Identification and characterisation of genes which are critical for murine gastrulation

Abstract

Gastrulation, a critical stage of embryogenesis, converts a group of relatively undifferentiated cells into the body plan of the growing foetus. Embryonic cells are allocated to one of three definitive germs layers and are organised with respect to three embryonic axes (anterior-posterior, dorsal-ventral and left-right). By the end of gastrulation the primordia of most organs and physiological systems have been established. In mammals, gastrulation occurs just after the embryo implants into the uterine wall. The small size of the mammalian gastrula and its inaccessibility means that, despite its central importance to embryonic development, gastrulation remains one of the most challenging and underexplored areas of mammalian biology. The study of other complex biological processes has been greatly facilitated by the use of genetic screens, which employ a powerful mutagen to introduce mutations into the germline of an organism and then examine descendants for altered biology (i.e. an abnormal phenotype) followed by post-hoc identification of the underlying genetic alteration. The utility of these screens is dependent upon quick, robust assays to identify pheno-deviants and it has long been thought that practical issues prevent this approach being applied to mammalian gastrulation. In this thesis, I examine this notion. In the first part of the thesis, I use phenotype analysis to infer the function of the gene altered in the katun mouse strain which was identified in a screen for mutations that alter embryonic development and was found to have a gastrulation phenotype. I show that the left-right axis defects in this strain are likely caused by dysregulated Wnt signalling at gastrulation which interferes with endoderm development. This confirms that mutants recovered from genetic screens can be used to infer gene function at gastrulation. In the second part of the thesis, I report results form a prototype, genome-wide, screen for recessive mutations that alter gastrulation. The screen was able to overcome the perceived difficulties of small litter sizes, of isolating a large number of gastrulation stage embryos, of distinguishing phenotypes from natural embryonic loss and of working with limiting amounts of genomic DNA. I demonstrate that mutant pedigrees can be identified, they can be maintained in the absence of genetic markers via test breeding protocols and that low-resolution meiotic mapping can be combined with next generation sequencing technologies to identify the causative mutation. This

confirms that gastrulation is amenable to genetic dissection by mutagenesis and I suggest a protocol that could be used to conduct a larger scale, efficient screen to identify the genes that direct mammalian gastrulation.