Search

Selection of test species for use in biosafety evaluation of genetically modified plants is challenging but important, as regulators in many jurisdictions require tests to determine the potential for adverse environmental impacts before the release of plants into the environment. This contribution provides an example of an evidence-based process whereby species from the receiving environment can be ranked in order of susceptibility to potential impact, and guide test species selection. The case study used for this example was ryegrass, a forage plant, which had been modified to produce elevated levels of the lipid triacylglyceride. The previously described priority ranking of nontarget invertebrates model (PRONTI), designed to rank invertebrates for biosafety testing, has been adapted for use with these plants, which could, potentially, be beneficial to invertebrate populations, and applied to data on 246 known pasture invertebrate species. The output from the model for the top 20 ranked pasture invertebrate species is discussed, the attributes of these are considered along with the level of uncertainty in the information used. Consideration is given to how the model output can be interpreted and used in a biosafety risk assessment. While some subjectivity is involved in establishing the scores, all invertebrate species are subjected to the same analysis, and treated equally. In this way, regulators have a method of a risk assessment that is evidence-based, and transparent in its assumptions thereby avoiding potential for bias. [ABSTRACT FROM AUTHOR]

Laboratory host specificity of 2 biological control agents, already introduced in New Zealand, was compared with actual field parasitism.The parasitoids were Microctonus aethiopoides Loan and Microctonus hyperodae Loan, braconids imported to control the curculionid forage pests Sitona discoideus Gyllenhal and Listronotus bonariensis (Kuschel), respectively. The nontarget weevil species tested included native,introduced, and beneficial species. M. aethiopoides oviposited in 11of the 12 species to which it was exposed and successfully parasitized 9 species. M. hyperodae oviposited in 5 of the 11 species to whichit was exposed and developed successfully in 4 species. Higher percentage parasitism was recorded with M. aethiopoides than with M. hyperodae. Field collections of weevils from Otago, Canterbury, and Waikato indicated that 10 New Zealand natives and 3 other nontarget species, including the weed biological control agent Rhinocyllus conicus (Froehlich), were parasitized by M. aethiopoides. M. hyperodae has been found parasitizing 1 native species, as well as Sitona lepidus Gyllenhal, which was accidentally introduced to New Zealand recently. In nontarget species, parasitism levels in the field of >70% have been recorded for M. aethiopoides and <5% for M. hyperodae. The results of this study suggest that laboratory host range testing is indicative of nontarget parasitism in the field. [ABSTRACT FROM AUTHOR]