Vertebrates have evolved a unique way to sense their environment: placodallyderived sense organs. These sensory structures emerge from a crescent-shaped domain, the preplacodal domain, which surrounds the anterior neural plate and generates the paired sense organs as well as the cranial ganglia. For decades, embryologists have attempted to determine the tissue interactions required for induction of various placodal tissues. More recently, technological advances have allowed investigators to ask probing questions about the molecular nature of placodal development. In this dissertation I largely focus on development of the otic placode. I utilize loss-of-function techniques available in the zebrafish model system to demonstrate that two members of the fibroblast growth factors family of secreted ligands, Fgf3 and Fgf8, are redundantly required for otic placode induction. I go on to show that these factors are expressed in periotic tissues from the beginning of gastrulation. These findings are consistent with a model where Fgf3 and Fgf8 signal to preotic tissue to induce otic-specific gene expression. This model does not address other potential inducers in otic induction. A study using chick explant cultures suggests that a member of the Wnt family of secreted ligands also has a role in otic induction. I therefore test the relative roles of Wnt and Fgf in otic placode induction. The results demonstrate that Wnt functions primarily to correctly position the Fgf expression domain and that it is these Fgf factors which are directly received by future otic cells. Lastly, I examine the function of the muscle segment homeobox (msx) gene family expressed in the preplacodal domain. This study demonstrates that Msx proteins refine the boundary between the preplacodal domain and the neural plate. Further, msx genes function in the differentiation and survival of posterior placodal tissues (including the otic field), neural crest and dorsal neural cell types. Loss of Msx function results in precocious cell death and morphogenesis defects which may reflect perturbed BMP signaling.

The inner ear is a complex sensory organ of interconnected chambers, each with a sensory epithelium comprised of hair cells and support cells for detection of sound and motion. This dissertation focuses on the development and regulation of sensory epithelia in zebrafish and utilizes loss of function, gain of function and laser ablation techniques. Hair cells and support cells develop from an equivalence group specified by proneural genes encoding bHLH transcription factors. The vertebrate Atoh1 bHLH transciption factor is a potential candidate for this role. However, data in mouse has led some researchers to conclude it does not have a proneural activity, but, rather, is involved in later stages of hair cell differentiation. In addition, the factors regulating Atoh1 are mostly unknown. We address these issues in zebrafish and show that the zebrafish homologs atoh1a and atoh1b are required during two developmental phases, first in the preotic placode and later in the otic vesicle. They interact with the Notch pathway and are necessary and sufficient for specification of sensory epithelia. Our data confirm atoh1 genes have proneural function. We also go on to show Atoh1 works in a complex network of factors, Pax2/5/8, Sox2, Fgf and Notch. Misexpression of atoh1 alters axial patterning and leads to expanded sensory epithelia, which is enhanced by misexpression of either fgf8 or sox2. Lastly, we examine the role of sox2 in sensory epithelia development and regeneration. Sox2 has been implicated in maintainence of pluripotent stem cells as well as cell differentiation. In the inner ear, Sox2 is initially expressed in the prosensory domain and is required for its formation. Eventually, Sox2 is downregulated in hair cells and maintained in support cells; however, its later role has not been determined. We show that in the zebrafish inner ear, sox2 is expressed after sensory epithelium development has begun and, like in mouse, expression is down regulated in hair cells and maintained in support cells. Our data demonstrate a role for sox2 in maintenance of hair cells and in transdifferentation of support cells into hair cells after laser ablation. Additionally, sox2 is regulated by Aoth1a/1b, Fgf, and Notch.

Vertebrates have evolved a unique way to sense their environment: placodallyderived sense organs. These sensory structures emerge from a crescent-shaped domain, the preplacodal domain, which surrounds the anterior neural plate and generates the paired sense organs as well as the cranial ganglia. For decades, embryologists have attempted to determine the tissue interactions required for induction of various placodal tissues. More recently, technological advances have allowed investigators to ask probing questions about the molecular nature of placodal development. In this dissertation I largely focus on development of the otic placode. I utilize loss-of-function techniques available in the zebrafish model system to demonstrate that two members of the fibroblast growth factors family of secreted ligands, Fgf3 and Fgf8, are redundantly required for otic placode induction. I go on to show that these factors are expressed in periotic tissues from the beginning of gastrulation. These findings are consistent with a model where Fgf3 and Fgf8 signal to preotic tissue to induce otic-specific gene expression. This model does not address other potential inducers in otic induction. A study using chick explant cultures suggests that a member of the Wnt family of secreted ligands also has a role in otic induction. I therefore test the relative roles of Wnt and Fgf in otic placode induction. The results demonstrate that Wnt functions primarily to correctly position the Fgf expression domain and that it is these Fgf factors which are directly received by future otic cells. Lastly, I examine the function of the muscle segment homeobox (msx) gene family expressed in the preplacodal domain. This study demonstrates that Msx proteins refine the boundary between the preplacodal domain and the neural plate. Further, msx genes function in the differentiation and survival of posterior placodal tissues (including the otic field), neural crest and dorsal neural cell types. Loss of Msx function results in precocious cell death and morphogenesis defects which may reflect perturbed BMP signaling.

<p>The semicircular canals of the vestibular system detect angular head rotations and play a fundamental role in guiding motor reflexes during locomotor behaviors. While extensive research has documented the relationship between the semicircular canal shape (i.e. radius of curvature and canal length) and locomotor behaviors, levels of intraspecific variation in primates are relatively unknown. Predictive models using these metrics to reconstruct locomotion in extinct animals are generally based on one individual per species. Furthermore, the influence of body size and to a lesser degree brain size heavily influences overall canal morphology.</p><p>This study documents intraspecific variation in the size, shape and orientation of the semicircular canals in relation to changes in function, brain size, and body size via analysis of high resolution CT scans of large samples of extant primate species. I test the hypothesis that the extent of intraspecific variation differs across a sample of primates, reflecting the intensity of selective pressure on canal shape in species that require agility during locomotion. I also examine whether spatial constraints resulting from the size of the skull (reflected by the size of the brain) affect canal radii of curvature and canal orthogonality more strongly than observed agility during locomotion. </p><p>To this end, data was gathered from high-resolution CT images of museum specimens. For the comparative analysis, 14-matched pairs of adult extant primate species were selected that contrast in agility and brain size in closely related genera. CT images of these specimens were used to measure functional measures of canal sensitivity (e.g., canal radii of curvature, orthogonality). This data was used to test hypotheses concerning intraspecific and interspecific variation in semicircular canal functional morphology. This data was then combined with a larger mammalian dataset culled from the literature, to further test hypotheses relating to body-size and brain size dependent variation in individual canal metrics. </p><p>Evaluation of levels of intraspecific variation support the hypothesis put forth by Billet et al. (2012), that selection on canal morphology is relaxed in animals with slow locomotor behaviors, who are observed to have higher levels of intraspecific variation. Analyses of interspecific variation provides tentative support for the use of canal orthogonality in reconstructive models, most especially in canal angles that seem least effected by other constraints—brain size, etc. However, locomotor signals are complex and brain/skull interactions can potentially produce misleading results when reconstructing locomotor behaviors. This work highlights the importance of critically assessing comparative groups used for inferring behaviors in both extinct and extant animals.</p> / Dissertation

Early prosensory specification to develop competence in the otic epithelium is disrupted by mutations of Eya1, Pax2, Sox2, Jag1, and Gata3. Mutations in these genes apparently disrupt sensory competence and may affect Atoh1 upregulation, a gene known to be necessary for sensory cell differentiation within the ear. How these genes interact with each other and other factors within the genetic network of the ear to refine and restrict sensory specification and impart competence to the developing organ of Corti is not known. These genes also interact with other factors expressed adjacent to or within the developing organ of Corti and provide the context to allow prosensory cells, after cell cycle exit, to appropriately respond to Atoh1 expression and differentiate as hair cells. Gata3 is expressed throughout the early placode. As ear development continues Gata3 is restricted to all prosensory areas except that of the saccule. In addition, it is expressed in a subset of delaminating neuroblasts. Gata3 continues to be highly expressed in the cochlear sensory epithelia as cells differentiate, and is expressed in all cells of the organ of Corti through adult. The human disorder caused by haploinsufficiency of Gata3 is known as Hypoparathyriodism, Deafness, and Renal dysplasia syndrome, and has been linked in mice to early hair cell death. I investigated the role of Gata3 in cochlear neurosensory specification utilizing a mouse Gata3 knockout model and a conditionally deleted Gata3 line combined with two cre driver lines (Foxg1cre and Pax2cre). Although both cre lines are expressed in the inner ear with only a slight difference in onset of expression there are major phenotypic differences. While the Foxg1cre:Gata3f/f deletion resulted in an ear closely matching that of the null mutant with a cochlear duct devoid of neurosensory cells, the Pax2cre:Gata3f/f cochlear duct contained patches of partially differentiated hair cells. Through the use of qRT-PCR and in situ hybridization of both mutants I was able to paint a picture of how Gata3 interacts with other prosensory genes to upregulate downstream genes. In particular, Atoh1, was downregulated but not absent with the loss of Gata3. Indicating that Gata3 is one of a set of factors necessary for the proper upregulation of Atoh1 in the cochlea.

<p>Proper development of the inner ear, including the cochlea, is necessary for normal hearing. Development of the inner ear requires many signaling molecules under both spatial and temporal control. These signaling molecules include the wingless-related integration site (Wnts) and the fibroblast growth factors (Fgfs) gene families. The embryonic chick inner ear was chosen as the model to study cochlear development due to its homology with the mammalian cochlea and the ease of access to the inner ear in ovo. Both the mammalian cochlea and the homologous chick basilar papilla contain two domains with their own type of hair cells and innervation. The neural side of the basilar papilla contains the tall hair cells innervated by the afferent axons which takes the noise signal to the brain. The abneural side of the basilar papilla contains the short hair cells innervated by the efferent axons which receive signals from the brain to turn down added gain.</p> <p>Previous research showed that virally induced cWnt9a overexpression within the basilar papilla generated a neural side phenotype across the basilar papilla (Munnamalai et al., 2017). These basilar papillas contained more tall hair cells and increased innervation at embryonic day 18 (E18) than their wild-type counterparts. Additionally, there were many differentially expressed genes found to be downstream of cWnt9a including cFgf3 and cFgf19. This project focused on determining the role of cFgf19 in inducing a neural side phenotype in the basilar papilla. First, in situ hybridization was used to determine the cFgf3 and cFgf19 mRNA transcript location with cWnt9a overexpression. Both Fgfs were found across the basilar papilla. Next, a possible cWnt9a receptor, cFzd4, which was upregulated with cWnt9a overexpression, was found in the neural side of the basilar papilla. cFgf19 was then overexpressed using one of two different vectors: RCAS(A)/EGFP-P2A-Fgf19 or RCAS(B)/Fgf19-P2A-EGFP in which the order of cFgf19 transcription was altered. RCAS(B)/Fgf19-P2A-EGFP was found to produce less GFP when transfected into DF-1 cells than RCAS(A)/EGFP-P2A-Fgf19. Additionally, RCAS(B)/Fgf19-P2A-EGFP transfected cells produced secreted fusion proteins of GFP and Fgf19, compared to RCAS(A)/EGFP-P2A-Fgf19 transfected cells which produced secreted individual proteins. The viruses were injected into the otocyst at E3 and the embryos harvested several days later including at E6, E10, and E14. Inner ears injected with either virus showed no changes in innervation, hair cells, proliferation, cartilage formation around the cochlear duct, cFgf3 expression, or phosphorylation of ERK. To determine understand where Fgf19 could be producing an effect, the location of a possible receptor, Fgfr4, was determined in wild-type embryos. At E6 and E8, cFgfr4 was found within the basilar papilla, but many more transcripts were found surrounding the cochlear duct. Overall, the role of Fgf19 in neural side fate of the basilar papilla was not determined. Possible reasons for the lack of phenotypic changes include nonfunctional Fgf19 being secreted which could not bind and induce downstream signaling, Fgf19 being responsible for an untested aspect of the cWnt9a overexpression model, or other misregulated genes would be needed for the phenotypic change to occur.</p>

Select epithelial regions of the bovine inner ear were established and maintained in cell culture. Marginal cells from the stria vascularis and dark cells from the posterior wall of the utricle were isolated, dissociated and placed in culture medium. Within 24 h, cellular islands of hexagonal-shaped, epithelial-like cells from both the stria vascularis and posterior utricular wall were readily identifiable by inverted light microscopy. Ultrastructural examination of both the cultured stria marginal cells and utricular dark cells revealed that both cell types had numerous microvilli on their apical surfaces and interdigitating infoldings of their basolateral surfaces. Apical tight junctional complexes were present between apposing cells. These findings demonstrate that inner ear bovine epithelial cells can be successfully isolated and maintained in culture, and that such cells retain certain of their in vivo morphological characteristics.

Indiana University-Purdue University Indianapolis (IUPUI) / Hearing loss and vestibular dysfunction are inner ear disease states that arise from an array of diverse etiologies that interfere with mechanosensory hair cell function, including: congenital syndromes, noise-induced trauma, ototoxic drugs, and aging. The investigation of normal inner ear development and the pathological aberrations that cause inner ear disease has been previously advanced through formation of an easily generated, scalable, accurate in vitro model system that readily facilitates experimental applications. This model utilizes a 3D floating cell culture protocol which guides differentiation of stem cell aggregates into inner ear organoids, which are vesicles containing a sensory epithelium with functioning mechanosensory hair cells. Inner ear organoid formation enables studying the effects of modulating the signaling pathways that guide developing inner ear structure and function. The Notch signaling pathway heavily influences the formation of the inner ear through two major mechanisms: lateral induction of sensory progenitor cells and lateral inhibition to determine which of those progenitors differentiate into mechanosensory hair cells. The effects of inhibiting Notch signaling within the inner ear organoid system were explored through application of the ɣ-secretase inhibitor MDL28170 (MDL) at a concentration of 25μM on day 8 of organoid culture. Aggregates were harvested on day 32, fixed, sectioned, and stained according to a standard immunohistochemistry protocol. Sections were stained for the mechanosensory hair cell markers Myosin7a (Myo7a) and Sox2. MDL-treated aggregates demonstrated statistically significant reductions in the total number of vesicles and the number of vesicles containing hair cells compared to control aggregates. In contrast to control aggregates which demonstrated two distinct organoid variants (protruding and embedded), MDL-treated aggregates only formed the embedded variant. Differences in the expression pattern of Sox2, which is also a marker of stemness and neural progenitor cells were also noted between the two conditions. MDL-treated aggregates demonstrated regions of ‘ectopic’ Sox2 expression whereas Sox2 expression in control aggregates was consistently expressed within Myo7a+ regions.

Indiana University-Purdue University Indianapolis (IUPUI) / Loss of the finite cochlear hair cells of the inner ear results in sensorineural deafness. Human cochlear hair cells do not regenerate, and there is no cure for deafness. Our laboratory has established a three-dimensional culture system for deriving functional sensory hair cells from human pluripotent stem cells. A major limitation of this approach is that derived hair cells exhibit a morphological and gene expression phenotype reflective of native vestibular hair cells. Previous studies have shown that establishment of localized domains of gene expression along the dorso-ventral axis of the developing otic vesicle is necessary for proper morphogenesis of both auditory and vestibular inner ear structures. Sonic hedgehog (SHH) signaling has been shown to play a key role in specification of the ventral otic vesicle and subsequent cochlear development. Here, SHH treatment was pursued as a potential strategy for inducing a patterning phenotype permissive to cochlear induction in vitro. Single-cell RNAsequencing analysis revealed that while treatment with the SHH pathway agonist Purmorphamine reduced expression of markers for the vestibular-yielding dorsal otic vesicle, upregulation of ventral otic marker genes was modest. More strikingly, the number of otic progenitors exhibiting a neuroprogenitor phenotype increased in response to Purmorphamine treatment. These results suggest that SHH pathway modulation in early-stage inner ear organoids may bias their differentiation toward a neural lineage at the expense of an epithelial lineage. The present study is the first to evaluate the patterning phenotype of human stem cell derived otic progenitors, and sheds light on the transcriptomic profile at this critical point of inner ear development. This study may also cultivate future efforts to derive cochlear cell types as well as inner ear neural cell types from human pluripotent stem cells, and contribute to the establishment of a more complete in vitro model of inner ear development. / 2021-08-21

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第21295号 / 医博第4384号 / 新制||医||1030(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 影山 龍一郎, 教授 浅野 雅秀, 教授 辻川 明孝 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DAM