Development of Oligonucleotide-directed proximity-interactome MAPping (O-MAP), for characterizing RNA-protein interactions and higher order subnuclear architecture in situ

Abstract

RNA-protein interactions underscore a broad array of regulatory mechanisms throughout biology, and dysregulation of these interactions contributes to a broad range of human pathologies such as neurodegenerative disorders and many cancers. Recent studies have discovered thousands of RNA-binding proteins that lack canonical RNA binding domains. Additionally, the human transcriptome harbors tens of thousands of novel RNAs that can be differentially regulated in disease. RNA-protein complexes, termed ribonucleoproteins (RNPs), can also serve as structural scaffolds for reorganizing local subcellular environments and assembling subnuclear bodies. A classic example of RNA-scaffolded structures is the nucleolus, a subnuclear body that serves as the site of ribosome biogenesis, cell cycle regulation, and stress responses. Nucleolar assembly is initiated de novo after each cell division—nucleated by pre-ribosomal RNA (pre-rRNA) and other nucleolar factors that are preserved during the disassembly of the nucleolus at the start of mitosis. This formation is thought to be driven by RNA-protein interactions that phase separate into the nucleolar tripartite structure. Other RNAs, such as X inactive specific transcript (XIST) and MALAT1, are thought to have an architectural role in organizing subnuclear Barr bodies and nuclear speckles for gene regulation. However, studying specific RNA and their protein interactomes is challenging by conventional methods. Most approaches for analyzing the composition of a subcellular structure rely on the biochemical purification of that structure or enriching for interacting proteins using antisense oligonucleotides of the target RNA bound to a resin. These approaches are either impossible for smaller structures or are plagued with nonspecific binding and low specificity after exposure to crude lysate. To address these problems, recent work has utilized proximity labeling as a way to study specific RNA-protein interactomes within the subcellular context; however, few methods exist to specifically target and enrich for RNA scaffolded subnuclear bodies. Utilizing existing RNA-fluorescence in situ hybridization (FISH) technologies and proximity labeling, this thesis focuses on the development of Oligonucleotide-directed proximity-interactome (O-MAP) to target specific RNAs within their subcellular context and probe for interacting proteins. This technique targets antisense oligonucleotides to a RNA of interest in fixed cells and recruits a proximity labeling enzyme, horse radish peroxidase (HRP), to the target RNA enabling the promiscuous biotinylation of nearby proteins. These biotinylated proteins can be isolated by streptavidin enrichment and analyzed by downstream omics such as mass spectrometry (MS). Using the 47S pre-ribosomal RNA, long noncoding RNA XIST and 7SK small nuclear RNA as models, O-MAP induces precise biotinylation of target RNA and nearby proteins that can then be systematically analyzed by mass spectrometry. O-MAP-MS at these RNA targets characterized these RNA scaffolded compartments and interacting proteomes, highlighting classes of proteins involved in nucleolar biology and nuclear speckles. Without the need for genetic manipulation, O-MAP is both easily portable to other cell lines, organoids, tissues as well as RNA targets of varying expression and lengths. Additionally, O-MAP can probe differentially-regulated proteomes in disease-relevant states including pancreatic cancer and cellular stress. These studies demonstrate the versatility and specificity of O-MAP as well as its potential to provide new characterizations in RNA interactome biology.