Potatoes, pathogens and pests: effects of genetic modifi cation for plant resistance on non-target arthropods

Currently, fungicides are necessary to protect potato crops against late blight, Phytophthora infestans, one of the world’s most damaging crop pathogens. The introgression of plant resistance genes from wild potato species targeted specifically to the late blight pathogen into susceptible potato varieties may alleviate the environmental impact of chemical control. Genetically modified plants are subject to an environmental risk assessment, and this includes testing for risks to the non-target arthropod community associated with the crop. The thesis begins with a review about the main plant defense responses and their role in influencing sequential interactions between herbivores and plant pathogens. The experimental chapters each focus on different aspects of the interaction between potato plants (both resistant and susceptible), the target pathogen (P. infestans) and several non-target insects. With each chapter, the scope widens: from the molecular gene expression in potato leaves in response to sequential attacks, to field scale biodiversity analyses. At the molecular level, one of the main findings was that the genomic position of the Rpi-vnt1 insertion conferring resistance to P. infestans influenced potato gene expression measured in leaves, when interacting with the non-target insect pests Myzus persicae (Green peach aphid) and Leptinotarsa decemlineata (Colorado potato beetle). Insect performance differed between the resistant GM and susceptible non-GM comparator. In the following chapter, the differences in insect performance were tested across a range of conventionally bred cultivars varying in resistance to P. infestans. Differences in M. persicae performance between several cultivars greatly outweighed the differences previously detected between the GM and non-GM comparator. These results are crucial in shaping the way risk is assessed in the context of GM crops, and these results are supported in our experiments assessing effects on biodiversity with pitfall traps in the field. The third trophic level was also addressed by comparing the performance of the parasitoid Aphidius colemani reared on GM and non-GM fed aphids, both with an without exposure to P. infestans. Differences in parasitoid performance were only found on the susceptible cultivar when inoculated with P. infestans. In the last experimental chapter the risk assessment is taken to the field comparing pitfall trap catches over two years and in two countries. Different methods for statistical analysis of biodiversity data were compared to arrive at recommendations for such analysis in the framework of environmental risk assessments. Drawing on these lessons, the discussion ends with ideas for the development of protocols for environmental risk assessments in the light of expected scientific progress in agricultural biotechnology.