UC Santa Cruz

Alzheimer’s Disease (AD) is the 6th leading cause of death in the United States and affects over 6 million Americans. Surprisingly, dementia-related deaths have increased by over 16% during the SARS-CoV-2 pandemic, making finding a cure more now important than ever. AD is characterized by two major pathological hallmarks: amyloid plaques, rich in the intrinsically disordered, aggregation-prone Amyloid-β (Aβ) peptide, and neurofibrillary tau tangles. The transmembrane protein that produces Aβ, the Amyloid-β Precursor Protein (AβPP), is cleaved by β- (BACE1) and γ-secretases. While much of the Aβ-focused therapeutic and academic efforts have targeted late-stage, insoluble Aβ fibrils, interest has shifted to the more toxic intermediate oligomers. These transient, rapidly interconverting oligomers are exceptionally challenging to study and therapeutically target, a fact made abundantly clear by the succession of devastating drug trial failures. AβPP can be alternatively processed by other lesser-known enzymes, such as α-secretase, to produce alternative peptidic fragments. One such fragment, the p3 peptide, is a C-terminal fragment of Aβ, and spans residues 17-40/42, the segment most attributed to Aβ’s amyloidogenicity. Despite this, p3 has traditionally been described as non-amyloidogenic and neuroprotective. Consequently, the biological and biophysical properties of p3 have been sparsely studied. The studies described in this thesis aim to provide an extensive in vitro characterization of p3, to better understand its role, if any, in AD. In Chapter 1, we aim to summarize and deconvolute the small pool of conflicting findings in the literature surrounding the p3 peptide. According to PubMed, since 1984, there have been 56,502 papers that mention Aβ, and only 921 that mention p3, an over 60-fold difference (Figure 2). Of the small pool of papers published discussing p3 since the mid-1980s, only a handful investigate the properties of the peptide, while most simply state that it is non-amyloidogenic, or that its production precludes the production of Aβ. Despite this, a few rarely discussed papers, primarily published in the 1990s and early 2000s, expose amyloidogenic properties of p3. Our work, discussed in Chapters 2 and 3, builds on these early studies. In Chapter 2, we employed advanced chemical biological techniques to assess whether p3 is truly non-amyloidogenic, as indicated by the literature. We found that p3 self-assembles to form oligomers and fibrils morphologically indistinguishable from Aβ, and that these resultant aggregates share confirmational similarities with Aβ. Additionally, we determined that the rate of p3 fibril formation is significantly faster than that of Aβ. These results highlight the solubilizing effect of the N-terminus of Aβ, and the importance of hydrophobic contacts in amyloid formation. In Chapter 3, we investigated the kinetic and biological consequences of mixing p3 and Aβ, to simulate the endogenous heterogeneity of amyloid aggregation in the brains of AD patients. We observed fibrillar colocalization of Aβ and p3, and enhanced aggregation propensity of Aβ upon introduction of p3. This enhancement in stable, insoluble, heterogenous fibril formation resulted in reduced cellular toxicity. We found that under fibril forming conditions, mixtures of Aβ and p3 produced unique oligomers not observed in the homogenous preparations. Additionally, fibril formation proved favorable even under oligomer forming conditions. The enhanced fibril formation resulted in suppression of toxicity and ROS production in both PC12 and SH-SY5Y cells. Additionally, we found that at an early timepoint, TAMRA-Aβ and TAMRA-p3 uptake was comparable, while at a later timepoint, internalization of labeled-peptide was nearly 6x higher for the TAMRA-Aβ treated cells. However, no augmentation of uptake was observed upon addition of unlabeled p3 into TAMRA-Aβ treated cells. In Chapter 4, we discuss challenges in the field of intrinsically disordered proteins (IDPs), and offer novel methods to improve reproducibility. We first propose the benefits of employing all mirror-image peptides to both rigorously control peptide quality, and to probe complicated mechanisms in aggregation and toxicity of Aβ and related peptides. Through comparing the uptake of L- and D-Aβ, we observed that cellular uptake of Aβ is highly stereospecific, indicating that Aβ uptake is likely a receptor-driven process. We also demonstrated “chiral inactivation”, a technique previously developed by the Raskatov lab to abolish toxicity of Aβ42, with the Aβ40 system, which we monitored with 1H NMR. This chapter also presents a structural study of the long-orphaned Pauling-Corey rippled β-sheet. Current knowledge on rippled sheets is limited to a small pool of studies that combined partial experimental structures with theoretical modeling. At the end of Chapter 4, we report a high-resolution crystal structure, in which racemic (L,L,L)- and (D,D,D)-triphenylalanine form dimeric antiparallel rippled sheets, packed into herringbone layers. Overall, the studies described herein highlight the challenges and controversies in probing IDPs, and a few ways to overcome them. Special attention is paid to p3, a peptidic fragment of Aβ previously described as non-amyloidogenic and innocuous. We urge the field of AD-related research to expand the Amyloid Cascade Hypothesis (ACH) in light of these findings, to account for alternative proteolytic fragments of AβPP, and their resultant biological properties.