The development of the heart and body wall muscles of the Drosophila embryo represents a powerful model for investigating transcriptional regulatory networks and the molecular underpinnings of organ formation. Moreover, the evolutionary conservation of key pathways involved in these processes insures that findings from this model organism will be relevant to other biological systems, including human. The Laboratory of Developmental Systems Biology uses an interdisciplinary approach to study heart and muscle development that combines genetics, genomics, biochemical, molecular, cellular and computational methods, thereby providing a systems-level understanding of organogenesis. Both the heart and somatic muscles are mesodermal derivatives arising from progenitor and founder cells (FCs) that are progressively determined by a well-orchestrated set of intrinsic and extrinsic regulators. The former include the tissue-restricted transcription factors (TFs), Twist (Twi) and Tinman (Tin), while the latter include intercellular signals mediated by Wnt, BMP, receptor tyrosine kinase/Ras and Notch pathways. Our prior studies suggested that these generic signals yield germ layer-specific outputs by functioning together with Twi and Tin to regulate the expression of target gene enhancers. For example, the combination of Pointed (Pnt, a Ras-activated TF), Twi and Tin serves as a regulatory code for a subset of FC genes. However, these findings do not explain how individual cell type-specific gene expression programs are generated. To address this critical question, we have turned to a more systematic analysis of the transcriptional regulatory networks that govern muscle and heart development. To identify a more comprehensive set of mesodermal TFs and their candidate target genes, we have undertaken flow cytometric purification and microarray-based RNA expression profiling of primary mesodermal cells isolated from wild-type embryos and from informative gain- and loss-of-function genotypes. A statistical meta-analysis of these datasets yielded groups of genes that are likely to be expressed in various subtypes of muscle and heart cells. These predictions were then validated by whole-mount embryo in situ hybridizations. By combining these findings with literature curation, we identified large numbers of relevant TFs and their potential downstream targets. Through a collaboration with Dr. Martha Bulyks laboratory at Harvard Medical School, we next used protein binding microarray (PBM) technology to determine the DNA binding specificities of a large number of mesodermal TFs at the level of individual k-mers and their relative affinities. This information was converted into a position weight matrix for each TF, and these sequence motifs were used to scan the entire Drosophila melanogaster genome for evolutionarily conserved clusters of biologically relevant TF binding sites. Windows of noncoding sequences containing such clusters were identified as candidate cis-regulatory modules, and transgenic reporter assays were used to validate these computational predictions. The statistical overrepresentation of particular Boolean combinations of binding sites among co-expressed muscle and heart genes was also assessed. In this manner, we found that a four-way AND combination of Pnt, Twi, Tin and homeodomain (HD) binding motifs comprises a potential regulatory code for subsets of muscle FC genes that are responsive to particular HD FC identity (HD-FCI) TFs. A comparison of PBM k-mer results for multiple TFs of this class further revealed that most bound k-mers are shared. However, some k-mers are preferentially bound by one member of the HD-FCI TF family, and these motifs are overrepresented among FC genes whose expression is responsive to this same TF. The presence in noncoding sequences of these preferred k-mers, together with Pnt, Twi and Tin binding sites, also contributed to the de novo capacity for predicting cognate HD-responsive FC genes when combined with a meta-analysis of informative expression profiling data. When a HD-FCI TF-preferred k-mer is mutagenized in a FC enhancer that is both dependent and responsive to this TF, as determined in appropriate loss- and gain-of-function genetic backgrounds, transcriptional activity of this element is lost in a FC that also expresses and requires this TF for its proper development. These experiments provide cis and trans evidence for the central role of a HD-FCI TF in the regulation of FC-specific gene expression. In addition, this is the first demonstration that the specificity of a HD TF can be influenced by binding to a unique DNA sequence that is not recognized by related members of the same TF family. Finally, the addition of HD binding sites to the known Pnt+Twi+Tin code extends our understanding of how FC-specific genetic programs are generated. Ongoing experiments are designed to evaluate whether additional classes of TFs contribute to the combinatorial regulation of gene expression in other mesodermal cells, and whether higher order regulatory codes than the four-way partnership represented by Pnt+Twi+Tin+HD can be derived for individual FCs. In addition to studying how TFs orchestrate gene regulatory networks, we are characterizing the downstream developmental pathways that are initiated by these proteins. Using RNAi and classical genetic methods, we found that a forkhead domain TF is essential for proper formation of the heart. In the absence of this TF, the Drosophila heart exhibits altered numbers and misalignment of myocardial cells, with some myocardial cells having atypical giant nuclei. Given that myocardial cells arise by asymmetric cell divisions that produce myocardial and pericardial cells and by symmetric cell divisions that produce only myocardial cells, we hypothesized that the observed phenotypes could be a consequence of one or more of the following mechanisms: (i) defects in asymmetric precursor cell division such that both daughter cells become myocardial;(ii) defects in asymmetric cell division of a precursor which results in both daughter cells becoming pericardial;(iii) defects in cytokinesis and/or mitosis during asymmetric cell division such that myocardial cell number does not change but giant nuclei result;and (iv) defects in cytokinesis and/or mitosis during symmetric cell division, resulting in a local reduction of myocardial cells and the formation of giant nuclei. Using informative cardiac cell-specific markers, we confirmed that all four of these mechanisms can occur in different embryonic segments. Intriguingly, we also found evidence suggesting that the same gene plays a role in an earlier round of asymmetric cell division that gives rise both to a somatic muscle founder cell and to the shared precursor of a myocardial and a pericardial cell. Experiments are in progress to confirm this possibility, which would additionally implicate the same gene in somatic muscle development and in the regulation of both cardiogenic and myogenic genes. Other ongoing work is focused on identifying the direct targets of the forkhead domain TF that is encoded by this gene in an effort to dissect an entire biological subnetwork from its generation by transcriptional mechanisms to its downstream developmental functions during organogenesis. Taken together, the above studies provide new insights into the transcriptional codes that regulate muscle and heart gene expression, and into the specific developmental roles played by individual TFs that specify cellular identity and control subsequent differentiation. These investigations also serve as an instructive experimental paradigm for studying related questions in other biological systems.
| Enriquez, Jonathan; Boukhatmi, Hadi; Dubois, Laurence et al. (2010) Multi-step control of muscle diversity by Hox proteins in the Drosophila embryo. Development 137:457-66 |
| Busser, Brian W; Bulyk, Martha L; Michelson, Alan M (2008) Toward a systems-level understanding of developmental regulatory networks. Curr Opin Genet Dev 18:521-9 |
| Michelson, Alan M (2008) Developmental biology. From genetic association to genetic switch. Science 322:1803-4 |
| Estrada, Beatriz; Michelson, Alan M (2008) A genomic approach to myoblast fusion in Drosophila. Methods Mol Biol 475:299-314 |
