In the track of new biopesticides, four genes namely cytA, cytB, cytC and cytD encoding proteins homologous to Bacillus thuringiensis (Bt) Cyt toxins have been identified in the plant pathogenic bacteria Dickeya dadantii genome. Here we show that three Cyt-like d-endotoxins from D. dadantii (CytA, CytB and CytC) are toxic to the pathogen of the pea aphid Acyrthosiphon pisum in terms of both mortality and growth rate. The phylogenetic analysis of the comprehensive set of Cyt toxins available in genomic databases shows that the whole family is of limited taxonomic occurrence, though in quite diverse microbial taxa. From a structure-function perspective the 3D structure of CytC and its backbone dynamics in solution have been determined by NMR. CytC adopts a cytolysin fold, structurally classified as a Cyt2-like protein. Moreover, the identification of a putative lipid binding pocket in CytC structure, which has been probably maintained in most members of the Cyt-toxin family, could support the importance of this lipid binding cavity for the mechanism of action of the whole family. This integrative approach provided significant insights into the evolutionary and functional history of D. dadantii Cyt toxins, which appears to be interesting leads for biopesticides. W eeds, pathogens and animal pests are potentially responsible for huge economic losses in agricultural production, and about 20% of these losses are due to animal pests 1. Among insects, aphids (Hemiptera: Aphidoidea) are one of the most injuring taxa for agricultural plants. They are difficult to control due to their specialized feeding mechanism and unusual reproductive biology 2,3. As a result, the management of aphid populations is quite challenging. Until now, most aphid pest control strategies rely on the use of specific sets of systemic chemical pesticides. But the extensive use of these pesticides had led to resistance to insecticides in several aphid species 4,5 , and cause significant environmental damage by targeting different guilds of beneficial insects (predators, parasitoids, and pollinators) 6,7. Therefore, it is highly desirable to develop biopesticides with low non-target effects. A substitute to current chemical pesticides is plant bioengineering; in order to be more selective to target pests, plants can be genetically modified to express insecticidal biomolecules within specific tissues 8. Nevertheless, as happened with conventional pesticides, plant bioengineering has also led to some field insect resistance 9. Hence, crops with more than one defensive protein, each with a different mechanism of action, have been proposed to delay insect resistance (gene pyramiding). However, very few genetically modified plants have yet been developed with resistance to sap-sucking insects, and none is used commercially 10–12. Therefore it seems essential to further develop suitable biopesticides, which could turn into candidate genes for the development of aphid-tolerant plants. In the track of such biopesticides, bacterial toxins did provide paradigmatic solutions, such as the crystal toxins encoded in plasmids of the soil bacterium Bacillus thuringiensis (Bt), which provided a vast diversity of Cry-like toxins for plant protection. Recently however, Bt-related toxins were found in genomes outside its original Gram1 bacterium: four genes namely cytA, cytB, cytC and cytD, encoding proteins homologous to Bacillus thuringiensis (Bt) Cyt toxins have been identified in the Dickeya dadantii (formerly Erwinia chrysanthemi) genome 13