Insect tolerance to the crystal toxins Cry1Ac and Cry2Ab is mediated by the binding of monomeric toxin to lipophorin glycolipids causing oligomerization and sequestration reactions.

Endotoxins from the soil bacterium Bacillus thuringiensis are used worldwide to control insect pests and vectors of diseases. Despite extensive use of the toxins as sprays and in transgenic crops, their mode of action is still not completely known. Here we show that two crystal toxins binding to different glycoprotein receptors have similar glycolipid binding properties. The glycolipid binding domain was identified in a recombinant peptide representing the domain II of the crystal toxin Cry1Ac (M-peptide). The recombinant M-peptide was isolated from bacterial lysates as a mixture of monomers and dimers and formed tetramers upon binding to glycolipid microvesicles from gut tissues and lipid particles from hemolymph plasma. Likewise, when mature toxins and M-peptides where mixed with plasma, these peptides bind to lipid particles and can be separated with lipophorin particles on low-density gradients. When mature toxin and M-peptides are added to lipid particles in increasing amounts, the peptide-particle complexes form higher aggregates that are similar to aggregates formed in low-density gradients in the presence of the toxin. This could indicate that glycolipids on lipid particles are possible targets for toxin monomers in the gut lumen, which upon binding to the glycolipids form tetramers and aggregate particles and thereby sequester the toxin inside the gut lumen before it can interact with receptors on the brush border membrane. The implication is that domain II interacting with glycolipids mediate tolerance to the toxin that is separate from interaction of the toxin with glycoprotein receptors causing toxicity.

Arsenic interactions with lipid particles containing iron.

While arsenic is toxic to all multicellular organisms, some organisms become tolerant by an unknown mechanism. We have recently uncovered an inducible tolerance mechanism in insects, which is based on a sequestration of toxins and pathogens by lipid particles. To examine whether arsenic interacts with lipid particles from mammals we compared binding of arsenic to lipid particles from insect and pig plasma after separation of lipid particles by low-density gradient centrifugation. Arsenic was found in both organisms in an area of the gradient, which corresponds to lipid-rich lipid particles. Since iron is known to affect arsenic toxicity in some organisms, we asked whether iron may be present in lipid particles. When low density cell (LDC) gradient fractions were analysed for the presence of iron we detected a peak in very low-density fractions similar to those that carried arsenic. This could indicate that arsenic interacts with lipid particles that contain iron and, if arsenic is removed from the plasma by lipid particles, that would also reduce iron-containing lipid particles at the time of arsenic emergence in the plasma. To test this assumption we measured the iron content in plasma at various time periods after the toxin ingestion. This time course revealed that iron is depleted in plasma fractions when arsenic shows a peak. Our data suggest that arsenic interacts with invertebrate and vertebrate lipid particles that are associated with proteins that may lead to detoxification by cell-free or cellular sequestration mechanisms.