In a report in the April issue of Nature Biotechnology, two British-based researchers look into scenarios of transgene escape from oilseed rape (Brassica napus) into its close wild relative, Brassica rapa. Their conclusion? Not very much. The possibility of transgene escape is negligible, but it nevertheless there, and as such would require careful watching. Sadly, this is unlikely to impress the public: "The bitter truth is that no matter what safeguards are put into place, the anti-GM lobby will never be appeased", complain Dean Chamberlain and Neal Stewart of the University of North Carolina in Greensboro, in a commentary accompanying the research.

In the report itself, Susan E. Scott and Mike J. Wilkinson of the University of Reading, UK, look at an ingenious form of genetic modification in which the risk of the transgene escaping is inherently low. Rather than placing foreign DNA in the nuclei of plant cells, the site of most of the host DNA, researchers could place the transgene within small bodies inside the cells called chloroplasts. These bodies, which contain the pigment cholorophyll responsible for the green colour of plants, also contain their own DNA, which is inherited maternally. Because of this, chloroplast DNA is never found in the widely dispersed male sex cells — the pollen — which means that transgenes in chloroplast DNA (in contrast to nuclear DNA) cannot spread through the environment in the wind or through the agency of insects. The bad news is that it is much harder to make transgenic plants through modification of chloroplast rather than nuclear DNA. So far, the only success has been with tobacco plants.

Scott and Wilkinson, undeterred by this, think that it is wise not to underestimate the march of technology, and it will do no harm to look ahead to the day when it is not only possible but routine to create 'transplastomic' oilseed rape — rape with genes introduced into chloroplast DNA.

In the absence of any trly transplastomic plants, though, the researchers assess the risks by plotting the spread of natural chloroplast genes between fields of cultivated rape and stands of feral rape and Brassica rapa growing nearby. Such situations appear to occur only extremely rarely — of 147 populations of B. rapa growing along 34 kilometres of the River Thames [in England] in 1997 and 1998, only one population in each of these years grew next to a field of rape. In addition, feral populations of rape itself are rare and do not often persist for more than a year or two. The chances of a transgene escaping from a transplastomic rape plant into the wild in such situations, therefore, are virtually nil.

However, such situations might arise in which, by chance, a transplastomic rape plant might find itself in a stand of B. rapa at just the right time when both are fertile. Overwhelmed by B. rapa pollen, the rape plant could give rise to a hybrid B. rapa plant which contained the transgene. And, given that the transgene might confer a selective advantage — such as resistance to infestation by some pest — the transgene might spread through the population. Future research, say Scott and Wilkinson, should now turn to questions of the specific effects of transgenes themselves.