Much of the genetic variation which makes us different from one another, including what makes some of us more susceptible to diseases, from autism to cancer, lies in non-coding regions of the genome which regulate gene-expression. These non-coding regions either directly interact with genes to alter their transcription or interact with one another to indirectly regulate gene-transcription. Interactions typically depend on the physical proximity of the sequences, and are thus determined by the three-dimensional (3-D) packaging of the genome. Changing the physical proximity of regulatory sequences changes which genes they influence. The 3-D organization which shapes these contacts is not well understood. Existing methods have analyzed either a single distance between a pair of points in a domain of interest in single cells (through fluorescent microscopy) or interactions among many points in a domain in population aggregates (through a ligation-based assay known as chromosome conformation capture, 3C). The first approach captures the large variation in genome structure between cells but provides very little information on what the structure looks like. The second approach provides more detailed and higher-resolution information about the average structure, but it is unclear to what extent individual cells resemble the average of all cells. Here we develop an imaging approach which measures thousands of inter-region distances per cell along the domain of interest. We resolve structural details previously only captured by the highest resolution 3C approaches requiring millions of cells, and do so with single-cell resolution. Moreover, our method scales to analyze thousands of cells per experiment, can be performed in complex tissue, and can measure nascent transcription and 3-D structure at the same time. We propose to combine this method with existing genetic tools to better understand how interactions among regulatory regions shape gene expression during development, focusing on genome regions with established major roles in development. We will perform an unbiased analysis of the contact interactions at these important loci across development in Drosophila and mouse embryos and relate differences in contact organization to differences in gene expression. We will also directly test several recent and influential hypotheses about gene regulation which are difficult to test with available methods. These include recently proposed models (derived from 3C data) describing how the genome is packaged in individual cells and new hypotheses about the processes which establish this packaging. We also aim to test models from genetic analyses which have postulated that regulatory elements compete for access to shared target genes and contrasting models which propose the existence of cooperative, hub-like interactions among regulatory elements. Our unprecedented characterization of the relation between spatial genome organization and transcriptional regulation in development will greatly enhance our understanding of the interactive behaviors of non-coding regulatory sequences, an essential component of our genomes and health.
The control of gene expression in shaping diverse processes, from embryonic patterning to memory formation, depends on numerous non-coding regulatory sequences in the genome. The behavior of these regulatory sequences is intimately linked to their three-dimensional spatial organization in the nucleus, but this organization is poorly understood. By developing revolutionary imaging tools to observe genome structure during embryogenesis, this proposal will decipher the links between genome spatial organization, gene expression, and embryonic patterning defects.
