Our sense of touch plays an important role in how we perceive the world. Touch sensation is the result of an intricate interplay between the nervous system and specialized sensory cells in the skin, one such example being the Merkel cell-neurite complex. Within the Merkel cell-neurite complex, Merkel cells (MCs) detect gentle touch signals in the skin and relay them to innervating neurites via synapse-like connections. Many aspects of the MC-neurite complex, including the molecules required for its formation and structure, remain poorly understood. Our lab recently discovered the presence of MCs in the transparent zebrafish skin, making the organism well-suited for study of MC-neurite complexes. Here, we show that Protocadherin-9 (pcdh9), a cell adhesion molecule important in synaptic structure and nervous system organization, is highly expressed in both zebrafish and mammalian MCs. Using a loss-of-function mutation in zebrafish pcdh9, we find a reduction in the number of MC-neurite complexes, but not in the number of MCs, compared to controls. This suggests a role for Pcdh9 in either the formation or maintenance of MC-neurite synapses. Additionally, we observe a higher rate of MCs contacting one another in pcdh9 mutant skin, consistent with a difference in MC spatial organization. In summary, our data indicate that Pcdh9 may regulate one or more aspects of MC-neurite complex formation. We are now in the process of developing tools to further investigate and quantify MC spatial arrangement, and to uncover the ways in which Pcdh9 may affect MC maturation and behavior.

Skin serves two key functions: hardened cells at the surface of the skin form a superficial layer to protect against the environment, while the inner layers of the skin are packed with diverse sensory machinery which allow us to perceive and navigate the world. Incredibly, the basal most layer of the epidermis houses stem cells which allow the skin to constantly renew itself, fortifying its protective function and maintaining somatosensation by replenishing all these diverse cell types. Perhaps unsurprisingly, these multipotent and highly active skin stem cells are emerging as an effective way to treat genetic skin conditions, promote wound healing, and rejuvenate ageing skin. To understand how skin stem cells contribute to these different functions, investigators are studying the many niches within the skin which may house diverse skin stem cells. Zebrafish are an excellent model to dissect this topic due to their translucent skin and the many genetic tools available. However, the anatomy and molecular characteristics of zebrafish skin is poorly described. Recently, we performed single cell RNA-sequencing of zebrafish skin and identified seven presumptive skin stem cell subpopulations. Informed by this data, I performed whole-mount hybridization chain reaction, a form of in-situ hybridization, to investigate molecular and spatial heterogeneity in zebrafish skin stem cells. My results have identified three novel skin stem cell subpopulations which occupy distinct spatial domains along the anterior-posterior axis. I found that the appearance of each subpopulation and the establishment of their spatial domain is dynamic throughout skin development. Finally, we have constructed a tool to interrogate their behavioral and functional differences. Moving forward, I aim to determine each subpopulation’s role in skin development, homeostasis, and regeneration, as well as whether they serve as specific progenitors for certain cell types.

The cerebellum (Cb) is typically associated with the regulation of motor behavior. However, recent studies have shown that the Cb has connections to other brain regions associated with higher-order cognitive functions, such as the Hippocampus (HPC), suggesting that the Cb is involved in modulating cognitive behavior. The literature also implies that dynamic changes in the levels of the neurotransmitters Dopamine (DA) and Norepinephrine (NE) may underlie the functionality of involved circuits. This project aims to characterize specific cerebellar circuits involved in the modulation of learning and memory. To test this question, I ran a Fear Conditioning (FC) paradigm in which a conditioned stimulus (CS+), a tone, was presented with an aversive unconditioned stimulus (US), a shock. An additional tone, CS-, was presented without shock. A cohort of 7 mice (4 control, 3 experimental) were run through the paradigm and injected with the optogenetic construct ChR2 to stimulate Purkinje Cell (PC) terminals that synapse on the Lateral Cerebellar Nuclei (LCN). Additionally, Fiber Photometry (FP) was used to record changes in DA and NE in the LCN. I used ezTrack’s python packages to analyze freezing % as the measure of learning, and coded python notebooks to process and visualize DA and NE signals. Mice that received ChR2 stimulation on PC terminals displayed better associative learning between CS+ and US compared to mice that did not receive ChR2 stimulation. These mice also extinguished the association of CS+ and US sooner than unstimulated controls. We also expect these mice to display elevated levels of DA and NE that is statistically significant compared to unstimulated controls. These experiments elucidating cerebellar circuitry in cognitive behaviors may serve to expand our understanding of neural substrate alterations in cognitive function. The LCN may also be a novel locus for targeted therapeutics in human affective disorders.

The auditory brainstem response (ABR) tests are used to objectively evaluate the clinical hearing threshold of infants and young patients. However, the ABR testing process can be time and resource-consuming, as audiologists have to test multiple frequencies. For each frequency, an ABR threshold (the lowest level at which a discernable ABR response is detected) must be determined by repeating the test for a multitude of levels. The efficiency of these tests depends on clinical expertise. Audiologists can expedite this process by utilizing their experience to quickly analyze the ABR waveform and jump to the next test, skipping redundant intermediary steps. Clinicians with this expertise might not be widely available. To address this issue, the long-term goal of this study is to create an automated system that can mimic the efficient testing procedure of experienced audiologists using machine learning. A set of clinical ABR data was leveraged for model development. Our baseline models operate by analyzing one waveform at a time and predicting the next stimulus a clinician would choose based on individual waveforms. We hypothesize that a neural network that treats ABR waveforms collected in a single session as a time series would outperform baseline models. We are comparing these baseline models with neural networks that hold memory, meaning they treat ABR waveforms collected in a single session as a sequence. Multiple models were built and evaluated, including multiple time series neural networks (e.g., Long-Short Term Memory model). Initial testing indicates that including sequential data ordered as time series results in better performance. The outcome of this research is likely to improve the efficiency of ABR testing without requiring real-time supervision of expert clinicians.