hi 2002 the first provisional version of the endomesoderm GRN was published (Davidson et al, 2002a;Davidson et al., 2002b), and in the past year the first drafts of the ectoderm GRN were published, work that included contributions from the McClay lab (Bradham and McClay, 2006);see http://sugp.caltech.edu/endomes/#EctoNetworkDiagram (Figure la, lb). The Network models were constracted by Eric Davidson based on data from many laboratories with major contributions from the Davidson and the McClay labs. I spent my sabbatical year in the Davidson lab in 2002, a stay that has resulted in a long-term collaboration with Eric Davidson and members of his laboratorv. To date 9 papers have come from that collaboration with a number of additional papers linked to this proposal, and to ongoing research (Amore et al., 2003;Davidson et al, 2003;Davidson et al., 2002a;Davidson et al., 2002b;Oliveri et al., 2003;Oliveri et al., 2006;Otim et al, 2004;Sodergren et al., 2006). The original goal was to constract a GRN that reflected the sequence of endomesoderm specification during the first 30 hours (up until the beginning of gastmlation). The network assembly was restricted to transcription factors and to signal transductions. This was because we knew that to include all of cell biology as well as physiology of development was beyond reach at that time. We also wanted to constract the GRNs in such a way that each edge and each node of the network could be authenticated experimentally. Technologies were developed so that each component of the current GRNs generally have at least three independent sources of experimental support. In some cases predicted franscriptional inputs were verified at the cis-regulatory level, a connection that """"""""hardwires"""""""" the previous predictions of the Network (those cis-regulatory coimections that have been verified in this way are indicated by the thick edges in Fig. la). As seen in the Davidson component of this Program Project, efforts continue to add more cis-regulatory confirmation, and to develop new strategies that allow more rapid analyses to this tedious, but cracial component of network solutions. The McClay component of the project has been to coimect signal transduction devices to the transcriptional network components, and to provide a number of experimental embryological approaches to validate network predictions, and connect the GRN to morphogenesis. That effort required a number of new assays and has led to the current view of the Network as outlined in this Proposal. Why, it might be asked, would one want such a detailed look at how an embryo is specified? After all, when the GRN is displayed to students, an audible gasp at the complexity is heard. Despite that reaction, the reality is that the mechanisms of specification in all cells are highly complex interactions of many transcription factors and many signaling devices. Inside each nucleus of every cell there is an operational network of transcription factors governing the progression of development and the physiology of that functioning cell. Each time two daughter cells assume different identities, a single network state must diverge into two different network states. The arrangement and distribution of cell network states in an embryo must constantly be coordinated and many signaling inputs have been discovered to accommodate those requirements. In short, to understand development it is essential to explore gene regulatory network assemblies, mechanisms of divergence, interconnection through signaling between cells, and subcircuits that control the progression of an embryo toward adulthood and a new reproductive cycle. This is a daimting challenge, but if one seeks to really know how the system operates, a detailed analysis of gene regulatory networks is essential. Accepting this, the next question is """""""" how best to analyze network circuitry?"""""""" Currently many laboratories ask this question. Some utilize microarray analyses, proteomic assays, ChlP-Chip assays and other high throughput approaches to identify candidate molecules for networks. While these approaches provide candidates for networks, and while they produce diagrams that look quite complex, often they are not authenticated cormections. The challenge and the goal must be to authenticate each connection as it actually works in the organism. Only when a network can be authenticated in the organism, with signaling inputs rationalized, can one begin to understand how the system actually works. That is the goal of this project. We seek to understand how the complexity and dynamics of gene regulatory networks program the early cells of the embryo and then drive those cells through the morphogenetic movements of gastmlation. By the time gastralation begins, the ectoderm, mesoderm, and endoderm in each deuterostome is at least partially specified. The long-term goal of this project has been to build and understand how Gene Regulatory Networks (GRNs) work in governing the specification of germ layers using the sea urchin embryo as a model. In this application, we extend that goal to understand how the ectoderm and endomesoderm GRNs cormect to, and control the events of archenteron invagination and ectoderm patteming. The sea urchin is used as a model for this effort because it is well suited for interrogation of the specification mechanisms, and the relative simplicity of gastmlation in this embryo is the prototype for deuterostome early development. In the first seven years as this project unfolded in the Davidson and McClay labs, with additional contributions from the sea urchin commxmity, more than 80 transcription factors (with perhaps on the order of 80 more yet to add) and a number of signal transduction inputs were identified and incorporated into a nuclear view of triploblastic specification in this organism. The GRN as currently modeled (Figure la,b below) provides the template for the next generation of studies that are proposed in this Program Project. Here, in this sub-project, three goals will advance the Network studies into novel areas to establish """"""""next generation"""""""" approaches. First, changes in the progression of the GRN currently are based on data collected at intervals. An important goal of this proposal is to establish tools for gathering GRN states in individual cells. This goal will better enable us to leam how the endoderm and mesoderm cells prepare for and then execute morphogenesis. As detailed in the Davidson project Genomicists view the entire GRN for their purposes (VfA), while developmental biologists prefer to view the subcircuits of that network that ran in each cell as development progresses (VfN). This is because the information relevant to the developmental biologist are the GRN states in each nucleus that progress toward, and control morphogenesis. A large number of pubhcations have provide anecdotal information on how archenteron invagination works (always with a black-box approach). Here the exciting challenge is to discover how those properties are controlled at the Network level. This effort will merely be a begiiming of what will be a major effort of many people to understand how a complex and dynamic rearrangement of cells is controlled at a franscriptional and signaling level. Further, we will expand the GRN exploration into a cormection with patteming. The ectoderm subdivides into oral and aboral halves with a ciliary band separating them. During the 2006 Genome Annotation project we led the effort to annotate all known signaling molecules.
The third aim will take advantage of that effort and the ability to identify signaling inputs functionally. Our effort, combined with the Davidson lab's advances in understanding the ectoderm GRN, will provide insight into how patteming information is produced and distributed between the ca. 500 cells of the ectoderm.
Showing the most recent 10 out of 163 publications
