BMP - a Key Signaling Molecule in Specification and Morphogenesis of Sensory Structures

BMP - a key signaling molecule in specification and morphogenesis of sensory structures

Vijay Kumar Jidigam

Umeå center for molecular medicine Umeå 2016

Responsible publisher under swedish law: the Dean of the Medical Faculty This work is protected by the Swedish Copyright Legislation (Act 1960:729) ISBN: 978-91-7601-468-4 ISSN: 0346-6612 Cover page: Confocal image of elecrtroprated olfactory invaginated placode showing GFP+ cells in green and nuclei stained with DAPI in grey. Elektronisk version tillgänglig på http://umu.diva-portal.org/ Tryck/Printed by: Print & Media Umeå, Sweden, 2016

To Kamla & Vaikan

Even the average person can accomplish a task if it were after his or her heart. But the intelligent ones are those who can convert every work into one that suits their taste. Vivekananda

Table of Contents

Table of Contents i Abstract iii Papers included in this thesis iv Abbreviations v Introduction 1 Cranial sensory placodes and their derivatives 1 Placode precursors in the neural plate border 2 Placode precursors are derived from a common pre-placodal ectoderm 4 Individual placode identity 7 Lens placode 7 Specification of lens placodal cells 8 Olfactory placode 10 Neurogenesis in the olfactory epithelium 11 Otic placode 12 Placode Invagination 14 Ectodermal thickening (placode) 14 Retina 16 Aims 19 Materials and methods 20 Functional experiments: 20 Electroporation 20 Chick explant assay 20 Whole embryo culture 21 Results 23 Paper I: BMP-Induced L-Maf Regulates Subsequent BMP-Independent Differentiation of Primary Lens Fibre Cells 23 Paper II: Neural retina identity is specified by lens-derived BMP signals 27 Paper III: Apical constriction and epithelial invagination are regulated by BMP activity 33 Discussion 39 Paper I: Lens 39 Paper II: Retina 42 Paper III: Placode invagination 46 Conclusions 54 Acknowledgements 55 References 58

Abstract

Cranial placodes are transient thickenings of the vertebrate embryonic head ectoderm that will give rise to sensory (olfactory, lens, and otic) and non-sensory (hypophyseal) components of the peripheral nervous system (PNS). In most vertebrate embryos, these four sensory placodes undergo invagination. Epithelial invagination is a morphological process in which flat cell sheets transform into three-dimensional structures, like an epithelial pit/cup. The process of invagination is crucial during development as it plays an important role for the formation of the lens, inner ear, nasal cavity, and adenohypophysis. Using the chick as a model system the following questions were addressed. What signals are involved in placode invagination? Is there any common regulatory molecular mechanism for all sensory placode invagination, or is it controlled by unique molecular codes for each individual placode? Are placode invagination and acquisition of placode-specific identities two independent developmental processes or coupled together? To address this we used in vivo assays like electroporation and whole embryo culture. Our in vivo results provide evidence that RhoA and F-actin rearrangements, apical constriction, cell elongation and epithelial invagination are regulated by a common BMP (Bone morphogenetic protein) dependent molecular mechanism. In addition, our results show that epithelial invagination and acquisition of placode-specific identities are two independent developmental processes.

BMP signals have been shown to be essential for lens development and patterning of the retina. However, the spatial and temporal requirement of BMP activity during early events of lens development has remained elusive. Moreover, when and how retinal cells are specified, and whether the lens plays any role for the early development of the retina is not completely known. To address these questions, we have used gain- and loss-of-function analyses in chick explant and intact embryo assays. Here, we show that during lens development BMP activity is both required and sufficient to induce the lens specific marker, L-Maf. After L-Maf upregulation the cells are no longer dependent on BMP signaling for the next step of fiber cell differentiation, which is characterized by up-regulation of δ-crystallin expression. Regarding the specification of retinal cells our results provide evidence that at blastula stages, BMP signals inhibit the acquisition of eye-field character. Furthermore, from optic vesicle stages, BMP signals emanating from the lens are essential for maintaining eye-field identity, inhibiting telencephalic character and inducing neural retina cells.

iii

Papers included in this thesis

This thesis consists of following papers

I. Pandit, T., V. K. Jidigam, Gunhaga, L. (2011). "BMP-induced L-Maf regulates subsequent BMP-independent differentiation of primary lens fibre cells." Developmental dynamics. 240(8):1917-28.

II. Pandit, T., Jidigam, V. K., Patthey, C., Gunhaga, L. (2015). "Neural retina identity is specified by lens-derived BMP signals." Development (Cambridge, England) 142(10): 1850-1859.

III. Jidigam, V.K., Srinivasan, R. C., Patthey C., Gunhaga L. (2015). "Apical constriction and epithelial invagination are regulated by BMP activity." Biology Open: bio. 015263.

Abbreviations CNS-Central nervous system PNS-Peripheral nervous system BMP-Bone morphogenetic protein FGF-Fibroblast growth factor PPE-pre-placodal ectoderm EMT-Epithelial mesenchymal transition Six- sine oculis homeobox Eya-eye absent Dach-Dachshund family transcription factor St-Stage Otx2-orthodenticle homeobox2 PMLC-phosphorylated myosin light chain ROCK-Rho associated coiled-coil containing protein kinases RhoA-Ras homolog gene family A Maf-musculoaponeurotic fibro sarcoma oncogene homolog Irx3- Iroquois homeobox gene Pax- Paired box Rx- Retinal anterior neural fold Sox-SRY related HMG box Dlx- Distal-less homeobox Gbx2-Gastrulation brain homeobox 2 OV-optic vesicle OVL-optic vesicle lens LR-lens retina RPE-retinal pigmented epithelium GFP-green fluorescent protein Ngn-neurogenin Hes5-Hairy and enhancer of split5 MSX- muscle segment homeobox gene Id3-inhibitor of DNA binding 3 SoHO1-sensory organ homeobox protein IKNM-interkinetic nuclear migration N-CAM-Neural cell adhesion molecule Mitf- micropthalmia associated transcription factor FoxG1-Forkhead boxG1 Smad-Sma-small body size, Mad-mother against decapentaplegic Vsx2-visual system homeobox 2 Emx2-Empty spiracles homeobox 2 Gastrula stage= stage 4 Blastula stage=stage 2 Neural fold stage= stage 8 Neural tube stage=around stage 11

Introduction

During early development, the vertebrate ectoderm becomes regionalized into neural, neural plate border and epidermal domains. The neural plate border region is located at the junction between the neural and epidermal ectoderm, and the derivatives of this intermediate domain include cranial sensory placodes and neural crest cells (Patthey, Edlund et al. 2009). The cranial placodes will give rise to several structures with various cell types of functional diversity, including our sensory organs. Although placode-derived structures are different in their functional properties, they do share certain morphological similarities during early development like columnar epithelia, invagination and delamination of neuroblasts (Begbie and Graham 2001). The focus of my thesis has been i) to understand at what stage and how the lens becomes independent of its initial inductive signals, ii) to understand when and how the retinal cells are specified, and iii) to identify mechanisms of placode invagination, i.e. whether it is regulated by common molecular machinery or is controlled by unique molecular codes for each individual placode.

Cranial sensory placodes and their derivatives

The cranial sensory placodes give rise to several important sensory structures in the vertebrate head (Streit 2004; Jidigam and Gunhaga 2013; Saint-Jeannet and Moody 2014). The most anterior placodes consist of the single midline adenohypophyseal and the two-sided olfactory and lens placodes. The hypophyseal placode gives rise to Rathke’s pouch, a dorsal out-pocketing in the roof of the oral ectoderm, which is located ventral to the hypothalamus. Within the adenohypophysis several hormone secreting cell types are generated that regulate gonadal, thyroid and adrenal functions, growth and metabolism (Baker and Bronner-Fraser 2001; Jidigam and Gunhaga 2013; Saint-Jeannet and Moody 2014). Lateral to the adenohypophyseal placode and close to the ventral telencephalon, the paired olfactory placodes develop and invaginate as pits. The olfactory placode will give rise to sensory epithelium and respiratory epithelium, and the latter eventually lines part of the nasal cavity. Respiratory epithelial cells are non-neural and mucus-generating cells that eliminate

dust particles from inhaled air. The sensory epithelium gives rise to several cell types including glial cells, horizontal cells, sustentacular cells, and globose cells as well as primary sensory neurons that transduce odour signals via axons of the first cranial nerve to the olfactory bulb in the forebrain. In some animals a separate domain of the olfactory placode contributes to the development of the vomeronasal organ, which is specific for pheromone detection (Akutsu, Takada et al. 1992; Norgren and Gao 1994). The lens placode is the only sensory placode that will not give rise to any neuronal cell types. Like other anterior sensory placodes, lens placodes also invaginate and later form a vesicle in response to signals within the placodal ectoderm and the adjacent growing optic cup. The anterior layer of the lens vesicle comprises lens epithelial cells that provide homeostatic support for the lens and act as stem cells to produce new fiber cells. The posterior layer of the lens vesicle differentiates into highly specialized lens fibers filled with crystallin proteins, which makes the organ transparent with the proper curvature. The lens is responsible for projecting the visual image on the central retina, and it is therefore important for vision (Simpson, Moss et al. 1995).

The more posterior located otic placode is formed at the level of rhombomere 5/6 of the hindbrain (Layer and Alber 1990). As the tissue develops, the otic placode invaginates and forms into a cup like structure, and the gradually invaginated cup separates from the overlying surface ectoderm and develops into an otic vesicle. The otic tissue undergoes complex morphogenetic movements to form the entire inner ear including supporting cells like mechanosensory hair cells that transmit balance and auditory information. The otic placode also gives rise to the neurons of the vestibuloacoustic ganglion of cranial nerve VIII (Fekete 1999; Brigande, Iten et al. 2000; Baker and Bronner-Fraser 2001).

Placode precursors in the neural plate border

Several signaling molecules have been shown to be vital for placode induction including members of the fibroblast growth factor (FGF), bone morphogenetic protein (BMP) and Wnt families, as well as sonic hedgehog (Shh), and Retinoic acid. Of these, two stand out (BMP and FGF) for having roles in the progression of

diverse placodes (Faber, Dimanlig et al. 2001; Sjodal, Edlund et al. 2007; Pandit, Jidigam et al. 2011). BMP signals are involved in the specification of olfactory and lens placodal cells at the gastrula stage in chick embryos. In vitro experiments from rostral neural plate border explants from the late gastrula stage (stage 4) show that short time exposure to BMP promotes olfactory identity and prolonged exposure to BMP signaling promotes lens character (Sjodal, Edlund et al. 2007; Saint-Jeannet and Moody 2014). Olfactory/lens placodal explants isolated from gastrula stage embryo confirm that the balance of both FGF and BMP signals is important for the generation of placodal cells, while FGF prevents from acquiring epidermal cell fate and BMP from acquiring neural fate (Sjodal, Edlund et al. 2007).

Figure 1: Schematic picture of a St 7 chicken embryo showing rostral placode territory in purple color and OEPD domain in the caudal region, indicated by red. OEPD, otic epibranchia l placode domain.

An interesting part of the caudal placode field is that at stage 6 otic and epibranchial placodal cells are intermingled in a single domain (Fig. 1), and both placodal cells require FGF signaling for their induction. Thus, although initially both require FGF signaling, other factors combine and give unique fate (identity) to intermingled placode precursors. One such factor is Wnt, which is inductive for otic placode and repressive for epibranchial placode (Ohyama, Mohamed et al. 2006; Freter, Muta et al. 2008).

Placode precursors are derived from a common pre- placodal ectoderm

Patterning events during gastrula and early neurula stages lead to the subdivision of the ectoderm into at least four distinct domains – neural plate, neural crest, pre- placodal ectoderm (PPE), and future epidermis. During the gastrula stage, a specific gene expression profile indicative of the PPE has not been observed (Bailey and Streit 2005). But by the early neurula stage, the PPE domain for the first time is detected by gene expression patterns (Streit 2002; Bhattacharyya, Bailey et al. 2004). The PPE is the precursor region of the placodes, which is identified as a U- shaped domain restricted to the rostral border of the neural plate (Fig. 2) (Schlosser 2010). Within this domain, precursors for different placodes are intermingled, although there is a rough subdivision along the rostral caudal axis into precursors that contribute to more rostral and caudal placodes, respectively (Streit 2002; Bhattacharyya, Bailey et al. 2004). The rostral part of the domain is divided into adenohypophyseal, olfactory and lens placodal precursors. The caudal part of the domain is further divided into trigeminal, otic and epibranchial placodes. At this stage, the rostral and caudal domains are made evident by differential expression of distinct transcription factor encoding genes like Pax6, Six3 and Otx2 in the rostral domain, and Pax2, Irx3 and Gbx2 in the caudal domain (Schlosser 2006; Patthey and Gunhaga 2011). Otx2 and Gbx2 mutually repress each other (Katahira, Sato et al. 2000). Otx2 target activation is required for olfactory and lens placodal precursors, whereas Gbx2 is required for otic placode specification (Steventon, Mayor et al. 2012). In the later stages of the embryonic development, the PPE successively separates into discrete patches of thickened ectoderm called placodes.

Figure 2: At 0-1 somite stage precursor region for all the placodes (pre-placodal ectoderm) is identified as a U-shaped domain detected by expression of Six1 and Eya genes, shown as blue region.

Numerous scientific studies have shown that the induction of sensory placodes at appropriate positions around the developing neural tube is a multi-step process that involves inductive signals from more than one tissue and the sequential or parallel action of several signaling pathways (Baker and Bronner-Fraser 2001; Streit 2004; Litsiou, Hanson et al. 2005). Attenuation of BMP and Wnt, together with activation of FGF signals are required for induction and positioning of the PPE (Litsiou, Hanson et al. 2005; Bailey, Bhattacharyya et al. 2006; McCabe and Bronner-Fraser 2009). Alterations in these signaling molecules will change the location of the neural plate border and further affect the development of the pre-placodal region.

On the other hand, Six, Eya and Dach families are generalized preplacodal markers commonly expressed both in the rostral and caudal placodal domain (Bailey and Streit 2005). These genes are known to be involved in preplacodal formation and in the control of several aspects of sensory organ development. In chick, the head mesoderm below the preplacodal region expresses FGF together with inhibitors of BMP and Wnt signals, which is sufficient and required for inducing preplacodal characteristic markers (Litsiou, Hanson et al. 2005). These factors act to guard preplacodal cells from antagonistic effects emanating from surrounding tissues. Mesoderm is not the only source of signals to induce an ectopic PPE (Seifert, Jacob et al. 1993). Experiments in Xenopus and chick support the idea that neural plate grafts induce Six1 and Xiro1 expression in the ventral epidermis (Glavic, Honoré et al. 2004; Bailey and Streit 2005; Litsiou, Hanson et al. 2005). It is also shown that the rostral region of the PPE is in close contact with underlying endoderm and the

prechordal plate (Seifert, Jacob et al. 1993), which might also play a role in the formation of the olfactory placode and to lesser extent for the induction of the lens (reviewed in Bailey and Streit 2005). With this it is clear that the induction of PPE is a multi-step process which is tightly involved with regulation of other derivatives from mesoderm, ectoderm and endoderm.

Even though the pre-placodal region displays rostral-caudal domains at the neurula stage, individualized placode identity will become more defined as neurulation proceeds. For example, Six1 and Eya2 expressed throughout the pre-placode region at the early neurula stage are down-regulated in the lens region but are maintained in the adenohypophseal and olfactory as neurulation progresses (Bailey and Streit 2005; Schlosser 2006). Moreover, Pax6 expressed in the entire rostral pre-placode region is strongly up-regulated in the lens and olfactory ectoderm and reduced from stage E4.5 of the chick adenohypophyseal region (Li, Yang et al. 1994; Bhattacharyya, Bailey et al. 2004). On the other hand, Pax2 becomes restricted to the otic and epibranchial placodes, and Pax3 to the prospective trigeminal ectoderm (Groves and Bronner-Fraser 2000; Bailey and Streit 2005).

Placode precursors are dispersed in such a way that is comparatively colinear with their final location (Kozlowski, Murakami et al. 1997; Streit 2002; Bhattacharyya, Bailey et al. 2004; Xu, Dude et al. 2008). Developmental studies on rostral placodes and caudal placodes using focal dyes and fate maps suggest that adjacent ectodermal placodal precursors can contribute to diverse placodes (Streit 2002; Bhattacharyya, Bailey et al. 2004; Bhattacharyya and Bronner-Fraser 2004; Xu, Dude et al. 2008). Thus, placodal precursors have been shown to overlap in fate map experiments. For example, around stage 6 prospective olfactory and lens precursors intermingle in the anterior border region (Bhattacharyya, Bailey et al. 2004; Sjodal, Edlund et al. 2007). In a similar manner, the trigeminal, otic and epibranchial precursors intermingle in the posterior border region at stage 5 (Streit 2002; Freter, Muta et al. 2008; Xu, Dude et al. 2008).

Individual placode identity

Lens placode

The vertebrate eye develops with synchronized communications between the neuroectoderm in the forebrain and the surface ectoderm (lens ectoderm). The neuroectoderm associated with the ventral forebrain evaginates, which further results in the formation of bilateral optic vesicles. The distal part of the optic vesicle will make contact with the overlying lens ectoderm, which is then thickened to form the lens placode (Fig. 3B) (Fuhrmann 2010). The lens placode later invaginates to form the lens pit (Fig. 3C). Successively the lens pit further deepens and the contact with the overlying surface ectoderm or the prospective cornea is abolished and ensues in the formation of the lens vesicle (Fig. 3D) (Gunhaga 2011). Until the formation of the lens vesicle the majority of the ectodermal lens cells proliferate. After this stage, the posterior-most cells in the vesicle exit the cell cycle and elongate to become primary lens fibre cells, whereas anteriorly situated cells form the lens epithelium and retain their high proliferative capacity. Eventually the rate of cell proliferation declines and becomes restricted to the equatorial zone, a small region of epithelial cells above the lens equator, which will remain mitotically active throughout life (Jarrin, Pandit et al. 2012).

Figure 3: Morphological changes occur during the early development of lens and retina during chick embryonic development. (A) Around neural tube stage (stage 11) prospective lens ectoderm (LE) in green comes in contact with evaginating optic vesicle in light red color. Once the optic vesicle comes in contact with the lens ectoderm, the ectoderm thickens to form the lens placode (LP) around stage 13. (B, C) From stage 13, the lens placode starts to invaginate

simultaneously with the underlying optic vesicle. (D) By stage 16, the lens ectoderm forms into a vesicle and the underlying optic vesicle forms into a bilayered optic cup. The inner layer of the optic cup becomes the neural retina and the outer layer forms the RPE. Lens cells in the posterior part close to the optic cup start elongating and form the primary fibre cells. MC, mesenchymal cells; RPE, retinal pigmented epithelium; NR, neural retina; LV, lens vesicle.

Specification of lens placodal cells

Rostral neural plate border cells are specified as rostral placodal progenitor cells, and they maintain the ability to acquire either lens or olfactory placodal fate (Fig. 1). Using explant assays it was shown that lens placodal cells are initially specified at the late gastrula stage, and that BMP signals play a key role in this process at least partly by inhibiting neural fate (Sjodal, Edlund et al. 2007). These results also provided evidence that beyond this stage lens placodal cells do not require any interactions with adjacent epidermal or neural ectoderm or underlying mesoderm; this proposes that maintenance and maturation of the lens placodal cells is dependent on signals from the prospective placode ectoderm itself (Sjodal, Edlund et al. 2007; Patthey, Gunhaga et al. 2008). At gastrula stages, Wnt activity in the caudal region inhibits lens and olfactory placodal specification while promoting generation of neural crest cells (Litsiou, Hanson et al. 2005; Patthey, Gunhaga et al. 2008; Gunhaga 2011). Slightly later at the neural fold stage prospective lens placodal explants cultured with premigratory or migratory neural crest cells fail to acquire lens character. Moreover, partial ablation of prospective neural crest cells from the forebrain to midbrain levels at the neural fold stage resulted in occasional ectopic lens formation, suggesting a lens-inhibitory activity in the neural crest that restricts lens formation to forebrain levels (Bailey, Bhattacharyya et al. 2006).

At the neural fold stage, pSmad1/5/8, a downstream transducer of BMP signaling is detected in the prospective lens ectoderm. During this stage prospective lens cells can switch between olfactory and lens placodal fate upon modulation of the levels of BMP signaling, suggesting that at the neural fold stage BMP signals promote generation of lens cells at the expense of olfactory placodal cells (Fig. 4) (Sjodal, Edlund et al. 2007). At later stages, Bmp7 is expressed in the lens ectoderm and optic vesicle, while Bmp4 is expressed in the optic vesicle (Dudley and Robertson 1997; Wawersik, Purcell et al. 1999). Bmp4 mouse mutant embryos display failed

lens placode formation, but the formation of olfactory placode appears to be normal. Although, Bmp4 knockout mice embryos lack a lens placode, the expression of Pax6 and Six3 is still maintained in prospective lens ectoderm, which is indicative of the presence of lens precursor cells (Furuta and Hogan 1998). Similar to Bmp4 mouse mutant, Bmp7 mutant embryos show disturbed lens placode invagination. Furthermore, deletion of genes encoding the type I BMP receptors (Bmpr1a and Acvr1) in mouse embryos prevented lens formation. But the deletion of BMP canonical transducers (Smad4, Smad1 and Smad5) have shown no effect on lens specific markers and lens development (Rajagopal, Huang et al. 2009). This demonstrates that both Bmp4, Bmp7 together with their type I receptors are required for the development of the lens (Dudley and Robertson 1997; Wawersik, Purcell et al. 1999). Further studies are needed to determine the downstream signaling of the BMP pathway for lens development.

Other signaling molecules like FGF have also been shown to play an important role during the development of the lens. In support, several FGF family members are expressed in the eye region and all four FGF receptors are expressed in the lens (Robinson 2006). As mentioned earlier, at the late gastrula stage FGF activity prevents prospective lens/olfactory cells in the rostral neural plate border from acquiring epidermal fate (Sjodal, Edlund et al. 2007). In addition, mis-expression of Fgf8 in the competent surface ectoderm at the neural tube stage inhibited the generation of epidermal cells, thereby enabling BMP signals to ectopically induce lens character as indicated by L-Maf expression (Kurose, Okamoto et al. 2005). Moreover, mouse eye explants from E8.5 to 9.5 cultured with an FGF inhibitor exhibited reduced Pax6 expression with significant size reduction in the lens placode and delayed placode invagination. In the mouse, conditional deletion of FGFR1/2/3 during the lens pit stage exhibited normal invagination with the expression of lens fiber cell maker α- and β-crystallins (Faber, Dimanlig et al. 2001; Zhao, Yang et al. 2008). A similar phenotype is also observed in mice mutants lacking Frs2α, an anchor protein mediating FGF signaling via the ERK pathway (Gotoh, Ito et al. 2004). Despite the fact that initial differentiation of the primary lens fiber cell and expression of the differentiation marker δ-crystallin are independent of FGF activity, further differentiation of lens fiber cells appears to require FGF activity (Lorén,

Schrader et al. 2009). Together, these results indicate that BMP signals together with FGF activity regulate the specification and development of lens placodal cells.

Olfactory placode

The olfactory placode lies in proximity to the ventral telencephalon, and gives rise to both sensory neural and respiratory non-neural epithelium (Maier, von Hofsten et al. 2010). As mentioned in the previous sections, the prospective olfactory and lens placodal cells are intermingled at the rostral border (Figure 1) (Bhattacharyya, Bailey et al. 2004). Consistently, explants cultured from this rostral border region generate cells of both lens and olfactory character in distinct domains (Sjodal, Edlund et al. 2007). At the neural fold stages, rostral placodal cells can change between lens and olfactory identities in response to alterations in BMP activity, in which BMP activity promotes lens character at the expense of olfactory placodal cells (Sjodal, Edlund et al. 2007). However, at this stage FGF activity is not sufficient to induce olfactory identity in lens progenitor cells, and inhibition of FGF activity does not prevent generation of olfactory cells (Bailey, Bhattacharyya et al. 2006; Sjodal, Edlund et al. 2007). In spite of this, FGF signaling is necessary for generation of olfactory placode derived neurons, in part by inhibiting the generation of epidermal cells (Sjodal, Edlund et al. 2007).

Figure 4: Schematic diagram of a St 8 (neural fold stage) chick embryo. At this stage prospective olfactory and lens cells are spatially segregated within the rostral placodal region, as indicated in green and blue, respectively. Within the caudal placodal region, the otic placode is indicated in orange. ANR, anterior neural ridge.

At stage 14, Fgf8 expression is detected in the anterior part of the olfactory placode and Bmp4 expression is identified in the posterior part of the placode. At around

stage 20, cells in the anterior portion of the olfactory pit express Fgf8, whereas weak Bmp4 expression can be detected at the olfactory rim. In the center part of the olfactory pit, cells at the rim express Fgf8 and Bmp4, and in the posterior part of the olfactory pit cells express Bmp4 with weak Fgf8 expression. The activated BMP transducer, pSmad1/5/8, is detected in more or less every cell in the posterior part and in scattered cells in the anteromedial part of the olfactory pit (Maier, von Hofsten et al. 2010).

The olfactory placode is first morphologically evident in chick around stage 14 (Bhattacharyya, Bailey et al. 2004; Maier and Gunhaga 2009). During olfactory placode patterning the specification of neurogenic versus non-neurogenic domains is a critical step that involves many signaling molecules and gene networks (Maier, Saxena et al. 2014). In chick and mouse, BMP signals promote respiratory epithelial cells, and FGF signaling is required for the generation of sensory epithelial cells (Maier, von Hofsten et al. 2010). Previous studies reveal that the spatiotemporal expressions of several transcription factors like Hes5, Dlx3, Dlx5, Sox2, Pax6, Msx1/2, and Id3 are important for olfactory placode formation (Cau, Gradwohl et al. 2000; Bhattacharyya, Bailey et al. 2004; Maier and Gunhaga 2009; Maier, von Hofsten et al. 2010). Hes5 and HuC/D are expressed in the sensory part of the epithelium, and Id3 and Msx1/2 are expressed in future respiratory epithelial cells. Together these genes determine the neurogenic and non-neurogenic regions of the olfactory placode (Maier, von Hofsten et al. 2010).

Neurogenesis in the olfactory epithelium

In the olfactory epithelium neurogenesis occurs throughout life and it is further categorized into two phases: 1) Primary neurogenesis occurs during embryonic development and 2) secondary neurogenesis is the homeostatic replacement of neurons in the adult to maintain the matured pseudostratified epithelium (Maier, Saxena et al. 2014). Primary neurogenesis begins with placode formation at around stage 14 in chick (Maier and Gunhaga 2009). As mentioned, primary neurogenesis forms the essential structure of the olfactory epithelium from which the axons of olfactory sensory neurons extend to the most rostral part of the telencephalon where

the future olfactory bulb forms (Treloar, Miller et al. 2010; Maier, Saxena et al. 2014)). During initial establishment of the olfactory placode, Hes5 is expressed in a cluster of neural progenitor cells and it is implicated in defining the neurogenic domain (Cau, Gradwohl et al. 2000). Ngn1 is expressed in immediate neuronal precursors in the basal region of the placode and it is also detected in the first early migrating neurons. In mice lacking Ngn1 (Ngn1−/−), basal progenitors fail to express several neuronal differentiation-marker genes including NeuroD. Furthermore, Ngn1 is required in the basal progenitors for the activation of the differentiation program for the olfactory sensory neurons (Cau, Casarosa et al. 2002; Maier and Gunhaga 2009).

Otic placode

Adjacent to the hindbrain region at the 4-6 somites stage, otic placode induction takes place (Jidigam and Gunhaga 2013). The otic placode gives rise to the inner ear and the eighth cranial nerve. The inner ear is a complex structure that is composed of several cell types including ciliated mechanoreceptors that detect sound. The signals are transduced to the central nervous system via the neurons of the eighth cranial nerve, the cochleovestibular nerve.

Several FGF family members are expressed in the developing inner ear and in adjacent tissues, and they regulate otic induction, neuronal production, and morphogenesis (Ladher, Anakwe et al. 2000; Alsina, Abelló et al. 2004). From the 8 somites stage on, Fgf10 is expressed in the anterior region, while Bmp7 expression is detected throughout the posterior region of the otic epithelium (Abello, Khatri et al. 2010).

Pax2 is one of the earliest known markers expressed during otic induction, with an onset at the neural fold stage (Christophorou, Mende et al. 2010). The Pax2 expressing domain of the surface ectoderm contains progenitors for both otic and epibranchial placodal cells (Groves and Bronner-Fraser 2000; Streit 2002; Streit 2004). This Pax2 positive domain has been referred to as the otic and epibranchial placode domain (OEPD) (Fig.1), the posterior placodal area (PPA), and the pre otic field.

Several studies have shown that members of the FGF family are required for the induction of the otic and epibranchial placode domains (Schimmang 2007). In chick, around stage 5, Fgf19 and Fgf3 are expressed in the mesoderm underlying the otic pre-placodal region, together with Fgf8 expressed in the endoderm (Ladher, Anakwe et al. 2000; Schimmang 2007). At the neural fold stage, this expression is expanded to the hindbrain adjacent to the developing otic placode. Although knockdown of Fgf19 did not show any significant effect, OEPD induction was blocked by inhibiting both Fgf19 and Fgf3 signals (Freter, Muta et al. 2008). Expression of Fgf10 (in mouse) and Fgf19 (in chick) seems to be under the control of Fgf8. That is why in the chick Fgf8 alone seems to be sufficient for otic induction (Ladher, Wright et al. 2005). By the neural tube stage (st10/11), Fgf10 and Fgf8 are expressed within the chick otic placode (Ladher, Anakwe et al. 2000; Karabagli, Karabagli et al. 2002; Schimmang 2007). During this stage, FGF family members are expressed in and around the developing otic placode; this suggests that FGF signals play a significant role in otic placode formation (Schimmang 2007). Cranial paraxial mesoderm positive for Fgf19 expression is essential for avian otic placode induction; by ablating the paraxial mesoderm the otic placode fails to develop (Kil, Streit et al. 2005). Conversely, overexpression of FGF signaling molecules like Fgf3 near the otic vesicle induced ectopic vesicle-like structures expressing otic related markers (Vendrell, Carnicero et al. 2000). In addition, placing FGF2 and FGF8 beads close to the otic region induces ectopic expression of otic placode markers like SOHo1 and Pax2. Taken together, it is clear that FGF signaling plays an important role for otic induction and further for otic placode formation.

Wnt signaling is also known to play an important role in otic induction (Ladher, Anakwe et al. 2000). Wnt activity in combination with FGF signals induce otic character, and excludes the epibranchial identity (Freter, Muta et al. 2008). In an explant assay, exposing the competent non-neural ectoderm to mesodermal signal FGF19 together with Wnt8 is sufficient to induce otic markers (Ladher, Anakwe et al. 2000). Consistent with sequential steps for otic induction, inhibiting Wnt signaling with Dkk1 blocked the expression of SOHo1 (inner ear marker), but did not inhibit the early otic marker Pax2 (Freter, Muta et al. 2008; Sai and Ladher 2015).

Placode Invagination

Ectodermal thickening (placode)

Thickening of the placodes occurs at different stages for different cranial placodes. Before the invagination process, cells in the placode elongate along the apical basal axis. In chick, the olfactory placode is shaped at stage 14, the lens at stage 13, and the otic placode is formed at stage 8/9. However, molecular mechanisms regulating the thickening of the placode epithelium have received less attention. It has been shown in the otic placode that Pax2 is a potential candidate for placode thickening (Christophorou, Mende et al. 2010), but detailed mechanisms for the formation of placodes are still unclear. However, a study in the mouse paves the path for investigating the members of the small GTPases family as potential candidates, which suggests that Rac1 is a main effector for cell elongation in the lens placode (Chauhan, Lou et al. 2011). Further studies are needed to determine if this small GTPases is essential for the thickening in the otic and olfactory placodes. Cells within the placode undergo cell division, through a process called interkinetic nuclear migration (IKNM), which permits the epithelial cells within the placode to be tightly packed for a higher number of cells (Sauer 1936). The strong association of placode formation and morphogenesis supports the idea that epithelial thickening might be essential for epithelial bending or placode invagination (Chauhan, Plageman et al. 2015).

Following thickening, placodes invaginate to form structures called pits or sacs. In chick, the initial invagination of the olfactory placode starts at stage 17, the lens at stage 14, and the otic placode at stage 10/11 (Jidigam and Gunhaga 2013). During these early stages of development, the morphological process of the lens epithelium is different when compared to olfactory and otic placode morphogenesis. Once the lens pit deepens, the lens epithelium fuses with itself, and the connection with the overlying surface ectoderm is broken. In all three placodes, transition from the placode to pit structure involves the cells in the epithelium to undergo a biphasic process – phase one being basal cell expansion and phase two apical constriction (reduced apical cell surface) (Sai and Ladher 2008; Christophorou, Mende et al. 2010; Plageman, Chung et al. 2010). Numerous well known transcription factors,

cell adhesion molecules and cytoskeleton elements have been connected with the regulation of placodal invagination (Hogan 1999; Jamora and Fuchs 2002), but the signaling molecules regulating them is still not known.

One of the driving forces behind morphogenesis is change of cell shape via rearrangements in the cytoskeleton, i.e. the network of actin, intermediate filaments and microtubules, which are the backbone for the structure of the cell (Lecuit and Lenne 2007). Both apical constriction and placodal invagination are known to be mediated by networks of F-actin, which contract by interaction with myosin II. At stage 10, F-actin is detected in both the apical and basal side of the otic placode, and by stage 11 F-actin is depleted basally and enhanced apically (Sai and Ladher 2008). Blocking myosin II and F-actin polymerization using blebbistatin and cytochalasin D, respectively, results in the failure of placode invagination (Sai and Ladher 2008; Borges, Lamers et al. 2011; Plageman, Chauhan et al. 2011). In addition, accumulation of phosphorylated Myosin and apical constriction requires apical localization of RhoA (Borges, Lamers et al. 2011; Plageman, Chauhan et al. 2011). Rho GTPases are known to act as molecular switches for transfer of information within the cell by cycling between a GTP bound “active state” and GDP bound “inactive state”. Rho GAP and Rho GEF are identified to promote the switching to the active and inactive states of Rho proteins, respectively (Hall 2012). Myosin is directly phosphorylated by ROCK, a downstream molecule of RhoA, and ROCK inhibition disrupts placode invagination (Amano, Ito et al. 1996; Borges, Lamers et al. 2011). Moreover, apical accumulation of RhoA is sufficient to induce apical constriction, and RhoA plays a key role in apical localization of Shroom3 (an actin scaffold protein). Blocking Shroom3 causes reduction in the apical distribution of both F-actin and myosin in the lens placode (Plageman, Chung et al. 2010). In cell culture experiments it is shown that Trio, a RhoA GEF molecule, is required for Shroom3 dependent apical constriction. Together this points to the idea that a Trio, RhoA, Shroom3 and ROCK pathway is required for apical constriction (Plageman, Chauhan et al. 2011). Molecular signals regulating this pathway are still not known.

Several transcription factors have been shown to play substantial roles in the formation and invagination of placodes. Pax2 plays a key role in otic placode

morphogenesis; in the absence of Pax2 cells fail to elongate due to the loss of apically localized N-cadherin and N-CAM. In addition, ectopic Pax2 in surface ectoderm did not promote cell elongation but induced N-Cadherin and N-CAM expression (Christophorou, Mende et al. 2010). On the other hand, Spalt4, a Zinc finger transcription factor that is strongly expressed in the olfactory and otic and weakly expressed in the lens placode, was shown to be involved in the early stages of otic placode development and to initiate cranial ectodermal invagination. Overexpressing Spalt4 in the non-placodal surface ectoderm from stages 8-11 induced ectopic invaginations. On the other hand, it was also shown that implantation of FGF2 beads into the area opaca (extra embryonic ectoderm in chick) induced Spalt4 expression (Barembaum and Bronner-Fraser 2007).

Retina

Once the neural plate is established, the anterior neural domain progressively becomes regionalized into different regions that later generate the telencephalon, eye field (optic), hypothalamus and diencephalon (Puelles and Rubenstein 2003). The eye-field gives rise to most structures of the eye, such as the neural retina, the retinal pigment epithelium (RPE) and the optic stalk. The cellular and molecular mechanisms required for the differential specification of these regions are not well characterized. In chick embryos, fate map studies at the neural plate stages show that prospective eye-field cells are located in a well-defined region in the anterior neural plate (Fernández-Garre, Rodríguez-Gallardo et al. 2002; Sánchez-Arrones, Ferrán et al. 2009). One of the initial markers expressed within the eye field region are the members of Rax/Rx family. At the time of initial specification of the eye field, prospective lens ectoderm and retina are not in close contact, and the initial contact is observed at the neural tube stage. The initial sign of morphogenesis in the eye region is the evagination of the optic vesicle from the ventral forebrain region at the neural tube stage. The neural retina is formed from the distal portion of the optic vesicle and the retinal pigment epithelium from the dorsal part (Kagiyama, Gotouda et al. 2005; Hirashima, Kobayashi et al. 2008).

Eye (retina) formation includes a complex series of morphogenetic events during vertebrate development. The basic morphology of the eye is visible when the optic vesicle evaginates and forms a bilayered optic cup in which the inner epithelium is neural and the outer is pigmented epithelium (Fig. 3C). Eventually, and consistent with a morphogenetic wave of differentiation, the sensory neural retina, ciliary body and iris are committed from this simple structure along the central to peripheral axis. Rax/Rx (homeodomain transcription factor) genes are shown to play vital roles in optic vesicle evagination. In zebrafish embryos lacking Rx3 (an ortholog of Rax1), the evaginated distal portion of the optic vesicle moving outwards is completely disturbed, which indicates the importance of Rx3 in extensive cell movements (Winkler, Loosli et al. 2000; Rembold, Loosli et al. 2006; Stigloher, Ninkovic et al. 2006). In Xenopus, the Rx gene is shown to regulate cell proliferation, and Optx2, which controls proliferation in the eye-field, is known to be dependent on Rx function (Zuber, Perron et al. 1999; Zuber, Gestri et al. 2003).

At the optic vesicle stage, neuroepithelial cells are bipotent, i.e. the presumptive retinal cells are competent to develop into either RPE or neural retinal cells (Rowan, Chen et al. 2004; Horsford, Nguyen et al. 2005; Westenskow, McKean et al. 2010). The earliest genes that are expressed in the retina and RPE domains are the homeobox gene Vsx2 (Chx10) and the bHLH transcription factor Mitf, respectively (Fuhrmann 2010). In the mouse it has been shown that Mitf is expressed in the whole optic vesicle and consequently reduced after Vsx2 expression is up-regulated (Nguyen and Arnheiter 2000). Both genes are needed for the maintenance of retinal cell identity, and in addition Vsx2 controls other aspects of the development like cell proliferation (Sigulinsky, Green et al. 2008). Lhx2 (LIM box transcription factor) is a known retinal patterning gene expressed in the eye-field that is known to be essential for Mitf and other retinal markers (Zuber, Gestri et al. 2003; Yun, Saijoh et al. 2009). It has been suggested that the neuroretinal-inducing signal comes from the surface ectoderm (Hyer, Kuhlman et al. 2003), suggesting a model whereby, signals from the ectoderm promotes the cells in the optic vesicle to become NR rather than RPE when the close contact is established between two tissues. It has been shown that Vsx2 might act directly by suppressing transactivation of the Mitf gene in the distal optic vesicle (Bharti, Liu et al. 2008). Another study has demonstrated that

Chx10 repression of Mitf is required for the maintenance of mammalian neural retina identity (Horsford, Nguyen et al. 2005). However, the signals by which the spatio- temporal expression of these transcription factors is regulated during the early stages of development remain to be described.

Although neural retina and telencephalon both arise within the forebrain, how these two domains are differentially specified is not well defined. Signaling pathways like FGF and BMP were shown to play important roles in several phases of early eye development. Studies in the mouse have shown that FGF signaling is necessary for the maintenance of neural retina characteristics, rather than the specification (Horsford, Nguyen et al. 2005). However, the signals required for the early specification of neural retinal cells are not known. Other Bmp mutant studies in the mouse have shown severe defects in early eye development (Furuta and Hogan 1998; Wawersik, Purcell et al. 1999). The spatio-temporal requirements of BMP signals during the early stages of the eye development are not known. Moreover, the molecular mechanisms that regulate the induction and maintenance of eye-field cells and the specification of neural retina cells are poorly understood.

Aims

The studies described in this thesis are about the requirement of BMP signals for induction and specification of lens and retina cells as well as its importance in the morphogenesis of sensory and non-sensory placode invagination.

Specific aims of this thesis are:

1. To determine at what stages BMP signals are required for initial identity of lens cells. 2. To examine when lens fibre cell differentiation becomes independent of BMP signalling. 3. To analyse the molecular signals involved in the specification of retinal cells, and at what stage retinal cells are specified 4. To determine what signalling molecules are regulating placode invagination. 5. To analyse if placode invagination is regulated by a common molecular machinery, or is controlled by unique molecular codes for each individual placode. 6. To examine whether placodes maintain their identity after the disruption of invagination.

Materials and methods

Functional experiments:

Electroporation

-minus +plus Olfactory

Lens

In ovo electroporation: Lens and olfactory regions of the chick embryo were electroporated in ovo at desired developmental stages to transfer a control green fluorescent protein (GFP) vector, alone or together with a construct of interest. DNA-constructs either to over-activate BMP or to inhibit the BMP signals were used. DNA constructs were transferred using an Electro Square Porator ECM 830 (BTX) by applying 5 pulses (9-15 V, 25 ms duration) at 1-s intervals. The electroporated embryos were allowed to grow to a desired developmental stage. Embryos with GFP expression within the desired region were selected for further analyses. The intact chick embryo assay allows to examine morphological processes and is a valuable tool to strengthen the physiological relevance of in vitro results.

Chick explant assay

Explant assay: Small pieces of tissue of interest (so called explants) are isolated from chick embryos at different developmental stages using tungsten needles. Thereafter the prospective placodal explants are cultured alone, or in the presence of inhibitors or activators of specific signalling pathways. All explants are cultured in vitro in collagen in serum-free conditions.

Whole embryo culture

Incubate at 37 degrees

Stage 5/7 Embryos in +/- Inhibitors

Ex ovo culture of chick embryos: Stage 5-7 embryos were dissected and folded in half along the midline into a semicircle, and cut along the outside edge. The embryos were allowed to rest undisturbed in Ringer’s solution at room temperature (RT) for approximately 45-60 min. Embryos were then transferred to a 4-well plate containing 500µl of medium (a 2:1 mixture of thin albumen and Ringer’s solution with penicillin and streptomycin 2 U/ml) alone or together with inhibitors. Each well contained one to six embryos. Embryos were cultured in the presence of BMP inhibitors Noggin, Dorsomorphin or ROCK inhibitor Y27632.

Evaluation of functional experiments: In both the explant and intact embryo studies, the identity of the cells was analysed by immunohistochemistry and in situ hybridization using specific antibodies and chick dioxigenin-labeled RNA probes, respectively. Changes in cell proliferation and cell death were analysed by using Caspase3, and phospho-HistoneH3. To obtain reliable statistics, explants and embryos were sectioned on consecutive slides and stained in multiple ways. The percentage of antigen-expressing cells was quantified by counting the number of marker-stained cells, and compared to the total number of cells, determined by counting DAPI-stained nuclei. Morphometric analysis: using confocal microscopy

both control and BMP inhibited embryos were analysed. Apical cell width, cell length, basal cell width were studied with the combination of confocal microscopy and Volocity 6.0 software.

Results

Paper I: BMP-Induced L-Maf Regulates Subsequent BMP- Independent Differentiation of Primary Lens Fibre Cells

Localization of BMP signals and lens specific markers in the developing eye region

To determine whether BMP signaling is active in the prospective lens from the optic vesicle stage to the developing optic cup, we examined expression of the potential BMP ligand, Bmp4, and pSmad1/5/8 (an active indicator of BMP signaling), together with lens specific markers L-Maf and δ-crystallin. At stage 11, Bmp4 and pSmad1/5/8 expression were found in the prospective lens ectoderm, but no L-Maf and δ-crystallin expression was detected. By stage 13, initial L-Maf expression was observed in the thickened epithelium (placode) with no δ-crystallin expression. Initial δ-crystallin expression was observed first at stage 17, and by this time no expression for Bmp4 was detected in the lens vesicle. This indicates that BMP signals are initially detected before placode formation and lens specific markers are detected after placode formation.

In in vitro conditions BMP activity is required for the initial expression of L- Maf in lens placode cells

To test if BMP activity is required for the expression of L-Maf and/or δ-crystallin, we dissected out explants containing the lens ectoderm together with optic vesicle tissue at stage 11; at this time point, L-Maf and δ-crystallin were not expressed in this region. These LR (lens/retina) explants were cultured in the presence or absence of Noggin, a BMP antagonist (Lamb, Knecht et al. 1993) for 42h, which in intact embryos corresponds to stage 21 chick embryos. Afterwards, explants were processed for sectioning and analyzed by immunohistochemistry using L-Maf and δ- crystallin markers. Stage 11 control explants generated L-Maf and δ-crystallin cells in the lens domain, characteristic of primary lens fibre cells. Explants cultured with

Noggin failed to generate cells expressing L-Maf and δ-crystallin but instead coexpressed HuC/D, Keratin and pan-Dlx, which was indicative of olfactory cell identity.

In vivo, BMP activity is required for the up-regulation of L-Maf expression

Next we examined the requirement of BMP signaling for L-Maf expression in intact chick embryos. To address this question, stage 11 prospective lens ectoderm was electroporated with GFP alone or together with a BMP antagonist, Noggin (Timmer, Wang et al. 2002), and cultured to stage 21. GFP positive cells within the lens region were selected for future analysis. Control embryos electroporated with only GFP exhibited normal lens morphology with expression of L-Maf and δ-crystallin. In contrast, Noggin treated embryos displayed two phenotypes – severely affected embryos showed complete loss of L-Maf and δ-crystallin with failure of optic cup formation; slightly less affected embryos exhibited a disturbed smaller lens with expression of L-Maf and δ-crystallin, while Keratin was detected in the surface ectoderm in both phenotypes. The severity of the lens phenotype was associated with the electroporation efficiency.

We wanted to better understand whether the mechanism behind the failure in lens formation after BMP inhibition is due to reduced proliferation or increase in cell death. We examined the cell proliferation marker pHH3 and the cell death marker aCaspase3 in control and Noggin treated embryos that were cultured to different time points. Embryos cultured to stage 13/14 did not show any significant changes in cell proliferation in Noggin treated embryos compared to control embryos. In embryos cultured to stage 15 and stage 17, a reduction in pHH3+ cells was observed. At all the stages examined, we did not observe any changes in the levels of aCaspase3+ in the Noggin and control treated embryos. These data indicate that the failure in invagination at stage 15 and stage 17 is partly associated with reduced cell proliferation. This indicates that BMP is required for the formation of the lens and for the up-regulation of L-Maf and δ-crystallin expression.

BMP activity is sufficient to induce L-Maf expression

At gastrula stages, Bmp4 expression was detected in the rostral border ectoderm region from where the lens cells arise. To investigate if BMP signals were sufficient to induce L-Maf expression, stage 4 neural explants were dissected and cultured for 43-45h in the presence or absence of BMP4 (35ng/ml). Neural explants cultured alone generated L5 and Sox1 positive cells, indicative of neural character. By contrast, addition of BMP4 blocked the generation of L5 and Sox1 positive neural cells and induced L-Maf positive lens cells. This suggests that BMP is sufficient to induce L-Maf expression.

Initial elongation of primary fibre cells is independent of BMP activity

We wanted to understand if BMP signals are required directly for δ-crystallin expression or indirectly required via L-Maf induction which in turn promotes δ- crystallin expression. To examine this issue, we dissected out stage 13 LR explants and cultured them alone or in the presence of Noggin to the time point equivalent to stage 21. In control explants both L-Maf and δ-crystallin were detected within the lens domain. Similar results were obtained when stage 13 LR explants were cultured in the presence of Noggin; this is characteristic of primary lens fibre cells. Taken together, after the specification of lens placodal cells or after the onset of L-Maf expression, the up-regulation of δ-crystallin and initial elongation of primary fibre cells was independent of BMP activity.

L-Maf expression is sufficient to induce lens fate independent of BMP signals

Our results provide evidence that after the onset of L-Maf, inhibition of BMP signaling did not prevent early fiber cell differentiation; this suggests that this event in primary fiber cell differentiation is dependent on L-Maf and not on BMP signals. To analyze this further, an L-Maf over-expression vector (Ogino and Yasuda 1998) was electroporated alone or together with Noggin in stage 9/10 olfactory epithelium and surface ectoderm, and cultured to stage 18 and 21. In the olfactory epithelium we detected δ-crystallin expression in the presence or absence of BMP signaling. Also in the head ectoderm L-Maf induced δ-crystallin expression in the absence of

BMP signaling. Taken together these results provide evidence that the up-regulation of δ-crystallin is independent of BMP activity; this implies that L-Maf acts as a regulator of primary lens fibre differentiation.

Paper II: Neural retina identity is specified by lens-derived BMP signals

Characterization of optic vesicle and forebrain markers in the developing chick head

To analyze when the eye-field cells acquire neural retina identity, we examined a set of markers in the different regions of the forebrain during early chick embryonic development from stage 9-21. At stage 9, Rax2 was expressed in cells of the evaginating prospective retina and in the prospective hypothalamus; also FoxG1 was expressed in prospective telencephalic cells. By stage 11, Rax2 expression was limited to the prospective optic vesicle and FoxG1 was detected in the periphery of the optic vesicle and a stronger expression in the telencephalon region. At stage 13, when the lens placode is formed and makes a contact with the underlying optic vesicle, Vsx2 was up-regulated in a region overlapping with that of Rax2, whereas Mitf was only detected in the prospective RPE. At stage 21, Rax2 and Vsx2 were expressed in the neural domain of the optic cup, with weaker expression of Rax2 in the tuberal hypothalamus. Both FoxG1 and Emx2 were expressed in the dorsal telencephalon, and no other regions of the forebrain area were shown to express these markers. Based on these expression patterns, Rax2 and Vsx2 can be used as retinal cell markers and FoxG1 and Emx2 together as dorsal telencephalic markers.

At blastula stages BMP inhibition is required for the generation of eye-field cells

A previous study in zebrafish has shown that BMP signals in the anterior neural ectoderm promote telencephalic identity at the expense of eye-field character (Bielen and Houart 2012). To understand if a similar mechanism was followed in the chick, we dissected out blastula stage medial (M) explants, and cultured them alone or in the presence of Noggin to a time point equivalent to stage 10. Subsequently, explants were then sectioned and processed for in situ hybridization and expression levels of Rax2 and FoxG1 were examined.

Stage 2 medial explants cultured alone showed FoxG1 expression, but no Rax2 expression was detected. This is in agreement with the previous findings that medial explants are specified as dorsal telencephalic character (Patthey, Edlund et al. 2009). In contrast, medial explants cultured in the presence of Noggin generated Rax2 and FoxG1 positive cells. This suggested that eye-field cells are not specified by the blastula stage, and in addition, BMP inhibition at this stage is required for the generation of eye-field cells within the forebrain region.

Prospective retinal cells acquire dorsal telencephalic identity during in vitro culture

To understand if the initial specification of eye-field cells and neural retina cells occur in single or distinct inductive events, we analyzed if the eye field cells can differentiate into Rax2 and Vsx2 positive cells in culture. To address this issue, we isolated stage 9, 10, 11 and 13optic vesicle (OV) and performed analysis at the onset of culture (0h), or after 46-54h in culture, which corresponds to around stage 21 in intact embryos. All explants isolated from stage 9, 10, 11, and 13 and cultured for 0h expressed Rax2 with few cells positive for FoxG1 in the boundary of the explants. But stage 9, 10 and 11 OV explants when cultured longer did not show any Rax2, Vsx2 or Mitf expression but instead generated FoxG1 and Emx2 positive cells, indicative of dorsal telencephalic character. This suggests that stage 9-11 OV explants acquire dorsal telencephalic identity when cultured longer. In contrast, stage 13 OV explants generated Rax2 and Vsx2 positive cells, but no Emx2 was detected when cultured to 38-40h, suggesting neural retinal cells are specified at stage 13. Thus prospective retinal cells require signals from adjacent tissue for maintenance of eye-field character and acquisition of neural retina character at the expense of telencephalic identity.

Stage 9-11 optic vesicle requires lens ectoderm and BMP signals

To determine if the signals emanating from the lens ectoderm are important for specification of neural retina characteristics in the underlying optic vesicle, we dissected out the optic vesicle together with lens (OVL) explants and cultured to a

time point equivalent to stage 21. Under these conditions, δ-crystallin positive lens fiber cells and Rax2 and Vsx2 positive retinal cells were detected in non-overlapping regions. No FoxG1, Emx2 and Mitf were detected in these explants. This suggests that lens ectoderm is sufficient for the optic vesicle to maintain Rax2 expression and further to acquire neural retina identity.

We have shown in our previous publications that Bmp4 is expressed in the prospective lens ectoderm (Pandit, Jidigam et al. 2011), and phosphorylated smad1 is detected in the optic vesicle (Belecky-Adams, Adler et al. 2002). These data suggest that BMP signals from the lens ectoderm is important for the maintenance of Rax2 expression in the optic vesicle and for subsequent generation of neural retina character. To further investigate this, we cultured stage 9/10 OVL explants in the presence of Noggin. After BMP inhibition, the generation of neural retina markers Rax2 and Vsx2 were suppressed and instead FoxG1 and Emx2 expression was induced; this is indicative of telencephalic character. In addition, we have previously shown that BMP inhibition blocked the generation of δ-crystallin positive lens cells (Sjodal, Edlund et al. 2007; Pandit, Jidigam et al. 2011). These results suggest that BMP signals emanating from prospective lens ectoderm are required for specification of neural retinal identity by suppressing dorsal telencephalic identity.

Retinal identity is dependent on BMP signaling emanating from the lens ectoderm

To examine if BMP signals are required for the induction of neural retina cells in ovo, stage 9/10 optic vesicles were electroporated with GFP alone or together with Noggin expressing vector (Timmer, Wang et al. 2002), and cultured to stage 15/16. Embryos with GFP staining within the retina region were selected and further sectioned and processed for in situ hybridization and immuno staining. All the non- electroporated side, as well as control GFP electroporated embryos, displayed normal retina morphology with normal expression of Rax2, Vsx2 and FoxG1. By contrast, optic vesicles electroporated with Noggin exhibited failed invagination and did not form bilayered optic cup. Moreover, expression of the neural retina marker Vsx2 was lost, with down-regulation of Rax2 and expansion of FoxG1 expression.

In, addition, the lens failed to invaginate and did not form a vesicle. This suggests that BMP activity is required for the early development of the neural retina.

To further analyze whether the BMP signals are derived from the lens or from other adjacent tissues, we co-electroporated Noggin together with GFP in the following various locations of the chick head: 1) in the lens ectoderm; 2) in the head ectoderm lateral to the prospective lens ectoderm; and 3) in the ventral midline. When Noggin was electroporated in the lateral surface ectoderm and ventral midline, no difference was observed in the eye morphology or expression of Rax2, Vsx2 and FoxG1 when compared to control embryos. In contrast, when Noggin was electroprated in the prospective lens ectoderm, the optic vesicle failed to invaginate and did not form a bilayered optic cup. In addition, Vsx2 expression was lost, and Rax2 expression was downregulated with no obvious change in FoxG1 expression. Altogether these data suggest that lens derived BMP signals are required for the induction of neural retina cells.

Direct requirement of BMP activity for specification of neural retina

To determine whether neural retina specification requires direct involvement of BMP signaling independent of lens ectoderm, we dissected out stage 13 OV explants that in vitro generate Rax2 and Vsx2 in the absence of lens ectoderm. When such explants were cultured in the presence of Noggin, all exhibited low levels of Rax2 with weak or no Vsx2 expression in half of the explants; in the other half Rax2 and Vsx2 were completely abolished with FoxG1 up-regulated. No Emx2 was detected in stage 13 OV explants when cultured in the presence of Noggin. Altogether this suggests that half of the explants acquired tuberal hypothalamus identity and the other half acquired FoxG1 positive forebrain cells, although distinct from dorsal telencephalic identity. These results indicate that a direct BMP activity is required for the specification of neural retina cells.

Next we analyzed if BMP is sufficient to suppress dorsal telencephalic identity and induce retinal characteristics. To test this, stage 9/10 dorsal telencephalic (dT) explants were dissected out and cultured alone or in the presence of BMP4. Stage

9/10 dT control explants generated FoxG1 and Emx2, indicative of dorsal telencephalic identity, with no Rax2, Vsx2 neural retinal cells and Mitf positive RPE cells. With addition of BMP4 protein to the cultured dT explants, dorsal telencephalic identity was completely inhibited, and instead Rax2 and Vsx2 positive neural retinal cells were induced, even with relatively low concentration of BMP4. However, Mitf positive RPE cells were not detected with lower levels of BMP4. Increased concentration of BMP4 did not induce RPE positive cells. These data show that BMP signals can substitute for the missing lens ectoderm, and provide evidence that low levels of BMP are sufficient to induce neural retina identity.

BMP induces Fgf8 expression in prospective retinal cells

FGF signals have been shown to play a role in promoting the identity of neural retinal cells (Pittack, Grunwald et al. 1997; Hyer, Kuhlman et al. 2003). In chick embryos Fgf8 expression is detected in the medial part of neural retina at stage 13- 14 and it is also detected at stage 21. To determine if the FGF signals act together with BMP activity in specification of neural retinal cells, we dissected out stage 9/10 OVL explants and cultured them alone or in the presence of Noggin. Control explants generated Fgf8 positive cells in the Rax2 positive domain of the explants. By contrast, in the presence of Noggin, Fgf8 expression was inhibited in the prospective retinal cells suggesting that BMP acts upstream of Fgf8.

To assess whether the loss of Rax2 and Vsx2 markers after BMP inhibition is due to the secondary effect of loss of FGF activity, we cultured stage 9/10 OVL explants together with Noggin and FGF8 (250 ng/ml). However, the addition of FGF8 could not rescue retinal identity in the absence of BMP signaling. In addition, when OVL explants were cultured with SU5402, a pharmacological inhibitor for FGF, Rax2 and Vsx2 markers were still maintained in the explants. These results suggest that early FGF signals are not required for the specification of neural retinal identity, and in the absence of BMP signals eye-field cells acquire telencephalic identity.

Wnt alone is not sufficient to specify neural retinal cells, but Wnt together with BMP signals induce RPE cells

Wnt signals have been suggested to be important for the specification of RPE cells (Fujimura, Taketo et al. 2009; Steinfeld, Steinfeld et al. 2013). Whether Wnt signals play a role in the specification of the neural retina has not been determined. To address this, we cultured stage 10 OVL explants and stage 13 OV explants together with a soluble frizzled receptor (Frizzled-conditioned medium) to inhibit Wnt activity (Hsieh, Rattner et al. 1999; Gunhaga, Marklund et al. 2003). Stage 10 OVL and stage 13 OV explants cultured alone or in the presence of Frizzled protein generated Rax2 and Vsx2 positive cells with no FoxG1 expression in stage 10 OVL explants, whereas there were few FoxG1 positive cells in stage 13 OV explants. This suggests that Wnt signals are not required for the specification of neural retinal cells.

To further analyze if the combination of Wnt together with BMP is sufficient to induce neural retina or RPE identity, Stage 9/10 OV explants were cultured with Wnt3A condition medium alone or together with low levels of BMP4. Explants cultured with Wnt3A alone generated FoxG1 and Emx2, indicative of dorsal telencephalic cells, while retinal markers like Rax2 and Vsx2 were not detected. By contrast, the combination of Wnt and BMP activity inhibited the generation of FoxG1 and Emx2 positive dorsal telencephalic cells, but instead induced Mitf positive RPE cells co-expressing with neural retinal markers Rax2 and Vsx2. Thus, Wnt activity is not sufficient to induce neural retina identity, but a combination of Wnt and BMP can induce RPE cells.

Paper III: Apical constriction and epithelial invagination are regulated by BMP activity

Characterization of Bmp expression in placode regions around the time of invagination

To study invagination we first characterized the timing of initial invagination in the olfactory, lens, and otic sensory placodes in the chick. The initial invagination of otic, lens and olfactory begins at stage 10/11, stage 14 and stage 17, respectively. It has been shown that BMP signals are required for the generation of placodes (Brugmann, Pandur et al. 2004; Glavic, Honoré et al. 2004; Sjodal, Edlund et al. 2007; Patthey, Gunhaga et al. 2008; Patthey, Edlund et al. 2009). To further investigate if BMP activity plays a role in placode invagination, we examined the expression pattern of Bmp4, Bmp7, and its transducer pSmad1/5/8 prior to and around the time of placode invagination. At stages 8-10 Bmp4, Bmp7, and pSmad1/5/8 were detected in the otic placode region. At stages 11-14 and stage 14- 17, Bmp4, Bmp7 expression and pSmad1/5/8 were observed in the lens and olfactory region, respectively. Thus, in the olfactory, lens and otic placodes BMP expression together with pSmad1/5/8 is indicative of BMP activity at the time of initial invagination of sensory placodes.

BMP signals are essential for morphogenesis of placode and placode invagination

To investigate whether BMP signals are involved in sensory placode invagination, we performed in vivo experiments by blocking BMP signals and analyzed for the changes in the sensory placode morphology. This was achieved by electroporating a Noggin expressing vector in the stage 10/11 olfactory and lens placode regions. A GFP vector was transfected alone or together with Noggin in the prospective ectodermal region before the onset of invagination. Electroporated embryos were cultured to stage 19/20 for olfactory, and stage 15/16 for lens targeted regions. GFP stainings within the olfactory and lens were further processed for sectioning and analyzed by in situ hybridization and immuno staining.

All control lens and olfactory electroporated embryos showed normal morphology of the lens and olfactory epithelium. By contrast, Noggin electroporated embryos displayed failed thickening of the lens placode and further invagination was suppressed in both lens and olfactory. Depending on the electroporation efficiency, occasionally a small olfactory invagination was detected in the GFP negative region. While the majority of the Noggin electroporated olfactory regions still possessed thickened epithelium, occasionally a few embryos completely lacked thickened epithelium. Using a whole embryo three dimensional OPT imaging technique, the disrupted invagination of the olfactory placode and the lack of pit formation was clearly visible after BMP inhibition as compared to the controls.

We next analyzed whether BMP signals played a similar role in the otic placode invagination as for the olfactory and lens. Electroporating stage 6/7 embryos in the otic region, which lies in close proximity of the heart region, resulted in heart malformation and arrested development, thus preventing analysis of otic placode invagination. To avoid this issue, we used a modified Cornish pasty method (Nagai, Lin et al. 2011) which is another type of ex ovo whole embryo culture technique. For the whole embryos culture, stage 6/7 embryos were dissected and cultured in the control medium alone or together with either Noggin or Dorsomorphin, to approximately stage 12/13. All the embryos cultured in the control medium exhibited normal morphology and underwent placode invagination. When BMP signaling was inhibited by either Noggin or Dorsomorphin, the majority of the embryos displayed normal placode thickening but disrupted placode invagination, while a few embryos displayed complete lack of otic placode formation. Altogether, these data suggest that the process of placode formation in the olfactory and otic placode is BMP independent, while lens placode formation requires BMP signals. But the invagination process for olfactory, otic, and lens all require BMP activity. This suggests that placode formation and invagination are two independent processes, and BMP signals are essential for the invagination process.

Disrupted placode invagination after BMP inhibition is not due to the loss of placode identity

Several studies have shown that BMP signals are important for the specification of lens fibre cells, partly by regulating L-Maf (Sjodal, Edlund et al. 2007; Pandit, Jidigam et al. 2011). This was also confirmed in the current studies. Therefore, we examined if the disrupted placode invagination after BMP inhibition might be a consequence of disturbed acquisition of placode cell identity in the olfactory and otic placodes, or if BMP signals are specifically important for the invagination process. To address this issue, we investigated whether cells in the placode still generated differentiated cell types in the absence of BMP signals by using a set of molecular markers that uniquely define specific placodes.

In general, the olfactory placode gives rise to both sensory and respiratory parts of the olfactory epithelium (Maier, von Hofsten et al. 2010). The cells in the sensory domain express stem-cell like Hes5 positive cells and post-mitotic HuC/D neurons, while cells in the respiratory region express Id3 (Maier, von Hofsten et al. 2010). Moreover, Id3 has been shown to be a direct target gene of BMP signaling, and can be used as a readout of BMP activity (Hollnagel, Oehlmann et al. 1999). Embryos electroporated with GFP alone in the olfactory placodal region expressed Hes5 and HuC/D cells in the sensory domain and Id3 in the respiratory region. In contrast, Noggin electroporation in the olfactory placodal region blocked generation of Id3 positive respiratory cell, which confirmed the suppression of BMP activity; this is in agreement with previous published data (Maier, von Hofsten et al. 2010). Even though BMP inhibition blocked the generation of the respiratory domain, expression of Hes5 and HuC/D characteristic of olfactory sensory identity was maintained.

Control cultured embryos expressed the non-neurogenic markers Irx1 and Lmx1b in the posterior otic placode as expected. In the BMP inhibited embryos, the otic placode failed to invaginate with Lmx1b expression throughout the otic epithelium, while Irx1 positive cells were detected in the posterior part of the otic region. This suggests that at this stage specification of otic placodal cells occur independent of BMP signals. Altogether, these results suggest that cells can differentiate beyond the placode stage and mature into differentiated cell types in the absence of invagination. This proposes that acquisition of placode specific cell types and invagination are two independent processes.

Failure in placode invagination after BMP inhibition is not due to changes in cell death or proliferation

We then examined whether disruption of placode invagination after BMP inhibition is due to the reduction of cell proliferation or increase in cell death. To examine this, cell proliferation marker pHH3 and cell death marker aCaspase3 were analyzed in the control and BMP-inhibited embryos (Noggin electroporated in the lens and olfactory, Dorsomorphin treated embryos for the otic region) that were cultured to different time points. We did not observe any significant changes in the levels of aCaspase3 positive cells in the Noggin electroporated lens and olfactory embryos compared to control embryos. Likewise, no significant changes for aCaspase3 positive cells were observed when otic regions were cultured in the presence or absence of Dorsomorphin. Regarding the proliferative studies, there was no significant change in pHH3/GFP double positive cells in the Noggin electroporated in the lens; similarly no changes in the levels of pHH3 were observed in the otic region treated with Dorsomorphin when compared to controls. In contrast, reduction in pHH3 positive cells was observed in the Noggin electroporated in olfactory region compared to controls. This suggests that reduced cell proliferation in the olfactory region might affect the invagination process, but in general reduced proliferation cannot explain failure in placode invagination after BMP inhibition.

BMP is required for cell shape changes in the prospective placodal cells

Cell shape changes like cell elongation and apical constriction were required for the invagination process. We next analyzed if BMP inhibition prevented any cell shape changes. We measured cell length, apical cell width and basal cell width in control and Noggin treated embryos. The olfactory placode was optimal for these studies, since BMP inhibition disrupted the placode invagination but not the placode formation, and the GFP electroporated cells were easily visualized in the failed olfactory placode. GFP positive cells from the medial invaginated part of the control olfactory were compared to Noggin treated embryos, with a total of 74 control cells and 61 cells from Noggin treated embryos.

BMP inhibition in the olfactory placode significantly increased the cell apical width (average of 2.8µm) compared to the controls (average of 1.2µm). Moreover, Noggin treated cells showed reduction in cell length (33µm compared to control average of 50µm). But no significant difference was observed with cell basal width in the control with an average of 5.5µm and Noggin treated cells with an average of 5µm. This resulted in an apical basal ratio of 0.56 for Noggin electroporated cells compared to 0.22 for the control treated cells. This suggests that variations in cell shape changes after BMP inhibition is one of the reasons for disruption of olfactory placode invagination.

Apical localization of RhoA and F-actin is disrupted after BMP inhibition

Apical constriction and placode invagination have been shown to be mediated by apical localization of RhoA that enables apical accumulation of F-actin, which in turn activates the acto-myosin network (Sai and Ladher 2008; Plageman, Chauhan et al. 2011). Using confocal microscopy, we analyzed if the apical accumulation of RhoA and F-actin were disturbed after the inhibition of BMP in the placode regions. In all the control invaginated placodes for the olfactory, otic and lens, RhoA and F- actin were apically localized. By blocking BMP signals in all the three sensory placodes, apical localization of RhoA and F-actin were disrupted. In the BMP inhibited embryos RhoA expression was completely lost and the cellular localization of F-actin was changed to the boundary of the cells instead of being apically localized as seen in the control embryos. These results provide evidence that BMP signals are required for the apical accumulation of RhoA and F-actin, and disruption of this accumulation is the reason for failed sensory placode invagination.

To determine if this mechanism is common to all placodes, we also analyzed the hypophyseal non-sensory placode by electroporating Noggin construct at stage 12 and cultured to stage 15/16. All the control GFP electroporated embryos exhibited normal placode invagination with apical localization of F-actin, while Noggin treated embryos demonstrated failed placode invagination with apical disruption of F-actin. RhoA was not expressed at the examined stages in the hypophyseal

epithelium. This suggests that BMP signals are required for the apical accumulation of F-actin and epithelial invagination in both sensory and non-sensory placodes.

ROCK activity is required for placode invagination

Rho-associated protein kinase (ROCK), a downstream effector of small GTPase Rho is known to be one of the major regulators of cytoskeleton elements. It is well known that RhoA forms a complex together with Trio, Shroom3 and Rock1/2, and blocking any of the activity disrupts apical constriction and placode invagination (Borges, Lamers et al. 2011; Plageman, Chauhan et al. 2011; Sai, Yonemura et al. 2014). To assess the importance of ROCK activity in all three sensory placodes, we used whole embryo culture together with Y27632, a ROCK inhibitor. During the control conditions the lens, otic, and olfactory showed normal apical localization of RhoA and F-actin. By contrast, embryos treated with ROCK inhibitor exhibited failed placode invagination in a similar manner as the embryos treated with BMP inhibition. Moreover, using confocal microscopy we were able to show that accumulation of RhoA was reduced or lost, and F-actin was located at the periphery of the cells instead of being apically localized as seen in the controls. This suggests that blocking ROCK activity results in the failure of placode invagination, a phenotype similar to that of BMP inhibited sensory placodes. Altogether these results indicate that BMP signals act upstream of ROCK during apical constriction and placode invagination.

Discussion

Paper I: Lens

Several studies in the chick and mouse have shown the BMP activity is required for the normal development of lens. Bmp4 or Bmp7 knockout mice revealed that BMP signaling is essential for lens induction, but the cellular mechanisms underlying this process are not well studied (Furuta and Hogan 1998; Wawersik, Purcell et al. 1999). Another study in mice demonstrates, inactivation of two type I BMP receptors (Bmpr1a and Acvr1) act redundantly in a smad independent pathway during lens formation (Rajagopal, Huang et al. 2009). Most of the BMP mice mutant studies have shown its requirement for the normal development of eye, but the spatio-temporal requirement of BMP signaling during lens development is not addressed. Our results now extend that knowledge and show that prior to lens placode formation, BMP signals are important for the initial induction of L-Maf, which acts as a key regulator for lens fiber cell identity and is also known to be a marker for early specification of the lens placode. In the absence of BMP signals, lens cells switch to olfactory fate characterized by co-expression of Keratin, HuC/D and Dlx. Furthermore, our results show that once L-Maf is up-regulated around stage 13 (lens placode stage), further lens development becomes independent of BMP signaling. At these stages, BMP inhibition did not exhibit any loss of the lens specific markers L-Maf and δ-crystallin. This defined time window of BMP dependent and BMP independent mechanisms, before and after placode formation is very crucial for the early development of the lens.

Previous studies have shown that L-Maf is required for the up-regulation of δ- crystallin which regulates cell cycle exit and lens fiber cell differentiation by activating p27kip1 (Reza, Urano et al. 2007). Moreover, it was shown that L-Maf expression is capable of ectopically inducing Prox1 and Crystallin genes in ovo, while dominant negative L-Maf represses these genes with complete loss of lens and optic vesicle invagination (Ogino and Yasuda 1998; Reza, Ogino et al. 2002; Reza, Urano et al. 2007). It was previously shown that L-Maf expression is dependent on

Pax6. Although Pax6 alone ectopically could not induce L-Maf expression, addition of Sox2 is sufficient to induce L-Maf (Reza, Ogino et al. 2002). We show that in the presence of L-Maf, δ-crystallin is upregulated with initial elongation of fiber cells. In our study we show that the upregulation of L-Maf is BMP dependent, and subsequent δ-crystallin upregulation with initial elongation of fiber cell is BMP independent.

Using mouse fibroblast cells it was shown that c-Maf is a potential candidate target of Pax6 (Sakai, Serria et al. 2001). c-Maf is a member of large Mafs, which were shown to be involved in lens fiber cell differentiation in the mouse (Kawauchi, Takahashi et al. 1999). Moreover, c-Maf knockout studies in mice have displayed reduced crystallin expression (Kim, Li et al. 1999). Although c-Maf and L-Maf have similar functions, they are not functionally identical since L-Maf plays an important role in lens placode formation in the chick, whereas c-Maf does not play a similar role in the mouse. Our study extends the knowledge that BMP directly or indirectly regulates L-Maf expression.

Otx2 and Pax6 are shown to maintain the competence of the surface ectoderm to respond to the signals from the optic vesicles (Zygar, Cook et al. 1998; Ashery- Padan, Marquardt et al. 2000). Many studies have shown that Pax6 plays a cell- autonomous role in surface ectoderm and it is required to respond to the lens inductive signal. Conditional deletion of Pax6 in the prospective lens-forming ectoderm prevents lens development (Ashery-Padan, Marquardt et al. 2000). One study of Pax6 conditional deletion in surface ectoderm prevented placode formation with substantial decrease in ECM (extra cellular matrix) between the optic vesicle and ectoderm (Huang, Rajagopal et al. 2011). One more study with placodal deletion of Sox2, another transcription factor involved in lens development, showed milder phenotypic changes with reduced crystallin expression in the invaginated lens placode (Smith, Miller et al. 2009). In mice lacking Bmp4 normal Pax6 expression was shown with reduced levels of Sox2. Moreover in the Bmp7 mutants, the expression of Pax6 is reduced with loss of Sox2 expression (Furuta and Hogan 1998; Wawersik, Purcell et al. 1999). Somewhat similar to that, our studies show that BMP inhibition has severe consequences for lens development in addition to the

loss of L-Maf and δ-crystallin expression, although expression of Pax6 and Sox2 was not affected after BMP inhibition. One possible explanation is that Sox2 and Pax6 have become independent of BMP signals at the time BMP was inhibited in the chicken experiments. Alternatively, Sox2 and Pax6 expression is not directly dependent on BMP, but the observed reduction in the mouse mutants might be because the germline deletion of Bmp gene also impairs the earlier development of ectodermal tissues (Rajagopal, Huang et al. 2009).

From our in vitro explant assay, it is evident that in the absence of BMP signals prospective lens ectodermal cells acquire olfactory cell fate. This is in agreement with previous studies at the neural fold stage, showing that continuous exposure of olfactory placodal cells to BMP signals promotes the generation of lens cells (Sjodal, Edlund et al. 2007). Studies in the chick spinal cord have shown that ectopic activation of Bmp signaling induced expression of MafB (Chizhikov and Millen 2004) indicating that large Maf genes are regulated by a common molecular mechanism. Altogether it can be suggested that Pax6 might play a potential role in the regulation of ECM deposition, as well as in the morphogenesis of lens ectoderm to come in close contact with optic vesicle which is crucial for further induction of lens placode, whereas BMP plays a potential role in cytoskeleton reorganization together with induction of lens placodal specific genes (Fig. 5). This proposes that both Pax6 and Bmp4 appear to be regulated and to function independently of each other during lens induction and both are required for proper lens development.

Fig. 5: BMP signals are required for induction of L-Maf. At the optic vesicle stage when the lens ectoderm (purple) is located close to the optic vesicle (green), BMP signals within the lens ectoderm are required for induction of L-Maf in response to inductive signals from the optic vesicle. Furthermore, upregulation of δ-crystallin is dependent on L-Maf. BMP signals are required for apical accumulation of RhoA and F-actin. Either BMP within the lens ectoderm is directly required for L-Maf induction or the signals from the optic vesicle induce it.

Paper II: Retina

The mechanism by which the eye field is specified within the anterior neural plate in early embryonic development is poorly described. Simultaneously, the mechanism by which the neural retina becomes regionalized from the forebrain region is not well defined. Our study extends the knowledge of the molecular mechanisms for the early specification of retinal cells and at the same time we provide evidence that signals from the lens ectoderm are necessary for the specification and maintenance of retinal identity. Briefly, our results provide evidence that at blastula stages the presence of BMP signals in the anterior neural ectoderm represses eye field characteristics by promoting subsequent acquisition of forebrain character. Conversely, at the optic vesicle stage lens derived BMP signals are important for the differential specification of the retina within the forebrain region, by inhibiting telencephalic identity. Furthermore, our results provide evidence that retinal cells are specified at around stage 13.

Using chick embryos, it has been shown that removal of prospective lens ectoderm at stage 12, resulted in optic vesicle formation but further failure to form an invaginated optic cup. In contrast, surgical removal of the lens placode at stage 13 did not show any adverse effects on optic cup formation or any effects on neural retinal identity. This indicates that prospective lens ectoderm is necessary for optic vesicle invagination (Hyer, Kuhlman et al. 2003). Although this study defined the time window for the optic vesicle invagination, specific signals or molecules important for the specification and maintenance of neural retina were not mentioned. In mouse, lens specific deletion of Pax6 was shown to allow induction of prospective lens ectodermal character and neural retina identity, but further development of the lens is arrested (Ashery-Padan, Marquardt et al. 2000). Another study in the chick showed that blocking of L-Maf and Pax6 function using dominant negative constructs in the lens ectoderm prevented both lens placode formation and optic vesicle invagination (Reza, Ogino et al. 2002). These studies however, do not provide evidence on the developmental windows at which these gene products are required for the specification and/or maintenance of neural retina identity. Our study provides evidence that prior to stage 13, BMP signals from the prospective lens

ectoderm are crucial for the specification of prospective retinal cells as indicated by Rax2 and Vsx2 expression; after stage 13, retinal cells develop independent of signals from the lens placode.

Our in vitro studies show that blocking BMP signals at stage 9/10 inhibited the generation of retinal cells, and up-regulated telencephalic markers. But the cell fate switching from retinal cell to telencephalic identity is only partially observed by blocking BMP signals at stage 13. In vivo studies exhibited similar results. Upon reduction of BMP signals cells express reduced levels of Rax2 with complete loss of the neural retinal marker Vsx2, and further expansion of FoxG1 expression in the optic vesicle. However, we do not see any switch in cell identity suggesting that in intact embryos signals from surrounding tissues like ventral midline or cephalic mesenchyme might compensate for the effect of reduced BMP signaling. This indicates that BMP signals are required for the specification of Vsx2+ neural retinal cells; in absence of BMP signals, cells acquire telencephalic character. From our study we showed that BMP signals are required for specification of neural retina and there have been studies in mouse indicating that BMP4 and BMP7 are essential for dorso-ventral patterning of the retina (Morcillo, Martínez-Morales et al. 2006; Yun, Saijoh et al. 2009). Our present studies as well as previous studies from the mouse indicate that BMP signals are essential for specification and/or maintenance of neural retinal cells along with patterning of the retina.

Another study in the mouse with conditional deletion of Lhx2 showed a severe phenotype with failed optic vesicle invagination towards the lens ectoderm, and further loss of retinal specific markers (Yun, Saijoh et al. 2009; Hägglund, Dahl et al. 2011). Rescue experiments in Lhx2 mutants have proven that the treatment of Lhx2 mutant explants with exogenously added BMP recovered some aspects of eye phenotype (Yun, Saijoh et al. 2009). Although Pax6 and Lhx2 mutants show similar morphology, molecularly they are distinct. In the absence of the Pax6, the eye field is still maintained through the optic vesicle stage as indicated by Rx, Six3, Pax2 and neural retinal marker Vsx2 and RPE marker Mitf (Bäumer, Marquardt et al. 2003). In Lhx2 mutants the eye field is still maintained with complete loss of neural retinal identity shown by Vsx2 marker. The mutants maintain optic vesicle formation

because of the Rx expression and this gene is known to be essential for the optic vesicle morphogenesis (Mathers, Grinberg et al. 1997). Our results after BMP inhibition show a similar kind of phenotype with failed optic cup formation. This is in agreement with a study by Murali et al, which showed that homozygous inactivation of two type I BMP receptors resulted in down-regulation of Vsx2 expression and failed optic cup formation. The morphological effect could be because of loss of the Vsx2 (Chx10) marker which was shown to have important functions for retinal development, such as regulation of proliferation (Green, Stubbs et al. 2003). Reduced Rax2 expression with loss of proliferative Vsx2 positive cells resulted in decreased dorsal ventral expansion which in turn had an impact on the invagination of the optic cup. In addition, it was shown that BMP signaling is also implicated in the regulation of apoptosis in chick and mouse eye development (Trousse, Esteve et al. 2001; Morcillo, Martínez-Morales et al. 2006). Further studies are needed to determine if reduction in cell death has any effect on the morphology. In contrast, using bead assays it was shown that ectopic BMP4 expression inhibited invagination of optic vesicle but not lens development (Hyer, Kuhlman et al. 2003). Our in vitro experiments with dorsal telencephalic and optic vesicle explants demonstrate that cells switch identity when cultured in the presence of low levels of BMP. However, Hyer et al, did not examine the requirement of BMP signals for the optic vesicle. We show that modulating the levels of BMP signals has adverse effects on the development of the retina, suggesting that a balance of BMP signals is required for the proper optic vesicle invagination. Another study in the mouse demonstrates that the cells within the optic vesicle change shape intensely, which is further accompanied by the transient alteration of the basal lamina composition (Svoboda and O'Shea 1987). Further studies are needed to determine if BMP signals play a role in the deposition of laminin during the optic vesicle to optic cup formation. As the main focus of our paper was on specification and maintenance of the eye field, a brief background is given about the possibilities of morphological difference caused after BMP inhibition.

Studies from 3D stem cell differentiation have shown that embryonic stem cells can either spontaneously acquire self-directed organization of the complex optic-cup and neural retina morphology (Eiraku, Takata et al. 2011) or dorsal telencephalic

identity (Lancaster, Renner et al. 2013). The difference between these behaviors has remained mysterious. Our study provides evidence that the absence or presence of low levels of BMP might switch cell identity between retina and telencephalon.

Paper III: Placode invagination

Our study provides mechanistic insight into cellular mechanisms that drive the morphogenetic process of the flat ectodermal sheet into invaginated cup like structures. We advocate that one essential module of morphogenetic cell behavior called placode invagination is not coupled with the acquisition of placode specific identity in three sensory (olfactory, otic and lens) placodes. Moreover, based on the studies from both sensory and non-sensory ectodermal placodes, our data show that the regulatory mechanisms involved in the placode bending are common for all the ectodermal placodes. Also, BMP signaling plays a central role in the sensory and non-sensory placode invagination. Our results provide evidence that disturbed placode invagination after BMP inhibition is not due to loss of placode identity, cell death or reduced proliferation, but mainly due to failed apical accumulation of RhoA and F-actin. Furthermore, our study shows that inhibition of ROCK, a molecule known to be essential for invagination of the otic placode, results in flat placode ectoderm.

Our expression analysis in the chick shows that the BMP ligands and its transducer pSmad1/5/8 are detected before placode formation and around the onset of placode invagination. In the present studies we provide evidence that BMP is required for the normal development of sensory placodes (olfactory, otic and lens) and one non- sensory placode (hypophyseal). Embryos in which BMP signals are blocked exhibit failure of placode invagination as a result of disrupted apical accumulation of RhoA and F-actin, which leads to a lack of apical constriction and reduced cell length. This is in agreement with previous observations in chick that defective BMP signaling also results in changes in olfactory and lens placode invagination (Maier, von Hofsten et al. 2010; Pandit, Jidigam et al. 2015). Consistently, both Bmp4 and Bmp7 germline mice mutants demonstrate failed lens placode formation and subsequent invagination (Furuta and Hogan 1998; Wawersik, Purcell et al. 1999). In addition, conditional deletion of Bmpr1a and Acvr1 receptors demonstrated defects in lens placode formation (Rajagopal, Huang et al. 2009).

Besides BMP, FGF signaling was shown to be important for placode invagination. Previous studies from chick and mouse embryos showed that both olfactory and otic placodes fail to invaginate after blocking FGF signaling (Zelarayan, Vendrell et al. 2007; Sai and Ladher 2008; Maier, von Hofsten et al. 2010). For otic morphogenesis, FGF signaling was shown to activate myosin II through PLCγ- mediated cascade which is essential for apical actin enrichment (Sai and Ladher 2008). Invagination in lens, olfactory and otic appears to be driven by both intrinsic and extrinsic forces; it is still controversial as to which force dominates. Conversely, whether these two processes are coordinated together is still unclear. Extrinsic forces originate not from the placode itself, but from the surrounding epidermis and mesenchyme. Another known extrinsic factor is the Rho family GTPase Cdc42, the activity of which is vital for the formation of filopodia (Nobes and Hall 1999; Ridley 2006). Conditional deletion of Cdc42 mutant mice resulted in reduced lens pit invagination (Chauhan, Disanza et al. 2009). In contrast, intrinsic forces generate through the apical constriction and cell elongation, but it is still a question if apical constriction and cell elongation rely only on intrinsic forces. Our studies from the three sensory placodes show that BMP signals regulate the process called apical constriction and this process is accompanied by actin-myosin cytoskeletal changes, and is supposedly a process of intrinsic forces.

BMP is required for apical accumulation of RhoA and actin-myosin contractility

Our results provide evidence that loss of BMP activity in prospective placodal regions of the head ectoderm disrupts accumulation of RhoA and F-actin at the apical side of the placodal cells as well as for subsequent apical constriction. It has already been shown that RhoA plays a central role in apical constriction, whereas another small GTPase, Rac, is essential for cell height. RhoA mouse mutant embryos show reduced lens pit curvature with increased cell length. Rac1 mutants show increased apical constriction reduced cell length in the lens placode (Chauhan, Lou et al. 2011). Based on this, it has been suggested that the RhoA mutant is indicative of Rac1 pathway gain of function, and the Rac1 mutant is indicative of RhoA pathway gain of function (Chauhan, Plageman et al. 2015). Similar mechanisms

have been shown in multinucleated giant cells (MNGC) in that the level of microtubule polymerization inversely regulates Rho and Rac activities. Depolymerization of microtubules leads to Rho activation by blocking Rac activity, whereas repolymerization of microtubule induces Rac activation by inhibiting Rho activity (Ory, Destaing et al. 2002). From our present studies, BMP inhibition in the lens and olfactory placodes cause defects of both apical constriction and cell height which suggests that BMP signaling might be upstream of both Rho and Rac signaling.

In one study, blocking MID (a microtubule associated protein) in neuroepithelial cells of Xenopus embryos caused failure of apical constriction and cell elongation, which involves both actin and microtubules (Suzuki, Hara et al. 2010). Similar to the role of Rac1, microtubules have long been hypothesized to be involved in maintaining cell height (Burnside 1973; Picone, Ren et al. 2010). In the amnioserosa (a squamous epithelial tissue in the dorsal-most region of the cellular blastoderm) of gastrulating Drosophila embryos, a cell shape conversion from columnar to squamous is observed due to change in the microtubule orientation from the apical- basal axis to the planar axis (Pope and Harris 2008). Moreover, it was shown that microtubules are needed for polarized localization of many factors like RhoA (Nakaya, Sukowati et al. 2008), polarity regulators (Siegrist and Doe 2007), RhoGEF (Rogers, Wiedemann et al. 2004) and adhesion molecules (Harris and Peifer 2007). Whether the microtubule requirement for localization of RhoA is direct or indirect is not known. Our studies provide evidence that BMP signals are required in the sensory placodes for apical accumulation of RhoA. However, microtubules regulating cell height are still under discussion.

An additional well studied mechanism for cell length was shown in the Drosophila ventral furrow formation, suggesting that apical constriction generates cytoplasmic flow and results in cell elongation (He, Doubrovinski et al. 2014). Our study in the olfactory placode showed that blocking BMP signals in the olfactory ectoderm led to reduced apical constriction with decreased cell length. If the reduced cell height is because of the increase in apical width then this indirectly means that the mechanism of cytoplasmic flow is also regulated by BMP signaling.

Although small Rho GTPases (Rho and Rac) have been shown to be important for the process of epithelial bending, invagination still occurs in absence of RhoA and Rac1, which suggests that other unknown potential molecules are possibly involved in the process of epithelial invagination (Chauhan, Lou et al. 2011; Sai, Yonemura et al. 2014). On the other hand, cells in both RhoA and Rac1 mutants display no significant changes in the basal surface area. Studies in chick embryos demonstrate that the cells in the otic placode undergo basal expansion first and apical constriction later. Furthermore, basally localized FGF signaling generates basal cell expansion in the otic placode by depletion of actin filaments. Activation of a RhoA-ROCK dependent pathway results in myosin II activation essential to direct the contraction of apical actin networks and subsequent apical constriction (Sai and Ladher 2008; Sai, Yonemura et al. 2014). Studies in mouse mutant embryos show that when FGF signaling is blocked the thickening of the lens placode is significantly reduced (Faber, Dimanlig et al. 2001). But it is not further analyzed, if the FGF signaling is essential for cell morphology in the lens placode invagination. Based on these studies, a possible reason might be that the apical constriction with elongated cells causes only slight bending, and basal cell expansion is further needed for deeper invagination. From the present studies, inhibition of BMP signaling in the lens ectoderm shows defective placode formation and subsequent invagination, whereas in the olfactory and the otic placode formation is normal but the invagination is shown to be affected. A possible reason might be that both the olfactory and otic placodes have two distinct domains and these regions are known to be regulated by FGF and BMP, respectively (Abello, Khatri et al. 2010; Maier, von Hofsten et al. 2010). In the absence of one domain, the placode still maintains its identity which is sufficient for thickening of the ectoderm, whereas for invagination both domains are required. Alternatively, a balance of both FGF and BMP activities is required for placode invagination (Fig. 6).

Figure 6: (A) schematic image showing Hes5 positive cells in the invaginated olfactory

placode and Id3 in the rim of the olfactory placode. (B) In the absence of BMP signaling Hes5 positive olfactory placode cell identity was maintained. (C) In the absence of FGF

signaling Id3 positive olfactory placode cell identity was retained (Maier et al., 2010). FGF, fibroblast growth factor; BMP, bone morphogenetic protein.

In addition, the basal lamina might play a prominent role in placode invagination. Using pull down assays it was demonstrated that laminin 10/11 activates Rac; in contrast activation of Rho is influenced by fibronectin (Gu, Sumida et al. 2001). In zebrafish fin epithelium it was shown that the laminin regulates cell shape changes via canonical Wnt signaling pathway (Nagendran, Arora et al. 2015). In Drosophila eye morphogenesis, integrins were shown to regulate apical constriction through microtubule stabilization (Fernandes, McCormack et al. 2014). There are several studies showing that integrins are cellular receptors for laminins, and they mediate anchoring to basal lamina (Zagris 2001; Zagris, Christopoulos et al. 2004). Integrins influence multiple functions by anchoring cells to the extracellular matrix (ECM). Integrins in the extracellular region act as a bridge between the laminin-containing ECM and the cytoskeleton of the cell in the cytosol, resulting in changes in cell polarity, shape and migration. Based on my personal unpublished observations, laminins are expressed as a thicker band before placode formation and later this lamina is refined to a thinner region as the placode invaginates (Fig.7). Apical constriction, increase in cell length, basal expansion and thinning of basal lamina altogether might play a potential role for complete placode invagination. Disparity in any of the stated events might show mild to severe defects in placode invagination. Further studies are needed regarding the role of laminin during placode invagination, and additional experiments have to be done to determine if the requirement for BMP

during sensory placode invagination is due to its impact on graded levels of laminin deposition.

Figure 7: At olfactory placode stage (st 14) Laminin was broadly expressed within the olfactory placode region. By stage 17, when the placode starts to invaginate, Laminin was concentrated in a thinner region at the basal side of the epithelium underneath the slightly invaginated placode. Cell nuclei stained with DAPI appear in blue and red indicates Laminin.

BMP activity acts upstream of ROCK

Vertebrate neural tube development is a typical model of epithelial invagination, since the neural tube is the first organ to develop. For neural tube closure, RhoA activation by ArhGEF11 is essential through activation of pMLC (Nishimura, Honda et al. 2012). Similarly, in the otic placode, RhoA is apically localized in the same way as during neural tube closure which is also known to be activated by ArhGEF11. ArhGEF11 and its putative downstream effector ROCK activate the levels of pMLC and F-actin at the apical junctions. Inhibiting RhoA, ArhGEF11and ROCK all reduced apical constriction with reduced levels of pMLC and F-actin in the apical junctions (Sai, Yonemura et al. 2014). During otic placode invagination, the PCP (planar cell polarity) marker Celsr1 is precisely expressed in the center of the otic placode; by blocking the function of this gene placodal cell displayed a significant increase in apical surface area. It has been suggested that Celsr1 lies upstream of RhoA during otic placode invagination (Sai, Yonemura et al. 2014). Although blocking Rho, ArhGEF, and Celsr1 shows reduced apical localization of pMLC and F-actin, invagination still occurs. Since ROCK is the downstream target of all the above mentioned genes, blocking of ROCK function using a

pharmacological inhibitor results in drastic effects on otic placode invagination (Chauhan, Lou et al. 2011; Sai, Yonemura et al. 2014). Altogether it can be suggested that Rho and GEF molecules are expressed specifically in placode cells, whereas ROCK is expressed in the placode as well as in the lateral ectoderm. By inhibiting ROCK, the force being generated from the lateral surface ectoderm is also inhibited. This suggests that the lateral surface ectoderm might be playing an essential role in sensory placode invagination. Our present study shows that ROCK inhibition exhibits failure in all the three sensory placode invagination with disrupted accumulation of F-actin. Moreover, our results provide evidence that morphologically the ROCK-inhibited placodal epithelia were indistinguishable from the BMP-inhibited flat placodal domain. This indicates that BMP activity acts upstream of the ROCK-regulatory molecular machinery to regulate epithelial invagination.

Another known suggested mechanism for epithelial invagination is apoptosis. Studies in fly embryos demonstrate that apoptotic cell extrusion accelerates dorsal closure (Toyama, Peralta et al. 2008). Another study in Drosophila epithelial cells displays that apoptotic cells actively influence their surrounding cells by accumulation of acto-myosin along their apical-basal axes and pulls the apical surface inward resulting in the transient depression of the apical surface (Monier, Gettings et al. 2015). However, our current study shows that failure in placode invagination after BMP inhibition is not caused by alterations in cell death or loss of placodal identity.

In contrast to the mechanism of apoptosis in epithelial invagination, it was suggested that a mitotic cell rounding and interkinetic nuclear migration (IKNM) play an important role in apical constriction and epithelial invagination (Messier 1978; Stewart, Helenius et al. 2011). In Drosophila tracheal placode invagination, mitotic cell rounding is known to play an important role, and using microtubule inhibitor colchicine it was shown that cell rounding, but not cell division, is responsible for tracheal placode invagination (Kondo and Hayashi 2013). However, in our study we do not see any significant difference in cell proliferation in lens and otic placodes. A possible reason for lack of placode invagination might be due to delayed mitosis

after BMP inhibition, which would affect invagination but not the number of proliferating cells. Since IKNM was to be required for neural tube closure, a primary question about IKNM is how different or similar the process is between tissue types.

Conclusions

1. Prior to the lens placode stage (stage 13) BMP signals are essential for acquisition of lens identity both in vitro and in vivo. Until stage 13, continuous exposure of BMP signals is required for induction of L-Maf (lens placode identity marker). In the absence of BMP activity, explanted lens cells acquire olfactory characteristics. 2. After the onset of L-Maf primary lens fibre cell differentiation indicated by δ-crystallin, becomes independent of BMP signals. 3. At optic vesicle stage BMP signals from the lens ectoderm are required for promoting retina cell fate in part by suppressing telencephalic identity. By stage 13, retinal cells are specified suggesting that prospective retina cells do not require lens tissue to maintain retina characteristics. 4. BMP signals are required for placode invagination. 5. BMP signals regulating the placode invagination is shown to be required for both sensory (olfactory, otic and lens) and non-sensory (hypophyseal) placode invagination. 6. Disruption in placode invagination after BMP inhibition is not due to loss of placode specific identity, but mainly to failure in apical localization of RhoA and F-actin.

Acknowledgements

There have been many people who have guided me, showed me opportunities during the last few years and I would like to thank each and every one of them. I would like to thank all the people who helped me reach my goal and please forgive me if i forget someone.

First of all, I wish to express my gratitude to my supervisor, Lena Gunhaga. Thank you for giving me the opportunity to be in your lab. The biggest dream of my life was not only possible but also a pleasure with your unlimited support and guidance. You are an excellent supervisor; I never hesitated to step into your office to discuss my projects and experimental problems. You gave me endless freedom to learn many things about the research which helped me improve my critical thinking and analyzing skills. It was not just at work, the support you have given when my health was deteriorating was amazing and unforgettable. I would like to thank from the bottom of my heart for everything. Thanks a lot for your help for my thesis writing. Am sure I will miss working with you in future.

I also wish to sincerely thank my co supervisor, Jonas Von Hofsten: I am really thankful for teaching me the electroporation technique and grateful for all the support during my tenure.

I wish to acknowledge the support received from the past and present lab members. Previous lab members; Tanushree: It was really nice working with you. I am really thankful for your help in teaching me the lab techniques in my initial joining days. I really like the way you discuss with interest in every scientific and nonscientific topics. I wish you all the best for your future. Miguel Jarrin: I miss the joy of working with you and still miss the fun talks we used to have daily! It was really fun discussing Spanish words with you. Esther: You are the first person who encouraged me to visit IKSU and now i realize the importance of it. Nevertheless, lately started visiting IKSU and would be thankful for the rest of my life for a good start. Walter: I always see you working without taking any breaks. I wonder where you get that energy to work like that without taking any breaks. Hanna Nord: I still remember the days when i was new to the lab and you responded to endless questions of mine with patience. I would like to thank you for all the help and support. I wish you all the best for your future. Helena Alstermark: It was really nice working with you during the initial days of work in this lab. I have tried my best to learn from you in organizing the stuff. All the masters and project students Nils, Harsha, Raghu, Manoj, Margaret, Ashleigh, Shori, Dåg, Emma, Miriam, Elsa, Parham it was fun getting to know all of you.

Current lab members: Cedric Patthey, It was a pleasure to work with you. You are very inspiring in several aspects of my PhD life. How to think rationally, clearly and reasonably when it comes to research is something i have learnt from you. You

have not only given your time generously, but also provided responses and questions that were valuable to my research. Though i have doubts, i hesitated to ask questions during several lab meetings and it was from you i have learnt to open up about my agreements or disagreements. Whenever i come to the lab during the late nights i always see you in the lab or in your office. Your dedication, swiftness and promptness inspired me a lot. I am really thankful for your help in correcting my thesis. Tami, Soufien, Alice, and Raj: I am really thankful to you all and it was really a great experience to know all of you. Raj i would really thank you for your help when i was in need. Alice and Shori it was really fun working with both of you. Tami and Soufien I really enjoyed nice discussions on th scientific and non scientific topics.

Lars Nilsson: You are one of the wonderful persons i met during my PhD. I really enjoyed discussions about the religious stuff and the atheism committee. You are never tired of discussing scientific and nonscientific subject anytime anywhere. Saba and Nawaz: It was really nice to know you guys. I felt really comfortable discussing with you when i am more stressed. I wish you all the best for your future. Agnes: Good luck with your Postdoc. It was really a nice dinner definitely i would like to have one more of it. Hope you are having nice time with cute little Sara. Christoffer Nord: As we were sharing the same office for thesis writing in the final stages it was really helpful from you in getting the advice in using several software programs including Volocity. Thank you for everything. Gowtham: I would like to thank for your help. P-A, Mirella, Lina: i would like to thank you all for making things easy for me in ordering and in the administrative documentation.

Jonathan Gilthorpe: I will never forget your help in suggesting some scientific ideas despite your busy schedule. I would like to thank you for everything. Iwan Jones: I am not sure if you remember you suggested me the tips for cloning. I still follow those tips with lots of successful cloning results. Thank you so much! Åke Renmen: It was you who made me more stable when my health was deteriorating and i can’t thank enough for that. Annelisa Wallius: My swimming pool friend. We had several nice discussions which was really helpful for me. Thank you. Anne & Ash: Thank you for inviting us to the get togethers.

Saveer: You are one of my childhood friends whom i will never forget. I was really happy when you came to Sweden for PhD. I will never forget those days whenever i feel low i use to call you share my feelings with you. Thanks alot for your support. Sudhakar: You are the person who suggested me to come to Sweden for PhD. This day would not come if you have not suggested me to come to Sweden. I am really thankful to you from the bottom of my heart. Srinivas N: I am glad we have done masters together and we still share each other's problems.

Dozens of people have helped me make my home away from home. I would like to thank Kiran & Sirisha for helping me put those first steps in the university and get accustomed to the new place. Very special thanks to Pramod & Deepthi for the

friendly chats, informal dinners, long walks and all the help during their stay in Umeå.

I would not have enjoyed little things in life if i hadn’t found friends outside work. Thank you so much for happily talking about things other than just papers. Heartful thanks to Ravi, Vidya; I am really thankful to Ravi for giving me the moral support during my tough times. Karthik, Gowthami & Shyam, Reshma for making our parenting journey easier. I would like to thank Karthik for his support during my house shifting. I would like to thank you for everything.

My time in Umeå was made enjoyable in large part due to many friends that became part of my life. I am grateful for the time spent with my roommate Edwin, sharing weekend lunches with Manoj, Atiq and for the fun talks we had. I would also like to acknowledge all friends who made my PhD journey memorable in Umeå. Thank you Uma, Bala, Philge, Nisha, Aaron, Madhavi, Mohan, Munindar, Sharvani. Especially thank Pramod for helping me in shifting my apartment. I would like to also thank Harsha, Chaitanya for helping in shifting. I would like to thank all the Indian friends Yugender, Venky, Srikanth Kolan, Shashank, and Rammurthy.

Amma, Nanna & Family: At every step of the way, they were with me even though they were physically thousands of miles away. Whatever i have achieved so far it is because of your support. I still remember how i was struggling to find a Phd position and the support you gave me was incredible, because of your support i have reached this far.

Kamla my wonderful wife... Her intelligence, honesty, and love have many times been the reason why I did not turn around and run away. I cannot imagine a more special soul-mate; I cannot imagine a better mother for my son. Support you gave me when i have to come to the lab during late nights was fantastic. Vaikan my cute little support system... There is nothing like the cute innocent face of a sleeping child to bring out the strongest determination to succeed. It was his magic that pulled me out of many intervals of depression. I love you both. Nothing would be possible without you.

References

Abello, G., S. Khatri, et al. (2010). "Independent regulation of Sox3 and Lmx1b by FGF and BMP signaling influences the neurogenic and non-neurogenic domains in the chick otic placode." Dev Biol 339(1): 166-178. Akutsu, S., M. Takada, et al. (1992). "Origin of luteinizing hormone-releasing hormone (LHRH) neurons in the chick embryo: effect of the olfactory placode ablation." Neuroscience letters 142(2): 241-244. Alsina, B., G. Abelló, et al. (2004). "FGF signaling is required for determination of otic neuroblasts in the chick embryo." Dev Biol 267(1): 119-134. Amano, M., M. Ito, et al. (1996). "Phosphorylation and activation of myosin by Rho-associated kinase (Rho-kinase)." Journal of Biological Chemistry 271(34): 20246-20249. Ashery-Padan, R., T. Marquardt, et al. (2000). "Pax6 activity in the lens primordium is required for lens formation and for correct placement of a single retina in the eye." Genes & development 14(21): 2701-2711. Bailey, A. P., S. Bhattacharyya, et al. (2006). "Lens specification is the ground state of all sensory placodes, from which FGF promotes olfactory identity." Developmental cell 11(4): 505-517. Bailey, A. P. and A. Streit (2005). "Sensory Organs: Making and Breaking the Pre‐Placodal Region." Current topics in developmental biology 72: 167-204. Baker, C. V. and M. Bronner-Fraser (2001). "Vertebrate cranial placodes I. Embryonic induction." Dev Biol 232(1): 1-61. Barembaum, M. and M. Bronner-Fraser (2007). "Spalt4 mediates invagination and otic placode gene expression in cranial ectoderm." Development 134(21): 3805-3814. Begbie, J. and A. Graham (2001). "The ectodermal placodes: a dysfunctional family." PHILOSOPHICAL TRANSACTIONS-ROYAL SOCIETY OF LONDON SERIES B BIOLOGICAL SCIENCES 356: 1655-1660. Belecky-Adams, T. L., R. Adler, et al. (2002). "Bone morphogenetic protein signaling and the initiation of lens fiber cell differentiation." Development 129(16): 3795-3802. Bharti, K., W. Liu, et al. (2008). "Alternative promoter use in eye development: the complex role and regulation of the transcription factor MITF." Development 135(6): 1169-1178. Bhattacharyya, S., A. P. Bailey, et al. (2004). "Segregation of lens and olfactory precursors from a common territory: cell sorting and reciprocity of Dlx5 and Pax6 expression." Dev Biol 271(2): 403-414. Bhattacharyya, S. and M. Bronner-Fraser (2004). "Hierarchy of regulatory events in sensory placode development." Current opinion in genetics & development 14(5): 520-526. Bielen, H. and C. Houart (2012). "BMP signaling protects telencephalic fate by repressing eye identity and its Cxcr4-dependent morphogenesis." Developmental cell 23(4): 812-822. Borges, R. M., M. L. Lamers, et al. (2011). "Rho signaling pathway and apical constriction in the early lens placode." Genesis 49(5): 368-379. Brigande, J. V., L. E. Iten, et al. (2000). "A fate map of chick otic cup closure reveals lineage boundaries in the dorsal otocyst." Dev Biol 227(2): 256-270. Brugmann, S. A., P. D. Pandur, et al. (2004). "Six1 promotes a placodal fate within the lateral neurogenic ectoderm by functioning as both a transcriptional activator and repressor." Development 131(23): 5871-5881. Burnside, B. (1973). "Microtubules and microfilaments in amphibian neurulation." American Zoologist 13(4): 989-1006. Bäumer, N., T. Marquardt, et al. (2003). "Retinal pigmented epithelium determination requires the redundant activities of Pax2 and Pax6." Development 130(13): 2903-2915. Cau, E., S. Casarosa, et al. (2002). "Mash1 and Ngn1 control distinct steps of determination and differentiation in the olfactory sensory neuron lineage." Development 129(8): 1871-1880. Cau, E., G. Gradwohl, et al. (2000). "Hes genes regulate sequential stages of neurogenesis in the olfactory epithelium." Development 127(11): 2323-2332. Chauhan, B., T. Plageman, et al. (2015). "Epithelial morphogenesis: the mouse eye as a model system." Curr Top Dev Biol 111: 375-399. Chauhan, B. K., A. Disanza, et al. (2009). "Cdc42-and IRSp53-dependent contractile filopodia tether presumptive lens and retina to coordinate epithelial invagination." Development 136(21): 3657-3667.

Chauhan, B. K., M. Lou, et al. (2011). "Balanced Rac1 and RhoA activities regulate cell shape and drive invagination morphogenesis in epithelia." Proc Natl Acad Sci U S A 108(45): 18289-18294. Chizhikov, V. V. and K. J. Millen (2004). "Control of roof plate formation by Lmx1a in the developing spinal cord." Development 131(11): 2693-2705. Christophorou, N. A., M. Mende, et al. (2010). "Pax2 coordinates epithelial morphogenesis and cell fate in the inner ear." Dev Biol 345(2): 180-190. Dudley, A. T. and E. J. Robertson (1997). "Overlapping expression domains of bone morphogenetic protein family members potentially account for limited tissue defects in BMP7 deficient embryos." Developmental Dynamics 208(3): 349-362. Eiraku, M., N. Takata, et al. (2011). "Self-organizing optic-cup morphogenesis in three-dimensional culture." Nature 472(7341): 51-56. Faber, S. C., P. Dimanlig, et al. (2001). "Fgf receptor signaling plays a role in lens induction." Development 128(22): 4425-4438. Fekete, D. M. (1999). "Development of the vertebrate ear: insights from knockouts and mutants." Trends in neurosciences 22(6): 263-269. Fernandes, V. M., K. McCormack, et al. (2014). "Integrins regulate apical constriction via microtubule stabilization in the Drosophila eye disc epithelium." Cell reports 9(6): 2043-2055. Fernández-Garre, P., L. Rodríguez-Gallardo, et al. (2002). "Fate map of the chicken neural plate at stage 4." Development 129(12): 2807-2822. Freter, S., Y. Muta, et al. (2008). "Progressive restriction of otic fate: the role of FGF and Wnt in resolving inner ear potential." Development 135(20): 3415-3424. Fuhrmann, S. (2010). "Eye morphogenesis and patterning of the optic vesicle." Current topics in developmental biology 93: 61. Fujimura, N., M. M. Taketo, et al. (2009). "Spatial and temporal regulation of Wnt/β-catenin signaling is essential for development of the retinal pigment epithelium." Dev Biol 334(1): 31-45. Furuta, Y. and B. L. Hogan (1998). "BMP4 is essential for lens induction in the mouse embryo." Genes & development 12(23): 3764-3775. Glavic, A., S. M. Honoré, et al. (2004). "Role of BMP signaling and the homeoprotein Iroquois in the specification of the cranial placodal field." Dev Biol 272(1): 89-103. Gotoh, N., M. Ito, et al. (2004). "Tyrosine phosphorylation sites on FRS2α responsible for Shp2 recruitment are critical for induction of lens and retina." Proc Natl Acad Sci U S A 101(49): 17144-17149. Green, E. S., J. L. Stubbs, et al. (2003). "Genetic rescue of cell number in a mouse model of microphthalmia: interactions between Chx10 and G1-phase cell cycle regulators." Development 130(3): 539-552. Groves, A. K. and M. Bronner-Fraser (2000). "Competence, specification and commitment in otic placode induction." Development 127(16): 3489-3499. Gu, J., Y. Sumida, et al. (2001). "Laminin-10/11 and fibronectin differentially regulate integrin-dependent Rho and Rac activation via p130Cas-CrkII-DOCK180 pathway." Journal of Biological Chemistry 276(29): 27090-27097. Gunhaga, L. (2011). "The lens: a classical model of embryonic induction providing new insights into cell determination in early development." Philosophical Transactions of the Royal Society B: Biological Sciences 366(1568): 1193-1203. Gunhaga, L., M. Marklund, et al. (2003). "Specification of dorsal telencephalic character by sequential Wnt and FGF signaling." Nature neuroscience 6(7): 701-707. Hall, A. (2012). "Rho family GTPases." Biochemical Society Transactions 40(6): 1378. Harris, T. J. and M. Peifer (2007). "aPKC controls microtubule organization to balance adherens junction symmetry and planar polarity during development." Developmental cell 12(5): 727-738. Hatada, Y. and C. D. Stern (1994). "A fate map of the epiblast of the early chick embryo." Development 120(10): 2879-2889. He, B., K. Doubrovinski, et al. (2014). "Apical constriction drives tissue-scale hydrodynamic flow to mediate cell elongation." Nature 508(7496): 392. Hirashima, M., T. Kobayashi, et al. (2008). "Anteroventrally localized activity in the optic vesicle plays a crucial role in the optic development." Dev Biol 317(2): 620-631. Hogan, B. L. (1999). "Morphogenesis." Cell 96(2): 225-233. Hollnagel, A., V. Oehlmann, et al. (1999). "Id genes are direct targets of bone morphogenetic protein induction in embryonic stem cells." Journal of Biological Chemistry 274(28): 19838-19845. Horsford, D. J., M.-T. T. Nguyen, et al. (2005). "Chx10 repression of Mitf is required for the maintenance of mammalian neuroretinal identity." Development 132(1): 177-187.

Hsieh, J.-C., A. Rattner, et al. (1999). "Biochemical characterization of Wnt-frizzled interactions using a soluble, biologically active vertebrate Wnt protein." Proceedings of the National Academy of Sciences 96(7): 3546-3551. Huang, J., R. Rajagopal, et al. (2011). "The mechanism of lens placode formation: a case of matrix- mediated morphogenesis." Dev Biol 355(1): 32-42. Hyer, J., J. Kuhlman, et al. (2003). "Optic cup morphogenesis requires pre-lens ectoderm but not lens differentiation." Dev Biol 259(2): 351-363. Hägglund, A.-C., L. Dahl, et al. (2011). "Lhx2 is required for patterning and expansion of a distinct progenitor cell population committed to eye development." PLoS One 6(8): e23387. Jamora, C. and E. Fuchs (2002). "Intercellular adhesion, signalling and the cytoskeleton." Nature cell biology 4(4): E101-E108. Jarrin, M., T. Pandit, et al. (2012). "A balance of FGF and BMP signals regulates cell cycle exit and Equarin expression in lens cells." Molecular biology of the cell 23(16): 3266-3274. Jidigam, V. K. and L. Gunhaga (2013). "Development of cranial placodes: Insights from studies in chick." Development, growth & differentiation 55(1): 79-95. Kagiyama, Y., N. Gotouda, et al. (2005). "Extraocular dorsal signal affects the developmental fate of the optic vesicle and patterns the optic neuroepithelium." Development, growth & differentiation 47(8): 523-536. Karabagli, H., P. Karabagli, et al. (2002). "Comparison of the expression patterns of several fibroblast growth factors during chick gastrulation and neurulation." Anatomy and embryology 205(5-6): 365-370. Katahira, T., T. Sato, et al. (2000). "Interaction between Otx2 and Gbx2 defines the organizing center for the optic tectum." Mechanisms of development 91(1): 43-52. Kawauchi, S., S. Takahashi, et al. (1999). "Regulation of lens fiber cell differentiation by transcription factor c-Maf." Journal of Biological Chemistry 274(27): 19254-19260. Kil, S.-H., A. Streit, et al. (2005). "Distinct roles for hindbrain and paraxial mesoderm in the induction and patterning of the inner ear revealed by a study of vitamin-A-deficient quail." Dev Biol 285(1): 252-271. Kim, J. I., T. Li, et al. (1999). "Requirement for the c-Maf transcription factor in crystallin gene regulation and lens development." Proceedings of the National Academy of Sciences 96(7): 3781-3785. Kimmel, C. B., R. Warga, et al. (1990). "Origin and organization of the zebrafish fate map." Development 108(4): 581-594. Kondo, T. and S. Hayashi (2013). "Mitotic cell rounding accelerates epithelial invagination." Nature 494(7435): 125-129. Kozlowski, D. J., T. Murakami, et al. (1997). "Regional cell movement and tissue patterning in the zebrafish embryo revealed by fate mapping with caged fluorescein." Biochemistry and cell biology 75(5): 551-562. Kurose, H., M. Okamoto, et al. (2005). "FGF19‐FGFR4 signaling elaborates lens induction with the FGF8‐L‐Maf cascade in the chick embryo." Development, growth & differentiation 47(4): 213-223. Ladher, R. K., K. U. Anakwe, et al. (2000). "Identification of synergistic signals initiating inner ear development." Science 290(5498): 1965-1967. Ladher, R. K., T. J. Wright, et al. (2005). "FGF8 initiates inner ear induction in chick and mouse." Genes & development 19(5): 603-613. Lamb, T. M., A. K. Knecht, et al. (1993). "Neural induction by the secreted polypeptide noggin." Science 262(5134): 713-718. Lancaster, M. A., M. Renner, et al. (2013). "Cerebral organoids model human brain development and microcephaly." Nature 501(7467): 373-379. Layer, P. G. and R. Alber (1990). "Patterning of chick brain vesicles as revealed by peanut agglutinin and cholinesterases." Development 109(3): 613-624. Lecuit, T. and P.-F. Lenne (2007). "Cell surface mechanics and the control of cell shape, tissue patterns and morphogenesis." Nature Reviews Molecular Cell Biology 8(8): 633-644. Li, H.-S., J.-M. Yang, et al. (1994). "Pax-6 Is First Expressed in a Region of Ectoderm Anterior to the Early Neural Plate: Implications for Stepwise Determination of the Lens." Dev Biol 162(1): 181-194. Litsiou, A., S. Hanson, et al. (2005). "A balance of FGF, BMP and WNT signalling positions the future placode territory in the head." Development 132(18): 4051-4062. Lorén, C. E., J. W. Schrader, et al. (2009). "FGF signals induce Caprin2 expression in the vertebrate lens." Differentiation 77(4): 386-394.

Maier, E. and L. Gunhaga (2009). "Dynamic expression of neurogenic markers in the developing chick olfactory epithelium." Dev Dyn 238(6): 1617-1625. Maier, E., J. von Hofsten, et al. (2010). "Opposing Fgf and Bmp activities regulate the specification of olfactory sensory and respiratory epithelial cell fates." Development 137(10): 1601-1611. Maier, E. C., A. Saxena, et al. (2014). "Sensational placodes: Neurogenesis in the otic and olfactory systems." Dev Biol 389(1): 50-67. Mathers, P., A. Grinberg, et al. (1997). "The Rx homeobox gene is essential for vertebrate eye development." Nature 387(6633): 603-607. McCabe, K. L. and M. Bronner-Fraser (2009). "Molecular and tissue interactions governing induction of cranial ectodermal placodes." Dev Biol 332(2): 189-195. Messier, P.-E. (1978). "Microtubules, interkinetic nuclear migration and neurulation." Experientia 34(3): 289-296. Monier, B., M. Gettings, et al. (2015). "Apico-basal forces exerted by apoptotic cells drive epithelium folding." Nature 518(7538): 245-248. Morcillo, J., J. R. Martínez-Morales, et al. (2006). "Proper patterning of the optic fissure requires the sequential activity of BMP7 and SHH." Development 133(16): 3179-3190. Nagai, H., M. C. Lin, et al. (2011). "A modified cornish pasty method for ex ovo culture of the chick embryo." Genesis 49(1): 46-52. Nagendran, M., P. Arora, et al. (2015). "Canonical Wnt signalling regulates epithelial patterning by modulating levels of laminins in zebrafish appendages." Development 142(2): 320-330. Nakaya, Y., E. W. Sukowati, et al. (2008). "RhoA and microtubule dynamics control cell–basement membrane interaction in EMT during gastrulation." Nature cell biology 10(7): 765-775. Nguyen, M. and H. Arnheiter (2000). "Signaling and transcriptional regulation in early mammalian eye development: a link between FGF and MITF." Development 127(16): 3581-3591. Nishimura, T., H. Honda, et al. (2012). "Planar cell polarity links axes of spatial dynamics in neural-tube closure." Cell 149(5): 1084-1097. Nobes, C. D. and A. Hall (1999). "Rho GTPases control polarity, protrusion, and adhesion during cell movement." J Cell Biol 144(6): 1235-1244. Norgren, R. B. and C. Gao (1994). "LHRH neuronal subtypes have multiple origins in chickens." Dev Biol 165(2): 735-738. Ogino, H. and K. Yasuda (1998). "Induction of lens differentiation by activation of a bZIP transcription factor, L-Maf." Science 280(5360): 115-118. Ohyama, T., O. A. Mohamed, et al. (2006). "Wnt signals mediate a fate decision between otic placode and epidermis." Development 133(5): 865-875. Ory, S., O. Destaing, et al. (2002). "Microtubule dynamics differentially regulates Rho and Rac activity and triggers Rho-independent stress fiber formation in macrophage polykaryons." European journal of cell biology 81(6): 351-362. Pandit, T., V. K. Jidigam, et al. (2011). "BMP-induced L-Maf regulates subsequent BMP-independent differentiation of primary lens fibre cells A balance of BMP and notch activity regulates neurogenesis and olfactory nerve formation." Dev Dyn 240(8): 1917-1928. Pandit, T., V. K. Jidigam, et al. (2015). "Neural retina identity is specified by lens-derived BMP signals." Development (Cambridge, England) 142(10): 1850-1859. Patthey, C., T. Edlund, et al. (2009). "Wnt-regulated temporal control of BMP exposure directs the choice between neural plate border and epidermal fate." Development 136(1): 73-83. Patthey, C. and L. Gunhaga (2011). "Specification and regionalisation of the neural plate border." European Journal of Neuroscience 34(10): 1516-1528. Patthey, C. and L. Gunhaga (2014). "Signaling pathways regulating ectodermal cell fate choices." Exp Cell Res 321(1): 11-16. Patthey, C., L. Gunhaga, et al. (2008). "Early development of the central and peripheral nervous systems is coordinated by Wnt and BMP signals." PLoS One 3(2): e1625. Picone, R., X. Ren, et al. (2010). "A polarised population of dynamic microtubules mediates homeostatic length control in animal cells." PLoS Biol 8(11): e1000542. Pittack, C., G. Grunwald, et al. (1997). "Fibroblast growth factors are necessary for neural retina but not pigmented epithelium differentiation in chick embryos." Development 124(4): 805-816. Plageman, T. F., B. K. Chauhan, et al. (2011). "A Trio-RhoA-Shroom3 pathway is required for apical constriction and epithelial invagination." Development 138(23): 5177-5188. Plageman, T. F., M.-I. Chung, et al. (2010). "Pax6-dependent Shroom3 expression regulates apical constriction during lens placode invagination." Development 137(3): 405-415. Plageman, T. F., Jr., B. K. Chauhan, et al. (2011). "A Trio-RhoA-Shroom3 pathway is required for apical constriction and epithelial invagination." Development 138(23): 5177-5188.

Pope, K. L. and T. J. Harris (2008). "Control of cell flattening and junctional remodeling during squamous epithelial morphogenesis in Drosophila." Development 135(13): 2227-2238. Puelles, L. and J. L. Rubenstein (2003). "Forebrain gene expression domains and the evolving prosomeric model." Trends in neurosciences 26(9): 469-476. Rajagopal, R., J. Huang, et al. (2009). "The type I BMP receptors, Bmpr1a and Acvr1, activate multiple signaling pathways to regulate lens formation." Dev Biol 335(2): 305-316. Rembold, M., F. Loosli, et al. (2006). "Individual cell migration serves as the driving force for optic vesicle evagination." Science 313(5790): 1130-1134. Reza, H. M., H. Ogino, et al. (2002). "L-Maf, a downstream target of Pax6, is essential for chick lens development." Mechanisms of development 116(1): 61-73. Reza, H. M., A. Urano, et al. (2007). "Sequential and combinatorial roles of maf family genes define proper lens development." Molecular vision 13: 18. Ridley, A. J. (2006). "Rho GTPases and actin dynamics in membrane protrusions and vesicle trafficking." Trends in cell biology 16(10): 522-529. Robinson, M. L. (2006). An essential role for FGF receptor signaling in lens development. Seminars in cell & developmental biology, Elsevier. Rogers, S. L., U. Wiedemann, et al. (2004). "Drosophila RhoGEF2 associates with microtubule plus ends in an EB1-dependent manner." Current biology 14(20): 1827-1833. Rowan, S., C.-M. A. Chen, et al. (2004). "Transdifferentiation of the retina into pigmented cells in ocular retardation mice defines a new function of the homeodomain gene Chx10." Development 131(20): 5139-5152. Sai, X. and R. K. Ladher (2008). "FGF signaling regulates cytoskeletal remodeling during epithelial morphogenesis." Curr Biol 18(13): 976-981. Sai, X. and R. K. Ladher (2015). "Early steps in inner ear development: induction and morphogenesis of the otic placode." Frontiers in pharmacology 6. Sai, X., S. Yonemura, et al. (2014). "Junctionally restricted RhoA activity is necessary for apical constriction during phase 2 inner ear placode invagination." Dev Biol 394(2): 206-216. Saint-Jeannet, J.-P. and S. A. Moody (2014). "Establishing the pre-placodal region and breaking it into placodes with distinct identities." Dev Biol 389(1): 13-27. Sakai, M., M. S. Serria, et al. (2001). "Regulation of c-maf gene expression by Pax6 in cultured cells." Nucleic acids research 29(5): 1228-1237. Sánchez-Arrones, L., J. L. Ferrán, et al. (2009). "Incipient forebrain boundaries traced by differential gene expression and fate mapping in the chick neural plate." Dev Biol 335(1): 43-65. Sauer, F. C. (1936). "The interkinetic migration of embryonic epithelial nuclei." Journal of Morphology 60(1): 1-11. Schimmang, T. (2007). "Expression and functions of FGF ligands during early otic development." International journal of developmental biology 51(6-7): 473-481. Schlosser, G. (2006). "Induction and specification of cranial placodes." Dev Biol 294(2): 303-351. Schlosser, G. (2010). "Chapter Four-Making Senses: Development of Vertebrate Cranial Placodes." International review of cell and molecular biology 283: 129-234. Seifert, R., M. Jacob, et al. (1993). "The avian prechordal head region: a morphological study." J Anat 183(Pt 1): 75. Siegrist, S. E. and C. Q. Doe (2007). "Microtubule-induced cortical cell polarity." Genes & development 21(5): 483-496. Sigulinsky, C. L., E. S. Green, et al. (2008). "Vsx2/Chx10 ensures the correct timing and magnitude of Hedgehog signaling in the mouse retina." Dev Biol 317(2): 560-575. Simpson, A., D. Moss, et al. (1995). "The avian eye lens protein δ-crystallin shows a novel packing arrangement of tetramers in a supramolecular helix." Structure 3(4): 403-412. Sjodal, M., T. Edlund, et al. (2007). "Time of exposure to BMP signals plays a key role in the specification of the olfactory and lens placodes ex vivo." Developmental cell 13(1): 141-149. Smith, A. N., L.-A. Miller, et al. (2009). "Stage-dependent modes of Pax6-Sox2 epistasis regulate lens development and eye morphogenesis." Development 136(17): 2977-2985. Steinfeld, J., I. Steinfeld, et al. (2013). "RPE specification in the chick is mediated by surface ectoderm- derived BMP and Wnt signalling." Development 140(24): 4959-4969. Stewart, M. P., J. Helenius, et al. (2011). "Hydrostatic pressure and the actomyosin cortex drive mitotic cell rounding." Nature 469(7329): 226-230. Steventon, B., R. Mayor, et al. (2012). "Mutual repression between Gbx2 and Otx2 in sensory placodes reveals a general mechanism for ectodermal patterning." Dev Biol 367(1): 55-65. Stigloher, C., J. Ninkovic, et al. (2006). "Segregation of telencephalic and eye-field identities inside the zebrafish forebrain territory is controlled by Rx3." Development 133(15): 2925-2935.

Streit, A. (2002). "Extensive cell movements accompany formation of the otic placode." Dev Biol 249(2): 237-254. Streit, A. (2004). "Early development of the cranial sensory nervous system: from a common field to individual placodes." Dev Biol 276(1): 1-15. Suzuki, M., Y. Hara, et al. (2010). "MID1 and MID2 are required for Xenopus neural tube closure through the regulation of microtubule organization." Development 137(14): 2329-2339. Svoboda, K. and K. O'Shea (1987). "An analysis of cell shape and the neuroepithelial basal lamina during optic vesicle formation in the mouse embryo." Development 100(2): 185-200. Timmer, J. R., C. Wang, et al. (2002). "BMP signaling patterns the dorsal and intermediate neural tube via regulation of homeobox and helix-loop-helix transcription factors." Development 129(10): 2459-2472. Toyama, Y., X. G. Peralta, et al. (2008). "Apoptotic force and tissue dynamics during Drosophila embryogenesis." Science 321(5896): 1683-1686. Treloar, H. B., A. M. Miller, et al. (2010). "Development of the olfactory system." Trousse, F., P. Esteve, et al. (2001). "Bmp4 mediates apoptotic cell death in the developing chick eye." J Neurosci 21(4): 1292-1301. Wawersik, S., P. Purcell, et al. (1999). "BMP7 acts in murine lens placode development." Dev Biol 207(1): 176-188. Vendrell, V., E. Carnicero, et al. (2000). "Induction of inner ear fate by FGF3." Development 127(10): 2011-2019. Westenskow, P. D., J. B. McKean, et al. (2010). "Ectopic Mitf in the embryonic chick retina by co- transfection of β-catenin and Otx2." Invest Ophthalmol Vis Sci 51(10): 5328-5335. Winkler, S., F. Loosli, et al. (2000). "The conditional medaka mutation eyeless uncouples patterning and morphogenesis of the eye." Development 127(9): 1911-1919. Xu, H., C. M. Dude, et al. (2008). "Fine-grained fate maps for the ophthalmic and maxillomandibular trigeminal placodes in the chick embryo." Dev Biol 317(1): 174-186. Yun, S., Y. Saijoh, et al. (2009). "Lhx2 links the intrinsic and extrinsic factors that control optic cup formation." Development 136(23): 3895-3906. Zagris, N. (2001). "Extracellular matrix in development of the early embryo." Micron 32(4): 427-438. Zagris, N., M. Christopoulos, et al. (2004). "Developmentally regulated expression and functional role of α7 integrin in the chick embryo." Development, growth & differentiation 46(3): 299-307. Zelarayan, L. C., V. Vendrell, et al. (2007). "Differential requirements for FGF3, FGF8 and FGF10 during inner ear development." Dev Biol 308(2): 379-391. Zhao, H., T. Yang, et al. (2008). "Fibroblast growth factor receptor signaling is essential for lens fiber cell differentiation." Dev Biol 318(2): 276-288. Zuber, M. E., G. Gestri, et al. (2003). "Specification of the vertebrate eye by a network of eye field transcription factors." Development 130(21): 5155-5167. Zuber, M. E., M. Perron, et al. (1999). "Giant eyes in Xenopus laevis by overexpression of XOptx2." Cell 98(3): 341-352. Zygar, C. A., T. Cook, et al. (1998). "Gene activation during early stages of lens induction in Xenopus." Development 125(17): 3509-3519.