Jeffrey N. Savas - US grants

DESCRIPTION (provided by applicant): The mammalian brain is an immensely complex organ in terms of diversity of resident cell types, anatomical organization, neuronal circuitry, modes of intercellular communication, and the regulation of protein expression. Further the effect aging and Alzheimer's disease (AD) has on the brain represents a colossal challenge for researchers. Previous mass spectrometry based proteomic technologies have been most adept at identifying protein-protein interactions or post-translational modifications of enriched proteins from a single cell type. The recent development of in vivo isotopic labeling of mammals and high resolution mass spectrometers has afforded the opportunity to determine the relative protein expression level of thousands of proteins from tissue. We aim to calculate changes in protein expression during Alzheimer's disease progression and regular aging in the mammalian brain by quantitative mass spectrometry. This project will yield a large-scale anatomic inventory of protein expression on an unprecedented level which can be mined for years to come. These quantitative proteomic studies will serve as a hypothesis-generating machine that will likely uncover new and unexpected data on the effect aging and AD have on the brain. Those in the advanced stages of AD become bed-bound and reliant on care 24/7 which in total results in 148 billion dollars in annual costs in the US alone. The effect of aging and AD on brain physiology has been historically investigated in candidate-based approaches. While these candidate approaches have significantly contributed to our understanding of aging and AD, they have been limited by their restricted scope. To broaden our knowledge base of these processes and eventually develop effective therapeutics for the treatment of AD we will calculate the expression level for thousands of proteins. Proteins with perturbed expression will serve as beacons for the identification of pathways relevant to AD pathology and aging. Determination of perturbed pathway(s) will in turn serve as fertile ground for in-depth analysis aimed at deciphering the molecular basis of AD and aging. More generally, this project will deliver a brain atlas of protein expression that occurs during aging and AD. AD represents a significant threat to the aging world population: AD is considered the world's most common neurodegenerative disease, affecting over 5 million aging Americans, and is a rising threat to public health. AD debilitates individuals by stripping them of the ability to reason, use language, and recall memories, resulting in a tremendous caretaking burden. Currently there is no cure or definitive treatment for AD and it remains the leading cause of dementia. AD is stratified by the age at which pathological onset occurs. The two forms of AD, early-onset and late-onset, both have genetic links. Identification of new genes and/or proteins that contribute to AD pathology could provide a critical first step for the development of effective AD treatments.

PROJECT SUMMARY Hearing impairment is one of the most common sensory disabilities, 250 million people worldwide have moderate to severe hearing loss (38), and significantly reduces quality of life due to the central role of verbal communication. On an economic scale, the total negative impact of hearing loss is greater than that of multiple sclerosis, spinal cord injury, stroke, epilepsy, Parkinson's and Huntington's disease combined and effects 4 times as many people (68). The most common causative factor among the defined hearing loss etiologies is excessive noise, and millions of people are exposed to dangerously loud noise at work. We hope that our research findings will aid in the reduction of noise-induced hearing loss by identifying significant proteins and pathways responsible for hearing loss. Specifically, we have developed a quantitative proteomic analysis platform to probe the effect of excess noise on the cochlear proteome. Our preliminary data shows this approach can accurately measure thousands of proteins from a single mouse and has already revealed proteins significantly perturbed after noise exposure. We have also made progress isolating the organ of Corti to ensure the accurate measurement of low abundant but potentially altered proteins in hair and adjacent support cells, and in cochlear nerve synapses. More specifically, we will quantitatively analyze inner ear extracts from mice exposed to multiple levels of noise. Through these comparative exposures, we will differentiate proteins with characteristics that are impacted by excess noise. The candidates from these proteomic experiments will be explored with bioinformatic tools and validated by traditional antibody based approaches. Next we will develop biochemical methods to ensure the accurate measurement of rare low abundance proteins. We will also test if protein?protein interactions are disrupted without significant changes in expression levels. Finally we will use bioactive molecules known to protect from NIHL and repeat the proteomic analysis to investigate the mechanisms by which these drugs are effective. In particular, we think that a comprehensive understanding of the inner ear proteome will accelerate the greater research field of hearing injury. In summary, we propose here to identify and investigate molecular defects in NIHL by applying quantitative proteomic tools that can simultaneously and sensitively investigate thousands of proteins in a single analysis. We believe this proposal represents the first ever application of quantitative proteomics to the investigation of NIHL and may hold the required analytical strength to kick start the development towards effective therapeutics to eventually treat and prevent NIHL.

Amyotrophic lateral sclerosis (ALS) is a progressive and untreatable neurodegenerative disease that is characterized by the selective death of upper and lower motor neurons (MNs). The overwhelming majority of the disease is sporadic in nature. However a relatively small (<12%) but highly informative fraction of patients suffer from familial forms of disease, which have enabled the identification of causative genetic variants that underlie their condition. Such genetic studies have demonstrated that ALS can be caused by mutations in genes that encode proteins involved in diverse set of cellular functions ranging from RNA processing, vesicle transport, cytoskeletal regulation, mitochondrial function, and protein quality control pathways. Nevertheless, ALS patients are uniformly characterized by a common pattern of progressive motor neurodegeneration. This raises the possibility that different disease initiating events could coalesce in one or more common molecular pathways. How the mutation of genes with dissimilar functions converge on MN degeneration has been and continues to be an outstanding question. Although all ALS patients exhibit neuropathological protein aggregates, the overall contribution of protein homeostasis in causing ALS has remained unclear. If we could identify a convergent mechanism, it may provide an opportunity to develop a broadly applicable therapeutic intervention strategy. In our preliminary studies, we conducted global analysis of protein degradation dynamics in mutant SOD1 and isogenic controls MNs derived from iPSC lines. Interestingly, we identified a number of proteins that are degraded at a slower rate in SOD1 MNs. Unexpectedly, this small panel of candidates included proteins whose genetic mutations cause ALS. In the proposed research we will use patient-derived neurons coupled with mass spectrometry analysis to determine the protein substrates, as well as the nature of the perturbation that arise as a result of mutations in the two most prevalent ALS genes: SOD1 and C9orf72. First, we will determine which proteins have reduced protein degradation dynamics. Second, we will determine which proteins have altered synthesis rates. Third, we will determine the overall degree of proteome-wide remodeling. Each of these approaches has strategic advantages over traditional work-flows and will allow us to determine not only which proteins have altered levels in ALS MNs but also the mechanism responsible for their perturbation. Taken together, our proposed aims will shed light into the cellular mechanisms compromised by changes in the proteostasis network in patient neurons and will likely uncover broadly relevant therapeutic targets for ALS.

The female reproductive system is an instructive model for studying aging mechanisms because it ages decades prior to other organs in the human body. Reproductive function begins to decline when women are only in their mid-30s and ceases completely at menopause. Female reproductive aging phenotypes include reduced endocrine function and decreased gamete quantity and quality. Together, these changes contribute to adverse fertility and general health outcomes, and such consequences are becoming more tangible as medical advances are extending lifespan and women worldwide are delaying childbearing. Several fundamental hallmarks of aging tissues have been identified including impaired protein homeostasis or proteostasis. Proteins are essential for the structure and function of all tissues and are involved in critical cellular processes. As such, regulatory mechanisms are in place to ensure that proteins are synthesized, folded, and modified properly. As proteins age, however, they accumulate various types of damage, and damaged proteins are typically turned over and replaced with newly synthesized functional versions to maintain proteostasis. Most proteins last only a total of two days or less. However, a unique class of proteins called extremely long lived proteins (ELLPs) can last the entire lifetime of an organism without being replaced. ELLPs are typically part of large complexes (e.g. histones, nuclear pores, structural networks) and underpin aging because accumulated damage compromises their function and may also elicit abnormal signaling pathways. The pathogenic properties of ELLPs are particularly problematic in post-mitotic cells because they are not diluted through cell division. In fact, neuronal ELLPs are implicated in aging and neurodegenerative conditions. Like neurons, the mammalian oocyte is also vulnerable to ELLP dysfunction because it is non-dividing but rather maintained in an extended prophase I arrest for up to months in mouse and decades in human. Thus, damaged ELLPs could accumulate, reduce gamete quality, and may even be passed onto the next generation through the embryo. While ELLPs provide a compelling intellectual framework for considering mechanisms of female reproductive aging, their identification and quantification at the single protein level has been historically challenging due to technical limitations especially within limited biological material. Here we propose to combine a two generation whole animal stable pulse- chase isotope labelling approach with advanced mass spectrometry-based approaches to identify and quantify the extremely long lived proteome in the ovary and oocyte (Aim 1) and to image and quantitatively analyze long lived molecules (proteins, lipids, nucleotides) in the oocyte and ovary within the context of the in vivo microenvironment (Aim 2). These experiments will be performed at two time points across the reproductive aging continuum and encompass the necessary pioneering steps to identify specific extremely long lived substrates and structures within the mammalian oocyte and ovary. The discoveries made here will lay the foundation for future mechanistic studies and have implications not only for reproductive aging but aging tissues more broadly.

ABSTRACT Synaptic dysfunction represents a core pathological hallmark of Alzheimer's disease (AD). Impairments in synapses underlie changes in the activity of neuronal circuits, which ultimately drive impaired cognition in AD. Despite this, we currently have an incomplete understanding of how and why synapses are altered in AD pathology. If we could advance our understanding of the underlying mechanisms that cause synaptic dysfunction in AD, it would be a significant advancement in the overall understanding of AD. Recent evidence has suggested that pathogenic amyloid beta and tau are enriched in extracellular vesicles and exosomes; but how these vesicles contribute to disease progression, particularly how they affect synaptic function, is still unknown. In preliminary studies, we provide evidence that activation of synaptic NMDA receptors triggers exosome release in neurons, and that this process regulates the level of many synaptic proteins. Importantly, an analysis of these complex datasets using multiple bioinformatic strategies, highlighted key protein-protein interaction networks that are regulated by exosomes including the presence of AMPA receptors, amyloid precursor protein (APP), and tau. Proteomic analysis of purified exosomes from stimulated neurons revealed the presence of several proteins known to cause neurodegeneration, and it suggests that this process may have a major role in the spreading of pathology. Inhibition of exosome synthesis in WT neurons eliminated synaptic potentiation demonstrating a previously unknown function of exosome signaling. Analysis of purified exosomes from the brains of multiple mouse models with AD-like pathology demonstrated an enrichment of APP and amyloid beta peptide as well as an alteration of exosome protein cargos. Based on this evidence, we hypothesize that normal exosome signaling is hijacked by AD pathology contributing to the synaptic dysfunction known to be prevalent in the disease. We propose that activity-induced exosomes, which normally support synaptic strengthening, are overwhelmed by aberrant enrichment of pathogenic molecules which results in disrupted synapses, in particular, impaired synaptic strengthening. We plan to test this hypothesis by combining orthogonal techniques and disciplines including electrophysiological measurement of synaptic function, non-biased proteomic approaches, and imaging of exosome release dynamics. The proposed research has the potential to transform our understanding of how altered activity-induced exosome signaling may contribute to synaptic dysfunction and spreading of AD- like pathology.

Synaptic plasticity in the hippocampus is critical to the formation, storage and retrieval of episodic memories. The separate regions of the hippocampus have evolved to play distinct roles in spatial navigation, contextual memories, social memories, and our ability to separate patterns or complete patterns to reconstruct partial memories. In particular the dentate and CA3 regions of the hippocampus are involved in our pattern separation that is vital to the integrity of episodic memories. At the center of this region are the mossy fiber afferents that make conditional detonator synapses onto CA3 pyramidal neurons, which have a distinct form of presynaptic cAMP dependent plasticity. Despite the importance of cAMP plasticity to memory formation and retrieval in the CA3 the exact molecular mechanisms underlying MF LTP have not been uncovered. The premise of this research builds upon our finding that there are at least two downstream cAMP effectors, PKA (protein kinase A) and Epac2 (exchange protein directly activated by cAMP 2), that contribute to cAMP dependent MF LTP. Despite these findings it is still not known how signaling by each of these effectors results in elevated release from MF synapses and, what are the important targets and substrates that are involved in MF LTP. Here we will use a comprehensive approach with leading edge proteomic, biochemical and electrophysiological approaches to determine the signaling partners of the cAMP effectors, and uncover the physiological mechanism of their actions. Thus, in Aim 1 we will take orthogonal approaches to find the interactors and substrates of PKA and Epac2 and validate and verify these by performing high resolution labeling in situ. In Aim 2 we will determine the exact physiological mechanism that underlie increases in release of neurotransmitter at MF synapses using a combined optogenetic-knockout/pharmacological strategy. In the final Aim we will answer the question of how these different but convergent mechanisms are engaged during naturalistic activity patterns, and whether selective disruption of these effectors impairs the ability of mice to separate similar patterns that underlie the formation and retrieval of episodic memories.

ABSTRACT Copy number variations (CNVs) are a major cause of neurodevelopmental disorders, but their biological investigation and pharmacological targeting pose many challenges. Deletions locus are among the most frequent causes of autism spectrum disorder and duplications at the 16p11.2 (ASD). However, alterations in the corresponding protein networks, especially at key cellular sites for pathogenesis, have not been investigated in this or other CNVs. We propose to use compartment-specific neuroproteomics, combined with bioinformatics, super-resolution microscopy, and drug repurposing, to understand and alter dendritic excitability phenotypes in 16p11.2 mouse and induced pluripotent stem cell (iPSC) models. Based on our extensive preliminary data, we hypothesize that altered expression of PRRT2, which likely regulates the trafficking of a subset of ion channels and receptors, drives and abnormal complement of ion channels and receptor on the plasma membrane, leading to abnormal excitability, excitatory/inhibitory (E/I) balance, and network properties in 16p11.2 models and patients. These phenotypes may be reversed by targeting ion channel function using FDA- approved anti-epileptic drugs or ERK signaling using repurposed cancer drugs. Our collaborative team, which includes experts in neurodevelopmental disorders (Penzes), neuroproteomics (Savas), molecular pharmacology (Barbolina), and ion channel physiology (George) will employ a powerful and multidisciplinary combination of highly innovative methodologies to pursue the following Specific Aims: (1) To chart the developmental regulation and determine molecular mechanisms underlying abnormal excitability in dup and del mice and human neurons. (2) To chart the developmental profile and determine the molecular mechanisms underlying the role of PRRT2 as a driver of excitability and seizure phenotypes. (3) Pharmacological reversal of 16p11.2 del and dup phenotypes. This proposal will be the first to demonstrate that cellular subcompartment-specific proteomics combined with super-resolution microscopy, informed by highly penetrant monogenic disease genes within a CNV, can identify novel disease mechanisms. Such phenotypes could be reversed globally by targeting network hubs using repurposed drugs, opening novel strategies for the treatment of neurodevelopmental disorders.

PROJECT SUMMARY Spread of neuroinvasive herpesviruses from sensory neurons to the eye, brain, or from mother to newborn, are significant causes of morbidity and mortality. Herpes simplex virus type 1 (HSV1) and pseudorabies virus (PRV) are representative members of the two genuses of mammalian neuroinvasive herpesviruses (simplexviruses & varicelloviruses). These viruses are dependent upon spread to the nervous system to establish life-long latent infections, yet very little is known regarding the neuroinvasive mechanism that underlies this remarkable trait. We propose to study the virus neuroinvasive machinery with the intent to: (i) decipher how these viruses invade the nervous system, (ii) understand the intrinsic barriers to neural infection that these viruses evade, and (iii) produce and characterize viruses lacking the neuroinvasive property as potential vaccines and recombinant vectors. These studies are designed to identify the virus-cellular interactions that promote virus genome delivery to the nuclei of non-neuronal and neuronal cells, and the corresponding intrinsic defenses that keep most pathogens at bay. We include preliminary data demonstrating that this path-breaking collaborative study has far-reaching medical and biological implications.