The relationships between small and large conductance channels formed by diphtheria toxin in planar lipid bilayers

Abstract

Diphtheria toxin is a dichain molecule in which a light chain, the active fragment (A), is linked by a disulfide bond to a heavy chain, the binding fragment (B). There is evidence that the toxin is internalized by receptor-mediated endocytosis and that acidification of the endocytic vesicle is required for translocation of the light chain across the vesicle membrane. The mechanism by which the active fragment crosses the vesicle membrane has remained unsolved.;The B fragment forms channels in planar bilayer membranes that have pH- and voltage-dependent properties. Channel formation is maximal when the protein-containing side of the bilayer is at acidic pH and the opposite side is at physiological pH, a pH gradient comparable to that across acidic vesicles.;Working with the intact toxin and its mutant CRM 45, noise analysis reveals the existence of more than one population of channels. Experiments at the single channel level show two kinds of channels with different conductances and a kinetics that explains the rectification and the noise observed for the macroscopic current. Reversal potential measurements in salt gradients show similar values for the small and large conductance channels; this finding and the flickering observed in the big channels suggest that the big channel is an aggregate of small channels. The number of big channels observed is a function of both toxin concentration and voltage. Experiments with low cis pH provide further evidence for the aggregation phenomenon.;Experiments with TEA{dollar}\sp+{dollar} as a large cation support the hypothesis of aggregation and suggest that this cation inhibits incorporation of small channels and the opening of big channels; the TEA{dollar}\sp+{dollar} effect is exerted from the trans side.;A working model of incorporation and gating of these channels is presented. The model involves two closed states, one open state, and an additional state of aggregation. The findings reported here suggest a model of diphtheria toxin membrane translocation in which both fragments of the toxin are involved in fragment A translocation, rather than the channel acting as a tunnel for that fragment.