Environmental Risks of GE: why you should take them seriously

E. Ann Clark, Plant Agriculture, University of Guelph (eaclark@uoguelph.ca)

Presented to the Forum of the Toronto Biotechnology Initiative, 15 April 1999, Toronto, ON

The goal of today's presentation is to substantiate the thesis that environmental risks of GE field crops are real, and that it is to your advantage in industry to face these risks forthrightly. No useful purpose is served by continuing to bypass, minimize, and dismiss the potential ecological harm from GE organisms. It is in the best interest of industry to openly acknowledge potential risks and to fund research to quantify, predict, and deal with them.

Risk? What Risk? So, what is all the fuss about? Anyone who has searched for evidence of government funding for risk assessment research (nil), or leafed through government GE promotional literature (CFIA, 1997) or even perused the new AIC Statement on Biotechnology could easily conclude that GE posed little or no risk to anyone or anything.

Winfield (personal communication, 1999) noted that roughly $700 million(1) in government money is allocated to GE research in Canada now, but none is specifically earmarked for risk assessment. No targeted federal or provincial program focuses explicitly on assessment of ecological or environmental risks or food safety hazards attributable to GE.(2) In the US, a paltry 1% of the USDA budget for biotechnology has been expended in risk assessment research starting in 1992 - but even that was imposed by the 1990 Farm Bill, rather than being undertaken in a responsible and pro-active fashion by research policy.

Of course, if you accept the principles of "familiarity" and "substantial equivalence", and evolve a risk assessment protocol founded upon chemical toxicology - a most unhelpful model(3)

- then it is likely that risks or hazards will be obscured. After all, if you are not looking for something, it is rather unlikely that you will find it. But while expedient, is it in your best commercial interest to really believe that GE poses no unusual risks - and liabilities?

Historical Precedents. In marked contrast to the present day, the early history of what I will call "genetic engineering" was characterized by widespread debate within both the scientific community and, thanks to a vigilent media, by the public at large. For example, Erwin Chargaff and Robert Sinsheimer were both members of the National Academy of Sciences who were respected for their fundamental contributions to molecular biology. They were disturbed by the ignorance of GE researchers regarding possible adverse impacts of their research and publicly challenged the wisdom of proceeding with GE research without a stronger understanding of risks. In the mid 1970's,

Sinsheimer wrote: "Is there a potential hazard and if so, is it large? ...many experiments introduce wholly unknown segments of DNA into these organisms, with patently unpredictable results...we know too little of the intricate ecology of E. coli to state with confidence that, in all circumstances, such organisms are at a disadvantage. The longer-term possible evolutionary consequences of introducing an appreciable genetic intercourse between the prokaryotic and eukaryotic worlds are similarly incalculable."

Chargaff wrote a letter to Science, published in June 1976, stating: "What seems to have been disregarded completely is that we are dealing here much more with an ethical problem than one in public health...whether we have the right to put an additional fearful load on generations that are not yet born...Our time is cursed with the necessity for feeble men, masquerading as experts, to make enormously far-reaching decisions. Is there anything more far-reaching than the creation of new forms of life?" (both quoted from Wright, 1996; p.182) 

Public concern mounted as people became aware of GE research already underway, sometimes in their own communities. For example, the City Council of Cambridge, MA imposed a 3-month moratorium on GE research within its boundaries - which included Harvard and MIT - to allow time for further investigation. The city-based Cambridge Experimentation Review Board published a report in January 1977 refusing to accept the NIH argument that GE was a technical issue best handled by technical experts. "Knowledge, whether for its own sake or for its potential benefits to humankind, cannot serve as a justification for introducing risks to the public unless an informed citizenry is willing to accept those risks." (quoted from Wright, 1996; p.222)

Thus, 25 years ago, the development of GE technology and industry was subject to much greater, and more critical scrutiny by all affected stakeholders than occurs today.

So what happened to this societal debate that was so vigorous, and appropriately so? Were the concerns of the rest of the scientific community including some dissident molecular biologists and citizens addressed and resolved? Or.....not?

According to Wright (1996), the open debate about risks from GE research was curtailed by the outcomes generated by a series of three, carefully crafted meetings held in the mid 70's(4). Outcomes of these meetings were heavily referenced in subsequent testimony, policy documents, and official statements to defuse the growing public and scientific debate and achieve closure. Without going into depth, some points are relevant to the question of how risks of GE are still being defined today, and why risk assessment and management is as weak as it unquestionably is. According to Wright (1996),

1. Conflict of Interest. All three meetings (Bethesda, Falmouth, and Ascot) were sponsored by the NIH (National Institutes of Health), which was the primary source of biomedical GE funding in the USA. Thus, the primary funding source - and recipients of that funding - played the dominant role in configuring the meetings, defining the scope and conduct of the meetings, and ultimately, determining the outcomes.

2. Self-Interest vs. Objective Science. The meetings were unannounced and by invitation only, with the chosen participants heavily weighted in favor of current or future recipients of GE funding. Other stakeholders, both within the scientific community and the environmental and consumer groups, were excluded. Apart from the two chairs, the list of participants at the Bethesda meeting remains secret to this day. At the Falmouth meeting, only two known critics of GE research were invited, one of which - Jonathan King of MIT - had to request permission to attend. At the Ascot meeting, even the British Genetic Modification Advisory Group (GMAG) was excluded, leading one GMAG member to state:

"It might be thought a discourtesy to run an international conference on an important policy question without involving the corresponding organization in the host country....indeed, it is hard to see why GMAG should have been excluded, except for the strong representation on GMAG of the members representing employees and public interest. Had GMAG been invited to participate, some of these would certainly have attended, and would have supplied a criticial presence." (quoted from Wright, 1996; p.230) 

The wider scientific community and public at large learned of these meetings only after the fact, and were outraged.

3. Political Intent. The intent of these "scientific" meetings was unabashedly political, as stated in the opening remarks of the chair of the Bethesda meeting, Wallace Rowe (31 August 1976).

"Part of the agenda today is to get you guys involved and get your voices heard, and maybe if the AInfectious Disease Society of America" comes out and says, "By God, if it's just insertion (of foreign DNA) you are talking about, nobody is worried about this mechanism." That carries a tremendous amount of weight, at least to me, if I could say that to the prophets of doom: "Look, these guys have come out and said there is nothing to worry about here, so let's really start and get on with more serious business." That's what I hope we can accomplish."

Rowe's conclusion that the third and final meeting at Ascot had been an "entirely scientific, analytical process" was disputed by a European participant: "It was very obviously a political meeting...The science was not too bad but I had a strong distaste for the way it was managed...We were being used in the name of being a disinterested group of virologists but it was fairly clear by the end of the meeting that (the organizers) wanted to go back with a result that could be exploited for deregulation." (quoted from Wright, 1996; p.243)

The stated objective of avoiding unwelcome external intervention into GE research directions and methods was achieved by channeling the discussion at the meetings with three narrowing assumptions:

  1. All GE research would be conducted on E. coli K12, a strain weakened by years of lab research and hence considered incapable of survival outside the lab.

  3. The only hazards to be considered were those outside the lab - the issue was not exposure but secondary spread - as in epidemics. Containment was key.

  5. GE research would occur only in technologically advanced countries, with advanced public health and sewage treatment facilities which would essentially prevent an epidemic from spreading.
The net effect of these and other initiatives was to defuse critical comment, from both within and without the scientific community, by framing "risks" in a narrow fashion that was readily discounted. The intent was to discourage external regulation and keep control within the research community itself. According to transcripts, one contributor at the Bethesda meeting stated: "...and somehow...try to get the word going around that informed people are really not worried about epidemics...". Another stated "Serious arguments are about this kind of low level thing, but in terms of the PR you have to hit epidemics, because that is what people are afraid of, and if we can make a strong argument about epidemics and make it stick then a lot of the public thing will go away." (quoted from Wright, 1996, p.234-235).

So, have the risks gone away in the intervening 20 odd years? Certainly public perception of risk has, thanks in no small part to the virtual absence of media coverage in both the US and Canada, in marked contrast to the rest of the world. For example, although it was front-page news everywhere else, one would have looked in vain for any significant coverage or debate on the Biosafety Protocol debacle in either the US or Canadian press.

But no, the risks have not gone away. Indeed, as our understanding of the ecological risks posed by GE organisms has deepened, scientific concerns have redoubled.

Blinkered Perceptions of Risk. Scientific discipline (e.g. genetics vs. ecology vs. entomology) seems likely to account at least in part for the inability of those sponsoring and conducting the research which culminated in commercial GE products to "see" ecological risks. A dairy farmer, for example, can look at a herd of 200 cows and even without an ear tag, tell you everything you would want to know (and likely, a great deal more!) about each individual cow - just from her conformation, coloration, and behavior. The farmer will know her sire and dam, her per year production, the performance of her progeny, and so on and so on. To you or me, they all look the same, but the farmer knows what to look for. The same applies in genetic engineering. If you don't "see" the potential risks, you are unlikely to develop a technology that avoids or minimizes them. This is the legacy of having intentionally excluded scientists of other disciplines from contributing to the technology we call genetic engineering.

But exclusion did not stop a host of scientists in the "other" disciplines from studying and expressing their scientific concerns.

It is particularly important to recognize that "science-based" assessments of genetic engineering raise quite serious concerns which must be dealt with. The concerns are biological and ecological in nature - not ethical, moral, social, or political - and it is counterproductive to pretend otherwise.

Tiedje et al. (1989), a group of ecologists funded by the Ecological Society of America, took on the task of synthesizing the then current knowledge of the potential ecological risks of GE organisms (Table 1).

In the interests of space, risk-related research which has been published subsequent to their review will be reported in just two of the several potential risk areas - Pest Creation/Augmentation (ESA 1 and 2), and Ecological Disruption (ESA 3, 4, and 5).

Pest Creation/Augmentation. Transfer of fitness-enhancing genes to compatible recipients, including wild/weedy relatives or neighboring crops, is of concern for at least two reasons: Table 1. Ecological risks of GE organisms, as defined by the Ecological Society of America (ESA) (adapted fromTiedje et al., 1989)
ESA No. Type of Risk
creation of new pests
enhancement of existing pests, through transferral of fitness-enhancing traits
harm to nontarget organisms, as through broadening a host range
disruptive effects on biotic communities, as through competition/interference
disruption of ecosystem processes, as the effect of expressing microbial ligninases on decomposition and cycling
incomplete breakdown of hazardous chemicals (as in reclamation applications)
squandering of priceless biotic resources, as Bt

Risks of outcrossing and gene transfer are commonly discounted by claiming (erroneously) that GE is just an extension of plant breeding, and that centuries of plant breeding have not created weed problems (also untrue; see Rissler and Mellon, 1996), and therefore, neither will GE. Without responding to these specific claims, suffice it to say for the moment that the potential for pest creation by GE entities is not well predicted by the outcomes of conventional breeding specifically because GE creates entirely new constructs. Indeed, given that a pest is any organism that is "out of context", GE entitites are considered more likely to be pestiferous than the products of conventional plant breeding:

"Organisms with novel combinations of traits are more likely to play novel ecological roles, on average, than are organisms produced by recombining genetic information existing within a single evolutionary lineage" (Tiedje et al., 1989).

Crops as Weeds. Volunteer transgenic crop plants are themselves weeds, particularly in crops following canola or other crops which readily shatter. As in the case of any crop that volunteers, the offending party and the "harmed" party are one and the same - the farmer who grew the GE crop.

However, for small-seeded crops, seed can move well beyond the sown field, adversely affecting the livelihoods of other farmers much like pesticide spray drift. And like spray drift, the offending party is culpable - but in this case, the offending party is at least arguably not the farmer who planted the crop but the industry which sold it to him/her. Risk of growing the GE crop is thus externalized to other farmers, involuntarily.

For example, Brown et al. (1996) surveyed 20 fields in Idaho and Washington, and reported finding escaped canola plants in both roadsides and ditches. They interpreted this finding as evidence that seed can move widely throughout a region. Thus, escaped crop plants may constitute a potential weed threat analogous to what has been claimed by a Saskatchewan farmer currently being sued by Monsanto for allegedly retaining GE seed.

Genetic pollution, by which pollen from GE crops inadvertantly pollinates neighboring fields of the same crop, is an additional vehicle by which crops become weeds (MacArthur, 1998a and b). As in the case of seed mobility, this "weed" is in someone else'field.

Outcrossing to pollinate neighboring fields did not originate with genetic engineering, but never before in the history of plant breeding have outcrossing traits been potentially deleterious, and hence, actionable on the part of neighboring farmers. No seed-keeping farmer would object to receiving crop genes for higher yield, reduced lodging, or disease resistance. Conversely, it is hard to imagine a farmer who would not object to receiving genes for herbicide resistance. Or for that matter, any farmer seeking to supply a market that demands GE-free seed would object to receiving any transgenes, of any kind.

Thus, crop plants themselves can be "weeds", either as volunteers in a sown field or as unwelcome seed or pollen interlopers in the fields of others. For agronomic and marketing reasons, the encroachment of GE seed or pollen are actionable by the offended farmer, every bit as much as spray drift or trespassing. The costs of growing GE crops must be borne by those claiming the benefits of the crop, not imposed involuntarily on a hapless community of neighboring farmers.

Superweeds? It is common, but incorrect, to state that there are no weedy ancestors of crop plants in North America. True, they didn't evolve here, but they are here now. Indeed, despite the fact that virtually all North American crops evolved elsewhere, wild or weedy relatives now exist in North America for many important crops (Table 2). Thus, contrary to some sources, weed generation or augmentation is a significant risk in North American agriculture.

Now, the presence of potentially crossable relatives does not necessarily mean that they are sexually compatible or that crossing does indeed occur, or that the trait would persist in the wild community.

Ellstrand (1996) reported, however, that of the 30 most important crops (in terms of acreage grown) in the United States, over half co-occur with at least one wild weedy relative somewhere in North America, and of these, some degree of sexual compatibility with the sown crops was common. He noted further that in some cases, as rye, sorghum, rice, and beets, hybridization with weedy relatives had already led to more aggressive weeds.

It should be further recognized that processes relating to making transgenic plants may, in fact, enhance the potential for outcrossing to wild ancestors. Bergelson et al. (1998) compared the ability of transgenic vs. mutant Arabidopsis thaliana plants to cross with wild-type plants. A. thaliana is a highly selfing species, for which risk of outcrossing is considered negligible. Both the transgenic and mutant "father" plants expressed the same allele - Csr1-1 - conferring resistance to the herbicide chlorosulphuron. Per-plant outcrossing rate was 0.3% for mutant fathers compared to 6% for transgenic fathers - a roughly 20-fold enhancement. Further studies revealed that two transgenic lines differed in their ability to outcross, with rates differing from 1.2% to 10.8% (a 30-fold enhancement).

While A. thaliana is not itself a weed, this study reveals two quite critical dimensions of GE technology:

Table 2. Examples of crops cultivated in the US which /have wild/weedy relatives (same species or genus) in the US (adapted from Rissler and Mellon, 1996)
Fiber Fruit/Nut Grass/ 
Legume Oil Crop Vegetable/Herb Misc.
cotton berries(5) alternative grains(6) alfalfa canola artichoke herbs(7) mustard
flax cherry forages(8) peas peanut asparagus lettuce sugarbeet
. peach millet . sunflower brassicas(9) pepper tobacco
. pecan rice . . carrot potato
. plum small grains(10) . . celery  radish
. walnut sorghum . . cucurbits(11) spinach
. . sugarcane . . eggplant tomato

Traits which are most likely to be retained in crop:weed hybrids are those which confer fitness, as by altering invasiveness, conferring winterhardiness, insect, or disease tolerance, or herbicide resistance. However, as noted by Rissler and Mellon (1996 p.50), neutral transgenes may also persist due to swamping (repeated flow caused by large scale repeated plantings), or by linkage to other valued transgenes. And many current transgenes can indeed be valuable to both volunteers and crop:weed hybrids. Stewart et al. (1996), for example, documented the enhanced fitness of Bt canola vs. unmodified Bt escapes, when subjected to moderate to heavy insect attack. Snow et al. (1996) observed the unintentional movement of rust resistance from cultivated to weedy sunflower, resulting in enhanced fecundity in the crop:weed hybrid where rust was problematic.

One of the ten worst weeds in the world, johnsongrass, hybridizes freely with cultivated sorghum and is likely to retain either neutral or positive transgenes in an agricultural setting (Arriola and Ellstrand, 1997). Paterson and Chandler (1996) emphasized that johnsongrass is problematic not simply in sorghum, but also in maize, soybean, cotton, and other crops. They consider that the exposure of transgenic sorghum to johnsongrass on 8 million acres in the US "presents an enormous opportunity for gene exchange, even if such events are relatively rare."

A few of the many examples of studies into crop:weed hybridization are presented in Table 3.

Table 3. Examples of gene flow, hybridization, and enhanced weediness attributable to transgenic crops cultivated in the US
Crop Weedy Relative or Recipient Distance of Pollen Travel Evidence of Crossing Fitness/Persistence of Hybrid Reference
canola (Brassica napus) wild radish (Raphanus raphanistrum L.) . Male-sterile transgenic plants pollinated by weedy relative - transgenic transmission was maternal F1 hybrids had poor fertility, but became more fertile with successive generations and resembled wild radish Chevre et al., 1997
. birdsrape mustard (B. rapa L.)(12) At least 26 m, but varied with weather and windspeed a few (1:1000 seedlings tested) were herbicide-resistant hybrids 2 F1 plants produced viable seed; weak F1 hybrids but when backcrossed to either parent, were as vigorous and high-yielding as either parent species Brown et al., 1996
. B. rapa . hybrid superior to wild parent during establishment phase . Linder and Schmitt (1995)
oilseed rape (B. rapa) B. campestris . transgenes for herbicide resistance expressed in F1 and BC1 . Mikkleson et al., 1996
radish (Raphanus sativa) wild radish . hybrid showed superior total seed set, compared to wild parent fitness correlates of hybrid exceeded those of wild parent Klinger and Ellstrand, 1994
sorghum (Sorghum bicolor) (2n) johnsongrass (S. halepense)(4n) At least 100 m in both years and at two separate research stations declined with distance, but as high as 2% at 100 m; frequency varied over test statio ns and years hybrids showed no significant increase or decrease in time to flowering; tiller, panicle, or seed production; pollen viability; or biomass - fitness unchanged from johnson
Arriola and Ellstrand, 1996 and 1997
strawberry (Fragaria x ananassa) F. virginiana . species co-exist, are interfertile, have overlapping flowering times, and share common pollinators . Spira et al., 1996
alfalfa (Medica-go sativa) alfalfa Greater than 400 m . . Skinner and Peaden, 1993
potato (Solanum tuberosum) various wild relatives cites reference for pollen movement on bumblebees up to 5 km crossability increased from Rotata to Tuberosa (wild) and Tuberosa (cultivated) . Jackson and Hanneman, 1996
. commercial potatoes at least 1100 m transgenic traits transferred to nontransgenic potatoes . Skogsmyr, 1994.
sunflower (Helian-thus annuus) wild sunflower (Helianthus annuus) At least 1000 m spontaneous hybridization common (a 6.4 km isolation zone recommended for sunflower seed nurseries (Smith, 1978) regional differences in comparing wild to crop/wild hybrid fitness; hybrids marginally less vigorous in some traits; however, hybridization inadvertantly transferred rust resistance which gave a 24% advantage in seed production to the hybrids when rust present Arias and Rieseberg, 1994; Snow et al., 1996
rice (Oryza sativa) wild weed rice (same species) . readily hybridizes progeny show marked morphological convergence to the weedy parent Langevin et al., 1990

In sum, risk of outcrossing to produce new or more problematic weeds is a real issue even in North America - to say nothing of the enormous risk of intentionally targeting the Third World - home of most crop plants - for GE release (see Clark, 1999). Reproductively compatible close relatives exist, and often co-occur, with many important North American crops. Weeds have likely been created or worsened by introgression of crop genes, and many of the genes targeted in GE crops will confer a selective advantage to weeds growing in an agricultural setting. Furthermore, the very process of genetic engineering itself may well exacerbate this problem by promoting outcrossing.

Ecological Disruption

To a very large extent, agriculturalists - including agricultural scientists - are unaware of our enormous dependence upon "nature" to make things work in the world, including in agriculture. For a good general review on this, see Spedding et al. (1981). More recently, a group of 13 ecologists, economists, and geographers headed by Costanza (1997) calculated the dollar value of all the "services" supplied by the natural environment to humankind. They came up with a figure of $33 trillion, in the form of 17 distinct service areas such as N-fixation, erosion control, pollination, and biological control of pests. They argue that it is essential to give due weight to the natural "capital" that supports and sustains so much of human endeavor, to avoid making potentially disastrous policy decisions.

Such matters need to be front-and-center in "risk assessment" discussions, because recent research is showing the potential of GE interventions to disrupt ecological processes, with potentially adverse outcomes. Too often, papers published by GE advocates clearly do not see these risks (Feitelson et al., 1992; Lambert and Peferoen, 1992; CFIA, 1997). A few brief examples will suffice to make the point:

Soil Processes: Genetically engineered Klebsiella planticola , a common soil bacteria, was modified to transform plant residues into ethanol. However, it not only competed well with parental strains, but also with beneficial soil mycorrhizae and actually killed the test wheat plants (Holmes et al., 1998). Unmodified parental strains did not have these effects..

Doyle et al. (1995) reported that Pseudomonas putida which was genetically engineered to break down 2,4-D not only broke down the herbicide but also killed soil fungi which are an essential component of both soil fertility and plant disease protection.

Tapp and Stotzky (1995) demonstrated that active BT-endotoxins (as are synthesized by Bt-plants) can persist for extended periods in the soil, bound to clay particles where they are resistant to breakdown by soil microbes. The accumulation of active BT endotoxins in the soil will have unknown long-term effects on soil processes specifically because the endotoxins in Bt-plant residues are active. Much of the specificity of Bt endotoxins in their natural form arises because they reside in inactive forms which require host-specific solubilization and enzymatic cleavage as well as host-specific intestinal receptors to become an effective and selective toxin. Because of this critical difference, Crecchio and Stotzky (1998) speculated that transgenic Bt crop residues may be much less selective than Bt sprays, indiscriminantly affecting a range of insects. Many species could be at risk which would otherwise have been shielded by the very specific relationship which has evolved between the original Bt organism and and target insect hosts. Unbalancing the soil biotic community could have unacceptable effects on such "natural" services as nutrient cycling and disease suppression which we take for granted(13).

Extracellular recombinant DNA was found to persist for at least 24 weeks in soil in a study reported by England et al. (1996) at Guelph. Similar findings were reported by Widmer et al. (1995), who compared stability of naked, linear plasmid DNA with that of DNA in plant tissues following introduction into non-sterile soil. They found that after an initial rapid rate of decay, a small fraction of recombinant DNA remained detectable in the soil for several months. The concern here is the ability of other organisms to take up extracellular, soil-bound DNA, including genes coding for antibiotic resistance and Bt, with unknown adverse effects on soil processes (Haack et al., 1996).

Donegan et al. (1997) studied the post-harvest effects of proteinase inhibitor I - another insecticidal protein - in buried GE tobacco residues. Compared to unmodified (parent plant) tobacco residues, transgenic residues altered the species composition of the soil biota responsible for organic matter decomposition and nutrient cycling. Species composition and balance in fungal feeding nematodes, which has been shown to influence N mineralization and plant growth, was altered by the transgenic residues.

In the aggregate, then, the potential is clear for GE entitities, both microbial and crop, to affect biological processes in ways which could compromise the ability of the soil to perform the "services" we demand of it.

Biological Control. A key concern with both GE as well as chemical approaches to pest control is unintentional and adverse effects on non-target organisms (see Clark, 1998). Indeed, target-specificity is reportedly one of the chief advantages of the Bt-approach to pest control. But does it work?

Chapman and Hoy (1991) showed that Bt var. tenebrionis wettable powder, which is specific to Coleoptera, was shown to affect not simply two-spotted spider mite (Tetranychus urticae Koch), but also western predatory mites (Metaseiulus occidentalis (Nesbitt). This was a powdered spray, and hence, would have been more selective than a transgenic application. The capacity of Bt-crops to affect beneficial as well as harmful insects would influence its compatibility with IPM and other biocontrol efforts.

In Scotland, Birch et al. (1997) found that ladybugs (Adalia bipunctata) which fed on peach potato aphids ((Myzus persicae) which had in turn fed on GE potatoes(14) produced up to 30% fewer progeny and lived only half as long as ladybugs feeding on aphids which had fed on conventional potatoes. Similar evidence of ecological ramification was reported from Hilbeck and colleagues at the Swiss Federal Research Station for Agroecology and Agriculture. They found that the BT endotoxin from transgenic corn killed most cornborers, but the green lacewings which fed on the cornborers were also killed. In subsequent studies, they found that 50% more lacewings died after consuming catepillers fed on Bt than consuming the purified Bt directly.

In an Internet communication, Altieri (1998) commented that the aphids in the Birch et al. (1997) study had apparently sequestered the BT endotoxin and succeeded in transferring it to predators, in this case, the ladybug. He cited further evidence that the phenomenon of toxin sequestration followed by transfer from herbivore to parasite was not uncommon in nature. Movement of the debilitating impact of BT endotoxins through soil invertebrate communities, with consequent impacts on soil organic matter and nutrient cycling, could have far reaching effects on ecosystem function.

Many other examples exist, particularly in regard to loss of the elegant selectivity of natural Bt when inserted transgenically. As a whole, the speed with which GE entities are being commercialized - without meaningful consideration of risk factors - bears striking parallels to the development and commercialization of biocides. The risks of biocides are being elucidated after the fact in part, tragically, using epidemiological studies on afflicted people (various in Benbrook, 1996; Colborn et al., 1996; Garry et al., 1996; Steingraber, 1996). It might have been hoped that contemporary GE decision-making would have benefitted from the lessons of biocide-based agriculture.


The historical record of scientific and societal scrutiny of the risks posed by genetically organisms gives little comfort to GE advocates and regulators. Risks identified and concerns expressed at the highest levels of the scientific community in the mid 70's have still not been addressed and resolved. Far from it. Yet, despite the virtual absence of government funding toward this end, a very wide range of research reports cited from the last 10-15 years not only support but amplify these historic environmental concerns.

The issues raised should not be pigeonholed as "ethical" or "social" or "consumer" based. Proponents of this technology need to acknowledge the biological and ecological basis for these concerns, and the strength of the scientific evidence which underlies them. Painting the opposition as hysterical greenies does you a disservice.

You are encouraged to take a pro-active stance on these issues. Fund the growing population of researchers who are willing and able to contribute to addressing these risks. Do this in an open and above-board fashion that unites, rather than polarizes, the scientific community and consuming public. Do this in your own best interest. It is time.


Arias, D.M. and L.H. Rieseberg. 1994. Gene flow between cultivated and wild sunflower. Theor. Appl. Gen. 89:655-660.

Arriola, P.E. and N.C. Ellstrand. 1996. Crop-to-weed flow in the genus Sorghum (Poaceae): spontaneous interspecific hybridization between Johnsongrass, Sorghum halepense, and crop sorghum, S. bicolor. Amer. J. Bot. 83(9):1153-1160

Arriola, P.E. and N.C. Ellstrand. 1997. Fitness of interspecific hybrids in the genus Sorghum: persistence of crop genes in wild populations. Ecological Applications 7(2):512-518

Benbrook, C.M. 1996. Pest Management at the Crossroads. Consumers Union, Yonkers, NY.

Bergelson, J., C.B. Purrington, and G. Wichmann. 1998. Promiscuity in transgenic plants. Nature 395:25 (3 Sept 98)

Birch, A.N.E., I.E. Geoghegan, M.E.N. Majerus, C. Hackette, and J. Allen. 1997. Interactions between plant resistance genes, pest aphid populations and beneficial aphid predators. Scottish Crop Research Institute Annual Report. pp. 68-72.

Brown, J., D.C. Thill, A.P. Brown, C. Mallory-Smith, T.A. Brammer, and H.S. Nair. 1996.

Gene transfer between canola (Brassica napus L.) and related weed species. (http://www.nbiap.vt.edu/brarg/brasym96/brown96.htm)

Chapman, M.H. and M.A. Hoy. 1991. Relative toxicity of Bacillus thuringiensis var. tenebrionis to the two-spotted spider mite (Tetranychus urticae Koch) and its predator Metaseiulus occidentalis (Nesbitt). J. Appl. Eng. 111:147-154.

Chevre et al. 1997. Gene flow from transgenic crops. Nature 389:924.

Clark, E. Ann. 1998 Environmental risks of genetic engineering in field crops. Presented to Factoring in the Environment for Decisions on Biotechnology in Agricultural Production. A workshop sponsored by the National Agricultural Environment Committee, August 98 Ottawa. (http://plant.uoguelph.ca/faculty/naec.htm)

Clark, E. Ann. 1999. Debunking the myths of genetic engineering in field crops.
Presented to Alternatives Food for Thought Lecture Series. March 1999. Kitchener.

Colborn, Theo, Diane Dumanoski, and John Peterson Myers. 1996. Our Stolen Future. Plume/Penguin Books, New York.

Costanza, R., R. d'Arge, R. de Groot, S. Farber, M. Grasso, B. Hannon, K. Limburg, S. Naeem, R. O'Neill, J. Paruelo, R.G. Raskin, P. Sutton, and M. van den Belt. 1997. The value of the world's ecosystem services and natural capital. Nature 387:253-260.

Crecchio, C. and G. Stotzky. 1998. Insecticidal activity and biodegradation of the toxin from Bacillus thuringiensis subsp. kurstaki bound to humic acids from soil. Soil Biology and Biochem. 30:463-470.

Donegan, K.K., R.J. Seidler, V.J. Fieland, D.L. Schaller, C.J. Palm, L.M. Ganio, D.M. Cardwell, and Y. Steinberger. 1997. Decomposition of genetically engineered tobacco under field conditions: persistence of proteinase inhibitor I product and effects on soil microbial respiration and protozoa, nematode and microarthropod populations. J. Applied Ecology 34:767-777.

Doyle, J.D., G. Stotzky, G. McClung, and C.W. Hendricks. 1995. Effects of genetically engineered microorganisms on microbial populations and processes in natural habitats. Adv. Appl. Micro. 40.

Ellstrand, N.C. 1996. Potential for crop transgene escape and persistence in weedy populations. (http://www.nbiap.vt.edu/brarg/cris/grant9403.html)

Feitelson, J.S., J. Payne, and L. Kim. 1992. Bacillus thuringiensis: insects and beyond. Bio/Technology 10:271-275.

Garry, V.F., D. Schreinemachers, M.E. Harkins, and J. Griffith. 1996. Pesticide appliers, biocides, and birth defects in rural Minnesota. Environmental Health Perspectives 104(4).

Haack, B.J., R.E. Andrews Jr., and T.E. Loynachan. 1996. Tn916 mediated genetic exchange in soil. Soil Biol. Biochem. 28:765-777.

Ingham, R.E., J.A. Trofymow, E.R. Ingham, and D.C. Coleman. 1985. Interactions of bacteria, fungi, and their nematode grazers: effects on nutrient cycling and plant growth. Ecol. Monog. 55(1):119-140.

Jackson, S.A. and R.E. Hanneman, Jr. 1996. Potential gene flow between cultivated potato and its wild tuber-bearing relatives: implications for risk assessment of transgenic potatoes. (http://www.nbiap.vt.edu/brarg/brasym96/jackson96.htm)

Klinger, T., and N.C. Ellstrand. 1994. Engineered genes in wild populations: fitness of weed-crop hybrids of Raphanus sativas. Ecological Applications 4:117-120.

Langevin, S.A., K. Clay, and J. Grace. 1990. The incidence and effects of hybridization between cultivated rice and its related weed red rice (Oryza sativa L.) Evolution 44:1000-1008.

Linder, R. And J. Schmitt. 1995. Potential persistence of escaped transgenes: performance of transgenic oil-modified Brassica seeds and seedlings. Ecological Applications 5:1056-1068.

MacArthur, M. 1998a. Canola crossbreeds create tough weed problem. Western Producer, 15 October 98.

MacArthur, M. 1998b. Resistant canola expected. Western Producer, 15 October 98.

Patterson, A.H. and J. M. Chandler. 1996. Risk of gene flow from sorghum to Ajohnsongrass@. (http://www.nbiap.vt.edu/brarg/brasym96/paterson96.htm)

Purrington, C.B. and J. Bergelson. 1995. Assessing weediness of transgenic crops: industry plays plant ecologist. TREE 10(8):340-342.

Regal, P.J. 1998. A brief history of biotechnology risk debates and policies in the United States. An Occasional Paper of The Edmonds Institute, Edmonds, WA.

Rissler, J. and M. Mellon. 1996. The Ecological Risks of Engineered Organisms. MIT Press, Cambridge, MA

Salama, H.S., A. El-Moursy, F.N. Zaki. R. Aboul-Ela and A. Abdel-Razek. 1991. Parasites and predators of the meal moth Plodia interpunctella Hbn. As affected by Bacillus thuringiensis Berl.

Skinner, D. and R.N. Peaden. 1993. Risk of transgenic alfalfa dissemination during seed production. (http://www.nbiap.vt.edu/brarg/cris/grant9302.html)

Skogsmyr, I. 1994. Gene dispersal from transgenic potatoes to conspecifics: a field trial. Theor. Appl. Genetics 88:770-774.

Snow, A.A., P. Moran-Palma, L.H. Rieseberg, and G.J. Seiler. 1996. Fitness of F1 wild-crop hybrids in sunflower: a progress report. (http://www.nbiap.vt.edu/brarg/brasym96/snow96.htm)

Spedding, C.R.W., J.M. Walsingham, and A.M. Hoxey. 1981. Biological Efficiency in Agriculture. Academic Press, London. 383 pp.

Spira, T.P., A.G. Abbott, and S. Rajapakse. 1995. Risks of gene escape in strawberry: gene flow, hybrid vigor, and introgression. (http://www.nbiap.vt.edu/brarg/cris/grant9503.html)

Steingraber, S. 1997. Living Downstream: An Ecologist Looks at Cancer and the Environment. Addison-Wesley, N.Y.

Stewart Jr., C.N., J.N. All, P.L. Raymer, and S. Ramachandran. 1996. Increased fitness of transgenic insecticidal canola under insect selection pressure. (http://www.nbiap.vt.edu/brarg/brasym96/stewart96.htm)

Tiedje, J.M. R.K. Colwell, Y.L. Grossman, R.E. Hodson, R.E. Lenski, R.N. Mack, and P.J.

Regal. 1989. The planned introduction of genetically engineered organisms: ecological considerations and recommendations. Ecology 70(2):298-315

Tapp, H. and G. Stotzky. 1995. Dot blot enzyme-linked immunosorbent assay for monitoring the fate of insecticidal toxins from Bacillus thuringeiensis in soil. Applied and Environmental Microbiology 61(2):602-609.

Winfield, M. (1999. Personal communication; Director of Research, Canadian Institute for Environmental Law and Policy, Toronto)

Wright, S. 1996. Molecular Politics Developing American and British Regulatory Policy for Genetic Engineering, 1972-1982. University of Chicago Press, Chicago.


1. $400 million federal, $250 million provincial, plus $55 million for the National Biotechnology Strategy - all of which, apparently, is contingent upon private sector contributions

2.  although risk management is receiving increased research support, given the pending demise of Bt crops.

3.  As currently applied the chemical toxicology model has also proven an inadequate predictor of biocide risks to human health because it ignores synergies among biocide residues, interactions with subject age and physiological state, and restricts consideration to a relative few of the various potentially adverse effects on immune system function, behavior, and reproduction. As discussed by Regal (1998), current GE environmental risk protocols also fail to assess risk in an ecologically and agronomically meaningful setting/context

4. Enteric Bacteria Meeting in Bethesda, MD in August 1976; Workshop on Risk Assessment of Recombinant DNA Experimentation with Eschericia coli K12 in Falmouth, MA in June 1977; and the US-EMBO Workshop to Assess the Containment Requirements for Recombinant DNA Experiments Involving the Genomes of Animal, Plant, and Insect Viruses, in Ascot, England, January 1978

5. Blackberries, blueberries, cranberries, raspberries, strawberries

6. Amaranth, quinoa

7. Cilantro, parsley

8. Bluegrass, fescue, ryegrass

9. Broccoli, cabbage, cauliflower, turnip

10. Barley, oats, rye, wheat

11. Cucumber, melon, pumpkin, squash

12. Under natural field conditions, black mustard, wild mustard, and birdsrape mustard were found to hybridize and produce seed more readily than canola x weed species

13. Should agriculturalists and others worry about unbalancing an ecosystem? See Clark, 1998 for explicit examples of the dire implications that have evolved from just such a situation.

14. Engineered to contain lectin from snowdrops, which is known to interfere with insect digestion