The latest thing in genetic manipulation is RNA interference: this is even less predictable and controllable than artificial DNA.
The first commercial GM crop, back in the 1990s, was a long-life tomato which had had one of its genes switched off. The engineered DNA which achieved this wasn't a gene, but a 'back-to-front' version of one of the plants own genes. The effect of the 'backwards' DNA was to interfere with the expression of a key gene necessary for the ripening process, causing a delay in softening of the tomatoes. The only artificial gene this plant contained was an antibiotic resistance gene, and the risk assessment successfully focused on this side issue and on the artificial DNA construct, ignoring any effects of the actual transformation process. Despite these limitations, all subsequent GM risk assessments seem to have been largely modelled on this first one. The long-life GM tomato was a biological and commercial failure.
Since then, the biotech industry has become pretty much fixated on three types of genetic manipulation: 'Bt' look-alike genes which are toxic to insect pests; toxin-neutralising genes which allow the crop to survive weedkillers; and paired fertility-adjusting genes which control cross-breeding of the GM plants. Commercial crops with other artificial characteristics inserted seem thin on the ground, a fact which has raised speculation that genetic transformation is an inefficiently unreliable technique which should be discontinued in favour of other more successful and less intrusive modern breeding technologies.
However, there are signs that genetic engineers are shifting their attention away from the pasting of re-invented genes into plants, and abandoning the unpredictable 'backwards' DNA to switch genes off. They are devising more sophisticated versions of DNA technology to interfere with the normal functioning of genes.
Far from being a few genes sitting in a sea of 'junk' DNA, the genome science is now uncovering is a plethora of different types of DNA with different functions.
Some DNA, the type mimicked by genetic engineers, causes a protein to be generated by the cell. The DNA achieves this by constructing and sending out a messenger molecule related to itself, referred to as 'RNA'.
A second type of DNA is not a gene and doesn't generate a protein, but it still functions by sending out RNA. The purpose of this type of RNA is to control all the on-going physiological processes in the cell which it does by interacting with the activities of the DNA and the other RNA.
Unlike DNA which is a highly stable molecule, these various RNAs don't usually hang around long in the cell: they act and are quickly destroyed. As a result, they have been very difficult to study.
One new focus of genetic engineering has been to insert artificial DNA which produces an RNA designed to interfere with the expression of a specific gene. Such techniques are receiving major attention in a clinical setting; for example, gene-therapy to knock out 'faulty' genes or genes associated with disease is being tried out. In the world of plants, biotech research has focused on how to keep the active RNA molecule, which is inherently unstable, intact long enough to be useful outside the plant.
The answer so far seems to be to create paired RNAs which stabilise each other by sticking together, and can even resist digestion. This type of 'double-stranded' RNA is found in nature only in certain viruses, referred to as 'retro-viruses'.
You are probably familiar with retro-viruses, because this group includes leukaemia- and AIDS-causing viruses. Some retro-viruses can integrate into the host genome and be passed to the next generation, and infectious particles are known to be passed on in breast milk.
They can, of course, only be pathogenic if the infected host's immune system fails to identify and destroy them. A plant or animal which recognises a piece of double-stranded RNA inside it will trigger a virus-alert and get rid of the intruder. Genetic engineers have, however, successfully introduced a variety of structural tricks in their novel RNAs to evade the immune response in both plants and animals.
GM crops of the future look likely to contain two pieces of artificial DNA, one for each RNA strand. It seems there is no need for them to be close to each other in the genome, but they will, of course, have to be accompanied by the usual consort of viral promoters to make them work, marker DNA to see if they're there, carrier DNA to get them in, and joining-up sections of DNA etc. The double-stranded RNA generated by the crop will be designed to interfere with a key gene in the metabolism of an insect pest, which will die after eating it.
The Institute of Science in Society has long been warning that, after the usual denial of any unwelcome science relating to GM, the RNA designed to selectively knock out one specific gene is not so specific after all, and interferes with a whole range of other, non-target, genes. Would the same stabilised and disguised RNA which kills insects quickly, make humans very ill, but slowly?
Also of concern is the creation of a crop generating large quantities of material peculiar to pathogenic viruses. This is made worse because they are attached to viral promoters which can be active in any organism, and the fact that retro-viruses are notorious for their rapid genetic variation. Even if defective, viruses can make use of other 'helper' viruses or foreign viral material in their surroundings to make up for deficiencies, in fact, many pathogenic viruses are defective.
It's not difficult to envisage how the links between natural pathogenic particles abounding in the environment and a plentiful supply of spare double-stranded RNA plus powerful viral promoters in crops, could lead to the emergence of widespread novel crop diseases which could obliterate our staple food supply. The inevitability of genetic pollution which will bring together different types of double-stranded RNA will make these crops a time-bomb.
Neither is it difficult to envisage how links between viruses abounding in ourselves (two per cent of the human genome is estimated to be made up of retroviral sequences) and a frequent exposure to spare, unusual, double-stranded RNA plus powerful viral promoters in our food could evolve novel viral human diseases.
Safety assessments on GM crops in the US remain essentially voluntary and are now much less exacting than the prototype which generated FDA approval of the long-life tomato. Unwisely, the days of escaping a potato famine by rushing off to another continent are over: problems over there will quickly become problems over here.
The EU has taken considerably more care than the rest of the world and has drafted carefully detailed regulations for GMOs, but our immediate worry is that the existing regulations have not been designed to deal with this type of genetic manipulation. Our system focuses on such concepts as the 'history of safe use' of the non-GM sources of all the materials used to construct the novel organisms, not on how the whole or the parts have been changed in the process. Safety assessment is focused on the novel protein generated, not the emergence of unexpected knock-on effects of the transformation. The double-stranded RNA does not generate a protein, nor does it really have a species of origin except that it is designed to complement a section of DNA in the target insect.
These novel, novel crops could be slipped onto your plate as just another insect-resistant crop. Or worse, as 'safer' non-GM crops because they don't contain any new genes.
(This article is adapted from an article which first appeared on GM-free Scotland in February 2008. View an archived copy of that article here.)
- Institute of Science in Society Press Release, 05.04.05
- The Biochemist online, 26:5 October 2004
- Shlomai and Shaul, Liver International, December 2004, 24:6 pp.526-31
- Monsanto Patent on 'Double-stranded RNA stabilized in planta', 01.11.07
- Baum et al. Letter in Nature Biotechnology, online 4.11.07
- Levin et al. Plant Molecular Biology, 44:6 December 2000
- Belinda Martineau, First Fruit, 2001, ISBN 0-07-140027-3
- University of Cape Town, Department of Medical Microbiology, 3rd Year Notes for Medical Students 1999
- Latham and Wilson, 2007, Molecular Plant Pathology, 8(6)