From non-target risk assessment to Bt resistance management: the example of Bt Brassica sp. and Pieridae

Insecticidal toxins produced by Bacillus thuringiensis (Bt) have been widely used in plant protection in two different ways: 1) by spraying bacterial Bt formulations, 2) by expressing their gene encoding in planta. In the second strategy, only the transgenic plants themselves are protected but insect- resistant plants could spread into natural habitats and/or the gene could be fixed in alternative hosts. It is necessary to estimate the magnitude of the possible effects on naturally occurring herbivorous insects but also on the target species, especially when they have several alternative hosts, as in the case of Pieridae. They are present both in agriculture where they could be considered pests, and non- agricultural habitats, where the larvae feed on several Brassicaceae. Moreover, oilseed rape is known to form feral populations in natural and semi-natural habitats, and that Bt- oilseed rape may out-compete insect-susceptible plants at high herbivore densities. Furthermore, high-density patches of highly damaged wild plants are the most vulnerable to Bt-transgene invasion. To assess the potential risk of Bt resistance in Pieridae, different parameters should be considered. In this contest we thought it worthwhile to study the population density of Pieridae, their fly period compared with the flowering period of potential Bt crops and relative wild species and its hosts preferences Moreover we report data on susceptibility to commercial Bt and to CryIAb toxin. Such data may both aid the design of further tests of related effects and aid in the assessment of any effects on the population outside the agricultural area.