Human gastruloids: an in vitro system for the study of human gastrulation

The discovery that mouse and human Embryonic Stem Cells (ESCs) can be coaxed in vitro to form tissues and organs has raised expectations that the associated protocols for guided differentiation in culture will provide a basis for the successful development of a rational Regenerative Medicine. However it is early days. At the moment the most successful experiments are being done with mouse ESCs because they allow a wide range of experiments in a well understood and standardized system. Also, while there are protocols that can generate organ like structures -called organoids- in vitro, the protocols are neither robust nor reproducible enough to be if much use. A reason for this is that while we have made great advances in our understanding of the genes that control the emergence of specific cell types, we have little understanding of how the expression of these genes is translated into the three dimensional structures that characterize tissues and organs. Understanding this requires a user friendly three dimensional experimental system that mimics the embryo. The situation is even more challenging in the case of human tissues and organs since, in this case, we have very little insight on the development of human embryos and negligible knowledge of their cellular and molecular underpinning. Making progress in this important area of research is also limited by obvious ethical issues and regulations associated with the use of human embryos.

ESCs provide an alternative experimental system but current studies with human ESCs restrict themselves to adherent cultures or use the mouse and mouse ESCs as a standard reference for human biology because of its superficial similarities in the underlying developmental processes. However, there are significant differences between the two species that, when considered in detail, highlight the need for a better understanding of the specific cellular and molecular features underlying human embryogenesis. Furthermore, adherent cultures do not reflect the complexities and interactions that emerge in three dimensional structures. Increasingly, other primates are being studied and used as referenced but this does not remove the ethical issues that are associated with this kind of study.

Over the last few years we have established an experimental system based on mouse ESCs which, surprisingly, reproduces to a very high degree the early stages of embryonic development and patterning, most notably, the process of gastrulation. The structures mirror the embryo until, approximately, stage 10 (almost half way through its development). An important feature of these organoids, that we call 'gastruloids' because of their ability to imitate gastrulation, is that they develop in the absence of the extraembryonic tissues that mediate the interactions between the embryo and the mother. A second feature is that they lack a head and brain and consist,only, of structures from the neck down.

Here we propose to build on these observations and preliminary results that we have obtained to establish a similar system for human ESCs. This will allow an unprecedented access to the mechanisms underlying the early stages of human development and, in particular, the cellular and molecular processes that underlie the establishment of primordial for tissues and organs. Furthermore, the lack of a brain and a head and importantly, the inability of these organoids to interact with the mother account for many of the ethical considerations that apply to embryos.

Over the last few years, together with our collaborator Matthias Lutolf at EPFL (Lausanne, Switzerland), we have developed a robust ESC based 3D model of mouse embryogenesis in culture. In our system, small and defined numbers of ESCs undergo symmetry breaking, germ layer specification and gastrulation like movements in a manner that mimics the embryo. These processes happen with spatial and temporal patterns that parallel the embryo and are both robust and reproducible. The resulting organoids, that we call 'gastruloids', develop the primordial of ectodermal, mesodermal and endodermal derivatives with a remarkable axial organization. We feel that we are now in a position to extrapolate this system to human ESCs and have obtained preliminary results suggesting that this is indeed feasible.

Our background work with mouse ESCs and the preliminary results with human ESCs form the basis for this project in which we propose to ask

1. What are the molecular events that underlie the establishment and elaboration of the embryonic axes in humans.
2. The sequence of cellular and molecular events underlying the process of gastrulation.
3. How do early developmental events in human and mouse embryos compare with each other and what are the relative roles that signal and transcription factor networks play in these differences.

The project will deliver a unique, robust, reproducible and ethically safe system for the study of post implantation human development, gastrulation and the specification of the primordial for tissues and organs. The experimental system will open up significant opportunities in the areas of disease modelling, iPSC validation, drug testing and, fundamentally, knowledge about human biology which, at the moment is inaccessible.

Understanding human development is a major goal of contemporary biological sciences. While there are good data basis for human embryos (Human Developmental Biology resource HBDR - http://www.hdbr.org/about/- , the Carnegie Collection - http://www.ehd.org/virtual-human-embryo/- and the Kyoto collection - http://bird.cac.med.kyoto-u.ac.jp/index_e.html-) with the exception of HBDR, these are descriptive of human embryology rather than a basis for the exploration of developmental events. The HBDR offers a range of opportunities to explore many aspects of normal and pathological development through gene and protein expression but the curated samples start at Carnegie Stage 12 (CS12, about 30 days postfertilization) which is much later than the stages proposed to study here, thus emphasizing the uniqueness and significance of the project.

The mechanistic understanding of the developmental processes requires cellular and molecular analysis on fresh tissue which, with the exception of preimplantation stages, it is not possible to obtain for obvious ethical reasons. This is particularly true of the process of gastrulation, a key event in the development of any organism at the end of which there is a spatial organization of the primordia for the tissues and organs of the organism. In humans, gastrulation starts at day 13 of development (CS5) and it is agreed that any experiments or invasive observations on human embryos should stop at this time. This rule, known as the day 13 rule, limits our ability to understand significant events of human biology. Developments in the field of human ESCs have opened up the possibility of differentiating these cells in adherent cultures and Embryoid Bodies but these contraptions while providing some insights into cell fate choices and basic genetic underpinning of different cell types, are far from the complexity and structural organization of an embryo. Our finding of how to coax mouse ESCs into 'gastruloids', embryonic organoids that mimic mouse embryos up to an equivalent of 9.5 days of development, represent a significant leap in our ability to study development in vitro.

Preliminary results presented in the main application and which form the basis for this application indicate that it will be possible to extrapolate this system to hESCs. In doing so we shall establish a unique system to study human development in vitro during and beyond gastrulation. The system is ethically sound as the outcome of the protocol are not whole embryos and lack the extraembryonic tissues, hence cannot interact with the mother. In any event, we are in contact with the local and national authorities to monitor the ethical implications of our experiments.

The experimental system that we propose to establish will open up unprecedented possibilities to understand the early stages of human development, test the developmental potential of iPSCs, create disease models -in particular for neonatal and infant diseases-, provide a platform for drug screening and also, understand the basis of early miscarriages which, in many instances, are thought to be related to the process of gastrulation.
The project will create opportunities for interactions with HBDR that we shall explore ways as the results suggest possibilities and ways to do this. Importantly there will be aspects of the research with commercial potential that we shall explore with Cambridge Enterprise (the University commercial arm) which is aware of our mouse work.