A three-dimensional atlas of gene expression during chick development with cross comparisons to the mouse

When people think of an atlas, they think of a world map, showing oceans, country borders, timelines, labelled with names of cities, rivers, etc. The atlas proposed here will show patterns of gene expression layered onto anatomical structures of the developing chick embryo, labelled with the names of structures, genes, etc. Just as a world atlas helps us to understand where we live, an atlas of the developing embryo will help us to comprehend how we form. We all develop from a single fertilized cell which multiplies into a mass of cells, this undergoes complex changes in shape, while growing to form the different organs and tissues that make up our bodies. By imaging chick embryos at different times during development, we will make a detailed map of anatomical structures as they form. The earliest stages are relatively simple as they are flat, but older embryos become increasingly more complicated and we will use a 3D imaging technique called 'Optical Projection Tomography' to image them. An atlas is not very useful without a systematic way to name its features so we will create a standard set of words, which will make it possible to query the atlas using tools based on computer science. But the overt structure of developing embryos hides a further level of anatomy, special groups of cells called 'organizers'. These organizers instruct cells around them so that the correct structures are made in the right place at the right time. Organizers are not always easy to identify; the 'polarizing region' responsible for patterning the digits of the limb for example looks just like the tissue all around it. About half a dozen organizers have been discovered, many through transplantation experiments in chick embryos, and we now know that they are best distinguished by specific genes that are active ('expressed) in their cells. In our project we plan to examine exactly which genes are expressed in four well defined organizers and produce a 3D map of their precise expression patterns in the whole embryo throughout development. Gene expression patterns of ~1000 genes will be mapped. This is a significant number of genes with which to begin to populate the chicken Atlas to be made publicly available to everyone over the internet. To determine what genes are expressed in these four different regions of early chick embryos (hypoblast, Hensen's node, floor plate of the neural tube and limb polarizing region) we will dissect out these tissues and use 'microarrays' to screen for all the genes they express and identify shared sets of genes. Genes expressed in the same place ('synexpression groups') are likely to be involved in the same biological process, so we hope to uncover sets of genes which work together to define an 'organizer'. But why focus on chicks rather than animals closer to humans? Amazingly, organizers and other signalling centres act in similar ways in different species as diverse as fish and man. Thus discoveries in the chick are relevant to human development and chicks are much easier to obtain and dissect than mouse embryos, so these two models are very complementary. The chick atlas however will be based on the same system developed for the mouse thus allowing comparisons. Conserved patterns of expression in chick and mouse will provide strong evidence for genes being functionally related while subtle differences can cast light on why a chick and mouse do not look the same. We will create a database to organize and manage this huge collection of data on gene expression patterns, anatomical structures, genes, etc. and develop new computer tools to query and analyse the data to discover new relationships and new functions for genes in development. This research will lead to a deeper understanding of the basic biological processes which will in turn help understanding of health issues such as congenital abnormalities, cancers and tissue repair.

This project will create a database of gene expression patterns during chick development coupled with a 3D anatomical atlas and ontology of the chick embryo. We will develop new computational tools to query and analyse these data to discover new relationships and functions for genes in development. The database and associated tools will be made available publicly over the internet to form the basis for creating in silico models of chick embryo development. This chick Atlas will be compared and linked to mouse gene expression data to provide new insights into the similarities/differences in the development of these two models. In order to establish this database and its tools, we will collect gene expression data on ~1000 genes expressed in known organizer regions and spatially map the expression patterns onto a series of reference chick embryos. Affymetrix microarray analysis will be undertaken on five embryonic tissues with well characterised organizer function: hypoblast, Hensen's node, notochord, floor plate of the neural tube and polarizing region of the limb bud compared to tissue with no organizer function. Analysis will reveal genes expressed in one or more organizers which will be confirmed by whole-mount RNA in situ hybridization in chick embryos. Expression data will be collected using photography and Optical Projection Tomography and annotated in a 3D Chick Atlas of gene expression patterns, based on the schema used to create the mouse 3D Atlas, EMAP. Emphasis will be on image quality and annotation of ~1000 gene expression patterns between stages XII and HH27. Analysis of 3D chick gene expression patterns and comparison to mouse data will allow precise definition of synexpression groups within known organizers and their conservation or otherwise between these two species, thus characterizing the basis of organizer function in vertebrate development.