Bacillus thuringiensis (Bt) est une bactérie produisant des inclusions protéiques cristallines à pouvoir insecticide et elle est largement exploitée à l'échelle industrielle. Dans cette étude, des souches de Bt ont été isolées du sol libanais. Nous avons étudié en premier la présence des principaux gènes cry1A codant pour des δ-endotoxines actives sur les lépidoptères. Les souches possédant ces gènes ont été testées pour leur toxicité sur des larves d'Ephestia kuehniella (E. kuehniella). Une souche nommée Lip, étant quatre fois plus toxique sur ces larves que la référence mondiale Bt subsp. kurstaki HD1, fut sélectionnée pour une étude plus approfondie. Après clonage et séquençage, nous avons identifié une nouvelle toxine de type Cry1Aa : Cry1Aa22 et une nouvelle variante de la toxine Cry1Ac. Ces dernières se sont montrées plus toxiques sur des larves d'E. kuehniella, et plus stables en présence des protéases intestinales de ces larves que Cry1Aa et Cry1Ac de HD1 permettant d'expliquer la toxicité élevée de la souche sauvage. D'autre part, nous avons optimisé la construction d'un modèle de membrane biomimétique incluant la membrane de la bordure en brosse intestinale (BBM) des larves d'E. kuehniella. Ces membranes nous ont servi à l'étude de l'interaction des toxines Cry1Aa et Cry1Ac de Lip et celles de HD1. Les toxines de Lip ont interagit différemment et avec une plus grande affinité avec ces modèles que celles de HD1.Tous ces résultats montrent que Lip est une souche intéressante pour une exploitation industrielle et que le modèle de membrane biomimétique est une alternative permettant la prédiction de l'affinité des toxines Cry. / Bacillus thuringiensis (Bt) is a bacterium that synthesizes insecticidal proteic crystallin inclusions and is widely used at an industrial scale. In this study, Bt strains were isolated from Lebanese soil. We studied the presence of the main cry1A genes encoding for δ-endotoxins active on Lepidoptera. Strains harboring these genes were tested for their toxicity against Ephestia kuehniella (E. kuehniella) larvae. The strain named Lip, being four folds more toxic to the larvae than the reference strain Bt subsp. kurstaki HD1, was selected for further study. We identified a novel Cry1Aa toxin, Cry1Aa22, and a variety of the Cry1Ac toxin after cloning and sequencing of the corresponding genes. These toxins were more toxic to E. kuehniella larvae and more stable in the presence of these larvae's intestinal midgut juice than Cry1Aa and Cry1Ac of HD1. Moreover, we optimized the construction of a biomimetic membrane model based on the intestinal brush border membrane (BBM) of E. kuehniella larvae. These models were used to study the interaction of Cry1Aa and Cry1Ac of Lip and HD1. Toxins of Lip interacted differently and with a greater affinity with these model membranes than toxins of HD1.These results show that Lip is an interesting Bt strain that could be exploited at an industrial scale. On another hand, the biomimetic membrane constructed in this study could be an alternative allowing the prediction of the Cry toxin's affinity.

The high-dose/refuge strategy is used globally to manage insect resistance development in genetically modified crops with insecticidal properties (Bt crops). The “refuge in a bag” (RIB) strategy is also being considered for deployment against several pest species. Busseola fusca, the target pest of Bt maize in South Africa, evolved resistance to Cry1Ab proteins. The objective of this study was to determine whether migrating B. fusca larvae are effectively controlled using the RIB strategy. A field study with a single-gene event (Cry1Ab) and a “pyramid” event (Cry1A.105 + Cry2Ab2) was conducted in which the migration patterns of B. fusca larvae in plots with different seed mixture treatments were studied. The experiment consisted of five seed mixture ratios (5%, 10%, 15%, 20% non-Bt seed and 100 % non-Bt seed as control). Natural infestation was augmented by artificial inoculation with neonate larvae into the central non-Bt maize plant of each plot. Rate of larval survival and migration, measured in terms of increase in number of plants per plot that exhibited borer damage was recorded at weekly intervals until flowering. A laboratory study was conducted to determine larval growth and survival when simulating migration between Bt and non-Bt maize plants. A feeding experiment in which larvae were reared on different types of maize (Bt and non-Bt) was conducted and larval survival and mass recorded after a 7-day feeding period. The incidence of damaged ears, stem damage and damaged internodes per stem were recorded and relationships between these variables determined by means of correlation analyses. A review was conducted in order to identify and discuss similarities and differences between the high-dose/refuge and seed mixture strategies. This was done to determine which strategy would be the most appropriate insect resistance management (IRM) strategy against B. fusca. The rate of survival and migration of B. fusca larvae was significantly higher in the plots with maize expressing Cry1Ab and control plots, than in plots with the pyramid Bt event. Older larvae exhibited improved growth and survival in the laboratory experiment when they were transferred from non-Bt to Bt plants. Positive correlations were found between early and late season damage, although some weaker than others. Plants of the “pyramid event” suffered less late-season damage than those of the single-gene event. Since the increase in number of damaged maize plants over time is associated with migration of older and larger larvae, the observed tendencies may indicate that the assumed high-dose does not kill larvae above a certain developmental stage. The high-dose refuge strategy seems to be the better option for delaying resistance development. / MSc (Environmental Sciences), North-West University, Potchefstroom Campus, 2014

The high-dose/refuge strategy is used globally to manage insect resistance development in genetically modified crops with insecticidal properties (Bt crops). The “refuge in a bag” (RIB) strategy is also being considered for deployment against several pest species. Busseola fusca, the target pest of Bt maize in South Africa, evolved resistance to Cry1Ab proteins. The objective of this study was to determine whether migrating B. fusca larvae are effectively controlled using the RIB strategy. A field study with a single-gene event (Cry1Ab) and a “pyramid” event (Cry1A.105 + Cry2Ab2) was conducted in which the migration patterns of B. fusca larvae in plots with different seed mixture treatments were studied. The experiment consisted of five seed mixture ratios (5%, 10%, 15%, 20% non-Bt seed and 100 % non-Bt seed as control). Natural infestation was augmented by artificial inoculation with neonate larvae into the central non-Bt maize plant of each plot. Rate of larval survival and migration, measured in terms of increase in number of plants per plot that exhibited borer damage was recorded at weekly intervals until flowering. A laboratory study was conducted to determine larval growth and survival when simulating migration between Bt and non-Bt maize plants. A feeding experiment in which larvae were reared on different types of maize (Bt and non-Bt) was conducted and larval survival and mass recorded after a 7-day feeding period. The incidence of damaged ears, stem damage and damaged internodes per stem were recorded and relationships between these variables determined by means of correlation analyses. A review was conducted in order to identify and discuss similarities and differences between the high-dose/refuge and seed mixture strategies. This was done to determine which strategy would be the most appropriate insect resistance management (IRM) strategy against B. fusca. The rate of survival and migration of B. fusca larvae was significantly higher in the plots with maize expressing Cry1Ab and control plots, than in plots with the pyramid Bt event. Older larvae exhibited improved growth and survival in the laboratory experiment when they were transferred from non-Bt to Bt plants. Positive correlations were found between early and late season damage, although some weaker than others. Plants of the “pyramid event” suffered less late-season damage than those of the single-gene event. Since the increase in number of damaged maize plants over time is associated with migration of older and larger larvae, the observed tendencies may indicate that the assumed high-dose does not kill larvae above a certain developmental stage. The high-dose refuge strategy seems to be the better option for delaying resistance development. / MSc (Environmental Sciences), North-West University, Potchefstroom Campus, 2014