PROJECT 1: SPATIOTEMPORAL GENOME ARCHITECTURE

点击翻译按钮获取中文摘要

PROJECT 1 – SPATIOTEMPORAL ARCHITECTURE OF THE GENOME SUMMARY A central challenge of molecular biology is to understand how transcriptional regulatory elements are selected from the genome thereby specifying cellular identity and cell-specific responses. In this project, we will use systems biology approaches and the maps-to-model paradigm to gain insights into general mechanisms responsible for the selection and function of cis-regulatory elements necessary for transcriptional responses to pathogens. These activities consist of the following four Specific Aims. First, we will generate a map and use this to model the selection of cell-specific enhancer landscapes. This aim is based on the hypothesis that relatively simple combinations of lineage-determining transcription factors (LDTFs) play dominant roles in selecting a large fraction of cell-specific enhancers. We will develop and test a mechanistic network model that begins with genome-wide predictions of binding sites for macrophage LDTFs and progresses to predict the genome-wide binding locations of other collaborating transcription factors that contribute to enhancer selection. Second, we will map and model the role of signal-dependent transcription factors (SDTFs) in regulating the enhancer landscape. This aim is based on the hypothesis that an essential feature of a functional enhancer is that it is actively transcribed. In this aim, we will extend the mechanistic network model developed in Aim 1 to predict binding of SDTFs and subsequent transcriptional activation of enhancers following lipopolysaccharide exposure or adenoviral infection. The predictive value of the model will be tested by loss-of-function studies of the SDTFs and by evaluation of effects of natural genetic variation. Third, we will try to predict transcriptional activity as a function of enhancer interactions. We propose extending the mechanistic network model achieved in Aim 2 to consider co-regulated transcriptional start sites in the context of spatial co-localization. This aspect of the model will be tested by evaluating loss-of-function mutants and the impact of natural genetic variation on enhancer- promoter interactions using a modified version of the Hi-C assay that focuses sequencing power on interactions involving promoters. Finally, we will map and model the 3D virus-host genome interaction hubs and transcriptional networks that determine the outcome of infection across tissues and species. We hypothesize that viral genomes target and subvert the 3D organization and interactions of the host genome to activate different viral and host transcriptional programs in the time course of infection. The proposed studies will map and model 3D genome interactions and transcriptional programs within different tissue types that determine viral tropism and replication. These studies are of particular interest because, while the `early' program of human adenovirus infection is intact in mouse cells, their productive lytic replication/expression is blocked `late' through mechanisms that are poorly understood. A molecular understanding of these mechanisms would be highly significant at both conceptual and practical levels.