Finding Order in Disorder: Probing Transient Functional States in the Amyloidogenic Alzheimer's Aβ Peptide Using the NMR Chemical Shift Covariance Analysis (CHESCA)

Abstract

Prior to self-association into toxic soluble oligomers, amyloidogenic peptides are often unstructured or only partially structured and they are best described by an heterogeneous conformational ensemble, in which multiple discrete structures are populated. Despite the complexity in the ensemble of conformations accessible to amyloidogenic peptides, it has been suggested that a key determinant of the propensity to aggregate is the transient access to a subset of functional conformations, i.e. conformers that are either aggregation competent or are incompetent but compete with aggregation competent states. However, due to the conformational ‘noise’ generated by the highly degenerate free-energy landscape typical of amyloidogenic peptides, the identification of minor populations of functional conformers relevant for aggregation remains experimentally challenging. Here, we propose a method to identify which residues within amyloidogenic peptides are involved in functional states based on the covariance analysis of NMR chemical shifts (CHESCA) [1]. CHESCA relies on linear correlations between residue-specific chemical shifts measured for the peptide system of interest subject to a set of perturbations that modulate a given functional property, e.g. the aggregation propensity in the case of amyloidogenic peptides. Such linear chemical shift correlations are useful to identify the residues involved in transient and minimally populated partially folded conformers. In addition, a distinct advantage of the chemical shift covariance method is that the relative order in which the perturbed states appear in the inter-residue correlations defines the functional relevance of the partially folded conformers. This general approach will be illustrated through its application to the Aβ peptide, which self-associates into insoluble β-sheet structured aggregates linked to Alzheimer's disease.1) Selvaratnam R, Chowdhury S, VanSchouwen B, Melacini G. Proc Natl Acad Sci U S A. 2011;108(15):6133.