Pollen-mediated gene flow in flax (Linum usitatissimum L.): can genetically engineered and organic flax coexist?

Coexistence allows growers and consumers the choice of producing or purchasing conventional or organic crops with known standards for adventitious presence of genetically engineered (GE) seed. Flax (Linum usitatissimum L.) is multipurpose oilseed crop in which product diversity and utility could be enhanced for industrial, nutraceutical and pharmaceutical markets through genetic engineering. If GE flax were released commercially, pollen-mediated gene flow will determine in part whether GE flax could coexist without compromising other markets. As a part of pre-commercialization risk assessment, we quantified pollen-mediated gene flow between two cultivars of flax. Field experiments were conducted at four locations during 2006 and 2007 in western Canada using a concentric donor (20 x 20 m) receptor (120 x 120m) design. Gene flow was detected through the xenia effect of dominant alleles of high alpha-linolenic acid (ALA; 18:3(cis Delta 9,12,15)) to the low ALA trait. Seeds were harvested from the pollen recipient plots up to a distance of 50m in eight directions from the pollen donor. High ALA seeds were identified using a thiobarbituric acid test and served as a marker for gene flow. Binomial distribution and power analysis were used to predict the minimum number of seeds statistically required to detect the frequency of gene flow at specific a (confidence interval) and power (1-beta) values. As a result of the low frequency of gene flow, approximately 4 million seeds were screened to derive accurate quantification. Frequency of gene flow was highest near the source: averaging 0.0185 at 0.1m but declined rapidly with distance, 0.0013 and 0.00003 at 3 and 35 m, respectively. Gene flow was reduced to 50% (O(50)) and 90% (O(90)) between 0.85 to 2.64 m, and 5.68 to 17.56 m, respectively. No gene flow was detected at any site or year > 35m distance from the pollen source, suggesting that frequency of gene flow was <= 0.00003 (P = 0.95). Although it is not possible to eliminate all adventitious presence caused by pollen-mediated gene flow, through harvest blending and the use of buffer zones between GE and conventional flax fields, it could be minimized. Managing other sources of adventitious presence including seed mixing and volunteer populations may be more problematic. Heredity (2011) 106, 557-566; doi:10.1038/hdy.2010.81; published online 16 June 2010