A number of crop species are adept at escaping cultivation and becoming feral forms that persist in natural communities. Genetically-engineered (GE) crops carry the additional risk of introducing transgenes into unmanaged populations. Nonetheless, we know little about the processes by which crops escape cultivation, become weeds, or transfer engineered genes into wild populations. Commercial canola (Brassica napus L.) has escaped cultivation and established feral populations on five continents. We have been tracking feral populations of GE herbicide tolerant canola (Brassica napus L.) in North Dakota, USA, for more than a decade.
Our objective was to answer the following questions: 1) Do feral populations persist outside of cultivation, 2) Have the frequencies of GE varieties changed through time; 3) Is there evidence of selection for or against transgenotypes? We surveyed more than 3000 miles of North Dakota highways in 2010 and again in 2021 for the presence of feral B. napus, stopping every five miles to census populations, test for the presence of transgenes, and collect a voucher. Our intent in 2021 was to repeat these methodologies, but conspicuous populations were encountered on stream sides and constructions sites and were added to the survey.
Results/Conclusions
Canola has escaped cultivation and has established persistent populations throughout North Dakota. Feral canola was present at nearly half of the survey sampling sites in 2010. Most plants sampled expressed at least one transgene (80%) and a small percentage (0.7%) expressed two. Our findings were similar in 2021 with a few notable exceptions. The incidence of transgenic glyphosate resistance dropped by 80%, and glufosinate resistance increased by 74%, though the overall incidence of GE canola was similar. Non-GE canola made up 20% of our sample, much higher than acres planted (3-7%). Canola was found growing beyond road verges to construction sites and stream sides, including populations in two protected areas (Sheyenne National Grassland, Upper Souris National Wildlife Refuge).
Substantial changes in GE phenotype frequencies between survey years could be due to shifts in cultivation practice, changing herbicide use in non-agronomic habitats, or fitness costs of glyphosate resistance. The high frequency of non-GE canola relative to production acres may indicate selection against GE canola outside of cultivation, or the long-term persistence of populations established before GE canola was grown commercially (1990s). Feral populations of transgenic canola merit further scrutiny.
