Towards an improved assessment of the risks associated with the dispersion of new genes into the environment

One of the main risks associated with the use of exotic (or non-native) species and, more recently, of genetically modified organisms (GMOs) is the potential for genetic contamination of related native species by genes introduced through gene flow. Resulting spontaneous hybrids could have long-term impacts on the genetic diversity of native species if they lead to introgression—the infiltration of genes of one species into the gene pool of another.

The possibility and effects of GMO introgression are difficult to measure directly. The Canadian Food Inspection Agency places strict confinement measures on genetically modified (GM) tree plantations and it is not possible to use such sexually mature plantations in a study. Indirect methods of risk assessment are needed. One such method is to simulate GMO introgression by measuring gene flow between exotic species in plantations or shelterbelts and populations of native species located on the periphery. At the Laurentian Forestry Centre, Canadian Forest Service, we designed an experiment to do just that.

We chose poplars and larches as our study trees. Exotic poplars and larches and their respective hybrids were introduced into Canada a number of decades ago for reforestation or ornamental purposes. In collaboration with researchers from Université Laval, we identified a number of sexually mature poplar and larch plantations adjacent to natural populations that were appropriate for measuring gene flow parameters. Sampling was conducted over three years; each harvest included a collection of catkins on poplars or cones on larches from about 30 native trees surrounding the plantations. Spontaneous hybridization rates were assessed using about 1 000 seeds per site per recipient species per year. We estimated hybridization rates using diagnostic molecular markers specific to each species; data were collected in different ecological settings and over a number of years to take into account potential spatio-temporal effects.

In poplar, the percentage of interspecific hybrids detected in the seeds varied from 5% to 66% (depending on the recipient species and the site); however, a portion of the detected hybrids were native (Populus deltoides × P. balsamifera). In larch, the hybridization rates were less than 2%. In all cases, the hybridization rates were not statistically different from year to year.

Our results clearly show the possibility of hybridization between trees from plantations of hybrids with exotic components and trees from populations of native species. This however does not directly mean that such gene flow immediately presents a risk if the plantation of hybrids would contain GMOs. The next step in the risk assessment is therefore to see whether the resulting hybrids manage to establish themselves in the natural populations and whether they reproduce. It is also essential to look how selectively advantageous genes spread over the natural populations. Meanwhile, we have produced a simulation model that helps to gain insight in the effects of transgene introgression, which has important implications for regulation and management.

This research project is part of the Canadian Regulatory System for Biotechnology program. Baseline data on ecological interactions in Canadian forest ecosystems will contribute to the body of scientific knowledge needed for regulating plants with novel traits (PNTs) in Canada.