Membrane Permeation by Islet Amyloid Polypeptide is Modulated by Disordered Residues Distal from the Helical Membrane Binding Domain

Several diseases, such as Alzheimer's and Parkinson's diseases, represent a class of degenerative disorders collectively termed amyloidoses, characterized by the prevalence of fibrous amyloid plaques consisting of a unique aggregation-prone protein. Despite the potentially high prevalence of plaque deposits often found in patients, cytotoxicity is most correlated with preamyloid oligomeric states known as toxic oligomers that exhibit the capacity to disrupt lipid bilayers. In type II diabetes, this cytotoxic protein is Islet Amyloid Polypeptide (IAPP), a peptide hormone secreted from pancreatic 6-cells.;A significant barrier in the understanding of amyloid protein-induced cell death is the currently incomplete description of the structures of membrane-bound oligomeric states of amyloid proteins and the mechanisms by which these states permeate lipid bilayers. For IAPP, this pursuit is hindered by apparent conflicts regarding the significance of membrane-bound peptide structure. While membrane-bound a-helical states are associated with permeation, IAPP induces membrane leakage cross-cooperatively with other membrane-active peptides of completely different sequence or of the opposite enantiomer. Such observations are more consistent with disruption that is not mediated by protein-protein interactions involving specific structures.;Here, we address this apparent discrepancy between structure and IAPPinduced membrane permeation by examining correlations between a-helical secondary structure formation and bilayer disruption for amyloidogenic and nonamyloidogenic mutants of IAPP. We find that membrane permeation cannot be fully accounted by secondary structure formation -within the N-terminal region of the peptide. We find that variable populations of virtually all C-terminal residues contact the membrane surface, and that the extent to which such contacts occur apparently correlates with bilayer disruption. We also show that truncation of these C-terminal residues can drastically reduce the rate and extent of IAPPinduced permeation. Finally, we show that residual bilayer disruption observed with truncated rIAPP, but not truncated hIAPP, is consistent with a cross-cooperative leakage mechanism. We interpret these findings using a lipocentric model in which peptide-induced transmembrane pore formation is the result of kinetics and thermodynamics associated with lipids within a bilayer.