The Ecology of Genetic Engineering

Stephen Nottingham GENESCAPES: The Ecology of Genetic Engineering (Zed Books, 2002) paperback £12.95/$19.95, hardback £12.95/$65

Genetic modification of plants has formed part of agricultural technology for many thousands of years. In recent decades, however, an exponential growth in understanding of molecular biology has enabled scientists to develop techniques which differ markedly from anything previously applied. In the past, farmers could choose to propagate plants from stock which exhibited desired traits such as high productivity, hardiness in the face of adverse conditions of, for example,  cold or drought, suitability for particular processes such as wine production or bread-making, or simply for flavour.  The mechanism of heredity was discovered by Gregor Mendel during the 1860s, though his work was not generally known until the twentieth century. Nevertheless, effective techniques had been developed on the basis of trial-and-error and careful observation, refined since the 17th century by scientific methodology. Indeed, Mendel himself was the product of a region and a family with a long history of involvement in plant “improvement”.

The possibility of using a growing knowledge of heredity as the basis for new methods of plant propagation and trait selection moved nearer to reality with the discovery of restriction enzymes by Salvador Luria in the early 1950s.  Bacteria, Luria found, were able to defend themselves against attack by viruses by slicing up the invaders’ DNA. In 1970 Hamilton Smith isolated one such enzyme and demonstrated how it worked. The enzyme attacked the DNA of an invading organism – the E.coli bacterium rather than a virus, but the principle was the same as that observed by Luria twenty years earlier – at particular points, leaving others unscathed. This ability to “recognise” specific sequences of bases provided the means for the genome mapping which, following another three decades of work, has now been successfully completed for a number of species, including our own.

The usefulness of restriction enzymes, however, goes far beyond their employment as a research tool. Having chopped DNA into fragments, the genetic engineer can then recombine these fragments, or combine fragments from different individuals or organisms of different species, producing the recombinant DNA around which modern agricultural biotechnology, and many other uses of genetic manipulation, revolve. Genetic engineers use vectors – usually a bacterium or virus - to transport and implant DNA from one organism into that of another.

The proponents of agricultural biotechnology argue that genetic modification through modern techniques such as gene-splicing is in essence no different to traditional practices. Some go so far as to assert that, because it makes possible greater precision in the selection of traits, it is inherently more controllable and therefore actually reduces any possibility of risk to the environment or to health. Notwithstanding the fact that there are indeed similarities between genetic modification through modern biotechnology and previous techniques of improvement, there are a number of reasons why this view is disingenuous. Stephen Nottingham, whose book attempts to take a dispassionate look at the issues raised by genetic modification of organisms, cites a number of these.

Though earlier methods also involved the rearrangement of genes in an attempt to improve on the product of natural selection, they dealt in almost all cases only with properties already present in the organism: bigger plants could be produced by breeding from the heftiest individuals, hardier ones by taking those which survived the worst frost or drought, and so on. Like is crossed with like. Supporters of genetic modification as an alternative to this traditional method are fond of pointing out exceptions, and it is true that there have been many. New species of plant occur in nature and can be produced by artificial means. In modern times, mutagens have been employed to create new species.  However, whilst it is true that the idea of a “species barrier” – an inviolable division between different species – is less an “iron law” than a rule of thumb, even if the traditional breeder succeeds in marrying two species of plant, he or she can do so only in a relatively haphazard way. Unless genetic engineering is employed, offspring will take 50% of their genes from each parent, and there is traditionally no way to predict what traits will be taken from which parent. Through genetic engineering, however, scientists can control precisely which genes they introduce to which plants. What is far more difficult, however, is predicting and controlling precisely how those genes will be expressed. A gene which causes one species to develop a certain trait may not have the same effect in another. Moreover, understanding of how genes affect an individual’s genotype remains limited. Numerous examples exist of genes which produce desirable effects in one set of environmental circumstances and undesirable ones in another. Eventually this problem may be at least partly solved, and genetic engineers will be able to add or remove traits from plants with the ease of a dextrous child building models from Lego. We are, however, some way from this level of understanding.  Scientists know how to add genes, remove them, or alter them. They can even create new ones, and are able with some degree of accuracy to predict what will happen as a result. Genetic engineering has the capability, arguably, to become a precise technology with its basis in a precise science, but much work must be done before it realises this potential. Unforeseen effects of genetic manipulation provide powerful support for the argument that, whatever the future may bring, the correct place for such experimentation is not an open field but a highly secure laboratory.

One thing which genetic modification through modern biotechnology does then have in common with traditional techniques is the unpredictability of the consequence of any intervention. This may seem an extraordinary statement, and would certainly be fiercely contested by GM’s supporters. One of the basic argument used in GM’s favour is that, unlike previous improvement techniques, it enables a high degree of precision. The properties of a gene can be analysed and that gene can then be introduced, lending those properties to a new species. This is true so far as it goes. Unfortunately for those who would convince us that GM offers untold benefits with minimal risks, this does not tell the whole story. If a genetic change, whether intentionally or unexpectedly, enables a plant to survive conditions lethal or discouraging to its parent, then it may become invasive, posing a threat to other varieties or species, whether cultivated or natural. This may occur as an unforeseen consequence of the modified plant itself, or by means of hybridisation with wild relatives growing in the region.

Evidence from the behaviour of crops in the wild and from accumulating experience of GM crops demonstrates that genes will certainly, in some cases, move from one to another species. This is more likely to happen with some plants than with others, and the dangers of some instances of gene-hopping are more evident than those presented by others. Almost all major crops hybridise with wild relatives where these grow nearby. If a crop plant has no relatives growing in the wild in a particular region, then the likelihood of gene transfer are vanishingly small or non-existent.  If wild relatives do flourish nearby, however, then gene transfer may occur, with the likelihood of its doing so determined by numerous climatic and topographical factors, as well, importantly, as the particular system of reproduction favoured by a species.

If all honest scientists admit that gene transfer sometimes occurs, then the important issue is simply whether or not this is dangerous and whether the risks it may pose are acceptable.

The first question which must be answered, then, is whether a transgene will survive in the wild. If it cannot reproduce as efficiently as the wild parent stock then it will be selected out, and is unlikely to cause problems. If, however, the transferred gene offers the inadvertently modified plant some advantage, then it will under normal conditions spread, possibly displacing its parent.  Pro-GM writers will tell you that this is unlikely. If, for example, a plant is modified to produce a pharmaceutical, or resist a certain herbicide, this might well be at the cost of some essential trait without which the modified version will not long survive in the wild. Other qualities, however, such as drought-hardiness, cold-hardiness, disease-resistance or the production by the plant itself of a pesticide, could confer advantages to contaminated plants. In one field trial, experimental modified sunflowers hybridised with wild sunflowers, bringing about a 50% increase in seed production, a feature which could lead to the displacement of the wild parent. (New Scientist, Aug 17.02, p.11) One of the scientists working on this experiment admitted that the increase had been quite unexpected and that it had occurred because of the destruction of parasitic insect grubs which lived in the stems of unmodified sunflowers. . (“Dangerous liaisons” New Scientist, Aug 31.02, p.40) The researchers were entirely ignorant of the presence of these insects when they began the experiment, a fine example of the fundamental problem underlying biotechnological research outside a secure laboratory environment. Outside the dishonest, blustering world of agricultural biotechnology’s corporate propagandists, discussion of the possible consequences of genetic modification is conducted entirely in the conditional and subjunctive modes: it abounds with words such as “could”, “possibly”, “probably”, “likely”, “unlikely”, “may” and so on. Honest opposition is no different. Opinion regarding GMOs is always in some measure speculative, and often purely so. We simply do not know what the full range of effects of a particular modification, deliberate or providential, will be.

Ecological scientist Stephen Nottingham lists some of the possible problems attendant upon the adventitious contamination of wild plants by genetically modified relatives. These include “cascading trophic interaction”, in which changes to a plant make it more, or less, suitable as food for a particular herbivore and this in turn affects the species which prey on the herbivore. In turn, other species, such as parasites on one or the other, may be affected. The cascade may fall away until it is quite out of sight, causing all manner of changes to an ecosystem, not all of which, probability and logic dictate, will be unnoticeable or benign.  Nottingham cites pollinating insects as a particular area of concern. “If transgenic crops were to harm honeybees, for example via an insecticidal protein expressed in their pollen,” he warns,  “the pollination of a wide range of crops and native flowering species could be affected.”

Despite such unknown risks, experiments in plant modification continue, entering areas where the dangers are clear to anyone whose mind is unclouded by anticipation of the wealth and prestige which their success might bring. To date, most genetic modification of crop plants has been aimed at making them resistant to herbicides. The goal of the companies involved in such research is to make farmers doubly dependent upon them. Farmers buying GM seeds are generally bound by contract not to save seed for replanting, and any plants resulting from the spread of GM seed to a neighbouring farmers land must be removed by that farmer if he or she is to avoid being sued for patent infringement. According to a Canadian judge, this is true even if the farmer was in no way instrumental in contaminating his or her own fields. At the same time, by conferring resistance only to a specific herbicide, a company can ensure that both seed and herbicide are supplied by themselves and no-one else.

The successful (in its own terms) application of modern biotechnology to agriculture requires ingenuity, patience, and hard work. Although patience is a virtue not readily associated with profit-hungry corporations, it can sometimes be fostered by the prospect of great wealth. Fortunately, no obvious application of genetic manipulation can guarantee this. The technology must then be adapted, aimed at producing the quickest and largest possible returns on research and development investment. This, rather than the claimed desire to “feed a hungry world” or “eradicate disease” is why most of the research effort of biotechnology corporations concerned with farming goes into developing means to trap farmers in a sort of modern version of the crop lien, the system of debt peonage which replaced cruder forms of slavery after the U.S. civil war.

Despite these drawbacks farmers may be attracted to such transgenic crops for their labour-saving and apparent input-reducing qualities. Herbicide specificity, the route to easy profits, is possible because herbicides work in a variety of very different ways. Most of the introduced genes have been isolated from soil-dwelling bacteria which have adapted through natural selection so that they are immune to the effects of certain herbicides.  Nottingham provides a useful description, accessible to the lay reader, of how various herbicides work.

Supporters of GM claim that herbicide resistance will allow more intensive spraying and by this means reduce total input, to the advantage of both environment and farmer. However, it is by no means certain that this will be the result, and any such effect could prove short-lived. Herbicide-resistant crops by definition remove one of the constraints on spraying, for if crops are not susceptible to damage there is no effective maximum amount which can be applied to them.  Though farmers will hardly buy more pesticide than they believe they need, if high-volume application is seen to increase yields and therefore income, then amounts sprayed will rise. In addition, the possibility of producing crops which are resistant to the effects of a herbicide risks creating entirely novel uses for that herbicide, increasing overall demand even if, in some cases, it may result in a reduced need measured per hectare.  Even if producers choose to concentrate on plants engineered to resist the least environmentally-damaging herbicides, there is no substance which does not create problems. Serious health problems for those working with it, the pollution of water courses, disruption of food chains, increased pressure on endangered plants, and toxic effects on the microscopic soil organisms which plants need to thrive, have all been associated with glyphosate, widely regarded as relatively benign and better known to the general public as Monsanto’s Roundup.

Pesticide resistance built into crop plants renders them immune, or partially so, to the effects of one particular weedkiller. Yet it is a recognised part of good farming practice that pesticides should be rotated, as this avoids the target unwanted plants’ themselves developing resistance. Alternatively, the destruction of one weed species can create an ecological niche into which another plant, already immune to the effects of the poison employed, can move. Indeed, one farmer’s weeds might be her neighbour’s proudly herbicide-resistant crop plant, or, through hybridisation with wild relatives, its parent.

Similar arguments can be used against the genetic modification of plants in response to the dangers presented by insect pests. The only such plants currently commercially available derive exclusively from the soil-dwelling bacterium Bacillus thuringiensis (Bt). Insects unaffected by Bt, and there are many, will still need to be controlled through the application of insecticide.  The pests themselves might develop immunity to Bt, and if such insects spread they could create havoc amongst crops and other plants for which the bacteria provide a naturally-occurring protection, a problem which would be particularly worrying to organic farmers but which might also threaten wild flora. As well as employing farming methods which encourage their crop plants to develop an effective armoury against pests, organic farmers also sometimes employ Bt directly as an insecticide.

Manufacturing corporations and regulatory authorities are aware of these potential problems and have devised strategies to counter them. Populations of the target pest species are maintained in GMO-free “refuges”. This will work, however, if at all, only if the refuges are sufficiently extensive, and there is currently no scientific consensus on what this might mean in practice. As a strategy, refuge-maintenance also rests on the assumption that insects which remain vulnerable to Bt will mate with those that are resistant, and that if they do so the immunity will be passed on to those of the next generation. If a gene is “dominant”, it will be passed on to offspring and “expressed” rather than simply carried in a form which has no effect on the individual’s genotype. If it is “recessive”, however, it will be passed on but never expressed, as when a mother carries a disease-inducing gene which leads to sickness only in males.

In terms of their potential to damage the environment through destroying the balance of particular ecosystems or pushing rival indigenous species towards extinction, the most dangerous traits which might be introduced are also amongst the most attractive to agricultural corporations.  If the modification is designed to improve a plant’s response to certain conditions, such as extreme cold or salinity, the dangers are clear. Invasive plants are nothing new: climate change, the migration of people or animals, trade and sheer chance have all led to the introduction of species into new environments with consequences ranging from the benign to the unnoticed through to the catastrophic.  As Nottingham explains, “If GMOs enter a new range, many of the constraints on individuals, for example those imposed by competitors or predators, are no longer present, and they could become invasive.” Of course, where technology creates problems it can sometimes provide the solutions to go with them. In this case, however, there has been very little research into either the problem or how to solve it.

Gene transfer does not occur exclusively between domesticated and wild species, but may allow one crop to contaminate another. This may also have unfortunate economic consequences, with unintended crossing damaging the crops of conventional or organic farmers nearby. Each plant species and individual, whether cultivated or wild, forms part of a complex of ecological relationships linking the tiniest virus to the predator which tops the particular ecology’s food chain. The results of even a small change are therefore extremely difficult to foresee and in many cases, short of some breakthrough in science comparable to those which made genetic modification through modern techniques possible, quite unknowable. Of course, the dangers attendant upon this do not come exclusively from GM species. A species modified by traditional techniques might also upset the balance of an ecology. Famously, the deliberate or accidental introduction of a new species can lead to devastation of the pre-existing balance. The addition of a powerful and only partially understood tool which enables us to generate new species does, however, greatly compound this danger. Insect geneticist Marjorie Hoy recently summed up the current situation perfectly when she said that “The science of putting genes in is far ahead of the risk assessment research.” (New Scientist, Aug 31.02, p.41)

More plausible than the wholly fatuous and transparently self-serving rhetoric of those who claim that GMOs are the key to ending world hunger, is the notion that through modern biotechnological techniques plants could be modified to produce a range of valuable substances including oils and other substances for use in industrial processes, so-called “functional foods” and pharmaceuticals.  Here, however, the dangers are gross and obvious, in that contamination by a food by another food is altogether a lower scale of threat than would be contamination by an intended pharmaceutical.

Nottingham’s excellent survey covers all of these dangers and more, with chapters on trees and on fish in which he deals with the specific problems associated with the genetic manipulation of these groups of organisms. It is well-organised and comprehensive, with a quiet, reflective tone which offers a welcome break from the usual decibel level of this particular non-debate. Indeed, if there is a criticism to be made of the book it is that this is here and there taken a little too far, that having grown tired of the tidal waves of rhetoric which come from all sides whenever this subject is raised, this reader found himself nevertheless longing for just a little passion to peep through Dr. Nottingham’s measured phrases. But this would be churlish. As a committed opponent of GMO cultivation, I have grown tired of listening to people who are supposed to be on my side but who are so ill-informed as to constitute a liability. The ones who think it is something to do with what is and isn’t “natural”, and those who concern themselves with the preferences of one or another supreme being are of course irrelevant. The rest of us should get up to speed with the science - which is more deliberately mystified than inherently difficult - the better to reveal the audacious dishonesty of those who would govern the world, or at least its harvests.

The reviewer, Steve McGiffen is Spectre’s editor and an environmental adviser to the European United Left/Nordic Green Left Group of Euro-MPs. He is currently working on a book for Pluto Press, the working title of which is Regulating Biotechnology: Corporate Power and the Public Interest

Stephen Nottingham is also the author of Eat Your Genes, which Food Magazine described as “As good an overview of the multiple problems that the new technology is spawning as one is likely to read.

You can find out more about these and other titles from Zed Books at