The advance of genetically modified crops and farm animals has opened up fears of ecological disaster if the engineered, or transgenic, organisms were to escape the confines of the farm. Assessing the environmental risk posed by transgenic populations requires an understanding of how they would compete with their wild counterparts under such circumstances. To that end, new laboratory research has found that wild salmon tend to experience reduced growth in the company of salmon engineered to attain a large body size. The presence of transgenic fish also increases the likelihood of population collapse when food is in short supply.
The study, published online today by the Proceedings of the National Academy of Sciences, involved growth hormone (GH) transgenic coho salmon, which have greater appetites and can grow up to seven times bigger than wild cohos. Robert H. Devlin of Fisheries and Oceans Canada and his colleagues divided their fish into three separate groups containing either all GH animals, all nontransgenic ones, or an equal mixture of the two. To examine the competition between the two types, the researchers varied the amount of food supplied to the fish.
The team found that all the salmon thrived as long as there was enough food to go around. Faced with food shortages, however, GH individuals in the mixed group outcompeted their wild tankmates, growing larger than both those fish and the ones living in GH-only groups; and the wild salmon in the mixed group exhibited reduced growth as compared to members of the wild salmon-only group. Furthermore, survival rates were significantly reduced in those tanks holding GH salmon--sometimes to the point of extinction. Some of the dead fish appeared to have died from attacks by other fish, and there were several instances of cannibalism. The fish that survived in these tanks were usually the most aggressive GH fish. Individuals in the wild salmon-only groups fared far better, experiencing a constant increase in population biomass over the 14-week period of low rations.
Because this research was conducted in an artificial setting, without all the environmental factors that affect survival, the scientists "caution against using such data prematurely in predictions of potential impacts until a broad understanding of the effects is known in many relevant habitats." --Michael Schirber