Date of Award

Spring 5-15-2017

Author's School

Graduate School of Arts and Sciences

Author's Department

Biology & Biomedical Sciences (Molecular Genetics & Genomics)

Degree Name

Doctor of Philosophy (PhD)

Degree Type

Dissertation

Abstract

Cell identity and function rely on intricately controlled programs of gene regulation, alterations of which underlie many diseases, including cancer. Epigenetic analyses of normal and diseased cells have started to elucidate different facets of epigenetic mechanisms for gene regulation. These include changes in nucleosome density, histone modifications, factor binding and chromosomal architecture. All of these aspects contribute to the activities of regulatory elements conferring promoter, enhancer and insulator functions and the cis-regulatory circuits formed by these elements. Despite this progress, an urgent need remains to profile these features and to study how they cooperatively function in normal and pathogenic settings. Here, using the mouse T cell receptor beta locus as a model, we first quantified 13 distinct features, including transcription, chromatin environment, spatial proximity, and predicted qualities of recombination signal sequences (RSS), to assess their relative contributions in shaping recombination frequencies of Vβ gene segments. We found that the most predictive parameters are chromatin modifications associated with transcription, but recombination efficiencies are largely independent of spatial proximity. These findings enabled us to build a novel computational model predicting Vβ usage that uses a minimum set of five features. Expanding on these results, we applied chromatin profiling and computational algorithms to other mouse antigen receptor loci, to classify and identify novel regulatory elements. We defined 38 chromatin states that reflect distinct regulatory potentials. One of these states corresponded to known enhancers and also identified new enhancer candidates in immunoglobulin loci. Indeed, all four candidate elements exhibited enhancer activity in B cells when subjected to functional assays, validating that our chromatin profiling and computational analyses successfully identified enhancers in antigen receptor loci. Finally, we translated these approaches to human B cell lymphoma to predict pathogenic cis-regulatory circuits composed of dysregulated enhancers and target genes. We then selected and functionally dissected a pathogenic cis-regulatory circuit for the mitosis-associated kinase, NEK6, which is overexpressed in human B cell lymphoma. We found that only a subset of predicted enhancers is required to maintain elevated NEK6 expression in transformed B cells. Surprisingly, a B cell-specific super-enhancer is completely dispensable to maintain NEK6 expression and chromatin architecture within its chromosomal neighborhood. Moreover, we showed that a cluster of binding sites for the CTCF architectural factor serves as a chromatin boundary, blocking the functional impact of a NEK6 regulatory hub on neighboring genes. These results emphasize the necessity to test predicted cis-regulatory circuits, especially the roles of enhancers and super-enhancers, when prioritizing elements as targets for epigenetic-based therapies. Our findings collectively pave the way for future investigations into the roles of cis-regulatory and architectural elements in regulating gene expression programs during normal development or pathogenesis.

Language

English (en)

Chair and Committee

Eugene Oltz

Committee Members

Douglas Chalker, Barak Cohen, John Edwards, Jacqueline Payton,

Comments

Permanent URL: https://doi.org/10.7936/K7N014ZQ