Title

Insights Into the Biogenesis of Stress-Induced Readthrough Transcripts

Date of Award

Spring 2021

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Molecular Biophysics and Biochemistry

First Advisor

Steitz, Joan

Abstract

Readthrough transcription, defined as transcription extending beyond the annotated termination site of genes, is induced by cellular stress and results in the production of thousands of long noncoding RNAs called DoGs (downstream-of-gene transcripts). The localization and induction patterns of these transcripts have been extensively characterized in different cellular contexts. However, the mechanisms that lead to DoG production upon cellular stress remain unknown. To gain a better understanding of the mechanisms that lead to DoG production, I investigated the transcriptional and proteomic landscapes that accompany the induction of these RNAs in human cells exposed to hyperosmotic stress.In collaboration with Dr. Matthew Simon’s lab, I studied the transcriptional profiles that accompany DoG induction in the context of hyperosmotic stress using transient transcriptome (TT) sequencing combined with TimeLapse (TL) chemistry. The results revealed widespread transcriptional repression induced by hyperosmotic stress from which very few genes escape. Analyses of expressed genes that do not overlap with DoG regions demonstrate that DoGs are produced regardless of the transcriptional regulation of their upstream genes. To assess how hyperosmotic stress impacts the protein interactome of RNA Polymerase (Pol) II, I collaborated with Dr. Jesse Rinehart’s lab to perform mass spectrometry and western blot analyses on anti-Pol II immunoprecipitates obtained from chromatin fractions of untreated and KCl-treated HEK-293T cells. Consistent with the transcriptional changes observed, mass spectrometry results revealed drastic alterations to the Pol II interactome upon stress. Interestingly, subunits of the Integrator complex were not detected among the Pol II interactors in cells exposed to hyperosmotic stress, while subunits of the cleavage and polyadenylation machinery were present in both untreated and stressed samples. ChIP-seq experiments performed with antibodies against Int11 and Int3 demonstrated a decreased occupancy of these subunits near the transcription start sites of DoG-producing genes and of non-DoG genes that was not dependent on a loss of Pol II binding at these sites. Furthermore, siRNA-mediated knockdowns of Int11, the catalytic subunit of the Integrator complex, revealed the induction of hundreds of DoGs. I compared the identity of genes that generate readthrough transcripts after Int11 knockdown to genes that produce DoGs after KCl-treatment and found an overlap of up to 25%. This suggests that the observed decrease of Int11 binding is important, but not solely responsible for DoG induction in the context of hyperosmotic stress. The findings presented in this dissertation provide insights regarding the biogenesis of a recently characterized class of long noncoding RNAs in human cells exposed to hyperosmotic stress. My observations also clarify the extent of transcriptional repression experienced by human cells upon exposure to hyperosmotic stress and shed light on a previously unappreciated role of the Integrator complex in transcriptional regulation.

This document is currently not available here.

COinS