Ecological impact of transgenic plants

One of the greatest fears associated with transgenic plants is what could happen to nearby ecosystems if the transgene escapes from the fields into the wild.  If transgenic plants mate with similar wild plants, the transgene could be incorporated into the offspring.  It is possible that these new plants with the transgene could become weedy or invasive , which could reduce biodiversity and might destroy entire ecosystems.

Transgenes will only have the potential for significant ecological impact if they can increase in frequency and persist in natural populations. Thus we can determine their potential for ecological impact by addressing each of the necessary conditions for transgene escape and spread in natural populations:

1. Can transgenes escape from the field into the wild?
2. Is the transgene expressed in hybrid plants?
3. Do the transgenic plants have an advantage over normal plants of the same species in the wild?

We will address each of these questions in turn.

Transgene escape

1.  Can the transgenes escape from the fields into the wild?

            The first and most important question is if the transgenic plants can escape from the fields.  Escape can happen several different ways.  First, the transgenic seeds can end up being carried out of the field by wind or animals, or can be spread through volunteer plants, which grow up from the seeds dropped by the previous year's harvest².  Second, transgenic plants in the field can hybridize with neighboring wild relatives, and spread the transgene that way³.  This is the method of transgenic plant escape that has the greatest potential to spread the transgene.

            The reason that hybridization is so much more worrisome is that it has the potential to create so many more transgenic plants.  This is because of the large number of seeds produced by most crop plants.

There are three possible avenues of hybridization:

1. Hybridization with non-transgenic crop plants of the same species and variety.
2. Hybridization with wild plants of the same species.
3. Hybridization with wild plants of closely related species, usually of the same genus.

However, there are a number of factors which must be present for hybrids to be created.

• The transgenic plants must be close enough to the wild species for the pollen to

reach the wild plants.

• The wild and transgenic plants must flower at the same time.
• The wild and transgenic plants must be genetically compatible.
• The hybrid offspring must be viable, and fertile.
• The hybrid offspring must carry the transgene.⁴

       Transgenic and non-transgenic plants of the same species only have to overcome the first barrier in order to produce hybrids because they are the same species and therefore interbreed easily ^2,5.

       Most transgenic species are also fertile with their closest wild relative, which is generally the same species as the cultivated and transgenic plants.  Thus where ever transgenic plants are grown along side wild populations of the same species, hybridization is virtually guaranteed ¹.

       This interbreeding between wild and cultivated plants has been seen in sunflowers ^5-7, oil-seed rape (Brassica napus) ², rice ³ and all varieties of beets ⁴, as well as a number of other species.

       Hybridization with closely related species is less common.  Some crops, such as rice, can hybridize with other species in the same genus ³.  However, cultivated sunflowers (Helianthus annuus) do not interbreed with other species of sunflower (H. petiolris) very well, and oil-seed rape (Brassica napus) does not hybridize with the closely related weed wild mustard (Sinapis arvenis) except under laboratory conditions ^{6, 8}. 

       Thus a studies suggest that the most likely escape route for transgenic plants will be through hybridization with wild plants of the same species.  Since we know that these hybrid plants exist, now let us move on to the next question.

Transgene expression

2. Is the transgene expressed in hybrid plants?

     To answer this question we need to evaluate the phenotype, or physical characteristics, of the hybrid plants. If the transgenic trait is expressed in the hybrid plants, we will be able to see evidence of it over the life cycle of the plant. 

     For a hybrid plant to express the transgene it must first receive the gene from one of its parents. Using genetic testing such as PCR scientists can determine if a plant has the gene in question by looking for genetic markers.

     The next question to ask is whether the hybrid plant expresses the gene.   This is tested by looking at what the gene is supposed to do in the plant.  For example, in the case of the oilseed rape plants with multiple herbicide resistances, seedlings were tested for herbicide resistance by applying a small amount of the herbicide in question to one leaf.  Resistant plants were not affected, while non-resistant plants had wilted leaves ².  While the exact rate of hybridization is not known, approximately three out of four hybrid plants carried the transgene.

     In another experiment sunflowers with transgenic resistance to wild mold were hybridized with wild sunflowers. The resulting offspring were examined genetically for the resistance gene and then exposed to white mold to judge their resistance ⁶.  In this case half of the hybrid plants carried the transgene.  There have been similar experiments with Bt pesticides in oilseed rape and Bt sunflowers ^{7, 9}.  These sunflowers were also found to produce hybrids where 49.3% carried the Bt gene⁹.  In addition to enetic testing for the transgene, the sunflowers were also shown to express the gene by having significantly less insect damage than the wild sunflowers⁹.  This means that the transgene continues to be effective in hybrid plants.

     The evidence shows that transgenic hybrids can be expected to have and express the transgene. So we know that transgenic hybrids will exist, and at least some will express the transgene.

Transgene fitness

3.  Do the transgenic plants have an advantage over other plants in the wild?

  This is a very important question to ask, because if the transgenic plants do not have any advantage due to the transgene then they are not at risk of becoming weedy or invasive. While this question is vital, it has not been studied extensively.  There are, however, a set of generally accepted hypotheses about the type of trait that will and will not lead to increased weedyness.

1. 1. Traits which confer a significant advantage over other plants with respect to non-living (abiotic) factors, such as drought and salt tolerance will increase rapidly in the population, and could cause the plant to become weedy or invasive.  At this point there are crop plants with some tolerance to salt, however none of these plants are transgenic.¹¹
   2. Traits which are only advantageous under selection pressure are likely to increase where ever there is selection for that trait, usually insect, disease, or herbicide resistance.  The frequency of these traits when there is no advantage to them will depend on the cost of the trait, and genetic drift.
      • The trait for Bt-enabled insect resistance is one such trait that should confer higher fitness when the plants are regularly stressed by insect attack.  In an experiment involving the survivorship and reproductive output of two lines of transgenic oilseed rape (rapeseed, canola, Brassica napus) with Bt resistance to insects, plants with the transgene suffered less damage from the insects, and plants with less insect damage were more likely to survive the winter and produced more seeds than the non-transgenic plants, which had serious insect damage⁹.  Thus the transgenic plants had greater fitness than the non-transgenic plants.
      • However, in a similar experiment involving hybrid sunflowers with a transgene (OxOx) which confers a resistance to white mold, the transgenic plants were found to have very little advantage over the wild plants.  The transgene did not always confer resistance to the white mold, and resistance did not significantly alter the amount of seeds produced⁵.  Because the reproductive output was the same, the transgenic plants had the same fitness as the wild plants.  It has been suggested that this resistance was not helpful to the hybrid plants because the wild sunflowers already have some kind of resistance (Personal communication J. M. Burke).
   3. Traits that reduce the plant’s fitness (ability to survive and produce fertile offspring) will not become more common in the population and will not become weedy or invasive¹.  This would include plants with genes for longer shelf life, greater nutritional content, or pharmaceuticals, all of which are costly for the plant to produce, but do not improve its survival.

            At this time there are no reported cases of any transgenic plant becoming weedy or invasive.

Summary

So what does this all mean for the impact of transgenic plants on the ecosystem? There are at least three reasonable conclusions:

1. It is known that most transgenic crop plants have been found to hybridize with wild counterparts, and that the hybrids will have the transgenes.
2. It is understood, as a basic part of population genetics, that the spread of a transgene in a wild population will be directly related to the fitness effects of the gene in addition to the rate of influx of the gene to the population.  Advantageous genes will spread rapidly, neutral genes will spread with genetic drift, and disadvantageous genes will only spread if there is a constant influx.
3. The ecological effects of transgenes are not known, but it is generally accepted that only genes which improve fitness in relation to abiotic factors would give hybrid plants sufficient advantages to become weedy or invasive.  Abiotic factors are parts of the ecosystem which are not alive, such as climate, salt and mineral content, and temperature.

This last conclusion might require some explanation.  The reason that transgenic resistance to drought, salt, or temperature would be much more likely to make weedy plants is that these factors cannot change in response to the new resistance by the plants.  Pathogen resistance works until the pathogens evolve resistance to the resistance.  Pesticides work until the pests become resistant.  And it has been seen that herbicide resistance simply spurs the development of new herbicides. 

            But abiotic factors do not change in response to the development of new plants, and so resistant plants will have a long term advantage over other species.  This long term advantage is what could have a very negative effect on ecosystems.

Preventing transgene escape

What should be done to prevent the spread of transgenes into the wild?

      At this time the containment of transgenic crops is generally limited to the same physical separation used for most crops.  However other methods are being developed to prevent the spread of the transgene itself.  For the most part these consist of developing transgenic plants which are sterile, and so cannot spread the transgene through reproduction¹⁰. 

      However, this method is not foolproof, and is totally useless when the desired product is the reproductive part of the plant, which is the case for almost all transgenic plants.  It also creates the issue of unfair supply control by the companies which produce the seed, since new seed would have to be bought every year, a serious problem for poor farmers.  Other suggested methods would limit the transgene to the particular tissues to be altered¹⁰.  Most of the proposed methods are still in the early stages of research and development.  At the present time the best method for containing transgenic plants looks to be simple physical containment, by planting the transgenic crop far from any other plants it might hybridize with. 

Future work

1. To prevent the creation of weedy transgenic plants we must study the base plant thoroughly.  Biocontainment measures such as distances between fields of compatible species will only work if we know how far pollen can travel¹⁰.  The case of the Canadian field of transgenic oilseed rape which developed multiple herbicide resistances due to cross-pollination with other transgenic varieties well demonstrates that the measures in use now are not sufficient to prevent gene movement within fields.  The only way to develop guidelines to prevent this hybridization is to conduct field studies such as the one used to determine the introgression (gene exchange) rate between oilseed rape and wild mustard ⁸.  The study looked at the rate of hybridization between plots of oilseed rape and plots of wild mustard, at a number of distances from each other.  While laboratory studies had shown that the two plants produce viable offspring, it was not until the field study that it was known that the actual rate of hybridization was known to be almost zero ⁸.  Therefore we cannot rely solely on laboratory studies to predict hybridization rates.
2. Study the real fitness effects of the transgene in hybrid plants.  As we saw with the white mold resistant sunflowers⁵, just because a trait is expected to increase fitness does not mean that it will necessarily do so in the wild. 
3. Investigate the actually effects of transgenic plants on an ecosystem.  This has not yet been recorded with any transgenic plant, although normal crop plants have been know to become weedy⁴.  While a simple release of transgenic crop plants or their hybrids is not wise, a better experiment might be of the common –garden type, where the transgenic plant would be introduced into a standing ecosystem which is open to the elements and yet is still contained.  The effects of the transgenic plant and its offspring could then be observed over time.  This would give us a better understanding of what happens when plants with novel traits are introduced into the environment.

References

1. Pilson, D. & Prendeville, H. R. Ecological Effects of Transgenic Crops and the Escape of Transgenes into Wild Populations. Annual Review of Ecology, Evolution, and Systematics 0 (0).

2. Hall, L., Topinka, K., Huffman, J., Davis, L. & Good, A. Pollen flow between herbicide-resistant Brassica napus is the cause of multiple-resistant B-napus volunteers. Weed Sci. 48, 688-694 (2000).

3. Olofsdotter, M., Valverde, B. E. & Madsen, K. H. Herbicide resistant rice (Oryza sativa L.): Global implications for weedy rice and weed management. Ann. Appl. Biol. 137, 279-295 (2000).

4. Ellstrand, N. C. in Dangerous liaisons? : when cultivated plants mate with their wild relatives 244 (Johns Hopkins University Press, Baltimore, 2003).

5. Burke, J. M. & Rieseberg, L. H. Fitness effects of transgenic disease resistance in sunflowers. Science 300, 1250-1250 (2003).

6. Rieseberg, L. H., Kim, M. J. & Seiler, G. J. Introgression between the cultivated sunflower and a sympatric wild relative, Helianthus petiolaris (Asteraceae). Int. J. Plant Sci. 160, 102-108 (1999).

7. Snow, A. A. et al. A Bt transgene reduces herbivory and enhances fecundity in wild sunflowers. Ecol. Appl. 13, 279-286 (2003).

8. Lefol, E., Danielou, V. & Darmency, H. Predicting hybridization between transgenic oilseed rape and wild mustard. Field Crops Res. 45, 153-161 (1996).

9. Stewart, C. N., All, J. N., Raymer, P. L. & Ramachandran, S. Increased fitness of transgenic insecticidal rapeseed under insect selection pressure. Mol. Ecol. 6, 773-779 (1997).

10. National Research Council (U.S.). Committee on Biological Confinement of Genetically Engineered Organisms. in Biological confinement of genetically engineered organisms 255 (National Academies Press, Washington, DC, 2004).

11. Netondo, G. W., Onyango, J. C. & Beck, E. Sorghum and salinity: I. Response of growth, water relations, and ion accumulation to NaCl salinity. Crop Sci. 44, 797-805 (2004).