Sharon G. Kujawa - US grants

DESCRIPTION (provided by applicant): Acoustic overexposure is a growing problem, and understanding the long-term consequences is critical to public health. Our recent work in a mouse model of noise and aging shows that moderate exposures, which initially appear reversible and cause no acute or chronic hair cell loss, elicit a slow-onset loss of spiral ganglion cells (SGCs) when followed for months post-exposure. Confocal immunohistochemistry suggests that many SGC peripheral terminals, and their synapses on inner hair cells, disappear within the first days or hours, consistent with acute excitotoxic effects of acoustic overexposure. We hypothesize that this acute dendritic retraction disrupts normal neurotrophin signaling among hair cells, supporting cells and neurons in the cochlear epithelium, and that this interruption initiates the slow cell-death cascade in SGCs. The proposed Aims test this hypothesis by characterizing the nature and time course of neuronal degeneration and associated pathophysiology (Aim 1), and by manipulating acute excitotoxicity (Aim 2) or neurotrophin expression (Aim 3) and assessing the effects on cochlear neurodegeneration. Quantification of neuropathy (Aim 1a), will track degeneration over post-exposure time as it progresses from synapse, to peripheral axon, to cell body. Correlations with pathophysiology at population-response and single-fiber levels (Aim 1b) will verify if we have identified the functionally important structural changes, and will test the hypothesis that noise causes a preferential loss of neurons with low spontaneous rates. To test if acute excitotoxicity is the key upstream elicitor of the neuropathy, we exploit our techniques for cochlear perfusion in mouse to either 1) block noise-induced excitotoxicity with the glutamate antagonist DNQX (Aim 2a), 2) mimic it with the glutamate agonist AMPA (Aim 2b) or 3) enhance it using mice with targeted deletion of the glutamate transporter GLAST (Aim 2c). To test the role of neurotrophins in the slow cascade of cell death, we will assay (via qRT-PCR, immunohistochemistry and a NT3-reporter mouse) gene expression levels of key molecules in the neurotrophin signaling pathway as a function of post-exposure time (Aim 3a), and attempt a rescue experiment (reduce/prevent loss of neurons;promote re-innervation of intact IHCs) using mouse lines with inducible neurotrophin overexpression in either hair cells or supporting cells (Aim 3b). Understanding the nature, etiology and possible prevention of slow-onset neurodegeneration in our noise- exposed mice has important ramifications for human hearing. It suggests that primary neuronal loss is a more common and important aspect of acquired sensorineural hearing loss than previously thought. It also raises important concerns re possible long-term consequences of apparently benign acoustic overexposures: the phenomenon of slow-onset noise-induced neurodegeneration may contribute in a major way to the main hearing-related complaint in aging humans, i.e. problems understanding speech in a noisy environment. PUBLIC HEALTH RELEVANCE Our recent work in mouse shows that noise-exposures, even those that appear to result in fully reversible threshold shifts, actually set in motion a slow cell death cascade leading to the ultimate loss of roughly half of the neural elements throughout large regions of the cochlea. If generally applicable to the mammalian ear, as there is every reason to believe it will be, the phenomenon of slow-onset, noise-induced, primary, cochlear-nerve loss is potentially a very common problem with significant public health implications. Our proposed experiments provide a powerful platform to study the phenomenon and to probe its mechanisms, using a cochlear insult (i.e. noise) that is highly relevant to the human condition.

Administrative Core Project Summary The Administrative Core for this P50 includes the 1) Program Director, 2) a Grants Manager, 3) a Subject Coordinator and 4) a Biostatistician co-investigator and his assistant. An External Advisory Committee comprised of three senior scientists with highly relevant expertise will provide oversight of the Center's scientific progress and will guide the research plan when appropriate. The purposes of this Administrative Core are to: Aim 1: Provide leadership and administrative oversight of projects and progress Aim 2. Foster inter-project collaboration and interactions with External Advisors Aim 3. Manage shared resources and resolve conflicts

Overall Project Summary New insights from animal studies of noise-induced and age-related hearing loss suggest that the most vulnerable elements in the inner ear are the synaptic connections between hair cells and sensory neurons. This primary neural degeneration, also called cochlear synaptopathy, does not elevate thresholds. Thus, it can be widespread in ears with intact hair cell populations and normal audiograms, where it has been called ?hidden? hearing loss. It likely contributes to difficulties understanding speech in a noisy environment and may be an instigating factor in the generation of tinnitus and hyperacusis. Cochlear synaptopathy may also be widespread in acquired sensorineural hearing loss of other etiologies and degrees of hair cell damage. Thus, it may be a major contributor to the well-known differences in auditory performance among people with identical audiometric patterns of ?overt? hearing loss. Our Research Center aims to take these paradigm-shifting ideas from animal models to human subjects. Based on the synthesis of many research threads from the study of overt and hidden hearing loss, we have devised a set of physiological, electrophysiological and psychophysical tests of hearing and cochlear function that we believe are most powerful in the diagnosis and understanding of cochlear synaptopathy in human subjects. In Project 1, we apply this test battery to gerbils exposed to noise or ototoxic drugs and test their diagnostic power by directly measuring the underlying cochlear histopathology in cases of overt or hidden hearing loss. In Project 2, we use immunostaining to directly assess the prevalence of cochlear synaptopathy in human temporal bones from subjects with overt or hidden hearing loss with a range of etiologies. In Project 3, we study hidden hearing loss in college students by applying the test battery to subjects with normal audiograms but a broad range of reported and measured sound exposures. In Project 4, we assess older adults with overt hearing loss by applying the tests to a subject pool with carefully matched down-sloping audiograms and by characterizing training-based improvements in speech-in-noise performance as reflected at different peripheral, brainstem, midbrain and cortical levels. Our preliminary studies of young adults show clear signs of hidden hearing loss in a group with repeated exposure to high-level music, suggesting the importance of this phenomenon to the public health. The success of neurotrophin-based approaches to the treatment of cochlear synaptopathy in animal models suggests that therapies may be on the horizon. Thus, the need for better understanding of the prevalence, diagnosis and functional consequences of cochlear synaptopathy is clear.

Project 1 Summary ? Abstract In common causes of human hearing loss like aging and noise exposure, permanent threshold losses are associated with permanent cochlear injury, often hair cell damage or loss. Recently, work in animal models has revealed what may be a more common consequence of these and other causes of acquired sensorineural hearing loss. This work has shown that synapses between inner hair cells (IHCs) and cochlear neurons are most vulnerable, with their loss interrupting sensory-to-neural communication long before loss of the hair cells themselves, and long before sensitivity losses appear on the threshold audiogram. The silencing of affected neurons that results is a likely contributor to a variety of auditory perceptual abnormalities, including speech-in- noise difficulties, tinnitus and hyperacusis that can occur with or without threshold sensitivity loss. As these findings are translated to the study of human hearing loss, animal models will continue to provide a powerful approach to test hypotheses, to characterize structural and functional consequences of carefully- titrated manipulations and to evaluate the sensitivity of the assessments to the underlying histopathology. Here, animal models of sensorineural hearing loss etiologies common in humans; exposure to noise, to aminoglycoside antibiotics and to platinum-containing chemotherapeutics, will be created. The models will address the mixed (sensory + neural) pathology that will likely be present in many of the humans and human temporal bones evaluated in the other Projects. The human test battery will be applied (Aim 2) and its diagnostic power assessed by directly measuring the underlying cochlear histopathology (Aim 1). Structure- function correlations will be probed further using detailed electrophysiologic assays that might be streamlined for future clinical use (Aim 3). Work will be performed in gerbil, a species with good low frequency hearing and can be trained to perform auditory tasks. By correlating performance on these complex listening tasks with electrophysiology in the same subjects and with explicit measurement of the underlying synaptopathy, the contribution of cochlear neuropathy to the perceptual declines can be quantitatively evaluated and results can be directly compared to those obtained in human subjects. An improved understanding of the extent to which synaptic mechanisms are damaged in common forms of human sensorineural hearing loss will have broad implications for efforts to identify drugs or other treatments with the potential to target these mechanisms for prevention or rescue. Practically, this knowledge will inform clinical diagnostics, the monitoring of new treatments for efficacy or the monitoring of individuals at risk of hearing compromise from drug and noise exposure. It also may help explain auditory performance differences among individuals with the same audiometric configurations, even for those with normal thresholds.