Auditory and vestibular function are dependent of the formation of a functional inner ear. While there are multiple components for both of these systems, this laboratory focuses on the development of the sensory epithelia, which contain mechanosensory hair cells and associated cells called supporting cells and on the innervation of those hair cells by neurons from the VIIIth (acousticovestibular) cranial nerve. All three of these cell types are derived from the otocyst, a placodal structure that forms adjacent to the hindbrain early in development. Identifying the factors the specify each of these cell types and then direct their assembly into functional units is a key goal of the Section on Developmental Neuroscience. During the previous year, different members of the laboratory have examined several different aspects of these developmental processes. A key step in development of the inner ear is the formation of mechanosensory hair cells. Previous work from our laboratory, as well as several others, has demonstrated that the transcription factor Atoh1 plays a key role in the induction of a hair cell fate. Based on these results, Atoh1 has been proposed as a possible candidate gene for the development of gene therapy approaches to potentially repair defects in auditory or vestibular function. However, an important un-answered question was whether forced expression of Atoh1 alone is sufficient to generate functional hair cells. To address this question, we used a transgenic mouse line in which expression of Atoh1 can be induced in adult supporting cells in vivo. After inducing Atoh1 expression in utricular supporting cells, we allowed maintained mice for a period of time before analyzing the effects of Atoh1 expression on hair cell formation. Results indicated that forced expression of Atoh1 is sufficient to induce hair cell formation, but the only one type of hair cell, type 2 hair cells are produced. Normally the utricle contains two types of hair cells, type 1 and type 2. The lack of type 1 hair cells suggests that Atoh1 alone is not sufficient to induce full recovery of utricular function. Hair cell stereociliary bundles are directionally sensitive such that only deflections in the direction of the tallest stereocilia lead to activation of mechanosensory transduction channels. As a result, the appropriate orientation of stereociliary bundles is crucial for normal function. Work from this laboratory and others identified a crucial role for the conserved planar cell polarity (PCP) pathway in stereociliary bundle polarization. More recently, changes consistent with disruption of PCP signaling have been reported in patients with mutations in genes related to genesis of or trafficking within cilia. In addition, some patients with ciliopathic syndromes suffer from hearing loss, suggesting a possible role for cilia in inner ear PCP. To directly address this possibility, we examined inner ear phenotypes in five different mouse lines with targeted mutations in cilia genes. Results were surprisingly variable with some lines demonstrating no obvious PCP defects. However, two deletions, Bbs8 and Ift20 (conditional to the inner ear) led to significant PCP-like defects in the inner ear. Bbs8 mutants displayed mis-oriented and misshapen bundles as well as mis-localized or missing kinocilia. Similar defects were observed in vestibular sensory epithelia. Conditional deletion of Ift20 beginning at the otocyst stage resulted in bundle phenotypes similar to those observed in Bbs8 mutants, even though cilia were completely absent on all inner ear cells. In addition, co-immunoprecipitation experiments demonstrated interactions between both Bbs8 and Ift20 and the crucial PCP protein Vangl2. Finally, sub-cellular localization of Vangl2 and another key determinant of PCP, Gai3, were shown to be disrupted in Bbs8 or Ift20 mutants, suggesting that both cilia-related genes act upstream of initial PCP signaling. The development of appropriate innervation patterns by spiral ganglion neurons (SGNs) is crucial for normal auditory function. Following their arrival at the sensory epithelium, SGNs must synapses with IHCs and OHCs. Most SGN fibers are considered to be Type Is meaning that they form synapses with IHCs, while a much smaller number (only 5 10 %) are considered to be Type II fibers which synapse with OHCs. Innervation patterns for the two types of SGNs differ markedly however the factors that mediate these patterns are largely unknown. A microarray screen that we performed several years ago indicated expression of multiple Semaphorins (Semas) within the embryonic cochlear epithelium. Since Semas regulate innervation patterns in other systems, we determined the cellular distribution of candidate Semas by in situ hybridization. Results indicate that Sema3F is expressed in a particularly intriguing pattern that includes a sharp medial boundary correlating with the separation between Type I and Type II fibers. Since Nrp2 is expressed by developing SGNs, it seemed possible that an inhibitory interaction could prevent extension of type I fibers into the OHC region. To test this hypothesis SGN innervation patterns were examined in Nrp2 and Sema3F mutants. In both cases, significantly increased numbers of fibers were observed in the OHC domain, confirming a role for this interaction in limiting the number of fibers that extend past the IHCs. These results suggest that the OHC domain of the embryonic OC is inhibitory for Type I fibers. To test this, we developed a live imaging technique that allowed visualization of SGN fibers as they extended into the cochlear sensory epithelium. Analysis of time-lapse images revealed numerous examples of fiber retraction from the OHC domain. The results of these experiments identify molecular interactions that are required for the formation of SGN innervation patterns and also have the potential to provide insights regarding interactions between cochlear implants and residual SGNs. The inner ear contains a comparatively large number of unique cell types, however the number of cells of each cell type present in a given ear is fairly limited. The recent development of new technologies for the isolation and transcriptional characterization of multiple single cells from a given tissue provides a novel methodology for identifying these cells and for understanding the development of the inner ear. As a first step towards characterizing different inner ear cell types, we isolated and characterized approximately 300 cells from the P1 cochlea and utricle. In order to provide quality control and validation, cells were isolated from triple-transgenic mice (LfngGfp;Gfi1cre;R26RtdTomato) in which virtually all supporting cells express Gfp and all hair cells express tdTomato. Following dissection of sensory regions, epithelial cells were dissociated and single cells were captured using a Fluidigm microfluidics platform. Each cell was then imaged prior to lysis, reverse transcription and initial amplification using SMARTer/Nextera kits. A unique, cell-specific, bar code was then applied to the cDNA from each sample and samples from 48 cells were then combined for multiplex RNA-Seq on an Illumina Hi-Seq. Individual reads were de-multiplexed and aligned to the Ensembl mouse genome. Average expression data from combined single cells showed a strong correlation with RNA-Seq results from tube controls containing 200-300 cells. Based on these results we were able to identify new lineage relationships within the inner ear as well as to model the changes in gene expression that occur during hair cell development. Ongoing studies will continue to generate more single cell data so that we can further refine our understanding of inner ear cells.
Showing the most recent 10 out of 51 publications
