Wild relatives of genetically engineered crops can acquire transgenic trait
s such as herbicide tolerance via spontaneous crop-wild hybridisation. The
fitness costs associated with a transgene conferring tolerance of glufosina
te were tested when introgressed into weedy Brassica rapa. Crosses were mad
e between transgenic Brassica napus ssp oleifera (oilseed rape) and wild B.
rapa from Denmark. The reproductive success of 457 BC3 progeny representin
g six full-sib families raised in growth rooms was quantified (plants were
pollinated by captive bumblebees). Segregation for herbicide tolerance was
1:1 overall, as expected for a dominant Mendelian trait. There were no sign
ificant differences between transgenic and non-transgenic plants in surviva
l or the number of seeds per plant, indicating that fitness costs associate
d with the transgene are likely to be negligible. Pollen fertility and seed
production of BC3 plants were as great as those of B. rapa raised in the s
ame growth rooms. These results suggest that transgenic herbicide tolerance
is capable of introgressing into populations of B. rapa and persisting, ev
en in the absence of selection due to applications of this herbicide.