Cantas Alev, M.D., Ph.D., Associate Professor

Institute for the Advanced Study of Human Biology (ASHBi), Kyoto University, Japan 

Pluripotent stem cells (PSCs) have increasingly been used to model different aspects of human embryogenesis and organ formation. Despite recent advances in the in vitro induction of major mesodermal lineages and mesoderm-derived cell types, experimental model systems that can recapitulate more complex biological features of human mesoderm development and patterning are largely missing. Here, we utilized induced pluripotent stem cells (iPSCs) for the stepwise in vitro induction of human presomitic mesoderm (PSM) and its derivatives to recapitulate distinct aspects of human somitogenesis. We focused initially on the in vitro recapitulation of the human segmentation clock, a major biological concept believed to underlie the rhythmic and controlled emergence of somites, which give rise to the segmental pattern of the vertebrate axial skeleton. We succeeded to observe oscillatory expression of core segmentation clock genes, including HES7 and DKK1, determined the period of the in vitro human segmentation clock to be around five hours and showed the presence of dynamic traveling wave-like gene expression within in vitro induced human PSM. We furthermore identified and compared oscillatory genes in human and murine PSC-derived PSM, which revealed species-specific as well as common molecular components and pathways associated with the mouse and human segmentation clocks. Subsequent analysis of patient-derived and patient-like iPSCs targeting genes associated with segmentation defects of the vertebrae (HES7, LFNG, DLL3, MESP2) revealed gene-specific alterations of different properties of the in vitro human segmentation clock. Taken together, these findings indicate that our in vitro system recapitulates key features of the human segmentation clock and may be used to provide novel insights into normal and abnormal development of the human axial skeleton.

Prof. Dr. Martin Bastmeyer

Zoological Institute, Cell and Neurobiology, Karlsruhe Institute for Technology (KIT), Germany

Cell behavior and differentiation are not only influenced by biochemical cues but also by physical properties like adhesive geometry, topography, and stiffness of the 3D extracellular environment. In this talk I will discuss how 3D laser nanoprinting can be applied to design 3D cellular microenvironments in the µm range with defined geometries and adjustable flexibility. To achieve a precise and patterned functionalization with biomolecules in 3D three approaches are chosen:
(i) By sequential printing of two different photoresists, composite-polymer scaffolds with distinct protein-binding properties can be fabricated and selectively bio-functionalised thereafter. Cells cultured in these scaffolds selectively form cell-adhesion sites with the functionalised parts, allowing for controlling cell adhesion and cell shape in 3D. Since the elastic modulus of the scaffold material varies between E=140-350 MPa, measurements of cell adhesion forces in relation to adhesion geometry are also feasible. In addition, these scaffolds can be used to mechanically stimulate cells at single defined adhesion sites.
(ii) By combining nanoprinting with an efficient surface photochemistry, also amenable to two-photon activation, it is possible to generate structurally complex 3D microstructures with 3D resolved chemical patterns. Microscaffolds with lattice constants of 10–20 microns can be patterned with protein ligands with a resolution close to one micron using a phototriggered cycloaddition. These techniques have been applied to guide cell attachment in 3D-microscaffolds selectively functionalized with two distinct adhesion proteins.
(iii) By using stimuli-responsive hydrogels, 3D scaffolds can be transferred from passive to dynamic systems. We have fabricated composite 3D scaffolds that allow for the micromanipulation of single cells. These scaffolds allow to directly correlate displacements to cellular forces and to quantify the effects with high throughput.
In summary, the above described 3D scaffolds enable to study the influence of spatial ligand-distribution on cellular differentiation, allow visualizing and measuring cell adhesion forces, and can be used to mechanically stimulate single cells at defined adhesion sites.


Seth Blackshaw, Ph.D., Professor

Department of Neuroscience, Johns Hopkins University School of Medicine, USA

The retina is widely used as a model system for functional studies of neural fate specification in model organisms such as mouse and zebrafish. The development of retinal organoids, moreover, potentially allow such studies to be extended to humans. However, we still lack a comprehensive picture of the gene regulatory networks that control both evolutionarily conserved and species-specific aspects of retinal development, and it is still unclear how well retinal organoids actually mirror the process of retinal development as it occurs in vivo. I will discuss our groups recent application of single cell RNA- and ATAC-Seq analysis to identify gene regulatory networks that control retinal development in zebrafish, mouse and human.  We will discuss functional studies that have arisen from this work, which have identified new genes that control temporal patterning, neurogenesis and specification of both photoreceptor and inner retinal cells.  In addition, we will present new work in which we have extended this analysis to human retinal organoids of various ages, and identified key similarities and differences between the gene regulatory networks that control retinogenesis in vitro and in vivo.

Cole A. DeForest

Cole A. DeForest, Ph.D., Assistant Professor

Department of Chemical Engineering & Department of Bioengineering, Institute for Stem Cell & Regenerative Medicine, University of Washington, USA

The extracellular matrix directs stem cell function through a complex choreography of biomacromolecular interactions in a tissue-dependent manner. Far from static, this hierarchical milieu of biochemical and biophysical cues presented within the native cellular niche is both spatially complex and ever changing. As these pericellular reconfigurations are vital for tissue morphogenesis, disease regulation, and healing, in vitro culture platforms that recapitulate such dynamic environmental phenomena would be invaluable for fundamental studies in stem cell biology, as well as in the eventual engineering of functional human tissue. In this talk, I will discuss some of our group’s recent success exploiting bioorthogonal photochemistry and chemoenzymatic reactions to reversibly modify both the chemical and physical aspects of polymeric cell culture platforms with user-defined spatiotemporal control. Results will highlight our ability to modulate intricate cellular behavior including stem cell differentiation, protein secretion, and cell-cell interactions in 4D.


Prof. Dr. Carsten Grashoff

Institute of Molecular Cell Biology, University of Münster, Germany

To investigate the molecular mechanisms underlying cell adhesion mechanics, we developed a set of single-molecule‒calibrated biosensors that are sensitive to physiologically relevant forces in the low piconewton range and characterized by fast folding/unfolding transitions and reversibility. All biosensors are genetically encoded and can be utilized to determine molecular forces acting across individual molecules in cells. Their application to the focal adhesion protein talin and the desmosomal molecule desmoplakin reveals intriguing differences in how distinct adhesion molecules modulate intracellular force transduction.


Prof. Dr. Jochen Guck

Max Planck Institute for the Science of Light, Germany

Tissues are defined not only by their biochemical composition, but also by their distinct mechanical properties, which cells can sense and respond to. Studying this mechanosensitivity in vivo is often descriptive and correlative. In vitro assays are either only 2D, or in 3D convolve mechanics with porosity and biochemical heterogeneity. This convolution renders testing the relative importance of mechanosensitivity in realistic environments challenging. Here, we present novel colloidal crystals as modular 3D scaffolds where these parameters are principally decoupled. By using monodisperse, protein-coated PAAm microgel beads with well-defined elastic properties as building blocks, variable stiffness regions can be realized by an additive process within one 3D colloidal crystal. Using these mechanically patterned colloidal crystals, we have demonstrated durotactic fibroblast migration and mechanosensitive neurite outgrowth of dorsal root ganglion neurons in 3D. Further, the PAAm hydrogel beads also find many other applications in mechanobiology, for example as standardized mechanical cell mimics for calibration of cell mechanics measurements and for assaying the importance of deformability in vascular circulation, or as cell-scale stress sensors in developmental processes.

Michael Heymann

Jun.-Prof. Dr. Michael Heymann

Institute of Biomaterials and Biomolecular Systems, University of Stuttgart, Germany

Living systems have a stunning ability to self-organize in space and time. Many remaining grand challenges in biology and medicine come from our inability to comprehend the underlying molecular-scale phenomena in a complex context such as a multi-component mixture or a cell. We advance two photon stereolithography based microfabrication to create and comprehend biomolecular structure and function across scales.

This entails novel ultracompact microfluidic approaches to time-resolved structural biology to record ‘molecular movies’ of macromolecular conformational changes at the atomic scale. This allows to determine the structures of transient states and thereby kinetic mechanisms of substrate turn-over during enzyme catalysis. We could follow the catalytic reaction of the M. tuberculosis β-lactamase with the 3rd generation antibiotic ceftriaxone with millisecond to second time resolution at 2 Å spatial resolution.

In extending this technology to synthetic biology, we can reconstitute functional biological and biomimetic systems from the bottom up with unprecedented precision and throughput. For instances to compartmentalize the E.coli MinDE protein oscillator, that positions the cell division machinery at mid-cell, into physiologically relevant three-dimensional model compartments, such as lipid vesicles exhibiting active shape changes. In current efforts, we are developing novel protein photoresists to nano-3D-print sub-cellular compartments with the highest achievable functional conformity to cellular structures in vivo. In first proof-of-principle experiments we structured a contractile eukaryotic cell division model.

Such model systems will allow to design and program dynamic biological states far from equilibrium to investigate spatiotemporal self-organization principles in biology that by lack of suitable tools have previously been inaccessible to experimental quantification. Our ultimate objective is to decipher the principles of synchronization, morphogenesis, and differentiation in confined geometries, as well as biochemical information processing and chemo-mechanical coupling at scales ranging from the nanoscale to the full organ. This will enable previously inconceivable avenues to investigate and to program fundamental aspects of biological self-organization and disease, to uncover new biophysical principles, as well as healthcare and biotechnology applications.


Majlinda Lako, Ph.D., Professor

Institute of Genetic Medicine, Newcastle University, United Kingdom

Prof. Majlinda Lako completed her PhD studies at the Human Genetics Department of Newcastle University in 1998. Following her postdoctoral training at Durham University, she returned to Newcastle to create her own independent research group in 2003 working in human pluripotent stem cells. The research aims of Lako’s group are to understand and define the early events occurring in human embryogenesis with special focus on eye formation and developing new treatments for eye disease.  We are engaged in several large research programmes that aim to define good manufacturing protocols for deriving functional corneal and retinal cells that can be used for drug testing, disease modelling and cell based replacement therapies.

In this talk, I will focus on our efforts to optimise the generation of light responsive retinal organoids and their application to disease modelling, photoreceptor transplantation and drug discovery/repurposing.

Fraunhofer Institute for Interfacial Engineering and Biotechnology IGB in Stuttgart. Jun. Prof. Dr. Peter Loskill, Attract Group Leader Organ-on-a-Chip.

Jun.-Prof. Dr. Peter Loskill

Fraunhofer Institute for Interfacial Engineering and Biotechnology IGB, Germany

Faculty of Medicine, Eberhard Karls University Tübingen, Germany

Drug discovery and development to date has relied on animal models, which are useful, but fail to resemble human physiology. The discovery of human induced pluripotent stem cells (hiPSC) has led to the emergence of a new paradigm of drug screening using human patient- and disease-specific organ/tissue-models. One promising approach to generate these models is by combining the hiPSC technology with microfluidic devices tailored to create microphysiological environments and recapitulate 3D tissue structure and function. Such organ-on-a-chip platforms (OoCs) or microphysiological systems combine human genetic background, in vivo-like tissue structure, physiological functionality, and “vasculature-like” perfusion.

Using microfabrication techniques, we have developed a variety of OoCs that incorporate complex human 3D tissues and keep them viable and functional over multiple weeks, including “Retina-on-a-chip”, “Choroid-on-a-chip”, “Heart-on-a-chip”, “Pancreas-on-a-chip and a “White adipose tissue(WAT)-on-a-chip”. The OoCs generally consist of three functional components: organ-specific tissue chambers mimicking in vivo structure and microenvironment of the respective tissues; “vasculature-like” media channels enabling a precise and computationally predictable delivery of soluble compounds (nutrients, drugs, hormones); “endothelial-like” barriers protecting the tissues from shear forces while allowing diffusive transport. The small scale and accessibility for in situ analysis makes our OoCs amenable for both massive parallelization and integration into a high-content-screening approach.

The adoption of OoCs in industrial and non-specialized laboratories requires enabling technologies that are user-friendly and compatible with automated workflows. We have developed technologies for automated 3D tissue generation as well as for the flexible plug&play connection of individual OoCs into multi-organ-chips. These technologies paired with the versatility of our OoCs pave the way for applications in drug development, personalized medicine, toxicity screening, and mechanistic research.

Matthias Lutolf

Prof. Dr. Matthias Lutolf

Laboratory of Stem Cell Bioengineering, Ecole polytechnique fédérale de Lausanne (EPFL), Switzerland

Bioprinting promises enormous control over the spatial deposition of cells in three dimensions, but current approaches have had limited success at reproducing the intricate micro-architecture, cell-type diversity and function of native tissues formed through cellular self-organization. We introduce a three-dimensional bioprinting concept that uses organoid-forming stem cells as building blocks that can be deposited directly into extracellular matrices conducive to spontaneous self-organization. By controlling the geometry and cellular density, we generated centimetre-scale tissues that comprise self-organized features such as lumens, branched vasculature and tubular intestinal epithelia with in vivo-like crypts and villus domains. Supporting cells were deposited to modulate morphogenesis in space and time, and different epithelial cells were printed sequentially to mimic the organ boundaries present in the gastrointestinal tract. We thus show how biofabrication and organoid technology can be merged to control tissue self-organization from millimetre to centimetre scales, opening new avenues for drug discovery, diagnostics and regenerative medicine.

Alfonso Martinez-Arias, Professor

Department of Genetics, University of Cambridge, United Kingdom

When small, specified numbers of mouse Embryonic Stem Cells are placed in defined culture conditions they aggregate and initiate a sequence of pattern forming events that mimic the events that take place in the embryo: they undergo symmetry breaking, gastrulation like movements, axial specification and germ layer organization. We can culture them for up to seven days to reach a stage comparable to E9.0 in the mouse embryo and exhibit a similar organization including three orthogonal axes with associated asymmetries. This experimental system can be used to gain insights into the process of gastrulation and axial organization as well as the emergence of the primordia for tissues and organs. I shall be discussing specific examples and the implications these have for the theoretical and practical understanding of developmental events in mammals and our efforts to extend the system to human Pluripotent Stem Cells.


  1. Beccari, L., Moris, N., Girgin, M., Turner, D., Baillie-Johnson, P., Cossy, A.C., Lutolf, M., Duboule, D. and Martinez Arias, A. (2018) Multiaxial self organization properties of mouse embryonic stem cells gastruloids. Nature https://www.nature.com/articles/s41586-018-0578-0
  2. Turner, D. et al. (2017) Anteroposterior polarity and elongation in the absence of extraembryonic tissues and spatially organized signaling in Gastruloids, mammalian embryonic organoids. Development 144, 3894-3906
  3. van den Brink, S. et al. (2014) Symmetry breaking, germ layer specification and axial organisation in aggregates of mouse ES cells. Development 141, 4231-4242.
11_Thomas A

Thomas A. Reh, Ph.D, Professor

Health Sciences Center, University of Washington, USA

The neural retina is subject to a wide range of neurodegenerative diseases, including inherited degeneration and dysfunction, as well as age related diseases. Although great progress has been made in understanding the genetics and risk factors for many of these diseases, translating these approaches to clinical trials and approved therapies remains challenging. This is in part due to the lack of a suitable in vitro screening system or cell line for identifying compounds that can effectively engage targets and for lead optimization and off-target effects. A key breakthrough in the field was the development of retinal organoids from iPSCs. Retinal organoids are a particularly attractive model, since these develop many features of normal retina; organoids mimic the lamination of the retina, and have been successful in generating the major retinal cell types. Retinal organoids have been shown to be quite similar to human fetal retina using histological methods and genomic approaches (eg. RNAseq at the bulk and single cell level, DNAseq, ATACseq, etc). Retinal organoids, however, are still not a perfect substitute for the human retina. Organoids lack non-neural cell types, a continuous layer of RPE, microglia and endothelial cells, cells that are well known to participate in major retinal pathologies. To better understand the limits of in vitro retinal development, we developed an organoid-like model using primary retinal tissue, which we call retinospheres. Retinospheres can be generated from any age of retina we have tested to date, and a detailed comparison using histology and scRNAseq between retinospheres and fetal retina have shown they are remarkably similar. Retinospheres also retain their lamination much better than stem cell derived organoids and they retain characteristic features of the region of the retina from which they originated, eg. fovea. Importantly, retinospheres contain the non-neural cells not typically present in organoids. Moreover, fusions between organoids and retinospheres allows us to better understand their differences and develop optimal models from stem cells.


Prof. Dr. Ute Schepers

Institute of Toxicology and Genetics (ITG), Karlsruhe Institute of Technology (KIT), Germany

During the last decade many approaches have been made to generate novel materials for the 3D recnstruction of neuronal tissues such as the blood brain barrier, the neurovascular junction or even the retina by tissue engineering through 3D bioprinting. The materials have to be tailored to resemble all the properties to allow for stem cells such as iPS cells to initiate their differentiation program to the neuronal tissue. In addition they have to be biocompatible and even degradable for long periods of time during tissue generation. 3D-printed tissues often resemble organoids from different cells without the appropriate vascularization enabling the tissue to be connected to nutrition. Endothilium of blood vessels also prime the tissue for the expression of maturation factors. Many groups ave established a model of 3D-retina organoids. However, they often miss the correct vascularizationand the reconstruction of the inner and outer bood-retina barrier. 3-D retina organoids are initiated by induced pluripotent stem cells outside of traditional 2D culture systems to result in multiple cell types that organize along three dimensions into microphysiological systems. One goal is the development of materials that allow for the stimulated and directed growth and differentiation of iPSCs into a 3D retinal tissue that resemble a nature like 3D neuronal environment. The blood-retinal barrier (BRB) is essential to maintaining the eye as a privileged site and is essential for normal visual function. The BRB, separated in inner and outer barrier, is a particularly tight and restrictive physiologic barrier that regulates ion, protein, and water flux into and out of the retina. The inner BRB being formed of tight junctions between retinal capillary endothelial cells and is similar to the blood brain barrier. Based on our experience on the reconstruction of the BBB we developed a blood vessel system that allowes together with our novel ultrafast curing material the reconstruction of the retina barriers by combined approached from direct laser writing and 3D bioprinting.


Prof. Dr. Ulrich Schwarz

BioQuant Center for Quantitative Biology / Institute for Theoretical Physics (ITP), Heidelberg University, Germany

Cells look and behave differently in three-dimensional scaffolds than on two-dimensional surfaces, but the most important underlying processes determining shape, mechanics and movement are the same: membrane protrusions due to actin polymerization and myosin-based contractility of the actomyosin cortex and stress fibers. I first will discuss the interplay between tension and elasticity that characterizes the cell envelope in two dimensions, and then extend this viewpoint to three dimensions. Next I will address models that allow us to formulate dynamical versions of tension-dominated systems, namely cellular Potts and phase field models. Given that we can model the forward problem, we finally can ask how to control cell shape by solving the inverse problem and designing scaffolds that result in a desired functionality. For 3D hybrid organotypic systems that mimic the retina, such a desired functionality might be light scattering determined by the contrast between cell nuclei and cytoplasm.

Magdalene Seiler_neu

Magdalene J. Seiler, Ph.D., Associate Professor

Department of Physical Medicine & Rehabilitation, Department of Ophthalmology, Sue and Bill Gross Stem Cell Research Center, University of California, Irvine, USA

Purpose. Human embryonic stem cell (hESC)-derived retinal organoids (ROs) improve visual function after transplantation into retinal degeneration (RD) models (McLelland et al, 2018, IOVS). Advanced imaging techniques (fluorescence lifetime microscopy [FLIM] and hyperspectral imaging [HSpec]) provides non-invasive data for quality control and long-term in vitro follow-up. In this work, hESC-derived retinal organoids, produced by a cGMP compatible process, were followed by 2-photon microscopy in vitro prior to surgical transplantation; and in vivo by OCT imaging after transplantation to the subretinal space of nude RD rho S334ter-3 (RN) rats.

Methods. A scalable cGMP compatible process was established for the generation and characterization of a Working Cell Bank (WCB) of CSC-14 hESCs (NIH 0284). hESC-derived retinal organoids were characterized by immunohistochemistry (IHC), flow cytometry and qPCR. 2-photon excitation microscopy (2PE) was used to collect metabolic information from intrinsic fluorophores: NADH (FLIM), and retinol (HSpec) inside organoids with subcellular resolution (Browne et al, 2017, IOVS) up to 6 months in vitro. FLIM images were taken using 740nm pulsed excitation (Zeiss LSM 710). HSpec fluorescence emissions were taken in the range of 420 nm to 690 nm. Data were analyzed by SimFCS (Global Software) via the phasor approach. Retinal organoid sheets (differentiation day 30-90) were transplanted to the subretinal space of RN rats (P31-51). Transplants were monitored in vivo by Optical Coherence Tomography (OCT). Visual function was accessed by optokinetic tests (OKT) and superior colliculus (SC) electrophysiology. Ex vivo sections through transplants were stained with hematoxylin & eosin (H&E), or processed for IHC to label human donor cells, retinal cell types and synaptic markers.

Results. The WCB of CSC-14 hESCs was characterized using the following metrics:  viability, identity (Oct4); karyotype stability; sterility and neural differentiation potential.  This WCB was used to generate all ROs. Long-term imaging data of retinal organoids (>180 days) demonstrated metabolic activities confirming overall cellular viability. Initially, a shift from glycolytic to oxidative metabolic activities was observed. As time progressed, glycolysis became predominant on the surface of the organoids. HSpec images showed retinol distribution on the surface. IHC of retinal organoids shows early lamination and development of retinal cell progenitors. Organoids selected for transplantation showed early lamination. Post-transplantation OCT imaging revealed transplant development and photoreceptor rosettes. Transplanted eyes showed vision improvement by OKT and SC recording. Transplants developed different retinal cells including photoreceptors; and integrated with the host retina.

Conclusions. A WCB of CSC-14 hESCs was successfully established and meets FDA requirements. Retinal organoids showed a metabolic shift in long term culture, from glycolytic (proliferative) to oxidative (differentiated) state, and back to the glycolytic surface (indicating a photoreceptor layer). Retinal organoids mature further after transplantation, develop photoreceptors, integrate into the host retina, and improve visual function.

Support: California Institute for Regenerative Medicine (CIRM) TR1-10995; RPB unrestricted grant to UCI Department of Ophthalmology; ICTS KL2 TR001416


Prof. Dr. Motomu Tanaka

Institute of Physical Chemistry (PCI), Heidelberg University, Germany

Dysfunction of the corneal endothelium reduces the transparency of the cornea and can cause blindness. Currently, the clinical treatment inevitably involves the transplantation of donor corneas, as human corneal endothelial cells have an extremely low proliferative capcity in vivo. The successful in vitro expansion of endothelial cells enables the restoration of a functional cornea via intraocular injection of endothelial cells in suspension [1], yet a substantial amount of the cultured cells is lost by destructive quality assessment.Recently, we established a quantitative measure (physical biomarker) by shedding light on the collective order of the cells by treating them as 2D colloidal assemblies [2]. The second derivative of potential of mean force (spring constant) calculated from phase contrast imaging and from specular microscopy can be used as a noninvasive index for the quality assessment of corneal endothelial cells in vitro and for the long-term prognosis of corneal restoration in patients in vivo, respectively.Our data suggest that this new biomarker may enable a shift from passive monitoring to pre-emptive intervention in patients with severe corneal disorders, which is a major health issue in the aging society.

Quantification of collective order of human endothothelial cells in in vitro culture (upper panels) and in restoring in vivo corneas in patients (lower panels) using one physical biomaker. 


[1] S. Kinoshita, N. Koizumi, M. Ueno,.. C. Sotozono and J. Hamuro, New Eng J Med 378 (2018) 995.

[2] A. Yamamoto, H. Tanaka,….C. Sotozono,.. S. Kinoshita, M. Ueno and M. Tanaka, Nat Biomed Eng, DOI:10.1038/s41551-019-0429-9 (2019).


Prof. Dr. Martin Wegener

Institute of Applied Physics (APH) / Institute of Nanotechnology (INT), Karlsruhe Institute of Technology (KIT), Germany, Spokesperson of the Cluster of Excellence “3D Matter Made of Order” (3DMM20)

In this talk, I shall gave an introduction into and an overview of the activities of the Excellence Cluster 3DMM2O. On the technology side, the scientific challenges pursued by the Cluster can be nicknamed as “finer, faster, and more”, i.e., advance molecular materials and technologies in terms of resolution, speed, and multi-material printing by orders of magnitude. On the application side, the Cluster aims at functional 3D hybrid optical and electronic systems, 3D artificial materials called metamaterials, and at reconstructing functioning organotypic systems by using 3D scaffolds for cell culture. In the talk, I will emphasize manufacturing technologies relevant for 3D organotypic systems, especially 3D laser nanoprinting.

Version 2

Prof. Dr. Joachim Wittbrodt

Centre for Organismal Studies (COS), Heidelberg University, Germany

The vertebrate eye is formed in a very stereotypic manner and the neuronal and non-neuronal cell types are arranged in an almost crystalline array. I will touch the molecular on morphogenetic aspects of vertebrate eye formation and will in particular address the life-long growth in non-mammalian species. In fish and amphibia the eye is formed in two phases, initially from multipotent retinal progenitor cells that evaginate from the lateral diencephalon. Those undergo a major morphogenetic rearrangement and this retinal flow eventually transforms the vesicle into the characteristic optic cup. Concomitant with the flow, differentiation of retinal cell types is initiated in the central retina by signals emanating form the optic stalk. The second phase of retina formation starts subsequently and is driven by a stem cell niche in the periphery of the just formed eye cup, the ciliary marginal zone. I will illustrate the properties of stem and progenitor cells and will eventually touch upon how the action of stem cells is particularly canalised by the immune system. It will be challenging to disentangle the evolutionarily tightly coupled processes into individual building blocks that, relying on the properties of self-organisation, can be recombined in a new context. Results on self-organising retinal organoids from diverse species give first encouraging insights and open a perspective for systematically tackling the impact of the physical environment.