Mechanisms of epithelial packing in proliferative tissues

DEPARTAMENTO DE BIOLOGÍA CELULAR FACULTAD DE BIOLOGÍA UNIVERSIDAD DE SEVILLA INSTITUTO DE BIOMEDICINA DE SEVILLA (IBiS) MECHANISMS OF EPITHELIAL PACKING IN PROLIFERATIVE TISSUES TESIS DOCTORAL Programa de doctorado en Biología Integrada Universidad de Sevilla Antonio Jesús Tagua Jáñez Sevilla, 2024 El Dr. Luis María Escudero Cuadrado, profesor titular del Departamento de Biología Celular e investigador principal del Instituto de Biomedicina de Sevilla CERTIFICA: Que la memoria titulada “MECHANISMS OF EPITHELIAL PACKING IN PROLIFERATIVE TISSUES”, desarrollada por Antonio Jesús Tagua Jáñez, graduado en Biomedicina Básica y Experimental, ha sido realizada bajo su supervisión y reúne requisitos necesarios para optar al título de Doctor por la Universidad de Sevilla. En Sevilla, a 20 de febrero de 2024. Fdo: Luis M. Escudero Cuadrado Director de tesis Indexes GENERAL INDEX FIGURES AND TABLES INDEX ................................................................................. iv SUMMARY .................................................................................................................. vi 1. INTRODUCTION ...................................................................................................... 1 1.1. Early animal development .................................................................................. 3 1.1.1. Embryogenesis ............................................................................................ 3 1.1.2. Mitosis in animal cells .................................................................................. 7 1.1.3. Epithelial morphogenesis ............................................................................ 8 1.2. Echinoderms as animal models for the study of embryonic processes ............ 10 1.2.1. Comparative development of sea stars and sea urchins ........................... 10 1.3. Influence of geometry on tissue organization ................................................... 13 1.4. Interplay between tissue mechanics and morphogenesis ................................ 15 1.6. Cell packing in curved epithelia ....................................................................... 17 2. OBJECTIVES ......................................................................................................... 22 3. MATERIALS AND METHODS................................................................................ 26 3.1. Sea star embryo model .................................................................................... 28 3.1.1. Animal husbandry ...................................................................................... 28 3.1.2. Live imaging ............................................................................................... 28 3.1.3. Embryo compression ................................................................................. 29 3.2. Segmentation method ...................................................................................... 30 3.3. Quantification of morphological and packing features ....................................... 32 3. 3.1 Cell measurements .................................................................................... 33 3. 3.2. Tissue measurements .............................................................................. 35 3. 4. Voronoi constructions ...................................................................................... 37 3. 5. Delimitation of outer and inner layers and automatic detection of scutoids from segmented regions .................................................................................................. 37 3. 6. Tracking of scutoids ........................................................................................ 38 3. 7. Normalization of developmental time ............................................................... 38 3. 8. Statistical analysis ........................................................................................... 38 3. 9. Data availability and graphic illustrations ........................................................ 39 4. RESULTS ............................................................................................................... 41 4. 1. The 3D epithelial packing changes during development in proliferative tissues. ................................................................................................................................ 43 4. 2. Compression of the embryo induce scutoids upon global increase in cell density and compaction. ...................................................................................................... 47 4. 3. Analysis of the influence of tissue geometry and cell proliferation rate on cell packing. ................................................................................................................... 51 4. 4. Regional differences in tissue compaction affect propensity of scutoid formation. ................................................................................................................................ 54 4. 5. Scutoid formation is temporally linked to cell divisions. .................................... 60 5. DISCUSSION ......................................................................................................... 67 6. CONCLUSIONS ..................................................................................................... 77 7. REFERENCES ....................................................................................................... 80 8. ANNEXES .............................................................................................................. 91 8.1. Supplementary figures ...................................................................................... 93 8.2. Supplementary tables ....................................................................................... 96 8.3. Supplementary movies ................................................................................... 114 FIGURES AND TABLES INDEX Figure 1. Embryogenesis of an organism ...................................................................... 6 Figure 2. Cell cycle in epithelial cells ............................................................................. 9 Figure 3. Echinoderms development ........................................................................... 11 Figure 4. Topology in packed tissues .......................................................................... 14 Figure 5. Tissue material properties ............................................................................ 17 Figure 6. 3D packing in curved epithelia ..................................................................... 18 Figure 7. Correlation between curvature anisotropy and 3D packing .......................... 20 Table 1. Detailed information about P.miniata sea star embryos ................................. 29 Figure 8. Deep learning-based 3D segmentation ........................................................ 32 Figure 9. Segmented sea star embryos ...................................................................... 33 Figure 10. Description of tissue and cellular features .................................................. 36 Figure 11. 3D segmentation of sea star embryos over time ........................................ 44 Figure 12. Changes of cell packing, size, and shape during early sea star embryo development ............................................................................................................... 46 Figure 13. Processing of physically constrained sea star embryos .............................. 48 Figure 14. 3D packing and morphological analysis at cellular level over time between WT and WT-comp embryos ........................................................................................ 49 Figure 15. Size analysis at cellular level over time between WT and WT-comp embryos ................................................................................................................................... 50 Figure 16. Shape analysis of WT and WT-comp embryos .......................................... 52 Figure 17. Analysis of cell proliferation ........................................................................ 53 Figure 18. Polarization of embryonic poles in WT condition ........................................ 54 Figure 19. Asynchronous compaction and scutoid formation in WT embryos .............. 55 Figure 20. Polarization of embryonic poles in WT-comp condition .............................. 57 Figure 21. Synchronous compaction and scutoid formation in WT-comp embryo ....... 58 Figure 22. Cell size comparison between animal and vegetal poles ............................ 59 Figure 23. 3D packing analysis between real embryos and theorical models .............. 61 Figure 24. Tracking scutoids over sea star embryo development ................................ 62 Figure 25. Analysis of the scutoids onsets throughout the interphase ......................... 64 Figure 26. Analysis of scutoid onsets and duration ..................................................... 65 Figure 27. Relationship between compaction, cell density, convexity and scutoids ..... 70 Figure S1. Development of sea star P. miniata ........................................................... 93 Figure S2. Cell compaction ......................................................................................... 94 Figure S3. Single-cell tracking of division .................................................................... 95 Table S1. Statistics of global comparisons .................................................................. 96 Table S2. Statistics of regional comparisons ............................................................... 98 Table S3. Morphological classification of the embryos .............................................. 100 Table S4. Statistics related to tracked “scutoidal” cells and duration of the embryo development ............................................................................................................. 101 Table S5. Tracked scutoids analysis ......................................................................... 109 Table S6. Detailed information about the normalization of the developmental time ... 113 Movie S1. Patiria miniata WT embryo, vegetal view. ................................................. 114 Movie S2. Patiria miniata WT-comp embryo, vegetal view ........................................ 114 Movie S3. Patiria miniata WT embryo, animal view. .................................................. 114 Movie S4. Patiria miniata WT-comp embryo, animal view. ........................................ 114 INTRODUCTION 2 INTRODUCTION 3 1.1. Early animal development The fundamental principles of developmental biology are founded upon three essential pillars: the regulation of growth, the organization of cells to form ordered structures, and the intrinsic capacity of cells to undergo differentiation resulting in the emergence of distinct cell types with specialized functions. These intricate processes drive the development of an animal from a single cell to a multicellular organism. Acquiring a proper understanding of these mechanisms could shed light on the underlying causes behind occasional aberrations in morphogenesis, which can lead to abnormalities or halting the progression of development (Gilbert, 2000; Wolpert et al., 2019). 1.1.1. Embryogenesis During embryogenesis, a consecutive series of dynamic processes gives rise to the formation of a fully developed embryo. These events can vary among animals; however, there are similarities that most of them share. Therefore, we can simplify the embryonic development into different stages. In this section, I will provide an overview of this process and the events shared by most animals. Firstly, after the formation of the zygote, a sequence of fast mitotic divisions called cleavages begins dividing the fertilized egg into smaller cells called blastomeres. Unlike traditional cell divisions, which occur during the growth of proliferative tissues, these events do not increase the total size. In other words, the sum of all blastomeres conserves the same size as the zygote. Cleavages are classified into various types (Fig. 1A) based on the quantity and distribution of yolk, which refers to the stored nutrients present in the cytoplasm of an egg cell. An uneven distribution of yolk influences the pattern and direction of cell divisions. When one pole of the egg has low concentration of yolk, cell division occurs there faster than at the opposing pole. The pole of the egg that generally contains most of the yolk is referred to as the vegetal pole and the opposite is commonly referred to as the animal pole. There are two main types of cleavages during animal development: ➢ Holoblastic cleavage: commonly observed in mammals and echinoderms. It involves the complete division and separation of the cellular cytoplasms. Based on the pattern and direction of cell divisions, holoblastic cleavage can be radial, spiral, bilateral or rotational. In many species, the holoblastic cleavage divisions occur in a relatively equal and symmetrical manner, leading to blastomeres of similar size. This type of cleavage is usually associated with a relatively low amount of yolk and a uniform INTRODUCTION 4 distribution throughout the egg (isolecithal). In other cases, the cleavage is unequal, and the divisions lead to blastomeres of different size (large macromeres and small micromeres). ➢ Meroblastic cleavage: commonly observed in insects, reptiles, or birds. It involves only the cleavage of a portion of the cytoplasm, with the rest of the cytoplasm contains most of the yolk. Within this category, there are two subgroups: o Discoidal cleavage: also known as telolecithal, type of meroblastic cleavage characterized by the accumulation of the yolk in one end of the egg while cell divisions occur only in a small portion free of yolk. o Superficial cleavage: also known as centrolecithal, type of meroblastic cleavage characterized by the accumulation of the yolk in the centre, while the divisions of the cytoplasm occur around the periphery. Commonly observed in insects as the fruit fly Drosophila melanogaster. Finally, this stage usually ends with the formation of a structure called the blastula. In many animals, the blastula is a symmetrical structure, where the cells on the outer surface form a single layer known as the blastoderm. Meanwhile, the fluid-filled cavity inside the blastoderm is called the blastocoel. However, it is important to note that not all animals follow this pattern. For example, in insects with superficial cleavage, the end of this stage typically leads to the formation of a narrow, elongated structure called germ band. Similarly, in mammals, birds, and reptiles, the blastula is known as a blastocyst, and it has a more complex structure with an inner cell mass and an outer trophoblast layer. After the formation of the blastula, the stage called gastrulation begins, marking the transition from a single-layered structure to a multi-layered embryo (Fig. 1B). While the specific mechanisms and details of gastrulation may vary among different organisms, it generally involves the formation of a small indentation called the blastopore, which marks the site of the future mouth or anus. Additionally, the cells within the blastula undergo dynamic movements, whether through invagination (inward folding), ingression (inward migration from the surface) or other processes. These movements are crucial for reorganizing the cells into the proper positions, forming a multi-layered structure known as the germ layers. The outermost layer, called the ectoderm, gives rise to the skin and nervous system. INTRODUCTION 5 The innermost layer, the endoderm, is responsible for the development of the digestive tract and its related organs. Between these two layers, in most organisms, lies the mesoderm, which plays a role in generating connective tissues and musculature. Following the establishment of the three primary germ layers, organogenesis initiates (Fig. 1B), leading to the development of specialized tissues and organs within the developing embryo. Cells interact with each other and reorganize to form complex structures such as tubes and cavities. After the embryonic stage, the organism progresses through further development and maturity, undergoing various morphological and physiological changes. In many animal groups, including insects, amphibians, and some fish, a profound morphological and physiological transformation known as metamorphosis occurs (Fig. 1B). This process involves the development of new organs, such as wings and compound eyes, as well as the breakdown and reorganization of existing organs. In summary, embryogenesis is a complex series of developmental stages, each of which plays a crucial role in laying the foundation for the development of specialized tissues and organs. The understanding of embryogenesis holds significant scientific relevance, offering insights into developmental disorders or regenerative medicine. This section has been constructed as a compendium of different sources (Gilbert, 2000; Gilbert & Raunio, 1997; Mueller et al., 2015; Wolpert et al., 2019). INTRODUCTION 6 Figure 1. Embryogenesis of an organism A) Examples of different types of cleavages. The yellow regions represent a higher concentration of yolk compared to the blue regions, where the yolk concentration is either lower or absent. B) Life cycle of a frog as a general example of the processes that occur from a zygote to adulthood. Adapted from Gilbert, 2000; Mueller et al., 2015. INTRODUCTION 7 1.1.2. Mitosis in animal cells The cell cycle plays a vital role in embryogenesis, precisely regulating cell division and growth to ensure the proper formation and organization of tissues and organs during embryonic development (Hardwick et al., 2018) The cell cycle is divided into two main stages (Fig. 2A): interphase, where cells grow, and mitosis, where they divide. Throughout the interphase, the cell is constantly monitoring its internal and external environment to ensure genetic stability during embryogenesis. The proper cell cycle progression and accurate DNA replication prevent the transmission of genetic abnormalities and promote normal development. Meanwhile, mitosis consists of the splitting of the previous duplicated DNA into two identical sets of chromosomes (karyokinesis) and ends with the distribution of the cytoplasm between the daughter cells (cytokinesis). In animals, the separation of the cytoplasm, known as cytokinesis, consist in the formation of a contractile ring composed of actin filaments and myosin-II. This contractile ring, situated at the centre between the daughter cells, constricts the plasma membrane leading to the formation of a cleavage furrow. In most animals, the constriction eventually leads to the formation of a narrow portion that links both cytoplasms, known as the midbody. The microtubules in the midbody slowly disassemble as the constriction further continues until total abscission. At this moment, the separation into two cells is complete (Alberts et al., 2013; Pollard et al., 2017; Ross & Pawlina, 2006; Yanagida, 2014). Additionally, the cell cycle is tightly linked to cell fate decisions. As cells progress through the cell cycle, they undergo specific molecular changes that determine their fate and differentiation potential. The duration of specific cell cycle phases can influence cell fate decisions, determining whether a cell will continue to proliferate, undergo differentiation, or enter a dormant state. In addition, the cell cycle facilitates the rapid proliferation of cells during embryonic development ensuring that there are enough cells available to differentiate into the various cell types and tissues necessary for the formation of complex structures (Alberts et al., 2013; Dalton, 2015; Pollard et al., 2017). Overall, the cell cycle is essential for orchestrating the dynamic processes that underlie the intricate and precise development of embryos. Its precise regulation ensures the orientation and rate of these cell divisions, which can result in significant changes that control tissue size and shape (Lecuit & Le Goff, 2007). INTRODUCTION 8 1.1.3. Epithelial morphogenesis During the development of functional tissues and organs, morphogenesis comprises a series of biological processes that establish the complex anatomical architecture and functional diversity observed in living organisms (Lecuit & Le Goff, 2007). In animal embryogenesis, epithelial cells play a crucial role (L. A. Davidson, 2012). Epithelia are characterized by tight bonds leaving minimal intercellular space between the epithelial cells. In some cases, epithelia consist of numerous layers of cells, forming a stratified structure, as seen in the epidermis (the outer layer of the skin). Alternatively, they may be a simple epithelium, formed by only a single cellular sheet. The epithelial cells can have various shapes, ranging from tall and columnar to cuboidal, or flat and squamous. In addition, within a specific layer, cells may consist of a uniform type or a combination of different types (Ross & Pawlina, 2006, Welsch et al., 2014). The functions of epithelia are very diverse: Some mainly act as protective barriers, while others have complex biochemical functions. Certain types also secrete specialized products like hormones, while others are involved in detecting diverse signals (Alberts et al., 2013, Welsch et al., 2014). Despite the variations, a standard set of structural features is shared by mostly all epithelia. One of them is the polarity that involves the presence of regions with specific structural and functional properties (Fig. 2B). The apical surface is often exposed to the external environment or a watery fluid, while the basal surface is generally connected to a layer of connective tissue known as the basal lamina (Alberts et al., 2013; Gilbert & Rice, 2016; Ross & Pawlina, 2006). Other common characteristics shared by epithelial tissues include the functions of cell-cell and cellmatrix adhesion, which enable cells to attach both to each other and to the basal lamina. The cell adhesion is facilitated by various types of junctions, such as the adherens junctions (Ross & Pawlina, 2006, Welsch et al., 2014). Morphogenesis underscores the dynamic nature of epithelia, wherein cells proliferate and simultaneously undergo changes in shape and organization to fulfil their functions (Guillot & Lecuit, 2013). During the onset of mitotic cycles, epithelial polarized cells typically increase their roundness (mitotic rounding), resulting in a temporary disengagement and deformation. Subsequently, the original shape is regained during the next interphase (Fig. 2C) (Godard et al., 2020; Ragkousi et al., 2017). Epithelial sheets, and the way their cellular components assemble, are essential during organogenesis, as they give rise to cavities or tubular complexes that are present in fully developed organs (Guillot & Lecuit, 2013). These tightly regulated processes are under the control of the acto-myosin complex and cell adhesion, involving a range of INTRODUCTION 9 interconnected events. This includes cell migration, rearrangements, modifications in cellular shape, changes in orientation and frequency of cell divisions, growth, and apoptosis (Lecuit & Le Goff, 2007; Munjal & Lecuit, 2014). Understanding how these different processes are coordinated and how they collectively contribute to organogenesis is a central area of study in developmental biology. The use of multidisciplinary approaches and specific model organisms pave the way to untangle the mechanisms of the dynamics of epithelial morphogenesis. Figure 2. Cell cycle in epithelial cells A) Scheme of the cell cycle where the coloured arrows indicate the progression through the different stages in a clockwise direction. B) Representation of a group of epithelial cells. The scheme shows the apical, lateral, and basal sides, the basal lamina, and the junctions between the polarized cells. C) Illustration of a cell division event in an epithelial layer. The cell increases in roundness upon entering mitosis. Then, after cytokinesis, the polarity and original shape are restored. Adapted from Alberts et al., 2013; Sorce et al., 2015. INTRODUCTION 10 1.2. Echinoderms as animal models for the study of embryonic processes Echinoderms have been used extensively as animal models in developmental biology (Briggs & Wessel, 2006). These organisms have provided fundamental insights into the cellular and molecular mechanisms involved in development, including the study of the genes that are expressed during development (E. H. Davidson & Erwin, 2006), signalling pathways controlling cell differentiation (Hörstadius, 1973; McClay, 2011) and cell-cycle regulation (Dorée & Hunt, 2002). From a practical standpoint, echinoderms are relatively easy to manipulate in the laboratory, have a relatively short generation time, and produce optically clear embryos that facilitate the use of live imaging techniques. They also have a unique larval stage characterized by bilateral symmetry and ciliated bands, enabling them to swim and feed. However, they eventually metamorphose into juvenile echinoderms with radial symmetry, developing the characteristic five-fold symmetry of adult echinoderms (Arnone et al., 2015). Most of the studies were previously performed almost exclusively in the sea urchin embryos (Lin & Su, 2016; McClay, 2011; Yanagida, 2014) such as Strongylocentrotus purpuratus (S, purpuratus). Despite this, various experimentally tractable echinoderm species display intriguing variations in essential aspects of morphogenesis (Barone et al., 2022; Poon et al., 2019), such as the formation of blastula and embryo compaction (i.e. the process of tissue sealing without leaving gaps between cells). Recent advances in understanding these variations are allowing us to unravel the processes behind tissue formation in echinoderms. Moreover, these advancements enable us to identify distinct dynamic cellular behaviours, such as changes in cell shape and cell adhesion, which are more likely to drive the evolution of novel and complex embryonic shapes (Barone et al., 2022; McCauley et al., 2012). 1.2.1. Comparative development of sea stars and sea urchins Alternatives to sea urchins have been studied within the Echinoderm phylum, including sea stars and sea cucumbers (Fig. 3A). Among the sea stars, the bat star Patiria miniata (P. miniata) has stood out in recent years for the analysis of essential aspects of embryonic development such as cell division or cell differentiation, particularly when compared to sea urchins (Barone & Lyons, 2022; Cary & Hinman, 2017; Poon et al., 2019). INTRODUCTION 11 Although the early development of sea stars and sea urchins shares many similarities, there are distinct differences in the shapes of their cells and embryos. Figure 3. Echinoderms development A) Representative classes of echinoderms. From left to right: sea star P. miniata, sea urchin S. purpuratus, and sea cucumber Parastichopus parvimensis. B) Scheme of fertilization in echinoderms. Sea star and sea INTRODUCTION 18 assumed that cells have a prismatic shape, and that curvature is achieved by prisms turning into frusta (i.e. truncated pyramids) (L. A. Davidson, 2012; Pearl et al., 2017). Cells that are prisms or frusta have the same set of neighbours on their apical and basal sides (Fig. 6A), but this is not always the case, as apico-basal intercalations can occur (Fig. 6B). Apico-basal intercalations refer to neighbours exchanges that develop not in time, but in space, along the apico-basal axis of a cell - also known as T1-space transitions (AB-T1s) (Gómez-Gálvez et al., 2021a; Rupprecht et al., 2017; Sanchez- Corrales et al., 2018). Whenever an AB-T1 occurs, the four cells involved in the transition are no longer shaped as prisms or frusta but as scutoids (Fig. 6A, B) (Gomez-Galvez et al., 2018). AB-T1s occur frequently in curved epithelia (Honda et al., 2008; Sun et al., 2017; Xu et al., 2016), especially in tubular monolayered epithelia (Gómez-Gálvez et al., 2022), allowing energetically favourable packing of cells in 3D (Gómez-Gálvez et al., 2018; Mughal et al., 2018). Figure 6. 3D packing in curved epithelia A) Illustrations of a planar epithelium where cells are represented as prismatic columns (left), a curved epithelia where the cells adopt the shape of truncated pyramids (centre), and a tubular epithelium where the cells acquire the scutoid conformation (right). B) Diagram showing an AB-T1 where cells exchange neighbours along the apico-basal axis. The green cells are in contact only on the apical surface, while, on the basal surface, only the blue cells are contacting. Adapted from Gómez-Gálvez et al., 2021a. Based on this principle of energy minimization, a model was developed (Gómez- Gálvez et al., 2018) to understand the 3D packing of cells in epithelial tissues (Fig. 7A). This model considers the relationship between the radii of curvature along the two main axes (transversal and longitudinal) along the apical (Ra) and basal (Rb) surfaces, and the INTRODUCTION 19 anisotropy of the surface ratio (Rb/Ra) that determines the shape of the tissue. AB-T1s are expected to form in any epithelium with anisotropic curvatures, e.g., ovoids and tubes (Gómez-Gálvez et al., 2018; Mughal et al., 2018) where the more prolate the shape, the proportion of scutoids is higher. In contrast, AB-T1s are not expected in spherical epithelia, where the curvature is isotropic (Gómez-Gálvez et al., 2018). Computational models have demonstrated agreement with measurements taken from biological samples under stationary conditions, especially in tubular shapes as the third instar larva salivary gland of Drosophila. In these cases, where one of the curvatures is null, there is a logistic relationship between the surface ratio and the 3D connectivity (Fig. 7B) (Gómez-Gálvez et al., 2022). However, despite those predictions, the scenario changes when dynamics are introduced, such as in the case of Drosophila egg chambers (Gómez-Gálvez et al., 2018). AB-T1s have in fact been observed in curved epithelia without pronounced anisotropy and in higher frequencies than expected considering only the tissue geometry (Gómez-Gálvez et al., 2018; Lou et al., 2023; Rupprecht et al., 2017) Therefore, it has been proposed that scutoid formation may also result from cell and tissue dynamics that produce different degrees of deformation on the apical and basal surfaces of epithelial tissues (Gómez-Gálvez et al., 2018; Lou et al., 2023; Rupprecht et al., 2017). For example, in the case of proliferation, cell divisions may temporarily alter the balance of forces acting on epithelial cells (Gómez et al., 2021; Ragkousi & Gibson, 2014; Ragkousi et al., 2017). Similarly, the deformations experienced by tissues undergoing morphogenesis may be different from those predicted according to geometry alone, due, for instance, to active cell movements within the tissue (Gómez-Gálvez et al., 2018; Sanchez-Corrales et al., 2018; Sun et al., 2017) or to pressures from outer structures (Lou et al., 2023; Rupprecht et al., 2017). In conclusion, determining whether dynamic processes underlie scutoid formation requires following those processes over time to observe how the cells change in relation to other morphogenetic events such as cell division, tissue rearrangements or cell deformation. INTRODUCTION 20 Figure 7. Correlation between curvature anisotropy and 3D packing A) Graphic summary showing how cells pack within tissues with different geometries. These tissues have two main axes of curvature, transversal (h) and longitudinal (w), where R represents the radius of curvature along each axis. This enables the assessment of surface ratio anisotropy for different geometries, quantified by the relative change of aspect ratio from basal to apical (colour code): (𝑅𝑏 ℎ/𝑅𝑎 ℎ)/ (𝑅𝑏 𝑤/𝑅𝑎 𝑤)−1. In all the representations of geometric shapes, basal is at the top (dark blue) and apical is at the bottom (light blue). The black dashed arrow in the main diagonal indicates the epithelium without anisotropy (ratio = 0). B) 3D rendering of a representative segmented salivary gland and 3D Voronoi tube (top). Quantitative principle that represents how the increment of surface ratio, (i.e., apico-basal coordinate) in tubular epithelia, leads to a logistic increment in the 3D connectivity (bottom). The colour gradient ranges from light blue at the apical surface to dark blue at the basal surface. Adapted from Gómez-Gálvez et al., 2022, 2018. INTRODUCTION 21 Objectives OBJECTIVES 23 OBJECTIVES 24 Previous studies have proposed that several geometric and biophysical factors have been associated with scutoid formation. However, the dynamic changes underlying their appearance and the subsequent reorganisation of 3D epithelial packing remain poorly understood. To solve this problem, this thesis attempts to unravel the biomechanical mechanisms that regulate cell packing in dynamic conditions. For that purpose, the objectives are organized as follows: ➢ Develop an optimal segmentation method for efficiently processing early sea star imaged embryos. ➢ Dissect the contributions of tissue compaction, cell density, and cell proliferation to the formation of AB-T1s across three different scales: global, regional, local. ➢ Analyse the effects on the dynamics of epithelial organization after applying physical compression to sea star embryos. ➢ Explore potential regional variations in compressed and wildtype embryos to determine which intrinsic factors may contribute to scutoid formation. ➢ Develop computational models for comparing theoretical 3D packing with the experimental results. ➢ Track the presence of scutoids over consecutive cellular cycles to analyse the influence of cell proliferation on AB-T1 formation. OBJECTIVES 25 Materials and methods MATERIALS AND METHODS 27 MATERIALS AND METHODS 34 Note that custom made MATLAB scripts were used to extract the following characteristics (Fig. 10, Table S1-2): ▪ Number of cells: selected cells whose centroid lay within 30 μm from the embryo surface closest to the objective. ▪ Average surface ratio anisotropy (or curvature anisotropy): proportional relationship between the radii of curvature along the two main axes, h and w (transversal and longitudinal respectively) of the apical (Ra) and basal (Rb) surfaces. This relationship is characterised by the differences between the two surface ratios (Gómez-Gálvez et al., 2018). The formula to calculate the surface ratio anisotropy was as follows: Where ℎ represents the axis of greatest curvature, 𝑤 represents the axis of least curvature, Ra, refers to the outer apical radius and Rb is the inner basal radius. To obtain the value of these parameters, we measured the two principal curvatures for each centroid of the selected cells in the segmented region of both surfaces. By using the formulas to calculate the curvature of a coordinate in an ellipsoid (Bektas, 2017), we obtained the values for the principal curvatures (kh, kw) and we computed the average value regarding the maximum and minimum radii of curvature (R= 1/k) for both apical and basal surfaces. A surface ratio anisotropy of 0 indicates a tissue with an isotropic curvature, while values greater than 0 suggest a tissue with more anisotropic curvature. Note that for this feature, the samples segmented were all the 6 WT embryos and 5 out of the 6 WT-comp embryos (embryos 7,8,9,11,12). ▪ Proportion of scutoids: frequency of “scutoidal” cells among the selected cells. We quantified the cells that acquire the scutoid conformation, marking the cells involved in AB-T1s, i.e., cells exchanging neighbours between apical and basal surfaces. ▪ Average cell volume (µm3): mean volume of individual cells. The volume was determined by counting the number of voxels belonging to segmented cells. ▪ Average cell height (µm): mean cell height of individual cells. Cell height was measured by calculating the Euclidean distance between the centroids of the basal and apical cell surfaces. MATERIALS AND METHODS 35 ▪ Average cell convexity ratio: mean ratio of the volume of the cell divided by the volume of the convex hull. The ratio ranges from 0 to 1, with values closer to 1 indicating a highly convex cell, and values closer to 0 indicating the opposite. Loss of convexity is likely attributable to how tightly cells are packed together (Fig. S2) and indicates increased cell compaction. 3. 3.2. Tissue measurements Besides the features at cellular level, we also quantified global measurements (Fig. 10, Tables S1-3) by making use of the segmented region of the embryos: ▪ Cell density of the region (cells/µm2): ratio between the selected cells and the total basal area they occupied. • Surface ratio of the region: proportional relationship between the areas of the outer and inner surfaces. It is a measure of how large the relative expansion of the outer surface (apical in this case) is with respect to the inner surface (basal) (Gómez- Gálvez et al., 2018). Higher the values, greater is the differences of areas between surfaces and the tissue is thicker. • Opening areas (µm2): regions of the embryo not occupied by the cells. Once we extracted the full projection of the image stacks, we used FIJI's polygon selection tool to quantify the sum of all opening areas in each sample. Note that for the following 3 characteristics, the samples segmented were all 6 WT embryos and 5 out of the 6 WT-comp embryos (embryos 7,8,9,11,12). • Embryo lengths (µm): dimensions of the three axes of symmetry of the embryo. Since in all cases the embryo surface closest to the objective is either animal or vegetal pole, we measured half of the animal-vegetal axis length. This was done by determining the distance from the surface closest to the objective to the half embryo using FIJI's orthogonal view and straight-line tool. The other two axes of the embryo were measured on the 2D slice that approximately represented this midpoint of the embryo using FIJI's polygon selection tool. MATERIALS AND METHODS 36 Figure 10. Description of tissue and cellular features Schemes and definitions of the different features extracted from the samples. ▪ Tissue aspect ratio: proportional relationship between the longest and shortest dimensions of a tissue. It can be defined as: 𝑎𝑠𝑝𝑒𝑐𝑡 𝑟𝑎𝑡𝑖𝑜 = 𝐿𝑒𝑛𝑔𝑡ℎ 𝑜𝑓 𝐿𝑜𝑛𝑔𝑒𝑠𝑡 𝐷𝑖𝑚𝑒𝑛𝑠𝑖𝑜𝑛 𝐿𝑒𝑛𝑔𝑡ℎ 𝑜𝑓 𝑆ℎ𝑜𝑟𝑡𝑒𝑠𝑡 𝐷𝑖𝑚𝑒𝑛𝑠𝑖𝑜𝑛 ⁄ MATERIALS AND METHODS 37 The aspect ratio provides information about the shape and elongation of the tissue. A value of 1 indicates the tissue is perfectly isotropic or symmetrical, while values greater than 1 suggest an elongation along the longest dimension. • Major axes length ratio: proportional relationship between the two longest dimensions of the tissue. It provides information about the relative elongation of the tissue along these two principal axes. A value of 1 indicates that the two longest dimensions of the tissue are similar in length, suggesting that the tissue is relatively symmetrical along the third axis. 3. 4. Voronoi constructions We have used MATLAB R2021a (MathWorks) as our computational tool to generate Voronoi constructions based on each imaged sea star embryo, maintaining similar geometrical characteristics of WT (n=150 time points, 6 embryos) and WT-comp datasets (n=150 time points, 6 embryos). These constructions were obtained by applying a 3D Voronoi algorithm to correctly mimic the epithelial packing (Voronoi 1908; Honda 1978). More specifically, we extracted the regions of the tissue that had been segmented from each imaged embryo in the form of binarized masks. These masks defined the bounding territory that the Voronoi cells could occupy in each construction. Then, we calculated the coordinates of the 3D centroids of all segmented cells which were then used as Voronoi seeds. The algorithm is based on tiling the space between these sets of Voronoi seeds by proximity, occupying the entirely binarized mask given in each construction (H. Honda, 1978; Voronoi, 1908) Once the Voronoi construction was generated, similar to our approach done with the embryos, we selected a subset of Voronoi cells for further analysis, i.e. the cells whose centroid lay within 30 μm from the region that would correspond to the embryo surface closest to the objective. 3. 5. Delimitation of outer and inner layers and automatic detection of scutoids from segmented regions The extraction of the outer and inner layers from the cellular segmented regions was performed using a custom MATLAB code (getInnerOuterLateralFromEmbryos), traversing the segmented region along the Z-axis from top to bottom (outer) and bottom to top (inner) selecting the pixels corresponding to the first labelled cells encountered in both scans. MATERIALS AND METHODS 38 The method for extracting the proportion of scutoids from the Voronoi constructions was adapted from CartoCell (Andrés-San Román et al., 2023). After extracting the outer and inner layers, we quantified the number of neighbours for each cell on both surfaces. To identify neighbours, we dilated each cellular surface, identifying those labelled cells which overlapped with the dilation, thereby obtaining the sets of neighbours for a specific cell on each layer. Finally, we quantified the proportion of scutoids as the relative number of cells that have different sets of neighbours on their apical and basal sides, as this situation can occur only in the case of an AB-T1 (Table S1). 3. 6. Tracking of scutoids Scutoids were manually tracked using Fiji/ImageJ (Schindelin et al., 2012). Interphase for each cell forming a scutoid was defined as the time between the end of cytokinesis and cell rounding, which marks the beginning of the next cell division (Fig. S3A). The time at which a cell adopted the “scutoidal” shape (scutoid onset) and for how long the cell maintained the “scutoidal” shape (scutoid duration) was recorded and normalised over interphase duration (Tables S4-5). Then, we classify scutoids into separate categories based on scutoid start times. Scutoids with onsets lower than 0.15 were grouped under "after mitosis", while the rest were grouped under “independent of mitosis” category (Table S5). 3. 7. Normalization of developmental time To compare the duration of mitotic waves and dynamics of epithelial closure across embryos, developmental time was normalized by the time elapsed between the beginning of the 128-cell stage (relative time 0) and the end of 512-cell stage (relative time 1). Relative time 0 was defined as the time when the first cell of the 64-cell stage embryo had divided and relative time 1 as the time when the first cell of the 512-cell embryo had divided (Table S6). Each stage was then split into a “mitotic wave” period, which ends when at least 50% of the cells have divided, and an interphase period, which ends when the first cell generated by the previous round of cell division divides again (Fig. S3B). 3. 8. Statistical analysis Statistical analyses were performed using software GraphPad Prism v.9.0.0. Shapiro- Wilk test was applied for normality determined our use of either standard Student t-test, ordinary one-way or two-way ANOVA tests (normally distributed data, equal variances) or non-parametric U-Mann-Whitney and Kruskal-Wallis tests (not normally distributed MATERIALS AND METHODS 39 data). Dunn multiple comparisons correction was used to compare most of the features over time (Fig. 12A-B, D-H, Fig. 14, Fig. 15, Fig. 16B-C, Fig. 19B-D, F, Fig. 21B-F, Fig. 22 and Fig. 23B-D). Bonferroni multiple comparisons correction was used to compare most of the features between WT and WT-comp conditions (Fig. 14, Fig. 15, Fig. 16B- C, Fig. 17B and Fig. 23B-D) or between animal and vegetal locations (Fig. 19B-F, Fig. 21B-F and Fig. 22). Tukey multiple comparisons correction was used to compare WT surface ratio anisotropy over time (Fig. 12C, Fig. 14C and Fig. 19E) and WT vegetal surface ratio and cell height over time (Fig. 19F, Fig. 22C). Student t-test was used to compare the average time elapsed between the beginning of the 128-cell stage and the end of 512-cell stage (Fig. 17A) and scutoids onsets after mitosis and independent of mitosis (Fig. 25A). Mann-Whitney tests were used to compare the average scutoids onset and duration (Fig. 25B-C). The different statistics used for each feature were shown in Table S1, S2 and S4. 3. 9. Data availability and graphic illustrations All codes utilized in the analyses are available in the public repository of the lab: https://github.com/ComplexOrganizationOfLivingMatter/seaStarProcessingSegmentatio n. The graphical representations of the data were created using GraphPad Prism v.9.0.0. The different panels of each figure were edited and combined using Adobe Photoshop CS6. MATERIALS AND METHODS 40 Results RESULTS 42 RESULTS 43 4. 1. The 3D epithelial packing changes during development in proliferative tissues To identify factors that couple cell proliferation and epithelial packing, we analysed the shape and 3D connectivity of cells during sea star (P. miniata) embryo development. After the initial rounds of cell division, this embryo develops into an approximately spherical blastula with a blastocoel encircled by a monolayered epithelium (Arnone et al., 2015; Barone & Lyons, 2022; Kominami, 1983). However, early blastomeres adhere loosely to one another in the sea star embryo. There are intercellular openings through which fluid can flow between the inside and outside of the embryo until about the 512- cell stage. Then, the epithelium closes to encircle the blastocoel (Barone et al., 2022; Barone & Lyons, 2022; Dan-Sohkawa & Satoh, 1978) (Fig. 11A, Fig. S1, Movie S1). After 512-cell stage, the cell divisions begin to be desynchronized until hatching (Barone & Lyons, 2022; Dan-Sohkawa & Satoh, 1978). Meanwhile, the fertilisation envelope forces the embryo to maintain a constant volume. As cells undergo several rounds of cell division, their volume is reduced changing their shape and cell-cell contact topology (3D packing) along the lateral membrane (Farhadifar et al., 2007; Lemke & Nelson, 2021; Nelson, 2016). Therefore, during the blastulation, many of the cellular processes characteristic of a dynamic epithelia can be readily observed. To determine how cellular characteristics vary during development we analysed timelapse imaging of wildtype (WT) sea star embryos expressing a membrane marker (mYFP) (Fig. 11A, Fig. S1) and a nuclear marker (H2B-CFP) (Movie S1). The movies were performed by our collaborators Vanessa Barone and Deidre Lyons (see materials and methods). Then, we processed the images with a deep learning-based 3D segmentation program developed in our group (Fig. 8, Fig. 9A, Fig. 11A, see materials and methods). To ensure accurate segmentation we limited our analysis to the portion of the imaged embryos with the highest signal/noise ratio (Fig. 9A, Fig. 11A, see materials and methods). The presence of "scutoidal" cells serves as a valuable indicator of changes in 3D packing, as scutoids are formed whenever an AB-T1 event occurs, resulting in an increase in 3D connectivity (Gómez-Gálvez et al., 2022, 2018; Okuda et al., 2019). Therefore, we used the processed images to identify whether scutoids are formed in this dynamic epithelial tissue, when they are formed, and which cell behaviours may be related to scutoid formation. RESULTS 50 Statistical tests: Mann-Whitney tests with Bonferroni multiple comparisons correction (black). Kruskal- Wallis tests with Dunn multiple comparisons correction (blue), except in 14C where one-way ANOVA test with Tukey multiple comparison correction was used (light blue). ns: non-significant; *: p-value < 0.05; **: p-value <0.01; ***: p-value <0.001. The comparison between the two conditions at each developmental stage highlighted interesting differences. First, we observed that the number of cells selected per embryo was greater in the WT-comp condition (Fig 14A). Furthermore, compression induced a higher proportion of AB-T1s (Fig. 14B) and their appearance at an earlier stage (128- cell in WT-comp vs 256-cell in WT). Concomitantly, WT-comp embryos showed higher surface ratio anisotropy (Fig. 14C), higher cell density (Fig. 14D) and higher surface ratio (Fig. 14F), but lower convexity ratio (Fig. 14E) than WT embryos. Additionally, the compressed embryos exhibited higher cell heights (Fig. 15B), while the cell volumes remained unchanged by the compression (Fig. 15A). Taken together, the experimental results suggested that either tissue curvature or cell size was insufficient to explain the dynamics of scutoid formation. On the other hand, it was best explained by the increase in cell density and tissue compaction. Figure 15. Size analysis at cellular level over time between WT and WT-comp embryos Quantifications of average cell volume (A) and average cell height (B). WT: n=150 time points, 6 embryos, 4 experiments; WT-comp: n=150 time points, 6 embryos, 5 experiments. Statistical tests: Mann-Whitney tests with Bonferroni multiple comparisons correction (black) and Kruskal-Wallis tests with Dunn multiple comparisons correction (blue). ns: non-significant; *: p-value <0.05; ***: p-value <0.001. RESULTS 51 4. 3. Analysis of the influence of tissue geometry and cell proliferation rate on cell packing Spatial constraints can potentially result in variations in tissue packing associated with tissue geometry or fluctuations in the cell proliferation rate. We explored whether these two factors could have an impact on the formation of scutoids in our system. First, using our imaging data, we were able to estimate the dimensions of the embryo (Table S3, see materials and methods), allowing us to explore the effect of tissue geometry on observed changes in cell packing. After obtaining the measurements, a morphological classification of the embryos was carried out based on the length of the three axes of symmetry. Then, four shape categories were established: spheres, prolate spheroids, oblate spheroids, and ellipsoids. We considered axes of symmetry to be similar when the difference in their lengths was less than 8%. Thus, we classified embryo as spheres when their three axes had similar lengths. On the other hand, when 2 of the 3 axes of symmetry were similar, but the other was different, we classified the embryos as spheroids. Spheroids were labelled as prolate when the differing axis was the major axis (similar in shape to a rugby ball), and oblate when it was the minor axis (like the shape of a tangerine). Ellipsoids were those embryos where all three axes were different. Our analysis showed that most WT embryos were classified as oblate spheroids after estimated their dimensions at the three cellular stages (128-, 256- and 512-cell). In all cases, the minor axis was the animal-vegetal (Fig. 16A, Table S3). To analyse changes in shapes over time, we compared the tissue aspect ratio and the length ratio between the two longest dimensions (see materials and methods). The shape of WT embryos did not change significantly during the developmental stages we analysed, as indicated by the consistent tissue aspect ratio over time (Fig. 16B). However, the lengths of the major axes became increasingly similar, approaching a ratio close to 1 at 256-cell stage (Fig. 16C). In contrast, WT-comp embryos were classified into more categories than WT: ellipsoids, prolate and oblate spheroids. However, like WT embryos, in all cases the minor axis was the animal-vegetal axis (Table S3) and, in most embryos, the same shape is maintained over time. As a result, there were no significant changes in the aspect ratio and the major axis length ratio at each subsequent stage (Fig. 16B-C). When we compared the relation between the embryo lengths in both conditions, we observed that there is a decreased ratio of major axis lengths and an increase in tissue aspect ratio compared to WT embryos (Fig. 16B-C). These findings suggested that the RESULTS 52 compression modified the tissue geometry leading to a variety of shapes, with some embryos becoming ellipsoids or even prolate spheroids, and this alteration was consistent over time. Therefore, these changes were responsible of the overall increase observed in the curvature anisotropy of WT-comp embryos (Fig. 14C). Figure 16. Shape analysis of WT and WT-comp embryos A) Orthogonal view (top) and maximum projection (centre) of a representative WT half embryo. Estimation of the complete shape (bottom). Scale bar, 50 μm. B-C) Quantifications of tissue aspect ratio (B) and major axes length ratio (C) of the whole embryos at 128-, 256- and 512-cell stages. WT: n=150 time points, 6 embryos, 4 experiments. WT-comp: n=125 time points, 5 embryos, 4 experiments. Statistical tests: Mann-Whitney tests with Bonferroni multiple comparisons correction (black) and Kruskal-Wallis tests with Dunn multiple comparisons correction (blue). ns: non-significant; **: p-value <0.01 ***: p-value <0.001. RESULTS 53 Regarding the proliferation rate, we conducted an analysis to determine whether compression had any effect on the rate of cell divisions that could influence the increasing cell density. For that purpose, we quantified the elapsed time between the stages analysed, i.e., from the beginning of 128-cell stage to the end of 512-stage (see materials and methods) at two scales: 1) comparing the entire time elapsed (Fig. 17A), and 2) comparing the duration of each consecutive interval of synchronous mitosis and interphase (Fig. 17B, Fig. S3). Thus, we were capable for searching local fluctuations on the proliferation rate. However, the differences observed between the conditions were not statistically significant in any case (Fig. 17A-B), leading to the conclusion that the compression did not accelerate or slow down the moment when the cells divide during embryo development. In conclusion, both the consistent tissue geometry over time and the steady cell proliferation rate across embryos were insufficient to explain the dynamic changes in the proportion of scutoids. Figure 17. Analysis of cell proliferation A) Quantification of the average time elapsed between the beginning of the 128-cell stage and the end of 512-cell stage. B) Quantification of the average time elapsed among each interphase and mitosis intervals WT: n=6 embryos, 4 experiments. WT-comp: n=6 embryos, 5 experiments. Statistical test: Student t test. ns: non-significant. RESULTS 54 4. 4. Regional differences in tissue compaction affect propensity of scutoid formation During blastulation, in the sea star embryo, several openings located in different regions close progressively (Fig. 18, Fig. S1, Movie S1, Movie S3) (Barone and Lyons 2022). To investigate the relation of this phenomenon with the tissue packing, we conducted a more detailed analysis by separately examining time-lapses where either the vegetal or the animal pole was imaged (Fig. 18, Movie S1, Movie S3, see materials and methods). We observed that closure was delayed on the animal pole: all openings are closed in vegetal poles of WT embryos by the 256-cell stage whereas they only close at the 512-cell stage in animal poles (Fig. 19A). Figure 18. Polarization of embryonic poles in WT condition A) Schematic representation of WT embryos highlighting animal and vegetal poles. B) Maximum projections of a representative animal pole and vegetal pole at 128- , 256- and 512-cell stages. At 128-cell stage, the two polar bodies can be clearly observed in the central opening area of the embryo with an animal orientation (top left). Scale bars, 50 μm. We then asked whether this asynchronous closure could result in local differences in cell density and compaction that may induce scutoid formation. When we measured cell density, we found that it was not significantly different between animal and vegetal poles (Fig. 19B). However, when using cell convexity as an indicator of compaction, we found that vegetal cells had lower convexity than animal cells (Fig. 19C). Interestingly, comparing the percentage of scutoids between both poles, we observed that they appeared earlier in the vegetal pole (256-cell stage in vegetal vs 512-cell stage in animal) and with a higher proportion (Fig. 19D). RESULTS 55 Figure 19. Asynchronous compaction and scutoid formation in WT embryos Quantifications of opening areas (A), cell density (B), average cell convexity ratio (C), proportion of scutoids (D), average surface ratio anisotropy (E), and surface ratio (F) at 128-, 256- and 512-cell stages. WT animal: n= 75 time points, 3 embryos, 3 experiments. WT vegetal: n= 75 time points, 3 embryos, 2 experiments. RESULTS 56 Statistical tests: Mann-Whitney tests with Bonferroni multiple comparisons correction (black) except in 19E, where two-way ANOVA test with Bonferroni multiple comparisons correction was used. Kruskal- Wallis tests with Dunn multiple comparisons correction (orange and green), except in 19E (orange and green) and 19F (green), where one-way ANOVA tests with Tukey multiple comparisons correction were used. ns: non-significant; *: p-value < 0.05; **: p-value <0.01; ***: p-value <0.001. In contrast, while we observed differences in surface ratio anisotropy between animal and vegetal cells in a subset of the analysed stages (Fig. 19E), they did not correlate consistently with changes in the proportion of scutoids. Meanwhile, we did not find significant differences in the tissue thickness (surface ratio) between both sides (Fig. 19F). Moreover, in both animal and vegetal poles, the trends observed over time mimic those described in the combined WT population (Fig. 12B-D, G): density doubled (Fig. 19B), convexity reduced (Fig. 19C), and scutoids notably increased, particularly at the 512-cell stage (Fig. 19D). Additionally, surface ratio anisotropy showed minimal change over time (Fig. 19E). On the other hand, surface ratio on the animal side did not decrease at the 512-cell stage in contrast to the vegetal side (Fig. 19F) and the combined WT population (Fig. 12I). These results suggested that, even though cell density remained similar between the poles, the presence of an opening allowed animal cells to maintain slightly less compactness for longer duration, thereby delaying the formation of AB-T1s compared to the vegetal pole. In this scenario, the differences in the propensities for scutoid formation between animal and vegetal cells would be lost if the opening in the animal pole would be closed earlier. To test this hypothesis, we analysed animal and vegetal poles in compressed embryos (Fig. 20, Movie S2, S4). We found that closure happens at the same time in animal and vegetal poles of WT- comp embryos, with openings being closed in both poles before the 256-cell stage (Fig. 21A). Cell density was similar between the two poles until the 512-cell stage, when it was higher in the vegetal pole compared to the animal pole (Fig. 21B). Moreover, there was no significant difference in convexity between animal and vegetal cells (Fig. 21C). Interestingly, there were no differences in the proportion of animal and vegetal cells acquiring the “scutoidal” shape at the 128- and 256-cell stages (Fig. 21D), when both cell convexity and cell density were similar. RESULTS 57 At the 512-cell stage, however, when cell convexity was not significantly different, but cell density was higher in the vegetal pole, a higher proportion of cells formed scutoids in the vegetal pole compared to the animal pole (Fig. 21D). Like WT unperturbed embryos, we observed differences in surface ratio anisotropy (Fig. 21E), but surface ratio did not vary significantly between animal and vegetal regions (Fig. 21F) indicating that these differences did not consistently correlate with changes in the proportion of scutoids. Figure 20. Polarization of embryonic poles in WT-comp condition A) Schematic representation of WT-comp embryos highlighting animal and vegetal poles. B) Maximum projections of a representative animal pole and vegetal pole at 128- , 256- and 512-cell stages. The two polar bodies can be clearly observed in the centre of the embryo with an animal orientation (top). Scale bars, 50 μm. In WT-comp embryos, the temporal variations in the features analysed differed between animal and vegetal poles. On the animal side, the proportion of scutoids, the surface ratio anisotropy and the surface ratio remained relatively constant across subsequent stages (Fig. 21D-F). Meanwhile, on the vegetal side, these features changed over time (Fig. 21D-F). However, in both regions, cell density increased, and cell convexity decreased throughout the developmental stages analysed (Fig. 21B-C). In summary, these results suggested that differences in compaction between regions of the embryo explained variation in the propensities for scutoid formation. Moreover, they showed that compaction and cell density can independently affect 3D packing, as differences in the proportion of scutoids between regions with similar cell density could be explained by variation in compaction (WT) and differences between similarly compact regions could be explained by variation in cell density (WT-comp). RESULTS 58 Figure 21. Synchronous compaction and scutoid formation in WT-comp embryo A) Quantifications of opening areas (A), cell density (B), average cell convexity ratio (C), proportion of scutoids (D), average surface ratio anisotropy (E), and surface ratio (F) at 128-, 256- and 512-cell stages. WT-comp animal: n= 75 time points, 3 embryos, 2 experiments. WT-comp vegetal: n= 75 time points, 3 embryos, 3 experiments except in 21E where n= 50 time points, 2 embryos, 2 experiments Statistical tests: Mann-Whitney tests with Bonferroni multiple comparisons correction (black) and Kruskal-Wallis tests with Dunn multiple comparisons correction (orange and green). ns: non-significant; *: p-value < 0.05; **: p-value <0.01; ***: p-value <0.001. RESULTS 59 Regarding cell size, in both regions of WT and WT-comp embryos, the trends were similar over time, as evidenced by halved volume (Fig. 22A-B) and reduced height (Fig. 22C-D) at subsequent stages. However, we observed how the cells from animal and vegetal poles differ more in their size in the case of the WT-comp embryos. Notably, we observed differences in cell volumes at 256-cell stage (Fig. 22B) and in cell height at each stage analysed (Fig. 22D). In contrast, in the WT embryos, we did not observe any differences in cell volumes (Fig. 22A) and only at 512-cell stage in cell heights (Fig. 22C) between animal and vegetal cells. Figure 22. Cell size comparison between animal and vegetal poles Quantifications of average cell volume (A-B) and average cell height (C-D) at 128-, 256- and 512-cell stages in WT and WT-comp embryos. WT animal: n= 75 time points, 3 embryos, 3 experiments. WT vegetal: n= 75 time points, 3 embryos, 2 experiments. WT-comp animal: n= 3, 3 embryos, 2 experiments. WT-comp vegetal: n= 75, 3 embryos, 3 experiments. RESULTS 66 Discussion DISCUSSION 68 DISCUSSION 69 Elucidating the mechanisms involved in 3D packing is crucial to understand how tissues and organs form during animal embryogenesis (Gómez-Gálvez, et al., 2021a; Lemke & Nelson, 2021). It has been previously proposed that AB-T1s are induced by i) curvature anisotropy (Gómez-Gálvez et al., 2018; Mughal et al., 2018) ii) steep curvature gradients that can lead to cell tilting (Lou et al., 2023; Rupprecht et al., 2017), and iii) cell migration and proliferation (Gómez-Gálvez et al., 2018). From a biophysics point of view, the formation of scutoids is clearly related to the surface tension parameters of the cells (Gómez-Gálvez et al., 2018; Mughal et al., 2018). In addition, Lou and colleagues propose that, in cylinder and ellipsoidal geometries, the interplay between mechanics (e.g. pressure, cell density and lateral tension) and cellular tilt is responsible for the appearance of neighbour rearrangements along the apico-basal direction (Gómez- Gálvez, et al., 2021b; Sánchez-Gutiérrez et al., 2016). However, distinguishing the specific involvement of each factor affecting 3D packing remains elusive. One possible reason is that previous studies analysing the factors contributing to the formation of scutoids applied theoretical models or examined tissues in steady-state conditions. Therefore, these previous approaches often overlooked the dynamics inherent in biological systems that can change 3D packing over time (Gómez-Gálvez et al., 2022, 2018; Lou et al., 2023; Rupprecht et al., 2017). Here, we take advantage of the characteristic development of the sea star embryo, coupled with its suitability for live imaging and mechanical manipulations to analyse tissue packing in 4D (3D + time). In addition, we use deep learning-based segmentation to dissect the contributions of tissue compaction, cell density, and cell proliferation to the formation of AB-T1s. In the sea star embryo, cells organise in a monolayered epithelium that is initially leaky. It is only with subsequent rounds of oriented cell divisions that the epithelium becomes progressively more compact. The term “compaction” is used to indicate how tightly cells are packed together. It is perfectly possible for a tissue to become more compact without the cells within it losing convexity. During the specific stages of sea star development that we analyse, the increase of compaction is a reflect of different events of the sea star embryo morphogenesis (Fig. 27A, B): 1) The intracellular space is progressively reduced until all cells are in direct contact with each other along the whole lateral surface. The less intracellular space available the more tightly packed cells are (“sealing” in Fig. 27B). We measure this aspect of compaction as the reduction of “opening areas” in Fig. 19A, Fig. 21A. DISCUSSION 70 2) Cell density increases, as measured in Fig. 12D, Fig. 14D, Fig. 19B and Fig. 21B. An increase in cell density when cells maintain the same volume is a good measure of compaction. This is the case, for instance, in the comparison between WT-comp and WT embryos that are at the same developmental stage (Fig. 14D). However, when comparing cells between stages, cell density is not necessarily a good measurement of changes in how tightly packed the cells are, because cell volumes are halved after each round of cell division (Fig. 12E, Fig. 14E). Figure 27. Relationship between compaction, cell density, convexity and scutoids Schematic representations of the process of compaction in sea star embryos. A) Compaction of an idealized tissue caused by external forces i.e., lateral compression. The illustration shows how increased compaction may result in several phenomena, including reduction of intercellular space, decreased convexity and increased cell density. B) Compaction driven by lateral expansion of the epithelium via oriented cell divisions. Initially, cell divisions cause the reduction of interstitial space, until the epithelium is sealed. In this phase, when cells have space to expand laterally, cell divisions do not drive the formation of AB-T1s. Once the epithelium is sealed, and lateral expansion is no longer possible, oriented cell divisions create lateral compression forces that result in lowered convexity and in cells adopting the “scutoidal” shape. The blue background represents the internal medium, which is shared with the external medium, when the tissue is not sealed. DISCUSSION 71 3) The segmentation of the individual cells of the embryonic epithelium has shown that they bulge into each other resulting in a loss of convexity (Fig. 27, Fig. S2). The extent of “lateral bulging” of cells into each other will be more pronounced the more tightly packed cells are (Fig. S2). Using these different measures, we aimed at describing and account for the different ways compaction occurs in this dynamic tissue (Fig. 27). Cell proliferation occurs initially in a loosely packed epithelium and, only at later stages, in the context of a tightly packed tissue (Fig. 27B). This phenomenon allows us to determine the relative contributions of proliferation and compaction to scutoid formation. In addition, the timing and extent of compaction is determined, at least in part, by how much space is available to the embryonic cells while the tissue expands laterally due to cell divisions. Therefore, we can perform two types of experiments: i) alter tissue compaction by reducing such space, i.e., by mechanically compressing the embryo; and ii) use computational methods to model 3D packing driven only by geometrical cues (minimizing the influence of proliferation). Taking advantage of this model system and experimental approaches, we unravel the mechanisms underlying changes in cell packing at three different scales: whole-embryo (Fig. 12-16), embryonic domains (Fig. 18-21) and local proliferation (Fig. 23-25). On a global scale, the analysis of the time-lapses shows that a significant amount of scutoids is formed only when the tissue is sufficiently compacted (starting at the 256-cell stage in WT embryos). Then, the number of cells with the “scutoidal” shape increase gradually in the subsequent developmental stages (Fig. 12B). This fact suggests that the increase in cell density per se is not sufficient to drive AB-T1s. Our results indicate that it is the coupling of the increase in cell density and the space constraint of a compact epithelium what induces scutoid formation (Fig. 12B, D-E). This hypothesis is supported by experimentally reducing the available space by compressing the embryo (Fig. 13). In this case, sealing happens at the 128-cell stage, and so does the formation of scutoids (Fig. 14B). While the development of P. miniata and this synchronicity of cell divisions is well known, to the best of our knowledge, the data presented here is the first that allows a careful quantification of the timing of cell divisions (Fig. 17A-B, Fig. S3) because one source of variation in cell density can then be in how synchronous each round of cell division is in different embryos. We do not find that the embryos in which scutoids form earlier, or in which the epithelium closes earlier, differ significantly in the synchronicity of their mitotic waves compared to the other DISCUSSION 72 embryos. The slightly different embryo behaviours may be attributed to natural variation in the epithelialization process. On a second level, to address whether intrinsic factors may underlie scutoid formation, we took advantage of a peculiarity of the sea star embryo that we discovered during our studies: compaction is heterogenous. Specifically, epithelial closure is delayed in the animal pole with respect to the vegetal pole, resulting in lower compaction, even though cell density is similar (Fig. 19A-B). The equally delayed appearance of scutoids in the animal pole strongly suggests that tissue closure is necessary for the reorganisation of 3D epithelial packing (Fig. 19A-B, D). This conclusion is supported by our experimental setup: compression of the embryos synchronises closure, equalises compaction and results in synchronous induction of scutoids, which is anticipated to the 128-cell stage in both animal and vegetal poles of compressed embryos (Fig. 21A, D). In addition, our results show that when both poles are sealed, and cells are equally compacted, higher cell density in the vegetal pole at the 512-cell stage results in higher proportion of scutoids (Fig. 21B, D). Therefore, this experiment shows that, once the epithelium is closed, cell density and compaction can affect 3D packing independently. In this context, we propose that tissue compaction, measured here as reduced cell convexity (Fig. 12G, 14E, 18C, 20C), is a proxy for the pressure acting between cells within a tightly packed proliferating epithelium. Taken together, our results suggest that the pressure exerted over the tissue (imposed either by the fertilisation envelope in WT embryos or by the embedding gel in compressed embryos) is driving the changes in 3D tissue organisation. Once the epithelium is closed, this pressure is combined with the pressure generated by an increase in cell density, resulting in higher compaction and inducing more AB-T1s. Still, we find that the cells in the WT vegetal pole of the embryo are more compact and form more scutoids than cells in the animal pole at the 512-cell stage (Fig. 19A, C- D). Given that the external spatial constraints imposed by the fertilisation envelope are the same, these differences in compaction could be due to differences in the material properties of the embryonic domains. Additionally, the higher regional differences in cell size found in WT-comp embryos (Fig. 22) can be attributed to these biophysical properties. Recent studies have shown that in vivo tissues undergo spatiotemporal transitions between fluid and solid states (Fig. 5) which present different properties, i.e. stiffness, cell motion and propensity of cellular rearrangements (T1 transitions) (Kuriyama et al., 2014; Mongera et al., 2018; Shellard & Mayor, 2023). Moreover, local fluidization or stiffening can change the balance of pressures acting on neighbouring DISCUSSION 73 tissues and ultimately drive morphogenesis (Barriga et al., 2018; Petridou et al., 2019). In contrast, cellular connectivity is another critical factor in determining tissue viscosity and stiffness (Petridou et al., 2021; Petridou & Heisenberg, 2019). The comparison between the poles of the embryo opens the way for an exciting set of experiments to determine whether the observed differences in scutoid formation are due to differences in the material properties of the animal and vegetal pole, so future studies will be needed to establish the relationship between AB-T1s and the physical properties of cells and tissues. Probing tissue stiffness via atomic force microscopy (AFM) is challenging, because the envelope will be situated between the AFM bead and the cell/tissue, thereby complicating measurements (Feng & Roos, 2024). Laser ablation may not be effective, because it is better suited to determine whether tissues are under tension or compression more than establishing material properties such as stiffness or viscosity (Sugimura et al., 2016). Another method suited to the task is pipette aspiration assay (Guevorkian & Maître, 2017), as the envelope can be pierced by a relatively small pipette without losing its overall shape and it allows probing cortical tension and viscosity both at the cellular and tissue level (Guevorkian & Maître, 2017). Finally, on a third level, the 3D epithelial packing comparison between real embryos and Voronoi models (Fig. 23) suggests that even accounting for tissue shape, the global tissue compaction and cell density is not sufficient to explain the high incidence of AB- T1s in this developing epithelium. Previous works have shown that in Voronoi tubular models and epithelial tubes, where one of the two main curvatures is null, there is a direct correlation between the ratio between apical and basal areas (surface ratio) and the proportion of scutoids (Gómez- Gálvez et al., 2022, 2018). In these cases, the Voronoi tubular models always showed a higher proportion of scutoids in comparison with the corresponding (same surface ratio) epithelial tubes. This was due to the absence of a resistance in the mathematical model that it is present in the real tissue. In the sea star embryo, that can be considered oblate spheroids (Fig. 16A-C, Table S3), we have found the opposite result. The Voronoi models present a lower proportion of scutoids than the tissue (Fig. 23B-C). In addition, our experiments and simulations discard a relevant contribution of curvature anisotropy in the induction of scutoids (Fig. 14B-C, Fig. 19D-E, Fig. 21D-E, Fig. 23B-C). DISCUSSION 74 Given that Voronoi models assume steady state conditions (Gómez-Gálvez, et al., 2021b; Sánchez-Gutiérrez et al., 2016), while the sea star embryo is a highly dynamic proliferating tissue, we have explored a possible role of local increase of cell density in scutoid formation. Even though the Voronoi models were constructed using the exact geometrical features, number and position of cells of real embryos, (so that each virtual cell had the exact same centroid position as the equivalent cell in a real embryo), these models predict lower incidence of scutoids than observed. We find that most cells acquiring a “scutoidal” shape do so shortly after cell division, both in WT and WT-comp embryos (Fig. 25, Fig. 26A). We have shown that compressing embryos causes scutoids to appear earlier in development and causes more cells to acquire the “scutoidal” shape. Yet, still, scutoids tend to form shortly after cell division in the compressed embryos at the same ratio as that in WT (Fig. 25, Fig. 26A). We think that this finding is in line with recent predictions from (Lou et al., 2023), who propose that cell division is a source of pressure on cells’ lateral membranes that could induce scutoid formation. Tracking individual cells after cell division in our time lapses shows that through the interphase most AB-T1s are resolved (Fig. 26D). In the compressed embryos, additional forces exerted on the embryo exacerbate the phenomenon, causing the “scutoidal” shape to be maintained for longer periods of time (Fig. 26C-D). We conclude that after division, the cells are likely to adopt the scutoid shape and then, the epithelium gradually accommodates the new cells and eventually change again into a frusta-like shape. Altogether, we propose that, in the proliferating sea star embryo we have the combination of two phenomena: i) rounds of cell divisions where there is an increase in cell density that leads to tissue compaction and ii) the sudden local appearance of new cells after cell division, creating the need to cope with new neighbours. Therefore, the induction of AB-T1s in the sea star embryo might be an efficient way to deal with the increasing pressure over the tissue at local and global levels. Interestingly, it has been previously observed that the “scutoidal” shape is better than prisms at withstanding compression forces in an architectural context (Dhari & Patel, 2022). It is tempting to speculate that a side effect of scutoid formation in tissues undergoing morphogenesis is that they can better withstand compression. DISCUSSION 75 Regarding future directions, our results add a new perspective on the regulation of 3D epithelial organisation and lay the foundation to identify the molecular mechanisms allowing cells to form and resolve AB-T1s in response to developmental changes. REFERENCES 82 1. Alberts, B., Bray, D., Hopkin, K., Johnson, A. D., Lewis, J., Raff, M., Roberts, K., & Walter, P. (2013). Essential Cell Biology. (4th ed.). Garland Publishing. 2. Andrés-San Román, J. A., Gordillo-Vázquez, C., Franco-Barranco, D., Morato, L., Fernández-Espartero, C. H., Baonza, G., Tagua, A., Vicente-Munuera, P., Palacios, A. M., Gavilán, M. P., Martín-Belmonte, F., Annese, V., Gómez-Gálvez, P., Arganda-Carreras, I., & Escudero, L. M. (2023). CartoCell, a high-content pipeline for 3D image analysis, unveils cell morphology patterns in epithelia. Cell Reports Methods, 3(10), 100597. https://doi.org/10.1016/j.crmeth.2023.100597 3. Arganda-Carreras, I., Kaynig, V., Rueden, C., Eliceiri, K. W., Schindelin, J., Cardona, A., & Sebastian Seung, H. (2017). Trainable Weka Segmentation: a machine learning tool for microscopy pixel classification. Bioinformatics, 33(15), 2424–2426. https://doi.org/10.1093/bioinformatics/btx180 4. Arnone, M. I., Byrne, M., & Martinez, P. (2015). Echinodermata. In A. Wanninger (Ed.), Evolutionary Developmental Biology of Invertebrates 6: Deuterostomia (pp. 1–58). Springer Vienna. https://doi.org/10.1007/978-3-7091-1856-6_1 5. Barone, V., Byrne, M., & Lyons, D. C. (2022). Lineage tracing shows that cell size asymmetries predict the dorsoventral axis in the sea star embryo. BMC Biology, 20(1), 179. https://doi.org/10.1186/s12915-022-01359-3 6. Barone, V., & Lyons, D. C. (2022). Live imaging of echinoderm embryos to illuminate evodevo. Frontiers in Cell and Developmental Biology, 10, 1007775. https://doi.org/10.3389/fcell.2022.1007775 7. Barriga, E. H., Franze, K., Charras, G., & Mayor, R. (2018). Tissue stiffening coordinates morphogenesis by triggering collective cell migration in vivo. Nature, 554(7693), 523–527. https://doi.org/10.1038/nature25742 8. Bektas, S. (2017). Curvature of the ellipsoid with Cartesian coordinates. Landsc. Archit. Reg. Plan, 2, 61–66. 10. Bertet, C., Sulak, L., & Lecuit, T. (2004). Myosin-dependent junction remodelling controls planar cell intercalation and axis elongation. Nature, 429(6992), 667–671. https://doi.org/10.1038/nature02590 11. Bi, D., Lopez, J. H., Schwarz, J. M., & Manning, M. L. (2014). Energy barriers and cell migration in densely packed tissues. Soft Matter, 10(12), 1885–1890. https://doi.org/10.1039/c3sm52893f 12. Briggs, E., & Wessel, G. M. (2006). In the beginning… Animal fertilization and sea urchin development. Developmental Biology, 300(1), 15–26. https://doi.org/10.1016/j.ydbio.2006.07.014 13. Campàs, O. (2016). A toolbox to explore the mechanics of living embryonic tissues. Seminars in Cell & Developmental Biology, 55, 119–130. https://doi.org/10.1016/j.semcdb.2016.03.011 14. Cary, G. A., & Hinman, V. F. (2017). Echinoderm development and evolution in the postgenomic era. Developmental Biology, 427(2), 203–211. https://doi.org/10.1016/j.ydbio.2017.02.003. REFERENCES 83 15. Dalton, S. (2015). Linking the Cell Cycle to Cell Fate Decisions. Trends in Cell Biology, 25(10), 592–600. https://doi.org/10.1016/j.tcb.2015.07.007. 16. Dan-Sohkawa, M., & Satoh, N. (1978). Studies on dwarf larvae developed from isolated blastomeres of the starfish. Asterina pectinifera. Journal of Embryology and Experimental Morphology, 46, 171–185. https://www.ncbi.nlm.nih.gov/pubmed/702029. 17. Davidson, E. H., & Erwin, D. H. (2006). Gene regulatory networks and the evolution of animal body plans. Science, 311(5762), 796–800. https://doi.org/10.1126/science.1113832. 18. Davidson, L. A. (2012). Epithelial machines that shape the embryo. Trends in Cell Biology, 22(2), 82–87. https://doi.org/10.1016/j.tcb.2011.10.005. 19. Dhari, R. S., & Patel, N. P. (2022). On the crushing behaviour of scutoid-based bioinspired cellular structures. International Journal of Crashworthiness, 27(4), 945–954. https://doi.org/10.1080/13588265.2020.1866859. 20. Dorée, M., & Hunt, T. (2002). From Cdc2 to Cdk1: when did the cell cycle kinase join its cyclin partner? Journal of Cell Science, 115(Pt 12), 2461–2464. https://doi.org/10.1242/jcs.115.12.2461. 21. Escudero, L. M., Costa, L. da F., Kicheva, A., Briscoe, J., Freeman, M., & Babu, M. M. (2011). Epithelial organisation revealed by a network of cellular contacts. Nature Communications, 2(1), 526. https://doi.org/10.1038/ncomms1536 22. Farhadifar, R., Röper, J.-C., Aigouy, B., Eaton, S., & Jülicher, F. (2007). The influence of cell mechanics, cell-cell interactions, and proliferation on epithelial packing. Current Biology: CB, 17(24), 2095–2104. https://doi.org/10.1016/j.cub.2007.11.049. 23. Feng, Y., & Roos, W. H. (2024). Atomic force microscopy: An introduction. Methods in Molecular Biology (Clifton, N.J.), 2694, 295–316. https://doi.org/10.1007/978-1-0716-3377- 9_14 24. Firmino, J., Rocancourt, D., Saadaoui, M., Moreau, C., & Gros, J. (2016). Cell Division Drives Epithelial Cell Rearrangements during Gastrulation in Chick. Developmental Cell, 36(3), 249–261. https://doi.org/10.1016/j.devcel.2016.01.007. 25. Franco-Barranco, D., Pastor-Tronch, J., González-Marfil, A., Muñoz-Barrutia, A., & Arganda-Carreras, I. (2022). Deep learning based domain adaptation for mitochondria segmentation on EM volumes. Computer Methods and Programs in Biomedicine, 222, 106949. https://doi.org/10.1016/j.cmpb.2022.106949. 26. Gibson, M. C., Patel, A. B., Nagpal, R., & Perrimon, N. (2006). The emergence of geometric order in proliferating metazoan epithelia. Nature, 442(7106), 1038–1041. https://doi.org/10.1038/nature05014. 27. Gibson, W. T., & Gibson, M. C. (2009). Cell topology, geometry, and morphogenesis in proliferating epithelia. Current Topics in Developmental Biology, 89, 87–114. https://doi.org/10.1016/S0070-2153(09)89004-2 28. Gibson, W. T., Veldhuis, J. H., Rubinstein, B., Cartwright, H. N., Perrimon, N., Brodland, G. W., Nagpal, R., & Gibson, M. C. (2011). Control of the mitotic cleavage plane by local epithelial topology. Cell, 144(3), 427–438. https://doi.org/10.1016/j.cell.2010.12.035. REFERENCES 84 29. Gilbert, S. F. (2000). Developmental biology. (6th ed.). Sinauer Associates, Sunderlan. 30. Gilbert, S. F., & Raunio, A. M. (1997). Embryology: Constructing the Organism. Sinauer Associates, Sunderlan. 31. Gilbert, S. F., & Rice, R. (2016). Principles of differentiation and morphogenesis. In Robert P. Erickson, and Anthony J. Wynshaw-Boris (Eds.), Epstein’s Inborn Errors of Development (pp. 9–21). Oxford University Press. https://doi.org/10.1093/med/9780199934522.003.0002. 32. Godard, B. G., Dumollard, R., Munro, E., Chenevert, J., Hebras, C., McDougall, A., & Heisenberg, C.-P. (2020). Apical relaxation during mitotic rounding promotes tension-oriented cell division. Developmental Cell, 55(6), 695-706.e4. https://doi.org/10.1016/j.devcel.2020.10.016 33. Godard, B. G., & Heisenberg, C.-P. (2019). Cell division and tissue mechanics. Current Opinion in Cell Biology, 60, 114–120. https://doi.org/10.1016/j.ceb.2019.05.007. 34. Gómez, H. F., Dumond, M. S., Hodel, L., Vetter, R., & Iber, D. (2021). 3D cell neighbour dynamics in growing pseudostratified epithelia. ELife, 10. https://doi.org/10.7554/eLife.68135. 35. Gómez-Gálvez, P., Anbari, S., Escudero, L. M., & Buceta, J. (2021b). Mechanics and selforganization in tissue development. Seminars in Cell & Developmental Biology, 120, 147–159. https://doi.org/10.1016/j.semcdb.2021.07.003. 36. Gómez-Gálvez, P., Vicente-Munuera, P., Anbari, S., Buceta, J., & Escudero, L. M. (2021a). The complex three-dimensional organization of epithelial tissues. Development, 148(1). https://doi.org/10.1242/dev.195669. 37. Gómez-Gálvez, P., Vicente-Munuera, P., Anbari, S., Tagua, A., Gordillo-Vázquez, C., Andrés-San Román, J. A., Franco-Barranco, D., Palacios, A. M., Velasco, A., Capitán- Agudo, C., Grima, C., Annese, V., Arganda-Carreras, I., Robles, R., Márquez, A., Buceta, J., & Escudero, L. M. (2022). A quantitative biophysical principle to explain the 3D cellular connectivity in curved epithelia. Cell Systems, 13(8), 631-643.e8. https://doi.org/10.1016/j.cels.2022.06.003. 38. Gómez-Gálvez, P., Vicente-Munuera, P., Tagua, A., Forja, C., Castro, A. M., Letrán, M., Valencia-Expósito, A., Grima, C., Bermúdez-Gallardo, M., Serrano-Pérez-Higueras, Ó., Cavodeassi, F., Sotillos, S., Martín-Bermudo, M. D., Márquez, A., Buceta, J., & Escudero, L. M. (2018). Scutoids are a geometrical solution to three-dimensional packing of epithelia. Nature Communications, 9(1), 2960. https://doi.org/10.1038/s41467-018-05376-1. 39. Gómez-González, M., Latorre, E., Arroyo, M., & Trepat, X. (2020). Measuring mechanical stress in living tissues. Nature Reviews. Physics, 2(6), 300–317. https://doi.org/10.1038/s42254-020-0184-6 40. Guevorkian, K., & Maître, J.-L. (2017). Micropipette aspiration: A unique tool for exploring cell and tissue mechanics in vivo. Methods in Cell Biology, 139, 187–201. https://doi.org/10.1016/bs.mcb.2016.11.012 41. Guillot, C., & Lecuit, T. (2013). Mechanics of epithelial tissue homeostasis and morphogenesis. Science, 340(6137), 1185–1189. https://doi.org/10.1126/science.1235249. REFERENCES 85 42. Guimarães, C. F., Gasperini, L., Marques, A. P., & Reis, R. L. (2020). The stiffness of living tissues and its implications for tissue engineering. Nature Reviews. Materials, 5(5), 351–370. https://doi.org/10.1038/s41578-019-0169-1 43. Hardwick, L. J. A., Azzarelli, R., & Philpott, A. (2018). Cell cycle-dependent phosphorylation and regulation of cellular differentiation. Biochemical Society Transactions, 46(5), 1083–1091. https://doi.org/10.1042/ BST20180276 44. Hodin, J., Heyland, A., Mercier, A., Pernet, B., Cohen, D. L., Hamel, J.-F., Allen, J. D., McAlister, J. S., Byrne, M., Cisternas, P., & George, S. B. (2019). Culturing echinoderm larvae through metamorphosis. Methods in Cell Biology, 150, 125–169. https://doi.org/10.1016/bs.mcb.2018.11.004. 45. Honda, H. (1978). Description of cellular patterns by Dirichlet domains: the two-dimensional case. Journal of Theoretical Biology, 72(3), 523–543. https://doi.org/10.1016/0022- 5193(78)90315-6 46. Honda, H., Nagai, T., & Tanemura, M. (2008). Two different mechanisms of planar cell intercalation leading to tissue elongation. Developmental Dynamics: An Official Publication of the American Association of Anatomists, 237(7), 1826–1836. https://doi.org/10.1002/dvdy.21609. 47. Hörstadius, S. (1973). Experimental Embryology of Echinoderms. Clarendon Press. 48. Ichbiah, S., Delbary, F., McDougall, A., Dumollard, R., & Turlier, H. (2023). Embryo mechanics cartography: inference of 3D force atlases from fluorescence microscopy. Nature Methods. https://doi.org/10.1038/s41592-023-02084-7 49. Kominami, T. (1983). Establishment of embryonic axes in larvae of the starfish, Asterina pectinifera. Journal of Embryology and Experimental Morphology, 75, 87–100. https://www.ncbi.nlm.nih.gov/pubmed/6886618 50. Kuriyama, S., Theveneau, E., Benedetto, A., Parsons, M., Tanaka, M., Charras, G., Kabla, A., & Mayor, R. (2014). In vivo collective cell migration requires an LPAR2-dependent increase in tissue fluidity. The Journal of Cell Biology, 206(1), 113–127. https://doi.org/10.1083/jcb.201402093. 51. Lecuit, T., & Le Goff, L. (2007). Orchestrating size and shape during morphogenesis. Nature, 450(7167), 189–192. https://doi.org/10.1038/nature06304. 52. Lecuit, T., & Lenne, P.-F. (2007). Cell surface mechanics and the control of cell shape, tissue patterns and morphogenesis. Nature Reviews. Molecular Cell Biology, 8(8), 633–644. https://doi.org/10.1038/nrm2222. 53. Lemke, S. B., & Nelson, C. M. (2021). Dynamic changes in epithelial cell packing during tissue morphogenesis. Current Biology: CB, 31(18), R1098–R1110. https://doi.org/10.1016/j.cub.2021.07.078. 54. Lewis, F. T. (1928). The correlation between cell division and the shapes and sizes of prismatic cells in the epidermis of cucumis. The Anatomical Record, 38(3), 341–376. https://doi.org/10.1002/ar.1090380305. REFERENCES 86 55. Lin, C.-Y., & Su, Y.-H. (2016). Genome editing in sea urchin embryos by using a CRISPR/Cas9 system. Developmental Biology, 409(2), 420–428. https://doi.org/10.1016/j.ydbio.2015.11.018. 56. Lou, Y., Rupprecht, J.-F., Theis, S., Hiraiwa, T., & Saunders, T. E. (2023). Curvature- Induced Cell Rearrangements in Biological Tissues. Physical Review Letters, 130(10), 108401. https://doi.org/10.1103/PhysRevLett.130.108401. 57. Lyons, D. C., Perry, K. J., Lesoway, M. P., & Henry, J. Q. (2012). Cleavage pattern and fate map of the mesentoblast, 4d, in the gastropod Crepidula: a hallmark of spiralian development. EvoDevo, 3(1), 21. https://doi.org/10.1186/2041-9139-3-21. 58. Machado, S., Mercier, V., & Chiaruttini, N. (2019). LimeSeg: a coarse-grained lipid membrane simulation for 3D image segmentation. BMC Bioinformatics, 20(1), 2. https://doi.org/10.1186/s12859-018-2471-0. 59. Mao, Y., Tournier, A. L., Hoppe, A., Kester, L., Thompson, B. J., & Tapon, N. (2013). Differential proliferation rates generate patterns of mechanical tension that orient tissue growth. The EMBO Journal, 32(21), 2790–2803. https://doi.org/10.1038/emboj.2013.197 60. McCauley, B. S., Wright, E. P., Exner, C., Kitazawa, C., & Hinman, V. F. (2012). Development of an embryonic skeletogenic mesenchyme lineage in a sea cucumber reveals the trajectory of change for the evolution of novel structures in echinoderms. EvoDevo, 3(1), 17. https://doi.org/10.1186/2041-9139-3-17. 61. Mongera, A., Rowghanian, P., Gustafson, H. J., Shelton, E., Kealhofer, D. A., Carn, E. K., Serwane, F., Lucio, A. A., Giammona, J., & Campàs, O. (2018). A fluid-to-solid jamming transition underlies vertebrate body axis elongation. Nature, 561(7723), 401–405. https://doi.org/10.1038/s41586-018-0479-2. 62. Mueller, W. A., Hassel, M., & Grealy, M. (2015). Stages and Principles of Animal Development: Terms of Developmental Biology. In: Development and Reproduction in Humans and Animal Model Species. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978- 3-662-43784-1_2. 63. Mughal, A., Cox, S. J., Weaire, D., Burke, S. R., & Hutzler, S. (2018). Demonstration and interpretation of ‘scutoid’ cells formed in a quasi-2D soap froth. Philosophical Magazine Letters, 98(8), 358–364. https://doi.org/10.1080/09500839.2018.1552806. 64. Munjal, A., & Lecuit, T. (2014). Actomyosin networks and tissue morphogenesis. Development, 141(9), 1789–1793. https://doi.org/10.1242/dev.091645. 65. Nelson, C. M. (2016). On Buckling Morphogenesis. Journal of Biomechanical Engineering, 138(2), 021005. https://doi.org/10.1115/1.4032128. 66. Okuda, S., Kuranaga, E., & Sato, K. (2019). Apical Junctional Fluctuations Lead to Cell Flow while Maintaining Epithelial Integrity. Biophysical Journal, 116(6), 1159–1170. https://doi.org/10.1016/j.bpj.2019.01.039. 67. Park, J.-A., Kim, J. H., Bi, D., Mitchel, J. A., Qazvini, N. T., Tantisira, K., Park, C. Y., McGill, M., Kim, S.-H., Gweon, B., Notbohm, J., Steward, R., Jr, Burger, S., Randell, S. H., Kho, A. T., Tambe, D. T., Hardin, C., Shore, S. A., Israel, E., … Fredberg, J. J. (2015). Unjamming and cell shape in the asthmatic airway epithelium. Nature Materials, 14(10), 1040–1048. https://doi.org/10.1038/nmat4357. REFERENCES 87 68. Pearl, E. J., Li, J., & Green, J. B. A. (2017). Cellular systems for epithelial invagination. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 372(1720). https://doi.org/10.1098/rstb.2015.0526. 69. Petridou, N. I., Corominas-Murtra, B., Heisenberg, C.-P., & Hannezo, E. (2021). Rigidity percolation uncovers a structural basis for embryonic tissue phase transitions. Cell, 184(7), 1914-1928.e19. https://doi.org/10.1016/j.cell.2021.02.017. 70. Petridou, N. I., Grigolon, S., Salbreux, G., Hannezo, E., & Heisenberg, C.-P. (2019). Fluidization-mediated tissue spreading by mitotic cell rounding and non-canonical Wnt signalling. Nature Cell Biology, 21(2), 169–178. https://doi.org/10.1038/s41556-018-0247-4. 71. Petridou, N. I., & Heisenberg, C.-P. (2019). Tissue rheology in embryonic organization. The EMBO Journal, 38(20), e102497. https://doi.org/10.15252/embj.2019102497 72. Pollard, T. D., Earnshaw, W. C., Lippincott-Schwartz, J., & Johnson, G. T. (2017). Cell Biology. (3rd ed). Elsevier. 73. Poon, J., Fries, A., Wessel, G. M., & Yajima, M. (2019). Evolutionary modification of AGS protein contributes to formation of micromeres in sea urchins. Nature Communications, 10(1), 3779. https://doi.org/10.1038/s41467-019-11560-8 74. Ragkousi, K., & Gibson, M. C. (2014). Cell division and the maintenance of epithelial order. The Journal of Cell Biology, 207(2), 181–188. https://doi.org/10.1083/jcb.201408044. 75. Ragkousi, K., Marr, K., McKinney, S., Ellington, L., & Gibson, M. C. (2017). Cell-Cycle- Coupled Oscillations in Apical Polarity and Intercellular Contact Maintain Order in Embryonic Epithelia. Current Biology: CB, 27(9), 1381–1386. https://doi.org/10.1016/j.cub.2017.03.064. 76. Roca-Cusachs, P., Conte, V., & Trepat, X. (2017). Quantifying forces in cell biology. Nature Cell Biology, 19(7), 742–751. https://doi.org/10.1038/ncb356 77. Roffay, C., Chan, C. J., Guirao, B., Hiiragi, T., & Graner, F. (2021). Inferring cell junction tension and pressure from cell geometry. Development (Cambridge, England), 148(18). https://doi.org/10.1242/dev.192773 78. Ross, M. H., & Pawlina, W. (2006). Histology: a text and atlas: with correlated cell and molecular biology. (5th ed). Lippincott Williams & Wilkins. 79. Rupprecht, J.-F., Ong, K. H., Yin, J., Huang, A., Dinh, H.-H.-Q., Singh, A. P., Zhang, S., Yu, W., & Saunders, T. E. (2017). Geometric constraints alter cell arrangements within curved epithelial tissues. Molecular Biology of the Cell, 28(25), 3582–3594. https://doi.org/10.1091/mbc.E17-01-0060. 80. Sánchez-Corrales, Y. E., Blanchard, G. B., & Röper, K. (2018). Radially patterned cell behaviours during tube budding from an epithelium. ELife, 7. https://doi.org/10.7554/eLife.35717. 81. Sánchez-Gutiérrez, D., Tozluoglu, M., Barry, J. D., Pascual, A., Mao, Y., & Escudero, L. M. (2016). Fundamental physical cellular constraints drive self-organization of tissues. The EMBO Journal, 35(1), 77–88. https://doi.org/10.15252/embj.201592374. REFERENCES 88 82. Santella, L., Vasilev, F., & Chun, J. T. (2012). Fertilization in echinoderms. Biochemical and Biophysical Research Communications, 425(3), 588–594. https://doi.org/10.1016/j.bbrc.2012.07.159. 83. Schamberger B., Ziege R., Anselme K., Ben Amar M., Bykowski M., Castro A.P.G., Cipitria A., Coles R.A., Dimova R., Eder M., Ehrig S., Escudero L.M., Evans M.E., Fernandes P.R., Fratzl P., Geris L., Gierlinger N., Hannezo E., Iglič A., Kirkensgaard J.J.K., Kollmannsberger P., Kowalewska Ł., Kurniawan N.A., Papantoniou I., Pieuchot L., Pires T.H.V., Renner L.D., Sageman-Furnas A.O., Schröder-Turk G.E., Sengupta A., Sharma V.R., Tagua A., Tomba C., Trepat X., Waters S.L., Yeo E.F., Roschger A., Bidan C.M., Dunlop J.W.C. (2023). Curvature in Biological Systems: Its Quantification, Emergence, and Implications across the Scales. Advanced Materials, 35(13), e2206110. https://doi.org/10.1002/adma.20220611 84. Schindelin, J., Arganda-Carreras, I., Frise, E., Kaynig, V., Longair, M., Pietzsch, T., Preibisch, S., Rueden, C., Saalfeld, S., Schmid, B., Tinevez, J.-Y., White, D. J., Hartenstein, V., Eliceiri, K., Tomancak, P., & Cardona, A. (2012). Fiji: an open-source platform for biological-image analysis. Nature Methods, 9(7), 676–682. https://doi.org/10.1038/nmeth.2019. 85. Shellard, A., & Mayor, R. (2023). Sculpting with stiffness: rigidity as a regulator of morphogenesis. Biochemical Society Transactions. https://doi.org/10.1042/BST20220826. 86. Sorce, B., Escobedo, C., Toyoda, Y., Stewart, M. P., Cattin, C. J., Newton, R., Banerjee, I., Stettler, A., Roska, B., Eaton, S., Hyman, A. A., Hierlemann, A., & Müller, D. J. (2015). Mitotic cells contract actomyosin cortex and generate pressure to round against or escape epithelial confinement. Nature Communications, 6(1), 8872. https://doi.org/10.1038/ncomms9872 87. Stooke-Vaughan, G. A., & Campàs, O. (2018). Physical control of tissue morphogenesis across scales. Current Opinion in Genetics & Development, 51, 111–119. https://doi.org/10.1016/j.gde.2018.09.002 88. Sugimura, K., Lenne, P.-F., & Graner, F. (2016). Measuring forces and stresses in situ in living tissues. Development (Cambridge, England), 143(2), 186–196. https://doi.org/10.1242/dev.119776 89. Sun, Z., Amourda, C., Shagirov, M., Hara, Y., Saunders, T. E., & Toyama, Y. (2017). Basolateral protrusion and apical contraction cooperatively drive Drosophila germ-band extension. Nature Cell Biology, 19(4), 375–383. https://doi.org/10.1038/ncb3497. 90. Sutherland, A., Keller, R., & Lesko, A. (2020). Convergent extension in mammalian morphogenesis. Seminars in Cell & Developmental Biology, 100, 199–211. https://doi.org/10.1016/j.semcdb.2019.11.002. 91. Tada, M., & Heisenberg, C.-P. (2012). Convergent extension: using collective cell migration and cell intercalation to shape embryos. Development, 139(21), 3897–3904. https://doi.org/10.1242/dev.073007. 92. Thompson, D. W. D. (1942). On Growth and form. (2nd edition). Cambridge University Press. 93. Vicente-Munuera, P., Gómez-Gálvez, P., Tetley, R. J., Forja, C., Tagua, A., Letrán, M., Tozluoglu, M., Mao, Y., & Escudero, L. M. (2020). EpiGraph: an open-source platform to REFERENCES 89 quantify epithelial organization. Bioinformatics, 36(4), 1314–1316. https://doi.org/10.1093/bioinformatics/btz683. 94. Voronoi, G. (1908). Nouvelles applications des paramètres continus à la théorie des formes quadratiques. Premier mémoire. Sur quelques propriétés des formes quadratiques positives parfaites. 1908(133), 97–102. https://doi.org/10.1515/crll.1908.133.97. 95. Welsch, U., Sobotta, J. and Delles, T. (2014). Histología (3th ed.). Médica Panamericana. 96. Wolny, A., Cerrone, L., Vijayan, A., Tofanelli, R., Barro, A. V., Louveaux, M., Wenzl, C., Strauss, S., Wilson-Sánchez, D., Lymbouridou, R., Steigleder, S. S., Pape, C., Bailoni, A., Duran-Nebreda, S., Bassel, G. W., Lohmann, J. U., Tsiantis, M., Hamprecht, F. A., Schneitz, K., … Kreshuk, A. (2020). Accurate and versatile 3D segmentation of plant tissues at cellular resolution. ELife, 9. https://doi.org/10.7554/eLife.57613. 97. Wolpert, L., Tickle, C., & Arias, A. M. (2019). Principles of Development. (6th ed.). Oxford University Press. 98. Xu, B., Washington, A. M., Domeniconi, R. F., Ferreira Souza, A. C., Lu, X., Sutherland, A., & Hinton, B. T. (2016). Protein tyrosine kinase 7 is essential for tubular morphogenesis of the Wolffian duct. Developmental Biology, 412(2), 219–233. https://doi.org/10.1016/j.ydbio.2016.02.029. 99. Yanagida, M. (2014). The role of model organisms in the history of mitosis research. Cold Spring Harbor Perspectives in Biology, 6(9), a015768. https://doi.org/10.1101/cshperspect.a015768. REFERENCES 90 Annexes ANNEXES 98 Table S2. Statistics of regional comparisons The colours indicate the values related to the orientation of the embryo, with orange representing animal view and green representing vegetal view. Multiple Mann-Whitney tests with Bonferroni multiple comparisons correction were used to compare the datasets from WT and WT-comp in the columns 128- cell stage, 256-cell stage, 512-cell stage. P-value marked (*) are exceptions where two-way ANOVA tests with Bonferroni multiple comparisons were used. Non-parametric Kruskal-Wallis tests with Dunn multiple comparisons correction were used in the columns 128/256, 256/512, 128/512 to compare the datasets among each cellular stage. 128-cell stage 256-cell stage 512-cell stage 128/256 256/512 128/512 condition Mean SD p Mean SD p Mean SD p p p p Scu WT 0.00 0.00 >0,99 0.00 0.00 <0.001 0.07 0.06 <0.001 >0,99 <0.001 <0.001 WT 0.00 0.00 0.06 0.08 0.16 0.07 0.08 <0.001 <0.001 WT-comp 0.07 0.11 >0,99 0.11 0.08 0.44 0.12 0.02 <0.001 0.42 0.59 0.03 WT-comp 0.08 0.13 0.13 0.10 0.27 0.11 0.55 0.001 <0.001 sr-A WT 0.149 0.031 0.008* 0.123 0.048 0.01* 0.107 0.023 0.15* 0.07** 0.22** 0.002** WT 0.106 0.040 0.094 0.043 0.087 0.036 0.59** 0.59** 0.38** WT-comp 0.196 0.070 >0,99 0.171 0.068 0.001 0.175 0.050 0.005 >0,99 >0,99 >0,99 WT-comp 0.194 0.063 0.106 0.044 0.129 0.047 <0.001 0.47 0.02 density (µm-2) WT 6.21E- 04 5.79E- 05 >0,99 1.13E- 03 1.24E- 04 0.92 2.16E- 03 2.35E- 04 0.86 0.003 <0.001 <0.001 WT 6.28E- 04 3.67E- 05 1.19E- 03 1.57E- 04 2.24E- 03 2.07E- 04 0.003 <0.001 <0.001 WT-comp 7.43E- 04 1.29E- 04 0.54 1.43E- 03 2.38E- 04 0.9 2.39E- 03 2.21E- 04 0.03 0.003 <0.001 <0.001 WT-comp 8.13E- 04 1.06E- 04 1.48E- 03 2.05E- 04 2.62E- 03 3.86E- 04 0.003 <0.001 <0.001 vol (µm3) WT 1.57E+ 04 2.76E+ 03 0.16 7.37E+ 03 1.63E+ 03 0.16 3.49E+ 03 6.90E+ 02 0.10 0.003 <0.001 <0.001 WT 1.44E+ 04 1.98E+ 03 6.73E+ 03 1.05E+ 03 3.00E+ 03 4.95E+ 02 0.003 <0.001 <0.001 WT-comp 1.39E+ 04 2.06E+ 03 0.38 6.39E+ 03 1.33E+ 03 0.002 3.38E+ 03 4.61E+ 02 0.32 0.003 <0.001 <0.001 WT-comp 1.68E+ 04 4.47E+ 03 8.10E+ 03 2.11E+ 03 3.87E+ 03 9.10E+ 02 0.004 <0.001 <0.001 convx WT 0.877 0.011 <0.001 0.844 0.009 <0.001 0.811 0.008 <0.001 0.003 <0.001 <0.001 WT 0.849 0.014 0.827 0.012 0.800 0.013 0.002 <0.001 <0.001 WT-comp 0.828 0.017 0.46 0.813 0.013 0.09 0.786 0.010 0.67 0.240 <0.001 <0.001 WT-comp 0.837 0.020 0.806 0.021 0.778 0.024 0.006 0.001 <0.001 sr WT 1.70 0.18 0.31 1.55 0.17 0.53 1.45 0.11 0.22 0.03 0.09 <0.001 WT 1.60 0.08 1.48 0.04 1.38 0.05 <0.001** <0.001** <0.001** WT-comp 1.92 0.38 >0,99 1.75 0.41 0.13 1.52 0.16 0.001 0.84 0.15 0.02 WT-comp 1.97 0.20 1.82 0.17 1.65 0.09 0.18 <0.001 <0.001 h (µm) WT 24.45 2.62 0.14 20.39 2.72 0.99 17.74 2.48 0.04 0.006 0.003 <0.001 WT 22.72 1.53 19.41 1.30 15.84 1.64 <0.001** <0.001** <0.001** WT-comp 25.01 3.64 0.03 22.10 2.20 0.001 19.03 1.14 <0.001 0.08 <0.001 <0.001 WT-comp 27.38 3.17 24.22 2.30 20.61 1.17 0.1 <0.001 <0.001 ANNEXES 99 P-value marked (**) are exceptions where ordinary one-way ANOVA tests with Tukey multiple comparisons correction were used. Bold values indicate p-value < 0.05. Scu: proportion of scutoids; SR-A: surface ratio anisotropy; density: cell density; vol: cell volume; convx: cell convex ratio. SR: surface ratio. h: cell height. ANNEXES 100 Major axis 2nd major axis Minor axis (µm) (µm) (µm) ID cell-stage condition Mean SD Mean SD Mean SD shape 128 209.6 0.4 202.4 0.9 176.1 8.4 oblate Embryo 1 256 WT 210.7 1.1 206.7 2.1 188.8 8.5 oblate 512 209.1 1.2 205.4 1.1 184.0 6.9 oblate 128 212.2 1.7 205.3 0.6 176.4 4.5 oblate Embryo 2 256 WT 210.4 1.5 207.1 2.1 174.4 14.7 oblate 512 209 2 206.8 1.9 176.6 6.4 oblate 128 233 0.7 215.2 1.4 169.0 5.4 oblate Embryo 3 256 WT 234.9 1.2 220.3 0.7 164.8 8.1 oblate 512 234.9 0.5 223.5 1.2 164.5 6.5 oblate 128 210.3 0.7 208 1.1 174.5 2.0 sphere Embryo 4 256 WT 213.1 1.7 209.8 1.5 174.1 5.5 oblate 512 211.7 1.2 210.3 1 177.8 5.9 oblate 128 202.8 1.2 197.2 2.4 181.3 5.8 oblate Embryo 5 256 WT 203.7 1.9 201 2.6 167.4 9.1 oblate 512 206 2.9 202.8 1.1 166.7 3.4 oblate 128 196.5 1.5 181.9 3.8 172.0 2.9 oblate Embryo 6 256 WT 199.4 1.3 196.6 1.9 174.8 4.2 oblate 512 196.6 3.6 193.3 4.3 173.0 11.6 oblate 128 217.9 2 210.7 2.8 127.8 5.4 oblate Embryo 7 256 WT-comp 214.6 4 210.6 5.9 146.0 16.8 oblate 512 213.1 3.4 210.2 3.4 148.1 11.8 oblate 128 215.1 1.6 190.9 1.3 165.9 8.5 ellipsoid Embryo 8 256 WT-comp 219.6 2.1 194.2 1.6 159.2 10.0 ellipsoid 512 219 2.4 195.6 1.8 167.7 6.0 ellipsoid 128 230.8 2.2 199.4 2.9 157.0 16.8 ellipsoid Embryo 9 256 WT-comp 234.9 0.5 223.5 1.2 184.0 6.9 oblate 512 235.1 1 203.3 1.1 151.6 5.7 ellipsoid 128 197.9 1.1 187.4 0.6 154.4 8.0 oblate Embryo 11 256 WT-comp 195.5 2.5 187.1 2 166.7 15.4 oblate 512 197.8 1 191.6 0.9 160.2 12.1 oblate 128 197.6 0.3 166.4 4.1 161.8 2.6 prolate Embryo 12 256 WT-comp 197.6 2.5 175.4 5 163.1 7.1 prolate 512 197.1 1.4 177.6 0.9 157.4 7.2 prolate Table S3. Morphological classification of the embryos Detailed information about the estimated length of the three symmetry axes of the embryos. In the column labelled condition, the colours indicate the orientation of the embryo, with orange representing animal view and green representing vegetal view. ANNEXES 101 The column shape denotes the classification depending on the conformation of the embryos by determining the proportional relationships among the three average length axes. When two axes exhibited a similarity of 92%, they were categorized as having equal sizes. Sphere: All three axes are equal in length; Oblate spheroid: two major axes and one minor axis; Prolate spheroid: one major axis and two minor axes; Ellipsoid: All three axes have different lengths. Feature Comparisons Mean SD p value relative scutoids onset WT 0.12 0.12 0.79 WT-comp 0.12 0.13 relative scutoids duration WT 0.53 0.24 0.01 WT-comp 0.61 0.25 duration from 128-cell stage to the end of 512-cell stage (minutes) WT 199.00 16.72 0.08 WT-comp 230.00 35.33 average scutoids onset across embryos After mitosis 0.63 0.27 0.02 Independent of mitosis 0.37 0.27 Table S4. Statistics related to tracked “scutoidal” cells and duration of the embryo development The colours indicate the values related to the condition, with light blue representing WT and dark blue representing WT-comp. Mann-Whitney tests were used to compare the normalized scutoids onset and duration (see materials and methods). Meanwhile, Student t tests were used to compare the minutes elapsed between the beginning of the 128-cell stage and the end of the 512-cell stage as well as the average scutoids onset between the categories after and independent of mitosis. Bold values indicate pvalue < 0.05. ANNEXES 102 ID_scutoid condition ID_embryo cell-stage onset duration end intP duration (minutes) 1 WT Embryo 1 512 0 0.471 0.471 84 2 WT Embryo 1 512 0 0.5 0.5 90 3 WT Embryo 1 512 0 0.6 0.6 102 4 WT Embryo 1 512 0.059 0.412 0.471 102 5 WT Embryo 1 512 0.059 0.471 0.529 102 6 WT Embryo 1 512 0.067 0.4 0.467 90 7 WT Embryo 1 512 0.067 0.467 0.533 90 8 WT Embryo 1 512 0.067 0.533 0.6 90 9 WT Embryo 1 512 0.067 0.6 0.667 90 10 WT Embryo 1 512 0.071 0.571 0.643 84 11 WT Embryo 1 512 0.077 0.538 0.615 78 12 WT Embryo 1 512 0.077 0.538 0.615 78 13 WT Embryo 1 512 0.077 0.615 0.692 78 14 WT Embryo 1 512 0.083 0.583 0.667 72 15 WT Embryo 1 512 0.125 0.375 0.5 96 16 WT Embryo 1 512 0.167 0.667 0.833 72 17 WT Embryo 2 512 0.167 0.5 0.667 108 18 WT Embryo 2 512 0.286 0.643 0.929 84 19 WT Embryo 2 512 0.286 0.643 0.929 84 20 WT Embryo 2 512 0.308 0.692 1 78 21 WT Embryo 3 512 0 0.444 0.444 108 22 WT Embryo 3 512 0.105 0.421 0.526 114 23 WT Embryo 3 512 0.111 0.444 0.556 108 24 WT Embryo 3 512 0.235 0.471 0.706 102 25 WT Embryo 4 256 0 0.333 0.333 54 26 WT Embryo 4 256 0 0.375 0.375 48 27 WT Embryo 4 256 0 0.375 0.375 48 28 WT Embryo 4 256 0.125 0.375 0.5 48 29 WT Embryo 4 512 0 0.75 0.75 78 30 WT Embryo 4 512 0 0.75 0.75 60 31 WT Embryo 4 512 0 0.75 0.75 72 32 WT Embryo 4 512 0 0.765 0.765 72 33 WT Embryo 4 512 0 0.769 0.769 72 34 WT Embryo 4 512 0 0.9 0.9 84 35 WT Embryo 4 512 0 0.929 0.929 102 36 WT Embryo 4 512 0.063 0.813 0.875 96 37 WT Embryo 4 512 0.067 0.667 0.733 90 38 WT Embryo 4 512 0.071 0.929 1 84 39 WT Embryo 4 512 0.083 0.75 0.833 72 40 WT Embryo 4 512 0.125 0.563 0.688 96 41 WT Embryo 4 512 0.143 0.643 0.786 84 42 WT Embryo 4 512 0.167 0.167 0.333 72 43 WT Embryo 4 512 0.167 0.75 0.917 72 44 WT Embryo 4 512 0.182 0.818 1 66 ANNEXES 103 ID_scutoid condition ID_movie cell-stage onset duration end intP duration (minutes) 45 WT Embryo 4 512 0.182 0.818 1 66 46 WT Embryo 4 512 0.25 0.167 0.417 72 47 WT Embryo 4 512 0.25 0.167 0.417 72 48 WT Embryo 4 512 0.25 0.167 0.417 72 49 WT Embryo 5 256 0 0.9 0.9 60 50 WT Embryo 5 256 0 0.9 0.9 60 51 WT Embryo 5 256 0 0.9 0.9 60 52 WT Embryo 5 256 0 1 1 54 53 WT Embryo 5 256 0.222 0.333 0.556 54 54 WT Embryo 5 256 0.222 0.333 0.556 54 55 WT Embryo 5 512 0 0.158 0.158 84 56 WT Embryo 5 512 0 0.5 0.5 90 57 WT Embryo 5 512 0 0.667 0.667 120 58 WT Embryo 5 512 0 0.714 0.714 78 59 WT Embryo 5 512 0 0.769 0.769 72 60 WT Embryo 5 512 0 0.923 0.923 78 61 WT Embryo 5 512 0 1 1 114 62 WT Embryo 5 512 0.056 0.667 0.722 108 63 WT Embryo 5 512 0.059 0.294 0.353 102 64 WT Embryo 5 512 0.063 0.625 0.688 96 65 WT Embryo 5 512 0.067 0.2 0.267 90 66 WT Embryo 5 512 0.067 0.333 0.4 90 67 WT Embryo 5 512 0.077 0.769 0.846 78 68 WT Embryo 5 512 0.091 0.909 1 66 69 WT Embryo 5 512 0.091 0.909 1 66 70 WT Embryo 5 512 0.105 0.632 0.737 114 71 WT Embryo 5 512 0.143 0.214 0.357 84 72 WT Embryo 5 512 0.231 0.231 0.462 78 73 WT Embryo 5 512 0.231 0.231 0.462 78 74 WT Embryo 5 512 0.25 0.25 0.5 72 75 WT Embryo 5 512 0.25 0.25 0.5 72 76 WT Embryo 5 512 0.25 0.25 0.5 72 77 WT Embryo 6 512 0.048 0.238 0.286 126 78 WT Embryo 6 512 0.056 0.333 0.389 108 79 WT Embryo 6 512 0.067 0.4 0.467 90 80 WT Embryo 6 512 0.067 0.933 1 90 81 WT Embryo 6 512 0.087 0.391 0.478 138 82 WT Embryo 6 512 0.111 0.333 0.444 108 83 WT Embryo 6 512 0.143 0.524 0.667 126 84 WT Embryo 6 512 0.211 0.105 0.316 114 85 WT Embryo 6 512 0.222 0.611 0.833 108 86 WT Embryo 6 512 0.227 0.091 0.318 132 87 WT Embryo 6 512 0.235 0.647 0.882 102 88 WT Embryo 6 512 0.25 0.25 0.5 120 ANNEXES 104 ID_scutoid condition ID_movie cell-stage onset duration end intP duration (minutes) 89 WT Embryo 6 512 0.25 0.25 0.5 96 90 WT Embryo 6 512 0.25 0.25 0.5 120 91 WT Embryo 6 512 0.25 0.45 0.7 120 92 WT Embryo 6 512 0.25 0.563 0.813 120 93 WT Embryo 6 512 0.278 0.5 0.778 108 94 WT Embryo 6 512 0.45 0.35 0.8 120 95 WT Embryo 6 512 0.467 0.133 0.6 90 96 WT Embryo 6 512 0.5 0.438 0.938 96 97 WT Embryo 6 512 0.5 0.438 0.938 96 98 WT-comp Embryo 7 256 0 0.455 0.455 66 99 WT-comp Embryo 7 256 0 0.636 0.636 60 100 WT-comp Embryo 7 256 0 0.636 0.636 60 101 WT-comp Embryo 7 256 0 0.636 0.636 60 102 WT-comp Embryo 7 256 0 0.818 0.818 60 103 WT-comp Embryo 7 256 0 0.909 0.909 66 104 WT-comp Embryo 7 256 0 0.909 0.909 66 105 WT-comp Embryo 7 256 0 1 1 66 106 WT-comp Embryo 7 256 0 1 1 66 107 WT-comp Embryo 7 256 0 1 1 66 108 WT-comp Embryo 7 256 0 1 1 66 109 WT-comp Embryo 7 256 0.071 0.5 0.571 84 110 WT-comp Embryo 7 256 0.071 0.5 0.571 84 111 WT-comp Embryo 7 256 0.083 0.833 0.917 72 112 WT-comp Embryo 7 256 0.091 0.455 0.545 66 113 WT-comp Embryo 7 256 0.091 0.455 0.545 66 114 WT-comp Embryo 7 256 0.091 0.909 1 66 115 WT-comp Embryo 7 256 0.2 0.7 0.9 60 116 WT-comp Embryo 7 256 0.2 0.7 0.9 60 117 WT-comp Embryo 7 512 0 0.75 0.75 72 118 WT-comp Embryo 7 512 0 0.778 0.778 72 119 WT-comp Embryo 7 512 0 0.875 0.875 78 120 WT-comp Embryo 7 512 0 0.923 0.923 96 121 WT-comp Embryo 7 512 0 0.933 0.933 84 122 WT-comp Embryo 7 512 0 1 1 90 123 WT-comp Embryo 7 512 0 1 1 96 124 WT-comp Embryo 7 512 0 1 1 108 125 WT-comp Embryo 7 512 0.056 0.667 0.722 108 126 WT-comp Embryo 7 512 0.063 0.688 0.75 96 127 WT-comp Embryo 7 512 0.118 0.706 0.824 102 128 WT-comp Embryo 7 512 0.125 0.75 0.875 96 129 WT-comp Embryo 7 512 0.133 0.8 0.933 90 130 WT-comp Embryo 7 512 0.188 0.688 0.875 96 131 WT-comp Embryo 7 512 0.267 0.733 1 90 132 WT-comp Embryo 7 512 0.267 0.733 1 90 ANNEXES 105 ID_scutoid condition ID_movie cell-stage onset duration end intP duration (minutes) 133 WT-comp Embryo 8 128 0 0.5 0.5 60 134 WT-comp Embryo 8 128 0 0.8 0.8 60 135 WT-comp Embryo 8 128 0.091 0.455 0.545 66 136 WT-comp Embryo 8 128 0.091 0.727 0.818 66 137 WT-comp Embryo 8 128 0.1 0.5 0.6 60 138 WT-comp Embryo 8 128 0.111 0.889 1 54 139 WT-comp Embryo 8 128 0.111 0.889 1 54 140 WT-comp Embryo 8 128 0.2 0.5 0.7 60 141 WT-comp Embryo 8 256 0 0.692 0.692 60 142 WT-comp Embryo 8 256 0 0.9 0.9 78 143 WT-comp Embryo 8 256 0.083 0.5 0.583 72 144 WT-comp Embryo 8 256 0.083 0.5 0.583 72 145 WT-comp Embryo 8 256 0.167 0.75 0.917 72 146 WT-comp Embryo 8 256 0.182 0.818 1 66 147 WT-comp Embryo 8 256 0.25 0.5 0.75 72 148 WT-comp Embryo 8 256 0.25 0.5 0.75 72 149 WT-comp Embryo 8 512 0 0.346 0.346 78 150 WT-comp Embryo 8 512 0 0.857 0.857 156 151 WT-comp Embryo 8 512 0 0.929 0.929 108 152 WT-comp Embryo 8 512 0 0.947 0.947 108 153 WT-comp Embryo 8 512 0 1 1 114 154 WT-comp Embryo 8 512 0 1 1 126 155 WT-comp Embryo 8 512 0 1 1 78 156 WT-comp Embryo 8 512 0 1 1 84 157 WT-comp Embryo 8 512 0.056 0.722 0.778 108 158 WT-comp Embryo 8 512 0.067 0.8 0.867 90 159 WT-comp Embryo 8 512 0.067 0.867 0.933 90 160 WT-comp Embryo 8 512 0.071 0.929 1 84 161 WT-comp Embryo 8 512 0.077 0.462 0.538 156 162 WT-comp Embryo 8 512 0.087 0.522 0.609 138 163 WT-comp Embryo 8 512 0.188 0.75 0.938 96 164 WT-comp Embryo 8 512 0.188 0.813 1 96 165 WT-comp Embryo 8 512 0.222 0.5 0.722 108 166 WT-comp Embryo 8 512 0.222 0.722 0.944 108 167 WT-comp Embryo 8 512 0.313 0.563 0.875 96 168 WT-comp Embryo 8 512 0.357 0.643 1 84 169 WT-comp Embryo 9 256 0.077 0.923 1 78 170 WT-comp Embryo 9 256 0.111 0.667 0.778 108 171 WT-comp Embryo 9 256 0.125 0.75 0.875 96 172 WT-comp Embryo 9 256 0.133 0.8 0.933 90 173 WT-comp Embryo 9 512 0 0.5 0.5 102 174 WT-comp Embryo 9 512 0 0.632 0.632 90 175 WT-comp Embryo 9 512 0 0.8 0.8 114 176 WT-comp Embryo 9 512 0 0.941 0.941 108 ANNEXES 106 ID_scutoid condition ID_movie cell-stage onset duration end intP duration (minutes) 177 WT-comp Embryo 9 512 0.048 0.571 0.619 126 178 WT-comp Embryo 9 512 0.059 0.529 0.588 102 179 WT-comp Embryo 9 512 0.063 0.688 0.75 96 180 WT-comp Embryo 9 512 0.083 0.5 0.583 144 181 WT-comp Embryo 9 512 0.111 0.889 1 108 182 WT-comp Embryo 9 512 0.143 0.762 0.905 126 183 WT-comp Embryo 9 512 0.2 0.8 1 120 184 WT-comp Embryo 9 512 0.278 0.667 0.944 108 185 WT-comp Embryo 9 512 0.294 0.706 1 102 186 WT-comp Embryo 9 512 0.333 0.667 1 108 187 WT-comp Embryo 9 512 0.429 0.143 0.571 126 188 WT-comp Embryo 10 256 0 0.667 0.667 54 189 WT-comp Embryo 10 256 0 0.667 0.667 54 190 WT-comp Embryo 10 256 0 0.889 0.889 54 191 WT-comp Embryo 10 256 0 0.889 0.889 54 192 WT-comp Embryo 10 256 0 1 1 48 193 WT-comp Embryo 10 256 0 1 1 48 194 WT-comp Embryo 10 256 0.1 0.4 0.5 60 195 WT-comp Embryo 10 256 0.111 0.444 0.556 54 196 WT-comp Embryo 10 256 0.111 0.444 0.556 54 197 WT-comp Embryo 10 256 0.111 0.444 0.556 54 198 WT-comp Embryo 10 256 0.111 0.667 0.778 54 199 WT-comp Embryo 10 256 0.111 0.667 0.778 54 200 WT-comp Embryo 10 512 0 0.769 0.769 78 201 WT-comp Embryo 10 512 0 0.833 0.833 72 202 WT-comp Embryo 10 512 0 0.909 0.909 66 203 WT-comp Embryo 10 512 0 0.909 0.909 66 204 WT-comp Embryo 10 512 0 0.917 0.917 72 205 WT-comp Embryo 10 512 0 0.917 0.917 72 206 WT-comp Embryo 10 512 0 1 1 66 207 WT-comp Embryo 10 512 0 1 1 66 208 WT-comp Embryo 10 512 0.071 0.571 0.643 84 209 WT-comp Embryo 10 512 0.083 0.583 0.667 72 210 WT-comp Embryo 10 512 0.083 0.833 0.917 72 211 WT-comp Embryo 10 512 0.091 0.636 0.727 66 212 WT-comp Embryo 10 512 0.091 0.727 0.818 66 213 WT-comp Embryo 10 512 0.091 0.909 1 66 214 WT-comp Embryo 10 512 0.091 0.909 1 66 215 WT-comp Embryo 10 512 0.091 0.909 1 66 216 WT-comp Embryo 10 512 0.1 0.3 0.4 60 217 WT-comp Embryo 10 512 0.1 0.3 0.4 60 218 WT-comp Embryo 10 512 0.154 0.538 0.692 78 219 WT-comp Embryo 10 512 0.154 0.538 0.692 78 220 WT-comp Embryo 10 512 0.167 0.25 0.417 72 ANNEXES 107 ID_scutoid condition ID_movie cell-stage onset duration end intP duration (minutes) 221 WT-comp Embryo 10 512 0.182 0.273 0.455 66 222 WT-comp Embryo 10 512 0.231 0.308 0.538 78 223 WT-comp Embryo 10 512 0.25 0.333 0.583 72 224 WT-comp Embryo 11 128 0 0.571 0.571 84 225 WT-comp Embryo 11 128 0 0.8 0.8 60 226 WT-comp Embryo 11 128 0.143 0.381 0.524 126 227 WT-comp Embryo 11 128 0.15 0.4 0.55 120 228 WT-comp Embryo 11 256 0.063 0.5 0.563 192 229 WT-comp Embryo 11 256 0.095 0.762 0.857 126 230 WT-comp Embryo 11 256 0.095 0.762 0.857 126 231 WT-comp Embryo 11 256 0.158 0.842 1 114 232 WT-comp Embryo 11 256 0.238 0.619 0.857 126 233 WT-comp Embryo 11 256 0.25 0.5 0.75 72 234 WT-comp Embryo 11 256 0.25 0.5 0.75 72 235 WT-comp Embryo 11 256 0.263 0.684 0.947 114 236 WT-comp Embryo 11 256 0.304 0.565 0.87 138 237 WT-comp Embryo 11 256 0.333 0.5 0.833 72 238 WT-comp Embryo 11 256 0.36 0.52 0.88 150 239 WT-comp Embryo 11 256 0.524 0.286 0.81 126 240 WT-comp Embryo 11 512 0 0.318 0.318 132 241 WT-comp Embryo 11 512 0 0.333 0.333 144 242 WT-comp Embryo 11 512 0 0.414 0.414 174 243 WT-comp Embryo 11 512 0 0.462 0.462 156 244 WT-comp Embryo 11 512 0 0.467 0.467 90 245 WT-comp Embryo 11 512 0 0.533 0.533 90 246 WT-comp Embryo 11 512 0 0.923 0.923 78 247 WT-comp Embryo 11 512 0.03 0.242 0.273 198 248 WT-comp Embryo 11 512 0.032 0.258 0.29 186 249 WT-comp Embryo 11 512 0.042 0.292 0.333 144 250 WT-comp Embryo 11 512 0.048 0.333 0.381 126 251 WT-comp Embryo 11 512 0.05 0.35 0.4 120 252 WT-comp Embryo 11 512 0.053 0.211 0.263 228 253 WT-comp Embryo 11 512 0.055 0.218 0.273 330 254 WT-comp Embryo 11 512 0.063 0.219 0.281 192 255 WT-comp Embryo 11 512 0.08 0.16 0.24 150 256 WT-comp Embryo 11 512 0.12 0.08 0.2 300 257 WT-comp Embryo 11 512 0.143 0.229 0.371 210 258 WT-comp Embryo 11 512 0.167 0.267 0.433 180 259 WT-comp Embryo 11 512 0.167 0.444 0.611 108 260 WT-comp Embryo 11 512 0.176 0.471 0.647 102 261 WT-comp Embryo 11 512 0.233 0.267 0.5 180 262 WT-comp Embryo 11 512 0.255 0.149 0.404 282 263 WT-comp Embryo 11 512 0.267 0.533 0.8 90 264 WT-comp Embryo 11 512 0.333 0.148 0.481 162 ANNEXES 114 8.3. Supplementary movies Movie S1. Patiria miniata WT embryo, vegetal view. Maximum projection of confocal time-lapse video of a WT embryo expressing a membrane marker (mYFP, yellow) and a nuclear marker (n-CFP, cyan) imaged between the 32- and 2000-cell stages. Vegetal view. Scale bars, 50 μm. Frame interval of 6 minutes, 7 fps. Movie S2. Patiria miniata WT-comp embryo, vegetal view. Maximum projection of confocal time-lapse video of a WT-comp embryo expressing a membrane marker (mYFP, yellow) and a nuclear marker (nRFP, cyan) imaged between the 32- and 2000- cell stages. Vegetal view. Scale bars, 50 μm. Frame interval of 6 minutes, 7 fps. Movie S3. Patiria miniata WT embryo, animal view. Maximum projection of confocal time-lapse video of a WT embryo expressing a membrane marker (mYFP, yellow) and a nuclear marker (n-RFP, cyan) imaged between the 32- and 2000-cell stages. Animal view (note the polar bodies). Scale bars, 50 μm. Frame interval of 6 minutes, 7 fps. Movie S4. Patiria miniata WT-comp embryo, animal view. Maximum projection of confocal time-lapse video of a WT-comp embryo expressing a membrane marker (mYFP, yellow) and a nuclear marker (nRFP, cyan) imaged between the 32- and 2000- cell stages. Animal view (note the polar bodies). Scale bars, 50 μm. Frame interval of 6 minutes, 7 fps.