Deployment of regulatory genes during gastrulation and germ layer specification in a model spiralian mollusc Crepidula.

BACKGROUND: During gastrulation, endoderm and mesoderm are specified from a bipotential precursor (endomesoderm) that is argued to be homologous across bilaterians. Spiralians also generate mesoderm from ectodermal precursors (ectomesoderm), which arises near the blastopore. While a conserved gene regulatory network controls specification of endomesoderm in deuterostomes and ecdysozoans, little is known about genes controlling specification or behavior of either source of spiralian mesoderm or the digestive tract. RESULTS: Using the mollusc Crepidula, we examined conserved regulatory factors and compared their expression to fate maps to score expression in the germ layers, blastopore lip, and digestive tract. Many genes were expressed in both ecto- and endomesoderm, but only five were expressed in ectomesoderm exclusively. The latter may contribute to epithelial-to-mesenchymal transition seen in ectomesoderm. CONCLUSIONS: We present the first comparison of genes expressed during spiralian gastrulation in the context of high-resolution fate maps. We found variation of genes expressed in the blastopore lip, mouth, and cells that will form the anus. Shared expression of many genes in both mesodermal sources suggests that components of the conserved endomesoderm program were either co-opted for ectomesoderm formation or that ecto- and endomesoderm are derived from a common mesodermal precursor that became subdivided into distinct domains during evolution.

Development of the annelid axochord: insights into notochord evolution.

The origin of chordates has been debated for more than a century, with one key issue being the emergence of the notochord. In vertebrates, the notochord develops by convergence and extension of the chordamesoderm, a population of midline cells of unique molecular identity. We identify a population of mesodermal cells in a developing invertebrate, the marine annelid Platynereis dumerilii, that converges and extends toward the midline and expresses a notochord-specific combination of genes. These cells differentiate into a longitudinal muscle, the axochord, that is positioned between central nervous system and axial blood vessel and secretes a strong collagenous extracellular matrix. Ancestral state reconstruction suggests that contractile mesodermal midline cells existed in bilaterian ancestors. We propose that these cells, via vacuolization and stiffening, gave rise to the chordate notochord.

SeaBase: a multispecies transcriptomic resource and platform for gene network inference.

Marine and aquatic animals are extraordinarily useful as models for identifying mechanisms of development and evolution, regeneration, resistance to cancer, longevity and symbiosis, among many other areas of research. This is due to the great diversity of these organisms and their wide-ranging capabilities. Genomics tools are essential for taking advantage of these "free lessons" of nature. However, genomics and transcriptomics are challenging in emerging model systems. Here, we present SeaBase, a tool for helping to meet these needs. Specifically, SeaBase provides a platform for sharing and searching transcriptome data. More importantly, SeaBase will support a growing number of tools for inferring gene network mechanisms. The first dataset available on SeaBase is a developmental transcriptomic profile of the sea anemone Nematostella vectensis (Anthozoa, Cnidaria). Additional datasets are currently being prepared and we are aiming to expand SeaBase to include user-supplied data for any number of marine and aquatic organisms, thereby supporting many potentially new models for gene network studies. SeaBase can be accessed online at: http://seabase.core.cli.mbl.edu.

Employing BAC-reporter constructs in the sea anemone Nematostella vectensis.

Changes in the expression and function of genes drive evolutionary change. Comparing how genes are regulated in different species is therefore becoming an important part of evo-devo studies. A key tool for investigating the regulation of genes is represented by bacterial artificial chromosomes (BAC)-reporter constructs. BACs are large insert libraries, often >100 kb, which thus capture the genomic sequences surrounding a gene of interest, including all, or nearly all, of the elements underpinning regulation. Recombinant BACs, containing a reporter gene in place of the endogenous coding sequence of genes, can be utilized to drive the expression of reporter genes under the regulatory control of the gene of interest while still embedded within its genomic context. Systematic deletions within the BAC-reporter construct can be used to identify the minimal reporter in an unbiased way, avoiding the risk of overlooking regulatory elements that may be many kilobases away from the transcription start-site. Nematostella vectensis (Edwardsiidae, Anthozoa, Cnidaria) has become an important model in regenerative biology, ecology, and especially in studies of evo-devo and gene-regulatory networks due to its interesting phylogenetic position and amenability to molecular techniques. The increasing interest in this rising model system also led to a demand for methods that can be used to study the regulation of genes in Nematostella. Here, we present our progress in employing BAC-reporter constructs to visualize gene-expression in Nematostella. Using a new Nematostella-specific recombination cassette, we made nine different BAC-reporter constructs. Although five BAC recombinants gave variable effects, three constructs, namely Nv-bra:eGFP::L10 BAC, Nv-dpp:eGFP::L10 BAC, and Nv-grm:eGFP::L10 BAC delivered promising results. We show that these three constructs express the reporter gene eGFP in 10.4-17.2% of all analyzed larvae, out of which 26.2-41.9% express GFP in a mosaic fashion within the expected domain. In addition to the expression within the known domains, we also observed cases of misexpression of eGFP and examples that could represent actual expression outside the described domain. Furthermore, we deep-sequenced and assembled five different BACs containing Nv-chordin, Nv-foxa, Nv-dpp, Nv-wnta, and Nv-wnt1, to improve assembly around these genes. The use of BAC-reporter constructs will foster cis-regulatory analyses in Nematostella and thus help to improve our understanding of the regulatory network in this cnidarian system. Ultimately, this will advance the comparison of gene-regulation across species and lead to a much better understanding of evolutionary changes and novelties.

Mesoteloblast-like mesodermal stem cells in the polychaete annelid Platynereis dumerilii (Nereididae).

Spiral cleavage is observed in animals that belong to the lophotrochozoa, a large group of marine invertebrates. As characteristic for spiral cleavage, the bulk of mesoderm forms from one cell, the "4d blastomere." This process has not yet been followed in cellular detail in annelids except in the leech, where "mesoteloblasts," a pair of mesodermal stem cells, generate two bands of mesoderm precursor cells in an iterative fashion. It is so far unknown whether such stem cell-like lineage is a general property of 4d-derived mesoderm in spiralian larvae. To address this, we have analyzed the cell lineage of the 4d blastomere in the polychaete annelid Platynereis dumerilii, an emerging model for lophotrochozoan and spiralian embryology (Fischer et al., 2010), by 4D microscopy, a semi-automated cell tracking technique based on differential interference contrast serial imaging (Schnabel et al. '97). Our data reveal that the two daughter cells of the 4d cell undergo seven consecutive rounds of unequal cell divisions. They bud off smaller cells in ventral-vegetal orientation and thus show mesoteloblast- and stem cell-like behavior. Based on these findings, we suggest that mesoteloblast-like mesodermal stem cells that form continuous mesodermal bands are part of the Errantia + Sedentaria ground pattern. In the course of annelid evolution, the number consecutive divisions of these cells would have been low initially with <10 division cycles, giving rise to larval segments only, and then increased up to 35 as observed in clitellates.

Evo-devo in the era of gene regulatory networks.

Advanced genomics tools enable powerful new strategies for understanding complex biological processes, including development. By extension, we should be able to use these methods in a comparative fashion to capture evolutionary mechanisms. This requires a capacity to go deep and broad, to analyze developmental gene regulatory networks in many organisms, especially nontraditional models. As we usher in a new era of next-generation GRN (gene regulatory network) analysis, it is important to ask how to evaluate the evolution of network interactions. Particularly problematic, as always, is defining "independence": Are two character traits found together because they are functionally linked or because of historical accident? The same basic question applies to understanding developmental GRN evolution. However, the essential difference here is that a GRN defines a causal chain of events. An understanding of causal relations--how Genes A and B work in concert to drive expression of Genes C and D to create a new Territory E--gives hope for establishing "trait independence" in a way that purely correlative arguments--the association of the expression of Gene D in Territory E--never could. Insight into causality provides the key to interpretation, as seen in this simplified scenario. Real-world networks bring new degrees of complexity, but the elucidation of causal relations remains the same. Has the day arrived when a single laboratory has the wherewithal to conduct multiorganism gene network projects in parallel? No. However, we argue that day is closer than one might suppose. We describe how a speedboat GRN project in one's favorite nonmodel organism(s) might look and provide a framework for comparative network analysis.

Germ band differentiation in the stomatopod Gonodactylaceus falcatus and the origin of the stereotyped cell division pattern in Malacostraca (Crustacea).

We analysed aspects of the embryonic development of the stomatopod crustacean Gonodactylaceus falcatus focusing on the cell division in the ectoderm of the germ band. As in many other malacostracan crustaceans, the growth zone in the caudal papilla is formed by 19 ectoteloblasts and 8 mesoteloblasts arranged in rings. These teloblasts give rise to the cellular material of the largest part of the post-naupliar germ band in a stereotyped cell division pattern. The regularly arranged cells of the genealogical units produced by the ectoteloblast divide twice in longitudinal direction. The intersegmental furrows form within the descendants of one genealogical unit in the ectoderm. Hence, embryos of G. falcatus share some features of the stereotyped cell division pattern with that in other malacostracan crustaceans, which is unique among arthropods. In contrast to the other malacostracan taxa studied so far, stomatopods show slightly oblique spindle direction and a tilted position of the cells within the genealogical units. The inclusion of data on Leptostraca suggests that aspects of stereotyped cell divisions in the germ band must be assumed for the ground pattern of Malacostraca. Moreover, Stomatopoda and Leptostraca share the lateral displacement of cells during the mediolateral divisions of the ectodermal genealogical units in the post-naupliar germ band. The Caridoida within the Eumalacostraca apomorphically evolved the strict longitudinal orientation of spindle axes and cell positions, reaching the highest degree of regularity in the Peracarida. The phylogenetic analysis of the distribution of developmental characters is the prerequisite for the analysis of the evolution of developmental patterns and mechanisms.

ZOONET: perspectives on the evolution of animal form. Meeting report.

What drives evolution? This was one of the main questions raised at the final ZOONET meeting in Budapest, Hungary, in November 2008. The meeting marked the conclusion of ZOONET, an EU-funded Marie-Curie Research Training Network comprising nine research groups from all over Europe (Max Telford, University College London; Michael Akam, University of Cambridge; Detlev Arendt, EMBL Heidelberg; Maria Ina Arnone, Stazione Zoologica Anton Dohrn Napoli; Michalis Averof, IMBB Heraklion; Graham Budd, Uppsala University; Richard Copley, University of Oxford; Wim Damen, University of Cologne; Ernst Wimmer, University of Göttingen). ZOONET meetings and practical courses held during the past four years provided researchers from diverse backgrounds--bioinformatics, phylogenetics, embryology, palaeontology, and developmental and molecular biology--the opportunity to discuss their work under a common umbrella of evolutionary developmental biology (Evo Devo). The Budapest meeting emphasized in-depth discussions of the key concepts defining Evo Devo, and bringing together ZOONET researchers with external speakers who were invited to present their views on the evolution of animal form. The discussion sessions addressed four main topics: the driving forces of evolution, segmentation, fossils and phylogeny, and the future of Evo Devo.