Selection of relevant non-target herbivores for monitoring the environmental effects of Bt maize pollen

¹ Botanical Garden, University of Konstanz, Universitätsstr. 10, 78457 Konstanz, Germany
² Robert Koch Institute, center for Gene Technology, Postfach 650280, 13302 Berlin, Germany
³ Federal Agency for Nature Conservation (BfN), Konstantinstr. 110, 53179 Bonn, Germany

Corresponding author: Gregor.Schmitz@uni-konstanz.de

Abstract

Genes of Bacillus thuringiensis var. kurstaki (Berliner) that encode lepidopteran-specific toxins were engineered into maize for protection against the European Corn Borer, Ostrinia nubilalis (Hbn.). Recent data suggest that Lepidoptera may be negatively affected, if maize pollen contains high amounts of Bt toxin and is diposited on host plants near maize fields. Monitoring the environmental effects of commercial Bt maize fields requires effective use of limited financial and logistical resources. The aim of this study was to develop and apply tools for selecting relevant herbivore species for the field monitoring of environmental Bt toxin effects via pollen deposition. We first present a theoretical selection tree based on “risk index of Bt pollen for herbivores” (I_Btp). Our index consists of five classes from zero (not relevant) to four (highly relevant) derived from data on potential temporal and spatial coincidence of pollen exposure (A), feeding mode (B), susceptibility to lepidopteran-specific Bt toxins (C) and hazard to rare and/or endangered species (“Red List”) (D). We then screened the Macrolepidoptera database LEPIDAT to identify relevant species in Germany. Finally, we also applied the index to species found in a local biocoenotic field study (Bonn, Western Rhineland, Germany). Approximately 7% of the German Macrolepidoptera species mainly occur in farmland areas and were selected as being potentially affected by Bt pollen exposure. Of these species, 14% (= 1% of total) were found to be potentially exposed on a regional scale. The combination of I_Btp and database screening enables us to pre-select species for monitoring purposes.