Effects of Lipids and Hydrophobic Surfaces on Aβ Aggregated Structures

Alzheimer's disease (AD) is part of a unique class of diseases because it is unlike the conventional diseases which are caused by viruses or bacteria. It is a "molecular disease" caused by the aggregation of a small protein (peptide) called β-amyloid (Aβ). Significant genetic, pathological, and biochemical evidence link AD to the aggregation (self-assembly) of the Aβ, especially the 40- or 42-residue isoforms (called Aβ(1-40) and Aβ(1-42) respectively) that are considered to be the most disease relevant. It is known that complex aggregation pathways produce a variety of aggregated Aβ structures, eventually producing amyloid fibrils (filamentous protein nanostructures) that deposit into plaques in the brain. Understanding the molecular structures of Aβ oligomers (aggregates that composed of 2 to roughly 100 molecules) and underlying assembly pathways will advance the fundamental understanding of AD at the molecular level. Furthermore, Aβ aggregation in vivo most likely occurs at peptide concentrations (nanomolar) which are much lower than concentrations required for Aβ aggregation in vitro (micromolar). One possible reason that has been proposed is that aggregation in vivo is catalyzed by interactions between Aβ and surfaces of lipid rafts or cell membranes. In this work, we have utilized solid state nuclear magnetic resonance (NMR) and molecular modeling to study the molecular structure of 150 kDa Aβ(1-42) oligomers formed by the interaction between sodium dodecyl sulfate (SDS) micelles and Aβ (1-42) peptide and compared to Aβ(1-42) fibril structures. Finite pulse radio frequency driven recoupling (fpRFDR) solid state NMR experiments on Aβ(1-42) oligomers reveal chemical shifts of labeled residues that are indicative of β-strand secondary structure. Results from 2D dipolar assisted rotational resonance (DARR) experiments indicate intermolecular proximity between I31 aliphatic and F19 aromatic carbons, which is contrary to models of Aβ oligomers proposed previously by other groups. In contrast, Aβ(1-42) fibrils have shown similar secondary and quaternary structures as oligomers. The most prominent structural differences between Aβ(1-42) oligomers and fibrils were observed through measurements of inter-molecular 13C-13C dipolar couplings observed in PITHIRDS-CT experiments. PITHIRDS-CT data indicate that, unlike fibrils, oligomers are not characterized by in-register parallel β-sheets. A hypothesized model is built based on the structural information we obtained from NMR characterization. We have further reported new solid state NMR constraints which indicate antiparallel intermolecular alignment β-strands within the 150kDa Aβ(1-42) oligomers and this result is consistent with our hypothesized model . Specifically, 150 kDa Aβ(1-42) oligomers with uniform 13C isotopic labels at I32, M35, G37 and V40 exhibit β-strand secondary chemical shifts in 2D fpRFDR NMR spectra, spatial proximities between I32 and V40 as well as M35 and G37 in 2D DARR spectra, and close proximity between M35 Hα and G37 Hα in 2D CHHC spectra. 2D DARR result of oligomer sample prepared with 30% labeled peptide further indicate the I32-V40, M35-G37 contacts are intermolecular. In addition, we employ molecular modeling to compare the newly derived experimental constraints with previously proposed geometries and come up with three candidate molecular models for arrangement of C-terminal region of Aβ(1-42) molecules into oligomers. Furthermore, we have developed a new technique to study how Aβ aggregation could be influenced by environmental factors such as interaction with different surfaces based on a surface patterning technique called microcontact printing (µCP). This approach allows easy measurement of thickness of the adsorbed layer by transferring the adsorbed layer form the polydimethylsiloxane (PDMS) stamp surface to a glass substrate. It also enables characterization of the face of the adsorbed peptide layer in contact with the hydrophobic stamp surface. Characterization of this face is normally not compatible with the conventional surface characterization techniques. Our results have shown that this face exhibits significant higher thioflavin T fluorescence than the face exposed to water, suggesting β-sheet formation induced by the hydrophobic PDMS surface. In addition, we have evaluated the intrinsic stability of the adsorbed layer by printing the layer on a sacrificial layer and have shown that by chemically crosslinking the adsorbed peptide, stable particulate microstructures in water can be obtained. Moreover, co-micropatterning of the different aggregated states of Aβ(1-40) (monomers and fibril fragments) is demonstrated, which may have potential applications for future study of Aβ aggregation or the interaction between different Aβ species.