Higher-Order Chromatin States and Nuclear Structures Regulating Gene Expression

Citation

Goronzy, Isabel Nadine (2024) Higher-Order Chromatin States and Nuclear Structures Regulating Gene Expression. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/8gm2-jn84. https://resolver.caltech.edu/CaltechTHESIS:07272023-175309910

Abstract

Although the same genome is present in every cell, each cell type orchestrates a distinct gene expression program, which can be rapidly adapted in response to stimuli. Accordingly, gene regulation is a highly complex, context-specific process that involves the dynamic interplay between numerous regulatory factors. Most methods to study these regulatory factors only measure pairwise interactions between molecules and are limited to mapping one regulatory protein at a time. Consequently, the combinatorial complexity of gene regulation at individual genomic loci and the functional consequence of many regulatory factors remain underexplored. To address this, we have developed new sequencing-based approaches and computational analyses to comprehensively profile, at unprecedented scale, the diverse gene regulatory landscape and directly establish the link between regulatory factors and transcriptional outcomes. In Chapter 2, we present Chromatin Immunoprecipitation Done-In-Parallel (ChIP-DIP), a highly multiplexed method for mapping hundreds of proteins to DNA within a single sample. ChIP-DIP increases the throughput of existing methods by > 100-fold and enables the production of consortium-scale, cell type-specific data within a single lab. Capitalizing on the scale and diversity provided by ChIP-DIP, we uncover unique quantitative combinations of histone modifications that define distinctive classes of regulatory elements. Specifically, we find features distinguishing classes of promoters that correspond to different polymerase activity, transcriptional levels, and gene types and find acetylation patterns distinguishing classes of enhancers that exhibit distinct activity states, induction potential, and regulatory potential. Next, in Chapter 3, we apply RNA-DNA SPRITE (RD-SPRITE), a method for simultaneous measurement of RNA and DNA organization, to investigate the functional relationship between genome structure and transcription. We demonstrate that RD-SPRITE precisely detects individual, nascent pre-mRNAs at their transcriptional locus and, as a result, can be used to assess the 3D genome structure present during active transcription. We find that RNA polymerase II transcription occurs within genomic structures previously thought to be inactive, such as the B compartment and DNA regions near the nucleolus. This suggests that active transcription can occur throughout the nucleus and argues against structural domains that preclude transcription. Overall, our findings highlight the ability of RD-SPRITE to establish a structure-function link. Finally, in Chapter 4, we apply RD-SPRITE to study the transcriptional dependence of nuclear organization. We demonstrate that transcriptional inhibition leads to the loss of high-order structure around multiple RNA-processing bodies — the nucleolus, the scaRNA hub and the histone locus body — that are responsible for essential nuclear functions such as RNA processing and gene regulation. These findings suggest a role for RNA and nascent transcription in the formation and maintenance of long-range 3D contacts and critical nuclear compartments. In summary, we have developed new approaches to explore epigenomic and organizational complexity within the mammalian nucleus and have uncovered genome-wide principles of gene regulation.

Thesis Files