Cunningham & Ow 1996, Glass 1997a.

Strategies for enhancing phytoremediation of organics are potentially more straightforward. Genes encoding biodegradative enzymes can be introduced and/or overexpressed in transgenic plants, leading to enhanced biodegradative abilities. Such genes can be of any origin -- bacterial, plant or even animal -- or can even be synthetic, perhaps making use of advanced techniques of protein engineering or directed mutagenesis to create optimized catalytic proteins. Transgenic plants altered in these ways are beginning to be created in academic and industrial laboratories

Among the reasons we expect to see use of transgenic plants in phytoremediation is a more favorable regulatory environment than that seen for engineered microorganisms. An exhaustive description of the regulatory framework for transgenic plants is beyond the scope of this report (see Glass 1997a), but the following is a summary of current status. Outdoor uses of transgenic plants have been regulated in the U.S., Canada, and other industrialized nations since the mid to late 1980s, under regulatory systems that have generally allowed small-scale field trials to take place under permits that were relatively easy to obtain. In data running through the end of 1996, several thousand field tests of almost two dozen different plant species had taken place in the U.S. and Canada, and 500 or more in Europe. Moreover, procedures are now in place for applicants to petition for and obtain permission to sell transgenic plants commercially, and through 1996, 27 varieties had received approval in the U.S. and 15 in Canada (Glass 1997a; all these numbers are considerably higher as of the present time).

The situation in the United States has recently gotten markedly simpler: a 1997 amendment to the regulations now allows almost all transgenic plants to be field tested at small scale without a permit, merely upon 30 days advance notice to the U.S. Department of Agriculture. The only exceptions are transgenic plants derived from noxious weeds, which would need a permit for field testing (this might affect some transgenics for phytoremediation). Field tests of transgenic plants have generated virtually no public controversies anywhere in the world, in contrast to public concerns over field uses of engineered microbes (note that we distinguish concerns over field testing from the current concerns in some European countries over food use of transgenic plants, an issue which, while serious, should not affect use of transgenics in phytoremediation).

Complying with the U.S. regulatory regime for transgenic plants has now gotten so simple and inexpensive that even academic laboratories can contemplate undertaking research-stage field tests. Indeed, much of the cutting-edge research with transgenic plants is taking place at universities, and many genes with potential utility in phytoremediation are available for licensing from universities or the federal government (e.g., the IRT1 gene is available from Dartmouth and the University of Minnesota and the MRP1 gene can be licensed from the University of Pennsylvania). The ability of a commercial entity to "license in" potentially useful genes may make it more economically feasible to contemplate use of transgenics in commercial phytoremediation. In spite of the regulatory clarity and compelling need to improve performance, economics remains an important factor in determining whether technological improvements will reach the commercial marketplace. This will be discussed below.

In summary, we expect that transgenic plants will eventually be used to a substantial degree in commercial phytoremediation. Engineered plants have already been constructed, and U.S. field tests or pilot commercial projects could begin as early as 1999 (in fact, one company has claimed to already have conducted trials of plants engineered for improved degradation of organics outside the United States). Once the utility of phytoremediation itself is better established in the market, one can expect to see transgenic plants beginning to be used commercially. As noted above, we believe it is not unrealistic to expect to see commercial use of transgenics within 2-4 years.

PHYTONET has this excellent Phytoremediation links page

PROSPECTS FOR USE AND REGULATION OF TRANSGENIC PLANTS IN PHYTOREMEDIATION [ 1997 © BATTELLE PRESS ]

David J. Glass (D. Glass Associates, Inc., Needham, Massachusetts)

ABSTRACT: The plant species currently being developed for phytoremediation seem capable of effective bioaccumulation of targeted contaminants, but efficiency might be improved through the use of transgenic (genetically engineered) plants. Transgenic plants were first field tested in the United States in 1986, and thousands of research field tests have since taken place in the U.S., Canada and Europe under reasonable regulatory regimes. Many specific transgenic varieties have been exempted from regulation based upon a record of safe research use, and many novel crop varieties are being sold and used commercially. It should be possible to routinely obtain government approvals for field testing and ultimate commercial use of transgenic plants in phytoremediation.

PROSPECTS FOR GENETIC ENGINEERING
All commercial and research activity to date in phytoremediation has used naturally occurring plant species. However, many of these are species that can be genetically engineered, including Brassica juncea, which is being investigated for phytoremediation of heavy metals from soils (Dushenkov et al., 1995), sunflower, Helianthus annuus, being tested for rhizofiltration of uranium (Dushenkov et al., 1995) and poplar trees (Populus deltoides nigra), being investigated for the accumulation of nitrates and other organic chemicals from soil (Schnoor et al., 1995). In general, any dicotyledonous plant species can be genetically engineered using the Agrobacterium vector system, while most monocotyledonous plants can be transformed using particle gun or electroporation techniques.
Genetic engineering might be used to improve heavy metal phytoremediation by introducing biochemical traits that enhance hyperaccumulation. Examples might include genes controlling the synthesis of peptides that sequester metals, like phytochelatins (e.g., the Arabidopsis cad1 gene of Howden et al. 1995), genes encoding transport proteins, such as the Arabidopsis IRT1 gene that encodes a protein that regulates the uptake of iron and other metals (Eide et al. 1996) or genes encoding enzymes that change the oxidation state of heavy metals, like the bacterial merA gene encoding mercuric oxide reductase (Rugh et al. 1996).

REGULATION OF TRANSGENIC PLANTS

United States. Genetically engineered plants are regulated in the United States by the U.S. Department of Agriculture (USDA) under regulations first promulgated in 1987 (52 Federal Register 22892-22915). Although these regulations arose from the debates over "deliberate releases" of genetically engineered organisms in the mid 1980s, field tests of plants have never been unusually controversial (see Glass 1991 for a historical review). Today these rules present only a minimal barrier against research field tests, and also allow commercial use of transgenic plants under a reasonable regulatory regime.
Under these regulations, USDA's Animal and Plant Health Inspection Service (APHIS) uses the Federal Plant Pest Act to regulate outdoor uses of transgenic plants. Originally, permits were required for most field tests of genetically engineered plants. Permit applications must include a description of the modifications made to the plant, data characterizing the stability of these changes, and a description of the proposed field test and the procedures to be used to confine the plants in the test plot. Submitters must also assess potential environmental effects, such as those shown in Table 1. APHIS review of these field test proposals has usually been completed well within the 120 days allowed by the regulations.
Table 1. Key Scientific Concerns:
Field Testing Transgenic Plants.

Introduced DNA
o Characterization of vector system, its stability and expression in the plant cell.
o Avoidance of infectious, pathogenic, toxic or deleterious functions encoded by introduced DNA.
Host Plant
o Reproduction and pollen/seed dispersal mechanisms.
o Ability to outcross with related species (particularly wild relatives).
o Status as a weed, characteristics involved in weediness (ability to compete, survive and spread in the environment).
Environmental Impacts
o Comparison of transgenic to wild type (e.g., competitiveness).
o Unintended effects (e.g. effect on birds, insects, etc.).
o Possibility of gene transfer to other plants.
Test Conditions
o Initial field trials may require monitoring, confinement procedures, pollination controls.
These regulations were substantially relaxed in 1993 (58 Federal Register 17044-17059) to create two procedures to exempt specific plants. Under the first, transgenic plants of six specific crops (corn, soybean, tomato, tobacco, cotton and potato) can be field tested merely upon notifying the agency 30 days in advance, provided the plants did not contain any potentially harmful genetic sequences and the applicant provided certain information and submitted annual reports of test results. The second procedure allowed applicants to petition that specific transgenic plant varieties be "delisted" following several years of safe field tests, to proceed to commercial use and sale without the need for yearly permits. In August 1995, APHIS proposed further revisions under which most transgenic plants could be field tested under a notification rather than a permit, and where annual reports would only be required if the tests showed unexpected adverse effects (60 Federal Register 43567-43573). At this writing, this regulation has not yet been adopted.
The USDA regulations have allowed a large number of field tests to be carried out with moderate levels of government oversight: through October 1995, APHIS had received about 1,800 permits or notifications for field tests of at least 17 different plant species, representing several thousand discrete experiments (USDA 1996, see Figure 1).
Through the end of 1996, 27 different transgenic varieties had been delisted for commercial use (USDA 1996). Although other government regulation may apply for plants used for traditional agricultural purposes (for example, Food and Drug Administration review of food crops), transgenic plants for phytoremediation could be commercialized upon USDA delisting, subject to regulation under the applicable hazardous waste laws.

Regulation in Other Countries. Under a regulatory scheme similar to hat of the U.S., Agriculture Canada has reviewed over 600 submissions since 1988 requesting permission for over 3,000 research field tests, about half of which involved canola (Agriculture Canada 1996). As of July 1, 1996, the agency had granted commercial approval for 15 different plant varieties (Yanchinski 1996).
In the European Union (EU), outdoor use of genetically engineered plants is governed by national laws adopted to implement an April 1990 directive (90/220/EEC). Member states can approve R&D uses within their jurisdiction, after notification to the European Commission (EC) and a 90 day review period. Applications for commercial use are to be made to a single member state, which would forward its recommendation to the EC within 90 days. A product approved by one nation could be marketed across the entire EU in the absence of objections from other states within 60 days. Individual national laws vary widely, with some countries (e.g. France) offering quick, inexpensive reviews and others like the U.K. and Germany requiring six month review periods (Lloyd-Evans and Barfoot 1996). Although environmental activism is still a problem in some EU nations, most has been directed against food uses. Nevertheless, there have been over 500 approved field releases in the EU between 1991 and 1995 (Lloyd-Evans and Barfoot 1996).

FIGURE 1. U.S. Field Tests of Transgenic Plants, 1987-1995, by type of plant (Source: USDA 1996).

In Japan, which has historically been very conservative about allowing open-field experiments with engineered organisms, there have now been five field tests of transgenic plants (OECD 1996).
USE OF TRANSGENIC PLANTS IN PHYTOREMEDIATION
There has not yet been a field test of a transgenic plant engineered for enhanced phytoremediation, although there have been field tests of tobacco plants engineered to express the metallothionein gene in their roots to avoid accumulation of metals in the leaves (USDA 1996). Proposals for research field trials of genetically engineered plants for phytoremediation should be handled routinely by the appropriate agency, and requests for commercial use could be expected to be straightforward once initial trials were conducted safely. The following are some considerations that applicants might keep in mind in preparing for such field tests.

Host Plant Biology. The two most important environmental issues relate to possible enhancement of the weediness of the plant, and its potential to outcross to related species. These will be discussed using examples of plant species likely to be used in phytoremediation: B. juncea and H. annuus.
Weediness. Single gene changes can enhance weediness, although more often multiple changes are needed (Keeler 1989, NRC 1989), but crops that have been subject to extensive agricultural breeding are less likely to revert to a weedy phenotype by simple genetic changes (NRC 1989). However, many plant species used in phytoremediation are not as well-characterized or as long-cultivated as agricultural crop species, and some may be, or be related to, weeds. For example, certain varieties of B. juncea are known to be weeds in the U.S., as are several related species like B. nigra and other species within the Brassica genus (USDA 1994). The genus Helianthus is also known to include a number of wild and weedy species (USDA 1991). It might be necessary to consider whether genes encoding an enhanced hyperaccumulation phenotype would confer any growth advantage or enhance weediness, either through literature evidence or laboratory experiments on the growth rates of the transgenic plants.
Cross-pollination to weedy relatives. A transgene introduced via cross-pollination into a weedy relative might give the recipient plant a competitive advantage or confer a weedy trait. Almost all plants have wild relatives (NRC 1989), so every plant species of commercial utility would have some potential to interbreed with wild, perhaps weedy, species. Commercially-cultivated plants are more likely to cross-breed when grown in the region of the world from which they originated, which is where close genetic relatives would most likely be found.
Different Brassica species can interbreed: for example, it is known that B. juncea and the oilseed crop B. napus can cross, at least where napus is the pollinator, and genetic material can be transferred using an intermediate species to bridge the gap between otherwise incompatible species (USDA 1994). B. juncea is believed to have originated in China, and other important Brassica species either in China, temperate Asia or Europe (NRC 1989), lessening the likelihood of outcrossing to wild relatives for uses in North America and regions of Europe. Brassica are generally pollinated by honeybees, but years of cultivation of rapeseed and other crops have led to a knowledge of the required isolation distances needed to prevent accidental outcrossing from one species to another (USDA 1994).
Cultivated sunflower is known to be capable of crossing to wild and weedy forms of H. annuus. Honeybees are the primary pollinator, and sunflower pollen can be transported very long distances, requiring long isolation distances for commercial breeding (USDA 1991). Helianthus is believed to have arisen in North America, increasing the chance of outcrossing to a wild relative (NRC 1989).
In small-scale agricultural field trials, the possibility of cross-pollination has generally been mitigated by preventing pollination, for example, by bagging or removing the pollen-producing organs or harvesting biomass before flowering. The need for pollen control has generally been relaxed for any given transgenic variety, as more is known about its safety. However, phytoremediation experiments are likely to take place in urban or industrial areas and are likely to be far from related plants. For those uses that happen to be near agricultural areas, the recognized procedures for pollen control can be utilized, particularly if the plants need to be harvested frequently to remove accumulated contaminants.
Cross-pollination to food crops. For phytoremediation, one must also be concerned over transfer of a hyperaccumulation phenotype into crop plants, causing contaminants to enter the food chain. Among the many Brassica species are a number that are used as foods or condiments (e.g., cabbage, broccoli, mustard) and Helianthus tuberosus is the Jerusalem artichoke. However, the proximity of such food crops to a hazardous waste site would itself be more immediately troubling than the remote chance of gene transfer from a transgenic.

Other Issues. For all proposed field tests, regulatory agencies would want to be certain that the products of the introduced genes are not toxic or pathogenic. One concern unique to phytoremediation might be the potential risks to birds and insects who might feed on plant biomass containing high concentrations of hazardous substances, particularly metals. Questions relating to the proper disposal of plants after use would also arise, and commercial approvals may require restrictions on the use of the harvested plant biomass for human or animal food. However, such concerns would be common to all uses of phytoremediation.

IMPLEMENTATION PLAN
Government regulation should not be a significant obstacle to the use of transgenic plants in phytoremediation. Applicants should consult with the appropriate regulatory agency in advance of any actual permit request, to ascertain the data the agency expects to see, which should not take longer than 2 to 3 months to obtain. Permit applications should be filed approximately 4 months before the planned starting date, but in most jurisdictions, approval for research field tests should be routine. As commercial use of a given variety approaches, close consultation with the regulators is also recommended. One can expect the agricultural review agency to consult with any applicable hazardous waste regulators, to ensure proper communication.

REFERENCES

Agriculture Canada. 1996. Field Trials and Registered Products of Biotechnology Under Agriculture and Agri-Food Canada's Acts, available at http://aceis.agr.ca/./fpi/agbiotec/fieldt96.html.

Glass, D. J. 1991. "Impact of Government Regulation on Commercial Biotechnology." In R. D. Ono, (Ed.), The Business of Biotechnology: From the Bench to the Street, pp. 169-198. Butterworth-Heinemann, Stoneham, MA.

Dushenkov, V., P. B. A. Nanda Kumar, H. Motto, and I. Raskin. 1995. "Rhizofiltration: The Use of Plants to Remove Heavy Metals from Aqueous Streams." Env. Sci. Tech. 29(5): 1239-1245.

Eide, D., M. Broderius, J. Fett, and M. L. Guerinot. 1996. "A Novel Iron-Regulated Metal Transporter from Plants Identified by Functional Expression in Yeast." Proc. Natl. Acad. Sci. 93(11):5624-5628.

Howden, R., P. B. Goldsborough, C. R. Anderson, and C. S. Cobbett. 1995. "Cadmium-Sensitive, cad1 Mutants of Arabidopsis thaliana are Phytochelatin Deficient." Plant Physiol. 107:1059-1066.

Keeler, K. H. 1989. "Can Genetically Engineered Crops Become Weeds?" Bio/Technology. 7(11): 1134-1139.

Lloyd-Evans, L. P. M., and P. Barfoot. 1996. "EU Boasts Good Science Base and Economic Prospects for Crop Biotechnology." Gen. Eng. News. 16(13): 16.

National Research Council. 1989. Field Testing Genetically Engineered Organisms: Framework for Decisions. National Academy Press, Washington, DC.

Organization for Economic Cooperation and Development. 1996. OECD's Database on Field Trials, available at http://www.oecd.org/ehs/biotrack.

Rugh, C. L., H. D. Wilde, N. M. Stack, D. M. Thompson, A. O. Summers, and R. B. Meagher. 1996. "Mercuric Ion Reduction and Resistance in Transgenic Arabidopsis Thaliana Plants Expressing a Modified Bacterial MerA Gene." Proc. Natl. Acad. Sci. 93(8): 3182-3187.

Schnoor, J. L., L. A. Licht, S. C. McCutcheon, N. L. Wolfe, and L. H. Carreira. 1995. "Phytoremediation of Organic and Nutrient Contaminants." Env. Sci. Tech. 29(7): 318A-323A.

USDA. 1996. Biotechnology Permits Homepage, available at http://www.aphis.usda.gov/bbep/bp.

USDA. 1991. Environmental Assessment and Finding of No Significant Impact, Permit 91-067-01 to Pioneer Hi-Bred, available at USDA (1996).

USDA. 1994. Response To Calgene Petition 94-090-01p For Determination Of Nonregulated Status For Laurate Canola Lines, available at USDA (1996).

Yanchinski, S. 1996. "Canada Watch." Gen. Eng. News. 16(18):11, 31.