Validation of ADAM10 metalloprotease as a Bacillus thuringiensis Cry3Aa toxin functional receptor in Colorado potato beetle (Leptinotarsa decemlineata).

Bacillus thuringiensis parasporal crystal proteins (Cry proteins) are insecticidal pore-forming toxins that bind to specific receptor molecules on the brush border membrane of susceptible insect midgut cells to exert their toxic action. In the Colorado potato beetle (CPB), a coleopteran pest, we previously proposed that interaction of Cry3Aa toxin with a CPB ADAM10 metalloprotease is an essential part of the mode of action of this toxin. Here, we annotated the gene sequence encoding an ADAM10 metalloprotease protein (CPB-ADAM10) in the CPB genome sequencing project, and using RNA interference gene silencing we demonstrated that CPB-ADAM10 is a Cry3Aa toxin functional receptor in CPB. Cry3Aa toxicity was significantly lower in CPB-ADAM10 silenced larvae and in vitro toxin pore-forming ability was greatly diminished in lipid planar bilayers fused with CPB brush border membrane vesicles (BBMVs) prepared from CPB-ADAM10 silenced larvae. In accordance with our previous data that indicated this toxin was a substrate of ADAM10 in CPB, Cry3Aa toxin membrane-associated proteolysis was altered when CPB BBMVs lacked ADAM10. The functional validation of CPB-ADAM10 as a Cry3Aa toxin receptor in CPB expands the already recognized role of ADAM10 as a pathogenicity determinant of pore-forming toxins in humans to an invertebrate species.

A membrane associated metalloprotease cleaves Cry3Aa Bacillus thuringiensis toxin reducing pore formation in Colorado potato beetle brush border membrane vesicles.

Insect proteases are implicated in Bacillus thuringiensis insecticidal proteins mode of action determining toxin specificity and sensitivity. Few data are available on the involvement of proteases in the later steps of toxicity such as protease interaction with toxin-receptor complexes and the pore formation process. In this study, a Colorado potato beetle (CPB) midgut membrane metalloprotease was found to be involved in the proteolytic processing of Cry3Aa. Interaction of Cry3Aa with BBMV membrane proteases resulted in a distinct pattern of proteolysis. Cleavage was demonstrated to occur in protease accessible regions of domain III and was specifically inhibited by the metalloprotease inhibitors 1,10-phenanthroline and acetohydroxamic acid. Proteolytic inhibition by a peptide representing a segment of proteolysis in domain III and the metalloprotease inhibitor acetohydroxamic acid correlated with increased pore formation, evidencing that Cry3Aa is a specific target of a CPB membrane metalloprotease that degrades potentially active toxin.

Role of toxin activation on binding and pore formation activity of the Bacillus thuringiensis Cry3 toxins in membranes of Leptinotarsa decemlineata (Say).

Binding and pore formation constitute key steps in the mode of action of Bacillus thuringiensis delta-endotoxins. In this work, we present a comparative analysis of toxin-binding capacities of proteolytically processed Cry3A, Cry3B and Cry3C toxins to brush border membranes (BBMV) of the Colorado potato beetle Leptinotarsa decemlineata (CPB), a major potato coleopteran-insect pest. Competition experiments showed that the three Cry3 proteolytically activated toxins share a common binding site. Also heterologous competition experiments showed that Cry3Aa and Cry3Ca toxins have an extra binding site that is not shared with Cry3Ba toxin. The pore formation activity of the three different Cry3 toxins is analysed. High pore-formation activities were observed in Cry3 toxins obtained by proteolytical activation with CPB BBMV in contrast to toxins activated with either trypsin or chymotrypsin proteases. The pore-formation activity correlated with the formation of soluble oligomeric structures. Our data support that, similarly to the Cry1A toxins, the Cry3 oligomer is formed after receptor binding and before membrane insertion, forming a pre-pore structure that is insertion-competent.

Mode of action of Bacillus thuringiensis PS86Q3 strain in hymenopteran forest pests.

The mode of action of Cry toxins has been described principally in lepidopteran insects as a multistep process. In this work we describe the mode of action of a Cry toxin active in the common pine sawfly Diprion pini (Hymenoptera, Diprionidae), considered a major forest pest in Europe. Strain PS86Q3 contains a long bipyramidal crystal composed of five major proteins. The N-terminal sequence shows that the 155 kDa protein corresponds to Cry5B toxin and the other proteins belong to the Cry5A subgroup. PCR analysis indicates the presence of cry5Ac and cry5Ba genes, suggesting that Cry5A protein should be Cry5Ac. Activation of protoxins with trypsin or with midgut content from D. pini and Cephacia abietis (Hymenoptera, Pamphiliidae) (spruce webspinning sawfly), another important hymenopteran forest pest, produced a single 75 kDa toxin that corresponded to Cry5A by N-terminal sequence and is responsible for the insecticidal activity. Homologous competition experiments with D. pini and C. abietis brush border membrane vesicles (BBMV) showed that the binding interaction of Cry5A is specific. Membrane potential measurements using a fluorescent dye indicate that Cry5A toxin at nM concentration caused immediate permeability changes in the BBMV isolated from both hymenopteran larvae. The initial response and the sustained permeability change are cationic as previously shown for Cry1 toxins. These results indicate that the hymenopteran specific Cry5A toxin exerts toxicity by a similar mechanism as Cry1 toxins.

Separation by FPLC chromatofocusing of UDP-glucosyltransferases from three developmental stages of Drosophila melanogaster.

Variation of UDP-glucosyltransferase activity, during Drosophila melanogaster development, was analyzed. The endogenous metabolite xanthurenic acid and the xenobiotic compounds 1-naphthol and 2-naphthol were used as substrates. Developmentally regulated differences were observed for the three substrates, suggesting the presence of UDP-glucosyltransferase isoenzymes. This was further confirmed by FPLC chromatofocusing on a Mono P column: seven peaks of UDP-glucosyltransferase activity (pHs: > or = 6.3, 5.8, 5.5, 4.9, 4.5, 4.2, < or = 4.0) with either single or overlapping substrate specificity were detected. A single xanthurenic acid:UDP-glucosyltransferase activity (pl 5.8) was found throughout development. In contrast, a gradual increase in the number of 2-napthol:UDP-glucosyltransferase-isoenzymes (pl from 6.3 to 4.0) was observed during development, whereas no isoenzymes specific for 1-naphthol were resolved. Based on the distribution and substrate specificity of the eluted peaks in the three developmental stages analyzed, the presence of seven or possibly eight UDP-glucosyltransferase isoenzymes is proposed.

A binding site for Bacillus thuringiensis Cry1Ab toxin is lost during larval development in two forest pests.

The insecticidal activity and receptor binding properties of Bacillus thuringiensis Cry1A toxins towards the forest pests Thaumetopoea pityocampa (processionary moth) and Lymantria monacha (nun moth) were investigated. Cry1Aa, Cry1Ab, and Cry1Ac were highly toxic (corresponding 50% lethal concentration values: 956, 895, and 379 pg/microl, respectively) to first-instar T. pityocampa larvae. During larval development, Cry1Ab and Cry1Ac toxicity decreased with increasing age, although the loss of activity was more pronounced for Cry1Ab. Binding assays with (125)I-labelled Cry1Ab and brush border membrane vesicles from T. pityocampa first- and last-instar larvae detected a remarkable decrease in the overall Cry1Ab binding affinity in last-instar larvae, although saturable Cry1Ab binding to both instars was observed. Homologous competition experiments demonstrated the loss of one of the two Cry1Ab high-affinity binding sites detected in first-instar larvae. Growth inhibition assays with sublethal doses of Cry1Aa, Cry1Ab, and Cry1Ac in L. monacha showed that all three toxins were able to delay molting from second instar to third instar. Specific saturable binding of Cry1Ab was detected only in first- and second-instar larvae. Cry1Ab binding was not detected in last-instar larvae, although specific binding of Cry1Aa and Cry1Ac was observed. These results demonstrate a loss of Cry1Ab binding sites during development on the midgut epithelium of T. pityocampa and L. monacha, correlating in T. pityocampa with a decrease in Cry1Ab toxicity with increasing age.

Effect of Bacillus thuringiensis toxins on the midgut of the nun moth Lymantria monacha.

Three steps of the proposed mode of action of Bacillus thuringiensis toxins have been studied in Lymantria monacha. We demonstrated that only the toxins that caused typical pathological changes in midgut epithelial cells and bound to the midgut brush border membrane were able to drastically reduce the midgut transepithelial voltage of the nun moth.