Transcriptional responses to FGF signalling during germ layer specification

All animals with backbones develop in similar ways. Although frogs would seem to be very different from humans, many of the processes and proteins involved in the development of their embryos are very similar. We use Xenopus frog embryos to study early development because they lay many hundreds of eggs which can easily be fertilized in a Petri dish. The large number of eggs produced is important because they provide lots of material for our experiments and chemical purifications. Furthermore, the embryos develop very rapidly and reach the swimming tadpole stage in just three days. The embryos are quite large (1 mm diameter) which means they can be injected with chemicals that either activate or inhibit the function of particular protein or gene involved in development. We are interested in a group of proteins known as the fibroblast growth factors (FGFs). FGFs are found in all animals and play an important role in allowing cells within the embryo and adult to communicate with each other. Such signals passing from one cell to another are important in development because they are involved in the process by which individual cells decide what they will form in the adult. FGFs are particularly important in the decision of whether a particular group of cells will form tissues such as muscle, bone, kidney or the nervous system. The FGFs tell a cell which genes to turn on and because each gene codes for a protein, this ultimately tells a cell which proteins to make. For example, FGFs can tell cells in the embryo to become muscle. In this situation the FGF signal tells a cell to turn on genes which allow the production of proteins needed for a muscle to contract. We already know some of the genes that FGF signals will activate but we would like to know them all. Recent advances in technology mean that we can now attempt to identify all those genes which are turned on by FGF signals. The experiments that we plan are very simple. We will take samples of very early frog embryos and stimulate FGF signalling in these samples. We will then compare which genes are activated in these samples versus samples in which we have not stimulated FGF signalling. This should for the first time give a large scale view of the genes that are activated by FGF signals in the developing embryo. Our work will be of interest to people studying the early development of animals and people interested in how the bodies of adults maintain and repair themselves. This is because the FGFs are involved in many processes in the adult as well as the embryo. Once we have identified the FGF target genes, which in itself is very useful information, we will begin to investigate the nature of the proteins made from these genes. We will also attempt to discover what these proteins do during development by over activating or inhibiting their production in the early frog embryo. We also plan to use the large amounts of information gained from these experiments to make computer based models of how all these genes interact with each other during development. Ultimately these models will allow us to understand better the decisions that cells need to make during the development of all animals including humans. They might also allow us to predict the consequences of inhibiting or activating a particular gene involved in development.

Fibroblast growth factors (FGFs) constitue a large family of polypeptide growth factors. During early development FGF signalling is involved in germ layer specification and patterning. Much work has focused on the role that FGFs play in regulating gene expression within the mesoderm. Furthermore, there is now good evidence that the FGFs are important regulators of mesodermal cell movements in both vertebrates and invertebrates. Data also indicate that FGF signalling is required for the induction and patterning of the vertebrate nervous system. In the adult FGFs are involved in homeostasis and have been shown to be positive regulators of angiogenesis and the wound healing response. Given the importance of FGF signalling the transcriptional targets of FGF signalling are of considerable interest. Previous studies have identified a number of FGF dependent gene regulatory pathways, including a highly conserved pathway involving the T-box transcription factor brachyury. The multiple FGF ligands and receptors have different biological activities in early development. However, the transcriptional responses that underlie these distinct activities are unclear at present. The amphibian Xenopus laevis is a classical model for the study of cell signalling and gene regulation. Much of the present understanding of FGF function in early development comes from experiments carried out in Xenopus. The utility of X.laevis as a model for studying gene regulation in development has been greatly enhanced by the recent introduction of large scale commercially available microarrays. In a set of preliminary experiments we have identified a group of about 100 genes, including 30 novel sequences, which are either significantly down regulated or up regulated in response to inhibiting FGF signalling in early gastrula stage Xenopus embryos. We plan to use X.laevis and the closely related species X.tropicalis to undertake a microarray based analysis of the transcriptional responses elicited by signalling through two different FGF receptors and ligands, which have previously been shown to have rather different biological activities in early development. Using a tissue explant system and drug inducible versions of FGFR1 and FGFR4 we will analyse gene expression activated by signalling through these receptors during blastula and gastrula stages. In a similar set of experiments we will examine differential gene expression activated by FGF4 and FGF8 ligands during blastula and gastrula stages. We will establish detailed temporal and spatial expression patterns for these genes. Part of this analysis will be to carry out microarray analysis of gene expression in normal embryos from fertilization until mid-neurula stage. These analyses will allow us to classify the FGF targets into synexpression groups of co-regulated genes, reference to co-regulated genes with known function will be a powerful tool in establishing possible functions for novel genes. Furthermore, the global analysis of gene expression will provide a valuable resource to the developmental gene regulation community. We have established collaborations which will allow us to best exploit these resources and integrate these large data sets into the 'Mesendoderm Regulatory Network Project' An aim of this project is to identify the function of FGF targets in relation to known developmental processes. We will identify full length open reading frames in novel cDNAs and will determine function for encoded proteins by a combination of overexpression and inhibition in Xenopus tropicalis. The reduced genetic polymorphism of the diploid genome of X.tropicalis versus the tetraploid genome of X.laevis means that it is a more favourable model in which to carry out anti-sense mediated knockdowns of gene function. This project will for the first time give an overview of the different transcriptional responses elicited by signalling activated by different members of the FGF ligand and receptor families.