3D Human Axial Development In Vitro: using novel human in vitro somitogenesis models to study birth defects with patient-relevant iPS cell lines

The vertebral column is a key element of the human body, made up of 33 individual bones arranged linearly in a continuous structure that houses the spinal cord and provides support for the ribs and associated muscles. This structure reveals the segmental nature of our body and has its origin in the process of somitogenesis, an event that takes place in the early embryo where elongation of the body axis leads to the sequential partitioning of a group of cells, the mesoderm, into discrete blocks, called somites. These somites will give rise to progenitors for bone and muscle derivatives, including the vertebrae. Structural abnormalities in vertebrae formation are a common pathology of early development, and span a range of severities that, in extreme forms, can have dramatic effects on the life of individuals. It is therefore critically important that we better understanding the basis of these disorders.

The process of somitogenesis has been studied in many different model organisms, including fish, chicken and mice. Genetic analysis has revealed that the periodic partitioning of the mesoderm is associated with oscillations in the expression of certain genes in space and time. Mutations in these genes lead to a variety of segmentation defects that manifest as aberrations in the organisation of the vertebral column and thoracic muscles, as seen in patients with vertebral column defects. However, there are differences in the development of mice and humans that make it difficult to infer causal, mechanistic relationships between mutations and specific syndromes.

Analysis of these genes in humans is hampered by the difficulty of accessing embryos at the stages of segmentation. One approach to this problem that has emerged over the last few years is Pluripotent Stem Cells (PSCs), which have been shown to recapitulate many developmental events in a laboratory setting. In particular, it is possible to use them to model gene expression patterns associated with somitogenesis in mouse and human. However, these models lack the spatial organization of the embryo that subdivides the length of the mesoderm into constantly proportioned territories and leads to the development of somite structures.

This project aims to develop a new model system to study normal and aberrant human somitogenesis in vitro. This will be achieved by bringing together two groups with complementary expertise. One, based in Kyoto, has established a somitogenesis model from human PSCs that recapitulates oscillations and differentiation into bone and muscle, but in a disorganized manner. A second group, based in Cambridge, has pioneered 'gastruloids', a three-dimensional PSC-based model of early mammalian development recently extended to human PSCs. Gastruloids recapitulate the basic spatial and temporal aspects of somitogenesis in 3D with an anteroposterior polarity. The project will bring the two groups together to optimize humans gastruloids for the study of the formation and differentiation of somites, and to test a number of iPSC lines derived from patients with vertebral defects. The outcome of the project will be the establishment of a new platform for the detection and analysis of pathologies associated with the segmentation of the mesoderm in human embryos, leading to a better understanding of the causes of these syndromes.

Familial or sporadic cases of congenital scoliosis and segmentation defects of the vertebrae (SDV) including spondylocostal dysostosis (SCD) are frequent in human populations. It is thought that the origin of these defects lies in the process of somitogenesis, a core developmental event that lays out the segmented body plan of vertebrate embryos. Our current understanding of this process is based on model organisms such as fish, chicken or mouse. In contrast, our understanding of human somitogenesis is limited to descriptive anatomical studies. This restriction dramatically hampers not only our understanding of normal human development but also of diseases. Recently, using human Pluripotent Stem Cells (PSCs), it has been possible to recapitulate a number of features of somitogenesis, including the oscillatory activity of segmentation clock genes and induction of paraxial mesoderm. Despite these advances, we still lack a human 3D model of axially organised somitogenesis that would allow elucidation of the causes of the segmentation defects observed in many patients.

Here we bring together two groups with complementary expertise and interests. The Alev group (Kyoto) has extensive experience of somitogenesis from the perspective of human disease and has recently described in vitro models based on patient-derived human induced PSCs. However, their systems lack the axial organisation that is a key feature of human somitogenesis. On the other hand, Martinez Arias and Moris (Cambridge) have developed a new human 'gastruloid' model whereby aggregates of human PSCs mimic the dynamic and axial organization of early human development. By bringing the two systems (disease-relevant hiPSC lines and human gastruloids) together, the proposed collaboration has the potential to create a powerful new approach to explore the mechanistic cause of several congenital spinal and muscle disorders.

We expect that the project will have a significant impact in biomedical research by showing how the gastruloid model system can be applied to the study of a specific developmental event with clinical implications. It is a proof of principle project but, on the basis of the results we have obtained with mouse gastruloids, we feel confident of success, certainly in the establishment of the experimental system. The test of patient-derived and disease-relevant human iPSC lines is a most important aspect of the project and, if successful, it could provide an important tool for diagnosis in the medical community, or as an assay for the pharmaceutical industry. We will therefore seek to engage with the clinical and medical community to provide access to information as to the functional and mechanistic implications of genetic observations.

We also expect that the project will have an impact on society more widely, both for the general public and those with experience of congenital abnormalities (patient families and support groups). It will therefore be important for us to engage with those who are interested or concerned with the use of embryonic stem cells in research, in order to highlight the value of in vitro models for understanding disease.

At a local level, the project will allow the establishment in the UK of a diagnosis tool. One of the potential outcomes of the study is an evaluation of the role that the genetic background plays in the expressivity of a mutation. Therefore, if successful it will provide a tool to link up to the genome projects that are ongoing in the UK. For Japan, it will introduce an important technology into a hub of healthy donor and patient derived iPSCs with the associated potential for assay system development and implementation. It will also benefit genetic research and genome projects that are ongoing in Japan. It will further foster and enhance active exchange of expertise and technological advances between biomedical research hubs and centres of scientific excellence located in the UK and Japan. We are confident that our proposed joint research project will help create a synergistic momentum for scientific and societal advances in both the UK and Japan.