Amyloidogenic diseases, such as Alzheimer's disease (AD) and Type II Diabetes (T2D), are characterized by the accumulation of amyloid aggregates. Despite having very different amino-acid sequences, the underlying amyloidogenic proteins form similar supramolecular fibril structures that are highly stable and resistant to physical and chemical denaturation. AD is characterized by two toxic lesions: extracellular amyloid β-peptide (Aβ) plaques and intracellular neurofibrillary tangles composed of microtubule-associated protein tau. Similarly, a feature of T2D is the deposition of islet amyloid polypeptide (IAPP) aggregates in and around the pancreas. The mechanisms by which Aβ, tau, and IAPP aggregate, and cause cell death is unknown; thus, gaining greater insight into the stabilizing forces and initial unfolding events is crucial to our understanding of these amyloidogenic diseases. This work uses molecular dynamics (MD) simulations to study the secondary, tertiary, and quaternary structure of Aβ, tau, and IAPP. Specifically, this work used the Drude polarizable force field (FF), which explicitly represents electronic polarization allowing charge distributions to change in response to perturbations in local electric fields. This model allows us to describe the role charge plays on protein folding and stability and how perturbations to the charge state drive pathology. Studies were conducted to address the following questions: 1) What are the stabilizing forces of fibril and oligomeric structures? 2) How do charge-altering mutations modulate the conformational ensemble and thermodynamic properties of Aβ? 3) How do charge-altering post-translational modifications of Aβ and tau modulate changes in the conformational ensembles? These studies establish that shifts in local microenvironments play a role in fibril and oligomer stability. Furthermore, these studies found that changes in protein sequence and charge are sufficient to disrupt and change the secondary and tertiary structure of these amyloidogenic proteins. Overall, this dissertation describes how charge modulates protein unfolding and characterizes the mechanism of those changes. In the long term, this work will help in the development of therapeutics that can target these changes to prevent protein aggregation that leads to cell death. / Doctor of Philosophy / Protein aggregation is the hallmark of many chronic diseases, such as Alzheimer's disease (AD) and Type II Diabetes (T2D). The formation of two toxic aggregates: amyloid β-peptide (Aβ) plaques and neurofibrillary tangles composed of microtubule-associated protein tau are some of the key characteristics of AD. In addition, the formation of islet amyloid polypeptide (IAPP) aggregates in the pancreas is thought to play a role in the development of T2D. The pathways by which the proteins Aβ, tau, and IAPP aggregate are unknown; thus, gaining a greater insight into the properties that may cause these diseases is necessary to develop treatments. By studying these proteins at the atomistic level, we can understand how small changes to these proteins alter how they misfold in a way that promotes toxicity. Herein, we used a computational technique called molecular dynamics (MD) simulations to gain new insights into how protein structure changes. We explored the dynamics of these proteins and investigated the role that charge plays in protein folding and described how charge modulates protein folding and characterized the mechanism of those changes. This work serves as a characterization of protein folding and sets the ground for future structural studies and drug development.

Studies presented in this dissertation were conducted to gain more insight into the role of phospholipid asymmetry and amyloid-β (Aβ)-induced oxidative stress in brain of subjects with amnestic mild cognitive impairment (aMCI) and Alzheimer disease (AD). AD is a largely sporadic, age-associated neurodegenerative disorder clinically characterized by the vast, progressive loss of memory and cognition commonly in populations over the age of ~65 years, with the exception of those with familial AD, which develop AD symptoms as early as ~30 years-old. Neuropathologically, both AD and FAD can be characterized by synapse and neuronal cell loss in conjunction with accumulation of neurofibrillary tangles and senile plaques. Elevated levels of oxidative stress and damage to brain proteins, lipids, and nucleic acids are observed, as well. Likewise, aMCI, arguably the earliest form of AD, displays many of these same clinical and pathological characteristics, with a few exceptions (e.g., no dementia) and to a lesser extent. Studies in this dissertation focused on the contributions of oxidative stress to the exposure of phosphatidylserine (PtdSer) to the outer-leaflet of the lipid membrane, how and when PtdSer asymmetric collapse contributes to the progression of aMCI, AD, and FAD, and the role played by methionine-35 (Met-35) of Aβ in oxidative stress and damage, as measured in a transgenic mouse model of Aβ pathology. Normally, the PtdSer is sequestered to the cytosolic, inner-leaflet of the bilayer by the adenosine triphosphate (ATP)-dependent, membrane-bound translocase, flippase, which unidirectionally transports PtdSer inward against its concentration gradient. Oxidative stress-induced modification of flippase and/or PtdSer, however, leads to prolonged extracellular exposure of PtdSer on the outer membrane leaflet, a known signal for both early apoptosis and selective recognition and mononuclear phagocytosis of dying cells. Within the inferior parietal lobule (IPL) of subjects with aMCI and AD, a significant collapse in PtdSer asymmetry was found in association with increased levels of both pro- and anti-apoptotic proteins, Bax, caspase-3, and Bcl-2. Moreover, a significant collapse in PtdSer asymmetry was also found in whole brain of human double-mutant knock-in mouse models of Aβ pathology, together with significantly reduced Mg2+ATPase activity, representing flippase activity, and increased levels of pro-apoptotic caspase-3. Significant PtdSer externalization corresponded to the age at which significant soluble Aβ(1-42) deposition occurs in this particular mouse model (9 months), and not of plaque deposition (12 months), suggesting that elevated levels of Aβ(1-42), together with increasing oxidative stress and apoptosis, may contribute to altered PtdSer membrane localization. Also in this dissertation, transgenic mice carrying Swedish and Indiana mutations on the human amyloid precursor protein (APPSw,In) and APPSw,In mice carrying a Met35Leu mutation on Aβ were derived to investigate the role of Met-35 in Aβ(1-42)-induced oxidative stress in vivo. Oxidative stress analyses revealed that Aβ-induced oxidative stress requires the presence of Met-35, as all indices of oxidative damage (i.e., protein carbonylation, nitration, and protein-bound 4-hydroxy-2-trans-nonenal [HNE]) in brain of Met35Leu mice were completely prevented. Moreover, immunohistochemical analyses indicated that the Met35Leu mutation influences plaque formation, as a clear reduction in Aβ-immunoreactive plaques in Met35Leu mice was found in conjunction with a significant increase in microglial activation. In contrast, behavioral analyses suggested that spatial learning and memory was independent of Met-35 of Aβ, as Met35Leu mice demonstrated inferior water-maze performance compared to non-transgenic mice. Differential expression and redox proteomic analyses to pinpoint proteins significantly altered by the APPSw,In and Met35Leu mutations was performed, as well. Expression proteomics showed significant increases and decreases in APPSw,In and Met35Leu mouse brain, respectively, in proteins involved in cell signaling, detoxification, structure, metabolism, molecular chaperoning, protein degradation, mitochondrial function, etc. Redox proteomics found many of these same proteins to be oxidatively modified (i.e., protein carbonylation and nitration) in both APPSw,In and Met35Leu mouse brain, providing additional insights into the critical nature of Met-35 of Aβ for in vivo oxidative stress in a mammalian species brain, and strongly suggesting similar importance of Met-35 of Aβ(1-42) in brain of subjects with aMCI and AD. Taken together, studies presented in this dissertation demonstrate the role of oxidative stress-induced alteration of PtdSer asymmetry and Met-35 in Aβ-induced oxidative stress in aMCI, AD, and FAD brain.