Three-dimensional mapping and modeling of combinatorial interactions underlying biomolecular condensates in olfactory neurons

Summary: Gene regulation is controlled by the dynamic localization of numerous regulatory factors to precise nuclear targets. Classic diffusion and affinity models alone cannot explain various aspects of how these processes occur, and growing observations suggest that regulation of gene expression is not a linear process that simply reflect the number of regulators and targets in a cell. Recently, spatial organization of regulatory molecules via formation of biomolecular condensates (BMCs) has emerged as a likely explanation for the non-linear regulation of gene expression. However, the functional roles of most condensates remain largely unknown because we lack methods to simultaneously measure the many different protein and RNA regulators that bind DNA, their 3D structural interactions in the nucleus, and gene expression within the same individual cell. To overcome this challenge, we will develop an integrative new framework consisting of cutting edge molecular and spatial measurements of DNA, RNA, and proteins within single cells combined with a novel machine learning approach to identify the critical molecular components required for organization of large, interconnected molecular interaction networks within BMCs. We will apply these approaches to dissect a long-standing question in neuroscience – how olfactory neurons stochastically express one, and only one, olfactory receptor gene out of the >1000 distinct gene located throughout the genome. The transition from polygenic to monogenic expression coincides with genomic transformations that organize the regulatory landscape of receptor transcription into competing multi-chromosomal enhancer hubs, localization of numerous regulatory proteins to distinct hubs, and a critical role for nascent receptor mRNA concentration in symmetry breaking between distinct hubs. The overall objective of this work is to develop generalized frameworks for measuring BMCs and for understanding and predicting relationships and causal components within them, as well as a detailed characterization of a foundational neuroscience model. We will accomplish this via the following Specific Aims: (1) Define the spatial and molecular composition and dynamics of BMCs within olfactory neurons, (2) Formulate a neural network model to identify critical nodes for BMC organization and function, (3) Validate the functional role of predicted causal “nodes” in olfactory receptor regulation. Our research team, with collective expertise in molecular mapping, computational biology, and olfactory system research, is uniquely positioned to address these questions.