Introduction

The development of an embryo is a beautiful and elaborate example of the sophistication of biological systems. Beginning with a single cell, rounds of cell division build a ball of cells called the blastula that begins, using combinations of molecular cues inherited from the mother and instructive cell signals of its own, to partition itself into distinct domains. This is the beginning of the process by which cells organize into the first primitive regions of the embryo, which will in turn grow and elaborate into the growing fetus. What is most striking to the casual observer is the gradual and progressive development of a recognizable physical form of the embryo that slowly shapes itself into a small, growing animal. The development of form is called morphogenesis, and its study is both a traditional part of developmental biology and the recent focus of renewed interest, driven by the availability of new imaging methods and growing molecular insight into its mechanisms. Developmental biologists have carefully characterized the stages of morphogenesis and have identified key transitional periods: blastula, gastrula, and neurula stages, which define the form of the embryo during its development and imply the changes that might take place between them. A great deal of attention has been given to the cellular processes that might enact these changes thought to involve such processes as changes in association between cells, the migrations of cells, and the reshaping of sheets of cells. A simple question to ask is “Where do cells of a particular region of the body originate, and how do they become located as they do?” A classic way to approach such a question is to generate a fate map. A fate map is an experimental tool to explore the reorganization of cells from one stage of development to a later stage. A cell is marked at a recorded location, then it or its progeny are relocated at a later stage. If cell rearrangements are stereotypical from embryo to embryo, then a map can be constructed by repeating this many times from different recorded positions. This approach has been widely used, employing a variety of means to label the cells and varying degrees of sophistication in classifying final cell fate, ranging from scoring their location by tissue or organ or analyses of gene expression patterns.

Classic Fate Maps: Hurdles of Movement and Resolution

In revealing the locations of cells during the morphological transitions of embryo to fetus, classic fate maps have proven fundamental to our understanding of the mechanisms of cell fate determination. Indeed, the fate map of the nematode Caenorhabditis elegans (1) has revealed an almost invariant cell lineage that can be traced in its entirety from the fertilized egg to the larva. This has made possible the manipulation of individual cells to examine the consequences on development, thereby demonstrating the importance of asymmetric inheritance and the interactions between both equivalent and nonequivalent cells in instructing cell fate. However, fate maps of higher organisms, including the frog Xenopus laevis (2), mouse (3), chick (4), and the teleost fish, zebrafish (Danio rerio) (5), and medaka (6) (Oryzias latipes), have revealed a less determinate development, which can be attributed in part to the extent of cell movement required to orchestrate morphogenesis. Cell mixing impedes the definition of clonal boundaries. Instead, populations of discrete fates overlap one another in the fate map, making the interpretation of tissue origins and their morphogenetic development difficult. Most illustrative of the spectrum of distortion caused by cell movement are the fate maps of Xenopus laevis and zebrafish. Despite the consistency of cell lineages, the slow mixing of muscle progenitors in the frog prevents elucidation of the origin of even the most simple muscular derivative, the somite (2,7), while cell mixing is so extensive during early embryogenesis in zebrafish that it has proven impossible to construct a reliable fate map before the onset of gastrulation (the establishment of the three principle body layers) (8,9). These examples reveal the inherent problems of classic fate maps: there is usually little or no direct information about how morphogenesis has occurred and, since at most only one or a few cells are analyzed perembryo, a complete map is an average view, limited in resolution by how well cell locations can be defined within the embryo. Consequently, cell movement is inferred, not measured, from a series of static snapshots. In complex organisms in which the possible routes to attain adult structure are many, this clearly compromises our ability to draw conclusions about the actual mechanisms involved. Indeed, nowhere has this lack of resolution been more evident than when considering axial patterning in the frog (10,11). For over a century, only the left-right and dorsoventral (back-to-belly) axes had been assigned to amphibian embryos. In an unprecedented turn of events, a series of careful lineage reconstructions in the frog evoked the reassignment of the dorso-ventral axis (10) and established, for the first time, the origin of the anterior-posterior (head-to-tail) axis (12). The implications of these revisions are widespread, forcing the reevaluation of developmental phenotypes in a number of model organisms, including the zebrafish (11).