The Mediterranean fruit fly, Ceratitis capitata (Wiedemann) (Diptera:Tephritidae), is considered the world’s most important fruit pest. Currently, the suitability of chemical control as the main tool to manage this pest is seriously threatened, for legal (prohibition of active materials), social (consumption trends) restrictions and also for intrinsic method limitations (resistance). Therefore, development and implementation of environmental friendly alternative methods is more and more important.

Bacillus thuringiensis (Berliner) (Bt) products have been successfully used in insect control for more than 50 years. The insecticidal activity of this bacterium primary lies on a diverse group of toxins known as delta-endotoxins. More than 3000 insect species have been found to be susceptible to at least one of these toxins. Moreover, Bt-products are highly selective and environmentally safe. Nevertheless, for the moment, no delta-endotoxin has been identified with high insecticidal activity towards C. capitata or other fruit flies (Tephritidae).

The goal of this doctoral thesis is the evaluation of B. thuringiensis and its toxins as potential C. capitata control agents. To that end, an exhaustive biodiversity analysis has been conducted on the citrus agro-ecosystem, a habitat in which the target insect (C. capitata) and the entomopathogen (B. thuringiensis) have coexisted for a long period. The application of different techniques (microscopy, protein electrophoresis, delta-endotoxin gene detection) has shown that this agro-ecosystem is an important B. thuringiensis reservoir, regarding both abundance and diversity. However, none of the strains from this and other collections (905 strains in total) showed high insecticidal activity towards C. capitata when culture crude extracts (spores and crystals or supernatants) were assayed.

The mode of action of delta-endotoxins is complex and includes several steps: in summary, after ingestion, delta-endotoxin crystals are solubilized in the insects’s midgut to release soluble protoxins that are proteolytically processed by digestive enzymes to produce active toxins that, after crossing the peritrophic membrane, interact with the epithelial membrane triggering a series of events that finally cause insect death. Limitations or alterations of any of these steps may explain that a particular delta-endotoxin is or is not toxic to a certain insect species. Accordingly, as a reaction to the lack of sensitivity of C. capitata towards delta-endotoxin crystals, in vitro treatment of these crystals has been performed emulating some steps of the mode of action mentioned above, as well as the evaluation of their insecticidal activity against C. capitata. Thus, presolubilization of the protoxins from 41 selected B. thuringiensis strains and one Bacillus sphaericus (Meyer and Neide) strain has shown that, only for the strains belonging to the subspecies israelensis (Bti), this treatment causes a gain of biological function over C. capitata larvae. Namely, at least for this subspecies, lack of activity of delta-endotoxins crystals is dependent on an insufficient solubilization in the insect’s gut, probably due to its slight alkalinity. Additionally, predigestion of Bti protoxins with an exogenous protease source (Culex pipiens L. extracts, a species that is naturally susceptible to Bti) causes a significant increase of its insecticidal activity (LC50 31.26 µg/cm2). This fact suggests that the processing of Bti protoxins in C. capitata midgut is not appropriate to promote their cytotoxic action.

Bti produces a mixture of 4 major protoxins: Cry4A, Cry4B, Cry11A and Cyt1A. The use of a recombinant strain only producing Cyt1Aa has shown that this protoxin is the determinant factor of Bti’s activity towards C. capitata. This study identifies, for the first time, a single delta-endotoxin that is toxic against this insect species. Cyt1Aa showed the highest activity reported in this study (LC50 4.93 µg/cm2), although only when their crystals are previously solubilized and exclusively against larvae (it causes sublethal effects on adults). To a certain extent, this issue restricts the potential use of this protein to control this pest.

Cyt toxins directly interact with certain membrane phospholipids; Cry toxins however, need specific receptors in the gut epithelium to be toxic. The results with native B. thuringiensis strains suggest that C. capitata lacks the appropriate receptors, at least for the array of Cry toxins that has been assayed. To overcome this limitation, this thesis proposes a new method based on the development of fusion proteins for which the conjugation to antibodies specifically raised against C. capitata membrane proteins emulates the affinity that Cry toxins show to their natural receptors. This method would require an intensive work and an important economic expense before confirming its viability; as a result, a model system has been put into practice to evaluate the strategy. The key elements of this model system are: Drosophila melanogaster (Meigen) transgenic larvae that express GFP in the gut epithelium and fusion proteins that conjugate different sections of Cry1Ab protoxin and the variable region of a heavy chain antibody specific to GFP (VHH anti-GFP). Four protein variants have been developed and GFP expression pattern has been determined in the D. melanogaster genetic line. However, deficiencies in binding of the fusion proteins to their antigen (GFP) have prevented, for the moment, the confirmation of this strategy. Additionally, regarding the potential application of this technique to C. capitata, an initial prospection has been conducted to detect membrane proteins that could be candidates for receptors of future fusion proteins. At least 160 different polypeptides have been found in the brush border membrane from the gut epithelium of C. capitata adults; and, after studying 11 proteins, at least 3 have been identified: V-ATPase subunits A and B, and alpha-tubuline. Future studies should address the suitability of this or other proteins as candidates for the new strategy.

In summary, this work has identified a single B. thuringiensis delta-endotoxin, Cyt1Aa, which is active towards C. capitata, although it is necessary to assess its potential to control this pest. Additionally, a B. thuringiensis strain collection has been developed and characterized and, although it does not show sufficient activity to the target insect, it has a great potential to control other pests. Finally, even if confirmation is still necessary, a new method has been set for developing of recombinant toxins to control C. capitata or  any other important pest.