nterests
The objective of our work is to understand early vertebrate development at the molecular level. We study this problem in both the amphibian Xenopus, and in the mouse. Xenopus embryos are large and easily manipulated, so that the function of various macromolecules, such as RNA and protein, can be assayed by microinjection into living embryos. Functional assays in Xenopus can then be complemented by genetic knockouts in the mouse, to gain fuller understanding of the normal requirements for gene action in the developing embryo.

Current Projects
Axis formation and Neural induction. Pattern formation in vertebrate embryos proceeds by a series of cell-cell interactions, or inductions. These inductions activate signal transduction pathways to change gene expression. The embryo is very sensitive to small changes in signaling, so that interesting bioactivities can be detected by their effects on embryonic phenotype. Starting with cDNA libraries made from embryos, we inject synthetic mRNA from small pools of library, and assay either embryo morphology or gene expression. Such functional assays allow us to find genes that we can then study to understand the molecular basis for embryonic inductions.
These screens have isolated various molecules that participate in embryonic signaling. Some encode secreted proteins, that either bind to receptors and activate signal transduction pathways, while some block the activity of other signaling molecules. One such molecule, Noggin, can act as a direct neural inducer, and illustrates the principle that blocking a signal can be as informative to cell fate as activating a pathway. Noggin and other similar molecules interact with yet other signals to produce the full range of neural cell types.
While there is an overall understanding of signaling pathways that contribute to formation of the mesoderm and neural plate, there are many aspects of neural and mesodermal patterning that remain poorly understood. Current projects include examination of signaling from the neural plate to the mesoderm, screens for molecules that affect anterior-posterior organization of the neural plate, and screens for new molecules involved in mesoderm induction and patterning.
A crucial aspect of axis formation is the coordinated movement of cells during gastrulation. The development of fluorescent protein fusions to mark cytoskeletal and cell surface changes will allow the functional dissection of this process. We have shown that the coordination of cell movements during convergence and extension of the Xenopus gastrula relies on a conserved Wnt signaling pathway, the planar cell polarity pathway.

Noggin expression in Xenopus laevis
A. Stage 9, vegetal view. Dorsal is up. Staining in the dorsal marginal zone.
B. Stage 9, lateral view. Staining in the dorsal marginal zone.
C. Stage 18, dorsal view, staining in notochord and head mesoderm (prechordal plate).
D. Stage 18. lateral view. Notochord and head mesoderm.

Mouse development. While Xenopus is extremely useful for embryology, it is not suited to genetic experiments. To complement the Xenopus embryology, we make mutations in mouse homologs of genes that we isolate. Thus we can not only study the activity of genes in Xenopus, but also analyze the requirement for the gene rigorously in the mouse. For example, noggin is expressed in several places in the mouse, and we find that noggin function is required for correct nervous system, somite and skeletal development. Noggin expression and function overlaps with that of other BMP antagonists, and we are in the process of analyzing mutations in other antagonists, individually and in combination with the noggin mutation.