Physiogenomic Control of Cell Fate in the Sea Urchin Embryo
The genetic program of development is encoded as a gene regulatory network (GRN), a contingency-based computational system specified in the genomic DNA sequence by the organization of transcription factors target sites in the cis-regulatory domains of genes (E.H. Davidson, Genomic Regulatory Systems, Academic Press, 2001). The architecture of this network determines the flow of genetic information during development. Thus, the reductionistic analysis of single genes and proteins that has been the predominant paradigm for biomedical research over the last half century, and which has been phenomenally successful in cataloging the genetic “toolkit” underlying developmental regulation, is inherently unable to engender a deep understanding of the genetic program of animal development. Rather, such an understanding will ultimately require elucidation of the architecture and dynamics of entire gene regulatory networks.
Among the model systems for biomedical research, one of the best suited to elucidation of GRN architecture is the sea urchin embryo. Currently the GRN that controls specification of the sea urchin larval endomesoderm is being mapped in the laboratory of Eric Davidson at the California Institute of Technology with additional contributions from several other laboratories. This large-scale collaborative effort, which has been greatly facilitated by the Strongylocentrotus purpuratus genome project at the Human Genome Sequencing Center at the Baylor College of Medicine, will soon lead to a deeper understanding of the genomic regulatory program underlying sea urchin development than has been achieved for any other animal.
A major advantage of the GRN approach pioneered by the Davidson lab is that it provides a systematic (as opposed to an ad hoc) method of generating testable hypotheses. That is, perturbation analyses are used to define hypothetical regulatory linkages, which can then be tested by chromatin immunoprecipitation (ChIP) and cis-regulatory analysis. A set of bioinformatics tools is now available to greatly facilitate the latter through the comparison of gene sequences from two species of sea urchin, S. purpuratus and L. variegatus. Using this approach, our research is currently focused on the following two developmental problems:
Problem 1: Developmental Coordination of Cell Proliferation and Differentiation
A fundamental prerequisite for the development and evolution of complex multicellular organisms is that the control of cell division be integrated into a higher-level program of organismal development. In animals, progression through (and exit from) the cell cycle is contingent upon signals that link cell cycle control to specific sub-elements of the developmental GRN. For example, mitogenic stimulation of G1 to S phase transit is controlled in part by the transcriptional regulation of cyclin D, which is linked to the GRN through numerous different signaling pathways that engage the cis-regulatory system of the cyclinD gene. In this way, cell division is coordinated with patterning of the embryo. While a number of labs are using various model systems to study the cell cycle in a developmental context, a comprehensive effort has yet to be launched toward defining the linkages within a developmental GRN that coordinates cell division with differentiation and pattern formation.
The sea urchin embryo has a very simple pattern of cell proliferation. In the late blastula stage embryo, actively dividing cells become confined to the endomesoderm and oral ectoderm, while cells in the aboral ectoderm and skeletogenic mesoderm exit the cell cycle to terminally differentiate. This pattern—global early followed by endomesoderm plus oral ectoderm late—is displayed by ~40% of the genes expressed in the sea urchin embryo, many of which are required for growth and cell division. Research in the Coffman lab is aimed at defining the GRN architecture that underlies this developmental pattern. The transcription factor SpRunt-1 is being used as an initial portal into this GRN, since we have shown that it is expressed in the proliferation-specific pattern described above and is required both for the transcriptional activation of cyclin D and for the normal program of cell proliferation. Currently, we are undertaking a cis-regulatory analysis of the SpCyclinD gene as well as analyzing the expression of additional cell cycle control genes, transcription factors, and differentiation markers at different stages of development in embryos that are depleted of SpRunt-1 by morpholino antisense-mediated knockdown. We are also applying a more systematic approach that uses microarrays to identify the global set of genes that are mis-expressed in the perturbed embryos. Putative regulatory linkages revealed by these perturbation analyses are being tested by ChIP, and they will also be explored in detail by analysis of the cis-regulatory systems of the most interesting target genes. Ultimately, these studies will facilitate the development of computational models of how cell proliferation and differentiation are coordinated during sea urchin embryogenesis, which will in turn generate questions and hypotheses for further experimental analysis.
Problem 2: Cell Physiology of Oral-Aboral Axis Specification in the Sea Urchin Embryo
The sea urchin pluteus larva is bilaterally symmetric, and is thus organized along two orthogonal axes: an invariant primary or animal-vegetal (AV) axis, and a labile secondary or oral-aboral (OA) axis. Development along the AV axis generates the three germ layers (ectoderm, endoderm, and mesoderm), while development along the OA axis further differentiates ectoderm into oral and aboral territories. The oral ectoderm is a complex neurogenic epithelium and the locus of mouth morphogenesis, while the aboral ectoderm is a simple squamous epithelium. It was recently demonstrated that localized zygotic activation of nodal during late cleavage stage sets in train the gene regulatory interactions that pattern the OA axis. However, the initial asymmetry that specifies OA polarity remains unknown. One of the earliest manifestations of OA polarity is a redox gradient reflecting an asymmetry in mitochondrial distribution. We have shown that imposing a respiratory gradient on embryos tends to entrain OA polarity, with the more active (oxidizing) side of the embryo having a strong bias to develop as oral (Coffman and Davidson, Dev. Biol. 230: 18-28, 2001). More recently we discovered that unfertilized eggs already contain an asymmetric distribution of mitochondria that is predictive of OA polarity, and that redistribution of the mitochondria by centrifugation or mitochondrial transfer also tends to entrain OA polarity (Coffman et al., Dev. Biol. 273: 160-171, 2004). Furthermore, specification of oral ectoderm via nodal activity is dependent on a relatively oxidizing redox state. These findings suggest a model in which nodal is precociously activated by redox-sensitive transcription factors in the cleavage stage blastomeres that contain the highest density of mitochondria, which leads to an initial asymmetry in nodal activity and consequent specification of the OA axis. We are testing this hypothesis by an analysis of the nodal cis-regulatory system. Ultimately, we expect that this work will converge with our work on the control of cell proliferation (since one of the ultimate manifestations of OA polarity is a difference in cell proliferation, as described above), as well as with work in other labs concerning the molecular basis of OA axis specification.
Research Publications
