Understanding the mechanisms underlying R-loop biogenesis and resolution in mammals

Project summary
During transcription, the nascent RNA can anneal with the template DNA strand behind the advancing RNA
polymerase and cause the formation of alternative DNA structures called R-loops. R-loop profiling studies have
revealed that these structures are prevalent in all genomes and form normally and dynamically. Under normal
conditions, R-loops serve important physiological roles. Yet, over the last decade, harmful R-loops that arise
when transcription is perturbed have been implicated as powerful triggers of genome instability from yeast to
humans. Harmful R-loops have also been linked to an increasing number of human disorders. What
distinguishes “good” R-loops from “harmful” R-loops remains mostly unknown. In this proposal, we aim to
dissect the mechanisms linking perturbed transcription, R-loop metabolism, and genome instability. This will be
accomplished by addressing three central questions. (1) What defines harmful R-loops? While harmful R-loops
have been proposed in many studies, they have never been directly defined at the genomic level. We will
leverage our unique expertise in R-loop profiling to characterize these proposed structures in the context of
well-defined human cellular models of RNA processing dysfunction. This work will define the diversity of
altered R-loop landscapes that result from defects in RNA splicing, termination, and export and will allow us to
identify how perturbed transcription results in altered R-loop distributions, boosting our knowledge of R-loop
biogenesis pathways. (2) Does genome instability result from harmful R-loops or from altered transcription
itself? While attention has been focused on harmful R-loops, the negative impacts of defective RNA processing
on transcription itself have seldom been considered. To disentangle possible R-loop effects from pure
transcriptional effects, we will carefully monitor transcriptional perturbations in cellular models of RNA
processing dysfunction. In addition, we will directly measure the accumulation of DNA damage markers in
relation to R-loops, allowing us to determine for the first time if altered R-loops are actually “harmful” or if they
simply reflect abnormal transcription. (3) What is the role of Ribonuclease H1 (RNase H1) in R-loop
metabolism? RNase H1 has a clear biochemical ability to resolve R-loops and its over-expression in cells
suppresses a variety of genome instability phenotypes attributed to harmful R-loops. Yet, little direct evidence
exists to show that cellular RNase H1 expression resolves nuclear R-loops. Furthermore, recent studies and
our preliminary data suggest that RNase H1 could instead work by mitigating the impact of altered transcription
itself. To address these two possibilities, we will develop cellular models of RNase H1 depletion and over-
expression in mammalian cells and conduct a broad characterization of the resulting genomic R-loop patterns
and transcriptional effects. Our work will resolve crucial knowledge gaps concerning the formation and roles of
putative harmful R-loops in genome instability in human cells. The function and targets of nuclear RNase H1
will also be clarified, possibly revealing this enzyme in a fundamentally new light. We expect that this work will
durably impact the field of genome maintenance and provide insights into a range of human disorders
characterized by genome instability and RNA processing dysfunction.

Grant Number: 5R35GM139549-05
NIH Institute/Center: NIH
Principal Investigator: Frederic Chedin