A pesky problem hovering around the current success of agricultural biotechnology is the potential risk that genes from transgenic crops may get transferred to weeds. Introduction of fitness related genes such as those for disease or pest resistance into weedy relatives may empower them to be more noxious and invasive. This fear is especially troublesome with herbicide resistance genes as the addition of such a trait can make the weeds more intractable to control by herbicides.

Many crops including sorghum, barley, sunflower, cucurbits and strawberry have close weedy relatives on farms where they are grown. Oilseed rape (Brassica napus), a target of intense biotechnology research, has a large number of crucifer relatives in its areas of cultivation. A report in the most recent issue of the journal Nature provides evidence that under field conditions, intergeneric gene flow can occur between oilseed rape and wild radish (Raphanus raphanistrutum), a common weed in oilseed rape fields (1).

Anne-Marie Chevre and colleagues from INRA, Le Rheu Cedex, France found that a single-copy 'bar' gene which confers resistance to the herbicide glufosinate ammonium (Basta) in transgenic oilseed rape was transferred to the distant weed. While gene flow between closely related species has been well documented earlier, the Nature paper provides evidence that flow of transgenes can occur even between different genera under field conditions and that hybrids can become 'weed-like'.

The French group grew male-sterile oilseed rape plants in a field surrounded by wild radish. Pollen from the weed plants pollinated the transgenic rape plants to produce hybrid seeds. A normal Mendelian transmission of the transgene from the parents to the hybrids was observed. Although the F1 hybrids had poor fertility, gradually over generations hybrids became more fertile and increasingly resembled the weedy parent in chromosome number and growth habit. The researchers report that "It is likely that under normal agricultural conditions this event is rare when the wild radish is the female parent".

Neal Stewart (University of North Carolina-Greensboro), who is studying the evolutionary fitness of oilseed rape plants engineered with insecticidal Bt genes, says "the most significant aspect of the Nature paper is that female fertility increased with successive generations. This paper should be an eye- opener to those who believe that the mustard family is full of clean species and transgenes will not eventually escape out of oilseed rape".

Chevre commented that while her studies were conducted under optimal conditions, gene flow under normal agronomic conditions is now being tested. In order to assess gene flow without pollen competition, male-sterile transgenic lines were used as female parents. It was not possible to study the reciprocal crosses in the same optimal conditions because of the variability of self-incompatibility in the wild radish.

An interesting feature of this study is that weeds served as a pollen source and transgene transmission was thus maternal. This raises questions as to whether chloroplast transformation of crop plants is as benign as claimed by its proponents. Chloroplast engineering has been proposed as a safer approach to the containment of transgenes as chloroplasts are not transmitted in the pollen of flowering plants.

The Chevre article is significant because it provides evidence that one potential consequence - gene flow from genetically engineered crops to wild relatives - can occur, and the authors present a model of gene flow. Apart from it strengths and weaknesses, at least the model will stimulate development of scientific tools by which genetically engineered organisms may be evaluated.

The authors predicated their work on the principle that gene flow from crops to related wild species must be considered when assessing the potential environmental impacts of cultivating genetically modified plants. At the same time, it's important to note that identifying events that can occur is only a part of risk assessment. The more important, and difficult, part is assessing the impact such events may have on natural and managed ecosystems. It's encouraging that we're beginning to see some scientifically credible studies that will support environmentally responsible decisions about the appropriate use of particular transgenic crops grown in particular areas.

Reference
1. Chevre, A-M, et al. (1997). Gene flow from transgenic crops. Nature 389:924.